Allelic variation of the Waxy gene in foxtail millet [Setaria italica (L.) P. Beauv.] by single nucleotide polymorphisms

The Waxy (Wx) gene product controls the formation of a straight chain polymer of amylose in the starch pathway. Dominance/recessiveness of the Wx allele is associated with amylose content, leading to non-waxy/waxy phenotypes. For a total of 113 foxtail millet accessions, agronomic traits and the molecular differences of the Wx gene were surveyed to evaluate genetic diversities. Molecular types were associated with phenotypes determined by four specific primer sets (non-waxy, Type I; low amylose, Type VI; waxy, Type IV or V). Additionally, the insertion of transposable element in waxy was confirmed by ex1/TSI2R, TSI2F/ex2, ex2int2/TSI7R and TSI7F/ex4r. Seventeen single nucleotide polymorphims (SNPs) were observed from non-coding regions, while three SNPs from coding regions were non-synonymous. Interestingly, the phenotype of No. 88 was still non-waxy, although seven nucleotides (AATTGGT) insertion at 2,993 bp led to 78 amino acids shorter. The rapid decline of r ² in the sequenced region (exon 1–intron 1–exon 2) suggested a low level of linkage disequilibrium and limited haplotype structure. K _s values and estimation of evolutionary events indicate early divergence of S. italica among cereal crops. This study suggested the Wx gene was one of the targets in the selection process during domestication. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Diverse origins of waxy foxtail millet crops in East and Southeast Asia mediated by multiple transposable element insertions

The naturally occurring waxy and low-amylose variants of foxtail millet and other cereals, like rice and barley, originated in East and Southeast Asia under human selection for sticky foods. Mutations in the GBSS1 gene for granule-bound starch synthase 1 are known to be associated with these traits. We have analyzed the gene in foxtail millet, and found that, in this species, these traits were originated by multiple independent insertions of transposable elements and by subsequent secondary insertions into these elements or deletion of parts of the elements. The structural analysis of transposable elements inserted in the GBSS1 gene revealed that the non-waxy was converted to the low-amylose phenotype once, while shifts from non-waxy to waxy occurred three times, from low amylose to waxy once and from waxy to low amylose once. The present results, and the geographical distribution of different waxy molecular types, strongly suggest that these types originated independently and were dispersed into their current distribution areas. The patterns of GBSS1 variation revealed here suggest that foxtail millet may serve as a key to solving the mystery of the origin of waxy-type cereals in Asia. The GBSS1 gene in foxtail millet provides a new example of the evolution of a gene involved in the processes of domestication and its post-domestication fate under the influence of human selection. Electronic Supplementary Material Supplementary material is available for this article at     

Ribosomal DNA variation in foxtail millet, Setaria italica (L.) P. Beauv., and a survey of variation from Europe and Asia

 Restriction fragment length polymorphism (RFLP) and the structure of ribosomal RNA genes (rDNA) were investigated in 117 landraces of foxtail millet, Setaria italica (L.) P. Beauv. Five RFLP phenotypes were found when the genomic DNA was digested with BamHI; these were named types I–V. Of these types I, II and III were the most frequent. Type I was mainly distributed in the temperature zone, type II in the Taiwan-Philippines Islands and type III in South Asia. Restriction mapping of the cloned rDNA and comparison with RFLP phenotypes showed that the different types originated from a polymorphism in the length within the intergenic spacer (IGS) and BamHI site changes within the IGS. Received: 28 August 1996 / Accepted: 28 February 1997     

The insertion/deletion polymorphisms in the waxy gene of barley genetic resources from East Asia

The length polymorphism in the waxy gene, which encodes a granule-bound ADP-glucose-glucosyl transferase [granule-bound starch synthase I (GBSS I), E.C. 2.4.1.11] in barley (Hordeum vulgare), was found. The 5′ leader sequence of the waxy gene of barley germplasm from Japan and Korea was analyzed by the polymerase chain reaction (PCR). The waxy gene of these genetic stocks had three types of length polymorphisms, suggesting that there are insertion/deletion mutations at the 5′ leader sequence of the waxy gene. DNA sequence analysis of the polymorphic PCR products showed that: (1) a 403-bp deletion mutation, which included a complete exon I, was found in the wax allele and a 193-bp insertion sequence was located in the intron I, and (2) the insertion sequence was also located in intron I of the Wax allele. The identity of the insertion sequence was completely conserved between the wax allele and the novel Wax allele. These finding s implying that the wax allele, which was found in indigenous waxy barley, originated in non-waxy barley with the novel Wax allele. Received: 12 January 2001 / Accepted: 17 April 2001     

Origin and evolution of the waxy phenotype in Amaranthus hypochondriacus: evidence from the genetic diversity in the Waxy locus

The existence of polymorphism in the Waxy locus in a large gene pool of 53 strains with various waxy phenotypes from samples of Amaranthus hypochondriacus collected from different regions was investigated in an origin-and-evolution study. First, we screened all strains for a mutation point (G–A polymorphism in exon 6) by using PCR–RFLP and/or direct sequence analysis. The results showed that the nonsense mutation in the coding region (exon 6) of the Waxy gene was responsible for the change in perisperm starch, leading to a waxy phenotype in all strains. Second, phylogenetic analysis, which was based on the Waxy variation, indicated diverse waxy types occurring separately and independently in certain domesticated regions in Mexico. Finally, we designated nine molecular types by comparing obvious structural variations in the coding region of the Waxy gene. Among the molecular types, A. hypochondriacus contained Type III in three subtypes with the waxy phenotype, with evolutionary routes that could originate from Type II in accordance with G–A polymorphism. In addition, these types had the same mutation points by which the Waxy gene was converted into the waxy phenotype. Therefore, the present results showed that the nonsense mutation is a unique event in the evolution of waxy phenotypes in this crop. This study will provide useful information for understanding the evolutionary process of the waxy phenotype.     

Molecular characterization of the waxy locus in sorghum

A comparison of approximately 4.5 kb of nucleotide sequence from the waxy locus (the granule-bound starch synthase I [GBSS I] locus) from a waxy line, BTxARG1, and a non-waxy line, QL39, revealed an extremely high level of sequence conservation. Among a total of 24 nucleotide differences and 9 indels, only 2 nucleotide changes resulted in altered amino acid residues. Protein folding prediction software suggested that one of the amino acid changes (Glu to His) may result in an altered protein structure, which may explain the apparently inactive GBSS I present in BTxARG1. This SNP was not found in the second waxy line, RTx2907, which does not produce GBSS I, and no other SNPs or indels were found in the approximately 4 kb of sequence obtained from RTx2907. Using one indel, the waxy locus was mapped to sorghum chromosome SBI-10, which is syntenous to maize chromosome 9; the waxy locus has been mapped to this maize chromosome. The distribution of indels in a diverse set of sorghum germplasm suggested that there are two broad types of non-waxy GBSS I alleles, each type comprising several alleles, and that the two waxy alleles in BTxARG1 and RTx2907 have evolved from one of the non-waxy allele types. The Glu/His polymorphism was found only in BTxARG1 and derived lines and has potential as a perfect marker for the BTxARG1 source of the waxy allele at the GBSS I locus. The indels correctly predicted the non-waxy phenotype in approximately 65% of diverse sorghum germplasm. The indels co-segregated perfectly with phenotype in two sorghum populations derived from crosses between a waxy and a non-waxy sorghum line, correctly identifying heterozygous lines. Thus, these indel markers or sequence-based SNP markers can be used to follow waxy alleles in sorghum breeding programs in selected pedigrees.     

Diversification of the rice Waxy gene by insertion of mobile DNA elements into introns.

