._ 3. _ ...... ._ C wasp. ._ ... . p. ,0 .5 J‘ r—.. :3 ... 3.}: .A 55.6.3 v . ’2 2... <'.rv-d .314... L! . \ u .- .mWM5 r.‘ ma J.“ L“ J’fr. ‘ '9 .... :55. ,. 3.5.5451..., ...5_ . ...., . it... ......T _ . ...5 31h .... . I. . ,. .. 5, , _ . . .. ... . . . . .5 o. . 1.22.: , . _ .. .1. 55.53.. 5. .._. . _ 1.. . . .. . ... ., . . . , . . . : .. 5— . A _ .. ...”.u.» J .3... .. . _ A , A. , 950.)“ . . ...Lwrxr. . .. 094v ,. J. .5.... s u . . - HI...‘ Il.ln .¢ .H5|.|«>w.~v5..ta."—IV2 _r.‘ 2 .. -533: ...5 9.45.: ,. L1,..flvfiflng ‘ “......“ QC . .4 ah U3 . . KL? .0. ‘ F ...-.5 .... 9 rug 9* fl' ._ _, . _ . ' id_--_’.'L 524.9. - m...)- M 474.53. ' ax. _ . . . ‘2.- , QBHJ‘I-Iu-a-~~-, ...... .. .p..\¢a. (J! 9 9 ' v. fi'ii?;fi'.3'r.u&‘ 1 This is to certify that the dissertation entitled The Isolation and Characterization of Mutants of Arabidopsis thaliana (L.) Deficient in Fatty Acid Desaturation presented by Peter John McCourt has been accepted towards fulfillment of the requirements for Ph.D. degreein Genetics and Botany Major professor Date Septembe 11.29.4286 MS U i: an Affirmative Action/Eq ual Opportunity Institution 0- 12771 MSU ‘ LlBRARlES —;—-— RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. THE ISOLATION AND CHARACTERIZATION OF MUTANTS OF ARABIDOPSIS THALIANA (L.) DEFICIENT IN FATTY ACID DESATURATION BY Peter John McCourt A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Genetics Program and Department of Botany 1986 ABSTRACT THE ISOLATION AND CHARACTERIZATION OF MUTANTS OF ARABIDOPSIS THALIANA (L.) DEFICIENT IN FATTY ACID DESATURATION BY Peter John McCourt In plant membranes each lipid class has a characteristic fatty acyl composition defined by chain length and degree, position, and stereochemistry of unsaturation. However, with few exceptions the functional significance of lipid acyl unsaturation remains uncertain. The major goal of this study was to isolate and characterize a series of mutants of the crucifer Arabidopsis thaliana (L.) Heynh. with specific alterations in leaf fatty acid composition. The mutants were isolated without selection by direct analysis of the leaf fatty acid composition of individual M2 plants using gas chromatography. From approximately 2000 plants examined by this procedure, seven lines were isolated which showed stably inherited changes in fatty acid composition and of these, two were further analyzed at both the biochemical and physiological level. The first mutant characterized was completely lacking the acyl group trans hexadecenoic acid due to a mutation at a single nuclear locus designated fagA. This fatty acid which is only found in chloroplast membranes in most higher plants was thought to play an Peter John McCourt important role in various aspects of photosynthesis. However, detailed study of photosynthetic function in the fadA mutant suggests the role of trans hexadecenoic acid in chloroplast function is, at best, subtle. The second mutant studied was characterized as deficient in both hexadecatrienoic (16:3) and linolenic (18:3) fatty acids. Both alterations were due to a single nuclear mutation at a locus designated fadg. This mutation affects the fatty acid composition both inside and outside the chloroplast and appears to be temperature sensitive. The multiple changes seen in the fggp mutant can be explained if the fagfl gene product is found in different cellular compartments or the product of the reaction it governs is transported between compartments. The fadfl mutation appears to have little or no functional effect on photosynthesis but does alter chloroplast size, number and ultrastructure. TABLE OF CONTENTS LIST OF TABLES .................................................. v LIST OF FIGURES ................................................. vii CHAPTER 1 LITERATURE REVIEW ..................................... 1 Introduction ................................................ 1 Lipid Struture ............................................ 3 Lipid Biosynthesis ........................................ 7 16:3 and 18:3 Type Plants ................................. 8 CHAPTER 2 MATERIAL AND METHODS .................................. 18 Growth Conditions ......................................... 18 Gas Exchange .............................................. 20 Preparation of Chloroplast Membranes ...................... 21 Electron Transport ........................................ 23 Room Temperature Flourescence Measurements ................ 24 Low Temperature Fluorescence ............................... 24 Movement of LHCP .......................................... 26 Temperature Induced Fluorescence Yield Enhancement ........ 27 Fluorescence Polarization Measurements .................... 27 Extraction and Analysis of Chl, Lipid and Protein ......... 28 Pigment Protein ElectrOphoresis ........................... 30 Electron Microscopy ....................................... 32 Fatty Acid Methyl Formation ............................... 32 Analysis of Lipid Composition ............................. 33 ii CHAPTER 3 MUTANT ISOLATION ...................................... 35 Introduction ................................................ 35 Results ..................................................... 38 Mutant Isolation .......................................... 38 Group I Mutants ........................................... 40 Group II Mutants .......................................... 42 Group III Mutants ......................................... 46 Double Mutants ............................................ 46 CHAPTER 4 PHYSIOLOGICAL AND BIOCHEMICAL STUDIES OF A MUTANT LACKING TRANS HEXADECENOIC ACID ................................. 53 Introduction ................................................ 53 Results ..................................................... 55 Biosynthesis of Trans-C 16:1 .............................. 55 Cellular and Biochemical Studies .......................... 58 Photosynthetic Studies .................................... 67 Fluorescence Spectra ...................................... 67 Fluorescence Induction .................................... 72 Effects of High Temperature on Fluorescence ............... 76 Discussion .................................................. 78 CHAPTER 5 A MUTANT DEFICIENT IN 18:3 AND 16:3 FATTY ACIDS ....... 83 Introduction ................................................ 83 Results ..................................................... 85 Biochemical Characterization .............................. 85 Fatty Acid Composition of Individual Lipids ............... 88 Lipid Composition of Roots, Seeds and Callus .............. 92 Discussion .................................................. 94 CHAPTER 6 PHYSIOLOGICAL CONSEQUENCES OF UNSATURATION ............ 100 Introduction ................................................ 100 Results ..................................................... 103 Growth Rate of Mutant and Wild Type ....................... 103 Effects of Trienoic Acid Composition on Chl, Protein and Lipid Content ......................................... 103 Chloroplast Ultrastructure ................................ 107 Photosynthetic Studies .................................... 109 Fluorescence Measurements ................................. 115 Membrane Fluidity ......................................... 117 Effects of Unsaturation on Protein Diffusion .............. 119 Effects of High Temperature on Fluorescence ............... 120 Discussion .................................................. 120 APPENDIX A ...................................................... 128 Temperature Induced Fluorescence .......................... 128 APPENDIX B ...................................................... 136 Chromosome Assignment ..................................... 136 BIBLIOGRAPHY .................................................... 139 iv 10 ll 12 13 LIST OF TABLES Fatty acid nomenclature ................................. List of plant lines ..................................... Concentrations of solutions used to isolate thylakoid membranes ..................................... Concentration of solutions used for electron transport assays ........................................ Concentration of solutions used in pigment-protein gel and buffer systems .................................. Percent of total fatty acids in Group I (lacking trans-C , ). Each value is the average of ten plants 49:! .............................................. Percent of total fatty acids in Group II mutants (deficient in 16:3 and 18:3) grown at 18 C and 27 C for three weeks ......................................... Percent of total fatty acids in wild type, J8], and F1 progeny from a JBI x wt cross grown at 28 C for three weeks ............................................. Percent of total fatty acids in Group III mutants deficient in 16:3 biosynthesis .......................... Percent of total fatty acids in double mutants LIPl (JBl x 0825) and LIP2 (JB60 x JBl) grown at 28 C .................................................... Percent of total fatty acids in double mutants LIPl (JBl x 0825) and LIP2 (JB60 x 081) grown at 19 C .................................................... Room temperature fluorescence induction parameters of isolated thylakoids from mutant 0860 and wild type Arabiggpsis in the presence and absence of 5mm MgCl2 .... Fatty acid composition of wild type and J8] thylakoid membranes from plants grown at 28 C ..................... 4 19 22 25 31 41 43 47 48 75 89 14 15 16 17 18 19 20 21 Fatty Acid compositions of leaf lipids from wild-type and mutant JBl Arabidopsis, grown at 28 C ............... Fatty acid composition of roots, seeds and callus tissue of wild type and J81 Arabidopsis grown at 23 C ........................................... Relative amounts of Lipid, Protein, and Chl in mutant J8] and wild type Arabidopsis leaves and thylakoids ..... Morphometric analysis of chloroplasts from mutant 081 and wild type Arabidopsis ........................... Number of chTOproplast per protoplast in both mutant 081 and wild type Arabidoosis grown at 19 C and 27 C .................................................... Comparison of photosynthetic activities in mutant J81 and wild type Arabidopsis ........................... Room Temperature (Fv/Fo) and Low Temperature (F685/F734) flourescence of Isolated thylakoids from mutant 081 and wild type Arabidoosis ............................... APPENDIX Temperature induced fluorescence breaks and unsaturation ratios in wild type plants grown 12 and 22 C ............................................. vi 9O 93 105 110 112 114 116 130 10 11 12 LIST OF FIGURES Lipid biosynthesis in 18:3-type plants ................. 10 Lipid biosynthesis in 16:3-type plants ................. 11 A typical gas chromatography tracing of fatty acids extracted from a single wild type leaf ................. 39 The effect of light intensity on the proportion of 18:2 (---) and 18:3( ) in leaves of mutant 081 (0) and wild type (0) Arabidopsis .................. 44 The effect of temperature on the proportion of trienoic fatty acids in leaves of mutant JBl (O) and wild type (0) Arabidopsis .............................. 45 Fatty acid composition of PG from wild type (a) and mutant and mutant JB60 (b) Arabidopsis ............. 57 Transmission electron micrographs of ulrathin (80 to 100 nm) sections of whole leaves from wild-tupe (A) and mutant JB60 (B) Arabidopsis at x22,500 magnification ............................... 59 Pigment protein complexes of mutant JB60 (lane 1) and wild type (lane 2) thylakoids ...................... 61 Effects of NaCl concentration on the proportion of Chl in LHCP oligomer in mutant JB60 (O) and wild type (0) Arabidopsis .............................. 63 Absorption spectra of mutant JB60 (---) and wild type (---) chlorophyll-protein complexes resolved by SOS—PAGE ................................... 65 Protein composition of mutant JB60 (lane 4) and wild type (lane 5) thylakoids and purified LHCPI and LHCP3 from JB60 and wild type separated by SOS-PAGE ............................................... 66 The effects of light fluence on photosystem I (a) and photosystem II (b) activity of thylakoids from wild type (0) and mutant JB60 (0) ................. 68 vii 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Chlorophyll fluorescence spectra of whole leaves from mutant (---) and wild type (---) Arabidoosis at 77K ................................................. Chlorophyll fluorescence spectra of chloroplasts from mutant (~--) and wild type ( ) Arabidopsis in the presence (A) and absence (8) of MgCl2 (SmM) ..... Ratio of fluorescence intensity from low temperature (77K) emission spectra of wild type (0) and mutant JB60 (0) thylakoids .................................... Temperature-induced fluorescence enhancement yield (F0) of wild type (0) and mutant JB60 (0) leaves ....... Ratio of the amount of trienoic fatty acids in mutant line 081 and wild type after shifting plants from 27 C to 19 C ...................................... Possible mechanism of 18:3 exchange and transport from the chloroplast for acylation to cytoplasmic lipids ................................................. Effect of temperature on the relative growth rate of wild type (0) and mutant (0) Arabidopsis ............... Chl and trienoic acid content following a shift of wild- -type (0) and mutant (0) plants from 19 C to 27 C OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO C0 fixation rates versus light fluence for mutant 1 (0) and wild type (0) Aradopisis on a chlorophlyll and fresh weight basis ................................. Effect of temperature on DPH fluorescence polarization by thylakoid membranes from wild-type (0) and mutant (0) 081 Arabidopsis .................................... Rate of ATP-induced change in the ratio of 77K fluorescence (F734/F685) from thylakoid membranes of wild-type (0) and mutant 081 (0) Arabidopsis ........ Temperature~induced fluorescence enhancement yield (fo) of wild-type (0) and mutant (0) leaves ................. APPENDIX Low temperature fluorescence (77K) emission spectra of wild type leaf samples incubated at 22 C ( ) or 53 C (~--) for 5 min before freezing of the samples Low temperature fluorescence (77K) emission spectra of wild type thylakoids incubated at 22 C ( ) and 53 C (---) for 5 min before freezing of the samples viii 70 71 73 77 87 98 104 111 118 120 122 131 133 27 Low temperature fluorescence (77K) emission spectra of wild type thylakoids incubated at 22 C ( ) or 2, 5, or 10 min at 53 C (---) before freezing the samples ................................................ ix 134 CHAPTER 1 LITERATURE REVIEW Introduction The most satisfactory explanation of how biological membranes can have enough stability to be an impermeabilty barrier between two compartments yet maintain the flexibility to selectively transport metabolites and macromolecules between such compartments has been the fluid-mosaic model of membrane structure proposed by Singer and Nicolson (1972). This model suggested that the lipids in biological membranes are organized as bilayers in which proteins needed for transporting ions, metabolites and macromolecules are embedded. While such a view implies heterogeneity at the protein level, it suggests that any lipid which can form a bilayer structure forming a barrier to polar or large nonpolar molecules under physiological conditions would be adequate for membrane function. Thus, for example, a phosphatidylcholine (PC) bilayer is not distinguished from a phosphatidylserine (PS) bilayer. However a survey of the lipid membrane composition of various cell types or even organelles within a particular cell type suggests that the situation is more complex. In plants, many of the lipids found in the thylakoid and envelope membrane are not found elsewhere in the cell and vice versa (Harwood, 1980). Furthermore, each lipid class has a characteristic fatty acyl composition which is defined by chain length and degree, position and sterochemistry of unsaturation. However with a few exceptions the functional significance of these complex lipids and fatty acids remains uncertain in both animals (Stubbs and Smith, 1984) and plants (Quinn and Williams, 1983). The major goal of this work was to use a genetic approach to understand the regulation and functional significance of lipid diversity. In principle, it should be possible to remove or to alter each individual class of lipid and fatty acid by mutation and then assay the physiological effect of this change on the organism. The use of such an approach would not only enable one to determine the role of specific lipids in membrane fuction but also, such mutants would be very useful in elucidating how these molecules are made. Because the enzymes that synthesize lipids and desaturate fatty acids are usually membrane-bound it has been difficult to characterize these proteins by conventional biochemical methods. This study describes the isolation and preliminary characterization of a number of mutants in Arabidopsis thaliana (L.) Heyhn. with altered leaf lipid metabolism. These mutants which are the first reported in higher plants have been used to formulate models of lipid biosynthesis and to evaluate the physiological role of particular lipids. Because the only changes in lipid metabolism in these particular mutants are in the leaf tissue, the review of glycerolipid biosynthesis has been limited to photosynthetic tissue. i i tr tur The major fatty acids of leaves are even numbered long chain hydrocarbons which have been classified into two distinct groups. The saturated fatty acids contain no double bonds in their carbon chains. Palmitate (hexadecanate) and stearate (octadecanoate) are the most prevalent members of this group making up roughly 10% and 2%, repectively of the total fatty acids of the leaf (Harwood, 1980). The other group, which are called unsaturated fatty acids, contain one or more double bonds in their acyl chains and are usually the most abundant fatty acids of the leaf. For example, linolenate (§1§_9,12,15 octadeca-trienoate) which has double bonds between the 9th and 10th, 12th and 13th, and 15th and 16th carbons of the fatty acid from the carboxyl end, can represent up to 80% of the total leaf fatty acid in some plant species (Harwood, 1980). For the purpose of this study a shorthand nomenclature will be used to symbolize the various fatty acids (Table 1). For example, palmitate which is a 16 carbon saturated fatty acid is designated 16:0 whereas oleate, which contains 18 carbons with one double bond in the 9th position, is designated 18:1. Normally fatty acids are attached to a glycerol molecule in two of the three hydroxyl positions. The third hydroxyl group is usually esterified with a sugar residue (glycolipid) or a phosphatidic acid derivative (phospholipid). These sugar and phosphate moieties, termed the head group, represent the polar part of the lipid and face out into the aqueous solution when in a bilayer. The presence of three esterified hydoxyl groups has allowed the naming of the two fatty acid esters by optical isomerism. In all cases studied in photosynthetic tissue the fatty acids are esterified to the Table 1 Fatty acid nomenclature Symbol Systematic Name Common Name Saturated 16:0 n - Hexadecanoic Palmitic 18:0 n - Octadecanoic Stearic Unsaturated trans-3 16:1 trans - Hexadecanoic cis 7 16:1 n - Hexadecamonoenoic Palmitoleic cis 7,10 16:2 n - Hexadecadienoic cis 7,10,13 16:3 n - Hexadecatrienoic cis 9 18:1 n - Octadecamonoenoic Oleic cis 9,12 18:2 n - Octadecadienoic Linoleic cis 9,12,15 18:3 n - Octadecatrienoic Linolenic sn-I and 53-2 positions of the glycerol backbone and the head group is attached to the 3rd position. For example, monogalactosyldiacylglycerol (MGD) contains fatty acids in position sn-l and sn-Z and a galactose moiety esterified to position sg-3. A survey of the leaf lipid and fatty acid composition from one plant species to another shows a very similar distribution (Harwood, 1980). The most striking feature is the amount of glycolipid present in leaves. These lipids, which can be classified into three categories based on head group, are predominantly found in the chloroplast. MGD and digalactosyldiacylglycerol (060), which have one and two galactose residues respectively, represent roughly 40% and 25% of the total lipid of the cell and sulfoquinovosyldiacylglcerol (SL), which has 6-sulfo-deoxy-D-glucose as a head group, represents 5 to 10% of the total cellular lipid (Harwood, 1980; Barber and Gounaris, 1986). The fatty acid composition of these three glycolipids varies substantually. For instance MGD and 060 are almost completely composed of 18:3 or 16:3 acyl chains while a high pr0portion of the acyl chains of SL are saturated (Quinn and Williams, 1983). The high degree of unsaturation on MGD gives this lipid a cone shaped structure which prevents the formation of bilayers from pure preparations (Murphy, 