Search

A natural N-linked glycosylation site (Asn-Val-Thr) at amino acid positions 18-20 (Kabat's numbering) was identified in the framework-1 (FR-1) region of the light chain variable (V kappa) domain of a murine anti-B cell lymphoma Ab, LL-2. Our earlier studies demonstrated that no contact between the V kappa-appended oligosaccharide and the Ag binding site was evident, because glycosylation at this site did not affect the Ag binding property of the Ab. By using the murine LL-2 F(ab')2 fragment (which is devoid of constant region-appended oligosaccharide) as substrate, as much as five bifunctional chelator molecules per F(ab')2 fragment could be site specifically conjugated at the V kappa-appended carbohydrate moiety with no reduction in immunoreactivity. The resulting conjugates labeled efficiently with both 90Y and 111In, with no significant effect on Ab affinity. In contrast, conjugation of less than five chelates/Ab fragment randomly at lysine residues resulted in a three- to fivefold reduction in affinity. By a single Arg to Asn mutation, an N-linked glycosylation site similar to that of LL-2 was introduced in the FR-1 segment of a nonglycosylated, humanized anti-carcinoembryonic Ag (CEA) Ab, MN-14 (hMN-14). Glycosylation at the engineered carbohydrate-addition site was demonstrated by SDS-PAGE analysis. Neither glycosylation nor site-specific conjugation of chelate at the V kappa-appended carbohydrate moiety resulted in the loss of immunoreactivity. The glycosylated hMN-14 conjugate labeled efficiently with 90Y.

MRL/Mp(-)+/+ mice produce antinuclear antibodies and develop a spontaneous autoimmune syndrome with lupus-like nephritis. We obtained a panel of seven histone-reactive IgG mAb from a single MRL/Mp(-)+/+ mouse. These antibodies do not react significantly with DNA or individual histones, but bind strongly to the histone H2A-H2B dimer and even more strongly to the H2A-H2B-DNA complex. These antibodies also bind to whole nuclei when tested by immunofluorescence, indicating that they recognize an epitope accessible in chromatin. The V region sequences of these antibodies have been determined. The H chain third complementarity-determining regions of these antibodies are similar to those found in anti-DNA antibodies even though the antibodies in our panel do not react with DNA in the absence of histones, suggesting that DNA is part of the subnucleosome epitope. Several of these antibodies are clonally related, supporting the hypothesis that the activation of these clones is Ag-driven. Analysis of the sequences of these antibodies indicates that they derive from autoreactive B cells that were clonally expanded and whose V region genes have undergone numerous somatic mutations.

Antibody humanization has eliminated or reduced the human antimouse antibody response associated with the administration of murine antibodies. We have successfully humanized three different antibodies: (a) hMN-3 (granulocyte targeting); (b) hMu-9 (colorectal cancer targeting); and (c) hWI2 (anti-idiotype to the anti-carcinoembryonic antigen antibody MN-14). All humanized antibodies demonstrated immunoreactivities comparable to their parent counterparts. Previously, we reported the generation of high productivity cell lines for hMN-14 and hLL2 using the amplifiable vector pdHL2. Through amplification, selection, and cloning procedures, cell lines capable of large scale production were established, and further enhancement of production was achieved by a fed-perfusion bioreactor process. Using a similar and improved approach, we have enhanced the production of the above-mentioned humanized antibodies by gene amplification induced by a stepwise increase in the concentration of methotrexate in the culture media. A reliable IgG determination method is essential to monitor amplification, especially at the final cloning stage, for the selection of the subclones with the highest productivity. We found that measurement of humanized IgG concentration in culture media supplemented with more than 1 microM methotrexate by a standard ELISA assay could be unreliable and misleading. Whereas the determination of antibody by adsorption/elution on protein A from a 100-ml culture is accurate and reproducible, the method is time-consuming, tedious, and labor intensive. We have recently developed a Western blot assay that enables us to monitor the productivity of the cultures. The assay is simple and sensitive, and it makes simultaneous determinations of relative antibody production from individual clones at the 96-well stage feasible. With this method, amplification, cloning, and adaptation to serum-free conditions of multiple cell lines can be monitored in an efficient manner.

