Advanced glycated end‐products inhibit dilation through constitutive endothelial RAGE and Nox1/4 in rat isolated skeletal muscle arteries

This study investigated the actions of advanced glycated end‐products (AGE), their receptors (RAGE), and NAD(P)H oxidase (Nox) subtypes 1, 2, and 4 on mechanisms of endothelium‐dependent dilation of the rat cremaster muscle artery (CMA).


| INTRODUC TI ON
Advanced glycated end-products (AGE) are formed by nonenzymatic reactions between glucose or glucose metabolites and long-lived macromolecules, particularly protein.AGE can cause cellular damage through their activation of signaling pathways involving inflammation, oxidative stress, and insulin resistance. 1,2nditions characterized by hyperglycemia and hyperlipidemia increase the formation of AGE, and as the vascular system is highly exposed, a role for AGE is implicated in many of the deleterious vascular effects of diabetes and obesity, including atherosclerosis and microvascular disease. 3,4Effects of AGE are commonly, but not exclusively, mediated through receptors for AGE (RAGE), pattern-recognition receptors whose expression is increased in AGE-rich or hypoxic environments.[7] A key early impact of hyperglycemia and diabetes is vascular endothelial dysfunction, where AGE can inhibit endothelium-dependent dilation of arteries in vivo and in vitro, [8][9][10][11] through activation of RAGE and subsequent ROS accumulation in vascular endothelial cells. 4,12ROS interact with nitric oxide (NO), a key mediator of endothelium-dependent vasodilation, forming peroxynitrite (ONOO − ) which depletes biologically active NO.Peroxynitrite also uncouples the active NOS dimer, preventing NO formation (and exacerbating ROS production) through depletion of the NOS co-factor BH 4. 4,8,13,14   While the effects of AGE and RAGE on NO-mediated vasodilation have been investigated, their effects on endothelium-derived hyperpolarization (EDH) of vascular smooth muscle are largely unknown.6][17] Previous studies have investigated the effects of AGE on component parts of EDH of smooth muscle.AGE inhibited Ca 2+ signaling in vascular endothelial cells, 18,19 vital for activation of Ca 2+ -sensitive K + -channels (K Ca ) which are the key mediators of EDH. 20,21Activation of RAGE inhibited voltage-sensitive K + -channels and large-conductance K Ca (BK Ca ) in smooth muscle cells isolated from rat cerebral arteries. 22But very few studies have investigated the effect of AGE and/or RAGE activation on functioning EDH of smooth muscle in an intact artery.Furthermore, the source of RAGE-induced ROS in the vascular endothelium is unclear.RAGE are thought to stimulate ROS-producing NAD(P)H oxidase (Nox) activity through protein kinase C-mediated phosphorylation. 4,12,23There are five known Nox subtypes (Nox1-5), 24,25 and those responsible for RAGE-linked endothelial dysfunction in intact arteries have not been investigated extensively.RAGE were coupled to increased Nox2 activity in coronary arteries, 14 and this pathway was linked to renovascular damage in rodent models of diabetes. 26,27her studies have linked increased RAGE expression and renal vascular dysfunction to increased Nox4 activity. 4,28e hypothesis of our study was that RAGE activation by AGE would inhibit EDH of smooth muscle through activation of NAD(P) H oxidase.The study investigated the effects of AGE (glycated albumin) and RAGE activation on endothelium-dependent dilation of the rat isolated cremaster muscle artery (CMA), with a particular focus on EDH of the vascular smooth muscle.The study also explored the subtypes(s) of Nox involved in mediating these effects using recently developed, subtype-selective inhibitors, setanaxib (Nox1/4), and GSK2795039 (Nox2).This particular artery (CMA) was studied as it is known to express RAGE. 29,30Furthermore, the mechanism of endothelium-dependent vasodilation in this artery is well-characterized and includes a substantial EDH of smooth muscle component, mediated by small and intermediate-conductance (S and I) K Ca associated with myo-endothelial gap junctions, and activation of smooth muscle BK Ca via endothelium-derived epoxy-eicosatrienoic acids (EETs). 31,32Therefore, the rat CMA was highly suitable for the proposed investigations.

