Intact NOX2 in T Cells Mediates Pregnancy-Induced Renal Damage in Dahl SS Rats

Abstract

Background:

Hypertensive disorders of pregnancy are associated with increased risk for cardiovascular disease, renal disease, and mortality. While the exact mechanisms remain unclear, T cells and reactive oxygen species have been implicated in its pathogenesis. We utilized Dahl Salt-Sensitive (SS), SS^CD247−/− (lacking T cells), and SS^{p67phox−/−} (lacking NADPH Oxidase 2 (NOX2)) rats to investigate these mechanisms in primigravida and multigravida states.

Methods:

We assessed blood pressure and renal damage phenotypes in SS, SS^CD247−/−, SS^{p67phox−/−} rats during primigravida and multigravida states. To investigate the contribution of NOX2 in T cells, we performed adoptive transfers of splenocytes or CD4+ T cells from either SS or SS^{p67phox−/−} donors into SS^CD247−/− recipients to determine pregnancy-specific alterations in phenotype.

Results:

Multigravida SS rats developed significant pregnancy-induced renal damage and renal functional impairment associated with elevated maternal mortality rates, whereas deletion of T cells or NOX2 garnered protection. During primigravida states, this attenuation in renal damage was observed, with the greatest protection in the SS^{p67phox−/−} rat. To demonstrate that NOX2 in T cells contributes to adverse pregnancy phenotypes, adoptive transfer of SS splenocytes into SS^CD247−/− rats resulted in significant pregnancy-induced renal damage whereas transfer of SS^{p67phox−/−} splenocytes garnered protection. Specifically, transfer of SS CD4+ T cells resulted in pregnancy-induced proteinuria and increases in UARI, an effect not seen with the transfer of SS^{p67phox−/−} CD4+ T cells.

Conclusions:

T cells and NOX2-derived reactive oxygen species thus contribute to end organ damage in both primigravida and multigravida pregnancies in the SS rat leading to increases in maternal mortality.

Graphical Abstract

Introduction

Hypertensive disorders of pregnancy, as classified by the American College of Obstetricians and Gynecologists (ACOG), consists of chronic hypertension, superimposed preeclampsia, gestational hypertension and preeclampsia¹. Superimposed preeclampsia is a pregnancy-specific syndrome that occurs in women with chronic hypertension prior to pregnancy or before the 20^th week of gestation and end organ damage such as proteinuria, thrombocytopenia, and liver dysfunction, which occurs in 20–30% of women with chronic hypertension². Women that are exposed to a hypertensive disorder of pregnancy are at an increased risk for developing chronic diseases later in life such as cardiovascular disease and chronic kidney disease^3,4. Furthermore, women that experience a hypertensive disorder of pregnancy during their first pregnancy are at an increased risk for developing adverse pregnancy outcome with subsequent pregnancies⁵. These women are more likely to develop severe proteinuria and adverse renal outcomes during pregnancy contributing to increasing mortality rates^6,7. Despite advancements in understanding the pathogenesis of disease progression, the rates of maternal mortality continue to climb in the United States with the greatest rates found in non-Hispanic Black women^8,9. To address these increasing rates of maternal mortality it is important to investigate the underlying mechanisms leading to adverse pregnancy outcomes.

While it is thought that the pathogenesis of superimposed preeclampsia is unknown, there are multiple risk factors contributing to the pathology including over-activation of the immune system¹⁰, oxidative stress¹¹, and genetic disposition^12,13. Our group recently demonstrated that the SS rat maintained on a casein-based diet (0.4% NaCl) is prone to developing a preeclamptic-like phenotype that is associated with a reduction in placental vascular gene expression¹⁴; however, the cause of these vascular changes is unknown. It has been demonstrated in preclinical models that contribution of both inflammation and oxidative stress participate in the preeclampsia/maternal syndrome^15–20. Since our group has previously demonstrated the contribution of inflammation and oxidative stress in the development of salt-sensitive hypertension and renal damage in SS rats^21–24, we speculate that T cells, in an NADPH oxidase 2 (NOX2)-dependent manner contribute to the adverse pregnancy phenotypes in the SS rat.

