Hypertension impacts peripheral blood leukocyte composition

In this issue of Hypertension, Kresovich et al.¹ investigated how the immune system is affected by the presence and occurrence of high blood pressure, comparing women enrolled in the Sister Study with and without a diagnosis of hypertension and those with a de novo diagnosis in a 15-year follow-up. This prospective cohort collected information from 4,124 women and prospectively identified incident cases of hypertension. A blood sample was collected at baseline, but no flow cytometry information was collected. Specifically, this manuscript adds epidemiological evidence to a potential association between an increase in B cell proportions and a reduction in CD4(+) T naïve cells and incident high blood pressure risk. In addition, they observed a higher proportion of neutrophils and lower proportions of CD4(+) T-cells, natural killers, and B cells in patients with hypertension compared to those without hypertension. The authors systematically evaluated the relationships between cell composition and systolic and diastolic blood pressure levels, the presence or absence of high blood pressure, and several groupings of the cells. The associations between increased B cell proportions and incidence of hypertension persisted in all the sensitivity analyses, similar to the association with systolic and diastolic blood pressure at baseline. Higher neutrophil-to-lymphocyte and lower lymphocyte-to-monocyte and CD4T/CD8T cell ratios were observed in prevalent hypertension.

In the absence of flow cytometry information, to conduct this investigation, the researchers used high throughput DNA methylation information (Illumina HumanMethylation microarrays) and resolved the cell composition using cell deconvolution. This technique utilizes unique “epigenetic fingerprints” related to cell-type identity as a reference and retrieves the relative cell proportions in the peripheral blood using a statistical method called constrained projection/quadratic programming². In this manuscript, the authors used, for the first time, an enhanced reference capable of resolving 12 different cell types³. Using a DNA substrate allowed the interrogation of the cell composition in stored samples that could not be obtained using flow cytometry techniques. DNA, unlike RNA or cell surface markers, is stable and less prone to critical biases due to cell degradation. Roughly, DNA methylation (a covalent methylation bond to specific cytosines in the DNA strand) demarcates active vs. inactive transcriptional areas, some unique to specific cell types. Cell deconvolution techniques offer new avenues for “DNA methylation cytometry” interrogation for researchers in different medical fields⁴.

Interactions between leukocytes and blood pressure are a critical component of immune regulation that develops in response to acute and chronic stress. High blood pressure arises from a complex dysregulation of blood volume (sodium-water retention/excretion balance), cardiac output, and vascular rarefaction. Traditionally, this is related to an overactivated sympathetic nervous system, releasing catecholamines, increasing the cardiac output, generating vasoconstriction, and secondary activation of the renin-angiotensin-aldosterone system (RAAS), further increasing vasoconstriction and promoting re-uptake of sodium/water in the kidney, expanding the blood volume. However, as with many complex diseases, other systems in the body are affected simultaneously, and more multifaceted endothelial-immune dysfunctions can play a role in the origin and evolution of high blood pressure.

The link between inflammation and endothelial dysfunction has been teased in human autopsies⁵ showing autoantibodies in major arteries from hypertensive subjects and from small observational studies monitoring immunoglobulins levels in subjects with hypertension⁶. Similarly, mouse models have shown an inconsistent role for B cells immunoglobulin production in these hypertension models^7,8. Several murine models have shown how T cell/mononuclear cell activation led to experimental hypertension, while experimental immunosuppression/increase in regulatory cells prevented high blood pressure in such models⁹. For the T-cell component, immunosuppressed mice have shown a role of CD4(+) and CD8(+) T cells^10,11, and a potential link with CD4(+) T cells expressing interleukin-17A with an accumulation of interferon-gamma producing T cells¹⁰. In particular, the role of CD8(+) T cells seems critical for the development of experimental hypertension in experimental models and should be further explored¹¹. Mechanistically, interleukin-17A is attractive as several tissues could be targeted simultaneously, including endothelium, vascular smooth muscle, fibroblasts, and kidney-specialized epithelium¹². Also experimentally, three mechanisms have been proposed to explain this: impaired vasodilation due to endothelial dysfunction, depletion of blood vessels due to inflammation, and vascular stiffening due to immune infiltration in the vessel wall¹³. Interestingly, in humans, the information is still scarce, with few examples of epidemiological studies associating immune subpopulation changes in normotensive and hypertensive subjects not taking any antihypertensive medications^14,15, and hypertensive patients receiving immunosuppressants related to autoimmune disorders¹⁶. These small studies looked at changes in the immune subpopulation averages that were different in patients with hypertension vs. those normotensive (even if they were not statistically significant) and how patients receiving immunosuppression showed some improvement in their blood pressure.

Kresovich et al.¹ associations are consistent with findings in the animal models⁹, as well as more complex Mendelian Randomization studies¹⁷. A significant limitation with human studies is the limited access to samples of target organs (kidney, brain, large blood vessels), where some of the alterations related to the immune cell responses may be regulated.^5,11,15. From the epidemiological perspective, Kresovich et al.¹ has some limitations in understanding differences in associations between different genetic ancestries (partially explored here using self-reported race as a proxy). This lack of population representation is an important issue that must be discussed in planned cohort studies; clarity on ancestry can help uncover potential associations that are differentially driven by any genetic component. Including more diverse populations also brings health disparities issues to the forefront for future work¹⁸. The inverse relation between natural killers and cytotoxic T-cells by reported race is a gap that must be addressed in future studies. Unfortunately, the method used to interrogate the cell proportions cannot establish finer cell activation layers relevant for hypertension, such as the presence of TH17 T cells. On the other hand, the fact that this study used a women’s cohort brings other exciting questions, given that some of them must have experienced pregnancy and that preeclampsia is also associated with autoantibodies and potentially higher risk of high blood pressure later in life¹⁹. Other inflammatory associations, such as the abnormal neutrophil activation in preeclampsia, are additional avenues of research that may benefit from these methodological approaches²⁰. At the same time, the results observed in this study may not be directly extrapolated to male populations, given sex-specific immune responses that have not been fully clarified in the literature due to a historically reduced representation of females in both experimental and clinical/epidemiological studies²¹.

In the future, additional applications of DNA methylation cytometry using cell count information will help uncover other important relations, particularly differences related to genetic ancestry and sex differences. This large observational study suggests that humans’ immune subpopulations are linked to hypertension. Hopefully, this or similar methods will be incorporated as part of clinical practice to complement the complete cell blood count. Understanding the more subtle layers that characterize inflammation in chronic diseases such as hypertension may offer new avenues for additional personalized therapies complementing our current approaches.