10
The measurement of biofilm formation was performed according to the method described (57). Overnight grown bacteria were diluted (1:100) in LB medium, and 150 l diluted bacteria were inoculated into each well of a 96-well microtiter dish (Orange Scientific, Belgium, cat #5530100 or TPP Scientific, America, cat #92096) and the plate incubated at 37°C for 24 hr or 48 hr to allow the biofilm formation. Each well was then washed with water and 150 l of 1% crystal violet was added and the incubation at room temperature continued for 30 min. After washing with water, 150
l of 1% SDS was subsequently added to each well and the microtiter dish was
shaken to dissolve the dye. The capability of biofilm formation was quantified by determining the absorbance at 595 nm (ELx800, BIO-TEK). The biofilm formation activity result represented the mean of three separate experiments.
9. Constructions of the recombinant His6-tagged proteins
The DNA fragments which respectively contains the major pilin of type 1 fimbriae, MrkI-HTH domain and MrkJ-EAL domain were PCR amplified from the genomic DNA of K. pneumoniae CG43S3 with primers wc27 /wc12 and wc21/wc08 (Table 3). The amplified PCR products were cloned into the cloning vector yT&A (Yeastern Biotech, Taiwan), and then subcloned using proper restriction enzymes specific enzyme and then ligated into pET30 expression vector. The recombinant plasmid was then transformed into E. coli NovaBlue(DE3) or E. coli BL21(DE3).
10. Overexpression and purification of insoluble the His6-tagged FimA
The bacterial cells were grown in 100 ml of LB medium at 37°C with shaking until OD600 reached 0.6. Isopropyl-1-thio-β-D-galactopyranoside (IPTG) was then added to a final concentration of 0.5 mM and the growth was continued for 4 hrs at 37°C. Subsequently, the cells were harvested by centrifugation at 8000 rpm for 10
11
min, resuspended in lysis buffer (50mM Tris-HCl [pH8.0], 1mM EDTA and 100 mM NaCl), and the cell suspension disrupted by sonication and then the cell debris removed by centrifugation at 13000 rpm for 10 min. The recombinant FimA was insoluble. Finally, the His6-tagged proteins were purified from the pellet via affinity chromatography using His-Bind resin (Novagen), and the elution was carried out with elution buffer (20 mM Tris-HCl, 0.5 M NaCl, 250 mM imidazole, 6N urea, [pH 7.9]).
Aliquots of the collected fractions were analyzed by SDS-PAGE and the fractions containing most of the purified His6-tagged protein were dialyzed against the 1 mL of 1X PBS buffer (8 g NaCl, 0.2 g KCl, 1.44 g Na2HPO4, 0.24 g KH2PO4, [pH 7.4]) containing 6N urea. The 5.26mg/ ml anti-rabbit FimA was generated by Kelowna International Scientific Inc.
11. Western blot analysis of the expression of type 1 and type 3 fimbriae
Total cellular lysates from the bacteria grown overnight in LB medium were resolved by 12% SDS-PAGE to determine the expression of type 1 and type 3 fimbriae in K. pneumoniae CG43. The proteins were then electrophoretically transferred onto polyvinylidene difluoride (PVDF) membrane (Millipore, Billerica, MA, USA). After incubation with 5% skim milk at room temperature for 1 hr, the membrane was washed 3 times with 1X PBS. Subsequently, the membrane was incubated at room temperature for 2 hrs with diluted anti-FimA or anti-MrkA serum.
After 3 washes with 1X PBS, a 5000-fold diluted alkaline phosphatase-conjugated anti-rabbit immunoglobulin G was added and the incubation continued for 1 hr. The blot was again washed and the bound antibodies were detected using the chromogenic reagents BCIP (5-bromo-4-chloro-3-indolyl phosphate), NBT (Nitro blue tetrazolium) and alkaline phosphatase buffer (10 mM, 5 mM and 100 mM Tris-HCl pH 9.5).
