Class switching and high-affinity immunoglobulin G production by B cells is dispensable for the development of hypertension in mice

Abstract

Aims 

Elevated serum immunoglobulins have been associated with experimental and human hypertension for decades but whether immunoglobulins and B cells play a causal role in hypertension pathology is unclear. In this study, we sought to determine the role of B cells and high-affinity class-switched immunoglobulins on hypertension and hypertensive end-organ damage to determine if they might represent viable therapeutic targets for this disease.

Methods and results 

We purified serum immunoglobulin G (IgG) from mice exposed to vehicle or angiotensin (Ang) II to induce hypertension and adoptively transferred these to wild type (WT) recipient mice receiving a subpressor dose of Ang II. We found that transfer of IgG from hypertensive animals does not affect blood pressure, endothelial function, renal inflammation, albuminuria, or T cell-derived cytokine production compared with transfer of IgG from vehicle infused animals. As an alternative approach to investigate the role of high-affinity, class-switched immunoglobulins, we studied mice with genetic deletion of activation-induced deaminase (Aicda^−/−). These mice have elevated levels of IgM but virtual absence of class-switched immunoglobulins such as IgG subclasses and IgA. Neither male nor female Aicda^−/− mice were protected from Ang II-induced hypertension and renal/vascular damage. To determine if IgM or non-immunoglobulin-dependent innate functions of B cells play a role in hypertension, we studied mice with severe global B-cell deficiency due to deletion of the membrane exon of the IgM heavy chain (µMT^−/−). µMT^−/− mice were also not protected from hypertension or end-organ damage induced by Ang II infusion or deoxycorticosterone acetate-salt treatment.

Conclusions 

These results suggest that B cells and serum immunoglobulins do not play a causal role in hypertension pathology.

Graphical Abstract

Translational perspective

Our results suggest that in most cases of essential hypertension, B cells are not a causal factor in the pathophysiology of disease. Thus, elevated serum immunoglobulins seen in hypertensive animals and humans may reflect a biomarker of aberrant immune activation in hypertension and not a therapeutic target. However, autoantibodies may cause hypertension in special cases, and more work is needed to determine whether specific B-cell subsets might play an important role that is masked by global B-cell deficiency.

1. Introduction

Hypertension is the leading risk factor for global mortality due to its damaging effects on the heart, kidney, and vasculature.¹ In the USA alone, nearly 50% of the adult population has hypertension, and the prevalence increases with age such that by the age of 75, ∼80% of people have hypertension.² Importantly, even in hypertensive individuals with reasonable blood pressure (BP) control, an elevated risk of cardiovascular events remains.^3,4 Thus, there is a critical need for novel therapeutics to target the end-organ damage associated with this disease.

While innate and adaptive immune cells, particularly T lymphocytes, play an important role in the pathophysiology of hypertension,⁵ the role of B lymphocytes and immunoglobulins in hypertension is poorly understood. Both experimental and human hypertension are associated with modest elevations in specific isotypes of serum immunoglobulins, particularly immunoglobulin G (IgG), but whether these immunoglobulins or the cells that produce them play a causal role in the pathophysiology of hypertension is unknown.^6–10 In fact, Khamis et al.¹¹ showed that in the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT), high total serum IgG levels were actually an independent predictor of freedom from adverse cardiovascular events in patients with hypertension, suggesting that IgG may play a protective role. In 2007, Guzik et al.¹² demonstrated that mice deficient in T and B cells develop blunted experimental hypertension, with the hypertensive response being restored by adoptive transfer of T but not B lymphocytes. These results suggest that B lymphocytes are not necessary for the development of hypertension. However, subsequently Chan et al.¹⁰ showed that pharmacologic depletion of B cells using anti-CD20 antibody or genetic deletion of B-cell activating factor receptor (BAFF-R) in mice attenuated angiotensin II (Ang II)-induced hypertension. Furthermore, they showed that adoptive transfer of wild type (WT) B cells into BAFF-R^−/− mice restored the hypertensive response.¹⁰ Of note, B cells are classified into two major subsets—B1 and B2 cells. Both of the models used by Chan et al. preferentially target B2 cells with some degree of retention of B1 cells. Dingwell et al.¹³ recently demonstrated that mice unable to produce functional B cells due to deletion of the gene for the heavy chain joining region (J_HT) have a modest reduction in baseline BP. However, these authors did not study the response to hypertensive stimuli in J_HT mice. Thus, the effect of a complete loss of B cells in hypertension is not known.

