Prevention of adoptively transferred diabetes in nonobese diabetic mice with IL-10-transduced islet-specific Th1 lymphocytes. A gene therapy model for autoimmune diabetes.

Four pancreatic islet-specific CD4+ helper T (Th) 1 (Th1) clones and two Th1 clones transduced with an SRalpha promoter-linked murine IL-10 (mIL-10) cDNA of 2.0-6.0 x 10(6) cells were adoptively transferred to nonobese diabetic (NOD) mice at age 8 d. Cyclophosphamide (CY) was administered at age 37 d (plus CY), and the incidence of diabetes and the histological grade of insulitis were examined at age 47 d. After the adoptive transfer of IL-10-transduced Th1 cells, polymerase chain reaction (PCR) and reverse-transcription (RT)-PCR detected the neo gene and the retrovirus vector-mediated IL-10 mRNA in situ in recipient islets, respectively. RT-PCR detected the decrease of IFN-gamma mRNA relative to IL-10 mRNA in IL-10-transduced Th1 clones in vitro and also in recipient islets. All four wild type Th1 clones plus CY induced the insulitis grade of 2.75 and diabetes in 66% of recipient NOD mice. IL-10-transduced two Th1 clones plus CY induced periinsulitis with the grade of 1.43 and diabetes in 8.0%. The 1:1 mixture of wild type Th1 cells and IL-10-transduced Th1 cells plus CY induced periinsulitis with the grade of 1.85 and diabetes in 20%. The suppression of diabetes through decreasing IFN-gamma mRNA by the tissue-specific delivery of IL-10 to pancreatic islets with IL-10-transduced Th1 cells affords us the starting basis to develop the gene therapy for autoimmune diabetes.

Role of p53 mutations in endocrine tumorigenesis: mutation detection by polymerase chain reaction-single strand conformation polymorphism.

To elucidate the molecular basis for endocrine tumorigenesis, p53 mutations in human endocrine tumors were analyzed by using polymerase chain reaction-single strand conformation polymorphism. Exons 5 through 10 of the p53 gene were studied in genomic DNAs from 134 primary endocrine tumors and 6 human endocrine cancer-derived cell lines. Mutations were detected and identified in 4 endocrine tumors, including one parathyroid adenoma and three thyroid carcinoma cell lines. The sites of these mutations were in exons 5 (codon 151 and 152) and 7 (codon 248 and 255). In all of three tumor cell lines, but not in a parathyroid adenoma, the normal allele encoding the p53 gene was lost. However, p53 mutations were not found in any other endocrine tumors or cell lines. Based upon these results, we concluded that the p53 gene may play a role in the tumorigenesis of a limited number of parathyroid adenoma and thyroid cancers, and that the p53 mutation with an allelic loss of the p53 gene is an important factor in malignant tumorigenesis of the thyroid gland.

Loss of the same alleles of HRAS1 and D11S151 in two independent pancreatic cancers from a patient with multiple endocrine neoplasia type 1.

The DNAs from two independent pancreatic cancers (tumors 1 and 2) in a patient with multiple endocrine neoplasia type 1 were analyzed. No amplification or gross rearrangement of 19 protooncogenes was observed. However, Southern blot analysis using polymorphic DNA probes revealed loss of heterozygosity at loci on chromosome 11p in both tumors. In tumor 1, an extensive region including the HRAS1, PTH, CALCA, and D11S151 loci was deleted, while in tumor 2 loss of heterozygosity was limited at the HRAS1 and D11S151 loci. Because loss of heterozygosity at other chromosomal loci in the two tumors was quite rare, loss of genes on 11p might be nonrandom. It is noteworthy that the same allele at the HRAS1 locus and also the same allele at the D11S151 locus were lost in the two independent tumors. These results suggest that loss of genes at the HRAS1 and/or D11S151 loci plays an important role unmasking the remaining sequences probably having a recessive mutation.

