hit. J . Cancer: 17, 396-406 (1976)

CHARACTERISTICS OF THREE STRAINS OF FELINE FIBROSARCOMA VIRUS GROWN IN CAT AND MARMOSET MONKEY CELLS

Ruth MCDONALD, Baku1 THAKKAR, Lauren G . WOLFE and Friedrich DEINHARDT Department of Microbiology, Rush-Presbyterian-St. Luke's Medical Center, 17.53 West Congress Parkway, Chicago, Ill. 60612; and Graduate College, University of Illinois at the Medical Center, Chicago, Ill. 60612, USA

Summary Two strains ojfeline fibrosarcoma virus (ST-FeSV and GA-FeSV) were found to induce tumors in cats and marmosets, and to transform feline and marmoset cells in vitro after primary inoculation. A third strain (SM-FeSV) failed to induce tumors or transform marmoset cells after primary inoculation; however, when SM-FeSV-infected marmoset cultures were passed 26 times in vitro, the cell cultures released infectious virus which transformed marmoset fibro- blasts but still failed to induce tumors in marmosets. ST-FeS V induced mainly round-cell type transfor- mation ( r foci) , GA-FeSV induced predominantly mixed round-fusiform cell type transformation (fr foci), and SM-FeSV induced r and f r type foci with a higher proportion of fusiform cells in the f r foci than seen with GA-FeSV. Transforming virus was obtained from r or mixed rjfr foci of ST-FeSV but not from f r foci; heat treatment changed the virus from producing almost exclusively r type foci to inducing an increased number of f r foci. Passage of FeSV in cat cells yielded viruses with a higher ratio of infectivity for feline vs marmoset cells, while passage of FeSV in marmoset cells yielded virus with a relatively higher infectivity ratio for marmoset cells; the three strains differed in the degree of change in the infectivity ratio. Despite the alteration of host range of SM-FeSV propagated in marmoset fibroblasts, the virus retained feline P-30 antigen by CF and FA assays, Neutraliza- tion tests did not indicate but also did not exclude an alteration of the surface antigens of ST-FeSV or SM-FeS V propagated in marmoset fibroblasts. The alteration of the relative infectivity of FeSV during passage in marmoset cells may be due to: ( 1 ) the selection of a variant present in the original hetero- geneous uncloned population; ( 2 ) mutation; or (3 ) recombination with some marmoset genetic material, possibly an as yet unident$ed endogenous marmoset virus.

Several strains of C-type RNA viruses have been isolated from naturally-occurring feline fibrosar- comas and examined for their oncogenicity in vivo and transforming property in vitro (Deinhardt ef al., 1972; Hampar et al., 1970; Lee, 1971; McAllister et al., 1971 ; McDonough et a[., 1971 ; Pearson et al., 1973; Rabin, 1971; Sarma et al., 1970, 1971a, b, 1972; Snyder, 1971; Snyder and Theilen, 1969; Theilen, 1971; Theilen et al., 1970; Ubertini et al.,

1971; Wolfe et al., 1971, 1972). In our laboratory, the Snyder-Theilen (ST-FeSV) (Snyder and Theilen, 1969) and Gardner-Arnstein (GA-FeSV) (Gardner et al., 1970, 1971) strains of feline fibrosarcoma virus consistently induced sarcomas in marmoset monkeys (Saguinus fuscicollis, S. nigricollis, S. oediyus) (Deinhardt et al., 1970, 1972; Rabin, 1971 ; Theilen et al., 1970; Wolfe et al., 1971,1972) and transformed feline and marmoset cells in vitro (McDonald et al., 1971,1972; Sarma et al., 1971a, 6,1972). In contrast, the McDonough (SM-FeSV) strain failed to induce tumors in marmosets (Wolfe, McDonald and Deinhardt, unpublished data) and on primary infection transformed feline but not marmoset cells in vitro. However, marmoset cells infected with SM-FeSV became transformed after several passages in vitro, and released virus that transformed both feline and marmoset cells in vitro, although the virus still failed to induce tumors in marmosets.

In this communication, we describe: (1) the tumor-cell lines established from cat and marmoset tumors induced by ST-FeSV and GA-FeSV; (2) the transformation of feline and marmoset fibroblasts in vitro by SM-FeSV, ST-FeSV and GA-FeSV; and (3) some of the characteristics of the virus released by the infected feline and marmoset fibroblasts.

MATERIAL AND METHODS

Virus

Stocks of ST-FeSV and GA-FeSV were prepared from medium plus cell extracts of 5-day-old cultures of cell lines established from cat and marmoset tumors (McDonald et al., 1972). SM-FeSV was kindly supplied by Dr. Sarma, National Cancer Institute, Bethesda, Md., USA, as clarified fluids from infected feline embryo cells (5 x lo4 FFU/ml). Additional stocks of SM-FeSV were prepared from medium plus cell extracts of feline and marmoset cultures infected with the SM-FeSV originally obtained from Dr. Sarma.

Cell cultures

Tumor-cell lines were established from cat and marmoset tumors that had been induced by cell-free

Received: October 20, 1975, and in revised form December 8, 1975.

FeSV IN CELL CULTURES 397

FeSV of cat origin or viable neoplastic tissue of marmoset origin (Wolfe et a!., 1972). The tumor-cell cultures were grown in medium RPMI 1640 sup- plemented with 20% fetal calf serum (FCS) and 2 mM glutamine.

Feline embryonic fibroblast cultures (FEF) and marmoset fibroblast cultures (MF) were established and grown as described previously (McDonald et al., 1972). Cultures were tested by fluorescent antibody (FA) tests for the species-specific P30 (gs-I) antigens of feline C-type RNA viruses (McAllister et al., 1972). Antiserum to the P30 antigen was provided by Dr. Gilden, Flow Laboratories, Rockville, Md., USA. Marmoset cell cultures were consistently negative for RD114 species-specific P30 antigen but all feline cell cultures, even if negative initially, became positive for this antigen during serial cell culture passages (usually between the 12th and 20th passages). Attempts were made to use only cells which were negative by FA for R D l l 4 antigens but this was not always possible. However, there was no indication in control experiments that the presence of RDI 14 antigen influenced in any way the growth of the sarcoma viruses.

Tumor-cell lines and FeSV-infected feline and marmoset fibroblast cultures were trypsinized and subcultured every 4 to 7 days by standard tissue culture techniques as described previously (McDonald et al., 1972).

