Female X-chromosome mosaicism for gp91phox expression diversifies leukocyte responses during endotoxemia

Abstract

Objective: To test the hypothesis, using an animal model, whether female X-chromosome mosaicism for inflammatory gene expression could contribute to the gender dimorphic response during the host response. X-chromosome-linked genetic polymorphisms present a unique biological condition because females display heterozygous cellular mosaicism, due to the fact that either the maternal or the paternal X chromosomes are inactivated in each individual cell in females. This is in contrast with the conditions in males who carry exclusively the maternal X chromosome.

Design: Prospective, randomized, laboratory investigation.

Settings: University research laboratory.

Subjects: Female mice deficient, heterozygous (mosaic) or WT for the X-linked gp91phox.

Interventions: We compared selected inflammatory markers among heterozygous (mosaics), WT and homozygous deficient animals in response to in vivo lipopolysaccharide (Escherichia coli, 20 mg/kg body weight). To test individual mosaic subpopulations of polymorphonuclear neutrophil responses, we also developed a flow cytometry assay that identifies the active parental X chromosomes in individual cells, using gp91phox expression as a marker.

Measurements and main results: Heterozygous mosaic mice presented white blood cell trafficking patterns similar to that observed in WT mice, despite the fact that the deficient subpopulation in mosaic animals displayed increased cell activation as reflected in elevated neutrophil CD11b expression and splenic infiltration. Mosaic animals also displayed splenic neutrophil infiltration, which was skewed toward the deficient subpopulation. Observations on splenic T-cell depletion and post lipopolysaccharide interleukin-10 responses indicated that the inflammatory response in mosaic animals does not simply display an average of the deficient and WT responses, but the mosaic subjects display a uniquely characteristic response.

Conclusions: The study supports the notion that female X chromosome mosaicism for polymorphic gene expression represents a unique condition, which may contribute to the gender dimorphic character of the inflammatory response. Mosaicism for X-linked polymorphisms may have clinical significance and needs consideration in genetic association or gender-related clinical studies.

Figure 1

Identification of mosaic cell subpopulations in heterozygous animals. A, Whole blood from gp91phox-deficient and WT animals was processed according to the BD phospho-lyse-fix-Perm-III buffer protocol. Samples were incubated with anti-CD11b-PERCP, as well as with either anti-gp91phox, or the corresponding isotype-IgG followed by incubation with PE-conjugated secondary antibody, as described in Materials and Methods. Staining intensity in gp91phox-deficient cells represents nonspecific background because it overlaps with the isotype-IgG staining. B, Myeloid cells from bone marrow (BM), blood, spleen, and peritoneal cavity from naive gp91phox heterozygous female mice were identified as described for part A. By the expression of gp91phox, two distinct cell populations carrying the active Xp or Xm can be readily identified in mosaic animals. C, Linear regression analysis of individual WT/deficient ratios BM and blood (circles) or BM and spleen (squares) from 11 animals. D, Heterozygous animals present two cell populations with different degree of oxidative burst after phorbol ester challenge. BM cells were preincubated with dihydrorodamine (DHR) for 20 mins followed by incubation in the presence or absence of 1μM of phorbol-myristate-acetate (PMA) for 20 mins. Subsequently, myeloid cells were gated and analyzed for DHR staining. A representative finding from a WT (left) and heterozygous mosaic animal (right) is shown. The moderate increase in PMA-induced fluorescence in the deficient mosaic subpopulation is most probably related to oxidant release from sources other than the nicotinamide adenine dinucleotide phosphate oxidase complex, which may include NOX isoenzymes or the mitochondria.

Figure 2

Lipopolysaccharide (LPS)-induced changes in bone marrow (BM) and blood cell composition in WT, gp91phox-deficient and mosaic animals. Twenty-four hours after LPS or saline controls, BM cells and blood were collected and processed for flow cytometry analyses, as described in Materials and Methods. Graphs A and B depict BM myeloid cell content identified by CD45/CD11b double positive (A) and B-cell content identified by CD45/CD19 double positive staining (B). Graphs C–F depict the numbers of circulating CD11b⁺ (mostly neutrophils) and different T-cell (CD4⁺ and CD8⁺) and B-cell (CD19⁺) subsets as indicated. Cell numbers were calculated from percent distribution of cells and total white blood cell yield. Statistically significant difference (p < .05): *compared with control within the same genotype; ^#compared with WT or heterozygous controls. Mean ± SEM, n = 7–8 animals in each group.

Figure 3

Lipopolysaccharide (LPS)-induced changes in splenic cell composition in WT, gp91phoxdeficient and mosaic animals. From control or LPS-injected mice, splenocyte suspensions were prepared and cell counts were determined. Based on CD marker staining and light scatter properties, the counts of splenic polymorphonuclear neutrophils (A), CD4⁺ T-cells (B), CD8⁺ T-cells (C), and CD19⁺ B-cells (D) were calculated from total cell yields and the percent distribution of respective cell populations. Statistically significant difference (p < .05): *compared with control within the same genotype; ^#compared with deficient in the same treatment group. Mean ± SEM, n = 7–8 animals in each group.

Figure 4

CD11b membrane expression is increased in the deficient polymorphonuclear neutrophilic leukocytes (PMN) subpopulations of mosaic animals and in spleen-infiltrated PMNs. CD11b staining (mean fluorescence intensity [MFI]) was measured in circulating (A) and splenic (B) neutrophils 24 hrs after LPS and compared among WT, deficient and mosaic animals. Graph C depicts CD11b expression levels in mosaic subpopulations identified by gp91phox expression, or lack of, as described for Figure 1. Statistically significant difference (p < .05): *compared with control within the same genotype; ^#compared with WT in the same treatment group; ^&compared with WT (paired comparison). Mean ± SEM, n = 7–8 animals in each group.

Figure 5

Gp91phox-deficient mosaic subpopulations dominate tissue cell composition changes after endotoxin. WT/deficient polymorphonuclear neutrophilic leukocytes (PMN) ratios were calculated, using the method described for Figure 1. Graphs A–C depict ratios in bone marrow (BM), blood, and spleen. Graph D presents the resident as well as the lipopolysaccharide (LPS)-induced newly infiltrated numbers of splenic PMNs, which were calculated from the WT/deficient (def) ratios combined with the cell infiltration values presented in Figure 3. *Statistically significant difference (p < .05) compared with control. Mean ± SEM, n = 7–8 animals in each group.

Figure 6

Heterozygous mosaic animals present robust interleukin (IL)-10 response post lipopolysaccharide (LPS). Blood IL-10 content was measured in plasma samples collected 24 hrs after LPS, using a commercial enzyme-linked immunosorbent assay kit. *Statistically significant difference (p < .05) compared with WT. Mean ± SEM, n = 6–7 animals in each group.