Human colonocytes in primary culture: a model to study epithelial growth, metabolism and differentiation

Human colonocytes in primary culture: a model to study epithelial growth, metabolism and differentiation
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Int J Colorect Dis (1994) 9:13 22 Col6reeial Disease 9 Springer-Verlag 1994 Human colonocytes in primary culture: a model to study epithelial growth metabolism and differentiation R. Fonti 1 G. LateHa 1 G. Bises 2 F. Magliocca 3 E Nobili 2 R. Caprilli 1 y. Sambuy 2 1 Cattedra di Gastroenterologia, Universit~i de L'Aquila, L'Aquila, Italy 2 Istituto Nazionale della Nutrizione (INN), Roma, Italy 3 Dipartimento di Biopatologia Umana, Universit/t "La Sapienza", Roma, Italy Accepted: October 1993 Abstract. The purpose of this work was to set up an in vitro model for the study of normal and pathological functions of the colonic epithelium. We have isolated colonic crypts by mild proteolytic digestion and mechan- ical dissociation of human biopsy material obtained dur- ing colonoscopy. The crypts, free of connective tissue, when placed in culture rapidly attached to the substrate and formed colonies containing over 95% of epithelial cells. Histochemical and ultrastructural characterization of the colonies showed the presence of both absorptive and secretory cells, exhibiting a high degree of differenti- ation. Proliferative activity occurred mostly during the first 24 h and progressively declined thereafter. The cells survived and maintained differentiated characteristics for at least three days in culture. This method can be used to study normal functions of the colonic epithelium. It may also be employed to investigate both noxious and protec- tive factors in pathological conditions such as inflamma- tory bowel disease and colorectal neoplasia. R~sum& Le but de ce travail a &6 d'6tablir un mod61e in vitro permettant d'6tudier les fonctions normales et pathologiques de l'~pith61ium colique. Nous avons isol~ des cryptes coliques par une digestion prot6olytique douce et par une dissection m6canique de mat6riel biop- sique pr61ev6 au cours de colonoscopies. Les cryptes d6- pourvues de tissu conjonctif, plac6es en milieu de culture, se fixent rapidement sur le substrat et forment des colonies qui renferment plus de 95 % de cellules 6pith6- liales. L'histochimie et la caract6risation ultrastructurelle des colonies ont montr6 la pr6sence de cellules d'absorbtion et de cellules secr6toires pr~sentant un haut degr~ de diff6rentiation. La proliferation active survient le plus souvent durant les premi6res 24 heures et diminue progressivement par la suite. Les cellules survivent et maintiennent les caract6ristiques de diff6rentiation durant au moins 3 jours en culture. Cette m6thode peut ~tre utilis6e pour 6tudier les fonctions normales de l'6pith61ium colique. Elles peuvent aussi atre utilis6es pour investiger les facteurs nocifs et protecteurs dans des situations pathologiques telles que les maladies inflam- matoires de l'intestin et les tumeurs colorectales. The epithelial cells lining the mucosa of the large bowel play a central role in absorptive and secretory functions and provides an effective barrier to the complex antigenic load of intracolonic contents. In addition, some of the most important pathological conditions of the large bow- el, such as neoplasia and inflammatory bowel disease (IBD), are known to be associated with alterations of the normal growth and functions of the colonic epithelium [11, 4, 151. Most studies on the normal physiology of the colonic epithelium have been performed on whole organ or mu- cosal preparations. The use of these approaches is limited both by rapid necrosis and degeneration of the epitheli- um and by the impossibility of distinguishing the func- tions of the different mucosal cytotypes. A model system of human colonocytes in culture would therefore be very valuable to study normal function and to define the met- abolic and growth alterations in colorectal neoplasia and IBD. Much effort has been devoted in recent years to the culture of human colonic epithelium and several tech- niques have been described, including organ cultures and primary or long-term cultures. Each, however has both advantages and limitations [1, 3, 5 7, 19, 27]. in general, better cell preservation and longer survival times have been reported for cultures of human fetal colonic epithe- lium, but success in inducing their differentiation in vitro has been limited [3, 23]. Other studies performed in human cell lines derived from colorectal carcinomas, in particular Caco 2, HT29 and T84 [20] present obvious limitations because of the neoplastic srcin of the cells and are therefore unsuitable for studies on the regulatory factors that influence nor- mal colonocyte function. The aim of this study was to develop an in vitro model of normal human colonocytes able to maintain differenti- ated functions in culture for sufficient time to allow for metabolic and molecular investigation.  14 Materials and methods Specimens and isolation of the crypts. Biopsy specimens (approx. 3-5 mm) of normal colorectal mucosa were obtained during en- doscopy in patients with hemorroids, irritable bowel syndrome or constipation. The procedures followed were in accord with the eth- ical standards of the Helsinki Declaration of 1975. All biopsies were taken from the descending or sigmoid colon. The specimens were transported and extensively washed in Hank's solution containing 1.2 mM CaC12 and with 50 gg/ml gentamicin (Schering Plough, Comazzo, Milan, Italy), 200 gg/ml sodium cefoperazone (Cefobid, Pfizer Italy, Rome, Italy), 200 gg/ml sodium piperacillin (Avocin, Lederle, American Cyanamid Co., Wayne, NJ, USA), 5 gg/ml am- photericin B (Flow Laboratories Int., Opera, Milan, Italy), the suspension was repeatedly centrifuged at low speed (650 rpm, 5 min at room temperature) to remove mucosal debris, single cells and mucous remaining in the supernatant. The specimens were then incubated in the same medium in the presence of 0.4 U/ml collage- nase P (Boehringer Mannheim Italia, Milan, Italy) and 1.2 U/ml dispase I (Collaborative Research, Bedford, MA, USA) for 5 min at 37~ After digestion, the material was mechanically broken up with syringe needles to separate the crypts and washed again five to six times as described above. At each step of the separation proce- dure samples of material were fixed in Bouin solution [13] and embedded in paraffin for histological sectioning and staining. Tis- sue morphology was evaluated by Mallory trichromic staining [13] on paraffin sections. Establishment of the cultures. The isolated crypts were seeded in 100 gl of growth medium on round glass coverslips (12 mm diame- ter) in 24 well culture dishes. For transmission electron microscopy (TEM) the crypts were seeded either on plastic petri dishes or on Cyclopore membrane cell culture inserts (0.45 gM pore diameter; Falcon, Becton Dickinson Labware, Lincoln Park, N J, USA). The growth medium consisted of Dulbecco Modified Minimum Essen- tial Medium (DMEM) containing 3.7 g/1 NaHCO3, 1.8 mM CaC12, 10 mM glucose, 4 mM L-glutamine, 1% fetal calf serum (FCS), 2% Luria's broth [5 g/1 NaC1, 10 g/1 bactotryptone, 5 g/1 yeast extract (Difco Laboratories, Detroit, MI, USA)] 10 gg/ml insulin, 10 gg/ml human transferrin, 10 ng/ml sodium selenite (all from Sigma Chem- ical Co., St. Louis, MO, USA), 50 gg/ml gentamicin, 200 gg/ml cefoperazone, 200 gg/ml piperaciUin. The cultures were incubated at 37~ in an atmosphere of 5% CO2/95% air at 90% relative humidity. After 24 h, when the crypts had attached to the growth surface, 400 gl of complete growth medium were added to each well. Cellu- lar viability was assessed by dye exclusion using 0.04% trypan blue in Hank's solution, both on the crypts in suspension and on the cultures, at 24 h intervals. All cell culture plastics were from Falcon Products (Becton Dickinson Italia, Milan, Italy) and culture reagents were from Flow Labs., unless otherwise stated. Cytochemistry and immunocytochembtry. The morphology of the cells srcinating from the crypts was evaluated by May Gruenwald Giemsa staining after a brief (20 min) fixation in 2% paraformalde- hyde in PBS +. The mucin-containing cells were localized by the periodic acid-Schiff (PAS) reaction [13]. For immunocytochemical localization of cytoskeletal and junctional proteins the colonies where fixed after 24, 48 or 72 h from seeding with methanol at -20 ~ for 5 min or with 2% paraformaldehyde (PFA) in PBS + for 30 min at room temperature and treated by conventional im- munofluorescence techniques using tetramethylrhodamine isothio- cyanate (TRITC)-labelled secondary antibodies (affinity purified IgG; Cappel, Organon Tecknika Corporation, Durham, NC, USA). Monoclonal antibodies against cytokeratin 18, vimentin, desmo- plakins l&2 were from Boehringer Mannheim. The anti fl-tubulin antibody was from ICN Biomedicals Inc. (Costa Mesa, CA, USA). T-lymphocytes were localized on the cultures, fixed with 1% PFA in PBS + for 15 min, and in the supernatant, 24 h after seeding, using an anti-human CD3 monoclonal antibody conjugated with fluores- cein isothiocyanate (FITC) (Dako A/S, Glostrup, Denmark). The organization of F-actin was studied labelling the cells with FITC- phalloidin (0.23 gM - Sigma Chemical Co.) after fixation with 2% paraformaldehyde in PBS § and permeabilization with 0.075% sa- ponin. For immunocalization of intermediate filaments on samples treated for autoradiography, the Biotin-streptavidin-peroxidase-an- ti peroxidase (PAP) technique was employed. Briefly, methanol- fixed cultures were incubated for 5 rain with 3% H202 to inactivate endogenous peroxidase activity and for 20 min with 0.1 mg/ml streptavidin (Gibco BRL, Paisley, Scotland) in PBS + with 0.2% bovine serum albumin and 1% normal goat serum (both from Sigma Chemical Co.) to block endogenous biotin. The cultures were then treated with biotinylated anti-mouse immunoglobulins for 30 rain, followed by streptavidin-peroxidase for an additional 30 min (both reagents were from Amersham International, Amer- sham, Buckinghamshire, UK). The peroxidase activity was local- ized by treating with 3% H20 z and 3% 3-amino-9-ethyl-carbazole in dimethylformamide (Sigma Chemical Co.) in acetate buffer, pH 5.2. The coverslips were then mounted cell-side upward and treated with photographic emulsion as described below. 3H-thymidine incorporation and autoradiography. The proliferative activity of the cultures was evaluated by 3H-thymidine incorpora- tion followed by autoradiography. Methyl-3H-thymidine (NEN, Du Pont de Nemours, Florence, Italy - specific activity 2.48 x 10 2 GBq/mmol) 1.85 x 104 Bq/ml in complete growth medium was add- ed to the cultures for 24 h, starting at different times after seeding. At the end of the incubation the cultures were washed, fixed in methanol at - 20 ~ for 5 min; the coverslips were mounted cell-side upwards on glass slides using transparent glue. The slides were dipped in the dark in photographic emulsion (NTB 2; Eastman Kodak, Rochester, NY, USA) and exposed at 4 ~ for 4 days. After developing and fixing with photographic reagents the cell nuclei were counterstained with 0.25~tg/ml bisbenzimide (H 33258, Boehringer Mannheim) in McIlvaine's buffer pH 5.5 [2]. The preparations were observed under the microscope (Diaplan; Leitz Italiana, Milan, Italy) and photographed in direct light to visualize the silver grains and in fluorescence to count the total number of nuclei in each colony. Transmission and scanning electron microscopy. For ultrastructural analysis the cultured colonocytes were prepared both for transmis- sion electron microscopy (TEM) and for scanning electron micros- copy (SEM). For TEM the crypts were seeded and allowed to grow on plastic petri dishes or on transparent polycarbonate filters. After two days the colonies were fixed for I h in 2.5% glutaraldehyde in 0.1 M phosphate buffer at pH 7,4, rinsed with the same buffer, postfixed in 1% OsO 4 for 30 rain and counterstained in 2% uranyl acetate. The samples were dehydrated through a series of graded alcohols and embedded in Agar 100 resin (Agar Scientific Ltd., Stansted, Essex, UK). Plastic grown cells were embedded in resin using (2-hydroxypropyl)-methacrylate (Merck, Darmstadt, Ger- many) as solvent, in place of propylene oxide. Ultrathin sections were cut (Ultracut E, Reichert Jung Optische Werke, Vienna, Aus- tria), stained in uranyl acetate and lead citrate, and observed in the electron microscope (EM 10; Carl Zeiss, Oberkochen, Germany). For SEM the cells were grown on glass coverslips and fixed in 1% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4, containing 5% sucrose, washed, postfixed and dehydrated as for TEM. The cells were then dried with liquid CO 2 in a critical point drying system (Balzers AG, Liechtenstein), attached to aluminium stubs, coated with gold in a sputter coater (S150; Edwards High Vacuum Inc., Crawley, W Sussex, UK), and examined in a field emission scanning electron microscope operating at 7-10 kV (S-4000, Hitachi, Tokio, Japan). Results The conventional histology of the biopsy specimens showed a normal pattern characterized by the deeply in-  15 Fig. 1. Histological section of paraffin- embedded biopsy specimen from the descending colon (A) and of isolated colonic crypts after enzymatic digestion (B). Mallory trichromic staining. Phase contrast image of isolated crypts after digestion, mechanical dissociation and washing, ready to be placed in culture (C). Morphological appearance of a colony of colonocytes after 48 h in cul- ture stained with FITC-conjugated phalloidin to show F-actin distribution in the cells growing and migrating out of a crypt residue (D) Fig. 2A, B. Colony after 48 h in cul- ture. About one third of the eolono- cytes show production and intracellular accumulation of mucin granules (PAS reaction, which localizes mucopoly- saccharides), View of the whole colony of cells (A). At higher magnification, the cytoplasmic accumulation of PAS- reactive material is more evident (B)  16 Fig. 3A-D. Expression and organisa- tion of cytoskeletal and junctional proteins by indirect mmunofluores- cence with monoclonal antibodies on 48 h cultures. The colonies contain over 95% of cytokeratin 18-positive cells (A) and a regular and abundant microtubular network labelled by an anti-/? tubulin antibody (B). Actin fila- ments, labelled by fluorescent FITC- phalloidin, show a belt-like distribution in many cells, which is typical of ep- ithelial cells (C). Desmoplakins 1&2, peripheral components of the desmo- somes, are present in all cells and ap- pear distributed at the cell periphery (I)) vaginated crypts, composed of numerous secretory gob- let cells, opening on the lumen, lined by surface absorp- tive cells (Fig. 1A). After short enzymatic digestion of the specimen, followed by gentle mechanical dissociation, the crypts were recovered intact and in good yield, free of interstitial tissue (Figs. 1B and 1C). Few contaminating blood cells remained loosely associated with the crypts at this stage (Fig. 1B), but were rapidly lost from the culture after the crypts had attached to the growth surface and the growth medium was changed (after approximately 16-24 h). Attempts to localize T-lymphocytes by im- munoreactivity with human CD3 antibody only detected the presence of few reactive cells in the culture superna- tant after 24 h from seeding, and never among the cells attached on the culture substrate (data not shown). When placed in culture the crypts rapidly attached and gave rise to colonies of polygonal cells, arranged in the typical epithelial cobblestone pattern. Fragments of the crypts often remained visible on the monolayer of cells growing out of the explant (Fig. 1D). Cellular viability assessed by trypan blue exclusion was very high (>98%) over the entire isolation procedure and during the first 72 h of culture, although progressively more dead cells were ob- served at the edge of the cultures from the third day onwards (data not shown). The colonocyte cultures were composed of at least two cell types, one of which showed characteristics of the secretory goblet cells. When stained with the PAS reac- tion, which identifies mucopolysaccharides, approxi- mately one third of the cells in culture exhibited abundant accumulation of PAS-positive material (Fig. 2A). At higher magnification this material appeared concentrat- ed in the cytoplasm, almost completely filling it (Fig. 2B), suggesting the presence of secretory granules filled with mucins. This pattern ofmucin expression remained stable for the entire period of the culture (72 h), nor did the incidence of PAS-positive cells change during this time. In order to characterize the expression and the organi- zation of cytoskeletal proteins in these cells we have used immunofluorescent techniques with specific monoclonal  17 Fig. 4A-E General view by SEM of colonocytes in culture for 48 h. A crypt residue at the top of the picture and the monolayer of cells surrounding it (A). The secretory goblet cells are irreg- ularly disposed among the absorptive cells (B). At higher magnification, the difference between the two cytotypes is clear: goblet cells have surface invagi- nations and sparser microvilli, while the surrounding cells are polygonal in shape and have densely packed micro- villi (C, D). SEM of normal colonic mucosa shows a similar pattern to that observed in culture (E, F) antibodies. In the case of actin localization we have used fluoresceinated phalloidin which tightly and specifically binds to F-actin, labelling it. Figure 3A shows the organi- zation of the cytokeratin 18, an epithelial-specific inter- mediate filament expressed in gastrointestinal epithelia [13]. Over 95% of the cells in each colony showed strong reactivity for cytokeratin 18 and only occasionally were cytokeratin-negative, presumably vimentin-positive cells, observed among the epithelial cells (Fig. 3A). Micro- tubules stained with an antibody for //-tubulin were abundant and regularly distributed in all cells (Fig. 3B). Actin filaments labelled by fluorescent phalloidin, con- versely, showed a belt-like distribution in many cells, which is typical of the epithelial phenotype [18] (Fig. 3C). The distribution of F-actin at the periphery of the cells was similar to the localization pattern of the junctional proteins desmoplakin 1 &2, peripheral components of the desmosomes (Fig. 3D). The staining and localization of the cytoskeletal and the junctional proteins remained un- altered for at least three days of culture. Ultrastructural studies by SEM showed that the cells arising from colonic crypts were arranged in a confluent monolayer of very tightly packed cells (Fig. 4A). At higher magnification, the presence of at least two cyto- types was clearly identified (Fig. 4B). The first, were polygonal in shape and their apical surface was covered by dense and high microvilli (Figs. 4C, D). The second cell type was round in shape and characterized by scat- tered microvilli, shorter than those of the former cell type, and by deep invaginations of the apical surface, presumably resulting from the release of secretory mate- rial (Fig. 4C). Ultrastructural analysis by TEM showed that the ap- pearance of the cultured cells differed markedly depend- ing on their position with respect to the crypt residue. As the cells remained aggregated in or around the crypts, they showed a columnar shape typical of the colonocytes (Fig. 5A). Cells were very tall, highly polarized with basal nuclei and apical Golgi complexes and organelles (Fig. 5B). The cells were coupled by junctional complex-
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks