Skip to main content
NCBI home page
Cancer Immunology, Immunotherapy : CII logoLink to Cancer Immunology, Immunotherapy : CII
. 2005 Feb 23;54(9):837–847. doi: 10.1007/s00262-004-0656-z

Role of tumor-derived transforming growth factor-β1 (TGF-β1) in site-dependent tumorigenicity of murine ascitic lymphosarcoma

V S Thakur 1, B Shankar 1, S Chatterjee 1, S Premachandran 1, K B Sainis 1,✉
PMCID: PMC11032987  PMID: 15726358

Abstract

An ascitic lymphosarcoma (LS-A) of Swiss mice that regressed spontaneously on subcutaneous (s.c.) transplantation was investigated for the mechanism of its progressive growth and host mortality on intraperitoneal (i.p.) transplantation. In vitro studies indicated significant inhibition of LS-A proliferation seeded at higher cell density (>104/ml). Culture supernatants of LS-A caused bi-modal growth effects, the early supernatants (24 h) caused stimulation and the late (72 h) supernatants inhibited LS-A proliferation. The 72-h supernatants also suppressed T and B cell response to mitogens in a dose-dependent manner. Pan anti-transforming growth factor-β antibody abrogated the inhibitory effects of supernatants. The supernatants contained both latent as well as bio-active form of transforming growth factor-β1 (TGF-β1) as determined by ELISA. Mice bearing i.p. ascites tumor had elevated serum TGF-β1, hemoglobulinemia, splenic lymphopenia, impaired response of the T cells to mitogen and reduced expression of transferrin receptor (CD71) on the bone marrow cells. However, mice which rejected s.c. transplants, did not show significant changes in these parameters. Our studies indicated profound influence of site of tumor growth on tumor progression and host immune system mediated by tumor-derived TGF-β1. It is possible that human tumors which secrete TGF-β1 may exhibit similar patho-physiological effects in the host depending on the anatomical site of the tumor.

Keywords: TGF-β, Immunosuppression, Site-dependent tumor growth, Hemoglobulinemia, Lymphopenia, Transferrin receptor (CD71)

Introduction

The site of tumor growth and its tissue microenvironment is an important factor in tumor progression, invasion, metastasis, host immune response and response to therapeutic treatments. Preclinical studies have shown differences in response to chemotherapy when the same tumor was implanted at several anatomical sites [8, 33, 37] suggesting profound influence of organ microenvironments in the control of tumor growth. Differential organ microenvironmental effects are also reported in the development of anti-tumor immunity by the host. Using orthotopic and ectopic transplantation of pancreatic tumor expressing human MUC1 transgene, the organ site was shown to be critical in the development of anti-MUC1 immune response in mice [20]. Understanding the host–tumor interactions of a particular tumor at different anatomical sites is, therefore, important so as to develop effective therapeutic treatments for primary as well as metastatic tumor growths.

Tumors which express immunogenic epitopes can elicit anti-tumor immune responses and often undergo rejection in syngeneic/autochthnous host and it is a cumulative result of the action of a number of cellular and humoral mediators [24]. Cytokines elaborated by the stromal tissue, the cells of immune system and the tumor itself can modify tumor microenvironment and hence host response to tumors. Many human tumors secrete cytokines that act as autocrine tumor growth factors. These include interleukin-6 in hematological tumors such as myeloma or hairy cell leukemia and renal cell carcinoma, tumor necrosis factor (TNF) in leukemia or neuroblastoma, interleukin-10 (IL-10) in lymphoma [5] and transforming growth factor-β (TGF-β) in case of carcinoma [1, 36]. The spectrum of tumor-derived cytokines has been shown to be different at different anatomical sites of tumor [14]. These cytokines may cause both generalized and specific inhibition of immune responses [6]. Earlier, we have reported that an ascitic lymphosarcoma (LS-A) tumor in Swiss mice, showed spontaneous regression on s.c transplantation and induced anti-tumor immunity that could be transferred adoptively by tumor-immune CD4+ T cells [35]. The anti-tumor immune response against LS-A tumor was not observed in mice transplanted by i.p. route and even a very small dose of LS-A tumor inoculum was lethal to the mice. The present studies were undertaken to examine the mechanism underlying i.p. site-dependent tumorigenicity of LS-A cells in the syngeneic Swiss mice and to ascertain the role of tumor-derived cytokines, if any.

Materials and methods

Chemicals

RPMI 1640 medium, sodium bicarbonate, 2-mercaptoethanol, fetal calf serum, anti-mouse CD71, anti-mouse-IgG-PE, pan TGF-β antibody and anti-rabbit-IgG-PE were obtained from Sigma Chemical Company, USA. TGF-β1, TGF-β2, TGF- α, Oncostatin M (OSM), Leukemia inhibitory factor (LIF) were purchased from Boehringer Mannheim, Germany. Concanavalin A (Con A) was procured from CalBiochem, USA. LPS was purchased from Difco Laboratories, USA. Purified rat anti-mouse, human, pig TGF-β1 monoclonal antibody (Rat IgG2a, κ; clone A75-2.1) and biotinylated rat anti-mouse, human, pig monoclonal antibody (Rat IgG2a, κ, clone A75-3.1) were procured from BD PharMingen, USA. 3H-thymidine (specific activity: 6.5 Ci/mM) was obtained from Board of Radiation and Isotope Technology, Mumbai, India. Penicillin and streptomycin were purchased from reputed local manufacturers.

Animals

Inbred Swiss albino male mice, 8–10 weeks old and weighing between 23–25 g were reared in the animal house facility of the Bhabha Atomic Research Centre (BARC). Inbred Swiss mice were obtained by breeding among littermates for more than 40 generations. The syngenicity was routinely checked by reciprocal skin grafts, which survived indefinitely. The guidelines issued by the Institutional Animal Ethics Committee of Bhabha Atomic Research Centre, Government of India, regarding the maintenance and dissection of experimental animals were followed.

