a. Aruba-«A— ' "’3 "3,3,5 . V ,. w” 1:2???593‘ «'3 . "’53“ off“? J " ‘ l E 31%.” r} '8‘. A , 253th .‘v '\. ’)-'.\>§‘ { d’ - _ V e :‘Hfééfifin‘qwflfia. . ‘ WA“ “x? w' ‘ " ' ‘ "7%“ WSW? N“. "3 . , t, , ‘wk ' ..“ |'-:\:E:}‘IV¢‘-n-f 6’ ,$% A.“ ' . 1129.5 ' ‘ h ‘1‘" ‘1‘}. g 2.5:. ,4: r r A A vi“; . ' .. .. k Ag, '1‘ - V ' ‘ aga- . I ‘. ,- '~ “" i .’ J E " '3; i M ‘ ‘ ‘5‘“ I I, ‘ gym“; . . o _ ‘. ' ’ ' ' " liv- ‘ ‘ . . ‘ f . fl ‘ . I ‘1 '71. J; ‘ N ‘ £1 IV A . . V; q .19 , v 3. : ~ not . 5 .5 V ‘ u 1‘; .5 ‘» IV}: 4‘ ‘ I . .‘ . l b y '0 1x ,3; ‘LV . 1": ’3 , -._ so - c‘ . 5. ~=-%93E?r4*€3i?-1 ., v . , ~ 4%. ' $4.1:L1w'1":1$’ L ‘ ‘ ‘ 9A; a,‘ v3... . _. an . 4' ...‘I . a 5%: .2 5, 1.1.0:” 3;; {i 9?: ‘ ' r, 1 ‘ 51' £ .nrx "I ‘f _ ' V. .1 ‘ ' b“ -‘ j» 9‘72“; ,. ~ ‘5 A 7~ iv 1‘" a $79.1 ' \ '4 1" ‘ ‘ ‘ ‘ ' J. “'4‘. . ' . {ma-.1- .r . w: i ‘s ‘EM‘A‘ ' , 3X 3‘? ‘ "‘35" J). . 242$ . 914-7“ :4 I" ‘ A‘ i W! '3 :37, .931; '~ .\ >1 :24? IMF "E’fi V: ’x‘ ‘ L‘Ht ‘ ’.~ ‘5 49". ‘1'! .4. SE ‘5' F f . ~ ~- .. . ”satay: ‘4 r .5“ _.V' ,' '3‘ ". din“, » d ‘ «a . r: "1317. w mint: - 3 S" 3‘}! 1‘ - ‘ Jr I: fié" . .m‘ ‘ ' , . , x J , .. . S. . . x TL ‘ E A“ ‘ 13%;?" £7" 1L— ’5‘ .. ,. We?” 1! 23%;; assEamE "Miam- §‘ 7+-i~“j.~é:{5{'""“ 3; ;.£-'.‘-‘. '~ - ‘»W‘fif “‘17: 4 . . ' " ‘ , Q I; 1351' "f a ‘ V241 . .43: z!- (A ,A 1. I‘v‘,‘.-:f'£; D, It” ¢ 'filfi-fi" I. ‘ If "2 E. g ’31. 1i.“ . 4 _, L'w'f H.u.w , - p7)" .' a“ ~ . '3. {Jr- ,_ ‘ %:5~J~r?:§?é‘ a rm! wrmfivfmjmf‘ rrut.“ 4’} 137'J’APJ’ "A"; ~ , 5,.Av’vj,“ wf‘ff- ~"' 3' f"5 fi'S’gwfifi-Ezfifl’w J ._...»r -". ~ ' -' "' .fi “#5:?! xé‘f“?=fig"z;ma«w " ‘ . -: a“: Vet ’ {1‘35 ‘E‘Wufififi .dw‘l: " ' {fig-n5. . -- kw - .6 i'm 7 ’1 J ”r ’2 ,. "ram *2: v .: . w-‘" ur.1.,.. ' ‘ III . In“ a ,,,.{’- -3-1.’L " .1. . .- ‘...¢pv . ,.- -r:{~'”r“‘g¢¢uwg;, ). $21., :30,” »- S '- ' ..5$'_7,,»w“ ,, . :v'fjgwad‘w. ".5 3' “2;. ' ' 6‘ ’ .- — . .- .Mzr" » ~ “34.2%" rm?" W M -. a 1:". Mr" .. ' '~--f--'-*:»:r"*"a~'-4‘ ”an 73:1. _ .- x. ‘ u M W 4"} 1,..3 .uw ‘ ~' " 1‘ ‘fl ,...5.,.. . "f4. «'24, v... " '17‘I‘3‘»~-‘- ’ w... .~ .. . -n r a w . ' W" "" “,L 414;, «3‘3” ”W‘CTTu-f , , . .f : FE?" .M, «>9 any .A 3-. n 3:," ... -1 ,0 , .. ' ' . ,xw'“ n. E‘; 11—9v;--'::‘-"'-,»"fi'-_~_pr»:- 1' o I35 rm;f~vv;f:..n-”;}W' ‘ ' git; Lye-Irv“ Jun-"‘5“. ' ‘ u“~a;’;fi5 1v: ,"w-“x. - N . J...- - flwavtifie’ “2:: tawny >~r"‘ w «:3. A; ”m "I; ’ f“; . ’3, J, . u ur “a" ‘w' .C-‘V’f‘ (“V J "A J‘ngrv .s’uiwzg, M" J. a 'v4‘ ‘1': - 'flqw. E." - " V .m' '37. "g" ” -"' :‘1' u; .v :73» - ”‘73-: .. 33'" "' . ’ .«n » v , - m!” «0“»~"' - . o’vxwv $c:~-::.=.-« ,8‘ ~-‘ 51S» m ~3 Starfish ' . u r: m .. .w .. “"v'mw :- 4:: .. ""2' ’50-'21!“ 7' "‘7' ‘ ,. .«wr-v $7; Jaz- ‘12 “4' ,. 3:, "MM 3v... ‘ ' -.§ . » ‘I -“ .33; .2d , ' - . ... POO...- - «O'Jl farm ‘5‘" Iiiiiiiiiiiiiiiiiiiiiil liiiiiiiiiii 3 1293 00788 397 MICI‘IEC‘mn «1,13 un:;@r33iy '50“-.- -m‘b- This is to certify that the dissertation entitled CYTOCHROME B5 GENE IN CHICKEN presented by Hong Zhang has been accepted towards fulfillment of the requirements for Ph. D degree in _G_en£._ti£5__ ' a o professor Date Aug- 2; 1989 012771 MS U i: an Affirmative Action/Equal Opportunity Institution NIVERSI SIBTYLI RARIES 2/? sfv ? {257 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE ii 4L; r—i ‘ i i r“? r f .. MSU Is An Affirmative ActlorVEquel Opportunity Institution ckchme-pd CYTOCHROHE 85 GENE IN CHICKEN By Hong Zhang A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements For the degree of DOCTOR OF PHILOSOPHY Genetics Program 1989 ABSTRACT CYTOCHROME 85 GENE IN CHICKEN By Hong Zhang Cytochrome b5 functions as an electron transport carrier in fatty acid desaturation in animal liver, in methemoglobin reduction in erythrocytes, and in cytochrome P-450 reduction. It exists in two forms: an amphipathic form, and a cytosolic form. The amphipathic form consists of a N-terminal hydrophilic domain which contains a functional heme as a catalytic site, and a C-terminal hydrophobic domain which anchors the protein in the microsomal membrane. The cytosolic form is equivalent to the hydrophilic domain of the amphipathic form. The possibility that the cytosolic form is derived from the amphipathic form by proteolytical processing was proposed in the literature. This dissertation describes the isolation and characterization of the chicken cytochrome b5 gene, and indirect support for the above hypothesis. Cytochrome b5 cDNA clones were isolated from chicken liver and erythrocytes by probing cDNA libraries with synthetic oligonucleotides designed from the chicken cytochrome b5 protein. The cDNA clones from liver and erythrocytes showed 100% homology and encoded a protein with an amphipathic form. The Northern analysis indicated only one kind of message present in liver total RNA, and this message is about the same size as the cDNA from erythrocytes. Furthermore, a lambda genomic clone was shown to contain a cytochrome b5 gene with an amphipathic form, and this clone produced the same hybridization pattern as the chicken genomic DNA did in Southern analysis. All the data suggested that there is only one copy of the cytochrome b5 gene in chicken. The presence of one gene excluded the possibility that different genes are reponsible for the two forms of cytochrome b5 protein in chicken. The complete identity of cDNA clones from liver and erythrocytes excluded the possibility that differential RNA splicing is the reason for two kinds of cytochrome b5 from a single gene. Posttranslational modifications appear to be the mechanism for synthesis of the cytosolic cytochrome b5. There may be one or more erythroid proteases which are responsible for the solubilization of amphipathic cytochrome b5 in erythrocytes to give a cytosolic protein. The data presented in this dissertation supports the existence of such proteases in erythrocytes. Copyright by HONG ZHANG 1989 To China Who Is Fighting For Democracy Acknoledgements I would like to thank all members of my guidance committee, Drs Jerry Dodgson, Tom Friedman, Chris Somerville, and Edward E. Talbert, for their continuously generous support and invaluble suggestions during my stay as a graduate student. I would like to thank Dr. Barry Chelm for his contribution in serving as a member of my guidance committee before he left us two years ago. I thank all my friends and colleagues in PRL and Genetics Program; their support made my life so much easier. There is no need to mention a particular person in C. Somerville’s lab for me to say "thank you"; you all made this lab a wonderful place to work and to learn. But, I would like to thank again to my major professor, Chris, for his patience, advices, and everything he provided in guiding me toward my Ph.D. Finally, I would like to thank my grandmother and my parents, their love, support and understanding have always been with me. vi TABLE OF CONTENTS Page LIST OF FIGURES -------------------------------------- i --------- x CHAPTER 1. LITERATURE REVIEW 0N CYTOCHROME B5 ---------------- 1 1. Introduction ------------------------------------------ 1 2. Functions of cytochrome b5 ---------------------------- 2 1) An electron carrier in fatty acid desaturation -------------------------------------- 2 2) Reduction of methemoglobin in erythrocytes -------- 4 3) Reduction of cytochrome P-450 --------------------- 5 3. Properties of cytochrome B5 --------------------------- 8 1) Primary structure of the protein from vertebrates --------------------------------------- 8 2) Plant cytochrome b5 structure --------------------- 9 3) Ternary structure --------------------------------- ll 4. Cytochrome bS-lipid interaction ----------------------- 14 5. Erythrocyte cytochrome b5 ----------------------------- 16 6. Acknowledgments --------------------------------------- 18 7. References -------------------------------------------- 18 CHAPTER II. THE PRIMARY STRUCTURE OF CHICKEN LIVER CYTOCHROME 85 DEDUCED FROM THE DNA SEQUENCE OF A cDNA CLONE ------------------------- 23 vii Abstract -------------------------------------------------- 24 Experimental procedures ----------------------------------- 24 Results and discussion ------------------------------------ 25 Acknowledgments ------------------------------------------- 28 References ------------------------------------------------ 28 CHAPTER III. CYTOCHROME 85 GENE IN CHICKEN ‘ ------------------- 29 Abstract -------------------------------------------------- 29 Introduction ---------------------------------------------- 30 Materials and methods ------------------------------------- 32 Results --------------------------------------------------- 35 Discussion ------------------------------------------------ 42 Acknowledgments ------------------------------------------- 45 References ------------------------------------------------ 45 CHAPTER IV. SEARCHING FOR THE PLANT CYTOCHROME 85 GENES -------------------------------------------- 48 1. Introduction ------------------------------------------ 48 2. Materials and methods ----------l ---------------------- 50 3. Results ----------------------------------------------- 52 1) Chicken liver cytochrome b5 clone as a probe ------ 52 2) Oligonucleotides as probes ------------------------ 59 4. Discussion -------------------------------------------- 63 5. References -------------------------------------------- 67 CHAPTER V. SUMMARY AND SUGGESTIONS --------------------------- 69 APPENDIX I. TRANSFER OF THE MAIZE TRANSPOSABLE ELEMENT Mul INTO ARABIDOPSIS THALIANA ----------------------------------------- 73 Abstract -------------------------------------------------- 74 viii Introduction ---------------------------------------------- 74 Materials and methods ------------------------------------- 75 Results ................................................... 75 Discussion ................................................ 31 Acknowledgments ------------------------------------------- 82 References -r .............................................. 32 APPENDIX II. TRANSFER OF THE Ac ELEMENT INTO ARABIDOPSIS BY SEED TRANSFORMATION -------------- 83 1. Introduction ------------------------------------------ 83 2. Materials and methods --------------------------------- 84 3. Results ----------------------------------------------- 86 1) Transforming vector construction ------------------ 86 2) Aqrobacterium transformation ---------------------- 87 3) Arabidopsis transformation ------------------------ 87 4. Discussion -------------------------------------------- 89 5. References -------------------------------------------- 90 APPENDIX III. SEARCHING FOR THE DESATURASE GENES IN ARABIDOPSIS AND AGMENELLUM --------------------- 91 1. Introduction ------------------------------------------ 91 2. Materials and methods --------------------------------- 94 3. Results ----------------------------------------------- 94 1) Yeast desaturase gene as a probe ------------------ 94 2) Oligonucleotides as probes ------------------------ 98 4. Discussion -------------------------------------------- 101 5. References -------------------------------------------- 102 APPENDIX IV. DOUBLE STRANDED DNA SEQUENCING AS A CHOICE FOR DNA SEQUENCING ----------------------- 103 ix LIST OF FIGURES FIGURE PAGE CHAPTER I l The absorption spectra in a and 6 region of the reduced cytochrome b5 in solution and crystals (3) ------------ l 2 The role of cytochrome b in fatty acid desaturation. Cyt.b R=Cytochrome b5 re uctase. Cyt.bssCytochrome b5. Des-D saturase ---------------------------------------- 3 3 The pathway of oxygenation by cytochrome P-450. Cytochrome b5 was proposed to provide the second electron for the P-450 catalyzed hydroxylation reactions (25) ---------------------------------------- S 4 Model for cytochrome b5 in cytochrome P-450 related reactions (28) ........................................ 5 5 The primary structures of cytochrome b proteins from different animal sources (1). [L] designates a protein from liver and [E] indicates an erythrocyte protein --- 7 6 NH -terminal amino acid sequence of pea cytochrome b5 (13) --------------------------------------------------- 11 7 Schematic diagram of the backbone chain of cytochrome b which was solubilized from liver microsome with p§nccreatic lipase (36) ------------------------------- 10 8 Models for insertion of cytochrome b5 in artificial membranes (38) ----------------------------------------- 15 CHAPTER II 1 The oligonucleotide mixtures used as probes for the cytochrome b5 gene ------------------------------------ 25 2 Composite nucleotide sequence of the cDNA for chicken cytochrome b and the deduced amino acid sequence. The amino acid sgquence is numbered from the alanine residue most commonly found at the amino terminus of the vertebrate protein. The regions of homology to the oligonucleotide probes extend from nucleotides 82 to 98 and 217 to 233 for b5-4 and b5-1, respectively -------- A comparison of the deduced amino acid sequence of chicken cytochrome b with the sequences obtained by direct amino acid seguencing of the microsomal cytochrome b from other vertebrates. The numbers in parentheses give the references for the sequences. The residues which differ from the most consensus sequence are enclosed in boxes --------------------------------- Hybridization of cytochrome b5 cDNA probes to chicken genomic DNA. The genomic DNA was digested with Eco RI (E), Hind 111 (H), and Bgl II (B). [A] The entire cDNA was used as probe. [8] The Hind III fragment encoding the hydrophobic domain was used as a probe ------------ CHAPTER III Oligonucleotides designed from chicken cytochrome b5 protein (bS-l, and b -4), and from chicken liver cytochrome b5 clone b5-7, and b5-8) ------------------- Sequence of a cDNA clone and the deduced amino acid sequence for cytochrome b from erythrocytes. Regions where oligonucleotides reéognize are marked ----------- The Northern Blotting of liver total RNA (lane 1) and erythrocyte total RNA (lane 2). Filter was first hybridized to cytochrome b gene (A), then the filter was washed before it was rghybridized to bete-globin gene (8) ----------------------------------------------- Restriction maps of three genomic clones of cytochrome b . Havy lines designate the fragments which hybridized t3 the indicated oligonucleotides --------------------- Restriction digests of chicken genomic DNA and the 95 clone probed with liver cytochrome b cDNA or oligonucleotides. Lanes A and B are genomic digests, C and D are 95 clone digests. A and C were digested with Hind III, B and D were with Xba I. A, B, and C were probed with cytochrome b5 gene, and D was probed with oligonucleotides bS-l, b -7, and b -8 sequentially. The bands recognized by speciEic oligonacleotides in lane D are marked -------------------------------------- CHAPTER IV Oligonucleotides designed from animal cytochrome b5 protein sequences. N represents T, A, C, and G ------- xi 26 27 27 31 36 38 39 41 49 10 11 12 A comparison of the deduced amino acid sequence of chicken cytochrome b with the sequences obtained by direct amino acid seguencing of the microsomal cytochrome b from other vertebrates. Regions where the oligonucleotides were designed are marked --------- The Eco RI digest of Arabidopsis genomic DNA was probed with chickenoliver cytochrome b cDNA clone. Lane A was washed at 37 C gor the final waéh, and Lane 8 was washed once more at 46 C ------------------------------------- Sequencing strategies for clones ABS-lo-O and ABS-13-0 - DNA Sequence of the clone Ass-13-0 .................... Result of the protein data bank search with reading frame 2 of the clone ABS-13-o ......................... Result of the protein data bank search with reading frame 3 of the clone ABS-13-0 ......................... DNA sequence of the clone ABS-IO-o .................... The Eco RI digests of Arabiggpsis genomic DNA were probed with oligonucleotide mixture b5-2 (lane A), b5-3 (lane 8), and b5-5 (lane C) --------------------------- Sequencing strategy for clone BS-ZFl ------------------ DNA sequences of 85-60 and 85-61. Both are part of the clone BS-ZFI .......................................... The sequence of the 33-amino acid motif from Arabjdgpsis clone ABS-13-0 and comparison to the consensus sequences from lin-12 of nematode, Notch of Drosophila, cdc 10 and SH 16 of yeast (13, 14) -------------------------------- APPENDIX 1 Map of the Mul-containing intermediate vector pHZl -~-- The arrangement of Mul in the cointegrate plasmid pGV3850::pHZ1. (A) The modified T-DNA region of pGV3850 [11] in which most of the T-DNA has been replaced by pBR322 which is joined to pTiC58 Hind III fragments 10 23. (8) Integration of pHZl into pGV3850. The arrows below the figure indicate the size of the corresponding restriction fragments in kbp --------------------------- Southern blot of DNA from wild-type and G-418-resistant tissue probed with Mul DNA. The DNA from all sources was restricted with Eco RI. The position and size (kbp) of molecular weight markers is indicated at the left xii 51 52 53 54 56 57 58 6O 60 62 64 76 77 side of the figure. Lanes: (A) DNA from eight G418- resistant calli; (B) DNA from leaf tissue of an R3 plant descended from a G418 resistant callus; (C) wild- type A. thaliana; (D) A mixture of equivalent amounts of DNA from A. tumefaciens C58Cl (pGV3850zzMu1) and the DNA preparation in lane 8 ------------------------------ Southern blot of DNA from transformed plants of A; thaliana probed with Mul DNA. The position and size (kbp) of molecular weight markers are indicated at the left side of the figure. Lanes: (A) wild-type A. thaliana digested with Eco RI; (8, C and D) DNA from 30 R2 plants digested with Nco I, Hind III, and Eco RI, respectively; (E, F, and G) DNA from one R2 plant digested with Nco I, Hind III and Eco RI, respectively - Southern blot of DNA from a variegated Mul-containing plant. Lanes A to E were digested with Eco RI. Lanes F to I were digested with Hind III. (A) pMJ9; (B and F) DNA from a single variegated R4 plant; (C, D, G and H) DNA from single non-variegated R4 plants; (E and I) wild-type A. thaliana ---------------------------------- APPENDIX 11 The construction of the transforming vector pBIN19::Ac Restriction pattern of two selected transforming Aqrobacterig. Strain l (A, C, E and G) and strain 2 (B, D, F and H) were digested with enzymes Pst I (A and B), Sal I (C and D), Hind III (E and F), and Bgl I(G and H). Only strain 2 gave the expected pattern ------- An example of some resistant T2 seedlings on Kanamycin plates ................................................ Restriction pattern of one putative transformant. Lane A, Ac element alone. Lanes 8 and C, wild-type Arabidopsis HS race and Columbia race cut with Eco RI. Lanes D and E, the putative transformant cut with Eco RI and Hind 111 respectively ------------------------------ APPENDIX III Comparison of the deduced amino acid sequence of yeast stearoyl-CDA desaturase gene with that of the rat stearoyl-CDA desaturase gene. Upper strand is yeast sequence ----------------------------------------------- Arabidopsis genomic DNA digests were probed with p433 (A) and p403 (8) both of which contain part of the yeast stearoyl-CDA gene. Lanes 1, 2, 5 and 6 are Columbia race; lanes 3, 4, 7 and 8 are HS race. Lanes 1, 3, 5 xiii 78 79 80 86 87 88 89 92 and 7 are Eco RI digests,; Lanes 2, 4, 6 and 8 are Bam HI digests ----------------------------------------- Oligonucleotides designed from the conserved regions of the stearoyl-CDA desaturase ........................... DNA sequence of the clone 02-62 ....................... Sequencing strategies for clone A1-5 ................... DNA sequences of A1-5-0 and A1-5-44. Both are parts of the clone Al-S ........................................ Genomic blotting of Agmenellum and yeast by oligonucleotide mixtures DIA, DlT, DZ, and 03 (A, B, C and D). Lanes 1, 3, 5 and 7 are yeast DNA; Lanes 2, 4, 6 and 8 are Agmenellum DNA. All DNA were digested with Eco R1 to completion (2 ug per lane) ------------------- Sequencing strategy for fragment 84-43 ----------------- DNA sequence of the fragment 84-43 -------------------- APPENDIX IV An example of double stranded DNA sequencing method ---- xiv 93 93 95 95 97 99 100 100 104 CHAPTER I LITERATURE REVIEW ON CYTOCHROME 85 1. Introduction Cytochrome b5 is a protohemoprotein which is present in high amounts in the microsomes of animal liver cells. In the reduced state, it shows an asymmetrical o-absorption band with a peak at 556 nm and a shoulder around 560 nm. See figure 1 below. 0.04 0.02 Crystals Absorboncy 500 520 540 560 580 600 Wavelength (nm) Figure l. The absorption spectra in a and Bregion of the reduced cytochrome b5 in solution and crystals (3). This cytochrome is bound to the membrane and reduced with NADH by a flavoprotein (cytochrome b5 reductase) which is also bound to the membrane. The primary structures and ternary structure of cytochrome b5 from several animal sources have been determined (1,2). Hagihara et al. (3) suggested that use of the term cytochrome b5 in referring to hemoproteins of biological materials other than liver microsomes, based simply on similarity in the wavelength of the a peak, may not be desirable unless: a, there is similarity of the low temperature spectrum in the reduced state (77°K or lower); b, there is similarity in amino acid sequence or immunochemical similarity; c, a similar reactivity to the microsomal cytochrome b5 reductase is shown. Besides the microsomes of animal liver, cytochrome b5 is contained in the outer membranes of mitochondria of the same tissue (4,5). A similar cytochrome contained in erythrocytes has been purified and sequenced (1, 6-8). Spectrally similar pigments are also found in yeast (9) and plants (10, 49, 50). 