Bruce L. Roberts

Investigation of the mechanism of action of alemtuzumab in a human CD52 transgenic mouse model

Alemtuzumab is a humanized monoclonal antibody against CD52, an antigen found on the surface of normal and malignant lymphocytes. It is approved for the treatment of B‐cell chronic lymphocytic leukaemia and is undergoing Phase III clinical trials for the treatment of multiple sclerosis. The exact mechanism by which alemtuzumab mediates its biological effects in vivo is not clearly defined and mechanism of action studies have been hampered by the lack of cross‐reactivity between human and mouse CD52. To address this issue, a transgenic mouse expressing human CD52 (hCD52) was created. Transgenic mice did not display any phenotypic abnormalities and were able to mount normal immune responses. The tissue distribution of hCD52 and the level of expression by various immune cell populations were comparable to those seen in humans. Treatment with alemtuzumab replicated the transient increase in serum cytokines and depletion of peripheral blood lymphocytes observed in humans. Lymphocyte depletion was not as profound in lymphoid organs, providing a possible explanation for the relatively low incidence of infection in alemtuzumab‐treated patients. Interestingly, both lymphocyte depletion and cytokine induction by alemtuzumab were largely independent of complement and appeared to be mediated by neutrophils and natural killer cells because removal of these populations with antibodies to Gr‐1 or asialo‐GM‐1, respectively, strongly inhibited the activity of alemtuzumab whereas removal of complement by treatment with cobra venom factor had no impact. The hCD52 transgenic mouse appears to be a useful model and has provided evidence for the previously uncharacterized involvement of neutrophils in the activity of alemtuzumab.

Vascular Gene Expression in Nonneoplastic and Malignant Brain

Malignant gliomas are uniformly lethal tumors whose morbidity is mediated in large part by the angiogenic response of the brain to the invading tumor. This profound angiogenic response leads to aggressive tumor invasion and destruction of surrounding brain tissue as well as blood-brain barrier breakdown and life-threatening cerebral edema. To investigate the molecular mechanisms governing the proliferation of abnormal microvasculature in malignant brain tumor patients, we have undertaken a cell-specific transcriptome analysis from surgically harvested nonneoplastic and tumor-associated endothelial cells. SAGE-derived endothelial cell gene expression patterns from glioma and nonneoplastic brain tissue reveal distinct gene expression patterns and consistent up-regulation of certain glioma endothelial marker genes across patient samples. We define the G-protein-coupled receptor RDC1 as a tumor endothelial marker whose expression is distinctly induced in tumor endothelial cells of both brain and peripheral vasculature. Further, we demonstrate that the glioma-induced gene, PV1, shows expression both restricted to endothelial cells and coincident with endothelial cell tube formation. As PV1 provides a framework for endothelial cell caveolar diaphragms, this protein may serve to enhance glioma-induced disruption of the blood-brain barrier and transendothelial exchange. Additional characterization of this extensive brain endothelial cell gene expression database will provide unique molecular insights into vascular gene expression.

Cloning and Characterization of the Genes Encoding the Murine Homologues of the Human Melanoma Antigens MART1 and gp100

The recent identification of genes encoding melanoma-associated antigens has opened new possibilities for the development of cancer vaccines designed to cause the rejection of established tumors. To develop a syngeneic animal model for evaluating antigen-specific vaccines in cancer therapy, the murine homologues of the human melanoma antigens MART1 and gp100, which were specifically recognized by tumor-infiltrating lymphocytes from patients with melanoma, were cloned and sequenced from a murine B16 melanoma cDNA library. The open reading frames of murine MART1 and gp100 encode proteins of 113- and 626-amino acids with 68.8 and 77% identity to the respective human proteins. Comparison of the DNA sequences of the murine MART1 genes, derived from normal melanocytes, the immortalized nontumorgenic melanocyte line Melan-a and the B16 melanoma, showed all to be identical. Northern and Western blot analyses confirmed that both genes encoded products that were melanocyte lineage proteins. Mice immunized with murine MART1 or gp100 using recombinant vaccinia virus failed to produce any detectable T-cell responses or protective immunity against B16 melanoma. In contrast, immunization of mice with human gp100 using recombinant adenoviruses elicited T cells specific for hgp100, but these T cells also cross reacted with B16 tumor in vitro and induced significant but weak protection against B16 challenge. Immunization with human and mouse gp100 together [adenovirus type 2 (Ad2)-hgp100 plus recombinant vaccinia virus (rVV)-mgp100], or immunization with human gp100 (Ad2-hgp100) and boosting with heterologous vector (rVV-hgp100 or rVV-mgp100) or homologous vector (Ad2-hgp100), did not significantly enhance the protective response against B16 melanoma. These results may suggest that immunization with heterologous tumor antigen, rather than self, may be more effective as an immunotherapeutic reagent in designing antigen-specific cancer vaccines.

