Branching in pea: double mutants of rms7 with rms1 through rms5

PISUM Genetics

2003—Volume 35

Research Papers

Branching in pea: double mutants of rms7 with rms1 through rms5

Murfet, I.C.

Plant Sci, Univ. of Tasmania, Hobart Tasmania, Australia

Mutants rms6 and rms7 in pea have increased branching from basal nodes (6, 10). In contrast, mutants rmslthrough rms5 have increased branching from both basal and aerial (upper stem) nodes (1-3). A start has beenmade on checking double-mutant phenotypes. In some cases, the double mutant expresses an additivephenotype with branching more strongly enhanced than in either single mutant, e.g. rms2-1 rms4-1, rms2-1rms5-2, rms3-1 rms6-2 and rms6-1 rms7-1 (6, 8-10). In other cases, epistasis occurs and the double-mutantphenotype does not transgress beyond the range of the single mutants, e.g. rms1-1 rms4-1 and rms2-1 rms3-1 (8).

In the present study, the phenotype of the double mutants of rms7 with rmsl through rms5 was examinedin tall (Le) plants grown in the glasshouse under an 18-h photoperiod (for details see 8). This strategy wasdesigned to allow identification of double-mutant plants regardless of whether they were clearly obvious froman additive double-mutant phenotype or hidden by epistasis of one or other mutant allele. The strategy makesuse of the following information gleaned from years of observation of branching in pea. 1) Basal branching isexpressed more strongly in dwarf (le) than tall (Le) plants (4, 7, 8). 2) In contrast to dwarf plants, tall plantswith WT (wild-type) branching genes invariably fail to produce secondary stems from a basal node under the18-h conditions used. 3) Tall rmsl through rms5 plants always produce aerial laterals under these 18-hconditions. Thus in an F₂ population, any tall plant with a major secondary stem and no aerial laterals couldbe considered as homozygous for rms7. In cases where double-mutant plants were not exposed in F₂ by anadditive phenotype, F₃ progenies could be grown from the homozygous rms7 plants and any F₃ plantsexpressing strong growth of aerial

branches would be exposed asdouble mutants.

In accordance with thisstrategy, dwarf line M3T-475(rms7 1) was crossed with tall linesWtl5240 (rms1-5, ex Kaliski),K524 (rms2A, ex Torsdag), K487(rms3-1 ex Torsdag), K164 (rms4-1, ex Torsdag) and HL298 (rms5-3). HL298 was specifically bred forthis purpose from a cross betweentall cv. Torsdag and Wtl5241(rms5-3, ex dwarf cv. Paloma).Further details on these mutantlines are given by Arumingtyas etal. (2).

The rmsl -5 rms7-1 doublemutant was found to have anadditive phenotype (Fig. 1). Tall F₂

plants of cross Wtl5240 (rms1-5) xM3T-475 (rms7-1) could bepartitioned into four branchingclasses corresponding to WT, rms7,rms1, and rms1 rms7 double-

Fig. 1. Distribution of the branching index 'ratio of lateral to main-stemlength' for cv. Terese, M3T-475 (rms7-1), cv. Kaliski, Wtl5240 (rms1-5), and tall F₂ plants from the cross Wtl5240 x M3T-475. The F₂ data aresubdivided into four branching phenotypes representing WT, rms7, rms1,and double-mutant rmsl rms7 plants. Data are from mature plants;photoperiod 18 h.

mutant plants. There was a quantum increase in the ratio of lateral to main-stem length from WT to rms7 torms1 to rms1 rms7. The observed F₂ numbers of 27 WT, eight rms7, eight rms1 and two rms1 rms7 plants arein good accordance with a di-hybrid 9: 3: 3: 1 ratio (P > 0.9). The additive phenotype of the double mutantwas confirmed by growing F₃ progenies from rms7 F₂ plants (data not shown). The tall rms1-5 rms7-1segregates in the F₃ population had a branching index 42 % higher than the WT15240 rms1-5 single-mutantcontrols grown with this generation.

