Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Journal of Cell Science 106, 771-783 (1993) Printed in Great Britain © The Company of Biologists Limited 1993 771 Differential expression of the human mucin genes MUC1 to MUC5 in relation to growth and differentiation of different mucus-secreting HT-29 cell subpopulations Thécla Lesuffleur1,*, Nicole Porchet2, Jean-Pierre Aubert2, Dallas Swallow3, James R. Gum4, Young S. Kim4, Francisco X. Real5 and Alain Zweibaum1 1INSERM U178, 16 avenue Paul-Vaillant Couturier, 94807 2INSERM U16, Lille, France 3MRC Human Biochemical Genetics Unit, London, UK 4Gastroenterology Research Laboratory, VA Hospital, San 5Institut Municipal d’Investigacio Mèdica, Barcelona, Spain Villejuif Cedex, France Francisco, CA, USA *Author for correspondence at Unité de Recherches sur la Différenciation Cellulaire Intestinale SUMMARY Mucin expression was analysed, in relation to cell growth, in parental HT-29 cells and in two populations of mucus-secreting HT-29 cells selected by adaptation to methotrexate (HT29-MTX) or 5-fluorouracil (HT29FU). These two populations express mature mucins that differ in their immunoreactivity to antibodies against gastric (HT29-MTX) or colonic mucins (HT29-FU). In the parental population, at late confluency, only very few cells produce mucins or the MUC1 glycoprotein, this being consistent with the low level of expression of the mRNAs corresponding to the MUC1 to MUC5C mucin genes. In the HT29-MTX and HT29-FU populations, the appearance of mucus droplets, as shown by histochemistry and immunofluorescence, starts a few days after confluency, progressively involving a greater proportion of cells and reaching a steady state at late confluency. The MUC1 glycoprotein appears earlier, already being detectable in preconfluent cells. Its distribution is restricted to the apical surface of the cells and is dis- tinct from that of the mucus droplets. In both populations the growth-related levels of MUC1 mRNA are concordant with the apparent levels of expression of the MUC1 glycoprotein. The levels of MUC2, MUC3, MUC4 and MUC5C mRNAs differ from one population to another and, within each population, according to the stage of the culture. The highest levels of MUC2 and MUC4 mRNAs are found in the HT29-FU cells, whereas the highest levels of MUC3 and MUC5C are found in the HT29-MTX cells, suggesting that the differences observed in the mature mucins expressed by either population may be related to which MUC genes are expressed. In both populations significant or even high levels of MUC mRNAs are already present in early cultures, i.e. at a stage when the mature mucins are not yet detectable, suggesting that mucin maturation is a later event. INTRODUCTION apomucins. MUC1 (Gendler et al., 1987; Lan et al., 1990) encodes a transmembrane mucin-like glycoprotein expressed in many epithelial cell types. MUC2 and MUC3 cDNA clones have been isolated from a human small intestine cDNA library (Gum et al., 1989, 1990, 1992) and from tracheobronchial libraries (Gerard et al., 1990; Jany et al., 1991). MUC4 and a family of other partial cDNAs included under the name of MUC5 have been isolated from a tracheobronchial library (Crepin et al., 1990; Nguyen Van Cong et al., 1990; Porchet et al., 1991; Aubert et al., 1991; Dufosse et al., 1993). The only full-length cDNA sequence currently available is that of MUC1 (Gendler et al., 1990; Lan et al., 1990), although most of the MUC2 cDNA has been sequenced (Gum et al., 1992). Mucin-encoding genes It has long been recognized that intestinal mucins are highly heterogeneous in their glycosylation, since this is readily detectable by a variety of biochemical and immunological techniques (Wesley et al., 1985; Podolsky, 1985a,b; Podolsky et al., 1986). In contrast, less is known about the synthesis and regulation of expression of apomucins or their structure and heterogeneity. Recent advances in two areas, namely the cloning of cDNAs encoding apomucins and the isolation of permanent lines of cultured mucus-secreting cell populations, have facilitated the analysis of this problem. It is now known that several distinct genes encode the Key words: mucins, MUC genes, HT-29 772 T. Lesuffleur and others seem to be characterized by a high level of polymorphism due to the presence of variable numbers of tandem repeat sequences (Swallow et al., 1987; Gendler et al., 1988; Griffiths et al., 1990; Fox et al., 1992; Nguyen Van Cong et al., 1990; Porchet et al., 1991; Toribara et al., 1991; Gross et al., 1992). The identification of several distinct mucinencoding genes raises a number of general questions regarding the biology of mucus-secreting cells. (1) How many apomucin genes can be expressed in one tissue? (2) Are multiple mucin genes expressed by a single cell? (3) What is the relationship between morphological differentiation and mucin mRNA and protein synthesis? (4) what is the relationship between apomucin structure and glycosylation? The mucus-secreting subpopulations isolated from the HT-29 cell line should help in answering these questions. This cell line is of particular interest with regard to mucins. Indeed, from the original population, which contains a small proportion (<0.5%) of goblet cells (Augeron and Laboisse, 1984; Lesuffleur et al., 1990, 1991c), several groups have isolated, through selective pressure, clones or subpopulations of mucus-secreting cells (Augeron and Laboisse, 1984; Huet et al., 1987; Hafez et al., 1990; Kreusel et al., 1991; Devine et al., 1991). There is some evidence that the mucins secreted by HT-29 cells are heterogeneous as to the oligosaccharide epitopes that they express. Mucins secreted by the clones isolated from galactose-adapted cells (Huet et al., 1987) were found to share epitopes with normal colonic mucins (Phillips et al., 1988), whereas those secreted by the clones isolated from sodium butyrate-treated cells (Augeron and Laboisse, 1984) are characterized by sialylated oligosaccharide chains of gastric type (Capon et al., 1992). Accordingly, the small proportion of goblet cells present