L-Arginine reduces downstream vascular contractility after flow-diverting device deployment: A preliminary study in a rabbit model

Abstract

Background

Flow diverters (FDs) are an effective treatment for intracranial aneurysms, though not free from hemorrhagic complications. A previous study demonstrated increased vascular contractility after FD-implantation as a potential mechanism of distal complications. Our study aimed to investigate whether L-arginine medication affects vascular contractility following FD deployment in a rabbit model.

Methods

FDs were implanted in the aorta of normal rabbits (+FD, n = 10), with sham-operated aorta as controls (n = 5). L-Arginine was given in the drinking water (2.25% L-arginine hydrochloride) of half of the +FD animals (+FD/+Arg). Force contraction vascular contractility studies were performed on the aortic rings proximal and distal to the FD using an organ bath. Total eNOS, eNOS(pS1177), eNOS(pT495), COX-2, and S100A4 were quantified by western analysis on total protein lysates from aortic segments, normalizing to GAPDH.

Results

Mean vascular contractility was 53% higher in distal relative to proximal aortic segments (P = 0.0038) in +FD animals, but were not significantly different in +FD/+Arg animals, or in sham-operated controls. The +FD animals expressed significantly reduced levels of eNOS(pS1177) than sham-operated controls (P = 0.0335), while both the +FD and +FD/+Arg groups had reduced levels of eNOS(pT495) relative to sham-operated controls (P = 0.0331 and P = 0.0311, respectively).

Conclusion

These results suggest that L-arginine medication reduces distal vascular contractility after FD treatment via nitric oxide production and thus might mitigate risk for downstream complications.

Introduction

Endovascular treatments for intracranial aneurysms are an effective alternative to invasive surgery. ¹ Though endovascular coiling is the most common method used for unruptured aneurysms, alternative or supplementary devices are being increasingly implemented to treat both ruptured and unruptured aneurysms, and to prolong the long-term success rates of such non-invasive procedures. ² Flow diverting devices (FDs) have shown superior occlusion rates relative to coils, though they are associated with a relatively high (17%) overall complication rate. ³ A particularly devastating complication is that of delayed, distal hemorrhages, which is noted in 2.4% of patients treated with flow diverters. ⁴ Understanding the mechanisms behind these vascular complications can help in the development of pre- and post-procedural therapies and allow for more choices in endovascular treatment options.

It was previously demonstrated that vascular contractility increased in the aorta of a rabbit model after FD-implantation as a potential mechanism of distal complications. ⁵ The cause of these distal vascular changes is still unknown, but several studies have suggested that the device itself may alter blood flow, causing changes in the vessel wall structure distal to the device. ⁶ Indeed, shear stress has been shown to increase vascular smooth muscle contractility,^7,8 and disrupted or turbulent blood flow can lead to changes in smooth muscle cell morphology causing abnormal vascular tone and vasospasms. ⁹

Nitric oxide is well known for its abilities to inhibit vasoconstriction and vascular remodeling, ¹⁰ and is generated by nitric oxide synthase, whose substrate, L-arginine, is found naturally in the environment. An early study demonstrated that long-term administration of L-arginine to rabbits after catheter balloon denudation injury reduced neointimal thickening and improved neoendothelial-dependent acetylcholine-induced relaxation. ¹¹ Thus, in this study, we aimed to investigate whether systemic L-arginine administration affects vascular contractility following FD deployment in a rabbit model.

Methods

In vivo studies

All animal research protocols were approved by our Institutional Animal Care and Use Committee. FDs (Pipeline Embolic Devices, Medtronic, Inc.) were implanted in the abdominal aorta of ten female New Zealand white rabbits (weight 2.6–3.4 kg) as described previously. ¹² Rabbits were premedicated two days prior to FD deployment with aspirin (10 mg/kg orally) and clopidogrel (10 mg/kg orally). Antiplatelet therapy was continued for 1 month following device implantation. Five of these animals received no other treatments (+FD, n = 5), and the other five received L-arginine medicated (2.25% L-arginine hydrochloride) drinking water ¹¹ (+FD/+Arg, n = 5) for the 8 week duration until sacrifice. Five separate animals were used as sham-operated controls (n = 5), where they underwent the same procedures as the FD-treated animals, but no FD was deployed.

Tissue harvest

All animals were euthanized eight weeks after FD deployment. Segments of the aorta proximal and distal to the implanted device, or the same areas of the control aortas, were extracted and divided for molecular and vascular contractility analyses. Segments were snap-frozen in liquid nitrogen for molecular analyses, or placed in oxygenated physiological salt solution PSS for vascular contractility analysis.

Blood plasma L-arginine levels

Blood was collected in sodium citrate blood collection tubes (BD Biosciences), and plasma separated via centrifugation. Plasma L-Arginine levels were measured by ultra performance liquid chromatography mass spectrometry (UPLC-MS) as previously described. ¹³ Briefly, plasma samples were spiked with an internal standard then deproteinized with cold methanol followed by centrifugation at 18,000 × g, 4 °C for 15 minutes. The supernatant was dried down and immediately derivatized with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate according to Waters’ MassTrak kit. A 10-point calibration standard curve underwent a similar derivatization procedure after the addition of the same internal standard. Both derivatized standards and samples were analyzed on a triple quadrupole mass spectrometer coupled with an Ultra Pressure Liquid Chromatography system. Data acquisition was done using select ion monitor (SRM). The concentration of Arginine of each unknown was calculated against its perspective calibration curve.

In vitro vascular contractility assay

After being cleaned of surrounding fat and connective tissues, aortic segments were cut into 2–3 mm long rings and placed into 4 mL of warm, oxygenated PSS in an organ chamber. Rings were connected to force transducers and isometric force was recorded continuously. The vascular rings were stretched to determine the optimal point of their length – tension. The output from the transducers was amplified by signal conditioners and transmitted to a computer for analog/digital conversion. For each aorta, two proximal and two distal rings were examined. The contractile response to a depolarizing concentration of potassium chloride (80 mM) provided a measure of maximal contractile responsiveness in each ring. All rings were then constricted with 1 μM phenylephrine. When maximal response produced by this agonist remained stable, the rings were washed with potassium chloride and allowed to return to resting tension. ¹¹

Immunoblot analysis of proteins

Total soluble proteins were extracted from aortic segments using RIPA buffer (MilliporeSigma) with mechanical homogenization in a Bullet Blender (Next Advance). Equal amounts of protein from each sample were run in experimental duplicates, and were separated and analyzed using a Wes automated western blot system (ProteinSimple). Protein targets included endothelial nitric oxide synthase (eNOS, Novus Biologicals), eNOS(phospho-Ser1177) (BD Biosciences), eNOS(phospho-Thr495) (BD Biosciences), cyclooxygenase-2 (COX-2, R&D Systems), S100 calcium binding protein A4 (S100A4, Atlas Antibodies), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH, Novus Biologicals). Each sample was multiplexed with up to three antibodies at once, always including GAPDH as a loading control.

Statistics

For vascular contractility analyses, descriptive statistics are provided as mean(±SD). The outcome variable for contractility was weight-normalized amplitude, calculated as (peak amplitude - baseline amplitude)/sample weight. Treatment groups and locations were compared to each other and to controls using mixed effects model with Sidak’s post-hoc test. Paired proximal and distal samples were taken from each test subject. Within-subject differences between locations are calculated as distal - proximal. P-values less than 0.05 were considered significant. Immunoblots were analyzed by normalizing the area under the curve for each target epitope’s relative fluorescence to that of GAPDH for each sample. Values provided are mean(±SEM). Outliers were removed using the ROUT method (FDR = 1%). Treatment groups were compared to controls using one-way ANOVA followed by Dunnett’s multiple comparisons test. P-values less than 0.05 were considered significant. All statistical analyses were performed using GraphPad Prism (version 8.0.0 for Windows, GraphPad Software, San Diego, California USA).

Additional information on materials, methods, and data may be obtained by request from the corresponding author.

Results

Of the 15 animals included at the start of the study, 14 were used for vascular contractility assays (93%, n = 5 per group for +FD/+Arg and controls, n = 4 for +FD). One +FD animal died due to complications. Data from 2 more animals were excluded in the immunoblot analyses of the distal segments (20%, one +FD and 2 controls) due to poor protein extraction, and data from one additional +FD animal was unusable for the eNOS immunoblot of the distal segments (27%, 2 +FD and 2 controls). There were no significant changes in the body weight of the animals between the groups at the time of device implantation and euthanasia.

Plasma L-Arginine levels

Blood plasma was analyzed from a total of N = 14 rabbits. The mean(SEM) plasma arginine levels were 169.9(49.65) µM in the +FD/+Arg medicated group, 71.4(1.44) µM in the +FD group, and 56.3(8.43) µM in the control group.

Vascular contractility

There were a total of N = 14 rabbits measured (5 control, 4 +FD, and 5 +FD/+Arg; See Table 1 for individual values). Of the treatment/location combinations, the +FD distal locations showed the highest weight-normalized amplitude (1.69(0.38)) and the +FD proximal locations the lowest (1.21(0.50)) (Table 2, Figure 1(a)). The mean amplitudes were not significantly different between sample groups per location (Table 3), but significant differences were seen comparing distal vs proximal locations overall (mean difference 0.25[90%CI 0.04–0.45], p = 0.011) and within the +FD treatment group alone (0.47[90%CI 0.19–0.75], p = 0.004) (Figure 1(b)). Thus, the mean amplitude of vascular contraction in the artery distal to FD was 21% lower in the +FD/+Arg group compared to +FD animals. The mean amplitude of vascular contraction was 53% higher in the artery distal to the FD compared to the artery proximal to FD in the +FD animals. However, the difference in vascular contractility between segments distal to FD and proximal to FD was 7% in the +FD/+Arg group (the error is high when comparing proximal and distal in individual cases).

Table 1.

Raw data values for vascular contractility measurements for each treatment group.

Group	Device + L-Arg					Device only					Control, control
Animal Number	1	2	3	4	5	6a	7	8	9	10	11	12	13	14	15
Vascular contraction-Distal to FD (Amplitude)	1.63	0.94	1.59	1.19	1.65	N/A	2.03	1.32	1.99	1.40	1.67	1.47	1.07	1.79	1.79
Vascular contraction- Proximal to FD (Amplitude)	1.56	0.95	1.56	1.47	1.11	N/A	1.70	1.05	1.52	0.59	1.67	1.47	1.07	1.79	1.79
% Vascular contraction-in Distal compared to Proximal	4.72	−1.15	1.57	−19.04	48.60		19.67	26.04	31.04	136.27	0.00	0.00	0.00	0.00	0.00

Note: For control, vascular contraction was measure in one arterial segment. Both distal and proximal are the same.

^aRabbit number 6 died.

Table 2.

Descriptive statistics for weight-normalized amplitude of aortic segments from untreated controls, flow diverter (“Device”) treated animals, and animals treated with flow diverter and medicated with L-arginine. Weight-normalized amplitude was calculated as (peak - baseline amplitude)/sample weight.

Treatment, Location	N	Mean (SD)
Control, control	5	1.56 (0.30)
Device only, distal	4	1.69 (0.38)
Device only, proximal	4	1.21 (0.50)
L-arginine, distal	5	1.40 (0.32)
L-arginine, proximal	5	1.33 (0.28)

Figure 1.

Vascular contractility results of aortic segments proximal to, and distal to, flow diverter placement. (a) Weight-normalized group average amplitudes (error bars represents a 90% confidence interval for the mean estimate). (b) Difference between mean amplitudes (error bars represent 90% confidence interval of mean) distal to, and proximal to, flow diverter placement. (Con=sham surgical untreated controls; +FD=flow diverter treated controls; +FD/+Arg=flow diverter treated animals medicated with L-arginine).

Table 3.

Pairwise comparisons between group means for vascular contractility. P-values are uncorrected for multiple testing. P-values are from Welch's t-test unless asterisked.

Comparison	Mean Diff.	90%CI Lower	90%CI Upper	p-value
Control vs device/distal	–0.13	–0.58	0.33	0.61
Control vs device/proximal	0.35	–0.23	0.92	0.28
Control vs L-arginine/distal	0.16	–0.20	0.53	0.44
Control vs L-arginine/proximal	0.23	–0.11	0.57	0.25
Proximal location: device vs L-arginine	0.29	–0.17	0.75	0.27
Distal Locations: device vs L-arginine	–0.12	–0.69	0.46	0.70
Both treatment groups (Distal–Proximal)*	0.25	0.04	0.45	0.06
Device only group (distal–proximal)*	0.47	0.19	0.75	0.03
L-Arginine group (distal–proximal)*	0.07	–0.21	0.35	0.63

*comparisons are from paired t-tests.

Immunoblot analysis of total protein extracts

A total of N = 12 distal aortic segments (3 controls, 4 +FD, and 5 +FD/+Arg) were used for protein analysis by immunoblot, except for eNOS analysis (N = 11, 3 controls, 3 +FD, 5 +FD/+Arg). Relative to controls, the +FD group showed significantly less eNOS(phospho-Ser1177) (P = 0.034) and eNOS(phospho-Thr495) (P = 0.033) (Figure 2). The +FD/+Arg group also showed significantly lower levels of eNOS(phospho-Thr495) (Figure 2, P = 0.031). Neither treatment group had statistically significant differences in eNOS relative to controls, but the +FD/+Arg was trending higher than both the +FD and control groups (Figure 2). The +FD group also showed lower levels of COX-2 and S100A4 than both the +FD/+Arg and control groups, though not significantly (Figure 2).

Figure 2.

Immunoblot results from analysis of proteins extracted from aortic segments distal to flow diverter. (a) Digital representation of immunoblot results from Wes automated protein analyzer. Each sample was loaded in duplicate (for consistency, only samples with clean results for all antibodies tested are shown; N = 3 control animals, N = 3 flow diverter treated animals, N = 4 flow diverter treated animals medicated with L-arginine). (b) Graphical representations of mean(SEM) relative luminescent units (RLUs) after normalizing to internal GAPDH control RLUs (N = 3 controls; N = 4 +FD, except eNOS N = 3; N = 5 +FD/+Arg, except S100A4 N = 4). +FD animals had significantly less eNOS(pSer1177) than controls (P = 0.0335), and both +FD and +FD/+Arg animals had significantly less eNOS(pThr495) than controls (P = 0.331 and P = 0.311, respectively). COX-2 and S100A4 levels appeared to be lower in +FD animals than both controls and +FD/+Arg animals, though not significantly. (*P < 0.05; Con=sham surgical untreated controls; +FD=flow diverter treated controls; +FD/+Arg=flow diverter treated animals medicated with Larginine; eNOS=endothelial Nitric Oxide Synthase; eNOS(pS1177)=eNOS phospho-Ser1177; eNOS(pT495)=eNOS phospho-Threonine 495; COX-2 = cyclooxygenase-2; S100A4= S100 calcium-binding protein A4).

Discussion

In this current study, we showed that systemic L-arginine treatment reduces vascular contractions after FD placement in an animal model. We also confirmed molecular differences between the animals treated with FD and the untreated, sham-operated controls, but found that protein levels appeared to normalize in three of the five nitric oxide signaling molecules we examined for the FD-treated rabbits medicated with L-arginine. We were also able to reproduce the significantly higher vascular contractility in aortic segments distal to the deployed FD in animals that did not receive supplemental L-arginine in their diet. These findings suggest that downstream vascular changes following FD implantation, which are implicated in severe hemorrhagic complications, may be mitigated with systemic therapies.

It has been postulated that the delayed intraparenchymal hemorrhage following FD treatment could be resulting from either the alteration of hemodynamics after FD implantation or the release of microembolic particles from the procedure. A recent study from Schob et al., ¹⁴ suggested that subacute vasospasm in the proximal and distal segments beyond the device could lead to a significant diminution of ipsilateral cerebral perfusion in flow diverter treated patients. That study also noted a positive association between device oversizing and the segmental vasospasm occurrence. Further, the high radial force of FD exerted on the vessel wall would potentially cause mechanotransduction along the arterial tree, resulting in vascular contraction. Significant increase in contractility and pulse wave velocity distal to FD in the aorta has also been reported. ⁵ It is possible that prolonged endothelial injury could be induced by implanted stents or ancillary devices during device implantation procedure. We demonstrated that FD implantation denude the endothelial layer in the parent artery. ¹⁵ Even though denuded arteries are re-endothelialized by surrounding endothelial cells (ECs) within days, re-endothelialization is not always complete and is often dysfunctional as the FD may alter the local fluid shear stress and mechanical forces, which could impact endothelial repair mechanisms in the device implanted arteries.^16,17 The observed near reversal of vascular hyper-contractility in Arginine treated animals provides a potentially promising approach for mitigation of this dangerous side-effect until physiological adaptation to the device can be attained. However, additional longitudinal studies are warranted to validate the relationship between ICH and vascular changes in the vessel wall.

The results from molecular analysis paralleled the vascular contractility findings in that the rabbits receiving L-arginine showed similar protein levels as the untreated controls for most proteins. In contrast, the FD-treated rabbits showed reduced levels of all investigated proteins, confirming changes in molecular signaling patterns in the aorta distal to the FD. The changes seen in the FD-treated animals support findings published previously demonstrating altered signaling patterns resulting from changes in shear stress, ⁸ which has been shown to be reduced in arteries distal to FD implants. ⁶ This reduction in shear stress is known to cause decreases in nitric oxide (NO) production in ECs, ¹⁸ among other signaling molecules, driving changes in gene expression leading to de-differentiation of the underlying vascular smooth muscle cells (VSMCs) and reduced vasodilation.^19,20 This phenotypic transition is associated with changes in the expression of several key proteins, including S100A4, which tends to be higher in the synthetic VSMC phenotype. ²¹ However, similar effects occur in response to endothelial injury resulting from catheter placement itself, due to endothelial injury and mechanical stretch. ²¹

We sought to attempt to rescue normal EC and VSMC signaling by providing system L-arginine, which is processed by eNOS in the ECs of the vessel walls, producing NO, and leading to reduced vasoconstriction in the VSMCs.^11,20 eNOS is activated by phosphorylation at several sites, including Serine 1177, which is primarily phosphorylated by PKA or Akt, typically in a calcium-dependent manner. Phosphorylation at Threonine 495, however, inhibits the enzyme’s activity. ²⁰ Thus, we analyzed levels of the total eNOS protein, and the levels of eNOS phosphorylated at the Ser1177 activation site, and the Thr495 deactivation site. To determine the effect of endothelial injury on observed changes, we also analyzed COX-2 levels, since COX-2 has been shown to play a role in regulating inflammation.^22,23 As expected, we did find increased levels of active eNOS (phospho-Ser1177) in the L-arginine medicated rabbits, but both FD-treated groups had reduced phosphorylation of eNOS at the Thr495 site, indicating dephosphorylation was occurring. However, since total eNOS levels were higher in the L-arginine medicated groups, the proportion of phospho-Thr495 eNOS was even lower than in the FD-treated controls. Although the COX-2 levels were not statistically different between groups, they did trend lower in the FD-treated controls. One possible reason for this is an increase in COX-2 in response to the increased levels of NO. This would also suggest a reduction of NO in the FD-treated controls relative to sham-treated negative controls and L-arginine medicated FD-treated animals.

Our study does suffer from a few limitations. First, outside of the mechanical properties of the vessel, we only assessed the biochemical and molecular changes. Studying the hemodynamic and other potential biophysical factors would provide additional insights into the mechanisms of downstream changes. Second, the FD was placed in a normal, straight, extracranial vessel, and we acknowledge that the physiology of cerebral circulation and vasculature are different from our animal model. Third, Arginine was administered though drinking water. Water consumption varies with each animal and thus the ingested dosage could be different among the animals. Further, the Arginine dose was selected from the literature. Optimizing a clinically relevant Arginine dose with a dose escalation study is necessary for clinical translation of this work. However, our studies are translatable to the patients who are exhibiting delayed thromboembolic complications after FD therapy. Unfortunately, there is no reliable blood biomarker reported for predicting thromboembolic complications. A comprehensive study will be required to identify the laboratory biomarkers for patients at risk of thromboembolic complications following FD implantation.

Conclusion

Our results demonstrate that altered distal vascular changes after flow diverter treatment can be reversed by arginine medication. Systemic pharmacotherapies can be used to reduce downstream vascular manifestations associated with flow diverter therapy.