Over half the tall WT F₂ plants of cross WT15240 x M3T-475 produced some lateral branches in contrastto the complete absence of branching in the tall WT control Kaliski (Fig.l). The WT F₂ plant with the ratherhigh branching index of 0.8 was confirmed to be WT by growing F₃. The branching in these WT plants wascomprised principally of short, late emerging, aerial laterals; no WT plant produced a secondary stem from abasal node. The late outgrowth of aerial laterals on many of the tall WT F₂ plants may partially be explainedby the fact that Terese and M3T-475 tended to flower at node 18 while Kaliski and Wt 15240 generallyflowered at node 16. A two-node delay in flower initiation allowed increased opportunity for aerial lateraloutgrowth. (NB. The late emerging aerial laterals referred to here are a normal part of the first reproductivecycle and not to be confused with second-growth laterals that emerge if plants fail to undergo monocarpicsenescence).

Growth of laterals from the cotyledonary node (node 0) was rare in the tall F₂ plants of crossWtl5240 x M3T-475 and occurred only in two of the eight rms7 plants and one of the two double-mutant plants.

The tall F₂ population of cross K524 (rms2-1) x M3T-475(rms7-1) gave a clear separationinto 21 WT, seven rms7, sevenrms2 and two probable rms2rms7 plants, numbers thatclosely fit a 9: 3: 3: 1 ratio (Fig.2). The two F₂ plants with thehigh branching indices of 3.8and 4.4 were backcrossed toK524 and M3T-475; the F₁results gave supporting evidencethese two plants had a double-mutant genotype (data notshown). Interestingly, thedouble-mutant F₃ plants had abranching index only 8% higherthan the K524 rms2-1 controls

grown with them (Fig. 2). Thusthere is evidence that the rms2-1 rms7-1 double mutant has anadditive phenotype. However,while the F₂ data indicated aclear quantum increase in

Fig. 2. Distribution of the ratio of lateral to main-stem length for M3T-475(rms7-l), K524 (rms2-l), and the F₁, F₂ and F₃ plants of cross K524 x M3T-475. F₂ data are subdivided into four branching phenotypes representing WT,rms7, rms2 and rms2 rms7 plants. F₂ and F₃ data are from tall plants only.The plants represented in the upper seven rows were sown July 31, 2000 andin the lower two rows July 30, 2001. Data are from mature plants;photoperiod 18 h.

branching, the F₃ data showedonly a small quantitativeincrease in branching over the rms2A single mutant.

The F₂ of cross K487 (rms3-1) x M3T-475 {rms7-1) gave no evidence of transgression (data not shown).An rms3-1 rms7-1 double-mutant line was obtained in F₃ and F₄ via a clear rms7 F₂ plant. The branchingindex of the double-mutant plants did not appear to be enhanced beyond the range of vigorous, single-

mutant, rms3-1 plants. However, the double mutant did combine features from both single mutants: the aeriallateral growth of rms3-1 plants and the tendency to produce laterals from the cotyledonary node shown byrms7-1 plants. Out of 73 tall rms3-1 rms7-1 plants, one third produced one or more laterals from node 0, andmany of these laterals continued growth into secondary stems. In contrast, basal laterals were not producedfrom node 0 of the K1487 rms3-1 mutant: when present, basal laterals grew from nodes 2 and/or 1 of K487plants.

No transgression for the ratio of lateral to main-stem length occurred in the F₂ of cross K164 (rms4-1) xM3T-475 (rms7-1) (data not shown). Homozygous rms4-1 rms7-1 plants were obtained in the F₃ and F₄ viaclear rms7 F₂ plants. The branching index of tall rms4-1 rms7-1 plants did not transgress beyond the upperrange of the K164 rms4-1 single mutant. Five per cent (3/59) of double-mutant plants produced a lateral fromthe cotyledonary node, a feature not seen

in the single mutant rms4-1.

F₂ data for cross HL298 (rms5-3) xM3T-475 (rms7-1) indicated an additivedouble-mutant phenotype (Fig. 3). Thetall F, population could be partitionedinto 33 WT, three rms7, ten rms5, andtwo probable rms5 rms7 double-mutantplants with a substantially higherbranching index than either single-mutant parent. The phenotypic contrastbetween the rms5 single mutant and thedouble mutant is not wholly revealed bythe index 'ratio of lateral to main-stemlength'. Expression of rms5-3 in a tallbackground seems fairly weak. Three outof six HL298 plants produced no basallateral of any consequence; likewise

three out of ten tall rms5 F₂ plants. Theserms5 plants were not really identifiableas ramosus mutants until outgrowth ofaerial laterals commenced three to fourweeks after sowing. In contrast, the

