Critical role for NHE1 in intracellular pH regulation in pancreatic acinar cells

David A. Brown, James E. Melvin, David I. Yule


The primary function of pancreatic acinar cells is to secrete digestive enzymes together with a NaCl-rich primary fluid which is later greatly supplemented and modified by the pancreatic duct. A Na+/H+ exchanger(s) [NHE(s)] is proposed to be integral in the process of fluid secretion both in terms of the transcellular flux of Na+ and intracellular pH (pHi) regulation. Multiple NHE isoforms have been identified in pancreatic tissue, but little is known about their individual functions in acinar cells. The Na+/H+ exchange inhibitor 5-(N-ethyl-N-isopropyl) amiloride completely blocked pHi recovery after an NH4Cl-induced acid challenge, confirming a general role for NHE in pHi regulation. The targeted disruption of the Nhe1 gene also completely abolished pHi recovery from an acid load in pancreatic acini in both Math-containing and Math-free solutions. In contrast, the disruption of either Nhe2 or Nhe3 had no effect on pHi recovery. In addition, NHE1 activity was upregulated in response to muscarinic stimulation in wild-type mice but not in NHE1-deficient mice. Fluctuations in pHi could potentially have major effects on Ca2+ signaling following secretagogue stimulation; however, the targeted disruption of Nhe1 was found to have no significant effect on intracellular Ca2+ homeostasis. These data demonstrate that NHE1 is the major regulator of pHi in both resting and muscarinic agonist-stimulated pancreatic acinar cells.

  • fluid secretion
  • exocrine glands
  • intracellular pH regulation
  • intracellular calcium

pancreatic acinar cells' primary physiological role is to synthesize, package in granules, and secrete zymogens in response to stimulation (57). In addition, pancreatic acinar cells also secrete a NaCl-rich primary fluid (21, 41). Although this fluid secretion is minor compared with the copious secretion by the ductal system of the pancreas (2, 21), the acinar fluid secretion plays an important role in hydrating the secreted contents of the granules and effectively “flushing” the zymogens from the secretory end piece into the ductal system. The currently held model for acinar fluid secretion has been most extensively developed in salivary acinar cells (7, 8, 29, 53); however, a majority of the processes involved are common to other acinar cells such as the exocrine cells of the pancreas (42). In the model, Ca2+-mobilizing agonists play a key role in initiating fluid secretion by activating Ca2+-sensitive Cl- channels located in the luminal membrane (3, 4, 29, 40). Subsequently, to maintain the electrical gradient for Cl- movement, Ca2+-activated K+ channels in the basal and lateral membranes are also activated (41, 42). To effectively sustain secretion, intracellular Cl- levels must be replenished, and this is accomplished by a basolateral Na+-K+-2Cl- cotransporter and by a basolateral Math exchanger (23, 24, 30). A consequence of the exit of Math is an intracellular acidification, as a result of increasing H+ levels in the cytoplasm (20, 21, 28, 32, 33, 43). The activity of NHE plays a major role in maintaining intracellular pH (pHi) balance by extruding H+ and thus plays a critical role in maintaining fluid secretion. In support of this concept, fluid secretion from mouse parotid acinar cells is reduced by ∼⅓ in NHE1-null transgenic animals (39). In addition, the activity of specific NHE may also play important roles in the transcellular movement of Na+ as well as in regulation of cell volume during secretagogue stimulation (55).

Genes encoding eight mammalian NHEs have been identified (17, 37, 55). The protein products have been demonstrated to play a role in pH homeostasis, cell volume regulation, transepithelial Na+ and water movement, together with roles in cell adhesion and proliferation (55). The NHE1 isoform appears to have a widespread distribution, including the exocrine pancreas, and is thought to play a housekeeping role by maintaining pHi and cell volume levels. In contrast, NHE2, NHE3, and NHE4 have more limited tissue distribution and are thought to be involved in NaCl absorption (10, 55). The NHE5 isoform has been found to be expressed at high levels in the brain, whereas an intracellular distribution of NHE6 expression in mitochondria has been reported (5, 38). Most recently, the NHE7 isoform has been found to be expressed in the trans-Golgi network (37), whereas NHE8 has recently been reported to be expressed in kidney (17).

