Pharmacological intervention in antibody mediated disease.

The use of single signal anergy to inactive pathological B cells in an antigen-specific manner is discussed. Cross-linking surface immunoglobulin, with a construct which contains oligovalent B cell epitopes on a non-immunogenic molecular framework can be used to inactivate the target B cells if the construct lacks T cells epitopes. An example of such a B cell toleragen is LJP 394, which inactivates anti-dsDNA-specific B cells in vivo in murine immunized and spontaneous disease models. The drug enhances survival and lowers renal pathology in BXSB mice. Appropriate definition of epitopes of pathological (auto) antibodies thus offers an opportunity for pharmacological intervention.

Immunospecific reduction of antioligonucleotide antibody-forming cells with a tetrakis-oligonucleotide conjugate (LJP 394), a therapeutic candidate for the treatment of lupus nephritis.

A discrete tetravalent conjugate, 7a (LJP 394), consisting of four oligonucleotides attached to a common carrier or platform was prepared. Single-stranded oligonucleotide 20-mers consisting of alternating deoxycytidine-deoxyadenosine nucleotides, (CA)10, were attached to a tetrabromoacetylated platform by displacement with sulfhydryl-terminated linkers. The tetrabromoacetylated platform 3a was synthesized in three steps using triethylene glycol bis-(chloroformate). The single-stranded conjugate was characterized by polyacrylamide gel electrophoresis, DNA sequencing, phosphate analysis, carbon and nitrogen combustion analysis, and correlation of stoichiometry to conversion in the conjugation process. HPLC and capillary electrophoretic methods were developed to evaluate purity. The tetrakis, single-stranded conjugate was annealed with a stoichiometric amount of a complementary single-stranded oligonucleotide 20-mer consisting of alternating thymidine-deoxyguanosine nucleotides, (TG)10. The double-stranded conjugate LJP 394 was characterized by melt temperature and hyperchromicity, phosphate analysis, and carbon and nitrogen combustion analysis. LJP 394 inhibits binding of DNA to anti-double-stranded oligonucleotide antibodies and reduces anti-oligonucleotide-specific plaque (antibody)-forming cells in an immunized mouse model by a proposed mechanism involving cross-linking B cell surface immunoglobins.

Conjugates of double-stranded oligonucleotides with poly(ethylene glycol) and keyhole limpet hemocyanin: a model for treating systemic lupus erythematosus.

Two types of oligonucleotides were synthesized with linker groups attached at the 5'-end. Both were repeating dimers of deoxyribocytidine and deoxyriboadenosine. A 20-mer was prepared with a thiol-containing linker, masked as a disulfide, and a 50-mer was prepared with a vicinal diol-containing linker. A tetraiodoacetylated poly(ethylene glycol) (PEG) derivative was synthesized and reacted with the thiol-containing 20-mer to provide an oligonucleotide PEG conjugate of precisely four oligonucleotides on each PEG carrier. The vicinal diol on the 50-mer was oxidized to an aldehyde and conjugated to keyhole limpet hemocyanin (KLH) to provide an oligonucleotide-KLH conjugate by reductive alkylation. The conjugates were annealed with complementary (TG)n strands. While the double-stranded oligonucleotide-KLH conjugate is an immunogen, eliciting the synthesis of antibodies against oligonucleotides, the PEG conjugate has the biological property of specifically suppressing (tolerizing) B cells which make antibodies against the immunizing oligonucleotide.

Monokine regulation of human immunodeficiency virus-1 expression in a chronically infected human T cell clone.

A T cell clone (ACH-2) derived from T cells infected with HIV-1 was found to produce HIV-1 in response to stimulation with a monokine-enriched supernatant prepared by culturing human monocyte/macrophages with bacterial LPS (LPS-MO SN). Monokine induction of ACH-2 cells resulted in augmented virus production reflected by an increase in reverse transcriptase activity and in the synthesis of all major viral proteins. Examination of the cells by indirect immunofluorescence revealed that 10 to 15% of uninduced cells constitutively expressed HIV proteins, whereas 100% showed positive immunofluorescence in response to LPS-MO SN. This induction of virus by LPS-MO SN resulted in approximately a 100-fold increase of infectious virus production over uninduced ACH-2 cells. LPS alone could not induce HIV-1 expression, whereas LPS-MO SN resulted in the greatest virus expression. Cell separation studies confirmed the source of the inducing factor(s) to be cells bearing the mature monocyte/macrophage marker, Leu M3. Biochemical fractionation of the LPS-MO SN suggested that one or more factors, having apparent Mr of approximately 45 kDa, were involved in this induction. Absorption of the LPS-MO SN with immunoaffinity gels specific for human TNF-alpha was shown to completely remove the HIV inducing activity for the ACH-2 cell line.

Molecular cloning of pertussis toxin genes.

We have cloned a 4.5 kb EcoRI/BamHI DNA fragment from Bordetella pertussis which contains at least two genes responsible for expression of pertussis toxin. The S4 subunit of the toxin was isolated by high pressure liquid chromatography and the NH2-terminal amino acid sequence determined. Using a mixed synthetic oligonucleotide probe designed by reverse translation of a portion of the protein sequence, a cloned DNA fragment was identified which contains the coding information for at least the S4 structural subunit of the toxin. Sequence analyses indicate that the mature protein is derived by proteolytic cleavage of a precursor molecule. Southern blot analyses of Tn5-induced B. pertussis toxin-deficient mutants show that the Tn5 DNA is inserted 1.3 kb downstream from the S4 subunit gene. This second gene could code for another subunit required for assembly of the mature toxin or a non-structural transport protein, possibly in the same polycistronic operon. The molecular cloning of pertussis toxin genes provides the basis for development of a safer recombinant "new generation" vaccine for whooping cough.