Antibody (Ab) fragments are preferred agents for imaging applications because of their rapid clearance from the blood, thereby providing high tumor:blood ratios within a few hours. Several preclinical studies have also suggested that Ab fragments might be preferred for therapeutic applications over an intact IgG. The purpose of this project was to develop engineered Ab fragments using a humanized anti-carcinoembryonic antigen and anti-CD22 Ab as the parent. Three types of variants were prepared: a deltaCH2 (deletion mutant missing the CH2), a gamma3 F(ab')2 containing the human IgG3 hinge, and three glycosylated variants. The gamma3 F(ab')2 and glycosylated variants were developed because of the potential for site-specific linkage to the Ab in its divalent or monovalent fragment. The gamma3 F(ab')2 variant contains 10 cysteine residues that could be used for direct coupling using thiol chemistry, whereas the glycosylated variants have N-linked glycosylation sites engineered in the CH1 domain (two variants) as well as the VK domain (one variant). All of these variants were successfully prepared and shown to react with the target antigen. All Abs could be purified to a single peak by size-exclusion HPLC, but the deltaCH2 variant showed two distinct peaks, which were believed to be both the divalent and monovalent forms of this fragment. The two CH1 glycosylated variants showed differences in the extent of glycosylation. Modeling studies suggest that one variant would be better suited for site-specific coupling than the other because the carbohydrate chain is extended further away from the antigen-binding site. The Abs were radioiodinated to determine their pharmacokinetic behavior in mice. All of the humanized Ab divalent fragments cleared nearly 20 times faster from the blood than the murine parent F(ab')2 over a 24-h period. The glycosylated fragments showed some added stability compared to the other fragments over 4 h, but by 24 h, they had cleared to the same extent. Size-exclusion high-performance liquid chromatography of blood samples indicated that the humanized Ab fragments were quickly degraded in the blood. Thus, there is an inherent instability of the divalent fragments from these humanized IgG1 constructs that may affect their utility in imaging or therapy applications.

Mu-9 is a murine monoclonal antibody that is directed against affinity-purified colon-specific antigen-p (CSAp). CSAp is a tumor-associated antigen that is present in 60% of colorectal carcinomas. Preclinical and clinical studies have shown Mu-9 to have excellent targeting abilities. However, as administration of the murine immunoglobulin G (IgG) provoked a human anti-mouse antibody response, chimerization of Mu-9 is warranted for decreasing immunogenicity.Polymerase chain reaction and cDNA library screening methods were used for the cloning of Mu-9 heavy and light chain variable regions for the construction of chimeric Mu-9.The functional chimeric Mu-9 antibody binds to the CSAp antigen in the GW-39 extracts. It has immunochemical properties similar to that of murine Mu-9.The V-region sequence information will be used for design of humanized Mu-9, which will be evaluated for targeting gastrointestinal carcinomas.

LL2 is a murine immunoglobulin (Ig)G2a-kappa anti-B-cell monoclonal antibody with proven targeting and therapeutic efficacy in the management of non-Hodgkin's lymphoma (NHL). The authors had previously generated a humanized LL2 (hLL2) that demonstrated binding properties identical to those of LL2. Nevertheless, the productivity of the cell line was insufficient for large-scale production of the antibody for clinical studies. Therefore, the authors chose an amplifiable system for the generation of hLL2.The hLL2 sequences were ligated into the expression vector pdHL2, which has a dhfr amplifiable gene, and were incorporated into the SP2/0 cells by electroporation. A methotrexate (MTX) resistant clone producing hLL2 was identified. Stepwise increases in MTX concentrations, from 0.1 to 5 microM, and subcloning of the cells by limiting dilution were performed.By amplifying the dhfr and hLL2 genes with stepwise increases in the MTX concentration, the antibody production was enhanced from its original 1.4 to 70 +/- 5 mg per liter of culture media. Subsequent subcloning further improved the productivity. Immunoreactivity of the antibody was conserved, as proven by enzyme-linked immunosorbent assay and cell-binding assays. By isoelectrofocusing, the isoelectric point (pI) of the antibody was measured at approximately 9.6. The productivity of the clone was not affected by culture conditions or storage of the cells in liquid nitrogen.By means of gene amplification, the authors have generated a high-producing hLL2-IgG clone suitable for production of the quantity of antibody necessary for clinical diagnostic and therapeutic trials of NHL patients.

