Reduction of the amount of intestinal secretory IgA in fulminant hepatic failure

Intestinal barrier dysfunction plays an important role in spontaneous bacterial peritonitis. In the present study, changes in the intestinal barrier with regard to levels of secretory immunoglobulin A (SIgA) and its components were studied in fulminant hepatic failure (FHF). Immunohistochemistry and double immunofluorescent staining were used to detect intestinal IgA, the secretory component (SC) and SIgA in patients with FHF (20 patients) and in an animal model with FHF (120 mice). Real-time PCR was used to detect intestinal SC mRNA in the animal model with FHF. Intestinal SIgA, IgA, and SC staining in patients with FHF was significantly weaker than in the normal control group (30 patients). Intestinal IgA and SC staining was significantly weaker in the animal model with FHF than in the control groups (normal saline: 30 mice; lipopolysaccharide: 50 mice; D-galactosamine: 50 mice; FHF: 120 mice). SC mRNA of the animal model with FHF at 2, 6, and 9 h after injection was 0.4 ± 0.02, 0.3 ± 0.01, 0.09 ± 0.01, respectively. SC mRNA of the animal model with FHF was significantly decreased compared to the normal saline group (1.0 ± 0.02) and lipopolysaccharide group (0.89 ± 0.01). The decrease in intestinal SIgA and SC induced failure of the intestinal immunologic barrier and the attenuation of gut immunity in the presence of FHF.

Secretory IgA (SIgA), which is composed of IgA, secretory component (SC) and the J chain, is an important component of the intestinal immunologic barrier.IgA is a predominant Ig in mucosal secretions and serves as the first line of humoral defense on intestinal mucosal surfaces.The majority of these IgA proteins are monomers, while the rest are predominantly polymeric to facilitate efficient transport to the mucosal secretions mediated via the polymeric Ig receptor (PIgR).SC is the extracellular component of the PIgR responsible for the transcytosis of newly synthesized IgA (polymeric IgA).SIgA and its components, SC and IgA, are important components of the intestinal immunologic barrier.Up to now, investigations of the intestinal barrier in patients with FHF have focused mainly on the mechanical barrier and barrier to the microorganism.In the present study, SIgA, SC and IgA were measured in patients with FHF and in an animal model of FHF in order to elucidate the state of the intestinal immunologic barrier.

Tissue samples
Paraffin-embedded sections of intestinal tissue from 20 patients with FHF who had died were obtained from Youan Hospital in Beijing.For controls, paraffin-embedded sections of normal intestine tissue from 30 patients who had died in a traffic accident were obtained from China Medical University, affiliated with Shengjing Hospital.All experimental procedures were approved by the Ethics Committee of China Medical University before the beginning of the study.

Immunohistochemistry of tissue samples
Intestinal IgA and SC of patients and mice with FHF were determined by immunostaining.Paraffin-embedded intestinal sections of patients and mice with FHF were deparaffinized, rehydrated, and stained with goat anti-human IgA or anti-mouse IgA (Sigma) and mouse anti-human SC (SigA; Sigma) or anti-mouse SC mAb (Bethyl, USA) by sequential incubation.The slides were rinsed three times with 0.02 M PBS, pH 7.4, after each incubation, and sections were counterstained with hematoxylin.The sections from the same patient or the mouse were processed without the primary antibody and then examined by the procedure detailed above as a control for nonspecific binding of the secondary antibody.The absorbance values of IgA and SC were determined with the Image analysis software (China) after scanning.

Double immunofluorescent staining of intestinal SIgA
Paraffin-embedded intestinal sections of patients with FHF were deparaffinized, rehydrated, and double-stained by immunofluorescence with rhodamine-conjugated goat anti-human IgA (Southern Biotech, USA) and fluoresceinconjugated mouse anti-human SC (SigA; ICN, USA) by sequential incubation.The slides were rinsed three times with 0.02 M PBS, pH 7.4, after each incubation, and sections were counterstained with hematoxylin.Sections from the same patient or the mouse were processed without the primary antibody and then examined using the procedure detailed above as a control for nonspecific binding of the secondary antibody.

Quantitative real-time polymerase chain reaction of intestinal SC mRNA of mice with FHF
Total RNA was measured as described earlier in Refs.10,11.Briefly, total RNA was extracted from mouse gut tissue with FHF using an RNA Mini Kit from Takara (Takara Biotechnology Co., Ltd., China).The quality of extracted RNA was determined by the number and size of the bands obtained with agarose gel electrophoresis.cDNA was synthesized using 100 ng RNA.The levels of individual RNA transcripts were quantified by the quantitative real-time polymerase chain reaction (PCR).The primers of SC were as follows: PIgR-F: 5'-CAGACATTAGCATGGCAGACTT CAA-3'; PIgR-R: 5'-TGCCGAGTAGGCCATGTCAG-3'; GAPDH-F: 5'-AATGGTGAAGGTCGGTGTG-3'; GAPDH-R: 5'-TGAAGGGGTCGTTGATGG-3'.
The primers and fluorescent probes for SC and standard were purchased from Takara.The PCR conditions were a preliminary cycle at 95°C for 10 s, followed by 45 cycles at 95°C for 5 s and at 60°C for 20 s, followed by 1 min at 60°C and 5 s at 95°C.We also confirmed that the efficiency of amplification for each target gene (GAPDH) was 100% in the exponential phase of PCR.The levels of SC mRNA and GAPDH mRNA were determined according to the standard (Takara).The mRNA levels were normalized to GAPDH mRNA by dividing SC gene copies of the samples by the SC gene copies of GAPDH.The intestinal RNA level of the NS group was assumed to be 1, with which the intestinal RNA levels of other groups were compared.

Statistical analysis
The statistical differences between treatment groups were determined by the t-test using the SPSS 10.0 software.

Intestinal IgA and SC staining in patients
In the normal control group, SC staining of the cytoplasm and membranes of intestinal epithelial cells was brown (Figure 1A).SC staining of the cytoplasm and of the membranes of intestinal epithelial cells in patients with FHF was pale and considerably weaker than in the normal control group (Figure 1B).IgA staining of the cytoplasm of plasma cells within the lamina propria and of intestinal epithelial cells was brown in the normal control group (Figure 1C).However, in patients with FHF, IgA staining of the cytoplasm of plasma cells within the lamina propria was pale brown and IgA staining of intestinal epithelial cells was pale.IgA staining in patient samples was notably weaker than that of the normal control group (Figure 1D).The mean absorbance values of SC and IgA in patients with FHF were significantly lower than those of the normal control group (P < 0.01; Figure 1E).

Double immunofluorescence staining
The results of intestinal double immunofluorescence showed that the cytoplasm of normal intestinal epithelial cells stained orange (SC, SIgA) and the membranes stained yellow (SC, IgA, SIgA), and that the plasma cells within the lamina propria stained orange (SIgA) and yellow (IgA, SIgA; Figure 2A).However, in patients with FHF, the staining of the intestinal epithelial cell cytoplasm and membranes as well as plasma cells within the lamina propria was green (Figure 2B).The staining results proved that intestinal SC, IgA and SIgA were decreased compared to the normal control group.

Intestinal IgA staining in the mouse model of FHF
IgA staining of the cytoplasm and/or membranes of intestinal epithelial cells, as well as of plasma cells within the lamina propria, was weaker in the mouse FHF group (Figure 3D) compared to the NS (Figure 3A), LPS (Figure 3B) and GalN groups (Figure 3C).The mean absorbance values of IgA in the mouse FHF group were significantly decreased from 2 to 12 h compared to the NS control group (P < 0. 01; Figure 3E).

Intestinal SC staining in mice with FHF
SC staining of the cytoplasm and/or the membranes of intestinal epithelial cells was significantly weaker in the mouse FHF group (Figure 4D) compared to the NS group (Figure 4A), LPS group (Figure 4B), and GalN group (Figure 4C).The mean absorbance values of SC in the mouse FHF group were significantly decreased from 2 to 12 h compared to the NS group (P < 0.01; Figure 4E).

SC mRNA of intestinal tissues in the animal model of FHF
The relative content of intestinal SC mRNA in mice with FHF at 2, 6, and 9 h after injection was 0.4 ± 0.02, 0.3 ± 0.01, 0.09 ± 0.01, respectively.In the mice with FHF, intestinal SC mRNA was significantly lower at 2 to 9 h compared to the NS group (1.0 ± 0.02) and LPS group (0.89 ± 0.01; P < 0.01).In the GalN group, intestinal SC mRNA (0.2 ± 0.00) was also lower compared to the NS group and LPS group (P < 0.01; Figure 5).

Discussion
SBP is a common illness in patients with cirrhosis and ascites, which occurs without any apparent focus of infection.Bacterial translocation, gut motility alterations, and intestinal barrier dysfunction, along with bacterial overgrowth and changes in intestinal permeability play important roles in SBP, and it is evident from a number of studies that the gut is a major source of these bacteria (12).Intestinal SIgA levels are inversely correlated with bacterial overgrowth, bacterial translocation and changes in intestinal permeability (13).SIgA may weaken bacterial translation and restrict bacteria through intestinal epithelia (14).Secretory IgA is the first line of defense against microorganisms through immune exclusion that prevents interaction of neutralized antigens with the intestinal epithelium.SIgA plays an important role in the immune balance of the intestinal epithelial barrier (15)(16)(17).SC plays a protective role in preventing the proteolytic degradation of polymeric IgA, thus enhancing the mucosal immunity provided by IgA at these sites.SC is a key defense against bacteria (18-21) and  parasites (22) as well as playing a critical role in the immunologically mediated neutralization of cholera toxin (23).In vitro, free SC has been shown to fix the complement component C3b, suggesting a role for SC in enhancing local immune responses (24).
In liver diseases, notably alcoholic cirrhosis of the liver, intestinal SIgA is decreased (25).In the present study, intestinal IgA, SC and SIgA were markedly decreased in patients with FHF, this possibly being the reason why gut immunity is weakened in patients with FHF.Changes in intestinal SIgA, IgA and SC in the animal model of FHF were consistent with those of the patients with FHF.SC mRNA was simultaneously detected by real-time PCR.A decrease in intestinal SC mRNA was observed in the animal model with FHF.These results confirm that at the mRNA and protein levels, SC is decreased in the animal model of FHF.It has been reported that in mildly inflamed ulcerative colitis lesions, SC levels showed no change but tended to decrease with increasing degree of inflammation.Decreased SC expression thus seemed to be related to intensified inflammatory activity (26).Therefore, we think that in patients with FHF, damaged intestinal mucus results in lower levels of intestinal SC and SIgA.Malnourished animals exhibit low levels of hepatic and intestinal pIgR.The concentration of biliary SC was also low in malnourished animals (27).In patients with FHF, the disease is usually complicated by malnutrition.Therefore, malnutrition may cause low concentrations of intestinal SC in FHF.
In conclusion, intestinal IgA and SC (SIgA) were lower in patients with FHF, with consequent failure of the intestinal immunologic barrier and attenuation of gut immunity.

Figure 1 .
Figure 1.Secretory component (SC) and IgA staining of intestinal epithelial cells in patients with fulminant hepatic failure (FHF).A, SC staining of intestinal epithelial cells in the normal control group (30 patients) was brown (see arrow).B, SC staining of intestinal epithelial cells in patients with FHF (20 patients) was pale and remarkably weaker than in the normal control group.The staining results indicated that intestinal SC was significantly decreased in patients with FHF.C, IgA staining of plasma cells within the lamina propria and of intestinal epithelial cells in the normal control group was brown (see arrows).D, IgA staining of the cytoplasm of plasma cells within the lamina propria was pale brown and IgA staining of intestinal epithelial cells in patients with FHF was pale and remarkably weaker than in the normal control group.The staining results indicated that intestinal IgA in patients with FHF was significantly decreased.Magnification bars = 50 µm.E, The absorbance values of SC and IgA were notably less in patients with FHF compared to the normal control group.Data are reported as means ± SD. *P < 0.01 compared to the normal control group (t-test).

Figure 2 .
Figure 2. IgA and secretory IgA (SIgA) staining of the intestinal epithelium by double immunofluorescence.A, The secretory component (SC) and SIgA stained orange in the cytoplasm and SC, IgA and SIgA on the membranes stained yellow in the normal intestinal epithelial cells.SIgA of the plasma cells within the lamina propria stained orange.IgA and SIgA of the plasma cells within the lamina propria stained yellow (see blue arrows).B, In patients with fulminant hepatic failure (FHF), the color of the intestinal epithelial cell cytoplasm, membranes and plasma cells within the lamina propria was green.The staining results indicated that intestinal SC, IgA and SIgA were notably decreased compared to the normal control group.Magnification bars = 50 µm.

Figure 3 .
Figure 3. IgA staining of the intestinal epithelium in the animal model with fulminant hepatic failure (FHF).A, In the normal saline (NS) group (30 mice), IgA staining of intestinal epithelial cell cytoplasm, membranes and plasma cells within the lamina propria was deep brown (arrows).B, In the lipopolysaccharide (LPS) group (50 mice), IgA staining of intestinal epithelial cell cytoplasm, membranes and plasma cells within the lamina propria was brown.C, In the galactosamine (GalN) group (50 mice), IgA staining of the intestinal epithelial cell cytoplasm, membranes and plasma cells within the lamina propria was pale brown.D, In the FHF group (120 mice), IgA staining of some intestinal epithelial cell cytoplasm and membranes was pale.Magnification bars = 50 µm.E, The absorbance values of IgA in the animal model with FHF (2, 6, 9, and 12 h) were significantly decreased compared to the normal saline (NS), lipopolysaccharide (LPS) and galactosamine (GalN) groups.Data are reported as means ± SD. *P < 0.01 compared to NS 9 h, LPS 9 h, and GalN 9 h (t-test).

Figure 4 .
Figure 4.Staining of the secretory component (SC) of the intestinal epithelium in the mouse model of fulminant hepatic failure (FHF).A, In the normal saline (NS) group (30 mice), the SC staining of the intestinal epithelial cell cytoplasm and membranes was deep brown (see arrows).B, In the lipopolysaccharide (LPS) group (50 mice), SC staining of the intestinal epithelial cell cytoplasm and membranes was brown.C, In the galactosamine (GalN) group (50 mice), SC staining of the intestinal epithelial cell cytoplasm and membranes was pale brown.D, In the FHF group (120 mice), SC staining of intestinal epithelial cells was pale.E, The absorbance values of SC in the animal model with FHF (2, 6, 9, and 12 h) were significantly decreased compared to the normal saline (NS), lipopolysaccharide (LPS) and galactosamine (GalN) groups.Data are reported as means ± SD. *P < 0.01 compared to NS 9 h, LPS 9 h, and GalN 9 h (t-test).

Figure 5 .
Figure 5. Relative content of the intestinal secretory component (SC) mRNA.Relative content of intestinal SC mRNA in the animal model with fulminant hepatic failure (FHF) was significantly lower compared to the normal saline (NS) and lipopolysaccharide (LPS) groups.The relative content of intestinal SC mRNA in the galactosamine (GalN) group was also lower compared to the NS and LPS groups.Data are reported as means ± SD. *P < 0.01 compared to NS and LPS (t-test).