Venugopal Kalavacharla1, Chunren Wu2 and S.  Roger Rimmer3

1Department of Plant Science, North Dakota State University, Fargo, ND 58105 USA.

2Limagrain Canada Seeds Inc., P.O.Box 250, Listowel, Ontario N4W 3H2 Canada

3Agriculture and Agri-Food Canada, 107 Science Place, Saskatoon, SK, S7N 0X2




A number of biological races of Albugo candida can be differentiated by certain accessions of crucifer species.  We use Brassica  juncea cv Burgonde and Brassica  rapa cv Torch to differentiate race 2 and race 7.  Races of A. candida are generally more virulent on their homologous host species but are often capable of infecting some genotypes of closely related species.  Race 2 of A. candida normally infects B. juncea and also infects some populations of B. rapa including accession CrGC1-18 but not Torch. Conversely, race 7 of A. candida normally infects B. rapa but is also capable of infecting some populations of B. juncea including accession UM3512 but not Burgonde.  The objective of this study was to determine the genetics of resistance of differential cvs Torch and Burgonde to race 2 and 7 of A. candida, respectively.  Analysis of crosses between Torch and CrGC1-18, using sib-mating, suggests that resistance to race 2 in Torch is determined by two duplicate dominant genes.  Analysis of F1, F2 and F3 generations of crosses between the resistant cv Burgonde and the susceptible line UM3512 suggests that resistance in Burgonde is controlled by two nuclear genes with dominant‑recessive interaction.




Albugo candida is a highly specialized oomycetous plant pathogen and at least 10 physiological races have been identified and classified based on specificity to different species of crucifers (Pound and Williams 1963; Hill et al. 1988).   Race 2 and race 7 of A.  candida can be differentiated based on their interaction phenotype on B.  rapa cv Torch and B.  juncea cv Burgonde.  However some genotypes of these species (B.  rapa accession CrGC1-18 and B.  juncea accession UM3512) are susceptible to both races (Table 1). 


Table 1.   Interaction phenotypes of race 2 and race 7 of Albugo candida on differentials and common suscepts of Brassica rapa and Brassica juncea.




Brassica rapa                          Brassica juncea



Pathogen                      Torch         CrGc1-18              Burgonde               UM 3512


race 2                                    -a                 +                                +                         +                     

race 7                                    +                 +                                -                          +         


        a+ = Compatible;  - = Incompatible


These common susceptible genotypes were used for crossing race 2 and race 7 to study the genetics of virulence of A. candida (Rimmer et al. 1999).  The genetics of virulence was studied in F2 populations of A. candida by testing F2 clones on the two differential cultivars Torch and Burgonde (Mathur et al. 1995, Rimmer et al. 1996).  To examine further the possibility of a gene-for-gene relationship in this host-pathogen system, and, since no information on the inheritance of resistance to race 2 and race 7 in both heterologous differentials is available, this study was undertaken to determine the genetic control of resistance to race 2 in B. rapa and race 7 in B. juncea.




Plant materials

B. rapa     Plants of Torch and CrGC1-18 were selected by screening against race 2 of A. candida and resistant plants of Torch and susceptible plants of CrGC1-18 were sib-mated for two generations to increase homozygosity.  Eight crosses (including two reciprocal crosses) between the selected parental lines were made to generate F1 seed.  At the time of crossing, pollen of both parents was collected on bee-sticks and stored in the freezer at -10EC for backcrossing (Williams, 1980).  Two resistant F1 plants were sib-mated to produce F2 populations, while one F1 plant was back-crossed to both parents.

B. juncea       Plants of Burgonde and UM3512 were selected by screening against race 7 of A. candida and resistant plants of Burgonde and susceptible plants of UM3512 were selfed for two generations. Seven crosses (including reciprocal crosses) were made between the selected lines of Burgonde and UM3512 to generate F1 seed.  Pollen of both parents was collected and stored as described above.  F1 plants were selfed to produce F2 populations while F1 plants were simultaneously back-crossed to the parents using pollen as stored above.  F3 families were derived from randomly selected highly susceptible F2 plants.

