Disruption of the meprin alpha and beta genes in mice alters homeostasis of monocytes and natural killer cells

Abstract

Meprin metalloproteases are implicated in inflammatory bowel disease, which involves dysfunction of immune cells. However, the roles of meprins in the immune and hematological system remain uncharacterized. In this report, we demonstrate that meprins were expressed in the hematological system, and meprin alpha/beta null (alpha(-/-)/beta(-/-)) mice had decreased prevalence of resident monocytes and natural killer (NK) cells in blood, with a concomitant accumulation of inflammatory monocytes and NK cells in bone marrow. In contrast, T and B lymphocytes were not affected by meprin deficiency. In response to acute inflammation induced by intraperitoneal injection of thioglycollate, meprin-deficient mice exhibited higher body temperature than wild-type mice, which was correlated with retention of inflammatory monocytes, but persistent low prevalence of NK cells in blood. These results indicate that meprin metalloproteases play important roles in the homeostasis of monocytes and NK cells, and possibly are involved in egress of these two type cells from bone marrow and homing to the periphery. Our findings are the first report to demonstrate that metalloproteases affect homeostasis of leukocytes, which have important implications for understanding physiology of and pathogenesis in the hematological system.

Fig. 1. Expression of meprin mRNA in the hematological system

RNA from blood and bone marrow mononuclear leukocytes was examined by RT-PCR for meprin expression. (A) Meprin alpha and beta mRNA were detected in mouse peripheral blood mononuclear cells (PBMC) and bone marrow mononuclear cells (BMMC), but not in skeletal muscle. (B) Human PBMC but not fibroblasts (SF) expressed meprin alpha and beta mRNA. N.Ctrl, no template control.

Fig. 2. Disruption of meprin genes in dKO mice

(A). DNA was extracted from bone marrow of the experimental mice and examined by PCR with primers flanking the insertion site of the neo gene in MEP1A (top) and MEP1B (bottom) genes. (*) indicates disrupted meprin gene fragments. N.Ctrl: no template negative control. P.Ctrl: positive control with DNA from meprin heterozygous (alpha^+/-/beta^+/-) mice. (B). Homogenates of kidney and skeletal muscle from dWT and dKO mice were examined by Western blot analysis for meprin expression. Pos. Ctrl: purified recombinant meprin proteins as positive controls.

Fig. 3. dKO, but not dWT, mice manifested a febrile reaction in response to acute peritoneal inflammation induced by TG

Anal temperature was measured for mice non-treated (A) or 24 hours after i.p. injection of TG (B). Data shown are from 2 independent experiments with a total of 12 pairs of mice.

Fig. 4. In comparison to dWT mice, dKO mice had low R-MC (CD11b⁺/Ly6c^-) in blood, but high I-MC (CD11b⁺/Ly6c⁺⁺) in bone marrow

Peripheral blood and bone marrow cells from dKO and dWT mice were stained with CD11b-FITC, Ly6c-PE and CD45-PerCP, followed by flow cytometry. Monocytes with the phenotype CD11b⁺/Ly6c^- are defined as resident monocytes (R-MC), and those of CD11b⁺/Ly6c⁺⁺ as inflammatory monocytes (I-MC). Representative dot plots (A-L) are to show cell populations compared in M. Darker events are the gated cells in CD11b/Ly6c plots. Granulocytes were excluded by gating in FSC/SSC plots. Data were from 2 independent experiments with a total of 7 mice in each group.

Fig. 5. Blood CD11b⁺/Ly6c⁺⁺ I-MC were enriched with CD49b⁺ cells

Peripheral blood from dKO and dWT mice was stained with CD11b-FITC, Ly6c-PE, CD45-PerCP and CD49b-APC, followed by flow cytometry. Representative dot plots (A, B) are to show cell populations compared in C. Events in A and B are CD11b⁺/Ly6c⁺⁺ and CD11b⁺/Ly6c^- cells, respectively. Granulocytes were excluded by gating in FSC/SSC plots (not shown). Data were from 2 independent experiments with a total of 7 mice in each group.

Fig. 6. In comparison to dWT mice, dKO mice had low CD14⁺/Ly6c^- R-MC in blood, but high CD14⁺/Ly6c⁺⁺ I-MC in bone marrow

Peripheral blood and bone marrow cells were stained with CD14-FITC, Ly6c-PE and CD45-PerCP, followed by flow cytometry. Representative dot plots (A-F) are to show cell populations compared in G. Darker events are the gated cells in CD14/Ly6c plots. Granulocytes were excluded by gating in the FSC/SSC plots (not shown). Data were from 2 independent experiments with a total of 7 mice in each group.

Fig. 7. In comparison to dWT mice, blood NK cells but not T and B lymphocytes were low in dKO mice, with simultaneous accumulation of NK cells in bone marrow

Blood and bone marrow cells from dKO and dWT mice were stained with antibodies against CD49b, NK1.1, CD11b and CD45 (A), or CD90 (blood only) or CD3e, CD19 and CD45 (B, C), followed by flow cytometry. Granulocytes were excluded by gating in FSC/SSC plots (not shown). Data were from 2 independent experiments with a total of 7 mice in each group.

Fig. 8. Meprin deficiency was correlated with distorted monocyte response to acute inflammation

Flow cytometry analysis of blood and bone marrow cells from dKO and dWT mice 24 hours after i.p. injection of TG for the relative prevalence of monocytes, NK cells, B cells and T cells. Note that dKO mice exhibited a higher level of blood I-MC and lower level of bone marrow R-MC than dWT mice. Data were from 2 independent experiments with a total of 12 mice in each group.

Fig. 9. Meprin deficiency differentially affected the absolute prevalence of blood leukocytes

Flow cytometry analysis of blood from dKO and dWT mice 24 hours after i.p. injection of TG for the absolute prevalence of monocytes, NK cells, B cells, T cells and granulocytes. The absolute leukocyte numbers were corrected with Fluorospheres as internal reference. Note that dKO mice exhibited a higher level of blood I-MC than their WT counterparts. Data were from 2 independent experiments with a total of 12 mice in each group.