2. Functions of cytochrome b5 1). An electron carrier in fatty acid desaturation Studies of an in vitro system employing microsomal membranes from animal liver capable of desaturating fatty acids showed that the overall reaction had a requirement for oxygen and NADH (11). The first studies examined the requirements for the introduction of the .A9 double bond by the microsomal fraction of rat liver, using stearic and palmitic acids as substrates. The same system was shown to be able to introduce a 116 double bond into oleic acid to form 18:2[56’9 (12). Oshino et al. suggested that the desaturation was associated with the microsomal electron transport chain and possibly involved cytochrome b5, not cytochrome P-450, because cyanide inhibited the desaturation, whereas carbon monoxide did not (13). The first definitive report implicating the NADH-specific microsomal electron transport chain showed that the rate of reoxidation of reduced cytochrome b5 was increased by stearoyl-CDA (51). The absolute requirement for cytochrome b5 was shown by removing endogenous cytochrome b5 from detergent-solubilized microsomes and observing the restoration of desaturase activity upon addition of the purified cytochrome b5 (14-16). The successful in vitro reconstitution of lipid desaturation was done by adding the purified cytochrome b5 reductase, cytochrome b5, and stearoyl-CDA desaturase together plus substrates in an artificial membrane (17, 18). The involvement of these microsomal electron transport components in other desaturase reactions has been demonstrated by the inhibition of the particular desaturase reaction by antibodies raised to the purified cytochrome b5. These results have led to the scheme in figure 2 below. NADH Cyt.bsRox Cyt.bsred Des H 0 + 18:1 CoA NAD Cyt.b R Cyt.bsox Des O2 5 red + 18:0 CoA red Figure 2. The role of cytochrome b in fatty acid desaturation. Cyt.b R=Cytochrome 3 reductase. Cyt.b5=Cytochrome b5. Des=Desaturasg. 2). Reduction of methemoglobin in erythrocytes The methemoglobin reduction system of red blood cells catalyzes the reduction of the four ferric ions of methemoglobin to ferrous ions (21). This reduction proceeds at a rate that is sufficient to maintain approximately 99% of the hemoglobin in its ferrous state, despite the continuous conversion of hemoglobin to methemoglobin by various oxidants of the cells. Under normal conditions, most of the methemoglobin reduction can be attributed to catalysis by an NADH-utilizing system, an NADH-dependent reductase. Since the rate of methemoglobin reduction catalyzed by purified reductase was slow relative to the rate observed in intact cells, the existence of a second component of the system was suggested. Also, there is no correlation between the rate of methemoglobin reduction in intact cells and the amount of NADH-specific reductase that can be detected in these cells (20). A soluble cytochrome b5 in erythrocytes markedly stimulates the catalysis of methemoglobin reduction by the reductase (21). At concentrations present in erythrocytes, cytochrome b5 serves as an effective substrate for erythrocyte NADH-reductase, and the resulting ferrocytochrome b5 then transfers an electron to methemoglobin as follows: reductase 1/2 NADH + Cyt.b5 (Fe3*) ---------- > 1/2 NADT + Cyt.b5 (FeZT) 4 Cyt.b5 (Fe2+) + Hemoglobin (Fe3+) ---------- > 4 Cyt.bS (Fe3+) + Hemoglobin (Fe2+) The name erythrocyte cytochrome b5 arises from the spectral and structural similarity of the protein to microsomal cytochrome b5 (22, 23). The reductase has been termed erythrocyte cytochrome b5 reductase because it acts upon erythrocyte cytochrome b5 and because it is enzymically similar to microsomal cytochrome b5 reductase (24). 3). Reduction of cytochrome P-450 The scheme in figure 3 has been proposed for the mechanism of cytochrome P-450 in hydroxylation reactions (25). RON RH Fe“ 9 m ‘ 'I 3’ iROH FeBO- ‘RH » Fe 8 fl» 9 R ‘ Fe~OH 3’ LRH Fe? is o 02 RH Fe-Of" 2, 2H‘ RH Fe3tqo; e ZRHIFeJ’OST Figure 3. The pathway of oxygenation by cytochrome P-450. Cytochrome b was proposed to provide the second electron' for the P-453 catalyzed hydroxylation reactions (25). 6 As indicated in the scheme, substrate binding to native, ferric P-450 is followed by reduction to the ferrous state, thereby allowing oxygen binding. A second reduction results in spliting of the oxygen-oxygen bond, one atom being lost as water. The other oxygen atom, presumably now an "activated oxygen," is inserted into a carbon-hydrogen bond of the substrate to produce the corresponding alcohol, which is then released with regeneration of the ferric form of the enzyme and completion of the catalytic cycle. Hildebrandt and Estabrook (26) suggested that cytochrome b5 may supply the second electron. Miki et al. purified a form of cytochrome P-450 with a high affinity for cytochrome b5 and showed that reduction of the P-450 by NADH required NADH-cytochrome b5 reductase, cytochrome b5, and suitable concentrations of detergents (27). More recently, Pompon and Coon proposed a new model for the involvement of cytochrome b5 in the P-450 related reactions (28) shown in figure 4. ( 7‘93 +02 (3) II ll Fe L Fe 02 —> [Fe02]H——> —> Fem- Vial-(202 (b) 111 2 Fe ‘ H202 (c) / [FeO I a) I' 2 P 3 III II III ‘ -. L 95 95 RH Fe FROH H20 (0) Figure 4. Model for cytochrome b5 in cytochrome P-450 related reactions (28). (I) 50 £0 30 S-K—S-T-H-L-I—L-R-H-K-V-Y-D-L-T-K-?-L-£-B-B-P-G-G-8-l-V-L- lulln(L) S~K-S-T-H—L-I-L-fi-H-K-V-Y-D-L-T-K-E-L-l-E-B-P-G—G—E-S-V—L— 20 A-v-k-r-Y-t-L-éia-:-o-k-n-N A-V-K-Y-Y-T-L-E-E-I-Q-K-H-N 10 iii A§;A;§:Q:§;Q-K-A-V~K~Y-Y-T-L-E-E-I-Q-K-R-N—N-S-K-S-T-H-L-I-L-H-H~K~V-Y-D—L-I-K-E-L-K~E-fl-P-Gncnl-S-V-L- Io:ctne(L) 1 AS:A:E:Q'5' Ac-A-E-Q-S- (I) K-V-Y-D-L-T-K-F-L-E-E-H-P-G-G-E-E-V-L- lovine(L) .(l) k-v-r-D-L-r-k-r-L-z-e-u-r-c-c-E-e-v-L- iii S-K-S-T-H-L-I-L-H-H-K-V-Y-D-L-T-K-F-L-l-E-H-P-G-G—E-B-V-L- Horle(b) S-K-S-T-H4EPIoL-H-H-K-V-Y-D-L-T-K-E-L-§-E-H-P-G~G-E-E-V—L- aac(t) s-k-s-r-u-L-I-L-u-u-x-v-r4!}L-r-x-r-t-a-s-a-r-c—c-s-:-v-L- Iabb1t(L) E-E-I-Q-K-H-N-N-S-K-S-T-H-L-I-L-H 4E k-HiEHD I v-x-r-t-r-L4§}E-t k-H-NjZi k-A-v-k-r-r-r-L-E1QIt4:}x-N-N-N-s-k-s-r-N-L-t-L-n V-K-Y-Y-T-L-E-E-I k-A-v-k-r-t-r-L- x-Kifi-{V-K-Y-Y-T-L-E-E—I-{Tq-K-H-N Ac-A-E-Q-s-D-K-A-v-k-r-r-r-L-E-E-I-o-k-H-N-N-s-k-s-r-wot-I-L-u-u-k-v-r-D-L-r-k-r-L-a-z-u-r-c-c-s-s-v-L- tdllllIlB-D-K x(A,Z,Z)S Ac-A-E-Q-s-D-k (I) lulIn(L) 100 J. .w TLMU . In: P? .- KK .N .- In 0.. 93 .. D. .. K .. PP .- “a .. LL .- ”mum mwr.r. ._ T.r; .. all. g. It .- KK .- .3.) iii r-a-D-v-c-N-s-r-D-A-a-e-L-s-x-r-r-t-t-c-z-L-u-P-D-D-n-s-x-t-A-x-r-s-e-t-L-t-r~ rotesnn(t) 7O 60 r n-s-o-A-c-c-D-A-t-E-N-r-E-D-v-c-a-s-r-D-A-a-E R-E-Q-A-G—G-D—A-I-E-N-F-E-D-V-C-H-S-T-D-A-R-E R-E-Q-A-G-G—D—A-T-E-N (I) (I) L-a-P-D-D-n-s-k-t-aj-x-r-s-Easl R-E-Q-A-C-C-D-A-T*E~N-F-E-D4IFG~H-S-T‘D-A-R-E-L-S-K-T-F-I-I-C-E-L-fl-P-D-D-R-S-K-I-A-K-P42FE-T-L-I-T- HotIOIL) -D-v-c-a-s-r-D-A-a-E-L-s-k-r-r-t-I-c-E-L-H-P-D-D-a-s-k-t-A-k-r-s-E-Efl R-E-Q-A-Coc-D-A-T-E-N-F-E-D-V-C—R-S-T—D-A-R-E-L-S-K-T4XPI-I-G-E-L-H-P-D—D-R-S-K k-e-Q-A-c-c—D.A-r-E4EIE-E-D-v—c-H-s-r-D-A-a-E-L-s-k-t-r-t-I-c-E-L-H-r-D-D-a-s-k-I4TLx-r-s-EAEZILI-r- Bovine(L) F-E l-E-Q-A-G-G-D—A-T-E~N-F-E-D-V-G-H-S-T-D—A-R-B-L—S-K—T-F-I-I-C-E- R-E-Q-A-C-G-D-A-T-E-N- -z-r-L-t-r- Rat(L) -P4EFE-T-L-I-T- labbit(L) K-P-S 251i: R-E-Q-A-G-G-D-A-T-E-N-F-E-D-V-G-H-S-I-D-A-R-I-L-S-K-T-F-l-I-G-E-L-H-P-D-D—l—S-K nu ) ) (L l- .l\ ( n. \It (on La. C1 (5 u v c_m p:u .u N x . 4 . E... J“ A. .. II 130 TIRE-H L-Y-T 120 r-vIE}s-u-s-s-u-u-r-N-u-v-t-P-A-t-s-A-L-v-v4§FL-H-Y 110 r-v-D-s-N-s-s-u-u-r-N-u-v-t-P-A-I-s-A4ELV-v-A-L-u-r-R4IPr-r-A-a4il Horec(L) t-v-D-s-N-s-s—u-u-r-N—u-v-t-p—A-t-s-A-L4I}v-AoL-n-r-s-L-r T-VJEFS~N-S~5~U~U~T~N~N~V-I~P~A-I~5-A~L-V~V—A~L~H~Y~R~L-Y r41}D-S-N4EL5-H-u-r-N-DII}t-?-A-t-s-A-L4E}v-A-L4ILY proteins a protein [L] designate? from liver and [E] indicates an erythrocyte protein. The primary structures of cytochrome b from different animal sources (1). Figure 5. On the basis of this model, competition between spontaneous decomposition of the ferrous dioxygen intermediate and its reduction by cytochrome D5 is believed to contribute to the partition between abortive hydrogen peroxide production and substrate hydroxylation in this enzyme system. 3. Properties of cytochrome b5 1). Primary structure of the protein from vertebrates The microsomal cytochrome b5 in its native state is an amphipathic protein with an Mr of 16,000. It contains two domains: a N-hydrophilic catalytic segment consisting of about 80 amino acid residues, and a C-hydrophobic segment that is required for binding the cytochrome b5 to the microsomal membrane (29, 30). Controlled proteolytic digestion of the native protein yields a water soluble cytochrome b5 with a Mr of 11,000. The sequence of this soluble cytochrome b5 has been determined (31) and is very similar to that of cytochrome b5 found in the supernatant fraction of erythrocytes. The complete microsomal cytochrome b5 sequences have been determined from 6 different animal species (1). Whereas the microsomal polypeptide is 133 amino acids long, the erythrocyte b5 is 97 residues long (1). Figure 5 shows a comparison of some known cytochrome b5 sequences (see the opposing page). The sequence homology between the various forms of cytochrome b5 is very striking. The sequence of residues 104-126 contains only hydrophobic or uncharged hydrophilic side chains. The basic and 9 acidic amino acid residues occur at the COOH-terminus and the peptide linking the membranous segment to the globular heme-carrying segment. Such residues may be expected to be present outside the hydrophobic milieu of the lipid membrane, thereby suggesting that the hydrophobic segment either penetrates the membrane or folds back on itself so that the COOH-terminus is near the cytoplasmic surface of the membrane. Although the soluble cytochrome b5 accepts electrons from cytochrome b5 reductase (33), the complete cytochrome b5, including the membranous segment, is necessary for functional reconstruction of the stearyl-CDA desaturase system (17, 18). The spectral properties of complete cytochrome b5 are essentially the same as those of the soluble form (heme peptide segment) (34). 2) Plant cytochrome b5 structure Microsomal cytochrome b5 has also been discovered and characterized in plants (10, 49, 50). Bonnerot et al. first purified cytochrome b5 from potato tubers by 350 fold and this protein is very similar to animal cytochrome b5 in terms of its Mr (16700) and its absorption spectrum (49). Later, Madyastha et al. reported the purification of a very similar cytochrome protein to 30% homogeneity from Cgtngrgntngs Lgsggs, and this protein has a Mr of 16500 (50). Jollie et al. (10) purified the microsomal cytochrome b5 from Eisgm ggtiygm, and sequenced the N-terminal part of this cytochrome b5 (figure 6). There is no similarity between this sequence and any animal cytochrome b5 protein. However, they presented results which indicate that the antibody raised against rat cytochrome b5 10 Figure 7. Schematic diagram of the backbone chain of cytochrome b which was solubilized from liver microsome with Banccreatic lipase (36). ll recognized the pea cytochrome b5 protein on western blots. This suggests that the conservation of this protein extends into the plant kingdom, at least at the epitope level. NH -Ala-Leu-Leu-Gln-Glu-Asp-Glu-Ala-Ile-Asp—Asp 2 -Phe-Asp-Phe-Asp—Asp-Gly-Ala-Lys-Asp—Asp-Asp-Gly Figure 6. NHz-terminal amino acid sequence of pea cytochrome b5 (10). Even though the protein was purified and partially sequenced, its function in plants is still not known. Possibly it is involved in reactions like those in animals, since the microsomal electron transport systems of higher plants are involved in a variety of central metabolic transformations including fatty acid desaturation and the mixed function oxidase activities of the cytochrome P-450 dependent monooxygenases. Microsomal cytochrome b5 is certainly a good candidate for a member of the electron transport system. 3) Ternary structure The X-ray crystallographic studies of calf liver cytochrome b5 solubilized by pancreatic lipase (with 93 amino acid residues) at 2.8 A and 2.0 A were carried out by Mathews et al (2, 35, 36). The ternary structure based on their analysis is presented in figure 7 on the opposing page. Like many other proteins, the interior of the molecule is distinctly nonpolar. The heme is buried in a hydrophobic crevice 12 with two vinyl groups lying deep in the interior of the molecule. One of the propionic acid groups lies on the surface of the molecule and is hydrogen-bonded to the ‘v-nygen and the peptide nitrogen of Ser-64, and the other projects outward into solution. The walls of the heme crevice are formed by two pairs of roughly antiparallel helices and the floor by the pleated sheet structure. The iron atom is coordinated by His-39 and His—63 which extend from the wall of the crevice. The nitrogens of the two histidines are hydrogen-bonded to the main chain carbonyl oxygens of Gly-4l and Phe-58, respectively. Furthermore, His-39 is in van der Haals contact with Leu-46 and His-63 lies close and paralled to Phe-58, suggesting a 1r -1r interaction between the latter pair of residues. Thus, the histidine residues are held firmly in place by the rigidity of the backbone structure and by a variety of interactions with the main and side chains. The core part (residues 3 to 86) contains the heme group at the top, lying in the hydrophobic crevice, and also has a narrow hydrophobic group open to the adueous environment. The residues principally involved in this latter group are Phe-35, Leu-70, and Phe-74. The site of the action of the NADH-cytochrome b5 reductase may be this group, and the two phenylalanines in the vicinity of the group may provide a path for an electron to the heme. von Bodman et al (44) chemically synthesized a gene coding for rat liver cytochrome b5 and expressed it in Escherichia coli. Transformants containing the soluble core of cytochrome b5 produced holoprotein containing the protoporphyrin IX prosthetic group in amounts up to 8% of the total cellular protein. The complete cytochrome bs gene including the membrane anchor domain was also 13 efficiently expressed in E; 5911 with incorporation of the holoprotein into the membrane fraction of the cell. The successful expression of cytochrome b5 in E; 9911 means that it is possible to construct mutant cytochrome b5 forms with alterations at specifically selected amino acids within the product protein. von Bodman et al. (44) replaced histidine-63 with alanine by cassette mutagenesis. The resulting protein failed to incorporate heme during fermentative growth and was not reconstituted with exogenous heme after purification of the apoprotein. Mutant cytochrome b5 protein with a methionine substituted at position 63 resulted in the production of the apo form of the cytochrome in high yield. Purification of this apoprotein following the identical procedure for the wild-type holoprotein allowed reconstitution with heme to form the intact mutant protein. This methionine-63 cytochrome b5 displayed an axial high spin ESR signal (9-6) and optical spectra in the ferric form, which was interpreted as evidence that the methionine sulfur was not bonded to the heme iron. Consistent with this interpretation, the reduced protein was found to readily bind carbon monoxide with a 420 nm Soret maxima similar to that observed for myoglobin and hemoglobin. It appears that the methionine-63 protein is a state five-coordinate heme protein. It will be very informative to see the results after the surface charge distributions, the composition of the hydrophobic membrane anchor domain, and the residues that potentially control the redox potential and electron transfer rate have been altered. l4 4. Cytochrome b5-lipid interaction There are two distinct mechanisms for the integration of de nova-synthesized polypeptides into cell membranes. One is specified by an "insertion" sequence and proceeds unassisted into any exposed cell membrane, resulting in the anchorage of a hairpin-loop domain of the polypeptide chain into the lipid bilayer; such a hairpin loop could easily extend into the hydrophilic milieu on the other side of the membrane. The other one is mediated by a "signal" sequence and is dependent on a signal-specific receptor that effects the translocation of a domain of the polypeptide from the biosynthetic compartment to the other side of a specific cell membrane. The membrane bound cytochrome b5 is first synthesized on free ribosomes (45), then bound to microsomal membranes without being recognized by any receptors (37). ' Using the binding of the cytochrome 65 to artificial phospholipid vesicles as a model system, Enoch et al (38) found two types of protein binding: one was capable of intermembrane transfer, the other was not. Based on these properties, they proposed a model for two different orientations of the protein in the membrane ( figure 8). 15 name pepr Id. “0.09. peonde transferable form nonpoior peptide nontransferable form Figure 8. Models for insertion of cytochrome b in artificial membranes (38). 5 The ability of cytochrome b5 to transfer from artificial membranes to other membranes, but not from biological membranes, may reflect a difference in the nature of the protein binding to the membrane. A nontransferable form of cytochrome b5, which may represent the microsomal type of binding, was obtained when cytochrome b5 was bound during the formation of phosphatidylcholine vesicles. A soluble, heme peptide fragment of cytochrome b5 was released when vesicles containing cytochrome b5 in the transferable form were incubated with carboxypeptidase Y. In contrast, the nontransferable form of cytochrome D5 in microsomes and artificial vesicles was not released by carboxypeptidase Y treatment (39, 40). When cytochrome b5 binds to pure, unperturbed bilayers, the loose binding form is predominately obtained. However, if the bilayer is in a perturbed state due to the presence of deoxycholate or another integral membrane protein (i.e., desaturase) in the bilayer, then 16 icytochrome 65 is inserted in the tight binding form. On the basis of their studies of cytochrome b5, Enoch et al. concluded that integral membrane proteins in general do not readily undergo intermembrane transfer between biological membranes (38). As to the mechanism and topology of the interaction between the hydrophobic domain of cytochrome b5 and the membrane, we know very little. The results of a predictive analysis for conformational features, according to the rules of Chou and Fasman (41,42), are similar for the amino acid sequence of the membranous segment from equine and bovine proteins. The sequence from 103 to 112, which contains a cluster of three tryptophanyl residues, seems to consist of 3-4 overlapping IS-turns. However, Jagow and Sebald stressed that the prediction must be viewed with caution, because the conformational parameters employed were derived from studies on globular hydrophilic proteins and may not be necessarily extendable to membranous peptides (43). 5. Erythrocyte cytochrome b5 0n the basis of studies done with bovine erythrocyte cytochrome b5, Hultquist et al. (32) proposed that the soluble erythrocyte cytochrome 65 is derived during erythropoiesis by proteolytic cleavage of the membrane-bound cytochrome b5 present in the endoplasmic reticulum of the proerythroblasts. Bovine erythrocyte cytochrome b5 is indistinguishable from protease-solubilized liver microsomal cytochrome b5 on the basis of spectral properties (22) and ability to react with other redox proteins (24). Cytosolic 17 cytochrome b5 is not present in an immature erythroid cell, but instead, a membranous form of the cytochrome b5 is present (45). An electron microscope study has shown that the endoplasmic reticulum disappears during erythroid maturation (32). There are reports of the existence of stromal proteases which are activated after hemolysis (46, 47). It is possible that a particular class of proteases convert membrane-bound cytochrome b5 into cytosolic cytochrome b5 during erythroid maturation. Since bovine liver lysosomal proteases can digest microsomal cytochrome b5 to produce hydrophilic segments which correspond to erythrocyte cytochrome b5 in vitro, these proteases can serve as a good model for the putative erythroid proteases which solubilize microsomal cytochrome b5 during erythroid maturation. Comparison of the cytochrome b5 sequences of both erythrocyte and liver forms in species such as bovine support the hypothesis of Hultquist and coworkers (32), because the sequence of erythrocyte cytochrome b5 is identical to liver cytochrome b5 from residue 1 to 97. The problem is that residue 97 is proline for human erythrocyte cytochrome b5 and serine for the porcine protein, while residues 97 for human and porcine liver cytochrome b5 are threonine. Three possibilities exist to explain the above problem. 