Antilymphocytic antibodies and marrow transplantation. IX. T-cell depletion in marrow donors with C1q high and low affinity antibodies for suppression of GVHD in fully mismatched mice.

Rat monoclonal antibodies (mAbs) of the same specificity (anti-Thy-1) but different immunoglobulin (Ig) subclass were investigated for their effect on suppression of graft-versus-host disease (GVHD) by depleting the marrow donors of T cells in vivo. Transplantation to homozygous, fully mismatched mice of spleen and bone marrow cells from unthymectomized mice injected with the mAbs revealed that two rat anti-Thy-1 mAbs with high affinity for C1q (IgG2b) suppressed and prevented acute and chronic mortality of GVHD. In contrast, rat mAbs with low affinity for C1q (IgM, IgG2c) barely delayed acute mortality. This correlated with findings on the degree of splenic T-cell depletion in donor mice with the IgG2b mAb, able to deplete 97%, and the IgG2c and IgM mAbs, only 83% and 75% of T cells, respectively. An effect akin to the one achieved with IgG2b was seen, however, when donor mice were thymectomized and then treated with three injections of IgG2c isotype. The rat IgM mAb was not immunosuppressive even under such conditions. Immunocytochemical and immunohistochemical examination of the donor lymph nodes after single injection of either mAb showed that only 84% of T cells were eliminated, and in contrast to the spleen, none of the tested antibodies could deplete T cells further. The thymus did not appear depleted at all, although the cortical thymocytes were coated with either of the injected mAbs.

Myeloid membrane markers. I. Expression of marker antigens specific for murine granulocytes and their precursor cells.

Bone marrow cells were investigated by immunoelectromicroscopy and by quantitative photometric immunoradioautography with a rabbit antiserum against murine bone marrow cells. The serum was absorbed with murine spleen, liver and thymus cells until it no longer reacted with thymocytes and lymph node cells in a quantitative complement fixation test. The antiserum stained granulopoietic but not erythropoietic or lymphopoietic cells. The density of the myeloid antigen on single cells increased with the differentiation from immature to mature granulopoietic cells. While the increase of label was statistically not significant at the level of differentiation from promyelocyte to myelocyte and metamyelocyte to band neutrophil, there was a remarkable gain of label from myeloblast to promyelocyte and from band neutrophil to segmented neutrophil. This was evident under the electron microscopy using peroxidase-labeled antibodies and could be measured quantitatively with photometric immunoradioautography using 125I-labeled antibodies.

Monoclonal antibodies that bind to hemopoietic stem cells: characterization and immunohistochemical localization of cells expressing gp50-65.

The cell surface of hemopoietic stem cells has been shown to express several antigens in common with more mature hemopoietic cells. One set of stem cell antigens is defined by a group of three monoclonal antibodies (13C6, 1C10, and 1A9), selected on the basis of binding to subpopulations of spleen colony-forming stem cells (CFU-S). These antibodies are shown to recognize cell surface glycoproteins of 50-65 kd (gp50-65) that occur widely on hemopoietic cells. Each cell type investigated shows a distinctive pattern of expression of these glycoproteins. To further investigate the presence of gp50-65 on stem cells, low-density bone marrow cells were labeled with 1C10 and sorted according to fluorescence intensity. Most (73%) of the stem cells (CFU-S10) were recovered in the two most highly fluorescent fractions containing 2.6% of starting marrow cells. Immunohistochemistry of frozen sections of normal spleen and spleens during repopulation after lethal irradiation and bone marrow transplantation showed that the most strongly 1C10-labeled cells occurred under the splenic capsule and along trabeculae. Although many of these cells were also alpha-naphthylacetate esterase positive and Mac-1 positive, indicating cells of the monocyte-macrophage lineage, a distinctive population of singular cells were stained with 1C10 alone. These cells were negative for surface Ig and closely corresponded with a small population of cells in a similar location that were doubly labeled with 1C10 and anti-Thy-1. These results show that stem cells express high levels of gp50-65 and suggest that stem cells can be identified by immunohistochemical methods using dual labeling procedures.

Serological identification and separation of canine lymphocyte subpopulations.