Variable major proteins of Borrelia hermsii. Epitope mapping and partial sequence analysis of CNBr peptides.

The variable major proteins (VMP) of serotypes 7 and 21 of the relapsing fever agent Borrelia hermsii were isolated by detergent extraction and high performance liquid chromatography. Cyanogen bromide (CNBr) digestion of the isolated VMP yielded two peptides of apparent molecular weights 20,000 (20 K) and 16 K from VMP7, and three peptides of 14.5, 14, and 7 K mol wt from VMP21. Serotype-specific monoclonal antibodies bound in Western blots to one of each of the two or three CNBr fragments from the homologous VMP. A single monoclonal antibody bound to the whole cells, the isolated VMP, and a CNBr fragment of both serotype 7 and serotype 21. (This crossreactive antibody did not, however, bind to any of four other serotypes examined.) Regional conservation of structure between VMP7 and VMP21 was also shown by amino acid sequence analysis of the N-termini of the five CNBr fragments. One pair of aligned fragments from VMP7 and VMP21 had 80% amino acid homology in sequence; a second pair had 40% homology. The partial amino acid homologies between two VMP suggest that these proteins are products of members of a polygene family.

Partial amino acid sequence and molecular cloning of the encoding gene for the major outer membrane protein of Chlamydia trachomatis.

The first 25 N-terminal amino acids of the major outer membrane protein of Chlamydia trachomatis serovar L2 were determined. The amino acid sequence was used to construct an oligonucleotide probe specific for the major outer membrane protein gene. Using this oligonucleotide as a hybridization major outer membrane protein gene. Using this oligonucleotide as a hybridization probe, we discovered one recombinant clone that produced a 15-kilodalton polypeptide which reacted with a monoclonal antibody directed against the major outer membrane protein type-specific epitope. In a separate set of experiments, we uncovered another recombinant clone that produced a 51-kilodalton polypeptide which was reactive with an anti-major outer membrane protein subspecies-specific monoclonal antibody. The expression of these recombinant DNA plasmids in Escherichia coli is discussed.

Isolation and characterizaiton of murine cell surface components. I. Purification of milligram quantities of Thy-1.1.

The Thy-l.1 molecule was isolated from the BW5147 murine lymphoblastoid cell line. The initial step in purification was the preparation of a crude plasma membrane fraction followed by acetone precipitation. The acetone pellet was solubilized using deoxycholate (DOC) and Thy-1.1 was purified by use of a Lens culinaris lectin affinity column and an AcA-34 gel filtration column. The purified glycoprotein with Thy-1.1 activity had a mol wt of approximately 25,000 daltons. The isolation of this molecule was effected by detecting Thy-I activity utilizing rabbit anti- mouse brain serum tested on rat thymocytes. Congenic anti-Thy-l.1 serum was ineffective in detecting Thy-l.1 after DOC solubilization. An antiserum prepared in rabbits to the purified Thy-1.1 was found to be cytotoxic to mouse and rat thymocytes. The cytotoxic activity of this antisera could be completely absorbed with AKR/Jax brain and thymus but was not absorbed by liver. In addition, AKR/Jax thymocytes totally absorbed all cytotoxic activity of the rabbit anti-purified Thy-1 serum for BW5147 cells suggesting that the cell line shares identical specificities with normal thymocytes. The purified Thy-1.1 molecule was able to totally absorb the cytotoxic activity of mouse congenic anti-Thy-1. These studies serve as a model for the isolation of other murine lymphoid cell surface components in quantities for detailed structural and functional analysis.

Mouse immunoglobulin heavy chains are coded by multiple germ line variable region genes.

The N-terminal 20 residues of 13 heavy immunoglobulin chains from myeloma protein of the BALB/c mouse are compared with the same residues of 15 other heavy chains described in the literature. Sixteen of 28 sequences are different from one another. These proteins fall into four major sets, with 18 of the proteins in the largest set being further divisible into at least five subsets. This pattern of diversity suggests there are at least eight germ line genes coding for the variable regions of mouse heavy chain. Many of the immunoglobulins from which these heavy chains are derived exhibit binding activity for various haptens. The differing hapten specificities are closely correlated with distinct primary amino-acid sequences.

Immunoglobulin structure: amino terminal sequences of mouse myeloma proteins that bind phosphorylcholine.

The amino terminal sequences of five light and heavy immunoglobulin chains from myeloma proteins of the BALB/c mouse with binding activity to phosphorylcholine are presented. Except for a single substitution in position 4, all five heavy chains have identical amino terminal sequences through the first hypervariable region. Proteins which share unique (idiotypic) antigenic determinants are identical through the first hypervariable region of their light and heavy chains. Proteins with differing idiotypic determinants have light chains of differing amino acid sequence. These observations suggest that the heavy chain plays a more important role than the light chain in determining the phosphorylcholine binding site.