1. There are two or more cytochrome b5 genes in those species, and the cytosolic cytochrome b5 and microsomal cytochrome D5 are encoded by two different, but closely related genes; 2. There is only one cytochrome b5 gene which gives rise to more than one form of cytochrome b5 protein by an alternative RNA splicing mechanism. It has been shown in mouse that four forms of myelin basic protein are 18 encoded by a single gene, and all four mRNAs are produced through an alternative splicing mechanism (48). 3. There is only one cytochrome b5 gene, but different cytochrome b5 proteins are due to posttranslational modifications. Proteolytic cleavage of the membrane-bound cytochrome DS to produce the cytosolic cytochrome b5 in erythrocyte cells can explain the bovine case. In human and porcine, there may be one more modification after the proteolytic cleavage, the addition of one amino acid to the C-terminal of the proteolytically processed protein. I am not aware of any precedents for the third possibility. The investigation of this problem represents one of the main goals of the work described in this study. 6. Acknowledgments I thank the authors and publishers for letting me use their figures in this chapter (figures 3 to 8 from 25, 28, 1, 10, 36, and 38). 7. References 1). K. Abe, S. Kimura, R. Kizawa, F. K. Anan, Y. Sugita (1985). J. Biochem. (Tokyo) 97:1659-1668 2). F. S. Mathews, P. Argos, and M. Levine (1972). Cold Spring Harbor Symp. Quant. Biol. 36:387-393 3). B. Hagihara, N. Sato, and T. Yamanaka (1975). The Enzymes. 11:549-593 4). 5). 6). 7). 8). 9). 10). 11). 12). l3). 14). 15). 15). 17). 19 1. Raw, N. Petragnani, and O. C. Nogueira (1960). J. Biol. Chem. 235:1517-1520 0. F. Parsons, G. R. Willians, W. Thompson, 0. F. Wilson, and B. Chance (1967). Mitochondrial Structure and Compartmentation (E. Quagliariella et al., eds.) p.5. Adriatice Editrice, Bari. D. E. Hultquist, D. W. Reed, P. G. Passon and W. E. Andrews (1971). Biochem. Biophys. Acta 229: 33-41 D. E. Hultquist, R. T. Dean, and R. H. Douglas (1974). Biochem. Biophys. Res. Commun. 60:28-34 0. A. Schafer, and D. E. Hultquist (1983). Biochem. Biophys. Res. Commun. 115:807-813 Y. Yoshida, Y. Tamura-Higashimaki, and R. Sata (1983). Arch. Biochem. Biophys. 220:467—476 0 O. R. Jallie, S. G. Sligar, and M. Schuler (1987). Plant Physiol. 8:457-462 J. B. Marsh, and A. T. James (1962). Biochem. Biophys. Acta 60:320-328 P. W. Hallway, R. Peluffa, and S. J. Hakil (1963). Biochem. Biophys. Res. Commun. 12:300-304 N. Oshino, Y. Imai, and R. Sata (1966). Biochem. Biophys. Acta 128:13-28 P. W. Hallow (1971). Biochemistry. 10:1556-1560 P. W. Hollow, and J. T. Wakil (1972). Biochemistry 11:3689-3696 T. Shimakata, K. Mihara, and R. Sata (1972). J. Biochem. (Tokyo) 72:1163-1174 P. Strittmatter, L. Spatz, 0. Rogers, M. J. Setlaw, and R. Redline (1974). Prac. Natl. Acad. Sci. USA 71:4565-4569 18) 19). 20). 21). 22). 23) 24). 25). 26). 27). 28). 29). 30). 20 M. R. Prasad, and V. C. Jashi (1979). J. Biol. Chem. 254:6362-6369 0. Hultquist (1979). Methods Enzymal. 52:463-473 Y. Kanazawa, M. Hattari, K. Kasaka, and K. Nakaa (1968). Clin. Chim. Acta 19:524 0. Hultquist, and P. G. Passon (1971). Nature New Biol. 229:252-254 P. G. Passon, D. W. Reed, 0. W. Reed, and D. E. Hultquist (1972). Biochem. Biophys. Acta 275:51-61 D. E. Hultquist, R. H. Daulas, and R. T. Dean (1975). In "Erythrocyte Structure and Function” (G. Brewer, ed.), p.297. Alan R. Liss, Inc., New York. P. G. Passon, and D. E. Hultquist (1972). Biochem. Biophys. Acta 275:62-73 R. E. White, and M. J. Coon (1980). Ann. Rev. Biochem. 49:315-356 A. Hildebrandt, R. W. Estabrook (1971). Arch. Biochem. Biophys. 143:66-79 N. Miki, T. Sugiyama, and T. Yamana (1980). In "Micrasames, Drug Oxidations, and Chemical Carcinogenesis", ed. M. J. Coon, A. H. Canney, R. W. Estabrook, H. Gelbain, J. R. Gillette, and P. J. O’Brien, pp. 469-478. New York: Academic. D. Pompon, and M. J. Caan (1984). J. Biol. Chem. 259:15377-15385 P. Strittmatter, M. J. Rogers, and L. Spatz (1972). J. Biol. Chem. 247:7188-7194 L. Spatz, and P. Strittmatter (1971). Prac. Natl. Acad. Sci. 31). 32). 33). 34). 35). 36). 37). 38). 39). 40). 41). 42). 43). 44). 45). 21 68:1042-1046 F.G. Nabrega, and a. Ozols (1971). J. Biol. Chem. 246:1706-1717 S. R. Slaughter, C. H. Williams, and D. E. Hultquist (1982). Biochem. Biophys. Acta 705:228-237 Y. Imai, and R. Sata (1977). Biochem. Biophys. Res. Commun. 75:420-426 P. Strittmater, P. Fleming, M. Connors, and D. Carcaran (1978). Methods Enzymal. 52:97-10] F. S. Mathews, M. Levine, and P. Argos (1971). Nature (London), New Biol. 233:15-16 F. S. Mathews, M. Levine, and P. Argos (1972). J. Mol. Biol. 64:449-464 0. J. Anderson, K. E. Mastav, and G. Blobel (1983). Prac. Natl. Acad. Sci. 80:7249-7253 H. G. Enoch, P. J. Fleming, and P. Strittmater (1979). J. Biol. Chem. 254:6483-6488 S. Tajima, and R. Sata (1980). J. Biochem (Tokyo) 87:123-134 K. Christiansen, and J. Carlsen (1985). Biochem. Biophys. Acta 815:215-222 P. Y. Chou, and G. D. Fasman (1977). J. Mal. Biol. 115:135-175 P. Y. Chou, and G. D. Fasman (1977). Trends Biochem. Sci. 2:128-131 G. van Jagaw, and W. Sebald (1980). Ann Rev. Biochem. 49:281-314 S. B. van Badman, M. A. Schuler, D. R. Jallie, and S. G. Sligar (1986). Prac. Natl. Acad. Sci. 83:9443-9447 S.R. Slaughter, and D.E. Hultquist (1979). J. Cell Biol. 46). 47). 48). 49). 50). 51). 22 83:231-239 S.L. Morrison, and H. Neurath (1953). J. Biol. Chem. 200:39-51 K.I. Altman (1959). Am. J. Med. 27:936 F. Ferra, H. Engh, L. Hudson, J. Kamhalz, C. Puckett, S. Malineaux, and R.A. Lazzarini (1985). Cell 43:721-727 C. Bonnerot, A. Galle, A. Jalliat, and J.C. Kader (1985). Biochem. J. 226:331-334 K. Madyastha, and N. Krishnamachary (1986). Biochem. Biophys. Res. Commun. 136:570-576 N. Oshino, Y. Imai, and R. Sata (1971). J. Biochem. (Tokyo) 69:155-167 CHAPTER II THE PRIMARY STRUCTURE OF CHICKEN LIVER CYTOCHROME BS DEDUCED FROM THE DNA SEQUENCE OF A cDNA CLONE ------ Reprint from ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS, Vol. 264, No. 1, July, pp343-347, 1988 23 ARCHIVES or BIOCHEMISTRY AND BIOPHYSICS Vol. 264, No. 1, July, pp. 343-347, 1988 24 COMMUNICATION The Primary Structure of Chicken Liver Cytochrome b5 Deduced from the DNA Sequence of a cDNA Clone HON G ZHANG AND CHRIS SOMERVILLEl Genetics Program and DOE Plant Research Laboratory, Michigan State University, East Lansing. Michigan 48821 Received January 25, 1988, and in revised form April 11, 1988 A cDNA clone encoding the chicken liver cytochrome 65 was isolated by probing a library with synthetic oligonucleotides based on a partial amino acid sequence of the protein. Determination of the DNA sequence indicated a 414-nucleotide open reading frame which encodes a 138-amina acid residue polypeptide. The open reading frame contains 6 amino acids at the amino terminus which were not present on any of the cytochrome 65 polypeptides for which the amino acid sequence has been determined directly, suggesting that the protein is proteolytically processed to the mature form. The results of genomic Southern analysis were consistent with the presence of two structurally different genes in the chicken genome, raising the possibility that the soluble and membrane-bound forms of the protein are the products of separate genes. 0 1988 Academic Press, Inc. Liver microsomal cytochrome b; is an amphipathic membrane protein consisting Of an N-terminal hy- drophilic domain which contains a functional heme as a catalytic site and a C-terminal hydrophobic do- main which anchors the pratein in the microsomal membrane (1, 2). It functions as a component of the microsomal stearyl-COA desaturase (3, 4), and is also involved in liver cytochrome P-450 reduction (5, 6). Determination of part or all of the amino acid se- quences of liver cytochrome b; from six different vertebrate species (7-14) has revealed that the pri- mary structures are highly conserved (15). The pro- teins characterized to date have a molecular mass of about 16 kDa and contain 133 amino acid residues. The protein has been extensively studied as a model for protein-protein interaction. protein-membrane interaction, and the dynamics of heme protein fold- ing (16-19). However, many questions remain con- cerning the mechanism and topology of the interac- tion of the cytochrome with membranes, the struc- ture and regulation of the genes which code for cytochrome 65, and the structure of the protein in nonvertebrates. ‘ To whom correspondence should be addressed. 343 Although no genes encoding cytochrome b. have been cloned previously, a gene encoding rat liver cy- tochrome b, has been synthesized and expressed in Escherichia coli (20). Since there are several ques- tions which cannot be addressed with a synthetic gene we have undertaken the cloning and sequencing of a chicken liver cDNA which encodes the mem- brane-associated cytochrome b,. EXPERIMENTAL PROCEDURES Materials. A chicken liver X-gtll cDNA library, constructed by blunt-end ligating EcoRI linkers to cDNA, was kindly provided by J . Dodgson (Michigan State University). Oligonucleotides were synthesized by the phosphoramidite method on an Applied Bio- systems 380A instrument. The plasmid pBluescript (KS‘) was purchased from Stratagene (San Di- ego, CA). Plaque screening. The cDNA library was plated on E. coli Y1090 and nitrocellulose plaque lifts were screened with the oligonucleotide mixtures by] and 65-4 (Fig. 1) which were end-labeled to an average specific activity of 10° dpm ug“ with [y-"P1ATP (3000 Ci mmol") and T4 polynucleotide kinase (21). Filters were prehybridized 3 to 5 h at 42°C in 6X SSC 0003-9861/88 $3.00 Copyright 0 1988 by Academic Press. Inc. All rights of reproduction in any form reserved. 344 GluAap PheGluAap Val GMGACTTCGMGACCT G T T C T bs-i Glu Val Gin Lya His Asn GMGTACMAMCATM G C G O C C T b5-4 FIG. 1. The oligonucleotide mixtures used as probes for the cytochrome b5 gene. (1X SSC is 150 mu NaCl, 15 mm sodium citrate ad- justed to pH 7.0). 50 mM NaPO. (pH 6.8), 5X Den- hardt’s solution (0.1% (w/v) Ficoll, 0.1% (w/v) poly- vinylpyrrolidone, and 0.1% (w/v) bovine serum albu- min), 100 pg ml" of sonicated herring DNA (Sigma). The hybridizations were carried out at 37°C for 24 to 30 h in the same solutions with the addition of 10% (w/v) dextran sulfate and 1-2 pmol ml" of labeled oligonucleotide. The temperature for final washing was based on an empirical formula (22) by assuming that all ambiguous positions contained A or T bases. The filters were washed four times for 5 min in 6X SSC at 22°C, twice for 30 min at 37°C, and once for 20 min at 42°C (for 65-1) or 40°C (for 125-4). DNA sequence analysis The 0.8-kb insert in a re- combinant phage was excised as a single fragment by cleavage with EcoRI and subcloned in both orienta- tions into the EcoRI site of pBluescript to produce the recombinant plasmids pHZS-l and pHZ5-2. A series of overlapping unidirectional deletions were made in plasmids pHZ5-l and pHZS-Z with exonucle- ase III and mung bean nuclease essentially as de- scribed (23). The deleted inserts from both orienta- tions were self-ligated to produce a series of overlap- ping plasmids which were then transformed into E. coli DH5a. The plasmids were sequenced on both strands as double-stranded DNA by the chain termi- nation method as described (24). Genomic Southern analysis. The chicken DNA was purchased from Clontech Laboratories (Palo Alto, CA). DNA (5 pg per lane) was digested to completion with restriction endonucleases, resolved by electro- phoresis in 0.8% agarose gels containing 89 mil Tris-berate (pH 8.2), 2 ma EDTA. and transferred to nitrocellulose filters as described (21). The filters were prehybridized for 4 to 6 h at 42°C in 50% (v/v) formamide, 5X SSC, 50 mm NaPO. (pH 6.8), 5X Den- hardt’s solution, 250 pg ml" of sonicated herring DNA. The hybridization was carried out at 42°C for about 2) h in the same solutions with the addition of 0.8 pg of probe DNA which was nick translated with [a-“PldCTP to a specific activity of about 10‘ dpm pg" (21). After hybridization, the filters were washed three times for 20 min at 42°C in 2X SSC, 0.1% SDS, then twice for 30 min at 65°C in 0.1x SSC, 0.1% SDS. 25 ZHANG AND SOMERVILLE RESULTS AND DISCUSSION From a partial amino acid sequence of the chicken liver cytochrome b, (13) we designed two nonoverlap- ping oligonucleotide mixtures of 17-mers (Fig. 1) which had the lowest possible degree of ambiguity. It was subsequently learned that the region of amino acid sequence used to design oligonucleotide mixture 65-1 erroneously contained an Asp residue instead of an Asn residue (14). However, this resulted in only one incorrect nucleotide in the oligonucleotide mix- ture and did not prevent the effective use of the mix- ture as a hybridization probe. These oligonucleotides were used to screen a th11_ cDNA library con- structed from chicken liver poly(A)* RNA. Among the 200,000 plaques screened, 41 hybridized to oligo- nucleotide mixture 65-1. Only 13 out of the 41 clones were also recognized by oligonucleotide mixture 65-4 which was derived from a sequence near the amino terminus of chicken cytochrome b;. The size of the inserts in these 13 phage were determined by restric- tion analysis, the largest insert, a 0.8-kb EcoRI frag- ment, from one of the 13 phage was subcloned in both orientations into the EcoRI site of pBluescript. and the DNA sequence was determined. The DNA sequence of the cDNA clone encoding chicken liver cytochrome b; and the deduced amino acid sequence of the open reading frame is shown in Fig. 2. The clone lacked a 3'-poly(A) sequence, sug- gesting incomplete methylation during library con- struction. Because the clone had only 20 nucleotides upstream of the first ATG codon, it appears likely that some of the mRN A leader sequence is also miss- ing. Thus, it is not possible to exclude the possibility that translation begins at a codon further upstream. However, the deduced amino acid sequence is in agreement at each residue with at least one of the two independently obtained partial amino acid se- quences of the chicken protein which were previously obtained from residue 8 to 91 (13. 14). The open read- ing frame of 414 nucleotides encodes a polypeptide of 15,544 Da containing 138 amino acid residues. All of the previously determined liver cytochrome 6. se- quences contain 133 amino acid residues and. where it has been unambiguously determined (15), begin with an N-acetylated alanine (designated Ala 1 in Fig. 2). Although the amino terminus for the chicken protein was not previously determined, the appar- ently ubiquitous presence of an N-terminal alanine on the other vertebrate proteins raises the possibility that the chicken and other forms of the protein are proteolytically processed from a larger precursor. Alignment of all available microsomal cytochrome 6., sequences indicates striking similarity between the chicken and other sources of cytochrome b; (Fig. 3). As with the cytochrome b5 from other vertebrates, most of the sequence heterogeneity is located at the N- and C-terminal ends. The sequence from residues 42 to 72, which forms the heme-binding site (25), is cDNA CLONE FOR CHICKEN LIVER CYTOCHROME 653 10 20 30 GGCTGTCTGTCAAGCGAGGAT ATG GTG GGC TCC Met Val Gly Ser -5 70 TAT CGG Tyr Arg 80 GAG GAG Glu Glu 15 140 TAC GAC Tyr Asp 35 90 GTG CAG AAG Val Gln Lys TAC Tyr 10 CTG Leu CGC Arg 130 CAC His 150 ATC ACC AAG Ile Thr Lys GTG Val CAC His 30 190 A66 Arg CGT Arg ATC Ile 200 CAA GCT GGG Gln Ala Gly 55 260 270 CTG TCG GAA ACA TTT Leu Ser Glu Thr Phe 75 210 GGA GAT GCT Gly Asp Ala GTC Val CTT Leu 50 GAG Glu 250 GCA Ala ACA Thr AGG Arg GAT Asp 70 310 CTT Leu GCG Ala 320 330 CCA GCA GAA ACT CTT Pro Ala Glu Thr Leu 95 CCG Pro AAG Lys 90 CAG Gln Lys 370 380 390 TGG ATC CCG GCA ATA GCA Trp TGG Trp TCC Ser 110 430 TCA Ser GTG Asn Val 115 440 450 TAC Tyr ATG Met 130 500 GAG Glu 510 520 530 CAT His TTC ACT ATT Ile Pro Ala Iie Ala Ala 26 345 40 GCC GGC Ala Gly l 100 AAC Asn 50 GAG GCG Glu Ala 5 60 C06 GGC Arg Gly AGT GAA Ser Glu GGT Gly TGG Trp 110 ABC ACC Ser Thr 25 170 CCT GGT Pro Gly 45 230 GAT GTT Asp Val 65 290 CAC CCG His Pro 85 350 TCT AAT Ser Asn 105 410 420 CTG ATG TAT CGT TCC Leu Met Tyr Arg Ser 125 120 ATC ATC Ile lie AAC Asn 20 ACC Ser TGG Trp CAG Gln 160 GAT Asp 180 GAA GAA Glu Glu CTG Leu 40 220 AAC Asn GAG CAC Glu ' His GGA Phe Gly 240 CAC TCT His Ser TTT Phe GAG Glu 60 GGC Thr 61 y 280 666 Gly 300 GAT AGA Asp Arg ATT lie 80 340 ACT Thr GAG Glu CTT Leu GAT Ile Asp 360 AGT TCA Ser Ser ATT lle ACC Thr 100 400 ATT Ile GTG Val CAG Gln TCC Ser GCA ATT Ile 120 460 GT0 GCC Val Ala 470 480 490 GCACCTTACTGAGAACTAATGCAAGAAGAGACTGATCTGGGAGAGAATAGAAGCAATCC 540 550 560 570 TAACCCAATATATTTCCTGACAAAAGCCTGATGTCTGAAGATAAATTCAACTTTTTCAGAAAACTGAACAATTCTTTTC 580 590 600 610 620 630 640 650 TGCTGTGCACTTTTCTTGATGTTGCCTTCTTATTTGCTGCACTGAAGTAATAAAAAGGCAGCATTTCTTTTCGTATAAC 670 700 710 720 730 660 680 690 AATATATTCTCTAATGAATGATTTGATAACTGTATTAGTTGCTGTATTAAAATAGTTTTGTAAGTAGCATTCTGATTCT 740 750 760 770 GGTTATATCTTTTTAATCTGTAATGGAGTCTGTCTTGCA FIG. 2. Composite nucleotide sequence of the cDNA for chicken cytochrome b. and the deduced amino acid sequence. The amino acid sequence is numbered from the alanine residue most com- monly found at the amino terminus of the vertebrate protein. The regions of homology to the oligonucleotide probes extend from nucleotides 82 to 98 and 217 to 233 for 65-4 and 65-1, respectively. completely conserved. The chicken polypeptide lacks one amino acid at the C-terminus which is present on all other known sequences. The overall amino acid sequence homologies of chicken liver cytochrome b, with the available sequences for this protein from other species are human 76.8%, porcine 77.4%, bo- vine 71.8%, rat 78.2% , and rabbit 79%. In vertebrate erythrocytes, a cytochrome b; is present in the soluble fraction where it is involved in the reduction of methemoglobin (26). The sequence of bovine erythrocyte cytochrome b5 was reported to be identical to the liver microsomal protein from resi- dues 1 to 97, suggesting that the erythrocyte protein was derived from the same gene product as the mi- crosomal protein by proteolytic processing during erythroid maturation (27). However, the presence of an amino acid difference at the C-terminal residue of the erythrocyte cytochrome b5 from human, porcine (15), and rabbit (7), suggests that mammalian eryth- rocyte cytochrome 65 is encoded by a different mRNA. Since only one amino acid difference was ob- served between the two forms, the two kinds of mRNA could arise from a single gene by differential mRNA splicing, or could be the products of highly conserved separate genes. In order to examine these possibilities we probed filters containing restriction digests of total chicken DNA with the complete cDNA and with a 3' HindIII fragment of the cDNA (nucleotides 307 to 774 in Fig. 2) which encodes only the hydrophobic domain (residues 91 to 132). When 2&46 -6 [M-v-c-s-S-EiA{§'G‘t‘I‘V‘R‘E E} ------- - A E o x(A z zaiégt x-Z-E-D Ac-A- -'-S—D-K xIJEJEJ - - 35 t «Ev-om - - - -H-K-V-Y-D- (Zix-v-v-o- -H-K-V-Y-D- -H-K-V-Y-D- -H-K-v-v4!} 70 i -G-H-S-T-D- -G-H-S-T-D- -G-H-S-T-D- -G-H-S-T-D- -G-H-S-T-D- -G-H-S-T-D- 105 * T-V S-N-S- T-V S-N-S iuu}D—s-N{E} T-V-D-S-N-S i-v(E}s-u~s T-V-D-S-N-S 2 7 ZHANG AND SOMERVILLE U 0 Q Chicken Porcine Bovine Horse Rat Rabbit 3:221: - Chicken - Porcine - Bovine {n- Horse - Rat - Rabbit T- Chicken T- Porcine T- Bovine T- Morse T- Rat T- Rabbit Chicken Porcine (8) Bovine (9) Horse (10) Rat (11) Rabbit (12) FIG. 3. A comparison of the deduced amino acid sequence of chicken cytochrome b; with the sequences obtained by direct amino acid sequencing of the microsomal cytochrome b5 from other vertebrates. The numbers in parentheses give the references for the sequences. The residues which differ from the most consensus sequence are enclosed in boxes. the entire cDN A was used as a probe we observed two E'coRI bands of 18 and 8.7 kb, three HindIII frag- ments of 3.5, 2.3, and 1.4 kb, and two BglII fragments of 16.5 and 2.3 kb (Fig. 4A). On the basis of other experiments (results not presented), we consider it likely that the slightly reduced intensity of the 18-kb EcoRI band was due to incomplete fragmentation by the acid treatment which resulted in incomplete transfer to the nitrocellulose filter. By contrast, when we probed the filters with the region of cDNA encoding the hydrophobic domain, we observed ho- mology only to one EcoRI fragment of 8.7, one HindIII fragment of 3.5 kg, and one BglII fragment of 16.5 kb (Fig. 48). There are no internal EcoRI or BglII sites and only one HindIII site in the cDNA clone. Thus, these results could be explained by the presence of two genes, one of which lacks homology to the region of the cDNA which encodes the hydro- phobic domain. These results are also consistent with the presence of one gene containing intron sequences of less than about 5.3 kb total length with one site each for EooRI, BglII, and HindIII. An unequivocal resolution of this problem will require the cloning and characterization of a cDNA for the soluble cy- tochrome b; from erythroid cells. In this respect, the E r1 8 E H 8 23.1- Q~ '. .- 9.4! - i... ' 6.6 - ‘l' 4.4! - - - 2.3 - Cu- 2.