Transforming growth factor Β (TGF-β) is a potent immunosuppressive cytokine that is frequently associated with mechanisms of tumor escape from immunosurveillance. We report that transplantation of murine bone marrow (BM) expressing a dominant-negative TGF-β type II receptor (TβRIIDN) leads to the generation of mature leukocytes capable of a potent antitumor response in vivo. Hematopoietic precursors in murine BM from donor mice were rendered insensitive to TGF-β via retroviral expression of the TβRIIDN construct and were transplanted in C57BL/6 mice before tumor challenge. After i.v. administration of 5 × 105 B16-F10 murine melanoma cells into TβRIIDN-BM transplanted recipients, survival of challenged mice at 45 days was 70% (7 of 10) versus 0% (0 of 10) for vector-control treated mice, and surviving TβRIIDN-BM mice showed a virtual absence of metastatic lesions in the lung. We also investigated the utility of the TGF-β-targeted approach in a mouse metastatic model of prostate cancer, TRAMP-C2. Treatment of male C57BL/6 mice with TβRIIDN-BM resulted in the survival of 80% (4 of 5) of recipients versus 0% (0 of 5) in green fluorescent protein-BM recipients or wild-type controls. Cytolytic T-cell assays indicate that a specific T-cell response against B16-F10 cells was generated in the TβRIIDN-BM-treated mice, suggesting that a gene therapy approach to inducing TGF-β insensitivity in transplanted BM cells may be a potent anticancer therapy.

Tumor immunotherapies to date have focused largely on the priming of immune responses to fight cancer, with mixed results and generally poor efficacy. In addition to immune stimulation, the issue of overcoming active immune suppression must also be considered when developing an immune-based strategy for cancer therapy (1, 2), particularly with regard to secreted soluble factors that are known to down-regulate immune function and antitumor response. Most significant of these is the pleiotropic cytokine TGF-β,3(3) which has previously been shown to act in a critical inhibitory fashion on most cells of the immune system and is secreted by a wide variety of tumor types, many of which down-regulate expression of their own TGF-β receptors (4, 5, 6, 7) to circumvent the growth-inhibitory activity of TGF-β signaling. Tumor-secreted TGF-β is capable of inhibiting the response of tumor-specific lymphocytes (8), including sites of metastatic tumor growth (9). The potency of TGF-β as an immunosuppressive cytokine makes it an attractive target as an anticancer therapy, because, as it has been suggested, a breakdown of self-tolerance mechanisms in the periphery may be a critical element in fighting nonimmunogenic tumors (10). We hypothesized that an immunotherapy strategy that specifically blocks TGF-β signaling in immune cells, regardless of tumor location or tumor microenvironment, could be highly successful in mediating an antitumor response.

We chose to use a retroviral-mediated gene therapy approach abrogating TGF-β signaling in hematopoietic stem cells in the BM, because this approach has been shown recently to be a successful protocol in the delivery of long-term transgene expression in immune effector cells (11). Here, we show that abrogation of TGF-β signaling in the immune compartment via retrovirus-mediated expression of a TβRIIDN in transplanted BM-derived stem cells elicits potent antitumor activity when treated animals are challenged i.v. with highly tumorigenic melanoma or prostate cancer cells.

Mice.

Male C57BL/6 mice, 6–8 weeks of age, were obtained from Jackson Labs (Bar Harbor, ME) and maintained in pathogen-free facilities at the Center for Comparative Medicine at Northwestern University Feinberg School of Medicine in accordance with established guidelines of the Animal Care and Use Committee of Northwestern University.

BM Isolation and Culture.

Donor mice were inhalation-anesthetized and were given injections i.p. of 5 mg of 5-fluorouracil (Sigma, St. Louis, MO). Five days later, mice were sacrificed by cervical dislocation and hind femora and tibiae were isolated and cleaned of tissue before being flushed aseptically with DMEM plus 10% fetal bovine serum (DMEM-10) using 26-gauge needles. The RBCs in the marrow preparation were then lysed using a hypotonic ammonium chloride solution (PharMingen, Becton-Dickinson, San Diego, CA). The processed marrow was resuspended in fresh DMEM-10 supplemented with 100 ng/ml stem cell factor, 50 ng/ml IL-6, and 20 ng/ml IL-3 (R&D, Minneapolis, MN) at 1–2 × 106 cells/ml, and were incubated at 37°C/5% CO2.

Construction of TβRIIDN-GFP Retroviral Vector.

