Nonmitogenic anti-CD3 monoclonal antibodies deliver a partial T cell receptor signal and induce clonal anergy.

Anti-CD3 monoclonal antibodies (mAbs) are potent immunosuppressive agents used in clinical transplantation. However, the activation-related adverse side effects associated with these mAbs have prompted the development of less toxic nonmitogenic anti-CD3 mAb therapies. At present, the functional and biochemical consequences of T cell exposure to nonmitogenic anti-CD3 is unclear. In this study, we have examined the early signaling events triggered by a nonmitogenic anti-CD3 mAb. Like the mitogenic anti-CD3 mAb, nonnmitogenic anti-CD3 triggered changes in the T cell receptor (TCR) complex, including zeta chain tyrosine phosphorylation and ZAP-70 association. However, unlike the mitogenic anti-CD3 stimulation, nonmitogenic anti-CD3 was ineffective at inducing the highly phosphorylated form of zeta (p23) and tyrosine phosphorylation of the associated ZAP-70 tyrosine kinase. This proximal signaling deficiency correlated with minimal phospholipase Cgamma-1 phosphorylation and failure to mobilize detectable Ca2+. Not only did biochemical signals delivered by nonmitogenic anti-CD3 resemble altered peptide ligand signaling, but exposure of Th1 clones to nonmitogenic anti-CD3 also resulted in functional anergy. Finally, a bispecific anti-CD3 X anti-CD4 F(ab)'2 reconstituted early signal transduction events and induced proliferation, suggesting that defective association of lck with the TCR complex may underlie the observed signaling differences between the mitogenic and nonmitogenic anti-CD3.

CD3 ligation on immature thymocytes generates antagonist-like signals appropriate for CD8 lineage commitment, independently of T cell receptor specificity.

The signals that direct differentiation of T cells to the CD4 or CD8 lineages in the thymus remain poorly understood. Although it has been relatively easy to direct differentiation of CD4 single positive (CD4+) cells using combinations of antibodies and pharmacological agents that mimic receptor engagements, equivalent stimuli do not induce efficient maturation of CD8+ cells. Here we report that, irrespective of the MHC-restriction specificity of the TCR, differentiation of mature CD8+ thymocytes can be induced by ligation of CD3 polypeptides on immature thymocytes with a F(ab')2 reagent (CD3fos-F(ab')2). The tyrosine phosphorylation patterns stimulated by CD3fos-F(ab')2 have been shown to resemble those delivered to mature T cells by antagonist peptides, which are known to direct positive selection of CD8+ cells, and we can show that this reagent exhibits potent antagonistic-like activity for primary T cell responses. Our results suggest a distinction in the signals that specify lineage commitment in the thymus. We present a model of thymocyte differentiation that proposes that the relative balance of signals delivered by TCR engagement and by p56lck activation is responsible for directing commitment to the CD8 or CD4 lineages.

Differential regulation of the two glyceraldehyde-3-phosphate dehydrogenase genes during Drosophila development.

Drosophila melanogaster contains two genes encoding glyceraldehyde-3-phosphate dehydrogenase, Gapdh-1 and Gapdh-2. The two genes are highly conserved in their coding sequences but not in their noncoding and flanking sequences. We report that both genes are expressed at higher levels in larval, late pupal, and adult stages than in embryonic, early, and midpupal stages. However, a major difference in the expression of the two genes is observed in the adult stage, during which the level of the Gapdh-1 transcript decreases over fourfold, while that of the Gapdh-2 transcript remains at a constant high level. In addition, the Gapdh-1 transcript appears highly enriched in the thorax section compared with the head and abdomen sections, while the Gapdh-2 transcript is evenly distributed. Analyses of the expression patterns of the two Gapdh hybrid genes, GAP1/2 and GAP2/1, revealed that the two genes have a distinct organization of their regulatory sequences. The principle regulatory sequences of Gapdh-2 reside upstream of the translation start, while the principle sequences specifying the level and developmental pattern of Gapdh-1 expression reside downstream of the translation start.

