Both high intratumoral microvessel density determined using CD105 antibody and elevated plasma levels of CD105 in colorectal cancer patients correlate with poor prognosis.

CD105 and its ligand transforming growth factor beta (TGFbeta) are modulators of angiogenesis, which drives tumour growth and metastasis. Tumour microvessel density (MVD) has proven to be an important determinant of prognosis. In this study, we have examined the prognostic value of MVD identified using Mabs to the pan-endothelial marker CD34 and to CD105 in 111 patients with colorectal cancer. The Mab to CD105 preferentially reacts with angiogenic endothelial cells. Of the 111 patients studied, 38 were alive and 73 had died of the disease. The median MVD values counted using anti-CD34 and anti-CD105 were 5 (range 1.40-9.00) and 3.10 (range 0.90-8.00), respectively. Kaplan-Meier survival analysis revealed that only MVD values obtained using CD105 Mab correlated with survival. Patients with a high MVD, above the median (3.10), showed the worst prognosis. A similar outcome was observed when MVD was divided into quartiles. In order to ascertain if this strong expression of CD105 in the tumour vasculature is reflected in patients' plasma, circulating levels of CD105, TGFbeta1 and TGFbeta3 together with the receptor-ligand complexes were quantified in patients with colorectal carcinoma and normal controls. Results showed that except for TGFbeta1, the levels of all other molecules were significantly elevated compared with controls. The levels of CD105 were positively correlated with Dukes' stages. A lower TGFbeta1 level was noted in patients with carcinoma over the controls. Furthermore, TGFbeta3 and CD105/TGFbeta3 complexes were markedly lowered in postoperative compared with preoperative plasma samples. Immunostaining revealed that TGFbeta1 was expressed in cancer cells but TGFbeta3 in the stromal cells, whereas CD105 was exclusively expressed in vascular endothelial cells of tumour blood vessels. In conclusion, this study demonstrates that MVD quantified using a Mab to CD105 is an independent prognostic parameter for survival of patients with colorectal cancer, and that plasma levels of CD105, TGFbeta1, TGFbeta3 and CD105/TGFbeta complexes may be useful markers for assessing disease progression. These data have led us to propose that quantification of these determinants may prove useful to monitor therapeutic efficacy in patients with colorectal cancer, especially those who are being treated with antiangiogenic therapies.

Antiangiogenic therapy of established tumors in human skin/severe combined immunodeficiency mouse chimeras by anti-endoglin (CD105) monoclonal antibodies, and synergy between anti-endoglin antibody and cyclophosphamide.

Endoglin (EDG; CD105) is a proliferation-associated cell membrane antigen of endothelial cells and is strongly expressed on the tumor-associated angiogenic vascular endothelium. Furthermore, EDG is essential for angiogenesis and a component of the transforming growth factor (TGF)-beta receptor complex. The present three anti-EDG monoclonal antibodies (mAbs), SN6f, SN6j, and SN6k, react strongly with proliferating human endothelial cells but cross-react very weakly with murine endothelial cells. Analysis of Scatchard plot of direct binding of these mAbs to proliferating human umbilical vein endothelial cells showed equilibrium constants of 8.3 x 10(9), 3.1 x 10(9), and 1.0 x 10(9) liter/mol, respectively, for SN6f, SN6j, and SN6k. These mAbs did not react with MCF-7 human breast cancer cells. To facilitate antiangiogenic tumor therapy by these mAbs in animal models, we used human skin/severe combined immunodeficiency (SCID) mouse chimeras bearing tumors of MCF-7. Blood vessels in the chimeras were analyzed by immunostaining with species (human or mouse)-specific anti-CD31 and anti-EDG mAbs including an antihuman EDG mAb termed SN6h. Blood vessels in the completely healed grafted human skins consisted of a mixture of human (43.5%) and murine (56.5%) vessels, whereas only murine vessels were detected in the adjacent murine skins and s.c. tissues. Therefore, murine vessels infiltrate into the human skin grafts from the adjacent murine tissues, whereas the growth of human vessels is limited within the boundary of human skins. Growth of human MCF-7 tumors in the human skin grafts increased the ratio of human:murine vessels. Analyses of the grafted skins before and after tumor transplantation showed that SN6h reacted with tumor-induced angiogenic blood vessels but not with nonangiogenic vessels, whereas antihuman CD31 mAb reacted with both angiogenic and nonangiogenic vessels. The results show that SN6h is capable of distinguishing the tumor-induced angiogenic vasculature from the nonangiogenic vasculature in the present model. Antiangiogenic therapy of the chimeras bearing established MCF-7 tumors was carried out by i.v. administration of a mAb(s) via the tail vein of mice. SN6j and SN6k were effective for suppressing the established tumors, whereas tumor suppression was weaker with SN6f. The results indicate an absence of a direct correlation between antigen-binding avidity and in vivo antitumor efficacy of anti-EDG mAbs and suggest the importance of other factors (e.g., epitopes) in antitumor efficacy. No significant toxicity of the mAbs was detected. Combination of SN6f and SN6k that define mutually nonoverlapping epitopes showed an additive antitumor effect. Combination of SN6j and cyclophosphamide using an antiangiogenic schedule of drug dosing showed synergistic antitumor efficacy. The combination therapy induced lasting complete regression of the established tumors in two of the eight treated chimeras. We examined human and murine blood vessels in large human tumors from the chimeras at the end of therapeutic experiment. The test showed that SN6j therapy resulted in complete suppression of human vessels in the tumors but resulted in only weak suppression of murine vessels. Cyclophosphamide was not effective for suppressing human vessels and only weakly suppressive against murine vessels. Combination of SN6j and cyclophosphamide was effective for completely suppressing human vessels and also effective for partial (i.e., 35%) suppression of murine vessels. The results show that systemic administration of naked antihuman EDG mAbs can suppress established tumors, and the efficacy is markedly enhanced by combining a chemotherapeutic drug using an antiangiogenic schedule of drug dosing. These mAbs should show stronger antitumor efficacy in patients whose tumors depend entirely on human blood vessels.

