NADPH-oxidase 4 gene over-expression in peripheral blood lymphocytes of the schizophrenia patients

Abstract

Introduction: Increased systemic oxidative stress is common in schizophrenia (SZ) patients. NADPH-oxidase 4 (NOX4) is the cell oxidoreductase, catalyzing the hydrogen peroxide formation. Presumably, NOX4 is the main oxidative stress factor in a number of diseases such as cardiovascular diseases and cancer. We hypothesized that NOX4 may be involved in the oxidative stress development caused by the disease in the schizophrenic patients' peripheral blood lymphocytes (PBL).

Materials and methods: The SZ group included 100 patients (68 men and 32 women aged 28 ± 11 years). The control group included 60 volunteers (35 men and 25 women aged 25 ± 12 years). Flow cytometry analysis (FCA) was used for DNA damage markers (8-oxodG, ɣH2AX), pro- and antiapoptotic proteins (BAX1 and BCL2) and the master-regulator of anti-oxidant response NRF2 detection in the lymphocytes of the untreated SZ patients (N = 100) and the healthy control (HC, N = 60). FCA and RT-qPCR were used for NOX4 and RNANOX4 detection in the lymphocytes. RT-qPCR was used for mtDNA quantitation in peripheral blood mononuclear cells. Cell-free DNA concentration was determined in blood plasma fluorimetrically.

Results: 8-oxodG, NOX4, and BCL2 levels in the PBL in the SZ group were higher than those in the HC group (p < 0.001). ɣH2AX protein level was increased in the subgroup with high 8-oxodG (p<0.02) levels and decreased in the subgroup with low 8-oxodG (p <0.0001) levels. A positive correlation was found between 8-oxodG, ɣH2AX and BAX1 levels in the SZ group (p <10-6). NOX4 level in lymphocytes did not depend on the DNA damage markers values and BAX1 and BCL2 proteins levels. In 15% of PBL of the HC group a small cellular subfraction was found (5-12% of the total lymphocyte pool) with high DNA damage level and elevated BAX1 protein level. The number of such cells was maximal in PBL samples with low NOX4 protein levels.

Conclusion: Significant NOX4 gene expression was found a in SZ patients' lymphocytes, but the corresponding protein is probably not a cause of the DNA damage.

Fig 1. NOX4 levels in the PBL of the SZ and HC groups.

A. The most typical examples of the NOX4 analysis in lymphocytes. B. Cumulative distribution of FL1- NOX4 for 63 randomly selected HC (green) and SZ (red) samples in a one experiment. The dashed curve is the background curve for the sample with the maximum background level. C. NOX4 levels in the lymphocyte of the SZ and HC groups. The average level of marker in the cell population was estimated by the FL-signal median value. In the graphs, each point is the mean for three FL-signal median values (three parallel measurements for the same PBMC sample). The median values were normalized to the maximum signal medians values in the analyzed total sample (N = 160). The relative standard error was 7 ± 3%. The significance of the observed differences was analyzed using non-parametric Mann–Whitney (p) test. D. RNANOX4 levels determined in lymphocytes of 30 SZ patients and 30 HC. E. ROC curve for the SZ/HC groups. Each point on the ROC curve represents a sensitivity/specificity pair. The area under the ROC curve (AUC) is a measure of how well a parameter can distinguish between two groups (SZ/HC).

Fig 2. 8-oxodG levels in the PBL of the SZ group and HC group.

A. The most typical examples of the 8-oxodG analysis in lymphocytes. B. Cumulative distribution of FL1-8-oxodG for 103 randomly selected HC (green) and SZ (red) samples in a single experiment. The dashed curve is the background curve for the sample with the maximum background level. C. 8-oxodG levels in the lymphocyte of the SZ- and HC groups. The average level of marker in the cell population was estimated by the FL-signal median value. In the graphs, each point is the mean for three FL-signal median values (three parallel measurements for the same PBMC sample). The median values were normalized to the maximum signal medians values in the analyzed total sample (N = 160). The relative standard error was 7 ±2%. The significance of the observed differences was analyzed using non-parametric Mann–Whitney (p) test. D. ROC curve. Each point on the ROC curve represents a sensitivity/specificity pair. The area under the ROC curve (AUC) is a measure of how well a parameter can distinguish between two groups (SZ/HC). E. The dependence of 8-oxodG on NOX4. Dotted lines indicate areas of points included in the SZ subgroups. F. (a) MtDNA CN in the HC and SZ groups. (b) Ratio 8-oxodG/mtDNA CN for the HC and SZ groups. G. (a) Cumulative distribution of mtDNA CN for HC and SZ subgroups. (b) Cumulative distribution of ratio 8-oxodG/mtDNA CN for HC and SZ subgroups. Arrows indicate compared subgroups. The significance of the observed differences was analyzed using the non-parametric Mann–Whitney test (p) and Kolmogorov–Smirnov test (D and α).

Fig 3. ɣH2AX levels in the lymphocyte of the SZ group and HC group.

A. The most typical examples of the ɣH2AX analysis in lymphocytes. B. Cumulative distribution of FL1- ɣH2AX for 60 randomly selected HC (green) and SZ (red) samples in a one experiment. The dashed curve is the background curve for the sample with the maximum background level. C. ɣH2AX levels in the lymphocyte of the SZ- and HC groups. The average level of marker in the cell population was estimated by the FL-signal median value. In the graphs, each point is the mean for three FL-signal median values (three parallel measurements for the same PBMC sample). The median values were normalized to the maximum signal medians values in the analyzed total sample (N = 160). The relative standard error was 5 ± 3%. The significance of the observed differences was analyzed using non-parametric Mann–Whitney (p) test. D. ɣH2AX(R) levels in the lymphocyte of the SZ- and HC groups. The significance of the observed differences was analyzed using non-parametric Mann–Whitney (p) test. E. The dependence of ɣH2AX on 8-oxodG. Linear regression data is shown at the top left of the graph. At the bottom right of the graph, italics show the correlation analysis data for the SZ subgroups and HC group. F. (a) Cumulative distribution of ɣH2AX for HC and SZ subgroups. (b) Cumulative distribution of ɣH2AX(R) for HC and SZ subgroups. Arrows indicate compared subgroups. The significance of the observed differences was analyzed using the non-parametric Mann–Whitney test (p) and Kolmogorov–Smirnov test (D and α).

Fig 4. BAX and BCL2 levels in the lymphocyte of the SZ and HC groups.

A. The most typical examples of the BAX analysis in lymphocytes. B. Cumulative distribution of FL1- BAX for 80 randomly selected HC (green) and SZ (red) samples in a one experiment. C. BAX levels in the lymphocyte of the SZ and HC groups. (a) The average level of marker in the cell population was estimated by the FL-signal median value. In the graphs, each point is the mean for three FL-signal median values (three parallel measurements for the same PBMC sample). The median values were normalized to the maximum signal medians values in the analyzed total sample (N = 160). The relative standard error was 8 ± 1%. The significance of the observed differences was analyzed using non-parametric Mann–Whitney (p) test. (b) Cumulative distribution of BAX for HC and SZ subgroups. D. The most typical examples of the BCL2 analysis in lymphocytes. E. Cumulative distribution of FL1- BAX for 60 randomly selected HC (green) and SZ (red) samples in a one experiment. The dashed curve is the background curve for the sample with the maximum background level. F. (a) BCL2 levels in the lymphocyte of the SZ and HC groups. The average level of BCL2 in the cell population was estimated by the FL-signal median value. In the graphs, each point is the mean for three FL-signal median values (three parallel measurements for the same PBMC sample). The relative standard error was 4 ± 2%. (b) Cumulative distribution of BCL2 for HC and SZ subgroups. G. (a) Ratio BAX/BCL2 for the SZ and HC groups. (b) Cumulative distribution of BAX/BCL2 for HC and SZ subgroups. Arrows indicate compared subgroups. The significance of the observed differences was analyzed using the non-parametric Mann–Whitney test (p) and Kolmogorov–Smirnov test (D and α).

Fig 5. Concentration of the circulating cell-free DNA (cfDNA) in blood plasma.

A. Concentration of the cfDNA in SZ- and HC groups. B. Cumulative distribution cfDNA for HC and SZ subgroups. The significance of the observed differences was analyzed using the non-parametric Mann–Whitney test (p) and Kolmogorov–Smirnov test (D and α). C. Dependence of cfDNA concentration in blood plasma on the size of lymphocyte fraction γH2AX (R) with damaged DNA. D. Scheme illustrating the change in the ratio of the size of the cell fraction γH2AX (R) and the concentration of cfDNA.

Fig 6. NRF2 levels in the lymphocyte of the SZ and HC groups.

A. The most typical examples of the NRF2 analysis in lymphocytes. B. NRF2 levels in the lymphocyte of the SZ and HC groups. The average level of marker in the cell population was estimated by the FL-signal median value. In the graphs, each point is the mean for three FL-signal median values (three parallel measurements for the same PBMC sample). The median values were normalized to the maximum signal medians values in the analyzed total sample (N = 160). The relative standard error was 5 ± 2%. The significance of the observed differences was analyzed using non-parametric Mann–Whitney (p) test. C. (a) Cumulative distribution of NRF2 for HC and SZ subgroups. (b) Cumulative distribution of ratio NRF2/NOX4 for HC and SZ subgroups. D. The dependence of NRF2 on NOX4 for the SZ and HC groups. E. Lymphocyte [HC (green) and SZ (red)] staining with two types of antibodies with different labels: NRF2(FITC) and NOX4(PC5.5).

Fig 7. Analysis of the subpopulations of the HC lymphocytes with a high and low γH2AX(R) index value.

(A and D) An example of the analysis of the population of lymphocytes. We divided the population into two subpopulations—the cells with relatively small FSC parameter values (red-fraction) and the rest (blue-fraction); (B and E) for each fraction, the cumulative distribution of several markers were determined separately. The analyzed marker is indicated along the X-axis. The significance of the observed differences was analyzed using the non-parametric Mann–Whitney test (p) and Kolmogorov–Smirnov test (D and α). (C and F) the ratio of the analyzed indices for red- and blue-subpopulations (mean ± SD for 10 lymphocyte samples).