The waxy (wx) gene of Oryza glaberrima was cloned, and its nucleotide sequence was determined. A waxy mutant of O. glaberrima showing a glutinous phenotype was found to contain a substitution mutation generating a termination codon in the coding region of the wx gene. The Wx sequence of O. glaberrima was different from that of Oryza sativa by substitutions and insertions/deletions, among which only a few substitutions occurred in several exons not to severely alter the amino acid sequence of the Wx protein. The most striking difference observed in introns was a 139-bp deletion (or insertion) in intron 10 of O. glaberrima (or O. sativa). In O. sativa, 125 bp of the 139-bp sequence was flanked by direct repeats of a 14-bp sequence. A sequence homologous to the 125-bp sequence was found in the region preceding exon 2; this sequence was also flanked by direct repeats of another 14-bp sequence. This result and the observation that the 125-bp sequence was interspersed in rice genomes indicate that they are SINEs (short interspersed elements) in the plant system. We also identified a DNA sequence with long terminal inverted repeats in intron 13 of both O. glaberrima and O. sativa. This sequence was present in multiple copies in rice genomes, suggesting that it is a transposable element. These results obtained suggest that mobile DNA elements have diversified the rice Waxy gene by inserting into introns, each of which may originally have a length of about 100 bp.     

Insertional mutagenesis of the cauliflower mosaic virus genome

A series of small insertions has been introduced into the various translational reading frames of the DNA of a "severe" strain of cauliflower mosaic virus (CaMV). A selectable gene (the kanamycin phosphotransferase gene of Tn903), flanked by a series of symmetrically arranged cloning sites taken from M13mp7, was used to prepare the site-specific mutants. In-phase insertions of 12 or 30 bp, which introduced unique SalI sites into reading regions I, III, IV, V and into the amino-proximal portion of region VI, destroyed infectivity. Insertions in the amino-distal portion of region VI, in the large intergenic region, and in region II retained infectivity. The amino-distal insertions in region VI reduced the severity of symptoms in plants. The insertion in region II destroyed aphid transmissibility. Longer DNA segments when inserted into region II or into the amino-distal portion of region VI destroyed infectivity, but similar insertions in the intergenic region were without effect on virus infection or development.     

Comparison of the waxy locus sequence from a non-waxy strain and two waxy mutants of spontaneous and artificial origins in barley

Molecular characterization of 3 alleles of the waxy gene from a non-waxy strain "Shikoku hadaka No. 84" (SH84), an indigenous waxy strain "Mochimugi D" (MMD), and an artificial waxy mutant strain "Shikoku hadaka No. 97" (SH97) of barley (Hordeum vulgare ssp. vulgare) was performed via a PCR direct sequencing strategy. The 3 haplotypes were analyzed in terms of single nucleotide polymorphisms, insertion/deletion mutations, and simple sequence repeat polymorphisms. In comparison with the barley non-waxy gene sequence deposited in the public DNA database, 110 polymorphic sites were found in the 5,190-bp sequenced region of the non-waxy strain SH84. A 418-bp deletion in the 5' non-coding sequence was identified in the indigenous waxy strain MMD. Except for the deletion in the promoter region, the spontaneous mutant wax allele and non-waxy allele were identical. Such highly conserved sequences provide evidence for the recent occurrence of a deletion event in the cultivated barley gene pool. Compared to the original variety SH84, induced waxy mutant SH97 had a base substitution of a C to T in the exon 5, which converting Gln-89 of the wild-type gene into a stop codon, suggesting the involvement of a nonsense-mediated mRNA decay. These results will be helpful for understanding the mechanism of the variable amylose content in waxy cultivars of cereal species.     

Analysis of genetic distance by SSR in waxy maize

We examined the genetic diversity of 80 inbred waxy maize lines using 22 SSR molecular markers that could be used to achieve heterosis in waxy maize. Eighty inbred waxy maize lines with different phenotypes, 40 yellow, 25 white, 13 black, and two red lines were analyzed by SSR molecular marker fingerprint and cluster analysis. Using a standard genetic distance of 0.55, the 80 waxy maize inbred lines were clustered into nine groups. Among them, group II, group V, groups VII and VIII, and group IX were divided into three subgroups at a genetic distance of 0.46, into two subgroups at 0.49, into two subgroups at 0.46, and into four subgroups at 0.493, respectively. All but one of the yellow waxy maize inbred lines were clustered in groups VI, VII, VIII, and IX. Group IX (30 lines) contained 28 yellow lines; the other 11 yellow lines were distributed among groups VI, VII and VIII. Among the 25 white lines, 21 were clustered in groups III, V, VI and the third subgroup of group II. The black line N72 was in a group of its own. The black lines N75, N76 and N78 were distributed in groups VII, VIII and IX, respectively. The other nine black lines were clustered in group II. The red lines were distributed in the second subgroup of group II and there was no difference in genetic distance between them. In conclusion, there were considerable genetic differences among waxy maize inbred lines of different colors. The mean genetic distance of inbred lines of the same color was significantly less than that of lines of different colors. Therefore, we concluded that it was more accurate to determine the difference between the populations using the highly stable DNA genetic markers.     

Comparative mapping reveals a complex relationship between the pearl millet genome and those of foxtail millet and rice

Comparative genetic maps were constructed of the pearl millet genome with foxtail millet and used to describe the homoeology between the genomes of pearl millet, foxtail millet and rice. Despite the close taxonomic relationship of pearl and foxtail millet, their genomes were highly, rearranged. A comparison of the millet and rice genomes indicated that most of these rearrangements were likely to have taken place in pearl millet. Two duplications were identified in pearl millet. A duplication between the distal segments of linkage groups 1 and 4 corresponds to the ancient duplication previously identified between rice chromosome arms 11S and 12S and foxtail millet chromosomes VII and VIII. The other putative duplication, also between regions of linkage groups 1 and 4, is likely to be species-specific. The exploitation of the comparative maps in pearl millet research is discussed. Received: 10 February 1999 / Accepted: 6 July 1999     

Site-selected transposon mutagenesis at the hcf106 locus in maize.

The High chlorophyll fluorescence106 (Hcf106) gene in maize is required for chloroplast membrane biogenesis, and the hcf106-mum1 allele is caused by the insertion of a Robertson's Mutator Mu1 element into the promoter of the gene. Seedlings homozygous for hcf106-mum1 are pale green and die 3 weeks after germination, but only in the presence of Mutator activity conferred by active, autonomous Mu regulatory transposons elsewhere in the genome. When Mutator activity is lost, the mutant phenotype is suppressed, and homozygous plants have an almost wild-type phenotype. To isolate derivative alleles at the hcf106 locus that no longer require Mutator activity for phenotypic expression, we have developed a method for site-selected transposon mutagenesis in maize. This procedure, first described for Caenorhabditis elegans and Drosophila, involves using polymerase chain reaction (PCR) to screen pools of individuals for insertions and deletions in genes of known sequence. Pools of seedlings segregating for the progenitor allele hcf106-mum1 were screened by PCR for insertions and deletions associated with Robertson's Mutator. In a 360-bp target region, two new insertions and one deletion were identified in only 700 Mu-active gametes screened. One of the insertions was in the progenitor hcf106-mum1 allele and the other was in the wild-type allele, but all three new alleles were found to have break-points at the same nucleotide in the first intron. Unlike the hcf-106-mum1 progenitor allele, the deletion and one of the insertions conferred pale green seedling lethal phenotypes in the absence of mutator activity. However, the second insertion had a weak, viable phenotype under these conditions.(ABSTRACT TRUNCATED AT 250 WORDS)     

Geographical distribution and genetic analysis of phenol color reaction in foxtail millet,Setaria italica (L.) P. Beauv.

