Intestinal alkaline phosphatase inhibits the translocation of bacteria of gut-origin in mice with peritonitis: mechanism of action

Abstract

Exogenous intestinal alkaline phosphatase (IAP), an enzyme produced endogenously at the brush edge of the intestinal mucosa, may mitigate the increase in aberrant intestinal permeability increased during sepsis. The aim of this study was to test the efficacy of the inhibitory effect of IAP on acute intestinal inflammation and to study the molecular mechanisms underlying IAP in ameliorating intestinal permeability. We used an in vivo imaging method to evaluate disease status and the curative effect of IAP. Two Escherichia coli (E.coli) B21 strains, carrying EGFP labeled enhanced green fluorescent protein (EGFP) and RFP labeled red fluorescent protein (RFP), were constructed as tracer bacteria and were administered orally to C57/B6N mice to generate an injection peritonitis (IP) model. The IP model was established by injecting inflammatory lavage fluid. C57/B6N mice bearing the tracer bacteria were subsequently treated with (IP+IAP group), or without IAP (IP group). IAP was administered to the mice via tail vein injections. The amount of tracer bacteria in the blood, liver, and lungs at 24 h post-injection was analyzed via flow cytometry (FCM), in vivo imaging, and Western blotting. Intestinal barrier function was measured using a flux assay with the macro-molecule fluorescein isothiocyanate dextran, molecular weight 40kD, (FD40). To elucidate the molecular mechanism underlying the effects of IAP, we examined the levels of ERK phosphorylation, and the expression levels of proteins in the ERK-SP1-VEGF and ERK-Cdx-2-Claudin-2 pathways. We observed that IAP inhibited the expression of Claudin-2, a type of cation channel-forming protein, and VEGF, a cytokine that may increase intestinal permeability by reducing the levels of dephosphorylated ERK. In conclusion, exogenous IAP shows a therapeutic effect in an injection peritonitis model. This including inhibition of bacterial translocation. Moreover, we have established an imaging methodology for live-animals can effectively evaluate intestinal permeability and aberrant bacterial translocation in IP models.

Fig 1. Colony-forming unit (CFU) counts in the animals.

Very dilute (1: 1 × 10⁶) feces samples from individual mice were cultured on agar plates containing ampicillin (100 μg/mL). Ampicillin-resistant microbes (presumably the exogenous tracer E. coli) formed clones. The numbers of clones were represented in this chart. There is no statistically significant difference between the three groups.

Fig 2. Animal model of peritonitis.

A schematic representation of the protocol used to generate the animal model carrying tracer bacteria. Tracer bacteria were administered to decontaminated C-57 mice via oral gavage. The validity of the model was confirmed in fecal cultures on the 4^th day. IP: Injective peritonitis; IAP: intestinal alkaline phosphatase.

Fig 3. Bacterial translocation to the blood stream was inhibited by intestinal alkaline phosphatase.

A: At 24 h after the induction peritonitis, blood was collected from the animals for quantitative culture. Ten microliters of whole blood was spread on a 90-mm regular LB agar plate; 12 h later, the numbers of clones were recorded. ^a P < 0.05 vs sham, ^b P < 0.05 vs injective peritonitis (IP); the data are from at least three individual animals. B: DNA from blood samples was additionally used for polymerase chain reaction (PCR) experiments. Primers specific for bacterial 16s RNA (466 bp) were used for PCR amplification in the PCR assay. A parallel PCR assay for actin was performed to ensure similar loading. C: The quantification of 16sRNA PCR experiments data; blots were analyzed using the Image J software. D: Fluorescent flow cytometry (FCM) to obtain blood bacterial count. Blood samples from groups receiving different treatments were subjected to the FCM assay. Fluorescent signal counting is represented in C; curves of typical FCM counts are shown at the top of each columns. ^c P < 0.05 vs. sham, ^d P < 0.05 vs. IP, IAP: intestinal alkaline phosphatase.

Fig 4. Blood BT assessment based on EGFP labeled tracer bacteria.

Peritonitis was induced in animals loaded with EGFP-BL21 E. coli. Twenty-four hours later, blood was collected from the animals. Plasma samples from animals receiving different treatment were subjected to an FCM assay. The fluorescent counts are shown in the bar chart. Typical FCM count curves are shown at the top of each column. ^a P < 0.05 compared with sham.

Fig 5. In vivo imaging assay showing the intestinal alkaline phosphatase-induced reduction in bacterial translocation.

A: At 24 h after the induction of peritonitis, mice were subjected to in vivo imaging, and red fluorescence was recorded. B: The fluorescence intensities in each organ area. (a) Fluorescence intensity in the liver, (b) fluorescence intensity in the lungs; ^a/c P < 0.05 vs. sham, ^b/d P < 0.05 vs. IP. The experiments were repeated in at least three individuals. IP: Injective peritonitis; IAP: intestinal alkaline phosphatase.

Fig 6. Western blotting to detect bacterial enhanced green fluorescent protein in liver and lung tissues.

At 24 hr after the induction of peritonitis, the mice were euthanized. Fresh protein samples were extracted from lung or liver tissue and were subsequently processed for Western blotting assays. A: Western blotting was performed with enhanced green fluorescent protein (EGFP)-targeted antibodies using protein from liver tissue; B: EGFP Western blotting in tissue. Three repeated experiments showed similar results. IP: Injective peritonitis; IAP: intestinal alkaline phosphatase. C: The quantification of Liver Western blotting data; blots were analyzed using the Image J software. D: The quantification of Lung Western blotting data; blots were analyzed using the Image J software.

Fig 7. Intestinal alkaline phosphatase prevented the increase in permeability following peritonitis.

At 24 h after the induction of peritonitis, intestinal permeability was examined using plasma FD40. ^a P < 0.05 compared with sham, ^b P < 0.05 compared injective peritonitis (IP). Experiments were repeated in at least three mice. IAP: Intestinal alkaline phosphatase.

Fig 8. Intestinal epithelial barrier function was enhanced by intestinal alkaline phosphatase.

Caco-2 cells were plated in the upper wells of transwell chambers. Once confluent monolayers formed, serial doses of intestinal alkaline phosphatase (0, 2, 4, 8, or 16 mIU) were added to the medium for 24 h. A: TEERs of treated monolayers were measured. B: Medium from the basal chamber was collected for HPLC measurements of the paracellular flux tracer FD-40. Data denote three independent experiments. *P<0.05 compared with controls.

Fig 9. Changes in protein expression and the time course of the changes in the ERK phosphorylation in Caco-2 cells pretreated with IAP.

A: Caco-2 cells were treated with increasing concentrations of IAP for 48 h (0, 4, 16 mIU). Whole-cell protein were extracted and subjected to Western blotting. Primary antibodies of ERK, p-ERK, SP-1, VEGF, Cdx-2 or Claudin-2 were used for the blotting assays. β-Actin immunoblotting was performed as an internal loading control. B: Caco-2 cells were treated with 16 mIU IAP for varying lengths of time (0, 0.5, 2 or 4 h). Upper figure, levels of phosphorylated ERK detected by Western blotting. C: The quantification of ERK and p-ERK Western blotting data of the dose-cause; blots were analyzed using the Image J software. D: The quantification of other proteins Western blotting data of the dose-cause; blots were analyzed using the Image J software. E: The quantification of ERK and p-ERK Western blotting data of the time-cause; blots were analyzed using the Image J software.

Fig 10. ERK phosphorylation levels and expression of related proteins following treatment with a phosphatase inhibitor in Caco-2 cells pretreated with IAP.

A: Caco-2 cells were treated with varying concentrations of IAP and sodium orthovanadate. Fresh protein samples were extracted from pretreated Caco-2 cells and were subsequently processed for Western blotting assays. Caco-2 cells were treated with increasing concentrations of IAP for 24 h (0, 4, 16 mIU) and with 15mM sodium orthovanadate. Whole-cell proteins were extracted and subjected to Western blotting. Primary antibodies of p-ERK, SP-1, VEGF, Cdx-2 or Claudin-2 were used for the blotting assays. β-actin immunoblotting was performed as an internal loading control. B: The quantification of the proteins Western blotting data; blots were analyzed using the Image J software.

Fig 11. Activity of the VEGF promoter in Caco-2 cells treated with IAP.

Caco-2 cells were treated with varying concentration of IAP. Cells were transfected with the VEGF promoter linked to firefly luciferase reporter and the renilla luciferase reporter, as a control, for 24 h and were treated with varying doses of IAP over the following 24 h. Luciferase reporter activities were examined. The activity of the firefly reporter in IAP-treated cells was divided by the activity of the Renilla luciferase in control cells. Values represent the mean ± SEM of data from three separate experiments. **P<0.01 compared with 0 mIU IAP group.

Fig 12. Activity of the VEGF promoter and the Claudin-2 promoter in Caco-2 cells treated with IAP.

A: The DNA binding domains of SP-1 and Cdx-1 respectively. B: The ChIP assay was performed to examine the DNA-binding activity of SP1 to the VEGF promoter and of Cdx-2 to the Claudin-2 promoter in Caco-2 cells receiving varying IAP treatments. C: The quantification of SP1 and VEGF promoter PCR data; the data were analyzed using the Image J software.

Fig 13. Activity of the Claudin-2 promoter in Caco-2 cells treated with IAP.

Caco-2 cells were treated with varying concentrations of IAP. Cells were transfected with the Claudin-2 promoter linked to a luciferase reporter and a Renilla luciferase reporter, as a control, for 24 h and were treated with varying dose of IAP over the following 24 h. Luciferase reporter activities were examined. The activity of the firefly reporter in the IAP-treated cells was divided by it’s the activity of the Renilla luciferase reporter in control cells. Values represent the mean ± SEM of data from three separate experiments. **P<0.05 compared with 0 mIU IAP group.

Fig 14. The change of VEGF level in Serum of mouse after treating with IAP.

IAP decreased the level of serum VEGF in injective peritonitis mice partially. *P<0.01 compared with Control. **P<0.05 compared with IP.

Fig 15. IAP inhibited VEGF and Claudin-2 expression in Caco-2 cells.

Caco-2 cells treated with IAP were used for immunohistochemical analyses of VEGF and Claudin-2. Photomicrographs were taken at 20x magnification.