in the parental line were shown to share antigenic properties with either colonic or gastric mucins as disclosed by their immunoreactivity to polyclonal antibodies raised against crude colonic or gastric mucins (Lesuffleur et al., 1990). We have recently isolated, through selective pressure with methotrexate (MTX) (Lesuffleur et al., 1990, 1991a; Dahyia et al., 1992) or 5-fluorouracil (FU) (Lesuffleur et al., 1991c), two distinct subpopulations of mucus-secreting HT-29 cells that maintain their ability to differentiate under normal culture conditions. These two populations differ as to the morphology of the cells and to the type of immunoreactivity of their mucins. Cells adapted to 10−6 M MTX form a homogeneous monolayer of polarized goblet cells (Fig. 1a) that exhibit a discrete apical brush border endowed with intestinal hydrolases and secrete mucins of gastric immunoreactivity (Lesuffleur et al., 1990, 1991a), two characteristics that are reminiscent of the differentiation of the human fetal colonic epithelium (Zweibaum et al., 1991; Bara et al., 1986). Cells adapted to 10−5 M FU form a mixed population involving a majority (80%) of polarized domeforming cells, devoid of a brush border, and a minority (20%) of mucus-secreting cells that, in contrast to MTXadapted cells, are organized into multilayered foci of nonpolarized goblet cells that accumulate mucins within the intercellular spaces (Fig. 1b); these mucins being of colonic immunoreactivity (Lesuffleur et al., 1991c). These differentiation characteristics, like those reported by the other groups, have mainly been analysed in late post-confluent cultures and nothing is known of the time-course of the goblet cell differentiation process in relation to cell growth, a relationship that has been well documented in enterocytelike differentiated cultured cells like Caco-2 cells or other HT-29 cell populations (Zweibaum et al., 1991). The aims of this study were to analyse the time course of differentiation of mucus-secreting MTX and FU cells, to examine the expression of the mucin gene mRNAs in relation to the growth and to the phenotype of these two populations compared with the parental line, and to determine whether the differences in phenotype observed between MTX and FU cells are associated with differences in the mucin genes that are expressed in these cells. MATERIALS AND METHODS Cell culture The parental HT-29 cell line (Fogh and Trempe, 1975), obtained from late Dr Jorgen Fogh (Memorial Sloan Kettering Cancer Center, Rye, NY), was established from a blood group A patient (J. Fogh, personal communication). The cell line was used between passages 169 and 182 and is referred to as HT-29. The subpopulations obtained by adaptation to 10−5 M 5-fluorouracil (Lesuffleur et al., 1991c) and 10−6 M methotrexate (Lesuffleur et al., 1990) are referred to as HT29-FU and HT29-MTX, respectively. Drug-adapted cells were used after several weekly passages (4 to 20) in the absence of drug. They were routinely grown in 25 or 75 cm 2 plastic T-flasks (Corning Glassworks, Corning, NY) in Dulbecco’s modified Eagles’s minimum essential medium supplemented with 10% heat-inactivated (56°C, 30 minutes) fetal bovine serum (Boehringer Mannheim Biochemicals, Mannheim, Germany) at 37°C in a 10% CO2/90% air atmosphere. For maintenance or experimental purposes all cells were seeded at 2×104 cells per cm2. The medium was changed daily. Antibodies and lectins Rabbit polyclonal antibodies against human gastric (L56) and colonic (L53) mucins were prepared as reported (Lesuffleur et al., 1990); a pool of mAbs against mucin M1 epitopes that specifically react with gastric mucins (Bara et al., 1986) was obtained from Dr J. Bara (INSERM U55, Paris); mAb ZE4 was raised against normal human small intestinal mucosa and was found to react with colonic and small intestinal mucins, but not with gastric mucins (D. Swallow, unpublished results); mAbs BC1, BC2 and BC3 (Xing et al., 1989), and LICR LON M8 (McIlhinney et al., 1985), which recognize the tandem repeat sequence of the MUC1 gene product were obtained from Dr McKenzie (University of Melbourne, Australia) and The Ludwig Institute for Cancer Research, Sutton, UK, respectively. Dipeptidylpeptidase-IV (DPPIV) was detected using mAb HBB 3/775/42 (Hauri et al., 1985) raised against the human small intestinal enzyme (obtained from Dr H. P. Hauri, Biocenter, Basel, Switzerland) and a rabbit immune serum raised against the porcine kidney enzyme (Darmoul et al., 1991) (obtained from Dr G. Trugnan, INSERM U239, Paris). Polyclonal rabbit antibodies against porcine villin (Robine et al., 1985) were obtained from Dr D. Louvard (Institut Pasteur, Paris). The following monoclonal antibodies were used for studying the saccharide epitopes associated with mucins: mAbs against AB blood group antigens from the Second International Workshop and Symposium on Monoclonal Antibodies against Human Red Blood Cells and Related Antigens (Oriol et al., 1990), obtained from Dr R. Oriol (INSERM U178, Villejuif); 7LE against Lea antigen (Bara et al., 1987), 2-25LE against Leb (Bara et al., 1986), and 12-4LE against Ley (Bara et al., 1988), obtained from Mucin gene expression in HT-29 cells 773 Dr J. Bara (INSERM U55, Paris); Cu-1 against Tn antigen (Takahashi et al., 1988) and B72.3 against sialyl-Tn (Nuti et al., 1982), obtained from Dr K. O. Lloyd (Memorial Sloan Kettering Center, New York, NY); G14 against αGalNAc (Dippold et al., 1987), obtained from Dr W.G. Dippold (Johannes Gutenberg Universität, Mainz, Germany); K21 against precursor type 1 (Rettig et al., 1985), obtained from Dr W. J. Rettig (Memorial Sloan Kettering Center, New York, NY); H15 against sialyl-Lea, obtained from Dr J. Sakamoto (Aichi Cancer Center, Nagoya, Japan). Ulex europaeus, Maackia amurensis and Sambucus nigra lectins labelled with rhodamine were from EY laboratories (San Mateo, CA, USA). tion buffer containing 50% formamide, 5× SSC, 1× Denhardt’s solution, 0.1% SDS, 20 mM sodium phosphate (pH 6.5), and 250 mg/ml denatured salmon sperm DNA. Filters were then hybridized with the 32P-labeled probe for 20 hours at 42°C in prehybridization buffer containing 10% dextran sulphate (Thomas, 1980). Blots were washed twice with 2× SSC, 0.1% SDS at room temperature, once with 0.1× SSC, 0.1% SDS at 55°C, and once, using the same buffer, at 65°C for 15 minutes. Blots were then processed for autoradiography. To normalize for RNA, filters were dehybridized and rehybridized with the actin cDNA probe. Immunofluorescence and histochemistry RESULTS Indirect immunofluorescence was performed on unpermeabilized cell layers and on cryostat sections of cell layer rolls as previously reported (Lesuffleur et al., 1990, 1991a,b). Cell layers and sections were fixed (15 minutes, room temperature) in either paraformaldehyde (3.5% in PBS Ca2+,Mg2+-free) or ethanol. Fluorescein-coupled sheep anti-rabbit and anti-mouse globulins were from Institut Pasteur Production (Marnes la Coquette, France). Rhodamine-coupled sheep anti-rabbit and anti-mouse globulins were from Boehringer Mannheim. Periodic acid/Schiff and Alcian blue staining (pH 2.5) of mucins was done on cryostat sections of cell layers fixed in ethanol following standard procedures. cDNA probes MUC1 was detected using PUM24P (Yonezawa et al., 1991), MUC2 with SMUC41 (Gum et al., 1989), MUC3 with SIB124 (Gum et al., 1990), MUC4 with JER64 (Porchet et al., 1991) and MUC5C with JER58 (Crepin et al., 1990; Aubert et al., 1991). Actin was detected with cDNA probe pA2 (Cleveland et al., 1986). Northern blot analysis Total RNA was extracted using guanidium isothiocyanate and centrifugation through a CsCl gradient (Chirgwin et al., 1979). Poly(A)+ RNA was purified by chromatography on oligo(dT)-cellulose (Davis et al., 1986). Samples (4 mg) of mRNAs were fractionated by electrophoresis in 1% agarose gels after denaturation in 1mM glyoxal (Thomas, 1980). Fractionated samples were transferred to reinforced nitrocellulose (Schleicher and Schuell, Dassel, Germany) in the presence of 2 M ammonium acetate, 0.02 M NaOH. Filters were incubated overnight at 42°C in prehybridiza- Characteristics of mucin expression in relation to growth and differentiation phenotype The morphological differentiation of the cells, as well as the secretion of mucins when it occurs, is a growth-related phenomenon, starting after the cells have reached confluency, i.e. under our culture conditions, between 7 and 9 days, depending on the cell populations In the parental HT-29 cells, goblet cells are detectable only after confluency, which is complete at day 7, and will never exceed 0.2% of the cells, even at late confluency, as already reported (Lesuffleur et al., 1990). Their mucins react either with anti-gastric or anti-colonic antibodies (Fig. 2). A few cells also express MUC1 (Fig. 2). HT29-MTX cells reach confluency on day 8. A gel becomes visible on the surface of the cell layer after 14 days of culture. This mucous gel becomes more conspicuous in the following days and continues to be present during all the complete period of culture (up to 60 days). This growth-related process is best visualized by indirect immunofluorescence of the mucus present on the cell surface at different times in culture (Fig. 3), as well as by histological staining of cryostat sections of the cell layer, which shows that the proportion of cells that express mucus increases progressively with the age of the culture, reaching 100% after 21 days and remaining stable thereafter (Fig. 4). The pattern of indirect immunofluorescence staining Fig. 1. Transmission electron microscopy of postconfluent cultures (day 28) of (a) HT29-MTX cells (bar, 6.25 mm) and (b) HT29-FU cells (bar, 7.7 mm). Sections are perpendicular to the bottom of the flask (arrows). 774 T. Lesuffleur and others Fig. 2. Indirect immunofluorescence staining of paraformaldehyde-fixed cryostat sections of late post-confluent cultures (day 28) of parental HT-29 cells with (a) polyclonal antiserum L-53 against colonic mucins; (b) polyclonal antiserum L-56 against gastric mucins; (c) mAb BC2 against MUC1 peptide. The same results as shown in a and b were obtained with mAb ZE4 and mAb against M1 antigen, respectively (not shown). Note that very few cells express immunoreactive material. Bar, 100 µm. Fig. 3. Detection by indirect immunofluorescence, with the polyclonal antiserum L-56, of mucus secreted on the cell surface of cultures of HT29-MTX cells in relation to cell growth; a, b, c, d, e and f correspond to days 7, 10, 12, 14, 21 and 28, respectively. Assays were performed on unpermeabilized cultures fixed with ethanol. In order not to remove the mucous gel, the medium was aspirated gently from the flask, which was kept horizontal, and ethanol was added without prior rinsing of the cell layer. The same results were obtained with mAbs against M1 antigens. The small mucin granules present in all the pictures do not correspond to intracellular mucins but to small clusters of mucins present on the cell layer surface as demonstrated by scanning electron microscopy or by their disappearance after extensive rinsing of the cell layer prior to fixation (not shown). Bar, 100 µm. Mucin gene expression in HT-29 cells with anti-gastric mucin antibodies exactly corresponds to the Alcian blue staining. These mucins appear to carry terminal oligosaccharides with the structure NeuAcα2- 775 3Galβ-, as indicated by double immunofluorescence labeling with Maackia amurensis lectin (Fig. 5 and Table 1). A high proportion of the mucin droplets also express termi- Fig. 4. Alcian blue staining of ethanol-fixed cryostat sections of the cell layer of HT29-MTX cells in relation to cell growth. The numbers indicate the day after seeding. Unlike the preparation shown in Fig. 3, the cell layer was rinsed with PBS prior to harvesting and freezing, so that the mucus stained corresponds to the intracellular mucus, as in Fig. 1a. The same pattern of mucus expression was observed by PAS staining or indirect immunofluorescence with mAbs against M1 antigens or polyclonal antiserum L-56 (not shown). Bar, 100 µm. 776 T. Lesuffleur and others Fig. 5. Double immunofluorescence