Suppression of H-2b-associated resistance to Friend erythroleukemia virus by a class I gene from the H-2d major histocompatibility complex haplotype.
ABSTRACT Mice homozygous for the H-2d haplotype at the major histocompatibility complex are markedly more susceptible to erythroleukemia induction by the Friend isolate of murine leukemia retrovirus (FV) than are congenic mice homozygous for the H-2b haplotype. The resistance conferred by the H-2b haplotype is recessive in this cross, since heterozygous F1 mice are as susceptible as parental strain H-2d homozygotes. However, H-2b-associated resistance is not an intrinsically recessive trait, since H-2b/H-2dm1 heterozygotes resemble H-2b homozygotes in their relative resistance to FV; the mutant H-2dm1 haplotype lacks the entire D region of the parental haplotype except for a single class I gene formed by the fusion of its terminal D-region genes to produce a class I gene differing from both parental genes, and thus this finding indicates that one or more D-region genes of the H-2d haplotype can actively suppress H-2b-associated resistance. Unlike H-2dm1, the mutant H-2dm2 haplotype, which retains only the class IDd gene in the D region of the H-2d haplotype, strongly suppresses resistance in H-2b/H-2dm2 heterozygotes, and the presence of Dd as a transgene significantly reduces the resistance of H-2b homozygotes. Since H-2b-associated resistance to FV appears to be due mainly to the capacity of Lb (also called Db), the only class I molecule encoded in the D region of the H-2b haplotype, to present viral epitopes for recognition by FV-specific cytotoxic T lymphocytes, suppression of resistance to FV by the Dd molecule implies that the presence of one class I molecule of the major histocompatibility complex can interfere with either the presentation of viral epitopes by another class I molecule or the generation of T cells that recognize viral epitopes so presented.
- SourceAvailable from: PubMed Central[show abstract] [hide abstract]
ABSTRACT: The H-2dm1 mutation is unique among all described H-2 mutations in that two transplantation antigens, the H-2Dd and the H-2Ld, are affected. Here, we show that the mutant gene, Ddm1, is formed by fusion of the 5' part of the Dd gene and the 3' part of the Ld gene, with the region in between deleted. The recombination junction is located in the third exon, which encodes the alpha 2 region of the protein. When the hybrid gene is transfected into mouse L cells, serological and biochemical analyses indicate the Ddm1 antigen expressed in the transformant line is identical to the mutant molecule in dm1 spleen cells. These results demonstrate that the D/L hybrid gene is most likely responsible for the dm1 mutant phenotype.Journal of Experimental Medicine 12/1985; 162(5):1588-602. · 13.21 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Sequential immunoprecipitates show that H-2dm1 mutant cells express a hybrid "H-2D/L" antigen exhibiting determinants normally associated with two different gene products of the parental d haplotype-i.e., the H-2Dd and H-2Ld antigens. The hybrid H-2D/Ldm1 antigen appears to consist of a portion of the NH2-terminal extracellular half of the H-2Dd antigen "fused" to a portion of the COOH-terminal extracellular half of the H-2Ld antigen. This structure is inferred from the reactivity of dm1 antigens with cytotoxic T lymphocytes specific for H-2Ld determinants and with monoclonal antibodies specific for determinants in the structural domains of H-2Ld or H-2Dd. The H-2D/Ldm1 molecule apparently retains all of the third external domain (C2 or alpha 3) and part of the second external domain (C1 or alpha 2) of H-2Ld, but its first external domain (N or alpha 1) derives from H-2Dd. From these findings and from previous peptide mapping studies, we propose that the H-2D/Ldm1 antigen is the product of a hybrid gene that has resulted from an unequal crossover between the parental H-2Dd and H-2Ld genes, leaving the N exon and part of the C1 exon of the H-2Dd gene joined to the H-2Ld gene beginning somewhere within its C1 exon.Proceedings of the National Academy of Sciences 09/1984; 81(16):5204-8. · 9.74 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Two types of quantitative response to the F-B strain of Friend virus in segregating generations of a cross involving a susceptible (DBA/2 or BALB/c; H-2(2)) and a resistant (C57BL/6; H-2(b)) mouse strain show a marked correlation with the H-2 type of the mice. Essential susceptibility, as determined by the splenomegalic response to high virus doses, is controlled by a single pair of alleles which segregates independently with respect to the H-2 locus. However, relative susceptibility, as determined by the incidence of the splenomegalic response at moderate or low levels of virus dosage, is significantly greater among mice homozygous or heterozygous for the H-2(d) allele than among H-2(b) homozygotes in these populations. In addition, the incidence of recovery from splenomegaly induced by a given level of virus dosage is significantly greater in H-2(b) homozygotes than in segregants of other H-2 types among their littermates. Possible mechanisms responsible for these effects are discussed.Journal of Experimental Medicine 04/1968; 127(3):465-73. · 13.21 Impact Factor
