Characterization of murine monoclonal anti-endothelial cell antibodies (AECA) produced by idiotypic manipulation with human AECA

The IgG fraction of human anti-endothelial cell antibodies (AECA) obtained from a patient with Wegener's granulomatosis was used as immunogen to raise AECA mAb in mice selected among those which developed vasculitis-like lesions after immunization. Three mAb (BGM, 3C8 and 7G2), selected by cyto-ELISA and flow cytometry analyses, featured a specific reactivity with human umbilical vein endothelial cells (HUVEC) and the mouse endothelial cell line H5V; on the contrary, HEp2 cells, the murine melanoma B16 cell line, the extracellular matrix as well as several other antigens tested were not recognized. BGM mAb, an IgG3 precipitating a 70 kDa structure from HUVEC, was able to induce endothelial cells to secrete amounts of IL-6 significantly higher than irrelevant controls or mAb binding different endothelial antigens (i.e. CD31, CD29, ICAM-1 and HLA class I). BGM mAb induced significant levels of antibody-dependent cell cytotoxicity (13 +/- 2.5 versus 0.6 +/- 0.03%). To the best of our knowledge, BGM is the first murine mAb specific for human endothelial cells generated by idiotypic manipulation; secondly, its biological properties further support the notion of a pathogenic role for AECA in autoimmune-mediated diseases.     

B7-1 and B7-2 monoclonal antibodies modulate the severity of murine Lyme arthritis

We assessed the role of B7-1 and B7-2 costimulatory molecules on the course of murine Lyme borreliosis because experimental Lyme arthritis is dependent, at least partially, upon the development of the host immune response and these costimulatory molecules have been implicated in CD4+ T-cell differentiation. We demonstrated that Borrelia burgdorferi infection upregulated the surface expression of B7-1 and B7-2 in macrophages and B7-2 expression in B cells. Anti-B7-2 monoclonal antibody (MAb) or both anti-B7-2 and anti-B7-1 MAbs produced a dose-dependent increase in the severity of Lyme arthritis in C3H/HeN mice. In contrast, the administration of an anti-B7-1 MAb reduced the degree of arthritis. These effects occurred independently of significant alteration in B. burgdorferi-specific immune responses, including splenocyte proliferative responses to B. burgdorferi, B. burgdorferi antibody levels and specificity, and mRNA levels of gamma interferon, interleukin-4 (IL-4), IL-10, and IL-12 in the spleen. These results demonstrate that signaling delivered by B7-1 and B7-2 plays a role in determining the severity of acute murine Lyme arthritis.     

Mechanism of cryoprecipitation of anti-blood group B murine monoclonal antibodies

In our experience, some examples of mouse monoclonal antibodies of anti-B blood group specificity develop a precipitate when stored at 4 degrees C. This poses problems during the preparation of blood grouping reagents containing anti-B, and in the storage and use of such reagents. Here we show that this problem can be circumvented by alteration of the glycan moiety of the secreted immunoglobulin, either by glycosidase treatment of the partially purified immunoglobulin, or by the addition of glycan processing inhibitors to the hybridoma cell cultures. These findings have importance for the manufacture of monoclonal antibodies, and highlight a possible new role for carbohydrate in immunoglobulin interaction and immune complex formation.     

Establishment of anti-idiotype monoclonal antibodies to human anti lung cancer monoclonal antibody (4G12)

The object of this study is to establish anti-idiotype (anti-ID) monoclonal antibodies (MoAbs) against a human MoAb (4G12) that highly reacts with lung squamous cell carcinomas. Two murine anti-ID MoAbs (2H1 and 2B12) were established by hybridoma technology. They showed specific reactivity with 4G12 but not with 3H12 human MoAb, human IgM and human IgG. These two MoAbs demonstrated more than 90% of binding inhibition of 4G12 to lung squamous cell cancer cell line (PC10). Moreover, cross inhibition test showed that 2H1 and 2B12 detect different idiotypes of 4G12 each other. Furthermore, specific reactivity of anti 2H1 and anti 2B12 sera to PC10 were observed by cell binding ELISA. These two anti-ID MoAbs had internal image of the original antigen.     

