GENETIC ASSOCIATION OF HERBICIDE TOLERANCE AND BLACKLEG RESISTANCE IN BRASSICA NAPUS L

Mohan R Thiagarajah, Gary R Stringam, Vipan K Bansal and

Delbert F Degenhardt

Dept. of Agricultural, Food & Nutritional Science, University of Alberta

Edmonton, Alberta, Canada T6G 2P5

ABSTRACT

The doubled haploid method is routinely used at the University of Alberta for developing new canola cultivars. As part of this program, closely related crosses were initiated with the objective of combining blackleg resistance and glufosinate tolerance into superior breeding lines. When 165 blackleg resistant doubled haploid progeny from these crosses were grown in the field in 1997 and sprayed with glufosinate, only 3 survived from an expected 50%. Reciprocal crosses were made between two of the herbicide resistant lines above and a susceptible breeding line. F₂ segregation data for herbicide tolerance and blackleg resistance, and herbicide tolerance data for F₁-derived DH lines of these crosses, indicate that although the original DH segregation data suggested linkage of the two traits, the apparent association may actually relate to other factors, rather than to genetic linkage. These issues are discussed.

KEYWORDS

Doubled haploids, glufosinate tolerance, transgene, differential selection, blackleg resistance

INTRODUCTION

Herbicide resistance in canola is becoming an important trait of new cultivars. At the University of Alberta, we have been developing breeding lines with resistance to glyphosate (Roundup Ready), glufosinate (Liberty Link), and imidazolinones (Pursuit Smart) using the doubled haploid (DH) method of breeding. As part of our routine screening program for blackleg (Leptosphaeria maculans) resistance, we conduct a greenhouse cotyledon test for this disease before any newly developed lines are tested in the field, thus eliminating susceptible types early in the program. In 1997, we planted 165 DH lines derived from closely related crosses from our Liberty Link breeding program, which had been classified as resistant to blackleg, but had not been tested for Liberty resistance. These lines were sprayed with Liberty at the appropriate stage of development in the field, and contrary to expectation, only 3 of the lines survived the herbicide application.

The present study was undertaken to determine whether the genes conferring resistance to blackleg and glufosinate are closely linked in the lines that survived the herbicide application in the field. Such an association could be expected if these lines were rare cross-over types combining blackleg resistance and glufosinate resistance, closely linked in repulsion in the original crosses.

MATERIALS AND METHODS

Two of the glufosinate and blackleg resistant DH lines above (P1 and P2), both containing the construct 'T177' for herbicide tolerance, were crossed reciprocally with an University of Alberta breeding line (P3), susceptible to both the herbicide and the pathogen. F₁ plants were grown in a growth cabinet maintained at 20/15°C day/night temperature and 16 h photoperiod. Photosynthetic photon flux density was 425-450 mE m^-2 s^-1. The growth medium used was soil-free (Stringam 1971), and the plants were fertilized weekly with a 20-20-20 nutrient solution. Microspore culture and subsequent operations to produce DH lines from the F₁ plants were performed as described by Coventry et al. 1988. The F₁ plants used to produce the DH lines were also self-pollinated to obtain F₂ seed for the study.

Herbicide applications were completed on greenhouse-grown F_2, and DH seedlings at approximately the 4 leaf stage. Spray was applied to the leaves until runoff, using a specially constructed automated greenhouse boom sprayer, delivering a 0.03% solution of 200g a.i./l  of commercially available Liberty. 'Tween 20' (0.01%) was used as the wetting agent. The cultivar 'Innovator' was the glufosinate tolerant control and the cultivar Quantum was the herbicide susceptible control for the herbicide tolerance screening. Seedlings from the DH lines P1 and P2 used as parents in the crosses, were included in the same screening. Plants that survived, two to three weeks after the application were classified as tolerant. These plants were discarded after rating them. A cotyledon test for blackleg resistance was conducted on a separate set of DH line seedlings, as described by Bansal et al. 1994.

RESULTS AND DISCUSSION

Glufosinate tolerance data from all four F₂ populations indicated a normal monogenic Mendelian segregation (Table 1), with the transgene conferring tolerance to the herbicide being expressed as a dominant gene.

Table 1. Glufosinate reaction in F₂ populations from crosses of P1, P2 (tolerant to glufosinate, and resistant to blackleg) and P3 (susceptible to both glufosinate and blackleg).

Cross	Number of plants		Ratio tested	c^2	P
	Tolerant	Susceptible
P1 x P3	113	51	3:1	3.25	0.2 - 0.05
P3 x P1	131	59	3:1	3.71	0.2 - 0.05
P2 x P3	156	55	3:1	0.13	0.8 - 0.50
P3 x P2	127	53	3:1	1.89	0.2 - 0.05

Although tolerance to the herbicide was expressed as expected in the F₂ populations, only a few F₁-derived DH lines survived the greenhouse herbicide application (Table 2). Only 2 DH lines were herbicide tolerant from 55 lines screened from the reciprocal crosses between P1 and P2. This contrasts with the 27 or 28 tolerant lines expected for a monogenic trait. Twenty-seven DH lines out of 96 DH lines tested from the reciprocal crosses between P2 and P3 showed tolerance to glufosinate. Most of the DH lines, which were classified as 'tolerant', exhibited some foliar symptoms of damage, but were alive at the time of scoring. The cultivar Innovator, and the glufosinate tolerant DH lines, which were used as the parents P1 and P2, were least affected by herbicide damage. The blackleg screening data suggest that the gene for glufosinate tolerance and the gene for blackleg resistance are likely not closely linked in P1 and P2, as originally suspected. Although the only two glufosinate resistant DH lines derived from the reciprocal crosses between P1 and P3 happen to be blackleg resistant as well, 50% of the glufosinate tolerant DH lines derived from the reciprocal crosses between P2 and P3 were found to be blackleg susceptible. There were many blackleg resistant DH lines among those that were susceptible to the herbicide.

Table 2. Glufosinate tolerance and blackleg resistance reactions in F₁-derived DH lines from crosses between P, P2 and P3.

Cross	Glufosinate tolerant DH lines		Glufosinate susceptible DH lines		Total number of DH lines tested
	Blackleg resistant	Blackleg susceptible	Blackleg resistant	Blackleg susceptible
P1 x P3	2	0	20	19	41
P3 x P1	0	0	9	5	14
P2 x P3	7	8	11	29	55
P3 x P2	6	6	4	25	41

In contrast to the low frequency of glufosinate tolerant DH lines found in our study, Baranger et al (1997) have reported that segregation for glufosinate tolerance/susceptibility followed the expected 1:1 Mendelian ratio in DH lines derived from an F₁ hybrid 'Westar T5' x 'Miyuki', hemizygous for the bar gene. The transgene in their study however is a different construct from the one in ours, and we are not certain at this time whether the glufosinate tolerance trait conferred by the 'T177' construct in our study is simply not being expressed in some of the DH lines. The presence or absence of the 'T177' construct in these glufosinate sensitive DH lines needs to be determined. Occasional loss of expression of phosphinothricin (PPT) tolerance in sexual offspring of transgenic oilseed has been reported by Metz et al. (1997), who found deviating segregation ratios for PPT tolerant:sensitive individuals in selfings and backcrosses involving transgenic PPT tolerant plants. They found that there was loss of expression of PPT tolerance despite the presence of the transgene.

The absence of differential selection in microspore culture is considered to be important for the successful use of doubled haploidy in plant breeding programs, as doubled haploids should be a random sample of the parental gametes. Most studies on Brassica species have concluded that DH populations are comparable to conventionally-derived populations (Lichter et al., 1988; Chen and Beversdorf, 1990; Thiagarajah and Stringam, 1993). However, Siebel and Pauls (1989) reported that the flowering trait in a B.napus DH population, measured in terms of 'days to flowering', tended to be skewed toward a larger proportion of early lines. It is possible that the 'T177' construct in our study caused differential selection against gametes carrying ‘T177’ during microspore culture, thus resulting in distorted segregation patterns.

As transgene expression can differ from the greenhouse to the field due to environmental effects (Meyer, 1995), all DH lines in the present study will be tested for glufosinate resistance in the field in the summer of 1999. Genetic segregation studies will also be undertaken among crosses involving parents with differing transgene constructs conferring resistance to glufosinate. If these constructs are to be effectively utilized in plant breeding programs, they must exhibit stable expression, and be reliably and easily transferred to new breeding lines.

ACKNOWLEDGEMENTS

Financial support for this study from Alberta Agricultural Research Institute and AgrEvo Canada is gratefully acknowledged.

REFERENCES

Bansal, V. K., P. D. Kharbanda, G. R. Stringam, M. R. Thiagarajah, and J. P. Tewari, 1994: Comparison of greenhouse and field screening methods for blackleg resistance in doubled haploid lines of Brassica napus. Plant Dis. 78: 276-281.

Baranger, A., R. Delourme, N. Foisset, F. Eber, P. Barret, P. Dupuis, M. Renard and A.M. Chevre, 1997. Wide mapping of a T-DNA insertion site in oilseed rape using bulk segregant analysis and comparative mapping. Plant Breed. 116: 553-560.

Chen, J. L. and W. D. Beversdorf, 1990. A comparison of traditional and haploid-derived populations of oilseed rape (Brassica napus L.) for fatty acid composition of the seed oil. Euphytica 51: 59-65.

Coventry, J., L. Kott, and W. D. Beversdorff, 1988. Manual for microspore culture technique for Brassica napus. OAC Publication 0489, University of Guelph, Guelph, Canada.

Lichter, R., E. De Groot, D. Fiebig, R. Schweiger and A. Gland, 1988. Glucosinolates determined by HPLC in the seeds of microspore-derived homozygous lines of rapesedd (Brassica napus L.). Plant Breed. 100: 209-221.

Metz, P. L. J., E. Jacobsen and W. J. Stiekema, 1997. Occasional loss of expression of phosphinothricin tolerance in sexual offspring of transgenic oilseed rape (Brassica napus L.). Euphytica 98: 189-196.

Meyer, P., 1995. Variation in transgene expression in plants. Euphytica 85: 359-366.

Siebel, J. and K. P. Pauls, 1989. A comparison of anther and microspore culture as a breeding tool in Brassica napus. Theor. Appl. Genet. 78: 473-479.

Stringam, G R., 1971. Genetics of four hypocotyl mutants in Brassica campestris L. Heredity 62: 248-250.

Thiagarajah, M. R. and G. R. Stringam,1993. A comparison of genetic segregation in traditional and microspore-derived populations of Brassica juncea L. Czern and Coss. Plant Breed. 111: 330-334.