Proc. Natl. Acad. Sci. USA
Vol. 88, pp. 9243-9247, October 1991
Suppression ofH-2b-associated resistance to Friend
erythroleukemia virus by a class I gene from the
H-2d major histocompatibility complex haplotype
DAVID POLSKY AND FRANK LILLY
Department of Molecular Genetics, Albert Einstein College of Medicine, Bronx, NY 10461
Contributed by Frank Lilly, July 12, 1991
the major histocompatibility complex are markedly more sus-
ceptible to erythroleukemia induction by the Friend isolate of
murine leukemia retrovirus (FV) than are congenic mice
homozygous for the H-2b haplotype. The resistance conferred
by the H-2b haplotype is recessive in this cross, since hetero-
zygous F1 mice are as susceptible as parental strain H-2d
homozygotes. However, H-2bassociated resistance is not an
intrinsically recessive trait, since H-2b/H-2dId' heterozygotes
resemble H-2b homozygotes in their relative resistance to FV;
the mutant H-2dIN haplotype lacks the entire D region of the
parental haplotype except fora single class Igene formed by the
fusion of its terminal D-region genes to produce a class I gene
differing from both parental genes, and thus this finding
indicates that one ormoreD-region genes oftheH-2dhaplotype
can actively suppress H-2b-associated resistance. Unlike
H-2dAIl, the mutant H-2'"'2 haplotype, which retains only the
class I D" gene in theD region of the H-2dhaplotype, strongly
suppresses resistance in H-2b/H-2dm2 heterozygotes, and the
presence of Dd as a transgene significantly reduces the resis-
tance ofH-2b homozygotes. SinceH-2bassociated resistance to
FV appears to be due mainly to the capacity ofLb (also called
Db), the only class I molecule encoded in the D region of the
H-2b haplotype, to present viral epitopes for recognition by
FV-specific cytotoxic T lymphocytes, suppression of resistance
to FV by the Dd molecule implies that the presence ofone class
I molecule of the major histocompatibility complex can inter-
fere with either the presentation of viral epitopes by another
class I molecule or the generation ofT cells that recognize viral
epitopes so presented.
Mice homozygous for the H-2d haplotype at
The murine major histocompatibility complex, H-2, is one of
several known genetic factors that can influence the resis-
tance or susceptibility to various retrovirus-induced diseases
in mice (1, 2). In most studies, the H-2b haplotype has proved
to confer relatively strong resistance to the virus-induced
disease. In the first such study, the capacity of the H-2b
haplotype to protect against lymphomas induced by Gross
murine leukemia virus was a genetically dominant trait, since
H-2b/H-2k heterozygotes in a segregating backcross popu-
lation showed a much lowerlymphoma incidence than that of
their H-2k-homozygous littermates (3). However, in the
second such study, resistance to Friend erythroleukemia
virus (FV) appeared to be recessively associated with H-2b,
since about 1/20th as much virus was required to induce
comparable levels of disease in H-2b/H-2d heterozygotes in
a backcross population as in their H-2b-homozygous litter-
It has long been clear that H-2 exerts this protective effect
mainly by its capacity to influence the hosts' immune re-
sponses to viral antigens; mice producing stronger, more
effective immune responses are relatively resistant to viral
pathogenesis. In most cases ofH-2-determined differences in
immune responsiveness, haplotypes governing higher levels
ofresponse have been genetically dominant in heterozygotes
over those governing lower levels (5, 6); thus it has been
puzzling that H-2b-associated resistance to FV is a recessive
trait. The present studies were based on the hypothesis that
the H-2d haplotype includes one or more genes that suppress
H-2b-associated resistance. The results show that the deter-
minant(s) of the suppressive effect are located in the H-2Dd
region and that theDdgene mediates the effect at least in part.
