| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Agronomy, Univ. of Wisconsin-Madison, 1575 Linden Drive, Madison, WI 53706
b USDA-ARS Cereal Crops Research Unit and Dep. of Agronomy, Univ. of Wisconsin-Madison, 1575 Linden Drive, Madison, WI 53706
c Dep. of Crop and Soil Science, Oregon State Univ., Corvallis, OR 97331
* Corresponding author (cahenson{at}wisc.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
-D-glucans and the pH activity optima of Morex and Steptoe rBmy2s were the same and not significantly different from barley rBmy1. The Morex rBmy2 was 7°C more thermostable than the Steptoe rBmy2, determined by differences in their T50 values, and is more thermostable than any reported wild-type ß-amylase1. Three amino acid differences were identified between the two Bmy2 sequences and the contributions to enzyme thermostability evaluated by site-directed mutagenesis. Examination of mutant enzymes with one amino acid substitution revealed that each of the three residues contributed 
3°C to the thermostability of the Morex wild-type rBmy2. Mutant enzymes with two amino acid substitutions contributed 5.6°C, and the triple amino acid mutant enzyme contributed 8.7°C to thermostability. To date, no quantitative trait loci (QTL) for malting quality traits have been associated with the bmy2 locus. Should an association be discovered, the Morex bmy2 allele, containing D238, M337, and Q362, provides a discrete signature of a thermostable ß-amylase2 that could be targeted for marker assisted selection.
Abbreviations: ESD, early seed development specific • MW, molecular weight • pI, isoelectric point • QTL, quantitative trait loci • SNP, single nucleotide polymorphism
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
-1,4-D-glucosidic bonds, releasing maltose from the nonreducing ends of a variety of polyglucans. ß-Amylase is one of the four carbohydrases involved in starch degradation during seed germination. Similarly, ß-amylase functions to degrade starch in the industrial processes malting and mashing, where the sugars produced are used to support fermentative metabolism by brewer's yeast. The mashing process takes place at temperatures high enough to render several of the carbohydrases, including ß-amylase, inactive. Hence, a concerted effort has been made to identify alleles encoding ß-amylases with enhanced thermostability or to create such enzymes via mutagenesis (Clark et al., 2003; Eglinton et al., 1998; Gunkel et al., 2002; Kaneko et al., 2001; Kihara et al., 1998; Ma et al., 2000a, 2001; Yoshigi et al., 1995). There are at least two genes encoding ß-amylases in barley (Kreis et al., 1987, 1988). The bmy1 gene located on chromosome 4H encodes ß-amylase1 (Bmy1), an enzyme unique to the endosperm tissue in the seeds of the Triticeae family of cereals [barley, wheat (Triticum aestivum L.), and rye (Secale cereale L.)] (Daussant et al., 1994; Kreis et al., 1988). Barley Bmy1 is well characterized due to its involvement in the starch degradation pathway that is the basis of the brewing industry (Allison and Swanston, 1974; Clark et al., 2003; Hejgaard, 1976; Kreis et al., 1987; Lauriere et al., 1986; Lundgard and Svensson, 1987; Ma et al., 2000a, 2000b; Nishimura et al., 1987; Ziegler, 1999). The ß-amylase2 (bmy2) gene, located on chromosome 2H (Kreis et al., 1988), encodes ß-amylase2 (Bmy2) that is present in all cereal plants and has been termed "ubiquitous ß-amylase" (Daussant et al., 1991, 1994). Shewry et al. (1988) showed that, in barley, a ß-amylase isozyme was expressed in all tissues tested, hence ubiquitous, and was distinct from the endosperm-specific ß-amylase that arises during seed development.
ß-Amylase2 from wheat leaves is composed of five isoforms (Wagner et al., 1999; Zemanova et al., 2002; Ziegler et al., 1994, 1997). These isoforms were derived from a single, full-length enzyme that underwent posttranslational proteolytic processing at the C-terminal end to generate smaller isoforms (Wagner et al., 1999; Zemanova et al., 2002; Ziegler et al., 1997). None of these ß-amylases were biochemically characterized other than to show that they produced maltose from amylopectin.
