Growth Hormone Stimulates, Through Tyrosine Kinase, Ion Transport and Proliferation in Human Intestinal Cells

Abstract

Background: 

Growth hormone (GH) stimulates intestinal growth and differentiation and promotes water and ion absorption in the rat intestine. Epidermal growth factor has similar effects, which involve tyrosine kinase activity. The effects of growth hormone on ion transport and cell growth and the role of tyrosine kinase in these effects were examined in a human-derived intestinal cell line (Caco-2).

Methods: 

For transport study, electrical parameters were measured in human intestinal Caco-2 cell monolayers mounted in Ussing chambers. Cell growth was monitored by counting and ³H-thymidine incorporation in the presence and absence of growth hormone. The role of tyrosine kinase was investigated by using its specific inhibitor genistein.

Results: 

The addition of growth hormone induced a rapid, Cl^--dependent, decrease in short-circuit current without affecting tissue conductance, which is consistent with an anion-absorptive effect. Incubation with growth hormone increased cell count by 85% and ³H-thymidine incorporation by 64% versus the count in control specimens. The absorptive and trophic effects of growth hormone were dose-dependent, and the maximum effective concentration was identical for each effect. Genistein blocked and growth effect on ion transport and cell growth.

Conclusions: 

Growth hormone stimulates ion absorption and cell growth in human enterocytes. Both effects result from a direct growth hormone-enterocyte interaction, and both require tyrosine kinase activity. Growth hormone may have therapeutic potential in intestinal diseases characterized by epithelial atrophy and loss of water and electrolytes.

Growth hormone (GH) plays a major role in somatic growth ⁽¹⁾. The existence of a widespread expression of GH receptor tanscripts in the fetal and the result intestinal human mucosal cells, particularly in enterocytes ⁽²⁾, suggests that GH exerts specific functions in the fetal and postnatal gastrointestinal tract. There is increasing evidence that GH exerts a trophic effect in the intestine. It stimulates enterocyte growth and differentiation in rats which have undergone small bowel resection ⁽³⁾, it is required for growth and differentiation of fetal rat intestinal transplant ⁽⁴⁾, and it induces a trophic effect in human mucosa cultured in vitro ⁽⁵⁾. Conflicting results are reported in animal models and in man on the efficacy of GH in enhancing intestinal adaptation in short-bowel syndrome ^(6-10).

Growth hormone also plays a role in intestinal ion transport processes. It increases water and ion absorption in the perfused rat intestine in vivo, and, consistent with an anion-absorptive effect, it decreases the short-circuit current in the unstripped rat intestine mounted in an Ussing chamber ⁽¹¹⁾. The absorptive effects are exerted throughout the rat intestine ⁽¹²⁾, but the intensity of the ion-absorptive effect decreases in a proximal-to-distal direction ⁽¹²⁾. This region-specific distribution of absorptive effects resembles the pattern of the trophic effect induced by GH in the intestine ⁽¹³⁾.

The earliest event in the intracellular signaling mechanisms subsequent to GH binding is the activation of a tyrosine kinase designated Janus kinase 2 ^(1,14).

Another growth factor, the epidermal growth factor (EGF), exerts effects similar to those of GH on intestinal growth and on ion transport through a direct interaction with the enterocytes ^(15,16). Tyrosine kinase activity is involved in the EGF-induced effects on NaCl absorption in rabbit ileum ⁽¹⁷⁾. It also plays an important role in cell growth and proliferation, and there is evidence showing an association between tyrosine kinase and EGF receptor ^(16-19). These data suggest that EGF and GH exert similar effects in the intestine.

The purpose of this study were to verify whether GH stimulation of intestinal ion absorption and cell growth results from a direct hormone-cell interaction and to test the hypothesis that tyrosine kinase activity is involved in either or both effects.

The Caco-2 cell line is an established experimental model that serves to investigate both ion transport and cell growth ^(20-21). Moreover, specific GH receptor has been detected on Caco-2 cells at the exponential phase of growth and in mature Caco-2 enterocyte 15 days after confluence ⁽²²⁾. We therefore used this cell line to investigate the effects of GH on enterocytes and their mechanisms.

MATERIALS AND METHODS

Cell Culture

CaCo-2 cells were obtained from the American Type Culture Collection (Rockville, MD). Cells were grown in Dulbecco's modified Eagle's medium with high glucose concentration (4.5 g/l) supplemented with 10% fetal calf serum (FCS), 1% non-essential amino acids, 50 mU/ml penicillin, and 50 mg/ml streptomycin and were incubated in 5% CO₂-95% air. Medium was changed daily.

Ion Transport Studies

Cells were grown on uncoated polycarbonate transwell filters, as previously described ⁽²⁰⁾, and were used for intestinal transport studies 15 days after confluence. The filter area was 4.9 cm². Each filter was mounted in an Ussing chamber (WPI, Sarasota, FL, U.S.A.) as a flat sheet between the mucosal and the serosal compartments. Each compartment contained 10 ml of Ringer's solution with the following composition (in millimoles per liter): 114 NaCl, 5 KCl, 1.65 Na₂HPO₄, 0.3 NaH₂PO₄, 1.25 CaCl₂, 1.1 MgCl₂, 25 NaHCO₃, and 10 glucose. The buffer was constantly gassed with 5% CO₂-95% O₂, and it was connected to a thermostat-regulated circulating pump to maintain a temperature of 37°C ⁽²²⁾.

The following electrical parameters were measured as described elsewhere ⁽²³⁾, before and after mucosal or serosal addition of GH: transepithelial potential difference, short-circuit current (Isc) and tissue ionic conductance (G). Isc is expressed as microamps per square centimeter, (µA/cm²), G as millisiemens per square centimeter (mS/cm²), and potential difference as millivolts (mV). To evaluate whether the electrical effects induced by GH depend on Cl^-, equimolar concentrations of SO₄- were substituted for Cl^-.

Genistein, a specific inhibitor of tyrosine kinase ⁽²⁴⁾, was used to investigate the role of tyrosine kinase in the effects exerted by GH. Monolayers of CaCo-2 cells were incubated for 30 minutes with 100 µM genistein (on the serosal or the mucosal side), after which 4 × 10^-9 M GH was added on the serosal side.

Cell viability was evaluated by measuring the electrical response to the serosal addition of theophylline (5 mM) added to the serosal side at the end of each experiment.

Growth Studies

Two methods were used to measure the trophic effects elicited by GH: ³H-thymidine incorporation and cell counts.

Thymidine Incorporation

Cells were seeded into 96-well microtiters plates (10⁴ cells/well) in DMEM with 10% FCS and incubated for 12 hours to allow entry to growth phase. The cells were incubated with FCS-free DMEM and with the test substances for a further 30 hours. ³H-Thymidine was added at 0.5 µCi/well, and incubation was continued for 18 hours. Cells were harvested (Titertek cell harvester; Flow Laboratories, Rickmansworth, UK), the resultant filters were dried, and β radioactivity was counted.

Cell Counting

Cells were plated in 24-well tissue culture plates (10⁴ cells/well), using two wells per concentration of each experimental condition, and were grown in DMEM supplemented with 10% FCS and antibiotics for 72 hours. Cells were then deprived of serum for 24 hours. Before the experiments, the medium was replaced with FCS-free medium containing 0.2% bovine serum albumin, added to stabilize the hormone ⁽²⁵⁾. The cells were harvested with 1% trypsin, 48 hours after the addition of each testing substance. The resultant cell suspensions were randomized to another investigator, and cells were counted in a blinded manner. Cell viability was determined by trypan blue exclusion. The difference in paired counts did not exceed 5%.

The role of tyrosine kinase was investigated using both thymidine incorporation and cell counting in the presence of 100 µM of the specific inhibitor genistein. Genistein was added to the medium 1 hour before the addition of GH, and it remained in the incubation medium throughout the experiment (48 hours).

Chemicals

All chemicals were to reagent grade and were purchased from Sigma (St. Louis, MO, U.S.A.). Genistein was purchased from Calbiochem (San Diego, CA, U.S.A.), ³H-thymidine from Amersham Italia (Milan, Italy), culture media from Life Technologies Gibco (Mascia e Brunelli, Milan, Italy), Transwell filters and supports from Costar (Costar Italia, Milan, Italy), and recombinant human GH from Serono (Industria Farmaceutica Serono, Rome, Italy).

Statistics

Each experiment was run in duplicate or triplicate and was repeated at least three times. Results are expressed as means ± standard error of the mean (SEM). Student's t-test was used for statistical analysis.

RESULTS

Intestinal Transport Studies

The addition of GH to the mucosal side of the cell monolayer had no effect on electrical parameters. On the contrary, the addition of the hormone to the serosal side caused Isc and potential difference to decrease but did not effect G. Isc peaked within 50 minutes of the addition of GH (Fig. 1). The effect on Isc was dose-dependent and was detected at a GH concentration as low as 2 × 10^-10 M (Fig. 2). The effect was at its maximum at 4 × 10^-9 M. Higher doses did not decrease Isc further.

FIG. 1: Time course of the effect on short-circuit current (Isc) and monolayer electrical conductance (G) of the addition of 4 × 10^-9 M of growth hormone (GH) to the serosal or the mucosal side of Caco-2 cells mounted in Ussing chambers. The conductance of cell monolayers was not modified by addition of GH to the serosal side and in the control specimens. Asterisks indicate that the difference between the Isc data point in cells receiving GH on the serosal side and Isc in control specimens was significant (p < 0.01). There was no difference in the Isc data point between cells receiving GH on the mucosal side and control specimens. Each data point is the mean of three observations.

Fig. 2: Short-circuit current (Isc) response to the addition of increasing concentrations of growth hormone (GH) to Caco-2 cells. ΔIsc values are expressed as the maximum difference between control cells and cells exposed for 60 minutes to GH on the serosal side. Negative values are consistent with an anion-absorptive effect. Each data point is the mean of three observations.

These experiments were repeated in Cl^--free buffer. The basal electrical parameters of cells incubated in modified Cl^--free Ringer's solution were not different from those of cells exposed to standard Cl^--containing Ringer's. However, in Cl^--free buffer, GH did not induce significant modifications of Isc (data not shown). This suggests that the observed GH decrease in Isc involved transepithelial Cl^- movement.

The effects of GH were abolished by the tyrosine kinase inhibitor genistein (Fig. 3). The addition of genistein to the serosal or mucosal side of the cell monolayer caused a transient increase in Isc but did not affect G. In cells pretreated with genistein, the subsequent addition of GH to the serosal side had no effect on Isc (Fig. 3).

Fig. 3: Effect of growth hormone (GH) and genistein, alone or in combination on short-circuit current (Isc) in Caco-2 cells. Amounts of 4 × 10^-9 M GH and 1 × 10^-4 M genistein were added to the serosal side of the Caco-2 cell monolayer mounted in Ussing chambers. The arrows indicate the time of addition of genistein or GH. Asterisks indicate values statistically significant compared with baseline Isc (p < 0.01). Data are expressed as mean ± standard error of the mean of three observations.

Cell Growth Studies

Continuous exposure of near-confluent Caco-2 cells to GH for 48 hours resulted in a dose-dependent increase in in ³H-thymidine incorporation (Fig. 4); the maximum increase of 64% was obtained with 4 × 10^-9 M GH, the identical concentration that induced the maximum effect on ion transport. The GH concentration at which the half-maximum effect on ion absorption and cell proliferation was observed was also identical (10^-9 M) (Fig. 4).

Fig. 4: Effect of increasing concentrations of growth hormone (GH) on ³H-thymidine incorporation in Caco-2 cells. Results are expressed as disintegrations per minute per microgram of cell protein. The short-circuit current values (Isc) are reported for comparison. Each data point is the mean of three observations.

Cell-counting experiments confirmed the results of thymidine incorporation. Exposure of near-confluent cells to GH for 48 hours induced a significant increase in cell number, with a maximum increase of 85% versus the count in control specimens (Fig. 5). The effect was dose-dependent. The minimum and maximum effects occurred at concentrations of 2 × 10^-10 M and 4 × 10^-9 M, respectively (data not shown), the same GH concentrations that caused the minimum and maximum effect, respectively, on ion transport and on thymidine incorporation.

Fig. 5: Effects of genistein on basal or growth hormone (GH)-stimulated cell proliferation and ³H-thymidine incorporation in Caco-2 cells. Cells were incubated with 4 × 10^-9 M GH and 1 × 10^-4 M genistein, either alone or in combination for 48 hours. ³H-Thymidine was added in the last 12 hours of incubation. Asterisks indicate that the difference in cell number between cells exposed to GH and control cells is statistically significant (p < 0.01), and ● indicates that the difference in cell number between cells exposed to GH alone and cells pretreated with genistein and subsequently exposed to GH is statistically significant (p < 0.01). Data are expressed as mean ± standard error of the mean of three observations.

To investigate whether tyrosine kinase is involved in the effects induced by GH on cell growth, GH was added to genistein-treated cells, as in transport studies. Cell-counting and thymidine incorporation experiments showed that the presence of genistein alone did not significantly affect the growth of cells under conditions in the current study (Fig. 5). Genistein significantly inhibited the trophic effect induced by GH on Caco-2 cells (Fig. 5).

DISCUSSION

We have demonstrated that GH directly stimulates both ion absorption and cell growth in human-derived enterocytes. The GH-induced decrease in Isc was similar to that observed in a rat model ^(11,12) and is consistent with an increased flux of an anion from the mucosal to the serosal compartment as a consequence of its increased absorption or decreased secretion. In fact, the abrogation of the Isc response in the Cl^--free Ringer's experiment suggests that Cl^- transport is involved in the Isc response. Growth hormone stimulated ion absorption only when it was added to the serosal side, which is in agreement with the finding that specific GH receptors are located on the serosal surface of intestinal epithelial cells ^(2,26). The similarity of the electrical responses observed with the rat intestine ^(11,12) and those seen in the cell line model, suggest that the effects of GH on ion transport are the result of a direct interaction of GH-enterocyte receptors.

Similarly, the data from cell growth experiments suggest that the trophic effect of GH is the result of a direct hormone-enterocyte interaction. Whether the effects of GH on enterocyte growth and proliferation are at least in part mediated by insulin-like growth factor-II which is also produced by Caco-2 cells ⁽²⁷⁾, or by other autocrine mediators, or whether they are directly related to GH itself, remains to be investigated.

Genistein is an isoflavone that inhibits tyrosine-specific protein kinase activity ⁽²⁴⁾. Genistein inhibits NaCl absorption induced by EGF in the rabbit intestine ⁽¹⁷⁾. In rabbit ileum, genistein has an absorptive effect when added to the mucosal side and a secretory effect when added to the serosal side ⁽¹⁷⁾. In contrast, in our experimental model, genistein increased Isc when added to either the mucosal or the serosal side. The difference may be because of a species-specific polarity of intestinal cells or because of the specimens used (i.e., whole intestinal tissue which contains inflammatory, neural, and endocrine-paracrine cells vs. that containing intestinal epithelial cells) ⁽¹⁷⁾.

However, the inhibition elicited by genistein on the GH-induced Isc decrease and on the proliferative response indicates that tyrosine kinase activity is a prerequisite for the effects exerted by the hormone on the enterocytes.

It has been shown that tyrosine kinase is involved in EGF-induced neutral NaCl absorption, which is mediated by the Na⁺/H⁺ exchanger ⁽¹⁷⁾. This raises the hypothesis that GH also could, through tyrosine kinase, stimulate ion absorption through an effect on the Na⁺/H⁺ exchanger. In addition, it is known that the Na⁺/H⁺ exchanger is also involved in the trophic effects exerted by EGF ^(28,29).

In this study, three lines of evidence point to an association between the effects exerted by GH on ion transport and on cell growth: the similar stoichiometry of both effects, their tendency toward saturation, and their dependence on tyrosine kinase activity. A dual action of growth factor in regulating ion transport processes and cell proliferation is not a novel concept: For instance, it is known that EGF is a potent mitogen for intestinal cells, and it is also involved in the control of intestinal transport processes ^{(15-17,28-30)}. An association between the two effects of GH is also suggested by their topographical distribution, which is consistent with a proximal-to-distal decreasing pattern in other animal models ^(12,13).

This evidences suggests the hypothesis that these two phenomenons are linked-that is, the events at the membrane level involved in the regulation of ion absorption may initiate proliferation. In fact, GH exerted a short-term effect on ion transport and a long-term effect on cell growth.

It should be noted that the effects of GH on ion transport were totally abolished by genistein, whereas a residual effect on cell proliferation was observed even after tyrosine kinase inhibition. This seems to suggest the co-existence of multiple mechanisms of cell growth stimulation elicited by GH, the main one of which could involve tyrosine kinase. Alternate signaling pathways have been demonstrated including protein kinase-C, phospholipase C activation, and calcium-dependent pathways ^(1,31,32). However, the different time settings of the experiments conducted to investigate the effects on growth and on ion transport may well explain quantitatively different results.

To conclude, this study showed that GH stimulates both ion absorption and cell proliferation in human enterocytes, both these effects results from a direct interaction the hormone and the intestinal epithelium and show stoichiometric similarities, and tyrosine kinase activity is involved in these effects. In light of the results of this study and the findings that GH inhibits the effects of potent secretagogues in the rat ileum ⁽¹¹⁾, it is feasible that GH can counteract intestinal damage and diarrhea. Controlled clinical studies are needed to establish whether this hormone may have a therapeutic role in selected intestinal diseases.

Acknowledgments: The authors thank Jean Gilder for substantial editing of this text.

This work was supported by a grant from the Ministero della Sanita' AIDS research project 1997 (Program 50a.0.29) and by a Grant 94.02505 CT04 from Consiglio Nationale Delle Ricerche, Rome, Italy.