Although the activity of NHE has been documented in pancreatic acinar cells (20, 21, 32, 33), the goal of this study was to identify the particular NHE isoform that is expressed and is physiologically important in pancreatic acinar cells. In addition, since changes in pHi can potentially impact the Ca2+-signaling machinery in nonexcitable cells (11, 31, 34, 35), we extended our study to investigate the effects of disruption of pHi regulation on Ca2+ signaling. Using transgenic knockout animals, we demonstrated that NHE1 is responsible for pHi recovery after an acid load and is upregulated during muscarinic stimulation. NHE1 is therefore a major regulator of pHi in both resting and secreting cells. Loss of this pH regulation, however, appears to have a minor impact on secretagogue-stimulated Ca2+ homeostasis.


Materials. The acetoxymethyl ester form of 2′-7′-bis(carboxyethyl)-5-carboxyfluorescein (BCECF-AM) and 5-(N-ethyl-N-isopropyl) amiloride (EIPA) were purchased from Molecular Probes (Eugene, OR). Monoclonal α-NHE1 was obtained from Chemicon (Temecula, CA). Fura-2 AM was purchased from TEFLABS (Austin, TX), and all other chemicals were purchased from Sigma (St. Louis, MO). Math-free solutions contained (in mM) 135 NaCl, 5.4 KCl, 0.4 KH2PO4, 0.33 NaH2PO4, 0.8 MgSO4, 1.2 CaCl2, 10 glucose, and 20 HEPES, pH 7.4. Math-containing solutions contained (in mM) 110 NaCl, 5.4 KCl, 0.4 KH2PO4, 0.33 NaH2PO4, 0.8 MgSO4, 1.2 CaCl2, 10 glucose, 20 HEPES, and 25 NaHCO3, pH 7.4, with NaOH after 30 min of gassing with 95% O2-5% CO2 to equilibrate. NH4Cl solution was used to induce an acid load, and it was prepared by replacing 30 mM NaCl with 30 mM NH4Cl in Math-free and Math-containing solutions. The high-K+ solution used to calibrate ratios into pH values contained (in mM) 120 KCl, 20 NaCl, 0.8 MgCl2, 20 HEPES, and 0.005 nigericin, and the pH was adjusted to between 5.6 and 8.4.

Wild-type and null mutant animals. 129/SvJ-Black Swiss mice were housed in microisolator cages in the University of Rochester vivarium on a 12:12-h light-dark cycle, and they were given access to laboratory chow and water ad libitum. The targeted disruption of murine NHE isoforms NHE1, NHE2, and NHE3 was carried out as described (9, 49, 50). Heterozygous animals were used to establish breeding colonies. Offspring were tail clipped, and genotypes were determined by PCR or by Southern blotting.

Isolation of mouse pancreatic acinar cells. Single and small groups of pancreatic acinar cells were isolated by collagenase digestion of freshly dissected pancreata from wild-type or NHE-deficient Black Swiss mice as previously described (15, 51, 56). Briefly, pancreata were removed and enzymatically digested with 400 units of collagenase type IV (Sigma) in minimum essential medium with 1% BSA and 1 mg/ml soybean trypsin inhibitor for 30 min and gently agitated while being gassed with 95% O2-5% CO2. The cells were then dispersed by trituration before being filtered through a 100-μm nylon mesh. Filtered cells were centrifuged for 2 min at 100 rpm and then resuspended in 2% BSA minimum essential medium solution.