Characterization of haemagglutinin-esterase protein (HE) of murine corona virus DVIM by monoclonal antibodies

We analyzed the characteristics of seven monoclonal antibodies (mAbs) raised against purified HE (hemagglutinin-esterase) glycoprotein of the murine coronavirus DVIM (diarrhea virus of infant mice). Immunocrossreaction of these mAbs with JHM and/or MHV-S suggest that antigenic epitopes of HE of DVIM are similar to those of JHM and/or MHV-S. Four mAbs (1b4, 3a28, 4c19, 10b7), designated as group A mAbs, strongly inhibited both HA and AE activities. On the other hand, three mAbs (5a3, 6a6, 13a4), referred to as group B, had a comparatively weak HA inhibition activity. These results indicate that the antigenic epitopes of this glycoprotein can be classified into at least two groups and that the functional sites of HA and AE activities are similar but not identical. Neutralizing activity was shown in group A mAbs exclusively, suggesting that the ratio of HA and/or AE activities may play important roles in the cell fusion activity of DVIM-infected cells.     

Homotypic aggregation of ovine leucocyte induced by anti-leucosialin (CD43) monoclonal antibodies

Four monoclonal antibodies directed against ovine/bovine leucocyte antigens (Co.44B8, Co.11F10, Co.33B3 and Co.26F4) were produced and all recognized an antigen with an apparent molecular weight of 105 kDa under reducing conditions. This molecule was considered to be the ovine/bovine analogue of human, rat and mouse CD43 or leucosialin. This conclusion was based on (a) the cellular distribution of the antigen which was similar to that described in other species, (b) a decrease of apparent M(r) (150 kDa) after treatment with neuraminidase, and (c) production of cell aggregation by the four monoclonal antibodies and the influence of temperature and metabolic inhibitors on this phenomenon. Co.44B8, Co.11F10, Co.33B3 and Co.26F4 all recognized and immunoprecipitated the molecule after neuraminidase treatment of the cells. The data indicate an important function of CD43 as a mediator of cell aggregation for ruminant cells.     

Activation of human complement by totally human monoclonal antibodies

A uniquely developed series of totally human monoclonal antibodies (mAbs) were examined for their complement fixing properties in comparison to human myeloma preparations and to commercially available human polyclonal immunoglobulins. C3b and C4b deposition was measured using a kinetic ELISA technique. When the IgG myeloma proteins were tested for classical pathway activation, our findings were similar to those previously described, where IgG1 and IgG3 were more potent activators of the classical pathway than IgG2 and IgG4. However, those same studies determined that IgG2 was the best activator of the alternative pathway followed by IgG1 and IgG3 while IgG4 does not activate complement via either pathway. In our studies of alternative pathway activation, the IgG2 myeloma exhibited strong activation of the alternative pathway, but, at levels lower than the other three IgG subtypes. Using this test system, we examined the complement activating potential of four totally human mAbs that were constructed from the peripheral blood lymphocytes of a colon carcinoma patient in long term remission. We found that our uniquely constructed totally human IgG2 mAbs (A3, E1, F6 and F8) were able to activate complement by both the classical and alternative pathways to varying degrees. In addition, we found that the complement activating ability of the human mAbs was greater than that of the human IgG2 myeloma immunoglobulins or normal human IgG2 preparations. This study represents the first report of complement activation by totally human mAbs and confirms more recent findings which indicate that levels of complement activation by human IgG immunoglobulins cannot be predicted based solely on their subclass identity.     

