Chloroplast genes transferred to the nuclear plant genome have adjusted to nuclear base composition and codon usage

Nucleic Acids Research, Vol. 18, No. 1 Chloroplast genes transferred to the nuclear plant genome have adjusted to nuclear base composition and codon usage J.L.Oliver*, A.Marfn1 and J.M.Martfnez-Zapater2 Unidad de Genetica, Facultad de Ciencias, Universidad de Granada, E-18071-Granada and 1Departamento de Genetica y Biotecnia, Facultad de Biologfa, Universidad de Sevilla, Aptdo. 1095, E-41080 Seville and 2Departamento de Protection Vegetal, CIT-INIA, Carretera Gral. de La Coruna Km 7, E-28040-Madrid, Spain Received October 17, 1989; Revised and Accepted November 30, 1989 ABSTRACT During plant evolution, some plastid genes have been moved to the nuclear genome. These transferred genes are now correctly expressed in the nucleus, their products being transported into the chloroplast. We compared the base compositions, the distributions of some dinucleotides and codon usages of transferred, nuclear and chloroplast genes in two dicots and two monocots plant species. Our results indicate that transferred genes have adjusted to nuclear base composition and codon usage, being now more similar to the nuclear genes than to the chloroplast ones in every species analyzed. INTRODUCTION The existence of adjustment of base composition to different genomic G + C contents was shown in homologous genes and noncoding sequences of microorganisms and mitochondrial genomes1. Demonstration of the existence of such adjustment has been hindered in higher eukaryotes because of the compartmentalization of their genomes. Only recently, it has been shown that there is a compositional adjustment for Alu repetitive sequences that are located in different human genome compartments2. Coding sequences that have moved between different genomes can be good candidates to probe the existence of compositional adjustment and to analyze the involved mechanisms. Such gene movements have recurrently occurred along plant evolution and most of the plastid genes have been transferred to the nuclear genome345'6. Since in most plant species chloroplast and nuclear genomes have different GC contents7, we think that plastid genes transferred to the plant nuclear genome can be an excellent model system to analyze if such compositional adjustment exists and, if so, how it works. We compared nucleotide composition and codon usage of nuclear, chloroplast and nuclear genes encoding chloroplast proteins (transferred genes) in two dicots (pea and tobacco) and two monocots (wheat and maize) species. The distributions of some relevant dinucleotides were also studied. Results indicate that — at the level of base composition, dinucleotide distribution and codon usage — transferred genes are more similar to nuclear genes than to chloroplast ones. We analyzed how they have adjusted their base composition and codon usage to that of the nuclear environment. DATA AND METHODS Gene sequences Sequences from nuclear and chloroplast genes were retrieved from the GenBank genetic sequence data bank8 (release 57), or directly taken from original publications. We selected the four species with higher numbers of nuclear and chloroplast genes sequenced: Nicotiana tabacum (Solanaceae), Pisum sativum (Leguminosae), Triticum aestivum and Zea mays (Poaceae). Nuclear genes encoding chloroplast proteins can be considered as transferred chloroplast genes5-6. Although this seems to be a common situation, two exceptions of nuclear encoded chloroplast proteins that probably evolved from nuclear genes have already been described9-10. Table 1 shows the genes we have identified as transferred genes. A list of the remaining nuclear and chloroplast genes used in this study is shown in the Appendix. Nucleotide composition data Before nucleotide composition and codon usage were analyzed, introns of all genes and the sequence coding for the