12
12. Construction of the LacZ reporter gene fusion
The putative promoters of fimA, fimB, fimE and mrkA were PCR amplified using the specific primers (Table 3) and the PCR products subcloned in front of the promoterless lacZ gene on placZ15 (48). The bacteria carrying each of the reporter plasmids were grown sharking overnight in LB medium, and the β-galactosidase activities were measured essentially as described (55). The data is representative of at last three independent experiments. Every sample was assayed in triplicate, and the average activity and standard deviation were presented.
13. β-galactosidase activity assay
β-galactosidase was assayed according to the method of Miller (54). The bacteria in the early or late logarithmic growth phase (optical density at 600 nm 0.5 or 0.8) were taken 100 l, and mixed with 900 l Z buffer (60 mM Na2HPO4 , 40 mM NaH2PO4, 10 mM KCl, 1 mM MgSO4, 50 mM β-mercaptoethanol), 17 l of 0.1%
SDS and 35 l chloroform and incubated for 15 min at 28°C. Subsequently, 200 l of 4 mg/ml ο-nitrophenyl-β-D-galactopyranoside (ONPG) was added and the mixture vortexed for 10 sec, then incubated at 28°C until yellow color was apparent. Finally, the reaction was stopped by adding 500 l of stop solution (1 M Na2CO3) and the absorbance of the supernatant was measured OD420. One unit of β-galactosidase is defined as the hydrolysis of 1 nmol ONPG per min per mg protein.
14. Motility assay
Essentially as described (49), 3 l overnight-grown bacteria was inoculated onto trypton swimming plate (0.3% Bacto Agar, 0.5% NaCl, and 1% tryptone) and the plate incubated at 30°C for 9 hrs. The diameter of the zone created by the swimming bacteria was measured.
13
15. Overexpression and purification of the His6-tagged MrkJ
The bacterial cells were incubated in 100 ml of LB medium at 37°C with shaking until OD600 reached 0.6. Isopropyl-1-thio-β-D-galactopyranoside (IPTG) was then added to a final concentration of 0.5 mM and the growth was continued for 4 hrs at 37°C. Subsequently, the cells were harvested by centrifugation at 8000 rpm for 10 min, resuspended in lysis buffer (50mM Tris-HCl [pH8.0], 1mM EDTA and 100mM NaCl), and the cell suspension disrupted by sonication and then the cell debris removed by centrifugation at 13000 rpm for 10 min. Finally, the His6-tagged proteins were purified from the supernatant via affinity chromatography using His-Bind resin (Novagen), and the elution was carried out with buffer A (20 mM Tris-HCl, 500 mM NaCl, 250 mM imidazole, [pH 7.9]). Aliquots of the collected fractions were analyzed by SDS-PAGE and the fractions containing most of the purified His6-tagged protein were dialyzed against the buffer containing 20 mM Tris-HCl [pH 8.5], 200 mM NaCl, and 10% glycerol.
16. Phosphodiesterase activity of MrkJ
Phosphodiesterase activity of the recombinant MrkJ was performed as previously described (39). In the assay buffer (50 mM Tris-HCl, 1 mM MnCl2 [pH 8.5]) supplemented with 5 mM bis(p-nitrophenol) phosphate (bis-pNPP), 20g of the purified MrkJ was added and the mixture incubated for 3 hrs at 37°C. Reactions were incubated for 3 hrs at 37°C, and the release of p-nitrophenol was measured at 410 nm.
14
Results
1. BLAST analysis and Pfam searches
As shown in Fig. 1A, the MrkH, MrkI, and MrkJ encoding genes are located downstream of the type 3 fimbriae operon mrkABCDF (Fig. 1A) and the intergenic region of mrkH-mrkI and mrkI-mrkJ are respectively 5 and 143 bp. The gene organization has been found to be conserved in the published K. pneumoniae genomes (97). As previously demonstrated by reverse transcription PCR (RT-PCR), the three ORFs mrkH, mrkI, and mrkJ could be transcribed in a transcriptional unit (96). These suggested a possibility of a coordinated expression of the physically linked genes mrkABCDF and mrkHIJ. The gene coding for csgD, which has been associated with bacterial virulence (31), is located within yggR, a putative ATPase, and yqgF which is an essential protein for Holliday junction resolvase (HJR) (4) (Fig. 1B).
