A Dominant Epistatic Gene which Inhibits Cellular Susceptibility to RSV(RAV-O)

Transcription

1 J. gen. Virol. 097I), x3, Printed in Great Britain 455 A Dominant Epistatic Gene which Inhibits Cellular Susceptibility to RSV(RAV-O) By L. N. PAYNE AND P. K. PANI Houghton Poultry Research Station, Houghton, Huntingdon AND R. A. WEISS* Department of Anatomy and Embryology, University College London, Gower Street, London, W.C. I (Accepted 6 August I97I) SUMMARY The susceptibility to infectious RSV(RAV-O) of pure-bred and incross-bred chick embryos from the inbred Reaseheath c and I lines was studied, and a genetic hypothesis is proposed to explain the observed segregation of susceptible and resistant embryos. The results suggest (I) the presence of a dominant autosomal gene for susceptibility to RSV(RAV-O), designated e s, in the I line, and of its recessive allele, e ~, in the c line, and (2) the presence in the ~ line of an unlinked dominant, autosomal epistatic gene I e, which inhibits the expression of e% and of its recessive allele, i t, in the c line. The two loci are designated 'tumour virus e (tve)" and 'inhibitor ~ (/e),. INTRODUCTION After solitary infection, the BRYAN strain of Rous sarcoma virus (RSV) replicates in transformed chick embryo cells to produce small amounts of either an infectious form of RSV with an unusual host range or a non-infectious variant of this virus (Vogt, I967; Weiss, I967, I969a; Hanafusa & Hanafusa, I968 ). The term RSV(O) was used to designate these newly discovered forms of RSV. Infectious RSV(O), synonymous with RSVp(O) (Hanafusa & Hanafusa, i968; Weiss, I969 a), is produced in cells which possess the 'natural' group-specific (gs) antigen (Weiss, I969a; Vogt & Friis, I97I ; Weiss & Payne, i97 I) or the 'chick cell associated helper factor (chf)' (T. Hanafusa, Miyamoto & H. Hanafusa, 197o[b; H. Hanafusa, Miyamoto & T. Hanafusa, 197o; Hanafusa, Hanafusa & Miyamoto, I97oa), whereas non-infectious RSV(O), synonymous with RSV ~, (O) (T. Hanafusa et al. I97oa), is produced in cells which lack the natural gs-antigen or chf. The natural gs-antigen, which is identical to a group-specific antigen found in avian RNA tumour viruses, is present in many apparently normal chick embryos in theabsence of infectious virus (Dougherty & Di Stefano, 1966; Dougherty, Di Stefano & Roth, I967) and its inheritance is controlled by a single dominant autosomal gene gs + (Payne & Chubb, 1968; Weiss & Payne, I971). Recent findings are consistent with the suggestion that genetically integrated leukosis provirus may be responsible for the presence of gs-antigenand chf, and under certain circumstances may be activated to produce a helper virus, designated RAV-O or RAV-6o, necessary for the production of infectious * Present address: Department of Microbiology, University of Southern California School of Medicine, 2o25 Zonal Avenue, Los Angeles, California 9oo33, U.S.A. 3o-2

