close

Вход

Забыли?

вход по аккаунту

?

944

код для вставкиСкачать
Int. J. Cancer: 65,221-229 (1996)
0 1996 Wiley-Liss, Inc.
Publication of the InternationalUnion Against Cancer
Publication de I'Union Internationale Contre le Cancer
THE MURINE Fc-GAMMA ( F c ~ )RECEPTOR TYPE I1
B1 IS A TUMORIGENICITY-ENHANCING FACTOR
IN POLYOMA-VIRUS-TRANSFORMED 3T3 CELLS
Tal zUSMAN'33, Ofra GOHAR',Ilan ELIASSI',Yechiam AVIVI',Ellen LISANSKY',
Catherine SAUTES~,
Jos EVEN^,
Christian BONNEROT~,
Wolf H. FRIDMAN~,
Isaac P. WITZ' and Maya RAN'
'Department of Cell Research and Immunology, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv, Israel;
and 21NSERA4U 255, Immunologie Cellulaire et Clinique, Institut Curie, Paris, France.
The murine receptor for the Fc portion of IgG is a molecule
expressed by cells of the immune system. This study suggests
the hypothesis that Fcy receptor type II B I (FcyRIIB I) functions
as a progression-enhancingfactor when expressed ectopically
on non-lymphoid tumor cells. It has been shown previously that
BALB/c 3T3 cells transformed in vitro with polyoma virus (PyV)
do not express FcyRll but acquire the expression of this
receptor following an in vivo passage in syngeneic mice. The
specific FcyRII transcript present in tumor cells was identified
in this report as FcyRllBl (BI). In order to determine whether
or not the ectopically expressed FcyRll plays a role in the
progression of these transformed cells, PyV-transformed 3T3
cells were transfected with B I-cDNA. The B I transfected cells
were tested for their ability to form local tumors in syngeneic
mice, as compared to transfected cells which express the
co-transfecting neomycine resistance (neoEs) DNA alone or
together with the IacZ gene. FcyRllBl expressors exhibited a
significantly higher tumorigenic phenotype than FcR-negative
controls, though both types of cells exhibited the same growth
curve in vitro. The ability of FcyRllBl to act as a potentially
tumorigenicity-enhancingfactor was also demonstratedas FcyRll
was expressed by tumor cells, originating from inoculated
FcyRllB Idransfected cells, or from inoculation of a mixture of
receptor-positive and -negative cells. B I-expressing cells dominated the tumor-cell population over non-expressors. This
dominance strengthened the hypothesis that FcR plays a role in
tumor progression in vivo.
o 1996 Wiley-Liss,Inc.
Tumor progression is a process involving multistep genetic
and epigenetic changes such as alterations in oncogenes and in
tumor-suppressor genes (Fearon and Vogelstein, 1990; Levine,
1993). However, progression is also enhanced by other factors,
for example, normal gene products that by themselves are
unable to transform cells, but when expressed or overexpressed by transformed cells show an increased malignant
phenotype (Herrlich et al., 1993). We suggest that FcyRIIBl is
such a factor.
The murine low-affinity receptor for the Fc portion of IgG,
FcyRII, is widely distributed on various cells of the immune
system and fulfils several immunological functions such as
regulation of Ig isotype synthesis (Fridman et al., 1992).
FcyRII gene products are the alternatively spliced FcyRIIBl
(Bl) and FcyRIIB2 (B2) transcripts (Ravetch et al., 1986;
Hogarth et al., 1991). The FcyRIIBl isoform contains a
specific intracellular insertion of 47 amino acids (exon 8)
(Fridman et al., 1992; Ravetch et al., 1986; Hogarth et al.,
1991). Both isoforms bind aggregated IgG of the IgG1,2a and
2b isotypes (Fridman et al., 1992) but differ in other ways.
While the B2 is expressed on macrophages in addition to the
low-affinity FcyRIII, B1 is the only isoform expressed on B
lymphocytes (Fridman et al., 1992). Transfection experiments
demonstrated that both the B1 and B2 transcripts are capable
of modulating surface-immunoglobulin-triggered B-cell activation. This required the intracellular domains common to both
isoforms. The B1-specific exon 8 is involved in the formation of
caps in response to ligand binding, resulting in its association
with the cytoskeleton (Amigorena et al., 1992) or to its
alignment with actin filaments (Miettinen et; al., 1992). Further-
more, since B2 functions in endocytosis and B1 does not, it was
tentatively concluded that the B1-specific exon 8 inhibits
endocytosis (Miettinen et al., 1992).
Previous work showed that FcyR is expressed by a subpopulation of an in vivo-passaged anaplastic carcinoma originally
induced by polyoma virus (PyV) (Ran et al., 1988). Ectopic
expression on these non-lymphoid cells was maintained by
tumors growing in vivo, and disappeared after explantation of
the tumor cells to culture (Ran et al., 1988).
Subsequent studies using another experimental model of
tumors that developed from BALB/c 3T3 cells transformed in
vitro with PyV confirmed these results (Ran et al., 1991). By
utilizing molecular and immunological probes, the receptor
was identified as the low-affinity Fcy RII (Ran et al., 1991). In
these studies, the type of transcript expressed by the tumor
cells was not identified.
It was further demonstrated that irz vitro transformation per
se was insufficient for the induction of FcyR expression by
these cells. Only transformed cells that had been passaged in
syngeneic animals as solid tumors acquired FcyRII expression.
This was true for several different cloned lines which were
originally derived from a parental PyV-transformed cell line
(Ran et al., 1991). One of these clones was used in the present
study.
The above experimental model demonstrated that host
factors, involved in the induction of receptor expression (Ran
et al., 1991), were also essential for the maintenance of FcyR
expression. As seen before (Ran et al., 1988), expression
decreased gradually when tumor-derived cells were explanted
and grown in culture (Ran et al., 1991). When explanted cells
were inoculated into mice forming second-passage tumors,
expression of FcyR occurred once more (Ran et al., 1991).