Tumor

These studies were carried out on an ascitic lymphosarcoma (LS-A), which had originated spontaneously in inbred Swiss mice [35]. The tumor had arisen spontaneously in a mouse in the Swiss mouse colony of our Research Centre. The mouse showed splenomegaly and lymph node enlargement but no ascites. The ascitic variant, LS-A, arose following serial i.p. passages of the original tumor cells of the spleen. The tumor was maintained by serial i.p. transplantation of 106 cells/animal, twice a month, into naïve mice.

Medium

All the cell suspensions were prepared in RPMI 1640 medium. The medium was supplemented with sodium bicarbonate (2 g/l), 2- mercaptoethanol (5×10−5 M), 10% heat-inactivated fetal calf serum, penicillin (100 U/ml) and streptomycin (100 μg/ml). This is referred to as complete medium.

Tumor cell suspension

Ascitic lymphosarcoma cells were obtained asceptically from peritoneal lavage of mice bearing 10–15-day-old ascitic tumor. The cells were washed twice in complete medium by centrifugation. Viability was determined by trypan blue dye exclusion method. It was usually >95%.

Generation of LS-A culture supernatants

Ascitic lymphosarcoma cells (2×106) were cultured in 10 ml complete medium at 37°C in a humidified 95% air/5% CO2 atmosphere, using tissue culture flasks (Falcon). Cell-free supernatants were collected at the end of 24-h and 72-h by centrifugation. Aliquots of supernatants were stored at −70°C and were used within 3 weeks.

Measurement of effect of culture supernatants on in vitro proliferation of LS-A cells

The effect of LS-A-derived culture supernatants on proliferation of LS-A cells was determined using proliferation assay based on the incorporation of 3H-thymidine [7] with minor modifications. LS-A cells (104/ml) were seeded in six replicates in 0.2 ml complete medium using 96 well microplates (Nunc, Denmark) in presence or absence of culture supernatants. The cultures were incubated in a CO2 incubator at 37°C for 72 h and were pulsed with 3H-thymidine (1 μCi/well) for additional 16 h. The cultures were harvested using a cell harvester (Dynatech; USA) on fibre glass filter papers. The scintillation fluid used was prepared using 4 g PPO and 0.3 g POPOP (Sigma Chemical Company, USA) in 1 L of toluene (Sarabhai Chemical Company, India). Radioactivity was measured using a liquid scintillation counter (LKB Wallac; Model: 1217 Rack beta).

Effect of anti TGF-β antibody on tumor growth

Ascitic lymphosarcoma cells were grown in presence of tumor-derived culture supernatants in presence or absence of pan TGF-β antibody (10 μg/ml) and their growth was assessed as mentioned earlier.

Sensitivity of LS-A tumor cells to recombinant cytokines

The sensitivity of LS-A cells to a spectrum of recombinant cytokines such as TGF-β1, TGF-β2, TGF-α, OSM, and LIF was determined by proliferation assay. LS-A cells (104/ml) were treated with different concentrations of cytokines in a total volume of 0.2 ml and cellular proliferation was determined by the method of 3H-thymidine incorporation as described earlier.

Estimation of TGF-β1 in culture supernatants and sera of mice bearing s.c. and i.p. tumors

Mice were injected with 106 LS-A cells by either s.c. or i.p. route (3 mice/group). Naïve, normal mice served as control. Blood samples were collected from retro-orbital plexus on different days (5, 10, 15 and 20); post tumor transplantation and sera were separated. The ELISA for TGF-β1 measures only the active form of the cytokine and acidification would add the latent amount to the un-acidified sample. Serum samples and culture supernatants stored at −70°C, were acidified with 1 N HCl and analyzed for TGF-β1 content by sandwich ELISA. Briefly, 96 well ELISA plates were coated with purified rat anti-mouse, human, pig TGF-β1 monoclonal antibody. Recombinant human TGF-β1 (from CHO cells) was used as a standard. Biotinylated, rat anti-mouse, human, pig TGF-β1, monoclonal antibody was used as detection antibody. Avidin-Horse radish peroxidase was used as an enzyme conjugate. The substrate used was 3,3′, 5,5′ tetramethylbenzidine (TMB). The reaction was stopped by addition of 1 N H2SO4 and optical density measurements were made on an automated ELISA reader (MediSpec Instruments India Ltd) at 450 nm.

Staining for intracellular TGF-β

Detection of intracellular TGF-β was carried out by the method of Garba et al. [10] with minor modifications. LS-A cells were fixed with 4% formaldehyde for 10 min in ice. Cells were washed with PBS and treated with 10% FCS for 20 min in ice to block non-specific binding of the antibody. The cells were then incubated with pan TGF-β antibody (1 μg/106  cells) for 30 min in ice. After extensive washing with PBS, cells were labeled with anti-rabbit-IgG-PE. The cells treated only with the secondary antibody served as negative control. Twenty thousand cells were acquired for each sample on flow cytometer (FACS Vantage, Becton Dickinson, USA). The per cent cells positive for intracellular TGF-β were calculated using CellQuest software (Becton Dickinson Immunocytometry Systems, USA).

Staining for transferrin receptor (CD71) on bone marrow cells of tumor bearing mice

Mice were injected with 106 LS-A cells either by s.c. or i.p. route. Naïve, normal mice served as control. Marrow cells were obtained from the femur bone of normal mice or s.c. and i.p. tumor (15-day-old tumor) bearing mice (3 mice/group). Labeling of CD71 on bone marrow cells was carried out according to the method of Schomaker et al. [30]. The cells were washed with PBS and were labeled with 1 μg/106 cells of anti-mouse CD71 for 30 min in ice. The isotype control used was mouse IgG1 (MOPC-21) at the same concentration. After extensive washing, the cells were further treated with anti-mouse-IgG-PE for 30 min in ice. Twenty thousand cells were acquired for each sample on flow cytometer. The percentage of cells positive for transferrin receptor and the mean fluorescence intensity at 585 nm were calculated using CellQuest software.