2 456 L.N. PAYNE, P. K. PANI AND R. A. WEISS RSV(O) (Payne & Chubb, I968; Weiss, I969a; T. Hanafusa et al. I97oa; Vogt & Friis, I97~ ; Weiss & Payne, ~97I). Infectious RSV(O) formed in the presence of the helper virus RAV-O has been renamed RSV(RAV-O) and the non-infectious variant generated in the absence of a helper virus has been renamed RSV(-) (Weiss, I97~). These new terms are used in this report. RSV(RAV-O) has a restricted host range in chickens which differs from that of viruses of subgroups A, B, C and D, and it has been provisionally allocated to a new subgroup, E, by T. Hanafusa et al. (197o a). The genetic basis for cellular susceptibility to RSV(RAV-O) has not been elucidated. In this report we present evidence derived from studies on two inbred lines of fowl and their crosses that genetic susceptibility to RSV(RAV-O) is dependent on the presence of a dominant autosomal gene which confers susceptibility and the concurrent absence of a dominant autosomal epistatic gene which masks the phenotypic expression of the susceptibility gene. METHODS Chick embryos. Chick embryos from the highly inbred Reaseheath c and I lines, the first backcross generations (cd' x IC9 and Ig x Icg), the F2 generation (~cg x Ic?), and the second backcross generation to the c line (c8 x (cg x It?), were examined for susceptibility to RSV(RAV-O). The first backcross generations to the c and I lines, and the F 2 generation, came from three pens of 6 IC (from ~ d ~ x c 9) dams mated to single c line, I line and m sires respectively. The second backcross to the c line came from 2o dams mated to 3 c line sires. These same dams were studied by Payne & Chubb (~968) in their investigation of the inheritance of the natural gs-antigen. Chick embryos from ten non-inbred commercial strains of fowl were kindly provided by F. and G. Sykes Ltd. Viruses. RSV(RAV-O) was prepared by isolating a single focus from gs-antigen positive Brown Leghorn cells infected with 5 focus-forming units (f.f.u.) of RSV(RAV-~) (Weiss, 1969a ). The transformed cells, designated SNP6, derived from this focus, were subcultured on Brown Leghorn feeder cells. Supernatant medium was harvested from these cultures, frozen and thawed once, and filtered through a o.22 #m. pore size cellulose nitrate filter. The filtrate was used as a virus stock. Apart from two preliminary experiments in which earlier batches of virus were used, all experiments were made using a single stock of RSV(RAV-O) which had a titre of I o a'4s f.f.u./ml, on quail cells; o. ~ ml. of a 10-1"3 dilution of this stock in phosphate buffered saline containing 2 ~ calf serum was inoculated on to the chorioallantoic membrane of I x-day-old embryos, and pocks were counted and the sex of each embryo determined 8 days later. Two stocks of RSV(-) were prepared in a similar way as RSV(RAV-O) by isolating two foci, designated CNP 8 and CNP I4, from gs-antigen negative cells from the Reaseheath c line infected with 5 f.f.u, of RSV(RAV-2). The transformed cells were separately subcultured on c line feeder cells, and virus was harvested in supernatant medium as described for RSV(RAV-O). The presence of non-infectious RSV(-) in the medium was detected by successful infection of Brown Leghorn cells after treatment of cells and virus inoculum with inactivated Sendal virus (Weiss, r969a ). These stocks were diluted to lo -1 before they were inoculated on to the chorioallantoic membrane.

3 Genetic susceptibility to RSV(RAV-O) 457 RESULTS The genetic hypothesis we propose to explain the results obtained predicates (I) the existence of a dominant autosomal gene for susceptibility to RSV(RAV-O) which we designate e ~ in accordance with accepted nomenclature (Crittenden et al. I967), and its recessive resistance allele, e ~, (2) the existence of a dominant autosomal gene P, which inhibits the expression of e s, and its recessive allele, i ~, (3) absence of linkage of the two loci. We suggest that the c line has the genotype i~i~ere " and the I line the genotype PPe~e ~. The genotypes of the F1, F2 and first backcross generations arise by recombination of genes in accordance with Mendelian principles, and are set out, with the corresponding phenotypes, in Table L Also shown are the comparisons between the expected segregation ratios of susceptible and resistant phenotypes and the observed ratios. The distribution of individual pock counts in the F2 and backcross generations formed two distinct populations (Fig. 0. Table I. Comparison of observed proportions of chick embryos susceptible to RSV(RA V-O) with those predicted by the 2-loci hypothesis Line or cross Genotype Phenotype* C ieieere r R I IefeeSe s R c x I (Ix) Iei'e~e~ R No. of Susceptible embryos (%) embryos c---~--~ Deviation tested Expected Observed X 2 I74 o o.6o o o -- Not tested o Not tested e- e rer 3 i~iees-- I ieieere r 25 I8"75 I8"80 O'OOO t Ieieere *" ieieese r ieieerer IXlC IeleeSeS l e I e es er i} IeieeSe s leieese r * S = susceptible, R = resistant. 22o z5'oo z3"6o o.zi8y 53 o o "~ X 2 (0"05, I D.F.) = 3"84- Embryos with pock counts of less than 5 were classified as resistant, and those with 5 or more were classified as susceptible. The numbers in each cross shown in Fig. ~ are smaller than the totals given in Tables I and 2 because results from two preliminary experiments with earlier batches of RSV(RAV-O) are excluded from the Figure. The proportions of susceptible F2 and backcross embryos obtained from individual dams are shown in Table 2. The validity of pooling results from individual dams to provide an overall segregation ratio for the cross was tested by means of heterogeneity X 2 tests in which goodness of fit of results from individual dams and total dams were compared with the expected ratio for each cross. Non-pedigreed embryos in each cross were included as an additional class. With the exception of a single susceptible c line embryo all embryos from the c line, I line and backcross to the I line were resistant to RSV(RAV-O), in agreement with expectation. The single susceptible c line embryo may indicate the presence in the c line of the e s gene in

