PMC:4236617 / 23135-40909 JSONTXT

{"project":"2_test","denotations":[{"id":"25070625-18029488-12189978","span":{"begin":1474,"end":1476},"obj":"18029488"},{"id":"25070625-18261863-12189979","span":{"begin":2603,"end":2605},"obj":"18261863"},{"id":"25070625-19047644-12189980","span":{"begin":5973,"end":5975},"obj":"19047644"},{"id":"25070625-19047644-12189981","span":{"begin":7992,"end":7994},"obj":"19047644"},{"id":"25070625-15034147-12189982","span":{"begin":9552,"end":9554},"obj":"15034147"},{"id":"25070625-17654362-12189983","span":{"begin":9612,"end":9614},"obj":"17654362"},{"id":"25070625-14530136-12189984","span":{"begin":9672,"end":9674},"obj":"14530136"},{"id":"25070625-17544232-12189985","span":{"begin":10288,"end":10290},"obj":"17544232"},{"id":"25070625-23677442-12189986","span":{"begin":12557,"end":12558},"obj":"23677442"},{"id":"25070625-19304849-12189987","span":{"begin":14310,"end":14312},"obj":"19304849"},{"id":"25070625-23354479-12189988","span":{"begin":14313,"end":14315},"obj":"23354479"}],"text":"Results\n\nAdherence of L. amylovorus strains to mucus\nThe adherence of the L. amylovorus strains to commercially available porcine gastric mucins and to mucus isolated from the small intestine of a 8-week old pig were examined using L. amylovorus cells labeled by the DNA-binding stain SYTO®9 (Figure 1). None of the strains bound extensively to porcine gastric mucins, i.e., typically less than 1% of the original amount of cells remained mucin-bound (Figure 1A). The same was true for porcine intestinal mucus, where the proportion of adhering bacterial cells was usually 2% or less (Figure 1B). The very high variation between the experiments and the lack of any consistent dose–response of binding (data not shown) supported the conclusions.\nFigure 1 Adherence of L. amylovorus strains to mucus. The adherence of L. amylovorus strains to immobilized porcine gastric mucin (type II, Sigma) (A) and to mucus purified from porcine small intestine (B) was studied using bacterial cells labeled with the fluorescent dye SYTO®9. The means and standard deviations of 5–8 independent experiments are shown, each with three technical replicates.\n\nAdherence of L. amylovorus strains to IPEC-1 cells\nIn contrast to mucus binding, clear differences were observed in the adherence of the L. amylovorus strains to porcine small intestinal epithelial cells as represented by the cell line IPEC-1 (Figure 2A). The previously reported binding of L. amylovorus DSM 16698 to IPEC-1 cells [26] was confirmed, and the adherence of the strains GRL 1112 and GRL 1115 was found to be within a similar range, though the three strains displayed high variabilities in different experiments. In contrast, the type strain of L. amylovorus, DSM 20531T, isolated from silage, and the rest of the porcine intestinal isolates were clearly less adhesive. An example of the dose–response of binding is shown in Figure 2B.\nFigure 2 Adherence of L. amylovorus strains to IPEC-1 cells. A) The adherence of 3H-labeled L. amylovorus strains (A600nm = 0.25) to IPEC-1 cells grown on Thincert™ wells showing the means and standard deviations of 3–7 independent experiments, each with three technical replicates. B) An example of the adherence of L. amylovorus DSM 16698, GRL 1112, GRL 1115 and GRL 1118 to IPEC-1 cells as the function of cell density. The means of three technical replicates from one representative experiment are shown. Dpm, disintegrations per minute.\n\nInhibition of F4-fimbriated ETEC adherence to IPEC-1 cells by L. amylovorus\nIPEC-1 cells have been shown to support the adherence of E. coli carrying F4-type fimbriae [38]. Next the L. amylovorus strains were tested in three different experimental set-ups (exclusion, displacement and competition as described in Methods) to evaluate if the observed differences in their adherence to IPEC-1 cells correlated with their abilities to inhibit ETEC adherence in the same model. The results were evaluated by comparing the adherence of ETEC in the presence or absence of the L. amylovorus strains. The strains DSM 16698, GRL 1112, GRL 1115 and GRL 1118 were able to inhibit pathogen adherence if they were added beforehand (exclusion, Figure 3A) or simultaneously with ETEC (competition, Figure 3B); the strain DSM 20531T achieved only a borderline inhibition when added beforehand (Figure 3A), and the rest of the strains had a negligible or even a slightly enhancing effect on ETEC binding in both assays (Figures 3A and B). Importantly, none of the strains was able to displace previously bound ETEC from IPEC-1 cells (displacement, Figure 3C).\nFigure 3 Inhibition of F4-fimbriated ETEC adherence to IPEC-1 cells by L. amylovorus. The inhibition of F4-fimbriated ETEC adherence to IPEC-1 cells by the indicated L. amylovorus strains in exclusion (A), competition (B) and displacement assays (C) was tested with 3H-labeled ETEC cells as detailed in Methods. The means and standard deviations of 3–7 independent experiments are shown, each with three technical replicates.\n\nInhibition of pathogen growth by the culture supernatants of L. amylovorus\nThe filter-sterilized culture supernatants of the L. amylovorus strains were assayed for their abilities to inhibit the growth of various intestinal pathogens (Figure 4). All the supernatants markedly inhibited the growth of the test pathogens. For instance, the supernatants of the strains DSM 20531T and GRL 1117 reduced the growth of F4-fimbriated E.coli by more than 100 000-fold and the growth of Salmonella typhimurium almost by a factor of 10 000. The growth of F4-fimbriated E. coli was most efficiently inhibited. The pH values of the supernatants varied from 3.8 to 4.5. It is notable that the reductions in pathogen counts inversely correlated with the pH values of the supernatants (Figure 4), and culture supernatants which had been adjusted to the pH of plain MRS lowered the pathogen counts by much less than tenfold (data not shown), indicating that the inhibition was mainly due to the low pH associated with lactic acid production.