Antisera

Antisera were obtained from cats and marmosets bearing tumors induced by ST-FeSV and GA-FeSV. The antisera were used in focus reduction and in indirect immunofluorescent antibody studies. The antiserum used in the fluorescent antibody studies came from a marmoset bearing a tumor induced by GA-FeSV and it contained antibodies to both the species- and interspecies-specific P30 (gs-1 and gs-3) antigens by double diffusion tests. Antisera obtained from marmosets bearing tumors induced by Simian sarcoma virus, type 1, (SSV-I) (Deinhardt et al., 1972) were also used in indirect fluorescent antibody studies.

Antisera were produced in guinea-pigs against tween-ether-disrupted feline leukemia virus (FeLV) obtained from a persistently infected F-422 cell line (Rickard et al., 1969) and against SSV-1, obtained from a cell line established from an SSV-I induced marmoset tumor. The antisera were adsorbed with FCS (75 mg/ml serum) until they no longer reacted with FCS by double diffusion. Antisera against F-422 and SSV-1 virus contained antibodies to both species-specific and interspecies-specific P30 (gs-1 and gs-3) antigens in double diffusion tests. The antibody titers of the individual sera were determined in

complement fixation tests (CF) with tween-ether- disrupted F-422 or SSV-1 virus.

Focus assay

ST-FeSV and GA-FeSV produced by the tumor- cell lines and ST-FeSV and SM-FeSV released from infected feline and marmoset cultures were assayed quantitatively for the amount of transforming virus on feline (FEF-R) and marmoset fibroblasts by their ability to induce foci using the agar overlay technique (McDonald et al., 1972). Cultures which had formed an approximately 60 % confluent monolayer were infected with FeSV, incubated for 2 h at 37"C, re-fed with liquid medium, incubated at 37°C for 12-72 h, and then overlaid with 1 % Difco purified bacto-agar. For all incubations, the flasks were sealed and incubated containing atmospheric air. The assays were done on cells not pre-treated with diethylaminoethyl (DEAE)-dextran except where indicated. The titers were expressed as focus-forming units per ml (FFU/ml), and all FFU/ml reported were based on the examination of 12 or more flasks. Foci were counted 14 to 21 days after infection in either viable or fixed cultures. For fixation, 2 ml of fixative were added to agar-overlaid cultures after two rinses with Hanks' balanced salt solution (BBS-H). The fixative consisted of 9 parts 10% formalin, 6 parts absolute ethanol and 2 parts glacial acetic acid. Transformed cells fixed in this manner retained their morphology and high refractility and could be counted at a later date.

Effect of DEAE-dextran

The effect of DEAE-dextran treatment of FEF-R and MF-69CBlM cultures on the number and morphology of FeSV-induced foci (large round cells (r) or mixture of small round and fusiform cells (fr)) was examined. FEF-R and MF-69CBlM cultures were seeded and 3-4 h later, 0.1 ml of DEAE-dextran (0.2 mg/ml BSS-H) was added and allowed to adsorb for 1 h. The cultures were then washed three times with BSS-H and infected with either ST-FeSV from marmoset tumor-cell line No. 2 or SM-FeSV from infected FEF-R or MF-69CB1 M cultures.

Stability of the viruses inducing r and f r foci

The viruses produced by the two morphological types of foci (r and fr foci) induced in MF-69CB1 M cultures by ST-FeSV from marmoset tumor-cell line No. 2 were examined to see whether they consistently induced foci of the same morphological types. The r and fr foci were collected with Pasteur pipettes; cells from each focus were then grown in four Leighton tubes, cell-free supernatants of these cultures were inoculated into MF-69CB1 M cultures and the cultures were examined for r, fr and mixed r/fr foci.

398 MCDONALD ET AL.

This procedure was performed serially four times for each original focus.

Virus-specific antigens

Test samples were prepared from approximately 6 ~ 1 0 ~ cells from 5-day-old cultures of cat and marmoset tumor-cell lines (passages 4 to 14) and SM-FeSV-infected feline and marmoset cells. The cells were scraped into a small amount of veronal buffered diluent (VBD), PH 7.4 and centrifuged at 8 0 0 x g for 20 min. Cell pellets were resuspended in 4 x volume of VBD, frozen and thawed three times, and again centrifuged as described above. Two-fold dilutions of supernatants were reacted with 2 units of guinea-pig complement and 2 units of antibody, prepared in guinea-pigs against tween-ether- disrupted, tissue-culture-grown FeLV or SSV-1 . The degree of the CF reaction was scored on a scale of 0 to 4 (4 = complete fixation). A sample was considered positive for viral antigen if a 3 - 1 or greater fixation of complement was attained in the absence of anticomplementary reactions and absence of reactivity with normal sera (McDonald et al., 1971). These studies were performed before radio- immunoassays for the antigens of FeLV and SSV-I had been developed.

Indirect fluorescent antibody studies were per- formed on the tumor-cell lines and infected cultures at the same passage levels at which they were examined by CF. Coverslip cultures were rinsed with BSS-H, fixed with chilled acetone for 7 min, air dried and stored at -120°C for 1 to 3 months before staining. Fixed cultures were stained for 90 min at 37" C and then 15-18 h at 4" C with antisera to GA-FeSV or SSV-I obtained from tumor-bearing marmosets; antisera were diluted 1 :4 with phosphate- buffered saline (PBS) containing 2.5 % bovine serum albumin (BSA), PH 7.2. The cultures were washed twice with PBS, reacted for 30min at 37" C with anti-human IgG conjugated with fluorescein isothio- cyanate (FITC) (diluted 1 :20 in PBS containing 1 % BSA), rinsed four times with PBS and mounted with elvanol, PH 8.2; strong cross-reactivity of anti-human IgG with marmoset IgG had previously been shown in our laboratory.

Neutralization tests

Neutralization tests were performed to compare the reactions of sera obtained from cats and marmo- sets, bearing tumors induced by ST-FeSV or GA- FeSV, against: ( 1 ) ST-FeSV isolated from cat or marmoset tumor-cell lines; and (2) SM-FeSV propagated in feline or marmoset fibroblasts. Approximately 3 0 0 FFU/ml of virus in RPMI 1640 medium were kept at 25" C for 1 h in a 1 :1 ratio with serial dilutions of heat-inactivated cat or marmoset sera (heat-inactivated 56" C, 3 0 min and

diluted in BSS-H). FEF-R cultures were infected with 0.4 mI of the virus-serum mixture. The neutraliz- ing antibody titer was recorded as the highest serum dilution which inhibited development of 50% of the foci. All values listed in Table VII are the averages of two or three experiments; 2-fold serum dilutions in each experiment were performed in duplicate.