Anti-idiotype monoclonal antibodies (Mabs) to mLL2, an anti-B-cell lymphoma and CD22-specific murine IgG2a-kappa Mab, were generated by hybridoma technology from splenocytes of Copenhagen rats immunized with mLL2 F(ab')2. Mab WN, an IgG2a-kappa, was selected based on its specific binding to mLL2 and not other IgG isotypes or anti-B-cell Mabs. In a radioimmunoassay, WN was found to inhibit the binding of 125I-labeled mLL2 to Raji cells and to have no effect on the binding of other B-cell-reactive antibodies. Using high performance liquid chromatography analysis, WN was shown to complex specifically with both mLL2 and mLL2 Fab'. Meanwhile, we have constructed chimeric (cLL2) and humanized (hLL2) versions of LL2. Both cLL2 and hLL2 were demonstrated to retain the original antigen specificity and affinity of mLL2 [S.O. Leung et al., Proc. Am. Assoc. Cancer Res., 2872 (abstract), 34: 481, 1993]. The specific binding of WN to either radioiodinated or peroxidase-conjugated mLL2 was inhibited in a dose-response manner, and to a similar extent by mLL2, cLL2, and hLL2. Since the mLL2 complementarity-determining regions are the only sequences common to mLL2, cLL2, and hLL2, the result confirms that WN is specific to the antigen-binding complementarity-determining regions. A WN binding assay is currently being evaluated as a substitute for the tedious, and sometimes inconsistent, Raji cell-binding assay for the determination of LL2 immunoreactivity. In conclusion, we have developed an anti-idiotype Mab, WN, to mLL2. Its potential use as a surrogate antigen for B-cell lymphoma is under investigation.

Despite the potential advantages of 32P over other isotopes for radioimmunotherapy, its development as a therapeutic has been hindered by the difficulty of the labeling chemistry. Recently, a heptapeptide [Kemptide (KPT)] has been chemically conjugated to antibodies, and the conjugates have successfully been labeled with 32P enzymatically by using bovine protein kinase. By using genetic engineering, we have produced a chimera (Fab.KPT) consisting of the Fab' moiety of the complementarity-determining region-grafted anti-carcinoembryonic antigen-monoclonal antibody, MN14, and a heptapeptide derivative of KPT (Trp-Arg-Arg-Ala-Ser-Leu-Gly). The recombinant protein was expressed in Escherichia coli as a soluble secretory product. The presence of the KPT derivative downstream of the COOH terminus of the hinge region did not impair the binding affinity of the antibody fragment. The Fab.KPT was enzymatically phosphorylated with 32P by bovine protein kinase, without significant effect on the resultant immunoreactivity; 100% of the 32P-labeled Fab.KPT was complexed with liquid carcinoembryonic antigen. The 32P-labeled humanized MN-14 Fab.KPT is expected to have longer blood circulation half-life, allowing for an improved therapeutic efficacy in radioimmunotherapy.

Site-specific introduction of metal-chelating groups into F(ab')2 fragments of an antilymphoma antibody (LL2) possessing a natural Asn-linked light chain carbohydrate and an anti-carcinoembryonic antigen antibody (hImmu-14-N) grafted with a light chain carbohydrate site is described. For this purpose, four yttrium- (and indium)-chelating agents were used, containing a primary amino group for antibody binding and 1-(4-substituted benzyl)diethylenetriaminepentaacetic acid as the metal-chelator, separated by structurally different additional linkers. Conjugates were prepared by reacting excess chelator with oxidized carbohydrate of F(ab')2 fragments, with or without a subsequent reduction step. The conjugates, with up to an average of 5.5 chelating groups attached to a F(ab')2 fragment, were readily labeled with 90Y and 111In and were found to retain antigen-binding ability in in vitro assays. Tumor targeting was demonstrated using a 88Y-labeled hImmu-14-N F(ab')2 carbohydrate-modified conjugate. 2-Pyridyldithiopropionic hydrazide was conjugated to the carbohydrate region, and the disulfide was selectively deprotected to the thiol group, which is reactive with reduced 99mTc. These initial experiments establish that light chain carbohydrate modification of F(ab')2 is as facile as with the Fc-region carbohydrate of intact IgG, and thereby offer the possibility of designing site-specifically substituted F(ab')2 fragments with favorable pharmacokinetic properties.