A ß-amylase2 from rye has also been studied. The bmy2 mRNA was found in the very early stages of kernel development and disappeared at 15 to 20 d after anthesis (Rorat et al., 1995). ß-Amylase activity increased then decreased early in development, corresponding to when the bmy2 mRNA appeared and disappeared (Rorat et al., 1995). The rye ß-amylase2 enzyme was not biochemically characterized.
A ß-amylase with unique temporal and tissue expression patterns was reported by Jung et al. (2001). This enzyme was termed the early seed development specific (ESD) ß-amylase because its mRNA was highly expressed at 5 d after anthesis, was undetectable during later stages of seed development, and was undetectable in germinating seeds, roots, and etiolated and fully expanded green leaves. The nucleotide and deduced amino acid sequences were more than 90% identical to the ubiquitous ß-amylases of wheat (Wagner et al., 1999) and rye (Sadowski et al., 1993), and only 76% identical to Bmy1 from barley seeds. The ESD ß-amylase enzyme from barley was not characterized.
The objective of the work presented here was to characterize the barley ß-amylase2 enzyme. To accomplish this, E. coli was transformed with bmy2 cDNAs, made from two barley genotypes, and recombinant proteins were produced. The thermostabilities, substrate specificities and pH activity profiles of the recombinant ß-amylase2 enzymes are presented here and compared with the properties of the endosperm-specific ß-amylases. The effects of the three amino acid differences between the two Bmy2 sequences on the thermostability of Bmy2 were determined using site-directed mutagenesis.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Sequencing and Sequence Analysis
The bmy2 cDNAs cloned from Morex (Genbank Accession no. AY835429) and Steptoe (AY835430) were sequenced by the Interdisciplinary Center for Biotechnology Research, University of Florida, Gainesville, FL, and aligned using the GAP program in the GCG Wisconsin Package Version 10.3 (Accelrys Inc., San Diego, CA). The theoretical molecular weights (MWs) and isoelectric point (pI) values for the Morex and Steptoe Bmy2 sequences were calculated using the Compute pI and MW tool on the ExPASy Molecular biology website (http://us.expasy.org/tools/pi_tool.html, verified 11 Apr. 2005). Sequence-based structure predictions were made based on Rasmol views of crystalline Bmy1 (Mikami et al., 1999) and by the use of Peptide Structure (GCG).
Alignment of ß-amylase sequences from multiple plant species was done using the Pileup program in GCG. Sequences examined for Table 1 are identified in the legend. Additional sequences studied were from alfalfa (AF026217, Medicago sativa L.), Arabidopsis (AJ250341, D43783, and M73467), Brassica napus L. (AF319168), potato (AF393847, Solanum tuberosum L.), rice (NM_197783, Oryza sativa L.), soybean [D50866, Glycine max (L.) Merr.], and sweet potato (D12882, Ipomoea batatas (L.) Lam. var. batatas). The percentage differences between the ß-amylase sequences were calculated using the Distances program in GCG.