MATERIALS AND METHODS
Mice. BALB/cAn (H-2d) and H-2-congenic strains
BALB.B (H_2b), BALB.G (H-25), and BALB.5R (H-2ls), as
well as C57BL/lOSn (B10, H-2b), were from ourown colony.
B10.D2-H-2dm1 (dml) breeding pairs were a gift from C. S.
David (Mayo Clinic, Rochester, MN); these mice carry an
H-2d haplotype with a mutation in the D region. BALB/c-
H-2dm2 (dm2) breeding pairs were purchased from The Jack-
son Laboratory; these mice carry an H-2d haplotype with a
different D-region mutation. Breeding pairs of the B6-tD8
(D8) strain, which are C57BL/6 mice (H-2b) expressing the
Dd transgene (7), were supplied by the courtesy of G. Jay
(American Red Cross, Rockville, MD). These H-2 haplo-
types are represented in Table 1. All experimental mice of
these strains and of F1 crosses among them were bred in our
Virus. Stocks ofFV were derived from mixtures ofculture
fluids from cells producing biologically cloned components of
FV. Culture fluids from F201NIH cells [a line ofNIH/3T3 into
which a molecularly cloned NB-tropic helper component of
FV, Friend murine leukemia virus FMuLVA, was introduced
(8)] and from SFFV-FRE/FMuLV cells [a line ofFRE cells
into which the defective pathogenic component of FV,
SFFVp, was biologically cloned and which was then super-
infected with FMuLV from F201NIH cells (9)] were mixed in
a 1:3 ratio; BALB/c mice received 1 ml i.v. of this mixture
['400 spleen focus-forming units (SFFU)/ml]. The virus was
passed in vivo twice more from mice infected 12-15 days
earlier. Virus stocks used in the present experiments were
clarified supernatants from 10% (wt/vol) homogenates ofthe
greatly enlarged spleens ofmice ofthe last passage, stored in
1-ml lots at -70°C for up to 6 months.
Experimental Procedures. Virus stocks were titrated by the
spleen focus method (10) in BALB/c mice (usual titer =104
SFFU/ml) and diluted to produce 10 or 20 SFFU/ml, de-
pending on the experiment. Male or female mice 6-16 weeks
of age received 1 ml of the desired virus dilution in the tail
Abbreviations: FV, Friend erythroleukemia virus; SFFU, spleen
focus-forming unit; CTL, cytotoxic T lymphocytes.
The publication costs of this article were defrayed in part by page charge
payment. This article must therefore be hereby marked "advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Medical Sciences: Polsky and Lilly
H-2 haplotypes used in these studies
b + Dd*
b (+ Dd)
vein. Spleen size was determined by palpation twice weekly
from day 7 to 28 and once weekly thereafter for at least 2
months. Mice were deemed splenomegalic if the spleen was
.0.5 g, but in most cases 0.5-g spleens rapidly progressed in
size to at least 1.5-2 g. Results are given in the text and tables
as number ofmice with splenomegaly/number ofmice in the
experimental group. Depending on the virus dose and the
genetic constitution of the hosts, some mice with the pro-
nounced splenomegaly, which usually peaked around 2-3
weeks after virus administration, showed regression to nor-
mal spleen size; after prolonged observation some regressor
mice relapsed into splenomegaly again. In experiments in-
cluding mice of different ages and sexes, older mice (>10
weeks) were slightly more resistant than younger ones, and
males were slightly more resistant than females, but in no
single experiment did these differences approach statistical
significance (P > 0.20).
