Ehrlichia canis (Jake Strain) gene sequences used in the study are available in the Integrated Microbial Genomes (IMG).¹ E. canis gene synthesis was performed (GenScript, Piscataway, NJ, USA) and the genes were cloned into a pIVEX-2.3d or pET-14b vector containing a T7 promoter/terminator and a 6× His-tag sequence. Plasmids with cloned genes were lyophilized and stored at −20°C prior to use.

Cell-free expression of the E. canis immunoreactive proteins was performed using the S30 T7 High-Yield Protein Expression System (Promega, Madison, WI, USA). Plasmids were transformed into Stellar competent cells (Takara, Mountain View, CA, USA) and plasmid was extracted and purified using QIAprep Spin Miniprep Kit (Qiagen, Germantown, MD, USA) as previously described (21). The recombinant plasmid was mixed with E. coli extract, a premix and reaction mixture and incubated at 37°C with agitation at 750 rpm for 3 h. The cell-free expressed protein was stored at 20°C until use.

A panel of 15 naturally infected dog sera gifted from United States, Colombia, and Brazil that were confirmed positive for E. canis antibody by IFA were used in this study. Sera collected from a dog experimentally infected with E. canis (needle inoculation) on days 0, 7, 14, 21, 28, 35, 42 and 56 post infection was used to assess temporal antibody responses to the proteins (16).

Immunoreactivity and diagnostic sensitivity of cell-free expressed E. canis proteins were evaluated by ELISA as described previously (21, 22). Briefly, anti-His-antibody coated ELISA plates (GenScript) were blocked with Starting Block Blocking Buffer (Thermo Fisher) at room temperature for 20 min with agitation (300 rpm). After washing twice with PBS-Tween 20 (0.05%) (PBST), plates were coated with cell-free expression lysate containing His-tagged recombinant proteins (50 μL) diluted (1:50) in blocking buffer (TBST, 2% nonfat dry milk) and incubated overnight at 4°C. The plates were washed five times and diluted dog sera (1:200, 50 μL) were added to each well and incubated at room temperature for 1 h with agitation. Plates were washed five times and incubated with alkaline phosphatase-labeled goat anti-dog IgG (H + L) secondary antibody (100 μL; 1:5000, KPL, Gaithersburg, MD), and incubated for 1 h at room temperature with agitation. The plates were washed and BluePhos substrate (100 μL; KPL) was added and incubated for 30 min in the dark with agitation. Color development was measured at A₆₅₀ on a VersaMax microplate reader (Molecular Devices) and data analyzed by Softmax Pro 7 software (Molecular Devices). The final optical density (OD₆₅₀) was determined after subtracting OD₆₅₀ value of the negative control (cell-free lysate from empty vector). Positive and negative dog sera were included as controls.

Linear epitopes in the E. canis TRP (Ecaj_0126) were mapped with overlapping 25 amino acid peptides (GenScript, Piscataway, NJ, USA) representing the entire open reading frame, except three 35 amino acid TRs which were synthesized separately. Lyophilized peptides were resuspended in molecular grade water (1 mg/mL) and ELISA plates (MaxiSorp; NUNC, Roskilde, Denmark) were coated (1 μg/mL) overnight at 4°C. Plates were washed three times with TBST and blocked with 10% horse serum in TBST for 1 h at room temperature with agitation. The ELISA was performed as previously described (14).

Ehrlichia canis (Jake strain) was propagated in DH82 cells (canine macrophage-like cells) with minimal essential medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (HyClone, Logan, UT, USA), 1% HEPES (Sigma Chemical Co., St. Louis, MO, USA), 1% sodium pyruvate (Sigma), and 1% nonessential amino acids (Sigma) at 37°C in a humidified 5% CO₂ atmosphere. ISE6,which is a tick Ixodes scapularis embryo-derived cell line, was obtained from Dr. Ulrike Munderloh (University of Minnesota) (23) and maintained in L15B300 medium supplemented with 10% fetal bovine serum (GeminiBio, Sacramento, CA, USA), 10% tryptose phosphate broth (BD, Sparks, MD, USA) and 1% bovine lipoprotein cholesterol concentrate (MP Biomedicals, Irvine, CA, USA) at 34°C as previously described (24). ISE6 cells were infected [multiplicity of infection (MOI) = 10] with host cell-free E. canis derived from infected DH82 cells.

Ehrlichia canis antigen for Western blot analysis was prepared as described previously (17). Briefly, infected cells (DH82 and ISE6) were collected when morulae were observed in all cells after Diff-Quik staining. Cells were then centrifuged at (5,000 × g for 15 min) and resuspended in phosphate-buffered saline (PBS). Cells were ruptured by sonication, twice (40 Hz) for 10 s, and large cell debris was pelleted by centrifugation (1,500 × g for 10 min) at 4°C. The supernatant was centrifuged (10,000 × g for 15 min) at 4°C to collect cell-free ehrlichiae. The pellet was washed in PBS, centrifuged (10,000 × g for 15 min) at 4°C, and resuspended in PBS. BCA protein assay (Pierce Biotechnology, Rockford, IL, USA) was performed to determine the protein concentration. Cell lysate prepared from the uninfected cells was used as a negative control.

Polyclonal rabbit antibodies were commercially generated by immunizing rabbits with peptides derived from respective E. canis immunoreactive proteins (GenScript). The antisera were used for Western blot analysis and immunofluorescence microscopy.

Purified E. canis antigen was solubilized in LDS sample buffer containing the reducing agent dithiothreitol (Invitrogen), heated at 70°C for 10 min, and separated by sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) in 3-N-morpholinopropanesulfonic acid (MOPS) running buffer under reducing conditions with 4 to 20% gradient Bis-Tris acrylamide gels (NuPAGE; Invitrogen). Proteins were transferred to a nitrocellulose membrane (Protran BA85, 0.45-μm pore size; Whatman, Florham Park, NJ, USA) using a semidry transfer apparatus (Bio-Rad, Hercules, CA, USA). Membranes were blocked for 1 h with TBST blocking buffer (5% nonfat milk). E. canis antisera were diluted (1:100) in the blocking buffer and incubated for 1 h at room temperature with shaking. Membranes were washed 3 times with TBST, 5 min each and an affinity-purified alkaline phosphatase-labeled goat anti-rabbit IgG (H & L) (KPL) secondary antibody (1:5000) in blocking buffer was applied and incubated for 1 h. After washing, 5-bromo-4-chloro-3-indolyl-phosphate and nitroblue tetrazolium (BCIP-NBT) substrate (KPL) was applied to visualize bound antibody. Densitometry was performed for Western blot bands of each protein using ImageJ software and the fold-changes of E. canis proteins were determined relative to cell actin.

Antigen slides were prepared from DH82 or ISE6 cells infected with E. canis. Infected cells were applied to 12-well Teflon coated slides, air dried and fixed with 4% paraformaldehyde (PFA) for 20 min at room temperature and washed twice in PBS. Cells were permeabilized with 0.2% Triton X-100 in PBS (BSA) for 15 min and washed. Image-It Signal Enhancer (Invitrogen) was applied to the cells for 30 min, followed by BlockAid (Invitrogen) for 30 min. Cells were then incubated with antigen-specific antisera (1:100) in blocking buffer for 1.5 h, washed three times with PBST, and incubated with goat anti-rabbit IgG (H + L) Alexa Fluorophore Plus 488 secondary antibody (1:200) for 30 min in the dark. The slides were washed thrice and mounted with ProLong Glass Antifade Reagent (Invitrogen). Immunofluorescence images were captured with an Olympus BX61 epifluorescence microscope and analyzed using Slidebook software (v.5.0; Intelligent Imaging Innovations, Denver, CO, USA).

DH82 and ISE6 cells were cultured in T-25 flasks (Cellstar) to 90–95% confluency and incubated with cell-free E. canis at a multiplicity of infection (MOI) of 10. Samples were collected daily for DH82 (5 days) and ISE6 cells (8 days), and the absolute E. canis dsb copy number was determined by real-time qPCR and plotted against the standard curve, as previously described (25). Briefly, cells were washed with PBS and lysed in SideStep Lysis and Stabilization Buffer (Agilent Technologies, Santa Clara, CA, USA), and real-time quantitative PCR (qPCR) amplification was performed using Brilliant II SYBR Green Mastermix (Agilent), forward primer (5-GCTGCTCCACCAATAAATGTATCCCT-3), and reverse primer (5-GTTTCATTAGCCAAGAATTCCGACACT-3), using a CFX96 Touch Real Time PCR System (BioRad).

Immunodominant E. canis proteins (n = 18) were recently identified by our laboratory in addition to those (i.e., TRPs, Anks, OMP-1) that have been previously reported (9, 11–13). In this study, we selected seven proteins that exhibited the strongest immunoreactivity for further evaluation and compared them to well defined immunoreactive proteins including, TRP19, TRP36 and TRP140. Consistent with our earlier findings, TRP19, TRP36 and TRP140 reacted with 15 naturally infected CME dog sera. By ELISA, TRP36 and two newly identified proteins (Ecaj_0919 and Ecaj_0126) reacted strongly with the dog sera (mean OD₆₅₀ > 2.0), and 5 others (Ecaj_0717, Ecaj_0920, Ecaj_0636, Ecaj_0073 and Ecaj_0151) were slightly lower and similar to TRP140 and TRP19 (mean OD₆₅₀ > 1.7). The sera from naturally infected dogs did not react with cell-free expressed negative control protein (OD₆₅₀ < 0.1) and the cell-free expressed E. canis proteins (including TRPs) did not react with sera from uninfected dog. Immunoreactivity of these proteins was determined by ELISA and ranked according to mean OD₆₅₀ (Figure 1A).

To understand the potential differences in diagnostic sensitivity for detection of antibodies generated early in infection, the E. canis proteins were examined using convalescent E. canis dog sera collected from an experimentally infected dog on days 0, 7, 14, 21, 28, 35, 42, and 56. Dog sera obtained prior to infection (day 0) was used as a negative control. As previously reported (15), TRP36 and TRP140 were most sensitive as both proteins reacted with antibodies in dog sera as early as day 14 (Figure 1B). TRP19 and Ecaj_0919 and Ecaj_0151 reacted with antibodies on day 21. All other immunodominant proteins reacted with antibodies at day 28 except Ecaj_0073, which reacted with antibody at day 35.

We investigated the genetic diversity of the E. canis. Proteins (including TRPs) using a BLAST sequence analysis. Protein sequences derived from E. canis Jake strain (USA) and E. canis YZ-1 (China), and orthologous sequences from E. minasensis are shown in Table 1. E. minasensis is a relatively new Ehrlichia species that is the closest relative to E. canis (26). Four of the 7 newly identified proteins in E. canis Jake strain were identical to the homologs in the E. canis YZ-1 strain, while three proteins had minor differences in the amino acid sequence length and percent identity. TRP19 was very conserved and TRP36 and TRP140 were less conserved as previously reported (18). Compared to orthologs in E. minasensis, the E. canis TRPs varied significantly in amino acid identity (53–70%). However, E. minasensis orthologs of two newly discovered proteins (Ecaj_0073 and Ecaj_0151) were identical in length and had 88 and 94% sequence identity, respectively with E. canis. Orthologs of five remaining newly identified E. canis proteins had variations in sequence length and their amino acid identity was lower (ranging 37–80%) between strains. These data indicate that many of the recently discovered antibody reactive proteins are highly conserved among E. canis strains, but diverse in the closest E. canis relative, E. minasensis.

Our recent study defining the E. canis immunome revealed a highly immunoreactive E. canis protein (Ecaj_0126) (22). In this investigation, we determined that Ecaj_0126 contained a C-terminal TR domain (Figure 2A). Ecaj_0126 has a predicted mass of ~70 KDa (671 amino acids) and contains three nearly identical 35 amino acid repeats and a fourth partial repeat (16 amino acids). To identify linear antibody epitopes in Ecaj_0126, we screened overlapping peptides (25 amino acids; 6 amino acid overlap) and the 3 full length repeats (35 amino acids) from the TR region (Table 2). Multiple strongly immunoreactive peptides were identified with Ecaj_0126 indicating the presence of linear antibody epitopes (Figure 2B). Further investigations of these immunoreactive epitopes with additional CME dog sera (n = 10) consistently found strong antibody reactivity with the TR peptides. We then screened the Ecaj_0126 R2 peptide with 8 CME dog sera collected from different North and South American countries (USA, Colombia, and Brazil). We observed consistent antibody reactivity with all dog sera, indicating that the TR epitope is conserved among geographically dispersed E. canis strains (Figure 2C).

By real time qPCR, E. canis replication increased more rapidly in DH82 cells, peaking at day 4 compared to ISE6 cells, which exhibited a slower (peak at day 7) more linear growth curve (Figure 3A). To examine differences in the protein expression levels of the E. canis TRPs and newly identified immunoreactive proteins, IFA and Western blot analyses were performed using E. canis-infected DH82 cells and ISE6 cells (Figure 3B). After determining the growth curve in both cell lines, we prepared E. canis antigen slides using DH82 cells (day 4) and ISE6 cells (day 7). By immunoblot, four proteins (TRP140, Ecaj_0126, 0920 and 0073) exhibited higher overall expression levels in ISE6 cells compared to DH82. In contrast, TRP19, TRP36, Ecaj_0636 and 0919 exhibited higher overall expression in DH82 cells. IFA images were consistent with Western immunoblot densitometry on days 4 (DH82) and 7 (ISE6) post infection.