\nFigure 4 Reductions in pathogen counts by L. amylovous culture supernatants. Six different swine intestinal pathogens were grown in TSB medium in the presence of the filter-sterilized supernatants of the L. amylovorus strains, and the reductions in pathogen counts, expressed as log CFU values, were estimated from the area reduction percentages (ARPs) of the pathogen growth curves by linear regression. The average pH values of the supernatants are shown above the histograms. The results are the means and standard deviations of three independent experiments, each performed with fresh culture supernatants with three technical replicates.\n\nCytokine induction in moDCs by L. amylovorus\nThe S-layer-carrying L. acidophilus strain NCFM interacts with human DCs eliciting an anti-inflammatory IL-10 response and it promotes the Th2-differentiation of T-cells through DC:s; the S-layer protein has been shown to have a role in this response [12]. Prompted by these findings, we examined the potential of the phylogenetically closely related, S-layer-carrying L. amylovorus strains to induce immune signaling in human DCs. As shown in Figure 5, when tested at the bacteria/DC ratio of 100:1, clear differences between the levels of cytokines induced by the strains were observed. Interestingly, the anti-inflammatory response induced by L. acidophilus NCFM was not observed with the L. amylovorus strains. Instead, our strains typically induced a mixed cytokine response with the release of both proinflammatory (TNF-α, IL-6, IL-1β, IL-12, IP-10/CXCL10) and anti-inflammatory (IL-10) cytokines from human DCs. Furthermore, the strain GRL 1116, which was most potent at inducing proinflammatory cytokines, induced also the highest levels of the anti-inflammatory cytokine IL-10. Analogously, the strain DSM 20531T and GRL 1115 were among the weakest inducers of both pro-and anti-inflammatory cytokines. At the lower MOI values of 1 and 10, no clear induction of any of the cytokines was observed in comparison to the negative control (data not shown).\nFigure 5 Cytokine induction in human dendritic cells by L. amylovorus. The extents of induction of TNF-α (A), IL-1β (B), IL-6 (C), IL-10 (D), IL-12 (E) and IP-10/CXCL10 (F) in human monocytic dendritic cells (moDCs) were tested after treating the cells with L. amylovorus strains for 24 hours at the bacteria/DC ratio 100:1. The data are presented as the means and standard deviations from one representative experiment out of three, performed with moDC:s of four donors.\n\nGenomic characterization of L. amylovorus Slp:s\nTo initiate comparative studies on the role of L. amylovorus surface layer proteins in the probiotic interactions described above, the numbers and sequence similarities of the slp genes in the genomes of the strains were initially analysed. The genomic investigation of the eight strains revealed several slp genes in each strain. Genes with homology to L. acidophilus NCFM slpA and slpB[12] were identified, and the homologous L. amylovorus genes were named slpA and slpB, respectively. Furthermore, slp-like genes of a third type were detected in all of the eight genomes and these were designated as slpC. The slp sequences, along with the deduced amino acid sequences, are shown in Additional file 1. All the eight strains studied carried only one slpA-homologue, except for GRL 1117, which had two distinct slpA-like genes (slpA1 and slpA2). Only one slpB-homologue was identified in GRL 1112, 1114, 1115, 1116, 1117 and 1118 as well as in DSM 16698, whereas DSM 20531T carried two slpB-like genes (slpB1 and slpB2). The highest variation was found in the number of slpC-type genes: strains DSM 20531T and GRL 1115 carried one, DSM 16698 possessed three (slpC1, slpC2, slpC3), and the rest of the strains had two slpC-type genes (slpC1 and slpC2). Exceptionally, the gene slpC3 of DSM 16698 was found to be located on a plasmid. A phylogenetic tree was constructed based on the deduced amino acid sequences of the slpA, slpB and slpC gene products (Figure 6). The tree clearly shows that the SlpA-like sequences have diversified most during evolution, while the SlpB-type proteins have remained more similar to each other whereas the predicted SlpC-type proteins could be clustered into three distinct groups.\nFigure 6 Phylogeny of L. amylovorus Slp:s. A neighbour-joining phylogenetic tree based on L. amylovorus Slp sequences was generated by creating a multiple amino acid sequence alignment of the predicted S-layer proteins with MUSCLE [39], by eliminating poorly aligned positions using GBLOCKS [40], and by generating phylogenies using the PhyML package [41]. Numbers 1–3 indicate the presence of several slp genes in the same strain. *, the corresponding gene is expressed.\n\nExpression analysis of slp genes and comparison of surface-located Slp:s in silico\nIn an attempt to reveal which of the identified slp genes encoded the S-layer protein bands seen in the surface protein profiles of the strains (Figure 7), either an amino-terminal sequencing or a peptide mapping analysis was performed for the proteins, and the results were compared with the genomic sequence data. In this study, the major S-layer protein bands of the L. amylovorus isolates GRL 1112-GRL 1118 [28] were all shown to be encoded by slpA-like genes. The surface protein profiles of the strains DSM 16698 and DSM 20531T also revealed one major protein band, approximately 45 kDa in size (Figure 7), and, based on N-terminal sequencing, this represented the protein encoded by slpA. The presence of an S-layer on the surface of L. amylovorus DSM 16698 and DSM 20531T was thus confirmed in this study. Furthermore, the two additional surface protein bands of DSM 16698, approximately 50 kDa and 40 kDa in size, were found to represent the products of slpB- and slpC-like sequences, respectively. Of the three slpC-type genes present in the DSM 16698 genome, the plasmid-borne version, slpC3, was found to be expressed. Despite the presence of the SlpC-encoding gene on a plasmid, the SlpC band was invariably present in the SDS-PAGE profile of DSM 16698. In indirect immunofluorescence assays, SlpA and SlpB of DSM 16698 were identified on the bacterial surface as predicted. In contrast, SlpC remained undetectable, suggesting that the location of SlpC is not accessible to antibodies due to shielding by other cell envelope components (data not shown). The expressed slp genes of the L. amylovorus strains are highlighted in Figure 6.\nFigure 7 SDS-PAGE analysis of L. amylovorus strains. Intact cells of L. amylovorus DSM 16698 (lane 1), DSM 20531T (lane 2), GRL 1112 (lane 3), GRL 1114 (lane 4), GRL 1115 (lane 5), GRL 1116 (lane 6), GRL 1117 (lane 7) and GRL 1118 (lane 8) from 50 μl of overnight cultures (A600nm = 6.4) were boiled in standard Laemmli sample buffer (extracting surface proteins) and the supernatants were analyzed by standard SDS-PAGE in a 12% gel. Arrowheads indicate SlpA (44 kDa), SlpB (50 kDa) and SlpC (40 kDa) of L. amylovorus DSM 16698 (lane 1) and SlpA (61 kDa) and B (49 kDa) of GRL 1117 (lane 7). The designations and calculated molecular weights of the S-layer proteins present on the bacterial surface are summarized in Table 1, and the deduced amino acid sequences of these proteins are found in Additional file 1. The analysis of the Slp amino acid sequences revealed the typical features of Lactobacillus S-layer proteins, including a high predicted pI value (9.1-9.6) and a very low proportion of sulfur-containing amino acids [8]. A pairwise comparison of the amino acid sequence similarities of these proteins is shown in Table 2. An amino acid sequence alignment of these Slp:s and the major, surface-located S-layer proteins of L. acidophilus NCFM (SlpA, GenBank AAV42070) and L. crispatus JCM 5810 (CbsA, GenBank AF001313) is shown in Additional file 2. All the L. amylovorus S-layer proteins, with the exception of SlpC, display significant overall similarity to the L. acidophilus NCFM and L. crispatus Slp:s, with the signal peptides and the carboxy-terminal thirds of the sequences being particularly well conserved.\nTable 2 Amino acid sequence similarities between L. amylovorus S-layer proteins present on the bacterial surface 1Pairwise scores were calculated for each pair of sequences by calculating the number of identities in the best CLUSTALW alignment, and by dividing by the number of residues compared (gap positions were excluded).\n\nThe role of S-layer proteins in adherence to IPEC-1 cells\nThe poor water-solubility of Lactobacillus S-layer proteins, resulting from the inherent self-assembly property of bacterial S-layers in vitro, sets limitations on what methods can be used to assess the adherence of S-layer proteins to a particular target. In order to avoid potential unspecific effects associated with protein precipitation in adhesion experiments, a protein presentation system, based on purified L. amylovorus cell wall fragments as S-layer protein carriers, was developed and used to study the role of the surface-located L. amylovorus Slp:s in adhering to IPEC-1 cells (see Figure 8B for an electron micrograph of purified CWF). This method is based on the inherent tendency of S-layer proteins to recrystallize in a native manner on CWF [42,43], which have been purified in such a way to remove all of the non-covalently attached components (including the endogenous S-layer proteins), but preserving the covalently attached polymeric components like teichoic acids and polysaccharides, thus ensuring the proper self-assembly of the recombinant Slp:s. However, purified cell wall fragments are of low density and have poor contrast, necessitating specific staining if one wishes to detect the CWF on epithelial cells. For Slp-coated CWF, an indirect immunofluorescence staining procedure with Slp-specific antibodies was used, but as we failed to obtain functional antibodies against purified cell wall fragments (data not shown), the detection of uncoated control CWF was based on their prior biotinylation and staining with labeled streptavidin after the adherence assay.