Therefore, the current study examined the contribution of T cells, NOX2, and intact NOX2 in T cells in the pathogenesis of superimposed preeclampsia. We tested the following hypotheses: 1) genetic deletion of T cells on the SS background (SS^CD247−/−) would protect from the development of superimposed preeclampsia, 2) SS^{p67phox−/−} rats which lack the ability to generate NOX2-derived ROS would also be protected from the development of superimposed preeclampsia, and 3) T cells with intact functional NOX2 are necessary for the renal damage phenotype observed in the SS rat.

Methods

A detailed description of all methods can be found in the Supplemental Material. The authors declare that all supporting data are available within the article and its online Supplemental Files.

Animals and Diet.

All protocols were approved by the Augusta University Institutional Animal Care and Use Committee (Protocol #: 2019–0990). Experiments were performed on rats that originated from colonies at Medical College of Wisconsin and are now maintained at Augusta University. These strains include inbred female Dahl SS (SS/JrHsdMcwi), SS^CD247−/−, and SS^p67phox-/−. As previously described, SS^CD247−/− rats lack functional T cells²³ and SS^{p67phox−/−} rats lack the p67phox subunit of NOX2 resulting in non-functional NOX2^21,25. All rats were fed a 0.4% NaCl purified casein-based AIN-76A diet (#113755, Dyets Inc), and animals were housed on an aspen wood bedding (aspen, Sani Chips, P.J. Murphy)²⁶. Animals were euthanized between 7:00 and 9:00 AM at the end of each study.

Statistical Analysis and Additional Details on Study Design.

For all statistical tests, a p-value <0.05 was considered significant utilizing GraphPad Prism 10. Two-way repeated measures ANOVA with a Tukey post-hoc test was used for blood pressure and urinary protein data. Two-way ANOVA was used for renal hypertrophy, indices of renal function and medullary protein casts in Figure 1. One-way ANOVA with a Tukey post-hoc test was used for comparisons among the three pregnancy groups. The survival curve was compared using the Mantel-Cox test between the SS, SS^CD247−/−, and SS^{p67phox−/−} rats. Phenotype data are expressed as the mean ± the standard error of the mean. Sample size calculations were determined a priori via power analysis (Sigma Plot 14.0) considering the type of statistical analysis, experimental outcome, and the expected minimum changes to be observed. Approximately 150 animals were utilized across the multiple studies in the manuscript. The number of animals utilized in each experiment is denoted in the figure legends. Data was excluded if any animal did not complete the entire course of the experimental timeline.

Figure 1. Genetic deletion of T cells or NOX2-derived reactive oxygen species prevents the increased mortality rate and renal damage observed in the Dahl SS rat.

(A) Percent survival rate in the Dahl SS, SS^CD247−/−, SS^{p67phox−/−} rats both virgin and mated rats. (B) Protein excretion rates throughout multiple pregnancies (C) Renal hypertrophy observed in Dahl SS mated group relative to other groups by normalizing kidney weight to body weight. (D) Blood urea nitrogen (BUN) levels following three successful pregnancies. (E) Creatinine clearance following three successful pregnancies. (F) Quantification of medullary protein casts and representative sections of kidneys at the end of the study. †p<0.05 vs SS Mated baseline *p<0.05, ***p<0.001 vs Virgin Control (within strain), ^p<0.05 vs SS^{p67phox−/−} mated rats at the same time point, #p<0.05 vs SS^CD247−/− mated rats at the same time point. n=3–8/ group. Statistics utilized (A) Mantel-Cox test, (B) Two-way ANOVA with repeated measures, (C-F) Two-way ANOVA.

Results

Increased maternal mortality and renal damage associated with multigravida in the Dahl SS is dependent on T cells and NOX2.