1982). However, when mixed with proteins and other lipids normally found in the chloroplast, MGD spontaneously integrates into the bilayer (Quinn and Williams, 1983). This has led to the suggestion that this lipid may play a role in regulating insertion and packing density of proteins in the membrane (Murphy, 1982; Gounaris and Barber, 1983) SL, which is a negatively charged lipid, has been proposed to associate closely with photosystem II and coupling factor, suggesting it might be a boundary lipid (Barber and Gounaris, 1986). A boundary lipid is usually defined as a specific lipid which associates with a protein causing a stimulation of catalytic activity. Although hypothesized in plants, direct proof of boundary lipids has not been obtained. The phospholipids in leaf tissue fall into five major groups, phosphatidylglycerol (PG), phosphatidylethanolamine (PE),phosphatidylserine (PS), phosphatidylcholine (PC), and phosphatidylinositol (PI). PG and PC which form 10 to 15% of the total lipid and PI which is roughly 3%, are found throughout the cell. However, PE and PS are not found in the chloroplast and are primarily located in the mitochondrion (Harwood, 1980). The PC and PI that are found in the chloroplast are located solely on the outer chloroplast membrane (Dorne et a1, 1985). The PG is evenly distributed in both inner and outer envelopes and the thylakoids. Chloroplastic PG is of additional interest due to the occurance of trans-3-hexadecenoic acid (trans-C16:l) at the second position of PG (Dubacq and Tremolieres, 1983). This fatty acid is unusual due to the position and conformation of the unsaturation (trans versus cis). In studies with phospholipases specific for the sn-2 position, it has been possible to show that trans-C16:l occurs only on the side of the thylakoid bilayer which faces the stroma (Unitt and Harwood, 1984). This specific location and the comigration of trans-C16:1 with the light harvesting complex (LHCP) in SDS polyacrylamide gels under conditions which maintain the integrity of chlorophyll-protein complexes have led to the suggestion that this fatty acid is essential for efficient membrane insertion and operation of this chlorophyll-protein complex in light capture (Dubacq and Tremolieres, 1983) Lipid Biosynthesis The observation that isolated chloroplasts readily synthesize 16:0, 18:0 and 18:1 from acetate in the light (Slack gt g1, 1977; Roughan and Slack, 1982) and that acyl carrier protein (ACP), a specific marker for fatty acid synthesis, is solely located in the chloroplast (Ohlrogge gt g1, 1979) has led to the conclusion that the chloroplast is the only site of gg novo fatty acid biosynthesis in higher plant leaves. Therefore, it follows that all fatty acids found outside the Chloroplastic compartment must be transported there from the chloroplast. Although the exact mechanisms that regulate the synthesis and distribution of fatty acids in plants are not clear, a number of studies have resulted in the formation of a rough outline of the overall process. Stumpf and colleagues partially purified the fatty acid synthetase (FAS) from Spinacia oleracea (spinach) and have shown this enzyme complex is comprised of six loosely associated polypeptides which can catalyse the condensation of malonyl CoA to give 16:0—ACP and 18:0-ACP thioesters (Shimakata and Stumpf, 1982). The molecular organization of the FAS in plants appears to be quite similar to the figttgrigng gglt FAS, and E. ggll ACP can be substituted for the spinach ACP in reconstitution studies even though the two forms show poor immunological crossreactivity (Ohlrogge gt g1, 1979; Simoni gt 11, 1967). Interestingly, the converse experiment in which E. ggll ACP is replaced by spinach ACP in a bacterial synthetase reconstitution system yields C12 through £18 fatty acids instead of the normal 16:0 and vaccinic (g1; 11,18:1) bacterial fatty acids (Simoni gt g1, 1967). This suggests ACP may play an important role in chain termination reactions of the FAS. Detailed studies of fatty acid synthesis in spinach have determined the existence of two B-keto-acyl-ACP synthases (Shimakata and Stumpf, 1982). One of these participates in the formation of 16:0 ACP from acetyl-ACP and the other, which can only use 14:0-ACP or 16:0-ACP as a substrate, produces only 18:0-ACP when added to a reconstitution system including both ACP synthases (Shimakata and Stumpf, 1982). The implication of such results is that regulation of these two B-keto-acyl-ACP synthases could control the ratio of 16 to 18 carbon fatty acids within the chloroplast. Although roughly 80% of the 16:0-ACP synthesized in the chloroplast is converted to 18:0-ACP by the second B-keto-acyl-ACP synthase almost all the 18:0-ACP is metabolized to 18:1 ACP via a highly active NADPH dependent stromal desaturase (Nagai and Block, 1968). Hence, the major labeled products of intact chloroplasts supplied with radioactive [14C]acetate are 16:0-ACP and 18:1-ACP. 1§t§_gnd 18:3 Type Plants The further metabolic fates of 16:0-ACP and 18:1-ACP made within the chloroplast appear to be dependent on the plant species. Following a detailed characterization of the leaf lipid composition of a variety of species, a pattern has emerged which has allowed the classification of plants into two groups. Those containing hexatrienoic acid (16:3-type plants) and those that do not (Roughan and Slack, 1984). Apiggggg, Chenopodiaceae, Solanaceae and the Brassicaceag, which all belong to the former group, contain 50 to 60% 18:3 and 5 to 40% 16:3. The latter group, which include the families Fabaceae, Asteraceae, and Egggggg, have no 16:3 and very high levels of 18:3 (80% or more) hence they are given the name 18:3 type plants (Heinz and Roughan, 1983). The reason for these differences in acyl chain composition is directly related to the way these two groups of plants partition their newly synthesized fatty acids within the cell (Figures 1 and 2). In short term labelling studies of intact leaves with [14C]acetate, 18:3 type plants (Figure 1) distribute the majority of the labeled fatty acids into phosphatidyl choline (PC) and a minor proportion into phosphatidyl glycerol (PG). 0n the other hand 16:3 type plants (Figure 2) produce approximately equal amounts of PC and the chloroplast-specific lipid monogalactosyldiacylglycerol (MGD) and a minor amount of PG under the same labelling conditions (Roughan and Slack, 1982). In leaf tissue, the endoplasmic reticulum is considered to be the major site of plant phospholipid metabolism and PC synthesis, in particular (Moore, 1984; Mudd, 1980). Hence, fatty acids made in the chloroplast appear to be exported to the ER where they become esterified to PC. In yttrg labelling studies provide some insight as to how this transfer might occur. Incubation of isolated chloroplasts with CoA and ATP cause the accumulation of 18:1-CoA esters in the outer envelope (Roughan gt g1. 1976; Roughan gt g1. 1979). This and the recent report of the localization of an acyl CoA synthetase to the outer envelope (Andrews and Keegstra, 1983) have led to the suggestion that newly synthesized fatty acids are liberated from the ACP and diffuse into the lipid bilayer. Upon movement to the outer envelope the acyl CoA synthetase 10 CYTOPLASM CHLOROPLAST PE DAG > PC I————T'_ 1"_'—T_ 10.1 18:1 18:1 18:1 (15.0) (111:0) PL PG H CDP‘ChOIine 1 1 : pA - COP fi f : DAG 15:2 15:2 1 : : . 15.2 15:2 (:15) ‘3' “1°, 1 (16:0) 1 GSP r——"r— 15:3 111:: MGD (16:0) acyl-CoA acyl-CoA , 15:2 15:2 1 (1620) l 1 1 18:0 ACP—>18:1 ACP——O FA MGD ————F DGD T/ r :15:1 15;c 15:3 15:3 15:3 15:3 (16:0) (16:01 16:0 AC? 63? FAS PG f——7"\_- 18:1 16201620 F—7T' 15.115;o11s;1_ r———/T‘ 15-215;0116;1 f__7"\_' 16 31601161 CHLOROPLAST Figure 1. Lipid biosynthesis in 18:3-type plants 11 CYTOPLASM CHLOROPLAST P18. DAG 15r_———r_ 1 15ft— 1 115 0) (15: 01 Pl PG ¢——>—\PI: CCP- choline COP t DAG 15:2 15: 2 15.1 15:1 . 15.2 15:2 (15:01 “1'0, 1 115:0) l GSP r———1—- 15:3 15:3 MGD (18:01 acyl-GOA acyl-CoA 15:2 15:2 18::0ACP—D181ACP—t PA LAC. >1181 18:0 1:8118:018:0 18:1 18:0 r———7"<- 18.1 181011821 1 r-—7R:— 18:218:0116:1 1 / 18 31810118;1 CHLOROPLAST G3? FAS UDP- gal UDP MGD -—+———+.DGD 18:3 18:3 18:3 10:3 (18:0) (18:0) 181 18: O 181 18 0 F——T— 18:1 - 18:1 1 1 1 .. l_———T_ 1"—"'1'_ 18:2 18:2 18:2 102'.) 1_'—T— r—r— 18:3 18:3 18 3 16.0 Figure 2. Lipid biosynthesis in 16:3-type plants 12 esterifies the fatty acid to 60A making it soluble for cytoplasmic transport to the ER. Although the mechanism of transfer of fatty acids across the envel0pe is unclear, the existence of a carnitine acyl transferase in the chloroplast suggests the mechanism may be similar to that characterized in animal mitochondria (McLaren gt g1, 1986). Until recently, the specificities of the enzymes which attach the exported 16:0 CoA and 18:1 CoA to glycerol-B-phosphate (G-3-P) in the cytoplasm could only be inferred from positional analysis of fatty acids attached to microsomal PC. However, purified microsomal fractions from Etggm ggttxgm (pea) leaves show two acyl CoA G-3-P acyltransferases (Frentzen gt g1, 1984). One acylates either 16 or 18 carbons to the gg-l position of G-3-P to form lysophosphatidic acid (LPA). The other acyl transferase, only esterifies 18 carbon fatty acids to the 53-2 position (Frentzen gt g1, 1984) giving phosphatidic acid (PA). Hence, PA formed via these cytoplasmic acyltransferases contains either 18 carbon fatty acids on the both 53-] and gg-Z positions or 16:0 on the gn-l position and an 18 carbon fatty acid on the gg-Z position (Figure 1). Both molecular species of PA can be used in the biosynthesis of PC and phosphatidylethanolamine (PE) but only gg-l 16:0 gn-Z 18:1 PA is used for phosphatidylinositol (PI) synthesis (Roughan and SLack, 1984). These results indicate that the fatty acid composition of a particular lipid can play an important role in channelling lipids into various pathways. The two molecular Species of PC are also further channelled according to their fatty acid composition. In 211g pulse chase experiments using [14C]acetate have demonstrated that radioactivity can be chased into chloroplastic MGD from PC in 18:3 plants (Roughan, 13 1975). Similar studies using double labelling in which fatty acids are labelled with 146 and the glycerolipid backbone with 3H have demonstrated that both labels are chased from PC into MGD at equal rates, suggesting that a complete lipid molecule is transferred in this process (Slack gt gt, 1977). However, 18:3 plants contain no 16:0 MGD indicating that tgg-l 16:0, gg-Z 18:1) PC is excluded by this transfer mechanism. The discovery of phospholipid exchange proteins (PLEP) in plants has suggested a possible mechanism of PC transport from the ER to the chloroplast (Ohnishi and Yamada, 1982). These proteins, are able to catalyse the exchange of phospholipids between microsomes and mitochondia and microsomes and chloroplasts (Kader gt g1. 1984) The evidence for PLEPs being involved in the transport of PC is that chloroplastic PC is only found in the outer envelope. By contrast, the apparent absence of phosphatidylcholine phosphatase activity and the inability of PLEPs to generate net membrane growth (since they can only exchange lipids), argues against such a mechanism. Results showing a rapid equilibrium between PC and diacylglycerol (DAG) in the ER of expanding maize leaves could mean that PC is converted to DAG in this compartment and that this species is transported to the chloroplast (Roughan and Slack, 1984). DAG would then be further metabolized to MGD via a well characterized chloroplast specific galactosyl transferase which uses cytoplasmic UDP-galactose as a second substrate (Block gt g1, 1983; Heemskerk gt g1, 1985). In principle, MGD can then be converted to digalactosyldiacylglycerol (DGD) by the addition of another galactose residue. However, this reaction has been difficult 14 to characterize in isolated chloroplasts and the exact nature of the galactose donor is not known. Acyl analysis from [l4CJacetate in vivo labeling experiments has suggested that PC is the major site of 18:1 desaturation rather than oleoyl-CoA (Slack gt g1 1977). This observation was verified directly in purified microsomal fractions by showing [14C]oleoyl-CoA was first esterified to lyso PC before it was desaturated (Murphy gt g1, 1983). Therefore in 18:3 plants it appears that 18:1 and 18:2 desaturations occur when the fatty acid is attached to the lipid. In summary, 18:3 plants appear to export the majority of their newly synthesized fatty acids (16:0, 18:1) from the chloroplast to the ER where they are attached to phospholipids. The acyl transferases which carry out these esterifications attach 16:0 only to position 1 and 18 carbon fatty acids to either position. Cytoplasmic 16:0 is not desaturated futher via this pathway while 18:1 can be converted to 18:2 and 18:3. Microsomal PC, which appears to be a sink for fatty acid acylation is also the major site in 18:3 plants for 18:1 to 18:2 desaturation. Upon desaturation, some form of lipid (possibly DAG with 18:2, 18:2 acyl side chains) is transported back to the chloroplast where it is converted to MGD and further desaturated to 18:3. MGD can be further modified to DGD by addition of another galactose residue to the head group. Because a large proportion of this pathway is located in the cytoplasm, it has been termed the eukaryotic pathway of lipid synthesis. In contrast to 18:3 type plants, 16:3 plants appear to have two ways of synthesizing galactolipids (Figure 2). The second pathway of synthesis is termed the prokaryotic pathway since it is limited to the 15 chloroplast (Roughan and Slack, 1982). After formation of 16:0 ACP, 18:0 ACP, and 18:1 ACP by the FAS, 16:3 plants can esterify the acyl chains directly to G-3-P within the chloroplast. Although these reactions also occur in 18:3 plants, the activities of the acyl transferases are relatively low compared to those of 16:3 species (Heinz and Roughan, 1983). The relatively low amount of PA that is made through this pathway in 18:3 plants is believed to be used only for chloroplastic PG synthesis (Andrews and Mudd, 1985). The acyl transferases that function in the prokaryotic pathway are quite different from the eukaryotic acyl transferases in terms of specificity. The prokaryotic enzyme is a stromal soluble protein which prefers 18:1-ACP to 16:0-ACP as a substrate (Gardiner gt g1, 1984). The product of this reaction, LPA, is soluble in the membrane and is converted to PA by an gg-Z specific membrane bound acyltransferase (Frentzen gt g1, 1983). Hence PA and subsequent lipids made from it via the prokaryotic pathway can be easily distinguished from lipids of eukaryotic origin by the position of the 16 carbon fatty acids. For example, 16:0 on the 53-1 position and an 18 carbon fatty acid on the 53-2 position of PA means it was synthesized by the eukaryotic pathway, while a lipid with the inverse arrangement of fatty acids would be synthesized via the prokaryotic pathway. The fact that the two prokaryotic acyltransferases only use fatty acids attached to ACP implies that removal of the ACP could cause the fatty acid to enter the eukaryotic pathway. The localization of an acyl-CoA thioesterase, which converts acyl-ACP to acyl-CoA, in the inner envelope of the chloroplast may play a role in the partioning of fatty acids between the two pathways (Andrews and Keegstra, 1983; Block 16 gt 31, 1983). Although prokaryotic PA is also used to make chloroplastidic PG in 16:3 plants (as in 18:3 plants) the majority is hydrolyzed to form DAG via a phosphatidic acid phosphatase (Gardiner and Roughan, 1983). This DAG which can readily equilibrate with eukaryotic DAG (Figure 2) in the chloroplast membrane can be further metabolized to MGD and DGD by a similar mechanism as described for the eukaryotic pathway (Roughan and Slack, 1982). By determining the 53 position of the 16 carbon fatty acids MGD and DGD, estimates suggest 50 to 60% of these lipids are made through the prokaryotic pathway in 16:3 type plants (Roughan and Slack, 1984). Furthermore, the 16:0 which is attached to position gg-Z of MGD is sequentially desaturated to 16:3. Although the 16:0 to gig 16:1 desaturation must be MGD specific, since it only occurs on that lipid, the subsequent 16:2 and 16:3 desaturations could be carried out by either 16 carbon specific desaturases or the desaturases which perform 18:] and 18:2 desaturations. As stated earlier, chloroplastic PG in both 18:3 and 16:3 plants is made through the prokaryotic pathway (Figures 1 and 2) and a number of studies have reported PG synthesis in isolated chloroplasts (Sparace and Mudd, 1982; Andrews and Mudd, 1985). Recently it has been shown that prokaryotic PA made in the envelope can be converted to GDP diacylglycerol using CTP and that this in turn reacts with G-3-P to form PG (Andrews and Mudd, 1985; Roughan, 1985). The synthesis of SL, which is not shown in Figures 1 or 2, has attracted considerable attention, but the biosynthetic pathway remains uncertain (Barber and Gounaris, 1986). It has not yet been established whether or not the chloroplast is autonomous for SL synthesis, and the 17 results of positional analysis of fatty acids esterified to SL varies from species to species. Therefore, a general statement about the biosynthetic pathway of this lipid cannot be made (Harwood, 1980). CHAPTER 2 MATERIALS AND METHODS Growth Conditions Argbidopsis thaliang (L.) Heynh., race Columbia, was used in this study. The seed source for the mutant search was an M2 population of ethyl methane sulfonate treated seed (Somerville and Ogren, 1979). A list of the plant lines used in this study which were either isolated from an M2 population or constructed from crosses is given in Table 2. MZD represents a particular batch of seed that was mutagenized. Before physiological studies were performed, the mutants were backcrossed to wild type and mutant lines were reisolated in the F2 progeny. This procedure was repeated and the subsequent reisolated mutant lines were advanced to the F4 generation. All plants were grown under continuous fluorescent illumination (150 - ZOO uEinsteins m'2 5.1) at 19 C and 60% relative humidity in a mixture composed of equal parts of vermiculite, perlite and sphagnum irrigated with mineral nutrients (Somerville and Ogren, 1979) 18 19 Table 2 List of plant lines Line Genotype Source / Derivation JBI tgg_ M2D seed JBZS ggg M2D seed JBZ7 fggA M2D seed JB60 tggA M2D seed JBlOl jggg gl;t J81 x CS2 JB60] tgdA 91;; JB60 x CSZ LIP1 ggg :ggg JBl x 0825 LIP2 thA thD JB60 x JBI MKl g_ 91 g; ql-l cer-Z Ms/ms M. Koorneef 20 For measurements of growth rates, seeds were germinated at 19 C on 100 x 25 mm petri plates containing 25 ml of mineral nutrients solidified with 0.7% (w/v) agar at a density of 16 evenly spaced plants per dish. The light intensity was 120 uE 1n'2 s'l. After seven days the temperature was adjusted as noted in the results. Because of the restricted air flow in the petri plates actual leaf temperature may have been slightly higher than the air temperature in the chamber. At three day intervals, for the next 21 days, plants were removed and the fresh weights of the aerial portion were determined. The relative growth rate was determined as the slope of the log of the fwt (mg) plotted against time since plating (days). For tissue culture studies, callus was induced from leaves by placing sterile tissue on PGl media of Negrutui gt g1 (1975). Plates were incubated in the light at 23 C for about two weeks until green callus was visible on the leaves. This callus was removed and placed on fresh PGl media and after one week the fatty acid composition was determined. Gas Exghgggg Measurements of CO2 fixation and dark respiration were obtained on intact plants of Arabidopsis by gently removing soil from the roots and placing the plants in a glass cuvette which had two ports to allow gas flow into and out of the system. Dry gas of a desired composition (350 ul 1'1 COZ’ 21% 02’ balance N2) was passed through the reference (navette of 5 Analytical Developement CO2 Infrared Gas Analyser (IRGA) Type 225. After leaving this cuvette the gas was humidified to approximately 70% RH by bubbling it through water at 20 C. The gas then 21 passed through the plant chamber which was immersed in a circulating water bath so that the temperature at which gas exchange occurs could be controlled. The gas stream was then dehumidified before entry into the analytical cuvette of the IRGA by passage through a cold finger at 4 C. The error due to transpired water is negligible if the analyser is calibrated under the same conditions of humidity and flow rate as used during photosynthesis measurements. Flow rates for this open system were determined by a Matheson mass flow meter. The rates were usually between 50 - 150 ml min'1 depending on the size of the plant. The light intensity and temperature is given with each experiment. Standardization of CO2 fixation rates were determined in three ways: leaf area which was measured on a portable area meter (LI-COR 3000, Lamba Instruments Corp), by fresh weight, or by chlorophyll concentration. Etgpgratjon Of Chloroplast Membranes For the following procedures all operations were carried out at 4 C. Usually one pot of three week old plants (4-6 grams fwt) was homogenized in 100 ml of grinding buffer by two 5 second bursts at high speed in a Osterizer galaxie blender. The homogenate was passed through four layers of cheese cloth and centrifuged at 3000 g for 5 min. The pellet was resuspended in a small volume of washing buffer using a paint brush and about 30 ml of the same buffer was then added to ensure good chloroplast breakage. The suspension was centrifuged at 5000 g for 5 min. The supernatant was then discarded and the pellet was resuspended in about 1 - 2 ml of resuspension buffer. The compositions of the various buffer solutions are given in Table 3. 