Analysis using BLAST (Basic Local Alignment Search Tool) (38) and Pfam database (protein family database) revealed that mrkH, mrkI, mrkJ and csgD respectively encode c-di-GMP binding protein (PilZ domain protein), LuxR-type transcription regulator, c-di-GMP phosphodiesterase (EAL domain protein) and LuxR-type transcription regulator (Fig.1C). As shown in Fig. 2, the conserved RxxxR motif and D/NxSxGG motif, which play essential roles for c-di-GMP binding in many PilZ domain proteins (3, 10, 31, 52, 61, 71), is found in MrkH. The conserved DDGF(T/A)GYSS motif and glutamate residue critical for phosphodiesterase activity of many EAL domain protein, is also present in MrkJ (Fig. 3). In K. pneumoniae, several EAL domain proteins including BlrP1 (6), YjcC (45), FimK (80), MrkJ (38) have been reported. As shown in Fig.4, the sequence alignment revealed K.
pneumoniae CsgD had 36% identity shared with E. coli CsgD and Salmonella enterica.
15
2. Generation of the mrkH, mrkI, mrkJ and csgD deletion mutants
The respective gene deletion was assessed using PCR analysis with the specific primer pairs, wc05/wc06 formrkH, wc07/wc08 formrkI, wc17/wc18 for mrkJ, and pcc226/pcc227 for csgD. As shown in Fig. 5, the amplicons of 1500-bp and 850-bp were obtained for wild type strain andmrkH strain; 2500-bp and 1900-bp for wild type strain andmrkI strain; 1300-bp and 620-bp for wild type strain andmrkJ strain; 1300-bp and 680-bp for wild type strain andcsgD strain, respectivrly which confirmed the individual deletion for each of the mutants. The selection percentage for mrkH, mrkI, mrkJ and csgD were respectivrly 57%, 64%, 10% and 21%.
To determine if the gene deletion effects the bacterial growth, growth curve of the mrkH, mrkI, mrkJ and csgD in LB or M9 medium were determined. The
four mutant strains appeared to show similar growth curve as the wild type strain in LB medium (Fig. 6A) or M9 medium (Fig. 6B).
3. Analysis of the deletion effects on the activity of type 1 and type 3 fimbriae The activity of type 1 fimbriae was assessed using yeast agglutination analysis.
As shown in Fig. 7A, agglutination could be observed for mrkAmrkH, mrkI and
mrkI[pKAS46-mrkHJ] strains. The mrkI deletion effect was able to be
complemented by introducing pKAS46-mrkHIJ into mrkI strain. Addition of 2%
mannose could inhibit the agglutination activity of mrkA, mrkI or
mrkI[pKAS46-mrkHJ] indicating a mannose-sensitive agglutination activity (Fig.
7B).
Expression of K. pneumonia type 3 fimbriae has been reported to be able to promote the biofilm formation (21, 36). Compared to wild type, the biofilm formation activity of mrkA, mrkH or mrkI was apparently reduced while the activity slightly decreased formrkJ in either type of microtiter dish (Fig. 8A and C or Fig. 8B and D)
16
and in 24 hrs or 48 hrs incubation (Fig. 8A and B or Fig. 8C and D). The deletion effect of mrkI was able to be complemented by introduction pKAS46-mrkHIJ into
mrkI strain. However, deletion of csgD did not affect biofilm formation activity except a reduced biofilm formation was observed in TPP-microtiter dish for 48 hrs.
4. Analysis of the deletion effects on the expression of FimA pilin of type 1 fimbriae and MrkA pilin of type 3 fimbriae
In order to obtain a good amount of the recombinant FimA protein, the recombinant plasmid pETfimA-23 (Table 2) was used to transform E. coli Novablue (DE3) and expression of the recombinant FimA was analyzed. As shown in Fig. 9A, an IPTG-induced overexpression of the His6-FimA could be observed, however, most of the recombinant proteins were in the pellet-fraction. Urea (6N) was employed to unfold the aggregated protein. The purified His6-FimA of approximately 28 kDa (Fig.
9A) was used to immunize rabbit to raise anti-FimA antibody. The specific of anti-FimA antibody was tested by a 10000-fold diluted anti-FimA at room temperature for 1 hr. As shown in Fig. 9B, the anti-FimA antibody could specifically bind to the recombinant FimA.
Western blot analysis using the prepared FimA antiserum and the MrkA antiserum obtained from Dr. HY Chang’s lab (NTHU, College of Life Science (96)) was then performed to determine the expression of type 1 and type 3 fimbriae. As shown in Fig. 10, approximately same amount of the bacterial total proteins were applied to gel stained by coomassie brilliant blue. Compared to wild type, expression of MrkA was abolished in mrkH, mrkI, mrkA while slightly decreased in csgD.
MrkA expression could be restored in mrkI[pKAS46-mrkHIJ] strain. By contrast, a slight increase on the expression of MrkA was observed inmrkJ strain. On the other hand, the expression of FimA was increased in mrkI and mrkA strains. Interestingly,
17
expression of FimA was observed in mrkI as well as in mrkA strains further suggesting a reciprocal expression of the two fimbriae.
5. MrkI affected the fimS inversion
As shown in Fig. 11A, two primer pairs pcc248/ pcc249 and pcc247/ pcc249 were designed to respectively assess ON-phase or OFF-phase of fimS. Compare to wild type, the level of ON-phase of fimS represented by a 478-bp amplicon in mrkI and mrkI[pKAS46-mrkHJ]strain increased (Fig. 11B). The mrkI deletion effect could be complemented by introduction of pKAS46-mrkHIJ into mrkI strain, suggesting MrkI negatively regulates the fimS promoter activity. By contrast, a slight decrease of ON-phase of fimS level was found for mrkJ and csgD strains. However, no apparent change of ON-phase or OFF-phase of fimS (599-bp amplicon) level in
mrkH was observed (Fig. 11B).
6. MrkA, FimA, FimB and FimE promoter activity analysis
As shown in Fig. 12A, the putative promoters of 551-bp, 358-bp, 400-bp and 270-bp noncoding DNA respectively located upstream of mrkA, fimA, fimB and fimE were individually fused with the promoterless lacZ gene of the reporter plasmid placZ15 (Table 2). Thus, the activity of PmrkA, PfimA, PfimB and PfimE could be determined by the activity of LacZ. The promoter reporter plasmids were then individually transformed into the parental strain lacZ (CG43S3Z01), or each of the specific gene deletion strains lacZmrkH (CG43S3Z01mrkH), lacZmrkI (CG43S3Z01mrkI) and lacZcsgD (CG43S3Z01csgD). The deletion of mrkH had no apparent effect on the activity of PfimA, PfimB or PfimE (Fig. 12A left panel) which is consistent with the result of fimS analysis (Fig. 11). On the contrary, the mrkH deletion dramatically decreased PmrkA activity (Fig. 12A right panel). As shown in Fig.
18
12B, similar deletion effect on the promoter activity was observed for mrkI compared to mrkH. However, mrkJ deletion had no apparent effect on either of the promoter activity (Fig. 12C). Although no apparent change of the PfimA, PfimB or PfimE activity by the csgD deletion (Fig. 12D left panel), reduced activity of PmrkA in lacZcsgD was observed (Fig. 12D right panel).