In vitro treatment of bone marrow grafts with absorbed rabbit-antidog thymocyte globulin (ATG) prevented graft-versus-host disease in a substantial number of the dogs differing by one DLA haplotype. Absorbed ATG has now been used for serological identification of canine lymphocyte populations. Specific labeling of canine T-lymphocytes by absorbed ATG could be demonstrated by (a) a distribution of ATG-positive cells in suspensions of canine lymphoid organs similar to that of T cells observed in other species and by specific staining of paracortical thymus-dependent lymph node areas in immunohistochemistry, (b) complementary labeling of nylon-wool-separated cells by ATG and antiimmunoglobulin sera, and (c) correlation of ATG surface labeling with responder activities in mixed lymphocyte cultures.

Immunohistochemical identification of T- and B-lymphocytes delineated by the unlabeled antibody enzyme method. I. Anatomical distribution of theta-positive and Ig-positive cells in lymphoid organs of mice.

The unlabeled antibody enzyme method was used to delineate the anatomical distribution of lymphocytes positive with a MBtheta and a-MIg in tissue sections of thymus, spleen, lymph nodes and Peyer's patches of mice. The known T- and B-cell areas were established. Moreover, individual B-cells are detected in T-cell regions, especially in the thymus medulla, and in the periarteriolar section of spleen white pulp. Similarly, individual T-cells occur in B-cell areas, namely in the marginal zone of spleen white pulp, in the medullar cores of lymph nodes, and in germinal centers.

A mouse model for the preclinical evaluation of immunosuppressive effector functions of human isotypes. The human IgG1 isotype is superior to IgG3.

Antibodies against T cells are widely used as immunosuppressive agents in clinical therapy. As effector functions of chimeric or humanized anti-T cell antibodies cannot be predicted in vitro, we compared T cell-depleting effects of human isotypes in vivo with their immunosuppressive consequences in a mouse BMT model. This system is based on chimeric antibodies with a mouse pan T cell specificity and human constant regions. To secure optimal immunosuppression, the specificity for Thy-1.2--one of the best-characterized T cell antigens--was selected, as Thy-1.2-specific antibodies prevent graft-versus-host disease in fully mismatched mice. Chimeric mouse anti-Thy-1.2 antibody with the human IgG1 Fc part was found to be equally effective in preventing graft-versus-host disease mortality as the highly protective anti-Thy-1.2 mouse IgG2a isotype, while human IgG3 was far less effective. This was not predictable by measuring the degree of T cell depletion in peripheral blood. T cell depletion in lymph nodes, however, exactly reflected the results obtained in the BMT system. In addition, this system offers the advantage of assessing the influence of reduced antigen density by using heterozygous Thy-1.2 mice.

Antilymphocytic antibodies and bone marrow transplantation. XI. Evidence that reduced Thy-1 expression in Thy-1.1 mice prevents suppression of graft-versus-host disease with anti-Thy-1 monoclonal antibodies.

Considerable variations in the suppression of graft-versus-host disease with monoclonal anti-Thy-1 antibodies were found to relate to substantial differences noted in the expression of mouse Thy-1 marker on lymph node and spleen cells of Thy-1.1 (AKR/J, C57BL/6.Thy-1.1) and Thy 1.2 (AKR/Cu, C57BL/6) mice. Thy-1.1 mice showed a population of 22% (AKR/J) or 13% (C57BL/6-Thy-1.1) of Thy-1 negative cells among peripheral T cells carrying Ly-1 marker. This was in sharp contrast with Thy-1.2 mice, where as expected practically all peripheral T cells expressed both Thy-1 and Ly-1. Double-marker analysis on FACScan revealed that the Thy-1-/Ly-1+ cell population identified in Thy-1.1 but not in Thy-1.2 mice doubtless represents T cells because they express CD3 and either the L3T4 (CD4) or Lyt2 (CD8) phenotype. Using quantitative fluorescence-measurement techniques, it was found in addition that the Thy-1 antigen-binding sites on Thy-1+ cells from Thy-1.1 mice are considerably fewer than those present in Thy-1.2 mice. In fact, the Thy-1 antigen-binding sites approximate the level of Ly-1 density. Consequences of the reduced expression of Thy-1 became apparent in vivo: (1) lymphnode and splenic T cell areas in Thy-1.1 mice were clearly less depleted when Thy-1.1 and Thy-1.2 mice had been injected with rat IgG2b anti-Thy-1 mAb; and (2) GvHD was prevented completely in fully mismatched mice by anti-Thy-1 mAb if the donor mice expressed Thy-1.2 but was barely delayed if the donors expressed Thy-1.1. Thus the present study provides a transplantation model for comparing differences in T antigen density and their consequences for antibody-induced immunosuppression.