Association of serum endoglin with metastasis in patients with colorectal, breast, and other solid tumors, and suppressive effect of chemotherapy on the serum endoglin.

In this report, we present data indicating that the increased serum endoglin (EDG; CD105) quantitated by a double-antibody sandwich assay is associated with metastasis in patients with solid tumors including colorectal and breast carcinomas. In addition, we show that chemotherapy exerts a suppressive effect on the serum EDG. EDG is a proliferation-associated cell membrane antigen of human vascular endothelial cells. Furthermore, EDG is essential for angiogenesis. We generated two anti-EDG monoclonal antibodies (mAbs), termed SN6a and SN6h, defining different epitopes of EDG and developed a double-antibody sandwich assay to quantitate serum EDG in patients with solid tumors. SN6h possesses an exceedingly high antigen-binding avidity (K, 1.38 x 10(11) liters/mol), whereas SN6a possesses an ordinary avidity for a mAb directed to a cell surface antigen (K, 2.85 x 10(8) liters/mol). We measured serum samples from 101 patients with solid tumors (34 colorectal cancers, 16 breast cancers, and 51 other cancers), 8 patients with benign diseases, and 31 healthy volunteers. The serum level of EDG was significantly elevated in the patients with metastatic cancers. The mean serum EDG in the 42 metastasis-negative patients was 34.0 +/- 26.8 ng/ml (median value, 27.9 ng/ml), whereas the value in the 59 metastasis-positive patients was 63.8 +/- 72.5 ng/ml (median value, 37.2 ng/ml). The difference in EDG levels between the two groups was statistically significant (P = 0.012). Of the colorectal cancer patients, the difference in EDG levels between the 19 metastasis-negative patients and the 15 metastasis-positive patients was statistically significant (P = 0.02). In addition, the difference between the normal control (n = 31) and the 15 metastasis-positive colorectal cancer patients was statistically significant (P = 0.04). Of the breast cancer patients, the difference in EDG levels between the 11 metastasis-positive patients and the normal control was statistically significant (P < 0.005). In additional studies, we found that chemotherapy suppressed serum EDG levels in cancer patients. Of the 54 metastasis-positive patients with solid tumors, the mean serum EDG in the 32 chemotherapy-receiving [chemotherapy(+)] patients was 44.7 +/- 41.9 ng/ml (median value, 36.1 ng/ml), whereas the value in the 22 chemotherapy(-) patients was 102.4 +/- 99.5 ng/ml (median value, 64.8 ng/ml). The difference in serum EDG between the two groups is statistically significant (P < 0.005). In the majority of metastasis-positive patients who were not receiving chemotherapy, serum EDG was elevated. The results suggest that serum EDG may be a useful marker for monitoring early signs of metastasis and cancer relapse in a long-term follow-up of solid tumor patients.

Angiogenesis and metastasis marker of human tumors.

Tumor growth and metastasis are dependent on angiogenesis. Therefore, certain angiogenesis markers may be useful as metastasis markers and/or the targets for antiangiogenic therapy. We and others have been studying endoglin(EDG; CD105) for such purposes. EDG is a proliferation-associated antigen of endothelial cells and essential for angiogenesis. In addition, EDG is a component of the transforming growth factor(TGF)-beta receptor complex. Expression of EDG is up-regulated in tumor-associated angiogenic vasculature compared with normal tissue vasculature. Microvessel density detected for EDG expression in breast cancer tissues showed a statistically significant correlation with overall and disease-free survival. In addition, elevated serum EDG was associated with metastasis in patients with colorectal, breast, and other solid tumors. On the other hand, We have been targeting EDG on tumor vasculature to suppress tumor growth and metastasis by systemic(i.v.) administration of anti-EDG monoclonal antibodies(mAbs) and immunoconjugates(IMCs). To thid end, we have been using three animal models, i.e., severe combined immunodeficient(SCID) mouse model of MCF-7 human breast cancer, human skin/SCID mouse chimera model bearing MCF-7 tumor, and syngeneic metastasis model of colon-26 adenocarcinoma cells in BALB/c mice. In addition, antiangiogenic activities of anti-EDG mAbs and IMCs were evaluated in mice using the dorsal air sac assay. The IMCs were prepared by coupling deglycosylated ricin A-chain or 125I to individual anti-EDG mAbs. These anti-EDG IMCs and mAbs showed substantial antitumor efficacy and antimetastatic activities without showing severe toxicity. Recently, we generated a recombinant human/mouse chimeric anti-EDG mAb to facilitate clinical application of the mAb.

