bs_bs_banner
doi:10.1111/jgh.13042
GASTROENTEROLOGY
Neutrophil gelatinase-associated lipocalin regulates gut microbiota of mice Katsuya Mori,*,† Takeshi Suzuki,* Shizuka Minamishima,* Toru Igarashi,* Kei Inoue,* Daisuke Nishimura,* Hiroyuki Seki,* Takashige Yamada,* Shizuko Kosugi,* Nobuyuki Katori,* Saori Hashiguchi* and Hiroshi Morisaki* *Department of Anesthesiology, Keio University School of Medicine and † Japan Society of the Promotion of Science, Tokyo, Japan
Key words bacteria, gut, lipocalin-2, microbiota, toll-like receptor. Accepted for publication 25 June 2015. Correspondence Dr H. Morisaki, Department of Anesthesiology, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan. Email [email protected] Conflict of interest All authors have nothing to declare any financial or personal conflict of interest that might influence the results or interpretation of our manuscript.
Abstract Background and Aim: Because neutrophil gelatinase-associated lipocalin (NGAL) is known to provide significant bacteriostatic effects during infectious conditions, we tested the hypothesis that this protein is up-regulated and secreted into the intraluminal cavity of the gut under critically ill conditions and is thus responsible for the regulation of bacterial overgrowth. Methods: With our institutional approval, male C57BL/6J mouse (6–7 weeks) were enrolled and applied for lipopolysaccharide or peritonitis model compared with naïve control. We assessed NGAL protein concentrations in intestinal lumen and up-regulation of NGAL expression in intestinal tissues in in vivo as well as ex vivo settings. Simultaneously, we examined the effects of NGAL protein administration on the growth of Escherichia coli (E. coli) in in vivo and in vitro experimental settings. The localization of NGAL in intestinal tissues and lumen was also assessed by immunohistological approach using NGAL antibody. Results: Both lipopolysaccharide and peritonitis insults evoked the marked up-regulation of NGAL mRNA and protein levels in gut tissues such as crypt cells. In addition, the administration of NGAL protein significantly inhibited the outgrowth of enteric E. coli under both in vitro and in vivo conditions, accompanied by histological evidence. Conclusion: Neutrophil gelatinase-associated lipocalin protein accompanied by apparent bacteriostatic action accumulated in the intestinal wall and streamed into the mucosal layer during critically ill state, thereby possibly shaping microbiota homeostasis in the gut.
Introduction 3+
Bacterial species need Fe for their growth and thereby possess an independent iron-uptake system and means of combating host iron-withholding defenses in gut to overcome the host’s ironwithholding potential.1 For example, siderophores produced by bacteria are a family of low-molecular weight iron chelators that play a pivotal role in the iron acquisition system of bacteria. Neutrophil gelatinase-associated lipocalin (NGAL) is up-regulated in several disease conditions including inflammation, ischemia, and cancer.2 Because the intensity of NGAL expression depends on the severity of disease, its detection in clinical specimens has been considered to be a diagnostic biomarker of several illnesses.2 On the other hand, NGAL exerts a bacteriostatic action by preventing bacterial growth from undergoing iron-siderophore uptake.3,4 A previous study showed that several gram-negative bacteria,5,6 induce the up-regulation of NGAL in the lung, liver, and systemic circulation, resulting in the strict regulation of bacterial overgrowth. Thus, NGAL is regarded to be essential in innate immune functions in hosts. Approximately 107 commensal bacteria, consisting of over 1000 species in the gut8 and called microbiota, are also essential for the
development of intestinal immune systems. Their growth is strictly regulated by immune function through antimicrobial peptides such as defensins and cathelicidins.9 Although the disruption of microbiota, such as the overgrowth of pathogenic bacteria, can be prevented by these immune functions, the processes could conversely augment inflammation by enabling passage through the intestinal epithelial barrier, thereby stimulating immune responses under critically ill conditions.10 Taken together, these findings suggest that the imbalance of microbiota could be associated with organ dysfunction and a poor outcome in critically ill patients because of gut-barrier dysfunction.11 In the present study, we examined whether NGAL contributes to the preservation of microbiota homeostasis and the development of gut immune function in critically ill state.
Methods This study protocol was approved by the Animal Care and Use Committee of Keio University School of Medicine in accordance with the National Institute of Health guidelines. The details of Methods section were described in Supplemental Materials.
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
145
K. Mori et al.
Lipocalin and gut immune function
Animal models and analyses of NGAL protein concentrations in intestinal lumen. Male mice (aged 6–9 weeks) were used as endotoxemia and peritonitis models to evoke a systemic inflammatory response. For the former model, lipopolysaccharide (LPS) was administered, and for the latter model, we applied cecal ligation and perforation (CLP) to evoke peritonitis.7,12 The contents of the ileum and colon were collected and homogenized, and the supernatant was used to measure the NGAL protein concentration, which was detected by ELISA-based assay. Histological analyses of NGAL and mucin in ileum and colon tissues. To identify NGAL-secreting cells in the ileum and colon, 5-μm-thick paraffin sections of each tissue were examined using immunohistochemistry. To identify the mucinsecreting cells in the gut, the remaining sections were stained with Alcian blue and hematoxylin staining, and the results were observed under a light microscope by an independent examiner. Isolation of gut crypt cells. To isolate gut crypt cells including Paneth cells, ileum and colon crypt cells were isolated using modified procedures, as described previously.13 A portion of the pellets were re-suspended in Hank’s balanced salt solution (HBSS) buffer and were stained with 0.25% amido black to label the secretory granules of Paneth cells in the crypts; the results were then observed using light microscopy. Stimulation of crypt cells ex vivo. To analyze the NGAL expression in isolated Paneth cells, crypt cells obtained from the ileum and colon were incubated with toll-like receptor (TLR4)-, TLR2-, and TLR9-ligands. After exposure to each TLR ligand, the stimulated crypt cells were deposited by centrifugation at 700 xg, and the total RNA was eluted. Analysis of Escherichia coli growth in vitro. To analyze the impact of NGAL on the growth of E. coli, feces were collected and cultured with 100 μL of phosphate-buffered saline (PBS) containing 125, 250, 500, or 1000 ng of recombinant NGAL/lipocalin 2. The number of bacteria colonies was then counted. Persistent NGAL infusion model in LPS-treated and CLP-treated animals. After intraluminal placement of a catheter, the mice were divided into control, LPS NS group and LPS NGAL group. At 24-h study period, the contents of the cecum and colon were collected for bacterial DNA analysis. As a separate series of experiments, the mice were divided into control, LPS NS group and LPS NGAL group after intra-abdominal placement of a catheter. To examine the survival time in a CLP model, normal saline or recombinant NGAL protein was administered via an inserted catheter. RNA/DNA preparation, reverse transcription, and real-time polymerase chain reaction. Total RNA from the ileum, colon, and crypt cells were extracted using a RiboPureTM kit. The quantity of eluted total RNA were measured using a NanoDrop 1000A spectrophotometer. For the removal of
146
contaminating DNA from the isolated RNA, deoxyribonuclease treatment was performed using a DNA-free kit. Complementary DNA synthesis from the total RNA was performed using a ReverTra Ace qPCR RT kit with a random primer. For the measurement of NGAL and TLR family expression, semi-quantitative real-time polymerase chain reaction (PCR) was performed using TaqMan-based Applied Biosystem gene expression assays. For the isolation of bacterial DNA, the ileum and colon contents were washed with PBS, and then extracted using the QIAamp DNA Stool kit. E. coli was quantified using TaqMan-based real-time PCR with a Quantification of E. coli standard kit. Statistical analyses. All data are expressed as mean ± standard deviation unless otherwise specified. Analyses were performed using the SPSS/21.0J (SPSS Inc., Chicago, IL, USA). Expression of NGAL at different LPS concentrations and enteric E. coli DNA copy numbers were analyzed by one-way analysis of variance (ANOVA), followed by Tukey’s post-hoc test. Mann–Whitney test was applied to compare expression of TLR family, NGAL, and alpha-defensin mRNA. Kaplan–Meier analysis and log-rank test were used for survival data. Differences were considered statistically significant if P value was less than 0.05.
Results NGAL expression in intestinal tissue in critically ill models. Lipopolysaccharide administration markedly increased the expression of NGAL mRNA in both tissues in a time-dependent manner, compared with the levels in a treatment-naïve animal (Fig. 1a,b). In particular, NGAL expression in the ileum tissue was approximately 7000-fold higher after a 24-h study period. Neutrophil gelatinase-associated lipocalin mRNA was significantly up-regulated in both the ileum and colon, compared with a sham-treated animal, at 24 h after the CLP insult (Fig. 1c,d). Furthermore, we showed that LPS-induced insults significantly induced the expression of NGAL mRNA in a dose-dependent manner, with a peak typically observed at 24 h, in both the ileum and colon tissues (Fig. 1e,f). While the discharge of NGAL protein was not detected in the intraluminal contents of either the ileum or the colon in treatment-naïve mice at 24 h, LPS insults evoked the up-regulation of NGAL discharge in both tissues in a dose-dependent manner (Fig. 1g,h). Localization of NGAL secretory cells in gut. In the ileum mucosa of treatment-naïve mice, NGAL-positive cells were found on a few Paneth cells as secretory granules in crypts (Fig. 2a; black arrowhead), but were not found in epithelial cells of villi (Fig. 2a). In contrast, LPS injection strongly induced the upregulation of NGAL in epithelial cells (Fig. 2b; black arrow) and crypts, including Paneth cells of the ileum mucosa (Fig. 2b; black arrowhead). Of note, NGAL evoked by LPS accumulated in the luminal area of the crypt cells (Fig. 2b; white arrow), whereas no accumulation was found in treatment-naïve mice. In the colon, goblet cells in treatment-naïve crypt cells were positive for NGAL (Fig. 2c; white arrowhead), whereas NGAL expression was slightly detected in a few epithelial cells from treatmentnaïve colon tissue (Fig. 2c; black arrow). After an LPS insult, the epithelial cells of the colon mucosa were stained more clearly with
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
K. Mori et al.
Lipocalin and gut immune function
Figure 1 Different insults induced the up-regulation of NGAL in the ileum and colon tissue. The expression of NGAL mRNA up-regulated by LPS in the ileum (a) and colon (b) was analyzed using semi-quantitative real-time PCR at different time points. The values of treatment-naïve animals were set at 1.00, and the data show the fold-increase as the mean ± SD. The expression of NGAL mRNA induced by CLP in the ileum (c) and colon (d) was analyzed using semi-quantitative real-time PCR after a 24-h treatment period. The values of sham animals were set at 1.00, and the data show the fold-increase as the mean ± SD. The expressions of NGAL mRNA mediated by different doses of LPS in the ileum (e) and colon (f) were analyzed using semi-quantitative real-time PCR after a 24-h treatment period. The values of treatmentnaïve animals (n = 3) were set at 1.00, and the data show the fold-increase as the mean ± SD. Different doses of LPS-induced discharge of NGAL protein in the ileum lumen (g) and colon (h) were measured using an ELISA. The data were expressed as the mean ± SD. The sample size of all the groups was 3. Abbreviations: CLP, cecum ligation and perforation; LPS, lipopolysaccharides; NGAL, neutrophil gelatinase-associated lipocalin; PCR, polymerase chain reaction; SD, standard deviation.
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
147
Lipocalin and gut immune function
K. Mori et al.
Figure 2 Representative pictures of NGAL localization in gut mucosa with or without insults. NGAL-positive cells were detected in the intestinal mucosa using immunostaining. NGAL-positive cells were stained brown. All scale bars in panel (a to g) indicate 50 μm. Upper panel (a) shows a representative villus in the ileum of a treatment-naïve mouse, whereas lower panel (a) shows crypt cells in the ileum. Upper panel (b) shows a villus in the ileum of an endotoxemic mouse, whereas lower panel (b) shows crypt cells in the ileum. The up-regulation of NGAL (brown) induced by LPS was apparent in both the epithelial and crypt cells (upper and lower panel (b)), compared with that in a treatment-naïve mouse (upper and lower panel (a)). Of note, NGAL accumulation was apparent in the luminal area of the crypt cells (arrow, lower panel (b)), whereas no accumulation was found in a treatmentnaïve mouse lower panel (a). Upper panel (c) shows a representative villus in the colon of a treatment-naïve mouse, whereas lower panel (c) shows crypt cells in treatment-naïve colon. Upper panel (d) shows a villus in the colon of an endotoxemic mouse, whereas lower panel (b) shows crypt cells in the colon. Compared with the ileum, the up-regulation of NGAL (brown) was visible even in the treatment-naïve mouse. Additionally, LPS apparently evoked an up-regulation of NGAL in the colon similar to that occurring in the ileum. Panel (e) showed representative pictures of crypt cells in the ileum (left panel) and colon (right panel). In both sections, the crypt cells were opened at the apex and accumulated NGAL (brown) was discharged into the lumen of the intestine after the LPS insult. Panel (f) shows that mucus stained by Alcian blue (lower panel) completely matched the accumulation of NGAL in the crypt cells of the ileum in an LPS-treated mouse. Dotted lines in the upper and the lower panels indicate that location of NGAL (brown) corresponded to Alcian blue staining in crypt cells. Panel (g) shows that similar findings were found in the colon. Dotted lines in the upper and the lower panels indicate that location of NGAL (brown) corresponded to Alcian blue staining in crypt cells. Abbreviations: LPS, lipopolysaccharides; NGAL, neutrophil gelatinase-associated lipocalin.
148
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
K. Mori et al.
Lipocalin and gut immune function
Figure 3 Expression of TLR family, NGAL, and alpha-defensin mRNA in isolated crypt cells. Panel (a) shows that cells isolated with EDTA were crypt cells. Paneth cell in crypt cells were stained with 0.25% amido black (black arrow). Bar means 50 μm. Panels (b–d) show the expressions of TLR 2 (b), TLR 4 (c), and TLR 9 (d) mRNA in crypt cells stimulated by each TLR ligand. Ileum (left panel) and colon (right panel) crypt cells from the control group (n = 3) and the TLR groups (n = 3 per each TLR) were treated with normal saline and TLR ligand (TLR 2, zymosan; TLR 4, LPS; and TLR 9, CpG-DNA), respectively. The data show the fold increase as the mean ± SD. Panel (e–g) shows the expression of α-defensin mRNA in ileum (left panel) and colon (right panel) crypt cells, which were stimulated with the TLR 2 (e), TLR 4 (f), or TLR 9 (g) ligands. The sample size of each group was 3, and the data show the fold-increase as the mean ± SD. Panel (h–j) shows the expression of NGAL mRNA in ileum (left panel) and colon (right panel) crypt cells, which were stimulated with the TLR 2 (h), TLR 4 (i), and TLR 9 (j) ligands. The sample size of each group was 3, and the data show the fold-increase as the mean ± SD. Abbreviations: EDTA, ethylenediaminetetraacetic acid; LPS, lipopolysaccharides; NGAL, neutrophil gelatinase-associated lipocalin; SD, standard deviation; TLR, .
the NGAL antibody (Fig. 2d; black arrow). While the expression of NGAL was observed in the crypts of treatment-naïve colon mucosa, LPS stimulation led to the accumulation of NGAL in goblet cells and crypts in colon (Fig. 2d; white arrowhead and white arrow). As shown in Figure 2e, several LPS-stimulated crypt cells from both the ileum and the colon were opened at the apex of these
cells, and a thick, positive region of NGAL simultaneously streamed from the crypts into the intestinal lumen (Fig. 2e; white arrow). We also found that the contents of the ileum crypt lumens, which stained positive for NGAL antibody, were simultaneously stained with Alcian blue (Fig. 2f). A blue signal from Alcian blue was accumulated
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
149
Lipocalin and gut immune function
Figure 3
150
K. Mori et al.
(Continued)
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
K. Mori et al.
Lipocalin and gut immune function
Figure 4 Effects of recombinant NGAL on enteric Escherichia coli and survival rate of peritonitis model. Panel (a) shows that NGAL suppressed the growth of enteric cultured E. coli in a dose-dependent manner except for 500 ng in vitro. Panel (b) shows the intestinal fistula model; a catheter was inserted into the cecum, and recombinant NGAL protein was infused via a catheter for 24 h. Panels (c) and (d) show the copy number of E. coli in the cecum and colon contents. Data are shown for the control, LPS with normal saline infusion (LPS-NS), and LPS with 15 μg of recombinant NGAL infusion (LPS-NGAL) groups (n = 3, each group). The in vivo administration of recombinant NGAL protein significantly inhibited the overgrowth of E. coli evoked by LPS injection in both the cecum and colon. Data are shown as the mean ± SD. Panel (e) shows the Kaplan–Meier curve of the survival time in an animal model of CLP-induced peritonitis that was treated with recombinant NGAL (n = 10, each group). , Control; , NGAL. Treatment with NGAL in the peritonitis mouse model did not improve the survival rate. Abbreviations: CLP, cecum ligation and perforation; LPS, lipopolysaccharides; NGAL, neutrophil gelatinase-associated lipocalin; NS, normal saline; SD, standard deviation.
