Gene Literature Dashboard

The development of effective screening tests for colorectal tumors is essential given the high frequency of these cancers in the general population, and more especially in various groups at risk. Sporadic and hereditary colorectal cancers result from the accumulation of mutations in oncogenes, such as ras, myc, neu/HER2, and in tumor suppressor genes such as apc, dcc, p53. The detection of ras or p53 mutations in DNA extracted from stool has been shown to be feasible and might be useful for the development of new screening tests. Many mutations in these genes can also be used as new prognostic factors. Identification of mutation in the apc gene responsible for familial polyposis, or its indirect detection through the study of polymorphism in such families, is completely changing the previously recommended medical attitude for the screening of this disease, and therefore may decrease or even avoid major medical follow-up. These changes are also true for the nonpolyposis hereditary colorectal tumors, also called Lynch syndrome, since the responsible hMSH2, hMLH1, hPMS1 and hPMS2 genes have recently been cloned. Mutations in these genes do not seem to be limited to families with Lynch syndrome, and could account for a predisposition of some patients to develop colorectal or other tumors.

Presumptive tumor suppressor genes may be localized to specific chromosomes by the procedure of microcell fusion, whereby individual chromosomes derived from normal human cells are introduced into tumor cells. Allelic loss on chromosome 18 is commonly seen in endometrial carcinoma, and the DCC gene on chromosome 18q is a potential human tumor suppressor gene. In this study, we investigated the hypothesis that a gene on chromosome 18, possibly DCC, is capable of suppressing the tumorigenicity of endometrial carcinoma cells. Microcells from the mouse A9 cell clone containing one human chromosome 18 tagged with the pSV2-neo plasmid were fused with the highly tumorigenic endometrial carcinoma cell lines HHUA and Ishikawa, and G418-resistant microcell hybrids containing and extra copy of chromosome 18 were isolated. Clones isolated from the HHUA cell line were completely suppressed for tumorigenicity in nude mice, and clones from the Ishikawa line were suppressed or inhibited for tumorigenicity. In contrast, growth rates in vitro were not significantly affected in clones from either parental cell line. DCC expression was elevated in most of the suppressed hybrids. These results indicate that a gene on human chromosome 18 is capable of suppressing the tumorigenicity of endometrial carcinoma cells, and that DCC is a candidate for this endometrial carcinoma tumor suppressor gene.

In situ hybridization was used to map spatiotemporal expression patterns of the four known intronless POU-III transcription factor genes Brn-1, Brn-2, Brn-4, and Tst-1 in the developing rat forebrain vesicle, beginning on embryonic day 10. The results indicate that the proliferation layers (ventricular and subventricular) and mantle layer of the forebrain neural tube each display a strikingly unique pattern of regionalized POU-III expression. Within a particular region, or layer within a region, none to all four of the mRNAs may be detected, and during development a particular mRNA in a particular region displays one of five expression patterns, or a combination of these patterns, which may be described as conserved, lost, transient, acquired, or redeployed expression. In the developing brain as a whole, Brn-1 and Brn-2 early on display somewhat different spatial expression patterns that converge to essential identity in the adult, whereas Brn-4 expression is initially broad and becomes much more restricted in the adult, and Tst-1 expression expands greatly through development. Usually, though not always, expression patterns tend to correlate with major histological features in the forebrain (often internal or external sulci associated with proliferation zones), and little evidence for waves of expression moving through the whole forebrain over time was obtained. Thus, clear differences in hybridization intensity often are observed between the cerebral cortex, basal telencephalic nuclei, hypothalamus, ventral thalamus, dorsal thalamus, and pretectal region. In contrast, transverse bands of hybridization extending from the roof to the floor of the forebrain, corresponding to proposed neuromeres, were not observed with these probes. The results suggest that POU-III transcription factors help define specific regions in the early neuroepithelium as well as different cellular phenotypes in the ventricular, subventricular, and mantle layers of specific regions later in development. Thus, the functions of these regulatory proteins may be different in proliferating neuroepithelial cells, young neurons, and mature neurons and appear to be region-specific.