The ectopic expression of FcyR is one of several characteristics that may be induced in tumor cells by host factors (Witz
and Ran, 1992). A murine in vitrolin vivo progression model
system was developed several years ago in this laboratory,
making it possible to identify and characterize gene products
of transformed cells involved in tumor progression and controlled by host factors (Halachmi and Witz, 1989; Ran et ab,
1991). In this model system, BALB/c 3T3 cells transformed in
vitro with polyoma virus (PyV) were cloned. These clones were
either maintained in culture (termed C for cultured) or
passaged once in vivo and then returned to culture (termed
CTC for culture-tumor-culture). The most prominent of these
characteristics is a higher tumorigenic phenotype expressed by
in vivo-passaged CTC cells compared to that expressed by their
C clonal ancestors (Halachmi and Witz, 1989). Since FcyRIIexpressing CTC cells tended to exhibit a higher malignant
phenotype than Fcy RII-negative CTC cells (Ben-Baruch/
3T0 whom correspondence and reprint requests should be addressed, at the Department of Cell Research and Immunology,
George S. Wise Faculty of Life Sciences, Tel Aviv University, 69978
Tel Aviv, Israel. Fax: 972 3 6422046.
Received: June 13,1995 and in revised form September 11,1995.
ZUSMAN ETAL
222
Langer et al., 1992), we hypothesized that FcyRII expression
may confer upon certain tumor cells a high tumorigenic
phenotype (Ran and Witz, 1985; Witz and Ran, 1985,1992).
In the present study we identified the specific transcript
expressed in PyV-transformed CTC cells as FcyRIIB1. In
order to evaluate the effect of FcyRIIBl on the tumorigenic
phenotype of tumor cells, cDNA for this particular transcript
was transfected into FcyRIIBl-negative C cells. Utilizing
FcyRIIBl transfectants, we provide direct evidence that the
B1 receptor is causally involved in augmenting the malignant
phenotype of tumor cells.
(Ran et al., 1991) using a Becton Dickinson (Mountain View,
CA) FACS analyzer. As second antibody either phycoerythrinelabeled goat anti-rat IgG antibody or fluorescein-isothiocyanate-labelled goat anti-mouse IgG antibody were used (Ran et
al., 1991). The macrophage cell line 5774 (Unkeles, 1979)
served as a positive control for FcyRII expression. Ligand
binding was detected using heat-aggregated IgG; for this
purpose 10mg/ml of mouse IgG (Jackson, West Grove, PA)
were heat-aggregated for 15 rnin at 65°C. These conditions led
to a 45% aggregation, as determined by optical density
measurements of the supernatant after 60 rnin of 100,000 g
precipitation.
MATERIAL AND METHODS
Growth curve
Growth was measured at various time intervals by MTT
colorimetric assay, measuring a mitochondria1 enzyme that
correlates with cell viability (He0 et al., 1990). The average OD
of 12 replicates was calculated.
Mice
BALB/c mice (H-2d) were bred and maintained in the
animal quarters of Tel Aviv University. In all experiments,
8-week-old female mice were used.
Cell lines
The 3T3-TSC polyoma-virus transformed cell line was
kindly provided by Dr. M. Fried (Imperial Cancer Research
Fund, London). This line was initiated from a BALB/c-3T3
cell line transformed in vitro by an incomplete viral genome of
a temperature-sensitive mutant of polyoma virus (PyV). This
mutation did not affect the expression of early antigens (Barra
et al., 1983). The PyV-transformed line was cloned and several
clones were propagated in culture. These were designated C
(cultured) cells. One of these C-cell cultures, the A9C clone
(Ran et al., 1991; Halachmi and Witz, 1989), was used in this
study for transfection experiments. Also utilized were FcyRIIpositive A9 CTC 220 cells which were derived from a tumor
formed 220 days after the inoculation of A9C cells (Ran et al.,
1991). The A9 CTC 220 clone was one of several clones that
acquired the expression of FcyR after one in vivo passage. All
cell lines were maintained in DMEM (Maagar, Beit Haemek,
Israel) supplemented with glutamin, antibiotics and 10% FCS
(Maagar).
DNA transfection
Expression vectors used were: FcyRIIBl cDNA-containing
pKC3 vector (Fridman et al., 1992) and the selection vector
pSV2neo (Maniatis et al., 1983). The calcium-phosphatemediated DNA transfection method (Maniatis et al., 1983) was
employed by mixing O.lml of O.lmg/ml DNA solution (pSV2neo
and B1 cDNA in pKC3 at a ratio of 1 to 10) with 0.06ml of 2M
CaC12. This mixture was added dropwise to a 0.84 ml of HBS
solution (140mM NaC1, 5mM KCI, 0.75mM Na2HP04,25mM
HEPES and 6mM dextrose) with continuous agitation. After
20 rnin at room temperature the mixture was added to lo6 A9C
cells which were incubated at 37°C for 4 hr in the presence of
5% FCS on 90-mm diameter dishes (Tissue Culture Miniplast,
Ein Shemer, Israel). For neomycine selection, cells were
incubated overnight and then fed twice weekly with DMEM
and 10% FCS containing lmg G418 (Sigma, Holon, Israel) per
ml.
The expression pattern of the 2.4G2 FcyRII epitope on
B1-transfected cells usually indicated a homogeneous population as tested by flow cytometry. When a non-homogeneous
pattern was observed, the culture was sub-cloned to achieve
homogeneity.
Stock cultures of each transfectant line were kept at -75°C.
Freshly thawed cells were maintained in culture for 1-month
periods and then discarded, following which another aliquot of
frozen cells was cultured.
Flow cytornetry analysis
The expression of the FcyRII epitope recognized by the
2.4G2 MAb (Unkeles, 1979) was revealed by flow cytometry
Turnorigenicity in vivo
Tumorigenicity of transformed cells was determined as
described previously (Ran et al., 1991; Halachmi and Witz,
1989). Control and transfected cells were trypsinized and
washed 3 times in cold DMEM. A dose of lo5cells in a volume
of 0.2ml per mouse was injected i.m. into a hind leg of
8-week-old syngeneic female BALB/c mice (12 mice per
group). The time of appearance of progressing tumors was
recorded by palpation. (No spontaneous tumor regressions
were observed). Histopathology confirmed the malignant nature of the tumors which were characterized as fibrosarcomas.
The statistical evaluation of differences between tumor appearance in the control and experimental mice was done by the
log-rank test (Peto et al., 1977). The difference in tumor
incidence or in median tumor latency between B1 and control
lines of all experiments was calculated according to Student's
t-test.
Northern-blot analysis
Total RNA was extracted by the guanidine-thiocyanate
method (Maniatis et al., 1983), then electrophoresed through
1% agarose-formaldehyde gel and transferred on to Hybond
N+ filters as described (Bonnerot et al., 1990). Probes labeled
with 32Pby the random priming method were used to detect
transcripts. The FcyRII-specific probe was a BclI-PstI fragment of FcyRIIBl cDNA (Bonnerot et al., 1990). A BamHI
fragment of rRNA genomic clone was used as a probe to detect
28s rRNA. Filters were washed under stringent conditions
(0.1 X SSC, 0.1% SDS, 65"C), then autoradiographed on XAR
films (Kodak, Rochester, NY) at -70°C.
RT-PCR
Extraction of total RNA was carried out from about lo7cells
by RNazol B kit (Cinna Biotecx, Houston, TX) following the
manufacturer's instructions. First-strand cDNA synthesis was
performed using an oligo dT primer. Two micrograms of RNA
were reverse transcribed for 60 min at 42°C by 50 units of
either AMV reverse transcriptase (Boehringer-Mannheim,
Indianapolis, IN) or M-MLV reverse transcriptase (Perkin
Elmer Cetus, Nonvalk, CT) for experiments recorded in
Figure 1 or Figure 2 respectively. Subsequent amplification
was carried out by polymerase chain reaction with 2.5 units of
Ampli TAQ DNA polymerase (Perkin Elmer Cetus) for 35
cycles (1 rnin at 95°C; 3 rnin at 60°C).
The 2 oligonucleotides used as PCR primers for the detection of FcyRII expression were identical to those used previously (Bonnerot et al., 1992). They were designed to amplify a
fragment from the 3' region of the published sequence of
FcyRIIBl (Ravetch et al., 1986) covering the transmembrane
and cytoplasmic domains and 3' untranslated region. This
FcrRIIBl IS A TUMORIGENICITY-ENHANCINGFACTOR
223
RESULTS
FIGURE1 - Characterization of the specific FqRII transcript
expressed by PyV-transformed tumor-derived cells. fa) Northernblot analysis using total RNA, and a BclI-PstI 32P-labeledFcyRIIB
probe. (1) A9CTC-220 tumor-derived cells; (2) A9C culturederived cells; (3) P388Dl-FcyRIIB2-expressingcells (Ravetch el
al., 1986); (4) A20/2J-FcyRIIB1-expressing
cells (Bonnerot et al.,
1990). (b) Northern-blot analysis of the amount of RNA is shown
by using 32P-labeled28s-specific probe. (1-4) are as in fa). (c)
Analysis of FcyRIIB expression by RT-PCR as described in
Material and Methods: (1) Hind 111 digest of pSVKl (Dorsett et
al., 1985), used as marker of molecular weights; (2) control of the
RT-PCR reaction; (3) A9-CTC-220 tumor-derived cells (as in a,
1); (4) A20/2J-FcyRIIBl-expressingcells (as in a, 4); (5) P388D1FcyRIIB2-expressingcells (as in a, 3).
FIGURE2 - FqRIIBl expression in B1-transfected cells. RTPCR analysis (performed as in Figure lc): (1) A9-CTC-220
tumor-derivedcells; (2-3) Controls of the RT-PCR reaction; (4-8
B1-transfected cells; (4) H513; ( 5 ) E2/3; (6) 44E; (7) 1F/4; (8
02Bll; (9) Hae I11 digest of +x174 used as a marker of molecular
weights.
1
produced DNA fragments of 550bp when FcyRIIBl was
expressed and of 400bp when FcyRIIB2 was expressed. The
ethidium-bromide-stained PCR products were analyzed on a
2% agarose gel.
Identification of Fc’cyRIIBlas the receptor acquired
in vivo by tumor cells
FcyRII is encoded by a single gene. As a result of alternative
splicing, 2 transcripts, FcyRIIBl and FcyRIIB2, are formed
(Ravetch et al., 1986; Hogarth et al., 1991). This leads in turn to
the formation of 2 variant receptor proteins that differ in their
intracellular portion. FcyRIIB2 lacks the 47 amino acids coded
for by exon 8 which is present in FcyRIIBl (Ravetch et al.,
1986; Hogarth et al., 1991).
By using antibodies recognizing the 2.462 epitope of both
FcyRII transcripts (Ravetch et al., 1986; Hogarth et al., 1991)
we found that tumor cells derived from several clones transformed in vitro with PyV acquired during their in vivo residence
period the expression of this receptor (Ran et al., 1991).
Receptor expression was gradually lost as a function of time
following explantation of tumor cells to culture conditions. A9
CTC 220 tumor-derived cells, maintaining receptor expression
for a relatively longer period following explantation compared
to other clones, were chosen for further characterization of the
receptor.
In order to identify which of the transcripts of FcyRII is
expressed by these tumor cells a Northern hybridization
analysis was performed for total RNA of tumor cells. A
BclI-PstI fragment comprising the intracellular portion and
the 3’UTR sequences of the FcyRIIBl cDNA was used as a
specific probe. Since the 2 FcyRII transcripts B1 and B2 are
highly homologous, the difference between them was determined by comparing their migration distance with reference to
control cell lines that express either B1 or B2. A message of the
size of B1 was detected in the tumor cells (Fig. la).
However, the level of mRNA for B1 was very low as
compared to the hybridization pattern with the 28s probe (Fig.
1b). These low B1 mRNA levels correlated with low protein
FcyRII levels as revealed by flow cytometric analysis (data not
shown). Therefore, an RT-PCR amplification was additionally
carried out using FcyRII-specific primers (Bonnerot et al.,
1992). The results obtained by this method (Fig. lc) confirmed
the Northern analysis. The amplified fragment of the transcript in tumor cells was about 550bp long, similar to that
detected for the transcript expressed by B1-positive control
cells, the B-cell lymphoma A20/2J (Bonnerot et aL, 1990).
BZexpressing control cells, the macrophage-like cell line
p388D1 (Bonnerot et al., 1990), had a shorter fragment of
about 400bp in length. These results indicate that the A9 CTC
220 cells acquired, during their in vivo residence period, the
expression of FcyRIIBl rather than of FcyRIIB2.
Stable transfection of FcyRlIBl cDNA into qtV-transformed
cells confers the expression of a functional receptor
In previous studies, we hypothesized that the ectopic expression of FcyRII on non-lymphoid tumor cells may facilitate and
enhance their in vivo development and malignant behavior
(Witz and Ran, 1992). Since the transcript acquired in vivo by
tumor cells was identified as FcyRIIBl (Bl), stable B1
transfectants of PyV-transformed A9C cells were generated.
These cells, like other in vitro-transformed clones, did not
express this receptor prior to transfection (Ran et aL, 1991).
The pKC3 expression vector containing B1 cDNA was cotransfected with the pSV2neo vector containing the neomycineresistance gene into the transformed cells. Three independent
transfection experiments were performed. G418-resistant transfectants were screened for FcyRII expression using flow
cytometry, and about 40-50% of these transfectants were
found to express various levels of the receptor. Each transfectant was characterized at the RNA level to confirm the
expression of B1 message. A typical RT-PCR analysis of
transfectants from these experiments is shown in Figure 2. For
ZUSMAN ETAL
224
Tumor Latency
Incidence
(median)
p<a.ai
S
days
l
80
-
60 -
40
W
~
$
20 -
-1
W
a
O
I
L
FCR+
0
100
10'
102
103
FLUORESCENCE
FIGURE3-Flow cytometric analysis of antibody and ligand
binding by FqRIIB1-transfected cells. (a) FcyRIIB1-transfected
cells (A9-1) were exposed to the 2.462 anti-mouse FcyRII MAb
(----),or to mouse aggregated IgG (....) or to diluent alone (-).
For details see text. (b) Control cells (AY-2 transfected with the
mores
gene only) undergoing the same treatment.
control cell lines, the same transfection procedure was applied
using either pSV2neo alone or a co-transfection with a nonrelevant plasmid containinglac2 gene (pSV2pgal). The FcyRIInegative pSV2pgal control transfectants were stained with
chromogenic substrate (Kerr et al., 1989) X-gal (data not
shown).
The functionality of B1 on the transfected cells was assessed
by flow cytometry using heat-aggregated mouse IgG as a
ligand. All transfected cell lines bound the ligand whereas
control transfectants did not. Additional flow cytometric analyses with the 2.462 MAb indicated that the transfected receptors were also antigenically intact. A typical cytometric analysis
of a B1-transfected line with either the 2.462 antibody o r the
mouse-aggregated IgG ligand is shown in Figure 3.
The tumorigenic phenotypes of transfected cells are reported in the following section. Regarding the level of expression, neither a positive nor a negative correlation could be
shown between the level of expression of the 2.4G2 epitope of
the transfectants and the tumorigenic phenotype. For example, transfectants expressing very low levels of membrane
FcyRIIBl or very high expression manifested a very high
tumorigenic phenotype. B1 transfectants expressed various
levels of expression that were equal to, higher or lower than
expression levels on cells acquiring expression by in vivo
passage (Ran et al., 1991).
FcyRIIBl -transfected Py V-transfomed cells express
an augmented tumorigenic phenotype
B1-expressing transfectants, derived from 3 independent
transfection series, were tested for their ability to form local
tumors when injected i.m. into syngeneic BALB/c mice. The
tumorigenicity of each clone was tested repeatedly 2-5 times,
by inoculating lo5cells per mouse. A dose of lo5cells, resulting
in about 80% incidence was selected by pilot experiments in
n
Controls
FCR+
Controls
FIGURE4 - Tumorigenicity of B1 or control transfectants in
BALB/c mice. Summarized data. Tumor incidence (%) and
median tumor latency (days) were calculated for the mouse groups
whose individual tumorigenicity data are shown in Table I. Final
tumor incidence was recorded 60 days after tumor incidence
reached maximal levels. The tumor incidence and median latency
of each of the individual inoculated groups are presented. Both
parameters-tumor incidence and tumor median-differed significantly between B1 and controls (j< 0.001, as determined by
Student's t-test).
which a dose response of B1 transfectants was determined.
Each in vivo experiment included a group of 12 mice inoculated with B1 transfectants as well as 12 mice inoculated with
FcyRII-negative control transfectants. Control transfectants
were either neorestransfected AYC, or A9C co-transfected with
pgal and neores.Occasionally, the same control cells served for
several experiments.
Figure 4 and Table I summarize results of tumorigenicity
experiments in which 8 different B1-transfectants and 6
different control transfectants, randomly paired, were inoculated into mice. Since the same control transfectant was paired
with more than one B1-transfectant, altogether there were 18
B1 experimental groups and 12 control groups. Data for
tumorigenicity of individual pairs are shown in Table I.
Tumorigenicity was analyzed by 2 parameters: (1) tumor
incidence, (2) median of the latency period needed to attain up
to 50% of tumor incidence. When maximal values of tumor
incidence were reached they did not usually change during at
least 60 days of follow-up. According to statistical analysis
(log-rank test) that takes into account both parameters, all 18
B1-expressing transfectants manifested a higher tumorigenic
phenotype than the control FcR-negative cells, as can be seen
in Table I. In 12 out of 18 experiments the differences were
statistically significant (p < 0.05 to p < 0.001). One experiment out of 18 could not be analyzed statistically because only
B1-transfectants produced tumors while control transfectants
did not. In 4 experiments out of 18, B1 exhibited a higher
tumorigenicity but the differences were not statistically significant.
Figure 4 provides combined tumor incidence and latency
data of the 18 B1 experimental groups and the 12 control
groups whose individual tumorigenic profile was tested. The
incidence in the B1 groups (X = 76%) was significantly higher
@ < 0.01) than that of the control groups (X = 36%). The
difference in median latency between B1 transfectants and
control cells was also highly significant. The median tumor
FcyRIIBl IS A TUMORIGENICITY-ENHANCING FACTOR
TABLE I - TUMORIGENICITYEXPERIMENTS OF B1 TRANSFECTANTS,
EACH COMPARED WITH A CONTROL TRANSFECTANT'
~
~
Clone
~
ment
1 A9-1
A9-2
02Bll
P2D6
02Bll
la12
02Bll
P2D6
1Fl4
N4H4
6 1016
la12
7 IF14
Z2H5
8 1H18
Z2H5
44E
Z2H5
10 E2l3
Z2H5
11 H513
Z2H5
12 1F/4
Z2H5
13 1H18
Z4H2
~
~1
~
Incidence
j
. %
expression B I
75
75
Control
50
10
Latency (days)*
BI
80-160 <0.05
15-90
100
75
95
80
100
85
10
60
10
90
55
'P
50-100
35-60
100
Control
30-90
40-65
85-105
110
50-1 00 < 0.001
50
<0.01
65-11
<O.O1
100
<0.001
70-145
<o'ool
45-80
50-65
<0.01
225
genic than the latter, supporting the results presented in
Figure 4.
Although B1 transfectants expressed an augmented in vivo
tumorigenic phenotype compared to control transfectants, this
in vivo advantage was not observed when the in vitro growth
curves of the 2 types of cells were compared. Figure 6 shows
the growth curves of pooled B1 transfectants and pooled
control transfectants, in complete as well as serum-free media.
Similar results were obtained with individual transfectants
(data not shown). Both B1 and control transfectants grew
more slowly in serum-free medium than in complete medium.
The cell number of both types of transfectants increased
between 48 and 72 hr in culture (Fig. 6). There was no growth
advantage in vitro for the highly tumorigenic B1-transfected
cells.
Based on the above results, we conclude that the expression
of FcyRII confers upon BALB/c cells transformed in vitro with
PyV a higher malignant phenotype than that of the same cells
which do not express this receptor.
125-200 <0'005
Tumors derived from a mkfure of Fc yRlIBl -expressing
and non-expressing transfectants are dominated by FcR
expressors
60
50-145
If B1-expressing tumor cells enjoy a growth advantage in vivo
55
125-200 NS
over B1-negative cells, the former cells should become domi70
80-110
nant in tumors formed by the inoculation of a mixture
55
125-200 N'S.
composed of B1-positive and -negative cells. In order to test
95
-4
55-75
this hypothesis, a mixture of a FcR-positive and -negative
0
transfectants was inoculated into syngeneic mice. Each clone
45-90
95
115
<0.001
was also inoculated separately.
20
Figure 7 shows the kinetics of tumor formation by B13&95
55
115
<0.025
positive transfectants alone, FcR-negative transfectants alone
l4 % I 2
20
or by a 1:l mixture of both cells. The B1-positive and -negative
E2/3
45
N.S.
cells manifested an expected kinetics of tumor appearance, i.e.
115
l5 Z2H5
20 130-160
high and low tumorigenicity respectively (p < 0.05 as tested by
75-130
85
16 H513
115 <0.001
log-rank test (Peto et al., 1977). The tumors formed by the
20
Z4H2
mixture of these cells showed a kinetics pattern similar to that
10-70
17 A9-1
74
of the B1-positive cells and different from that of the negative
60-90 < 0.02
30
Z2H5
cells. Although the tumorigenicity of the mixed population was
1Fl4
85
40-60
non-significantly different from that of both lines separately,
108
<0.001
10
l8 Z4H2
the similarity of the tumorigenic pattern of the mixture to that
'Eighteen individual experiments were performed in which of the B1 positive lines is evident. For example, the median
BALB/c syngeneic mice were inoculated i.m. with a B1 transfec- latency period of the mixture was 50 days, that of the
tant versus a control transfectant [transfected either with neores B1-expressing line was 35 days and that of the control cells was
alone (experiments 1-6) or together with the lac2 gene (experi- 140 days.
ments 7-18)]. Tumorigenicity was determined as described in the
We then approached the dominance hypothesis by testing
text.JThe period between the appearance of the first and the last
palpable t~mor.-~Calculatedby the log-rank test (see text).sNo B1 expression on tumor-derived cells. FcyRII expression was
statistical analysis was performed since no tumors appeared in the assayed on explanted tumor cells derived from B1-expressing
control group.
transfectants, from B1-negative control transfectants and from
the 1:l mixture. These explanted tumor cells had been grown
in culture for at least 10 in vitro passages prior to analysis to
latency period of B1 transfectants (83 days) was significantly ensure infiltrate depletion. Figure 8 shows a flow cytometric
analysis of FcyRII expression of these 3 cell populations
shorter than that of control transfectants (129 days).
before
and after inoculation. FcyRII expression by in vivoB1 transfectants and control transfectants may vary in their
passaged FcyR-positive or -negative transfectants was similar
tumorigenic profile in different experiments. This is in accor- to that seen prior to inoculation. In contrast, tumors formed by
dance with results of Nicolson et al. (1992) who demonstrated the mixture show a FcyR expression pattern completely
that the transfection procedure itself may cause a rapid different from that of the in vitro-formed mixture prior to
stimulation of clonal diversification. In order to neutralize the inoculation. While, as expected, the in vitro-maintained mixeffect of clonal or biological variation between the different ture contains 2 clearly distinctive subpopulations of B1transfected lines, the strategy of pooling individual transfec- positive and -negative cells, the in vivo population derived from
tants, used previously by others (Theodorescu et al., 1990; Kim this mixture is clearly dominated by B1-positive cells. Similar
and Cohen, 1994), was employed in this study. The tumori- results were obtained in another experiment, where B1genic phenotype of pooled lines of B1 transfectants was positive transfectants were inoculated in a mixture with control
compared to that of pooled control transfectants. Figure 5 negative transfectants, containing the lacZ gene. Again, the
shows the tumorigenic profile of a pool of 14 different B1 dominance of FcyRII-positive cells was seen, while no detecttransfectants, in comparison with a pool of 14 different control able lacZ-expressing cells were found in tumors arising from
transfectants. The former pool was significantly more tumori- the mixture (data not shown).
25
55
90-135
125-200
NS'
ZUSMAN ETAL.
226
100 I
80
-
n
$
Q)
60
lo0
0
z:
10‘
id
1d
Fluorescence
Q)
20
40
20
p<o.o1
.
+
H
-!JP
I
II
I
40
60
80
Days after inoculation
01
20
I
B1 M ,
100
Cells.
Control 5-U
FIGURE5 - Kinetics of tumor appearance in BALB/c mice inoculated i.m. with a pool of B1 versus a pool of control transfectants.
Fourteen B1 transfectants and 14 control transfectants, from 3 separate transfection series, were each pooled and inoculated into a
group of 15 BALBic mice. Expression of the 2.462 epitope by the pooled B1 transfectants was ascertained (see insert). Pooled B1 lines
showed an enhanced tumorigenicity compared to pooled controls (p < 0.01, as determined by log-rank test).
6
4
n
8
v
2
80-
W
60-
4
8
40-
2
z
z
W
0
0
24
48
72
P
I
0
20 -
0
I
I
I
I
I
I
I
TIME (Hours)
FIGURE
6 - Growth rate of pooled B1 transfectants (0-0)
and
control transfectants (m-B) as measured by MTT colorimetry.
Results are represented as the ratio of the average OD value at
indicated time versus the average OD value at time zero. Cells
were grown in (1)medium containing 10% FCS or (2) serum-free
medium. 0,Pool pl; M, pool LacZ.
DISCUSSION
The role of alterations in oncogenes and in tumor-supressor
genes with respect to turnorigenesis and tumor progression is
well established (Fearon and Vogelstein, 1990; Levine, 1993).
It seems, however, that a third set of factors may also be
involved in these processes. These “progression-enhancing
factors” are normal gene products expressed by certain normal
cells. By themselves, progression factors are unable to transform a cell. However, when expressed or overexpressed by
transformed cells, they may drive such cells towards increased
malignancy even when these gene products are genetically
unaltered. This could also happen when such molecules are
ectopically expressed by transformed cells from a different
lineage than the one normally expressing this molecule.
Certain splice variants of CD44 (Herrlich et al., 1993) are
appropriate examples of such progression-enhancing factors.
The results of this study prove that murine FcyRIIBl, when
ectopically expressed on non-lymphoid tumor cells, functions
as a progression-enhancing factor. The hypothesis that Fcy RII
may function as a progression factor (Witz and Ran, 1992) was
FcvRIIBl IS A TUMORIGENICITY-ENHANCING FACTOR
in vitro
A
1
6
lo0
B
ex vivo
0
101
227
~
)
ld
102
1tP
10'
102
id
.
150
0
lo0
C
101
ld
ld
150
lo0
101
lo0
10'
102
150
0
0
lo0
101
ld
ld
FLUORESCENCE
FIGURE8 - Expression of the FcyR epitope 2.4G2 by tumor-derived B1-transfected cells mixed with control transfectants. In
vitro-maintained transfectants or cells derived from tumors induced by transfectants ( a vivo) were analyzed for FcyRII expression using
the 2.462 anti-mouse FcyRII MAb (---),control diluent alone (-),
and secondary FITC-labeled antibody: (a) A 1:l mixture of B1
and control cells before inoculation (in vitro) or after tumor excision (avivo). (b) B1-transfected cells before inoculation and after tumor
cells before inoculation and after tumor excision.
excision. (c). Control (neoTes)-transfected
based upon previous studies demonstrating that, whereas
BALB/c 3T3 cells transformed in vitro with PyV do not express
FcyRII, expression of this receptor was acquired when the
transformed cells were passaged in vivo as tumors in syngeneic
mice (Ran et aZ., 1991). The in vivo passage also considerably
increased the tumorigenic phenotype of the cells (Ben-Baruch
Langer et al., 1992). These results demonstrated that FcyRII
expression was induced by in vivo operating factors and
suggested a possible cause-and-effect relationship between the
increased tumorigenicity of in vivo-passaged cells and the
acquisition of FcyRII expression by these cells. The present
study determined the identity of the specific receptor involved
in this phenomenon as FcyRIIB1. Stable transfection of this
receptor into non-expressing cells performed in this study
established that FcyRIIBl indeed conferred upon PyVtransformed cells a highly tumorigenic phenotype. This was
shown by comparing the tumorigenicity of individual B1 with
that of control transfectants (Table I, Fig. 4) as well as that of
pooled ones (Fig. 5). In spite of the clonal variation which was
apparent in tumorigenicity of transfectants tested individually,
the overall effect of B1 transfection as tumor-enhancing factor
has been determined by both the increase in tumor incidence
and the shortening of tumor latency.
The transfection of FcyRIIBl did not change the antigenic
expression of a PyV-related surface antigen (Ben-Baruch
Langer et aZ., 1992) as well as that of 2 MHC class-I antigens
tested (data not shown). Therefore, and based on in vivo
tumorigenicity results described herein, augmented malignant
phenotype of transfectants can be referred to as a direct result
of FcyRIIBl expression. The dominance of FcyRIIBl expressors over FcR-negative cells when inoculated in a mixture
strengthens the idea that FcyRIIB1-expressing tumor cells
enjoy an in vivo advantage.
Although the function of FcyRII as a tumor-enhancing
factor was not investigated in many other tumor systems other
228
ZUSMAN ETAL.
than PyV, two facts should be noted. The first is that FcyRIIBl
conferred a highly tumorigenic phenotype upon cells transformed by an entirely different mode of transformation. Stable
transfectants originating from c-H-ras-transformed NIH 3T3
cells were more tumorigenic than neorescontrol transfectants
(Witz and Ran, 1992; Gonen et al., 1988, 1992). The second is
that several non-hematopoietic experimental tumors or cancers of human origin contain tumor cells expressing receptors
for the Fc of IgG (Milgrom et al., 1968; Tonder et al., 1974;
Biran et al., 1979; Noltenius, 1981) or IgE (Biran et al., 1979,
reviewed in Witz and Ran, 1992). However, reports on FcR
expression by tumor cells are relatively scarce. The probable
reason for this is that the continuous expression of the receptor
depends upon regulatory in vivo operating host factors. In the
absence of such factors, receptor expression is gradually lost
following the explantation of such cells to culture conditions
(Ran et al., 1991), except for 2 human neuroblastoma cell lines
expressing FcyR receptors (Gorini et al., 1992).
It has been demonstrated that certain splice variants of
CD44 (Herrlich et al., 1993) or FcyRIIBl (this study), being
molecules usually expressed by lymphocytes, function as progression-enhancing factors. Could similar molecules such as
IL-2 receptors, which may be ectopically expressed by certain
non-hematopoietic tumor cells (Rimoldi et al., 1993), also
function as progression-enhancing factors and be involved in
conferring a highly malignant phenotype upon tumor cells?
This question remains unanswered.
What is the mechanism by which FcyRIIBl confers a highly
tumorigenic phenotype upon transformed cells? The answer to
this question is largely unknown at the present time. However,
the information to hand indicates that, under the experimental
conditions described above, FcyRIIBl does not confer an in
vitro growth advantage upon cells expressing it, but B1expressing cells enjoy a growth advantage only under in vivo
conditions. Such a situation fits a receptor-ligand interaction
where the ligand is present only in vivo but not in vitro. Since
the most likely ligand for FcyRII is complexed IgG (Fridman et
al., 1992), preliminary experiments were performed to measure possible signals delivered by complexed IgG to FcyRIIexpressing tumor cells. Exposure of FcyRIIBl-positive transfectants to complexed IgG brought about capping of the
IgG-B1 complex (data not shown). Furthermore, incubation
of the transfectants with the 2.402 anti-FcyR MAb resulted in
a transiently up-regulated expression of the receptor (Ran et
al., 1992). These preliminary results suggest that ligandFcyRIIBl interactions may transduce signals to tumor cells
bearing these receptors. Such ligand-mediated signals could
regulate the expression of genes that may be involved in
enhancement of tumor progression.
In addition to signal-transduction-mediated mechanisms,
other mechanisms accounting for the augmented tumorigenicity of B1 expressors may operate (Witz and Ran, 1992). For
example, the expression of FcyR on tumor cells could influence interactions between such cells and components of the
immune system such as complement or ADCC-mediating cells.
FcR expression could also sterically block membrane components on the tumor cells, especially when complexed to ligand.
Finally, soluble FcyR (Fridman et al., 1992) released from
tumor cells (Billaud et al., 1989) could exert immunoregulatory
functions similar to those exerted by soluble FcyR secreted
from immunocytes under physiological conditions (Fridman et
al., 1992), eg., inhibition of IgG antibody production in vivo
(Fridman et al., 1992). Interestingly, B1 but not B2 is maintained on the cell membrane upon ligand binding, thus being
exposed for longer periods to soluble FcR-generating proteolysis. A secretion of soluble FcR by cells which are normally not
present may create an imbalance in the functions regulated by
these molecules.
Further clues to the mechanism by which FcyRIIBl exerts
its progression-enhancing effects will emerge when tumorigenicity data from tumor cells transfected with a variety of
receptor mutants become available. For example, B1 mutants
deleted in the cleavage site of proteolytic enzymes generating
soluble receptor (Fridman et al., 1992), or in the intracellular
domain responsible for association with the cytoskeleton
(Amigorena et al., 1992; Miettinen et al., 1992) will provide
information as to which portions of B1 are contributing to its
function as a tumor-progression enhancing factor. Such studies, comparing the tumorigenic phenotype of cells transfected
with wild-type B1 and with mutated forms of B1 as well as with
wild-type B2, are well in progress. The results, showing that the
B1 receptor enhances tumorigenicity much more than B2 (data
not shown), establishes the importance of the B1-specific
cytoplasmic portion for enhancement of tumorigenicity.
ACKNOWLEDGEMENTS
This work was supported by the United States-Israel Binational Science Foundation, the Israeli Ministry of Health,
Chief Scientist’s Office, the Israel Cancer Research Fund, the
Israel Cancer Association, the Ela Kodesz Institute for Research on Cancer Development and Prevention, the Fainbarg
Family Fund, Orange County, California, and the Pikovsky
Fund, Jerusalem. I.P. Witz is the incumbent of the David
Furman Chair in Immunobiology of Cancer. We thank Ms. R.
Anavi and Mr. Y. Shlomo for skillful technical assistance. The
efforts of Mr. R. Shabat in establishing and maintaining the
BALB/c colony are greatly appreciated. We thank Dr. 0.
Elroy-Stein, from the Department of Cell Research and
Immunology, for kindly providing the pSV2pgal plasmid. The
statistical advice of Ms. I. Gelernter from the Statistical
Laboratory, School of Mathematics, Tel Aviv University, is
highly appreciated.
REFERENCES
AMIGORENA,
S., BONNEROT,
C., DRAKE,J.R., CHOQUET,D., HUNZIKER,W., GUILLET,
J.-G., WEBSTER,
P., SAUTES,
C., MELLMAN,
I. and
FRIDMAN,
W.H., Cytoplasmic domain, heterogeneity and functions of
IgG Fc receptors in B lymphocytes. Science (Wash.). 256, 1808-1812
(1992).
D. and MEYER,G., Characterization of
BARRA,Y., ITO,Y., DEPETRIS,
tumor specific transplantation antigen induced in cells infected or
transformed by polyomavirus.J. nut. CancerInst., 70,283-290 (1983).
BEN-BARUCH
LANCER,A., EMMANUEL,
N., EVEN,J., FRIDMAN,
W.H.,
GOHAR,O., GONEN,
B., KATZ,B.-Z., RAN,M., SMORODINSKY,
N.I. and
WITZ.I.P., Phenotvuic Drooerties of 3T3 cells transformed in vitro with
polyoma virus and- Bssaged once in syngeneic animals. J. Immunobiol.,
185,281-291 (19927.
N., RousBILLAUD,
M., BUSSON,
P., HUANG,D., MUELLER-LANTZCH,
G., PAVLISH,
O., WAKASUGI,
H., SEIGNEURIN,
J.M., TURSZ,T.
and LENOIR,M., Epstein-Barr virus (EBV)-containing nasopharyngeal
carcinoma cells express the B-cell activation antigen blast2/CD23 and
low levels of the EBV receptor CR2. J. ViroL, 63,41214128 (1989).
BIRAN,H., MAVLIGIT,G.M. and MOAKE, J.L., Receptor sites for
complement and for immune complexes on human nonhemopoietic
tumor cells. Cancer, 44,131-135 (1979).
BONNEROT,
C., AMIGORENA,
S., CHOQUET,
D., PAVLOVICH,
R., CHOUKROUN,
V. and FRIDMAN,
W.H., Role of associated y-chain in tyrosine
kinase activation via murine FcTRII. EMEO J., 11,2747-2758 (1992).
BONNEROT,
C., AMIGORENA,
S., FRIDMAN,
W.H., EVEN,J. and DAERON,
M., Unmethylation of specific sites in the 5’ region is critical for the
expression of murine aFcyR gene. J. Immunol., 144,323-328 (1990).
E., Circular and linear
DORSETT,D., DEICHITE,I. and WINOCOUR,
SELET,
FcyRIIBl IS A TUMORIGENICITY-ENHANCING FACTOR
229
Simian Virus 40 DNAs differ in recombination. Mol. cell. Biol., 5, NICOLSON,G.L., GALLICK,G.E., SPOHN,W.H., LEMBO,T.M. and
869-880 (1985).
TAINSKY,M.A., Transfection of activated c-H-rasE.’./pSV2neo or
FEARON,E.R. and VOGELSTEIN,
B., A genetic model for colorectal pSVneo genes into rat mammary cells: rapid stimulation of clonal
diversification of spontaneous metastatic and cell-surface properties.
tumorigenesis. Cell, 61,759-767 (1990).
Oncogene, 7,1127-1135 (1992).
FRIDMAN.
W.H.. BONNEROT.
C.. DAERON.M.. AMIGORENA.
S.. TEILH.W., Fc and complement receptors on malignant tumor
LAUD, J-L. and SAUTES,
C., Structural bases of’Fcy receptor functions. NOLTENIUS,
cells. Cancer, 48,1761-1767 (1981).
Immunol. Rev., 125,49-75 (1992).
GONEN,B., KAHANA,
0. and WITZ,I.P., In vivo tumorigenicity and in PETO, R., PIKE, M.C., ARMITAGE,P., BRESLOW,N.E., Cox, D.R.,
K., PETRO,J. and SMITH,
vitro sensitivity to tumor-necrosis-factor 01 mediated killing of c-Ha-ras- HOWARD,S.V., MANTEL,N., MCPHERSON,
transformed cells. Cancer Immunol. Immunother., 35, 388-394 (1992). P.G., Design and analysis of randomized clinical trials requiring
prolonged observation of patient. Brit. J. Cancer, 35,l-39 (1977).
GONEN,B., RAN,M., WRESCHNER,
D.H., CAHALON,
L., ANAVI,R.,
LANGER,A,, ELIASSI,I., GOHAR,O., GONEN,
SHLOMO,
Y. and WITZ,I.P., Some immunological properties of high RAN,M., BEN-BARUCH
S., FRIDMAN,
W.H., TEILLAUD,
J.-L. and WITZ,I.P.,
and low tumorigenic cellular variants of c-H-ras-transformed 3T3 cells. B., GRADSZTAJN,
Possibilities of interference with the immune system of tumor bearers
Nut. Immun. cell. Growth Regul., 7,144-154 (1988).
by non lymphoid FcyRII expressing tumor cells. Immunobiology, 185,
GORINI,G., CIOTTI,M.T., STARACE,
G., VIGNETI,
E. and RASCHELLA, 415-425 (1992).
G., Fcy rece tors are expressed on human neuroblastoma cell lines:
E., TEILLAUD,
J.-L., EVEN,
lack of correration with N-myc oncogene activity. Int. J. Neurosci., 62, RAN,M., KATZ,B., KIMCHI,N., HALACHMI,
J., BERKO,Y., ATLAS,E., FRIDMAN,
W.H. and WITZ,I.P., The in vivo
287-297 (1992).
acquisition of FcyRII expression on polyoma virus transformed cells
HALACHMI,
E. and WITZ,I.P., Differential tumorigenicity of 3T3 cells derived from tumors of long latency. Cancer Res., 51,612-618 (1991).
transformed in vitro with polyoma virus and in vivo selection of high
RAN,M., TEILLAUD,
J.-L., FRIDMAN,
W.H., FRENKEL,
H., HALACHMI,
tumorigenic clones. Cancer Res., 49,2383-2389 (1989).
E., KATZ,B., GIPS, M., SHLOMO,Y . , BARZILAY,
J. and WITZ, I.P.,
HEO, D.S., PARK,J.-G., HATA,K., DAY, R., HERBERMAN,
R.B. and Increased expression of Fc receptor in cancer patients and tumor
WHITESIDE,
T., Evaluation of tetrazolium-based semiautomatic colori- bearing mice. Molec. Immunol., 25,1159-1167 (1988).
metric assay for measurement of human antitumor cytotoxicity.Cancer
RAN, M. and WITZ, I.P., FcR derived from without the immune
Res., 50,3681-3690 (1990).
system-a potential escape mechanism for cells propagating in a
HERRLICH,
P., ZOLLER,M., PALS,S.T. and PONTA,H., CD44 splice hostile immunological environment. Contr. Gynec. Obsfet., 14, 83-89
variants: metastases meet lymphocytes. Immunol. Today, 14, 395-399
(1985).
(1993).
RAVETCH,J.V., LUSTER,A.D., WEINSHANK,
R., KOCHAN,J., PAVHOGARTH,
P.N., WITORT,E., HULETT,M.D., BONNEROT,
C., EVEN,J., LOVEK,A., PORTNOY,
D.A., HULMES,J., PAN,Y-C.E. and UNKELESS,
FRIDMAN,
W.H. and MCKENZIE,I.F.C., Structure of the mouse p Fc J.C., Structural heterogeneity and functional domains of murine
receptor II gene. J. Immunol., 146,369-376 (1991).
immunoglobulin G Fc receptors. Science {Wash.), 234,718-725 (1986).
KERR,W.G., NOLAN,G.P., SERAFINI,
A.T. and HERZENBERG,
L.A., RIMOLDI,D., SALVI,S., HARTMAN,F., SCHREGER,M., BLUM,S.,
Transcriptionally defective retroviruses containing lac2 for the in situ ZOGRAFOS,
L., PLAISANCE,
S., AZZARONE,
B. and CARREL,S., Expresdetection of endogenous genes and developmentally regulated chroma- sion of IL-2 receptors in human melanoma cells. Anticancer Res., 13,
tin. Cold SpringHarbor Symp. quant. Biol., LIV, 767-776 (1989).
555-564 (1993).
KIM,T.S. and COHEN,E.P., Interleukin-2-secreting mouse fibroblasts THEODORESCU,
D., CORNIL,I., FERNANDEZ,
B. and KERBEL,R.S.,
transfected with genomic DNA from murine melanoma cells prolong Overexpression of normal and mutated forms of HRAS induces
the survival of mice with melanoma. CancerRes., 54,2531-2535 (1994). orthotopic bladder invasion in a human transitional cell carcinoma.
LEVINE,A.J., The tumor suppressor genes. Ann. Rev. Biochem., 62, Proc. nut. Acad. Sci. (Wash.), 87,9047-9051 (1990).
TONDER,
O., MORSE,P.A., JR. and HUMPHREY,
L.J., Similarities of Fc
623-651 (1993).
.
,
MANIATIS,
T., FRITSCH,E.F. and SAMBROOK,
J., Molecular cloning: a receptors in human malignant tissue and normal lymphoid tissue. J.
laboratbry manual, Cold Spring Harbor Laboratory, Cold Spring Immunol., 113,1162-1169 (1974).
UNKELES,J.C., Characterization of a monoclonal antibody directed
Harbor, NY (1983).
MIETTINEN,H.M., MATTER, K., HUNZIKER,W., ROSE, J-K. and against mouse macrophage and lymphocyte Fc receptors. J. exp. Med.,
.
,
MELLMAN,
I., Fc receptor endocytosis is controlled by cytoplasmic 150,580-596 (1979).
domain determinant that actively prevents coated pit localization. J. WITZ, I.P. and RAN, M., Could Fc-receptor facilitate the escape of
immunogenic premalignant cells from host defense? A hypothesis.
Cell Biol., 116,875-888 (1992).
MILGROM,
F., HUMPHREY,
L.J., TONDER,O., YASUDA,
J. and WITEB- Ann. Inst. Pasteur, 136C, 423-428 (1985).
WITZ,I.P. and RAN,M., FcR may function as a progression factor of
SKY, E., Antibody mediated hemadsorption by tumor tissues. Int. Arch.
nonlymphoid tumors. Immunol. Res., 11,283-295 (1992).
Allergy appl. Immunol., 33,478-492 (1968).
Документ
Категория
Без категории
Просмотров
3
Размер файла
1 052 Кб
Теги
944
1/--страниц
Пожаловаться на содержимое документа