ITS sequence variation in wild species and cultivars of pea

Polans, N.O. and Saar, D.E.                     Dept. of Biological Sci. and Plant Molecular Biology Center

                                                                                                  Northern Illinois Univ., Dekalb, IL

      An often powerful approach to characterizing the relationships among plant taxa is to compare the nucleotide sequences of their ribosomal DNA. Nuclear ribosomal DNA (nrDNA) is organized as distinct chromosomal units that are repeated thousands of times in most higher plant genomes. Each of these units contains the genes that encode the 18S, 5.8S and 26S ribosomal RNA subunits, as well as several different spacer DNA regions. The nucleotide sequence variation found in both of the internal transcribed spacer regions (ITS-1 and ITS-2, Fig. 1) is routinely used for the systematic analysis of closely related taxa, at least in part due to the high rate of evolutionary change characterizing these DNA regions (1).

        In our preliminary study of pea ITS regions (6), ITS-1 and ITS-2 DNA sequence variation was assessed for five pairs of wild and cultivated pea taxa selected to approximate the range of Pisum. The objective of that investigation was to examine the similarity of the sequences within paired accessions, the overall level of genetic variation found across the entire genus, and the topological relationships established among the five selected groups of taxa. It resulted in the following six observations: 1) very close genetic affinities throughout Pisum, with P. fulvum exhibiting the greatest degree of divergence, 2) support for the established taxonomic categories of the genus based upon identical or near identical sequences within group pairs, 3) the assignment of JI1794 as a “northern” humile, 4) the validity of northern and southern humile as closely-related, but distinct, lines, 5) the apparent independent evolution of a pea chromosomal translocation and 6) a close relationship between elatius and the cultivated sativum. Additionally, when Vicia montbrettii was included as an outgroup to Pisum in both the preliminary and present studies, phylogenetic analyses indicated that P. fulvum remained not only the most divergent pea taxon but also the most basal taxon relative to the sativum group (data not shown).

      The goal of the present study is to extend the use of ITS variation as a comparative tool to an additional 55 wild and cultivated pea taxa, both to validate our preliminary findings among a more diverse sample of the genus and to include previously unexamined pea types in these analyses.

Materials and Methods

      Pisum isolates 701-722 are from the Ben Ze’ev and Zohary (1973) collection (courtesy of J.G. Waines), JI accessions are from the John Innes collection (courtesy of M. Ambrose), cv. Alaska is from J. Mollema and Son, Inc. (Grand Rapids, MI) and cv. (Morse’s) Progress #9 is from Ferry-Morse Seeds (Mountain View, CA). P. sativum Syriacum was graciously provided by R. Jorgensen, and accessions 82-14n, A1078-234 and PI 179449 were kindly provided to this project by G. Marx and N. Weeden.

      DNA extraction, PCR amplification, gel purification, and ITS primers (ITS2, ITS3, ITS4 and ITS5m) are described elsewhere (6).  DNA sequencing is performed with either an Applied Biosystems model 373 DNA sequencer or a Beckman Coulter CEQ 2000 XL DNA analysis system. Forward and reverse DNA sequences are compared to resolve ambiguities using PC Gene software and the resulting sequences are aligned with the Clustal X computer program. Sequence data are analyzed using the PAUP computer package (7).

Results and Discussion

      The pea 18S rRNA, ITS-1, 5.8S rRNA, ITS-2 and 26S rRNA regions examined in this study contain 27, 238, 164, 213 and 22 alignable base pairs (bp), respectively, totaling 664 bp (including 451 bp of spacer DNA) for all but one of the 65 plants analyzed. The only exception to these results involves a P. sativum Syriacum accession that contains an additional guanine at ITS-2 position number 582. Ambiguous or polymorphic pyrimidine and purine sites are denoted by the IUPAC/IUB symbols “Y” and “R,” respectively. Of the 664 total bp sequenced for each of the individual plants, 640 (>96%) of these sites are constant among the 64 pea taxa. Of the 451 ITS bp sequenced, 428 (>94%) of these sites are constant. Only 24 of the total sites are polymorphic (and only 21 are parsimony informative), reaffirming both the very close evolutionary relationships that must exist within the genus and the limited ITS information available with which to differentiate pea taxa. In this study, ITS-1 contains 14 of the polymorphic sites, as compared with nine found for ITS-2 and one polymorphic site located just within the 5.8S rRNA coding region (Table 1).

Table 1. Variable ITS sites for wild and cultivated taxa of pea.

 *     In the 5'->3' direction (see Fig. 1) beginning with those bases nearest primer ITS5m. Position 267 is assigned to the 5.8S rRNA coding region.

(1)  JI241 is listed as ssp. humile, but it displays ssp. sativum ITS characteristics.

(2)  JI198 and JI1398 are listed as ssp. elatius, but they display ssp. sativum ITS characteristics.

(3)  JI1096 and JI2055 are listed as ssp. elatius, but they display unique ITS variation at several sites

Parentheses around four JI accessions indicate taxonomic nomenclature not supported in this table.

       A compilation of the 24 variable nrDNA sites is delineated for all 65 pea taxa in Table 1, accompanied by corresponding GenBank accession numbers for the retrieval of complete sequences. The table is organized in accordance with the two commonly recognized species of pea (2-4), the more divergent P. fulvum and the typically cultivated P. sativum. The former is represented by eight identical nrDNA sequences, while the latter is differentiated as four subspecies: humile, elatius, abyssinicum and sativum. Subspecies humile is further subdivided by northern and southern populations as described by (2). There are five pea accessions characterized as questionable taxonomic assignments solely based on their nrDNA variation, and there are also differences distinguishing from one another the two “Syriacum” accessions surveyed. The four subspecies and 52 assigned accessions of P. sativum are further arranged in Table 1 by the number of unambiguous base changes each possesses relative to the invariant P. fulvum accessions. The number of base differences separating fulvum from the 52 sativum accessions ranges from 11 to 17, with 10 of these sites being unique to fulvum. JI1096, an elatius accession displaying unique ITS variation at several sites, shows 18 base differences with fulvum. The subdivisions of P. sativum are listed in the following order based on their base pair differences with fulvum: northern humile (11 base changes), southern humile (11-12 base changes), elatius and abyssinicum (13-14 base changes each), and sativum (14-17 base changes). Named cultivars of sativum usually display 15 or 16 base changes.

      A Neighbor Joining (NJ) distance analysis of these data is presented in Fig. 2 to provide a basic illustration of the associations suggested in Table 1, while also including such influences as the multiple polymorphisms found at ITS-1 sites 132 and 234. No attempt is made, however, to infer evolutionary relationships among the 65 taxa, given the relatively few parsimony informative sites available to the analysis. In the figure, only fulvum, northern and southern humile and a pair of elatius accessions maintain distinct group associations. Ten of the 21 parsimony informative sites differentiate fulvum from the much larger sativum ingroup. Within sativum, the two northern humile accessions display completely identical nucleotide sequences (at 664 sites), while the southern humile differ at a single site and show ambiguity at several others. Only two elatius accessions (JI 1096 and JI 2055), displaying four unique sites and the largest overall numbers of sequence differences with fulvum, group separately from the remaining sativum subspecies. These remaining accessions group roughly based on possessing 14, 15 or 16 base differences with fulvum. Most of the other elatius and all four abyssinicum are found in the first group, along with approximately a dozen sativum and the single questionable humile accession. The latter two groups principally comprise sativum, including most of the named cultivars.

Fig. 2. Neighbor Joining phylogram of 65 wild and cultivated pea taxa based on 24 variable nrDNA sites (23 ITS and one 5.8S rRNA). Number of base pair differences from P. fulvum (as shown in Table 1) are indicated in parentheses. Branch length distances are drawn with reference to the 0.1 length standard.

According to Fig. 2, elatius and abyssinicum are the closest taxa to the cultivated sativa, despite the fact that northern humile has been postulated the closest wild progenitor of the cultivated pea based in part on a shared chromosomal translocation (2) and detailed chloroplast studies (5). Other, larger data sets (not shown) place northern humile closer to sativum, but they do not support northern humile as the taxon closest to the cultivars. Thus, the present study largely supports the conclusions from our previous work (6): generally very close relationships within Pisum, with P. fulvum clearly displaying the greatest divergence; JI 1794 classified as a “northern” humile; northern and southern humile as closely-related, but distinct, taxa; and the independent evolution of a pea chromosomal translocation. The study also supports distinct taxonomic categories for fulvum and for northern and southern humile; however, the ITS sequence variation obtained from this investigation is too limited to separate unambiguously the very close relationships among elatius, abyssinicum and sativum. Further efforts are needed to resolve these relationships and to clarify the taxonomic assignments of the few questionable accessions addressed in this study.

Acknowledgement: We thank Scott Grayburn for his DNA sequencing skills. This work was supported by funds from the Department of Biological Sciences and the Plant Molecular Biology Center, Northern Illinois University.

 1.      Baldwin, B.G., Sanderson, M.J., Porter, J.M., Wojciechowski, M.F., Campbell, C.S. and Donoghue, M.J.  1995.  Ann. Mo. Bot. Gard. 82: 247-277.

2.      Ben Ze’ev, N. and Zohary, D.  1973.  Israel J. Bot. 22: 73-91.

3.      Hoey, B.K., Crowe, K.R., Jones, V.M. and Polans, N.O.  1996.  Theor. Appl. Genet. 92: 92-100.

4.      Marx, G.A.  1977.  In: Sutcliffe, J.F. and Pate, J.S. (eds.).  Physiology of the Garden Pea.  Academic Press, New York, pp 21-43.

5.      Palmer, J.D., Jorgensen, R.A. and Thompson, W.F.  1985.  Genetics 109: 195-213.

6.      Saar, D.E. and Polans, N.O.  2000.  Pisum Genetics 32: 42-45.

7.      Swofford, D.L.  1998.  PAUP, Version 4.0b4a. Sinauer Associates, Sunderland, MA.