Dual Angiotensin Receptor and Neprilysin Inhibitor Ameliorates Portal Hypertension in Portal Hypertensive Rats.

Background: Portal hypertension is characterized by exaggerated activation of the renin-angiotensin-aldosterone axis. Natriuretic peptide system plays a counter-regulatory role, which is modulated by neprilysin. LCZ696 (sacubitril/valsartan) is a dual angiotensin receptor and neprilysin inhibitor. This study evaluated the effect of LCZ696 on portal hypertensive rats. Methods: Portal hypertension was induced by partial portal vein ligation (PVL) in rats. LCZ696, valsartan (angiotensin receptor blocker), or normal saline (control) was administered in PVL rats for 10 days. Then, hemodynamic and biochemistry data were obtained. The hepatic histology and protein expressions were surveyed. On the parallel groups, the portal-systemic shunting degrees were determined. Results: LCZ696 and valsartan reduced mean arterial pressure and systemic vascular resistance. LCZ696, but not valsartan, reduced portal pressure in portal hypertensive rats (control vs. valsartan vs. LCZ696: 15.4 ± 1.6 vs. 14.0 ± 2.3 vs. 12.0 ± 2.0 mmHg, control vs. LCZ696: P < 0.05). LCZ696 and valsartan improved liver biochemistry data and reduced intrahepatic Cluster of Differentiation 68 (CD68)-stained macrophages infiltration. Hepatic endothelin-1 (ET-1) protein expression was downregulated by LCZ696. The portal-systemic shunting was not affected by LCZ696 and valsartan. Conclusion: LCZ696 and valsartan reduced mean arterial pressure through peripheral vasodilation. Furthermore, LCZ696 significantly reduced portal pressure in PVL rats via hepatic ET-1 downregulation.


Introduction
In chronic liver diseases, an increase of intrahepatic resistance develops due to chronic inflammation and fibrosis, which impedes hepatic blood flow then induces portal hypertension [1]. At the same time, peripheral vasodilation takes place, followed by renin-angiotensin-aldosterone system (RAAS) activation and sodium and fluid retention, with the initial aim to compensate for inadequate intravascular effective volume [2]. The RAAS regulates vascular tone and affects systemic and portal hemodynamics through mainly angiotensin II and aldosterone. However, exaggerated RAAS activation

Measurement of Portal Venous and Superior Mesentery Arterial Hemodynamics
The measurements of PVf (mL/min/100 g BW) and SMAf (mL/min/100 g BW) were performed as previously described [15]. In brief, after carefully dissecting the portal vein and superior mesentery artery from their surrounding soft tissue, PVf and SMAf were measured using a nonconstrictive perivascular ultrasonic transit-time flow probe (l RB, 1-mm diameter, Transonic Systems, Ithaca, NY, USA). SMAR (mmHg/mL/min/100 g BW) was calculated by dividing (MAP minus PP) by SMAf.

Portal-Systemic Shunting Analysis
The portal-systemic shunting degree was determined using the color microsphere technique as described previously [16]. In brief, 30,000 15-µm color microspheres (Dye Track, Triton Technology, San Diego, CA, USA) were slowly injected into the spleen. In "normal" condition, the microspheres would get into the liver through the splenic vein, being trapped in the liver. However, in portal hypertension, most of the microspheres bypass the stiff liver and escape through portal-systemic shunts into the lung where they become trapped. After the microspheres were injected into the spleen, the rats were euthanized and the livers and lungs were dissected and placed in new polypropylene centrifuge tubes. The numbers of microspheres were determined following the protocol provided by the manufacturer. The degree of portal-systemic shunting was calculated as the number of microspheres in the lung divided by the sum of microspheres in the liver and lung. Assuming a worst-case scenario in which two-thirds of the microspheres were trapped in the spleen, this technique could still detect a minimum shunt of 3.5%.

Histopathological and Immunohistochemical Staining of Liver
The liver was dissected free and fixed in 10% formalin solution. The sections were stained with hematoxylin and eosin (H&E) and examined by light microscopy. The immunohistochemical staining study was performed with anti-Cluster of Differentiation 68 (CD68) antibody (diluted 1:200, ab31630, Abcam, Cambridge, UK) to detect intrahepatic CD68-positive stained macrophages, which could indicate the severity of intrahepatic inflammation. The numbers of CD68-positive cells per high-power field (magnification 200×) were counted by a semiquantification method [17].

Drugs
LCZ696 and valsartan were purchased from Novartis, Taiwan. LCZ696 and valsartan were dissolved in normal saline. All solutions were freshly prepared on the days of experiments.

Data Analysis
The results are expressed as mean ± standard deviation. Statistical analyses were performed by one-way ANOVA test with post hoc Tukey analysis and the survival curve analysis using log-rank test. Results were considered statistically significant at a two-tailed P value less than 0.05.

Discussion
The major findings of the present study were that (1) LCZ696 and valsartan reduced mean arterial pressure, which may be attributed to, at least partly, a decrease of systemic vascular resistance in sham-operated and portal hypertensive rats; (2) LCZ696, but not valsartan, reduced portal pressure and SMAR in portal hypertensive rats; (3) LCZ696 and valsartan improved hepatic biochemical data and reduced CD68-stained macrophage infiltrations in the liver.
Our data showed that LCZ696 and valsartan significantly reduced MAP and SVR, which was in accordance to the previous studies in the hypertensive patients and experimental animal models [18,19]. Regarding portal hypertension, we found that LCZ696, but not valsartan, reduced PP and SMAR in portal hypertensive rats, implicating the superior portal hypotensive effects of LCZ696 compared to valsartan. Portal pressure was determined by PVf and intrahepatic resistance. Since the present study revealed that PVf was not significantly different among the LCZ696-and valsartan-treated and control groups, the portal hypotensive effect of LCZ696 might be related to the reduced intrahepatic resistance. The superiority of LCZ-696 compared to valsartan comes from its potent vasodilatory effects. LCZ696 can not only augment NPs but also influence other vasoactive agents [20]. Our data showed that LCZ696, but not valsartan, significantly downregulated the hepatic ET-1 protein expression. ET-1 is the main vasoconstrictor controlling intrahepatic resistance in portal hypertensive status, which accounts for, at least partly, the LCZ696-induced reduction of intrahepatic resistance. Indeed, the impacts of ARNI on ET-1 are different among various tissues. It is worth noting that neprilysin inhibition upregulated the ET-1 level, particularly in neprilysin-rich organs like the kidney [21]. However, another study showed that ARNI reduced the plasma ET-1 level in rats with pulmonary hypertension [22]. The various findings may be related to different research subjects and experimental settings.
On the other hand, the current study found that LCZ696 and valsartan treatments downregulated the hepatic eNOS protein expressions in portal hypertensive rats. Since nitric oxide is a potent vasodilator, the paucity of intrahepatic nitric oxide synthase plays an important role by increasing intrahepatic resistance [1]. Initially, we postulated that LCZ696 treatments could upregulate eNOS protein to reduce intrahepatic resistance in PVL rats. However, a contradictory data showed LCZ696 and valsartan downregulated the hepatic eNOS protein expressions in portal hypertensive rats. We reviewed the literatures about LCZ696 and valsartan in the function of nitric oxide and vascular endothelium. In a study comparing the effect of LCZ696 to valsartan monotherapy, LCZ696 significantly ameliorated the impairment of acetylcholine-induced vascular relaxation, while it was not exerted by valsartan [23]. However, another study showed that LCZ696 was as effective as valsartan in improving the impaired endothelium-dependent hyperpolarization-mediated responses during hypertension, while no difference was observed in acetylcholine-induced, nitric oxide-mediated relaxations [24]. Indeed, the ARNI-induced hepatic ET-1 and eNOS downregulation in portal hypertensive rats has not been reported in the previous literature, and the ET-1 downregulation may play a dominant role in the LCZ696-induced portal hypotensive effect according to our data. This interesting finding suggests a complicated interplay of vasoactive substances during chronic LCZ696 treatment.
Regarding other agents acting via modulating ET-1 and nitric oxide (NO) levels with potential portal hypotensive effects, bosentan is a nonselective ET-1 blocker and sildenafil is a phosphodiesterase which upregulates endothelial nitric oxide synthase. Bosentan has been documented to improve portal hypertension through amelioration of intrahepatic vasoconstriction [25]. On the other hand, low-dose sildenafil treatment for 1 week decreased intrahepatic resistance in rats with biliary cirrhosis, which was related to improve NO bioavailability [26]. However, a clinical study revealed that sildenafil did not influence hepatic venous pressure gradient in patients with cirrhosis [27]. Therefore, further large-scale clinical investigations may be required.
In the present study, both LCZ696 and valsartan treatments improved liver biochemistry data, as evidenced by the decreased plasma levels of alanine aminotransferase. Although the H&E staining of the liver appeared similar among control and valsartan-and LCZ696-treated groups, the numbers of intrahepatic Cluster of Differentiation 68 (CD68)-stained macrophages were significantly decreased by LCZ696 and valsartan treatments. Although LCZ696 and valsartan did not influence the protein expressions of inflammatory markers as phosphorylated NF-κB, IκBα, and iNOS, the infiltration of macrophages in the liver is an important indicator of hepatic inflammation [28]. A previous report showed that ARNI ameliorated pulmonary inflammation and reduced the release of inflammatory mediators such as interleukin (IL)-6, IL-1β, and tumor necrosis factor (TNF)-α [29]. It has been noted that angiotensin II plays an active role in inflammation [30]. Furthermore, valsartan significantly inhibited adipose tissue macrophage infiltration and inflammation in rats treated by long-term high-fat diet [31]. In addition, valsartan has been shown to reduce CD68-stained cell adhesion to endothelial cell in spontaneously hypertensive rats [32]. Therefore, the antiinflammatory effect of ARNI and ARB might be contributed mainly by the blockade of angiotensin.
RAAS blockade is commonly used to attenuate the progression of chronic kidney disease. Emerging data showed that LCZ696 is more effective than valsartan therapy alone in delaying the progression of kidney disease via antiinflammation, antioxidant, and antifibrosis effects [33]. In early stages of diabetic nephropathy, Habibi et al. [34] showed that LCZ696 was superior to valsartan in reducing proteinuria, renal ultrastructure, and tubular injury in a murine model. Furthermore, a recent study demonstrated that the renoprotection effect of LCZ696 was attributed by limiting podocyte injury [35]. In the present study, we found that LCZ696 and valsartan treatments did not affect the plasma level of creatinine, an indicator of renal injury. However, the comprehensive impact of LCZ696 on the renal angiotensin system of portal hypertension still awaits further clarification.
Since ARNI elevates the NPs levels, some researchers raise concerns about the potential adverse effects of NPs in cirrhotic and portal hypertensive status [36,37]. It has been noted that the level of BNP is upregulated in cirrhotic patients and it is significantly related to complications, including esophageal varices and ascites [36]. In addition, the serum level of CNP concentration is also elevated in patients with chronic liver disease and is associated with unfavorable prognosis of cirrhotic patients [37]. However, in PVL rats, Jonas et al. [38] showed that the plasma ANP level was reduced in PVL rats compared to the sham rats. In addition, the ANP mRNA level was downregulated by 40% to 60% in the heart of PVL rats [38]. The discrepant results from portal hypertensive animals and patients disclose the complexity of vasoactive agents in portal hypertension. Patients with portal hypertension exhibit peripheral vasodilatation and inadequate intravascular effective volume, which elicit compensatory responses including enhanced sympathetic activity, RAAS activation, and elevated circulating vasopressin and ET-1 levels [39]. Therefore, the elevated NPs in patients with long-term portal hypertension might be merely the compensatory action to counteract hyperdynamic circulation. On the contrary, ARNI treatment started at the early stage may exert a therapeutic potential to block the development of portal hypertension.
The dreadful complications of portal hypertension, including variceal bleeding and hepatic encephalopathy, are closely linked with portal-systemic collaterals. Emerging evidences show that BNP promotes vessel growth by increasing the number of endothelial progenitors and enhancing their functional properties, which may upregulate angiogenesis [40]. However, CNP has a contrary effect by counteracting vascular endothelial growth factor (VEGF) on angiogenesis [41]. CNP activation has been shown to inhibit key events of the angiogenic cascade, such as migration and proliferation of endothelial cells [42]. Regarding ANP, it has been documented to block VEGF signaling in vitro and to reduce VEGF-induced blood-retinal barrier leakage in vivo [43]. These divergent evidences show that NPs have pro-and antiangiogenesis properties in different tissues and situations. On the other hand, ARB also exerts antiangiogenesis activity. Valsartan markedly decreased capillary density in hamsters with cardiomyopathy by downregulating VEGF expression [44]. In rats with steatohepatitis, ARB significantly inhibited activated hepatic stellate cells and decreased the formation of new vessels both in vitro and in vivo [45]. Although our data showed that LCZ696 and valsartan treatments had a neutral effect on portal-systemic collateral formation and they did not affect intrahepatic VEGF protein expression in portal hypertensive rats, the impact of ARNI treatment on cirrhotic status awaits further exploration.
In conclusion, LCZ696 and valsartan reduce mean arterial pressure, which is related to a reduction of systemic vascular resistance. Besides, LCZ696, but not valsartan, exerts the portal hypotensive effect and a decrease of intrahepatic resistance, which might be contributed by hepatic ET-1 downregulation. Both LCZ696 and valsartan improve liver biochemistry data and reduce hepatic CD68-positve staining macrophages. The effects of LCZ696 on portal hypertensive animals suggest its therapeutic potential in this field.