Estimation of blood hemoglobin and total nucleated cells in the spleen

Mice were injected with LS-A cells (106/animal) by s.c. or i.p. route. Naïve, normal mice served as control group. Three mice were used per group per time point. Blood samples were collected, under anesthesia individually from the retro-orbital plexus on different days (5, 10, 15, 20 and 23), following tumor transplantation. Hemoglobin content in the heparinized blood samples was estimated by Drabkin’s method [2]. Spleens were also dissected out from these animals on different days (5, 10, 15, 20 and 23) following tumor transplantation and cell suspensions were prepared individually in the medium. Total nucleated cells per spleen were enumerated by using methylene blue:acetic acid diluent.

Lymphocyte proliferation assay

The effect of LS-A-derived culture supernatants on the function of normal T cells and B cells and those obtained from s.c. and i.p. LS-A tumor bearing mice was determined using polyclonal mitogens. Spleen cells were obtained from either normal mice or mice bearing 15-day-old s.c. and i.p. tumors (3 mice/group). Red blood cells were lysed by treatment with 0.01 M Tris–NH4Cl (0.83%) solution (pH 7.2). Viability was determined by trypan blue dye exclusion method. Spleen cells were stimulated with either concanavalin A (2.5 μg/ml per 106 cells) or E. coli lipopolysaccharide (LPS) at 50 μg/ml per 106 cells, in a total volume of 0.2 ml complete medium. The cultures were pulsed with 3H-thymidine (1 μCi/well) for 16 h at the end of 72 h and 48 h, respectively.

Statistical analyses

The data were analyzed by applying Student’s t test. A P value of 0.05 and less was considered significant.

Results

LS-A growth in vitro

The effect of initial cell density on LS-A proliferation in vitro is shown in Fig. 1. Optimal 3H-thymidine incorporation at the end of 72 h was observed with initial cell densities of 103–104/ml, above which significant (P<0.01) inhibition of 3H-thymidine incorporation was observed. Total inhibition of 3H-thymidine incorporation was observed when 106 LS-A cells/ml were initially seeded and cultured for 72 h (Fig. 1). It was possible that the tumor cells seeded at higher densities did grow initially. However, at the end of 72 h when 3H-thymidine was added, no dividing cells remained due to the inhibitory action of some tumor-derived soluble factor.

Fig. 1.

Fig. 1

Open in a new tab

Effect of initial cell density on the growth of LS-A cells in vitro. Cells were cultured for 3 days in complete medium in a 95% air/5% CO2 atmosphere before 3H thymidine (1 μCi/well) was added and they were cultured further for 16 h. Cellular proliferation was estimated in terms of 3H-thymidine incorporation. Each value represents mean CPM ± SE of six replicates. Two independent experiments were carried out. **P<0.01, as compared to the optimal incorporation at 104/ml

To ascertain this possibility, LS-A cells were seeded at the initial density of 2×105/ml. Culture supernatants were obtained at two different time points. The effects of early supernatant (24 h) and late supernatant (72 h) were tested on in vitro proliferation of LS-A cells. It was observed that the early supernatant caused significant (P<0.01) stimulation of proliferation (Fig. 2a) up to 1:4 dilution. At higher dilution (1:8), 3H-thymidine incorporation decreased but remained higher than the control (Fig. 2a). Figure 2b shows the effect of late supernatants (72 h) on LS-A proliferation. At lower dilution (1:2), significant inhibition (P<0.01) of proliferation of LS-A cells was observed. This inhibition decreased at higher dilution (1:8) of the supernatant and 3H-thymidine incorporation was almost at the level of the control (Fig. 2b).

Fig. 2.

Fig. 2

Open in a new tab

Proliferation of LS-A cells (1×104/ml) in vitro in the presence of different dilutions of culture supernatants. a Early supernatant (24-h culture) and b late supernatant (72-h culture). Proliferation of cells was measured as explained in Fig. 1. Each value represents mean CPM ± SE of six replicates. Four independent experiments were carried out and one representative experiment is shown. **P<0.01, as compared to 3H-thymidine incorporation in control group (cells grown in complete medium)

Immunosuppressive effects of LS-A supernatants

The effect of LS-A culture supernatants on the mitogen-induced stimulation of host lymphocytes is shown in Fig. 3. Polyclonal mitogens viz. con A and LPS stimulated splenic T and B lymphocytes respectively, of normal mice (Fig. 3). When the spleen cells were co-cultured with 24-h supernatant in presence of con A, there was no significant suppression of T cell proliferation. However, co-culture with 72-h supernatant significantly (P<0.01), suppressed T cell proliferation. A similar trend of suppression was observed in the case of LPS-induced proliferation of B cells co-cultured with 72-h culture supernatant (Fig. 3).

Fig. 3.

Fig. 3

Open in a new tab

Effect of 24-h and 72-h culture supernatants on the response of normal T and B cells to polyclonal mitogens. Spleen cells (1×106/ml) were stimulated with con A (2.5 μg/ml) or LPS (50 μg/ml) in 0.2 ml complete medium and cultured for 72 h and 48 h, respectively before the addition of 1 μCi of 3H-thymidine per well. Each value represents mean CPM ± SE of six replicates. Four independent experiments were carried out and one representative experiment is shown. **P<0.01, as compared to mitogen treated control. Filled rectangle cells only, line shaded rectangle cells + mitogen, gray shaded rectangle cells + mitogen + 24 h sup, open rectangle cells + mitogen + 72 h sup

This immunosuppressive effect was also seen in vivo. Figure 4a shows lymphocyte counts in spleens of normal mice and mice bearing i.p. or s.c. tumors. In normal mice, the lymphocyte count per spleen did not change significantly with time. In case of s.c. and i.p. tumor bearing mice, there was an initial increase in lymphocyte count, probably due to immune response mounted by the host. In tumor bearing mice, this initial surge in the lymphocyte count declined only by day 23, after tumor transplantation. On the contrary, i.p. tumor bearing mice showed sharp decline in lymphocyte count with time. Severe lymphopenia was observed in these mice on day 23, a time point when the mice had full-blown ascites.