A recombinant, membrane-acting immunotoxin.

The anti-Tac antibody is known to bind to the p55 chain of the human interleukin 2 receptor. An immunotoxin was produced by genetically linking Clostridium perfringens phospholipase C (PLC) to the Fab domain of anti-Tac. For this purpose, the PLC gene, with its own promoter and signal sequence, was fused to the 5' end of the VHCH1 segment of the anti-Tac heavy chain gene. The anti-Tac light chain gene, with an attached bacterial signal sequence, was made part of the same transcriptional unit. Escherichia coli transformed with the construct secreted a recombinant immunotoxin, anti-Tac(Fab)-PLC, in an active form. Anti-Tac(Fab)-PLC bound to cells expressing the interleukin 2 receptor and inhibited protein synthesis, with a 50% inhibitory concentration of 0.02 nM (1.8 ng/ml).

IL-2Ralpha-Directed monoclonal antibodies provide effective therapy in a murine model of adult T-cell leukemia by a mechanism other than blockade of IL-2/IL-2Ralpha interaction.

Adult T-cell leukemia (ATL) develops in a small proportion of human T-cell lymphotrophic virus-I infected individuals. The leukemia consists of an overabundance of activated T cells, which are characterized by the expression of CD25, or IL-2Ralpha, on their cell surface. Presently, there is not an accepted curative therapy for ATL. We developed an in vivo model of ATL in non-obese diabetic/severe combined immunodeficient (NOD/ SCID) mice by introducing cells from an ATL patient (MET-1) into the mice. The leukemic cells proliferated in these mice that lack functional T, B, and natural killer (NK) cells. The MET-1 leukemic cells could be monitored by measurements of both serum soluble Tac (IL-2Ralpha) and soluble human beta2-microglobulin (beta2mu) by ELISA. The disease progressed to death in the mice after approximately 4-6 weeks. The mice developed grossly enlarged spleens and a leukemia involving ATL cells that retained the phenotype and the T-cell receptor rearrangement and human T-cell lymphotrophic virus-I integration pattern of the patient's ATL leukemia cells. This model is of value for testing the efficacy of novel therapeutic agents for ATL. The administration of humanized anti-Tac (HAT), murine anti-Tac (MAT), and 7G7/B6, all of which target IL-2Ralpha, significantly delayed the progression of the leukemia and prolonged the survival of the tumor-bearing mice. In particular, HAT induced complete remissions in 4 of 19 mice and partial remissions in the remainder. It appears that the antibodies act by a mechanism that had not been anticipated. The prevailing view is that antibodies to the IL-2Ralpha receptor have their effective action by blocking the interaction of IL-2 with its growth factor receptor, thereby inducing cytokine deprivation apoptosis. However, although both HAT and MAT block the binding of IL-2 to IL-2Ralpha of the high affinity receptor, the 7G7/B6 monoclonal antibody binds to a different epitope on the IL-2Ralpha receptor, one that is not involved in IL-2 binding. This suggested that the antibodies provide an effective therapy by a mechanism other than induction of cytokine deprivation. In accord with this view, the MET-1 cells obtained from the spleens of leukemic mice did not produce IL-2, nor did they express IL-2 mRNA as assessed by reverse transcription-PCR. Another possible conventional mechanism of action involves complement-mediated killing. However, although MAT and 7G7/B6 fix rabbit complement, HAT does not do so. Furthermore, in the presence of NOD/SCID mouse serum, there was no complement-mediated lysis of MET-1 cells. In addition, the antibodies did not manifest antibody-dependent cellular cytotoxicity with NOD/SCID splenocytes that virtually lack NK cells as the effector cells as assessed in an in vitro chromium-release assay. However, in contrast to the efficacy of intact HAT, the F(ab')2 version of this antibody was not effective in prolonging the survival of mice injected with MET-1 ATL cells. In conclusion, in our murine model of ATL, monoclonal antibodies, HAT, MAT, and 7G7/B6, appear to delay progression of the leukemia by a mechanism of action that is different from the accepted mechanism of IL-2 deprivation leading to cell death. We consider two alternatives: the first, antibody-dependent cellular cytotoxicity mediated by FcRI- or FcRIII-expressing cells other than NK cells, such as monocytes or polymorphonuclear leukocytes. The second alternative we consider involves direct induction of apoptosis by the anti-IL-2R antibodies in vivo. It has been shown that the IL-2R is a critical element in the peripheral self-tolerance T-cell suicide mechanism involved in the phenomenon of activation-induced cell death.

Synthesis of human insulin gene. VIII. Construction of expression vectors for fused proinsulin production in Escherichia coli.

We have constructed two families of plasmids suitable for the cloning of genes and for directing the synthesis of large amounts of fused proteins in Escherichia coli. The plasmids include the E. coli lac promoter and a portion of the coding sequence for beta-galactosidase, which can code for approx. 590 or 450 amino acids. The truncated beta-galactosidase gene ends with a poly-linker region at the 3' end, which can be cleaved by any one of the eight common restriction enzymes and joined to the gene coding for any desired protein. Each family includes three plasmids that enable fusion to be made in all three of the translational reading frames. We have cloned a synthetic human proinsulin gene into these plasmids, and 30% of the total E. coli protein was represented by the 590 amino acid-long truncated beta-galactosidase fused to proinsulin. The yield of proinsulin in this system is more than twice the amount produced by using a 1007 amino acid-long beta-galactosidase gene for fusion.

HuM291(Nuvion), a humanized Fc receptor-nonbinding antibody against CD3, anergizes peripheral blood T cells as partial agonist of the T cell receptor.

BACKGROUND: Humanized Fc receptor (FcR)-nonbinding antibodies against CD3 are promising immunosuppressive agents that may overcome both the neutralizing response to and the cytokine release syndrome seen with conventional monoclonal antibodies against CD3. In addition, evidence from several murine models suggests that these recombinant antibodies may actively induce T cell unresponsiveness by a mechanism other than modulation of the T cell receptor (TCR) or T cell depletion. We hypothesized that FcR-nonbinding antibodies against CD3 could induce T cell unresponsiveness by acting as partial agonist ligands of the TCR and thus, inducing T cell anergy. METHODS: To test this hypothesis, we examined the signaling and functional effects of HuM291 (Nuvion), a FcR-nonbinding humanized antibody against CD3, on primary human T cells. RESULTS: Short exposure of human peripheral blood T lymphocytes to HuM291 caused a partial agonist type of signaling through the TCR characterized by incomplete phosphorylation of TCR zeta, failure to activate ZAP-70 and to phosphorylate LAT but activation of ERK-1/-2 and subsequent up-regulation of CD69 expression. These changes correlated with a dose-dependent induction of anergy in human, primary resting T cells, which was reversed by exogenous interleukin-2. CONCLUSIONS: The tolerogenic properties of FcR-nonbinding monoclonal antibodies against CD3 correlate with its ability to reproduce the biochemical and functional effects of TCR partial agonist ligands. Thus, generation of engineered antibodies against CD3 with low TCR oligomerization potential may provide a clinically applicable partial agonist-based strategy for the prevention of polyclonal T cell responses.

HuM291, a humanized anti-CD3 antibody, is immunosuppressive to T cells while exhibiting reduced mitogenicity in vitro.

BACKGROUND: OKT3, a mouse monoclonal antibody (Ab) specific for the human CD3 complex on T cells, is a potent immunosuppressive agent used for the treatment of acute allograft rejection. The utility of the drug has been limited by a neutralizing anti-mouse Ab response and adverse side effects resulting from T cell activation and systemic cytokine release. T cell activation is caused by OKT3-mediated cross-linking of T cells and Fc receptor-bearing cells. Studies in the mouse model have shown that global T cell activation is not necessary for immunosuppression, as Fc receptor-nonbinding anti-CD3 Abs can suppress graft rejection in the absence of the activation effects seen with Fc receptor-binding Abs. Thus, a humanized anti-CD3 antibody with a low affinity for Fc receptors might improve immunosuppressive therapy by reducing the side effects associated with OKT3. METHODS: We developed a mouse monoclonal Ab, M291, which competes with OKT3 for binding to T cells. Humanized, complementary-determining region-grafted versions of M291 featuring various Fc were engineered, including a previously described IgG2 mutant deficient in Fc receptor binding (HuM291). RESULTS: Compared with OKT3 and HuM291-IgG1, HuM291 was significantly less mitogenic to T cells in vitro and induced the release of much lower levels of the cytokines tumor necrosis factor-alpha, interferon-gamma, and interleukin-10. Despite this reduction in T cell activation, HuM291 retained the ability to modulate the CD3 complex and inhibit the mixed lymphocyte reaction. CONCLUSIONS: When evaluated in vivo, HuM291 may be an immunosuppressive agent associated with less of the acute toxicity and immunogenicity seen with OKT3 therapy.

Chemical modifications of Serratia marcescens anthranilate synthase component I.

Serratia marcescens anthranilate synthase Component I (AS I) was purified from a plasmid-containing Escherichia coli strain. Residues essential for AS I function were studied by chemical modification reactions. Phenylglyoxal and 1,2-cyclohexanedione modified 2-5 arginine residues and inactivated AS I. The substrate chorismate reduced the rate of inactivation. Analysis of inactivation data indicated that 1 arginine residue is essential for activity. Histidine residues in AS I were modified by ethoxyformic anhydride and by photooxidation. Enzyme inactivation accompanied modification of histidine residues. Inactivation was prevented by substrate. Comparison of the number of carbethoxy groups incorporated between substrate-protected and unprotected AS I indicated that 1 histidine residue is required for activity. AS I was also inactivated by bromopyruvate. Substrate retarded inactivation by bromopyruvate. A differential labeling experiment indicated that the loss of AS I activity was correlated with alkylation of 1 cysteine residue. A tryptic peptide containing the essential cysteine residue was isolated. The peptide has the amino acid sequence of Ile-Cys-Gln-Ala-Gly-Ser-Arg.

Cloning and expression of the phospholipase C gene from Clostridium perfringens and Clostridium bifermentans.

The phospholipase C gene from Clostridium perfringens was isolated, and its sequence was determined. It was found that the structural gene codes for a protein of 399 amino acid residues. The NH2-terminal residues have the typical features of a signal peptide and are probably cleaved after secretion. Escherichia coli cells harboring the phospholipase C gene-containing plasmid expressed high levels of this protein in the periplasmic space. Phospholipase C purified from E. coli transformants was enzymatically active, hemolytic to erythrocytes, and toxic to animals when injected intravenously. The phospholipase C gene from a related organism, Clostridium bifermentans, was also isolated. The two phospholipase C genes were found to be 64% homologous in coding sequence. The C. bifermentans protein, however, was 50-fold less active enzymatically than the C. perfringens enzyme.

Glutamine phosphoribosylpyrophosphate amidotransferase from cloned Escherichia coli purF. NH2-terminal amino acid sequence, identification of the glutamine site, and trace metal analysis.

Glutamine 5-phosphoribosylamine pyrophosphate phosphoribosyltransferase (amidophosphoribosyltransferase) was purified in large amounts from an Escherichia coli strain harboring a purF hybrid plasmid. Purified E. coli amidophosphoribosyltransferase lacks iron as well as other trace metals as determined by x-ray fluorescence spectrometry. The NH2-terminal amino acid sequence of the enzyme was determined and is in agreement with that deduced from the DNA sequence. [6-14C] Diazo-5-oxo-norleucine (DON), an active site-directed affinity analog of glutamine, selectively inactivated the glutamine-dependent amidophosphoribosyltransferase. Inactivation was accompanied by incorporation of 1 eq of [6-14C]DON per enzyme subunit. A 10-residue cyanogen bromide peptide labeled by [6-14C]DON was isolated and sequenced. The NH2-terminal cysteine of amidophosphoribosyltransferase was determined to be the residue alkylated by [6-14C]DON. These results establish that the NH2-terminal cysteine is the active site residue required for the glutamine amide transfer function of the enzyme. The experiments reported in this and the preceding article (Tso, J. Y., Zalkin, H., van Cleemput, M., Yanofsky, C., and Smith, J. M. (1982) 257, 3525-3531) demonstrate the application of affinity labeling, rapid peptide purification by high pressure liquid chromatography, and nucleotide sequence determination of a structural gene to localize an amino acid residue, peptide fragment, or functional domain in a long protein chain.

Mechanism of inactivation of glutamine amidotransferases by the antitumor drug L-(alpha S, 5S)-alpha-amino-3-chloro-4,5-dihydro-5-isoxazoleacetic acid (AT-125).

L-(alphaS, 5S)-alpha-Amino-3-chloro-4,5-dihydro-5-isoxazoleacetic acid (AT-125), an antitumor drug isolated from Streptomyces sviceus, is an active site-directed affinity analog of glutamine. It selectively inactivates the glutamine-dependent activities of two bacterial glutamine amidotransferases, anthranilate synthase and glutamate synthase. A reversible noncovalent complex is formed prior to irreversible enzyme modification. Inactivation of anthranilate synthase results from incorporation of approximately 1 eq of AT-125/enzyme protomer. Active site cysteine-83 in Serratia marcescens anthranilate synthase Component II is the residue alkylated by AT-125. Anthranilate synthase is rapidly inactivated by AT-125 IN S. marcescens cells. In vivo inactivation is by the same mechanism as in vitro.

Primary structure of Serratia marcescens anthranilate synthase component II.

The amino acid sequence of anthranilate synthase component II (AS II) from Serratia marcescens was determined. The cysteine residue essential for glutamine utilization was alkylated selectively by iodo [1-14C]acetamide prior to separation of the two protein components of anthranilate synthase. The isolated AS II then was subjected to cleavage by cyanogen bromide and by trypsin after citraconylation to obtain overlapping fragments. AS II is a single polypeptide chain of 192 residues having a calculated molecular weight of 20,956. The active site region is virtually identical to that of the Pseudomonas putida AS II enzyme (Kawamura, M., Keim, P.S., Goto, Y., Zalkin, H., and Heinrikson, R.L. (1978) J. Biol. Chem. 253, 4659-4668). Overall amino acid sequence similarity is 43%.

Structure of two unlinked Drosophila melanogaster glyceraldehyde-3-phosphate dehydrogenase genes.

Two Drosophila genes that code for the enzyme glyceraldehyde-3-phosphate dehydrogenase (Gapdh) have been isolated and their structures determined by DNA sequence analysis. The two genes, Gadph-1 and Gapdh-2, are homologous to each other in their coding regions but differ entirely in the 5' and 3' flanking regions. Both genes are functionally expressed in adult flies as determined by Northern blot analysis using gene-specific probes. Gapdh-1 is mapped by in situ hybridization at position 43E-F on the right arm of the second chromosome and Gapdh-2 at position 13F on the left arm of the X chromosome. Transcription initiation sites as well as polyadenylation sites for both Gapdh transcripts have also been determined. Gapdh-1 lacks a sequence homologous to the TATA box in its -30-base pair region that is characteristic of many RNA polymerase II transcribed promoters. In contrast, Gapdh-2 contains a consensus TATA box sequence as well as a CAAT box in its promoter region. Furthermore, a sequence element ATTTGCAT (dc) and nontandem multiple direct repeats have been found in the -35 to -155-base pair 5' flanking region. Other than the intron located in the 5' noncoding region of Gapdh-2, both genes lack intervening sequences.

Human IgG2 variants of chimeric anti-CD3 are nonmitogenic to T cells.

The mouse anti-human CD3 mAb OKT3 is a potent immunosuppressive agent used for the treatment of acute transplant rejection. OKT3 therapy is associated with acute toxicity resulting from in vivo T cell activation and systemic cytokine release, and a human anti-mouse Ab response. T cell activation is thought to be triggered by CD3 cross-linking mediated by the Abs bridging T cells and Fc receptor-bearing cells. Recent studies in a mouse model indicate that anti-mouse CD3 Abs with low affinity for Fc receptors can achieve immunosuppression without T cell activation, toxicity, or an anti-Ab response. To obtain an analogous Ab to improve the current anti-human CD3 therapy, a humanized Ab with low affinity for Fc receptors is needed. In this study, we introduced mutations into the upper CH2 region of IgG2 and expressed the altered Fc as chimeric OKT3 Abs. Compared with chimeric OKT3 IgG1, IgG2, IgG3, and IgG4, the IgG2 mutants were less mitogenic to T cells, and they did not induce the release of TNF-alpha, IFN-gamma, or IL-2. In parallel, we observed no functional interaction of the IgG2 mutant Abs with K562 cells, which express the IgG2-binding Fc receptor on their surface. Despite no measurable T cell activation, the mutant Abs could still modulate the CD3 complex. When coupled to a humanized anti-CD3, the IgG2 variant may provide a drug with less acute toxicity and immunogenicity, but may still retain potent immunosuppressive properties.

Formation of a bispecific antibody by the use of leucine zippers.

A new method is described for the production of bispecific F(ab')2 heterodimers using leucine zippers. Two heterodimer-forming "zipper" peptides derived from the Fos and Jun proteins were respectively linked to the Fab' portions of two different mAb by gene fusion. The antibodies used were 145-2C11, which binds to murine CD3, and anti-Tac, which binds to the p55 chain of the human IL-2R. Anti-Tac Fab'-Jun and anti-CD3 Fab'-Fos were expressed individually as F(ab'-zipper)2 homodimers in the mouse myeloma cell line Sp2/0. When these homodimers were reduced at the hinge region to form monomers and then reoxidized together, the resulting end products were mostly F(ab'-zipper)2 heterodimers. Bispecific anti-CD3 x anti-Tac F(ab'-zipper)2 heterodimers produced by this method were shown to be highly effective in recruiting cytotoxic T cells to lyse IL-2R-bearing HuT-102 cells in vitro.

Construction of a human Ig combinatorial library from genomic V segments and synthetic CDR3 fragments.

A naive combinatorial Ig library was constructed from semi-synthetic V genes consisting of human genomic V segments and synthetic CDR3 fragments. VH and V kappa segments were amplified from human genomic DNA by polymerase chain reaction using V subgroup-specific primers. The amplified VH and V kappa segments were combined with synthetic oligonucleotides containing a J region and CDR3 with amino acid sequence variations, resulting in complete V genes. These V genes were cloned into a phagemid expression vector in a single-chain form fused to the carboxyl-terminus of the M13 minor coat protein III. Phagemid particles displaying the single chain hybrid proteins on their surface were screened with Con A as Ag. Several clones showing specific binding to Con A were obtained after four rounds of selection and were further analyzed for their binding properties and DNA sequences. This method provides a novel way to create a naive combinatorial library without using mRNA from B lymphocytes as template. The method should be useful to isolate human antibodies that react with self-Ag.

Preparation of a bispecific F(ab')2 targeted to the human IL-2 receptor.

The heterodimer-forming leucine zippers Fos and Jun can efficiently mediate the formation of bispecific F(ab')2 when they are fused separately to two different Fab' fragments. This recombinant method can be used in conjunction with the humanization process to yield humanized bispecific F(ab')2. The potential immunogenicity of the leucine zippers can be eliminated by their removal using pepsin digestion. This method has been scaled up to produce hundreds of milligrams of a bispecific F(ab')2 that targets two subunits of the human IL-2 receptor.