| Animals and blood vessel dissection
Male Sprague-Dawley rats (~6 weeks old/180 g; Animal Resources Centre, Perth, WA) were used for the study.All animal care and research protocols were approved by the UNSW Animal Care and Ethics Committee (approved projects #06/29B, #09/43B, and #18/86B).Animals were housed within the Biological Resources Centre at UNSW, in standard housing conditions with a temperature-controlled environment at 21°C under a 12 h light/dark cycle with food (standard rat chow) and water ad libitum.Animal studies are reported in compliance with the ARRIVE guidelines. 33ts were anesthetized with sodium thiopentone (100 mg kg −1 , intraperitoneally; Abbott, Kurnell, NSW).The aorta, brain, segments of mesentery, and cremaster muscles were removed and transferred to a cold (4°C) dissecting dish containing dissecting buffer (in mM: Bovogen Biologicals, Essendon, Vic).The aorta was cleaned of extraneous tissue and segments of middle cerebral, third-order mesenteric and first-order cremaster muscle arteries extracted from their respective tissues for subsequent studies.A total of 55 rats were used in the study.
Antibody controls involved the use of positive and negative expressing tissue, peptide block and western blotting, per previous studies. 32,34,35

| In vitro protein glycation
In order to produce AGE for experimental use, BSA (10 mg/mL) was incubated with either glucose-6-phosphate (G6P, 250 mM for 60 days) or methylglyoxal (MGO, 9 mM for 5 days) in PBS at 37°C, in the dark.Control samples were generated by incubating BSA alone in PBS under the same conditions.The pH of G6P incubations was checked every 3 weeks and did not go >7.7.To terminate the reaction by removing unreacted sugar, all aliquots (including control) were dialyzed extensively (4 times over 6 h) in the dark using PBS at 4°C.Then, aliquots were sterile-filtered through a Millipore Express 0.22 μm filter prior to storage at −80°C.The final protein concentration was determined using Bradford assay.Generation of AGE was assayed in the sugar co-incubated BSA samples through measuring sample fluorescence and absorbance as previously described 19,36 (Figure S1).BSA samples (100 μL; 1 mg/mL) were placed in a 96-well microtiter plate, and fluorescence was measured at 370 nm excitation/440 nm emission (FLUOstar Optima, BMG Labtech, CA, USA).

| ROS assay
The dissected arteries were cut into segments and incubated, together with the glycated BSA (AGE) and nonglycated BSA (control) with both intra-and extraluminal application in a 37°C heating block for 40 min.The chemiluminescent probe L-012 (100 μM) was then added to the tubes for measurement of O − 2 .L-012 addition was carried out in the dark due to the photosensitive nature of the compound.Chemiluminescence was measured immediately after L-012 addition, using a luminescence reader (GloMax® 20/20 Single Tube Luminometer, Promega, WI, USA).A preaddition reading (without artery segments) was performed for background measurements.
H 2 O 2 measurement was carried out using an Amplex Red Assay.
After incubating the arteries with either glycated or nonglycated BSA, both intra-and extraluminally, in a 37°C heating block for 40 min, the arteries were carefully removed and placed in a 96-well plate containing the supplied reaction mixture (15 μM Amplex® Red reagent, 0.1 U/mL Horseradish Peroxidase (HRP) in Krebs buffer solution; see below for Krebs composition) that had been preheated for 10 min at 37°C.The plate was then incubated for an additional 10 min.Well fluorescence was measured using a fluorescence reader (FLUOstar Optima, BMG Labtech) at ex540 nm/em570 nm.A H 2 O 2 standard curve (0-0.125 μM) was also performed with each assay to quantify the H 2 O 2 produced.All H 2 O 2 assay material and protocol was handled and carried out in the dark due to the photosensitive nature of the compounds.
All luminescence and fluorescence counts were corrected for artery weight and in the case of the H 2 O 2 assay, the amount of H 2 O 2 in the sample was obtained via the derived H 2 O 2 standard curve.
Aortic segments were dried overnight by incubating at 37°C after the day's experiment and weighed the next day.Arteries other than the aorta were very small and therefore pooled together over the span of all experiments and then dried overnight on pieces of aluminum foil in a 37°C incubator.Dried tissues were weighed the next day, and the average weight of an individual segment was estimated by dividing the total weight by the number of pooled arteries.The isolated arteries were cannulated onto glass micropipettes and secured with 10-0 nylon sutures (Alcon Laboratories, Frenchs Forest, NSW).Arteries were checked for leaks and only vessels able to hold 120 mmHg for at least 10 s used.Vessels were stretched to optimal length (straight and not "bowed") and superfused (3 mL/min) with Krebs solution (34°C) and initially pressurized at 40 mmHg before being equilibrated at 70 mmHg until myogenic tone developed (~15 min).Vessel intraluminal diameter was measured using video microscopy and calibrated video calipers.Pressure was measured with a pressure transducer (Abbott Critical Care Systems, NSW).

