Anne Fischbeck

Protein tyrosine phosphatase nonreceptor type 2 regulates autophagosome formation in human intestinal cells

Background: Autophagy is a process of central importance for maintaining cell homeostasis, survival, and the regulation of inflammation. Recent studies associated variants within the gene loci, encoding protein tyrosine phosphatase nonreceptor type 2 (PTPN2), and autophagy genes, such as autophagy‐related 16‐like 1 (ATG16L1), with chronic inflammatory disorders, such as Crohns disease (CD). We show that PTPN2 regulates autophagy in human intestinal epithelial cells (IEC) and primary colonic lamina propria fibroblasts (CLPF). Methods: Protein analysis in IEC and CLPF was performed by western blotting. Autophagososme formation was assessed by LC3B immunofluorescence or immunohistochemistry. Human intestinal tissue samples were obtained from noninflammatory bowel disease (IBD) control or from CD patients and genotyped for disease‐associated PTPN2 or ATG16L1 variations. Results: Knockdown of PTPN2 causes impaired autophagosome formation and dysfunctional autophagy resulted in increased levels of intracellular Listeria monocytogenes (LM) and elevated IEC apoptosis in response to tumor necrosis factor (TNF) and interferon gamma (IFN‐&ggr;). Similar findings were observed in primary CLPF derived from CD patients carrying the CD‐associated PTPN2 variant. Presence of the ATG16L1 variant prevented the cytokine‐induced rise in PTPN2 protein, finally resulting in impaired LC3B‐II levels in IEC. Actively inflamed intestinal biopsies from CD patients carrying either ATG16L1 or PTPN2 genetic variants revealed aberrant LC3B expression patterns when compared with samples from non‐IBD control patients. Conclusions: Our results demonstrate that PTPN2 regulates autophagosome formation in human intestinal cells. We provide a model of how a dysfunction of the CD susceptibility genes, PTPN2 and/or ATG16L1, may contribute to the onset and perpetuation of chronic intestinal inflammation. (Inflamm Bowel Dis 2011;)

Crohn's disease-associated polymorphism within the PTPN2 gene affects muramyl-dipeptide-induced cytokine secretion and autophagy.

Background: The single nucleotide polymorphism (SNP) rs2542151 within the gene locus region encoding protein tyrosine phosphatase non‐receptor type 2 (PTPN2) has been associated with Crohns disease (CD), ulcerative colitis (UC), type‐I diabetes, and rheumatoid arthritis. We have previously shown that PTPN2 regulates mitogen‐activated protein kinase (MAPK) signaling and cytokine secretion in human THP‐1 monocytes and intestinal epithelial cells (IEC). Here, we studied whether intronic PTPN2 SNP rs1893217 regulates immune responses to the nucleotide‐oligomerization domain 2 (NOD2) ligand, muramyl‐dipeptide (MDP). Materials and Methods: Genomic DNA samples from 343 CD and 663 non‐IBD control patients (male and female) from a combined German, Swiss, and Polish cohort were genotyped for the presence of the PTPN2 SNPs, rs2542151, and rs1893217. PTPN2‐variant rs1893217 was introduced into T84 IEC or THP‐1 cells using a lentiviral vector. Results: We identified a novel association between the genetic variant, rs1893217, located in intron 7 of the PTPN2 gene and CD. Human THP‐1 monocytes carrying this variant revealed increased MAPK activation as well as elevated mRNA expression of T‐bet transcription factor and secretion of interferon‐&ggr; in response to the bacterial wall component, MDP. In contrast, secretion of interleukin‐8 and tumor necrosis factor were reduced. In both, T84 IEC and THP‐1 monocytes, autophagosome formation was impaired. Conclusions: We identified a novel CD‐associated PTPN2 variant that modulates innate immune responses to bacterial antigens. These findings not only provide key insights into the effects of a functional mutation on a clinically relevant gene, but also reveal how such a mutation could contribute to the onset of disease. (Inflamm Bowel Dis 2011;)

Analysis of Sphingomyelin in Meat Based on Hydrophilic Interaction Liquid Chromatography Coupled to Electrospray Ionization−Tandem Mass Spectrometry (HILIC-HPLC-ESI-MS/MS)

The amount of sphingomyelin in different kinds of meat was analyzed by hydrophilic interaction liquid chromatography (HILIC-HPLC) coupled with electrospray ionization-tandem mass spectrometry (ESI-MS/MS). Analysis comprised sphingomyelin species with the five different sphingoid bases dihydrosphingosine (d18:0), sphingosine (d18:1(Delta4)), 4,8-sphingadienine (d18:2(Delta4,8)), 4-hydroxysphinganine (phytosphingosine (t18:0)), and 4-hydroxy-8-sphingenine (t18:1), and fatty acids with 12-26 carbon atoms as well as their (poly)unsaturated (up to four double bonds) and monohydroxylated analogues. Most sphingolipids contained sphingosine (d18:1) as the predominant sphingoid base, while stearic acid and palmitic acid were found as prevalent fatty acids. Total amounts vary from 361-471 mg/kg, whereas the meat of the wild animals showed considerably lower amounts.

Sphingomyelin and phosphatidylcholine contrarily affect the induction of apoptosis in intestinal epithelial cells.

SCOPE The major alimentary sources for the plasma membrane lipid sphingomyelin (SM) are dairy products, eggs, and meat. We recently reported that the SM metabolite ceramide induces cathepsin D mediated apoptosis in murine intestinal epithelial cells (IECs) and increases inflammation in acute colitis. We investigated the impact of SM and phosphatidylcholine on apoptosis in human IECs and point out BH3-interacting death agonist (BID) as link between cathepsin D and apoptosis. METHODS AND RESULTS HT-29 and isolated human IECs were stimulated with SM or phosphatidylcholine. SM treatment resulted in increased apoptosis. Phosphatidylcholine showed contrary effects. Western revealed higher amounts of cathepsin D and BID activation upon lipid stimulation. Western blotting revealed BID activation through SM in both an induced and a spontaneous mouse model of colitis. CONCLUSION Dietary phospholipids may induce or abolish apoptosis in IECs and seem to play a role in the pathogenesis of inflammatory bowel diseases. This nutritional factor might be considered when evaluating the pathogenesis of inflammatory bowel diseases. Effects of SMase- and SM treatment on inflammation in dextran sulfate sodium induced animal models of colitis and in vitro experiments are discussed as controversial. Variable sources of SM, feeding techniques, and mouse strains might play a role.

Regulation of the expression of chaperone gp96 in macrophages and dendritic cells.

The chaperone function of the ER-residing heat shock protein gp96 plays an important role in protein physiology and has additionally important immunological functions due to its peptide-binding capacity. Low amounts of gp96 stimulate immunity; high quantities induce tolerance by mechanisms not fully understood. A lack of gp96 protein in intestinal macrophages (IMACs) from Crohn`s disease (CD) patients correlates with loss of tolerance against the host gut flora, leading to chronic inflammation. Since gp96 shows dose-dependent direction of immunological reactions, we studied primary IMACs and developed cell models to understand the regulation of gp96 expression. Induction of gp96-expression was higher in in vitro differentiated dendritic cells (i.v.DCs) than in in vitro differentiated macrophages (i.v.MACs), whereas monocytes (MOs) expressed only low gp96 levels. The highest levels of expression were found in IMACs. Lipopolysaccharide (LPS), muramyl dipeptide (MDP), tumour necrosis factor (TNF), and Interleukin (IL)-4 induced gp96-expression, while IL12, IL-17, IL-23 and interferon (IFN)-γ were not effective indicating that Th1 and Th17 cells are probably not involved in the induction of gp96. Furthermore, gp96 was able to induce its own expression. The ER-stress inducer tunicamycin increased gp96-expression in a concentration- and time-dependent manner. Both ulcerative colitis (UC) and CD patients showed significantly elevated gp96 mRNA levels in intestinal biopsies which correlated positively with the degree of inflammation of the tissue. Since gp96 is highly expressed on the one hand upon stress induction as during inflammation and on the other hand possibly mediating tolerance, these results will help to understand the whether gp96 plays a role in the pathophysiology of inflammatory bowel disease (IBD).

Knock-Out of β-Glucosidase 2 Has No Influence on Dextran Sulfate Sodium-Induced Colitis

Background/Aims: The non-lysosomal glucosylceramidase, β-glucosidase (Gba2), hydrolyzes glucosylceramide to glucose and ceramide (Cer). Cer is a potent second-messenger lipid that plays an important role in signaling cascades involved in apoptosis. The aim of this study was to investigate whether Gba2 knock-out (Gba2–/–) affects the extent of dextran sulfate sodium (DSS)-induced colitis in mice. Methods: Acute colitis was induced in wild-type (WT) and Gba2–/– mice by administration of 2% DSS in drinking water. After 7 days, mice underwent colonoscopy and were sacrificed. Results: Both DSS-treated WT (n = 10) and Gba2–/– (n = 12) mice showed elevated histological and endoscopic scores compared to respective H2O controls (n = 9 each). However, no significant differences between the DSS groups were detected. Flow cytometric analysis of propidium iodide staining, cleavage of caspases-3 and -8, indicative for apoptosis, as well as Cer levels were not altered in DSS-treated WT or Gba2–/– mice. Gba2–/– resulted in slightly decreased expression of glucocerebrosidase (Gba1) as well as in upregulation of proteins being involved in cellular regeneration, such as STAT3 (signal transducer and activator of transcription), JNK and iNOS, upon DSS treatment. Conclusion: We demonstrate that Gba2–/– does not affect the extent of DSS-induced inflammation in mice, however, it might be involved in tissue regeneration in response to toxic agents.

Effects of retinoids in mouse models of colitis: benefit or danger to the gastrointestinal tract?

Background:In vitro and in vivo data have shown that retinoid treatment promotes an anti-inflammatory milieu with few adverse effects toward the gastrointestinal tract. The in vivo studies reported here further evaluate retinoid effects in 2 mouse models of inflammatory bowel disease. Method:Chronic dextran sulfate sodium colitis was induced in age- and weight-matched C57Bl/6 mice by 4 cycles of dextran sulfate sodium administration (6–8 animals/group). At cycle 4, animals were administered 13-cis-retinoic acid (isotretinoin, 30 mg/kg) or vehicle (oral gavage) or 4-oxo-13-cis-retinoic acid (15 mg/kg, intraperitoneal) daily. T-cell transfer colitis was induced in CB17 SCID mice by transfer of naive CD4+CD62L+ T cells and treated by transfer of regulatory CD4+CD25+ T cells (4–6 animals/group); isolated from BALB/c mice after treatment with isotretinoin or vehicle, as above, for 2 weeks. Assessments included endoscopic and histological scores, myeloperoxidase activity, serum cytokines, and plasma isotretinoin levels. Results:Retinoid-treated animals with colitis showed comparable changes in myeloperoxidase activity, and endoscopic and histological scores, versus untreated animals with colitis. Modest and comparable changes were seen in body weight and colon length in animals injected with naive T cells from isotretinoin-treated donors versus those injected with T cells from vehicle-treated donors. Retinoid treatment was consistently associated with lower interleukin-12 levels, which, after the transfer of naive T cells from isotretinoin-treated donors, supported isotretinoin-mediated predisposition of naive T cells toward reduced proinflammatory cytokine expression. Colitis had no effect on isotretinoin exposure. Conclusions:Retinoids attenuate the proinflammatory cytokine response in vivo, with only modest effects on body weight and parameters of gastrointestinal morphology.

Exogenous Heat Shock Protein gp96 Ameliorates CD4+CD62L+ T-Cell–mediated Transfer Colitis

Background:The heat shock protein gp96 is an endoplasmic reticulum chaperone involved in endoplasmic reticulum stress reactions. gp96 binds antigens and is secreted into extracellular space on cell stress. After reinternalization by antigen presenting cells, antigens can be transferred to major histocompatibility complex molecules. In recent studies, we found induction of gp96 during differentiation of intestinal macrophages, whereas it was absent in intestinal macrophages of patients with Crohns disease. Methods:To study immuno-modulating effects of gp96 in T-cell transfer colitis BALB/c donor mice were injected with 2 × 100 &mgr;g gp96. After 1 week, 2.5 × 105 CD4+CD62L+ cells were isolated from spleens and injected into severe combined immunodeficiency recipients. Another group received cells from untreated donors and was treated with 100 &mgr;g gp96 after transfer. Control groups received cells from untreated donors, or buffer alone. Results:After transfer of CD4+CD62L+ T cells from gp96-pretreated donors, mice (TBT gp96) showed an initial weight loss, but after 3 weeks, they recovered and reached the starting weight after 5 weeks. Mice treated with gp96 after transfer (TAT gp96) showed a delayed weight loss in comparison with the CD4+CD62L+ group. The histological scores in CD4+CD62L+ mice were 2.6 ± 0.1, in TBT gp96 mice 1.3 ± 0.3 (CD4+CD62L+ versus TBT gp96: P < 0.05) and in mice treated after transfer 1.9 ± 0.1 (CD4+CD62L+ versus TAT gp96: P < 0.05). Conclusions:These findings indicate an essential role of gp96 in the maintenance of tolerance against luminal antigens in the intestinal mucosa. The absence of gp96 in intestinal macrophages of patients with Crohns disease might provoke loss of this tolerance mediating mechanism.

Loss of Heat Shock Protein gp96 Does Not Impair Toll Like Receptor Signaling In Vitro

Background: The heat shock protein gp96 is an endoplasmic reticulum chaperone for multiple protein substrates. We showed that a lack of gp96 in intestinal macrophages of Crohns Disease (CD) patients is correlated with loss of tolerance against the host gut flora, leading to chronic inflammation. A recent manuscript suggested gp96 to be the major chaperone for TLRs in murine macrophages. Therefore, we studied the impact of gp96 knockdown on TLR function in the monocytic cell line Mono Mac 6 (MM6). Methods: For stable gp96 knockdown a lentiviral system was used. Stimulations were performed with lipopolysaccharide (LPS), a ligand of TLR 4. Internalization was shown using FITC-labeled LPS. TLR folding and functionality was investigated by Western blotting and flow cytometric analysis. IL-8 secretion upon stimulation with LPS was determined by ELISA. Results: Flow cytometric analysis of TLR 2 and 4 showed similar patterns on the cell surface of gp96knockdown cells as well as on control MM6 cells. Internalization of LPS by gp96-knockdown cells was shown using a FITC-labeled derivative. No reduction by gp96 knockdown was observed. Western blot analysis of phospho-IκB/IκB and phospho-NFκB/NFκB ratios did not reveal a significant difference in TLR mediated NFκB signaling between gp96 knockdown and control cells. Induction of IL-8 secretion upon stimulation with LPS was comparable between knockdown and control cells. Conclusions: In contrast to recent reports loss of gp96 does not have pivotal effects on functionality of TLRs in human MM6 cells. These findings indicate that the loss of tolerance against commensal gut flora is caused by different mechanisms yet to be identified.

The Crohn's Disease Candidate Gene, Protein Tyrosine Phosphatase Non-Receptor Type 2, Regulates Muramyldipetide-Induced NOD2-Dependent Effects in Human Monocytes and Fibroblasts

Background: Mutations within the gene loci, encoding protein tyrosine phosphatase nonreceptor type 2 (PTPN2) and nucleotide oligomerization domain 2 (NOD2), have been associated with Crohns disease (CD). We have recently shown that PTPN2 limits proinflammatory effects in human intestinal epithelial cells (IEC) and monocytes. A dysregulated immune response to bacterial antigens, such as the NOD2-ligand, muramyl-dipeptide (MDP), has been strongly implicated in the pathogenesis of CD. The transcription factor T-bet is upregulated in NOD2-deficient mice, activated during CD and controls the secretion of proinflammatory cytokines, such as IFNγ. Here, we studied whether PTPN2 controls NOD2mediated effects in human THP-1 monocytes. Methods: Protein analysis was performed by Western blotting, mRNA analysis by real-time PCR, cytokine secretion by ELISA and visual imaging by immunofluorescence (IF) studies. PTPN2 or NOD2 knock-down was induced by siRNA. Primary intestinal colonic lamina propria fibroblasts (CLPF) were isolated from surgical specimens derived from patients with active CD (n=5) and were genotyped for the CD-associated PTPN2 variant. Results: MDP (100 ng/ml) increased PTPN2 mRNA (p<0.05; n=3) and protein expression (p<0.05; n=3) by 24 h treatment. This effect was absent in NOD2 knock-down cells (n=3). Levels of the transcription factor T-bet (p<0.01) were also enhanced by MDP stimulation and this effect was potentiated by PTPN2 knock-down (p<0.05; n=3). While MDP treatment induced secretion of TNF, IL-8 and IFNγ (p<0.01; n=3 each) in THP-1 cell supernatants, knock-down of PTPN2 by siRNA resulted in diminished secretion of all three cytokines in response to MDP (p<0.01; n=3 each). We next studied whether PTPN2 regulates NOD2-mediated autophagy. PTPN2 knock-down prevented the MDP-induced expression of microtubule-associated protein 1 light-chain 3 B-II (LC3B-II), a marker of autophagy activity, in THP-1 cells (n=3; p<0.05). Though MDP was able to increase PTPN2 and LC3B-II protein levels in PTPN2 wild-type (WT) CLPF (n=3), in CLPF carrying the CD-associated PTPN2 variant (n=2), MDP was unable to increase PTPN2 or LC3B-II expression. By IF studies we found elevated numbers of LC3B+ vesicles, indicating autophagosome assembly, in MDP-treated PTPN2-WT cells. However, in PTPN2-variant cells LC3B+ vesicles were less numerous, but larger in size, when compared to PTPN2WT cells, suggesting that defective autophagosome formation occurred. Conclusions: We demonstrate that PTPN2 is regulated by MDP via a NOD2-dependent mechanism and controls NOD2-mediated cytokine secretion and autophagy. These findings demonstrate for the first time a functional interaction between the two CD candidate genes, PTPN2 and NOD2, and suggest that the CD-associated PTPN2 variant could favour a dysregulated immune response to bacterial antigens.