A Bacteroides fragilis surface glycoprotein mediates the interaction between the bacterium and the extracellular matrix component laminin-1

A Bacteroides fragilis surface glycoprotein mediates the interaction between the bacterium and the extracellular matrix component laminin-1
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Research in Microbiology 157 (2006) 960– A  Bacteroides fragilis  surface glycoprotein mediates the interaction betweenthe bacterium and the extracellular matrix component laminin-1 Eliane de O. Ferreira a , ∗ , Leandro Araújo Lobo a , Débora Barreiros Petrópolis c ,Kátia Eliane dos S. Avelar b , Maria Candida Ferreira a , Fernando Costa e Silva Filho c ,Regina Maria C.P. Domingues a a  Departamento de Microbiologia Médica, UFRJ, Instituto de Microbiologia Prof. Paulo de Góes, Ilha do Fundão, CEP.: 21941-590, Rio de Janeiro, Brazil b  Departamento de Bacteriologia, FIOCRUZ, Instituto Oswaldo Cruz, Av. Brasil, 4365, CEP.: 21040-361, Rio de Janeiro, Brazil c Programa de Bioengenharia & Biotecnologia Animal, UFRJ, Instituto de Biofísica Carlos Chagas Filho, CCS-Bloco G, Ilha do Fundão, CEP.: 21949-900, Rio de Janeiro, Brazil Received 29 November 2005; accepted 14 September 2006Available online 26 October 2006 Abstract Theadherenceof   Bacteroidesfragilis strainstoimmobilizedlaminin-1(LMN-1)wasinvestigatedusingthisproteinadsorbedontoglass.Amongthe 27 strains isolated from infectious processes and assayed, 13 presented strong adherence to LMN-1. Among them, two strains, MC2 and 1081,showed the strongest association, and for that reason they were selected for further studies in which adherence to this protein was confronted withboth physical-chemical and enzymatic treatments, along with concurrence assays with the LMN-1 molecule itself and the LMN-1-residing aminoacid sequences (RGD, IKVAV, YIGSR, AG73, A13 and C16). The chemical and enzymatic treatments resulted in sharp decreases in bindingrates of those strains, and competition experiments with LMN-1- residing amino acids revealed that, except for RGD and A13, all the otherswere effective at reducing bacterial binding of the bacteria. The outer membrane proteins (OMPs) of   B. fragilis  were extracted and assayed ontodot-blotted LMN-1, and when the extracts were chemically treated, especially with metasodium periodate, a drastic reduction in bacterial bindingoccurred. Results of the latter assays clearly indicate that bacterial molecules involved in both recognition and binding of   B. fragilis  to LMN-1are present in OMP extracts. Taken together, our results strongly indicate that a  B. fragilis  surface glycoprotein may play a key role in bacterialassociation with LMN-1. © 2006 Elsevier Masson SAS. All rights reserved. Keywords: Bacteroides fragilis ; Extracellular matrix components; Laminin-1; Glycoprotein 1. Introduction  Bacteroides fragilis  is the Gram-negative strictly anaerobicbacterium most frequently isolated from infectious processes,especially from patients suffering from intraabdominal infec-tions [8] and bacteremia [44]. Its virulence determinants havebeenthesubjectofmanyinvestigations[24]focusingonsurfacestructures, extracellular enzymes and toxins [26]. The capsularpolysaccharide complex (CPC) is the major virulence factor in  B. fragilis  and it is responsible for abscess formation in animalmodels [8,14]. * Corresponding author.  E-mail address: (E. de O. Ferreira). Although  B. fragilis  is a member of the intestinal micro-biota, under certain circumstances it might invade the epithe-lium, reaching and crossing the barrier formed by the com-ponents of the basal lamina. The extracellular matrix (ECM)is a stable macromolecule structure, underlying epithelial andendothelial cells and surrounding connective tissue cells, thatbecomes exposed when tissue integrity is disturbed [45]. It iscomposed of a complex assortment of glycoproteins and pro-teoglycans that can serve as signals for the attachment of sev-eral microorganisms and may also be partly responsible forthe tissue tropism of many infectious microorganisms [15,35].Microbial pathogens may use adhesins to bind to ECM com-ponents