Fig. 4.

Fig. 4

Open in a new tab

a Total nucleated cell count per spleen taken at different time points following i.p. and s.c. LS-A transplantation. Each value represents mean for nine mice obtained from data of three experiments each comprising of three mice per time point. open square normal,filled circle s.c. tumor, filled triangle i.p. tumor. b Response of splenic lymphocytes obtained from i.p. and s.c. tumor bearing mice to con A. Spleen cells were taken 15 days after tumor transplantation and were stimulated with con A (2.5 μg/ml/106 cells/ml) and cultured for 72 h before the addition of 1 μCi of 3H-thymidine per well. Each value represents mean CPM ± SE of six replicates. Three mice per group were used in each experiment and two such independent experiments were carried out. One representative experiment is shown. **P<0.01 as compared to mitogen treated cells of normal mice. filled rectangle Cells, open rectangle cells + con A

The functional capability of splenic lymphocytes of s.c. and i.p. tumor bearing mice was evaluated ex vivo. Fifteen days after tumor transplantation, the spleen cells were stimulated with con A. Lymphocytes from mice bearing s.c. tumors showed about 30% reduction in 3H-thymidine incorporation as compared to control mice. Strikingly, lymphocytes of i.p. tumor bearing mice exhibited 95% reduction (P<0.01) in response to con A as compared to control mice (Fig. 4b).

Identification of LS-A derived soluble factor

As a preliminary step towards identification of the soluble factor(s) present in the supernatant, different recombinant cytokines were added to LS-A cultures and their effect on proliferation was measured. Fig. 5a shows that TGF-β1 caused bi-modal effects on the growth of LS-A, i.e., at lower concentration (0.01 ng/ml), significant (P<0.01) stimulation was observed but higher concentrations of the cytokine (>10 ng/ml) significantly (P<0.01) suppressed LS-A proliferation. The isoform TGF-β2 caused inhibition of proliferation at all the concentrations used (Fig. 5b). Other cytokines viz. TGF-α, LIF and OSM showed no significant effect on LS-A proliferation at any of the concentrations used (Fig. 5b). Incubation of LS-A cells with supernatants (72 h) treated with pan TGF-β antibody abrogated proliferation inhibitory effect of these supernatants almost completely (P<0.01; Fig. 5c), indicating that the supernatant contained TGF-β as the growth-modulating factor.

Fig. 5.

Fig. 5

Open in a new tab

Identification of the tumor-derived cytokine of LS-A cells. Effect of a recombinant TGF-β1 b r-TGFα (open square), r-TGF-β2 (filled circle), r- LIF (open triangle), and r-OSM (Down filled triangle) on the proliferation of LS-A cells in vitro. Untreated LS-A cells incorporated 223778±8419 cpm. c Effect of pan-anti TGF-β antibody on proliferation of LS-A cells in presence of 72-h culture supernatants. Each value represents mean CPM ± SE of six replicates. Two independent experiments were carried out and one representative experiment is shown. **P<0.01, as compared to respective controls

The above results suggested the involvement of TGF-β in the bimodal action of LS-A culture supernatant. Furthermore, intracellular TGF-β was detected in LS-A cells obtained from i.p. tumor ascites using flow cytometry. About 58% of LS-A cells expressed TGF-β, whereas negative control cells showed less than 5% staining (Fig. 6).

Fig. 6.

Fig. 6

Open in a new tab

Flow cytometric detection of intracellular TGF-β in LS-A cells. Overlapped histograms of LS-A cells labeled with pan-anti-TGF-β antibody and phycoerythrin labeled secondary antibody. M2 population represents the per cent TGF-β positive cells. M2 values for different groups are given in paranthesis below, dotted line unlabeled (0.49%), thin line negative control (4.19%), bold line TGF-β (57.64%)

Estimation of TGF-β1 in LS-A supernatants and sera of tumor bearing mice

Figure 7 shows the amounts of active and latent form of TGF-β1 present in the early (24 h) and the late (72 h) supernatants. To estimate total TGF-β1 (active plus latent), supernatants were acidified to release the bioactive form. Both, un-acidified and acidified supernatants were analyzed for TGF-β1 content using ELISA. The acidified supernatants (24 h and 72 h) contained higher quantity of TGF-β1 as compared to their respective non-acidified counterparts, which indicated the presence of both the active and latent form of TGF-β1 in the supernatants. Further, the 72-h supernatants contained almost 20 times more active TGF-β1 as compared to the 24-h supernatants (Fig. 7).

Fig. 7.

Fig. 7

Open in a new tab

Estimation of active and total TGF-β1 in 24-h and 72-h culture supernatants of LS-A cells using sandwich ELISA. Total TGF-β1 was estimated before and after acidification of culture supernatants, which released the active form. Each value represents mean ± S. E. from three replicates. Two independent experiments were carried out and one representative experiment is shown. open rectangle unacidified, shaded rectangle acidified

Transforming growth factor-β1 levels were determined in the acidified sera of mice bearing LS-A tumors at the two anatomical sites (i.p. and s.c.) at different time points after tumor transplantation. Serum TGF-β1 level gradually increased in mice bearing i.p. ascites (Fig. 8). These levels were significantly (P<0.01) higher than that of normal mice or those bearing s.c. tumors, at all the time points. However, TGF-β1 levels did not increase significantly during s.c. tumor growth as a function of time and they were comparable to that of normal mice (Fig. 8).

