DNA polymorphism in active gene and pseudogene of the cytosolic phosphoglucose isomerase (PgiC) loci in Arabidopsis halleri ssp. gemmifera

DNA variations in two PgiC loci were investigated in 15 strains of Arabidopsis halleri ssp. gemmifera. In a 5.5-kb region of the PgiC1 locus, 127 nucleotide substitutions and 33 length variations were observed. In a 6.0-kb region of the PgiC2 locus, 138 nucleotide substitutions and 33 length variations were observed. Frame shift, novel stop codons, and large length variations were observed in the PgiC2 coding region. These findings suggested that PgiC2 may be a pseudogene. The nucleotide diversities (pi) for the entire regions of both PgiC loci were approximately 0.0033. Tajima's test of both PgiC loci yielded significantly negative results. In the coding regions, the high proportions of replacement substitutions caused significant deviations from neutrality in McDonald and Kreitman's test. An excess of singletons and a high proportion of replacement polymorphic sites have been observed in the Adh and ChiA regions of A. halleri ssp. gemmifera. Thus, the A. halleri ssp. gemmifera population may not have reached equilibrium, and thus nonneutral patterns of DNA polymorphism were observed.     

DNA polymorphism at the cytosolic phosphoglucose isomerase (PgiC) locus of the wild plant Arabidopsis thaliana

DNA variation in a 4.7-kb region of the cytosolic phosphoglucose isomerase (PgiC) locus was investigated for 21 ecotypes of Arabidopsis thaliana. The estimated nucleotide diversity was 0.0038, which was one-third of those in previously investigated loci. Since most of the nucleotide variations (93%) were singleton and doubleton, Tajima's test statistic was significantly negative. About 50% of nucleotide polymorphisms in exons were replacement, which caused significance in McDonald and Kreitman's test when compared with Arabis gemmifera and Cardaminopsis petraea. These results indicated that DNA polymorphism at the PgiC locus was not under neutrality. There were two divergent sequence types in the PgiC region, which were associated with allozyme variation. The Fast allozyme was shown to have originated from the Slow allozyme, since two outgroup species had the Slow form. A phylogenetic tree of ecotypes with the Fast allozyme had the shape of a star phylogeny. Mismatch distribution of the Fast allozyme ecotypes resembled that expected under an expanding population model. These results suggest positive selection for the Fast allozyme of the PGIC in A. thaliana.     

Molecular Characterization of Duplicate Cytosolic Phosphoglucose Isomerase Genes in Clarkia and Comparison to the Single Gene in Arabidopsis

The nucleotide sequence of PgiC1-a which encodes a cytosolic isozyme of phosphoglucose isomerase (PGIC; EC 5.3.1.9) in Clarkia lewisii, a wildflower native to California, is described and compared to the previously published sequence of the duplicate PgiC2-a from the same genome. Both genes have the same structure of 23 exons and 22 introns located in identical positions, and they encode proteins of 569 amino acids. Exon and inferred protein sequences of the two genes are 96.4% and 97.2% identical, respectively. Intron sequences are 88.2% identical. The high nucleotide similarity of the two genes is consistent with previous genetic and biosystematic findings that suggest the duplication arose within Clarkia. A partial sequence of PgiC2-b was also obtained. It is 99.5% identical to PgiC2-a in exons and 99.7% in introns. The nucleotide sequence of the single PgiC from Arabidopsis thaliana was also determined for comparison to the Clarkia genes. The A. thaliana PgiC has 21 introns located at positions identical to those in Clarkia PgiC1 and PgiC2, but lacks the intron that divides Clarkia exons 21 and 22. The A. thaliana PGIC protein is shorter, with 560 amino acids, and differs by about 17% from the Clarkia PGICs. The PgiC in A. thaliana was mapped to a site 20 cM from restriction fragment length polymorphism marker 331 on chromosome 5.     

