Introduction

In the genus Cajanus with 32 species and 11 related genera, Cajanus cajan (L.) Millspaugh is the only species cultivated throughout Asia and Africa for its leguminous proteins. Although many of the closely related wild species easily cross with C. cajan, various studies have shown that pigeonpea originated from its closest wild relative C. cajanifolius (Haines) van der Maesen (Ladizinsky and Hammel 1980; van der Maesen 1980, 1986; Krishna and Reddy 1982; Pundir and Singh 1985; Panigrahi et al. 2007), most probably in India and later it spread to the continents of Africa and Australia, where some wild relatives of pigeonpea still exist.

There are many published reports indicating that C. cajanifolius is the proginator species of cultivated pigeonpea (van der Maesen 1980; Panigrahi et al. 2007). It would not only be interesting but useful to know the difference/diversity between the two. Studies revealed that there are at least 5–6 traits that separate the two species such as flower morphology, pod color and morphology, pod constriction, seed color and strophiole, and 100 seed weight that separate C. cajan from C. cajanifolius. Molecular studies using SSRs markers showed that a genetic dissimilarity index value ranging from 0.81 to 0.94 exists between the two species.

Materials and methods

Plant morphology

Four accessions of C. cajanifolius namely ICPW 28, 29, 30 and 31 together with pigeonpea cultivar ICPL 85010 were grown in a glasshouse. Morphological parameters such as branching pattern, growth habit, plant height, number of primary branches, stem color, leaf shape, color of basal petal, pattern of streaks on the petal, pod color, pod constriction, pod size, seed color, seed shape, 100-seed weight and presence/absence of strophiole were studied on randomly selected plants.

Crossability

Emasculations followed by pollinations were carried out in the morning using C. cajan cultivar ICPL 85010 as the female parent and C. cajanifolius accessions as the pollen donor. Care was taken to allow only cross pollinated pistils to remain and grow in an axil, removing all other self pollinated pistils or immature buds. Application of gibberellic acid (75 mg/l), as a cotton swab wrapped around the pistil soon after pollinations, increased pod set. Mature pods were harvested upon maturity.

Cytology

Flower buds from F1 hybrids ICPW 28 × ICPL 85010 and ICPW 29 × ICPL 85010 were squashed in 2% aceto-carmine and well spread preparations were examined. To study different stages of meiosis such as metaphase, anaphase and tetrad, at least 20 pollen mother cells (PMCs) were examined and means were calculated from them. Pollen fertility analysis was carried out by staining mature pollen grains in 2% aceto carmine. Well stained grains were counted as fertile grains and partial to unstained grains were counted as sterile.

SSR analysis

PCR amplification of microsatellite loci using 14 fluorescent-dye-labeled primer pairs was carried out in 15 μl volume. The reaction mixture contained 10 mM Tris-HCl, 50 mM KCl, 10 ng of genomic DNA, 2–4 mM MgCl2, 300–400 μM of dNTP, 1 unit of Taq DNA polymerase. Amplified products were pooled as per multiplex plan and separated on an ABI 3700 fragment analyzer. The results were evaluated using the software package Genotyper 3.7 (Applied Biosystem).

Data analysis

Analysis of the data was performed using data of 14 SSRs markers. Genetic polymorphism was measured in terms of number of alleles per locus, expected and observed heterozygosity, average genetic distance between accessions (Dg) and the polymorphic information content (PIC) using Powermarker V3 (Liu and Muse 2005). Genetic distance is a measure of the dissimilarity of genetic material between different species or individuals of the same species. Depending upon the difference and correcting the values of genetic distances for known rates of evolution, genetic distance is used as a tool to construct cluster diagrams. Genetic diversity analysis was carried out by using the program DARwin version (Perrier and Jacquemoud-Collet 2006).

Results

Morphological studies

Four accessions of C. cajanifolius ICPW 28, 29, 30 and 31 had the presence dense small trichomes on their leaves making them velvety to touch, in comparison to cultivated pigeonpea ICPL 85010, which had trichomes on their leaves but they were not velvety to touch. Variation was observed with respect to flower color. The keel petal of ICPW 28 was more yellowish and comparable to that of ICPL 85010 than in the other three C. cajanifolius accessions which had more of orange tinge in them (Fig. 1). All the four accessions of C. cajanifolius and C. cajan were erect with semi-spreading branching (Table 1) pattern. Accession ICPW 31 was taller than all the other three accessions of C. cajanifolius and C. cajan. The number of primary branches varied from 3 to 4 in all accessions of C. cajanifolius compared to 9 in C. cajan. The color of the keel petal and the streaks on them also varied between C. cajanifolius accessions and C. cajan. Pod constriction was prominent on C. cajan whereas it was slight on all the accessions of C. cajanifolius. Pod size too varied from 2.9 to 4.4 cm in different accessions of C. cajanifolius compared to a pod size 5.4 cm of C. cajan. Prominent difference between the accessions of C. cajanifolius and C. cajan was the seed color (Fig. 1) and seed strophiole. Seeds of C. cajanifolius were ash brown to black in color (Fig. 1) and it was light brown (beige) in C. cajan. Seed strophiole was prominent in all accessions of C. cajanifolius and it was absent in C. cajan. Pod morphology varied between the accessions of C. cajanifolius and C. cajan. Pods were flat in C. cajanifolius compared to C. cajan pod. The locules between the seeds were more prominent in C. cajanifolius accessions with clear cut demarcations between individual locules. Pod shattering was observed in all C. cajanifolius accessions compared to C. cajan where mature dry pod did not shatter.

Fig. 1
figure 1

C. cajanifolius accessions’s flower and seed morphology (from left to right: ICPW 28, ICPW 29, ICPW 30 and ICPW 31). a, b Flower morphology and striations on the keel petal. c Seed morphology and presence of strophiole on all the accessions

Full size image
Table 1 Study of some morphological characters in some C. cajanifolius accessions
Full size table

A dendrogram was drawn based on morphological traits and ICPW 29 and 30 showed closer relationship to each other compared to ICPW 28. ICPW 31 showed closer relationship to C. cajan than any of the accessions of C. cajanifolius. But the distance between C. cajanifolius accessions and C. cajan was distant enough for species differentiation (Fig. 2).

Fig. 2
figure 2

Dendrogram showing diversity between C. cajanifolius accessions based on morphological traits

Full size image

Molecular diversity

SSR markers were able to distinguish all the four accessions of C. cajanifolius (Fig. 3) from C. cajan. Accessions ICPW 28 showed a diversity index number of 0.44 with ICPW 30, compared to the index number of 0.61 with ICPW 29 and 0.94 with ICPW 31. This shows that accession ICPW 28 is closer to ICPW 30 than to either ICPW 29 or 31. The diversity index number between all the accessions of C. cajanifolius and C. cajan varied between 0.81 and 0.94 showing difference between C. cajanifolius and C. cajan (Table 2).

Fig. 3
figure 3

Genetic diversity in C, cajanifolius accessions as revealed by SSR marker Ccm0407. (The top line is ICPW 30, second line is ICPW31, third line is ICPW 28 and the fourth line is ICPW 29)

Full size image
Table 2 Molecular diversity between four C. cajanifolius accessions based on SSR markers
Full size table

Crossability between C. cajanifolius accessions and C. cajan

Crossability between C. cajanifolius accession ICPW 28, 29, 30 and 31 and C. cajan varied from 0.03 to 0.20% with the formation of mature pods as a result of cross pollinations. The response was much low when the 50 ppm GA3 was not applied to the base of pollinated pistils soon after cross pollinations. A maximum of 9 pods were obtained as a result of 68 pollinations in the cross ICPL 85010 × ICPW 29 and a minimum of 2 pods were obtained from 45 cross pollinations from the cross ICPL 85010 × ICPW 28 (Table 3). Formation of mature pods from cross pollinations which was less than 1%, shows that the two species are closely related but distant enough as percent pod set is low compared to pollinations between accessions of C. cajan which ranges between 15–18%, depending upon the accessions used in cross pollinations (data not shown in the table).

Table 3 Summary of the crossability between pigeonpea cultivar ICPL 85010 and four accessions of C. cajanifolius
Full size table

Cytological analysis of the F1 hybrids

Meiocytes from the crosses ICPL 85010 × ICPW 29 and ICPL 85010 × ICPW 31 showed the formation of 11 bivalents (Fig. 4a showing total homology between the parental species. Twenty percent of the meiocytes showed the formation of 7 bivalents and 2 tetravalents showing two chromosomes which are totally homologous between the parental species (Fig. 4a, b). In the cross ICPL 85010 × ICPW 28, 10% of the meiocytes showed the presence of 2 univalents which signifies that one chromosome in each parent did not have a homologous chromosome in the other parent or the divergence of one chromosome in one of the parent. Such an anomaly was present only in 10% of the meiocytes (Fig. 4c). Tetrads were observed (Fig. 4d) in both the crosses and pollen fertility in the F1 hybrids varied between 48–62%, showing closer relationship between the two species.

Fig. 4
figure 4

Meiotic study of the F1 hybrid between C. cajan × C. cajanifolius. a Eleven bivalents in F1 hybrid from the cross ICPL 85010 × ICPW 29. b Anaphase from the cross ICPL 85010 × ICPW 29. c Anaphase showing laggards in the F1 hybrid ICPL 85010 × ICPW 28. d Tetrads in the crosses ICPL 85010 × ICPW 28 and ICPL 85010 × ICPW 29

Full size image

Discussion

Morphological traits such as plant morphology, leaf glabrous-ness, pod shape, pod shattering, seed shape and color, presence of strophiole are distinguishing characters between C. cajanifolius and C. cajan. These traits separate C. cajanifolius from cultivated pigeonpea. Morphological studies showed that ICPW 31 is closer to ICPL 85010 than the other three accessions and this is evident when plant height is studied. Both are taller than the other material used in the study.

With respect to cytology cross ICPL 85010 × ICPW 29 and ICPL 85010 × ICPW 31 showed the formation of 11 bivalents and the presence of two tetravalents shows that two pairs of chromosomes are probably identical to each other but might be evolving into separate pair as tetravalents were seen in only 20% of the meiocytes. In the cross ICPL 85010 × ICPW 28 the presence of two univalents in 10% of the meiocytes shows that there may be one pair of chromosome in each parent which is distinct without a partner in the other parent. It is known that as the species evolves due to crossover the pairs of chromosomes will also evolve. Pundir and Singh (1985) did not observe seed set when C. cajan was crossed with C. cajanifolius whereas Reddy et al. (1980) obtained seed set. In the present study it was observed that pod/seed set was low (<1%), nevertheless mature pod set was observed. Low mature pod set shows that although the two species are closely related there might be other extraneous factors causing low pod set. Regular meiosis between the two species shows the closer relationship between the two.

Molecular analysis showed a greater diversity ranging between 0.81 and 0.94, between C. cajanifolius accessions and cultivated pigeonpea, and this cannot be explained by an inversion separating the two species as suggested by Pundir and Singh (1985). Published literature show that C. cajanifolius is closely related to C. cajan, and according to the present report there were at least five traits that separated C. cajanifolius from C. cajan. This is expected as species evolve, there are bound to be differences and in case of cultivated pigeonpea domestication coupled with selection could have resulted in the differences. Parani et al. (2000) have concluded that C. scarabaeoides (L.) Thouars is the progenitor species of C. cajan but their analysis does not include C. cajanifolius. Apart from the above mentioned traits that separate C. scarabaeoides from C. cajan, the trailing growth habit is an additional trait separating C. scarabaeiodes from C. cajan, showing a more distant relationship with cultivated pigeonpea. In conclusion, although C. cajanifolius is the progenitor species of C. cajan, there are at least five evident traits that separate the two species. Since molecular diversity between the two species is large, it can be concluded that there might be some minor traits which are not very evident such as leaf size and glabrous-ness etc., which might add up to the diversity between the two species.