signal peptide present in transferred genes were removed. Nucleotide sites subject to silent changes ('silent sites') are calculated according to reference 1 [N = A, C, G, or T(U); R = A or G; Y = C or T(U)]: A, third positions of all codons, plus A in first positions of AGR codons; C, third positions of all codons, plus C in first positions of CTR and CGR codons; G, third positions of all codons, minus G in third positions of ATG and TGG codons; T, third positions of all codons, plus T in first positions of TTR codons. * To whom correspondence should be addressed 65 66 Nucleic Acids Research Codon usage data arginine, leucine and serine, each with two codon groups. We ., ,„ . , , „ _ , , . . exclude from die analysis termination codons and single-codon The following strategy was used to study codon usage in nuclear . ••• \ i_ \ TT.- 1 m F ... .B. c- . AC A .u . groups (methiorune and tryptophan). This leaves 21 codon groups and chloropast genes. First, we define as codon groups the sets •*. * c en * c A . J o r r . _,^ • i • . . • J . •_, with a total of 59 codons. Second, we count codon appearances of synonymous codons differing only in the third nucleotide. . , , ^ u i »• r t u A —,-•,_, ;. •. r in each gene and compute the relative frequency of each codon There is a single codon group for each ammo acid, except for . , r.. , ,,. » /-?. . . .• •. A u 6 6 v ' f in eacn of the C(Xjon groups (the count of that codon divided by Table 1. ChJoroplast genes transferred to the nucleus of pea (PEA), tobacco (TOB), wheat (WHT) and maize (MZE) used in this study. Sequences were retrieved from GenBank (Release 57) or, when no GenBank LOCUS name is specified, directly from the indicated source. GENE GenBank SYMBOL LOCUS PROTEIN Chlorophyll a/b-binding protein Major light harvesting protein AB80 RuBisCo small subunit Ferredoxin-NADP+ induced protein Early light-induced protein Chloroplast ribosomal protein (CL18) (CL24) (CL25) " (CL9) Chloroplast GAPDH-A Chloroplast GAPDH-B RuBisCo, small subunit Acetolactate synthase Major chlorophyll a/b-binding protein RuBisCo, small subunit RuBisCo, small subunit Glyceraldehide-3-phosphate dehydrogenase B Chlorophyll a/b-binding protein (1) Newman, B.J. and Gray, J.C. (1988) Plant Mol. Biol. 10, 511 -520. (2) Kolanus, W., Scharnhorst, C, Kiihne, U. and Herzfeld, F. (1987) Mol. Gen. Genet. 209, 234-239. (3) Gantt, J.S. (1988) Curr. Genet. 14, 519-528. (4) Mazur, B.J., Chui, C.F. and Smith, J. (1987) Plant Physiol. 85, 1110-1117. (5) Matsuoka, M., Kano-Murakami, Y., Tanaka, Y., Ozeki, Y. and Yamamoto, N. (1987) J. Biochem. 102, 673-676. (6) Brinkmann, H., Martinez, P., Quigley, F., Martin, W. and Cerff, R. (1987) J. Mol. Evol. 26, 320-328. (7) Matsuoka, M., Kano-Murakami, Y. and Yamamoto, N. (1987) Nucl. Acids Res. 15, 6302. Table. 2 Nucleotide composition and GC content in replacement (RS) and silent (SS) sites (see text) of chloroplast (CP), transferred (TF) and nuclear (NUC) genes. Species abbreviations are as in Table 1. PEA: cab 15 cab80 rubpl5 fnr elip rpsl8 rps24 rps25 rps9 TOB: gapA gapB rbpco als WHT: cab rbca MZE: rbcs gapB cab PEACAB15 PEACAB80 PEARUBP15 (1) (2) (3) TOBGAPA TOBGAPB TOBRBPCO (4) WHTCAB WHTRBCA (5) (6) (7) GENOME PEA: NUC TF CP TOB: NUC TF CP WHT: NUC TF CP MZE: NUC TF CP TOTAL G+C .43 .44 .41 .45 .48 .41 .57 .59 .38 .57 .67 .42 A .31 .30 .25 .27 .28 .27 .28 .25 .31 .26 .26 .30 T .21 .22 .28 .23 .22 .24 .20 .24 .24 .22 .22 .21 C .19 .19 .20 .21 .19 .20 .27 .20 .19 .25 .21 .19 RS G .29 .29 .27 .29 .31 .29 .25 .31 .26 .27 .31 .30 G+C .48 .48 .47 .50 .50 .49 .52 .51 .45 .52 .52 .49 A .29 .29 .27 .24 .21 .30 .20 .07 .32 .14 .02 .33 T .37 .34 .44 .40 .36 .42 .13 .19 .41 .19 .02 .38 C .20 .19 .18 .23 .25 .17 .42 .48 .15 .40 .67 .18 SS G .14 .18 .11 .13 .18 .11 .25 .26 .12 .27 .29 .11 G+C .34 .37 .29 .36 .43 .28 .67 .74 .27 .67 .96 .29 Nucleic Acids Research 67 the total of codons in the pertinent codon group); this