Ehrlichia chaffeensis Transcriptome in Mammalian and Arthropod Hosts Reveals Differential Gene Expression and Post Transcriptional Regulation

Background Human monocytotropic ehrlichiosis is an emerging life-threatening zoonosis caused by obligately intracellular bacterium, Ehrlichia chaffeensis. E. chaffeensis is transmitted by the lone star tick, Amblyomma americanum, and replicates in mononuclear phagocytes in mammalian hosts. Differences in the E. chaffeensis transcriptome in mammalian and arthropod hosts are unknown. Thus, we determined host-specific E. chaffeensis gene expression in human monocyte (THP-1) and in Amblyomma and Ixodes tick cell lines (AAE2 and ISE6) using a whole genome microarray. Methodology/Principal Findings The majority (∼80%) of E. chaffeensis genes were expressed during infection in human and tick cells. There were few differences observed in E. chaffeensis gene expression between the vector Amblyomma and non-vector Ixodes tick cells, but extensive host-specific and differential gene expression profiles were detected between human and tick cells, including higher transcriptional activity in tick cells and identification of gene subsets that were differentially expressed in the two hosts. Differentially and host-specifically expressed ehrlichial genes encoded major immunoreactive tandem repeat proteins (TRP), the outer membrane protein (OMP-1) family, and hypothetical proteins that were 30–80 amino acids in length. Consistent with previous observations, high expression of p28 and OMP-1B genes was detected in human and tick cells, respectively. Notably, E. chaffeensis genes encoding TRP32 and TRP47 were highly upregulated in the human monocytes and expressed as proteins; however, although TRP transcripts were expressed in tick cells, the proteins were not detected in whole cell lysates demonstrating that TRP expression was post transcriptionally regulated. Conclusions/Significance Ehrlichia gene expression is highly active in tick cells, and differential gene expression among a wide variety of host-pathogen associated genes occurs. Furthermore, we demonstrate that genes associated with host-pathogen interactions are differentially expressed and regulated by post transcriptional mechanisms.


Introduction
Human monocytotropic ehrlichiosis (HME) is a life-threatening emerging tick-borne zoonosis caused by obligately intracellular bacterium, Ehrlichia chaffeensis [1]. HME is a systemic disease characterized by clinical presentation that includes fever, headache, myalgia, anorexia, chills and laboratory abnormalities including leucopenia, thrombocytopenia, anemia and elevation of serum hepatic aminotransferases [1]. The severity of the disease varies from asymptomatic seroconversion to a fatal multisystem failure [2]. E. chaffeensis is transmitted by the lone star tick, Amblyomma americanum, and maintained in nature by persistent infection of mammalian hosts [1]. In the mammalian host, E. chaffeensis replicates primarily within mononuclear phagocytes forming membrane-bound cytoplasmic microcolonies called morulae that are resistant to innate immune destruction [3].
Bacterial pathogens survive by expressing genes necessary for transmission, invasion and persistence, and evasion of innate and adaptive defenses [4]. Among these include surface proteins of Borrelia burgdorferi and Yersinia pestis, secreted effectors of Shigella flexneri and transcriptional regulator of Bordetella pertussis [5][6][7]. Moreover, host-specific gene expression by Anaplasma phagocytophilum has been reported in human and tick cells [8], and the E. chaffeensis p28 outer membrane protein encoded by the OMP-1 multigene locus is differentially expressed in human and tick cells [9][10][11]. Furthermore, it is recognized that E. chaffeensis propagated in tick cells has a distinct antigen expression profile from that of mammalian phagocyte grown ehrlichiae [12].
Molecular interactions between TRP47 and the mammalian host identified numerous host cell targets with distinct cellular functions associated with signaling, transcriptional regulation, vesicle trafficking and cellular proliferation and differentiation [21]. TRP120 has been shown to play an important role in binding and internalization [22], and its expression is regulated by the second messenger cyclic di-GMP and protease HtrA [23]. It is also associated with novel molecular protein-protein, protein-DNA interactions suggesting that it is involved in modulating host cell processes and gene transcription [24,25]. E. chaffeensis Ank200 was recently detected in the mammalian host cell nuclei and interacts with an adenine-rich motif in promoter and Alu elements [26].
The macrophage transcriptome during E. chaffeensis infection has been previously determined [27]; however, investigation of E. chaffeensis gene expression in distinct hosts has been limited to genes encoding the OMP-1 multigene family. In this study, we analyzed the E. chaffeensis transcriptome in human monocytes (THP-1), tick cells from the known arthropod vector (A. americanum; AAE2 cells) and the vector of A. phagocytophilum (Ixodes scapularis; ISE6 cells) and determined that well characterized ehrlichial proteins involved in host-pathogen interactions were differentially expressed.

E. chaffeensis genes upregulated in the human monocytes
There were 50 E. chaffeensis genes upregulated (.2 fold; p,0.05) in the THP-1 cells compared to both AAE2 and ISE6 cells (Table 1), and 19 additional genes upregulated in THP-1 compared to AAE2 cells. In contrast, only five additional genes were upregulated in THP-1 compared to ISE6 cells. When classifying the genes according to the Clusters of Orthologous Groups (COGs) [28], the E. chaffeensis genes upregulated in the THP-1 cells were grouped into the metabolic and cellular process (C, G, P, Q, D); transcription, translation and DNA repair (J, K, L); cell envelope biogenesis and outer membrane (M); posttranslational modifications (O); general function predicted or unknown (R,S); trafficking and secretion (U) (Fig. 2B, black bars). The majority of these genes (54%) were classified as hypothetical with unknown functions.

E. chaffeensis genes upregulated in the tick cells
There were 193 E. chaffeensis genes upregulated (.2 fold; p,0.05) in tick cells compared to human cells (Table S1). The largest proportion (32%) belonged to the COG with general function predicted or unknown (R, S), 30% were involved in metabolism and cellular process, and 7% of the genes were associated with translation (J) (Fig. 2B, grey bars). The remaining E. chaffeensis genes (31%) were distributed in the other COGs.

Hyper-expressed genes in the human monocytes
There were ten E. chaffeensis genes expressed in the THP-1 cells with expression levels 10-15 times higher (hyper-expressed) than other genes identified as highly expressed. These genes included TRP47 (the highest expressed gene), TRP32, ribosomal proteins, malonyl CoA-acyl carrier protein transacylase, and hypothetical proteins (ECH_0166, ECH_1059, ECH_0570, ECH_0253).
E. chaffeensis genes associated with protein trafficking and secretion were expressed in tick and human cells; however, several of these genes were upregulated in the tick cells compared to human cells, including SecF, TatC, TatA and members of the type IV secretion systems. Additionally, in the tick cells, several of the genes associated with posttranslational modification, and protein turnover were upregulated including several ATP-dependent proteases and chaperones.

Expression of hypothetical genes
A large percentage (42%) of annotated E. chaffeensis genes encode hypothetical proteins with unknown functions [13]. In this study, we determined that most of these genes were differentially expressed in human and tick cell lines. There were 27 hypothetical genes (COG; R, S) that were highly expressed in the THP-1 cells, including TRP32 and TRP47 (Table 1). However, 11 of these 27 genes were not expressed by E. chaffeensis in the AAE2 and ISE6 cells ( Table 2). Most of these genes (9/11) encoded peptides (30-70 amino acids) that do not have orthologs. Host-specific expression of these genes suggested that they are required exclusively for adaptation and survival within the mammalian host.
There were a larger number of E. chaffeensis genes that were differentially expressed in the tick cells, including 63 genes categorized as hypothetical (COG; R, S). Some of these genes (n = 18) were expressed only in the tick cells (Table 3). Of these genes, ECH_0114, ECH_0249, ECH_0258, ECH_0889, ECH_1030, ECH_1048 were highly expressed (expression values 3-5 times greater than the average expression value). Of the 18 genes differentially expressed in the tick cells, seven did not have orthologs and six were peptides (30-80 aa). ECH_0114 was predicted to be a secreted protein, and ECH_0526 and ECH_1038 were predicted as outer membrane proteins (CELLO subcellular localization predictor) [30]. In a previous study, ECH_0526 protein expression was detected in both AAE2 and ISE6 tick cell lines [31].

Expression of tandem repeat proteins
Expression of three tandem repeat proteins; TRP32, TRP47 and TRP120 were analyzed in AAE2, ISE6 and THP-1 cells. TRP120 transcript was detected in all three cell lines, and the protein was also expressed in human monocytes and tick cells (Fig. 4). Interestingly, transcripts for hyper-expressed genes TRP32 and TRP47 were detected in tick cells by microarray and qRT-  Metabolism/Cellular Processes PCR; however, these proteins were not detected by western immunoblot in E. chaffeensis-infected AAE2 and ISE6 cell lysates (Fig. 4).

Discussion
Defining the dynamic changes in pathogen gene and protein expression that occurs in infected hosts is essential to understanding pathobiology and having a rational basis for vaccine development. This investigation was conducted because of our lack of knowledge regarding the relative E. chaffeensis gene expression in mammalian and arthropod hosts, which is a major impediment to understanding which genes are essential for ehrlichial adaptation. In this study, we demonstrated that the expression of many E. chaffeensis genes was influenced by the host environment. In addition, we examined E. chaffeensis gene expression in tick cell lines from the established vector species and another common tick that is not a vector. Significant differences in E. chaffeensis gene expression were not observed between the two tick cell lines, and similar expression patterns were observed in E. chaffeensis genes involved in metabolic and cellular processes between human and tick cells. Differentially expressed genes identified were primarily hypothetical genes and genes associated with translation and posttranslational modification. Furthermore, we have also found evidence of posttranscriptional regulation of select ehrlichial genes involved in host-pathogen interactions in mammalian and arthropod hosts.
In mammalian cells, the Ehrlichia developmental cycle occurs within 72 hrs and is characterized by entry of the dense-cored form, replication of the reticulate cells and transformation to infectious dense-cored ehrlichiae [32]. In this study, enriched bacterial RNA was extracted from E. chaffeensis infected THP-1, AAE2 and ISE6 cells when 90% of the cells were infected. Although the infection was not synchronized to evaluate a specific phase of the developmental cycle, cells were harvested when 90% of the cells had E. chaffeensis morulae within their cytoplasm; therefore the data presented in this study likely includes genes expressed during all phases of development, but may be more representative of ehrlichial transcription during the later developmental stages that is dominated by dense-cored ehrlichiae [32].
The I. scapularis cell line, ISE6, has been routinely used in previous investigations involving arthropod-borne pathogens including Ehrlichia, Rickettsia, Anaplasma and Borrelia [33]; however, I. scapularis is not a natural vector of E. chaffeensis. Recently, Munderloh et. al, developed the A. americanum, AAE2 cell line from tick embryos (Munderloh, U. and Davidson, W.R., unpublished data), and E. chaffeensis protein expression in AAE2 tick cells has been investigated [31], but a comprehensive analysis of E. chaffeensis gene expression has not been determined in the AAE2 cell line. Therefore, we investigated gene expression in both cell lines to determine if significant differences existed. Notably, we did not identify any significant differences in E. chaffeensis gene expression; hence, it appears that either cell line could be used for vector-pathogen studies for E. chaffeensis. However, the AAE2 cells grew more rapidly and appeared to support more robust growth of ehrlichiae, which are also important considerations. Transcripts were detected for ,80% of the E. chaffeensis genes in human monocytes and tick cells. This level of transcript detection is slightly higher (,10%) than that previously reported for A. phagocytophilum in human (HL-60) and tick (ISE6) cells [8]. This difference may be related to the fact that we utilized pathogenenriched RNA rather than total RNA. Most E. chaffeensis genes (81-95%) involved in metabolic and cellular process, transcription, translation, DNA repair, cell envelope biogenesis, outer membrane proteins, posttranslational modifications, general function predicted or unknown, trafficking and secretion were expressed in all three cell lines. There were 77 genes for which transcripts were not detected in any of the three cell lines, and the function of the majority of these genes is unknown. It is possible that these genes are required during stages of infection not depicted in this study such as transmission of the pathogen from one host to the other, reactivation of the pathogen after a blood meal in the tick, or in the presence of tick saliva.
Although similar numbers of genes were expressed by E. chaffeensis in the human and tick cells, the most striking discovery is that 38% of the E. chaffeensis genes were differentially expressed. When compared to human cells, E. chaffeensis was transcriptionally more active in the tick cells, and there were a larger number of genes with high expression levels in the tick cells. The functions of these genes were associated with protein modification, energy production and conversion and nutrient transport. Similar genes were upregulated in Rickettsia conorii and R. rickettsii under conditions of nutrient limitations and lower temperatures, when the metabolism of the host cells slows [34,35]. In contrast, the majority of the genes had a moderate expression level in human cells. Additionally, there were several genes involved in metabolism, cellular process, and translation that were upregulated in tick cells. The upregulation of these genes in the arthropod host suggests that Ehrlichia has higher metabolic activity in the tick. The number of genes differentially expressed by E. chaffeensis between the two host cells was similar to that observed for A. phagocytophilum [8]. However, only minimal differences were observed in gene expression when R. rickettsii grown in ISE6 was compared to R. rickettsii grown in Vero cells [35]. Although all three organisms have evolved to adapt within both arthropod and mammalian cells, there appears to be significant differences between Rickettsia compared to Ehrlichia and Anaplasma suggesting that they have different adaptation mechanisms and pathobiology.
In contrast to the total expression level (,80%) of genes from the other functional groups, the 437 E. chaffeensis genes with unknown function were expressed at a lower level (,67%) in each cell line and the majority were differentially expressed. We identified 11 genes highly expressed in the human cells that were not expressed in the tick cells and identified 18 genes exclusively expressed in the tick cells, and the majority of these hypothetical genes encoded peptides, 30-80 amino acids in length. Additionally, these peptides are E. chaffeensis-specific and do not have orthologs in other ehrlichial species. Nearly half of the genes (n = 243) annotated as hypothetical proteins in the E. chaffeensis genome contain fewer than 100 amino acids, and recently peptides were detected for 66% of these proteins during infection in THP-1 cells [36]. Host-induced expression of these E. chaffeensis peptides suggests that they are required exclusively for adaptation and survival within the mammalian host. Further studies are needed to characterize these genes and their role in host-specific adaption and survival.
In human cells, there were ten ehrlichial genes that were hyperexpressed, defined as genes with expression values 10-15 times greater than other highly expressed genes. Among these genes were two major immunoreactive tandem repeat proteins, TRP47 and TRP32. TRP47, the most highly expressed E. chaffeensis gene in human cells, contains seven 19-mer tandem repeats that dominate the C-terminal region and several N-terminal tyrosine phosphorylation sites [21,37]. A recent study to examine molecular interactions between TRP47 and the host identified several interactions with specific host cell proteins that have distinct cellular functions associated with signalling, transcriptional regulation, vesicle trafficking, and cellular proliferation and differentiation [21]. The hyper-expression of TRP47 in human cells, the absence of TRP47 in tick cells, and our recent findings regarding molecular host-pathogen interactions, suggests that TRP47 is a multifunctional effector that is required for ehrlichial intracellular survival within the mammalian host. Unlike TRP47 which is differentially expressed by the dense-cored form of E. chaffeensis, TRP32 is extracellularly associated with the morular fibrillar matrix and the morula membrane, indicating that this protein is secreted. TRP32 does not have homology with other known proteins [18]; however, we have recently demonstrated that TRP32 interacts with proteins with functions similar to those that interact with TRP47 [38]. In the tick cells, although transcripts were detected for the TRP47 and TRP32, the proteins were not detected suggesting that they are regulated posttranslationally. E. chaffeensis TRP120 is a well characterized protein that is differentially expressed on the surface of the dense-cored E. chaffeensis. Similar levels of TRP120 transcripts were detected in human and tick cells, and the protein was detected in both cell lysates. Our findings regarding TRP120 expression were in contrast to a previous study that examined macrophage-and tick cell-derived proteins of E. chaffeensis, in which TRP120 was not detected in macrophages, but was detected in tick cell lysates [31]. However, numerous other studies have reported TRP120 expression in ehrlichiae cultivated in human cells [20,22,23]. TRP120 has known functional properties including binding and internalization, and its surface expression is regulated by second messenger cyclic di-GMP and interacts with host cell proteins associated with biological processes similar to TRP47 [22][23][24]. Furthermore, we recently, demonstrated that TRP120 binds host cell DNA and targets genes associated with biological processes known to be altered during E. chaffeensis infection [25]. Although TRP120 has important functions in the mammalian host related to pathobiology, the role of TRP120 in the arthropod host is unknown. The expression of TRP120 in the tick cells suggests that it may have similar functions in the arthropod host.
The OMP-1/P28 multigene family of E. chaffeensis have been well studied and host cell-specific expression of these genes has been previously reported [10,11]. The function of these immunoreactive outer membrane proteins has been associated with immune evasion; however, Rikihisa et. al. recently demonstrated porin activity for OMP-1F and P28 [16,39,40], suggesting an important functional role in nutrient acquisition. Consistent with previous studies, we determined that Omp-1B and p28 were expressed in human and tick cells. The upregulation of p28 (p28- 19) in human cells and the high expression of OMP-1B (p28-14) in tick cells were also consistent with previous in vitro studies [11,12], and expression of OMP-1B (p28-14) transcript has been detected in all three developmental stages of the tick vector, A. americanum [11]. However, our finding that OMP-1N (p28-1) was upregulated in tick cells has not been previously reported. Transcripts were not detected for OMP-1H (p28-11) and OMP-1W (p28-7) in the human cells. The absence of OMP-1W (p28-7) expression was also consistent with the fact that it could not be detected in dogs experimentally infected with E. chaffeensis [11]. However, OMP-1H (p28-11) was detected in experimentally infected dogs and DH82 cells (canine cell line) [11,41], but not in the human and tick cells suggesting that there are other host factors that contribute to expression of OMP-1H. Similarly, although transcripts for OMP-1 family members have been detected in several studies, OMP-1B is the only OMP-1 paralogue detected by proteomics in E. chaffeensis  cultured in ISE6 cells [31]. The role of host-specific OMP expression is not clear, but our findings suggest that vaccines targeting ehrlichial OMPs expressed in the tick should include OMP-1N. Host cell specific expression of these genes could be related to adaptation to different host environments and for nutrient acquisition.
It is generally recognized that regulation of bacterial gene expression is controlled by transcriptional and posttranscriptional mechanisms [42][43][44]. Several recent studies have investigated the mechanisms involved in the mRNA and protein stability and translational regulation in prokaryotes, and their dependence on environmental conditions and growth phase, especially with virulence factors [45,46]. Bacterial protein expression is not only dependent on levels of mRNAs but also on other RNA species. Regulatory RNAs such as small RNAs (sRNA) controlling virulence and pathogenesis have been demonstrated in other Gram-negative bacteria including Escherichia coli, Pseudomonas aeruginosa, Salmonella typhimurium, and Chlamydia trachomatis [47,48]. In this investigation, transcripts were detected for TRP47 and TRP32 in the tick cells, yet the proteins were not detected suggesting that their expression is, in part, controlled by posttranslational mechanisms in response to host cell environments, potentially by regulatory RNAs. Similarly, OMP-1B (P28- 14) transcripts have been routinely detected in mammalian cells, yet the protein has not been detected in numerous proteomic studies [9,11,31,49]. Therefore, there is evidence that posttranslational mechanisms are involved in TRP expression and could also be involved in regulating OMP expression. Further investigation of posttranslational regulation mechanisms in Ehrlichia survival in mammalian and arthropod hosts is needed to understand how ehrlichial protein expression is regulated and its role in host adaptation.
Understanding the molecular survival strategies within the distinct hosts and the mechanisms involved in host adaptation will lead to novel prophylactic and therapeutic targets to prevent transmission and infection. We determined that some TRPs, OMPs, and hypothetical proteins are differentially expressed, and thus, appear to be important for adaptation to each host. Additionally, the hyper-expression of the TRP32 and TRP47 genes in the human cells and absence of the expression of these proteins in the tick cells demonstrate their significance in the mammalian host. The subset of E. chaffeensis hypothetical genes identified exclusively in each of the host cells in this study should be examined and their functions determined.

RNA extraction
Total RNA was purified from uninfected and E. chaffeensisinfected (90% infected) THP-1, AAE2 and ISE6 cells (10 7 cells per sample) using Tri reagent (Ambion, Austin, TX). Genomic DNA was eliminated by treatment with Turbo DNA-free (Ambion) according to the manufacturer's protocol. Polyadenylated host mRNA was removed using oligo (dT) columns (Oligotex, Qiagen, Valencia, CA) as previously described [52], and bacterial RNA was enriched using Terminator (Epicenter Biotechnologies, Madison, WI) that selectively digests RNAs with 59-monophosphates that are present only on ribosomal RNA. RNA concentration was determined by NanoDrop (Thermo Scientific, Wilmington, DE) and quality confirmed by Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA) at the UTMB Genomics Core Facility. The RNA quality requirements were established as: A260/A280 $1.8, A260/A230 $1.8 and concentrations $ 1 mg/ul.

Microarray design
The genome sequence of E. chaffeensis (GenBank accession no. CP000236) [13] was submitted to Roche NimbleGen Systems (Madison, WI) for custom 4-plex microarray design. The arrays were manufactured using maskless, digital micromirror technology. Five replicates of the genome were included per chip, with an average of 12 different 60-base oligonucleotides (60-mer probes) representing each open reading frame in the E. chaffeensis genome. Unique probes were designed for 1031 of the 1158 ORFs (1.18 Mbp genome). Three biological replicates were included for E. chaffeensis cultivated in each cell line (THP-1, AAE2 and ISE6). Additionally, for each cell line, RNA was extracted from uninfected cells (negative controls) and was processed similarly, and these samples were used to establish background subtraction thresholds.

Hybridization
Enriched E. chaffeensis RNA (10 mg) was used for cDNA synthesis using random hexamer primers and the SuperScript Double-Stranded cDNA Synthesis Kit (Invitrogen) according to the NimbleGen Arrays User's Guide (Gene Expression Analysis v3.2). Labeling and hybridization was done at the MD Anderson Cancer Center, Genomics Facility (Houston, TX). Briefly, doublestranded cDNA was random-prime labeled with Cy3-nonamers and hybridized to the microarray for 16 hrs at 42uC. The arrays were washed, dried and scanned using a GenePix 400B microarray scanner (Molecular Devices, Sunnyvale, CA).

Microarray data analysis
Data were extracted from the scanned array images using NimbleScan software (Roche NimbleGen). Quantile normalization was performed across replicates within the 4-plex arrays, and RMA (Robust Multichip Average) analysis was performed to generate gene expression values [53]. The genes expressed were determined by subtracting expression values obtained from uninfected cells from those of infected cells from the same cell line. Analysis and visualization of the expression data were performed using ArrayStar4 software (DNASTAR Inc., Madison, WI), using mean log 2 expression values for the three biological replicates for each cell line. F-test (ANOVA) was used to compare the mean gene expression values for replicates (within same cell line) and groups of replicates (between cell lines) for a given gene. The microarray data generated in this study have been deposited in NCBI's Gene expression Omnibus [54]. The data are accessible through GEO series accession number GSE29109 (http://www. ncbi.nlm.nih.gov/geo/query/acc.cgi?acc = GSE29109).

Real time quantitative PCR
Real time PCR of selected E. chaffeensis genes was performed with gene specific primers designed using Lasergene 8 (DNAS-TAR) ( Table 4). RNA (1 mg) was used as template for cDNA synthesis using iScript cDNA synthesis kit (Bio-Rad Laboratories, Hercules, CA) according to the manufacturer's instructions. qPCR was performed using iQ SYBR Green supermix (Bio-Rad), genespecific primers and thermal cycling protocol that consisted of an initial denaturation step of 95uC for 2 min, and 40 cycles of 95u for 10 s, 55uC for 30 s, and 65uC for 30 s using a Mastercycler EP Realplex 2 S (Eppendorf). DNA from infected cells was used as positive control. Samples lacking cDNA and cDNA from uninfected cells were used as negative control. qPCR data were converted by subtracting the Ct value from the number of cycles (40 cycles) to obtain values.