Lithocholic acid inhibits the expression of HLA class I genes in colon adenocarcinoma cells. Differential effect on HLA-A, -B and -C loci.

Loss of HLA antigen expression is considered to be one of the mechanisms whereby tumor cells escape immune surveillance. We recently observed reduced or lost expression of HLA antigens during human colon carcinogenesis. We studied the effect of bile acids (BAs), long implicated in the pathogenesis of colon cancer, on the expression of HLA class I antigens in human colon adenocarcinoma cells. Lithocholic acid (LCA) decreased by 42% the expression of HLA class I antigens on the surface of these cells. This dose-dependent reduction was specific for both the target genes and the chemical structure of LCA, and was not evident in cultured liver cells. None of the other BAs that were tested manifested this effect. LCA, and to a lesser extent deoxycholic acid (DCA), decreased steady-state HLA class I mRNA levels. LCA decreased the rate of transcription of HLA-B (64%) and HLA-C (87%) but not HLA-A; DCA had a similar but less pronounced effect. In transient gene expression (CAT assays) experiments, we evaluated the role of a 0.6-0.7 kb EcoRI/XbaI sequence from the 5' flanking region of HLA-A2, -B7 and -Cw7 genes in the regulation of class I gene expression by LCA. LCA down-regulated by 70% the expression of the reporter gene for all three genes. We interpret these results as indicating a differential regulation of the three HLA loci by LCA. Our findings, demonstrating a profound effect of LCA on HLA class I gene regulation, raise the possibility that such a mechanism may be operative in vivo.

Cloning, sequence analysis, and expression of the bacteriophage T4 cd gene.

The cd gene of bacteriophage T4, which encodes the enzyme deoxycytidylate deaminase, was isolated as a 1.9-kilobase DNA fragment and completely sequenced. The deduced amino acid sequence was found to be 193 residues long compared with 188 for the corresponding enzyme from bacteriophage T2. There were nine amino acid differences between the two enzymes in addition to a 5-residue insert near the carboxyl terminus of the T4 deaminase which was not present in the T2 deaminase. The cd-containing fragment also contained all of gene 31 (Nivinskas, R., and Black, L. W. (1988) Gene (Amst.) 73, 251-257) and thus precisely locates the two genes relative to one another within the T4 phage genomic map. Attempts to place the cd gene within a high expression vector have not been successful so far due to possible toxic effects of the gene product. However, placement of the gene within pUC18 resulted in a degree of expression which is about 10-20 times that found in T4-infected Escherichia coli. The enzyme was purified to homogeneity and found to possess properties similar to T2 phage deoxycytidylate deaminase.

HLA-JY328: mapping studies and expression of a polymorphic HLA class I gene.

The JY328 clone was identified in a human genomic library using cDNA corresponding to mRNA for HLA-B7 as a probe. The L/328 cell line was established by cotransformation of mouse Ltk- cells with the herpes thymidine kinase gene and clone JY328. On Northern blots, RNA from L/328 strongly hybridized to an HLA class I probe, and an antigen was recognized by an anti-HLA class I framework antibody on the cell surface. A DNA probe corresponding to a segment of intron 7 was developed by comparing the nucleotide sequence of clone JY328 with that of other HLA class I-type genes. Using the radiolabeled probe to screen Southern blots of DNA from families with siblings exhibiting intra-HLA recombinations, a restriction fragment length polymorphism was revealed--a 1.4 kb BstE II band not present in all individuals. A corresponding fragment was apparent in the base sequence of clone JY328. The occurrence of this band on Southern blots established that JY328 maps distinct from and centromeric to the HLA-C locus and near to the HLA-B locus. Antibody absorption studies and cytotoxicity tests indicated that the JY328 gene product was not an HLA-B antigen but that it did specifically absorb CW7-specific antibody. In sum, these results suggest a novel, polymorphic HLA class I gene which expresses a product serologically similar to HLA-Cw7 but which does not map within the corresponding locus.

Molecular organization of the class I genes of human major histocompatibility complex.

