Molecular Immunology 64 (2015) 228–234
Contents lists available at ScienceDirect
Molecular Immunology journal homepage: www.elsevier.com/locate/molimm
Short communication
The antigens contributing to the serological cross-reactions of Proteus antisera with Klebsiella representatives Agata Palusiak ∗ Department of General Microbiology, Institute of Microbiology, Biotechnology and Immunology, University of Łód´z, Banacha 12/16, 90-237 Łód´z, Poland
a r t i c l e
i n f o
Article history: Received 1 October 2014 Received in revised form 20 November 2014 Accepted 21 November 2014 Available online 12 December 2014 Keywords: Cross-reaction Lipopolysaccharide Protein Proteus Klebsiella
a b s t r a c t Proteus sp. and Klebsiella sp. mainly cause infections of the urinary and respiratory tracts or wounds in humans. The representatives of both genera produce virulence factors like lipopolysaccharide (LPS) or outer membrane proteins (OMPs) having much in common in the structures and/or functions. To check how far this similarity is revealed in the serological cross-reactivity, the bacterial masses of 24 tested Klebsiella sp. strains were tested in ELISA with polyclonal rabbit antisera specific to the representatives of 79 Proteus O serogroups. The strongest reacting systems were selected to Western blot, where the majority of Klebsiella masses reacted in a way characteristic for electrophoretic patterns of proteins. The strongest reactions were obtained for proteins of near 67 and 40 kDa and 12.5 kDa. Mass spectrometry analysis of the proteins samples of one Proteus sp. and one Klebsiella sp. strain showed the GroEL like protein of a sequence GI number 2980926 to be similar for both strains. In Western blot some Klebsiella sp. masses reacted similarly to the homologous Proteus LPSs. The LPS contribution in the observed reactions of the high molecular-mass LPS species was confirmed for Klebsiella oxytoca 0.062. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction Proteus sp. and Klebsiella sp. bacteria were described and named in the same year, 1885, by Trevisan to honor the German microbiologist Edwin Klebs – Klebsiella, and by Hauser with reference to the Greek deity and the swarming growth phenomenon – Proteus (Brisse et al., 2006; Penner, 1992). The genus Proteus includes three unnamed Proteus genomospecies 4, 5 and 6 and five named species: P. mirabilis, P. vulgaris, P. penneri, P. hauseri, P. myxofaciens [proposed to be named Cosenzaea myxofaciens comb. nov.] (Giammanco et al., 2011; O’Hara et al., 2000). The genus Klebsiella consists of five species: Klebsiella pnaeumoniae with three subspecies—pneumoniae, ozaenae and rhinoscleromatis, K. oxytoca, K. mobilis, K. variicola and K. granulomatis. Klebsiella and Proteus sp. are ubiquitous in nature, where they are found in surface waters, sewage, soil, manure and intestinal tracts of different
Abbreviations: d-GalNAc, 2-acetamido-2-deoxy-d-galactose; d-GlcNAc, 2acetamido-2-deoxy-d-glucose; ELISA, enzyme-linked immunosorbent assay; Gal, galactose; GalA, galacturonic acid; GI, GenInfoIdentifier; Glc, glucose; GlcN, glucosamine; l-Ara4N, 4-amino-4-l-deoxyarabinose; l-FucNAc, 2-acetamido-2,6dideoxy-l-galactose; LPS, lipopolysaccharide; OD, optical density; OMPs, outer membranes proteins; (R)-Lac, (R)-lactic acid; SDS PAGE, sodium dodecyl sulphate polyacrylamide gel electrophoresis; S-Lac, (S)-lactic acid. ∗ Tel.: +48 42 635 44 69. E-mail address: [email protected] http://dx.doi.org/10.1016/j.molimm.2014.11.016 0161-5890/© 2014 Elsevier Ltd. All rights reserved.
˙ animals (Brisse et al., 2006; Podschun and Ullmann, 1998; Rózalski et al., 2007). The representatives of both genera colonize the human intestinal tract and, as opportunistic pathogens, may cause some infections especially the hospital-acquired ones (Brisse et al., 2006; O’Hara et al., 2000). Among Proteus and Klebsiella species the most important etiological agents are P. mirabilis (70–90% of all Proteus ˙ infections) (Penner, 1992; Rózalski et al., 2007) and K. pneumoniae (75–86% of Klebsiella sp. infections) (Hansen et al., 2004). K. pneumoniae and P. mirabilis nosocomial infections are mostly implicated in infections of the urinary and respiratory tracts, burns, wounds infections, meningitis, bacteremia and affect immunocompromised patients especially in neonatal units. These infections are often caused by both species representatives, e.g. P. mirabilis and K. pneumoniae species were isolated from abdominal wound of a man with colon carcinoma (Krajden et al., 1987). The virulence of the representatives of both species is connected with many factors, some of which have much in common in their structures and/or functions e.g. lipopolysaccharide (LPS) or outer membrane proteins (OMPs). LPS is regarded to play a major role due to its great contribution to the septic shock and its huge immunogenicity. The representatives of both genera produce lipid A molecules consisting of a bisphosphorylated -1-6 linked GlcN disaccharide substituted by similar types of fatty acids in a similar arrangement (Helander et al., 1996; Kotełko, 1986). A small difference between the structures of lipid A moieties in P. mirabilis R45 and K. pneumoniae O3 mutant OM-5 LPSs concerns the position of
A. Palusiak / Molecular Immunology 64 (2015) 228–234
4-amino-4-deoxy-l-arabinose (l-Ara4N): in K. pneumoniae both 4 phosphate and glycosidically linked phosphates are substituted by l-Ara4N, whereas in P. mirabilis lipid A this residue is mainly bound to the ester-linked phosphate group (Helander et al., 1996; Vinogradov et al., 1994). The inner core region structures of LPS of both genera are also very similar and consist of a disaccharide of 3-deoxy-d-manno-oct-2-ulosonic acid, a trisaccharide of d-glycero-d-manno-heptose, where each heptose carries a different subsituent: -glucose (-Glc), ␣- or -galacturonic acids (␣-GalA or -GalA) residues. Additionally, the outer core region of LPS of both species often contains an ␣-glucosamine (␣-GlcN) residue bound to ␣-GalA of the inner part of the core region (Holst, 2007; Vinogradov et al., 2002b). As for O-polysaccharide, there are single mono- or disaccharide components common to this part of LPS in the representatives of both species (Knirel et al., 2011; Vinogradov et al., 2002a). In general, OPS of K. pneumoniae LPS is characterized by lower structural diversity (11 different structures) than P. mirabilis OPS (45 different structures) (Knirel et al., 2011; Vinogradov et al., 2002a). Other molecules which may play a role of common surface antigens in both species are OMPs, which are important for membrane integrity. K. pneumoniae strains produce two non-specific major porins, OmpK35 and OmpK36, forming water-filled; channels which regulate the diffusion of antimicrobials like ceftazidime (Ananthan and Subha, 2005; Nikaido, 2003). P. mirabilis strains produce mainly a single porin with the subunit size of approximately 35 to 37 kDa (Nikaido, 2003). In the OM of P. mirabilis three other proteins have also been found: 39 kDa (OmpA), 36 kDa ˙ (peptidoglycan-associated matrix protein) and 17 kDa (Rózalski et al., 1997). OmpA and peptidoglycan-associate lipoprotein are characterized by a high level of analogy within Enterobacteriaceae and are major OMPs released by Gram-negative bacteria during sepsis (Hsieh et al., 2013). Many common features of the representatives of both genera encouraged us to examine how far this similarity is revealed in the cross-reactivity between the unique collection of polyclonal rabbit sera against Proteus representatives of 79 O serogroups of the genus and the bacterial masses of clinical Klebsiella sp. isolates from the Łódz´ area (Poland).