| Experimental protocols for functional studies
For experiments shown in Figure 3, cumulative concentration-dilation curves were constructed to acetylcholine and adenosine, with continual artery superfusion.Arteries were then incubated with unglycated BSA (BSA) or glycated BSA (AGE) on the intraluminal surface of the vessel only, for 15 min prior to a second concentration-dilation curve being performed.The pipette holder used for BSA/AGE loading was modified to include an additional access port, which minimized dead space and enabled rapid pipette solution exchange while maintaining isobaric conditions. 37Responses to sodium nitroprusside were performed in separate artery segments to limit endothelial exposure to BSA or AGE.
For experiments shown in Figures 4 and 5, control concentration-dilation curves were constructed to one of acetylcholine, SNP or NS1619.The various enzyme or ion channel inhibitors or their vehicle solutions were incubated with the tissue for 30 min, both intraluminally (perfused) and in the bathing solution (superfused) prior to a second concentration-dilation relationship being obtained and remained for the duration of the experiment.Following a 15min incubation with AGE, a third concentration-dilation relationship was obtained.Time controls confirmed no significant alteration of responses to any of the agents across these protocols (not shown).
At the conclusion of all experiments, arteries were incubated in a nominally Ca 2+ -free Krebs solution (no added CaCl 2 plus 2 mM EGTA) for 20 min in order to obtain the maximum (passive) diameter at 70 mmHg.

| Materials
All drugs and chemicals were obtained from Sigma-Aldrich, Castle Hill, NSW, Australia unless otherwise stated.

| Data and statistical analysis
Data from functional studies were not used if the pressurized arteries failed to generate sufficient myogenic tone (≤60% of the initial diameter) or if the initial or control maximum relaxation response to acetylcholine, SNP or adenosine was <70% maximum diameter (criteria were different for NS1619 due to low selectivity at high concentrations).Data from three of the 76 arteries used were excluded due to these criteria.
Treatment of arterial segments with drugs was randomized.
Blinding of the operator was not feasible, but data analysis was performed semi blinded by an independent analyst.
The number of experiments in each experimental group was selected on the basis of observing a statistically significant effect while using the minimum number of animals (3R principles) and on the experience of previous studies.The number of tissue samples in each concentration-response group was 5-8 (exact numbers per treatment group are shown in tables and Figure legends).Five rats (total 20 arteries including non-CMA types) were used in ROS assay studies and four rats (total 16 arteries) for IF studies.For IF/ imaging studies, up to 20 images of each artery were taken for nonquantitative studies.The experimental group size is the number of independent values, and statistical analysis was done using these independent values.Outliers were included in data analysis and presentation, except where data were excluded under criteria described above (Experimental protocols).Basal diameter data were analyzed using one-way ANOVA (all analyses with GraphPad Prism v7.02, California, USA; RRID:SCR_002798).
For all analyses, values of p < .05were considered significant.

| RAGE expression in arteries
IF studies suggested a high level of diffuse RAGE expression in the endothelial cells (EC) of the CMA (Figure 1B), and both diffuse and punctate labeling in MCA endothelial cells, with large low-density puncta alternating with smaller high-density puncta, the latter apparently limited to individual endothelial cells (Figure 1D).RAGE IF was comparatively low in the endothelium of the MA and aorta (Figure 1A,C, respectively).In the vessel types under study, RAGE expression was not detected on the smooth muscle cells/medial layer (e.g., insets Figure 1A-D) or adventitial layer (not shown).

| Effects of AGE and Nox inhibition on ROS production in arteries
Basal or unstimulated superoxide (O − 2 ) accumulation, as measured by L-012 luminescence, was significantly greater in MCA compared with the other vessels, between which there was no difference in basal O − 2 accumulation.(Figure 2A).Basal hydrogen peroxide (H 2 O 2 ) accumulation, assayed by Amplex Red fluorescence, tended to be greater in the MCA than in the other vessels but this was not statistically significant (Figure 2B).Regarding