Irnmunoelectro-osmophoresis (IEOP)

Counterimmunoelectrophoresis for measuring serum antibody titers against viral proteins was performed as described previously (Hoekstra and Deinhardt, 1971).

Electron microscopy

Cell monolayers were scraped from the culture flasks, suspended in maintenance medium and centrifuged for 10 min at 800 xg. Cell pellets were fixed for 3 0 min in 2 % glutaraldehyde buffered with 0.1 M sodium cacodylate, postfixed in 1 % osmium tetroxide, dehydrated in graded alcohols and embedded in Epon 812. Thin sections were stained with lead citrate and uranyl acetate; unless virus was readily found, at least 100 cells from each preparation were examined for virus.

RESULTS

Three cat and four marmoset ST-FeSV tumor-cell lines were easily distinguishable from four cat and five marmoset GA-FeSV tumor-cell lines. The ST-FeSV tumor-cell lines contained large round cells, small round cells and fusiform cells, whereas the GA-FeSV tumor-cell lines contained small round cells and fusiform cells but only a few large round cells (McDonald et al., 1972). This morphological distinction was stable in all cell lines during the observation period of 10 to 87 passages.

Feline sarcoma-leukemia P30 group-specific (gs) antigen was demonstrated by C F and/or FA in all 18 tumor-cell lines examined (Table I). Antigen was present even when no virions were found by electron microscopy and when no transforming virus was recovered. As shown in Table I, ST-FeSV and GA-FeSV tumor-cell lines derived from cat tumors had higher antigen titers than the marmoset tumor- cell lines. Furthermore, no correlation existed between the antigen titer and the amount of infectious virus produced by a tumor-cell line. The antigens detected with the particular sera used were predomi- nantly P30 but the presence of additional viral glycoprotein or protein antigens was not studied further.

Extracellular C-type virus and virus budding from plasma membranes were found in 17 of 19 tumor-cell lines. Morphological descriptions of the cells and

FeSV IN CELL CULTURES 399

TABLE I

ANTlGEN CONTENT A N D VIRUS PRODUCTION OF FeSV-INDUCED TUMOR-CELL LINES

Virus production

FFU/ml Infectivity ratio : Tumor-cell Gs antieen ' strain species EM FEF-R/MF-69CBIM Virus line

number FEF-R MF-69CBIM (cat) (marmoset) CF FA

ST-FeSV Cat 1 2 3

Marmoset 1

3 4

GA-FeSV Cat 1 2 3 4

Marmoset 1 2 3 4 5 6 7

2 5

> 1 :32 >1:32 >1:32

ND 1 :8 I :8 I :8

1 :32 > 1 :32 >1:32 > 1 :32

1 :8 1 :I6 1 :16 1 :16 1 :I6 1 :16 1 :8

+ + + + + + + + + - t + + + + f f + t + + + - t

+ - f - t + - t ND t- + + + t + -

2,200 20,750 3 1,000

3,450 5,150

41,500 43,750

0 163 875

1,300

0 0

15 83 95

113 700

28 3 70 375 160 883

1,800 2,100

0 0

50 483

23 0

750 60

500 10

230

80 56 83

22 6

23 21 -

>163 18 3

<0.04 0 0.02 1.4 0.19

1 1 3

~

Tumor-cell lines were established from cat and marmoset tumors induced by ST-FeSV and GA-FeSV, and cultures were examined between passages 4 to 14. - a CF = antigen titer in complement fixation test; FA = positive by fluorescent antibody test. For details see " Material and Methods ". - 'Virus titers (FFU/ml) are the average of three experiments with a minimum of 6 flasks per dilution. EM = virus particles identified by electron microscopy. - ' ND = not done. - ST-FeSV isolated from marmoset tumor-cell line No. 2 is referred to as CO-I-ST- FeSV in a previous paper (McDonald et al.. 1972).

virus particles have been reported elsewhere (Wolfe et al., 1972).

All ST-FeSV tumor-cell lines released more trans- forming virus than the GA-FeSV tumor-cell lines during passages 4 to 14 (Table I). Regardless of the species of origin, cat or marmoset tumor-cell lines induced by ST-FeSV yielded approximately equiva- lent titers of transforming virus when assayed on cat cell cultures; however the viruses recovered from either cat or marmoset tumor cells differed in their ability to induce transformation of marmoset cells, i.e. ST-FeSV isolated from cat tumor-cell lines was less infectious for marmoset cells than the viruses produced by marmoset tumor-cell lines. The same was true for GA-FeSV induced tumors.

Like the virus titration curves for FeSV from the marmoset ST-FeSV tumor-cell line No. 2 reported previously (McDonald et al., 1972), the virus titration patterns of FeSV from cat ST-FeSV tumor- cell line No. 1, cat GA-FeSV tumor-cell line No. 4 and marmoset GA-FeSV tumor-cell line No.7 showed one-hit kinetics. This suggested that the FeSV preparations from the four tumor cell lines were either competent or that the preparations contained an excess of " helper " virus, ',which is consistent with previous reports (Sarma et af.,

1971a, b, 1972; McDonald et al., 1972) that ST- FeSV and GA-FeSV produced by infected and transformed feline or marmoset (ST-FeSV only) fibroblasts showed one-hit kinetics and contained helper virus.

FeSV-induced transformation in vitro

Virus isolated from a1 four types of tumor-cell lines (ST-FeSV and GA-FeSV from cat and marmoset tumors) induced foci on both feline embryonic fibroblasts (FEF-R) and marmoset fibroblasts (MF- 69CBlM) (Table I). Virus isolated from cat and marmoset ST-FeSV tumor-cell lines induced foci consisting of large round cells (r foci) or a loose meshwork of fusiform and small round cells (fr foci) whereas virus isolated from cat and marmoset GA- FeSV tumor cell lines induced predominantly fr foci. Additional cultures (1 human, 4 marmoset and 8 feline cultures) transformed by virus from marmoset ST-FeSV tumor-cell line No. 2 formed either r and fr foci or r foci alone (Table 11). There was no relationship between the species of origin of the cultures and the morphology of the foci induced.

