Toxicologic Pathology, 42: 784-791, 2014 Copyright # 2014 by The Author(s) ISSN: 0192-6233 print / 1533-1601 online DOI: 10.1177/0192623314525687

Hemodynamic Correlates of Drug-induced Vascular Injury in the Rat Using High-frequency Ultrasound Imaging TERRI A. SWANSON2, TERI CONTE1, BEN DEELEY1, SUSAN PORTUGAL2, JOHN M. KREEGER2, LESLIE A. OBERT2, E. CLIVE JOSEPH2, TODD A. WISIALOWSKI2, SHARON A. SOKOLOWSKI2, CATHERINE RIEF3, PAUL NUGENT2, MICHAEL P. LAWTON2, AND BRADLEY E. ENERSON2 1

FUJIFILM VisualSonics, Inc., Toronto, Ontario, Canada Pfizer Worldwide Research and Development, Groton, Connecticut, USA 3 Pfizer Worldwide Research and Development, Andover, Massachusetts, USA 2

ABSTRACT Several classes of drugs have been shown to cause drug-induced vascular injury (DIVI) in preclinical toxicity studies. Measurement of blood flow and vessel diameter in numerous vessels and across various tissues by ultrasound imaging has the potential to be a noninvasive translatable biomarker of DIVI. Our objective was to demonstrate the utility of high-frequency ultrasound imaging for measuring changes in vascular function by evaluating blood flow and vessel diameter in the superior mesenteric arteries (SMA) of rats treated with compounds that are known to cause DIVI and are known vasodilators in rat: fenoldopam, CI-1044, and SK&F 95654. Blood flow, vessel diameter, and other parameters were measured in the SMA at 4, 8, and 24 hr after dosing. Mild to moderate perivascular accumulations of mononuclear cells, neutrophils in tunica adventitia, and superficial tunica media as well as multifocal hemorrhage and necrosis in the tunica media were found in animals 24 hr after treatment with fenoldopam and SK&F 95654. Each compound caused marked increases in blood flow and shear stress as early as 4 hr after dosing. These results suggest that ultrasound imaging may constitute a functional correlate for the microscopic finding of DIVI in the rat. Keywords:

drug-induced vascular injury; high-frequency ultrasound imaging; biomarker; CI-1044; SK&F 95654; fenoldopam.

Published cases of DIVI, principally in rat mesenteric and canine coronary arteries, have demonstrated an association with marked decreases in blood pressure with reflex tachycardia following administration of compounds, for example, adenosine agonists and phosphodiesterase 3 (PDE3) inhibitors, with vasodilator activity (Enerson et al. 2006; Isaacs et al. 1989; Joseph et al. 1996). These changes in systemic hemodynamics that lead to reflex tachycardia correlate with vascular injury and have been used to monitor for DIVI in the clinic (Kerns et al. 2005). However, there are also examples in which compounds cause DIVI in these same arterial beds without correlative changes in systemic hemodynamics outside the normal physiological ranges (Mesfin et al. 1989). It is hypothesized that localized changes in blood flow and arterial resistance occur in target blood vessels and are involved in the pathogenesis of lesion formation with vasoactive compounds. For example, endothelin receptor antagonists evoke >6-fold increases in dog coronary blood flow without major changes in heart rate and blood pressure (Mesfin et al. 1989; Louden et al. 2000), and the PDE4 inhibitor CI-1044 causes significant local changes in rat arterial blood flow in vessels that are a target of DIVI (Korkmaz et al. 2009). Significant increases in rat mesenteric blood flow have also been observed with the PDE3 inhibitor SK&F 95654 at doses that induce DIVI (Gardiner and Joseph, unpublished observations) and the D1 agonist fenoldopam (Lappe, Todt, and Wendt 1986); at doses that induce DIVI, however, these latter compounds also produce systemic hemodynamic changes.

INTRODUCTION Acute arterial drug-induced vascular injury (DIVI) is a common finding in preclinical toxicity testing of drugs in dogs and rats. However, similar findings have not been documented in clinical trials or in the clinical histories of patients treated with drugs that demonstrated this effect preclinically, making the relevance of this finding unclear with respect to human risk (Kerns et al. 2005). Concerns about the safety of these drugs persist primarily because current methods of clinical monitoring cannot accurately detect early vascular damage in humans (or, in animals). Determining the relevance of preclinical DIVI in human risk assessment requires a better understanding of the mechanisms of injury, in addition to the development of sensitive and specific biomarkers for the clinical diagnosis of acute vascular damage and inflammation. The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author(s) received no financial support for the research, authorship, and/or publication of this article. Address correspondence to: Bradley E. Enerson, Pfizer Worldwide Research and Development, Eastern Point Road, MS8274-1234, Groton, CT 06340, USA; e-mail: [email protected]. Abbreviations: DIVI, drug-induced vascular injury; EDV, end diastolic velocity; ECG, electrocardiogram; FMD, flow-mediated dilation; HF, high frequency; IACUC, Institutional Animal Care and Use Committees; IVC, inferior vena cava; MRI, magnetic resonance imaging; NIH, National Institutes of Health; NO, nitric oxide; NOS, nitric oxide synthase; PDE, phosphodiesterase; PSV, peak systolic velocity; RRA, right renal artery; SMA, superior mesenteric artery. 784

Downloaded from tpx.sagepub.com at DALHOUSIE UNIV on May 17, 2015

Vol. 42, No. 4, 2014

HEMODYNAMIC CORRELATES OF DRUG-INDUCED VASCULAR INJURY IN RATS

One potential mechanism for lesion formation is alterations in shear stress (Joseph et al. 1996). The vascular endothelium is generally subjected to two types of shear (frictional) forces that are due in part to blood flow patterns within the vessel itself: unidirectional laminar force and disturbed (turbulent or oscillatory) force. Endothelial cells respond differently to these types of shear forces and subsequently transfer these stimuli into distinctive phenotypes (Gimbrone, Nagel, and Topper 1997). Endothelial cells exposed to a steady laminar shear stress adopt a quiescent anti-inflammatory, anti-thrombotic, and antiapoptotic phenotype (Malek, Alper, and Izumo 1999; Chappell et al. 1998). In contrast, endothelial cells exposed to disturbed shear stresses and nonlaminar flow (oscillatory or turbulent) conditions display an activated phenotype (one implicated in the pathogenesis of vascular injury and disease) by rapidly transcribing a repertoire of genes for proteins involved in inflammation, coagulation, and proliferation (Brooks, Lelkes, and Rubanyi 2002; Chappell et al. 1998; Davies et al. 1986; Garcia-Cardena et al. 2001). Ultrasound imaging is a noninvasive technique that can measure blood flow and vessel diameter, which can then be used to calculate shear stress in numerous vessels in multiple tissues. In cardiovascular research, the measurement of flow-mediated dilation (FMD) using ultrasound represents a widely applied and prospectively validated clinical tool to assess vascular health (Charakida et al. 2010). Until now, there have been few studies using ultrasound to measure similar endpoints in rodents to establish translation of vascular health across species. This has been due in part to technical limitations of relatively low-frequency ultrasound in the context of the small vessel size in rodents. The recent development of high-frequency transducers (with axial resolution of 30–50 mm), together with automated edge detection and analysis software, makes it possible to monitor and measure reproducibly, small changes in vessel diameter and flow in rat peripheral arteries with vessel diameters as small as 0.2 to 0.5 mm. The promise of ultrasound imaging in a preclinical setting is that it provides a noninvasive functional measurement of changes in blood flow, associated shear stress, and other parameters that may correlate to lesion formation in specific tissues, thus potentially providing a mechanistic biomarker of DIVI that may also be translatable to man. The main objective of the studies described herein was to demonstrate the utility of high-frequency ultrasound imaging to evaluate functional changes in the superior mesenteric artery (SMA) of rats and to correlate these results with microscopic findings in response to treatment with vasodilators known to cause vascular injury. Although DIVI is typically found to occur in the second or third order vessels of the rat mesenteric arteries, we used the SMA as a surrogate for these smaller vessels as the SMA can be reproducibly located and measured using high-frequency ultrasound. Our findings demonstrate that localized increases in blood flow and shear stress in the SMA precede microscopic vascular lesions in the mesentery of rats treated with compounds known to cause DIVI in rats.

MATERIALS

AND

785

METHODS

Animals Male Sprague-Dawley (SD) rats were purchased from Charles River Labs (Wilmington, MA) and were approximately 200 g on receipt. All animals were maintained according to the National Institutes of Health (NIH) standards established in the Guide for the Care and Use of Laboratory Animals. The Pfizer Institute of Animal Care and Use Committee (IACUC) approved all experimental protocols. Rats were single-housed, had free access to water, and were fed a standard commercial laboratory certified rodent diet 5002 (PMI Feeds, Inc.). The testing facility maintained 12-hr light/dark cycle, with controlled temperature, humidity, and air changes. In-life Studies For ultrasound imaging, rats were anesthetized and maintained with 2% isoflurane/oxygen, and the abdomen and thorax were shaved and depilated of all fur (Nair, Church & Dwight Co). Body temperature was measured by rectal probe and supported using a homeothermic imaging stage (VisualSonics, Toronto, ON) that also provided electrocardiogram (ECG) and respiration monitoring. Ultrasound images of the superior mesenteric arteries and right renal arteries were collected before and after drug treatment. Following baseline imaging, typically performed 24 hr prior to dosing, rats were randomized into groups (n ¼ 6–8) based on the diameter of the SMA. 



Study 1: Separate groups of rats were dosed via oral gavage (PO) with either vehicle (5% methylcellulose) or 160 mg/kg CI-1044 (Pfizer Inc) and by subcutaneous (SC) route with either vehicle (saline) or 100 mg/kg fenoldopam mesylate (Sigma-Aldrich, St. Louis, MO). Study 2: Rats were dosed SC with either vehicle (dimethyl sulfoxide, 100%) or 200 mg/kg SK&F 95654 (Pfizer Inc).

All rats were imaged at baseline and then at 4, 8, and 24 hr after dosing. In study 1, separate groups of rats were euthanized at each time point following imaging. In study 2, rats were euthanized after imaging only at the 24-hr time point. Following euthanasia, blood was collected for biomarker analysis, and mesenteric tissues were collected for histological analysis. Histology Rats were euthanized under isoflurane gas anesthesia followed by exsanguination. At necropsy, the mesenteric arteries were collected and fixed in 10% neutral buffered formalin for 24 to 48 hr. After trimming, the tissues were processed on a Sakura Tissue Tek VIP through graded alcohols and cleared with xylene before being infiltrated with Paraplast Plus. The tissues were placed into embedding molds, covered with Paraplast Plus, and cooled on a chill plate. Blocks were cut at 5 microns, dried overnight, and stained with hematoxylin and

Downloaded from tpx.sagepub.com at DALHOUSIE UNIV on May 17, 2015

786

SWANSON ET AL.

TOXICOLOGIC PATHOLOGY

eosin (H&E). Microscopic examination was performed and the lesions were graded as minimal (1), mild (2), moderate (3), marked (4), or severe (5). Grade 1—Minimal: a histopathologic change ranging from inconspicuous to barely noticeable but so minor, small, or infrequent as to warrant no more than the least assignable grade. Grade 2—Mild: a histopathologic change that is a readily noticeable but not a prominent feature of the tissue and/or may be considered to be of no functional consequence. Grade 3—Moderate: a histopathologic change that is a prominent but not a dominant feature of the tissue and/or may be considered to have limited impact on organ function. Grade 4—Marked: a histopathologic change that is a dominant feature and may be an overwhelming component of the tissue and/or may cause significant tissue or organ dysfunction. Grade 5—Severe: a histopathologic change that is consistently an overwhelming component or represents an end-stage feature of the tissue (i.e., total organ failure).

Imaging Procedure The Vevo2100 (Fujifilm VisualSonics) system was used to obtain color Doppler, pulsed wave Doppler, and B-mode images of the superior mesenteric arteries, right renal arteries, and short axis of the heart (M-mode) in all study animals. Using Vevo2100 software, the diameter of the SMA was measured in B-mode at each time point by placing a measurement caliper on the endothelium of the vessel anterior wall and the endothelium of the posterior wall as previously described (Gutierrez et al. 2002). All vascular diameters were measured at the same pulse time representing the maximal diameter of any wall segment. This timing is the maximum diameter or when the maximum pulse is traveling through the measurement area. Doppler flow was measured on 3 heartbeats that were free of respiration noise for the analysis of peak systolic velocity (PSV), end diastolic velocity (EDV), and velocity time integral. M-mode analysis of the short axis of the heart was done by tracing the heart walls, which provides measures of cardiac output, and ejection fraction. Flow was calculated as:  (diameter/2)2  peak flow velocity (V). V is calculated by the ultrasound system as velocity time integral  duration of heart cycle. Wall shear stress was calculated for readings over 3 heart cycles as: 8  m V/diameter, where blood viscosity (m) was assumed to be constant at .035 dyns–1cm–2. Flow velocity represents the peak angle-corrected Doppler-flow velocity.

Statistical Analyses Statistical analyses were conducted on the baseline-adjusted values of shear stress, blood vessel diameter, EDV, and PSV measured in the superior mesenteric arteries at 4, 8, and 24 hr post-dose to compare the compound-treated rats with the vehicle-treated rats. Separate analyses were conducted at each post-dose time point. For each study, comparisons of the compound-treated rats with the vehicle-treated rats were performed using one-sided Dunnett’s test.

FIGURE 1.—Both fenoldopam and SK&F 95654 cause drug-induced vascular lesions in rats 24 hr after dosing. (A) Rat mesenteric artery from a vehicle-treated animal with no vascular injury present, H&E, 20 magnification. (B) Rat mesenteric artery 24 hr after fenoldopam treatment showing mild to moderate perivascular accumulations of mononuclear cells and lesser numbers of neutrophils in the adventitia and superficial tunica media along with minimal hemorrhages in the media, H&E 10 magnification. (C) Rat mesenteric artery 24 hr after treatment with SK&F 95654 showing mild hemorrhage and minimal apoptosis in media with mild cellular infiltrates in the adventitia, H&E 10 magnification.

Downloaded from tpx.sagepub.com at DALHOUSIE UNIV on May 17, 2015

Vol. 42, No. 4, 2014

HEMODYNAMIC CORRELATES OF DRUG-INDUCED VASCULAR INJURY IN RATS

FIGURE 2.—Anatomy of rat abdomen in cross section from B-mode ultrasound imaging showing location of the descending aorta, portal vein, and inferior vena cava (IVC) at the level of the liver.

787

FIGURE 3.—Ultrasound image illustrating the descending aorta and branching of the superior mesenteric artery (SMA) and right renal artery (RRA) in cross section. Diameter measures (L) are collected from this view as shown.

RESULTS Microscopic Findings In study 1, selected rats were euthanized at 4, 8, and 24 hr (n ¼ 6) following imaging. Microscopic evidence of injury in the mesenteric arteries was observed at 24 hr post-dose and only in fenoldopam-treated animals (Figure 1). Lesions were characterized by mild to moderate perivascular accumulations of mononuclear inflammatory cells and fewer neutrophils, the latter were occasionally present within blood vessel walls. Mural hemorrhage and necrosis were also frequently seen. Vehicle-treated and CI-1044-treated rats had no histologic abnormalities. In study 2, injury in the mesenteric arteries was observed in SK&F 95654–treated animals at 24 hr post-dose with minimal hemorrhage and apoptosis of the media smooth muscle predominating. Minimal vacuolation of the medial smooth muscle and minimal to mild perivascular cellular infiltrates were evident at 24 hr. There were no histologic abnormalities in vehicle-treated animals. Ultrasound Imaging Analysis Although the histological findings were located in the mesenteric arteries, these vessels cannot be imaged due to their small size, gut motion, and presence of food/air in the gut. Instead, the SMA was used as a surrogate vessel for monitoring changes in blood flow into the mesenteric vascular bed. The SMA was identified by locating the descending aorta near the portal vein and vena cava (Figure 2) and following the aorta caudally to the SMA (Figure 3). Once the B-mode acquisition was established, color Doppler was used to identify the region of highest blood flow, which was identified by turbid or mixedcolor Doppler, and spectral Doppler was used to collect images on a minimum of 3 consecutive heartbeats without respiration interference (Figure 4). The diameter of the SMA was measured using the ultrasound system B-mode measurement software (Figure 3). These data were used to randomize rats to

FIGURE 4.—Image of the superior mesenteric artery (SMA) in the rat with color Doppler showing turbulent (mixed color) flow. Doppler measures are collected from this region and analysis of the Doppler data are shown.

the different dosing groups, ensuring similar average vessel diameters were used in each cohort. Statistical analysis of the various ultrasound measurements taken in studies 1 and 2 are shown in Table 1. The table shows the mean and standard errors of the measurements, with the statistically significant comparisons noted. In study 1, ultrasound measurements of the SMA from fenoldopam-treated rats at 8 hr post-dose revealed significant increases in shear stress from pre-dose levels (Figure 5), and significant decreases in vessel diameter from pre-dose levels when compared to the vehicletreated rats. In addition, the ultrasound measurements from fenoldopam-treated rats demonstrated significant decreases in EDV from pre-dose levels at 4, 8, and 24 hr post-dose compared to the vehicle-treated rats. Cardiac output remained relatively stable and did not change significantly in response to either treatment. Blood flow in the right renal artery (RRA)

Downloaded from tpx.sagepub.com at DALHOUSIE UNIV on May 17, 2015

788

SWANSON ET AL.

TOXICOLOGIC PATHOLOGY

TABLE 1.—Hemodynamic parameters of rats treated with compounds that cause vascular injury. PSV (cm/s) Treatment Study 1

Vehicle

Fenoldopam

CI-1044

Study 2

Vehicle

SK&F95654

EDV (mm/s)

Time (hr)

SMA sheer stress (dynes/sec)

SMA

RRA

SMA

0 4 8 24 0 4 8 24 0 4 8 24 0 4 8 24 0 4 8 24

32.186 (4.807) 44.718 (3.853) 41.75 (15.192) 32.074 (7.497) 34.949 (4.816) 65.158 (17.871) 87.838* (20.168) 59.752 (15.500) 37.746 (4.570) 37.213 (6.173) 23.180 (4.723) 31.286 (5.082) 85.710 (8.954) 90.333 (12.966) 76.367 (15.886) 81.940 (17.290) 86.084 (13.111) 131.633* (20.153) 139.783* (24.255) 92.992 (23.961)

63.283 (7.220) 72.067 (10.774) 78.295 (10.518) 65.122 (13.718) 78.008 (5.120) 77.800 (18.980) 101.506 (6.443) 110.981 (12.749) 78.903 (5.369) 89.287 (7.427) 65.523 (12.536) 95.919 (9.175) 195.399 (23.384) 215.076 (13.019) 172.976 (33.105) 180.551 (32.591) 189.649 (21.556) 283.929** (22.399) 293.100* (36.985) 207.354 (39.500)

54.204 (.543) 55.205 (6.820) 52.863 (3.780) 57.474 (8.284) 54.073 (1.142) 80.069 (6.416) 97.057 (5.982) 81.530 (9.765) 58.350 (.216) 77.937 (10.753) 76.645 (9.435) 76.168 (5.941) 152.551 (22.885) 135.401 (12.529) 133.043(32.820) 109.868 (15.416) 132.856 (14.083) 189.407 (10.536) 195.660 (25.148) 101.435 (14.618)

17.930 (1.850) 23.388 (4.037) 21.041 (1.385) 22.991 (4.994) 27.528 (3.178) 10.046** (2.547) 9.344** (3.280) 7.261** (2.361) 23.600 (2.334) 15.875 (3.363) 18.877 (3.019) 16.526* (3.038) 44.002 (5.150) 36.077 (6.730) 29.914 (5.895) 41.729 (8.208) 50.321 (7.952) 35.801 (5.401) 38.081 (7.375) 29.389 (11.929)

RRA 18.244 22.519 14.533 18.027 15.334 17.660 15.001 14.987 19.070 18.392 18.126 28.410 47.403 27.491 31.219 39.133 42.487 18.218 20.586 15.730

(.944) (4.372) (2.151) (3.622) (.824) (3.105) (2.945) (1.213) (1.382) (2.715) (1.772) (4.703) (5.966) (4.605) (5.823) (6.326) (6.686) (4.739) (4.687) (4.872)

Cardiac output (ml/min)

SMA diameter (mm)