A natural N-linked glycosylation site (Asn-Val-Thr) at amino acid positions 18-20 (Kabat's numbering) was identified in the framework-1 (FR-1) region of the light chain variable (V kappa) domain of a murine anti-B cell lymphoma Ab, LL-2. Our earlier studies demonstrated that no contact between the V kappa-appended oligosaccharide and the Ag binding site was evident, because glycosylation at this site did not affect the Ag binding property of the Ab. By using the murine LL-2 F(ab')2 fragment (which is devoid of constant region-appended oligosaccharide) as substrate, as much as five bifunctional chelator molecules per F(ab')2 fragment could be site specifically conjugated at the V kappa-appended carbohydrate moiety with no reduction in immunoreactivity. The resulting conjugates labeled efficiently with both 90Y and 111In, with no significant effect on Ab affinity. In contrast, conjugation of less than five chelates/Ab fragment randomly at lysine residues resulted in a three- to fivefold reduction in affinity. By a single Arg to Asn mutation, an N-linked glycosylation site similar to that of LL-2 was introduced in the FR-1 segment of a nonglycosylated, humanized anti-carcinoembryonic Ag (CEA) Ab, MN-14 (hMN-14). Glycosylation at the engineered carbohydrate-addition site was demonstrated by SDS-PAGE analysis. Neither glycosylation nor site-specific conjugation of chelate at the V kappa-appended carbohydrate moiety resulted in the loss of immunoreactivity. The glycosylated hMN-14 conjugate labeled efficiently with 90Y.

In H-2S mice, the administration of mercuric chloride results in the development of antinucleolar autoantibodies. These mice represent a valuable model to study the role of environmental factors in the development of systemic autoimmunity. We have obtained seven antinucleolar mAb from mercury-injected A.SW mice and characterized their specificities and V genes. All mAb immunoprecipitate the U3 and U8 ribonucleoprotein particles (RNP) and some (but not all) react with fibrillarin, the only currently characterized protein component of mammalian nucleolar RNP. Several VH and V kappa genes are recurrently used by these antinucleolar RNP mAb and their H chain CDR3 segments contain several acidic residues that may be important for binding to the cationic proteins composing the nucleolar RNP. Our results support the concept that in H-2S mice administration of mercury induces a specific loss of tolerance to nucleolar RNP.

MRL/Mp(-)+/+ mice produce antinuclear antibodies and develop a spontaneous autoimmune syndrome with lupus-like nephritis. We obtained a panel of seven histone-reactive IgG mAb from a single MRL/Mp(-)+/+ mouse. These antibodies do not react significantly with DNA or individual histones, but bind strongly to the histone H2A-H2B dimer and even more strongly to the H2A-H2B-DNA complex. These antibodies also bind to whole nuclei when tested by immunofluorescence, indicating that they recognize an epitope accessible in chromatin. The V region sequences of these antibodies have been determined. The H chain third complementarity-determining regions of these antibodies are similar to those found in anti-DNA antibodies even though the antibodies in our panel do not react with DNA in the absence of histones, suggesting that DNA is part of the subnucleosome epitope. Several of these antibodies are clonally related, supporting the hypothesis that the activation of these clones is Ag-driven. Analysis of the sequences of these antibodies indicates that they derive from autoreactive B cells that were clonally expanded and whose V region genes have undergone numerous somatic mutations.

We previously demonstrated that patients with carcinoembryonic antigen [CEA]-producing neoplastic tumors, treated with murine monoclonal antibody to CEA, produced antibodies directed against the constant regions [human anti-mouse antibody (HAMA)] and the idiotypes [anti-Id] of these murine immunoglobulins. In this study, we describe a method for analyzing the presence of such antibodies in the sera of these patients. The HAMAs were measured by enzyme immunoassay and removed by immunoadsorption on Affi-Gel mouse IgG. The unabsorbed fraction contained the anti-Id antibodies; their presence was demonstrated by binding to the CEA monoclonal antibody (Ab1). The specificity of the binding was assessed by preincubating the sera with Ab1 and measuring the residual nonspecific binding. When specific binding was detected, the anti-Id antibodies were isolated by adsorption and elution on Affi-Gel Ab1. The anti-Id antibodies were fixed on enzyme immunoassay plates and incubated with a panel of mouse anti-human immunoglobulin to determine their isotypes. In a first series of 24 patients, HAMAs were found in 20 cases and anti-Id antibodies in 19 cases. The isolation of a specific IgG to Ab1 was achieved in 2 cases. In an ongoing series, the HAMA and anti-Id antibodies were detected in all five patients given injections of another monoclonal antibody to CEA. In two patients an IgG1 kappa anti-Id was isolated from the serum. The potential therapeutic effect of these antibodies is under investigation.

