Bacterial growth and purification of R. akari
R. akari reference strain Hartford was propagated in egg yolk sacs of pathogen-free chicken embryos in BSL-3 containment as described previously [55] with some modifications. Briefly, six days old chicken embryos were inoculated and incubated at 35°C. The bacterial inoculum was adjusted; therefore, most embryos died between 8 to 9 days after inoculation. Yolk sacs were collected, homogenized in 2M NaCl solution and the suspension was centrifuged at 22 000 x g at 4°C. Pellet was resuspended in PBS and centrifuged again at low speed (200 x g for 10 min) to remove debris. The supernatant was then overlaid on 25% w/w sucrose cushion and centrifuged at 22 000 x g at 4°C to collect bacterial pellet. Resulted pellet was re-suspended in sucrose phosphate glutamate (SPG) buffer and purified by two rounds of Ultravist 370 (Bayer, Germany) discontinuous gradient centrifugation by using 32, 36, and 42% layers at 90 000 × g at 4°C for 50 min. Light bands containing intact bacteria were collected, pooled in SPG and centrifuged at 22 000 x g for 30 min [55]. Finally, the pellet was re-suspended in PBS and stored at -80°C until analysis. Giménez staining technique was employed to evaluate the purity of rickettsial cells.
Gel-free proteomics
The bacterial cells of R. akari (4 mg/mL) was pelleted by centrifugation (18 000 x g; 20 min; 4˚C) and washed with 300 µl PBS. The resulting pellets were re-suspended in 100 µl of 50 mM Tris pH 7.5 containing 0.1% RapiGestTM SF (Waters, UK) [56] and then incubated for 10 min at 95°C. After cooling, 200 µl of 0.1% RapiGestTM SF in 8 M guanidinium chloride (Sigma-Aldrich, USA) was added and incubated for another 20 min. Then, filter aided sample preparation – (FASP) [57] protocol was applied. Briefly, inactivated samples were transferred onto Amicon® Ultra – 10 kDa filters (Millipore) and washed twice with 100 mM ammonium bicarbonate (Sigma-Aldrich). Subsequently, proteins were quantified by bicinchoninic acid assay (QuantiPro™ BCA Assay Kit, Sigma-Aldrich) [58]. The samples were then reduced with 100 mM Tris (2-carboxyethyl) phosphine hydrochloride (TCEP, Sigma-Aldrich) and alkylated with 300 mM iodoacetamide (Sigma-Aldrich). Finally, the samples were digested with 2 μg of sequencing grade trypsin (Promega) overnight at 37 °C. Empore™ SPE Cartridges, C18, standard density, bed I.D. 4 mm (Sigma-Aldrich) were used to desalt peptide mixtures before drying to completion in a speed-vac. Prior to mass spectrometry analysis, the samples were re-suspended in 30 μl of 2% acetonitrile (ACN)/0.1% trifluoroacetic acid.
The samples were further analyzed by Ultimate 3000 RSLCnano system controlled by Chromeleon software (Dionex, USA), involving targeted mass spectrometry and LFQ as described earlier [59] with some modifications. Briefly, extracted peptide mixtures were loaded onto a PepMap100 C18, 3 µm, 100 Å, 0.075 × 20 mm trap column (Dionex) at 5 µl/min for 5 min. Separation was performed on a PepMap RSLC C18 column (0.075 x 150 mm, particle size 2.0 µm; Dionex) using 68 min gradient of 4-34% mobile phase B (80% ACN, 0.1% FA) with mobile phase A (0.1% formic acid, FA) and 21 min gradient of 34-55% mobile phase B at flow rate of 0.3 μl/min. Eluted peptides were electrosprayed into a Q-Exactive mass spectrometer using a Nanospray Flex ion source (Thermo Scientific, Bremen, Germany) to obtain positive ion full-scan MS spectra in the range 350-1650 m/z.
Acquired raw files were further processed in MaxQuant (version 1.6.7.0) [60]. Andromeda search engine [61] software was applied to identify proteins against the Rickettsia akari strain Hartford databases downloaded from Uniprot (September 21st, 2019). Identifications were accepted if at least two distinct reliable peptides matched the protein sequence, or the sequence coverage achieved at least 15%. Relative quantification was performed using the default parameters of the MaxLFQ algorithm [62], with the minimum ratio count set to 2.
Protein preparation and 2-D electrophoresis (2-DE)
The rickettsial pellet (2 mg wet-weight) was re-suspended in 4 mL of lysis buffer [28 mM Tris-HCl, 22 mM Tris-base, 200 mM dithiothreitol (DTT, Promega, USA), 2% SDS, protease inhibitor] and incubated for 30 min at 4°C under shaking. After incubation sample was boiled for 5 min at 100°C and cooled down on the ice for 5 minutes, followed by incubation with benzonase nuclease (Sigma-Aldrich, USA) 1ul/mL during 1 hour at room temperature. The supernatant was collected after centrifugation at 14 000 x g for 20 min at 4°C and proteins precipitated with an equal volume of chloroform, 5 volumes of methanol, and 3 volumes of MilliQ water was added prior centrifugation at 14 000 x g for 5 min at 4°C. After phase separation, the upper phase was discarded, the same volume of methanol was added and vortexed. The protein pellet was collected by centrifugation at 14 000 x g for 20 min at 4°C, dried with N2 and stored at -80°C. Samples were dissolved in lysis buffer No. 7 (8M urea, 2 M thiourea, 85 mM DTT, 2.5% Triton X-100) prior to rehydration.
The IPG strip (18 cm, pH 3-10 NL; GE Healthcare, USA) was rehydrated overnight with approx. 140 μg of total protein with 1.0% v/v IPG Buffer (GE Healthcare) and 0.001% bromophenol blue in buffer No.7. The isoelectric focusing (IEF) was carried out in the Ettan IPGphor 3 IEF System apparatus (GE Healthcare, USA). After IEF each IPG strip was incubated in equilibration buffer composed of 6M urea, 2% SDS, 30% glycerol, 375 mM Tris-HCl (ph8.8) in the presence of 1.0% DTT for 15 min, followed by equilibration for 15 min in 3.75% iodoacetamide (IAA, Sigma-Aldrich, USA) solution in the same equilibration buffer. Equilibrated strips were transferred onto 13% polyacrylamide gels to perform second dimension SDS-PAGE. After electrophoresis, gels were stained with Coomassie brilliant blue R-250 (Serva, Germany) and stored at 4°C until spot excision and trypsin digestion.
In-gel peptide digestion and MS analysis
Protein spots were excised, transferred into 1.5 mL Eppendorf tubes and distained in 50mM ammonium bicarbonate buffer containing 50% acetonitrile (ACN, Merck, Germany), reduced with 10 mM DTT in 100 mM ammonium bicarbonate, followed with alkylation in 50 mM IAA in 100 mM ammonium bicarbonate buffer [63]. Proteins were digested with porcine trypsin in 10mM ammonium bicarbonate with 10% acetonitrile at 37°C overnight. The enzymatic reaction was stopped with a 70% acetonitrile solution containing 1.0% trifluoroacetic acid (TFA, Merck, Germany), and peptides were subsequently extracted. Supernatants were concentrated on SpeedVac (Eppendorf, Germany) to a final volume of 20μl and analyzed by automated nanoflow reverse-phase (RP) liquid chromatography coupled to a Q-TOF (quadrupole time-of-flight) Premier Electrospray Ionisation (ESI) tandem mass spectrometer (Waters, USA). Separation of peptides was carried out on RP column BEH 130 C18 (200 mm x 75 µm, particle size 1.7 µm; Waters) using a 60 min gradient elution of 5-40% acetonitrile with 0.1% (w/w) formic acid at a flow rate of 0.3 μl/min. Samples were nanosprayed at 3.4 kV capillary voltage to Q-TOF detector and spectra were recorded from alternate scans at low (4 eV) and high (20-40 eV ramp) collision energies to obtain full-scan mass in ion range 50-1950 m/z. Finally, the data were processed by ProteinLynx Global Server 3.0.3 (Waters). For peak picking the following thresholds were applied: low energy 140 counts and high energy 30 counts. Precursors and fragment ions were coupled, using correlations of chromatographic elution profiles in low/high energy traces. Spectra were searched against the assembled Rickettsiaceae proteome database (downloaded on January 20th, 2019). Workflow parameters for the protein identification queries were: i) maximum one possible trypsin miscleavage; ii) a fixed carbamidomethyl cysteine, variable oxidized methionine and deamidated asparagine/glutamine; iii) the precursors and fragments mass tolerance was automatically determined by the software; iv) peptide matching was limited to less than 4% false discovery rate against the randomized database. Identifications were accepted if at least two distinct reliable peptides matched the protein sequence or the sequence coverage achieved at least 15%.
Membrane protein enrichment methods
The first method is based on the Triton X-114 phase separation [30]. Briefly, the purified R. akari cells were resuspended in Triton X-114 extraction solution (1% Triton X-114, 10 mM Tris pH 7.5, 5 mM EDTA) and incubated on rotating platform at 4°C for 4 hours. After the extraction sample was centrifuged at 15 000 x g at 4°C for 15 min, resulting in cell debris containing pellet and supernatant. This supernatant was layered over a chilled sucrose cushion, incubated at 37°C for 30 min and centrifuged at 500 x g for 20 min at 30°C to separate the lower detergent and the upper aqueous phase. The proteins from the Triton X-114 and aqueous fractions were precipitated in acetone at -20°C prior to separation.
As a second membrane enrichment technique, the cell surface biotinylation method was applied. Briefly, purified R. akari bacteria were labeled with Sulfo-NHS-LC-Biotin (Thermo Scientific, USA) and the biotinylated proteins were captured on streptavidin agarose resin (Thermo Scientific) as described previously [64]. The captured proteins were eluted from the streptavidin resin with a 5% solution of 2-mercaptoethanol-PBS at 30°C for 30 min, repeated 3 times, and the eluted proteins were precipitated in acetone at -20°C overnight. All protein fractions were then dissolved in Laemmli sample buffer and subjected to SDS-PAGE on 12% polyacrylamide gel. The whole lane was cut and processed by in-gel trypsin digestion and analyzed by LC-MS/MS as described above.
Preparation of recombinant proteins
The genes groEL, dnaK and 44 kDa uncharacterized protein (locus: A1C_04610) were amplified from the genomic sequence of R. akari (GenBank accession number: CP000847) with groEL primer pairs, 5'- TATA-CCATGG(NcoI)-CAACAAAACTTATTAAACACG -3' and 5'- TATA-AGATCT(BglII)-GGAAATCCATACCGCCCATA -3' and dnaK primer pairs, 5'- TATA-CCATGG(NcoI)-GAAAAGTAATAGGTATTGACCTTGG -3' and 5'- TATA-AGATCT(BglII)-TCTTCTTCGCTACATCCTGAAAATCG -3' and 44 kDa uncharacterized protein (locus: A1C_04610) primer pairs, 5'- AG-CCATGG(NcoI)-GTAAATTAAATAAATTAAATTTAACTATTGC -3' and 5'- CTGCAG(PstI)-CTAAATCTAATTTTAACCCTGCTCTAA -3'. Each PCR amplified products was inserted into the pEcoli-Cterm6xHN vector (Clontech, Takara Bio USA) and then competent E. coli BL21(DE3) (Novagen, Merck Bioscience) cells were transformed with each recombinant plasmid and cultivated in LB broth (Amresco, USA) with addition of carbenicillin (Sigma-Aldrich). Expression of recombinant proteins were induced with 0.1 mM isopropyl-β-D-thiogalactopyranoside (IPTG, Bioline, UK) at OD600 = 0.5-0.6, overnight 25°C under shaking.
Recombinant proteins were purified by a combination of two methods: in first-round E. coli pellets (200 mg wet-weight per each clone) were lysed and purified according to the manufacturer’s instruction with HisTalon buffer set (Clontech, USA) and using Talon Metal affinity resin (Clontech, USA). Additional purification of protein eluates was achieved with ÄKTA pure protein purification system (GE Healthcare, USA), using HisTrap FF crude (GE Healthcare, Sweden) 1 mL columns according to the manufacturer’s instruction. Elution was done with a gradient of 20 mM to 200 mM imidazole in a buffer of 20 mM sodium phosphate pH 7.4 containing 0.5 M NaCl. Concentrations of dialyzed protein eluates were determined by Pierce bicinchoninic acid assay kit (BCA, Thermo Scientific, USA) [58].
Human and animal sera
Rabbit serum and serum from BALB/c mice boosted with live Rickettsia akari and rickettsial 44 kDa recombinant protein, respectively, were prepared as described previously [27] with modifications. Briefly, 108 live rickettsial cells and 30 µg of recombinant purified protein mixed with complete Freund’s adjuvant were intraperitoneally administered into the rabbit and mice, respectively. Follow by booster immunizations with 108 live rickettsial cells and 30 µg of recombinant purified protein mixed with incomplete Freund’s on day 28. Blood samples were collected 7 days after the booster. The titers of IgG were determined by ELISA using their corresponding antigen. All sera were produced in our experimental animal facility at Biomedical Research Center of the Slovak Academy of Sciences.
Five rickettsial positive patients were diagnosed base on clinical signs. The patient’s blood was dispatched to our Department which is the National Reference Centre for Rickettsioses (established by Regional Authority of Public Health by the Ministry of Health of Slovak Republic) for rickettsia screening. Sera were tested by an indirect immunofluorescent assay using a panel of rickettsial antigens, including SFG rickettsia (R. slovaca, R. conorii, R.helvetica, R. rickettsii), R. typhi and R. akari. IFA test was considered positive if antibody titers were at a cutoff of 1/128 for IgG and 1/64 for IgM. A fourfold increase in titer was observed on three sera against R. akari antigen only. Two sera from healthy blood donors were obtained from the serum collection of our Department of Rickettsiology.
Serological analysis of R. akari antigen and recombinant proteins using western blot
R. akari proteins resolved by 2-DE were transferred (at 100 V/90min) onto 0.45μm PVDF membrane (Pall Life Sciences, USA) and blocked in 5% non-fat dried milk (Bio-Rad, USA) in PBS with 0.1% Tween-20 at 4°C overnight. Membranes were incubated with a serum of rabbit infected with R. akari or serum from dinfected patients at 1:1000 dilutions. After incubation, the membranes were probed with horseradish peroxidase-conjugated polyclonal swine anti-rabbit (Dako Denmark A/S, 1:3000) or polyclonal rabbit anti-human IgG (Dako Denmark A/S, 1:3000) secondary antibody, respectively. Visualization was carried out with a method of enhanced chemiluminescence (ECL) development [65].
In order to characterize the immunogenic properties of recombinant proteins, sixteen micrograms of each purified proteins were resolved by SDS PAGE 12% gels and immunoblotted. Membranes were sliced into strips, and each lane was incubated separately with different sera from infected patients at 1:1000 dilutions. After incubation and repeated washings membrane strips were probed and visualized as described above.
Preparation of “pre-absorbed” serum against rickettsial 44 kDa protein
The absorption procedure was described by Zhang et al. [66] with minor modifications. E. coli BL 21DE3 overnight cultures in LB broth with a density of 108 cells/ml were centrifuged, washed twice and re-suspended in 100 µl of rickettsial 44 KDa protein mouse serum, incubated for 2 h at 37°C and then overnight at 4 °C on a rotator. E. coli cells were centrifuged at 10 000 x g for 30 min; the supernatant was collected and used for Immunofluorescence assay.
Dual fluorescence staining of R. akari and 44 kDa uncharacterized protein (A8GP63)
Purified R. akari cells on a coverslip were fixed and permeabilized as described earlier [67]. Rickettsial cells were washed three times in PBS (pH 7.2) containing 2% bovine serum albumin (BSA; mixtures, PBSA) and then blocked with 5% BSA in PBS for 1 hour at 37°C. After washing, bacterial cells were incubated with a “pre-absorbed” mouse anti-serum against 44 kDa hypothetical protein (1:100) diluted in 2% PBSA for 1 hour at 37°C. After washing, cells were incubated with goat anti-mouse IgG secondary antibody conjugated with Alexa fluor 488 (Life Technologies, USA) diluted 1:1000 in PBS containing 2% PBSA. The cells were washed three times with PBSA and blocked again with 5% BSA in PBS for 1 hour at 37°C. R. akari cells were then stained by using polyclonal rabbit antiserum against live R. akari diluted 1:200 and a goat anti-rabbit IgG conjugated with Rhodamine (Life technologies, USA) diluted 1:2000. After five times washing with PBS, the coverslip was dried and mounted with Vectashield (Vector Laboratories) and viewed with fluorescence microscopy (model Eclipse Ni, Nikon Japan).
Database use and in silico analyses
Mass spectra of peptides were extracted by ProteinLynx Global Server version 3.0 and compared to UniProt database (https://www.uniprot.org/proteomes/UP000291740). Prediction of membrane proteins was performed using four independent programs: signal peptide prediction of identified peptides was completed by the program SignalP-5.0 (http://www.cbs.dtu.dk/services/SignalP/)[68]. Predictions of the subcellular localization of identified proteins was performed by program PSORTb version 3.0.2 (https://www.psort.org/psortb/)[31] and SOSUI-GramN (http://harrier.nagahama-i-bio.ac.jp/sosui/sosuigramn/sosuigramn_submit.html)[32]. Prediction of beta-barrel outer membrane proteins was made by a hidden Markov model PRED-TMBB program (http://bioinformatics.biol.uoa.gr/PRED-TMBB/input.jsp)[33]. Homology of identified protein with other rickettsial proteins was compared using the Basic Local Alignment Search Tool (BLAST) (https://blast.ncbi.nlm.nih.gov/Blast.cgi)[69]. Protein classification into COGs functional classes was carried out using public database EggNOG v5.0 (http://eggnog5.embl.de/#/app/home)[70]. Analysis of the 44 kDa protein was submitted to the bcell program (http://tools.immuneepitope.org/bcell/) which predicts flexible linear B-cell epitopes. The sequence was submitted with a signal peptide and a threshold 0.35 [71].