Differential behavior of VEGF receptor expression and response to TNP-470 in two immortalized human endothelial cell lines.

Angiogenesis consists of endothelial cell proliferation, migration and tube formation. It is useful to investigate endothelial cell behavior using immortalized endothelial cell lines. We characterized cell growth property, growth factor dependency and response to angioinhibitory drugs; TNP-470, staurosporine, radicicol and genistein, using human umbilical vein endothelial cells (HUVECs) immortalized by human papilloma virus (HPV)-16 E6-E7, named HUVECs/E6-E7, and HUVECs/E6-E7 transformed by v-Ki-ras gene, named HUVECs/E6-E7/ras. The dependency to vascular endothelial cell growth factor (VEGF) and basic fibroblast growth factor (bFGF) for cell proliferation decreased in HUVECs/E6-E7, but were restored in HUVECs/E6-E7/ras. Flow cytometric analysis demonstrated that a VEGF receptor KDR/flk-1 was down-regulated in HUVECs/E6-E7 but not in HUVECs/E6-E7/ras. Expression of another VEGF receptor flt-1 was consistent in all cells including HUVECs, HUVECs/E6-E7 and HUVECs/E6-E7/ras. According to the analysis of the angioinhibitory drugs, HUVECs/E6-E7 was obviously resistant to TNP-470, but HUVECs/E6-E7/ras showed similar response compared to HUVECs which suggests that v-Ki-ras signaling pathway is associated with VEGF receptor expression and make HUVECs/E6-E7 sensitive to TNP-470 by modulating the signal transduction cascade. In conclusion, HPV-16 E6-E7 and v-Ki-ras genes have unique growth properties and these immortalized cells are useful for investigating signal transduction pathways of endothelial cells, and for screening of angioinhibitory drugs.

Antiangiogenic radioimmunotherapy of human solid tumors in SCID mice using (125)I-labeled anti-endoglin monoclonal antibodies.

Endoglin (CD105), which is a component of the TGF-beta receptor complex, is highly expressed at the surface of proliferating human endothelial cells such as those of tumor vessels. In the present study, we tested the antitumor efficacy of (125)I-labeled anti-endoglin monoclonal antibodies (MAbs), SN6f and SN6j, against s. c. tumors of MCF-7 human breast cancer cells in SCID mice by i.v. administration. SN6f and SN6j cross-react weakly with mouse endothelial cells, but show no significant reactivity with MCF-7 tumor cells. These MAbs are effectively internalized into the cells after binding to the cell surface antigen of endothelial cells. Four groups of SCID mice (n = 10 or 9 in each group) inoculated s.c. with 8 x 10(6) MCF-7 cells were treated with (125)I-SN6f (10 microCi), (125)I-SN6j (10 microCi), a (125)I-labeled isotype-matched control IgG (10 microCi) or PBS. The systemic therapy was performed in 2 series, i.e., on days 3, 5, 7 and days 58, 60, 62. Both (125)I-SN6f and (125)I-SN6j showed significant growth suppression of the tumors, whereas the (125)I-labeled control IgG did not show any significant antitumor efficacy. No significant toxicity or weight loss was observed in mice treated with either (125)I-SN6f or (125)I-SN6j. After 100 days of observation, autopsies revealed no significant organ damage. Our results show the possible usefulness of antiangiogenic radioimmunotherapy using (125)I-labeled anti-endoglin MAbs.

Induction of lasting complete regression of preformed distinct solid tumors by targeting the tumor vasculature using two new anti-endoglin monoclonal antibodies.

Endoglin (EDG, CD105) is a proliferation-associated antigen on endothelial cells. In this study, two new anti-EDG monoclonal antibodies (mAbs) Y4-2F1 (or termed SN6j) and P3-2G8 (SN6k) were generated and used for treating distinct preformed tumors. These mAbs, both IgG1-kappa antibodies, cross-reacted weakly with mouse endothelial cells but defined epitopes different from the epitope defined by a previously reported anti-EDG mAb K4-2C10 (B. K. Seon et al., Clin. Cancer Res., 3: 1031-1044, 1997). SN6j and SN6k reacted strongly with human endothelial cells and vascular endothelium of malignant human tissues but showed no significant reactivity with tumor cells per se. The deglycosylated ricin A chain (dgRA) conjugates of the two mAbs showed a weak but specific cytotoxic activity against murine endothelial cells in vitro. In the therapeutic studies, severe combined immunodeficient mice were inoculated s.c. with MCF-7 human breast cancer cells and left untreated until palpable tumors of distinct size (4-6 mm in diameter) appeared. Mice with the distinct tumors were treated by i.v. administration of individual anti-EDG conjugates, unconjugated mAbs, or a control conjugate. Long-lasting complete regression of the tumors was induced in the majority of tumor-bearing mice (n = 8 for each conjugate) when 40 microg of the individual conjugates were administered three times via the tail vein. It is remarkable that the tumors remained regressed without further therapy for as long as the mice were followed (i.e., 100 days). Control conjugate did not induce regression of the tumors in any of the treated mice, although weak nonspecific effects were observed in some of the mice (n = 8). The effects of unconjugated mAbs were small with the dose used, i.e., 34 microg three times. The anti-EDG conjugates showed antiangiogenic activity in the dorsal air sac assay in mice. The results suggest good potential of these conjugates for the clinical application.