The interaction of DNA with type I to VI collagens and laminin was studied in vitro in systems in which the connective tissue components were immobilized, as well as when in solution. In studies on immobilized components, significant binding of DNA was observed only for type V collagen, and the binding of radiolabeled DNA to this component could be effectively inhibited in a concentration-dependent manner by the addition of unlabeled DNA. Similar results were observed in solution assays in which it was observed that DNA binding to type V collagen was dependent on the native triple-helical conformation of the collagen. The preferential binding of DNA to native type V collagen may be due to the relative basicity of type V collagen chains, as well as the unique spatial arrangement of amino acid side chains in the native molecules. The data are of potential clinical relevance in that binding of DNA to type V collagen may represent at least one component of the mechanism whereby DNA and its immune complexes are deposited in connective tissues in certain pathologic conditions.

Concentrations and rates of synthesis of phosphoribosylpyrophosphate (PP-Rib-P) and purine nucleotides were compared in fibroblasts cultured from 5 males with PP-Rib-P synthetase superactivity, 3 normal individuals, and 2 children with severe hypoxanthine-guanine phosphoribosyltransferase deficiency. Although all cell strains with PP-Rib-P synthetase superactivity showed increased PP-Rib-P concentration and generation, increased rates of PP-Rib-P-dependent purine synthetic pathways, and increased purine and pyrimidine nucleoside triphosphate concentrations, two subgroups were discernible. Three fibroblast strains with isolated catalytic defects in PP-Rib-P synthetase showed milder increases in PP-Rib-P concentration (2.5-fold normal) and generation (1.6- to 2.1-fold) and in rates of purine synthesis de novo (1.6- to 2.2-fold) and purine nucleoside triphosphate pools (1.5-fold) than did cells from 2 individuals with combined kinetic defects in PP-Rib-P synthetase, both with purine nucleotide inhibitor-resistance. Values for these processes in the latter two strains were, respectively, 5- to 6-fold, 2.6- to 3.2-fold, 4- to 7-fold, and 1.7- to 2.2-fold those of normal cells. In contrast to cells with catalytic defects, these cells also excreted an abnormally high proportion of labeled purines and resisted purine base-mediated inhibition of PP-Rib-P and purine nucleotide synthesis. Hypoxanthine-guanine phosphoribosyltransferase-deficient cells showed normal regulation of PP-Rib-P synthesis and normal nucleoside triphosphate pools despite increased rates of purine synthesis de novo and of purine excretion. Cells with PP-Rib-P synthetase superactivity thus synthesize purine nucleotides at increased rates as a consequence of increased PP-Rib-P production, despite increased purine nucleotide concentrations. These and additional findings provide evidence that regulation of purine synthesis de novo is effected at both the PP-Rib-P synthetase and amidophosphoribosyltransferase reactions.

Phenotypic expression of 5-phosphoribosyl 1-pyrophosphate (PRPP) synthetase superactivity was examined in lymphoblast lines derived from six unrelated male patients. Fibroblasts from these individuals have increased rates of PRPP and purine nucleotide synthesis and express four classes of kinetic derangement underlying enzyme superactivity: increased maximal reaction velocity (catalytic defect); inhibitor resistance (regulatory defect); increased substrate affinity (substrate binding defect); and combined catalytic and regulatory defects. Lymphoblast lines from three patients with catalytic defects and from three normal individuals were indistinguishable with respect to enzyme activities, PRPP concentrations and generation, and rates of purine synthesis. Enzyme in lymphoblasts from a patient with combined defects also showed normal maximal reaction velocity but expressed purine nucleotide inhibitor resistance. A second regulatory defect and a substrate binding defect were also demonstrable in lymphoblasts and were identical to the enzyme defects in fibroblasts from the respective patients. Regulatory and substrate binding defects in lymphoblasts were accompanied by increased rates of PRPP and purine nucleotide synthesis. Among explanations for selective expression of enzyme superactivity, reduced concentrations of catalytically superactive enzymes seemed unlikely: immunoreactive PRPP synthetase was comparable in normal-derived and patient-derived cells. Activation of normal enzyme in transformed lymphocytes was also unlikely because absolute specific activities of lymphoblast PRPP synthetases corresponded to those of normal fibroblast and erythrocyte enzymes. Abnormal electrophoretic mobilities and thermal stabilities, identified in certain catalytically superactive fibroblast PRPP synthetases, were not found in the corresponding lymphoblast enzymes. Thus, lymphoblast PRPP synthetases from patients with catalytic superactivity appeared to differ structurally and functionally from their fibroblast counterparts.

Superactive phosphoribosylpyrophosphate (PRPP) synthetases were characterized in fibroblasts and erythrocytes from 5 unrelated men with gout and/or hyperuricemia and uric acid overproduction. The kinetic basis of enzyme superactivity in all patients was increased maximal reaction velocity. Affinities of the enzymes for substrates and activators and responsiveness to inhibitors were normal, and levels of immunoreactive enzyme in patient and control fibroblast and erythrocyte extracts were comparable. Enzymes purified to homogeneity from 2 patients confirmed the presence of isolated catalytic defects. Altered physical properties of certain of the superactive enzymes suggested the presence of several distinctive structural defects among the aberrant forms. Fibroblasts from each affected patient showed increased PRPP concentration and generation, as well as accelerated rates of all PRPP-requiring purine nucleotide synthetic pathways. These findings support the concept that enzyme superactivity results in uric acid overproduction as a consequence of increased rates of PRPP and purine nucleotide synthesis. Cultured cells from female relatives of 2 patients showed evidence for the heterozygous carrier state, as measured both by enzyme activities and by rates of PRPP and purine synthesis. The clinical phenotype in 4 patients was limited to early adult-onset gout and its consequences, whereas the fifth patient expressed a familial constellation of hyperuricemia, sensorineural deafness, ataxia, and renal insufficiency. The severity of the derangements in PRPP synthetase and in PRPP and purine synthesis in cells from the 5 patients, however, was comparable. The neurologic accompaniments of enzyme superactivity found in 1 family described here, and in 2 others described previously, thus may not necessarily be consequences of primary defects in PRPP synthetase.

Concentrations and rates of synthesis of phosphoribosylpyrophosphate (PP-Rib-P) and purine nucleotides were compared in fibroblasts cultured from 5 males with PP-Rib-P synthetase superactivity, 3 normal individuals, and 2 children with severe hypoxanthine-guanine phosphoribosyltransferase deficiency. Although all cell strains with PP-Rib-P synthetase superactivity showed increased PP-Rib-P concentration and generation, increased rates of PP-Rib-P-dependent purine synthetic pathways, and increased purine and pyrimidine nucleoside triphosphate concentrations, two subgroups were discernible. Three fibroblast strains with isolated catalytic defects in PP-Rib-P synthetase showed milder increases in PP-Rib-P concentration (2.5-fold normal) and generation (1.6- to 2.1-fold) and in rates of purine synthesis de novo (1.6- to 2.2-fold) and purine nucleoside triphosphate pools (1.5-fold) than did cells from 2 individuals with combined kinetic defects in PP-Rib-P synthetase, both with purine nucleotide inhibitor-resistance. Values for these processes in the latter two strains were, respectively, 5- to 6-fold, 2.6- to 3.2-fold, 4- to 7-fold, and 1.7- to 2.2-fold those of normal cells. In contrast to cells with catalytic defects, these cells also excreted an abnormally high proportion of labeled purines and resisted purine base-mediated inhibition of PP-Rib-P and purine nucleotide synthesis. Hypoxanthine-guanine phosphoribosyltransferase-deficient cells showed normal regulation of PP-Rib-P synthesis and normal nucleoside triphosphate pools despite increased rates of purine synthesis de novo and of purine excretion. Cells with PP-Rib-P synthetase superactivity thus synthesize purine nucleotides at increased rates as a consequence of increased PP-Rib-P production, despite increased purine nucleotide concentrations. These and additional findings provide evidence that regulation of purine synthesis de novo is effected at both the PP-Rib-P synthetase and amidophosphoribosyltransferase reactions.