() - .a “ A 8 FIG. 4. Hybridization of cytochrome b, cDNA probes to chicken genomic DNA. The genomic DNA was digested with EcoRI (E), HindIII (H), and BglII (B). [A] The entire cDN A was used as a probe. [B] The HindIII fragment encoding the hydrophobic domain was used as a probe. cDNA CLONE FOR CHICKEN LIVER CYTOCHROME bss chicken is a favorable experimental organism be- cause it has nucleated erythroid cells. The availabil- ity of the cDNA clone described here should facilitate a resolution of this and several other problems con- cerning the structure and function of cytochrome ()5. ACKNOWLEDGMENTS We are grateful to J. Dodgson for the gift of the cDNA library. This work was supported in part by grants from the US. Department of Agriculture (86- CRCR-l-2046) and the US. Department of Energy (ACO2-76ER01338). REFERENCES 1. SPATZ, L., AND STRI’I'I‘MATTER. P. (1971) Proc. Natl. Acad. Sci. USA 68. 1042-1046. 2 SPATZ, L., AND STRI'I'I‘MA‘I‘TER, P. (1973) J. Biol. Chem. 248, 793-799. 3. OSHINO, N., IMAI, Y., AND SATO, J. (1971) J. Bio- chem. (Tokyo) 69. 155-167. 4. STRITTMATTER, P.. SPATZ, L., CORCORAN, D., ROGERS, M. J., SETLOW, B.. AND REDLINE, R. (1974) Proc. Natl. Acad. Sci. USA 71, 4565-4569. 5. WHITE, R. E., AND COON, M. J. (1980) Annu. Rev. Biochem 49, 315-356. 6. HILDEBRANT, A., AND EsrAsaoox, R. W. (1971) Arch. Biochem Biophys. 143. 66-79. 7. KIMURA, S., ABE, K., AND SUGrrA, Y. (1984) FEBS Lett. 169. 143-146. 8. OZOLS. L., AND GERARD, C. ( 1977) Proc. Natl. Acad. Sci. USA 74, 3725-3729. 9. FLEMING, P. J ., DAILEY, H. A., CORCORAN, D., AND STRITTMATTER, P. (1978) J. Biol. Chem. 253, 5369-5372. 10. OZOLS. J., AND GERARD, C. (1977) J. Biol. Chem. 252. 8549-8553. 11. Oz0Ls, J., AND HEINEMANN, F. S. (1982) Biochi m. Biophys. Acta 704. 163-173. 12. 13. 14. 15. 16. 17. 18. 19. 21. EL? 28 347 KDNDo, K.,TAJ1MA. S., SATO. R.. AND NARITA, K. (1979) J. Biochem (Tokyo) 86. 1119-1128. TSUGITA, A., KOBAYASHI, M.. TANI, S., KYO, S., RASHID, M. A., YOSHIDA, Y., KAJIHARA, T., AND HAGIHARA, B. (1970) Proc. Natl. Acad. Sci. USA 67, 442-447. NOBREGA, F. G., AND Oz0Ls, J. (1971) J. BiolChem. 246. 1706-1717. ABE, K., KIMURA, S., KIZAWA, R., ANAN, F. K., AND SUGITA, Y. (1985) J. Biochem. (Tokyo) 97. 1659-1668. DAILY, H. A., AND STRI'I'I‘MAT’I‘ER, P. (1979) J. Biol. Chem. 254. 5388-5396. ENOCH. H. G., FLEMING, P. J., AND STRITTMAT- TER, P. (1979) J. Biol. Chem. 254. 6438-6488. INOKO, Y. (1980) Biochim. Biophys. Acta 599, 359-369. BENDZKO, P.. AND PrEIL, W. (1983) Biochim. Biophys. Acta 742, 669-676. . BECK v0N BODMAN, S., SCHULER, M. A., JOLLIE, D. R., AND SLIGAR, S. G. (1986) Proc. Natl. Acad. Sci. USA 83, 9443-9447. MANIATIS, T.. FRITSCN, E. P.. AND SAMBROOK. J. (1982) Molecular Cloning, a Laboratory Man- ual. Cold Spring Harbor Laboratory, Cold Spring Harbor. NY. . SMITH, M. (1983) in Methods of DNA and RNA Sequencing (Weissman, S. M., Ed.), Praeger. New York. . HENIKOPF, S. (1984) Gene 28, 351-359. . ZHANG. H., SCHOLL, R., BROWSE. J., AND Sousa. VILLE, C. (1988) Nucleic Acids Res. 16. 1220. 25. MATHEWS. F. S., Aacos, P.. AND LEVINE, M. 26. (1972) Cold Spring Harbor Symp. Quant. Biol. 36, 387-393. HULTQUIST, D. E., AND PASSON, P. G. (1971) Na- ture New Biol. 229, 252-254. . SLAUGHTER. S. R.. WILLIAMS. C H., AND HULT- QUIS‘I‘, D. E. (1982) Biochim. Bioph ya Acta 705. 228-237. CHAPTER III CYTOCHROME 85 GENE IN CHICKEN Abstract A cDNA clone coding for chicken cytochrome b5 has been isolated from an erythrocyte cDNA library using synthetic oligonucleotides based on a partial amino acid sequence of the protein and the DNA sequence of the previously described chicken liver cytochrome b5 cDNA clone. The complete homology between the erythrocyte cDNA and the liver cDNA suggests that they are transcribed from the same gene. Both genomic blotting data and the mapping of cytochrome b5 genomic clones support the notion that there is only one cytochrome b5 gene in chicken. This gene appears to be responsible for the two forms of cytochrome b5 protein discovered in different organisms. This suggests that, at least in chicken, the formation of soluble erythrocyte cytochrome b5 occurs by proteolytic processing of membrane-bound cytochrome b5 found in liver. 29 30 Introduction Cytochrome b5 is a heme protein (1,2) which is involved in the fatty acid desaturation in animal liver (3), methemoglobin reduction in erythrocytes (4,5) and cytochrome P-450 reduction (6).. It exists in two forms: an amphipathic form in the microsomal membrane of animal liver, and a cytosolic form in erythrocytes. The amphipathic form, which is 133 amino acid residues long, consists of an N-terminal hydrophilic domain which contains a functional heme as a catalytic site and a C-terminal hydrophobic domain which anchors the protein in the microsomal membrane. The cytosolic form is equivalent to the hydrophilic domain of the amphipathic form. Both forms have been purified and sequenced from several species (1, 7). The amino acid sequences of the two forms in a given species are either the same or differ by only one amino acid residue at the C-terminus of the cytosolic form. For example, the primary structure of bovine erythrocyte cytochrome b5 is identical to its liver form from residues 1 to 97. However, residue 97 is proline for human erythrocyte and serine for porcine erythrocyte forms, whereas residue 97 of both human and porcine liver forms is threonine. This raises possibility that two forms of cytochrome b5 come from two different mRNAs. But the question as to the whether those mRNAs are transcribed from a single gene or two different genes cannot be answered with available data. Isolation of a cytochrome b5 cDNA clone from chicken liver has provided new information about cytochrome b5 primary structure (8). The genomic Southern analysis using this liver cDNA clone as a probe 31 Glu Asp Phe Glu Asp Val b -l GAA GAC TTC GAA GAC GT 6 T T C T Glu Val Gln Lys His Asn b5-4 GAA GTA CAA AAA CAT AA 6 C G G C G T Gly Arg Tyr Tyr Arg Leu Glu b5-7 (3’)CCG GCG ATG ATA GCC GAC CTC(S’) Val Ile Pro Ala Ile Ala Ala b5-8 (3’)CAC TAG GGC CGT TAT CGT CGT(5’) Figure l. Oligonucleotides designed from chicken cytochrome b protein (b5 -1, and b -4), and from chicken live; cytochrome 5b5 clone b5 7, and b5 -8). 32 indicated that either one gene or at most two genes encode cytochrome D5 in chicken. In this chapter, we describe the isolation and characterization of cDNA clones from erythrocyte cells and genomic clones of cytochrome b5. The results presented here suggest that there is only one gene in chicken which is responsible for all forms of cytochrome b5. This suggests that posttranslational modification is the mechanism responsible for the synthesis of the two'forms of cytochrome D5 in chicken. Materials and methods Materials: A chicken erythrocyte lambda gtll cDNA library was constructed by blunt-end ligating EcoRI linkers to cDNA (26). A chicken Charon 4A genomic library was constructed by collecting 7-23 kb fragments from a partial digestion of genomic DNA with enzymes AluI and HaeIII, then ligating them to EcoRI linkers (25). Both libraries were kind gifts from Dr. J. Dodgson (Department Of Microbilogy and Public Health, Michigan State University). The total RNA used for Northern blots and the beta-globin gene used as an internal standard were from D. Browne and J. Dodgson. The plasmid pBluescript (KS+) was purchased from Stratagene (San Diego, CA). Oligonucleotides (figure 1) were synthesized by the phosphoramidite method on an Applied Biosystems 380A instrument. 33 Plaque screening: The cDNA library was plated on 5; £911 Y1090 and nitrocellulose plaque lifts were screened with the oligonucleotide mixtures bs-l, b5-4, b5-7, and b5-8 (fig. 1) which were end-labeled to an average specific activity of 109 dpm ug'l with ( 7-32P)ATP (3000 Ci mmol'l) and T4 polynucleotide kinase (9). Filters were prehybridized 3 to 5 h at 42°C in 6 X SSC (1 X SSC is 150 mM NaCl, 15 mM sodium citrate adjusted to pH 7.0), 50 mM NaPO4 (pH 6.8), 5 X Denhardt’solution (0.1% (w/v) Ficoll, 0.1% (w/v) polyvinylpyrrolidone, and 0.1% (w/v) bovine serum albumin) and 100 ug ml'1 of sonicated herring DNA (Sigma). The hybridizations were carried out at 37°C for 24 to 36 h in the same solution with the addition of 1-2 pmol ml.1 of labeled oligonucleotide. The temperature for final washing was based on an empirical formula (10) by assuming that all ambiguous positions contained A or T bases. The filters were washed four times for 5 min in 6 x ssc at 37°C, twice for 30 min at 42°C (for bs-l), 40°C (for b5-4), and 60°C (for b5-7 and b5-8). The charon 4A library was plated on g; £911 803 sgpfi and nitrocellulose plaque lifts were screened with the chicken liver cDNA clone which was labelled by random-priming (24) to at least 109 dpm ug"l with (12-32P)dCTP (3000 Ci mmol'l). Filters were prehybridized 2 to 6 h at 42°C in 6 X SSC, 30% formamide (v/v), 50 mM NaPO4 (pH 6.8), 5 X Denhardt’s solution, 250 ug ml'l sonicated hering DNA. The hybridizations were carried at 42°C for 24-46 h in the same solution with the addition of 0.2 ug of labeled probe. The washing conditions were as follows: once in 6 X SSC plus 30% formamide at 42°C for 10 34 min, twice in 2 X SSC plus 0.5% SDS at 55°C for 20 min each time, twice in 0.1 X SSC plus 0.1% SDS at 55°C for 20 min each time. DNA sequence analysis: The 1.5 kb insert in a recombinant phage was excised as a single fragmentby cleavage with EcoRI and subcloned in both orientations into the EcoRI site of pBluscript to produce the recombinant plasmids pHZ-3 and pHZ-4. A series of overlapping unidirectional deletions were made in plasmids pHZ-3 and pHZ-4 with exonuclease III and mung bean nuclease essentially as described (11). The deleted inserts from both orientaions were self-ligated to produce a series of overlapping plasmids which were sequenced on both strands as double-stranded DNA by the chain termination method as described (12). Genomic Southern analysis: The chicken DNA was purchased from Clontech Laboratories (Palo Alto, CA). DNA (5 ug per lane) was digested to completion with restriction endonucleases, resolved by electrophoresis in 0.8% agarose gels containing 89 mM Tris-borate (pH 8.2), 2 mM EDTA, and transferred to nitrocellulose filters as described (9). The filter was prehybridized for 4 to 6 h at 42°C in 50% (v/v) formamide, 5 X SSC, 50 mM NaPO4 (pH 6.8), 5 X Denhardt’s solution, 250 ug ml"1 sonicated herring DNA. The hybridization was carried out at 42°C for about 20 h in the same solutions with the addition of 0.8 ug of probe 35 DNA which was nick translated with («z-32P)dCTP to a specific activity of about 108 1 dpm ug‘ (9). After hybridization, the filter was washed three times for 20 min at 42°C in 2 X SSC, 0.1% SDS, then twice for 30 min at 65°C in 0.1 X SSC, 0.1% SDS. Northern analysis: Total RNA (10 ug per lane) was electrophoresed in a 0.8% agarose gel containing formaldehyde (13), then blotted onto a nitrocellulose filter. The filter was prehybridized and hybridized under exactly the same conditions as in the genomic Southern analysis. The filter was first probed with the cytochrome b5 cDNA clone of erythrocytes, exposed to films, then rehybridized to a chicken13-globin gene (25). Results 1. Cloning and sequencing of a cytochrome bs gene from erythrocyte cells Oligonucleotides bs-l and b5-4 were previously used to obtain a cDNA clone for cytochrome b5 from chicken liver (8). These oligonucleotides were also used to screen a chicken erythrocyte cDNA library. Out of 200,000 plaques screened, 19 hybridized to b5-l, and two of these also hybridized to b5-4. These two clones, designated lambda HZ-3 and lambda HZ-S, were also recognized by oligonucleotides b5-7 and b5-8 which were based on regions of sequence from the hydrophilic domain and hydrophobic domain of 36 1 GTG TGG TGA GTC GCG GCG GCG TTG GGC TGT CTG TCA AGC GAG GAT ATG 48 Met 1 85-7 49 GTG GGC TCC AGT GAA GCC GGC GGT GAG GCG TGG CGG GGC CGC TAC TAT 96 2 Val Gly Ser Ser Glu Ala Gly Gly Glu Ala Trp Arg Gly Arg Tyr Tyr 17 85-4 97 C66 CTG GAG GAG GTG CAG AAG CAT AAC AAC AGC CAG AGC ACC TGG ATC 144 18 Arg Leu Glu Glu Val Gln Lys His Asn Asn Ser Gln Ser Thr Trp Ile 33 145 ATC GTG CAC CAC CGT ATC TAC GAC ATC ACC AAG TTC CTG GAT GAG CAC 192 34 Ile Val His His Arg Ile Tyr Asp Ile Thr Lys Phe Leu Asp Glu His 49 193 CCT GGT GGA GAA GAA GTC CTT AGG GAG CAA GCT GGG GGA GAT GCT ACT 240 50 Pro Gly Gly Glu Glu Val Leu Arg Glu Gln Ala Gly Gly Asp Ala Thr 65 85-1 241 GAG AAC TTT GAA GAT GTT GGC CAC TCT ACA GAT GCA AGG GCG CTG TCG 288 66 Glu Asn Phe Glu Asp Val Gly His Ser Thr Asp Ala Arg Ala Leu Ser 81 289 GAA ACA TTT ATT ATT GGG GAG CTT CAC CCG GAT GAT AGA CCG AAG CTT 336 82 Glu Thr Phe Ile Ile Gly Glu Leu His Pro Asp Asp Arg Pro Lys Leu 97 337 CAG AAA CCA GCA GAA ACT CTT ATT ACC ACT GTG CAG TCT AAT TCC AGT 384 98 Gln Lys Pro Ala Glu Thr Leu Ile Thr Thr Val Gln Ser Asn Ser Ser 113 85-8 385 TCA TGG TCC AAC TGG GTG ATC CCG GCA ATA GCA GCA ATT ATT GTG GCC 432 114 Ser Trp Ser Asn Trp Val Ile Pro Ala Ile Ala Ala Ile Ile Val Ala 129 433 CTG ATG TAT CGT TCC TAC ATG TCA GAG TGA GCA CCT TAC TGA GAA CTA 480 130 Leu Met Tyr Arg Ser Tyr Met Ser Glu "' 481 ATG CAA GAA GAG ACT GAT CTG GGA GAG AAT AGA AGC AAT CCT AAC CCA 528 529 ATA TAT TTC CTG ACA AAA GCC TGA TGT CTG AAG ATA AAT TCA ACT TTT 576 577 TCA GAA AAG TGA ACA ATT CTT TTC TGC TGT GCA CTT TTC TTG ATG TTG 624 625 CCT TCT TAT TTG CTG CAC TGA AGT AAT AAA AAG GCA GCA TTT CTT TTC 672 673 GTA TAA CAA TAT ATT CTC TAA TGA ATG ATT TGA TAA CTG TAT TAG TTG 720 721 CTG TAT TAA AAT AGT TTT GTA AGT AGC ATT CTG ATT CTG GTT ATA TCT 768 769 TTT TAA TCT GTA ATG GAG TCT GTC TTG CAT ATG AAT TTT ATA GCT TTA 816 817 AAT TAG TAG CAA AAC TTT GTA CAT GTA TTT GTC CAT GTA CAC AAC CTA 864 865 ACT TAA AAA TCA TGT TGT CGT CTT AAA TCT AGA ATG TTT GAG TAA GAG 912 913 GCT AAT TAA AAT AAA CAT AAT GGA AGA AGC TGA GTA TAG TAA TGA GTA 960 961 CAG GTG CCT GTA AAT GGT TGG GTC CTG CCA GTC AGG CTA TAA GAA GAT 1008 1009 AAC TTT CCT TCC CTC CTG CCA TGT GGT CTT AGA GTT GTT ACA GGT ACT 1056 1057 CCT GCT GGC AAG CTG TTG TTT GAC TGC CAT GGG AAA ATT AAA GTA AAA 1104 1105 TAT GAA ATC CAC TGG CCG AGT TAT GTC CAT CTC CGT TTT GTG AAC TGT 1152 1153 TGA ACT GTT CTG CAA AAA AGG CAG AAA GTG CTG TGT AAA TTC CAC TAC 1200 1201 AGG TAA TAT AAC TGC TAC TAA TAC TGT TCT TGC CAA GCA CTC AGG TGA 1248 1249 CTC TGA AAC TTG TTC TGG AAC TTC TAG ACT TGT ATA CAA TCT TCA ACT 1296 1297 TTA TCA TGG TAT GTG CTG ATG GGG TGG AAA AAG TGA TGG TTC TGA CTG 1344 1345 TTC TGT TAT GTG CTC CTT GGT GCT TTA CTA TGG AGA GAT GAG CAT TTT 1392 1393 CTG TGC TAA ATA CAG GAC AAC TGA AAG TCT GCA TTT TGT GGT GAA TTT 1440 1441 TTT TTT TAT TTT TAT TTT TTA GTG ATG CAT AAA TGA TCA TGA ATA AAA 1488 1489 GTT TAA TTG CTT ACT CTT T Figure 2. Sequence of a cDNA clone and the deduced amino acid sequence for cytochrome b from erythrocytes. Regions where oligonucleotides recogniEe are marked. 37 cytochrome b5 respectively. The fact that the hydrophobic domain probe (b5-8) hybridized to the two clones suggested that cytochrome b5 mRNA in erythrocytes encodes a hydrophobic domain comparable to that in cytochrome b5 mRNA from liver. The inserts in lambda HZ-3 lambda HZ-5 were subcloned into the EcoRI site of pBluescript to produce plasmids designated pHZ-3 and pHZ-5. The larger insert (1.5 kb) was in pHZ-3. The complete sequence of the insert in pHZ-3 is shown in figure 2. This cDNA clone was approximately twice as large as the cDNA clone from chicken liver (8). The region of the cDNA from nucleotides 27 to 799 was 100% homologous to the liver cDNA clone. The open reading frame extends from nucleotide 48 to 462, so most of the extra sequence in pHZ-3 is at the 3’ untranslated region. 2. Expression of the cytochrome b5 gene in liver and erythrocyte cells The complete sequence identity between the cDNAs from erythrocyte and liver cells suggested that there is only one kind of cytochrome b5 message in both liver and erythrocyte cells. In order to examine this, a Northern blot of total RNA from liver and erythrocyte cells was probed with the cDNA clone from liver (figure 3). The cytochrome b5 probe hybridized to a 1.6 kb mRNA from liver which is approximately the same size as the cDNA clone from erythrocytes. The intensity of signal was relatively high as indicated by the fact that panel A in Figure 3 is an 24 h exposure. By contrast, there was no apparant hybridization of the probe to the 38 12 12 288- 188- A 8 Figure 3. The Northern Blotting of liver total RNA (lane 1) and erythrocyte total RNA (lane 2). Filter was first hybridized to cytochrome b5 gene (A), then the filter was washed before it was rehybridized to bete-globin gene (8). 39 H HH 1; E E .95 '2‘ .E H F '21 ”'1 92 917 I m "I m r]: Figure 4. Restriction maps of three genomic clones of cytochrome b Navy lines designate the fragments which hybridized t8 the indicated oligonucleotides. 4O ierythrocyte RNA lane, even after 3 days exposure. Presumably the abundance of cytochrome b5 mRNA is relatively low in this cell type. This is consistant with the fact that only two clones out of 200,000 plaques screened were recovered from the erythrocyte cDNA library. In order to ensure that the mRNA was not degraded, the filter was rehybridized to the beta-globin gene. In this case, the probe hybridized strongly to an mRNA of the correct size (0.6 kb). This internal standard indicates that the RNA from erythrocytes was intact and was accurately quantitated. These results are consistent with the hypothesis that the membrane-bound form (amphipathic form) and the cytosolic form (hydrophilic form) of cytochrome b5 are the products of posttranslational modification of a polypeptide produced from a common mRNA. 3. Cloning and mapping the genomic sequences The chicken liver cDNA clone was used as a probe to screen a chicken Charon 4A genomic library. Twelve clones were isolated from 600,000 plaques. Analysis of the restriction pattern of these phage indicated that only three were independent. A partial restriction map for each of these 3 clones, designated 92, 95, and 917, is presented in figure 4. The devised restriction maps overlapped, indicating that they all contain a common region of the chicken genome. 41 A1 B» 1C 23-1 "' 23.1— D 9.4— 23 1 9.4— 6'6_ 9:4— 6-6— . . d—bs" ' 4.4— '4 fi—b5-7 iL3- 203— 2.0— . 2.0- 203— 2.0— 0.5— 0.5— Figure 5. Restriction digests of chicken genomic DNA and the 95 clone probed with liver cytochrome b cDNA or oligonucleotides. Lanes A and B are genomié digests, C and D are 95 clone digests. A and C were digested with Hind III, B and D were with Xba I. A, B, and C were probed with cytochrome b5 gene, and D was probed with oligonucleotides b -l, b -7, and b -8 sequentially. The bands recognizgd by gpecific oTigonucleotides in lane 0 are marked. 42 4. Genomic southern analysis In a previous experiment (see Chapter II), chicken genomic DNA was digested with 3 restriction enzymes and probed with the chicken liver cytochrome b5 cDNA clone. The results were consistent with 2 possible explanations. 1. There are two genes in the chicken genome which give rise to two kinds of cytochrome b5; 2. There is only one gene in the chicken genome, but within the gene, there is approximately 5.3 kb of intron DNA which harbors one site each for EcoRI, BglII, and HindIII. Hhen clone 95 was digested with HindIII and XbaI and probed with the chicken liver cDNA clone, the hybridization pattern produced was exactly like that of genomic digests (figure 5). The results of this experiment indicate that there is only one cytochrome b5 gene in the chicken genome which is entirely contained in clone 95, because the 18 kb fragment of lambda clone 95 generated tha same hybridization pattern as the total chicken genomic DNA did. Clone gS encodes one copy of microsomal cytochrome b5 gene. From analysing the size of the fragments which hybridize to the oligonucleotides, it is apparent that there are introns within the gene, and the introns are more than 5.3 kb long, but less than 9.3 kb. Discussion It has been suggested that the bovine erythrocyte cytochrome b5 protein is derived from the same gene product as the microsomal 43 protein by proteolytic processing during erythroid maturation (7). This is based on the observation that the cytosolic cytochrome b5 is not present in an immature erythroid cell. Instead, a membrane-bound form of cytochrome b5 is present (23). Electron microscopy also showed that the endoplasmic reticulum disappears during erythroid maturation. Microsomal cytochromeb5 from liver cells can, when treated with liver lysosomal proteases, produces two hydrophilic segments one of which was identical to the form II of bovine erythrocyte cytochrome b5 (7). Erythrocyte cytochrome b5 I and II are equivalent to residues 1-97 and 1-95, respectively, of microsomal cytochrome b5 in bovine. The existence of lysosomal proteases capable of converting microsomal cytochrome b5 to the cytosolic protein nurtured the idea that the putative erythroid proteases are responsible for the solubilization of microsomal cytochrome b5 in erythrocyte cells. Several lines of evidence presented here support the above hypothesis. First, a cDNA clone from erythrocytes was about 1.5 kb long. This is comparable in size to the homologous RNA from liver. Second, the cDNA clones from erythrocyte and liver cells are 100% homologous. Both of the erythrocyte cDNA clones contained the hydrophobic membrane-binding domain, indicating that they both encode a protein which is exactly the same as the microsomal cytochrome b5 found in liver. Third, all the fragments of chicken genomic DNA with homology to the cDNA clone are carried on a single clone (95). Furthermore, analysis of the structure of this clone by probing with oligonucleotides from various regions of the coding sequence are consistent with the existence of only one cytochrome b5 44 gene in chicken. Therefore, this gene must be responsible for synthesis of two forms of cytochrome b5. Unfortunately, there has been no direct characterization of the cytochrome b5 protein in the chicken erythrocyte. Thus, it is not possible to directly compare the sequence of the erythrocyte protein to the deduced sequence of the cDNAs. However, if, as seems certain, a soluble form exists in chicken, it must be derived by posttranslational modification. As to the analogous situations in human and porcine, we cannot rule out the possibilities that there are two different genes, or one gene with alternative RNA splicing products, or posttranslational modification. The level of gene expression is regulated in an organ-specific manner, presumably reflecting the degree of necessity. Liver is the place where fatty acids, cholesterol, and other products are being actively synthesized (3, 14-17). As a component of those biosynthetic processes, cytochrome b5 plays an obligatory role. In erythrocytes, however, only a relatively small quantity of cytochrome b5 may be required for methemoglobin reduction, and thus the level of its mRNA is much lower in red cells as compared to liver. No other functions have been documented for cytochrome b5 in erythrocytes. Of course, it is also possible that the rate of turnover of cytochrome b5 may be much lower in erythrocyte cells than that in liver, so that the actual concentration of cytochrome b5 molecules could be comparable. There are several recent reports concerning the cytochrome b5 cDNA structure in rabbit and bovine (18, 19). The rabbit and chicken liver cytochrome b5 mRNAs are 63% homologous. No studies of the regulation of the cytochrome b5 gene are available. Since terminal 45 Adesaturase activity was shown to be regulated by hormones and diets (20-22), it would be interesting to test whether or not cytochrome b5 is subject to any kind of regulation besides organ-specific regulation. It is unlikely that the cytochrome b5 gene is subject to a simple "On or Off" form of regulation, since it is involved in so many vitally important biological processes. Acknowledgments He thank Dr. J. Dodgson for his generous gifts of plasmid, libraries, andespecially the advice. References 1). K. Abe, S. Kimura, R. Kizawa, F. K. Anan, Y. Sugita (1985). J. Biochem. (Tokyo) 97:1659-1668 2). F. S. Mathews, P. Argos, and M. Levine (1972). Cold Spring Harbor Symp. Quant. Biol. 36:387-393 3). N. Oshino, Y. Imai, and R. Sato (1966). Biochem. Biophys. Acta. 128:13-28 4). D. Hultquist, and P. G. Passon (1971). Nature (London). 