2. Materials and methods 2.1. Bacterial strains and antisera 24 Klebsiella sp. strains including 12 K. pneumoniae ssp. pneumoniae (nr 0.0: 1–9, 12, 15, 19) and 12 K. oxytoca (nr 0.0: 10, 11, 13, 21, 23, 30, 42, 54, 55, 60–62) were obtained from urine of patients in the Łódz´ area. The species of each strain was determined by the use of a commercial api 20 E test (BIOMERIEUX) recommended for Enterobacteriaceae identification. The ambiguous results with a lower percentage of identification were confirmed by the shortened biochemical tests proposed by Edwards and Ewing: lysine and ornithine decarboxylation, indole, methyl red, Voges–Proskauer and malonate (Hansen et al., 2004). Bacteria were cultured in Luria Broth and stored with glycerol (1:1) at −80 ◦ C. The set of polyclonal rabbit sera against the representatives of Proteus O serogroups, Proteus strains and their LPS homologous to the tested sera were obtained from the Department of General ´ Poland. They were obtained by Microbiology, University of Łódz, intravenous immunization of rabbits with boiled (for 2 h) bacterial cells at three doses (five-day intervals): 0.25, 0.5 and 1 mL bacterial suspension (1.5 × 1010 cfu mL−1 ). Seven days after the last injection the rabbits were bled. Dried bacterial cells of Klebsiella and Proteus sp. were obtained as previously described (Palusiak et al., 2008). Proteinase K
229
(10 mg/mL) was added (1:8) to bacterial mass or LPS preparations in a loading buffer (after boiling them for 10 min) and incubated for 1 h at 60 ◦ C. 2.2. LPS extraction and purification The LPS extraction was performed from dried bacterial mass in 45% phenol for 5 min at 67 ◦ C by means of a modified method by Westphal and Jann (1965). After centrifugation (4845 × g) the water phase containing LPS was collected and subjected to the repeated extraction procedure. The water phases were combined, dialyzed for 48 h and centrifuged (4845 × g). The supernatant was concentrated to 5 mL and adjusted to 2% CH3 COONa. Crude LPS was precipitated by the addition of 2 v of 96% ethanol (24 h, 4 ◦ C), collected by centrifugation (4845 × g, 20 min, 20 ◦ C), suspended in 5 mL of distilled water, dialyzed (24 h, 4 ◦ C) and lyophilized. Klebsiella sp. LPSs were purified with proteinase K as described above (Section 2.1) and Proteus sp. LPSs—by a treatment with cold aqueous 50% CCl3 COOH. Both methods appeared to be equally efficient in removing the potential protein contaminations, so LPSs used in this work were not contaminated with proteins. 2.3. Serological assays An enzyme-linked immunosorbent assay (ELISA), sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS PAGE) and the Western blot procedure were carried out as previously described with some modifications (Palusiak et al., 2008). In ELISA 5000 ng of bacterial mass concentration was used per well, antisera were serially diluted in a phosphate buffer solution from the initial concentration 1:4000. 3–9 g of the LPS or bacterial mass preparations in a loading buffer (1:1) were added per lane of polyacrylamide 3% stacking gel. The sera were diluted: 1:100 in dot-blot-10% skimmed milk buffer (Western blot). 2.4. Gel staining The polyacrylamide gel with separated samples was fixed for 30 min in methanol:acetic acid:water (5:1:9), stained with 0.3 mM Coomassie brilliant Blue R (Serva) in 10% acetic acid for an hour and destained in 10% acetic acid. The stained samples were transferred into nitrocellulose papers (Whatman Schleicher & Schuell) for 60 min at 100 V and then subjected to Western blot technique. The unstained SDS PAGE Protein Marker 6.5–200 kDa (Serva) was used. 3. Results and discussion The bacterial masses of 24 tested Klebsiella strains were checked in ELISA with polyclonal rabbit antisera specific to the representatives of 79 Proteus O serogroups (1824 tested sets of an antigen and a serum). Serum reactivity titers >1:32 000 were obtained for 67 tested sets. Each of the tested Klebsiella strains reacted with one serum at least. The masses of three strains: K. pneumoniae 0.015, K. oxytoca 0.010 and 0.011 reacted only with one or two antisera. The biggest numbers of reactions to the titer ≥1:32 000 were observed for four tested strains: K. pneumoniae 0.03 (32 cross-reactions), K. oxytoca: 0.030, 0.062 and 0.042 (31, 24 and 19 cross-reactions, respectively). There were no reactions observed for antisera specific to the Proteus representatives of O11, O41, O43 and O48 serogroups. Among the systems which in ELISA reacted to significantly high titers, exemplary ones [Table 1] were selected to the further stage of work. The bacterial masses were separated in polyacrylamide gel, transferred into nitrocellulose papers and checked in Western blot with the respective Proteus antisera. The LPSs and bacterial masses of the Proteus strains homologous to the respective sera were used
230
A. Palusiak / Molecular Immunology 64 (2015) 228–234
Table 1 ELISA data for the homologous and cross-reacting systems of Proteus antisera and Klebsiella sp. masses selected for the further serological studiesa,b,c . Antiserum against strain P. mirabilis (O8) P. mirabilis 10703 (O40) P. mirabilis 1Bm (O78) P. myxofaciens (O60) P. vulgaris TG155 (O55) P. penneri 8 (O67) P. penneri 19 (O64a,b,c) P. penneri 31 (O19a,b) P. penneri 60 (O70) a b c
Masses from investigated Klebsiella sp. strains 0.03, 0.07, 0.030, 0.042 0.03, 0.08, 0.012, 0.030, 0.062 0.03, 0.013, 0.019, 0.030, 0.042 0.01, 0.02, 0.03*, 0.04, 0.012, 0.030*, 0.042 0.01, 0.02, 0.03, 0.06, 0.08, 0.09, 0.012, 0.062* 0.03, 0.013, 0.042, 0.062 0.062 0.03, 0.012, 0.062* 0.054, 0.062*
Reciprocal titer for LPSs homologous to Proteus antiserum
Cross-reacting Klebsiella sp. masses
128 000 256 000 512 000 512 000 128 000 512 000 512 000 512 000 128 000
32 000 32 000 32 000 32 000, 128 000* 32 000, 128 000* 32 000 64 000* 32 000, 64 000* 32 000, 128 000*
Bacterial masses reacting in Western blot in a way characteristic for non-protein antigens are bolded. The end titer for the reactions was taken as the highest dilution of antiserum yielding A405 > 0.2. Results for the strongest cross-reacting systems are marked with *.
Fig. 1. Western blot reactions of selected Klebsiella sp. masses with Proteus antisera specific to strains: (A.1 and A.2.) P. mirabilis (O8); (B.1 and B.2) P. myxofaciens (O60); (C.1 and C.2) P. mirabilis 1Bm (O78); (D.1 and D.2) P. mirabilis 10703 (O40); (E.1 and E.2) P. vulgaris TG155 (O55); (F.1 and F.2) P. penneri 8 (O67); (G.1 and G.2) P. penneri 60 (O70); (H.1 and H.2) P. penneri 19 (O64a,b,c); (I.1 and I.2) P. penneri 31 (O19a,b). Homologous masses and LPSs were used as the controls of the reactions specificity. One and two stand for bacterial masses untreated or treated with proteinase K, respectively.
A. Palusiak / Molecular Immunology 64 (2015) 228–234
231
Fig. 2. Western blot reactions of selected Klebsiella sp. masses with Proteus antisera specific to strains: (A) P. myxofaciens O60, (B) P. mirabilis O8 after SDS PAGE of the samples and staining them with Coomassie brilliant Blue R. The unstained SDS PAGE Protein Marker 6.5–200 kDa was used. The protein group tested by mass spectrometry is marked in the orange frame. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
as a control of the reaction specificity. This method revealed the unspecific character of a few strong ELISA reactions, which were not observed in Western blot or were very weak. For example, the following systems: K. oxytoca 0.054 mass with P. penneri 60 antiserum and K. pneumoniae 0.012 mass with P. penneri 31 antiserum reacted in ELISA to the serum titers 1:32 000 [Table 1], while in Western blot the reactions appeared to be very weak [Fig. 1G.1 and I.1]. The majority of Klebsiella sp. masses reacted in a way characteristic for electrophoretic patterns of proteins, noticeable in Fig. 1A.1–F.1 as single bands. To confirm the contribution of protein antigens to the observed reactions, the tested systems were checked once more in Western blot but using bacterial masses treated with proteinase K. As suspected, single patterns well visible in Fig. 1A.1–F.1 Western blot before using proteinase K, disappeared when the bacterial mass was treated with the enzyme [Fig. 1A.2–F.2]. To determine the masses of proteins groups contributing to the reactions observed in Western blot, two Proteus antisera (against P. myxofaciens O60 and P. mirabilis O8) and masses of ten Klebsiella strains were selected. For these systems the banding patterns characteristic for protein antigens were the clearest and common for both species representatives [Fig. 1A.1 and B.1]. The bacterial masses and homologous LPSs were separated in polyacrylamide gel, stained with Coomassie brilliant Blue R, transferred
into nitrocellulose papers and checked with an appropriate antiserum in the Western blot technique. The protein antigens which were not recognized by specific antibodies remained blue, whereas these reacting with immunoglobulins became brown. The reactions with Proteus sp. masses and the LPSs were seen also in the brown color. Approximate sizes of the protein groups, seen in Fig. 2 as single banding patterns, were determined with reference to the unstained SDS PAGE Protein Marker 6.5–200 kDa. In the case of both tested sera, strong reactions were obtained only for proteins of the sizes between 6.5 and 12.5 kDa. The strong cross- and homologous reactions of the P. myxofaciens antiserum were also obtained for protein groups of the sizes: nearly 67 and 40 kDa. The reactions with the former protein group were also well visible for 0.030 K. oxytoca mass and P. mirabilis O8 antiserum. Both sera reacted with the proteins group of the approximate size of 21 kDa but the reactions were stronger with the P. myxofaciens antiserum both in homologous and heterologous systems. Bigger differences in the reactions with particular Klebsiella sp. masses were observed for the P. mirabilis O8 antiserum. For example, there was no reaction observed for the K. pneumoniae 0.07 bacterial mass at the level corresponding to the proteins groups of the sizes 40–67 kDa, which distinguished this strain from the other used Klebsiella masses. To check what particular proteins may be responsible for the observed reactions, single strips corresponding to the proteins
232
A. Palusiak / Molecular Immunology 64 (2015) 228–234
Table 2 Selected proteins detected by mass spectrometry in the protein samples obtained for K. pneumoniae 0.03 and/or P. myxofaciens (O60). Protein hits GroEL proteins gi2980926 gi197286390 gi491046433 gi493583140 gi496408604 Outer membrane proteins gi197284675 gi496193676 gi440895 gi516438429 gi516224454 gi493584382 gi197286195 gi197283959
K. pneumoniae 0.03
The name of the protein identified: P. myxofaciens (O60)
Similar to GroEL protein – – – –
Similar to GroEL protein Molecular chaperone GroEL [P. mirabilis HI4320] Chaperone Hsp60 (GroEL) [Providencia rettgeri] Molecular chaperone GroEL [P. penneri] Molecular chaperone GroEL [Neisseria wadsworthii]
– – – – – – – –
Outer membrane protein A [P. mirabilis HI4320] Outer membrane protein A [Providencia sneebia] OmpA = 39 kDa outer membrane protein [P. mirabilis, peptide partial, 34 aa, segment 5 of 6] Outer membrane porin protein C [P. mirabilis] Outer membrane porin protein C [P. penneri] Outer membrane channel protein [P. mirabilis HI4320] Outer membrane protein [P. mirabilis HI4320]
group (of approximate mass 40 kDa) of K. pneumoniae 0.03 and P. myxofaciens O60 strains were cut from the polyacrylamide gel stained with Coomassie brilliant Blue R. These groups of proteins strongly reacted in Western blot with P. myxofaciens antiserum and the reactions were observed for all tested Klebsiella and homologous masses [Fig. 2A]. K. pneumoniae 0.03 strain was selected since its antigens contributed to the numerous cross-reactions with Proteus sp. sera [Table 1]. This particular band of proteins (approximate size 40-kDa, marked in the orange frames in Fig. 2A) was chosen because in the range of protein masses from 38 to 40kDa a few OMPs had been previously found in OMP samples for the representatives of both genera (Hofstra and Dankert, 1979; Witkowska et al., 2006). One of them was 38-kDa OMP, the major Enterobacteriaceae protein recognized by the human immunological system (Witkowska et al., 2009). The immunoblot analysis of healthy human sera with OMP extracts from different Enterobacteriaceae representatives showed that the predominating reactivity occurred with the band corresponding to 38-kDa OMP for all enterobacterial strains tested, including strains of K. pneumoniae and P. vulgaris (Witkowska et al., 2006). Also, in the PAGE profiles of the OM fractions of K. pneumoniae and P. vulgaris strains the most visible bands appeared at levels corresponding to products of 39-kDa (K. pneumoniae) and 38-, 41-kDa (P. vulgaris) (Hofstra and Dankert, 1979). Identification of protein contents of the samples was performed by the Laboratory of Mass Spectrometry, the Institute of Biochemistry and Biophysics, the Polish Academy of Sciences in Warsaw (Poland). What was interesting, among numerous proteins detected only one was common to both strains, similar to GroEL protein and had a sequence GI (GenInfoIdentifier) number 2980926. GroEL is an oligomeric particle containing 14 identical subunits and together with the heptameric protein GroES plays an important role in numerous processes like protein folding, protease activity, stress protection, or oligomer assembly. Proteins of that type were also implicated in immunological response (Bochkareva et al., 1992; Govezensky et al., 1991). For instance, K. pneumoniae GroELlike protein (HSP60Kp) was recognized by IgG antibodies in the sera from HLA-B27 positive patiens with ankylosing spondylitis and HLA-B27 negative ones (Cancino-Díaz et al., 2000). In a sample of P. myxofaciens proteins more that one chaperon GroEL were detected [Table 2] but only one was similar to that of K. pneumoniae. The detection in one Proteus species sample of GroEL chaperones previously found in other representatives of the genus or other genera (also for Providencia rettgeri or Niesseria sp.) [Table 2], indicates this type of proteins as a potential vaccine antigen common to many bacteria.
In spite of preliminary assumptions that OMPs may have occurred in both tested samples, they were detected only in that of P. myxofaciens [Table 2] but not in the sample of K. pneumoniae. Among these OMPs also OmpA of 39-kDa was found, which had ˙ been previously described for P. mirabilis strains (Rózalski et al., 1997). Apart from the reactions typical for protein antigens, two Klebsiella masses, K. pneumoniae 0.03 and K. oxytoca 0.062, gave in Western blot [Fig. 1D.1–I.1] the reactions (marked with frames in Fig. 1) similar to those obtained for the Proteus LPS homologous to an appropriate serum. Such reactions were observed in the systems presented in Table 1 as bolded ones. It is worth mentioning that the majority of the reactions are characterized by the highest reciprocal titers of Proteus sera compared to those obtained for other Klebsiella sp. masses cross-reacting with each serum [Table 1]. Using proteinase K in the samples preparation did not cause disappearance of the reaction (previously seen in Western blot before using the enzyme) but only made it weaker [Fig. 1D.2–I.2, the reactions are marked with frames]. Most of the antibodies present in the sera obtained by immunization with bacterial cells are specific to LPS, especially its O-polysaccharide part, which is reflected in ELISA by high titers of the tested sera reactivity with homologous purified Proteus LPSs: 1:128 000–1:512 000 [Table 1] and strong binding-patterns concerning high-molecular-mass O-polysaccharide-containing LPS species observed for these systems in Western blot [Figs. 1D.2 and
Fig. 3. Reactivity of six Proteus sp. sera with K. oxytoca 0.062 LPS in ELISA.
A. Palusiak / Molecular Immunology 64 (2015) 228–234
233
Fig. 4. Western blot reactions of K. oxytoca 0.062 LPS with Proteus antisera specific to strains: (A) P. penneri 8 (O67), (B) P. penneri 19 (O64a,b,c), (C) P. penneri 31 (O19a,b), (D) P. penneri 60 (O70), (E) P. mirabilis 10703 (O40), (F) P. vulgaris TG155 (O55). Homologous masses and LPSs were used as the controls of the reactions specificity. 1 and 2 stand for LPSs untreated or treated with proteinase K, respectively.
E.1-F.1, H.1, I.1 and 4]. All these facts indicate the LPS as a possible antigen contributing to the above-mentioned reactions. To confirm this hypothesis, the LPSs were extracted from the dried bacterial cells of K. pneumoniae 0.03 and K. oxytoca 0.062 by means of modified Westphal’s method (Westphal and Jann, 1965). These LPSs were checked in ELISA with these Proteus antisera, which reacted in Western blot with the bacterial masses of both tested strains in a way characteristic for a LPS molecule [Fig. 1D.1,2–I.1,2]. Opposite to Western blot results [Fig. 1I.1,2], the K. pneumoniae 0.03 mass did not react with the P. penneri 31 antiserum, but it should be mentioned that the reaction in Western blot was weak after treating the mass with proteinase K [Fig. 1I.2]. The other mass, K. oxytoca 0.062, reacted with all used sera to the titers ranging from 1:16 000 [sera against: P. penneri 8 (O67), 19 (O64a,b,c) and 60 (O70), P. mirabilis 10703 (O40)] to 1:32 000 [sera against: P. vulgaris TG155 (O55) and P. penneri 31 (O19a,b)] [Fig. 3]. These reactions are weaker compared to the reactions in an appropriate homologous system [Table 1], which suggests that antibodies in tested Proteus antisera recognize only some fragments of heterologous K. oxytoca 0.062 LPS. To confirm the specificity of the ELISA reactions, the K. oxytoca 0.062 LPS (treated and untreated with proteinase K) was tested with the mentioned sera in the Western blot [Fig. 4A–F]. The LPSs homologous to the appropriate serum were used as a control. In all cases the reactions concerned the high molecular species containing LPS moieties substituted with O-polysaccharides and were identical for both LPSs treated and untreated with proteinase K (LPSs were not contaminated with the proteins). In tested heterologous systems, except of P. penneri 31 antiserum [Fig. 4C], the ladder-like banding pattern was noticed, which is characteristic for O-polysaccharide repeating units. K. oxytoca LPS was recognized by specific antibodies in P. penneri 60 serum in a way almost identical to P. penneri 60 LPS [Fig. 4D]. The comparison of the ELISA results [Fig. 3] revealed that the reactions in heterologous systems differed slightly between each other. This observation and the differences in binding patterns observed in Western blot [Fig. 4] for heterologous systems (different position of bands and different strength of the reactions) suggest that a few epitopes may contribute to the reactions. The ladder-like banding patterns observed in Western blot for K. oxytoca 0.062 LPS [Fig. 4A, B, D and F] (except the reaction with P. penneri 31 antiserum) suggest that common epitopes may be found in the OPS of K. oxytoca 0.062 LPS and Proteus sp. LPSs homologous to each of the tested serum. These six Proteus LPSs present a different serotype of the O-antigen and some of them share common fragments (Knirel et al., 2011): P. penneri 8 (O67) and 31 (O19a,b) share the following trisaccharide structure ␣d-GalpNAc-(1 → 3)-␣-l-FucpNAc-(1 → 3)--d-GlcpNAc, where the -d-GlcpNAc is substituted by lateral EtnP. A similar fragment