The foci induced on FEF-R and MF-69CB1 M by virus from the ST-FeSV tumor cell line No. 2 appeared to arise by multiplication of transformed

400 MCDONLAD ET AL.

TABLE I1

MORPHOLOGY OF FOCI INDUCED IN VARIOUS CELL CULTURES BY ST-FeSV DERIVED FROM MARMOSET TUMOR-CELL LINE No. 2

~~ ~~ ~ ~~~~~

Origin of cell line Designation Morphology of culture M ~ ~ ~ ~ ~ ~ g Y

Species Tissue *

Human Embryonic lung Wi-38 Fibroblast-like r and fr

Marmoset Lung MF-1283 r r

r and fr

epithelial4 ke r and fr

(S. fuscicollis) Skin MF-69CB2S Muscle MF-69CB1 M Kidney Secondary cultures Mixed fibroblast and

Muscle FEF-NBM Fibroblast-like r Embryonic kidney FEF-CRFK-FZ-FL-4 r Tongue FEF-Tg 7050 (code No. FFc 3 Tg) r Embryo FEF-F.L. 9716 (code No. FFc 71 WF) r

r r and fr r and fr r and fr

FEF-F.L. 9717 (code No. FFc 3 Tg) FEF-R FEF-NCE FEF-F.L. 9202 (code No. FFc 9 WF)

Cat

' For more details on the origin of the tissues refer to McDonald el a/. (1972)

cells rather than by the spread of virus through the medium. When marmoset ST-FeSV antisera (0.25 % final concentration) were added to the medium and agar overlays, 1-2h after infection, there was no reduction in the number of foci indicating that spread of virus through the medium was not necessary for focus formation.

The effect of DEAE-dextran treatment of FEF-R and MF-69CB1 M cultures on transformation by virus from the marmoset ST-FeSV tumor-cell line No. 2 and the cat ST-FeSV tumor-cell line No. 1 was examined. DEAE-dextran treatment did not alter the ratio of r to f r foci induced by ST-FeSV although it increased the titer 10- to SO-fold.

consistently induced foci of the same morphological type (Table 111). Virus from each r focus induced only r foci, with one exception which yielded a mixture of 10 r, 2 fr and 4 mixed r/fr foci on the 4th serial inoculation on MF-69CB1 M. No transforming virus was isolated from 7 fr foci whereas virus isolated from mixed r/fr foci induced r foci, fr foci and mixed r/fr foci. Although the virus inducing the r foci consistently induced r foci, it lost this stability with heat treatment (McDonald et al., 1972). Virus from r foci incubated for 30min a t 37" C prior to inoculation on MF-69CBl M showed a shift toward inducing fr foci rather than r foci with a fr:r ratio of 2.42: 1.

Virus produced by the two morphological types of foci, r and fr foci, induced in MF-69CBlM cultures by ST-FeSV were examined to see whether they

Cells from r and fr foci were examined to see whether they induced r and fr clones respectively. Cells of foci induced in MF-69CBlM cultures by

TABLE 111

STABILITY OF r OR fr TYPE FOCUS INDUCTION BY FeSV

Viral inocula Drepared

Number of r. fr and mixed r/fr foci induced during four serial subcultures of MF-69BIM ' from foci Number Passage 4 Passage 4

not-heat-treated heat-treated inducedon of Passage 1 Passage 2 Passage 3 MF-69CBIM inocula by ST-FeSV

Type of focus r fr r/fr r fr r/fr r fr r/fr r fr r/fr r fr rjfr Mixed Mixed Mixed Mixed Mixed

r 9 215(8) 0 0 46 (5) 0 0 25(3) 0 0 40 2 4 15 14 19 fr 7 0 0 0 Mixed r/fr 6 5 2 3 (1) 1 0 l ( 1 ) 2 i 2(2) 3 1 2 1 8 0

' The figures in brackets indicate the number of foci (picked from infected MF-69CBlM). from which virus was isolated and inoculated on MF-69CBIM in the next passage. - ' At the third passage, virus inocula were prepared from 3 r foci and 2 mixed r/fr foci. Half of each inoculum was then heat-treated (37" C. 30 min) and inoculated on MF-69CBIM. - a 10 r, 2 fr and 4 mixed r/fr foci in passage 4 not heat-treated were induced by virus prepared from one r focus of passage 3.

FeSV IN CELL CULTURES 401

ST-FeSV from marmoset tumor-cell line No. 2 were cloned in soft agar. Nine r foci produced 67 r clones; 5 mixed r/fr foci produced 11 mixed r/fr clones, 19 r clones and 7 fr clones; and 3 fr foci produced 9 f r clones.

SM-FeSV received from Dr. Sarma induced transformation of feline embryonic fibroblasts (FEF-R and FEF-NCE) after primary inoculation but marmoset fibroblasts became transformed only after serial propagation of the infected cells. The transformed FEF cultures contained both r and fr foci, with the formation of more fusiform cells than observed in cultures transformed by either ST-FeSV or GA-FeSV. The transformation was evident in cultures overlaid with agar 8 to 13 days after inocu- lation and appeared earliest in cultures inoculated with the highest dose of virus (lo4 FFU/ml).

Cat and marmoset fibroblast cultures infected with the original SM-FeSV obtained from Dr. Sarma were propagated for 35 passages. Unlike the feline cultures which became transformed within 10-14 days after inoculation and before subculture, the marmoset cultures were transformed only after 14 or more cell culture passages, with the transformation resembling that seen in the feline embryonic fibroblast cultures.

Characteristics of' virus produced by transformed cultures

The release of virus from MF-69CBlM cultures infected with 8 . 8 ~ 1 0 ~ FFU/ml of ST-FeSV from marmoset tumor-cell line No. 2 and subcultured every 5 to 7 days was examined. Very low titers of

infectious virus (about 0.5 x 10' FFU/ml) persisted in the infected cultures for 2-12 h after inoculation, then disappeared and reappeared in very small amounts 36-48 h after inoculation (about 0.5 x 10' FFU/ml). The titer of infectious virus in preparations of frozen and thawed cells suspended in the original culture medium increased with each subculture of the infected cells and paralleled the increasing fraction of transformed cells in the cultures; by the 6th subculture, the titer had increased to 7.8 x lo2 FFU/ml. FA studies of the infected MF-69CBlM cultures using marmoset sera to GA-FeSV showed small amounts of viral antigen appearing in the infected cells as early as 24 h post inoculation. Similar results were obtained with GA-FeSV- infected marmoset cell cultures.