Huntington's disease (HD) results from the expansion of a polyglutamine encoding CAG repeat in a gene of unknown function. The wide expression of this transcript does not correlate with the pattern of neuropathology in HD. To study the HD gene product (huntingtin), we have developed monoclonal antibodies raised against four different regions of the protein. On western blots, these monoclonals detect the approximately 350 kD huntingtin protein in various human cell lines and in neural and non-neural rodent tissues. In cell lines from HD patients, a doublet protein is detected corresponding to the mutated and normal huntingtin. Immunohistochemical studies in the human brain using two of these antibodies detects the huntingtin in perikarya of some neurons, neuropiles, varicosities and as punctate staining likely to be nerve endings.

The Huntington's disease (HD) gene encodes a novel protein with as yet no known function. In order to identify the functionally important domains of this protein, we have cloned and sequenced the homologue of the HD gene in the pufferfish, Fugu rubripes. The Fugu HD gene spans only 23 kb of genomic DNA, compared to the 170 kb human gene, and yet all 67 exons are conserved. The first coding exon, the site of the disease-causing triplet repeat, is highly conserved. However, the glutamine repeat in Fugu consists of just four residues. We also show that gene order may be conserved over longer stretches of the two genomes. Our work describes a detailed example of sequence comparison between human and Fugu, and illustrates the power of the pufferfish genome as a model system in the analysis of human genes.

The relationship between the specific neuronal loss observed in Huntington's disease and the mutation in the IT15 gene responsible for this disease remains obscure. Using an antipeptide antibody against amino acids 3114-3141 of the human huntington protein, we demonstrate that striatal injection of quinolinic acid in mice induces increased immunoreactivity for huntington in some remaining neurons but not in glial cells. This increase is apparent in both neuronal cell bodies and in cell processes in the white matter six hours after excitotoxic challenge. This finding suggests that huntington may be involved in the response to excitotoxic stress in these neurons.

This study measured serotonin transporter (5-HTT) sites in the dorsal raphe nucleus (DRN) and hippocampus in Alzheimer's disease (AD). 5-HTT sites were significantly decreased in the DRN in AD, with significant reductions occurring in the lateral wings of the DRN complex. A significant reduction in 5-HTT sites were also observed in the CA2 subfield of the hippocampus and in the entorhinal cortex in AD. The results indicate that the integrity of 5-HT neurons in the DRN may be compromised in AD, and that region specific alterations in 5-HTT sites may occur in the hippocampal complex in AD.

Allelic loss is a common mechanism of inactivation of tumour-suppressor genes in colorectal carcinomas. A number of known or putative tumour-suppressor genes including NF1, BRCA1, NME1, NME2 and prohibitin are present on the long arm of chromosome 17, and this region has not been extensively analysed in colorectal tumours. In this study 72 colorectal carcinomas were examined for allelic loss at eight loci on chromosome 17. Allelic loss was frequent both at the p53 locus, which is known to be important in colorectal carcinoma, and also telomeric to p53 on 17p. Allelic loss continued to be present in more than 50% of cases in the pericentromeric region and on proximal 17q to the marker LEW101 (D17S40) at 17q22-23. The most telomeric markers on 17q showed lower rates of allelic loss. Analysis of cases with partial deletions which did not include the p53 locus showed a common region of overlap of the deletions centred on D17S40. This suggests the target of allelic loss on 17q is a tumour-suppressor gene in this region.

Human colorectal cancer tissue and matched uninvolved mucosa from 21 patients were examined by radioligand displacement for the presence of binding sites for bombesin-like peptides. Five cancers, but no uninvolved mucosa, expressed high-affinity, low-capacity bombesin binding sites (Kd = 6.53 nM, Bmax = 58.6 fmol mg-1 protein) of the gastrin-releasing peptide (GRP)-preferring subtype (IC50 4.8 nM). Bombesin-like peptides may have a role in the pathogenesis of colorectal cancer, and bombesin receptor antagonists may be of value in the treatment of receptor-positive tumours.