4 45 8 L.N. PAYNE, P. K. PANI AND R. A. WEISS very low frequency, or may have been due to an error in marking or reading the pedigree of the egg. No F1 embryos were available for testing. The observed proportions of I8.8o susceptible embryos in the F2 and 23"6o ~ susceptible embryos in the c x Ic agree well with the respective expectations of 18"75 ~ and 25"o ~. The segregations were independent of sex of the embryos. The susceptible F2 embryos tended to have a higher pock count than the susceptible c xm embryos (Fig. I). A gene dosage effect may be responsible, iei~e~e s embryos, only present in the F2 generation, being more susceptible than iqeese ~ embryos ~ i i i I i I F 2 embryos 10 E 0 d z 132 c x IC embryos , STNC Pock count Iog~o (x+l) Fig. t. Distribution of pock counts on chorioallantoic membranes of F2 and c ic generation chick embryos inoculated with RSV(RAV-O). TNC = pocks too numerous to count. The titre of the RSV(RAV-O) stock based on the response of susceptible embryos was about ten-fold higher than the titre of io 3"48 f.f.u./ml, found on quail cells. The titration on quail cells was made without the addition of polycations, and it is to be expected that their addition would have increased the titre (Toyoshima & Vogt, I969). Additional support for our hypothesis is given by the segregation ratios observed in the second backcross generation to the c line (Table 3). The c x lc dams used in making the second backcross are of four possible genotypes of equal frequency: I~ieere ~" and iei~ere ~

5 Genetic susceptibility to RSV(RA V-O) 459 which give all resistant progeny, Ieiee~e r which give 25 ~ susceptible progeny, and ieieese " which give 5o ~ susceptible progeny. These three progeny response patterns were observed. If the I I dams which had I o or more progeny are considered, this being the minimum number of progeny required to differentiate between a 3: I ratio and all resistant embryos at the 5 ~ level of probability, it is seen that two dams had progeny about 5o ~ susceptible, two had progeny which were about 25 ~ susceptible, and the remaining seven dams had all resistant progeny. This is in good agreement with expectations of 2"75, 2"75 and 5"5 dams respectively. Table 2. Proportions of chick embryos from individual dams susceptible to RSV(RA V-O) in the F2 and first backcross generations No. of No. susceptible Heteroembryos, ~ - -, Deviation geneity Cross Sire no. Dam no. tested Expected* Observed X2t X2:~ IC X IC (F~) IC! " " t 21 3" o.ooi " ' ' o'ooi " " Non "75oo I7 o' pedigreed Total 25o 46" o.ooo 5'695 c IC c "oooo 6 o'ooo -- 39I I5 3" o.ooo I o'oi " ' oooo 4 0" Non- 9I 22" o' pedigreed Total "000o "955 I x IC o o I 2 0 O IO O O o o o o Non- I9 o o Pedigreed Total 53 o o * I8"75% for IC IC, 25"OO~o for cxlc, 0% for t X ~ (0"05, I D.r.) = 3"84. X 2 (0'05, 6 D.F.) = I2"59. I X IC. The genetic control described is specific for RSV(RAV-O). No susceptible embryos were found among 68 C XlC embryos inoculated with the non-infectious variant, RSV(-). Resistance to RSV(RAV-O) was widespread among the non-inbred commercial strains of chick embryos. In a survey for susceptible individuals in groups of about 12 embryos from IO commercial strains of chickens we found resistance in all embryos except 2/11 and 1/I2 from two strains.

