Pisum Genetics Volume 27 1995 Research Reports pages 26-28

Sym2 and nod-3 are independent but closely linked genes influencing nodule development in pea

Temnykh, S. V.*, Department of Horticultural Sciences

Weeden N. F. Cornell University, Geneva, NY 14456, USA


LaRue, T. A. Boyce Thompson Institute for Plant Research

Ithaca, NY 14853-1801, USA

* Present address: Dept. of Plant Breeding and Biometry, Cornell Univ., Ithaca, NY 14853-1801, USA

Gene sym2 confers strain-specific nodulation in a few primitive pea accessions from the Middle East. It was initially described as a recessive gene responsible for partial resistance of cv. 'Afghanistan' to European strains of Rhizobium leguminosarum bv. vicia (2 8). Resistance to nodulation, conferred by sym2, is not complete and depends on genetic and environmental factors. Gene sym2 can act as recessive or semidominant depending on the strain of rhizobium used, and at higher temperatures nodulation occurs with higher efficiency (9). It was recently shown that Sym-1, which originally was described as a dominant gene responsible for temperature-dependent nodulation of cv. 'Iran' (7, 9), is an allele of sym2 (6).

Recessive mutation nod-3, induced in cv. 'Rondo', is characterized by an excessive number of nodules even in the presence of nitrate (3) and by shortened lateral roots (10). The mutant phenotype is not strain-specific (4); however, nod-3 appears to map very near sym2. Gene sym2 was localized at 17 cM from marker d (anthocyanin ring at the base of stipules) (13) and at 25 cM from locus Idh (isocitrate dehydrogenase) (12) which is about 10 cM from d. Our data indicate that nod-3 and Idh are separated by about 20 cM (11). In addition, fine mapping experiments demonstrated that these two genes map close to the same RAPD marker (our unpublished data). In order to determine unambiguously whether nod-3 is a phenotypically different allele of sym2 or a different locus we attempted to identify the double mutant phenotype in progeny of crosses between the two mutants.

Parental material included lines Nod3 (homozygous for allele nod-3 in a 'Rondo' genetic background) and BC-sym2 (a near-isogenic line produced as described by Kneen et al. (5) by six repeated backcrosses of line 'Afghanistan' to cv. 'Sparkle' and a final self-pollination), and a single F2 plant, heterozygous for sym2 and derived from a cross between BC-sym2 and multiple-marker line JI73. In our experiments, when plants were grown in vermiculite at 18-20C and inoculated with R. leguminosarum strain 128C53, line BC-sym2 formed no nodules (non-nodulating), the Nod3 line displayed 300 70 nodules (hypernodulating), and lines with normal nodulation ('Rondo', 'Sparkle') displayed 50 to 100 nodules per plant.

In the cross between lines BC-sym2 and Nod3, the F1 hybrids formed from 2 to 40 nodules per plant (low nodule number), indicating that in this cross gene sym2 behaved as incompletely dominant. Similar results have been observed in a cross between line 'Afghanistan' and cv. 'Sparkle' tested under similar conditions (5). The F2 progeny of the cross BC-sym2 x Nod3 segregated as follows:

8 hypernodulating : 18 low nodulating : 18 non-nodulating : 6 normal.

The eight hypernodulating plants also all displayed shortened roots, similar to the Nod3 line. We postulate that these represented nod-3/nod-3 genotypes. Thirty-six out of fifty F2 plants formed a few or no nodules, presumably due to the semidominant effect of gene sym2. The F2 plants with normal nodulation were suspected to be recombinant between genes sym2 and nod-3. In the F3 progeny of three of these plants with normal phenotype, normal and hypernodulating plants segregated in a 3:1 ratio. These parental F2 plants thus possessed the genotype Sym2, nod-3/Sym2, Nod-3. Progeny of the other three plants consisted of non-nodulating, low-nodulating and normally nodulating plants in proportion 1:2:1, consistent with a genotype of sym2, Nod-3 / Sym2, Nod-3.

From these data we conclude that sym2 and nod-3 are different loci. The apparent recombination frequency between these genes was about 25%, much higher than predicted based on earlier results. However, the uncertainty in scoring sym2 as semidominant and the small size of the population compelled us to investigate another segregating population, which produced recombinants between sym2 and nod-3 with a double mutant phenotype.

This second cross (between mutant line Nod3 and the F2 plant heterozygous for sym2 and derived from a cross between BC-sym2 and multiple-marker line JI73) produced nine F1 hybrids, which all displayed normal nodule number. Five of these plants produced F2 progeny that segregated only for nod-3 and therefore had genotype Sym2, nod-3/Sym2, Nod-3. The remaining four plants produced F2 progeny that included non-nodulating phenotypes (Table 1), suggesting that those four F1 hybrids were heterozygous for both sym2 and nod-3 and that the two mutants complemented each other in these hybrids.

Table 1. Phenotypic segregation in an F2 population derived from Sym2, nod-3/sym2, Nod-3 hybrids.


Number of F2 plants

Lateral roots normal; non nodulating


Lateral roots normal; low nodulating


Lateral roots normal; normal nodulation


Lateral roots short; hypernodulation


Lateral roots short; few nodules


In this cross the proportion of plants with normal root morphology and normal nodulation was high and did not seem to represent a recombinant class. Ten plants of this category were subjected to progeny analysis and, in contrast to the results from the first cross, all were heterozygous at both sym2 and nod-3, resembling the initial F1 hybrids. Thus, in this second cross, the semidominant effect of sym2 was less pronounced and the class of heterozygotes for sym2 consisted of plants with nodule number ranging from relatively few (5-10) to that expected for a wild type.

Three plants with a novel phenotype were found among the F2 population. These plants were characterized by a low number of nodules and shortened roots. The progeny of these plants consisted of 1/4 hypemodulating, 1/2 low nodulating and 1/4 nonnodulating plants, all having the same altered root morphology. When two of these plants with the novel phenotype were crossed with the parental line Nod3, all eight F1 hybrids had an excessive number of nodules and shortened roots. These data indicate that the recombinants were homozygous for mutation nod-3 and heterozygous for gene sym2 (sym2, nod-3/Sym2, nod-3), further confirming that syml and nod-3 are different loci.

The expected probabilities of progeny phenotypes in the second cross, according to the epistatic model (1), were 0.25 non-nodulating, 0.5 low-nodulating, (1-r)2/4 hypernodulating, and (2r r2)/4 hypernodulating plants with short roots, where r is the recombination frequency between the sym2 and nod-3 loci. Solution of the likelihood equation

HYPER x 2/(r-1) + RECOMB x 2 (1-r)/(2r-r2) = 0

gave an estimated r of 0.05 0.05, when HYPER and RECOMB were replaced by the observed counts of these classes, 31 and 3, respectively.

The data obtained in this study clearly demonstrate that sym2 and nod-3 are different loci. Although the estimates of genetic distance between the two loci varied in two different crosses we analysed, our data in combination with previous mapping results (6, 11, 12, 13) suggest that these two genes influencing the same developmental process are closely linked. The occurrence of recombinants with few nodules but shortened roots shows that hypernodulation and short roots in the nod-3 mutant are regulated separately, and that in the nodulation process, sym2 is epistatic to nod-3. The determinants of semidominant versus recessive expression of sym2 have yet to be elucidated but appear to partially involve the genotype at other loci affecting nodule development.

Acknowledgements. We sincerely thank James Clare Nelson for assistance with calculating the recombination frequencies and Ian Murfet for his many helpful editorial suggestions.

1. Allard, R.W. 1956. Hilgardia 24:237-278.

2.           Holl, F.B. 1975. Euphytica 24:767-770.

3.                 Jacobsen, E. 1984. Pisum Newsletter 16:23-24.

4.           Jacobsen, E. and Nijdam, H. 1983. Pisum Newsletter 15:31-32.

5.                 Kneen, B.E., LaRue, T.A. and Weeden, N.F. 1984. Pisum Newsletter 16:31-34.

6.           Kozic, A., Heidstra, R., Horvath, B., Kulikova, O., Tikhonovich, I., Ellis, T.H.N.,
van Kammen, A., Lie, T.A. and Bisseling, T. 1995. Plant Sci. 108:41-49.

7.           Lie, T.A. 1971. Plant and Soil 34:751-752.

8.                 Lie, T.A. 1978. Ann. Appl. Biol. 88:462-465.

9.           Lie, T.A. 1984. Plant and Soil 82:415-425.


10.                Postma, J.D., Jacobsen, E. and Feenstra, W.J. 1988. J. Plant Physiol. 132:424-430.

11.                Temnykh, S.V., Kneen, B.E., Weeden, N.F. and LaRue, T.A. 1995. J. Hered.

12.                Weeden, N.F., Kneen, B.E. and LaRue, T.A. 1990. In Nitrogen Fixation:
Achievements and Objectives. Chapman and Hall, London, pp.323-330.

13.                Young, J.P.W. 1985. J. Hered. 76:207-208.