A new RFLP in intron 1 of the human c-Ha-ras1 gene and its close relationship with the variable tandem repeats in the region 3' to the gene.

Based on restriction fragment length polymorphism (RFLP) analysis of BamHI or MspI fragments, the human c-Ha-ras1 alleles could be divided into two major groups, one having about 80 copies of a 28 base pair (bp) sequence in the variable tandem repeat (VTR) region 1.4-kilobase pairs (kb) downstream from the end of the coding exon and the other having 40 copies of the sequence. We found a second RFLP in intron 1 of the c-Ha-ras1 gene at a position about 80 bp upstream from the 5'-end of exon 1. The size of the PstI fragments carrying this region is either 371 or 359 bp depending on the numbers of a hexanucleotide sequence, GGGCCT. In larger fragments, the unit sequence was repeated four times, while in smaller fragments it was repeated twice. Unexpectedly, we found that the alleles with 80 copies of the 28 bp sequence in the VTR region all carried two repeats of GGGCCT in intron 1, while alleles with 40 copies all had four repeats of the GGGCCT sequence.

Molecular cloning and characterization of the ABC transporter expressed in Trachea (ATET) gene from Drosophila melanogaster.

A novel member of the ATP-binding cassette (ABC) transporter proteins has been cloned from Drosophila melanogaster. The gene is designated as ABC Transporter Expressed in Trachea (Atet), because the transcript was localized to the respiratory system by in situ hybridization analysis of whole-mount embryos using digoxigenin-labeled RNA probes. The hybridization signal was also observed in amnioserosa. Northern blot analysis identified a single 4.5 kb mRNA expressed in all developmental stages at a relatively constant level. The Atet gene mapped to 24E on the left arm of the second chromosome. The Atet protein shows extensive homology with the Drosophila white gene product, which is reported to form heterodimers with the brown or scarlet gene product to transport guanine or tryptophan into the pigment cells in the compound eye. By analogy, Atet is suggested to be involved in transporting a small molecule after dimerization with a partner protein.

Differential regulation of the human insulin gene transcription by GG1 and GG2 elements with GG- and C1-binding factors.

Using a human growth hormone reporter system, the introduced mutations in GG1 alone or both GG elements of GG1 and GG2 in the human insulin promoter abolished 94 or 96% of the beta-cell-specific transcriptional activity in a pancreatic islet beta-cell line of MIN6, while the mutations in GG2 or its total deletion abolished 85 or 86% of the transcriptional activity. When linked to the thymidine kinase promoter, mutations in GG1 or both GG elements abolished 74% of the transcriptional activity in MIN6 cells, while the mutations in GG2 or its total deletion abolished 55 or 54%. In the electrophoretic mobility shift assay (EMSA), one nuclear factor was shown to interact with two GG elements, and another C1-binding factor with GG1 and C1. The differential effects of deletions or selective mutations in the GG2 or GG1 sequence in the oligonucleotide probes on the binding activity of GG- or C1-binding factors in EMSA proved the requirement of both GG1 and GG2 or both GG1 and C1, respectively, for the transaction of these two factors. The molecular size of the GG-binding factor was estimated about 30 kDa. Based on these, we conclude that two GG elements contribute, with GG1 more critically than GG2, to the beta-cell-specific transcription of the human insulin gene through transaction with the GG- and C1-binding factors.

cDNA cloning and expression of Bacillus thuringiensis Cry1Aa toxin binding 120 kDa aminopeptidase N from Bombyx mori.

Bacillus thuringiensis Cry1Aa toxin binds to a 120 kDa putative receptor protein in the Bombyx mori midgut. Recently, this protein was purified and identified as glycosyl-phosphatidylinositol (GPI) anchored aminopeptidase N (APN). In this study, a full-length cDNA thought to encode this 120 kDa APN was isolated and sequenced. It has a 2958 bp ORF encoding 986 amino acids. In the deduced amino acid sequence, we identified GPI-anchor and zinc-metallopeptidase signals, which are the same as those of APNs of other insects that are reported to be putative Cry1 toxin receptors. The B. mori APN amino acid sequence also has a high similarity with those of the other APNs. Subsequently, the recombinant APN was expressed by Escherichia coli and its Cry1Aa toxin binding ability was analyzed. Ligand blotting showed that Cry1Aa toxin bound to the recombinant APN.

