Intestinal Na⁺ Loss and Volume Depletion in JAK3-Deficient Mice

JAK3 (Janus kinase 3), a tyrosine kinase, contributes to the signaling of hematopoietic cell cytokine receptors [1,2,3,4,5]. The kinase further participates in the response to energy depletion, hypoxia and ischemia-reperfusion [6,7,8,9]. JAK3 stimulation fosters lymphocyte proliferation and suppresses lymphocyte apoptosis [10,11,12]. Along those lines, JAK3 inhibitors trigger lymphocyte apoptosis [13,14]. In contrast, JAK3 activation fosters dendritic cell apoptosis [15]. A gain of function mutation of JAK3 (^A572VJAK3) [16] is associated with acute megakaryoplastic leukemia [17,18].

JAK3 deficiency has been shown to trigger inflammatory bowel disease [19], which is in turn known to be associated with downregulation of several Na⁺-transporting proteins [20,21] including the colonic epithelial Na⁺ channel ENaC [22,23,24,25,26]. Downregulation of ENaC in inflammatory bowel disease has been attributed to transcriptional suppression of the channel protein by proinflammatory cytokines such as tumor necrosis factor (TNF)-alpha [25]. Moreover, ENaC activity is decreased in macroscopically non-inflamed colon from patients with Crohn´s disease [26]. As ENaC is important for colonic Na⁺ reabsorption [26,27], the decreased ENaC activity contributes to diarrhea in those patients [26]. Moreover, decreased ENaC transcription and activity contributes to the diarrhea of ulcerative colitis [22,23]. Decreased ENaC activity is also involved in the diarrhea of interleukin-2-deficient mice [24].

At least in theory, ENaC activity and extracellular volume regulation may be similarly compromized in mice lacking JAK3. However, ENaC activity and Na⁺ metabolism of those mice have never been studied. The present study thus explored whether JAK3 deficiency influences colonic ENaC activity. To this end, experiments were performed in gene-targeted mice lacking functional JAK3 (jak3^-/-) and in age- and sex-matched wild type mice (jak3^+/+).

Experiments were performed in female and male gene-targeted mice lacking functional JAK3 (jak3^-/-) and in corresponding wild type mice (jak3^+/+) at the age of 12-16 weeks. The mice had free access to water and control food (Ssniff, Soest, Germany) [28].

All animal experiments were conducted according to the German law for the welfare of animals and the surgical procedures on the mice were reviewed and approved by the respective government authority of the state Baden-Württemberg (Regierungspräsidium) prior to the start of the study.

Systolic arterial blood pressure was determined utilizing the non-invasive tail-cuff method (IITC Life Science, Woodlands Hills, CA) [29]. As reviewed earlier [30], the tail cuff approach to determine arterial blood pressure requires certain precautions to reduce the stress of the animals, including appropriate training of the mice over multiple days, prewarming to an ambient temperature of 34°C, as well as measurement in a quiet and semidarkend environment at a defined day time. All these precautions were taken in the present study. Readings of two days were averaged to obtain the systolic blood pressure for the respective mouse. All recordings and the data analysis were obtained by utilizing a computerized data acquisition system and software (PowerLab 4/26 and chart5; ADIstruments).

For evaluation of renal excretion, mice were placed individually in metabolic cages (Techniplast, Hohenpeissenberg, Germany) for 24 hours urine collection as described previously [31,32]. The mice were maintained on a standard diet and had free access to tap water before the experiment. They were allowed a two day habituation period. Subsequently, urine and feces were collected daily for four days. To assure quantitative urine collection, metabolic cages were siliconized, and urine was collected under water-saturated oil. Feces were collected in separated tubes.

To collect blood specimen, animals were lightly anesthetized and about 50 - 200 μl of blood was collected into serum tubes by puncturing the retro-orbital plexus [33]. The serum aldosterone concentration was determined using a commercial ELISA kit (Alpha Diagnostics International, Texas; USA). Serum and urinary concentrations of Na⁺ were measured by flame photometry (ELEX 6361, Eppendorf, Hamburg, Germany). Serum and urinary creatinine concentrations were assessed utilizing an enzymatic colorimetic method (Labor & Technik, Berlin, Germany). Fecal dry weight was obtained by drying the collected sample at 80°C for three hours. The fecal samples were prepared for determination of electrolyte content by dissolving in nitric acid (0.75 M HNO₃) and 48 hours at 50°C with continuous shaking. The homogenized samples were centrifuged at 3,500 g for 10 min and 1 ml of the supernatants were again centrifuged at 10,000 g for 5 min. Aliquots from the second supernatants were diluted and the Na⁺ content of the supernatant was determined by flame photometry. The measured electrolyte concentrations were calculated to obtain the fecal sodium excretion in µmol per g of feces excreted within 24 hours.