We have studied families C and A in which superactivity of PRPP synthetase (E.C. 2.7.6.1) is associated with gout and uric acid overproduction in affected hemizygous males. PRPP synthetase catalyzes synthesis of PRPP, a regulatory substrate in purine synthesis de novo. Activities of the enzyme in erythrocyte and fibroblast extracts from the male index cases, T.C. and R.A., were nearly threefold greater than normal at each Pi concentration tested. PRPP synthetase superactivity was accompanied by increased intracellular PRPP concentration and generation in erythrocytes and fibroblasts from these patients, and enhanced rates of PRPP-dependent purine synthesis reactions, including purine synthesis de novo, were demonstrable in their fibroblasts. These findings suggested that increased intracellular synthesis dut to enzyme superactivity underlay purine nucleotide and uric acid overproduction in these patients. Similar studies in cells from the sister of T.C. and the mother of R.A. showed increased values that were, however, intermediate between normal values and those of the affected males, indicating that these women are heterozygous carriers of the traits for enzyme superactivity. The enzymatic basis for increased PRPP synthetase activity in both families was investigated. Immunochemical studies in dialyzed erythrocyte lysates and highly purified erythrocyte enzyme preparations provided evidence for increased enzyme activity per molecule of immunoreactive enzyme. In addition, purified T.C. and R.A. PRPP synthetases showed 3.1- and 2.8-fold greater enzyme specific activities, respectively, than comparably purified normal enzymes. Kinetic constants of purified T.C. and R.A. PRPP synthetases for substrates, activators, and inhibitors were indistinguishable from normal, and increased maximal reaction velocity alone appeared to account for enzyme superactivity. Despite an apparently similar kinetic mechanism for superactivity, the diminished electrophoretic mobility of T.C. PRPP synthetase and increased thermal lability of R.A. PRPP synthetase suggested distinct structural alterations leading to enzyme superactivity in families C and A.

The high energy sugar phosphate 5-phosphoribosyl 1-pyrophosphate (PRPP) is synthesized from MgATP and ribose-5-phosphate in a reaction catalyzed by PRPP synthetase (E.C. 2.7.6.1). In vitro studies of this enzyme,1–6 as well as evaluations of PRPP production in intact cells, 7,8 indicate that PRPP synthetase activity is regulated in a complex fashion involving the interaction of substrates, activators, reaction products and diverse inhibitors. Enzyme activity is strictly dependent upon the presence of inorganic phosphate (Pi) and Mg2+ which serve both as activators and cofactors.3 Among inhibitors of PRPP synthetase activity are purine, pyrimidine and pyridine nucleotide end-products of the pathways of PRPP utilization, the reaction products (PRPP and AMP), and 2,3-DPG.3,6

Phosphoribosyl pyrophosphate ( PP Rib P ) synthetase activity was studied in cultured fibroblasts and lymphoblasts from a male child (patient 2-A) in whom inherited purine nucleotide and uric acid overproduction are accompanied by neurological deficits. Chromatographed or partially purified preparations of the child's enzyme showed 5–6-fold increased inhibitory constants ( I 0.5 ) for the noncompetitive inhibitors GDP and 6-methylthioinosine monophosphate but normal responsiveness to the competitive inhibitors ADP and 2,3-diphosphoglycerate. Activation of the PP Rib P synthetase of patient 2-A by P i was also abnormal with 3–4-fold reduced apparent K D values for P i . Superactivity of the PP Rib P synthetase of this child thus appeared to result from a combination of regulatory defects; selective resistance to noncompetitive inhibitors and increased responsiveness to P i activation. Selective growth of the patient's fibroblasts in medium containing 6-methylthioinosine confirmed the functional significance of the in vitro inhibitor resistance of the aberrant enzyme. Fibroblasts and lymphoblasts derived from patient 2-A showed increased concentrations and rates of generation of PP Rib P as well as increased rates of the pathways of purine base salvage and purine nucleotide synthesis de novo. The magnitudes of these increases in the child's cells exceeded those in cells with catalytically superactive PP Rib P synthetases. These alterations as well as the in vitro kinetic abnormalities in the patient 2-A enzyme were expressed to a reduced degree in fibroblasts from the child's affected mother, supporting the proposal that this woman is a heterozygous carrier for X-linked enzyme superactivity.

Superactivity of phosphoribosylpyrophosphate (PRPP) synthetase is one of several hereditary enzyme abnormalities associated with gout and excessive uric acid excretion. Although measurement of PRPP synthetase activities in erythrocyte lysates should provide a practical means to detect abnormalities of the enzyme, reported values for normal individuals have varied considerably. We describe a radioisotopic procedure for measurement of PRPP synthetase activities in dialyzed hemolysates under a variety of conditions permitting evaluation of enzyme catalytic function and responsiveness to inhibitors and activators. Utilizing this procedure, enzyme activities for normal individuals were higher than generally reported, a difference attributable in part to the following measures undertaken to assure accuracy in activity determinations: precise control of pH of reaction mixtures, provision of verified excesses of the auxiliary enzyme adenine phosphoribosyltransferase, and measurement of all of the radiolabeled products of the assay. Under each condition of measurement, enzyme activities in 44 normal individuals, 13 patients with gout and normal uric acid excretion, and 10 patients with gout and uric acid overproduction were indistinguishable. In four additional individuals with uric acid overproduction, however, excessive enzyme activities were identifiable at all inorganic phosphate concentrations, but responses to purine nucleotide inhibitors were normal. In hemolysates from a patient with an inhibitor-resistant PRPP synthetase, an altered pattern of inorganic phosphate activation and diminished nucleotide inhibitor response was demonstrated. Our studies confirm the ability of the assay procedure to detect kinetically distinct variant forms of PRPP synthetase. Application of this procedure should aid in evaluation of the prevalence of derangements of PRPP synthetase among patients with gout and uric acid overproduction.

Antibodies reactive with left-handed Z-DNA arise spontaneously in the sera of patients with SLE and rheumatoid arthritis and in autoimmune MRL mice. However, the precise specificity of these autoantibodies has not been established. In this report, we have characterized four monoclonal anti-Z-DNA antibodies from unimmunized MRL/Mp-lpr/lpr mice that do not cross-react with B-DNA and can discriminate between different types of left-handed helices. Two of the monoclonal antibodies (Za and Zi) behaved similarly in that they bound to two forms of Z-DNA (Br-poly(dG-dC).poly(dG-dC) and AAF-poly(dG-dC).poly(dG-dC) but not to two other Z-form DNA (poly(dG-5BrdC).poly(dG-5BrdC) or poly(dG-5MedC).poly(dG-5MedC)). Neither antibody (Za or Zi) bound significantly to B-DNA or to denatured DNA. A third antibody (Ze) exhibited similar binding characteristics for the Z-DNA preparations, but also recognized denatured DNA. In contrast, a fourth antibody (3-7.3) bound preferentially to poly(dG-5BrC).poly(dG-5BrdC) in Z conformation. These results provide the first evidence for anti-Z-DNA autoantibodies in autoimmune mice that do not cross-react with native or denatured DNA and indicate that these antibodies exhibit considerable heterogeneity in their fine binding specificity.

The interaction of DNA with type I to VI collagens and laminin was studied in vitro in systems in which the connective tissue components were immobilized, as well as when in solution. In studies on immobilized components, significant binding of DNA was observed only for type V collagen, and the binding of radiolabeled DNA to this component could be effectively inhibited in a concentration-dependent manner by the addition of unlabeled DNA. Similar results were observed in solution assays in which it was observed that DNA binding to type V collagen was dependent on the native triple-helical conformation of the collagen. The preferential binding of DNA to native type V collagen may be due to the relative basicity of type V collagen chains, as well as the unique spatial arrangement of amino acid side chains in the native molecules. The data are of potential clinical relevance in that binding of DNA to type V collagen may represent at least one component of the mechanism whereby DNA and its immune complexes are deposited in connective tissues in certain pathologic conditions.

Superactivity of PRPP synthetase (PS) is an X chromosome-linked disorder associated with gout and uric acid overproduction. Fibroblasts from affected males show PS superactivity and increased rates of PRPP and purine nucleotide synthesis. Kinetic aberrations underlying PS superactivity include: catalytic defects; regulatory defects; substrate (ribose-5-P) binding defects; and combined catalytic and regulatory defects. We studied PS in B lymphoblasts derived from blood lymphocytes (PBL) of 6 males whose fibroblasts express the 4 classes of PS superactivity. Neither excessive PS activity nor increased rates of PRPP or purine synthesis were found in lymphoblasts from 3 patients with catalytic defects in fibroblasts, despite PS superactivity in PBL from these patients. Catalytic superactivity was also absent in lymphoblasts from a patient with combined defects in fibroblast PS. These cells and cells from 2 other patients expressed the specific regulatory and substrate binding defects found in fibroblast PS from the respective patients and also showed increased rates of PRPP and purine synthesis. Immunoreactive PS differed