Skip to main content
NCBI home page
Toxicological Sciences logoLink to Toxicological Sciences
. 2017 Jul 29;159(2):461–469. doi: 10.1093/toxsci/kfx151

Editor’s Highlight: Pulmonary Vascular Thrombosis in Rats Exposed to Inhaled Sulfur Mustard

Matthew D McGraw *,†, Christopher M Osborne ‡, Emily J Mastej *, Jorge A Di Paola *, Dana R Anderson §, Wesley W Holmes §, Danielle C Paradiso §, Rhonda B Garlick *, Tara B Hendry-Hofer *, Raymond C Rancourt *, Russell W Smith *, Carol Burns *, Gates B Roe *, Jacqueline S Rioux *, Carl W White *,†, Livia A Veress *,†,1
PMCID: PMC5837673  PMID: 28962529

Abstract

Sulfur mustard (SM) is a chemical warfare agent. When inhaled, SM causes significant injury to the respiratory tract. Although the mechanism involved in acute airway injury after SM inhalation has been well described previously, the mechanism of SM’s contribution to distal lung vascular injury is not well understood. We hypothesized that acute inhalation of vaporized SM causes activated systemic coagulation with subsequent pulmonary vascular thrombi formation after SM inhalation exposure. Sprague Dawley rats inhaled SM ethanolic vapor (3.8 mg/kg). Barium/gelatin CT pulmonary angiograms were performed to assess for pulmonary vascular thrombi burden. Lung immunohistochemistry was performed for common procoagulant markers including fibrin(ogen), von Willebrand factor, and CD42d in control and SM-exposed lungs. Additionally, systemic levels of d-dimer and platelet aggregometry after adenosine diphosphate- and thrombin-stimulation were measured in plasma after SM exposure. In SM-exposed lungs, chest CT angiography demonstrated a significant decrease in the distal pulmonary vessel density assessed at 6 h postexposure. Immunohistochemistry also demonstrated increased intravascular fibrin(ogen), vascular von Willebrand factor, and platelet CD42d in the distal pulmonary vessels (<200 µm diameter). Circulating d-dimer levels were significantly increased (p < .001) at 6, 9, and 12 h after SM inhalation versus controls. Platelet aggregation was also increased in both adenosine diphosphate - (p < .01) and thrombin- (p < .001) stimulated platelet-rich plasma after SM inhalation. Significant pulmonary vascular thrombi formation was evident in distal pulmonary arterioles following SM inhalation in rats assessed by CT angiography and immunohistochemistry. Enhanced systemic platelet aggregation and activated systemic coagulation with subsequent thrombi formation likely contributed to pulmonary vessel occlusion.

Keywords: CT angiography, hypercoagulability, inhalation injury, sulfur mustard, thrombosis

Sulfur mustard (bis 2-chloroethyl sulfide [SM]) is a vesicating chemical warfare agent. First introduced during World War I as a significant battlefield agent, SM has caused more casualties than all other chemical warfare agents combined (Ghanei etal., 2008). Despite nearly 100 years of use, SM remains a significant threat used in recent wars and terrorist attacks (Ghanei etal., 2012; Kehe etal., 2009; Wattana and Bey, 2009). It was used extensively in the Iran-Iraq War against both soldiers and civilians. More recently, there has been a resurgence of SM use by terrorist groups (Wattana and Bey, 2009).

SM causes significant injury to the skin, eye, and respiratory tract (Kehe etal., 2009; Rowell etal., 2009). Injury to the respiratory tract is the leading cause of mortality after SM exposure (Newman-Taylor and Morris 1991, Rees etal., 1991). At lower SM concentrations or when SM doses do not cause significant mortality acutely, SM causes acute edema, airway epithelial sloughing, and airway obstruction from cast formation (Newman-Taylor and Morris 1991). The mechanism involved in acute airway obstruction after SM inhalation has been described previously in Veress etal. (2010). This mechanism involves injury to the bronchial circulation adjacent to conducting airways leading to increased vascular permeability. Once these vessels are injured, pro-coagulant plasma fluid leaks out of the bronchial circulation into the airway lumen forming airway casts.

Airway injury after SM exposure has been well described (Gould etal., 2009; Rancourt etal., 2014; Ray etal., 2010), but the mechanism of SM’s contributing to distal parenchymal injury is not as well understood. In long-term survivors exposed to SM, histopathology demonstrate bronchiolitis with and without obliterans, as well as distal lung parenchymal remodeling with pulmonary fibrosis (Emad and Emad, 2007; Ghanei etal., 2010). Thus, SM inhalation causes both proximal and distal lung injury with long-term sequelae.

Pulmonary thrombi formation has been documented previously by Iranian veterans exposed to SM (Balali-Mood etal., 2008), and may contribute to the parenchymal lung sequelae after SM inhalation. In vitro, previous authors have shown direct endothelial injury in co-culture with epithelial cells after SM exposure (Emmler etal., 2007). When exposed, endothelial cells lost cellular adherence, provoking vascular permeability and ultimately inducing apoptosis or necrosis (Emmler etal., 2007). In an effort to repair the injured lung, endothelial cells activate platelets and other pro-coagulant signaling cascades (Pant and Vijayaraghavan, 1999). Additionally, other anti-coagulant factors are decreased after SM analog exposures, tipping the lung into a prothrombotic state (Rancourt etal., 2014).

This study’s purpose was to assess for pulmonary microthrombi formation using CT chest angiography and immunohistochemistry after inhaled SM exposure in the rat. We hypothesized that acute inhalation of vaporized SM causes activated systemic coagulation with subsequent pulmonary arteriolar thrombi formation after SM inhalation exposure.

METHODS

Animals

The Institutional Animal Care and Use Committee of the University of Colorado approved these studies. Adult male (250–300 g) Sprague Dawley, Charles River rats (Charles River Laboratories, Wilmington, Massachusetts) were maintained in an animal care facility fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. The “Guide for the Care and Use of Laboratory Animals” (1996) was followed.

Chemicals

SM (purity confirmed by NMR at 99.4%) was synthesized at the University of Colorado. Due to international treaties pertaining to Schedule 1 chemicals, methods used for SM synthesis and purity ascertainment in Denver are not detailed here. With the exceptions of the barium-gelatin angiography and platelet aggregation studies, all other SM experiments were performed at U.S. Army Medical Research Institute of Chemical Defense (USAMRICD) in Aberdeen, MD. SM for those studies was synthesized at USAMRICD—purity confirmed to be 97.4% by NMR and is a certified standard Chemical Agent Standard Analytical Reference Material.

Inhalation exposure to ethanol and SM vapor in rats

Our methods for exposure of rats to SM vapor were reported previously (Anderson etal., 2000; Veress etal., 2015). For all exposures, 3.8 mg/kg of SM was diluted in ethanol and given as per our previous studies for high dose exposure. Male Sprague Dawley rats weighing 250–300 g were anesthetized with a mixture of ketamine 80 mg/kg and xylazine 10 mg/kg; intubated under direct visualization with a modified glass pipette, and connected to individualized vaporization chambers in a fume hood. For those rats in the control arm, they received vehicle only (100 µl of 100% ethanol). For the exposure arm, SM (3.8 mg/kg in 100% ethanol) was utilized. Each substance was placed into the vaporization chamber, and animals were exposed in the supine position for 50 min. Rats were then extubated and recovered on a heating blanket in their cages.

The LDt50 for SM inhalation exposures in humans has been estimated to be 1500 mg-min/m3 (U.S. Department of the Army, 1974). At a dose 3.8 mg/kg, each rat received a total dose of 0.95 mg. A rat’s volume at this weight is about 0.0003 m3. Therefore, the average mustard exposure is about 1500 mg-min/m3 over 20 min or 750 mg-min/m3 for 10 min, which is similar to LDt50 for SM inhalation exposures in humans.

Euthanasia

Animals were euthanized either at end of study (12 h after exposure), or if there was the concomitant presence of oxygen saturation below 70% and a clinical distress score >7 (maximum, 9). This score was used as previous data demonstrated an animal’s inability to recover from this level of morbidity after SM exposure. Rats were euthanized with a combination of ketamine (75 mg/kg), xylazine (7.5 mg/kg), and acepromazine (1.5 mg/kg), and exsanguinated by aortic puncture. As our previous studies have demonstrated, 100% mortality occurs at 12 h after exposure following inhalation of this high level of SM. Of note, all animals exposed to 100% ethanol were euthanized at 12 h after exposure, and there was no mortality prior to that time in this group.

Thoracotomy, lung fixation, and removal

Rats were anesthetized with ketamine 75 mg/kg, xylazine 7.5 mg/kg, and acepromazine 1.5 mg/kg when they met previously defined euthanasia criteria. The trachea was cannulated, and animals were exsanguinated by puncture of the aorta. Diaphragms were pierced to achieve lung deflation. Lungs were fixed by tracheal infusion of 4% paraformaldehyde in phosphate-buffered saline at 20 cm water pressure for 30 min before excision of whole lungs.

Barium-gelatin pulmonary angiography

CT angiography was performed at 6 h post exposure to assess for vascular thrombi formation. At 6 h, hypoxia developed in all exposed rats. Twenty minutes prior to the time of euthanasia, both naive and SM-exposed rats were given heparin (2000 U/kg) intraperitoneally. Previous experiments comparing the effect of ethanol on lung injury versus naïve controls showed no significant difference in this model (Rancourt etal., 2013, 2014; Veress etal., 2015). Barium angiography was performed as described by Ohar et al. (1998). In brief, after terminal anesthesia, the trachea and pulmonary artery each were cannulated. Phosphate-buffered saline (pH 7.4) containing heparin (1 U/ml) was instilled via the pulmonary artery at 73 mmHg. This was followed by instillation of barium sulfate (30%)/gelatin (4%) mixture at 73 mmHg. Each instillation was continued for 5 min. Thereafter, the main pulmonary artery was tied off, and the lungs were inflation fixed via the trachea with paraformaldehyde (4% in PBS) at 20 cm water. All computed tomography (CT) studies were performed in the University of Colorado Cancer Center/Clinical and Translational Science Award (UCCC/CTSA) Animal Imaging Shared Resources (AISR). Scans were performed on a Siemens Inveon microCT scanner using Inveon Acquisition Workplace software (IAW v1.5). The following parameters were used: total rotation: 270 degrees; rotation steps: 180; field of view (FOV) = 2048 × 2048; binning: 2; with 80 kV and 500 microA; exposure time: 1200 ms; magnification: low-med; effective pixel size: 47.93 µm; total scan time: 10 min. Ultra-fast Siemens postprocessing “COBRA” workstation was used for image reconstruction with a down sample of 1. Multiplanar reconstruction images were produced using Siemens Inveon Research Workplace (IRW v1.5) software.

CT chest angiography quantification

Two-dimensional images were created from projections of each CT chest angiography. All images were binarized to scale from zero to one. No significant vascular attenuation was seen on microCT angiography in the larger pulmonary arteries, definitive for no complete occlusion of a main vessel. Barium remained in the arterial system, and thus, the major veins, venules, or capillaries were unassessed with the current imaging technologies. Large vessels were segmented out of all images to minimize their contribution to the estimation of distal lung vasculature. Similar methods for pulmonary vasculature quantification have been validated in rodents (Badea etal., 2007; Shingrani etal., 2010).

Binary masks of the original image were created through applied thresholding. A threshold value of 0.35 was used to minimize noise and to optimize the desired vessel area (Shingrani etal., 2010). Total vasculature area (pixels2) was calculated for each binary image. All image analysis was performed using Matlab, a commercial software (MATLAB 6.1, The MathWorks Inc., Natick, Massachusetts, 2000).

Histology

Tissue sections were cut from formalin-fixed, parrafin-embedded naive, ethanol-exposed, and SM-exposed lung tissue, and standard IHC protocol was performed. Primary antibodies used in this study included von Willebrand factor (VWF) (1:500, Abcam, Cambridge, Massachusetts) fibrin(ogen) (1:2000, DAKO, Carpinteria, California) and CD42d (1:500, BD Pharmingen, San Jose, California). Color reaction was obtained via antibody-appropriate biotinylated secondary antibodies (Southern Biotech, Birmingham, Alabama) followed by streptavidin-HRP (Thermo Scientific, Waltham, Massachusetts) and then DAB (Vector Laboratories, Burlingame, California). Images were obtained using an Axioskop 2 (Carl Zeiss, Germany).

Histologic quantification

Tissue sections of naïve and SM exposed animals stained for fibrin(ogen) were quantified for clot burden in the distal pulmonary vessels. All quantification was done blinded at 10× magnification. Quantification was performed on 6 randomly selected distal lung sections without conducting airways in the FOV. The average percent fibrin(ogen) positive vessels per total number of assessed vessels (<200 µm in diameter) was calculated for each animal (Yun etal., 2016). A minimum of 50 vessels were assessed per animal, and then the average percent positive vessels for fibrin(ogen) per total number of assessed vessels was calculated.

d-dimer

Enzyme-linked immunosorbent assay (ELISA) was performed on circulating blood plasma from naive and SM-exposed rats (Diagnostica Stago, Parsippany, New Jersey). Sixteen samples were tested: 4 naive and 4 animals each that met euthanasia criteria at 6, 9, and 12 h postexposure.

Thrombin-antithrombin complex levels

ELISA was performed on circulating citrated plasma from naive and SM-exposed rats (Enzygnost microassay; Siemens; Tarrytown, New York). Sixteen samples were tested: 4 naive and 4 animals at 6 and 12 h postSM exposure.

Platelet aggregometry

Light transmission aggregometry of naive, ethanol-exposed, and SM-exposed platelets was performed using a Chrono-Log Corp Model 700 Aggregometer (Chrono-Log Corp, Havertown, Pennsylvania) and Aggrolink 8, Version 1.1.0 software (Chrono-Log ). 10% sodium citrated plasma was centrifuged, at room temperature, for 10 min at 100g. Supernatant was collected to use as platelet rich plasma (PRP). The remaining blood was centrifuged, at room temperature, for 10 min at 3000g. The resulting supernatant was collected to use as platelet poor plasma (PPP). PRP platelet counts were obtained using a Hemavet (Drew Scientific, Dallas, Texas) and the samples were diluted with corresponding PPP to make a final platelet concentration of 250 000/µl. Thrombin (Chrono-Log cat no. 386), at a final concentration of 2 U/ml, and adenosine diphosphate (ADP) (Chrono-Log cat no. 387), at a final volume of 50 µl per sample, were used as aggregation-inducing agents. After transfer of the adjusted PRP in cuvettes and addition of the aggregation-inducing agents, recorded light transmission was under permanent agitation of the samples. Results were obtained as a percent of the maximal aggregation.

Statistics

Statistics were performed using Prism 6.01 software (GraphPad, La Jolla, California). A non-parametric Mann-Whitney test was used to compare the median surface area of the CT angiography images between naïve controls and SM exposed animals. The interquartile range (IQR) was expressed for each group. One-way ANOVA followed by Tukey’s post hoc analysis was used for comparison of d-dimer assay, thrombin-antithrombin (TAT) complex assay and platelet aggregation assay results. Mean and standard deviation of the mean was reported. A p-value < .05 was considered to be statistically significant.

RESULTS

CT Chest Angiography

CT chest angiography was utilized to demonstrate the distribution and burden of thrombi formation in the lung after SM exposure. A 3D reconstruction of a naïve and SM-exposed rat lung can be viewed on the online supplement (Supplementary Videos E1 and E2, respectively). In naïve controls, barium injected into the pulmonary artery demonstrated a normal vascular branching pattern (Figure 1A). At 6-h postSM exposure, the pulmonary arterioles of the SM-exposed rat lungs were attenuated compared with naive controls (Figure 1B). This process appeared diffuse, particularly involving the distal pulmonary arterioles.

Figure 1.

Figure 1

Open in a new tab

Whole lung barium-gelatin perfused CT angiograms of naïve (A) and SM exposed rats (B). (A) demonstrates the homogeneous branching pattern of the distal lung in an naive rat. In contrast, (B) represents significant pulmonary vasculature attenuation of the rat’s pulmonary vascular 6 h after SM exposure.

CT Chest Angiography Quantification

To quantify the decrease in number of affected pulmonary arteriole branches after acute SM inhalation, each CT chest angiography was converted to a grayscale image and total vascular area was calculated. The median vascular area of naïve, non-exposed rat lungs was 4.576 × 105 pixels2 (Figure 2A; IQR: 4.070–4.835 × 105; n = 4). Comparatively, the median vascular area of SM-exposed rat lungs was 2.452 × 105 pixels2 (IQR: 1.414–3.256 × 105; n = 4). Thus, the median intravascular area of the distal pulmonary arterioles was significantly reduced after SM exposure by CT chest angiography (Figure 2B;p < .05).

Figure 2.

Figure 2

Open in a new tab

Quantification of pulmonary vasculature area from chest CT angiography images of naive and SM-exposed rat lungs. The median pulmonary vascular area of naïve rat lungs was 4.576 × 105 pixels2 (IQR: 4.070–4.835 × 105; n = 4). The median vascular area of SM-exposed rat lungs was 2.452 × 105 pixels2 (IQR: 1.414–3.256 × 105; n = 4). The median intravascular area was significantly reduced after SM exposure by CT chest angiography (*p < .05).

Histopathology and Special Stains

Since the number of distal branching arterioles was significant reduced after SM exposure by CT angiography, immunohistochemistry was utilized to detect markers of coagulation initiation in naive, ethanol-exposed and SM-exposed tisues, suggestive of clot formation. These markers included fibrin(ogen), vWF, and CD42d. When evaluating for fibrin(ogen) in the pulmonary vasculature of SM-exposed rats, a number of small pulmonary arterioles stained intensely for fibrin(ogen) (Figs. 3C and D, black arrows), which was absent in naive (Figure 3A, red arrow) and IgG controls (Figure 3B). The intensity of fibrin(ogen) staining was patchy and nonhomogeneous as denoted by similar-sized vessels with intense fibrin(ogen) staining (black arrow; Figs. 3C and D) and without intense fibrin(ogen) staining in most histologic sections (red arrows; Figs. 3C and D). The most commonly affected pulmonary vessels were those < 200 µm in cross-sectional diameter.

Figure 3.

Figure 3

Open in a new tab

Immunohistochemistry of naive (A), IgG control (B), and SM exposed (C, lower magnification; D, higher magnification) of rat lung with fibrin(ogen) staining. Naive control lung sections demonstrate a lack of intense fibrinogen staining in the small pulmonary vessels (red arrow, A). Similar staining was absent in IgG control (B). In SM-exposed rat lungs, staining for fibrin(ogen) was intense in certain small pulmonary arterioles (black arrow, C). Intense staining for fibrin(ogen) was patchy and heterogeneous as seen by adjacent vessels with lower intensity staining of fibrinogen (red arrow, C). Patchy areas of intense fibrin(ogen) staining was more prominent at higher magnification (size bar: 50 µm, D).

As fibrin(ogen) alone does not constitute a vascular thrombus, we also evaluated for the presence of vWF, a glycoprotein primarily synthesized and stored within endothelial cells involved with hemostasis, and CD42d, a soluble fragment released from platelets or megakaryocytes after thrombin cleavage that interacts with the vWF receptor on endothelial cells. In small pulmonary vessels where the intensity of fibrin(ogen) staining was increased (Figure 4C), both vWF (Figure 4D) and CD42d (Figure 4B) staining were also increased compared with IgG controls (Figure 4A). vWF staining along the vessel wall or in deposits bound to endothelial cells as well as CD42d of the platelets were in close proximity to fibrin(ogen) (white arrows; Figs. 4D and B, respectively), suggestive of endothelial cell injury with subsequent thrombosis formation.

Figure 4.

Figure 4

Open in a new tab

Immunohistochemistry of naive (A) and SM exposed rat lung displaying increased signaling of thrombi surface markers, vWF (B), fibrin(ogen) (C), and platelet marker CD42d (D) within a small pulmonary arteriole 10 h after SM exposure. vWF staining was increased in the injured pulmonary vessel after SM exposure (white arrow; B). An image taken of the same pulmonary arteriole after SM injury showed increased signaling of fibrin(ogen) at the site of vWF staining in the adjacent endothelium (white arrow; C). Platelet staining for CD42d (white arrow, D) was also increased at the location of vWF and fibrin(ogen).

Histology Quantification

To quantify the clot burden of the distal pulmonary arterioles, tissue sections of naïve and SM exposed animals were stained for fibrin(ogen). The average percent fibrin(ogen) positive arterioles per total number of assessed arterioles was calculated for each animal (Figure 5). In naïve rats, the average percent fibrin(ogen) positive vessels was 18% (IQR: 9%–22%, n = 5). In SM-exposed rats, the average percent fibrin(ogen) positive vessels was 42% (IQR: 32%–58%, n = 5). Thus, the average percent fibrin(ogen) positive vessels was significantly increased in SM exposed animals compared with naïve controls (p < .01).

Figure 5.

Figure 5

Open in a new tab

Average percent fibrin(ogen) positive pulmonary vessel (<200 µm diameter) per total number of assessed arterioles for naïve and SM exposed animals. In naïve rats, the average percent fibrin(ogen) positive vessels was 18% (IQR: 9–22%, n = 5). In SM exposed rats, the average percent fibrin(ogen) positive vessels was significant increased at 42% (IQR: 32%–58%, n = 5) compared with naïve controls (p < .01).

Plasma d-Dimer Levels

Since both CT angiography and immunohistochemistry were suggestive of thrombus formation in the distal pulmonary vasculature, we measured plasma d-dimer, a small protein fibrin degradation marker, at 6, 9, and 12 h after SM exposure as a surrogate marker of activated systemic coagulation. In naïve controls, plasma d-dimer levels were 53.8 ± 3.8 ng/ml (Figure 6A). At 6 h, plasma d-dimer levels were significantly elevated compared with naïve controls (169.3 ± 9.4 ng/ml; p < .0001) and highest among the tested time points. At both 9 and 12 h after SM exposure, plasma d-dimer levels were decreased from 6 h but remained significantly elevated compared with naïve controls (128.4 ± 6.4 and 111.2 ± 6.8 ng/ml, respectively; p < .0001). Thus, elevated d-dimer was present in circulating plasma when CT angiography and immunohistochemistry demonstrated distal vascular attenuation with thrombosis formation.

Figure 6.

Figure 6

Open in a new tab

Effect of SM inhalation on plasma d-dimer levels (A) and plasma TAT complex levels (B) in rats. A, Plasma d-dimer levels were significantly increased at 6, 9, and 12 h after SM exposure compared with the naive controls (p < .001; n = 4 per group). B, Plasma TAT complex levels were significantly increased at 12 h after SM exposure (p < .05; n = 4 per group).

TAT Complex Levels

To further validate the activation of systemic coagulation, we measured TAT complex levels, which are known to be sensitive marker of activated intravascular coagulation. In naïve control rats, the average plasma TAT complex level was 2.5 ± 2.8 µg/l. At 6 h after SM exposure, average plasma TAT levels were 6.7 ± 2.5 µg/l. At 12 h after SM exposure, average plasma TAT levels were significantly increased at 12.91 ±7.4 µg/l (n = 4 for each group; p < .05).

Platelet Aggregation

To further assess for systemic markers of hypercoagulability, we measured platelet aggregation after ADP (Figure 7A) and thrombin (Figure 7B) stimulation, 2 potent agonists for platelet aggregation. The maximal light transmission was measured for each naïve, ethanol-exposed, and SM-exposed plasma sample using platelet aggregometry. With the addition of APD, the maximal light transmission did not differ significantly between naïve and ethanol-exposed controls (p > .05). In contrast, the maximal light transmission of rats exposed to SM was significantly increased (44.3% ± 2.6%; p < .01) compared with both naïve and ethanol-exposed controls (30.5% ± 0.3% and 33.3% ± 0.4%).

Figure 7.

Figure 7

Open in a new tab

Effect of SM inhalation on ADP-mediated (A) and thrombin-mediated (B) platelet aggregation 1 h after SM exposure. ADP-mediated platelet aggregation was significantly increased in the SM-exposed compared with ethanol and naïve controls (p < .01; n = 4 per group, A). Thrombin-mediated platelet aggregation was also significantly increased in the SM-exposed compared with the ethanol and naïve controls (p < .001; n = 4 per group, B).

Following the addition of thrombin, there was also a significant increase in maximal light transmission of rats exposed to SM (82.3% ± 3.1%, p < .0001) as compared with naive and ethanol-exposed controls (56.5% ± 1.4% and 58.8% ± 1.9%, respectively). Thus, circulating platelet activation was increased in rats exposed to SM compared with naïve and ethanol-exposed.

DISCUSSION

Within hours of SM inhalation injury, rats developed intravascular thrombi in the distal pulmonary vasculature. This manuscript is the first paper to demonstrate microvascular clot formation in the distal pulmonary vasculature after SM inhalation, supported both radiologicalliy with CT angiography and histologically with fibrin(ogen), vWF, and CD42d staining. Pulmonary thrombi formation developed within hours of exposure in the setting of activated intravascular coagulation measured by an acute rise in plasma d-dimer levels and platelet activity levels followed by an incremental rise in plasma TAT complex levels postSM exposure.

In previous work with an SM analog, plasma TAT complex levels were increased while mean clotting time in plasma was decreased within hours of SM exposure. Both of these findings support systemic activation of intravascular coagulation (Rancourt etal., 2013). In bronchoalveolar lavage fluid, TAT complex levels were increased in the setting of fibrin-laden bronchial cast formation visualized by microdissection. Within 4 h of SM analog exposure, bronchial cast formation was present and was 100% fatal at 12 h. Similar lethality from bronchial cast formation has been demonstrated in the rat model used here after SM (Veress etal., 2015). Other authors have shown similar airway injury after SM inhalation in intubated rodents (Gao etal., 2011; McClintock etal., 2006; Weber etal., 2010). The current manuscript adds to this previously published work demonstrating that microthrombi formation developed in the distal pulmonary vasculature after systemic activation of the coagulation cascade with elevated plasma d-dimer and TAT complex levels within hours of SM exposure.

One potential therapy to ameliorate systemic thrombi formation is tissue plasminogen activator (tPA). tPA is a fibrinolytic agent that not only prevents further clot formation but also actively breaks down previously formed clots through the activation of plasmin. When given intratracheally after SM analog exposure in a rat model, acute mortality was significantly improved (Veress etal., 2015). Intratracheal delivery of tPA was chosen over systemic administration in these prior experiments as the main cause of mortality after SM exposure is bronchial cast formation. Systemic administration of tPA after SM has yet to be performed, but the manuscript’s authors are hesitant to advocate for systemic administration of fibrinolytics considering that most SM exposures are during combat when the risk of traumatic brain injury is high, and systemic administration of a fibrinolytic such as tPA would significantly increase the risk of intracranial hemorrhage.

Chest CT with angiography is the gold standard for assessment of a pulmonary embolism in a hemodynamically unstable patient (Bankier etal., 1998; Kennedy etal., 2015). Until recently, the use of CT angiography in small animal models has been limited by technical challenges, including low image quality, imaging speed and low vessel contrast compared with the surrounding tissues (Hu etal., 2016; Ott etal., 2016). To our knowledge, this is the first study to use micro-CT angiography to assess for vascular changes after acute SM inhalation exposure. CT angiography was performed at 6-h postSM exposure as all exposed animals develop hypoxia at this time point. At the exposed dose of 3.8 mg/kg, SM is 100% fatal at 12 h due to bronchial cast formation (Veress etal., 2015). Thus, assessment for intravascular thrombi burden at or later than 12 h was impossible at this dose. One limitation of this study was that CT angiography was not performed at sequential time points after hypoxia developed. Future studies assessing intravascular thrombi burden before or after 6 h using CT angiography may be helpful in determining optimal timing for therapeutic targeting of intravascular coagulation abnormalities.

Thrombosis is not the only cause for perfusion attenuation with CT angiography. Classically, pulmonary vascular remodeling due to pulmonary hypertension and/or impaired lung vascular development is a known cause of pulmonary vascular filling restriction after barium infusion of the pulmonary artery (Latus etal., 2016; Rajaram etal., 2015; Siegel and Mirpuri, 2014). However, these are chronic disorders that develop over weeks to months, rather than minutes to hours. Certainly, acute pulmonary hypertension can occur with pulmonary vascular thrombosis or embolism in which elevated pulmonary arterial pressures occur almost instantaneously. Similar filling defects would be anticipated in this situation. However, pulmonary hypertension associated with acute respiratory distress is less associated with chronic pulmonary vascular remodeling.

To validate the changes seen on CT angiography, histologic staining of rat lung sections assessing the distal pulmonary vascular for fibrin(ogen), vWF, and CD42d, all pro-thrombic proteins, was performed. When compared with naïve controls, the average percent fibrin(ogen) positive arterioles was significantly increased in SM exposed animals (p < .01) As fibrin(ogen) alone does not validate the presence of a fibrin clot, we also performed staining for vWF, a glycoprotein primarily synthesized within endothelial cells and an important component of hemostasis, and CD42d, a surface marker on platelets possessing the complex role of thrombosis initiation. In histologic lung sections with increased intravascular staining for fibrin(ogen), adjacent vessels demonstrated increased vWF and CD42d staining. Collectively, these results support that animals exposed to SM developed intravascular thrombi within hours of SM inhalation exposure.

The development of pulmonary vascular thromboses after SM inhalation exposure is a novel finding in animal models with potentially significant contribution to, both acute and chronic, morbidity and mortality associated with SM exposure. Concomitant obstruction of both central airways by bronchial cast formation and pulmonary vasculature by thrombi will diminish ventilation/perfusion matching and contribute to worsening hypoxia and gas exchange previously noted after SM inhalation. Further work is needed to understand the exact mechanism of how SM contributes to activated intravascular coagulation with subsequent thrombi formation as this hypercoagulable state will likely contribute to both acute and chronic morbidity associated with SM inhalation injury.

SUPPLEMENTARY DATA

Supplementary data are available at Toxicological Sciences online.

Supplementary Material

Supplementary Video 1
Click here for additional data file. (2.3MB, mov)
Supplementary Video 2
Click here for additional data file. (2.3MB, mov)

ACKNOWLEDGMENTS

The authors thank the following people for their contributions to our study: Steven Abman, MD, from the Department of Pediatrics, Pulmonary Medicine, at the University of Colorado for his expert advice and guidance regarding the development of pulmonary hypertension and methodologies in barium perfused CT angiograms. They also thank their External Advisory Committee members from the Countermeasures against Chemical Threats Research Network, for their support and advice. Last, the authors would like to thank Natalie Serkova PhD and Kendra Huber of the UCCC/CTSA, Animal Imaging Shared Resources (AISR) on grants UCCC P30 and CCTSI U01 for their expert guidance and assistance with the CT barium perfusion angiograms. The authors are also grateful to Joan Loader for additional technical assistance.

FUNDING

CounterACT Program, National Institutes of Health (NIH), Office of the Director, and the National Institute of Environmental Health Sciences (NIEHS) (1U54 ES027698-01 to C.W.W. and L.A.V.; 5U54 ES015678-09 to C.W.W.). NIH (T32HL 007670-27 To M.D.M.).

REFERENCES

  1. Anderson D. R., Byers S. L., Vesely K. R. (2000). Treatment of sulfur mustard (HD)-induced lung injury. J. Appl. Toxicol. 20(Suppl. 1), S129–S132. [DOI] [PubMed] [Google Scholar]
  2. Badea C. T., Hedlund L. W., De Lin M., Mackel J. S., Samei E., Johnson G. A. (2007). Tomographic digital subtraction angiography for lung perfusion estimation in rodents. Med. Phys. 34, 1546–1555. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Balali-Mood M., Mousavi S., Balali-Mood B. (2008). Chronic health effects of sulphur mustard exposure with special reference to Iranian veterans. Emerg. Health Threats J. 1, e7.. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bankier A., Herold C. J., Fleischmann D., Janata-Schwatczek K. (1998). [Spiral CT angiography in diagnosis of acute pulmonary embolism. What factors modify implementation of standard algorithms?]. Radiologe 38, 248–255. [DOI] [PubMed] [Google Scholar]
  5. Emad A., Emad Y. (2007). Levels of cytokine in bronchoalveolar lavage (BAL) fluid in patients with pulmonary fibrosis due to sulfur mustard gas inhalation. J. Interferon Cytokine. Res. 27, 38–43. [DOI] [PubMed] [Google Scholar]
  6. Emmler J., Hermanns M. I., Steinritz D., Kreppel H., Kirkpatrick C. J., Bloch W., Szinicz L., Kehe K. (2007). Assessment of alterations in barrier functionality and induction of proinflammatory and cytotoxic effects after sulfur mustard exposure of an invitro coculture model of the human alveolo-capillary barrier. Inhal. Toxicol. 19, 657–665. [DOI] [PubMed] [Google Scholar]
  7. Gao X., Anderson D. R., Brown A. W., Lin H., Amnuaysirikul J., Chua A. L., Holmes W. W., Ray P. (2011). Pathological studies on the protective effect of a macrolide antibiotic, roxithromycin, against sulfur mustard inhalation toxicity in a rat model. Toxicol. Pathol. 39, 1056–1064. [DOI] [PubMed] [Google Scholar]
  8. Ghanei M., Ghayumi M., Ahakzani N., Rezvani O., Jafari M., Ani A., Aslani J. (2010). Noninvasive diagnosis of bronchiolitis obliterans due to sulfur mustard exposure: Could high-resolution computed tomography give us a clue?. Radiol Med 115, 413–420. [DOI] [PubMed] [Google Scholar]
  9. Ghanei M., Harandi A. A., Tazelaar H. D. (2012). Isolated bronchiolitis obliterans: High incidence and diagnosis following terrorist attacks. Inhal. Toxicol. 24, 340–341. [DOI] [PubMed] [Google Scholar]
  10. Ghanei M., Tazelaar H. D., Chilosi M., Harandi A. A., Peyman M., Akbari H. M., Shamsaei H., Bahadori M., Aslani J., Mohammadi A. (2008). An international collaborative pathologic study of surgical lung biopsies from mustard gas-exposed patients. Respir. Med. 102, 825–830. [DOI] [PubMed] [Google Scholar]
  11. Gould N. S., White C. W., Day B. J. (2009). A role for mitochondrial oxidative stress in sulfur mustard analog 2-chloroethyl ethyl sulfide-induced lung cell injury and antioxidant protection. J. Pharmacol. Exp. Ther. 328, 732–739. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hu M. Q., Li M., Liu Z. Y., Huang M. P., Liu H., Liang C. H. (2016). Image quality evaluation of iterative model reconstruction on low tube voltage (80 kVp) coronary CT angiography in an animal study. Acta Radiol. 57, 170–177. [DOI] [PubMed] [Google Scholar]
  13. Kehe K., Thiermann H., Balszuweit F., Eyer F., Steinritz D., Zilker T. (2009). Acute effects of sulfur mustard injury-Munich experiences. Toxicology [DOI] [PubMed] [Google Scholar]
  14. Kennedy N., Jayathissa S., Healy P. (2015). Investigation of suspected pulmonary embolism at hutt valley hospital with CT pulmonary angiography: Current practice and opportunities for improvement. Adv. Med. 2015, 357576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Latus H., Kuehne T., Beerbaum P., Apitz C., Hansmann G., Muthurangu V., Moledina S. (2016). Cardiac MR and CT imaging in children with suspected or confirmed pulmonary hypertension/pulmonary hypertensive vascular disease. Expert consensus statement on the diagnosis and treatment of paediatric pulmonary hypertension. The European Paediatric Pulmonary Vascular Disease Network, endorsed by ISHLT and DGPK. Heart 102(Suppl. 2), ii30–ii35. [DOI] [PubMed] [Google Scholar]
  16. McClintock S. D., Hoesel L. M., Das S. K., Till G. O., Neff T., Kunkel R. G., Smith M. G., Ward P. A. (2006). Attenuation of half sulfur mustard gas-induced acute lung injury in rats. J. Appl. Toxicol. 26, 126–131. [DOI] [PubMed] [Google Scholar]
  17. Newman-Taylor A. J., Morris A. J. (1991). Experience with mustard gas casualties. Lancet 337, 242.. [DOI] [PubMed] [Google Scholar]
  18. Ohar J. A., Waller K. S., Williams T. J., Luke D. A., Demello D. E. (1998). Computerized morphometry of the pulmonary vasculature over a range of intravascular pressures. Anat Rec. 2521, 92–101. [DOI] [PubMed] [Google Scholar]
  19. Ott S., Struffert T., Saake M., Golitz P., Adamek E., Doerfler A. (2016). Influence of different reconstruction parameters in the visualization of intracranial stents using C-arm flat panel CT angiography: Experience in an animal model. Acta Radiol. 57, 233–240. [DOI] [PubMed] [Google Scholar]
  20. Pant S. C., Vijayaraghavan R. (1999). Histomorphological and histochemical alterations following short-term inhalation exposure to sulfur mustard on visceral organs of mice. Biomed. Environ. Sci. 12, 201–213. [PubMed] [Google Scholar]
  21. Rajaram S., Swift A. J., Wild J. M., Kiely D. G. (2015). Response to: ‘CT assessment for pulmonary hypertension requires systematic assessment of cardiac, vascular and parenchymal signs’ by Marloes etal., Thorax 70, 1087–1088. [DOI] [PubMed] [Google Scholar]
  22. Rancourt R. C., Ahmad A., Veress L. A., Rioux J. S., Garlick R. B., White C. W. (2014). Antifibrinolytic mechanisms in acute airway injury after sulfur mustard analog inhalation. Am. J. Respir. Cell Mol. Biol. 51, 559–567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Rancourt R. C., Veress L. A., Ahmad A., Hendry-Hofer T. B., Rioux J. S., Garlick R. B., White C. W. (2013). Tissue factor pathway inhibitor prevents airway obstruction, respiratory failure and death due to sulfur mustard analog inhalation. Toxicol. Appl. Pharmacol. 272, 86–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Ray R., Simbulan-Rosenthal C. M., Keyser B. M., Benton B., Anderson D., Holmes W., Trabosh V. A., Daher A., Rosenthal D. S. (2010). Sulfur mustard induces apoptosis in lung epithelial cells via a caspase amplification loop. Toxicology 271, 94–99. [DOI] [PubMed] [Google Scholar]
  25. Rees J., Harper P., Ellis F., Mitchell D. (1991). Mustard gas casualties. Lancet 337, 430.. [DOI] [PubMed] [Google Scholar]
  26. Rowell M., Kehe K., Balszuweit F., Thiermann H. (2009). The chronic effects of sulfur mustard exposure. Toxicology [DOI] [PubMed] [Google Scholar]
  27. Shingrani R., Krenz G., Molthen R. (2010). Automation process for morphometric analysis of volumetric CT data from pulmonary vasculature in rats. Comput. Methods Prog. Biomed. 97, 62–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Siegel Y., Mirpuri T. (2014). Pulmonary hypertension detection using dynamic and static measurable parameters on CT angiography. J. Comput. Assist. Tomogr. 38, 586–590. [DOI] [PubMed] [Google Scholar]
  29. U.S. Department of the Army. (1974). Chemical agent data sheets. Defense Technical Information Center, Alexandria, VA., U.S. Department of the Army. 1. [Google Scholar]
  30. Veress L. A., Anderson D. R., Hendry-Hofer T. B., Houin P. R., Rioux J. S., Garlick R. B., Loader J. E., Paradiso D. C., Smith R. W., Rancourt R. C., Holmes W. W., White C. W. (2015). Airway tissue plasminogen activator prevents acute mortality due to lethal sulfur mustard inhalation. Toxicol. Sci. 143, 178–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Veress L. A., O’Neill H. C., Hendry-Hofer T. B., Loader J. E., Rancourt R. C., White C. W. (2010). Airway obstruction due to bronchial vascular injury after sulfur mustard analog inhalation. Am. J. Respir. Crit. Care Med. 182, 1352–1361. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Wattana M., Bey T. (2009). Mustard gas or sulfur mustard: An old chemical agent as a new terrorist threat. Prehosp. Disaster Med. 24, 19–29. discussion 30-11. [DOI] [PubMed] [Google Scholar]
  33. Weber W. M., Kracko D. A., Lehman M. R., Irvin C. M., Blair L. F., White R. K., Benson J. M., Grotendorst G. R., Cheng Y. S., McDonald J. D. (2010). Inhalation exposure systems for the development of rodent models of sulfur mustard-induced pulmonary injury. Toxicol. Mech. Methods 20, 14–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Yun E. J., Lorizio W., Seedorf G., Abman S. H., Vu T. H. (2016). VEGF and endothelium-derived retinoic acid regulate lung vascular and alveolar development. Am. J. Physiol. Lung Cell Mol. Physiol. 310, L287–L298. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Video 1
Click here for additional data file. (2.3MB, mov)
Supplementary Video 2
Click here for additional data file. (2.3MB, mov)

Articles from Toxicological Sciences are provided here courtesy of Oxford University Press

RESOURCES