Ehrlichia Wnt SLiM ligand mimic deactivates the Hippo pathway to engage the anti-apoptotic Yap-GLUT1-BCL-xL axis

ABSTRACT Ehrlichia chaffeensis TRP120 effector has evolved short linear motif (SLiM) ligand mimicry to repurpose multiple evolutionarily conserved cellular signaling pathways, including Wnt, Notch, and Hedgehog. In this investigation, we demonstrate that E. chaffeensis and recombinant TRP120 deactivate Hippo signaling, resulting in the activation of Hippo transcription coactivator Yes-associated protein (Yap). Moreover, a homologous 6 amino acid (QDVASH) SLiM shared by TRP120 and Wnt3a/5a ligands phenocopied Yap and β-catenin activation induced by E. chaffeensis, rTRP120, and Wnt5a. Similar Hippo gene expression profiles were also stimulated by E. chaffeensis, rTRP120, SLiM, and Wnt5a. Single siRNA knockdown of Hippo transcription co-activator/factors, Yap, and transcriptional enhanced associate domain (TEAD) significantly decreased E. chaffeensis infection. Yap activation was abolished in THP-1 Wnt Frizzled-5 (Fzd5) receptor knockout cells (KO), demonstrating Fzd5 receptor dependence. In addition, the TRP120-Wnt-SLiM antibody blocked Hippo deactivation (Yap activation). Expression of anti-apoptotic Hippo target gene SLC2A1 (encodes glucose transporter 1; GLUT1) was upregulated by E. chaffeensis and corresponded to increased levels of GLUT1. Conversely, siRNA knockdown of SLC2A1 significantly inhibited infection. Higher GLUT1 levels correlated with increased B cell lymphoma-extra large (BCL-xL) and decreased BCL2-associated X, apoptosis regulator (Bax) levels. Moreover, blocking Yap activation with the inhibitor Verteporfin induced apoptosis that corresponded to significant reductions in GLUT1 and BCL-xL levels and activation of Bax and Caspase-3 and -9. This study identifies a novel shared Wnt/Hippo SLiM ligand mimic and demonstrates that E. chaffeensis deactivates the Hippo pathway to engage the anti-apoptotic Yap-GLUT1-BCL-xL axis.

. Ligand mimicry enables pathogens to impersonate the function of eukaryote ligands for infection (11).
We previously demonstrated that E. chaffeensis activates canonical Wnt signaling by direct interaction between a TRP120-Wnt-SLiM ligand mimic and the host cell Wnt Frizzled 5 (Fzd5) receptor (7). Interactions between canonical Wnt ligands and the Fzd5 receptor are known to stimulate Wnt transcriptional factor β-catenin but can also result in the deactivation of Hippo signaling, which coincides with the activation of transcription regulator Yes-associated protein (Yap) through Wnt-Hippo/Fzd receptor crosstalk (12). SLiMs are short (3-11 amino acids) linear sequences typically found within intrinsically disordered protein domains that are responsible for mediating various cellular processes (13,14). Interestingly, there are 23 predicted SLiMs in 14 proteins involved in Wnt signal transduction, including Axin, Dishevelled, and β-catenin (15). However, Wnt ligand SLiMs that mimic endogenous eukaryotic ligands, leading to pathway activation, have only recently been identified in Ehrlichia (7).
The Hippo pathway, discovered in Drosophila in 2003, is evolutionarily conserved in metazoans and universally recognized as a key regulator in embryogenesis, organ size, tissue homeostasis, cell proliferation, apoptosis, and tumorigenesis (16)(17)(18)(19)(20). Typically, when the Hippo pathway is active, the downstream transcriptional co-activator Yap is phosphorylated and deactivated, preventing nuclear translocation and activation of Hippo gene targets. Recently, investigations have linked pathway crosstalk between the Wnt and Hippo signaling pathways to control cell fate, demonstrating that Wnt5a and Wnt3a ligands bind the Fzd5 receptor to deactivate Hippo signaling and activate Yap (12,20,21). Additional crosstalk occurs between Yap and Wnt transcriptional regulator, β-catenin. When Hippo is active, phosphorylated Yap remains within the cytoplasm, where it sequesters β-catenin, leading to the degradation of β-catenin and inhibition of Wnt signaling (20). Thus, Hippo deactivation through interactions between Wnt5a/ Wnt3a and Fzd5 receptor must occur to support Wnt signaling. Notably, E. chaffeensis is a known β-catenin activator and utilizes a TRP120-Wnt-SLiM to activate β-catenin for Wnt gene regulation (7).
Regulation of apoptosis as a survival strategy is well documented during E. chaffeensis infection (22). Mitochondria are the primary regulators of apoptosis by both intrinsic and extrinsic pathways; thus, inhibition of mitochondrial outer membrane permeabilization (MOMP) is required to prevent apoptosis. It is known that E. chaffeensis utilizes the eukaryotic translation termination factor-1 (Etf-1) effector to stabilize mitochondria by regulating mitochondrial matrix protein manganese superoxide dismutase (MnSOD) to induce antioxidative protection and inhibit apoptosis (22). Further, E. chaffeensis utilizes a TRP120 Hedgehog SLiM ligand mimic to activate Hedgehog signaling, which prevents intrinsic apoptosis by maintaining B-cell lymphoma 2 (BCL2) levels and mitochondrial stability (10). However, there are likely other mechanisms E. chaffeensis engages to stabilize mitochondria, such as modulating the anti-apoptotic BCL2 family of proteins, including B-cell lymphoma-extra large (BCL-xL), which is regulated by the Hippo pathway (23).
The Hippo pathway is well known for its role in cancer but has recently been implicated in viral infections, including Hepatitis B virus (HBV), Hepatitis C virus (HCV), Human papillomavirus (HPV), Epstein-Barr virus (EBV), and Kaposi Sarcoma-associated herpesvirus (KSHV) (34). However, there are only a few reports of Hippo exploitation by parasites, fungi, and bacteria (35)(36)(37). Yap activation by viruses has been reported, but the precise mechanism whereby deactivation of Hippo signaling occurs to activate Yap remains unclear (34). Studies have shown that Yap activation during HBV infection triggers hepatocarcinogenesis and pathogenesis of the liver and may cause HBV-induced hepatocellular carcinoma. Additionally, HBV infection in Alb-preΔS2 transgenic mice increases the expression of Hippo target genes BIRC5, ANKRD1, CTGF, and CYR61 (34,38). HPV E6 major oncoprotein inhibits active Yap degradation, and Yap knockdown impairs E6-mediated cell proliferation, indicating that Yap activation plays a role in the proliferation of cervical cancer cells (39).
Although the Hippo pathway is targeted by multiple pathogens, the pathogen-host interactions and mechanisms involved in Hippo pathway exploitation have not been defined. We have previously identified an Ehrlichia Wnt SLiM ligand mimic that activates Wnt signaling. Therefore, since Hippo signaling is initiated through Wnt Fzd recep tors, we considered that Hippo signaling may be regulated through the same ligandreceptor complex during infection (40). This investigation reveals a molecular mecha nism whereby E. chaffeensis utilizes a eukaryotic Wnt SLiM ligand motif interaction with the Fzd5 receptor to deactivate Hippo signaling, thereby activating the Yap-GLUT1-BCL-xL axis to promote an anti-apoptotic cellular environment.

E. chaffeensis activates Yap and Hippo gene expression
Hippo deactivation mediated by Wnt ligand engagement of the Fzd5 receptor results in Yap activation and nuclear translocation, where it binds the transcription factor TEAD to regulate Hippo gene targets (12,20,(41)(42)(43). Recent studies demonstrate that E. chaffeensis directly interacts with the Fzd5 receptor to activate β-catenin (7). To investigate whether E. chaffeensis activates Yap via Hippo-Wnt ligand-receptor crosstalk, confocal microscopy was used to detect active Yap in the nucleus of infected THP-1 cells within 4 h post-infection (hpi). Moreover, significant nuclear accumulation of active Yap was observed over 48 hpi compared to uninfected controls based on intensity graphs ( Fig. 1A, C, and D). Further, active Yap accumulated in the nucleus of E. chaffeensis-infec ted primary human monocytes (PHMs; 10 hpi) compared to the uninfected control ( Fig.  1B and E). To examine the role of E. chaffeensis in Hippo deactivation (Yap activation), we examined Hippo pathway gene transcription using a human Hippo signaling PCR array (Fig. 1F). Significant activation of Hippo pathway component genes was detected during E. chaffeensis infection, with the majority (63%) of Hippo genes being upregulated, including major Hippo and Wnt components YAP, TAZ, TEAD1, TEAD2, TEAD3, TEAD4, and DVL2 compared to controls (Fig. 1F).

TRP120 activation of Yap and Hippo gene targets
To examine the role of TRP120 in Hippo deactivation, THP-1 cells and PHMs were incubated with recombinant TRP120 protein (rTRP120-FL), and Yap activation was examined using confocal microscopy ( Fig. 2A and B). Active Yap accumulated in the nucleus of THP-1 cells ( Fig. 2A and C) and PHMs ( Fig. 2B and D) at 6 and 10 h post-treat ment (hpt), respectively, consistent with Yap activation by recombinant Wnt5a (rWnt5a). To further confirm the role of TRP120 in Hippo regulation, cells were stimulated with rTRP120-FL or rWnt5a for 24 h, and a transcriptional analysis was performed ( Fig. 2E and F). Hippo genes (45%) were significantly upregulated compared to the rTrx negative control, including genes important for Hippo and Wnt signaling (YAP, TAZ, TEAD4, and DVL2), and 16% were downregulated (Fig. 2E). In comparison, cells treated with rWnt5a had significant transcriptional upregulation of Hippo genes (65%), including YAP, TAZ, TEAD1, TEAD2, TEAD3, TEAD4, WNT1, and DVL2, and 22% of genes were significantly downregulated (Fig. 2F). Though there were differential expression patterns of genes in TRP120 and Wnt5a-treated cells, we found that 34 Hippo target genes, including YAP, TAZ, TEAD4, and DVL2, were upregulated in both rTRP120-FL and rWnt5a treat ments. Together, these data demonstrated that treatment of THP-1 cells with rTRP120-FL resulted in intracellular Yap accumulation and changes in Hippo pathway gene transcript levels that were similar, but not identical, to those observed after incubation with rWnt5.  (D and E) Intensity graphs demonstrate the mean nuclear accumulation of active Yap in THP-1 cells and PHMs, respectively. Analysis was performed using ImageJ to determine the mean gray value from randomized areas/slides (n = 10), and data are shown as mean ± SD (*P < 0.05; ***P < 0.001). (F) The table represents the normalized expression of significantly regulated Hippo array genes between E. chaffeensis-infected and uninfected cells at 24 h. The scatterplot represents the expression of all Hippo array genes. The top and bottom lines depict a twofold upregulation or downregulation, respectively, compared to an uninfected control.
Scatterplots are representative of three (n = 3) biological and technical replicates.

TRP120-Wnt-SLiM inactivates Hippo signaling and activates Yap
TRP120 contains a tandem repeat domain (TRD), with four tandem repeats, flanked by Nand C-terminal domains. Various TRP120 SLiMs have been reported within the TRD and (C and D) Intensity graphs demonstrate the mean nuclear accumulation of active Yap in the respective THP-1 cells and PHMs. Analysis was performed using ImageJ to determine the mean gray value from randomized areas/slides (n = 10), and data are shown as mean ± SD (***P < 0.001). (E) The table represents significantly regulated Hippo signaling PCR array gene expression in THP-1 cells stimulated with rTRP120-FL (1 µg/mL) after normalization to control cells treated with rTrx (1 μg/mL) at 24 h. The respective normalized expression of rTRP120-FL-regulated Hippo array genes was performed with three biological and technical C-terminus that are relevant to E. chaffeensis pathobiology, including posttranslational modification motifs, DNA-binding motifs, and ubiquitin ligase catalytic motifs (7). We previously reported that E. chaffeensis TRP120 TRD utilizes SLiMs to regulate Wnt, Notch, and Hedgehog signaling pathways (7,9,10). The discovery of a TRP120-Wnt-SLiM that activates Wnt signaling was previously reported, and homology was identified between TRP120 and Wnt5a (7). However, based on a revised BLAST analysis of TRP120 with both Wnt5a and Wnt3a (Fig. 3A), we identified a shorter Wnt SLiM (QDVASH) shared by both ligands (60% and 83% similarity, respectively) within the previously identified TRP120-Wnt-SLiM (IKDLQDVASHESGVSDQ).
Further, we treated THP-1 cells with a single amino acid TRP120-Wnt-SLiM histidine deletion mutant (QDVAS) and observed no significant activation of Yap or β-catenin, indicating that the 6 amino acid TRP120-Wnt-SLiM containing histidine is essential for activation (Fig. S2).

Hippo co-activator and transcription factors influence infection
Although the Hippo pathway is widely recognized for its role in embryogenesis and tumorigenesis, it also plays a key role in regulating apoptosis, which is crucial for successful ehrlichial intracellular infection (4,10,16,17,22). To determine whether E. chaffeensis survival depends on Hippo transcriptional components, we used RNAi to individually silence genes for YAP, TEAD1, TEAD3, and TEAD4 (Fig. 5A). Ehrlichial load was significantly reduced in all transfection groups 24 h post-RNAi transfection compared to the scramble siRNA-transfected controls (Fig. 5B).

A TRP120-Wnt-SLiM antibody blocks Yap activation
To further elucidate the role of TRP120-Wnt-SLiM during E. chaffeensis infection, we investigated whether blocking E. chaffeensis infection or the TRP120-Wnt-SLiM with a TRP120-Wnt-SLiM-targeted antibody would inhibit Yap activation. E. chaffeensis-infected and TRP120-Wnt-SLiM-treated cells in the presence of α-TRP120-Wnt-SLiM demonstrated a significant reduction in active Yap relative to E. chaffeensis-infected and TRP120-Wnt-SLiM-treated cells in the presence of α-TRP120-PIS antibody (control) (Fig. 6A through C).   (Fig. S1B). However, unlike Yap, there was significant activation of β-catenin compared to control cells, likely due to the contribution of other Fzd receptors known to interact with TRP120 (7).

Hippo target gene SLC2A1 is upregulated during E. chaffeensis infection
To understand the basis and downstream effects of Hippo regulation during E. chaffeensis infection, the Hippo target gene and the anti-apoptotic SLC2A1 were investigated. SLC2A1 encodes the glucose transporter GLUT1, which is necessary for preventing apoptosis through the Yap-GLUT1-BCL-xL axis (19,23,26,33,45). During E. chaffeensis infection, significant upregulation of SLC2A1 was detected at 3 and 24 hpi (Fig. 8A). Further, TRP120-Wnt-SLiM upregulated SLC2A1 in a concentration-dependent manner at 6 hpt (Fig. 8B). To determine whether E. chaffeensis infection relies on SLC2A1 for survival, we used RNAi to silence SLC2A1 in THP-1 cells. The ehrlichial load was significantly reduced (24 hpi) in SLC2A1-siRNA-transfected cells compared to the scramble control transfected cells (Fig. 8C).

E. chaffeensis TRP120-Wnt-SLiM-mediated regulation of GLUT1, BCL-xL, and Bax
It is well documented that Hippo signaling promotes cell proliferation and prevents cell apoptosis through the Yap-GLUT1-BCL-xL axis (19,23,26,33,45). Further, BCL-xL is involved in the inhibition of mitochondria-mediated pro-death pathway by directly inhibiting Bax and subsequent caspase activation (45)(46)(47). Based on our results, we hypothesized that E. chaffeensis deactivates Hippo signaling and activates Yap to increase GLUT1 and BCL-xL and decrease Bax levels. To examine this question, THP-1 cells were infected with E. chaffeensis or treated with TRP120-Wnt-SLiM and TRP120-Wnt-SLiM-mut. E. chaffeensis and TRP120-Wnt-SLiM significantly increased GLUT1 and was identified using NCBI BLAST between the TRP120 TR and Wnt3a/Wnt5a ligands (activators of Yap) amino acid sequences. The complete amino acid sequence of one TR is shown, with homologous Wnt SLiM identified in green and percent homology to the right of the sequence. (B) The table displays the various TRP120 peptide amino acid sequences used in the TRP120-Wnt-SLiM study. TRP120-Wnt-SLiM represents the homology sequence identified through BLAST.
TRP120-Wnt-SLiM-mut contains glycine and alanine substitutions in the Wnt-SLiM region and is used as a negative control. TRP120-TR-Wnt5a is a 19-amino acid sequence that contains the identified TRP120-Wnt homology sequence. TRP120-TR (−) is a sequence within TRP120-TR that does not contain the defined TRP120-Wnt homology sequence. (C) Confocal immunofluorescence microscopy of untreated (−) or peptide-treated THP-1 cells (1 μg/mL). THP-1 cells were stained with active Yap antibody, and the micrograph shows increased levels of active Yap (green) in TRP120-TR-Wnt5a and TRP120-Wnt-SLiM-treated, but not in untreated, TRP120-TR (−) or TRP120-Wnt-SLiM-mut-treated THP-1 cells (6 hpt were selected to detect active Yap nuclear translocation. (E and F) Intensity graphs demonstrate the mean nuclear accumulation of active Yap in the respective THP-1 cells and PHMs. Analysis was performed using ImageJ to determine the mean gray value from randomized areas/slides (n = 10). Data are represented as means ± SD (***P < 0.001).
BCL-xL and decreased Bax levels compared to controls, consistent with a Yap-dependent anti-apoptotic profile induced by E. chaffeensis ( Fig. 9A through D).

TRP120-Wnt-SLiM-mediated regulation of GLUT1, BCL-xL, and Bax during Yap inhibition
Our results support the importance of the anti-apoptotic Yap-GLUT1-BCL-xL axis during infection. Further, we hypothesized that E. chaffeensis deactivates Hippo to regulate GLUT1, BCL-xL, and Bax. To test this hypothesis, we used a Yap inhibitor (Verteporfin) to determine whether E. chaffeensis infection or TRP120-Wnt-SLiM regulated GLUT1, BCL-xL, and Bax levels during Yap inhibition. During infection, there was a significant reduction in GLUT1 in the presence of Verteporfin, demonstrating that E. chaffeensis depends on Yap activation to increase GLUT1 (Fig. 10B). Additionally, during Verteporfin treatment, BCL-xL levels were insignificant during E. chaffeensis infection compared to the control (Fig. 10C). Further, Bax levels significantly increased during E. chaffeensis infection in the presence of Verteporfin compared to the control (Fig. 10D).

Yap inhibition induces an apoptotic profile during E. chaffeensis infection
Our results support the importance of the anti-apoptotic Yap-GLUT1-BCL-xL axis during infection and demonstrate that E. chaffeensis infection and TRP120-Wnt-SLiM engage the Hippo pathway to regulate GLUT1, BCL-xL, and Bax. Further, we hypothesized that E. chaffeensis regulates the anti-apoptotic Yap-GLUT1-BCL-xL axis during infection to prevent subsequent Caspase-9 and -3 activation and intrinsic apoptosis. To test this hypothesis, E. chaffeensis-infected and uninfected THP-1 cells were treated with the Yap inhibitor Verteporfin or DMSO (Fig. 11A). E. chaffeensis-infected Verteporfin-treated cells demonstrated a significant increase in cytoplasmic condensation (a precursor to apoptosis) at 24 hpi compared to uninfected Verteporfin-treated cells and E. chaffeensisinfected and uninfected DMSO-treated cells, supporting the conclusion that E. chaffeensis activates Yap to prevent apoptosis. Additionally, ehrlichial survival was significantly reduced in the presence of Verteporfin compared to the control (DMSO) (Fig. 11B). Further, cell viability significantly decreased in E. chaffeensis-infected cells treated with Verteporfin (Fig. 11C). To define a direct mechanism by which E. chaffeensis activates Yap to prevent apoptosis, we evaluated levels of pro-and cleaved Caspase-9 and -3 during infection in the presence of Verteporfin (Fig. 11D through E). E. chaffeensis-infected cells treated with inhibitor showed a significant decrease in pro-Caspase-9 and -3 levels, while cleaved Caspase-9 and -3 levels significantly increased during E. chaffeensis infection in the presence of Verteporfin compared to DMSO-treated E. chaffeensis-infected cells (control).

DISCUSSION
Based on the premise of Hippo-Wnt crosstalk and the regulation of Wnt signaling by the TRP120-Wnt-SLiM, we sought to identify whether the TRP120-Wnt-SLiM deactivates Hippo, leading to Yap activation. Indeed, we reveal that TRP120-Wnt-SLiM regulates Hippo signaling and identifies the downstream effects directed at inhibiting intrinsic host-cell apoptosis. E. chaffeensis contains a Wnt-SLiM and depends on the Wnt-Fzd5 receptor to activate the transcription co-activator Yap, which promotes a significant upregulation in genes critical for Hippo and Wnt signaling. This is the first report of a single eukaryotic SLiM mimic in bacteria that can regulate multiple conserved signal ing pathways. This study also reveals a novel strategy utilized by obligate intracellu lar bacteria to extend host cell lifespan and highlights the importance of pathogen utilization of eukaryotic cellular signaling motifs for reprogramming the host cell to promote infection. Although Hippo signaling has been studied during viral infection, little is known regarding Hippo signaling during bacterial infection. We investigated whether E. chaffeensis regulates Hippo signaling during infection. Indeed, we confirmed that infection induces Yap activation and transcriptional induction of Hippo pathway genes. Although there was differential expression of Hippo pathway genes in TRP120 and Wnt5a-treated cells, we discovered that many Hippo target genes were upregulated by Full-Length Text both TRP120 and Wnt5a, which supports TRP120 mimicry of Wnt5a. Some differences between TRP120 and Wnt5a were expected since the biological functions of various Wnt ligands differ despite a highly similar amino acid sequence (48). Additionally, TRP120 also contains Notch and Hedgehog SLiMs, which may also influence gene expression due to the intricate crosstalk between the pathways (49,50).
To further establish the direct mechanism of Hippo regulation during infection, we determined that TRP120-Wnt-SLiM sufficiently induces active Yap. Notably, E. chaffeensis, TRP120, and TRP120-Wnt-SLiM induced similar Yap activity in THP-1 cells and PHMs, which is important to note since primary cells have a limited lifespan and THP-1 cells are a more practical alternative for laboratory studies. Additionally, we further defined the previously reported Wnt-SLiM (7), shortening the SLiM to 6 aa (from 17 aa) using BLAST analysis to detect a short region of homology among TRP120 and Wnt5a/3a ligands. Additionally, we determined that the histidine residue is essential for Yap and β-catenin activation. Various studies have indicated the importance of histidine in protein-protein (targets TRP120 sequence DLQDVASHESGVSDQPAQV) or α-TRP120-PIS (neg ctrl) (1.5 μg/mL) for 1 h or overnight, respectively, before incubation with THP-1 cells. THP-1 cells were harvested at 6 hpt, immunostained with active Yap antibody (green), and visualized by confocal fluorescence microscopy (scale bar = 10 μM). Randomized areas/slides (n = 10) were selected to detect active Yap nuclear translocation. (B) The intensity graph demonstrates the mean nuclear accumulation of active Yap in respective THP-1 cells. Analysis was performed using ImageJ to find the mean gray value from randomized areas/slides (n = 10).
(C) Western blot analysis of treatment groups with GAPDH as a loading control with a bar graph of Western blot analyzed from densitometry values normalized to GAPDH (A-C); α-TRP120-Wnt-SLiM inhibits active Yap upregulation in cells with E. chaffeensis or TRP120-Wnt-SLiM compared to α-TRP120-PIS. Untreated cells were incubated with α-TRP120-Wnt-SLiM or α-TRP120-PIS as negative controls. Experiments were performed with three biological and technical replicates, and significance was determined through t-test analysis. Data are represented as means ± SD (*P < 0.05; **P < 0.01; ***P < 0.001).
interactions. In fact, histidine is known as the most active and versatile amino acid, is often the key residue in enzyme catalytic reactions, and is essential for protein interac tions (51).
The shorter TRP120-Wnt-SLiM highlights shared amino acids between Wnt5a/3a that may be critical in binding Fzd receptors and activating signaling. In our previous study, we defined the TRP120-Wnt-SLiM based on sequence and functional similarities between TRP120 and Wnt8, since Wnt8 activates β-catenin and structural studies have defined many residues for Wnt8-Fzd binding (7,52). However, many of the Wnt residues necessary for binding Fzd receptors are not conserved among the Wnt ligands (53). Additionally, Wnt5a and Wnt3a residues for Fzd binding are not well defined; however, these ligands are relevant to this study since they activate Yap (12). Identification of a SLiM with the capability to affect multiple pathways is new to science and will have a significant impact on how these ligand-receptor interactions are viewed by cell biologists and others. Analysis was performed using ImageJ to determine the mean gray value from randomized areas/slides (n = 10). (C) Western blot analysis of treatment groups to determine active Yap, Fzd5, and Dsb levels with GAPDH as a loading control. Western blot bar graph was analyzed from densitometry values normalized to GAPDH. (A-C) Experiments were performed with three biological and technical replicates, and significance was determined through t-test analysis. Data are represented as means ± SD (***P < 0.001).

Full-Length Text
In our investigation, TRP120-Wnt-SLiM exhibited stronger upregulation of Hippo gene targets than TRP120. This is likely due to higher molar concentrations of the SLiM sequence present in the TRP120-Wnt-SLiM treatment. To further support our results, we used Wnt-SLiM (QDVASH) to target Wnt signaling and determined that it does activate both Hippo and Wnt signaling, consistent with known Hippo-Wnt receptor overlap and crosstalk (20). Additionally, we used an anti-SLiM antibody that blocked Yap activation during E. chaffeensis infection and TRP120-Wnt-SLiM treatment, demonstrat ing the importance of the SLiM in Hippo regulation during infection and confirming that TRP120-Wnt-SLiM is the only SLiM utilized by E. chaffeensis to activate Yap.
In recent years, our laboratory has determined that TRP120 contains multiple SLiMs within the intrinsically disordered TRD that act as ligand mimics to regulate Wnt, Notch, Hedgehog, and Hippo signaling. E. chaffeensis likely contains multiple pathways activating SLiMs due to the intricate crosstalk between the pathways and the role each plays in regulating apoptosis to promote infection (20,50,54). SLiMs are disor dered, short, linear sequences that contain a limited number of specificity-determining residues (55). Few mutations are necessary for the generation of new SLiMs, allowing rapid convergent evolution of SLiMs within proteins de novo, enabling rapid functional flexibility (56,57). E. chaffeensis has likely convergently evolved TRP120 SLiMs to engage multiple cellular signaling pathways for redundancy and to influence anti-apoptotic signaling through different pathways. All defined TRP120 SLiMs activate conserved signaling pathways known to prevent apoptosis, which may be a strategy executed by E. chaffeensis to ensure host cell survival and productive infection.
TRP120 is a Wnt ligand mimic and directly interacts with the Fzd5 receptor (7). Wnt5a and Wnt3a ligands interact with the Fzd5 receptor, which can lead to the activation of Hippo and Wnt transcriptional regulators Yap and β-catenin, respectively. Further, while only Fzd-1, -2, and -5 are associated with Yap activation, most Fzd receptors are known to activate β-catenin (7,12,42,43). Additionally, the co-expression of Fzd5 with the co-receptor tyrosine kinase ROR1 potentiates Fzd5 receptor-induced Yap activation (12). Previously, we demonstrated that E. chaffeensis survival depends on ROR1, which may be due to its role in the co-activation of Yap (7,12). To better understand why E. chaffeensis interacts with the Fzd5 receptor and how it relates to Yap activation, we utilized Fzd5 receptor KO to demonstrate that the Fzd5 receptor is essential for Yap activation during infection. We found that the activation of E. chaffeensis and TRP120-Wnt-SLiM Yap is solely dependent on the Fzd5 receptor. Yap activation has been associated with Fzd-1, -2, and -5 in HEK293 cells (12,21); however, the fact that Yap activation induced by E. chaffeensis and TRP120-Wnt-SLiM depends solely on the Fzd5 receptor may be related to fundamental differences in cell types (innate immune phagocyte vs. epithelial kidney cell). In contrast to our finding that Hippo relies completely on the Fzd5 receptor for signaling, β-catenin activation was only significantly reduced (~50%) in the Fzd5 receptor KO cells. This is consistent with reports demonstrating that multiple Wnt ligands and Fzd receptors are involved in β-catenin activation. Similarly, we have observed interactions between TRP120 and other Fzd receptors known to activate β-catenin (7). Cellular apoptosis plays an important role as an innate defense mechanism against microbial infection. During infection, cells utilize apoptotic mechanisms for process ing infected apoptotic bodies containing pathogens to facilitate antigen presentation and protective immunity (58). Preventing apoptosis is critical for obligate intracellular bacteria since maintaining a replicative niche is essential to complete the infection cycle. Obligate intracellular pathogens, including Rickettsia, Anaplasma, Mycobacterium, Chlamydia, and others, have evolved multiple regulatory mechanisms to inhibit host cell apoptosis, including regulation of mitochondria-mediated intrinsic apoptosis (58)(59)(60)(61)(62)(63)(64)(65). Additionally, intracellular bacteria regulate the BCL-2 family of proteins to stabi lize mitochondria and promote host cell survival. Recently, we demonstrated that E. chaffeensis activates the Hedgehog pathway to regulate mitochondria-mediated intrinsic apoptosis via BCL-2 and extend the host cell lifespan (10). Chlamydia trachomatis upregulates MCL-1 to inhibit Bax-induced apoptosis (66), and M. tuberculosis upregulates BCL-2 in macrophages during infection to prevent apoptosis (67).
Recently, investigations have demonstrated a major role for Hippo signaling in glucose metabolism to preserve mitochondrial stabilization and prevent apoptosis. To prevent apoptosis, the cell deactivates Hippo signaling to activate the transcriptional co-activator Yap to upregulate Hippo gene targets, including SLC2A1, which encodes the glucose transporter GLUT1. The upregulation of GLUT1 promotes glucose metabolism, which subsequently promotes the upregulation of BCL-xL (23,(26)(27)(28). Previous studies demonstrate that a reduction in GLUT1 protein expression increases Bax, Bak, Bim, and Bid (pro-apoptotic) and inhibits MCL-1 and BCL-xL (33). Additionally, E. chaffeensis infection and TRP120-Wnt-SLiM treatment increased GLUT1 and BCL-xL and decreased Bax levels. Further, we show that a small-molecule Yap inhibitor prevents E. chaffeensis from regulating GLUT1, BCL-xL, and Bax and induces a pro-apoptotic profile. These results reveal a novel anti-apoptotic mechanism by which E. chaffeensis modulates the Hippo pathway for infection by extending the host cell lifespan using glucose metabo lism, which is consistent with the role of Hippo signaling in cell biology. Remarkably, and Bax levels during TRP120-Wnt-SLiM, TRP120-Wnt-SLiM-mut-treated (1 μg/mL) and untreated THP-1 cells (24 hpt) with GAPDH as a loading control. (C and D) The blots were obtained from the same Western blot membrane. As a result, the GAPDH in these panels is identical. (A-E) Bar graphs depict Western blot densitometry values normalized to GAPDH or α-actinin. Experiments were performed with three biological and technical replicates, and significance was determined through t-test analysis. Data are represented as mean ± SD (*P < 0.05; ***P < 0.001).

Full-Length Text
E. chaffeensis regulates Hippo and Hedgehog to target various BCL-2 family proteins and inhibit intrinsic apoptosis, a remarkable redundancy resulting in comprehensive regulation of anti-apoptotic signaling for intracellular survival.
The current study reveals a model of molecular mimicry in which a single bacte rial SLiM phenocopies endogenous ligands to regulate multiple conserved signaling pathways. The discovery of SLiMs that phenocopy endogenous ligands provides a valuable tool for various fields to study cell signaling and cancer biology. Here, we characterize a TRP120-Wnt-SLiM that utilizes Hippo-Wnt pathway crosstalk to engage the Yap-GLUT1-BCL-xL axis to promote an anti-apoptotic profile (Fig. 12). This study demonstrates the importance of Hippo signaling in preventing apoptosis for ehrlichial replication and provides a potential new target for therapeutic development. The potential to use E. chaffeensis as a model to define the role of SLiM ligand mimicry and an evolutionary conserved eukaryotic signaling pathway will lead to a broader understanding of intracellular pathogen biology and provide mechanistic targets for intervention.

Cell culture and E. chaffeensis cultivation
Human monocytic leukemia cells (THP-1; ATCC TIB-202) or PHMs were propagated in RPMI 1640 with L-glutamine and 25 mM HEPES buffer (Invitrogen, Carlsbad, CA), supplemented with 10% fetal bovine serum, and incubated at 37°C in a 5% CO 2 atmosphere. Peripheral blood mononuclear cells were obtained from deidentified healthy human donors (Gulf Coast Regional Blood Center, Houston, TX), and PHMs were isolated using MACS negative selection (Miltenyi Biotec, Cambridge, MA). E. chaffeensis (Arkansas strain) was cultivated in THP-1 cells and PHMs as previously described (10).

Protein sequence analysis
The NCBI Protein Basic Local Alignment Search Tool (Protein BLAST) was utilized for sequence alignment of TRP120 (NCBI accession number: AAO12927.1) and Wnt5a and Wnt3a amino acid sequences (NCBI accession numbers: AAH74783 and EAW69829).

RNAi and Ehrlichia quantification
All siRNAs were ON-TARGETplus SMARTpool (Dharmacon, Lafayette, CO). siRNA KD was performed as previously described (7,10). Scrambled RNAi was used as an siRNA control. THP-1 cells were infected with cell-free E. chaffeensis (MOI 100) 24 h post-transfection. Cells were harvested at 24 hpi, and the ehrlichial load was determined using qPCR. All knockdowns were performed with three biological and technical replicates, and significance was determined using t-test analysis.

Human Hippo signaling pathway PCR array
The human Hippo signaling target PCR array (Qiagen) was used to determine the expression of 84 key Hippo target genes. PCR arrays were performed according to the manufacturer's protocol (Qiagen). Real-time PCR was performed using the RT 2 Profiler PCR array and SYBR green master mix (Qiagen) using the QuantStudio 6 Flex Real-Time PCR system (Thermo Fisher Scientific). PCR data analysis was performed as previously described (7).

Real-time qPCR
The analysis of SLC2A1 gene expression during infection was determined using real-time qPCR. THP-1 cells were infected with E. chaffeensis (MOI 100). Cells were harvested at 0, 3, and 24 hpi to examine gene expression during the entry and early replication phases. The fold change in SLC2A1 from 0 to 3 or 24 hpi was calculated using the 2 −ΔΔCT method and C T values for host SLC2A1 and GAPDH genes as previously described (69).

Hippo inhibitor infection analysis
E. chaffeensis-infected (MOI 50), uninfected, TRP120-Wnt-SLiM-and TRP120-Wnt-SLiMmut-treated THP-1 cells were incubated with DMSO or Verteporfin (7 µg/mL) for 24 h, then cells were harvested for Western blot and Diff-Quik staining (Thermo Fisher Scientific). Diff-Quik images are in their original form, taken with a DP25 camera using cellSens. However, this software can create artifacts. Ehrlichial load was determined using qPCR as described above. Cell counts and viability were determined by the Cellometer Mini Brightfield Automated Cell Counter (Nexcelom, Lawrence, MA).

ADDITIONAL FILES
The following material is available online.