Fig. 3. Distribution of the ratio of lateral to main-stem length for M3T-475 (rms7-l) and HL298 (rms5-3), and tall F₂ plants from the crossHL298 x M3T-475. The F₂ data are subdivided into four branchingphenotypes representing WT, rms7, rms5, and double mutant rms5rms7 plants. Data are from mature plants; photoperiod 18 h.

presumed double-mutant plant with

index 3.5 (Fig. 3) had four basal laterals exceeding 10 mm by day 11 (two from node 0, one from node 1 andone from node 2) and all four shoots continued growth to become secondary stems. These observationsprovide further support for the view that rms5-3 rms7-1 plants have an additive phenotype that combines theaerial branching of rms5-3 plants with the enhanced basal branching of rms7-1 plants.

Segregation for branching phenotype in the F₂ of cross HL298 x M3T-475 (Fig. 3) is in accordance with a9: 3: 3: 1 ratio (P > 0.1). However, numbers in the rms7 class are below expectation and it seems highly likelythat some tall rms7 plants in this F₂ failed to produce a basal lateral and were indistinguishable from WTplants. In a tall background, results from several crosses showed that rms7-1 behaved as a weak mutant lackingthe full penetrance of classic Mendelian mutants. In the F₃ of cross Wtl5240 {rms1-5) x M3T-475 (rms7-1),only 38% (8/21) of tall rms7 plants had a phenotype distinguishable from WT. In the F₃ of crosses K164(rms4-I) x M3T-475 and K487 (rms3-1) x M3T-475, penetrance of rms7-1 fell to 15% (3/20) and 6% (2/36),respectively. With a penetrance that low, the rms7-1 mutation may well have escaped detection had it beeninduced in a tall cultivar. We were also fortunate to have obtained rms7 numbers right on Mendelianexpectation in two F₂ populations (Figs 1 and 2). Clearly, penetrance of rms7-1 varied from planting toplanting.

In summary, a clearly additive double-mutant phenotype was observed for rms1-5 rms7-1 (Fig. 1) andrms5-3 rms7-1 (Fig. 3), rms2-1 rms7-1 showed variable levels of transgression (Fig. 2), and rms3-1 rms7-1 andrms4-1 rms7-1 did not transgress, respectively, beyond the upper range of the rms3-1 and rms4-1 single-mutantparents. Interestingly, rms1 and rms5 both produced an additive phenotype with rms7 (Figs 1 and 3), whichfits well with evidence that RMS1 and RMS5 regulate the same novel branching signal (5). However, rms1-5and rms5-3 had a lower branching index than rms2-1, rms3-1 and rms4-1 when all five mutants were plantedand grown together (data not shown). Thus rms1-5 and rms5-3 have more room to show enhanced branchingand an additive phenotype in combination with rms7-1.

The basal branching mutants rms6-1 and rms7-1 produced an additive double-mutant phenotype,suggesting that RMS6 and RMS7 may operate in different pathways (6). That view is supported by the factthat rms3-1 rms6-2 was found to have an additive phenotype with strongly enhanced branching (10), whereasno evidence of transgression was found here for the rms3-1 rms7-1 double mutant.

Acknowledgements: This work was funded by a grant from the Australian Research Council. I thank Dr Suzanne Morrisfor help with Figs. 1-3, and Ian Cummings and Tracey Winterbottom for technical assistance.

1. Apisitwanich, S., Swiecicki, W.K. and Wolko, B. 1992. Pisum Genetics 24: 14-15.

2. Arumingtyas, E.L., Floyd, R.S., Gregory, M.J. and Murfet, I.C. 1992. Pisum Genetics 24: 17-31.

3. Blixt, S. 1976. Agri Hort. Genet. 34: 83-87.

4. Floyd, R.S. and Murfet, I.C. 1986. Pisum Newslett. 18: 12-15.

5. Morris, S.E., Turnbull, C.G.N., Murfet, I.C. and Beveridge, C.A. 2001. Plant Physiol. 126: 1205-1213.

6. Morris, S.E., Beveridge, C.A., Murfet, I.C, Prioul, S. and Rameau, C. 2003. Pisum Genetics 35: 10-14.

7. Murfet, I.C. and Reid, J.B. 1993. In: Casey, R and Davies, D.R. (eds.) Peas: Genetics, MolecularBiology and Biotechnology. CAB International, Wallingford, UK, pp 165-216.