but containing only ␣-l-FucpNAc-(1 → 3)-␣-d-GlcpNAc is present in OPS of P. penneri 60 (O70) LPS; P. penneri 19 (O64a,b,c) and P. mirabilis 10703 (O40) OPSs share a similar fragment—-dGlcpNAc3(S-Lac)-(1 → 3)-␣-d-Galp {-d-GlcpNAc4(R-Lac) is found in P. mirabilis O40 OPS}; the last OPS that probably shares a common fragment with K. oxytoca 0.062 OPS is that of P. vulgaris TG 155 (O55) LPS, but it has no polysaccharide fragment common with OPSs of the above mentioned Proteus sp. The differences in the OPS structures of tested Proteus LPSs are the next premise indicating that a few epitopes contribute to the observed reactions. It is possible that one epitope is responsible for the cross-reaction of K. oxtytoca 0.062 with more than one Proteus anisera (very similar ELISA results and banding patterns in Western blot were observed for the systems: K. oxytoca 0.062 LPS with P. penneri 8 and 60 antisera) [Figs. 3 and 4A and D]. An indication of these potential molecular bases for the observed cross-reactions is aimed to show that a few epitopes, not one, may contribute to the reactions. However, the main purpose of this work was to show the screening of several cross-reactions between Proteus sp. antisera and Klebsiella sp. masses and determine their character. A search for more proteins responsible for the observed reactions is planned in the future. Finding the LPS moiety or protein like GroEL common to the representatives of Klebsiella and Proteus genera is of high importance in the process of looking for the crossreactive vaccine antigens, especially considering the polimicrobial character of the infections caused by the Klebsiella and Proteus species. Acknowledgements This study was supported by the grant from the Ministry of Sciences and Higher Education (Poland) for the young scientists and doctoral students, no. B1311000000061.02. The Author thanks the scientists from the Laboratory of Mass Spectrometry, the Institute of Biochemistry and Biophysics, the Polish Academy of Sciences in Warsaw (Poland), where mass spectrometry analyses of proteins samples were performed. References Ananthan, S., Subha, A., 2005. Cefoxitin resistance mediated by loss of a porin in clinical strains of Klebsiella pneumoniae and Escherichia coli. Indian J. Med. Microbiol. 23, 20–23. Bochkareva, E.S., Lissin, N.M., Flynn, G.C., Rothman, J.E., Girshovich, A.S., 1992. Positive cooperativity in the functioning of molecular chaperone GroEL. J. Biol. Chem. 267, 6796–6800. Brisse, S., Grimont, F., Grimont, P.A.D., 2006. The genus Klebsiella. In: Dworkin, M., Falkow, S., Rosenberg, E., Schleifer, K.-H., Stackebrandt, E. (Eds.), The Prokaryotes. Springer, New York, NY, pp. 159–196.
234
A. Palusiak / Molecular Immunology 64 (2015) 228–234
Cancino-Díaz, M., Ayala-Narváez, H., Burgos-Vargas, R., Reyes-López, A.S., Tovar-Castillo, L., Domínguez-López, L., Arreola, J.G., Jiménez-Zamudio, L., García-Latorre, E., 2000. Recognition of B cells epitopes of the Klebsiella pneumoniae GroEL-like protein by HLA-B27 positive subjects. Microb. Pathog. 28, 211–220. Giammanco, G.M., Grimont, P.A.D., Grimont, F., Lefevre, M., Giammanco, G., Pignato, S., 2011. Phylogenetic analysis of the genera Proteus, Morganella and Providencia by comparison of rpoB gene sequences of type and clinical strains suggests the reclassification of Proteus myxofaciens in a new genus, Cosenzaea gen. nov., as Cosenzaea myxofaciens comb. nov. Int. J. Syst. Evol. Microbiol. 61, 1638–1644. Govezensky, D., Greener, T., Zamir, A., 1991. Involvement of GroEL in Klebsiella pneumoniae nif gene expression and nitrogenase assembly. Dev. Plant Soil Sci. 8, 43–48. Hansen, D.S., Aucken, H.M., Abiola, T., Podschun, R., 2004. Recommended test panel for differentiation of Klebsiella species on the basis of a trilateral interlaboratory evaluation of 18 biochemical tests. J. Clin. Microbiol. 42, 3665–3669. Helander, I.M., Kato, Y., Kilpeläinen, I., Kostiainen, R., Lindner, B., Nummila, K., Sugiyama, T., Yokochi, T., 1996. Characterization of lipopolysaccharides of polymyxin-resistant and polymyxin-sensitive Klebsiella pneumoniae 03. Eur. J. Biochem. 237, 272–278. Hofstra, H., Dankert, J., 1979. Antigenic cross-reactivity of major outer membrane proteins in Enterobacteriaceae species. J. Gen. Microbiol. 111, 293–302. Holst, O., 2007. The structures of core regions from enterobacterial lipopolysaccharides—an update. FEMS Microbiol. Lett. 271, 3–11. Hsieh, P.-F., Liu, J.-Y., Pan, Y.-J., Wu, M.-C., Lin, T.-L., Huang, Y.-T., Wang, J.-T., 2013. Klebsiella pneumoniae peptidoglycan-associated lipoprotein and murein lipoprotein contribute to serum resistance, antiphagocytosis, and proinflammatory cytokine stimulation. J. Infect. Dis. 208, 1580–1589. Knirel, Y.A., Perepelov, A.V., Kondakova, A.N., Senchenkova, S.N., Sidorczyk, Z., Rozalski, A., Kaca, W., 2011. Structure and serology of O-antigens as the basis for classification of Proteus strains. Innate Immun. 17, 70–96. Kotełko, K., 1986. Proteus mirabilis: taxonomic position, peculiarities of growth, components of the cell envelope. Curr. Top. Microbiol. Immunol. 129, 181–215. Krajden, S., Fuksa, M., Petrea, C., Crisp, L.J., Penner, J.L., 1987. Expanded clinical spectrum of infections caused by Proteus penneri. J. Clin. Microbiol. 25, 578–579. Nikaido, H., 2003. Molecular basis of bacterial outer membrane permeability revisited. Microbiol. Mol. Biol. Rev. 67, 593–656.
O’Hara, C.M., Brenner, F.W., Miller, J.M., 2000. Classification, identification, and clinical significance of Proteus, Providencia, and Morganella. Clin. Microbiol. Rev. 13, 534–546. ˛ Palusiak, A., Dzieciatkowska, M., Sidorczyk, Z., 2008. Application of two different kinds of sera against the Proteus penneri lipopolysaccharide core region in search of epitopes determining cross-reactions with antibodies. Arch. Immunol. Ther. Exp. 56, 135–140. Penner, J.L., 1992. The genera Proteus, Providencia and Morganella. In: Balows, A., Truper, H.G., Harder, W., Schleifer, K.H. (Eds.), The Prokaryotes. Springer-Verlag, Berlin, pp. 2849–2853. Podschun, R., Ullmann, U., 1998. Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and pathogenicity factors. Clin. Microbiol. Rev. 11, 589–603. ˙ ˛ Rózalski, A., Kwil, I., Torzewska, A., Baranowska, M., Staczek, P., 2007. Bakterie z rodzaju Proteus–cechy i czynniki chorobotwórczo´sci. Postepy Hig. Med. Dosw. 61, 204–219. ˙ Rózalski, A., Sidorczyk, Z., Kotełko, K., 1997. Potential virulence factors of Proteus bacilli. Microbiol. Mol. Biol. Rev. 61, 65–89. Vinogradov, E., Frirdich, E., MacLean, L.L., Perry, M.B., Petersen, B.O., Duus, J.Ø., Whitfield, C., 2002a. Structures of lipopolysaccharides from Klebsiella pneumoniae. Elucidation of the structure of the linkage region between core and polysaccharide O chain and identification of the residues at the non-reducing termini of the O chains. J. Biol. Chem. 277, 25070–25081. Vinogradov, E., Sidorczyk, Z., Knirel, Y.A., 2002b. Structure of the lipopolysaccharide core region of the bacteria of the genus Proteus. Aust. J. Chem. 55, 61–67. Vinogradov, E.V., Thomas-Oates, J.E., Brade, H., Holst, O., 1994. Structural investigation of the lipopolysaccharide from Proteus mirabilis R45 (Re-chemotype). J. Endotoxin Res. 1, 199–206. Westphal, O., Jann, K., 1965. Bacterial lipopolysaccharides. Extraction with phenol–water and further applications of the procedure. Methods Carbohydr. Chem. 5, 83–91. Witkowska, D., Barty´s, A., Gamian, A., 2009. Białka osłony komórkowej pałeczek jelitowych i ich udział w patogenno´sci oraz odporno´sci przeciwbakteryjnej. Postepy Hig. Med. Dosw. 63, 176–199. Witkowska, D., Masłowska, E., Staniszewska, M., Szostko, B., Jankowski, A., Gamian, A., 2006. Enterobacterial 38-kDa outermembrane protein is an age-dependent molecular marker of innate immunity and immunoglobulin deficiency as results from its reactivity with IgG and IgA antibody. FEMS Immunol. Med. Microbiol. 48, 205–214.
Contents lists available at ScienceDirect
Molecular Immunology journal homepage: www.elsevier.com/locate/molimm
Short communication
The antigens contributing to the serological cross-reactions of Proteus antisera with Klebsiella representatives Agata Palusiak ∗ Department of General Microbiology, Institute of Microbiology, Biotechnology and Immunology, University of Łód´z, Banacha 12/16, 90-237 Łód´z, Poland
a r t i c l e
i n f o
Article history: Received 1 October 2014 Received in revised form 20 November 2014 Accepted 21 November 2014 Available online 12 December 2014 Keywords: Cross-reaction Lipopolysaccharide Protein Proteus Klebsiella
a b s t r a c t Proteus sp. and Klebsiella sp. mainly cause infections of the urinary and respiratory tracts or wounds in humans. The representatives of both genera produce virulence factors like lipopolysaccharide (LPS) or outer membrane proteins (OMPs) having much in common in the structures and/or functions. To check how far this similarity is revealed in the serological cross-reactivity, the bacterial masses of 24 tested Klebsiella sp. strains were tested in ELISA with polyclonal rabbit antisera specific to the representatives of 79 Proteus O serogroups. The strongest reacting systems were selected to Western blot, where the majority of Klebsiella masses reacted in a way characteristic for electrophoretic patterns of proteins. The strongest reactions were obtained for proteins of near 67 and 40 kDa and 12.5 kDa. Mass spectrometry analysis of the proteins samples of one Proteus sp. and one Klebsiella sp. strain showed the GroEL like protein of a sequence GI number 2980926 to be similar for both strains. In Western blot some Klebsiella sp. masses reacted similarly to the homologous Proteus LPSs. The LPS contribution in the observed reactions of the high molecular-mass LPS species was confirmed for Klebsiella oxytoca 0.062. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction Proteus sp. and Klebsiella sp. bacteria were described and named in the same year, 1885, by Trevisan to honor the German microbiologist Edwin Klebs – Klebsiella, and by Hauser with reference to the Greek deity and the swarming growth phenomenon – Proteus (Brisse et al., 2006; Penner, 1992). The genus Proteus includes three unnamed Proteus genomospecies 4, 5 and 6 and five named species: P. mirabilis, P. vulgaris, P. penneri, P. hauseri, P. myxofaciens [proposed to be named Cosenzaea myxofaciens comb. nov.] (Giammanco et al., 2011; O’Hara et al., 2000). The genus Klebsiella consists of five species: Klebsiella pnaeumoniae with three subspecies—pneumoniae, ozaenae and rhinoscleromatis, K. oxytoca, K. mobilis, K. variicola and K. granulomatis. Klebsiella and Proteus sp. are ubiquitous in nature, where they are found in surface waters, sewage, soil, manure and intestinal tracts of different
Abbreviations: d-GalNAc, 2-acetamido-2-deoxy-d-galactose; d-GlcNAc, 2acetamido-2-deoxy-d-glucose; ELISA, enzyme-linked immunosorbent assay; Gal, galactose; GalA, galacturonic acid; GI, GenInfoIdentifier; Glc, glucose; GlcN, glucosamine; l-Ara4N, 4-amino-4-l-deoxyarabinose; l-FucNAc, 2-acetamido-2,6dideoxy-l-galactose; LPS, lipopolysaccharide; OD, optical density; OMPs, outer membranes proteins; (R)-Lac, (R)-lactic acid; SDS PAGE, sodium dodecyl sulphate polyacrylamide gel electrophoresis; S-Lac, (S)-lactic acid. ∗ Tel.: +48 42 635 44 69. E-mail address: [email protected] http://dx.doi.org/10.1016/j.molimm.2014.11.016 0161-5890/© 2014 Elsevier Ltd. All rights reserved.
˙ animals (Brisse et al., 2006; Podschun and Ullmann, 1998; Rózalski et al., 2007). The representatives of both genera colonize the human intestinal tract and, as opportunistic pathogens, may cause some infections especially the hospital-acquired ones (Brisse et al., 2006; O’Hara et al., 2000). Among Proteus and Klebsiella species the most important etiological agents are P. mirabilis (70–90% of all Proteus ˙ infections) (Penner, 1992; Rózalski et al., 2007) and K. pneumoniae (75–86% of Klebsiella sp. infections) (Hansen et al., 2004). K. pneumoniae and P. mirabilis nosocomial infections are mostly implicated in infections of the urinary and respiratory tracts, burns, wounds infections, meningitis, bacteremia and affect immunocompromised patients especially in neonatal units. These infections are often caused by both species representatives, e.g. P. mirabilis and K. pneumoniae species were isolated from abdominal wound of a man with colon carcinoma (Krajden et al., 1987). The virulence of the representatives of both species is connected with many factors, some of which have much in common in their structures and/or functions e.g. lipopolysaccharide (LPS) or outer membrane proteins (OMPs). LPS is regarded to play a major role due to its great contribution to the septic shock and its huge immunogenicity. The representatives of both genera produce lipid A molecules consisting of a bisphosphorylated -1-6 linked GlcN disaccharide substituted by similar types of fatty acids in a similar arrangement (Helander et al., 1996; Kotełko, 1986). A small difference between the structures of lipid A moieties in P. mirabilis R45 and K. pneumoniae O3 mutant OM-5 LPSs concerns the position of
A. Palusiak / Molecular Immunology 64 (2015) 228–234
4-amino-4-deoxy-l-arabinose (l-Ara4N): in K. pneumoniae both 4 phosphate and glycosidically linked phosphates are substituted by l-Ara4N, whereas in P. mirabilis lipid A this residue is mainly bound to the ester-linked phosphate group (Helander et al., 1996; Vinogradov et al., 1994). The inner core region structures of LPS of both genera are also very similar and consist of a disaccharide of 3-deoxy-d-manno-oct-2-ulosonic acid, a trisaccharide of d-glycero-d-manno-heptose, where each heptose carries a different subsituent: -glucose (-Glc), ␣- or -galacturonic acids (␣-GalA or -GalA) residues. Additionally, the outer core region of LPS of both species often contains an ␣-glucosamine (␣-GlcN) residue bound to ␣-GalA of the inner part of the core region (Holst, 2007; Vinogradov et al., 2002b). As for O-polysaccharide, there are single mono- or disaccharide components common to this part of LPS in the representatives of both species (Knirel et al., 2011; Vinogradov et al., 2002a). In general, OPS of K. pneumoniae LPS is characterized by lower structural diversity (11 different structures) than P. mirabilis OPS (45 different structures) (Knirel et al., 2011; Vinogradov et al., 2002a). Other molecules which may play a role of common surface antigens in both species are OMPs, which are important for membrane integrity. K. pneumoniae strains produce two non-specific major porins, OmpK35 and OmpK36, forming water-filled; channels which regulate the diffusion of antimicrobials like ceftazidime (Ananthan and Subha, 2005; Nikaido, 2003). P. mirabilis strains produce mainly a single porin with the subunit size of approximately 35 to 37 kDa (Nikaido, 2003). In the OM of P. mirabilis three other proteins have also been found: 39 kDa (OmpA), 36 kDa ˙ (peptidoglycan-associated matrix protein) and 17 kDa (Rózalski et al., 1997). OmpA and peptidoglycan-associate lipoprotein are characterized by a high level of analogy within Enterobacteriaceae and are major OMPs released by Gram-negative bacteria during sepsis (Hsieh et al., 2013). Many common features of the representatives of both genera encouraged us to examine how far this similarity is revealed in the cross-reactivity between the unique collection of polyclonal rabbit sera against Proteus representatives of 79 O serogroups of the genus and the bacterial masses of clinical Klebsiella sp. isolates from the Łódz´ area (Poland).
2. Materials and methods 2.1. Bacterial strains and antisera 24 Klebsiella sp. strains including 12 K. pneumoniae ssp. pneumoniae (nr 0.0: 1–9, 12, 15, 19) and 12 K. oxytoca (nr 0.0: 10, 11, 13, 21, 23, 30, 42, 54, 55, 60–62) were obtained from urine of patients in the Łódz´ area. The species of each strain was determined by the use of a commercial api 20 E test (BIOMERIEUX) recommended for Enterobacteriaceae identification. The ambiguous results with a lower percentage of identification were confirmed by the shortened biochemical tests proposed by Edwards and Ewing: lysine and ornithine decarboxylation, indole, methyl red, Voges–Proskauer and malonate (Hansen et al., 2004). Bacteria were cultured in Luria Broth and stored with glycerol (1:1) at −80 ◦ C. The set of polyclonal rabbit sera against the representatives of Proteus O serogroups, Proteus strains and their LPS homologous to the tested sera were obtained from the Department of General ´ Poland. They were obtained by Microbiology, University of Łódz, intravenous immunization of rabbits with boiled (for 2 h) bacterial cells at three doses (five-day intervals): 0.25, 0.5 and 1 mL bacterial suspension (1.5 × 1010 cfu mL−1 ). Seven days after the last injection the rabbits were bled. Dried bacterial cells of Klebsiella and Proteus sp. were obtained as previously described (Palusiak et al., 2008). Proteinase K
229
(10 mg/mL) was added (1:8) to bacterial mass or LPS preparations in a loading buffer (after boiling them for 10 min) and incubated for 1 h at 60 ◦ C. 2.2. LPS extraction and purification The LPS extraction was performed from dried bacterial mass in 45% phenol for 5 min at 67 ◦ C by means of a modified method by Westphal and Jann (1965). After centrifugation (4845 × g) the water phase containing LPS was collected and subjected to the repeated extraction procedure. The water phases were combined, dialyzed for 48 h and centrifuged (4845 × g). The supernatant was concentrated to 5 mL and adjusted to 2% CH3 COONa. Crude LPS was precipitated by the addition of 2 v of 96% ethanol (24 h, 4 ◦ C), collected by centrifugation (4845 × g, 20 min, 20 ◦ C), suspended in 5 mL of distilled water, dialyzed (24 h, 4 ◦ C) and lyophilized. Klebsiella sp. LPSs were purified with proteinase K as described above (Section 2.1) and Proteus sp. LPSs—by a treatment with cold aqueous 50% CCl3 COOH. Both methods appeared to be equally efficient in removing the potential protein contaminations, so LPSs used in this work were not contaminated with proteins. 2.3. Serological assays An enzyme-linked immunosorbent assay (ELISA), sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS PAGE) and the Western blot procedure were carried out as previously described with some modifications (Palusiak et al., 2008). In ELISA 5000 ng of bacterial mass concentration was used per well, antisera were serially diluted in a phosphate buffer solution from the initial concentration 1:4000. 3–9 g of the LPS or bacterial mass preparations in a loading buffer (1:1) were added per lane of polyacrylamide 3% stacking gel. The sera were diluted: 1:100 in dot-blot-10% skimmed milk buffer (Western blot). 2.4. Gel staining The polyacrylamide gel with separated samples was fixed for 30 min in methanol:acetic acid:water (5:1:9), stained with 0.3 mM Coomassie brilliant Blue R (Serva) in 10% acetic acid for an hour and destained in 10% acetic acid. The stained samples were transferred into nitrocellulose papers (Whatman Schleicher & Schuell) for 60 min at 100 V and then subjected to Western blot technique. The unstained SDS PAGE Protein Marker 6.5–200 kDa (Serva) was used. 3. Results and discussion The bacterial masses of 24 tested Klebsiella strains were checked in ELISA with polyclonal rabbit antisera specific to the representatives of 79 Proteus O serogroups (1824 tested sets of an antigen and a serum). Serum reactivity titers >1:32 000 were obtained for 67 tested sets. Each of the tested Klebsiella strains reacted with one serum at least. The masses of three strains: K. pneumoniae 0.015, K. oxytoca 0.010 and 0.011 reacted only with one or two antisera. The biggest numbers of reactions to the titer ≥1:32 000 were observed for four tested strains: K. pneumoniae 0.03 (32 cross-reactions), K. oxytoca: 0.030, 0.062 and 0.042 (31, 24 and 19 cross-reactions, respectively). There were no reactions observed for antisera specific to the Proteus representatives of O11, O41, O43 and O48 serogroups. Among the systems which in ELISA reacted to significantly high titers, exemplary ones [Table 1] were selected to the further stage of work. The bacterial masses were separated in polyacrylamide gel, transferred into nitrocellulose papers and checked in Western blot with the respective Proteus antisera. The LPSs and bacterial masses of the Proteus strains homologous to the respective sera were used
230
A. Palusiak / Molecular Immunology 64 (2015) 228–234
Table 1 ELISA data for the homologous and cross-reacting systems of Proteus antisera and Klebsiella sp. masses selected for the further serological studiesa,b,c . Antiserum against strain P. mirabilis (O8) P. mirabilis 10703 (O40) P. mirabilis 1Bm (O78) P. myxofaciens (O60) P. vulgaris TG155 (O55) P. penneri 8 (O67) P. penneri 19 (O64a,b,c) P. penneri 31 (O19a,b) P. penneri 60 (O70) a b c
Masses from investigated Klebsiella sp. strains 0.03, 0.07, 0.030, 0.042 0.03, 0.08, 0.012, 0.030, 0.062 0.03, 0.013, 0.019, 0.030, 0.042 0.01, 0.02, 0.03*, 0.04, 0.012, 0.030*, 0.042 0.01, 0.02, 0.03, 0.06, 0.08, 0.09, 0.012, 0.062* 0.03, 0.013, 0.042, 0.062 0.062 0.03, 0.012, 0.062* 0.054, 0.062*
Reciprocal titer for LPSs homologous to Proteus antiserum
Cross-reacting Klebsiella sp. masses
128 000 256 000 512 000 512 000 128 000 512 000 512 000 512 000 128 000
32 000 32 000 32 000 32 000, 128 000* 32 000, 128 000* 32 000 64 000* 32 000, 64 000* 32 000, 128 000*
Bacterial masses reacting in Western blot in a way characteristic for non-protein antigens are bolded. The end titer for the reactions was taken as the highest dilution of antiserum yielding A405 > 0.2. Results for the strongest cross-reacting systems are marked with *.
Fig. 1. Western blot reactions of selected Klebsiella sp. masses with Proteus antisera specific to strains: (A.1 and A.2.) P. mirabilis (O8); (B.1 and B.2) P. myxofaciens (O60); (C.1 and C.2) P. mirabilis 1Bm (O78); (D.1 and D.2) P. mirabilis 10703 (O40); (E.1 and E.2) P. vulgaris TG155 (O55); (F.1 and F.2) P. penneri 8 (O67); (G.1 and G.2) P. penneri 60 (O70); (H.1 and H.2) P. penneri 19 (O64a,b,c); (I.1 and I.2) P. penneri 31 (O19a,b). Homologous masses and LPSs were used as the controls of the reactions specificity. One and two stand for bacterial masses untreated or treated with proteinase K, respectively.
A. Palusiak / Molecular Immunology 64 (2015) 228–234
231
Fig. 2. Western blot reactions of selected Klebsiella sp. masses with Proteus antisera specific to strains: (A) P. myxofaciens O60, (B) P. mirabilis O8 after SDS PAGE of the samples and staining them with Coomassie brilliant Blue R. The unstained SDS PAGE Protein Marker 6.5–200 kDa was used. The protein group tested by mass spectrometry is marked in the orange frame. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
as a control of the reaction specificity. This method revealed the unspecific character of a few strong ELISA reactions, which were not observed in Western blot or were very weak. For example, the following systems: K. oxytoca 0.054 mass with P. penneri 60 antiserum and K. pneumoniae 0.012 mass with P. penneri 31 antiserum reacted in ELISA to the serum titers 1:32 000 [Table 1], while in Western blot the reactions appeared to be very weak [Fig. 1G.1 and I.1]. The majority of Klebsiella sp. masses reacted in a way characteristic for electrophoretic patterns of proteins, noticeable in Fig. 1A.1–F.1 as single bands. To confirm the contribution of protein antigens to the observed reactions, the tested systems were checked once more in Western blot but using bacterial masses treated with proteinase K. As suspected, single patterns well visible in Fig. 1A.1–F.1 Western blot before using proteinase K, disappeared when the bacterial mass was treated with the enzyme [Fig. 1A.2–F.2]. To determine the masses of proteins groups contributing to the reactions observed in Western blot, two Proteus antisera (against P. myxofaciens O60 and P. mirabilis O8) and masses of ten Klebsiella strains were selected. For these systems the banding patterns characteristic for protein antigens were the clearest and common for both species representatives [Fig. 1A.1 and B.1]. The bacterial masses and homologous LPSs were separated in polyacrylamide gel, stained with Coomassie brilliant Blue R, transferred
