Hypertensive Brain Injury as a Journey Not a Destination

Hypertension affects nearly half of adults in the United States, with its global prevalence rising most rapidly in low‐ and middle‐income countries. Uncontrolled hypertension substantially increases the risk of brain damage, making it one of the most important modifiable cardiovascular risk factors for all‐cause dementia. ¹ Several studies have shown the beneficial effect of controlling blood pressure when it comes to the risk of dementia. ² , ³ Yet, hypertension remains difficult to manage with ∼40% of patients living with uncontrolled blood pressure despite available therapies. This underscores the urgent need to uncover novel mechanisms of disease progression that could inform the next generation of targeted pharmacotherapies.

Long before irreversible brain injury occurs, the cerebral microvasculature undergoes critical molecular and cellular changes that may represent a therapeutic window. Particularly, cerebral small vessel disease (cSVD) is a key contributor to cognitive decline in hypertension. Although hypertension alone increases the risk of dementia, cSVD is particularly impactful as it is strongly associated with cognitive decline, mood disturbances, and balance problems and contributes to ∼25% of ischemic strokes. ¹ Among imaging markers used to diagnose cSVD, white matter hyperintensities, cerebral microbleeds (CMBs), and enlarged perivascular spaces are most prominent in hypertension. ¹ In this study, Arndt et al. ⁴ used stroke‐prone spontaneously hypertensive rats (SHRSP) that develop both CMBs and enlarged perivascular spaces alongside high blood pressure and cognitive decline to better understand how the length of hypertension affects brain health. The SHRSP model was chosen because these animals reliably develop hypertension and cSVD without additional interventions, making them relevant for modeling vulnerable patients in the early and preclinical stages of disease. The authors approached hypertension as a continuum and characterized molecular and cellular events driving disease progression, offering insights into when and how interventions might be most effective. Focusing on the frontal cortex, a region that is particularly vulnerable in patients with hypertension across multiple modalities, including cognitive testing, gene expression studies, and neuroimaging, the authors dissected temporal mechanisms across distinct stages of hypertension (Figure): initial (6–8 weeks old), early chronic (14–15 weeks old), and late chronic (32–34 weeks old) hypertension. Wistar controls at each stage allow for separating the effects of aging versus hypertension length. Of note, the rats studied in both early and late chronic hypertension groups were stroke‐free. Because 70% to 80% of SHRSP develop stroke by 9 months, ⁵ , ⁶ the findings at each time point may reflect distinct disease trajectories. Namely, many rats in the “initial” hypertension group will later develop a stroke, whereas those in the “late chronic” group have survived stroke free to this age. Although future studies are warranted in hypertensive models that develop cSVD but are not stroke prone, these findings nonetheless provide important insights into the molecular mechanisms underlying brain injury with increasing history of hypertension.

Figure 1. Neurovascular and immune changes across length of hypertension.

In the initial stage of hypertension in stroke‐prone spontaneously hypertensive rats (6–8 wk of age), microglia exhibit enlarged somata and there are more VAM. By the late chronic stage, the increase in VAM is predominantly in junction‐associated VAM and parenchymal microglia exhibit increased branching. By the early chronic stage (24–25 wk), BBB disruption, CMBs, and leukocyte infiltration into the parenchyma emerge. Enlarged perivascular spaces are present at all stages and become more prevalent as hypertension continues. Although motor activity is decreased in the initial stage of hypertension, cognitive impairment is first detected in the early chronic stage. BBB indicates blood–brain barrier; CMB, cerebral microbleed; and VAM, vessel‐associated microglia.

As expected, SHRSP animals developed cognitive deficits with increasing length of hypertension history, including increased time in the center during open field test and impaired prepulse inhibition. Prepulse inhibition is a measure of attention and sensorimotor gating, which refers to the ability to filter out competing or conflicting irrelevant stimuli, and is primarily regulated by the frontal cortex. Prepulse inhibition deficits have been previously shown in several neurocognitive disorders, including Alzheimer disease. These findings suggest that chronic hypertension is marked by impaired cognition and cortical circuit dysfunction, highlighting a critical window to investigate mechanisms underlying disease progression before apparent cognitive impairments. The progression of cognitive decline was associated with increased prevalence of both CMBs and enlarged perivascular spaces. Although these findings are generally in accordance with previous studies by these authors and other groups showing age‐dependent increase in CMBs and enlarged perivascular spaces in SHRSP, ⁷ minor differences were noted. For example, this group had previously described that CMBs were not observed in animals younger than 32 weeks of age, ⁸ whereas here they report the presence of CMBs in animals younger than 20 weeks of age. A key distinction between the studies lies in the distribution and timing of microbleeds. In their previous work, microbleeds appeared irregularly across multiple brain regions and at varying time points, whereas the current study focused specifically on the frontal cortex. This regional heterogeneity highlights differences in pathological progression. Together, these observations underscore the contributions of cerebrovascular pathology to cognitive decline, offering valuable insight into the temporal dynamics of hypertension‐related brain injury.

The study also reports a shift in neuroimmune markers with disease progression. As previously shown, ⁹ microglia developed enlarged somata in the initial stage of hypertension with increased branching at the late chronic stage. Notably, vessel‐associated microglia (VAM) increased in the initial stage, persisted throughout the course of disease, and were preferentially located at vascular junctions in the late chronic hypertensive stage. The increase in VAM is highly relevant since they regulate cerebrovascular tone ¹⁰ and preferentially express AT1aR (angiotensin II receptor type 1), ¹¹ which promotes proinflammatory signaling in these microglia. Indeed, microglia association to microvessels has been suggested to be related to decreased cerebral blood flow in 12 weeks of age in SHRSP ¹² and increased blood–brain barrier (BBB) leakage and inflammation. ¹³ In this study, the increase in VAM preceded BBB disruption, suggesting recruitment driven by changes in neurovascular tone. BBB was most severe during the early chronic stage and was accompanied by robust infiltration of peripheral leukocytes into the cortex without further changes in VAM. However, quantifying BBB leakage by immunoglobulin G extravasation may underestimate early BBB disruption, given its high molecular weight (150 kDa). We have previously used a small molecular weight (3 kDa) tracer to show that BBB disruption is present early in disease in the BPH/2 mouse model of chronic hypertension even before the development of cognitive impairment, ¹⁴ raising the possibility that microglial recruitment in the initial stage also reflects subtle BBB changes missed here. These findings suggest that early microglial recruitment to the vasculature may reflect subtle neurovascular disturbances that precede overt BBB disruption, highlighting VAM as potential early therapeutic targets in hypertension‐induced brain injury.

Another interesting point reported by the authors is that the BBB disruption observed in the early chronic hypertensive stage was absent in the late chronic hypertensive stage. This reversal in BBB disruption is unexpected, but the authors attribute it to an enrichment of the extracellular matrix and cell adhesion molecules, including Cdh2 and Cdh5, known to promote BBB integrity. However, their prior work showed increasing BBB disruption in pooled analysis of the hippocampus, basal ganglia, cortex, and corpus callosum of SHRSP animals as they age, ⁸ consistent with our findings in the BPH/2 mouse model of chronic hypertension. ¹⁴ Reconciling these results, it is possible that chronic hypertension differentially affects the BBB depending on the brain region, highlighting the need to further investigate the regional vulnerability of the BBB. Thus, the therapeutic window to target BBB disruption across brain regions remains to be determined.

The authors next performed bulk RNA‐sequencing on laser microdissected vascular or neural/parenchymal tissue and found significant differences in arterial smooth muscle cell, fibroblast, and endothelial specific gene transcripts between early and late chronic disease. Consistent with BBB disruption, early chronic hypertension showed increased expression of blood microparticle transcripts in neural tissue, which can access the parenchyma only through a leaky barrier. In line with heightened leukocyte infiltration, this stage was also enriched for immunological processes related to chemotaxis and innate immune cell migration. The authors also found enriched transcripts of the complement cascade including C1qa and C1qc during early chronic hypertension. The complement cascade plays a key role in enhancing immune clearance through phagocytosis and inflammation; however, its dysregulation has been implicated in cognitive decline during Alzheimer disease pathology. ¹⁵ , ¹⁶ Particularly, C1q and C3 have been shown to mediate synaptic pruning in various mouse models of Alzheimer disease. In agreement with this role of the complement cascade, the authors found a reduction in synapse‐specific genes across both chronic hypertension stages, including a downregulation of genes associated with the modification of synaptic structure and downregulation of genes involved in synapse density. Therefore, the immune activation in the SHRSP model may underlie the cortical circuit dysfunction contributing to the cognitive deficits observed in this model.

Interestingly, compensatory vascular mechanisms emerged in late chronic hypertension. Endothelial fluorescent activated cell sorting analysis and quantitative polymerase chain reaction of isolated cortical microvessels revealed increased angiogenesis and upregulation of endothelial markers, suggesting an adaptive remodeling response. Upregulated genes supported increased vessel density and endothelial proliferation, in line with enhanced angiogenesis. In contrast, downregulated genes were predominantly neuronal, involving synaptic and axonal compartments, consistent with altered neuronal transcriptomic profiles. These findings suggest that chronic hypertension is associated with impaired remodeling of neural and vascular circuits, possibly contributing to the observed development of cognitive deficits.

Crucial for disease pathology, Arndt et al., showed a decrease in genes associated with cerebrospinal fluid circulation and regulation of blood vessel size during late chronic hypertension, suggestive of an impaired glymphatic system. In agreement with a possible glymphatic impairment in this model, it has been previously reported that the parent strain of SHRSP, the SHR, develop impaired glymphatic transport kinetics, which includes parenchyma waste clearance, starting at 7 to 9 weeks and continues with age. ¹⁷ Paradoxically, despite decrease cerebrospinal fluid circulation genes, SHRSP animals exhibited an increase in genes associated with beta‐amyloid clearance in the neural tissue. Gene ontology analysis also revealed transcripts associated with heightened metabolic demand. Consistent with this, ¹⁸F‐fluorodeoxyglycose positron emission tomography showed a ~30% increase in glucose uptake during early chronic hypertension particularly in the cingulate, retrosplenial cortex, motor, and somatosensory cortices. These changes were not regionally shifted but significantly attenuated during late chronic hypertension. Interestingly, previous work found no significant differences in glucose metabolism in 3‐ or 10‐month‐old male SHRSP without stroke symptoms. ¹² Furthermore, glucose uptake is also not altered in young SHR (12–16 weeks of age), which do not develop CMBs, and actually decreases in 15 brain regions in aged SHRs (64–68 weeks of age). ¹⁸ In humans, there is a strong association between hypertension and a progressive decrease in glucose metabolism, which is strongest in apolipoprotein E ɛ4 carriers. ¹ Indeed, glucose metabolism is also reduced in cSVD, ¹⁹ , ²⁰ but it may be important to differentiate between cSVD pathologies. As the authors mention, glucose consumption may increase during experimental stroke and intracerebral hemorrhage, associated with an activation and infiltration of immune cells. Thus, the increased glucose uptake observed here may likely be a result of the CMBs and not chronic hypertension itself.

Taken together, this report provides valuable insight into the molecular mechanisms contributing to cognitive deficits in chronic hypertensive cSVD, briefly summarized in the Figure 1. Particularly, this work highlights the importance of vascular‐immune interactions and how they change with disease progression. Given the strong relationship between midlife hypertension and late‐life dementia, unraveling the pathophysiological mechanisms underlying cognitive decline in hypertension remains a public health priority. Here, the authors offer insight into the link between hypertension and dementia. They report that 7 out of the top 45 Alzheimer disease risk genes were altered in either the hypertensive cerebrovascular or neural tissue, and most of these differentially expressed genes are highly expressed by microglia. This finding emphasizes the importance of understanding how vascular‐associated microglia differ from their parenchymal counterparts and identifying their role in both hypertension and dementia disease progression. In summary, the current study provides a comprehensive understanding of the disease course of hypertensive cSVD and provides the foundation to identify novel molecular targets to prevent hypertensive brain injury, specifically, if the disease is caught early in its development.

Disclosures

None.