Microbiota transplants from feces or gut content attenuated portal hypertension and portosystemic collaterals in cirrhotic rats

Liver cirrhosis and portal hypertension is the end of chronic liver injury with hepatic, splanchnic and portosystemic collateral systems dysregulation. Liver injury is accompanied by gut dysbiosis whereas dysbiosis induces liver fibrosis, splanchnic angiogenesis and dysregulated vascular tones vice versa, making portal hypertension aggravated. It has been proved that intestinal microbiota transplantation alleviates dysbiosis. Nevertheless, the influences of microbiota transplantation on cirrhosis related portal hypertension are not so clear. Liver cirrhosis with portal hypertension was induced by bile duct ligation in rats. Sham rats were surgical controls. Rats randomly received vehicle, fecal or gut (terminal ileum) material transplantation. The results showed that microbiota transplantation from feces or gut material significantly reduced portal pressure in cirrhotic rats ( P = .010, .044). Hepatic resistance, vascular contractility, fibrosis and relevant protein expressions were not significantly different among cirrhotic rats. However, microbiota transplantation ameliorated splanchnic hyperdynamic defined by whole mesenteric window vascular density, in both transplantation groups Portosystemic shunts determined by splenorenal shunt flow decreased in both transplantation groups ( P = .037, .032). Shunting severity assessed by microsphere distribution method showed consistent results. Compared to sham rats, cirrhotic rats lacked Lachnospiraceae. Both microbiota transplants increased Bifidobacterium. In conclusion, microbiota transplantation in cirrhotic rats reduced portal pressure, alleviated splanchnic hyperdynamic circulation and portosystemic shunts. The main beneficial effects may be focused on portosystemic collaterals-related events, such as hepatic encephalopathy and gastroesophageal variceal hemorrhage. Further clinical investigations are mandatory.


Introduction
Portal hypertension is a pathological condition linked to liver damages and most commonly, liver cirrhosis. Along with the elevation of portal pressure, lethal complications develop with high morbidity and mortality [1]. Portal pressure is determined by hepatic, splanchnic and collateral vascular systems. During the progression of liver cirrhosis, abnormal collagen fiber deposition in the liver interferes with the outflow of portal blood flow. The dysregulation of hepatic vascular contractility further enhanced hepatic vascular resistance. Furthermore, the splanchnic inflow increases due to pathological splanchnic vasodilatation and angiogenesis. Both of them increase portal venous inflow and further contribute to a high pressure in the portal system [2]. To divert the portal blood flow, portosystemic collateral vessels and complications develop. A recent human study from Baveno VI-SPSS group showed that large spontaneous porto-systemic shunts correlated with overt hepatic encephalopathy and poor 1-year survival [3].
Gut microbiota consists of 10-100 trillion bacteria and is considered an extremely complex system [4]. The microbiome and liver orchestrate to maintain the body homeostasis. However, the gut-liver mutually beneficial relationship becomes unbalanced when dysbiosis develops. The break down leads to the aggravation of cirrhosis and portal hypertension though multiple aspects [5]. In the liver, gut dysbiosis enhances inflammation, fibrogenesis and even vascular resistance [6]. In splanchnic system, dysbiosis promotes angiogenesis [7]. Reversion of dysbiosis is thus a conceivable way to control portal hypertension.
Fecal microbiota transplantation (FMT) is the transferring of fecal microbiota from healthy individuals. It has been noted that in Clostridium difficile-associated disease, FMT alleviated dysbiosis and restored normal flora in patients with dysbiosis Downloaded from http://portlandpress.com/clinsci/article-pdf/doi/10.1042/CS20210602/925584/cs-2021-0602.pdf by guest on 07 December 2021 Clinical Science. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/CS20210602 [8]. Regarding the relevant study in liver diseases, a phase 1 study showed that FMT significantly improved hepatic encephalopathy [9]. Another phase 1, randomized, placebo-controlled trial performed in decompensated cirrhotic patients revealed that FMT alleviated dysbiosis and hepatic encephalopathy [10]. However, in both studies, liver function as assessed by model for end-stage liver disease (MELD score) was not affected by FMT. Furthermore, a comprehensive survey of the effects of FMT on cirrhosis itself and portal hypertension-related derangements is still lacking [11].
Until now, it has not been confirmed in which part of the alimentary tract do microbiota affect the development and aggravation of liver cirrhosis and portal hypertension the most. In cirrhosis, the bile acid metabolism and bacteria translocation occur primarily in terminal ileum. A recent study showed that in cirrhotic rats, dysbiosis in terminal ileum resulted in immune dysregulation. While decontamination with antibiotics reversed the process [12]. Although the potential role of FMT in cirrhosis has drawn attention, the optimal source of microbiota in the alimentary tract is still unclarified.
This study therefore aimed to find out the optimal way of gut microbial transplantation accompanied by mechanistic survey from the perspective of basic research. We tested the effects of microbiota transplants from different part of intestine, including feces or gut materials. Furthermore, the impacts of microbiota transplantations on portal hypertension-related parameters, especially of the splanchnic and collateral systems, were studied. donor microbiota, antibiotics were administered on the 3rd, 4th and 5th days after operations. Sham or BDL rats were then randomly divided into three groups: 1.
Control (vehicle), 2. Fecal material transplantation (FMT), 3. Gut material transplantation (GMT). Transplantations were performed for 5 consecutive days since the 7 th day. Experiments were performed on the 28 th day. Parallel groups received sham or BDL operation without antibiotics and transplantation were employed to exclude the effects of antibiotics.
To evaluate the therapeutic effect of FMT, BDL rats were randomly divided into

Preparation of donor gut and feces material
The gut material was collected from donor rats at the time when they were just sent from the animal breeding center without any treatment. After anesthesia, the gut material of terminal ileum and fecal pallets were collected respectively. The collected material was placed in transfer buffer (pre-reduced sterile phosphate buffered saline containing 0.05 % cysteine HCl, 2 mL/g) on ice. The material was then immediately homogenized and centrifuged at 800 g for 2 minutes. After that, the supernatant was collected, mixed with glycerol to make the concentration of glycerol equal to 10%, and stored at -80 ℃ until being used [14]. Before the transplantation, the recipient rats received antibiotics including imipenem (50 mg/kg/day), va Forty-eight hours after the last dose of antibiotics treatment, recipient rats were oralgavaged with fecal supernatant (2.4 ml/kg, equivalent to 0.43 g/kg of fecal or gut material) for 5 consecutive days [15]. During the treatments, body weight (BW), food/water intake and any adverse response were monitored daily.

Measurement of systemic and portal hemodynamics
The right carotid artery was cannulated with a PE-50 catheter that was connected to a pressure transducer. Continuous recordings of mean arterial pressure and heart rate were performed on a multi-channel recorder (MP45, Biopac Systems Inc., Goleta, CA, U.S.A.). The external zero reference was placed at the level of the mid-portion of the rat. To measure portal pressure, the abdomen was then opened with a mid-line incision, and the mesenteric vein was cannulated with a 18G catheter connected to the transducer [16].
Superior mesenteric artery (SMA) was identified at its aortic origin and a 5-mm segment was gently dissected free from surrounding tissues. Then a pulsed-Doppler flow transducer (TS420, Transonic system Inc., Ithaca, NY, U.S.A.) was placed to measure the SMA flow [16]. Hepatic inflow via the portal vein was also measured by placing a flow probe around the portal vein as proximal to the liver as possible.
Cardiac output was measured by thermodilution, as previously described. 16 Briefly, a thermistor was placed in the aortic arch just distal to the aortic valve and the thermal indicator (100 μ L of normal saline) was injected into the right atrium through a PE-50 catheter. The aortic thermistor was connected to a cardiac output computer

Hepatic fibrosis determination with Sirius red staining
Liver paraffin section was stained with Sirius red staining kit (Polysciences Inc., Warrington, PA, U.S.A.). Image J was used to measure the percentage of Sirius redstained area. Briefly, grayscale image was used, then the stained red collagen was isolated using thresholding function. After that, the thresholded area was measured and shown as the percentage of thresholded area in the whole liver section. temperature. The image within the whole mesenteric window was thresholded by Image J software (available for download from the National Institutes of Health (http://rsb.info.nih.gov/ij/)). The mesenteric vascular density was shown as the percentage of thresholded area within the whole window.

SMA perfusion
The in situ perfusion technique was modified from the in vitro SMA perfusion [17].

Liver perfusion
The in situ perfusion system was performed as previously described with some modification [16]. Briefly, both jugular veins were cannulated with 16-gauge Teflon cannulas to ensure an adequate outflow without any resistance even at the highest flow rates. Heparin (200 U/100 g body weight) was injected through one of the cannulas.  Both the jugular vein cannulas were simultaneously opened to allow a complete washout of the blood. Pneumothorax was created by opening slits through the diaphragm. All the experiments were performed 15 minutes after starting perfusion at a constant rate of 40 ml/min.

Portosystemic collateral system perfusion
The in situ perfusion system was performed as previously described [18]. Briefly, both jugular veins were cannulated with 16-gauge Teflon cannulas. The abdomen was then opened and an 18-gauge Teflon cannula was inserted in the distal SMV and fixed with cyanoacrylate glue. To exclude the liver from perfusion, the second loose ligature around the portal vein was tied. All the experiments were performed 15 minutes after starting perfusion at a constant rate of 12 ml/min.

Portal-systemic shunts analysis: microsphere distribution method
Portal-systemic shunts was determined using the technique described by Chojkier and Groszmann and substituting color for radioactive microspheres; 30,000 of 15-μm yellow microspheres (Dye Track; Triton Technology, San Diego, CA, U.S.A.) was slowly injected into the spleen [19]. The rats were euthanized, and the livers and lungs were dissected and placed into new polypropylene centrifuge tubes. The number of microspheres in each tissue was determined following the protocol provided by the manufacturer. In brief, 3,000 blue microspheres (Dye Track) were added to each tube as an internal control. Tissue was digested overnight with 1 M KOH at 60 °C and thoroughly sonicated. After centrifugation, the supernatant was removed, and the pellets were washed once with 10 % Triton X-100 and twice with acidified ethanol. At the end of the process, a minimum pellets containing the Assuming a worst-case scenario in which two-thirds of the microspheres remain trapped in the spleen, this technique can detect a minimum shunt of 3.5 %. Studies using color microspheres have been shown to provide results similar to those using radioactive microspheres [20].

Western analysis
Tissue was immediately frozen in liquid nitrogen and stored at -80℃ until required.
The protein extracts were made by pulverization in grinder with liquid nitrogen, then a ratio of 1 ml of lysis buffer (phosphate-buffered solution containing 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS) was added.

Fecal sample collection, DNA extraction and 16S rRNA gene sequencing
16S rRNA is made up of conserved and hypervariable regions. The former are not significantly different across microbial strains, whereas the later are genus or speciesspecific, and they differ in accordance to phylogenetic difference. Therefore, 16S rDNA are frequently used to identify species and are important for microbial phylogeny and taxonomic identification.
Fecal samples were collected and fecal DNA was extracted according to the previous literature [21]. In brief, sample (250 mg) was suspended in 250 μl of guanidine thiocyanate, 0.1M Tris (pH 7.5), 40 μl of 10% N-lauroyl sarcosine, and 500 μl 5% N-lauroyl sarcosine. Mechanical disruption with beads were applied for DNA extraction. 2 μl of a 10-mg/ml solution of RNAase was added for RNA removal.
Alcohol precipitation was performed for nucleic acids recovery. Standard PCR was

Statistical analyses
The data other than sequence analyses were expressed as mean ± S.E.M. Statistical analyses was performed using repeated t-test (for in situ perfusion study) or one-way ANOVA as appropriate. LSD was used for the post-hoc test. SPSS version 21 software for Windows (SPSS Inc., Chicago, IL, U.S.A.) was used for analyses. Results is considered statistically significant at two-tailed P-values of less than 0.05.
For sequence analysis, pairwise comparisons were performed using OTU tables generated from each sample. Samples that contained fewer reads than the rarefaction depth were removed for the alpha and beta diversity analyses. Richness provided by alpha diversity was computed with Simpson and Shannon index. Sample clustering was performed using PCA methods based on UniFrac metrics. The Shapiro-Wilk test was used to check normality of the data. UniFrac weighted distance was analyzed by

Microbiota transplantation ameliorated portal hypertension
Rats received BDL to induce liver cirrhosis and portal hypertension. After antibiotics treatment, sham-operated (S) or BDL (B) rats randomly received vehicle as control

Microbiota transplantation did not influence the hepatic vasoconstriction and fibrosis
In hepatic system, hepatic vascular contractility and fibrosis are main components affecting portal hypertension. Figure 2A

Microbiota transplantation attenuated overt vasodilatation and angiogenesis in splanchnic system
The effects of microbiota transplantation on splanchnic system are shown in

Microbiota transplantation improved dysbiosis in BDL rats
The hypertension in a rat model of NASH. However, there was no liver cirrhosis in the rats [23]. Following that, there were several comprehensive reviews pointed out the relationship between microbiota and portal hypertension [24]. Nevertheless, the impact of MT on cirrhosis-induced portal hypertension and the relevant hemodynamic derangements has not been clearly addressed. In this study, microbiota transplantation with fecal or gut material effectively attenuated portal hypertension and portosystemic collaterals. In addition, splanchnic hyperdynamic circulation, mesenteric pathological angiogenesis and eNOS phosphorylation were ameliorated. However, hepatic vascular resistance, severity of liver fibrosis and hepatic vascular constriction were not affected by microbiota transplantation. Taken together, microbiota transplantation from both feces or gut effectively control portal hypertension and extrahepatic derangements.
Liver cirrhosis induced by BDL significantly changed microbiome. Compared to the sham group, BDL rats significantly lacked Lachnospiraceae. It is worth noting that microbiota transplantation did not restore microbiome in cirrhotic rats towards that of the sham condition. Instead, the microbiota shifted to a new status. In both transplantation groups, Bifidobacterium increased significantly. These findings were consistent with several FMT studies in cirrhotic patients [9][10]25]. Interestingly, microbiota transplantation from feces or gut exerted similar results in cirrhotic rats. In addition, there was no marked differences in the diversity of microbiota between the primarily regulated by NO via the eNOS pathway [26]. In this study, we found that BDL rats, as compared to sham rats, had significantly higher hepatic resistance. Liver fibrosis severity and related protein expressions, including α-SMA, TIMP and TGFβ also upregulated significantly in the BDL rats. On the other hand, eNOS phosphorylation downregulated significantly in BDL rats. Taken together, the structural and functional components alter in the hepatic system of BDL rats, which are compatible with the features of cirrhosis. However, the aforementioned changes could not be observed in BDL rats receiving microbiota transplantation. In addition, in situ liver perfusion showed that microbiota transplantation did not influence hepatic inflow [27]. It has been documented that VEGF, a potent proangiogenic factor, plays a pivotal role in the process. In splanchnic organs, VEGF and receptor overexpression have been identified in cirrhotic rats [28]. Inhibition of VEGF receptor significantly attenuated splanchnic vascularization and portosystemic collaterals in animal models [27]. NO, a vasodilator, has also been recognized as a potent proangiogenic factor [29]. Its role in splanchnic vasodilation and angiogenesis in portal hypertension has also been widely surveyed [30]. In this study, splanchnic blood flow and vasodilation decreased significantly in microbiota transplantation groups. In addition, eNOS phosphorylation was downregulated. In brief, microbiota attenuated angiogenesis and vasodilation in splanchnic system that resulted in reduction of portal pressure and was through at least partly, the inhibition of eNOS activation.
It is worth noting that portosystemic collaterals decreased in microbiota transplantation groups. In this study, shunting severity was evaluated by two methods.
The blood flow of SRS, the most prominent collateral vessel in cirrhotic rats, was measured directly. The results showed that SRS flow decreased significantly and consistently in both transplantation groups. Shunting severity was further evaluated by microsphere method. By injecting the microsphere into the spleen and measuring the distribution of microsphere in lung and liver, the shunting severity could be intraperitoneal injection or oral route. Nevertheless, intraperitoneal injection of TAA aggravates mesenteric angiogenesis and may interfere with the interpretation of the current findings on extrahepatic effects of the treatments. TAA delivered through oral gavage also inevitably impacts microbiota. On the other hand, the severity of portosystemic shunting in CCL 4 model is not prominent enough as compared to that of BDL [33]. Taken together, although the animal models of liver cirrhosis and portal hypertension have been extensively adopted and BDL is the one that fits the main theme of the current study the most, validation of the results in clinical study is absolutely necessary.
In conclusion, microbiota transplantations from both feces or gut effectively control portal hypertension, which is related to the amelioration of splanchnic hyperdynamic circulation, mesenteric angiogenesis and eNOS phosphorylation. The portosystemic shunting degree also reduced at the same time but the liver was unaffected. The clinical application of microbiota transplantation for the control of portosystemic collaterals-related complications deserves further investigation.   vasoconstrictors. There was no difference in hepatic vascular pressure change between the two groups. n = 6, 6 (C) Liver fibrosis severity evaluated by Sirius red stained area in whole section of the liver showed that liver fibrosis aggravated in BDL group. Microbiota transplantation did not affect the severity of liver fibrosis. Scale bars, left panels: 1 mm, right panels, 200 μm; n = 5, 6, 6, 6, 7; *P < .05, **P < .01, ***P < .001 higher SMA flow and lower SMA resistance as compared to SC group. SMA flow was reversed in BG group as compared to BC group. n = 6, 7, 7, 6, 6, 6, 7, 7 (B) The AVP (arginine vasopressin)-induced SMA territory vasoconstriction was significantly enhanced in cirrhotic rats receiving microbiota transplantation. n = 6, 6 (C) Splanchnic angiogenesis was determined by vascular density in whole frame of mesenteric window to minimize selection bias. The mesenteric vascular density increased significantly in BC group, which was attenuated by microbiota transplantations (BF, BG). Scale bar: 1 mm; n = 6, 7, 7, 6, 6, 6, 7, 7; *P < .05, **P < .01, ***P < .001 Clinical Science. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/CS20210602 microbiota transplantation significantly enhanced vascular contractility of collateral vessels. n = 6, 6; *P < .05, **P < .01, ***P < .001 There was no significant difference in liver fibrosis among three groups. Scale bars, left panels: 1 mm, right panels, 200 μm; *P < .05, **P < .01, ***P < .001 Mesenteric angiogenesis decreased significantly in BF group even delivered since the stage of liver fibrosis.
It is worth noting that antibiotics treatment exerted no observable effects in this experiment. n = 6, 7, 7; Scale bar: 1 mm. (B) The SMA morphology was further evaluated to determine the effects of microbiota transplantation on splanchnic vessels. The cross section of SMA was stained with α-SMA to highlight the smooth muscle layer. The SMA wall thickness and area increased significantly in BF group. The intensity of phospho-myosin over smooth muscle layer may reflect vasoconstriction. The result revealed that BF group had increased phospho-myosin intensity. Scale bar: 100 μm; n = 7, 7; *P < .05, **P < .01, ***P < .001