In Cftr^{TgH(neoim)Hgu} mice the exon 10 of the Cftr gene is disrupted by the insertion of the vector pMCIneoPolyA 4. This construct was crossed from Cftr^{TgH(neoim)Hgu} transgenic CF mice into the DBA/2J and C57BL/6J backgrounds via heterozygous carriers of the mutation. The vector insert is known to be excised at low frequency in heterozygous carriers of mutant and wild type alleles 9. Hence all CF mice of this study that were subjected to electrophysiological analysis were typed at the Cftr locus in order to verify their CF genotype. All typed D2-CF/3, B6-CF/3, CF/1 and CF/3 mice were carrying the pMCIneoPolyA insertion at the same integration site in exon 10 of the Cftr gene and were homozygous for the same intragenic Cftr 3-marker microsatellite haplotype like their CF/3 founders [see Additional file 1]. This data demonstrated that the two congenic mouse lines shared the organization of the Cftr locus with that of CF/1 and CF/3 mice.

Next, the genetic background of DBA/2J, C57BL/6J and the four CF lines was compared in 26 SNPs whereby each murine chromosome was represented by one or two SNPs [see Additional file 1]. The congenic D2-CF/3 and B6-CF/3 mice shared the SNP genotypes with wild type DBA/2J and C57BL/6J mice. The exception was SNP rs3023064 located 0.5 Mbp upstream of Cftr where all typed B6-CF/3 mice had the same genotype as the CF/3 strain [see Additional file 1]. The shared SNP genotype profile indicated that D2-CF/3 and B6-CF/3 were congenic mouse lines. This conclusion could however have been flawed by the low informativeness of the marker set. The six inbred strains were identical in genotype for 7 of the 26 SNPs. To allow an unequivocal discrimination between the inbred strains, we selected a set of five informative microsatellites for further genotyping [see Additional file 1]. This marker set concurrently differentiated the closely related CF/1 and CF/3 and the unrelated DBA/2J, C57BL/6J and CF/3 from each other. D2-CF/3 and B6-CF/3 shared homozygous microsatellite genotypes with DBA/2J and C57BL/6J wild type, respectively. Thus, the profile of marker genotypes indicated that both D2-CF/3 and B6-CF/3 strains were congenic and were harbouring the same sequence as CF/3 mice at Cftr and adjacent loci.

Male mice had significantly higher body weight than females of the same age in all investigated mouse strains but D2-CF (Tables 1 and 2). With the exception of C57BL/6J males, body weight was similar for wild type and the corresponding congenic CF mice of the same gender indicating that the 'leaky' insertion mutation 5 did not significantly affect growth. Inbred CF/1 and CF/3 mice were heavier than the congenic CF mice of the same age and gender in accordance with the lower body weight of wild type inbred lines DBA/2J and C57BL/6J compared to outbred HsdOla:MF1 (Tables 1 and 2). Body weight of CF/1 and CF/3 mice of both sexes was similar to that of DBA/2J males, but higher than that of DBA/2J females. Similarly, CF/1 females were heavier than C57BL/6 females. In summary, gender and genetic background were more important than the Cftr^{TgH(neoim)Hgu} mutation in shaping body weight.

The inbred CF/1, CF/3 and B6-CF/3 mice were fertile in the homozygous CF state. D2-CF/3 females, however, were not fertile. Hence, the line was propagated by crossing D2-CF/3 males with female heterozygous carriers of the Cftr^{TgH(neoim)Hgu} mutation. Tables 3 and 4 provide information about the reproductive fitness of the investigated strains. Wild type C57BL/6 mice generated the highest number of offspring. Litter size of the congenic B6-CF/3 strain was comparable, but the number of litters per breeding pair was lower than that of the wild type strain (P < 0.01). CF/1 and CF/3 mice produced the lowest number of litters. Litter size, however, was normal. CF/3 mice generated more progeny per litter than DBA/2J and as many as C57BL/6J mice (Table 3).

Survival rates beyond weaning were 81% and 91% for wild-type DBA/2J and C57BL/6J, both 87% for the two congenic strains D2-CF/3 and B6-CF/3, 86% for CF/1 and 91% for CF/3. Under our housing conditions death rates during the perinatal and weaning periods were not higher than in wild type mice for all four inbred CF mouse lines.

Aberrant epithelial ion flow is the hallmark of the basic defect in CF 1. We investigated the secretory responses of nasal, ileal and colon epithelia upon exposure to secretagogues and channel blockers in the four inbred Cftr^{TgH(neoim)Hgu} strains and their corresponding wild type controls. The outcome of the Ussing chamber measurements is summarized in the histograms of Figure 1.