Summary Phenol color reaction was examined in a total of 376 strains of foxtail millet, Setaria italica (L.) P. Beauv., collected from different areas throughout Eurasia. Positive and negative phenotypes, and no intermediate type could be recognized by the phenol color reaction. Of 376 strains examined, 50 were positive, 319 were negative, five were mixtures of both phenotypes, and the coloration in two strains with blackish lemmata and paleae could not be distinguished. The strains that showed the positive phenotype of phenol color reaction were found in rather limited regions, while those with the negative phenotype occurred in almost all the regions. The positive phenotype occurred more frequently in the lower latitudinal regions of Asia. Genetic analysis of the F₁ and F₂ generations between the two phenotypes showed that the phenol color reaction is controlled by a single gene, and that the positive phenotype is dominant.Contribution No. 28 from the Plant Germ-plasm Institute, Faculty of Agriculture, Kyoto University, Kyoto (Japan)     

Quick Identification of the Members of the Glutamine Synthetase Gene Family from Sunflower by Simultaneous Amplification of cDNA with Degenerate Primers

A single degenerate glutamine synthetase (GS)-specific primer was used to amplify the 3′ end of cDNAs derived from different GS genes that are expressed in leaves and roots of sunflower (Helianthus annuus L. cv. Peredovic). Four types of GS cDNA (I, II, III and IV) were simultaneously amplified from leaves and five types (I, II, V, VI, VII) from roots with a minimum investment of time and experimental work. cDNAs II, III and IV encode chloroplastic isoforms as deduced by the presence of chloroplastic GS-specific features in their sequences. The rest of cDNAs codifies cytosolic isoforms. Using cDNA-specific probes and primers, homologous sequences to all GS cDNAs amplified from cv. Peredovic, except to cDNAs III and IV, were detected in the inbred line R41. This result strongly suggests that the three cDNAs for chloroplastic isoform are allelic sequences from the same locus, and since cDNA type IV contains sequences derived from cDNAs II and III, it indicates a recombinational origin. The results presented are consistent with the existence of a GS gene family in sunflower with at least five members. Four of them, named ggs1.1 to ggs1.4, codify for the cytosolic isoforms (cDNAs I, V, VI and VII). A fifth member, named ggs2, from which three allelic sequences (cDNAs II, III and IV) have been cloned, encodes the chloroplastic isoform. This revised version was published online in July 2006 with corrections to the Cover Date.     

Multiplex PCRs for assignment of Staphylocoagulase types and subtypes of type VI Staphylocoagulase

Staphylocoagulases (SCs) have been classified by the differences in antigenicity using a serological method. We have developed a system to classify them based on the nucleotide differences in SC genes (coa). The system was composed of three multiplex PCRs (M-PCRs): M-PCR:A, identifying types III, IV, VII, and VIII; M-PCR:B, identifying types I, II, V, and VI; M-PCR:C, identifying three subtypes of type VI. In this study, we found that coa genes of the serotype VI were not identical, but classified into three subtypes based on the nucleotide differences, especially in D2 and the central region: VIa, the coa gene carried by stp12 from human; and VIb and VIc, the coa genes carried by strains IFH556 and IFH514 isolated from bovine raw milk. The primer pair used in M-PCR:B was designed to identify all three subtypes of type VI coa. The results showed that coa types of 154 out of 155 Staphylococcus aureus strains from various origins assigned by M-PCR:A and B were identical to those obtained by serological methods, leaving a serotype IV strain unclassifiable. All 73 type VI strains were classified into one of three subtypes by M-PCR:C. Furthermore, we found that type VIa and VIb strains carried characteristic pyrogenic toxin superantigen genes, while no toxin genes were identified in type VIc strains, suggesting the correlation between the subtype of type VI coa gene and the carriage of genomic islands. Our results showed that these M-PCRs are convenient methods for SC typing that might be useful for epidemiological studies.     

Development and utilization of novel intron length polymorphic markers in foxtail millet (Setaria italica (L.) P. Beauv.)