B-cell immunoglobulin production is dependent on a subset of B cells called germinal centre (GC) B cells. These cells undergo somatic hypermutation and/or immunoglobulin class switching (e.g. IgM → IgG), a process dependent on the enzyme activation-induced cytidine deaminase (AID), to produce high-affinity immunoglobulins.¹⁴ This GC reaction is driven by interleukin 21 (IL-21) and T follicular helper (Tfh) cells, both of which we recently demonstrated play a critical role in hypertension and hypertensive end-organ damage.⁹

In this study, we sought to determine whether immunoglobulins play a causal role in the pathophysiology of hypertension using adoptive transfer studies and AID deficient mice (Aicda^−/−), which are unable to produce high-affinity class-switched immunoglobulins. We found that adoptive transfer of IgG from hypertensive animals did not induce hypertension in recipient mice and that Aicda^−/− mice were not protected from Ang II-induced hypertension and end-organ damage. We then investigated whether B cells may have immunoglobulin independent functions in hypertension by studying mice with global deficiency of B cells due to deletion of the IgM heavy chain (µMT^−/−). Interestingly, these mice were also not protected from Ang II or deoxycorticosterone acetate (DOCA)-salt hypertension. Thus, it appears that B cells are not required for the development of experimental hypertension.

2. Methods

2.1. Mice

WT C57Bl/6J mice and immunoglobulin M heavy chain deficient (µMT^−/−) mice (Stock No. 002288) were obtained from Jackson Laboratory (USA). Activation-induced cytidine deaminase (Aicda^−/−) mice on a C57Bl/6N background were obtained from Genentech Inc. and generated as described in Sun et al.¹⁵ For all studies, age- and sex-matched mice were used starting at 10–12 weeks of age. Mice were anaesthetized with isoflurane via nose cone or ketamine/xylazine (90–120 mg/kg + 10 mg/kg; 1:1 volume) via intraperitoneal injection. Anaesthetization was performed once before implanting osmotic mini-pumps or once before uninephrectomy and implantation of DOCA pellets. All protocols were conducted in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals and the Institutional Animal Care and Use Committee at Vanderbilt University Medical Center.

2.2. Experimental hypertension

Hypertension was induced using Ang II or DOCA-salt treatment. For Ang II-induced hypertension, 28-day osmotic mini-pumps (Alzet, DURECT, model 2004) containing Ang II (490 ng/kg/min; Sigma Catalog #A2900) or vehicle (0.08 M sodium chloride + 1% acetic acid solution) or 14-day osmotic mini-pumps (Alzet, DURECT, model 2002) containing Ang II (140 ng/kg/min) or vehicle were implanted subcutaneously under 2–3% isoflurane anaesthesia via nose cone. For DOCA-salt hypertension, uninephrectomy was performed and a DOCA pellet (100 mg; Innovative Research of America) was implanted subcutaneously under anaesthesia with ketamine/xylazine (90–120 mg/kg + 10 mg/kg; 1:1 volume) through intraperitoneal injection. In addition, the drinking water was supplemented with 1% NaCl for 21 days after uninephrectomy and DOCA pellet implantation. In memory experiments, we first implanted 28-day pumps containing Ang II (490 ng/kg/min), then removed the pumps and let the mice rest for 14 days (wash-out period), and then implanted 14-day pumps containing low-dose Ang II (140 ng/kg/min) or vehicle. In the IgG transfer experiments, we implanted 14-day pumps containing low-dose Ang II (140 ng/kg/min), and purified IgG (prepared as described below) was injected intraperitoneally on Days 2, 5, 8, and 11 of Ang II infusion. Mice were euthanized by CO₂ inhalation at the end of experiments.

2.3. BP measurement

BP was measured non-invasively using tail-cuff (Hatteras Instruments) twice a week or invasively using carotid radiotelemetry 3 days a week as previously described.¹⁶ After implantation of telemeters, mice were allowed to recover for 10 days prior to recording baseline BPs and implanting osmotic mini-pumps.

2.4. Quantification of serum immunoglobulins

Serum immunoglobulins were quantified using a LEGENDplex Mouse Immunoglobulin Isotyping Panel (Biolegend, Cat. #740493) according to the manufacturer’s instructions.

2.5. Serum IgG purification and adoptive transfer

Serum samples were obtained from WT mice treated with 28 days of Ang II (490 ng/kg/min) or vehicle. IgG was purified from serum using a Melon Gel IgG Spin Purification Kit (Thermo Scientific, Cat. #45206) after diluting the serum 1:10 in Melon Gel Purification Buffer. We then dialysed the flow-through from the column with endotoxin-free phosphate-buffered saline (PBS) using Slide-A-Lyzer Dialysis Cassettes (Thermo Scientific, Cat. #2160728), and further concentrated using Vivaspin 6 (GE Healthcare, 28-9323-17) at 4000 g, 15 min. The amount of purified IgG was quantified by LEGENDplex (BioLegend, Cat. #740493). One hundred fifty micrograms purified IgG in 200 µL PBS was injected intraperitoneally to WT mice on Days 2, 5, 8, and 11 of low-dose (140 ng/kg/min) Ang II infusion.

2.6. Splenic T-cell isolation, culture, and cytokine quantification

Spleens were homogenized and filtered through a 40 µm cell strainer to obtain single-cell suspensions followed by depletion of red blood cells (RBCs) using RBC lysis buffer (eBioscience). The single-cell suspension from each spleen was split in half for isolation of CD4⁺ or CD8⁺ T cells using Miltenyi negative selection cell separation kits following the manufacturer’s instructions. After isolation, splenic T cells were cultured with RPMI 1640 media containing 10% FBS, 1% penicillin/streptomycin, and 50 µM β-mercaptoethanol in non-tissue culture-treated flat-bottom 96-well plates coated with mouse anti-CD3 (2 µg/mL) and mouse anti-CD28 (2 µg/mL) antibodies (BD Biosciences) at a density of 200 000 cells/100 µL/well for 72 h. Mouse interleukin 17 A (IL-17A), IL-21 and interferon gamma (IFNγ) were measured from the cell culture supernatants using ELISA kits (Invitrogen, Cat. #88-7371-88, #88-8210-88, and #88-7314-88).

2.7. Vascular reactivity studies

Isometric tension studies were performed on 2-mm segments of third-order mesenteric arterioles dissected free of perivascular fat using a small vessel horizontal wire myograph (Danish Myo Technology, models 610 M and 620 M) as previously described.¹⁶ Vessel segments were pre-constricted with norepinephrine before testing endothelium-dependent and endothelium-independent vascular relaxations using increasing doses of acetylcholine or sodium nitroprusside, respectively.

2.8. Urinary albumin and creatinine measurements

Albumin and creatinine concentrations were measured on spot urine samples using ELISA kits (Exocell, Cat. #1011 and #1012). Albumin concentration was divided by the creatinine concentration and reported as the albumin/creatinine ratio.

2.9. Flow cytometry of renal, aortic, bone marrow, and splenic leukocytes

Single-cell suspensions of thoracic aorta and kidney were prepared as previously described.¹⁷ Thoracic aortas with surrounding perivascular fat were first minced and digested in RPMI 1640 media containing 10% FBS, 1 mg/mL collagenase A, 1 mg/mL collagenase B and 100 μg/mL DNAse I for 30 min at 37°. Kidneys were homogenized and digested in RPMI 1640 media containing 10% FBS, 2 mg/mL collagenase D, and 100 μg/mL DNAse I for 20 min at 37°. After digestion, tissue homogenates were filtered through 40 µm cell strainers. Bone marrow from femur and tibia was flushed out with RPMI 1640 media containing 10% FBS and filtered through 40 µm cell strainers. Spleens were homogenized and filtered through 40 µm cell strainers. RBC lysis was used to deplete RBCs from the spleen samples. We used LIVE/DEAD Fixable Violet Dead Cell Stain (Life Technologies) for viability staining and surface antibodies as outlined in Supplementary material online, Tables S1–S6. Flow cytometry was performed on BD FACSCanto II system and analysed using Flowjo software. Gates were set according to fluorescence minus one controls. Results were normalized using 123count eBeads (eBioscience).

2.10. Histological staining and imaging

Mice were perfused with 10% neutral buffered formalin after euthanasia. Aortas were incubated in Decalcifying Solution-Lite (Sigma-Aldrich) for 24 h, embedded in paraffin, and cut into 5 µm cross-sections. For collagen staining, slides were stained with Picrosirius Red, scanned using a Leica SCN400 Slide Scanner, and quantified by Digital Image Hub software (Leica Biosystems).

2.11. Statistics

All data are presented as mean ± SEM or as box-and-whisker plots, with the line in the box representing the median and the whiskers representing the minimum to maximum. Data were analysed in GraphPad Prism using Student’s t-test (for two group comparisons) or two-way ANOVA (for multiple group comparisons). Non-parametric two group data were analysed by Mann–Whitney U test. P < 0.05 was considered significant.

3. Results

3.1. Repeated hypertensive stimulus induces formation of memory B cells

We previously demonstrated that hypertension is associated with increased GC B cells in secondary lymphoid organs.⁹ GC B cells differentiate into long-lived memory B cells and immunoglobulin secreting plasma cells that home to the bone marrow and can be rapidly reactivated to produce large amounts of high-affinity immunoglobulins. To determine whether repeated hypertensive stimuli increase memory B cells and plasma cells in the bone marrow, we treated C57Bl/6J WT mice with an initial exposure of Ang II (490 ng/kg/min) for 4 weeks, followed by a 2-week washout and then a low dose of Ang II (140 ng/kg/min) or vehicle for 2 weeks. We previously showed that this low dose of Ang II increases BP in mice that received an initial pressor dose of Ang II.¹⁸ As expected, WT mice showed a significant increase in BP during the last 2 weeks of low-dose Ang II infusion compared to vehicle treatment (Figure 1A). We then isolated bone marrow from the two groups of mice and found a significant increase in frequency of memory B cells and a trend for increased plasma cells in mice that received the second dose of Ang II compared to vehicle (Figure 1B and C and Supplementary material online, Figures S1 and S2). These data suggest that hypertension induces a memory B-cell response.

Figure 1.

Repeated hypertensive stimulus induces formation of memory B cells. (A) Systolic blood pressure (SBP) was measured weekly by tail-cuff over 56 days of repeated Ang II infusion. WT mice were infused with 4 weeks of Ang II (490 ng/kg/min), followed by a wash-out period of 2 weeks, and then 2 weeks of Ang II (140 ng/kg/min) or vehicle (n = 4–5). (B) Percentage of B memory cells (IgG⁺CD38⁻CD80⁺) of total CD45⁺ cells in bone marrow from WT mice infused with repeated Ang II or vehicle (n = 7–10). (C) Percentage of plasma cells (CD19⁻B220⁻CD138⁺IgK⁺) of total CD45⁺ cells in bone marrow from WT mice infused with repeated Ang II or vehicle (n = 7–10). Data are expressed as mean ± SEM (A) or box-and-whisker plots (B and C); *P < 0.05, **P < 0.01 by two-way ANOVA with repeated measures (A) or Student’s t-test (B and C).

3.2. IgG from hypertensive mice is not sufficient to induce a hypertensive response in recipient mice treated with a low dose of Ang II

To determine whether IgG from hypertensive mice is sufficient to induce hypertension in recipient mice in the setting of low-dose Ang II infusion, we purified IgG from the serum of male WT mice after 4 weeks of either vehicle or Ang II (490 ng/kg/min) infusion (referred to as vehicle IgG and HTN IgG, respectively) and injected the IgG intraperitoneally twice weekly into male WT mice that received a low dose of Ang II (140 ng/kg/min) for 2 weeks (Figure 2A and B). Although the mice that received HTN IgG had a modest increase in aortic total leukocytes, T cells, CD8⁺ T cells, macrophages, and NK cells (Figure 2C), there was no effect of HTN IgG on BP, endothelium-dependent or -independent relaxation of mesenteric arterioles, or glomerular injury as assessed by albuminuria (Figure 2D–F). Furthermore, HTN IgG had no effect on renal inflammatory cell content (Supplementary material online, Figure S3). To determine whether HTN IgG affects T-cell cytokine secretion, we isolated splenic CD4⁺ and CD8⁺ T cells and cultured them in the presence of anti-CD3 and anti-CD28 antibodies for 3 days. We found no difference in the production of interleukin 17 A (IL-17A) and IL-21 from CD4⁺ T cells or interferon gamma (IFNγ) from splenic CD8⁺ T cells between the two groups (Figure 2G). We previously demonstrated that these cytokines play an important role in hypertension.^9,17,19 Taken together, these data indicate that IgG from hypertensive mice compared to IgG from normotensive mice is not sufficient to induce a hypertensive response in recipient mice.

Figure 2.

IgG from hypertensive mice is not sufficient to induce a hypertensive response in recipient mice treated with a low dose of Ang II. (A and B) Schematic diagram of IgG transfer study. IgG was purified from serum of WT mice infused with 4 weeks of Ang II (490 ng/kg/min; HTN IgG) or vehicle (Vehicle IgG). Purified IgG was administered to WT mice twice weekly during 2 weeks of low-dose Ang II (140 ng/kg/min) infusion. (C) Flow cytometric quantification of total leukocytes (CD45⁺), T cells (CD3⁺), CD4⁺ T cells, CD8⁺ T cells, B cells (CD19⁺), macrophages (F4/80⁺), NK cells (NK1.1⁺), and neutrophils (CD11b⁺Ly6G⁺) in the aorta from both groups (n = 10). (D) Systolic BP was measured by tail-cuff weekly over 14 days of Ang II infusion (n = 12). (E) Endothelium-dependent relaxation in response to increasing doses of acetylcholine (Ach) (left) and endothelium-independent relaxation in response to increasing doses of sodium nitroprusside (SNP) (right) were measured in isolated mesenteric arterioles (n = 11). (F) Albumin:creatinine ratio was measured in both groups by ELISA (n = 11–12). (G) Splenic CD4⁺ T-cell production of IL-17A and IL-21 and CD8⁺ T-cell production of IFNγ were quantified in both groups by ELISA (n = 7–9). Data are expressed as box-and-whisker plots (C, F, and G) or mean ± SEM (D and E); *P < 0.05 by Student’s t-test or Mann–Whitney U test (C).