In contrast, although feline cultures infected with SM-FeSV released transforming virus, infected marmoset cultures released transforming virus only after 26 serial passages of infected cells (12 passages after transformation was first observed). Viruses isolated after 35 passages from SM-FeSV infected feline and marmoset cultures still induced the same type of morphological transformation (r and fr) of feline cultures as the original virus. Furthermore, SM-FeSV isolated from infected and subcultured feline fibroblasts, passage 35, resembled the original virus in that it failed to transform marmoset cultures but did transform feline fibroblasts. In contrast, SM- FeSV isolated from marmoset cultures after 26, 30 and 35 passages of SM-FeSV-infected cells trans- formed both feline and marmoset fibroblasts, and the

TABLE I V

TITRATION OF SM-FeSV PREPARED FROM INFECTED FELINE A N D MARMOSET CULTURES

Cells infected with SM-FeSV

Focus assav on FEF-R and MF-69CB1M cells (FFUIml) . . _ FEF-R MF-69CBlM SM-FcSV from

cell passage level ' -DEAE a +DEAE -DEAE +DEAE

MF-69CBl M (marmoset)

FEF-R (cat)

2 8

14 16 26 30 35

2 13 18 23 26 35

0 0 0 0

6.5 X l o a 9.0 x loa 3.5 x lo3

i .9 x 104 2.3 x 104 1 .0~104 1 .1~104

1.1 xi04 1 . 3 ~ 1 0 ~

0 ND

0 0

ND

1.ox 1 0 4

0

1.2 x 103

7.0 x 104 4.0 x 104 2.8 x 10' 6.4 x 104

ND

0 0 0 0

0.9 x 10' 3.0~ 10'

ND

0 0 0 0 0 0

0 ND

0 0

ND 3.5 x 1 0 2

ND

0 0 0 0 0

ND

' Number of cell passages after infection with FcSV. - Each FFU/ml reported is the average of two or three experiments with at least 6 flasks per experiment. - a fDEAE = treatment of cultures with diethylaminoethyl-dextran (Sigma Chemical Co.); -DEAE = no treatment with DEAE-dextran. - ' N D = not done.

402 MCDONLAD ET AL.

TABLE V

INFECTIVITY OF SM-FeSV, PROPAGATED IN FELINE AND/OR MARMOSET CELL CULTURES

Virus inocula

Focus assay on FEF-R and MF-69CBIM cells

(FFUlrnl)

FEF-R MF-69CBIM

SM-FeSV propagated in marmoset cell cultures for 0.4 x I 0' 30 passages 7.5 x lo2 0 . 4 ~ loL

SM-FeSV propagated in marmoset fibroblasts for 30 passages and sub- sequently in feline cell 9.8 x lo2 1.1 x lo1 cultures for 6 passages 7.4 x lo2 1.0 x 10'

SM-FeSV propagated in feline cell cultures for 23 passages 1.2x104 0

8.0 x lo2

morphological transformation resembled that seen with the original virus on feline cultures (r and fr). The titer of SM-FeSV from infected marmoset cultures, passages 26,30 and 35, was 20- to 160-fold higher on feline fibroblasts than on marmoset fibroblasts, but still 15- to 160-fold lower on feline fibroblasts than titers produced by SM-FeSV derived from infected feline cells (Table IV). DEAE-dextran treatment of the cultures did not alter the ratio of r to fr foci but it did increase the titer of SM-FeSV on feline and marmoset cells 3- to 40-fold.

The stability of the altered host range of SM-FeSV propagated in marmoset fibroblasts was examined in two ways. First, SM-FeSV prepared from infected and subcultured marmoset cells, passage 26, was inoculated into fresh marmoset cultures and the cultures were propagated for 26 additional passages.

Virus prepared from these cultures (after the additional 26 subcultures) transformed feline and marmoset cultures with the same efficiency as the virus prepared from the original passage series in marmoset cells. Secondly, SM-FeSV prepared from infected and subcultured marmoset cells, passage 30, was inoculated into feline (FEF-R) cultures and the infected feline fibroblasts were subcultured for six passages. At the 6th subculture, virus was isolated and tested for transformation on feline and mar- moset fibroblasts. As shown in Table V, virus prepared from the 6th subculture in feline cells had the same relative titers on feline and marmoset fibro- blasts as the SM-FeSV isolated from the infected and subcultured MF-69CB1 M marmoset cells, passage 30.

Although SM-FeSV propagated in marmoset cells for 26 passages or more could transform marmoset cells whereas the virus propagated in feline fibroblasts could not, SM-FeSV propagated in marmoset cells did not show a change of its species-specific antigens. SM-FeSV propagated in marmoset cells for 26 passages or more retained the feline virus antigens and lacked the primate SSV-1 antigens by CF and FA (Table VI).

Neutralization tests

Sera from cats and marmosets with ST-FeSV and GA-FeSV induced tumors were examined for neutralizing antibodies against ST-FeSV and SM- FeSV obtained from infected feline and marmoset cells (Table VII). The highest titers were obtained with sera from marmosets infected with GA-FeSV when tested against ST-FeSV or SM-FeSV of either cat or marmoset origin; sera from marmosets bearing tumors induced by ST-FeSV or sera from cats bearing tumors induced by either ST-FeSV or

TABLE VI

VlRAL ANTIGENS IN SM-FeSV-INFECTED MARMOSET AND FELINE CELLS

Cell cultures FeSV antigens SSV-I antigens

C F ' F A a C F S FA ' Number of cell culture passages

SM-FeSV-infected MF-69CB1 M 12, 16 and 22 tundil. - < I :4 - (marmoset) 28 and 30 1 :4 + MF-69CB1 M (marmoset) control 10 and 45 <undil. - 1 :4 SM-FeSV-infected FEF-NCE (cat) 17 >1:16 + < I :4 FEF-NCE (cat) control 14 and 22 I :4 ND <1:4 -

<1:4 -

- -

' For the CF assay, 2 antibody units of guinea-pig antisera against tween-ether-treated F-422 feline leukemia virus and 2 units of guinea-pig complement were used. The antisera contained antibodies to feline leukemia virus, P30, PI4 and PI0 antigens and did not react with SSV-I antigens. Titers were recorded as the highest dilution giving a 3 f fixation of 2 units of complement. - Serum from a marmoset bearing a tumor induced by GA-FeSV was used for the FA assay. The serum was broadly reacting with the major viral proteins of FeSV but not of SSV-I. - 'For the CF assay, antisera were prepared in guinea-pigs against tween-ether-treated SSV-I: these sera did not react with FeLV or FeSV antigens. - For the FA assay, serum was obtained from a marmoset bearing a tumor induced by SSV-I ; this serum reacted with the three major proteins of SSV-I (P30, F14, PIZ) and also with the envelope glycoproteins but not with FeLV or FeSV antigens. - ' ND = not done.