229:252-254 5). D. Hultquist (1979). Methods Enzymol. 52:463-473 6). R. E. Hhite, and M. J. Coon (1980). Ann. Rev. Biochem. 49:315-356 7). S. R. Slaughter, C. H. Williams, and D. E. Hultquist (1982). Biochem. Biophys. Acta 705:228-237 8). 9). 10). 11). 12). 13). 14). 15). 16). 17). 18). 19). 20). 46 H. Zhang, and C. Somerville (1988). Arch. Biochem. Biophys. 264:343-347 T. Maniatis, E. F. Fritsch, J. Sambrook (1982). Molecular Cloning, a Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY M. Smith (1983). In Methods of DNA and RNA Sequencing (Heissman, S. M., Ed.), Praeger, New York. 5. Henikoff (1984). Gene 28:351-359 H. Zhang, R. Scholl, J. Browse, and C. Somerville (1988). Nucleic Acid Res. 16:1220 F. M. Ausubel, R. Brent, R. E. Kingston, 0. D. Moore, J. G. Seidman, J. A. Smith, K. Struhl (1988). Current Protocols In Molecular Biology. 4.9.1-4.9.4 S. R. Keyes, J. A. Alfano, I. Jansson, and D. L. Cinti (1979). J. Biol. Chem. 254:7778-7784 V. R. Reddy, D. Kupfer, E. Caspi (1977). J. Biol. Chem. 252:2797-2801 F. Paultauf, R.A. Porugh, B.S.S. Masters, and J.M. Johnson (1974). J. Biol. Chem. 249:2661-2662 F.F. Kadlubar, and D.M. Ziegler (1974). Arch. Biochem. Biophys. 162:83-92 N. Dariush, C.H. Fisher, and A.H. Steggles (1988). Prot. Seq. Data Anal. 1:351-353 R.J. Cristiano, and A.H. Steggles (1989). Nucleic Acids Res. 17:799 P.M. Lippiello, C.T. Holloway, S.A. Garfield, and P.H. Holloway (1979). J. Biol. Chem. 254:2004-2009 21). 22). 23). 24). 25). 26). 47 V.C. Joshi, and L.P. Aranda (1979). J. Biol. Chem. 254:11779-11782 N. Oshino, and R. Sato (1972). Arch. Biochem. Biophys. 149:369-377 S.R. Slaughter, and D.E. Hultquist (1978). J. Cell Biol. 83:231-239 ' A.P. Feinberg, and B. Vogelstein (1983). Anal. Biochem. 132:6-13 J.B. Dodgson, J. Strommer, and J.D. Engel (1979). Cell 17:879-887 N.S. Yew, H.-R. Choi, J.L. Gallarda, and J.D. Engel (1987). Proc. Natl. Acad. Sci. 84:1035-1039 CHAPTER IV SEARCHING FOR THE PLANT CYTOCHROME 85 1. Introduction The presence of cytochrome b5 in microsomal membranes of higher plants has been documented in the literature, and several groups have reported the purification of cytochrome b5 proteins (1-3). Jollie et al. determined a short amino acid sequence from the N-terminus of pea cytochrome D5 which is completely unrelated to any known sequences of cytochrome b5 from animals. Since the sequence of cytochrome b5 is so highly conserved among animals (4), we considered it possible that the plant cytochrome b5 genes may be isolated by exploiting sequence homology instead of purifing the protein first. Toward this end, two approaches were tried. Oligonucleotides corresponding to the most conserved regions of the cytochrome b5 protein were used as probes to screen plant genomic or cDNA libraries, and the chicken cytochrome b5 gene was used as a heterologous probe. 48 Figure l. 49 Trp Trp Thr Asn Trp Val TGG TGG ACN AAC TGG GT T His His Lys Val Tyr Asp CAT CAT AAA GTN TAT GA C C G C Glu Glu Ile Lys Lys His Asn GAA GAA ATT CAA AAA CAT AA G G C G G C A Oligonucleotides designed from animal cytochrome 65 protein (see figure 2). N represents T, A, C, and G. 50 2. Materials and methods The Arabigopsis lambda gt10 cDNA library, constructed from leaf mRNA, was from N. Crawford (Stanford, CA). The EMBL4 genomic library was from E. Meyerowitz (Caltech, CA). The conditions for oligonucleotide end-labelling, hybridization, and screening were the same as in chapter 11. The heterologous probing was carried out as follows: prehybridization was done in 30% formamide, 5 X SSPE, 5 X Denhardt’s solution, 50 mM NaPO4 (pH 6.5), 0.2 mg/ml sonicated herring DNA for 12 hrs at 42°C. Hybridization was done in the same solution except using E; 5911 DNA as carrier DNA and adding dextran sulfate to 5% for 48 hrs at 42°C. The filters were washed twice in 5 X SSPE plus 0.5% SDS for 20 min each time at 37°C, then twice in 2 X SSPE plus 0.1% SDS for 20 min each time at 37°C and finally once in 0.2 X SSPE, 0.1% SDS at 42°C for 30 min (1 X SSPE is 150 mM NaCl, 10 mM NaH2P04, and 1.3 mM EDTA adjusted to pH 7.4). The construction of overlapping deletions and DNA sequencing were done as described in (6, 7). All the other DNA manipulations were performed following the standard methods (5). The oligonucleotide mixtures in figure I were deduced from animal protein sequences, as indicated in figure 2, and they were used as probes to screen Arabjdgpsis cDNA libraries. 51. [M-V-G-S-S-EiAiG-G-E-A-H-R-G-R} from were with the sequences obtained by direct of the microsomal cytochrome b3 5 Regions where the oligonucleotid A comparison of the deduced amino acid sequence of chicken cytochrome b amino acid sequencin other vertebrates. Figure 2. designed are marked. 52 3. Results: 1). Chicken liver cytochrome b5 clone as a probe Hhen the chicken liver b5 cDNA was used as a probe to hybridize to an Arabjdgpsis genomic EcoRI digest, several fragments hybridized to the probe (figure 3). 23.1— 212: 4.4- 2.3— 2:0 Figure 3. The Eco RI digest of Arabidoosis genomic DNA was probed with chicken liver cytochrome b cDNA clone. Lane A was washed 3t 37 C for the final wash, and Lane B was washed once more at 46 C. 0.5— (an. After a 46°C wash, the 1.7 kb and 4.0 kb bands still gave strong signals, suggesting significant homology to the probe. Therefore, the cDNA clone was used as a probe to screen both genomic and cDNA libraries. Three clones out of 200,000 plaques from the genomic library, and four out of 150,000 plaques from the cDNA library were 53 isolated which hybridized to the probe. The DNAs were made from all these clones, and digested with EcoRI. It turned out that all three genomic clones covered a common fragment (1.7 kb) which hybridized to the probe. This fragment (ABS-IO-O) together with other two larger cDNA clones (AB5-11 and ABS-l3-0) were subcloned into pBluescript plasmids, and ABS—lO-O and ABS-I3-0 were sequenced in one orientation as double stranded DNA (6, 7). Figure 4 shows the sequencing strategies these two clones. # ——# fl fl 7 ———-——o ——-——o ———> 0.1 kb h—i genomic clone ABS-io-O fl ———fl ———d —’ _——, —, ———a —) ————) . 4 _____. PolyA O 1 kb cDNA clone ABS-130 Figure 4. Sequencing strategies for clones ABS-lO-O and ABS-I3-0. There was no particular reason to choose the cDNA clone AB5-13-0 to sequence first, since the sequencing of cDNA clone ABS-II was also planned (later the plan was droped). Both cDNA clones hybridized to the chicken probe equally well under low stringency condition. SI 10 AAACGCTCAG 90 ATTAGATGAA 170 GAAGGACAAG 250 GGAAAGTTGA 330 GGAAATTTCT 410 AACCCTAGAT 490 CAAGAAATCC 570 TTGCCACATC 650 CACAATCTTG 730 TACTAACTAC 810 AACACATCAG 890 ACTGCACGTT 970 ACAAGGATGG 1050 AAGAGTTTCC 1130 AAGTACACTC 1210 AGTGGGTCCA 1290 ATCCATTCTT Figure 5. 20 GACAGGGACG 100 TCCGAGATTG 180 CGCAAAATAG 260 CTTGAATGAC 340 CCGACGACAG 420 GGGCAGTTTA 500 GACGGCCCTC 580 AAAAAAATGG 660 ATATCAATGC 740 CTGCTGAGGG 820 CTCCCACAAT 900 GCAGTACAGG 980 GTTAACTCCG 1060 ACTTAGCAGA 1140 ACTTATGAGA 1220 CTGTCTCACT 1300 AAAGAAAATA 54 30 GGTGGACTCC 110 AGCCCGAGAA 190 CGCAGGAGAT 270 TTTCTCACGT 350 TGGAGCGTCA 430 CGGAAAGGGA 510 GAAAGCTGCT 590 CTTCCTCTTC 670 AACCGATGTG 750 AATCGGCAAA 830 AAAACTTCTC 910 CCAGAAGAAG 990 CTTGGGCTTT 1070 CACAAGAAGA 1150 ACCTGAGAAA 1230 TCCTTAAATT 1310 TCTGAAAATA DNA Sequence 40 ACTGCACGTT 120 CCTCGTGCCA 200 GGAGTTCGGG 280 ACAAGGAAGC 360 AGATTGGAGA 440 TTCGACCACG 520 TTCAAAAGAA 600 ACACACTGGC 680 GGCGGCTTGA 760 TCCATTTGTT 840 CTACTGTATA 920 CGACATTGTA 1000 GCTCTACCTT 1080 GATTGGTAAC 1160 AGGAGATGGA 1240 TGGTTTGCTG 1320 AATAAAAAGT 50 GCAGTACAGG 130 GAGGAATGGA 210 GTTCGGGTGG 290 CAAGTTGGCT 370 GACCGCTGTA 450 TTGCCAAGTT S30 GAGAAGTTTA 610 AGCATGTGGA 690 CAGTACTTCA 770 CTTGATGACG 850 ACGCTGATAT 930 AAGCTTCTTT 1010 GGAAGAGAGA 1090 AACAGATGAA 1170 GGTAAAGGTG 1250 TTAGTCTTAT 60 CAAGAAGAAT 140 GGGATATCAG 220 AGAAGAAGAG 300 CAATTGAGGC 380 TCATCTCCCA 460 CTTCAATAGC S40 TGCTCAATAG 620 GAGTTTTATC 700 CCGAGCAATC 780 AAGGTGCGAC 860 AAACGCTCAG 940 TGATAAAAGG 1020 TAAGGACGTA 1100 GATATTGAAT 1180 ATGATTAGGG 1260 CCATCGATTT AATACATAT 3' of the clone ABS-IB-O. 70 CGGCGTGGCG 150 GCGGAGGTGA 230 GCAAGGGCTA 310 CTGTCATTCT 390 GCGAGCGAGT 470 GACAAGTACG 550 CCGGAATCCT 630 TGGTTGATTC 710 ATTGGTAAGA 790 CTTGACGCAC 870 GACAGGGACG 950 GGCGGACATA 1030 TGAGGTGATG 1110 AGTCCTTCAA 1190 CATTGGAACC 1270 TGGATATTTA 80 GGGCTAAGAG 160 ATCTGACGAA 240 ATTCCGCTGA 320 CGATAAACCG 400 .GGCTCCTAAG 480 ATCCCAGCGA 560 GACCTAGCCG 640 CTTGCTAAAG 720 AGCAGGCTAT 800 TATGCTGTGA 880 GGTGGACTCC 960 GAAGTGAAGA 1040 AAGCTGTTGA 1120 TTTCAGCTTG 1200 TCGGAGTCGG 1280 TCACAACTTG 55 V Figure 5 shows the complete sequence of ABS-13-0. There might be some mistakes in this sequence, since it was only sequenced in one orientation. The DNA data bank (Gene Bank R58.0, December, 1988 and EMBL Bank R17.0, November 1988) was searched with HIBIO DNASISTM program. No sequences were identified which had similarity more than 50% to AB5-l3-0. According to Doolittle (12), the range of chance similarities between two unrelated sequences can exceed 50% if we allow gaps in the sequences. Similarities below 50% were not considered informative. All three reading frames have been used to screen the protein data bank (Protein Identification Resource R18.0 September 1988) to look for homologous proteins with HIBIO PROSISTM program. No cytochrome b5 protein from any sources was found to be homologous. However, there is one protein from Qrgsgphila called Notch in which part of its sequence is similar to reading frames 2 and 3 at about 25% identity (figures 6 and 7). Possibly there are some mistakes in the sequence of ABS-I3-0 which shifted the reading frame and split the similarities into two reading frames. However either match extends over 100 residues. It is very likely they are related. The significance of this observation will be discussed later. The genomic clone ABS-lo-O was also sequenced in one orientation (figure 4), and the sequence was analyzed the same way as the clone ABS-13-0 except all 6 reading frames were analyzed (3 for each direction). Because it was a genomic clone, we don’t know which was the sense strand. No information can be deduced concerning this clone after searching the DNA bank and the protein bank. The chance of finding a similar DNA sequence in the DNA bank is very low, A8513RF2.AMI A24768 A8513RF2.AM1 A24768 A8513RF2.AMI A24768 A8513RF2.AM1 A24768 A8513RF2.AMI A24768 A8513RF2.AMI A24768 A8513RF2.AM1 A24768 A8513RF2.AMI A24768 Figure 6. 56 PROSIS HOMOLOGY SEARCH 24.6% identity in 130 aa overlap 10 20 30 40 50 60 NAODROGUTPLHVAVOARRIGVAGLRDMNPRLSPRTSCORNGGISGGGESDEEGGAGNSA PKRORSOPVSGVGLGNNGGYASOHTMVSEYEEADGRVUSOAHLDVVDVRAIMTPPAHODG 1840 1850 1860 1870 1880 1890 70 80 90 100 110 120 GDGVRGSGGEEEARANSAEESLELSHVOGSOVGSIEACHSRTGKFLRROHSVKIGETAVS GKHDVDARGPCGLTPLHlAAVRGGGLDTGEDIENNEDSTAOVISDLLAOGAELNATMOKT 1900 1910 1920 1930 1940 1950 130 140 150 160 170 180 SPSERVAPKNPRUAVYGKGFDHVAKFFNSDKYDPSDKKSDGPRKLLSKEEKFMLNSRNPD GETSLHLAARFARADAAKRLFHAGAOANCGDNTGRTPLHAAVAADAMGVFGILLRNRATN 1960 1970 1980 1990 2000 2010 190 200 210 220 230 240 LAVATSKKULPLHTLAACGEFYLVDSLLKHNLDINATDVGGLTVLHRAIIGKKOAITNYL LNARMHDGTTPLILAARLAIEGMVEDLITADADINAADNSGKTALHUAAAVNHTEAVNIL 2020 2030 2040 2050 2060 2070 250 260 270 280 290 LRESANPFVLDDEGATLTHYAVKHISSHNKTSPTVRY KRSGOGRVDSTARCSTGOKKRH LMHHAHROAODDKDETPLFLAAREGSYEACKALLDNFANREITDHMDRLPRDVASERLHH 2080 2090 2100 2110 2120 2130 300 310 320 330 340 350 CKASFDKRGGHRSEEOGUVNSAUALLYLGREIRTYEVMKLLKEFPLSRHKKRLVTTDEDI 0IVRLLDEHVPRSPOMLSMTPOAMIGSPPPGOOOPOLITOPTVISAGNGGNNGNGNASGK 2140 2150 2160 2170 2180 2190 360 370 380 390 400 410 ESFNFSLKYTHLEPEKRRHRRLGHHNLGVGVGPLSHFLKFGLLLVLSIDFGTLSOLDPFL OSNOTAKOKAAKKAKLIEGSPDNGLDATGSLRRKASSKKTSAASKKAANLNGLNPGOLTG 2200 2210 2220 2230 2240 2250 420 430 KKISENKKVIH GVSGVPGVPPT 2260 Result of the protein data bank search with reading frame 2 of the clone ABS—13-0. Figure 7. A8513RF3.AM1 A24768 A8513RF3.AMI A24768 A8513RF3.AHI A24768 A8513RF3.AMI A24768 A8513RF3.AHI A24768 A8513RF3.AM1 A24768 A8513RF3.AM1 A24768 A8513RF3.AM1 A24768 2130 57 10 20 30 40 50 60 TLRTGTGGLHCTLOYROEESAURGEIRlRDAREPRARGMEGYOAEVNLTKKDKRKIAOEM AHGVTHFPEGFRAPAAVMSRRRRDPHGOEMRNLNKOVAMOSOGVGOPGAHUSDDESDHPL 1780 1790 1800 1810 1820 1830 70 80 90 100 110 120 EFGVRVEKKRGGLlPLRKVDLNDFLTYKEAKLAOLRPVILDKPGNFSDDSGASRLERPLY PKRQRSDPVSGVGLGNNGGYASDHTMVSEYEEADORVUSOAHLDVVDVRAIHTPPAHOOG 1840 1850 1860 1870 1880 1890 130 140 150 160 170 180 HLPASEULLRTLDGGFTERDSTTLPSSSIATSTIPATRNPTALESCFOKKRSLCSIAGlL GKHDVDARGPCGLTPLMIAAVRGGGLDTGEDlENNEDSTAOVISDLLAOGAELNATMDKT 1900 1910 1920 1930 1940 1950 190 200 210 220 230 240 TPLPHQKNGFLFTHUOHVESFIULIPCSTILISMOPMUAAOYFTEOSLVRSRLLLTTCGN GETSLHLAARFARADAAKRLFHAGADANCODNTGRTPLHAAVAADAMGVFOILLRNRATN 1960 1970 1980 1990 2000 2010 250 260 270 280 290 300 R01HLFLMTKVRPRTMLNTSAPTIKLLLLYNADINAODRDGUTPLHVAVOARRSOIVKLL LNARMHDGTTPLILAARLAIEGMVEDLITADADINAADNSGKTALHUAAAVNNTEAVNIL 2020 2030 2040 2050 2060 2070 310 320 330 340 350 360 LIKGADlEVKNKDGLTPLGLCSTLEERGRMRSCKSFHLADTRRDUOOMK1LNSPSISAST . I ... O . .. ... O. 0...... ... . O O. ..... ...... I . LMHHANRDAODOKDETPLFL-'°AAREGSYEACKALLDNFANREITDHMDRLPRDVASER 2080 2090 2100 2110 2120 370 380 390 400 410 420 LTYENLRKGDGGKGDDGIGTSESEHVHCLTSLNLVCCSYPSILDIYHNLIHSRKYLKINK LHHDIVRLLDEHVPRSPONLSMTPOAMIGSPPPGOOOPOLITOPTVISAGNGGNNGNGNA 2140 2150 2160 2170 2180 KYI SGK 2190 Result of the protein data bank search with reading frame 3 of the clone ABS-13-0. 10 5' CATACATTAA 90 TCGGTCAATA 170 GCTTGGGTCG 250 TGAACGGATC 330 AGGTCTAAAT 410 AAAAGAAGCT 490 GCCAAAAGGC 570 AATGTTGTGG 650 AATTTATATA 730 GTCTTTACAA 810 AAAACTCTTT 890 CAGTGACAAC 970 ACAAGTACGA 1050 TCAAAGGATT 1130 TTTGTGATAT 1210 AAGCTCGCAA 1290 ACAGCCGATA 1370 TTGGTTTTTG 1450 TCGATCAATA 1530 AGGATGAGAA 1610 GGTTCATATG CATATC 3' Figure 8. 20 CGTTAGCTTC 100 AACCTAGCTC 180 TCAAGTGCTG 260 AGAGTAGAAT 340 CGACTTCGTA 1.20 ' TCCTCATCGG 500 TTCTCCATTG 580 AGAGGAAGCC 660 GACTAACGTT 740 GTTTAGTTTA 820 TGCTTAGTGT 900 AAATTAAAAC 980 AATTGTTGAA 1060 AAATTAAGAA 1140 TAACTGTTCA 1220 TAAAGTGAAA 1300 CTACCGTCCT 1380 GTATAAACTA 1460 ACGGCCTGAG 1540 GGTGATTCTT 1620 GCTATTTTCG 30 GAAAGACTTT 110 AGGCGGCAAG 190 GTGCACAGAA 270 GCCTTTTTCC 350 CTCGATGTCG 430 TGAGAGCCGC 510 TTCTTCGGGT 590 TATAATGAAA 670 TAAAGGGTAC 750 GTTTTACCAT 830 TGTGATTTGT 910 GTTTTCTTGT 990 AATTTTAATT 1070 AATGATTCCA 1150 ATATGTCAAC 1230 AGAAAAAATT 1310 ATCTCCGCCA 1390 TACCATTCTT 1470 ACTCTACGGT 1550 CTGGCACATA 1630 TTATTGGTGA 523 40 TCGGAATAAC 120 TATCCGGCTA 200 AACTCCTTTG 280 AAGCTTGAAG 360 TCGGGAAGAT _ 440 TGCGAATAGA 520 CATGGTGATG 600 CGAAGGCAAA 680 GAAAATAGCA 760 TGTTGAAATT 840 GAATATGTCA 920 AAAATTGCTG 1000 TTTGTTAAGA 1080 ACTAGTATAT 1160 ACCTCGTTAT 1240 CGGTTCGATC 1320 CGCCACAACC 1400 GAAATCTCAA 1480 TGGTTGCGGA 1560 GTTTGAGTGG 1640 AAGAGTTTCA 50 TCCGCAAAAC 130 TATCCGCATG 210 TAAGAACACA 290 GGCAATGTAA 370 CTCCGTTTTC 450 AAAAAATGGA 530 AAAAAACTTG 610 CAGAAACCAT 690 AACTATTGTA 770 TGGACGACCA 850 AATTTTATTT 930 ATACTGTAGG 1010 AAAAAGACTT 1090 CTGTTTTTAT 1170 TTAGTTTCAC 1250 GAAAACAAAT 1330 CGCAACATAA 1410 GGCCATAACG 1490 GTCTATTGGG 1570 GGCTTGCTAT 1650 GCACTGGATG 60 CTGTTGGAAT 140 GTTAAGGTCA 220 CGTCTGGACC 300 GCCCTTTTAA 380 TGATAATGCC 460 GATGGTGAAA 540 GAAATGGCTC 620 TGGATTGTTT 700 GCTAATTTAG 780 ATATTTTTTA 860 GACGTTTGCG 940 CTATATTTAT 1020 TCTTAGTAGT 1100 AATCACTTTT 1180 TATATATTGT 1260 AGTTACTTAA 1340 CCGGGCCCAT 1420 TAACTGCGGT 1500 CCCTTGATGG 1580 TCCTAAGGCC 1660 ATCAAGAACA 70 TAACGTGAAA 150 ATCCCCGCCA 230 AACCCAATTA 310 GTCTGTTATT 390 AAAAGTTGGC 470 CAAAAGAGGA 550 TATGGCTTGC 630 TGCTATTTTT 710 AATTAGAGGT 790 CAGTCTGGAA 870 TTTTGAACTC 950 TGACACAACC 1030 ATAGTATTTG 1110 AAGAATGATC 1190 TTTGAACATT 1270 CCTCAATTCT 1350 TTTGTGATGG 1430 CAACTTAGCG 1510 GCTAATGGAG 1590 ATGGAGTTGT 1670 ACCCTCCTAA DNA sequence of the clone ABS-lO-O. 80 CAAGGCGACA 160 CAACCAATAC 240 GCCGCAGTGT 320 AGCAAATTTG 400 GTCTTGTAAG 480 GAAGGAAGCA 560 TTAATAGATG 640 GGTATAACAA 720 CGTTGACAAC 800 AAAAAAAGTA 880 TTTGGGAAAA 960 AAAGACGTTT 1040 AAGAAAATTT 1120 AAATAATGTT 1200 CGGTCATCAG 1280 TGTTATATTG 1360 TTCATGGAGC 1440 GCATCCAGAA 1520 AGTCTAGGTG 1600 TCTACAAAAA 1680 CCACGTGGAG 59 because on the average, 25% of the residues of any aligned sequences would be identical. Actually there would be a dispersion around that mean expectation, and a predictable fraction of random cases would be as much as 35% identical (12). The best match for protein similarity searches was 25% identity over 56 amino acid residues. The significance of this match is considered to be "improbable" according to Doolittle (12). The lack of similarity to any sequences in the protein bank could reflect the fact that no similar protein has been sequenced, or, perhaps much of the genomic fragment covers introns which substantially decreased the chances to find similar sequences in the protein bank with reasonable confidence. Figure 8 shows the sequence of the clone ab5-10-0. The reason why those two clones were isolated by using chicken cytochrome b5 gene as a probe is not clear, possibly the very low stringency conditions for hybridization and washing allowed this happened. The sequence comparison showed 45.3% identity in 685 bp overlap between ABS-lO-O and chicken cytochrome b5 gene, and 46% identity in 658 bp overlap between ABS-13-0 and chicken cytochrome b5 gene. In both cases, there is a thirteen nucleotides identity continuously between the compared sequences. 2). Oligonucleotides as probes Oligonucleotide mixtures b5-2, b5-3, b5-5 were used to probe an Arabidopsis EcoRI digest of DNA (figure 9). 60 2 .l_ I 3%: 23.1: . W' 6:6- 1 2.3 4.4— r 2.0“ r 2.3— ‘ 2.0- “r" ! EV ‘1 A B Figure 9. The Eco RI digests of Arabidopsis genomic DNA were probed with oligonucleotide mixture bS-Z (lane A), b5-3 (lane 8), and b5-5 (lane C). (12kb ‘7’ ——-—-§ —-) —> ——D ——e —-b # 7* 35-2 F1 r———-—-4 F a 85-60 ; 85-61 b 2 Figure 10. Sequencing strategy for clone BS-2F1. 61 As can be seen from lanes 8 and C, oligonucleotide mixtures b5-3 and bS-S recognized a common fragment of approximately 2 kb. When both genomic and cDNA libraries were screened with the oligonucleotide mixtures as probes, no plaques hybridized with b5-3, but a few hybridized with bS-S. Those clones isolated by b5-5 alone were not sequenced, because the extremely strong hybridization to the 2.3 kb fragment would be expected to totally mask the signal from the 2.0 kb fragment during screening. The cDNA library was also screened with oligonucleotide b5-2 and two clones were isolated out of 100,000 plaques. The inserts of these two cDNA clones were subcloned into pBluescript plasmids, and the smaller insert, designated b5-2F1, was deleted and sequenced as described (7, 8). Figure 10 shows the sequencing strategy of bS-ZFI. Figure 11 shows the sequences of 85-60 and 85-61, partial sequences of 85-2Fl. Even though this sequence is not complete, and sequenced in one orientation only, the message it carries is probably enough for us to do some computer analysis. Unfortunately, after searching both the DNA data bank and the protein data bank in the same way as that for clone A85-13-0, no significantly similar sequences with known function can be found. The best match from the DNA data bank search is less than 50% identity, and the best match from the protein data bank search is 26.8% over 56 amino acid residues. Neither is informative. The sequence recognized by oligonucleotide mixture b5-2 is marked in figure 10, and it is only 88% homologous to b5-2. 