The present study reports on the frequency of MDM2 gene amplification and MDM2 protein expression in a series of 100 breast carcinomas and its association with accumulation of the p53 protein. Of the 100 cases, frozen samples for 82 cases were available for Southern blotting. Three of the 82 (4%) demonstrated MDM2 gene amplification of up to 6-fold. Immunohistochemical analysis of the formalin-fixed, paraffin-embedded tumours demonstrated that 7/97 (7%) had nuclear expression for MDM2 in 10-50% of the tumour cells (type 2 staining) and were denoted MDM2+. Two of the MDM2-amplified samples were MDM2+ with one of the two tumours also displaying type 2 p53 nuclear staining. Finally at the protein level, MDM2+ tumours were significantly associated with tumours having low levels of p53 staining (0-10% cells positive) (P = 0.03). We conclude that MDM2 gene amplification occurs at a lower frequency in breast cancer than in non-epithelial tumours. Alterations in MDM2 and p53 may represent alternative pathways in tumorigenesis, but they are not mutually exclusive in all cases.

The mechanisms underlying iron-induced liver fibrogenesis in patients with genetic hemochromatosis are poorly understood. We studied signs of Kupffer cell activation and inflammatory responses in liver biopsy specimens obtained from 15 patients with untreated and six patients with treated hemochromatosis. Immunohistochemistry was performed on 11 of the untreated and all treated patients. Three of the untreated patients (20%) had cirrhosis and eight (53%) had fibrosis. None had chronic active hepatitis (CAH). Immunohistochemistry indicated that 55% of the untreated patients had sparse intercellular adhesion molecule-1 (ICAM-1) expression by hepatocytes, and all of these had Kupffer cell iron overload. No ICAM-1 expression was seen by hepatocytes in treated patients or healthy controls. ICAM-1 was strongly expressed by hepatocytes from control patients with inflammatory liver disease. HLA-DR reactivity was seen on sinusoidal cells in all groups, but not on hepatocytes except for two of the control patients with CAH. Twenty-seven percent of the untreated hemochromatosis patients displayed moderate infiltration by CD3-positive lymphocytes. Electron microscopy of samples from untreated hemochromatosis patients showed hypertrophic Kupffer cells containing iron-rich remnants of phagocytosed hepatocytes. Fat-storing cells close to iron-laden hepatocytes contained multiple lipid droplets and adjacent collagen fibril bundles. Thus, in patients with untreated genetic hemochromatosis and Kupffer cell iron overload, hepatocytes occasionally express ICAM-1. In regions with heavy iron overload, Kupffer cell hypertrophy and transition of fat-storing cells are seen. Our findings indicate that release of factors from iron-loaded, activated Kupffer cells is of importance for the transformation of fat-storing cells and increased collagen deposition seen in genetic hemochromatosis.

This study reports on the biological data of ten patients with acute venous thrombo-embolism. They were treated for 5 days with continuous intravenous infusion of a fixed dose (0.05 mg/kg/hr) of a recombinant hirudin (r-H HBW 023 Behringwerke, Germany). The plasma level of r-H (HBW 023), assessed by an anti-factor IIa amidolytic activity, was stable after Day 2 and showed considerable individual variations. It correlated with APTT ratio, suggesting that this test is a reliable tool to monitor therapy. In contrast, thrombin time was constantly over 120 sec (control 15 sec) and consequently was not a useful parameter. Prothrombin time showed a slight, but significant, prolongation, which was correlated with the increase of APTT ratio. There was no bleeding time prolongation, platelet count, or ATIII level decrease. Levels of thrombin-antithrombin III complexes, and D-dimers, which were high in all patients on admission, decreased during the course of the treatment but remained abnormal on Day 5, showing an ongoing hemostasis and fibrinolysis activation: this is consistent with the delayed, but only slightly decreased thrombin generation evidenced by thrombin generation test performed on Day 3. These results suggest that thrombin inhibition by rH-hirudin at this dosage is only partial, which allows the generation of traces of thrombin needed for the feed-back thrombin production generated by factor V and VIII activation.

The putative tumor suppressor gene deleted in colorectal carcinoma (DCC), located on human chromosome band 18q21, is deleted or inactivated in many solid tumors. Its role in the pathogenesis of non-Hodgkin's lymphoma (NHL) has not been studied. Recently, inactivation of this gene was reported in cases of leukemia with monosomy 18. As monosomy 18 is frequently observed in low-grade NHL, we investigated the incidence of altered DCC gene expression in patients with NHL, and correlated it with the number of copies of chromosome 18. Fifteen unselected cases of NHL were studied for evidence of DCC gene expression by reverse transcriptase-polymerase chain reaction. The results were correlated with Southern blot analysis of the DCC gene and with the number of copies of chromosome 18 determined by fluorescent in situ hybridization (FISH). The controls were tissues from normal colon mucosa and normal tonsils. Eight of 15 (53%) NHL cases lacked DCC mRNA, and one expressed substantially less than normal. Southern blot analysis showed normal configuration of the DCC gene in all samples. Two copies of chromosome 18 were found in 9 of 11 samples studied by FISH: one case had a subpopulation of cells with monosomy 18 and one had trisomy 18. All controls expressed DCC. We conclude that DCC gene expression is frequently absent or decreased in NHL and may be involved in the pathogenesis of NHL. Monosomy 18 was not required for DCC inactivation.

A novel immunoglobulin-type protein expressed in blood vessels has been identified. The cDNA for AAMP (angio-associated, migratory cell protein) was first isolated from a human melanoma cell line during a search for motility-associated cell surface proteins. Upon analysis of the tissue distribution of AAMP, it was found to be expressed strongly in endothelial cells, cytotrophoblasts, and poorly differentiated colon adenocarcinoma cells found in lymphatics. The sequence of AAMP predicts a protein (M(r) 49,000) with distant identity (25%) to known proteins. It contains immunoglobulin-like domains [one with multiple homologies to deleted in colon carcinoma (DCC) protein], the WD40 repeat motif, and a heparin-binding consensus sequence. A 1.6-kilobase mRNA transcript of AAMP is detected in tissue culture cell lines and tissues. Affinity-purified polyclonal antibodies, anti-recombinant AAMP, and anti-peptide 189 (AAMP derived) recognize a M(r) 52,000 protein in human tissue and cellular extracts. The protein size is in keeping with the mRNA and predicted sequence. The AAMP-derived peptide, P189, contains a heparin-binding domain (dissociation constant, 14 pmol) and mediates heparin-sensitive cell adhesion. The shared expression of AAMP in endothelial cells, trophoblasts, and tumor cells implies a common function in migrating cells.

In voltage-dependent Ca2+ channels, the beta subunit interacts with the alpha 1 subunit via a cytoplasmic site. A biochemical assay has been developed to quantitatively describe the interaction between both subunits. In vitro synthesized 35S-labeled beta subunits specifically bind to a glutathione S-transferase (GST) fusion protein containing the alpha 1A interaction domain (AIDA, located between the amino-acids 383 and 400 of the cytoplasmic loop between the hydrophobic domains I and II). Kinetic analysis demonstrates that the association of 35S-labeled beta 1b subunit to the AIDA GST fusion protein occurs with a fast rate constant at 4 degrees C. The binding is almost irreversible as demonstrated by the absence of dissociation observed after an 8-h incubation with an 18-amino acid synthetic AIDA peptide. The alpha 1-beta binding site does not seem to be a target for cytoplasmic regulation. The interaction is mostly unaffected by changes in ionic strength, pH, and Ca2+ concentration or by protein kinase C phosphorylation. The specificity of subunit interaction in voltage-dependent Ca2+ channels was also followed by saturation analyses. The data obtained show that the AIDA GST fusion protein binds to a single site on the beta 1b with an apparent Kd of 5 nM. The affinities of AIDA GST fusion protein for various beta subunits was measured and demonstrate that beta subunits associate with different affinities to each alpha 1 interaction domain. The rank order of AIDA affinity for each beta subunit is as follows: beta 4 > beta 2a > beta 1b >> beta 3. The binding of the beta subunit to alpha 1 subunit can be inhibited in vitro by the AIDA synthetic peptide with an apparent Ki of 285 nM. This interaction can also be prevented in heterologous Ca2+ channels by the injection of the AIDA GST fusion protein into Xenopus oocytes. Our results demonstrate that the site of interaction between AID and beta subunit is responsible for anchoring the beta subunit to the alpha 1 subunit and thus allowing the beta subunit to modify Ca2+ channel activity.

Huntington's Disease (HD) is caused by expansion of a CAG repeat within a putative open reading frame of a recently identified gene, IT15. We have examined the expression of the gene's protein product using antibodies developed against the N-terminus and an internal epitope. Both antisera recognize a 350 kDa protein, the predicted size, indicating that the CAG repeat is translated into polyglutamine. The HD protein product is widely expressed, most highly in neurons in the brain. There is no enrichment in the striatum, the site of greatest pathology in HD. Within neurons, the protein is diminished in nuclei and mitochondria and is present in the soluble cytoplasmic compartment, as well as loosely associated with membranes or cytoskeleton, in cell bodies, dendrites, and axons. It is concentrated in nerve terminals, including terminals within the caudate and putamen. Thus, the normal HD gene product may be involved in common intracellular functions, and possibly in regulation of nerve terminal function. The product of the expanded allele is expressed, consistent with a gain of function mechanism for HD at the protein level.

The gene defective in Huntington's disease encodes a protein, huntingtin, with unknown function. Antisera generated against three separate regions of huntingtin identified a single high molecular weight protein of approximately 320 kDa on immunoblots of human neuroblastoma extracts. The same protein species was detected in human and rat cortex synaptosomes and in sucrose density gradients of vesicle-enriched fractions, where huntingtin immunoreactivity overlapped with the distribution of vesicle membrane proteins (SV2, transferrin receptor, and synaptophysin). Immunohistochemistry in human and rat brain revealed widespread cytoplasmic labeling of huntingtin within neurons, particularly cell bodies and dendrites, rather than the more selective pattern of axon terminal labeling characteristic of many vesicle-associated proteins. At the ultrastructural level, immunoreactivity in cortical neurons was detected in the matrix of the cytoplasm and around the membranes of the vesicles. The ubiquitous cytoplasmic distribution of huntingtin in neurons and its association with vesicles suggest that huntingtin may have a role in vesicle trafficking.

OX40 is a member of the TNF/NGF-receptor family expressed on activated T cells, whose ligand is found on activated T and B cells. In the present study, we show that cross-linking of OX40L on CD40L-stimulated B cells, alpha IgD dextran-stimulated B cells, or both results in a significantly enhanced proliferative response with no change in the cell survival rate. Furthermore, OX40 stimulation increases immunoglobulin heavy chain mRNA levels and immunoglobulin secretion, which could not be blocked by anti-cytokine antibodies. In additional molecular studies, we show that OX40L cross-linking results in the down-regulation of the transcription factor BSAP. This, in turn, leads to a change in the in vivo binding pattern of the immunoglobulin heavy chain gene 3' alpha enhancer, suggesting its activation. This effect may thus be one mechanism for OX40-induced increase in immunoglobulin secretion. In conclusion, our data suggest that the OX40-OX40L interaction is a novel pathway in T cell-dependent B cell proliferation and differentiation.

DNA replication of the adenovirus genome complexed with viral core proteins is dependent on the host factor designated template activating factor I (TAF-I) in addition to factors required for replication of the naked genome. Recently, we have purified TAF-I as 39- and 41-kDa polypeptides from HeLa cells. Here we describe the cloning of two human cDNAs encoding TAF-I. Nucleotide sequence analysis revealed that the 39-kDa polypeptide corresponds to the protein encoded by the set gene, which is the part of the putative oncogene associated with acute undifferentiated leukemia when translocated to the can gene. The 41-kDa protein contains the same amino acid sequence as the 39-kDa protein except that short N-terminal regions differ in both proteins. Recombinant proteins, which were purified from extracts of Escherichia coli, expressing the proteins from cloned cDNAs, possessed TAF-I activities in the in vitro replication assay. A particular feature of TAF-I proteins is the presence of a long acidic tail in the C-terminal region, which is thought to be an essential part of the SET-CAN fusion protein. Studies with mutant TAF-I proteins devoid of this acidic region indicated that the acidic region is essential for TAF-I activity.

We have ectopically expressed transcription factor ETS1 in two different highly tumorigenic human colon cancer cell lines, DLD-1 and HCT116, that do not express endogenous ETS1 protein and have obtained several independent clones. The expression of wild-type ETS1 protein in these colon cancer cells reverses the transformed phenotype and tumorigenicity in a dose-dependent manner. By contrast, expression in DLD-1 cells of a variant form of ETS1, lacking transcriptional activity, did not alter the tumorigenic properties of the cells, suggesting that the reduction in tumorigenicity in these clones was specific for the wild-type ETS1 gene products. Since these colon cancer cells have multiple genetic alterations, the system described in this paper could be a good model to study the suppression of tumorigenicity at a transcriptional level, which could lead to the design and development of novel drugs for cancer treatment.

Tryptophan and 5-bromotryptophan (5-BrTrp) are relatively potent inhibitors of sickle-haemoglobin polymerization. The binding sites of these compounds to normal and sickle haemoglobin (HBA and HBS) have been suggested, but not firmly established, through the use of spin-labelled derivatives and/or computer modeling. In the present study we approached the problem by utilizing the technique of photoaffinity labelling. The cyanomet forms of HBA and HBS were subjected to photoaffinity labelling with N alpha-(4-azidotetrafluorobenzoyl)tryptophan and N alpha-(1-ethyl-2-diazomalonyl)-5-bromotryptophan respectively. Both irradiated samples of HBA and HBS were denatured, digested with trypsin, and then separated by reversed-phase HPLC. A labelled tryptic peptide was isolated from the photolabelling of HBS with N alpha-(1-ethyl-2-diazomalonyl)-5-bromotryptophan. The peptide was identified to be Val1(alpha)-Lys7(alpha), with the label attached to Val1(alpha), by virtue of amino acid analysis and sequencing, in conjunction with fast-atom-bombardment MS. The binding mode of N alpha-(1-ethyl-2-diazomalonyl)-5-bromotryptophan is proposed and its relevance to the potency of the 5-BrTrp-based anti-sickling agents is discussed.

Inhibitors of rat and human Alpha- and Mu-class glutathione S-transferases that effectively inhibit the glutathione (GSH) conjugation of bromosulphophthalein in the rat liver cytosolic fraction, isolated rat hepatocytes and in the rat liver in vivo have been developed. The GSH analogue (R)-5-carboxy-2-gamma-(S)-glutamylamino-N-hexylpentamide [Adang, Brussee, van der Gen and Mulder (1991) J. Biol. Chem. 266, 830-836] was used as the lead compound. To obtain more potent inhibitors, it was modified by replacement of the N-hexyl moiety by N-2-heptyl and by esterification of the 5-carboxy group with ethyl and dodecyl groups. In isolated hepatocytes, the branched N-2-heptyl derivatives were stronger inhibitors of GSH conjugation of bromosulphophthalein than the N-hexyl derivatives. The ethyl ester compounds were more efficient than the corresponding unesterified derivatives. The dodecyl ester of the N-2-heptyl analogue was the most effective inhibitor in isolated hepatocytes, but was relatively toxic in vivo. However, the corresponding ethyl ester was a potent in vivo inhibitor: GSH conjugation of bromosulphophthalein (as assessed by biliary excretion of the conjugate) was decreased by 70% after administration of a dose of 200 mumol/kg. The isoenzyme specificity of the inhibitors towards purified rat and human glutathione S-transferases was also examined. The unesterified compounds were more potent than the esterified analogues, and inhibited Alpha- and Mu-class isoenzymes of both rat and human glutathione S-transferase (Ki range 1-40 microM). Other GSH-dependent enzymes, i.e. GSH peroxidase, GSH reductase and gamma-glutamyltranspeptide, were not inhibited. Thus (R)-5-ethyloxycarbonyl-2-gamma-(S)-glutamylamino-N-2-hept ylpentamide, the in vivo inhibitor of GSH conjugation, may be useful in helping to assess the role of the Alpha and Mu classes of glutathione S-transferases in cellular biochemistry, physiology and pathology.

Low plasma heparin cofactor II (HCII) levels are associated with a thrombotic tendency, and we have previously shown these to be decreased in a variety of haemolytic conditions. The risk of thrombosis is recognized to be increased in both thalassaemia major (TM) and intermedia (TI), although the exact mechanisms are poorly understood. HCII levels have therefore been compared in 20 untransfused patients with TI and 20 regularly transfused TM patients to determine the influence of transfusion on HCII. Additionally, untransfused TI patients have been commenced on regular red cell transfusion and the effects on correction of low HCII levels investigated. HCII levels were significantly lower in the untransfused TI patients (mean 0.56 +/- 0.06 U/ml) compared to TM patients (mean 0.85 +/- 0.1 U/ml; P < 0.001). Levels in TI were significantly less than in healthy age-matched controls (P < 0.001) and correlated with Hb values (r = 0.8), whereas levels in TM were at the lower end of the normal range. ATIII values were within the normal reference range in both TI and TM, and HCII antigen showed a parallel reduction to HCII activity, indicating that reduction in HCII is not a consequence of increased thrombin consumption. Three patients with TI were commenced prospectively on hypertransfusion programmes which resulted in a slow normalization of their levels taking 2-3 months. These findings support a hypothesis that the low HCII levels are related to increased red cell turnover and can be normalized once this turnover has been suppressed by hypertransfusion. The thrombotic risk to patients with low HCII levels in the presence of haemolysis might in principle be decreased by such transfusion regimes.

South Asians in Britain have a high incidence of ulcerative colitis and a low incidence of colorectal cancer. The pattern of mucus production in 12 South Asian and 16 European colitics and a control group of 19 South Asians was studied. Three types of mucin were identified after organ culture of colonic biopsy specimens with a dual label of [3H]-glucosamine and sodium [35S]-sulphate: type A had a high [35S]:[3H] ratio and high incorporation ([3H] dpm/micrograms DNA > 500); type B had a low ratio and high incorporation; and type C had low incorporation but with either high (C1) or low (C2) ratios. European colitic mucins show a significant reduction in the level of sulphation detected by mucin histochemistry with high iron diamine/Alcian blue staining, together with predominantly type B or C2 mucins (low sulphation). South Asian colitics showed histochemically normal patterns of high sulphation and largely type A and C1 mucins (high sulphation). There was no correlation of mucin type with disease activity index in either ethnic group. The appearance of apparently normal mucin in patients with ulcerative colitis may be a useful marker for the identification of a subgroup at low risk of colorectal cancer.