**1. Introduction**

The mast cell, part of the innate immune system, generally resides in tissues such as the lung and the skin to protect against pathogens like bacteria and parasites. Mast cells have also been described to participate in diseases such as asthma, allergies, and rheumatoid arthritis. Over the last decades, the mast cell has been implicated in cardiovascular diseases as well, for example in atherosclerosis [1,2], the underlying pathology of acute cardiovascular diseases including peripheral artery disease (PAD) [3,4]. Interestingly, mast cell numbers in advanced human atherosclerotic plaques obtained after endarterectomy surgery were seen to be of predictive value for the incidence of a secondary cardiovascular event [5]. In those plaques, mast cell density associated with the number of CD31+ microvessels [5]. Mast cells have also been associated with arteriogenesis and collateral formation [6]. Mast cells can secrete growth factors and pro-inflammatory cytokines that can recruit immune cells such as neutrophils and monocytes to the site of inflammation, thus further enhancing a

pro-inflammatory response [6–8]. Patients with systemic mastocytosis, a disease that is characterized by the excessive accumulation of mast cells in tissue or organs, experienced an increased prevalence of cardiovascular disease events, such as myocardial infarction, stroke, and importantly, peripheral artery disease, which is actually caused by the development of atherosclerosis in the arteries that supply oxygen to the extremities [9]. These associative data sugges<sup>t</sup> that the mast cell may actively contribute to these underlying pathologies, and preclinical studies have been performed to elucidate underlying mast cell pathways that are causally related to vascular remodeling and the related disease outcome.

We speculate that mast cells may a ffect atherosclerotic plaque progression and PAD by increasing angiogenesis. Plaque angiogenesis has been shown to be related to atherosclerotic plaque progression and also, during tumor development, mast cells have been implicated in angiogenesis [10]. Mast cells have been shown to induce tumor endothelial proliferation by the release of Vascular Endothelial Growth Factor (VEGF) in response to the hypoxic environment in the tumor [10], which may be translatable to muscles in PAD patients, in which hypoxia occurs as well [11,12]. Although induction of neovascularization, in particular angiogenesis, maybe unfavorable for atherosclerosis progression, in PAD, the induction of neovascularization by mast cells, more precisely by inducing collateral formation and angiogenesis, may act as a repair pathway to resupply the ischemic limb tissue with oxygen. This neovascularization process requires a pro-inflammatory response, which is thus, on one hand, beneficial to resolve ischemia, but may on the other hand lead to enhanced progression of atherosclerosis. This two-faced process is also known as the Janus phenomenon [13]. Recently, the contribution of mast cells to arteriogenesis and collateral formation in a shear-stress induced mouse model of hind limb ischemia (HLI) has been described by Chillo and colleagues [6]. In that study, mast cells were systemically activated with the mast cell activator compound 48/80 upon ligation of the femoral artery, which resulted in increased hind limb perfusion. Treatment with the mast cell stabilizer cromolyn prevented the mast cell-induced e ffects on arteriogenesis and the investigators identified the neutrophil as a prominent e ffector cell involved in the mechanism behind mast cell-induced arteriogenesis. However, in that study, mast cell activation-dependent e ffects on angiogenesis in the ischemic muscles were not reported.

In this study, we aimed to investigate whether local induction of mast cell activation can stimulate blood flow recovery in a mouse hind limb ischemia model by inducing angiogenesis as well as arteriogenesis.

First, we analyzed the number and activation status of mast cells in human ischemic tissue, obtained after limb amputation. Next, we induced hind limb ischemia in mice by ligation of the femoral artery and at time of ligation, we activated mast cells in the hind limb by a skin sensitization/challenge protocol using a pluronic gel to apply the hapten DNP locally at the ligation site as described before [7,14], after which blood perfusion was measured using laser Doppler imaging. Mast cell activation increased reperfusion of the hind limb by not only increasing arteriogenesis, but also angiogenesis, which is at least partly induced by increasing the pro-inflammatory monocyte response.

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

#### *2.1. Tissue Collection from Patients with End-Stage Peripheral Artery Disease*

Sample collection of tissues from patients with end-stage peripheral artery disease undergoing limb amputation was approved by the Medical Ethics Committee of the Leiden University Medical Center (Protocol No. P12.265). Written informed consent was obtained from the participants. Inclusion criteria were a minimum age of 18 years and lower limb amputation, excluding ankle, foot, or toe amputations. Exclusion criteria were suspected or confirmed malignancy and inability to give informed consent. Gastrocnemius and soleus muscle samples obtained from 15 patients were formalin fixed, processed, para ffin embedded sectioned, and stained for mast cells. From these patients, *n* = 1 had type I diabetes and *n* = 7 su ffered from type II diabetes.

#### *2.2. Hind Limb Ischemia Model*

This study was performed in accordance with the Directive 2010/63/EU of the European Parliament and Dutch governmen<sup>t</sup> guidelines. All experiments were approved (reference number 14185) by the Leiden University and Leiden University Medical Center committee on animal welfare (Leiden, the Netherlands). Wild-type C57Bl/6J mice were bred in our in-house breeding facility. Male mice aged 8 to 12 weeks were housed in groups with free access to water and regular chow.

Before the unilateral hind limb ischemia, mice were anesthetized by i.p. injection of midazolam (8 mg/kg, Roche Diagnostics, Basel, Switzerland), medetomidine (0.4 mg/kg, Orion, Espoo, Finland), and fentanyl (0.08 mg/kg, Janssen Pharmaceuticals, Beerse, Belgium). Hind limb ischemia was induced by electrocoagulation on two locations of the left femoral artery; the first ligation proximal to the superficial epigastric artery and the second proximal to the bifurcation of the popliteal and saphenous artery [15,16]. After surgery, anesthesia was antagonized with with atipamezol (2.5 mg/kg, Orion, Espoo, Finland) and flumazenil (0.5 mg/kg, Fresenius Kabi, Bad Homburg vor der Höhe, Germany).and buprenorphine (0.1 mg/kg, MSD Animal Health, Keniworth, NJ, USA) was provided as a painkiller. For the time course, 5 mice per time point were used, whereas for both the long-term (t28) and short-term (t9) HLI experiments, 8–9 mice per group were used.

#### *2.3. Local Mast Cell Activation with DPN treatment*

Mice were skin-sensitized on the shaved abdomen and paws for 2 consecutive days with a dinitrofluorobenzene (DNFB (D1529) solution (0.5% *v*/*v* in acetone:olive oil (4:1), Sigma-Aldrich, St. Louis, MO, USA) as described previously to sensitize the mice for the hapten DNP [7,14]. In the control mice, a vehicle solution of acetone:olive oil (4:1) was applied. At the end of the hind limb ischemia procedure, which was scheduled one week after the skin-sensitization procedure, 50 μg dinitrophenyl hapten (DNP (D198501), Sigma-Aldrich, St. Louis, MO, USA) in a pluronic gel (25% *<sup>w</sup>*/*<sup>v</sup>*, Sigma-Aldrich, St. Louis, MO, USA) was applied around the ligated areas of the left hind limb to locally activate the mast cells. Empty pluronic gel was applied in the control mice. This model has been previously applied [7,14,17,18] and has been shown to specifically induce mast cells activation upon local hapten application.

#### *2.4. Laser Doppler Perfusion Measurements*

Before and directly after surgery and at 3, 7,10, 14, 21, and 28 days after surgery, blood flow recovery to the ligated hind limb and the unligated control paw were measured using Laser Doppler Perfusion Imaging (LDPI) (Moor Instruments, Axminster, UK). Before the LDPI measurements, mice were anesthetized by i.p. injection of midazolam (8 mg/kg) and medetomidine (0.4 mg/kg,). Next, mice were placed in a double glazed pot, perfused with water at 37 ◦C for 5 min [19]. After LDPI, anesthesia was antagonized by subcutaneous injection of flumazenil (0.7 mg/kg). LDPI measurements in the ligated paw were normalized to measurements of the unligated paw, as an internal control. At sacrifice, after the last LDPI measurement, analgesic fentanyl (0.08 mg/kg) was administered, blood was drawn, and mice were sacrificed via cervical dislocation. The adductor and soleus muscles were harvested and fixed in 4% formaldehyde. For a second short experiment, animals were sacrificed at day 9, where blood was collected by orbital bleeding and the inguinal lymph nodes were isolated for further analyses. Again, the adductor and soleus muscles were harvested and fixed in 4% formaldehyde for histology analysis.