Mapping of Resistance to FV Using Recombinant H-2 Hap-
lotypes. Groups of mice of various H-2-congenic BALB
strains and crosses received 10 SFFU of FV and were
monitored by palpation for the development of splenome-
galy. Three categories of response emerged from these stud-
ies according to the latencies and the final incidences for
splenomegaly development (Fig. 1 and Table 2, experiment
1). BALB/c and BALB.5R mice and their F1 crosses with
BALB.B were the most highly susceptible, showing pro-
Days post infection
H-2-congenic strains and crosses after receiving 10 SFFU of FV on
day 0. Group I (solid lines): BALB/c, (BALB/c x BALB.B)F1,
BALB.5R, and (BALB.5R x BALB.B)F1. Group II (long dashes):
BALB.G and (BALB.G x BALB.B)F1. Group III (short dashes):
BALB.B. See Table 2, experiment 1.
Cumulative incidences ofsplenomegaly in female mice of
nounced splenomegaly by day 24. The common feature ofthe
H-2 types of these mice was that none was homozygous for
theD region ofthe H-2b haplotype; otherwise they carried 0,
1, or 2 sets of alleles from the K and I regions of the H-2'
haplotype and 0 or 1 set from the D region of H-2b. In a
category of intermediate susceptibility, BALB.G mice and
their F1 crosses with BALB.B, all homozygous for the D
region ofH-2b but homozygous or heterozygous for genes in
the K and I regions of H-2d, showed an incidence of sple-
nomegaly ofonly 36% atday 24, although 100%6 ofthese mice
haddeveloped disease by day 38. Only theH-2b-homozygous
BALB.B mice ofthe third category showed relatively strong
resistance to FV, with a final incidence of 50%1o attained only
on day 38. In all cases the phenotypes ofF1 mice were those
of the more susceptible parent strain. Thus the ability of the
H-2b haplotype to confer resistance to FV appears to be a
recessive trait in these crosses.
Mapping Studies Using Variant H-2d Haplotypes. To study
the effects ofcertain mutant H-2d haplotypes and transgenes
available only on the C57BL background [which includes a
Susceptibility to FV splenomegaly in H-2-congenic BALB mice
Experiment 1: 10 SFFU
Experiment 2: 20 SFFU
% with splenomegaly
Strain or cross
(BALB/c X BALB.B)F
(BALB.5R x BALB.B)F1
(BALB/c x BALB.B)F1
(BALB/c-dm2 x BALB.B)F1
(BALB.G x BALB.B)F1
Female mice age 7-10 weeks, 8-15 per strain or cross, received FV i.v. on day 0. Splenomegaly was
assessed on the indicated days after infection; results are pooled for each group.
*The K and I-A regions of the H-2'5 haplotype originate from H-2b, and the I-E region is from H-2k.
77 100 100
Proc. Natl. Acad. Sci. USA 88(1991)
Proc. Natl. Acad. Sci. USA 88 (1991)
recessive allele, Fv-2r, that confers essentially complete
resistance to FV (11)], we verified that H-2b-associated
resistance to FV was also recessive in H-2b/H-2d heterozy-
gotes of the hybrid BALB x C57BL constitution. The
recessive nature of this resistance was even more pro-
nounced in these hybrids than in mice of the homozygous
BALB genetic background, an effect most likely due to the
presence of a dominant independently segregating allele
(Rfr-3r) in the CS7BL genome that augments this resistance
(12). After a dose of 20 SFFU of FV, H-2b homozygotes of
this hybrid genetic background showed 0-3% incidences of
splenomegaly in different experiments, whereas H-2biH-2d
heterozygotes showed incidences of 80-95% (Table 3, ex-
perinments 1 and 2).
In sharp contrast to the H-2d haplotype, the mutant H-2dml
haplotype (13) retained little capacity to abrogate H-2"-
associated resistance. This mutation consists of a fusion of
the two terminal class I genes oftheD region ofH-2d, Ddand
Ld (Fig. 2), such that the only remnant of the parental D
region is a class I gene encoding a product that differs in its
antigen-presenting domains from any class I molecule of the
parental haplotype (14-16). Whereas (BALB/c x B1O)F1
mice (H-2b/H-2d) showed a peak incidence of splenomegaly
of53/56 (95%) after adose of20 SFFU ofFV, only 5/99 (5%)
of (BALB.B x B10-dml)F1 mice (H_2b/H_2dml) developed
splenomegaly, an incidence comparable to that in (BALB.B
x B1O)F1 (2/63, 3%) (Table 3, experiment 1). This finding
indicates that H-2b-associated resistance to FV is not intrin-
sically recessive, but rather that one or more genes of the D
region of H-2d can suppress the phenotype.
