Main

H,K-ATPase is the membrane-bound enzyme responsible for electroneutral exchange of cytoplasmic H+ and external K+ coupled with ATP hydrolysis(1) and therefore contributes to acid secretion. The enzyme is composed of two subunits, namely, α and β(2, 3). Different isoforms of the catalytic α subunit are expressed in stomach, colon, kidney, and uterus(4). The α subunit of H,K-ATPase belongs to a family of proteins with multiple membrane-spanning domains termed“P-type” ATPases. The α subunit of H,K-ATPase shows the greatest levels of sequence similarity with the α subunit of the Na,K-ATPase(4). In the stomach, the H,K-ATPase is localized in intracellular tubulovesicular structures of the parietal cells(5). At the onset of acid secretion, the morphology of the parietal cells changes dramatically: the tubulovesicles decrease in number, whereas the apical membrane of the secretory canaliculi increases in area because of the insertion of the enzyme into the canalicular membrane(6).

In the rat, the H,K-ATPase activity in gastric cells is low until the 3rd wk of life and then rapidly increases after weaning(7). Analysis of the α and β subunit expression seems to indicate that the maturation of enzyme activity in the weaning period is due to increased protein availability(8, 9). Little is known about the hormones that regulate H,K-ATPase maturation. It has recently been shown that corticosteroids play an important role in gastric acid secretion in fetal animals(10, 11) and in postnatal development(12–14), as well as in adulthood(15). In premature babies, the gastric pH does not appear to be sufficiently low to support peptic activity(16). However, it has been reported that treatment with dexamethasone in preterm babies may result in gastroduodenal perforation(17, 18).

In the rat, the maturation of several enzymes is accelerated by GCs in the preweaning period, at the time when the level of circulating GCs rapidly increases(19). It has been shown that GCs also accelerate gastric maturation(20, 21). It is therefore likely that the maturation of gastric H,K-ATPase is stimulated by GCs in the preweaning period. The aim of this study was therefore to investigate the influence of GCs on gastric H,K-ATPase development.

METHODS

Animals. Experiments were performed on Sprague-Dawley rats of different ages (ALAB, Sollentuna, Sweden). Timed pregnant rats were housed individually and given free access to food and water. At gestational age 20 d(i.e., 2 d before expected parturition), the pregnant rats were anaesthetized intraperitoneally with thiobutabarbital (8 mg/100 g of body weight), and the fetuses were extracted within 10-15 min. Ten- and 20-d-old rats were kept with their mothers until the day of the experiment, and adult rats were maintained on regular chow with free access to water. Rats were anesthetized with thiobutabarbital. The stomachs were quickly removed, cut open, emptied of all contents, and washed in ice-cold buffer (NaCl 135 mM, KCl 3 mM, Na2HPO4 10 mM, KH2PO4 2 mM, pH 7.4) and saved for further analysis. After dissection, we pooled stomachs from three to five fetuses and from 10-d-old rats.

In a separate protocol, 10-d-old rats were injected once intraperitoneally with betamethasone (60 μg/100 g of body weight) (Glaxo Lab, Ltd, Greenford, England) or vehicle. After 6, 12, 18, or 24 h, the rats were anesthetized, and the stomachs were immediately removed.

H,K-ATPase activity assay. The H,K-ATPase activity was measured by the hydrolesis of [γ-32P]ATP and was calculated as the difference between total ATPase activity and SCH-28080 insensitive activities according to Kaminski et al.(22, 23). The stomachs were gently homogenized on ice with a Teflon pestle in a buffer containing (in mM): Tris-HCl 75, MgCl2 12.5, EDTA 1.5, pH 7.5. All procedures were performed at 4 °C. Protein concentration was determined by the Bio-Rad method. Aliquots of proteins were preincubated for 30 min in the ice-cold bath with sodium deoxycholate (2 mmol/30 μg of protein), a concentration predetermined to yield maximal activation of H,K-ATPase. The total ATPase activity was measured in 100 μL of a solution containing 10μL of crude membrane preparation (≅30 μg of protein), and 90 μL of a buffer containing: 150 mM Tris-HCl, 10 mM MgCl2, 1 mM EGTA, 2 mM ouabain, 2 mM N-ethylmaleimide, 2 mM sodium azide, 20 μg/mL oligomycin, and 12 mM vanadium-free ATP (pH 7.2) as well as 3 mM[γ-32P]ATP in the presence or absence of 4 mM KCl. To determine H,K-ATPase activity, 200 mM SCH-28080 (a gift from the Schering Corp., Kenilworth, NJ) was added to the buffer. After incubation for 15 min at 37°C, the reaction was terminated by the addition of 700 μL of ice-cold stop solution containing 10% charcoal and 5% trichloric acid. The samples were left on ice for 60 min and then centrifuged with a table centrifuge. The phosphate liberated by the hydrolysis of [γ-32P]ATP was determined in the supernatant. All samples were run in triplicate, and appropriate corrections for blanks and the spontaneous hydrolysis of ATP were made. Enzyme activity was expressed as micromoles of Pi/mg of protein/h.