| Effect of AGE on myogenic tone and vasodilation of pressurized arteries
The mean passive diameter of the arteries used (at 70 mmHg) was 160.4 ± 2.0 μm (n = 73).Isolated, pressurized CMA developed myogenic tone, such that they constricted to ~50% of their maximum or passive diameter.The standard pressure utilized for vaso-relaxation experiments was 70 mmHg, but the effect of AGE on myogenic tone and responsiveness across a range of pressures was also examined.
Relative to recordings in normal Krebs buffer, neither BSA nor AGE altered myogenic tone over the pressure range 50-70-120 mmHg (Figure 3A).
The effect of AGE was examined on vasodilation initiated by three agents.Intraluminal incubation with AGE (0.1 mg/mL) for 15 min inhibited CMA vasodilation caused by the endothelium-dependent vasodilator acetylcholine (ACh), and the NO-donor sodium nitroprusside (Figure 3B,C; Table 1; Figure S2).The same AGE incubation protocol did not inhibit adenosine-induced vasodilation of the CMA, which is endothelium-independent in this vessel (Figure 3D).

Table 2).
The role of BK Ca was examined in the inhibitory effects of AGE on CMA vasodilation.AGE strongly inhibited dilation of the CMA induced the BK Ca activator NS1619 (0.1-300 μM; Figure 5A).
The BK Ca inhibitor iberiotoxin (0.1 μM) alone cause constriction of the CMA and inhibited ACh-induced dilation of the vessel.In the presence of iberiotoxin, AGE did not cause a further inhibition of ACh-induced dilation (Figure 5B; Table 2).Further experiments with NS1619 examined responses over a narrower range of concentrations (0.01-3 μM), as higher concentrations are not selective for BK Ca in the CMA. 32The RAGE antagonist, FPS-ZM1, alone inhibited responses to NS1619 but prevented the inhibition of NS1619-induced dilation of the CMA caused by intraluminal AGE (Figure 5C).Setanaxib (5 μM) alone had no significant effect upon NS1619-induced dilation but prevented further inhibition of this dilation by AGE (Figure 5D).

| Effect of ROS inhibitors on myogenic tone
The effects of ROS inhibitors and other enzyme or ion channel inhibitors on myogenic/basal tone of the arteries are shown in Figure 6.
Neither α-lipoic acid nor apocynin altered myogenic tone of the CMA at 70 mmHg (Figure 6).Setanaxib alone caused a small but significant constriction, both L-NAME and iberiotoxin alone caused a significant constriction of the CMA at 70 mmHg (Figure 6).

| DISCUSS ION
The present study demonstrated that AGE inhibited endothelium-dependent vasodilation of cremaster muscle arteries (CMA) from normal (i.e., non-hyperglycemic/diabetic) male rats through activation of endothelial RAGE in the vessel.The effect of AGE and RAGE activation was mediated by reactive oxygen species derived from Nox1/4 activity and inhibition of both NO-and BK Ca -mediated vasodilation in the artery.

| RAGE expression in arteries
Immunofluorescence demonstrated expression of RAGE in the CMA endothelium at a high level in contrast to the other arteries examined.Among these other arteries, there was also a gradation in endothelial RAGE expression, at a very low level in the aorta and at a low level in the mesenteric artery, but showing distinct patterns in the MCA.The contrast between the MCA and other vessels in this study may reflect the organization of individual RAGE monomers into dimers or oligomers required for RAGE activity 38,39 ; although why this pattern should be evident only in the MCA among the arteries investigated is not clear.This may reflect a distinct endothelial phenotype in the MCA contrasted with the CMA, MA, or aorta 18,20 ; perhaps reflective of the MCA producing relatively high levels of ROS.RAGE expression in general is thought to be at a low level in healthy states, but increased by elevated plasma AGE and conditions, including hypoxia, diabetes, and inflammation. 40,41Several studies have identified constitutive RAGE concentrated in vascular tissue, 42 and RAGE expression has been reported in renal and cerebral vascular smooth muscle. 22,43

| ROS accumulation and AGE
Basal or unstimulated superoxide and hydrogen peroxide accumulation was highest in the MCA, less in the CMA and MA and almost undetectable in the aorta.Similar findings regarding ROS accumulation being higher in cerebral as opposed to peripheral arteries have been reported previously and linked to cerebrovascular function. 44,45Notably, smaller arteries like the CMA and MA produced a significant level of basal ROS compared with larger vessels, including the aorta or carotid arteries. 44In the present study, AGE increased hydrogen peroxide accumulation in the CMA, but not in the other vessels tested, which had far lower RAGE expression, and the ac- and/or genetic deletion of Nox4 (but not Nox1) were preventive of endothelial and/or vascular injury in various models of diabetes, particularly diabetic renal microvascular disease. 28,47,48FPS-ZM1 also downregulated renal Nox4 expression in spontaneously hypertensive rats. 26

| Effects of AGE on vasoactivity of isolated CMA
The effects of AGE on functional responses of the CMA were examined.AGE did not alter the contractile response to transmural/intralumenal pressure (myogenic tone).AGE inhibited endothelium-dependent vasodilation of the CMA but did not alter adenosine-induced dilation, which is endothelium-independent in this vessel. 49The ability of AGE, via ROS production, to inhibit NO-mediated vasodilation is established. 4,8,12,14Thus, ROS such as O − 2 or H 2 O 2 react with NO, forming peroxynitrite (ONOO − ) which depletes biologically active NO.Peroxynitrite also uncouples the active NOS dimer, preventing NO formation (and exacerbating ROS production) through depletion of the NOS co-factor BH 4 . 13,50is idea is supported by the findings of the current study showing the NOS inhibitor L-NAME and the antioxidants apocynin and lipoic acid prevented the effect of AGE on endothelium-dependent vasodilation.
TA B L E 2 Effects of RAGE, NOS, Nox, and BK Ca inhibitors, alone and in the presence of AGE, on pEC 50 and maximum dilations (relative to passive diameter at 70 mmHg) of the rat isolated, pressurized (70 mmHg) CMA caused by acetylcholine (ACh).Numbers represent mean ± SEM ('n' = number of arteries from individual animals).In smaller, resistance-type arteries and arterioles, endothelium-dependent vasodilation of the type induced by acetylcholine is mediated primarily by EDH of vascular smooth muscle, with a lesser role for NO. 15,16In the rat CMA, there is a substantial role for BK Ca in EDH. 31,34The current study provides functional evidence that AGE are able to inhibit vascular BK Ca ; AGE strongly inhibited vasodilation caused by the BK Ca activator NS1619, while the BK Ca blocker iberiotoxin prevented the (further) inhibitory actions of AGE on endothelium-dependent vasodilation.Inhibition of BK Ca also could explain the inhibitory effect of AGE on SNP-induced dilation, as BK Ca mediate a substantial part of SNP's vasodilator effect in the rat CMA. 322][53][54] In contrast, peroxynitrite inhibited BK Ca in cerebral and coronary vascular smooth muscle cells. 55,56ese observations support those of the present study, whereby in- (i.e., there was no apparent RAGE on the smooth muscle).Activation of endothelial RAGE likely had no effect on STOCs, as AGE did not alter myogenic tone of the CMA which is directly affected by changes in smooth muscle STOC frequency. 57  signaling in BAECs, suggesting potential downstream inhibition of Ca 2+ -dependent endothelial K Ca , although no such acute functional effect was observed in the current study.

| Effects of Nox1/4 inhibition and antioxidants on myogenic tone of isolated CMA
1][62] Setanaxib increased myogenic tone of the CMA maintained at 70 mmHg, but neither apocynin nor lipoic acid had any effect on tone.Recent investigations suggest Nox1 or Nox2 in vascular smooth muscle may regulate myogenic tone through inhibition of Ca 2+ sparks and BK Ca , although contrasting effects between the CMA and cerebral arteries were again observed. 61The effects of Nox1 inhibition on myogenic tone were transient and possibly due to off-target effects of the agents used. 61,62The protocol of the current study with intraluminal addition of drugs prevented observation of transient effects, but inhibition of BK Ca directly (with iberiotoxin) did increase myogenic tone.Therefore, there was no clear evidence that Nox1/4 isoforms or ROS modulated myogenic tone in the rat CMA.It is conceivable that ROS modulate pressure-induced myogenic tone in cerebral arteries, given the comparatively high amount of ROS produced in this vessel.
Reactive oxygen species data are expressed as mean ± SEM and analyzed using one-way ANOVA.Arterial diameter in response to vasodilator agents was calculated as % dilation (((diameter -basal diameter)/(passive diameter at 70 mmHg -basal diameter)) × 100).Myogenic tone or basal diameter was calculated as a percentage of the maximum artery diameter at 70 mmHg (Dmax) as determined in zero-Ca 2+ Krebs solution (D = (D/Dmax) × 100).Group data are expressed as mean ± SEM.Concentration-response relationships were analyzed using two-way ANOVA followed by Tukey's multiple comparisons test where F in ANOVA achieved the level of statistical significance.Negative-log of the half-maximal effective concentration (pEC 50 ) was calculated by four-parameter logistic (4PL) nonlinear regression analysis of log [agonist concentration] vs. response graphs (GraphPad Prism v7.02, California, USA; RRID:SCR_002798).Maximum responses were determined from experimental data.
subsequently applied AGE samples, successful glycation of BSA incubated with MGO was indicated by an increase in MGO-BSA sample fluorescence and absorbance over time, and relative to BSA incubated without MGO (FigureS1).AGE (1 mg/mL glycated BSA) increased H 2 O 2 accumulation relative to control treatment (1 mg/mL unglycated BSA) in the CMA, but not the MA or MCA (Figure2C).The AGE-induced increase in H 2 O 2 in the CMA was prevented by preincubation of the vessels with the RAGE receptor antagonist FPS-ZM1 (1 μM) or the NAD(P)H oxidase (Nox) inhibitor apocynin (500 μM; Figure2D).Further studies employed Nox subtype-selective inhibitors setanaxib (Nox1/4) and GSK2795039 (Nox2).Setanaxib (5 μM) inhibited AGE-stimulated H 2 O 2 accumulation in isolated CMA, but GSK2795039 (20 μM) did not significantly affect AGE-induced H 2 O 2 accumulation (Figure2D).