7. The recombinant MrkJ exhibited a phosphodiesterase activity
To examine whether mrkJ encodes a functional phosphodiesterase, the MrkJ expression plasmid was transformed into E.coli MG1655 for motility analysis. As shown in Fig.13, E. coli MG1655[pRK415-MrkJ] exhibited the highest level of motile activity, while E. coli MG1655[pRK415-Ydeh] which expresses c-di-GMP cyclase activity (95), had the lowest level of motility activity. This implied that MrkJ encodes a functional phosphodiesterase activity to reduce the cellular c-di-GMP leading to increase the bacterial swimming activity.
The in vitro analysis was performed using the purified recombinant MrkJ and phosphodiesterase-specific substrate bis(pNPP). As shown in Fig. 14, an IPTG-induced overexpression of the His6-MrkJ could be observed in E. coli BL21 (DE3), however, the recombinant proteins were found mostly in pellet-fraction but some in supernatant fractions. The recombinant MrkJ was then purified from the supernatant fraction for the assay of phosphodiesterase activity. The purified His6-MrkJ of approximately 34 kDa was found to exhibit a 40-fold increase of the p-nitrophenol release compared to the reaction with BSA (Fig. 15). This further supported that MrkJ is a phosphodiesterase playing a role in modulation of the level of the secondary messenger c-di-GMP.
19
Discussion
Type 1 and type 3 fimbriae are important factors for bacterial invasion, biofilm formation, cell motility and persistence in specific cell surface. Regulation of type 1 fimbriae is well known which is mediated by the DNA recombinase FimB and FimE to control the inversion DNA sequence of fimS (34). By contrast, the regulation of type 3 fimbriae is poorly understood. Downstream to the type 3 fimbrial gene clusters, mrkH, mrkI, and mrkJ have recently been demonstrated to be transcribed in a transcription unit (96). This also implies that mrkHIJ acts as a regulatory operon for the expression of type 3 fimbriae. If mrkHIJ operon is also involved in the type 1 fimbriae expression is hence investigated.
1. MrkH is a positive regulator for type 3 fimbriae
The deletion of mrkH from K. pneumoniae CG43S3 caused an increased of mannose-resistant yeast agglutination implying that MrkH controls an unknown type of sugar-mediated binding adhesion activity. On the other hand, a reduced level of biofilm formation was found for the mrkH deletion strain which suggesting a positive regulatory role on the expression of type 3 fimbriae. A decreased expression of MrkA and PmrkA activity in the mrkH strain further supports that MrkH play a positive role on the expression of type 3 fimbriae through influencing PmrkA activity. Science MrkH is a putative c-di-GMP binding protein, how the second messenger-mediated regulation carried out remains be investigated.
2. An inverse regulatory role of MrkI on the expression of type 1 and type 3 fimbriae
The deletion of MrkI from K. pneumoniae CG43S3 caused an increase of
20
mannose-sensitive yeast agglutination, ON-phase fimS inversion, and FimA expression. This implies MrkI plays an inhibitory role for the expression of type 1 fimbriae via altering of the fimS direction. However, promoter activity measurement (Fig. 12B) revealed that none of the promoters PfimA, PfimB, or PfimE were affected by the deletion of mrkI. If MrkI indirectly affects the expression of type 1 fimbriae remains to be clarified. On the other hand, a reduced level of biofilm formation, PmrkA
activity, promoter activity and MrkA pilin expression were found for the mrkI deletion strain suggesting MrkI plays as an activator at the transcription level for the expression of type 3 fimbriae. RT-qPCR analysis of the mrkI deletion effect and an electrophoresis mobility shift assay (EMSA) of MrkI binding to PmrkA performed by Dr. Ching-Ting Lin (School of Chinese Medicine, China Medical University) indicated that the recombinant MrkI was able to bind PmrkA further supporting that
MrkI reciprocally regulates the expression of the fimbriae at the transcription level.