To explore further the effect of the H-2Dd region on
H-2bassociated resistance to FV, we studied two additional
variant strains. The mutantH-2' haplotype (16), carried in
the congenic BALB/c-dm2 strain, differs from the parental
H-2d by a deletion comprising all D-region class I genes
except Dd (16, 17) (Fig. 2). In a preliminary experiment,
BALB/c, BALB/c-dm2, BALB.G, and BALB.B mice and
F1 crosses among them were tested for susceptibility to 20
SFFU ofFV. No difference was apparent between the H-2d
and H-2d?2 haplotypes in their abilities to suppress H-2
associated resistance to FV (Table 2, experiment 2). In a
second experiment in congenic mice of the hybrid 1310 x
BALB genetic background, no H-2b homozygotes developed
splenomegaly after 20 SFFU of FV, but 80% of H_2b/H_2d
heterozygotes and 72% ofH-2b/H-2d12 heterozygotes did so
different H-2 haplotypes
FV susceptibility of mice of BALB x C57BL crosses of
b/b + Dd
b/b + Dd
BALB.B x B10
BALB/c x B10
BALB.B x dm1
BALB/c x dml
BALB.B x B10
BALB/c x B10
B10 x dm2
BALB.B x D8
BALB.B x B10
BALB/c x B10
BALB.B x D8
Mice received 20 (exp. 1), 10 (exp. 2), or 7 (exp. 3) SFFU of FV
i.v. from preparation4 (exps. 1 and 2)or 5(exp. 3). Susceptible mice
had .0.5-g spleens on day 14 after virus inoculation.
theD regions ofthe H_2d, H-2dml, H-2dl?2, and H-2b haplotypes. The
figure is based on data in ref. 14.
Schematic representation of the class I genes encoded in
(Table 3, experment 2). This finding suggests that expression
of the Dd gene-the only known gene remaining in the D
region of H-2dmQ-accounts for at least a large part of the
ability of the H-2d haplotype to suppress H-2b-associated
resistance to FV.
This conclusion was confirmed in studies utilizing trans-
genic C57BL/6-Dd (D8) mice. Since these mice are homozy-
gous for both the complete native H-2b haplotype and the Dd
transgene, the role of Dd could be evaluated in the absence
of any other portion of the H-2d haplotype. In two separate
experiments in which H-2b-homozygous (BALB.B x 1310)F1
mice, which do not carry the Dd transgene, were completely
resistant to 20 SFFU of FV, (BALB.B
(H_2bl/H-2b plus Dd) showed peak splenomegaly incidences
of 31% and 26%6 (Table 3, experiments 2 and 3). Thus the Dd
gene was capable of partial suppression of resistance to FV
even in homozygous H_2b mice.
x B6-tD8)Fl mice
In H-2-congenic mice on both the inbred BALB and the
hybrid BALB x B10 genetic backgrounds, H-2b/H-2d het-
erozygotes showed the high susceptibility to FV of homozy-
gous H-2d mice ratherthan the relative resistance to the virus
of H-2b homozygotes. In studies using recombinant H-2
haplotypes, comparison of the levels of resistance seen in
BALB.B mice (Fig. 1 and Table 2, experiment 1) suggested
that, while genes in both the I(/I region and the D region of
H-2b confer some degree of resistance, the
gene(s) did so only in mice homozygous for the D-region
gene(s). This conclusion is consistent with the demonstration
that at least two H-2b-linked genes influence resistance to FV
(17, 18). The critical determinantofH-2b-relatedresistance to
FV appeared to be associated with the D region of the
haplotype, but the resistance was compromised in mice
heterozygous for theD region ofthe H-2dhaplotype.The loss
ofresistance to FV in H-2b/H-2d heterozygotes could be due
either to the intrinsic recessiveness of the H-2b-associated
trait or to the presence in the H-2d haplotype of a dominant
suppressor of the resistance.