Apoptosis of malignant human B cells by ligation of CD20 with monoclonal antibodies

CD20 is a nonglycosylated 33 to 37 kD phosphoprotein involved in B-cell signaling that subserves important functions in the regulation of B-cell proliferation and differentiation. In addition, this B-cell surface antigen has been shown recently to be an effective target for immunotherapy of B-cell malignancies using chimeric (mouse/human) or radiolabeled murine monoclonal anti-CD20 antibodies. In this report we show that extensive crosslinking of CD20 with murine anti-CD20 monoclonal antibodies (MoAbs) in the presence of either goat anti-mouse IgG or Fc receptor (FcR)-expressing cells directly inhibits B-cell proliferation, induces nuclear DNA fragmentation, and leads to cell death by apoptosis. The apoptotic effects of these MoAbs can be inhibited by chelation of extracellular or intracellular Ca2+ by EGTA or Bapta AM, indicating that anti-CD20-mediated apoptosis may be related to changes in Ca2+ concentration. These findings suggest that ligation of CD20 in vivo by anti-CD20 antibodies in the presence of FcR-expressing cells may initiate signal transduction events that induce elevation of [Ca2+]i and lead to apoptosis of malignant B cells, thereby contributing to the impressive tumor regressions observed in mouse models and clinical trials using anti-CD20 MoAbs.     

Generation of neutralizing human monoclonal antibodies against parvovirus B19 proteins

Infections caused by human parvovirus B19 are known to be controlled mainly by neutralizing antibodies. To analyze the immune reaction against parvovirus B19 proteins, four cell lines secreting human immunoglobulin G monoclonal antibodies (MAbs) were generated from two healthy donors and one human immunodeficiency virus type 1-seropositive individual with high serum titers against parvovirus. One MAb is specific for nonstructural protein NS1 (MAb 1424), two MAbs are specific for the unique region of minor capsid protein VP1 (MAbs 1418-1 and 1418-16), and one MAb is directed to major capsid protein VP2 (MAb 860-55D). Two MAbs, 1418-1 and 1418-16, which were generated from the same individual have identity in the cDNA sequences encoding the variable domains, with the exception of four base pairs resulting in only one amino acid change in the light chain. The NS1- and VP1-specific MAbs interact with linear epitopes, whereas the recognized epitope in VP2 is conformational. The MAbs specific for the structural proteins display strong virus-neutralizing activity. The VP1- and VP2-specific MAbs have the capacity to neutralize 50% of infectious parvovirus B19 in vitro at 0.08 and 0.73 microgram/ml, respectively, demonstrating the importance of such antibodies in the clearance of B19 viremia. The NS1-specific MAb mediated weak neutralizing activity and required 47.7 micrograms/ml for 50% neutralization. The human MAbs with potent neutralizing activity could be used for immunotherapy of chronically B19 virus-infected individuals and acutely infected pregnant women. Furthermore, the knowledge gained regarding epitopes which induce strongly neutralizing antibodies may be important for vaccine development.     

Dual recognition of lipid A and DNA by human antibodies encoded by the VH4-21 gene: a possible link between infection and lupus

The VH4-21 (V4-34) gene segment, a member of the VH4 family, is expressed early in B-cell maturation and is utilized by approximately 6% of normal adult B lymphocytes. This prevalence indicates an importance of VH4-21 in the B-cell repertoire. The gene also encodes certain autoantibodies being mandatory for pathological IgM anti-red cell antibodies directed against the I/i antigen, and also capable of encoding anti-DNA antibodies. Recognition of I/i antigen or DNA appears to be via two distinct sites on VH, with I/i binding mediated by sequences in the framework region, and DNA binding correlating with the presence of positively charged amino acids in complementarity-determining region 3. However, these positively charged residues appear to suppress the ability of the framework region to interact with I/i, rendering a single sequence monospecific for I/i or DNA. The IgM anti-DNA antibodies also recognize bacterial lipid A, whereas the anti-I/i antibodies do not, indicating that CDR3 may be involved in binding the negatively charged lipid A. Structural similarities between the DNA backbone and lipid A provide a possible explanation for this cross-reactivity. This dual recognition of bacterial antigen and autoantigen provides a potential link between infection and autoimmunity.     