in crypt lumen of both ileum and colon (Fig. 2f,g; white arrow) and goblet cells of colon (Fig. 2g; white arrowhead) that exhibited positive signals for NGAL; simultaneously, a thick positive region streaming from the crypts into the intestinal lumen was also apparent (Fig. 2g; black arrow in colon), suggesting that mucin including NGAL discharged into intestinal lumen.
Toll-like receptor signal-induced NGAL expression in crypt cells. We further isolated intestinal crypt cells containing secretory granules, such as Paneth cells specifically stained using amido black, with a range of 70–80% (Fig. 3a). To examine whether each agonist induced the expression of TLR 2, 4, and 9 mRNA in isolated crypt cells, the cells were cultured with each
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
151
K. Mori et al.
Lipocalin and gut immune function
TLR ligand for 3 h to analyze the expression of each TLR mRNA. While the TLR 2 mRNA level was not increased in ileum crypt cells stimulated with zymosan as a TLR2-ligand, stimulation with zymosan significantly up-regulated the TLR 2 mRNA level in isolated colon crypt cells (Fig. 3b). Regarding TLR 4-stimulation, LPS as a TLR4-ligand induced a marked expression of TLR 4 mRNA in ileum crypt cells, but suppressed TLR 4 mRNA expression in colon crypt cells (Fig. 3c). As shown in Figure 3d, TLR 9 mRNA mediated by CpG-DNA was not up-regulated significantly in the ileum crypt cells, whereas a significant elevation of TLR 9 mRNA expression was found in the colon crypt cells. None of the TLR ligands evoked a significant increase in alphadefensin expression in either the ileum or the colon (Fig. 3e–g). To further examine whether stimulation with each TLR ligand evoked the up-regulation of NGAL mRNA in crypt cells, the NGAL expression induced by each TLR ligand was assessed on isolated crypt cells. Upon TLR 2-stimulation, NGAL mRNA was significantly expressed in colon crypt cells, but not in the ileum (Fig. 3h). The TLR4-ligand induced a significant up-regulation of NGAL mRNA, the level of which was greater than that induced by the other TLR ligands, in both ileum and colon crypt cells (Fig. 3i). For TLR 9, a significant increase in NGAL expression as a result of stimulation with CpG-DNA was found only in colon crypt cells (Fig. 3j). NGAL prevents overgrowth of E. coli but does not improve mortality in a peritonitis model. Figure 4a shows the number of enteric E. coli cultured with different doses of NGAL at 24 h, indicating that NGAL treatment in an in vitro study apparently prevented the growth of enteric E. coli in a dose-dependent manner. Using qRT-PCR analyses of E. coli, LPS injection was shown to evoke a significant overgrowth of E. coli in both the cecum and the colon contents (Fig. 4c,d). Of note, the overgrowth of E. coli was markedly depressed by the persistent administration of recombinant NGAL protein, compared with the administration of normal saline (Fig. 4c,d). We used the CLP model with the continuous administration of recombinant NGAL protein into the intestine to perform a survival rate analysis. The survival times of the saline-treated and the NGAL-treated mice were not significantly different (P = 0.77) (Fig. 5).
Discussion In the present study, we demonstrated that both LPS-induced endotoxemia and CLP-induced peritonitis are likely to evoke the up-regulation of the NGAL protein at both intestinal epithelial cells and crypt cells in a manner that is proportional to the degree of the insult. Besides, we found that pattern recognition receptors such as TLR 2, TLR 4, and TLR 9 were some of the main receptors for the up-regulation of NGAL in crypt cells. Our findings suggest that TLR ligands are a potent stimulator of NGAL secretion in intestinal crypt cells. Furthermore, we demonstrated that the persistent administration of recombinant NGAL protein into the intestinal lumen prevented the overgrowth of E. coli during LPS-induced changes in the microflora. Taken together, these findings indicate that intestinal crypt cells preserved NGAL protein under homeostatic conditions, and once inflammatory invasion
152
occurred, NGAL was up-regulated in secretory cells of the gut, such as epithelial and crypt cells, resulting in its accumulation in crypt lumens and its discharge into the intestinal lumen. To our knowledge, this is the first study to show an increase in the secretion of NGAL into the intestinal lumen according to the degree of disease severity and the role of NGAL as a potent antibacterial glycoprotein in an animal model of endotoxemia. In critically ill, alterations in microflora can easily occur through gut barrier dysfunction and/or environments caused by the discharge of inflammatory mediators, or antibiotic therapy.14–16 These alterations can subsequently induce the overgrowth of pathogenic bacteria, resulting in bacterial translocation and further development of organ injury,17,18 despite the presence of complex defense systems in the gut. Among these defense systems, Paneth cells, recognized by the presence of secretory granules in intestinal crypt cells, discharge such antimicrobial proteins and/or peptides as alpha-defensin, lysozyme, and secretory phospholipase A2 into the intestinal mucus layer, thereby preventing the colonization of pathogenic bacteria and preserving microflora homeostasis in the gut lumen.19,20 Our study demonstrated that insults caused by either LPS or peritonitis evoked a marked up-regulation of the NGAL mRNA and protein levels in gut secretory cells. In addition, histological analyses showed that NGAL protein accumulated in the crypt lumen and streamed into the mucosal layer under endotoxemic conditions, as previously described.5 Furthermore, the persistent administration of NGAL into the intestinal lumen markedly prevented the overgrowth of enteric E. coli elicited by LPS (Fig. 4), indicating that direct NGAL infusion exerted a bacteriostatic effect and improved aberrant alterations under critically ill conditions. A recent study demonstrated that NGAL was also detected in the feces of mice with colitis, accompanied by the distribution of the bacterial community in the gut.21,22 Collectively, intestinal NGAL secreted from crypt cells might exert anti-microbial effects to preserve microflora homeostasis, possibly depending on the level of NGAL. We also found that stimulation of the TLRs pathway evoked the up-regulation of NGAL mRNA as well as TLRs mRNA in an ex vivo study. Our study showed that TLR ligands induced the release of NGAL mRNA from crypt cells, especially in the colon, followed by a slight increase in alpha-defensin, an abundant antimicrobial peptide in intestinal crypt cells, expression 19 (Fig. 3). The TLRs pathway has been shown to play a pivotal role not only in the induction of the inflammatory status but also in the regulation and maintenance of the homeostatic condition in the gut under inflammatory conditions. Regarding the regulation of NGAL expression, peptidoglycans, such as the TLR2-ligand, strongly modulated the up-regulation of NGAL mRNA in an in vitro study using cultured distal convoluted tubular cells.23 Furthermore, TLR4 / mice did not express NGAL mRNA in either blood cells or peritoneal cells, even in the presence of endotoxemia.4 On the other hand, previous study indicated that inflammatory cytokines evoked by microbial infection augmented NGAL expression via the activation of the NF-κB pathway.24,25 Crypt cells involving TLRs have been reported to increase the discharge of anti-microbial peptides via the stimulation of the TLR pathway, thereby contributing to mucosal defense systems against enteric bacteria.13,26,27 Collectively, the overgrowth of gram-negative bacteria induced by LPS or peritonitis is likely to augment the expression of NGAL in the maintenance of microflora homeostasis
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
K. Mori et al.
through TLR pathways in intestinal crypt cells, contributing to the negative feedback that suppresses the overgrowth of pathogenic bacteria. Thus, the TRLs pathway might play an important role in regulating the homeostasis of intestinal microflora through NGAL secretion. There are several limitations to interpret the present data herein. First, although we demonstrated that the up-regulation of NGAL expression was induced in proportion to the degree of disease severity, it remains to be clarified if this proportional increase in NGAL expression is merely a marker of disease severity and if this increase in NGAL secretion contributes to the regulation of pathogenic bacteria in the intestinal lumen under disease conditions. However, our LPS model showed that NGAL is involved in the depression of E. coli overgrowth, suggesting that the increased secretion of NGAL may contribute to the maintenance of the intestinal environment. Second, although we did not measure the crypt length and density of mucosal area, these anatomical structures could be altered in disease conditions.28 In addition, NGAL involved N-glycosylation site and simultaneously possessed to form heterodimer with matrix metalloproteinase-9,29,30 both of which might be associated with the alterations of NGAL properties via such post-translational modifications. Further study was warranted to evaluate the association between function and structure in crypt cells, as well as post-translational modifications of NGAL. Third, because this study focused on the expression of NGAL induced by gram-negative bacteremia and the role of NGAL on the growth of E. coli, we were unable to indicate on how this up-regulation of NGAL function may affect other pathogenic bacteria or what the role of this glycoprotein with a biostatic capacity might be under other disease conditions. Finally, several issues such as clinical implications of bacterial overgrowth in the gut of rodent model and different responses between ileum and colon found in the present study remain to be fully determined. In conclusion, the secretion of NGAL was up-regulated in the intestinal lumen in an endotoxemia or peritonitis model in proportion to the severity of the insults, possibly through the TRLs pathway. NGAL infusion in an LPS-induced endotoxemia model suppressed the growth of E. coli without any effects on the survival rate. The present study sheds light on a new aspect of the importance of NGAL secreted into the intestine and the growth of pathogenic bacteria in response to inflammatory insults. Further study should be warranted to examine the exact role of NGAL in greater detail and in more clinically relevant situations.
Acknowledgement The authors would like to acknowledge the funding support from Japan Society of the Promotion of Science.
References 1 Cassat JE, Skaar EP. Iron in infection and immunity. Cell Host Microbe 2013; 13: 509–19. 2 Chakaborty S, Kaur S, Guha S, Batra SK. The multifaceted role of neutrophil gelatinase associated lipocalin (NGAL) in inflammation and cancer. Biochim. Biophys. Acta 2012; 1826: 129–69. 3 Borregaard N, Cowland JB. Neutrophil gelatinase-associated lipocalin, a siderophore-binding eukaryotic protein. Biometals 2006; 19: 211–5.
Lipocalin and gut immune function
4 Flo TH, Smith KD, Sato S et al. Lipocalin 2 mediates an innate immune response to bacterial infection by sequestrating iron. Nature 2004; 432: 917–21. 5 Chan YR, Liu JS, Pociask DA et al. Lipocalin 2 is required for pulmonary host defense against Klebsiella infection. J. Immunol. 2009; 182: 4947–56. 6 Guglani L, Gopal R, Rangel-Moreno J et al. Lipocalin 2 regulates inflammation during pulmonary mycobacteria infection. PLoS One 2012; 7: e50052. 7 Remick DG, Newcomb DE, Bolgos GL, Call DR. Comparison of the mortality and inflammatory response of two models of sepsis: lipopolysaccharide vs. cecal ligation and puncture. Shock 2000; 13: 110–6. 8 Macpherson AJ, Harris NL. Interactions between commensal intestinal bacteria and the immune system. Nat. Rev. Immunol. 2004; 4: 478–85. 9 Muniz LR, Knosp C, Yeretssian G. Intestinal antimicrobial peptides during homeostasis, infection, and disease. Front Immunol 2012; 3: 310. 10 Deitch EA, Xu D, Franko L, Ayala A, Chaudry IH. Evidence favoring the role of the gut as a cytokine-generating organ in rats subjected to hemorrhagic shock. Shock 1994; 1: 141–5. 11 Shimizu K, Ogura H, Hamasaki T et al. Altered gut flora are associated with septic complications and death in critically ill patients with systemic inflammatory response syndrome. Dig. Dis. Sci. 2011; 56: 1171–7. 12 Rittirsch D, Huber-Lang MS, Flierl MA, Ward PA. Immunodesign of experimental sepsis by cecal ligation and puncture. Nat. Protoc. 2009; 4: 31–6. 13 Ayabe T., Satchell DP, Wilson CL, Parks WC, Selsted ME, Ouellette AJ. Secretion of microbicidal alpha-defensins by intestinal Paneth cells in response to bacteria. Nat. Immunol. 2000; 1: 113–8. 14 Li Q, Zhang Q, Wang C, Liu X, Li N, Li J. Disruption of tight junctions during polymicrobial sepsis in vivo. J. Pathol. 2009; 218: 210–21. 15 Dantzker DR. The gastrointestinal tract. The canary of the body? JAMA 1993; 270: 1247–8. 16 Papoff P, Ceccarelli G, d’Ettorre G et al. Gut microbial translocation in critically ill children and effects of supplementation with pre- and pro biotics. Int J Microbiol. 2012; 2012: 151393. 17 Lousi K, Netea MG, Carrer DP et al. Bacterial translocation in experimental model of multiple organ dysfunctions. J. Surg. Res. 2013; 183: 686–94. 18 Shimizu K, Ogura H, Goto M et al. Altered gut flora and environment in patients with severe SIRS. J. Trauma 2006; 60: 126–33. 19 Ouellette AJ. Paneth cells and innate immunity in the crypt microenvironment. Gastroenterology 1997; 113: 1779–84. 20 Bevins CL, Salzman NH. Paneth cells, antimicrobial peptides and maintenance of intestinal homeostasis. Nat. Rev. Microbiol. 2011; 9: 356–68. 21 Winter SE, Lopez CA, Bäumler AJ. The dynamics of gut-associated microbial communities during inflammation. EMBO Rep. 2013; 14: 319–27. 22 Chassaing B, Srinivasan G, Delgado MA, Young AN, Gewirtz AT, Vijay-Kumar M. Fecal lipocalin 2, a sensitive and broadly dynamic non-invasive biomarker for intestinal inflammation. PLoS One 2012; 7: e44328. 23 Eller K, Schroll A, Banas M et al. Lipocalin-2 expressed in innate immune cells is an endogenous inhibitor of inflammation in murine nephrotoxic serum nephritis. PLoS One 2013; 8: e67693. 24 Cowland JB, Muta T, Borregaard N. IL-1beta-specific up-regulation of neutrophil gelatinase-associated lipocalin is controlled by IkappaBzeta. J. Immunol. 2006; 176: 5559–66. 25 Sommer G, Weise S, Kralisch S et al. Lipocalin-2 is induced by interleukin-1beta in murine adipocytes in vitro. J. Cell. Biochem. 2009; 106: 103–8. 26 Clevers HC, Bevins CL. Paneth cells: maestros of the small intestinal crypts. Annu. Rev. Physiol. 2013; 75: 289–311.
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
153
K. Mori et al.
Lipocalin and gut immune function
27 Rumio C, Besusso D, Palazzo M et al. Expression of toll-like receptor3 is enhanced in active inflammatory bowel disease and mediates the excessive release of lipocalin 2. Clin. Exp. Immunol. 2013; 173: 502–11. 28 DeRoche TC, Xiao SY, Liu X. Histological evaluation in ulcerative colitis. Gastroenterol Rep 2014; 2: 178–92.
154
29 Rudd PM, Mattu TS, Masure S et al. Glycosylation of natural human neutrophil gelatinase B and neutrophil gelatinase B-associated lipocalin. Biochemistry 1999; 38: 13937–50. 30 Opdenakker G1, Van den Steen PE, Van Da006Dme J. Gelatinase B. A tuner and amplifier of immune functions. Trends Immunol. 2001, 22: 571–9.
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
doi:10.1111/jgh.13042
GASTROENTEROLOGY
Neutrophil gelatinase-associated lipocalin regulates gut microbiota of mice Katsuya Mori,*,† Takeshi Suzuki,* Shizuka Minamishima,* Toru Igarashi,* Kei Inoue,* Daisuke Nishimura,* Hiroyuki Seki,* Takashige Yamada,* Shizuko Kosugi,* Nobuyuki Katori,* Saori Hashiguchi* and Hiroshi Morisaki* *Department of Anesthesiology, Keio University School of Medicine and † Japan Society of the Promotion of Science, Tokyo, Japan
Key words bacteria, gut, lipocalin-2, microbiota, toll-like receptor. Accepted for publication 25 June 2015. Correspondence Dr H. Morisaki, Department of Anesthesiology, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan. Email [email protected] Conflict of interest All authors have nothing to declare any financial or personal conflict of interest that might influence the results or interpretation of our manuscript.
Abstract Background and Aim: Because neutrophil gelatinase-associated lipocalin (NGAL) is known to provide significant bacteriostatic effects during infectious conditions, we tested the hypothesis that this protein is up-regulated and secreted into the intraluminal cavity of the gut under critically ill conditions and is thus responsible for the regulation of bacterial overgrowth. Methods: With our institutional approval, male C57BL/6J mouse (6–7 weeks) were enrolled and applied for lipopolysaccharide or peritonitis model compared with naïve control. We assessed NGAL protein concentrations in intestinal lumen and up-regulation of NGAL expression in intestinal tissues in in vivo as well as ex vivo settings. Simultaneously, we examined the effects of NGAL protein administration on the growth of Escherichia coli (E. coli) in in vivo and in vitro experimental settings. The localization of NGAL in intestinal tissues and lumen was also assessed by immunohistological approach using NGAL antibody. Results: Both lipopolysaccharide and peritonitis insults evoked the marked up-regulation of NGAL mRNA and protein levels in gut tissues such as crypt cells. In addition, the administration of NGAL protein significantly inhibited the outgrowth of enteric E. coli under both in vitro and in vivo conditions, accompanied by histological evidence. Conclusion: Neutrophil gelatinase-associated lipocalin protein accompanied by apparent bacteriostatic action accumulated in the intestinal wall and streamed into the mucosal layer during critically ill state, thereby possibly shaping microbiota homeostasis in the gut.
Introduction 3+
Bacterial species need Fe for their growth and thereby possess an independent iron-uptake system and means of combating host iron-withholding defenses in gut to overcome the host’s ironwithholding potential.1 For example, siderophores produced by bacteria are a family of low-molecular weight iron chelators that play a pivotal role in the iron acquisition system of bacteria. Neutrophil gelatinase-associated lipocalin (NGAL) is up-regulated in several disease conditions including inflammation, ischemia, and cancer.2 Because the intensity of NGAL expression depends on the severity of disease, its detection in clinical specimens has been considered to be a diagnostic biomarker of several illnesses.2 On the other hand, NGAL exerts a bacteriostatic action by preventing bacterial growth from undergoing iron-siderophore uptake.3,4 A previous study showed that several gram-negative bacteria,5,6 induce the up-regulation of NGAL in the lung, liver, and systemic circulation, resulting in the strict regulation of bacterial overgrowth. Thus, NGAL is regarded to be essential in innate immune functions in hosts. Approximately 107 commensal bacteria, consisting of over 1000 species in the gut8 and called microbiota, are also essential for the
development of intestinal immune systems. Their growth is strictly regulated by immune function through antimicrobial peptides such as defensins and cathelicidins.9 Although the disruption of microbiota, such as the overgrowth of pathogenic bacteria, can be prevented by these immune functions, the processes could conversely augment inflammation by enabling passage through the intestinal epithelial barrier, thereby stimulating immune responses under critically ill conditions.10 Taken together, these findings suggest that the imbalance of microbiota could be associated with organ dysfunction and a poor outcome in critically ill patients because of gut-barrier dysfunction.11 In the present study, we examined whether NGAL contributes to the preservation of microbiota homeostasis and the development of gut immune function in critically ill state.
Methods This study protocol was approved by the Animal Care and Use Committee of Keio University School of Medicine in accordance with the National Institute of Health guidelines. The details of Methods section were described in Supplemental Materials.
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
145
K. Mori et al.
Lipocalin and gut immune function
Animal models and analyses of NGAL protein concentrations in intestinal lumen. Male mice (aged 6–9 weeks) were used as endotoxemia and peritonitis models to evoke a systemic inflammatory response. For the former model, lipopolysaccharide (LPS) was administered, and for the latter model, we applied cecal ligation and perforation (CLP) to evoke peritonitis.7,12 The contents of the ileum and colon were collected and homogenized, and the supernatant was used to measure the NGAL protein concentration, which was detected by ELISA-based assay. Histological analyses of NGAL and mucin in ileum and colon tissues. To identify NGAL-secreting cells in the ileum and colon, 5-μm-thick paraffin sections of each tissue were examined using immunohistochemistry. To identify the mucinsecreting cells in the gut, the remaining sections were stained with Alcian blue and hematoxylin staining, and the results were observed under a light microscope by an independent examiner. Isolation of gut crypt cells. To isolate gut crypt cells including Paneth cells, ileum and colon crypt cells were isolated using modified procedures, as described previously.13 A portion of the pellets were re-suspended in Hank’s balanced salt solution (HBSS) buffer and were stained with 0.25% amido black to label the secretory granules of Paneth cells in the crypts; the results were then observed using light microscopy. Stimulation of crypt cells ex vivo. To analyze the NGAL expression in isolated Paneth cells, crypt cells obtained from the ileum and colon were incubated with toll-like receptor (TLR4)-, TLR2-, and TLR9-ligands. After exposure to each TLR ligand, the stimulated crypt cells were deposited by centrifugation at 700 xg, and the total RNA was eluted. Analysis of Escherichia coli growth in vitro. To analyze the impact of NGAL on the growth of E. coli, feces were collected and cultured with 100 μL of phosphate-buffered saline (PBS) containing 125, 250, 500, or 1000 ng of recombinant NGAL/lipocalin 2. The number of bacteria colonies was then counted. Persistent NGAL infusion model in LPS-treated and CLP-treated animals. After intraluminal placement of a catheter, the mice were divided into control, LPS NS group and LPS NGAL group. At 24-h study period, the contents of the cecum and colon were collected for bacterial DNA analysis. As a separate series of experiments, the mice were divided into control, LPS NS group and LPS NGAL group after intra-abdominal placement of a catheter. To examine the survival time in a CLP model, normal saline or recombinant NGAL protein was administered via an inserted catheter. RNA/DNA preparation, reverse transcription, and real-time polymerase chain reaction. Total RNA from the ileum, colon, and crypt cells were extracted using a RiboPureTM kit. The quantity of eluted total RNA were measured using a NanoDrop 1000A spectrophotometer. For the removal of
146
contaminating DNA from the isolated RNA, deoxyribonuclease treatment was performed using a DNA-free kit. Complementary DNA synthesis from the total RNA was performed using a ReverTra Ace qPCR RT kit with a random primer. For the measurement of NGAL and TLR family expression, semi-quantitative real-time polymerase chain reaction (PCR) was performed using TaqMan-based Applied Biosystem gene expression assays. For the isolation of bacterial DNA, the ileum and colon contents were washed with PBS, and then extracted using the QIAamp DNA Stool kit. E. coli was quantified using TaqMan-based real-time PCR with a Quantification of E. coli standard kit. Statistical analyses. All data are expressed as mean ± standard deviation unless otherwise specified. Analyses were performed using the SPSS/21.0J (SPSS Inc., Chicago, IL, USA). Expression of NGAL at different LPS concentrations and enteric E. coli DNA copy numbers were analyzed by one-way analysis of variance (ANOVA), followed by Tukey’s post-hoc test. Mann–Whitney test was applied to compare expression of TLR family, NGAL, and alpha-defensin mRNA. Kaplan–Meier analysis and log-rank test were used for survival data. Differences were considered statistically significant if P value was less than 0.05.
Results NGAL expression in intestinal tissue in critically ill models. Lipopolysaccharide administration markedly increased the expression of NGAL mRNA in both tissues in a time-dependent manner, compared with the levels in a treatment-naïve animal (Fig. 1a,b). In particular, NGAL expression in the ileum tissue was approximately 7000-fold higher after a 24-h study period. Neutrophil gelatinase-associated lipocalin mRNA was significantly up-regulated in both the ileum and colon, compared with a sham-treated animal, at 24 h after the CLP insult (Fig. 1c,d). Furthermore, we showed that LPS-induced insults significantly induced the expression of NGAL mRNA in a dose-dependent manner, with a peak typically observed at 24 h, in both the ileum and colon tissues (Fig. 1e,f). While the discharge of NGAL protein was not detected in the intraluminal contents of either the ileum or the colon in treatment-naïve mice at 24 h, LPS insults evoked the up-regulation of NGAL discharge in both tissues in a dose-dependent manner (Fig. 1g,h). Localization of NGAL secretory cells in gut. In the ileum mucosa of treatment-naïve mice, NGAL-positive cells were found on a few Paneth cells as secretory granules in crypts (Fig. 2a; black arrowhead), but were not found in epithelial cells of villi (Fig. 2a). In contrast, LPS injection strongly induced the upregulation of NGAL in epithelial cells (Fig. 2b; black arrow) and crypts, including Paneth cells of the ileum mucosa (Fig. 2b; black arrowhead). Of note, NGAL evoked by LPS accumulated in the luminal area of the crypt cells (Fig. 2b; white arrow), whereas no accumulation was found in treatment-naïve mice. In the colon, goblet cells in treatment-naïve crypt cells were positive for NGAL (Fig. 2c; white arrowhead), whereas NGAL expression was slightly detected in a few epithelial cells from treatmentnaïve colon tissue (Fig. 2c; black arrow). After an LPS insult, the epithelial cells of the colon mucosa were stained more clearly with
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
K. Mori et al.
Lipocalin and gut immune function
Figure 1 Different insults induced the up-regulation of NGAL in the ileum and colon tissue. The expression of NGAL mRNA up-regulated by LPS in the ileum (a) and colon (b) was analyzed using semi-quantitative real-time PCR at different time points. The values of treatment-naïve animals were set at 1.00, and the data show the fold-increase as the mean ± SD. The expression of NGAL mRNA induced by CLP in the ileum (c) and colon (d) was analyzed using semi-quantitative real-time PCR after a 24-h treatment period. The values of sham animals were set at 1.00, and the data show the fold-increase as the mean ± SD. The expressions of NGAL mRNA mediated by different doses of LPS in the ileum (e) and colon (f) were analyzed using semi-quantitative real-time PCR after a 24-h treatment period. The values of treatmentnaïve animals (n = 3) were set at 1.00, and the data show the fold-increase as the mean ± SD. Different doses of LPS-induced discharge of NGAL protein in the ileum lumen (g) and colon (h) were measured using an ELISA. The data were expressed as the mean ± SD. The sample size of all the groups was 3. Abbreviations: CLP, cecum ligation and perforation; LPS, lipopolysaccharides; NGAL, neutrophil gelatinase-associated lipocalin; PCR, polymerase chain reaction; SD, standard deviation.
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
147
Lipocalin and gut immune function
K. Mori et al.
Figure 2 Representative pictures of NGAL localization in gut mucosa with or without insults. NGAL-positive cells were detected in the intestinal mucosa using immunostaining. NGAL-positive cells were stained brown. All scale bars in panel (a to g) indicate 50 μm. Upper panel (a) shows a representative villus in the ileum of a treatment-naïve mouse, whereas lower panel (a) shows crypt cells in the ileum. Upper panel (b) shows a villus in the ileum of an endotoxemic mouse, whereas lower panel (b) shows crypt cells in the ileum. The up-regulation of NGAL (brown) induced by LPS was apparent in both the epithelial and crypt cells (upper and lower panel (b)), compared with that in a treatment-naïve mouse (upper and lower panel (a)). Of note, NGAL accumulation was apparent in the luminal area of the crypt cells (arrow, lower panel (b)), whereas no accumulation was found in a treatmentnaïve mouse lower panel (a). Upper panel (c) shows a representative villus in the colon of a treatment-naïve mouse, whereas lower panel (c) shows crypt cells in treatment-naïve colon. Upper panel (d) shows a villus in the colon of an endotoxemic mouse, whereas lower panel (b) shows crypt cells in the colon. Compared with the ileum, the up-regulation of NGAL (brown) was visible even in the treatment-naïve mouse. Additionally, LPS apparently evoked an up-regulation of NGAL in the colon similar to that occurring in the ileum. Panel (e) showed representative pictures of crypt cells in the ileum (left panel) and colon (right panel). In both sections, the crypt cells were opened at the apex and accumulated NGAL (brown) was discharged into the lumen of the intestine after the LPS insult. Panel (f) shows that mucus stained by Alcian blue (lower panel) completely matched the accumulation of NGAL in the crypt cells of the ileum in an LPS-treated mouse. Dotted lines in the upper and the lower panels indicate that location of NGAL (brown) corresponded to Alcian blue staining in crypt cells. Panel (g) shows that similar findings were found in the colon. Dotted lines in the upper and the lower panels indicate that location of NGAL (brown) corresponded to Alcian blue staining in crypt cells. Abbreviations: LPS, lipopolysaccharides; NGAL, neutrophil gelatinase-associated lipocalin.
148
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
K. Mori et al.
Lipocalin and gut immune function
Figure 3 Expression of TLR family, NGAL, and alpha-defensin mRNA in isolated crypt cells. Panel (a) shows that cells isolated with EDTA were crypt cells. Paneth cell in crypt cells were stained with 0.25% amido black (black arrow). Bar means 50 μm. Panels (b–d) show the expressions of TLR 2 (b), TLR 4 (c), and TLR 9 (d) mRNA in crypt cells stimulated by each TLR ligand. Ileum (left panel) and colon (right panel) crypt cells from the control group (n = 3) and the TLR groups (n = 3 per each TLR) were treated with normal saline and TLR ligand (TLR 2, zymosan; TLR 4, LPS; and TLR 9, CpG-DNA), respectively. The data show the fold increase as the mean ± SD. Panel (e–g) shows the expression of α-defensin mRNA in ileum (left panel) and colon (right panel) crypt cells, which were stimulated with the TLR 2 (e), TLR 4 (f), or TLR 9 (g) ligands. The sample size of each group was 3, and the data show the fold-increase as the mean ± SD. Panel (h–j) shows the expression of NGAL mRNA in ileum (left panel) and colon (right panel) crypt cells, which were stimulated with the TLR 2 (h), TLR 4 (i), and TLR 9 (j) ligands. The sample size of each group was 3, and the data show the fold-increase as the mean ± SD. Abbreviations: EDTA, ethylenediaminetetraacetic acid; LPS, lipopolysaccharides; NGAL, neutrophil gelatinase-associated lipocalin; SD, standard deviation; TLR, .
the NGAL antibody (Fig. 2d; black arrow). While the expression of NGAL was observed in the crypts of treatment-naïve colon mucosa, LPS stimulation led to the accumulation of NGAL in goblet cells and crypts in colon (Fig. 2d; white arrowhead and white arrow). As shown in Figure 2e, several LPS-stimulated crypt cells from both the ileum and the colon were opened at the apex of these
cells, and a thick, positive region of NGAL simultaneously streamed from the crypts into the intestinal lumen (Fig. 2e; white arrow). We also found that the contents of the ileum crypt lumens, which stained positive for NGAL antibody, were simultaneously stained with Alcian blue (Fig. 2f). A blue signal from Alcian blue was accumulated
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
149
Lipocalin and gut immune function
Figure 3
150
K. Mori et al.
(Continued)
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
K. Mori et al.
Lipocalin and gut immune function
Figure 4 Effects of recombinant NGAL on enteric Escherichia coli and survival rate of peritonitis model. Panel (a) shows that NGAL suppressed the growth of enteric cultured E. coli in a dose-dependent manner except for 500 ng in vitro. Panel (b) shows the intestinal fistula model; a catheter was inserted into the cecum, and recombinant NGAL protein was infused via a catheter for 24 h. Panels (c) and (d) show the copy number of E. coli in the cecum and colon contents. Data are shown for the control, LPS with normal saline infusion (LPS-NS), and LPS with 15 μg of recombinant NGAL infusion (LPS-NGAL) groups (n = 3, each group). The in vivo administration of recombinant NGAL protein significantly inhibited the overgrowth of E. coli evoked by LPS injection in both the cecum and colon. Data are shown as the mean ± SD. Panel (e) shows the Kaplan–Meier curve of the survival time in an animal model of CLP-induced peritonitis that was treated with recombinant NGAL (n = 10, each group). , Control; , NGAL. Treatment with NGAL in the peritonitis mouse model did not improve the survival rate. Abbreviations: CLP, cecum ligation and perforation; LPS, lipopolysaccharides; NGAL, neutrophil gelatinase-associated lipocalin; NS, normal saline; SD, standard deviation.