3.3. High-affinity class-switched immunoglobulins are not necessary for a hypertensive response in female or male mice

To further investigate whether high-affinity class-switched immunoglobulins might be important for the full development of hypertension and hypertensive end-organ damage, we studied both female and male mice with genetic deletion of AID (Aicda^−/−) on a C57BL/6N background. AID is the key enzyme required for somatic hypermutation to generate high-affinity immunoglobulins and for class switch recombination of immunoglobulin heavy chain genes. Upon antigen activation, B cells in Aicda^−/− mice are able to produce IgM and IgD and proliferate but are unable to produce IgG and other class-switched immunoglobulins such as IgA. We infused female and male Aicda^−/− mice and Aicda^+/+ littermate controls with 490 ng/kg/min of Ang II or vehicle. We first confirmed that female Aicda^−/− mice were deficient in IgG and IgA production by quantifying serum immunoglobulins after 4 weeks of Ang II infusion. Indeed, IgG subtypes (including IgG1, IgG2a, IgG2b, and IgG3) as well as IgA were virtually undetectable in Aicda^−/− mice compared to Aicda^+/+ mice and there was a likely compensatory increase in IgM (Supplementary material online, Figure S4A). Systolic BP, diastolic BP, and heart rate (HR) were not significantly different between female Aicda^+/+ and Aicda^−/− mice as measured by telemetry at baseline or during 4 weeks of Ang II infusion (Figure 3A). Furthermore, at the end of 4 weeks of Ang II infusion, there was no difference in aortic or renal adaptive and innate immune cell infiltration, Ang II-induced aortic medial hypertrophy, or albuminuria (Figure 3B–F and Supplementary material online, Figure S4B). To determine if high-affinity class-switched immunoglobulins contribute to the rise in BP in response to a repeated hypertensive stimulus, we used the repeat Ang II stimulation model described in Figure 1. We found that female Aicda^−/− mice are not protected from a repeat hypertensive stimulus (Figure 3G).

Figure 3.

High-affinity class-switched immunoglobulins are not necessary for a hypertensive response to Ang II in female mice. (A) SBP, diastolic BP (DBP), and HR were measured invasively weekly using carotid radiotelemetry over 28 days of Ang II (490 ng/kg/min) infusion in female Aicda^+/+ and Aicda^−/− mice (n = 4–5). (B and C) Flow cytometric quantification of total leukocytes (CD45⁺), T cells (CD3⁺), B cells (CD19⁺), macrophages (F4/80⁺), NK cells (NK1.1⁺), and neutrophils (CD11b⁺Ly6G⁺) in the aorta from both groups (n = 19–22 for B and n = 6–9 for C). (D) Representative images of bright-field aortic media thickness by Picrosirius Red staining. Scale bar: 100 µm. (E) Quantification of aortic media thickness in Aicda^+/+ and Aicda^−/− mice infused with vehicle or Ang II for 4 weeks (n = 5–6). (F) Albumin:creatinine ratio was measured in both groups by ELISA (n = 10–16). (G) SBP was measured by tail-cuff weekly in female Aicda^+/+ and Aicda^−/− mice infused with 4 weeks of Ang II (490 ng/kg/min), followed by a wash-out period of 2 weeks, and then 2 weeks of Ang II (140 ng/kg/min) (n = 9). Data are expressed as box-and-whisker plots (B, C, E, and F) or mean ± SEM (A and G); ***P < 0.001,****P < 0.0001 by two-way ANOVA (E).

Similar to female mice, male Aicda^−/− mice had virtually undetectable serum IgG and IgA and a compensatory increase in IgM after 4 weeks of Ang II infusion (Supplementary material online, Figure S5A). We also quantified adaptive immune cells in the spleen of male Aicda^−/− and Aicda^+/+ mice. Consistent with prior reports,²⁰ we found an increase in GC B cells in Aicda^−/− mice (Supplementary material online, Figure S5B) which likely reflects a compensatory reaction. There was no change in T cells or T-cell subsets (CD4⁺, CD8⁺, and Tfh) (Supplementary material online, Figure S5B). Like female mice, male Aicda^−/− mice were not protected from Ang II-induced hypertension (Figure 4A). There was also no difference in infiltration of innate and adaptive immune cells in the aorta or kidney, CD4⁺ T-cell production of IL-17A and IL-21, CD8⁺ T-cell production of IFNγ, or albuminuria following 4 weeks of Ang II infusion (Figure 4B–D and Supplementary material online, Figure S5C). As in female mice, AID deficiency did not affect the hypertensive response to repeat Ang II infusion (Figure 4E). Taken together, these data indicate that B-cell immunoglobulin production likely does not play a causal role in hypertension pathophysiology.

Figure 4.

High-affinity class-switched immunoglobulins are not necessary for a hypertensive response to Ang II in male mice. (A) SBP was measured by tail-cuff weekly over 28 days of Ang II infusion (490 ng/kg/min) in male Aicda^+/+ and Aicda^−/− mice (n = 14–15). (B) Flow cytometric quantification of total leukocytes (CD45⁺), T cells (CD3⁺), B cells (CD19⁺), macrophages (F4/80⁺), NK cells (NK1.1⁺), and neutrophils (CD11b⁺Ly6G⁺) in the aorta of both groups (n = 4–7). (C) Splenic CD4⁺ T-cell production of IL-17A and IL-21 and CD8⁺ T-cell production of IFNγ were quantified in both groups by ELISA (n = 4–7). (D) Albumin:creatinine ratio was measured in both groups by ELISA (n = 10–11). (E) SBP was measured by tail-cuff weekly in male Aicda^+/+ and Aicda^−/− mice infused with 4 weeks of Ang II (490 ng/kg/min), followed by a wash-out period of 2 weeks, and then 2 weeks of Ang II (140 ng/kg/min) (n = 9–11). Data are expressed as box-and-whisker plots (B, C, and D) or mean ± SEM (A and E). n.s., not significant.