Fig. 8.

Fig. 8

Open in a new tab

Estimation of serum total TGF-β1 in mice bearing i.p. and s.c. LS-A tumors using sandwich ELISA. Each value represents mean of six replicates. Three mice per group were used. Two independent experiments were carried out and one representative experiment is shown. **P<0.01, as compared to normal. Open rectangle day 10, Shaded rectangle day 15, filled rectangle day 20.

Tumor-induced hemoglobulinemia

Hemoglobin levels were measured in peripheral blood samples of normal and tumor (i.p. or s.c) bearing mice. Both normal mice and s.c. tumor bearing mice had comparable blood hemoglobin profile (Fig. 9a). Their levels also remained steady with time. Interestingly, mice bearing i.p. tumors developed severe hemoglobulinemia, concomitant with tumor progression. On day 23, a significant reduction (62.5%) as compared to hemoglobin level of corresponding normal mice was observed. Since one of the reasons for hemoglobulinemia could be the reduced uptake of iron carrier transferrin due to reduced expression of transferrin receptor (CD71) on hematopoietic erythroid progenitor cells, CD71 expression on bone marrow cells was determined using flow cytometry. The percentage of cells expressing CD71 was not significantly different among normal, s.c. and i.p. tumor bearing mice (Fig.9b, Table 1). However, the mean fluorescence intensity which indicates the transferrin receptor density was significantly reduced (P<0.01) on the bone marrow cells of i.p. tumor bearing mice as compared to normal or s.c. tumor bearing mice (Fig. 9b, Table 1).

Fig. 9.

Fig. 9

Open in a new tab

a Total hemoglobin levels in peripheral blood taken at different time points following i.p. and s.c. tumor transplantation. Each value represents mean for nine mice obtained from pooling three experiments comprising of three mice per time point. filled square normal, open circle s.c. tumor, filled triangle i.p. tumor. b Flow cytometric profiles showing CD71 expression on bone marrow cells of normal mice and mice bearing s.c. and i.p. tumors

Table 1.

Expression of CD71 on bone marrow cells of normal mice and mice bearing s.c. and i.p. tumors

Group Mean fluorescence intensity (Percentage of bone marrow cells positive for transferrin receptor CD71)
Normalmice 169.20±5.0 (32.97±0.8)
s.c. tumor bearing mice 151.61±4.8 (35.49±3.2)
i.p. tumor bearing mice 120.05±1.4 (29.89±1.8)
Open in a new tab

Discussion

Earlier, we have shown that LS-A cells transplanted by i.p. route formed ascites and caused mortality in the recipient mice at the lowest dose (102 cells/animal) used. However, solid tumor formed by ectopic s.c. transplantation of LS-A cells, regressed spontaneously even when a much higher dose of 106 cells/animal was used. These animals subsequently developed anti-tumor immunity [35]. Similar site-dependent growth of tumors has been shown by other investigators using different tumor models. Ectopic transplantation of human pancreatic tumor cells (AsPC-1) into nude mice, caused slow growth of tumor [34] AK-5, a rat histiocytoma that grew as ascites on i.p. transplantation, regressed spontaneously after s.c. transplantation [15]. Human colon carcinoma cells (KM12) metastasized only following cecal wall transplantation but not when transplanted at the subcutaneous site [22] indicating organ microenvironment as a decisive factor in tumor progression. Anti-tumor responses against human MUC1 antigen expressed as transgene by pancreatic cancer cell line syngeneic to C57BL/6 mice, were less effective at the orthotopic site than at the ectopic s.c. site [20]. From these observations, it is seen that tumor cells showed varied growth response in the host depending on the site of transplantation. Some investigators have shown the involvement of cytokines in this site dependent growth of tumor cells [4, 11, 19, 21, 25, 31]. The magnitude of angiogenic cytokine production was higher when the human tumor xenografts were grown as ascites in the peritoneal cavity than at the s.c. site [14].

Ascitic lymphosarcoma cells also showed cell density-dependent growth under in vitro conditions (Fig. 1). An optimum incorporation of 3H-thymidine was observed at initial densities of 103 and 104 LS-A cells/ml. At higher cell densities, significant suppression of 3H-thymidine incorporation was seen. Since, 3H-thymidine was added at 72 h after initiation of cultures, it was likely that after the initial growth, the proliferation of LS-A cells was inhibited probably by some LS-A derived soluble factor in a concentration-dependent manner, suggesting an autocrine regulatory loop in the growth of this tumor. Cell-free supernatants obtained at different time points from LS-A cells grown in vitro, caused bi-modal effects on the growth of LS-A cells supporting the autocrine regulatory loop in LS-A tumor growth. The early (24 h) supernatant caused stimulation and the late (72 h) supernatant inhibited the proliferation of LS-A cells (Figs. 2a and 2b). It was also found that the 24-h supernatant did not have a significant effect on either T cell or B cell responses to mitogens (Fig. 3). However, the 72-h supernatant significantly suppressed both T and B cell responses (Fig. 3). These observations suggested that the putative factor present in the supernatant was not only controlling the growth of LS-A cells but also responsible for immunosuppression in the LS-A tumor bearing mice. This was further supported by our observations that mice bearing i.p. LS-A tumors showed severe lymphopenia in the spleen concomitant with tumor progression as against the lack of lymphopenia in mice bearing s.c. tumors (Fig. 4a). The initial upsurge of lymphocyte counts in the spleen of i.p. tumor bearing mice could be due to host sensitization by immunogenic LS-A tumor, which later failed to eliminate the tumor, possibly due to paracrine immunosuppressive effects of the putative LS-A-derived soluble factor. Further, a severe impairment of splenic T cells response to con A in i.p tumor bearing mice as against a marginal impairment in s.c. tumor bearing mice (Fig. 4b), suggested an immunosuppressive role of LS-A-derived soluble factor in tumor bearing mice.