22 Table 3 Concentrations of solutions used to isolate thylakoid membranes Final Concentration Stock Conc. Stock Volume/100 ml Grindinq,Buffer 50 mM Tricine-KOH (pH 8.4) 1.0 M 5.0 ml 10 mM NaCl 1.0 M 1 0 ml 400 mM Sorbitol 2.0 M 20.0 ml 10 mM EDTA (pH 8.0) 0.5 M 2.0 ml Washinq,Buffer 10 mM Hepes-KOH (pH 7.8) 1.0 M 1.0 ml 10 mM NaCl 1.0 M 1.0 ml 5 mM EDTA (pH 8.0) 0.5 M 1.0 ml Resuspension Buffer 20 mM Hepes-KOH (pH 7.8) 1.0 M 2.0 ml 10 mM NaCl 1.0 M l 0 ml 300 mM Sorbitol 2.0 M 15.0 ml 5 mM MgCl 1.0 M 0.5 ml 2 23 For chlorophyll determinations, 100 ul of sample was removed and added to 5 ml of 80% (v/v) acetone. This solution was shaken and centrifuged for 1 min at 1000 g in a clinical centrifuge. The chlorophyll concentration was determined by measuring the absorbance of the supernatant at 645 nm and 663 nm and transformed according to the formula (A645 * 20.2) + (A * 8.02) - mg Chl/ml (McKinney, 1941). 663 Electron Transport Electron transport was measured in a Rank 02 electrode by adding an aliquot of thylakoid membranes so that the final concentration was 20 ug chl ml"1 to one ml of resuspension buffer. Whole chain rates were measured by the reduction of methyl viologen (MeV). Because of MeV low redox potential MeV accepts electrons exclusively from photosystem 1 (PSI) and passes them to 02 to make a superoxide radical which spontaneously dismutates to H202 and 02. In the presence of NaN3, a catalase inhibitor, the transport of electrons from water to PSI is measured as 02 uptake (Mehler reaction). To uncouple electron transport from ATP synthesis NH4Cl and Gramicidin-D were added to the reaction. For photosystem 11 (P811) measurements, 2,5-dimethyl-p- benzoquinone (DMQ) was added as an artifical electron acceptor and to eliminate electron transport reactions associated with PSI the plastoquinone antagonist 2,5-dibromo-3-methyl-6-isopropyl -p-benzoquinone (DBMIB) was added. PSI electron transport rates were measured by inhibiting PSII activity with 3-(3,4-dichlorophenyl)-1,l-dimethylurea (DCMU). Electrons were supplied directly to PSI by reducing dichloroindophenyl (DCIP) to DCIPH with ascorbate (Asc). MeV was used as an electron acceptor from O 24 PSI as described above. Because Asc can cause superoxide formation in the prescence of MeV, superoxide dimutase ($00) was added. The solutions for the partial and whole chain assays are given in Table 4. Room Tempgtgture Fluorescence Measurements Thylakoids were diluted in resuspension buffer to a final concentration of 5 ug Chl ml'l. Illumination was by a Unitron microscope illuminator powered by a model C5-6, 6A stable output supply (Power One, Cornello CA). Chl fluorescence was measured with a photodiode positioned at a 900 angle to the actinic light source. This source was filtered through a broadband blue optical filter (Corning 4-96). The onset of illumination was controlled by an electronic shutter (Vincent Associates, Rochester NY). The fluorescence was measured through a red cut off filter (Corning 2-64) to protect against scattered blue light from the actinic light souce. The voltage output of the photodiode was stored on a Nicollet model 206 digital recording oscilloscope. Low Temperature Fluorescencg Isolated thylakoids were divided into two aliquots of resuspension buffer, one of which was supplemented with SmM MgCl2 and allowed to dark adapt for 30 min. The samples were then diluted to a concentration of 10 ug Chl ml'1 in a 60% glycerol (v/v) solution of the same buffers. Two hundred ul of this suspension was added to heat-sealed 11 inch pasteur pipets (5mm id), and kept frozen in liquid nitrogen in the dark until measurements were taken. 25 Table 4 Concentration of solutions used for electron transport assays Final Concentration Stock Solution Stock Solution/100 ml Whole Chain 0.1 mM NaN3 25 mM 4.0 ul 0.1 mM MeV 25 mM 4.0 ul 1.0 mM NH4Cl 150 mM 6.7 ul 1.0 uM Gramiciden 1 mM 1.0 ul £§_11 1.0 uM DBMIB 20 mM 5.0 ul 0.25 mM DMQ 100 mM (Methanol) 2.5 ul £§_1 2.5 mM Asc 1 M 2.5 01 1.0 ul DCMU 20 mM (Ethanol) 5.0 ul 0.5 DCIP 100 mM 5.0 ul 0.07 mg/ml $00 7 mg/ml 10.0 ul 0.1 NaN3 25 mM 4.0 ul 0.1 mM MeV 25 mM 4.0 ul 26 For whole-leaf measurements a wet leaf was placed on a metal spatula and held in place by water proof tape. The spatula was immersed in liquid NZ to freeze the sample. For measurements, the spatula was rotated while in the fluorometer to an angle approximately 45° to the excitation light. All fluorescence emission spectra were recorded using an SLM 4048 scanning spectrofluorometer (SLM Instruments, Urbana IL) operating in the ratiometric acquisition mode. Excitation was usually 480 or 440 nm light with a half band width of 8 nm. Fluorescence emission was scanned in 1.0 nm increments from 650 - 800 nm with a half band width of 2 nm. Acquisition, storage and mathematical manipulation of spectra were performed by an on line Hewlett Packard 9825 computer. Movement,0f LHCP For phosphorylation experiments of LHCP, thylakoids were resuspended in resuspension buffer to a final concentration of 200 ug Chl ml'l. To this, ATP at a concentration of 200 uM (10 ul from a 20 mM stock) and NaF to a concentration of 10 mM (20 ul of a 0.5 M stock) were added and the sample was kept in the dark for 30 min. The NaF was added to inhibit phosphatases. The samples were placed into a plexiglas waterbath maintained at 23 C and were continuously stirred on a magnetic stir plate. The assay was initiated by illumination from the 2 side at a fluence of 300 uE m' 5']. At specific times 20 ul was removed and quickly added to 1 ml of 60% (v/v) glycerol, 5 mM MgCl2 resupension buffer giving a concentration of 10 ug chl ml'l. These samples were frozen for low temperature fluorescence work and measured as described above. Movement of LHCP was determined by the change in F685 versus F734. Fluorescence at 685nm is believed to represent LHCP 27 associated with PS 11 core particles (Staehelin and Arntzen, 1983). With phosphorylation of LHCP a decrease in the ratio of F685 to F734 due to movement of the antenna away from the PSII reaction centers is observed. Measurements were done on three independent thylakoid preparations. Temperature Induced Fluorescence Yield Enhancement The method used to monitor fluorescence yield (F0) in whole detached leaves at various temperatures was a modification of the method used by Schreiber and Berry (1977). A leaf was positioned between two 45mm x 10mm strips of acetate and placed diagonally into a water filled cuvette. The cuvette was placed into the spectroflorometer so that the leaf was at a 450 angle to the excitation light. A thermocouple was attached to the leaf to monitor the temperature. The excitation light which was at 480 nm with a 16 nm half band pass was filtered with neutral density filters so that the intensity was 0.3 uE 111'2 s'l. Fluorescence emission from the leaf surface was monitored at 700 nm with a 16 nm half band width. Sample temperature was increased 1 at a rate of roughly 1.5 C min' . Measurements were made by simultaneously recording leaf temperature and fluorescence intensity. Fluorescence Polarization Measurements An estimation of the microviscosity of the thylakoid membranes was achieved by determining the steady state fluorescence polarization of the hydrophobic probe 1,6-diphenyl-1,3,5-hexatriene (DPH) inserted in this membrane. Because conditions which favour thylakoid stacking inhibit the uptake of DPH into the membrane (Barber gt g1, 1984) 28 thylakoids were isolated as usual but without MgCl2 in the resuspension buffer. DPH was added from a 3 mM stock solution in tetrahydrofuran directly to a suspension of thylakoids (50 ug ml'l chl) to a final concentration of 50 uM. After incubation in the dark for 40 min at room temperature, thylakoids were centrifuged (3000 g, 3 min), the supernatant was discarded and the pellet was respended in resuspension buffer containing no DPH to a final concentration of 10 ug chl ml'l. Fluorescence polarization was carried out on an SLM 4048 spectrofluorometer in the T format. In this conformation the intensities of the parallel and perpendicular components are measured simultaneously using two separate emission polarizers. The formula used to calculate fluorescence polarization (p) values was Ivv - Ivh (Ihv/Ihh) Ivv + Ivh (Ihv/Ihh) where I is the intensity and v and h are the positions of the polarizers (Barber gt g1, 1984). For example Ivh corresponds to a vertical excitation polarizer and a horizontal emission polarizer. Excitation was provided by 360 nm light with a half band pass of 16 nm. the two emission polarizers were at 460 nm with a half band width of 8 nm. Extraction and Analysis of Chl. Lipid and Proteins Leaves were harvested at the rosette stage (3 weeks) and either fresh weight or leaf area were determined. Individual leaves were ground in 2 ml of 95% ethanol and centrifuged at 10009 for 1 min to remove the insoluble material. Chl concentrations were determined 29 according to the formula (A654 * 39.8)/1000 - mg chl/ml (Hintermans and Demots, 1965). To the Chl fraction 0.01 mg of 17:0 fatty acid methyl ester (FAME) in one ml of hexane was added and the suspension was dried down under nitrogen gas. Fatty acid methyl esters of the dried samples were prepared by transesterification in hot methanolic-HCl as described elsewhere. The known amount of 17:0 standard added was used as an internal standard to determine the amount of total fatty acid in the leaf. For measurements of Chl, lipid and protein of fractionated samples, leaves from a pot of plants (4 to 6 g) were harvested and homogenized as was done for thylakoid preparations except the supernatants of both the grinding and washing steps were saved, combined and centrifuged at 80,000 g in an SH-40 rotor for 60 min at 4 C to sediment extrachloroplast membranes. The resulting pellet was termed the extrachloroplast or 80,000 g fraction. Determinations of Chl and fatty acid amounts in the chloroplast fraction were done as described above. Quantities of fatty acids in the extrachloroplast fraction were corrected for the presence of chlor0plast membranes by assuming that Chl in this fraction was associated with chloroplast lipid. This assumes the Chl in the 80000 9 fraction was associated with the same specific lipids as Chl in the chloroplast fraction. The correction was made by subtracting the chloroplast contamination (mol chl in 80000 9 fraction x mol fatty acid per mol chl in cloroplast fraction) from the total for the 80000 9 fraction and adding that amount to the chloroplast fraction. Protein concentrations of the extrachloroplast fraction were determined according to Markwell gt g1 (1981). 30 Pigment1Protein Electrophoresis Pigment-protein electrophoresis was performed by slight modifications of the method of Anderson gt g1 (1978). Chloroplast membranes were isolated as described above except MgCl2 and NaCl were omitted from all buffers. Membranes were solubilized before electrophoresis by incubation for 5 min at 22 C in a volume of solubilization buffer (300 mM Tris-Cl (pH 8.8), 10% (v/v) glycerol, and 5% (w/v) SDS) which gave an $05 : Chl ratio of 1 : 10. In some instances as noted in the results, NaCl was added to the solubilization buffer. The acrylamide to N,N’~methylenebisacrylamide ratio was 30 : 0.8 for both the lower and stacking gels. The concentrations of the reagents and buffers used are given in Table 5. After the gel had set (30 min), 10 - 20 ul of sample was loaded and run at 12 mA for 50-60 min at 4 C giving a migration distance of 4.5 cm for the free pigments. The absorbance of the pigment - protein containing bands in the gels was determined at 600 nm using a Gelman ACD-18 automatic computing densitometer. LHC was isolated by density gradient ultracentrifugation by the method of Burke gt g1 (1978). For rerunning LHCP oligomers and monomers by $05 - PAGE for investigations of protein composition,slices of the bands were removed from the gel with a razor blade and homogenized in a tenbrock. Two gel slices of the oligomeric band were reloaded for SDS-PAGE (Laemmli, 1970). For the monomeric band only one slice was used to prevent overloading. 31 Table 5 Concentration of solutions used in pigment-protein gel and buffer systems Buffer System Buffer Amgggt Titrant Upper Gel 0.56M Tris (pH 6.14) 6.7Bg/100ml H2S04 Lower Gel 1.68M Tris (pH 9.5) 20.349/100ml HCl Upper Resevoir 0.04M Tris (pH 8.64) 4.84g/l Boric Acid ‘ Lower Resevoir 0.42M Tris (pH 9.5) 51.359/l HCl Gel System Lower Gel (8%) StackinqeGel (4%) H20 9.2 ml 3.7 ml Buffer 5.0 ml 0.5 ml 5% SDS 0.4 ml - 0.1 ml 30% acry-0.8% Bis 5.3 ml 0.67 ml APS (IOOmg/lOOml) 100 ul 50 ul Temed 10 ul 2.5 ul 32 Electron Microscopy All stages of preparative electron microscopy were carried out at 1 C. The leaf tissue was fixed in 2% (v/v) glutaraldehyde in sodium cacodylate buffer (pH 7.2) for 1 hour and postfixed in 1% (w/v) osmium tetroxide in the same buffer. After dehydration in a graded ethanol series, the specimens were embedded in Spurr’s Epoxy resin (Spurr, 1969). Ultrathin sections were stained with uranyl acetate and lead citrate and examined in a JEOL 100 CX electon microscope. For obtaining quantitative measurements of the amount of membrane from electron micrographs a map wheel was used. Measurements were made on sections of 20 chloroplasts from both wild type and mutant lines. Egtty Acid Methyl Ester Formation Single Arabidgpsis leaves (5-50 mg fwt) were placed in 13 x 100 mm screw capped glass tubes. Three molar methanolic-HCl (Sulpelco) was diluted to 1 M with reagent grade methanol, one ml was added to each sample and the tubes were sealed with Teflon lined caps and heated to 80 C for 1 h. After cooling, 0.3 to 1.0 ml of hexane and 1 ml of 0.9% NaCl were added and the fatty acid methyl esters (FAMES) were extracted into the hexane phase by vigorous shaking of the tubes. The tubes were centrifuged (1000g for 305) to break any emulsion formed and to completely separate the phases. One or two ul of the hexane layer was injected directly onto the gas chromatograph (GC). GLC analysis was carried out on a Varian 3700 instrument equipped with a flame ionization detector. A six meter column (5% DEGS PS on 100/120 1 Supelcoport; Supelco, Bellefonte, PA.) with 30 ml min' N2 as a carrier gas was used. Operating temperatures for the column, injection port, 33 and detector were 175 C, 210 C and 210 C respectively. Peaks were quantified by a Spectra Physics 2100 Auto lab integrator. Peaks were identified by comparing retention times with known standards. Agglysis OfeLipid Composition Leaves from 4 to 6 g fwt of Arabidopsis plants were pulverized in liquid nitrogen with a mortar and pestle. The pulverized powder was transferred to a 50 ml polypropylene SS-34 centrifuge tube and 30 ml of chloroform-methanol (1:1 v/v) was added. This slurry was homogenized in a Polytron for approximately 30 sec and filtered through Hhatman No.3 MM filter paper under vacuum. The funnel and paper were washed several times with chloroform- methanol giving a final volume of about 50 ml. To the filtered mixture 5 ml of 0.9% NaCl was added. The sample was vigorousrly shaken and separated into phases by centrifugation (10009 for 305). The lower chloroform phase was removed for further analysis. For complete separation of all the lipid species the chloroform extract was applied in 2 ml aliquots to a Biosil A (BIO-RAD) column (0.5 x 3.5 cm) in chloroform. The column was eluted with 10 ml chloroform (neutral fraction), 20 ml acetone (glycolipid fraction) and 10 ml methanol (phospholipid fraction). All three fractions were dried under nitrogen and respended in chloroform : methanol (1:1 v/v) for thin layer chromatography (TLC). For TLC separations, Silica G TLC plates (Baker Si 250 PA, 200 microns) were activated by heating them to 100 C for 1 hour before spotting the sample. For lipid separations two solvent systems were employed. The first was acetone-benzene-water (91:30:8) (Khan and Hilliams, 1977). This system was excellent for separating PG from PE 34 but involved soaking the plates in 0.15M (NH4)ZSO4 for 15 min before activation. One problem with this system was the variabilty in Rf values from plate to plate. The second system was chloroform-acetone-methanol-acetic acid-water (100:40:30:10:4) which was more reproducible but gave poor separation of SL, 006 and PC. However this was not a problem if the lipids were first separated into glycolipid and phospholipid fractions by silicic acid chromatography. After development, the plates were allowed to dry in the fume hood and standards were detected using an stream of air passed through a pasteur pipet containing iodine crystals. Usually one lane of sample was also exposed to the iodine vapours to allow detection of lipids for which standards were not availiable. Lipids were marked and the corresponding area on the plate which had not been exposed to iodine was scraped off into a I3x100 mm screw cap test tube. Three ml of chloroform : methanol (1:1 v/v) were added to each tube and then sealed with a teflon lined cap. The tubes were mixed and allowed to stand for a minimum of 15 min. One ml of 0.9% NaCl was added which upon gentle mixing caused a separation of chloroform from the methanol-water phase with the silica remaining at the interface. The chloroform layer was removed and the procedure was repeated with another ml of chloroform. The combined chloroform phases were dried under nitrogen and 1 ml of methanolic HCl was added to each tube. FAMEs of the lipids were made as described for fresh leaf samples. If quantification of the amount of lipid was desired an internal standard of 17:0 FAME (0.1 ug/ml) was added to the lipid sample before extraction into chloroform. CHAPTER 3 MUTANT ISOLATION Introduction Although the use of mutational analysis to dissect the roles of lipids and fatty acids in microorganisms has led to considerable insight into the function of these macromolecules in membrane assembly and function, such an approach has not been considered widely applicable to higher plants. This generally reflects the inability to adapt microbial methodologies of mutant isolation to multicellular organisms. Bacterial and yeast mutants with altered lipid composition are usually isolated by two methods. One involves screening for mutants auxotrophic for lipid precursors by rescue with nutritional supplementation (Keith gt g1, 1969; Atkinson gt g1, 1980). However this method depends upon the ability to replica plate cells and the organisms ability to takes up the precursors readily. Neither of these criteria have been achieved in higher plants. The alternative to supplementation is to screen a large population of individuals for temperature sensitive (tg) conditional mutants. However a t; mutation in any particular gene is generally much less 35 36 frequent than a null mutation. Therefore, some form of mutant enrichment, usually involving conditions which kill actively growing cells at nonpermissive temperatures, is required (Raetz, 1978). This in turn allows enough individuals to be screened to make the approach feasible. However adaption of such negative selection screens has met with limited success in plant tissue culture (Horsch and King, 1986) and is mechanistically difficult to do at the whole plant level. A alternative approach for the isolation of well defined biochemical mutants at the whole plant stage has involved a detailed characterization of the regulation of the pathway of interest under a variety of experimental conditions. If conditions can be defined which appear to change the regulation of the pathway, mutants with altered growth response under these conditions might be defective in the characterized pathway. Such an approach has been used successfully to isolate mutants defective in photorespiratory metabolism (Somerville and Ogren, 1982). In the case of lipid biosynthesis, however, the results of biochemical and physiological studies on the regulation of these pathways have been conflicting (Kuiper, 1986). Early work on chilling sensitive and chilling resistant species of plants led to the proposal that the ratio of unsaturated to saturated fatty acids was instrumental in maintaining membrane integrity at various temperatures (Lyons and Raison, 1970). Although subsequent analysis of these ratios in a larger collection of species did not support this theory most studies did show that for any particular plant a change in unsaturation does occur upon shifting to different temperatures. For example, in Brgssjgg ngggg upon transfer to 5 C an increase in 18:3 was detected in both roots and leaves (Smolenska and Kuiper, 1977). This increase is 37 most evident in phosphatidylcholine and phosphatidylethanolamine in the leaves and neutral lipids in the roots. Similar biochemical changes in 18:3 levels in Arabidopsis, which is also a member of the Brassjgaggae, led to the suggestion that a class of mutants sensitive to chilling temperatures might also be disrupted in some aspect of fatty acid unsaturation. Based on the theory that the degree of lipid unsaturation affects the resistance of a plant to chilling, two chilling sensitive mutants of Arabidopsis were isolated; however, neither showed alterations in fatty acid metabolism (McCourt, 1983). The low number of mutants isolated did not allow evaluation of this method as a potential approach for the isolation of mutants with altered lipid membrane composition. Because of the failure of this method to result in the isolation of plant lipid mutants, it was decided that a direct screen for changes in fatty acid composition under nonselective conditions should be attempted. The rationale for such a screen was as follows. All successful methods of lipid mutant isolation in microbial systems assume such defects are lethal to the organism. Hence, almost all of these mutants have a detectable phenotype. However, E. ggli mutants completely defective in the synthesis of cyclopropane fatty acids show no impairment of growth under a variety of environmental conditions (Taylor and Cronan, 1976; Grogan and Cronan, 1986) suggesting changes can be made to the membrane lipid composition which are not essential for the life cycle of the organism. Therefore, an assay based on fatty acid methyl ester formation and subsequent gas chromatographic analysis was devised which permits rapid and quantitative analysis of fatty acids containing 16 or more carbons from a single leaf sample (see 38 Materials and Methods). An example of the results of a typical analysis of a leaf from the wild type is presented in Figure 3. Results Mutant Isolation This single leaf assay was applied to a population of M2 Arabidoosis plants grown under normal laboratory growth conditions. From approximately 2000 plants examined in this way, 89 were retained for future analysis due to anomalies in fatty acid composition. Four to ten individual M3 progeny from each of the 89 lines were analysed for inheritance of the altered fatty acid composition. In those cases where all the progeny exhibited the mutant phenotype the most vigorous plant was saved and advanced by single seed descent. In instances where some of the plants appeared to be wild type more plants were analysed to see if the population was segregating for the original phenotype. This allowed the reselection of the mutation in the homozygous state. By repeating this process for several generations eight of the orginal 89 lines were found to have stably inherited changes in fatty acid composition and were given strain numbers beginning with 081 (Table 2). These 8 lines were grouped according to phenotype based on whole leaf fatty acid composition. Although the exact reason for the low recovery of mutant phenotypes in the M3 generation is not known, in most cases the 81 lines which did not show a reproducible phenotype had very subtle changes in fatty acid composition and probably reflect physiological variation within a growing population. 39 9 on 03 c o c. a) on ... 0: :2 L o ...-o 0 N on a '1" . P 01 D g 3 .. NP 0 i ,3 1‘2 Elution time (min) Figure 3. A typical gas chromatography tracing of fatty acids extracted from a single wild type leaf 40 Group I Mutants Two mutants (0827 and 0860) were isolated with nondetectable levels of the fatty acid 3-ttggg-hexadecenoic acid (ttggg-Clfizl) and corresponding increases in the amounts of 16:0. Normally less than 5% of the wild-type levels can be detected by this method. Otherwise the mutants were indistinguishable from wild-type in both fatty acid composition (Table 6) and viablity. The genetic basis of the phenotype in the line 0860, was determined by measuring the fatty acid composition of F1 and F2 progeny from crosses between 0860 and wild-type. The F1 progeny had approximately 50% as much tgggg-C16:l as wild-type (0.89 1 0.25 mol % in the F1 progeny versus 1.74 i 0.21 mol % in the wild-type, n-lO plants) suggesting a simply inherited codominant nuclear mutation. The frequency of the homozygous mutant phenotype in an F2 generation was determined by gas chromatographic analysis of the fatty acid composition of leaves from 57 F2 plants. Of these 13 completely lacked ttggg-C16:l. This is a very good fit to the 3:1 hypothesis (X2 (1)-0.14; P>0.9) indicating the deficiency is due to a single nuclear mutation at a locus that has been designated fggA (fatty acid desaturation gene A). The thA mutation was mapped to chromosome 4 by F2 mapping (see Appendix B) from a cross of line MKl by JB60. Complementation analysis of 0860 with 0827 was carried out using a glabirous derivative of line 0860 LfadA, ql-l) as the female parent. The recessive leaf marker assured selfing had not occurred in the cross. The F1 progeny showed no detectable ttggg-Clszl indicating these two mutations are allelic. 41 Table 6 Percent of total fatty acids in Group I mutants (lacking trans-C16,1). Each value represents the average of ten plants. Fatty Acid Composition (mol %1 Fatty Acid HT 0827 0860 16:0 15.8 17.0 18.3 t16:1 1.8 0.0 0.0 C16:1 0.2 0.3 0.2 16:2 0.7 0.8 0.8 16:3 12.6 12.5 11.6 18:0 1.0 0.9 1.0 18:1 2.7 2.8 . 2. 18:2 18.9 19.3 18.9 18:3 47.5 47.5 47.2 42 Group II Mutants One mutant (line 081) was isolated with reduced levels of both 18:3 and 16:3 fatty acids and had corresponding increases in 18:2 and 16:2 (Table 7). On the basis of evidence presented later, this line is considered to carry a mutation at a single nuclear locus designated fggfl. Upon subsequent analysis of this mutant it was observed that the fatty acid composition, although always different from wild-type, showed variability between plants grown within the same chamber. Subsequent experiments in which light intensity, light quality and temperature were varied, were carried out to determine the cause of this variation. No obvious difference could be detected in trienoic acid composition of plants grown for 3 weeks at 25 C under varying light intensities (Figure 4). By contrast, a marked effect was seen in the levels of both 18:3 and 16:3 at temperatures above about 23 C in the 081 line, suggesting the ngQ mutation confers temperature sensitivity upon dienoic acid desaturation (Table 7, Figure 5). In order to determine the genetic basis for the reduced trienoic acid content, 317 F2 plants resulting from a cross of wild-type to 081 were grown at 23 C and analysed for fatty acid composition. Of these, 244 F2 progeny resembled wild-type in fatty acid content and 73 had a fatty acid composition indistinguishable from the mutant parent grown at the same temperature. This segregation pattern is a good fit (X2(1)-0.66; P>0.9) to the 3:1 hypothesis. The cosegregation of reduced 16:3 and 18:3 content strongly sugests both changes are caused by a single nuclear mutation. The fggp mutation was mapped to chromosome 3 by F2 mapping (see Appendix B) from a cross of line MKl by 081. 43 Table 7 Percent of total fatty acids in Group II mutants (deficient in 16:3 and 18:3) grown at 18 C and 27 C for three weeks. Each value represent average of ten plants. Fatty Acid Composition (mol %) H (I) ("1 N \l ("a Fatty Acid HT 081 HT 081 16:0 13.3 13.4 15.8 17.4 t16zl 1.4 1.1 2.2 3.1 c16:1 0.3 0.3 0.2 2.7 16:2 0.2 3.4 0.3 6.0 16:3 13.8 6.9 10.7 2.0 18:0 0.4 0.4 1.3 1.4 18:1 1.1 2.4 3.7 9.2 18:2 12.7 18.1 17.0 39.3 18:3 55.6 52.0 48.7 18.6 44 mwonEQmHa 83 5&0. 0:3 Ba 83 an £32: 00 8253 5 TIDE: Em Allv mama mo :ofiuhomoum 05 :0 Ban 8:053 no nomuum 05. .v 90de . Amlm Trams; 33:25 Em: 00¢. 00” CON OO _. O _ _ _ _ 91114111291 u; 8:81 Jo 5:91 % 45 I 1 A 60 ' 4 8? 99 + Wild-type ‘7’. g 40 : ‘ E Mutant O \ (U .2 \\ 8 20 '- ' .2 ; 1: ' i 32 3 , i O 1 l 1 l P J 14 18 22 26 30 - Temp (°C) Figure 5. The effect of temperature on the proportion of trienoic fatty acids in leaves of mutant 081 (O) and wild type (D) Arabidopsis 46 When F1 plants from a cross of 081 by wild-type were grown at 23 C and the fatty acid composition of these plants was analysed, no distinguishable difference from wild-type could be detected. However F1 plants grown at 28 C had intermediate levels of 18:3 and 16:3 compared to wild-type (Table 8). Therefore, as with the thA mutant, the level of 18:3 and 16:3 is probably regulated by the amount of active enzyme present. Group III Mutants Four mutants were isolated which showed similar whole leaf fatty acid alterations (Table 9). All had reduced amounts of 16:3 and no 16:2 or g_s~16:l. These changes appeared to be compensated by increases in 18:1, 18:2, and 18:3 fatty acids. Although no further work was carried out on these mutants in this study, L. Kunst has shown that 0825 is also a simple nuclear mutation at a locus provisionally designated pgp. Double Mutants Construction of double mutants for the testing of epistatic relationships between various mutations can help provide detailed information about the sequence of steps affected by these mutations in a biosynthetic pathway. For example, if one mutation occurs earlier in a linear pathway than another it should be epistatic to the second mutation. If the mutations fail to show epistasis then they probably do not affect the same pathway or there are alternate pathways. If double mutants are obtained which have a fatty acid composition different from both parents, although this makes the interpretation of the 47 Table 8 Percent of total fatty acids in wild type, 081, and F1 progeny from a 081 x wt cross grown at 28 C for three weeks. Each value represents the average of ten plants. Fatty Acid Composition (mol %1 Fatty Acid HT 081 F1 (WT x 0B1) 16:0 17.4 1 0.2 16 8 i 0.1 17.6 i 0.3 t16:1 2.0 i 0 2 1 9 i 0.3 1 8 i 0.3 c16:1 0.5 i 0.1 0 7 i 0.1 0.9 i 0.2 16:2 1.0 i 0.1 7.6 i 1.2 2.9 i 0.9 16:3 10.0 i 0 8 2.2 i 1 4 7.7 i 1.3 18:0 1.0 i 0 1 1 4 i 0 1 1 4 i 0.1 18:1 4.4 i 0 1 5 2 i 0.5 4 6 i 0 1 18:2 22.3 i 0 3 39.8 i 2.9 28 3 i 2.6 18:3 39.1 i 0 4 22.3 i 2.8 33.7 i 2.7 48 Table 9 Percent of total fatty acids in Group III mutants (deficient in 16:3 biosynthesis. Each value represents the average of ten plants. Fatty,Acid Composition 1mol %1 Fatty Acid HT 083 0819 0825 0828 16:0 18.0 10.5 13.8 14.6 12.8 t16:1 1.7 1.4 1.0 1.1 0.9 c16:l 0.3 0.0 0.0 0.0 0.0 16:2 0.8 0.0 0.0 0.0 0.0 16:3 12.9 0.0 4.9 0.0 2.2 18:0 0.9 0.6 0.2 0.6 0.8 18:1 2.0 6.7 4.8 . 8.6 6.8 18:2 11.7 17.9 19.1 19.0 19.7 18:3 51.6 62.3 55.2 55.4 57.4 49 biosynthetic pathway more difficult it does provide completely new membrane compositions upon which to carry out physiological studies. Because group-II (tgdfl) and group-III (9gp) fatty acid phenotypes both appear to have altered 16:3 levels, it was of interest to determine their epistatic relationship. Of 49 F2 plants from a 081 (:ggp) by 0825 (9gp) cross, one plant line showed a fatty acid composition similar to both parents. This phenotype was inherited stably for three generations, and the line was given the strain designation LIP1. As with 0825, the LIP1 line had no detectable 16:3, 16:2 or cis 16:1. However LIP1 also had reduced amounts of 18:3 and increased 18:2 in plants grown at 27 C characteristic of the 081 (Table 10). The lack of 16 carbon unsaturated fatty acids was independent of temperature but the 18:3 levels were not (Table 11). The LIP1 line is of added interest because although it appears that the pgp mutation is epistatic to the thD mutation at the level of 16:3 synthesis the opposite is true for 18:3 synthesis. This suggests some features of the 16:3 and 18:3 biosynthetic pathway(s) are shared while others are not. Of the F2 progeny that were screened from the 081 by 0825 cross, the ratio of classes were 27 HT : 5 0825 : 16 081 : l LIP1. This ratio is close to the 9:3:3:l ratio expected for independent assortment (X2(3)-5.59, P>0.9). This result in combination with normal 3:1 ratios for the single loci (FadD x2(1)-2.45, pgg x2(1)-4.25, P>0.9) suggest these two loci are unlinked. The F2 progeny from a 0860 (group I) by 081 cross were also screened for double mutants. Eight of 120 plants lacked both trggs-Cl6:l and had reduced levels of 16:3 and 18:3 (Table 10). All lines were tested for the inheritance of the phenotype and one strain was kept for further 50 Table 10 Percent of total fatty acids in double mutants LIP1 (081 x 0825) and LIP2 (0860 x 081) grown at 28 C. Each value represents the average of four plants. Fatty Acid Composition (mol %) Fatty Acid HT 081 0825 LIP1 LIP2 16:0 14.7 13.2 14.6 10.5 16.2 t16:1 2.7 2.7 1.1 2.0 -- c16:1 0.1 3.1 -- -- 0.2 16:2 1 2 10.5 -- -- 9 0 16:3 11.5 2.0 -- -- 1.4 18:0 1.1 1.4 0.6 ' 0.8 1.0 18:1 2.3 9.2 8.6 6.7 6.5 18:2 14.2 37.7 19.0 65.7 37.9 18:3 52.1 18.9 55.4 15.9 25.0 51 Table 11 Percent of total fatty acids in double mutants LIP1 (081 x 0825) and LIP2 (JB60 x 081) grown at 19 C. Each value represents the average of four plants. Fatty,Acid Composition (mol %) Fatty Acid HT 081 0825 LIP1 LIP2 16:0 15.0 13.4 14.6 11.7 17.8 t16:1 1.7 1.1 1.1 1.2 -- c16:1 0.3 .3 -- -- 0.4 16:2 0.8 3 4 -- -- 5.0 16:3 12.9 6.7 -- -- 6.6 18:0 0.9 0.4 0.6 ' 0.6 0.4 18:1 2.0 2 4 8.6 9 0 2.7 18:2 11.7 18.1 19.0 29.6 21.6 18:3 53.6 52.0 55.4 45.5 48.7 52 analysis (LIP2). Again the 18:3 and 16:3 fatty acid compositions were affected by temperature as is expected of the 081 phenotype but the ttggs-Clazl was absent at both temperatures (Table 11). The ratio of classes from the screen were 73 HT : 19 081 : 20 0860 : 8 LIP2. This excellent fit to the 9 3 3:1 hypothesis (x2(3)-1.3o, p>0.9) suggests that the two mutations are unlinked. This had also been established by F2 mapping of both thA and ngD to known chromosomal markers (Appendix B). The lack of any epistasis between ngA and ngQ was expected since trans-C15.1 is only synthesized on PG and is not further metabolized (Dubacq and Tremolieres, 1983). CHAPTER 4 PHYSIOLOGICAL AND BIOCHEMICAL STUDIES OF A MUTANT LACKING TRANS HEXADECENOIC ACID Introduction Aside from the family Orchidaceae (Roughan, 1986) the chloroplast membranes of all photosynthetic eukaryotes contain the unusual fatty acyl group 3'LEADS'C15.1 which is always found esterified to the second position of phosphatidyl glycerol (Dubacq and Tremolieres, 1983). The fatty acid is atypical because of the tgggs configuration, and because of the position of the double bond near the carboxyl rather than the methyl end of the fatty acid. A specific role for the acyl group in photosynthesis has frequently been proposed because iriflfi'c16zl'PG occurs only in chloroplast membranes (Dubacq and Tremolieres, 1983), and is present in relatively low amounts in etioplasts but accumulates upon light-induced chloroplast development in parallel with the accumulation of the LHCP and the development of appressed membranes (Dubacq and Tremolieres, 1983; Galey gt g1, 1980; Mackender, 1979). Also, removal of trans-C from PG by 16:1 phospholipase-AZ treatment of isolated thylakoids was reported to alter the efficiency of light capture and to change the kinetics of 53 54 fluorescence induction (Duval gt g1, 1979). However, a specific role in photosynthesis has not been demonstrated (reviewed in Dubacq and Tremolieres, 1983). Recently, evidence pertaining to a possible role for tgggs—C16:l was obtained from experiments in which the lipid content of isolated Chl-protein complexes was characterized. When thylakoid Chl-protein complexes are solubilized in low amounts of SDS and electrophoresed on polyacryamide gels which also contain low concentrations of $05, the Chl-protein complexes separate into a characteristic pattern of about six major Chl-containing bands (Anderson gt g1, 1978). Hhen these bands were extracted from the acrylamide gel and the lipid composition of each band measured, it was found that the LHCP band which is believed to correspond to an oligomeric form of LHCP, was significantly enriched with ttg_s-C16:l (Tremolieres gt g1, 1981). The possible importance of the lipid was also suggested by experiments in which treatment of thylakoids with phospholipase-AZ before solubilization and electr0phoresis caused the disappearance of the LHCP oligomer, which has been proposed to be the native form of the complex 15 yiyg (Kuhlbrandt, 1984). Although it may be only coincidental, there is approximately enough trans-615:1 in the chloroplast membranes to satisfy a stoichiometry of one molecule per LHCP oligomer (Dubacq and Tremolieres, 1983). A role in LHCP oligomer formation or stabilization is also suggested by recent experiments showing that the rate of reconstitution of LHCP oligomer in liposomes is stimulated by the presence of LEQDS-C15.1 (Remy gt al, 1984). The mutant line of Argbidopsis thaliana (L.), 0860 which specifically lacks trans-C16,1 has a compensating increase in palmitic 55 acid (16:0). The mutant is, therefore, believed to lack a specific desaturase which converts palmitic acid at the sn-2 position of PG to ttggs-Clezl. As mentioned in a previous section the mutant, which has no obvious phenotype, was isolated by analyzing the fatty acid composition of several thousand randomly selected individuals from a mutagenized population. In a preliminary analysis of thylakoid ultrastructure and function we were unable to establish a difference between the mutant and the wild-type. Here, we describe the results of experiments designed to test the role(s) of ttggt-C16:l in formation of LHCP oligomer and in the functional association of LHCP and the photochemical reaction centers. Although fluorescence measurements suggest normal LHCP function, the LHCP oligomer appears less stable to dissociation by SDS in the mutant. A similar effect on the CPla complex suggests that lxéfl§'c16:1 also stabilizes the presumed oligomeric form of the PSI Chl-protein complex. Results ngsynthgsis gf (rans-C16 1 The observation that 16:0 levels were higher in the fadA mutant than in wild-type plants suggests this fatty acid is a precursor of the trans isomer (Table 6). Support for this idea can also be inferred from radiotracer studies of trans-616,1 synthesis in Chlorellg yulqaris (Nichols gt g1, 1964). Upon adding [14C]palmitate to light grown cultures, radioactive trans-C16,1 was exclusively found esterified to PG. Moreover the observation that [14C]trans-C16_l when added to 56 cultures is randomly distributed among the lipids implies PG is the substrate for tgggs unsaturation. Thus, it was of interest to compare the fatty acid composition of PG from the mutant and the wild-type. The results of these measurements (Figure 6), which substantially increased the limit of detection, confirmed that the mutant completely lacked ttggs-Clezl. The decrease in ttggs-C16:1 was compensated by a proportional increase in the 16:0 content of PG. These observations suggest that the mutant is deficient in activity for a proposed desaturase which specifically converts 16:0 at position two of PG to Liéflé-C15.l (Galey gt g1, 1980). If this interpretation is correct, the presence of ttggs-Clszl in the heterozygote would suggest that the amount of this fatty acid is regulated directly by the amount of enzyme activity rather than by some mechanism which senses and responds to the absolute concentration of this fatty acid in the membrane. Since this desaturase activity has not, as yet, been demonstrated by in ytyg assay of cellular extracts or by tracer studies with intact chloroplasts (Guillot-Salomon gt g1, 1982) the precise enzymatic lesion in the mutant cannot be evaluated by these means at present. The absence of LEADé-C16:1 in a barley mutant deficient in chloroplast ribosomes (Haworth gt g1. 1983), and the inhibition of tgggg-C16:l synthesis by chloroplast protein synthesis inhibitors (Henry gt g1, 1983) have been interpreted as possible evidence that a gene for ttggs-C16:1 synthesis is plastome-encoded. The Mendelian segregation of the {egg mutation does not support this concept. However, our results do not exclude the possibility that one or more proteins encoded by the chloroplast genome are also required for trans-C16:1 synthesis. 57 Detector Response 6 . 1. 111 0 4 8 12 0 4 8 12 Elution time (min) Figure 6. Fatty acid composition of PG from wild type (a) and mutant JB60 (b) Arabidogis. The position of trans-C16:1 in the chromatogram from the wild-type is indicated by the arrow. 58 Cellular and Biochemical Studies Because ttggs-Clszl-PG is not present in etiolated tissue but accumulates during light-induced chloroplast devel0pment (MacKender, 1979) and occurs only in chloroplast membranes, it has been inferred that this lipid has a specific role associated with the light reactions of photosynthesis. Recently, attention has been focused on an apparent association of this lipid with the light harvesting chlorophyll a/b protein complex (LHCP) (Lam gt g1, 1984; Lynch and Thompson, 1984), which also accumulates in thylakoid membranes during light-induced chloroplast development and is thought to have an important role in the formation of the appressed membranes of the grana (Mackender, 1979). In order to critically evaluate the possibility that ttggs-C15:1 is also involved in this process, thylakoid ultrastructure of mutant and wild-type chloroplasts was analyzed by electron microscopy of thin sections of whole leaves. The micrographs (Figure 7) showed no obvious differences in either the size or extent of granal development or any other major ultrastructural feature of the chloroplast, and are considered compelling evidence against the obligatory involvement of ttggs-C16:l in the development of thylakoid structure. Recent models for the native structure of LHCP based on Fourier analysis of high resolution electron micrographs of two-dimensional LHCP crystals have indicated that LHCP is a trimer of three structurally equivalent subunits (Kuhlbrandt, 1984). It is believed that this oligomeric structure corresponds to a high molecular weight form of LHCP observed following electrophoresis of thylakoid proteins in acrylamide gels under conditions in which thylakoid proteins are solubilized with low amounts of SDS so that Chl-protein associations 59 Figure 7. Transmission electron micrographs of uHFathin (80 to 100 nm) sections of whole leaves from wild-tupe (A) and mutant 0860 (8) Arabidopsis at x22,500 magnification. 60 remain intact (Lam gt g1, 1984; Lynch and Thompson, 1984). Furthermore, previous studies have shown that ttggs-C16zl-PG comigrates with the LHCP oligomer in gels run under these conditions (Anderson gt g1, 1978). It was, therefore, of interest to examine the effect of the thA mutation on the pattern of Chl-protein complexes resolved by this method. Separation of the Chl-protein complexes from wild-type Arabidopsis extracts revealed five major Chl-containing bands (Figure 8) CP1a, CPI, LHCP], LHCP3 and free Chl. Comparison of the electrophoretic separation patterns of the Chl-protein complexes from the wild-type and the mutant under standard conditions revealed that the mutant lacked the two Chl-containing bands designated CPla and LHCP1 (Figure 8). These bands are believed to represent the oligomeric forms of CPI (the P700-Chl a-protein complex) and LHCP3 (the presumed LHCP monomer), respectively (Anderson gt g1, 1978). The absence of the oligomeric form of LHCP in the mutant mimics similar results obtained following removal of the acyl group at position two of PG by phospholipase-AZ treatment of thylakoids (Remy gt g1, 1984). The reduction of the amount of CPla in the extracts of the mutant was unexpected since ttggs-Clezl-PG has not previously been reported to be a component of this Chl-protein complex. However, a Chl a/b protein complex associated with the PSI complex has recently been reported (Haworth gt g1, 1983; Kyle gt g1, 1984; Lam gt g1, 1984). The results presented here raise the possibility that this complex has trggg-Clszl-PG specifically associated with it as a boundary lipid. Alternatively, the presence of ttggs-Clszl-PG in the membrane may exert a nonspecific effect on both LHCP and CPla stability. 61 Figure 8. Pigment protein complexes of mutant JB60 (lane 1) and wild type (lane 2) thylakoids. FP denotes free pigment 62 Studies of cation effects on Chl-protein complexes have shown that removal of cations from solubilization buffers increased the proportion of Chl found in CPla and LHCP following electrophoresis in SDS polyacryiamide gels (Argyroudi-Akoyunoglou, 1981; Argyroudi-Akoyunoglou and Thomou, 1981) The addition of either MgCl2 or NaCl converted an increased proportion of the oligomers into their respective monomers. Therefore, we examined the effect of cation concentration on the proportion of Chl associated with LHCP to determine if conditions could be found which would stabilize the oligomers of the mutant. The solubilization of thylakoid membranes in solutions of SDS containing very low concentrations of NaCl revealed that the thylakoids of the jggg mutant contained normal levels of the LHCP oligomer (Figure 9). As the NaCl concentration was increased from 0 to 100 mM, the amount of LHCP in both mutant and wild-type decreased from a maximum of about 7% to zero (Figure 9). However the membranes from the mutant were much more sensitive to salt-induced dissociation of LHCP oligomer than those of the wild-type. The concentration of NaCl which gave 50% reduction in LHCP concentration was about 13 mM in the mutant as compared to about 37 mM in the wild-type (Figure 9). Thus, it appears that trggt-C16:1 is not required for LHCP formation but in some way stabilizes the oligomer so that it is less susceptible to SDS-mediated dissociation. Since it was now possible to visualize the four pigment protein complexes under low salt conditions in both mutant and wild-type, the bands were removed and analyzed by absorption spectroscopy (Figure 10). The absorption spectra of the gel slices derived from extracts of the two genotypes were very similar indicating that lack of trans-C16:1 in 63 1 v 010 ?o 30 40 so 100 Amount of Oligomeric LHCP (% of total Chl) o. 3 7/ NaCl Concentration (mM) Figure 9. Effects of NaCl concentration on the proportion of Chl in LHCP oligomer in mutant 0860 (O) and wild type (I) Arabidopsis 64 the complex does not cause a major change in the orientation of the chlorophyll on the proteins under these conditions. CPI and CPIa both showed a red maximum at 675 nm and a blue maximum at 438 nm which is characteristic of a P700 chlorophyll a-protein complex (Thornber and Highkins, 1974). LHCP] and LHCP3 also showed similar spectra as would be expected for related complexes. The enhanced absorption at 672 nm and 472 nm is due to high levels of chl b relative to chl a which is commonly seen in chlorophyll a/b binding complexes (Anderson gt g1, 1978). If the oligomeric bands (LHCPI) were rerun under SDS denaturing gel conditions (PAGE) no difference in protein composition of this complex could be detected (Figure 11, lane 6 wt, lane 3 JB60). To further characterize the protein composition of thA thylakoids and LHCP, SDS PAGE was carried out on thylakoids and LHCP purified by density gradient centrifugation (Figure 11). The thylakoid protein composition of 0860 (lane 4) and wild-type (lane 5) were indistinguishable. The density gradient purified LHCP monomers from wild-type (lane 7) and 0860 (lane 2) were also identical. The oligomeric LHCP from mutant could not be purified under these conditions but wild-type oligomers appear to have the same protein composition as monomers (lane 1). The major polypeptide for the Arabidopsis LHCP appears to have a molecular weight of 26 kd which is close to that published for the major LHCP polypeptides from other species (Darr, 1985). In summary, the lack of traps-615:1. although causing instability in the oligomeric form of LHCP under certain gel conditions, does not seem to effect polypeptide or chlorophyll composition of this complex under less harsh isolation conditions. 65 mu<¢-mom an um>Fommc mmxm_qsou =_muocq-_Pacqoco_zu TI; 33 3.3 new 71v. 8%. 23:... so «.50QO 53335 .3 3:3... 255182.“: m><>> CON. 0mm 00m 0mm com Om»N 0mm COO 0mm 00m Omv 00V _ _ _ _ _ _ v.0. md N._. 0;. 66 +68 ltd «4511:! Figure 11. Protein composition of mutant JB60 (lane 4) and wild type (lane 5) thylakoids and purified LHCPI and LHCP3 from 0860 and wild type separated by SDS-PAGE. Lane 1 is density gradient purified wild type LHCPI. Lane 2 and 7 are density gradient purified 0860 and wild type LHCP3 monomers. Lane 3 and 6 are gel purified wild type and JB60 LHCPI. 67 Photosynthetic Studies Since the function of LHCP is to enhance the capture of light energy by PSII, we examined the effect of the thA mutation on the irradiance response curve for the light reactions catalyzed by isolated thylakoids (Figure 12). A major role for ttggg-C16:1 in LHCP function would be expected to result in a reduced rate of electron transport in the mutant lines at low irradiance. However, no difference was observed between the photosynthetic activities of mutant and wild-type. FluorescencegSDectra Chl fluorescence emission spectra are sensitive indicators of the efficiency of energy distribution between the Chl-containing components of the photosynthetic membranes. Preferential excitation of Chl b using 480 nm light results in most of the energy being distributed between the two photosystems. Detachment of LHCP from one or both of the reaction centers results in relatively increased fluorescence emission from LHCP and associated Chl-protein complexes. Thus, it is possible to monitor the relative extent of interactions between LHCP and the reaction centers in thylakoid membranes by comparing the fluorescence emissions at 685, 695, and 735 nm which have been attributed to LHCP, PSII, and PSI, respectively (Butler, 1978; Papageorgiou, 1975). To test for the presence of an in yiyg difference between LHCP function in mutant and wild-type we first compared the spectrum of Chl fluorescence from whole leaves at 77K (Figure 13). An alteration in the efficiency of exciton transfer from LHCP to the photosystems in the mutant would have been expected to result in a change in the ratio of 68 69 LHCP fluorescence (685 nm) relative to the other emission maxima. However, no significant difference was apparent between the mutant and the wild-type by this criterion (Figure 13). In an attempt to relate the effects of cations on LHCP1 stability to a functional property of LHCP, we examined the effect of cations on low temperature (77K) fluorescence emission spectra of isolated thylakoid membranes. Assuming that the LHCP oligomer is the native form in gitg (Kuhlbrandt, 1984), it might be expected that if cations induced dissociation of the LHCP oligomer in intact membranes, this would be reflected in less efficient transfer of excitons from LHCP to the reaction centers. In this case one would expect more fluorescence at 685 nm and less at 734 nm. It should be noted, however, that a cation-induced change in the ratio of PSI to PSII fluorescence has previously been attributed to changes in the spatial organization of the Chl-proteins rather than to changes in the quaternary structure of individual proteins (Staehelin and Arntzen, 1983). The Chl fluorescence spectrum of chlorOplast membranes isolated in the absence of cations is presented in Figure 14. The spectrum, which is qualitatively very similar to that obtained with whole leaves, was not significantly different with respect to the wavelength of the emission maxima or the relative distribution of fluorescence between the two photosystems in mutant and wild-type. Addition of 5 mM MgCl2 to the thylakoids caused an increase in the ratio of PSII to PSI fluorescence (Figure 14) as expected from previous studies concerning the effect of divalent cations on membrane appression and fluorescence (Butler, 1978; Staehelin and Arntzen, 1983). However, the ratio of PSI 7O Relative Fluorescence Intensity 650 700 V V 250 . I 500 Wavelength (nm) Figure 13. Chlorophyll fluorescence spectra of whole leaves from mutant (---) and wild type (--) Arabidopsis at 77K 71 [N Relatlve Fluorescence Intensity 650 700 750 800650 700 750 - A 500 Wavelength (nm) Figure 14. Chlorophyll fluorescence spectra of chloroplasts from mutant (---) and wild type (-—-) Arabidopsis in the presence (A) and absence (8) of MgCl (SmM). The same preparation of chloroplasts were usgd for A and B. 72 to PSII fluorescence in the mutant was not significantly different from that of the wild-type. A more extensive analysis of cation effects was performed by examining the effects of a range of NaCl concentrations on the photosynthetic lamellae. Incubation of thylakoids in increasing concentrations of NaCl in the range from 0 to 100 mM caused an essentially linear increase in the ratio of fluorescence at 685 nm to that at 734 nm (Figure 15). As noted above, a cation stimulated increase in PSII activity at the expense of PSI has previously been observed and is attributed to a decrease in energy spillover from PSII to PSI due to cation-induced changes in the spatial separation of PSI and PSII, which decreases the probability of exciton migration from PSII to P51. In contrast to the results from the SDS-acrylamide gel experiments, no differential cation effect on the transfer of energy from LHCP to PSI in the mutant versus the wild-type was detected in the range of O to 100 mM NaCl (Figure 15). Energy transfer between LHCP and PSII and between the photosystems does not appear to be impaired as one might expect if LHCP structure was substantially altered (Staehelin and Arntzen, 1983) Fluorescence Induction In previous studies of the role of_ttggs-C16:1-PG, thylakoid membranes were depleted of LLAQS-C15:l by treatment with phospholipase-AZ (Duval gt g1, 1979; Rawyler and Siegenthaler, 1981; Remy gt g1, 1984). Membranes treated in this way exhibited altered fluorescence induction kinetics which were interpreted as a reduction in the efficiency of light capture and the rate of plastoquinone 73 1.10-0 0.90-0 0.70-4 Ratio of Fluorescence Intensity (1585/1734) o 20 4o 50 8'0 100 NaCl (mM) ' Figure 15. Ratio of fluorescence intensity from low temperature (77K) em1531on spectra of wild type (0) and mutant JB60 (0) thylak01ds. Each point represents the mean t SE (n=3). 74 reduction (Duval gt g1, 1979). To reexamine the relevance of these observations to understanding the role of ttggs-C16:1-PG, the function of PSII and LHCP in wild-type and the mutant 0860 lacking ttggg-C16:l-PG was compared by examining the kinetics of induction of room temperature fluorescence at 700 nm. Room temperature fluorescence primarily represents fluorescence emitted from PSII (Patterson and Arntzen, 1982). When electron transport is blocked with DCMU, the rise of the variable fluorescence (Fv) is a measure of the time required to close (cause a turnover of) all PSII reaction centers (thereby reaching maximal fluorescence, Fm). In this respect, the rise time of Fv is a relative measure of both the number of Chl active in transferring excitation energy to PSII reaction centers and of the efficiency of transfer. The minimum level of fluorescence, F0, is due to emission from the antenna Chl of PSII which occurs before the excitation energy is trapped by the reaction centers (Butler, 1978). F0 and the proportion of Chl active in photochemistry (Fv/Fo) appeared to be identical in the mutant and the wild-type in the absence of MgCl2 (Table 12). Addition of 5 mM MgCl2 to the membranes resulted in a dramatic increase in Fm due to Mg-induced changes in the spatial organization of the membranes (Staehelin and Arntzen, 1983) and a resulting decrease in the amount of spillover of excitation energy to PS1. The effect of cations on fluorescence characteristics of mutant and wild-type membranes was quantitatively and qualitatively indistinguishable under these conditions. These observations, in conjunction with previous studies showing that the mutant and the wild-type have indistinguishable rates of electron transport, suggest that the two genotypes have indistinguishable PSII photochemistry 75 Table 12 Room temperature fluorescence induction parameters of isolated thylakoids from mutant 0860 and wild type Arabidopsis in the presence and absence of 5mM MgClz. Line F F F /F 0 m V 0 Wild Type (+Mg) 1164 i 114 4746 i 97 3.01 i 0.14 0860 (+Mg) 1122 i 35 4735 i 225 3.22 i 0.07 Wild Type (-Mg) 1027 i 35 2333 i 71 1.27 i 0.02 0860 (~Mg) 1010 i 13 2302 i 89 1.28 i 0.07 76 efficiency. This implies that the P511 antenna must be structurally similar. It is, therefore, apparent that the interpretation of previous studies employing lipase modification of membrane structure was confounded by the lack of specificity of the experimental approach (Duval gt g1, 1979; Rawler and Siegenthaler, 1981). Effects of High Tempergture on Fluorescence Several authors have interpreted increases in Chl fluorescence which occur upon heating of leaves as an indicator of temperature-induced changes of photosynthetic membrane stability (Armond gt g1, 1978; Lynch and Thompson, 1984; Schrieber and Berry, 1977, see Appendix A). The heat-induced rise in F0 which has been interpreted in terms of a breakdown in energy transfer from LHCP antenna pigments to PSII centers and related inhibition of photochemistry, has been taken as an indicator of the thermal stability of the PSII pigment system. More precisely, the fluorescence rise has been attributed to the physical separation of the LHCP from the PSII core, thereby blocking excitation energy transfer and leading to reemission of excitation energy from LHCP as fluorescence (Armond gt g1, 1978). The temperature at which enhanced fluorescence occurs may vary in response to environmental adaptation and appears to be affected by the lipid environment in which the proteins are embedded (Lynch and Thompson, 1984). In this respect, Chl fluorescence may be considered an intrinsic probe of lipid-protein interaction. The effect of temperature on mutant and wild-type leaves was measured by appressing leaves to a temperature-controlled metal block 77 ‘ Relative Fluorescence 25 35 45 55 Temperature (°C) Figure 16. Temperature-induced fluorescence enhancement yield (F ) of wild type (I) and mutant 0860 (0) leaves. Plants were grown at 21 C. The arrow indicates the threshold temperature at which fluorescence is enhanced. Each point represents the mean 13E (n24). - 78 and measuring fluorescence continuously as the temperature of the heating block was increased from 25 to 56 C at a rate of about 1.5 C/min. At approximately 37 to 38 C a transition in the level of fluorescence was observed in both wild-type and mutants leaves (Figure 16). This response is similar to that observed by comparable experimental approaches with other species (Armond gt g1, 1978; Lynch and Thompson, 1984; Schrieber and Berry, 1977). There was no significant difference in the threshold temperature or magnitude of the fluorescence response of the mutant as compared to the wild-type. Thus, it does not appear that the absence of ttggg-C16:l-PG has a significant effect on the thermal stability of the LHCP-PSII core association. Discussion The normal growth, chloroplast ultrastructure, and rate of photosynthesis in the thA mutant grown under standard conditions indicate that the role of ttggg-C16:1 is subtle. The most striking effect attributable to the mutation is the relatively reduced amount of LHCP oligomer which was recovered from mutant membranes following detergent-mediated thylakoid solubilization (Figures 8 and 9). This observation suggests a role for trans-Clszl-PG in stabilizing the LHCP oligomer. The simplest hypothesis to explain the apparent instability of LHCP oligomer in the mutant would seem to be that PG containing treat-C16:1 is more effective at preventing SDS from penetrating the subunit contact sites of the LHCP oligomer than PG alone . This is 79 consistent with the results from previous studies showing that phospholipase-AZ treatment of membranes leads to loss of the LHCP oligomer (Remy gt g1, 1982), and with studies showing that the LHCP oligomer extracted from SDS gels appears to be specifically enriched in PG containing trans-C16:1 (Tremolieres gt g1, 1981). The results of experiments in which the presence of ttggg-C16:l-PG in artificial liposomes enhanced the rate (but not the amount) of reconstitution of LHCP oligomer (Remy gt g1, 1984) also supports the concept that the lipid facilitates formation or stability of the LHCP oligomer. However, these observations are difficult to interpret since, for example, the addition of Triton X-100 to the SDS-solubilization buffer was also reported to increase the amount of LHCP1 at the expense of LHCP3 (Anderson gt al, 1978). Similarly, the solubilization of thylakoids with the nonionic detergent octyl-B,D-glucoside rather than with SDS resulted in loss of the apparent ttggs-Clszl-PG/LHCPl association (Henry gt g1, 1983). Also, analysis of the lipid composition of mechanically isolated stroma and grana lamellae revealed that the stroma lamellae, which contained low amounts of LHCPl had higher levels of ttggg-C16:l-PG than PSII granal vesicles (Guillot-Salomon, 1982). Thus, it appears that the effect of the loss of iiflfli'c16;1‘PG on LHCP1 stability may reflect a nonspecific change in the overall properties of the photosynthetic lamellae rather than a specific effect on LHCP pg; gg. The observation that the CPla oligomer is also less stable in the mutant (Figure 9) lends credence to this view. It is well established that cations stimulate thylakoid stacking (Izawa and Good, 1966), a process which involves LHCP (Staehelin and 80 Arntzen, 1983). Thus the possibility was considered that the differential cation enhancement of SDS-mediated dissociation of LHCP oligomer in the mutant was related to properties of the LHCP oligomer involved in bringing about membrane appression (Argyroudi-Akoyunoglou, 1981). We analyzed the effect of cations on LHCP function by measuring the effect of cation concentration on the efficiency of exciton transfer from LHCP to PSI (Figure 15). In principle, cation-induced dissociation of LHCP oligomers might be expected to lead to less efficient exciton transfer to PSI and, therefore, to increased fluorescence from LHCP at 685 nm. Although both mutant and wild-type showed a change in the ratio of PSI to PSII fluorescence, there was no significant difference between the two lines at any cation concentration. The absence of a differential effect of NaCl on the ratio of Chl fluorescence at 685 and 734 nm is considered evidence against an important role for ttggs-C16:1 in conferring unique functional properties to the LHCP oligomer 1g yiyg. Similarly, the absence of a differential effect of divalent cations on fluorescence induction kinetics in mutant versus wild-type (Table 12) indicated that the photochemical efficiency of PSII reaction centers are indistinguishable in the two lines. The effect of NaCl on the proportion of Chl found in LHCPl is, therefore, probably due to a stimulation of the activity of SDS rather than a specific effect on LHCP quaternary structure. We must conclude that the lipid has no significant 13_yiyg effect on LHCP quaternary structure. The apparent absence of an effect of the thA mutation on PSI or PSII activity contrasts with the results of experiments involving lipase treatment of thylakoid membranes which were designed to examine 81 the role of ££35§'C16:1' The lipase treatment was intended to exploit the head group and positional specificity of phospholipase-A2 to catalyze preferential removal of the acyl group at position two of PG and phosphatidyl choline (Rawler and Siegenthaler, 1981). In one study, phospholipase-A2 treatment increased the amount of light required to saturate the Hill reaction, decreased the variable fluorescence and increased the time required to reach maximal fluorescence (Duval gt g1, 1979). However, the implications of these observations were disputed by Rawyler and Siegenthaler (1981) who showed that both PG and phosphatidylcholine were affected by phospholipase-A2 to varying degrees, depending on the source of enzyme, and that PSII activity was severely depressed by phospholipase treatment. Whatever the reason for the effects of the lipase treatments the discrepancy between the functional properties of lipase-treated thylakoids and those of the thA mutant illustrate the limited utility of lipolytic analysis in attempting to determine the functional significance of specific acyl groups. The results of several studies have provided evidence that membrane lipid composition may exert an important influence on the stability of the association of LHCP with the PSII core (Armond gt al, 1978;Lynch and Thompson, 1984; Schreiber and Berry, 1977). Indeed, on the basis of correlations between adaptive changes in lipid composition and the threshold for temperature-induced fluorescence, it has been suggested that Lnflnfi'c16:1'PG could play a role in mediating thermal stability of the LHCP-PSII complex (Lynch and Thompson, 1984). However, the absence of any differential effect of the thA mutation on the threshold 82 temperature for temperature-induced fluorescence (Figure 16) renders a specific role for trans-C16,1 in thermal adaptation untenable. In conclusion, although we have independently reproduced the evidence for an effect of trans-C16,1 on in vitro LHCP oligomer stability, we have not observed any functional significance associated with the absence of L£§fl§'cl6:1' On this basis we propose that ttggg-Clszl—PG normally has no effect on the function of the photosynthetic lamellae. One possibility is the role is either restricted to an different environmental circumstance that we have not investigated, or to a specific phase of development. For example, since ttggt-C16:1 accumulates concommitently with LHCP accumulation it seems possible that it facilitates insertion of proteins into the thylakoid membranes and thereby leads to more rapid membrane assembly. The observation that the rate of LHCPl formation is enhanced in liposomes containing ttggs-Clezl (Remy gt g1, 1982) may be considered preliminary evidence in favour of this concept. Alternatively it is possible that this lipid is an element of the fine tuning mechanisms (Butler, 1978) which have evolved to optimize the efficiency of photosynthetic electron transport. This possibility would provide an example of the principle that many components of organisms may not be absolutly required, but serve very subtle functions which might only give a marginal selective advantage to the organism at best. CHAPTER 5 A MUTANT DEFICIENT IN 18:3 AND 16:3 FATTY ACIDS Introduction The synthesis of -linolenate (18:3) in higher plants occurs by the sequential desaturation of stearate. The first double bond is inserted by a soluble chloroplast enzyme which utilizes stearoyl-ACP as its substrate and is closely associated with the fatty acid synthetase (Nagai and Bloch, 1968). The second and third double bonds are introduced only after the fatty acid has been incorporated into a glycerolipid molecule. As described in Chapter 1, glycerolipid synthesis is thought to involve two discrete pathways which have been designated the ’prokaryotic’ and ’eukaryotic’ pathways (Roughan land Slack, 1982). In this scheme, the 16:0 and 18:1 fatty acids synthesized gg ggyg in the chloroplast may either enter the prokaryotic pathway in the chloroplast envelope or be exported as CoA esters to enter the eukaryotic pathway at extrachloroplast sites predominantly localized in the endoplasmic reticulum. In ’16z3 species’ such as Argbidopsis ttgjjgug both pathways contribute to the production of chloroplast membrane lipids (Browse gt g1, 1986). 83 84 It is not yet apparent how many distinct desaturases are active in the leaves of 16:3 plants. Isolated intact chloroplasts of 16:3 plants are able to synthesize MGD by the prokaryotic pathway (Mckee and Hawke, 1978; Roughan gt al, 1979), and 18:1 esterified to position gg-l of this MGD is desaturated to 18:2 and 18:3. Similarly, 16:0 at position sg-Z of MGD is converted by sequential desaturations to 16:3 (Roughan gt al, 1979). Desaturation of 18:1 and 18:2 on lipids synthesized by the prokaryotic pathway is not confined to MGD since [14C]-18:l-PG synthesized by chloroplasts is sequentially converted to labelled 18:2- and 18:3-PG (Roughan, 1985). For lipids synthesized by the eukaryotic pathway, PC located in the endoplasmic reticulum is the predominant substrate for desaturation of 18:1 to 18:2 (Roughan and Slack, 1982; Slack gt g1, 1976). MGD of the eukaryotic pathway is thought to be the main substrate for the desaturation of 18:2 to 18:3 in leaves (Hawke and Stumpf, 1979; Roughan and Slack, 1982), although microsomal PC is probably the substrate for this reaction in developing seeds (Browse and Slack, 1981; Stymne and Appelquist, 1980). Unsaturated C-16 fatty acids are not produced to any extent by the eukaryotic pathway in leaves. Although a broad outline of the pathways of lipid desaturation is available, many uncertainties remain. In particular, it is not yet established which lipids are substrates for desaturation, how many distinct desaturases exist, and whether 18:2 to 18:3 conversion occurs outside as well as inside the chloroplast. Given the predominance of trienoic fatty acids in leaf lipids and their suggested importance to photosynthesis and other plant functions (Gouraris and Barber, 1983; Quinn and Williams, 1983; Raison, 1980) it is important 85 to fully understand the operation and control of the fatty acid desaturases. However, each desaturation reaction is believed to involve the interaction of several membrane-bound components (Okayasu gt g1, 1981; Slack gt g1, 1976; Strittmatter gt g1, 1974) which have not yet been characterized. This chapter describes the biochemical characterization of a group II mutant ttggg) which is deficient in 18:3 and 16:3 in the leaves and contains increased amounts of 18:2 and 16:2. Results i h m h r r a io The probable nature of the biochemical lesion in the {gap mutant is inferred from the observation that in leaves of plants grown at 26 C, the decreased amount of 16:3 and 18:3 fatty acids was accompanied by an increase of similar magnitude in the amounts of 16:2 and 18:2 (Table 7). Thus, although other possibilities are considered (see below and next chapter), the simplest hypothesis is that the mutant is deficient in a fatty acid desaturase which is normally responsible for introducing the double bond at position 15 of 18-carbon acyl groups and at position 13 of 16-carbon acyl groups. If this hypothesis is true the site for insertion of these double bonds is determined relative to the methyl end of the chain since it is n-3 in both cases. Furthermore, it seems likely that this desaturase is located in the chloroplast because chloroplast MGD is believed to be the substrate for 16:3 synthesis (Roughan gt g1, 1981). 86 As noted in Chapter 1 during the preliminary characterization of the mutant we observed some variability in the proportion of both 18:3 and 16:3 from one experiment to another. Analysis of the effects of various environmental influences led to the recognition that the amount of these fatty acids is dramatically affected by temperature in the mutant line but is much less affected in the wild-type (Figure 5). When grown at 18 C the fatty acid composition of the mutant is similar to that of the wild-type. By contrast, when grown at 26 C the mutant has only about 35% as much trienoic fatty acids (ie., 16:3 + 18:3) as the wild-type grown at the same temperature (Figure 5). As noted in a later section, the trienoic acid synthesis which occurs above 26 C may be due to the action of a second desaturase. The simplest explanation for this observation is that the genetic lesion in the fng mutant renders a desaturase or a regulator of the desaturase(s) temperature sensitive so that it is almost normally functional at low temperatures but is largely inoperative at temperatures above about 26 C. To further characterize the temperature effects on trienoic acid synthesis in the fng mutant, temperature shift experiments were performed in which plants were grown at nonpermissive conditions for trienoic acid desaturation (28 C) for three weeks and then shifted to permissive conditions (19 C). The new synthesis of 16:3 can be detected within 48 hours after shifting and in the case of 18:3, changes in the total pools can be detected as early as 18 hours. (Figure 17) The difference in the rate of recovery of 16:3 versus 18:3 can not be explained at this time due to the lack of any knowledge of the product of the thQ gene or the kinetics of desaturation. However, it could in mutant and wild type lenoic fatty aci Ratio of tr Figure 17. 87 .0 O) l ' 1 l l O 25 50 75 Time (hr) ' Ratio of the amount of trienoic fatty acids in mutant line 081 and wild type after shifting plants from 27 C to 19 C 88 reflect different substrate specificities of the n-3 desaturase for 16:2 and 18:2. The rapid increase in 18:3 synthesis upon shifting the mutant to 19 C is of interest with respect to questions concerning sites of synthesis and final location of trienoic acid. Although the thylakoid lamellae, which are highly enriched for 18:3 and 16:3 (Table 13), account for almost 80% of all membranes in higher plants (Harwood, 1980), it is believed that the major site of synthesis of these fatty acids is the chloroplast envelope (Douce and Joyard, 1980). If this is true, upon shifting the thQ mutant to permissive conditions the thylakoid membranes must import large amounts of 18:3 and 16:3 from the envelope. Unless there is an exchange of fatty acids out of the thylakoids for the incomming trienoic acids, this mechanism would demand massive membrane growth. Alternatively the thylakoid membrane may also contain n-3 desaturation activity which is affected by the ngQ mutation and hence the desaturation of 18:2 may occur directly in the thylakoid without lipid transfer. Egtty Acid Composition of Individual Lipids Analysis of the fatty acyl composition of individual lipids from leaf tissue of mutant and wild-type plants grown at 28 C revealed that all the major polar lipids are affected by the thQ mutation (Table 14). However, the proportions of various polar lipids in the extracts were essentially the same for the mutant and the wild-type. The decreased levels of 18:3 and 16:3 and an increase in the corresponding dienoic fatty acids are the most striking differences between the wild-type and the fadQ mutant. In addition, the amount of 18:1 appears 89 Table 13 Fatty acid composition of wild type and 081 thylakoid membranes from plants grown at 28 C Fatty Acid Composition (mol %l Fatty Acid WT 081 16:0 10.5 9.2 t16:1 4.8 4.4 c16:1 0.8 1.0 16:2 2.6 13.8 16:3 '13.8 1.4 18:0 0.9 0.7 18:1 2.4 4.5 18:2 7.4 37.7 18:3 46.3 20.7 9O qmcdm 3a «mane >nsa nosuomsnsosm om dmmm sateen macs :sda-weum use scams" em— >1mchOUmsm. meet: an mmon. --—--- . - 2 o o- o ' I 1 I L i 0 3 6 0 3 6 Time (days) Figure 20. Cnl and trienoic acid content following a shift of wild-type (O) and mutant (0) plants from 19 C to 27 C. Plants grown at 27 C were transferred to 19 C (A and C) or maintained at 27 C (B and D). Each point represents the mean 1- SD (n=3). 109 similar but less dramatic changes in thylakoid ultrastructure to those observed in SAN9785 treated barley seedlings (Leech gt g1, 1985) (Table 17). The granal width was increased 20% with a similar reduction in stromal thylakoids and a slight reduction in grana/plastid. However the most striking feature of the morphometric analysis was the overall reduction in plastid size (Table 17). In the mutant, chloroplasts were roughly half the size of wild-type and both granal and stromal membranes were reduced to approximately 73% and 64% of wild-type amounts, respectively. This overall reduction in thylakoid and envelope membrane in the mutant is consistent with the lipid, Chl, and protein ratios shown in Table 16. To determine if the smaller chloroplast size was compensated to any degree by increased chloroplast number, protoplasts were isolated from both wild-type and fng leaves and the number of chloroplasts per protoplast were counted under a light microscope (Table 18). The mutant appears to have more chloroplasts per protoplast than wild-type raising the possibilty that smaller chloroplasts are the result of increased chloroplast divisions. Phgtosynthetig Studies. A preliminary test of the effects of the reduced trienoic acid content on photosynthesis was performed by measuring photosynthetic gas exchange in mutant and wild-type plants at various light intensities (Figure 21). Although COZ-fixation rates of mutant and wild-type were indistinguishable when expressed on a Chl basis at all light intensities, the mutant showed a 20% reduction in COZ-fixation rates when measured on the basis of fwt (Figure 21,Table 19). This is 110 Table 17 Morphometric analysis of chloroplasts from mutant J81 and wild type Arabidopsis. Measurements were made on 20 chloroplasts from each line. WT J81 grana/plastid 54.1 i 13.0 37.2 i 8.2 thylakoids/granum 5.5 i 3.3 4.8 i 2.5 granal width (um) 0.40 i 0.03 0.48 i 0.03 stroma thylakoids/plastid (um) 67.0 i 20. 55.7 i 14.2 stroma thylakoid length (um) 0.26 1 0.03 0.20 i 0.03 total grana (um/plastid) 119.0 85.7 total stroma (um/plastid) 17.4 11.1 total thylakoids (um/plastid) 136.4 96.8 grana/stroma 6.8 I 7.7 surface area (umz/plastid) 9.7 i 2.0 5.3 i 1.6 111 '- r l 1 I 1 I .z: T 150r- _. E . 0 a e a 8 «100.. 3 . ° a U - .5 a E ‘3 50r- . 2 g e u. N o 1 1 l 1 J 8 200 _ 400 600 800 1000 lrradiance (1L5 111-2 3.1) 1 l r T I T -'-' 7.5r- .. T. E E I C 945.0?" l ¢ é - c ‘I U I: .3 ? C E 2.5:- - . LL a 1 C o o I I l I . 200 400 600 800 1000 Irradiance (an m‘2 s") Figure 21. C0 fixation rates versus light intensity for mutant 081 (0) , an wild type (0) Aradogisis on a chlorophlyll and fresh weight basis. Each point represents 3 plants. 112 Table 18 Number of chloproplast per protoplast in mutant 081 and wild type Arabidopsis grown at 19 C and 27 C. (n = 50) Temperature HT 081 19 C 34.6 i 11 32.4 i 10 27 C 40.1 i 15 57.9 i 25 113 consistent with the overall reduction in the photosynthetic membranes in the mutant. In order to measure the effect of lipid composition on the light reactions catalyzed by isolated thylakoids, plants were grown at 27 C to ensure maximum reduction of trienoic acid content and then assayed at various temperatures. Whole chain electron transport rates were very similar in mutant and wild-type at 6 C, 14 C and 25 C suggesting reduced unsaturation is not a rate limiting factor in plastoquinone diffusion (Table 19). Uncoupled PSI and PSII rates were also not dramatically different between mutant and wild-type, indicating that a high concentration of trienoic acid is not required to support these activities. These results contrast with studies in which thylakoids were exposed to hydrogenation in the presence of a water soluble paladium catalyst which reduced the level of unsaturated fatty acids (Vigh gt g1, 1985). Such exposure caused loss of whole chain but not the partial reactions of electron transport (Vigh gt g1, 1985). Our results suggest the loss of whole chain activity can not be attributed to trienoic acid content. It probably reflects the nonspecificity of the hydrogenation treatment which also reduces 18:2 to 18:1. Futhermore the observation that non water-soluble catalysts which cause near identical reductions in fatty acid unsaturation, do not affect electron transport rates also suggests that water soluble hydrogenation catalysts are nonspecific in action (Restall gt g1, 1979). 114 Table 19 Comparison of photosynthetic activities in mutant J81 and wild type Arabidopsis vi 031 C02 fixation (umol coz mg chI‘1 h'l) 116.0 i 2.5 114.1 _ 2.5 (mg co2 mg fwt h'l) 6.8 i 0.6 5.5 0.2 Electron transport rates (umol 02 mg chl'1 h’l) Whole chain 25 c 261.4 i 7.1 270.7 g 19.1 14 c 202.3 i 5.9 200.0 _ 9.3 6 c 132.9 g 7.5 123.8 _ 11.3 PSII 25 c 261.8 i 14.5 ~265.7 _ 4.3 PS1 25 c 375.4 i 9.5 422.4 _ 16.7 115 Fluorescenge Measurements. As mentioned earlier, spectral analysis of fluorescence emitted from thylakoids excited with 440 nm light at 77 K can resolve three major peaks at 685, 695 and 734 nm. These peaks have been attributed to LHCP associated with PSII, PSII reaction centers, and PS1 respectively (Bose, 1983; Butler, 1978). Interestingly SAN9785 treated plants show large enhancements of fluorescence at 685 and 695 nm relative to 734 nm at low temperature (Leech gt g1, 1985). The ratio of fluorescence at 685 nm/734 nm is considered a measure of PSII to PSI stoichiometry and in herbicide treated plants might reflect the altered ratio of appressed to non-appressed membranes seen in electron micrographs (Leech gt al, 1985). A comparison of the F685/F734 ratio in mutant and wild-type thylakoids did not demonstrate any change in PSII/PSI organization in the absence or presence of MgCl2 (Table 20). The fact that we did not observe any difference between the two genotypes under either condition implies that the Chl-protein complexes are structurally similar and the efficiency of energy transfer between the major complexes is normal. This conclusion was substantiated by measuring variable fluorescence (F0) at room temperature in the mutant and wild-type. Room temperature fluorescence is a kinetic measure of PSII activity and therefore it is a sensitive indicator of actual light capturing and P511 photochemistry. However, variable fluorescence was indistinguishable between mutant and wild-type in both high and low MgCl2 implying normal PSII photochemistry efficiency in the mutant (Table 20). 116 Table 20 Room Temperature (F /F ) and Low Temperature (F /F ) fluorescence in isolated thylakoias from mutant 081 and wiléstygg4Ar abidopsis Fv/Fo F685/F734 Wild Type (+Mg) 2.73 1 0.11 1.49 1 0.08 081 (+Mg) 2.61 1 0.05 1.52 1 0.08 Wild Type (-Mg) 1.03 1 0.07 0.87 1 0.01 J81 (+Mg) 1.26 1 0.16 0.88 1 0.07 117 Membrgne Fluiditx To estimate the effects of trienoic acid composition on the fluidity of thylakoid membranes fluorescence polarization measurements were carried out on freshly isolated thylakoids from mutant and wild-type. The principle of the measurement is that the fluorophor DPH readily partitions into thylakoid membranes (Barber gt g1, 1984), and upon excitation with polarized light emits polarized fluorescence. The orientation of the probe upon excitation and the plane at which the polarized fluorescence is re-emitted is directly dependent on the rotational diffusion of the probe during its excitation lifetime. This diffusive motion is in turn dependent on the viscosity of the membrane. The more fluid a membrane the faster the rotational diffusion of the probe during its excitation lifetime and hence the more depolarized the re-emitted fluorescence thus, high fluidity is reflected in low fluorescence polarization values (P). The P values obtained for Arabidgpsis thylakoids are very similar to those of other species (Barber gt 11, 1984) and in general indicate a very fluid membrane (Figure 22). Although the mutant shows higher polarization values at every temperature tested in comparison to wild-type samples the difference was just at the limit of statistical significance (Figure 22). These results suggest that considerable changes in the level of unsaturation by removal of the n-3 double bond caused only a small decrease in membrane fluidity. Physical studies on melting temperatures of 18:3 versus 18:2 and 18:1 are consistent with these findings and imply that the position and not the number of double bonds is important in maintaining fluidity (Stubb and Smith, 1984). 118 .. b O. c: f . 6 6 43 (L13- I f Y 5 _ - L 1 I .9 I - I o I l 0.10 2:. - d) 0 c: d.) S 30.1» - O E. u. 1 L I I I 1 0'0 -o 10 2o so 40 so Temp (°C) Figure 22. Effect of temperature on DPH fluorescence polarization by thylakoid membranes from wild-type (0) and mutant (0) 081 Arabidopsis. Each point represents three independent replicates. ‘ 119 Considered together, the role of trienoic acids in membrane fluidity is relatively minor at this level of resolution. Effects of Unsaturation on Protein Diffusion. It has recently been recognized that light harvesting antenna complexes can undergo reversible lateral diffusion between appressed and non-appressed regions of thylakoid membranes under the appropriate environmental conditions (Staehelin and Arntzen, 1983). This phosphorylation-dependent process can be artificially regulated in isolated thylakoids by the addition of ATP and light and appears to be controlled by a protein kinase (Staehelin and Arntzen, 1983). The physical movement of LHCP in turn leads to a redistribution of energy between PSI and PSII. With the lateral movement of LHCP away from PSII in the grana towards PSI in the stromal lamellae, a greater portion of the incident radiation is partitioned to PSI thereby giving increased fluorescence at 734 nm. Because phosphorylation and subsequent movement of the complexes from PSII to PS1 is a relatively slow process, the increased fluorescence from PS1 (F734) relative to PSII (F685) can be monitored by 77K fluorescence emission spectra. Analysis of low temperature fluorescence emission of PSII and PS1 spectra indicate that LHCP phosphorylation does lead to a redistribution of absorbed energy between these photosystems in Atattdggsis and therefore LHCP diffusion but the kinetics of diffusion are nearly identical between mutant and wild-type samples (Figure 23). Therefore reduced unsaturation does not seem to hinder movement of 120 .0 o: I ‘ 3- P \l I .0 01 I 1 1 0'5 0 6 10 15 20 Time (min) Ratio of Fluorescence Intensity ('734/f685) Figure 23. Rate of ATP-induced change in the ratio of 77K fluorescence (F734/F685) from thylakoid membranes of wildatype (I) and mutant 081 (0) Arabidopsis. Each point represents the mean of three samples. 121 macromolecules such as proteins to any major extent in thylakoid membranes. Effects of High Temperature on Fluorescence As with the thA mutant, the stability of pigment-protein complexes in the new lipid enviroment caused by the fadfl mutation were assayed by heat induced fluorescence yield enhancement. However there was no significant difference in the temperature at which fluorescence started to increase in the mutant as compared to wild-type or in the magnitude of the responses (Figure 24). Thus the stability of the LHCP-PSII core association is not significantly affected by the reduced 18:3 and 16:3 composition of the membranes. Discussion Although the ngQ mutation has no apparent effect on growth rate under controlled conditions the mutant is somewhat chlorotic by comparison with the wild-type due to a major reduction in the amount of chloroplast membranes per cell when grown at temperatures above 26 C. This alteration is expressed both as a reduction in the size of chloroplasts and in the amount and distribution of lammelar membranes within the chloroplasts (Table 17). Futhermore, although the chl/lipid and protein/lipid ratios are unchanged, the number of chloroplasts per cell appears to be substantially higher in the mutant (Table 14, Table 18). 122 Ooo ‘ o o O O .00 O O o O O o O O 8 O O C O O 3 on g, o. o o. 2 0. LL 0 0 00. 7: 0000000000000000 o°° 0.. m C O. 1: 1 1 1 1 1 I 1 30 40 50 Temp (°C) Figure 24. Temperature-induced fluorescence enhancement yield (f0) of wild-type (0) and mutant (0) Leaves. Plants were grown at 27 C. Each point represents the mean of 4 indeperxient measurements. 123 The redistribution of the thylakoid lamellae, although similar to those reported in SAN9785 treated barley seedlings are less pronounced (Leech gt at, 1985). Ultrastructural studies with 6 day old barley seedlings treated with sublethal doses of the herbicide SAN9785 reduced the amount of 18:3 from 76 mol % to 28% and increased the 18:2 levels from 7 mol % to 49%, in thylakoid membranes but caused no change in total lipid or chl levels/cell (Leech gt g1, 1985). In the treated plants, however, the grana were approximately 60% wider than the controls with fewer thylakoids/stack and less stromal lamellae (Leech gt 11, 1985). Although these structural changes were supported by lower chl a/b ratios and altered fluorescence emission, other chemicals which do not specifically effect 18:3 levels can cause similar ultrastructural and biochemical changes (Bose and Mannan, 1984; Festke gt al, 1977). Therefore we believe that many treatments which cause perturbation of the photosynthetic apparatus during development can cause alterations in granal/stromal ratios irrespective of lipid composition. By using chl/fwt ratios as a convenient indication of membrane growth it was possible to show that this parameter is affected by temperature in a similar way to that of trienoic acid synthesis (Figure 20). Along with the observation that these two traits co-segregate from an F2 population derived from a wild-type x fgdfl cross we believe that the fng gene product must somehow regulate both phenomena. There are two simple possibilities. l) The fng gene product is some cellular component which controls the assembly and growth of chloroplast membranes and the n-3 desaturase is one of the enzymes regulated by this gene product. 2) The fadD gene product is the structural gene for 124 the n-3 desaturase and the reduction in trienoic acids due to the mutation causes the reduction in chloroplast membrane. The former hypothesis would explain one of the discrepancies between this study and that in which 18:3 was reduced by treating seedlings with SAN9785. The changes in photosynthetic function attributed to decreased 18:3 content in the herbicide experiments are considered artifactual since we did not observe any similar change in the thQ mutant. Using the same reasoning it can be argued that reduced chloroplast membrane size observed in the mutant is not directly due to the trienoic acid levels since this does not occur in SAN9785 treated barley. Thus the effect measured is probably due to some other function affected by the altered {gap gene product. Moreover if the {gap gene product does regulate a number of functions the fact that the mutation is temperature sensitive and somewhat leaky might be because a null mutation in such a regulatory gene is lethal to the plant. The direct testing of the above hypotheses of thQ action is at this time not possible but the fact that n-3 desaturation is closely regulated with membrane growth in the fng mutant strongly suggests a role for 18:3 and 16:3 in membrane biogenesis. Support for this proposal can be inferred from developmental and greening studies of plants which have shown a direct correlation of trienoic acid synthesis and membrane biogenesis (Leech and Leech, 1976). Although the fgdfl mutation translates into decreased unsaturation and structurally modified chloroplasts these changes have little effect on any functional aspect of this organelle. Photosynthesis as assayed by C02 fixation, electron transport (Table 19), and room temperature fluorescence (Table 20) showed no significant differences. Net C02 125 fixation was reduced on a fwt basis but this is only a reflection of the smaller photosynthetic apparatus since this difference was not detected on a chl basis. These results are in sharp contrast to studies carried out with chemically modified membranes (Leech gt al, 1985; Bose and Mannan, 1984; Raison, 1980). We therefore suggest these methods induce non-specific changes and are not by themselves adequate to determine the functional significance of trienoic acids in vivo. It has been suggested that the structural organization of the chloroplast involves lateral separation of the two photosystems (Anderson and Melis, 1983; Staehelin and Arntzen, 1983). Membrane fractionation studies suggest LHCP-PSII protein complexes are primarily located in granal or appressed regions of the thylakoid lamellae and PSI is associated with non-appressed or stromal membranes. The spatial separation of the photosynthetic light reactions necessitates the movement of plastoquinone and probably plastocyanin as long range carriers of electrons. Also, it is now well established that control of energy distribution between P511 and PSI under different light conditions involves the movement of LHCP between the reaction centers (Staehelin and Arntzen, 1983). Although each case involves very different diffusion constants, both movements are dependent on the membrane fluidity (Barber gt 11, 1984; Chapman gt g1, 1983). In a more general sense the lateral movement of all the membrane components is also determined by rates that are optimal for the overall system. Because of the high proportion of polyunsaturated fatty acids in the thylakoids, and in particular trienoic acids, it has been suggested that these molecules play a major role in maintaining an extremely fluid matrix for lateral movement of photosynthetic components. The 126 introduction of double bonds into linear fatty acids causes kinking of the chains thereby disordering hydr0phobic interactions between the fatty acids which in turn causes a more fluid bilayer. The relatively low fluorescence polarization values for thylakoid membranes of Arabidopsis (Figure 22) and other plant species (Barber gt al, 1984) does reflect the very fluid environment of this compartment compared to other bilayers. The reduced trienoic acid in the fggp mutant, however did not decrease the fluidity to any great extent (Figure 22). Rates of plastoquinone and plastocyanin diffusion as assayed by PSII and PSI partial reactions (Table 19) and protein diffusion as assayed by LHCP movement from PSII to PS1 (Figure 23) support this view. Taken together with results obtained from model membrane systems (Stubbs and Smith, 1984) the conclusion drawn is that major changes in unsaturation may often result in only minor changes in fluidity. The lack of any major effect of the fadfl mutation on photosynthesis probably reflects the bias of a nonselective screen for fatty acid alterations. However the structural changes described above have defined a new class of mutants for chloroplast developmental studies. To date the majority of previously isolated mutants which affect chloroplast membrane development also disrupt photosynthetic function. This is largely a reflection of the screening procedure for isolation of such mutants which has relied on chlorosis or variegation as a phenotype. By contrast, mutant classes such as fadfl were screened by looking for specific changes in a chloroplast component. By using such a criterion one can isolate mutations which alter chloroplast structure without disrupting function. In this sense, changes in fatty 127 acid metabolism may be an excellent method for the isolation of mutants with altered membrane growth and organization. In conclusion it appears the high proportion of trienoic acid found in chloroplasts plays little if any role in the function of photosynthesis as assayed by a number of criteria. The observation that the growth of both chloroplast envelopes and thylakoids is inhibited in the fng mutant however leads us to conclude that their is a function for trienoic acids in membrane biogenesis. The isolation of a collection of mutations which affect 18:3 and 16:3 levels would be very useful in further determining this relationship. Not only would such a collection allow the determination of the number of loci involved in the synthesis and regulation of these fatty acids but the recovery of a nontemperature sensitive null mutant which totally lacks 18:3 and 16:3 might shed some light on the nature of the ing mutation. For example, if the fgdfl phenotype is due to change in a multifunctional regulator of chloroplast development it should be possible to isolate mutations in the structural gene(s) for trienoic acid synthesis which do not show these pleiotropic effects. Furthermore because the fgdfl mutation is somewhat leaky for trienoic acid synthesis it can be argued that many of the functions in which these fatty acids play a role can still occur. APPENDIX A Temperature Induced Fluorescence As a method for estimating the stability of pigment-protein complexes in thylakoid membranes the technique of temperature induced fluorescence has become very popular (Schreiber and Berry, 1977; Armond gt al, 1978). A number of studies have now been done to see if changes in fatty acid composition commonly seen in plants acclimating to various temperatures can be correlated with changes in fluorescence emmission (Raison gt 31, 1980; Lynch and Thompson, 1984): The general approach to using this technique is as follows. Detached leaves or isolated thylakoids are placed in a temperature controlled cuvette and 2 5‘1, 480 nm) causing the sample excited with weak blue light (0.3uE m' to fluoresce at a low constant level (F0). F0, which represents photochemically inactive chlorophyll at room temperature, and is mostly PSII fluorescence from chlorophyll a (Armond gt gt, 1978). As the temperature is increased a sudden increase in F0 is observed at temperatures which appear to be dependent on the temperature at which the plant was grown (Armond gt gl, 1978). Plants grown at higher temperatures have fluorescence breaks at higher temperatures than plants grown at lower temperatures. From experiments using different 128 129 excitation wavelengths of light on plants grown at various temperatures it has been suggested that the increased fluorescence is due to an energy block between chlorophyll b (chl b) and chlorophyll a (chl a) (Armond gt g1, 1978). Normally the effiency of transfer between these two pigments is nearly 100%. Further studies involving low temperature fluorescence spectra of leaves heated above their break point showed a new 660 nm peak which probably represents free chl b. These results have led to a model suggesting that upon heating of the sample the chlorophyll a/b binding protein complex which contains all the chl b in a plant becomes dissociated from the PSII reaction centre thereby not allowing energy transfer to occur between chl b and chl a. Because such a system could be useful as an assay of the effects of fatty acid alterations on membrane stability a more detailed characterization of heat induced damage on chloroplast function was carried out on wild-type Arabidopsis. As a first test of the use of this system, plants were grown at 12 C and 22 C until leaves were large enough to be assayed (2 to 3 weeks). Plants grown at cooler temperatures do show a lower break point in fluorescence emission than 22 C grown plants (Table 21). Although this change in fluorescence does correlate with changes in unsaturation ratios, the differences were not dramatic (Table 21). Further analysis of the nature of the fluorescence rise was studied using low temperature fluorescence spectra of wild type leaves subjected to temperatures above the normal break point of 22 C grown plants (Figure 25). A typical spectrum shows increased 734 nm (F734) and 695 nm (F695) fluorescence peaks relative to 685 nm (F685) fluorescence (Figure 25). It is generally accepted that F685 belongs to 130 Table 21 Temperature induced fluorescence breaks and unsaturation ratios in wild type plants grown 12 and 22 C 12 C 22 C WT-l 40.4 45.2 WT-2 43.8 46.2 WT-3 42.9 45.0 AVERAGE 42.9 1 1.8 45.5 1 0.6 Unsaturation Ratio 5.47 4.99 18:3+18:2+18:1+16:3+16:2+16:1 Unsaturation Ratio = ............................. 16:0+18:0 131 RELATIVE FLUORESCENCE I — M‘— 600.00 _ 800.00 WAVELENGTH (nm) Figure 25. Low temperature fluorescence (77K) emission spectra of wild type leaf samples incubated at 22 C (—) or 53 C (--) for 5 min before freezing of the samples. Both spectra were normalized to 685 nm. 132 antenna of PSII and probabaly LHCP. That F695 is from the antenna of PSII and the core antenna of PSI and that F734 is fluorescence from the peripheral antenna of PS1 (Butler, 1978; Bose, 1982). Because of the quenching of F685 and F695 by the high level of chlorophyll in the leaf and to determine which peaks are actually increasing or decreasing a similar experiment was carried out using thylakoids to which an internal standard was added (Carbofluor 1 ug ml'l, F max - 500 nm). Comparisons of unheated and heated samples normalized to the internal stadard show that PSI external antenna are not dramatically altered by the heat treatment although the maxima was slightly blue shifted (Figure 26). In contrast the PSII-LHCP associated fluorescence peaks are significantly affected, showing decreased fluorescence yield and shifts in peak maxima (Figure 26). This result supports the Armond hypothesis that PSII is preferentially affected by heating (Armond gt 31, 1978). The increased F695 and decreased F685 suggests that the antenna of PSII has somehow been affected. One possible cause for these changes could be the separation of LHCP from the rest of the PSII core particle. Recently emission spectra from purified LHCP have been shown to have an emission maximum of 680 nm (Bose, 1982). Therefore separation of LHCP from PSII should be reflected in a 5 nm shift in the 685 nm peak towards the blue. To see if such a shift was occurring in thylakoids, samples were heated for various times and frozen in liquid nitrogen for low temperature fluorescence analysis. The 680 to 690 nm region of the spectra was expanded for more detailed study (Figure 27). From the results it can be seen that as the sample is subjected to longer heating a shift in the F685 occurs towards 680 nm at the same time as 133 LU 0 F 2! 1.1.1 _ . (J , "i I E; ' I 0 I'I ’ :3 - / I I c—l f\ \ I u. \l LU '- 1 V Z r F- \ l \ a \ \ 0C \\. \ \ \ \J \\ _____ ’// 111111111111L11111111111L'IIIIII (480.00 800.00) WAVELENGTH (nm) Figure 26. Low temperature fluorescence (77K) emission spectra of wild type thylakoids incubated at 22 C ( ) and 53 C (---) for 5 min before freezing of the samples. Carbofluor (Fmax 500 nm) was added as an internal standard to which both spectra were normalized. 134 .8: 0mm op E: owe co mommy 05 5 x600 0235 “mos 05. 05 B confluence mums» mean—mom .mmHQEom 05 Eugen“ 96mg AIIIV 0 mm um 52 3.. 00 .m .N ..8 Ale 0 mm um Bug messaged. 0%» UH?» mo in cowmmflfim $2.5 mocwommuooau 03% 3nd .hm madman .EE Eczmswis I. 9 .9 Ir 6 I. C. O O O _ 33013382180019 HAHN—138 .:_E CF .55 m .EE N 135 an increase in F695 is happening. This indicates that indeed there is a separation of LHC from the PSII core and that this is a possible cause of the increased fluorescence emission upon heating leaves. Therefore this system for measuring protein stability in thylakoid membranes should be useful in analysis of Arabidopsis mutants with altered lipid membrane composition. APPENDIX B Chromosome Assignments Because of the difficulty of constructing test crosses for mapping purposes in Arabidopsis, the 11g mutants were assigned to chromosomes by scoring aberrant independent assortment of F2 phenotypes from F1 x F1 crosses. The tgd mutants were crossed to an Arabidopsis strain (MKl) which contains a visible marker for each chromosome. These mutations are listed in Table 2. For assignment to a chromosome each mutation was scored for 3:1 segregation and then 9:3:3:1 independent assortment. Any significant departure from these ratios not due to aberrant segregation was used to assign the fad mutations to chromosomes. The observed and expected frequencies and X2 values are reported below for the fadA and fadg mutations. 0860 (+ + + + + + fadA) x MKI (an py gl cer ms +) + an + 01 + 91 + car + as + fadA obs 120 35 119 39 117 38 126 28 108 35 116 39 exp 116 39 119 40 116 39 115 39 107 36 116 39 x2(1) 0.52 0.08 0.02 3.81 0.05 0.02 136 Chromosome 1 081 (+ + + + + + + an obs 62 24 exp 65 21 x20) 0.39 13 obs exp gy obs exp gl obs exp ggt obs exp mg obs exp * signifies a significant difference + Independent assortment of fa A L_/+ obs exp obs exp obs exp obs exp obs exp 4- /+ 53 48 60 48 49 48 53 46 49 54 75 65 21 6.84 Independent assortment of fadD +/- 9 16 9 16 15 16 14 15 9 15 +/+ 91 87 87 87 89 87 89 87 80 80 + 137 g1 64 21 64 21 0.03 +/- 29 29 31 29 29 29 37 29 28 27 fadfl/+ 20 16 15 16 21 16 12 15 14 15 67 61 27 29 26 29 25 27 £93 2.56 fadD/- #01 (”N 010 ON 01% 14 20 fadA/- 10 10 fadQ) x MKl (an py gl cer ms +) + E§ 58 19 58 19 0.004 x213) 4.87 * 7.25 3.76 2.61 x213) 0.74 0.54 0.18 * 8.96 0.30 + £9.14 65 16 61 20 0.19 138 All the marker and thA mutations segregated as expected for a simple mendelian trait (X2(l)<3.84, p<0.05). This was also seen for independent assorment of the markers except ggt and 1115 which gave a high value. Egdg was therefore assigned to chromosome 4. In the case of fagQ two markers showed aberrant independent assortment. However the p1 marker also showed aberrant segregation from ngQ, therefore, fadQ was assigned to chromosome 3. 139 BIBLIOGRAPHY Anderson JM, JC Waldron, SW Thorne 1978 Chlorophyll-protein complexes of spinach and barley thylakoids. FEBS Lett 92:227-233 Anderson J.M, A. Melis 1983 Localization of different photosystems in separate regions of chloroplast membranes. Proc. Natl. Acad. Sci. USA 80:745-749. Andrews J, K Keegstra 1983 Acyl-CoA synthetase is located in the outer membrane and acyl-CoA thioesterase in the inner membrane of pea chloroplast envelopes. Plant Physiol 72:735-740 Andrews J, B Mudd 1985 Phosphatidylglycerol synthesis in pea chloroplasts. Plant Physiol 79:259-265 Argyroudi-Akoyunoglou JH 1981 Monovalent and divalent cation-induced transformation of the oligomeric to monomeric forms in the pigment-protein complexes of the thylakoid. In G Akoyunglou, ed, Structure and Molecular Organization of the Photosynthetic Apparatus III. Balaban International, Philadelphia, PA, pp 547-558 Argyroudi-Akoyunoglou JH, H Thomou 1981 Separation of thylakoid pigment-protein complexes by SDS sucrose density gradient centrifugation. FEBS Lett 135:177-181 Armond PA, U Schreiber, O Bjorkman 1978 Photosynthetic acclimation to temperature in the desert shrub Larrea divarigata. Plant Physiol 61: 411-415 Atkinson KO, 8 Jensen, AI Kolat, EM Storm, SA Henry, S Fogel 1980 Yeast mutants auxotrophic for choline or ethanolamine. J Bacteriol 141: 558-564 Barber J, K Gounaris 1986 What role does sulfolipid play in the thylakoid membrane? Photosynthesis Res 9:239-250 Barber J., R.C. Ford, R.A.C. Mitchell, P.A. Millner 1984 Chloroplast thylakoid membrane fluidity and its sensitivity to temperature. Planta 161:375-380. Block MA, AJ Dorne, J Joyard, R Douce 1983 The acyl CoA synthetase and the acyl CoA thoiesterase are located respectively on the outer and on the inner membrane of the chloroplast envelope. FEBS Lett 153:377-381 Bose S, RM Mannan 1984 Increased synthesis of Photosystem II in Triticum vulqatg when grown in the presence of BAS 13-338. Z. Naturforsch 39:510-513. Browse JA, CR Slack 1981 Catalase stimulates linolenate desaturase activity in microsomes from developing lineseed. FEBS Lett 131:111-114 140 Browse JA, N Warwick, CR Somerville, CR Slack 1986 Fluxes through the prokaryotic and eukaryotic pathways of lipid synthesis in the 16:3 plant Arabidopsis thaliana. Biochem J 235:25-31 Butler WL 1978 Energy distribution in the photochemical apparatus of photosynthesis. Annu Rev Plant Physiol 29:345-378 Chapman DJ, J DeFelice and J Barber 1983 Influence of winter and summer growth conditions on leaf membrane lipids of Pisum gatiyum L. Planta 157:218-233. Darr S 1985 Structure and function of the light harvesting chlorophyll a/b protein complex: Investigations using reconstitution and monoclonal antibodies Ph.D Thesis Michigan State University Doonan S, E Marra, S Passarella, C Saccone, E Quagliariello 1984 Transport of proteins into mitochondria. Int Rev Cytol 91:141-186 Dubacq J-P, D Orapier, A Tremolieres, J-C Kader 1984 Role of phospholipid transfer protein in the exchange of phospholipids between microsomes and chloroplasts. Plant Cell Physiol 25:1197-1204 Dubacq J-P, A Tremolieres 1983 Occurrence and function of phosphatidyl- glycerol containing 3-trans-hexadecanoic acid in photosynthetic lamellae. Physiol Veg 2:293-312 Douce R, J Joyard 1980 Plant Galactolipids. In PK Stumpf, EE Conn, eds. The Biochemistry of Plants, Vol 4. Academic Press, New York, pp 321-362 Duval JC, A Tremolieres, JP Dubacq 1979 The possible role of trans hexadecenoic acid and phosphatidylglycerol in the light reactions of photosynthesis. FEBS Lett 106:414-418 Ellis RJ 1981 Chloroplast proteins: Synthesis transport and assembly. Annu Rev Plant Physiol 32:111-137 Festke C, G Deichgraber, E Schnepf 1977 Herbicide induced changes in wheat chloroplast ultrastructure and chlor0phyll a/b ratio. Biochem Physiol Pflanzen 171:307-312. Ford RC, DJ Chapman, J Barber, JZ Pedersen, RP Fox 1982 Flourescence polarization and spin-label studies of the fluidity of stromal and granal chloroplast membranes. Biochim Biophy Acta 681:145-151 Frentzen M, W Hares, A Schiburr 1984 Properties of the microsomal glycerol-3-phosphate and monoacylglycerol-3-phosphate acyl transferase from leaves. In PA Siegenthaler, W Eichenberger, eds,Structure, Function and metabolism of Plant Lipids, Elsevier, Amsterdam, pp 105-110 Frentzen M, E Heinz, TA McKeon, PK Stumpf 1983 Specificities and selectivities of glycerol-3-phosphate acyltransferase and monoacyl- 141 glycerol-3-phosphate acyltransferase from pea and spinach chloroplasts. Eur J Biochem 129:629-636 Galey J, B Francke, J Bahl 1980 Ultrastructure and lipid composition of etioplasts in developing dark-grown wheat leaves. Planta 149:433-439 Gardiner SE, PG Roughan, J Browse 1984 Glycerolipid labelling kinetics in isolated intact chloroplasts. Biochem J 224:637-643 Gardiner SE, PG Roughan 1983 Relationship between fatty acid composition of diacylgalactosylglycerol and turnover of chloroplast phosphatidate. Biochem J 210:949-952 Gounaris K, J Barber 1983 Monogalactosyldiacylglycerol: the most abundant polar lipid in nature. Trend Biochem Sci 8:378-381 Grogen 0W, JE Cronan 1986 Characterization of Escherighjg golj mutants completely defective in synthesis of cyclopropane fatty acids. J Bacteriol 166:872-877 Guillot-Salomon T, C Tuquet, N Farineau, J Farineau, M Signol 1982 Lipoprotein associations in chlorophyll-containing complexes associated by non-ionic detergents. In JFGM Wintermans, PJC Kuiper, eds, Biochemistry and Metabolism of Plant Lipids. Elsevier, Amsterdam, pp 373-376 Harwood JL 1980 Plant acyl lipids: Structure, distribution and analysis. In PK Stumpf, EE Conn, eds, The Biochemistry of Plants, Vol. 4. Academic Press, New York, pp 2-48 Haverkate F, LLM van Deenen 1965 Isolation and chemical characterization of phophatidylglycerol from spinach leaves. Biochim Biophys Acta 106:78-92 Hawke, JC, PK Stumpf 1980 The incorporation of oleic and linoleic acids and their desaturation products into glycerolipids of maize leaves. Arch Biochem Biophys 203:296-306 Haworth P, JL Watson, CJ Arntzen 1983 The detection, isolation and characterization of a light-harvesting complex which is specifically associated with photosystem 1. Biochim Bi0phys Acta 724:151-158 Heemskerk JWM, G Bogermann, TJM Peeters, JFGM Wintermans 1984 Spinach chloroplasts: Localization of enzymes involved in galactolipid metabolism In PA Siegenthaler W Eichenberger eds, Structure function and metabolism of plant lipids Vol 9 Elservier Amsterdam pp 119-122 Heinz E, PG Roughan 1983 Similarities and differences in lipid metabolism of chloroplasts isolated from 18:3 and 16:3 plants. Plant Physiol 72:273-279 Henry LEA, JD Mikkelsen, BL Moller 1983 Pigment and acyl lipid composition of photosystem I and II vesicles and of photosynthetic mutants of barley. Carlsberg Res Comm 48:131-148 142 Henry SA 1982 Membrane lipids of yeast : Genetic and biochemical studies In JN Strathern, EW Jones, JR Broach eds, The molecular biology of the yeast Saccharomyces Cold Spring Harbor pp 101-158 Hopper AK, AH Furukawa, HO PHam, NC Martin 1982 Defects in modification of cytoplasmic and mitochondrial transfer RNAs are caused by single nuclear mutations. Cell 28:543-550 Horsch R8, J King 1986 Arsenate counterselective enrichment for auxotrophic plant cells works in theory but not in practice. Can J Bot 63:2115-2120 Izawa S, NE Good 1966 Effects of salt and electron transport on the conformation of isolated chloroplasts. Plant Physiol 41:544-552 Joyard J, R Douce, HP Siebertz, E Heinz 1980 Distribution of radioactive lipids between envelope and thylakoids from chloroplasts labelled in ijg. Eur J Biochem 108:171-176 Kader JC, M Julienne, C Vergnolle 1977 Purification and characterization of a spinach leaf protein capable of transfering phospholipids from liposomes to mitochondria or chloroplasts Eur J Biochem. 139: 411-416 . Khan M, JP Williams 1977 Improved thin layer chromatographic method for the separation of major phospholipids and glycolipids from plant extracts and phosphatidylglycerol and bis(monoacylglycerol) phosphate from animal lipid extracts. J Chromatogr 140:179-185 Keith A0, MR Resnick, AB Black 1969 Fatty acid desaturase mutants of saggharomyggs ggrgvjsiag. J Bacteriol 98:415-420 Kosmac U, J Feierabend 1985 Control of plastidic glycolipid synthesis and its relation to cholorophyll formation. Plant Physiol 79:646-652. Kuhlbrandt W 1984 Three dimensional structure of the light-harvesting chlorophyll a/b protein complex. Nature 307:478-480 Kuiper PJC 1985 Environmental changes and lipid metabolism of higher plants. Physiol Plant 64:118-122 Kyle DJ, TY Kuang, JL Watson, CJ Arntzen 1984 Movement of a subpopulation of the light-harvesting complex LHCII from grana to stroma lamellae as a consequence of its ph05phorylation. Biochim Biophys Acta 765:89-96 Lam E, W Oritz, S Mayfield, R Malkin 1984 Isolation and characterization of a light-harvesting chlorophyll a/b protein complex associated with photosystem 1. Plant Physiol 74:650-655 Leech 8M, RM Leech 1976 Sequential changes in lipids of developing proplastids isolated from green leaves. Plant Physiol. 57:789-794. Leech RM, CA Walton, NR Baker 1985 Some effects of 4-chloro-5- 143 (dimethylamino)-2-phenyl-3(2H)-pyridazinone (SAN9785) on development of thylakoid membranes in Hordeum vulqare L. Planta 165:277-283. Lynch 0V, GA Thompson 1984 Chloroplast phospholipid molecular species alterations during low temperature acclimation in Qunaliglla. Plant Physiol 74:198-203 Lyon JM, JK Raison 1970 Oxidation activity of mitochondria isolated from plant tissue sensitive and resistant to chilling injury. Plant Physiol 45:386-389 Mackender R0 1979 Galactolipids and chlorophyll synthesis and changes in fatty acid composition during the greening of etiolated maize leaf segments of different ages. Plant Sci Lett 16:101-109 Mackinney G 1941 Absorption of light by chlorophyll solutions. J Biol Chem 140:315-322 Markwell JP, JP Thornber, RT Boggs 1979 Higher plant chloroplsts: Evidence that all the chlorophyll exists as chlorophyll-protein complexes: Proc. Natl. Acad. Sci USA 76:1233-1235. Markwell MAK, SM Haas, NE Tolbert, LL Bieber 1981 Protein determination in membrane and lipoprotein samples: manual and automated procedures. Methods in Enzymology 72:296-303. Mazliak P 1977 Glyco- and phospholipids of biomembranes in higher plants. In M Tevini, HK Lichtenthaler, eds, Lipids and Lipid Polymers in Higher Plants. Springer-Verlag, Berlin, pp 49-74 McCourt PJ 1983 A cold sensitive mutant of Arabidopsis MSc. Thesis University of Alberta McKee JWA, JC Hawke 1979 The incorporation of 14C-acetate into the constituent fatty acids of monogalactosyldiglyceride by isolated spinach chlor0plasts. Arch Biochem Biophys 197:322-332 McLaren I, C Wood, MNH Jahil, BSC Yong, DR Thomas 1985 Carnitine acyltransferases in chloroplasts of Bisum sativum (L.) Planta 163:197-200 Moore TS 1982 Biochemistry and biosynthesis of plant acyl lipids In PA Siegenthaler, W Eichenberger eds, Structure function and metabolism of plant lipids 9 Elsevier Amsterdam pp 83-92 Mudd JB 1980 Phospholipid biosynthesis In P Stumpf, E Conn eds, The biochemistry of plants 4 Academic Press New York pp 250-280 Murphy DJ 1982 The importance of non-planar bilayer regions in photosynthetic membranes and their stabilisation by galactolipids. FEBS Lett 150:19-26. 144 Murphy DJ, IE Woodrow, E Latzko, K0 Murkherjee 1983 Solubilisation of oleoyl-CoA thioesterase, oleoyl-CoA:phosphatidylcholine acyltransferase and oleoyl phosphatidylcholine desaturase. FEBS Lett 162:442-446 Nagai J, K Bloch 1968 Enzymatic desaturation of stearyl ACP. J Biol Chem 243:4626-4633 Natsoulis G, F Hilger, G Fink 1986 The HTSl gene encodes both the cyt0plasmic and mitochondial histidine tRNA synthetases of 5. ggrgvjsiag Cell 46:235-243 Nichols BW, P Harris, AT James 1964 The biosynthesis of trans 16:1 hexadecenoic acid by Chlorella vulqaris. Biochem Biophys Res Comm 21:473-479 Norman H, JB St John 1986 Metabolism of unsaturated monogalactosyl- diacylglycerol molecular species in Arabidopsis thaliana reveals different sites and substrates for linolenic acid synthesis. Plant Physiol 81:731-736 Ohlrogge J8, ON Kuhn, PK Stumpf 1979 Subcellular localization of acyl carrier protein in leaf protoplasts of Spinagia glgtgggg. Proc Natl Acad Sci USA 76:1194-1198 Ohlrogge JB, WE Shineg PK Stumpf 1978 Characterization of plant acyl-ACP and acyl CoA hydrolases Arch. Biochem Biophys 189: 382-391 Ohnishi J, M Yamada 1982 Glycerolipid synthesis in Avgng leaves during greening of etiolated seedlings 111. Plant Cell Physiol 23:767-773 Okayasu T, M Nagao, T Ishibashi, Y Imai 1981 Purification and partial characterization of linoleoyl-CoA desaturase from rat liver microsomes. Arch Biochem Biophys 206: 21-28 Papageorgiou G 1975 Chlorophyll fluorescence: An intrinsic probe of photosynthesis. In Govindjee, ed, Bioenergetics of Photosynthesis, Academic Press, New York, pp 319-322 Paterson DR, CJ Arntzen 1982 Detection of altered inhibition of photosystem 11 reactions in herbicide resistant plants. In M Edelman, R Hallick, NH Chua, eds, Methods in Chloroplast Molecular Biology, Elsevier, Amsterdam, pp 109-119 Praje E, B Guiard 1986 One nuclear gene controls the removal of transient pre-sequences from two yeast proteins: One encoded by the nuclear the other by the mitochondrial genome. EMBO 5:1313-1317 Quinn PJ, WP Williams 1983 The structural role of lipids in photosynthetic membranes. Biochim Bi0phys Acta 737:223-266 Raetz CRH 1978 Enzymology, genetics and regulation of membrane phosholipid synthesis in Escherichia gglj Microbiol Rev 42:614-659 145 Raison J 1980 Membrane lipids: structure and function. In PK Stumpf, EE Conn, eds, The Biochemistry of Plants, Vol 4, Academic Press, New York, pp 57-83 Rawyler A, PA Siegenthaler 1981 Transmembrane distribution of phospholipids and their involvement in electron transport as revealed by phospholipase A2 treatment of spinach thylakoids. Biochim Biophys Acta 635:348-368 Restall CJ, P Williams, MP Percival, PJ Quinn, 0 Chapman 1979 The modification of membrane fluidity by hydrogenation processes. III. The hydrogenation of biomembranes of spinach chloroplasts and a study of the effects of this on photosynthetic electron transport. Biochem Biophys Acta 555:119-130. Remy R, A Tremolieres, JC Duval, F Ambard-Brettevile, JP Dubacq 1982 Study of the supramolecular organization of light harvesting chlorophyll protein (LHCP). FEBS Lett 137:271-275 Remy R, A Tremolieres, F Ambard-Bretteville 1984 Formation of oligomeric light harvesting chlorophyll a/b protein by interaction between its monomeric form and liposomes. Photobiochem Photobiophys 7: 267-276 Reynolds ES 1963 The use of lead citrate at high pH as an electron opaque stain in elctron microscopy. J. Cell Biol. 17:208-212. Roughan PG 1975 Phosphatidyl choline: donor of C18 unsaturated fatty acids for glycerolipid synthesis. Lipids 10:609-614 Roughan PG 1985 Cytidine triphosphate-dependent acyl-CoA-independent synthesis of phosphatidylglycerol by chloroplasts isolated from spinach and pea. Biochim Biophys Acta 835:527-532 Roughan PG 1986 Plant Cell Envir in press Roughan PG, CR Slack R Holland 1976 High rates of [14C]acetate incorporation into lipids of isolated spinach chloroplasts. Biochem J 158:593-601 Roughan PG, J8 Mudd, TT McManus, CR Slack 1979 Linoleate and linolenate synthesis by isolated spinach (Spinacea oleracea) chloroplasts. Biochem J 184:571-574 Roughan PG, CR Slack 1982 Cellular organization of glycerolipid metabolism. Annu Rev Plant Physiol 33:97-132 Roughan G, R Slack 1984 Glycerolipid synthesis in leaves. Trend Biochem Sci 383-386 Roughan PG, CR Slack, R Holland 1978 Generation of phospholipid artifacts during extraction of developing soybean seeds with metabolic solvents. Lipids 7:497-503 146 Schreiber 0, JA Berry 1977 Heat induced changes in chlorophyll fluorescence in intact leaves correlated with damage in the photosynthetic apparatus. Planta 136:233-238 Shimakata T, PK Stumpf 1982 Fatty acid synthetase of Spinagja glgragae leaves. Plant Physiol 69:1257-1262 Shimakata T, PK Stumpf 1982 Isolation and function of spinach leaf B-acyl-[acyl-carrier-protein] synthases. Proc Natl Acad Sci USA 79: 5808-5812 Siefermann-Harms 0, JW Ross, KH Kaneshiro, HY Yamamoto 1982 Reconstitution by Monogalactosyldiacylglycerol of energy transfer from light harvesting chlorophyll a/b protein complex to the photosystem in Triton X-100 solubilized thylakoids. FEBS Lett 149:191-196. Simoni RO, RS Criddle, PK Stumpf 1967 J Biol Chem 242:573-581 Singer SJ, GL NIcolson 1972 The fluid mosaic model of the structure of membranes Science 175:720-731 Slack CR, PG Roughan, N Balasingham 1977 Labglling of glycerloipids in the3cotyledons of developing oil seeds by [ Clacetate and [2- H]glycerol. Biochem J 162:289-296 Slack CR, PG Roughan, J Terpstra 1976 Some properties of a microsomal oleate desaturase from leaves. Biochem J 155:71-80 Smolenska G, P Kuiper 1977 Effects of low temperature upon lipid and fatty acid composition of roots and leaves of winter rape plants. Physiol Plant 41:29-35 Somerville CR, WL Ogren 1982 Isolation of photorespiration mutants of Arabidopsis. In M Edelman, R8 Hallick, NH Chua, eds, Methods in Chloroplast Molecular Biology. Elsevier, Amsterdam pp 129-138 Somerville Cr, SC Somerville, WL Ogren 1981 Isolation of photosynthe- ticially active protoplasts and chloroplasts from Aragiggpsjs thaliana. Plant Sci Lett 21:89-96 Sparace SA, J8 Mudd 1982 Phosphatidylglycerol synthesis in spinach chloroplasts: Characterization of the newly synthesized molecule Plant Physiol 701:1260-1264 Spurr AR 1969 A low viscosity epoxy embedding medium for electron microscopy. J Ultrastruct Res 26:31-43. Stubbs C0, A0 Smith 1984 The modification of mammalian poly- unsaturated fatty acid composition in relation to membrane fluidity and function. Biochem Biophys Acta 779:89-137. Staehlin LA, CJ Arntzen 1983 Regulation of chloroplast membrane function: Protein phosphorylation changes the spatial organization of membrane components. J Cell Biol 97:1327-1337 147 Strittmatter P, L Spatz, D Corcoran, MJ Rogers, 8 Setlow, B Redline 1974 Purification and properties of riat liver microsomal stearyl coenzyme A desaturase. Proc Natl Acad Sci USA 71:4565-4569 Stymne S, Applequist L-A 1980 The biosynthesis of linoleate and linolenate in homogenates from developing soya bean cotyledons. Plant Sci Lett 17:287-293 Taylor F, JE Cronan 1976 Selection and properties of Escherichia coli mutants defective in the synthesis of cyclopropane fatty acids. J Bacteriol 125:518-523 Thornber JP, HR Highkins 1974 Composition of the photosynthetic apparatus of normal barley leaves and a mutant lacking chlorophyll b. Eur J Biochem 41:109-116 Tremolieres A, JP Dubacq, F Ambard-Bretteville, R Remy 1981 Lipid composition of chlorophyll protein complexes. FEBS Lett 130:27-31 Unitt MD, JL Harwood 1982 Lipid topography of thylakoid membranes. In JFGM Wintermans and PJC Kuiper, eds, Biochemistry and Metabolism of Plant Lipids, Elsevier, Amsterdam pp 359-362 Vigh L, F Joo, M Dr0ppa, LI Horvath, G Horvath 1985 Modulation of chloroplast membrane lipids by homogeneous catalytic hydrogenation. Eur. J. Biochem 147:477-481. Williams JP 1980 Galactolipid synthesis in yicia faba leaves V. Radistribution of C-labelling in the polar moieties and the C-labelling kinetics of the fatty acids of the molecular species of monoiygnd digalactosyl diacylglycerols. Biochim Biophys Acta 618: 461- Wintermans JFGM, A DeMots 1965 Sepctrophotometric characteristics of chlorophylls a and b and their pheophytins in ethanol. Biochim Biophys Acta 109:448-453