Cross-linking of antigen receptor via Ig-beta (B29, CD79b) can induce both positive and negative signals in CD40-activated human B cells.

Antigen-dependent activation of B lymphocytes is mediated through surface immunoglobulins and their associated molecules Ig-alpha (CD79a, Mb1) and Ig-beta (CD79b, B29). Here we show that an antibody directed against the extracellular part of human Ig-beta can, when cross-linked by CD32-transfected L cells, induce an IL-2-dependent proliferation of tonsil B cells. With the use of L cells stably transfected with both CD32 and CD40L, anti-Ig-beta activation of B cells was combined with CD40 triggering, an important component of the T cell-dependent B cell activation. This dual cellular activation resulted in two different phases, with initially synergistic proliferative effects, both without and with IL-2 or IL-10. Then, after 5-6 days of culture, cells stimulated with both anti-Ig-beta and CD40L underwent massive cell death, in contrast to B cells activated with CD40L alone. Cell death was not prevented by the addition of IL-2 or IL-10, but was prevented by the addition of IL-4. These results are discussed in the context of positive and negative selection of mature B cells.

Improvement of the chronic lymphocytic leukemia scoring system with the monoclonal antibody SN8 (CD79b).

A scoring system, based on the immunophenotypic analysis of a panel of five membrane markers (CD5, CD22, CD23, FMC7, SmIg) was shown to be useful in the distinction between chronic lymphocytic leukemia (CLL) and other B-cell lymphoproliferative diseases (non-CLL). We investigated whether the monoclonal antibody SN8 (CD79b) could improve our previous scoring system. Peripheral blood samples of 298 patients with CLL and 166 patients with non-CLL were analyzed by flow cytometry. Using the five standard markers, the accuracy of the scoring system was 91.8%, using a cutoff of 4 points or higher, to distinguish CLL from non-CLL. This was increased to 96.6% if SN8 was added and a cutoff of 4 points or higher was also used. A similar accuracy, 96.8%, was observed if CD22 was excluded and a cutoff of 3 points or higher was used. Thus, the replacement of CD22 by SN8 in the original scoring system significantly increases its potential to discriminate between CLL and other B-cell lymphoproliferative diseases.

Long-lasting complete inhibition of human solid tumors in SCID mice by targeting endothelial cells of tumor vasculature with antihuman endoglin immunotoxin.

In the present study, we developed an antitumor immunoconjugate that appears to be promising as a novel curative antitumor agent against a variety of human solid tumors. We generated a new antihuman endoglin (EDG) monoclonal antibody (mAb) K4-2C10 (or termed SN6f) that cross-reacts with mouse endothelial cells. Such cross-reactive anti-EDG mAbs have not been reported previously. This mAb was used to target tumor-associated vasculature in SCID mice inoculated with human tumors. No anti-EDG mAb or its immunoconjugates have previously been successfully used for targeting vasculature in vivo. In this study, MCF-7 human breast cancer cells were inoculated s.c. into SCID mice. K4-2C10 did not react with the MCF-7 cells but showed a weak reactivity with mouse endothelial cells. The mAb reacted with the proliferating endothelial cells more strongly than with the quiescent endothelial cells. The mAb exhibited much stronger reactivity (>10-fold) with human endothelial cells than with mouse endothelial cells and reacted strongly with vascular endothelium of tumor-associated blood vessels in a variety of human malignant tissues. Conjugates of K4-2C10 with ricin A chain (RA) and deglycosylated ricin A chain (dgRA) showed a weak but specific cytotoxic activity against murine endothelial cells in vitro; the 50% inhibitory dose of the RA and dgRA conjugates was 54 nm and 29 nm, respectively. Remarkable antitumor efficacy was observed when a small amount (a total of 60 microgram corresponding to 24% of the LD50 dose) of the dgRA conjugate was administered i.v. into SCID mice that had been inoculated s.c. with MCF-7. Unconjugated mAb K4-2C10 was not significantly effective in the inhibition of the tumor growth. The immunotoxin (IT) completely inhibited growth of the tumor in all of the treated mice (n = 8). Furthermore, similar antitumor efficacy was observed when the IT was administered i.v. into the tumor-inoculated SCID mice that had been pretreated with unconjugated K4-2C10 to block the potentially available weak binding sites of normal tissues. The strong therapeutic effects of the IT were reproduced in another set of therapeutic experiments. No significant side effects were observed in the mice. The differences in the tumor growth between the control group and the IT-treated groups were statistically significant. The IT showed antiangiogenic activity in the dorsal air sac method. The results indicate that K4-2C10 IT effectively treated the tumor-bearing mice by selectively inhibiting the tumor-associated blood vessels and by disrupting tumor-associated angiogenesis. The strong antitumor efficacy of the K4-2C10 IT is remarkable in view of the fact that K4-2C10 and its IT showed only a weak reactivity with mouse endothelial cells, and a relatively small amount of the IT was administered i.v. to treat s.c. tumors. We anticipate that the K4-2C10 IT will show much stronger antitumor efficacy and antiangiogenic activity in patients with solid tumors and other angiogenesis-associated diseases. The present results demonstrate for the first time that an anti-EDG mAb or its immunoconjugate can effectively target tumor-associated vasculature in vivo.