3.4. Global B-cell deficiency does not affect the hypertensive response to Ang II or DOCA-salt

Aicda ^−/− mice still produce IgM and in fact produce higher levels of IgM compared to Aicda^+/+ mice (Supplementary material online, Figures S4A and S5A). Thus, to determine whether IgM or non-immunoglobulin-dependent functions of B cells contribute to hypertension, we studied mice with targeted disruption of the membrane exon of the immunoglobulin µ heavy chain gene (µMT^−/−). Due to this targeted gene disruption, B cells in these mice are mostly arrested at the pre-B-cell maturation stage. µMT^−/− mice are considered to be more severely B cell deficient than BAFF-R^−/− mice.²¹ We first treated male µMT^−/− and WT mice with 490 ng/kg/min Ang II for 4 weeks. Systolic BP, diastolic BP, and HR were not different between the two groups by telemetry at baseline or during 4 weeks of Ang II infusion (Figure 5A). Aortic flow cytometry following 4 weeks of Ang II infusion demonstrated that µMT^−/− mice had virtually no B cells leading to reduced total aortic leukocytes. There was no change in aortic T cells, T-cell subsets, or macrophages (Figure 5B). Furthermore, there was no difference in Ang II-induced aortic medial hypertrophy, CD4⁺ T-cell production of IL-17A and IL-21, a slight reduction in CD8⁺ T-cell production of IFNγ, and no change in albuminuria following 4 weeks of Ang II infusion (Figure 5C–F). Finally, global B-cell deficiency did not affect the hypertensive response to repeat Ang II infusion (Figure 5G).

Figure 5.

Global B-cell deficiency does not affect the hypertensive response to Ang II. (A) SBP, DBP, and HR were measured invasively weekly using carotid radiotelemetry over 28 days of Ang II infusion (490 ng/kg/min) in male WT and µMT^−/− mice (n = 3–5). (B) Flow cytometric quantification of total leukocytes (CD45⁺), T cells (CD3⁺), CD4⁺ T cells, CD8⁺ T cells, B cells (CD19⁺), and macrophages (F4/80⁺) in the aorta of both groups (n = 8–9). (C) Representative images of bright-field aortic media thickness by Picrosirius Red staining. Scale bar: 100 µm. (D) Quantification of aortic media thickness in male WT and µMT^−/− mice infused with vehicle or Ang II for 4 weeks (n = 5–7). (E) Splenic CD4⁺ T-cell production of IL-17A and IL-21 and CD8⁺ T-cell production of IFNγ were quantified in both groups by ELISA (n = 5–8). (F) Albumin:creatinine ratio was measured in both groups by ELISA (n = 10–13). (G) SBP was measured by tail-cuff weekly in male WT and µMT^−/− mice infused with 4 weeks of Ang II (490 ng/kg/min), followed by a wash-out period of 2 weeks, and then 2 weeks of Ang II (140 ng/kg/min) (n = 21–24). Data are expressed as box-and-whisker plots (B, D, E, and F) or mean ± SEM (A and G); **P < 0.01, ***P < 0.001, ****P < 0.0001 by Student’s t-test or Mann–Whitney U test (B and E) or two-way ANOVA (D).

We then investigated the effect of global B-cell deficiency in an independent model of salt-sensitive hypertension characterized by uninephrectomy, implantation of a DOCA pellet, and 1% NaCl in the drinking water for 3 weeks. BP was not different between male µMT^−/− and WT mice during DOCA-salt treatment (Figure 6A). Furthermore, there was no difference in endothelium-dependent or -independent relaxation of mesenteric arterioles after 3 weeks of DOCA-salt treatment (Figure 6B). Aortic flow cytometry demonstrated virtual absence of B cells and thus total leukocytes as well as a modest reduction in macrophages but no change in T cells or T-cell subsets following DOCA-salt treatment (Figure 6C). Finally, glomerular injury as assessed by albuminuria was also not different between the two groups (Figure 6D). Taken together, these data suggest that global deficiency of B cells has no effect on experimental hypertension or hypertensive end-organ damage.

Figure 6.

Global B-cell deficiency does not affect the hypertensive response to DOCA-salt. (A) SBP was measured weekly by tail-cuff over 21 days of DOCA-salt treatment in male WT and µMT^−/− mice (n = 5–8). (B) Endothelium-dependent relaxation in response to increasing doses of Ach (left) and endothelium-independent relaxation in response to increasing doses of SNP (right) were measured in isolated mesenteric arterioles (n = 3). (C) Flow cytometric quantification of total leukocytes (CD45⁺), T cells (CD3⁺), B cells (CD19⁺), macrophages (F4/80⁺), CD4⁺ T cells, and CD8⁺ T cells in the aorta of both groups (n = 6–8). (D) Albumin:creatinine ratio was measured in both groups by ELISA (n = 6–8). Data are expressed as mean ± SEM (A and B) or box-and-whisker plots (C and D); *P < 0.05, **P < 0.01, ***P < 0.001 by Student’s t-test or Mann–Whitney U test (C). n.s., not significant.

4. Discussion

For over 50 years, subsets of immunoglobulins, particularly IgG, have been shown to be elevated in hypertensive animals and humans,²² but whether these immunoglobulins are causal to the pathophysiology of disease or simply a biomarker reflective of immune activation is unclear. Here we show that adoptive transfer of IgG from hypertensive mice is not sufficient to raise BP in recipient animals or induce significant end-organ damage. Furthermore, loss of high-affinity class-switched immunoglobulins also has no effect on hypertension development. Contrary to two other recent reports using different models of B-cell deficiency,^10,13 we found no effect of severe global B-cell deficiency on baseline BP or development of hypertension and hypertensive end-organ damage in response to two different hypertensive stimuli—Ang II and DOCA-salt. Taken together, our findings suggest that elevated serum immunoglobulins are a bystander or biomarker of immune activation in hypertension but not necessarily causal to disease pathology.

Several studies in humans have shown elevated levels of serum immunoglobulins, particularly IgG, in hypertensive patients compared with normotensive controls.^6–8 Furthermore, serum IgG was shown to positively correlate with BP in untreated patients.⁷ Murine studies by our group and others corroborated increased serum immunoglobulins in hypertension and demonstrated IgG deposition in hypertensive mouse aortas.^9,10 Renal infarction-induced hypertension in rats showed elevated serum antibodies which bound to arteries and kidneys, implicating a possible causative role for these antibodies.²³ In contrast, a recent study from the ASCOT demonstrated that high total serum IgG levels were an independent predictor of freedom from adverse cardiovascular events in patients with hypertension, suggesting that IgG may play a protective role.¹¹

B-cell immunoglobulin production is primarily dependent on a GC reaction in which GC B cells undergo proliferation and differentiation into immunoglobulin secreting plasma cells and long-lived memory B cells. Somatic hypermutation (to generate high-affinity antibodies) and isotype class switching (e.g. IgM → IgG) occurs in germinal centres and is dependent on the enzyme AID. The most potent inducer of a GC reaction is the cytokine IL-21, which is produced by Tfh cells and Tfh-like peripheral helper T (Tph) cells. We recently demonstrated that Ang II-induced hypertension is associated with an increase in Tfh, Tph, and GC B cells in the aorta, increased serum IgG1, and increased T-cell production of IL-21. Pharmacologic or genetic deletion of IL-21 reduced aortic inflammation, IgG1 production, BP, and vascular damage in response to hypertensive stimuli.⁹ Based on this study, we postulated that the effect of IL-21 on hypertension was in part through promoting a GC reaction and immunoglobulin production. However, based on our current findings, it is likely that the GC reaction and increased IgG1 production is a consequence of increased IL-21 signaling that does not actually contribute to disease pathology. Rather, the effects of IL-21 on hypertension are likely through non-B cell-related pathways. In this regard, we demonstrated that IL-21 has profound effects on T-cell cytokine production, promoting both IL-17A and IFNγ production by T cells, and that IL-21 has direct effects on endothelial nitric oxide production.⁹ Further studies are needed to determine the role of Tfh and Tph cells in hypertension and why and how these cells are activated by hypertensive stimuli.

There are important differences in B-cell subsets between our µMT^−/− and Aicda^−/− mouse models and the anti-CD20 monoclonal antibody and BAFF-R^−/− mouse models used by Chan et al.¹⁰ B cells are classified as B1 or B2 cells. B1 cells are innate-like cells found predominantly in the peritoneal and pleural cavities which produce natural IgM antibodies. Conventional B cells are called B2 cells. Peritoneal B1 cells are relatively resistant to anti-CD20 treatment.²⁴ In addition, BAFF-R^−/− mice have a preferential reduction in B2 B cells.²⁵ Thus, both of these models retain some degree of B1 cells, and this increased B1/B2 ratio may be protective in hypertension. In keeping with this, B2 cells are pathogenic and B1 cells are protective in mouse models of atherosclerosis.^26,27 The µMT^−/− mice are severely B cell deficient with almost complete absence of both B1 and B2 cells. Furthermore, B1 cells were recently shown to express AID, and thus Aicda^−/− mice may also have defects in B1 cell function.²⁸ Therefore, our models are characterized by virtually absent or defective B1 and B2 cells while those used by Chan et al.¹⁰ retain functional B1 cells. Further studies are needed to determine if these innate-like B1 cells play a protective role in hypertension that might be masked by global B-cell deficiency.

Dingwell et al.¹³ studied a slightly different mouse model of global B-cell deficiency, namely mice with deletion of the gene for the heavy chain joining region (J_HT). While these authors showed a modest (10 mmHg) reduction in baseline anaesthetized BP in J_HT mice, they did not assess conscious BP or their response to hypertensive stimuli such as Ang II and DOCA-salt. The primary findings from that paper were in the context of c-myb deficiency which is a significant confounding variable.

While the µMT^−/− mouse is a commonly used model of global B-cell deficiency, they are not necessarily completely devoid of all B cells. These mice retain the ability to produce some B cells using a non-canonical pathway that bypasses the mu-chain route leading to production of non-specific IgG and IgE under extenuating circumstances, such as following sensitization and repeated challenge with live fungus.²⁹ To determine if a hypertensive stimulus is sufficient to induce this phenomenon, we quantified serum immunoglobulins as well as B cells in the peritoneal cavity and mesenteric lymph nodes of WT and µMT^−/− mice following 28 days of Ang II infusion (Supplementary material online, Figure S6). We found that in response to Ang II infusion, serum IgG subtypes, IgM, and IgA are absent in µMT^−/− mice compared to WT mice. Furthermore, in the peritoneal cavity, B cells compose on average 0.2% of total lymphocytes in µMT^−/− mice vs. 15% in WT mice. In the mesenteric lymph node, B cells compose on average <0.01% of total lymphocytes in µMT^−/− mice compared to 43% in WT mice. Our results are consistent with that of Tay et al.³⁰ showing that in the context of ApoE deficiency and 8 weeks of high-fat diet, µMT^−/− ApoE^−/− mice have no detectable levels of B cells in the peripheral blood, lymph nodes, spleens, and peritoneal cavities. Nevertheless, we cannot rule out that a small number of immature B cells may exist in these animals and contribute to hypertension development.

In addition, the µMT^−/− mice tend to have defects in other aspects of the immune system that can potentially mask a protective effect of removing B cells. Spleen micro-architecture is partially dependent on B cells, and Ngo et al.³¹ showed that splenic T cell and dendritic cell numbers are reduced in µMT^−/− mice. However, we detected roughly equal numbers of T cells in the aorta of µMT^−/− mice compared to WT mice in response to Ang II and DOCA-salt-induced hypertension. There is also evidence for defects in CD4⁺ and CD8⁺ T-cell responses in µMT^−/− mice.³² Nevertheless, we observed no significant change in splenic T-cell cytokine production of IL-17A and IL-21, suggesting that production of these key pro-hypertensive cytokines are not altered in our model and may be sufficient to drive hypertension in the absence of B cells.

Our findings are consistent with those of Guzik et al.¹² showing that adoptive transfer of B cells into recombination activating gene 1 (RAG1) deficient mice, which lack both T and B cells, did not restore the hypertensive response to Ang II infusion but adoptive transfer of T cells did, suggesting that B cells do not play an important role in hypertension. One criticism of that experiment is that B cells rely on T cells for help, and thus it is unclear whether the adoptively transferred B cells received the proper activation signals. However, it is important to note that the RAG1^−/− mice that only received T cells (and thus remained deficient in B cells) were not protected from Ang II-induced hypertension or endothelial dysfunction.

It is possible that B cells and immunoglobulins contribute to hypertension in specific cases such as in autoimmune diseases or preeclampsia. Autoantibody immune complexes can be found in the glomerular and tubular basement membranes in lupus but whether these autoantibodies contribute to SLE-associated hypertension is unclear.²² Preeclampsia is associated with activating Ang II type-1 receptor autoantibodies (AT1R-AA) with antibody titre correlating with disease severity. Interestingly, administration of human AT1R-AA to pregnant mice did induce hypertension in the recipient animals.²²

In summary, while our results suggest that class switching and high-affinity IgG production is dispensable for the development of hypertension, we cannot rule out modulatory effects of individual B-cell subsets. Furthermore, given the limitations and risk of compensatory changes that can occur with genetically modified animals, future studies are needed using acute pharmacological depletion of individual B-cell subsets or inducible genetic B-cell deletion in adult animals to determine whether B cells are a viable therapeutic target for hypertension.

Data availability

All data are incorporated into the article and its online supplementary material.

Supplementary material

Supplementary material is available at Cardiovascular Research online.

Conflict of interest: none declared.

Funding

This work was supported by the National Institutes of Health [DP2HL137166 to M.S.M.], the American Heart Association [EIA34480023 and IPLOI34760558 to M.S.M., 17PRE33460032 to B.L.D.], and the China Scholarship Council [to Y.C.].

 

Time for primary review: 34 days