Further
studies investigated the inhibitory signaling mechanisms of AGE on vasodilation of the CMA.Incubation of the vessels (intra-and extraluminal) with the RAGE antagonist FPS-ZM1(1 μM) prevented further inhibition of ACh-induced dilation by AGE, with FPS-ZM1 alone causing a significant inhibition of AChinduced dilation of the CMA (Figure4A; Table2).The NOS inhibitor L-NAME (100 μM) caused a significant, if modest, inhibition of ACh-induced vasodilation and in the presence of L-NAME, AGE did not cause a further inhibition of ACh-induced dilation of the vessels (Figure4B; Table2).The NAD(P)H oxidase inhibitor apocynin (500 μM) and the antioxidant α-lipoic acid (2 μM) also prevented AGE-induced inhibition of ACh-induced vasodilation.Neither apocynin nor lipoic acid alone had a significant effect on F I G U R E 1 RAGE expression in the in situ endothelial cells (EC) of male Sprague-Dawley rat aorta (A), first-order cremaster muscle artery (B), third-order mesenteric artery (C), and middle cerebral artery (D).Confocal immunofluorescence was used to determine RAGE protein distribution in EC or smooth muscle cells (SM; in insets).Outlined part of the image in panel (D) indicates a single EC of the middle cerebral artery with a high density of small puncta in contrast to the larger, low-density puncta present in other EC in the image.Longitudinal vessel axis left-to-right (with EC long axis aligned parallel to and SM long axis at ~90° to this axis, respectively) with the same vessel segments but at different focal planes in inset (SM) images.Propidium iodide (PI) indicates cell layer patency; PI staining in panel A shows SM (upper inset) representative of SM insets in panels B, C and D, and EC nuclei (lower inset).Control image for cremaster muscle artery (application of secondary antibody only) is shown in lower inset, panel (B).Also shown in inset is an auto-fluorescent image of the internal elastic lamina of the mesenteric artery (IEL; C).Images are representative of vessels from four different animals (n = 4).Scale bar = 25 μm.ACh-induced dilation (Figure 4C,D; Table