Molecular cloning of a human cDNA for the 41-kDa phosphoribosylpyrophosphate synthetase-associated protein.

A human cDNA encoding 41-kDa phosphoribosylpyrophosphate (PRPP) synthetase (PRS)-associated protein (PAP41) was cloned from two expressed sequence tag (EST) clones having the nucleotide similarity of 61.5 and 70.0% to human PAP39 cDNA. The predicted open reading frame of 1107 base pairs (bp) has the nucleotide identity of 91.8% to rat PAP41 and encodes a protein of 369 amino acids with a calculated molecular weight (MW) of 40,925. The deduced amino acid sequence exhibits the 98.9% identity to rat PAP41 and 72.2, 50.6, and 50.0% identity with human PAP39, PRS I, and PRS II, respectively, but lacks the PRPP binding site. Southern blot analysis suggested that the PAP41 gene exists as a single copy in the human genome. The single PAP41 mRNA of about 2.1 kb was shown to be present in five human cell lines by Northern blot analysis.

Rat genomic structure of amidophosphoribosyltransferase, cDNA sequence of aminoimidazole ribonucleotide carboxylase, and cell cycle-dependent expression of these two physically linked genes.

Genomic structure of rat amidophosphoribosyltransferase (ATase; EC 2.4.2.14), which catalyzes the first committed step in de novo purine nucleotide synthesis, was determined by polymerase chain reaction (PCR)-based methods. There are 11 exons and all exon-intron boundaries were conserved among rat, human, and chicken ATase genes. A rat aminoimidazole ribonucleotide carboxylase (AIRC) cDNA encoding a bifunctional enzyme of AIRC (EC 4.1.1.21) at step 6 and SAICAR synthetase (EC 6.3.2.6) at step 7 in de novo purine nucleotide synthesis was cloned and sequenced. The size of the cloned rat AIRC cDNA was 1329 bp, and amino acid identity with human and chicken AIRC was 96 and 85%, respectively. The intergenic sequence using a phage clone and the PCR product disclosed that ATase and AIRC genes are physically linked with the 736 bp sequence between the translation start sites, and the determination of the transcriptional start sites by the primer extension assay for these genes disclosed that distance between the two major transcriptional start sites is 585 bp. The amount of mRNAs of both genes showed approx. 5-6-fold increase in G1/S phase of the cell cycle over those in G0 phase in synchronized rat 3Y1 fibroblasts.

p53 gene mutations in human skin cancers and precancerous lesions: comparison with immunohistochemical analysis.

Mutations of exons 3 through 9 of the p53 gene in skin lesions were screened in 23 cases of squamous cell carcinoma (SCC), 25 cases of basal cell carcinoma (BCC), two cases of Bowen's disease, 10 cases of solar keratosis, and five cases of keratoacanthoma by polymerase chain reaction--single strand conformation polymorphism analysis. Mutations of the p53 gene were detected in seven of 23 SCCs (30%), three of 25 BCCs (12%), and none in all cases of Bowen's disease, solar keratosis, or keratoacanthoma. Of 23 cases of SCC, mutations were detected in four of 15 SCCs (27%) that originated in the sunlight-exposed skin region, in two of three SCCs (67%) that originated in the scar tissue, and in one of three SCCs (33%) that originated in radiation dermatitis. Mutations of C-->T transition predominated in SCC and BCC that originated in the sunlight-exposed skin region. Mutations of C-->A or CC-->AT observed in tumors that originated in the predisposed conditions, presumably unrelated to UV light, are different from those found in UV light-related SCC or BCC. Twelve cases of SCC were comparatively analyzed with the immunohistochemical staining with anti-p53 antibody. Two of four cases with positive staining had missense mutations, and three of eight cases with negative staining had nonsense mutations. Based on these findings, immunohistochemical results do not necessarily mean the presence or absence of p53 gene mutations in skin tumors, and sequence analysis is essential for determining whether the gene is mutated.