Cloning of the goat β-casein-encoding gene and expression in transgenic mice

The goat beta-casein-encoding gene (CSN2), which encodes the most abundant protein of goat milk, has been cloned and sequenced. The intron/exon organization of the 9.0-kb goat CSN2 gene is similar to that of other CSN2 genes. Expression of the goat gene was principally restricted to the mammary gland of lactating transgenic animals. A low level of expression was also observed in skeletal muscle and skin. In contrast to a rat CSN2 transgene [Lee et al., Nucleic Acids Res. 16 (1988) 1027-1041], the goat gene was expressed to a high degree in the lactating mammary gland. Differences in the content or context of regulatory elements may account for the enhanced performance of the goat relative to the rat CSN2 gene in transgenic mice.

Design, construction and function of a multicopy display vector using fusions to the major coat protein of bacteriophage M13

Incorporation of numerous copies of a heterologous protein (bovine pancreatic trypsin inhibitor; BPTI) fused to the mature major coat protein (gene VIII product; VIII) of bacteriophage M13 has been demonstrated. Optimization of the promoter, signal peptide and host bacterial strain allowed for the construction of a working vector consisting of the M13 genome, into which was cloned a synthetic gene composed of a lac (or tac) promoter, and sequences encoding the bacterial alkaline phosphatase signal peptide, mature BPTI and the mature coat protein. Processing of the BPTI-VIII fusion protein and its incorporation into the bacteriophage were found to be maximal in a host bacterial strain containing a prlA/secY mutation. Functional protein is displayed on the surface of M13 phage, as judged by specific interactions with antiserum, anhydrotrypsin, and trypsin. Such display vectors can be used for epitope mapping, production of artificial vaccines and the screening of diverse libraries of proteins or peptides having affinity for a chosen ligand. The VIII display phage system has practical advantages over the III display phage system in that many more copies of the fusion protein can be displayed per phage particle and the presence of the VII fusion protein has little or no effect on the infectivity of the resulting bacteriophage.

Human Peripheral Blood Mononuclear Cells Exhibit Heterogeneous CD52 Expression Levels and Show Differential Sensitivity to Alemtuzumab Mediated Cytolysis

Alemtuzumab is a monoclonal antibody that targets cell surface CD52 and is effective in depleting lymphocytes by cytolytic effects in vivo. Although the cytolytic effects of alemtuzumab are dependent on the density of CD52 antigen on cells, there is scant information regarding the expression levels of CD52 on different cell types. In this study, CD52 expression was assessed on phenotypically distinct subsets of lymphoid and myeloid cells in peripheral blood mononuclear cells (PBMCs) from normal donors. Results demonstrate that subsets of PBMCs express differing levels of CD52. Quantitative analysis showed that memory B cells and myeloid dendritic cells (mDCs) display the highest number while natural killer (NK) cells, plasmacytoid dendritic cells (pDCs) and basophils have the lowest number of CD52 molecules per cell amongst lymphoid and myeloid cell populations respectively. Results of complement dependent cytolysis (CDC) studies indicated that alemtuzumab mediated profound cytolytic effects on B and T cells with minimal effect on NK cells, basophils and pDCs, correlating with the density of CD52 on these cells. Interestingly, despite high CD52 levels, mDCs and monocytes were less susceptible to alemtuzumab-mediated CDC indicating that antigen density alone does not define susceptibility. Additional studies indicated that higher expression levels of complement inhibitory proteins (CIPs) on these cells partially contributes to their resistance to alemtuzumab mediated CDC. These results indicate that alemtuzumab is most effective in depleting cells of the adaptive immune system while leaving innate immune cells relatively intact.

In Vitro Priming with Adenovirus/gp100 Antigen-Transduced Dendritic Cells Reveals the Epitope Specificity of HLA-A*0201-Restricted CD8+ T Cells in Patients with Melanoma

Replication-deficient recombinant adenovirus (Ad) encoding human gp100 or MART-1 melanoma Ag was used to transduce human dendritic cells (DC) ex vivo as a model system for cancer vaccine therapy. A second generation E1/E4 region deleted Ad which harbors the CMV immediate-early promoter/enhancer and a unique E4-ORF6/pIX chimeric gene was employed as the backbone vector. We demonstrate that human monocyte-derived DC are permissive to Ad infection at multiplicity of infection between 100 and 500 and occurs independent of the coxsackie Ad receptor. Fluorescent-labeled Ad was used to assess the kinetics and distribution of viral vector within DC. Ad-transduced DC show peak transgene expression at 24–48 h and expression remains detectable for at least 7 days. DC transduced with replication-deficient Ad do not exhibit any unusual phenotypic characteristics or cytopathic effects. DC transduced with Ad2/gp100v2 can elicit tumor-specific CTL in vitro from patients bearing gp100+ metastatic melanoma. Using a panel of gp100-derived synthetic peptides, we show that Ad2/gp100v2-transduced DC elicit Ag-specific CTL that recognize only the G209 and G280 epitopes, both of which display relatively short half-lives (∼7–8 h) on the surface of HLA-A*0201+ cells. Thus, patients with metastatic melanoma are not tolerant to gp100 Ag based on the detection of CD8+ T cells specific for multiple HLA-A*0201-restricted, gp100-derived epitopes.