While the mated SS rats exhibited increased maternal mortality, both the SS^CD247−/− and SS^{p67phox−/−} mated groups did not exhibit any maternal mortality when subjected to three pregnancies (Figure 1A). Furthermore, this maternal mortality is pregnancy-specific since no mortality was observed in any of the virgin-control groups throughout the duration of the study. For the animals that experienced three successful pregnancies, we tracked the progression of renal damage throughout this timeline as well as markers of renal function at the end of the study. The mated SS rats developed a significant and progressive increase in urinary protein excretion compared to the virgin controls (Figure 1B). Conversely, the mated SS^CD247−/− rats, which lack T cells, had an attenuation in protein excretion rates during the three pregnancies compared to the SS mated rats. The mated SS^{p67phox−/−} rats, which globally lack functional NOX2, were even more protected from pregnancy-induced renal damage since their urinary protein excretion rates were comparable to all the virgin control groups, highlighting the role of T cells and NOX2 to the progression of renal disease in the SS rat. Following the three pregnancies, renal hypertrophy was assessed in each of these groups. The greatest renal hypertrophy was observed in the SS mated group which was significantly increased relative to virgin controls as well as the other mated groups (Figure 1C). The mated SS^CD247−/− rats did exhibit a significant increase in renal hypertrophy relative to the virgin SS^CD247−/− rats; however, it was to a lesser degree than the SS group. The mated SS^{p67phox−/−} rats exhibit no renal hypertrophy relative to virgin controls. In terms of indices of renal function assessed at the end of the study, BUN levels were significantly elevated in both the virgin and mated SS rat relative to both the SS^CD247−/− and SS^{p67phox−/−} groups (Figure 1D). Similarly, creatinine clearance was significantly reduced in mated SS rats relative to virgin controls, the mated SS^CD247−/−, and mated SS^{p67phox−/−}, emphasizing the importance of T cells and NOX2 to this severe renal disease phenotype in the SS rats (Figure 1E). Lastly, these effects were also evident in the renal histological damage where the mated SS rats had a significant increase in the presence of medullary protein casts relative to the other experimental groups (Figure 1F).

Impact of T cells and NOX2 to renal damage in the Dahl SS rat during pregnancy.

To understand the potential mechanisms whereby T cells and NOX2 mediate long-term mortality and renal damage, subsequent studies were then performed to examine blood pressure and renal disease during an initial pregnancy in 9–10-week-old rats. To demonstrate the impact of T cells and NOX2 to the development of pregnancy-induced renal damage, we compared the mated groups from the SS^CD247−/− and SS^{p67phox−/−} rats to the SS rats exposed to a primigravida pregnancy protocol. While the blood pressure was not significantly different between the SS and SS^CD247−/− at any point throughout the experiment, SS^{p67phox−/−} rats exhibited a significantly lower blood pressure throughout the study (Figure 2A). Despite the lack of difference in blood pressure, both SS^CD247−/− and SS^{p67phox−/−} rats had less proteinuria than SS at the early and mid-pregnancy time points (Figure 2B). At gestational day 20, both circulating and renal immune cell profiles were examined in all three groups. There was no difference in the number of circulating CD45+ total leukocytes or CD11b/c+ monocytes/macrophages among the groups (Figure 2C). The SS^{p67phox−/−} rats had a significant increase in the number of circulating CD3+ T cells relative to SS and SS^CD247−/−, but there was no difference in CD45R+ B cells (Figure 2C). There was no difference in the number of infiltrating renal CD45+, CD11b/c+, or CD45R+ cells among the groups (Figure 2D). The only significant difference was the absence of CD3+ T cells in the SS^CD247−/− rats, as expected. To demonstrate any pregnancy-specific alterations in phenotypes, virgin controls for the SS^CD247−/− and SS^{p67phox−/−} rats were compared to the mated animals seen in Figures S1 and S2, respectively. In terms of fetal outcomes, there was no difference in the number of pups per litter or placental weight among these three groups (Figure S3). The SS^CD247−/− fetuses weighed more than both SS and SS^{p67phox−/−} at GD20, and the SS^{p67phox−/−} fetuses weighed significantly less than the SS fetuses. The SS litters exhibited more fetal demise than either the SS^CD247−/− or SS^{p67phox−/−} rats at GD 20.

Figure 2. Dahl SS rats exhibit an increase in proteinuria during pregnancy compared to SS^CD247−/− and SS^{p67phox−/−} mated rats.

(A) Daily mean arterial pressure (MAP) and (B) Proteinuria averages throughout pregnancy. (C) Circulating immune cell profile and (D) infiltrating renal immune cell profile at GD20. ***p<0.001, vs Dahl SS Baseline, *p<0.05 vs SS, ^p<0.05 vs SS^{p67phox−/−} mated rats at the same time point, #p<0.05 vs SS^CD247−/− mated rats at the same time point. %p<0.01, vs SS^CD247−/− Baseline. n=8–15/group. Statistics utilized (A-B) Two-way ANOVA with repeated measures, (C-D) Two-way ANOVA.

Adoptive transfer of Dahl SS splenocytes into SS^CD247−/− rats results in significant renal damage and vascular impairment through a NOX2-dependent mechanism.

To investigate the effects of T cells, our group developed an adoptive transfer approach utilizing SS^CD247−/− rats lacking T cells as recipients^22,24. We transferred ~10 million splenocytes from either male SS wildtype or SS^{p67phox−/−} rats into female SS^CD247−/− pups at postnatal day 5 before the thymus is fully formed. This resulted in two experimental groups: 1) SS^CD247−/−+SS Splenocytes: those that received SS WT splenocytes, and 2) SS^CD247−/−+ SS^{p67phox−/−} Splenocytes: those that received SS^{p67phox−/−} splenocytes. As a negative control, PBS (200uL) was administered to SS^CD247−/− pups at postnatal day 5 resulting in the SS^CD247−/−+PBS group. To examine the role of T cells and how intact NOX2 in T cells might contribute to renal damage in pregnancy these animals were bred with uninjected SS^CD247−/− males, and any phenotypic changes that occur could be attributed to the T cells that engrafted. The adoptive transfer of SS or SS^{p67phox−/−} splenocytes into the SS^CD247−/− females resulted in an increase in blood pressure relative to the PBS injected females (Figure 3A). While there was no statistical difference in blood pressure response between the groups treated with SS or SS^{p67phox−/−} splenocytes, there was a significant increase in protein excretion rate in rats that received SS splenocytes compared to those that received PBS or SS^{p67phox−/−} splenocytes (Figure 3B). At the end of the study on GD20, creatinine clearance was significantly reduced in the animals injected with SS splenocytes compared to the PBS and SS^{p67phox−/−} injected animals (Figure 3C), indicating functional impairment of the kidney. This reduction in creatinine clearance can be attributed, at least in part, due to the T cell’s ability to produce reactive oxygen species since there was no difference observed between the PBS and SS^{p67phox−/−} groups. These results are consistent with the renal histological damage shown in Figure 3D, where the SS splenocyte treated animals exhibited a significant increase in the presence of protein casts relative to the PBS and SS^{p67phox−/−} groups. Lastly, on GD20 when uterine artery blood flow was measured utilizing Doppler ultrasound, the rats receiving SS splenocytes exhibited a significant increase in the uterine artery resistance index suggesting an impairment in blood flow to the fetal-placental units (Figure 3E). In contrast, the animals receiving either PBS or SS^{p67phox−/−} splenocytes do not exhibit an increase in UARI, evidenced by the lack of a notch-like formation in the waveform.

Figure 3. Adoptive transfer of Dahl splenocytes into T cell deficient rats results in pregnancy-induced significant renal damage, impairment in renal function and increased in uterine artery resistance index through a NOX2-dependent mechanism.

(A) Daily mean arterial pressure (MAP) and (B) Proteinuria averages throughout pregnancy. (C) Creatinine clearance at gestational day 20. (D) Quantification of medullary protein casts and representative sections of kidneys at GD20. (E) Calculated uterine artery resistance index and representative uterine artery blood flow tracings at gestational day 20. *p<0.05 vs SS^CD247−/−+ SS Splenocytes Baseline, #p<0.05 vs SS^CD247−/−+ PBS, ^ p<0.05 vs SS^CD247−/−+SS^{p67phox−/−} Splenocytes, n=5–15/group. n=3–8 for Figure 3E. Groups: SS^CD247−/−+ PBS, SS^CD247−/−+ SS Splenocytes, SS^CD247−/−+SS^{p67phox−/−} Splenocytes. Statistics utilized (A-B) Two-way ANOVA with repeated measures, (C-E) One-way ANOVA.

On GD20, animals were euthanized, and tissues were collected for flow cytometry to analyze the engraftment of T cells from the adoptive transfer. Shown in the left graph of Figure 4A, there was no difference in the number of circulating total CD45+ leukocytes, CD11b/c+ monocytes/macrophages, and CD45R+ B cells among the three transfer groups. Shown on the right graph, in terms of repopulation of T cells, the rats injected with either SS splenocytes or SS^{p67phox−/−} splenocytes had the same number of CD3+ T cells in the blood, and these cells were primarily of the CD4+ T helper lineage, which we have demonstrated previously^22,24. Rats injected with splenocytes had significantly more T cells than the group injected with PBS. Representative flow cytometry plots are shown for the blood immune cells in Figure 4B. In terms of infiltrating immune cells in both the renal (Figure 4C and 4D) and placental tissues (Figure 4E and 4F), the only differences observed were the number of T cells present in the transfer groups versus the PBS injected rats. The number of engrafted T cells were comparable between the rats injected with either SS splenocytes or SS^{p67phox−/−} splenocytes. Additionally, these results are pregnancy-induced since there was no difference in blood pressure, protein excretion, or creatinine clearance between the virgin controls as seen in Figure S4. The number of pups and placental weights were all comparable among these groups at GD20; however, the pups from the SS^CD247−/−+ SS^{p67phox−/−} splenocyte injected dams weigh significantly more than the pups from the SS^CD247−/−+SS splenocyte injected dams (Figure S5A–C). The SS^CD247−/−+SS splenocyte injected dams had higher rates of fetal demise than either the PBS or SS^{p67phox−/−} Splenocytes injected groups (Figure S5D).

Figure 4. Immune cell profiles in target tissues following adoptive transfer of splenocytes into T cell deficient rats.

(A, B) Number of circulating immune cell profiles on gestational day 20 with representative cytometry plots of T cells for each group (SS^CD247−/−+ PBS, +SS Splenocytes, and + SS^{p67phox−/−} Splenocytes). (C, D) Number of infiltrating immune cells into kidneys of pregnant rats on gestational day 20 with representative cytometry plots of T cells for each group. (E, F) Number of infiltrating immune cells into placentas of pregnant rats on gestational day 20 with representative cytometry plots of T cells for each group. *p<0.05 vs SS^CD247−/−+ PBS within each cell type, n=4–17, Flow cytometry data for placental tissue analyzed only one placenta per dam. Statistics utilized (A, C, E) One-way ANOVA per cell type.

To assess potential differences in T cells from the adoptive transfer, we isolated T cells from kidneys at GD20 of SS^CD247−/− rats that were either injected with SS or SS^{p67phox−/−} splenocytes and utilized a PCR array approach to examine over 80 genes related to antioxidant, reactive oxygen species metabolism, and oxidative response genes. When comparing renal T cells isolated from the two injected groups, there was an overall trend for genes to be downregulated in the T cells isolated from the rats that received SS^{p67phox−/−} splenocytes relative to the SS splenocytes (Figure S6). There were only 4 genes that reached statistical significance when comparing the T cells from SS^{p67phox−/−} versus those from the SS. Both Ercc2 (Excision repair cross-complementing rodent repair deficiency) and Slc38a1 (solute carrier family 38, member 1, glutamine transporter) were significantly upregulated while Cat (Catalase) and Nqo1 (NADPH dehydrogenase, quinone 1) were significantly downregulated in the T cells from SS^{p67phox−/−} injected rats compared to the SS injected.

Adoptive transfer of CD4+ T cells into SS^CD247−/− rats results in pregnancy-induced renal damage