3. MrkJ exerted a PDE activity
MrkJ has been reported in K. pneumoniae IApc35 as a functional c-di-GMP phosphodiesterase (38). Here, overexpression of MrkJ in E. coli appeared to increase the motility further supporting that MrkJ function as a PDE to decrease the cellular c-di GMP level. In the mrkJ-deletion mutant, a slightly increased of MrkA product was found whereas no obvious effect on the expression of FimA. This implied MrkJ plays a negative role in regulating the expression of type 3 fimbriae. Nevertheless, biofilm forming activity was decreased by the deletion of mrkJ indicated that MrkJ may play an indirect role to affect type 3 fimbriae activity.
4. CsgD is also a positive regulator for the expression of type 3 fimbriae
Deletion of csgD which encoding a LuxR-type transcription regulator slightly
21
reduced the MrkA expression and PmrkA activity, suggesting CsgD is a positive regulator at the transcription level for the expression of type 3 fimbriae. However, no apparent effect of the csgD deletion on type 1 fimbriae was observed. Different regulatory role from the CsgD of E. coli or Salmonella is speculated because K.
pneumoniae is non-flagellated bacteria.
In summary, this study indicated that MrkH, MrkI and CsgD play as activators whereas MrkJ plays as an inhibitor for the expression of type 3 fimbriae. In addition, MrkI played a negative role for the expression of type 1 fimbriae. The deletion of mrkA has been shown to cause an increase of type 1 fimbriae expression, however, by an unknown mechanism (82). It is concluded that MrkI is probably the regulator determining the reciprocal expression between the two types of fimbriae.
22
References
1. Abraham, J. M., C. S. Freitag, J. R. Clements, and B. I. Eisenstein. 1985.
An invertible element of DNA controls phase variation of type 1 fimbriae of Escherichia coli. Proc Natl Acad Sci U S A 82:5724-7.
2. Allen, B. L., G. F. Gerlach, and S. Clegg. 1991. Nucleotide sequence and functions of mrk determinants necessary for expression of type 3 fimbriae in Klebsiella pneumoniae. J Bacteriol 173:916-20.
3. Amikam, D., and M. Y. Galperin. 2006. PilZ domain is part of the bacterial c-di-GMP binding protein. Bioinformatics 22:3-6.
4. Aravind, L., K. S. Makarova, and E. V. Koonin. 2000. SURVEY AND SUMMARY: holliday junction resolvases and related nucleases: identification of new families, phyletic distribution and evolutionary trajectories. Nucleic Acids Res 28:3417-32.
5. Arnqvist, A., A. Olsen, and S. Normark. 1994. Sigma S-dependent growth-phase induction of the csgBA promoter in Escherichia coli can be achieved in vivo by sigma 70 in the absence of the nucleoid-associated protein H-NS. Mol Microbiol 13:1021-32.
6. Barends, T. R., E. Hartmann, J. J. Griese, T. Beitlich, N. V. Kirienko, D.
A. Ryjenkov, J. Reinstein, R. L. Shoeman, M. Gomelsky, and I.
Schlichting. 2009. Structure and mechanism of a bacterial light-regulated cyclic nucleotide phosphodiesterase. Nature 459:1015-8.
7. Benach, J., S. S. Swaminathan, R. Tamayo, S. K. Handelman, E.
Folta-Stogniew, J. E. Ramos, F. Forouhar, H. Neely, J. Seetharaman, A.
Camilli, and J. F. Hunt. 2007. The structural basis of cyclic diguanylate signal transduction by PilZ domains. EMBO J 26:5153-66.
8. Blomfield, I. C. 2001. The regulation of pap and type 1 fimbriation in Escherichia coli. Adv Microb Physiol 45:1-49.
9. Blumer, C., A. Kleefeld, D. Lehnen, M. Heintz, U. Dobrindt, G. Nagy, K.
Michaelis, L. Emody, T. Polen, R. Rachel, V. F. Wendisch, and G. Unden.
2005. Regulation of type 1 fimbriae synthesis and biofilm formation by the
2005. Regulation of type 1 fimbriae synthesis and biofilm formation by the