Enlisting the Ixodes scapularis Embryonic ISE6 Cell Line to Investigate the Neuronal Basis of Tick—Pathogen Interactions

Neuropeptides are small signaling molecules expressed in the tick central nervous system, i.e., the synganglion. The neuronal-like Ixodes scapularis embryonic cell line, ISE6, is an effective tool frequently used for examining tick–pathogen interactions. We detected 37 neuropeptide transcripts in the I. scapularis ISE6 cell line using in silico methods, and six of these neuropeptide genes were used for experimental validation. Among these six neuropeptide genes, the tachykinin-related peptide (TRP) of ISE6 cells varied in transcript expression depending on the infection strain of the tick-borne pathogen, Anaplasma phagocytophilum. The immunocytochemistry of TRP revealed cytoplasmic expression in a prominent ISE6 cell subpopulation. The presence of TRP was also confirmed in A. phagocytophilum-infected ISE6 cells. The in situ hybridization and immunohistochemistry of TRP of I. scapularis synganglion revealed expression in distinct neuronal cells. In addition, TRP immunoreaction was detected in axons exiting the synganglion via peripheral nerves as well as in hemal nerve-associated lateral segmental organs. The characterization of a complete Ixodes neuropeptidome in ISE6 cells may serve as an effective in vitro tool to study how tick-borne pathogens interact with synganglion components that are vital to tick physiology. Therefore, our current study is a potential stepping stone for in vivo experiments to further examine the neuronal basis of tick–pathogen interactions.


Introduction
The North American black-legged deer tick Ixodes scapularis and the European castor-bean tick Ixodes ricinus are both medically important arthropod vectors. These two allopatric tick species are well recognized for transmitting a wide spectra of bacterial, viral, and protozoan pathogens [1,2]. Both tick species infect hosts with the bacterium Anaplasma phagocytophilum, consists of approximately 100 neuropeptides encoded by at least 34 genes that correspond to invertebrate neuropeptide orthologs [37,43]. Tick neuropeptides are identified with immunohistochemistry, proteomics, and genomics [37,38,44,45], but their function is almost exclusively implicated by discovering their transport pathways and/or the localization of their cognate receptors in target organ(s) [37,46,47]. Tachykinin-related peptides (TRP) constitute a large family of pleiotropic neuropeptides found across bilaterians [48]. On the one hand, vertebrate TRPs play important roles in pain, inflammation, sensory processes, immune systems, gut function, or hormonal regulation [48]. On the other hand, invertebrate TRPs are known for regulating the central nervous system and gut function, while olfactory processing, locomotion, food seeking, vasodilatation, nociception, and metabolic stress are also described [48].
For our current study, we confirmed the presence of an entire Ixodes neuropeptidome in ISE6 cells that led to several questions. Are neuropeptide-expressing ISE6 cells peptidergic neurons? Are Ixodes TRPs expressed in vivo (i.e., I. scapularis synganglion)? How are Ixodes TRPs respectively distributed in vitro versus in vivo? Given that A. phagocytophilum induces transcriptional reprograming in ticks [24][25][26], will infection affect Ixodes trp gene expression? The cumulative answers to these questions propose that ISE6 cells may serve as an effective in vitro tool for studying the nature of tick neuropeptidergic cells and their interactions with various TBPs.

Neuropeptidome of ISE6 Cells
Extensive BLAST searches in the ISE6 genome predict 38 distinct genome-based neuropeptide genes and their representative genomic scaffolds ( Table 1). The presence of 37 neuropeptide transcripts was also confirmed in the ISE6 Sequence Read Archive (SRA) databases (BioProject: PRJNA239331). The transcript for natalisin was the only neuropeptide not detected in our in silico searches (Table 1). We also failed to experimentally amplify the natalisin transcript using three different sets of primers (Supplementary Table S1). Then, six of the SRA-confirmed neuropeptides (i.e., sulfakinin, kinin, CCHamide, short neuropeptide F, and FGLamide-related allatostatin) were selected for qRT-PCR validation using RNA extracted from ISE6 cells (Supplementary Figure S1).

Structure of the Gene-Encoding Ixodes TRP
The ISE6 genome BLAST yielded a predicted transcript (VectorBase accession number ISCI008383) encoding a putative TRP. The I. ricinus TRP transcript (GenBank accession number MW082607) was molecularly identified in this study. Nucleotide and protein alignments of the I. scapularis TRP (ISCI008383) with I. scapularis TRP EST (EL516783) and I. ricinus TRP (MW082607) reveal an incorrect computational prediction of ISCI008383. Specifically, the 5 -end of ISCI008383 is incomplete with an incorrect translated putative signal peptide. The 3 -end reading frame is also shifted, causing an improper conceptual TRP translation (Supplementary Figure S2). A BLAST using the I. scapularis EST (EL516783) encoding TRP, against the ISE6 genome confirms a relationship to the PKSA02005591.1 scaffold, resulting in a better representation of the I. scapularis TRP genomic organization ( Figure 1A). In ISE6 cells genome, the I. scapularis trp gene ( Figure 1A) is composed of four exons (≈140, 87, 328, and 170 bp) interrupted by three introns (189857, 5018, and 1862 bp). The I. scapularis trp ORF is 492 bp, spanning exons 2-4 ( Figure 1A). Varroa destructor [49] and the fruit fly Drosophila melanogaster TRPs ( Figure 1C). I. scapularis, I. ricinus, R. sanguineus, R. microplus and D. silvarum share identical TRP1 and TRP2 amino acid sequence ( Figure 1C).  The translated TRP ORF of I. scapularis (169 residues) and I. ricinus (172 residues) share 96.5% amino acid identity ( Figure 1B). The TRP prepropeptide contains dibasic cleavage sites for three putative TRPs (plus one repeat) for both tick species that are characterized by a general conserved carboxy-terminus motif, F-x 1 -G/A-x 3 -Ramide ( Figure 1B,C). This TRP motif is typical for other tick species such as Rhipicephalus sanguineus, Rhipicephalus microplus, and Dermacentor silvarum, as well as the parasitic mite Varroa destructor [49] and the fruit fly Drosophila melanogaster TRPs ( Figure 1C). I. scapularis, I. ricinus, R. sanguineus, R. microplus and D. silvarum share identical TRP1 and TRP2 amino acid sequence ( Figure 1C).

Expression of ISE6 TRP in Response to A. phagocytophilum Infection
Compared to uninfected ISE6 cells (control), A. phagocytophilum infection causes disparate trp transcript levels that are strain-dependent ( Figure 2A). Specifically, trp levels significantly decrease 0.5-fold in ISE6 cells infected with human strain NY18, but they significantly increase 2.5-fold with bovine strain BV49. No significant changes in trp transcript levels were detected in ISE6 cells infected with A. phagocytophilum ovine strain NV2Os ( Figure 2A). Therefore, A. phagocytophilum NV2Os was selected to infect ISE6 cells for subsequent immunocytochemistry (ICC) detection of TRP, since trp transcript levels were stable ( Figure 2B-G). Immunochemical analyses were facilitated by an antibody against the D. melanogaster neuropeptide natalisin (DromeNTL4) [50], (also see Section 4) that is a sister group of TRP. The DromeNTL4 C-terminal sequence (FPATRamide) is highly similar to the Ixodes TRP3 (FVATRamide) ( Figure 1C). Therefore, the term TRP-like immunoreaction (TRP-like IR) is used hereafter.

Expression of ISE6 TRP in Response to A. phagocytophilum Infection
Compared to uninfected ISE6 cells (control), A. phagocytophilum infection causes disparate trp transcript levels that are strain-dependent ( Figure 2A). Specifically, trp levels significantly decrease 0.5-fold in ISE6 cells infected with human strain NY18, but they significantly increase 2.5-fold with bovine strain BV49. No significant changes in trp transcript levels were detected in ISE6 cells infected with A. phagocytophilum ovine strain NV2Os ( Figure 2A). Therefore, A. phagocytophilum NV2Os was selected to infect ISE6 cells for subsequent immunocytochemistry (ICC) detection of TRP, since trp transcript levels were stable ( Figure 2B-G). Immunochemical analyses were facilitated by an antibody against the D. melanogaster neuropeptide natalisin (DromeNTL4) [50], (also see Section 4) that is a sister group of TRP. The DromeNTL4 C-terminal sequence (FPATRamide) is highly similar to the Ixodes TRP3 (FVATRamide) ( Figure 1C). Therefore, the term TRP-like immunoreaction (TRP-like IR) is used hereafter.   1 Possible allelic forms of two scaffolds. 2 The gene likely spans multiple scaffolds (and multiple predictions); * Incorrect, partially predicted transcript (see Supplementary Figure S2); ND-not detected. Note that XM, XR, and AXL predicted transcripts are from NCBI databases of ISE6 cell and were not detected in VectorBase ISE6 datasets. Nomenclature of the neuropeptides was used according to Coast and Schooley (2011) [51].   1 Possible allelic forms of two scaffolds. 2 The gene likely spans multiple scaffolds (and multiple predictions); * Incorrect, partially predicted transcript (see Supplementary Figure S2); ND-not detected. Note that XM, XR, and AXL predicted transcripts are from NCBI databases of ISE6 cell and were not detected in VectorBase ISE6 datasets. Nomenclature of the neuropeptides was used according to Coast and Schooley (2011) [51]. The ICC revealed cytoplasmic TRP-like IR in the majority of uninfected ISE6 cells ( Figure 2B,C). Although processes such as axon-like filopodia were apparent in a predominant cell subpopulation of both uninfected and infected ISE6 cells, these filopodia were absent of any TRP-like IR ( Figure 2B-G). The levels of NV2Os used (60% and 80%) are considered a high infection status that may cause a decrease in cell population-as qualified by cells infected with 80% NV2Os compared to uninfected and 60% infection ( Figure 2D-G). As expected by the stable trp transcript levels, and comparable to uninfected cells (Figure 2A), cytoplasmic TRP-like IR was also observed in ISE6 cells from both NV2Os infection levels ( Figure 2D-G). However, there are more compact clusters of TRP-like IR in uninfected and 60% NV2Osinfected cells than at 80% infection ( Figure 2B-G). Negative control in uninfected and infected cells did not reveal any TRP-like IR ( Figure 2H-M).

Expression of TRP in I. scapularis Synganglion
The in situ hybridization (ISH) determined trp transcript distribution in specific neurons of the I. scapularis synganglion ( Figure 3A

Discussion
The relationships between ticks and TBPs are complex, and understanding their molecular determinants is crucial for developing effective control strategies. Here, we introduce the existence of neuropeptide transcripts in ISE6 cells and present their neuropeptidergic features. These findings are supported by previous studies indicating that ISE6 cells are predominantly neuron-like [18].
The in silico prediction of neuropeptide genes from the ISE6 cell genome [52] reveals the same set of genes identified in the I. scapularis genome [37,43], but previous proteomic approaches failed to identify mature neuropeptides in ISE6 cells [18]. In the current study, we confirmed the expression of a complete Ixodes neuropeptidome in ISE6 cells. Considering that neuropeptides are regulators of all tick physiological processes and pathogens modulate tick physiology [32], our study suggests that ISE6 cells are an effective in vitro archetype for investigating TBP interactions with vital elements (i.e., neuropeptides) of the tick synganglion. Therefore, investigating tick-pathogen interactions by enlisting parallel, yet similar, cell-types (i.e., ISE6 cells and tick synganglion) may contribute to advancing tick control strategies to prevent TBP transmission. Attempts have been made to exploit components of the tick nervous system for control measures, but these studies did not achieve any information for pathogen infection or transmission [53][54][55].
Our ICC analyses detect distinct TRP-like IR in ISE6 cells, indicating the effective translation process of trp transcripts. It is not known whether the TRP-like IR in ISE6 cells is specific to neuropeptide epitopes within the prepropeptide, propeptide, or mature neuropeptides. However, evidence shows that mature neuropeptides are transported via the axons to their terminals [41], while pre/propeptides, theoretically, are proximate to the cell cytoplasm. Our study shows that TRP-like IR was strictly localized within the cytoplasm of ISE6 cells, while no staining was expressed in the axon-like filopoda. The ISE6 cells develop axon-like projections under unknown factors in the hemocoel of partially fed ticks [18]. Therefore, an interesting examination is if these unknown factors facilitate the axonal guidance of specific neuropeptidergic-type ISE6 cells. The subsequent immunodetection of additional neuropeptides may also elucidate the qualitative distribution of neuropeptidergic-type ISE6 cells.
The African species Rhipicephalus appendiculatus was the first tick experimentally used to confirm TRP-like IR in synganglion neurons [38]. The R. appendiculatus Pd 1 SG neurons are named Pd 1 DL neurons in I. scapularis where we localized, among other synganglion cells, trp transcripts using ISH. Our IHC approach confirms TRP-like IR in the majority of ISH synganglion stained cells, thereby highlighting specific I. scapularis TRP-producing neurons. The molecular characterization of the I. ricinus TRP encoding sequence also confirms an evolutionary relationship with the I. scapularis TRP previously identified [56] and verified here. The commonality in TRP-like expression for R. appendiculatus and I. scapularis synganglion, and the sequence identity of two out of three mature TRP between I. scapularis, I. ricinus, R. microplus, R. sanguineus and D. silvarum, suggests common attributes of TRP signaling in hard tick lineage.
In addition to the TRP-like IR in neuronal bodies, we identified several TRP-like IR axons exiting the I. scapularis synganglion, suggesting multiple visceral targets for TRPaxonal signaling. Localized TRP-like IR in LSOs, and their associated axons, supports previous findings that LSO cells are peptidergic and/or serve as neurohemal sites [37,38]. Furthermore, the arborization of TRP-like axon terminals on the paraganglionic sheath of the dorsal synganglion surface suggests a possible neurohemal site for discharging TRP to the hemolymph-thereby acting as neurohormones. Knowledge on TRP roles in ticks may help to localize their cognate receptors. The Ixodes genome possesses 10 putative TRP receptors (TRP-Rs) [43], whereas insects, at most, possess two TRP-Rs per species [57]. Since a TRP function in insects is diuresis [57], the increased number of Ixodes TRP-Rs may reflect the acute need for eliminating excess water and metabolic waste during tick feeding. However, whether these Ixodes TRP-Rs are functional and/or possess affinity to TRP ligands awaits experimental confirmation.
A. phagocytophilum induces tissue-specific transcriptional reprogramming, thereby affecting different cellular functions in infected tick [21][22][23][26][27][28]. A. phagocytophilum infection also alters tick physiology and behavior [32]. For example, A. phagocytophilum-infected ticks were more fitted to survive in cold temperatures [34] or desiccating conditions [35] compared to uninfected ticks. The differential regulation of heat shock protein transcripts (i.e., hsp20 and hsp70) upon A. phagocytophilum infection in ticks was associated with an increase in questing activity [35]. Our present study shows that trp levels in ISE6 cells are differentially regulated in response to infection by different A. phagocytophilum strains with specificities for bovine (BV49), ovine (NV2Os), or human (NY18) hosts. Although the precise molecular mechanism(s) by which particular A. phagocytophilum strains interact with TRP is unknown, our data suggest that A. phagocytophilum modulates TRP transcript levels in a strain-specific manner. Invertebrate TRPs have multiple functions in the central nervous system and intestine [48], but the specific functions of many tick neuropeptides have yet to be determined. Therefore, investigations are necessary to conclude if A. phagocytophilum affects TRP expression in infected ticks and thus alter their physiology.
Future research will confirm if other neuropeptide transcripts differentially respond to the presence of Anaplasma to elucidate the complex cascade of physiological features potentially modified by this pathogen. We anticipate that identifying the crucial neuronal components in tick-pathogen interactions will present key targets for developing novel tick management strategies applicable to a broad spectrum of TBPs. Thereby, these novel strategies will reduce the negative impacts TBPs have on human and animal health.

In Silico Identification of Neuropeptides Genes in ISE6 Databases
We used 34 I. scapularis preproneuropeptide query sequences previously identified in the genome project [43]. The list of queries was enriched by previously identified I. scapularis neuropeptide sequences natalisin [49] and elevenin [58], and the insulin-like peptide 2, identified using Drosophila subobscura sequence (GenBank access. no. XP_034657457.1). In addition, an isoform b of I. scapularis crustacean hyperglycemic hormone-(CHH)-related ion transport peptide (CHH/ITP) was also included to the query list. Thus, 38 preproneuropeptides were BLAST against publicly available ISE6 cell line databases in NCBI (www.ncbi.nlm.nih.gov) and VectorBase (www.vectorbase.org) to identify a computed predicted transcript of homologous neuropeptides. To identify genomic scaffolds encoding neuropeptides, VectorBase databases were exclusively used for BLAST. To reveal the expressed neuropeptide transcripts in ISE6 cells, we provided homology BLAST searches of the ISE6 Sequence Read Archive (SRA) in BioProject PRJNA239331 that contains 33 experimental datasets from Illumina transcript reads.

Culture of ISE6 Cells
The ISE6 embryonic tick cell cultures were maintained according to Munderloh et al. [17]. Healthy, uninfected ISE6 cells were propagated in a flask of 25 cm 2 with 5 mL of L15B300 medium. Infected cells were cultured in L15B300 medium supplemented with 0.1% NaHCO 3 and 10 mM HEPES with an adjusted pH at 7.5. Both uninfected and infected ISE6 cells were maintained at 34 • C.

Neuropeptides Quantitative Real-Time PCR in ISE6 Cells
Total RNA was extracted from ISE6 cells by Trizol Reagent (Invitrogen, Carlsbad, CA, USA) and complementary DNA (cDNA) was obtained by reverse transcription using the High Capacity cDNA Reverse Transcription kit (Invitrogen, Carlsbad, CA, USA). The template cDNAs were analyzed for amplification using the SYBR Green Master Mix (Roche, Basel, Switzerland) on a LightCycler ® 480 thermocycler (Roche, Basel, Switzerland). The primers for quantitative RT-PCR (qRT-PCR) are listed in Supplementary Table S1. Then, relative transcript levels were calculated using the ∆∆Ct ratio [59]. The ribosomal protein S4 (rps4) (GenBank accession number DQ066214) was used as a reference gene [60]. The statistical significance of normalized Ct values between groups was evaluated by Student's t-test with unequal variance in the GraphPad 5 Prism program (GraphPad Software Inc., San Diego, CA, USA). Differences were considered significant when p < 0.05. Three technical and two biological replications were performed.

Gene Cloning and Sequence Analyses
The I. scapularis EST sequence (EL516783) [44] was used to amplify the full open reading frame (ORF) of I. ricinus TRP. The forward and reverse primers used for amplification were 5 -AGTGATAAGCAAACCCGGTG-3 and 5 -CACGGCTTGGGGAATCTTCT-3 , respectively. The predicted full-length ORF of trp was amplified by PCR using cDNA isolated from unfed I. ricinus adult synganglia. The PCR amplicon of trp ORF was inserted into the pGEM-T Easy vector (Promega) followed by the transformation of competent DH5α bacteria (prepared using the Mix & Go kit, Zymo Research). Plasmid DNA was purified using the Nucleospin Plasmid kit (Macherey-Nagel, Düren, Germany). Recombinant plasmids were commercially sequenced (Eurofins, Luxemburg). We used the Signal P3.0 server to predict the signal peptide of TRP precursor [61].

Pathogen Infection of ISE6 Cells
Either the A. phagocytophilum bovine strain BV49 [62], ovine strain NV2Os [63], or human strain NY18 [64] were propagated in ISE6 tick cells as described before [19]. A. phagocytophilum infection was propagated by transferring 1/10th of an infected ISE6 cell culture to a new flask of healthy cells once infection reached 70%. To determine the level of infection, 300 µL of media with 100 µL of suspended cells were mixed, and 60 µL of the mixture were concentrated on a slide using the Shandon Cytospin (Thermo Fisher Scientific, Kalamazoo, MI, USA). Subsequently, the Hemacolor ® staining kit (Merck, Darmstadt, Germany) provided observation of cells under an Olympus BX53 light microscope (Olympus, Hamburg, Germany).

Immunocytochemistry of TRP in Uninfected and A. phagocytophilum-Infected ISE6 Cells
The cells were cultured in tubes of 7 cm 2 with 1 mL of L15B300 medium. Trac bottles containing internal glass slides (Dutscher, Brumath, France) were maintained at 34 • C as described in the section above. After one week, cells attached on the glass slides were washed with phosphate-buffered saline (PBS, 137 mM NaCl, 1.45 mM NaH 2 PO 4 .H2O, 20.5 mM Na 2 HPO 4 , pH 7.2) and fixed for 30 min at room temperature (RT) in 4% paraformaldehyde. After three washes with PBS + 0.01%Triton X-100 (PBST), cells were incubated at 4 • C overnight with anti-rabbit antibody (diluted 1:1000 in PBST) against D. melanogaster TRPlike neuropeptide, natalisin 4 (DromeNTL4) [50]. Then, the cells were washed three times for 5 min with PBST and incubated in dark conditions for 3 h at RT with goat anti-rabbit Alexa 488 conjugated secondary antibody (Life technologies) diluted at 1:1000. After three PBST washes, the cells were mounted in antifade media containing DAPI for nuclei staining (ProLong™ Diamond Antifade Mountant with DAPI, Thermo Fisher). The immunoreaction was observed under a Leica DMi8 confocal microscopy. The images were assembled in Adobe Photoshop CS4 (Adobe, Mountain View, CA, USA).

In Situ Hybridization
Validated ISH protocol previously developed by Šimo et al. [45,65] for tick synganglia was used. Briefly, a digoxygenin (DIG) probe synthesis kit (Roche Diagnostic, Germany) was used to synthesize a single-stranded DIG-labeled DNA probe for trp (662 bp). The I. ricinus trp ORF insert in the pGM-T Easy plasmid was used as a template (see the section Gene cloning and sequence analyses in Materials and Methods). Asymmetric PCR using either reverse (5 -CACGGCTTGGGGAATCTTCT-3 ) or forward (5 -AGTGATAAGCAAACCCGGTG-3 ) primer was performed to generate respective antisense and sense probes. Synthesized DIGlabeled probes were gel-purified and stored at -20 • C. Synganglia of unfed I. scapularis females were dissected in cold PBS and fixed with 4% paraformaldehyde for 2 h at RT. After cell membrane permeabilization with Proteinase K (New England BioLabs), synganglia were incubated with single-stranded DIG-DNA probes for 27 h at 48 • C. Then, specimens were incubated with mouse anti-digoxygenin/AP (Alkaline phosphatase; Roche Diagnostics, Germany) overnight at 4 • C. The reaction with hybridized DIG probes was developed by the addition of substrate/chromogen ready-to-use NBT-BCIP tablets (Roche Diagnostics, Germany). Finally, samples were incubated 5 min in 50% glycerol and subsequently mounted into 100% glycerol and observed by light microscopy (Olympus BX53). Images were assembled and enhanced in Adobe Photoshop CS4.

Wholemount Immunohistochemistry of Ixodes Synganglion
We slightly modified the IHC protocols previously described by Šimo et al. [38,45,47]. Briefly, Ixodes synganglia were dissected from unfed adult females, fixed with 4% paraformaldehyde solution for 2 h at RT, and then washed with PBS + 0.5% Triton X-100 (PBST). Tissues were incubated for 3 days at 4 • C with polyclonal anti-rabbit antibody against D. melanogaster NTL4 diluted 1:1000 in PBST. After three washes in PBST, the specimens were incubated two days at 4 • C with a goat anti-rabbit Alexa 488 conjugated secondary antibody (Life technologies) diluted at 1:1000. Samples were mounted in Prolong Antifade Diamond Mountant containing DAPI (Life Technologies) and analyzed by a Leica DMi8 confocal microscopy. Image assemblage was performed in Adobe Photoshop CS4. For neuronal cells of tick synganglia, we used nomenclatures as per Šimo et al. [38].

Supplementary Materials:
The following are available online at https://www.mdpi.com/2076-0 817/10/1/70/s1, Figure S1: Relative expression of ribosomal protein S4 (rps4) transcript and six different neuropeptide transcripts in uninfected ISE6 cells, Figure S2: Protein (A) and nucleotide (B) alignment of Ixodes scapularis and Ixodes ricinus TRP, Table S1: List of primers used in our qRT-PCR experiments. Funding: This research was funded by DIM1Health-Région Île-de-France (Acronym of the project: NeuroPaTick), the purchase of confocal microscopy used in this study was supported by the DIM1Health-Région Île-de-France (Acronym of the project: SAMiCAI). The NP was funded by French Institute of Slovakia/French embassy of Slovakia, the JJV was supported by Project FIT (Pharmacology, Immunotherapy, nanoToxicology), which was funded by the European Regional Development Fund and acknowledge a grant for the development of a research organization RVO: RO0516.

Author
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.