|
The QuikChange II Site Directed Mutagenesis Kit (Stratagene, La Jolla, CA) was used to generate all other mutations. Mutant enzymes are identified by the first letter representing the original residue, the number identifying the position of the mutation in the sequence and the second letter representing the replacement residue. To generate the D238G single mutant, the Morex wild-type bmy2 cDNA was the template with the oligonucleotides TGGAGAGTACAATGGCACCCCTGAGAAAACCC and GGGTTTTCTCAGGGGTGCCATTGTACTCTCCA as the forward and reverse primers, respectively. To generate the M337T single mutant, the Morex wild-type bmy2 cDNA was again the template and with oligonucleotides AGGCATCATGCTAGCACGAACTTCACTTGTGCAG and CTGCACAAGTGAAGTTCGTGCTAGCATGATGCCT as the forward and reverse primers, respectively. To generate the Q362R single mutant, the Morex wild-type bmy2 cDNA was the template with the oligonucleotides CAGAAGAACTAGTCCGACAGGTGCTGAGTGCT and AGCACTCAGCACCTGTCGGACTAGTTCTTCTG as the forward and reverse primers, respectively. To create the D238G/M337T double mutant, the D238G single mutant DNA was the template and the M337T forward and reverse oligonucleotides were the mutagenic primers. The M337T/Q362R double mutant was created using the M337T single mutant DNA as the template and the Q362R forward and reverse oligonucleotides as the mutagenic primers. The D238G/Q362R double mutant was created using the Q362R single mutant DNA as the template and the D238G forward and reverse oligonucleotides as the mutagenic primers. Finally, the D238G/M337T/Q362R triple residue mutant was created using the D238G/M337T double mutant DNA as the template and the Q362R forward and reverse oligonucleotides as the mutagenic primers. All mutations created were confirmed by sequencing.
Expression in Escherichia coli
The bmy2 cDNA from Morex was subcloned into the pET21b expression vector (Novagen Inc., Madison, WI) at the BamHI/HindIII sites and the bmy2 cDNA from Steptoe was subcloned into the pET21b vector at the BamHI/EagI/NotI sites in the multiple cloning site. Expression of rBmy2 proteins was induced by 1 mM isopropyl ß-D-thiogalactopyranoside. Cultures were grown for 6-h postinduction at 37°C in an orbital shaking incubator. Cells were concentrated by centrifugation (4000 x g, 20 min, 4°C) and then suspended in 25% of the original culture volume of cold 20 mM Tris-HCl (pH 8). The suspension was centrifuged a second time under the same conditions, and the resulting cell pellets were frozen at –80°C. Cells were lysed with the BugBuster protein extraction reagent (Novagen, Inc., Madison, WI) using a ratio of 5 mL reagent to 1 g of wet cell paste. The slurry was shaken at room temperature for 20 min. Cellular debris was removed by centrifugation (16700 x g, 20 min, 4°C). The supernatant was used to characterize the rBmy2 enzymes.
Thermostability Determination
Four to eight separate colonies expressing rBmy2 were used to get crude preparations for determination of enzyme thermostability. Crude extracts were prepared as described in the "Expression in Escherichia coli" section. The crude extracts were dialyzed against 50 mM sodium acetate (pH 5.2), hereafter referred to as Buffer A, for 18 h at 4°C. To obtain initial velocities, Morex rBmy2 crude extracts were diluted 1:1000 and Steptoe extracts were diluted 1:10 with Buffer A containing 1 mg mL–1 Bis (trimethylsilyl) acetamide, hereafter referred to as Buffer B. Temperature treatments were for 10 min at temperatures ranging from 0 to 70°C. After cooling to room temperature, samples were immediately assayed at 40°C for residual activities using the Betamyl kit (Megazyme International Ireland Ltd., Co. Wicklow, Ireland). It was determined before the rBmy2 thermostability experiments that E. coli transformed with the pET21b vector alone did not produce enzymes able to degrade the Betamyl substrate.
Samples of the rBmy2 single, double, and triple mutants were prepared as described for wild-type rBmy2 in the "Expression in Escherichia coli" section. The mutant enzymes were diluted 1:1000 with Buffer B with the exception of two colonies of the M337T/Q362R double mutant that were diluted 1:500. The thermostabilities of the mutant rBmy2s were determined as described for wild-type rBmy2s. Enzyme activity data at each temperature were expressed as a percentage of the activity remaining after incubation at 0°C, which was set equal to 100%. The average rBmy2 activities at 0°C were 0.12 ± 0.04 µmoles of p-nitrophenol min–1mL–1 crude Morex preparation and 0.002 ± 0.002 µmoles of p-nitrophenol min–1 mL–1 crude Steptoe preparation.