Development of a severe combined immunodeficiency (SCID) mouse model consisting of highly disseminated human B-cell leukemia/lymphoma, cure of the tumors by systemic administration of immunotoxin, and development/application of a clonotype-specific polymerase chain reaction-based assay.

A new severe combined immunodeficiency (SCID) mouse model consisting of highly disseminated human B-cell leukemia/lymphoma was developed by i.v. inoculation of BALL-1a, an in vivo adapted malignant B-cell line. A 100% transplantability was achieved in nonpreconditioned SCID mice using various BALL-1a doses between 2.5 x 10(4) and 6 x 10(6) cells. Hind-leg paralysis preceded the death of the mice. Utility of the developed tumor model for the therapeutic studies was investigated by i.v. administration of an anti-B-cell monoclonal antibody SN7 (IgG1) and its conjugate with deglycosylated ricin A chain (dgRA). The therapy was initiated 2, 4, or 6 days after tumor inoculation using 4 x 24 microg of SN7-dgRA or 4 x 20 microg of SN7; the total dose (96 microg) of SN7-dgRA corresponded to 14% of the LD50 dose. SN7-dgRA showed a strong antitumor efficacy in all groups of treated mice. All of the day-2 group mice (n = 7) and six (66.7%) of the day-4 group mice (n = 9) survived healthily for as long as followed (240 days), whereas four (57.1%) of the day-6 group mice (n = 7) survived healthily for as long as followed (200 days). Unconjugated SN7 showed a significant antitumor efficacy but was less effective than SN7-dgRA. A PCR-based assay specific for the clonogenic BALL-1a tumor was developed and applied to determine tumors in various organs of BALL-1a-bearing SCID mice. The assay was highly sensitive in screening for trace quantities of residual tumors in various organs of SCID mice, and it could detect 1 malignant cell/2.5 x 10(5) tissue cells. The PCR-based assay was shown to be much more powerful than the conventional histological analysis in detecting residual tumors. Furthermore, we could estimate quantities of the detected tumors by the PCR-based assay. It is remarkable to find that all examined organs of some of the SN7-dgRA-treated mice were tumor-free as determined by the clonotype-specific PCR-based assay. The present results show the usefulness of the newly developed SCID mouse model, SN7-dgRA, and the clonotype-specific PCR-based molecular assay for the study of therapy of human B-cell leukemia/lymphoma.

Expression of the immunoglobulin-associated protein B29 in B cell disorders with the monoclonal antibody SN8 (CD79b).

We studied the expression of the immunoglobulin-associated membrane protein B29 in 499 cases of chronic B cell diseases using the monoclonal antibody SN8 (CD79b). SN8 was positive in 5% (17/330) of chronic lymphocytic leukemia (CLL) and 100% (15/15) of B prolymphocytic leukemia. The expression of B29 in other B cell disorders was, as a rule, significantly higher than in CLL. Two thirds of non-Hodgkin's lymphomas in leukemic phase were SN8 positive, including lymphoplasmacytic (45%), follicular (83%), mantle cell (92%) and splenic lymphoma with villous lymphocytes (74%) while only 25% of hairy cell leukemias were SN8 positive. Within CLL, 2.3% of typical cases were SN8+ while 16% of cases with atypical morphology and an increased number of prolymphocytes were SN8+. Our results suggest a useful role for SN8 in the immunophenotypic differentiation of B cell disorders as a marker for non-CLL diseases. The analysis of B29 expression may throw light into the structure of the B cell antigen receptor in B cell malignancies while the distinctive reactivity profile of SN8 has direct applications to diagnosis.

Differential expression of B29 (CD79b) and mb-1 (CD79a) proteins in acute lymphoblastic leukaemia.