labeling of an ethanol-fixed frozen cryostat section of a late post-confluent culture (day 28) of HT29MTX cells with (a) polyclonal antibodies L-56 and fluorescein-labelled globulins; and (b) Maackia amurensis labelled with rhodamine. Bar, 40 µm. nal αGalNAc and sialyl-Le a (Table 1). They do not express blood group A, H, Lea and Le b antigens (Table 1). At late confluency (after 21 days) a small proportion of mucin droplets (<2%) also reacts with anti-colonic antibodies (not shown). In contrast to the mucous secretions, MUC1 is already expressed in 50% of the cell population in preconfluent cultures, with this proportion increasing to the entire cell layer after 14 days and remaining stable thereafter (Fig. 6). MUC1 is not colocalized with the mucous droplets, but is associated with the apical brush border, as substantiated by its colocalization with DPP-IV or villin (Fig. 7). The same growth-related expression of mucus secretion is also observed in HT29-FU cells,which reach confluency on day 9. In this population, however, the proportion of mucus-secreting cells never exceeds 20%, even in late postconfluent cultures (Fig. 8). These mucins react with polyclonal and monoclonal antibodies against colonic mucins (L-53 and ZE4) and appear to express unfucosylated linear A antigen (Fig. 9) as deduced from the analysis of their immunoreactivity to the panel of anti-blood group A antigens tested, as well as NeuAcα2-3Gal, as indicated by their reactivity to Maackia amurensis (Table 1). A proportion of these mucins also express Lea (Fig. 9), Tn, sialyl-Tn, αGalNAc and sialyl-Lea (Table 1). As with HT29-MTX Table 1. Immunoreactivity of mucins from HT29-M and HT29-FU cells to antibodies and lectins that recognise the ABH and Lewis blood group structures Approximate % of labelled mucin droplets mAbs and lectins 1, 2, 5, 6, 12, 14, 16, 19, 20* 13* 9* 48, 49, 50* 26, 28* 7Le 2.25Le 12.LE Cu-1 B72-3 G14 K21 H15 Ulex europaeus Maackia amurensis Sambucca nigra Specificity ALey, HT29-M HT-29-FU 0 100 0 0 0 0 0 0 5 10 80 0 60 0 100 0 100 100 0 0 0 50 0 0 50 50 100 0 50 0 100 0 ALeb, A type1, 2, 3, 4, 5, 6, unfucosylated linear A, Forssman A type2, 6, ALey, unfucosylated linear A A type 3, 4, 5 A type 2, ALey, B B Le a Le b Le y Tn Sialyl-Tn Terminal αGalNAc Precursor type 1 Sialyl Le a H NeuAcα2-3Galβ-R NeuAcα2-6Galβ-R The results were obtained by immunofluorescence double-labelling assays with anti-gastric and anti-colonic mucin antibodies. *The numbers correspond to the mAbs listed by Oriol et al. (1990). Mucin gene expression in HT-29 cells 777 Fig. 6. Immunofluorescence detection of MUC1 peptide with mAb BC2 in paraformaldehyde-fixed cryostat sections of cultures of HT29MTX cells in relation to cell growth; a, b and c correspond to days 7, 14 and 21, respectively, and are from the same culture as shown in Fig. 3. Note that MUC1 is already expressed at day 7, i.e. when no mucins can be detected by Alcian blue staining. Bar, 100 µm. Fig. 7. Phase-contrast microscopy and double immunofluorescence staining of paraformaldehyde-fixed cryostat sections of late postconfluent cultures (day 21) of HT29-MTX cells; (a), (b) and (c), the same section labelled with: (b) the polyclonal antiserum L-56 against gastric mucins; and (c) the mAb BC2 against MUC1 peptide; (d), (e) and (f), same section labelled with: (e) the polyclonal antiserum against DPP-IV; and (f) the mAb BC2. Note the difference in distribution of MUC1 and mucins and the apical colocalization of MUC1 and DPP-IV. The same results as in (c) and (f) were obtained with the mAbs BC1, BC3 and LICR LON M8 (not shown). The same results as in (e) were obtained with anti-villin antibodies (not shown). Bar, 40 µm. 778 T. Lesuffleur and others Fig. 8. Alcian blue staining of ethanol-fixed cryostat sections of HT29-FU cells in relation to cell growth. The numbers indicate the days after seeding. The same pattern of mucin expression was observed with PAS staining or indirect immunofluorescence with the mAb ZE4 or the polyclonal antiserum L-53 against colonic mucins (not shown). Bar, 100 µm. cells, MUC1 is already expressed at the apical surface of a high proportion of cells in preconfluent cultures (day 7, not shown), i.e. when mucous secretions are not yet detectable by histological or immunological techniques. The proportion of cells expressing MUC1 increases during the days following and is stable after day 14 (Fig. 9). Expression of MUC1, as demonstrated by double-immunofluorescence staining (not shown) is restricted to these polarized cells, organized into a monolayer, which form the majority of the HT29-FU population. Differential and growth-related expression of mucin genes mRNAs Days 7, 14 and 21 were chosen for a comparative analysis of the expression of the apomucin mRNAs in the three cell populations, since these represent stages at which no, little or extensive, mucus is produced by the cultures of MTX and FU cells. As shown in Fig. 10, the presence and the levels of expression of the MUC mRNAs vary with the time in culture and the phenotype of the cells. Two transcripts of 6.5 kb and 4 kb of MUC1, presumably corresponding to the two MUC1 alleles, are expressed in all three populations. They are already expressed at a high level in early cultures (day 7) in all populations. Except for HT-29 cells, which show a decrease of these transcripts with time, their levels remain high in the two differentiated populations. MUC2 mRNA is expressed at a low level in the parental HT-29 cells and its level decreases with time. In the HT29FU cells it is expressed at much higher levels, but also decreases with time. In the HT29-MTX cells the level of expression is low in early cultures but increases with time. The highest levels of MUC3 mRNA are found in the MTX cells. In these cells the transcript is already detectable at day 7; it greatly increases at day 14, remaining stable thereafter. It is poorly expressed in HT-29 and not detectable in the FU cells. MUC4 mRNA is the most highly expressed in the HT29FU cells, with its level of expression decreasing from day 7 to day 