such as fibronectin, collagens, laminins (LMNs) andvitronectin [27,44]. LMN is an ECM glycoprotein found in na- 0923-2508/$ – see front matter  © 2006 Elsevier Masson SAS. All rights reserved.doi:10.1016/j.resmic.2006.09.005   E. de O. Ferreira et al. / Research in Microbiology 157 (2006) 960–966   961Table 1  B. fragilis  strains selected for this studyStrains Source43859 b , 23745, 25285 ATCC a 1037,1039,1081,1384-b, 1417,HC02-1, M1, A1, 1241, 1058-4Bacteremia058109, MC3, 1386-4, 1034 Abscess1048-a, 1304-2, 1033, B3, 048205 Intraabdominal infectionMC2 Soft tissue infection1032 Pleural fluid1077 Spinal liquid1097 Pos-operatory secretion wound079298 b Diarrhea a American Type Culture Collection. b  B. fragilis  strain positive for enterotoxin production (enterotoxigenic  B. fragilis  or ETBF). ture in more than15 isoforms. Isoform 1 of LMN or LMN-1 is a900 kDa cross-shaped molecule [6] mainly associated with po-larized epithelial cells and found just below the basal domain of the plasma membrane of intestinal epithelial cells [42]. At thebasal lamina, the glycoprotein is found in close association withcollagen IV, both of which form a molecular ground for epithe-lial cells [34]. The short arms of LMN ( β  and  γ  chains) andthe long one ( α   chain), including some peptide sequences, arein turn differentially recognized by prokaryotic [5,16,30] andeukaryotic pathogens [4,38,39]. Some peptide sequences fromthe short and long arms of LMN can be recognized by bacteria[22,39]. The bacterial surface adhesins that mediate microbialrecognition of the ECM are termed “microbial surface compo-nents recognizing adhesive matrix molecules” (MSCRAMMs)[27]. The ability of the bacteria to bind to ECM components isa very important phenomenon, since it may be directly relatedto bacterial virulence for some species [10,29,37].It has been reported that  B. fragilis  may both recognize andbind to each of the LMNs and fibronectin, vitronectin and colla-gens [5,18,41]. However, no data is available on the role playedby such molecular recognition and binding in the infectiousprocesses caused by this species.The present study focuses on the molecular basis for the in-teraction between  B. fragilis  strains and LMN-1 found to beimmobilized on glass coverslips. Attempts were made to iden-tify the LMN-1 residing sequence recognized by  B. fragilis,  aswell as the chemical nature of the bacterial molecule involvedin this recognition process. 2. Materials and methods 2.1. Bacterial strains B. fragilis  strains assayed here (Table 1) were from the Cul-ture Collection of the Anaerobe Biology Laboratory (Institutode Microbiologia Prof. Paulo de Góes, Universidade Federal doRio de Janeiro). The bacterial strains were collected and iso-lated from patients attending at Brazilian university hospitalsduring the 1980’s and 1990’s (10 from patients presenting bac-teremia, 13 from those suffering from other infections), exceptfor 4 strains, ATCC 23745, ATCC 25285 and ATCC 43859,which were obtained from the American Type Culture Col-lection (Rockville, MD, USA) and the 079298 strain, kindlydonated by Dr. Lyle Myers (Culture Collection, Montana Uni-versity, MT, USA). 2.2. Culture conditions Several days before the experiments, all bacteria were inoc-ulated and grown in brain heart infusion (BHI) (Sigma Chem.Co) previously reduced and anaerobically sterilized (PRAS)medium [11,12]. After 24 h of cultivation at 37 ◦ C, the microor-ganisms were collected by centrifugation at 3000  g  for 10 min,washed three times with 0.05 M phosphate-buffered 0.15 MNaCl (PBS) pH 7.4 and resuspended at desired densities (10 9 ,10 8 and 10 7 CFUmL − 1 ) in the same buffer. 2.3. LMN-1 LMN from an Engelbreth–Holm–Swarm mouse tumor EHS(LMN-1; Invitrogen) was used throughout and it was immobi-lized onto glass coverslips as previously described [46]. Briefly,from 5 to 20 µgmL − 1 LMN-1 were resuspended in 0.01 Mphosphate-buffered 0.15 M NaCl (PBS) and immobilized or notonto glass coverslips (precleaned with 70% ethanol and 2% Ex-tran) for 1 h at room temperature. Immediately following this,each of the uncoated and LMN-1-coated coverslips was care-fully washed with PBS containing 0.1% (w/v) bovine serumalbumin (BSA) in order to remove unbound LMN-1, avoidingnon-specific association of   B. fragilis . 