In this brief review, our main emphasis has been on the analysis of the sequence diversity among various class I genes and their functional implications. The availability of complete nucleotide sequences of 7 different genes representing different loci allowed us to derive a consensus sequence. One mouse MHC Class I gene was included in these comparisons as a representative of H2 genes Evolutionary patterns can be seen on the basis of divergence of various genes from the derived consensus sequence. At least 1 human gene which has a promoter similar to that of H2 genes and which contains a single initiation codon following this promoter (unlike all other human genes and like all the H2 genes) has been identified. Both variable and homology regions can be identified in the entire length of the gene. While exons show relatively strong conservation of sequences, the introns have many variable regions, introns 6 and 7 being the most heterogeneous. Stretches of conserved nucleotide sequences are noticed at the 3' regions of most introns. Estimation of total number of class I genes is presented on the basis of cloning experiments, and the abundance of 1 particular pseudogene is discussed.

Structure and polymorphism of class I MHC antigen mRNA.

We have used cDNA cloning and primer extension techniques to determine the complete nucleotide sequence of HLA-B7 mRNA. The 5'-untranslated sequence of the mRNA is rather short and the putative promoter has weak homology to the conventional "TATA" sequences. The 5' end and the polyadenylation site define the transcription unit of the B7 gene to be about 3.5 kb long. Comparison of the translated nucleotide sequence of this cDNA with the amino acid sequence of the heavy chain of the B7 antigen showed two amino acid differences. In addition, comparison with the sequences of the coding and untranslated regions of several HLA and H-2 genes showed that the class I histocompatibility molecules consist of four variable segments separated by three regions of homology. Analysis of the DNA polymorphisms revealed that in the variable segments the majority of the nucleotide substitutions are nonsilent, while in the homology regions the majority of substitutions are silent. Further analysis of the nature of the amino acid substitutions revealed the predominance of nonconservative replacements/changes in both the variable segments and the homology regions except the transmembrane part of the molecule. Selection at both the protein and the codon levels contributes to the pattern of mutations found in class I histocompatibility molecules.

An HLA-B locus probe clarifies endonuclease polymorphism of major histocompatibility complex class I genes.

A DNA probe specific for the HLA-B locus has been isolated from a broadly cross-reactive HLA class I genomic clone. Locus specificity of the probe appears to be derived primarily from a stretch of approximately 180 nucleotides comprising the last (7th) intron of the original B7 gene. Use of the probe to analyze Southern blots of genomic DNA from unrelated individuals provides the first direct demonstration of intragenic localization of an HLA allele-specific restriction endonuclease site. Availability of this probe should make practicable the study of HLA-B locus restriction fragment length polymorphism as genetic markers of disease susceptibility, and should provide a model for developing probes specific for other HLA class I loci.

RNA polymerase I in hepatoma 3924A: mechanism of enhanced activity relative to liver.

Two major cyclic nucleotide-independent protein kinases, NI and NII, have been identified in Morris hepatoma 3924A and rat liver. When expressed per unit DNA, the activities of protein kinase NI and NII were 1.3 and 12 times greater, respectively, in the hepatoma than in liver. Protein kinase NII, but not NI, was capable of phosphorylating and activating the DNA-dependent RNA polymerases I and II. Phosphorylation of RNA polymerase I was accompanied by an increase in average size of the RNA synthesized in vitro, whereas phosphorylation of RNA polymerase II was concomitant with an elevation in the number of RNA chains initiated. RNA polymerase I polypeptides of Mr 120,000, 65,000 and 25,000 were phosphorylated by protein kinase NII; RNA polymerase II polypeptides of Mr 214,000, 140,000 and 21,000 were modified by this kinase. In contrast to the purified hepatoma enzyme, RNA polymerase I activity in nuclear lysates was not affected by addition of protein kinase NII. In vitro phosphorylation of the tumor lysate followed by immunoprecipitation of RNA polymerase I polypeptides indicated little or no phosphate transfer to the 65,000 Mr polypeptide of the enzyme. These data suggested that the tumor enzyme, particularly the 65,000 Mr polypeptide, was highly phosphorylated in vivo, but becomes dephosphorylated during purification. Unlike the tumor enzyme, RNA polymerase I in the liver lysate responded to protein kinase addition; phosphorylation of the liver polymerase I polypeptides of Mr 120,000, 65,000 and 25,000 was observed. These observations indicate that the liver enzyme is not completely phosphorylated (activated) in vivo and that the relatively rapid rate of ribosomal RNA synthesis in the rapidly growing hepatoma may result, at least in part, from a polymerase I which is maximally phosphorylated.

Spermine-mediated phosphorylation of RNA polymerase I and its effect on transcription.

A nuclear protein kinase, designated NII, was purified essentially to homogeneity from the Morris hepatoma 3924A. In the presence of excess Mg2+, phosphorylation of casein by the kinase was stimulated by spermine (1-5 mM) and was inhibited completely by 0.1 microgram/ml heparin. The apparent Km for casein was reduced in the presence of spermine. Spermine preferentially augmented phosphorylation of threonine residues. The kinase was also associated with highly purified RNA polymerase I and appears to correspond to two polypeptides (Mr 42,000 and 24,600) of the polymerase. RNA polymerase I polypeptides of Mr 120,000 (S2), Mr 65,000 (S3) and Mr 24,600 (S5) were phosphorylated by the endogenous kinase. Spermine enhanced phosphorylation of the RNA polymerase I subunits as much as 20-fold. Phosphorylation activated RNA polymerase I; the phosphorylated enzyme synthesized longer product with no apparent effect on the number of RNA chains initiated.

Activation of purified hepatoma RNA polymerase I by homologous protein kinase NII.

We have recently purified a cyclic nucleotide-independent, heparin-sensitive nuclear protein kinase (NII) from Morris hepatoma 3924A and demonstrated an apparent relationship of this kinase to the two subunits (Mr = 42,000 and 24,600) of RNA polymerase I. When homogeneous protein kinase NII was recombined with purified homologous RNA polymerase I containing limiting quantities of endogenous kinase, RNA synthesis was stimulated as much as 5-fold during a 90-min incubation. The enhanced RNA synthesis was due to an increase in the average RNA chain length; protein kinase did not alter the number of RNA molecules synthesized by the polymerase. Phosphorylation of RNA polymerase occurred at serine and threonine moieties. Unlike the NII kinase, purified homologous NI kinase did not phosphorylate RNA polymerase I and, as a result, did not alter transcription. These data indicate that 1) RNA polymerase I is activated by protein kinase NII, 2) endogenous protein kinase NII remaining with highly purified RNA polymerase I does not fully phosphorylate RNA polymerase I in vitro, and 3) protein kinase NII is capable of regulating RNA polymerase I activity by preventing premature termination of RNA chains.

Transcriptionally active RNA polymerases from Morris hepatomas and rat liver. Elucidation of the mechanism for the preferential increase in the tumour RNA polymerase I.

The amount and/or activity of DNA-dependent RNA polymerase I, Ii and III from resting liver, regenerating liver and a series of Morris hepatomas (5123D, 7800, 7777, 3924A) were determined after extraction of the enzymes from whole tissue homogenates and subsequent fractionation by DEAE-Sephadex column chromatography. When compared with resting liver, the tumours exhibited a characteristic enzyme pattern in which polymerase I, but not II, was increased. The increase in RNA polymerase I was proportional to the tumour growth rates. Alterations in polymerase III were confined to the most rapidly proliferating hepatomas. By contrast, all classes of RNA polymerase were found to be increased during liver regeneration. Relative to resting liver, the fastest growing tumour, 3924A, exhibited the highest activities and/or amounts of RNA polymerase I (8-fold) and III (5-fold) per g of tissue. These alterations in the tumour RNA polymerases were reflected in corresponding increases in the transcriptionally active (bound or chromatin-associated) enzyme population. The mechanisms underlying the augmented synthesis of RNA in vitro by bound polymerase I from hepatoma 3924A were elucidated by product analysis. The results indicated that, relative to liver RNA polymerase I, the tumour enzyme produced more nascent RNA chains and elongated these chains at a faster rate. The number of 3'-termini, as measured by incorporation into uridine, was higher in the hepatoma even under conditions which prevented re-initiation. suggesting increased amount of transcriptionally active RNA polymerase I in the tumour.