The finding that a mutant H-2dhaplotype, H-2dmnl,has lost
the ability to suppress H-2b-associated resistance in het-
erozygotes (Table 3, experiment 1) indicates that this resis-
tance is not an intrinsically recessive trait. Rather, it appears
that the resistance was suppressed in H-2b/H-2d heterozy-
gotes mainly if not entirely by one or more genes located in
theD region ofthe H-2d haplotype. Although theD region of
x BALB.B)F1, (BALB.G x BALB.B)F1, and
9246Medical Sciences: Polsky and Lilly
the H-2b haplotype includes only one class I gene, Lb (also
called Db), five class I genes have been identified in the D
region ofthe H-2dhaplotype: Dd, D2d, D3d, D4d, and Ld(Fig.
2). The mutation in the H-2dmI haplotype consists of fusion
of the Dd and Ld genes with deletion of all intervening DNA
including the D2d, D3d, and D4d genes (19). As a result, the
sole remaining class I gene, D:Ld, is a recombinant whose
product differs from the products ofboth parental genes in its
antigen-presenting domain (14, 15); in effect, no intact gene
from the D region of the parental H-2d haplotype is present
in the H-2dml haplotype.
Although these results locate the suppressor gene(s) within
the H_2Dd region, they do not identify the gene. While it is
possible that one or more of the class I genes themselves
serve this function, the 150 kilobases (kb) of DNA that
intervene between the five class I H-2Ddgenes have not been
analyzed for potential coding sequences of other types, and
it is conceivable that the suppressor gene is located there.
The results of our studies using the H-2d?2 haplotype
indicate that at least a major portion ofthe suppressive effect
of the H-2Dd region is due to a gene(s) located in a more
narrowly defined portion oftheD region. The mutation in the
haplotype consists ofa deletion that eliminates a large
fraction oftheD region, including the class I genes D2d, D3d,
Dd, and Ld, leaving only the Dd gene and up to half of the
S50 kb of DNA between Dd and D2d (14, 20). Since we
observed a similar high level of susceptibility to FV in both
H_2b/H_2d`2 and H_2b/H_2d mice, again in contrast to highly
resistant H-2b homozygotes, it appears that either Dd or an
unidentified neighboring gene participates strongly in the
suppression of H-2b-associated resistance to the virus.
That the Ddgene itself is responsible at least in part for the
suppression ofresistance was indicated by the finding that its
presence as a transgene in mice otherwise similar to highly
resistant H-2b homozygotes conferred a significant level of
susceptibility to FV-induced splenomegaly (Table 3, exper-
iments 2 and 3). The fact that these Dd transgene carriers
were, however, markedly less susceptible than the H-2b/H-
2dti2 heterozygotes studied in the same experiment (26% vs.
72% peak incidences, Table 3, experiment 2) could be due to
any of several differences between them. Although the only
known gene from the H-2Dd region expressed in both kinds
of mice was Dd, the transgenic mice possessed two intact,
resistance-conferring H-2b haplotypes (H-2bIH-2b + Dd) in
contrast to the single one ofthe H-2b/H-2d72 heterozygotes.
In addition, the latter mice possessed an intactK and Iregion
from the H-2d haplotype. These differences in dosage of K,
I, and D region genes could have reduced the efficacy of
Dd-mediated suppression of resistance to the virus in the
transgenic mice. It is also conceivable that yet-unidentified
D-region genes of the H-2d?2 haplotype, in addition to Dd,
contributed to the suppression. On the hypothesis that theDd
gene is solely responsible for the suppression ofresistance to
FV in H-2b/H_2d heterozygotes, it would be important to
compare the levels ofresistance in H-2b/H-2dmI mice that do
or do not carry Dd as a transgene; at present both the H_2dm)
haplotype and theDd transgene are available only in C57BL-
background mouse strains, crosses of which would be ho-
mozygous for the recessive Fv-2r allele and thus absolutely
resistant to FV (11). We are currently establishing a
BALB.B-Dd transgenic strain for eventual use in this exper-
Previous studies have established that FV-specific cyto-
toxic T lymphocytes (CTL) generated in H-2b homozygotes
recognize FV antigens only in the context ofthe D-region Lb
(also called Db) product (21, 22), and it seems likely that this
fact is the basis of H-2b-associated resistance to the virus.
