Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

The stomach consists of many types of cells, including smooth muscle cells, mesenchymal cells, vessel forming cells, nerve cells, blood cells, including immune cells and gastric gland cells. Gastric epithelial cells can be further subdivided into at least 11 different cell types, ranging from highly differentiated cells to actively proliferating undifferentiated cells (1). Chief cells are characterized by production and secretion of pepsinogen, parietal cells have a specialized function as acid secreting cells, and neck cells and pit cells (surface mucous cells) are recognized as mucous producing cells. In addition, there are several kinds of endocrine cells producing gastrin, somatostatin, and histamine. These cells are considered to be terminally differentiated. On the other hand, premature forms of these cells such as pre-pit cells and pre-parietal cells also exist in the gastric gland. Interestingly and importantly, all of these different cell types are known to be originated from a single “stem cell”.

When researchers try to culture gastric epithelial cells, they are faced with at least two major problems. First, is the problem of the “purity” of the cells. As mentioned above, the stomach or gastric gland consists many different cell types. Therefore, it is difficult to get a highly purified culture, consisting only of a single cell type. Soll et al. (2) first solved this problem. They made isolated cell suspensions from canine fundic mucosa and fractionated the cells by their size, using a counterflow elutriation technique followed by a Percoll density gradient. Using this method, our knowledge of parietal cells, ECL cells, D cells, G cells, and chief cells has been greatly enhanced. Cell purity obtained by this method is sufficient for certain experiments such as acid output and gastrin or somatostatin secretion.

However, for more precise experiments such as elucidation of intracellular signal transduction mechanisms, contamination of the culture by other cell types is a critical problem. Kinoshita et al. succeeded in culturing primary pit (surface epithelial) cells based on this method (3). Fortunately, in the case of primary culture of gastric pit cells, they are present in high numbers and attach and grow on a culture plate relatively faster than other types of gastric gland cells (4). The methods we describe here finely utilize these characteristics of gastric pit cells.

Secondly, is the problem of “differentiation”. Immediately after gastric cells are separated and inoculated onto plastic plates, they start to de-differentiate and rapidly lose or decrease their terminally differentiated characteristics. Although culture conditions, such as cell density in culture medium, additives in the culture medium, and plate coating materials, can modify the tendency towards de-differentiation, careful observation of cultured cells and close comparison with in vivo models are indispensable.

Here, we introduce two methods of primary culture of rat gastric epithelial cells and two cell lines derived from nontransformed rat or mouse gastric glands.

1.1 Primary Culture of Gastric Epithelial Cells from Newborn Rat Stomach

There have been a number of attempts to establish primary cultures of gastric epithelial cells (5-7), but the methods developed by Terano et al. (8) have been most widely adopted. Using this method, over 90% of the cultured cells have the characteristics of epithelial cells. Although previous researchers tried to reduce the number of fibroblasts by 20% by administration of pentagastrin (7), this method using newborn rat stomach is more efficient. Even using this method, however, fibroblast overgrowth can be observed after subculturing on d 4. This overgrowth can be prevented by collagenase treatment or using F-12 medium containing D-valine. The mitotic index is maximum on d 2 (2.0%).

After the success of primary culture of gastric epithelial cells from newborn rat stomach, many attempts were made to obtain a large number of cells from a single procedure from adult rat stomach. Gastric epithelial cell culture systems from adult rat (9), rabbit (10,11), and guinea pig (12) were reported. Generally, epithelial cultures from adult stomach contain more fibroblasts than those from newborn stomach. To prevent contamination with a large number of fibroblasts, Matsuoka et al. (10) inverted the stomach to expose the mucosa to proteinase E to digest only surface epithelial cells efficiently and then scraped off the mucous layer before enzymatic digestion with collagenase (12).

2 Materials

2.1 Primary Culture of Gastric Epithelial Cells from Newborn Rat Stomach

  1. 1.

    1- to 2-wk-old Sprague-Dawley rats.

  2. 2.

    Hank’s balanced salt solution (HBSS) containing 100 U/mL of penicillin and 100 μg/mL of streptomycin.

  3. 3.

    Enzyme solution: HBSS supplemented with 0.1% collagenase and 0.05% hyalu-ronidase (both from Sigma).

