Ne amapimod Induces Vasodilation in Resistance Mesenteric Arteries by Inhibiting p38 Mitogen- Activated Protein Kinase Alpha (MAPKα) and Heat- Shock Protein 27 (Hsp27) Phosphorylation


 Neflamapimod, a selective inhibitor of p38 MAPKα, is under clinical investigation for its efficacy in Alzheimer’s disease (AD) and dementia with Lewy Bodies (DLB). Here, we investigated if neflamapimod-mediated acute inhibition of p38 MAPKα could induce vasodilation in resistance-size rat mesenteric arteries. Our pressure myography data demonstrated that neflamapimod produced a dose-dependent vasodilation in mesenteric arteries. Our Western blotting data revealed that acute neflamapimod treatment significantly reduced the phosphorylation of p38 MAPKα and its downstream target heat-shock protein 27 (Hsp27) that is involved in cytoskeletal reorganization and smooth muscle contraction. Likewise, non-selective inhibition of p38 MAPK by SB203580 attenuated p38 MAPKα and Hsp27 phosphorylation, and induced vasodilation. Endothelium denudation or pharmacological inhibition of endothelium-derived vasodilators such as nitric oxide (NO) and prostacyclin (PGI2) had no effect on such vasodilation. Neflamapimod-evoked vasorelaxation remained unaltered by the inhibition of smooth muscle cell K+ channels. Altogether, our data for the first time demonstrates that in resistance mesenteric arteries, neflamapimod inhibits p38 MAPKα and phosphorylation of its downstream actin-associated protein Hsp27, leading to vasodilation. This novel finding may be clinically significant and is likely to improve systemic blood pressure and cognitive deficits in AD and DLB patients for which neflamapimod is being investigated.

Introduction p38, a member of the mitogen-activated protein kinase (MAPK) family, is traditionally known to be a stress sensor that is activated by physical and chemical stressors, in ammatory mediators, vasoactive compounds, and growth factors. [1][2][3][4] Among the four isoforms of p38 MAPK, namely α, β, γ and δ, [5,6] p38 MAPKα is present in most cell types and is best characterized for its role in chronic in ammatory processes linked to degenerative diseases of the nervous system, musculoskeletal system, and cardiovascular system [7] . In microglia, p38 MAPKα activation is associated with the release of proin ammatory cytokines. Stimulation of p38 MAPKα signaling in neurons regulates tau localization and neuronal plasticity [8][9][10] . In addition, p38 MAPKα regulates the function of a Ras-related protein Rab5, which is a key regulator of early endosomal tra cking involved in a wide variety of cell signaling and cellular processes. Due to the involvement of p38 MAPKα in different neuronal pathologies, p38 MAPKα inhibitors have recently attracted considerable attention as potential therapeutics for in ammatory and neurodegenerative diseases, such as Alzheimer' Disease (AD), Dementia with Lewy Bodies (DLB) and Huntington' Disease (HD) [11,12] . Ne amapimod, a selective inhibitor of p38 MAPKα, is under clinical investigation for its e cacy in improving cognitive de cits in AD and DLB. Preclinical and recent phase II clinical trial data show that ne amapimod treatment is effective in reversing cognitive decline in DLB [13] , however, its e cacy in reversing AD-related cognitive de ciencies remains somewhat unclear. Although the exact mechanism is not fully known, it is established that hypertension and vascular dysfunction drive neurodegeneration and cognitive dysfunction [14] . Hypertension and altered vascular tone are implicated in vascular cognitive impairment that encompasses all sorts of cognitive de cits arising from vascular dysfunction [14][15][16][17] . Since vascular tone is a key regulator of peripheral resistance and systemic blood pressure, drugs that relax small resistance arteries are expected to lower systemic blood pressure and enhance cognitive function.
Although p38 MAPKα has been extensively studied in the context of neuroin ammation and neurodegeneration, little is known about its role in vascular contractility. According to a previous study [18] , high-glucose or diabetes activates p38 MAPK in rat aortic smooth muscle cells. As p38 MAPK activation often re ects gene expression changes leading to cell growth or apoptosis [19][20][21] , glucose-or diabetesassociated activation of p38 MAPK was suggested to regulate vascular in ammation, thrombosis and other pathological changes [4,22] . Subsequent studies reported that p38 MAPK activation also regulates the contractility of resistance arteries. Vasoconstrictors such as angiotensin II and norepinephrine were reported to cause p38 MAPK activation leading to smooth muscle contraction [7,23,24] . This raises the possibility that acute inhibition of p38 MAPK could stimulate vasodilation, which could reduce vascular resistance and systemic blood pressure. Non-selective p38 MAPK inhibition by SB203580 relaxed high K + -contracted renal arteries and enhanced acetylcholine-induced vasodilation [4] . It has been suggested that p38 MAPK activation may in uence the function of the cardiovascular system by regulating systemic and renal hemodynamics. To our knowledge, no previous studies have investigated what effect selective p38 MAPKα inhibition by ne amapimod may have on the contractility of resistance mesenteric arteries. Here, we examined if ne amapimod could relax resistance-size rat mesenteric arteries and its mechanism of action. Our data demonstrates that acute ne amapimod treatment stimulates dosedependent vasodilation in mesenteric arteries, in a manner consistent with p38 MAPKα inhibition and reduced Hsp27 phosphorylation, and independent of endothelial signaling and smooth muscle cell K + channels. Such novel vasodilatory action of ne amapimod may be clinically translatable and is likely to improve systemic blood pressure and cognitive de cits in AD and DLB patients.

Ne amapimod stimulates vasodilation in resistance mesenteric arteries
To examine if acute ne amapimod treatment could regulate arterial contractility, we performed pressurized arterial myography on resistance 3rd to 5th order mesenteric arteries from male Sprague Dawley rats. Cannulated arterial segments were maintained in a temperature-controlled perfusion chamber (Living Systems), continuously perfused with 37°C PSS, gassed with a mixture of 21% O2, 5% CO2, and 74% N2. Intravascular pressure was gradually increased to 80mmHg to stimulate the development of myogenic tone. At 80mmHg, mesenteric arteries developed ~24% myogenic tone, after which a cumulative concentration response (0.0001-10µM) to ne amapimod was performed and diameter changes recorded. Our myography data shows that, within 2-3 minutes of application, ne amapimod produces a fully reversible and concentration-dependent vasodilation ( Fig. 1a-d). The calculated EC 50 was found to be ~0.2 µM (Fig. 1c). Of note, the C max reported in a preclinical study with rats is ~2 µM [11] . Ne amapimod at 10µM concentration potently relaxed mesenteric arteries by ~41µm, or by ~17% of arterial diameter from a baseline diameter at 80 mmHg (Fig. 1d). Considering the reported C max in a previous study, and for a better diameter read out in the presence of vasomotion, we used 10µM concentration for subsequent mechanistic studies. Altogether, our data for the rst time demonstrate that acute ne amapimod application induces vasodilation in resistance-size mesenteric arteries.

Ne amapimod inhibits p38 MAPKα phosphorylation in mesenteric arteries
As ne amapimod is a selective inhibitor of p38 MAPKα, we asked if ne amapimod could inhibit p38 MAPKα phosphorylation and its kinase activity, in mesenteric arteries to stimulate vasodilation. To do so, we incubated mesenteric arteries for 30 minutes with either ne amapimod or SB203580, a non-selective inhibitor of p38 MAPK. Our Western blotting data reveals that mesenteric arteries have a basal p38 MAPKα phosphorylation, as detected by phospho-p38 MAPKα antibody (Fig. 2a, vehicle control, lane 1). Importantly, ne amapimod and SB203580 reduced p38 MAPKα phosphorylation to 33.74±0.3.7% and 32.25±0.1.8% of vehicle-treated control, respectively (Fig. 2a, 2b). This data suggests that both drugs are equally effective in reducing basal p38 MAPKα phosphorylation at the concentrations used. Overall, this data demonstrates that mesenteric arteries have baseline p38 MAPKα phosphorylation, and treatment with 10µM ne amapimod potently suppresses p38 MAPKα phosphorylation and, therefore, its activity in mesenteric arteries.

Ne amapimod reduces phosphorylation of actin-associated protein Hsp27
A critical downstream target of p38 MAPK is Hsp27 which, when phosphorylated, promotes actin lament formation and smooth muscle contraction. Earlier studies reported that reduction of Hsp27 phosphorylation by non-selective p38 MAPK inhibition with SB203580 stimulates vasodilation in mesenteric arteries [7,23,24] . This led us to test the hypothesis that ne amapimod-mediated selective inhibition of p38 MAPKα may also decrease Hsp27 phosphorylation to elicit vasodilation. To test this, we processed mesenteric arteries that had been treated with either ne amapimod or SB203580 for Western blotting. Our data revealed that ne amapimod-treated arteries had mean 37.75±1.19% phosphorylation of Hsp27, or 62.25% less compared to that in control arteries (Fig. 3a, 3b). Likewise, non-selective p38 MAPK inhibitor SB203580 also reduced arterial Hsp27 phosphorylation to 31.57±3.47% or by 69%. Combined application of ne amapimod and SB203580 reduced Hsp 27 phosphorylation to 29.5% ±5.66%, or by 71% (Fig. 3a, 3b), which is similar to the inhibition produced by either drugs alone. These data suggest that ne amapimod-induced selective inhibition of p38 MAPKα reduces the phosphorylation of its downstream target Hsp27, an actin-associated protein that takes part in cytoskeletal reorganization and smooth muscle contraction.
We next asked if the inhibition of basal Hsp27 phosphorylation by selective p38 MAPKα inhibitor ne amapimod and non-selective p38 MAPK inhibitor SB203580 correlate with their vasodilatory actions in mesenteric arteries. Our pressure myography data showed that, like ne amapimod, SB203580 application relaxed mesenteric arteries by ~44 µm, which is slightly higher than that produced by ne amapimod (Fig 3c, 3d). When ne amapimod and SB203580 were applied together, they produced a mean arterial dilation of ~50 µm. Importantly, these data correlates with the magnitude of reduction of Hsp27 phosphorylation in mesenteric arteries. Overall, our data demonstrate that ne amapimod selectively inhibits p38 MAPKα and downstream Hsp27 phosphorylation to induce vasodilation in resistance mesenteric arteries.

Ne amapimod-induced vasodilation does not depend on endothelial signaling
Endothelium is an important regulator of arterial contractility. Endothelial cells induce vasodilation primarily through the production of vasodilators and their paracrine action on underlying smooth muscle cells. Therefore, we next assessed the role of endothelial signaling in ne amapimod-elicited vasodilation.
To determine the role of endothelium, ne amapimod-induced vasodilatory responses in endotheliumintact and -denuded mesenteric arteries were compared. To this end, we performed endothelium denudation by slowing passing air bubbles through the vessel lumen and con rming the absence of 1µM acetylcholine (ACh)-mediated dilation as described previously (Fig 4a) [25][26][27][28] . Both endothelium-intact and -denuded vessels were preconstricted with 1µM phenylephrine (PE). Application of 1µM ACh fully reversed PE constriction in endothelium-intact arteries. In contrast, endothelium-denuded vessel showed only 5% reversal, which is 95% less than that shown by endothelium-denuded arteries (Fig 4a, 4b).
However, application of 1µM sodium nitroprusside (SNP), a nitric oxide (NO) donor, fully reversed PE constriction in both endothelium-intact and -denuded vessels, suggesting that endothelium denudation does not affect smooth muscle responses to NO (Fig 4b). 60K-induced vasoconstriction was also similar between endothelium intact and denuded arteries (Fig 4c). Since ACh-induced vasodilation is primarily dependent on endothelial NO production, selective loss of ACh-evoked vasodilation demonstrates successful endothelium denudation. After validating endothelium denudation, we recorded ne amapimod-evoked vasorelaxation in both endothelium-intact and -denuded mesenteric arteries. Our myography data on endothelium-intact mesenteric arteries reveal that application of 10µM ne amapimod dilated mesenteric arteries by ~43 µm (Fig. 4d, 4e). On the other hand, ne amapimod application on endothelium-denuded mesenteric arteries produced a dilation of ~38 µm, which is ~89% or similar, compared to endothelium-intact mesenteric arteries (Fig 4d, 4e). This data suggests that endothelium denudation does not affect ne amapimod-evoked vasodilation.
Consistently, pharmacological inhibition of endothelial nitric oxide synthase (eNOS) with L-NNA and the production of the most important physiological vasodilator molecule NO do not reduce ne amapimodevoked vasodilation (Fig. 5a, 5b), precluding the role of NO in this process. Likewise, ne amapimodevoked vasodilation remained unchanged by the application of indomethacin which inhibits cyclooxygenase (COX) and synthesis of PGI 2 , another important endothelium-derived vasodilator (Fig. 5a,   5b). These data suggest that endothelial signaling has no role in acute ne amapimod-evoked vasodilation.
Ne amapimod-induced vasodilation is not mediated by smooth muscle cell K + channels Opening of arterial smooth muscle cell K + channels and K + e ux induce smooth muscle cell hyperpolarization, leading to vasodilation. Arterial smooth muscle cells express a variety of K + channels including voltage-gated K + channels (K V ), large-conductance Ca 2+ -activated K + channels (BK Ca ), and ATPsensitive K + channels (K ATP ). We therefore assessed the role of the major K + channels in ne amapimodinduced mesenteric artery vasodilation. Application of 4-aminopyridine (4AP), an inhibitor of K V channels, produced a mean dilation of ~37 µm, which is 90% of that produced by ne amapimod (Fig. 6a, 6b). Similarly, application paxillin, a BK Ca channel blocker, and glibenclamide, a blocker of K ATP channels, both relaxed mesenteric arteries by ~35 µm (Fig. 6a, 6b). Application of TEA, a general K + channel blocker, dilated mesenteric arteries by ~36 µm, which is ~88% of that produced by ne amapimod (Fig. 6a,   6b). Overall, these data suggest that smooth muscle cell K + channels do not mediate ne amapimodinduced vasodilation in resistance mesenteric arteries.

Discussion
In this study, we demonstrate for the rst time that acute ne amapimod application stimulates vasodilation in resistance-size mesenteric arteries. Our data suggests that ne amapimod reduces basal phosphorylation of p38 MAPKα and its downstream target Hsp27 to induce vasorelaxation. Our data also indicates that ne amapimod-evoked vasodilation is not dependent on endothelial signaling or smooth muscle cell K + channels.
While the vast majority of previous studies focused on the role of p38 MAPKα in in ammation, its role in regulating arterial contractility remained virtually unexplored. Clues about the involvement of p38 MAPK(α/β) activity in arterial contractility surfaced from earlier reports of angiotensin II-and norepinephrine-induced phosphorylation of p38 MAPK and Hsp27, leading to vasoconstriction [7,23,24] .
Hsp27 is an actin-associated protein that regulates actin cytoskeleton reorganization, which is a critical step in smooth muscle contraction. Vasoconstrictors such as angiotensin II, norepinephrine and endothelin-1 enhance Hsp27 phosphorylation and actin lament formation via p38 MAPK activation. Inhibition of p38 MAPK or disruption of cytoskeleton blocks or reverses these vasoconstrictor responses [29][30][31] .
Our nding is in agreement with previous studies in that selective p38 MAPKα inhibition, like non-selective p38 MAPK inhibition, reduces the phosphorylation of Hsp27. Reduced phosphorylation of Hsp27 decreases its actin capping ability, leading to smooth muscle relaxation. Komers et al. demonstrated that non-selective inhibition of α and β isoforms of p38 MAPK by SB203580 relaxed renal arteries and enhanced ACh-induced vasodilation [4] . However, whether selective inhibition of the α isoform of p38 MAPK could relax resistance arteries of the systemic circulation remained unknown. Our study provides pharmacological and biochemical evidence that selective inhibition of p38 MAPKα and downstream Hsp27 phosphorylation underlie ne amapimod-induced vasodilation in mesenteric arteries. According to the Poiseuillie's equation (Pugsley and Tabrizchi, 2000), blood ow through a vessel is related to diameter(d)/radius (r), pressure gradient (δP), blood viscosity( ) and vessel length (l) as follows: Ne amapimod at 10µM concentration induced ~17% relaxation of mesenteric artery diameter from baseline. Therefore, such vasodilation upon acute ne amapimod application may lead to a signi cant drop in arteriolar resistance and systemic blood pressure in vivo. Mesenteric arteries are part of the splanchnic circulation that receive approximately 25% of cardiac output. The splanchnic circulation regulates the distribution of cardiac output and acts as a blood reservoir. Thus, vasodilation in small mesenteric arteries is likely to lower peripheral vascular resistance and systemic blood pressure. However, future studies will be required to investigate these possibilities.
Earlier studies suggest that non-selective inhibition of p38 MAPK improves endothelial dysfunction, enhances NO bioavailability and improves ACh-induced vasodilation, presumably by reducing vascular in ammation. Based on the ndings in our study and previous reports, it is possible that ne amapimod treatment may improve long-term blood pressure control by reducing both vascular resistance and vascular in ammation.
Due to the involvement of p38 MAPKα in in ammation, there is great enthusiasm for developing selective p38 MAPKα inhibitors for the treatment of in ammatory diseases. As a selective inhibitor of p38 MAPKα, ne amapimod is currently being investigated for its e cacy in reversing neuroin ammation, neurodegeneration and cognitive de cits in DLB and AD. In clinical trials, ne amapimod treatment appears to be promising in rescuing cognitive de cits in DLB and likely AD. Hypertension, through a variety of mechanisms, may lead to cognitive dysfunction in these patients. Proposed mechanisms by which hypertension leads to dementia include microvascular damage and dysfunction that results in white matter disease, microinfarcts, microhemorrhages, poor clearance of amyloid-β peptide among others [14,16,17] . In addition, elevated systemic blood pressure negatively impacts cerebral perfusion, which alters brain structure and function by promoting ischemic injury, oxidative stress and endothelial dysfunction [14] . Although hypertension and increased vascular tone are known drivers of cognitive dysfunction [14][15][16][17] , it remains unknown if ne amapimod treatment-associated improvement in cognitive function in DLB and AD could be due to a reduction in peripheral vascular resistance as a result of direct vasorelaxation of the systemic resistance vessels such as the resistance-size mesenteric arteries. Our nding of a direct vasodilatory effect of ne amapimod suggests that the observed clinical bene t with ne amapimod pharmacotherapy may result from direct vasodilation and reduction of vascular resistance, as well as the reduction of in ammation of the cardiovascular and nervous systems as proposed by others earlier.
This study laid the foundation for future studies that acute inhibition of p38 MAPKα could induce vasodilation, which in turn, could enhance regional organ blood ow. Our future study will examine if ne amapimod directly relaxes resistance cerebral arteries that control brain perfusion and cognitive function.
In conclusion, our study demonstrates that acute ne amapimod application relaxes resistance mesenteric arteries via selective inhibition of smooth muscle cell p38 MAPKα and the phosphorylation of its downstream target Hsp27. Inhibition of Hsp27 phosphorylation decreases its actin capping ability, resulting in smooth muscle relaxation and vasodilation. Ne amapimod-evoked vasodilation appears to independent of endothelial signaling and smooth muscle cell K + channels. As elevated blood pressure and altered vascular tone contribute to cognitive dysfunction, vasodilatory effect of this drug is likely to have positive impact on the cognitive performance. We propose that the vasodilatory action of ne amapimod may be clinically translatable and is likely to improve systemic blood pressure and cognitive de cits in AD and DLB patients for which ne amapimod is currently being investigated.

Animals
Animal protocols were reviewed and approved (Ref. No. A2107011) by the Mercer University Institutional Animal Care and Use Committee (IACUC), and animal experiments were performed according to the guidelines set by the United States National Institutes of Health (NIH). We used male Sprague Dawley (SD) rats between 7 and 10 weeks of age. Rats were purchased from Charles River Laboratories (Wilmington, MA, USA) and acclimatized for one week at the Mercer University College of Pharmacy Vivarium prior to experimentation. Rats were individually caged in a temperature-regulated room (temperature 22 ± 2°C; 55% humidity; 12-hour light/dark cycles) and had access to standard chow diet and drinking water ad libitum. Animals were randomized for all experments [32] .
Tissue preparation CO 2 inhalation was used as per the '2020 AVMA Guidelines for the Euthanasia of Animals' to euthanize rats followed by decapitation. Entire mesenteric artery bed was dissected out and placed into ice-cold PSS containing (in mM) KCl 6.0, NaCl 112, NaHCO 3 1.18, MgSO 4 1.18, KH 2 PO 4 1.18, CaCl 2 1.8, and glucose 10. Third and fourth order branches of mesenteric arteries (150-250µm) were dissected cleaned of adventitial tissue in ice-cold PSS. Cleaned arterial branches were cut into 1-2 mm long segments that were kept in ice-cold PSS until cannulated for pressure myography [25,26,28] . PSS for vessel dissection, isolation and preparation contained (in mM) KCl 6.0, NaCl 112, NaHCO 3 1.18, MgSO 4 1.18, KH 2 PO 4 1.18, CaCl 2 1.8, and glucose 10. pH of the PSS was adjusted to and maintained at 7.4 by gassing with a mixture of 21% O 2 , 5% CO 2 , and 74% N 2 . 60 mM K + -PSS (60K) was prepared by equimolar replacement of NaCl with KCl in the PSS. [32] Pressure myography Arterial segments were cannulated in a temperature-controlled perfusion chamber (Living Systems Instrumentation, St. Albans, VT), continuously perfused with 37°C PSS, and gassed with mixture of 21% O 2 , 5% CO 2 , and 74% N 2 to maintain pH 7.4. 60 mM K + -PSS (60K) was prepared by equimolar replacement of NaCl with KCl in the PSS and used for checking the viability of the cannulated arteries at 10 mmHg. Using a pressure servo controller with peristaltic pump (PS-200, Living Systems) intraluminal pressure was gradually increased to 80 mmHg with increments of 20 mmHg at each step. Mounted arteries were maintained for at least 10 minutes at each pressure step so that the arteries reach a stable diameter. Mesenteric arterial segments developed a stable myogenic tone at 80mmHg, after which a cumulative concentration response to ne amapimod (0.0001-10 µM) was performed [32] . Vessel diameter was read at 1 Hz using a charge-coupled device (CCD) camera connected to a Nikon Ts2 microscope and the automatic edge-detection function of IonWizard software (IonOptix, Milton, MA, USA). Myogenic tone (%) was calculated as: 100 × (1 -D active /D passive ), where D active is active arterial diameter in the presence of Ca 2+ and D passive is the arterial diameter determined in Ca 2+ -free PSS supplemented with 5 mM EGTA [25][26][27][28] . Endothelium denudation was performed by slowly passing air bubbles through the vessel lumen for 1-2 minutes. Arteries exhibiting at least 80% reduction of ACh-induced vasodilation were considered endothelium denuded [25][26][27] . Endothelium denudation selectively abolishes ACh-induced vasodilation without altering artery responses to sodium nitroprusside (SNP) and 60K [32] .
Arterial lysates were centrifuged at 12,000 rpm for 10 minutes at 4ºC and supernatants collected in clean eppendorf tubes. After measuring protein concentrations using BCA method, 40µg protein for each sample was boiled with 2X SDS-sample buffer (Bio-Rad) for 5 minutes and resolved by SDS-PAGE.

Statistical analysis
OriginLab software version 9.55 (2018b, URL: https://www.originlab.com/index.aspx? go=SUPPORT&pid=3325) was used for statistical analysis. Values were expressed as mean± standard error of mean (SEM). One-way analysis of variance (ANOVA) along with Newman-Keuls post-hoc test was used for multiple group comparison. p < 0.05 was considered statistically signi cant [26,34] .   Role of endothelial NO and PGI2 in ne amapimod-induced vasodilation. (a) Original traces illustrating vasodilatory responses produced by ne amapimod (ne a, 10µM) alone, and together with L-NNA (10µM) or indomethacin (indo, 1mM) in endothelium-intact mesenteric arteries. (b) Bar graph showing that the application of L-NNA and indo did not reduce ne amapimod-induced vasorelaxation in mesenteric arteries. Figure 6