Metabolism of 17-(allylamino)-17-demethoxygeldanamycin (NSC 330507) by murine and human hepatic preparations

17-(Allylamino)-17-demethoxygeldanamycin (17AAG), a compound that is proposed for clinical development, shares the ability of geldanamycin to bind to heat shock protein 90 and GRP94, thereby depleting cells of p185erbB2, mutant p53, and Raf-1. Urine and plasma from mice treated i.v. with 17AAG contained six materials with absorption spectra similar to that of 17AAG. Therefore, in vitro metabolism of 17AAG by mouse and human hepatic preparations was studied to characterize: (a) the enzymes responsible for 17AAG metabolism; and (b) the structures of the metabolites produced. These materials had retention times on high-performance liquid chromatography of approximately 2, 4, 5, 6, 7, and 9 min. When incubated in an aerobic environment with 17AAG, murine hepatic supernatant (9000 x g) produced each of these compounds; the 4-min metabolite was the major product. This metabolism required an electron donor, and NADPH was favored over NADH. Metabolic activity resided predominantly in the microsomal fraction. Metabolism was decreased by approximately 80% in anaerobic conditions and was essentially ablated by CO. Microsomes prepared from human livers produced essentially the same metabolites as produced by murine hepatic microsomes, but the 2-min metabolite was the major product, and the 4-min metabolite was next largest. There was no metabolism of 17AAG by human liver cytosol. Metabolism of 17AAG by human liver microsomes also required an electron donor, with NADPH being preferred over NADH, was inhibited by approximately 80% under anaerobic conditions, and was essentially ablated by CO. Liquid chromatography/mass spectrometry analysis of human and mouse in vitro reaction mixtures indicated the presence of materials with molecular weights of 545, 601, and 619, compatible with 17-(amino)-17-demethoxygeldanamycin (17AG), an epoxide, and a diol, respectively. The metabolite with retention time of 4 min was identified as 17AG by cochromatography and mass spectral concordance with authentic standard. Human microsomal metabolism of 17AAG was inhibited by ketoconazole, implying 3A4 as the responsible cytochrome P450 isoform. Incubation of 17AAG with cloned CYP3A4 produced metabolites 4 and 6. Incubation of 17AAG with cloned CYP3A4 and cloned microsomal epoxide hydrolase produced metabolites 2 and 4, with greatly decreased amounts of metabolite 6. Incubation of 17AAG with human hepatic microsomes and cyclohexene oxide, a known inhibitor of microsomal epoxide hydrolase, did not affect the production of metabolite 4 but decreased the production of metabolite 2 while increasing the production of metabolite 6. These data imply that metabolite 2 is a diol and metabolite 6 is an epoxide. Mass spectral fragmentation patterns and the fact that 17AG is not metabolized argue for the epoxide and diol being formed on the 17-allylamino portion of 17AAG and not on its ansamycin ring. These data have implications with regard to preclinical toxicology and activity testing of 17AAG as well as its proposed clinical development because: (a) production of 17AG requires concomitant production of acrolein from the cleaved allyl moiety; and (b) 17AG, which was not metabolized by microsomes, has been described as being as active as 17AAG in decreasing cellular p185erbB2.     

Epitope-specific monoclonal antibodies against human C-terminal procollagen alpha1(III)-propeptide

We have generated monoclonal antibodies against recombinant C-terminal human procollagen alpha1(III) propeptide (PIIICP), produced in E. coli in high yields. The monoclonal antibodies were screened for epitope specificity using recombinant truncated PIIICP. Several antibodies were identified which recognized different regions of the PIIICP molecule. The ability of the antibodies to detect PIIICP antigens in human cell line lysates and supernatants was demonstrated. As PIIICP antigens are a key marker of extracellular matrix metabolism, the monoclonal antibodies described here should be of value for clinical and basic research.     