into nitrocellulose papers and checked with an appropriate antiserum in the Western blot technique. The protein antigens which were not recognized by specific antibodies remained blue, whereas these reacting with immunoglobulins became brown. The reactions with Proteus sp. masses and the LPSs were seen also in the brown color. Approximate sizes of the protein groups, seen in Fig. 2 as single banding patterns, were determined with reference to the unstained SDS PAGE Protein Marker 6.5–200 kDa. In the case of both tested sera, strong reactions were obtained only for proteins of the sizes between 6.5 and 12.5 kDa. The strong cross- and homologous reactions of the P. myxofaciens antiserum were also obtained for protein groups of the sizes: nearly 67 and 40 kDa. The reactions with the former protein group were also well visible for 0.030 K. oxytoca mass and P. mirabilis O8 antiserum. Both sera reacted with the proteins group of the approximate size of 21 kDa but the reactions were stronger with the P. myxofaciens antiserum both in homologous and heterologous systems. Bigger differences in the reactions with particular Klebsiella sp. masses were observed for the P. mirabilis O8 antiserum. For example, there was no reaction observed for the K. pneumoniae 0.07 bacterial mass at the level corresponding to the proteins groups of the sizes 40–67 kDa, which distinguished this strain from the other used Klebsiella masses. To check what particular proteins may be responsible for the observed reactions, single strips corresponding to the proteins
232
A. Palusiak / Molecular Immunology 64 (2015) 228–234
Table 2 Selected proteins detected by mass spectrometry in the protein samples obtained for K. pneumoniae 0.03 and/or P. myxofaciens (O60). Protein hits GroEL proteins gi2980926 gi197286390 gi491046433 gi493583140 gi496408604 Outer membrane proteins gi197284675 gi496193676 gi440895 gi516438429 gi516224454 gi493584382 gi197286195 gi197283959
K. pneumoniae 0.03
The name of the protein identified: P. myxofaciens (O60)
Similar to GroEL protein – – – –
Similar to GroEL protein Molecular chaperone GroEL [P. mirabilis HI4320] Chaperone Hsp60 (GroEL) [Providencia rettgeri] Molecular chaperone GroEL [P. penneri] Molecular chaperone GroEL [Neisseria wadsworthii]
– – – – – – – –
Outer membrane protein A [P. mirabilis HI4320] Outer membrane protein A [Providencia sneebia] OmpA = 39 kDa outer membrane protein [P. mirabilis, peptide partial, 34 aa, segment 5 of 6] Outer membrane porin protein C [P. mirabilis] Outer membrane porin protein C [P. penneri] Outer membrane channel protein [P. mirabilis HI4320] Outer membrane protein [P. mirabilis HI4320]
group (of approximate mass 40 kDa) of K. pneumoniae 0.03 and P. myxofaciens O60 strains were cut from the polyacrylamide gel stained with Coomassie brilliant Blue R. These groups of proteins strongly reacted in Western blot with P. myxofaciens antiserum and the reactions were observed for all tested Klebsiella and homologous masses [Fig. 2A]. K. pneumoniae 0.03 strain was selected since its antigens contributed to the numerous cross-reactions with Proteus sp. sera [Table 1]. This particular band of proteins (approximate size 40-kDa, marked in the orange frames in Fig. 2A) was chosen because in the range of protein masses from 38 to 40kDa a few OMPs had been previously found in OMP samples for the representatives of both genera (Hofstra and Dankert, 1979; Witkowska et al., 2006). One of them was 38-kDa OMP, the major Enterobacteriaceae protein recognized by the human immunological system (Witkowska et al., 2009). The immunoblot analysis of healthy human sera with OMP extracts from different Enterobacteriaceae representatives showed that the predominating reactivity occurred with the band corresponding to 38-kDa OMP for all enterobacterial strains tested, including strains of K. pneumoniae and P. vulgaris (Witkowska et al., 2006). Also, in the PAGE profiles of the OM fractions of K. pneumoniae and P. vulgaris strains the most visible bands appeared at levels corresponding to products of 39-kDa (K. pneumoniae) and 38-, 41-kDa (P. vulgaris) (Hofstra and Dankert, 1979). Identification of protein contents of the samples was performed by the Laboratory of Mass Spectrometry, the Institute of Biochemistry and Biophysics, the Polish Academy of Sciences in Warsaw (Poland). What was interesting, among numerous proteins detected only one was common to both strains, similar to GroEL protein and had a sequence GI (GenInfoIdentifier) number 2980926. GroEL is an oligomeric particle containing 14 identical subunits and together with the heptameric protein GroES plays an important role in numerous processes like protein folding, protease activity, stress protection, or oligomer assembly. Proteins of that type were also implicated in immunological response (Bochkareva et al., 1992; Govezensky et al., 1991). For instance, K. pneumoniae GroELlike protein (HSP60Kp) was recognized by IgG antibodies in the sera from HLA-B27 positive patiens with ankylosing spondylitis and HLA-B27 negative ones (Cancino-Díaz et al., 2000). In a sample of P. myxofaciens proteins more that one chaperon GroEL were detected [Table 2] but only one was similar to that of K. pneumoniae. The detection in one Proteus species sample of GroEL chaperones previously found in other representatives of the genus or other genera (also for Providencia rettgeri or Niesseria sp.) [Table 2], indicates this type of proteins as a potential vaccine antigen common to many bacteria.
In spite of preliminary assumptions that OMPs may have occurred in both tested samples, they were detected only in that of P. myxofaciens [Table 2] but not in the sample of K. pneumoniae. Among these OMPs also OmpA of 39-kDa was found, which had ˙ been previously described for P. mirabilis strains (Rózalski et al., 1997). Apart from the reactions typical for protein antigens, two Klebsiella masses, K. pneumoniae 0.03 and K. oxytoca 0.062, gave in Western blot [Fig. 1D.1–I.1] the reactions (marked with frames in Fig. 1) similar to those obtained for the Proteus LPS homologous to an appropriate serum. Such reactions were observed in the systems presented in Table 1 as bolded ones. It is worth mentioning that the majority of the reactions are characterized by the highest reciprocal titers of Proteus sera compared to those obtained for other Klebsiella sp. masses cross-reacting with each serum [Table 1]. Using proteinase K in the samples preparation did not cause disappearance of the reaction (previously seen in Western blot before using the enzyme) but only made it weaker [Fig. 1D.2–I.2, the reactions are marked with frames]. Most of the antibodies present in the sera obtained by immunization with bacterial cells are specific to LPS, especially its O-polysaccharide part, which is reflected in ELISA by high titers of the tested sera reactivity with homologous purified Proteus LPSs: 1:128 000–1:512 000 [Table 1] and strong binding-patterns concerning high-molecular-mass O-polysaccharide-containing LPS species observed for these systems in Western blot [Figs. 1D.2 and
Fig. 3. Reactivity of six Proteus sp. sera with K. oxytoca 0.062 LPS in ELISA.
A. Palusiak / Molecular Immunology 64 (2015) 228–234
233
Fig. 4. Western blot reactions of K. oxytoca 0.062 LPS with Proteus antisera specific to strains: (A) P. penneri 8 (O67), (B) P. penneri 19 (O64a,b,c), (C) P. penneri 31 (O19a,b), (D) P. penneri 60 (O70), (E) P. mirabilis 10703 (O40), (F) P. vulgaris TG155 (O55). Homologous masses and LPSs were used as the controls of the reactions specificity. 1 and 2 stand for LPSs untreated or treated with proteinase K, respectively.
E.1-F.1, H.1, I.1 and 4]. All these facts indicate the LPS as a possible antigen contributing to the above-mentioned reactions. To confirm this hypothesis, the LPSs were extracted from the dried bacterial cells of K. pneumoniae 0.03 and K. oxytoca 0.062 by means of modified Westphal’s method (Westphal and Jann, 1965). These LPSs were checked in ELISA with these Proteus antisera, which reacted in Western blot with the bacterial masses of both tested strains in a way characteristic for a LPS molecule [Fig. 1D.1,2–I.1,2]. Opposite to Western blot results [Fig. 1I.1,2], the K. pneumoniae 0.03 mass did not react with the P. penneri 31 antiserum, but it should be mentioned that the reaction in Western blot was weak after treating the mass with proteinase K [Fig. 1I.2]. The other mass, K. oxytoca 0.062, reacted with all used sera to the titers ranging from 1:16 000 [sera against: P. penneri 8 (O67), 19 (O64a,b,c) and 60 (O70), P. mirabilis 10703 (O40)] to 1:32 000 [sera against: P. vulgaris TG155 (O55) and P. penneri 31 (O19a,b)] [Fig. 3]. These reactions are weaker compared to the reactions in an appropriate homologous system [Table 1], which suggests that antibodies in tested Proteus antisera recognize only some fragments of heterologous K. oxytoca 0.062 LPS. To confirm the specificity of the ELISA reactions, the K. oxytoca 0.062 LPS (treated and untreated with proteinase K) was tested with the mentioned sera in the Western blot [Fig. 4A–F]. The LPSs homologous to the appropriate serum were used as a control. In all cases the reactions concerned the high molecular species containing LPS moieties substituted with O-polysaccharides and were identical for both LPSs treated and untreated with proteinase K (LPSs were not contaminated with the proteins). In tested heterologous systems, except of P. penneri 31 antiserum [Fig. 4C], the ladder-like banding pattern was noticed, which is characteristic for O-polysaccharide repeating units. K. oxytoca LPS was recognized by specific antibodies in P. penneri 60 serum in a way almost identical to P. penneri 60 LPS [Fig. 4D]. The comparison of the ELISA results [Fig. 3] revealed that the reactions in heterologous systems differed slightly between each other. This observation and the differences in binding patterns observed in Western blot [Fig. 4] for heterologous systems (different position of bands and different strength of the reactions) suggest that a few epitopes may contribute to the reactions. The ladder-like banding patterns observed in Western blot for K. oxytoca 0.062 LPS [Fig. 4A, B, D and F] (except the reaction with P. penneri 31 antiserum) suggest that common epitopes may be found in the OPS of K. oxytoca 0.062 LPS and Proteus sp. LPSs homologous to each of the tested serum. These six Proteus LPSs present a different serotype of the O-antigen and some of them share common fragments (Knirel et al., 2011): P. penneri 8 (O67) and 31 (O19a,b) share the following trisaccharide structure ␣d-GalpNAc-(1 → 3)-␣-l-FucpNAc-(1 → 3)--d-GlcpNAc, where the -d-GlcpNAc is substituted by lateral EtnP. A similar fragment
but containing only ␣-l-FucpNAc-(1 → 3)-␣-d-GlcpNAc is present in OPS of P. penneri 60 (O70) LPS; P. penneri 19 (O64a,b,c) and P. mirabilis 10703 (O40) OPSs share a similar fragment—-dGlcpNAc3(S-Lac)-(1 → 3)-␣-d-Galp {-d-GlcpNAc4(R-Lac) is found in P. mirabilis O40 OPS}; the last OPS that probably shares a common fragment with K. oxytoca 0.062 OPS is that of P. vulgaris TG 155 (O55) LPS, but it has no polysaccharide fragment common with OPSs of the above mentioned Proteus sp. The differences in the OPS structures of tested Proteus LPSs are the next premise indicating that a few epitopes contribute to the observed reactions. It is possible that one epitope is responsible for the cross-reaction of K. oxtytoca 0.062 with more than one Proteus anisera (very similar ELISA results and banding patterns in Western blot were observed for the systems: K. oxytoca 0.062 LPS with P. penneri 8 and 60 antisera) [Figs. 3 and 4A and D]. An indication of these potential molecular bases for the observed cross-reactions is aimed to show that a few epitopes, not one, may contribute to the reactions. However, the main purpose of this work was to show the screening of several cross-reactions between Proteus sp. antisera and Klebsiella sp. masses and determine their character. A search for more proteins responsible for the observed reactions is planned in the future. Finding the LPS moiety or protein like GroEL common to the representatives of Klebsiella and Proteus genera is of high importance in the process of looking for the crossreactive vaccine antigens, especially considering the polimicrobial character of the infections caused by the Klebsiella and Proteus species. Acknowledgements This study was supported by the grant from the Ministry of Sciences and Higher Education (Poland) for the young scientists and doctoral students, no. B1311000000061.02. The Author thanks the scientists from the Laboratory of Mass Spectrometry, the Institute of Biochemistry and Biophysics, the Polish Academy of Sciences in Warsaw (Poland), where mass spectrometry analyses of proteins samples were performed. References Ananthan, S., Subha, A., 2005. Cefoxitin resistance mediated by loss of a porin in clinical strains of Klebsiella pneumoniae and Escherichia coli. Indian J. Med. Microbiol. 23, 20–23. Bochkareva, E.S., Lissin, N.M., Flynn, G.C., Rothman, J.E., Girshovich, A.S., 1992. Positive cooperativity in the functioning of molecular chaperone GroEL. J. Biol. Chem. 267, 6796–6800. Brisse, S., Grimont, F., Grimont, P.A.D., 2006. The genus Klebsiella. In: Dworkin, M., Falkow, S., Rosenberg, E., Schleifer, K.-H., Stackebrandt, E. (Eds.), The Prokaryotes. Springer, New York, NY, pp. 159–196.
234
A. Palusiak / Molecular Immunology 64 (2015) 228–234
Cancino-Díaz, M., Ayala-Narváez, H., Burgos-Vargas, R., Reyes-López, A.S., Tovar-Castillo, L., Domínguez-López, L., Arreola, J.G., Jiménez-Zamudio, L., García-Latorre, E., 2000. Recognition of B cells epitopes of the Klebsiella pneumoniae GroEL-like protein by HLA-B27 positive subjects. Microb. Pathog. 28, 211–220. Giammanco, G.M., Grimont, P.A.D., Grimont, F., Lefevre, M., Giammanco, G., Pignato, S., 2011. Phylogenetic analysis of the genera Proteus, Morganella and Providencia by comparison of rpoB gene sequences of type and clinical strains suggests the reclassification of Proteus myxofaciens in a new genus, Cosenzaea gen. nov., as Cosenzaea myxofaciens comb. nov. Int. J. Syst. Evol. Microbiol. 61, 1638–1644. Govezensky, D., Greener, T., Zamir, A., 1991. Involvement of GroEL in Klebsiella pneumoniae nif gene expression and nitrogenase assembly. Dev. Plant Soil Sci. 8, 43–48. Hansen, D.S., Aucken, H.M., Abiola, T., Podschun, R., 2004. Recommended test panel for differentiation of Klebsiella species on the basis of a trilateral interlaboratory evaluation of 18 biochemical tests. J. Clin. Microbiol. 42, 3665–3669. Helander, I.M., Kato, Y., Kilpeläinen, I., Kostiainen, R., Lindner, B., Nummila, K., Sugiyama, T., Yokochi, T., 1996. Characterization of lipopolysaccharides of polymyxin-resistant and polymyxin-sensitive Klebsiella pneumoniae 03. Eur. J. Biochem. 237, 272–278. Hofstra, H., Dankert, J., 1979. Antigenic cross-reactivity of major outer membrane proteins in Enterobacteriaceae species. J. Gen. Microbiol. 111, 293–302. Holst, O., 2007. The structures of core regions from enterobacterial lipopolysaccharides—an update. FEMS Microbiol. Lett. 271, 3–11. Hsieh, P.-F., Liu, J.-Y., Pan, Y.-J., Wu, M.-C., Lin, T.-L., Huang, Y.-T., Wang, J.-T., 2013. Klebsiella pneumoniae peptidoglycan-associated lipoprotein and murein lipoprotein contribute to serum resistance, antiphagocytosis, and proinflammatory cytokine stimulation. J. Infect. Dis. 208, 1580–1589. Knirel, Y.A., Perepelov, A.V., Kondakova, A.N., Senchenkova, S.N., Sidorczyk, Z., Rozalski, A., Kaca, W., 2011. Structure and serology of O-antigens as the basis for classification of Proteus strains. Innate Immun. 17, 70–96. Kotełko, K., 1986. Proteus mirabilis: taxonomic position, peculiarities of growth, components of the cell envelope. Curr. Top. Microbiol. Immunol. 129, 181–215. Krajden, S., Fuksa, M., Petrea, C., Crisp, L.J., Penner, J.L., 1987. Expanded clinical spectrum of infections caused by Proteus penneri. J. Clin. Microbiol. 25, 578–579. Nikaido, H., 2003. Molecular basis of bacterial outer membrane permeability revisited. Microbiol. Mol. Biol. Rev. 67, 593–656.
O’Hara, C.M., Brenner, F.W., Miller, J.M., 2000. Classification, identification, and clinical significance of Proteus, Providencia, and Morganella. Clin. Microbiol. Rev. 13, 534–546. ˛ Palusiak, A., Dzieciatkowska, M., Sidorczyk, Z., 2008. Application of two different kinds of sera against the Proteus penneri lipopolysaccharide core region in search of epitopes determining cross-reactions with antibodies. Arch. Immunol. Ther. Exp. 56, 135–140. Penner, J.L., 1992. The genera Proteus, Providencia and Morganella. In: Balows, A., Truper, H.G., Harder, W., Schleifer, K.H. (Eds.), The Prokaryotes. Springer-Verlag, Berlin, pp. 2849–2853. Podschun, R., Ullmann, U., 1998. Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and pathogenicity factors. Clin. Microbiol. Rev. 11, 589–603. ˙ ˛ Rózalski, A., Kwil, I., Torzewska, A., Baranowska, M., Staczek, P., 2007. Bakterie z rodzaju Proteus–cechy i czynniki chorobotwórczo´sci. Postepy Hig. Med. Dosw. 61, 204–219. ˙ Rózalski, A., Sidorczyk, Z., Kotełko, K., 1997. Potential virulence factors of Proteus bacilli. Microbiol. Mol. Biol. Rev. 61, 65–89. Vinogradov, E., Frirdich, E., MacLean, L.L., Perry, M.B., Petersen, B.O., Duus, J.Ø., Whitfield, C., 2002a. Structures of lipopolysaccharides from Klebsiella pneumoniae. Elucidation of the structure of the linkage region between core and polysaccharide O chain and identification of the residues at the non-reducing termini of the O chains. J. Biol. Chem. 277, 25070–25081. Vinogradov, E., Sidorczyk, Z., Knirel, Y.A., 2002b. Structure of the lipopolysaccharide core region of the bacteria of the genus Proteus. Aust. J. Chem. 55, 61–67. Vinogradov, E.V., Thomas-Oates, J.E., Brade, H., Holst, O., 1994. Structural investigation of the lipopolysaccharide from Proteus mirabilis R45 (Re-chemotype). J. Endotoxin Res. 1, 199–206. Westphal, O., Jann, K., 1965. Bacterial lipopolysaccharides. Extraction with phenol–water and further applications of the procedure. Methods Carbohydr. Chem. 5, 83–91. Witkowska, D., Barty´s, A., Gamian, A., 2009. Białka osłony komórkowej pałeczek jelitowych i ich udział w patogenno´sci oraz odporno´sci przeciwbakteryjnej. Postepy Hig. Med. Dosw. 63, 176–199. Witkowska, D., Masłowska, E., Staniszewska, M., Szostko, B., Jankowski, A., Gamian, A., 2006. Enterobacterial 38-kDa outermembrane protein is an age-dependent molecular marker of innate immunity and immunoglobulin deficiency as results from its reactivity with IgG and IgA antibody. FEMS Immunol. Med. Microbiol. 48, 205–214.