FeSV IN CELL CULTURES 403

TABLE VII

ANTIGENIC COMPARISON OF ST-FeSV GROWN IN FELINE OR PRIMATE CELLS BY NEUTRALIZATION TESTS

50% inhibition o f focus formation ' SM-FcSV '

from cell cultures of ST-FeSV '

from cell cultures of Antisera Antibodies to kAE F ~ L V by IEOP antigens '

Virus strain Species Cat Marmoset Cat Marmoset

ST-FeSV Cat 1 ND 1 :13 1:144 <1:8 <1:8 2 - 1 :28 1:360 <1:8 <1:8

GA-FeSV

Marmoset 1 +/- 1 :265 1:375 <1:8 <1:8 2 +/- 1 :212 1 :302 1 :12 1 :16 3 +I- < l : 8 <1:8 4 +I- 1 :8 1 :19

Cat 1 - 150 1:294 <1:8 <1:8 2 - 154 1:383 <1:8, <1:8

Marmoset 1 + 1 : 1,074 1 :2,374 2 + 1:1,124 1 :685 1 : 1,024 1 :690 3 + 1 :1,024 4 + 1 :2,370 1 :1,280

Control cat sera 1 - <I :8 <1:8 <1:8 <1:8

Control marmoset sera 1 - <1 :8 <1:8 <1:8 <1:8

The 50% inhibition of focus formation is listed as the dilution of antisera which prevented development of 50% o f the foci on FEF-R. A 1 :8 dilution of the antisera was the lowest dilution tested. - ' Each value listed is an average of two or three experiments with a minimum of two flasks per serum dilution in each experiment. - a Antibodies to FeLV antigens determined by immunoelectro-osmophoresis (IEOP) against tween-ether-treated F-422 virus. Animals whose sera had antibody activity at one bleeding but lacked it at other blecdings were scored a s + / - . - ' ST-FcSV inocula were prepared from cat tumor-cell line No. 1 and marmoset tumor-cell line No. 2. SM-FeSV inocula were prepared from infected and propagated FEF-R cells, passages 26 and 35. and from infected and propagated MF-figCBlM, passages 30 and 35. - N D = not done.

GA-FeSV contained lower neutralizing antibody titers than sera from marmosets infected with GA- FeSV. Cats inoculated with either ST-FeSV or GA- FeSV did not develop measurable neutralizing antibodies to SM-FeSV, whereas some sera from marmosets inoculated with either ST-FeSV or GA- FeSV did cross-react with SM-FeSV.

DISCUSSION

Three strains of feline fibrosarcoma virus (ST- FeSV, GA-FeSV and SM-FeSV) were examined for their ability to induce tumors in marmosets and to transform marmoset cells in vitro. Two of the three strains, ST-FeSV and GA-FeSV, induced tumors in marmosets and transformed marmoset cells in vitro after primary inoculation; the third strain, SM-FeSV, did neither. However, after the SM-FeSV-infected marmoset cell cultures were passaged 14 times in vitro, the cells became transformed and after 26 passages, the cell cultures released infectious virus which transformed marmoset and feline fibroblasts but failed to induce tumors in marmosets. The latter may however have been due to low-titered inocula (lo2 to lo3 FFU per inoculum) as >lo3 FFU per inoculum are generally needed for tumor induction with other C-type viruses in marmosets (Wolfe and Deinhardt, 1972).

The three strains of FeSV induced transformation of feline and marmoset cells of the large round-cell type (r foci), mixed small round-fusiform cell type (fr foci) and mixed r/fr foci. ST-FeSV induced mainly r type transformation, GA-FeSV induced predominantly fr type with some large r cells and SM-FeSV induced mainly fr /type transformation with a higher proportion of fusiform cells than was seen with GA-FeSV. Transforming virus was isolated from r type foci but not from fr type foci of ST-FeSV and the virus derived from r type foci induced almost exclusively r type foci over several serial passages. Incubation at 37" C for 30 to 60 min changed the virus, and the treated virus induced an increased number of fr foci. These observations were consistent with the finding that the amount of trans- forming virus produced by FeSV-infected cell lines derived either from tumors or from cells infected in vitro was related to the number of r cells in such cultures, i.e. the higher the percentage of r cells, the greater the amount of transforming virus produced. The amount of non-transforming virus produced by r or fr type transformed cells has not as yet been determined. These data indicate that the induction of r or fr type transformation is controlled by the infecting virus and that this virus genome function is stable under natural conditions but altered when extracellular virus is exposed to as low a temperature

404 MCDONALD ET AL.

as 37" C. It is most likely that fr type transformation is caused by partially inactivated sarcoma virus particles which can still effect the fr type trans- formation of cells but which can no longer produce transforming progeny virus. Passage of FeSV in marmoset cells yielded viruses with a quantitatively altered host range; i.e. virus grown in cat cells had a higher ratio of infectivity for feline vs marmoset cells than virus grown in marmoset cells. The three strains differed however in the degree of their change of infectivity for feline or marmoset cells, ST-FeSV from marmoset or feline tumor-cell lines showed a higher infectivity on feline than on marmoset cells but the relative infectivity for marmoset cells was increased by passage in marmoset cells (6- to 23-fold), whereas GA-FeSV from marmoset tumor-cell lines frequently showed a higher infectivity on marmoset than on cat cells. SM-FeSV grown in feline cells infected but did not transform marmoset cells after primary infection. However, SM-FeSV was produced when SM-FeSV-infected marmoset cells were passaged several times in vitro and this virus trans- formed both feline and marmoset cells in vitro, but was still 30 to 200 times more infective for feline than for marmoset cells. The stability of the change in host range of FeSV was tested for SM-FeSV produced by marmoset cells and was found unaltered when the virus was passaged six times in feline cells.

Despite the alteration of host range of SM-FeSV propagated in marmoset fibroblasts, the virus retained feline virus-specific antigens and was negative for the antigens of a non-human primate sarcoma virus (SSV-I) by C F and FA assays. Also, the neutralization tests did not indicate any alteration of the surface antigens of ST-FeSV or SM-FeSV propagated in marmoset fibroblasts. FeSV propa- gated in either marmoset or feline fibroblasts showed similar neutralizing titers with antisera from cats and marmosets bearing tumors induced by ST-FeSV and GA-FeSV. However, a distinct difference in neu- tralization by the same sera was observed between ST-FeSV and SM-FeSV, and ST-FeSV grown in

marmoset cells was neutralized by both cat and marmoset sera at a somewhat higher serum titer than virus grown in cat cells. The latter could be due to a smaller amount of FeSV-associated virus present in the FeSV stocks grown in marmoset cells but this was not further evaluated. ST-FeSV and SM-FeSV both contain viruses of subgroups A and B (Sarma and Log, 1973; Sarma et al., 1971b, 1972) but it is possible that during passage of the three FeSV strains in our laboratory either A- or 8-type viruses propagated preferentially which could explain the difference in neutralization titers between ST-FeSV and SM- FeSV, although it was shown that ST-FeSV grown in marmoset cells still consisted of a mixture of A- and B-type virus (Sarma, personal communication). Marmosets generally developed higher titered and more broadly reacting antibodies than cats; this was particularly evident in animals infected with GA- FeSV and perhaps explained by partial tolerance of cats to some of the FeSV envelope antigens.