For statistical analysis a two-tailed Student's t test was performed with each colony as a separate replication and assuming unequal variance. The mean T50 and T0 values and standard deviations were calculated using each colony as a separate replication.
Determining Enzyme Activity as a Function of pH
Three separate colonies expressing rBmy2 were used to get crude preparations, as described in the "Expression in Escherichia coli" section, which were dialyzed against Buffer A. The dialyzed rBmy2 extracts were incubated in the presence of 2% (w/v) buffered (pH 3–10) boiled soluble starch for 5 and 20 min at 30°C. After incubation, reactions were boiled and the amount of reducing sugars produced determined (Nelson, 1944). The nmoles of maltose produced were calculated using standard curves of maltose solutions at pH 3 to 10. The means and standard deviations at each pH were calculated using each colony as a separate replication.
The samples used in control assays for this experiment were dialyzed extracts of colonies transformed with the pET21b vector without a bmy2 cDNA insert. These assays were conducted for 0 and 60 min at 30°C. These colonies had a maximum of 0.3% of the activity of the Steptoe rBmy2 and a maximum of 0.01% of the activity of the Morex Bmy2 (data not shown). The Morex and Steptoe rBmy2 rates were not corrected for the control rates because it was a small percentage of the rBmy2 rates.
Determining Enzyme Substrate Specificity
Three separate colonies expressing rBmy2 were used to get three enzyme samples which were dialyzed against Buffer A and diluted with Buffer B. Assays were with a range of substrates, all at 5 mg mL–1 concentrations, at 40°C for 2.5 and 7.5 min. Reactions were terminated by boiling for 5 min. Reaction mixtures were separated on a CarboPac PA-1 column (250 by 4 mm, Dionex, Sunnyvale, CA) using a Shimadzu HPLC (VP series HPLC system) following the method of Clark et al. (2003). Maltooligosaccharides were identified by co-chromatography with standards. Means and standard deviations were calculated assuming each colony as a separate replication.
The samples used in controls assays for these experiments were dialyzed extracts of colonies transformed with the pET21b vector without a bmy2 cDNA insert. Control assays were incubated for 2.5 and 30 min at 40°C with each substrate. The assays were boiled to stop any reaction and the amounts of carbohydrates were determined using the HPLC as described above. The maximal hydrolysis rate of the control assays was 1.7% of the rate of maltotetraose hydrolysis by Steptoe rBmy2. All other control assays had rates of hydrolysis < 0.5% of either the Morex or Steptoe rBmy2 hydrolysis rates.
bmy2 and Malting Quality QTL
The bmy2 locus has been mapped directly to chromosome 2H in the Steptoe x Morex mapping population and assigned to BIN7 (http://barleygenomics.wsu.edu/arnis/linkage_maps/maps-svg.html, accessed 10 Dec. 2004, verified 11 Apr. 2005). To determine if genetic variation at the bmy2 locus is associated with malting quality QTL effects, we used search functions for BINs 6, 7, and 8 in two resources: the Barley QTL Summary (http://www.barleyworld.org/northamericanbarley/qtlsummary.php, accessed 10 Dec. 2004, verified 11 Apr. 2005) and BeerGenes (http://gnome.agrenv.mcgill.ca/bg/, accessed 10 Dec. 2004, verified 11 Apr. 2005).
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The Bmy2 sequences from barley, rye, wheat, rice, corn (Zea mays L.), and the barley ESD ß-amylase (shown in italics in Table 1) exhibited greater allelic variation than did the Bmy1 sequences. The allelic variation between the Triticeae Bmy2 sequences ranged from 0.59% (between the Morex and Steptoe Bmy2s) to 8.35% (between the ESD and rye Bmy2s). Comparing the Triticeae and non-Triticeae grasses (corn and rice) Bmy2 sequences revealed allelic variation ranging from 11.7 to 17.6%.
The Bmy1 and Bmy2 sequences from Triticeae and non-Triticeae grasses differed 18.4 to 23.5% (shown in black standard font in Table 1). Although two different genes encode Bmy1 and Bmy2, the sequences only vary by a maximum of 23.5%. ß-Amylase sequences from soybean, alfalfa, sweet potato, potato, rice, Arabidopsis, and Brassica napus were also compared (data not shown). The percentage differences between these sequences ranged from a minimum of 33.7% to a maximum of 63.6%, suggesting different gene(s) encode these proteins. Recent work has shown the presence of nine different genes encoding ß-amylases in Arabidopsis (Kaplan and Guy, 2004).
The multiple sequence alignment also revealed several motifs (Table 2) that distinguish Bmy1 from Bmy2. In addition to the well-documented glycine rich repeat sequence at the C-terminal end of Bmy1 that is absent from Bmy2, there were three other motifs that differentiated Bmy1 from Bmy2 sequences. Most of these differences were single or double residue differences. On the basis of data presented in Tables 1 and 2 and the work of Jung et al. (2001), we conclude that the ESD ß-amylase is a ß-amylase2.
|
|
Thermostability Comparison of Morex and Steptoe Wild-Type rBmy2 Enzymes
The thermostability profiles of the two wild-type rBmy2s are shown in Fig. 2A. The Morex rBmy2 was significantly more thermostable at 57 and 60°C (P = 1.1 x 10–16 and 2.1 x 10–18, respectively) than the Steptoe rBmy2. The Morex rBmy2 retained >90% of its activity at 57°C, whereas the Steptoe rBmy2 had only 5% of its activity remaining at the same temperature. The T50s (the temperature at which 50% of the enzyme's activity remains) were 62.4 ± 0.8 and 55.7 ± 0.4°C for Morex and Steptoe rBmy2s (Table 3), a difference of nearly 7°C. The Morex rBmy2 T50 is the highest T50 reported for any wild-type ß-amylase from barley.
|
|
The T50 and T0 (the temperature at which no enzyme activity remains) values for the three single mutant rBmy2s were calculated. The T50 and T0 of the D238G, M337T, and Q362R single mutant enzymes decreased significantly (P < 0.01 in all cases) compared with the Morex wild-type rBmy2 values (Table 3). A comparison of the T50 values of the three single mutant enzymes shows that M337T was the most thermostable with Q362R being the second most thermostable and D238G being the least thermostable of the three single mutant enzymes.
No Bmy2 has been crystallized; therefore, secondary structure predictions were made for Positions 238, 337, and 362 of Bmy2 and compared with the known structure of Bmy1, which is 80% identical to Bmy2. For Positions 238 and 362, the predicted structure of Bmy2 was the same as the actual structure found in Bmy1. The residue at Position 238 was located in a ß-turn on the surface of the protein. The residue at Position 362 was located in the middle of an -helix in a cleft on the surface of Bmy1. The residue at Position 337 was located in a ß sheet in the interior of the protein in the actual structure of Bmy1 while the prediction for Position 337 of Bmy2 was in a turn.
The residues at Positions 238 and 362 were hypothesized to be more influential on the thermostability differences between the Morex and Steptoe rBmy2 enzymes than the residues at Position 337. This was based on the fact that at 238, the change was from a charged residue in Morex to an uncharged residue in Steptoe. And, the change at 362 was from an uncharged residue in Morex to a positively charged molecule in Steptoe. These changes were hypothesized to disrupt noncovalent interactions with neighboring side chains and result in loss of secondary structure and lower thermostability for the Steptoe enzyme. In contrast, the change at Position 337 was from one amino acid with a fairly unreactive side chain to another amino acid with an unreactive side chain. The changes at Positions 238 and 362 did reduce thermostability of the Morex rBmy2 more than the change at Position 337 (Table 3, single mutants).
The thermostability profiles of the three double residue mutant enzymes (D238G/M337T, M337T/Q362R, and D238G/Q362R) are shown in Fig. 2C. The double mutants showed even greater decreases in thermostability compared to the Morex wild-type rBmy2 than did the single residue mutants. The T50 and T0 values of the three double mutant enzymes decreased 5.5 to 5.7 and 4.9 to 5.6°C, respectively, compared with the Morex wild-type rBmy2 T50 and T0 (Table 3). No double mutant enzyme exhibited the thermostability of the Morex wild-type rBmy2, demonstrating that all three residues are required to realize the Morex wild-type rBmy2 phenotype.
The thermostability profile of the triple residue mutant rBmy2 is shown in Fig. 2D. This mutant had significantly (P < 0.01) reduced thermostability at 50 to 63°C compared with the Morex wild-type enzyme. The triple mutant enzyme's thermostability was also significantly reduced (P < 0.05) at 50 to 55°C compared with the Steptoe wild-type rBmy2. The T50 of the triple mutant decreased 8.7°C and the T0 decreased 8.9°C compared with the Morex wild-type rBmy2 (Table 3).
The Morex wild-type rBmy2 enzyme had 16 extra amino acids attached to its N-terminal end, artifacts from subcloning into the expression vector, while the Steptoe wild-type rBmy2 had six extra amino acids on its N-terminal end. Therefore, in addition to the amino acid differences at Positions 238, 337, and 362, there were 10 more amino acids on the N-terminal end of the Morex rBmy2 than the Steptoe rBmy2. The triple mutant, which converted the Morex sequence to a Steptoe sequence, showed that the extra 10 amino acids present at the N-terminal end of the Morex rBmy2 do not impart additional thermostability to the enzyme. A comparison of the thermostability of the triple mutant to that of the Steptoe rBmy2, which have the same residues at Positions 238, 337, and 362, showed the extra 10 amino acids were likely responsible for the reduced thermostability of the triple mutant. The high thermostability of the Morex wild-type recombinant ß-amylase2 enzyme was a result of the three amino acids present at Positions 238, 337, and 362 in its sequence.
Effect of pH on ß-Amylase2 Activity
The effect of pH on the Morex and Steptoe rBmy2 activities was determined (Fig. 3). There were no significant differences between the rBmy2 activities as a function of pH. Both enzymes showed a broad peak of activity from pH 5.0 to 6.5. Industrial starch conversion typically takes place at a pH between 5 and 6 (Kunze 1999). Both enzymes would be operating within their optimal pH during industrial starch hydrolysis.
|
|
-D-glucans by the Morex and Steptoe rBmy2s are shown in Table 4. The Morex rBmy2 hydrolyzed the maltodextrin series with 3 to 7 glucoses and boiled soluble starch at relative rates of 0, 125, 144, 214, 104, and 100%, respectively. Maltotriose does not induce the correct fit of enzyme-substrate complex that leads to hydrolysis likely because this oligomer is too short to occupy the required number of subsites within the active site. Thoma and Koshland (1960) showed that sweet potato ß-amylase had four glucose binding sites within its active site. The Morex enzyme did not hydrolyze oyster glycogen or liver glycogen at detectable levels. Glycogen, the glucose storage polymer found in animals, is more highly branched than starch, the glucose storage polymer in plants, and ß-amylase is unable to hydrolyze the -1,4 linkages that are next to branch points. The oyster and liver glycogens likely have too many branch points to allow the active site of ß-amylase2 to bind the -1,4-glycosidic linkages in these polymers in productive enzyme-substrate complexes. The Steptoe rBmy2 hydrolyzed the maltodextrins of 3-7 glucoses and boiled soluble starch at relative rates of 0, 117, 188, 216, 117, and 100%, respectively. Again, maltotriose is likely too short of a glucose chain to be an effective substrate for barley ß-amylase2. The Steptoe enzyme also did not hydrolyze oyster glycogen and liver glycogen at detectable levels. |
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication December 13, 2004.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