Frequent p53 accumulation in the chronically sun-exposed epidermis and clonal expansion of p53 mutant cells in the epidermis adjacent to basal cell carcinoma.

p53 expression was studied immunohistochemically to identify a precursor lesion of basal cell carcinoma (BCC) in the epidermis adjacent to BCC. With two different anti-p53 antibodies of CM1 and DO7, p53 expression was frequently detected in the epidermis adjacent to BCCs arising on the face and in the normal epidermis with usual sun exposure. In the epidermis adjacent to BCC, stained cells were occasionally clustered in a small area, but no cluster was found in the normal epidermis with usual sun exposure. The expression was less frequent in the normal epidermis with rare sun exposure. Ten cases of normal skin with usual sun exposure, showing CM1 staining in the epidermis, were screened for p53 gene mutations with polymerase chain reaction-single-strand conformation polymorphism analysis using DNAs obtained from the epidermis. No mutation was detected in exons 2 to 10 of the p53 gene in these 10 cases. The epidermis flanking three BCCs that was stained with CM1, on the other hand, carried a missense mutation of C to G transversion at a dipyrimidine site of codon 249. This alteration replaced arginine with threonine. The mutation of codon 249 was not detected in the three BCCs. Our results first suggest that ultraviolet light irradiating the skin in a daily life induces p53 accumulation in the epidermis and secondly that the frequent clonal expansion of p53 mutant cells occurs in the epidermis adjacent to BCCs. This clonal expansion of mutant p53 may provide a molecular basis for high risk of developing subsequent new skin cancers in patients with BCC.

Bacillus thuringiensis Cry1Aa toxin-binding region of Bombyx mori aminopeptidase N.

The Bacillus thuringiensis Cry1Aa toxin-binding region of Bombyx mori aminopeptidase N (APN) was analyzed, to better understand the molecular mechanism of susceptibility to the toxin and the development of resistance in insects. APN was digested with lysylendopeptidase and the ability of the resulting fragments to bind to Cry1Aa and 1Ac toxins was examined. The binding abilities of the two toxins to these fragments were different. The Cry1Aa toxin bound to the fragment containing 40-Asp to 313-Lys, suggesting that the Cry1Aa toxin-binding site is located in the region between 40-Asp and 313-Lys, while Cry1Ac toxin bound exclusively to mature APN. Next, recombinant APN of various lengths was expressed in Escherichia coli cells and its ability to bind to Cry1Aa toxin was examined. The results localized the Cry1Aa toxin binding to the region between 135-Ile and 198-Pro.

The lipopolysaccharide-binding protein participating in hemocyte nodule formation in the silkworm Bombyx mori is a novel member of the C-type lectin superfamily with two different tandem carbohydrate-recognition domains.

We recently isolated and characterized the lipopolysaccharide (LPS)-binding protein, BmLBP, from the larval hemolymph of the silkworm Bombyx mori. BmLBP is a pattern recognition molecule that recognizes the lipid A portion of LPS and participates in a cellular defense reaction. This paper describes the cDNA cloning of BmLBP. The deduced amino acid sequence of BmLBP revealed that BmLBP is a novel member of the C-type lectin superfamily with a unique structural feature that consists of two different carbohydrate-recognition domains in tandem, a short and a long form.

Radioimmunoassay of an analog of luteinizing hormone-releasing hormone, [D-Ser(tBu)]6des-Gly-NH2(10) ethylamide (Buserelin).

