Fatal gastrointestinal obstruction and hypertension in mice lacking nitric oxide-sensitive guanylyl cyclase

Abstract

The signaling molecule nitric oxide (NO), first described as endothelium-derived relaxing factor (EDRF), acts as physiological activator of NO-sensitive guanylyl cyclase (NO-GC) in the cardiovascular, gastrointestinal, and nervous systems. Besides NO-GC, other NO targets have been proposed; however, their particular contribution still remains unclear. Here, we generated mice deficient for the beta1 subunit of NO-GC, which resulted in complete loss of the enzyme. GC-KO mice have a life span of 3-4 weeks but then die because of intestinal dysmotility; however, they can be rescued by feeding them a fiber-free diet. Apparently, NO-GC is absolutely vital for the maintenance of normal peristalsis of the gut. GC-KO mice show a pronounced increase in blood pressure, underlining the importance of NO in the regulation of smooth muscle tone in vivo. The lack of an NO effect on aortic relaxation and platelet aggregation confirms NO-GC as the only NO target regulating these two functions, excluding cGMP-independent mechanisms. Our knockout model completely disrupts the NO/cGMP signaling cascade and provides evidence for the unique role of NO-GC as NO receptor.

Fig. 1.

Gene inactivation of the β₁ subunit of NO-sensitive GC. (A) Targeted disruption of the β₁ subunit of NO-sensitive GC. loxP sites are depicted as white triangles, and exons (black squares) are numbered from the first coding exon. TK/Neo, thymidine kinase/neomycin cassette; DT, diphteria toxin gene; WT, wild-type loci of the β₁ subunit; S, SpeI; N, NheI; B, BamHI; brackets indicate destroyed restriction sites, and newly created ones are in italics. For more information, see SI. (B) Genotyping of mice. PCR and Southern blotting were performed as described in Materials and Methods. In the Southern blot, the β₁-KO band is larger than that of the WT because the NheI site in intron 10 was lost by the insertion of the TK/Neo cassette. (C) Western blot analysis of the expression of the β₁, α₁, and α₂ subunits of NO-GC in lung homogenates. (D) NO-stimulated cGMP-forming activity in brain and lung homogenates in the presence of diethylamine-NO (100 μM). Data are from one representative experiment. The cGMP-forming activity in KO tissues was zero under nonstimulated and NO-stimulated conditions. Because the deletion of the β₁ subunit results in deletion of NO-GC, we will henceforward use the term “GC-KO.”

Fig. 2.

Postnatal survival and appearance of mice deficient in NO-GC. (A) Postnatal survival of WT, heterozygous, and GC-KO mice. WT (n = 243) and heterozygous (β₁^−/+; n = 483) mice were observed for ≈4 weeks. The survival of GC-KO mice (n = 126) was monitored until death. Feeding with fiber-free diet improved the survival of GC-KO mice (GC-KO+diet; n = 58). (B) WT and GC-KO siblings at ≈21 days of age. Body weight (BW) ratio of WT (100%; n = 24) and GC-KO (n = 28) animals are shown (Right). Three-week-old siblings of the same sex were compared for BW determination.

Fig. 3.

Gastrointestinal tract and total gut transit time of WT and GC-KO mice. (A) The gastrointestinal tract of 3-week-old WT and GC-KO mice on regular diet. Arrows indicate the caecum, which in the GC-KO mouse is enlarged and displaced. (B) Comparison of esophagus (arrows) and stomach. (C) Lower gastrointestinal tract (caecum and parts of colon) of a 3-week-old GC-KO mouse immediately after dying. The arrows indicate perforations with adjoining hemorrhages. (D) Total gut transit time (see Materials and Methods). After weaning, mice were either fed fiber-containing standard diet (Left) or fiber-free diet (Right). In the case of standard diet, the measurement was discontinued for four KO mice (indicated by the asterisk) that did not produce red-colored feces even after 8 h. The lines indicate the respective median value. Statistical analysis is described in detail in Materials and Methods. Kruskal–Wallis global test was followed by Mann–Whitney U test for the comparison of two individual groups. The P values for the Mann–Whitney U test are given. ns, not significant.

Fig. 4.

Cardiovascular effects in mice lacking NO-GC. (A) Systolic blood pressure (SBP) was measured in conscious WT and GC-KO mice by tail-cuff plethysmography as described in Materials and Methods (Left, n = 7–8 for each genotype). Blood pressure data are given as mean ± SD. In a different set of experiments, mice (nine WT and six GC-KO of either sex) were anesthetized with ketamine/hydrazine. After i.p. injection of 5 mg/kg GTN, the subsequent change in blood pressure was recorded. Data represent mean ± SEM. (B) Representative original registration of aortic rings from WT and GC-KO mice. Rings were precontracted with 1 μM phenylephrine (PE). After reaching steady state, relaxation was induced by GSNO (10 and 100 μM). The membrane-permeable 8-pCPT-cGMP was added to confirm the integrity of the GC-KO rings and to show intact signaling beyond NO-GC. 8-pCPT-cGMP induced a similar relaxing response in WT rings (data not shown in this trace). (C) Relaxation curves for GSNO in aortic rings from WT and GC-KO mice (n = 5–8 per genotype). Error bars are smaller than the size of the symbols. (D) Relaxation curves for diethylamine-NO in aortic rings from WT and GC-KO mice (n = 5–8 per genotype).

Fig. 5.

NO does not inhibit aggregation of platelets from GC-KO mice. (A) Western blot analysis of the expression of the β₁, α₁, and α₂ subunits of NO-GC in WT and GC-KO platelets and, as positive control for the α₂ subunit, in WT brain. (B) Phosphorylation of PDE5 and vasodilator-associated phosphoprotein in intact WT and GC-KO platelets upon stimulation with GSNO (NO; 100 μM) and 8-pCPT-cGMP (8-cG; 40 μM). The lack of PDE5 phosphorylation by 8-pCPT-cGMP-activated PKG is explained by the fact that PDE5 requires bound cGMP to change its conformation for being phosphorylated. (C–F) Aggregation of platelets. Platelets from WT (C and E) and GC-KO mice (D and F) were stimulated with collagen (4 μg/ml; indicated by the arrow). To inhibit aggregation, the NO donor Proli-NO (4 mM; C and D) was administered simultaneously with collagen, and 8-pCPT-cGMP (40 μM; E and F) was added 20 min before collagen.