The alteration of FeSV during passage in marmoset cells could be due to selection of a variant present in the original heterogeneous uncloned population or more probably to mutation or recombination with some marmoset genetic material, possibly an unidentified endogenous marmoset virus. Particularly the latter mechanisms have been postulated to explain similar observations made with avian (Altaner and Svec, 1966; Altaner and Temin, 1970; Bauer and Graf, 1969; Kuwata, 1964), murine (Aaronson, 1971; Scolnick and Parks, 1974) or feline (Chan et al., 1974a, b) C-type RNA viruses.

ACKNOWLEDGEMENTS

These studies were supported by Contract N01- CP-33219 within the Virus Cancer Program of the National Cancer Institute, National Institutes of Health, US Public Health Service. We thank the Board of Health, City of Chicago, for providing space for housing many of our experimental animals.

CARACTERISTIQUES DE TROIS SOUCHES DE VIRUS DU FIBROSARCOME FELIN CULTIVEES DANS DES CELLULES DE CHAT ET DE MARMOUSET

On a constate que deux des trois souches de virus diifibrosarcome de chat (ST-FeSVet GA-FeSV) induisent des tumeurs chez les chats et les marmousets, et transforment in vitro les cellules de chat et de marmouset a p r h l'inoculation primaire. Une troisiime souche (SM-FeSV) n'a pas induit de tumeur ni transforme' les cellules de marmouset apris I'inoculation primaire; cependant, lorsque des cultures de cellules de marmouset infecte'es avec le virus SM-FeSV ont Pte'passees 26 fois in vitro, les cellules cultive'es ont produit des virus qui ont trans- form& les fibroblastes de marmouset mais n'ont pu induire de tumeurs chez ces animaux. Le virus ST-FeSV a induit principalement des transformations de type b cellules rondes (foyers r ) , le virus GA-FeSV a induit d'une maniire prgdominante des types mixtes b cellules rondes et fusiformes (foyers f r ) , et le virus SM-FeSV a induit les deux types cellulaires r et fr avec une plus grande proportion de cellules fusifarmes que la souche GA- FeSV. On a obtenu des virus produisant la transformation apartir des foyers r ou des foyers mixtes rlfr induits

FeSV IN CELL CULTURES 405

par le virus ST-FeS V, mais pas a partir des foyers f r ; le traitement par la chaleur a changt le virus produisant presque exclusivement des foyers r en tin virus qui induit un plus grand nombre Lie foyers f r . Le passage drr virus FeSV dam des cellules de chat a produit des virus posse'dant un plus grand taux d'infectivite' pour les cellules de chat que pour celles de marmouset, tandis que le passage dans des cellules de marmouset a produit des virus ayant un taux d'infectivitt relativement plus grand pour les celliiles de marmouset. Le degre' de change- rnent du faux d'infectivitt t ta i f diffe'rent pour chaque souche. En dPpit de I'alfe'ration de I'infecfivife' du virus SM-FeSV envers les fibroblastes de marmouset, ce virus a retenu un antigene P-30, comme I'ont montrk les Ppreuves de fixation du complkment et des anticorps puorescents. Les tests de neutralisation n'ont pas indiquk mais n'ont pas exclu une altkration des antigenes de surface des virus ST-FeSV ou SM-FeSV cultivks dans les fibroblastes de marrnouset. Les altkrations de I'infectivitP relative du virus FeSV lors du passage dam les cellules de marrnouset suggkrent trois explications possibles: ( 1 ) la stlection d'un variant du virus prksent dans la population multiclonale originale, ( 2 ) une mutation ou ( 3 ) une combinaison avec le matkrial ge'nttique du marmouset, peut-etre un virus endoghe iiu marmouset non encore identifit.

REFERENCES

AARONSON, S. A,, Common genetic alterations of RNA tumor viruses grown in human cells. Nnture (Lond.), 230, 445-447 (1971).

ALTANER, C., and SVEC, F., Virus production in rat tumors induced by chicken sarcoma virus. J. nat. Cancer Inst., 37,

ALTANER, C., and TEMIN, H. M., Carcinogenesis by RNA sarcoma viruses. XII. A quantitative study of infection of rat cells in vitro by avian sarcoma viruses. Virology, 40, 118-134 (1970).

BAUER, H., and GRAF, T., Evidence for the possible existence of two envelope antigenic determinants and corresponding cell receptors for avian tumor viruses. Virology, 37, 157-161 (1969).

CHAN, E. W., SCHIOP-STANSLY, P. E., and OTONNOR, T., Rescue of cell transforming virus from a non-virus-producing bovine cell culture transformed by feline sarcoma virus. J. nat. Cancer Inst., 52, 469-472 (19740).

CHAN, E. W., SCHIOP-STANSLY, P. E., and O'CONNOW, T., Mammalian sarcoma-leukemia viruses. 1. Infection of feline, bovine and human cell cultures with Snyder-Theilen feline sarcoma virus. J. nut. Cancer Inst., 52, 473-481 (19746).

DEINHARDT, F., WOLFE, L. G., NORTHROP, R., MARCZYNSKA, B., OGDEN, J., MCDONALD, R., FALK, L., SHRAMEK, G., SMITH, R., and DEINHARDT, J., Induction of neoplasms by viruses in marmoset monkeys. J . med. Primatol., 1, 29-50 (1972).

SNYDER. S. P., ST-feline fibrosarcoma virus: induction of tumors in marmoset monkeys. Science, 167, 881 (1970).

GARDNER, M . B., ARNSTEIN, P., JOHNSON, E., RONGEY, R. W., CHARMAN, H. P., and HUEBNER, R. J., Feline sarcoma virus induction in cats and dogs. J. Amer. vet. Med. Ass., 158, 1046-1053 (1971).

GARDNER, M. B., RONGEY, R. W., ARNSTEIN, P., ESTES, J. D., SARMA, P., HUEBNER, R. J., and RICKARD, C. G., Experi- mental transmissioh of feline fibrosarcomas to cats and dogs. Nature (Lond.), 226, 807-809 (1970).