8. Murfet, I.C. and Symons, G.M. 2000. Pisum Genetics 32: 33-38.

9. Murfet, I.C. and Symons, G.M. 2000. Pisum Genetics 32: 59-60.

10. Rameau, C, Murfet, I.C, Laucou, V., Floyd, R.S., Morris, S.E. and Beveridge, C.A. 2002. Physiol.Plant. 115:458-467.


	mutant, rms3-1 plants. However, the double mutant did combine features from both single mutants: the aeriallateral growth of rms3-1 plants and the tendency to produce laterals from the cotyledonary node shown byrms7-1 plants. Out of 73 tall rms3-1 rms7-1 plants, one third produced one or more laterals from node 0, andmany of these laterals continued growth into secondary stems. In contrast, basal laterals were not producedfrom node 0 of the K1487 rms3-1 mutant: when present, basal laterals grew from nodes 2 and/or 1 of K487plants. No transgression for the ratio of lateral to main-stem length occurred in the F₂ of cross K164 (rms4-1) xM3T-475 (rms7-1) (data not shown). Homozygous rms4-1 rms7-1 plants were obtained in the F₃ and F₄ viaclear rms7 F₂ plants. The branching index of tall rms4-1 rms7-1 plants did not transgress beyond the upperrange of the K164 rms4-1 single mutant. Five per cent (3/59) of double-mutant plants produced a lateral fromthe cotyledonary node, a feature not seen

	in the single mutant rms4-1. F₂ data for cross HL298 (rms5-3) xM3T-475 (rms7-1) indicated an additivedouble-mutant phenotype (Fig. 3). Thetall F, population could be partitionedinto 33 WT, three rms7, ten rms5, andtwo probable rms5 rms7 double-mutantplants with a substantially higherbranching index than either single-mutant parent. The phenotypic contrastbetween the rms5 single mutant and thedouble mutant is not wholly revealed bythe index 'ratio of lateral to main-stemlength'. Expression of rms5-3 in a tallbackground seems fairly weak. Three outof six HL298 plants produced no basallateral of any consequence; likewise

	three out of ten tall rms5 F₂ plants. Theserms5 plants were not really identifiableas ramosus mutants until outgrowth ofaerial laterals commenced three to fourweeks after sowing. In contrast, the	Fig. 3. Distribution of the ratio of lateral to main-stem length for M3T-475 (rms7-l) and HL298 (rms5-3), and tall F₂ plants from the crossHL298 x M3T-475. The F₂ data are subdivided into four branchingphenotypes representing WT, rms7, rms5, and double mutant rms5rms7 plants. Data are from mature plants; photoperiod 18 h.

	presumed double-mutant plant with index 3.5 (Fig. 3) had four basal laterals exceeding 10 mm by day 11 (two from node 0, one from node 1 andone from node 2) and all four shoots continued growth to become secondary stems. These observationsprovide further support for the view that rms5-3 rms7-1 plants have an additive phenotype that combines theaerial branching of rms5-3 plants with the enhanced basal branching of rms7-1 plants. Segregation for branching phenotype in the F₂ of cross HL298 x M3T-475 (Fig. 3) is in accordance with a9: 3: 3: 1 ratio (P > 0.1). However, numbers in the rms7 class are below expectation and it seems highly likelythat some tall rms7 plants in this F₂ failed to produce a basal lateral and were indistinguishable from WTplants. In a tall background, results from several crosses showed that rms7-1 behaved as a weak mutant lackingthe full penetrance of classic Mendelian mutants. In the F₃ of cross Wtl5240 {rms1-5) x M3T-475 (rms7-1),only 38% (8/21) of tall rms7 plants had a phenotype distinguishable from WT. In the F₃ of crosses K164(rms4-I) x M3T-475 and K487 (rms3-1) x M3T-475, penetrance of rms7-1 fell to 15% (3/20) and 6% (2/36),respectively. With a penetrance that low, the rms7-1 mutation may well have escaped detection had it beeninduced in a tall cultivar. We were also fortunate to have obtained rms7 numbers right on Mendelianexpectation in two F₂ populations (Figs 1 and 2). Clearly, penetrance of rms7-1 varied from planting toplanting.