ENaC activity was estimated from the amiloride-sensitive current across the colonic epithelium. After removing the outer serosal and muscular layer of the late distal colon under a microscope, the tissue was placed onto a custom-made mini-Ussing chamber with an opening area of 0.00769 cm². Transepithelial potential difference (Vte) was determined continuously and transepithelial resistance (Rte) estimated from the voltage deflections (ΔVte) elicited by imposing rectangular test currents of 1 µA and 1.2-s duration at a rate of 8/min. The serosal and luminal perfusate contained (in mM) 145 NaCl, 1 MgCl, 2.6 Ca-gluconate, 0.4KH₂ PO₄, 1.6 K₂HPO₄, 5 glucose. To assess ENaC-mediated current, 50 μM amiloride (in DMSO; Sigma, Schnelldorf, Germany) was added to the luminal perfusate.

Data are provided as means ± SEM, n represents the number of independent experiments. All data were tested for significance using unpaired Student t-test. Only results with p < 0.05 were considered statistically significant.

Ussing chamber experiments were performed in order to study colonic ENaC activity in JAK3-deficient mice (jak3^-/-) and in wild type mice (jak3^+/+). ENaC activity was estimated from the effect of ENaC blocker amiloride (50 µM) on the transepithelial potential difference and current. As illustrated in Fig. 1, amiloride application was followed by a lumen-positive voltage deflection and a change of current, which was significantly lower in jak3^-/- mice than in jak3^+/+ mice.

Decreased ENaC activity is expected to compromize colonic Na⁺ reabsorption. Thus, fecal Na⁺ excretion was determined. As shown in Fig. 2, the fecal excretion of Na⁺ was higher in jak3^-/- mice than in jak3^+/+ mice.

Intestinal Na⁺ loss could result in extracellular volume depletion with decrease of blood pressure and increase in aldosterone release. Thus, blood pressure was determined utilizing the tail cuff method. As illustrated in Fig. 3A, the systolic arterial blood pressure was significantly lower in jak3^-/- mice than in jak3^+/+ mice. According to ELISA, plasma aldosterone concentration was significantly higher in jak3^-/- mice than in jak3^+/+ mice (Fig. 3B).

The secondary hyperaldosteronism is expected to impact on renal tubular Na⁺ reabsorption and thus on urinary Na⁺ exretion. We therefore analyzed jak3^-/- mice and jak3^+/+ mice in metabolic cages. As illustrated in Fig. 4, both absolute (Fig. 4A) and fractional (Fig. 4B) urinary Na⁺ excretion were lower in jak3^-/- mice than in jak3^+/+ mice.

The present study reveals that JAK3 deficiency was followed by a marked decrease of amiloride-sensitive current reflecting ENaC activity in colonic epithelium. Accordingly, the amiloride-induced voltage deflection was significantly smaller in gene-targeted mice lacking functional JAK3 (jak3^-/-) than in wild type mice (jak3^+/+). As JAK3 deficiency leads to inflammatory bowel disease [19], which is in turn known to decrease ENaC activity [22,23,24,25,26], the inflammatory bowel disease of those animals presumably contributes to or even accounts for the decline of ENaC activity. The present observations do, however, not rule out additional, more direct effects of JAK3 on ENaC. Possibly, further transport systems [20,21] are similarly impaired in jak3^-/- mice.

The impaired intestinal Na⁺ absorption led to enhanced fecal Na⁺ output. Apparently, the intestinal Na⁺ loss resulted in extracellular volume contraction with decrease of blood pressure and increase in aldosterone release. In inflammatory bowel disease the effect of aldosterone on intestinal ENaC activity is blunted or even completely lost [22]. Thus, aldosterone release could not restore colonic Na⁺ absorption. However, the increased aldosterone release contributed to or even accounted for the enhanced renal tubular Na⁺ reabsorption resulting in marked antinatriuresis. Again, we cannot rule out that additional JAK3-dependent mechanisms contribute to the antinatriuresis of JAK3 deficient mice.

At least in theory, JAK3 mutations could similarly impact on Na⁺ metabolism in humans. Gain of function JAK3 mutations may lead to malignancy [34,35] and loss of function JAK3 mutations are associated with severe immune deficiency [5,35,36]. Moreover, pharmacological JAK3 inhibition could, similar to genetic knockout of JAK3, impact on intestinal Na⁺ absorption and extracellular volume homeostasis. JAK3 inhibitors are considered for the treatment of malignancy as well as several inflammatory and immunological diseases, such as rheumatoid arthritis, psoriasis, ulcerative colitis and dry eye disease [5,35,37,38]. Moreover, JAK3 inhibitors are considered for treatment of kidney transplant patients [37,39]. It remains to be shown, whether application of those inhibitors affects Na⁺ metabolism.

JAK3 deficiency led to decreased activity of ENaC and possibly further intestinal and colonic transport systems. The fecal Na⁺ loss presumably caused extracellular fluid volume depletion with decrease of blood pressure, secondary hyperaldosteronism and renal Na⁺ retention.

The authors acknowledge the technical support by Elfriede Faber. This study was supported by the Deutsche Forschungsgemeinschaft (FO 695/1-1) and the Open Access Publishing Fund of Tuebingen University.