Fluorescence measurement of pHi. Pancreatic acinar cells were loaded with pH-sensitive fluoroprobe by incubation for 30 min at room temperature with BCECF-AM (2 μM). BCECF-loaded cells were gassed continuously with 95% O2-5% CO2 as previously described (14, 36). BCECF-loaded pancreatic acinar cells were allowed to settle and subsequently adhere to a coverslip forming the base of a superfusion chamber on the stage of a Nikon Diaphot 200 microscope interfaced with an Axon Imaging Workbench system (Axon Instruments, Foster City, CA). Experiments were performed at room temperature. Cells were then alternately excited at 440 and 490 nm by using a DG-4 filter changer (Sutter Instruments, Novato, CA) every 10 s, and emitted fluorescence was captured at 530 nm by using a Cooke Sensicam (Cooke, Auburn Hills, MI) 12-bit frame transfer digital camera. A ratio of the fluorescence at 490 vs. 440 nm was computed. pHi was then estimated by using an in situ calibration (52), where external pH was changed in the presence of high K+ and the ionophore nigericin. The fluorescence ratios computed for pH solutions over the physiological range of 6.4-7.6 were linear. Recovery consisted of an initial near-linear increase in pHi, followed by a slower recovery. The linear portion was used to calculate the initial rate of pHi recovery (pHi units/min). All data are representative of three or more experimental runs. In individual experiments, multiple cells from an individual acini were imaged, and the means ± SD for all cells measured in an experimental run are shown in each figure.

Fluorescence measurement of Ca2+ concentration. Pancreatic acinar cells were loaded with the Ca2+-sensitive dye fura-2 AM (2 μM) by incubation for 30 min at room temperature. Fura-2-loaded cells were allowed to adhere to a glass coverslip, which was perfused in a HEPES-buffered physiological saline solution that contained (in mM) 137 NaCl, 0.56 MgCl2, 4.7 KCl, 1 Na2HPO4, 10 HEPES, 5.5 glucose, and 1.26 CaCl2, pH 7.4, that was gravity fed. Imaging was performed by using an inverted epifluorescence Nikon microscope with a ×40 oil immersion objective lens (numerical aperture, 1.3). Fura-2-loaded cells were excited alternately with light at 340 and 380 nm by using a monochrometer-based illumination system, and the emission at 510 nm was captured by using a high-speed, digital frame transfer charge-coupled device camera (T.I.L.L. Photonics). Images were acquired every 500 ms with an exposure of 20 ms. Intracellular Ca2+ concentration ([Ca2+]i) was calculated from the fluorescence ratios by using the equation of Grynkiewicz (18). The maximum and minimum fluorescence ratios were obtained by imaging a 5-μl droplet of physiological saline containing 1 and 0 mM [Ca2+], respectively. All imaging experiments were performed at room temperature, essentially as previously described (15, 51). Traces are from a single cell, representative of multiple individual cells from at least one imaged acini in a particular experimental run, and n represents the number of experimental runs.

Membrane preparation. The purified membrane preparation was prepared by mincing the pancreas in ice-cold homogenizing buffer followed by further homogenization and centrifugation, essentially as described previously for parotid tissue (36). Aliquots were quickly frozen in liquid nitrogen and stored in a -85°C freezer until later use. Protein concentration was measured by using Bio-Rad protein assay reagent (Bio-Rad, Hercules, CA).

Western blot analysis. Equal amounts of purified pancreatic plasma membrane proteins from either wild-type mice or NHE-deficient mice (60-100 μg/lane) were resolved on 7.5% SDS-PAGE and transferred to nitrocellulose membranes (Bio-Rad) as previously described (15). Membranes were incubated with the indicated primary antibody and then detected with a horseradish peroxidase-linked secondary antibody (Pierce, Rockford, IL) and the Super Signal detection system (Pierce) exposed on XAR film (Eastman Kodak, Rochester, NY).