Since we have found no significant difference between the
levels of cell-surface expression of Lb in H-2b-homozygous
mice that do or do not carry the Dd transgene (unpublished
result), the finding that expression ofthe class I Ddmolecule
is capable of at least partially suppressing this resistance
suggests that it does so either (i) at the level of the virus-
infected cell by interfering with the antigen-presenting func-
tion of the Lb molecule or (ii) at the level of the T-cell
response by interfering with the ability of mice to produce T
lymphocytes with receptors that recognize one or more viral
epitopes presented by the Lb molecule. If the interference is
at the level ofthe antigen-presentingfunction ofLb, it is likely
that the Dd molecule competes strongly with Lb for FV
peptide epitopes at some point during'antigen processing. If
the interference is due to the failure to produce T lympho-
cytes bearing appropriate T-cell receptor molecules, it would
appear that the D
molecule, perhaps armed with some
endogenously produced peptide epitope, can mimic the Lb
molecule armed with an epitope of FV that is critical for
resistance to the virus, so that the Lb-positive host is immu-
nologically tolerant to this important epitope and thus mark-
edly less resistant to the virus.
Lb-restricted anti-FV CTL recognize at least two viral
epitopes (22), one ofwhich maps in the viral envgene (23) and
another in the gag-pol region. H-2b/H-2d heterozygotes do
generate a CTL response after immunization with FV-
infected H-2b-homozygous cells, but their response is weak
compared with that in H-2b homozygotes (D.P., unpublished
results) and is apparently inadequate to confer significant
resistance to the virus. It is possible that the generation or
effectiveness of Lb-restricted CTL specific for a particular
one of these epitopes is of overriding importance for confer-
ring resistance to the virus in H-2b homozygotes. It would
presumably be this component of the response that is sup-
pressed in the presence of the Dd molecule. Further studies
in this system will be necessary to elucidate the mechanisms
of this suppression of resistance to FV.
It is interesting to note a mirror-image symmetry between
the present findings in the FV system and resultsfrom studies
(24) of H-2-associated resistance to the murine acquired
immunodeficiency syndrome (MAIDS) induced by the LP-
BM5 virus. In these latter studies (24), homozygosity for the
H-2dhaplotype conferred relative resistance to the diseaseby
comparison with the high susceptibility in H-2b homozy-
gotes. As in the present studies, resistance appeared to be
recessive in F1 heterozygotes; the D region of the H-2d
haplotype determined resistance, and the Dd gene was re-
sponsible at least in part for the resistance, since H-2b-
homozygous D8 transgene carriers were significantly more
resistant than parental strain mice lacking the transgene.
Thus the Dd gene appears to have opposite effects in the two
sets of studies, conferring relative resistance to the MAIDS
virus and suppressing resistance to FV.
It is also noteable that studies using the Dd transgene-
carrying D8 strain have demonstrated that the transgene is a
major determinant of the phenomenon of hybrid resistance
(25, 26). Whereas classical transplantation rejection depends
on recognition by the recipient of antigens present on en-
grafted cells but absent in the host, the presence of the Dd
transgene in H-2b-homozygous recipients confers the ability
to reject hemopoietic grafts from genetically identical donors
lacking the transgene (27). In particular, Rauscher murine
leukemia virus-induced erythroleukemia cells derived from
H-2b-homozygous mice lacking the transgene were rejected
more weakly by syngeneic recipients than by otherwise
identical carriers of the Dd transgene (28). However, graft
rejection due to the presence of the Dd gene in the host
appears to be mediated by natural killer cells, while mature
T cells are fully capable of adoptively transferring H-2b-
associated resistance to FV (ref. 29 and D.P., unpublished
results). Although the coincidence oftwo unusualproperties
of the Dd molecule-the abilities to confer hybrid resistance
and to suppress resistance to a retroviral disease-is intrigu-
Proc. NatL Acad Sci. USA 88(1991)
Proc. Natl. Acad. Sci. USA 88 (1991)
ing, it is not clear how the mechanisms of the two effects
might be related.
We thank Stanley Benjamin and Lillie Lopez for technical assis-
tance and Drs. C. S. David and G. Jay for mouse breeding pairs. This
work was supported by National Institutes ofHealth Grant CA19931.
F.L. is supported in part by an American Cancer Society Research
Professorship and National Institutes ofHealth Grant P30-CA13330.
Lilly, F. & Pincus, T. (1973) Adv. Cancer Res. 17, 231-277.
Chesebro, B., Miyazawa, M. & Britt, W. J. (1990) Annu. Rev.
Immunol. 8, 477-499.
Lilly, F., Boyse, E. A. & Old, L. J. (1964) Lancet A, 1207-1209.
Lilly, F. (1968) J. Exp. Med. 127, 465-473.
McDevitt, H. 0. & Sela, M. (1965) J. Exp. Med. 122, 517-531.
Melchers, I., Rajewsky, K. & Shreffler, D. C. (1973) Eur. J.
Immunol. 3, 754-761.
Bieberich, C., Scangos, G., Tanaka, K. & Jay, G. (1986) Mol.
Cell. Biol. 6, 1339-1342.
Troxler, D. H. & Scolnick, E. M. (1978) Virology 85, 17-27.
Evans, L. H., Duesberg, P. H., Troxler, D. H. & Scolnick,
E. M. (1980) Cold Spring Harbor Symp. Quant. Biol. 44,
Axelrad, A. A. & Steeves, R. A. (1964) Virology 24, 513-518.
Lilly, F. (1970) J. Natl. Cancer Inst. 45, 163-169.
Chesebro, B. & Wehrle, K. (1979) Proc. Natl. Acad. Sci. USA
Egorov, I. K. (1967) Genetika (Moskva) 9, 136-144.
Stephan, D., Sun, H., Fischer Lindahl, K., Meyer, E., Ham-
merling, G., Hood, L. & Steinmetz, M. (1986)J. Exp. Med. 163,
Sun, H., Goodenow, R. S. & Hood, L. (1985) J. Exp. Med. 162,
Melvold, R. W. & Kohn, H. I. (1976) Immunogenetics 3,
Chesebro, B., Wehrle, K. & Stimpfling, J. (1974) J. Exp. Med.
Chesebro, B. & Wehrle, K. (1978) J. Immunol. 120, 1081-1085.
Burnside, S. S., Hunt, P., Ozato, K. & Sears, D. (1984) Proc.
Natl. Acad. Sci. USA 81, 5204-5208.
Hansen, T. H., Cullen, S. E., Melvold, R., Kohn, H. I., Fla-
herty, L. & Sachs, D. H. (1977) J. Exp. Med. 145, 1550-1558.
Plata, F. & Lilly, F. (1979) J. Exp. Med. 150, 1174-1186.
Holt, C. A., Osorio, K. & Lilly, F. (1986) J. Exp. Med. 164,
Ruan, K.-S. & Lilly, F. (1991) Virology 181, 91-100.
Masahiko, M., Morse, H. C., III, Fredrickson, T. N. & Hart-
ley, J. W. (1990) J. Immunol. 144, 4347-4355.
Cudkowicz, G. & Bennett, M. (1971) J. Exp. Med. 134, 82-102.
Bennett, M. (1987) Adv. Immunol. 41, 333-445.
Ohldn, C., Kling, G., Hoglund, P., Hansson, M., Scangos, G.,
Bieberich, C., Jay, G. & Karre, K. (1989) Science 246,666-668.
Hoglund, P., Ljunggren, H.-G., Ohlen, C., Ahrlund-Richter,
L., Scangos, G., Bieberich, C., Jay, G., Klein, G. & Kdrre, K.
(1988) J. Exp. Med. 168, 1469-1474.
Johnson, C. S., Thurlow, S. M. & Furmanski, P. (1986) Cancer
Res. 46, 183-189.