ENTEROCOCCUS FAECALIS AND/OR ENTEROCOCCUS FAECIUM ANTIGEN

Abstract

The present invention generally relates to the field of detecting and preventing infectious diseases caused by Enterococcus faecalis and/or Enterococcus faecium. More specifically, the invention relates to an Enterococcus faecalis and/or Enterococcus faecium antigen which comprises at least one unit having the following general formula:

##STR00001##

Claims

1. An isolated antibody, or an antibody fragment, to an Enterococcus faecalis and/or Enterococcus faecium antigen, wherein the antigen comprises at least one unit having the following general formula: ##STR00004## wherein R.sub.1 are independently of one another selected from the group consisting of H, OH, OCH.sub.3, OAc, OC.sub.nH.sub.m, OFo, OAcyl, SH, SCH.sub.3, SC.sub.nH.sub.m, SFo, SAc, SAcyl, NH.sub.2, NHCH.sub.3, NHC.sub.nH.sub.m, NHFo, NHAc, NHAcyl, PO.sub.2(OR.sub.2).sub.2, F, Cl, Br and I, wherein Fo is formyl, R.sub.2 is independently selected from the group consisting of H, C.sub.nH.sub.m, C.sub.nH.sub.m-1NHR.sub.3 and C.sub.nH.sub.m-1N(CH.sub.3).sub.3, wherein R.sub.3 is independently selected from the group consisting of H, Fo, Ac and acyl, X are independently of one another selected from the group consisting of O, S, CH.sub.2, NH and POOR.sub.4, wherein R.sub.4 is H, C.sub.nH.sub.m, C.sub.nH.sub.m-1NHR.sub.3 and C.sub.nH.sub.m-1N(CH.sub.3).sub.3, Y are independently of one another selected from the group consisting of O, S, CH.sub.2 and HPO.sub.4, CA is independently of one another selected from the group consisting of C.sub.1-C.sub.6-acyl radicals, in particular hydroxyacyl radicals, preferably lactyl, H, OH, OCH.sub.3, OAc, OC.sub.nH.sub.m, OFo, OAcyl, SH, SCH.sub.3, SC.sub.nH.sub.m, SFo, SAc, SAcyl, NH.sub.2, NHCH.sub.3, NHC.sub.nH.sub.m, NHFo, NHAc, NHAcyl, PO.sub.2OR.sub.2, F, Cl, Br and I, wherein Fo is formyl, R.sub.2 is independently selected from the group consisting of H, OC.sub.nH.sub.m, OC.sub.nH.sub.m-1NHR.sub.3 and OC.sub.nH.sub.m-1N (CH.sub.3).sub.3, wherein R.sub.3 is independently selected from the group consisting of H, Fo, Ac and acyl, wherein m=2n+1 and n is selected from the set of natural numbers from 1 to 10.

2. The antibody, according to claim 1, wherein the antigen has two sugars that are in the D configuration.

3. The antibody, according to claim 1, wherein the antigen comprises a disaccharide consisting of a furanose Gal and a pyranose Glc has a structure that is selected from the following: .fwdarw.-v)-D-Galf-(1.fwdarw.-z)-D-Glcp-(1.fwdarw.-, .fwdarw.-v)-D-Galf-(1.fwdarw.-z)-D-Glcf-(1.fwdarw.-, .fwdarw.-v)-D-Galp-(1.fwdarw.-z)-D-Glcp-(1.fwdarw.- or .fwdarw.-v)-D-Galp-(1.fwdarw.-z)-D-Glcf-(1.fwdarw.-, wherein v and z are in each case 1, 2, 3, 4, 5 or 6.

4. The antibody, according to claim 1, wherein the antigen has the following formula: ##STR00005##

5. The antibody, according to claim 1, wherein, for the antigen, R.sub.1 is OH, X is O, Y is O and CA is lactyl.

6. The antibody, according to claim 1, wherein the antigen has a molecular weight of approximately 50,000 to 150,000 Da.

7. The antibody, according to claim 1, said unit of the antigen occurs at least 100 times.

8. The antibody, according to claim 1, that is a recombinant antibody, or a recombinant fragment thereof.

9. A composition comprising an antibody according to claim 1.

10. The composition as claimed in claim 9, comprising a pharmaceutically acceptable carrier.

11. The composition as claimed in claim 9, characterized in that it is a vaccine against Enterococcus faecalis and/or Enterococcus faecium.