144.42 (5.90) 135.69 (33.49) 176.03 (31.16) 181.52 (31.04) 159.17 (10.65) 168.75 (31.52) 195.81 (13.07) 167.77 (33.70) 146.24 (9.76) 159.34 (52.42) 230.75 (28.77) 158.74 37.60 76.96 (5.33) 100.32 (16.86) 75.96 (3.34) 77.45 (6.83) 69.65 (4.05) 92.57 (13.74) 93.57** (4.39) 110.91* (17.93)

0.95 (0.03) 0.87 (0.04) 0.95 (0.05) 0.95 (0.05) 0.96 (0.03) 0.85 (0.05) 0.76** (0.04) 0.91 (0.05) 0.91 (0.02) 0.97 (0.04) 0.98 (0.03) 1.05 0.04 0.94 (0.02) 0.98 (0.05) 0.96 (0.04) 0.95 (0.03) 0.95 (0.03) 0.93 (0.04) 0.93 (0.03) 0.98 (0.05)

Note: SMA ¼ superior mesenteric artery, PSV ¼ peak systolic velocity, EDV ¼ end diastolic velocity, RRA ¼ right renal artery. All values represent group mean; standard error of mean is given in parentheses. *Significantly different from vehicle at 0.05 significance level. **Statistically significantly different from vehicle at 0.01 significance level.

FIGURE 5.—(a) Changes in shear stress in response to fenoldopam and CI-1044 treatment reveal significant changes at 8 and 24 hr post-dose leading to vascular lesions in the fenoldopam-treated rats. (b) In study 2, changes in shear stress were significant at 4 and 8 hr post-dose, with a return to baseline at 24 hr.

showed an increase from baseline at 4 and 8 hr post-dose, but was less than the flow increases observed in the SMA and was not associated with any lesion formation in the renal arteries. High-frequency ultrasound measurements from CI-1044treated rats revealed a decrease in EDV from pre-dose values at 24 hr post-dose when compared to vehicle-treated rats. In study 2, SK&F 95654-treated rats experienced significant increases in PSV and shear stress relative to baseline at 4 and 8 hr post-dose (Figure 5). Vessel diameter was not significantly altered with SK&F 95654 treatment. Cardiac output remained stable at 4 hr post-dose, but significantly increased at 8 and 24 hr post-

dose in the SK&F 95654-treated rats compared to the vehicletreated rats. Unlike animals treated with fenoldopam, shear stress in these rats returned to baseline levels at 24 hr post-dose. DISCUSSION Within the pharmaceutical industry, the lack of qualified and translatable biomarkers for DIVI often results in the termination of compounds that cause vascular injury in preclinical toxicology studies. Heart rate and blood pressure have historically been used in the clinic to monitor compounds that cause

Downloaded from tpx.sagepub.com at DALHOUSIE UNIV on May 17, 2015

Vol. 42, No. 4, 2014

HEMODYNAMIC CORRELATES OF DRUG-INDUCED VASCULAR INJURY IN RATS

FIGURE 6.—Upgraded imaging technology allows for improved visualization of the superior mesenteric artery (SMA; white arrows) in the rat. (A) An earlier system (Vevo770, top) employed a single focal zone and did not have the ability to clearly distinguish the SMA. (B) The current system (Vevo2100 bottom) employs broadband technology that helps to identify this vessel.

DIVI due to the link between systemic hemodynamic changes and vascular injury caused by drugs such as minoxidil (Kerns et al. 2005). However, compounds causing DIVI without concomitant and marked systemic hemodynamic effects are not easily monitored in a similar manner, prompting the need for novel preclinical biomarkers of vascular injury. We describe an ultrasound imaging approach that, with further validation, may be a useful noninvasive mechanism–based biomarker for DIVI. In this report, we utilize three compounds (fenoldopam, CI-1044, and SK&F 95654) that are vasodilators known to cause systemic changes in blood pressure as proof of concept tools for occasions when regional changes may be expected. This approach is expected to be most useful when localized changes in hemodynamics are suspected in the pathogenesis of the vascular injury based on the nature of the lesion and/or the mechanism of action.

789

The technological advances in ultrasound imaging systems such as the Vevo2100 have provided a more precise tool for preclinical researchers to noninvasively quantify a wide variety of translatable imaging and functional biomarkers. This new broadband technology allows for identification and imaging of the superior mesenteric arteries that was previously not possible when using a single focal zone high-frequency (HF) ultrasound (Vevo770, VisualSonics; Figure 6). In the current study, vascular and cardiac scanning were performed with the Vevo 2100 micro ultrasound, which can discriminate to 30 to 50 mm resolution and allows for monitoring small changes in vessel diameter of rodents. Doppler signals collected in the SMA using this platform are more reliable than those captured by laser Doppler techniques where the lack of concurrent imaging may result in signal acquisition from other abdominal vessels. Magnetic resonance imaging (MRI) and other imaging tools have not yet reached the temporal resolution of HF ultrasound that is required to capture the often subtle changes in blood flow velocity and sheer. High throughput ultrasound imaging allows for acquisition of data from each animal in less than 5 min total time allowing for reduced time under anesthesia. In addition, capability to monitor cardiac function and the ability to prescreen vessels with the Vevo 2100 in B-mode for randomization of vessel diameter is an additional benefit of ultrasound, thus ensuring consistency of diameters across cohorts, and ability to preselect cohorts with consistent cardiac output and renal function. The results of our studies demonstrate the utility of highfrequency ultrasound imaging for measuring changes in vascular function in arterial beds that are susceptible to vascular injury. The mesenteric vascular bed is often the site of predilection for DIVI in rats. Ultrasound evaluation of the superior mesenteric arteries of rats treated with D1 agonist fenoldopam or the PDE3 inhibitor SK&F 95654 demonstrated that functional changes in shear stress, vessel diameter, and PSV or EDV precede the microscopic evidence of vascular injury (either the earlier time points assessed in the fenoldopam arm of this study in the fenoldopam-treated animals or the established literature-based timing of SK&F 95654 lesion formation). The most significant change following administration of SK&F 95654 to rats was increased PSV and shear stress relative to baseline at 4 and 8 hr post-dose. Similarly, fenoldopam treatment resulted in significant increases in shear stress from pre-dose levels at 8 hr postdose. At 24 hr after treatment with fenoldopam and SK&F 95654, microscopic vascular lesions were evident. In this study, histological samples from earlier time points were not examined microscopically for SK&F 95654. However, in our experience and based on previous reports, focal hemorrhage can be observed as early as 12 hr post administration of either SK&F 95654 or fenoldopam (Joseph 2000). The observations using HF ultrasound to image blood flow and vessel parameters in the SMA of rats is fully consistent with data previously reported with SK&F 95654. In that study, SK&F 95654, administered at the same dose of 200 mg/kg used in the current investigation, resulted in a significant increase in mesenteric blood flow in the presence of a decreased blood

Downloaded from tpx.sagepub.com at DALHOUSIE UNIV on May 17, 2015

790

SWANSON ET AL.

pressure, demonstrating increases in mesenteric arterial conductance (Joseph 2000). The current investigation expands on this information by demonstrating an increase in shear stress in this arterial bed. Shear stress, together with hoop stress (the force on the vascular wall that is tangential to the lumen), has been postulated to be involved in the pathogenesis of vasodilator-mediated arteriopathy (Joseph 2000). While shear stress and blood flow were demonstrated to increase in the SMA, the resulting shear and flow patterns in the secondary branches of the mesenteric arteries were not studied here because of ultrasound resolution limitations. Therefore, it is not known whether the secondary and smaller branches of mesenteric arteries were subjected to the types of altered flow and disturbed shear that could contribute to lesion formation. It is also not known whether the flow pattern in the SMA in the current study was disturbed, or the magnitude of the disturbance. A recent study in dogs administered the potassium channel opener, ZD6169, or the endothelin receptor antagonist, ZD1611, demonstrated that both drugs induced very substantial changes in bidirectional oscillatory flow patterns in coronary blood vessels with increased velocity during the cardiac cycle that preceded vascular injury (Jones, Bjorkman, and Schofield 2013). Because all the compounds tested here are vasodilators, and rat mesenteric resistance arteries are sensitive to vasodilators (Lappe, Todt, and Wendt 1986), we suspect that the secondary and lesser branches of the mesenteric vascular bed are likely dilated. Vasoconstrictors such as midodrine and methoxamine are also known to induce vascular lesions in rat mesenteric arteries (Dalmas et al. 2011). It is hypothesized that administration of a vasoconstrictor would be accompanied by decreased blood flow in the SMA due to vasoconstriction of the downstream mesenteric resistance arteries. Studies are ongoing to determine ultrasound imaging correlates to these types of vasoconstrictors (a-adrenergic agonists). The CI-1044 data presented herein are inconsistent with a previous study in rats at similar dose levels, in which flow changes of *2-fold (compared to controls) were observed in mesenteric arteries 4 hr post-dose using an indirect blood flow measurement with microbeads (Korkmaz et al. 2009). In the current study, there were slight increases in PSV (*22%) and decreases in EDV (*20%) at 24 hr after treatment with CI1044. It was expected that CI-1044 would cause larger increases in blood flow parameters in the current study; the reasons for the discrepancies between the studies are not fully understood. However, it should be noted that there were also no microscopic findings with CI-1044 in this study. Some inconsistencies have been observed in the response to CI1044 across studies in our laboratory, and occasionally lesions are not apparent until 48 hr after dosing. The lack of microscopic findings, together with slight changes in flow parameters, may indicate that a potential threshold (magnitude and/or duration) in flow or shear is needed to elicit microscopic evidence of vascular lesions. Therefore, it is feasible that the slight nonsignificant increase in PSV observed with CI-1044 represents a subthreshold DIVI value. Further studies may help define the threshold blood flow and vascular parameters that

TOXICOLOGIC PATHOLOGY

lead to microscopic findings. Moreover, the use of anesthesia in the current study needs to be considered when comparing these data to other published work. More studies may be needed to explore the effects of anesthesia on blood flow in this paradigm. Common mechanisms likely underlie DIVI produced by vasoactive drugs in rats and recent reports have detailed a link between nitric oxide (NO) and these preclinical vascular lesions in mesenteric arteries. NO is a potent vasodilator, and vascular injury caused by vasoactive drugs may be the result of exaggerated pharmacology partially, or fully, mediated by NO signaling (Brott, Richardson, and Louden 2012). A role for NO in the pathogenesis of DIVI has been suspected due to the correlation between elevated levels of serum nitrite and vascular injury in SD rats treated with CI-1044 (Sheth et al. 2011). Studies conducted in separate laboratories with CI-1044 or fenoldopam demonstrate that addition of a NO donor will exacerbate DIVI lesions, while a nitric oxide synthase (NOS) inhibitor will ameliorate the formation of vascular lesions (Brott, Richardson, and Louden 2012; Sheth et al. 2011). These independent reports suggest the possibility of a common mechanism for drugs that cause vascular injury. The specific mechanism that explains NO mediation of vascular injury in rats has not been identified and applying these ultrasound imaging approaches to similar studies with NO donors and inhibitors could help establish the contribution of NO to the altered hemodynamics associated with DIVI. The findings described here represent an initial step toward a noninvasive functional imaging marker for DIVI. The translation of an ultrasound-based hemodynamic measurement from the preclinical toxicity setting to the clinic will be made easier by the widespread use of ultrasonic imaging techniques for quantification of FMD in the clinic. In the clinic, brachial artery FMD is used as an indicator of endothelial dysfunction and cardiovascular disease risk, and reductions in brachial FMD are correlated with elevations of circulating biomarkers of inflammation (Alber et al. 2007; Charakida et al. 2010; Filer et al. 2003; Kathiresan et al. 2006). Furthermore, a technique for ultrasonic assessment of FMD in femoral arteries has been described for rats (Heiss et al. 2008). However, rat femoral arteries are much less sensitive to DIVI producing vasodilators such as CI-1044 when compared to mesenteric resistance arteries, so the suitability of this model for DIVI would require further study (Korkmaz et al. 2009). More research is also needed to determine if preclinical DIVI results in endothelial dysfunction that is widespread and can be sensitively measured in vessels distant from the site of the lesion such as with FMD, or whether the endothelial dysfunction is localized to the affected arteries or vascular beds. In conclusion, preclinical DIVI is frequently associated with compounds that have a primary or secondary hemodynamic effect in toxicity studies. This association has been documented for vasodilators of different pharmacological classes, including potassium channel openers, PDE3 inhibitors, PDE4 inhibitors, D1 agonists, and endothelin antagonists. A common feature of these compound classes is increases in regional blood flow at sites of predilection for induction of arteriopathy. In this report, we

Downloaded from tpx.sagepub.com at DALHOUSIE UNIV on May 17, 2015

Vol. 42, No. 4, 2014

HEMODYNAMIC CORRELATES OF DRUG-INDUCED VASCULAR INJURY IN RATS

describe a noninvasive method for measuring changes in regional hemodynamics in rats and demonstrate that these alterations occur early and precede microscopic findings of vascular injury. Changes in vascular tone, function, or physiology can be easily measured in preclinical studies and integrated with histopathology results to potentially monitor early onset, progression, or reversibility of DIVI caused by small molecules with vasoactive properties and perhaps effects with other pharmacologic classes of compounds. This technology can also be used to investigate the mechanism of action of vascular lesions if altered blood flow is suspected as the underlying causal factor in the pathogenesis. The approach described here may also have potential use as part of a preclinical screening strategy to identify candidates in chemical series that lack specific vascular changes or identify candidates with expanded safety margins for vascular liabilities. ACKNOWLEDGMENTS The authors would like to thank Kristin Reibling for anesthesia support and Dr. Matthew Walker III for his early contributions to this work. REFERENCES Alber, H. F., Frick, M., Sussenbacher, A., Dorler, J., Dichtl, W., Stocker, E. M., Pachinger, O., and Weidinger, F. (2007). Effect of atorvastatin on peripheral endothelial function and systemic inflammatory markers in patients with stable coronary artery disease. Wien Med Wochenschr 157, 73–78. Brooks, A. R., Lelkes, P. I., and Rubanyi, G. M. (2002). Gene expression profiling of human aortic endothelial cells exposed to disturbed flow and steady laminar flow. Physiol Genomics 9, 27–41. Brott, D. A., Richardson, R. J., and Louden, C. S. (2012). Evidence for the nitric oxide pathway as a potential mode of action in fenoldopaminduced vascular injury. Toxicol Pathol 40, 874–86. Chappell, D. C., Varner, S. E., Nerem, R. M., Medford, R. M., and Alexander, R. W. (1998). Oscillatory shear stress stimulates adhesion molecule expression in cultured human endothelium. Circ Res 82, 532–9. Charakida, M., Masi, S., Luscher, T. F., Kastelein, J. J., and Deanfield, J. E. (2010). Assessment of atherosclerosis: The role of flow-mediated dilatation. Eur Heart J 31, 2854–61. Dalmas, D. A., Scicchitano, M. S., Mullins, D., Hughes-Earle, A., Tatsuoka, K., Magid-Slav, M., Frazier, K. S., and Thomas, H. C. (2011). Potential candidate genomic biomarkers of drug induced vascular injury in the rat. Toxicol Appl Pharmacol 257, 284–300. Davies, P. F., Remuzzi, A., Gordon, E. J., Dewey, C. F., Jr., and Gimbrone, M. A. Jr., (1986). Turbulent fluid shear stress induces vascular endothelial cell turnover in vitro. Proc Natl Acad Sci USA 83, 2114–17. Enerson, B. E., Lin, A., Lu, B., Zhao, H., Lawton, M. P., and Floyd, E. (2006). Acute drug-induced vascular injury in beagle dogs: Pathology and correlating genomic expression. Toxicol Pathol 34, 27–32. Filer, A. D., Gardner-Medwin, J. M., Thambyrajah, J., Raza, K., Carruthers, D. M., Stevens, R. J., Liu, L., Lowe, S. E., Townend, J. N., and Bacon, P. A. (2003). Diffuse endothelial dysfunction is common to ANCA associated systemic vasculitis and polyarteritis nodosa. Ann. Rheum Dis 62, 162–67.

791

Garcia-Cardena, G., Comander, J., Anderson, K. R., Blackman, B. R., and Gimbrone, M. A. Jr., (2001). Biomechanical activation of vascular endothelium as a determinant of its functional phenotype. Proc Natl Acad Sci USA 98, 4478–85. Gimbrone, M. A., Jr., Nagel, T., and Topper, J.N. (1997). Biomechanical activation: An emerging paradigm in endothelial adhesion biology. J Clin Invest 100, S61–65. Gutierrez, M. A., Pilon, P. E., Lage, S.G., Kopel, L., Carvalho, R.T., and Furuie, S.S. (2002). Automatic measurement of carotid diameter and wall thickness in ultrasound images. Computers in Cardiology 29, 359–62. Heiss, C., Sievers, R. E., Amabile, N., Momma, T. Y., Chen, Q., Natarajan, S., Yeghiazarians, Y., and Springer, M. L. (2008). In vivo measurement of flow-mediated vasodilation in living rats using high-resolution ultrasound. Am J Physiol Heart Circ Physiol 294, H1086–1093. Isaacs, K. R., Joseph, E. C., and Betton, G. R. (1989). Coronary vascular lesions in dogs treated with phosphodiesterase III inhibitors. Toxicol Pathol 17, 153–63. Jones, H. B., Bjorkman, J. A., and Schofield, J. (2013). Coronary and systemic arterial physiology and immunohistochemical markers related to early coronary arterial lesions in beagle dogs given the potassium channel opener, ZD6169, or the endothelin receptor antagonist, ZD1611. Toxicol Pathol 41, 722–35. Joseph, E. C. (2000). Arterial lesions induced by phosphodiesterase III (PDE III) inhibitors and DA(1) agonists. Toxicol Lett 112–113, 537–46. Joseph, E. C., Rees, J. A., and Dayan, A. D. (1996). Mesenteric arteriopathy in the rat induced by phosphodiesterase III inhibitors: An investigation of morphological, ultrastructural, and hemodynamic changes. Toxicol Pathol 24, 436–50. Kathiresan, S., Gona, P., Larson, M. G., Vita, J. A., Mitchell, G. F., Tofler, G. H., Levy, D., Newton-Cheh, C., Wang, T. J., Benjamin, E. J., and Vasan, R. S. (2006). Cross-sectional relations of multiple biomarkers from distinct biological pathways to brachial artery endothelial function. Circ 113, 938–45. Kerns, W., Schwartz, L., Blanchard, K., Burchiel, S., Essayan, D., Fung, E., Johnson, R., Lawton, M., Louden, C., MacGregor, J., Miller, F., Nagarkatti, P., Robertson, D., Snyder, P., Thomas, H., Wagner, B., Ward, A., and Zhang, J., and Expert Working Group on Drug-Induced Vascular, I. (2005). Drug-induced vascular injury—a quest for biomarkers. Toxicol Appl Pharmacol 203, 62–87. Korkmaz, S., Maupoil, V., Sobry, C., Brunet, C., Chevalier, S., and Freslon, J. L. (2009). An increased regional blood flow precedes mesenteric inflammation in rats treated by a phosphodiesterase 4 inhibitor. Toxicol Sci 107, 298–305. Lappe, R. W., Todt, J. A., and Wendt, R. L. (1986). Effects of fenoldopam on regional vascular resistance in conscious spontaneously hypertensive rats. J Pharmacol Exp Ther 236, 187–91. Louden, C. S., Nambi, P., Pullen, M. A., Thomas, R. A., Tierney, L. A., Solleveld, H. A., and Schwartz, L. W. (2000). Endothelin receptor subtype distribution predisposes coronary arteries to damage. Am J Pathol 157, 123–34. Malek, A. M., Alper, S. L., and Izumo, S. (1999). Hemodynamic shear stress and its role in atherosclerosis. Jama 282, 2035–42. Mesfin, G. M., Piper, R. C., DuCharme, D. W., Carlson, R. G., Humphrey, S. J., and Zins, G. R. (1989). Pathogenesis of cardiovascular alterations in dogs treated with minoxidil. Toxicol Pathol 17, 164–81. Sheth, C. M., Enerson, B. E., Peters, D., Lawton, M. P., and Weaver, J. L. (2011). Effects of modulating in vivo nitric oxide production on the incidence and severity of PDE4 inhibitor-induced vascular injury in Sprague-Dawley rats. Toxicol Sci 122, 7–15.

For reprints and permissions queries, please visit SAGE’s Web site at http://www.sagepub.com/journalsPermissions.nav.

Downloaded from tpx.sagepub.com at DALHOUSIE UNIV on May 17, 2015