The dCas9 gene from Staphylococcus aureus was PCR-amplified following the manufacturer's guidelines for Phusion DNA polymerase (New England Biolabs, Ipswich, MA) using the plasmid, pX603-AAV-CMV::NLS-dSaCas9(D10A,N580A)-NLS-3xHA-bGHpA (a gift of Dr. F. Zhang; Addgene plasmid #61594), as template and primers 5′- ATATAACCGGTATGAAGCGGAACTACATCCTGGGCCT and 5′-ATATTCGGCCGTTAGCCCTTTTTGATGATCTGAGGGT. The underlined sequences correspond to AgeI and EagI sites, respectively, and the bolded nucleotide indicates the beginning of the gene specific sequence. The purified PCR product was digested with the indicated enzymes (Fastdigest from Fermentas/ThermoFisher) and ligated into the AgeI and EagI digested plasmid pL2-LtetO (kind gift of Dr. P.S. Hefty, Univ. Kansas), which was gel-purified and treated with alkaline phosphatase, using T4 DNA ligase (Fermentas). The ligation was transformed into chemically competent E. coli XL1 using standard techniques. The resulting colonies were screened for the correct plasmid, pL2-LtetO-Sa_dCas9, which was isolated and sequenced. The gRNA cassette targeting the 5′ region upstream of incA was synthesized by Integrated DNA Technologies (Coralville, IA). The gRNA sequence targeting the template strand was 5′-AATTTTTATCATATAAAGCCC (PAM = TAGGAT). A second gRNA sequence targeting the non-template strand, incA_IGR2, was also designed and tested: 5′-GGGAGATGGAGGAGTCACGAT (PAM = TAGAGT). For the non-targeting construct, only the gRNA scaffold was expressed. The gRNA transcription was designed to be under the control of a constitutive but weakened promoter driving dnaK expression in Chlamydia (Schaumburg and Tan, 2003). The sequence of the synthetic gene cassette(s) is listed in Supplementary Material. The gRNA was PCR-amplified using the synthetic gene cassette as template and primers 5′-gtaaattgattgtacaaggTCTTGAACGGTGGAGACG and 5′-aatttcgtctaacttacgTAAAACGAAAGGCCCAGTC. The lowercase letters correspond to the plasmid specific sequences for insertion into BamHI-digested pL2-LtetO-Sa_dCas9 via the HiFi DNA Assembly kit (New England Biolabs) according to the manufacturer's instructions and transformed into E. coli NEB5α. The resulting colonies were screened for the correct plasmid, pCRISPRi::L2 (incA_IGR, incA_IGR2, or non-targeting), which was isolated and sequenced. For the plasmid pL2-LtetO-Sa_gRNA, the gRNA cassette was PCR-amplified and inserted as above into pL2-LtetO previously digested with BamHI.

The murine fibroblast cell line McCoy (a kind gift of Dr. H. Caldwell, NIH) was routinely cultured at 37°C with 5% CO₂ in DMEM (Gibco/ThermoFisher, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, UT) and 10 μg/mL gentamicin (Gibco/ThermoFisher). The plasmidless (-pL2) strain of C. trachomatis serovar L2 was a kind gift of Dr. I. Clarke (Univ. Southampton) and was propagated in McCoy cells as described (Wang et al., 2011) for use in transformations.

Transformations were performed as described previously (Mueller and Fields, 2015) using demethylated plasmids prepared from the dam-/dcm- strain of E. coli from New England Biolabs. Briefly, 2 μg of demethylated plasmid was added to 2.5 × 10⁶ C. trachomatis serovar L2 (lacking the endogenous plasmid; -pL2) in Tris-CaCl₂ buffer (Wang et al., 2011) and incubated at room temperature for 30 min. McCoy cells plated the day before at 10⁶ per well in a six-well plate were subsequently infected with the transformation mix (one well per transformation). One U/mL penicillin (Millipore-Sigma; St. Louis, MO) was added at 8 h post-infection (hpi). Infected cells were harvested at 48 hpi, centrifuged at 20 k × g for 30 min at 4°C, and resuspended in 1 mL of Hank's Balanced Salt Solution (HBSS; Gibco). The suspension was centrifuged for 5 min at 400 × g at 4°C, and the supernatant was added to 1 mL HBSS to infect a new monolayer of McCoy cells. This process of infecting and harvesting infected cells was repeated until wild-type inclusions were visible and the penicillin-resistant bacteria isolated. When possible, organisms were plaque-purified one time as described elsewhere (Matsumoto et al., 1998) with individual plaques being picked from wells that had 5–10 plaques in total.

McCoy cells were seeded in 24-well tissue culture plates on coverslips at a density of 70,000 cells per well. The following day, cells were infected with the indicated strains at a multiplicity of infection (MOI) of 0.2–2. At the indicated hours post-infection (hpi), 10–50 nM anhydrotetracycline (aTc) was added or not to induce expression of the dCas9. Infected cells were fixed at the indicated hpi in methanol at room temperature for 5 min. In a subset of samples, the aTc was washed out at 16 hpi, and the cells were subsequently fixed at 24 hpi to monitor recovery of the target protein expression. Chlamydia was labeled with a guinea pig primary antibody (kind gift of Dr. E. Rucks, UNMC), and IncA, the target of the CRISPRi, was labeled with a rabbit primary antibody (kind gift of Dr. T. Hackstadt, Rocky Mountain Labs, NIH). Separately, a rabbit antibody against Sa_Cas9 (Abcam, Cambridge, MA) was used to visualize this target in some experiments. Alexa fluor-conjugated goat secondary antibodies (ThermoFisher) were used to detect the primary antibodies. The coverslips were imaged on an epifluorescent upright Olympus BX60 microscope equipped with a Nikon DS-Qi1Mc camera at a magnification of 40x or, alternatively, a Zeiss Imager.Z2 equipped with an Apotome2 and an Axiocam 506 monochromatic camera using a 63x Apochromat objective. Images were processed equally for contrast and brightness in Adobe Photoshop Creative Cloud 2017.

As originally constructed for E. coli, the CRISPRi system relied on the co-transformation of two plasmids: one encoding the dCas9 from Streptococcus pyogenes under the control of the Tet promoter and a second encoding the gRNA under the control of a constitutive promoter (Qi et al., 2013). The PAM sequence for S. pyogenes Cas9 is NGG. Given the inherent difficulties in transforming Chlamydia, the low efficiency of the process, the lack of compatible origins of replication, and the limited antibiotic selection agents, a single-plasmid version of this system was designed. However, although it was possible to repress expression of a target sequence in E. coli with the CRISPRi plasmid (data not shown), it was not possible to successfully transform C. trachomatis with it. The reason for this is not clear, but one possibility may have been that leaky expression of the S. pyogenes dCas9 was not well-tolerated.

Next a single-plasmid CRISPRi construct based on the smaller dCas9 from Staphylococcus aureus (Sa_dCas9; Figure 1), which utilizes a different PAM sequence, NNGRRT (Ran et al., 2015), was created. This plasmid encoded the Sa_dCas9 under the control of the Tet promoter and the gRNA under the control of a weakened promoter for dnaK that is expressed from an early time post-infection. Because of Chlamydia's developmental cycle, there is no bona fide constitutive promoter. As a proof of principle, the Sa_gRNA was designed to target the region upstream from the incA start site (incA_IGR). IncA is an inclusion membrane protein that is known to be dispensable and for which there are antibody reagents (Suchland et al., 2000; gift of Dr. T. Hackstadt). IncA was chosen as a target so as to better explore the likelihood of leaky expression that might be deleterious for an essential gene and, indirectly, off-target effects of the system (see below for further discussion on these possibilities). Additional control plasmids were constructed encoding only the inducible Sa_dCas9, only the “constitutive” Sa_gRNA targeting the incA intergenic region, or the gRNA scaffold without a targeting sequence and the Sa_dCas9.

To ensure that the Sa_dCas9 had no deleterious effects on chlamydial growth, C. trachomatis was transformed with the plasmid encoding only the dCas9. Upon isolation of a pure population of transformants, an experiment was prepared to test the effects of induction of Sa_dCas9 on chlamydiae. McCoy cells were infected with the transformant, and the expression of Sa_dCas9 was induced at 8 h post-infection (hpi), before the first division of the bacterium, by the addition of 10 nM anhydrotetracycline (aTc). Infected cells were fixed at 20 hpi and processed for immunofluorescence for IncA or Sa_dCas9 (both rabbit primary antibodies) in combination with a marker for chlamydiae. As seen in Figure 2A, Sa_dCas9 was clearly present as a cytosolic antigen within chlamydiae, as expected. In addition, IncA staining was clearly visible in replicate samples, thus, as predicted, expression of the dCas9 without the gRNA has no impact on IncA (or presumably any other target sequence). Further, given the large number of organisms present within the inclusions, the Sa_dCas9 was well-tolerated and did not impede chlamydial growth or division. It should be noted that, in the absence of inducer, a faint signal for the dCas9 could be detected by eye in the organisms when visualized on the microscope but was not easily imaged without significantly altering the acquisition settings or artificially enhancing the signal post-acquisition (data not shown, see also Supplementary Figure 1). Therefore, it is likely that the Tet promoter is not completely repressed by the Tet repressor in Chlamydia in the absence of inducer.

To ensure that the constitutive expression of the gRNA alone could not act as a small RNA to block incA expression, C. trachomatis was transformed with the plasmid encoding only the gRNA targeted to the incA intergenic region. Although penicillin-resistant transformants were isolated, it was not possible to isolate a pure population as a mixture of penicillin-resistant and—sensitive bacteria were recovered after multiple passages and even after re-infection from a single “resistant” inclusion (i.e., after lysis, adjacent inclusions were of a mixed phenotype). The reason for this apparent plasmid instability is not clear. McCoy cells were infected with the transformant, fixed at 20 hpi, and imaged for IncA. Nevertheless, in bacteria resistant to penicillin, IncA expression was detected (Figure 2B). As expected, IncA was also present on inclusions derived from penicillin-sensitive bacteria. Consequently, constitutively expressed gRNA was insufficient to inhibit expression of the target IncA as a small RNA.

As another control, a plasmid lacking a targeting gRNA sequence but expressing constitutively the gRNA scaffold and inducible Sa_dCas9 (i.e., non-targeting) was created and used to transform C. trachomatis L2. Induction of dCas9 failed to eliminate IncA, as expected (Figure 3). The leakiness of the dCas9 was demonstrated by capturing an extended exposure time. Again, this transformant could not be isolated as a pure population.

Given the observations that each component of the CRISPRi system alone was incapable of suppressing expression of the target IncA, the ability of the complete system to inducibly block IncA expression was evaluated. As with the gRNA construct alone, it was not possible to isolate a pure population of penicillin-resistant bacteria. McCoy cells were infected with the transformant and aTc was added at 12 hpi to induce Sa_dCas9. At 16 and 20 hpi, cells were fixed and imaged for the presence of IncA. In a subset of wells, the aTc was washed out at 16 hpi and replaced by fresh medium (with penicillin), and cells were incubated an additional 8 h (24 hpi) before fixation and processing. At both 16 (data not shown) and 20 hpi (Figure 4; 8 h Pulse), and for all penicillin-resistant bacteria, IncA was completely absent, indicating that the CRISPRi system was highly efficient at repressing IncA expression in contrast to samples without aTc where IncA remained on the inclusion (Figure 4). In penicillin-sensitive bacteria (i.e., bacteria without the plasmid conferring resistance), IncA expression was detected, thus showing the dependence of suppression on the CRISPRi plasmid. Similar results were noted using a gRNA targeting a different upstream region of incA (incA_IGR2; data not shown). Importantly, removal of the aTc inducer led to the detection of IncA in penicillin-resistant inclusions after 8 h (4 h Pulse/8 h Chase). Therefore, the effects of CRISPRi were reversible. However, the reversibility of IncA repression was not uniform as IncA negative inclusions were visualized in roughly half of penicillin-resistant inclusions after removal of inducer in the pulse-chase samples (data not shown). Presumably, longer chase times would result in a greater percentage of IncA-expressing inclusions albeit with the caveat of leaky dCas9 expression (see below).

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.