Global mapping of the Chlamydia trachomatis conventional secreted effector – host interactome reveals CebN interacts with nucleoporins and Rae1 to impede STAT1 nuclear translocation

Chlamydia trachomatis (C.t.), the leading cause of bacterial sexually transmitted infections, employs a type III secretion system (T3SS) to translocate two classes of effectors, inclusion membrane proteins and conventional T3SS (cT3SS) effectors, into the host cell to counter host defense mechanisms. Here we employed three assays to directly evaluate secretion during infection, validating secretion for 23 cT3SS effectors. As bioinformatic analyses have been largely unrevealing, we conducted affinity purification-mass spectrometry to identify host targets and gain insights into the functions of these effectors, identifying high confidence interacting partners for 21 cT3SS effectors. We demonstrate that CebN localizes to the nuclear envelope in infected and bystander cells where it interacts with multiple nucleoporins and Rae1, blocking STAT1 nuclear import following IFN-γ stimulation. By building a cT3SS effector-host interactome, we have identified novel pathways that are targeted during bacterial infection and have begun to address how C.t. effectors combat cell autonomous immunity.


INTRODUCTION
To usurp host defense mechanisms and establish a favorable replicative niche, intracellular pathogens are tasked with remodeling the host cell using secreted virulence factors, termed effector proteins.Identification of the host proteins and pathways targeted by secreted proteins during active infections has been exceedingly difficult for obligate intracellular pathogens owing to their genetic intractability 1 .Several obligate intracellular pathogens, including Chlamydia trachomatis (C.t.), are the etiological agents of important human diseases for which no vaccines exist 2 .C.t. is the leading cause of infectious blindness and is the most common bacterial sexually transmitted infection worldwide 3 .Untreated infection can result in severe complications including pelvic inflammatory disease, ectopic pregnancy, sterility, and increased risk of developing cervical and ovarian cancer 4,5 .The incidence and prevalence of C.t. infections are rapidly rising, and reinfection rates are high due to a lack of long-term protective immunity and treatment failure following antibiotic therapy 2 .Understanding how C.t. co-opts the host cell to form its unique replicative niche is vital to developing new therapies.
All Chlamydiae share a biphasic developmental cycle in which the bacteria alternate between two forms: the infectious elementary body (EB) and the replicative reticulate body (RB) 3 .Upon contact with a target host cell, the EB delivers a set of pre-synthesized type III secretion system (T3SS) effector proteins into the eukaryotic cell to drive cytoskeletal rearrangements and membrane remodeling, triggering endocytosis of the pathogen [6][7][8][9] .The plasma membrane-derived compartment in which the EB resides is rapidly modified by the pathogen to form a unique replicative niche, termed the inclusion.The inclusion quickly dissociates from the endolysosomal pathway 10 , trafficking along microtubules to the peri-Golgi region 11 where the EB differentiates into an RB and initiates replication.Following multiple rounds of division, RBs convert to EBs, and the bacteria are released by lysis or extrusion to begin the infection cycle anew 12 .While it is well established that formation of an intact replicative niche is vital for C.t. proliferation and chlamydial disease 13,14 , how C.t. accomplishes such feats remains incompletely understood.
Despite its extremely small genome (1Mb), C.t. is predicted to secrete an astonishing 10-15% of its proteome through the T3SS 1,15 .A large subset of these secreted proteins, termed inclusion membrane proteins (Incs), possess one or two bi-lobed hydrophobic domains of ~40 amino acids, allowing for intercalation into the inclusion membrane in such a way that the Nand C-termini of the protein are oriented into the host cell cytosol 16 .Inc proteins play a vital role at the host-pathogen interface; however, the other subset of effectors, conventional T3SS (cT3SS) effector proteins, are secreted into the host cell cytosol and likely also play vital roles in chlamydial infection at distal sites from the inclusion.Identification and functional characterization of cT3SS effector proteins has been challenging as most lack functional domains indicative of effector function, and they do not share homology to proteins with known functions.Until recently, Chlamydiae were recalcitrant to genetic manipulation, necessitating a reliance on surrogate systems to identify candidate secreted effectors 1 .To date, over 40 cT3SS effector proteins have been shown to be secreted in heterologous systems that employ Yersinia pseudotuberculosis, Shigella flexneri, or Salmonella enterica serovar Typhimurium as surrogate hosts [17][18][19] .While still challenging, advances in Chlamydial genetics, including the ability to transform C.t. with a stably maintained plasmid that enables inducible expression of epitope-tagged proteins and the adoption of multiple reporter constructs have allowed for validation of effector protein secretion during infection [20][21][22] .Significantly, several recent studies have confirmed that while useful, secretion by a surrogate organism does not necessarily correlate with secretion in the native organism 20,23 , necessitating validation of secretion by C.t.
Here we leveraged cutting-edge chlamydial genetics, in conjunction with large-scale unbiased affinity purification-mass spectrometry (AP-MS), to comprehensively define which cT3SS proteins are secreted by C.t. during infection and to map the host pathways targeted by these effector proteins.We identified 23 proteins that were secreted by C.t. in at least one assay and identified high-confidence interacting partners for 21 cT3SS proteins.Intriguingly, we show that CT584, which we have renamed Chlamydia effector blocking nuclear transport (CebN), binds to a subset of Phe-Gly (FG) nucleoporins (NUPs) and the mRNA export factor Rae1targets previously associated with viral infection [24][25][26][27][28] .Our data indicate that CebN predominately localizes to the nuclear envelope in both infected and bystander cells and is sufficient to inhibit STAT1 import into the nucleus following interferon (IFN)-γ stimulation.This work significantly contributes to our understanding of C.t. cT3SS effectors and their host targets, providing a key stepping-stone for elucidating how these effector-host interactions contribute to the pathogenesis of C.t. infections.

MATERIALS AND METHODS
Bacterial and cell culture.Chlamydia trachomatis serovar L2 (LGV 434/Bu) was propagated in HeLa 229 cells (American Type Culture Collection) and EBs were purified using a gastrografin density gradient as previously described 29 .HeLa cells were grown in RPMI 1640 medium (Thermo Fisher Scientific) supplemented with 10% Fetal Bovine Serum (Gibco) at 37°C with 5% CO 2 .
Plasmid construction.To assess secretion of candidate cT3SS effectors, chlamydial genes were PCR amplified from L2/434/Bu genomic DNA and each orf was cloned into the NotI/KpnI site of pBomb4 CyaA, pBomb4 BlaM, and pBomb4 GSK-FLAG 20 S1.

Transformation of C.t. C. trachomatis serovar L2 (
LGV 434/Bu) EBs were transformed as previously described 30 with minor modifications.Briefly, plasmid DNA (5 µg), fresh C.t. lysates (~2 x10 6 EBs), and 10 µl 5X transformation mix (50 mM Tris pH 7.4 and 250 mM CaCl 2 ) were gently mixed and the final volume was adjusted to 50 µl with tissue-culture grade water.Mixtures were incubated at room temperature for 30 min.RPMI with 10% FBS (4 ml) was then added to each transformation mix and 2 ml was applied to 2 wells of a 6-well plate containing a confluent HeLa cell monolayer.Plates were centrifuged at 900 x g for 30 min and at 18h post-infection, the media was replaced with RPMI with 10% FBS containing 0.3 µg/ml penicillin G (PenG).Infectious progeny were harvested every 48h and used for infections of a fresh HeLa cell monolayer until viable inclusions were evident (~2-3 passages).Expression of individual fusion proteins was confirmed by western blotting.
Adenylate Cyclase (CyaA) secretion assay.HeLa cell monolayers were infected at an MOI of 5 with C.t. transformants, and expression of the effector-CyaA fusion protein was induced using anhydrotetracycline hydrochloride (aTc; 10 ng/ml) as previously described 20 .
cAMP production in host cells was quantified by ELISA according to the manufacturer's guidelines (Abcam).Effector secretion was determined by comparing the levels of cAMP in cells infected with C.t. pBomb4 CyaA (negative control vector) to those infected with the C.t. CyaAeffector fusion strains.
Beta-lactamase (BlaM) assay.HeLa cells (2 x 10 4 /well) were seeded into black, clear bottom 96-well plates (Greiner).Cell monolayers were infected at an MOI of 5 and effector-βlactamase (BlaM) fusion protein expression was induced using 10 ng/ml aTc as previously described 20 .At 24h post-infection (hpi), cells were washed three times with 1X PBS and loaded with CCF4-AM using the alternative loading protocol per the manufacturer's instructions (ThermoFisher Scientific).Samples were incubated in the dark for 1h at room temperature and were then read on a Tecan Infinite M200 Pro plate reader.To quantify effector translocation, the background was subtracted, the ratio of 460nm to 535nm (blue:green) was determined, and expression relative to cells infected with C.t. pBomb4 BlaM (negative control vector) was calculated as previously described 31 .

Immunofluorescence (IF) microscopy.
To determine the subcellular localization of secreted effector proteins, HeLa cells were transfected with pcDNA3.1-GFPplasmids using Lipofectamine LTX (Thermo Fisher Scientific).Cells were fixed with 4% formaldehyde 18h post-transfection and the nucleus was stained using DAPI (Invitrogen).Images were captured on a Nikon Ti2 immunofluorescent microscope.
For visualization of CebN by stimulated emission depletion (STED) microscopy, HeLa cells were transfected with pcDNA3.1-GFPplasmids containing an empty vector or CebN.Cells were fixed with 2% formaldehyde and permeabilized with 0.1% Triton-X 100 at 24h posttransfection and stained with DAPI and NUP specific antibodies: anti-NUP54, anti-NUP153, or anti-NUP214.Images were captured on a Leica SP8 inverted microscope.Images were deconvoluted using Imaris Professional Software.
For visualization of CebN during infection, HeLa cells were infected at an MOI of 2 with WT C.t. or C.t. strains expressing a FLAG-tagged empty vector, CebN-FLAG, or TmeA-FLAG.
Expression was induced for 24h using 10 ng/ml aTc added at the time of infection.Cells were fixed with 4% formaldehyde 24h post-infection and stained with DAPI, anti-FLAG (Cell Signaling), and anti-C.t.HSP60 (Sigma).For STAT1 translocation experiments, HeLa cells were transfected with pcDNA3.1-GFPplasmids containing empty vector, CebN, or TmeA using Lipofectamine LTX (Thermo Fisher Scientific).Six h post transfection, the media was changed with half the samples receiving normal RPMI media and half with RPMI plus 600U/ml IFN-γ as previously described 32 .Cells were fixed with 4% formaldehyde 18h post-transfection and stained with DAPI and anti-STAT1 (Cell Signaling) antibody.Images were captured using a Leica DFC7000T confocal microscope equipped with Leica software.Nuclear translocation of STAT1 was quantified from 15 images per coverslip with three coverslips per biological replicate.
Affinity Purification (AP): HeLa cells, in three T175 flasks, were infected at an MOI of 2 with C.t. strains expressing a FLAG-tagged effector protein.Expression was induced for 24h using 10 ng/ml aTc, added at the time of infection.Four h prior to lysis, 10 µM MG132 (Millipore Sigma) was added to the media.Cells were subsequently lysed in ELS with Halt cocktail protease inhibitor.Lysates were centrifuged at 12,000 x g for 20 min, and the supernatants were incubated with 60 µl preclearing beads (mouse IgG agarose, Millipore Sigma) for 2h at 4°C.The precleared lysate was then incubated with 30 µl FLAG beads (anti-FLAG M2 Affinity Gel, Millipore Sigma) overnight at 4°C.The beads were washed six times with ELS without detergent.For mass spectrometry, samples were stored in 50 mM ammonium bicarbonate prior to digestion and analysis as previously described 33 .For western blotting, proteins were eluted from the beads in 4X NuPAGE LDS Sample Buffer (Thermo Fisher Scientific) and boiled for 5 minutes.

Mass Spectrometry (MS).
MS was performed as previously described 33 , with the following adjustments.Beads containing samples were washed with 25 mM ammonium bicarbonate and digested with 0.5 micrograms trypsin (Pierce, Thermo Fisher Scientific, MS Grade) using a CEM microwave reactor for 30 min at 55°C.Digested peptides were extracted twice using 50% acetonitrile plus 5% formic acid, lyophilized to dryness, and resuspended in 5% acetonitrile plus 0.1% formic acid.For LC/MS, samples were injected into an UltiMate 3000 UHPLC system coupled online to a Thermo Scientific Orbitrap Fusion Tribrid mass spectrometer.Peptides were separated by reversed-phase chromatography using a 50-cm MicroPac Nano C18 column (Thermo Fisher Scientific) with mobile phases of 0.1% formic acid in water and 0.1% formic acid in acetonitrile; a linear gradient from 4% to 35% Acetonitrile over the course of 45 min was employed for peptide separations.The mass spectrometer was operated in a data-dependent acquisition (DDA) mode, employing precursor scans from 300 to 1,500 m/z (120,000 resolution) followed by collision induced dissociation (CID) of the most intense precursors over a maximum cycle time of 3 s (35% NCE, 1.6 m/z isolation window, 60-s dynamic exclusion window).Raw LC-MS/MS data were converted to peak lists using Mascot Distiller 2.8 and searched against a database containing UniProt_Human and Chlamydia_trachomatis_L2434Bu using Mascot 2.8 (Matrix Science).Tryptic digestion was specified with a maximum of two missed cleavages, while peptide and fragment mass tolerances were set to 10 ppm and 0.6 Daltons, respectively.Label-free Quantitation was performed utilizing the Mascot Average method on Mascot Distiller 2.8.2.
Gene Ontology and pathway analysis.Localization, Reactome pathway, biological process, and molecular function were determined for each host prey with a MiST 34 score ≥0.69 using Uniprot and GeneCards.Dot plots for visualization were generated using R package ggplot2.STRING (12.0) was used to generate protein-protein interaction networks for each effector using those host prey with a MiST score ≥0.69.Cytoscape (3.10.1) was used for visualization of STRING networks.
Statistics: Statistical analysis was performed using GraphPad Prism 10.1.1 software.

Identification of cT3SS effector proteins that are secreted during C.t. infection. Over 29 putative C.t. cT3SS effectors have been identified by their T3SS-dependent secretion in
genetically tractable surrogate bacterial hosts [17][18][19]35,36 . Althogh useful, screens using surrogate hosts can yield false positives and negatives 20,23 .Thus, we applied newly developed C.t. genetics to directly test whether 29 candidate effectors (Table 1) are secreted into the host cell during C.t.
infection.Additionally, we included several candidates from prior reports that were listed as negative for secretion due to incorrect product size during western blotting (CT016), low protein expression (CT309), or inconclusive results (CT330, CT338, CT386, CT504, and CT631) 36 .We also included TmeA and CteG in our assays as positive controls as they have been confirmed to be secreted by C.t 21,22 .By applying 3 standard secretion assays that fuse the candidate protein of interest to an secretion dependent reporter construct, adenylate cyclase (CyaA), β-lactamase (BlaM), and GSK-FLAG (glycogen synthase kinase FLAG-tag) 20,21,23,37 , we sought to evaluate the secretion of 36 candidate secretion substrates during C.t. infection (Table 1).Prior to conducting the CyaA and BlaM reporter assays, expression of each candidate effector was confirmed by western blotting (Fig. S1).While we sought to evaluate secretion for all candidates in each of the 3 assays, some were not tested due to the inability to generate a clone, lack of C.t.

transformants, or lack of expression (Table 1).
Using TmeA-CyaA and CteG-CyaA as positive controls, we demonstrate that infection of HeLa cells with these C.t. strains yield significantly elevated intracellular levels of cAMP relative to cells infected with C.t. expressing CyaA alone (Fig. 1A).These results confirm that secretion of C.t. cT3SS effectors can readily be detected using this approach and can be applied to identify other secreted factors.Applying this approach to the candidate secretion substrates, we show that CT016, CT053, CT143, CT144, CT161, CT606.1,CT621, CT671, and CT711 are capable of being translocated into the host cell during C.t. infection (Fig. 1A, Table 1).To evaluate candidate effector secretion using a second assay, we fused BlaM to the C-terminus of each candidate and confirmed the functionality of this assay using TmeA-BlaM as a positive control (Fig. 1B), which was previously shown to be secreted using this assay 21 .Here we demonstrate that CT053, CT144, CT620, CT622, CT652.1,CT656, CT671, CT738, and CT849 are secreted by C.t. using the BlaM assay (Fig. 1B, Table 1).Lastly, we employed the GSK assay to evaluate effector secretion as this assay has the advantage of using a small, 13-residue tag that can be phosphorylated in the host cell cytosol.Using this approach, we demonstrate that CT016, CT053, CT142, CT143, CT144, CT161, CT311, CT386, CT504, CT583, CT620, CT621, CT622, CT631, CT671, CT711, CT712, CT738, and CT848 are secreted (Fig. 1C, Table 1).The GSK-fusion protein assay appeared to be the most sensitive, likely owing to the smaller size of the tag allowing for more efficient translocation.For subsequent analysis, we required that the effectors be secreted in at least 2 assays to be classified as "secreted," in one assay to be identified as "possibly secreted," and "not secreted" if negative in at least 2 assays and not positive in any assay.Collectively, of the 36 candidates tested, we identified 11 secreted and 12 possibly secreted effectors (Table 1).

Subcellular localization of transfected cT3SS effector-GFP fusion proteins.
Following translocation into a host cell, many effector proteins localize to specific subcellular compartments where they interact with host proteins [38][39][40][41] .To identify host organelles targeted by the 11 secreted and 12 possibly secreted effectors, as well as CebN (which we have previously shown to be secreted 20 ), we used IF microscopy.Transient transfection of HeLa cells with GFPfusion proteins revealed that 12/23 cT3SS effectors displayed a pattern distinct from GFP alone (Fig 1D).CT311, CebN, and CT652.1 localized to the nucleus, with CebN being enriched at the nuclear envelope.CT620, CT621, CT622, CT631, CT711, and CT712 were excluded from the nucleus and displayed a cytosolic pattern.CT142, CT738, and CT849 also localized to the cytoplasm, but accumulated in punctate-like structures (Fig. 1D).

Identification of putative host proteins and pathways targeted by cT3SS effector proteins during C.t. infection. A few cT3SS effector proteins have been functionally
characterized and shown to modulate diverse host cell signaling pathways 6,8,9,19,33,[42][43][44] .However, the function of most of these putative virulence factors remains unknown.Affinity purificationmass spectrometry (AP-MS) has emerged as a powerful technique to comprehensively map protein-protein interactions (PPIs) between bacterial effectors or viral proteins and host proteins, yielding key mechanistic insights into how these pathogens establish their unique replicative niches 34,[45][46][47][48] .While informative, most of these studies have been undertaken by overexpressing a single effector protein in a mammalian cell at non-physiological levels and in the absence of additional bacterial or viral factors that might promote or hinder PPIs.With the increasing genetic tractability of C.t., we are poised to evaluate effector-host PPIs in the context of infection.
Here we sought to perform AP-MS on a total of 33 cT3SS effectors: 23 from this study, 9 additional recently identified effectors 20 , and TmeB 21 .Of these, 24 were successfully expressed in C.t. and the remaining effector proteins were excluded from further analysis due to the inability to obtain chlamydial transformants or due to the lack of detectable expression by western blotting.Only prey proteins with at least 2 unique peptides that were present in at least 2 of the replicates were further pursued.CT311 and CT161 were excluded from further analysis due to lack of detection of the bait protein following AP-MS.To identify high-confidence PPIs, we analyzed the complete data set using Mass Spectrometry interaction STatistics (MiST) (Table S3), which evaluates prey reproducibility, abundance, and specificity to generate scores between 0 and 1 34 .Using a cut-off score of ≥ 0.69, we identified 241 putative host interacting partners for 21 cT3SS effectors.While CT144 was detected in the AP-MS, and multiple preys were identified, none of these were predicted high-confidence interactions when scored by MiST.
To further define the potential function of the C.t. cT3SS effector proteins, Gene Ontology (Genecards) and pathway analysis (Reactome and Uniprot) was performed for each of the 241 MiST high-confidence interactors (Table S3) (Fig. 2A, B).As shown in Fig. 2A, most of the host proteins targeted by C.t. effectors reside in the cytoplasm, nucleus, endoplasmic reticulum, and mitochondria; however, a few host proteins associated with the Golgi apparatus, ribosomes, cytoskeleton, and plasma membrane were also noted.Pathway analysis revealed that these C.t. cT3SS effector proteins target host protein associated with vesicle-mediated transport, translation, stress response, mRNA splicing, and the immune system (Fig. 2B).In agreement with our pathway analysis, the majority of the MiST high-confidence interactors function in protein or RNA binding (Fig. S2A) to facilitate protein transport, transcription, or translation (Fig. S2B).
To further delineate potential effector function, we employed STRING and Cytoscape to map PPI networks and identify effector proteins that associated with multiprotein complexes (Fig. 2C, D, 3A, and S3).Through assessment of whether prey proteins that were pulled down by individual effector proteins interacted with one another we show that the cT3SS effector proteins CT620, CT386, and CebN associated with multiprotein complexes (Fig. 2C, D, 3A).Notably strong connections between proteins that are structural components of the ribosome and are important for translation were noted for CT620 (Fig. 2C), suggesting that this effector might target the translation machinery (Fig. 2B).Analysis of the predicted high confidence host protein interactors of CT386 identified a subset of RNA-binding proteins associated with mRNA processing (Fig. 2A, D), suggesting CT386 might also interfere with translation.
Ectopically expressed CebN binds to multiple nucleoporins and Rae1.Our infection AP-MS screen identified 68 high-confidence interactors for CebN (Table S3).Using STRING and Cytoscape, we assembled a PPI network map for CebN, revealing that many of its targets are involved in transport of ribonucleoproteins into the nucleus, mRNA export, or transcription (Fig. 3A).Transfected CebN-GFP predominately localized to the nuclear envelope (Fig. 1D), a pattern that aligns with it binding to host proteins involved in nucleocytoplasmic transport.To confirm these interactions, and to rule out the requirement of additional bacterial proteins that contribute to the CebN infection interactome, we performed AP-MS on Strep-tagged CebN as previously described 48 .This approach identified 30 high-confidence interactors (MIST ≥ 0.69) for CebN (Table S3), of which 19 overlapped with the infection IP (Table 2, Fig. 3B).Notably, 9 nucleoporins (NUP58, NUP214, NUP98, NUP54, NUP62, NUP88, NUP153, POM121/NUP121 and RANBP2/NUP358) and the mRNA export factor Rae1 were present in both the CebN infection and transfection interactomes.
Nucleoporins (NUPs) are a family of ~30 proteins that form the nuclear pore complex (NPC) and play an important role in regulating import and export of small molecules into and out of the nucleus 49 .The NPC is organized into an inner pore ring, the nuclear and cytoplasmic rings, the nuclear basket, and the cytoplasmic filaments, each of which are enriched for select NUPs (Fig. 3C) 49 .Intriguingly, while most of the NUPs that CebN binds make up the cytoplasmic filaments, interactions with NUPs in other subcomplexes of the NPC were noted (Fig. 3C), suggesting that secreted CebN may play a broad role in modulating NPC function.
Rae1 is an mRNA export factor that binds to NUP98 to aid in the transport of messenger ribonucleoprotein (mRNP) complexes through the nuclear pore complex 50 .Several viral proteins target NUPs and Rae1 to promote replication of their genomic information and to dampen the host response to infection by blocking import of important transcription factors [25][26][27][28] .To the best of our knowledge, no bacterial protein has been identified that targets NUP proteins or Rae1, making CebN an intriguing effector protein to study.
CebN binds to and co-localizes with NUPs and Rae1.To confirm CebN binding to NUP proteins and Rae1, we immunoprecipitated FLAG tagged CebN from C.t. infected cells and probed with antibodies specific to NUPs and Rae1.We focused on NUP54, NUP153, and NUP214 due to their high peptide counts in the AP-MS (Table S3).NUP54, NUP153, NUP214, and Rae1 IP with CebN but not with vector or TmeA (Fig. 3D), an effector previously shown to bind to N-WASP 6,8 .Processing of NUP153 and NUP214 was noted on these blots in infected samples.We determined that CPAF, a broad-spectrum protease produced by C.t. was responsible for this post-lysis cleavage, as this processing was absent in lysates derived from HeLa cells infected with a CPAF mutant 51 (Fig. S4).
To additionally confirm the interaction between CebN and NUPs, HeLa cells were transfected with GFP-CebN or GFP-vector, fixed, and stained using anti-NUP54, NUP153, and NUP214 antibodies.Imaging by stimulated emission depletion (STED) microscopy confirmed that CebN colocalized (white) with specific NUP proteins, whereas no colocalization was noted with GFP-vector (Fig. 4A, B).Pearson's correlation coefficient was calculated as a measure of colocalization, and a significant difference was found between GFP-vector and GFP-CebN transfected cells for each individual NUP (Fig. 4B).Taken together our results indicate that the cT3SS effector protein CebN localizes to the nuclear envelope where it binds to multiple NUPs and to Rae1.

CebN localizes to the nuclear envelope of infected and bystander cells. Most cT3SS
effector proteins are not readily visualized by microscopy, and thus their subcellular localization is generally assessed by transfection.Due to CebN's unique localization to the nuclear envelope, we assessed CebN localization directly in C.t. infected cells.In line with our ectopic expression data, CebN-FLAG was found to localize to the nuclear envelope of infected cells (Fig. 4C).Intriguingly, we also observed CebN on the nuclear envelopes of bystander cells (Fig. 4C), suggesting this effector can translocate into adjacent cells.

The C-terminus of CebN is required for interactions with NUPs and Rae1 as well as
for its localization to nuclear envelope.To delineate the region of CebN that is necessary for interaction with NUPs and Rae1, we generated 20-40 amino acid sequential truncations from the C-terminus of the 183 amino acid protein and expressed these truncations as FLAG-tagged constructs in C.t. Immunoprecipitation of these truncations, followed by subsequent western blotting, showed that the C-terminal 23 amino acids of CebN are necessary for interactions with NUP54, NUP153, NUP214, and Rae1 (Fig. 5A).We further confirmed the importance of this region by performing confocal microscopy on HeLa cells infected with the C.t. strains expressing the CebN FLAG-tagged deletion constructs (Fig 5B).As noted above, full length CebN-FLAG localizes to the nuclear envelope of infected cells, as well as the nuclear envelopes of bystander cells (Fig. 5B).However, none of the truncated versions of CebN localized to the nuclear envelope in infected or bystander cells (Fig. 5B).In total, the C-terminus of CebN is necessary for its interaction with NUPs and Rae1 and for its localization to the nuclear envelope in infected and bystander cells.

CebN attenuates STAT1 import into the nucleus following interferon-γ stimulation.
Viral proteins from HIV, SARS-CoV-2, Kaposi's sarcoma-associated herpesvirus, and vesicular stomatitis virus interact with and rearrange the nuclear pore complex to modulate nuclear import of transcription factors required for the anti-viral response 25,26,28,[52][53][54][55][56][57][58] .Similarly, C.t. attenuates nuclear import of STAT1 following IFN-γ stimulation 32 .We hypothesized that CebN interactions with NUPs and Rae1 could be responsible for perturbing STAT1 nuclear import.To test this, HeLa cells were transfected with GFP-CebN, GFP-empty vector, or GFP-TmeA, treated with IFN-γ, and imaged by confocal IF microscopy using an anti-STAT1 antibody (Fig. 6).We observed a significant decrease in the frequency of cells with nuclear translocated STAT1 between CebN-transfected cells compared to those cells transfected with empty vector or TmeA.Thus, CebN by itself is sufficient to block the translocation of STAT1 into the nucleus.Taken together, our results suggest that CebN, through interactions with nucleoporins and Rae1, plays a key role in dampening the host response to C.t. infection by blocking nuclear translocation of a key transcriptional regulator of the host innate immune response, STAT1.

DISCUSSION
In this study, we combined newly developed C.t. genetics with AP-MS to definitively identify C.t. cT3SS effectors and to generate the first cT3SS effector-host interactome.Our approach successfully identified high confidence interacting host partners for 21 of the 36 uncharacterized cT3SS effectors.Our work is especially valuable as it not only begins to build a compendium of proteins that are secreted into the host cell during active infection, but also provides a launch point for detailed mechanistic characterization of these effector proteins.
Importantly, screens such as AP-MS, can reveal novel pathways targeted by intracellular bacteria [45][46][47] .Here we discovered that CebN targets NUPs to inhibit STAT1 nuclear import, revealing a potential mechanism by which C.t. neutralizes the host innate immune response to survive intracellularly.Altogether, these studies will give us a better understanding of how obligate intracellular pathogens remodel the host to form their unique replicative niches.
One of the most striking findings from our AP-MS screen was the interaction between CebN and multiple nucleoporins and Rae1.Pointing to its importance in C.t. intracellular infection, CebN is 100% conserved amongst C.t. serovars, including trachoma (serovars A-C), urogenital (serovars D-K), and lymphogranuloma venereum (L1-L3) isolates and is highly conserved (83% identical and 91% similar) in C. pneumoniae 59 .Previous biophysical studies on CebN were interpreted as this protein associating with the tip of the T3SS needle apparatus 59 , whereas studies using Yersinia enterocolitica as a surrogate host indicated CebN might function as a chaperone 35 .Subsequent protein-protein interaction studies indicated that CebN interacts with core components of the T3SS, including the needle protein CdsF, the ATPase CdsN, and the C-ring CdsF 60,61 , suggesting it might be a T3SS effector protein.In agreement with this observation, we have shown that CebN is secreted into the host cell during C.t. infection 20 and in the current study, we begin to mechanistically dissect the role of CebN in co-opting NUP and Rae1 functions.
While our study is, to the best of our knowledge, the first time a bacterial effector has been shown to interact with host nucleoporin proteins, several viral proteins have been identified that co-opt NUPs and Rae1.ORF6 of SARS-CoV-2, ORF10 of Kaposi's sarcoma-associated herpesvirus, and M protein of vesicular stomatitis virus all bind to the NUP98-Rae1 complex, whereas the HIV-1 capsid binds to multiple nucleoporins leading to altered NUP expression and localization 25,26,28,[52][53][54][55][56][57][58] .While HIV-1 manipulates NUPs to facilitate viral import and integration of its genome into the host genome 62 , other viral proteins interact with NUPs to alter nucleocytoplasmic transport of key transcription factors like STAT1.During SARS-CoV-2 infection, ORF6-Rae1-NUP98 interactions block STAT1 import into the nucleus and mRNA export, resulting in a significantly diminished host response 24,26,28,57 .The ability of CebN to perturb STAT1 import during C.t. infection and our observation that it binds to NUPs-Rae1 suggest it might perturb host defense mechanisms by an analogous mechanism to that of ORF6 during SARS-CoV-2 infection.
Similar to how viruses dampen the immune response, C.t. has been shown to antagonize interferon pathways 63 .C.t. infection induces production of the proinflammatory cytokine IFN-γ, which is aimed at curtailing the infection, however the bacteria can attenuate this response establishing a persistent infection 64 .IFN production by the host cell activates the JAK-STAT signaling pathway.Activation leads to phosphorylation and homodimerization of STAT1, which is imported into the nucleus by karyopherin alpha 1 and karyopherin beta 1 heterodimers 49 .
Once in the nucleus, the STAT1 homodimer complex binds to gamma-activated site promoter elements to drive expression of a subset of ISGs meant to impede the infection.Recent studies have shown that following IFN-γ stimulation, nuclear translocation of STAT1 is reduced 32  Lining the NPC are intrinsically disordered NUPs that harbor numerous Phe-Gly (FG) repeats separated by a hydrophilic spacer of 5-30 amino acids 49 .Movement of large cargo across the NPC requires highly specific interactions between these so-called FG-NUPs and transporters of the karyopherin family, which enables entry and rapid diffusion of the cargo-transporter complex through the NPC.Of the 11 NUPs identified as putative binding partners of CebN, 9 are classified as FG-NUPs (Table 2, S3).Recognition of FG motifs within these select NUPs might explain how CebN binds to multiple nucleoporins.
Crystallization of CebN revealed that the N-terminus consists of a four α-helix bundles (α1-α2-α3-α4), followed by a three-stranded antiparallel β-sheets (β1-β2-β3) 66 .The C-terminus contains two α-helices (α5 Val135-Lys154 and α6 Pro156-Leu179) arranged in a kinked antiparallel manner 66 .AlphaFold modeling predicts that the C-terminus of CebN harbors a coiled-coil domain (amino acids 153-181).Truncation of the last 23 amino acids of CebN would disrupt this predicted coiled-coil, thus abrogating binding to NUPs and Rae1.While analysis of CebN did not identify motifs known to be required for interactions with NUPs or Rae1, new motifs are constantly being discovered, and it is possible that CebN possesses a previously undefined motif.
Ectopically expressed CebN appeared to concentrate at the nuclear envelope, thus we sought to evaluate the localization of CebN during infection.Upon doing so, not only did we observe CebN at the nuclear envelope of infected cells, but we also observed localization to the nuclear envelope of bystander cells.This unique localization in apparently uninfected bystander cells has only ever observed once before with the C. psittaci cT3SS effector SINC, which similarly targets the nuclear envelope through interactions with lamins 67 .Two mechanisms for SINC exit from infected cells into presumably uninfected bystander cells were proposed: packaging within exosomes for release at the cell surface and tunneling nanotubes.Prior work with Mycobacterium tuberculosis has revealed that mycobacterial proteins are packaged into vesicles and released via calcium-regulated lysosomal exocytosis.These proteins are then trafficked to uninfected bystander cells, expanding the microorganism's sphere of influence to other cells without the need to infect them 68 .Tunneling nanotubes are transient cellular connections that play a role in cell-to-cell communication and facilitate exchange of molecules between cells.Analogous to viruses, previous work has shown that C.t. may spread cell-to-cell by nanotubules 69 .Thus, it is conceivable that effector proteins may also be transmitted to adjacent cells via this mechanism.If an infected cell undergoes cell division, the inclusion is partitioned into one cell, leaving the other daughter cell "uninfected 70  Of the 36 effectors we sought to test for secretion in C.t., 23 were found to be secreted in at least one assay.Due to challenges generating clones, obtaining chlamydial transformants, and/or lack of expression during C.t. infection, we were unable to draw conclusions for 8 of these candidates.Of note, CT082, NUE, and CT795 were found to be not secreted in our assays (Table 1).NUE was previously shown to be secreted using Shigella flexneri as a surrogate host and mechanistic characterization revealed it localizes to the nucleus and interacts with chromatin 42 .While screening in a surrogate host is useful, especially in genetically intractable organisms, we 20 and others 23 have shown that secretion by a surrogate organism does not necessarily correlate with secretion in the native host.Our new results suggest NUE might not be secreted by C.t.However, it is possible that localization to the nucleus hinders evaluation of secretion and that assays such as subcellular fractionation of infected cells could tell whether it truly localizes in the nucleus.
As with all screens, false positives and negatives can result.To add rigor to our AP-MS data set analysis, we employed MiST, which combines metrics of reproducibility, specificity, and abundance across the entire data set to identify putative host binding partners more accurately and stringently.Using this technique, we identified high confidence targets for 21 of the secreted effectors tested herein.In our study, we sought to define PPIs for all the previously uncharacterized cT3SS effector proteins and included CT695 (TmeB) as at the onset of this study TmeB had no function ascribed to it.Recent work has since shown that ectopically expressed TmeB targets the ARP2/3 complex 71 .While we did not find components of the ARP2/3 complex in our infection AP-MS, we did identify an actin-binding protein, inverted formin 2 (INF2).
INF2 belongs to the formin family of proteins, which function to both polymerize and depolymerize actin filaments 72 .Formins and the ARP2/3 complex act in parallel to regulate the actin cytoskeleton 73 .Differences in experimental set-up between transfection of TmeB-FLAG and infection of a C.t. strain expressing TmeB-FLAG could contribute to these differences in identified putative host binding partners.Additionally, our timepoint of 24 hours post infection may correlate with functions of TmeB beyond invasion.
By combining chlamydial genetics with large-scale screens, we have begun to define the compendium of proteins secreted during active C.t. infection and further have identified putative host proteins and pathways targeted by each of these important factors.Our analysis uncovered a wealth of unique high-confidence host interactors, which will lay the groundwork for detailed mechanistic characterization of cT3SS effector proteins.Importantly, our approach has identified targets previously not associated with bacterial infection, including nucleoporins, which are commonly targeted by viral proteins to dampen the host response to viral infection [25][26][27][28] .We propose a model whereby secretion of CebN leads to inhibition of STAT1 nuclear translocation through interaction of CebN with NUPs and Rae1 (Fig. 7), which we hypothesize ultimately         Additionally, CebN is translocated to neighboring cells, potentially priming these cells for infection through altered nuclear import and/or export.
. For AP-MS, each validated cT3SS effector was cloned into the NotI/KpnI site of pBomb4-tet-mCherry with a FLAG-tag added to the C-terminus of each orf by PCR.CebN truncations were similarly cloned into pBomb4-tet-mCherry as FLAG-tagged fusions.For ectopic expression by transfection into human cells, secreted effectors were cloned into the KpnI/XhoI site of pcDNA3.1-GFP.The integrity of all constructs was verified by DNA sequencing at McLab.All primers used in this study are listed in Table and inhibition of the JAK-STAT pathway correlates with lower mRNA and protein levels of key interferon response elements in C.t. infected cells compared to uninfected cells 65 .Furthermore, this difference was dependent on C.t. de novo protein synthesis, supporting the role for a C.t. effector protein in this process 65 .Our new data showing that CebN binds NUPs and Rae1 and perturbs nuclear translocation of STAT1 may be the missing link needed to understand how C.t. establishes a persistent infection in spite of robust IFN-γ production.While the exact mechanism for how CebN-NUP-Rae1 interactions lead to blocked STAT1 nuclear translocation and whether this leads to diminished ISG production remains to be elucidated, our new observations support a role for CebN in attenuation of the IFN-γ response during C.t. infection.Future studies employing a CebN deletion mutant may yield additional information about its role in antagonizing the host defense system in the context of infection.However, as C.t. is predicted to use multiple mechanisms to antagonize the host immune response, our CebN transfection studies highlight the importance of this individual effector protein to this process.
." One intriguing possibility is that secreted effector proteins are left behind and continue to function in the absence of infection.The ability of C.t. effectors to access neighboring cells has huge implications for infection, potentially allowing C.t. to prime neighboring cells for infection prior to the invasion of the bacterium.Future work will involve delineating the exact mechanism by which C.t. is able to transport effector proteins into neighboring, uninfected cells and further whether effectors besides CebN and SINC can access adjacent cells.
alters host cell transcription to downregulate the innate immune response.More research is needed to know exactly what genes are being regulated.Moreover, the ability of CebN to translocate to bystander cells may provide a mechanism by which C.t. can prime nearby cells for infection.Future study of how CebN co-opts NUPs and Rae1 will not only enhance our understanding of how C.t. establishes a persistent infection despite a robust host response but may also identify druggable targets applicable to both bacterial and viral infections.N.J.K: P50 GM082250; PO1 AI090935, PO1 090935,P50 GM081879, PO1 091575, U19 AI106754, U54AI081680, DARPA-10-93-Prophecy-PA-008), University of Iowa Training in Mechanisms of Parasitism T32 AI007511 (B.S. and X.T.), and the University of California San Francisco Pathogenesis and Host Defense T32 AI060537 (K.M.M).We also acknowledge support from the University of Iowa Stead Family Scholars Program (M.M.W.), the University of Iowa Stinski fellowship (S.E.A), UCSF Program for Breakthrough in Biomedical Research (J.E.and N.K.), and the American Heart Association (K.M.M.).The Krogan Laboratory has received research support from Vir Biotechnology, F. Hoffmann-La Roche, and Rezo Therapeutics.Nevan Krogan has a financially compensated consulting agreement with Maze Therapeutics.He is the President and is on the Board of Directors of Rezo Therapeutics, and he is a shareholder in Tenaya Therapeutics, Maze Therapeutics, Rezo Therapeutics, GEn1E Lifesciences, and Interline Therapeutics.

Figure 2 .
Figure 2. Host pathways identified as targets of C.t. secreted effector proteins using AP-MS.

Figure 4 .
Figure 4. CebN localizes to the nuclear envelope in transfection and infection conditions.(A)

Figure 5 .
Figure 5.The C-terminus of CebN is necessary for interaction with NUPs and Rae1 as well as

Table 1 :
Candidate cT3SS effectors evaluated for secretion in C.t. ND-Not Determined, a unable to generate clone, b unable to obtain chlamydial transformants, c not expressed, d skipped, positive already in two.

Table 2 :
Host proteins with significant MiST score (>0.69) common to the infection and transfection IPs of CebN.