Nucleotide polymorphism at the Atmyb2 locus of the wild plant Arabidopsis thaliana

DNA variation was studied in a 2.2 kb region of the regulatory gene Atmyb2 using 20 ecotypes of Arabidopsis thaliana and one accession each of Arabis gemmifera and Arabidopsis himalaica. Nucleotide diversity (pi) in the region was 0.0027, which was lower than for other loci in A. thaliana. The MYB domain of the Atmyb2 gene (pi = 0.0036) had a larger variation than the non-MYB region (pi = 0.0013). Tajima's test and Fu and Li's test did not give a significant result. In contrast to the low level of polymorphism, the degree of divergence of the Atmyb2 region was higher between A. thaliana and A. gemmifera (K = 0.0730) than for other loci. The MYB domain (K = 0.0436) had smaller divergence than the non-MYB region (K = 0.0939). The HKA test detected significant discordance in the ratio of polymorphism to divergence in some comparisons. The pattern of low polymorphism and high divergence, which is mainly observed in the non-MYB region of the gene, is inconsistent with the neutral mutation theory. Strong purifying selection after establishment of A. thaliana and a species-specific adaptive process could be invoked to account for this pattern of polymorphism and divergence of Atmyb2.     

A recently silenced, duplicate PgiC locus in Clarkia

Previous electrophoretic analysis showed that 17 diploid species of the wildflower Clarkia (Onagraceae) have two cytosolic isozymes of phosphoglucose isomerase (PGIC; EC 5.3.1.9), whereas 15 other diploid species have a single PGIC. Molecular studies revealed that the two isozymes in the former species are encoded by duplicate genes, PgiC1 and PgiC2, whereas the single isozyme in the latter is always encoded by PgiC1. Phylogenetic analysis of the nucleotide sequences implied that PgiC2 was silenced four times independently in the genus. Here we describe a psi PgiC2 from C. mildrediae, a species in which only PgiC1 is expressed. The discovery of the psi PgiC2 is significant because it confirms a formal prediction of the phylogenetic analysis. The psi PgiC2 includes 5,039 nucleotides corresponding to 18 of the 23 exons of PgiC, as well as the intervening introns and 3' nontranslated region. The absence of an increase of nucleotide substitutions in its "exons" suggests that the gene was silenced recently. The present study appears to be the first to establish that a specific duplicate gene locus regularly expressed in a group of related plant species has been silenced in one of them. The multiple independent silencings of PgiC2 suggest that it remained functional but inessential in ancestral lineages. We discuss the possibility that PgiC2 may have been preserved in these lineages by selection against mutants causing defective PGIC1- PGIC2 heterodimers.      

Diverse selective modes among orthologs/paralogs of the chalcone synthase (Chs) gene family of Arabidopsis thaliana and its relative A. halleri ssp. gemmifera

As a model system, Arabidopsis thaliana and its wild relatives have played an important role in the study of genomics and evolution in plants. In this study, we examined the genetic diversity of the chalcone synthase (Chs) gene, which encodes a key enzyme of the flavonoid pathway and is located on chromosome five, as well as two Chs-like genes on the first and fourth chromosomes of Arabidopsis. The objectives of the study are to determine if natural selection operates differentially on the paralogs of the Chs gene family in A. thaliana and Arabidopsis halleri ssp. gemmifera. The mode of selection was inferred from Tajima's D values from noncoding and coding regions, as well as from the ratio of nonsynonymous to synonymous substitutions. Both McDonald-Kreitman and HKA tests revealed the effects of selection on the allelic distribution, except for the chromosome 1 paralog in ssp. gemmifera. The Chs gene on chromosome 5 was under purifying selection in both species. Significant, negative Tajima's D values at synonymous sites and positive Fay and Wu's H values within coding region, plus reduced genetic variability in introns, indicated effects of background selection in shaping the evolution of this gene region in A. thaliana. The Chs paralog on chromosome 1 was under positive selection in A. thaliana, while interspecific introgression and balancing selection determined the fates of the paralog and resulted in high heterogeneity in ssp. gemmifera. Local adaptation differentiated populations of Japan and China at the locus. In contrast, the other Chs-paralog of chromosome 4 was shaped by purifying selection in A. thaliana, while under positive selection in ssp. gemmifera, as indicated by dn/ds>1. Moreover, these contrasting patterns of selection have likely resulted in functional divergence in Arabidopsis, as indicated by radical amino acid substitutions at the chalcone synthase/stilbene synthase motif of the Chs genes. Unlike previous studies of the evolutionary history of A. thaliana, the high levels of genetic diversity in most gene regions of Chs paralogs and nonsignificant Tajima's D in the intron sequences of the Chs gene family in A. thaliana did not reflect the effects of a recent demographic expansion.     