Recognition of two epitopes of an antigen present on canine T cells but not on hemopoietic progenitors by four monoclonal antibodies.

Pairs of murine monoclonal antibodies, which recognize 2 different epitopes on a single antigen are described. These antibodies (MdT-P1, -P2, -Q1, -Q2) defining a canine pan-T cell antigen, were raised against dog thymocytes. In immunoblotting of solubilized and polyacrylamide gradient gel electrophoresis in sodium dodecyl sulphate (SDS-PAGE) fractionated dog thymocytes, they revealed a strong heterogeneous antigen. Competitive inhibition of binding of directly labeled mouse-antidog T lymphocytes monoclonal antibodies (MdT-mAbs) to solubilized dog thymocytes indicates that 2 different antigenic epitopes (P, Q) are recognized. In indirect peroxidase immunocytochemistry, MdT monoclonal antibodies recognized up to 95% thymocytes, 69% blood lymphocytes, 76% lymph node lymphocytes, and approximately 2% bone marrow lymphocytes; they were nonreactive with surface immunoglobulin positive blood cells, monocytes, platelets, cells of myelo- and erythropoietic lineage in the bone marrow. Immunohistochemistry on thymus, lymph nodes, and spleen sections revealed that MdT-mAbs had labeled cortical and medullary thymocytes, paracortical T cell areas in lymph nodes and the periarteriolar zone of spleen white pulp, whereas B cell areas remained unstained. The antibodies lysed dog thymocytes in the presence of complement. Lethally irradiated dog receiving bone marrow autograft depleted of MdT-P1 positive cells ex vivo showed engraftment and complete recovery of marrow function. Studies of antibody activity on canine hemopoietic progenitor cells in granulocyte-macrophage progenitors (CFUGM) also showed no reduction of CFUGM in MdT-P1-depleted bone marrow.

Antilymphocytic antibodies and marrow transplantation. VI. Absence of immunosuppression in vivo after injection of monoclonal antibodies blocking graft-versus-host reactions and humoral antibody formation in vitro.

The in vivo and in vitro effectiveness of several monoclonal antimouse T and B cell antibodies, of anti-Th-1 and of Iak serum, as well as of ATG were compared. The parameters were prolongation of skin graft survival, prevention of graft-versus-host disease (GVHD), antibody and primary and secondary plaque formation against sheep redblood cells (RBCs), and T cell depletion of lymphoid tissues. In general, in vitro effectiveness of the monoclonal antibodies exceeded their in vivo effectiveness. Skin graft survival was prolonged by ATG, but not by monoclonal anti-T, or anti-T plus anti-B antibody. GVHD was prevented by in vitro incubation of donor bone marrow with monoclonal anti-Th-1, but in vivo treatment of marrow donors was ineffective. Treatment with ATG was successful. Anti Iak antibody blocked plaque formation by spleen cells incubated with sheep RBCs, but had no effect on secondary plaque formation when given in vivo. Neither was there any in vivo effect of anti-Iak or anti-Th-1 on antisheep RBC agglutinin formation. ATG was effective in both of these assays, although its cytotoxic and complement-fixing titer did not exceed that of anti-Th-1 or anti-Iak. Although anti-Th-1 was cleared more rapidly from the serum of mice expressing the corresponding Th-1 alloantigen, than from mice with the noncorresponding alloantigen and although anti-Th-1 was shown to bind to the T cell areas of the lymphoid tissue, it did not--unlike ATG--deplete these areas of T cells. Possible reasons for the difference in effectiveness of in vitro and in vivo application of these monoclonal antibodies are discussed.

Anatomical distribution of call antigen expressing cells in normal lymphatic tissue and in lymphomas.

Anatomical distribution of common acute lymphoblastic leukemia antigen (CALLA) was studied in lymphomas as well as in normal lymphatic organs using the monoclonal antibody VIL-A1. Twelve lymphomas were labelled by VIL-A1. Three of the 12 tumours also had T-cell marker, six lymphomas also showed immunoglobulin staining and only three tumours were pure CALLA lymphomas. Tonsils showed a distinct CALLA labelling of many germinal centre cells and of singular cells in interfollicular T-cell regions. Children's thymuses showed rare distinctly labelled cells in the cortex and medulla and slightly more cortical cells stained faintly by VIL-A1. Foetal thymuses of about the twelfth week of gestation contained many heavily labelled cells. The findings are discussed as evidence for the presence of CALLA on immature B as well as T lymphocytes. They favour the idea of CALLA as a common lymphocyte differentiation antigen although other possibilities of interpretation are also discussed.

