**1. Introduction**

Feline aortic thromboembolism (ATE) is a severe complication that can occur in 11.6–21% of cats with myocardial disease [1–6]. It happens when a thrombus usually

**Citation:** Cheng, W.-C.; Wilkie, L.; Kurosawa, T.A.; Dobromylskyj, M.; Priestnall, S.L.; Luis Fuentes, V.; Connolly, D.J. Immunohistological Evaluation of Von Willebrand Factor in the Left Atrial Endocardium and Atrial Thrombi from Cats with Cardiomyopathy. *Animals* **2021**, *11*, 1240. https://doi.org/10.3390/ ani11051240

Academic Editor: Lysimachos Papazoglou

Received: 22 March 2021 Accepted: 19 April 2021 Published: 26 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

originating in the left atrium (LA) or left atrial appendage (LAA) dislodges and obstructs a branching artery of the aorta. Common clinical signs frequently relate to the pelvic limbs and include severe pain, paraparesis, paraplegia, absence or reduced strength of femoral pulses, and cyanotic paws. ATE in cats carries a poor prognosis, with over 60% of affected cats being euthanised on presentation at first opinion practices [7].

Virchow's triad describes three factors that contribute to venous thrombosis: blood stasis, endothelial injury, and hypercoagulability [8]. Abnormal findings associated with Virchow's triad have been reported in cats with a variety of cardiomyopathies, which provides insights into the potential pathogenesis of feline ATE. For instance, development of spontaneous echo contrast (SEC) in the LA of cats with cardiomyopathy is associated with decreased LAA blood velocity and blood stasis [9]. A hypercoagulable state has been suggested in cats with HCM [10] and histopathologic evidence of LA endothelial damage has been observed in cats with CHF [11]. A further indication of endothelial dysfunction in cats with ATE is that plasma arginine, the precursor to nitric oxide critical for endothelial health, is lower in these cats than in than cats with cardiomyopathy alone [12]. However, the contribution of endothelial injury in the left atrial endocardium to the development of ATE has not been studied at the sub-cellular level in cats with cardiomyopathy at different stages of their disease process and only to a limited extent in humans with heart disease.

vWF is synthesised in endothelial cells and megakaryocytes and is an adhesive protein that plays an important role in thrombosis through the formation of multimers [13,14]. Endothelial vWF is either secreted into the plasma constitutively, stored in rod-shaped specialised compartments called Weibel-Palade bodies within the endothelial cells, or deposited in the subendothelium [15]. vWF of platelet origin is stored in alpha-granules within the platelet and is released upon activation. The main functions of vWF are to aid binding of platelets to exposed subendothelium and to assist in platelet aggregation via the glycoprotein Ib receptor on platelets [16,17]. The majority of plasma vWF originates from the endothelium, and in human patients the plasma concentration of vWF increases in various thrombogenic diseases reflecting endothelial injury [18]. Similarly, in cats with ATE and cardiomyopathy, the elevation in circulatory vWF has been suggested to be associated with endothelial damage [10].

In human patients with atrial distension due to a variety of cardiac conditions, expression of circulating and LA endocardial vWF protein is increased and associated with the severity of LA blood stasis and atrial remodelling and has been suggested as a predisposing factor for thrombogenesis [19–23].

Previous studies have also confirmed the pathogenic role of vWF in both arterial and venous thrombosis. High haemodynamic forces present in the systemic arterial system result in platelets binding to vWF through integrin αIIbβ3 for the initiation of aggregation [13,24], while patients with vWF deficiency as a result of von Willebrand disease are partly protected against arterial thrombosis [25]. vWF also contributes to thrombosis in veins, through formation of an extracellular scaffold with platelets, leukocytes, and fibrin to trap erythrocytes [26]. Mice deficient in vWF were protected from induced deep vein thrombosis [26]. Although fibrin is the best studied factor involved in trapping red blood cells (RBCs) in slow blood velocity environments [27–30], a recent in vitro study showed that vWF can bind to RBCs directly and the degree of binding is increased when blood stasis is present [31].

Currently, there is limited information regarding the contribution of endocardial vWF to thrombosis in cats with myocardial disease. However, given the pro-thrombotic action of vWF in both rodent models and human patients described above, it is likely that vWF also plays an important role in the development of LA thrombosis in cats. The aim of this study was to investigate the expression of vWF in the LA of cats at different clinical stages of myocardial disease.

#### **2. Materials and Methods**

#### *2.1. Study Population*

Cats with clinical signs related to cardiomyopathy were enrolled based on their clinical presentation of ATE, CHF or both. The control cats and those with preclinical cardiomyopathy (without clinical signs of disease) were recruited from referral and first opinion cases euthanised for a variety of reasons. Cardiomyopathy was confirmed based on characteristic cardiac structural changes on gross and histopathology using criteria previously described [32,33] (see supplementary material Table S1 for histopathological diagnostic criteria for feline cardiomyopathies) in addition to complete or point of care (POC) echocardiographic examination [34–36] (see Table S2 for echocardiographic diagnostic criteria for feline cardiomyopathies). Briefly for full echocardiographic examinations, all measurements were taken over three different cardiac cycles and averaged. Measurements taken included left atrium to aortic ratio (LA/Ao), maximal left ventricular freewall thickness in diastole (LVFWd), maximal interventricular septum thickness in diastole (IVSd), presence of systolic anterior motion of the mitral valve (SAM), presence of spontaneous echo contrast (SEC), and presence of a formed thrombus in the LA. SAM was defined as anterior motion of either septal or both mitral valve leaflets during systole toward the LVOT using the right parasternal long axis five chamber view on review of 2D cineloops [37]. All echocardiographic examinations were performed by a veterinary cardiology diplomate or resident under direct supervision (see supplementary material Table S5 for echocardiographic views used [38,39]). POC examinations were performed in the emergency setting by a veterinary ECC diplomate or resident in training and facilitated measurement of LA/Ao as described above and a subjective assessment of left ventricular wall thickness in diastole.

Cats were divided into four groups: (1) Control group: cats without structural and histopathological cardiac changes, (2) Subclinical group: cats with subclinical cardiomyopathy (some cats had mild LAE), (3) CHF group: cats with CHF attributable to cardiomyopathy with LAE, (4) ATE group: cats with ATE attributable to cardiomyopathy with LAE, irrespective of whether presenting with concurrent CHF.

Inclusion and exclusion criteria are listed in Table 1.


**Table 1.** Inclusion and exclusion criteria for case selection of cats by pathological examination.

> Diagnosis of ATE was based on clinical signs including acute fore/hind limb(s) paresis or plegia, loss of palpable femoral pulses, cold limbs, and cyanotic nail beds in association with myocardial disease identified by cardiac imaging [40]. CHF was diagnosed based on radiographic evidence of cardiogenic pulmonary oedema, ultrasonographical evidence of pleural effusion in association with cardiomyopathy, and left or bi-atrial enlargement. Cats

with cardiomyopathy identified by cardiac imaging and gross and histopathology but showing no clinical signs of heart disease were determined to have subclinical cardiomyopathy.

#### *2.2. Sample Collection*

The heart was harvested within 30 min of euthanasia and flushed with slowly running tap water and each chamber was further gently flushed using a 20 mL syringe to ensure all blood was removed to facilitate optimal fixation in 10% buffered formalin. The LA samples were collected from the LA free wall. Where a formed thrombus in the LA was identified, it was carefully removed and placed into 10% buffered formalin, while an in situ thrombus in the LAA was harvested without being removed from the LAA. After gross pathological examination, sections of the left atrial free wall and cross-sections of the ventricles at the heart base, midwall, and apex were embedded in paraffin wax in a routine manner for preparation of slides for haematoxylin and eosin and Masson's trichrome staining for histopathological examination. The LA sample blocks were used to prepare slides for immunostaining.

## *2.3. Fluorescent Immunostaining*

After dewaxing, rehydration, antigen retrieval with citric acid (pH6) at 95 ◦C for 10 min, and blocking with 10% goa<sup>t</sup> serum, slides were incubated with primary antibodies (1:500 Rabbit polyclonal IgG against vWF, Sigma, Gillingham UK; 1:25 Mouse monoclonal IgG1 against CD41 platelet marker integrin αIIb, clone B-9, Santa Cruz, CA, USA) at 4 ◦C overnight. An hour-long incubation of slides with 1% Bovine serum albumin/Tris-buffered saline (TBS) suspended <sup>4</sup>,6-diamidino-2-phenylindole (DAPI) (1:100 nuclei stain, Sigma) and secondary antibodies (conjugated with Cyanide Dyes, Cy2 1:100, Cy3 1:500, Jackson ImmunoResearch, Ely, UK) was completed at room temperature. IgG isotype control (1:400 Rabbit polyclonal IgG, Abcam, Cambridge, UK) was used to replace primary antibody against vWF to ensure specific binding. For assessment of nonspecific binding of secondary antibodies, reagen<sup>t</sup> control was carried out by omitting all primary antibodies.

#### *2.4. Image Analysis*

All the slides were examined under a Leica DMRA2 microscope (Leica, Wetzlar, Germany) connected to a monochrome camera (AxioCam, Oberkochen, Germany) and 3 images covering the entire length of the LA endocardium were taken. The green fluorescence shown in grey scale from endocardial endothelium was selected free-hand and measured using ImageJ (https://imagej.net/Welcome, accessed on 28 February 2018). The three measurements were then averaged to give a final number representing the detected fluorescence of the endocardial sample. Exposure time was fixed for channels that detected fluorescence from DAPI and vWF for all slides. All the slides were coded so the observer (WCC) was fully blinded when analysing the images and the order in which they were viewed was randomised by a second person (DJC). The fluorescent signals from both IgG isotype control and reagen<sup>t</sup> control were graphed for reference.

A Leica DM4000B with DFC550 colour microscopy camera (Leica, Wetzlar, Germany) was used for light microscopy. The microscopes and cameras were controlled using the Leica Application Suite Version 4.12 (Leica, Wetzlar, Germany).

#### *2.5. Statistical Analysis*

All statistical analyses were performed on GraphPad (version 8) (GraphPad, San Diego, CA, USA). Histogram and Shapiro-Wilk test were used for inspection of data distribution and normality. ANOVA test with Tukey post-hoc test and Chi-squared test with Yate correction were used to assess the difference in age and sex. Kruskal-Wallis test with Dunn's multiple comparison test was used to compare the fluorescent intensity of labelled vWF between groups. Difference with a *p*-Value < 0.05 was considered significant.