The nature of the soluble factor in LS-A culture supernatant could be deduced from the response of LS-A cells to different recombinant cytokines. r-TGF-β1 showed bi-modal effect on tumor growth (Fig. 5a), similar to that seen with early and late supernatants of LS-A cultures (Fig. 2a, b). r-TGF-β2 showed growth inhibition at all the concentrations used (Fig. 5b). Proliferation of LS-A cells was not affected by other cytokines r-TGF-α, r-LIF and r-oncostatin M (Fig. 5b). Further, pan anti-TGF-β antibody completely abolished the inhibitory effect of 72-h culture supernatant on LS-A cells which confirmed the role of TGF-β as the major growth modulator present in the LS-A supernatants (Fig. 5c). TGF-β was also detected intracellularly (Fig. 6) in LS-A cells using flow cytometry providing an additional evidence that this cytokine is released in the supernatant. About 60% of LS-A cells were positive for TGF-β, indicating heterogeneity of LS-A tumor, which was seen earlier by karyotyping [17].

Both latent and active forms of the TGF-β1 were detected in LS-A derived supernatant. The 24-h supernatant contained lower concentration (0.4 ng/ml) and the 72-h supernatant contained higher concentration (9 ng/ml) of active TGF-β1 (Fig. 7). These concentrations of active TGF-β1 in the culture supernatants (Figs. 2a, b, respectively) matched with stimulatory and inhibitory effects respectively observed with r-TGF-β1 (Fig. 5a).

The cytokines belonging to the TGF-β family are secreted by many types of normal and malignant cells [26, 28]. Different isoforms identified as TGF-β1 to TGF-β 5 are constitutively produced in a latent form that contains mature TGF-β and its NH2-terminal propeptide called latency-associated peptide (LAP). The TGF-β/LAP complex is noncovalently attached to members of latent TGF-β binding proteins (LTBPs) in most cells forming the large latent complex [23, 29]. However, several cell lines established from malignant tissues have been shown to secrete active form of TGF-β [3, 9, 16]. TGF-β is reported to play a key role in embryonic development, hematopoisis, wound healing, matrix formation, immune function as well as disease states including chronic inflammation, immune disorders and cancer [1, 27]. In the present study, we could detect a biologically active form of TGF-β1 in LS-A tumor-derived supernatants and its elevated serum levels in progressive i.p. tumor bearing mice (Figs. 7, 8, respectively). We have observed decreased tumor cell density in peritoneal ascites, at late stage of tumor growth and proliferation-inhibitory activity of cell-free supernatants of ascites (Data not shown). High concentration of TGF-β1 in i.p. transplanted mice led to host mortality subsequently. The death of i.p. tumor bearing mice may be due to several reasons including (1) tumor load (2) immunosuppression and subsequent infections (3) paracrine effects of TGF-β, etc.

The amount of TGF β1 is very high in i.p. tumor bearing mice. Normally, one would expect a rapid hepatic elimination of this active cytokine. However this may not be so in the progressively growing tumor bearing mice as, among others, hepatic function may also be compromised. Even in clinical situation, e.g., patients with colorectal carcinoma, high levels of TGF β1 (53±12 ng/ml) have been observed before surgery. After tumor resection the level dropped to 36±6 ng/ml indicating elevated TGF β1 as a sign of progressive disease [32]. In other carcinomas, an association was found between high tumor burden and elevated circulating plasma levels of TGF-β1 [1].

This cytokine is reported to impair a number of immune functions such as T cell growth, cytotoxicity and T cell cytokine production necessary for potentiation of CTL activity and antigen presentation [5]. It was also observed that mice bearing i.p. tumors developed progressive hemoglobulinemia which was absent in mice transplanted by s.c. route (Fig. 9a). This could be partly explained by the reduced expression of transferrin receptor (CD71) on the bone marrow cells of i.p. tumor bearing mice as compared to that of s.c. tumor bearing mice (Fig. 9b). Transferrin receptor is a marker on proliferating cells and is involved in uptake of iron loaded transferrin which determines intracellular iron levels [12]. Reduction in the expression of CD71 have been shown in activated human T lymphocytes treated with TGF-β [13] under in vitro conditions. It is possible that LS-A tumor-derived TGF-β1 can impair uptake of transferrin via transferrin receptor down-regulation on the hemopoietic erythroid progenitors and thus contribute to hemoglobulinemia observed in i.p. tumor bearing mice. TGF-β1 or the tumor itself may enhance the tumoricidal activity of macrophages [15, 18] but this mechanism may be more relevant to subcutaneously transplanted tumor.

To conclude, our studies identified TGF-β1 as the endogenous growth factor involved in LS-A growth. It displayed bimodal growth effects on LS-A proliferation in vitro depending on the concentration. The cytokine caused patho-physiological effects like lymphopenia, hemoglobulinemia, reduction in CD71 expression on bone marrow cells and impaired the host T and B cell function by paracrine effects. Tumor progression and mortality in i.p. transplanted mice could be attributed to elevated levels of serum TGF-β1. Further, the data suggested that the tumorigenicity or immunogenicity is dependent on the route of transplantation. This is the first report showing tumor-derived TGF-β1 as a site-specific modulator of murine tumor of spontaneous origin.

Acknowledgements

The authors, thank Shri S. Santosh Kumar for critically going through the manuscript and Shri N.S. Sidnalkar, also, Shri K.S. Munankar for their expert technical help in animal experiments.

References

  • 1.Akhurst RJ, Derynck R. TGF-β signaling in cancer- a double-edged sword. Trends Cell Biol. 2001;11:S44–S51. doi: 10.1016/s0962-8924(01)02130-4. [DOI] [PubMed] [Google Scholar]
  • 2.Balasubramaniam P, Malathi A. Comparative study of hemoglobin estimated by Drabkin’s and Sahli’s methods. J Post Grad Med. 1992;38:8–9. [PubMed] [Google Scholar]
  • 3.Bodmer S, Srommer K, Frei K, Siepl C, de Tribolet N, Heid I, Fontana A. Immunosuppression and transforming growth factor beta in glioblastoma. Preferential production of transforming growth factor beta2. J Immunol. 1989;143:3222–3229. [PubMed] [Google Scholar]
  • 4.Burbridge MF, Kraus-Berthier L, Naze M, Pierre A, Atassi G, Guilbaud N. Biological and pharmacological characterization of three models of human ovarian carcinoma established in nude mice: use of the CA 125 tumor marker to predict antitumor activity. Int J Oncol. 1999;15:1155–1162. doi: 10.3892/ijo.15.6.1155. [DOI] [PubMed] [Google Scholar]
  • 5.Chouaib S, Asselin-Paturel C, Mami-Chouaib F, Caignard A, Blay JY. The host-tumor immune conflict: from immnosuppression to resistance and destruction. Immunol Today. 1997;18:493–497. doi: 10.1016/S0167-5699(97)01115-8. [DOI] [PubMed] [Google Scholar]
  • 6.Czarniecki W, Chiu HH, Wong GH, Mc Cabe SM, Palladino MA. Transforming growth factor beta 1 modulates the expression of class II histocompatibility antigens on human cells. J Immunol. 1988;140:4217–4223. [PubMed] [Google Scholar]
  • 7.De Coteau JF, Knaus PI, Yankelev H, Reis MD, Lowsky R, Lodish HF, Kadin ME. Loss of functional cell surface transforming growth factor beta (TGF-β) type I receptor correlate with insensitivity to TGF-β in chronic lymphocyte leukemia. Proc Natl Acad Sci USA. 1997;94:5877–5881. doi: 10.1073/pnas.94.11.5877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Donelli MG, Russo R, Garattini S. Selective chemotherapy in relation to the site of tumor transplantation. Int J Cancer. 1975;32:78–88. doi: 10.1002/ijc.2910020503. [DOI] [PubMed] [Google Scholar]
  • 9.Fischer JR, Darjes H, Schindel M, Drings P, Krammer PH. Constitutive secretion of bioactive transforming growth factor beta1 by small cell lung cancer cell lines. Eur J Cancer. 1994;30:2125–2129. doi: 10.1016/0959-8049(94)00364-B. [DOI] [PubMed] [Google Scholar]
  • 10.Garba L, Frelinger JA. Intracellular cytokine staining for TGF-beta. J Immunol Methods. 2001;258:193–198. doi: 10.1016/S0022-1759(01)00491-4. [DOI] [PubMed] [Google Scholar]
  • 11.Hamilton TC, Young RC, Louie KG, Behrens BC, McKoy WM, Grotzinger KR. Characterization of a xenografts model of human ovarian carcinoma which produces ascites and intraabdominal carcinomatosis in mice. Cancer Res. 1984;44:5286–5290. [PubMed] [Google Scholar]
  • 12.Hentze MW, Kuhn LC. Molecular control of vertebrate iron metabolism: mRNA-based regulatory circuits operated by iron, nitric oxide and oxidative stress. Proc Natl Acad Sci USA. 1996;93:8175–8182. doi: 10.1073/pnas.93.16.8175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kehrl JH, Wakefield LM, Roberts AB, Jakowlew S, Alvarez-Mon M, Derynck R, Sporn MB, Fauci AS. Production of transforming growth factor β by human T lymphocytes and its potential role in the regulation of T cell growth. J Exp Med. 1986;163:1037–1050. doi: 10.1084/jem.163.5.1037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Keyes KA, Mann L, Teicher B, Alvarez E. Site-dependent angiogenic cytokine production in human tumor xenografts. Cytokine. 2003;21:98–104. doi: 10.1016/S1043-4666(03)00015-2. [DOI] [PubMed] [Google Scholar]
  • 15.Khar A. Characterization of spontaneous immunological rejection in a histiocytoma. I. Role of NK cell and anti tumor antibodies. Int J Oncol. 1993;3:393–398. doi: 10.3892/ijo.2.3.393. [DOI] [PubMed] [Google Scholar]
  • 16.Kloen P, Jennings CL, Gebhardt MC, Springfield DS, Mankin HJ. Expression of transforming growth factor beta (TGF-β) receptors, TGF-β1 and TGF-β2 production and autocrine growth control in osteosarcoma cells. Int J Cancer. 1994;58:440–445. doi: 10.1002/ijc.2910580323. [DOI] [PubMed] [Google Scholar]
  • 17.Kumar M, Seshadri M, Poduval TB, Chauhan PS, Sundaram K. Cytogenetic studies on murine lymphosarcoma. Ad Biosci. 1987;6:35–44. [Google Scholar]
  • 18.Liaw FD, Denkins YM, Collins JH, Van Arsdall M, Chang JL, Chakrabarty S, Nguyen D, Kruzel E, Fidler IJ. Type-1 transforming growth factor-beta differentially modulates tumoricidal activity of murine peritoneal macrophages against metastatic variants of the B16 murine melanoma. J Exp Ther Oncol. 2002;2:286–297. doi: 10.1046/j.1359-4117.2002.01053.x. [DOI] [PubMed] [Google Scholar]
  • 19.Mesiano S, Ferrara N, Jaffe RB. Role of vascular endothelial growth factor in ovarian cancer: Inhibition of ascites formation by neutralization. Am J Pathol. 1998;153:1249–1256. doi: 10.1016/S0002-9440(10)65669-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Morikane K, Tempero R, Sivinski CL, Kitajima S, Gendler SJ, Hollingsworth MA. Influence of organ site and tumor cell type on MUC1-specific tumor immunity. Int Immunol. 2001;13:233–240. doi: 10.1093/intimm/13.2.233. [DOI] [PubMed] [Google Scholar]
  • 21.Morris PC, Scholten V. Osmotic lysis of tumor spill in ovarian cancer: a murine model. Am J Obstet Gynecol. 1996;175:1489–1492. doi: 10.1016/s0002-9378(96)70095-9. [DOI] [PubMed] [Google Scholar]
  • 22.Nakajima M, Morikawa K, Fabra A, Bucana CD, Fidler IJ. Influence of organ environment on extracellular matrix degradative activity and metastasis of human colon carcinoma cells. J Natl Cancer Inst. 1990;82:1890–1898. doi: 10.1093/jnci/82.24.1890. [DOI] [PubMed] [Google Scholar]
  • 23.Oklu R, Hasketh R. The latent transforming growth factor beta binding protein (LTBP) family. Biochem J. 2000;352:601–610. doi: 10.1042/0264-6021:3520601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Papac RJ. Spontaneous regression of cancer. Cancer Treat Rev. 1996;22:395–423. doi: 10.1016/S0305-7372(96)90023-7. [DOI] [PubMed] [Google Scholar]
  • 25.Paterson C, Davies M, Stone A, Huntley S, Smith E, Pring M, Eveson JW, Robinson CM, Parkinson EK, Prime SS. TGF-beta 1 acts as a tumor suppressor of human malignant keratinocytes independently of Smad 4 expression and ligand-induced G(1) arrest. Oncogene. 2002;21:1616–1624. doi: 10.1038/sj.onc.1205217. [DOI] [PubMed] [Google Scholar]
  • 26.Roberts B, Flanders KC, Kondaiah P, Thompson NL, Obberghen- Schilling E, Wakfield L, Rossi P, de Cambrugghe B, Heine U, Sporn MB. growth factor beta: biochemistry and roles in embriogenesis, tissue repair and remodeling and carcinogenesis. Recent Prog Horm Res. 1988;44:157–168. doi: 10.1016/b978-0-12-571144-9.50010-7. [DOI] [PubMed] [Google Scholar]
  • 27.Roberts B, Flanders KC, Heine UI, Jakowlew S, Kondaiah P, Kim SJ, Sporn MB. Transforming growth factor-beta: multifunctional regulator of differentiation and development. Philos Trans R Soc Lond B Biol Sci. 1990;327:145–154. doi: 10.1098/rstb.1990.0050. [DOI] [PubMed] [Google Scholar]
  • 28.Rodeck U, Bosller A, Graeven U, Fox FE, Nowell PC, Knabbe K, Kari C. Transforming growth factor beta production and responsiveness in normal human melanocytes and melanoma cells. Cancer Res. 1994;54:575–581. [PubMed] [Google Scholar]
  • 29.Saharinen J, Hyytiainen M, Taipale J, Keski-Oja J. Latent transforming growth factor beta binding proteins (LTBPs)—structural extracellular matrix proteins for targeting TGF beta action. Cytokine Growth Factor Rev. 1999;10:99–117. doi: 10.1016/S1359-6101(99)00010-6. [DOI] [PubMed] [Google Scholar]
  • 30.Schomaker SJ, Clemo FA, Amacher DE. Analysis of rat bone marrow by flow cytometry following in vivo exposure to cyclohexanone oxime or daunomycin HCl, Toxicol. Appl Pharmacol. 2002;185:48–54. doi: 10.1006/taap.2002.9512. [DOI] [PubMed] [Google Scholar]
  • 31.Schwarz RE, McCarty TM, Peralta EA, Diamond DJ, Ellenhorn JD. An orthotopic in vivo model of human pancreatic cancer. Surgery. 1999;126:562–567. doi: 10.1067/msy.2099.100267. [DOI] [PubMed] [Google Scholar]
  • 32.Shim KS, Kim KH, Han WS, Park EB. Elevated serum levels of transforming growth factor-beta1 in patients with colorectal carcinoma: its association with tumor progression and its significant decrease after curative surgical resection. Cancer. 1999;85:554–561. doi: 10.1002/(SICI)1097-0142(19990201)85:3&#x0003c;554::AID-CNCR6&#x0003e;3.0.CO;2-X. [DOI] [PubMed] [Google Scholar]
  • 33.Staroselsky N, Fan D, O’Brian CA, Bucana CD, Gupta KP, Fidler IJ. Site-dependent differences in response of the UV-2237 murine fibrosarcoma to systemic therapy with adriamycin. Cancer Res. 1990;50:7775–7782. [PubMed] [Google Scholar]
  • 34.Tan MH, Chu TM. Characterization of the mouse tumorigenic and metastatic properties of a human pancreatic tumor cell line (AsPC-1) implanted orthotopically into nude mice. Tumor Biol. 1985;6:89–98. [PubMed] [Google Scholar]
  • 35.Thakur VS, Seshadri M, Shah DH, Poduval TB. Adoptive immunotherapy of murine lymphosaroma: Comparative study using allo-immune or tumor-immune T cells. Ind J Exp Biol. 1997;35:581–588. [PubMed] [Google Scholar]
  • 36.Vanky F, Nagy N, Hising C, Sjovall K, Larson B, Klein E. Human ex vivo carcinoma cells produce transforming growth factor-β and thereby can inhibit lymphocyte function in vitro. Cancer Immnol Immunother. 1997;43:317–323. doi: 10.1007/s002620050339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Wilmanns C, Fan D, O’Brian CA, Bucana CD, Fidler IJ. Orthotopic and ectopic organ environments differentially influence the sensitivity of murine colon carcinoma cells to doxorubicin and 5-fluorouracil. Int J Cancer. 1992;52:98–104. doi: 10.1002/ijc.2910520118. [DOI] [PubMed] [Google Scholar]

Articles from Cancer Immunology, Immunotherapy : CII are provided here courtesy of Springer

RESOURCES