Northern and dot blot hybridization. Total RNA was isolated from the tissues, as previously described(24). After dissection, we pooled stomachs from three to five fetuses and from two to three 10-d-old rats. For statistical analysis, RNAs from six independent extractions were used in each group. The integrity of the RNA was routinely evaluated by Northern blot (Fig. 1). To quantify the H,K-ATPase mRNA level, 4 μg of total RNA were denatured in ice-cold 10 mM NaOH and blotted in triplicate under a vacuum onto a nylon filter, as previously described(25). DNA probes were random primelabeled with 32P (Megaprime DNA Labeling System, Amersham, Buckinghamshire, UK). Prehybridization (20 min) and hybridization (3 h) were performed at 65 °C with Amersham Rapid Hybridization buffer. Filters were washed at 65 °C to a final stringency of 0.1 × SSPE, 0.1% SDS. The filters were subjected to autoradiography at -70 °C, and the results were analyzed by laser densitometry. Data were corrected by the intensity of an internal standard (pooled mRNA from control rats) to which an arbitrary value of 1 was given. Expression of actin mRNA was also routinely determined to evaluate appropriate loading. Actin cDNA was purchased from Clonthech, Palo Alto, CA). The gastric H,K-ATPase α probe is a 572-base pair fragment(-1981 to -2552 nucleotides), and the β probe is a 196-base pair fragment(-165 to -350), which were synthetized from rat gastric mRNA by RT-PCR. PCR primers of the α probe (20 mer) and of the β probe (22 mer) were designed according to the published sequence(2, 26). The sequences of the primers were checked against the Entrez databases to ensure that there was no homology with unrelated sequences. RNA (2 μg) was reverse-transcribed in a 20-μL reaction containing 50 mM Tris-HCl, 75 mM KCl, 3 mM MgCl2, 1 mM DTT, 2 mM dNTP, 25 mg/mL oligo(dT), 30 U of RNasin, and 100 U of Moloney murine leukemia virus RT. After incubation for 30 min at 42 °C, the sample was used for PCR amplification. Reactions were performed in 100 μL of a buffer containing 10 mM Tris-HCl, 50 mM KCl, 1.5 mM MgCl2, 0.2 mM dNTP, 50 pM of each primer, and 2 U of Taq DNA polymerase. The following temperature profile was used: 1 min at 94°C, 1 min at 55 °C, and 2 min at 72 °C (35 cycles). Gel analysis of the PCR product showed a unique band of the predicted size. The band was eluted from the gel and used for hybridization. Northern blot reveals a unique band in stomach for both α and β subunits and no signal in the colon (Fig. 1), indicating that the probes do not cross-react with colon H,K-ATPase nor with Na,K-ATPase, or other ATPase.

Figure 1
figure 1

Representative Northern blot mRNA analysis of stomach in rats of different ages from fetal life (20-d gestational age,F20) to postntal life (10, 20, and 50 d), and of colon from 10-d-old rats. A total of 10 μg of total RNA were loaded in each lane and probed with gastric α and β H,K-ATPase cDNA. Positions of ribosomal 18 and 28 S are indicated. Ethidium bromide staining of total RNA is shown in the lower panel.

Full size image

Western blot analysis. A homogenate was prepared as that for enzyme activity, and the protein levels were determined as previously described(27). Total protein concentration was determined in duplicate with the Bio-Rad method. A constant amount of the homogenate (2 μg) was resuspended in a buffer containing 1.6% SDS, 2%β-mercaptoethanol, 10% glycerol, Tris 50 mM, and separated on a 8% SDS-polyacrylamide gel. The amount of protein was predetermined as within the linear range of detection. Proteins were then transferred to a nitrocellulose filter. The blots were quenched with 5% dry milk, 0.1% Tween 20 in PBS for 30 min at room temperature. After rinsing in PBS, the filter was incubated with the primary antibody for 60 min, washed, and incubated with alkaline phosphatase-conjugated anti-rabbit secondary antibody (Promega, Madison, WI), always at room temperature. Washing preceded detection with the Protoblot Kit(Promega). Samples from the treated group and the control group were determined on the same filter for purposes of comparison. Quantification of the immunoblot bands was done with laser densitometry. Data were corrected by the intensity of an internal standard (pooled proteins from 10-d-old rats) to which an arbitrary value of.1 was given. The antibody against gastric H,K-ATPase was kindly given to us by Dr. M. J. Caplan(28).

Statistical analysis. Results are expressed as means ± SEM. The statistical analysis was performed with the t test(n = 3-6 in each group).

RESULTS

The present study showed that the gastric H,K-ATPase enzyme matures postnatally. Both α and β H,K-ATPase mRNAs are already expressed in 20-d gestational age fetuses, and no changes are observed during the first 10 postnatal days (Fig. 2). The expression of both H,K-ATPase mRNAs increases about 20 d of life. In adults the α and β mRNAs are 4.7- and 1.8-fold more abundant than in fetuses, respectively (p< 0.01 for both). Similar postnatal changes were observed in enzyme activity and protein abundance (Table 1). The gastric H,K-ATPase activity is low until 20 d of life and increases thereafter to adult levels. The expression of the H,K-ATPase α protein changes during the postnatal period in parallel with enzyme activity (Table 1): there is no significant change between 10 and 20 d of life. In adult life the expression of the α protein is ≅4-fold higher than in young rats (p < 0.01).

Figure 2
figure 2

Ontogenic changes of gastric α and β H,K-ATPase mRNA in normal rats from fetal life (20 d gestational age,F20) to postnatal life (10, 20, and 50 d). A total of 4 μg of total RNA were loaded in triplicate for dot blot analysis. An arbitrary value of 1 has been given to mRNA abundance in fetal rats. Values are means ± SEM (n = 6 in each group).

Full size image
Table 1 Ontogenic changes of H,K-ATPase activity andα protein expression in normal rats
Full size table

In the rat, the circulating level of GCs is very low until the preweaning period, i.e. ≅15 d(24). To study the role of GCs in the maturation of gastric H,K-ATPase, we treated 10-d-old rats with a single injection of betamethasone, before the physiologic upsurge of circulating GCs occurs. In a previous study we have shown that this dose of betamethasone maximally up-regulated Na,K-ATPase in infant tissues(25, 29). The effect of betamethasone on gastric H,K-ATPase activity was examined 24 h after injection. In treated animals, the enzyme activity was significantly higher (2.6-fold, 0.26 ± 0.02versus 0.10 ± 0.05 μmol of Pi/mg of protein/h,p < 0.05) than in control animals (Fig 3A). To ascertain whether GCs stimulate the synthesis of new enzymes or whether they activate preexisting units, we determined by Western blot the abundance of the α catalytic subunit of H,K-ATPase (Figs. 3B and4). The results indicate that the GC treatment increases the abundance of H,K-ATPase α subunit to a similar extent (2.5-fold, p < 0.01). To ascertain whether GCs rapidly stimulate the synthesis of new enzymes by increasing the mRNA abundance, the time-dependent effect of betamethasone on α and β H,K-ATPase mRNA was determined 6, 12, 18, and 24 h after treatment (Figs. 5 and6). GCs did not significantly alter the mRNA abundance during the first 18 h. Twenty-four hours after injection, the gastric H,K-ATPase mRNA was significantly increased, compared with controls (2.8- and 2.2-fold increase for α and β subunit, respectively, p < 0.01).

Figure 3
figure 3

Effects of betamethasone treatment on gastric H,K-ATPase activity (A) and on the α H,K-ATPase protein level(B). Ten-day-old rats were injected with vehicle (control) or betamethasone (treated) 24 h before the experiment. An arbitrary value of 1 has been given to activity and protein abundance in control rats. Values are means ± SEM (n = 6 in each group). Theasterisk indicates a significant difference compared with control rats.

Full size image
Figure 4
figure 4

Representative Western blot analysis of stomach from 10-d-old rats. Rats were injected with vehicle (C) or betamethasone(T) 24 h before the experiment. A total of 2 μg of protein were loaded in each lane and probed with gastric α H,K-ATPase antibody. Molecular size markers are indicated.

Full size image
Figure 5
figure 5

Representative Northern blot analysis of stomach from 10-d-old rats. Rats were injected with vehicle (C) or betamethasone(GC) 24 h before the experiment. A total of 10 μg of total RNA were loaded in each lane and probed with gastric α and β H,K-ATPase and actin cDNA. Positions of ribosomal 18 and 28 S are indicated. Ethidium bromide staining of total RNA is shown in the upper panel.

Full size image
Figure 6
figure 6

Time-dipendent effects of betamethasone treatment on gastric α and β H,K-ATPase mRNA. Ten-day-old rats were injected with vehicle (control) or betamethasone (treated). An arbitrary value of 1 has been given to mRNA abundance in control rats. Values are means ± SEM(n = 3-6 in each group).

Full size image

DISCUSSION

This study indicates that GCs accelerate the maturation of gastric H,K-ATPase by stimulating enzyme synthesis. The time lag between the GC injection and the effect on mRNA abundance indicates that the synthesis of an intermediary protein is probably required.

We found that the α and β H,K-ATPase mRNAs are already expressed at low levels in the fetal stomach and that they rapidly mature during the weaning period. Morley et al.(8) determined the protein levels of both α and β subunits from fetal to adult life in the mouse. Their results indicate that the protein H,K-ATPase abundance is low until the 2nd wk of life, and then it rapidly increases. The present study shows that both α and β H,K-ATPase mRNAs are up-regulated in 20-d-old rats and that this increase is followed by an increase in protein abundance and enzyme activity. Taken together, these studies suggest that the H,K-ATPase synthesis is up-regulated about the time of weaning, and protein synthesis is probably the rate-limiting factor for the rapid maturation of gastric H,K-ATPase occurring during weaning. In rats, the weaning usually takes place at the end of wk 3. Many studies show that several intestinal enzymes are up-regulated during the period preceding the transition from milk to solid food(19, 30). In the rat, the circulating levels of GCs are low during the first 2 wk of life and then rapidly increase during the 3rd wk when they stimulate enzyme maturation in several organs. There is evidence that GCs also contribute to gastric maturation in the preweaning period(20). The present study shows that a single injection of a specific GC agonist at the 10th d of life,i.e. before the physiologic upsurge of circulating GCs, accelerates the maturation of gastric H,K-ATPase by increasing protein availability. It is well established that GCs act by regulating the transcription rate of target genes(31, 32). In this study GCs did not increase H,K-ATPase mRNA until 24 h after treatment. We have shown that GCs can rapidly(≅30 min) increase the Na,K-ATPase mRNA in several infant tissues, such as the kidney, the lung, and the heart(33). In all tissues, the maximal effect on Na,K-ATPase mRNA was achieved within 6 h. More recently, we have demonstrated that GCs also rapidly activate Na,K-ATPase mRNA in the stomach(34), which suggests that the absence of a rapid effect on H,K-ATPase mRNA is not due to reduced tissue bioavailability of the hormone. GCs usually bind to cytoplasmic receptors that, after ligand binding, are transported into the nucleus where they activate the transcription rate of target genes by interacting with specific DNA sequences(GRE)(32, 35). Morley et al.(8) identified several half-consensus GREs on the α H,K-ATPase promoter and speculated about their possible physiologic relevance. The present finding that the H,K-ATPase mRNA did not increase until 24 h after injection does not support the hypothesis of a direct effect. It remains to be evaluated whether the GC receptor binds to the α H,K-ATPase promoter and regulates its gene transcription rate. If this is the case, however, it is likely that an auxiliary factor, whose synthesis is activated by GCs, is required for transcriptional activation. Alternatively, GCs may stimulate the synthesis of a protein that directly regulates gastric α H,K-ATPase mRNA. It has been shown that the injection of GCs in infancy prematurely increases the amount of gastric mucosal pepsinogen(36). The number of gastrin receptors is also increased by GCs, and adrenalectomy delays the normal development of gastrin receptors and the tissue level of the hormone(37). Corticosterone has also been shown to precociously induce sensitivity of the acid secretory mechanism to carbachol and histamine as well as to pentagastrin(36). It remains to be demonstrated if one of these hormones mediates the GC effect on gastric H,K-ATPase mRNA by activating gene transcription rate or by stabilizing the mRNA.

In preterm babies, the capacity to acidify the gastric fluid is impaired compared with older babies(16). Corticosteroids in high doses have been effectively used to treat the infants with bronchopulmonary dysplasia, but not without serious adverse side effects, such as perforated peptic ulcer(17, 18). Furthermore, it also seems likely that the incidence of gastritis or other irritation of the gastrointestinal tract is also increased after GC treatment. The observation that premature infants cannot make gastric acid(16, 38) may lead to the misconception that neonates are at lower risk for the gastrointestinal complications of steroid treatment, because in the developing human fetal stomach from the end of the first trimester, parietal cells are present in a mature, functional form with the potential to secrete both gastric acid and intrinsic factor(39). Our finding that the maturation of gastric H,K-ATPase is indirectly accelerated by GCs may indicate that the increased gastric acid secretion could be a common side effect during GC treatment.