Development of TEM-1 β-lactamase based protein translocation assay for identification of Anaplasma phagocytophilum type IV secretion system effector proteins

Anaplasma phagocytophilum, the aetiologic agent of human granulocytic anaplasmosis (HGA) is an obligate intracellular Gram-negative bacterium with the genome size of 1.47 megabases. The intracellular life style and small size of genome suggest that A. phagocytophilum has to modulate a multitude of host cell physiological processes to facilitate its replication. One strategy employed by A. phagocytophilum is through its type IV secretion system (T4SS), which translocates bacterial effectors into target cells to disrupt normal cellular activities. In this study we developed a TEM-1 β-lactamase based protein translocation assay and applied this assay for identification of A. phagocytophilum T4SS effectors. An A. phagocytophilum hypothetical protein, APH0215 is identified as a T4SS effector protein and found interacting with trans-Golgi network in transfected cells. Hereby, this protein translocation assay developed in this study will facilitate the identification of A. phagocytophilum T4SS effectors and elucidation of HGA pathogenesis.


Results
Development of a TEM-1 β-lactamase based protein translocation assay for A. phagocytophilum T4SS. T4CP specifically interacts with its cognate substrates through its AAD domain, meanwhile it can interact functionally with heterologous T4SS. Based on this observation, previously a Cre recombinase reporter assay for translocation (CRAfT) was established to determine the translocation of heterologous effector proteins using E. coli as recipient 8 . In this assay, TraJ, the T4CP of conjugation plasmid pKM101 T4SS, was remodeled by swapping the cytoplasmic domain of TraJ with the counterpart of VirD4 of T4SS in A. tumefaciens, A. phagocytophilum, or Wolbachia pipientis to create chimeric T4CP, TraJ-VirD4, in which the NTD of TraJ is kept for interaction with other VirBs of pKM101 T4SS. When co-expressed with Cre-tagged heterologous effector proteins in E. coli, cytoplasmic domain of TraJ-VirD4 recognizes its cognate substrates and delivers them to E. coli recipient cells through pKM101 T4SS. In recipient cells, Cre-tagged substrate proteins cleave loxP sites flanking the reporter gene, leading to the phenotypic change. By this assay, several T4SS effectors were confirmed, including Ats-1 of A. phagocytophilum. Followed the same rationale, we employed another broad-host-range plasmid RP4 and TEM-1 β-lactamase reporter to establish a more straightforward system for assaying the translocation of A. phagocytophilum effector proteins to eukaryotic recipient cells. Promiscuous plasmid RP4 can transfer substrates to a broad range of host cells including bacteria, fungi, and eukaryotic cells 32,33 . TEM-1 β-lactamase is widely used as reporter in applications such as mammalian gene expression assay, and bacterial effector protein translocation assay in eukaryotic cells 30,31 . To establish this assay system, we utilize a E. coli strain SM10 λpir as donor since it hosts integrated plasmid RP4 34 . Two types of plasmids were constructed, one expresses NTD of T4CP (TraG) of RP4 T4SS and cytoplasmic domain of VirD4 of A. phagocytophilum fusion protein (TraG-VirD4) 8,35 , the other expresses TEM-1. TraG-VirD4 gene was cloned into plasmid pMal-C5X under control of promoter tac. The ampicillin resistance gene, bla in pMal-C5X was then replaced with streptomycin resistance gene, Sm R to avoid interference with TEM-1 in assay, leading to the generation of plasmid pTraG-VirD4 (Fig. 1A,C). TEM-1 β-lactamase gene expression cassette with promoter tac and transcription terminator rrnBT1 and T2 is cloned into plasmid pACYCDuet-1 for replacement of T7-based protein expression elements, leading to the generation of plasmids pTEM-1 and pTEM-1-X (X: A. phagocytophilum proteins) (Fig. 1B,D). For investigation of effector translocation, SM10 λpir expressing TraG-VirD4 and TEM-1 or TEM-1-X were incubated with Chinese hamster ovary K1 (CHO K1) cells, followed by the loading with CCF2-AM, a fluorescent substrate for β-lactamase. If TEM-1-X is transported into CHO K1 cells, it cleaves CCF2-AM, leading to emission of blue light, instead of green light, under fluorescent microscope. We first use a confirmed T4SS effector, Ats-1 to validate the system. CHO K1 cells incubated with SM10 λpir expressing TraG-VirD4 and TEM-1-Ats-1 have higher percent of blue cells than those incubated with SM10 λpir expressing TraG-VirD4 and TEM-1 (0.367 ± 0.012% vs. 0.040 ± 0.010%) (Fig. 2), suggesting that TEM-1-Ats-1 is transported into CHO K1 cells. To determine whether the translocation of TEM-1-Ats-1 is T4SS dependent, plasmid pSD VirD4 which expresses only cytoplasmic domain of VirD4, was constructed by deletion of DNA sequence encoding NTD of TraG in pTraG-VirD4. The NTDdeleted T4CP loses interaction with T4SS 9 , thereby it fails to translocate substrates. CHO K1 cells incubated with SM10 λpir expressing SD VirD4 and TEM-1-Ats-1 have lower percent of blue cells, compared to those incubated with SM10 λpir expressing TraG-VirD4 and TEM-1-Ats-1 (0.020 ± 0.010% vs. 0.367 ± 0.012%) (Fig. 2), suggesting the translocation of TEM-1-Ats-1 is T4SS dependent. Thus TEM-1 β-lactamase based protein translocation reporter system for A. phagocytophilum T4SS is established.

Discussion
Bacterial T4SS plays an important role in the pathogenesis by dysregulating physiological processes of host cells through its transported effector proteins. Although a myriad of effector proteins was identified in intracellular bacteria, such as L. pneumophila, C. burnetii, and Brucella 31,36,37 , a very limited number of T4SS substrates was revealed in A. phagocytophilum. Most T4SS effector proteins bear typical signals, such as the positive charge www.nature.com/scientificreports www.nature.com/scientificreports/ at their C-termini. Based on these signals or features, T4SS effector proteins were initially identified by bioinformatic analysis, followed by the verification in effector protein translocation assay 31 . Several translocation assays were established based on Cre-loxP recombination, or the enzyme activities of the adenylate cyclase (Cya) domain of the Bordetella pertussis Cya toxin, and TEM-1 β-lactamase. Compared to Cre-loxP-based system which observes the gene transcription, and Cya-based system which measure intracellular cAMP level, TEM-1 β-lactamase offers a simple, flexible assay by observing the conversion of fluorescent substrate. Because of the www.nature.com/scientificreports www.nature.com/scientificreports/ difficulty in genetic manipulation of A. phagocytophilum, the verification of protein translocation was routinely performed in surrogate hosts including A. tumefaciens, L. pneumophila, C. burnetii, and recently E. coli. This newly developed Cre-loxP-based pKM101 conjugation system in E. coli has advantage over other systems since it uses T4CP for its cognate substrates as showed by demonstration of translocation of several rickettsial effector proteins, including A. phagocytophilum Ats-1 into E. coli recipient 8 . Although successfully used for identification of AnkA and HGE-14, two effectors of A. phagocytophilum, the coupling proteins of A. tumefaciens VirB/VirD4 and L. pneumophila Dot/Icm were not engineered, which might cause the lower translocation efficiency of A. phagocytophilum T4SS substrates compared to those in which the engineered coupling proteins are used, especially for L. pneumophila Dot/Icm, which belongs to phylogenetically distant T4BSS type. Meanwhile pKM101 conjugation system only transfers its substrates to bacterial recipient 8 , which prevents the use of TEM-1 β-lactamase as reporter in eukaryotic cells. In this study we take advantage of broad-host-range of RP4 conjugation plasmid to establish heterogeneous protein translocation assay using TEM-1 β-lactamase as reporter, and eukaryotic cells as recipient. Ats-1, a known effector protein was showed to be transported into CHO K1 cells in this assay, validating that this assay system can work with A. phagocytophilum T4SS effector proteins. We further applied this assay system to determine the translocation of two T4SS substrate feature-bearing A. phagocytophilum proteins. Sec24 domain-containing APH0215, not APH0756 was found to be transported into CHO K1 cells in this assay. Of note, due to the VirB protein difference between RP4 T4SS and A. phagocytophilum T4SS, this established protein translocation assay using chimeric coupling protein might not be suitable for all of A. phagocytophilum T4SS substrates. Sec24 is an important component of coat protein complex II (COPII) which promotes the formation of transport vesicles from ER 38 . It was reported that A. phagocytophilum extensively interacts with ER in vertebrate and invertebrate host cells, and recruits vesicles derived from ER and trans-Golgi network to A. phagocytophilum-containing compartments 21,26 . We found that APH0215 vesicles were colocalized with TGN38, suggesting there is the interaction between APH0215 and trans-Golgi network. However, whether APH0215 is involved in A. phagocytophilum interaction with ER and trans-Golgi network in A. phagocytophilum infection needs further investigation. In sum, we here developed a straightforward protein translocation assay for A. phagocytophilum T4SS, which will facilitate identification of A. phagocytophilum T4SS effector proteins and elucidation of HGA pathogenesis.

Materials and Methods
Construction plasmids pTraG-VirD4 and pTEM-1. To make chimeric coupling protein, TraG-VirD4 which is composed of the N-terminal transmembrane domain (NTD) of RP4 T4CP, TraG (1 st -102 nd amino acids of TraG) 35 and cytoplasmic domain of A. phagocytophilum VirD4 (103 rd -740 th amino acids of VirD4) 8 , the DNA fragment encoding TraG-VirD4 was chemically synthesized (Talen-Bio Scientific Co, Shanghai, China) with two extra DNA sequences located at 5′-end (5′-CAATTGACCAACAAGGACCATAGATTATG-3′; derived from plasmid pMal-c5X (New England Biolabs), containing MfeI endonuclease site and ribosomal binding site; Italicized nucleotides: MfeI site; Underline nucleotides: start codon) and 3′-end (5′-TAGGGATCC-3′; Underline nucleotides: stop codon; Italicized nucleotides: BamHI site) respectively. This DNA fragment was digested with MfeI and BamHI, followed by ligation with MfeI, BamHI-digested pMal-c5X and transformation into competent E. coli DH5α strain (Vazyme Biotech Co, Nanjing, China), leading to the generation of plasmid pTraG-VirD4-bla, in which original malE gene is replaced by traG-virD4, and the β-lactamase gene bla is still kept. Since β-lactamase expressed from pTraG-VirD4-bla will interfere with TEM-1 expressed from another plasmid, its gene is replaced by streptomycin resistance gene (Sm R ). DNA fragment including coding region and promoter region of Sm R was amplified by PCR using a pair of primers (SM-F and SM-R) (Table S1), and plasmid pCDFDuet-1 (Novagen) as template. This DNA fragment was digested with NdeI and NotI, and ligated with NdeI, NotI-digested PCR product amplified using pTraG-VirD4-bla as template, and a pair of primers (pMal-F and pMal-R) ( Table S1). The ligated product was transformed into E. coli DH5α, leading to the generation of plasmid pTraG-VirD4-Sm R (pTraG-VirD4), in which original β-lactamase gene bla is replaced by Sm R gene. To make plasmid pVirD4 , which only expresses the cytoplasmic domain of VirD4 of A. phagocytophilum, the PCR product amplified using a pair of primers (VirD4-F and VirD4-R) (Table S1) and pTraG-VirD4 as template, was digested with SalI and self-ligated, followed by transformation into E. coli DH5α, leading to the generation of pSD VirD4, in which the coding region of NTD of RP4 TraG was deleted.
To construct plasmid that expresses TEM-1 β-lactamase reporter, DNA fragment containing pMal-c5X-derived promoter tac and transcription terminators rrnB T1 and T2, 6x His and bacteriophage T7 tag coding sequence, TEM-1 coding sequence, and multiple cloning sites (MCS) was commercially synthesized (Talen-Bio Scientific Co, Shanghai, China). By recombination-based cloning using ClonExpress II One Step Cloning Kit (Vazyme), the PCR product amplified using this synthesized DNA fragment as template and a pair of primers (TEM-1-F and TEM-1-R) (Table S1) was cloned into pACYCDuet-1 (Novagen) linearized by PCR using a pair of primers (pACYCDuet-F and pACYCDuet-R) (Table S1), leading to the generation of plasmid pTEM-1, in which original phage T7-based expression elements is replaced by ptac-based expression system for expressing TEM-1 and TEM-1-X fusion proteins. For constructing plasmid which expresses TEM-1-Ats-1, TEM-1-APH0215, or TEM-1-APH0756, DNA fragments encoding full length of Ats-1, APH0215, and APH0756 were amplified by PCR using A. phagocytophilum genomic DNA as template and responding primer pairs (Table S1), and cloned by using ClonExpress II One Step Cloning Kit into pTEM-1 linearized by NdeI digestion, leading to the generation of plasmids pTEM-1-Ats-1, pTEM-1-APH0215, and pTEM-1-APH0756. To construct pTEM-1-APH0215ΔC, in which the DNA sequence encoding the C-terminal 21 amino acids of APH0215 was deleted, the PCR was performed using a pair of primers (215-Delta-C-R and 215-Delta-C-F) (Table S1) and pTEM-1-APH0215 as template. The PCR product was digested with BamHI and self-ligated, leading to the generation of pTEM-1-APH0215ΔC.
For easy view, plasmid map of pTraG-VirD4 and pTEM-1 were made with the application SnapGene (GSL Biotech, Chicago, IL,USA) after their sequences were assembled.
Protein translocation assay. CHO K1 cells (Invitrogen) were propagated in F-12K medium (Gibco) supplemented with 10% fetal bovine serum (Gibco). Cultures were incubated at 37 °C in a humidified 5% CO 2 /95% air atmosphere. The day before the assay, the CHO K1 cells were split onto cover slips in 24-well plate with 5 × 10 4 cells/well. On the day of assay, SM10 λpir harboring responding plasmids were cultured in Luria-Bertani (LB) medium containing spectinomycin (50 μg/mL) and chloramphenicol (50 μg/mL) to 0.3 OD 600 , followed by the addition of Isopropyl β-D-1-thiogalactopyranoside (IPTG) to final concentration of 1 mM to induce protein expression. After 2 h induction, 1 mL of SM10 λpir cultures were assigned to Western blot analysis to determine protein expression. The remaining cultures were centrifuged at 5000 × g for 5 min, and the bacterial pellets were resuspended with the same volume of completed F-12K medium supplemented with 1 mM IPTG. CHO K1 cells with 80% confluence were washed with 1 x PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO4, and 2 mM KH 2 PO4, pH 7.4), followed by the addition with SM10 λpir suspension at 0.5 mL/well (ratio of bacteria/cells is 200:1) and centrifugation at 250 × g for 10 min. After CHO K1 cells were incubated with SM10 λpir for additional 2 h, the cells were washed by PBS for 3 times to remove bacteria, followed by the loading of fluorescent sub-  (Table S1). For protein expression, the PCR products were digested with EcoRI and NotI, ligated with pET28a (+) (prokaryotic expression vector, Novagen), and transformed into E. coli BL21 (DE3) (Vazyme). Recombinant E. coli BL21 (DE3) were cultured in LB medium containing kanamycin (34 μg/mL) at 37 °C with shaking at 250 rpm. After bacteria were cultured to reach the mid-logarithmic phase of growth (OD 600 value of 0.4), IPTG was added at the final concentration of 1 mM to induce recombinant protein expression. After induction at 37 °C for 4 h, bacterial cultures were centrifuged at 10 000 × g for 10 min at 4 °C. The pellet was lysed and recombinant protein was affinity-purified with Ni-NTA resin (Qiagen) under denaturing condition as instructed by manufacturer. Purified recombinant protein was subjected to commercial antibody production in rabbit host (Abgent Biotech, Suzhou, China).

Transfection and immunofluorescence labeling.
To determine the cellular localization of APH0215 in transfected cells, the plasmid pAPH0215-GFP was constructed and transfected into HeLa cells, followed by immunofluorescence labeling. Briefly, DNA fragment encoding full-length APH0215 was amplified by PCR using a pair of primers (APH0215-GFP-F and APH0215-GFP-R) (Table S1) and A. phagocytophilum genomic DNA as template, followed by digestion with XhoI and BamHI, and ligation with eukaryotic expression vector pEGFP-N1 (Clontech), leading to the generation of plasmid pAPH0215-GFP. pAPH0215-GFP and pEGFP-N1 were then transfected into HeLa cells by Polyethylenimine MAX in EZ Trans transfection kit according to manufacturer's instruction (Shanghai Life iLab Biotech Co, Shanghai, China), respectively. For IFA, the transfected cells were harvested at 36 h post-transfection, and subjected to fixation with 2% paraformaldehyde in PBS at RT for 30 min and washing with PBS for three times, followed by permeabilization and blocking with PS solution (PBS containing 0.8% BSA and 0.3% saponin) at RT for 10 min. Immunofluorescence labeling was performed using mouse monoclonal antibody against TGN38 (Clone B-6, Santa Cruz Biotechnology) and Alexa Fluor 555-conjugated goat anti-mouse IgG (Invitrogen). Cells were observed under Leica TCS SP8 confocal microscope (Leica).
Western blot analysis. SM10 λpir and SM10 λpir harboring plasmids were centrifuged at 10, 000 × g for 1 min after induced with 1 mM IPTG for 2 h in the protein translocation assay. Bacterial pellet from 1 mL culture was dissolved in 100 μl of 2 × SDS-PAGE loading buffer (4% SDS, 135 mM Tris-HCl [pH 6.8], 10% glycerol, and 10% β-mercaptoethanol). Samples were separated by SDS-PAGE with 12% polyacrylamide resolving gels, then transferred to nitrocellulose membrane using mini Trans-blot cell (Bio-Rad). The membrane blocking was performed with blocking buffer (5% (wt/vol) skim milk (BD Difco) in PBS buffer) for 30 min at RT, followed by incubation with mouse monoclonal anti-T7 antibody (Millipore, diluted in blocking buffer at ratio 1:5000), rabbit anti-VirD4 antibody (diluted in blocking buffer at ratio 1:2000), or mouse monoclonal anti-E. coli GroEL antibody (abcam, diluted in blocking buffer at ratio 1:5000) at RT for 1 h. After washing for three times with PBS (10 min each time), the membranes were probed with peroxidase-conjugated goat anti-mouse IgG, or anti-rabbit IgG secondary antibody (KPL, Gaithersburg, MD, diluted in blocking buffer at ratio 1:5000) at RT for 1 h. The membranes were washed four times with PBS (10 min each time), followed by ECL chemiluminescence. The images were captured by Tanon 4200 chemiluminescent imaging system (Tanon, Shanghai, China). Similarly, to determine the expression of APH0215-GFP, GFP, and TGN38 in transfected cells, HeLa cells transfected with pAPH0215-GFP or pEGFP-N1 were harvested at 36 h post-transfection, and subjected to probing with mouse monoclonal antibodies against GFP (TransGen Biotech, Beijing, China), TGN38, and β-tubulin (Beijing CoWin www.nature.com/scientificreports www.nature.com/scientificreports/ Biotech Co) in Western blot analysis. E. coli GroEL and human β-tubulin were used as internal controls for sample loading. Due to the abundance of GroEL in E. coli and GFP in transfected cells, which shadows other weak bands in image capture by CCD camera, sequential incubation with antibodies were performed. To determine the expression VirD4 and TEM-1 expression in SM10 λpir, nitrocellulose membranes were first probed with anti-VirD4 antibody or anti-T7 tag antibody, and developed, followed by incubation with anti-GroEL antibody and development. For determination of APH0215-GFP and GFP in transfected cells, nitrocellulose membranes were sequentially incubated with antibodies against β-tubulin and GFP. For determination of TGN38 in pAPH0215-GFP or pEGFP-N1-transfected cells, nitrocellulose membranes were concurrently incubated with antibodies against TGN38 and β-tubulin.
Statistical analysis. Three independent experiments were conducted and data were expressed as mean ± S.D. Statistical significance was evaluated by the Student t-test.
The materials and data generated in the current study are available from the corresponding author on reasonable request.