Pathogen Isolate

The same isolates of race 2 (MIAc2_B1) and race 7 (MSAc7A) of A. candida were used in this study as in a previous genetic study of A. candida (Rimmer et al. 1996).  The isolates were increased by inoculation on the appropriate susceptible cultivar i.e. Burgonde (race 2) or Torch (race 7).  Mature zoosporangia were collected in gelatin capsules (Parke-Davis Size 00) and stored in glass screw-cap vials at -10EC.  For inoculation, inoculum was prepared according to the methods of Liu et al. (1996).   The number of zoospores were quantified and adjusted to 2 x 105 zoospores per millitre.

Inoculation Techniques

Seeds were planted in 12 cell multipots containing soilless mix (Metro mix, W. R. Grace & Co., Canada Ltd., Ajax, ON) and kept in a growth room at day/night temperature of 22/17EC, with a 16 h photoperiod.  Cotyledons of 6-day-old seedlings were inoculated by applying a 10Fl drop of zoospore suspension with an Eppendorf repeater pipette, and inoculated seedlings were incubated in a mist chamber for 48 h before returning them to the growth room.

Phenotype Evaluation

Interaction phenotypes were scored on a 0-9 scale 8-10 days after inoculation (Williams, 1985).  Cotyledons which showed no symptoms or small necrotic flecks on the adaxial surface with no sporulation were scored IP 0 or 1 and considered resistant, whereas those showing scattered or coalescing pustules on the abaxial or adaxial surface were scored IP 7 or 9 and considered susceptible. The chi-square test (X2) was used to evaluate the data from segregating generations.




Resistance in Torch to race 2

F1 Progeny   Eight families of F1 plants from six crosses (including reciprocals) of Torch ´ CrGC1-18 were screened against race 2 of A. candida.  Almost all families segregated for resistance and susceptibility to race 2.  In those families which did not segregate resistance was dominant.

F2 Progeny   Four of eight F2 families segregated for 15 resistant : 1 susceptible indicative of two duplicate dominant genes for resistance to race 2 in Torch.  One family segregated in a ratio of 7 resistant : 1 susceptible suggesting that the resistant parent was heterozygous at one of the two resistance gene loci.  Three families did not fit one gene or two gene models (Table 2).  Backcross data (not presented) from families which did not fit a two duplicate dominant gene model nonetheless segregated 3 resistant : 1 susceptible when backcrossed to the susceptible parent.

Table 2.  Observed segregation for host reaction, expected segregation ratios, chi-square values (X2), and probabilities (P) for goodness of fit for F2 plants from crosses between accessions of Brassica rapa following cotyledon inoculation with race 2 of Albugo candida


                                   Host Reaction


Cross               Ra               S                Total                Ratio          X2              P         


Cr1b                 196           13              209                  15:1           0.003         0.95-0.98

Cr2                  132             9              141                  15:1           0.004         0.95-0.98

RCr2                193            44              237                    3.1           5.22           0.02-0.05

Cr3                  318           16              334                  15:1          1.20           0.20-0.30

RCr3                220            48              268                   3:1           7.17           0.001-0.01

Cr4                  257            47              304                    7:1           2.43           0.10-0.20

Cr5                  259              5              264                  15:1           8.53           0.001-0.01

Cr6                  144           12              156                  15:1           0.54           0.30-0.50


aNumber of resistance (R) and susceptible (S) plants.    bCr=cross; Rcr=reciprocal cross


Resistance in Burgonde to race 7

F1 Progeny   Plants from 7 F1 crosses and reciprocal crosses of UM3512 ´ Burgonde were screened against race 7 of A. candida.  All plants tested were found to be resistant suggesting that resistance is dominant and under nuclear control.

F2 Progeny   F2 data from seven (1 through 7) crosses between susceptible parent UM3512 and resistant parent Burgonde and two reciprocal crosses (1R and 3R) were analyzed for all possible segregation ratios for a one-gene and all possible two-gene models (Table 3).  F2 progeny of crosses 3, 4, 6 and 7 fit a 3 resistant :1 susceptible segregation ratio.  F2 progeny of Cross 1, reciprocal Cross 1R and 3R fit 13:3.  Two F2 progeny (Cross 2 and 5) fit both a 13:3 and 3:1 resistant/susceptible segregation ratio.  Thus, F2 results suggest that resistance in the B. juncea cv. Burgonde is controlled by two genes with dominant recessive epistatic interaction.  This two gene model was confirmed by segregation for resistance in F3 families of susceptible F2 plants from those crosses which segregated in 13:3 ratio.

F3 Progeny   The hypothesis of two genes with dominant recessive epistatic interaction was tested further by evaluating F3 families derived from susceptible F2 plants from the populations of cross 1 (U3 x B1), which fit a 13 resistant : 3 susceptible ratio, and also from susceptible F2 plants of cross 3 (U8 x B3) which fit a 3 resistant : 1 susceptible ratio.  As predicated by the hypothesis, F3 families derived from the susceptible F2 plants of cross 1 fit either a 1:3 or 0:1 segregation ratio of resistant/susceptible plants.  However, three families of 22 tested fit a 3 resistant : 1 susceptible ratio which is not explainable by the hypothesized two-gene model.  All F3 families derived from susceptible F2 plants of cross 3 (U8 x B3) fit only a 0 resistant :1 susceptible segregation ratio.




Results from this study indicate that resistance to race 2 of A. candida in B. rapa cv Torch is controlled by two duplicate dominant genes and resistance to race 7 of A. candida in B. juncea cv Burgonde is controlled by two genes with a dominant-recessive epistatic interaction.  Presumably, both genes in Torch recognize a single avirulence gene in A. candida identified in crosses of race 2 with race 7 of A.  candida (Rimmer, Liu Mathur and Wu, unpublished data).  The situation is more complex for B. juncea.  Avirulence to Burgonde in F2 of A. candida crosses was a single dominant gene in some families but segregated as a single recessive gene in other families.  How these genes interact with the dominant and recessive genes for resistance observed here is not known.  Testing of F3 families with segregating families of A. candida should clarify these interactions.


Table 3.  Observed segregation for host reaction, expected segregation ratios, chi-square values (X2), and probabilities (P) for goodness of fit for F2 plants from crosses between accessions of Brassica juncea following cotyledons inoculation with race 7 of Albugo candida


                                    Host Reaction                                                   X2



Cross               Ra         S          Total                3:1 (R:S)    P                     13:3(R:S)      P


1 (U3xB1)        450        96      546                  16.03         <0.0001            0.52            0.30-0.50

1R (B1xU3)     346        70      416                  14.83         <0.0001            1.01            0.30-0.50

2 (U4xB2)        135        37      172                    1.12         0.20-0.30          0.86            0.30-0.50

3 (U8xB3)        413      152      565                    1.09         0.20-0.30        24.65            <0.0001  

3R (B3xU8)     261        55      316                    9.72         <0.01                0.37            0.50-0.70

4 (U6xB4)        131        49      180                    0.48         0.30-0.50          8.48            <0.01      

5 (U10xB5)      229        58      287                    3.33         0.05-0.10          0.41            0.50-0.70

6 (U9xB6)        210      78       288                    0.63         0.30-0.50        13.13            <0.0001

7 (U7xB7)        222      73       295                    0.01         0.90-0.95          6.97            <0.01      


aNumber of resistance (R) and susceptible (S) plants.




Hill, C.B., Crute, I.R., Sheriff, C., and Williams, P.H.  1988.  Specificity of Albugo candida and Peronospora parasitica pathotypes towards rapid-cycling crucifers.  Cruciferae Newsletter. 13:112-113.

Liu Q.J., Parks, P., and Rimmer, S.R.  1996.  Development of monogenic lines for resistance to Albugo candida from a Canadian Brassica napus cultivar.  Phytopathology 86:1000-1004.

Mathur, S., Wu, C.R., Liu, J.Q., and Rimmer, S.R.  1995.  Inheritance of virulence and fungicide resistance in Albugo candida (Abstr.).  Can. J. Plant Pathol. 17:360.

Pound, G.S., and  Williams, P.H.  1963.  Biological races of Albugo candida.  Phytopathology 53:1146-1149.

Rimmer, S.R., Mathur, S., Wu, C.R., and Liu, J.Q.  1996.  Genetics of virulence to Brassica species in crosses of race 2 and race 7 of Albugo candida (Abstr.).  Downy Mildew Newsletter 9:6.

Williams, P.H.  1985.  Crucifer Genetics Cooperative (CrGC) Resource Book. Department of Plant Pathology, University of Wisconsin, Madison, WI.

Williams, P.H.  1980.  Bee-sticks, an aid in pollinating cruciferae.  HortScience 15:802-803.