6 460 L.N. PAYNE, P. K. PANI AND R. A. WEISS Table 3. Proportions of chick embryos from individual dams susceptible to RSV(RAV-O) in the second backcross generation to the C line Presence of No. of Susceptible embryos gs-antigen embryos Dam no. gene, gs + tested No. % * ii 2 i I o o o o o o o o I I6" t8 o o Io o o i o o 34I I' ~7 o " i o o I4 o o I5 o o I5 o o +, Heterozygous for gs-antigen gene, gs+gs-; -, homozygous recessive for gs-antigen gene, gs-gs- (see Payne & Chubb, I968; Weiss & Payne, I971). DISCUSSION Resistance to RSV(RAV-O) appears to be controlled by two pairs of genes, and may be due either to absence of the susceptibility gene e ~, as in the c line, or to the presence of the epistatic gene I e, to which e ~ is hypostatic, as in the I line. We suggest that the two loci should be termed 'tumour virus e (tve)' and 'Inhibitor (P)'. The two types of resistance could be differentiated by use of a susceptible tester line of genotype i~iee'~e s, of which line 7 embryos of phenotype C/A, which are susceptible to RSV(RAV-O), may be an example (Vogt, I967; Hanafusa & Hanafusa, I968; Vogt & Friis, 197[ ). Production of only susceptible embryos in the F1 would denote the genotype of the other line as Pi ere r, whereas only resistant F1 progeny would indicate genotypes of either IeleeSe ~ or IeFere ~. These could be discriminated between on the basis of finding resistance : susceptibility ratios in the F~ generation of 3 : I or 13 : 3 respectively. In addition to the dominant epistatic gene postulated in this paper, the I line also carries a dominant autosomal gene, gs +, absent from the c line, which is associated with the presence of the natural gs-antigen (Payne & Chubb, I968; Weiss & Payne, I97~). Embryos which possess the gs-antigen carry a latent helper virus needed for replication of RSV(RAV-O) (Weiss, I969a; T. Hanafusa et al. x97oa; Vogt & Friis, I97X ; Weiss &Payne, I970. We have considered the possibility that the gs + gene could be identical to the I ~ gene, because blocking of RSV(RAV-O) receptor sites with virus products controlled by the gs + gene could explain the mechanism of phenotypic expression of the P gene. Since gs + may be integrated viral genome, resistance of gs-antigen positive embryos to RSV(RAV-O) might be similar to the immunity of lysogenic bacteria to homologous phage. Some of the c Ic dams in Table 3 were known to be heterozygous for the gs-antigen gene, whereas the others were homozygous recessive (Payne & Chubb, 1968). Four out of seven dams producing ten or more all resistant

7 Genetic susceptibility to RSV(RAV-O) progeny, of genotype Ieiee~e " or iei~e~e ~, carried gs +. The two dams with more than ten progeny which were 25 ~ susceptible, of putative genotype Fi~e~e ~, both possessed gs +, whereas the two dams which gave 5o ~ susceptible progeny, ofgenotype iq"e~e ~, both lacked the gs + gene. These results are consistent with the suggestion that the gs + gene and the I" gene may be identical. Other evidence suggests that the association may be fortuitous. Line 7 (C/A) embryos, of postulated genotype iei~e~e ~, are susceptible to RSV(RAV-O) but gs-antigen positive; however, their susceptibility is ~ to 3 logs10 less than quail cells, and it is possible that they are I~I"e~e~ type with incomplete epistasis (Hanafusa & Hanafusa, I968; Vogt & Friis, 1971 ; Vogt, personal communication). Weiss (~ 969 a) reported two gs-antigen positive embryos out of nine susceptible embryos, but the possibility that the presence of gs-antigen was a consequence of infection with RSV(RAV-O) cannot be excluded. The results of H. Hanafusa et al. 097o) suggest that susceptibility to RSV(RAV-O) and presence or absence of chf are independent characteristics. In spite of these apparent contradictions we believe that the possibility that the gs + gene and I ~ gene are identical is worth further investigation. Several other observations require further study in the light of our findings. In certain strains or crosses of fowl there exists an association between susceptibility to subgroup B virus and RSV(RAV-O) (Vogt, I97o; L. B. Crittenden, personal communication) which might be due to linkage of the tvb and tve loci or to the existence of alleles ofb ~ which control susceptibility to both B and E subgroup viruses. The b ~ gene is apparently not identical to e ~ because the c line, of genotype i~i"e~e~ according to our hypothesis, is b*b s. There does appear to be a close relationship between subgroup B and subgroup E viruses because subgroup B (and related D) viruses interfere with RSV(RAV-O) (Vogt, L 967; Weiss, t 969 b; T. Hanafusa et al. 197o a). Another finding that may be relevant to the nature of resistance to RSV(RAV-O) ~ is the induction of susceptibility to RSV(RAV-O) in resistant cells infected with subgroup A and C leukosis viruses, RAV I and RAV 49 (Ishizaki & Shimizu, I97o; Weiss & Vogt, unpublished results). These viruses may act by interfering with the function of I e or by enhancing susceptibility in e~e ~ cells. In this connexion it is of interest that Vogt (quoted by Ishizaki & Shimizu, I97O) found that loss of resistance to RSV(RAV-O) was shown only by some resistant c/o chick cells preinfected with subgroup A leukosis virus, which supports the suggestion of the existence of more than one type of resistance to RSV(RAV-O). The tve locus described in this report appears to be comparable to the tva, tvb and tvc loci which control responses to avian tumour viruses of subgroups A, B and C, resistance being a recessive trait (Crittenden et al. ~967; Payne & Biggs, 197o). The I ~ gene we believe inhibits expression of e s genes at the tve locus, and may be an example of a class of dominant genes which explain instances where resistance to turnout viruses is inherited as a dominant trait (Waters & Burmester, I963; Dhaliwal, ~963). We wish to thank Mr K. Howes for technical assistance. 46I REFERENCES CRITTENDEN, L. B., STONE, H. A., REAMER, R. H. & OKAZAKI, W. 0967). Two loci controlling genetic cellular resistance to avian leukosis-sarcoma viruses. Journal of Virology z, 898. DHALIWAL, S. S. (1963). Resistance of the chorio-allantoic membrane of chick embryos to Rous sarcoma and MH 2 retieuloendothelioma viruses. Journal of the National Cancer Institute 3o, 323. DOUGHERTY, R. M. & OI STEFANO, H. S. 0966). Lack of relationship between infection with avian leukosis virus and the presence of COFAL antigen in chick embryos. Virology 29, 586. DOUGHERTY, R. M., DI STEEANO, H. S. & ROTH, F. K. (1967). Virus particles and viral antigens in chicken tissues free of infectious avian leukosis virus. Proceedings" of the National Academy of Sciences of the United States of America 58, 808.