HAMPAR, B., KELLOF, G. J., MARTOS, L. M., OROSZLAN, S., GILDEN, R. V., and WALKER, J. L., Replication of murine and feline RNA-containing C-type viruses in human lympho- blastoid cells. Nature (Lond.), 228, 857-859 (1970).

HOEKSTRA, J., and DmwARDT, F., Counterimmunoelectro- phoresis-rapid method for detecting gs antigen and anti- bodies associated with oncogenic RNA viruses. Appl. Microbiol., 22, 1172-1173 (1971).

KUWATA, T., Studies of adaptation of Rous sarcoma virus to ducks. Cancer Res., 24, 947-956 (1964).

745-752 ( I 966).

DEINHARDT, F., WOLFE, L. G., THEILEN, G. H., and

LEE, K. M., Comments on feline leukemia virus. J. Amer. vet. Med. Ass., 158, 1037-1039 (1971).

MCALLISTER, R. M., FILBERT, J. E., NICOLSON, M. O., RONGEY, R. W., GARDNER, M. B., GILDEN, R. V., and HUEBNER, R. J., Transformation and productive infection of human osteosarcoma cells by a feline sarcoma virus. Nature New Biol., 230, 279-282 (1971).

MCALLISTER, R. M., NICOLSON, M., GARDNER, M. B., RONGEY, R. W., RASHEED, S., SARMA, P. S., HUEBNER, R. J., HATANAKA, M., OROSZLAN, S., GILDEN, R. V., KABIGTING, A., and VERNON, L., C-type virus released from cultured human rhabdomyosarcoma cells. Nature New Biol., 235, 3-6 (1972).

MCDONALD, R., WOLFE, L. G., and DEINHARDT, F., In vitro assay for ST-feline fibrosarcoma virus. Fed. Proc., 30, 302 (1971).

MCDONALD, R., WOLFE, L. G., and DEINHARDT, F., Feline fibrosarcoma virus : quantitative focus assay, focus mor- phology and evidence for a " helper virus ". In t . J. Cancer, 9, 57-65 (1972).

MCDONOUGH, S. K., LARSEN, S., BRODEY, R. S., STOCK, N. D., and HARDY, W. D., JR., A transmissible feline fibrosarcoma of viral origin. Cnncer Res., 31, 953-956 (1971).

PEARSON, L. D., SNYDER, S. P., and ALDRICH, C. D., Oncogenic activity of feline fibrosarcoma virus in newborn pigs. Amel-. J . vet. Res., 34. 405-409 (1973).

RABIN, H. L., Assay and pathogenesis of oncogenic viruses in non-human primates. Lab. anim. Sci., 21, 1032-1049 (1971).

RICKARD, C. G., POST, J. E., NORONHA, F., and BARR, L. M., A transmissible virus-induced lymphocytic leukemia of the cat. J. nat. Cancer Inst., 42, 987-1014 (1969).

SARMA, P. S., BASKAR, J. F., GILDEN, R. V., GARDNER, M. B., and HUEBNER, R. J., I n vitro isolation and characterization of the GA strain of feline sarcoma virus. Proc. SOC. exp. Biol. ( N . Y . ) , 137, 1333-1336 (1971a).

SARMA, P. S., HUEBNER, R. J., BASKER, J. F., VERNON, L., and GILDEN, R. V., Feline leukemia and sarcoma viruses: susceptibility of human cells to infection. Science, 168, 1098-1100 (1970).

SARMA, P. S., and LOG, T., Subgroup classification of feline leukemia and sarcoma viruses by viral interference and neutralization tests. Virology, 34, 160-169 (1973).

SARMA, P. S., LOG, T., and THEILEN, G. H., ST-feline sarcoma virus. Biological characteristics and in vitro propagation. Proc. Soc. exp. Biol. ( N . Y . ) , 137, 1444-1448 (19716).

SARMA, P. S., SHANAR, A. L., and MCDONOUGH, S., The SM strain of feline sarcoma virus. Biologic and antigenic

406 MCDONALD ET AL.

characterization of virus. Proc. Soc. exp. Biol. ( N . Y . ) , 140,

SCOLNICK, E. M., and PARKS, W. P., Harvey sarcoma virus: a second murine type-C sarcoma virus with rat genetic information. J. Virol., 13, 1211-1219 (1974).

SNYDER, S. P., Spontaneous feline fibrosarcomas: transrnis- sibility and ultrastructure of associated virus-like particles. J . nut. Cancer Inst., 47, 1079-1085 (1971).

SNYDER, S. P., and THEILEN, G. H., Transmissible feline fibrosarcoma. Nature (Lond.), 221, 1074-1075 (1969).

THEILEN, G. H., Continuing studies with transmissible feline fibrosarcoma virus in fetal and newborn sheep. J. Amer. vet. Med. Ass., 158, 1040-1045 (1971).

THEILEN, G. H., SNYDER, S. P., WOLFE, L. G., and LANDON, J. C., Biological studies with viral induced fibro- sarcomas in cats, dogs, rabbits and non-human primates. In: R. M. Dutcher (ed.), Comparative leukemia research 1969, No. 36, Bibl. ffaematol., pp. 373-400, S . Karger, New York (1 970).

1365-1368 (1972). UBERTINI, T., NORONHA, F., POST, J. E., and RICKARD. C. G., A fluorescent antibody technique for the detection of the group-specific antigen of feline leukemia virus in infected tissue-culture cells. Virology, 44, 219-222 (1971).

WOLFE, L. G., and DEINHARDT, F., Oncornaviruses associated with spontaneous and experimentally induced neoplasia in non-human primates. In ; E. I. Goldsmith and J. Moor- Jankowski (ed.), Medical primatology, part I I I , Proc. 3rd Con$ Exp. Med. Surg. Primates, 1972, pp. 176-196, Karger, Basel (1 972).

WOLFE, L. G., MARCZYNSKA, B., RABIN, H., SWTH, R., TISCHENDORF, P., GAVITT, F., and DEINHARDT, F., Viral oncogenesis in non-human primates. In : E. I. Goldsmith and J. Moor-Jankowski (ed.), Proc. 2nd Conf. Exp. Med. Surg. Primates, 1969, pp. 671-682, S. Karger, New York (1971).

WOLFE, L. G., SMITH, R. D., HOEKSTRA, J., MARCZYNSKA, B., SMITH, R. K., MCDONALD, R., NORTHROP, R. L., and DEINHARDT, F., Oncogenicity of feline fibrosarcoma viruses in marmoset monkeys: pathologic, virologic and immuno- logic findings. J. nut. Cancer Inst. , 49, 519-533 (1972).