A sensitive and specific radioimmunoassay is described for plasma and urinary levels of [D-Ser(tBu)]6des-Gly-NH2(10) ethylamide (buserelin). No appreciable cross-reaction (less than 0.05%) was observed with LH-RH and its analogs other than buserelin fragments (1.6-45%). The sensitivity was 3 pg per tube. At buserelin concentrations of 125, 250 and 500 pg/ml, the intra- and inter-assay coefficients of variation were 7.9, 10.0 and 10.0%, and 19.0, 7.8 and 6.8% respectively. Recovery of buserelin added to plasma was quantitative (62.5 pg/ml, 101.6%; 125 pg/ml, 76.8% and 250 pg/ml, 63.4%). A dose of 5 micrograms buserelin injected subcutaneously into 5 normal male adults, reached a peak plasma level in 45 min (mean value 119.3 +/- 47.3 pg/ml) and remained detectable for at least 4 h. The half disappearance time was 118.8 +/- 26.0 min. Between 9 and 16% of the administered dose was excreted in the urine within 24 h. Buserelin could also be detected in the plasma after intranasal administration of doses of 150, 300 and 450 micrograms. There was a significant difference in the area under the curve (AUC) for plasma levels after subcutaneous injection of 5 micrograms and intranasal administration of 150 micrograms, but not between the AUC values after the three intranasal doses. These results indicate that this method for radioimmunoassay of buserelin is suitable for analyzing the pharmacokinetics and bioavailability of buserelin in man.

Multiple fluorescence-based PCR-SSCP analysis.

Multiple fluorescence-based polymerase chain reaction single-strand conformation polymorphism (MF-PCR-SSCP) was developed. The target sequence was amplified by PCR using forward and reverse primers labeled with two different fluorescent dyes at their 5' ends. The amplified products were then heat-denatured, mixed with internal standard DNA markers labeled with a third fluorescent dye and applied to a temperature-controlled gel in an automated DNA sequencer, with a gel-temperature-controlling system. Mutations were detected as positional shifts of two-colored peaks in the electrophoretogram. The image data were analyzed by the computer program GENESCAN 672. The peak positions were standardized to internal DNA size markers. MF-PCR-SSCP analysis of 7 human tumor cell lines with 7 different single base mutations of the human K-ras oncogene detected all mutations even under the same electrophoresis conditions. Complete loss of heterozygosity was detected in two cell lines simultaneously. A gel temperature at 20 degrees C and polyacrylamide concentration of 10% gave the best separation. MF-PCR-SSCP is superior to the current PCR-SSCP in several ways: it does not involve radioactivity, migration patterns are standardized to internal standard DNA markers, there is a strict temperature-controlling system and the higher percentage of the gel enables better separation with resultant 100% detection of mutations most likely under one set of electrophoresis conditions.

An end-trimming method to amplify adjacent cDNA fragments by PCR.

We describe a method for cDNA cloning by PCR, which we named "an end-trimming method." This method can be used for PCR amplification and cloning of unknown cDNA fragments adjacent to a short stretch of a known sequence by using a combination of a sequence-matched primer with an ATCG sequence added to the 5' end (5'-ATCG-primer) and an adaptor-(dT)17-primer (dT-primer). The fragments amplified by PCR using a 5'-ATCG-primer, which were modified to have a 5'-ATC overhang by blocking the G site, were exclusively cloned into pUC19 with the vector having a 3'-TAG-5' complementary overhang with a confined direction of inserted fragments. Practical application of this method for the determination of rat amidophosphoribosyltransferase cDNA resulted in successful cloning of adjacent cDNA fragments.

Multiple fluorescence-based PCR-SSCP analysis using internal fluorescent labeling of PCR products.

The method to internally label PCR products with multiple colored fluorescent dyes was developed and applied to multiple fluorescence-based PCR single-stranded conformational polymorphism (MF-PCR-SSCP) analysis. PCR-amplified fluorescent DNA fragments, which were internally labeled by adding fluorescent dUTPs ([F]dUTPs) to the PCR mixture, were heat-denatured and applied to a nondenaturing polyacrylamide gel (SSCP gel) set on an automated DNA sequencer with a gel temperature-controlling system. The image data were analyzed by GENESCAN 672 software. In spite of differences in species and amount of integrated [F]dUTPs, the SSCP profiles were not significantly affected, even by the different labeling methods used-internal labeling or post-labeling at the 3' ends-in regard to the three different [F]dUTPs examined. However, the salt concentration of the solution containing the DNA samples affected the SSCP profiles. The internally labeled [F]dUTP-containing DNA fragments beyond 1000 bp in length were successfully digested with restriction endonucleases and subjected to SSCP analysis. MF-PCR-SSCP analysis with internal fluorescence labeling affords a simple and sensitive method to detect alterations in DNA sequences.

Two new nonsense mutations in type Ia antithrombin III deficiency at Leu 140 and Arg 197.

Using polymerase chain reaction-single strand conformation polymorphism (PCR-SSCP) and DNA sequencing, the molecular basis of hereditary type Ia antithrombin III (AT III) deficiency was disclosed in two families. One mutation was a change from T to A in the codon of TTA for Leu 140 forming a stop codon of TAA, which was confirmed by mutated primer-mediated PCR-HindIII digestion. The application of this method demonstrated that all four affected members had the mutant allele in a heterozygous state and that none of unaffected subjects had this mutation. Another mutation in the second family was a change from C to T in the codon of CGA for Arg 197 also forming a stop codon of TGA, which was confirmed by PCR-HaeIII digestion. Based on these, it was concluded that the two new nonsense mutations in the AT III gene in a heterozygous state are the molecular basis of hereditary type Ia AT III deficiency.

p53 gene mutation in N-butyl-N-(4-hydroxybutyl)nitrosamine-induced urinary bladder tumors and N-methyl-N-nitrosourea-induced colon tumors of rats.

We analyzed p53 mutations in 17 N-butyl-N-(4-hydroxybutyl) nitrosamine-induced bladder transitional cell carcinomas (TCCs) with or without areas of squamous cell carcinoma (SCC) of Long-Evans Cinnamon (LEC) and F344 rats, and in 7 N-methyl-N-nitrosourea-induced colon adenocarcinomas of LEC rats by polymerase chain reaction-single strand conformation polymorphism analysis and DNA sequencing. Of these bladder tumors, one TCC with moderately differentiated SCC had a T to G transversion mutation at codon 141, leading to a Val to Gly amino acid change. No p53 mutation was found in colon adenocarcinomas. Thus a p53 gene mutation seems infrequent in these rat bladder and colon carcinogenesis models even in the late stage.

Infrequent mutations of p16INK4A and p15INK4B genes in human pituitary adenomas.

The p16INK4A and p15INK4B genes on chromosome 9p21 encode the p16 and p15 inhibitors of cyclin D/cyclin-dependent kinase 4 complexes respectively. Mutations and deletions of the p16INK4A gene have been found in melanomas and many other types of tumors. To assess the role of the p16INK4A and p15NK4B genes in tumorigenesis of the pituitary gland, 31 sporadic pituitary adenomas and 2 pituitary adenomas in familial acrogigantism were examined for loss of heterozygosity on 9p21-22 and screened for mutations in the p161NK4A and p15INK4B genes. To identify pituitary adenomas which had lost 9p21-22, pituitary adenomas were genotyped with markers flanking the p16INK4A and p15INK4B loci. The frequency of mutations in coding regions of the p16INK4A and the p15INK4B genes in pituitary adenomas was determined with polymerase chain reaction-single strand conformation polymorphism analysis and sequencing of variants. Of the 33 pituitary adenomas, two revealed loss of 9p21-22 sequences, but none of them had tumor-specific mutations. We conclude that mutations of the p16INK4A and p15INK4B genes are not required for tumorigenesis of the pituitary gland.