Introns are noncoding sequences in a gene that are transcribed to precursor mRNA but spliced out during mRNA maturation and are abundant in eukaryotic genomes. The availability of codominant molecular markers and saturated genetic linkage maps have been limited in foxtail millet (Setaria italica (L.) P. Beauv.). Here, we describe the development of 98 novel intron length polymorphic (ILP) markers in foxtail millet using sequence information of the model plant rice. A total of 575 nonredundant expressed sequence tag (EST) sequences were obtained, of which 327 and 248 unique sequences were from dehydration- and salinity-stressed suppression subtractive hybridization libraries, respectively. The BLAST analysis of 98 EST sequences suggests a nearly defined function for about 64% of them, and they were grouped into 11 different functional categories. All 98 ILP primer pairs showed a high level of cross-species amplification in two millets and two nonmillets species ranging from 90% to 100%, with a mean of ～97%. The mean observed heterozygosity and Nei's average gene diversity 0.016 and 0.171, respectively, established the efficiency of the ILP markers for distinguishing the foxtail millet accessions. Based on 26 ILP markers, a reasonable dendrogram of 45 foxtail millet accessions was constructed, demonstrating the utility of ILP markers in germplasm characterizations and genomic relationships in millets and nonmillets species.     

Deletion commonly found in Waxy gene of Japanese and Korean cultivars of Job’s tears (Coix lacryma-jobi L.)

We isolated the entire sequence of the coding region of Waxy gene of a non-waxy accession of Job??s tears (Coix lacryma-jobi) by PCR-based methods. We also compared the entire sequences of the gene between two non-waxy accessions and three waxy cultivars and found a 275-bp deletion in the coding region (exons 10?C11) of this gene specific to waxy cultivars. We showed by PCR genotyping that this deletion is commonly found in Japanese and Korean cultivars and confirmed that this deletion resulted in lack of Wx protein. We also confirmed that this polymorphism of the gene co-segregates with phenotypes in endosperm and pollen. These results suggest that this PCR-based marker will be useful in breeding of Job??s tears and that genetic information obtained in other grass species will be also useful in genetics and breeding of Job??s tears.     

Erythrocyte glyoxalase I polymorphism in the owl monkey,Aotus

Three erythrocyte glyoxalase I phenotypes were observed in a sample of 235 karyotypically defined New World owl monkeys, Aotus. The selective distribution of glyoxalase I allele (GLO¹, GLO²) is related to the karyotype of each animal. Owl monkeys with a karyotype VI had an equal distribution of GLO¹ and GLO² genes in the population. Aotus with karyotype II, III, IV, or V had, exclusively, the GLO² allele (expressed as the fast electrophoretic phenotype), in contrast with monkeys with karyotype I or VII, which had only the GLO¹ allele (expressed as the slow electrophoretic phenotype).     

Estimating loop-helix interfaces in a polytopic membrane protein by deletion analysis.

Insertions of amino acids into transmembrane helices of polytopic membrane proteins disrupt helix-helix interactions with loss of function, while insertions into loops have little effect on transmembrane helices and therefore little effect on activity [Braun, P., Persson, B., Kaback, H. R., and von Heijne, G. (1997) J. Biol. Chem. 272, 29566-29571]. Here the inverse approach, amino acid deletion, is utilized systematically to approximate loop-helix boundaries in the lactose permease of Escherichia coli. Starting with deletion mutants in the periplasmic loop between helices VII and VIII (loop VII/VIII), which has been defined by immunological analysis and nitroxide-scanning electron paramagnetic resonance spectroscopy, it is shown that mutants with single or multiple deletions in the central portion of the loop retain significant transport activity, while deletion of amino acid residues near the loop-helix boundaries or within the flanking helices leads to complete inactivation. Results consistent with hydropathy analysis are obtained with loops VI/VII, VIII/IX, and IX/X and the flanking helices. In contrast, deletion analysis of loops III/IV, IV/V, and V/VI and the flanking helices indicates that this region of the permease differs from hydropathy predictions. More specifically, evidence is presented supporting the contention that Glu126 and Arg144 which are charge paired and critical for substrate binding are within helices IV and V, respectively.