Na+/H+ exchange inhibitor EIPA blocks pHi recovery. In rat acinar cells, recovery from acid load has been shown to be mediated by NHE family proteins (32, 33). To confirm that pHi recovery in the mouse pancreas is accomplished by similar mechanisms, the selective Na+/H+ exchange inhibitor EIPA was used. NHE activity was assessed by acid loading the cells by using the Math prepulse technique (45, 46). A representative pHi trace in Math-free buffer in response to a NH4Cl pulse is shown in Fig. 1A. Resting pHi under these conditions averaged 7.59 ± 0.05 (n = 12 experiments). The addition of 30 mM NH4Cl resulted in a rapid alkalinization of 0.53 ± 0.01 pH units. Subsequent removal of NH4Cl led to an intracellular acidification below basal pH levels by 0.42 ± 0.03 units, followed by a pHi recovery back to the resting values. The initial rate of recovery was 0.18 ± 0.04 pH units/min (Fig. 4). Incubation of cells with 10 μM EIPA resulted in complete inhibition of the recovery from acidification following removal of NH4Cl (n = 5 experiments; Fig. 1A; initial recovery rates are summarized in Fig. 4). This inhibition of NHE activity was completely reversible on removal of EIPA (Fig. 1A). Figure 1B, inset, shows an overlay of control recovery vs. the use of 10 μM EIPA. One minute after removal of NH4Cl, the pHi of the EIPA-treated cells continued to acidify. These data suggest in a similar fashion to data from parotid gland cells that the NHE family of proteins is responsible for pHi recovery in pancreatic acinar cells (14).

Fig. 1.

5-(N-ethyl-N-isopropyl) amiloride (EIPA) reversibly inhibits Na+/H+ exchanger (NHE) activity in wild-type mouse pancreatic acini. 2′-7′-Bis(carboxyethyl)-5-carboxyfluorescein (BCECF)-loaded mouse pancreatic acini from wild-type animals were acid loaded by the addition and subsequent removal of 30 mM NH4Cl (Formula prepulse technique) indicated by the closed bars. A: acini from wild-type animals rapidly recover from intracellular acidification toward the resting intracellular pH (pHi). Incubation of cells with NHE inhibitor EIPA (10 μM) significantly reduced the rate of intracellular recovery (3rd prepulse). The 4th prepulse shows that acini are capable of recovering pHi following the washout of EIPA, indicating that EIPA inhibition is reversible. B: overlay of traces indicating the difference in kinetics between control pHi recovery and EIPA-treated pHi recovery. Each trace shown is representative of 3 or more experiments.

Fig. 4.

Summary of initial rates of recovery. Initial rates of recovery were calculated as described in methods. Data show that initial rates of pHi recovery are not significantly different in wild-type, Nhe2-/-, and Nhe3-/- pancreatic acini. In contrast, no recovery from acid load occurs in either Nhe1-/- acini or wild-type acini treated with EIPA.

Loss of pHi regulation in acini isolated from NHE1-null mutants. To identify which particular isoform of the NHE family is responsible for pH recovery in pancreatic acini, we performed experiments using transgenic animals in which Nhe genes had been disrupted. We initially focused on the NHE1 protein because the expression of this protein has been reported in pancreatic tissue. As shown in Fig. 2A and Fig. 1, after an NH4Cl pulse, wild-type cells recover from an acid challenge within minutes. In contrast, no recovery from an acid load was observed in acini prepared from Nhe1-/- mutant mice (Fig. 2B; initial recovery rates are summarized in Fig. 4). The inset confirms by Western blot analysis that the NHE1 isoform is absent in pancreatic acini from these transgenic animals. However, the resting pHi (7.57 ± 0.03; n = 10 experiments) and the extent of alkalinization (0.55 ± 0.1 pH units) and of subsequent NH4Cl-induced acidification (0.42 ± 0.06 pH units) was not significantly different in the knockout compared with wild-type. Essentially, the lack of recovery of pHi after acidification in NHE1-deficient cells mimicked the inhibition of pHi recovery by EIPA, as demonstrated in Fig. 1.

Fig. 2.

NHE1-dependent pH recovery in mouse pancreatic acini. BCECF-loaded mouse pancreatic acini prepared as described in Fig. 1 legend from wild-type and Nhe1-/- animals. Experiments were performed in both the absence (A and B) and presence (C and D) of Formula. A: acini from wild-type animals recover from intracellular acidification in the absence of Formula. B: recovery from NH4Cl-induced intracellular acid load is inhibited completely in acini from Nhe1-/- animals in the absence of Formula. C: acini from wild-type animals recover from intracellular acidification in the presence of Formula. D: even in the presence of Formula, acini from Nhe1-/- animals do not recover from an NH4Cl-induced intracellular acid load. Each trace shown is representative of 4 or more experiments. B, inset: immunoblot analysis of pancreas membrane proteins (100 μg/lane) using anti-NHE1 antibodies. Membranes were prepared from either wild-type (+/+) or Nhe1-/- animals as indicated. Arrow indicates location of NHE1 protein.