Suppression of normal and neoplastic human T cells with unconjugated monoclonal antibodies in SCID mouse chimeras.

The aim of this study was to establish a preclinical in vivo model to evaluate the suppressive effect of unconjugated anti-human T (CD3, 5, 7)-cell monoclonal antibodies (mAb) of mouse IgG2a or rat IgG2b isotype. Therefore, severe combined immunodeficient (SCID) mice were transplanted with human peripheral blood lymphocytes (PBL) of healthy donors (hu-PBL-SCID) or with neoplastic T cells of the human T-ALL cell line Jurkat. In preselected hu-PBL-SCID mice with substantial T cell chimerism single antibody injection caused prompt suppression of circulating human T lymphocytes within 2 days followed by occasional T cell recovery during the following weeks. Furthermore, antibody-mediated T cell suppression was measured by prolonged survival of SCID mice that had been injected with Jurkat cells preadapted to cause 100% mortality within 40 days. Injection of preadapted Jurkat cells caused fatal metastasis in lymphoid as well as non-lymphoid organs. The progression of leukemic cells was successfully suppressed when cells and anti-T cell mAb were given i.p. In contrast to control mice, tumor mortality of antibody-treated animals was delayed or completely suppressed. We conclude that SCID mice with reproducible human T cell chimerism are a relevant animal model to test the suppressive effect of anti-human T cell mAb in preclinical studies.

Lymphoproliferations in the bone marrow: identification and evolution, classification and staging.

Bone marrow biopsies from 3229 patients with lymphoproliferative disorders and 1156 patients with benign or reactive lymphoproliferations were investigated and criteria for distinguishing between them are given. Bone marrow involvement was found in 89% of multiple myeloma, 64% of non-Hodgkin's lymphomas and 8% of Hodgkin's disease. According to the predominant proliferative cell type there were five major entities in multiple myeloma and non-Hodgkin's lymphomas: (1) plasmacytic; (2) lymphocytic; (3) hairy cell; (4) immunocytic; (5) centrocytic. These were further classified into distinct subtypes each of which had independent prognostic significance. The mode of spread of the lymphoproliferative disorders in the bone marrow showed one of six architectural patterns, which together with the quantity of infiltration in the biopsy (reflecting the tumour cell burden) had significant predictive value. These results demonstrate the value of bone marrow biopsies in the identification, classification and staging of lymphoproliferative disorders, as well as in monitoring the course of disease and the response to therapy.

Cutaneous immunoblastic T-cell lymphoma.

The case of a 69-year-old male patient with an unusual type of malignant lymphoma is presented. Clinically, it was at first characterized by follicular papules and erythematous patches, later, by the development of cutaneous tumors and enlarged lymph nodes, and by a severe, finally excruciating pruritus. Treatment with PUVA (psoralen-ultraviolet-A) combined with 40--80 mg prednisolone and then with chemotherapy [COPP regimen (cyclophosphamide, vincristine, procarbacine, prednisone), high-dosage methotrexate followed by citrovorum factor rescue] was not successful. The patient died of pneumonia 2.5 years after the onset of the first clinical symptoms. An immunoblastic infiltrate was observed histologically and electromicroscopically in the initial lesions of the skin. Therefore, the diagnosis of a cutaneous immunoblastic T-cell lymphoma was tentatively made at the beginning, which was later confirmed in numerous biopsies and laboratory investigations. Immunocytologically and enzymecytochemically, the infiltrating cells were shown to be immature T cells; in the lymph nodes, numerous immunoblasts and large Sézary cells was observed in the peripheral blood, though there were no very large Sézary cells or blast cells. In the autopsy, a systemic involvement with an atypical lymphoid infiltration was found in numerous internal organs. The special nature of this case justifies its classification as high-grade malignant lymphoma and its differentiation from normal cases of mycosis fungoides. In contrast, mycosis fungoides generally fulfils criteria typical of low-grade malignant lymphomas.