The procedure for the construction of the TβRIIDN viral vector has been described earlier (12). Briefly, a truncated sequence of the human TGF-β type II receptor was cloned into a mouse stem-cell virus-based bicistronic retroviral vector coexpressing GFP under the control of the 5′ long terminal repeat viral promoter. The truncated receptor contained both the extracellular domain and the transmembrane domain but lacked the cytoplasmic kinase domain. The control empty vector was designated as the GFP vector.

Production of Infectious TβRIIDN-GFP Retrovirus.

Pantropic GP293 retroviral packaging cells (Clontech, San Diego, CA) were seeded at a density of 2.5 × 106 cells in collagen-I-coated T-25 flasks (BIOCOAT; BD Biosciences, Mountain View, CA) 24 h before plasmid transfection in antibiotic-free DMEM-10, such that the cells were ∼70–90% confluent at the time of transfection, at which point the cells were rinsed with PBS to remove residual serum. A mixture of 2 μg of retroviral plasmid and 2 μg of VSV-G envelope plasmid were cotransfected in serum-free DMEM using LipofectAMINE-Plus (Invitrogen, Gaithersburg, MD) according to the manufacturer’s protocols with the following modifications. Cells were transfected for 12 h followed by the addition of an equivalent volume of DMEM-20 and reincubation for an additional 12 h. After 24 h of total transfection time, the supernatant was aspirated, the cells were rinsed gently in PBS, and 3 ml of fresh DMEM-10 was added to each flask. After 24 h, virus-containing supernatant was collected and used to infect target cells.

Western Blotting for SMAD-2 Phosphorylation.

The infected primary mouse BM cells were treated with or without 10 ng/ml TGF-β1 for 30 min in culture to test the functionality of the TGFβ signaling pathway (12). Proteins in the cell lysate were subjected to electrophoresis (Novex/10% acrylamide gel) and blotted onto a polyvinylidene difluoride membrane. Blots were probed using monoclonal antibody against phosphorylated SMAD-2. Blots were stripped and reprobed with antibodies against SMAD-2 and then glyceraldehyde-3-phosphate dehydrogenase (GAPDH).

Retroviral Infection and Transplantation of Murine BM.

Cultured murine BM cells were infected on days 2 and 3 postisolation via spin infection as follows: an aliquot of 1 ml of viral supernatant was added to each well of a 24-well plate containing BM cells in the presence of a minimum concentration of 4 μg/ml Polybrene (Sigma), spun at 1000 × g for 90 min, and supplemented with 1 ml of fresh cytokine-supplemented DMEM-10. On day 4–5, cells were examined for GFP expression, washed two times in PBS, and injected into the lateral warmed tail veins of irradiated (1200 rads) recipient C57BL/6 mice. Transplanted mice were maintained on sulfamethoxazole/trimethoprim for a minimum of 2 weeks to prevent opportunistic infection.

i.v. Inoculation of Tumor Cells into Mice after BM Transplant.

C57Bl/6 mice receiving TβRIIDN, GFP, or nontransduced BM transplants were challenged i.v. with 5 × 105 B16-F10 cells (n = 10 mice/group) or TRAMP-C2 cells (n = 5 animals/group) 2 months after transplant. The B16-F10-challenged mice were monitored for morbidity and mortality for 6 weeks, and the TRAMP-C2-challenged mice were monitored for 8 weeks. At the conclusion of each experiment, all of the animals were inspected for the presence of metastases. Statistical analysis was conducted on a Kaplan-Meier survival curve, using the log-rank test (13).

Functional Status of TGF-β Signaling in Transfected BM.

Transfection efficiency into primary BM cells using the above approach was consistently greater than 90% as assayed by GFP expression (12). Results of the functional analysis of these transfected BM cells have been reported earlier (12). Briefly, when the TβRIIDN BM cells were treated with 10 ng/ml TGF-β1 in culture, the expression of the dominant negative receptor resulted in an absence of SMAD-2 phosphorylation. SMAD-2 phosphorylation was observed in similarly treated mock-infected cells or cells infected with the control vector expressing GFP alone (12). Furthermore, at 6-months posttransplant, the results of flow cytometry data indicated that there was no significant reduction of GFP expression in BM cells of either TβRIIDN- or GFP-transduced mice (12).

Increased Survival and Decreased Metastases in TβRIIDN-BM-treated Mice.

C57BL/6 mice receiving TβRIIDN, GFP, or nontransduced BM transplants (n = 10 mice/group) were challenged with 5 × 105 B16-F10 cells i.v. and monitored for morbidity and mortality for a period of ∼6 weeks. Whereas 100% of wild-type and GFP transplant recipients were dead by 22 days postchallenge, there was no mortality observed in the TβRIIDN-BM recipient group (Fig. 1,A) by this time. The TβRIIDN-BM control group was monitored for a total period of 45 days postchallenge, at which point surviving (7 of 10) mice were sacrificed and their lung tissue removed for macroscopic examination to determine whether metastatic lesions comparable with those observed in the wild type-BM and GFP-BM control groups were present. As shown in Fig. 1 B, the lung tissue of untreated control mice was characterized at the time of death by numerous black melanoma metastases throughout the tissue. However, the TβRIIDN-BM-treated group had fewer metastatic lesions in the lungs of nonsurviving mice and virtually no discernable lesions in the lungs of mice surviving throughout the duration of the experiment. These results strongly suggest that mice transplanted with BM with targeted blockade of TGF-β signaling generate potent antitumor immunity in C57BL/6 mice challenged with highly metastatic, nonimmunogenic tumor cells.

To determine the efficacy of the TβRIIDN-BM treatment on metastatic tumor formation in a model of prostate cancer, we subsequently challenged TβRIIDN-BM treated male C57BL/6 mice with i.v. administration of 5 × 105 TRAMP-C2 cells and monitored the mice similarly as described above. At 3 weeks postchallenge, macroscopic tumor formation was difficult to detect in either the treated or untreated controls, indicating that the TRAMP-C2 tumor cells were not as aggressive in their formation of metastatic lung foci as were the B16-F10 tumor cells. However, on further examination of histological specimens of mice sacrificed at 21 days post tumor challenge, micrometastatic lesions were already visible in the GFP group but not in the TβRIIDN group (data not shown). A second group of mice was tumor challenged and monitored for a period of 8 weeks, by which point the survival of the wild-type and GFP control mice was 0% (0 of 5, each group by week 7; Fig. 2,A), whereas the survival of the TβRIIDN-BM treated cohort was 100% (5 of 5). By week 9, one animal in the TβRIIDN-BM group died, leaving the overall survival rate of 80% (4 of 5) for this group. Results of statistical analysis, using the log-rank test, indicated P < 0.05 between the TβRIIDN-BM and the other two control groups. Postmortem analysis of the untreated or vector-control-treated animals indicated a significant tumor burden evident in the lung tissue of each mouse (Fig. 2 B), whereas the lungs of TβRIIDN mice remained metastases free. From these data, we conclude that targeting immune TGF-β signaling with BM-directed retroviral therapy is an effective means of preventing metastatic prostate tumor growth in mice.

TβRIIDN Mice Generate Specific Antitumor CTLs in Vivo.

To determine whether the antitumor response generated by transplant of TβRIIDN-BM is tumor-specific, we collected splenocytes from TβRIIDN-BM- and GFP-BM-tumor-challenged mice at 3 weeks post-tumor challenge and assayed the ability of CTLs to lyse B16-F10 cells in vitro using a standard 51Cr release assay. Results from the CTL assay indicated a significant increase in tumor-specific lysis of melanoma cells in splenocytes from TβRIIDN-BM-transplanted mice compared with GFP control-treated counterparts (Fig. 3,A), suggesting that the antitumor phenotype in TGF-β signaling pathway-deficient mice is at least partially caused by CTL activity and not simply a result of broader, nonspecific immune stimulation of treated mice. Likewise, a 51Cr release assay performed on labeled TRAMP-C2 cells by splenocytes recovered from TβRIIDN-BM- and GFP-BM-transplanted mice (Fig. 3 B) indicate that tumor-specific cytolysis is generated by the retroviral blockade of TGF-β signaling.

Results of the present study demonstrate that disruption of the TGF-β signaling pathway in BM cells using a gene therapy approach confers an antitumor phenotype on treated mice. Targeting of TGF-β-mediated immunosuppression has been used previously to show that the blockade of normal TGF-β signaling pathways confers an antitumor effect in a variety of tumor models, either via modulation of tumor TGF-β production in a tumor vaccine approach or via the systemic down-regulation of available TGF-β cytokine in the serum, and has been used in a variety of tumor therapies to combat both primary and secondary tumor growth. Ex vivo transfer of an antisense TGF-β construct into isolated tumor cells followed by reimplantation into the brain of rats with established gliomas has been shown to result in complete eradication of the tumors in vivo(14), and a similar approach has been used successfully to confer immunogenicity to a prostate tumor model in the Dunning rat (15). Systemic administration of anti-TGF-β antibody and IL-2 shows a significant decrease in number and size of metastatic B16 tumor lesions (16), suggesting that TGF-β immunosuppression can be at least partially overcome by a general TGF-β signal blockade. This latter approach, including similar approaches such as soluble TGF-β type II receptor therapy (17), although providing a rationale for a TGF-β-targeted approach in cancer therapy, may be ultimately limited in its ability to mediate antitumor effects at sites in which the delivery of a soluble therapeutic agent may be insufficient to block TGF-β present at high concentrations in tumor microenvironments.

In the present study, we demonstrated the therapeutic efficacy of targeting progenitors of leukocyte populations in the BM with retroviral particles that specifically blocked TGF-β signaling by expressing a dominant negative TGF-β type II receptor with a truncated cytoplasmic domain. The lack of formation of metastatic lesions in TβRIIDN-BM-treated mice after i.v. administration with highly metastatic B16-F10 cells emphasizes the importance of the TGF-β signaling pathway to tumorigenicity in vivo, even in the case of tumor cells with aggressive growth properties and little natural immunogenicity. Likewise, the lack of metastatic lesion formation in TβRIIDN-BM-treated animals after a challenge with TRAMP-C2 cells, a murine model of prostate cancer, supports the idea that this antitumor approach is viable in a range of cancers of different tissue origins.

The potency of TGF-β as an immunoregulatory cytokine that is critical for the maintenance of immune homeostasis also necessitates the careful application of perturbations in the TGF-β signaling processes for cancer immunotherapy. The potential for the generation of widespread autoimmunity and inflammation, which is generated in the absence of functional TGF-β pathways in immune cells (12), makes it essential that the approach described here be maximized for its utility as an antitumor therapy but modified so as to minimize potential autoimmune side effects against host tissue. Mice that are deficient in TGF-β1 cytokine display a massive auto-inflammatory phenotype and quickly succumb to systemic damage in a variety of tissues (18, 19), whereas other transgenic models, restricted to TGF-β-signal abrogation in the immune compartment or single lineages including T (20) and B cells (21), similarly result in dysregulation of immune function. The retroviral approach to therapeutic gene delivery can be enhanced by vectors that offer a regulatory mechanism to control expression of the transgene and/or survival of transgene-positive cells, whether through the use of on/off systems responsive to pharmacological agents (e.g., tetracycline) or through the use of suicide gene elements present in the integrated viral genome.

We submit that the results presented here represent a viable approach to the problem of tumor escape from immune surveillance using readily available retroviral gene transfer technology, and we suggest that this approach could potentially be coupled with other immunostimulatory protocols that generate tumor-specific lymphocyte responses but that, to date, have had only mixed results because of a lack of cytotoxic effector activity, particularly with regard to distant metastatic tumor foci, as a result of TGF-β-mediated immunosuppression. The hematopoietic stem-cell gene therapy approach, already established as a viable means for the delivery of therapeutic genes to cells of the immune system, provides a legitimate and characterized target for TGF-β signaling-directed therapy for a potentially wide variety of cancers.

Fig. 1.

Antitumor capacity of mice receiving transplant of TβRIIDN-BM. A, Kaplan-Meier survival curve of C57BL/6 mice challenged with 5 × 105 B16-F10 melanoma cells via tail vein injection after transplantation with 2–4 × 106 syngeneic BM cells transduced with TβRIIDN-expressing retrovirus, GFP control virus, or uninfected wild-type BM cells (n = 10/group; P < 0.01 by the log-rank test for the TβRIIDN group versus GFP or control group; Ref. 13). B, lungs of mice 3 weeks post-tumor challenge from TβRIIDN-transplanted mice (left) or GFP control mice (right). The GFP control lung is covered with black, melanin-producing tumor cells. The lung in the TβRIIDN-treated group is devoid of any tumor.

Fig. 1.

Antitumor capacity of mice receiving transplant of TβRIIDN-BM. A, Kaplan-Meier survival curve of C57BL/6 mice challenged with 5 × 105 B16-F10 melanoma cells via tail vein injection after transplantation with 2–4 × 106 syngeneic BM cells transduced with TβRIIDN-expressing retrovirus, GFP control virus, or uninfected wild-type BM cells (n = 10/group; P < 0.01 by the log-rank test for the TβRIIDN group versus GFP or control group; Ref. 13). B, lungs of mice 3 weeks post-tumor challenge from TβRIIDN-transplanted mice (left) or GFP control mice (right). The GFP control lung is covered with black, melanin-producing tumor cells. The lung in the TβRIIDN-treated group is devoid of any tumor.

Close modal
Fig. 2.

TβRIIDN-BM-treated mice showing antitumor capacity against TRAMP-C2 mouse prostate cancer tumor challenge. 5 × 105 TRAMP-C2 prostate adenocarcinoma cells were injected via the tail vein into TβRIIDN-BM-treated mice, and the mice were monitored for morbidity and mortality. A, survival of wild-type (untreated), GFP, and TβRIIDN-transplanted mice post-tumor challenge (n = 5/group), expressed as the Kaplan-Meier curve. (P < 0.05 by the log-rank test for the TβRIIDN group versus the control or GFP group; Ref. 13). B, lung tissue from TβRIIDN-BM- and GFP-BM-treated mice at 6 weeks post-tumor challenge indicating metastatic tumor foci (arrows).

Fig. 2.

TβRIIDN-BM-treated mice showing antitumor capacity against TRAMP-C2 mouse prostate cancer tumor challenge. 5 × 105 TRAMP-C2 prostate adenocarcinoma cells were injected via the tail vein into TβRIIDN-BM-treated mice, and the mice were monitored for morbidity and mortality. A, survival of wild-type (untreated), GFP, and TβRIIDN-transplanted mice post-tumor challenge (n = 5/group), expressed as the Kaplan-Meier curve. (P < 0.05 by the log-rank test for the TβRIIDN group versus the control or GFP group; Ref. 13). B, lung tissue from TβRIIDN-BM- and GFP-BM-treated mice at 6 weeks post-tumor challenge indicating metastatic tumor foci (arrows).

Close modal
Fig. 3.

Generation of tumor-specific killing in TβRIIDN-BM-transplanted mice. Splenocytes from tumor-challenged mice were collected and stimulated for 5 days with irradiated B16-F10 mouse melanoma cells (A) or with TRAMP-C2 mouse prostate carcinoma cells (B) before being cocultured with 51Cr-labeled targets at the indicated E:T ratios. Samples were analyzed in duplicate (A) or triplicate (B) wells.

Fig. 3.

Generation of tumor-specific killing in TβRIIDN-BM-transplanted mice. Splenocytes from tumor-challenged mice were collected and stimulated for 5 days with irradiated B16-F10 mouse melanoma cells (A) or with TRAMP-C2 mouse prostate carcinoma cells (B) before being cocultured with 51Cr-labeled targets at the indicated E:T ratios. Samples were analyzed in duplicate (A) or triplicate (B) wells.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported in part by GrantDAMD17-99-1-9009 from the Department of Defense.

3

The abbreviations used are: TGF-β, transforming growth factor β; GFP, green fluorescent protein; TβRIIDN, dominant negative type II TGF-β receptor; BM, bone marrow; TRAMP, TGF-β-targeted approach in a mouse metastatic model of prostate cancer; IL, interleukin.