method draws attention to the specific choices made by the organism among different options (the synonymous codons) regardless of the frequencies of the different amino acids in its proteins. Third, we compute the overall difference in codon usage between any two genes through a distance algorithm which is a version of the 'Manhattan metric' often used by numerical taxonomists. The codon usage distance between genes A and B is simply the sum of the absolute values of the differences in codon frequencies: i=59 D (A,B) = where x(i,A) and x(i,B) are the frequencies of the ith codon in genes A and B, respectively. RESULTS Similarity estimates between non-homologous genes Measurement of similarity between nuclear and chloroplast genes requires comparisons of non-homologous sequences. Grantham" proposed the combination of four indexes, based on codon usage and GC content, to estimate the similarity among non-homologous genes from very different sources. When applied to our data, this method was unable to differentiate among the nuclear, transferred, and chloroplast gene sets from each species (data not shown); only the GC content of the third codon position, taken individually, was able to clearly differentiate chloroplast from nuclear genes in all four species analyzed (data not shown). Chloroplast genes transferred to the nuclear genome were always grouped among the nuclear genes using this index. Nucleotide composition analysis Changes in nucleotide composition of transferred genes after relocation in the nuclear genome were analyzed by studying base Table 3. Distributions of CpG and TpA doublets in nuclear (NUC), transferred (TF) and chloroplast (CP) genes of different species. The ratio of the observed to the expected frequencies for these dinucleotides are given. Deviations from expectations were tested by Chi-square. Gene abbreviations are as in Table 1 and the Appendix. TOB GENE NUC: gapC pr-la pr-lb pr-lc ech pox thaur TF: gapA gapB rbpco als CP: psaA psaB psbA psbC psbD atpA atpB atpE atpF atpH atpl rps2 rps4 rpsl4 rpsl6 rpoB rbcl petA petB CpG 0.26 0.59 0.55 0.60 ' 0.56 0.42 ' *** i.** *** 0.39 *** 0.52 *** 0.48 *** 0.48 * 0.65 *** 0.59 *** 0.66 *** 0.64 * 0.60 ** 0.80 0.97 0.99 0.73 1.35 0.85 0.60* 0.94 1.39 1.02 1.35 0.81 * 0.85 0.83 1.10 TpA 0.45 *** 0.84 0.80 gsl 0.86 gs2 0.62 ** 0.75 * 0.67 * 0.35 *** 0.52 *** 0.63 0.74 ** 0.87 0.85 * 0.96 0.95 0.82 0.93 0.94 0.84 0.66 * 1.01 0.97 0.82 0.90 0.46 ** 0.64 0.81 *** 0.88 0.84 1.07 PEA GENE abn2 legJ 0.40 *** 0.25 *** gs3 0.39 *** lecA 0.52 ** vie 0.45 *** adh-1 hsp cab 15 cab80 rubpl5 fnr elip rpsl8 rps24 rps25 rps9 atpA cyf psbD psbC CpG 0.86 0.59 *** 0.71 0.59 0.55 ' 0.72 ' 0.57 ' 0.39' C* |:** |t* 0.43 ** 0.43 *** 0.44 *** 0.50 * 0.50 ** 0.15 *** 0.43 0.56 0.95 0.38 ** 0.96 0.66 * 0.71 * 0.60 TpA 0.65 ** 0.48 *** 0.51 *** 0.55 *** 0.71 0.53 ** 0.39 ** 0.54 *** 0.60** 0.29 *** 0.75 0.36 ** 0.59 ** 0.95 0.82 0.87 1.03 * P < 0.05; ** P < 0.01; *** P < 0.0001 68 Nucleic Acids Research Table 3 (continued) MZE GENE NUC: actlG adhlF ant eg2R h3 h4 susysG zel9A ze22A ze22B zea20M zea30M b32 gapA pepC gst cat cl TF: rbcs gapB cab CP: atbB atbE atpB rps4 rubp rpsl4 rps8 CpG 0.58 *** 0.69 ** 0.46 *** 0.81 1.08 1.15 0.73 *** 0.34 *** 0.46 *** 0.57 ** 0.34 *** 0.34 *** 0.88 0.70 ** 0.92 1.09 1.11 1.14 0.98 1.17 1.05 0.92 0.70 0.64 1.24 0.93 1.32 0.96 TpA 0.59 *** 0.40 *** 0.62 ** 0.50 * 0.21 * 0.76 0.51 *** 0.68 * 0.79 0.75 0.75 0.72 0.47 ** 0.46 *** 0.48 *** 0.57 0.68 * 0.41 ** 0.91 0.25 *** 0.48 * 0.86 0.79 0.94 0.85 0.89 0.87 0.91 WHT GENE glgB gliABA glumrA h3 h4 g'r amy em cab rbca atp atps cytf cytb rps2 frxB CpG 0.37 *** 0.42 *** 0.51 * 1.00 1.19 0.92 0.98 0.96 0.75 * 1.22 0.45 1.19 0.85 1.29 0.96 0.90 TpA 0.31 *** 0.49 *** 1.01 0.44 0.97 0.57 *** 0.59 ** 0.37 0.41 *** 0.59 1.03 0.70* 0.74 * 1.04 0.71 * 0.79 * P < 0.05; ** P < 0.01; *** P < 0.0001 composition of silent and replacement sites. Table 2 shows the average base frequencies in silent and replacement sites of nuclear, transferred and chloroplast genes of each species. Total G + C of the analyzed genes is higher in the nucleus than in the chloroplast for all four species. This difference is specially high in the two monocots species. Table 2 also shows that transferred genes have reached similar GC content than nuclear genes by increasing their GC content mainly in silent sites. Note that in our sample of genes, GC content at replacement sites is very similar among dicot and monocot species but both groups clearly differ in their GC content at silent sites: GC content at silent sites is lower than at replacement sites in dicots but higher in monocots. Dinucleotide distributions Nuclear plant DNA has a high content of 5-methylcytosine in both the CG dinucleotide and also in the C(A/T)G trinucleotides12 and CpG methylation is not found in the chloroplast genome13. Since chloroplast genes transferred to the nucleus show an increase in GC content (Table 2), we analyzed the distribution of methylation sites in transferred genes to find out if the increase in GC parallels an increase in the methylation sites available. CTG and CAG trinucleotides were found at expected frequencies in most genomes (data not shown). The gene to gene distributions of the other methylation target, CpG, are shown in Table 3. Chloroplast genes generally show the expected frequencies of this dimer in the four species; on the contrary, most of the nuclear and transferred genes show significant deficiencies. This is specially true for dicots, while in monocots some nuclear genes show the CpG expected frequencies or even an excess of this dinucleotide. TpA is other dinucleotide of special relevance, since a general avoidance of this dimer has been reported in most genomes14'15'16'17. Table 3 shows its distribution for every gene. Nuclear and transferred genes show very variable TpA ratios while chloroplast genes are more uniform. Significant avoidances of TpA were more often found in nuclear and transferred genes of the four species than in chloroplast genes. Codon usage analysis Codon usage distances. A triangular matrix containing all pairwise comparisons of codon usage distances among all genes was computed for each species by means of the distance algorithm described in the Data and Methods section; all pairwise distances were then categorised into six groups (nuclear vs nuclear, nuclear vs transferred, etc.) and the average distance for each group computed (Table 4). Codon usage distances between chloroplast and nuclear genes are small in dicots and higher in monocot species. In monocots this is paralleled by higher distances between chloroplast and transferred genes. This method gives a global idea of codon usage in genes from the three different gene sets but does not consider the variation in codon usage within genes from the same group. Correspondence analysis. To get a better representation of the variability for codon usage within the different groups of genes, Nucleic Acids Research 69 Table 4. Codon usage distances ( S.E.) among nuclear (NUC), transferred (TF) and chloroplast (CP) genes. Species abbreviations are as in Table 1. NUC TF CP 8 1 3 1 2 . 75 . OS . 44 * ± 0 0 0 . 28 . 3 1 . 38 1 5 1 5 . 25 . 77 ± 0 ± 0 . 69 . 49 I I 1 11.81 z 0 PEA . 90 | NUCCP NUC TF CP 1 1 2 1 2 1 3 . 6 1 . 77 . 1 2 x 0 ± 0 ± 0 . 53 . 68 . 24 1 1 14 . 99 . 52 ± 1 ± 0 . 45 . 46 I I10. 46± 0 TOB . 22 | NUCCP NUC TF CP 15 . 13 . 23 . 38 59 73 ± ± ± 0 0 0 . 93 . 84 . 76 1 | 1 1 1 | 23 . 22 . 8 1 I £ 0.SO | 13.94 WHT ±0.73 | NUC NUC TF CP 13 13 22 . 06 . 57 . 18 ± 0 ± 0 I 0 . 32 . 73 . 38 4 29 . 65 ± .84 ± 1 0 . 1 1 . 5 1 I I13. 1 5 ± 0 MZE .97 | NUCCP we performed a factorial correspondence analysis on a data matrix containing the n-1 codon frequencies at each synonymous codon group for each gene (see ref. 18 for methodology involved in correspondence analysis of codon usage). Differences in codon usage cluster chloroplast genes separately from the nuclear ones in monocot (Fig. 1) but not in dicot species (data not shown). In all the species nuclear and transferred genes are always mixed together, being more widely scattered than the chloroplast ones. DISCUSSION Nucleotide composition of transferred genes GC content is lower in the chloroplast genomes than in the nuclear ones7; Table 2 shows that at gene level the differences are more extreme in maize and wheat than in the dicot species; this is due to the higher GC content of monocot nuclear genes. Since there is a correlation between genomic GC content and GC level in the three codon positions of genes119-20, it would be interesting to investigate if an adjustment of base composition occurred in transferred genes adapting them to the high GC content of the nucleus. Table 2 shows that chloroplast genes relocated into the nuclear genome have now similar GC content to nuclear genes. Increase in GC has been more pronounced at the silent sites than at the replacement sites. At silent sites, transferred genes show higher GC content than nuclear genes, while at replacement sites GC contents are very similar between nuclear and transferred glumrA • gliABA * rbca . glgB 1 • h3 .em • M • any • cab . gir • cytb rps2 • • cytf • atps • frxB • atp 1 B I atpB h3 • * cab rbcB * * gapB . gst pepC • • gopA eg2R • cat • susyBG • a: • b32 • odhlF actlG zea20M zea30M • 2B22B ze22A • rubp rps4 • • atbZ • atbB rpsM • rps8 Figure 1. Correspondence analysis of codon use in different genes of wheat and maize. The n-1 codon frequencies in each synonymous group were used for each gene. Gene and species abbreviations are as in Table 1 and the Appendix. The plots of Fl (vertical) xF2 (horizontal) factors explain the 59% of the variability in codon usage of WHT and the 60% in MZE. (• =nuclear gene; • =chloroplast gene; *=transferred gene). 70 Nucleic Acids Research PEATOB 60 55 50 45 40 35 30 o a a a o r a a a CD = 0.21 D (NS) 0.10.2 0.3 0.4 CG0.50.60.7 WHTMZE 80 70 50 40 30 r = 0.76 (P ( 0.95) S = 0.19 0.2 0.4 0.6 0.8 CG1.2 1.4 80 70 60 40 30 r - 0.93 (P < 0.99) S - 0.27 0.20.4 0.6 CG0.81.2 Figure 2. Plots of %(G+C) and CG (Obs/Exp ratio) of nuclear and transferred genes in each species. genes. Similar adjustments at silent and replacement sites provoked by GC pressure have been previously observed when homologous genes and noncoding sequences are compared in bacterial and mitochondrial genomes with different GC contents (see reference 1 and references therein). The differences in GC content at silent sites between monocots and dicots (Table 2) are probably indicating the existence of a higher GC pressure in the two monocots studied. The rise in GC content of transferred genes of dicots is due to similar increments in both bases G and C (Table 2); however, in the two monocots the rise in GC content of transferred genes has been due preferentially to increases in C. This is a similar situation to what was found earlier in the genomes of warmblooded vertebrates1921. Distributions of TpA and CpG dinucleotides The compositional adjustment of transferred genes to the nuclear environment is also reflected by a stronger avoidance of the doublet TpA. This avoidance is commonly found in most eukaryotic genomes1415, including the nuclear genomes of plants16-17. The higher CpG avoidance in nuclear than in chloroplast genes is probably due to the existence of CpG methylation in the nuclear genome12 but not in the chloroplast13'22. Transferred genes also follow a similar pattern in CpG avoidance as nuclear genes, thus adjusting their base composition to that of the nuclear genome. Nuclear genes of dicot species analyzed here always show avoidances of TpA and CpG dinucleotides. However, a different situation was found in the two monocot species analyzed, where some genes do not show any avoidance. We think that these differences between species are reflecting the different compositional organization of their genomes. The genomes of tobacco and pea are far more homogenous in base composition than the genomes of wheat and maize20-23. Having this in mind, the lack of CpG shortage in four out of five transferred genes of monocots could be due to the location of these genes in GC rich chromosome regions with a decreased discrimination against CpG doublets. Bernardi et al.24 have shown that CpG shortage decreases in degree when increasing genomic GC level in both vertebrates and their viruses. This could also be happening in the nucleus of wheat and maize, since in these two species — but not in pea and tobacco — we found that the CpG doublet Nucleic Acids Research 71 level is strongly correlated with overall GC content of different genes (Fig. 2). Codon usage comparisons In all four species the codon usage of transferred genes has consistently become undistinguishable from that of the nucleus where they are integrated. Table 4 shows that the codon usage of transferred genes is more distantly related to the chloroplast genome from which they derive than to the nuclear genome in which they are now located. This is particularly clear in the two monocots species. Since the GC contents of the dicot nuclear genes analyzed are more similar to the chloroplast ones (Table 2), codon usage does not differentiate so well genes from one or the other compartment. Multivariate analyses shown in Fig. 1 allow to visualize the heterogeneity present for codon usage within the nuclear genomes. The dispersion found for nuclear genes contrasts with the homogeneity found for the chloroplast ones. Since constraints on codon usage imposed by the aminoacid composition of proteins can be discarded due to the method we used to compute codon frequencies, this is probably a reflection of the compositional organization of the plant genomes20. Transferred genes always appeared mixed with the nuclear genes by this analyses (Fig. 1), which again supports their adjustment to the nuclear environment. A strong bias in codon usage for the nuclear encoded chloroplast GAPDH of maize has been reported6. These authors hypothesize that the strong codon bias found could be a consequence of a selection for higher expressitivity. However, the same bias is not found for the same gene in other species. A similar pattern of stronger biases in codon usage in monocots than in dicot species have also been reported17 for two transferred genes, rbcs (maize) and cab (wheat). Our more global results indicate that this bias could be the consequence of the increase in GC content that transferred genes have suffered to reach the level of the new host genome. Since this increase is mainly supported by changes in silent sites (Table 2), it produces a strong bias in the codon usage of these genes. Bias is extremely high in species like maize where differences in GC content between chloroplast and nuclear genome are very high. As a general conclusion, chloroplast genes transferred to the nucleus seem to have adjusted their base composition, dinucleotide distribution and codon usage according to the characteristics prevailing in their new host genomes, and thus they behave as polite DNA2526. Table 2 reveals the clear trend of transferred genes to achieve similar GC content as the nuclear genomes where they are integrated. Consequently codon usage is affected and since changes in nucleotide sequence affect, almost exclusively, to the silent sites, protein sequences can remain mainly unmodified. Therefore, the GC increase can occur without changing the coding capacity of transferred genes, which at the aminoacid level are still homologous to their prokaryotic counterparts5. Because the mosaic organization of the eukaryotic genome19'20'24-27, a certain level of variability should be expected in base composition and codon usage among genes from the nuclear genome and this is in fact found (Fig. 1). As expected from the isochore organization in these four species, variation is higher in wheat and maize that show a wider compositional heterogeneity20. The distributions of TpA and CpG doublets in transferred genes also reflect the conditions of every nuclear genome or genome compartment. It seems clear from these results that there is an evolution towards compositional homogenization within the different compartments of the nuclear plant genome. If this evolution is based on a selective advantage due to improved expressitivity or to other compositional modifying mechanisms not related with gene expression, such as different mutational bias of DNA polymerases in germline cells28 or variation in mutation patterns along the replication timing of different chromosomal regions in the germline29, is not known at this moment. Experiments testing the expressitivity of coding sequences with different base composition and codon usage are required to elucidate the presence of any selective advantage. ACKNOWLEDGEMENTS We are most grateful to Drs. M. Ruiz Rejdn and J. Salinas by the critical reading of the manuscript. This work was partially supported by the DGICYT (PB87-0881) and the INIA (#7556) of the Spanish Government. REFERENCES 1. Jukes, T.H. and Bhushan, V. (1986) J. Mol. Evol. 24,39-44. 2. Filipski, J., Salinas, J. and Rodier, F. (1989) J. Mol. Biol. 206,563-566. 3. Palmer, J.D. (1985) Ann. Rev. Genet. 19,325-354. 4. Martin, W. and Cerff, R. (1986) Eur. J. Biochem 159:323-331. 5. Shih, M.-C, Lazar, G. and Goodman H.M. (1986) Cell 47,73-80. 6. Brinkmann, H., Martinez, P., Quigley, F., Martin, W. and Cerff, R. (1987) J. Mol. Evol. 26, 320-328. 7. Boudraa, M. (1987) Genet. Sel. Evol. 19,143-154. 8. Bilofsky, H.S., Burks, C, Fickett, J.W., Goad, W.B., Lewitter, F.L., Rindone, W.P., Swindell, CD. and Tung, C.S. (1986) Nucl. Acids Res. 14,1-4. 9. Tingey, S.V., Tsai, F.-Y., Edwards, J.W., Walker, E.L. and Coruzzi, G.M. (1988) J. Biol. Chem. 263:9551-9657. 10. Vierling, E., Nagao, R.T., DeRocher, A.E. and Harris, L.M. (1988) EMBO J. 7, 575-581. 11. Grantham, R. (1978) FEBS Utters 95(1),1-11. 12. Gruenbaum, ¥., Naveh-Many, T., Cedar, H. and Razin, A. (1981) Nature 292, 860-862. 13. Tewari, K.K. and Wildman, S.G. (1966) Science 153,1269-1271. 14. Grantham, R., Greenland, T., Louail, S., Mouchiroud, D., Prato, J.L., Gouy, M. and Gautier, C. (1985) Bull. Inst. Pasteur 83, 95-148. 15. Ohno, S. (1988) Proc. Natl. Acad. Sci. USA 85:9630-9634. 16. Boudraa, M. and Pen-in, P. (1987) Nucl. Acids Res. 15,5729-5743. 17. Murray, E.E., Lotzer, J. and Eberle, M. (1989) Nucl. Acids Res. 17,477-498. 18. Holm, L. (1986) Nucl. Acids Res. 14,3075-3087. 19. Bemardi, G. and Bernardi, G. (1986) J. Mol. Evol. 24,1-11. 20. Salinas, J., Matassi, G., Montero, L.M. and Bernardi, G. (1988) Nucl. Acids Res. 16,4269-4285. 21. Marin, A., Bertranpetit, J., Oliver, J.L. and Medina, J.R. (1989) Nucl. Acids Res. 17,6181-6189. 22. Ngernprasirtsiri, J., Kobayashi, H. and Akazawa, T. (1988) Proc. Natl. Acad. Sci. USA 85:4750-4754. 23. Matassi, G., Montero, L.M., Salinas, J. and Bemardi, G. (1989) Nucl. Acids Res. 17,5273-5290. 24. Bemardi, G., Olofsson, B., Filipski, J., Zerial, M., Salinas, J., Cuny, G., Meunier-Rotival, M. and Rodier, F. (1985) Science 228,953-958. 25. Zuckerkandl, E. (1986) J. Mol. Evol. 24, 12-27. 26. Holmquist, G.P. (1989) J. Mol. Evol. 28, 469-486. 27. Aota, S. and Ikemura, T. (1986) Nucl. Acids Res. 14,6345-6355. 28. Filipski, J. (1987) FEBs Lett. 217,184-186. 29. Wolfe, K.H., Sharp, P.M. and Li, W.-H. (1989) Nature 337,283-285. 72 Nucleic Acids Research APPENDIX List of the remaining nuclear and chloroplast genes from pea (PEA), tobacco (TOB), wheat (WHT) and maize (MZE) used in this study. Sequences were retrieved from GenBank (Release 57) or, when no GenBank LOCUS name is specified, directly from the indicated source. GENE SYMBOLGenBank LOCUSPROTEIN PEA (nucleus): abn2 legJ gsl gs2 gs3 lecA vie adh-1 hsp PEA (chloroplast): atpA cyf psbD psbC TOB (nucleus): gapC pr-la pr-lb pr-lc ech pox thaur TOB (chloroplast): psaA psaB psbA psbC psbD atpA atpB atpE atpF atpH atpl rps2 rps4 rpsl4 rpsl6 rpoB rbel petA petB WHT (nucleus): gliABA glumrA h3 h4 gir amy em WHT (chloroplast): atp atps cytf cytb rps2 frxB MZE (nucleus): actlG adhlF PEAABN2 (1) PEAGSR1 PEALECA (2) (3) (4) PEACPATPG PEACPCYF PEACPD2 PEACPD2 TOBGAPC (5) TOBPR1CR TOBECH TOBPXDLF TOBTHAUR TOBCPCG WHTGLGB WHTGLIABA WHTGLUMRA WHTH3 WHTH4 WHTGIR WHTAMYA WHTEMR WHTCPATP WHTCPATPS WHTCPCYF WHTCPCYTB (6) (7) MZEACT1G MZEADH1F Albumin 2 'Minor' legumin polypeptide Glutamine synthase 1 2 II II "1 Seed lectin A Vicilin Adh-1 Chloroplast hsp ATP synthase subunit a (aa 1 —247) Cytochrome f propeptide PSII D2 protein PSII 44kDa reaction center protein Cytosolic GAPDH-C Pathogenesis related protein la Pathogenesis related protein lb Pathogenesis related protein lc Endochitinase Lignin-forming peroxidase TMV induced protein homologous to thaumatin PSI P700 apoprotein Al A2 PSII 32kD protein " 44kD protein " D2 protein ATPase alpha subunit " beta subunit " epsilon subunit " I subunit " III subunit " a subunit Ribosomal protein S2 " protein S4 " protein SI4 " protein SI6 RNA polymerase beta subunit RuBisCo large subunit Cytochrome f b Gamma-gliadin B Alpha-beta-gliadin A-II High-M-r gluten polypeptide H3 histone H4 histone Gibberellin responsive wheat gene Alpha-amylase EM protein ATP synthase proton-translocating subunit ATP synthase CF-0 subunit I prepeptide Cytochrome f Cytochrome b-559 (aa 1-83) Ribosomal protein S2 frxB gene Actin 1 Alcohol dehydrogenase (ADH1-F) Nucleic Acids Research 73 APPENDIX (continued) GENE SYMBOLGenBank LOCUSPROTEIN ant eg2R h3 h4 susysG zel9A ze22A ze22B zea20M zea30M b32 gapA pepC gst cat cl MZE (chloroplast): atbB atbE atpB rps4 rubp rpsl4 rps8 REFERENCES MZEANT MZEEG2R MZEH3C2 MZEH4C14 MZESUSYSG MZEZE19A MZEZE22A MZEZE22B MZEZEA20M MZEZEA30M (8) (9) MZEPEPCR MZEGST3A (10) (11) MZECPATBE MZECPATPB MZECPRPS4 MZECPRUBP (12) (12) ATP/ADP translocator Endosperm glutelin-2 Histone H3 Histone H4 Sucrose synthase Zein 19 kD protein " 22 kD protein " 26.99 kD protein " (clone a20) " (clone a30) b-32 protein Glyceraldehide-3-phosphate dehydrogenase A Phosphoenolpyruvate carboxylase Glutation-S-transferase GSTIII Catalase Regulatory cl locus Coupling factor complex, beta subunit " " " , epsilon subunit ATPase, beta subunit (aa 1-25) Ribosomal protein S4 RuBisCo, large subunit Ribosomal protein L14 Ribosomal protein S8 1. Gatehouse, J.A., Bown, D., Gilroy, J., Levasseur, M., Castleton, J. and Ellis, T.H.N. (1988) Biochem. J. 250, 15-24. 2. Watson, M., Lambert, N., Delauney, A., Yarwood, J.N., Croy, R.R.D., Gatehouse, J.A., Wright, D.J. and Boulter, D. (1988) Biochem. J. 251, 857-864. 3. Llewellyn, D.J., Finnegan, E.J., Ellis, J.G., Dennis, E.S. and Peacock, W.J. (1987) J. Mol. Biol. 195, 115-123. 4. Vierling, E., Nagao, R.T., DeRocher, A.E. and Harris, L.M. (1988) EMBO J. 7, 575-581. 5. Matsuoka, M., Yamamoto, N., Kano-Murakami, Y., Tanaka, Y., Ozeki, Y., Hirano, H., Kagawa, H., OsHima, M. and Ohashi, Y. (1987) Plant Physiol. 85, 942-946. 6. Hoglund, A.S. and Gray, J.C. (1987) Nucl. Acids Res. 15, 10590. 7. Dunn, P.P.J. and Gray, J.C. (1988) Nucl. Acids Res. 16, 348. 8. Di Fonzo, N., Hartings, H., Brembilla, M., Motto, M., Soave, C, Navarro, E., Palau, J., Rhode, W. and Salamini, F. (1988) Mol. Gen. Genet. 212, 481-487. 9. Brinkmann, H., Martinez, P., Quigley, F., Martin, W. and Cerff, R. (1987) J. Mol. Evol. 26, 320-328. 10. Bernards, L.A., Skadsen, R.W. and Scandalios, J.G. (1987) Proc. Natl. Acad. Sci. USA 84, 6830-6834. 11. Paz-Ares, J., Ghosal, D., Wienand, U., Peterson, P.A. and Saedler, H. (1987) EMBO J. 6, 3553-3558. 12. Markmann-Mulisch, U. and Subramanian, R. (1988) Eur. J. Biochem. 170, 507-514.