Anatomical distribution of T- and B-lymphocytes in Marek's disease--an immunohistochemical study.

Neural lesions of Marek's disease, Marek's disease tumours in the ovary, liver, and kidney, as well as spleen and bursa of Fabricii of chickens bearing Marek's disease tumorous infiltrations, were examined by a new immunohistochemical technique basing upon Sternbergers unlabelled antibody enzyme method which allows the exact localization of lymphoid cells based on their surface antigens. Type C neural lesions contained T-lymphocytes almost exclusively. Type B neural lesions had relatively high proportions of T- and B-lymphocytes, and severe type A neural lesions possessed one part of heavily labelled T-lymphocytes and a number of cells stained weakly by rabbit-anti-chicken-T-cell-globulin. Tumorous infiltrations had similar characteristics as type A neural lesions. Spleen and interfollicular spaces of bursa of Fabricius were infiltrated by T-lymphocytes.

Identification and characterization of a mouse monoclonal antibody (M10) directed against canine (dog) CD8+ lymphocytes.

Monoclonal antibodies (mAb) were produced by immunizing BALB/c mice with non-adherent dog lymphocytes. M10 was specific for a subset of dog lymphocytes. M10 belonged to the IgG1 subclass and reacted with 26% of dog peripheral blood lymphocytes, 24% of spleen lymphocytes, 81% of thymus cells, 1.2% of bone marrow cells (5.8% of bone marrow lymphocytes) and 23% of PHA-stimulated lymphocytes. Immunohistology of snap-frozen thymus and spleen showed that the spleen B-cell area stained negative, whereas the spleen T-cell area and the thymus medulla exhibited positive reaction in 20-30%. The thymus cortex was strongly positive. M10 diminished cell lysis by 58% in cell mediated lysis assays (CML). Immunoblot assays revealed that M10 recognized an antigen with a molecular weight of 76 kD under non-reducing and 33 kD under reducing conditions. Finally, M10 bound to a canine CD8 alpha transfected rat T-cell line (NB2). These findings characterize M10 as an antibody directed against the dog CD8 antigen.

Use of two virustatica (AZT, PMEA) in the treatment of FIV and of FeLV seropositive cats with clinical symptoms.

In the present study the therapeutic efficacy and the side effects of two antiretroviral compounds used in human acquired immunodeficiency syndrome (AIDS) research, 3'-azido-2',3'-dideoxythymidine (AZT, zidovudine, Retrovir) and 9-(2-phosphonylmethoxyethyl)adenine (PMEA), were investigated in the treatment of cats naturally infected with feline immunodeficiency virus (FIV) and cats naturally infected with feline leukemia virus (FeLV). AZT was administered subcutaneously at a dose of 5 mg kg-1 body weight every 12 h and PMEA was administered subcutaneously at a dose of 2.5 mg kg-1 body weight every 12 h during a 3 week hospitalization. The therapeutic efficacy of both compounds was investigated. There was a stronger potency of PMEA than of AZT on the regression of stomatitis in FIV and in FeLV infected cats. In addition, in FIV infection PMEA had a stronger effect on the improvement of the general clinical status. Both antiretroviral compounds were potent agents to improve the immunologic status of FIV infected cats by raising the CD4/CD8 ratio. In FeLV infection PMEA and AZT appeared to reduce antigenemia. The hematological side effects caused by PMEA were severe and stronger than those of AZT. Therefore the advantage of PMEA in clinical and immunologic improvement was diminished by the hematologic disorders, which do not allow long term treatment with this drug in the dose used.

Comparison of T-cell subpopulations in cats naturally infected with feline leukaemia virus or feline immunodeficiency virus.

T-cell subsets were studied by flow cytometry in 58 feline leukaemia virus (FeLV)-positive cats with naturally acquired FeLV infection to determine whether the changes in CD4+ or CD8+ T cell populations differed from those observed in 55 feline immunodeficiency virus (FIV)-positive cats with naturally acquired FIV infection. The sole criterion for inclusion into the study was seropositivity. Mean (SD) CD4+ T cell values of FeLV positive cats were decreased to 31.1 (8.0) per cent and their CD8+ T cell values were increased to 22.8 (6.3) per cent in comparison with uninfected control cats (37.9 [9.5] per cent CD4+; 15.2 [6.3] per cent CD8+). The CD4+/CD8+ ratio was reduced to 1.5 (0.7), compared with 3.0 (1.5) in 39 FeLV-and FIV-negative control cats. Differences from control values were significant, but there was no significant difference between CD4+ and CD8+ lymphocytes of FeLV-versus FIV-infected cats. These findings indicate that FeLV and FIV have similar effects on T lymphocyte subsets. Both retrovirus infections can induce immunodeficiency, both viruses infect a broad range of lymphohaemopoietic cells, despite having different primary target cells, and can induce the killing of lymphocytic cells in vitro. It is concluded that a decreased CD4+/CD8+ ratio is not restricted to FIV infections but may also occur in FeLV infection.