The adoptive transfer of splenocytes demonstrated a role for CD4+ T cells with intact NOX2 to induce renal damage during pregnancy; however, the adoptive transfer of splenocytes also includes other cell types that can generate ROS such as macrophages. Since the adoptive transfer of splenocytes resulted in primarily CD4+ T cells being engrafted, we adoptively transferred CD4+ T cells into SS^CD247−/− rats to investigate the contributions of a single cell type. Following the same experimental timeline, we transferred CD4+ T cells isolated from the spleens of male Dahl SS WT rats (SS^CD247−/−+SS CD4+ T Cells) or SS^{p67phox−/−} rats (SS^CD247−/−+ SS^{p67phox−/−} CD4+ T Cells). Rats injected with PBS (SS^CD247−/−+PBS) were again utilized as a control. As observed in the splenocyte adoptive transfer, there was an increase in mean arterial pressure in the rats injected with either CD4+ T cells from the SS or SS^{p67phox−/−} rats relative to the PBS-injected group (Figure 5A). There was a significant increase in protein excretion rate in the rats injected with SS CD4+ T cells compared to the rats injected with PBS (Figure 5B). There was no difference in proteinuria levels in the group that were injected with SS^{p67phox−/−} CD4+ T cells relative to the PBS injected group. There was no difference in creatinine clearance (Figure 5C) or renal histological damage (Figure 5D) among the three experimental groups at GD20. There was a significant increase in uterine artery resistance index in the animals injected with SS CD4+ T cells compared to animals injected with PBS, and this increase was absent in the group injected with SS^{p67phox−/−} CD4+ T cells (Figure 5E).

Figure 5. Adoptive transfer of Dahl CD4+ T cells into T cell deficient rats results in pregnancy-induced significant renal damage, impairment in renal function and increased in uterine artery resistance index through a NOX2-dependent mechanism.

(A) Daily mean arterial pressure (MAP) and (B) Proteinuria averages throughout pregnancy. (C) Creatinine clearance at gestational day 20. (D) Quantification of medullary protein casts and representative sections of kidneys at GD20. (E) Calculated uterine artery resistance index and representative uterine artery blood flow tracings at gestational day 20. *p<0.05, ***p<0.001 vs SS^CD247−/−+ SS CD4+ Baseline, #p<0.05 vs SS^CD247−/−+ PBS, ^ p<0.05 vs SS^CD247−/−+SS^{p67phox−/−} CD4+. n=6–7/group. Statistics utilized (A-B) Two-way ANOVA with repeated measures, (C-E) One-way ANOVA.

To verify the engraftment of CD4+ T cells in these animals, flow cytometry was performed on blood (Figure 6A and 6B), kidney (Figure 6C and 6D) and placental (Figure 6E and 6F) tissues at GD20. Similar, to the adoptive transfer of splenocytes, there was no difference in the number of total leukocytes, monocytes/macrophages and B cells in any tissue. The only significant difference observed was in the T cell population where rats injected with either SS CD4+ T cells or SS^{p67phox−/−} CD4+ T cells that exhibited greater number of CD4+ relative to PBS controls. These data demonstrate that intact NOX2 specifically in CD4+ T cells is needed to increase proteinuria and during pregnancy.

Figure 6. Immune cell profiles in target tissues following adoptive transfer of CD4+ into T cell deficient rats.

(A, B) Number of circulating immune cell profiles on gestational day 20 with representative cytometry plots of T cells for each group (SS^CD247−/−+ PBS, +SS CD4+, and + SS^{p67phox−/−} CD4+). (C, D) Number of infiltrating immune cells into kidneys of pregnant rats on gestational day 20 with representative cytometry plots of T cells for each group. (E, F) Number of infiltrating immune cells into placentas of pregnant rats on gestational day 20 with representative cytometry plots of T cells for each group. *p<0.05 vs SS^CD247−/−+ PBS within each cell type, n=5–7, Flow cytometry data for placental tissue analyzed only one placenta per dam. Statistics utilized (A, C, E) One-way ANOVA per cell type.

Lastly, the number of pups and placental weights were all comparable among these groups at GD20; however, the pups from the SS^CD247−/−+ SS CD4+ injected dams weigh significantly less than the pups from the SS^CD247−/−+SS^{p67phox−/−} CD4+ injected dams (Figures S7A, 7B). Placentas from both injected groups weighed significantly less than the placentas from the PBS controls (Figure S7C). The PBS control group did not exhibit fetal demise; however, both injected groups experienced higher rates of fetal demise (Figure S7D).