12. A kit for detecting Enterococcus faecalis and/or Enterococcus faecium in a sample, comprising an antibody according to claim 1, optionally bound to a support.

13. The kit according to claim 12, characterized in that the antigen or the antibody is labeled with a marker.

14. A method for immunizing a subject against Enterococcus faecalis and/or Enterococcus faecium, wherein said method comprises administering to the subject an antibody of claim 1.

15. An isolated antibody, or an antibody fragment, to an Enterococcus faecalis and/or Enterococcus faecium antigen, wherein the antigen comprises at least one unit having the following general formula: ##STR00006## wherein R.sub.1 is OH, X is O, Y is O and CA is lactyl.

16. A method for immunizing a subject against Enterococcus faecalis and/or Enterococcus faecium, wherein said method comprises administering to the subject an antibody of claim 15.

Description

[0112] In the figures,

[0113] FIG. 1 depicts the .sup.1H NMR spectrum of the capsular polysaccharide from E. faecalis strain type 5. The spectrum was recorded at 600 MHz and 27° C. The letters refer to the carbohydrate radicals as depicted in FIG. 3, and the Arabic numerals refer to the protons on the corresponding radicals; LA, lactic acid.

[0114] FIG. 2 depicts sections of the ROESY spectrum of the capsular polysaccharide from E. faecalis strain type 5. The spectrum was recorded at 600 MHz and at 27° C. The letters refer to the carbohydrate radicals as depicted in FIG. 3, and the Arabic numerals refer to the protons on the corresponding radicals. The NOE contacts between the radicals are underlined and depicted in italics.

[0115] FIG. 3 depicts the chemical structure of the repetitious unit of the capsular polysaccharide of E. faecalis strain type 5. LA, lactic acid. The Galf radical is assumed to have the D configuration.

[0116] FIG. 4 depicts binding of a rabbit antiserum to the bacterial cells of E. faecalis strain type 5. The antigens used are indicated in the legend. The values indicated represent in each case the average of at least two measurements.

[0117] FIG. 5 depicts the inhibition of opsonophagocytosis of E. faecalis strain type 5 bacterial cells by rabbit antiserum. All antisera were used as 1:200 dilutions. The inhibitors are indicated in the legend. The data indicated represent an average of at least 4 measured values, and the error bars represent the SEM. The opsonophagocytotic activity without inhibitor was >70% for all antisera.

[0118] FIG. 6 depicts the binding of antibodies generated in rabbits to capsular polysaccharide of E. faecalis type 2 and type 5 to the corresponding purified antigen.

[0119] FIG. 7 depicts the determination of curing by opsonophagocytosis of E. faecalis type 2, which were produced against the purified antiserum. The values clearly demonstrate that protective antibodies were produced in rabbits, which resulted in the in vitro assay to destroy the bacteria.

EXAMPLE 1

[0120] Elimination of Enterococcus faecalis by opsonophagocytosis is accomplished inter alia by antibodies which are directed to carbohydrate antigens of the cell wall and of the bacterial capsule. Recently, lipoteichoic acid (LTA) has been identified as target of the opsonizing antibodies in E. faecalis strain 12030. However, serum raised against purified LTA does not kill any bacterial strains having the CPS-C and -D serotypes.

[0121] The present example comprises isolating a novel capsular polysaccharide from the E. faecalis type 5 strain, a CPS-D strain, by enzymatic digestion of the cell wall and by gel permeation and anion exchange chromatography.

[0122] The isolated polysaccharide is analyzed by sugar analysis, one-dimensional and two-dimensional homonuclear and heteronuclear .sup.1H and .sup.13C NMR spectroscopy.

[0123] A novel E. faecalis capsular polysaccharide is identified which includes an unusual .fwdarw.6)-3-O-[1-carboxyethyl]-β-Galf-(1.fwdarw. unit in the repetitious unit.

[0124] A rabbit antiserum induced by means of immunization of heat-inactivated E. faecalis type 5 bacteria cells contained both antibodies specific to the novel capsular polysaccharide and antibodies to LTA of said strain.

[0125] However, opsonophagocytosis of E. faecalis type 5 by this antiserum was inhibited only by the purified polysaccharide but not by LTA.

[0126] This example therefore illustrates a possibility of how to identify a novel capsular polysaccharide as antigen in E. faecalis type 5 that is immunogenic and can serve as target for opsonizing antibodies.

[0127] This example further illustrates how to assay the immunogenicity of antigens derived from E. faecalis.

EXAMPLE 2

[0128] 2a) Bacteria Strains and Cultures

[0129] Capsular polysaccharides were isolated from E. faecalis type 5 (Maekawa, S., et al., Microbiol. Immunol., 36:671-681, 1992) a CPS-D strain, using a recently described serotyping system (Hufnagel, M., et al., J. Clin. Microbial. 42:2548-2557, 2004). Bacteria cells were cultured from the starter cultures in a Columbia nutrient solution (Becton Dickinson, Sparks, MD, USA) enriched with 1% glucose, without agitation at 37° C. for 2 hours.

[0130] 2b) Antisera

[0131] Antisera to whole bacteria cells of E. faecalis type 5 have been described in the past (Hufnagel, M., et al., J. Clin. Microbiol. 42:2548-2557, 2004). The antiserum to LTA was produced with LTA purified from the strain 12030, as described previously (Theilacker, C., et al., Infect. Immun. 74, 2006). A female white New Zealand rabbit was immunized subcutaneously with 100 μg of LTA suspended in complete Freund's adjuvant, and thereafter with the same LTA dose suspended in incomplete Freund's adjuvant seven days later, and then with 10 μg booster doses every three days in the following week.

[0132] 2c) Preparation and Characterization of the Capsular Colysaccharide.

[0133] E. faecalis LTA was isolated as described previously (Huebner, J., et al., Infect. Immun. 67:1213-1219, 1999; Theilacker, C., et al., Infect. Immun. 74, 2006). The antigen described herein was prepared as follows: briefly, bacterial cells were harvested by centrifugation and digested by adding mutanolysin and lysozyme (in each case 100 pg/ml, Sigma Chemicals, St. Louis, Mo., USA in PBS enriched with 5 mM MgCl.sub.2, 1 mM CaCl.sub.2 and 0.05% NaN.sub.3) at 37° C. for 18 hours. Insoluble material was removed by centrifugation, and the supernatant was treated with nucleases (DNase I and RNase A, 100 pg/ml) at 37° C. for 4 hours, followed by 18 hours of adding proteinase K (100 pg/ml, all available from Sigma Chemicals) at 56° C. The supernatant was precipitated by adding ethanol (final volume 80%) and then collected by centrifugation. After dialysis against deionized H.sub.2O, the material was lyophilized. For gel permeation chromatography, the material was dissolved in 0.01 M ammonium bicarbonate buffer solution and applied to a Sephacryl S-400 column (1.6×90 cm) (GE Healthcare, Uppsala, Sweden). Fractions eluting at a K.sub.av of about 0.45 were combined, dialyzed and lyophilized. The material was resuspended in 20 mM of NaHCO.sub.3, pH 8.4 and applied to an anion exchange column (Sepharose Q FF, GE Healthcare). Bound antigen was eluted from the column by way of a linear NaCl gradient, and fractions comprising polysaccharides were identified by a Dubois assay (Dubios, M., et al., Anal. Chem. 28:350-356, 1956.) and immunoblotting using a rabbit anti-type 5 immune serum. Immunoreactive material eluting at 450 mM NaCl was combined, dialyzed and lyophilized. The final purification step performed was a gel permeation chromatography on a 1.5×75 cm Toyopearl HW-40 (Tosoh Corporation, Tokyo, Japan) column. The purity of the isolated material was confirmed by means of SDS PAGE using a 10% Bis-Tris gel and an MOPS running buffer (Invitrogen, Karlsruhe, Germany) and by Coomassie (Invitrogen) and PAS (Sigma) staining according to the manufacturer's instruction.

[0134] Furthermore, a Western blot of material removed by SDS PAGE was stained using an anti-type 5 rabbit antiserum.

[0135] 2d) Result of the Purification of the Capsular Polysaccharide

[0136] Capsular polysaccharide of the E. faecalis strain type 5 was mobilized by enzymatic digestion of peptidoglycan from the bacterial cells. The extracted material eluted in the form of two carbohydrate-containing fractions in Sephacryl S-400 gel chromatography. One fraction eluting in the void volume comprised LTA as determined by .sup.1H NMR analysis (data not shown). A large, second fraction at a K.sub.av of around 0.45 was further purified by anion exchange chromatography using Q Sepharose. Small amounts of immunoreactive material eluted at 450 mM NaCl, which material comprised only glucose and galactose, and which was subjected to further analysis after gel permeation chromatography on Toyopearl HW-40S. The SDS PAGE of this purified material showed a single broad band at 100 kDa which was stained by PAS but not by Commassie blue. A Western blot stained with anti-type 5 antiserum likewise showed a single broad band at 100 kDa and no further bands (data not shown).

EXAMPLE 3

[0137] Provision of LTA

[0138] LTA was prepared by butanol extraction and hydrophobic interaction chromatography as described previously (Theilacker, C., et al., Infect. Immun. 74, 2006). The purity of the LTA preparations was evaluated by SDS PAGE and Western blot analysis using the corresponding antiserum to whole bacterial cells (cf. above). The structural identity of LTA was confirmed by NMR spectroscopy as recently described (Theilacker, C., et al., Infect. Immun. 74, 2006).

[0139] General and Analytical Methods

[0140] Hydrolysis was carried out using 2 M trifluoroacetic acid (120° C., 3 h). Monosaccharides were converted to alditol acetates and analyzed by GC in a Hewlett-Packard 5890 chromatograph with an SPB-5 column (30 m×0.25 mm×0.25 pm, Supelco, Munich, Germany), using a temperature program of 150° C. for 3 minutes, then 3° C. min.sup.−1 up to 300° C. The absolute arrangement of the sugar radicals was determined as described previously (Haseley, S. R., et al., Eur. J. Biochem. 244:761-766, 1997; Leontein, K., et al., Carb. Res. 62:359-362,1978).

[0141] NMR Spectroscopy

[0142] The sample was substituted three times with 99.0% .sup.2H.sub.2O, lyophilized and redispersed in 99.9% .sup.2H.sub.2O. All one- and two-dimensional spectra were recorded on a Bruker DRX Avance 600 MHz spectrometer (working frequencies of 600.31 MHz for .sup.1H NMR and 150.96 MHz for .sup.13C NMR) using a conventional Bruker Software (Bruker, Rheinstetten, Germany) at 27° C. The chemical shifts are indicated in relation to acetone (δH 2.225; δC 31.45). Correlation spectroscopy (COSY), and total correlation spectroscopy (TOCSY), and ROESY were recorded using datasheets (t1×t2) containing 4096×512 points, and 32 scans were carried out.

[0143] TOCSY and ROESY were carried out in a phase-sensitive manner according to the method of States et al., with a mixing time of 100 ms being used for TOCSY (States, D. J., et al., J. Magn. Reson. 48:286-292, 1982). The .sup.1H, .sup.13C correlations were measured in .sup.1H detection mode by means of a multiple quantum coherence (HMQC) with proton decoupling in the .sup.13C domain using datasets containing 2048×256 points, with 128 scans being recorded for each t.sub.1 value (Bax, A., et al., J. Am. Chem. Soc. 109:2093-2094, 1986; Summers, M. F., et al., J. Am. Chem. Soc. 108:4285-4294, 1986).

[0144] Chemical Analysis and NMR Spectroscopy

[0145] In the low field region, the .sup.1H NMR spectrum of the purified polysaccharide (FIG. 1) showed two anomeric signals at δ 5.315 (radical A, [.sup.3J.sub.H1,H2<2 Hz]), and at δ 4.542 (radical B, [.sup.3J.sub.H1,H2=7.8 Hz]), which were identified as β-Galf and β-D-Glcp. Furthermore, the Dublett at δ 1.361 was identified as a methyl group which belongs to a lactic acid radical (LA) (Knirel, Y. A., et al., Carbohydr. Res. 259, 1994; , Knirel, Y. A., et al., Carbohydr. Res. 235:C19-23, 1992). The presence of a D-Glcp radical was confirmed by chemical analysis, assigning of the absolute configuration, and by NMR spectroscopy data. The Galf radical substituted with lactic acid in position C-3 was identified only by NMR data (Beynon, L. M., et al., Eur. J. Biochem. 250:163-167, 1997; Knirel, Y. A., et al., Carbohydr. Res. 259, 1994; , Knirel, Y. A., et al., Carbohydr. Res. 235:C19-23, 1992). All .sup.1H and .sup.13C chemical shifts of the capsular polysaccharide of E. faecalis type 5 (Table 1) were determined from .sup.1H,.sup.1H COSY and TOCSY, and from .sup.1H,.sup.13C HMQC spectra.

TABLE-US-00001 TABLE 1 Chemical shift .sup.1H and .sup.13C [δ] H1 H2 H3 H4 H5 H6.sup.a Radical C1 C2 C3 C4 C5 C6 H6.sup.b A 5.315 4.346 3.932 4.225 4.040 3.768 4.019 .fwdarw.6)-β- 109.26 80.26 84.96 82.41 70.55 71.90 Galf- B 4.542 3.460 3.663 3.465 3.500 3.742 3.930 .fwdarw.3)-β- 103.22 74.05 82.44 68.80 76.26 61.26 D-Glcp- LA 4.038 1.361 Lactic 181.29 77.72 19.21 acid

[0146] Carbon atom signals shifted to low field indicated substitutions on C-6 and C-3 of β-Galf (radical A, δ 71.90 and δ 84.96) and substitutions on C-3 of β-D-Glcp (radical B, δ 82.44). The sequence of the radicals in the repetitious unit was determined by ROESY experiments. Strong NOE contacts between the radicals were found between the protons A1 (δ 5.315) and B3 (δ 3.663), and B1(δ 4.542) and A6a (δ 3.768) (FIG. 2). Thus the structure of the repetitious unit of the isolated polysaccharide was as depicted in FIG. 3.

EXAMPLE 4

[0147] ELISA Studies

[0148] ELISA experiments were carried out by customary methods as described previously (Theilacker, C., et al., Infect. Immun. 74, 2006). Briefly, microtiter plates were coated with various carbohydrate antigens derived from E. faecalis (10 pg/ml in 0.04 M phosphate buffer, pH 7.0), and left at 4° C. for 18 hours. Washing steps were carried out using PBS containing 0.05% Tween 20. The plates were blocked with 3% skimmed milk in PBS-0.02% sodium azide at 37° C. for 2 hours. The secondary antibody used was a goat anti-rabbit IgG alkalinephosphatase conjugate (Sigma), diluted to 1:1000, with p-nitrophenyl phosphate being used as substrate (Sigma). After incubation at 37° C. for 60 minutes, absorption was measured at 405 nm.

EXAMPLE 5

[0149] Opsonophagocytosis Assay

[0150] An opsonophagocytosis assay was performed as described previously (Theilacker, C., et al., Infect. Immun. 74, 2006). Baby rabbit serum (Cedarlane Laboratories, Hornby, Ontario, Canada) absorbed with the target bacterial strain was used as complement source. The opsonic activity of the immune sera was compared to that of the controls containing normal rabbit serum. The immune serum was heat-inactivated at 56° C. for 30 minutes before use. Negative controls comprised sample tubes which either did not contain any polymorphonuclear leucocytes or any complement or any serum. The opsonic activity of the serum was calculated as follows: [1−(CFU immune serum at 90 min/CFU preimmune serum at 90 min)]×100.

[0151] Opsonophagocytosis inhibition was studied by incubating antiserum in a concentration of 1:200 with from 0.08 to 100 μg/ml at 4° C. for 60 minutes. After incubation, the absorbed serum was added and the opsonophagocytosis assay was continued as described above. Without inhibition, all sera had a minimum opsonophagocytosis activity of >70% of the inoculum.

EXAMPLE 6

[0152] Immunochemical Characterization

[0153] Serum against bacteria cells of E. faecalis type 5 (Maekawa, S., et al., Microbiol. Immunol. 36:671-681, 1992) was reactive to the purified polysaccharide (FIG. 4). The antiserum also contained high levels of IgG antibodies to E. faecalis LTA (FIG. 4). Since anti-LTA antibodies mediate opsonophagocytosis of the E. faecalis strain 12030, we intended to determine the specificity of opsonizing antibodies for the E. faecalis type 5 strain. Purified LTA did not inhibit the opsonophagocytosis activity of anti-type 5 serum (FIG. 5), in agreement with the observation that there is little cross-reactivity of opsonizing antibodies between E. faecalis type 5 and E. faecalis 12030 (Hufnagel, M., et al., J. Clin. Microbiol. 42:2548-2557, 2004). However, the purified capsular polysaccharide was a potent inhibitor of opsonizing antibodies to type 5 (FIG. 5). Rabbit serum against purified LTA, which promotes opsonophagocytosis of strain 12030 in the opsonophagocytosis assay, was non-opsonizing against the type 5 strain, confirming the results from our inhibition studies (data not shown).

EXAMPLE 7

[0154] Antibodies to the two polysaccharides of type 2 and of type 5 were raised in rabbits: a female white New Zealand rabbit was immunized subcutaneously with 100 μg of purified polysaccharide of type 2 and type 5 which had been suspended in complete Freund's adjuvant, and was then immunized with the same dose of polysaccharide suspended in incomplete Freund's adjuvant seven days later, and thereafter with 10 μg booster doses every three days for the following week.

[0155] The polysaccharides were obtained by the method described under 2d). In each case two rabbits were used for immunization and the antiserum was subsequently obtained from the rabbits. The antisera obtained from the rabbits were studied in an ELISA. To this end, both the polysaccharide of type 2 and the polysaccharide of type 5 were bound to the microtiter plates. Surprisingly, antibodies were found to be produced both against purified polysaccharide of type 5 and against polysaccharide of type 2. This demonstrated that the antigen is immunogenic, a fact that was not necessarily expected in view of the fact that it is a “T cell-independent” antigen. Production of the antibodies is depicted in FIG. 6.

EXAMPLE 8

[0156] Opsonophagocytic Assay

[0157] This in vitro assay involved combining granulocytes, antiserum and bacteria and testing whether it was possible to kill the Enterococcus faecalis bacteria by the antibodies generated. The exact conditions of the assay mixture were as follows: fresh whole blood was obtained from healthy donors and admixed with a heparin dextran buffer. The white blood cells were then purified and adjusted to a defined number (5×10.sup.6 cells/ml). The bacteria to be studied were removed by centrifugation from a culture at average logarithmic growth and adjusted spectrophotometrically likewise to 5×10.sup.6 cells/ml. The complement source used was lyophilized baby rabbit serum diluted 1:15 with cell culture medium and absorbed with the target bacteria strain in order to remove existing antibodies to the Enterococcus strain used. The rabbit serum was likewise diluted with cell culture medium in accordance with the experimental set up. For the experiment, in each case 100 μl of the bacterial suspension, 100 μl of leucocytes (bacteria: leucocytes ratio of 1:1), 100 μl of the complement source and 100 μl of the corresponding antibody dilution were mixed. The initial bacterial count was determined by dilution and plating out, and the experimental mixture was then incubated in an end over end rotator at 37° C. for 90 minutes. At the end of the experiment, the bacteria in the experimental mixture were likewise diluted again and plated out, and the number of colonies was counted on the next day. The reduction of the bacterial count between inoculum and the bacterial count at the end of the experiment was expressed as opsonophagocytosis-mediated killing in percent. This value represents the best surrogate marker for a protective immuno response to bacterial infectious pathogens.

[0158] The results of the experiment are depicted in FIG. 7. They demonstrate that serum dilutions of from 1:50 to 1:200 result in a marked killing of approx. 80%, while this effect is lost with further dilutions of the serum. This indicates that higher antibody titers to this antigen represent a protection of the organism from infection with E. faecalis.