21. It is poorly expressed in HT-29 and HT29MTX cells. The highest levels of MUC5C mRNA are found in the HT29-MTX cells, with the level increasing dramatically between 7 and 14 days of culture. A similar increase in expression with time was not detected in the other cell populations where the expression remains low. Because of the stable expression of mucins in the late post-confluent cultures of HT29-MTX cells and the expression of some mucins of colonic phenotype at very late confluency, poly(A)+ RNAs prepared from cultures of Mucin gene expression in HT-29 cells 779 Fig. 9. Immunofluorescence detection of mucins and the MUC1 peptide in cryostat sections of post-confluent cultures (day 21) of HT29FU cells. In (a) and (b), ethanol-fixed sections were double-labelled with: (a) the polyclonal antiserum L-53 against colonic mucins; and (b) the mAb 021 against blood group A antigen (see Table 1), showing that all the mucin droplets express both types of immunoreactive material. In (c) and (d), ethanol-fixed sections were double-labelled with: (c) the polyclonal antiserum L-53; and (d) the mAb 7LE against Lea antigen showing that only half of the mucin droplets express Lea. In (e) paraformaldehyde-fixed sections labelled with mAb BC2, showing an apical expression of MUC1. Bar, 100 µm. HT29-MTX cells harvested over a 60-day period were hybridized with MUC2, MUC3 and MUC5C cDNAs. As shown in Fig. 11, the levels of MUC3 and MUC5C mRNAs are stable from day 21 onwards. Interestingly, the level of MUC2 mRNA increases in late cultures. DISCUSSION Although a number of mucus-secreting HT-29 clones or sublines have been isolated and used as in vitro models for the analysis of functions associated with goblet cells, little is known about the kinetics of differentiation of these cells or which genes encode the mucins they secrete. Most studies so far have been restricted to the ultrastructural or histochemical analysis of cells at late confluency (Augeron and Laboisse, 1984; Huet et al., 1987; Hafez et al., 1990; Lesuffleur et al., 1990, 1991a,c; Kreusel et al., 1991), or to the characterization of the oligosaccharide structures associated with their mucins (Phillips et al., 1988; Capon et al., 1992). The differences in the oligosaccharide structures observed by Phillips et al. and Capon et al. would suggest, notwithstanding the different approaches used (immuno- logical or biochemical) that the clones or subpopulations analysed are different from one laboratory to another. This is not surprising in view of the heterogeneity of the parental cell line. Even less is known about which MUC genes are expressed in these cells. Ogata et al. (1992) did not detect any MUC1 to MUC4 mRNA in Northern blots of total RNA from parental HT-29 cells. In a study on preconfluent cultures Dahyia et al. (1992) observed, by slot blot analysis of total RNA, an increased expression of MUC1, but not of MUC2 and MUC3, in MTX cells in comparison with the parental HT-29 cells. The MUC1, but not MUC2 glycoprotein, was detected, by Western blot analysis, in the mucins extracted from the gel secreted by a population of mucus-secreting cells derived from sodium butyrate-treated HT-29 cells (Devine et al., 1991). The results reported here describe the growth-related appearance of mucins, the corresponding time-course of expression of the mucin gene mRNAs and the relationship between the type of mucins secreted by the two different populations of mucus-secreting HT-29 cells analysed and the type of mucin genes that they express. The appearance of mucus in both MTX and FU cells, as disclosed by histological and immunohistological methods, 780 T. Lesuffleur and others MUC1 -actin MUC2 MUC3 MUC4 MUC5C -actin Fig. 10. Northern blot hybridization of MUC1, MUC2, MUC3, MUC4 and MUC5C cDNAs with poly(A)+ RNAs from cultures of the parental HT-29 (P), HT29-FU (F) and HT29-MTX (M) cells in relation to cell growth; (a), (b) and (c) correspond to days 7, 14 and 21, respectively. The results presented for MUC2, MUC3, MUC4 and MUC5C represent successive hybridisation of a single filter. The MUC1 results shown correspond to another filter. Similar results were obtained from three different Northern blots corresponding to experiments with cells at different passages. Fig. 11. Growth-related expression of MUC2, MUC3 and MUC5C mRNAs in long-term cultures of HT29-MTX cells; cDNAs were hybridized with poly(A)+ RNAs using the same filter for all hybridizations; a, b, c, d, e, f and g correspond to days 7, 14, 21, 28, 40, 50 and 60, respectively. appears to be a growth-related process that starts at confluency and progressively involves an increasing proportion of cells. This process differs in its time-course with each of the mucins examined, the epithelial mucin MUC1 being detected much earlier than the mucous droplets. It is also associated with quantitative and qualitative differences in MUC gene expression. In HT29-MTX and FU cells the MUC1 protein is expressed on the apical surface of the cells, like in other systems of polarized cells (Ormerod et al., 1981; Corcoran and Walker, 1990), and its localization is distinct from that of the mucous droplets as disclosed by immunofluorescence. In contrast to the late appearance of mucous droplets in these cells, the MUC1 protein is already present in preconfluent cultures. This high expression of the MUC1 protein is consistent with the observation that the levels of MUC1 mRNA are high and almost constant during cell growth in the two populations. In the parental line, however, unlike in the differentiated populations, the level of MUC1 mRNA is high in early cultures, but decreases with the time in culture. What mechanism is responsible for the decrease in the steady-state level of MUC1 mRNA in these cells remains to be elucidated. The expression of MUC2, MUC3, MUC4 and MUC5C mRNAs differs from one population to another and, within each population, according to the stage of the culture. MUC2 and MUC4 are expressed at the highest levels in FU cells, the cells that express mucins of colonic Mucin gene expression in HT-29 