in crypt lumen of both ileum and colon (Fig. 2f,g; white arrow) and goblet cells of colon (Fig. 2g; white arrowhead) that exhibited positive signals for NGAL; simultaneously, a thick positive region streaming from the crypts into the intestinal lumen was also apparent (Fig. 2g; black arrow in colon), suggesting that mucin including NGAL discharged into intestinal lumen.
Toll-like receptor signal-induced NGAL expression in crypt cells. We further isolated intestinal crypt cells containing secretory granules, such as Paneth cells specifically stained using amido black, with a range of 70–80% (Fig. 3a). To examine whether each agonist induced the expression of TLR 2, 4, and 9 mRNA in isolated crypt cells, the cells were cultured with each
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
151
K. Mori et al.
Lipocalin and gut immune function
TLR ligand for 3 h to analyze the expression of each TLR mRNA. While the TLR 2 mRNA level was not increased in ileum crypt cells stimulated with zymosan as a TLR2-ligand, stimulation with zymosan significantly up-regulated the TLR 2 mRNA level in isolated colon crypt cells (Fig. 3b). Regarding TLR 4-stimulation, LPS as a TLR4-ligand induced a marked expression of TLR 4 mRNA in ileum crypt cells, but suppressed TLR 4 mRNA expression in colon crypt cells (Fig. 3c). As shown in Figure 3d, TLR 9 mRNA mediated by CpG-DNA was not up-regulated significantly in the ileum crypt cells, whereas a significant elevation of TLR 9 mRNA expression was found in the colon crypt cells. None of the TLR ligands evoked a significant increase in alphadefensin expression in either the ileum or the colon (Fig. 3e–g). To further examine whether stimulation with each TLR ligand evoked the up-regulation of NGAL mRNA in crypt cells, the NGAL expression induced by each TLR ligand was assessed on isolated crypt cells. Upon TLR 2-stimulation, NGAL mRNA was significantly expressed in colon crypt cells, but not in the ileum (Fig. 3h). The TLR4-ligand induced a significant up-regulation of NGAL mRNA, the level of which was greater than that induced by the other TLR ligands, in both ileum and colon crypt cells (Fig. 3i). For TLR 9, a significant increase in NGAL expression as a result of stimulation with CpG-DNA was found only in colon crypt cells (Fig. 3j). NGAL prevents overgrowth of E. coli but does not improve mortality in a peritonitis model. Figure 4a shows the number of enteric E. coli cultured with different doses of NGAL at 24 h, indicating that NGAL treatment in an in vitro study apparently prevented the growth of enteric E. coli in a dose-dependent manner. Using qRT-PCR analyses of E. coli, LPS injection was shown to evoke a significant overgrowth of E. coli in both the cecum and the colon contents (Fig. 4c,d). Of note, the overgrowth of E. coli was markedly depressed by the persistent administration of recombinant NGAL protein, compared with the administration of normal saline (Fig. 4c,d). We used the CLP model with the continuous administration of recombinant NGAL protein into the intestine to perform a survival rate analysis. The survival times of the saline-treated and the NGAL-treated mice were not significantly different (P = 0.77) (Fig. 5).
Discussion In the present study, we demonstrated that both LPS-induced endotoxemia and CLP-induced peritonitis are likely to evoke the up-regulation of the NGAL protein at both intestinal epithelial cells and crypt cells in a manner that is proportional to the degree of the insult. Besides, we found that pattern recognition receptors such as TLR 2, TLR 4, and TLR 9 were some of the main receptors for the up-regulation of NGAL in crypt cells. Our findings suggest that TLR ligands are a potent stimulator of NGAL secretion in intestinal crypt cells. Furthermore, we demonstrated that the persistent administration of recombinant NGAL protein into the intestinal lumen prevented the overgrowth of E. coli during LPS-induced changes in the microflora. Taken together, these findings indicate that intestinal crypt cells preserved NGAL protein under homeostatic conditions, and once inflammatory invasion
152
occurred, NGAL was up-regulated in secretory cells of the gut, such as epithelial and crypt cells, resulting in its accumulation in crypt lumens and its discharge into the intestinal lumen. To our knowledge, this is the first study to show an increase in the secretion of NGAL into the intestinal lumen according to the degree of disease severity and the role of NGAL as a potent antibacterial glycoprotein in an animal model of endotoxemia. In critically ill, alterations in microflora can easily occur through gut barrier dysfunction and/or environments caused by the discharge of inflammatory mediators, or antibiotic therapy.14–16 These alterations can subsequently induce the overgrowth of pathogenic bacteria, resulting in bacterial translocation and further development of organ injury,17,18 despite the presence of complex defense systems in the gut. Among these defense systems, Paneth cells, recognized by the presence of secretory granules in intestinal crypt cells, discharge such antimicrobial proteins and/or peptides as alpha-defensin, lysozyme, and secretory phospholipase A2 into the intestinal mucus layer, thereby preventing the colonization of pathogenic bacteria and preserving microflora homeostasis in the gut lumen.19,20 Our study demonstrated that insults caused by either LPS or peritonitis evoked a marked up-regulation of the NGAL mRNA and protein levels in gut secretory cells. In addition, histological analyses showed that NGAL protein accumulated in the crypt lumen and streamed into the mucosal layer under endotoxemic conditions, as previously described.5 Furthermore, the persistent administration of NGAL into the intestinal lumen markedly prevented the overgrowth of enteric E. coli elicited by LPS (Fig. 4), indicating that direct NGAL infusion exerted a bacteriostatic effect and improved aberrant alterations under critically ill conditions. A recent study demonstrated that NGAL was also detected in the feces of mice with colitis, accompanied by the distribution of the bacterial community in the gut.21,22 Collectively, intestinal NGAL secreted from crypt cells might exert anti-microbial effects to preserve microflora homeostasis, possibly depending on the level of NGAL. We also found that stimulation of the TLRs pathway evoked the up-regulation of NGAL mRNA as well as TLRs mRNA in an ex vivo study. Our study showed that TLR ligands induced the release of NGAL mRNA from crypt cells, especially in the colon, followed by a slight increase in alpha-defensin, an abundant antimicrobial peptide in intestinal crypt cells, expression 19 (Fig. 3). The TLRs pathway has been shown to play a pivotal role not only in the induction of the inflammatory status but also in the regulation and maintenance of the homeostatic condition in the gut under inflammatory conditions. Regarding the regulation of NGAL expression, peptidoglycans, such as the TLR2-ligand, strongly modulated the up-regulation of NGAL mRNA in an in vitro study using cultured distal convoluted tubular cells.23 Furthermore, TLR4 / mice did not express NGAL mRNA in either blood cells or peritoneal cells, even in the presence of endotoxemia.4 On the other hand, previous study indicated that inflammatory cytokines evoked by microbial infection augmented NGAL expression via the activation of the NF-κB pathway.24,25 Crypt cells involving TLRs have been reported to increase the discharge of anti-microbial peptides via the stimulation of the TLR pathway, thereby contributing to mucosal defense systems against enteric bacteria.13,26,27 Collectively, the overgrowth of gram-negative bacteria induced by LPS or peritonitis is likely to augment the expression of NGAL in the maintenance of microflora homeostasis