2.4. LMN-1 and   B. fragilis  interaction First, screening with all bacterial strains was performed.Suspensions containing 10 9 CFUmL − 1 of   B. fragilis  (200 µL)were allowed to interact from 30 to 60 min at room tempera-ture with uncoated or LMN-1-coated coverslips which were inturn placed in 24 wells. The interaction was carried out in 0.1M PBS pH 7.4. At the end of the experiment, samples weresequentially washed once with PBS, fixed with methanol for5 min, further washed with PBS and stained with 0.1% crystalviolet [46]. Each experiment was carried out in triplicate (3 sep-arated experiments) and the association of the bacteria (ratioof adherence) to the coverslips was established by counting 10random fields of each coverslip (1000 × magnification). The re-sulting association was scored from + 4 (more than 70 bacteriaper field = high association) to 0 (less than 10 bacteria per field =  no association). For some experiments, prior to the interac-tion with each of the LMN-1-coated and uncoated coverslips,bacteria were incubated or not for 60 min at 4 ◦ C with each(2.5 µgmL − 1 ) of the peptides listed in Table 2, which werekindly provided by Dr. Motoyoshi Nomizu (National Instituteof Health, Bethesda, MD, USA). 2.5. Physical and chemical treatments of the bacteria Prior to interacting with uncoated or LMN-1-coated cov-erslips, bacteria were submitted or not to each of the follow-ing treatments: [a] incubation with proteinase K (5 µgmL − 1 ),  962  E. de O. Ferreira et al. / Research in Microbiology 157 (2006) 960–966  Table 2Purified peptides and positions in the LMN-1 moleculeSynthetic peptides Sequence Active sequence LocalizationAG73 RKRLQVQLSIRT LQVQLSIR  α  1 chain globularDomainA13 RQVFQVAYIIIKA Not known  α  1 chainA208 AASIKVAVSADR IKVAV  α  1 chain– IKVAV IKVAV  α  1 chain– RGD RGD  α  1 chain– YIGSR YIGSR  β 1 chainC16 KAFDITYVRLKF YVRL  γ 1 chainThe synthetic peptides [19–21] correspond to different fragments of thepolypeptide chains that form the LMN-1 molecule. (Kindly provided byDr. Motoyoshi Nomizu from the National Institute of Health, Bethesda, MD,USA.) trypsin (20 µgmL − 1 ) or pronase E (5 µgmL − 1 ), all diluted in0.1 M PBS pH 7.4 for 1 h at 37 ◦ C, [b] heating at 80 ◦ C for10 min and [c] incubation with 100 mM sodium periodate madein sodium acetate buffer (pH 5.0) for 1 h at 37 ◦ C. After eachof these treatments, bacteria were washed three times with PBSand assayed throughout (paragraph 2.4). 2.6. Extraction of outer membrane proteins (OMPs)and immunoblotting assays Toobtainanenrichedfractionof   B.fragilis OMPs,amethod-ology according to Bölin, Norlander and Wolf-Watz (1982)was used with some alterations. Briefly, following cultivation,the bacteria were washed twice with 0.1 M PBS (pH 7.2),centrifuged at 4000  g  (10 min) and the resulting pellet wereresuspended in a 10 mM Tris–HCl buffer (pH 8.0), containing1 mM EDTA. The bacteria were then disrupted in a sonicator(2 min, 4 ◦ C). Any remaining intact cells were removed by cen-trifugation (4000  g , 10 min). The crude envelope fraction wascollected from the supernatant by centrifugation (100000  g ;30 min;4 ◦ C). The pelletwastreatedtwicewitha 0.3% sarcosylsolution (Sigma Chem. co.) and the outer membrane fractionwas recovered by centrifugation (10000  g ; 10 min) followed byfreeze-drying to concentrate, and was kept at − 20 ◦ C. In orderto investigate the possible involvement of OMPs in the bacterialinteraction with LMN-1, extracts of OMPs from both MC2 and1081 strains, chemically treated with proteinase K, trypsin andsodium periodate (paragraph 2.5) or not, were used for dot blot-ting analyses. OMP extracts were blotted onto a nitrocellulosemembrane. After moistening the membrane with TBST 10 mMTris, 150 mM NaCl, 0.1% Tween 20 (pH 7.4), for a few secondsand drying, the extracts (3 µL) were dropped onto the mem-brane and the nitrocellulose blots were incubated overnight inTBST blocking buffer (1% BSA and 5% skim milk) at roomtemperature in an