Monoclonal antibodies to complement components without the need of their prior purification. II. Antibodies to mouse C3 and C4.

Rat monoclonal antibodies (MAbs) to mouse complement components C3 and C4 were produced by immunizing rats with cell-bound C3 and C4. This principle involves: a) using mouse thymocytes coated with syngeneic rat antibody isotypes that show high affinity to C1q, b) the intercalation of C1q from serum and c) the subsequent activation of the classical complement pathway leading to deposition of cell-bound complement components. Screening for anti-complement antibodies was performed on antibody coating microtiter plates with mouse serum as source of complement. The reactivity of the MAbs was determined by variations of the ELISA screening system using EDTA-serum to inhibit complement activation by C1 dissociation, serum rendered deficient of functionally active C3 by treatment with cobra venom factor (CVF) or serum of genetically C5-deficient mice. The specificity of the MAbs was confirmed by affinity chromatography followed by SDS-PAGE and immunoblotting. We were able to establish a panel of anti-C3 and anti-C4 MAbs of various isotypes.

Distribution of injected anti-Thy-1 monoclonal antibodies in mouse lymphatic organs: evidence for penetration of the cortical blood-thymus barrier, and for intravascular antibody-binding onto lymphocytes.

The localization of monoclonal anti-Thy-1 binding in mouse thymus, spleen, and lymph nodes was studied at early intervals after intravenous (i.v.), intraperitoneal (i.p.) and subcutaneous (s.c.) injection of a single dose of the monoclonal antibody (MAb). Five minutes after i.v. injection, anti-Thy-1 was bound to cortical thymocytes surrounding capillaries in the thymic cortex, to thymic cells beneath the thymic capsule and to medullary thymocytes around venules of the thymus medulla. When anti-Thy-1 was injected i.p. or s.c. the MAb was first deposited in capillary walls in the thymus cortex and did not appear on thymocytes outside of capillaries until 60 min after injection. These findings suggest that thymic cortical capillaries are permeable for anti-Thy-1 MAb contrary to the generally accepted principle of a blood thymus barrier to antigens in thymic cortex. Some cortex capillaries also became permeable for peroxidase when injected 15 min after anti-Thy-1 MAb. Anti-Thy-1 MAb penetration into spleen white pulp and lymph node paracortex occurs along the circulatory pathway of the vascular system in the spleen and of lymphatics in lymph nodes. But those lymphocytes with a strong anti-Thy-1 MAb loading always appeared along the pathways of lymphocyte circulation indicating that the most intense contact between anti-Thy-1 MAb and T-lymphocytes occurs not in the lymphatic organs but during the intravascular period of recirculation of lymphocytes.

Monoclonal antibodies to complement components without the need of their prior purification. I. Antibodies to mouse C1q.

Rat monoclonal antibodies (MAbs) reactive with mouse complement subcomponent C1q were raised applying a principle that requires minute amounts of serum and circumvents purification of serum-derived C1q. The principle involves a) using the high affinity of certain cell-bound antibody isotypes for intercalating C1q from serum of various species, b) selecting such antibodies as are syngeneic to the immunized animal species, thus avoiding the production of antibodies against the intercalating antibodies and c) screening for the anti-C1q MAb in microtiter plates coated with C1q-intercalating MAb isotypes that are heterogeneic to the immunized animal species. We could establish 3 MAbs of IgM subclass, whose reactivity to mouse C1q was shown by ELISA techniques. One of them (RmC13C9) crossreacted with human C1q and was shown by SDS-PAGE and subsequent immunoblotting to recognize the C-chain of mouse and human C1q. The other two MAbs are directed against SDS-sensitive epitopes on mouse C1q and were, therefore, further characterized by native agarose gel electrophoresis and immunohistochemical studies.