| Background&ObjectivesBiointerfaces between gastric mucosal epithelium and its surroundings such as microbe and food play an important role in digestion and absorption of nutrition and prevention from pathogenic factor. Especially, due to the serious consequences after Helicobacter pylori(H.pylori) infection, studying on adhesive behaviours of H.pylori-gasttic mucosal epithelium (H.pylori-GME) biointerface is critical for understanding mechanisms of H.pylori infection and prevention and treatment of gastroenterological diseases induced by H.pylori infection.Clinical-pathological studies demonstrate that about21%of total H.pylori tightly contact with gastric mucosal epithelial cells after invasion to our stomach. By the method of molecular physiology, we found that the interactions of outer membrane proteins (OMPs) of H.pylori and their ligands on the surface of GME cells can facilitate stable adhesion of H.pylori to GME cells. For instance, BabA and SabA commonly expressed on surface of H.pylori can keep stable adhesion and initiate inflammation of GME through specially recognizing Lewisb and sialylated-Lewisx, respectively. CagL can initiate type IV secretion system of H.pylori though interaction with integrin β1, which would deliver cytotoxin-associated gene A protein (CagA) into GME cells. Other adhesins such as HopZ, AlpA/AlpB and OipA as targets for immunity adaptive against H.pylori infection also participate in the adhesive interaction between H.pylori and GME cells. But other properties of H.pylori-GME biointerface would also influence adhesive behaviours of H.pylori on the GEM.Through topography, chemical molecular and viscoelastic patterns on them, biointerfaces match molecules at nano-scales and match cells, tissues at micro-scales to keep integrity of medical functions. Chemical molecular is the smallest unit which determines architectures and properties of our tissue. From nano-to micro-scale, chemical molecular ultimately self-assemble to micro/-nanostructures (topography) and reflect special mechanic characters (viscoelastic patterns), which influences macroscopic functions of our tissues and even organs. This is so-called the theory of "mirco/-nanotopography-coupled-mechanical (TCM) action. Therefore, besides key interaction of OMPs and their ligands, effects of the topographical and mechanical properties of H.pylori-GME biointerface on bacterial adhesive behaviours should be appreciated. But due to the limitations of technologies and methods, there is little recognition in these fields.In this work, we mimicked the characteristic topography of H.pylori-GME biointerface by nanometer-resolution fabrication techniques and then study its effects on bacterial adhesion in vitro. In addition, we also simulated the mechanical properties of extracellular matrix of GME with stiffness-controlled macromolecular materials. And we combined mechanical characterization with molecular biology to explore the mechanism of its effects on H.pylori adhesion to gastric mucosal epithelium. MethodsFirstly, we ascertained micro/-nanotopographical parameters of gastric mucosal epithelium under different pathological stages according to reports of the clinical pathology associated with infection of H.pylori. Then we fabricated mimic nano-structures of H.pylori-GME biointerface by reactive ion etching technology (RIE) in vitro and explored the underlying mechanism based on the adherent assay and morphological analysis of three species of bacteria(Escherichia coli, Staphylococcus aureus and Helicobacter pylori).Secondly, we also mimicked the stiffness of the extracellular matrix under gastric mucosal epithelium with polyacrylamide (PAAm) hydrogel. And we studied its effects on the adhesion of H.pylori to GES-1cells and explored the underlying mechanism by atomic force microscopy (AFM) and molecular biological methods.At last, we used different morphological gold nanoparticles to characterize proteins on the surface of A549cells based on imaging advantages of Environmental scanning electronic microscopy (ESEM) for biomaterial samples, which would be a new method for further study