Intestinal Bicarbonate Secretion in Cystic Fibrosis Mice

Lane L Clarke*, Xavier Stien and Nancy M Walker
Dalton Cardiovascular Research Center and Department of Biomedical Sciences, University of Missouri. Columbia, MO, USA
Corresponding Author
Lane L Clarke
University of Missouri-Columbia
Dalton Cardiovascular Research Center
Research Park Drive
Columbia, MO 65211-3300
Phone +1-573-882.7049
Fax +1-573-884-4232
E-mail [email protected]


Visit for more related articles at JOP. Journal of the Pancreas


Gene-targeted disruption of the cystic fibrosis transmembrane conductance regulator (CFTR) in mice results in an intestinal disease phenotype that is remarkably similar to bowel disease in cystic fibrosis patients. In the intestinal segment downstream from the stomach (i.e., the duodenum), CFTR plays an important role in bicarbonate secretion that protects the epithelium from acidic gastric effluent. In this report, we examine the role of CFTR in cAMP-stimulated bicarbonate secretion in the murine duodenum and the mechanisms of acid-base transport that are revealed in CFTR knockout (CF) mice. Ion substitution, channel blocker and pH stat studies comparing duodena from wild-type and CF mice indicate that CFTR mediates a HCO3 - conductance across the apical membrane of the epithelium. In the presence of a favorable cellto-lumen HCO3 - gradient, the CFTR-mediated HCO3 - current accounts for about 80% of stimulated HCO3 - secretion. Exposure of the duodenal mucosa to acidic pH reveals another role of CFTR in facilitating HCO3 - secretion via an electroneutral, 4,4’-diisothiocyanatostilbene-2,2’ disulfonic acid (DIDS) sensitiveCl- /HCO3 - exchange process. In CF duodenum, other apical membrane acid-base transporters retain function, thereby affording limited control of transepithelial pH. Activity of a Cl- - dependent anion exchanger provides nearconstant HCO3 - secretion in CF intestine, but under basal conditions the magnitude of secretion is lessened by simultaneous activity of a Na+ /H+ exchanger (NHE). During cAMP stimulation of CF duodenum, a small increase in net base secretion is measured but the change results from cAMP inhibition of NHE activity rather than increased HCO3 - secretion. Interestingly, a small inward current that is sensitive to the anion channel blocker, 5-nitro- 2(3-phenylpropyl amino)-benzoate (NPPB), is also activated during cAMP stimulation of the CFTR-null intestine but the identity of the current is yet to be resolved. Studies to identify the proteins involved in non-CFTR mediated HCO3 - secretion are on-going and potentially will provide targets to correct deficient HCO3 - secretion in the CF intestine.


Antiporters; Chloride Channels; Cyclic AMP; Cystic Fibrosis Transmembrane Conductance Regulator; Duodenum; Sodium- Hydrogen Antiporter


CF: CFTR knockout mice; CFTR: cystic fibrosis transmembrane conductance regulator; DIDS: 4,4’- diisothiocyanato-stilbene-2,2’ disulfonic acid ; EIPA: 5-(N-Ethyl-N-isopropyl) amiloride; Gt: total tissue conductance; Isc: short-circuit current; Jsm HCO 3: serosal-to-mucosal bicarbonate flux; KBR: Krebs bicarbonate Ringers solution; NHE: Na+/H+ exchanger: NPPB: 5-nitro-2(3- phenylpropyl amino)-benzoate; WT: wild-type

In cystic fibrosis patients and CFTR knockout mice, duodenal bicarbonate transport is greatly diminished, resulting in abnormal pH regulation at the mucosal surface. Two transport pathways at the apical cell membrane are involved in bicarbonate secretion - an anion conductance(s) and Cl-/HCO3 - (OH-) exchanger(s). Stimulation of intracellular cAMP yields electrogenic bicarbonate secretion that requires the activity of CFTR to either provide a bicarbonate conductance and/or a chloride conductance that recycles Cl- entering the cell via a Cl-/HCO3 - exchanger. Although patch clamp and apical membrane preparation studies have shown that CFTR is moderately permeable to HCO3 - (PCl:PHCO3 = about 0.25) [1, 2, 3] it has been difficult to determine whether CFTR mediates a bicarbonate conductance under physiological conditions in native duodenal epithelium. Two lines of evidence have emerged. First, in vivo measurements and pH stat studies have shown that CFTR is required for electrogenic bicarbonate secretion under conditions that inhibit apical membrane Cl-/HCO3 - exchange activity [4, 5, 6, 7, 8]. Second, anion substitution studies of intact duodenal mucosa, such as the study of murine duodenum shown in Figure 1, indicate that cAMP-stimulated CFTR can carry a bumetanide-insensitive HCO3 - current when other anions that have significant permeability in CFTR are removed from the bathing medium.
Interestingly, careful examination of CFTR knockout duodenum reveals a finite increase in bicarbonate (base) secretion in response to cAMP stimulation. As shown in Figure 2a, bicarbonate secretion (Jsm HCO 3) is minimal under basal conditions. Following treatment with forskolin (cAMP), Jsm HCO 3 increases by 1 μEq/cm²·h and this change is accompanied by a similar increase in short-circuit current (Isc).
Note, however, that the cAMP-induced changes in the bioelectric properties include a significant increase in total tissue conductance (Gt), a measure of the paracellular pathway in the intestine [9]. To evaluate the possibility that the cAMP-induced D Jsm HCO 3 is a consequence of both the imposed transepithelial bicarbonate gradient (see Methods) and an increase in paracellular permeability, the cAMP-induced DJsm HCO 3 was correlated with the DGt. As shown in Figure 2b, no relationship existed between the DJsm HCO 3 and DGt under either condition. This finding confirms earlier studies of murine duodenum showing that Jsm HCO 3 does not correlate with Gt during cAMP stimulation [10].
The above findings indicated an active bicarbonate secretory process, therefore, the involvement of luminal Cl-/HCO3 - exchange activity during cAMP stimulation of the CF duodenum was investigated by replacing Cl- in the luminal bath with the poorly permeable solute, isethionate. Interestingly, this maneuver resulted in net acid secretion during the basal period that was then abolished by cAMP treatment (Figure 3a). The basal Isc of the duodenum was greatly accentuated under these conditions but did not increase with cAMP treatment. Acid secretion during inhibition of luminal Cl-/HCO3 - exchange is consistent with activity of luminal Na+/H+ exchange activity as recently suggested by pH measurements of the luminal content in wild-type and NHE3 knockout mice [11]. We tested this hypothesis by exposing the luminal membrane to the NHE exchange inhibitor, 5-(N-Ethyl-N-isopropyl) amiloride (EIPA), at a concentration (100 μM) that inhibits the major intestinal isoforms, NHE2 and NHE3, in media containing physiological concentrations of Na+ [12]. As shown in Figure 3b, EIPA treatment increased the basal Jsm HCO 3 to about 1 μEq/cm²·h and prevented the increase in Jsm HCO 3 during cAMP stimulation. Thus, the increase in Jsm HCO 3 measured during cAMP stimulation of the CF duodenum is likely due to cAMP inhibition of NHE activity, which reveals activity of a Cl-- dependent anion exchanger(s). This latter conclusion was confirmed by the lack of an EIPA effect on the CF duodenum during Clsubstitution in the luminal bath (data not shown).
Although the cAMP change in Jsm HCO 3 was abolished by the amiloride analog EIPA, forskolin stimulated a small but significant increase in the Isc of the CF duodenum. Thus, the cAMP-induced DIsc in the CFTR knockout duodenum was dissociated from the DJsm HCO 3 in both the luminal Cl- substitution and EIPA experiments, indicating that the current is not carried by HCO3 -. Evaluation of the individual cAMP-induced DIsc indicates the presence of a subpopulation of CF mice that have robust responses (see Figure 4a), suggesting that a cohort of CF mice surviving to adulthood may be selected for the expression of an alternate conductance. Inhibitor studies of the cAMP stimulated Isc in CF duodenum indicate partial blockade by the anion conductance inhibitors, DIDS and NPPB (Figure 4b). Although the identity of the conductive pathway has not been resolved, these findings indicate the presence of an alternate cAMP-sensitive anion channel that may modify the physiological consequences of gene-targeted deletion of CFTR in murine intestine.


Animals. Wild-type (WT) and CFTR knockout (CF) mice 2-4 months of age were used. The mice were fasted overnight before experimentation (water was provided adlibitum).
Ussing chamber studies. Freshly-excised duodenum was stripped of the underlying muscle layers and mounted on standard Ussing chambers with 0.25 cm² exposed surface area. All sections were treated with 1 μM indomethacin and 0.1 μM tetrodotoxin (serosal) prior to experimentation. The duodenal sections were voltage-clamped using an automatic voltage clamp (Physiologic Instruments, San Diego, CA, USA).
pH stat. The duodenal studies consisted of two sequential 30 min flux periods: a basal period and a treatment period using either 10 μM forskolin (cAMP) or 100 μM EIPA (EIPA). All drugs were obtained from Sigma Chemicals (St. Louis, USA). The luminal surface of duodenum was bathed with 4 mL of an unbuffered NaCl solution containing (in mM): Na+, 144.0; Cl-, 154.0; K+, 5.2; Ca2+, 1.2; Mg2+, 1.2. The mucosal bath pH was clamped at 7.4 by neutralizing the appearance of base with 5 mM HCl or acid with 5 mM NaOH using an automatic titrator (Radiometer, Radiometer Analytical, Lyon, France). The mucosal solution was gassed with 100% O2. The serosal surface was bathed with Krebs bicarbonate Ringers solution (KBR) containing (in mM): Na+, 140.0; Cl-, 120.0; HCO3 -, 25.0; H2PO4 -, 0.4; HPO4 2-, 2.4; K+, 5.2; Ca2+, 1.2; Mg2+, 1.2; glucose, 10; pH 7.4 (gassed with 95% O2: 5% CO2). Both solutions were warmed to 37 °C by water-jacketed reservoirs. For Cl- free lumen experiments, chloride was replaced with 91 mM gluconate and either 25 mM SO4 2- plus 25 mM mannitol or 50 mM isethionate.
Statistics. Paired t-test was used to compare two sequential treatment periods and an unpaired t- test was used to compare two treatment groups. Repeated measures ANOVA was used to compare three sequential treatment periods and a one-way ANOVA was used to compare multiple treatments groups. A p value less than 0.05 was considered statistically significant. All data are given as means±SEM.


The study was funded by grants from the Cystic Fibrosis Foundation and National Institutes of Health (DK48816).

Figures at a glance

Figure 1 Figure 2 Figure 3 Figure 4
Figure 1 Figure 2 Figure 3 Figure 4


Select your language of interest to view the total content in your interested language

Viewing options

Post your comment

Share This Article

Flyer image

Post your comment

captcha   Reload  Can't read the image? click here to refresh