These experiments were repeated in a more physiological, Math-buffered solution. Under these conditions, basal pHi was significantly lower compared with Math-free solutions (7.18 ± 0.04; n = 5 experiments), as previously reported (32, 33). However, wild-type cells recovered from an acid challenge with similar kinetics to that observed in a Math-free environment (Fig. 2C). In a similar manner to experiments in Math-free solutions, acini prepared from NHE1-defi-cient animals showed no pHi recovery following an acid challenge (Fig. 2D). These data indicate that a similar mechanism, i.e., the activity of NHE1, is used by pancreatic acini to recover pHi after an acid load whether in Math-containing or Math-free medium and suggest that Na+/Math exchange is not a major mechanism for pHi regulation under these conditions.

The data in Fig. 2 suggest that NHE1 has an exclusive role in pHi recovery in response to an acid challenge in pancreatic acini. It is, however, formally possible that disruption of the Nhe1 gene could have resulted in lowered levels of other NHE isoforms. To address this issue, we investigated pHi recovery in transgenic mice in which the Nhe3 or Nhe2 gene had been disrupted, resulting in a null mutant animal. Acini prepared from either of these animals retained their ability to recover from an acid challenge in a similar fashion to wild-type mice; indeed, the rates of recovery were not significantly different from wild-type animals (Fig. 3, A and B; initial recovery rates are summarized in Fig. 4; n = 3 experiments each). Furthermore, when membranes prepared from NHE2-deficient animals were probed with NHE1 antiserum, similar amounts of NHE1 protein were evident (Fig. 3, inset). These data collectively suggest that disruption of an Nhe gene per se does not appear to result in decreased levels of other NHE family members. More importantly, these results demonstrate that the NHE1 isoform is the dominant regulator of pHi in unstimulated mouse pancreatic acinar cells. This is consistent with the role of NHE1 in other tissues, such as the parotid, lacrimal, and sublingual glands (14, 22, 36, 48).

Fig. 3.

Pancreatic acini from Nhe3-/- and Nhe2-/- mice retain NHE activity. BCECF-loaded mouse pancreatic acini were prepared as described in Fig. 1 legend from Nhe3-/- or Nhe2-/- animals. Acini isolated from both Nhe3-/- (A) and Nhe2-/- (B) mice recover from an intracellular acidification with kinetics similar to that seen in wild-type animals. Traces are representative of 6 or more experiments. C: NHE1 levels are similar in wild type and Nhe2-/- animals, which suggests that disrupting one Nhe gene does not effect other NHE proteins. Arrow indicates location of NHE1 protein.

Muscarinic stimulation of NHE1 activity. NHE activity is frequently increased or “upregulated” during agonist stimulation (6, 22, 32). We therefore specifi-cally investigated whether NHE1 activity, in addition to functioning in unstimulated acini, was responsible for the increased activity in pancreatic acinar cells during secretagogue stimulation. Experiments were performed in which acini from wild-type animals were incubated with 10 μM CCh during the period of recovery from acid load (Fig. 5A). Incubation with CCh resulted in a significant increase in the rate of recovery in the absence of Math, which averaged 152 ± 15% faster for CCh-treated trials vs. paired controls in the same cells (P = 0.03 by paired t-test; see Fig. 5, inset). In addition, CCh treatment resulted in an alkaline shift or “overshoot” above the initial resting rate of 0.16 ± 0.03 pHi units (n = 3 experiments). A similar overshoot was observed in Math-containing media (0.18 ± 0.06 pHi units; Fig. 5C; n = 3 experiments). In contrast, CCh did not cause any change in the rate of pHi recovery in the Nhe1-/- animals and there was no overshoot in either the presence (Fig. 5D) or absence (Fig. 5B) of Math (n = 3 experiments). These data demonstrate that the activity of NHE is upregulated on secretagogue stimulation in pancreatic acini. Furthermore, this affect can be attributed to the increased activity of the NHE1 isoform, in particular during muscarinic receptor agonist-induced fluid secretion.

Fig. 5.

Upregulation of NHE1 activity during muscarinic stimulation. Experiments were performed as described previously, i.e., application of 10 μM CCh (open bar) immediately following the NH4Cl prepulse. Each wild-type experiment was part of a paired experiment as shown in the inset. A: application of CCh in Formula-free solutions in wild-type cells resulted in upregulation of NHE activity because after pHi recovery an intracellular alkalinization relative to the initial resting pHi (represented by the dashed line) was evident. B: evidence of increased NHE activity in Formula-free solutions is abolished in CCh-stimulated acini prepared from Nhe1-/- mice. C: application of CCh in Formula-containing solutions retained the increased activity of NHE and intracellular alkalinization relative to the resting pHi. D: result of CCh treatment of Nhe1-/- acini in HCO3-containing solutions was identical to Formula-free data.

Role of NHE1 in buffering the intracellular acid load resulting from Math efflux. The results described above demonstrate that NHE1 directly regulates the recovery from an acid load in both stimulated and unstimulated pancreatic acinar cells. Experiments were then performed to determine whether the activity of NHE1 was altered by agonist stimulation under relatively physiological conditions in the absence of an artificial acid load. Exposure of acini to low concentrations of CCh (100-500 nM) resulted in no discernible changes in pHi in either wild-type or Nhe1-/- cells. These data suggest that any changes in pHi that occur under these conditions are below the threshold for detection or represent local changes. However, when cells were stimulated with a higher concentration of CCh (5-10 μM) a rapid alkalinization of 0.22 ± 0.05 pHi units occurred (Fig. 6; n = 4 experiments), indicating that pHi is dynamically regulated during secretagogue-induced fluid secretion. This alkalinization was presumably the result of upregulation of NHE1 because the effect was absent in Nhe1-/- mice (Fig. 6). Stimulation of Nhe1-/- mice with CCh resulted in a slow acidification, which after 3 min averaged 0.29 ± 0.04 pHi units (n = 3 experiments), presumably as a result of Math loss from the cells via the basolateral Cl-/Math exchanger or exit via the luminal Cl- channel. Thus loss of NHE1 activity results in disruption of the cellular regulation of pHi during the secretory process.

Fig. 6.

Effects of muscarinic stimulation on resting pH levels. In the absence of extracellular Formula, 5 μM CCh addition to wild-type BCECF-loaded acini induced an intracellular alkalization (top trace, filled symbols). In the Nhe1-/- acini (bottom trace, open symbols), 10 μM CCh failed to produce the same alkalinization and in fact resulted in a significant acidification. Each trace shown is representative of 3 or more experiments.

Does loss of pHi regulation impact [Ca2+]i signaling? The data presented above indicate that secretagogue stimulation of pancreatic acinar cells results in an upregulation of NHE1 activity. A critical signal in the overall process that gives rise to fluid secretion during muscarinic stimulation is an increase in the [Ca2+]i. Because the cellular machinery that results in an increase in [Ca2+]i is a rich source of potential sites for regulation by changes in pHi (11, 31, 34, 35), the consequences of the loss of this pHi regulation during Ca2+ signaling was investigated next. An experimental paradigm was designed to initially assess any effect of disordered pHi regulation on both a physiological, oscillatory signal and a peak and plateau response to a higher concentration of agonist. In addition, since we have shown a marked pHi change in response to the high concentration of secretagogues, which is essentially nonreversible over the time course of the experiment in the Nhe1-/- mice (Fig. 6), we assessed the effects of this change in pHi on a subsequent oscillatory response by restimulating with a low concentration of CCh. Figure 7 shows wild-type (Fig. 7A) and Nhe1-/- (Fig. 7B) pancreatic acinar cells responding to a low and then subsequently a high concentration of CCh, followed by a repeat exposure to a low concentration of CCh. Acini from both wild-type and Nhe1-/- animals responded with very similar Ca2+-signaling profiles, characteristic of the concentration of CCh used. When the characteristics of the signals were analyzed in detail, no statistical differences were observed in the response to maximal CCh concentration. These experiments revealed that the initial peak used as a measure of maximal Ca2+ release {[Ca2+]i over basal was 335 ± 36 nM (n = 6; 29 cells) vs. 379 ± 54 nM (n = 4; 22 cells) for wild-type vs. Nhe1-/-, respectively} or the plateau height measured after 5 min as an indicator of depletion-activated Ca2+ entry {Δ[Ca2+]i was 133 ± 18 nM (n = 6; 29 cells) vs. 121 ± 11 nM (n = 4; 22 cells) for wild-type vs. Nhe1-/-, respectively} were not different. In a similar fashion, there were no obvious differences in the initial peak height or oscillation frequency when comparing the first or second response to 300 nM CCh in the wild-type or Nhe1-/- acini. For example, the initial elevation over basal stimulated by the first application of agonist averaged a Δ[Ca2+]i of 290 ± 34 nM (n = 12; 59 cells) vs. 356 ± 42 nM (n = 8; 43 cells) for wild-type vs. Nhe1-/-, respectively, and the average oscillation frequency was 4.32 ± 0.24 vs. 4.7 ± 0.31 oscillations/min for wild-type vs. Nhe1-/- acini, respectively. These parameters, although decreased on second application of 300 nM CCh, did not change relatively. For example, the initial Δ[Ca2+]i was 55 ± 9% (n = 4; 17 cells) of the first response in wild-type and 40 ± 5% in the Nhe1-/- acini (n = 3; 18 cells), whereas the oscillation frequency was 90 ± 4% of the first response vs. 106 ± 8% for wild-type vs. Nhe1-/-. Together, these data indicate that although NHE1 activity is critically important in pHi regulation in pancreatic acini, disruption of this regulation does not appear to exert a major effect on [Ca2+]i-signaling events.

Fig. 7.

Disruption of the NHE1 protein does not affect Ca2+ signaling. Fura-2 loaded mouse pancreatic acini from wild-type and Nhe1-/- animals were stimulated with either a low (300 nM) or high (5 μM) concentration of CCh, which in the Nhe1-/- acini resulted in a sustained acidification after removal of the agonist. After removal of CCh, the acini were stimulated by a second challenge with 300 nM CCh. A: 300 nM CCh treatment caused repeated oscillations in wild-type pancreatic acini. Higher concentrations of CCh caused an increase in the initial peak of the intracellular Ca2+ concentration ([Ca2+]i) response and resulted in a sustained plateau where [Ca2+]i levels remained elevated during agonist application. A second application of 300 nM CCh again results in an oscillatory Ca2+ signal, of somewhat reduced magnitude. The trace is representative of 4 experiments. B: identical paradigm applied to Nhe1-/- pancreatic acini resulted in [Ca2+]i signals with similar characteristics to wild type. The trace is representative of 5 experiments.


Expression of a sodium-dependent proton exchanger that is regulated during secretion in pancreatic acinar cells has been demonstrated in a number of studies (21, 23, 32, 33). However, the molecular identity of this exchanger had not, to this point in time, been elucidated. At least eight genes encode members of the NHE family of proteins, and multiple members of this family have been identified in exocrine cells (1, 14, 17, 30, 39, 47). In particular, although the presence or absence of other isoforms has not been determined, mouse pancreatic cells have been reported to express NHE1 and NHE4 based on their profile of inhibition by various agents and by immunocytochemistry (1, 47). In addition to the predicted localization to the basolateral membrane of acinar and ductal cells, the NHE1 and NHE4 isoforms have also been reported to be expressed intracellularly on zymogen granule membranes (1, 47). A role has been proposed for these intracellular exchangers in regulated exocytotic secretion. Other related exocrine glands such as salivary gland acinar cells express NHE1, NHE2, and NHE3 in a manner that appears to be species and/or gland specific (14, 19, 25, 39).