on molecule mechanisms of topographical and mechanical effects on adhesion of H.pylori to gastric mucosa epithelial cells.Statistical analysis in our study was performed using Spss13.0software. Continours variables were showed as mean±standard deviation (x±s) and their statistical significance were evaluated by accoresponding methods of paremeter analyses. Conversely, non-continours variables were expressed by median and quartiles (M (Q1; Q3)) and their statistical significance were evaluated by accoresponding methods of non-paremeter analyses.Results1. Effects of mirco/-nanotopography of mimic H.pylori-GME biointerface on bacterial adhesion. 1.1. Topographical and mechanical properties of mimic H.pylori-GME biointerface.According to clinical-pathological studies on H.pylori infection, the mirco/-nanotopography of normal is consisted of microvilli arrays with about200nm in diameter,700nm in height and200nm in spacing. But GEM microvillus would be shortened, ablated and transformed to intestinal ones after H.pylori infection. Thus, the characteristic topography of H.pylori-GME biointerface is the microvilli arrays with gradual changes of the height and spacing. For sake of mimicking the characteristic topography of H.pylori-GME biointerface, we have successfully fabricated PET films to nanopillar arrays with the diameter of about250nm and different heights (200nm,700nm and1000nm) and interpillar spacings (50nm,200nm and400nm) by reactive ion etching (RIE).The physical properties of different PET substrates were eveluated by AFM and the sessile drop method. The results were showed as x±s. Our results showed that PET nanopillar arrays was rougher and less hydrophobic (Ra=17.4-257.9; Rq=21.1～333.4; WCA=100.3-117.4) than the PET smooth film (95%confidence interval for Ra, Ra, and WCA were1.8-3.6,2.3-4.4and100.3-117.4, respectively). Moreover, through the factorial analysis of physical properties of PET nanopillar arrays, we found that the roughness and wetting of PET nanopillar arrays were influenced by the nanopillar height (Ra F=143.480, P<0.001; Rq F=182.213, P <0.001; WCA F=139.532, P<0.001) and interpillar spacing (RaF=887.891, P<0.001; Rq F=928.372, P<0.001; WCA F=45.894, P<0.001). Furthermore, the results from multiple comparisons using the Bonferroni or Dunnett T3method showed that increasing nanopillar height increased the roughness of nanopillar arrays with interpillar spacing of200and400nm (Ra P=0.014, Rq P=0.063; Ra P <0.001, Rq P<0.001), but there was no statistically significant difference in the roughness of nanopillar arrays with interpillar spacing of50nm (Ra P=0.209; Rq P =0.393). In addition, there was statistically significant difference in the roughness of nanopillar arrays with different interpillar spacings (P<0.001). And the roughness of nanopillar arrays rised with increasing interpillar spacings (all P<0.050). On the other hand, increasing nanopillar spacing increased the WCA of nanopillar arrays with nanopillar height of700and1000nm (P=0.004, P<0.001), but there was no statistically significant difference in the roughness of nanopillar arrays with height of200nm (P=0.548). In addition, there was statistically significant difference in the WCA of nanopillar arrays with different nanopillar height (F=139.532, P<0.001). And the WCA of nanopillar arrays rose with increasing nanopillar height (all P<0.001).1.2. Bacterial adhesion assay.PET nanopillar arrays with gradual changes of the nanopillar height and interpillar spacing influenced the adhesion of three species of bacteria. The results were showed as x±s. Through the factorial analysis of feature sizes of PET nanopillar arrays, we found there was no statistically significant difference in the number of bacteria adherent to the PET nanopillar arrays with different nanopillar height (H.pylori43504F=0.151, P=0.860; S.aureus6538F=1.508, P=0.350; E.oli BL21F=0.019, P=0.982). But effects of interpillar spacing on the number of bacteria adherent to the PET nanopillar arrays had the same trend (H.pylori43504F=480.631, P<0.001; S.aureus6538F=879.873, P<0.001; E.coli BL21F=250.481, P<0.001). Furthermore, the results from multiple comparisons using the Dunnett T3method showed that there were more bacterial cells of H.pylori43504and S.aureus6538adherent to PET nanopillar arrays with interpillar spacing of50nm compared with the PET nanosmooth substrate (104.6±12.7vs69.7±9.3, P<0.001;190.4±21.8vs134.0±11.7, P<0.001). But for E.coli BL21, it showed no statistical significance (46.8±12.5vs51.7±14.9, P= 0.830). In addition, increasing interpillar spacing inhibited bacterial adhesion to nanopillar arrays (H.pylori4350458.4±7.6and45.3±6.6vs104.6±12.7, P<0.001; S.aureus653854.9±12.5and86.9±12.5vs190.4±21.8, P<0.001; E.coli BL2126.9±7.1and5.5±3.7vs46.8±12.5, P<0.001). But compared with interpillar spacing of200nm, PET nanopillar arrays with interpillar spacing of400nm promoted adhesion of S.aureus6538(54.9±12.5vs86.9±12.5, P<0.001).We also detected the morphological variances of each species adherent to PET nanopillar arrays and nanosoomth substrates (control) using CellTool software for the quantitative analyses of bacterial morphology. The results were expressed by M (Q1;Q3)-Through non-paremeter analysis of the Kruskal-Wallis H test, we found that topography of PET nanopillar arrays also influenced morphologies of adherent bacteria(E.coli BL21X2=26.428, P<0.001; S.aureus6538X2=73.002, P<0.001; H.pylori43504X2=20.257, P<0.001). Furthermore, the results from comparisons between any two groups using the Wilcoxon rank sum test showed that the major morphological variation of E.coli BL21adherent to different PET substrates was the centerline length of bacterial cells. Compared to nanosmooth substrates, the average of the centerline length of the E.coli BL21adhesive on nanopillar PET pattern was shorter (P<0.001). And the centerline length decreased with the increasing interpillar spacing. Though there were three highly explanatory modes of shape variation according to the result of principle component analyses (PCA), the mean diameter of S.aureus6538was selected to explain the major morphological variation because that the most principle component was the shape size and another two modes could be expressed partly by the parameter. The mean diameters of S.aureus6538on the nanopillar arrays were larger than the one on the nanosmooth substrates (P<0.001). And there was decreasing trend of it with the increasing interpillar spacing. Just as the result of S.aureus6538, more than one highly explanatory modes of shape variation of H.pylori43504were found. The centerline length and the side curvature of the bacterial cells were the major morphological parameters to express the shape variation. Therefore, the ratio of side curvature to length (RCL) of the bacterial cells was used to explain integrated variations instead. The RCL of the bacterial cells on the nanopillar arrays with400nm interpillar spacing was larger than the one on any other PET substrates (P<0.001). And there was increasing trend of it with the increasing interpillar spacing.2. Effects of the stiffness of H.pylori-GME biointerface on H.pylori adhesion.2.1. Effects of the stiffness of PAAm hydrogel substrates on the adhesion of Kpylori to GES-1cells.The extent of H.pylori adherent to GES-1cells on different stiffness of PAAm hydrogel substrates was evaluated by ESEM and the results were showed as x±s. H.pylori tended to adhere to GES-1cells at the edge of cytoplasm by the interaction with microvillus. Through the one-way ANOVA, we found that increasing stiffness of PAAm hydrogel substrates facilitated bacteria adhesion to cells (F=55.880, P<0.001). Furthermore, the results from multiple comparisons using the Dunnett T3method showed that there was the maximum numbers of adherent H.pylori to single cell on the PAAm hydrogel substrate with intermediate stiffness (3.8kPa)(17.6±4.4vs5.9±2.2, P<0.001;17.6±4.4vs14.2±3.6, P=0.003;17.6±4.4vs14.3±4.0, P=0.005). But the number of H.pylori on the cells surface showed no significant difference between PAAm hydrogel substrates with of8.4kPa and21.5kPa (14.2±3.6vs14.3±4.0, P=1.000). In addition, the substrates with different stiffness didn’t affect the adhesion of H.pylori directly. Few H.pylori adhered to the PAAm hydrogel substrates functionalized with Collagen I (X2=1.738, P=0.628). 