LMN-1 solution (Invitrogen; 10 µgmL − 1 in blocking buffer). The membrane was washed again (3 × )with TBST and incubated with goat anti-LMN IgG (1:100)(Santa Cruz Biotechnology) in blocking buffer for 60 min. Af-ter washing 3 times with TBST, the membrane was incubatedat room temperature for 60 min with anti-goat secondary anti-body (1:8000 in blocking buffer; Sigma Chem. Co.) conjugatedto peroxidase. Blots were finally washed 3 times more in TBSTand developed. In the negative control, no LMN-1 was added Table 3Adherence of   B. fragilis  strains to LMN-1Ratio of adherence a Percentage of   B. fragilis  strains capable of adhering to LMN-1– 3.70% (1 strain: HC02-1)1 +  22.22% (6 strains: 1039, 1384-b, 1417, ATCC25285, ATCC 43859 and ATCC 23745)2 +  25.93% (7 strains: MC3, 1386-4, 1241,1058-4, A1, M1 and 1037)3 +  40.74% (11 strains: 1032, 1077, 1097,1048-a, 1304-2, 1033, B3, 048205, 079298,05109 and 1034)4 +  7.41% (2 strains: 1081 and MC2) a (–): no association (less than 10 bacteria per field); (1 + ): low middle asso-ciation (10–30 bacteria per field); (2 + ): middle association (30–50 bacteria perfield); (3 + ): high middle association (50–70 bacteria per field; (4 + ): high as-sociation (more than 70 bacteria per field). Negative control: each strain wasincubated with BSA (2%). (All tests were performed in triplicate.) and as positive control, LMN-1 (1 mgmL − 1 ) was used. Testswere made in duplicate. 2.7. Statistical analysis Foralladherenceassays,Student’s t   testwasusedandvalues P <  0 . 05 were considered significant. 3. Results 3.1. Association of   B. fragilis  with immobilized LMN-1 Data shown in Table 3 reveal that 26 out of 27  B. fragilis strains studied here were able to associate with immobilizedLMN-1. As can be clearly seen in data presented in the ta-ble, the numberof bacterial strains associated with immobilizedLMN-1 varied greatly, from no association (0) to a high one(4+) in the following order: 3.70% (1 strain; no association),22.22% (6 strains; low middle association), 25.93% (7 strains;middle association), 40.74% (11 strains; high middle associa-tion) and 7.41% (2 strains; high association). The two strainsthat presented the highest levels of association, 1081 and MC2,were further investigated. 3.2. Binding of the 1081 and MC2microorganisms to immobilized LMN-1 The 1081 and the MC2 microorganisms were each allowedto interact with coverslips coated or not with different amountsof LMN-1 (5, 10, 15 or 20 µgmL − 1 ). This set of experimentsresulted in two dose-response curves (Fig. 1). Coating of thecoverslips with LMN-1 amounts higher than 15 µgmL − 1 didnot induce an increase in the binding of   B. fragilis  to the glyco-protein. Furthermore, we tested the binding of each strain at dif-ferent concentrations (10 7 to 10 9 CFUmL − 1 ) with 15 µgmL − 1 immobilized LMN-1 (Fig. 2). As expected, the number of 1081or MC2 microorganisms bound to LMN-coated coverslips wasdirectly related to the bacterial densities. A slight increase to10 7 CFUmL − 1 was observed in the association of MC2 withimmobilized LMN-1.   E. de O. Ferreira et al. / Research in Microbiology 157 (2006) 960–966   963Fig. 1. Adherence of   B. fragilis  strains 1081 and MC2 (10 9 CFUmL − 1 ) todifferent concentrations of LMN. Student’s  t   test was used and all experimentswere performed in triplicate; BSA was used as control.Fig. 2. Interaction of different concentrations of the MC2 and 1081 strains (10 7 to 10 9 CFUmL − 1 ) to LMN-1 (15 µgmL − 1 ). For negative control the strainswere incubated with BSA (2 mgmL − 1 ). 3.3. Chemical nature of LMN binding groups present at the bacterial surface Attempts were made to identify the chemical nature of thebacterial surface molecule involved in the binding of   B. frag-ilis  strains to LMN-1. As shown in Fig. 3, previous treatmentsof 1081 and MC2 whole cells with proteinase K, trypsin andpronase E, heat and sodium periodate resulted in sharp de-creases in the binding rates to the bacteria srcinally found.This bacterial binding was strongly inhibited when MC2 and1081 microorganisms were pretreated with proteinase K andheating. Those treatments resulted in a remarkable decrease inbacterial binding to LMN-1, about 90% and 95%, respectively,to the 1081 strain. On the other hand, adhesion to the MC2strain was completely inhibited. Although treatments focusingon protein digestion were much more effective at reducing oravoiding bacterial binding to LMN-1, treatment of strains withsodium periodate decreased by about 40% their ability to bindto LMN-1. 3.4. LMN-1-related peptides competed with binding of   B. fragilisExperiments were designed to access the so-called LMN-1-residing adhesion sequences for MC2 microorganisms, which Fig. 3. Inhibition adhesion tests of the strains 1081 and MC2 with chemi-cal, physical and enzymatic treatments. (a) Positive control (10 9 CFUmL − 1 + 15 µLmL − 1 LMN-1); (b) proteinase K; (c) trypsin; (d) pronase E; (e) heating;(f) sodium periodate; (g) negative control (BSA 2%). The Student’s  t   test wasused  (P <  0 . 05 )  and all experiments were performed in triplicate. were revealed to be more susceptible. The bacteria (10 9 CFUmL − 1 ) were previously incubated in the presence of eachof peptides A208, AG 73, RGD, IKVAV, YIGSZ, A13 and C16.Ourresultsdemonstratethat,exceptforRGDandA13peptides,all other assayed peptides effectively reduced the binding of theMC2 strain to immobilized LMN-1 by 40% (Figs. 4b, 4c, 4gand 4h). Interestingly, previous treatment of the bacteria withthe A208 peptide resulted in a bacterial binding decrease of about 50% (Fig. 4e). 3.5. Immunoblotting analysis In an attempt to identify bacterial structures involved inadhesion to LMN-1, immunoblotting analyses were madewith enriched OMPs extracts. Analysis with extracts of 1081(Fig. 5a) and MC2 (Fig. 5b) strains confirmed the involve-ment of components present in these extracts in adhesion toLMN-1. When 1081 and MC2 strains were chemically treated(Section 3.3) with proteinase K and trypsin and sodium perio-date, a high or a mild decrease in bacterial binding to LMN-1was shown, respectively. Nevertheless, when enriched extractsof OMPs were submitted to the same treatments, we found thatsodium periodate strongly reduced bacterial binding to LMN-1. 4. Discussion The clinical importance of   B. fragilis  in human anaerobicinfections is well known. Evidence has been focusing on theexistence of particular properties in the pathogenicity of thespecies. The adhesion ability of   B. fragilis  has already beensuggested to be one of the most important virulence factors[24,32]. Among surface structures described in  B. fragilis  andinvolved in adhesion to human tissue, CPC and fimbriae mightplay important roles [31]. Little is known about the possible in-volvement of other surface molecules, especially the OMPs, inthe pathogenicity of   B. fragilis , but it has been proposed thatthey might play an important role in such crucial events as ad-hesion, invasion and evasion of host defenses and antimicrobialagents [25,47].Adhesion of pathogens to host tissue is a critical early stepin the process of infection, and their ability to adhere to dif-  964  E. de O. Ferreira et al. / Research in Microbiology 157 (2006) 960–966  Fig. 4. Inhibition tests of strain MC2 with synthetic peptides of LMN-1.The bacterial suspension was previously incubated for 1 h with: (b) IK-VAV; (c) YIGSR; (d) RGD; (e) A208; (f) A13; (g) AG78; (h) C16; (i) LMN(10 µgmL − 1 , final concentration). All treatments with peptides showed P <  0 . 05, except for RGD and A13 where  P >  0 . 05. Controls: (a) positive con-trol  ( 10 9 CFUmL − 1 + 15 µgmL − 1 ) ; (j) negative control (10 9 CFUmL − 1 + BSA 2%).Fig.5.Dotblotting withOMPextractsofstrains1081andMC2,nottreated(C),incubated with LMN and probed with anti-LMN antibody. Strain OPM extractswereincubatedwithmeta-sodium periodate(P),proteinase K(P–K) andtrypsin(T). Each sample was tested in duplicate; LMN (1 mgmL − ) was used as pos-itive control, and primary antibody and second antibody without laminin asnegative control. ferent surfaces depends on the expression of surface adhesins.Some of these are responsible for colonization of host surfaces,translocation through the endothelial tissue and invasion of ad- jacent tissues eventually reaching the bloodstream with dissem-ination of the infection [36]. Bacteria adhesins can recognizeand bind to some ECM proteins. One of the host ECM com-ponents known to support bacterial adherence, as previouslymentioned, is LMN, which forms a family of heterotrimericmolecules ubiquitous in basement membranes [1]. LMN hasdifferent cell binding residing sites located on the short arms aswell as on the long one of the protein cross [34]. Other surfacestructures of the LMN sequences are involved in the fixation of this protein to basement membranes by connecting it to otherECMs, such as collagen IV and heparan sulfate sulfated pro-teoglycans. The major function is related to the control of cellbehavior by interacting with integrins [1], which are, in turntransmembrane receptors involved in the bidirectional transferof signals betweenthe extra- and intracellularenvironments[3].The MSCRAMMS, microbial surface adhesins that can spe-cific recognize molecules in the ECM, have been described inmany important pathogenic bacteria, including  Staphylococcusaureus  [28],  Mycobacterium  spp. [16],  Escherichia coli  [7] and  Enterococcusfaecalis [43].Inprotozoa,thisinteractionhasalsobeen studied, for instance, in  Trichomonas vaginalis  [4],  Enta-moeba histolytica  [17] and  Trichomonas foetus , which involvean exchange of signals from the ECMs to parasites and fromparasites to ECMs [40].Few studies have been carried out to test the ability of members of the  Bacteroides  genus to interact with ECM com-ponents. Eiring and co-workers (1995) studied the capacityof   Bacteroides  species to adhere to laminin. In that study,a total of 55 strains were examined and  B. fragilis  strainsshowed a pronounced binding activity to LMN when comparedwith the other two clinical predominant species of the group,  B. vulgatus  and  B. thetaiotaomicron . In 1996, Szöke and co-workers examined the binding of different ECM proteins (vit-ronectin, collagen-IV and fibronectin) to a collection of anaer-obic bacteria srcinating from various infectious processes andfrom healthy individual stools. The authors demonstrated thatnot only do the strains contain components that recognize fi-bronectin and collagen-I, but also that different culture mediaand growth conditions influence the expression of cell surfaceproteins in anaerobic species.Our results demonstrated the capacity of some  B. fragilis strainsto adhereto LMN-1. After assaying27 strains, it was ob-served that 26 were capable of adhering to LMN-1. MC2 and1081 strains exhibited the strongest adhesion capacity to thisECM component, and for that reason they were chosen for fur-ther studies, in an attempt to understand and to characterize theadhesive molecule.Previous studies demonstrated that, to achieve a superfi-cial concentration of 2 pmol of LMN in adhesion assays,20 µgmL − 1 would be necessary [45]. In our experiments, theMC2 and 1081 strains showed a dose-dependent adhesion toLMN-1 and the highest adherence capacity was at a concentra-tion of 15 µgmL − 1 . Although higher concentrations of LMN-1had been tested, the differences detected were not significant( P >  0 . 1). For that reason, it seemed reasonable for us to per-formtheexperimentswith15µgmL − 1 ofLMN-1,whichwouldachieve1.5pmolof thismoleculeasasuperficialconcentration.Even at lower concentrations of 5 µgmL − 1 , those two strainsstill showed strong adhesion when compared to the negativecontrol (BSA). The bacterial densities were also tested, rangingfrom 10 7 to 10 9 CFUmL − 1 . As expected, adherence to immo-bilized LMN-1 was also dose-dependent in those assays and10 9 CFUmL − 1 was chosen as the ideal inoculum.Treatment of strains 1081 and MC2 with heating and pro-teinase K strongly affected adhesion to LMN-1. Although,trypsin and pronase E did not show the same effect, inhibitionwas still strong. Sodium periodate, a chemical agent that oxi-dizes sugars, was also used in treatment of whole bacteria, andthis approach also resulted in a reduction in bacterial adher-ence, but in a slight way. Nevertheless, treatment of enriched
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks