Observation of a phenomenon resembling hybrid dysgenesis,
in a wild pea subspecies Pisum sativum ssp. elatius

Bogdanova, V.S. and Berdnikov, V.A.     Inst. of Cytology and Genetics, Russian Academy of Sciences

                                                                                                                         Novosibirsk, Russia

      In a series of crosses of wild peas with cultivated forms we found that one line, VIR320, was peculiar in that the characters of hybrid plants produced when VIR320 was used as a female parent differed from those when it was used as the male parent. This line has been assigned to P. sativum ssp. syriacum (which is a synonym of P. sativum ssp. elatius var. pumilio). It originated from Palestine and was collected on an early expedition of N.I.Vavilov. When VIR320 was used as the female parent in crosses, the majority of the F₁ plants were abnormal to some extent. Hybrid plants pro-duced from the reciprocal cross were normal. Observed abnormalities included chloro-phyll pigmentation (sectors of chlorotic tissue on the normal background and vice versa), reduced leaves, and sterile flowers (Fig.1).

      We made crosses between VIR320 and 47 accessions, using the former as the female parent. The number of F₁ plants obtained from each cross varied from 1 to 15. Of the alternate parents, 32 were from non-cultivated acces-sions and 15 from cultivated types. In only six cases (L100, VIR2521, JI1794, L99, P008 and WT301) abnormalities were not observed in the F₁ generation, and in all these cases the alternate parent represented a wild form of P. sativum or P. fulvum (WT301). The degree of abnormality observed varied, depending on the accession used as the pollen parent. There was a positive correlation between the degree of the abnormality and the fraction of F₁ plants exhibiting an abnormality, with the frequency of abnormal plants varying between 0 and 20%.

      It proved useful to divide all accessions into two groups, those 'compatible' with VIR320 (producing only normal progeny in crosses) and those 'incompatible' with VIR320 (producing abnormal progeny in crosses). The phenotypes resulting from crosses with these two types of alternate parent are schematically presented in Fig.2. One of the accessions that caused the most severe anomalies in the progeny was VIR1451 (Arkhangelsk region) which was used as a tester in the experiments described below.

Fig.2.  Schematic model of phenotypes observed in seedlings from crosses with incompatible (A) and compatible (B) lines.

      We investigated the mode of inheritance of the ability of line VIR320 to induce aberrations in the progeny using two types of crosses: 1) crosses with incompatible lines; 2) crosses with compatible lines.  We investigated segregation in F₂, and the trait we scored was the ability of an F₂ plant to produce aberrant progeny when pollinated with foreign pollen.  In the presentation of results we use the ♀ and ♂ symbols to denote maternal and paternal lines, respectively.  The line VIR320 is abbreviated to 320.  In all crosses, the maternal parent is indicated first.

Crosses with incompatible lines

        Each plant from an F₂ population of 120 plants derived from the cross WL1238 x 320 was pollinated with VIR1451. None of the 821 plants produced from these crosses was phenotypically abnormal. We can summarize these results in the following scheme:

F₂[♀ x 320] x VIR1451 —> normal

      When VIR320 plants were pollinated using 43 F₂ plants from the cross WL1238 x 320 the F₂ plants appeared to resolve into three classes (Table 1), one that produced only abnormal progeny, one that produced both normal and abnormal progeny, and one that produced only normal progeny.

      It appears that in incompatible crosses either the presence of VIR320 genes in the pollen or the presence of VIR320 cytosol is the factor responsible for the appearance of the observed abnormalities.

Crosses with compatible lines

      In this type of cross the ability of VIR320 to induce aberrations in the progeny was approximately the same regardless of the direction of the cross (Table 2). The line L100 was used as the compatible parent.

Table 2. Results of pollinations of F₂ plants from compatible crosses with VIR1451 pollen.

      In the experiment with the ♀ x 320 F₂ population, the segregation of phenotypes is very close to a 1:2:1 ratio, suggesting that there is one (or several closely linked) nuclear factor responsible for aberrations observed.  We conclude that the ability of the line VIR320 Pisum sativum ssp. elatius to cause aberrations in the progeny is determined by the nuclear genome, but this effect can be observed only in the presence of compatible cytoplasm. Abnormal progeny occurs in crosses where maternal plants refer to the line VIR320 or its hybrids with compatible lines.

      Phenotypic aberrations observed by us are very similar to those described by Lutkov (4) in distant hybrids of pea. However, in experiments of Lutkov, abnormal plants were observed in F₂ and F₃ generations, while we could observe the phenomenon only in the F₁ after pollination of VIR320 (or its hybrids) with a tester pollen.

      There are some types of crosses where phenotypes of the resulting progeny depend on the direction of the cross. For example, cytoplasmic male sterility (CMS), manifested as inability to produce fertile pollen or functional anthers, is maternally inherited although it is influenced by nuclear genes. CMS is commonly associated with changes in mitochondrial gene expression (reviewed in 6). In fact, VIR320 line acts like a source of sterility-inducing cytoplasm, whereas compatible lines perform as if carrying nuclear fertility restorer genes. However, segregation in the F₂ of compatible crosses indicates that we are dealing with a nuclear factor that can be transmitted both maternally and paternally, although the source of cytoplasm plays an important role. The issue appears to be involved in the phenomenon of cytoplasmic incompatibility (CI) described for some arthropod species and associated with endocellular bacteria of Wolbachia species. CI is maternally transmitted and appears as reproductive aberrations when an infected male is crossed with a female either devoid of Wolbachia or harboring another strain of bacteria (see for example, 1). This aberration can be overcome by antibiotic treatment against microbial invasions. We treated seeds of VIR 320 and tester lines with doxycycline (25 μg/ml and 40 μg/ml). The resulting seedlings were pale-green indicating that the treatment was effective; however, the treatment did not abolish aberrations in the progeny.

      Thus, we suppose that the phenomenon observed most closely resembles hybrid dysgenesis described in Drosophila (3). In flies, the dysgenic syndrome appears in hybrid progeny when mobile genetic elements of certain type come from the male parent and the hybrid genome conflicts with cytoplasm of the oocyte produced by the female lacking active copies of this mobile element. The resulting mobilization of the transposon in the hybrids causes multiple genetic disorders including complete or partial sterility. It is known that plant genomes may contain numerous copies of transposable elements (5). By analogy with Drosophila, we can suppose that the genomes of the majority of pea samples are saturated with some type of transposons which are, however, inactive due to accumulation in the cytoplasm of a repressor encoded on the DNA of the mobile element. Male gametes lack the repressor or contain very little of it.  A cross involving two such lines does not cause dysgenesis because mobile elements delivered by the male gamete are controlled by the repressor produced by transposons of the egg. We suppose that the genome of the line VIR320 is devoid of mobile elements capable of repressor production, therefore, when an egg of this line meets with a male gamete of an incompatible line, transposons of the latter begin active transpositions and repressor production. As for compatible wild samples, we suppose that their genome harbors defective mobile elements not capable of active transposition but producing a factor of repression which is accumulated in cytoplasm thus making it non-reactive. Therefore, crosses of these lines with common testers produce normal progeny, and the off-spring from the crosses with VIR320 also does not suffer from genetic disorders caused by transpositions of mobile elements.

      It should be noted that crosses of Pisum sativum with other pea taxa such as P. sativum ssp. abyssinicum and P. fulvum are successful (with rare exceptions) only when P. sativum is used as the female plant (2, 7). In addition, an F₂ may segregate for aberrant plants (4) similar to those described in the present study. It is possible, therefore, that nuclear-cytoplasmic conflict contributes to interspecies isolation and, thus, to speciation.

Acknowledgement: This work was supported by the Russian State Program 'Russian Fund for Fundamental Research', grant No 99-04-49970.

1.   Bourtzis, K., Androniki, A., Markakis, G. and Savakis, C.  1996.  Genetics 144: 1063-1073.

2.   Dyachuk, P.A.  1975.  In: Khvostova, V.V. (ed.) Genetics and Selection of Pea,. Novosibirsk. Nauka, pp 197-223 (in Russian).

3.   Engels, W.R.  1979.  Genet. Res. 33: 219-236.

4.   Lutkov, A.N.  1930.  Proceedings of All-Union Congress on Genetics, Selection and Seed Breding, Leningrad, 1929-30, Vol. 2, p.353-367 (in Russian).

5.   Pearce, S.R., Harrison, G., Li, D., Heslop-Harrison, J.S., Kumar, A. and Flavell, A.J.  1996.  Mol. Gen. Genet. 250: 305-315.

6.   Schnable, P.S. and Wise, R.P.  1998.  Trends Plant Sci. 3: 175-180.

7.   Sobolev, N.A. and Agarkova, S.N.  1975.  In: Khvostova, V.V. (ed.). Genetics and Selection of Pea, Novosibirsk. Nauka, pp 141-160 (in Russian).