Chronic murine typhoid fever is a natural model of secondary hemophagocytic lymphohistiocytosis

Abstract

Hemophagocytic lymphohistiocytosis (HLH) is a hyper-inflammatory clinical syndrome associated with neoplastic disorders especially lymphoma, autoimmune conditions, and infectious agents including bacteria, viruses, protozoa and fungi. In both human and veterinary medicine, hemophagocytic histiocytic disorders are clinically important and frequently fatal. HLH in humans can be a primary (familial, autosomal recessive) or secondary (acquired) condition, with both types generally precipitated by an infectious agent. Previously, no mouse model for secondary HLH has been reported. Using Salmonella enterica serotype Typhimurium by oral gavage to mimic naturally-occurring infection in Sv129S6 mice, we characterized the clinical, hematologic and morphologic host responses to disease thereby describing an animal model with the clinico-pathologic features of secondary HLH as set forth by the Histiocyte Society: fever, splenomegaly, cytopenias (anemia, thrombocytopenia), hemophagocytosis in bone marrow and spleen, hyperferritinemia, and hypofibrinogenemia. Disease severity correlates with high splenic and hepatic bacterial load, and we show disease course can be monitored and tracked in live animals. Whereby secondary HLH is known to occur in human patients with typhoid fever and other infectious diseases, our characterization of a viable natural disease model of secondary HLH offers an important means to elucidate pathogenesis of poorly understood mechanisms of secondary HLH and investigation of novel therapies. We characterize previously unreported secondary HLH in a chronic mouse model of typhoid fever, and novel changes in hematology including decreased tissue ferric iron storage that differs from classically described anemia of chronic disease. Our studies demonstrate S. Typhimurium infection of mice is a natural infectious disease model of secondary HLH that may have utility for elucidating disease pathogenesis and developing novel therapies.

Figure 1. Hematology of S. Typhimurium-infected mice: acute, then chronic active inflammatory response; microcytic anemia, persistent microcytosis.

Mice were orally gavaged with 9.1×10⁸ CFU of S. Typhimurium (n = 8) or sterile PBS (n = 7). Complete blood counts were monitored over 16 weeks. X  =  S. Typhimurium-infected mice; circle  =  mock-infected control mice. Mean and standard deviation are shown. (A) neutrophils, (B) monocytes, (C) lymphocytes, (D) hematocrit (HCT), (E) mean cell volume (MCV). *P<0.05 (Student's t-test).

Figure 2. S. Typhimurium-infected mice have increased hemophagocytic macrophages, myeloid hyperplasia, regenerative microcytic anemia and erythrocyte fragmentation.

(A) Control bone marrow cytology, 3 weeks post mock-infection. (B) Infected mouse bone marrow cytology, 3 weeks post-infection; myeloid hyperplasia with increased blasts and monocytes, hemophagocytic macrophage (white arrow), and foamy macrophage (black arrow). (C) Bone marrow histology, 6 weeks post-infection; hypercellularity with myeloid hyperplasia. (D) Mouse bone marrow cytology, 3 weeks post-infection; erythrocyte (arrow and inset) within hemophagocytic macrophage. (E) Control mouse blood film, 3 weeks post mock-infection. (F) Blood film from highly infected mouse, 3 weeks post-infection; marked erythrocyte anisocytosis, increased polychromasia and markedly fragmented erythrocytes (mature and polychromatophilic). Polychromatophilic erythrocytes (arrow-heads), fragmented polychromatophilic erythrocytes (carats), fragmented erythrocytes (arrows), L = small lymphocyte. Wright stain (A, B, D, E, F). H & E stain (C). Original magnifications 500× (A–B, D), 200× (C), or 1000× (E–F).

Figure 3. S. Typhimurium-infected mice have tissue inflammation and thrombosis, increased hematopoiesis, and decreased splenic iron.

(A) Mouse liver, 6 weeks post-infection; inflammation and necrosis (arrow). (B) Mouse spleen, 3 weeks post-infection; extramedullary hematopoiesis (EMH; arrow, megakaryocytes), histiocytic infiltration (I) throughout the red pulp, and thrombus (T). H&E stain (A, B). (C) Spleen, mock-infected (left) and infected mouse (right), 3 weeks post-infection; markedly decreased ferric iron staining in red pulp. (D) Spleen, mock-infected (left) and infected mouse (right), 6 weeks post-infection; markedly decreased splenic ferric iron in red pulp. Perl's Prussian Blue stain (C, D). (E) Hemophagocytic macrophage in mouse spleen 3 weeks post-infection that had 10-fold more macrophages and 43-fold more 6N+ macrophages than control mouse spleen. CD11b (red), DAPI (blue), TER119 (green). N  =  endogenous macrophage nucleus, E1  =  nucleated erythrocyte, E2  =  non-nucleated erythrocyte. Confocal fluorescent micrograph. (F) Representative histogram overlay of TER119 expression on DAPI+ splenocytes from a mock-infected (red) and infected mouse (blue) 3 weeks post-infection. Filled gray histogram corresponds to the isotype control. The infected mouse had 11.5-fold more TER119^med pro-erythroblasts and 5.5-fold more TER119^high erythroblasts than the mock-infected mouse. (G) Mean numbers of TER119^med and TER119^high splenocytes from three mock-infected (white bars) and four infected (gray bars) mice. Mean number of TER119^med pro-erythroblasts per spleen increased 6.8-fold in infected mice, while the mean number of TER119^high cells, corresponding to all nucleated erythroblasts subsequent to the pro-erythroblast stage , increased 3.6-fold. (P<0.05) Error bars = SD. Original magnifications 100× (A–B), 200× (C–D), and 1000× (E).