2.2. Effects of the stiffness of PAAm hydrogel substrates on the morphological and mechanical properties of GES-1cells.It was easier for GES-1cells to spread on stiffer PAAm hydrogel substrates. On the softest substrates (0.7kPa), GES-1cells had the sphere morphology with more and short microvillus on their surfaces. When the stiffness of substrates rose up to8.4kPa, the cells spread completely and there were less microvillus on the surface and more filopodia at the edge of cytoplasma.The physical properties of micro-region on the GES-1cell were eveluated by AFM and the results were showed as M (Q1; Q3). Through non-paremeter analysis of the Kruskal-Wallis H test, we found that the stiffness of micro-region on the GES-1cell showed the statistically significant difference on PAAm with different stiffness (the margin of cytoplasma, X2=47.025, P<0.001, the nucleus, X2=14.075, P=0.003). Furthermore, the results from comparisons between any two groups using the Wilcoxon rank sum test showed that the stiffness of micro-region on GES-1cell increased with stiffer substrates. But the stiffness of the margin of cytoplasma did not show remarkable difference on the8.4kPa and21.5kPa hydrogel substrates (P=0.193), while the stiffness of nucleus did (P=0.003). In addition, the results from paired comparisons using the Wilcoxon signed rank test showed that the stiffness of cytoplasma margin was always larger than that of the nucleus on each substrate (t=3.762, P<0.001;t=2.680, P=0.007; t=4.008, P<0.001; t=5.199, P<0.001), the adhesion work between GES-1and microsphere showed statistical significance (the margin of cytoplasma, X2=52.163, P<0.001; the nucleus, X2=46.775, P<0.001). In addition, the adhesion work at the margin of cytoplasma was larger than that of nucleus (t=2.584, P=0.010; t=3.895, P<0.001; t=5.470, P<0.001), except on the hydrogel substrates of3.8kPa (t=0.355, P=0.723). 2.3. Effects of the stiffness of PAAm hydrogel substrates on the expression of integrinβ1in GES-1.The expression of integrin β1in GES-1cultured on PAAm hydrogel substrates with different stiffness was detected by the means of western blot and immunofluorescence microscopy and the results were showed as x±s. Through the one-way ANOVA, we found that expression of integrin β1in GES-1was influenced by stiffness of PAAm hydrogel substrates (F=250.372, P<0.001). Furthermore, the results from multiple comparisons using the Dunnett T3method showed that there was the highest expression level of integrin β1on the substrate of intermediate stiffness (8.4kPa)(2.3±0.2, all P<0.050), which was the twice of that on the substrate of0.7and21.5kPa.3. Characterization of multi-proteins within biointerface.By using immunized gold nanoparticles with different morphologies, we successfully characterize insulin receptor and caveolin-1on the surface of A549cells. We found that immunized gold nanoparticles could recognize these two proteins respectively and there was the overlap between their distributions after exclusion of non-special absorption.Conclusions1. The micro/-nanotopography of H.pylori-GME biointerface is the microvilli arrays with gradual changes of the height and spacing. H.pylori-GME biointerface is rougher and more hydrophobic with increasing height and spacing. And it inhibited bacterial adhesion and selected bacterial cells with special morphologies when spacing between microvillus increased.2. The mechanical properties and expression level of key proteins within H.pylori-GME biointerface were influenced by the stiffness of extracellular matrix of GME. Adhesion of H.pylori to gastric epithelial cells were promoted by rising the expression level of integrin (31rather than increasing the stiffness of gastric epithelial cells, which was regulated by the stiffness of extracellular matrix of GME.3. The method of characterizing multi-proteins with immunized gold nanoparticles of different morphologies is a feasible tool for study on interaction between proteins within biointerfaces. |
PDF Full Text Request