cells immunoreactivity, whereas the highest levels of MUC3 and MUC5C are found in the MTX cells, the cells that express mucins of gastric immunoreactivity. It has been postulated that the type of mucin peptide may determine the type of carbohydrate present in the final mature mucin (Gum et al., 1989). Although the epitopes recognized by the polyclonal and monoclonal antibodies against gastric and colonic mucins used in this study have not been identified, it is clear that mucins from MTX and FU cells carry different oligosaccharide structures. There is indirect evidence that the oligosaccharide species associated with the mucins from MTX cells are the same as those that have been characterized on the mucins produced by clone 16E (Capon et al., 1992): they react with the lectin from Maackia amurensis, which binds to the major oligosaccharide species found in Cl.16E mucins, and we have previously shown that mucins from Cl.16E immunoreact with the polyclonal antibodies against gastric mucins (L-56) used in this study (Lesuffleur et al., 1991b). Although further compositional studies are required to determine precisely which oligosaccharide structures are associated with the mucins from each population, it is clear that MUC2 and MUC4 are expressed concomitantly with the mucins of colonic-like phenotype (from FU cells), while MUC3 and MUC5C are expressed concomitantly with the mucins of gastric-like phenotype (from MTX cells). These results are interesting in view of recent studies using in situ hybridization (Audié et al., 1993) and apomucin-specific antibodies (Gambus et al., 1993; Ho et al., 1993; and unpublished observations) which indicate that high levels of MUC2 are expressed in the colon, but not in the stomach, and the reciprocal pattern is observed with MUC5C. It is noteworthy that the expression of the MUC mRNAs occurs earlier than the production of mucus (detected histologically) in the differentiated cells. This discrepancy would suggest that the process of mucin biosynthesis in these cells involves a growth-related time-lag between the activation of the mucin genes, which are responsible for the biosynthesis of the mucin peptides, and the onset of their glycosylation. It would therefore be of interest to detect the newly synthesized apomucins in early cultures. However, using the monoclonal antibody LDQ10 (Gambus et al., 1993) against MUC2 peptide, or polyclonal antibody M3P against MUC3 repeat peptide (Gum et al., 1990) we were unable to detect either of these these peptide antigens by immunofluorescence, in 7-day-old cultures of FU and MTX cells, respectively (unpublished results). This will be further investigated by electron microscopy and immunogold labeling, since Egea et al. (1993) have successfully demonstrated MUC2 in the RER in post-confluent MTX cells by this method, even though it is undetectable by immunofluorescence. It is also possible and likely that other mucin genes are expressed in these cells, such as the human mucin gene MUC6, which has been identified recently (Toribara et al., 1993), or other unidentified mucin genes. Thus, notwithstanding the fact that they are malignant, and may show abnormal apomucin synthesis and glycosylation, these two different populations of mucus-secreting cells isolated from the HT-29 cell line appear to be relevant in vitro models for studying the sequence of regula- 781 tory events that are associated with the onset and maintenance of the biosynthesis of mucins, and the relationship between the type of mucin gene expressed and the type of glycosylation of the mature mucin. This work was supported in part by Association pour la Recherche sur le Cancer, Fondation pour la Recherche Médicale Française, INSERM-CSIC Cooperation Agreement, grant SAL900853 from the Plan Nacional de I+D (Ministerio de Educacion, Madrid), the Veterans Affairs Medical Research Service, and NATO grant 0789/88. Part of this work has been presented at the 2nd International Workshop on Carcinoma-Associated Mucins (Cambridge, August 1992). REFERENCES Aubert, J. P., Porchet, N., Crepin, M., Duterque-Coquillaud, M., Vergnes, G., Mazzuca, M., Debuire, B., Petitprez, D. and Degand, P. (1991). Evidence for different human tracheobronchial mucin peptides deduced from nucleotide cDNA sequences. Amer. J. Resp. Cell. Mol. Biol. 5, 178-185. Audié, J. P., Janin, A., Porchet, N., Copin, M. C., Gosselin, B. and Aubert, J. P. (1993). Expression of human mucin genes in respiratory, digestive and reproductive tracts ascertained by in situ hybridization. J. Histochem. Cytochem. (in press). Augeron, C. and Laboisse, C. (1984). Emergence of permanently differentiated cell clones in a human colonic cancer cell line in culture after treatment with sodium butyrate. Cancer Res. 44, 3961-3969. Bara, J., Daher, N., Mollicone, R. and Oriol, R. (1987). Immunohistological pattern of 20 monoclonal antibodies against non-A non-B glycoconjugates in normal human pyloric and duodenal mucosae. Blood Transf. Immunohaematol. 30, 685-692. Bara, J., Gautier, R., Daher, N., Zaghouani, H. and Burtin, P. (1986). Monoclonal antibodies against oncofetal mucin M1 antigens associated with precancerous colonic mucosae. Cancer Res. 46, 3983-3989. Bara, J., Mollicone, R., Herrero-Zabeleta, E., Gautier, R., Daher, N. and Oriol, R. (1988). Ectopic expression of the Y (Ley) antigen defined by monoclonal antibody 12-4LE in distal colonic adenocarcinomas. Int. J. Cancer 41, 683-689. Capon, C., Laboisse, C. L., Wieruszeki, J. M., Maoret, J. J., Augeron, C. and Fournet, B. (1992). Oligosaccharide structures of mucins secreted by the human colonic cancer cell line CL. 16E. J. Biol. Chem. 267, 1924819257. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J. and Rutter, W. J. (1979). Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18, 5294-5299. Cleveland, D. W., Lopata, M. A., McDonald, R. J., Cowan, N. J., Rutter, W. J. and Kirschner, M. W. (1986). Number and evolutionnary conservation of α- and β-tubulin and cytoplasmic β- and α-actine genes using specific cloned cDNA probes. Cell 20, 95-105. Corcoran, D. and Walker, R. A. (1990). Ultrastructural localization of milk fat globule membrane antigens in human breast carcinomas. J. Pathol. 161, 161-166. Crepin, M., Porchet, N., Aubert, J. P. and Degand, P. (1990). Diversity of the peptide moiety of human airway mucins. Biorheology 27, 471-484. Dahiya, R., Lesuffleur, T., Kwak, K. S., Byrd, J. C., Barbat, A., Zweibaum, A. and Kim, Y. S. (1992). Expression and characterization of mucins associated with the resistance to methotrexate of HT-29 human colonic adenocarcinoma cell line. Cancer Res. 52, 4655-4662. Darmoul, D., Rouyer-Fessard, C., Blais, A., Voisin, T., Sapin, C., Baricault, L., Couvineau, A., Laburthe, M. and Trugnan, G. (1991). Dipeptidylpeptidase IV expression in rat jejunal crypt-villus axis is controlled at mRNA level. Amer. J. Physiol. 261 (Gastrointest. Liver Physiol. 24), G763-G769. Davis, L. G., Dibner, M. D. and Battey, J. F. (1986). Basic Methods in Molecular Biology. Elsevier Science, New York. Devine, P. L., Layton, G. T., Clark, B. A., Birrell, G. W., Ward, B. G., Xing, P. and McKenzie, I. F. C. (1991). Production of MUC1 and MUC2 mucins by human tumor cell lines. Biochem. Biophys. Res. Commun. 178, 593-599. 782 T. Lesuffleur and others Dippold, W. G., Klingel, R., Bernhard, H., Dienes, H. P., Knuth, A. and Meyer Zum Buschenfelde, K. H. (1987). Secretory epithelial cell marker on gastrointestinal tumors and in human secretions defined by a monoclonal antibody. Cancer Res. 47, 2092-2097. Dufosse, J., Porchet, N., Audie, J. P., Guyonnet-Duperat, V., Laine, A., Van-Seuningen, I., Marrakchi, S., Degand, P. and Aubert, J. P. (1993). Degenerate 87 base pair tandem repeats create hydrophilic/hydrophobic alternating domains in human mucin peptides mapped to 11p15. Biochem. J. (in press). Egea, G., Franci, C., Gambus, G., Lesuffleur, T., Zweibaum, A and Real, F. X. (1993). cis-Golgi resident proteins and O-glycans are abnormally compartmentalized in the RER of colon cancer cells. J. Cell Sci. (in press). Fogh, J. and Trempe, G. (1975). New human tumor cell lines. In Human Tumor Cells In vitro (ed. J. Fogh), pp. 115-141. Plenum Press, New York. Fox, M. F., Lahbib, F. Pratt, W., Attwood, J., Gum, J., Kim, Y. and Swallow, D. M. (1992). Regional localization of the intestinal mucin MUC3 to chromosome 7q22. Ann. Hum. Genet. 56, 281-287. Gambus, G., de Bolos, C., Andreu, D., Franci, C., Egea, G. and Real, F. X. (1993). Detection of a peptide epitope of the MUC2 gene product with a mouse monoclonal antibody. Gastroenterology 104, 93-102. Gendler, S. J., Burchell, J. M., Duhig, T., Lamport, D., White, R., Parker, M. and Taylor-Papidimitriou, J. (1987). Cloning of partial cDNA encoding differentiation and tumor-associated mucin glycoproteins expressed by human mammary epithelium. Proc. Nat. Acad. Sci. USA 84, 6060-6064. Gendler, S. J., Lancaster, C. A., Taylor-Papadimitrou, J., Duhig, T., Peat, N., Burchell, J., Pemberton, L., Lalani, E. and Wilson, D. (1990). Molecular cloning and expression of the human tumourassociated polymorphic epithelial mucin, PEM. J. Biochem. 265, 1528615293. Gendler, S. J., Taylor-Papadimitriou, J., Duhig, T., Rothbard, J. and Burchell, J. (1988). A highly immunogenic region of a human polymorphic epithelial mucin expressed by carcinomas is made up of tandem repeats. J. Biol. Chem. 263, 12820-12823. Gerard, C., Eddy, R. L., Shows, T. B. (1990). The core polypeptide of cystic fibrosis tracheal mucin contains a tandem repeat structure. Evidence for a common mucin in airway and gastrointestinal tissue. J. Clin. Invest. 86, 1921-1927. Griffiths, B., Mathews, D. J., West, L., Attwood, J., Povey, S., Swallow, D. M., Gum, J. R. and Kim, Y. S. (1990). Assignment of the polymorphic intestinal mucin gene (MUC2) to chromosome 11p15. Ann. Hum. Gen. 54, 277-285. Gross, M. S., Guyonnet-Duperat, V., Porchet, N., Bernheim, A., Aubert, J. P. and Nguyen, V. C. (1992). Mucin 4 (MUC4) gene: regional assignment (3q29) and RFLP analysis. Ann. Genet. 35, 21-26. Gum, J. R., Byrd, J. C., Hicks, J. W., Toribara, N. W., Lamport, D. T. A. and Kim, Y. S. (1989). Molecular cloning of human intestinal mucin cDNAs. J. Biol. Chem. 264, 6480-6487. Gum, J. R., Hicks, J. W., Swallow, D. M., Lagace, R. L., Byrd, J. C., Lamport, D. T. A., Siddiki, B. and Kim, Y. S. (1990). Molecular cloning of cDNAs derived from a novel human intestinal mucin gene. Biochem. Biophys. Res. Commun. 171, 407-415. Gum, J. R., Hicks, J. W., Toribara, N. W., Rothe, E. M., Lagace, R. E. and Kim, Y. S. (1992). The human MUC2 intestinal mucin has cysteinerich subdomains located both upstream and downstream of its central repetitive region. J. Biol. Chem. 267, 21375-21383. Hafez, M. M., Infante, D., Winawer, S. and Friedman, E. (1990). Transforming growth factor β1 acts as an autocrine-negative growth regulator in colon enterocytic differentiation but not in goblet cell maturation. Cell Growth Differ. 1, 617-626. Hauri, H. P., Sterchi, E. E., Bienz, D., Fransen, J. A. M. and Marxer, A. (1985). Expression and intracellular transport of microvillus membrane hydrolases in human intestinal epithelial cells. J. Cell Biol. 101, 838-851. Ho, S. B., Niehans, G. A., Lyftogt, C., Yan, P. S., Cherwitz, D. L., Gum, E. T., Dahyia, R. and Kim, Y. S. (1993). Heterogeneity of mucin gene expression in normal and neoplastic tissues. Cancer Res. 53, 641-651. Huet, C., Sahuquillo-Mérino, C., Coudrier, E. and Louvard, D. (1987). Absorptive and mucus-secreting subclones isolated from a multipotentent intestinal cell line (HT-29) provide new models for cell polarity and terminal differentiation. J. Cell Biol. 105, 345-358. Jany, B. H., Gallup, M. W., Yan, P. S., Gum, J. R., Kim, Y. S. and Basbaum, C. B. (1991). Human bronchus and intestine express the same mucin gene. J. Clin. Invest. 87, 77-82. Kreusel, K. M., Fromm, M., Schulze, J. D. and Hegel, U. (1991). Cl − secretion in epithelial monolayers of mucus-forming human colon cells (HT-29/B6). Amer. J. Physiol. 261 (Cell Physiol. 30), C574-C582. Lan, M. S., Batra, S. K., Qi, W., Metzgard, R. S. and Hollingsworth, M. A. (1990). Cloning and sequencing of a human pancreatic tumor mucin cDNA. J. Biol. Chem. 265, 15294-15299. Lesuffleur, T., Barbat, A., Dussaulx, E. and Zweibaum, A. (1990). Growth adaptation to methotrexate of HT-29 human colon carcinoma cells is associated with their ability to differentiate into columnar absorptive and mucus-secreting cells. Cancer Res. 50, 6334-6343. Lesuffleur, T., Barbat, A., Luccioni, C., Beaumatin, J., Claire, M., Kornowski, A., Dussaulx, E., Dutrillaux, B. and Zweibaum, A. (1991a). Dihydrofolate reductase gene amplification-associated shift of differentiation in methotrexate-adapted HT-29 cells. J. Cell. Biol. 115, 1409-1418. Lesuffleur, T., Kornowski, A., Augeron, C., Dussaulx, E., Barbat, A., Laboisse, C. and Zweibaum, A. (1991b). Increased growth adaptability to 5-fluorouracil and methotrexate of HT-29 sub-populations selected for their commitment to differentiation. Int. J. Cancer 49, 731-737. Lesuffleur, T., Kornowski, A., Luccioni, C., Muleris, M., Barbat, A., Beaumatin, J., Dussaulx, E., Dutrillaux, B. and Zweibaum, A. (1991c). Adaptation to 5-fluorouracil of the heterogeneous human colon tumor cell line HT-29 results in the selection of cells committed to differentiation. Int. J. Cancer 49, 721-730. Mc. Ilhinney, R. A. J., Patel, S. and Gore, M. E. (1985). Monoclonal antibodies recognizing epitopes carried on both glycolipids and glycoproteins of the human fat globule membrane. Biochem. J. 227, 155162. Nguyen, Van Cong, Aubert, J. P., Gross, M. S., Porchet, N., Degand, P. and Frezal, J. (1990). Assignment of human tracheobronchial mucin gene(s) to 11p15 and a tracheobronchial mucin-related sequence to chromosome 13. Hum. Genet. 86, 167-172. Nuti, M., Teramoto, Y. A., Mariani-Constantini, R., Horan Hand, P., Colcher, D. and Schlom, J. A. (1982). A monoclonal antibody (B72. 3) defines patterns of distribution of a novel tumor-associated antigen in human mammary carcinoma cell populations. Int. J. Cancer29, 539-545. Ogata, S., Uehara, H., Chen, A. and Itzkowitz, S. H. (1992). Mucin gene expression in colonic tissues and cell lines. Cancer Res. 52, 5971-5978. Oriol, R., Samuelsson, B. E. and Messeter, L. (1990). ABO antibodies: serological behaviour and immuno-chemical characterization. J. Immunogenet. 17, 279-299. Ormerod, M. G., Monoghan, P., Easty, D. and Easty, G. C. (1981). Asymmetrical distribution of epithelial membrane antigen on the plasma membranes of human breast cell lines in culture. Diagn. Histopathol. 4, 89-93. Phillips, T. E., Huet, C., Bilbo, P. R., Podolsky, D. K., Louvard, D. and Neutra, M. (1988). Human intestinal goblet cells in monolayer culture: characterization of a mucus-secreting subclone derived from the HT29 colon adenocarcinoma cell line. Gastroenterology 94, 1390-1403. Podolsky, D. K. (1985a). Oligosaccharide structures of human colonic mucin. J. Biol. Chem. 260, 8262-8271. Podolsky, D. K. (1985b). Oligosaccharide structures of isolated human colonic mucin species. J. Biol. Chem. 260, 15510-15515. Podolsky, D. K., Fournier, D. A. and Lynch, K. E. (1986). Human colonic goblet cells. Demonstration of distinct subpopulations defined by mucinspecific monoclonal antibodies. J. Clin. Invest. 77, 1263-1271. Porchet, N., Nguyen, V. C., Dufosse, J., Audie, J. P., Guyonnet-Duperat, V., Gross, M. S., Denis, C., Degand, P., Bernheim, A. and Aubert, J. P. (1991). Molecular cloning and chromosomal localization of a novel human tracheobronchial mucin cDNA containing tandemly repeated sequences of 48 base pairs. Biochem. Biophys. Res. Commun. 175, 414422. Rettig, W. J., Cordon-Cardo, C., No, J. S. C., Oettgen, H. F., Old, L. J. and Lloyd, K. O. (1985). High molecular weight glycoproteins of human teratocarcinoma defined by monoclonal antibodies to carbohydrate determinants. Cancer Res. 45, 815-821. Robine, S., Huet, C., Moll, R., Sahuquillo-Merino, C., Coudrier, E., Zweibaum, A. and Louvard, D. (1985). Can villin be used to identify malignant and undifferentiated normal digestive epithelial cells? Proc. Nat. Acad. Sci. USA 82, 8488-8492. Swallow, D. M., Gendler, S., Griffiths, B., Corney, G., TaylorPapadimitriou, J. and Bramwell, M. E. (1987). The human tumourassociated epithelial mucins are coded by an hypervariable gene locus PUM. Nature 328, 82-84. Mucin gene expression in HT-29 cells Takahashi, H. K., Metoki, R. and Hakomori, S. (1988). Immunoglobulin G3 monoclonal antibody directed to Tn antigen (tumor-associated α-Nacetylgalactosaminyl epitope) that does not cross-react with blood group A antigen. Cancer Res. 48, 4361-4367. Thomas, P. S. (1980). Hybridization of denaturel mRNA and small DNA fragments transferred to nitrocellulose. Proc. Natl. Acad. Sci. USA 77, 5201-5205. Toribara, N. W., Gum, J. R., Culhane, P. J., Lagace, R. E., Hicks, J. W., Petersen, G. M. and Kim, Y. S. (1991). MUC2 human small intestinal mucin gene structure. J. Clin. Invest. 88, 1005-1013. Toribara, N. W., Roberton, A. M., Ho, S. B., Kuo, W. L., Gum, E., Hicks, J. W., Gum, J. R., Byrd, J. C., Siddiki, B. and Kim, Y. S. (1993). Human gastric mucin. Identification of a unique species by expression cloning. J. Biol. Chem. 268, 5879-5885. Wesley, A., Mantle, M., Man, D., Qureshi, R., Forstner, G. and Forstner, J. (1985). Neutral and acidic species of human intestinal mucin. J. Biol. Chem. 260, 7955-7959. 783 Xing, P. X., Tjandra, J. J., Reynolds, K., McLaughlin, P. J., Purcell, D. F. J. and McKenzie, I. (1989). Reactivity of anti-human milk fat globule antibodies with synthetic peptides. J. Immunol. 142, 3503-3509. Yonezawa, S., Byrd, J. C., Dahiya, R., Ho, J. J. L., Gum, J. J., Griffiths, B., Swallow, D. M. and Kim, Y. S. (1991). Differential mucin gene expression in human pancreatic and colon cancer cells. Biochem. J. 276, 599-605. Zweibaum, A., Laburthe, M., Grasset, E. and Louvard, D. (1991). Use of cultured cell lines in studies of intestinal cell differentiation and funciton. In Intestinal Absorption and Secretion. Handbook of Physiology, section 6, The Gastrointestinal System, vol. IV (ed. M. Field and R. A. Frizzell), pp. 223-255. Amer. Physiol. Soc. Bethesda, MD. (Received 27 May 1993 - Accepted 27 July 1993)