Stimulation of angiotensin II type I receptors (AT1R) elicits vasoconstriction (VC) that may be occurring through the activation of a pathogenic vascular pathway such as Rho kinase (ROCK). We hypothesize that reflex cutaneous VC to whole body cooling (mean skin temperature = 30.5°C) in older humans relies in part on AT1R activation, which may explain greater ROCK activity attendant with aging. Two microdialysis (MD) fibers were placed in the forearm skin of 10 young (Y; 24 ± 1 yr) and 10 older (O; 70 ± 2 yr) individuals for infusion of 1) lactated Ringer's solution (switched to fasudil, a ROCK antagonist, after cooling); and 2) AT1R blockade with losartan. Laser Doppler flux (LDF) was measured over each MD site and cutaneous vascular conductance (CVC) was calculated (CVC = LDF/mean arterial pressure) and expressed as percent change from baseline (%ΔCVCBASELINE). In older individuals the VC response to whole body cooling was blunted (Y = −34 ± 2, O = −17 ± 3%ΔCVC) and was further attenuated at the losartan site (Y = −34 ± 3, O = −9 ± 3%ΔCVC; P < 0.05). The VC response to an exogenous 10-μM dose of angiotensin II (Y = −27 ± 3, O = −42 ± 5%ΔCVC) was completely blocked in sites pretreated with losartan or with fasudil. These data suggest that AT1R activation contributes to the reflex VC response in aged but not young skin. Furthermore, the angiotensin II component of the VC response appears to occur primarily through a ROCK-mediated mechanism.
- skin blood flow
- temperature regulation
- Rho kinase
reflex cutaneous vasoconstriction (VC) is an immediate and sustained response to whole body cold exposure that minimizes convective heat loss to the environment. This response is mediated by multiple parallel signaling pathways that ultimately increase intracellular calcium in vascular smooth muscle. Loss of redundancy and impaired function in older men and women result in a blunted reflex VC response to cold exposure and increased susceptibility to excess heat loss and hypothermia even in mildly cool ambient conditions (11, 19, 42).
The RhoA/Rho-kinase (ROCK) pathway is a proconstrictor vascular mechanism that may remain functionally intact or potentially upregulated in aged skin (40). It was demonstrated that ∼50% of the reflex cutaneous VC response in aged skin was mediated by ROCK (23). Despite activation of this pathway, the VC response in older adults remains blunted. Although upregulated ROCK may strengthen the effectiveness of cold-induced VC from a thermoregulatory standpoint, this may be occurring at the expense of microvascular function. Increased ROCK activity precedes several vascular pathologies including atherosclerosis, stroke, diabetes, endothelial dysfunction, and hypertension (5, 12, 24, 29–32, 34, 43). Thus increased ROCK activation may confer some thermoregulatory compensation benefit in a cold environment, but this may also indicate a shift toward preclinical signaling changes in the vasculature.
Although the source for ROCK activation is unclear, animal and in vitro studies have identified that ROCK activation increases in the presence of angiotensin II (ANG II) (6, 16, 31). The classic view is that ANG II is produced from an endocrine pathway involving multiple organ systems that originates with renin secretion from the kidney. However, it is increasingly evident that human skin alone expresses a complete renin angiotensin system (RAS) for the production and activation of ANG II (35). Thus ANG II can be synthesized locally and have paracrine effects independently of systemic RAS and may have a more central role in cutaneous vasomotor function. This is supported by evidence indicating that elevated ANG II is in part responsible for blunted nitric oxide (NO)-mediated cutaneous vasodilation (VD) in low-flow postural tachycardia syndrome (POTS) patients (37). However, the importance of ANG II in the VC response or as a trigger for increased ROCK activation in the cutaneous microvasculature has yet to be elucidated.
Increased ANG II signaling and activation of downstream targets such as ROCK may provide a causative link between primary aging and microvascular dysfunction. In light of that, the purpose of this study was to determine the functional role of ANG II in the reflex cutaneous VC response in young and aged skin. We hypothesize that intradermal administration of an angiotensin type I receptor (AT1R) antagonist, losartan, will have a greater effect in reducing the reflex VC response to whole body cooling [mean skin temperature (Tsk) = 30.5°C] in aged skin. We further hypothesize that ANG II may be eliciting cutaneous VC in part through downstream activation of ROCK.
All experimental procedures were approved by the Des Moines University Institutional Review Board and conformed to the standards set by the Declaration of Helsinki. After providing voluntary verbal and written informed consent, 10 young (24 ± 1 yr; 4 men, 6 women) and 10 older (70 ± 2 yr; 3 men, 7 women) subjects participated in the study (Table 1). All subjects were normotensive, not obese, nondiabetic, normal lipid profile, nonsmokers, and not taking any prescription medications that would otherwise alter thermoregulatory or vascular function (e.g., statin or blood pressure lowering drug). Young women were normally menstruating and tested during the low hormone phase (days 2–7) of the menstrual cycle or during the placebo week of oral contraceptives (n = 3) (7). All older women were postmenopausal and not taking hormone replacement therapy. On the day of the experimental visit, subjects reported to the laboratory after abstention from alcohol and caffeine (>12 h) as well as strenuous physical activity or vitamin supplement (>24 h).
Subjects arrived to the laboratory (room temperature = 23°C) in the morning (0800–1300) and were positioned comfortably in a semirecumbent position for instrumentation. Two microdialysis (MD) fibers (10 mm, 30-kDa molecular mass cutoff membrane, MD 2000; Bioanalytical Systems, West Lafayette, IN) were aseptically placed in the dermis of the ventral forearm. Before placement, the entry and exit site for each fiber were marked ∼2.5 cm apart in an area of skin free of larger superficial veins. Ice was applied to the forearm for 5 min to temporarily anesthetize the skin (17). Immediately thereafter, two 25-gauge needles were inserted, one for each fiber site, in approximation to the marked areas of skin. Then, a MD fiber was threaded through the beveled end of each of the needles and then withdrawn leaving the membrane component (10 mm) of the MD fiber submerged in the dermis. After which, MD fibers were secured with tape and then perfused for 60–90 min at 2.0 μl/min (Baby Bee Syringe Drive and Hive Controlled; Bioanalytical Systems) to allow resolution of localized hyperemia from needle insertion trauma.
To control skin temperature, subjects donned a water-perfusion suit that covered the entire body with the exception of the face, feet, hands, and instrumented forearm. The unweighted Tsk was collected from six copper-constantan thermocouples taped on the skin of the thigh, calf, abdomen, chest, back, and arm. Local skin temperature was controlled by placing a heater (moorVMS-HEAT; Moor Instruments, Axminster, UK) directly over each MD fiber site. The local temperature was set at 33°C for the duration of the experimental protocol. This was important to ensure that any vasomotor change detected is reflex in origin and not confounded by locally derived stimuli.
Red blood cell flux [laser Doppler flux (LDF)], an index for skin blood flow, was continuously measured by multifiber integrated laser Doppler probes (moorVMS-LDF; Moor Instruments) fitted into the local heaters such that the probe was positioned perpendicular to the skin surface. Arterial blood pressure was measured every 5 min on the contralateral arm using an automated cuff (Suntech Tango M2) that was verified with brachial auscultation. The mean arterial pressure (MAP) was calculated as the diastolic blood pressure plus one-third the pulse pressure. Heart rate was monitored using a lead II electrocardiogram (CT-1000 cardiotachometer; CWE). Cutaneous vascular conductance (CVC) was calculated as the ratio of LDF to MAP and expressed as percent change from baseline (%ΔCVCBASELINE).
After instrumentation, MD fiber sites were randomly assigned for continuous perfusion of either 1) lactated Ringer's solution serving as the control site, or 2) 2 μg/l losartan, an angiotensin II receptor type I (AT1R) antagonist. All drugs were weighed, dissolved in lactated Ringer's solution, and filtered with syringe microfilters (Acrodisc; Pall, Ann Arbor, MI) immediately before use. Drug dosages were based from previous studies and verified with pilot experiments (23, 38). The optimal ANG II dose was determined as the dose that maximally affected the VC response, and the losartan dose ensured complete blockade of ANG II-mediated VC without altering baseline CVC. After 60 min of drug perfusion, baseline values were collected while Tsk was maintained at 34°C.
Following baseline, cold water was circulated through the suit to induce a reflex cutaneous VC response. Skin temperature was gradually reduced from 34 to 30.5°C over a 30-min period and maintained at 30.5°C for an additional 10 min. The peak cooling stimulus was not severe enough to elicit a shivering response, and this was verified by asking participants to describe their thermal discomfort or propensity to shiver. Rewarming for ∼30 min followed to return Tsk to 34°C. Then, 10 μM angiotensin II were perfused through both MD fibers for 10 min to verify the blockade established by losartan. After a 20-min washout period, MD fibers were reassigned such that the losartan site was switched to perfuse lactated Ringer's solution and the control site was switched to receive 3 mM fasudil to inhibit ROCK. Following 60–90 min of drug perfusion, the 10 min ANG II perfusion was repeated. After a 20-min washout period, vascular responsivity was tested by briefly (∼3 min) increasing the local heaters to 42°C at each site until an increase in LDF was observed.
Data acquisition and analysis.
sk during cooling, and during ANG II perfusion. Normal distribution was evaluated using Shapiro-Wilk tests. A three-way mixed model repeated measures ANOVA followed by Bonferroni post hoc tests was conducted (SPSS version 22; SPSS, Chicago, IL) to detect age and treatment effects during both whole body cooling and ANG II administration. Physical characteristics between age groups were analyzed using Student's unpaired t-tests. All tests were two-sided and statistical significance was set at α = 0.05. An a priori sample size analysis was performed, which indicated that 10 participants were needed to detect significant differences at P < 0.05 and 80% power. Values are represented as means ± SE.
The physical characteristics of the age groups are presented in Table 1. Groups were well matched with respect to anthropometric parameters, resting MAP, and lipid profile. Although total and LDL cholesterol was higher in older adults (P < 0.05), there was no difference with respect to HDL cholesterol or cholesterol ratio.
The absolute CVC values (calculated as LDF/mmHg) for each MD fiber site is represented in Table 2. At the losartan site, CVC values during ANG II perfusion differed from the control site in older adults (P < 0.05). After MD fiber sites were reassigned so that fasudil was perfusing skin, CVC was elevated both before (i.e., baseline) and during ANG II perfusion (P < 0.05). Moreover, at the control site, CVC during baseline and ANG II perfusion was lower in aged as opposed to the control site in young skin (P < 0.05).
In Fig. 1, the VC response represented as the reduction in CVC from baseline was illustrated at every 0.5°C drop in Tsk during whole body cooling. In younger subjects (Fig. 1A), the VC response was unaffected in MD sites treated with losartan. In contrast, older adults (Fig. 1B) exhibited a blunted response to cooling in the losartan site (Tsk ≤ 33.5°C; P < 0.05). Localized administration of losartan further attenuated VC in older skin (Tsk ≤ 32.5°C; P < 0.05) but had no effect in young skin.
The peak VC response to the localized 10-μM dose of angiotensin II is illustrated in Fig. 2. Older adults exhibited a more pronounced VC response to ANG II (Y = −27 ± 3, O = −42 ± 5%ΔCVCBASELINE; P < 0.05). The ANG II-mediated VC was completely blocked in both age groups at the losartan as well as the fasudil-treated sites.
The primary novel finding from this study was that endogenous ANG II contributes to the reflex VC response in aged but not young skin. During gradual whole body cooling (Tsk = 30.5°C), AT1R inhibition with losartan reduced the VC response by ∼50% in aged skin whereas younger skin remained unchanged. Additionally, the VC response to localized exogenous ANG II perfusion not only was more sensitive in older skin but was also completely blocked in sites pretreated with fasudil, a ROCK inhibitor. Collectively, these data indicate that ANG II importantly contributes to the reflex cutaneous VC response in older individuals and this appears to operate predominantly through a ROCK-mediated mechanism.
ANG II is produced from a multiorgan endocrine cascade or locally released within tissues to elicit endocrine and paracrine/autocrine effects, respectively. Specifically, a localized angiotensin-converting enzyme system has been identified in skin and may serve as an important source for ANG II biosynthesis (35). The potent VC effect of ANG II is mediated through AT1R binding. These receptors have been identified in skin and may be located presynaptically on postganglionic sympathetic nerves as well as on the vessel (10, 35). From a functional standpoint, there is evidence that ANG II blunts cutaneous VD in POTS patients (37). The findings from the present study extend the functional role of ANG II in skin by revealing a contributory role in the reflex VC response in older adults.
Aging is associated with an increase in AT1R density (13, 33). For a given infused dose of ANG II, older individuals exhibited a greater reduction in conduit vessel blood flow (3, 44). Moreover, this age-associated increase in AT1R sensitivity was eliminated after concomitantly infusing phentolamine (3). This supports a presynaptic role of AT1R by potentiating norepinephrine (NE). This presynaptic component accounted for ∼20% of the exogenous ANG II-mediated VC in older adults. In the cutaneous microvasculature, we also observed a marked increase in the VC response to an exogenous dose of ANG II in older adults, thereby indicating an increase in cutaneous AT1R sensitivity with age. Although ANG II is not necessarily compensating for the blunted VC response in older adults, it is plausible that ANG II may be functioning to support a compromised noradrenergic pathway.
AT1R activates downstream intracellular targets such as ROCK and NADPH oxidase (NOX) that may elicit VC (6, 15, 16, 28, 31). Specifically, ROCK activation promotes VC by 1) inhibiting myosin light chain phosphatase; 2) increasing α2C-translocation from the Golgi apparatus to the plasma membrane; and 3) reciprocally inhibiting NO synthase (1, 9, 18, 25, 27, 39). In aged skin, whole body and local cooling responses are increasingly dependent on ROCK to elicit VC (23, 40). In the present study, ∼50% of the reflex cutaneous VC response was attenuated after AT1R inhibition. This was nearly identical to the magnitude of attenuation achieved with ROCK inhibition during a similar whole body cooling protocol in healthy older adults (23). Additionally, the VC response to an exogenous dose of ANG II was completely blocked in skin sites pretreated with fasudil. Collectively, this supports a role for ANG II as a source for ROCK activation and upregulation in aged skin.
Various mediators may serve as a source for ROCK activation in skin. During localized skin cooling, it has been demonstrated that superoxide anion can directly increase ROCK activity (2), and the VC response is attenuated with intradermal antioxidant supplementation using ascorbate (45). Both xanthine oxidase and NOX, a downstream target of ANG II, serve as significant sources of reactive oxygen species (ROS) in skin (26). Thus increased ROCK activity may be occurring directly from AT1R activation or indirectly from ROS generated from AT1R-mediated stimulation of NOX (6, 15, 16, 28, 31). Considering that increased oxidative stress has been implicated as a causative factor in the age-associated reduction in cutaneous function (21, 22, 41), it is plausible that increased reliance on AT1R and its downstream activation of ROCK may underlie multiple age-related impairments in cutaneous microvascular function.
It is well established that the reflex cutaneous VC during whole body cold exposure is blunted in aged skin (19, 21, 42). This is consistent with the findings from the present study. In young skin, the reflex VC response relies entirely on sympathetic adrenergic nerve fibers; however, only ∼60% of the VC response in mediated by NE (36). The remaining ∼40% is dependent on cotransmitters released from the sympathetic nerve terminal. However, in aged skin 1) the cotransmitter component of VC is functionally absent (21, 42); 2) noradrenergic biosynthesis and receptor sensitivity is attenuated (22, 41); and 3) the extant VC response is more reliant on ROCK (23). Many of these impairments may be explained in part by the influence of elevated oxidative stress attendant with aging. The reason that ROCK remains either intact or upregulated in aged skin may be due to the fact that it is activated by ROS and the AT1R pathway.
Reflex VC during whole body cold exposure is an important effector mechanism that functions to minimize heat loss and maintain body core temperature. From a thermoregulatory standpoint, increased VC from ROCK activation may be beneficial in mitigating excess heat loss during cooling. However, this may also signify a more pathogenic shift in microvascular function, one that results in compromised vasodilatory function and increased generation of ROS. Increased ROCK activity precedes the pathogenesis of several cardiovascular diseases (5, 12, 24, 29–31, 34). Increased AT1R sensitivity may play an important role in this shift in function. This has been elegantly demonstrated in a study where the compromised VD to local skin warming in POTS patients was partially corrected with AT1R blockade (37). Furthermore, the current study has identified a contribution of AT1R to the reflex VC in aged skin that is mediated by ROCK; however, in young adults the AT1R pathway does not appear to be functionally relevant in this response. Both AT1R and ROCK may serve as preclinical markers indicating more globalized microvascular dysfunction.
There is evidence indicating that the inhibition of AT1R may serve as an effective strategy to improve vascular function. Homozygous mice deficient for AT1R have less oxidative stress, develop fewer atherosclerotic lesions, and have less cardiac injury during aging (4). Oral supplementation of losartan elicits VD and a reduction in systemic vascular resistance in heart failure patients (14, 20). Losartan also increased NO-mediated VC in conduit vessels of type II diabetics (8). In skin, losartan improved the cutaneous VD response in POTS patients (37) and explained a significant component of the VC response to whole body cooling in older adults.
The ROCK antagonist used in the study, fasudil, elicits a considerable baseline VD response. Thus it is possible that the drug alone is stimulating other signaling pathways that may obfuscate our interpretation of the findings. Specifically, enhanced baseline dilation may blunt or mask the effect of VC. However, this is not evident from other studies that have used fasudil (23, 34, 40). Additionally, the inhibition of ANG II-mediated VC at the fasudil site was evident when normalizing as a percent change from baseline or as a change in absolute CVC. More selective inhibitors could not be used as they are currently unavailable for in vivo use in humans.
It is possible that ANG II may be binding to the AT2R. AT2R mediates VD and tends to counteract the effects of AT1R. Although AT2R is found within the cutaneous vasculature, it does not appear to be functionally relevant at physiologic concentrations of ANG II (35, 38).
In summary, the reflex cutaneous VC response to whole body cooling was attenuated with AT1R antagonism in aged skin. The effect was not observed in young skin. Furthermore, intradermal administration of ANG II resulted in a more robust VC in aged skin, which was completely blocked after ROCK inhibition. Collectively, this suggests that older adults not only have increased reliance on ANG II to elicit reflex VC but are also more sensitive to ANG II. This ANG II-mediated VC appears to rely on ROCK activation. Thus ANG II may be serving as a primary source of ROCK activation in aged skin and contributing to a more pathogenic shift in microvascular function.
This work was supported with resources from Des Moines University; Iowa Osteopathic and Education Research Grant IOER #05-13-05.
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: J.A.L. conception and design of research; J.A.L. and K.E.K. performed experiments; J.A.L. and K.E.K. analyzed data; J.A.L. and K.E.K. interpreted results of experiments; J.A.L. prepared figures; J.A.L. and K.E.K. drafted manuscript; J.A.L. and K.E.K. edited and revised manuscript; J.A.L. approved final version of manuscript.
We gratefully acknowledge the subjects who participated in the study. We specially thank Dr. Terri Plundo for medical assistance as well as Kevin Smaller and Bindiya Shah for assistance with data collection.
- Copyright © 2015 the American Physiological Society