Janus Kinase Inhibitors Ameliorated Gastrointestinal Amyloidosis and Hypoalbuminemia in Persistent Dermatitis Mouse Model

Abstract

Malnutrition is not only regarded as a complication of rheumatoid arthritis and inflammatory bowel disease but also that of inflammatory skin disease; however, the mechanisms and efficacy of its treatment have not been elucidated. Using a mouse model of dermatitis, we investigated the pathophysiology of malnutrition in inflammatory skin conditions and efficacy of its treatment. We employed spontaneous skin inflammation mice models overexpressing human caspase-1 in the epidermal keratinocytes. Body weight, nutrition level, and α1-antitrypsin fecal concentration were measured. The gastrointestinal tract was histologically and functionally investigated. Fluorescein isothiocyanate (FITC)-dextran was forcibly fed on an empty stomach, and plasma FITC-dextran was measured. The treatment efficacy of antibodies against tumor necrosis factor-α (TNF-α) and interleukin (IL)-α/β as well as Janus kinase (JAK) inhibitors was investigated. Compared with wild-type littermates, the inflammatory skin mice models showed a lowered body weight, reduction of serum albumin level, amyloid deposition in the stomach, small intestine, and large intestine, and increased α1-antitrypsin fecal concentration. However, the plasma FITC-dextran was unchanged between the dermatitis models and wild-type littermates. The over-produced serum amyloid A1 in the liver was detected in the plasma in the dermatitis model. Antibodies against TNF-α and IL-α/β showed partial effects on amyloid deposition; however, JAK inhibitors improved gastrointestinal amyloidosis with the improvement of skin symptoms. Chronic dermatitis is closely related to secondary amyloidosis in the gastrointestinal tract, resulting in hypoalbuminemia. Therefore, active control of skin inflammation is essential for preventing gastrointestinal complications.

Keywords: JAK inhibitor; absorption; amyloidosis; cytokine; dermatitis; emaciation; gastro-intestinal tract; hypoalbuminemia; inflammatory skin mouse model; nutrition.

Figure 1

KCASP1Tg mice showed weight loss and hypoalbuminemia. (a) Body weight of the KCASP1Tg mice decreased significantly more than that of the WT littermates from 12 weeks of age. (c) TP was not significantly different between all the groups; however, Alb decreased from 10 weeks of age in the KCASP1Tg mice (b,d). WT; wild-type C57BL/6, KCASP1Tg; K14/caspase-1 transgenic mouse. Values shown as means ± SDs and statistically significant differences (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001) are indicated.

Figure 2

Histological analysis showed the amyloid deposition in KCASP1Tg mice. At 16 weeks of age, the stomach, small intestine, and large intestine were collected from the WT and KCASP1Tg mice and stained with Congo red. Compared with the WT mice, the KCASP1Tg mice had more severe amyloid deposition in the gastric mucosa and intestinal epithelial cells. Scale bar = 50 μm. There was a significant increase in the Congo red stain-positive area in the KCASP1Tg mice. All data were expressed as means ± SDs. *** p < 0.001 between KCASP1Tg and wild-type mice by Mann–Whitney test.

Figure 3

Gastrointestinal tract barrier function. (a) α1-antitrypsin in feces was significantly more in the KCASP1Tg mice than in the WT littermates (n = 6 per group). (b) However, intestinal permeability was not considered to be enhanced after evaluating the plasma concentration of FITC-dextran in the KCASP1Tg mice (n = 6 per group). All data were expressed as means ± SDs. ** p < 0.01 between KCASP1Tg and wild mice by Mann–Whitney test.

Figure 4

SAA levels in organs and plasma. (a) The mRNA purified from various organs in the KCASP1Tg and WT mice revealed that there was a higher production of SAA1 and SAA2 in the liver of the KCASP1Tg than in the WT mice. (b) An increased level of SAA3 was detected in the skin of the KCASP1Tg mice. (c,d) The plasma levels of SAA1 were higher in the KCASP1Tg mice than in the WT littermates; however, the plasma levels of SAA3 were similar between the two groups. All data are expressed as means ± SDs. *** p < 0.001, **** p < 0.0001 compared to wild mice by Mann–Whitney test.

Figure 5

The effect of anti-TNF-α and IL-1α/β in the KCASP1Tg mice. KCASP1Tg mice were treated with neutralizing antibodies against TNF-α or IL-1α/β to prevent gastrointestinal amyloidosis, and the Congo red stain-positive area was partially ameliorated in the stomach and small intestine. Scale bar = 50 μm. * p < 0.05 compared to wild mice by Kruskal–Wallis test.

Figure 6

The effect of administration of JAK inhibitors in the KCASP1Tg mice. KCASP1Tg mice were treated with baricitinib or cerdulatinib (baricitinib and cerdulatinib, respectively). The clinical phenotype was ameliorated by both JAK inhibitors (a). Plasma Alb levels were measured from 10 to 16 weeks of age. Plasma Alb was recovered in baricitinib- and cerdulatinib-treated mice (b). At 16 weeks of age, the stomach, small intestine, and large intestine were collected and stained with Congo red. Amyloid deposition was reduced in the gastric mucosa and intestinal epithelial cells of the baricitinib- and cerdulatinib-treated KCASP1Tg mice compared with those of the non-treated KCASP1Tg mice (c). Scale bar = 100 μm. * p < 0.05, ** p < 0.01, *** p < 0.001 compared to KCASP1Tg mice by Kruskal–Wallis test.