Ehrlichia Notch signaling induction promotes XIAP stability and inhibits apoptosis

ABSTRACT Ehrlichia chaffeensis has evolved multiple strategies to evade innate defenses of the mononuclear phagocyte. Recently, we reported the E. chaffeensis tandem repeat protein (TRP)120 effector functions as a Notch ligand mimetic and a ubiquitin ligase that degrades the nuclear tumor suppressor, F-box and WD repeat domain-containing 7, a negative regulator of Notch. The Notch intracellular domain (NICD) is known to inhibit apoptosis primarily by interacting with X-linked inhibitor of apoptosis protein (XIAP) to prevent degradation. In this study, we determined that E. chaffeensis activation of Notch signaling increases XIAP levels, thereby inhibiting apoptosis through both the intrinsic and executioner pathways. Increased NICD and XIAP levels were detected during E. chaffeensis infection and after TRP120 Notch ligand mimetic peptide treatment. Conversely, XIAP levels were reduced in the presence of Notch inhibitor DAPT. Cytoplasmic and nuclear colocalization of NICD and XIAP was observed during infection and a direct interaction was confirmed by co-immunoprecipitation. Procaspase levels increased temporally during infection, consistent with increased XIAP levels; however, knockdown (KD) of XIAP during infection significantly increased apoptosis and Caspase-3, -7, and -9 levels. Furthermore, treatment with SM-164, a second mitochondrial activator of caspases (Smac/DIABLO) antagonist, resulted in decreased procaspase levels and increased caspase activation, induced apoptosis, and significantly decreased infection. In addition, RNAi KD of XIAP also decreased infection and significantly increased apoptosis. Moreover, ectopic expression of TRP120 HECT Ub ligase catalytically defective mutant in HeLa cells decreased NICD and XIAP levels and increased caspase activation compared to HeLa cells with functional HECT Ub ligase catalytic activity (TRP120-WT). This investigation reveals a mechanism whereby E. chaffeensis modulates Notch signaling to stabilize XIAP and inhibit apoptosis.

Two apoptosis pathways, extrinsic and intrinsic, have been defined and are well characterized. The extrinsic pathway is activated through a death ligand receptor resulting in the activation of Caspase-8 and induction of the execution pathway leading to apoptosis (26)(27)(28). In comparison, the intrinsic pathway is initiated by various nonreceptor-mediated stimuli that result in mitochondrial changes, specifically mitochondrial permeability transition (MPT). MPT results in cytochrome c release, triggering the formation of a complex known as an apoptosome and subsequent Caspase-9 activation resulting in apoptosis (29,30). Execution of apoptosis occurs when Caspase-8 and/or -9 cleave inactivated executioner Caspase-3/7 into active forms, leading to the cleavage of various downstream targets important for cell survival (29,31,32). Modulation of several genes that regulate host apoptotic mitochondrial events (intrinsic apoptosis) occurs during E. chaffeensis infection including Bcl-2, BirC3, and downregulation of apoptotic inducers, such as Bik, BNIP3L, and hematopoietic cell kinase (HCK) (33). The E. chaffeensis effector, ECH0825, is known to inhibit Bax-induced apoptosis by increasing mitochondrial manganese superoxide dismutase (MnSOD) to reduce reactive oxygen species-mediated damage (25). Although the manipula tion of intrinsic apoptosis as a survival mechanism for E. chaffeensis has been previ ously reported, there remain significant unanswered questions about the mechanisms involved.
Our laboratory has recently reported that E. chaffeensis evasion of monocyte host defenses involves activation of conserved host signaling pathways, including Wnt, Notch, and Hedgehog (2,13,14). Notably, TRP120 activates the evolutionarily conserved Notch signaling pathway using a novel molecularly defined pathogen-encoded Notch SLiM ligand mimic found within the tandem repeat (TR) domain (13). Notch signaling plays significant roles in cellular homeostasis, MHC Class II expression, B-and T-cell develop ment, and modulation of innate immune mechanisms such as autophagy and apoptosis (34)(35)(36)(37)(38)(39). Recently, we have reported that E. chaffeensis TRP120-induced Notch signal ing results in downregulation of toll-like receptor (TLR) 2/4 expression (5). Moreover, we determined that TRP120 degrades the Notch negative regulator, F-box and WD repeat domain-containing 7 (FBW7), resulting in increased levels of several oncopro teins, including the Notch intracellular domain (NICD), which regulates cell survival and apoptosis (15). Therefore, E. chaffeensis induced Notch signaling, and increased levels of NICD during E. chaffeensis infection may play an important role in inhibiting apoptosis.
In this study, we reveal a novel mechanism whereby E. chaffeensis inhibits apoptosis through Notch activation resulting in NICD stabilization of XIAP. Inhibition of apopto sis through modulation of Notch signaling provides further evidence that E. chaffeen sis hijacks evolutionarily conserved signaling pathways to evade innate host defense mechanisms.

E. chaffeensis infection and TRP120 increases XIAP levels
We recently demonstrated that NICD levels temporally increase during E. chaffeensis infection (15). Increased levels of NICD were associated with TRP120 ubiquitination and degradation of a Notch negative regulator, FBW7. NICD is known to directly bind the XIAP BIR-RING domain and prevent XIAP autoubiquitination and degradation (47), thereby inhibiting apoptosis. To investigate potential XIAP upregulation during infection, THP-1 cells were incubated with E. chaffeensis [multiplicity of infection (MOI) 50], and XIAP protein and transcription levels were analyzed by immunoblot and qPCR. Levels of XIAP were unchanged in uninfected THP-1 cells (Fig. 1A). In comparison, increases in XIAP protein levels were demonstrated over the course of infection, with significant increases detected at 24, 48, and 72 h post-infection (hpi) (Fig. 1A). Interest ingly, additional cleaved fragment was observed at 48 and 72 hpi and identified as the BIR3-RING domain of XIAP (Fig. 1A) (48). When cleaved, BIR3-RING also acts as a potent inhibitor of the intrinsic apoptotic pathway by binding the Caspase-9 monomer preventing its cleavage and heterodimerization (48). Moreover, transcriptional levels of XIAP were also shown to be significantly upregulated in a temporal manner (Fig.  1B). XIAP expression was evaluated in THP-1 cells treated with recombinant TRP120 TR (rTRP120-TR) domain, or the recently described TRP120 Notch ligand memetic SLiM (TRP120-TR-P6) peptide (13). A significant temporal increase in XIAP levels was detected in cells treated with rTRP120-TR or TRP120-TR-P6 peptide ( Fig. 1C and E) compared to rTRX or TRP120-TR-P5-treated cells ( Fig. 1D and F), demonstrating E. chaffeensis TRP120 promotes increased XIAP levels.

Notch activation and XIAP stabilization by NICD
To confirm that the increase in XIAP levels was a direct result of Notch activation, THP-1 cells were pretreated with DAPT (5 µg/mL; 1 h), a Notch γ-secretase inhibitor. Cells pretreated with DAPT inhibitor and infected with E. chaffeensis (MOI 50) exhibited decreased XIAP levels at 24 and 48 hpi (Fig. 3A). Similar results were also shown with SAHM1 treatment, which prevents assembly of the active Notch transcriptional complex (Fig. S1). To further determine the direct relationship between NICD and XIAP levels during E. chaffeensis infection, uninfected and E. chaffeensisinfected THP-1 cells were pretreated alone or in combination with DAPT (5 µg/mL; 1 h) and SM-164 (100 nM; 12 h), a second mitochondrial activator of caspases (Smac/DIABLO) mimetic compound that antagonizes IAPs to promote activation of caspases and apoptosis (45). DAPT/SM-164 pretreatment was given prior to induction of cell death by tumor necrosis factor alpha (TNF-α; 100 ng/mL; 12 h), followed by subsequent infection with E. chaffeensis (MOI 50). interaction between XIAP and NICD at 24 hpi compared to the IgG-negative control. Western blot analysis was normalized to GAPDH expression. NICD and XIAP levels were measured from the Co-IP. Quantification of NICD or XIAP levels from one representative Co-IP experiment is shown. ns (not significant), P > 0.05, * P < 0.05, **P < 0.01, *** P < 0.001, **** P < 0.0001. Experiments were performed in triplicate (n = 3) and representative images are shown.

Full-Length Text Infection and Immunity
A 22% reduction in cell viability was observed by trypan blue exclusion in uninfected THP-1 cells treated with TNF-α alone (Fig. S2). Hematoxylin and eosin (H&E) staining of THP-1 cells treated with DAPT/SM-164 did not contain morulae, displayed significant cell death, and had significantly decreased ehrlichial infection compared to uninfected, E. chaffeensis-infected, dimethyl sulfoxide (DMSO) alone, or with DAPT or SM-164 alone treated cells ( Fig. 3B and C).

XIAP is required for antiapoptotic activity and increased infection
The effect of XIAP on E. chaffeensis infection was examined using small interfering RNA (siRNA) knockdown (KD) of XIAP in THP-1 cells. siRNA KD of XIAP resulted in a 77% KD efficiency (Fig. 4A). E. chaffeensis infection was significantly decreased in XIAP-KD cells (24 hpi) compared to scrambled control siRNA-treated cells ( Fig. 4B and C). To determine whether the decrease in ehrlichial load was caused by the induction of apoptosis due to XIAP destabilization, cell viability was determined by flow cytometry with the Muse Count & Viability Kit. XIAP-KD cells exhibited a viability of ~8% compared (D) Quantification of Caspase-3/CPP32 activity determined at an absorbance of 405 nm. Bar graphs represent mean ± SD. ns (not significant), ** P < 0.01, *** P < 0.001, **** P < 0.0001. Experiments were performed in triplicate and representative images are shown.

Full-Length Text
Infection and Immunity to 81% in scr-KD cells (Fig. 4D). Furthermore, XIAP-KD cells displayed morphological changes associated with apoptosis including shrinkage of the cell, fragmentation into membrane-bound apoptotic bodies, and nuclear fragmentation (Fig. 4D).

XIAP inhibits apoptosis during E. chaffeensis infection
Smac/DIABLO is a cytosolic antagonist of IAPs (50). To determine the significance of IAPs during E. chaffeensis infection, THP-1 cells were pretreated with SM-164 (100 nM; 12 h), a Smac/DIABLO mimetic compound. Cell death was induced by TNF-α (100 ng/mL; 12 h), followed by subsequent infection with E. chaffeensis. Treatment with SM-164 resulted in a significant reduction in ehrlichial infection as determined by confocal microscopy and qPCR of the dsb gene ( Fig. 5A and B). Importantly, cells treated in combination with SM-164/TNF-α had 10-20% viability and exhibited morphological changes in apoptosis including membrane blebbing, nuclear fragmentation, and cell shrinkage ( Fig. 5A and C; Fig. S3A). In comparison, untreated or DMSO-treated cells had cell viability ranging from 82% to 92% and 63% to 82%, respectively, with lower viability of DMSO-treated cells due to the addition of TNF-α ( Fig. 5C; Fig. S3A). In addition, there were unremarkable morphological changes associated with untreated or DMSO-treated cells (
To demonstrate that XIAP was directly associated with downstream caspase inhibition, THP-1 cells were pretreated with SM-164 (100 nM; 12 h) and TNF-α  (100 ng/mL; 12 h), infected with E. chaffeensis infection (MOI 50) and Caspase-9 and -3 levels were determined. The SM-164/TNF-α-treated cells had significantly lower levels of pro-Caspase-9 and -3, and significantly increased levels of cleaved (active) Caspase-9 and -3 (Fig. 6B). To demonstrate that XIAP was directly associated with downstream caspase inhibition, THP-1 cells were treated with SM-164 in combination with TNF-α, infected with E. chaffeensis infection (MOI 50) and multi-caspase levels were determined by flow cytometry. A significant increase in the percentage of caspase+/dead cells with SM-164/ TNF-α treatment was detected (Fig. 6C). These data demonstrate increased XIAP levels inhibit caspase activation and apoptosis during E. chaffeensis infection.

DISCUSSION
Inhibition of host cell apoptosis is an important survival strategy utilized by E. chaffeensis (33,52,53). Previous studies have demonstrated that the T4SS effector protein, ECH0825, inhibits host cell apoptosis in human monocytes (54). ECH0825 localizes to mitochondria and inhibits Bax-induced apoptosis by increasing mitochondrial MnSOD and reducing reactive oxygen species-mediated damage (54). Furthermore, upregulation of apoptotic inhibitor genes during E. chaffeensis infection, including BCL-2 and BIRC3, and downregu lation of apoptotic inducers, such as BIK, BNIP3L, and hematopoietic cell kinase have also been reported (33). However, there is little information related to E. chaffeensis modulation of caspase activation. In this study, we have identified a mechanism by which E. chaffeensis TRP120 effector activates the Notch signaling pathway to inhibit caspase-dependent apoptosis through NICD and XIAP interaction. We recently identified a TRP120 Notch SLiM ligand memetic motif responsible for Notch activation during E. chaffeensis infection (13). In addition, we demonstrated that E. chaffeensis and rTRP120 activate Notch signaling to downregulate TLR 2/4 expression for intracellular survival (5). TRP120 Notch activation was recently confirmed to occur through a TRP120 TR Notch SLiM ligand mimic that directly binds to the Notch recep tor in a region containing the known ligand binding region (13). Although a Notch SLiM mimic has been recently identified, its role in E. chaffeensis survival has not been fully elucidated. In this study, we investigated the functional implications of TRP120 Notch SLiM mimic activation of Notch signaling during E. chaffeensis infection. A novel antiapoptotic mechanism involving IAP proteins potently inhibiting the catalytic activity

Full-Length Text
Infection and Immunity of caspases through regulation of Notch signaling was defined and is illustrated in the graphical abstract (Fig. 8).
Pathogens have evolved various means of subverting innate immune defense of the host for survival (2,43,(55)(56)(57)(58)(59)(60). One of the most well-studied mechanisms of bacterial pathogens is targeting intracellular signal transduction cascades. Pathways such as MAPK and NF-κB are manipulated by various pathogens (39,(61)(62)(63). Manipulation of these innate immune and inflammatory pathways is regulated through bacterial effector proteins and host-pathogen interactions (3,64). In addition, inhibition of apoptosis is well documented as a mechanism used by bacteria and viruses to subvert innate immune defense (2,56,62,65,66). Although intracellular bacterial pathogens manipu late apoptosis by various mechanisms, exploitation of the Notch signaling pathway is a strategy that has not been reported.

Full-Length Text
Infection and Immunity Interestingly, Notch has been shown to inhibit apoptosis by directly interfering with the ubiquitination of the most potent inhibitor of apoptosis, XIAP (47). NICD directly binds the BIR3-RING domain of XIAP to inhibit autoubiquitination. Inhibition of XIAP autoubiquitination results in stabilization of XIAP levels, leading to inhibition of apoptosis (47). In this study, we demonstrated increases in XIAP expression over the course of E. chaffeensis infection. A cleavage product was observed at later time points which correlated with the XIAP BIR3-RING domain. Studies have demonstrated this domain to be a potent inhibitor of the intrinsic apoptotic pathway through direct binding to Caspase-9 (48). XIAP sequesters Caspase-9 in a monomeric state, which serves to prevent catalytic activity (76). Previous studies have demonstrated that E. chaffeensis inhibits apoptosis through the intrinsic apoptotic pathway by blocking the BCL-2 pathway (33). In addition, we have recently demonstrated that engagement of the BCL-2 antiapoptotic cellular programming during E. chaffeensis infection is caused by activation of the Hedgehog signaling pathway (2). Induction of BCL-2 resulted in inhibition of Caspase-3 and -9, preventing activation of intrinsic apoptosis (2). Therefore, evidence supports the inhibition of intrinsic apoptosis as a survival mechanism for E. chaffeensis. Importantly, both rTRP120 and the TRP120 Notch SLiM ligand memetic peptide upregulated XIAP expression in a time-dependent manner, with significant upregulation occurring at later time points, as demonstrated with E. chaffeensis-infected cells. These data support the role of the TRP120 induced Notch signaling activation leading to increased XIAP levels. Full-Length Text NICD and XIAP levels increased simultaneously and temporally. We have previously determined that NICD levels increase during infection, attributed in part to E. chaffeensis TRP120 ubiquitinating and degrading the Notch negative regulator, FBW7 (15). Here, we demonstrate NICD to both colocalize and directly bind with XIAP at later time points of E. chaffeensis infection. Interestingly, colocalization of NICD and XIAP occurred in both the cytoplasm and the nucleus during E. chaffeensis infection. Selective localization of pro-caspases in different subcellular compartments has been previously demonstrated (77). Pro-caspase and active Caspase-3, -7, and -9 are mainly found in the cytosolic fraction; however, Caspase-3 and -9 are also found in the mitochondrial fraction, while Caspase-7 is also found in the microsomal fraction in untreated Jurkat T lymphocytes (77). Caspase-3 is the only major caspase found in the nucleus (77). XIAP expression has been found to be mainly cytoplasmic; however, it is also present in the nucleus in specific cell types. XIAP nuclear translocation has been previously associated with aberrant cell division and anchorage-independent growth (78,79). Previous data have demonstrated that increases in XIAP are not associated with stimulation of XIAP transcription by NICD (47). Therefore, further investigation is needed to determine the functional implications of NICD/XIAP colocalization in the nucleus during E. chaffeensis infection (47). Inhibition of Notch activation by DAPT, a γ-secretase inhibitor, reversed increases in XIAP levels during E. chaffeensis infection. DAPT inhibits Notch receptor enzymatic hydrolysis, NICD release, and downstream transcriptional activation by inhibiting γ-secretase activity. Hence, inhibition of NICD is directly associated with decreases in XIAP levels.
Apoptosis is an important innate defense mechanism against microbial infection; however, various intracellular pathogens hijack apoptosis by inhibiting either extrinsic or intrinsic apoptosis through different mechanisms (2,(80)(81)(82). We demonstrated the importance of XIAP expression in inhibition of apoptosis during E. chaffeensis infec tion. siRNA KD of XIAP significantly reduced E. chaffeensis infection. This finding was associated with apoptosis, as demonstrated by Muse Count & Viability Assays and microscopy demonstrating cellular morphological hallmarks of apoptosis. siRNA-treated cells showed significant cell blebbing, shrinkage of the cell, and nuclear fragmentation. XIAP siRNA-treated cells contained a significant reduction in morulae compared to scrambled siRNA (scrRNA) cells. Interestingly, Anaplasma phagocytophilum also appears to inhibit apoptosis by preventing XIAP degradation (43). Cleaved fragments of XIAP were not detected in A. phagocytophilum-infected neutrophils (43), suggesting that XIAP degradation is blocked during A. phagocytophilum infection. In contrast, we detected an increase in XIAP cleavage product (30 kDa), which we identified as XIAP BIR3-RING. As previously stated, the XIAP BIR3-RING cleavage product has been shown to strongly inhibit intrinsic apoptosis. Differences in the presence of the cleavage product observed between E. chaffeensis and A. phagocytophilum are not well understood and need further investigation. These findings indicate that modulating IAPs to inhibit apoptosis may be a conserved mechanism utilized by various intracellular bacterial pathogens for survival.
Various studies have demonstrated XIAP as the most potent endogenous inhibi tor of caspases due to weaker binding and inhibition of caspases by other IAP pro teins. Interestingly, XIAP has been shown to inhibit both the executioner and intrinsic apoptotic pathways using various domains found within its structure. XIAP inhibits the executioner pathway by directly binding to Caspase-3 and -7 through the linker region between the BIR1 and BIR2 domains (51). As previously mentioned, XIAP also directly binds to Caspase-9 via the BIR3 domain (45,46,76). We have demonstrated levels of pro-Caspase-3,-7, and -9 temporally increase during E. chaffeensis infection. However, there were only minor changes in Caspase-8 levels during infection, demonstrating that mitochondrial-mediated apoptosis is the predominantly targeted for inhibition. In comparison, inhibition of Caspase-8 activation and Bid cleavage has been demonstrated in A. phagocytophilum-infected human neutrophils (43).
Previous data have shown that transcriptional levels of caspases do not change at earlier timepoints during E. chaffeensis infection (83). In comparison, our data show transcriptional regulation at 12 hpi. Pro-Caspase-3 transcript and protein levels increased at 24 hpi but protein levels decreased at 48 and 72 hpi. Cleavage of pro-Caspase-3 levels coincides with BIR3-RING cleavage products observed at 48 and 72 hpi. Activated Caspase-3 has previously been demonstrated to cleave XIAP (48), resulting in BIR1-2 and BIR3-RING fragments. The BIR1-2 fragment inhibits Caspase-3 and -7; however, BIR1-2 is a less potent IAP than full-length XIAP and may also be susceptible to further degra dation (84). In comparison, the BIR3-RING fragment blocks activation of Caspase-9 by directly binding to and inhibiting activity (45,46,76). Therefore, activation of Caspase-3 may lead to XIAP BIR3-RING fragments that inhibit intrinsic apoptosis through direct Caspase-9 binding; however, more studies are needed to fully elucidate this mechanism. Interestingly, similar evidence of inhibition of Caspase-3 and -9 activation during A. phagocytophilum infection has been reported where activation of Caspase-3 and -9 was linked to inhibition of XIAP degradation (43).
Inhibition of caspase activation by XIAP is mediated by the endogenous IAP inhibitor, SMAC/Diablo. During induction of apoptosis, SMAC/Diablo is processed and released from the mitochondria where it binds to the BIR2 and BIR3 domains of XIAP to antag onize XIAP activity (45,50). SM-164 is a bivalent, SMAC mimetic that induces apopto sis (85). Treatment with SM-164 in the presence of TNF-α significantly reduced cell viability and ehrlichial load. An increase in caspase-positive apoptotic cells was shown with SM164/TNF-α treatment. Importantly, SM-164/TNF-α-treated cells pre-treated with Caspase-9 inhibitor, Z-LEHD-FMK TFA, blocked full induction of apoptosis during E. chaffeensis infection. TNF-α has been demonstrated to inhibit apoptosis through both the extrinsic and intrinsic apoptotic pathways. Activation of the extrinsic pathway results in the cleavage of cytosolic BID to truncated p15 BID (tBID), which translocates to mitochondria and triggers cytochrome c release (86). Therefore, reversal of cell death by Caspase-9 inhibitor, Z-LEHD-FMK TFA, may occur through BID activation. In addition, SM-164 treatment resulted in decreased pro-caspase and increased levels of cleaved Caspase-9 and -3 during infection. These results demonstrate direct correlation of XIAP activity and caspase inhibition during E. chaffeensis infection. Previous studies have shown that the reduction in XIAP either does not occur or takes place at later time points in various cancer cell models, several hours after robust apoptosis induction, suggesting that the degradation of XIAP is not required for apoptosis induction by Smac mimetics (45,85,87). Therefore, in our model, induction of apoptosis is dependent on the activity of XIAP on caspase inhibition alone, rather than XIAP degradation.
During E. chaffeensis infection, TRP120 ubiquitinates Notch negative regulator, FBW7, resulting in degradation (15). Degradation of FBW7 is known to result in increased NICD levels (88). HeLa cells transfected with TRP120-C520S catalytic mutant displayed an increase in cleaved Caspase-3 and -9, and a decrease in XIAP levels, demonstrat ing a direct relationship between FBW7 stabilization of NICD levels and subsequent increased XIAP expression and caspase inhibition. Collectively, this study serves to provide insight into the molecular details of how TRP120 Notch signaling leads to increased XIAP expression through direct interaction with NICD, leading to inhibition of caspase activation and apoptosis for E. chaffeensis survival.
There are multiple questions that remain to be answered regarding E. chaffeensis regulation of apoptosis. IAP proteins have been previously shown to interact with one another to form IAP-IAP complexes that inhibit apoptosis (89). Many of the IAP-IAP complexes consist of one or more of four key IAPs: c-IAP1, c-IAP2, XIAP, and survivin. Whether XIAP functions in a complex with other IAPs to inhibit apoptosis during E. chaffeensis infection remains unknown. Evolutionarily conserved signaling pathways, such as Notch and Hedgehog, play key roles in regulation of apoptosis (35,90,91). TRP120 has been demonstrated to inhibit apoptosis by activation of Hedgehog signaling (2). Further investigation is needed to understand potential crosstalk between Notch, Hedgehog, and potentially other signaling pathways that are activated during E. chaffeensis infection and associated with apoptosis regulation.
In conclusion, we demonstrated E. chaffeensis Notch activation results in an XIAPmediated antiapoptotic program. Our findings reveal an E. chaffeensis initiated, Notch signaling regulated, antiapoptotic mechanism involving IAP proteins that inhibit caspase activation (Fig. 8). This study gives further insight into the molecular mechanisms used by obligate intracellular pathogens to exploit conserved signaling pathways to suppress innate defenses and promote infection.

Immunoblot analysis
Cells were infected or treated as indicated in text and figure legends and subsequently lysed with Triton-X 100 supplemented with protease inhibitor cocktail, Halt phosphatase, and phenylmethylsulfonyl fluoride for 30 min, with lysing by pipetting every 10 min on ice. Lysates were cleared by centrifugation at 14,000 × g (4°C) for 20 min. Protein concentration of cleared lysates was determined by bicinchoninic acid assay. Laemelli buffer was added to lysates then boiled for 5 min at 95°C. Lysates were then subjected to SDS-PAGE and transferred to nitrocellulose membrane. Membranes were blocked using 5% nonfat milk in Tris-buffered saline with 0.1% Tween 20 (TBST) and then exposed to α-XIAP, α-NICD, α-Casp-3,-7, or -9 or α-GAPDH antibodies overnight at 4°C. Membranes were washed thrice in Tris-buffered saline containing 1% Triton for a total of 30 min followed by 1 h of incubation with horseradish peroxidase-conjugated antirabbit or antimouse secondary antibodies (SeraCare, Milford, MA, USA) (diluted 1:10,000 in 1% nonfat milk in TBST). Proteins were visualized with ECL using a ChemiDoc It 2 imager (UVP) and densitometry performed with VisionWorks software (ver. 8.1). All blots are presented in their original form but may have loss of resolution due to the image compression from the dock system, Chemidoc-it2 Imager.

Co-IP
Magna ChIP™ A/G Chromatin Immunoprecipitation kit (MilliporeSigma, Burlington, MA, USA) was used to investigate XIAP and NICD interactions during E. chaffeensis infection. Briefly, THP-1 cells were infected with E. chaffeensis (MOI 100) or left uninfected (control) for 24 h. Cells were harvested, and Co-IP was performed, according to the manufacturer's protocol. XIAP and NICD antibodies (Cell Signaling Technology) were used to determine interactions. IgG purified from normal serum was used as control antibody. Bound antigen was eluted, solubilized in 4x SDS sample loading buffer, and processed for immunoblot analysis. The membrane was probed with XIAP or NICD antibody to confirm pulldown. Co-IP was performed in triplicate experiments.

Transfection
HeLa cells (1 × 10 6 ) were seeded in a 60 mm culture dish 24 h prior to transfec tion. All proteins were expressed in a pcDNA3.1+C-6His vector. TRP120 full-length (pcDNA3.1+TRP120_FL_C-6His) and its HECT Ub ligase catalytic inactive mutant (pcDNA3.1+TRP120_C520S_C-6His) were cloned into the pcDNA3.1+C-6His vector at NheI/XbaI sites. pcDNA3.1+C-6His empty vector was used as a control. All vectors were added to Opti-MEM and Lipofectamine 3000 mixture and incubated for 20 min at 37°C. Lipofectamine/plasmid mixtures were added to HeLa cells and incubated for 4 h at 37°C. The medium was aspirated 4 h post-transfection and fresh medium was added to each plate and incubated for 24 h.

Immunofluorescent confocal microscopy
THP-1 cells (1 × 10 6 ) were infected with E. chaffeensis (MOI 100) for indicated time intervals at 37°C. Cells were collected and fixed using ice-cold 4% formaldehyde and washed with sterile 1× phosphate-buffered saline (PBS) five times for 5 min. Cell samples were permeabilized and blocked in 0.5% Triton X-100 and 2% bovine serum albumin in PBS for 30 min. Cells were washed with sterile 1× PBS three times for 5 min and probed with XIAP, NICD, or DSB antibodies for 1 h at room temperature. Cells were washed with sterile 1 × PBST (0.1% Tween) three times for 5 min and probed with Alexa Fluor IgG (H + L) or Alexa Fluor IgG (H + L) for 30 min at room temperature, washed three times with sterile 1× PBST, and mounted with ProLong Gold antifade reagent with DAPI (Molecular Probes). Slides were imaged on a Zeiss LSM 880 confocal laser scanning microscopy. Mander's correlation coefficients were generated by ImageJ software to quantify the degree of colocalization between fluorophores.

TRP120 recombinant protein and peptide treatment
Recombinant TRP120 containing the TR or thioredoxin (TRX) was expressed in a pBAD expression vector and purified as previously published (93)(94)(95). rTRP120-TR was dialyzed in 1× PBS and tested for bacterial endotoxins using the limulus amebocyte lysate test. TRP120 synthetic peptides were produced commercially (Genscript, Piscataway, NJ, USA). THP-1 cells were treated with 2 µg/mL of rTRP120-TR or TRX, or 1 µg/mL of synthetic TRP120 peptides for 0-72 h time points. Cells were collected post-treatment and immunoblot and qPCR analysis was performed.

RNAi KD
Cells (1.0 × 10 6 ) were transfected with ON-TARGETplus SMARTpool XIAP siRNA (3 µL) (Dharmacon, Lafayette, CO, USA) using Lipofectamine 3000 (7.5 µL) (Invitrogen, Waltham, MA, USA), according to the manufacturer's instructions. Scrambled siRNA was utilized as a control in both uninfected and E. chaffeensisinfected THP-1 cells. The siRNA and Lipofectamine mixture were added to 250 µL of MEM medium (Invitrogen), incuba ted for 12 min at room temperature and added to cells in a 6-well plate. KD was assessed by immunoblot analysis as previously described. Cells were knocked down for 24 h and infected with E. chaffeensis (MOI 100) for 24 h. Cells were then harvested after 24 hpi. Proliferative/cell death analysis was performed on all KD cells. Ehrlichial load was determined by qPCR of dsb gene as previously described (8). All siRNA KDs were performed with triplicate technical and biological replicates and significance was determined using a t-test analysis.
1. Trypan blue exclusion. Cell samples (20 µL) were collected and mixed with an equal volume of trypan blue. Samples were incubated at room temperature for 2 min and then read using the Nexcelom Cellometer Mini (Nexcelom Bioscience LLC, Lawrence, MA, USA). 2. Caspase-3/CPP32 Assay Kit. Caspase-3/CPP32 Assay Kit (Colorimetric) [NBP2-54838] was utilized to assess the activity of Caspase-3, according to the manufacturers' protocol. 3. H&E. H&E stain was utilized to assess cell morphological changes associated with cell death. Cells were collected and washed with 1× DPBS. Fresh RPMI media were added to the cell samples and fixed onto slides by cytospin (800 x g, 5 min). Cells were then fixed by acetone (1 min) and stained with H&E (1 min/stain). Slides were rinsed with DI water and dried prior to analysis using light microscopy. Images were taken using the Olympus CellSens software. 4. Guava Muse Cell Analyzer. Various Muse assays were utilized according to the manufacturers' protocol to determine apoptosis: a. The Muse Count & Viability Kit was used to determine cell counts (cells/mL) and viability (%) (Part Number MCH100102). b. The Muse MultiCaspase Kit (Part Number: MCH100109) was used to determine caspase activation and cellular plasma membrane permeabiliza tion, or cell death. The percentage of live, caspase+, caspase+ and dead, total caspase+, and dead cells was determined. c. The Muse Annexin V & Dead Cell Kit (Part Number: MCH100105) was used to determine live, early, and late apoptosis and cell death. The percentage of live, early apoptotic, late apoptotic, total apoptotic, and dead cells was determined.

Statistical analysis
All data are represented as the means ± standard deviation of data obtained from at least three independent experiments done with triplicate biological replicates. Experiments performed with technical replicates are indicated in figure legends and the material and methods section. Analyses were performed using a two-way ANOVA or two-tailed Student's t-test (GraphPad Prism 6 software, La Jolla, CA, USA). P < 0.05 was considered statistically significant.

AUTHOR CONTRIBUTIONS
LaNisha L. Patterson

ADDITIONAL FILES
The following material is available online.