F I G U R E 2
Basal and AGE-stimulated superoxide/O − 2 (A) and hydrogen peroxide/H 2 O 2 (B-D) accumulation in male rat isolated segments of middle cerebral (MCA), cremaster muscle (CMA), and mesenteric (MA) arteries and aorta.O − 2 accumulation was assayed by L-012 luminescence, H 2 O 2 through Amplex Red fluorescence.Artery segments were incubated with 1 mg/mL BSA (Con) or glycated BSA (AGE) for 30 min.Data in panel D show the effects of various agents on H 2 O 2 accumulation in the CMA; the RAGE antagonist FPS-ZM1 (FPS, 1 μM), the antioxidant and Nox inhibitor apocynin (Apo, 500 μM), the Nox1/4 inhibitor setanaxib (Set, 5 μM), and the Nox2 inhibitor GSK2795039 (GSK, 20 μM).Columns represent mean ± SEM; n = 5 for all.* in (A) represents significantly greater accumulation of basal O − 2 in the MCA compared with the other vessels; in (C) and (D), * represents significant AGE-induced increase in H 2 O 2 accumulation compared with control, while Δ indicates significant inhibition of AGE-induced H 2 O 2 accumulation by treatment (p < .05,one-way ANOVA with Tukey's multiple comparisons test).F I G U R E 3 Effect of unglycated protein (BSA) and AGE (0.1 mg/mL for both) on pressure-induced myogenic tone (A) and vasodilation induced by acetylcholine (ACh, B), sodium nitroprusside (SNP, C), and adenosine (D) in the rat isolated, pressurized (70 mmHg) cremaster muscle artery with myogenic tone.Concentration-response relationships were constructed to the vasodilator stimuli in the absence (Control) or presence of unglycated BSA (BSA) or glycated BSA (AGE; 0.1 mg/mL for each).Symbols represent the mean ± SEM; A, n = 8 for all; B, n = 8 for Con, AGE and n = 9 for BSA; C, n = 5 for all; D, Con n = 8, AGE n = 7, BSA n = 9; E, n = 4 for both.* in (B) and (C) represents significant AGE-induced inhibition of responses to acetylcholine and SNP, respectively, relative to BSA (p < .05,two-way ANOVA with Tukey's multiple comparisons test).

F I G U R E 4
Effect of various enzyme inhibitors on AGE-inhibition of ACh-or SNP-induced dilation of the rat isolated, pressurized (70 mmHg) cremaster muscle artery with myogenic tone.Shown are the effects of FPS-ZM1 (A, FPS, 1 μM), the nitric oxide (NO) synthase inhibitor L-NAME (B, 100 μM), apocynin (C and E, Apo, 500 μM), the antioxidant α-lipoic acid (D, LA, 2 μM) and setanaxib (F, Set, 5 μM).Concentration-response relationships were constructed to acetylcholine in the absence (Control) or presence of the inhibitor, then the inhibitor plus AGE.Symbols represent the mean ± SEM; A, n = 6 for all; B, n = 8 for Con, n = 6 for L-NAME and L-NAME + AGE; C, n = 5 for all; D, n = 5 for all; E, Con n = 5, Apo and Apo + AGE n = 4 each; F, n = 6 for all).* in (A), (B), and (F) represents significant effect of the drug/ enzyme inhibitor alone on responses to ACh (p < .05,two-way ANOVA with Tukey's multiple comparisons test).

hibition of NOS or H 2 O 2 /
ROS formation, either of which would forestall peroxynitrite accumulation, prevented the inhibitory action of AGE on BK Ca -mediated vasodilation of the rat CMA.Recent patchclamp studies showed AGE (glycated fibronectin) increased ROS accumulation and inhibited BK Ca -mediated spontaneous transient outward currents (STOCs) in RAGE-expressing vascular smooth muscle cells isolated from rat cerebral arteries. 22In the present study, RAGE were found on the endothelium of the intact CMA only F I G U R E 5 (A) Effect of AGE on dilation of the rat isolated, pressurized (70 mmHg) cremaster muscle artery induced by the BK Ca activator NS1619.(B) Effect of the BK Ca blocker iberiotoxin (IbTx, 0.1 μM) on ACh-induced dilation of the pressurized CMA in the absence and presence of AGE (0.1 μM).(C) The effect of FPS-ZM1 (FPS, 1 μM) and (D) setanaxib (Set, 5 μM) on AGE-inhibition of NS1619-induced dilation of the pressurized CMA.Concentration-response relationships were constructed to acetylcholine in the absence (Control) or presence of the inhibitor, then the inhibitor plus AGE.Symbols represent the mean ± SEM; A, n = 6 for both; B, n = 8 for Con, n = 5 for IbTx and AGE + IbTx; C, n = 6 for all; D n = 5 for all.* in (A), (B) and (C) represents significant effects of AGE, iberiotoxin and FPS-ZM1, respectively, on responses to NS 1619 or ACh (p < .05,two-way ANOVA with Tukey's multiple comparisons test).
in the present study.In the absence of AGE, both FPS-ZM1 and setanaxib alone inhibited acetylcholine-induced vasodilation.FPS-ZM1 also partially inhibited NS1619-induced dilation of the CMA, suggesting a possible inhibitory effect of the drug on BK Ca activity.Setanaxib alone did not significantly alter NS1619induced dilation of the CMA, but inhibited acetylcholine-induced, endothelium-dependent dilation.Munoz et al. showed setanaxib and plumbagin, a nonselective Nox4 blocker, inhibited endothelium-dependent dilation of rat isolated renal interlobar arteries from both lean and obese Zucker rats.59These investigators suggested a role for Nox4-derived H 2 O 2 in acetylcholine-induced dilation of this artery.59However, there is little evidence for such a mechanism in the rat CMA.31 Furthermore, in the present study neither apocynin nor lipoic acid altered endothelium-dependent dilation of the CMA, further supporting the idea that H 2 O 2 or other ROS are not involved in this response.A portion of the endothelium-dependent vasodilation of the CMA was not affected by AGE, this is likely the proportion of EDH of smooth muscle mediated by endothelial I-and SK Ca through myo-endothelial gap junctions, as demonstrated previously. 31,32There are few studies in the area of AGE and EDH of vascular smooth muscle; a study in mesenteric arteries reported longer-term incubation with AGE (3 h) decreased vasodilation induced by acetylcholine and NS309, a nonselective activator of I-and SK Ca , and decreased I-and SK Ca expression in HUVECs. 11Our previous investigation 22 showed glycated albumin or extracellular matrix proteins could impair Ca 2+ Glycated proteins (AGE) inhibited endothelium-dependent dilation of a rat skeletal muscle artery through endothelial RAGE-linked activation of Nox1 and/or 4. The ROS thus generated inhibited F I G U R E 6 Effect of various enzyme and ion channel inhibitors on myogenic tone (basal diameter) of the rat isolated, pressurized (70 mmHg) cremaster muscle artery.Shown are the effects of α-lipoic acid (αLA, n = 6), apocynin (Apo, n = 6), setanaxib (Set, n = 11), L-NAME (n = 8) and iberiotoxin (IbTx, n = 8).Bars represent the mean ± SEM before (Control) and 5 min after (Treat) addition of the drug.*represents significant effects of drug on myogenic tone (p < .05,one-way ANOVA with Tukey's multiple comparisons test).
3 MOPS, 145 NaCl, 5 KCl, 2.5 CaCl 2 , 1 MgSO 4 , 1 NaH 2 PO 4 , 0.02 EDTA, 2 pyruvate, 5 glucose and 1% bovine serum albumin (BSA; Experiments with selective Nox inhibitors suggested RAGE increased H 2 O 2 accumulation by increasing Nox1 and/or Nox4 activity in the CMA.Notably, the Nox1/4 inhibitor setanaxib did not fully prevent the AGE-stimulated increase in H 2 O 2 , in contrast to both the RAGE antagonist and apocynin, suggesting another potential source of RAGE-stimulated H 2 O 2 .Nox3 expression is low or absent in vas- 1,4ulation in the CMA was prevented by the RAGE antagonist, suggesting a fundamental role of endothelial RAGE in AGE-stimulated ROS accumulation in the vessel.cularendothelialcellswhilerodents (unlike humans) do not express Nox5.46RAGEstimulatemultipleinflammation-linked signaling pathways which could conceivably produce ancillary ROS apart from Nox activation, or AGE may increase ROS accumulation without activating RAGE.1,4Some previous studies have suggested vascular or endothelial RAGE are coupled to Nox4 on the basis that setanaxib

Treatment Con pEC 50 ; n Treatment pEC 50 ; n Treatment + AGE pEC 50 ; n Con Max (%) Treatment Max (%)
Note: Concentration-response relationships were constructed to acetylcholine in the absence (Control) or presence of the inhibitor (Treatment) and then the inhibitor plus AGE (Treatment + AGE).*indicates significant difference from control (Con; p < .05,one-way ANOVA).
57elial cholinergic receptors to smooth muscle BK Ca activation in the rat CMA.31Such an interaction between H 2 O 2 and CYP 450 has been demonstrated in coronary arteries,58but does not explain the effect of AGE on NS1619-induced dilation of the CMA in the current study, presuming the drug activates the channel directly.In their study on smooth muscle RAGE and STOCs, Yang et al. also observed that glycated albumin (the same form of AGE used in our study) did not alter BK Ca activity, in contrast to glycated fibronectin.22Previousstudiesshowedseveral differences in regulation of BK Ca activity between rat cerebral arteries and CMA,57including the channel's regulation in