In this study, we have shown that NHE1 accounts for a majority, if not all, of Na+/H+ exchange activity in mouse pancreatic acinar cells under conditions of imposed acid load. Moreover, this observation holds under both Math-free conditions where this activity is isolated and under more physiological conditions in the presence of Math where Cl-/Math exchange activity could contribute to the acid loading of the cell and where Na+-Math cotransport could potentially play a role in recovery from acid challenge. Since mouse pancreatic acinar cells from Nhe1-/- mice were incapable of pHi recovery from an acid load, it follows that other NHE isoforms expressed cannot substitute for NHE1 activity, presumably because of localization to an intracellular compartment. Furthermore, these data indicate that Na+-Math cotransport appears to play little if any role in recovery from acid load. In addition, we demonstrate that the activity of this particular isoform is increased during muscarinic secretagogue stimulation both under secretory conditions and on recovery from an acid load. These data are in agreement with the idea that the ubiquitous expression of NHE1 functions as the “housekeeping” isoform and more specifically is consistent with reports from both sublingual and parotid acinar cells demonstrating an important role for NHE1 in secretagogue-stimulated fluid secretion (14, 36).

The mechanism responsible for upregulation of NHE activity by secretagogues was not addressed in this study. However, an abundant literature on epithelial cell types indicates that stimulation of NHE1 activity is mimicked acutely by maneuvers that result in elevations of [Ca2+]i but not by activation of protein kinase C (26, 27, 55). This is indicated as upregulation and can be mimicked by Ca2+ ionophore treatment but is not duplicated by diacylglycerol analogs. NHE activation by Ca2+ is thought to occur either by the direct interaction of Ca2+ or by the rapid, high-affinity binding of Ca2+-calmodulin to region A of the exchanger (54). Although additional mechanisms such as cell shrinkage or Cl- depletion (44) could contribute to the regulation of NHE1 in pancreatic acinar cells, regulation by changes in [Ca2+]i would be entirely consistent with intracellular events known to be stimulated by CCh treatment in this cell type.

The machinery responsible for initiating, propagating, and clearing an increase in [Ca2+]i on stimulation by pancreatic secretagogues is a rich potential source of loci for regulation by changes in pHi. For example, inositol-1,4,5-trisphosphate receptors are exquisitely sensitive to changes in pHi (11), the binding of inositol-1,4,5-trisphosphate to its receptor being markedly potentiated at basic pH. In addition, store-dependent Ca2+ entry in pancreatic acinar cells is also markedly inhibited by relatively small changes in pHi to more acidic values (34). This potential modulation of Ca2+ release and influx by changes in pHi is consistent with reports that changes in pHi markedly affect the wave speed of propagating Ca2+ signals in pancreatic acinar cells (16). In addition, aberrant pHi regulation occurs in disease states such as cystic fibrosis as consequence of decreased Math efflux (12, 13). We thus investigated the characteristics of [Ca2+]i signals in pancreatic acinar cells from Nhe1-/- mice, where pHi regulation is markedly perturbed. When [Ca2+]i was measured on stimulation with either a physiological or supermaximal concentration of CCh, various characteristic parameters such as peak height or oscillation frequency were not different when comparing Nhe1-/- with wild-type. This suggests that, even in the face of a marked (∼0.3 pH unit) acidification during CCh stimulation in Nhe1-/--derived pancreatic acinar cells, the pH change is not sufficient to adversely impact [Ca2+]i-signaling events (Fig. 6).

In summary, this study has demonstrated for the first time that NHE1 is responsible for NHE activity in pancreatic acinar cells under conditions of acid load. In addition, the activity of this specific protein is solely responsible for enhanced Na+/H+ exchange activity during secretagogue stimulation. Surprisingly, the disruption of pHi regulation in Nhe1-/- mice failed to markedly alter the muscarinic receptor-associated changes in Ca2+ mobilization.


Grants from the National Institutes of Health to J. E. Melvin (RO1-DE-08721) and D. I. Yule (DK-54568) supported this study. D. A. Brown was supported by National Institute of Dental Research Training Grant T32-DE-07202.


We thank Drs. Ha-Van Nguyen, Keith Nehrke, and Trevor Shuttleworth for helpful discussion during the course of this study.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract