Protein candidates for the serodiagnosis of rickettsioses

Abstract

Four Rickettsia species frequently cause incapacitating, life-threatening illnesses: Rickettsia prowazekii (agent of epidemic typhus, ET), Rickettsia typhi (agent of murine typhus, MT), Rickettsia conorii (agent of Mediterranean spotted fever, MSF), and Rickettsia rickettsii (agent of Rocky Mountain spotted fever, RMSF). The clinical manifestations of most rickettsioses are characterized by a continuous spectrum in which some cases reemerge worldwide. However, some examples showed inconsistent clinical manifestations hardly correlated with geographical context, which makes clinical diagnosis uncertain.

The laboratory diagnosis of rickettsioses is based on serological testing (La Scola & Raoult, 1997), cell culture and molecular techniques (La Scola & Raoult, 1997). In the French National Reference Center (FNRC) for rickettsioses, the gold diagnostic standard is serology (immunofluorescence assay, IFA) followed by real-time PCR, which allows identification of Rickettsiae at the species level (Renvoise et al., 2011). ELISA (Fuller Laboratories) methods are also currently available for non-referenced laboratories which are focused on more specific reactivities to the lipopolysaccharide (LPS) antigens (IgG screening) and rOmpB protein antigens (IgG/IgM). However, to date, serological testing only allows identification of Rickettsiae at the group or clade level. The advantage of Western blot and antigen adsorption is identification of Rickettsiae at species level (Sompolinsky et al., 1986; Raoult & Dasch, 1989a, b, 1995). In addition, the diagnosis of suspected rickettsiosis requires several steps. The choice of complementary approaches depends on clinics and the results of serology. Diagnosing rickettsioses is still a great challenge. Research has been focused on new diagnostic applications that can increase sensitivity and specificity. The aim of the present work was to propose a diagnostic test based on recombinant proteins for discrimination of Typhus group (TG) Rickettsiae from Spotted fever group (SFG) Rickettsiae. We described here the preliminary results of screening of 20 rickettsial recombinant proteins by ELISA with sera of infected patients with R. typhi and R. conorii, respectively.

The choice of protein targets was defined according to previous studies with a focus on immunogenic proteins (Renesto et al., 2005, 2006) and proteins involved in physiopathological processes: RickA (Balraj et al., 2008a, b) and rOMPB, rOMPA, adr2 (Renesto et al., 2006). The nucleic acid sequences of ORFs were extracted from the NCBI genomic library. The predicted signal peptide (http://bp.nuap.nagoya-u.ac.jp/sosui/sosuisignal/SOSUIsignal_submit.html) sequence was removed.

DNA of R. prowazekii strain Madrid E and R. rickettsii strain Sheila Smith was extracted using a commercially available kit (Qiagen, Chatsworth, CA) according to manufacturer's instructions.

Cloning and expression of 20 targets (Gateway Cloning Technology/Invitrogen Life Technologies) were performed as previously described (Sekeyova et al., 2010). The vectors used to generate an entry clone were pDONR201 and pETG-20A, which generated expression clones containing an N-His6 tag and a fusion protein thioredoxin (TRX) (Canaan et al., 2004; Vincentelli et al., 2011). The resulting entry and expression clones were transformed into Escherichia coli DH5α cells, and constructions were confirmed by PCR screening and DNA sequencing, respectively.

All steps of expression and purifications were performed as previously described (Sekeyova et al., 2010; Vincentelli et al., 2011). Briefly, expression vectors carrying the 20 targets were transformed into E. coli strain Rosetta (DE3) pLysS (Novagen). The growth conditions, induction and harvest were done as previously described (Vincentelli et al., 2011). The proteins were purified by affinity chromatography based on the affinity of the histidine tag (HHHHHH) with nickel ions as previously described (Sekeyova et al., 2010). The identity of recombinant protein was confirmed by mass spectrometry.

Modified ELISA assay was performed as previously described (Sekeyova et al., 2010). The human sera were diluted 1/1000 in phosphate-buffered saline containing 5% milk and 0.05% Tween-20 (PBST-milk). Alkaline phosphatase-conjugated goat anti-human IgG (whole molecule; Sigma; 1/5000), alkaline phosphatase yellow para-nitro-phenyl phosphate (pNPP) (Sigma) were used as described (Sekeyova et al., 2010). The reaction was read with a microplate reader (Multiskan EX; Labsystems, Thermo Fisher Scientific, Waltham, MA) at a wavelength of 405 nm and data were analysed with graphpad prism (San Diego, CA). A positive control consisted of positive serum actively infected with R. typhi and R. conorii; a negative control consisted of negative serum. Each serum sample was tested at least in duplicate. The cut-off was determined as described by Sekeyova et al. (2010). Any samples exhibiting absorbance above the cut-off value were considered positive.

In this study, sera from 10 patients with an infection due to R. typhi (TG group) and sera from 28 patients with an infection due to R. conorii (SFG group) diagnosed at FNRC for rickettsioses were included in this study after giving informed consent. The diagnosis was based on serology (IFA) and real-time PCR assays. Titer cut-off values were ≥ 128 for IgG and 64 for IgM, as previously described (Murphy et al., 2011). A negative titer was reported when the initial serum screening result was negative. A titer of 0 was reported when an initial screening result was positive but no Ig was detected. The first real-time PCR screening was systematically performed with a set of primers and probe targeting SFG Rickettsia (‘1029’ and ‘Rco7’) as described by Renvoise et al. (2011). When TG Rickettsia was suspected from the clinical and serological results, a second run of PCR was done with set ‘274’ (Socolovschi et al., 2009). A control group (group HBD) consisted of 10 healthy blood donors.

Here, we describe the preliminary results obtained with 20 recombinant proteins of R. prowazekii and R. rickettsii which may be useful tools in the detection of Rickettsiae in clinical samples.

Our first attention was drawn to using these proteins as versatile discriminatory markers between the TG and the SFG Rickettsiae group, as both pathologies occur frequently in our Mediterranean district.

To discriminate between R. typhi and R. conorii patients, we screened 20 recombinant proteins (10 of TG R. prowazekii and 10 of SFG R. rickettsii). Until now, very few studies have focused on rickettsial recombinant proteins. There has been no large-scale screening, as done for another intracellular bacterium, Coxiella burnetii (Vigil et al., 2010, 2011) or Chlamydia trachomatis (Cruz-Fisher et al., 2011). Protein array is a widely used technique for numbered bacteria, but has not yet been performed for Rickettsiae.

Four recombinant proteins were found to cross-react with sera infected with R. typhi and with R. conorii (Table 1). This result is not surprising because they belong to well conserved bacterial proteins: groEL, Adr2, murC and EF-Tu (Table 1). Adr2 is ubiquitously present in Rickettsiae and acts as one of the putative ligands recognized by host cell surface proteins. Rickettsial entry into the host cell is mediated by the rOmpB protein, which attaches to the host cell receptor Ku70, a component of the DNA-dependent protein kinase (Uchiyama et al., 2006). Several tested recombinant proteins were used to distinguish between infections due to R. typhi, despite their origin, targets of R. prowazekii (RP016, groEL, RP173) and R. rickettsii (PLD, Sca10, EF-Tu, A1G_00215), supporting an already documented cross-reactivity among Rickettsia species. Interestingly, among these targets we found Sca10 protein, belonging to a large family of outer membrane proteins known as the surface cell antigen (Sca) family proteins, and PLD, involved in rickettsial adherence and invasion of Vero cells. However, these results may be underestimated considered a small cohort of patients infected with R. typhi. Infection due to R. typhi is sporadic in Europe (Bechah et al., 2008) and our cohort (n = 10) therefore represents a 1-year collection of patients diagnosed in FNRC for rickettsioses. Almost all of our patients are imported MT, caused by travel to endemic zones (Parola et al., 2005; Bitam et al., 2009). Thus, it will be suitable to validate these diagnostic targets on a larger study population.

In conclusion, no individual protein was enough to be used to diagnose SFG, except for the three targets already used for the diagnosis of both MT and MSF: groEL, adr2 and EF-Tu. Considering these preliminary results carefully, optimization of recombinant protein-based ELISA may be an interesting alternative for the diagnosis of rickettsial diseases. The advantage of ELISA is rapidity, low cost and possible development of high throughput screening requiring only a small amount of the patient's sera (< 1 µL). ELISA may be also be recommended as a complementary diagnostic technique, in addition to IFA (reference method), adsorbed Western blot and molecular tools.

Acknowledgements

This work was supported by Direction Générale de l'Armement (No. CP209812DGA0004), 91710 Vert-le-Petit, France. The authors would like to thank Madame Françoise Ramisse and Monsieur Pascal Rameil for all their cooperation in this project.

References