Candida Administration in 5/6 Nephrectomized Mice Enhanced Fibrosis in Internal Organs: An Impact of Lipopolysaccharide and (1→3)-β-D-Glucan from Leaky Gut

Uremic toxins and gut dysbiosis in advanced chronic kidney disease (CKD) can induce gut leakage, causing the translocation of gut microbial molecules into the systemic circulation. Lipopolysaccharide (LPS) and (1→3)-β-D-glucan (BG) are the major gut microbial molecules of Gram-negative bacteria and fungi, respectively, and can induce inflammation in several organs. Here, the fibrosis in the kidney, liver, and heart was investigated in oral C. albicans-administered 5/6 nephrectomized (Candida-5/6 Nx) mice. At 20 weeks post 5/6 Nx, Candida-5/6 Nx mice demonstrated increased 24 h proteinuria, liver enzymes, and serum cytokines (TNF-α, IL-6, and IL-10), but not weight loss, systolic blood pressure, hematocrit, serum creatinine, or gut-derived uremic toxins (TMAO and indoxyl sulfate), compared to in 5/6 Nx alone. The gut leakage in Candida-5/6 Nx was more severe, as indicated by FITC-dextran assay, endotoxemia, and serum BG. The areas of fibrosis from histopathology, along with the upregulated gene expression of Toll-like receptor 4 (TLR-4) and Dectin-1, the receptors for LPS and BG, respectively, were higher in the kidney, liver, and heart. In vitro, LPS combined with BG increased the supernatant IL-6 and TNF-α, upregulated the genes of pro-inflammation and pro-fibrotic processes, Dectin-1, and TLR-4 in renal tubular (HK-2) cells and hepatocytes (HepG2), when compared with LPS or BG alone. This supported the pro-inflammation-induced fibrosis and the possible LPS–BG additive effects on kidney and liver fibrosis. In conclusion, uremia-induced leaky gut causes the translocation of gut LPS and BG into circulation, which activates the pro-inflammatory and pro-fibrotic pathways, causing internal organ fibrosis. Our results support the crosstalk among several organs in CKD through a leaky gut.


Introduction
Chronic kidney disease (CKD) has been acknowledged as a significant global burden [1], leading to an accumulation of different metabolic substances known as "uremic toxins". Such toxins, which are primarily produced by food components or metabolic processes in the body, can lead to several consequences, including CKD progression, cardiovascular complications, and pulmonary disorders [2]. A few circulating uremic toxins, including trimethylamine N-oxide (TMAO), indoxyl sulfate, p-cresol sulfate, hippuric acid, and phenylacetic acid, are produced in the gastrointestinal GI tract, and are referred to as gut-derived uremic toxins [3]. The accumulated toxins in advanced CKD are compensatorily excreted into the intestinal system, which promotes the proliferation of pathologic bacteria, referred to as "gut dysbiosis" [4]. During gut dysbiosis, there is an increased production of gut-derived uremic toxins, and the toxins (both gut and non-gut derivatives) might damage intestinal epithelial tight junctions, which causes the so-called "gut leakage" or "leaky gut" or "gut translocation"-translocation of microbial molecules from the gut into the bloodstream [5]. Although gut-derived uremic toxins originate in the intestine, they can spread throughout the body and cause damage to various tissues, including the intestines, heart, liver, and kidneys [6][7][8]. The term "gut-kidney axis" refers to the vicious loop in which uremic toxin buildup and gut dysbiosis intensify as CKD progresses [9]. Particularly, when the microbial compounds, especially lipopolysaccharide (LPS; a major molecule on the cell wall of Gram-negative bacteria) and (1→3)-β-D-glucan (BG; the main component of the Candida cell wall), from the intestinal translocation activate their receptors, it leads to the enhanced inflammatory responses that further facilitate the progression of CKD [5]. Notably, Gram-negative bacteria and Candida albicans are the microbes with the most and the second most abundance in the human intestine. Therefore, increases in both molecules through a leaky gut might lead to clinical adverse effects [10].
However, fungi are less prominent in the mouse gut than in the human intestine, and the influence of intestinal fungi in mouse models is possibly underestimated [11]. Contrary to what has been found in human feces [12], mouse feces do not have enough Candida spp. to be detectable in stool cultures [11]. Although the presence of Candida spp. in the gut does not directly cause disease, gut fungi alter the gut microbiota and provide a higher BG in gut content [13], which possibly worsens systemic inflammation following a gut barrier malfunction (gut leakage) through the systemic immune responses against BG [14][15][16][17][18][19][20]. Indeed, the possible fungal mechanisms that interfere with the growth of specific bacteria (mostly the lower virulence) in the gut are mentioned, including some bactericidal molecules (yeast killer toxins, Candida exotoxin, and endogenous alcohol) [21][22][23][24], and the competition for certain nutrients [25]. Meanwhile, Candida spp. possibly facilitate the growth of specific bacteria with the glucanase enzyme, which can digest BG from the fungal cell wall [26]. Despite the possibility of both symbiotic and antagonistic impacts of Candida on gut bacteria, gut fungi alter microbial homeostasis, and the oral administration of Candida causes gut dysbiosis, with leaky gut and inflammatory reactions observed in several mouse models [27][28][29][30]. Indeed, explorations into the impacts of gut fungi through oral administration in mice are possible due to the naturally lower abundance of fungi in the mouse gut. Indeed, the increase in fecal fungi after oral administration in mice is reported [31,32].
Gut translocation of LPS and BG causes systemic inflammation [10], and chronic systemic inflammation possibly facilitates fibrosis in several organs as increased serum BG in chronic alcoholism (humans and mice) enhances liver fibrosis, as previously demonstrated [33]. Interestingly, the activation of Dectin-1, a recognition receptor of BG, induces fibrosis in several organs (heart, liver, and lung) [34][35][36][37]. For kidney fibrosis, there are only a few investigations on the impact of gut fungi and BG on CKD. As such, our previous publication demonstrates no difference in renal fibrosis between in CKD mice with or without Candida gavage at 16 weeks after 5/6 nephrectomy (5/6 Nx), despite a synergistic pro-inflammatory effect of LPS plus BG [18]. However, the pro-fibrotic impact of LPS and BG on other organs, and the influence on the kidney after extended observation, might be different. Due to (i) the worsening renal fibrosis and CKD progression by systemic inflammation [38] and endotoxemia [39,40], (ii) the induction of gut leakage by systemic inflammation and uremia [5] with the attenuation of leaky gut by probiotics [28][29][30][31][32][33][34][35][36][37][38][39][40][41][42], and (iii) the association among kidney and other internal organs (gut-liver-kidney and gutheart axis) [10,43], the presence of fungi in the gut with uremia-induced leaky gut might worsen fibrosis in several organs.
Here, 5/6 Nx and Candida gavage were observed for 20 weeks with the exploration of fibrosis in other internal organs (liver and heart) along with in vitro experiments. To understand the pathophysiologic effects of fungi on fibrosis, BG and LPS (the main cell wall components of fungi and Gram-negative bacteria, respectively) were tested on hepatocytes (HepG2 cells) and renal tubular cells (HK2 cells).

Results
2.1. The Presence of Candida in the Uremic Gut Enhanced Injury in Internal Organs (Kidney, Liver, and Heart) through Systemic Inflammation Despite the lack of mortality among 5/6 Nx mice regardless of Candida administration during the 20 weeks observation, more prominent systemic inflammation and organ injuries in Candida-administered 5/6 Nx mice, compared with 5/6 Nx alone, were demonstrated by the increased 24 h proteinuria (a more sensitive renal injury biomarker than serum creatinine) [44], liver enzymes (alanine transaminase), and serum cytokines (TNF-α, IL-6, and IL-10), but not weight loss, systolic blood pressure, hematocrit, serum creatinine, or gutderived uremic toxins (TMAO and indoxyl sulfate) ( Figure 1A-K). The enhanced systemic inflammation might partially be due to the more severe leaky gut in Candida-administered 5/6 Nx mice than in 5/6 Nx alone, as indicated by FITC-dextran assay, endotoxemia, and increased serum BG (Figure 2A-C). Notably, the molecular weight (MW) of FITCdextran is 4.4 kDa, while the MWs of LPS and BG were higher than 0.5-5 kDa, which normally cannot pass through the enterocyte tight junction [10]. In parallel, the increased presentation of LPS and BG, the pathogen-associated molecular patterns (PAMPs) with foreignness properties against the host cells [10], upregulated gene expression of TLR-4 and Dectin-1, the pattern recognition receptor of LPS and BG, respectively, in kidney, liver, and heart, along with the enhanced fibrosis in these organs (Figures 2D-L and 3-5), supporting systemic inflammation-induced injury in the kidney, liver, and heart [45][46][47].

An Additive Effect of Lipopolysaccharide (LPS) and (BG) on Inflammation-Induced Fibrosis
Although inflammatory maladaptation from uremia and LPS alone can cause fibrosis and organ dysfunction [6,48,49], an additive pro-inflammatory impact of BG in fibrosis might also be important. Indeed, LPS plus BG (LPS + BG) induced the highest supernatant IL-6 and TNF-α, with more prominent upregulation of the IL-8 gene when compared with LPS or BG alone in renal tubular cells ( Figure 6A-C). Additionally, the enhanced fibrosis after LPS-induced inflammation was demonstrated through the upregulated HIF-1α with collagen type III and IV, but not TGF-β, fibronectin-1, α-SMA, or collagen type I, in renal tubular cells (HK2) ( Figure 6D-J). However, the additive effect of BG on LPSinduced fibrosis was indicated only by the upregulation of collagen type III, but not by other genes in these cells ( Figure 6D-J). There were subtle responses against BG alone only in supernatant TNF-α and upregulated fibronectin and collagen type III without other parameters in HK2 cells ( Figure 6A-J). Interestingly, there was an upregulation of Dectin-1 (a receptor for BG) after induction by either LPS or BG, while TLR-4 was upregulated only with LPS and LPS + BG stimulation ( Figure 6K,L), implying a possible LPS-BG synergy through the enhanced BG responses from the LPS-induced Dectin-1 upregulation in renal tubular cells.
Likewise, LPS and BG also enhance inflammation in hepatocytes because LPS and BG from gut translocation are directly transported to the liver through the portal vein [10]. Here, at 1 h after stimulation, supernatant TNF-α, but not other parameters, was elevated, which was more prominent in LPS + BG than the activation by each molecule alone in hepatocytes (HepG2 cells) ( Figure               Oral Candida administration in mice is used to examine the influence of gut fungi in several models, taking advantage of the lower abundance of C. albicans in mouse feces compared with humans (Candida in mouse feces is detectable only by polymerase chain reaction (PCR) [50], but not by culture methods [11], which differs from human conditions [12]. Here, Candida-5/6 Nx mice had more severe gut leakage (FITC-dextran assay), with a higher level of serum endotoxin and BG (glucanemia) compared to non-Candida 5/6 Nx mice, perhaps due to the direct intestinal invasion of fungi, gut dysbiosis, and proinflammatory enterocytes [18]. Although enterocytes are naturally resistant to pathogen molecules (LPS and BG), the presence of certain uremic toxins enhances the inflammatory responses in Caco-2 cells (an enterocyte cell line), which might cause a leaky gut [42]. Despite Candida-induced gut dysbiosis in uremic mice, as previously described [42], only serum BG, but not gut-derived uremic toxins and endotoxemia, in Candida 5/6 Nx were higher than non-Candida 5/6 Nx, and serum BG might be the main factor responsible for the hyper-inflammation in Candida uremic mice. Additionally, the inflammatory activation of LPS and BG in the blood of 5/6 Nx mice might be responsible for the increased proteinuria, elevated liver enzyme, and organ damage (liver, kidney, and heart), as indicated by histology. Although Candida did not alter systolic blood pressure in 5/6 Nx mice, cardiac fibrosis was more prominent in Candida-5/6 Nx than 5/6 Nx alone, implying an impact of some non-blood pressure-related profibrotic factors possibly, which was due to the direct myocardial activation by LPS and BG. An upregulation of TLR-4 and Dectin-1 in these organs (heart, liver, and kidney) supported an inflammatory synergy of LPS and BG in Candida-5/6 Nx mice. Likewise, enhanced fibrosis in the liver might be correlated with LPS-BG from a leaky gut that was directly transported to the liver through the portal vein [51,52]. Notably, the duration of Candida administration is important as renal fibrosis at 20 week, but not at 16 weeks of the experiment [18], demonstrated higher fibrosis than the 5/6 Nx control. The enhanced level of BG in serum after Candida administration (both heat-killed and viable cells) was also previously demonstrated [31]. It is interesting to note that persistent chronic and severe acute inflammation can frequently lead to maladaptation with organ fibrosis instead of the regular healing process [53]. Here, the chronic inflammation from LPS and BG in serum for 20 weeks is possibly potent enough to generate fibrosis in several internal organs in Candida-5/6 Nx mice, supporting the importance of gut fungi in the condition with systemically chronic inflammation. Thus, pro-fibrosis of the internal organs in other conditions with chronic inflammation from leaky gut, such as obesity, inflammatory bowel disease, and cirrhosis, would be interesting to explore further.

The Additive Inflammatory Effect of LPS Plus BG, a Key Pro-Inflammatory Factor in Leaky Gut
The synergy or additive effect of BG presentation upon LPS responses is intensively demonstrated through the activation of several innate immune cells (neutrophils, macrophages, and dendritic cells) [5,17,30,[54][55][56][57][58][59][60][61], possibly with the crosslink between TLR-4 and Dectin-1, the pattern recognition receptors (PRRs) for LPS and BG, respectively [62]. The similar downstream signaling through the NF-κB transcription factor of TLR-4 and Dectin-1 in these immune cells may be another explanation of the LPS-BG additive impact on several organs through the activation of immune cells (especially macrophages) inside these organs. Interestingly, trained immunity (enhanced immune response to a second unrelated challenge) through preconditioning BG administration before LPS activation, which is more profound than LPS activation without preconditioning, is also referred to as the LPS-BG synergistic pro-inflammatory effect [63]. However, TLR-4 and Dectin-1 are not only present in immune cells, but also in other parenchymal cells (kidney, liver, and heart) [64][65][66][67][68], which are additively associated with profound inflammation and fibrosis in these organs [34,[69][70][71]. Here, the additive inflammation in LPS-BG in hepatocytes might be due to the upregulation of Dectin-1 by LPS, possibly responsible for the higher responses against BG when simultaneously presented with LPS. The LPS-BG responses of hepatocytes were severe enough for the more profound upregulation of HIF-1α and fibronectin, the important liver profibrotic genes [72,73], when compared with LPS activation alone. On the other hand, the additive effect of LPS-BG over LPS alone in renal tubular cells was different from the hepatocytes because LPS did not upregulate hepatic Dectin-1, but the inflammatory re-sponses against LPS-BG were potent enough for a more prominent collagen type III upregulation compared with LPS activation alone [74]. Hence, the underlying mechanisms of the additive LPS-BG pro-inflammatory effect might differ among various organs. Nevertheless, our in vitro data support a possible impact of gut fungi in uremia, partly through the translocation of BG from the gut into the blood circulation, which sub-sequentially facilitated LPS-induced systemic inflammation, possibly through various mechanisms.

Clinical Aspect and Future Experiments
Our data support the crosstalk among several organs in CKD through leaky gut, which might be a common situation in several medical conditions (Figure 8). Therefore, measurement of the abundance of LPS and BG in gut content, perhaps through PCR or culture, might be an interesting factor in predicting the severity of systemic inflammation during leaky gut. Additionally, our results support the importance of gut fungi in uremic conditions, especially in patients with severe uremia, before performing kidney replacement therapies (dialysis and renal transplantation). In patients with CKD with positive residual renal function (still producing urine), the determination of leaky gut using an oral administration of non-absorbable carbohydrate before detection in the urine [15,75] might be an interesting biomarker that demonstrates a possibly persistent chronic inflammation from uremia-induced leaky gut. The interventions for the attenuation of leaky gut, such as probiotics and other renal replacement therapies, act through the reduction in uremic toxins (gut-derived and non-gut-derived toxins). The earlier replacement therapies in patients with constant uremia-induced leaky gut might avoid unnecessary leaky-gut-induced hyperinflammation that will lead to injury and fibrosis in several internal organs. Further studies on these topics are warranted for an adaptation of leaky gut in real clinical practice.

Animals and Animal Model
The approved protocol of the Institutional Animal Care and Use Committee of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand (CU-ACUP No. The proposed working hypothesis of the enhanced fibrosis from chronic kidney disease through leaky-gut-induced pro-inflammation that is activated by lipopolysaccharide (LPS) together with (1→3)β-D-glucan (BG). Firstly (1), the increased uremic toxins in the intestine, and the substitution of kidney excretion, affects gut organisms and causes the intestinal barrier defect. Secondly (2), the LPS and BG from the gut translocate into the blood (leaky gut). Thirdly (3), LPS and BG additively activate pro-inflammation in several organs, partly through the activation of Toll-like-receptor 4 (TLR-4; LPS receptor) and Dectin-1 (BG receptor). The enhanced inflammatory responses from the leaky gut during uremia possibly worsen several uremic complications. The diagram was created using Biorender.com.

Animals and Animal Model
The approved protocol of the Institutional Animal Care and Use Committee of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand (CU-ACUP No. 018/2562), according to the National Institutes of Health (NIH) criteria [76], using 8-weekold male C57BL/6 mice, purchased from Nomura Siam (Pathumwan, Bangkok, Thailand), was followed.

Candida-Administered Chronic Kidney Disease Model
First, 5/6 nephrectomy (5/6 Nx) surgery was performed via flank approach under isoflurane anesthesia by removing the upper and lower poles of the left kidney before the right nephrectomy 1 week later, as previously described [18,42]. Only mice with a weight of the removed fragments from the left kidney to the right kidney weight in a ratio of 0.55-0.72 were included to ascertain that removal of the left kidney mass is optimum for CKD development [77]. After that, the 5/6 Nx mice were orally administered phosphate buffer solution (PBS) (5/6 Nx PBS) or Candida albicans (5/6 Nx + Candida) to explore the impact of gut fungi using C. albicans from the American Type Culture Collection (ATCC 90028) (Fisher Scientific, Waltham, MA, USA), which was cultured overnight on Sabouraud dextrose broth (SDB) (Oxoid, Hampshire, UK) at 35 • C for 48 h before enumeration using a hemocytometer. The C. albicans at 1 × 10 6 CFU in 0.5 mL PBS or PBS alone was orally administered on alternate days starting from 4 weeks after the right nephrectomy until 20 weeks. Another group of mice underwent the sham operation to identify renal vessels before abdominal closure (Sham group). The Candida-administered sham mice have not been demonstrated here because of the lack of difference from the sham control reported in a previous publication [5].

Gut Permeability Determination
Because the detection of fluorescein isothiocyanate (FITC)-dextran (an intestinal nonabsorbable carbohydrate) in serum after an oral administration, or the spontaneous elevation in serum of LPS or BG without systemic infection, indicate gut permeability defects (gut leakage), these parameters were used as previously described [30,58,84,85]. As such, 12.5 mg of FITC-dextran (4.4 kDa) (FD4; Sigma-Aldrich ® , St. Louis, MO, USA) was orally administered before measuring FITC-dextran in serum 3 h later by fluorospectrometer (NanoDrop™ 3300; Thermo Fisher Scientific™, Wilmington, DE, USA). Meanwhile, serum LPS and BG was evaluated by HEK-Blue LPS Detection Kit 2 (InvivoGen™, San Diego, CA, USA) and Fungitell ® assay (Associates of Cape Cod, Falmouth, MA, USA), respectively, and values less than 0.01 EU/mL (for LPS) and 7.8 pg/mL (for BG) were recorded as 0, due to the lower limit of the standard curve of the test.

The In Vitro Experiments
To explore a possible pro-fibrotic effect of LPS and BG (a major cell wall component of Gram-negative bacteria and Candida, respectively, on kidneys and liver, HK2 renal proximal tubular cells (ATCC 237 CRL-2190) and HepG2 hepatocytes (American Type Culture Collection, Manassas, VA, USA) were used as the representatives. Briefly, HK2 cells or HepG2 at 1 × 10 6 cells/well maintained in Dulbecco's modified Eagle Medium (DMEM) were incubated LPS (Escherichia coli O26:B6) (Sigma-Aldrich) at 100 µg/mL with or without BG (CM-Pachyman) (Megazyme, Bray, Ireland) at 100 µg/mL for 24 h (HK2 cells) or 4 and 72 h (HepG2 cells) under 5% CO 2 at 37 • C before the determination of supernatant cytokines (TNF-α and IL-6) by ELISA (Invitrogen, Carlsbad, CA, USA). Additionally, the expression of several genes related to inflammatory responses and fibrosis was examined by quantitative reverse transcription polymerase chain reaction (qRT-PCR) relative to βactin (a house-keeping gene) with the 2 −∆∆CT method using cDNA (SuperScript™ Vilo™ cDNA synthesis assay) (Invitrogen™) prepared from 50 ng of TRIzol-extracted total RNA (invitrogen™) by a qPCR machine (LightCycler ® 2.0) (Roche Diagnostics) with the primers (Table 1).

Statistical Analysis
Analyzed data are presented as mean ± standard error (SE), determined using Graph-Pad Prism version 9.0 software (La Jolla, CA, USA). Statistical significance was determined by one-way analysis of variance (ANOVA) followed by Tukey's analysis. The time-point experiments were analyzed by repeated measures ANOVA. All statistical analyses were performed with Stata 16.0 software (StataCorp, TX, USA) and Graph Pad Prism version 7.0 software (La Jolla, CA, USA). A p-value of <0.05 was considered statistically significant.

Conclusions
Our results suggest the crosstalk among several organs in CKD through a leaky gut. Uremia-induced leaky gut in advanced CKD causes the translocation of LPS and BG from the gut into the systemic circulation. As such, LPS and BG additively induce the proinflammatory and pro-fibrotic pathways through the activation of TLR-4 and Dectin-1, causing internal organ fibrosis. These enhanced inflammatory responses during leaky gut could worsen uremic complications in patients with advanced CKD. Gut abundance of Gram-negative bacteria and fungi and/or the levels of LPS and BG in the blood might be interesting factors in predicting the severity of systemic inflammation during CKD-induced leaky gut.  Institutional Review Board Statement: The study was conducted in accordance with the National Institutes of Health (NIH) criteria, and the animal study protocol was approved by the Institutional Animal Care and Use Committee of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand (CU-ACUP No. 018/2562).