HLA (A*0201) mimicry by anti-idiotypic monoclonal antibodies

Soluble MHC Ags and anti-Id (anti-anti-MHC) Abs have both been shown to inhibit MHC alloantigen-specific B cell responses in vivo. We hypothesized that some anti-idiotypic Abs function as divalent molecular mimics of soluble HLA alloantigen. To test this idea, we studied two well-defined anti-idiotypic mAbs, T10-505 and T10-938, elicited in syngeneic BALB/c mice by immunization with CRll-351, an HLA-A2,24,28-specific mAb. Each anti-Id induced "Ab-3" Abs in rabbits that cross-reacted with HLA-A2 but not with HLA-B Ags. Furthermore, each anti-Id could bind to and block Ag recognition by Ha5C2.A2, a human homologue of mAb CRll-351. Both anti-Id mAb displayed weak reactivity with the human mAb SN66E3, which recognized an overlapping but distinct determinant of HLA-A2 Ags; neither reacted with human mAb MBW1, which recognized a nonoverlapping HLA-A2 determinant. Amino acid sequence comparison of mAb CRII-351 heavy and light chain variable region complementarity-determining regions (CDRs) with those of mAb Ha5C2.A2 and SN66E3 revealed short regions of homology with both human mAb; a large insert in the light chain CDR1 of mAb SN66E3 distinguished it from both CRll-351 and Ha5C2.A2. The amino acid sequences of mAb T10-505 and T10-938, which differed markedly from each other, revealed no homology to the alpha2 domain sequence of HLA-A*0201 that contains the CRll-351 mAb-defined epitope. We conclude that structurally different anti-Id Abs can mimic a polymorphic conformational epitope of an HLA Ag. In the case of T10-505 and T10-938 mimicry was not based on exact replication of the epitope by the hypervariable loops of the anti-Id mAb.     

Cytotoxicity and cell-proliferation induced by the nephrocarcinogen hydroquinone and its nephrotoxic metabolite 2,3,5-(tris-glutathion-S-yl)hydroquinone

Hydroquinone, an intermediate used in the chemical industry and a metabolite of benzene, is a nephrocarcinogen in the 2-year National Toxicology Program bioassay in male Fischer 344 rats. Current evidence suggests that certain chemicals may induce carcinogenesis by a mechanism involving cytotoxicity, followed by sustained regenerative hyperplasia and ultimately tumor formation. Glutathione (GSH) conjugates of a variety of hydroquinones are potent nephrotoxicants, and we now report on the effect of hydroquinone and 2,3,5-(tris-glutathion-S-yl)hydroquinone, on site-selective cytotoxicity and cell proliferation in rat kidney. Male Fischer 344 rats (160-200 g) were treated with hydroquinone (1.8 mmol/kg or 4.5 mmol/kg, p.o.) or 2,3,5-(tris-glutathion-S-yl)hydroquinone (7.5 micromol/kg; 1.2-1.5 micromol/rat, i.v.), and blood urea nitrogen (BUN), urinary gamma-glutamyl transpeptidase (gamma-GT), alkaline phosphatase (ALP), glutathione-S-transferase (GST) and glucose were measured as indices of nephrotoxicity. Hydroquinone (1.8 mmol/kg, p.o.) is nephrotoxic in some rats, but not others, but cell proliferation (BrDU incorporation) in proximal tubular cells of the S3M region correlates with the degree of toxicity in individual rats. At 4.5 mmol/kg, hydroquinone causes significant increases in the urinary excretion of gamma-GT, ALP and GST. Pretreatment of rats with acivicin prevents hydroquinone-mediated nephrotoxicity, indicating that toxicity is dependent on the formation of metabolites that require processing by gamma-GT. Consistent with this view, 2,3,5-(tris-glutathion-S-yl)hydroquinone, a metabolite of hydroquinone, causes increases in BUN, urinary gamma-GT and ALP, all of which are maximal 12 h after administration of 2,3,5-(tris-glutathion-S-yl)hydroquinone. In contrast, the maximal excretion of GST and glucose occurs after 24 h. By 72 h, BUN and glucose concentrations return to control levels, while gamma-GT, ALP and GST remain slightly elevated. Examination of kidney slices by light microscopy revealed the presence of tubular necrosis in the S3M segment of the proximal tubule, extending into the medullary rays. Cell proliferation rates in this region were 2.4, 6.9, 15.3 and 14.3% after 12, 24, 48 and 72 h, respectively, compared to 0.8-2.4% in vehicle controls. Together with the metabolic data, the results indicate a role for hydroquinone-thioether metabolites in hydroquinone toxicity and carcinogenicity.     

Anti-CD4 monoclonal antibodies suppress murine collagen-induced arthritis only at the time of primary immunisation

We have examined the ability of a mixture of two anti-CD4 mAbs to protect against collagen-induced arthritis. Anti-CD4 mAbs, administered around the time of primary immunisation with type II collagen in adjuvant, reduced the subsequent incidence of arthritis from 67 to 16% (P < 0.01 by Fisher exact test). However, anti-CD4 treatment 3 weeks after the primary immunisation did not significantly affect the incidence of arthritis. This result extends earlier findings concerning the lack of efficacy of anti-CD4 treatment in established collagen-induced arthritis. Next, the ability of anti-CD4 treatment to induce tolerance to bovine type II collagen (and hence protect against arthritis) was evaluated using a regime known to be capable of inducing tolerance to human gamma-globulin. Anti-CD4 treatment completely failed to induce tolerance to type II collagen, as judged by levels of anti-collagen antibody, or protect against collagen-induced arthritis. These findings highlight the potential limitations of anti-CD4 mAb depleting treatment in immunotherapy.     

Telomerase activity is induced in human peripheral B lymphocytes by the stimulation to antigen receptor

To understand the molecular events for the proliferation of B cells, we studied the induction of telomerase activity in vitro after stimulation to B-cell antigen receptor (BCR) on human peripheral B cells. Although unstimulated purified B cells of tonsils and peripheral blood from healthy volunteers do not express detectable telomerase activity, anti-IgM beads induce telomerase activity in these B cells. Soluble anti-IgM antibody (Ab) alone does not induce telomerase activity, but the second signal, given by either one of the cytokines of interleukin-2 (IL-2), IL-4, and IL-13 or by anti-CD40 monoclonal Ab (MoAb), is effective as the costimulation for the induction of the activity. Stimulation with anti-IgM Ab and anti-CD40 MoAb induces telomerase activity in most mature B cells of the tonsils and peripheral blood. The stimuli to both IgM and IgD receptors similarly induce the activity. Induction of telomerase activity is accompanied with the proliferation of B cells, but is not absolutely correlated with the extent of B-cell growth. Phorbol dibutylate (PDB) plus calcium (Ca) ionophore (PDB/Ca), which replace the activation through BCR and the costimulatory molecules, also induce telomerase activity. Moreover, it is suggested that phosphoinositide (PI) 3-kinase plays a role for the induction of telomerase activity in B cells stimulated with anti-IgM Ab and anti-CD40 MoAb. These results suggest that telomerase activity is induced in the B-cell activation of the antigen specific immune response.     

Protection against lethal endotoxemia by anti-lipid A murine monoclonal antibodies: comparison of efficacy with that of human anti-lipid A monoclonal antibody HA-1A

The protective capacities of murine anti-lipid A monoclonal antibodies (MAbs) 8-2 and 26-20 were examined and compared with those of the human MAb HA-1A with respect to inhibition of lipopolysaccharide (LPS) priming of human polymorphonuclear leukocytes (PMNL) in vitro and protection against lethal endotoxemia in mice. HA-1A did not prevent the priming effect of either rough or smooth LPS, while MAb 26-20 effectively inhibited LPS priming of human PMNL. Also, both murine MAbs protected mice against an otherwise lethal challenge with rough Re LPS of S. minnesota R595 as well as with smooth LPS of E. coli O111:B4. HA-1A exerted no protection against the lethal effects of Re LPS in this in vivo model. The enhanced survival in mice by treatment with MAbs 8-2 and 26-20 was associated with decreased levels of LPS-induced tumor necrosis factor. Neutralization of lipid A as a mechanism of protection was strongly suggested by efficacious inhibition of LPS priming of human PMNL by MAbs 8-2 and 26-20 in vitro.     