Intra- and interspecific DNA variation and codon bias of the alcohol dehydrogenase (Adh) locus in Arabis and Arabidopsis species

Sequence variation at the alcohol dehydrogenase (Adh) locus was analyzed for six species each of the genera Arabis and Arabidopsis. Phylogenetic analysis showed that investigated species were grouped into three clusters, and the generic classification did not correspond to the clusterings. The results indicated that the genera could not be distinguished on the basis of the Adh variation. A significant difference in the ratio of silent to replacement sites was detected by MK test in two comparisons, with Arabidopsis thaliana polymorphism due to excess silent divergence. Silent changes were predominant in the evolution of the Adh locus in Arabis and Arabidopsis. To infer evolutionary significance of silent substitutions, codon bias was studied. The degree of codon bias of the Adh region was relatively constant over Arabis and Arabidopsis species. "Preferred" codons of A. thaliana were determined. No evidence of natural selection on codon change was detected in the Adh regions of A. thaliana and Arabis gemmifera.     

DNA polymorphism at the ACAULIS5 locus of the wild plant Arabidopsis thaliana

Nucleotide variation in the ACL5 gene region, which encodes spermine synthase, was analyzed for 21 Arabidopsis thaliana ecotypes and one accession of Arabis gemmifera. In A. thaliana, dimorphism was also detected in the ACL5 region, as in other nuclear genes of this plant. The nucleotide diversity (pi) of the entire region, exon and intron was 0.0163, 0.0042 and 0.0293, respectively. The level of nucleotide variation in this region was among the highest of those reported for genes in this plant species. The neutrality tests of Tajima, and Fu and Li did not detect significant deviation from test assumptions for the polymorphism data. However, the HKA test indicated that the level of polymorphism in the intron was significantly high, compared with A. gemmifera. The high nucleotide variation in the intron is responsible for the high level of nucleotide variation in the entire region. These results can be explained by elevated mutation rate in the ACL5 region in the A. thaliana lineage after the two species were split.     

DNA variation in the acidic chitinase locus (ChiA) region in Arabis gemmifera and its related species

We analyzed DNA variation at the acidic chitinase (ChiA) locus of Arabis gemmifera and among its eight related species. Nucleotide diversity (pi) of the entire locus in A. gemmifera was 0.0032, which was one third that of A. thaliana. In A. gemmifera, an excess of unique polymorphisms yielded significantly negative results with the tests of Tajima and Fu and Li. The McDonald and Kreitman test revealed that the ratio of nucleotide replacement to synonymous changes in A. gemmifera was significantly greater than those between A. gemmifera and A. glabra, A. gemmifera and A. griffithiana, A. gemmifera and A. korshinskyi, A. gemmifera and A. wallichii, and A. gemmifera and A. himalaica. These results indicated that the neutrality assumption, the equilibrium population assumption, or both, could not be applied to the ChiA locus of A. gemmifera. The small size and relative isolation of local subpopulations of A. gemmifera could explain the excess of unique polymorphisms and the high proportion of replacement changes. The specific sampling scheme of this study, where one strain each was sampled from each local subpopulation might also have led to an excess of singletons. Interspecific comparison among Arabidopsis, Arabis and Cardaminopsis species showed that Ka was always lower than Ks, providing evidence against the adaptive evolution of ChiA. However, Ka/Ks was greater between closely related species than between more distant related species. ChiA had a higher level of replacement divergence and a lower level of synonymous divergence compared than did Adh. We suggest that both the mutation rate at the nucleotide level and the selective constraints at the protein level are lower in ChiA than in Adh.     

Dual-domain, dual-targeting organellar protein presequences in Arabidopsis can use non-AUG start codons

The processes accompanying endosymbiosis have led to a complex network of interorganellar protein traffic that originates from nuclear genes encoding mitochondrial and plastid proteins. A significant proportion of nucleus-encoded organellar proteins are dual targeted, and the process by which a protein acquires the capacity for both mitochondrial and plastid targeting may involve intergenic DNA exchange coupled with the incorporation of sequences residing upstream of the gene. We evaluated targeting and sequence alignment features of two organellar DNA polymerase genes from Arabidopsis thaliana. Within one of these two loci, protein targeting appeared to be plastidic when the 5' untranslated leader region (UTR) was deleted and translation could only initiate at the annotated ATG start codon but dual targeted when the 5' UTR was included. Introduction of stop codons at various sites within the putative UTR demonstrated that this region is translated and influences protein targeting capacity. However, no ATG start codon was found within this upstream, translated region, suggesting that translation initiates at a non-ATG start. We identified a CTG codon that likely accounts for much of this initiation. Investigation of the 5' region of other nucleus-encoded organellar genes suggests that several genes may incorporate upstream sequences to influence targeting capacity. We postulate that a combination of intergenic recombination and some relaxation of constraints on translation initiation has acted in the evolution of protein targeting specificity for those proteins capable of functioning in both plastids and mitochondria.     

Identification and functional validation of a 5' upstream regulatory sequence in the human tyrosinase gene homologous to the locus control region of the mouse tyrosinase gene

Comparison analysis of the sequences of the mouse and human genomes has proven a powerful approach in identifying functional regulatory elements within the non-coding regions that are conserved through evolution between homologous mammalian loci. Here, we applied computational analysis to identify regions of homology in the 5' upstream sequences of the human tyrosinase gene, similar to the locus control region (LCR) of the mouse tyrosinase gene, located at -15 kb. We detected several stretches of homology within the first 30 kb 5' tyrosinase gene upstream sequences of both species that include the proximal promoter sequences, the genomic region surrounding the mouse LCR, and further upstream segments. We cloned and sequenced a 5' upstream regulatory sequence found between -8 and -10 kb of the human tyrosinase locus (termed h5'URS) homologous to the mouse LCR sequences, and confirmed the presence of putative binding sites at -9 kb, homologous to those described in the mouse tyrosinase LCR core. Finally, we functionally validated the presence of a tissue-specific enhancer in the h5'URS by transient transfection analysis in human and mouse cells, as compared with homologous DNA sequences from the mouse tyrosinase locus. Future experiments in cells and transgenic animals will help us to understand the in vivo relevance of this newly described h5'URS sequence as a potentially important regulatory element for the correct expression of the human tyrosinase gene.     

Chromosomal arrangement of leghemoglobin genes in soybean.

A cluster of four different leghemoglobin (Lb) genes was isolated from AluI-HaeIII and EcoRI genomic libraries of soybean in a set of overlapping clones which together include 45 kilobases (kb) of contiguous DNA. These four genes, including a pseudogene, are present in the same orientation and are arranged in the order: 5'-Lba-Lbc1-Lb psi-Lbc3-3'. The intergenic regions average 2.5 kb. In addition to this main Lb locus, there are other Lb genes which do not appear to be contiguous to this locus. A sequence probably common to the 3' region of Lb loci was found flanking the Lbc3 gene. The 3' flanking region of the main Lb locus also contains a sequence that appears to be expressed more abundantly in root tissue. Another sequence which is primarily expressed in root and leaf is found 5' to two Lb loci. Overall, the main leghemoglobin locus is similar in structure to the mammalian globin gene loci.     

The introgression of a functional nuclear gene from Poa to Festuca ovina

In sheep's fescue, Festuca ovina, genes coding for the cytosolic enzyme phosphoglucose isomerase, PGIC, are not only found at the standard locus, PgiC1, but also at a segregating second locus, PgiC2. We have used PCR-based sequencing to characterize the molecular structure and evolution of five PgiC1 and three PgiC2 alleles in F. ovina. The three PgiC2 alleles were complex in that they carried two gene copies: either two active genes or one active and one pseudogene. All the PgiC2 sequences were very similar to each other but highly diverged from the five PgiC1 sequences. We also sequenced PgiC genes from several other grass species. Phylogenetic analysis of these sequences indicates that PgiC2 has introgressed into F. ovina from the distant genus Poa. Such an introgression may, for example, follow from a non-standard fertilization with more than one pollen grain, or a direct horizontal gene transfer mediated by a plant virus.