Enhanced efficacy of melanoma vaccines in the absence of B lymphocytes.

Provoking a specific cellular immune response against tumor-associated antigens is a promising therapeutic strategy to treat cancers with defined antigens such as melanoma. In recent clinical trials, however, immune responses against melanoma antigens have been elicited without consistent clinical responses, suggesting the need for approaches that potentiate the specific cellular immune response. Since B lymphocytes have been reported to exert a negative effect on the cellular arm of the immune response in certain model systems, the authors compared the protective immunity elicited by melanoma antigens in B cell-deficient μMT mice to that obtained in fully immunocompetent C57BL/6 mice. Immunization with melanoma-associated antigens was accomplished using recombinant adenovirus (Ad) vectors encoding human gp100 (Ad2/gp100) or murine TRP-2 (Ad2/mTRP-2). A single dose of Ad2/gp100 or Ad2/mTRP-2 inhibited the growth of established subcutaneous B16 melanoma tumors in B cell-deficient but not wild-type C57BL/6 mice. The enhanced tumor protection observed in B cell-deficient mice appeared to be associated with potentiation of the magnitude and longevity of the specific cellular immune response. Natural killer (NK) cells were also found to be essential to the protective immune response in μMT mice because NK cell depletion with anti-asialo-GM1 antibody resulted in both the loss of tumor growth suppression and attenuation of the specific cellular immune response. The authors conclude that the protective cell-mediated immunity provoked by Ad-based cancer vaccines is enhanced in the absence of B cells, suggesting that a therapeutic regimen that includes depletion of B lymphocytes may be beneficial to cancer vaccine therapy.

Protein tyrosine phosphatase PRL-3 in malignant cells and endothelial cells: expression and function

Protein tyrosine phosphatase PRL-3 mRNA was found highly expressed in colon cancer endothelium and metastases. We sought to associate a function with PRL-3 expression in both endothelial cells and malignant cells using in vitro models. PRL-3 mRNA levels were determined in several normal human endothelial cells exposed or unexposed to the phorbol ester phorbol 12-myristate 13-acetate (PMA) and in 27 human tumor cell lines. In endothelial cells, PRL-3 mRNA expression was increased in human umbilical vascular endothelial cells and human microvascular endothelial cells (HMVEC) exposed to PMA. An oligonucleotide microarray analysis revealed that PRL-3 was among the 10 genes with the largest increase in expression on PMA stimulation. Phenotypically, PMA-treated HMVEC showed increased invasion, tube formation, and growth factor–stimulated proliferation. A flow cytometric analysis of cell surface markers showed that PMA-treated HMVEC retained endothelial characteristics. Infection of HMVEC with an adenovirus expressing PRL-3 resulted in increased tube formation. In tumor cells, PRL-3 mRNA levels varied markedly with high expression in SKNAS neuroblastoma, MCF-7 and BT474 breast carcinoma, Hep3B hepatocellular carcinoma, and HCT116 colon carcinoma. Western blotting analysis of a subset of cell line lysates showed a positive correlation between PRL-3 mRNA and protein levels. PRL-3 was stably transfected into DLD-1 colon cancer cells. PRL-3-overexpressing DLD-1 subclones were assessed for doubling time and invasion. Although doubling time was similar among parental, empty vector, and PRL-3 subclones, invasion was increased in PRL-3-expressing subclones. In models of endogenous expression, we observed that the MCF-7 cell line, which expresses high levels of PRL-3, was more invasive than the SKBR3 cell line, which expresses low levels of PRL-3. However, the MDA-MB-231 cell line was highly invasive with low levels of PRL-3, suggesting that in some models invasion is PRL-3 independent. Transfection of a PRL-3 small interfering RNA into MCF-7 cells inhibited PRL-3 expression and cell invasion. These results indicate that PRL-3 is functional in both endothelial cells and malignant cells and further validate PRL-3 as a potentially important molecular target for anticancer therapy. [Mol Cancer Ther 2006;5(2):219–29]

Protease inhibitor display M13 phage: selection of high-affinity neutrophil elastase inhibitors

We report display of the complete protease inhibitor (Kunitz) domain, BPTI, on the surface of bacteriophage M13 as a fusion to the gene III product. Phage that display BPTI bind specifically to anti-BPTI antibodies, trypsin and anhydrotrypsin. A point mutation of BPTI [Lys15-->Leu(K15L)] alters the binding specificity of fusion phage such that a human neutrophil elastase-binding phenotype is conferred while a trypsin-binding phenotype is eliminated. Phage were eluted from an immobilized protease with step gradients of decreasing pH. Phage that display Kunitz domains having higher affinity for the immobilized protease exhibit characteristic pH elution phenotypes, indicating that bound display phage can be selectively recovered from an affinity matrix. Utilization of this technology should enable the selection of remodeled protease inhibitors exhibiting novel binding specificities.