CD79 is a heterodimeric molecule comprising two polypeptide chains, B29 (CD79b) and mb-1 (CD79a). It is physically linked in the surface of B cells to membrane immunoglobulin, forming the B cell antigen receptor complex. Expression of the mb-1 (CD79a) chain has been studied in leukaemias and shown to be present in most B lineage acute lymphoblastic leukaemias (ALL). In contrast, little is known about the expression of B29 (CD79b) in this condition. Two monoclonal antibodies (MoAb) were used in this study by immunocytochemistry and flow cytometry: HM57, against an intracellular epitope of the mb-1(CD79a) chain, and SN8, reacting with an extracellular epitope of B29 (CD79b). Our aim was to investigate the expression of B29 (CD79b) in the various immunological subtypes of B lineage ALL and compare its cytoplasmic and membrane expression. Seventy-nine cases were studied, including 13 chronic myeloid leukaemia in B lymphoid blast crisis (CML-BC) and 66 ALL, subclassified as early B (two), common (28), pre-B (23), mature (five) and biphenotypic with B lymphoid commitment (eight). Most cases expressed mb-1 (CD79a) in the cytoplasm. B29 (CD79b) was expressed in the cytoplasm in 65% (15/23) of pre-B-ALL and in 14% (4/28) common-ALL but it was detected in the cell membrane in only three cases of mature B-ALL, being negative in all other B lineage subtypes ALL. Three of the biphenotypic leukaemias coexpressed cytoplasmic B29 (CD79b) and mu-chain. This was also seen in two cases of CML-BC, while four cases expressed only cytoplasmic B29 (CD79b) without mu-chain. Our results suggest that during B cell differentiation, B29 (CD79b) is expressed later than mb-1 (CD79a) in the cytoplasm and parallels the cytoplasmic expression of mu-chain. B29 (CD79b) is present in the membrane at a later stage compared to its cytoplasmic expression and found in mature B blasts (B-ALL) that express membrane Ig as it is in normal and leukaemic B lymphocytes.

Immune responses in patients with T-cell lymphoma treated with an anti-idiotype antibody mimicking a highly restricted T-cell antigen.

We generated an IgG1 murine monoclonal anti-idiotype antibody (Ab2) to a highly restricted T-cell antigen designated glycoprotein (gp) 37 that is found on T-cell malignancies but not on normal cells. gp37 is identified by the murine monoclonal antibody SN2 (Ab1) against which the Ab2 was raised. Each of four patients with T-cell lymphoma predominantly confined to the skin received a minimum of four intracutaneous injections of aluminum hydroxide precipitated anti-idiotype murine monoclonal antibody (1 mg/injection) given every 2 weeks. For responding patients, injections were continued on a monthly basis. All tumors were measured along orthogonal major and minor axes, using a ruler and/or calipers, by the same observer. Tumor sizes were documented photographically. Three of the four patients developed specific idiotypic humoral immune responses, and two of the four patients also demonstrated idiotypic cell-mediated responses. Humoral responses included binding of the patients' sera to the anti-idiotype antibody as well as specific inhibition of binding of the SN2 antibody (Ab1) to the anti-idiotype antibody (Ab2). Anti-anti-idiotypic (Ab3) antibody from one patient's serum bound specifically to the gp37-positive cell line MOLT-4 and also to semipurified gp37 antigen. Cell-mediated responses were demonstrated by specific proliferative response to the aluminum hydroxide precipitated anti-idiotype antibody by patients' peripheral blood mononuclear cells. While three of the four patients had extensive disease and did not have clinical responses, one of the patients who had nine discrete skin tumors and peripheral blood involvement without other detectable disease had virtually complete disappearance of the tumors lasting over 11 months. Our results demonstrate that this particular anti-idiotype antibody can induce humoral and cellular immune responses, and at least in one patient led to a meaningful therapeutic response. Future trials should focus on immunocompetent patients with minimal disease.

Establishment of new SCID and nude mouse models of human B leukemia/lymphoma and effective therapy of the tumors with immunotoxin and monoclonal antibody: marked difference between the SCID and nude mouse models in the antitumor efficacy of monoclonal antibody.

BALL-1, a human B leukemia/lymphoma cell line, was transplanted into nude and SCID mice under various conditions. The transplantation was substantially improved by preadaptation of BALL-1 by serial passages in newborn and young nude mice. We were able to establish the desirable conditions where 100% of SCID and nude mice that were inoculated i.p. with various doses of the adapted BALL-1 (termed BALL-1a) developed tumors. Tumors in SCID mice were disseminated to various tissues in a manner analogous to tumors in patients with B leukemia/lymphoma, whereas tumors in nude mice were not as widely disseminated and grew mainly as ascites. Flow cytometric analyses showed that all of the 11 tested cell surface markers of the parental BALL-1 were well maintained on the tumor cells recovered from the SCID and nude mice. The utility of the developed tumor models for the therapeutic studies was investigated by i.p. or i.v. administration of an anti-B leukemia/lymphoma monoclonal antibody, termed SN7 (IgG1 kappa), and SN7 immunotoxin (IT) that was prepared by conjugating SN7 to ricin A chain (RA) or deglycosylated RA (dgRA). In the nude mouse model study, SN7-RA that had been administered i.p. suppressed the tumor growth completely in all of the treated mice (n = 5) without any sign of tumor or undesirable side effects for as long as followed (i.e., 350 days), whereas unconjugated SN7 showed only a slight therapeutic effect. A control RA conjugate was not effective. In the SCID mouse model studies, several sets of experiments were carried out by i.p. or i.v. administration of IT, monoclonal antibody, or control IT. In the first three sets of experiments, SCID mice inoculated with 1.1 x 10(6) BALL-1a cells received an i.p. administration of phosphate-buffered saline or three different doses (i.e., 4 x 10 micrograms, 4 x 20 micrograms, and 4 x 30 micrograms) of therapeutic agents (SN7-RA and SN7). Virtually an identical result was obtained from the three experiments. All of the phosphate-buffered saline control group mice (n = 15) died within 35 days post tumor inoculation. In contrast, all of the mice that were treated with SN7-RA (n = 19) or with SN7 (n = 15) survived for as long as followed (i.e., 250 days). However, the unconjugated SN7 was less effective than SN7 IT for tumor suppression in SCID mice that were inoculated with a larger tumor burden (i.e., 4 x 10(7) BALL-1a cells).(ABSTRACT TRUNCATED AT 400 WORDS)

Isolation and chemical characterization of the human B29 and mb-1 proteins of the B cell antigen receptor complex.

A disulfide-linked heterodimeric antigen from human B leukemia cells was detected by radioimmunoprecipitation and Western blot analysis using a monoclonal antibody (mAb). The mAb was generated against a cell membrane antigen preparation from human B prolymphocytic leukemia cells and found to define an extracellular epitope of the smaller component (beta chain) of a heterodimeric antigen on human B leukemia cells. The antigen from BALL-1, a human B leukemia cell line, and fresh (uncultured) B prolymphocytic leukemia cells was found to consist of a 44-49 kDa (alpha chain) and a 36-40 kDa (beta chain) component. An additional minor component of 34 kDa was detected in the reduced antigen from BALL-1. For chemical identification of the antigen, we isolated the antigen from a Triton X-100 lysate of BALL-1 by immunoaffinity chromatography using the mAb. Determination of the amino-terminal amino acid sequences of the alpha and beta chains unequivocally identified them as the human mb-1 and B29 proteins, respectively. The sequence analyses indicate the molecular heterogeneity of the mb-1 protein and also perhaps the heterogeneity of the B29 protein. We detected three forms of the mb-1 protein which share an identical amino-terminal amino acid sequence and probably two forms of the B29 protein which share an identical amino-terminal sequence. Our sequence data allowed us to establish the authentic amino-terminal amino acid sequence of the human B29 protein which is different from those proposed by others based on cDNA sequence analyses. Comparison of the amino-terminal sequences of the human mb-1 and B29 proteins with those of the mouse mb-1 and B29 proteins showed that the majority of the conserved amino acids in the mb-1 proteins are hydrophobic amino acids. Such conservation of hydrophobic amino acids is not observed in the amino termini of the human and mouse B29 proteins. A beta-tubulin-like protein was found to be co-purified with the mb-1-B29 antigen in the present study. In addition, we found that there is a strong amino acid sequence homology between the microtubule-binding domains of certain human microtubule-associated proteins and an intracellular segment of the human mb-1 protein.

Biodistribution and in vivo antitumor efficacy of the systemically administered anti-human T-leukemia immunotoxins and potentiation of their efficacy by alpha-interferon.

In this study, the systemically administered anti-human T-leukemia immunotoxins (ITs) are shown to be effective for tumor suppression in Ichikawa T-leukemia-bearing nude mice. In addition, their antitumor efficacy was markedly potentiated by recombinant human IFN-alpha. The combination of ITs and IFN-alpha effectively killed the tumor in the majority of the treated mice; 9 of the 12 treated mice survived tumor-free for as long as they were followed, i.e. for 140 days. Two different ITs, SN1-ricin A chain (RA) and SN2-RA, were used together to minimize the problem of tumor heterogeneity; monoclonal antibodies SN1 and SN2 are directed toward two different human T-leukemia associated cell surface antigens. In the biodistribution experiments, the paired label technique was used to include a reliable internal control. In an experiment, equal amounts of 125I-SN1-RA and 131I-labelled isotype-matching control IgG (IgG1-kappa)-RA were mixed and administered i.v. into tumor-bearing nude mice. In a separate experiment, a mixture of equal amounts of 125I-SN2-RA and 131I-control IgG-RA was administered i.v. This technique allowed us to distinguish the immunospecific uptake from the non-immunospecific uptake of ITs into individual organs. The present results clearly show that both SN1-RA and SN2-RA are specifically localized in tumors after systemic administration. For instance, 24 h after the administration of a radiolabelled mixture, the ratio of 125I/131I in the tumor was 7.0 and 23.5, respectively, for the SN1-RA/control IgG-RA mixture and the SN2-RA/control IgG-RA mixture. Such high ratios of 125I/131I were detected in the tumors throughout the experiments between 30 min and 24 h after the administration of the paired label mixture.

Three new monoclonal antibodies that define a unique antigen associated with prolymphocytic leukemia/non-Hodgkin's lymphoma and are effectively internalized after binding to the cell surface antigen.

Prolymphocytic leukemia (PLL) is closely related to chronic lymphocytic leukemia (CLL), but present with distinctive clinical/laboratory features and associated with much worse prognosis. In this study, we generated three new IgG1-kappa monoclonal antibodies (MoAbs), termed SN8, SN8a and SN8b, by use of an unconventional approach, ie, by using an isolated B PLL antigen preparation to immunize mice. These MoAbs, particularly SN8, showed a highly selective reactivity to B PLL and B non-Hodgkin's lymphoma (NHL) among various human leukemia-lymphoma specimens tested; eg, SN8 was capable of effectively distinguishing B PLL from B CLL as well as from hairy cell leukemia (HCL) cell specimens. The cell surface antigen defined by the three MoAbs was determined to be a covalently linked heterodimeric glycoprotein complex (gp49/40) consisting of a 49,000 dalton (alpha-chain) and a 40,000-dalton component (beta-chain). Epitope comparison showed that the epitope defined by SN8 (SN8 epitope) is in close proximity to SN8a epitope but in a distant position from SN8b epitope. Western blot analysis showed that both SN8 and SN8a epitopes are on the beta-chain, but SN8b epitope was not detected on either the alpha- or the beta-chain of the reduced antigen in the same analysis. Binding of either SN8 or SN8b to the cell surface gp49/40 did not cause significant downregulation of the antigen expression whereas binding of SN8a to the antigen caused small (approximately 20%) decrease in the antigen expression. Among the various normal peripheral blood cells, only a subpopulation (6.0% to 24.2% among different specimens derived from different donors) of B cells reacted with the SN8 series MoAbs; these MoAbs showed no significant reactivity against T cells, granulocytes, monocytes, erythrocytes, and platelets. Minimal or no significant reactivity (0 to 2.6% among different specimens) was detected against normal bone marrow cells. Ricin A-chain conjugates of the three MoAbs are all strongly effective for specific killing of SN8 antigen-expressing leukemia cells in the absence of any potentiators; furthermore, the addition of 10 mmol/L NH4Cl, a potentiator, enhanced strongly the cytotoxic activities of the SN8, SN8a, and SN8b conjugates. Thus, each of the three MoAbs was effectively internalized after binding to the cell surface antigen.

Monoclonal antibody SN10 which shows a highly selective reactivity with human B leukemia-lymphoma and is effectively internalized into cells.

The monoclonal antibody termed SN10 (IgG1-k) which was generated and characterized in the present study shows a highly selective reactivity with fresh (uncultured) human leukemia-lymphoma cells. The antigen defined by SN10 is a cell surface glycoprotein composed of a single polypeptide chain of Mr 36,000 and designated as gp36. The primary reactivity of SN10 is against mature B-lineage leukemia-lymphoma cells. For instance, SN10 reacted with all of the 17 B non-Hodgkin's lymphoma specimens, all of the 15 B chronic lymphocytic leukemia specimens, both of the 2 B prolymphocytic leukemia specimens, all of the 3 B hairy cell leukemia specimens, and 2 of the 3 B acute lymphoblastic leukemia specimens tested. Of normal peripheral blood cells, only a marginal reactivity of SN10 was detected with a minor subpopulation (less than 1-4% among different specimens) of isolated B-cells from healthy donors. No significant reactivity of SN10 was detected against any other isolated normal peripheral blood cells which include T-cells, granulocytes, monocytes, erythrocytes, and platelets. Furthermore, no significant reactivity of SN10 was detected against normal bone marrow specimens. In immunohistological studies using frozen tissue sections, SN10 reacted well with malignant lymphomas and showed varying patterns of reaction with hyperplastic reactive lymph nodes. Various normal human tissues tested were unreactive with SN10. In general, glycoprotein 36 was more abundantly expressed on fresh (uncultured) leukemia-lymphoma cells than on cultured leukemia-lymphoma cell lines. No significant amount of circulating SN10 antigen was detected in the plasma of leukemia-lymphoma patients or normal healthy donors. Scatchard plot analysis of direct binding of radiolabeled SN10 to a fresh (uncultured) B non-Hodgkin's lymphoma cell specimen, a fresh B chronic lymphocytic leukemia cell specimen, and DND-39 (an American Burkitt's lymphoma cell line) showed equilibrium constants of 5.2, 5.8, and 6.8 x 10(8) liters/mol, respectively. Thus, SN10 shows a high binding avidity to each of the 3 B leukemia-lymphoma cell specimens tested. Ricin A chain conjugate of SN10 killed leukemia-lymphoma cells effectively, whereas the same conjugate showed no cytotoxicity against control cells. Thus, SN10 bound to target antigen on the cell surface was effectively internalized into the cell. The present results suggest the potential of SN10 for therapy as well as for diagnosis of various forms of leukemia-lymphoma, particularly mature B-lineage leukemia-lymphoma.