CF nasal epithelium was exposed to the sodium channel blocker amiloride (Fig. 1A). Whereas all wild type mouse strains showed a minute response to amiloride, nasal tissues of all four CF mouse strains responded with a decrease of the short circuit current within the range of 1 μEqh^-1cm^-2. Next, forskolin was added which activates adenylate cyclase, production of cAMP and subsequent opening of cAMP-dependent chloride conductances. Forskolin evoked a stronger chloride secretory response in wild type than in the corresponding CF nasal epithelium in accordance with the expectation that forskolin should elicit a larger luminal flow of chloride in Cftr-proficient than in Cftr-compromised mice (Fig. 1B). This difference was particularly pronounced in the two congenic CF strains, although – due to the large scatter of values of individual DBA/2J mice – it was only significant for the comparison between C57BL/6J and B6-CF/3 mice. The genetic background was as least as important as the non-CF vs. CF state for luminal chloride flow. The CF strains CF/1 and CF/3, for example, showed a similar chloride secretory response as the non-CF DBA/2J and C57BL/6J mice.

Next, the chloride secretory responses upon stimulation with carbachol or with the cAMP agonist forskolin and the phosphodiesterase inhibitor genistein were measured in the mouse ileum which is the site of the most profound pathology in most CF mouse models, namely the fatal intestinal obstruction syndrome 2311. This complication was absent in all four strains CF/1, CF/3, B6-CF/3 and D2-CF/3. The proximal intestine is known to express numerous anion transporters, exchangers and channels, of which CFTR is the major respondent upon activation with forskolin 12. The addition of forskolin and genistein stimulated statistically indistinguishable short circuit currents in the ileum of all seven tested mouse strains (Fig. 1D). The interindividual response within a strain was more variable than the mean interstrain difference suggesting that at least in the ileum all four CF strains produced sufficient Cftr and/or other chloride conductances to mount a normal or close-to-normal chloride secretory response.

Carbachol was the other agonist to stimulate chloride secretion. Carbachol initiates the cholinergic activation of chloride secretion by increasing the intracellular Ca²⁺ concentration through stimulation of Ca²⁺ influx and mobilization of intracellular Ca²⁺ stores 13. This causes Ca²⁺ dependent K⁺ efflux, which acts as the electrogenic driving force for apical chloride secretion 14. Moreover, carbachol activates CFTR in the apical membrane by increasing the formation of diacylglycerol, thus stimulating the protein kinase C – dependent signaling pathway 15. A large transient chloride secretory response upon addition of carbachol, i.e. an increase of Isc, is characteristic for intestinal epithelia and is caused mainly by CFTR-mediated chloride secretion 16.

The mean chloride secretory response of the tested mice to serosal application of carbachol was 1.3- to 3-fold stronger in wild-type than in their corresponding Cftr^{TgH(neoim)Hgu} strain(Fig. 1C). However, pairwise comparisons revealed significantly larger Isc only between the outbred HsdOla:MF1 and the four CF strains (Fig. 1C). The ileal epithelium of CF/1 and CF/3 showed a similar Isc response as that of wild type C57BL/6J mice. In other words, genetic loci other than Cftr influenced the response to carbachol.

The high levels of Cftr mediated chloride secretion in the ileum of the congenic CF mice were unexpected, although it corresponded to their uniform mild phenotype. To address the issue if a substantial carbachol-induced chloride secretory response was a general feature of the intestine of congenic mice, we also measured the carbachol response in the colon. The number of animals in each group was increased in order to unequivocally answer the question whether or not isogenic non-CF and Cftr^{TgH(neoim)Hgu} mouse strains could be differentiated by their transient chloride secretory response to carbachol which is the basis for the diagnosis of CF in intestinal current measurements in humans 1718. Moreover the protocol was modified so that the carbachol response should almost exclusively be caused by Cftr-mediated chloride secretion. The proximal and distal specimens from the colon ascendens were preincubated with amiloride, TEA and BaCl₂ in order to block the apical Na⁺ and K⁺ conductances. Subsequently, the tissues were preincubated with DIDS prior to carbachol addition. DIDS does not affect the Cftr channel from the extracellular side at physiological pH, but it blocks other chloride conductances such as the CaCC and ORCC 1920.

Taking a size of ten animals per group and following the modified protocol, the change of short circuit current in colonic epithelium induced by carbachol was significantly lower in all CF strains than in their wild type controls (Fig. 1E), although the values recorded for individual CF and non-CF mice were overlapping. Mean Isc values of the D2-CF/3 and the B6-CF/3 strain were 63% and 51% of those of DBA/2J and C57BL/6J mice, respectively. In contrast to the findings in the nasal and ileal epithelia, the mice with the DBA/2J and C57BL/6J backgrounds mounted stronger chloride secretory responses in the colon to carbachol than HsdOla:MF1 and the derived CF/1 and CF/3 mice. Carbachol-stimulated chloride secretion was within the same range in colon and ileum of the three latter strains with MF1 background. In contrast, the response was 4-fold higher in the colon than in the ileum of the other four investigated strains (Figs. 3C, 3E). Hence, the genetic backgrounds divergently influenced absolute and relative levels of Cftr-mediated chloride secretion in the proximal and distal intestine.

Offspring (generation F4) of the original Cftr^{TgH(neoim)Hgu} mouse population with divergent genetic backgrounds originating from the 129/Sv, C57BL/6 and HsdOla:MF1 strains 4 were used to generate the CF/1 and CF/3 inbred mouse strains by brother × sister mating at the Central Animal Facility of the Hannover Medical School 9. Congenic mouse strains were generated on DBA/2J and C57BL/6J backgrounds by using Cftr^{TgH(neoim)Hgu} mice as a donor strain establishing the D2-CF/3 and B6-CF/3 strains 928 by speed congenic production 29.

All experiments were approved by the local government as well as by the local Institutional Animal Care and Research Advisory Committee. All mice were bred in the Central Animal Facility of the Hannover Medical School, except for the HsdOla:MF1 strain. Mice of this strain had been bred by Harlan Winkelmann GmbH (Borken, Germany) and were shipped to the Central Animal Facility of the Hannover Medical School four weeks before the beginning of the experiments.

The C57BL/6J strain was kept under specified pathogen-free conditions. All other mouse strains were kept in individually ventilated cages. The temperature was 21°C, the relative humidity 55 ± 5%. The C57BL/6J strain was fed an irradiated standard chow (ssniff R-Z V1324-300) and autoclaved water ad libitum. The other mouse strains were fed a different irradiated standard chow (altromin 1314) and filtered water also ad libitum. Litter size and number of litters were recorded for all breeding pairs housed at Hannover Medical School.

High molecular weight DNA was isolated from murine liver tissue based on the protocol by Gross-Bellard et al. 30. Mice were genotyped by a low density SNP scan (Kbiosciences, Hoddesdon, England), microsatellites and PCR spanning the integration sites of the pMC1neoPolyA vector in exon 10 of Cftr 9. Microsatellites were amplified by PCR from genomic DNA, separated by direct blotting electrophoresis 9 and evaluated as outlined by Mekus et al. 31.

During the generation of the congenic strains, the insertion of the vector was verified for all litters by Southern blot hybridizations 49. This control was necessary because the spontaneous excision of the vector has been observed at a frequency of 4–5% in heterozygous carriers of the insertion mutation 9.

Freshly excised mouse ileum, colon and nasal epithelium of CF/1, CF/3, D2-CF/3 and B6-CF/3 homozygous animals were used for short circuit current (Isc) measurements and compared with wild type controls. Experiments were performed at 37°C. The perfusion solution was saturated with 95% O₂ and 5% CO₂, pH 7.4.

Modified Meyler's solution (128 mM NaCl, 4.7 mM KCl, 1.3 mM CaCl₂, 1.0 mM MgCl₂, 20.2 mM NaHCO₃, 0.4 mM NaH₂PO₄, 0.33 mM Na₂HPO₄, 10 mM HEPES) was used for the measurements on ileum and nasal epithelium, and Krebs-Henseleit solution (113.6 mM NaCl, 5.4 mM KCl, 1.2 mM CaCl₂, 1.2 mM MgCl₂, 0.6 mM NaH₂PO₄, 2.4 mM Na₂HPO₄, 21.0 mM NaHCO₃, 10.0 mM glucose and 19.7 mM mannitol) was used for the measurements on colon epithelium.

Mouse ileum or colon was excised under hypnorm/diazepam anaesthesia and reverted on a plastic rod. The muscle layer was cut longitudinally using a blunt razor blade and was stripped of fat manually. The stripped tissue was mounted in a holder with the mucosal side up (exposed tissue area 0.2 or 0.5 cm²). After insertion of the holder into the Ussing chamber the tissue was allowed to recover for 10–20 min and to reach a stable baseline. The tissues were then short-circuited for the whole experimental period.

In case of ileum tissue specimens, glucose (10 mM) and indomethacin (10 μM) were added to the serosal side. After equilibrium, the following compounds were added consecutively to the mucosal (M) or serosal (S) side of the tissue: Forskolin (10 μM, S), genistein (100 μM, M+S) and carbachol (200 μM, S). All compounds were present throughout the experiment. Experiments on the colon were performed with a proximal and a distal specimen of colon ascendens in each wild type and CF mouse. After equilibration the following drugs were sequentially added: amiloride (100 μM, M), TEA (tetraethylammonium, 5 mM, M), Ba²⁺ (BaCl₂, 1 mM, M), DIDS (4,4'-diisothiocyanostilbene-2,2'-disulfonic acid, 200 μM, M) and carbachol (100 μM, S).

Mouse nasal epithelium was isolated as described by Grubb et al. 32. In brief, the mice were sacrificed by cervical dislocation and the skin of the head was peeled back in order to reach the underlying paired nasal bones. These were removed and the two sheets of the nasal epithelia, separated by the septum, were isolated independently. The sheets of epithelia were mounted immediately between the Ussing chambers (exposed area 1.13 mm²). The chambers were filled with gassed modified Meyler's solution supplemented with glucose (10 mM) and indomethacin (10 μM) and the short circuit measurements were started. After equilibration amiloride (10 μM, M) was added followed by forskolin (10 μM, S, after stabilization of the current).

Data are presented as means ± standard deviation (SD), n represents the number of mice. Unpaired t-test was used for comparisons between two groups. One-way ANOVA followed by a post hoc Tukey's test was applied for comparisons within one group. P-values < 0.05 were considered to be significant. The statistical analysis was performed using GraphPad Prism version 4.00 for Windows.