.H. .d.. 3. « .murén‘okmm “a.” _ Siva... rug. . )5 a u . : “filtflr I! J ‘ Jun}... t. n.3,? . . 5L1: ‘ .e e . a 3» Iv . .uufih. Vgfisfiz , 11 . 1.1. a ,3.“ h 1. . 1......15 I. . .vhflmr Shw‘runru l . I... 533.2. - 3.! Ac i=1?” . f . .31 ”an; " .-'? LIBRARIES j MICHIGAN STATE UNIVERSITY m, 3,1 EAST LANSING, MICH 48824-1048 (0 35 I? x J / a This is to certify that the dissertation entitled GENETIC ANALYSIS or SUCROSE ACCUMULATION IN SUGAR BEET (Beta vulgan's L.) presented by DANIELE TREBBI has been accepted towards fulfillment of the requirements for the PhD. degree in Crop and Soil Sciences - Plant Breeding and Genetic Program ,1 / Major PfofesSor’s Signature ’l fl/(m/s 5 Z 6703/ Date MSU is an Affinnative Action/Equal Opportunity Institution PLACE IN RETURN BOX to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 2/05 alcfiC/Dateouejnddp. 15 GENETIC ANALYSIS OF SUCROSE ACCUMULATION IN SUGAR BEET (Beta vulgaris L.) By Daniele Trebbi A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Crop and Soil Sciences 2005 ABSTRACT GENETIC ANALYSIS OF SUCROSE ACCUMULATION IN SUGAR BEET (Beta vulgaris L.) By Daniele Trebbi Genetic maps have been widely used to characterize important agronomic traits. The first objective of this research was to develop a sugar beet (Beta vulgaris L.) genetic map for quantitative trait loci (QTL) analysis of root sucrose content using AFLP markers generated with both EcoRI/Msel and Pstl/Msel restriction enzyme pair combinations. The genetic map was generated from a segregating population derived from an intra- specific cross between a sugar beet and a table beet line. The map showed a high density of markers on the n = 9 chromosomes of B. vulgaris. PstI/MseI-derived AF LP markers presented lower proportions of clustering, of segregation distortion and a higher proportion of markers linked on linkage groups with respect to EcoRI/MseI-derived markers. PstI/MseI-derived AFLP markers were highly efficient in generating the genetic map. QTL analysis of the compounded trait root sucrose content expressed as percent of the fresh weight (% SucF W) was more efficiently analyzed considering the combination of root dry matter content (% DM) and sucrose content expressed as percent of the dry matter (% SucDM) separately. The second objective of this research was to analyze QTL locations for % DM and % SucDM in relation to % SucFW to understand the relative importance of these two traits to the final values of % SucFW. Traits were evaluated on field-grown F3 progeny-tests derived from the mapping population for two museum ' "I: 4 . , mm. 0. title sit m ., .. 301 1‘9““ Usduu: Well} r- x T - 13.13“; gm: curs: ‘fimy _ .3 C, ”353‘ t I‘..1\,,~, T ." 15“.”...gua 0 45.14.?“ L vL m 4' ~ . at.» ‘J' ~41.- I‘M; \ A- g. 5‘ ‘5: '51. . Q . (a A U‘~ \JJI ‘- Kl a u ‘\\ v 5.8 ‘- nil en 93‘ p-' T ‘UU’ - k ‘u “:7“: ‘* .1‘ "M ‘ .- consecutive years. A total of seven QTL were detected for % SucFW, which all co- localized or with QTL for % DM (three) or with % SucDM (three) or with both (one), while six more QTL were only identified for % DM (one) and for % SucDM (five) but not identified for % SucFW. Overall, both DM and SucDM influenced SucFW and a total of 13 QTL appeared to control root sucrose content. The third objective of this research was to characterize the dynamics of sucrose accumulation and the changes in gene expression during early root development, and to identify genes differentially expressed between developmental phases characterized by different rates of sucrose accumulation. Sucrose accumulation dynamics were characterized in greenhouse- and field-grown plants for two consecutive years, and gene expression profiles were analyzed with cDNA-AF LP during early root development. Identification of genes differentially expressed during development was performed by sequencing clones derived from a subtractive cDNA library. Estimates of the level of up and down-regulation of differentially expressed genes was performed using Beta vulgaris L. and Arabidopsis thaliana L. cDNA and oligonucleotide microarrays. Sucrose started accumulating afier the 3rd week after emergence (WAE) and reached levels comparable to mature roots at the 7‘” WAE in greenhouse-grown plants. During the same period of time a major change of gene expression was also observed. Genes differentially expressed between developmental phases critical for sucrose accumulation were represented by regulatory genes more than structural genes. Early over-expressed transcripts were associated with cell cycle metabolism and osmotic regulation pathways, while genes up-regulated during active root sucrose accumulation were involved in transcriptional regulation and signal transduction pathways, in energy metabolism and in metabolite transports. Copyright by DANIELE TREBBI 2005 ‘ caret ‘1‘ I1.“ 3‘ Dedicato ai miei genitori Paola ed Alberto e a mia moglie Costanza per i! Ioro infinito supporto ed incoraggiamento. Dedicated to my parents Paola and Alberta and to my wife Costanza for their everlasting support and encouragement. Mould 11in ‘ I T""1‘.fi“.3 r - NwODD ‘1‘]: i 3‘ r-gpx - A" ‘kpcusutsb t1“ ‘fiA‘LJ' “d . ~st ad £ 5 Embers of m. ‘1’:«‘.l. '. u when t: 23‘30 ~. V ‘H ‘Ur “4:1 N“. 3. iut‘ . ' 36- F Ha‘[¥.‘t t. ACKNOWLEDGEMENTS I would like to thank my major professor and mentor Dr. J. Mitchell McGrath for his guidance and support over the past five years. I am particularly thankful for sharing his experiences and forcing me to think at ‘the big picture’, avoiding me doing major mistakes and getting lost in minor details. I would like to express my gratitude to the members of my guidance committee Dr. James F. Hancock, Dr. James D. Kelly, and Dr. Jack Preiss for their constant help and productive discussions. I’m also grateful to Dr. Dechun Wang for his advices on genetic mapping and QTL analysis, and to Dr. Alexandra ‘Sasha’ Kravchenko for her guidance during the statistical analyses. I am particularly thankful to Dr. Britta Schulz from KWS SAAT AG for providing critical genetic markers, and to Dr. Irwin L. Goldman fiom the University of Wisconsin for supplying the table beet parental line. I would also like to thank Dr. Jeff Landgraf, Dr. Annette Thelen, and Dr. Robert Halgren from the Genomics Technology Support Facility at MSU for their help and advices during gene expression analyses. A particular thank to the ‘Sugar Beet Lab’ group. I would like to thank Dr. Benildo de los Reyes, Susan Myers, Subashini Nagendran, Scott R Shaw, Gina Vergara, Shiranee Gunasekera, Tim Duckert, Teresa Koppin, Bob Sims, the students Azeza, Reid and Kevin, and Missy Stielfel for their help. I am particularly thankful to the USDA-ARS, the Beet Sugar Development Foundation, the Graduate School, the GREEEN Project, the Plant Breeding and Genetics Program, and the Crop and Soil Sciences Department for financial support. My sincere gratitude goes also to my parents Paola and Alberto for their support and understanding, and to my wife Costanza for her continuous encouragement. Thank you. vi llSl 0F TA usromc KEY TO St: KRUOXAU Lil: CHAPTER 1 Fl't'OROME Air 111:. Me: TABLE OF CONTENTS LIST OF TABLES ix LIST OF FIGURES xiii KEY TO SYMBOLS AND ABBREVIATIONS xviii RA'I‘IONALE AND OBJECTIVES OF THE RESEARCH .......................... 1 Literature cited .................................................................. 6 CHAPTER 1 FLUOROMETRIC SUCROSE EVALUATION FOR SUGAR BEET ............ 9 Abstract ........................................................................... 10 Introduction ...................................................................... 11 Materials and Methods ......................................................... 14 Plant material and growth conditions .............................. 14 Sugar extraction ...................................................... l4 Chromatography and refractometry ............................... 15 Enzymatic-fluorogenic assay ....................................... 16 Results ............................................................................ 18 Enzymatic-fluorogenic assay ....................................... 18 Comparison of analytical methods ................................. 19 Discussion ........................................................................ 28 Literature cited .................................................................. 31 CHAPTER 2 GENETIC MAP DEVELOPMENT IN Beta vulgaris L. ............................. 34 Abstract ........................................................................... 34 Introduction ...................................................................... 35 Materials and Methods ......................................................... 38 Plant material ......................................................... 38 AF LP markers ........................................................ 38 Linkage analysis ...................................................... 41 Results ............................................................................ 44 AF LP markers ........................................................ 44 Genetic map ........................................................... 45 Discussion ........................................................................ 55 Conclusions ..................................................................... 61 Literature cited .................................................................. 62 vii CHAPTER I ANALYSIS Bi'a vul’gtzn Al‘l In: Mr Lite CR‘APTER 4 PHENOIYP'. ACC'L‘ML'LA PHASE CHA Abs‘ liiift Mv Ht CHAPTER 3 ANALYSIS OF LOCI AFFECTING ROOT SUCROSE CONTENT IN Beta vulgaris L. ........................................................................... Abstract ........................................................................... Introduction ...................................................................... Materials and Methods ......................................................... Plant material and experimental design ........................... Traits evaluation ...................................................... Heritability estimate ................................................. QTL analysis ......................................................... Results ............................................................................ Traits evaluation ...................................................... QTL analysis ......................................................... Discussion ........................................................................ Conclusions .................................................................... Literature cited .................................................................. CHAPTER 4 PHENOTYPIC AND GENOTYPIC ANALYSES OF ROOT SUCROSE ACCUMULATION DURING JUVENILE TO ADULT DEVELOPMENTAL PHASE CHANGE IN SUGAR BEET ................................................... Abstract ........................................................................... Introduction ...................................................................... Materials and Methods ......................................................... Plant material and growth conditions .............................. Samples collection and traits analysis .............................. Nucleic acid purification ............................................ cDNA-AF LP protocol ............................................... Sugar beet root developmental cDNA library .................... Enrichment of differentially expressed genes ................... Library hybridization and cDNA sequencing .................... Beta vulgaris cDNA microarray ................................... Arabidopsis thaliana cDNA microarray .......................... Arabidopsis thaliana oligonucleotide microarray ............... Results ............................................................................ Phenotypic analysis .................................................. cDNA-AF LP ......................................................... Library hybridization and cDNA sequencing .................... Beta vulgaris cDNA microarray ................................... Arabidopsis thaliana cDNA microarray .......................... Arabidopsis thaliana oligonucleotide microarray ............... Discussion ........................................................................ Conclusions ....................................................................... Literature cited .................................................................. SUMMARY AND CONCLUSIONS .................................................... viii 67 67 68 73 73 74 75 75 77 77 78 105 109 1 10 113 113 114 117 117 118 119 120 123 123 125 128 134 135 137 137 140 141 142 146 148 210 217 219 225 mt Cicely. :UCEL‘CIidSS ID ‘3 o. )L.fct.‘-)C 33:31 9 .. 9.: 3 ‘ (g .331; fin). .‘h $513.1. F. 1 $2336, Sac. 13‘ Q11 are» “it 3.2 F. p ‘ J rsfl '0 S“ P ‘ G A tn, “\L «a, a“. ‘Jrl'qéhirus AH" r. . . t“ ‘ I - ' A: ‘e ks: [re \“:T=fir hfiao, " ‘ “‘“St 3““«3‘44 ‘ \\ “int”.- LIST OF TABLES Table 1.1. Mean dry matter sucrose contents measured with HPLC and EF A in five germplasm lines at seven developmental points. .......................................... 22 Table 2.1. AF LP adaptors and primers nucleotide sequences used to generate AF LP markers. (N) represents the number of selective nucleotides used during selective amplifications. .............................................................................. 43 Table 2.2. AF LP primer combination analysis. Rare cutters used with MseI (M) in primer combination (pc), refer to EcoRI (E) and PM (P), generating E/M and P/M pc, respectively. The three nucleotides following the E and M primer sequences, and the two nucleotides following the P primer sequence, represent selective nucleotides used during selective amplification. ..................................................................... 48 Table 2.3. Marker distribution fi'equency, segregation distortion, and clustering of markers per linkage group. AF LP markers were subdivided in two classes depending to the restriction enzyme pair used: EcoRI/MseI (E/M) or PstI/Msel (P/M). Non-AF LP markers were RFLP, SSR, EST-UTR and phenotypic markers (see text). Chromosome assignment according to Butterfass (1964). ................................................... 53 Table 3.1. F 3 progeny test — Year 2002. Root dry matter (% DM), sucrose content as dry matter (% SucDM) and sucrose content as fresh weight (% SucF W) estimates were used for QTL analysis. Ranking of each trait was performed with the multiple comparison with the best treatment (MCB) procedure (Hsu, 1984), where ‘b’ indicates no significant difference with the best, ‘w’ indicates no significant difference with the worst, ‘bw’ indicates no significant difference either with the best or with the worst. SD = standard deviation; nd = non detected; Rep. = number of replications. ........................ 82 Table 3.2. F 3 progeny test — Year 2003. Root dry matter (% DM), sucrose content as dry matter (% SucDM) and sucrose content as fresh weight (% SucF W) estimates were used for QTL analysis. Ranking of each trait was performed with the multiple comparison with the best treatment (MCB) procedure (Hsu, 1984), where ‘b’ indicates no significant difference with the best, ‘w’ indicates no significant difference with the worst, missing letter indicates significant difference either with the best or with the worst. SD = standard deviation; nd = non detected; Rep. = number of replications. ............... 84 ix Table 3.3. ATI {"r‘e D11 ). sue (It SucFW). d. = mime-J years; if = LIL" labia 3.4. C. sucrose some 306W) €1.31 coats-.1119, D}. P533311 “rig." Ritual allele: 13ml and Cd 33.11.10] {01' '. Téble 4.l_ Nu: Pratt Combzn“ T ' 1 2‘3 ‘ w,“ 1001 dwelt.) «:5 AL) A" If.“ 4 Rt: 7.5 8 is. ' t’ju h. an .‘ ~ w: a. V ““i 12%.: * . ‘!\.;‘ . . U. .‘ ‘rl~\§‘.‘ "~. "‘UD “.2 ., LALJ‘ \ft ‘0 Table 3.3. AN OVA table for broad sense heritability (h’B) estimates of root dry matter (% DM), sucrose content as dry matter (% SucDM), and sucrose content as fresh weight (% SucFW). 0'23 = genetic variance; 023, = variance due to genotype x year interaction; 02.; = variance due to experimental error, r = number of replications; y is the number of years; df = degrees of freedom; SS = sum of squares; MS = mean squares. ............... 90 Table 3.4. Composite interval mapping analysis of root dry matter content (% DM), sucrose content as dry matter (% SucDM), and sucrose content as fresh weight (°/o SucF W) evaluated on field-grown F3 progeny tests in 2002 and 2003. .............. 102 Table 3.5. Comparative phenotypic analysis (CPA) and parental effect of root dry matter content (% DM), sucrose content per dry matter (% SucDM), and sucrose content perfresh weight (% SucF W) in F3 families grouped based on homozygosity for each parental alleles in putative QTL genomic regions estimated with single marker analysis (SMA) and composite interval mapping (CIM). Significance of CPA at p < 0.1, 0.05, and 0.01 for *, ”, and “‘3 respectively. n.s. = non significant. SD = standard deviation ........................................................................................................ 103 Table 4.1. Number of transcript-derived fragments obtained with different cDNA-AF LP primer combinations. up = non performed. ......................................... 159 Table 4.2. Transcript sequences cluster analysis. Transcripts were selected based on plaque hybridization with early root development enriched transcripts (3rd WAE), with late root development enriched transcripts (7m WAE) or with both probes (3&7 WAE) ................................................................................................ 162 Table 4.3. Beta vulgaris cDNA microarray analysis: Number of clone hybridizations ....................................................................................... 165 Table 4.4. Beta vulgaris cDNA microarray analysis: Number of transcripts and variation of gene expression during root development detected by Beta vulgaris subtracted-library cDNA microarray. ............................................................................. 166 Table 4.5. Beta vulgaris cDNA microarray analysis: Number of differentially expressed genes that hybridized with different classes of subtracted probes. Levels of gene expression were evaluated for each biological replication (BR) of the two sugar beet lines USH20 and SR96. Differential gene expression refers to regulation during root development: Up indicates up-regulation at 7th with respect to 3rd WAE; Down indicates down-regulation at 7th with respect to 3“l WAE. ......................................... 167 Table 4.6. Beta vulgaris subtracted library microarray: clustered up-regulated transcripts. For each clone, differential gene expression was estimated independently for both USH20 and SR96 sugar beet lines (DGEHM) averaging the level of expression of the two biological replications, as follows: DGEune = [(t1 x DGE.) + (t; x DGE2)] / (t; + t;), where t1 and t; are the number of technical replications of the first and second biological replications, and DGE] and DGEz are the samples fold difference of gene expression of the first and second biological replications of the same line. Also, for each clone a mean differential gene expression (DGEm) was estimated combining data from the two sugar beet line as fOIIOW1 DGE". = [(tlineA X DGEIimA) + (tunes X DGEIWBH / (tlineA + tlineB), where um and tlineB are the total number of technical replications in both biological replications of the first and the second lines respectively, and the DGEmA and DGEmB are the DGE of the first and the second lines, respectively. :bCIgo indicates 90 % confidence intervals. Asterisks indicate clones sequenced in duplicate. nd indicates not detected. ...................................................................................... 168 Table 4.7. Beta vulgaris subtracted library microarray: singleton up—regulated transcripts ........................................................................................... 172 Table 4.8. Beta vulgaris subtracted library microarray: clustered down-regulated transcripts. .......................................................................................... 176 Table 4.9. Beta vulgaris subtracted library microarray: singleton down-regulated transcripts. Captions as in Table 4.6. .................................................. 181 Table 4.10. Arabidopsis thaliana cDNA microarray analysis: Number of transcripts and variation of gene expression during active sucrose accumulation. ........................ 185 Table 4.11. Arabidopsis thaliana cDNA microarray analysis: Genes up-regulated during active sucrose accumulation (7"I week after emergence). Average fold up-regulation (up- reg.) was estimated averaging two technical replications and 90 % confidence intervals (i C190) for the mean fold difference were estimated with the two-sided t-test. Hybridization levels refers to low (L), medium (M), high (H), and very high (VH) if the average hybridization mean intensities of the dyes of both replications were < than 2000, between 2000 and 4000, between 4000 and 10,000, or > 10,000 units, respectively. .............. 186 Table 4.12. Arabidopsis thaliana cDNA microarray analysis: Genes down-regulated during the late (seventh week after emergence) root development. Average fold down- regulation (down-reg.) was estimated averaging two technical replications and 90 % xi f l cortidcnce ir sided l-test .' (1'11) if the a' tan 3000. bc ltSpt‘cm'el}. Table 4.13. .-.’ will too high 353: express; .r ASmNnu TabietH. A; @5515: Gent- 13564.15. ,4. \‘zlbii 01 in; 15513 mil-61's It, tarragon : mm " I “.1 OUU a: 73326 4116. An its “We-gt? 1th refers 10 11 I35“ Olihe pm p34 1r... “a U110. 0r> l 8151: 417. .4,“ ‘ \ufs doun.refl‘ "'51: 1H? 3° hI-Sdizali) fever o "la b .. x309- 6r In“ W 80.: confidence intervals (3: C190) for the mean fold difference was estimated with the two- sided t-test. Hybridization levels refers to low (L), medium (M), high (H), and very high (V H) if the average hybridization mean intensities of the dyes of both replications were < than 2000, between 2000 and 4000, between 4000 and 10,000, or > 10,000 units, respectively. ............................................................................. 189 Table 4.13. Arabidopsis thaliana cDNA microarray analysis: Identification of clones with too high hybridization signals to be used to calculate relative ratio and change of gene expression between samples. These clones saturated the signal detector for both dyes in both technical replications. ........................................................... 195 Table 4.14. Arabidopsis thaliana oligonucleotide microarray, conventional (default) analysis: Genes up- and down-regulated during active sucrose accumulation. . ....199 Table 4.15. Arabidopsis thaliana oligonucleotide microarray, non-conventional analysis: Number of transcripts subdivided by hybridization signal intensities. Hybridization levels refers to low (L), medium (M), high (H), and very high (V H) if the average hybridization means of probe sets were between 400 and 500, between 500 and 600, between 600 and 1000, or > 1000 units, respectively. ................................ 200 Table 4.16. Arabidopsis thaliana oligonucleotide microarray, non-conventional analysis: Genes up-regulated during active sucrose accumulation (7“l WAE). Hybridization levels (HL) refers to medium (M), hi (H), and very high (VH) if the average hybridization mean of the probe sets of the 7 WAE samples were between 500 and 600, between 600 and 1000, or > 1000 units, respectively. .................................................. 201 Table 4.17. Arabidopsis thaliana oligonucleotide microarray, non-conventional analysis: Genes down-regulated during active sucrose accumulation (7lh WAE). Hybridization levels (HL) refers to medium (M), high (H), and very high (VH) if the average hybridization mean of the probe sets of the 3"I WAE samples were between 500 and 600, between 600 and 1000, or > 1000 units, respectively. ................................ 204 xii Figure 1.1 . composition. represent 121'. 5 Q 9" . 1.011312 'Etlufi LIST OF FIGURES Figure 1.1. Range of 142 overlaid HPLC chromatograrns analyzed for carbohydrate composition. Peaks at 6.94 ($0.02), 7.86 ($0.01), 9.72 (1:002), and 11.43 (:l:0.07) min represent raffinose, sucrose, glucose, and fructose, respectively. Carbohydrate concentrations were determined from peak areas. .......................................... 23 Figure 1.2. Determination of optimal range of glucose concentrations for a standard curve. The inner graph represents the glucose concentration range (0 - 16 uM) selected for the standard curve. ..................................................................... 24 Figure 1.3: Determination of sugar extract dilution optimum from a stande 60 mM sucrose solution. Bars represent :1: standard deviations of four replications. ...... 25 Figure 1.4. Comparison of sucrose quantification methods. (A) Correlation between sucrose contents as dry matter (DM) estimated with chromatographic (HPLC) and refractometric analyses (R2 = 0.950; y = 0.8801: + 0.451). (B) Correlation between sucrose contents as dry matter (DM) basis estimated with HPLC and Enzymatic fluorometric assay (R2 = 0.944; y = 0.984x - 0.015). Triangles and circles represent young greenhouse-grown and mature field-grown samples, respectively. ............... 26 Figure 1.5. Mean sucrose content comparison between analytical methods (HPLC vs. EFA) for each line at each sampled time point (data from Table 1.1.). Error bars represent Standard Deviations. ............................................................ 27 Figure 2.1. Frequency of primer combinations (pc) per number of polymorphisms generated. ....................................................................................... 49 Figure 2.2. Influence of AT content in selective nucleotides versus the total and polymorphic number of amplified fragments. Bars represent standard deviations (n 2 4). ................................................................................................ 49 Figure 2.3 Genetic map of Beta vulgaris. Chromosome nomenclature according to Butterfass (1964). Map distances in cM estimated using the Kosambi (1944) function. Black and grey parentheses indicate intervals of EcoRI/MseI-derived and PstI/Msel- xiii dented clustel imitate mark: respectively . . . Figure 2.4. M climates Bir 311:1ch for: market :er-d 11933.1. In expressed as c SR96 and LS} line 39a did 11. not ”Prefix-1e Figure 3.2. 1; jléds are ex 05139111165 5; 33113210115, 1:. 00"}: Years. an 1‘ , . see i ."'>,_ 3.6- 5 5 ‘4 £315“. 1;.“ F. derived clustered markers, respectively. Markers followed by C“), ("), and (***) indicate markers with distorted segregation at p < 0.05, < 0.01, and < 0.001, respectively .......................................................................................... 50 Figure 2.4. Marker inter-distances. A. Frequency of markers based on marker inter- distances. Bins sizes are equal to 0.1 cM for marker inter-distances between 0 and 4 cM, and 1 cM for marker inter-distances > 4 cM. B. Cumulative markers frequency based on marker inter-distances. ..................................................................... 54 Figure 3.1. Indexed root yield of F3 progeny tests during 2002 and 2003. Root yields are expressed as comparison (indexed root yield) with the averaged root yields of the lines SR96 and USH20 (index = 100) per each year. Bars represent standard deviations. F3 line 29a did not have enough roots for reliable estimate of root yield in both years, and is not represented in the graph. ..................................................................... 86 Figure 3.2. Indexed sucrose yield of F3 progeny tests during 2002 and 2003. Sucrose yields are expressed as comparison (indexed sucrose yield) with the average sucrose yield of the lines SR96 and USH20 (index = 100) per each year. Bars represent standard deviations. F3 line 29a did not have enough roots for reliable estimate of sucrose yield in both years, and is not represented in the graph. .......................................... 87 Figure 3.3. Correlation between root dry matter and root sucrose content as dry matter in F3 progeny tests. Black and grey lines represent correlation lines during 2002 (y = 20.95 - 0.04 x; R2 = 0.006) and 2003 (y = 11.95 + 0.12 x; R2 = 0.080), respectively. ...... 88 Figure 3.4. Analysis of correlation between root sucrose content as fresh weight and A. root dry matter, and B. sucrose content as dry matter. A. Black and grey lines represent correlation lines during 2002 (y = 5.68 + 1.24 x; R2 = 0.666) and 2003 (y = 7.49 + 1.02 x; R2 = 0.733), respectively. B. Black and grey lines represent correlation lines during 2002 (y = 39.39 + 1.59 x; R2 = 0.263) and 2003 (y = 35.08 + 2.12 x; R2 = 0.543), respectively. ................................................................................................ 89 Figure 3.5. Analysis of correlation of root dry matter content and sucrose content as dry matter between years. A. Black line represents correlation line (y = 16.24 + 0.14 x; R2 = 0.028). B. Black line represents correlation line (y = 32.47 + 0.46 x; R2 = 0.208). ...... 91 Figure 3.6. Meteorological data daily recorded at the Michigan State University Agronomy Farm, East Lansing, MI, during the growing seasons 2002 and 2003. A. Maximum and minimum temperatures (°C) during 2002. B. Differences of maximum xiv and minimum t: precipitations (1 Figure 3.7. Qu rater 1% Sucf (100 scores as 30215 sis ($31.31 (C111. full lines 0.5. 0.01. and C iig‘eie 3.8. Ge moseeonten‘ 80:111.). Map Been “ (l9t’ makers nomet F. 1333? 4. 1. Su 131120. SR95. anti emergenc and minimum temperatures (°C) during 2003 compared to 2002. C. Cumulative precipitations (mm) during 2002 and 2003. ................................................... 92 Figure 3.7. Quantitative trait analyses of root dry matter (% DM), sucrose content as dry matter (% SucDM), and sucrose content as fresh weight (% SucF W) evaluated in 2002 (LOD scores as grey lines) and 2003 (LOD scores as black lines), using single maker analysis (SMA), interval mapping (1M, dotted lines), and composite interval mapping (CIM, full lines). Map distance in cM. ‘, ", *“ and “** indicate difference at p < 0.1, 0.5, 0.01, and 0.001, respectively ............................................................. 93 Figure 3.8. Genomic locations of quantitative trait loci for root dry matter (% DM), sucrose content as dry matter (% SucDM), and sucrose content as fresh weight (% SucF W). Map distances in cM. Chromosomes (Ch.) nomenclature according to Butterfass (1964). Marker names have been omitted for clearness (refer to figure 3.7 for markers nomenclature). .................................................................... 104 Figure 4.1. Sucrose content as fresh weight of five sugar beet lines (SR87, SR96, USH20, SR95, and SR97) grown under greenhouse condition for a period of nine weeks after emergence during 2002. Bars represent :1: standard deviations. .............. 150 Figure 4.2. Root dry matter content of five sugar beet lines (SR87, SR96, USH20, SR95, and SR97) grown under greenhouse condition for a period of nine weeks afier emergence during 2002. Bars represent i standard deviations. ......................................... 151 Figure 4.3. Sucrose content as dry matter of five sugar beet lines (SR87, SR96, USH20, SR95, and SR97) grown under greenhouse condition for a period of nine weeks after emergence during 2002. Bars represent 3: standard deviations. ....................... 152 Figure 4.4. Sucrose content as fresh weight of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W357B) grown under greenhouse condition for a period of seven weeks after emergence during 2003. Bars represent i standard deviations. ..... 153 Figure 4.5. Root dry matter content of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W357B) grown under greenhouse condition for a period of seven weeks after emergence during 2003. Bars represent :1: standard deviations. . .154 XV Figure 4.6. C869) and 2 set en week Figure 4.7. C869) and a weeks afier Figure 4.8. 33:13 table 1 emergence c Figure 4.9, C8091 and a Weeks afifl’ 1 Here 4.10. dented 111:; the $31 5 \2 “Med T1 10‘ Obsen ed 53-31358 fi’OE ‘M‘Sis. Ne 5332363}. 02 ha . ““h 01 th hM\.eh figure 4. 13 (a; 4_ Mafieni_ 1‘ ‘ 1‘s :fire 4 3 814:3: 1.) r. gsg‘l‘a ~. 5“\4 I4 '1 “‘< I 'K ' ‘0' 3.- . S‘Wr“- "] a‘ i “(‘1‘ Figure 4.6. Sucrose content as dry matter of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W357B) grown under greenhouse condition for a period of seven weeks after emergence during 2003. Bars represent :L- standard deviations. ..... 155 Figure 4.7. Sucrose content as fresh weight of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W357B) grown under field condition for a period of 20 weeks after emergence during 2003. Bars represent :1: standard deviations. .............. 156 Figure 4.8. Root dry matter content of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W357B) grown under field condition for a period of 20 weeks after emergence during 2003. Bars represent :1: standard deviations. ....................... 157 Figure 4.9. Sucrose content as dry matter of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W357B) grown under field condition for a period of 20 weeks after emergence during 2003. Bars represent :1: standard deviations. .............. 158 Figure 4.10. cDNA-AF LP differential gene expression analysis of 3302 transcript- derived fragments (TDFs). Early-expressed TDFs were present in at least two samples in the first five weeks after emergence (WAE), and not observed successively. Lately- expressed TDFs were only present in at least two samples in the last five WAE, and were not observed before. Transitionally-expressed TDFs were expressed in at least two samples from the 3rd to the 7'” WAE but not during the first or last two weeks of the analysis. Note that y-axis starts at 2150 (2181 TDFs were monomorphic between all the samples). ...................................................................................... 160 Figure 4.11. Cluster analysis of transcript-derived fragments (TDFs) differentially expressed between the l” and 9th week after emergence (WAE). White and black colors of each of the 1121 TDFs per each WAE indicate detected or undetected TDF, respectively. ...................................................................................... 161 Figure 4. 12. Beta vulgaris L. cDNA microarray analysis: Functional classification of differentially expressed genes during active sucrose accumulation. ....................... 184 Figure 4. 13. Arabidopsis thaliana cDNA microarray analysis: Functional classification of differentially expressed genes during active sucrose accumulation. .............. 197 Figure 4.14. Arabidopsis thaliana cDNA microarray analysis: Frequency of differentially expressed transcripts during active sucrose accumulation. .............. 198 xvi Figure 4.15 unlysis: F accumulate Figure 4. It analysis: F I accumulatic Figure 4.15. Arabidopsis thaliana oligonucleotide microarray, non-conventional analysis: Frequency of differentially expressed transcripts during active sucrose accumulation. ............................................................................. 208 Figure 4. l6. Arabidopsis thaliana oligonucleotide microarray, non-conventional analysis: Functional classification of differentially expressed genes during active sucrose accumulation. ...................................................................................... 209 xvii ‘P I 1 ’1. p e) ' — .; (‘1 1.? — v . f . —— " . KEY TO SYMBOLS OR ABBREVIATIONS Symbol or abbreviation Description % DM ......................... Dry matter content expressed as percent of the fresh weight % SucDM .................... Sucrose content expressed as percent of the dry matter % SucFW .................... Sucrose content expressed as percent of the fresh weight AFLP Amplified fragments length polymorphism cDNA Complementary DNA Cleo 90 % Confidence interval ClM Composite interval mapping cM CentiMorgan CPA Comparative phenotypic analysis cRNA Complementary RNA Cy3 Cyanine3dye Cy5 Cyanine5dye DM Dry matter DGE Differential gene expression DNA Deoxyribonucleic acid E ............................... EooRl restriction enzyme E/M ............................ EooRl / Msel restriction enzyme combination EFAEnzymatic fluorogenic assay EST-UTR ..................... Espessed sequence tag - untransleted region FW Fresh weight GGT Graphical genotype in”. Broad-sense heritability HPLC High Performance Liquid Chromatography lM Interval mapping LOD Logarithm ofodds MCB Multiple comparisons with the best treatment mRNA.........................Messenger RNA P ................................ Pstl restriction enzyme P/M ............................ Pstl / Msel restriction enzyme combination pc Primercombination PCR Polymerase chain reaction QTL............................Quantitative trait loci RAPD Random amplified polymorphic DNA RFLP Restriction fragments length polymorphism RNA Ribonucleic acid RY ..............................Rootyield SD..............................Standard deviation SMA Single marker analysis SSH Suppression subtractive hybridization SSR Simple sequence repeat SY Sucrose yield TDF ............................ Transcript-derived fragment WAE Week after emergence x2 Chisquare xviii RA Sucrose p1 translocated p Sucrose conic. ceriuhmraze \ predation. grt bios nth-e515 in 105333011. 8 meahiiism acii ‘ .-. -‘ d.._1‘91~11~1. sync - L?» Van- ' ' ”:1r 0931 If; RATIONALE AND OBJECTIVES OF THE RESEARCH Sucrose plays a pivotal role in plant growth and development since it is the major translocated product of photosynthesis and an important compound for energy storage. Sucrose content of plant tissues represents a balance between source tissues that supply carbohydrate via photosynthesis and sink tissues that require sugars for energy production, growth and storage. Sink tissues play an important role in regulating sucrose biosynthesis in source tissue and transport in the phloem (Chiou and Bush 1998, Paul and Foyer 2001). Sucrose plays also a regulatory role as a general regulator of cellular metabolism acting at the level of gene expression (Koch 1996, Jang et al. 1997, F arrar et al. 2000, Smeekens 2000). Sugar beet (Beta vulgaris L.) is the world’s most cultivated crop for the production of sucrose for human consumption afier sugarcane (Saccharum officinarum L.). Sugar beet is grown on about 5.9 million ha worldwide, mainly in Europe (3.5 million ha), Russia (920,000 ha), North America (550,000 ha), China (400,000 ha), and in other areas around the Mediterranean Sea and the Middle East (FAO 2003, http://www.fao.org). The capability to accumulate sucrose in the root is the most important agronomic trait for the beet sugar industry since the economic return of the crop derives from the total root yield produced combined with the amount of sucrose content in the roots. Sugar beet has become a cultivated crop only in the last two centuries (Winner 1993). During the first 100 years of cultivation, root sucrose content in beets increased from an estimated 6 % ,expressed as percentage of the fresh weight (FW), to over 12 % of the FW through selective breeding of open pollinated varieties. Subsequently, with continued selection for 10.33} 's h} br Die bioel plans are m biosynthesis. exceptional 3. $311151} ( l a be reguhzj "P. Wu ' -| I Am; ‘ 3“ .. F :5A31\S \iC‘V-I ‘ ‘15 for is. Nth ii. - HM.':S hd‘ selection for higher sucrose levels, sucrose has increased to more than 18 % of the FW in today's hybrids. The biochemical pathways and the major enzymes involved in sucrose metabolism in plants are well understood (Huber and Huber 1996, Winter and Huber 2000). Sucrose biosynthesis, transport, and storage in sugar beet likely occur by mechanisms similar to other plants (Kovtun and Daie 1995, Avigad and Dey 1997, Martin et al. 1997, Bush 1999), but specific regulatory and structural metabolic pathways that allow this exceptional accumulation of sucrose in the roots remain to be identified. Savitsky (1940) specifically addressed the question on how many genes are involved in the regulation of sucrose content in sugar beet, analyzing the phenotypic segregation of progenies derived from controlled crosses between beets of different B. vulgaris ssp. vulgaris groups. Savitsky concluded that five or more major genes might explain the differences at the level of sucrose content between sugar beet (16 to 18 % of the FW) and either red beet (8 to 10 % of the FW) or fodder beet (10 to 12 % of the FW). Although sucrose content is a highly heritable quantitative trait in sugar beet with genes acting in additive fashion (Culbertson 1942, Powers 1957, Powers et a1. 1963, Zhao et al. 1997), few studies have been attempted to understand the dynamics and genetic regulation of sucrose accumulation and little is known about quantitative trait loci (QTL) analysis for root sucrose content. The development of new genetic maps to be used for QTL analysis has been very useful in order to identify genes responsible of several important traits in sugar beet (Schafer-Pregl et al. 1999, Nilsson et al. 1999). Schneider et al. (2002) showed that five QTL for root sucrose content as FW were present on five different linkage groups in sugar beet. ‘ increasing 5 number 01‘ n 5 ‘r ,. Well 35 JIM-‘1 trieren gene ‘33 . . :‘T-Ol‘i NS _ a - ‘ sugar beet, while Maughan et a1. (2000) revealed that seven QTL were responsible for increasing sucrose content during seed development in soybean. It appears that the number of major genes regulating sucrose accumulation as storage compound in plant is limited, and that a re-analysis of the early Savitsky’s studies in B. vulgaris with new molecular techniques will yield significant and useful information, which could be directly applicable in breeding programs during selection for higher root sucrose content. Increases in sucrose content and introgression of other important agronomic traits, such as disease resistance, has been obtained through crosses and successive selection of different genotypes of the species B. vulgaris. The subspecies B. vulgaris ssp. vulgaris is divided in four groups with different morphology and sucrose storability potential: Swiss- chard (leaf beet), red beet (garden beet), fodder beet and sugar beet (Frese et al. 2001). All groups of B. vulgaris ssp. vulgaris can be intercrossed without major reproductive barriers, offering the possibility for the introgression of favorable genes from wild genotypes. Broadening the genetic variability of breeding populations by analyzing different sources of germplasm may further increase sucrose content in sugar beet if variability at those sucrose content loci can be identified. However, root sucrose content usually drastically drops after crosses between genetically distant beet lines, such as during introgression of favorable genes, and fast recovery to acceptable sucrose level is not always achievable. Traditionally, root sucrose content has been expressed as percentage of sucrose of the FW (% SucFW) by beet sugar industries and breeding programs and current methods of analysis only refer to sucrose content as FW. However, sucrose content as FW could be more precisely estimated from its two components (i) root dry matter content (% DM) and iii) sucri minim] m: in order to be line SIC CS'im v CA. I I 9mm seas. . 5 11.111.317.16 of: 11 15 “10“?) 1}: ,x. r ‘ . “4153an co .T‘n. $31011 311d \\ 3‘3"???" ' _ “Ned-.31: Q n \ .Z'HTLL ““x-I, ,. t-¢‘.n6n[ l i ‘ ‘Slgil‘lfi'fiaun’ ‘ roli\‘.‘d 77",; v . “\ (In: fi‘r “.13“; 11 10 3.. . 1‘ .2?"- 5 ‘ ‘1.~V LHGIA‘l- . "i“ 3'.“ “.313: ,4“ H.) m: 1" fr I ll '1_._ H' and (ii) sucrose content expressed as percentage of the dry matter (% SucDM). New analytical methods able to independently analyze % DM and % SucDM will be necessary in order to be able to rapidly screen and independently select for these two traits on a high number of individuals to quickly recover high level of root sucrose content as FW. In the majority of beet breeding programs, sucrose content as F W of each individual line are estimated from a representative sample of roots harvested at the end of the growing season. A major limitation of this method of analysis is that it ignores the dynamics of sucrose accumulation during plant developmental stages across the season. It is known that sucrose is concentrated with the innermost six or seven of the 10 to 12 concentric cortical rings of the tap root, around the point of maximum root girth, and it mainly accumulates in vacuoles of parenchymatic cells adjacent to the floematic tissue (Elliott and Weston 1993, Bell et al. 1996). However, the dynamics of sucrose accumulation in the root during the growing season and the genes that are involved during accumulation remain to be identified. In this study I hypothesized that some of the early developmental phases of the crop could play very important roles for the future root storage capability and for the establishment of the crop, and identification and characterization of these critical developmental phases could be used as early selective tools in breeding programs. The objectives of this research were: (i) to develop and evaluate a new sucrose quantification assay able to analyze on a high number of individuals and in a short period of time sucrose content as proportion of the root dry matter; (ii) to develop a new sugar beet genetic map analyzing a segregating population derived in content be iiiil to i. do matter. content as 1 iii} to a: £161 eiopmcr identiii uh; beet derived from a wide B. vulgaris cross, in order to maximize the difference of root sucrose content between individuals; (iii) to identify QTL responsible for root dry matter content and for sucrose content as dry matter, in order to evaluate the relative importance of these two traits on root sucrose content as fresh weight; (iv) to analyze the dynamics of sucrose accumulation during the early root developmental phases and to correlate it with changes of gene expression, in order to identify which metabolic pathways are activated during sucrose accumulation in sugar beet. Avigad. G: biochemi: 304. Bell. C .1.; .\1 plants and Deltlter. 1n Blah. DR. S 187.101, Chiou. 1.1.; B .\ at} A, “a 1' £1.1an .\l. C: Sign 13,... 1’1"” ‘1 C: du‘ LOULIUT S. ‘I'Fijry ‘ 1 ‘ o. Literature cited Avigad, G.; Dey, P.M. Carbohydrate metabolism: storage carbohydrates. In: Plant biochemistry. Dey P.M., Harbome JB. (Eds), Academic Press, New York 1997, 143- 204. Bell, C.I.; Milford, G.F.J.; Leigh, R.A. Sugar beet. In: Photoassimilate distribution in plants and crops: source-sink relationships. Zamski E, Schaffer AA. (Eds), Marcel] Dekker, Inc. 1996, Ch29 691-707. Bush, DR. Sugar transporters in plant biology. Current Opinion in Plant Biol. 1999, 2, 187-191. Chiou, T.J.; Bush, DR. Sucrose is a signal molecule in assimilate partitioning. Proc. Natl. Acad. Sci. USA 1998, 95, 4784-4788. Culbertson, J. O. Inheritance of factors influencing sucrose percentage in Beta vulgaris. .1. Ag. Res. 1942, 64, 153-172. Elliott, M. C.; Weston, G. D. Biology and physiology of the sugar-beet plant In: The sugar beet crop: Science into practice. Cooke DA, Scott RK (Eds), Chapman and Hall, London 1993, 37-66. Farrar, J.; Pollock, C.; Gallagher, J. Sucrose and the integration of metabolism in vascular plants. Plant Science 2000, 154, 1-11. Frese, L.; Desprez, B.; Ziegler, D. Potential of Genetic Resources and breeding strategies for base-broadening in Beta. IPGRI/FAO 2001 Eds HD Cooper, C Spillane and T. Hodgkin 2001, Ch] 7, 295-309. Huber, S.C.; Huber, J.L. Role and regulation of sucrose-phosphate synthase in higher plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 1996, 47, 431-444. Jang, J.C.; Leon, P.; Zhou, L.; Sheen, J. Hexokinase as a sugar sensor in higher plants. Plant Cell 1997, 9, 5-19. Komm. Y.; 1 beet plant and enhcm. Koch K.E. C .l/Ol. Bitil Martin. 1.; 11; mutants in Meaghan. PJ , .,. controtnno 105-111. Nilsson N04 11.; Rodin: \ “Instance 1r. p;‘r" i all. J._\I; F0‘ BUM-'7} 20:51 I- trier-S _ ' ,f L; \"r Kovtun, Y.; Daie, J. End-product control of carbon metabolism in culture-grown sugar- beet plants - molecular and physiological evidence on accelerated leaf development and enhanced gene expression. Plant Physiol. 1995, 108, 1647-1656. Koch, K.E. Carbohydrate-modulated gene expression in plants. A nnu. Rev. Physiol. Plant Mol. Biol. 1996, 4 7, 509-540. Martin, T.; Hellmann, H.; Schmidt, R.; Willmitzer, L.; Frommer, W.B. Identification of mutants in metabolically regulated gene expression. PlantJ. 1997, 11, 53-62. Maughan, P.J.; Saghai Maroof, M.A.; Buss, G.R. Identification of quantitative trait loci controlling sucrose content in soybean (Glycine max). Molecular Breeding 2000, 6, 105-111. Nilsson, N.O.; Hansen, M.; Panagopoulos, A.H.; Tuvesson, S.; Ehlde, M.; Christiansson, M.; Rading, I.M.; Rissler, M.; Kraft, T. QTL analysis of Cercospora leaf spot resistance in sugar beet. Plant Breeding 1999, 118, 327-334. Paul, J.M.; Foyer, H.C. Sink regulation of photosynthesis. Journal of Experimental Botany 2001, 52, 1383-1400. Powers, L. Identification of genetically-superior individuals and the prediction of genetic gains in sugar beet breeding programs. .1. Am. Soc. Sugar Beet Tech. 1957, 9, 408-432. Powers, L.; Schmehl, W.R.; Federer, W.T.; Payne, M.G. Chemical genetic and soil studies involving thirteen characters in sugar beets. J. Am. Soc. Sugar Beet Tech. 1963, 12, 393-448. Savitsky, V.F. Genetics of sugar beets. USDA-National Agricultural Library Translation 5825. Translated from Russian, original source unavailable. 1940, pp. 1 35 Schafer-Pregl, R.; Borchardt, D.C.; Barzen, E.; Glass, C.; Mechelke, W.; Seitzer, J.F.; Salamini, F. Localization of QTL for tolerance to Cercospora beticola on sugar beet linkage groups. Theor. Appl. Genet. 1999, 99, 829-836. Schneider. 1 sucrose CI related m; Smeekens. S P6141”, .‘JlUt, Winner. C. 1 1993.(‘/121". inter. 11.; lli and regular 1120.3; 31...: lUll‘Slb fam 11. [twist it Schneider, K.; Schafer-Pregl, R.; Borchardt, D.C.; Salamini, F. Mapping QTLs for sucrose content, yield and quality in a sugar beet population fingerprinted by EST- related markers. Theor. Appl. Genet. 2002, 104, 1107-1113. Smeekens, S. Sugar-induced signal transduction in plants. Annu. Rev. Plant Physio]. Plant Mol. Biol. 2000, 51, 49-81. Winner, C. History of the crop. In: The sugar beet crop. Cooke DA, Scott RK (eds). 1993, Ch], 1-32 Winter, H.; Huber, S.C. Regulation of sucrose metabolism in higher plants: localization and regulation of activity of key enzymes. Crit. Rev. Plant Sci. 2000, 19, 31-67 Zhao, B.; Mackay, I.J.; Caligari, P.D.S.; Mead, R. The efficiency of between and within full-sib family selection in a recurrent selection program in sugar beet (Beta vulgaris L). Euphytica 1997, 95, 355-359. Trebbi. D. an. Journal oil-1;; CHAPTER 1 Trebbi, D. and McGrath, J. M. F luorometric sucrose evaluation for sugar beet. Journal of Agricultural and Food Chemistry 2004, 52: 6862-6867. le'ORO.“ Sucrose is If in 10“ sucrose radiated spec inrogression. inexpensixe en/ agar beet root t +7 f4 ..e.c-gro\m mot L CHAPTER 1 FLUOROMETRIC SUCROSE EVALUATION FOR SUGAR BEET Abstract Sucrose is the economic product from sugar beet. Disease resistance is often available in low sucrose genotypes and, prior to deploying such novel genes as available into the cultivated spectrum, selection for increased sucrose content is required during introgression. The objective of this work was to evaluate a relatively rapid and inexpensive enzymatic—fluorometric microtiter plate assay for sucrose quantification in sugar beet root dry matter, both for progeny testing in the greenhouse and evaluation of field-grown mother roots. As determined using HPLC, sucrose content in diverse populations of sugar and table beet assayed over various developmental stages ranged from 0.213 to 2.416 mmol g'1 dry matter, and these values were used as references for both refractometry and enzymatic-fluorometric assay. As expected, refractometric analysis generally overestimated sucrose content. Enzymatic-fluorometric analyses were reasonably well correlated with HPLC results for young greenhouse-grown root tissues (R2 = 0.976), and less so with older field-grown roots (R2 = 0.605), for unknown reasons. Enzymatic-fluorometric assays may be best deployed for progeny testing young seedlings. 10 Sucar beet . sweetener in increased from (ll'inner. 1903 eermplasm mil and novel gen: $3011.95 backcrt‘ ‘ l - $163101] {Ur ~"~‘1£1crossjn S- 15 \Q ettzciencies in 111 m generall Nil-’lmetr}. ( Le Introduction Sugar beet (Beta vulgaris L.) has been selected for high sucrose content as a source of sweetener in human diets for the last two centuries, where sucrose in fresh beets increased from ca. 6% (F W) in early selections to >18% (FW) in many modern hybrids (Winner, 1993). Public sugar beet breeding programs generally focus on improving germplasm with resistance to biotic and abiotic stress. Introgression of favorable alleles and novel genes from lower sucrose content wild genotypes, usually accomplished through backcross breeding, invariably reduces sucrose content in elite germplasm. Re- selection for sucrose content, while maintaining characters of interest during backcrossing, is required in the return to agronomic and economic performance, and here efficiencies in the selection process are needed. The generally accepted method for industrial sugar quantification has been polarimetry (Le Docte, 1927; ICUMSA, 1964). Polarimetry can be cumbersome, especially during selection in early generations where seed quantity and quality may be limited, since a relatively large number of beets (ca. 10-15) are needed for adequate juice sample volumes, and measurements can be skewed by optical activity of other sugars, particularly glucose and raffinose (McGinnis, 1971). Refractometry is frequently used to estimate soluble solid content in crude extracts (Pollach et al., 1991), but the lack of sucrose specificity and additional interfering compounds (Mulcock er al., 1985 a) makes refractometry less suitable than polarimetry for determining sucrose content in sugar beets. Both polarimetry and refractometry are available for analyses of unprocessed beets, 11 F;- houeier “ 31 sucrose conte (negatively) environment: has been sug; selection metlt expected to re the return to e1 ”dill-€5.15 1L) beet roots his practiced becau a Li‘hri ~ . ct.ic‘3 Opkldlr it.) ‘16“ 51.13053 liar , .u LC) (\Imln ( NIH-.1 e. 1"»..‘3- ' . . 3“an {\qu i ? L11 If? H ' aiioruar . I 4 rs... 1&5 9“- . b t mlnL‘d *th n S'Jar )- however water content in fresh sugar beet roots (75 — 80 %) has a major influence on sucrose content as well as root yield. Sucrose content in fresh roots has been consistently (negatively) correlated with root yield across elite breeding lines and over multiple environments (Pritchard, 1916; Powers, 1957; Carter, 1987), however this relationship has been suggested to be a pseudo-correlation (Simmonds, 1994). If so, breeding and selection methods geared towards analyses of sucrose and non-sucrose dry matter may be expected to reveal additional opportunities for genetic gains, as well as perhaps assist in the return to elite breeding line status from crosses with wild and unadapted germplasm. Methods to examine sucrose as a proportion or as a total yield of dry matter in sugar beet roots has received some attention (Milford, 1973 and 1976), but is not routinely practiced because an extra processing step is required (e.g. drying the tissues) and sugar factories operate exclusively on fresh post-harvest materials. Among analytical methods to view sucrose as proportion of dry matter (DM), highest sensitivity and specificity is achieved by chromatographic analyses (e.g. High Performance Liquid Chromatography, HPLC) (Martin et al., 2001). Time and labor costs for sample preparation and sequential analyses (ca. 12 min per sample) limit use of HPLC to analyses of relatively small populations (Mulcock et al., 1985 b). Enzymatic assays, based on phosphorylation or hydrolysis of sucrose, have been used to determine dry matter sucrose content in plant tissues (Jones et al., 1977; Birnberg and Brenner, 1984). Their application, particularly in microtiter plate formats (Holmes, 1997; Campbel et al., 1999; Spackman and Cobb, 2001), allows an ability to enhance breeding efficiency both by increasing the number of samples examined, but perhaps more importantly, to routinely assess dry matter sucrose content in sugar beet individuals and populations. Here we report the development, 12 eraluation. and for dry matter 5 553} (EFA’ m. needs to be min field-groun root The purpose i determination ir recumulation prr acceptable mater determining suer breeding pro-gm: «13, .' ' damnation 1 evaluation, and potential limitations of a fluorogenic, multiplate-format enzymatic assay for dry matter sucrose quantification of sugar beet taproots. The enzymatic-fluorogenic assay (EFA) may be useful when high sensitivity is needed or when processing time needs to be minimized, but may be subject to as yet undetermined artifacts when mature, field-grown roots are sampled. The purpose of this work was to (i) ascertain the sensitivity of EF A for sucrose content determination in immature beet roots, which could aid characterization the sucrose accumulation process during plant development and perhaps allow earlier selection for acceptable mature root sucrose content, (ii) assess the feasibility and utility of EFA for determining sucrose content in a wide range of germplasm typically encountered in a beet breeding program, and (iii) assess the practicality of EFA for rapid DM sucrose determination. Ultimately, the later aim will need to address a separate problem area, that of rapidly sampling and dehydrating many individual beet roots, in order to measure and select for sucrose content as a proportion of the total dry matter of the root. 13 Plant materi: Sugar 12661. foe and Hog: .,‘. 3 U.-- m uood These lines ant l£~ in 1... 1.5m State ”lift" ' usrng star. ueetly From {1; c 2'}. ‘7. “ml 1mm I005 h 331329165) 8 Hear extraction Pig-1115 “CR1 hd 2331 , ‘ 4 at “b 1113an i need ‘ 50 at \k'cckh i 39 .3 “\Ele 01.61" :1. l fier- Ma 3. ~ Cm thick It 4.1 “r155, lit 0 . R1711 nun). Fur ’~ 1: {'7}; h .. ' ‘3 10m - 11713 Material and Methods Plant material and growth conditions Sugar beets C869, USH20, and SR96, and table (red) beet W357B (Lewellen, 2004; Coe and Hogaboam, 1971; McGrath, 2003; Goldman, 1996) were greenhouse grown in 0.25 m2 wooden boxes with 15 cm soil depth, 8-16 dark-light cycle, at 15-20 °C, and irrigated daily with fertilization twice a month, for nine weeks after emergence (WAE). These lines and F3 plants derived from the cross C869 x W357B, were also grown at Michigan State University Agronomy Farm, East Lansing, MI, USA during 2002 and 2003 using standard agronomic practices. Samples were obtained from roots collected weekly from the third through ninth WAE from greenhouse-grown plants (48 samples), and from roots harvested at 18th (2003) or 19th (2002) WAE from field-grown plants (94 samples). Sugar extraction Plants were harvested, leaves removed, roots washed and weighed, and 10 g of root tissue was frozen in liquid nitrogen and stored at — 80 °C. Progressively fewer roots were needed at weekly intervals, ranging from 58 roots per accession at 3 weeks to three roots at 9 weeks of age (Table 1). Samples of older roots (18th and 19th WAE) were obtained from a 2 cm thick transverse section at the widest part of the root. For greenhouse grown samples, two replications were done of the complete experiment (e.g. biological replication). For field-grown samples, one section from each root was taken and two samples from this root were analyzed independently (e.g. technical replication). 14 Following th- dryness 1' held then ground It “as resuspenc Inc. FL). plac smpension “a ext-action solu the same soluzn Chromatograp An aliquot l ; wulllg allql H 31723. Following the method of Spackman and Cobb (2001), samples were lyophilized to dryness (held at <1 mTorr for at least 3 h) and reweighed to determine water content, then ground to a fine powder with mortar and pestle. Pulverized dried tissue (100 mg) was resuspended in 4 mL of 80% ethanol in a 5 mL fluted-cap tube (USA Scientifics, Inc., FL), placed horizontally on an orbital shaker (50 rpm) at 40 °C for 16 h, and the suspension was centrifuged at 3000 g for 10 min to obtain the clarified ethanol sugar extraction solution. The clarified ethanol sugar extract was used directly for EFA, and the same solution was dried and resuspended for HPLC and refractometer analyses. Chromatography and refractometry An aliquot (1.5 mL) of the clarified ethanol sugar extract was vacuum dried, the pellet was resuspended in 1.5 mL of high-resistivity water (18 MOhm cm'l), and the solution passed through a 0.2 pm filter (Puradisc 25 TF, Whatman). An aliquot of the water- resuspended sugar extract (0.3 mL) was analyzed with a Rudolph .1157 Automatic Refractometer (Rudolph Research Analytical, NJ), read at 589.3 nm (20 °C). The remaining aliquot of water-resuspended sugar extract (1.2 mL) was used for HPLC analyses with a 6.5 mm x 300 mm steel cartridge Waters Sugar-Pakl carbohydrate column (WAT085188, Waters Co., Milford, MA). The mobile phase was 134 uM Na2Ca EDTA set at constant flow at 0.5 mL min", 90 °C, 12 min run time, and quantified with a Waters 410 Differential Refractometer held at 35 °C, as per manufacturer’s literature (Dorschel, 1984). Concentration standards for sucrose (2.92 - 46.74 mM), glucose and fructose (0.22 - 4.44 mM), raffinose (0.13 - 1.34 mM) and stachyose (0.12 - 1.20 mM) were used to generate standard curves. System control and data management were 15 7‘ 1H)- E111) mailC'l ln summ; glucose [0 I exolution 0f reacléd uith Eggcne. OR) 1 product Nil) .13 6! at,“ 199‘), Oxidase Assa} according to ma. game and T’CSL‘ a!» . same in the sa" \ 4. 26" It'- accomplished using Empower Chromatography Manager Software (Waters Co., Milford, MA). Enzymatic-Fluorogenic Assay In summary, sucrose was hydrolyzed via invertase, followed by conversion of D- glucose to D-gluconolactone by the action of glucose oxidase, with concomitant evolution of H202. Hydrogen peroxide, in the presence of horseradish peroxidase, reacted with Amplex Red (10-acetyl-3,7-dihydroxyphenoxazine, Molecular Probes, Eugene, OR) to generate resorufin (Mohanty et al., 1997), a red-fluorescent oxidation product with absorption and emission maxima of 563 nm and 587 nm, respectively (Zhou et al., 1997). These reagents were included in the Amplex Red Glucose / Glucose Oxidase Assay Kit (A-22189, Molecular Probes, Eugene, OR) and were deployed according to manufacturer’s instructions. The reaction stoichiometry of sucrose-derived glucose and resorufin is 1:], thus quantification of resorufin is directly proportional to glucose in the sample. Specifically, in deep-well microtiter plates with 3 replicates for each sample, 100 uL of the clarified ethanol sugar extract was added to 1.9. mL of reaction buffer (50 mM NaHzPO4, pH 7.5) for native glucose determination. From this, 100 uL was transferred to a second well for sucrose determination, to which was added 100 uL of freshly prepared 10 mg mL'1 invertase (1-4504, Sigma) in 100 mM Na acetate pH 4.5 (Horvath et al., 2002). Plates were sealed with aluminum foil (Microseal “F” Foil, MJ Research, Waltham, MA), incubated at 55 °C for 90 min, and 1 mL of reaction buffer was added to neutralize the pH. Native glucose and sucrose-derived glucose measures were quantified l6 in 96 “ell Labfinechm included in E at) to 5‘9”” {primed as we 1.39554“ sucrose content ( Red reactions. an- All 51:1" ..stieal goal of if 1656 C03 552.} results for tl -~..l... Km uere used a in 96 well-format plates (Fluorotrac600, black, 96-well flat bottom plate, Greiner Labortechnik, Germany) by adding 4 uL of sample to 46 uL of reaction buffer. Also included in each plate were glucose standards (1 , 2, 4, 8, and 16 uM in reaction buffer; 50 uL) to generate calibration curves. Fifty microliters of oxidant solution was then added (prepared as manufacturer’s directions, containing 100 pM Amplex Red, 0.2 U mL'l horseradish peroxidase, and 2 U mL'l glucose oxidase, Molecular Probes), and plates were incubated at room temperature protected from light for 30 min. A Wallace VICTOR2 V plate reader (PerkinElmer, Inc., Wellesley, MA) was used in the stabilized energy mode with D531 excitation/D572 emission filters to measure fluorescence. Native sucrose content was initially calculated by subtracting native glucose readings from sucrose-derived glucose amounts estimated from hydrolyzed sucrose, but later abandoned because of low native glucose concentrations in all tested samples relative to sucrose content (see results) and to minimize the expense of running duplicate Amplex Red reactions, and all results reported here were considered as sucrose-derived glucose. All statistical analyses were performed using JMP version 5.0.] (SAS Institute). The goal of these comparisons was to compare refractometric and enzymatic-fluorometric assay results for their ability to predict HPLC determined sucrose contents, thus HPLC results were used as the baseline for comparison. 17 HPLC anal) ofage showed 11; Table 1.1). groom with ax 01.1.89 mmol g and greatest in s i}; O :h as echcrc, C360 Vs. 18“ 0 l0 Cross. only Siimp $123058 COHIL‘DI Results HPLC analysis of 142 samples derived from 1,046 roots taken from three to 19 weeks of age showed a wide range of sucrose content, from 0.23 to 2.02 mmol g'1 DM (Figure 1.1; Table 1.1). During development, sucrose content was least at the earliest stages of growth, with average of 0.29 mmol g'l DM at 3 weeks, increasing with age to an average of 1.89 mmol g'1 DM at 18 — 19 weeks. Sucrose content was least in table beet W357B and greatest in sugar beet, as expected, although variation among sugar beet was not as high as expected based on their fresh weight sucrose values (ca. 15% for USH20 and C869 vs. 18% for SR96; data not shown). F3 plants derived from sugar beet x table beet cross, only sampled from field-grown individuals at 19 weeks of age, had intermediate sucrose content values, as expected. In all germplasm considered, low amounts of glucose and fructose were consistently detected by HPLC (from 5.5 to 44.4 mol g'1 DM) at the earliest stage of development, and in only 15% of the samples at root maturity (5.5 to 83.3 umol g'1 DM). Raffinose (3.9 to 13.9 umol g'I DM) was only present at root maturity and was observed in all lines. Stachyose was not detected. Enzymatic — Fluorogenic Assay Samples whose carbohydrate content was determined via HPLC were re-tested using an enzymatic-based assay with fluorometric detection of glucose. Briefly, sucrose was hydrolyzed to glucose and fructose using invertase, and then total glucose content was assayed using a resorufin-based detection assay. In a standard two-fold dilution series of glucose (0 — 128 uM), fluorometric signal strength increased over the interval of glucose 18 concentratui was $56“ in mmnfllfiwt Sucrose C samples 110-” alculated to dilution setter of a standard more appropr Diufions l0\\ through satura: concentrations from 0 to 64 uM, but decreased at 128 uM (Figure 1.2). Near linearity was seen in the 0 to 16 uM glucose range (R2 = 0.996), and this was chosen as the target interval for the standard curve for sucrose EFA of sugar beet. Sucrose content of sugar beet exceeded 2.4 mmol g'1 DM in some HPLC analyzed samples from individual roots, and here sucrose ethanol-extract concentrations were calculated to be as high as 60 mM. Dilution of the extracts was thus necessary, and a dilution series (240, 480, 600, 960, 1200, 2400, 3000, 4800, 6000, 9600, and 12000-fold) of a standard 60 mM sucrose solution, replicated four times, was used to identify the more appropriate dilution ranges required to maintain assay linearity (Figure 1.3). Dilutions lower than 3000-fold underestimated sucrose concentration, presumably through saturation of the assay’s enzyme kinetics. Dilution factors of 6000 and above gave equivalent estimates, and 6000-fold dilution of sample extracts was chosen for sugar beet sucrose EF A. lnvertase-hydrolyzed sucrose and glucose standard solutions, each ranging from 1 to 16 M, were tested individually. Both sucrose and glucose showed a high linear correlation (R2 = 0.985; p < 0.0001) between estimated and expected results, with their slopes and intercepts not significantly different from one and zero, respectively. Fructose and raffinose standard solutions, each ranging from 1 to 16 M, were tested individually afier invertase treatment to test their native reactivity in the assay. Fructose and raffinose were not detected across the range of concentrations tested. Comparison of analytical methods Overall, bivariate analyses comparing matched pairs of individual readings (by 19 If:- memod) shO‘ refractometric Rgfractomelr.‘ content b." C5 regression line respecrufil.‘ ll estimates of SL [19.8% - 0.015 1 Correlations of benteen methoc 3.91;}th for FF :1: Differences in Mose greenho Deserted. The St at: or location 1 Ct gianhOLLfie-grmm = 0.976; p < 0,000 3‘“. $1111 TOOLS {R2 a u. ‘ IP‘ 1 While," method) showed that sucrose content determined by HPLC was correlated with both refractometric (R2 = 0.950; p < 0.0001) and EFA (R2 = 0.944; p < 0.0001) (Figure 1.4). Refractometry was relatively imprecise, and overestimated HPLC-determined sucrose content by ca. 0.2 mmol g‘1 (as dry weight), with both slope and intercept of its regression line (y = 0.880x + 0.451) being significantly different from one and zero, respectively (Figure 1.4A). EFA and HPLC sugar determinations showed comparable estimates of sucrose content, and the slope and intercept of its linear regression (y = 0.984x - 0.015) did not significantly differ from one and zero, respectively (Figure 1.4B). Correlations of mean sucrose values (based on variety and WAE, Figure 1.5) compared between methods were slightly higher than individual matched-pairs (R2 = 0.985; p < 0.0001 for EFA). Differences in the magnitude of variation observed with EF A between younger, lower sucrose, greenhouse-grown roots, and mature, higher sucrose, field-grown roots were observed. The source of this uncertainty (e.g. as related to either age, sucrose content, and/or location) could not be estimated from these data. Bivariate statistics for younger, greenhouse-grown roots indicated a high correlation between EF A and HPLC results (R2 = 0.976; p < 0.0001; y = 1.049x — 0.053), but this correlation degenerated for older, field- grown roots (R2 = 0.605; p < 0.0001; y = 1.016x — 0.092) when these groups were considered separately. For field-grown roots, mean dry matter sucrose determined with HPLC was 1.84 mmol g" (31) = 0.181; CI 0.95 = 1.80 — 1.87), whereas from the EFA average of three replications was 1.77 mmol g'l (SD = 0.236; CI 095 = 1.73 — 1.82). For field-grown roots, EFA underestimated HPLC sucrose values by 0.06 mmol g'1 (as dry weight) (so = 0.149; CI 0.95 = 0.03 — 0.09). 20 All field; “as no obVi HPLC and li > 0.95). and espectitel} 1. this progress heterozx'gosit} co. elations. 3: terms 11 = 62. 1 In general. 1 replicates “as p .Jureated less \8 Rip). It = 0.995 ETD-'9‘ " ‘ Ins 01116th s \ 0.881 Rep] \I: "Q'tn- ~‘\ NW sucrtwe "item, . ‘Jug. Erc€hf 5 AL 1 I. I ,1 1 US\ “45 [O . 1 e \p, n All field-grown genotypes demonstrated a level of uncertainty with EFA, thus there was no obvious varietal component, although correlations (all with p < 0.0001) between HPLC and EFA were higher with USH20 and W357B (R2 > 0.97), lower with SR96 (R2 > 0.95), and low or very low with C869 and the F 3 population (R2 = 0.938 and 0.500, respectively). With the exception of USH20, the only commercial hybrid in this dataset, this progression follows an increasing level of genetic diversity (e.g. increased heterozygosity) within the populations. Root color did not appear to influence correlations, as green roots showed as much variability as red roots (n = 83, R2 = 0.928 versus n = 62, R2 = 0.910, respectively). In general, three replications of each EFA were performed, and variability among replicates was present. Bivariate analyses between replicates of greenhouse grown roots indicated less variation (Repl vs. Rep2, R2 = 0.994; Rep 2 vs. Rep3, R2 = 0.997; Repl vs. Rep3, R2 = 0.995; all with p < 0.0001). In contrast, variability in results from replicated samples of field-grown roots was higher (Repl vs. Rep2, R2 = 0.792; Rep 2 vs. Rep3, R2 = 0.881; Repl vs. Rep3, R2 = 0.895; all with p < 0.0001). Increased variability in estimated sucrose content using EFA observed in mature, field-grown roots relative to young, greenhouse-grown roots was not expected, and from measurements to date, no clear trends to explain this difference were evident. 21 Table 1.1. .31 gamphynlu rim \\ Ah SR96 ISHZO 1 C809 '10 O /0 “3‘8 J.- k.) (.1- Lii‘tg'l 19 < in. : t I- _, Q dII‘gr e ‘r-Cfifi 4 .mcr. Table 1.1. Mean dry matter sucrose contents measured with HPLC and EF A in five germplasm lines at seven developmental points. Entry WAEa No. of No. of HPLC SDc HPLC SD EFAd SD EFA SD samples roots/ % mmol/g % mmol/g sampleb SR96 3 2 58+54 10.75 1.19 0.31 0.03 10.30 0.78 0.30 0.02 4 2 27+25 27.10 0.29 0.79 0.01 26.00 0.34 0.76 0.01 5 2 6+5 38.48 0.53 1.12 0.02 37.25 0.92 1.09 0.03 6 2 3 48.07 2.53 1.40 0.07 48.39 0.91 1.41 0.03 7 2 3 57.70 0.95 1.69 0.03 55.86 1.20 1.63 0.04 9 2 3 58.19 2.13 1.70 0.06 58.95 0.74 1.72 0.02 18 10 0.56 68.08 4.50 1.99 0.13 66.42 8.38 1.94 0.24 USH20 3 2 47+45 10.64 2.11 0.31 0.06 10.05 1.06 0.29 0.03 4 2 24+22 23.84 2.42 0.70 0.07 24.39 0.29 0.71 0.01 5 2 10+7 48.36 16.32 1.41 0.48 47.94 16.64 1.40 0.49 6 2 3 55.75 3.70 1.63 0.11 54.57 2.09 1.59 0.06 7 2 3 53.55 1.07 1.56 0.03 58.05 3.04 1.70 0.09 9 2 3 58.07 4.10 1.70 0.12 59.08 6.75 1.73 0.20 18 10 05° 69.12 3.56 2.02 0.10 69.17 3.13 2.02 0.09 C869 3 2 45+42 10.13 0.01 0.30 0.00 8.99 0.48 0.26 0.01 4 2 18+16 30.56 0.70 0.89 0.02 29.74 1.34 0.87 0.04 5 2 6+5 38.18 4.93 1.12 0.14 38.22 3.50 1.12 0.10 6 2 3 49.10 2.11 1.43 0.06 48.42 0.53 1.41 0.02 7 2 3 58.23 0.89 1.70 0.03 64.35 7.12 1.88 0.21 9 2 3 55.32 2.69 1.62 0.08 62.96 6.17 1.84 0.18 18 10 05° 65.10 3.19 1.90 0.09 67.88 6.76 1.98 0.20 W357B 3 2 52+47 8.04 0.92 0.23 0.03 6.50 1.04 0.19 0.03 4 2 27+24 14.83 0.23 0.43 0.01 12.62 1.05 0.37 0.03 5 2 7 35.30 1.50 1.03 0.04 34.60 2.31 1.01 0.07 6 2 5 42.62 2.11 1.25 0.06 40.34 1.06 1.18 0.03 7 2 3 48.44 4.59 1.42 0.13 46.41 5.89 1.36 0.17 9 2 3 62.92 22.18 1.84 0.65 62.95 20.08 1.84 0.59 18 10 0.5b 60.92 1.71 1.78 0.05 57.64 4.39 1.68 0.13 F3pop'l 19 54 6 61.62 7.18 1.80 0.21 58.40 6.84 1.71 0.20 " Weeks after emergence; b Different number of roots were used in each sample; ° Standard deviation; d Enzymatic fluorometric assay; ° two samples were taken per root. 22 Peak h -ight (mV) 4‘ 'd l J ‘- 70.0 60.0 50.0 40.0 30.0 Peak height (mV) 20.0 10.0 0.0 7.0 8.0 9.0 10.0 11.0 12.0 Retention time (min) Figure 1.1. Range of 142 overlaid HPLC chromatograms analyzed for carbohydrate composition. Peaks at 6.94 ($0.02), 7.86 ($0.01), 9.72 (i002), and 11.43 (i007) min represent raffinose, sucrose, glucose, and fructose, respectively. Carbohydrate concentrations were determined from peak areas. 23 I’lm tresccncc (.5 72 nm) 240 293 120 240 _ 200 1 160 3 Q a 120 1 o O e: co 8 2 8° 1 o 2 1.1.. 40 1 0 2 4 6 810121416 0 I 7 T F I I I O 16 32 48 64 80 96 112 128 Glucose concentration (01M) Figure 1.2. Determination of optimal range of glucose concentrations for a standard curve. The inner graph represents the glucose concentration range (0 — 16 11M) selected for the standard curve. 24 l 0 (1 Q‘. 2.22: 22.222.392.10 97,35,427. _..4.....:.Z.r....~ Se 501111100 511:11:1‘ 60- -1 é+ % fi 50 4 ¥§ E .. E .5 40 i ‘5 b r: 3 § 30 e 8 o ’5 :1 m 20 7 “o 3 «1 .§ E 10 ~ 01 T T 1 1 fl 1 0 2000 4000 6000 8000 1 0000 12000 Dilution factor Figure 1.3: Determination of sugar extract dilution optimum from a standard 60 mM sucrose solution. Bars represent i standard deviations of four replications. 25 (111111111pj'hM) Huclutnclrlc suctnsc qlmutiliculinn R L' - 0 tcutu )n ) I4 Ian m inn-trio sucn we: qtmnlil' 'II (HHHUI g"I)M) 00 1'11 A 1’ ‘JIS ' & A 'me”ln I 1 .. ‘1 1L a\- . ”:11 . “‘dt‘ 5 Q ‘1“ ' ‘ qMilt) 3.0 2.5 C .2 3 2.0 i: A E 2 g P 3' '60 ‘3’. E. :5 1.0 Q) E O 8 4: 0.5 Q) I! 0.0~'~-—- .- m. g 2.5 8 g 2.0 U SE ‘8' s 3- o 1.5 Q '7 -:-.° f9: E .9. E 1.0 Ln 6 E 9. E 0.5 0.0 . 7 . . . 0.0 0.5 1.0 1.5 2.0 2.5 HPLC sucrose quantification (mmol g'lDM) Figure 1.4. Comparison of sucrose quantification methods. (A) Correlation between sucrose contents as dry matter (DM) estimated with chromatographic (HPLC) and refractometric analyses (R2 = 0.950; y = 0.8801: + 0.451). (B) Correlation between sucrose contents as dry matter (DM) basis estimated with HPLC and Enzymatic fluorometric assay (R2 = 0944; y = 0.984x - 0.015). Triangles and circles represent young greenhouse-grown and mature field-grown samples, respectively. 26 Sucrnsc ("1"“)ng I)M) Cir... - were I). \ 35‘} {Or eath I] r8136018 1II C l 00 Q I]. C) 000 ’ £111an 2.50 HPLC l - EFA l 2.00 .l. l l. 1 l l I. 1 I l ‘L g 1.50 I I Q 1 - 1 TOD 73' E 1.00 E. 3 2 g 0.50 CD 0,00 __-1ll.-1..ll Willi 3456791834567918345679183456791819 SR96 USH20 C869 W357B F3 Pop‘l Weeks after emergence per entry Figure 1.5. Mean sucrose content comparison between analytical methods (HPLC vs. EFA) for each line at each sampled time point (data from Table 1.1.). Error bars represent Standard Deviations. 27 lliree me trefractometr} emrnatic-tlu. satings of sue; 01‘ individuals. expressed as :1 10111112 plants. 1' 1'ir-m- J - ....drngma:er1 410:2 and rm.» if; 3038151101. Discussion Three methods to determine sucrose content in dry sugar beet root tissues (refractometry, chromatography, and EFA) were compared, specifically to validate the enzymatic-fluorometn'c assay reported here for a microtiter plate format. Time and labor savings of such an assay could be substantial in a breeding program examining thousands of individuals, if validated, and concomitantly allow an estimation of sucrose content expressed as a proportion of total dry matter content. Progeny tests for sucrose content in young plants could assist breeders by obviating field screening in early generation breeding materials, since sucrose accumulation appears to be initiated early in growth and varietal sucrose content may be approximated by as few as ten weeks after emergence (Klotz and Finger, 2002; Trebbi and McGrath, 2003). The relative straightforward nature of the enzymatic-fluorometric assay suggests semi-automation could increase sample throughput. Reasonable sucrose content estimates were obtained using the enzymatic-fluorometric assay within a wide range of sample dilutions (4,800 to 12,000-fold), indicating high sensitivity. Enzymatic digestions were considered a potential source of error. For glucose oxidase, results obtained with different incubation times were uniform within the range of 30 to 80 min (data not shown). For invertase, initial times and temperatures of incubation were chosen to allow the reaction to proceed to near completion, and results suggested this was the case. An additional concern was the low pH optima for invertase activity, which is usually different from that needed during detection of liberated glucose and requires an extra-step for pH adjustment (Bergmeyer and Bernt, 1974). In the assay 28 11616. pH 1% (15 531111520116 Pr“ As expecte correlation W4 sucrose contcn earl) detelopn' )orxig plants tolanmetric an relincse could field-groan r00 tltorometric and here, pH was raised from 4.5 to neutrality during the required dilution of the extract, saving one processing step. As expected, refractometry was not suitable for precise sucrose determination. The correlation was reasonable for a rapid estimation, but it overestimated HPLC-determined sucrose content. Enzymatic-fluorometric and HPLC results were well correlated in the early development of the sugar beet root, which might be important in progeny-testing young plants where sufficient tissue mass has not yet accumulated for reliable polarimetric analyses, or where optically active compounds such as free glucose and raffinose could interfere with polarimetric determinations (Martin et al., 2001). In older field-grown roots with higher sucrose content, the correlation between enzymatic- fluorometric and HPLC results was less convincing. It is unclear exactly the source of higher variability between field-grown enzymatic- fluorometn'c and HPLC results. F ield-grown samples consisted of single roots, where as greenhouse-grown roots were pooled samples, and genetic heterogeneity is prevalent in beet germplasm (McGrath et al., 1999). However, this cannot explain variation within samples extracted from the same source. Variability was relatively high among replicated readings of the same field-grown sample, but did not account for all of the variability observed among field-grown roots, suggesting some inter-root or inter- population influence on the overall variation, however no statistically informative relationships were uncovered. Variability between field-grown roots for the presence or absence of enzyme inhibitors of invertase, glucose oxidase, and horseradish peroxidase might explain these results, as could variability in roots for inhibitors of fluorescence per se (e.g. quenching). The detection of raffinose only in field-grown roots suggests the 29 possibility tl appeared to introduced b; the same as u In summai sucrose deterr. 11.7.. 19.77; Birr and Cobb. 3011 here not peril 9316011211 criteri I‘EL'IICUlCiIl} }OL.‘ 5001201111 11 g 10 1 fromazieq‘igm li‘l ‘ 1 Sucrose FW) 311C591} dorm.» P 1r. A l ‘\ \ possibility that unique biochemicals are produced in these conditions, although raffinose appeared to have little or no effect itself on the detection procedures. Variability introduced by dilution or other technical errors seem unlikely since these operations were the same as used with greenhouse-grown roots. In summary, caution should be applied in using enzymaticofluorometric assays for sucrose determination of field-grown beet roots. Enzymatic-colorimetric assays (Jones 81 al., 1977; Bimberg and Brenner, 1984; Holmes, 1997; Campbel et al., 1999; Spackman and Cobb, 2001) may be less sensitive to the postulated interference seen using EF A, but were not performed here because sensitivity to predict sucrose in young roots as a potential criterion for early selection was a higher priority. Greenhouse-grown beets, particularly young beets (<10 weeks old) appeared less sensitive to unknown influences contributing to variability introduced in the assay. The level of precision afforded by the enzymatic-fluorometric methods was not ideal but the magnitude of the differences (est. 1% sucrose F W) was not as large as to render the method useless for breeding purposes, especially during backcross breeding of unadapted materials with elite breeding lines. The sources of variability among field-grown roots are puzzling, and their experimental determination may reveal growth-environment properties not previously appreciated. Conversely, increased dilution of sample extracts may be one trivial approach to minimize observed uncertainties. 30 Ber-gm?) er. } 1' Eds; ACNE mecrg. P. R. Campbell. J. A; enzymatic as FW.}dxlg’riC. l trier. J. .\'. Sue sugarbeets. J. foe. 0. E ; Hogal Dorsehel. C \ . .‘ ll Highliri ‘ “11.157.” 198 ISA. 4 031mm, 1 1.. A breeding ] 1110ng Homes. E 11' 1'7. - .COU 991.344.11.13- Literature Cited Bergmeyer, H. U.; Bernt, E. Sucrose. In Methods of enzymatic analysis; Bergmeyer H. U. Eds; Academic press: New York, NY, 1974; Vol. 3, 1176-1179. Birnberg, P. R.; Brenner, M. L. A one-step enzymatic assay for sucrose with sucrose phosphorylase. Anal. Biochem. 1984, 142, 556-561. Campbell, J. A.; Hansen, R. W.; Wilson, J. R. Cost-effective colorimetric microtitre plate enzymatic assay for sucrose, glucose and fructose in sugarcane tissue extract. J. Sci. Food Agric. 1999, 79, 232-236. Carter, J. N. Sucrose production as affected by root yield and sucrose concentration of sugarbeets. J. Am. Soc. Sugar Beet Technol. 1987, 24, 14-31. Coe, G. E ; Hogaboam, G. J. Registration of USH20 sugar beet. Crop Sci. 1971, 1, 942. Dorschel, C. Analysis of sugars 1: Retention times on Sugar-Pak I. In Waters Lab Highlight 1984, LAH 0210, 2. Waters Co., 34 Maple Street, Milford, MA, 01757, USA. Goldman, I. L. A list of germplasm releases from the University of Wisconsin table beet breeding program, 1964-1992. HortScience 1996, 31, 880-881. Holmes, E. W. Coupled enzymatic assay for the determination of sucrose. Anal. Biochem. 1997, 244, 103-109. Horvath D. P.; Chao W. S.; Anderson J. V. Molecular analysis of signals controlling dormancy and growth in underground adventitious buds of leafy spurge. Plant Physiol. 2002, 128, 1439-1446. ICUMSA methods of sugar analysis; de Whalley, H. C. S. Eds.; Elsevier Publishing Co.: New York, NY, 1964; p 153. 31 Jones. 31. G. 1 sucrose in l K1012. K. L; sucrose cat. Le Docte. A. DOCIC pr 11cc Lewellen. R, 1 C309CMS 51 during File 5: l g - _ .VLGIMIS‘ R_ 1 Te", 7' '14, 4 . 97,10er . F011 COllln5_ ( hlill‘t‘ird‘ G I: .1 351-257 and; llei‘OCk ‘ .A. . 311311515 MCGfa h. 111.1 \1. R i'.) ._ ‘- . .lLC'TZiIh~ .l .\l I1 .9, 9‘ ‘ . 68-9. .‘11111‘, 11.1,} d. G F J . "#81771 19... ‘r Jones, M. G. K.; Outlaw, W. H. J.; Lowry, O. H. Enzymic assay of 10'7 to 10'” moles of sucrose in plant tissues. Plant Physiol. 1977, 60, 379-383. Klotz, K. L.; Finger, F. L. Contribution of invertase and sucrose synthase isoforms to sucrose catabolism in developing sugarbeet roots. J. Sugar Beet Res. 2002, 39, 1-24. Le Docte, A. Commercial determination of sugar in the beet root using the Sachs-Le Docte process. Int. Sugar J. 1927, 29, 488-492. Lewellen, R. T. Registration of rhizomania resistant, monogerrn populations C869 and C869CMS sugarbeet. Crop Sci. 2004, 44, 357-358. Martin, S. S.; Narum, J. A.; Chambers, K.H. Sugarbeet biochemical quality changes during pile storage. I. Sugars. J. Sugar Beet Res. 2001, 38, 35-53. McGinnis, R. A. Chemistry of the beet and processing materials. In Beet-Sugar Technology, 3lrd edition; McGinnis R.A. Eds; Beet Sugar Development Foundation: Fort Collins, CO, 1971; 25-63. Milford, G. F. J. Sugar concentration in sugar beet: varietal differences and the effect of soil type and planting density on the size of the root cells. Arm. Appl. Biol. 1976, 83, 251-257. Mulcock, A. R; Moore, 8.; Barnes, F .; Hickey, B. The analysis of sugars in beet part III: HPLC analysis. Int. Sugar J. 1985, 87, 203-207. McGrath, J. M. Registration of SR96 and SR97 smooth-root sugarbeet germplasm with high sucrose. Crop Sci. 2003, 43, 2314-2315. McGrath, J. M.; Derrico, C. A.; Yu, Y. Genetic diversity in selected, historical USDA sugarbeet germplasm releases and Beta vulgaris ssp. maritima. Theor. Appl. Genet. 1999, 98, 968-976. Milford, G. F. J. The growth and development of the storage root of sugar beet. Ann. Appl. Biol. 1973, 75, 427-438. 32 110113110? 1' ‘ djh1dr0-‘J. \lulcoclt. A. preparailor. Pollaclt G-1 l quality usin Potters. L. ldet gains in suc. 433. 5 Pitchard F. J. C 01‘sugar beet. Simmonds X. \l' Spackman. V. ‘11 sucrose. cluco~ m. ‘ . ..sflian est sto; ; Mohanty, J. G.; Jaffe, J. S.; Schulman, E. S.; Raible, D. G. A highly sensitive fluorescent micro-assay of H202 release from activated human leukocytes using a dihydroxyphenoxazine derivate. J. Immunol. Methods 1997, 202, 133-141. Mulcock, A. P.; Moore, S.; Barnes, F. The analysis of sugars in beet part 1: Sample preparation. Int. SugarJ. 1985, 87, 172-175. Pollach, G.; Hein, W.; Rtisner, G.; Berninger, H.; Kernchen, W. Assessment of beet quality using a refractometric method. Zuckerindustrie 1991 , 116, 689-700. Powers, L. Identification of genetically-superior individuals and the prediction of genetic gains in sugar beet breeding programs. .1. Am. Soc. Sugar Beet Technol. 1957, 9, 408- 432. Pritchard F. J. Correlation between morphological characters and the saccharine content ofsugar beet. Am. J. Bot. 1916, 3, 361-376. Simmonds N. W. Yield and sugar content in sugar beet. Int. Sugar .1. 1994, 96, 413-416. Spackman, V. M.; Cobb, A. H. An enzyme-based method for the rapid determination of sucrose, glucose and fructose in sugar beet roots and the effects of impact damage and postharvest storage in clamps. J. Sci. Food A gric. 2001, 82, 80-86. Trebbi, D.; McGrath, J. M. Sucrose accumulation during early sugar beet development. Proceedings of the Joint Meeting of the International Institute for Beet Research and the American Society of Sugar Beet Technologists 2003, 267-271. Winner, C. History of the crop. In The sugar beet crop: Science into practice; Cooke, D. A., Scott, R. K. Eds; Chapman and Hall: London, UK, 1993; 1-35. Zhou, M.; Diwu, Z.; Panchuk-Voloshina, N.; Haugland, R. P. A stable nonfluorescent derivative of resorufin for the fluorometric determination of trace hydrogen peroxide: Application in detecting the activity of phagocyte NADPH oxidase and other oxidases. Anal. Biochem. 1997, 253, 162-168. 33 GENETIC‘ Genetic mar" there is not an} Alain/‘1‘. Mouth: Of! AI.\ generated with b to it used for 1 eialuite and con' generating inform “is generated fro: ‘vx ..rss betueen a su 53': ‘1‘ e da high den iii-‘70“ Mk Cl. Pstl .1} let's...‘ * t-GllOI. 311111) er 1 35552:" c B H ‘ ' . ton dtstt‘ftmt iiiLgef" .icr'enct in "e1 CHAPTER 2 GENETIC MAP DEVELOPMENT IN Beta vulgaris L. Abstract Genetic maps have been widely used to characterize important agronomic traits, but there is not any publicly available genetic map for sugar beet (Beta vulgaris L.). The objective of this research was to develop a B. vulgaris genetic map using AF LP markers generated with both EcoRI / MseI and Pstl / Msel restriction enzymes pair combinations, to be used for further quantitative trait loci analysis of root sucrose content and to evaluate and compare the efficiency of EcoRI/Msel- and Pstl/MseI-derived markers in generating informative genetic maps for sugar beet breeding programs. The genetic map was generated from the analysis of a segregating population derived from an intraspecific cross between a sugar beet and a table beet line. The map, spanning a total of 496.2 CM, showed a high density of markers on all 9 chromosomes (1.5 cM average marker inter- distance). Pstl/MseI-derived markers presented a lower proportion of clustering behavior, a higher probability to be linked to other markers, and a lower frequency of segregation distortion with respect to EcoRI/MseI-derived markers, showing an overall high efficiency in generating the genetic map. 34 One of genome “a and root cc Oven and R breeding is r to detect p0: n“ - ‘9. Introduction One of the first attempts to link important agronomic traits to the Beta vulgaris L. genome was based on the analysis of phenotypic markers such as monogermy, hypocotyl and root color, bolting behavior, male fertility, and disease resistance (Keller, 1936; Owen and Ryser, 1942; Savitsky, 1952). However the use of phenotypic markers in plant breeding is restricted by their limited availability. Isozyme markers increased the ability to detect polymorphism between individuals but the low number of markers available was not sufficient to develop genome-wide genetic maps (Smed et al., 1989; Van Geyt et al., 1990; Wagner and Wricke, 1991; Wagner et al., 1992). The advent of molecular markers strongly increased the ability to associate important traits with molecular markers. Restriction Fragment Length Polymorphism (RFLP) markers were the first genetic markers used to construct high density sugar beet genetic maps in which important agricultural traits could be precisely located (Pillen et al., 1992 and 1993; Barzen et al., 1992; Boudry et al., 1994; Heller et al., 1996; Hallden et al., 1996 and 1997). Random Amplified Polymorphic DNA (RAPD) markers were also used for linking phenotypic markers and to integrate existing RFLP genetic maps (Uphoff and Wricke, 1992 and 1995; Barzen et al., 1995; Laporte et al., 1998). Other molecular markers used to further integrate B. vulgaris genetic maps were microsatellites and Amplified Fragment Length Polymorphism (AFLP) markers (Schondelmaier et al., 1996; Barnes et al., 1996; Schumacher et al., 1997; Schafer-Pregl et al., 1999, Rae et al., 2000). Despite the large number of published sugar beet genetic maps, a publicly available B. vulgaris genetic map does still not exist. Attempts were made to unequivocally identify 35 linkage gYOU sugar beet ge used tuo set characterized by different 1 inbred lines u 198.7). Hone and. as pfOpO) chromosome r or the are: '«Y‘d in .\F1P Ci Ends ’7‘! linkage groups with individual chromosomes in order to compare results of different sugar beet genetic maps developed by different laboratories. For this purpose researchers used two sets of sugar beet trisomic series that had been previously developed; one characterized by an heterogeneous genetic background and with trisomic lines identified by different plant morphologies (Butterfass, 1964), and the other set established from inbred lines with trisomic lines identified by karyotype differences (Romagosa, 1986 and 1987). However, some of the Romagosa trisomic lines were lethal (Romagosa, 1987) and, as proposed by Schondelmaier and Jung (1997), I adopted the sugar beet standard chromosomic nomenclature based on the Butterfass (1964) trisomic series. Of the different methodologies used to generate and visualize genetic polymorphisms, AF LP produces the largest number of genetic markers per reaction (Powell et al., 1996). However, a resulting problem of the increase of the number of detectable markers was the possibility that several markers would cluster together, reducing the overall efficiency of the AFLP analysis. EcoRI and Msel are the most common restriction enzymes pair used in AFLP analysis in several species (Vos et al., 1995), but this pair combination tends to produce markers that cluster together in hyper-methylated regions of chromosomal centromers, as opposed to other restriction enzymes pairs such as Pstl and Msel (Young et al., 1999; Vuylsteke et al., 1999; Castiglioni et al., 1999). In sugar beet, clustering is known for RFLP and RAPD markers (Pillen et al., 1992; Nilsson et al., 1997) but little information is known on the clustering behavior of AFLP markers. Furthermore, no information on the usefulness of AFLP markers generated with Pstl and Msel restriction enzymes pair on sugar beet genome has been published to my knowledge. 36 The objec: using AFLP 1 pair combina‘ genetic anal}: beet and (ii Pstl Msel-den breeding prim- The objective of this study was to develop a publicly available sugar beet genetic map using AFLP markers generated with both EcoRI/Msel and Pstl/Msel restriction enzymes pair combinations in order (i) to generate a high-quality genetic framework for further genetic analyses such as quantitative trait loci analysis of root sucrose content in sugar beet, and (ii) to evaluate and compare the efficiency of EcoRI/MseI-derived and Pstl/MseI-derived markers in generating informative genetic maps for sugar beet breeding programs. 37 Plant mater The mfii‘l between the “.3578 (GO background .‘ Line C 869 see, allendelian c sucrose comer 103102} gous characterized l“ AFLP Markers AFLP WOIOCL: '.~';l 1. Tu 0 di ffe Eaters; 11k, . ‘1 537151111 e Pstl ,t/(t, Materials and methods Plant material The mapping population utilized for this study was derived from an intraspecific cross between the diploid sugar beet line C869 (Lewellen, 2004) and the diploid table beet line W357B (Goldman, 1996). Parents were chosen because of their different genetic background and large phenotypic variability (Bosemark, 1971; McGrath et al., 1998). Line C869 segregates for a Mendelian recessive gene conferring male-sterility and carries a Mendelian dominant gene for self-fertility, and it is characterized by medium-high root sucrose content as fresh weight (16 %) and a white, conical shaped root. Line W357B is homozygous dominant for both self-fertility and male-sterility genes, and it is characterized by a low root sucrose content as fresh weight (10 %) and a dark-red, ball shaped root. Crosses were made by hand pollination between a sterile C869 and a fertile W3 578 lines and 128 F 2 plants, obtained from self-pollination of a single fertile F1 plant, were genotyped for genetic map development. AF LP markers AFLP protocol of Vos et al. (1995) was applied with modifications (Myburg et al., 2001). Two different restriction enzyme pair combinations were used to generate AF LP markers: the C-methylation-insensitive EcoRI/Msel (E/M) and the C-methylation- sensitive Pstl/Msel (P/M) combinations. For the E/M pair combination, genomic DNA (300 ng) was double digested at 37 °C for 3 h with EcoRI and Msel restriction enzymes in 30 or. of restriction solution [0.17 U rtL" EcoRI; 0.17 U pL" Msel; 10 mM Tris (pH 38 7.5.1: 10 m Resm'ciion 6 Double 5 phosphorilate adapters. eqU 70 °C to 25 °C [CURI ard adaptor: 1.75 7.51; 11)le .‘ round of PC'R {5‘11 and .lr'xr’ extending into 7.5); 10 mM Mg Acetate; 50 mM K Acetate; 5 mM DTT; 0.05 11g 1.1L" BSA]. Restriction enzymes were deactivated with a treatment at 70 °C for 15 min. Double stranded AF LP adapters were produced from two single stranded non- phosphorilated linkers. To prepare non-phosphorilated double stranded EcoRI and Msel adapters, equimolar amounts of the two complementary linkers, were slowly cooled from 70 °C to 25 °C in a one-hour period, after an initial treatment at 95 °C for 3 min. EcoRI and Msel adapters were ligated in 40 11L of ligation solution [0.35 1.1M EcoRI adaptor; 1.75 11M Msel adaptor; 1 mM ATP; 0.025U 11L" T4 ligase; 10 mM Tris (pH 7.5); 10 mM Mg acetate; 50 mM K acetate] at room temperature overnight. The first round of PCR cycle of amplifications (pre-amplification) was performed with EcoRI (E+1) and Msel (M+1) primers each with one selective nucleotide (adenine or cytosine) extending into the restricted-ligated genomic sequences. One 11L of restricted-ligated solution was used to prepare the 20 1.1L pre-amplification solution [0.5 uM E+1-primer; 0.75 11M M+1-primer; 0.188 mM dNTPs; 2.5 mM MgC12; 2 mM Tris-HCl; 10 mM KCl; 0.01 mM EDTA; 0.1 mM DDT; 0.025 U ttL" Taq polymerase], which was amplified for 20 cycles of 94 °C denaturation (30 s), 56 °C annealing (30 s) and 72 °C extension (60 s), with initial steps at 72 °C for two minutes and 94 °C for 1 min, and last extension period at 72 °C for 10 min. Pre-amplified solutions were diluted five-fold with low-TE buffer [10 mM Tris; 0.1 EDTA; pH 8.0] and stored at 4 °C. The second round of PCR amplifications (selective amplification) was performed using simultaneously (multiplexing) one of each IRD700 and IRD800 fluorescence-labeled EcoRI primers with three selective nucleotides ('RDXOOE+3-primers), a single unlabeled Msel primer also with three selective nucleotides (M+3), and 1 uL of the diluted PCR product from the pre- 39 o amplification. primer: 17 n.\1 31:10:: 3 iii“ F. polymerase] m cycle") annealir denaturation 1, 3f initial denaturati For the P .l/ r digested with Pet atpiification pro Tt‘0r6h in 30 1 7.51;] W MgC conditions. gucw *1 left at 37 mph Ican‘on. pv' . 1 reflect - . Itch 1) - . - 11m amplification. Fifteen microliters of selective amplification solution [17 nM IRD700E+3- primer; 17 nM IRDBOOE+3-primer; 0.5 uM Msel +3-primer; 0.188 mM dNTPs; 2.5 mM MgC12; 2 mM Tris-HCl; 10 mM KCl; 0.01 mM EDTA; 0.1 mM DDT; 0.025 U 11L'l Taq polymerase] were amplified for 13 cycles of 94 °C denaturation (30 s), 65 °C (-0.7 °C cycle") annealing (30 s), and 72 °C extension (90 s), followed by 23 more cycles of 94 °C denaturation (30 s), 56 °C annealing (30 s) and 72 °C extension (90 5, +2 5 cycle’l), with initial denaturation at 94 °C for 30 s, and last extension period at 72 °C for 10 min. For the P/M restriction enzyme pair combination, genomic DNA samples were double digested with Pstl and MseI, with few modification with respect to the E/M digestion and amplification protocol. Genomic DNA (300 ng) was first digested with only Pstl at 37 °C for 6 h in 30 UL of a modified restriction solution [0.17 U 1.1L'l Pstl; 5 mM Tris (pH 7.5); 1 mM MgC12; 5 NaCl; 1 mM DTT; 0.1 pg 11L" BSA] to optimize Pstl restriction conditions. Successively, Msel was added at a 0.17 U 11L" concentration, and the solution was left at 37 °C for an additional 1.5 h. During pre-amplification and selective amplification, PstI+0/MseI+1 and lRDPstI+2/Msel+3 primer combinations were used, respectively. Primer sequences used to generate AF LP markers are presented in Table 2.1. A total of 16 and 20 different primer combinations were analyzed for E/M and P/M restriction enzyme pair combinations, respectively. Seven 11L of stop/loading buffer [formamide 95 “/0 v/v; 10mM EDTA; 0.1 % basic fuchsin; 0.01 % bromophenol blue; pH 9.0] were added to the 15 11L selective amplification product, and DNA was denatured at 96 °C for 5 min, then chilled in ice before gel separation. One microliter of the denatured DNA was analyzed on 4200 L1- COR IR2 automated DNA sequencer (LI-COR, Lincoln, NE) in a denaturing 25 cm, 0.2 40 mm thick. 3 in“ EDT for-1 11. Gel ima‘e scored for l marker non‘ selective nut polymorphic polymorphic .litel restricttt selectite print 7,: anabsis I“:Specntelt. Linkage anah s In order to ir 1.111.er 33R, CFC 3'54“ ‘LL‘ 1,4" . ‘ “#5 (1964; LJ‘ ' are' i A ll 0011 i).- - “w. ~13» mm thick, 7 % acrylamide gel [7 M urea, 1x TBE (89 mM Tris-base; 89 mM boric acid; 2 mM EDTA), 3 mM ammonium per-sulfate, 4.4 mM TEMED] run at 1,500 V, at 45 °C for 4 h. Gel images were collected and saved as image files and amplified bands were visually scored for polymorphism to obtain binary datasets (band presence/absence). AF LP marker nomenclature was derived by the restriction enzyme combination and the selective nucleotides used to generate the marker, and by the base pair length of the polymorphic fragment. As example, the marker ECATMCCA148 was represented by a polymorphic band of 148 nucleotides, generated after genomic digestion with EcoRI and Msel restriction enzymes and selective amplification with EcoRI+CAT and MseI+CCA selective primers. Expected Mendelian segregation ratio for 12:1 and 3:1 were tested by )6 analysis for co-dominant and dominant polymorphic amplified fragments, respectively. Linkage analysis In order to increase the number of markers available to develop the genetic map, 28 RFLP, 20 SSR, 19 EST—UTR, and 7 phenotypic markers previously obtained from the same F2 population were integrated with the 316 newly generated AFLP markers. Sixteen additional SSR markers, kindly tested by KWS (KWS SAAT AG, Einbeck, Germany), were used for chromosome identification of linkage groups accordingly to Butterfass (1964) nomenclature. Linkage analysis was performed with JoinMap 3.0 software (Van Ooijen and Voorrips, 2001). LOD grouping threshold was equal or higher than 4.5. Marker order was calculated using pair-wise data of loci estimated with REC 41 threshold of 0.35 and LOD threshold equal to 3.0. Genetic distances were calculated using the Kosambi function (Kosambi, 1944). Graphical representation of linkage groups was obtained with MapChart 2.1 software (Voorrips, 2002). Distribution of markers over the sugar beet genome (between chromosomes) was analyzed independently for both E/M—derived and P/M-derived AFLP markers, and for all the other markers combined named as non-AFLP markers, using x2 analysis considering as expected number of markers per chromosome the total number of linked markers divided by the gametic number of B. vulgaris (x = 9). Clustering of AFLP markers within each chromosome was analyzed independently for both E/M-derived and P/M-derived markers, using the Poisson-model distribution function: p(y) =(xt/y! we" where p(y) represents the probability mass function that in any defined map interval the number of markers is y, and 1. is the average number of markers in the population per that defined map interval (Young et al., 1999). Clusters of AFLP markers were identified by a sliding 5 cM interval over the linkage groups and counting the number of markers per interval. Markers in regions showing a number of markers per interval equal to or higher than the calculated marker density threshold estimated with Poisson distribution were designed as clustered markers. 42 Table 2.1. AF LP adaptors and primers nucleotide sequences used to generate AF LP markers. (N) represents the number of selective nucleotides used during selective amplifications. Adaptors and primers Sequence EcoRI adaptor: LinkerEl S'CTCGTAGACTGCGTACC‘T . LinkerE2 3 CATCTGACGCATGGTTAAS Msel adaptor: Linkeer S'GACG'ATGAGTCCTGAGy . LinkerM2 3TACTCAGGACTCAT5 Pstl adaptor: LinkerP] . S'CTCGTAGACTGCGTACC3' LinkerPZ 3 ACGTGAGCATCTGACGCA5 EcoRI primer: S'GACTGCGTACCAATTC(NNN)3' 5' 3' Mselprimer. GATGAGTCCTGAGTAA(NNN) Pstl primer: S'CAGTCTACGAGTGCAG(NN)3' 43 Results In order to develop a high-quality sugar beet genetic map, AFLP markers were generated using both EcoRI/Msel (E/M) and Pstl/Msel (P/M) restriction enzymes pair combinations. Comparison of the efficiency of EcoRI/MseI-derived and Pstl/Msel- derived markers in generating informative genetic maps for sugar beet breeding programs was also evaluated. AF LP markers Overall, 36 different restriction enzymes primer combinations (pc) generated 1558 amplified fragments (43.3 bands pc'l), of which 316 (20.3 % = 8.8 bands pc") were polymorphic between parents (Table 2.2). Specifically, the 16 E/M primer combinations yielded from 15 to 79 bands pc", averaging 45.5 bands pc'I (d: 16.9 SD), of which 10.4 bands pc'I (i 7.1 SD) were polymorphic (22.9 %). The 20 P/M primer combinations yielded from 17 to 93 bands pc", averaging 41.5 bands pc'l (:t 19.7 SD), of which 7.5 bands pc'1 (i 4.8 SD) polymorphic (18.0 %). The 316 AF LP polymorphic bands equally derived from the two parents without any significant differences between restriction enzyme pairs used: 147 polymorphisms derived from the sugar beet parent and 162 from the red beet one, while the remaining seven AFLP markers, which all derived from E/M pc ()52 = 4.67, p < 0.05), showed codominant behavior (Table 2.2). The frequency of primer combinations based on the number of polymorphisms generated showed that 58.3 % (21/36) of pc yielded between 5 and 16 polymorphisms pc' I, but a conspicuous 33.3 % (12/36) had 5 4 polymorphism pc'l and only an 8.3 % (3/36) 44 had _>_ 17 polymorphism pc" (Figure. 2.1). The percentage of AT content present in the selective nucleotides was not statistically correlated either with the number of total amplified fragments or with the percentage of polymorphisms (Figure 2.2). Genetic map Of the 406 markers used to construct the genetic map, 331 (81.5 %) were grouped on 9 linkage groups, spanning a total of 496.2 cM (Fig. 2.3). The number of markers and the length of individual linkage groups varied from 26 to 47, and from 36.8 cM to 69.7 cM, respectively (Table 2.3; Figure 2.3). Linkage groups were identified with sugar beet chromosomes by the location of chromosome-specific SSR markers (coded KWS-SSR). The distribution of markers over the sugar beet genome showed that the number of E/M, P/M and non-AFLP derived markers observed per chromosome did not statistically vary with respect to their expected frequencies (Table 2.3). The proportion of E/M-derived unlinked markers (49/ 167 = 29.3 %) was higher (x2 = 16.6, p < 0.001) than those of P/M- derived (17/149 = 11.4 %) and non-AF LP (9/90 = 10.0 %) unlinked markers (Table 2.3). Sixty-three of the 331 linked markers (19.0 %) showed segregation distortion (p<0.05) (Table 2.3; Figure 2.3), and all chromosomes had at least one segregation-distorted marker. However, chromosome V showed a significant (x2 =19.0, p<0.001) higher proportion of segregation-distorted markers (23/31 = 74.2 %) with respect of the proportion of chromosomes IX, 1, VIII and VII, which varied between 20 and 25 %, and of those of chromosomes IV, 11, VI and III, which were lower that 5 % (Table. 2.3; Figure 2.3). Considering the distribution of segregation distortion between classes of linked markers, P/M—derived markers showed a lower, but not significant, proportion of 45 segregation-distorted markers (19/ 132 = 14.4 %), with respect to those of E/M-derived (26/118 = 22.0 %) and non-AFLP (18/81 = 22.2 %) markers. As for linked markers, P/M- derived unlinked markers showed a lower, but not significant, proportion of segregation- distorted markers (7/ 17 = 41.2 %) with respect to those of non-AF LP (5/9 = 55.6 %) and E/M—derived (31/49 = 63.3 %) unlinked markers. Overall, considering linked and unlinked markers combined, P/M-derived markers showed a significant (12 =12.2, p<0.001) lower proportion of segregation-distorted markers (26/ 149 = 17.4 %), with respect to that of E/M-derived markers (57/ 167 = 34.1 %), while non-AFLP markers had intermediate behavior (23/90 = 25.5 %). Considering the classes of markers combined, a significantly (p<0.05) higher proportion of segregation distortion was observed in unlinked (43/75 = 57.3 %) versus linked (63/331 = 19.0 %) markers. Three markers with inter-marker distances higher than 10 cM were excluded from the marker clustering analysis to avoid cluster over-estimation by Poisson distribution. For clustering estimation the total map length was consequently reduced to 445.1 cM and a total of 117 E/M-derived markers (7.5M = 1.314 markers/ 5 cM), and 131 P/M-derived markers (MM = 1.471 markers/ 5 cM) were considered. Using these parameters, Poisson distribution indicated that a density of y = 6 or more markers in a 5 cM interval for each of E/M or P/M markers, represented the lower threshold of the significant deviation (p<0.001) from the expected markers density. E/M markers clustered on six chromosomes (1, 111, V, VI, VIII and 1X), showing 51.7 % (61/118) of E/M markers in clusters (Table 2.3; Figure 2.3). Differently, P/M markers clustered on just two chromosomes (III and IX), showing only 14.4 % (19/ 132) of P/M markers in clusters (Table 2.3; Figure. 2.3). 46 Average inter-marker distances of individual chromosomes varied from 1.05 to 1.95 cM, with overall genetic map average inter-marker distance of 1.50 cM (Table 2.3). Inter-markers distances were not normally distributed (p < 0.001) but 58.1 % of markers had an inter-markers distance smaller than 1 cM (Figure 2.4). 47 Table 2.2. AF LP primer combination analysis. Rare cutters used with Msel (M) in primer combination (pc), refer to EcoRI (E) and Pstl (P), generating E/M and P/M pc, respectively. The three nucleotides following the E and M primer sequences, and the two nucleotides following the P primer sequence, represent selective nucleotides used during selective amplification. Primer combination: Total Polymorphic Polymorphic Polymorphic Polymorphic Polymorphic Rare cutter Msel amplified bands bands sugar beet table beet codominant primer + primer bands (%) parent origin parent origin origin E+ACA M+ACA 58 21 36.2 9 10 2 E+ACA M+ACG 35 7 20.0 5 2 0 E+ACA M+CAT 60 10 16.7 4 4 2 E+ACA M+CCA 79 1 6 20.3 1 3 3 0 E+ACA M+CGG 15 5 33.3 1 4 0 E+ACA M+CTT 58 1 5 25.9 6 8 1 E+ACG M+ACC 27 1 3.7 0 0 1 E+ACT M+ACA 46 3 6.5 2 1 0 E+ACT M+CAG 29 4 13.8 2 2 0 E+ACT M+CTT 54 1 5 27.8 9 6 0 E+AGC M+CAG 58 24 41 .4 8 16 0 E+AGC M+CAT 30 2 6.7 0 2 0 E+CAT M+ACC 56 16 28.6 8 8 0 E+CAT M+CAT 55 14 25.5 8 5 1 E+CAT M+CCA 39 3 7.7 0 3 0 E+CTC M+CAG 29 1 1 37.9 7 4 0 Average E/M total 728 167 22.9 82 78 7 Average E/M per pc 45.5 10.4 P+AC M+CAG 28 7 25.0 2 5 0 P+AC M+CAT 20 1 5.0 1 0 0 P+AC M+CCA 24 6 25.0 2 4 0 P+AC M+CGA 17 2 1 1.8 0 2 0 P+AG M+ACA 48 4 8.3 2 2 0 P+AG M+AGC 37 1 1 29.7 5 6 0 P+AG M+CAG 30 8 26.7 3 5 0 P+AG M+CAT 46 9 19.6 5 4 0 P+AG M+CCA 22 4 18.2 2 2 0 P+AG M+CTT 50 9 18.0 5 4 0 P+CA M+ACA 59 10 16.9 6 4 0 P+CA M+AGC 51 7 13.7 3 4 0 P+CA M+CAG 56 13 23.2 5 8 0 P+CA M+CAT 93 22 23.7 8 14 0 P+CA M+CCA 52 8 15.4 3 5 0 P+CA M+CTT 76 1 1 14.5 6 5 0 P+TC M+CAG 28 2 7.1 0 2 0 P+TC M+CAT 36 8 22.2 3 5 0 P+TC M+CCA 34 3 8.8 1 2 0 P+TC M+CGA 23 4 17.4 3 1 0 Average P/M total 830 149 18.0 65 84 0 Average P/M per pc 41.5 7.5 Total amplified bands 1558 316 20.3 147 162 7 Average bands per pc 43.3 8.8 48 1:1 Pstl/Msel Frequency (# pc) 0 l 2 3 4 5 6 7 8 9 10 ll 12 13 14 15 16 17 18 19 20 21 22 23 24 Polymorphic fragments Figure 2.1. Frequency of primer combinations (pc) per number of polymorphisms generated. 60 _.... .. ”po1ymdrph1c ....................... 50. ...... Q ..... Total ................................... .. ........ ........ Amplified fragments ‘6’ 10 - 0 I I I I I I I I I 30 35 40 45 50 55 60 65 70 AT content in selective nucleotides (%) Figure 2.2. Influence of AT content in selective nucleotides versus the total and polymorphic number of amplified fragments. Bars represent standard deviations (n 2 4). 49 30300030: .256 v 0:0 .56 v .36 v Q 00 :oumwflwqm 00:20:. 5:5 $001.08 08065 A.,...L 0:0 A19 AL .3 330:8 000102 59:50:00: .3815: 000320 00>t0040€$=£ 0:0 w0>€0vr~0€$ao0m mo 20>..0HE 03065 0000553: afiw 0:0 005 .:o:0:£ $3: Efimmox 05 gm: 0000850 20 E 80:82: :02 Ago: flambtsm 3 wEEooom 0:320:38: 0:888:20 0.0.80.3: 80m 00 :08 0:0:00 m.~ 05E: w2<0uawfi . a...” a . o 8.5: . .. oo ..Eaeeeuoee- - - a ESEcE . 2.. A a %fiwfifi , .3“ . i. .. ., \ mwowwgowm- - - _S £00300: -. - Mm“ 385x28» -33. wowwuwde . . mm 98.55.85. 1 . E 820543.: r 5.3. 2~< ..r _ 2mm?! « mm C ce_h 080850.50 > 08308030 >— 05805050 51 «bony NanO—mman— 3.8(58 _wo 0582.350 7. 0hm0<§0_ 00505:me m: 36 v m 80 55. 0050555 5. x0 n 5 8535 80m .5 55:5: 0:05» 05 5 58035 000.0 :5 88—55 585: 5 55:5: _88 05 00 580550 53 0505525 :0: 88—5.: 5 55:5: 580098 0.: ’005055550 50505 88.55 5 £5.05 555505 5m Q U855 5:5 50 8 585: 5: 88—55 5 55:52 .0. £855 550 >5 8 055: 0:0:0w 53650 881055 59:5 Z 2. om._ om; mm; mm; 3.. mo; 3.. E..— _w._ mo; A2805805L8550—52 N53 m.mm Em ..mm wém 0.9m Wow 5% 9: 06m A23 550‘— ow «m m _ - o o - 2 - x. 5:855 02 m0 8 : 2 h _ m N _ _ h 8:820 8:00.205 cow mm _mm 3 3 mm mm _m 3 :0 mm mm 52550 55H 55% mm m w. v m _ - 0 - - - N 55805 535053 oa o _w m m _ o m N. 2 5 o w 502005 555‘ ASH—(T5: 3 _ _ - - . - - m - - 5:856 cm N. 2 m _ - _ o m - - m 5:585 :oumw8wom a: 2 N2 M: o .2 _ _ S om om _ _ N_ 502005 :59 5E - G 2 2 - o c - _ _ - n 55285—0 mm _m cm v v o - w - _ _ N 5052585 555055 no. 3 f _ 2 mm _ _ 2 m _ _ 2 o _ _ 6.53505 .05... 5Q - 8855 as": 02 E> :> _> > >_ E : _ 15% 05855: #0855 0589550 50.55 .5 05—0 Avg: 008555 8 555000 505305 05055050 A58 008 8855 05bo55 v5 ~53-me .Mmm m3: 803 88—55 A5.55.552 A353 EVESBK 5 $33 5,5538% 505 55 0:550 55500: 05 8 55:55 00055 93 5 502233 0:03 88:05 mqm< .955 055: :5 8855 5 5:855 55 £535.05 5550500 $05508 552505 5502 .m.~ 050,—. 53 40-.- ............. . ............................................................. . .................................................... 30¢. ............................................................................................ , ....................... . ............. 20.4. ..... _._. .................................................................................................................... Frequency (# markers) NNMMMMMMMMMMY IIIIIIIIIIIIOIIIIII mo—vamonwoé— mvwenwa w '00 _ ............. ‘ ......................... . ............ _ ........................... _ ...................................................... Cumulative markers frequency (%) Marker inter-distance (cM) Figure 2.4. Marker inter-distances. A. Frequency of markers based on marker inter- distances. Bins sizes are equal to 0.1 cM for marker inter-distances between 0 and 4 cM, and 1 cM for marker inter-distances > 4 cM. B. Cumulative markers frequency based on marker inter-distances. 54 Discussion The average number of E/M-derived AFLP scorable bands (45.5 bands pc") and the percentage of polymorphisms obtained in this study (22.9 % = 10.4 polymorphic bands pc'l) are in agreement with the results of Hansen et a1. (1999). In an exhaustive evaluation of E/M-derived AF LP markers in the genus Beta, the authors reported and average of 44.3 bands pc", of which 27 % polymorphic (l2 polymorphic bands pc'l). In comparison, Schondelmaier et a1. (1996) found an average of 61 amplified bands pc", of which 50 % polymorphic, but the authors only considered four selected E +3/M+3 primer combinations. Using 16 primer combinations of the HindIII/Msel restriction enzymes pair, Schafer-Pregl et al. (1999) also observed an average of 11 polymorphic bands pc". My study demonstrated that a similar number of amplified bands and polymorphism could also be obtained using Pstl/Msel restriction enzymes pairs with P+2/M+3 primer combinations. P+3/M+3 primer combinations were initially tested on the mapping population, but were not used because of the low number of amplified bands and polymorphisms obtained with these primer combinations. Overall, this sugar beet by table beet cross appeared to show an average number of band amplification and polymorphism normally present between beet lines (Hansen et al., 1999), with few primer combinations (3/36 = 5.5 %) yielding high polymorphisms pc'1 and several primer combinations (12/36 = 33.3 %) yielding low polymorphisms pc". Keim et al. (1997) and Hansen et al. (1999) found a strong positive correlation between the AT content of the selective nucleotides of the primers used during selective amplification and the number of amplified and polymorphic bands generated. In my 55 study the AT content in the selective nucleotides did not show any positive correlation either with the total number of amplified bands or with the percentage of polymorphisms. However, a positive trend, particularly between AT content and total amplified bands, was observed. The high variability of the results and the lower number of primer combinations tested in my study with respect to those in Hansen et al. (1999), could explain the lack of correlations between these parameters. In order to evaluate the genome coverage of a genetic map, average marker density (average inter-marker distance) is a better indicator than the total length of the genetic map. Early sugar beet genetic maps, mainly generated by phenotypic, isozymes and RFLP markers, were characterized by maps covering a thousand or more cM with a relatively low number of markers, resulting in very high average inter-markers distances (Pillen et al., 1992; Pillen et al., 1993; Barzen et al. 1995). The integration of AF LP and SSR markers increased marker density but did not increase the length of the genetic maps (Schondelmaier et al., 1996; Schumacher et al., 1997; Rae et al., 2000). Therefore, the average markers inter-distance, with respect of the expected number of linkage groups and the clustering behavior of the markers, would be a good indicator of the genome coverage of a genetic map. In this study, the uniformity and very high density of markers found in every chromosome and the expected number of linkage groups observed, other than the relatively high total length of the genetic map, indicate an overall good genetic map coverage of the Beta vulgaris genome. However, the relatively high proportion of unlinked markers, and the increase of marker density mainly caused by markers clustering, still suggests that a relevant fraction of the genome is not represented by this genetic map. Other good indicators of the quality of the genetic map are that all the 56 KWS-SSR markers were mapped and they identified all nine chromosomes. Also, two phenotypic markers were mapped on the expected chromosomes based on previous literature. Precisely, hypocotyl color was mapped on Butterfass chromosome II and multigermy on Butterfass chromosome IV (Barzen et al., 1995). However, integration of more markers will be of critical importance to incorporate unlinked markers and to increase marker density and genomic coverage of the map. Marker coverage of each chromosome was also uniform. The expected number of markers per chromosome was estimated based on the total number of chromosomes (9), and not based on the relative length of each linkage group (cM), because of the similarity in size of the B. vulgaris karyotype (de Jong et al., 1985; Nakamura et al., 1991). High frequencies of marker segregation distortion (19.0 %) were observed in my study, particularly in E/M-derived markers. Segregation distortion can be caused by selection forces that act directly on the markers or on hidden lethal loci linked to the markers. Distortion segregation in sugar beet is a known phenomenon. Wagner et al. (1992) and Pillen et al. (1992 and 1993) found that more than 15 % of markers showed segregation distortions, which the authors mainly attribute to the selection of nine lethal loci present on six linkage groups. Barzen et al. (1992 and 1995) found that 19.3 % of markers distributed on eight linkage groups showed segregation distortion, and attributed the causes for this high proportion to the presence of lethal loci, to the presence of structurally abnormal chromosomes, and to gametic self-incompatibility (SI), for which four loci have been described in sugar beet (Larsen et al., 1977). The possibility that in my study these same causes might have generated the high degree of marker segregation distortion in the mapping population can not be excluded, and further analysis will be 57 necessary to confirm or deny the relative importance of these factors to generate segregation distortion in the genetic map. However, the cross between the genetically distant sugar beet and red beet could also have increased the occurrence of chromosome rearrangements or created less favorable allelic combinations in the progeny, increasing the frequency of segregation distortion in the mapping population. In this study, AFLP markers, particularly E/M-derived, tended to cluster on the central part of chromosomes. Pillen et a1. (1993) and Nilsson et al. (1997) reported a similar behavior for RFLP and RAPD markers in B. vulgaris genetic maps, but little information is known on clustering tendency of AFLP markers in sugar beet. In soybean, Young et a1. (1999) analyzed the distribution of two types of AFLP markers generated with Msel in combination with rare cutter restriction enzymes that were insensitive (EcoRI) or sensitive (Pstl) to cytosine methylation. The authors observed that 34 % of EcoRI- derived markers clustered around the centromeric regions of the chromosomes, while Pstl-derived markers did not cluster and were evenly distributed throughout the soybean genome. Identical behavior of EcoRI-derived and Pstl-derived markers was observed by Vuylsteke et al. (1999) and Castiglioni et al. (1999) in corn, who also noticed that Pstl was more efficient in generating polymorphisms than EcoRI. Clustering behavior of markers derived from these two different restriction enzyme pair combinations can be explained if the centromeric regions of the genome are characterized by DNA hyper-methylation and with low frequency of recombination, other than by low gene-density (heterochromatic regions), and in these parts of the chromosomes Pstl does not digest the DNA, failing to generate markers. Contrary, Pstl cleaves DNA mainly in euchromatic region, characterized by DNA hypo-methylation and 58 with high frequency of recombination, other than high gene-density (euchromatic regions). Differently, EcoRI generates markers from all over the genome, independently from the level of DNA methylation. My study confirms the clustering behavior of EcoRI-derived markers in sugar beet, as observed in other crops. Clusters of EcoRI- derived markers in the central part of sugar beet chromosomes could be considered as putative centromeric locations. A possible explanation for the existence of three chromosomes without any EcoRI cluster in this study could be if the frequencies of recombination of the centromeric regions could be higher than expected, or the number of markers generated in this study was not sufficient to detect clustering with the Poisson parameters considered. Differently from what was observed in soybean and corn, Pstl- derived markers showed some clustering behavior in sugar beet. The Pstl cluster on chromosome IX co-localized with the EcoRI cluster, and could be caused by a lower than expected level of methylation of the centromeric region. The second Pstl cluster flanks the EcoRI cluster on chromosome III, and could also be caused by hypo-methylation centromeric region, or, more probably due to its distal location, by a lower than expected frequency of recombination of the euchromatic region. High frequencies of recombination in euchromatic regions could also explain the higher inter-markers distances observed between markers mapped on distal regions of the chromosomes respect those inter-markers distances of markers mapped on the central part of the chromosomes. The clustering behavior of AF LP-derived markers determined the reduction of the average inter-markers distances. However, a general increase of average inter-markers distance was mainly caused by the presence of 3 distal markers that increased the overall map length of more than 50 cM. 59 The comparison of the efficiency of E/M—derived and P/M—derived markers in generating informative genetic maps revealed a remarkably higher usefulness of Pstl- derived markers. Even if the average total number of bands and polymorphisms obtained by each primer combination was higher with E/M with respect to P/M restriction enzyme pair, E/M-derived markers showed a significant higher proportion of unlinked markers with respect to P/M-derived markers and, of those markers that were linked, E/M-derived markers showed a very high proportion (51.7 %) of clustered markers with respect to P/M-derived markers (14.4 %). In order to quantify these differences and understand the relative importance of these two sets of markers for genetic map development, two simulated genetic partial-maps were constructed using the same set of markers used for the construction of the genetic map, but eliminating only the P/M-derived markers in the first simulated “E/M-based partial map”, and eliminating only the E/M-derived markers in the second simulated “P/M-based partial map”. Of the 257 markers left (406 - 149 P/M-derived markers) in the first simulated E/M-based partial map, only 162 markers (63.0 %) were mapped, covering a total of 293 cM (59.1 % of original map). Differently, of the 239 markers left (406 - 167 E/M-derived markers) in the second simulated P/M- based partial map, 180 markers (75.3 %) were mapped, covering a total of 354 cM (71.3 % of original map). Overall, from the complete set of markers used to generate the genetic map a smaller number of E/M-derived markers, with also a lower efficiency to cover the entire genome, were used to develop the genetic map with respect to the P/M- derived markers. Further advantages of P/M-derived markers were that they showed a significant lower proportion of segregation distortion with respect to E/M-derived markers and that they are believed to represent gene-rich regions of the genome. 60 W (i) A new publicly available Beta vulgaris L. genetic map based on AF LP markers was developed. The genetic map covered a total of 4962 CM in length and showed a high density of markers on all 9 chromosomes, with average marker inter-distance of 1.5 cM; (ii) Pstl/MseI-derived AF LP markers presented a lower proportions of clustering, an higher proportion of linked markers, and a lower segregation distortion with respect to the EcoRI/MseI-derived markers, and were very efficient in generating the genetic map. Overall, the high marker density, quality of the map and uniformity of the genome coverage make this B. vulgaris genetic map a suitable starting framework for further genetic analyses, such as QTL analysis, map-based cloning, and integration of genetic and physical maps. 61 Literature cited Barnes, 8.; Massaro, G.; Lafebvre, M.; Kuiper, M.; Verstege, E. A combined RFLP and AFLP genetic map for sugar beet. Proceedings of the 59th IIRB Congress, 1996, Brussels (B) 1996, 555-560. Barzen, E.; Mechelke, W.; Ritter, E.; Schulte-Kappert, E.; Salamini, F. An extended map of the sugar beet genome containing RFLP and RAPD loci. Theor. Appl. Genet. 1995, 90, 189-193. Barzen, E.; Mechelke, W.; Ritter, E.; Seitzer, J. F.; Salamini, F. RFLP markers for sugar beet breeding: chromosomal linkage maps and location of major genes for rhizomania resistance, monogermy and hypocotyl colour. The Plant Journal 1992, 2, 601-611. Bosemark, N. O.; Bormotov, V. E. Chromosome morphology in a homozygous line of sugar beet. Hereditas 1971 , 69, 205-212. Boudry, P.; Wieber, R.; Saumitou-laprade, P.; Pillen, K.; van Dijk, H.; Jung, C. Identification of RFLP markers closely linked to the bolting gene B and their significance for the study of the annual habit in beets (Beta vulgaris L.). Theor. Appl. Genet. 1994, 88, 852-858. Butterfass, T. Die chloroplastenzahlen in verschiedenartigen zellen trisomer zuckerruben (Beta vulagaris L.). Z. Bot. 1964, 52, 46-77. Castiglioni, P.; Ajmone-Marsn, P.; van Wijk, R.; Motto, M. AFLP markers in a molecular linkage map of maize: codominant scoring and linkage roup distribution. Theor. Appl. Genet. 1999, 99, 425-431. de Jong, J. H.; Speckmann, G. J.; De Bock, T. S. M.; Van Voorst, A. Monosomic additions with resistance to beet cyst nematode obtained from hybrids of Beta vulgaris and wild Beta species of the selction Patellares. II. Comparative analysisof the alien chromosomes. Z. Pflanzenzuecht 1985, 95, 84-94. Goldman, I. L. A list of germplasm releases from the University of Wisconsin Table Beet Breeding Program, 1964-1992. HortScience 1996, 31, 880-881. 62 Hallden, C.; Hjerdin, A.; Rading, I. M.; Sall, T.; Fridlundh, B.; Johannisdottir, G.; Tuvesson, S.; Akesson, C.; Nilsson, N. O. A high density RF LP linkage map of sugar beet. Genome 1996, 39, 634-645. Hallden, C.; Sall, T.; Olsson, K.; Nilsson, N. O.; Hjerdin, A. The use of bulk segregant analysis to accumulate RAPD markers ner a locus for beet cyst nematode resistance in Beta vulgaris. Plant Breeding 1997, 116, 18-22. Hansen, H.; Kraft, T.; Christiansson, M.; Nilsson, N. 0. Evaluation of AFLP in Beta. Theor. Appl. Genet. 1999, 98, 845-852. Heller, R, Schondelmaier, J.; Steinrucken, G, Jung, C. Genetic localization of four genes for nematode (Heterodera schachtii Schm. ) resistance in sugar beet (Beta vulgaris L. ). Theor. Appl. Genet. 1996, 92, 991-.997 Keim, P.; J. M.Schupp; S. E. Travis; K. Clayton; T. Zhu; L. Shi; A. Ferreira; D. M. Webb. A high-density soybean genetic map based on AFLP markers. Crop Sci. 1997, 37, 537-543. Keller, W. Inheritance of some major colour types in beets. J. Agric. Res. 1936, 52, 27- 38. Kosambi, D. D. The estimation of map distances from recombination values. Ann. Eugen. 1944, 12,172-175. Laporte, V.; Merdinoglu, D.; Saumintou-Laprade, P.; Butterlin, G.; Vemet, P.; Cuguen, J. Identification and mapping of RAPD and RFLP markers linked to a fertility estores gene for a new source of cytoplasmatic male sterility in Beta vulgaris ssp. maritima. T hear. Appl. Genet. 1998, 96, 989-996. Larsen, K. Self-incompatibility in B. vulgarisL. 1. Four gametophytic, complementary S- loci in sugar beet. Hereditas 1977, 85, 227-248. Lewellen, R. T. Registration of rhizomania resistant, monogerm populations C869 and C869CMS sugarbeet. Crop Science 2004, 44, 357-358. 63 McGrath, J. M.; C. A. Derrico; Y. Yu; and R. A. Kitchen. Field evaluation of emergence I: Replicated trails of a range of sugar beet and related germplasm. Sugar Beet Res. 1998, E3-E5. Myburg, A. A.; Remington, D. L.; O'Malley, D. M.; Sederoff, R. R.; Whetten, R. W. High-throughput AF LP analysis using infrared dye-labeled primers and an automated DNA sequencer. BioTechniques 2001, 30, 348-357. Nakamura, C.; Skaracis, G. N.; Romagosa, I. Cytogenetics and breeding in sugar beet. In Chromosome engineering in plants: genetics, breeding, evolution; T. T. and P. K. Gupta, Eds.; Elsevier: Amsterdam, 1991; pp 295-313. Nilsson, N. O.; Hallden, C.; Hansen, M.; Hjerdin, A.; Sall, T. Comparing the distribution of RAPD and RFLP markers in a high density linkage map 03 sugar beet. Genome 1997, 40, 644-651. Owen, F. W.; Ryser, G. K. Some Mendelian characters in Beta vulgaris L. and linkages observed in the Y-R-B group. J. Agric. Res. 1942, 65, 153-171. Pillen, K.; Steinrucken, G.; Herrmann, R. G.; Jung, C. An extended linkage map of sugar beet (Beta vulgaris L.) including nine putative lethal genes and the restorer gene X. Plant Breeding 1993, 111, 265-272. Pillen, K.; Steinrucken, G.; Wricke, G.; Herrmann, R. G.; Jung, C. A linkage map of sugar beet (Beta vulgaris L.). T heor. Appl. Genet. 1992, 84, 129-135. Powell, W.; Morgante, M.; Andre, C.; Hanafey, M.; Vogel, J.; Tingey, S.; Rafalsky, A. The comparison of RFLP, RAPD, AFLP and SSR (mirosatellite) markers for germplasm analysis. Mol. Breed. 1996, 2, 225-238. Rae, S. J.; Aldam, C.; Dominguez, 1.; Hoebrechts, M.; Barnes, S. R.; Edwards, K. J. Development and incorporation of microsatellite markers into the linkage map of sugar beet (Beta vulgaris spp.). Theor. Appl. Genet. 2000, 100, 1240-1248. 64 Romagosa, 1.; Haecker, R. J.; Tsuchiya, T.; Lasa, J. M. Primary trisomic in sugar beet. 1. Isolation and morphological characterization. Crop Science 1986, 26, 243-249. Romagosa, I.; Cistue, L.; Tsuchiya, T.; Lasa, J. M.; Hecker, R. J. Primary trisomic in sugar beet. II. Cytological inentification. Crop Science 1987, 27, 435-439. Savitsky, V. A genetic study of monogerm and multigerm characters in beets. Proc. Am. Soc. Sugar Beet T echnol. 1952, 7, 331-338. Schafer-Pregl, R.; Borchardt, D. C.; Barzen, E.; Glass, C.; Mechelke, W.; Seitzer, J. F .; Salamini, F. Localization of QTLs for resistance to Cercospora bieticola on sugar beet linkage groups. T heor. Appl. Genet. 1999, 99, 829-836. Schondelmaier, J .; Jung, C. Chromosomal assignment of the nine linkage groups of sugar beet (Beta vulgaris L.). T heor. Appl. Genet. 1997, 95, 590-596. Schondelmaier, J .; Steinrucken, G.; Jung, C. Integration of AFLP markers into a linkage map of sugar beet (Beta vulgaris L.). Plant Breeding 1996, 115, 23 1-23 7. Schumacher, K.; Schondelmaier, J.; Barzen, E.; Steinrucken, G.; Borchardt, D. C.; Salamini, P.; Jung, G; Weber, W. E. The combination of different linkage maps in sugar beet (Beta vulgaris L.) into one map. Plant Breeding 1997, 116, 23-38. Smed, E.; Van Geyt, J. P. C.; Oleo, M. Genetical control and linkage relationships of isozyme markers in sugar beet (B. vulgaris L.). Theor. Appl. Genet. 1989, 78, 97-104. Uphoff, H.; Wricke, G. Random amplified polymorphic DNA (RAPD) markers in sugar beet (Beta vulgaris L.): Mapping the genes for nematode resistance and hypocotyl colour. Plant Breeding 1992, 109, 168-171. Uphoff, H.; Wricke, G. A genetic map of sugar beet (Beta vulgaris L.) based on RAPD markers. Plant Breeding 1995, 114, 355-357. Van Geyt, J. P. C.; Smed, E.; Oleo, M. Genetical control and linkage relationships of isozyme markers in sugar beet (Beta vulgaris L.). Theor. Appl. Genet. 1990, 80, 593- 601. 65 Van Ooijen, J. W. and R. E. Voorrips JoinMap 3.0, Software for the calculation of genetic linkage maps. Plant Research International, Wageningen, the Netherlands. 2001 Voorrips, R. E. MapChart: software for the graphical presentation of linkage maps and QTLs. J. Hered. 2002, 93, (1), 77-78. Vos, P.; Hogers, R.; Bleeker, M.; Reijans, M.; van de Lee, T.; Homes, M.; Frijters, A.; Pot, J.; Peleman, J.; Kuiper, M.; Zabeau, M. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 1995, 23, 4407-4414. Vuylsteke, M.; Mank, R.; Antonise, R.; Bastiaans, E.; Senior, M. L.; Stuber, C. W.; Melchinger, A. E.; Lubberstedt, T.; Xia, X. C.; Stam, P.; Zabeau, M.; Kuiper, M. Two high-density AFLP linkage maps of Zea mays L.:analysis of distribution of AFLP markers. T heor. Appl. Genet. 1999, 99, 921-935. Wagner, H.; Weber, W. E.; Wricke, G. Estimating linkage relationship of isozyme markers and morphological markers in sugar beet (Beta vulgaris L.) including families with distorted segregations. Plant Breeding 1992, 108, 89-96. Wagner, H.; Wricke, G. Genetical control of five isozyme systems of sugar beet (Beta vulgaris L.). Plant Breeding 1991, 107, 124-130. Young, W. P.; Schupp, J. M.; Keim, P. DNA methylation and AFLP marker distribution in the soybean genome. Theor. Appl. Genet. 1999, 99, 785-790. 66 CHAPTER 3 ANALYSIS OF LOCI AFFECTING ROOT SUCROSE CONTENT IN Beta vulgaris L. Abstract Quantitative trait loci (QTL) analysis of compounded traits like root sucrose content as fresh weight (% SucF W) could be more efficiently analyzed considering the two component traits that influence % SucF W (i) root dry matter content (% DM) and (ii) sucrose content as dry matter (% SucDM), separately. The objective of this study was to analyze QTL locations for % DM and % SucDM relative to the location of QTL for % SucFW to understand the relative importance of the two components on the %SucF W. Progeny-tests of segregating populations derived from an intraspecific cross between a sugar beet and a table beet lines were field evaluated for two consecutive years, and QTL analysis was performed on all three traits considered. A total of 13 genomic regions were identified on seven chromosomes that might play a role in sucrose content in sugar beet. Of these regions, seven QTL directly influenced % SucF W, while five QTL influenced % DM and nine QTL influenced % SucDM. All the QTL influencing % SucF W co- localized with QTL for % DM or for % SucDM or for both traits. In populations derived by wide genetic crosses % DM and % SucDM were not correlated and both traits influenced root sucrose content. 67 Introduction Since the beginning of sugar beet cultivation two centuries ago, root sucrose content has been expressed as percentage of sucrose of the root fresh weight (% SucF W) by beet sugar industries and breeding programs. Sucrose yield (SY), which represents the amount of sucrose produced per unit area (i.e. t ha'l) and it is the most important trait for beet sugar production, is ideally calculated considering the amount of roots produced per unit area (root yield, RY), and the % SucFW measured by polarimetry using the equation: SY = RY x (% SucFW/ 100) [3.1] The simplicity of the measurement and its direct relation with SY has made % SucFW the standard way to express root sucrose content in sugar beet. The % SucFW can be more precisely evaluated from its two components: (i) root dry matter content (% DM) and (ii) root sucrose content expressed as percentage of the root dry matter (% SucDM) using the equation: % SucFW = (% DM x % SucDM) / 100 [3.2] The determination of these two components would increase precision in the estimate of sucrose expressed on the fresh weight basis, and would reveal potential differences in dry matter content and dry matter partitioning between breeding lines, which would remain hidden with the sole polarimetric analysis, increasing the understanding of factors influencing sucrose accumulation in sugar beet. Surprisingly, during screening and selection of superior lines in public and private sugar beet breeding programs, % SucF W is the only selection parameter used, and its components % DM and % SucDM are rarely considered even if their direct and positive relationship with SY appears evident. 68 Reasons not to consider and not to routinely use % DM and % SucDM as selective parameters in beet breeding programs could only be explained if these two traits are reciprocally correlated, or if one of them is invariable in the breeding populations. If this were the case, the direct use of % SucFW as selective parameters, whose analysis is cheaper and less time consuming, would be justified, but little information is known on the correlation and variability of % DM and % SucDM within Beta vulgaris, and only few studies indirectly addressed this problem (Bergen, 1967; Wise, 1979 and 1980; Milford et al., 1988). However, % DM and % SucDM are believed to be correlated or one of the two traits is believed to be invariable, because of the presumed existence of another (negative) correlation between RY and % SucFW parameters. If this negative correlation does exist, by the equations [3.1] and [3.2] (i) it would imply the existence of a direct correlation between % DM and % SucDM because their product should always be high or low in lines with low or high RY, respectively, or (ii) it would imply the invariability of one of the two % DM or % SucDM and the simultaneous negative correlation of the other trait with RY, such as a negative correlation between % DM and RY and constant % SucDM, or a negative correlation between % SucDM and RY and constant % DM between breeding populations. Several studies indicated a strong negative correlation between RY and % SucFW, resulting in an overall limited variation in SY between sugar beet varieties (Pritchard 1916; Powers, 1957; Hecker, 1967; Oldemeyer, 1975; Doney et al., 1981; Geidel et al., 2000). Bergen (1967) found that differences between a high RY/low % SucF W and a low RY/high % SucF W sugar beet lines were mainly determined by differences in root % DM rather than by difl‘erences in % SucDM, which always remained at similar levels. Carter (1987), analyzing several 69 lines in different years and locations, obtained similar results confirming the negative correlation between % DM and RY and constancy of % SucDM. However, Simmonds (1994) suggested that this inverse relation between RY and % SucF W, which he referred as ‘pseudo-correlation’, was a consequence of simultaneous traits selection resulting in a sort of statistical artifact instead of a true biological relationship, and could only exist in populations derived by crosses between elite breeding lines, where a higher % SucDM was already present and fixed (homozygosity) in the parental lines. Indeed, this pseudo- correlation between non correlated traits is typical of plant material in the late stage of breeding programs, when only superior lines are intercrossed and selected (Gravois et al., 1991). Furthermore, Schneider et al. (2002) observed a weak positive correlation between RY and % SucFW. Therefore, if RY and % SucFW are not always correlated, % DM and % SucDM might not be correlated themselves or one of the two might not be always invariable in breeding populations, particularly in those populations derived from crosses between different genetic backgrounds. These findings reopen the questions on which selective parameters for sucrose content should be used in sugar beet breeding programs. Selection based on % SucFW could be the most efficient in genetically homogeneous populations in the late stages of breeding programs, where changes in root sucrose content are mainly influenced by the relative ratio between root water/dry matter content (changes in % DM). However, if % DM and % SucDM are not correlated and are variables in genetically heterogeneous populations, the use of both % DM and % SucDM as selective parameters for sucrose content in the early stages of breeding programs would be the most efficient way to select for superior individuals. Particularly after the introgression of favorable genes, such as biotic and 70 abiotic resistance genes, from wild parental lines into elite genetic backgrounds, the recovery of high level of root sucrose content, which usually drops drastically in the first generations after the wide crosses, could be accelerated selecting for both % DM and % SucDM independently. Furthermore, analyzing the two % SucFW components independently would reduce the risk to discard individuals with putative favorable genes. Two individual beets with extremes but opposite values of % DM and °/o SucDM, i.e. one root with very high % DM (favorable gene) but very low % SucDM and the other root with very low % DM but very high % SucDM (favorable gene), will probably have the same % SucFW (similar to the average % SucFW of the population from which the two roots derived) and would not be selected. Other than for practical reasons, an approach aimed to subdivide the compounded trait % SucFW in to its components would also help to better understand the physiology of sucrose accumulation and sucrose content in sugar beet. Biochemical pathways involved in water/dry matter relationship (% DM) in root tissue could be independent from other pathways involved in dry matter partitioning (% SucDM). Knowledge of the biochemistry of those pathways could reveal opportunity to increase sucrose content in sugar beet. The objectives of this study were (i) to evaluate the correlation and variability of % DM and % SucDM in individuals derived from a wide genetic cross in order to evaluate the most appropriate selective parameters during breeding programs, and (ii) to analyze QTL location for % DM, % SucDM and % SucF W in order to estimate the total number of genomic regions influencing % SucFW, to understand the relative importance of % 71 DM, % SucDM on final sucrose content, and to ultimately use marker information to facilitate selection. 72 Material and methods Plant material and experimental design The genetic map used for quantitative trait loci (QTL) analysis was derived from an intraspecific cross between the sugar beet line C869 and the table beet line W357B (see Chapter 2 of this Dissertation). Progeny tests of the mapping population replicated over two years were used to evaluate QTL map location. F 3 populations derived from self- pollination of the F2 plants used to construct the genetic map were field-grown in 2002 and 2003, and average F3 family values were used to analyze the quantitative traits. On 1 May 22‘”, 2002, 54 F3 families were planted in triplicate single-row in a randomized complete block design at the Michigan State University Agronomy Farm, East Lansing, MI. Two sugar beet lines (USH20 and SR96) were also included in the experimental design and used as reference lines. Samples were harvested on October 22'”, 2002 and immediately processed for analysis. Due to the insufficiency of seeds, on May 215‘, 2003, 13 F3 families were planted in triplicate and 14 families in duplicate single-row in a randomized block design, respectively, while the remaining 27 F3 families were planted as un-replicated plots at the Michigan State University Agronomy Farm, East Lansing, MI. Four beet lines (USH20, SR96, and the two parental lines C869 and W357 of the genetic map) were included in the experimental design (three replications) and used as reference lines. Samples were harvested on October 13‘“, 2003, and immediately processed for analysis. Rows were 6.0 m long, row inter-distance was 0.76 m and common agronomic practices were employed in 2002 and 2003. 73 Traits evaluation Plot length (cm), plot weight (kg), and total number of roots per plot were measured, and from these values, average root weight (kg root'l), average root yield (RY, t ha"), and average sucrose yield (SY, t ha") of each line were estimated. Per each year, average RY and SY of the lines were expressed as comparison with the averages (index = 100) of the root and sucrose yields combined as the average of the USH20 and SR96 sugar beet lines. RY (t ha") was estimated for each line using the following equation: RY = [plot m / (plot 1, x plot id)] x 10 where plot wt is the average root plot weight (kg), plot n the average plot length (m) and plot id was the row inter-distance (m). SY (t ha") was estimated for each line using equation [3.1]: SY = RY x (% SucFW/ 100) Traits measured for QTL analysis were (i) root dry matter content and (ii) root sucrose content as dry matter, while (iii) root sucrose content as fresh weight was calculated from the two previous traits. For each plot, 6 representative roots were sawn longitudinally, root pulp was homogenized and 25 g of sample were frozen in liquid N2 and lyophilized. Root dry matter content (% DM) was measured by comparison of weights before (Wtbeforc) and after (Wtafier) lyophilization of the sample, using the following equation: % DM = (wtmer / Wtbcfore) x 100 Root sucrose content as dry matter (% SucDM) was measured via Enzymatic- Fluorometric Assay (Trebbi and McGrath, 2004; also Chapter 1 of this Dissertation), and 74 root sucrose content as fresh weight (% SucF W) was estimated using the following equation: % SucFW = (% DM x % SucDM)/ 100 Per each year, ranks of the F3 families for % DM, % SucDM, and % SucFW traits were performed following the multiple comparisons with the best treatment (MCB) procedure (Hsu, 1984). Heritability estimates Heritability was estimated for % DM, % SucDM, and % SucFW traits on a plot basis (Fehr, 1987), using variances components calculated from the expected mean square values estimated with PROC MIXED, Type 3 Analysis of Variance (ANOVA) (SAS Institute,l994). Broad-sense heritability (th) was estimated combining the two years data obtained in single location using the following equation: h23=ozg/[ozg+(ozgy/r)+(oze/ry)] where 0'23 is the genetic variance, 623,. is the variance due to genotype x year interaction, 02¢ is the variance due to experimental error, r is the number of replications, and y is the number of years (Johnson et al., 1955). QTL analysis QTL analysis was performed for % DM, % SucDM, and % SucFW traits, analyzing the 2002 and 2003 data separately. QTL analysis was performed using Windows QTL Cartographer 2.0. Single marker analysis (SMA), interval mapping (1M), and composite interval mapping (CIM) were carried out and results were compared. IM was run with 75 model 3 and walking interval of 0.5 cM, while CIM was run with model 6, walking interval of 0.5 CM, window size of 10 cM, and with 10 markers as background control detected through forward and backward stepwise regression. The likelihood value for the identification of QTL was expressed as likelihood of the odds (LOD) score calculated as the logm of the Ll/Lo ratio, where L1 and Lo are the maximized probabilities of the model with and without the putative QTL, respectively. Per each trait considered, the LOD score thresholds for significance of QTL detection was estimated after 1000 permutations (Churchill and Doerge, 1994). QTL estimate positions corresponded to the markers interval containing the highest LOD score value. Phenotypic variances explained by each QTL were estimated from the partial correlation coefficients (R2) obtained from the Windows QTL Cartographer output. Map and QTL graphical representation was obtained with MapChart 2.1 software (Voorrips, 2002) and Graphical Genoype (GGT) sofiware (van Berloo, 1999). QTL information derived from SMA, IM and CIM analyses were combined, and the influence of each locus in determining trait differences and the effects of the parental allele combinations were estimated by comparative phenotypic analysis (CPA) of each putative QTL. F3 families genotypes were analyzed with GGT software and were grouped based on their genotypes in the QTL regions. For each QTL, average trait value of the group of families showing homozygosity for the sugar beet-derived allele was compared with the average trait value of the group of families showing homozygosity for the table beet-derived allele. Significance of trait differences between homozygous parental alleles combinations for each putative QTL was estimated with a pooled t-test. 76 Results Traits evaluation Similarity of traits of the two reference sugar beet lines USH20 and SR96 between the two years allowed comparing F3 data of both years with the C869 and W357B parental performances in 2003 (Tables 3.1 and 3.2). Averages of F3 families for % DM, % SucDM and % SucF W were similar among the two years of investigations and showed continuous variations of intermediate values with respect to the two parental lines. For all traits considered, individual F3 family transgressive segregation was rarely observed and no F3 family showed transgressive segregation in both years. Averages of F3 families for RY and SY also showed the continuous variations of intermediate values with respect to the two parental lines, which is typical of quantitative traits (Figures 3.1 and 3.2). The analysis of correlation between % DM and % SucDM in the F3 population did not show any significant correlation between the two traits in either year (Figure. 3.3). During 2002, the coefficient of determination between the two traits was equal to R2 = 0.006, with % DM varying fi'om 15.23 to 24.89, and % SucDM varying from 49.24 to 67.43 %. Similarly, in 2003 the coefficient of determination between the two traits was equal to R2 = 0.080, with % DM varying from 15.77 to 23.26, and % SucDM varying from 48.41 to 65.57 %. These values were intermediate between those of the two parental lines in 2003. Both root dry matter content and sucrose content as dry matter showed a positive correlation with the overall sucrose content as fresh weight (Figure 3.4). Coefficient of determinations of R2 = 0.666 and R2 = 0.733 were observed between % DM and % 77 SucFW in 2002 and 2003 respectively, while coefficient of determinations of R2 = 0.263 and R‘2 = 0.543 were observed between % SucDM and % SucFW in 2002 and 2003 respectively (Figure 3.4). Broad-sense heritabilities estimated from family-plot average values were equal to 13., = 0.44 for % DM, 1%., = 0.68 for % SucDM, and 1223 = 0.50 for % SucFW (Table 3.3). The higher influence of the environmental component on % DM with respect to % SucDM, which determined the lowest value of heritability for % DM, was also evident from the comparison traits between years, showing a lower coefficient of determinations for % DM (R2 = 0.028) with respect to that for % SucDM (R2 = 0.208) (Figure 3.5). Different meteorological conditions were observed between years, with 2002 being on average more than 1 °C warmer for both maximum and minimum temperatures and wetter during the growing season with respect to 2003 (Figure 3.6). QTL analysis Single marker analysis (SMA) for root dry matter content (% DM) showed that of the 331 markers of the genetic map (see chapter 2 of this dissertation for details) a total of 69 markers were correlated (p < 0.05) with % DM. Twenty-three markers on chromosomes 1, III, and IV were correlated with % DM in both years (Figure 3.7). SMA for sucrose content as dry matter (% SucDM) showed a total of 52 markers correlated with the trait, of which 19 markers on chromosomes VII were correlated with “/0 SucDM in both years (Figure 3.7). SMA for sucrose content as fresh weight (% SucF W) showed a total of 89 markers correlated with the trait, of which nineteen markers on chromosomes I and VII were correlated with % SucFW in both years (Figure 3.7). 78 SMA analysis for co-localization of significant markers between different traits showed that markers on the genomic region from 41.0 to 47.2 cM on chromosome I were both correlated with % DM and with % SucF W, in both years. Similarly, sbcDOlO RF LP marker on chromosome VII (16.1 cM) was correlated with % SucDM and with % SucFW, in both years. With SMA, no marker was simultaneously correlated either with % DM and % SucDM or with all three traits in both years. LOD score threshold values estimated after 1000 permutations were equal to 4.19, 4.21 and 4.12 for % DM, % SucDM, and % SucF W, respectively. CIM detected 3, 5 and 3 significant QTL for % DM, % SucDM, and % SucF W, respectively (Table 3.4; Figure 3.7), while IM did not detect any QTL for either trait considered (Figure 3.7). With CIM, two QTL for % DM were detected in 2002 (on V and VII) and one in 2003 (on III). The QTL at 31.3 cM on V (LOD 6.36) and at 30.0 cM on VII (LOD 5.10) respectively explained 27.5 and 22.9 % of the phenotypic variance in 2002, while the locus at 44.7 cM on III (LOD 6.57) only explained 8.8 % of the phenotypic variance in 2003 (Tab. 3.4). Three QTL for % SucDM were detected in 2002 (on IV, VII and IX) and two in 2003 (on V and VII) with CIM. The QTL at 4.0 cM on IV (LOD 4.23), at 5.1 cM on VII (LOD 4.56), and at 2.5 cM on IX (LOD 5.58) explained 20.0, 18.30, and 21.3 °/o of the phenotypic variance in 2002, respectively. The QTL at 43.0 cM on V (LOD 7.45) and at 16.2 cM on VII (LOD 4.80) explained 33.0 and 27.6 % of the phenotypic variance in 2003, respectively (Table 3.4). Three QTL for % SucFW were detected in 2003 (on IV and two on VII) and none in 2002. The QTL at 15.0 cM on IV (LOD 4.57), at 4.1 cM on VII (LOD 5.32) and at 16.2 cM on VII (LOD 7.07) explained 22.3, 39.0 and 31.3 % of the phenotypic variance in 2003, respectively (Table 3.4). 79 CIM analysis for co-localization of significant QTL between different traits showed that loci on the chromosomic regions from 4.1 to 5.1 cM (in different years) and at 16.2 cM (in 2003) on chromosome VII were correlated with both % SucDM and % SucFW. No QTL were simultaneously observed either for % DM and % SucFW% or for % DM and % SucDM in either year. The combination of SMA and CIM information revealed a total of 13 genomic regions, spread over seven chromosomes, that could be involved in at least one of the traits under investigation (Table 3.5; Figure 3.8). Comparative phenotypic analysis (CPA) of groups of F3 families, selected with GGT software and homozygous for each parental allele in the putative QTL intervals, confirmed (p < 0.1) allelic effects on the traits in 12 instances out of the 13 genomic regions considered (Table 3.5). Overall, five QTL were observed for % DM, nine for % SucDM, and seven for % SucF W. Co- localization of loci was observed between % DM and % SucFW (three), between % SucDM and % SucF W (three) and between all three traits (one), while only one locus was specific for % DM, and five QTL were specific for % SucDM. There was not any locus only observed for % SucFW. The F3 families CPA also revealed the parental allele combination influences of the traits (Table 3.5). Considering as ‘favorable’ those allele combinations that increased % SucFW, of the five QTL for % DM, four favorable alleles derived from the sugar beet parent while the fifth allele combination (interval 30.0 to 31.4 cM on V) did not show any significant difference between parents (T able 3.5). In contrast, of the nine QTL for % SucDM, five favorable alleles derived fiom the sugar beet parent, one did not show any significant difference between parents (interval 28.2 to 30.1 cM on VII), and three loci showed favorable allelic combinations derived from the 80 table beet parent (intervals 4.0 to 5.0 cM on IV, 11.5 to 17.8 cM on IV, and 0 to 8.6 cM on IX) (Table 3.5). For the seven QTL for % SucFW, six favorable alleles derived from the sugar beet parent, and one did not show any significant difference between parents (interval 11.5 to 17.8 cM on IV) (Table 3.5). 81 9n ...N ...N «.2 a... 3 mm. an... a n..m 8N0 3 we... 5.... m «2......— .... c...N N.N ..n. o... 3 N... 3.... 3 «a... .nNn 3.. mmN 0.....N m N.N»... ..m Mn. .... 9N. ..v 3 or... 3.3 3 2.. New 3 an... 3.... m :Nhtmn. o... ..N. ...o 5.... v... .. mm. 3.. . a NY... 3.8 3.. an... 3.3. m 2.3m... a... Md. 0.. 0.... ..v a K... we... 3 mm. NwNn 3.. .N.N N.....N m $4.... 0... S. N.N N.N. N... 3 ..N. 2.... 3 an... 2.3 3 3.... as... m e..Nim... rt. 3. ...N N.N. rev 3 2... NN.... 3 v..." 2.3. 3 8.. 8.... m «3...... ..m ...w. ...N m... N... 3 3.... N. .... 3 SN a. 6n 3 N. .N N...». m 3.1»... .... ...w. v. «..N. n... 3 2.... 2.6 3 3.. 8.3 3 8... 5...... n Fin... o... ...N e... m. . n... 3 8... 3.... 3.. SN .3». 3 A2.... 2 .2 n 8.9. n. ...N. n... v... n... 3 ..N... 2.6 3.. S... 3.? 3 an... mud. n coin... ..N 5. v. c... N... 3 m... on... 3.. ...N 3.3 3 EN 2.... n «2...... ..N ..N. a. Q: N... a on. ..v... 3 8d 2.3 3.. we... 9....N m «NYE 9n ...N. N.. ...o N... 3 mm. v. .o 3 N... $.nn 3 ca... 8.: m 37m... ...n N... ...N n... N.m 3.. 2... N»... 3 SN 3.: 3.. 8.. an...N m 3.»... .... c... nN Qn ..m a ..N... NYN. 3.. m...m ”man 3.. c...” .a...N m «mum... n. m...— h. n6 .4... 3 c. .N as... 3 e. .v nN.nn 3.. 3.... Nnd. m 2.5.. ...v m... nN ed 3. 3 on... .8... 3 ...N ENn 3 3... N... m «3.4.... ..m Em. w. a.» ..v 3 .5... 2.... 3.. new when 3 av... ...6. m «whim... 9n ...3 ...N mt... o... 3 on... N. .o. 3 and o. .3 3 N.... 8.2 m «a. .-m... m... 0: N.N ed ...M 3.. 8... 8... 3 and v. .9». 3.. on. .5... m main... o.. ...N. ...N ...o On p v. ... NE. . a 3.... 3.8 3.. 3.... med. m «3...... ...N ...v. a... n... m... 3 ...N ..N... 3 ... N «new 3 on. .NN. m «NN..m... Wm m... ...N N.” ...v 3 an... 0...... 3 ...N 3.3 3 3... mod. n aim... c... Q». a. v... v... 3 ..N... co... 3 3.. on... 3 no... MNd. m «35... ..N ...m. n .N No ...v 3 on... co.» 3 an... ..v. . n 3 we... ...N. m Sim“. ..n ...a. N.N ...o ..v .. 3N .NN. 3.. N...N 3.3. a N... on..N m 5...... n.N ..N. a... ...w m... 3 SN Ne... 3 3N 3.3 3 2... Nwd. m whim... N.. ...N ...N n... v... 3 3.. 3.: 3 0.... ...N... 3 ms... and. m «3...... ea .3. .E. .2. . .03 5mg. :m 38.. am a: a: mo: am 3.25 .\. mu: am 2:65 .\. :02 am 2: .\. a3. 2.: .m:o..ao....o. .6 55:9: 0 day. 58080.. :8 n 5: 30.3.5: 585:3... u Gm .683 05 5.3 .o .mon 05 5.3 85.0 3:20....5 .5853? 6: 3.8.5:. .3... .503 05 5.3 00:805.. €35.53 0: 3.8.5:. .3. 48.. 05 5.3 3:98....5 Emu—mam... 0: 3.8.5:. b. 22.3 ASK. an... 9:538: 803.. .5839: .mon 05 5.3 :8..:a:.oo 03.5.8 05 5.3 508.83.. 9.3 ..E. :03 .0 33:5. 23.53 5.0 8.. 50m: 203 33:53 gmosm e\e. ...w.o3 58.... m: .:o.:oo 08.68 5:: 9.03m e\e. 3.3:. .3. m: .:o.:oo 8983 .95 .\e. .032: be .03. .N....N 80> I .8. Fame... n... . . .m 0.9:... 82 lau 1‘ ! an. a... c a s 2 ._ c: : a an... 3. he 3 NE. :..E m 3:-.. .37. Am... :2. is I 9:..~ 5% 2—3-03 C-aA-U- I _ A zflhg—dm : any: “F’a zflh .9 I0 . . . . 2...). AF! . . E... 2.42 27 3:57 .0 I: . Cm 3661 :7 2.... a..:.............. .n n.s.... ad ...w. ...m N.nN dd 8.. 8d. d». 3.65 Nd... nnNN m cam: h... nd. mé ddN n... as... 56. N... 8.3 3... dvéN m 595 a... 3.... he...” .vén 2... .dd. own—26 n... n.N m.m n . . ..N .v: 3 9.... 5nd 3 5N6 VNdv >5 5.... Nc.cN m «dNA... vé 5.... ad ad. :6 3 N... ...d. 35 n . .5 nmdm 3 and 3.5. m an. .-.N... ..N m... ...N No. n6 5 .5... and. 5 on... 3.8 >5 v. .N .o..N m can... 9m cd. 5.. N... né 5 8.. ON. . 5 an... 8N5 3 do... 5...... m «SA... n.. ...N ...N Na. mi >> nNd med. 3 v. .. mddn >5 5EN and. m 8.9. 5... ...5. ... 5.... N6 5 3.. do. . >5 N¢.m mmdn >5 m . .m dad. m no.4”... ad ...w. bd v.“ . m... >> .d... N... . 5 5.... 3.5 B mN.N 59.”. m SA... n6 fin. 5.9 dd. Ni 3 Nu... m5d >5 N. .v 3.3 3 3.... mad. m 87m... 0m c.5— 5.N ...v. V... 5 and 3.. . 35 dad 3.3 >5 3.. 2d. m ..N .-m... N.. m5. «.5 5.5. v... 3 .5... and. 3 3N .dén 3 3.. and. m nN .-N... n. ...5. dd in. mi 5 N». 3.2 >5 2.5 dadn 5 8d on..N m .N.-m... ..N m... 5... N. . . v6 5 ”N.N nN.v. >5 8.... ..N...m 5 .v... dde m .572 5... ...5. ..N nd ed 3 no... 5N6. >> SN «”3 >5 ”..N and. m ndné... N.v ...m. N.N 9n. tv >5 3.. h. ... B NnN .53 >5 add 8.8 m 57m... N.. m.n. 5.. N6. mé >> nNd and. >5 SN w . Kn >> MN... and. m «a. ...m... 9N cd. Nd m6. n6 3 3... add >5 3... 55.3 3 mN... and. m 32-2 N. . «:5. ON «.6. N... 3 Z... 8.... 3 3.5 N...Nn 3 ”...N 3.”. m 5a-”. Wm cd E. ...w 9n 5 5 . .. %N. 3 N5. .vén 5 No... d5.mN m «No.9. n.N md. N... ...N. N6 5 N5. 3.. . 5 me... . ..No >5 nmé v. d. m ..Nonn. ad «...N a. ..v. m... 3 de 3d >5 med 3.5m 3 Nu... med. m mm..m... dd ...N. 5.5 dd QM 3 rd... 3.... 3 N56 .13 3 «...N 5.». m .. ..m". ..N n. . N d... d... mi 3 ..N... 9...: >> .n... Nnén >5 3.: 3d. n «whom... 5.. 98 ad ...m. n6 3 ..N. .dd 3 an. N99 3 ...N 3a». m 3%... n.N 6.... w... d... 5.9 .5 mod and. 3 . ..c no.9. >5 5. .n 86N m gum... n. ...o. ..N ad dd >> m. .. 8... 5 use 9.5 3 N9... .nd. m 3. .-m... 3... .9: .8. .2. 392: 598. Gm v.83. Gm .0... .0... mos. Gm 295 x. mus. Gm SEQ—m 0} m0: Gm 2A. «a do... 8.... 93.80% . . .m 0.5:... 83 [111-211]... 11.11:. I 1..: .1555... :c: u 1: .c:.:=.5.... .tEEEm I CO .923 .2.— 5..3 ..0 E05 0... 5:3 .050 00:000....5 .cauc.:m.m 00.00.00. .020. $5002: .5003 0:. 5.3 000000.85 0:00....030 0: 93005:. .3. .605 05. 5.3 00:000....5 3005030 0: $800.02.. .5. 20:3 433. 50:. 0030000.... «$.03: 000.500.: .000 05 5.5.» 08.000500 0.03.55 05 5.3 00500.00... n03 «.00 p.000 ..0 $50.00! 6.9233 4&0 00.— 00%: 0095 00.65.30 «gloat X.» 5303 chuck .3 00.3000 30.5.3 0:6 2200.5 o\... 0030:. bv an 000.000 0.309% .320 XL 00:3: .00 .00! 6.00m 30> I 2...... >000ch ...~ Nth 033s. m n ...N N.N «.2 ed 3 I... 2.... a 3... 3.? 3 cc... 2.... m «2...... - c «N - . .N m6 3 - «a... 3 . 8.: - 3.2 . NW”. N m n v. v... N v. ..v 3 N... Nod 3 3... .03. 3 on... NE... m “N.N... .. o n - od 5. a . «9n. n - 3.9 A. . «SN . «...N... . c N. . ..« «.. - «ad. 3 .. 8.3 . No.2 . 3...... . c . v: v: a - 3:: n .. Nn.No A. . ..N.nN . «5...... . o 2 - «.: m n . 3.: n . n63 . .«.«. . in". « N o 2 . c «.n. u v Na... 9.... 3 RN «0.3 ..«d oNdN N «219. N o n o. v N ...2 h m 3.... 3...: a .5... No.3 3 ..«d N. .2 N Rum... . c ..N - n.« o m n . 3.2 a . oN.No - oodN . Sum... n o n «N m n «.2 . n mm o 8.: a ..«6 San m«.o «.22 N 52?. . c v - ...N N . .. 8.: a - o. . .w 3 - no.2 . «8-..... . o ..N - n... a v - «ed. 3 - .hfin . 2 .2 . «NYM... m n o m. n v «d. e v 3 c «...N. p «..N 3.3 3.... 3.2 m 3.-.”. . c « .. ...« v N - «...N. .. «3% a - «...N . 3...”... m « c o. a 0 ad m n n. 2. c 2N. .. 3... 3.8 8... 3.2 N «2.-.... . c .. - 9.2 N m - 9.: 3 . 3.3 A. - 8..N . «N-N... N . o « e v ...o o m 3.. N... a a... Nv.«n 3 ma... «9.... m «3.4”... n... ma N.N v6 a... 3 No... .a.« 3 .ad .09». 3 .N... :6. N «34.... . 3.. .. ..« ...N - «N... a - .osn - 9:2 . «2 .-N... . c... - v... «.. 3 - 3... A. - N«...« 3 - .«.n. . 2A". .. ...N. Wm 06 c... . 3.. . n .. 90:3 - 3.3 N “3...“... . c m. - o6. . v - Nnd. . 3.3 . .n.«. . 32-..... - c.« . «.m. o.— 3 . on... 3 .. 2.3 3 . 3.... . 2.-.”... .. o6. . a... QN . and. n .. 8.3 3 - 3.«. . «3-2 N m o ..N « c 00. n v 3.. «ad. 3 on... 3.3 $6 5.2 N 3...... . n N 5N o m N... b n ««.o «N... a 3N than n... ..«.«. m 2.-.“... - o h. - as « N 3 . «n... 3 . 3.2 3 . 8.2 . 3-..... « N 0 cm .2 0.2 « n 3 «n... cod 3 ..N.N Nn.vn 3 NM... «0.... N «$4... 8.. .9: .E. .2. £303 5?». Gm 903. Gm 8... .0... mos. Om 3.25 .\o mos. am Znoam .\. mos. Om 2n. .\. do... 0:... .80..ao..n.o.. ..o 0055.: .I. .95. 63080.. no: u .0: 33330.. 53.05% H Gm .683 05 5.3 8 60.. 05 5.3 55.0 00:20.5... .5552... «0.3.5:. .550. «53.8 4.53 05 5.3 3.8.—ob... Enema..." o: 3.8.5:. .3. $2. 05 5.3 00:95.“...0 2:85.56 on 8.8.5:. b. 22.3 ...«2 .85 23.08.... 502. 2.2535 62. 05 5.3 sagas 0322. 05 a? 8.89:2. 83 as 58 .o 32.5. 3.35 do .5. BS 225 .3258 92.2% § its fie. 3 88:8 8825 .05 .3.ng «\ov .388 5 ma 5:088 0888 ASE a... .8508 be .03. .mgN 30> I 68 manna... m... .N.... 0.5.... 84 .2 (u ~ ~ .u .. I: 3; 2.3 .3 c... ~. I . I G i C h V I Rx:— I!!! am .2 w :4: . . ‘ 3.: ~ 4.. 3.; 3.7: .2: ~ boa 2.3.93 5::0. Cm 58v— CZ 2:.— .O_& 3U: 27. 3103?. .x :02 Ana zfiuuamv. : at: 2.... we: Jr. in! 05.. N}. $33-...My. m. 43d w. M1 QN ...N 3.3 véN 3.3 3... N93. 33.. «3.33 we... onNN 3 cNmmD 3.. ...N 3.3 ...N 3.3 V3... 3...... 33.. 8.30 2.6 3v.3N 3 03.3 ..3 3.3. ...N N... 3... .v... 8.3 o... 3.3.3 3... 3.3. 3 E333 ..N ...N 3.3 ...NN 3.3 33... 33.3. .v. 3.63 33... $.3N 3 330 ...N. ac... 3.3 3.33 N3.. 33.3. 03.96 m... c: v: u: a: v: v: . 8N3... ad ad Nd ...N. N... ..N... 8.. . a and N33 3... 33.8 N a... .-3... 3.3 ca 3.3 3: N3 3.. 33... a 3.3 3.8 3.. 3.3. 3 «33.3... ...c 3.3 3.. ..N. 3.3 «N. K... a 2.3 2N3 33... 8.3. N 33.3... oé ...N 3.3 «...N ...3 a nu... NwN. a ..N... 8. .3 a on... 8. .N n 3.3... ad 3.. N.N 3.... ..v 33... .3... a and $.33 s. .c on... N 33.3... 3.3 ...N N.3 QNN 3.3 3... 8.. . A. 3.. 3.3.8 03... 33.3. 3 30.3... «xv ...N. 3.3 3.2 3.3 3 3.. 336. a g.— 3333 3 o. .N 3.... n 32.3... v: v: v: .E v: v: . 32-3... . c6. .. c3. N.v a . NaN. .. - 2.3 A. . nadN . 32.3... - c.3N - ad. 3... - 3.... a - 33.3 3 - 3.6. . .N.-3... 3.3 ...N. ad 3 .o 3.3 .3... ..N. . a 3.3 33.33 3.... N33. 3 .c . .3... - Q. . - 3 .3 ...N - 3.. . 3 - 3.33 a - 8.8 . «33-3... 3.3 c... N.. 3.3 3.3. 9.. 3.... a 3N3 33.33 3 3... .33.... 3 573... .. ch - ...N .3 - 2... . a - a... .3 - 3.3. . a... .-m... . 3.3 - v.33 3.3 . 3.... a - 3N3 3 . N..a. . «32.3”. - c... .. ..3 3.3 3 - h . d. 3 . 33.3 - 3.3. . 33-3... . ...N . ...N a: 3 - 33.3 3 . 33.33 3 . 2.... . «3-3... 3.3 3.3 . v... N... 3.3 NN... v. .N. a 33.. 33.3 ...c 33.3. N 3N3... 3N OVN 9v ...N «.3 we... 3.. . a «...N 3N.No 33 .o 3.3. m 33.3“. - c6 - 3.3 o... a . 33.N. A. - 53.33 - 3...: . ...me 3N 3.3N 3.. 3.». 3.3 $6 3.... A. we... ..N..o 3 3o... 3.... 3 ~33... - o...N .. ...N 3.3 3 .. nos 3 .. 3.3.. 3 - 33.3. . 3-3... 3N ...N. h... v.3 . 93 3 N3... 5.» 3 .3... 33.63 3 33.. 3N...— N 35-3... .. ad. 3.3 cd ...v .. 33.... a . «3.33 3 - 3 . .3. N no. .-3... a... .9: ..5 .2. 23.25 598. Q3 £03. Gm .0... .0... mos. am E3 «3 mos. am 3533 «3 no: Om 3.0 «3 d3. 0:3 8.38. .3 29:. 85 .fiflw 05 E 32.88% «o: mm 33 £8» 53 5 Born 82 mo 89.58 053—2 3m 88.. £955 96: «on Ev «am on: a Saratov 28:5” 582%. gm .5» £8 “a 82 u 58c 8sz as 3% 85. 28a 3.2» 32 Bugs 2: a? £0: 88 3595 newt—«9:00 ma Bap—axe 98 ago?» “com .88 98 Sea waist 88. 30on an me 203 .02 3x35 .~.m 0.5mm..— on 8 plagfi 100.1 paxapul 8— cm— 86 3cm U ween l x «X» x «VN \ ll. - 2002 [:3 2003 ployC osomns poxopul 87 Line deviations. F3 line 29a did not have enough roots for reliable estimate of sucrose yield in both years, and is not represented in the amph- Figure 3.2. Indexed sucrose yield of F3 progeny tests during 2002 and 2003. Sucrose yields are expressed as comparison (indexed sucrose yield) with the average sucrose yield of the lines SR96 and USI-IZO (index = 100) per each year. Bars represent standard 26* 24‘ 11. A." 20< Root dry matter ("/o) 26 l o 2002 o 2003 g 24 - . 22 ~ 20 - Root dry matter (%) 18 - .r 16 ‘ '4 I r I 1 I 45 50 55 60 65 70 Sucrose as dry matter (%) Figure 3.3. Correlation between root dry matter and root sucrose content as dry matter in F3 progeny tests. Black and grey lines represent correlation lines during 2002 (y = 20.95 — 0.04 x; R2 = 0.006) and 2003 (y = 11.95 + 0.12 x; k1 = 0.080), respectively. 4.. 2 m We Ae\ev 3:5: be :51 I6 Ae\ev L023: XL? :0 UmCLUSW 26 W ‘1 2002 C) 2003 .1 Root dry matter (%) B 1? E, ‘- i G O 3 § 5 m 45 I I I I I I I 1 7 8 9 10 11 12 13 14 15 Sucrose on fresh weight (%) Figure 3.4. Analysis of correlation between root sucrose content as fresh weight and A. root dry matter, and B. sucrose content as dry matter. A. Black and grey lines represent correlation lines during 2002 (y = 5.68 + 1.24 x; R2 = 0.666) and 2003 (y = 7.49 + 1.02 x; R2 = 0.733), respectively. B. Black and grey lines represent correlation lines during 2002 (y = 39.39 + 1.59 x; R2 = 0.263) and 2003 (y = 35.08 + 2.12 x; R2 = 0.543), respectively. 89 Table 3.3. (% DM), 1 (% SucFW 0'2, = varie years; df = Trait % DM “/3 SucDM “/0 SUCF W Table 3.3. ANOVA table for broad sense heritability (1123) estimates of root dry matter (% DM), sucrose content as dry matter (% SucDM), and sucrose content as fresh weight (°/o SucFW). 02, = genetic variance; 02,, = variance due to genotype x year interaction; 02.. = variance due to experimental error, r = number of replications; y is the number of years; df = degrees of freedom; SS = sum of squares; MS = mean squares. Trait hZB Source df SS MS Expected MS % DM 0.44 genotype 53 324.9 6.129553 oz, + 1.8946 023,.» 3,6778 02. gen. x year 52 214.1 4.117052 02¢ +185510’23y residual 116 399.5 3.444247 02, % SucDM 0.68 genotype 53 1781.7 33.616795 62, + 1.7858 62., + 3.5342 6’. gen, x year 53 794.7 14.994400 0’2. + 1.7843 61.. residual 104 1562.0 15.019332 6’. % SucFW 0.50 genotype 53 167.8 3.165458 03. + 1.6823 (1’2 gy + 3.2717 0}; gen. x year 52 104.3 2.006358 62¢ + 1.6532 0'2 .y residual 88 125.0 1.420365 62. 9O q 1 ‘77 q q 0 8 7. 11 TX; NOON 5086:. she—v «COM 1 6 [l1 4 4 q u 1 q 1 ,M w «A v. Ae\ev NOGN 5025: b3 :0 virtual 24W O A 22- 2% ° 0 S 00 o 00 o 3 20- o 0 0 2.. 2 O o 0 O 0 Q) a n W: 0 o E 18 O O 080 000 " O E 8 8:. o o o 0 8 a: 0 o ‘6‘ o 0 I4 I I f I 1 14 16 18 20 22 24 B Root dry matter 2003 (%) 68- 8°. 64- N O C N 1'5 60‘ a”: E E a 56- o 8 8 g 52- U) 0 O 48 I I I I I 48 52 56 60 64 68 Sucrose on dry matter 2003 (%) Figure 3.5. Analysis of correlation of root dry matter content and sucrose content as dry matter between years. A. Black line represents correlation line (y = 16.24 + 0.14 x; R2 = 0.028). B. Black line represents correlation line (y = 32.47 + 0.46 x; R2 = 0.208). 91 5 3 w 2 2 41 4| AOV oeaamewaEwk q q 5 5 . 2 1 «UV oeaumcoarcmu 00:99.20 0505050505 21 ...1...QQ wmwm 221141 3 «E5» (036230er 82233530 Difference temperature (C) Temperature (C) Cumulative precipitation (mm) -5 _ -10 _ -15 5 -20 5 — 2002 T. max A 2002 T. min. l l I l l l — Diff. T. max (2003 -2002) B Diff. T. min. (2003 - 2002) WW W 4M -25 300 i — 2002 250 i 200 1 150 - 100 - 50- — — — 2003 0 May Jun Jul Aug Sep Oct Nov Month Figure 3.6. Meteorological data daily recorded at the Michigan State University Agronomy Farm, East Lansing, MI, during the growing seasons 2002 and 2003. A. Maximum and minimum temperatures (°C) during 2002. B. Differences of maximum and minimum temperatures (°C) during 2003 compared to 2002. C. Cumulative precipitations (mm) during 2002 and 2003 92 N ~<~CN NawaN tv~ .C(. v35 :; V... \\ :6 My. ll 0 <2 .I. h 0:59.12: 9.. .9. V fl i. 32 BEN s S ‘ 3 . . v 3 25m ~33 r \L 4 ..J 5. l. 0 5 .Q a) 3.. I: 0 :_.V\ z- <27. z-.V\ .2— <:% E—.V\V<~ :AhUDV. JA- EAH 9" 5 BROS” «\c 53:8,? .83 as. .85 .2 .3 v a a 885%. 38.3 :2. e5 .2. .3 .. .2... a 85% %2 28% =3 .28 means ates 2.8.288 Ba .285. 8:8 .2: manage 3323 .229 ”was... coo—«E 29:» mama: A83 383 mm 888 906 8cm v5 A82:— moew mm 8.68 003 «com E veg—“>0 SEQ—m .xu E303 €29 me 28:80 8983 98 .9505 .xev has: Eu 3 88:8 3983 A35 .5 .832: .3. 82 mo £35 :8. 2583550 Hm 85E 83829: oi 88.82% \22 35.059. / «.2 8.5659. -/ / . \_ S . L”? O . x, 8 28525.. ,./// 3 54.329... 0.7 on Ov—F~ 2=°I°nfl°L£LU Efvx 2— ('27. @283 .3 03mm $830M woo M I . . o o o_o £380 97 ...I .3:<.Z . _. ,,. .. I _ u 2.55 y. 5?. 3.. . a c «cam Nccw netm NCSN MCCN «can Ervx 3: <27. 2..u\ z— (27. 3:.vx.)= (IV. if /‘ Blazw o\o 2003‘. : ~> §°.°I=EIQLV €353 gum onE SzmmMme. M3... .3320: “.2 33.85:. N.N” $63255... , D «on oE 05305930 98 C II In ...... :7 . .\ O M z, 00 O. ... ‘:X»f.. :4 vexi \ \ C r1 u _ O . V. 000 000 _ M al.»; ( .3: (a, VM~ ~ ~ 0 . m . . M . . ... 0 . . ‘ I. NN-.~ (r V‘(<. E a H» 3 2 l. 0 MOON NOON 5 4 3 2 I. 0 moan N8 5 4 3 2 I. 0 MSN 25m 2-.v\ 2- (27. 2-.V\ u): (2?. 2~.v\ 2~ <27. BLUSW 9» 2°03“ 90 In x -> OEOOQEPQ‘HV $.88. .2 8%.... o _ 99.2; ) ..mm a 9.30% v.7. w m 9.09.255. f kw“ .. #352098 1\ 1 / ..... . $383... \\ /n.$ m 3.39.. \\1 [a 9 _ «805.209... ...... ..fio _ u. . ..... . . . \ .. ”82020.... ...n . .. .0 . . __ \ .. «Buzzwfi/Mu Kw...“ ... u _ .8082... u .. .. n n w.\ 889.29... V. 1 A23 : ...»... / v ... . 38520.... \UA n fi/ffifi co .3 . /; . c o —QNO 0.26-3.55:9 8.2.8. .3 as»... . 8521‘ i] L: H . vSo83 .:U E: : :> .....v 2 > .£.v >~ .£.V -~ \\ ..F v \\ .C.V e \ 6.830588 866—88 88 Nam oSwm 8 Saab $2520 3.“ @0280 :02. 26: 88a: hoe—.82 Avg: mmmtousm 2 $5888 8820580: 35v 88808030 .20 8 mecca—mac 32 .3535 .XL 8803 58m 8 8088 8825 can .32ng $9 5.88 5 mm 8688 0888 .95 $9 888 S poo.— 88 62 83 3238838 me 80:82 68266 dam onE I D l n BmosmxflU MEnosmfienH_ m35.x.l l f .20 3 co mm Om Ll I E l ,E l I l L I l I IIJI I II II I 'l : /// Vlllll ov mm 9:. mm on m— o— ; nHU mm um IL Ir o 5 .5 :5 .80 => :0 > :0 Z :0 E .80 H 80 I_IIIII 1111111 1111 II I II I /[A I\\“\\\\I II \I//// I \I D I W [l I IjT Ill 11 \\ IJ ll 11 [1\ II 11 I] l] I I IIIIIIII_I l l\\\\\\\\\ ’ I I III IIIIII w‘“§\W//Kfi l 1 II III I I II III 11 Lllllll III 1 11 I II V) 104 The \ expressic SucDM) control 01 by the co segregatic the limiter the em'iro 1.118 IOW-m intennedia %SmDM betWeen °/'b bem'een 9/0 the % SUCF % of the re fraction 0ft An impc betweefl % implilng the SucDM trait: “4‘34 . Discussion The wide cross between sugar and table beet lines maximized the variability of expression of root dry matter content (% DM) and sucrose content as dry matter (% SucDM) in the field-grown F3 progeny tests in both years. The assumed polygenic control of root dry matter content and dry matter partitioning in sugar beet was confirmed by the continuous variability observed in both traits. Those rare cases of transgressive segregation were usually associated with high level of variability of the samples due to the limited number of replications in 2003. This variability also indicates an influence of the environment on the traits, particularly on root dry matter content. The combination of the low-medium heritability of % DM and high heritability of °/o SucDM determined an intermediate heritability of the combined trait % SucFW, as expected. Both % DM and % SucDM were positively correlated with % SucF W. However a higher correlation between % DM and % SucF W was observed in both years with respect to the correlation between % SucDM and % SucF W. This observation could be explained considering that the % SucF W is more influenced by variations of the % DM, which represents 15 to 25 % of the root weight, with respect to variations of % SucDM, which only represents a fraction of the % DM. An important finding of this research was the complete absence of a correlation between % DM and % SucDM and their wide ranges of variability in both years, implying that it is possible to increase % SucFW by selecting for both % DM and % SucDM traits independently. Particularly during the early stages of breeding programs after the introgression of favorable genes from wild un-adapted germplasm, it should be recommended to consider and analyze both % DM and % SucDM independently in the 105 599 61 % Suc relatec already which selectic The several Havey e content 3 lines ove on Chrotr QTL werc Similar detected f. HOWCVeL 1 0n the 33111: 00mm“ m SUCF W c0~1 manilation er al. (2002 phosphate 3 SUcDM R'er. segregating populations. The first cycles of selection should mainly focus on increasing % SucDM because of its higher heritability, and trying to increase the more environment- related % DM in the later cycles. During the late stage of the breeding programs, when already selected materials are evaluated between different environments, the % SucF W, which would mainly reflect root dry matter content, would represent the most economic selection parameter. The analysis of QTL to dissect complex traits like sucrose content has been used for several crops (Maughan et al., 2000; Ming et al., 2001 and 2002; Natoli et al., 2002; Havey et al., 2004). Recently, Schneider et al. (2002) mapped QTL related to sucrose content as fresh weight (% SucF W) in sugar beet using populations derived from elite lines over several locations. The authors found a total of five QTL for % SucF W spread on chromosomes 1, 11, VI, VII, and IX, but high variability were observed and not all QTL were always detected between locations. Similar variability was observed during my study and a total of seven QTL were detected for % SucF W on chromosomes I (one), 111 (one), IV (two), and VII (three). However, analysis of co—localization of these QTL with those of Schneider et al. (2002) on the same chromosomes I and VII was not possible because of the lacking of reference common markers between the two genetic maps. Interestingly, all seven QTL for % SucFW co-localize with % DM (three) or % SucDM (three) or both (one), giving further information on possible biochemical pathways causing high sucrose content. Schneider et al. (2002) associated the enzymes sucrose synthase on chromosome VII and sucrose phosphate synthase on chromosome VIII, two chromosomes for which QTL for % SucDM were detected in my study. However, QTL analysis should not only be expected 106 to detec factors. enzyme- and Fed. %SmD detected identific: weight I in the p0 of both Another expressic Seems to the stabil SucFW It Most SUcF“r. to detect location of structural genes, but also regulatory genes such as transcriptional factors, kinases and gene promoter regions, making the identification of QTL with enzyme-related candidate-genes more difficult (McMullen et al., 1998; Mitchell-Olds and Pedersen, 1998). During my research, one more QTL for % DM and four more for % SucDM were also detected during the analysis, and these QTL would not have been detected with the sole analysis of % SucFW. Several reasons could explain the identification of QTL for one or both of the two traits that influence sucrose as fresh weight, but not for % SucF W itself. It is possible that the variability of one of the traits in the population is too low to be detected when combined as % SucFW, or the variability of both traits behave with Opposite effects on % SucFW in a particular population. Another reason could be if the variability of one trait is masked by the stability of expression of the other trait that mainly influences the level of % SucFW. This later case seems to have occurred in my study for the five QTL only detected for % SucDM, where the stability of level of expression of % DM reduced the phenotypic variability at the % SucF W level. Most of the allelic combinations derived from the sugar beet parent increased % SucFW. Of the seven QTL detected for sucrose as fresh weight, six alleles derived from the sugar beet parents increased the trait and for one locus there was not statistically difference between allele combinations of the two parents. The same was observed for % DM, where four alleles combinations from the sugar beet parent increased % SucF W and one was not statistically different between parents. Surprisingly, for three of the nine QTL detected for % SucDM, the allelic combinations derived from the table beet parent increased the level of sucrose content in the root dry matter. Two of these QTL were co- 107 loca was limit reach found germ}: Cat were c detecte. evaluati imponai moderatc Conditior. CV61] if a COHSitler 1 11111 Uencin‘ localized on chromosome IV, and one was only detected for % SucDM while the other was also detected for % SucF W by CIM in 2003. This finding suggests that the higher limit of sucrose dry matter partitioning in sugar beet might not have already been reached, and potential to increase % SucDM, and consequent sucrose yield, could be found in different QTL allelic combinations present in un-adapted Beta vulgaris L. germplasm. Caution in the interpretation of QTL results must always be used. Even if some QTL were constantly detected in both years by SMA and by CPA, other QTL were only detected in single year, particularly considering CIM analysis. Further years of field- evaluation in different locations will be necessary in order to elucidate the relative importance of the 13 QTL detected in my study on root sucrose content sugar beet. The moderate level of heritability of the traits under study and the differences of growing conditions between years could explain the level of traits variability observed. However, even if moderate for some of the traits, the level of heritability was sufficiently high to consider QTL analysis a valid approach to detect the locations and the effect of genes influencing root sucrose content in B. vulgaris. 108 (i) I rootdr vans v dunng fionixv d1)" mat quickly the selec (n) A influenci hBfimmes influtnch 9.2m (i) In breeding populations derived by crosses from genetically diverse germplasm, the root dry matter content and the root sucrose content as dry matter were not correlated and traits variability followed a quantitative pattern. This finding implies that, particularly during the early stages of a breeding program afier the introgression of favorable genes from wild germplasm, simultaneous and independent screening and selection based on dry matter content and on root sucrose content as dry matter would be more efficient in quickly recovering individuals with high sucrose content as fresh weight, with respect to the selection exclusively based on sucrose content as fresh weight itself. (ii) A total of 13 genomic regions located on seven chromosomes were identified influencing root sucrose content. These 13 loci influenced root dry matter content in five instances and root sucrose content as dry matter in nine instances, with one locus influencing both traits. Seven of these loci were directly detected with the analysis of sucrose content as fresh weight, while six more loci would not be detected without analyzing the trait separately for % DM (one) and % SucDM (five). Overall, 13 QTL influencing water/dry matter relationship and partitioning of root dry matter between sucrose and non-sucrose metabolites might influence the amount of sucrose stored in the root and the sucrose yield capability of sugar beet breeding lines. Markers associated with these regions could be used to drive the selection of superior individuals during the early and late stage of breeding programs via marker-assisted selection, while information of the metabolic and regulatory pathways influencing sucrose content would increase our knowledge on sucrose accumulation in sugar beet. 109 Berg: Butte; (Be Churcl map; Carter. sugar Doney, sucros Literature cited Bergen, P. Seasonal patterns of sucrose accumulation and weight increase in sugar beets. J. Am. Soc. Sugar Beet Technol. 1967, 14, 538-545. Butterfass, T. Die chloroplastenzahlen in verschiedenartigen zellen trisomer zuckerruben (Beta vulagaris L.). Z. Bat. 1964, 52, 46-77. Churchill, G. A., and R. W. Doerge. Empirical threshold values for quantitative trait mapping. Genetics 1994, 138, 963-971 Carter, .1. N. Sucrose production as affected by root yield and sucrose concentration of sugarbeet. J. Am. Soc. Sugar Beet T echnol. 1987, 24, 14-31. Doney, D. L.; Wyse, R. E.; Theurer, J. C. The relationship between cell size, yield, and sucrose concentration of the sugarbeet root. Can. J. Plant Sci. 1981, 61 , 447-453. F ehr, W. R. 1987. Principles of cultivar development. Macmillan, New York. Geidel, H.; Weber, W. E.; Mechelke, W.; Haufe, W. Selection for sugar yield in sugar beet, Beta vulgaris, using different selection indices. Plant Breeding 2000, 119, 188- 190. Havey, M. J .; Galmarini, C. R.; Gocke, A. F .; Henson, C. QTL affecting soluble carbohydrate concentrations in stored onions bulbs and their association with flavor and health- enhancing attributes. Genome 2004, 47, 463-468. Hecker, R. J. Evaluation of Three Sugar Beet Breeding Methods. J. Am. Soc. Sugar Beet Technol. 1967, 14, 309-318. Hsu, J. C. 1984. Constrained simultaneous confidence intervals for multiple comparisons with the best. Annals of Statistics 12, 1136-1144 110 Johnson in soy Maugh contro 105-l McGra 1: Re Resea McMulle W. Quz 1996-2 Milford. matter . Ming. R. COmple Resear. Ming. R. dissecti TEIaIed MilChell. central Johnson, H. W.; Robinson, H. F .; Comstock, R. E. Genotypic and phenotypic correlations in soybeans and their implications in selection. Agron. J. 1955, 47, 477-483. Maughan, P. J .; Saghai Maroof, M. A.; Buss, G. R. Identification of quantitative trait loci controlling sucrose content in soybean (Glycine max). Molecular Breeding 2000, 6, 105-111. McGrath, J. M., C. A. Derrico, Y. Yu, and R. A. Kitchen. Field evaluation of emergence I: Replicated trails of a range of sugar beet and related germplasm. Sugar Beet Research 1998, E3-E5. McMullen, M. D.; Byme, P. F .; Snook, M. E.; Wiseman, B. R.; Lee, E. A.; Widstrom, N. W. Quantitative trait loci and metabolic pathways. Proc. Natl. Acad Sci. USA 1998, 95 , 1996-2000. Milford, G. F. J.; Travis, K. 2.; Pocock, T. O.; Jaggard, K. W.; Day, W. Growth and dry- matter partitioning in sugar beet. J. agric. Sci, Camb. 1988, 110, 301-308. Ming, R.; Liu, S. C.; Moore, H. P.; Irvine, E. J .; Paterson, H. A. QTL analysis in a complex autopolyploid: genetic control of sugar content in sugarcane. Genome Research 2001, I I, 2075-2084. Ming, R.; Wang, Y. W.; Draye, X.; Moore, H. P.; Irvine, E. J .; Paterson, H. A. Molecular dissection of complex traits in autopolyploids: mapping QTLs affecting sugar yield and related traits in sugarcane. Theor. Appl. Genet. 2002, 105, 332-345. Mitchell-Olds, T.; Pedersen, D. The molecular basis of quantitative trait variation in central and secondary metabolism in Arabidopsis. Genetics 1998, 149, 739-747. Natoli, A.; Gorni, C.; Chegdani, F.; Marsan, P. A.; Colombi, C.; Lorenzoni, C.; Marocco, A. Identification of QTLs associated with sweet sorghum quality. Maydica 2002, 47, 31 1-322. Oldemeyer, R. K. Introgressive Hybridization as a Breeding Method in Beta vulgaris. J. Am. Soc. Sugar Beet Technol. 1975, 14, 269-273. 111 Power gani 432. Pnuflu ofsu SASln Sdmdc sucro rekue ShnHKn 1994, Trebbi, ] (Went leBefl. 90.328 Wise. R. Swfln-E “3).&’ R. SUpply ‘ Powers, L. Identification of genetically-superior individuals and the prediction of genetic gains in sugar beet breeding programs. J. Am. Soc. Sugar Beet Technol. 1957, 9, 408- 432. Pritchard, F. J. Correlation between morphological characters and the saccharine content of sugar beets. American Journal of Botany 1916, 3, 361-376. SAS Institute. 1994. The SAS System for Windows. Release 8.0. SAS Inst., Cary, NC. Schneider, K.; Schafer-Pregl, R.; Borchardt, D. C.; Salamini, F. Mapping QTLs for sucrose content, yield and quality in a sugar beet population fmgerprinted by EST- related markers. Theor. Appl. Genet. 2002, 104, 1107-1113. Simmonds, N. W. Yield and sugar content in sugar beet. International Sugar Journal 1994, 96, 413-416. Trebbi, D.; McGrath, J. M. F luorometric sucrose evaluation for sugar beet. J. Agric. Food Chem. 2004, 52, 6862-6867. van Berloo, R. GGT: software for the display of graphical genotypes. J. Hered. 1999, 90, 328-329. Wyse, R. Parameters controlling sucrose content and yield of sugarbeet roots. J. Am. Soc. Sugar Beet Technol. 1979, 20, 368-385. Wyse, R. Partitioning within the taproot sink of sugarbeet: effect of photosynthate supply. Crop Science 1980, 20, 256-258. 112 PHEl SL’CI DEVI The obj accumul Characte; identify sucrose Characteri dllring to different}; accumulai hybridtzat exPIESSed Arabdops; | accumula: name rm“ CHAPTER 4 PHENOTYPIC AND GENOTYPIC ANALYSES OF ROOT SUCROSE ACCUMULATION DURING JUVENILE TO ADULT DEVELOPMENTAL PHASE CHANGE IN SUGAR BEET Abstract The objectives of this research were (i) to understand the dynamics of root sucrose accumulation during the early root developmental phases in sugar beet, (ii) to characterize gene expression profiles during these early developmental phases, and (iii) to identify genes differentially expressed between developmental phases associated with sucrose accumulation. Greenhouse and field-grown plants were phenotypically characterized and compared for two consecutive years, and gene expression profiles during root development were analyzed with cDNA-AF LP. Identification of genes differentially expressed between developmental phases characterized by different sucrose accumulation activity was performed by sequencing clones obtained from a subtractive hybridization library. Quantification of levels of up and down-regulation of differentially expressed genes during root development was performed with Beta vulgaris L. and Arabdopsis thaliana L. cDNA and oligonucleotide microarrays. Sucrose started accumulating after the 3rd week after emergence (WAE) and reached level comparable to mature root at the 7"I WAE. Regulatory genes were differentially expressed between these two stages of the root development in sugar beet. 113 j uven are m betwei pathwz endoge 1131151111 plant 11: Characte Introduction The plant life cycle can be subdivided in four chronological phases: embryonic, juvenile vegetative, adult vegetative, and reproductive (Poethig, 1990). Generally there are not specific physiological or morphological characteristics that indicate transition between phases. Phase changes are usually controlled by an array of signal transduction pathways triggered by the interaction of exogenous (i.e. temperature and light) and endogenous (i.e. hormones) factors. Environmental and genetic factors influencing the transition from adult vegetative to reproductive phases have been well characterized in plant using Arabidopsis thaliana as model organism, while phenotypic and genotypic characterizations during the transition from juvenile to adult vegetative phases has also been analyzed considering the shoot system of Zea mays (Poethig, 1990; Orkwiszewski and Poethig, 2000; Vega et al., 2002; Poethig 2003). Little information is known on juvenile to adult vegetative phases transition of the plant root system. However, the timing and intensity of these biological changes, particularly related to sucrose accumulation during early root developmental of Beta vulgaris, could influence the rate of establishment of the crop, and could represent good indicators of productivity of breeding lines and be used as selective parameters. Theories on the dynamics of sucrose accumulation in sugar beet have been contradictory in early studies. Root development was believed to occur late during the growing season after the complete development of the leaves, and sucrose accumulation triggered by lower temperatures at the end of the growing season (Bouillene et al., 1940; Ulrich, 1952 and 1955). Watson and Selman (1938) and van Ginneken (1959) assumed that early development was mainly dominated by leaves expansion, but were unable to 114 disting 1970's sucros: growin and 19: Exp] during] Willenb 1996; K and con rarely b CXpressic genes pr:— The 0 cOntent ix phases in accumula Phases an. genes dif; A Phen (hiring the content 31‘ distinguish separate phases of growth and sucrose accumulation in the root. During the 1970’s the theory that photosynthates partitioning between leaves and root apparatus and sucrose accumulation in the root occur simultaneously and continuously during the growing season was accepted (Bergen, 1968; Follett et al., 1970; Milford, 1973, 1976, and 1988; Wise, 1979 and 1980). Expression analysis of specific candidate genes involved in sucrose accumulation during root development in sugar beet has been addressed by several authors (Fieuw and Willenbrink, 1990; Chaubron et al., 1995; Hesse et al., 1995; Hesse and Willmitzer, 1996; Klotz and Finger, 2002; Klotz and Campbell, 2004). However, a comprehensive and comparative gene expression analysis during root development in sugar beet has rarely been investigated. Bellin et al. (2002) used a genome-wide comparative expression analysis between different sugar beet plant organs and were able to identify genes preferentially expressed in the storage root. The objectives of this research were (i) to examine the dynamics of root sucrose content in greenhouse and in field grown sugar beet lines during the early developmental phases in order to identify morphological and physiological changes influencing sucrose accumulation, (ii) to characterize gene expression profiles during the early developmental phases and to relate these to the dynamics of sucrose content changes, and (iii) to identify genes differentially expressed between developmental phases characterized by different sucrose accumulation activity. A phenotypic analysis of greenhouse and field grown sugar beet lines was performed during the early root developmental phases to understand the dynamics of root sucrose content and to identify critical stages for sucrose accumulation. Gene expression analysis 115 (cDb early expre develt vulgar develo microa (cDNA-AF LP) was also performed to characterize gene expression profiles during these early developmental phases to correlate the dynamics of sucrose content with gene expression changes. Identification of genes differentially expressed between developmental phases critical for sucrose accumulation was finally performed using Beta vulgaris cDNA microarray selected by subtractive hybridization of a sugar beet root developmental cDNA library and Arabidopsis thaliana cDNA and oligonucleotides microarrays. 116 Plant Ar IICICI'} sucros % to n for a random time of Materials and Methods Plant material and growth conditions Analyses were conducted in greenhouse-grown plants during 2002 and 2003, and on field-grown plants during 2003. In 2002, five sugar beet lines characterized by different sucrose contents at root maturity (USH20, SR87, SR95, SR97, and SR96 ranging from 15 % to more than 17 % of sucrose as fresh weight, respectively), were grown in greenhouse for a period of nine weeks after emergence (WAE). Experimental design was a randomized complete block design with three replications. In order to synchronize the time of emergence, seeds were treated with a 0.3 % H202 aqueous solution for 48 h (McGrath et al., 2000), and germinated seeds showing root tip erupting the seed were hand-transplanted in 0.25 m2 wooden boxes with 15 cm soil depth, spacing the seeds at 5 cm within and between rows, with an 8-16 dark-light cycle, at 15 to 20 °C, irrigated daily and fertilized twice a month. Time of emergence (March 4th, 2002) was estimated as 96 h after transplanting, when approximately 95 % of the plants had emerged. During 2003 two of the previously tested beet lines (USH20 and SR96) and the two parental lines of the mapping population (Chapter 2 of this Dissertation), sugar beet C869 (Lewellen, 2004) and table beet W357B (Goldman, 1996), were grown in the greenhouse for a period of seven WAE. Experimental design was a randomized complete block design with two biological replications per each of the three replications. Time of emergence for the 2003 experiment was March 26‘". The same four beet lines tested in greenhouse during 2003 (USH20, SR96, C869, and W357B) were also analyzed for the dynamics of root sucrose accumulation in plants 117 grow consec and tit plants East It Sam p11 Sam respecti WAE It root suc CXpressc matter c Root commit follom‘n Sucre Via 6112M, ChaPIEr mlmateg grown under field conditions during the summer of 2003. Each line was planted in 7 consecutive double-rows (6 m long, with 0.76 m row inter-distance) on May 21“, 2003, and time of emergence (June 2nd, 2003) was estimated when approximately 50 % of the plants had emerged. Plants were grown at Michigan State University Agronomy Farm, East Lansing, MI, using standard agronomic practices. Samples collection and traits analysis Samples were collected weekly from the 2'ml to the 9"I and from the 3rd to the 7th WAE, respectively for the 2002 and 2003 greenhouse experiments, and fi'om the 3"l to the 20‘“ WAE for the 2003 field experiment. Traits measured were root dry matter content and root sucrose content expressed as proportion of the dry matter, while sucrose content expressed as proportion of the fresh weight (% SucF W) was estimated from the root dry matter content and the root sucrose content as dry matter. Root samples were lyophilized, and root dry matter content (% DM) was estimated by comparison of weights before (wtbcfm) and after (wtmfl) drying the samples, using the following equation: % DM = (wtmer / Wtbefore) x 100 Sucrose content as dry matter (% SucDM) was measured via HPLC for the 2002 and via enzymatic-fluorometric assay for the 2003 samples (Trebbi and McGrath, 2004; also Chapter 1 of this Dissertation), while sucrose content as fresh weight (% SucFW) was estimated using the following equation: % SucFW = (% DM x % SucDM)/ 100 118 remai % SUI regres CXUaClit and fror han'este Wildie biOIOgiC.’ NUCICIQ Tout (Cat No Purified! During the early weeks after emergence roots from 50 to 60 plants per each sample (5 g) were collected, and progressively reduced to 5 plants per samples (20g) during the final weeks. Plant collection for the greenhouse experiments was carried out in order to leave the remaining plants thinned and equally spaced within each box, while plant collection in the field was performed by randomly selecting roots and leaving the remaining plants equally spaced within plots. Phenotypic data of % DM, % SucDM and % SucFW of 2002 and 2003 experiments were analyzed and plotted with the non-linear regression model (logistic curve) with SigmaPlot 2001 (SPSS Inc., Chicago, IL). Analyses of the variation of gene expression (cDNA-AF LP experiments) and identification of differentially expressed genes during early root developmental stages of sugar beet (microarrays experiments) were performed on greenhouse-grown plants during 2002 and 2003, simultaneously to the phenotypic analysis. Root samples for RNA extraction were collected weekly from the 1“ to the 9th WAE from USH20 during 2002, and from the 3" to the 7'" WAE from USH20 and SR96 during 2003. Plants were harvested and 5 g of roots were washed from soil, separated from leaves and hypocotyls, immediately frozen in liquid nitrogen and stored at — 80 °C before RNA extraction. Two biological replications were collected per each line during 2003. Nucleic acid purification Total RNA was extracted using the Concert"I Plant RNA Reagent Kit (Cat. No. 12322-012, Invitrogen Life Technologies, Carlsbad, CA) and purified with RNeasy® Kit (Cat. No. 74104, Qiagen, Valencia, CA) according to manufacturer’s instructions. Purified total RNA was quantified using the RiboGreenO RNA Quantification Kit (Cat. 119 No. I pre-c accor RVA Carls Grade manut synthe l 1917- N0. .\3 cDNA DNA C manufat No. R-11490, Molecular Probes, Eugene, OR) and RNA quality was assessed on Reliant‘D pre-casted 1.25 % agarose gels (Cat. No. R-54948, Cambrex, East Rutherford, NJ) according to manufacturer’s instructions. Messenger RNA (mRN A) isolation from total RNA was obtained using the mTRAPm Total Kit (Cat. No. 23012, Active Motif, Carlsbad, CA) performing the optional DNase treatment with DNaseI Amplification Grade (Cat. No. 18068-015, Invitrogen Life Technologies, Carlsbad, CA) according to manufacturer’s instructions. Double-stranded complementary DNA (ds-cDNA) was synthesized using the SuperScript" Double-Standed cDNA Synthesis Kit (Cat. No. 11917-010, Invitrogen Life Technologies, Carlsbad, CA) employing the Oligo dT (Cat. No. N420-01, Invitrogen Life Technologies, Carlsbad, CA) as primer for first-strand cDNA synthesis, and quantification of the ds-cDNA was performed with the PicoGreen‘ID DNA Quantification Kit (Cat. No. P-l 1496, Molecular Probes, Eugene, OR), following manufacturer’s instructions. cDNA-AFLP protocol For the sugar beet USH20 samples collected in 2002 (1“ to 9m WAE), cDNA-AF LP analysis was performed using the EcoRI/Msel (E/M) restriction enzyme combination and following the protocol of Bachem et al. (1996 and 1998) with modifications. Complementary DNA (150 ng) was double digested at 37 °C for 3 h with EcoRI (Invitrogen, Carslbad, CA) and Msel (New England BioLabs, Beverly, MA) restriction enzymes in 30 1.1L of restriction solution [0.17 U 1.1L" EcoRI; 0.17 U 11L" Msel; 10 mM Tris (pH 7.5); 10 mM Mg acetate; 50 mM K acetate; 5 mM DTT; 0.05 pg 3.1L" BSA]. Restriction enzymes were deactivated with a treatment at 70 °C for 15 min. 120 pho: phos peric ligatt' nLM . acetat aInpli: selecti micro] solutio Ix.) mM polyme anm‘éilir °C for 1 diluted f The s ”Sing sin Iabeled (. tinexemdi (M2), at microliter Double stranded AF LP adapters were produced from two single stranded non- phosphorylated linkers (MWG Biotech Inc., High Point, NC). To prepare non- phosphorylated double stranded EcoRI and Msel adapters, equimolar amounts of the two complementary linkers, were slowly cooled down from 70 °C to 25 °C in a one-hour period, after an initial treatment at 95 °C for 3 min. EcoRI and Msel adapters were ligated in 40 uL of ligation solution [0.35 uM EcoRI adaptor; 1.75 pM Msel adaptor; 1 mM ATP; 0.025U 11L" T4 ligase; 10 mM Tris (pH 7.5); 10 mM Mg acetate; 50 mM K acetate] at room temperature overnight. The first PCR cycle of amplifications (pre- amplification) was performed with EcoRI (E +0) and Msel (M+0) primers without any selective nucleotide extending into the restricted-legated genomic sequences. Two microliters of restricted-ligated solution was used to prepare the 20 uL pre-amplification solution [0.5 uM E+0 primer; 0.75 uM M+0 primer; 0.188 mM dNTPs; 2.5 mM MgC12; 2 mM Tris-Hcl; 10 mM KCl; 0.01 mM EDTA; 0.1 mM DDT; 0.025 U ILL" Taq polymerase], which was amplified for 20 cycles of 94 °C denaturation (30 s), 56 °C annealing (30 s) and 72 °C extension (60 s), with initial steps at 72 °C for two min and 94 °C for l min, and last extension period at 72 °C for 10 min. Pre-amplified solutions were diluted five-fold with low-TE buffer [10 mM Tris; 0.1 EDTA; pH 8.0]. The second PCR cycle of amplifications (selective amplification) was performed using simultaneously (multiplexing) one of each IRD700 and IRD800 fluorescence- labeled (LI-COR Biosciences, Lincoln, NE) EcoRI primers with each three selective nucleotides (”DE +3), a single unlabeled Msel primer with two selective nucleotides (M+2), and 1 uL of the diluted PCR product from the pre-amplification. Fifieen microliters of selective amplification solution [17 nM ImmE+3-primer; 17 nM 121 biologij EXpres manufa (1.11) r With lo Primer t Amp el‘cct0pl EDTA; Selectit.E gel gem 1R2 auto “It 0.2 ‘. bode acii IRD800E+3-primer; 0.5 M M+2 primer; 0.188 mM dNTPs; 2.5 mM MgC12; 2 mM Tris- Hcl; 10 mM KCl; 0.01 mM EDTA; 0.1 mM DDT; 0.025 U ttL'l Taq polymerase] were amplified for 13 cycles of 94 °C denaturation (30 s), 65 °C annealing (30 s; -0.7 °C cycle’ I), and 72 °C extension (90 s), followed by 23 more cycles of 94 °C denaturation (30 s), 56 °C annealing (30 s) and 72 °C extension (90 3; +2 8 cycle”), with initial denaturation at 94 °C for 30 s, and last extension period at 72 °C for 10 min. A total of 134 different primer combinations were analyzed combining 9 E+3 with all 16 possible M+2 primers (Tab 4.1). For the USH20 and SR96 samples collected in 2003 (from the 3“| to the 7‘“ WAE; two biological replications per sample), cDNA-AFLP was performed using the AFLP“ Expression Analysis Kit (Cat. No. 830-06518, LI-COR Biosciences, Lincoln, NE) as manufacturer’s instructions, starting from 150 ng of ds-cDNA and using the T an/Msel (T/M) restriction enzyme pair combination. Pre-amplified solutions were ten-fold diluted with low-TE buffer solution [10 mM Tris; 0.1 EDTA; pH 8.0]. A total of 8 different primer combinations were analyzed combining 5 T+2 with 5 M+2 primers. Amplified fragments (transcript-derived fragments, TDFs) were separated via gel electophoresis. Two microliters of stop/loading buffer [formamide 95% v/v; 10mM EDTA; 0.1% basic fuchsin; 0.01% bromophenol blue; pH 9.0] were added to 5 1.1L of selective amplification product, and the solution was denatured at 96 °C for 5 min before gel separation. One microliter of the denatured solution war analyzed on 4200 LI-COR IR2 automated DNA sequencer (LI-COR, Biosciences, Lincoln, NE) in a denaturing 25 cm, 0.2 mm thick, 7 % acrylamide gel [7 M urea, 1x TBE (89 mM Tris-base; 89 mM boric acid; 2 mM EDTA), 3 mM ammonium persulfate, 4.4 mM TEMED] run at 1,500 V 122 at4 10C Sag: ch51 311a Suga toanz Obuun greenh 7'h WA at 45 °C for 4 h. Gel images were collected, saved as image files and TDFs were scored to obtain binary datasets of polymorphic expression (band presence/absence) with the SagaMx AF LP Analysis Software (LI-COR, Biosciences, Lincoln, NE). Hierarchical cluster analysis using Pearson correlation similarity matrix was performed with Cluster 3.0 and data visualized with Java TreeView 1.0.8 (Eisen et al., 1998). Sugar beet root development cDNA Library A it Uni-ZAP XR Vector cDNA library was used (Amplicon Express, Pullman, WA) to analyze gene expression during early root developmental stages. The library was obtained from 5 mg of total RNA derived from USH20 (4.5 mg) and SR96 (0.5 mg) lines greenhouse-grown during 2003, each representing equal amounts of 3'“, 4‘“, 5‘”, 6‘“, and 7"I WAE samples. Enrichment of differentially expressed genes A Suppression Subtractive Hybridization (SSH) technique (Diatchenko et al., 1996) was employed to enrich for differentially regulated transcripts between samples, using the PCR-Select cDNA Subtraction Kit (Cat. No. K1804-1, DB Biosciences Clontech, Palo Alto, CA) according to manufacturer’s recommendations. Two different RNA samples were used for the SSH: a 3rd WAE sample obtained pooling equal amounts of total RNA from USH20 and SR96 lines greenhouse-grown in 2003, and a 7th WAE sample obtained pooling equal amounts of total RNA from USH20 and SR96 lines greenhouse-grown in 2003. Two different and independent transcript enrichments were performed: one that only enriched transcripts expressed at the 3rd but not at 7‘h WAE (early root development 123 enric nota 1001 WAE Rsal (teste subdi' hybrit conce. them 1 ”HBO the set fTom tl hrbn'tl ends v Prhner develo the 7‘h Were I reSlLSp. deVeIo enriched transcripts), and the other that only enriched transcripts expressed at the 7‘“ but not at 3rd WAE (late root development enriched transcripts). Specifically, to obtain early root development (3rd WAE) enriched transcripts, 2 pg of mRNA of each 3rd and 7‘“ WAE were used for ds-cDNA synthesis. Both 3rd and 7m WAE cDNA samples were RsaI digested, but adaptors ligation was only performed on the 3rd WAE digested cDNA (tester cDNA) but not on the 7th WAE digested cDNA (driver cDNA). Tester cDNA was subdivided in two pools and each pool was ligated to a different adaptor. A first hybridization was then performed using an excess of driver cDNA (IO-fold higher concentration) to each pool of tester cDNA, by heat-denaturing the samples and allowing them to anneal. Most of the transcripts common to both samples formed ds-cDNA, while transcripts only present in the tester samples mainly remained as ss-cDNA, available for the second hybridization, which was performed mixing together the two samples derived from the first hybridization with new excess of denatured driver cDNA. After the second hybridization new differentially expressed ds-DNA hybrid with different adaptors at each ends were formed in solution, which were selectively amplified using adaptors-specific primers with two consecutive PCR amplifications. Similarly, to obtain late root development (7"' WAE) enriched transcripts, the same procedure was performed using the 7th WAE sample as tester cDNA and the 3"I WAE sample as driver cDNA. Samples were phenol: chlrophorrn: isoamyl alcohol (25:24:1) purified, ethanol precipitated and resuspended in 30 [L sterile RNase- and DNase-free water. Early and late root development enriched transcripts were used as probes for cDNA library hybridization. 124 Libr Tl plaqu (Stratz tetraC) L"; pl SUppler MgSOr °C for 1 concentr dilutions ODémt‘ Library hybridization and cDNA sequencing The cDNA primary library was titered to estimate the original concentration of plaques forming units (pfu) following the Uni-ZAP® XR Premade Library Protocol (Stratagene, La Jolla, CA). Ecoli host bacteria XLl-Blue MRF' strain was grown on LB- tetracycline agar plate [NaCl 10 g L"; tryptone 10 g L"; yeast extract 5 g L"; agar 20 g L"; pH 7.0; tetracycline 15 mg L"] at 37 °C overnight, and 50 mL of LB broth with supplement [NaCl 10 g L"; tryptone 10 g L"; yeast extract 5 g L"; pH 7.0; 10 mM MgSO;; 0.2 % (w v") maltose] were inoculated with a single colony and incubated at 37 °C for 6 h. Cells were precipitated and resuspended in 10 mM MgSOr solution at a concentration equal to an optical density at 600 nm (ODooo) of 0.5. Four ten-fold dilutions of the primary library were incubated with 200 uL of XLl-Blue MRF' cells at or)600 = 0.5 for 15 min at 37 °c and plated on NZY agar [NaCl 5 g L"; Mgso. - 7 H20 2 g L"; yeast extract 5 g L"; NZ amine 10 g L"; agar 15 g L"; pH 7.5] plates with 15 uL 0.5 M IPTG and so uL X-gal (250 mg mL"). Bacteria were grown overnight at 37 °c and white and blue plaques were counted. Ten microliters of primary library, representing an estimate of 25,500 pfu, were plated with XLl-Blue MRF' strain cells on ten 150 mm diameter NZY agar Petri dishes (Cat. No. 25384-326, VWR, Bristol, CT) and grown at 37 °C overnight, obtaining an average density of ~15 pfu cm". Plaque lifts was performed with Nytran® 0.2 pm nylon membranes (Cat. No. 78112, Schleicher & Schuell, Keene, NH) placing a dry membrane onto the NZY agar for 5 min. A denaturing step was performed by placing the membrane face-up on 3MM whatmat filter papers (Whatrnan International, Kent, England) pre-soaked in 1.5 M NaCl and 0.5 M NaOH solution for 5 min. A neutralizing step was performed transferring the 125 inert (pll [12.115 SSC? for 3( order membrane face-up on 3MM wathman papers pre-soaked in 1.5 M NaCl and 0.5 M Tris (pH 8.0) solution for 5 min, followed by a final rising step where the membrane was transfer face-up on 3MM wathman papers pre-soaked in 0.1 M Tris-HCl (pH 7.5) and 2x SSC solution for an additional 5 min. Nucleic acids were UV cross-linked (0.12 J cm'2 for 30 s) to the membranes. Two plaque lifts were performed per each Petri dish. In order to facilitate plaque transfer to the membrane during the second lift, membranes were pre-wetted the a 5x SSC solution, dried on 3MM wathman papers for 3 min and plaque transfer from agar was allowed for 10 min. NZY agar plates were stored at 4 °C for post-hybridization plaques collection. Early and late root development enriched transcripts obtained from SSH technique were 32P labeled and used as probes for membrane hybridization. Amplified transcripts (100 ng) were denatured at 100 °C for 5 min and 32P was incorporated in the 50 uL labeling solution [32PdCTP so pCi; dATP, dTTp and dGTP 0.6 mM; tris 1.5 mM; MgCl 0.15 mM; 2-mercaptoethanol; BSA 0.4 ng ltL"; Hepes 0.2 M; Klenow enzyme 0.05 U uL'l] containing pd(N)6 random examer (Cat. No. 27-2166-01, Amersham Pharmacia, Piscataway, NJ) as per manufacturer’s instructions. The labeling solution was incubated overnight at room temperature. Two separate reactions were performed for each of the 3" and 7th WAE enriched transcripts. Unincorporated 32PdCTP was removed from the samples by filtration and pre-hybridization was performed placing the membranes (20) in 2 L of pre-hybridization solution [6x SSPE; 6x Denhardt’s solution; 0.5 % SDS; 100 pg mL'l denatured fish sperm] at 60 °C for 20 h. Membrane hybridization was performed separately for early and late labeled enriched transcripts. The 10 membranes derived from the first plaque lifts were hybridized with 126 late t labelt deriv: WAE solutit SDSs 65 °C RPN 3C The mu 10 the p. Plaqt classes t vthich re enfiched P331611: p “C1 pH 7 pBIUScn’p] helper th Premade I eluted fro“ MR? ' SIFair a1 70 0C for urged DBl late enriched transcripts (7‘h WAE) in 100 mL pre-hybridization solution, at which 3 mL labeled probe were added, and incubated at 60 °C for 16 h. Similarly, the 10 membranes derived from the second plaque lifts were hybridized with early enriched transcripts (3“I WAE). Post-hybridization membranes washes were performed in 2x SSC and 1 % SDS solution at room temperature for 5 min, followed by a second wash in 2x SSC and l % SDS solution at 65 °C for 15 min, and a final wash in 2x SSC and 0.1 % SDS solution at 65 °C for 15 min. Membranes were exposed to X-ray films (Hyperfilmm MP, Cat No. RPN30K, Amrsham Pharmacia, Piscataway, NJ) for 7 days before film development. The number and relative positions in the NZY agar Petri dishes of plaques that hybridized to the probes were analyzed. Plaques were classified based on which kind of probe they hybridized with. Three classes were observed: 3" WAE plaques, 7‘h WAE plaques, and 3" & 7" WAE plaques, which respectively hybridized only to the 3" WAE enriched probe, only to the 7"I WAE enriched probe, or to both probes. Plaques were collected from NZY agar with sterile Pasteur pipette in 500 uL SM buffer [NaCl 5.3 g L"; MgSOa-7 H20 2 g L"; 50 mM tris- HCl pH 7.5; 0.01 % w v" gelatin] and 25 uL chloroform. Single clone excision of the pBluscript phagemid from the Uni-ZAP XR vector was performed using the ExAssist helper phage and the E. coli host bacteria SOLR strain, following the Uni-ZAP" XR Premade Library Protocol (Stratagene, La Jolla, CA). pBluscript-containing phages eluted from the collected plaques were incubated with E. coli host bacteria XLl-Blue MRF’ strain and ExAssist helper phage at 37 °C overnight. Cell suspension was heated at 70 °C for 20 rrrin, centrifuged at 1000g for 15 min, and the supernatant containing the excised pBluscript was used to inoculate E. coli host bacteria SOLR strain on LB- 127 Tecl Genl detec the identi Beta v ampicillin agar plates at 37 °C overnight. Single colonies were transfer to LB freezing media [NaCl 10 g L"; tryptone 10 g L'l; yeast extract 5 g L"; pH 7.5; 4 % glycerol; KzHPOr 36.2 mM; KH2P04 13.2 mM; sodium citrate 1.7 mM; MgSOa 4 mM; ammonium sulfate 6.8 mM; ampicillin 100 mg L"] and sequenced (Genomic Support Technology Facility at Michigan State University). Sequences were submitted to GenBank (http://www.ncbi.nlm.nih.gov/). Similarity between sequences was analyzed to detect frequency of clustering, and unique non redundant sequences were compared to the NCBI public databases (http://www.ncbi.nlm.nih.gov/) for putative gene identification (Altschul et al., 1990). Beta vulgaris cDNA microarray Sequenced cDNA clones selected with enriched transcripts were also spotted on nricroarrays and hybridized with the original non-enriched total RNA samples from early and late root developmental stages in order to verify their temporal variation in expression and quantify their relative abundance. RNA samples from two different biological replications of each 3" WAE and 7" WAE samples of both USH20 and SR96 lines greenhouse-grown during 2003 were used as probes. To spot cDNA clones on microarrays, 2 uL of the plasmid solution used for clone sequencing were used to PCR amplify cDNA inserts using T7 (5'-TAA TAC GAC TCA CTA TAG GG-3') and T3 (5'-ATT AAC CCT CAC TAA AGG GA-3') primers for 30 cycles of 94 °C denaturation (20 s), 52 °C annealing (20 s) and 72 °C extension (90 s), with initial steps at 94 °C for 2 min and last extension period at 72 °C for 5 min. Amplified DNA was precipitated in 200 11L of precipitation solution [150 mM Na 128 8C3 was 0.5 1 each In 17 m tubuli. betwet microa data an Ami: hrlichigg SMltI, microan Pins (Pro diatneter) A 10121. again“ the 3rd WAE a, two for the SR96; and 1 ICle Cau‘ OTIS acetate; 90 % ethanol] at — 80 °C for 2 h, centrifuged at 5000 g for 1 h, precipitate was washed with 200 ILL 80 % ethanol solution, centrifuged at 5000 g for 30 min, and air- dried overnight. Pellet was resuspended in 25 uL of slide-printing solution (3x SSC) and 0.5 uL was run on 1 % agarose gel to verify that only a single insert was amplified from each clone. In addition to the B. vulgaris cDNA selected from the library via subtractive probes, 17 more clones, representing different classes of polyubiquitins, ribosomal proteins, tubulirr, GAPDH, and actins, which did not show any changes in gene expression between the 3" and 7th WAE RNA samples used in the A. thaliana oligonucleotide microarray (see below), were selected as controls for spot intensity normalization during data analysis. Amplified clones were submitted to the Genomic Support Technology Facility at Michigan State University and printed in triplicate on SuperAmine slides (Product No. SMM, TeleChem International Inc., Sunnyvale, CA) using an OmniGrid 100 nricroarrayer (Genomic Solutions, Ann Arbor, MI) with eight ChipMaker Micro Spotting Pins (Product No. CMP3, TeleChem International Inc., Sunnyvale, CA). Spots (100 um diameter) were spaced 300 um (center to center) and laid out in eight 13 x 12 subarrays. A total of 8 microarrays were analyzed overall, two for comparing the 3" WAE against the 7‘“ WAE samples fiom the first biological replications of USH20; two for the 3" WAE against the 7‘h WAE samples fiom the second biological replications of USH20; two for the 3" WAE against the 7"I WAE samples from the first biological replications of SR96; and two for the 3" WAE against the 7" WAE samples from the second biological replications of SR96. For each one of these four comparisons two technical replications 129 we lab Wt con T(T 10 it over techr Corp. Label instruc M ED‘ the Q1, PE and 80% et. Wort] Cat No. COUpIing ; ”8111. A s remOVe UH was TESusp, [ramming 0 SDS in a Co were performed by inverting the cyanine dyes between samples (i.e. 3" WAE-Cy3 labeled and 7‘h WAE—Cy5 labeled for the first slide, and 3rd WAE—Cy5 labeled with 7th WAE—Cy3 labeled for the second slide). Specifically, 20 pg of total RNA were combined with 6 ug of oligo dT [an equimolar mixture of individually syntlretized T(T).,TA, T(T).,TC, and T(T).,TG] in a final 18.5 uL volume and incubated at 70 °C for 10 min. First strand cDNA synthesis with uracil-aminoallyl incorporation was performed overnight at 42 °C, using the SuperScript 11 RT Kit (Cat. No. 18064-014, Invitrogen life technologies, Carlsbad, CA) and aminoallyl-dUTP (Cat. No. A-0410, Sigma-Aldrich Corp., St. Louis, MO) in a 30 uL reaction solution following the TIGR Aminoallyl Labeling of RNA for Microarray Protocol (http://www.tigr.org) as per manufacturer’s instructions. RNA strands were hydrolyzed with 10 pL of l M NaOH and 10 uL of 0.5 M EDTA and incubatied at 70 °C for 15 min. Unincorporated d-UTP was removed with the QIAquick PCR Purification Kit (Cat. No. 28106, Qiagen, Valencia, CA), replacing PE and EB buffers from Qiagen Kit with Phosphate Wash Buffer [5 mM KPOa, pH 8.5, 80% ethanol] and Phosphate Elution Buffer [4 mM KPOa, pH 8.5], respectively. Aminoallyl-cDNA was coupled with cyanine dyes Cy3 ester and Cy5 ester (respectively Cat. No. PA23001 and PA25001, Amersham Pharmacia, Piscataway, NJ) in 9 uL coupling solution [0.1 M NazCO3, pH 9] for 1 h at room temperature and protected from light. A second purification with the QIAquick PCR Purification Kit was performed to remove unlinked dyes as manufacturer’s instructions, and labeled single strand cDNA was resuspended in 4 uL of 10 mM EDTA for slide hybridization. Pre-hybridization treatments of the microarray slides were performed with two consecutive washes in 0.1 % SDS in a Coplin jar for 2 min each, followed by two more washes in deionized filtered 130 water for 2 min and final treatment in 100 °C water for 3 min. Slides were then re- washed in deionized filtered water at room temperature and dried by centrifugation at ~ 200 g for 5 min. Slide hybridization was performed with SlideHybTM Buffer (Cat. No. 8861, Ambion, Austin, TX) following manufacturer’s recommendations. The 4 uL of resuspended labeled cDNA were denatured at 96 °C for 10 min and mixed with 70 uL of pre-warmed (65 °C) SlideHybTM Buffer. Solutions were loaded in the microarrays and hybridization was performed at 54 °C for 16 h. Three 15 min post-hybridization washing treatments at increasing stringency were performed after hybridization with low [2x SSC, 0.5 % SDS], medium [0.1x SSC, 0.2 % SDS], and high [0.1x SSC] stringency solutions. Data analysis was performed with GenePix Pro 3.0 software (Axon Instrument, Union City, CA) with the same threshold parameters used in the B. vulgaris cDNA nricroarray experiment. Normalization of clone signal intensities between microarrays was calculated using average intensities of the 17 controls. Slide hybridization was detected with Affymetrix 428 Array Scanner (Affymetrix, Santa Clara, CA), data were analyzed with GenePix Pro 3.0 (Axon Instruments, Union City, CA). In addition to the software default thresholds parameters used to define presence versus absence of hybridization on each printed spot, three more threshold parameters were employed to eliminate doubtful hybridizations: both technical replications were discarded from further analysis (i) if in at least one replication both Cy3 and Cy5 mean signal intensities were lower than 1000 units, or (ii) if they were lower than the double of the Cy3 and Cy5 background median signal intensities, or (iii) if the area of both Cy3 and CyS with mean signal intensities lower than background median signal intensities plus 131 one background standard deviation were higher than 50 % of the spot surfaces. For those spots with positive hybridization, fold-level differences of gene expression between early and late root development were estimated from the ratios of the intensities of the dye signals per each gene. These differences were calculated as the ratio between the Cy5 hybridization mean intensity (CySH...) minus the Cy5 background median intensity (CySBmed) and the Cy3 hybridization mean intensity (Cy3flm) minus the Cy3 background median intensity (Cy3gmed) as follow: (CySHm - CySBM) / (Cy3Hm — Cy33med) if ratio was 21 or - 1 /[(Cy5Hm — CySBmed) / (Cy3r;m — Cy3w)] if ratio was <1 Individual clone average signal hybridization intensity and standard deviation for confidence intervals estimation were calculated from the 6 replications (3 spots per each of the 2 slides) of the same biological replication. Clones were considered present if hybridization was observed in at least one of the four treatments (two biological replications per each sugar beet line), with at least two of the six replicated spots per biological replication (3 spots per each of the 2 slides with dye swap) with signal hybridization intensity higher than threshold values. Within each of the four biological replications, genes were considered differentially expressed if the lower C190 was _>_ 1.1-fold expression difference between the two developmental stages (up-regulated genes during active sucrose accumulation), or if the higher C190 was 5 -l.1-fold expression difference (down-regulated genes during active sucrose accumulation). Experimental-wise, genes were considered differentially expressed if they showed at least two of the four biological replications with significant 132 differences in the level of gene expression. If a gene showed one biological replication differentially expressed in one developmental phase, it was considered as variable (V) if another replication showed opposite variation of gene expression in the same developmental phase, or it was considered as not changed (NC) if none of the other biological replications showed significant signal hybridizations. To summarize line-specific and overall variation of gene expression, two estimates of differential gene expression (DGE) were performed. For each clone, differential gene expression was estimated independently for each USH20 and SR96 (DGEtim) averaging the level of expression of the two biological replications, as follows: DGEline = [(tl X DGEI) ‘1’ 02 X [30132)] / (ti +12) where t; and t; are the number of technical replications of the first and second biological replications of a specific line, and DGE. and DGE; are the samples fold difference of gene expression of the first and second biological replications of the same line. For each clone a mean differential gene expression (DGE...) was also estimated combining data from the two sugar beet line as follow: DGEm = [(tlineA X DGEIineA) + (tlinen X DGEIineB)] / (11ch + tunes) where tit.“ and tuner; are the total number of technical replications, of both biological replications = (t; + t;), of the first and the second beet lines respectively, and the DGEtm and DGEHMB are respectively the differential gene expression of the first and the second beet lines, as estimated in the previous equation. Functional classification of differentially expressed genes was performed using the Munich Information Center for Protein Sequences database (http://mips.gtf.de). 133 Arabidopsis thaliana cDNA microarray Complementary DNA Microarrays carrying 12,580 PCR-amplified and purified cDNA derived from Arabidopsis thaliana developing seeds (White et al., 2000) were provided by the Genomic Support Technology Facility at Michigan State University. Of these 12,580 cDNA, 5791 identified transcripts with known or putative functions, 2322 with proteins with unknown functions, 3572 with hypothetical proteins, and 895 had no sequence similarity to other known nucleotide sequences in public databases. Sugar beet total RNA samples from the 3" and 7"I WAE of the USH20 line greenhouse-grown during 2002 were used to hybridize the A. thaliana cDNA nricroarrays. RNA samples preparation and nricroarray pre-hybridization, hybridization and washes were performed as described for the cDNA Beta vulgaris nricroarray experiment, with the exception for the hybridization temperature that was set at 48 °C for 16 h. Data normalization was performed using the Statistics for Microarray Analysis Software 0.5 (http://www.stat.berkeley.edu) and the R Statistical Package (http://lib.stat. cmu.edu/R/CRAN) (Yang et al., 2002). Fold expression differences at both 3" and 7th WAE were averaged between technical replications and 90 % confidence intervals (Clog) for the mean fold difference was estimated with the two-sided t-test. Genes were considered differentially expressed if the lower C190 was 2 1.1-fold expression difference between the two developmental stages (up-regulated genes during active sucrose accumulation), or if the higher C190 was 5 -1.1- fold expression difference (down-regulated genes during active sucrose accumulation). In order to comparatively estimate the quantity of transcripts from the hybridization intensity signals (based on the 428 Array Scanner scale ranging from 0 to 65,000 signal 134 units), clones were subdivided as having low, medium, high, or very high hybridization signal, if the average hybridization mean intensities of the dyes of both replications were < than 2000, between 2000 and 4000, between 4000 and 10,000, or > 10,000 signal units, respectively. Functional classification of differentially expressed genes was performed using the Munich Information Center for Protein Sequences database (http://mips.gtf.de). Arabidopsis thaliana oligonucleotide microarray Sugar beet total RNA samples from the 3" WAE and the 7" WAE of the USH20 line greenhouse-grown during 2003 were used to hybridize A. thaliana oligonucleotide nricroarrays following the Affirnetrix GeneChip" Expression Analysis Protocol. Specifically, 16 pg of purified total RNA were used for ds-cDNA synthesis using the SuperScript 11 Double Stranded cDNA Synthesis Kit (Cat. No. 11917-010, Invitrogen life technologies, Carlsbad, CA) and the T7-Oligo(dT) Promoter Primer Kit (Cat. No. 900375, Affymetrix, Santa Clara, CA) as manufacturer’s recommendations. Complementary DNA was purified adding an equal volume (162 pL) of phenol: chlrophorm: isoamyl alcohol (25:24: 1) (Cat. No. 15593-031, Invitrogen life technologies, Carlsbad, CA) and separating the phases in Phase Lock Gel Heavy 2 mL (Cat. No. 955154045, Eppendorf, Hamburg, Germany). Complementary DNA was ethanol precipitated adding 0.5 volume (81 pL) of 7.5 M ammonium acetate, 3.5 volumes (567 pL) of ice-cold absolute ethanol and 4 pL of 40x glycogen (Cat. No. 901393, Roche, Penzberg, Germany), and resuspended in 12 pL RNase-free water. Complementary RNA (cRNA) was synthesized from the purified ds-cDNA template with T7 RNA polymerase and biotin-labeled nucleotides using the BioArraym Higtheld1m RNA Transcript 135 Labeling Kit (Cat. No. 42655-10, ENZO Life Sciences, Farmingdale, NY) as manufacturer’s recommendations. Complementary RNA was purified with RNeasy® Kit (Cat. No. 74104, Qiagen, Valencia, CA), quantified with RiboGreenO RNA Quantification Kit (Cat. No. R-11490, Molecular Probes, Eugene, OR) and cRNA quality was controlled on Reliant® pre-casted 1.25 % agarose gels (Cat. No. R-54948, Cambrex, East Rutherford, NJ) as manufacturer’s instructions. Hybridization of the fragmented cRNA was performed at the Genomic Support Technology Facility at Michigan State University using the Arabidopsis Genome ATHl Array (Cat. No. 900385, Affymetrix, Santa Clara, CA). Per each 3" and 7" WAE cRN A sample a single array was used. Data analysis was performed using two different approaches. Positive hybridizations and changes in gene expression were derived (i) using the Affymetrix Suite software (Affymetrix, Santa Clara, CA) with default threshold parameters optimized for A. thaliana transcripts hybridization ((11 = 0.05; at; = 0.065; 71L = yrH = 0.0045 ; 12L = yzH = 0.006 ; ‘t = 0.015; d =l.l; TGT = 100), and (ii) using a non-conventional analysis based on the level and the ratio of probe set signals hybridization of the two microarrays to estimate presence and changes of expression between samples. For the non-conventional analysis, normalization of probe set signal intensities between arrays was calculated using average intensities of 64 internal controls. Probe set was considered present if signal hybridization was _>_ 400, and hybridization strength was estimated as low, medium, high and very high, if signal hybridization was between 400 to 500, between 500 to 600, between 600 to 1000, or > 1000 units, respectively. Functional classification of differentially expressed genes was performed using the Munich Information Center for Protein Sequences database (http://mips.gtf.de). 136 Results A phenotypic analysis of greenhouse and field grown sugar beet lines during the early developmental phases was performed to understand the dynamics of root sucrose content and to identify critical stages for initiation of sucrose accumulation. A Gene expression analysis (cDNA-AF LP) was performed simultaneously with the phenotypic analysis to characterize gene expression profiles during the early developmental phases and to correlate the dynamics of sucrose content with gene expression changes. Identification of genes differentially expressed between developmental phases critical for sucrose accumulation was performed using a Beta vulgaris cDNA microarray whose sequences were selected by subtractive hybridization of a sugar beet root developmental cDNA library, and Arabidopsis thaliana cDNA and oligonucleotides nricroarrays. Phenotypic analysis All the five sugar beet lines grown under greenhouse conditions for a period of nine WAE in 2002 showed the same trend of sucrose accumulation expressed as sucrose content as fresh weight (% SucF W) (Figure 4.1). Sucrose accumulated in roots fiom the 3" to the 7" WAE, increasing from less than 1 % to more than 10 % SucF W. Differences (p < 0.05) between lines were observed at the 9"I WAE between the higher sucrose accumulator line SR96 (13.5 % SucFW) and the lower sucrose accumulator line USH20 (10.7 % SucF W). This difference of root sucrose content as fresh weight was influenced more by differences in root dry matter content (% DM) between lines rather than differences in sucrose content as dry matter (% SucDM) (Figures 4.2 and 4.3). Root dry matter showed the same trend of sucrose accumulation as fresh weight, and varied 137 from an average of 7 % DM at the 2"d WAE to an average of 22 % DM at the 9m WAE, when a significant difference (p < 0.05) between SR96 (23.3 % DM) and USH20 (19.5 % DM) was observed (Figure 4.2). Conversely, sucrose accumulation expressed as dry matter varied from an average of 5 % SucDM at the 2"“I WAE to an average of 56 % SucDM at the 9”I WAE, and no significant differences were observed between lines (Figure 4.3). For the 2003 greenhouse experiment only the two extremes SR96 and USH20 lines were analyzed, using two biological replications per line, and specifically focusing on the period between the 3" and the 7th WAE. A similar trend of sucrose accumulation per fresh weight basis was observed in 2003 with respect to the previous year, with sucrose content as fresh weight increasing from 1 % SucF W at the 3" WAE to more than 12 % SucF W at the 7" WAE (Figure 4.4). The level of root sucrose content reached at the 7" WAE in 2003 was similar to the level reached at the 9‘II WAE in 2002, but in 2003 no difference was observed between lines (Figure 4.4). Similarly to the results obtained in 2002, SR96 accumulated significantly more root dry matter (23.8 % DM) with respect to USH20 (21.5 % DM) at the 7‘” WAE in 2003 (Figure 4.5). The two parental lines of the mapping population were also analyzed in the 2003 greenhouse experiment to analyze phenotypic differences between parent lines of the population used for genetic analyses (see Chapter 2 of this Dissertation). The sugar beet line C869 had similar trends and levels of dry matter and sucrose accumulation, both expressed as fresh weight and as dry matter, with respect to the two other sugar beet lines, but showed a significant lower root dry matter content (20.1 % DM) at the 7" WAE with respect to them. The table beet line W357B accumulated significantly less (p < 138 0.001) sucrose as fresh weight (8.4 % SucFW at the 7‘h WAE) with respect to the sugar beet lines, determined by a decrease of both root dry matter content (18.2 % DM at the 7‘h WAE) and sucrose content as dry matter (46.4 % SucDM at the 7th WAE) (Figures 4.4, 4.5 and 4.6). To evaluate levels of root dry matter and sucrose accumulation in field-grown plants and to compare these levels with the results obtained in greenhouse-grown conditions, SR96, USH20, C869 and W357B lines were field-grown during 2003 and analyzed weekly from the 3" to the 20" WAE. Root sucrose content as fresh weight increased fiom ~ 2 % in all lines at the 3rd WAE to between 16 % and 18 % SucFW at the 20‘h WAE in the sugar beet lines and to 9 % SucFW in the table beet line (Figure 4.7). No differences were observed between sugar beet lines, which showed higher (p < 0.05) sucrose content as fresh weight compared to the table beet line starting at the 6" WAE. Root dry matter increased from ~ 12 % DM at the 3'd WAE to between 23 % and 25 % DM at the 20‘“ WAE in the sugar beet lines and to 16 % DM in the table beet line (Figure 4.8), with a significant difference between sugar and table beet lines but no significant difference between the sugar beet lines. Similarly, sucrose content as dry matter basis increased from an average of ~ 13 % SucDM at the 3" WAE to between 70 % and 75 % SucDM at the 20"I WAE in the sugar beet lines and to 55 % SucDM in the table beet line (Figure 4.9), with a significant difference between sugar and table beet lines but no significant difference between the sugar beet lines. Overall, greenhouse-grown plants predicted in the early stage of development % DM and % SucFW traits values observed in field-grown plants late in the season. Values observed at the 9"I WAE in 2002 and at the 7"I WAE in 2003, were usually detected 139 between the 13th and the 16" WAE in field-grown plants for both % DM and % SucFW traits. Conversely, values of sucrose content as dry matter in greenhouse-grown plants were indicative of similar values observed in field-grown plants after similar growing periods. cDNA-AFLP Differential gene expression analyses of root tissues sampled from the l" to the 9'" WAE during 2002 showed that the 134 different primer combinations yielded a total of 3302 transcript-derived fragments (TDF s) (24.6 TDFs per primer combination) (Table 4.1). A total of 2181 (67 %) TDFs did not show any differential level of expression between samples (monomorphic TDFs), while 1121 (33%) showed a difference (absence versus presence) of expression in at least two of the nine samples analyzed (polymorphic TDF s). Overall, the total number of TDFs per week decreased from an average of ~2900 in the first three WAE to an average of ~2700 fi'om the 4" to the 9"I WAE (Figure 4.10). Of the polymorphic fiagments, 199 (6.0%) were expressed in at least two samples of the first five WAE and never observed later (early-expressed TDFs), 109 (3.3 %) were expressed in at least two samples in the last five WAE during active sucrose accumulation and never expressed earlier (lately -expressed TDFs), while 50 (1.5 %) were expressed in at least two samples between the 3" and 7" WAE but never during the first or last two weeks of analysis (transitionally expressed TDFs) (Figure 4.10). Cluster analyses based on similarity of gene expression profiles of the different periods of root development showed continuous changes from the l" to the 5“I WAE and from the 6‘" to the 9" WAE, with a marked shift in gene expression between the 5" and 140 the 6" WAE (Figure 4.11). The eight different primer combinations analyzed on the two biological replications of USH20 and SR96 between the 3" and 7" WAE in 2003, overall confirmed the results observed in the 2002, and further estimated the potential differences between lines. SR96 showed a higher number of TDFs and polymorphism with respect to USH20. The eight Tan/Msel primer combinations used in 2003 yielded a total of 457 TDFs, of which 373 were common between lines, 34 were specific for USH20 and 50 were specific for SR96. A total of 127 TDFs (27.8 %) were polymorphic, of which 80 were common between lines, 11 specific for USH20 and 36 specific for SR96. On average, only 4.2 % of TDFs showed a differences in expression between biological replications (presence and absence of TDFs in the two biological replications) for both lines. Clustering analysis confirmed the same trend of gene expression observed in 2002, with a shift of expression between the 4th and the 6" WAE, corresponding with the phenotypic increase of root sucrose accmnulation. Library hybridization and cDNA sequencing The root cDNA library showed an average insert size of 1.2 kb and an estimated titer of 2.55 x 106 pfu mL", of which 2.25 x 106 pfu mL'1 were recombinant. A total of 966 plaques hybridized to the SSH probes and were collected and their inserts sequenced. Of these, 587 hybridized only to early enriched transcripts (3rd WAE), 313 hybridized only to late enrich transcripts (7" WAE), while 66 hybridized to both probes. Sequence quality controls revealed that a total of 710 inserts were successfully sequenced, showing an average reading length of 769 (SD i156) nucleotides, while the other clones mainly matched vector sequences or presented very short readings. These 710 sequences were 141 deposited in GeneBank and represented 442 (GeneBank No. CV301334 to CV301775), 219 (GeneBank No. CV301776 to CV301994), and 49 (GeneBank No. CV301285 to CV301333) transcripts that hybridized only with early (3rd WAE), only with late (7th WAE), or with both developmental phases enriched probes, respectively. To estimate confidence of sequence data some of the clones were sequenced in duplicate or triplicate, obtaining 63 sequences fi'om 29 clones. Considering these 29 sequences plus the 647 sequences for which single-pass sequence was performed, clustering analysis revealed that of these 676 transcripts, 267 grouped in 84 clusters, ranging from 2 to 14 sequences per cluster (Table 4.2). Sixty-two of these sequences grouped in 5 clusters representing transcripts that mostly hybridized with the 3" WAE (glycine-rich RNA-binding protein, jacalin lectin family protein similar to agglutinin, elongation factor l-alpha, and an unknown protein) or mostly with the 7'h WAE (pentatricopeptide (PPR) repeat-containing protein) enriched probes (Table 4.2). Other than the 84 transcripts represented by the 267 clustered sequences, another 247, 137, and 25 non-clustered (singletons) transcripts hybridized to the 3", 7", or to both subtractive enriched probes respectively, giving a total of 493 non-redundant (clustered + singletons) sequences. Beta vulgaris cDNA microarray The overall level of clone hybridization in the subtracted library-derived microarray showed that of the 676 B. vulgaris individual clones, 551 (81.5 %) were detected in the original total RNA samples (Table 4.3). The majority of the clones (366 = 54.1%) were detected in all four biological replications, while only 21 clones (3.1 %) were only detected in one biological replication and were not considered for differential gene 142 expression analysis. No difference in the frequency of detected versus undetected was observed between the 267 clustering sequences (217 = 81.3 %) and the 409 singleton sequences (334 = 81.7 %). However, a higher proportion of clustering sequences hybridized to all four biological replications (178/217 = 82.0 %) with respect to the singleton ones (188/334 = 56.3 %) (Table 4.3). Of the 551 clones detected, 95 (17.2 %) and 110 (20.0%) were respectively up- and down-regulated during active sucrose accumulation, while 21 (3.8 %) clones had variable pattern of gene expression between biological replications, and 325 (59.0 %) did not show any significant variation of expression (Table 4.4). Most of the differentially expressed clones hybridized to all four replications, 88 (92.6 %) 99 (90.0 %) and 20 (95.2 %) for the up-, down-, and variable-regulated respectively, while only 159 (48.9 %) of those clones that did not show any variation of expression hybridized to all replications (Table 4.4). Transcripts that hybridized with early and late root development enriched probes were expected to respectively show down- and up-regulation during active sucrose accumulation in the subtracted library microarray experiment. The comparison of the expression levels of differentially expressed genes with the classes of subtracted probe the clone hybridized with, indicated a good accuracy of the SSH technique, particularly on clones that hybridized to at least three biological replications (T able 4.5). Specifically, of the 95 up-regulated clones during active sucrose accumulation, 61 (48 + 13 = 64.2 %) hybridized to late root development enriched transcripts (7" WAE), while of the 110 down-regulated clones during root development, 93 (87 + 6 = 84.5 %) hybridized to early root development enriched transcripts (3" WAE), as expected. The 143 proportions increase to 47/60 (78.3 %) for up-regulated and to 57/63 (90.5 %) for down- regulated genes if considering only clones that hybridized to at least three biological replications. Interestingly, 38 transcripts, 12 up-regulated and 26 down-regulated during root development, were differentially expressed only in SR96, while only 2 transcripts (up-regulated) were specific of USH20 (Table 4.5). The 95 up-regulated clones during active sucrose accumulation represented a total of 55 non-redundant sequences, 23 derived from 62 clustered sequences, and from 32 singleton sequences (Tables 4.6 and 4.7). Of the clustered sequences 12 sequences, which all hybridized with the 7th WAE enriched probes, were represented by the pentatricopeptide (PPR) repeat-containing protein cluster, showing an average level of 2.6-fold up-regulation during late root deve10pment (Table 4.6). Similarly, all sequences of the donnancy/auxin-repressed family protein (five sequences 6.9-fold up-regulated), of the nucleoporin family protein (four sequences 7.4-fold up-regulated), of the DNA repair- recombination protein (four sequences 2.8-fold up-regulated), of the dormancy associated protein (two sequences 4.5-fold up-regulated), and of the DCl domain-containing protein (two sequences 4-fold up-regulated) only hybridized with the late root development enriched probes (Table 4.2). The level of expression of the singleton sequences (Table 4.7) ranged from 1.1- to 6.3- fold up-regulation. Two of these sequences matched a pentatricopeptide (PPR) repeat- containing protein (1.6-fold up-regulated) and a DCl domain-containing protein (2.1-fold up-regulated) with different sequences respect those that were clustered. Three ATPases were also detected showing a 1.2-, 1.8-, and 2.4-fold up-regulation during active sucrose accumulation. 144 The 110 down-regulated clones represented a total of 65 non-redundant sequences; 36 derived from 80 clustered sequences, and from 29 singleton sequences (Tables 4.8 and 4.9). Of the clustered sequences 11, 9 and 8 sequences, which mainly hybridized with the 3" WAE enriched probes, represented an unknown protein (3-fold down-regulated), a jacalin lectin family protein (2.4-fold down-regulated), and a glycine-rich RNA-binding protein (2-fold down-regulated), respectively (Table 4.8). Similarly, all sequences of the PRLI factor (two sequences 8.6-fold down-regulated), of the no apical meristem family protein (two sequences 5.3-fold down-regulated), of the Bet v I allergen family protein (two sequences 4.2-fold down-regulated), of the C2 domain-containing protein (four sequences 2.5-fold down-regulated), of the glycine-rich protein (three sequences 2.1-fold down-regulated), and of the 608 ribosomal protein (three sequences 1.6-fold down- regulated) only hybridized with the early root development enriched probes (Table 4.2). The level of down regulation of the singleton sequences ranged from 1.2- to 5-fold differences (Table 4.9). Sucrose synthase (SUSl), a gene directly involved in sucrose metabolism, showed a 1.4-fold down-regulation during root development. Nine sequences matched ribosomal subunit proteins ranging from 1.4- to 2-fold down- regulation, while five genes involved in osmotic regulation (Cation-Cl cotransporter; cation efflux family; osmotin—like protein; dehydratation-responsive protein; and a dessication-responsive protein) were strongly down regulated during active sucrose accumulation. Overall, during the first seven weeks of root developmental, expression of genes involved in cell cycle and osmotic regulation were reduced, while genes involved in transcriptional regulation, signal transduction, metabolite transport and energy 145 metabolism were activated during sucrose accumulation phases (Figure 4.12). Differently, genes involved in carbohydrate and protein metabolism were constantly active during root development. A high proportion of genes down-regulated (37.8 %) and up-regulated (26.5 %) during root development had unknown function (Figure 4.12). Arabidopsis thaliana cDNA microarray Of the 12,580 A. thaliana cDNA clones printed on the arrays, 6697 (53.2 %) showed hybridization with probes derived from B. vulgaris transcripts, of which 1956 (15.5 %), 2081 (16.5 %), 1799 (14.3 %), and 861 (6.9 %) had low, medium, high, and very high hybridization signals, respectively (Table 4.10). Overall, a total of 332 differentially expressed clones were observed of which 120 (1.8 %) and 212 (3.2 %) clones were up- and down-regulated during active sucrose accumulation respectively, while 6365 (95.0 %) did not show any significant variation of expression between samples. Of the 120 clones up-regulated at the 7th WAE, only 47 (39.2 %) had transcripts with known or putative functions, while 29 (24.2 %), 27 (22.5 %), and 17 (14.1 %) matched transcripts with proteins with unknown functions, hypothetical proteins, and with nucleotide with no sequence similarity with other sequences in public databases, respectively (Tables 4.10 and 4.11). Similarly, of the 212 clones down-regulated at the 7th WAE, only 106 (50.0 %) had transcripts with known or putative functions, while 35 (16.5 %), 50 (23.6 %), and 21 (9.9 %) matched transcripts with proteins with unknown functions, hypothetical proteins, and with nucleotides with no sequence similarity with other sequences in public databases, respectively (Tables 4.10 and 4.12). 146 Considering the levels of transcript hybridization, of the 120 clones up-regulated during late root development, only 43 had transcripts with low hybridization signals, while the other 77 (64.2 %) matched transcripts with at least medium hybridization signals (Tables 4.10 and 4.11). Similarly, of the 212 clones down-regulated during late root development, only 53 had transcripts with low hybridization signals, while the other 159 (75.0 %) matched transcripts with at least a medium hybridization signals (Tables 4.10 and 4.12). Similar proportions were observed considering the 6365 clones that did not show any change of expression between early and late root developmental stages, and an higher than expected frequency was observed for clones with very high (835 = 13.1 %) hybridization signal (Table 4.10). However, of the total 861 clones with very high hybridization intensity, 68 showed a too high hybridization signal to be used to calculate relative ratios (saturation of the detector), and gene expression levels were considered as unchanged between samples (Table 4.13). Functional classification of differentially expressed genes revealed that genes involved in cell cycle, osmotic regulation, signal transduction and stress responses were down- regulated during root development, while genes of carbohydrate, protein and lipid metabolisms and of transcriptional regulation were weakly up-regulated during active sucrose accumulation (Figure 4.13). A very high proportion of down-regulated (50.9 %) and up-regulated (60.0 %) genes during root development had unknown function. Furthermore, the level of fold difference of gene expression variation between early and late root development was very low, with the majority of transcripts being down- regulated (84.4 %) or up-regulated (80.8 %) only less than 1.4-fold difference (Figure 4.14). 147 Arabidopsis thaliana oligonucleotide microarray The analysis performed using default threshold parameters showed that of the 22,746 probe sets present in the ATHl array, 1056 (4.6 %), 228 (1.0%), and 21,462 (94.4 %) were respectively identified as present, marginal and absent. Of those (1284) identified as present or marginal, 1234 (96.1 %) did not show any change of expression, while only 4 (0.3 %) and 46 (3.6 %) probe sets showed an increase and decrease of expression during active sucrose accumulation, respectively (Table 4.14). Three of the four up- regulated genes showed more than a lO-fold difference expression between samples (NADH dehydrogenase D3, cytochrome b6-f complex subunit V, and an unknown protein), while a ribosomal protein L2 showed a 4-fold up-regulation (Table 4.14). Differently, the level of variation between the 46 down-regulated genes was not higher than 2.5 fold difference (Table 4.14). Using the non-conventional analysis based on the level of probe set hybridization signals, 845 (3.8 %) and 21,901 (96.2 %) transcripts were respectively identified as present (if present in at least one of the samples) or absent (Table 4.15). Of those transcripts identified as present, only 224 (26.5 %) were present in both 3" ad 7“I WAE samples, while 375 (44.4 %) and 246 (29.1 %) were only present during the early and late phases of root development, respectively (Table 4.15). In order to estimate differences of level of gene expression, of the 845 signals detected, 274 and 184 transcripts were discarded from the analysis because only present at low signals hybridization in one of the two samples (Table 4.15). Of the remaining 387 probes, at signal hybridization ratio threshold values of > 1.2 for up-regulated and < -1.2 for down-regulated genes, 105 and 148 164 transcripts showed increase and decrease of expression during active sucrose accumulation, respectively (Tables 4.16 and 4.17; Figure 4.15). Differently from the results obtained with default analysis, most (96/105 = 91.4 %) of the differentially up- regulated transcripts showed level of expression smaller than 4-fold difference, and only two of the four previously detected up-regulated genes (NADH dehydrogenase D3 and cytochrome b6-f complex, subunit V) with conventional analysis were also identified as present and up-regulated with the non-conventional analysis (Table 4.16). Functional classification of differentially expressed genes revealed that genes involved in cell cycle and signal transduction were down-regulated during root development, while genes of osmotic regulation and of transcriptional regulation were weakly up-regulated during root development (Figure 4.16). A relatively high proportion of down-regulated (37.8 %) and up-regulated (31.4 %) genes during root development had unknown function. 149 l4 .. ...................................................................................................................... a O SR87 T __ O SR96 12 1 ..... .v. ..... USHZO .......................................................................... M .. .............. Q V SR95 ' e; 10 1 ..... y ..... SR97 .............................................................................. .' .............. . 4°30 . SR96 fitted line 'g — USH20 fitted line 8 .. ...................................................................... “I .............................................. g 0 d: a 6 .. ................................................................ " ................................................ 8 8 s 4 .. .................................................. ‘ ................................................................ m I 2 .. ..................................... L. ................................................................................ 3 0 I I I I T I I I 2 3 4 5 6 7 8 9 Week after emergence Figure 4.1. Sucrose content as fresh weight of five sugar beet lines (SR87, SR96, USH20, SR95, and SR97) grown under greenhouse condition for a period of nine weeks after emergence during 2002. Bars represent :1: standard deviations. 150 25q ................................................................................................................................... o SR87 o SR96 . v USH20 T v sass i . 20 ..( ..... l. ...... S 3.9.7 ....................................................................................... u .............. SR96 fitted line gas —- USH20 fitted line I ' . v l-r " is a ,, . .......................................................... g .............................................................. ‘2' ad ,0 . ............................. .g ....................................................................................... 5 I I l I T I I r 2 3 4 s 6 7 8 9 Week after emergence Figure 4.2. Root dry matter content of five sugar beet lines (SR87, SR96, USH20, SR95, and SR97) grown under greenhouse condition for a period of nine weeks after emergence during 2002. Bars represent i standard deviations. 151 O SR87 60 -( ..... 6 ..... S R96 ................................................. . ................................ i ............. i V USH20 V SR95 ? so .. ..... .. ..... S R97. ................................................................................................. é SR96 fitted line 33 -- USH20 fitted line E 40 .. ...................................................... ; .......................................................... IS 30 .. ..................................................................................................................... 8 o (I) a I :1 20 .. ...................................... , ................................................................................ m 10 a ..................... 1 ............................................................................................ 0 I I I r I f I I Week after emergence Figure 4.3. Sucrose content as dry matter of five sugar beet lines (SR87, SR96, USH20, SR95, and SR97) grown under greenhouse condition for a period of nine weeks after emergence during 2002. Bars represent :1: standard deviations. 152 14 .4 ........... .C869 ...................................................................................... O SR96 12 a ........... YUSHZO ......................................................................... V W357B § SR96 fitted line / 2’ 1o .... ----USH20fittedline ............................................................. jgn — — — C869 fitted line / 0 W357B fitted line 3 8 . .................................................................................. 7.1.4 ...................... E 8 6 .. ................................................................... . / ............. y ........................... 3 8 3 4 . .................................................... / ............................................................ m 2 .. ......................... I ...................................................................................... o I I I I I 3 4 5 6 7 Week after emergence Figure 4.4. Sucrose content as fresh weight of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W3 57B) grown under greenhouse condition for a period of seven weeks after emergence during 2003. Bars represent :1: standard deviations. 153 25.. ......................................................................................................................... W357B .1. 20 q .......................................................................................... USHZO fitted line — — — C869 fitted line - W357B fitted line ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Root dry matter (%) ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo 3 4 5 6 7 Week after emergence Figure 4.5. Root dry matter content of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W357B) grown under greenhouse condition for a period of seven weeks after emergence during 2003. Bars represent :t standard deviations. 154 0 C869 60 . ........... .0. ............. $396. ..................................................................................... V USH20 V W357B 50 . ........................... USHZOfittedlme .................................. <3 SR96 fitted line 2; — — — C869 fitted line H . g 40 w ........................... W357}? fitted 1.1.09. ............. 5 30 a 8 o m g :3 20 a m 10 1 o I I I I I 3 4 5 6 7 Week afler emergence Figure 4.6. Sucrose content as dry matter of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W357B) grown under greenhouse condition for a period of seven weeks afier emergence during 2003. Bars represent i standard deviations. 155 0 C869 0 SR96 20... .......... I ............. USH20 .................................................................................. v W357B , SR96fittedline z; USH20 fitted line H 9. e, -— — — 0869 fitted line / --«—-' .. H 15 d .......................... “/3573 fittéd'fihé ........................ l... ..................... g V I” 7 o O " o / 3 . ' O 7?: - ’ ch 10... ....................................... . ..N/.v ........................................................ i. ‘6’ v E E E T 8 o ' / V V . ‘7 i E . i m 5.... ..... a ......... v ................. i ....................................................... I/ v v I 0 I I I I I I I l T I I I I I I I I I 3 4 5 6 7 8 9 10 ll 12 l3 14 15 l6 17 18 19 20 Week after emergence Figure 4.7. Sucrose content as fresh weight of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W3 57B) grown under field condition for a period of 20 weeks afier emergence during 2003. Bars represent i standard deviations. 156 3oq ......................................................................................................................... 0 C869 C) SR96 ‘V IJSHDO V W357B 25 . .......................... SR96..fi.t.ted.l.i.ne .............................................. 13.... . USH20 fitted line 0 .... " i" — — -— C869 fitted line V 0.. ' Q W357B fitted line - a, J 8 20a ................................................. Y.....T .......... l ..................................... 3: ' I 4. a / ' V $ é . . . . ............. r'......u....vl.Y .................. 2 ......ii .......... e i V. -* 4' V 5 I I I I I I I I I I I I I I f I I I 34567891011121314151617181920 Week afier emergence Figure 4.8. Root dry matter content of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W357B) grown under field condition for a period of 20 weeks after emergence during 2003. Bars represent as standard deviations. 157 80d ........................................................................................................................ ‘ 0 6 a I I‘ I, .‘I.-—-""'*“‘—' A .. ............................... .6...‘:.. . .’.. ..... §3 60 J ‘,p 23 3E g: T: T ‘ v 0 -- é; I-o 4” V' 1! .. ' ‘V E as " V V i .5 4o ... ....................... $ .............................. a ................................................. 33 .- 0 C869 g __ o SR96 ‘6 2 v USH20 5 20 ._ ; v W7? ........... q I ................................................................................ SR96fi-tt-edlcine- USH20fittedline — — — C869 fitted line W357Bfittedline 0 I I I I I I I I I r I I I I I I I I 34S67891011121314151617181920 Figure 4.9. Sucrose content as dry matter of three sugar beet lines (SR96, USH20, and C869) and a table beet line (W357B) grown under field condition for a period of 20 weeks after emergence during 2003. Bars represent i standard deviations. 158 QVN odN 9mm M.M.N mam mdw QVN Wow 90.. «.2 Q3 adw mNN v.2. Yon Qmm 0.3 omflm>< Q? B dd dd dd dd dd dd dd dd dd dd 9 3 mm 3 3. ._.<0 98 mm 8 t «N 9 on mu «N om vw mm mm 9 av mm Nv 00< Qmm R 9 mu 5 R «N on 9 _. F or m— mm mm um mm mm 004. v.9. m vw cm 5 NF 9 mm t. 9 9 5 m o_. m? 3 E. ._.o< v.2. 3 mu 2. mm mm mm mm m 9 m 9 mm F F um 5 mm 006. 0.8 2. Q mm 3 av mm 3 or a. N_. 9 mm m? ..n 3 3. oo< 08 mm mm «N or m? R E 9 5 3 mm mm or nu «N on < .E. 0... 0... (... .60 00 00 <0 ...0 00 00 <0 ._.< 0< 06. << 82630:: 03828 .8: deducted do: u dd .mdewmfiafioo 38E m4...?Ttov¢dtom§u 0290538 5: 161 - use. 92 oz _ o N N _quN 655: _Nm seen .583: 8... 82.33. _ o N N Navmoa 532a 22E: ”census 25% 883?. c o N N 3mg 2 .25. aged wssscséaoc Na< 2 $33. a c m N 8.9: E805 endgame. £293 £395 sotéfiubm 9.69.va c o n m 8+mvN 52% 3:5. 32 05.585 83ng a o m n 8.86 See States; 3:55-528 Nm< SNWNNNE o c N N NE: 529a 832%... 8:33. o o N m 8.92 §_§b_o2§ 285554.592 \ 3:33 $25855“ 8932 a o N m 23:: some 03385 egaseofiofimofi 2 a» :< o o N M 2-8.... 8ng 2 532a 3583: 223 m8 83.33. c c N m «N73. 82.3 2 582; 3583: 228 m8 8N8N0< a o n m SE: £38.. £5 €2.85... 80%» _.< v o o e 8+m. v 823: seen ..ozusassgia do $23 5D madtomfib donuts accede—26v 88 8m— 55 Am<3 Emv fidtomdg vosotdo “define—gov 80.. been 53» douuamvtnxn nag—d :0 v3.3 vogue—om 203 Viv—8:8... duh—m5 downs—o moose—.63 Eton—ac. .m.v 28¢. 162 NYmoN 9.5.385 8.5.582 <25 . 5325 5.55 5.5. N 885 8528 .888 888mg .. a N N N38. ANNE. 38.88.58...» 33.3.83 . 82.8.... 383 8.882 c c N N N25... §§=.NE.NO§..2..N€§_N§5-N ..NNN .82 e o N N No-8.N ”35v. 528.. ..NNNNNEN c o N N ...-ch 92.8 :5 522.. .25 <30. :8... 2.5 23.32 c c N N 5...... £55. 83:03. . 03.955 058852..-.8825 . 085888588... 83282 e o N N 8+8... 320.882.8558... 25.3 . 2.82. 688.28.388.85 059..» ..NNN 32... c o N N N38... 620. 532.. N5u5n. 3 883.2 a c N N 8.ch 522.. 852%... 8.832 c c N N 8+..NN 522.. M55.55535558 .8 8.8%.... e c N N 2-8.N 5.85. 28.8583 . .558... 532.. 83293 cNNNNm..< .. c N N 35.3 353.2. NN.. 532.. .882... m8 8; .32 a c N N ....m... 825...: ...). 532.. 5885. 8.. 28%.... N c o N N33 532.. 5.55 .555.... . > 8.. 883.2 N c . N 37...... 5.5.... 5.... 33...... Nv ”858.82. 8... N32 N a . N 52-..... .59.. 35.83.... 8.58%... 55532 ..NNNNNNE N c . N ....moN ..mz... . 3.8.082... 83.53 N o . N 8.8... 52.8 ....55 852 3.088 5.8.5.58. 8323... . . . N 8+8... N 03.58 883.2 . a N N N NQN N 523. 2.2. 88.0 5 a. 8.3.555. 8.888 35>... 8.5.885 :88... .V 85.3. ..N 3% .o .382 3.283 NV 035,—. 163 8+&c.N 522.. 353258 .8 23ng N o o N 3&3 2528 $33 Sosa 3383-338 88%.? N o c N 3&3 seamen: _Nmszfmééz 2 3:5 seem N :33 3.2023 3.4.83 SEN»? N o c N 3&3 £395 225 3a: 8:33. _ _ o N 3&3 Ed 522; £23 233.2% 83:25? 0:22; cN_NNNu< o N c N 2 7&3 $.25: 53% 3355 322.5 8333. _ a _ N 8&3 .& €38 93...? 22% «N asafisi 522.. 05:853.?“ 2 RN»? _ o _ N : 7&3 55$ N easy. 05883 8233 _ c _ N N2-&¢._ 2.385 _ 5395 a9. 2 2E8 «.85 .8: 8232‘ _ c _ N 2&3 828. 3:35 8.6a 2 2% 5.0 2 52% 832.52% 9:83 :28. 83.32 _ a _ N «2&3 522.. 3308338383. 82% same 5:33:22. 8532 _ c _ N 8&3 532; £3 58225 8332 _ o _ N 21%... 532.. 2.3 3:828 83525.5: £338? 0.5.... 2 $33 _ o _ N 3.9... 23%: :3 522.. 1.5ch m8 c8233 _ c _ N 2&3. 522.. Biameéifiem 2.9332 _ o _ N 8&3 U fi— n._ _d E d _d n._ d n.N _d n.n n nd a; : awn—dng a; n._ _d _._ d _d n._ d d.N dd d.n n nd d._ d «dN_dn>U Q— d._ _d n._ d dd m; d d.N Nd ~.n n wd ..N d n _n_dn>U d.— h; dd :4 n nd a: d N.N a; d.N n nd d.N : onNn_dn>U d.N d; dd n. _ d d. _ _.N n n.N ad d.N N dd n.N d .NNn _dn>U Q— n.— _d N._ d Nd r... d d.N _d n.n n md d._ d _mw_dn>U din—3?. n_ 389:: 052: Eur— 058:8 ”odd—3:0 .953 030.30:— 5035 :03 :00 5:05.05 .§0:600 30.50803 522.. 5.35.2: «:0. «on 00051.3:53850 2 3:5: :22: £5: E: 532: 5:5: 29.0023 833.35 88.2: 0 d.N n.N dd _.n d hd c: d 0N ad d.N n ... d.N d .hNd-dn>U n.n d.N nd d.N d dd N.N n d.N :._ n.N n - 0N _ odd—dng d.N n.N nd d.N n n._ ..N d d.N Nd d.n n pd _.N d nhd_dn>U d.N n.N dd d.N d dd d.N d d.N dd d.N n n: d.N m ode_dn>U r." d.N d._ ..n 0 nd h: d n.n 5.: dd N d._ d.n d d5.dn>U d.N d.N dd _.n 0 dd 5.. : d.n d._ n.n n d: d.n d Nan—$50 ”..N n .N ad ..n n dd d: : n.n dd nd n d._ d.n d 3d_dn>u d.N N.N nd d.N 0 5d a: d n.n ad ad n n: n.n : d_d_dn>U d.n n.N nd d.N d nd Na— d d.n «N _.n n n._ d.n d nun—dn>o d.n d.N dd n.n 0 ed d._ d n.n w: dd n n._ d.n d ndw_dn>U d.n n.N dd d.N : dd N... d d.n d._ dd n w: n.n : n.n—dn>U fin nN md d.n d dd d.N n n.n d._ dd n a: N.n d 32dn>o n.n n.n d.N d.n d wd d.N n ..n _d d.n N ... d.n n dnd.dn>o anhanE 30:2 MEA— unnfldmm 0E2: 50.5 0:338 50.2: $3550.38“: 95.: 0::do::0£8:0: N— umu: 1:00: 25 ~00: N. 30 .00: _. _aasmc: 30 £0: N. 30a J00: : _ E 0.5080 e ...: 8:0 2:0: 53:80: 88.0 * :2003: .0: 832:5 :: 030:3: :_ 30:03:00 00:20 0308:: 30:034.: £3202: 00:09.38 .\. dd 8303:: 850" ..N—3:00:00: .00:: ::008 0.: :5 3.6 05 :0 :00 05 0.3 02=mOQ :5 <8..mOQ 05 :5 $333300: 0.0:: ::0000. 05 :5 3:: 05 do 2230:“?— 3030.03 53 E 0:33:30: 32:58 :0 Baas: .88 0.: 0.3 08.: :5 <85 20:3 A085 + 65$ \ 20858 x 02:5 + A5258 x <85: n 5:8 ”30:0: .3 0:: 30: 3mg. :25 05 Ed: :3: $35800 :038300 0:31:09 5300308 0:0w 330.0%: :02: a 0:20 :030 :0: .002 0:: 089. 05 :0 823230.. 303205 ::0000 :5 E: 05 no :200298 0:0» :0 00:93:: 20.: 003800 05 0.3 $8 :5 7&8 :5: .m:om30:d0: 3030—2: ::0000 :5 an: 05 do 0:230:30: 32:52 :0 89:5: 05 03 N. :5 a 20:3 .3 + 5 \ Eman— x 5 + d.n—OD x J: u .....mOQ ”030:8 m: .m:o_30::0._ 3030.03 03. 05 :0 5600.508 do 35. 05 w:_w80>18__m0ndm0:: ~00: 3w:m ddxm ::: szmD 50: .5: $532.35 33.530 003 5302950 0:0w _:_::0:0.c_: 6:20 :30 3.: fiatowswb :03_:w0._.:: :20320 Swamp—0E 33.5: 3:00.53. 33M? 30% dd 030,—. 168 ~ In. ..llAhE-JK Ch \V. :2 if ~ 50:55.0: .02 90:20 i $0.259. 6% 030g .2 S 3 S n .3 3 o 3 a... 3 m 2. 3 0 8285 3 3 2 3 v S 3 n 3 .... 2 N 2 3 m 2235 3 q. 3 S o 3 2 n 3 3 3 n f 3 m c2285 93832 @532 ”.3832: A .33 QB .8585 _ 3.83232 33% 38.253: 2 3.82 525 _ 88.8:3t m 3 3 8 3 o 3 3 e 3 3 ..m n 3 E o 5390 3 3 3 3 o 3 S o 3 E 3 m 3 E e 3285 3 3 3 3 o S 2 e 3 2 3 m 3 3 o 2235 823.2 _§__§ 3.9.823. 5.. Z 08 3282.. S 8832 3385 2 Eucfi 35:. as: Somme 3.. 3.322. m I 3 3 N..- e ..e 2 e 3 to 3 n 2. 2 o 3:85 3 2 .... ... o ..o 3 e 3 3 ...M m I. 2 o 33:56 3 S 3 I e S 2 e 3 ..o 3 n 3 3 o 8:85 2 E ..c 2 e 3 3 o 3 no 3 m I. a. e 3890 22.3.2 3: 8.36. sea «32. _ 8 522.. 555.2,. 2 3.2% 52% 2:5. ..ucééea v 3 3 3 3 e 3 E e 3 2 3 m ... 3 o 2285 3 3 3 3 o 3 3 o 3 3 S. m 3 3 a $390 3 3 3 3 e S 3 o 3 3 3 m I 3 o 8285 3 3 S 3 o S. S o 3 3 3. m 2 3 e 3890 2233 .333. Eqfifii 52. 3:: .20 522.. =ou8_neo§.._aa 6 N.N o N- m _ cm>U 958 a: 02 _ o." m.N 56 ed 0 “no _.N 0 9m N._ 9m m ... EN c 8:230 958 a: 02 _ a; m._ _d N; o _d v; o YN #6 9n n v6 a; o wan—250 a.— w._ Nd w; e ... ...N v _.N v._ vN m «6 6N o n.n—$50 cN_NNwm~< m 388.. 3:95 ENE 23:8 6:4:me 353 030603: 589:. =9: :8 snug—PE ..N—3:338 50an83 52% soc2__eq «9. 3m mmuiuzi _ $805 9 has; 532.. gee Ed 522; £35 29.22% 8853.2 88.2.. N N.N N.N - N.N _ 3. 3. m N.N 2 N.N N 3. nN m 5.390 n.n n.n 3 3. e N._ ..N o 3 n.n 3 N S N.N c 93826 2.3.22 532.. 32:? £2025 62.8": 0E9... N .5: 23:8 "3&5 33:2. $3855“ .838... e 532.. Engagesasec 38:5 2 3:5: 2532. £32; u>_§&2.=_=8£em N.N ..N N.N S N 2. N.N v 2N S N.N N - S _ N385 E n.n 3 e.“ o S ....n o 3 9N 3 N 9N N.“ m .8890 c8332 2: Man—m o-Cen Eat 9:855 58th MEESCOZQEOQ —UQ N 3 .... 3 3.. o 2 N.N e a.“ «.2 3 N ...N N.“ v 3890 5 3. N._ E e 2 N.N o 3. 3. n.n N N._ 3. c 82950 833.2 82.— 3869.8 5.3335:an 633mm “...-«95 En: 25:3 ”—5323“. .555 Eat Samoan—O E295 3369.3 -.mucascov 2 3:87. ”83330 _uzuzci ammonia—x. £329 vofiuommaAocaNEo—v 2 18:52 2329 2583 .5395 “98383935....3 —.— A: «6 e; n 06 o. _u e m._ md ...N m ed _. _u 0 8n _ 350 N.— N._ md m: o m6 «.7 e N._ ad md m md QT c .82350 n; ~._ 2. o; n 56 m. T v n._ c.— n.n m 56 N. _ - 9 5n _ 350 n.— N._ m6 n; n 56 _._- v V._ ad nd n 06 _._- o 1.2.—ans o3¢~w§< Sagan—0 _unwaui aaouuafii rag—my £32; £23... 05.5an «Ema—a 8 33:67. mach» bo> 2583 .539... 2E5 2.25:8:— aEuaE n amen _flumooaa Nmoo N. a6 Eon : _aasmoo flu «moo fl. a? Nmos 4 #838 a ...w 8% 8:3 83:35 88.0 * 9.283 .o.v 035. 170 {Cl . s f: flaccuc QM] :59 cm. 5%: 5:3,?»an «0:30! 9.30% 6% 03$ I S 3 S n 3 2 m 3 to ..N n 3 2 m 3.890 8&3? 522m 2 3:8... ”532; 33.335828 3769 2:82 53338 _ ... E 3 3 o 2. 2 o 2 I. S m I. S m 2:85 @8223. Fags: ESEEE see 3585 522; 9.39:8 £58 «.2 2 _8_.=8_ E .85: N. .25 532; 2.5589528 «.2 _ N..- _ .3890 833?. 8.8323» Ea 282,.» fimm 2 3:5 "_uzufiuto: 355: hmwnwnvua 324D. £38m 25582: 3.3988 083.05. 3 3:8? ”Ens—ow ugh-Bog _ 33:50 £32m 38239 Nu— w._ ed 9; m 06 _.N m N. T v: v: v: a; 9N md a; my ad _.N o N..— v._ N.N m «ac v; n 33250 0.58 :2 OZ _ S ...N 3 3 o S 3 e 2 to mm m I. 2 o 8:85 882: oz _ a; o._ vd ”A e 56 ..N e a; a; «..N m ed 0.. o «me—cm>0 v5.8 5.— oZ _ ...N S 3 S n I. ..N m N.N ed N.N m 2. 2 3 R285 £332 .253” .5 382.23.. 25%.» déafig.~xase§__% sesbmfiaofié : 2.3 om: 8%..» Renghofiéw—Bm 3 3:82 828» _ do: _afimondfi «mom a. flu doc : _aazmoa a6 «moo N. an: ion _. .950 e ...w 8% 8sz Santana 88.9 * 9.288 .cé 035. 171 .o.v 939 5,. ms cocgau .mEtumcsb was—:wnziz 23532.5 ”52305:: ASE: $32253 h2§3§ 3:5 N .V 935‘ ...N 3 - ”..N _ 3 n.n m M: ..o N.N N to Z N 3285 8833 389: $5 §fi< saga .5: 25:8 532,. 3:5. 8...? ab.<<< ...N a.” Z 3 o S 3 c 1 I. ..N m S N... m 2285 8332 £32.. ham \ <5. 5 258 8&3. ”889... .528 a: 588.: 362.. E: 25:8 ”_§u~§~ afififii 52. 8222c 3232. hamioe E9. 3 .8352 Anzme 3:33 .532; $333. n.n w.~ _._ QM ~ m... n.n m a; a: v: v: wd a; m 332.50 cos—nan? _ 5&2 .m 25 oxzémmcaum 3393 .m £82m agent—m 338.3 .8383 3.8%: 03908.3 382m ”mo—c2; Ea: «58:8 5205 miss—_ooéflfio—v m 58an 3 a.” S 2 o 3 ..N c ..N a... 3 n 3 N... ... 8:36 25.22 XNSE 52.8 $5 fieoweom 33:23.2 9.8.2.. a“: £28 :2. 835 5%»: SEN. BE. 2 3:5; 522.. 2:5. Ed 532.. 58.5 29322» 833.3%. 9.8.2.. n.n ed 06 “..N e m._ 1N n n.n «6 EN n N; ad v 3.2230 883.2 .3623. finnehBEE _wvmmnéumc £896 565?: 8 .8252 £829 vommoaxm a." ...N We N.m 9 rd N..— o n.n n6 mi m A: «N 0 $3330 958 «2 oz fin cN _._ n.n m z «Q N n.n e; a; N in ed N chm—230 gnaw—2 use; 0533208 <5: 5.29: 2%... SE 23:8 x3332 3: 68:22.. :22: §...~... om: < 3.2.; Sarasaaa <72, 3:385 2 3E 52.5 £5. 35% usufiogaa <5: n.n ow. n... oi N n.n 9n n 9n n.n 2. m - ~2— _ 2.2350 c3232 V5.33 3.8.: 833229 5395 3.3088 hose." 03.80.50:— 8385 3035.5ch o... 8:8? $395 33353 nd fie a; ad. m EN 3. 0 m6 _.m We m 9N 9m o entom>u v5.8 .2 c2 amen _flwmoa 26 Nmoo N. 26 _moa .. _aadmoo a6 “moo N. flu Amoo : A 9 359.5 a. .Q 8% 8%: 8258a .cé 038 8m mm 5930 .mfitomcab “98253.33 cage—wan $8.820?— bfin: vogombasm £333, Sum saw 03.; 172 :25: Iii Cnv~rr ~ affix a :u. :53 {(1 N. .v 036K. a; a; 06 EN n Y: n._ e - v: v: v: v: v: 2. Sn Sm>0 agnwflt‘ 82:8... .336 39: SN. :8: 85:5 u 3.2: 522$ 52% EEEQN A: z 2 5E: 333 cages... u as: 532.. \ 33.... £22.. NEEfiEN N... N.N 3 ..N n N... N.N v .... . N.N _ N... .... N N585 2832 35.2. “_aoafihs 52.. SREHmo 2885 520 7223.2...“ 5.33 2 5:82 55: 520 $223.51... 5.3.; ...N _.N ..o N.N v No 2 N S - 9N _ 3 N._ N 8225 33mg :25 28 Sofia... 68.83: ”Enoch—m oan ENE 23:8 £3330 55.335 .5333. 6892.20 .82 333.5 Eat «3.283: 0239 2 3:5? 02.82. dun—cab. 02me N.N 9N 2 ...N c 3 ..N v ... Z 2 N E N..- N 3285 8.8»? 52.8 maisézo a? $88.. “scan .5: 25:8 28:6 .98.. Sites: :25 a: N.N ...N no N.N v 3 M: e ON a... N.N N Na ...N N 82:90 8333 .353. afifizui 3N8 532.. Ngvcsézé 8382.. £3225 seeaoozész A: N.N $255 2 .8252 $8.. N 8 530:. Nscczézx \Eaeozo sseafiscst we. % n.n ed ed n.n n 0.: a; 0 Wm 56 ed m .2 ON 0 mac—oma consumz FEES: ...NEQEQEE 57:6 .293: Emu—.5: <~ £38m 3.1653230:— Swngupc.8_3w 2 .8252 23.3: 93.8.6 .5085 50:028—363. Ya ..m 3 3. v «6 w; v ~._ 2. 0.. N Nd ...- m chmEM>U 32.3w? £085 382qu v.« N.N vd Om n v._ a; v «N 2 ad m 9.. Wm N 55.350 c833? 5085 commoaxm b; 5. W: N..— m ad N... m b.— m._ mN a Nd N... v .3235 ¢nom~wfl< v." n; n .c n.~ v ed 2 n ..m N._ 9v m 56 «N e 38 Sm>U onew~w9< .5588 555.5 mo 52:8 U 08:: £289 Euwbzv a .5256 _ 0Q £32: Ewe... 2i 0:.— 333 «58:8 MES: Ema: uEN GI.— xwoB 25:8 35:3: 383%»: Eat «Kant—G 589:. .3358»: 3 5:63 5085 33350-5250 .00 amen _flumooaa «mom N. a6 _moa : _gmoo a6 68 N. 30 .moo : 325950 e ...w 8% 85: SEE @253 .5. use. 173 .... E 3 E m 3 S n .... . n... _ S ...- n £225 88%.2 .853. ESEEE $933.9 .8592a 3.532.. ..qu 2 5E... fi§§§ 33828:. 53:53 7858.... mm“... 88.2.. 2 52% 322.888: 1 3 no 2 o S 2 e ... 3 2 m ..o :2 on 8285 c3232 5:92 @332 _w 2 _3u§_ 50$ =_2eq.§8_&o§e< v.— 9.. nd a; m 2. v._ o «4 v6 9N N N6 2. n .3 Sm>0 annomwnz _Efisa. 23:. Eat 382.30 _ £82m xgwct 3 3:87. Baa—2é_o§n x3.w=_m 3 2 ..o ... m 3 2 e ...N o... 2 n 3 3 o 8236 833.2 3&2 Em ”was“: “.52a .5: 23:8 522.. €523.22 2%: 8.52.8335 9; n._ m6 N... 9 m6 «... n 5.. n.n Q. N v6 0.. n can 5n>0 3.33mi _mavwaifl aura: 3.53% :8: 0:89.? .omfioEomm 82.305385 9 bra—.87. macaw 03.82. 6:83? .omEuES_ Susana—308$ 3 2 E n... e 2. E o S 3 3 n 2. 3 o $225 38»? «mafifioxamim .35: RE 8.533 E 883...... 2 i.e.“ ”Sanimzfinngmna .5822 833.92. 08322. 2 5:...” 3:53 guise; a; c.~ Na 9. m n.s md n m._ «2: ed m We c..- @ nae—ems 2:9.me 58.: £58 3.5.. fiduwsem Bauéefi 3385 En: 83:8 6.2.9.”.0 7.33.3 §§§BiaE 023258 «9. Z 8 3:56 5305 bind 3.5V 5085 bum—ab EgEowESm voomtognEE 3.38& K..— S nd a; m «d 9. v 2 N.N S N - ... _ n$_om>u 853mm; 03.83 .0353 £82; unfinEoEmcg 5&2 sutéfiuaoq Q. ed 2. v: c: c._ ...N v n._ c: v: v: ..o n._ m coo—$59 Begum; _Ncovm<<_nw_om_vwm _ E E5383» 83:8335 80¢ .233 «03.2: unamh< :20... 3.2.23 2 3:83 268.5 gnaw—2 .885 3.2.23 3 3 3 E n S f ... 3 S ...M m 2 2 n $285 3522 .833. 3.833: 52.. 82:3 <59. 88.2.. _25 .333. ”.3832: 323 um. 3252a <59. 03522.. ......»iu 8235 s .8382 8.32.. .25 \ 2.85 aésoa 86:6 ..mwn _aamwaaa Eon a. 35 Jon F _aawmmoa 36 «m8 N. 30 £09 A. 325950 e ...V 8% 8:3 82:85 9.303 .56. 03.; 174 E S 2 3 e 2. 2- o 2 no ..N n 8 S- o $288 831%? 8.33: ”6 .233.” :35. in 52% magpézo 2 age: 3:53 Sosa m=_2_...u «N. ...N. . ..N- . N... N... N N.N. .... .... .... .... N.N. .. 828R. N.N. N.N- N... N... o N. ..N- N N... N... ... N .... N.N- .. NNN...N>u N.N. ...N. N... N.N. N ...N N.N. N ...N. N... ...- N .... N.N- .. NNN...N>u N.N. ...N- N... N..- e ...N ...N. e N..- N... ... N N... ...N- .. NNN...N>u N.N. ...N- .... N..- . N.N ...... o N..- .... .... N .... N.N- .. 3285 ..N. ..N- N... N... . N.N ...... e ...N. N... N..- N .... N.N- o 8.890 ...N. ..N- N... N... .. N.N N.N. .. N... N... N... N .... N.N. .. 82.56 ...N. N.N. N... .... e ...N N... e N... .... .... N N... ...N. o 9.285 28.32 ESE... .58. 8:75.32. “ass"... 2:0... :3... 25:8 .mwwwNnmNna .53.»... 3.3.89. :.:..=_wwa 8 8:8... 589... 3.8.... E60. ....32 a N.N. w. _. ..c o. T n o... N.N. m 9N. a.N v.N.. N ad EN. 0 new .350 ...N. a.N. Nd o. T o «..N ma.- 9 v.N- n... w. T m a... N.N- o .8 . cm>U N.N. a.N- v... e. ... c m.N «:7 o «N. m... ...N- m a... ”N- c Sue—350 a.N. ..m- m... a. T o N.N n6. 0 n.N- v... ...N. m a... N.N- o nmo . cm>U a.Nu N.N- v... e. .. n N.N n6. e 9N. n... ...N- m o... ...N- o an 5mg a.N. ..N. Nd a. T o ...N mic o N.N. m... N.N. m a... ...N. 0 ran . 36 En. N.N. m... o. T o a.N ...? e a.N- .... N.N. N N. . N.N. n NE. 3m>0 Nd. 9m. 9... ...N. 0 ON Nd- e 9N. n6 N.N- m a... N.N. 6 can . cm>U Nd. 9m. Nd a. T o ...M n.n. o ...N. n... N.N. n ad a.N. o §.¢m>0 v.7 Em- md ...N. e ...N n.n- c ...M. 5... ON. m ad ..m- 0 ram Sm>0 :6: m6. .6 n. 7 v ...M .6. o n.n- 06 ...N- m v.N mi. 0 _an 3mg 8§w9< .2383... 522.. .55.... s 3......” 522.. 3...... . . 176 .muiflwmooa... don N. a... ....co . 433...... flu Nmoo N. 01.6.. _moo .. _ 9.258.. ...... 3% .55: Eats... 8...... .. o... 0.9.... ..N... mm guano .fiqtomcwb 38.53....38 3.89.... $8.32ch 5.... BEN—53 «E33. Sum .m... 035... ad- 06- ad Nun- o N.N e.a- o a.m- a.N d6- N ad nd- 0 _am.dm>0 Nd? m.N.- .... a8- 0 N.v d.n? o v.a- dd ed? m n6 a.n- o $m.dm>0 saga—3N ENESEN. 385...? 5.5—Boo .NNNNNN... 2%... ...N... 2.2.8 uNNNNN. . ...N. .353... N332: .. .88.. Egg-...... ... 5...... 03.3.... .. .32.. guess-...N... N..- N... N... N... N N.. N.N. N N... N... N... N N... N... N .8590 N..- N..- N... N..- N N... N..- N N..- N.. N..- N N... N... N NNN...N>u N..- N..- N... N... N N... ...N- N N..- .... N... N N... N..- N NNN...N>u ...NNNN..< .NN.N.N. NNNN._.N .38.st 83...? as... 95.5. «...... £32.. .2.ch ... 3.8... ...82 82...... N... 532.. .383: NNN N..- N... N... ... N N... ...N- N N... N... ... N N... N... N NNN...N>u N..- N... N... N..- N N... N.N- N N..- N... N..- N N... N.N- N NNN...N>u N..- N... .... N... N .... N.N- N N..- .... .... N N... N.N. N 2285 ..NNN.NN.< NNNNSNNNONNNEZ. .2358 582.... ..N: 2......N 23:8 .NNNSNNENN .338. Not-8.2.0. 2......8 .NNNNNN.<<.NN_NNNN.NN_.N €33.38» =8.E§3. 538.. N.N-5835 2 ....EN 522.. $5. ...N-5...» . ...-...N... ...N... .39. ...N-2.63.. ...N- ...N- N... N..- N N... N.N- N N..- N... ... N N... N.N- N NNN...N>o ”N. N.N. N... N..- N N... N.N. N N.N- N... ....- N N.. N.N- N NNN...N>u N.N. ..N- N... N..- N N... N.N. N N.N- N... .... N N.. N.N. N 3385 ..NNNNNN.< NN.NNNNNN... €83... $.36 Sec... 8.2%.. N=2... 532.. ...N-2.3.... N..- N..- N... N..- N N... N..- N ... .... .... .... ...N ... N .585 N..- N... N... N... N N... ...N. N N... N... .... N N... N... N N.N...N>u N..- N..- N... N... N N... ...N- N N..- .... ...- N N... N... N NNN...N>... 8NN..NN.< ...N... 8.26. NNNNNNHN... .... sec... .382... 0.28 NNN 2:32. 2 .....e... 89...... ..N. 522.. .382... 222. NNN 3. N... N... N.. N N... N... N N..- N... ... N N... N..- N 2235 N.N. N.N. N... N... N ..N N.N. N N.N- N... ...N- N N... N.N. N .3890 N.N. N.N- N... N... N N.. ...N- N N.N- N... N.N- N N... N.N. N 8285 ...N.- N.N- N... N..- N N.N N.N. N N.N- N... ..N. N N... N.N. N NNN...N>u ..NNNNNN.< .3526... 2.53939. 8...... 39.0.2.6 £32.. 82...... 9 5.6.... ...-32.. w...:.3:3-=.«Eou NU ...-5.. gmgafi qNon. N. a... doc N. 13.-amen. a6 don .1 a? Joe .. 7.25%... 9...... NNNN 85.. 8.3.8.8.. 8...... N 8.503 ”..N. a.N—u... 177 a... ...N- a... a..- N ..N. _ m..- - ... _ v... N..- N 32950 5.7 ...N.- nd N..- n a... N.N- o n..- n... m. N n... a...- o oar-8.50 oamtmmz NNNNNo.No....-NN.3N .58... 3:35.... N 8505...? .0203. .N onEo-chgimocova 2.2.2502. N 8305...? aficoififiewocogm 8 SEEN 03.8.... ...N-805...? o=_=o_:8E_Nmo:ova-w N N..- N..- .... N..- N N... N..- N N..- N... ... N N... ..N- N N.N...N>... N..- N..- N... N..- N N.. ..N- N N..- .... ....- N N... N... N 32.90 ...N. .3... ...N... 8N. ..NNN......o N.N 522.. 558...... NNN ... 5...... 8N.N..... N.N ......E 558...... N..N N N..- N..- N... N..- N N... ..N- N N..- N... N.. N N... N..- N 2285 ...N- N..- N... N... N N... N.N. N ...N- N... .... N N... N.N. N NNN...N>o NNNNNNN... 32...... 553.2... .2228... 2...... .35. 555.... NNNN.N..No....NN.>NN ... 50.25... 2......-... . 2...... ...N... 855...... N N..- N..- N... N... N .... N.N. N N..- N... N... N .... ..N- N NNN...N>u N.N- N.N- N... N... N N.. N.N. N N... N... N... N N... N.N- N 22.90 ..NN.NNN.< .....ENN ...NE... 82......3 .NNN.......... ......58 SEE 5.5.8 .....o... N........ .5... .28...N_..5... N N.N. N.N. N.. ..N- N .... N.N. N ...N- N... N... N N... N.N. N 8285 N.N. N.N- .... N... N N.. N.N. N N.N- N... N... N N.. ..N. N NNN...N>o ...NNNNN... 25:88.... 05.25828. 8:. 880:3. ...... NNN...... 2...... .5... 8.3:... u.NN..NNNN. H.... 5.8.2.... 25......- E.... N............ 05...... E... .8558. ..NNNNNNN..N. N22858:... .353... N.NQQVEEV Sat 9......— D 58...... 982.8 85.. reggae. S ...—.5... £22.. NEE... 50:38.... EEENE 3.73.833. 3.8... N N... N.N- N... N.N- N N.. N.N. N ..N- N.N N.N. N N... N.N- N ..N...N>o ...N- N.N- N... N.N. N ...N N.N. N N.N. N.. ..N- N N.N ..N- N 5.85 .322... ....5. . > ..N. .....o... 55.2....85NN5... ”83...... ...-...... .5... 2.3.... .NNNNNNN..o=.........5 ..N. 86 ... 5...... .....o... ....5. 5N5... . > ...N. N N.N. ...N- N.. N.N. N N.. N.N. N N.N- .... .... .... N... N.N. N 8225 N.N. N.N. N. N.N. N N.. N.N- N N.N- .... N.N. N N... ...N- N 228?. ..NNNNN..< 52...... 93.2.. 508.38 .8... ..z . NNNNN... .5... ... .... .8... 2.3:... H......5 5.5. .NNNNNNN..N.. ......o... 5.2.... ... 5...... ...so... ....5. .322. 55...... .8... ..z N 178 ...N... $5885.... N80.. N. 2.6 .80.. .. .3288 ...N ...-X. N. an... 8.... .. 32585.. a. .... NN..N ..sz: 8......85 8...... ... ......8. N.N ......N fin. N.N- v.0 N.N- m m... n.n- n o. T n. n..- N Nd Wu- N non Sm>U Oman—uni. 33...... 22.2.55... .0508. 82...... 2..........2.. 5.8375538. 05...... 82...... 8.8.5.... .§8&§.N.0 2.25.3 .... .....N 00. 82.05.»? 8.§.8§T§§=.0§.eua§§§_€2¥n «880:. ... .808... 82.0. 82...... 8.826... 20000000000050.0000 \ 0005...? 00.00.50... .000000005-N_m.0_§_> \ 00800290350... 00.083052038803532802035832 N.N. ...N. 2. N.N. .. ... N.N. .. ...- .... .... N n... ...N. 0 «3.890 80.3.... 3.80.80.00.30 .320... 28.0: -90... .3....:..0......§. 28.855.020.22. .32....55 ... 5...... 02.3.... .§.p.............a ...86...§.....§.z . 2.05.... 0.00855... 2. .... m... 0... v .... ...N. ... N... m... ...M. N a... ...N. N 88890 2...... 2... ..z ..N. ..N. .... .... 0 ..N ...N- 0 8..- a... ...- ... 0... N.N. 0 0.0890 8820.... 80.23.... 2.5.0.88008... H. .88... 2...... .5... 2.5:... 0.5.3.»... 28:3... .0380... :0... ... 00. 8......35 03.20.802.22... 882.... ... 002...... 2...... 00...: 2.3.... $003.8 00.. . 2.3.... 03.0.3.8 agafiossaa... ...N. ...N. ... ...N- v 2. N.N. 0 ...N. a... v... N n... ..N. 0 5.890 8280...... N00. 880090.30 .8003. 0.83%.... .N 822....0 82...... N 820.2558... 05....st N 85.5.... 98.80.035.088... ... .808... 32%. N 85.5.... 2.0.02.2...5880 ...N. ...N. 8... ..N. .. ...... ..N. N ...N. .... .... .... . ..N- . 08890 2...... 2... ..z N... n.n. .... ...N- N N... ...N. N ...N- .... .... .... .... ...N- N .8890 2...... 2... ..z ...N. ...N- .... 2.. 0 .... n.n. 0 ...... .... n.n- N N.. N... 0 .8890 9...... a... ..z ...... ...N. n. ...N. .. .... ...... m . .... .... .... .... .... .... 5.890 830.... ............. 38...... ...... ......E 8.8.... 2...... .5... 00.3000 "wen—3.00 €000.03 00.3.0.3 02.00080 .3325 0.00.9 3 3:88 0.0.0.0 9:50.. 5.03.— . 0.0.0.0 2.0.0.. .233... 0.30.9.0 0.... 0.0.0.5. ..000 1000030 N000 N. 0.0 .000 _. _088000 0.0 N000 N. a.0« .000 4 .8253 0. .... 8.... ..sz0 8......890 8...... .. 0.....8. .... 2.3 179 N..- ....- .... N..- N .... N..- N .... N... .... N N... ....- N 3285 NNNNNNNE NNNN . 8.0 .853... ......8........... 80%... 70......888 8.288..N ......eoz. ... .895... ....o... ...-.2882... 3N.NN8N.N .3828. . N..- N..- N... N..- N .... N.N. N N... N. N..- N N... N... N NNN...N>u .....N.N:< NNDmN... ...... 6...... NS 532.. 128...... .u ..NNN.N...< ..NNNNN..o .858... ......N...N....... N.No... N 50...... 2.8... <2... ...N-8.8.3 £56. 532.. N..........-o ....NN.NN.< 33.5.8... 8.8.. .....N 2855...... 532.. 8...... 8:2...8 2......ng ... 5...... .88... N...N... 8...... 8:22...0 3.8.8.85 . N..- ..N- N... ....- N N.. N.N. N N..- N... N..- N .... N..- N NNN...N>,.. ..NNNNN...< NNNSNNEE 8......8 8.2... 522.. 853:8 .N.NNN<u 28.9.52 . ....- ..N- N... N... .. N... ...N. .- N... Ne N..- N N... N..- N 3.89.. ..NNNNNN... 883...... .. 522.. .382... NNN ... 85...... N88 23...... ....3 522.. .288... N8 . ...N- ..N- I. N..- N N... ..N. N N..- N... .... N N... N.N- N ..NN...N>o 3.8.9.2 3.....3. ......8............ .8328... 683...... Egg-.8888. .... .N.N om. 08......NNSEN...N88..< NNNNNOEN ... ......8... i=5. 8882...... . 82.5.... N=.N.NN8...N..-..-..858-N88.888.12.83 . ...N- N..- N... N..- N .... N.N- N ...N- N.. N.N. N N... N..- N 5.36 ..NNNNNN.< .NNN 0.............No HNN.N....... £25.. ...N... 8.3.8 .83... ......aN .8: E9285 . Inn... 38.-...“. N...N... 8 N40 Nmoo .. _aamoo 8.0 «moo N. 8.? _moN. .. .9893 a. ...N. NNNN 8...... 8......88. 88... ... ......8. N... N.N...N 180 ..N. N.N. n... 8... n 8.: 3- .. q:- ...8 m... n ... ...N. 8 88890 8.238.... 288:: 882: :3: 23:8 .....33. 88:32... .. a. 22:5. 8.0% :32: 8382328829 2 38:8. :32: 822:3 ..N. ...N. 8.: ..N. n :8 8.“- 8 ....- m... .... m ...: N.N. 8 2:890 8828?. @888: 882: :3: 23:8 .3838. 938.2. .§b::8<. 82.28:: 232822-2883. .28.: 0E §§§§5§< 0.2.8 ==_m:o:m2-:c_§o_moe 532: we. no 33v:_.8380::§.33 8 bra—Ea“: x33 88:8 3.88.882: o>fi§obm 2- N... ...: S n 8 E- v a..- a: 2 N 2 2. n 8285 283.2 .233: 3:93:83 :5: 2 32an 2385 520 12:32...“ 5.33 s .828. 55: 520 7223.22. 5.3.: 2- s. _- .3 o. _- m I. S- v 2 . 2 _ 3 3 n 83:90 2 :33 £32: 332%“.— 2- v. _- 8 2- o 2 2- e Z. 3 2 ... 2. h. _. o 2:85 8. 9:3 5sz :83}: E§§E$ AaeeégESBw n.n—b.8239 AmzéN 08 3:59AM 823m Eva—Em 3 bra—Ev. wcobm 258:: .goumfihimouaw :039226 \ 258:: 685:? 3.20am v..- 2- ...c ...- m 2 o..- n 2. 2 c... N no 3. n 8285 $333 8: .3 see: 283: 83%: 3 532: 3283: m8 2- a..- 3 Z- n S S v 3. ...e m... n 2 ...N- n 2285 8833 3:55: .3 E2: gaming—m d—oan—O $382: .25: <30: 8:522: 05393 2 bra—Ev. 28:8 258:: .0382: .25 \ 258:: 68582: 05295 n... S. 3 S- e I. S- v ...- I. 2. n a: ...- n $890 :88»? 8828: 428.5 x «Ewe: .5 £32.. .383: 832:. Rm see: 18%: 8.. m..- 3. S S- o 8 a..- e I- I. N... n 3 n..- n 2.23.5 83:3 322232: .§§3§ 89.5»:qu .:m 522.. .283: 9125.3 5 £32: 2582: we do: _flwmo: as do: a. a6 Am5: _h _fiamo: a6 Nmo: N. 25 dmo: 4. J E “.8950 e 2‘ 8% 85.: 8.353: 9.253 .3. 2%... 183 Down-regulated genes: Osmotic regulation 3.6% Carbohydrate met. Cell cycle 0 26.1% 2'” Protein met. 9.9% Stress response 4.5% Transport «an: ._ p 4.5% Signal transduction 0.9% Unknown function 37.8% Transcriptional regulation 9% U -re ulated enes: Stress response Pmin met 61% Energy met. I 1.2% 3.1% Carbohydrate met. 3 l"/ Transpon . 0 10.2% Signal transduction Cc" cycle 7.1% 8.2% Unknown function 26.5% Transcriptional regulation 24.5% Figure 4. 12. Beta vulgaris L. cDNA microarray analysis: Functional classification of differentially expressed genes during active sucrose accumulation. 184 Table 4.10. Arabidopsis thaliana cDNA microarray analysis: Number of transcripts and variation of gene expression during active sucrose accumulation. Gene expression Hybridization signal L M H VH Total Up—regulated Known or putative function 12 13 17 5 47 Unknown function 14 7 8 0 29 Hypothetical protein 11 9 6 1 27 No sequence similarity 6 4 4 3 17 Total 43 33 35 9 120 Down-regulated Known or putative function 25 32 43 6 106 Unknown function 8 11 13 3 35 Hypothetical protein 14 18 l 5 3 50 No sequence similarity 6 4 6 5 21 Total 53 65 77 17 212 No Changes 1860 1983 1687 835 6365 Total 1959 2081 1799 861 6697 185 Table 4.11. Arabidopsis thaliana cDNA microarray analysis: Genes up-regulated during active sucrose accumulation (7th week after emergence). Average fold up-regulation (up- reg.) was estimated averaging two technical replications and 90 % confidence intervals (:1: C190) for the mean fold difference were estimated with the two-sided t-test. Hybridization levels refers to low (L), medium (M), high (H), and very high (V H) if the average hybridization mean intensities of the dyes of both replications were < than 2000, between 2000 and 4000, between 4000 and 10,000, or > 10,000 units, respectively. Average tClgo Hybr. TIGR ID Definition Blast up-reg. level score 2.22 0.47 L No hit found - - 2.00 0.64 M At5g49190 sucrose synthase 1.15E-94 1.77 0.18 M At1g22710 putative sucrose transport protein, SUC2 1.68-180 1.73 0.38 H At1g59820 chromaflin granule ATPase 11 homolog, putative 3.98E-54 1.70 0.12 M A8362010 hypothetical protein 0 1.64 0.34 L At1g33270 protein of unknown function 0 1.56 0.21 M At2g46500 protein of unknown function 8E-123 1.51 0.27 M At5g42810 hypothetical protein 4.9E-109 1.50 0.37 M At5g44120 legumin-like protein 6.0E-142 1.49 0.05 M Atlg30720 putative reticuline oxidase-like protein 0 1.47 0.19 H At1g18300 protein of unknown function 1.1E-126 1.46 0.24 M At2g30830 putative dioxygenase 5.7E-159 1.46 0.35 M At1g72970 protein of unknown function 5.9E-168 1.45 0.11 M At4g31150 hypothetical protein 3.105-94 1.44 0.15 M No hit found - - 1.43 0.13 L No hit found - - 1.42 0.23 L At1g19660 protein of unknown function 1.19E-88 1.42 0.08 L No hit found - - 1.41 0.13 L At3g22560 hypothetical protein 0 1.41 0.25 L At5g43750 protein of unknown function 2.2E-138 1.41 0.09 L At1g22850 protein of unknown function 2.88E-78 1.40 0.23 L At1g73480 lysophospholipase homolog, putative 7.6E—146 1.40 0.09 L No hit found - - 1.38 0.03 11 At5g44380 berberine bridge enzyme-like protein 1.94E-45 1.37 0.03 M At4g31720 Transcription factor 11 homolog 1075-60 1.37 0.06 M No hit found - - 1.37 0.19 L At3g10260 protein of unknown function 2.87B-98 1.37 0.25 H At1g13390 protein of unknown function 1.67E-93 1.37 0.23 M Ath41950 hypothetical protein 1.3E-130 1 .36 0.05 H At1g59890 hypothetical protein 3. 1913-95 1.36 0.26 H At2316380 putative phosphatidylinositol phosphatidylcholine transfer. 4.1713-82 1.36 0.01 H At3g56310 alpha-galactosidase-like protein 2.91 1340 1.35 0.25 L At3g46220 hypothetical protein 838-119 1.35 0.18 L At5g16270 hypothetical protein 3.048-69 1.35 0.21 M Atlg24360 putative 3-oxoacy1 (acyl-carrier protein) reductase 3.38-116 1.33 0.02 H No hit found - - 1.33 0.18 H At4g23650 calcium-dependent protein kinase (CDPK6) 1.9E-114 186 Table 4.11. (cont’d) Average tClgo Hybr. TIGR 10 Definition Blast up-reg. level score 1.33 0.03 L At5g04750 F 1F0-ATPase inhibitor-like protein 8.3OE-81 1.32 0.04 L At1g53210 protein of unknown function 2113-12] 1.32 0.17 VH No hit found - - 1.31 0.04 M At4g35470 hypothetical protein 1.65-170 1.31 0.09 VH At1g05160 putative cytochrome P450 2308-56 1.31 0.09 M At2g21240 protein of unknown function 13725-90 1.30 0.02 H At2g04280 protein of unknown function 1.1E-1 18 1.30 0.03 H At1g64350 protein of unknown function 3078-98 1.30 0.03 VH At3g57390 MADS transcription factor-like protein 1.68-124 1.30 0.11 L Atlg74970 ribosomal protein S9, putative 0 1.30 0.10 M At5g60600 hypothetical protein 1265-54 1.30 0.02 H At5g10860 hypothetical protein 2.6E-120 1.29 0.1 1 M At5g18630 triaeylglycerol lipase-like protein 7.1E-102 1.29 0.12 L Atl g70580 putative alanine aminotransferase 1.79E-83 1.29 0.09 M At5g06240 protein of unknown function 1.1813-69 1.29 0.04 H At1g75220 integral membrane protein. putative 6.81E-42 1.29 0.05 L At3g53180 nodulin glutamate-ammonia ligase-like protein 4.58-119 1.29 0.17 M At1g78610 hypothetical protein 4.7E-155 1.29 0.13 M At2g42880 putative MAP kinase 0 1.28 0.11 M At3g11420 protein of unknown function 288-163 1.28 0.02 M At2g03690 putative ubiquinone biosynthesis protein 4.5E-165 1.27 0.06 M At1g29930 photosystem 11 type 1 chlorophyll a b binding protein 6.44E-55 1.27 0.04 H No hit found - - 1.27 0.07 L At3g22640 protein of unknown function 26813-79 1.27 0.12 L No hit found - - 1.27 0.13 H At5g13410 hypothetical protein HOE-70 1.27 0.17 H At4g01690 protoporphyrinogen oxidase 0 1.27 0.15 L At4g11420 hypothetical protein 0 1.26 0.11 H No hit found - - 1.26 0.01 H At2342680 protein of unknown function 5.398-80 1.26 0.10 L At1g10490 protein of unknown function 1925-90 1.26 0.08 L Atlg53880 protein of unknown function 1.59E-19 1.25 0.06 L At1g02640 beta-xylosidase, putative 0 1.25 0.13 VH No hit found - - 1.25 0.15 L At3g23190 protein of unknown function 1.32E-75 1.25 0.03 L At5g48180 hypothetical protein 1.513-177 1.25 0.09 H At1g18270 protein of unknown function 8.14E-68 1.24 0.01 VH At5g47720 aeetoacyl-CoA-thiolase 2.6415-43 1.24 0.05 H At1g10760 protein of unknown function 1.6213-96 1.24 0.1 l M No hit found - - 1.24 0.1 1 L At3g11900 putative amino acid transporter protein 7.7E-151 1.24 0.11 H At4g37980 cinnamyl-alcohol dehydrogenase EL13-1 1.2415-79 1.24 0.06 L A6349320 hypothetical protein 2.38-124 187 Table 4.11 (cont’d) Average tClgo Hybr. TIGR 1D Definition Blast up-reg. level score 1.23 0.06 H At3g12090 senescence-assocated protein, putative 3.0E-139 1.23 0.02 L At1g09140 putative SF2 ASF splicing modulator, Srp30 9945-76 1.22 0.05 H At5gl7400 ADP ATP translocase-like protein 662-133 1.22 0.10 L At5g13040 hypothetical protein 1.28-115 1.22 0.10 L At5g11450 hypothetical protein 268-157 1.22 0.03 M No hit found - - 1.22 0.03 L At5g06140 sorting nexin-like protein 8.03E-77 1.21 0.08 H No hit found - - 1.21 0.08 H At1g20630 hypothetical protein 0 1.21 0.11 VH No hit found - - 1.21 0.02 L At1g06570 4-hydroxyphenylpyruvate dioxygenase (H PD) 2285-57 1.21 0.03 H At2g25070 putative protein phosphatase 2C 2.062-65 1.20 0.00 L At5g42320 protein of unknoLn function 5.96E-84 1.20 0.07 L At5g14970 seed maturation-like protein 0 1.19 0.09 H At1g60940 putative serine threonine-protein kinase 1.66E-71 1.19 0.09 H At4g28520 128 cruciferin seed storage protein 5.6913-33 1.18 0.05 VH At5g62470 MYB96 transcription factor-like protein 7.4E-129 1.18 0.04 M At3g56860 hypothetical protein 2.48-166 1.18 0.04 H At1g07750 globulin-like protein 0 1.18 0.08 L No hit found - - 1.18 0.06 H At1g19680 protein of unknoLn function 1.14E-72 1.18 0.06 L At1g80960 protein of unknoLn function 4.88-116 1.17 0.03 L At3g46890 hypothetical protein 2.7E-179 1.17 0.03 L At1g50740 protein of unknoLn function 3.85-150 1.17 0.07 M At4g21160 hypothetical protein 1.0813-38 1.16 0.03 VH At4g18470 RNA helicase-like protein 2.09E-44 1.16 0.05 M At2g38960 protein of unknoLn function 1318-75 1.16 0.04 L At3g19860 putative myc-like DNA-binding protein 1.58-149 1.16 0.05 H At5g67250 hypothetical protein 2.0E-145 1.16 0.03 H At1g48380 root hairless 1 (RHLI) 0 1.15 0.05 M At1g62080 protein of unknoLn function LOB-130 1.15 0.00 L At3g53700 hypothetical protein 0 1.15 0.01 H At4g00400 hypothetical protein 0 1.14 0.01 H At4g26200 l-aminocyclopropane-l-carboxylate synthase-like protein 3.75-135 1.14 0.01 L At4g10060 hypothetical protein 1.38-167 1.13 0.01 M At5g56300 S-adenosyl-L-methioninezsalicylic acid carboxyl CHg'me. 0 1.13 0.01 H At2g01880 putative purple acid phosphatase 2.913-161 1.13 0.03 M At3g47340 glutamine-dependent asparagine synthetase 3.49E-70 1.12 0.00 L At3g16860 protein of unknoLn function 0 1 . 12 0.01 VH At1g29050 hypothetical protein 0 188 Table 4.12. Arabidopsis thaliana cDNA microarray analysis: Genes down-regulated during the late (seventh week after emergence) root development. Average fold down- regulation (down-reg.) was estimated averaging two technical replications and 90 % confidence intervals (:t C190) for the mean fold difference was estimated with the two- sided t-test. Hybridization levels refers to low (L), medium (M), high (H), and very high (V H) if the average hybridization mean intensities of the dyes of both replications were < than 2000, between 2000 and 4000, between 4000 and 10,000, or > 10,000 units, respectively. Average iClgo Hybr. TIGR lD Definition Blast down-reg. level score -1.93 0.62 L At5g14230 ankyrin-like protein 0 -1.86 0.07 VH At3g29779 pseudogene, hypothetical protein 0 -1.66 0.33 M At1g29530 hypothetical protein 713-127 -1.61 0.29 H At4g15800 OEP8 like protein 4E-70 -1.60 0.00 M At5g40140 hypothetical protein 515-162 -1.60 0.41 H At5g17170 protein of unknown function 8B-154 -l.57 0.15 H At4g30710 hypothetical protein 4E-135 -1.56 0.45 L At4g187‘70 myb-Iike protein lE-150 -1.56 0.08 M At5g35590 multicatalytic endopeptidase complex a subunit-like 113-50 -1 .54 0.32 VH At2g28740 histone H4 8B—176 -1.52 0.07 VH At5g55810 hypothetical protein 813-154 -1.50 0.23 H At5g05750 DnaJ-like protein lE-l 13 -1.50 0.18 H At4g03090 NDXl homeobox protein homolog 713-124 -1.50 0.09 M At1g50480 10-forrny1tetrahydrofolate synthetase 3E-173 -1.49 0.19 L No hit found - - -1.49 0.05 M No hit found - - -1.48 0.32 H At4g38840 auxin-induced protein-like 58-131 -1.46 0.08 H At5g02260 expansin precursor-like protein 313-163 -1.45 0.02 H At1g68010 hydroxypyruvate reductase (HPR) 0 -1.44 0.30 M At5g25590 hypothetical protein 215-151 -1.44 0.29 H At5g05360 hypothetical protein 18-122 -1.44 0.33 H At2g26100 protein of unknown function 6132-50 -1.43 0.01 H At5g15780 proline-rich protein 3E-138 -1.43 0.19 H At1g71950 protein of unknown function 113-85 -1.43 0.30 H At4g39920 hypothetical protein BEE-92 -1.42 0.07 H At1g53310 phosphoenolpyruvate carboxylase 1, putative 213-105 -1.42 0.03 H At5g48180 hypothetical protein 613-170 -1 .41 0.15 M At1g68020 putative trehalose-6-phosphate synthase 213-106 -1 .41 0. 12 H At5g39950 thioredoxin 1 E-94 -1 .41 0.22 VH No hit found . . .1 .40 0.23 H No hit found - . -1.40 0.07 M Atlg56290 protein of unknown function 313-150 -1.40 0.28 H At1g49240 actin 8 713-168 -1.37 0.14 L At3g55740 proline transporter 2 0 -1.37 0.13 H At1g03310 putative isoamylase lE-96 -1.37 0.11 H A3gL8940 putative chlorophyll a b-binding protein 513-93 189 Table 4.12. (cont’d) Average 1C1” Hybr. TIGR 1D Definition Blast down-reg. level score -1.37 0.13 L At3g26650 glyceraldehyde 3-phosphate dehydrogenase A subunit lE-60 -1.36 0.26 H At5g17060 ADP-ribosylation factor-like protein 713-90 -1.36 0.25 M At5g61820 hypothetical protein 113-68 -1.36 0.09 M At1g61040 protein of unknown function 15-171 -1.36 0.09 H At3g01700 protein of unknown function 0 -1.36 0.06 H At1g13260 DNA-binding protein RAVI 1E-99 -1.35 0.10 M At5g48000 cytochrome P450-1ike protein 313-157 -1.35 0.09 L At4g29380 hypothetical protein 0 -1.35 0.12 H At4g26580 hypothetical protein 413-113 -1.35 0.25 M At2g24280 putative prolylcarboxypeptidase 113-153 -1.34 0.00 L No hit found - - -1.34 0.05 VH Atlg7l880 sucrose transport protein SUC 1 SEE-60 -1.34 0.12 M At4g02680 hypothetical protein 58-87 -1.34 0.04 H No hit found - - -1.34 0.04 L No hit found - - -1.33 0.00 L At3g58730 v-ATPase subunit D (vATPD) 213-88 -1.33 0.19 VH At5g17020 Exportinl (XPOI) protein 0.0006 -1.33 0.05 H At5g04750 FIFO-ATPase inhibitor-like protein 2131-90 -1.33 0.07 M At1g12110 putative NPKl-related protein kinase 2 613-131 -1.32 0.21 L At1g64330 protein of unknown function 0 -1.32 0.19 H At2g41430 ERDIS protein 25-46 -1.32 0.16 M At3g29180 protein of unknown function 25-108 -1.32 0.05 M At1g74030 putative enolase 315-151 -1.32 0.05 L At3g62730 hypothetical protein 9E-170 -1.32 0.18 H No hit found - - -1.32 0.10 VH At1g62710 beta-VPE 113-114 -1.31 0.18 L At4g24160 hypothetical protein 382-142 -1 .31 0.19 L At3g06483 putative pynrvate dehydrogenase kinase 0 -1.31 0.14 L At1g27640 heat-shock protein 90, putative 913-179 - l .31 0.12 VH No hit found - - -1.31 0.21 H At4g00730 homeodornain protein AHDP 0 -1.31 0.13 M At4g31080 hypothetical protein 613-114 -1.31 0.16 H At5g42870 hypothetical protein 58-51 -1.31 0.08 H At5g05000 GTP-binding protein (gb AAD09203. 1) 981-59 -1.31 0.02 H At2g37940 protein of unknown function 3E-101 -1.31 0.19 L At5g54380 receptor-protein kinase-like protein 113-1 14 -1.31 0.15 M At2g19740 60$ ribosomal protein L31 SE-113 -1.31 0.11 H At1g67700 F12A21.16 613-146 -1.30 0.19 VH Atlg76960 protein of unknown function 713-65 -1.30 0.12 H At5g48160 hypothetical protein 313-157 -1.30 0.05 H At5g66850 MAP protein kinase 315-120 -1.30 0.08 M At3g51550 receptor-protein kinase-like protein 0 -1.30 0.09 L At5g12950 hypothetical protein 113-106 190 Table 4.12. (cont’d) Average tClgo Hybr. TIGR lD Definition Blast down-reg. level score -1.30 0.13 M At3g56130 hypothetical protein 313-91 -1.30 0.14 H At5g02890 hypothetical protein 3E-160 -l.30 0.03 M At2g18020 60$ ribosomal protein L2 28-161 -1.30 0.18 H At5g64330 non-phototropic hypocotyl 3 (gb AAF05914. 1) 0 -1.30 0.13 H At5g49840 CLP protease regulatory subunit CLPX-like 313-173 -1.30 0.16 M Atlg27l30 glutathione transferase, putative 1E-140 -1.29 0.17 L At1g52980 putative GTP-binding protein lE-56 -1.29 0.16 M At3g55440 cytosolic triosephosphatisomerase 413-39 -1.29 0.14 H At5g65760 lysosomal Pro-X carboxypeptidase ZE-132 -1.29 0.02 M At4g30660 stress responsive protein homolog lE-137 -l.29 0.03 VH No hit found - - -l.29 0.01 M AtZg46170 protein of unknown function 413-94 -1.28 0.04 L At5g61590 ethylene responsive element binding factor-like 6E-174 -1.28 0.07 L At1g10890 protein of unknown function 413-178 -1.28 0.05 VH At3gl4400 protein of unknown function 2E-59 -1 .28 0.02 H At4gl6830 nuclear antigen homolog IE-l 19 -1.28 0.13 L Atlg14920 signal response protein (GM) 0 -1.28 0.18 H No hit found - - -1.28 0.14 H At5g65620 oligopeptidase A SE-l 14 -1.28 0.00 H No hit found - - -l .28 0.10 M At2g40900 putative integral membrane protein nodulin 113-115 -l.27 0.06 H At4g15080 hypothetical protein 5E-143 -1.27 0.04 M At5g15710 hypothetical protein 0 -1.27 0.15 H At5g04340 putative c2h2 zinc finger transcription factor 3E-172 -1.27 0.13 H At5g05100 hypothetical protein 813-65 -1.27 0.13 L No hit found - - -1.27 0.04 M At3g62410 CP12 protein precursor-like protein 715-180 -1.26 0.11 H At2g17440 protein of unknown function 513-81 -1.26 0.02 H At1g74270 putative ribosomal protein 415-104 -1.26 0.12 M At4g17390 608 ribosomal protein L15 homolog lE-85 -1.26 0.02 H At1g68100 protein of unknown function 213-82 -l.26 0.01 H At1g76340 protein of unknown function 0 -1.25 0.02 H No hit found - - -1.25 0.03 L At1g69120 homeotic protein boilAPl, putative 413-175 -1.25 0.03 VH At1g79550 phosphoglycerate kinase, putative 113-60 -1.25 0.01 M No hit found - - -l .25 0.05 H At2g27690 putative cytochrome P450 0 -1.25 0.13 M At5g63460 protein of unknown function 7E-65 -1.24 0.10 M At4g28300 predicted proline-rich protein 0 -1.24 0.11 M At1g64400 acyl-CoA synthetase, putative 613-124 -1.24 0.01 H At4g13160 hypothetical protein 0 -1.24 0.12 L At3g46970 starch phosphorylase H (cytosolic form)-like protein 6E-90 -1.24 0.10 H At5g11680 hypothetical protein 2E-58 191 Table 4.12. (cont’d) Average tClgo Hybr. TIGR 1D Definition Blast down-reg. level score -1.24 0.12 M At5g49210 protein of unknown function ZED-111 -1.24 0.13 L At1g14700 purple acid phosphatase, putative 3E-86 -1.24 0.03 L At3g61200 hypothetical protein 113-131 -1.24 0.12 H At5g26240 CLC-d chloride charmel protein 0 -1.24 0.10 L At4g00870 awaiting functional assignment 352-172 -1.23 0.01 H At1g06630 protein of unknown function 6E-106 -1.23 0.11 M At5g48250 hypothetical protein 113-90 -1.23 0.09 M At5g10270 cdc2-like protein kinase 28-62 -l.23 0.02 H At5g53570 GTPase activator protein of Rab-like small GTPases-like p. SEE-150 -1.23 0.08 M No hit found - - -1.23 0.12 L At5g65480 hypothetical protein 913-81 -1.23 0.04 H At2g25450 putative dioxygenase 8E-164 -1.23 0.03 L Atlg13300 protein of unknown function 6B-88 -1.23 0.12 H At4g00620 putative tetrahydrofolate synthase 113-178 -1.23 0.06 M At2g29980 omega-3 fatty acid desaturase 313-113 -1.23 0.04 L At1g43190 nuclear ribonucleoprotein, putative 913-121 -1.22 0.04 L At5g61450 hypothetical protein 513-59 -1.22 0.11 H At1g50240 hypothetical protein 213-115 -1.22 0.01 M At4g01120 GBFZ, G-box binding factor 2E-48 -1.22 0.05 L At5g23920 protein of unknown function 0 -1.22 0.01 H At1g12810 protein of unknown function 513-99 -1.22 0.06 H Atlg52370 chloroplast 508 ribosomal protein L22, putative 0 -1.22 0.10 L At1g68190 putative zinc finger protein 113-75 -1.22 0.12 M No hit found - - -1.22 0.11 H At5g04850 protein of unknown function 213-48 -1.22 0.02 L No hit found - - -1.22 0.12 M At1g28280 hypothetical protein 4B-128 -1.22 0.07 M At3g22600 protein of unknown function 1E-138 -1.21 0.07 VH No hit found - - -1.21 0.06 VH At5g62880 Arac10 213-145 -1.21 0.03 L At3g54720 Peptidase-like protein 613-1 11 -l.21 0.1 1 L At4g25560 myb-like protein 3E-l40 -1.21 0.10 L At3g05320 protein of unknown function 7E-139 -1.21 0.00 L At5g04470 hypothetical protein 25-139 -1.21 0.11 H At3g21080 protein of unknown function 513-77 -l.21 0.08 L At3363170 hypothetical protein 413-166 -1 .21 0.01 M At2g30530 protein of unknown function 0 -1.20 0.10 VH No hit found - - -l.20 0.01 M At4g19210 RNase L inhibitor-like protein 615-109 -1.20 0.09 L At2g45960 aquaporin (plasma membrane intrinsic protein 18) 115-128 -1.20 0.06 L At5g22430 protein of unknown function 3E-111 -l.20 0.06 L At4g14880 cytosolic O-acetylserine(thiol)lyase (EC 4.2.99.8) 2E-40 -l.20 0.08 H A_t1g62770 protein of unknown function 213-79 192 Table 4.12. (cont’d) Average :tClgo Hybr. TIGR ID Definition Blast down-reg. level score -1.20 0.06 L At5g23060 hypothetical protein 813-106 -1.20 0.07 M At2g47850 protein of unknown function 913-160 -1.20 0.03 M At4gl4880 cytosolic O-acetylserine(thiol)lyase (EC 4.2.99.8) 2E-56 -1 . 19 0.01 M At4g33610 hypothetical protein 0 -1. 19 0.04 M At2g46830 MYB-related transcription factor (CCA 1) 0 -1.19 0.01 M At1g03440 protein of unknown function 0 -1.19 0.08 H At5g10290 protein serine threonine kinase-like protein 4E-166 -l.18 0.00 M At2g3 7790 putative alcohol dehydrogenase 9E-83 -1.18 0.06 H At3g13750 galactosidase, putative 613-68 -1. 18 0.07 H At 1 374520 AtHVA22a 1E-60 -1.18 0.01 L At1g05990 calcium-binding protein, putative 513-162 -1.18 0.08 H At4g28240 putative wound induced protein 0 -1.18 0.07 L No hit found - - -1.18 0.06 M At2g40510 40S ribosomal protein 826 113-64 -l.18 0.00 H A8309190 pseudogene, lectin GB:AAD56334 313-144 -1.18 0.01 L At4g03030 predicted OR23 protein of unknown function 215-91 -1.17 0.07 M At5g08210 hypothetical protein 713-170 -1.17 0.05 M At2g33040 mitochondrial F l-ATPase, 7 subunit (ATP3_ARATH) 913-86 -l.17 0.02 L At3gl5950 protein of unknown function 4E-29 -1. 17 0.02 H At3g02230 reversibly glycosylated polypeptide-l 6E-127 -1.17 0.04 M At5g14120 nodulin-like protein 113-140 -1.17 0.05 H At3g49210 hypothetical protein 7E-59 -1.16 0.04 M At5g53300 ubiquitin-conjugating enzyme E2-17 kD 10 413-31 -1 . 16 0.03 VH At1g48620 protein of unknown function 7E-84 -1.16 0.05 M At1g45332 mitochondrial elongation factor, putative 813-102 -1.15 0.04 L At5g18860 hypothetical protein 413-85 -1.15 0.05 H At5g20230 blue copper binding protein 513-143 -1.15 0.04 L At3g53480 ABC transporter-like protein 28-118 -l.15 0.04 M At5g44180 hypothetical protein 113-143 -1.15 0.02 M At1g57990 protein of unknown function 213-146 -1.15 0.01 L At5g66930 hypothetical protein 1132-67 -l.14 0.03 M At5g41940 GTPase activator protein of Rab-like small GTPases-like p. 313-93 -1.14 0.01 M At4g32280 hypothetical protein 213-126 -1.14 0.03 H At3g45190 hypothetical protein 3E-138 -1.14 0.02 M At1g50240 hypothetical protein 213-131 -1.13 0.00 VH At4g10000 hypothetical protein lE-134 -1.12 0.02 L At3g56270 hypothetical protein 115-91 -1 . 12 0.01 M At5g39570 hypothetical protein 213-85 -1.12 0.02 H At1g16190 UV-sensitive rad23, putative 113-66 -1.12 0.02 L At3g02420 protein of unknown function 215-68 -1.12 0.01 M At1g61620 hypothetical protein 28-98 -1.11 0.01 M At3g51520 hypothetical protein 313-102 -1.1 l 0.02 L @55460 putative RNA binding protein 28—31 193 Table 4.12. (cont’d) Average iClgo Hybr. TIGR 1D Definition Blast down-reg. level score -1 .11 0.00 L At2g44610 putative small GTP-binding protein 3E-111 -l.11 0.01 H At2g44350 citrate synthase 3E-55 -1 .1 l 0.01 H At4g00500 putative calmodulin-binding heat shock protein 413-156 -1.10 0.00 M Ammo putative purine-rich single-stranded DNA-bindingprotcin 113-57 194 Table 4.13. Arabidopsis thaliana cDNA microarray analysis: Identification of clones with too high hybridization signals to be used to calculate relative ratio and change of gene expression between samples. These clones saturated the signal detector for both dyes in both technical replications. TIGR 1D Definition Blast score At2g19590 l-aminocyclopropane-l-carboxylate oxidase 5.218-111 At5g55070 2-oxoglutarate dehydrogenase 82 subunit 220948-57 At3g49360 6-phosphogluconolactonase-like protein 0 At3g06810 acetyl-coA dehydrogenase, putative 2.0858-130 At5g42250 alcohol dehydrogenase 880888-91 At4g08920 AI fiavin-type blue-light photoreceptor (SW:Q43125) (Pfam: PFOO875) 1.3268—136 At4g141 10 COP9 protein 1.78768-53 At5g45340 cytochrome P450 9.5498-151 At3g11670 digalactosyldiacylglycerol synthase 0 At1g11545 endo-xyloglucan transferase, putative 1.2858-122 At1g07810 ER-type Ca2+-pump protein 1.4448-127 At5g14780 forrnate dehydrogenase (F DH) 2.7038-109 At1g32470 glycine cleavage system H protein precursor, putative 7.8768-110 At4g12720 growth factor like protein 4.083 8- 1 77 At3g61890 homeobox-leucine zipper protein ATHB-12 5.02038-69 At1g01380 Myb homolog (CPC), putative 2.0628-135 At4g08770 peroxidase C2 precursor like protein 9.7018-100 At4g15410 phosphatase like protein 1.48018-92 At4g02530 predicted protein of unknown function 1.4198-115 At2gl8390 putative ADP-ribosylation factor 7.6758-176 At2g43670 putative beta-1,3-glucanase, C terminal fragment 3.20568-85 At2g30950 putative fisH chloroplast protease 2.1418-173 At2g17870 putative glycine-rich, zinc-finger DNA-binding protein 1.0728-152 At2g27810 putative membrane transporter 2.7178-122 At1g14140 putative mitochondrial uncoupling protein 1.0668-102 At2g22420 putative peroxidase 1.8248-123 At2g02800 putative protein kinase 7.3428-104 At1g30730 putative reticuline oxidase-like protein 7.1948-133 At2g47620 putative SW1 SNF family transcription activator 6.9238-155 At1g08340 rac GTPase activating protein, putative 3.3288-116 At1g77940 ribosomal protein L30, putative 9.74958-49 At5g25280 serine-rich protein 3.3258-147 Atlgl 7100 SOUL-like protein 8.97398-88 At3307020 UDP-glucose:sterol glucosyltransferase 1.7148-93 At3g15090 zinc-binding dehydrogenase, putative 1.9698-176 At2g47270 protein of unknown function 3.2218-172 At3g17l60 protein of unknown function 2.248-126 At1g16040 protein of unknown function 5.1198-100 At1g29950 protein of unknown function 1.6928-114 A_t2g96530 protein of unknown function 1.4288-155 195 Table 4.13. (cont’d) TIGR 1D Definition Blast score At5g52840 protein of unknown function 1.3878-121 At2g39050 protein of unknown function 7.07438-87 At1g24090 protein of unknown function 2.67338-41 At1g10030 protein of unknown function 2.2458-83 At1g78280 protein of unknown function 1.1598-159 At4g16060 hypothetical protein 4.2418-45 At1g65510 hypothetical protein 1.34568-86 At1g56210 hypothetical protein 4.1248-121 At3g11100 hypothetical protein 1.0298-102 At4g17480 hypothetical protein 0 At4g13500 hypothetical protein 1.31258-94 At5g61020 hypothetical protein 4.6498-125 At5g64670 hypothetical protein 0 At4g39680 hypothetical protein 6.35438-78 At3g56230 hypothetical protein 9.4878-176 At3g47430 hypothetical protein 0 At3g51250 hypothetical protein 4.6898-111 At3g60650 hypothetical protein 7.5578-173 No hit found (10) - - 196 Down-regulated genes: Cell cycle Osmotic regulation 6.6% r Carbohydrate met. /‘ 4.2% Protein met. 4.7% Lipid met. 1.4% Sress response 3% Unknown function 50.9% Energy met. 6.6% Transport 0.9% Signal transduction 6.6% Others Transcriptional regulation 6. 1% 5 7% Up—regulated genes: Osmotic regulation 0.8% Cell cycle Protein met. 5.8% Lipid met. 2.5% Stress response 0.8% Energy met. 6.7% Transport 0.8% Signal transduction 5.0% Transcriptional regulation Unknown function 60.0% 2.5% Figure 4. l3. Arabidopsis thaliana cDNA microarray analysis: Functional classification of differentially expressed genes during active sucrose accumulation. 197 - Up-regulated 3A .32 as: «:2 Es... 3a.. m first? 2:: . up. 3%.». m3 .._:,_.. «in... W.;“: a... Q. VAR.— m. {N.— N42.— N. 72. 7 m. 7R4- v. 73.7 n. 7?.7 0.7%..- h. 73.7 n. 7R.7 a. 73.7 «9.73.7 a.Nv Mint. .. r. “.4. «I ._ ... bu. out a. .3 in _ O r . 0 0 0 4 2 5:258..— analysis: Frequency of microarray Fold-change of expression 198 differentially expressed transcripts during active sucrose accumulation. Figure 4.14. Arabidopsis thaliana cDNA Table 4.14. Arabidopsis thaliana oligonucleotide microarray, conventional (default) analysis: Genes up- and down-regulated during active sucrose accumulation. Fold up-reg. Hyb. S. Probe set Definition > + 20 310.8 259190_at unknown protein est hit, predicted by genscan > + 20 732.4 245012_at NADH dehydrogenase D3 + 11.17 600.7 244966__at cytochrome b6-f complex, subunit V + 4.00 478.4 244987_s_at ribosomal protein L2 Fold down- reg. Hyb. S. Probe set Definition -2.50 326.7 263235_at serine/threonine protein phosphatase, PP2A, catalytic subunit -2.22 7814.5 258712_s_at putative 408 ribosomal protein S23 similar to 408 ribosomal protein 823 (812) -2.10 1496.4 245356_at adenosylhomocysteinase :supported by full-length cDNA: Ceres: 16846. -2.10 226.3 245541_at hypothetical protein -2.10 253.8 252026_at serine protein kinase - like serine protein kinase SRPKI, Homo sapiens -2.10 491.3 249674_at unknown protein -2.00 264.3 266129_at hypothetical protein predicted by genscan -2.00 272.2 255293_at putative zinc finger protein -2.00 260.4 255706_at putative ribosomal protein 813 similar to ribosomal protein S13 -2.00 367.4 261393_at NAM (no apical meristem)-like protein OsNAC4, 01:6730938 from [0. saliva] -2.00 298.9 251672_at MADS-box transcription factor-like DEFH125 - Antirrhinum majus -2.00 295.5 259007_at putative MYB family transcription factor C-term similar to C GB:Q08759 -2.00 2671 1.4 247644_s_at translation elongation factor eEF-l alpha chain (gene A4) -2.00 290.9 263089_at putative retroelement pol polyprotein -2.00 282.7 2461 16__at putative protein predicted proteins - Arabidopsis thaliana -1.91 426.5 245687_at unknown protein -1.91 424.9 266705_at 408 ribosomal protein S30 -1 .91 239.9 264700_at hypothetical protein predicted by genemark.hmm -1.83 915.2 251843_x_at extensin precursor -like protein extensin precursor, Kidney bean, P1R:T10863 -l .83 260.2 246007_at Expressed protein ; supported by full-length cDNA: Ceres: 31366. -1.83 1429.4 251409_at Expressed protein ; supported by full-length cDNA: Ceres: 42545. -1.83 186.5 245921_at putative protein predicted protein, Arabidopsis thaliana -1.83 277.1 254862_at putative transport protein Na(+) dependent transporter (Sbf family) - A aeolicus -1.83 329.9 258615_at putative aspartyl protease contains Pfarn profile: PF 00026 -1.83 330.1 246488_at steroid 5alpha-reductase-like protein steroid 5alpha-reductase - R norvegicu: -1.76 953.8 250683_x_at putative protein similar to unknown protein (pii1lT14195) -1.76 342 258979_at heat-shock protein (At-hsc70-3) identical to (At-hsc70-3) (cytosolic Hsp70) -l .76 257 249918_at putative protein predicted protein, Arabidopsis thaliana -1.69 166.3 266715_at putative RNA-binding protein -1.69 206.3 247705_at putative protein ;supported by full-length cDNA: Ceres:22152. -1.69 408.4 257029_at dem-like protein similar to dem GB:CAA‘73973 from [Lyeopersicon exculentum] -1.69 291.1 249540_at 4-coumarate-COA ligase -1ike protein 4-coumarate—C0A ligase, A. thaliana -1.69 273.4 266605_at putative SNFZ subfamily transcriptional activator -1.63 71.9 251604_at putative protein 608 RIBOSOMAL PROTEIN L21 — A. thaliana -1.63 271.8 2463 59_x_at hypothetical protein -1.63 181.6 247052_at homeodomain transcription factor-like -1.63 188.3 257804_at protein kinase, similar to somatic embryogenesis receptor-like kinase D. carom -1.58 264.3 251787_at 2-oxog1utarate dehydrogenase, E1 subunit - Arabidopsis thaliana -1.58 117.7 245376_at peroxidase like protein -1.53 222.3 246092_at glutaredoxin ;supported by full-length cDNA: Ceres:115597. -1.53 169.8 259717_at putative cleavage and polyadenylation specificity factor 73 kDa [H. sapiens] -1.53 1134.9 265963_s_at 40S ribosomal protein S5 identical to GP:3043428 -1.53 272 254877_at putative protein threonine dehydratase - Escherichia coli,P1R1 :DWECT D -1.49 355.3 263608_at putative glycine-rich RNA-binding protein -1.45 213.6 249558_at hypothetical protein -1.38 361.9 259130 at putative ribosomal protein L39 GB:P51424 [Arabidopsis thalianath2 199 Table 4.15. Arabidopsis thaliana oligonucleotide microarray, non-conventional analysis: Number of transcripts subdivided by hybridization signal intensities. Hybridization levels refers to low (L), medium (M), high (H), and very high (V H) if the average hybridization means of probe sets were between 400 and 500, between 500 and 600, between 600 and 1000, or > 1000 units, respectively. 3rd WAE 7th WAE Absent Present Total 1. M H VH Total Absent 21,901 184 42 20 o 246 22,147 Present L 274 48 25 8 2 83 357 M 84 36 14 12 o 62 146 H 17 14 14 26 3 57 74 VB 0 o 2 6 14 22 22 Total 375 98 55 52 19 224 Total 22,276 282 97 72 19 22,746 200 Table 4.16. Arabidopsis thaliana oligonucleotide microarray, non-conventional analysis: Genes up-regulated during active sucrose accumulation (7'h WAE). Hybridization levels (HL) refers to medium (M), hi (H), and very high (VH) if the average hybridization mean of the probe sets of the 7 WAE samples were between 500 and 600, between 600 and 1000, or > 1000 units, respectively. Fold up- HL Probe set Definition _regulation 7 WAE 26.85 H 245012_at NADH dehydrogenase D3 23.53 M 255017_at hypothetical protein 12.41 H 263289_at ubiquitin extension protein (UBQZ) 9.06 M 257148_at cytochrome c 8.38 M 256887_at hypothetical protein 5.94 M 249310_at putative protein contains similarity to 40S ribosomal protein 810 4.22 M 249237_at putative protein similar to unknown protein (sp|P37707) 4.11 H 244966_at cytochrome b6-f complex, subunit V 4.04 M 260597_at hypothetical protein 3.33 M 262287_at unknown protein 3.28 M 246302_at Ca2+/H+-exchanging protein-like Arabidopsis thaliana 3.27 H 249220_at putative protein 3.10 H 249492_at germin-like protein (GLP2a)copy1 3.05 VH 244929_at NADH dehydrogenase subunit 4 2.78 M 264012_at hypothetical protein 2.71 M 255397_at putative transposon protein 2.62 H 261229_at RAC-like GTP-binding protein ARAC4 2.62 H 262581_at unknown protein 2.61 M 264789_at putative glycine-rich, zinc-finger DNA-binding protein 2.57 M 245152_at putative mitochondrial carrier protein 2.46 H 252764_at nucleic acid binding protein-like 2.46 M 256481_at elicitor response protein 2.43 M 262851_at putative RNA helicase Contains DEAD-box subfamily ATP-dependent 2.36 H 247762_at cell wall protein 2.35 M 250027_at putative protein similar to unknown protein 2.34 H 253843_at putative protein MLL protein 2.34 VH 245002_at PSll D2 protein 2.33 H 245026_at ATPase 111 subunit 2.17 M 258488_at unknown protein 2.17 H 250797_at unknown protein 2.17 M 255992_at unknown protein 2.09 H 264164_at Expressed protein 2.05 H 258715_at putative 608 ribosomal protein L1 2.05 M 257122_at RNA-binding protein 1.96 M 266709_at unknown protein 1.96 H 257435_at putative RSZp22 splicing factor 1.91 M 245281_at DEF (CLAl) protein 1.90 H 252643_at acidic ribosomal protein P2 -Iike 1.89 H 256373_at hypothetical protein 1.88 M 259907_at GCN4-complementing protein 1.85 M 267590_at putative expansin 1.84 M 245718_at putative protein 1.80 M 256617__at unknown protein 1.79 M 262090_at CREE-binding protein 1.74 H 248362 at fibrillarin 1 (AtFibl) 201 Table 4.16 (cont’d) Fold up- HL Probe set Definition _regulation 7 WAE 1.74 H 260970_at unknown protein 1.73 H 267368_at citrate synthase similar to GB:X17528 1.71 H 250534_at NADH dehydrogenase 1.68 M 264130_at hypothetical protein 1.66 M 263 l95_at non-specific lipid transfer protein 1.65 M 264084_at putative kinesin light chain 1.65 M 263489_at putative inositol polyphosphate 5 -phosphatase 1.63 H 258256_at unknown protein 1.62 M 253357_at Dem -like protein Dem (defective embryo and meristems) 1.62 M 247233_at mitochondrial carrier protein-like 1.59 VH 266533_s_at putative plasma membrane intrinsic protein 1.58 H 265187_at putative ADP-ribosylation factor 1.58 M 244912_at cytochrome c biogenesis ort382 1.56 H 251776_at eukaryotic translation initiation factor 6 (ElF-6) 1.56 M 2453l6_at hypothetical protein 1.51 M 263806_at hypothetical protein 1.49 H 252824_at histone H3.3 1.45 M 264607_at putative K+ channel, beta subunit 1.44 M 261956_at oxidoreductase 1.43 H 264644_at hypothetical protein 1.43 M 246131_at molybdopterin biosynthesis CNXl protein 1.43 M 258020_at unknown protein 1.42 M 2536l8_at nodulin-like protein MtNZl gene product 1.42 M 265232_s_at hypothetical protein 1.40 M 258128_at chloroplast thylakoidal processing peptidase 1.40 H 255940_at prolyl endopeptidase 1.38 H 255089_at nucleoside-diphosphate kinase 1.38 M 254l40_at putative protein 1.38 M 251112_s__at pyruvate decarboxylase-like (EC 4.1.1.1) 1.36 H 262341_at E2, ubiquitin-conjugating enzyme 1.36 M 252107_at sugar transporter-like protein sugar transporter, Arabidopsis thaliana 1.36 H 254227_at putative protein chS-Rex-b 1.36 VH 252056_at ubiquitin extension protein (UBQl) 1.36 M 266690_at malate oxidoreductase (malic enzyme) 1.35 M 246829_at putative protein pyruvate water dikinase 1.35 H 254239_at water channel - like protein plasma membrane intrinsic protein lo 1.34 M 258284__at putative ribosomal protein similar to ribosomal protein L37 1.34 M 261397_at hexose transporter 1.34 M 265319_at auxin-regulated protein (1AA8) 1.33 VH 258569_at ribosomal protein L17 1.33 VH 259077.s_at reversibly glycosylated polypeptide-1 1.32 M 256152_at ethylene-responsive RNA helicase 1.30 H 266082_at hypothetical protein 1.29 M 252327_at MTN3-like protein MtN3 gene product 1.29 M 244943_at NADH dehydrogenase subunit 9 l .28 H 248779__at aeetoacyl-CoA-thiolase 1.28 H 246538_at 4OS ribosomal protein 819 1.28 H 245939_at oxoglutarate/malate translocator-like protein 1.28 M 266815_at F-box protein family, AtFBXS 1.27 M 253 706_at putative protein D-threonine dehydrogenase 1.26 VH 252198 x at putative protein predicted protein 202 Table 4.16 (cont’d) Fold up- HL Probe set Definition regulation 7 WAE 1.24 M 260625_at storage protein 1.24 M 264728_at unknown protein 1.23 H 261416__at ribosomal protein S15 1.22 M 259837_at aquaporin, putative similar to delta tonoplast integral protein 01:1 145697 1.22 M 266396_at unknown protein 1.21 M 256145_at lipid transfer protein 1.21 M 266158_at putative ABC transporter 1.20 M 245915_s_at tubulin alpha-5 chain-like 1.20 M 247265_at putative protein 203 Table 4.17. Arabidopsis thaliana oligonucleotide microarray, non-conventional analysis: Genes down-regulated during active sucrose accumulation (7th WAE). Hybridization levels (HL) refers to medium (M), high (H), and very high (VH) if the average hybridization mean of the probe sets of the 3th WAE samples were between 500 and 600, between 600 and 1000, or > 1000 units, respectively. Fold down- HL Probe set Definition _regulation 3 WAE -15.37 M 264804_at putative receptor kinase, CLVl -12.62 M 266348_at putative MAP kinase -8.01 M 249815_at 608 ribosomal protein L13 -5.92 M 263974_at hypothetical protein -5.75 M 259540_at nodule inception protein -5.46 M 267105_at hypothetical protein -4.86 M 250816_at coatomer delta subunit (delta-coat protein) (delta-COP) 4.03 M 258535_at unknown protein -3.80 H 248862_at unknown protein -3.46 M 264962_at auxin transport protein ElRl -3.42 M 245462_at transcription factor like protein -3.23 M 253742_at putative protein retrofit -3. 12 M 249800_at MtN3-like protein -3.07 H 247936_at lycopene epsilon cyclase -3.07 M 260778_at splicing factor -2.98 M 250366_at putative protein predicted proteins in castor bean -2.95 M 248785_at unknown protein -2.91 M 260168_at unknown protein contains zinc finger, C3HC4 type -2.87 M 263560_s_at unknown protein -2.76 M 252481_at putative protein DCL PROTEIN, Chloroplast precursor -2.72 M 252255_at pectinesterase -2.70 VH 253291_at expressed protein -2.70 M 256990_at hypothetical protein -2.70 H 259130_at putative ribosomal protein L39 -2.60 M 255220_at polyubiquitin (UBQIO) -2.60 M 260896_at flower pigmentation protein ATANl l -2.54 M 256149_at zinc finger protein -2.48 M 249708_at unknown protein -2.48 H 260816_at hypothetical protein -2.46 M 251614_at hypothetical protein -2.44 M 25189l__at aintegumaenta-like protein ovule development protein -2.42 M 262364_at disease resistance protein -2.40 M 246521_at N2,N2-dimethylguanine tRNA methyltransferase-like protein -2.36 M 246669__at galactinol synthase, putative -2.26 M 260775_at 14—3-3 protein GF14omega (grt2) -2.24 M 260417_at putative chromomethylase -2.23 H 258588_s_at glyceraldehyde-3-phosphate dehydrogenase C subunit (GapC) -2.23 M 252043_at putative protein mttC protein -2.19 H 255140_x_at extensin-like protein hydroxyproline-rich glycoprotein precursor -2. 13 M 245695__at rec - like protein -2.11 M 249114_at putative protein -2.08 H 253230_at putative protein -2.07 H 263517_at unknown protein ~2.00 VH 250790_at putative protein -l.98 M 258489 at 14-3-3 protein GF14nu (grfl) 204 Table 4.17 (cont’d) Fold down- HL Probe set Definition regulation 3 WAE -1.97 M 249772_at unknown protein -1.97 M 258972_at hypothetical protein -l.94 H 260780_at valyl-tRNA synthetase -1.92 M 246815_at putative protein integrin analogue -1.92 M 2534l3_at putative protein Fe(11) transport protein -1.92 M 256112_at guanine nucleotide regulatory protein -1.92 M 252337_at Cell division control protein 2 homolog A -1.92 M 265750_x_at putative Athila retroelement ORFl -1.91 M 264811_at hypothetical protein -1.91 M 246132_at Rad51-like protein -l.90 M 255981_at hypothetical protein -1.90 M 265394_at predicted protein -1.89 M 249540_at 4-eoumarate—CoA ligase -like protein -1.87 H 266266_at putative enolase (2-phospho-D-glycerate hydroylase) -1.84 H 257790_at gda-l -1.83 H 257173_at S-adenosyl-L-homocysteinas -1.81 M 264890_at unknown protein -1.80 H 263608_at putative glycine—rich RNA-binding protein -l.79 H 259224_at putative thymidine kinase similar to thymidine kinase -1 .79 M 266344_at hypothetical protein -1.78 H 254879_at hypothetical protein -1.78 M 260689_at hypothetical protein -1.78 M 251672_at MADS-box transcription factor-like protein -1.78 H 248833_at Bax inhibitor-1 like -1.77 M 251853_at putative protein -1.75 M 245825_at protein kinase -1.74 H 255932_at mutator-like transposase -1.74 H 252945_at putative protein -1.73 M 262009_at hypothetical protein -1.73 M 259408_at protein phosphatase 2A 65 kDa regulatory subunit -1.72 VH 261639_at tubulin alpha-2/a1pha-4 chain -1.70 M 251536_at ketol-acid reductoisomerase -1.69 M 262718_at hypothetical protein similar to putative non-LTR retroelement reverse transcriptase -1.68 VH 264421_at ribosomal protein -1.68 H 245479_at extensin like protein -1.66 M 256397__at putative dual-specificity protein phosphatase -1.65 M 267560_at putative cytochrome P450 - l .64 M 259800_at Expressed protein -1.64 M 253717_at putative protein putative suppressor protein - Arabidopsis thaliana,PlD:g3687246 -1.62 H 255789_at 608 ribosomal protein L23 -1.61 M 255928__at unknown protein -1 .61 M 255104_at expressed protein -1.60 M 245378_at GLABRAZ like protein -1.60 M 255704_at putative proline-rich protein -1.60 M 248877_at putative protein -1.59 H 256035_at Ran-binding protein (atranbpla) -1.59 H 245606_at hypothetical protein -1.59 H 247818_at contains similarity to GTP-binding protein CGPA -1.59 M 253050_at putative protein probable arabinogalactan protein precursor -l.59 VH 263838_at putative s-adenosylmethionine synthetase -1.58 H 253018 at putative protein Zn firger protein BBF2aO -Nicotiana tabacum 205 Table 4.17 (cont’d) Fold down- HL Probe set Definition _regulation 3 WAE -1 .57 M 264685__at endo-1,4-beta-glucanase -l.55 M 267223_at hypothetical protein -1.55 M 260295_at putative arninopeptidase N (alpha-aminoacylpeptide hydrolase) -1.53 M 262529_at putative receptor protein kinase (disease resistance proteins) -1.52 M 266768_s_at ubiquitin extension protein (UBQ6) -1.52 VH 245989_s_at polyubiquitin (UBQ4) -1.50 M 261304_at unknown protein -1.50 M 25589l_at hypothetical protein -1.50 H 265357_at E2, ubiquitin-conjugating enzyme -1.48 M 261591_at protein kinase -1.48 M 256024_at unknown protein -1 .48 H 252157_at hypothetical protein - l .47 M 265644_at hypothetical protein -1.47 H 263177_at hypothetical protein -1.46 M 255694_at putative transcriptional regulator -1.46 H 247154_at receptor protein kinase-like protein -1.45 VH 265963_s_at 408 ribosomal protein 85 identical to GP:3043428 -1.45 M 254363_at pectinesterase like protein pectinesterase - 1 .44 M 245268_at hypothetical protein -1.44 M 254621_at gene “-1 protein - like -1.43 H 253002_at phosphoinositide-specific phospholipase C - l .43 M 248886_at phosphate/triose-phosphate translocator precursor -1.42 H 250434_at histone H3 - like protein -1.42 H 249357_at ribonucleoprotein -|ike -l.42 H 249818_at Expressed protein -1.42 M 252926_at H+-transporting ATPase 16K chain P2, vacuolar -1.40 H 254446_at tubulin beta-9 chain -1.40 M 263990_at putative phosphoprotein phosphatase -1.39 H 250433_at histone H3 - like protein -1.39 M 249497_at putative protein predicted proteins -1.39 M 266366_at unknown protein -1.38 M 245342_at ribosomal protein -1.38 M 249519_at putative protein -1.37 VH 250683_x_at putative protein -1.35 M 267306_at Ran binding protein (AtRanBPlb) -l.35 M 248026__at putative protein similar to unknown protein (gblAAC63638. 1) -1.33 M 255946_at glycine hydroxymethyltransferase -l .33 M 256607_at unknown protein -1.32 H 263298__at unknown protein -l.32 H 249674_at unknown protein -1.30 H 263552_x_at unknown protein proline, tyrosine, and serine-rich protein -1.30 M 263554_at putative Tal l-like non-LTR retroelement protein -1.29 M 264248__at hypothetical protein -1.29 M 248967_at unknown protein -1.28 H 265169_x_at unknown protein -1.28 M 257029_at dem-like protein similar to dem GB:CAA73973 -1.28 H 253754_at glycine-rich protein like -1.27 VH 249236_at putative protein -1.27 H 267638_at unknown protein -1.26 M 245862_at putative protein -1.26 M 259775 at hypothetical protein 206 Table 4.17 (cont’d) Fold down- HL Probe set Definition _rggulation 3 WAE -l.26 M 251388_at putative protein protein regulating cytokinesis 1 -l.26 H 246185_at 5-methyltetrahydropteroyltriglutamate-homocysteine S-methyltransferase -1.25 M 266705_at 40S ribosomal protein 830 -l .25 M 267136_at putative GTP-binding protein (extra large) -1.25 H 260765_at actin 8 almost identical to actin 8 G1: 1669389 from [Arabidopsis thaliana] -1.25 M 263164_at putative glutamate/aspartate-binding peptide -1 .24 M 248962_at putative protein similarity to FKBP-type peptidyl-prolyl cis-trans isomerase -1.24 H 261939_at cytochrome c oxidase subunit -1.24 H 259361_at glyceraldehyde-3-phosphate dehydrogenase -1.24 M 263183__at putative glucan synthase -1 .23 H 245435_at hypothetical protein -1.23 H 259114_at 26S proteasome AAA-ATPase subunit RPTSa -1 .23 M 262560_at hypothetical protein -l.21 H 255752_at secretory carrier membrane protein -1.21 M 245687_at unknown protein -1.20 VH 261129_at tubulin alpha-2/alpha-4 chain -1.20 VH 251257 at ADP-ribomation factor-like protein ADP-ribosylation factor 1 207 40.1 ..................................................................................................................... - Down-regulated -Up-regulated 30.1 ................................................ .. ..... ,‘l; ...................................................... > o C 8 O”20.. ...................................................................................................... 9 u. E}, .~ LI 10.. ......................... .. .. .. .. .. ..... ”1.. ........................................ ' iiF‘l‘iil» la 1‘ 1 I ll 0.] l l’l qqqquzqnqataaqwq 2221222222===sss véqqnqnqxd‘ii‘fiffifin «777°37'77": I Fold-change of expression Figure 4.15. Arabidopsis thaliana oligonucleotide microarray, non-conventional analysis: Frequency of differentially expressed transcripts during active sucrose accumulation. 208 Cell cycle Down-regulated genes: 14.6% Osmotic regulation 0.6% Carbohydrate met. 4.9% Protein met. 6.1% Lipid met. 1.2% Stress response 0.6% Energy met. 3. 7% Unknown function 37.8% Transport 2.4% Signal transduction 9.8% 13 .4% Transcriptional regulation 4.9% Cell cycle 9.5% Osmotic regulation 5.7% Up-regulated genes: Carbohydrate met. 5.7% Protein met. 4.8% Lipid met. 2.9% Unknown function 31.4% Stress response 1.0% Energy met. 11.4% Others 7.6% Transport Transcriptional regulation Spa] transdtction 2.9% 9- 7.6% 5% Figure 4. l6. Arabidopsis thaliana oligonucleotide microarray, non-conventional analysis: Functional classification of differentially expressed genes during active sucrose accumulation. 209 Discussion Most of the concentric cambium rings that form the majority of the sugar beet root volume at maturity are formed during the first weeks afier emergence (Elliot and Weston, 1993). A chronological analysis of the biological changes that occur during the early root developmental phases in sugar beet, particularly considering the changes in the dynamics of sucrose accumulation, could yield information on overall difference in productivity between different breeding lines, which could be used to develop new selection parameters during the early phases of plant development. Comparison between sugar beet lines with different sucrose accumulation capability (SR96 > USH20 > C869) showed that these difference could be detected in greenhouse-grown plant as early as at the 7th week after emergence (WAE) in 2003 and after the 9th WAE in 2002. These differences, also observed in field-grown plants after three to four months after emergence, were mainly associated with differences in dry matter content more than in sucrose content as dry matter, as expected by the comparison between elite sugar beet lines. During the field experiment the higher variability observed and the limited number of samples collected may have masked expected differences between sugar beet lines. Differences in trait values between 2002 and 2003 greenhouse experiments were probably caused by different time of transplanting (three weeks later) in 2003, accompanied by increase of the average temperatures and photoperiod during the early plant developmental phases. Other than showing the possibility to detect differences between breeding lines, the greenhouse experiments also highlighted that, during the first two month after 210 emergence, all biochemical pathways involved in sucrose metabolism that influence sucrose content at maturity are already present and fully active. Particularly between the 3rd and the 7‘h WAE a sharp change in root sucrose content occurs in both the dry matter and fresh weight, reaching levels comparable to those observed at root maturity. These results are consistent with earlier observations of Bergen (1967) and Milford (1973 and 1988) for field-grown plants, and Klotz (2002 and 2004) for greenhouse-grown plants. Several techniques for a comprehensive time course analysis of differentially expressed genes have been developed during the last years (Kuhn, 2001; Green et al., 2001; Donson et al., 2002). Of these techniques, cDNA-AFLP has mainly been used to study gene expression changes during developmental phases of species where little sequence information was available, such as in Beta vulgaris L. (Bachem et al., 1996; Durrant er al., 2000; Jones et al., 2000; Okamuro er al., 2000; Breyne et al., 2003). Reliability of cDNA-AF LP for genome-wise analysis of gene expression was confirmed in my study by the similarity of results obtained in the two years and by the high ratio of similarity between biological replications analyzed during the second year of greenhouse experiment, particularly considering that differences between biological replications might be expected. The results of this research indicated a change of expression of at least 25 % of the genes normally expressed during the first two months of root development. Gene expression profiles of early developing roots (first three WAE) showed the highest number of expressed transcripts, associated with very little sucrose accumulation capability. Root tissues in this early developing phase were predominantly involved in cell division and cell expansion, and no functional differentiation for sucrose storaging of 211 the root tissues was observed. This stage seemed to gradually change during the following two weeks (4“1 and 5m WAE), with the activation of some of the transitional- expressed genes and the deactivation of the 199 early-expressed genes, causing an overall decrease in the total number of expressed transcripts. This transitional phase is characterized by a sharp increase in root sucrose accumulation. Consequently to the deactivation of the early-expressed genes, starting at the 5th WAE several late-expressed genes appear to be activated in the active sucrose accumulating root system. Particularly between the 5‘h to the 6“I WAE, profiles of expressed genes changed drastically, while similar profiles of expressed genes were observed from the 6‘” through the 9th WAE. This pattern of gene expression suggested a subdivision of the first two months of root development in distinct developmental phases: a pre-sucrose accumulation phase of root tissues development during the first three weeks where no functional differentiation for sucrose storaging was observed, and an active sucrose accumulation phase after the 6"I WAE where root tissues appear to be already functionally differentiated. A transitional phase, when root tissues started accumulating occurred from the 4"I to the 6'" WAE. Because no morphological or physiological markers specifically exist to differentiate juvenile to adult vegetative phases, and because this transition occurs gradually during development, it would be plausible to consider the overall gene expression changes as good indicator of this phase switch, and the functional differentiation of the tissues for sucrose storaging as further confirmation of this juvenile to adult vegetative phases change in the sugar beet root system. The use of microarray technology to precisely identify genes differentially expressed has been widely used in plants and animals (Lemieux et al., 1998; Lipshutz er al., 1999; 212 Brown and Botstein, 1999; Kuhn, 2001). Oligonucleotide microarrays are generally used to analyze gene expression profiles and characterize genes in species for which genomic sequences information is known (Harmer et al., 2000; Schaffer et al., 2001) but are rarely used for cross-species hybridization studies using transcripts from different species (Kayo et al., 2001). In contrast, cDNA-based microarrays can also be used for species for which little genomic sequence information is known, using transcripts selected from cDNA libraries (Wullschleger and Difazio 2003; Casu et al., 2003). My study combined the advantage of the suppression subtractive hybridization technique to select only those transcripts differentially expressed fi'om a complex cDNA library (Pearson et al., 2001), and the ability to confirm and quantify their relative levels of expression with the cDNA microarray in a species with little sequence information available such as in B. vulgaris. The relative abundance of clustered transcripts indicated high levels of expression and possible functional differentiations of the root tissue. However, the majority of the clustered genes represented regulatory genes rather than structural genes, implying that several metabolic changes could be active during early root development, as opposed of the variation of only few specific biosynthetic pathways, such as those involved in sucrose biosynthesis and accumulation. Clustered genes down- regulated during root development were mainly involved in the regulation of cell cycle, like ribosomal proteins, or in signal transduction, like for the 8-fold down-regulated PRLl interacting factor. Interestingly, PRLl is a negative regulator of the Arabidopsis SNFl kinase, a key regulator of glucose signaling (Bhalerao et al., 1999). Pentatricopeptide (PPR) repeat-containing protein, one of the most abundant clustered transcripts up-regulated during active sucrose accumulation, is supposed to have RNA- 213 binding sites and be involved in RNA processing and stabilization (Small and Peeters, 2000). Functional analysis of singletons confirmed the observations made for the clustered genes. Overall, transcripts involved in cell cycling and osmotic regulation were mainly expressed during the early stage of root development, while genes involved in transcriptional regulation, signal transduction, energy metabolism and metabolite transports were mainly expressed during the sucrose accumulating phase of root development. Structural genes involved in proteins, lipids, and particularly carbohydrate metabolism were not found to be differentially expressed between phases, with the exception of sucrose synthase (SUSl), an enzyme directly involved in the catabolism of sucrose, that was weakly over-expressed before sucrose started accumulating in the root. However SUSl high expression during early root development can be explained considering that SUS 1 , interacting with cellulose synthase in the plasma membrane, plays also an important function in cell wall biosynthesis (Chourey et al., 1998). A limited expression of transcripts for carbohydrate metabolism was also observed in maturing stems of sugarcane, while sugar transporters were mainly expressed (Casu et al., 2003). Several lines of evidences supported the quality and reliability of the results obtained with the B. vulgaris subtracted library cDNA microarray. The majority of the transcripts selected by the subtracted probes hybridized to all four biological replications of the two sugar beet lines, and the levels of gene expression between biological replications were similar to each other. Furthermore, similarity of levels of gene expression observed between independently arrayed clustered sequences support the reliability of the results. Overall, suppressive subtractive hybridization was demonstrated to be an efficient technique able to select differentially expressed genes, particularly for abundant 214 (clustered) transcripts showing more than 3-fold difference of expression between the two time-point samples considered for the analysis. Considering the Arabidopsis thaliana cDNA microarray experiment, the level of hybridization between B. vulgaris probes and A. thaliana targets was relatively high (53.2 %). In a similar experiment, Horvath et al. (2003) found levels of hybridization to A. thaliana cDNA microarray of 23 %, 34% and 47 % for wild cat (Avena fatua), poplar (Populus deltoidries) and spurge (Euphorbia esula), respectively. In spite of the high level of hybridization, the low number of differentially expressed genes (5 %) and the relatively low differences in gene expression (< 2-fold), indicated the inability of the Arabidopsis-based microarray to precisely identify differentially expressed genes during root development in sugar beet. This result could be explained considering the broad evolutionary distance between brassicaceae and chenopodiaceae families with consequent decrease of sequences homology between A. thaliana and B. vulgaris genomes. Overall, genes involved in the cell cycle, osmotic regulation and stress responses were uporegulated during early root development, as observed in the B. vulgaris cDNA microarray experiment. Conversely, sucrose synthase was found to be up-regulated during the active sucrose accumulation phase, and two distinct sucrose transporters SUCl and SUC2 were found down-regulated (-1.34 fold) and up-regulated (+ 1.77 fold) during root development, respectively. During the Arabidopsis thaliana oligonucleotide microarray the level of transcript detection strongly decreased, and only 4.6 % of the probe set for the conventional analysis and 3.8 % of the probe sets for the non-conventional analysis showed a detectable hybridization signals. Affymetrix gene chips are developed to be analyzed 215 only with conventional and standardized analytical software, and results from data analysis using non-conventional parameters must be considered very carefully, and only used as general indicator of differentially expressed genes. Cell cycle related genes were also found to be up-regulated during the early phase of root development, while genes related with energy metabolism and transcriptional regulation were up-regulated during the sucrose accumulation phase. Results obtained with microarrays experiment confirmed the overall decrease of expressed transcripts between the 3Ird and the 7'“ WAE, as observed with cDNA-AF LP technique. 216 Conclusions (i) Phenotypic analysis of the dynamics of root sucrose accumulation revealed a sharp increase in sucrose content from the 3rd to the 7th week after emergence (WAE) in greenhouse-grown plants. Comparison between sugar beet lines with known difference of sucrose storage capability at maturity showed proportional sucrose content differences after seven and nine WAE in greenhouse-grown plants. These differences in sucrose levels as fresh weight were caused by differences in root dry matter content and not by differences in sucrose content as dry matter. Similar differences between lines for sucrose content as fresh weight and root dry matter were observed after 13 to 16 WAE in field- grown plants, indicating these traits as potential selective indicators for early greenhouse screening during breeding programs. (ii) Gene expression profiles revealed an early root developmental phase during the first three WAE characterized by the highest number of expressed transcripts and absence of accumulation of sucrose in the root. This early phase was followed by a transitional phase from the 4m to the 5th WAE when the expression of the early-expressed transcripts terminated and the expression of late-expressed transcripts started. This transitional phase was characterized by a sharp increase of sucrose accumulation in the root system. The transitional phase was followed by a later root developmental phase after the 6‘h WAE, characterized by late-specific expressed transcripts and root sucrose content values on dry and fresh weight basis similar to those observed at root maturity. The highest change in gene expression profiles was observed between the 5“I and the 6th WAE. Considering the dynamics of sucrose accumulation and of gene expression profiles 217 together, it was hypothesized that there was a developmental phase change from the juvenile vegetative phase, characterized by the lack of sucrose accumulation activity, to the adult vegetative phase, characterized by active sucrose accumulating tissues and full activity of those metabolic pathways that influence sucrose content at root maturity. This developmental phase change occurred between the 4"I and the 6th WAE in the sugar beet root system. (iii) Genes differentially expressed between developmental phases characterixed by different sucrose accumulation activity were mainly represented by regulatory genes as opposed to structural genes. Transcripts up-regulated during early root development (juvenile phase) were mainly associated with cell cycle metabolism and osmotic regulation pathways, while genes up-regulated during active root sucrose accumulation (adult phase) were mainly involved in transcriptional regulation and signal transduction pathways, in energy metabolism and in metabolite transports. Arabidopsis-based microarrays were not completely efficient in detecting hybridizations and change of transcripts expression during Beta vulgaris L. gene expression analyses. These findings would increase our understanding on which metabolic pathways regulate the early stages of root development and sucrose accumulation in sugar beet. 218 Literature cited Altschul, S. F.; Gish, W.; Miller, W.; Myers, E. W.; Lipman, D. J. Basic local alignment search tool. J. Mol. bioI. 1990, 215, 403-410. Bachem, C. W. B.; van der Hoeven, R. S.; de Bruijn, S. M.; Vreugdenhil, D.; Zabeau, M.; Visser, R. G. F. Visualization of differential gene expression using a novel method of RNA fingerprinting based on AF LP: Analysis of gene expression during potato tuber development. The Plant Journal 1996, 9, 745-753. Bachem, C. W. B.; Oomen, R. J. F. J .; Visser, R. G. F. Transcript imaging with cDNA- AF LP: a step-by-step protocol. Plant Molecular Biology Reports 1998, 16, 157-173. Bellin, D.; Werber, M.; Theis, T.; Schulz, B.; Weisshaar, B.; Schneider, K. EST Sequencing Annotation and Macroarray Transcriptome Analysis Identify preferentially Root-Expressed Genes in Sugar Beet. Plant Biology 2002, 4, 700-710. Bergen, p. Seasonal patterns of sucrose accumulation and weight increase in sugar beets. J. Am. Soc. Sugar Beet T echnol. 1967, 14, 538-545. Bhalerao, R. P.; Salchert, K.; Bako, L.; Okresz, L.; Szabados, L.; Muranaka, T.; Machida, Y.; Schell, J.; Koncz, C. Regulatory interaction of PRLI WD protein with Arabidopsis SNF l-like protein kinases. Proc. Natl. Acad Sci. USA 1999, 96, 5322- 5327. Bouillene, R.; Kronacher, P. C.; de Roubaix, J. Etapes morphologiques et chimiques dans le cycle vegetatif de la batterave sucriere. Publications Institut Belge pour l’Amelioration de la Batterave 1940, 8, 87-166. Breyne, P.; Dreesen, R.; Cannoot, B.; Rombaut, D.; Vandepoele, K.; Rombauts, S.; Vanderhaeghen, R.; Inze, D.; Zabeau, M. Quantitative cDNA-AF LP analysis for genome-wide expression studies. Mol. Gen. Genet. 2003, 269, 173-179. Brown, P. 0.; Botstein, D. Exploring the new world of the genome with DNA microarrays. Nature Genetics 1999, 21, 33-37. 219 Casu, R. E.; Grof, C. P. L.; Rae, A. L.; McIntyre, C. L.; Dimmock, C. M.; Manners, J. Identifiation of a novel sugar transporter homologue strongly expressed in maturing stem vascular tissue of sugarcane by expressed sequence tag and microarray analysis. Plant Molecular Biology 2003, 52, 371-386. Chaubron, F .; Harris, N.; Ross, H. A.; Davies, H. V. Partial purification and characterization of fructockinases from developing taproots of sugar beet (Beta vulgaris L.). Plant Science 1995, 110, 181-186. Chourey, P. S.; Taliercio, E. W.; Carlson, S. J .; Ruan, Y. L. Genetic evidence that the two isoenzymes of sucrose synthase present in deve10ping maize endosperrn are critical, one for cell wall integrity and the other for starch biosynthesis. Mol. Gen. Genet. 1998, 259, 88-96. Donson, J .; Fang, Y.; Espiritu-Santo, G.; Xing, W.; Salazar, A.; Miyamoto, S.; Armendarez, V.; Volkmuth, W. Comprehensive gene expression analysis by transcript profiling. Plant Molecular Biology 2002, 48, 75-97. Durrant, W. E.; Rowland, 0.; Piedras, P.; Hammond-Kosack, K. E.; Jones, J. D. G. cDNA-AFLP reveals a striking overlap in race-pecific resistance and wound response gene expression profiles. Plant Cell 2000, 12, 963-977. Diatchenko, L.; Lau, Y. F. C.; Campbell, A. P.; Chenchik, A.; Moqadarn, P.; Huang, B.; Lukyanov, S.; Lukyanov, K.; Gurskaya, N.; Sverdlov, E. D.; Siebert, P. D. Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries. Proc. Natl. Acad Sci. USA 1996, 93, 6025-6030. Eisen, B. M.; Spellrnan, P. T.; Brown, P. O.; Botstein, D. Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci. USA 1998, 95, 14863-14868. Elliot, M. C.; Weston, G. D. Biology and physiology of the sugar beet plant. In The sugar beet crop: science into practice; D. A. Cooke and R. K. Scott, Eds; Chapman and Hall: London, 1993; pp 37-66. Fieuw, S.; Willenbrink, J. Sugar transport and sugar-metabolizing enzymes in sugar beet storage roots (Beta vulgaris ssp. altissima). J. Plant Physiol. 1990, l 3 7, 216-223. 220 Follett, R. F .; Schmehl, W. R.; Viets, J. F. G. Seasonal leaf area, dry weight, and sucrose accumulation by sugarbeet. J. Am. Soc. Sugar Beet T echnol. 1970, 16, 23 5-252. Goldman, I. L. A list of germplasm releases from the University of Wisconsin Table Beet Breeding Program, 1964-1992. HortScience 1996, 31 , 880-881. Green, C. D.; Simons, J. F.; Taillon, B. E.; Lewin, D. A. Open system: panoramic views of gene expression. Journal of Immunological Methods 2001, 250, 67-79. Harmer, S. L.; Hogenesch, J. B.; Straume, M.; Chang, H. S.; Han, B.; Zhu, T.; Wang, X.; Kreps, J. A.; Kay, S. A. Orchestrated transcription of key pathwys in Arabidopsis by the circadian clock. Science 2000, 290, 2110-2113. Hesse, H.; Sonnewald, U.; Willmitzer, L. Cloning and expression analysis of sucrose- phosphate synthase from sugar beet (Beta vulgaris L.). Mol. Gen. Genet. 1995, 24 7, 515-520. Hesse, H.; Willmitzer, L. Expression analysis of a sucrose synthase gene fiom sugar beet (Beta vulgaris L.). Plant Molecular Biology 1996, 30, 863-872. Horvath, D. P.; Schaffer, R.; West, M.; Wisman, E. ArabidOpsis microarrays identify conserved and differentially expressed genes involved in shoot growth and development from distantly related plnts species. The Plant Journal 2003, 34, 125- 134.Jones, C. S.; Davies, H. V.; Taylor, M. A. Profiling of changes in gene expression during raspberry (Rubus idaeus) fruit ripening by application of RNA fingerprinting techniques. Planta 2000, 211, 708-714. Kayo, T.; Allison, D. B.; Weindruch, R.; Prolla, T. A. Influences of aging and caloric restriction on the transcriptional profile of skeletal muscle from rhesus monkeys. Proc. Natl. Acad. Sci. USA 2001, 98, 5093-5098. Klotz, K. L.; Campbell, L. G. Sucrose catabolism in developing roots of three beta vulgaris genotypes with different yield and sucrose accumulating capacities. Journal of Sugar Beet Research 2004, 41, 73-88. 221 Klotz, K. L.; Finger, F. Contribution of invertase and sucrose synthase isoforms to sucrose catabolism in developig sugarbeet roots. Journal of Sugar Beet Research 2002, 39, 1-24. Kuhn, E. From library screening to microarray technology: strategies to detemine gene expression profiles and to identify differentially regulated genes in plants. Annals of Botany 2001, 87, 139-155. Lemieux, B.; Aharoni, A.; Schena, M. Overview of DNA chip technology. Molecular Breeding 1998, 4, 277-289. Lewellen, R. T. Registration of rhizomania resistant, monogerm populations C869 and C869CMS sugarbeet. Crop Science 2004, 44. Lipshutz, R. J .; Fodor, S. P. A.; Gingeras, T. R.; Lockharrt, D. J. High density synthetic oligonucleotide arrays. Nature Genetics 1999, 21, 20-24. McGrath, J. M.; Derrico, C. A.; Morales, M.; Colpeland, L. 0.; Christenson, D. R. Germination of sugar beet (Beta vulgaris L.) seed submerged in hydrogen peroxide and water as a means to discriminate cultivar and seedlot vigor. Seed Sci. and Technol. 2000, 28, 607-620. Milford, G. F. J. The growth and development of the storage root of sugar beet. Ann. app]. Biol. 1973, 75, 427-438. Milford, G. F. J. Sugar concentration in sugar beet: varietal differences and the effect of soil type and planting density on the size of the root cells. Ann. appl. Biol. 1976, 83, 251-257. Milford, G. F. J.; Travis, K. Z.; Pocock, T. 0.; Jaggard, K. W.; Day, W. Growth and dry- matter partitioning in sugar beet. J. agric. Sci, Camb. 1988, 110, 301-308. Okamuro, J.; Jofuku, D.; Pannell, R.; Chen, Z.; Dang, V. D.; Donson, J.; Fang, Y.; Volkmuh, W.; F lavell, R. B. A comprehensive analysis of gene expression fiom ovule to seed and from flower to fruit. In Arabidopsis Genomics; C. S. Harbor, Ed.: NY, 2000; pp 7-10. 222 Orkwiszewski, A. J. J.; Poethig, R. S. Phase identity of the maize leaf is determined after leaf initiation. Proc. Natl. Acad. Sci. USA 2000, 97, 10631-10636. Poethig, R. S. Phase change and the regulation of shoot morphogenesis in plants. Science 1990, 250, 923-930. Pearson, G.; Serrao, E. A.; Cancela, M. L. Suppressive subtractive hybridization for studing gene expression during aerial dessication in fucoid algea. Eur. J. Phycol. 2001, 36, 359-366. Schaffer, R.; Landgraf, J .; Accerbi, M.; Simon, V.; Larson, M.; Wisman, E. Microarray analysis of diurnal and circadian-regulated genes in Arabidopsis. The Plant Cell 2001, 13, 113-123. Small, 1. D.; Peeters, N. The PPR-motif - a TPR-related motif prevalent in lant organellar proteins. Trend in Biochem. Science 2000, 25, 46-47. Trebbi, D.; McGrath, J. M. F luorometric sucrose evaluation for sugar beet. J. Agric. Food Chem. 2004, 52, 6862-6867. Ulrich, A. The influence of temperature and light factors on the growth and development of sugar beet in controlled climatic environments. Agronomy Journal 1952, 44, 66- 73. Ulrich, A. Influence of night temperature and nitroger deficiency on the growth, sucrose accumulation and leaf minerals of sugar beet plants. Plant Physiology 1955, 30, 250- 257. Watson, D. J .; Selman, I. W. A comparative phisiological study of sugar beet and mangold with respect to growth and sugar accumulation. 11. Changes in sugar content. Annals of Botany 1938, 2, 827-846. van Ginneken, P. J. H. The developmental of the sugar content of sugar beet during the growth period. Mededeelingen van het Instituut voor Suikerbeitenteelt 1959, 29, 119- 260. 223 Vega, S. H.; Sauer, M.; Orkwiszewski, A. J. J .; Poethig, R. S. The early phase change gene in maize. The Plant Cell 2002, 14, 133-147. WhiteJ.A., Todd,J., Newman,T., Focks,N., Girke,T., Martinez de Ilarduya,O., Jaworski,J.G., OhlroggeJ., Benning,C. A new set of Arabidopsis expressed sequence tags from developing seeds. The metabolic pathway fi'om carbohydrates to seed oil. Plant Physiol. 2000124 (4): 1582-1594 Poethig, R. S. Phase change and the regulation of developmental timing in plants. Science 2003, 301, 334-336. Wullschleger, S. D.; Difazio, S. P. Emerging use of gene expression microarrays in plant physiology. Comparative and Functional Genomics 2003, 4, 216-224. Wyse, R. Parameters controlling sucrose content and yield of sugarbeet roots. J. Am. Soc. Sugar Beet Technol. 1979, 20, 368-385. Wyse, R. Partitioning within the taproot sink of sugarbeet: effect of photosynthate supply. Crop Science 1980, 20, 256-258. Yang, Y. H.; Dudoit, S.; Luu, P.; Lin, D. M.; Peng, V.; Ngai, J.; Speed, T. P. Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucleic Acids Research 2002, 30, e15. 224 SUMMARY AND CONCLUSIONS (1) A new enzymatic-fluorometric assay for sucrose quantification as dry matter basis was developed. The microtiter plate format and the accuracy of the assay would allow to enhance breeding efficiency both by increasing the number of samples examined and by routinely assessing the dry matter sucrose partitioning of individuals sugar beet roots of breeding populations. (ii) A new publicly available Beta vulgaris L. genetic map based on AF LP markers was developed. The genetic map covered a total of 496.2 cM in length and showed a high density of markers on all 9 chromosomes, with average marker inter-distance of 1.5 cM. Pstl/MseI-derived AF LP markers presented a lower proportion of clustering, a higher proportion of linked markers, and a lower segregation distortion with respect to the EcoRI/MseI-derived markers, and were very efficient in generating the genetic map. (iii) In breeding populations derived by crosses from genetically diverse germplasm, the root dry matter content and the root sucrose content as dry matter were not correlated and traits variability followed a quantitative pattern. This finding implies that, particularly during the early stages of a breeding program after the introgression of favorable genes from wild germplasm, simultaneous and independent screening and selection based on dry matter content and on root sucrose content as dry matter would be more efficient in quickly recovering individuals with high sucrose content as fresh weight, with respect to the selection exclusively based on sucrose content as fresh weight itself. 225 (iv) A total of 13 genomic regions located on seven chromosomes were identified influencing root sucrose content. These 13 loci influenced root dry matter content in five instances and root sucrose content as dry matter in nine instances, with one locus influencing both traits. Seven of these loci were directly detected with the analysis of sucrose content as fresh weight, while six more loci would not be detected without analyzing the trait separately for % DM (one) and % SucDM (five). (v) Phenotypic analysis of the dynamics of root sucrose accumulation revealed a sharp increase in sucrose content fi'om the 3rd to the 7th week after emergence (WAE) in greenhouse-grown plants. Comparison between sugar beet lines with known difference of sucrose storage capability at maturity showed proportional sucrose content differences after seven and nine WAE in greenhouse-grown plants. These differences in sucrose levels as fresh weight were caused by differences in root dry matter content and not by differences in sucrose content as dry matter. Similar differences between lines for sucrose content as fresh weight and root dry matter were observed after 13 to 16 WAE in field- grown plants, indicating these traits as potential selective indicators for early greenhouse screening during breeding programs. (vi) Gene expression profiles revealed an early root developmental phase during the first three WAE characterized by the highest number of expressed transcripts and absence of accumulation of sucrose in the root. This early phase was followed by a transitional phase from the 4‘" to the 5‘“ WAE when the expression of the early-expressed transcripts terminated and the expression of late-expressed transcripts started. This transitional phase was characterized by a sharp increase of sucrose accumulation in the root system. The transitional phase was followed by a later root developmental phase afier the 6"l 226 WAE, characterized by late-specific expressed transcripts and root sucrose content values on dry and fresh weight basis similar to those observed at root maturity. The highest change in gene expression profiles was observed between the 5th and the 6th WAE. Considering the dynamics of sucrose accumulation and of gene expression profiles together, it was hypothesized that there was a developmental phase change from the juvenile vegetative phase, characterized by the lack of sucrose accumulation activity, to the adult vegetative phase, characterized by active sucrose accumulating tissues and full activity of those metabolic pathways that influence sucrose content at root maturity. This developmental phase change occurred between the 4th and the 6th WAE in the sugar beet root system. (vii) Genes differentime expressed between developmental phases characterixed by different sucrose accumulation activity were mainly represented by regulatory genes as opposed to structural genes. Transcripts up-regulated during early root development (juvenile phase) were mainly associated with cell cycle metabolism and osmotic regulation pathways, while genes up-regulated during active root sucrose accumulation (adult phase) were mainly involved in transcriptional regulation and signal transduction pathways, in energy metabolism and in metabolite transports. Arabidopsis-based microarrays were not completely efficient in detecting hybridizations and change of transcripts expression during Beta vulgaris L. gene expression analyses. These findings would increase our understanding on which metabolic pathways regulate the early stages of root development and sucrose accumulation in sugar beet. In conclusion, 13 genomic regions influenced water/dry matter relationship and sucrose partitioning of root dry matter, which could be easily estimated with the 227 enzymatic-fluorometn'c assay, that combined influenced the amount of sucrose stored in the root and the sucrose yield capability of different lines. The sugar beet genetic map developed for the QTL analysis would also be a suitable starting framework for further genetic analyses, such as map-based cloning, and integration of genetic and physical maps. Both phenotypic analysis and molecular markers associated with quantitative trait loci could be used to drive the selection of superior individuals during the early stage of breeding programs after introgression of favorable alleles, while information of the regulatory and structural metabolic pathways influencing sucrose content would increase knowledge on sucrose accumulation in sugar beet. 228 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII lllllllllllllllllllllllllllllllllllllll