Discussion

These studies demonstrated the contribution of T cells, NOX2-derived reactive oxygen species, and intact NOX2 in T cells to the development of pregnancy-induced renal damage. These same mechanisms further exacerbated significant renal disease observed in the SS rat subjected to multigravida leading to an increase in maternal mortality and renal impairment. By genetically deleting T cells or functional NOX2 on the SS background, the pregnancy-induced renal damage was attenuated in these animals. To further investigate how intact NOX2 in T cells induces renal damage during pregnancy, we performed adoptive transfer of splenocytes into SS rats lacking T cells (SS^CD247−/−). Rats that received SS splenocytes develop significant renal damage during pregnancy, while those rats that received either PBS or SS^{p67phox−/−} splenocytes were protected. The adoptive transfer of splenocytes resulted in primarily CD4+ T cells being engrafted in the SS^CD247−/− rats, but the splenocyte transfer also included other NOX2 containing cell types. To specifically isolate the effects of the adoptive transfer to CD4+ T cells, we performed an adoptive transfer of CD4+ T cells from either SS or SS^{p67phox−/−} into SS^CD247−/− rats. The transfer of CD4+ SS T cells into SS^CD247−/− rats resulted in significant increase in protein excretion rate throughout pregnancy and an increase in uterine artery resistance index, while transfer of SS^{p67phox−/−} CD4+ T cells did not. These studies indicate that intact NOX2 in CD4+ T cells induced pregnancy-related renal damage and vascular impairments that could lead to lasting renal impairments and increased risk for maternal mortality.

As the rates of maternal mortality continue to rise in the United States^8,27, it is vital to understand the underlying mechanisms to help combat these climbing rates. The majority of these deaths are cardiovascular-related mortalities and can occur up to one year postpartum²⁸. The risk for pregnancy-related mortalities is greater in non-Hispanic Black women relative to non-Hispanic white women demonstrating the racial disparity in pregnancy-related mortalities^29,30. Our work demonstrated that multigravida SS rats exhibited an increased mortality, and rats that survived three pregnancies developed significant renal damage and impaired renal function. A recent publication by Havers-Borgersen demonstrated that primiparous women who experience a preeclamptic pregnancy exhibited increased risk of developing preeclampsia with subsequent pregnancies relative to women who had a normotensive pregnancy³¹. Similarly, an observation study investigating the long-term renal outcomes of a preeclamptic pregnancy reported higher risk of developing chronic kidney disease, chronic hypertension, and reduction in eGFR relative to women that had a normotensive pregnancy³². While these reports did not investigate mechanisms, our data suggests that these lasting consequences of adverse pregnancies could, in part, be attributed to T cells and NOX2-derived ROS, which could serve as potential therapeutic targets to mitigate disease and begin to reduce maternal mortality rates.

In order to better understand the development of these adverse outcomes, we investigated the phenotype of a primigravida pregnancy in the SS rat. In support of these findings in the multigravida studies, both T cells and intact NOX2 contribute to the renal damage that occurs during a singular pregnancy in the SS rat. A role for both T cells^33–35 and reactive oxygen species^36–40 in the development of preeclampsia has been demonstrated in both humans and pre-clinical models. In pre-clinical models of preeclampsia, CD4+ T cells have been identified as playing a detrimental role in the disease state^16,20. It has been demonstrated that a single injection of CD4+ T cells isolated from the placenta of preeclamptic women and activated in vitro induces a preeclamptic-like phenotype in nude-athymic rats at GD13⁴¹. More recently, Deer et al demonstrated that injection of CD4+ T cells from rats exposed to placental ischemia transferred into normal Sprague Dawley rats at GD13 resulted in increases in circulating NK cells and placental hydrogen peroxide production⁴². Our work supports the role of CD4+ T cells in promoting end organ damage during pregnancy, and through our adoptive transfer experiments, appears to be dependent on intact NOX2. Since we adoptively transferred these naïve cells at postnatal day 5 into SS^CD247−/− pups, we believe this approach allows the T cells to as a naïve/inactivated T cell during pregnancy to gain insight into how these cells respond to an adverse pregnancy. It appears there is a NOX2-independent mechanism at play regarding blood pressure during pregnancy in the adoptive transfer experiments since there was no difference in blood pressure between the transferred groups. This will be the focus of future experiments.

While the presence of reactive oxygen species is critical for the proper development of the placenta, an imbalance of the pro-oxidant and antioxidant ratios can lead to the production of oxidative stress^43,44. It can be difficult to target this imbalance due to either an increase in pro-oxidants or a decrease in antioxidant capabilities or the various sources that can produce oxidative stress. In a recent meta-analysis of 32 studies, oral antioxidants in the treatment and prevention of preeclampsia were examined⁴⁵. The only protective effect observed in these studies was on the incidence of intrauterine growth restriction in the offspring. The lack of maternal effects could be attributed to variations in types and concentrations of antioxidants administered in the various studies. In animal models, there are reports of oxidative stress contributing to the development of preeclampsia^{17,42,46–49}. In our study, by genetically deleting the critical subunit of NOX2 on the SS background, these animals exhibited the greatest protection from pregnancy-induced end organ damage. NOX2 is the most prominent superoxide-producing enzyme found in phagocytic cells like macrophages, but it is also expressed in T cells^50,51. NOX2-generated ROS has been shown to cause differentiation and activation of T helper cells and is dependent on transcription factors GATA-3 and STAT5⁵¹. Based upon our adoptive transfer data, the number of engrafted CD4+ T cells did not differ between those recipients injected with SS cells versus those injected with SS^{p67phox−/−} cells; however, intact NOX2 in CD4+ T cells was necessary to induce proteinuria and increased in UARI during pregnancy.

While the data demonstrates that intact NOX2 in T cells contributes to pregnancy-induced renal damage, there are limitations to consider. All transferred cells were from male donors which could have implications on results. Our group has recently demonstrated that the sex of the transferred cells affects the severity of salt-sensitive hypertension and kidney damage⁵². When female SS^CD247−/− recipients received male SS splenocytes, there was a greater exacerbation in hypertension and kidney damage compared to females receiving female SS splenocytes. The impact of sex on hematopoietic transfers in humans has also been observed. There have been reports that adoptive transfers with male donor cells are associated with a greater risk for developing post transplantation hypertension, albuminuria, and mortality^53–55. Future studies will examine how the sex of transferred cells could influence pregnancy-related phenotypes. While reconstitution of SS CD4+ T cells resulted in proteinuria and increases in UARI, CD4+ transfer did not induce alterations in creatinine clearance or renal histological damage, differing from what was observed in the splenocyte transfer experiments. Perhaps the small percentage of CD8+ T cells reconstituted following SS splenocyte transfer could be contributing to the full renal damage phenotypes. Future studies will also examine CD8+ T cell transfers to examine that contribution of intact NOX2 in this population as well. Lastly, since the multigravida experiments were run concurrently with the SS control animal that has been previously published¹⁴, we felt it was appropriate to plot the knockout data against the SS controls.

Perspectives

Since superimposed preeclampsia is a multi-systems disorder, there can be multiple mechanisms contributing to the underlying pathology. While there was no blood pressure difference observed in our work, these findings have shown that intact NOX2 in T cells contributes to pregnancy-induced renal damage, and perhaps the ROS-producing capabilities of T cells could be a therapeutic target for renal related pathologies.

Novelty and Relevance.

What is New?

Both T cells and NOX2 contribute to renal damage and renal function impairment in multigravida Dahl SS rats contributing to increased maternal mortality.

Adoptive transfer of splenocytes or CD4+ T cells with intact NOX2 are necessary for pregnancy-induced proteinuria.

What is Relevant?

This study demonstrates that intact NOX2 in T cells contributes to pregnancy-induced renal damage.

Clinical/Pathophysiological Implications?

Intact NOX2 in T cells may contribute to renal damage phenotypes in women that experience adverse pregnancies such as superimposed preeclampsia.

Nonstandard Abbreviations and Acronyms

SS

Dahl Salt-Sensitive

SS^CD247−/−

Dahl Salt-Sensitive CD247 knockout rat

NOX2

NADPH Oxidase-2

SS^{p67phox−/−}

Dahl Salt-Sensitive p67^phox (NOX2) knockout rat

GD

gestational day

EP

early pregnancy

MP

mid pregnancy

LP

late pregnancy

UARI

uterine artery resistance index