1
Wojtowicz-Praga S. Reversal of tumor-induced immunosuppression: a new approach to cancer therapy[see comments].
J. Immunother.
,
20
:
165
-177,  
1997
.
2
Shah A. H., Lee C. TGF-β-based immunotherapy for cancer: breaching the tumor firewall.
Prostate
,
45
:
167
-172,  
2000
.
3
Letterio J. J., Roberts A. B. Regulation of immune responses by TGF-β.
Annu. Rev. Immunol.
,
16
:
137
-161,  
1998
.
4
Kim I. Y., Ahn H. J., Zelner D. J., Shaw J. W., Sensibar J. A., Kim J. H., Kato M., Lee C. Genetic change in transforming growth factor β (TGF-β) receptor type I gene correlates with insensitivity to TGF-β1 in human prostate cancer cells.
Cancer Res.
,
56
:
44
-48,  
1996
.
5
Guo Y., Jacobs S. C., Kyprianou N. Down-regulation of protein and mRNA expression for transforming growth factor-β (TGF-β1) type I and type II receptors in human prostate cancer.
Int. J. Cancer
,
71
:
573
-579,  
1997
.
6
Kim Y. S., Yi Y., Choi S. G., Kim S. J. Development of TGF-β resistance during malignant progression.
Arch. Pharm. Res. (Seoul)
,
22
:
1
-8,  
1999
.
7
Park K., Kim S. J., Bang Y. J., Park J. G., Kim N. K., Roberts A. B., Sporn M. B. Genetic changes in the transforming growth factor β (TGF-β) type II receptor gene in human gastric cancer cells: correlation with sensitivity to growth inhibition by TGF-β.
Proc. Natl. Acad. Sci. USA
,
91
:
8772
-8776,  
1994
.
8
Inge T. H., Hoover S. K., Susskind B. M., Barrett S. K., Bear H. D. Inhibition of tumor-specific cytotoxic T-lymphocyte responses by transforming growth factor β1.
Cancer Res.
,
52
:
1386
-1392,  
1992
.
9
Lee H. M., Timme T. L., Thompson T. C. Resistance to lysis by cytotoxic T cells: a dominant effect in metastatic mouse prostate cancer cells.
Cancer Res.
,
60
:
1927
-1933,  
2000
.
10
Nanda N. K., Sercarz E. E. Induction of anti-self-immunity to cure cancer.
Cell
,
82
:
13
-17,  
1995
.
11
Cavazzana-Calvo M., Hacein-Bey S., de Saint Basile G., Gross F., Yvon E., Nusbaum P., Selz F., Hue C., Certain S., Casanova J. L., Bousso P., Deist F. L., Fischer A. Gene therapy of human severe combined immunodeficiency (SCID)-X1 disease[see comments].
Science (Wash. DC)
,
288
:
669
-672,  
2000
.
12
Shah A. H., Tabayoyong W. B., Kimm S. Y., Kim S.-J., van Paris L., Lee C. Reconstitution of lethally irradiated adult mice with dominant negative TGF-β type II receptor-transduced bone marrow leads to myeloid expansion and inflammatory disease.
J. Immunol.
,
169
:
3485
-3491,  
2002
.
13
Freedman L. S. Tables of the number of patients required in clinical trials using the log rank test.
Stat. Med.
,
1
:
121
-129,  
1982
.
14
Fakhrai H., Dorigo O., Shawler D. L., Lin H., Mercola D., Black K. L., Royston I., Sobol R. E. Eradication of established intracranial rat gliomas by transforming growth factor β antisense gene therapy.
Proc. Natl. Acad. Sci. USA
,
93
:
2909
-2914,  
1996
.
15
Matthews E., Yang T., Janulis L., Goodwin S., Kundu S. D., Karpus W. J., Lee C. Down-regulation of TGF-β1 production restores immunogenicity in prostate cancer cells.
Br. J. Cancer
,
83
:
519
-525,  
2000
.
16
Wojtowicz-Praga S., Verma U. N., Wakefield L., Esteban J. M., Hartmann D., Mazumder A., Verma U. M. Modulation of B16 melanoma growth and metastasis by anti-transforming growth factor β antibody and interleukin-2[erratum appears in J. Immunother. Emph. Tumor Immunol., 19: 386, 1996].
J. Immunother. Emph. Tumor Immunol.
,
19
:
169
-175,  
1996
.
17
Won J., Kim H., Park E. J., Hong Y., Kim S. J., Yun Y. Tumorigenicity of mouse thymoma is suppressed by soluble type II transforming growth factor β receptor therapy.
Cancer Res.
,
59
:
1273
-1277,  
1999
.
18
Kulkarni A. B., Huh C. G., Becker D., Geiser A., Lyght M., Flanders K. C., Roberts A. B., Sporn M. B., Ward J. M., Karlsson S. Transforming growth factor β1 null mutation in mice causes excessive inflammatory response and early death.
Proc. Natl. Acad. Sci. USA
,
90
:
770
-774,  
1993
.
19
Shull M. M., Ormsby I., Kier A. B., Pawlowski S., Diebold R. J., Yin M., Allen R., Sidman C., Proetzel G., Calvin D., Annunziata N., Doetschman T. Targeted disruption of the mouse transforming growth factor-β1 gene results in multifocal inflammatory disease.
Nature (Lond.)
,
359
:
693
-699,  
1992
.
20
Lucas P. J., Kim S. J., Melby S. J., Gress R. E. Disruption of T cell homeostasis in mice expressing a T cell-specific dominant negative transforming growth factor β II receptor.
J. Exp. Med.
,
191
:
1187
-1196,  
2000
.
21
Cazac B. B., Roes J. TGF-β receptor controls B cell responsiveness and induction of IgA in vivo.
Immunity
,
13
:
443
-451,  
2000
.