8 462 L.N. PAYNE, P. K. PANI AND R. A. WEISS rianavusa, H. & nanafusa, W. (I968). Further studies on RSV production from transformed cells. Virology 34, 63o. IIANAFUSA, H., MIYAMOTO, T. & HANArUSA, T. (I970). A cell-associated factor essential for formation of an infectious form of Rous sarcoma virus. Proceedings of the National Academy of Sciences of the United States of America 66, 314. HANAEUSA, T., HANAFUSA, H. & MIYAMOTO, T. (1970 a). Recovery of a new virus from apparently normal chick cells byinfection with avian tumor viruses. Proceedings of the Nationat Academy of Sciences of the United States of America 67, I797. HANArtJSA, W., ~IYAMOTO, T. & HANAEUSA, I-I. (t 970 b). A type of chick embryo cell that fails to support formation of infectious RSV. Virology 4 o, 55. ISHlZAKI, R. & SHIMIZU, T. (I970). Observations on the envelope properties of RSV(O). Virology 4 o, 415. PAYNE, L. N. & BIGGS, P. M. (1970). Genetic resistance of fowl to MH2 reticuloendothelioma virus. Journal of General Virology 7, I77. VAYNE, L. N. & CHUBB, R. C. (1968). Studies on the nature and genetic control of an antigen in normal chick embryos which reacts in the COFAL test. Journal of General Virology 3, 379. TOYOSHIMA, K. & VOGT, P. K. (I 969). Enhancement and inhibition of avian sarcoma viruses by polycations and polyanions. Virology 38, 414. root, 1 ~. K. (I967). A virus released by 'non-producing' Rous sarcoma cells. Proceedings' of the National Academy of Sciences o.f the United States of America 58, 8Ol. VOGT, P. K. (I970). Envelope classification of avian RNA tumor viruses. Comparative Leukemia Research, I969. Bibliotheca Haematologica, no. 36 (ed. R. M. Dutcher), pp. I53-I67. Basel, Miinchen, Paris, New York: Karger. VOLT, V. K. & FRnS, R. R. (1971). An avian leukosis virus related to RSV(O) : properties and evidence of helper activity. Virology 43, 223. WATERS, N. F. & BURMESTER, B. R. (I963). Mode of inheritance of resistance to induced erythroblastosis in chickens. Poultry Science 42, 95. WElSS, R. A. (I967). Spontaneous virus production from 'non-virus producing' Rous sarcoma cells. Virology 32, 719. WEISS, R. A. (1969a). The host range of Bryan strain Rous sarcoma virus synthesized in the absence of helper virus. Journal of General Virology $, 511. WEXSS, R. A. (I969b). Interference and neutralization studies with Bryan strain Rous sarcoma virus synthesized in the absence of helper virus. Journal of General Virology 5, 529. WEISS, a. A. (1971). Helper viruses and helper cells. Proceedings of a Symposium on RNA Viruses and Host Genome in Oncogenesis. Amsterdam: North Holland Publishing Co. (In the Press.) WEISS, R. A. & PAYNE, L. N. (1971). The heritable nature of the factor in chicken cells which acts as a helper virus for Rous sarcoma virus. Virology" (in the Press). (Received 15 June 197I)