\nFigure 8 Adherence of S-layer protein-coated cell wall fragments to IPEC-1 cells. A) The IPEC-1 cell adherence of CWF, coated or uncoated by the indicated S-layer proteins, expressed by quantitative means. The mean number of adherent CWF was quantitated from 20 randomly selected fields of 3.5 x 104 μm2 and the results are presented as means and standard deviations from one representative experiment out of three; letters above the bars refer to Figures C-H below. B) An electron micrograph of purified CWF of L. amylovorus DSM 16698. Scale bar, 0.5 μm. C-H) Examples of the adherence of Slp-coated and uncoated L. amylovorus CWF to IPEC-1 cells as detected by fluorescence. The figures show the adherence of uncoated L. amylovorus DSM 16698 CWF (C) and the adherence of the following Slp/CWF complexes: DSM 16698 CWF/SlpA (D), DSM 16698 CWF/SlpB (E), DSM 16698 CWF/SlpC (F), DSM 20531T CWF/SlpA (G) and GRL 1117 CWF/SlpA (H). The rightmost figures display the corresponding fields viewed with phase contrast optics. The inset in (D) shows a magnified image of a cell wall fragment. Arrowheads in (G) indicate precipitated Slp. Scale bars, 10 μm. Figure 8A shows the adherence of CWF, coated or uncoated by L. amylovorus cell surface-located Slp:s, to IPEC-1 cells. In Figures 8C-H, micrographs illustrating the results of the binding assay in (A) are shown. The adherence of all uncoated CWF was negligible, as exemplified by the adherence of the CWF of the strain DSM 16698 in Figure 8C. The major Slp:s of the L. amylovorus strains DSM 16698 (Figure 8D), GRL 1112 and GRL 1115 adhered poorly to IPEC-1 cells, although the intact cells of these strains were adhesive (Figure 2). The minor S-layer like protein SlpB of DSM 16698 exhibited some adhesiveness (Figure 8E), when compared to SlpA (Figure 8D) or SlpC (Figure 8 F) of the same strain. Surprisingly, the S-layer protein SlpA of the weakly adhering strain GRL 1117 (Figure 8H) and, to a lesser extent, the Slp:s of some of the other weakly adhering strains, e.g. DSM 20531T (Figure 8G) and GRL 1118 also displayed affinity for IPEC-1 cells.As detailed in Methods, special care was taken to minimize the formation of S-layer protein precipitates during the coating procedure of CWF. However, the presence of small Slp aggregates, as indicated by the small, dot-like, immunoreactive material among the coated cell walls, could not be completely avoided (see Figure 8G as an example). However, the quantification of the result by microscopic counting made it possible to ignore this undesirable signal, probably originating from unspecific and/or irrelevant binding.\n"}

{"project":"MicrobeTaxon","denotations":[{"id":"T140","span":{"begin":3199,"end":3203},"obj":"562"},{"id":"T141","span":{"begin":3409,"end":3413},"obj":"562"},{"id":"T142","span":{"begin":3528,"end":3532},"obj":"562"},{"id":"T145","span":{"begin":2568,"end":2575},"obj":"562"},{"id":"T147","span":{"begin":3234,"end":3251},"obj":"1423723"},{"id":"T148","span":{"begin":2463,"end":2467},"obj":"562"},{"id":"T149","span":{"begin":2875,"end":2879},"obj":"562"},{"id":"T150","span":{"begin":2966,"end":2970},"obj":"562"},{"id":"T151","span":{"begin":3032,"end":3049},"obj":"695563"},{"id":"T123","span":{"begin":22,"end":35},"obj":"1604"},{"id":"T124","span":{"begin":74,"end":87},"obj":"1604"},{"id":"T125","span":{"begin":232,"end":245},"obj":"1604"},{"id":"T126","span":{"begin":122,"end":129},"obj":"9823"},{"id":"T127","span":{"begin":345,"end":352},"obj":"9823"},{"id":"T128","span":{"begin":486,"end":493},"obj":"9823"},{"id":"T129","span":{"begin":208,"end":211},"obj":"9823"},{"id":"T130","span":{"begin":545,"end":554},"obj":"2"},{"id":"T131","span":{"begin":1155,"end":1168},"obj":"1604"},{"id":"T132","span":{"begin":1279,"end":1292},"obj":"1604"},{"id":"T133","span":{"begin":1304,"end":1311},"obj":"9823"},{"id":"T134","span":{"begin":1769,"end":1776},"obj":"9823"},{"id":"T135","span":{"begin":1433,"end":1456},"obj":"1604"},{"id":"T136","span":{"begin":1700,"end":1713},"obj":"1604"},{"id":"T137","span":{"begin":1715,"end":1725},"obj":"1423723"},{"id":"T138","span":{"begin":1433,"end":1456},"obj":"695563"},{"id":"T139","span":{"begin":2497,"end":2510},"obj":"1604"},{"id":"T143","span":{"begin":2617,"end":2630},"obj":"1604"},{"id":"T144","span":{"begin":3005,"end":3018},"obj":"1604"},{"id":"T146","span":{"begin":3032,"end":3049},"obj":"1604"},{"id":"T152","span":{"begin":4067,"end":4080},"obj":"1604"},{"id":"T153","span":{"begin":4131,"end":4144},"obj":"1604"},{"id":"T154","span":{"begin":4364,"end":4382},"obj":"1423723"},{"id":"T155","span":{"begin":4432,"end":4438},"obj":"562"},{"id":"T156","span":{"begin":4564,"end":4571},"obj":"562"},{"id":"T157","span":{"begin":4483,"end":4505},"obj":"90371"},{"id":"T158","span":{"begin":5707,"end":5720},"obj":"1604"},{"id":"T159","span":{"begin":6825,"end":6842},"obj":"1423723"},{"id":"T160","span":{"begin":6090,"end":6103},"obj":"1604"},{"id":"T161","span":{"begin":6413,"end":6426},"obj":"1604"},{"id":"T162","span":{"begin":5784,"end":5789},"obj":"9606"},{"id":"T163","span":{"begin":6142,"end":6147},"obj":"9606"},{"id":"T164","span":{"begin":6629,"end":6634},"obj":"9606"},{"id":"T165","span":{"begin":5742,"end":5768},"obj":"272621"},{"id":"T166","span":{"begin":6194,"end":6202},"obj":"2"},{"id":"T167","span":{"begin":6367,"end":6386},"obj":"272621"},{"id":"T168","span":{"begin":7584,"end":7597},"obj":"1604"},{"id":"T169","span":{"begin":7651,"end":7664},"obj":"1604"},{"id":"T170","span":{"begin":8032,"end":8045},"obj":"1604"},{"id":"T171","span":{"begin":7958,"end":7977},"obj":"272621"},{"id":"T172","span":{"begin":8549,"end":8558},"obj":"1604"},{"id":"T173","span":{"begin":8737,"end":8746},"obj":"1604"},{"id":"T174","span":{"begin":8890,"end":8899},"obj":"1604"},{"id":"T175","span":{"begin":8568,"end":8578},"obj":"1423723"},{"id":"T176","span":{"begin":8692,"end":8710},"obj":"1423723"},{"id":"T177","span":{"begin":8549,"end":8558},"obj":"695563"},{"id":"T178","span":{"begin":8737,"end":8746},"obj":"695563"},{"id":"T179","span":{"begin":8890,"end":8899},"obj":"695563"},{"id":"T180","span":{"begin":10246,"end":10259},"obj":"1604"},{"id":"T181","span":{"begin":10928,"end":10937},"obj":"1604"},{"id":"T182","span":{"begin":11132,"end":11141},"obj":"1604"},{"id":"T183","span":{"begin":11199,"end":11208},"obj":"1604"},{"id":"T185","span":{"begin":12209,"end":12218},"obj":"2"},{"id":"T186","span":{"begin":12419,"end":12432},"obj":"1578"},{"id":"T187","span":{"begin":12761,"end":12780},"obj":"272621"},{"id":"T188","span":{"begin":13002,"end":13021},"obj":"272621"},{"id":"T189","span":{"begin":12810,"end":12822},"obj":"47770"},{"id":"T190","span":{"begin":13026,"end":13038},"obj":"47770"},{"id":"T191","span":{"begin":11474,"end":11487},"obj":"1604"},{"id":"T192","span":{"begin":12896,"end":12909},"obj":"1604"},{"id":"T193","span":{"begin":10377,"end":10394},"obj":"1604"},{"id":"T194","span":{"begin":10399,"end":10409},"obj":"1423723"},{"id":"T195","span":{"begin":10645,"end":10655},"obj":"1423723"},{"id":"T196","span":{"begin":10617,"end":10640},"obj":"1604"},{"id":"T197","span":{"begin":10747,"end":10756},"obj":"1604"},{"id":"T198","span":{"begin":10377,"end":10394},"obj":"695563"},{"id":"T199","span":{"begin":10617,"end":10640},"obj":"695563"},{"id":"T200","span":{"begin":10747,"end":10756},"obj":"695563"},{"id":"T201","span":{"begin":10928,"end":10937},"obj":"695563"},{"id":"T202","span":{"begin":11132,"end":11141},"obj":"695563"},{"id":"T203","span":{"begin":11199,"end":11208},"obj":"695563"},{"id":"T204","span":{"begin":13578,"end":13591},"obj":"1578"},{"id":"T205","span":{"begin":17173,"end":17183},"obj":"1423723"},{"id":"T206","span":{"begin":13664,"end":13673},"obj":"２"},{"id":"T207","span":{"begin":13965,"end":13978},"obj":"1604"},{"id":"T208","span":{"begin":14092,"end":14105},"obj":"1604"},{"id":"T209","span":{"begin":16358,"end":16371},"obj":"1604"},{"id":"T210","span":{"begin":16610,"end":16626},"obj":"1604"},{"id":"T211","span":{"begin":16664,"end":16695},"obj":"1604"},{"id":"T212","span":{"begin":16871,"end":16880},"obj":"1604"},{"id":"T213","span":{"begin":16610,"end":16626},"obj":"695563"},{"id":"T214","span":{"begin":16664,"end":16695},"obj":"695563"},{"id":"T215","span":{"begin":16871,"end":16880},"obj":"695563"},{"id":"T297","span":{"begin":1914,"end":1927},"obj":"1604"},{"id":"T298","span":{"begin":1984,"end":1997},"obj":"1604"},{"id":"T299","span":{"begin":2209,"end":2232},"obj":"1604"},{"id":"T300","span":{"begin":2209,"end":2232},"obj":"695563"},{"id":"T301","span":{"begin":3650,"end":3663},"obj":"1604"},{"id":"T302","span":{"begin":3745,"end":3758},"obj":"1604"},{"id":"T303","span":{"begin":3616,"end":3620},"obj":"562"},{"id":"T304","span":{"begin":3697,"end":3701},"obj":"562"},{"id":"T305","span":{"begin":3856,"end":3860},"obj":"562"},{"id":"T306","span":{"begin":5074,"end":5086},"obj":"1604"},{"id":"T307","span":{"begin":5123,"end":5128},"obj":"9823"},{"id":"T308","span":{"begin":5236,"end":5249},"obj":"1604"},{"id":"T314","span":{"begin":9342,"end":9355},"obj":"1604"},{"id":"T315","span":{"begin":9410,"end":9423},"obj":"1604"},{"id":"T316","span":{"begin":11556,"end":11569},"obj":"1604"},{"id":"T317","span":{"begin":11595,"end":11618},"obj":"1604"},{"id":"T318","span":{"begin":12030,"end":12053},"obj":"1604"},{"id":"T319","span":{"begin":11629,"end":11639},"obj":"1423723"},{"id":"T320","span":{"begin":11595,"end":11618},"obj":"695563"},{"id":"T321","span":{"begin":12030,"end":12053},"obj":"695563"},{"id":"T322","span":{"begin":15659,"end":15682},"obj":"1423723"},{"id":"T323","span":{"begin":15867,"end":15890},"obj":"1423723"},{"id":"T324","span":{"begin":15761,"end":15774},"obj":"1604"},{"id":"T325","span":{"begin":15953,"end":15962},"obj":"1423723"},{"id":"T326","span":{"begin":15977,"end":15986},"obj":"1423723"},{"id":"T327","span":{"begin":16001,"end":16010},"obj":"1423723"},{"id":"T328","span":{"begin":16025,"end":16035},"obj":"1604"},{"id":"T329","span":{"begin":16025,"end":16035},"obj":"695563"},{"id":"T184","span":{"begin":11232,"end":11241},"obj":"2"},{"id":"T292","span":{"begin":768,"end":781},"obj":"1604"},{"id":"T293","span":{"begin":817,"end":830},"obj":"1604"},{"id":"T294","span":{"begin":854,"end":861},"obj":"9823"},{"id":"T295","span":{"begin":924,"end":931},"obj":"9823"},{"id":"T296","span":{"begin":970,"end":979},"obj":"2"},{"id":"T309","span":{"begin":7114,"end":7119},"obj":"9606"},{"id":"T310","span":{"begin":7139,"end":7152},"obj":"1604"},{"id":"T311","span":{"begin":7341,"end":7354},"obj":"1604"},{"id":"T312","span":{"begin":7259,"end":7264},"obj":"9606"},{"id":"T313","span":{"begin":7383,"end":7391},"obj":"2"},{"id":"T331","span":{"begin":12209,"end":12218},"obj":"2"}],"namespaces":[{"prefix":"_base","uri":"http://purl.bioontology.org/ontology/NCBITAXON/"}],"text":"Results\n\nAdherence of L. amylovorus strains to mucus\nThe adherence of the L. amylovorus strains to commercially available porcine gastric mucins and to mucus isolated from the small intestine of a 8-week old pig were examined using L. amylovorus cells labeled by the DNA-binding stain SYTO®9 (Figure 1). None of the strains bound extensively to porcine gastric mucins, i.e., typically less than 1% of the original amount of cells remained mucin-bound (Figure 1A). The same was true for porcine intestinal mucus, where the proportion of adhering bacterial cells was usually 2% or less (Figure 1B). The very high variation between the experiments and the lack of any consistent dose–response of binding (data not shown) supported the conclusions.\nFigure 1 Adherence of L. amylovorus strains to mucus. The adherence of L. amylovorus strains to immobilized porcine gastric mucin (type II, Sigma) (A) and to mucus purified from porcine small intestine (B) was studied using bacterial cells labeled with the fluorescent dye SYTO®9. The means and standard deviations of 5–8 independent experiments are shown, each with three technical replicates.\n\nAdherence of L. amylovorus strains to IPEC-1 cells\nIn contrast to mucus binding, clear differences were observed in the adherence of the L. amylovorus strains to porcine small intestinal epithelial cells as represented by the cell line IPEC-1 (Figure 2A). The previously reported binding of L. amylovorus DSM 16698 to IPEC-1 cells [26] was confirmed, and the adherence of the strains GRL 1112 and GRL 1115 was found to be within a similar range, though the three strains displayed high variabilities in different experiments. In contrast, the type strain of L. amylovorus, DSM 20531T, isolated from silage, and the rest of the porcine intestinal isolates were clearly less adhesive. An example of the dose–response of binding is shown in Figure 2B.\nFigure 2 Adherence of L. amylovorus strains to IPEC-1 cells. A) The adherence of 3H-labeled L. amylovorus strains (A600nm = 0.25) to IPEC-1 cells grown on Thincert™ wells showing the means and standard deviations of 3–7 independent experiments, each with three technical replicates. B) An example of the adherence of L. amylovorus DSM 16698, GRL 1112, GRL 1115 and GRL 1118 to IPEC-1 cells as the function of cell density. The means of three technical replicates from one representative experiment are shown. Dpm, disintegrations per minute.\n\nInhibition of F4-fimbriated ETEC adherence to IPEC-1 cells by L. amylovorus\nIPEC-1 cells have been shown to support the adherence of E. coli carrying F4-type fimbriae [38]. Next the L. amylovorus strains were tested in three different experimental set-ups (exclusion, displacement and competition as described in Methods) to evaluate if the observed differences in their adherence to IPEC-1 cells correlated with their abilities to inhibit ETEC adherence in the same model. The results were evaluated by comparing the adherence of ETEC in the presence or absence of the L. amylovorus strains. The strains DSM 16698, GRL 1112, GRL 1115 and GRL 1118 were able to inhibit pathogen adherence if they were added beforehand (exclusion, Figure 3A) or simultaneously with ETEC (competition, Figure 3B); the strain DSM 20531T achieved only a borderline inhibition when added beforehand (Figure 3A), and the rest of the strains had a negligible or even a slightly enhancing effect on ETEC binding in both assays (Figures 3A and B). Importantly, none of the strains was able to displace previously bound ETEC from IPEC-1 cells (displacement, Figure 3C).\nFigure 3 Inhibition of F4-fimbriated ETEC adherence to IPEC-1 cells by L. amylovorus. The inhibition of F4-fimbriated ETEC adherence to IPEC-1 cells by the indicated L. amylovorus strains in exclusion (A), competition (B) and displacement assays (C) was tested with 3H-labeled ETEC cells as detailed in Methods. The means and standard deviations of 3–7 independent experiments are shown, each with three technical replicates.\n\nInhibition of pathogen growth by the culture supernatants of L. amylovorus\nThe filter-sterilized culture supernatants of the L. amylovorus strains were assayed for their abilities to inhibit the growth of various intestinal pathogens (Figure 4). All the supernatants markedly inhibited the growth of the test pathogens. For instance, the supernatants of the strains DSM 20531T and GRL 1117 reduced the growth of F4-fimbriated E.coli by more than 100 000-fold and the growth of Salmonella typhimurium almost by a factor of 10 000. The growth of F4-fimbriated E. coli was most efficiently inhibited. The pH values of the supernatants varied from 3.8 to 4.5. It is notable that the reductions in pathogen counts inversely correlated with the pH values of the supernatants (Figure 4), and culture supernatants which had been adjusted to the pH of plain MRS lowered the pathogen counts by much less than tenfold (data not shown), indicating that the inhibition was mainly due to the low pH associated with lactic acid production.\nFigure 4 Reductions in pathogen counts by L. amylovous culture supernatants. Six different swine intestinal pathogens were grown in TSB medium in the presence of the filter-sterilized supernatants of the L. amylovorus strains, and the reductions in pathogen counts, expressed as log CFU values, were estimated from the area reduction percentages (ARPs) of the pathogen growth curves by linear regression. The average pH values of the supernatants are shown above the histograms. The results are the means and standard deviations of three independent experiments, each performed with fresh culture supernatants with three technical replicates.\n\nCytokine induction in moDCs by L. amylovorus\nThe S-layer-carrying L. acidophilus strain NCFM interacts with human DCs eliciting an anti-inflammatory IL-10 response and it promotes the Th2-differentiation of T-cells through DC:s; the S-layer protein has been shown to have a role in this response [12]. Prompted by these findings, we examined the potential of the phylogenetically closely related, S-layer-carrying L. amylovorus strains to induce immune signaling in human DCs. As shown in Figure 5, when tested at the bacteria/DC ratio of 100:1, clear differences between the levels of cytokines induced by the strains were observed. Interestingly, the anti-inflammatory response induced by L. acidophilus NCFM was not observed with the L. amylovorus strains. Instead, our strains typically induced a mixed cytokine response with the release of both proinflammatory (TNF-α, IL-6, IL-1β, IL-12, IP-10/CXCL10) and anti-inflammatory (IL-10) cytokines from human DCs. Furthermore, the strain GRL 1116, which was most potent at inducing proinflammatory cytokines, induced also the highest levels of the anti-inflammatory cytokine IL-10. Analogously, the strain DSM 20531T and GRL 1115 were among the weakest inducers of both pro-and anti-inflammatory cytokines. At the lower MOI values of 1 and 10, no clear induction of any of the cytokines was observed in comparison to the negative control (data not shown).\nFigure 5 Cytokine induction in human dendritic cells by L. amylovorus. The extents of induction of TNF-α (A), IL-1β (B), IL-6 (C), IL-10 (D), IL-12 (E) and IP-10/CXCL10 (F) in human monocytic dendritic cells (moDCs) were tested after treating the cells with L. amylovorus strains for 24 hours at the bacteria/DC ratio 100:1. The data are presented as the means and standard deviations from one representative experiment out of three, performed with moDC:s of four donors.\n\nGenomic characterization of L. amylovorus Slp:s\nTo initiate comparative studies on the role of L. amylovorus surface layer proteins in the probiotic interactions described above, the numbers and sequence similarities of the slp genes in the genomes of the strains were initially analysed. The genomic investigation of the eight strains revealed several slp genes in each strain. Genes with homology to L. acidophilus NCFM slpA and slpB[12] were identified, and the homologous L. amylovorus genes were named slpA and slpB, respectively. Furthermore, slp-like genes of a third type were detected in all of the eight genomes and these were designated as slpC. The slp sequences, along with the deduced amino acid sequences, are shown in Additional file 1. All the eight strains studied carried only one slpA-homologue, except for GRL 1117, which had two distinct slpA-like genes (slpA1 and slpA2). Only one slpB-homologue was identified in GRL 1112, 1114, 1115, 1116, 1117 and 1118 as well as in DSM 16698, whereas DSM 20531T carried two slpB-like genes (slpB1 and slpB2). The highest variation was found in the number of slpC-type genes: strains DSM 20531T and GRL 1115 carried one, DSM 16698 possessed three (slpC1, slpC2, slpC3), and the rest of the strains had two slpC-type genes (slpC1 and slpC2). Exceptionally, the gene slpC3 of DSM 16698 was found to be located on a plasmid. A phylogenetic tree was constructed based on the deduced amino acid sequences of the slpA, slpB and slpC gene products (Figure 6). The tree clearly shows that the SlpA-like sequences have diversified most during evolution, while the SlpB-type proteins have remained more similar to each other whereas the predicted SlpC-type proteins could be clustered into three distinct groups.\nFigure 6 Phylogeny of L. amylovorus Slp:s. A neighbour-joining phylogenetic tree based on L. amylovorus Slp sequences was generated by creating a multiple amino acid sequence alignment of the predicted S-layer proteins with MUSCLE [39], by eliminating poorly aligned positions using GBLOCKS [40], and by generating phylogenies using the PhyML package [41]. Numbers 1–3 indicate the presence of several slp genes in the same strain. *, the corresponding gene is expressed.\n\nExpression analysis of slp genes and comparison of surface-located Slp:s in silico\nIn an attempt to reveal which of the identified slp genes encoded the S-layer protein bands seen in the surface protein profiles of the strains (Figure 7), either an amino-terminal sequencing or a peptide mapping analysis was performed for the proteins, and the results were compared with the genomic sequence data. In this study, the major S-layer protein bands of the L. amylovorus isolates GRL 1112-GRL 1118 [28] were all shown to be encoded by slpA-like genes. The surface protein profiles of the strains DSM 16698 and DSM 20531T also revealed one major protein band, approximately 45 kDa in size (Figure 7), and, based on N-terminal sequencing, this represented the protein encoded by slpA. The presence of an S-layer on the surface of L. amylovorus DSM 16698 and DSM 20531T was thus confirmed in this study. Furthermore, the two additional surface protein bands of DSM 16698, approximately 50 kDa and 40 kDa in size, were found to represent the products of slpB- and slpC-like sequences, respectively. Of the three slpC-type genes present in the DSM 16698 genome, the plasmid-borne version, slpC3, was found to be expressed. Despite the presence of the SlpC-encoding gene on a plasmid, the SlpC band was invariably present in the SDS-PAGE profile of DSM 16698. In indirect immunofluorescence assays, SlpA and SlpB of DSM 16698 were identified on the bacterial surface as predicted. In contrast, SlpC remained undetectable, suggesting that the location of SlpC is not accessible to antibodies due to shielding by other cell envelope components (data not shown). The expressed slp genes of the L. amylovorus strains are highlighted in Figure 6.\nFigure 7 SDS-PAGE analysis of L. amylovorus strains. Intact cells of L. amylovorus DSM 16698 (lane 1), DSM 20531T (lane 2), GRL 1112 (lane 3), GRL 1114 (lane 4), GRL 1115 (lane 5), GRL 1116 (lane 6), GRL 1117 (lane 7) and GRL 1118 (lane 8) from 50 μl of overnight cultures (A600nm = 6.4) were boiled in standard Laemmli sample buffer (extracting surface proteins) and the supernatants were analyzed by standard SDS-PAGE in a 12% gel. Arrowheads indicate SlpA (44 kDa), SlpB (50 kDa) and SlpC (40 kDa) of L. amylovorus DSM 16698 (lane 1) and SlpA (61 kDa) and B (49 kDa) of GRL 1117 (lane 7). The designations and calculated molecular weights of the S-layer proteins present on the bacterial surface are summarized in Table 1, and the deduced amino acid sequences of these proteins are found in Additional file 1. The analysis of the Slp amino acid sequences revealed the typical features of Lactobacillus S-layer proteins, including a high predicted pI value (9.1-9.6) and a very low proportion of sulfur-containing amino acids [8]. A pairwise comparison of the amino acid sequence similarities of these proteins is shown in Table 2. An amino acid sequence alignment of these Slp:s and the major, surface-located S-layer proteins of L. acidophilus NCFM (SlpA, GenBank AAV42070) and L. crispatus JCM 5810 (CbsA, GenBank AF001313) is shown in Additional file 2. All the L. amylovorus S-layer proteins, with the exception of SlpC, display significant overall similarity to the L. acidophilus NCFM and L. crispatus Slp:s, with the signal peptides and the carboxy-terminal thirds of the sequences being particularly well conserved.\nTable 2 Amino acid sequence similarities between L. amylovorus S-layer proteins present on the bacterial surface 1Pairwise scores were calculated for each pair of sequences by calculating the number of identities in the best CLUSTALW alignment, and by dividing by the number of residues compared (gap positions were excluded).\n\nThe role of S-layer proteins in adherence to IPEC-1 cells\nThe poor water-solubility of Lactobacillus S-layer proteins, resulting from the inherent self-assembly property of bacterial S-layers in vitro, sets limitations on what methods can be used to assess the adherence of S-layer proteins to a particular target. In order to avoid potential unspecific effects associated with protein precipitation in adhesion experiments, a protein presentation system, based on purified L. amylovorus cell wall fragments as S-layer protein carriers, was developed and used to study the role of the surface-located L. amylovorus Slp:s in adhering to IPEC-1 cells (see Figure 8B for an electron micrograph of purified CWF). This method is based on the inherent tendency of S-layer proteins to recrystallize in a native manner on CWF [42,43], which have been purified in such a way to remove all of the non-covalently attached components (including the endogenous S-layer proteins), but preserving the covalently attached polymeric components like teichoic acids and polysaccharides, thus ensuring the proper self-assembly of the recombinant Slp:s. However, purified cell wall fragments are of low density and have poor contrast, necessitating specific staining if one wishes to detect the CWF on epithelial cells. For Slp-coated CWF, an indirect immunofluorescence staining procedure with Slp-specific antibodies was used, but as we failed to obtain functional antibodies against purified cell wall fragments (data not shown), the detection of uncoated control CWF was based on their prior biotinylation and staining with labeled streptavidin after the adherence assay.\nFigure 8 Adherence of S-layer protein-coated cell wall fragments to IPEC-1 cells. A) The IPEC-1 cell adherence of CWF, coated or uncoated by the indicated S-layer proteins, expressed by quantitative means. The mean number of adherent CWF was quantitated from 20 randomly selected fields of 3.5 x 104 μm2 and the results are presented as means and standard deviations from one representative experiment out of three; letters above the bars refer to Figures C-H below. B) An electron micrograph of purified CWF of L. amylovorus DSM 16698. Scale bar, 0.5 μm. C-H) Examples of the adherence of Slp-coated and uncoated L. amylovorus CWF to IPEC-1 cells as detected by fluorescence. The figures show the adherence of uncoated L. amylovorus DSM 16698 CWF (C) and the adherence of the following Slp/CWF complexes: DSM 16698 CWF/SlpA (D), DSM 16698 CWF/SlpB (E), DSM 16698 CWF/SlpC (F), DSM 20531T CWF/SlpA (G) and GRL 1117 CWF/SlpA (H). The rightmost figures display the corresponding fields viewed with phase contrast optics. The inset in (D) shows a magnified image of a cell wall fragment. Arrowheads in (G) indicate precipitated Slp. Scale bars, 10 μm. Figure 8A shows the adherence of CWF, coated or uncoated by L. amylovorus cell surface-located Slp:s, to IPEC-1 cells. In Figures 8C-H, micrographs illustrating the results of the binding assay in (A) are shown. The adherence of all uncoated CWF was negligible, as exemplified by the adherence of the CWF of the strain DSM 16698 in Figure 8C. The major Slp:s of the L. amylovorus strains DSM 16698 (Figure 8D), GRL 1112 and GRL 1115 adhered poorly to IPEC-1 cells, although the intact cells of these strains were adhesive (Figure 2). The minor S-layer like protein SlpB of DSM 16698 exhibited some adhesiveness (Figure 8E), when compared to SlpA (Figure 8D) or SlpC (Figure 8 F) of the same strain. Surprisingly, the S-layer protein SlpA of the weakly adhering strain GRL 1117 (Figure 8H) and, to a lesser extent, the Slp:s of some of the other weakly adhering strains, e.g. DSM 20531T (Figure 8G) and GRL 1118 also displayed affinity for IPEC-1 cells.As detailed in Methods, special care was taken to minimize the formation of S-layer protein precipitates during the coating procedure of CWF. However, the presence of small Slp aggregates, as indicated by the small, dot-like, immunoreactive material among the coated cell walls, could not be completely avoided (see Figure 8G as an example). However, the quantification of the result by microscopic counting made it possible to ignore this undesirable signal, probably originating from unspecific and/or irrelevant binding.\n"}