Addressing the tertiary structure of human parathyroid hormone-(1-34)

Parathyroid hormone (PTH) regulates mineral metabolism and bone turnover by activating specific receptors located on osteoblastic and renal tubular cells and is fully functional as the N-terminal 1-34 fragment, PTH-(1-34). Previously, a "U-shaped" conformation with N- and C-terminal helices brought in close proximity by a turn has been postulated. The general acceptance of this hypothesis, despite limited experimental evidence, has altered the direction of the design of PTH-analogs. Examining the structure of human PTH-(1-34) under conditions that encompass the different environments the hormone may experience in the approach to and interaction with the G-protein-coupled receptor (including benign aqueous and saline solutions and in the presence of dodecylphosphocholine), we observe no evidence for a U-shape conformation or any tertiary structure. Instead, the N- and C-terminal helical domains, which vary in length and stability depending on the conditions, are separated by a highly flexible region of undefined conformation. These observations are in complete accord with recent conformational studies of PTH-related protein analogs containing lactams (Mierke, D. F., Maretto, S., Schievano, E. , DeLuca, D., Bisello, A., Mammi, S., Rosenblatt, M., Peggion, E., and Chorev, M. (1997) Biochemistry 36, 10372-10383) or a model amphiphilic alpha-helix (Pellegrini, M., Bisello, A., Rosenblatt, M., Chorev, M., and Mierke, D. F. (1997) J. Med. Chem. 40, 3025-3031). Reliable structural data from different environmental conditions are absolutely requisite for the next step in the design of non-peptide PTH analogs.     

A comparison of two murine monoclonal antibodies humanized by CDR-grafting and variable domain resurfacing

The variable domain resurfacing and CDR-grafting approaches to antibody humanization were compared directly on the two murine monoclonal antibodies N901 (anti-CD56) and anti-B4 (anti-CD19). Resurfacing replaces the set of surface residues of a rodent variable region with a human set of surface residues. The method of CDR-grafting conceptually consists of transferring the CDRs from a rodent antibody onto the Fv framework of a human antibody. Computer-aided molecular modeling was used to design the initial CDR-grafted and resurfaced versions of these two antibodies. The initial versions of resurfaced N901 and resurfaced anti-B4 maintained the full binding affinity of the original murine parent antibodies and further refinements to these versions described herein generated five new resurfaced antibodies that contain fewer murine residues at surface positions, four of which also have the full parental binding affinity. A mutational study of three surface positions within 5 A of the CDRs of resurfaced anti-B4 revealed a remarkable ability of the resurfaced antibodies to maintain binding affinity despite dramatic changes of charges near their antigen recognition surfaces, suggesting that the resurfacing approach can be used with a high degree of confidence to design humanized antibodies that maintain the full parental binding affinity. By comparison CDR-grafted anti-B4 antibodies with parental affinity were produced only after seventeen versions were attempted using two different strategies for selecting the human acceptor frameworks. For both the CDR-grafted anti-B4 and N901 antibodies, full restoration of antigen binding affinity was achieved when the most identical human acceptor frameworks were selected. The CDR-grafted anti-B4 antibodies that maintained high affinity binding for CD19 had more murine residues at surface positions than any of the three versions of the resurfaced anti-B4 antibody. This observation suggests that the resurfacing approach can be used to produce humanized antibodies with reduced antigenic potential relative to their corresponding CDR-grafted versions.