*3.2. Histamine- and Bradykinin-Induced Microvascular Permeability Is Mediated by Di*ff*erent Signaling Pathways*

Histamine-induced (1 µM) permeability increases (3.4 ± 1.0 cm·s <sup>−</sup><sup>1</sup> <sup>×</sup> <sup>10</sup>−<sup>6</sup> ) were blocked by L-NAME (−0.1 ± 0.1 cm·s <sup>−</sup><sup>1</sup> <sup>×</sup> <sup>10</sup>−<sup>6</sup> ) but unaffected by the free radical scavengers superoxide dismutase and catalase (Figure 3A). In contrast, as shown in Figure 3B, the permeability increase induced by bradykinin (100 nM, 2.2 ± 0.2 cm·s <sup>−</sup><sup>1</sup> <sup>×</sup> <sup>10</sup>−<sup>6</sup> ) was unaffected by L-NAME (1.6 ± 0.4 cm·s <sup>−</sup><sup>1</sup> <sup>×</sup> <sup>10</sup>−<sup>6</sup> ) but blocked by co-application of superoxide dismutase and catalase (100 U·mL−<sup>1</sup> ; −0.1 ± 0.1 cm·s <sup>−</sup><sup>1</sup> <sup>×</sup> <sup>10</sup>−<sup>6</sup> *Antioxidants*  ). **2020**, *9*, x FOR PEER REVIEW 7 of 17

**Figure 3.** Involvement of reactive oxygen species and nitric oxide in histamine- and bradykinininduced permeability. Changes in venular permeability following application of histamine (**A**) or bradykinin (**B**) at 1 µM in the absence or presence of superoxide dismutase (SOD, 100 U·mL<sup>−</sup>1) and catalase (CAT, 100 U·mL<sup>−</sup>1) or L-NAME (10 µM). Data from paired measurements in 4 venules from **Figure 3.** Involvement of reactive oxygen species and nitric oxide in histamine- and bradykinin-induced permeability. Changes in venular permeability following application of histamine (**A**) or bradykinin (**B**) at 1 <sup>µ</sup>M in the absence or presence of superoxide dismutase (SOD, 100 U·mL−<sup>1</sup> ) and catalase (CAT, 100 U·mL−<sup>1</sup> ) or L-NAME (10 µM). Data from paired measurements in 4 venules from 4 different animals. Data were analyzed using a paired Student's *t*-test.

#### 4 different animals. Data were analyzed using a paired Student's *t*-test. *3.3. Bradykinin-Induced Microvascular Permeability Is Potentiated by IL-1*β

permeability responses (Figure 4B).

*3.3. Bradykinin-Induced Microvascular Permeability Is Potentiated by IL-1β* Bradykinin applied abluminally to the cremaster microcirculation induced a dose-dependent Bradykinin applied abluminally to the cremaster microcirculation induced a dose-dependent increase in permeability to FITC-albumin (Figure 4A,B). Acute treatment with IL-1β (30 pM) for 10 min,

increase in permeability to FITC-albumin (Figure 4A,B). Acute treatment with IL-1β (30 pM) for 10 min, followed by wash-off of IL-1β, resulted in a significant potentiation of bradykinin-induced

followed by wash-off of IL-1β, resulted in a significant potentiation of bradykinin-induced permeability responses (Figure 4B). *Antioxidants* **2020**, *9*, x FOR PEER REVIEW 8 of 17

**Figure 4.** Acute treatment with IL-1β potentiates bradykinin-induced microvascular permeability. (**A**) Experimental protocol for dose–response curves to bradykinin (Bk) in the presence of kininase inhibitors. Bradykinin dose–response curves were obtained in the absence of IL-1β and after acute application of IL-1β (30 pM) for 10 min, followed by wash-off and reapplication of bradykinin applications to the same post-capillary venule. The numbers in panel A indicate the time in minutes for each phase of the protocol. (**B**) Bradykinin permeability response curve following IL-1β **Figure 4.** Acute treatment with IL-1β potentiates bradykinin-induced microvascular permeability. (**A**) Experimental protocol for dose–response curves to bradykinin (Bk) in the presence of kininase inhibitors. Bradykinin dose–response curves were obtained in the absence of IL-1β and after acute application of IL-1β (30 pM) for 10 min, followed by wash-off and reapplication of bradykinin applications to the same post-capillary venule. The numbers in panel A indicate the time in minutes for each phase of the protocol. (**B**) Bradykinin permeability response curve following IL-1β preapplication was significantly greater than all other responses. Data denote mean ± SEM of measurements in 8 vessels from 8 different animals, repeated measures, analysis of co-variance.

#### preapplication was significantly greater than all other responses. Data denote mean ± SEM of measurements in 8 vessels from 8 different animals, repeated measures, analysis of co-variance. *3.4. A Role for NADPH Oxidase and Reactive Oxygen Species in the Potentiation of Bradykinin-Induced Microvascular Permeability by IL-1*β

*3.4. A Role for NADPH Oxidase and Reactive Oxygen Species in the Potentiation of Bradykinin-Induced Microvascular Permeability by IL-1β* Figure 5 summarizes changes in permeability obtained in single post-capillary venules in response to bradykinin (100 nM), IL-1β (30 pM), or bradykinin (100 nM) following 10 min treatment Figure 5 summarizes changes in permeability obtained in single post-capillary venules in response to bradykinin (100 nM), IL-1β (30 pM), or bradykinin (100 nM) following 10 min treatment with IL-1β (30 pM). Apocynin, co-applied with IL-1β, effectively prevented the potentiation of bradykinin-induced permeability (Figure 5). Free radical scavenging by a mixture of by superoxide dismutase and catalase completely blocked the permeability response to bradykinin following IL-1β.

with IL-1β (30 pM). Apocynin, co-applied with IL-1β, effectively prevented the potentiation of bradykinin-induced permeability (Figure 5). Free radical scavenging by a mixture of by superoxide dismutase and catalase completely blocked the permeability response to bradykinin following IL-1β.

**Figure 5.** IL-1β potentiates bradykinin-induced microvascular permeability and involves NADPH oxidase and reactive oxygen species. The potentiated response to bradykinin (100 nM) application, following application of IL-1β (30 pM, 10 min), was prevented by co-application of apocynin (1 µM) with IL-1β (30 pM). Scavenging reactive oxygen species with superoxide dismutase (100 U/mL) and **Figure 5.** IL-1β potentiates bradykinin-induced microvascular permeability and involves NADPH oxidase and reactive oxygen species. The potentiated response to bradykinin (100 nM) application, following application of IL-1β (30 pM, 10 min), was prevented by co-application of apocynin (1 µM) with IL-1β (30 pM). Scavenging reactive oxygen species with superoxide dismutase (100 U/mL) and catalase (100 U/mL) completely blocked the permeability response to bradykinin. Data denote mean ± SEM, n = 10 venules from 10 animals. One-way ANOVA with Tukey's multiple comparison test.

#### catalase (100 U/mL) completely blocked the permeability response to bradykinin. Data denote mean ± SEM, n = 10 venules from 10 animals. One-way ANOVA with Tukey's multiple comparison test. *3.5. Pretreatment of Animals with Simvastatin*

*3.5. Pretreatment of Animals with Simvastatin*  Figure 6A demonstrates that in non-treated animals, application of IL-1β (30 pM, 10 min) in the absence of bradykinin resulted in a small permeability increase, which was prevented by the pretreatment of animals with simvastatin (5 mg·mL−1) 24 h before. In subsequent experiments, animals were pretreated with simvastatin (5 mg·mL−1) 24 h before an acute application of IL-1β (30 pM), followed by wash-off of IL-1β and application of bradykinin (100 nM). Pretreatment with simvastatin did not alter hyperpermeability induced by bradykinin alone (*p* = 0.411; Figure 6B). As shown in Figure 6B, pretreatment with simvastatin abolished the potentiation of bradykinin-induced microvascular permeability by IL-1β, with no significant effect on the permeability response to bradykinin alone. To examine whether the simvastatin induced loss of IL-1β potentiation of the bradykinin permeability response was due to an upregulation of HO-1, the cremaster preparation was treated with tin protophoryrin IX (SnPP), a known inhibitor of HO-1 [37,38]. Notably, inhibition of HO-1 with SnPP could not restore IL-1β mediated potentiation of bradykinin-induced permeability (Figure 6C). Apocynin (1 µM), a specific inhibitor of NADPH oxidase, also had no effect Figure 6A demonstrates that in non-treated animals, application of IL-1β (30 pM, 10 min) in the absence of bradykinin resulted in a small permeability increase, which was prevented by the pretreatment of animals with simvastatin (5 mg·mL−<sup>1</sup> ) 24 h before. In subsequent experiments, animals were pretreated with simvastatin (5 mg·mL−<sup>1</sup> ) 24 h before an acute application of IL-1β (30 pM), followed by wash-off of IL-1β and application of bradykinin (100 nM). Pretreatment with simvastatin did not alter hyperpermeability induced by bradykinin alone (*p* = 0.411; Figure 6B). As shown in Figure 6B, pretreatment with simvastatin abolished the potentiation of bradykinin-induced microvascular permeability by IL-1β, with no significant effect on the permeability response to bradykinin alone. To examine whether the simvastatin induced loss of IL-1β potentiation of the bradykinin permeability response was due to an upregulation of HO-1, the cremaster preparation was treated with tin protophoryrin IX (SnPP), a known inhibitor of HO-1 [37,38]. Notably, inhibition of HO-1 with SnPP could not restore IL-1β mediated potentiation of bradykinin-induced permeability (Figure 6C). Apocynin (1 µM), a specific inhibitor of NADPH oxidase, also had no effect in simvastatin pretreated animals, suggesting that pretreatment with simvastatin was sufficient to prevent the assembly of NADPH oxidase induced by IL-1β (Figure 6D).

in simvastatin pretreated animals, suggesting that pretreatment with simvastatin was sufficient to

prevent the assembly of NADPH oxidase induced by IL-1β (Figure 6D).

**Figure 6.** Pretreatment with simvastatin abolishes potentiation of bradykinin-induced microvascular permeability by IL-1β. (**A**) IL-1β (30 pM) application itself for 10 min resulted in a small permeability increase in non-treated rats (control), which was abrogated by the pretreatment with simvastatin (5 mg·mL<sup>−</sup>1) 24 h before. (**B**) Potentiation of bradykinin-induced (100 nM) permeability by IL-1β (30 pM) was compared in cremaster muscle post-capillary venules from control and simvastatin pretreated (5 mg/kg; i.p.) animals. (**C**) Inhibition of HO-1 with SnPP (5 µM) did not restore IL-1β potentiation of bradykinin-induced permeability in simvastatin pretreated (5 mg/kg; i.p.) animals. Data were analyzed using a paired Student's *t*-test. (**D**) Apocynin had no effect on simvastatin-treated animals, **Figure 6.** Pretreatment with simvastatin abolishes potentiation of bradykinin-induced microvascular permeability by IL-1β. (**A**) IL-1β (30 pM) application itself for 10 min resulted in a small permeability increase in non-treated rats (control), which was abrogated by the pretreatment with simvastatin (5 mg·mL−<sup>1</sup> ) 24 h before. (**B**) Potentiation of bradykinin-induced (100 nM) permeability by IL-1β (30 pM) was compared in cremaster muscle post-capillary venules from control and simvastatin pretreated (5 mg/kg; i.p.) animals. (**C**) Inhibition of HO-1 with SnPP (5 µM) did not restore IL-1β potentiation of bradykinin-induced permeability in simvastatin pretreated (5 mg/kg; i.p.) animals. Data were analyzed using a paired Student's *t*-test. (**D**) Apocynin had no effect on simvastatin-treated animals, suggesting that the pretreatment with simvastatin was sufficient to prevent the assembly of NADPH oxidase induced by IL-1β. One-way ANOVA with Tukey's multiple comparison test. Data denote mean ± SEM of paired measurements in 3–6 venules from 4–6 different animals in each group.

#### oxidase induced by IL-1β. One-way ANOVA with Tukey's multiple comparison test. Data denote **4. Discussion**

mean ± SEM of paired measurements in 3–6 venules from 4–6 different animals in each group. **4. Discussion**  The present study in an intact skeletal muscle microvasculature provides the first evidence that simvastatin prevents small permeability increases induced by IL-1β alone, as well as IL-1β mediated potentiation of bradykinin-induced microvascular permeability, highlighting the importance of pleiotropic effects of statins. Importantly, inhibition of Nox2 assembly by apocynin [37] or scavenging The present study in an intact skeletal muscle microvasculature provides the first evidence that simvastatin prevents small permeability increases induced by IL-1β alone, as well as IL-1β mediated potentiation of bradykinin-induced microvascular permeability, highlighting the importance of pleiotropic effects of statins. Importantly, inhibition of Nox2 assembly by apocynin [37] or scavenging of reactive oxygen species with superoxide dismutase and catalase abolished the microvascular hyperpermeability induced by IL-β and bradykinin, strongly implicating Nox2 mediated free radical generation in increased microvascular permeability.

suggesting that the pretreatment with simvastatin was sufficient to prevent the assembly of NADPH

of reactive oxygen species with superoxide dismutase and catalase abolished the microvascular hyperpermeability induced by IL-β and bradykinin, strongly implicating Nox2 mediated free radical generation in increased microvascular permeability. Our study confirms our previous findings in cerebral microvessels in vivo that acute bradykinin application results in a reactive oxygen species mediated increase in microvascular permeability. We report here that basal skeletal muscle microvascular permeability is reduced by scavenging reactive oxygen species, and that an increased permeability observed following inhibition of nitric oxide generation is abrogated by superoxide dismutase and catalase (Figure 2). This finding indicates that Our study confirms our previous findings in cerebral microvessels in vivo that acute bradykinin application results in a reactive oxygen species mediated increase in microvascular permeability. We report here that basal skeletal muscle microvascular permeability is reduced by scavenging reactive oxygen species, and that an increased permeability observed following inhibition of nitric oxide generation is abrogated by superoxide dismutase and catalase (Figure 2). This finding indicates that constitutive NO generation effectively scavenges basal formation of reactive oxygen species. There are numerous indications in the literature that NOS inhibition exacerbates inflammatory conditions, and this may provide an explanation for this.

constitutive NO generation effectively scavenges basal formation of reactive oxygen species. There

Bradykinin-induced microvascular permeability has been associated with increased NO production and vasodilation [39,40], and a key role for reactive oxygen species generated following bradykinin receptor activation has been reported in cultured endothelial cells in vitro [18,41] and in rat cerebral microvessels in vivo [19]. Further studies in vivo, using scavengers of reactive oxygen species, confirmed these findings and showed that superoxide generation contributed to the vasodilation [42] and increased permeability following bradykinin application [19,20]. Similar to these findings, we have shown that bradykinin-induced permeability in rat cremaster muscle post-capillary venules was inhibited by superfusion with superoxide dismutase and catalase (Figure 3B). In addition, the fact that L-NAME did not inhibit bradykinin-induced permeability in cremaster muscle venules argues against a role for NO and supports findings in rat mesentery [43] and brain [20].

Histamine has been shown to increase cGMP production in endothelial cells via endothelial derived NO production, with increased vascular permeability and vasodilation mediated via activation of soluble guanylyl cyclase [44,45]. In this context, treatment of the cremaster muscle preparation with L-NAME allowed us to establish that histamine-induced permeability increases were NO-dependent but unaffected by scavenging of reactive oxygen species.

Although intracellular signaling pathways underlying reactive oxygen species mediated permeability increases were not studied, it is likely that bradykinin may induce permeability changes via the generation of free radicals during arachidonic acid metabolism leading to Ca2<sup>+</sup> entry through areas of lipid peroxidation, as we previously reported for brain pial microvessels in vivo [20]. The attenuation of bradykinin-induced permeability responses in the presence of superoxide dismutase and catalase suggests that bradykinin-induced permeability increases are linked to free radical generation in rat cremaster muscle. This finding is consistent with previous reports from our laboratory that permeability responses to bradykinin in the brain microvasculature in vivo involve the generation of reactive oxygen species [19,20].

Statins have been described to improve endothelial function, reduce vascular permeability and edema formation in different experimental and clinical studies [9,46–50]. A clinical study with hypercholesterolemic patients assessed transcapillary albumin escape rate as an index of macromolecular permeability, and notably simvastatin treatment over 1 month normalized increases in transvascular albumin leakage independently of lipid levels in these patients [51]. Using an Evans blue dye exclusion test, simvastatin treatment for 1 month reduced vascular leakage in the aorta of hyperlipidemic rabbits [52]. Moreover, simvastatin treatment for 5 weeks improves endothelial barrier permeability changes in the brain, retina and myocardium of streptozotocin-induced diabetes rats [53].

Notably, administration of simvastatin 24 h before and along with intratracheal injection of lipopolysaccharide (LPS) attenuates vascular leak and inflammation in a murine inflammatory model of acute lung injury [7]. Simvastatin reduced approximately 50% of albumin levels in the bronchoalveolar lavage, and leakage of albumin conjugated with Evans blue dye into the pulmonary parenchyma in a murine inflammatory lung injury model [7]. Additionally, acute oral administration of simvastatin reduces brain edema formation and blood–brain barrier permeability after traumatic brain injury in rats [9]. In a model of experimental intracerebral hemorrhage in rats, simvastatin treatment increases cerebral blood flow in the injured region of the brain and reduces blood-brain barrier (BBB) permeability and cerebral edema [10]. Simvastatin also acutely protects the neurovascular unit, reducing blood–brain barrier permeability, when administered subcutaneously 30 min after transient cerebral ischemia induced by middle cerebral artery occlusion [8]. It is important, however, to highlight that most of these previous studies evaluated permeability changes using indirect methods, such as the Evans blue dye test. Our findings establish that simvastatin has the potential to protect the endothelial barrier and reduce vascular permeability; however, further studies are necessary to elucidate the mechanisms involved in these processes and measuring permeability coefficients.

It has been reported that lovastatin induces expression of bradykinin type 2 receptors in cultured human coronary artery endothelial cells [54]. However, in order to confirm these in vitro findings, additional in vivo studies with statin treatment in humans and in animal models are required. Simvastatin was chosen in the present study based on its potency and pharmacokinetic properties. The potency rank order for HMG-CoA reductase inhibition among the second-generation statins is simvastatin > pravastatin > lovastatin mevastatin [55]. Furthermore, lipophilic statins, such as simvastatin, are considered more likely to enter endothelial cells by passive diffusion in contrast to hydrophilic statins, such as pravastatin and rosuvastatin, which are primarily targeted to the liver [56]. Hydrophilic statins have been described to exert similar effects on the vasculature to lipophilic statins suggesting that specific mechanisms may exist for the uptake of the former; however, this may take longer than the lipophilic statins [57].

Bradykinin has been shown to play a key role in blood–brain barrier disruption and edema formation in different pathophysiological processes, including stroke [58,59]. IL-1β is rapidly released from the brain parenchyma after an ischemic event, triggering NADPH activation and thereby potentiating bradykinin-induced microvascular permeability [60]. Moreover, the release of bradykinin and IL-1β contribute to reactive oxygen species generation in the early stages of cerebral ischemia and reperfusion injury [19]. IL-1β application increases superoxide anion release from human endothelial cells and increases reactive oxygen species generation from mitochondria and NADPH oxidase in cultured retinal epithelial cells [61]. Additionally, bradykinin may act as a potential mediator of vascular leakage and pulmonary edema in patients with COVID-19 [12–14]. In this context, IL-1β release has been proposed as one of the key inflammatory mediators in COVID-19 [23–25] and could potentially exacerbate bradykinin-induced vascular permeability in these patients. Thus, employing drugs already in clinical use, such as simvastatin, could offer a therapeutic strategy for decreasing bradykinin- and/or IL-1β-induced pulmonary edema in patients with COVID-19.

In accordance with previous studies [19,62], we observed that concomitant application of IL-1β with apocynin, a specific inhibitor of NADPH oxidase, abolished the potentiation of bradykinin-induced microvascular permeability by IL-1β (Figure 5). Apocynin rapidly prevents the assembly of NADPH oxidase, by blocking the cytosolic subunit p47phox translocation to the cell membrane [37]. Furthermore, apocynin had no effect on simvastatin pretreated rats, suggesting that simvastatin pretreatment was sufficient to prevent the assembly of NADPH oxidase induced by IL-1β (Figure 6C). Pretreatment with simvastatin was effective in inhibiting IL-1β actions on bradykinin-induced permeability, suggesting that protection afforded by simvastatin against microvascular hyperpermeability may in part be due to inhibition of Nox2. Furthermore, it has been reported that IL-1β alone rapidly (within 10 to 15 min of its application) increases superoxide release in both cultured endothelial cells [63] and retinal epithelial cells, with the latter study suggesting that NADPH oxidase activation was involved [61]. Similarly, we have also demonstrated that IL-1β itself results in a small permeability increase (see Figure 6A), which was abrogated by simvastatin. These findings strengthen the proposition that simvastatin pretreatment prevents IL-1β stimulation of ROS generation via Nox2 assembly. Nevertheless, additional studies are necessary to investigate whether other pro-inflammatory cytokines, such as IL-2 and IL-6, could also increase bradykinin-induced microvascular permeability and whether statins could modulate the profile of these cytokines.

By inhibiting reactive oxygen species generation and reducing the NAD+/NADH ratio, statins will reduce cellular oxidative stress [64–66]. Thus, protective cardiovascular effects of statins may be directly associated with their cellular antioxidant properties, independent of the cholesterol-lowering effects of these agents. As statins have been reported to activate the redox sensitive transcription factor Nrf2 and upregulate the cytoprotective antioxidant enzyme HO-1 [29–33], we postulated that loss of IL-1β potentiation of bradykinin-induced permeability may be a consequence of enhanced HO-1 activity. Notably, inhibition of HO-1 with SnPP did not restore the IL-1β-induced potentiation (see Figure 6B), suggesting that simvastatin probably acts via reducing NADPH oxidase activity. Statins have been reported to reduce NADPH oxidase activity by inhibiting isoprenylation of the protein Rac1 [28,66–68].

Isoprenylated Rac1 is essential for assembly of the NADPH oxidase enzymatic complex on the cell membrane [69]. In patients with heart failure, statin treatment reduces Rac1 function, NADPH oxidase activity and levels of reactive oxygen species [70], a finding consistent with our observation that simvastatin pretreatment reduces IL-1β/bradykinin mediated microvascular hyperpermeability. Reactive oxygen species have been reported to negatively regulate cell–cell adhesion controlled by intercellular adhesion molecules, such as VE-cadherin and β-catenin, which are linked to transmembrane molecules and the actin cytoskeleton. In addition to a role for reactive oxygen species, RhoA activation is important for bradykinin-induced permeability [71]. RhoA-GTP activation leads to actin cytoskeleton contraction, resulting in the breakdown of the endothelial barrier [72]. In this context, statins protect the endothelial barrier, reduce oxidative stress and inhibit isoprenylation and activation of RhoA and Rac1 [52,66].

In the present study, protection afforded by simvastatin against increased microvascular permeability in cremaster muscle venules in response to IL-1β and bradykinin may be associated with inhibitory effects on the assembly of NADPH oxidase subunit, leading to diminished NADPH oxidase mediated superoxide release. Although not investigated in the present study, other cytokines such as such as IL-6, TNF-α and IL17 may similarly potentiate bradykinin-induced microvascular permeability. It has been reported that simvastatin inhibits IL-6, IL-8 and IL-1β production in vitro [73,74], which may contribute to its protective role in cardiovascular diseases. We have now demonstrated that a key anti-inflammatory action of simvastatin is to prevent IL-1β mediated potentiation of bradykinin-induced permeability in skeletal muscle microvasculature. This study highlights a novel action by which simvastatin prevents the potentiation of bradykinin-induced permeability by IL-1β, possibly by targeting the assembly of NADPH oxidase subunits. The approach undertaken in this study was functional, and future studies focusing on the molecular pathways are needed to elucidate the exact mechanism by which simvastatin reduces NADPH oxidase assembly.

## **5. Conclusions**

Simvastatin could play an important role in the prevention and/or treatment of patients with a high predisposition to microvascular hyperpermeability mediated by pro-inflammatory cytokines potentiating the actions of bradykinin, with implications perhaps for vascular leakage and pulmonary edema.

**Author Contributions:** F.F., P.A.F. and G.E.M. designed the experiments and critically discussed and analyzed experimental data. F.F. and M.S. conducted all experiments. F.F. drafted the manuscript and P.A.F. and G.E.M. revised the manuscript. F.F., E.T., P.A.F. and G.E.M. read and approved the final manuscript and all authors agree to accept accountability for all aspects of the work. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by CNPq, Brazil (205398/2014-6/SWE, to F.F) and British Heart Foundation (FS/15/6/31298, to G.E.M.).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


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

© 2020 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 (http://creativecommons.org/licenses/by/4.0/).

*Article*