85-60 58 SI 10 CCATGGAATT 90 ATTGAAAACA 170 CAGGGCAGGC 250 TCGAGTATGA 85-61 10 CCTCGTGGGA 80 GACTCAAAAG 150 GGGTAATATG 220 ACACGAAACC 290 AAAGAAGTAT 360 CAATTCATTG 430 TGGATCCAGG 500 TTCTGAAGAT 570 CAACAGAATA 640 NTCAGATATA 710 GGCGTCGTGG 780 TCAACAACAT 850 CCATAGCTTC 920 CTTCTTTGGT 990 GCAAATTTTC 1060 TAATTGTTAT Figure 11. Both are part of the clone 85—2FI. 20 AGATGATTCA CATGATGAAG AGGTATTGAA 100 CTCCAGATCC TTTAGCAGGG GAACAGACAT 180 TAAAGAAGAA cAcnccscs 260 TGATGTAGGA GTCACACAGA GTAGCAGATG 20 TCCAAATGAG 90 CTTGTGCTCA 160 TACGGCTGCA 230 AATTATAGGA 300 GATGGTTACG 370 CAAGGCCGGT 440 GTGTTGTCNM 510 TTCTGAAGGG 580 ATCTTAAAGA 650 TGTTCCACAA 720 TCGCGTGAAG 790 GACGTAGTGT 860 TCTGATTCGA 930 AATTTATGGT 1000 TTCTTTATTG 1070 GAGAAAAATA 30 110 190 270 30 TCTTTACCTC 100 AGCAAGCCCT 170 TATTAAAGAA 240 TTTGGGCTTC 310 AAGCTCCTAT 380 ATTTGCGACC 450 GTTGGCTTCA 520 TCGGATCCTC 590 AGATAATATT 660 CCCCGAAGAC 730 GAACCAGTAG 800 GCATGAACTT 870 ATTTTTTAAT 940 GGTGGTTTAG 1010 CTCACTATAC CTTAAAT 3' AGGACTCAGA 40 120 200 280 40 AGGACTATGC 110 GAAGATGGAA 180 GTCCTCTTGG 250 TGCAGCATGA 320 TAAAACAAAA 390 GACAATTTCA 460 ATATACGGCC 530 GTGCCATTGC 600 AACTGGGTAC 670 GTGAAATGGT 740 GCACACATGG 810 GTACAAGCGC 880 GAAGATGATG 950 TGGAAGTCCT 1020 ATATACATTT DNA sequences of 85-60 62 50 130 210 290 50 TAGAATTTTT 120 GAAGAAGACA 190 TGCTGCCTCT 260 ATCCAAGATG 330 GAAGAGCTAA 400 GCTCAGACAA 470 CCATATCTAT 540 TGCCCTTGGT 610 CCTCAGAGAG 680 TCAAGCCTGT 750 GGCAATGAAA 820 GCTTATCCCA 890 ATGTTCAAGT 960 CCAGTTTTGT 1030 GGAATATGGG and 85- 60 140 220 300 60 GCTTTTGATA 130 GAGATGATTG 200 AACTATCTTC 270 TCGGTTCTGC 340 TGTTCCATGT 410 GCACAAGATG 480 CCCACCCCTT 550 TCCTTGAAAA 620 TATCGAAAAT 690 TGAAGTATGG 760 TGCATATTCA 830 AGTHGCCTGA 900 TTCTCTGAAT 970 CCTGTATGTG 1040 GTTGAACCAT 61. 70 150 230 AGTGACGATA CATGATCCGT AGTA 3' 70 ATGTCGCAAG 140 TGTACCAATT 210 TCTTGTGAAC 280 ATTTTAGCGT 350 TGGTTTCCGT 420 GAGAGATTCC 490 CCTTTGGTAG 560 GCGTAGAATC 630 GAAAGCTTCA 700 TCAAAATGCG 770 GTGGAGTGGT 840 ACGTCTGTAC 910 TGACGACATT 980 ATAAAAATAT 1050 TTGATAAGCT 80 GTCTCTGGTT CCTGATCCCA TGAAACAAGA GCCTTTAGTA 160 GGCCANCAGA GGAAGAATGG CTGAGGCTGA CAAATCAAAG 240 TATAGCTGCT GATGTGATGA CAGTGAGAGA TCGTATGTAT 63 4. Discussion Both the chicken gene as a probe and the oligonucleotides as probes have been been tried, neither succeeded. The use of heterologous probes entails the risk that the approach may not work. Two clones isolated by using the chicken gene as a heterologous probe are not cytochrome b5 genes as judged by computer analysis of homology to known cytochrome b5 genes at the DNA and deduced amino acid sequence levels. The clone A85-13-0 has three open reading frames coding for proteins with MW of 11 KD, 14 KD, and 16 KD respectively. The open reading frame 2 encodes a protein of 11 KD which has 24.6% identity over 130 amino acid residues to Notch protein of Drosophilg (10). Sequence similarity is as high as 66% if we include substitutions of similar amino acids. If the mistakes in the sequence were corrected, the open reading frame 2 could possibly encode a larger protein without decreasing the percentage of the similarity to the Notch protein. Since another slightly higher identity (25.7% over 101 amin acid residues) was found in reading frame 3 which overlaps a little bit with the end of open reading frame 2. I estimated the significance of these two matches according to Doolittle’s method (12). The result is that they are "probably" related, but not "certainly". The product of the Notch locus in Drosophila melanoqaster is involved in a cell interaction which controls the accurate differentiation of certain tissues during development, including the embryonic nervous system (10). Since Notch protein is a much larger protein with 2703 amino acid residues, it is difficult 64 I 5 10 15 20 25 30 AbS-lB-O I N A T D V G G L T V L H R A I I G K K 0 A I T N Y L L R E S A N Tin-12 0 G R T A L H A A N V Y L A CONSENSUS M Notch, N 0 G T L H A A R ~V L A N cdc 10, 0 SN 16 CONSENSUS Figure 12. The sequence of the 33-amino acid motif from Arabidopsis clone A85-13-0 and comparison to the consensus sequences from lin—12 of nematode, Notch of Drosophila, cdc 10 and SH 16 of yeast (13, 14). 65 to see how clone A85-13-0 can encode for a protein with a similar function. However, a 33-amino acid segment within the open reading frame 2 was found to be similar to segments not just in Notch gene, but also in cell-cycle genes cdc 10, SH 16 of yeast and lin-IZ gene of nematode Caggorhabditis eleqans. The 33-amino acid motif repeats 6 times in Tin-12 of C elegggs (13), 5 times in Notch and 2 times in both cdc 10 and SN 16 of yeast (14). Figure 12 shows the sequence of the 33-amino acid motif of the clone A85-l3-0 and the consensus sequences of the motifs from Notch protein, cdc 10 gene, SH 16 gene, and lin-12 gene. As Breeden and Nasmyth predicted (14), this motif is conserved in a set of evolutionarily distant organisms and it will be found in other organisms. The presence of such motifs in genes of animal, plant, yeast and nematode suggests that it plays an important role in the cell. The clone 85-2F1 was not completely sequenced. The gap which separates the sequenced parts, 85-60 and 85-61, is about 200 nucleotides long. But unlike A85-10-0, this is a cDNA clone, and was isolated by oligonucleotide b5-2. The sequence which recognized the oligonucleotide is located on 85-61. The coding sequence of 85-61 is completely different from animal cytochrome b5 except for a region of five amino acid residues of identity from which we derived our oligonucleotide. The clone ABS-2F] is not likely to be a plant cytochrome b5 gene. None of the clones sequenced (or semi-sequenced) were assigned a function. This indicates that the sequence of the plant cytochrome b5 gene has diverged from that of the animal cytochrome b5 gene to the point that other sequences have greater homology to the animal 66 gene at low stringency. After the rabbit and bovine liver cytochrome b5 cDNA sequences were published recently (9, 11), I analyzed the nucleotide homology between the rabbit and bovine and between rabbit and the chicken liver cytochrome b5. Even though the homology between rabbit and bovine is about 77%, the homology between rabbit and chicken is only 63%, whereas the amino acid sequence homology among animal cytochrome b5 proteins is at least above 70%. He can predict that the homology between plant and animal cytochrome b5 will be lower. This is probably why it is so difficult to clone the plant cytochrome b5 by this approach. It would be better to select the "right" clone to sequence, instead of sequencing every possible clone. This can be done by using the candidate clones to hybrid-select mRNA, in vitro translating the hybrid-selected mRNA, using anti-cytochrome b5 antibodies to precipitate the translated protein. Those clones which hybrid-select the cytochrome b5 mRNA will be the right clones to sequence. Jollie et al. (1987) showed that the antibody raised againt the rat cytochrome b5 can cross react with the plant cytochrome b5 (3). This raised the possibility that the conservation of cytochrome b5 might be extended to plant, and might well be at the nucleotide level. Unfortunately, this is a "Horse Behind Artillery" (a Chinese Saying which means It Is Too Late, And lt’s Over). The use of mixed oligonucleotides overcomes the problems of codon usage and should have been a usable approach, if the plant cytochrome b5 amino acid sequence is homologous to the animal proteins. This approach failed because of the failure of the b5-3 probe to hybridize to anything in both genomic and cDNA libraries. The lack of recovery 67 of clones which hybridized to b5-3 could be due to a bias in the library, or some other problems. Additional work would be needed to determine the causes. The 2.0 kb fragment detected on genomic Southern blots by oligonucleotide mixtures b5-3 and b5-5 is the best candidate for plant cytochrome b5 gene. 1). 2). 3). 4). 5). 6). 7). 8). 9). 10. References C. Bonnerot, A. Galle, A. Jolliot, and J. C. Kader (1985). Biochem. J. 226:331-334 K. Madyastha, N. Krishnamachary (1986). Biochem. Biophys. Res. Commun. 136:570-576 0. R. Jollie, S. G. Sligar, and M. Schuler (1987). Plant Physiol. 85:457-462 K. Abe, S. Kimura, R. Kizawa, F. K. Anan, and Y. Sugita (1985). J. Biochem. (Tokyo) 91:1659-1668 T. Maniatis, E. F. Fritsch, and J. Sambrook (1982). Molecular Cloning, a Laboratory Manual. Cold Spring Harbor Laboratory. Cold Spring, New York S. Henikoff (1984). Gene 28:351—359 H. Zhang, R. Scholl, J. Browse, and C. Somerville (1988). Nucleic Acids Res. 16:1220 H. Zhang, and C. Somerville (1988). Arch. Biochem. Biophys. 264:343-347 R.J. Cristiano, and A.w. Steggles (1989). Nucleic Acids Res. 17:799 S. Artavanis-Tsakonas (1988). Trends In Genetics 4:95-100 11). 12). l3). 14). 68 N. Dariush, C.N. Fisher, and A.N. Steggles (1988) Prot. Seq. Data Anal. 1:351-353 R.F. Doolittle (1986). Of Urfs And Orfs. ---A Primer on How to Analyze Derived Amino Acid Sequences. Univerity Science Books. Mill Valley, CA, USA J. Yochem, K. Weston, and I. Greenwald (1988). Nature 335:547-550 L. Breeden, and K. Nasmyth (1987). Nature 329:651-654 CHAPTER V SUMMARY AND SUGGESTIONS The goals of this research initially were: 1. Clone the cytochrome b5 genes from chicken and study its structure, expression, and the relationship of erythrocyte cytochrome b5 mRNA and liver cytochrome b5 mRNA; 2. Use the cloned chicken gene as a probe to clone the plant cytochrome b5 genes, then study its regulation and expression in plant and possibly explore the relationship of its structure and function. I have successfully accomplished the first goal, but failed in the second. The reason for this failure was discussed in chapter IV. The findings of this research are summarized as follows: 1. The cloning of the chicken liver cytochrome b5 cDNA was the first cloned cytochrome bS gene. The availability of the cDNA sequence revealed several new features about cytochrome b5 structure. The chicken cytochrome b5 gene encodes a protein which is-l38-amino acid long, not 133-amino acid residues found for all the sequenced 69 7O cytochrome b5 proteins. More heterogeneities were found at the N- and C-terminal ends. The chicken polypeptide lacks one amino acid at the C-terminus which is present on all the other known sequences. Several errors in the protein sequences were identified by comparison of the sequence deduced from the cDNA sequences and the published amino acid sequences. 2. Chicken erythrocyte cytochrome b5 cDNA was also isolated and found to be identical in sequence to the cDNA from liver, except it was much longer. The cDNA from liver lacked a 3’ poly(A) sequence, possibly due to premature termination during the second strand synthesis from mRNA. Northern Analysis of the size of the liver cytochrome b5 mRNA indicated it was of the same size as the cDNA sequence of erythrocyte cytochrome b5 mRNA. 3. The cloning and mapping of the genomic cytochrome b5 gene suggested that there is only one cytochrome b5 gene in chicken. This implys that all forms of cytochrome bs proteins are produced from the same gene. 4. Cytochrome b5 mRNA is much more abundant in the liver than in erythrocytes, possibly because the liver is a very active tissue biosynthetically. A large quantity of cytochrome b5 may have to be present since cytochrome b5 plays a vital role for many biological reactions. The turnover of cytochrome b5 in erythrocytes might be much slower than that in liver. In this way, the amount of cytochrome b5 can be maintained at levels adequate to carry out all biological reactions therein. 71 Further research may be targetted at: 1. Using site-specific mutagenesis to alter the amino acid residues within hydrophobic region, study the interaction between the mutated protein and the membrane. This might tell us about the functions of amino acid residues within the hydrophobic domain and their contributions toward the two kinds of membrane binding of cytochrome b5. Since our results indicate that it is posttranslationally modified, it might be interesting to attempt to identify the protease. This may represent a new class of protease since it must have substantial sequence specificity and, more importantly, it appears to insert a new amino acid residue to the C-terminal of the proteolytically processed microsomal cytochrome b5 protein. 2. Further characterize the cytochrome b5 genomic clones by completely sequencing them to locate all the introns, exons, and regulatory sequences. 3. Consider additional strategies to clone the plant cytochrome b5 genes. If the cytochrome b5 gene was cloned, it would be possible to study its expression and regulation a in plant systems. It might then be possible to use reverse genetics methods to explore its functions in plant. These methods could include the use of antisense mRNA to explore the mechanisms of organ-specific regulation or developmental regulation, and the use of the cloned gene to over express cytochome b5 so that antibodies can be raised against it. Since cytochrome b5 is about loo-fold less abundant in plants than in animals, it would be useful to have a more abundant 72 source for biochemical studies. Expression in yeast or E.goli might be suitable in this regard. Alternatively, it might be possible to overexpress the gene in transgenic plants. APPENDIX I TRANSFER OF THE MAIZE TRANSPOSABLE ELEMENT Mul INTO ARABIDOPSIS THALIANA ----- Reprint from PLANT SCIENCE, 48(1987), pp 165-173 73 Plant Science, 48 (1987) 165-173 Elsevier Scientific Publishers Ireland Ltd. 74 165 TRANSFER OF THE MAIZE TRANSPOSABLE ELEMENT M U1 INTO ARABIDOPSIS THALIANA HONG ZHANG and CHRIS R. SOMERVILLE' MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing. MI 48824 (U. S.A.) (Received June 25th, 1986) (Revision received October 27th, 1986) (Accepted October 29th, 1986) The maize transposeble elunent Mul was transferred to Arabidopsis thaliana by Ti plasmid-mediated trans- formation and a fertile line containing Mal was regenerated. Southern analysis of transformed ti-ue indicated that the Mu! DNA remained entirely within a segment of T-DNA during three sexual generation. The results of a search for spontaneous mutation in a large number of Mal containing seedlings suggests that the presence of Mu] did not cause a major increase in the spontaneous mutation frequency. Key words: Arabidopsis thaliana; transpoeable element; transformation; Ti plasmid; plant regeneration; Robertson's Mutator Introduction The isolation of genes by transposon tagging has recently been accomplished in several plant species [1—3] . Unfortunately, the application of this potentially powerful approach is currently limited to those few species in which active transposable elements have been characterized at the genetic and molecular levels [4,5]. In order to extend this approach to other species it would be neces- sary to either identify endogenous transpos- able elements or to introduce a suitable element by transformation. Although the latter approach has recently become feasible for many plant species, there are many uncertainties concerning the autonomy of the plant transposable elements and the mecha- nisms involved in transposition [5] . The maize transposable element Mu] is ‘To whom correspondence should be sent. Abbreviations: 2,4-D, 2,4-dichlorophenoxyacetic acid;kbp, kilobese pair: SDS, sodium dodecyl sulfate. 0618-9452/87/803.50 Printed and Published in Ireland thought to represent a member of a family of transposable elements which are present only in certain maize lines which are said to carry Robertson’s Mutator [6,7]. These lines pro- duce spontaneous mutations at a variety of loci at rates which are 30—50-fold higher than non-Mutator lines [7] . The Mul element was identified as the causal agent of spontan- eous unstable mutants at the Adh-I locus in a Robertson’s Mutator line of maize [8]. Determination of the DNA sequence of Mu] [9] revealed an overall structure which is consistent with it being a transposable element. In particular, Mul has inverted terminal repeats and creates direct repeats of target DNA at the site of insertion. Each strand has two open reading frames but no transcription product has been identified. The MuI element has recently been intro- duced into tobacco and tomato by Ti plasmid- mediated transformation [10]. The prelimi- nary results indicated that MuI DNA was present in the tobacco transformants in higher copy number than the T-DNA, and the pattern of restriction fragments which hybridized to Mal were different than those Elsevier Scientific Publishers Ireland Ltd. 166 of the transforming vector. However an unambiguous interpretation of the results was not possible. We report here the results of similar experiments in which Mu] was introduced into Arabidopsis thaliana by Ti plasmid-mediated transformation. A. thaliana was considered advantageous for' such studies because the small size and rapid generation time permitted a facile analysis of the effect of Mal on the spontano eous mutation rate. Materials and methods Plant material and growth conditions The Columbia wild-type of A. thaliana (L.) Heynh. used for these studies was originally provided by G.P. Redei. Plants were generally grown on an artificial potting mixture irri- gated with mineral nutrients at 23°C in natural light. For some purposes sterile plants were grown from surface-sterilized seed in unsealed Petri plates (90 X 23 mm) containing mineral salts supplemented with 10 g/l sucrose and solidified with 7 g/l agar. Bacterial strains and plasmids Agrobacterium tumefaciens C58C1 riff carrying the nononcogenic Ti plasmid pGV3850 [11] was obtained from J. Schell. The plasmid pMJ9 [9], which carries Mu], was obtained from M. Freeling. The plasmid pRK2013 and the intermediate vector pMON200 [12] which carries a bacterial spectinomycin resistance gene, a nopaline synthase gene and a neomycin phosphotrans- ferase (NPTII) II gene was obtained from S. Rogers (Monsanto). The growth of plasmid-bearing bacterial strains and the procedures for triparental matings were as described [12] . The Escherichia coli strain LE392 (T hst supE supF lac gal metB tIpR) was used as the host for all E. coli plasmids described here. Tissue culture media The media used for tissue culture were 75 modifications of those described by Nemtiu et al. [13] for A. thaliana. The basal medium contained the mineral salts of Murashige and Skoog [14] supplemented with 20 g/l sucrose, 0.4 mg/l glycine, 0.1 mg/l nicotinic acid, 1 mg/l thiamine-H01, 0.1 mg/l pyridoxine- HCl, 7 g/l agarose at pH 5.8. Phytohormones were added to the media after autoclaving in the following amounts: MSal contained 1 mg/l 2,4-D, 0.05 mg/l kinetin'; MSa2 con- tained 0.5 mg/l 2,4-D, 0.1 mg/l kinetin; PR29 contained 5.0 mg/l naphthaleneacetic acid, 0.5 mg/l benzyladenine; PR33i contained 1 mg/l isopentenyl adenine, 0.1 mg/l indole acetic acid; MSa4 contained one fifth the concentration of salts, 5 g/l sucrose, normal levels of vitamins but no phytohormones. Transformation of callus tissue Callus was established by placing sterile leaves on MSal for 3 weeks in continuous low light. The callus was then subcultured on MSa2 for at least 2 weeks before use. Approximately 0.5 g of callus was chopped into pieces of about 2 mm diameter and covered with 10 ml of liquid MSa2. Two drops of a fresh saturated culture of A. tumefaciens was added to the cell suspen- sion which was left shaking at 40 rev./min for 20 h at 20°C in low light. At this time the liquid was discarded and replaced with fresh MSa2 containing 250 ug/ ml Cefotaxime (Calbiochem) and the incubation continued for 5 days with slow shaking. The cell slurry was then transferred to solid MSa2 medium containing 250 ug/ml Cefotaxime and 25 ug/ ml G418 (Gibco) and incubated at 23°C in continuous low light. After several subcul- tures on the same medium over a period of 4—6 weeks, the G418-resistant callus was transferred to PR29 for 4 weeks without antibiotic selection then transferred to PR33i to induce shoot formation. The transfer to PR29 before attempting shoot regeneration appears to enhance the frequency of shoots in long-term callus cultures. Shoots were rooted by dipping the base of the shoots in 0.1 mg/ml NAA then transferring the shoots to MSa4. After roots had developed the plants were transferred to potting mixture. Opine assays The presence of nopaline in transformed tissue was determined by the method of Rogers et al. [12]. Nopaline was visualized by grinding single leaves of approximately 0.5 cm2 surface area in 3 ul of water and applying the entire extract to the paper. DNA Manipulations Arabidopsis DNA was prepared as described previously [15]. Southern hybridizations were performed at 68°C for 36 h in 6 X SSC (SSC is 150 mM NaCl, 15 mM sodium citrate, pH 7.0), 0.5% (w/v) SDS, 0.1% (w/v) Ficoll, 0.1% (w/v) polyvinyl pyrrolidone, 0.1% (w/v) bovine serum albumin 0.1 mg/ ml denatured salmon sperm DNA and 10 mM EDTA. The filters were washed at 68°C as follows: 5 min in 2 X SSC, 0.5% SDS; 15 min in 2 X SSC, 0.1% SDS; 2 h in 0.1 X SSC, 0.5% SDS; 30 min in 0.1 X SSC, 0.5% SDS. All other II “II I 14 76 167 nucleic acid manipulations were performed by standard methods [16] . Results Plasmid constructions Mu] was introduced into the non-oncogenic Ti plasmid pGV3850 in two steps. First, the 2.9 kbp BamHI-I-lindIII fragment from pMJ9 was ligated into the BglII—HindIII sites of pMON 200 and transformed into E. coli strain LE392. The resulting plasmid, designated pHZl (Fig. 1), was then introduced into the A. tumefaciens strain C58CI(pGV3850) by a triparental mating in which E. coli strain LE392(pRK2013) provided the mobilization functions [12]. Since pMON200 derivatives do not replicate in A. tumefaciens, a cointe- grate between pHZl and pGV3850 was obtained by selecting for stable expression of the spectinomycin resistance gene carried on pHZl in A. tumefaciens. There are two regions of homology be- tween pMON 200 (or pHZl) and the modified Fig. 1. Map of the Mal-containing intermediate vector pHZi. 77 168 da: 5 Soon—mac c0323.? magnoneoang 23 go on: 23 3335 Sam: 2: 36:5 £50.: 25. .cmmm>0n 35 52.. no 5:335 a: .3 ...: 2 35595 ESE 30:... 3 35¢. a :92; «$me .3 982%. :25 8.. <29... 2: no so... 523 5 I: 84:50.. so new: «.29.. 3:3... 3... A5 4533 8825.. 253a assesses 2: 5 3: no Essence: 2:. .«sfi ell-ulsteTe 0 «a v Tile-Ills e z... e 8.. .3 8.. .33.... .. 2.. ....n 3.8 .3 Ila" SBJfiaamalnlelsll I al TH ...-UBWLI a? a h. t ad a _ _ =s _ _ H mm m a x mm” mm Em m 8. .3 zeal: ... all~ ......lu Ll s_HIII_| I: . . _ ... m m m m m m ... < T-DNA region of pGV3850. Both plasmids contain the nopaline synthase gene and the origin or replication of pBR322. Thus, if cointegration occurs by a single crossover between the two circular plasmids, there are two possible configurations for the organiza- tion of DNA sequences in the cointegrate pGV 3850 ::pHZ1. Southern hybridization analysis of DNA from C58Cl(pGV3850:: pHZl) using the internal ’I‘th111I fragment from MuI as a probe was consistent with the orientation shown in Fig. 23 (results not presented) in which Mill is positioned be- tween the left and right border sequences of p’l‘iC58 carried on HindIII fragments 23 and 10, respectively [11]. Thus, the configu- ration is that expected if cointegrate forma- tion occurred by a single recombinational event between the regions of pBR322 homo- logy on pHZ1 and pGV3850. Transformation of Arabidopsis Previous studies had shown that A. thaliana could be transformed with tumorigenic strains of A. tumefaciens [17]. However, as there were no previous reports of transformation of A. thaliana with a disarmed Ti plasmid we explored several approaches to transform- ation. We initially observed a very low frequency of transformation of A. thaliana by the leaf disc procedure used with other species [12] but obtained a satisfactory frequency of transformants with minor modifications of the methods used by Muller et al. [18] to transform tobacco callus. By this method, 5—7% of the calli exposed to A. tumefaciens C5801(pGV3850: :pI-IZ1) developed outgrowths which were resistant to at least 25 ug/ml G418. We obtained a very low frequency of shoot regeneration from the G418-resistant calli using published procedures [13]. We are uncertain whether this is due to the properties of the accession of A. thaliana we have used or some other reason. However, it has previ- ously been noted that the regeneration poten- tial of A. thaliana callus declines sharply after several months in culture [13]. Whatever 78 169 the case, from several dozen G418-resistant callus lines we obtained only five shoots and from these we recovered only one fertile plant which accumulated nopaline (results not presented). This line, which was very vigorous, was designated MSU252 and the original regenerant is designated the R1 generation. Analysis of Mu1 DNA in transformants Southern Analysis of the restriction pattern of DNA from G418-resistant callus tissue using the internal TthlllI fragment fromMuI as a hybridization probe indicated that the G418-resistant calli contained uences homologous to Mu] This is evident from the results in Fig. 3 in which the combined DNA from eight independently transformed G418- resistant calli was cleaved with EcoRI to Fig. 8. Southern blot of DNA from wild-type and G418-resistant tissue probed with Mul DNA. The DNA from all sources was restricted with EcoRI. The position and size (kbp) of molecular weight markers is indicated at the left side of the figure. Lanes: (A) DNA from eight G418-resistant calli; (B) DNA from leaf tissue of an R3 plant descended from a G418 resistant callus; (C) wild-type A. thaliana; (D) A mixture of equivalent amounts of DNA from A. tumefaciens 05801 (pGV3850itMul) and the DNA preparation in lane 3. 170 release the Ma] DNA and surrounding sequences on a 6.8 kbp fragment (Fig. 2). DNA from leaf tissue of the wild-type showed no homology to Mu] under the conditions used. The only band of homology in DNA from the callus tissue was the same size as the major band of homology in DNA from A. tumefaciens carrying the Ti-plasmid pGV3850::pHZ1, and from leaf tissue of a single R3 plant descended from a transformed regenerant. Thus, it is inferred that the Ma] element did not rearrange or transpose from the surrounding vector DNA during the early cell divisions following the initial transforma- tion events that gave rise to the eight callus lines. A more detailed analysis was conducted on DNA from plants of the R2 and R4 genera- tions of the Mal-containing line MSU252. DNA from individual plants or the combined DNA from 30 plants was cleaved with Ncol, EcoRI and HindIII transferred to nitrocellu- lose and probed with the Tthllll fragment from Mu] (Figs. 4 and 5). The apparent mole- cular weights of the fragments which hybri- dized with the probe are consistent with those expected from the analysis of the orientation of Mu] sequences in pGV 3850 :: pHZl (Fig. 2). Thus, when cleaved with EcoRI, a single 6.8 kbp fragment showed homology to Mill When the DNA was cleaved with NcoI, which cuts within Mal, two fragments of 1.6 and 2.8 kbp were the most prominent bands, and when cleaved with HindIII the major signal was a 12.8 kbp band. A less intense signal corresponding to a HindIII fragment of about 4.0 kbp in the HindIII digests of the transformants was also found in digests of A. tumefaciens C58Cl- (pGV3850) (results not presented). On the basis of this observation, and considering the size and intensity of this band, it is probably due to contamination of the gel- purified probe with vector sequences which hybridize with the pBR322 sequences present in the T-DNA region of pGV3850. Thus, there was no evidence that MuI DNA has rearranged or moved from the adjoining 79 ABCDEFG 23.1 — 9.4— Q . 6.6- . ..._ V we - - - 2.3 — 2.0—— Fig. 4. Southern blot of DNA from transformed plants of A. thaliana probed with Mal DNA. The position and size (kbp) of molecular weight markers are indicated at the left side of the figure. Lanes: (A) wild-type A. thaliana digested with EcoRI; (B, C and D) DNA from 30 R2 plants digested with NcoI, HindIII and EcoRI, respectively; (E, F and G) DNA from one R2 plant digested with NcoI, Hind III and EcoRI, respectively. vector sequences during the four generations following transformation. Inheritance of introduced DNA On the basis of the intensity of the hybridie zation signal obtained with Mu] DNA as the probe, the Mulocontaining line MSU252 carried approximately 2—5 copies of Mul (results not presented). The inheritance of the introduced DNA in line MSU252 was followed for four generations by scoring for nopaline production or antibiotic resistance. Twenty-eight randomly chosen R2 progeny resulting from the self-fertilization of the R1 plant were examined for the presence of nopaline and were all found to be positive. Since a single insertion of T-DNA would be expected to behave as a single dominant mendelian locus, this observation suggested ABCDEFGH Fig. 5. 80 Southern blot of DNA from a variegated Mal-containing plant. Lanes A to E were digested with EcoRI. Lanes F to I were digested with HindIII. (A) pMJ9; (B and F) DNA from a single variegated R4 plant;(C, D, G and B) DNA from single non-variegated R4 plants; (E and I) wild-type A. thaliana. that the original transformation event was more complex. One hundred R2 progeny were also scored for G418-resistance by germinating surface-sterilized seed on agar- solidified mineral medium to ensure that spurious problems with seed viability did not bias the results, then transferring the seedlings to minimal medium containing 25 ag/ml G418. Ninety-seven R2 progeny were G418- resistant and three were sensitive. This pattern of inheritance is an acceptable fit to the hypothesis that resistance is due to two unliked dominant nuclear genes which were present in heterozygous condition in the original regenerant(x2 = 1.80;P> 0.1). The R3 progeny resulting from self- fa'tilization of nine R2 individuals were also scored for G418-resistance in the same way. Two lines had all G418-resistant progeny. The progeny of two other lines had segre- gation ratios which were excellent fits to a 3:1 (resistant/susceptible) pattern, and the other five lines had patterns of resistance which were an acceptable fit to a 16:1 (resistant/susceptible) pattern of inheritance (results not presented). These results are also consistent with the proposal that the original regenerant was heterozygous for two unliked copies of the NPTII gene each of which could confer G418-resistance. In order to obtain a line of A. thaliana with simple inheritance of G41&resistance a number of the R3 individuals from the two lines showing 3:1 segregation of the resistance marker were self-fertilized and the resulting R4 progeny scored for G418- resistance. This resulted in the recovery of two lines, designated MSU253 and MSU254, which produced only G418-resistant R5 progeny and are, therefore, presumed to be homozygous for a simply-inherited G418- resistance determinant. Spontaneous mutation rate In order to examine the possibility that the Mul element might enhance the spontan- eous mutation rate we examined approxi- mately 30 000 R3 seedlings descended from self-fertilization of 227 phenotypically wild- type R2 individuals for mutations affecting chlorophyll content. Ten variegated indivi- duals were observed in the R3 progeny whereas none were observed among a similar number of wild-type seedlings. No other obvious mutant phenotypes were observed among the 30 000 R3 seedlings. Thus, aside from the variegated phenotype, no increase in mutation frequency was observed in this experiment. A recessive nuclear mutation which causes 172 variegation has previously been recovered from a plant regenerated horn untransformed callus tissue of A. thaliana [19]. Similarly, in the R2 generation of line MSU252, four of 231 individuals were variegated. Most and possibly all of the progeny resulting from self-fertilization of the variegated individuals appeared to be either variegated or completely chlorotic. Several progeny from these plants did not appear to be variegated, but because the amount of chlorotic tissue varied widely hem individual to individual, small variegated sectors may have been missed on these individuals. The possibility that the variegation was directly related to Mill was investigated by Southern analysis of DNA from leaf tissue of a heavily variegated plant. However, no difference was observed in the restriction pattern of the sequences with homology to the Tth1111 fragment from MuI (Fig. 5). Thus, the variegation does not appear to be due to a rearrangement of MuI DNA in the A. thaliana genome. Discussion Bennetzen [6] has estimated that in order to account for the observed increase in spontaneous mutation rate in Robertson’s Mutator lines of maize, each of the large number of Mu] sequences present in these lines would have to transpose, on the average, once every plant generation. Since we did not observe any evidence for a transposition event among more than 30 individuals separated by two or three generations from the original transformant it appears that, if Mu] is an autonomous element, the frequency of transposition is markedly reduced in A. thaliana as compared with maize. In this regard our results substantiate similar experi- ments in which Mu] did not appear to trans- pose in tomato [10]. In order to examine the possibility of Mu] transposition by genetic criteria we examined a relatively large number of seedlings for spontaneous mutations. There are more than 81 100 loci in both barley and maize at which mutations giving rise to chlorophyll-deficient phenotypes are known to occur. Thus, the frequency of chlorotic mutations is a rela- tively sensitive indicator of mutation rate. Since no chlorotic mutations were observed in approximately 30 000 seedlings it is apparent that the presence of Mu] DNA did not cause a major increase in the spontaneous mutation rate. The only genetic anomaly observed was the occurrence of variegated plants in the Mal-containing line. However, as we did not observe any rearrangement of the Ma] DNA in variegated leaf tissue, the variegation appears to be an unrelated pheno- menon which may, nevertheless, be relevant in another context to the search for a trans- posable element in A. thaliana. As with most negative results this experi- ment does not provide any information as to the reason that Mu] does not appear to transpose. As noted by others [20] it may be that MuI is not the autonomous element in Robertson’s Mutator lines or host-specific factors may be required. Alternatively, since the frequency of transposition of Ala] is correlated with copy number [9] it may be that the copy number of Mu] in the trans- formants we obtained is too low. Whatever the case, on the basis of the results presented here we do not consider Mu] a promising candidate for a broad-host-range plant transposon. As there have not been previous reports of transformation of A. thaliana with non- oncogenic Ti plasmids, a few comments on the methodology seem appropriate. First, it should be noted that the protocol described here for obtaining transformed plants is relatively slow and the conditions we have employed may not be optimal. The methods recently developed by others appear to substantially more efficient [21]. Second, although the NPTII gene from pMON200 does confer resistance to G418 in A. thaliana, there is a relatively narrow range of anti- biotic concentration at which the wild-type tissue does not grow and the transformed tissue will grow. Thus we believe that alter- nate selectable markers which provide greater discrimination between transformed and non-transformed tissue would be very useful. Acknowledgments We thank R. Horsch for advice concerning the transformation of Arabidopsis and J. Schell. M. Freeling, W. Taylor, J. Bennetzen and S. Rogers for the generous gifts of plasmids. This work was supported in part by grants from the National Science Foundao tion (No. PCM8351595) and the US. Depart- ment of Energy (ACOZ-76ER01338). References 1 N. Fedoroff, D.E. Furtek and O.E. Nelson Jr., Proc. Natl. Acad. Sci. U.S.A., 81 (1984) 3825. 2 C. Martin, R. Carpenter, H. Sommer, H. Saedler and ES. Coen, EMBO J., 4 (1985) 1625. 3 C. O’Reilly, N.S. Shepherd, A. Pereira, Z. Schwarz-Somrner, I. Bertram, D.S. Robertson, P.A. Peterson and H. Saedler, EMBO J., 4 (1985) 877. 4 M. Freeling, Ann. Rev. Plant. Physiol. 35, (1984) 277. 5 P. Nevers, N.S. Shepherd and H. Saedler, Adv. Bot. Res., 12 (1986) 103. 6 J.L. Bennetzen, J. Mol. Appl. Genet, 2 (1984) 519. 7 D.S. Robertson, Mutat. Res., 51 (1978) 21. 10 11 12 13 14 15 16 17 18 19 20 21 82 173 J.N. Strommer, S. Hake, J.L. Bennetzen, W.C. Taylor and M. Freeling, Nature (London), 300 (1982) 542. R.F. Barker, D.V. Thompson, D.R. Talbot, J. Swanson and J.R. Bennetzen. Nucleic Acids Res., 12 (1984) 5955. M. Lillis, A. Spielmann and R.B. Simpson, in: M. Freeling (Ed. ), Plant Genetics, Alan R. Liss, New York, 1985, p. 213. P. Zambryski, H. Joos, C. Genetello, J. Leemans, M. Van Montagu and J. Schell. EMBO J., 2 (1983)2143. S.G. Rogers, R.B. Horsch and R.T. Fraley, Methods Enzymal., 118 (1986) 627. . I. Negrutiu, F. Beeftink and M. Jacobs, Plant Sci. Lett., 5 (1975) 293. T. Murasbige and F. Skoog, Physiol. Plant, 15 (1962)473. L.S. Leutwiler, B.R. Bough-Evans and E.M. Meyerowitz, Mol. Gen. Genet, 194 (1984) 15. T. Maniatis, R.F. Fritsch and J. Sambrook, Molecular Cloning, Cold Spring Harbor, 1982. M. Aerts, M. Jacobs, J.P. Hernalsteens, M. Van Montagu and J. Schell, Plant Sci. Lett., 17 (1979) 43. A. Muller, T. Manzars and P.F. Lurquin, Bio- chem. Biophys. Res. Commun., 123 (1984) 458. ‘ J. Martinez-Zapater, S.C. Somerville and GR. Somerville, in: M. Freeling (Ed. ), Plant Genetics, Alan R. Liss Inc., New York, 1985, p. 828. L.P. Taylor and V. Walbot, EMBO J., 4 (1985) 869. AM. Lloyd, AR. Barnason, S.G. Rogers, M. Byrne, R.T. Fraley and R. Horsch, Science, 234 (1986) 464. APPENDIX II TRANSFER OF THE Ac ELEMENT T0 ARABIDOPSIS BY SEED TRANSFORMATION 1. Introduction The Ac transposable element is an autonomous element in Maize (1, 2), and has been successfully used to clone bz locus by transposon tagging (3). Furthermore Ac was recently shown to excise from chromosomal DNA in a heterologous transgenic plant system (4). Thus, Ac is a good candidate to be put into Arabidopsis where there is no endogenous element available for transposon tagging. Although it was posssible to obtain transgenic plants (Appendix I), Arabidopsis transformation had been very difficult until recently. In order to try and improve the frequency, Feldmann developed the seed transformation protocol (5). With this simple protocol, transformation seems extremely easy. Germinating seeds of Arabidopsis were cocultivated with an Aqrobacterium for 24 hrs before being planted in pot. Seedlings resulting from this treatment were called T1 generation. The selection is applied in the T2 generation on a selective petri dishes. Transformants will grow with suitable antibiotics selection. 83 84 The following experiments were carried out in an attempt to transfer Ac to Arabidopsis by this transformation protocol. 2. Materials and methods Plasmid pAc9 contained intact Ac element, and was from N. Fedoroff (Carnegie Institution of Washington). Vector pBINl9 (6) was from M. Bevan (Plant Preeding Institute, Cambridge). Agrobacterium strain LBA 4404 was from M.D. Chilton (Ciba-Geigy Biotecnology). Arabidopsis seeds (NS race) were from Feldmann ( Sandoz Crop Protection). Aqrobacterium transformation protocol: Grow 2 ml of LBA 4404 in YEPa plus 50ug/ml rifampicin overnight at 28°C Innoculate 50 ml YEP plus rifampicin with entire 2 ml culture Continue to shake vigorously at 28°C for 4 hrs (00600>l.0) Pellet in sterile Oak Ridge tubes at 7000 rpm 4°C for 10 minutes Nash each tube in 2.5 ml of 150 mM NaCl (ice cold) Pool in one tube and pellet again Resuspend in 0.5 ml of ice cold 75 mM CaCl Add 0.2 ml cells to 20 ul pBIszAc DNA (10-20 ug) in Eppendorf tube Leave on ice 30 min, then freeze in dry ice/ethanol 5 min Thaw at 37°C for 3 min, then innoculate into 5 ml of YEP Incubate at 28°C 1 hour, then pellet, resuspend in 1 ml YEP 85 Plate entire transformation mixture over 4-8 selective platesb a YEP (per liter): b Selective Plate (per ml) 10 g peptone 50 ug Kanamycin 10 g yeast extract 50 ug Rifampicin 5 9 NaCl 200 ug' Streptomycin Arabidopsis seed transformation protocol: Surface sterilize seeds with 30% bleach plus lul/ml of 20% . Triton-X100 Imbibe 3000-5000 seeds in 50ml of Murashige and Skoog (MS) salts (8), 4% sucrose, vitaminc for 12 hrs, Shake imbibition mix, maintain temp at 22°C in dim light (about 1000 lux) Add 3-5ml overnight culture of Agrobacterium (3-5 X 109 cells of LBA 4404::Ac) into the imbibition mix Shake resultant culture for 24 hrs at 28°C, in dim light Nash seeds on a Buchner funnel with water, then leave seeds on 3MM papers Allow them to dry in a hood until just dry enough to plant Plant about 200 seeds in one pot, place all pots in 4°C for 24 hrs Grow plants in greenhouse to harvest and collect seeds in bulk, usually get about ISO—200 seeds per plant Plate T2 progeny seeds on medium containing 100 ug/ml of Kanamycin 86 CVitamin (per liter): 10.0 mg Thiamine 0.5 mg Pyridoxine 0.5 mg Nicotinic Acid 100.0 mg Inositol All DNA manipulations were performed according to the methods in molecular cloning (7). 3. Results: 1). Transforming vector construction pAc9 plasmid (pKP32::Ac9) was cut with SalI, and ligated into pBIN19 at SalI site. See figure 1 below. Ac9 (4.8 kb) pKP 32 (3.5 kb) pBlN 19 (10 kb) Figure l. The construction of the transforming vector pBIN19::Ac. 87 2). Aqrobacterium transformation The transforming vector pBIN19::Ac made from above construction was directly transformed into Aqrobacterium LBA4404. Three KamR transformants were obtained. However only one showed to have the unrearranged pBIN19::Ac plasmid. See the expected restriction hybridization pattern in figure 2 below. IX 8» (I E) E? F (3 ti - ...... —. ' III! thL I III we poem 00 Figure 2. Restriction pattern of two selected transforming Agrobacteria. Strain l (A, C, E and G) and strain 2 (B, D, F and H) were digested with enzymes Pst I (A and 8), Sal I (C and D), Hind III (E and F), and Bgl I(G and H). Only strain 2 gave the expected pattern. 3). Arabidopsis Transformation The Aqrobacterium which contained the correct pBIN19::Ac vector was used to infect approximately 20,000 germinating seeds of Arabidopsis Ns race according to the protocol developed by Feldmann. From approximately 300,000 ‘T2’ progeny of treated seedings, about 30 appeared to be more resistant than wild type to grow on selection plates. Figure 3 showed some resistant ‘T2’ seedlings on Kanamycin 88 Figure 3. An example of some resistant T2 seedlings on Kanamycin plates. 89 plates. However, only 5 seedings have survived after transplanting from the selective medium to potting mixture. I isolated the DNA from the most vigorously growing putative transformant, and cut it with EcoRI and HindIII, then hybridized it to Ac element. Unfortunately, the hybridization pattern did not match the expected pattern. See figure 4 below. The result simply can not be explained. 23. '— Figure 4. Restriction pattern of one putative transformant. Lane A, Ac element alone. Lanes 8 and C, wild-type Arabidopsis WS race and Columbia race cut with Eco RI. Lanes 0 and E, the putative transformant cut with Eco RI and Hind III respectively. 4. Discussion The result in figure 4 can not be explained by premature breakage during single stranded T-DNA transfer, because within 4 kb there is no other EcoRI site from either side of the EcoRI site in the Ac construct (see figure 1). Where did the 2.6 kb EcoRI band from? We can only assume that there was a strange rearrangement which happened 90 after T-DNA transfer (or just before). I have to believe that this putative transformant must have some DNA sequences which hybridize to Ac element, because the control DNAs (both Columbia and Ws wild type of Arabidopsis) did not show any signals at all! If they are true transforants, the transformation frequency would be 1/10,000 or l/60,000 (from 30/300,000 or 5/300,000), which is much lower than 4/1,000 Feldmann originally reported (5). Nevertheless, the technical simplicity of this seed transformation method seems very atractive, even though there are tremendous amount of work involved. 5. References 1). H.-P. Doring and P. Starling, Ann. Rev. Genet. (1986). 20:175-200 2). N. V. Fedoroff, Cell (1989). 56:181-191 3). N. Fedoroff, D. B. Furtek, and 0. E. Nelson, Jr (1984). Proc. Natl. Acad. Sci. USA 81:3825-3829 ' 4). 8. Baker, J. Schell, H. Lorz,‘and N. Fedoroff (1986). Proc. Natl. Acad. Sci. USA 83:4844-4848 5). K. A. Feldmann, and M. 0. Marks (1987). Mol. Gen. Genet. 208:1-9 6). M. Bevan (1984). Nucleic acids Res. 12:8711-8721 7). T. Maniatis, E. F. Fritsch, and J. Sambrook (1982). Molecular Cloning, a Laboratory Manual, Cold Spring Harbor Laboratory, Cold Sporing Harbor, New York 8). T. Murashige, and F. Skoog (1962). Physil. Plant. 15:473-497 APPENDIX III SEARCHING FOR THE DESATURASE GENE IN ARABIDOPSIS AND AGMENELLUM I. Introduction A cDNA clone coding for rat liver stearoyl-CoA desaturase gene had been isolated and sequenced by Thiede et al (1). Unfortunately, this clone did not recognize any sequences in Arabidopsis by genomic Southern analysis (Hugly, personal comm.). Martin et al cloned and sequenced the yeast stearoyl-CoA desaturase gene by complementing the 91g 1 mutation in yeast. (personal comm.). By comparing the coding sequence of the yeast gene with that of the rat gene, regions of homology can been found (figure I). We believe that these homologies represent the conservation of the functional domains of stearoyl-CoA desaturase, and may extended to plants and cyanobacteria. We designed mixed oligonucleotides from these conserved regions and examined their use as probes to clone the plant and blue—green desaturase genes. In addition, we also used the yeast gene as a heterologous probe, because, from preliminary Southern blotting, the yeast gene seemed to recognize a few sequences from Arabidpsis (figure 2). 91 92 45 55 65 75 85 95 105 N- *ID*RPXSKG*LCQQWHFDEVDLTEANILAIGLNKKAPRIVNCFGSLMGSKEMVSVEFDKKGNEKKSNID * * i * - LNTHPVRQEGRFPSAAPPHIISAIGK*SBQPTAEMPAHMIQEISSSYTTTTTTTEPPSGNIQNGREKMKK 10 20 30 40 50 60 70 115 125 135 145 155 165 ' 175 RLLEKDNQEKEEAKTKIHISEXPWTINNWHQHIN‘WLNMVLVCGMPMIGWYFALSGKVPIHINVFIFSVF * g g t ** ** * *g:*;: g g * *: ;* VPIXIEEDIRPEMREDIHDPSYQDEEGPPPKIEYVWRNIILMRLLHVGALXGITLIPSSKVYFTILNGIF 80 90 100 110 120 130 140 185 195 205 215 225 235 245 YYAVCGVSITAGYHRIWSHRSYSAHWPXRLFYAIFGCASVEGSAKWWGHSXRIHHRYTDTLRDPYDARRG ** g : **** *******;* *3 * *g* t * g * **;:::* ** *** YYIISALGITAGAHRLWSHRTYKARLPIRIFLIDNIDflUK3flIhHafluuiflUQEHG§flIfiflu3§flflfififi3 150 160 170 180 190 200 210 255 265 275 285 295 305 315 I3nEflDIWWflIBENPKYKARADITDMTDDWTIRr-FQXXHY--IILMIII%£VUIHEICENTTNDYMG ;:**:**;*:; * * : **°* 0 ** * -*** :*:;***; * g 0 * FFFSHVGWLLVRKHPAVKEKGGKIEMBDLKAEKINMFQRRYYKPGIIlMé-CFIIETINGWDKJK§NHHQ 220 230 240 250 260 270 280 325 335 345 355 365 375 385 GIIYAGFYDPTKIRVFVIQQATFCINSMAHYIGDQPFDDRRIPRDNWITAIVTFGEGYHNFHHEFPTDMR :: ;* 3; :ii: ;** **/\* *;* *g* g g : *** **: ** ** ** HSLF--VSTTIR8TININATWINNSAAHIXGYRPYDKNIQSRENIINSIGSVGEGFHNYHHAFPYDYS 290 300 310 320 330 340 350 395 405 415 425 NAI-KWYQVI--IYIESINCEAHIKBWESQNAIEEALIQQEQKKINKKKAKINWGP -C :***** ** * *: ASEYRWHINFTTFFIDCMAALGIAXDRKKVSKAAVIARIKRIGDGSHKSS*VLG'-C 360 370 380 390 Figure 1. Comparison of the deduced amino acid sequence of yeast stearoyl- CoA desaturase gene with that of the rat stearoyl- -CoA desaturase gene. Upper strand is yeast sequence. 93 I234 567 Figure 2. Arabidopsis genomic DNA digests were probed with p433 (A) and p403 (8) both of which contain parts of the yeast stearoyl-CoA gene. Lanes 1, 2, 5 and 6 are Columbia race; lanes 3, 4, 7 and 8 are WS race. Lanes 1, 3, 5 and 7 are Eco RI digests,; Lanes 2, 4, 6 and 8 are Bam HI digests. His DIT CAC ' His 01A CAC Glu 02 GAA Leu D3 CTI Arg AGI Arg AGI Gly 661 Ala GCI Leu CTI Leu CTI Phe TAC TT Tyr TAC Trp Ser TGG TGI C Trp Ser TGG AGI C His Asn CAC AAC T T Asp Arg GAC IGI His CAC His CAC Tyr AT Lys Arg AG Arg AG His CA Lys Val AAA TT Figure 3. Oligonucleotides designed from the conserved regions of the stearoyl—CoA desaturase. 94 2. Materials and methods: The plasmids p433 and p403 both containing part of the yeast desaturase gene were from D.C. Martin (Rutgers, New Jersey). The conditions for oligonucleotide hybridization, southern blotting, and sequencing were essentially the same as in chapter II. The heterologous probing using yeast gene to screen Arabidopsis libraries was carried out under conditions exactly like that used in oligonucleotide hybridization. The Arabidopsis lambda gth cDNA library was from N. Crawford (stanford, CA). The Arabidopsis lambda gtIO A1 library was made by digesting Arabidopsis genomic DNA with EcoRI, and collecting fragments from 0.4-3.5kb, then ligating them into lambda gth arms. I made an Agmenellum lambda gtIO Agl library the same way as A1 libyary, except I collected fragments from 2.7-5kb. An Agmenellum lambda 2000 library was made by completely digesting Agmenellum DNA with EcoRI, then directly packaging them into lambda 2000 arms. The 5 original titers for all these libraries were at least above 10 pfu. The oligonucleotides designed from the homologies between the yeast and rat desaturase genes are shown in figure 3. 3. Results 1). Yeast desaturase gene as a probe We screened both an Arabidopsis cDNA library and the A1 genomic library by using the yeast desaturase gene as a heterologous probe. 95 10 20 30 40 50 60 70 5' ACATCGATCC AAGAGATGTT TAGGCGGGTG AGCGAGCAGT TCACTGCTAT GTTCAGGAGG AAAGCTTTCT 90 100 110 120 130 140 150 CACAGGTGAA GGAATGGACG AGATGGAGTT TACTGAAGCT GAGAGCAACA TGAACGATCT AGTCTCAGAG 170 180 190 200 210 220 230 ACCAAGACGC AACTGCAGAT GACGAAGGCG AGTATGAAGA AGACGAGGAT GAAGAAGAGA TATTGGATCA 250 260 270 280 290 300 310 AAAAGAGCTG ATATTACCGA TTTTTAAATA CCTCTCTTAT CTTCTTTTCG TTTGGTCGGT ATATGTTTAT 330 340 350 360 370 380 390 80 TGCATTGGTA 160 TACCAGCAAT 240 TGAGTGAGTG 320 TGAGTTTCAT 400 GTATTGTTTG TGATGGTCTG TGTGTAATAG TGAGGTCGGA TCTAAACTTT TATGCGTGGT TTTTATATGA AATTGTCCAT TGTGGTT 3' Figure 4. DNA sequence of the clone 02-62. ——> —-> —5 £3 Xi; H xp P E 0.4 kb . . . . . t —: u——-—a A1 -5-O A1 -5-44 Figure 5. Sequencing strategies for clone A1-5. 96 I isolated 1 clone from the cDNA library and 7 clones from the Al library. Among the 7 clones from A1 library, only three are independent fragments because they migrated differently during electrophoresis (results not shown). The clone from the cDNA library was named 02-65, because it was recognized by 02 oligo also. The insert was excised by digestion with EcoRI and cloned into the EcoRI site of pBluescript to produce plasmid pD2-65. This clone is a very short sequence, only 410 bp. The deletion and sequencing were done as described in Methods and Materials section of previous chapters. The DNA sequence of this clone is shown in figure 4. After we searched the protein bank (PIR R18.0 September 1988) with all three reading frames of this clone, to our surprise, reading 3 was found to encode beta-Tubulin. The homology between the coding sequence of 02-65 and sea urchin beta-Tubulin is as high as 85% in 73 amino acids overlap. The beta-tubulin gene family of Arabidopsis has been characterized by Oppenheimer et al (2). However, the clone 02-65 appears to represent a new member of the gene family, since the published sequences do not completely match the sequence of D2-65. When I compared the sequence of clone 02-65 with that of the yeast desaturase gene, I found that there is a 18 nucleotides identity continuously between the two sequences. The oligonucleotides DZ has 77% identity (14/18) to the sequence from nucleotides 149 to 167 of clone D2-65. That is why the clone D2-62 was picked up by yeast desaturase gene and oligonucleotides 02. A clone designated A1-5 among the 7 isolated from Arabidopsis A1 library has an insert of 1.6 kb, and was chosen for sequence analysis 97 A1-5-0 10 20 so 40 so 60 70 so 5' CTCAGTTTGG TGAAGTCGAC ATTGATGGAT CACTTGTGGC AGCTCAAACT GCGGGACGGA GGATATAATG ATGCTTAACA 90 100 110 120 130 140 150 160 ATGGCTGTCT GTGTTGTACT GTTAGGGGTG ATCTTGTGAG GATGATTTCT GAAATGGTCC AGACCAAGAA AGGAAGGTTC 170 180 190 zoo 210 220 GACCATATCG TTATTGAGAC GACAGGTTCA ATATTTCTAT CACTCTGGAG CTACATGATT CGTTAAGAA 3' Al-5-44 10 20 so so so so 70 so 5' TTCTATGCTG AGGATGAAAT TTTCAATGAT GTCAAGCTGG ATGGGGTTGT CACTCTGGTT GATGCTAAAC ATGCTCGTTT 90 100 110 120 130 140 150 160 GCATCTAGAT GAGGTCAAAC CTGAAGGCTA TGTCAATGAG GCGGTTGAAC AAATAGCTTA CGCGGATCGT ATCATTGTTA 170 130 190 zoo 210 220 230 240 ACAAGGTATT GTGAAGTTCT ATTTCGACCT AATTTTCATG ATTAACATCA CCATAACTTG TATTTGAAAT GTGTGGTGGA 250 260 270 280 290 zoo 310 320 TGCTTATAGT AGAAGCTTTT CAATGGTCTG CTATGTATGT TCTCCTCGAC ATATCATACA AGCTGAAAAT TGTTTGTACT 330 340 350 360 370 330 390 400 TTGTTGTAGA CTGATCTTGT TGGTGAGCCA GAACTAGCTT CAGTGATGCA GCGGATAAAG GTAGGTTCTG CTTCTTCTGA ‘10 420 430 440 «so «so 470 480 TTCTTTCTTT TTTCTCTTTG AACATCTATT TTTCTCTGTG CACTGCGTTC AGTTTATGAT GTGGTTATTT CCAGACCATA 490 500 510 520 530 540 550 560 AACCAGCATG GCTCACATGA AGCGGACAAA GTATGGGAAG GTTGACTTGG ATTATGTTCT TGGAATTGGA GGGTTTGATC 570 530 590 600 610 620 630 640 TAGAAAGGTC CGGTTTTATT TTACGTCTTA CTAAAAATCT TATTACGAGT CACTCCGAAA ACTTCTGTCC AAAGTGAATT 650 660 670 680 690 700 710 720 GTAGCTCCTT ACATTTGTTA TGGTTTATTG GTTGTGGATT TCCTGGTTTA AGAGGGCTGA TTTTGTTACT CTGATTTATT no no 750 760 no no 790 300 GCTGGTGGCC AACTATTTTA TCTTTCCAGA ATTGAAAGCT CTGTGAATGA AGAAGAGAAA GAAGATCGCG AGGGTCATGA 810 820 830 340 850 860 370 880 TGATCATCAC CATGGTCATG ACTGCCATGA TCACCACAAT GAGCATGAGC ATGAGCATGA ACACGGTATG TTATAGTGTT 890 900 910 920 930 940 950 960 ACCAAATAAT GGCCTTATTA CTTAATAACG ATCGCCTACA TGCATTGATT AGTTTGCTTC TGAAACTGCA GAGATCACCA 970 930 990 1000 1010 1020 1030 1040 TTCTCATGAT CACACCCATG ACCCTGGTGT TGGTTCAGTC AGTATAGTTT GCGAAGGAGA CTTAGACCTC GAGAAGGTAT 1050 1060 1070 1080 1090 1100 1110 1120 TAACCCAATA TGGCTCAAGT TGTGTCACAT GTGATGCATA AAACTGTAGC ACCTGACGTG TTACTTATAA TTAAAACGAT 1130 1140 1150 1160 1170 1130 1190 1200 TGTTGAGATG TGCGATTGAT TACTTGCATC ATCTATAACC TTAGTTTTGA AAATGGATCA GGCTAACATG TGGCTTGGGG 1210 CGCTATTGTA CCAACGT 3' Figure 6. DNA sequences of Al-5-0 and Al-5-44. Both are parts of the clone Al-S. 98 because of its slightly higher hybridization signal. The insert was released from the recombinant phage with EcoRI and then cloned into the EcoRI site of pBluescript to give rise to the plasmid pAl-S. The Al-5 was sequenced in one orientation and was not completely sequenced, figure 5 shows the sequencing strategy. Two sets of sequences generated, Al-5-0 and A1-5-44, were separated by a 200 bp gap. The two DNA sequences and all the possible reading frames of these two sequences were analyzed in the same way as the clone Ab5-10 (Chapter IV), no informative similarities with any known DNAs and proteins could be found. The best match for the DNA data bank search is below 50%, and the best match for the protein data bank search is 26% identity over 47 amino acid residues. The 200 bp gap which has not been sequenced does not seem to cover the conserved sequences, because both in rat and yeast genes, the sequences corresponding to 01 and D3 oligonucleotides are approximately 500 bp apart. The figure 6 shows part sequences of the clone Al-S. 2). Oligonucleotides as probes Genomic Southern analysis indicated that in Agmenellum there are two EcoRI fragments of approximately 3.8 and 10 kb recognized by both DI (01A and DIT) and D3 (figure 7). 99 l 2 3 4 5 6 7-8 23:1? -- - e .I 22:" .... _. 1'"? Z - . ' a 53': “" A B C D Figure 7. Genomic blotting of Agmenellum and yeast by oligonucleotide mixtures DIA, DIT, 02, and 03 (A, B, C and D). Lanes 1, 3, 5 and 7 are yeast DNA; Lanes 2, 4, 6 and 8 are Agmenellum DNA. All DNA were digested with Eco RI to completion (2 ug per lane). Another 5 kb fragment was recognized by the D3 oligo also. After extensive screening of 4 independent libraries ( 2 Agmenellum lambdo gt10 libraries and 2 Agmenellum lambdo 2000 libraries), not a single clone could be isolated by using oligonucleotides 01A and DIT as probes. By contrast, more than 100 clones were isolated by using oligo 03 as a probe, none of these hybridized to oligo 01A or DIT at all. DNA was made from 25 clones isolated on the basis of homology to probe 03, and all of these clones contained a 5 kb EcoRI fragment with homology to D3 probe. This is presumably the fragment evident on the Southern blot, which did not hybridize to probe DlT or DlA. A 1 kb fragment called 84-43 from this 5 kb clone was shown to hybridize to probe 03, and was subsequently sequenced. Figure 8 is the sequencing strategy and figure 9 is the sequence for the fragment 100 Q___ <—— <——— 4—— e— ‘—— 0.2 kb g . , . 84-43 FRAGMENT E PB H S H E 1 kb vg :4 :fi it i 4 P————Q 5 kb FRAGMENT Figure 8. SequenCing strategy for fragment 84-43. 10 20 30 40 50 60 70 80 5' AGCTTTACGC ITACCACAGA GTTGATAAAT GTAGTAGAAA GCGGCCCCAG CAATATAGGT TCCCAGTAAA AAATTAAAGA 90 100 110 120 130 140 150 160 GTCCAGGTGC GCCAATGGTG GGAACATGAG GACGACAAAA ATAACGGTGA GGATCCCCGG CACGAGATAG GCTTTCTGGA 170 180 190 200 210 220 230 240 GAGGTCTTTC CCGGTGGCGG CGATGGGAAA AAGTTGGGTC AGGGTGAGGT TGGAGTTGGG ACTGACAATG GGGCCGGGGA 250 260 270 280 290 300 310 320 CAGGGATATG GGAACCCGGA TCGAGGGGAG CCTGGGGATT GTGGCCTTGC TCTGGAGCTC GAAACTAAAG CTGGGGCCAT 330 340 350 360 370 380 390 400 TGTCTCCGAG GGTGAGGCGA TCGCTGCTTG CAGATTCGTG AACCTTGTAA CCGTTGACCA TTGATAAAGG TGCCATTCGC 410 420 430 440 450 460 470 480 ACTGCCGAGA TCAACGGCTT CCCAACCTTG ACCCTGGGGA CGAATCAGAA CGTGGCGACG GGAGACAACG GTGTAAAGAT 690 500 510 520 530 $40 550 560 TGGGGTTGAG GGCAATCTGG CAACTGGGTT CGCGGCCAAT GAGAATTTCT TGACGATGAT CAAGGGAAAA ATCGGGCAGC 570 580 590 600 610 620 630 640 AGTGGGGCTT GAACACCGCC AGTGGAAATT TGGCGCAAAA TACCAGAAGC TTGCATCGTA ATAATTTTTT TCTAAAACGC 650 660 670 680 690 700 710 720 AAAAGACGGA GGAGCGCCCC GCCTCAACTT TAGTTATAAC GATAGATTGC CGGATCTTAA AGATATTTTT GTAAATTGAT 730 740 750 760 770 780 790 800 ACGGGCTGTT GAACGCCTAG TCTTGATTGG CGCGAGTATC TTGGACGGCC CACAAAAAAG AGGCAATGGT CAGGGGAATA 810 820 830 840 850 860 870 880 CTGAGGGCCA GGATAATGGT GGTATTGGTC GCACCAAGAT CGGGTTCGCC GGAACTGAGT TCAAACACGC AACCCACAGC 890 900 910 920 930 940 950 960 GGCGATCGAA AGACGCAGCA TAGTAATAAT AAAACACCAC TTTTGGGGGT CATCGAAACC ATGGTTTTAT GCCTCAAAGA 970 980 990 1000 1010 TTTGGTTTAG GGGATTATAA AAAAGAGTGA GATAAGGGCA AAACAGAGGT 3' Figure 9. DNA sequence of the fragment 84-43. 101 84-43. Unfortunately, as we expected (because it did not hybridize to DlA or DlT), it is not desaturase gene. As before, analysis of the sequence failed to indicate any significant similarity to proteins or genes of known function. Thus we cannot assign any functions for this clone. 4. Discussion Although we have not been able to clone desaturase genes from either plant or Agmenellum, it remains possible that, with additional efforts, the approach may work. The reason that the two Agmenellum fragments recognized by both oligos on Southern blots can not be recovered in plaque screening could be due to a variety of reasons. First, it is possible that the sequences are killing E;ggli. The sequences could encode protein products which are toxic to Engli. Second, there may be some bias in the library so that the sequences not be packaged into lambda or the recombinant phage fail to produce plaques. These problems may be circumvented by making libraries using different restriction enzymes or making plasmid libraries. The problem with using heterologous probes to isolate genes is that there is no way to know what stringency to use. Yeast genes have been used successfully to clone the corresponding plant genes (3,4). However the probability of identifing a specific gene depends on how conserved the specific gene is between yeast and plants. The yeast desaturase gene does recognize a few Arabidopsis sequences under the condition used. The two clones I have sequenced are not desaturase genes. However, this does not mean there is no need to 102 sequence the rest of the clones. I do sincerely believe that it is possible to clone the desaturase genes by this approach, but I don’t know how long it will take and how many clones we have to sequence before we can really get what we want. 5. Refferences l). M. A. Thiede, J. 02015, and P. Strittmatter (1986). J. Biol. Chem. 261:13230-13235 2). D. G. Oppenheimer, N. Haas, C. D. Silflow, and D. P. Snustad (1988). Gene. 63:87-102 3). R. L. Scholl, H. Zhang, Y. Kim, and C. Somerville (1989). Gene. Summitted 4). B. J. Mazur, C. F. Chui, and J. K. Smith (1987). Plant Physiol. 85:1110-1117 APPENDIX IV DOUBLE STRANDED DNA SEQUENCING AS A CHOICE FOR DNA SEQUENCING ----- Reprint from NUCLEIC ACIDS RESEARCH, Vol. 16, No. 3, 1988, p 1220 103 104 Volume 16 Number 3 1988 Nucleic Acids Research llmbhsflmmkd[HUKamuummgasatmmkefiw[nutsnmumkg Hong Zhang. Randy Scholl‘. John Browse and Chris Somerville Genetics Program. Michigan State University. E. Lansing. MI 48824 and ‘Genctics Department. Ohio State University. Columbus. OH 43210. USA Submitted October 23. I987 Double stranded DNA sequencing (1) is favored because of its simplicity, and convenice. However it is only recently that the quality of this method has become comparable with single stranded DNA sequencing. We believe that two factors have limited the popularity of double stranded DNA sequencing: 1, the quality of the template DNA: 2, the inherent propertyMof the DNA polymerase. The modified T7 DNA polymerase (Sequenase ) has several properties which make it more suitable for sequencing (2). We replaced the Klenow Polymerase I with Sequenase in our double stranded DNA sequencing, and by paying careful attention to the conditions used to denature and recover the template DNA, we are now routinely producing sequencing results which are as good as single stranded DNA sequencing (see figure). Here, we present a step-by-step protocol for the alkaline denaturation of template DNA, and our recommendations for the sequencing reaction. I. Template preparation: 1. use 3ug of CsCl purifed plasmid DNA, add NaOH to 0.2M and EDTA to 0.2mM (total volume 20ul), incubate at room temperature for 5min 2. neutralize by adding 2u1 of 2H NH Ac (pH4.65, mix quickly, then add 60ul of 100% etHanol (-20 C), mix well on ice 3. p ecipitate the DNA in a microfuge f8r 20min at 4 C, wash once with 80% ethanol (-20 C), spin again for another 5min 4. carefully draw away ethanol, and dry the DNA in a vacum desicator for 10min, then use immediately II. Sequencing Reaction: .."EF :5: See Sequenase Manual (3) for details. This “- 2": manual was originally designed for single stranded :F" DNA sequencing, for double stranded DNA sequencing, .. we recommend the following modifications: 1. Sng (about 1pmol) primer will be enough, excess of primer often decreases the sequencing quality 2. anneal template and primer at 65 C for 3-5min, then cool at room temperggure for about 30min 3. 1u1 of 400 Ci/mmol of ( S)-dATP per reaction will give satisfactory results A i ..I I I '5‘ References: 1. Chen, E. Y., Seeburg, P. H. (1985) DNA 4, 165-170 2. Tabor, S., Richardson, C. C. (1987) Proc. Natl. Acad. Sci. 84, 4767-4771 3. Step-By-Sffip Protocols For DNA Sequencing With Sequenase (United States Biochemical Corporation, P.O.Box 22400, Cleveland, Ohio 44122) Ii ~° in. "I Ea? {i 1220 © IRL Press Limited, Oxford, England. “11111111111111“