  4. 4.

    Nylon mesh 200 (Nakarai Tesque, Japan).

  5. 5.

    Growth medium: Coon’s modified Ham’s F-12 medium supplemented with 10% fetal bovine serum (FBS), 15 mol/L HEPES, 100 U/mL fibronectin, 100 U/mL penicillin, 100 μg/mL streptomycin, 100 μg/mL gentamycin.

  6. 6.

    Sodium pentobarbitol: 5 mg/mL stock; use 10μL/g bodyweight.

2.2 Primary Culture of Gastric Epithelial Cells from Adult Rat Stomach

  1. 1.

    8-wk-old Wistar rats.

  2. 2.

    Perfusion solution: calcium- and magnesium-free HBSS supplemented with 50mMEDTA.

  3. 3.

    Digestion solution: HBSS supplemented with 0.75% type IV collagenase, 0.1% hyaluronidase.

3 Methods

3.1 Primary Culture of Gastric Epithelial Cells from Newborn Rat Stomach

  1. 1.

    Sacrifice the rats by anethestizing with an ip injection of diluted sodium pento-barbital (5 mg/mL stock; 10 μL/bodyweight g).

  2. 2.

    Resect the stomachs from the rats (see Note 1 ).

  3. 3.

    Place in HBSS containing penicillin and streptomycin in a 10-cm plastic plate at room temperature.

  4. 4.

    Excise the fundic area (this is usually recognized as the area with folds) with fine scissors from the stomach.

  5. 5.

    Cut the fundic tissue into strips.

  6. 6.

    Rinse the strips 3 times with HBSS and then mince into 2- to 3-mm3 pieces (see Note 2 ).

  7. 7.

    Place the minced tissue into enzyme solution.

  8. 8.

    Incubate this suspension at 37°C in a shaking water bath for 60 min (see Note 3 )

  9. 9.

    Pipet the tissues up and down several times to complete dispersion of the cells.

  10. 10.

    Incubate for a further 15 min and then pipet again.

  11. 11.

    Filter through nylon mesh.

  12. 12.

    Centrifuge the filtrate, containing cell clumps, at 1000g for 5 min.

  13. 13.

    Wash the pellet in growth medium and centrifuge as before.

  14. 14.

    Resuspend the pellet in growth medium.

  15. 15.

    Maintain the cultures at 37°C with 5% CO2 in air in a humidified atmosphere. When cells are cultured on a plastic plate coated with type I collagen, they attach and grow better than on a noncoated plate. Generally, culture medium (Dul-becco’s modified Eagle medium [DMEM]/F12 with 10% FBS) is changed every other day. Commercial trypsin-EDTA solution can be used when subculturing these cells. However, in comparison with other cultured cells, these cells need more time to detach completely. After 4 to 5 passages, most cells lose their growth activity and die.

3.2 Primary Culture of Gastric Epithelial Cells from Adult Rat Stomach

Even using the methods described in Subheading 3.1. , contamination of the epithelial cell cultures with fibroblasts is still a major problem (see Note 4 ). Ichinose and his group partially solved this problem (13,14) using the following protocol. This method makes it possible (i) to obtain epithelial cultures with less fibroblast contamination, and consequently (ii) to observe in detail the relationship between gastric epithelial cells and mesenchymal cells.

They also investigated the role of substratum (type I collagen, type IV collagen, fibronectin, and laminin) on epithelial cell attachment and proliferation. They reported that gastric epithelial cells obtained by this method were able to form a monolayer and proliferate on plastic plates coated with those substratum when cultured with Ham’s F-12 medium supplemented with only 0.1% of bovine serum together with epidermal growth factor (EGF), cholera toxin, hydrocortisone, and insulin. Unlike intestinal epithelial cells obtained by the same method (15), response to transferrin was not significant in gastric epithelial culture.

  1. 1.

    Anesthetize the rats with pentobarbital.

  2. 2.

    Cut the abdomen with scissors in the middle to upper left upper region.

  3. 3.

    Insert a needle (18 or 16 G) connected to a silicon-coated tube into the right atrium of the heart.

  4. 4.

    Cut 5 mm from the left ventricle.

  5. 5.

    Perfuse the rat using ice-cold perfusion solution, using a perfusion pump (5 mL/min or slower) until the whole liver looks pale.

  6. 6.

    After perfusion, the stomach can easily be separated into epithelium and mesen-chyme under a dissecting microscope.

  7. 7.

    Add digestion solution to the stomach epithelium.

  8. 8.

    Shake the epithelium and digestion solution in a flask in a water bath (100 cycle/min) at 37°C for 15 min (this time will vary depending on the experiment).

  9. 9.

    Check small samples of epithelium under light microscopy while shaking, until digestion is seen to be complete.

  10. 10.

    After removing the digestion solution by centrifugation, culture cells on type I collagen-coated plastic plates in DMEM/F-12 with 10% FBS.

3.3 An Epithelial Cell Line Derived from Nontransformed Rat Stomach (see Note 5 )

Although the methods were improved, primary cell culture of gastric mucosal cells is time-consuming, and the resulting cells cannot reach 100% purity. In many cases, these cells are not suitable for DNA transfection. Furthermore, because these cells have begun apoptosis as soon as they have lost the physiological 3-dimensional relationship that they have within a tissue, they are basically not suitable for experiments that examine apoptotic events.

Matsui has established a cell line derived from normal rat gastric mucosa, RGM1 (16). The stomach was harvested from an anesthetized 4-wk-old Wistar rat and inverted so that it was inside-out. After washing the mucosa with phosphate-buffered saline (PBS) at 4°C, the inverted pouch was immersed in a 0.2% pronase E solution at 37°C. The solution was then changed every 15 min and centrifuged to collect the exfoliated gastric cells. Thereafter, the cells were washed twice with PBS and cultivated in a 1:1 mixture of DMEM and Ham’s F-12 medium supplemented with 20% FBS (RGM1 could be cultured with 10% fetal calf serum [FCS]). When the cells were at the tenth passage, the cell line was named RGM1.

They examined characteristics of RGM1 using cells at passage 30-40. RGM1 cells are homogeneous epithelial-like cells with large oval nuclei and a polygonal shape. They grow as a monolayer with a doubling time of 15.7 h and a saturation density of 1.97 ± 0.38 x 105 cells/cm2. They stop proliferating when they become confluent and do not grow in multiple layers. RGM1 cells do not form colonies and form single cells in soft agarose. Notably, RGM1 DNA was found to have diploid pattern by flow cytometry. These features are the characteristics of untransformed cells. Prostaglandins, ICAM1, insulin-like growth factor (IGF)-II, des-1-IGF-II, IGF binding protein-2, SPARC, and β2-microgloblin are produced in RGM1 cells.

During the past few years, RGM1 has been used as the standard nontrans-formed epithelial culture model and precise characteristics of RGM1 have been reported. For example, Miyazaki showed that heparin-binding EGF-like growth factor is an autocrine growth factor for rat gastric epithelial cells (17). Hassan found that prostaglandin (PG) E2 plays a role on mucin synthesis through PG EP4 receptor, not EP1 and EP3 (18). Jones investigated the expression of COX-2 in RGM1 and the stimulatory effect of hybridoma growth factor (HGF) on COX-2 expression (19). Pai reported effects of Helicobacter pylori vacuolating cytotoxin (VacA) on re-epithelialization of wounded gastric epithelial monolayers (20).

RGM1 is registered at RIKEN cell bank in Japan (RCB0876). The RIKEN home page is (http://www.rtc.riken.go.jp

3.4 Epithelial Cell Line Derived from Nontransformed Mouse Stomach (see Notes 7 and 8 )

Although this chapter is on “rat” gastric epithelial cell culture, we would like to describe GSM06 cell line, a nontransformed epithelial cell line derived from mouse stomach. Many investigators have suggested that immortalization of cells by a temperature-sensitive simian virus 40 (stSV40) large T-antigen gene retains more or less stable cell type specific function of the original cells and that the oncogene products are rapidly degraded at nonpermissive temperature but functions at the permissive temperature. Obinata and his group had established transgenic mice harboring a tsSV40 large T-antigen gene, and Tabuchi had established a surface mucous cell line GSM06 from the stomachs of these transgenic mice (21,22).

Gastric fundic mucosal cells from the transgenic mouse were isolated as a modification of a method for rats described by Schepp (23). The isolated gastric fundic mucosal cells were suspended in DMEM/F12 medium supplemented with 2% FBS, 1% ITES, 10 ng/mL EGF, and antibiotics (100 U/mL penicillin, 100 μg/mL streptomycin, 25 μg/mL amphotericin B), and seeded onto a collagen-coated plastic culture dish and incubated at 37°C for 24 h in a humidified incubator in a 5% CO2 atmosphere. The cells were then cultured under similar conditions except for a temperature change to 33°C. When the cells were used for experiments, they were cultured in DMEM/F12 medium supplemented with 10% FBS, 1% ITES, and 10 ng/mL EGF in a humidified atmosphere. Like RGM1, GSM06 forms a confluent monolayer and shows characteristics of untransformed pit cells (see Note 6 ). GSM06 cells grown at 33°C (permissive temperature) and 37°C (intermediate temperature) having a doubling time of about 29 h and a saturation density of 2.76 ± 0.19 x 105 cells/cm2. In contrast, at the nonpermissive temperature (39°C), GSM06 cells did not grow, but when the temperature of the culture was lowered to 33°C, cell growth was restored. Chromosome analysis showed that the chromosome number in GSM06 cells was distributed widely (2n = 35 - 102). In contrast, primary culture cells from the gastric mucosa of normal mice or transgenic mice had 38-43 or 38-42 chromosomes, respectively (for mouse, 2n = 40).

By modifying culture conditions, a wide variety of different types of pit cells can be simulated using GSM06. When cells were cultured in a tightly confluent monolayer or at 39°C, GSM06 shows more differentiated features. On the contrary, when they are at a nonconfluent cell density or at 33°C, the GSM06 are less differentiated (24-26). This change of differentiation is significant and reliable, however, compared to terminally differentiated pit cells in vivo mucous granules in GSM06 cultured in confluence or at 39°C are small in number. Like the primary cultured gastric epithelial cells, GSM06 contain periodic acid Schiff reaction (PAS)-positive granules and are able to synthesize and secrete glycoprotein. In spite of production of such glycoprotein, primary cultured cells usually do not produce a glycoprotein sheet as is seen in the case of the gastric surface mucosa in vivo (8,10). However, GSM06 could produce mucous sheets.

Further information on GSM06 can be obtained at Exploratory Research Laboratories III, Daiichi Pharmaceutical Co. Ltd., 16-13, Kitakasai 1-chome, Edogawa-ku, Tokyo 134, Japan. Fax+81-3-5696-8334.

4 Notes

  1. 1.

    It is important to keep the cells at a low temperature during the whole procedure.

  2. 2.

    Do not allow the cells to dry out. Even while in the procedure of mincing, a small amount of culture medium should be added to the tissue.

  3. 3.

    When cells are treated with enzyme, the treatment period must be as short as possible.

  4. 4.

    Be careful to avoid fibroblast contamination. It takes more than 10 min for the cells mentioned above to attach to a plastic plate. Fibroblasts can attach more quickly, so cells can be preplated, the fibroblasts allowed to adhere, and then the epithelial cells removed and replated.

  5. 5.

    Neither primary cultured gastric pit cells nor the gastric pit cell line are sufficient material for investigating the physiological function of pit cells. Although they are derived from normal or nontransformed animals, once they start growing as cultured cells, many of their physiological functions will become different from normal pit cells in vivo.

  6. 6.

    It is important to note that primary pit cell cultures contain not only other types of epithelial cells, but also nonepithelial cells such as fibroblasts and blood cells, and the type and percentages of such contaminating cells can vary from experiment to experiment. This is especially important for sensitive experiments, such as polymerase chain reaction (PCR).

  7. 7.

    Gastric epithelial cells migrate upward rapidly and finish their life in 3 d (2 27). On the other hand, RGM1 cells are more resistant to apoptotic stimulation than most of the cultured gastric cancer cell lines.

  8. 8.

    It should be emphasized that in the living stomach, epithelial cells are continuously affected by the signals from mesenchymal cells, extracellular matrix, and influencing luminal milieu. Experiments using cultured cells could exclude this “noise”, although no epithelial cells can exist without them under physiological conditions.