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
K. Mori et al.
through TLR pathways in intestinal crypt cells, contributing to the negative feedback that suppresses the overgrowth of pathogenic bacteria. Thus, the TRLs pathway might play an important role in regulating the homeostasis of intestinal microflora through NGAL secretion. There are several limitations to interpret the present data herein. First, although we demonstrated that the up-regulation of NGAL expression was induced in proportion to the degree of disease severity, it remains to be clarified if this proportional increase in NGAL expression is merely a marker of disease severity and if this increase in NGAL secretion contributes to the regulation of pathogenic bacteria in the intestinal lumen under disease conditions. However, our LPS model showed that NGAL is involved in the depression of E. coli overgrowth, suggesting that the increased secretion of NGAL may contribute to the maintenance of the intestinal environment. Second, although we did not measure the crypt length and density of mucosal area, these anatomical structures could be altered in disease conditions.28 In addition, NGAL involved N-glycosylation site and simultaneously possessed to form heterodimer with matrix metalloproteinase-9,29,30 both of which might be associated with the alterations of NGAL properties via such post-translational modifications. Further study was warranted to evaluate the association between function and structure in crypt cells, as well as post-translational modifications of NGAL. Third, because this study focused on the expression of NGAL induced by gram-negative bacteremia and the role of NGAL on the growth of E. coli, we were unable to indicate on how this up-regulation of NGAL function may affect other pathogenic bacteria or what the role of this glycoprotein with a biostatic capacity might be under other disease conditions. Finally, several issues such as clinical implications of bacterial overgrowth in the gut of rodent model and different responses between ileum and colon found in the present study remain to be fully determined. In conclusion, the secretion of NGAL was up-regulated in the intestinal lumen in an endotoxemia or peritonitis model in proportion to the severity of the insults, possibly through the TRLs pathway. NGAL infusion in an LPS-induced endotoxemia model suppressed the growth of E. coli without any effects on the survival rate. The present study sheds light on a new aspect of the importance of NGAL secreted into the intestine and the growth of pathogenic bacteria in response to inflammatory insults. Further study should be warranted to examine the exact role of NGAL in greater detail and in more clinically relevant situations.
Acknowledgement The authors would like to acknowledge the funding support from Japan Society of the Promotion of Science.
References 1 Cassat JE, Skaar EP. Iron in infection and immunity. Cell Host Microbe 2013; 13: 509–19. 2 Chakaborty S, Kaur S, Guha S, Batra SK. The multifaceted role of neutrophil gelatinase associated lipocalin (NGAL) in inflammation and cancer. Biochim. Biophys. Acta 2012; 1826: 129–69. 3 Borregaard N, Cowland JB. Neutrophil gelatinase-associated lipocalin, a siderophore-binding eukaryotic protein. Biometals 2006; 19: 211–5.
Lipocalin and gut immune function
4 Flo TH, Smith KD, Sato S et al. Lipocalin 2 mediates an innate immune response to bacterial infection by sequestrating iron. Nature 2004; 432: 917–21. 5 Chan YR, Liu JS, Pociask DA et al. Lipocalin 2 is required for pulmonary host defense against Klebsiella infection. J. Immunol. 2009; 182: 4947–56. 6 Guglani L, Gopal R, Rangel-Moreno J et al. Lipocalin 2 regulates inflammation during pulmonary mycobacteria infection. PLoS One 2012; 7: e50052. 7 Remick DG, Newcomb DE, Bolgos GL, Call DR. Comparison of the mortality and inflammatory response of two models of sepsis: lipopolysaccharide vs. cecal ligation and puncture. Shock 2000; 13: 110–6. 8 Macpherson AJ, Harris NL. Interactions between commensal intestinal bacteria and the immune system. Nat. Rev. Immunol. 2004; 4: 478–85. 9 Muniz LR, Knosp C, Yeretssian G. Intestinal antimicrobial peptides during homeostasis, infection, and disease. Front Immunol 2012; 3: 310. 10 Deitch EA, Xu D, Franko L, Ayala A, Chaudry IH. Evidence favoring the role of the gut as a cytokine-generating organ in rats subjected to hemorrhagic shock. Shock 1994; 1: 141–5. 11 Shimizu K, Ogura H, Hamasaki T et al. Altered gut flora are associated with septic complications and death in critically ill patients with systemic inflammatory response syndrome. Dig. Dis. Sci. 2011; 56: 1171–7. 12 Rittirsch D, Huber-Lang MS, Flierl MA, Ward PA. Immunodesign of experimental sepsis by cecal ligation and puncture. Nat. Protoc. 2009; 4: 31–6. 13 Ayabe T., Satchell DP, Wilson CL, Parks WC, Selsted ME, Ouellette AJ. Secretion of microbicidal alpha-defensins by intestinal Paneth cells in response to bacteria. Nat. Immunol. 2000; 1: 113–8. 14 Li Q, Zhang Q, Wang C, Liu X, Li N, Li J. Disruption of tight junctions during polymicrobial sepsis in vivo. J. Pathol. 2009; 218: 210–21. 15 Dantzker DR. The gastrointestinal tract. The canary of the body? JAMA 1993; 270: 1247–8. 16 Papoff P, Ceccarelli G, d’Ettorre G et al. Gut microbial translocation in critically ill children and effects of supplementation with pre- and pro biotics. Int J Microbiol. 2012; 2012: 151393. 17 Lousi K, Netea MG, Carrer DP et al. Bacterial translocation in experimental model of multiple organ dysfunctions. J. Surg. Res. 2013; 183: 686–94. 18 Shimizu K, Ogura H, Goto M et al. Altered gut flora and environment in patients with severe SIRS. J. Trauma 2006; 60: 126–33. 19 Ouellette AJ. Paneth cells and innate immunity in the crypt microenvironment. Gastroenterology 1997; 113: 1779–84. 20 Bevins CL, Salzman NH. Paneth cells, antimicrobial peptides and maintenance of intestinal homeostasis. Nat. Rev. Microbiol. 2011; 9: 356–68. 21 Winter SE, Lopez CA, Bäumler AJ. The dynamics of gut-associated microbial communities during inflammation. EMBO Rep. 2013; 14: 319–27. 22 Chassaing B, Srinivasan G, Delgado MA, Young AN, Gewirtz AT, Vijay-Kumar M. Fecal lipocalin 2, a sensitive and broadly dynamic non-invasive biomarker for intestinal inflammation. PLoS One 2012; 7: e44328. 23 Eller K, Schroll A, Banas M et al. Lipocalin-2 expressed in innate immune cells is an endogenous inhibitor of inflammation in murine nephrotoxic serum nephritis. PLoS One 2013; 8: e67693. 24 Cowland JB, Muta T, Borregaard N. IL-1beta-specific up-regulation of neutrophil gelatinase-associated lipocalin is controlled by IkappaBzeta. J. Immunol. 2006; 176: 5559–66. 25 Sommer G, Weise S, Kralisch S et al. Lipocalin-2 is induced by interleukin-1beta in murine adipocytes in vitro. J. Cell. Biochem. 2009; 106: 103–8. 26 Clevers HC, Bevins CL. Paneth cells: maestros of the small intestinal crypts. Annu. Rev. Physiol. 2013; 75: 289–311.
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
153
K. Mori et al.
Lipocalin and gut immune function
27 Rumio C, Besusso D, Palazzo M et al. Expression of toll-like receptor3 is enhanced in active inflammatory bowel disease and mediates the excessive release of lipocalin 2. Clin. Exp. Immunol. 2013; 173: 502–11. 28 DeRoche TC, Xiao SY, Liu X. Histological evaluation in ulcerative colitis. Gastroenterol Rep 2014; 2: 178–92.
154
29 Rudd PM, Mattu TS, Masure S et al. Glycosylation of natural human neutrophil gelatinase B and neutrophil gelatinase B-associated lipocalin. Biochemistry 1999; 38: 13937–50. 30 Opdenakker G1, Van den Steen PE, Van Da006Dme J. Gelatinase B. A tuner and amplifier of immune functions. Trends Immunol. 2001, 22: 571–9.
Journal of Gastroenterology and Hepatology 31 (2016) 145–154 © 2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd
