Creating and testing a flexible gene-silencing system to study bacterial control in the squid-bacteria partnership

To assay the reversal of CRISPRi repression of the mRFP test gene when IPTG inducer is removed, cultures of ES114:JMP1189/psgRNA (RR1) grown overnight with 2 mM IPTG induction of repression were re-cultured without IPTG inducer and assayed for restoration of RFP expression (Fig. 2B).

Bioluminescence is generated in V. fischeri when the autoinducer 3-oxo-C6 (Eberhard et al. 1981), activates the regulatory protein LuxR, which in turn activates transcription of the luxICDABEG operon. This operon contains luxI, (synthesis of the autoinducer 3-oxo-C6), luxAB (encoding dual-protein components luciferase enzyme), luxCDE (enzymes for synthesis of the luciferase aldehyde substrate) and luxG, a flavin reductase (Engebrecht et al. 1983; Engebrecht and Silverman 1984; Miyashiro and Ruby 2012; Nijvipakul et al. 2008). Since it has been shown that transcriptional repression caused by the steric hindrance of dcas9:sgRNA complex bound to a target gene is polar and blocks transcription of further downstream genes in an operon (Peters et al. 2016), we reasoned that targeting the luxC gene for repression would induce repression of the CDABEG genes of the operon and thus repression of bioluminescence.

The utility of ES114JMP1189/psgRNA (LC1) for regulating bioluminescence production from V. fischeri in its host environment during symbiosis was demonstrated by inoculating newly hatched aposymbiotic Euprymna scolopes juveniles with ES114JMP1189/psgRNA (LC1) cells that were induced by IPTG before and during the initial colonization of the juvenile squid. By manipulating the time-course of V. fischeri luminescence production during initial colonization, we can probe the time-dependence of the "winnowing" selection that prevents non-luminous symbionts from maintaining a successful colonization (Visick et al. 2000). Earlier research using non-luminescent luxA knockout mutants established that while mutant V. fischeri could successfully initiate colonization of the light organ, by 48 h post colonization the number of symbionts remaining in the light organ declined precipitously. This decline in light organ symbiont numbers then continued till no mutants were left, leading to an unsuccessful symbiosis (Visick et al. 2000).

As a final component of a generalized CRISPRi tool suite for V. fischeri, we investigated designs to combine the expression of multiple sgRNA transcripts from our psgRNA vector, which would allow for simultaneous repression of multiples genes-of-interest, or to target a single genetic locus with multiple sgRNA transcripts. Targeting multiple non-overlapping sgRNAs to a single gene can additively increase CRISPRi repression of transcription of that gene (Qi et al. 2013). We estimated that having two nearly homologous sgRNA cassettes in the same plasmid would lead to recombinational instability in V. fischeri (which has a generally high level of recombinational functionality (Thompson et al. 2007)) and therefore adopted a design that utilized nonhomologous sgRNA cassette component sequences that can be cloned into a single expression plasmid with minimal risk of recombination (Reis et al. 2019). A software sgRNA design tool (ELSA Calculator; De Novo DNA) was used to find combinations of non-homologous sgRNA cassette components (synthetic minimal promoters, RBS, sgRNA frameworks, transcription terminators) to enable the design of a sequence with three independent sgRNA cassettes that was cloned into our ES114:JMP1189 single sgRNA plasmid, replacing the sgRNA/lacI (BsaI) cassette. This generated the multimeric sgRNA cloning plasmid pMMsgRNA (3IIS). The three Type IIS spacer cloning sites can be sequentially restriction digested and cloned with tailed annealed oligomers, using the same protocol used for cloning spacer elements into the single psgRNA (BsaI) cloning plasmid. We generated multiple multiplexed sgRNA plasmids with combinations of 20-nt spacer targeting sequences and validated their function and utility in vitro. Two of the targeting spacers combined in multiplex format were the previously validated spacers for mRFP (RR1) and luxC (LC1), along with a spacer targeting the master regulator of the flagella production regulon, flrA (FA1) (Millikan and Ruby 2003; Wolfe et al. 2004). To test for enhancement of single gene repression through multiplexed 20-nt spacer targeting, we compared the repression of the test mRFP reporter in ES114:JMP1189 using the single 20-nt pMMsgRNA (RR1) with a version expressing two sgRNA to mRFP (pMMsgRNA (RR1, RR2). A triple sgRNA multiplex with targeting spacers for mRFP, luxC, and flrA was also able to cause significant fluorescence repression (Fig. 5).

Work in genetic tool development for marine bacteria is continually expanding, reflecting their wide-spread associations in aquatic microbiomes. α-proteobacteria and γ-proteobacteria symbionts have been subject to genetic modification for research in phytoplankton (Sunagawa et al. 2015), coral (Bourne et al. 2016), tubeworm (Vijayan et al. 2019), and sepiolid squid (Dunn et al. 2006; Visick et al. 2018) symbiotic systems. Recently, a novel CRISPRi plasmid tool was used to demonstrate the ability to modulate gene expression in the symbiont marine bacterium Pseudoalteromonas luteoviolacea during the course of its symbiosis with the model tubeworm, Hydroides elegans (Alker et al. 2023). In their study, CRISPRi repression of the macB gene, which is essential for inducing host tubeworm metamorphosis, was shown to lead to significantly reduced levels of metamorphosis in the host.

To demonstrate the utility of our CRISPRi system for modulating endogenous V. fischeri gene expression in symbionts within the host squid, we chose to target bioluminescence production during squid hatchling colonization. The production of bioluminescence by V. fischeri during the initial colonization of host tissues, and latter during the maintenance of the symbiosis within the mature light organ, is intricately coordinated by multiple environmental conditions and cues within host microenvironments (Pan et al. 2015; Chavez-Dozal et al. 2012; Nourabadi and Nishiguchi 2021), to ensure that bioluminescence is rhythmically produced in a diurnal pattern (Schwartzman et al. 2015). Bioluminescence produced by V. fischeri has also been found to act as a cue for changes in the tissue development of the immature colonized light organ (Visick et al. 2000), with hosts colonized by non-luminescing mutant V. fischeri strains failing to undergo normal light organ epithelial tissue maturation. These "dark" mutants are also actively rejected by the host after initial colonization, implying that the regulation of bioluminescence by microenvironmental conditions/cues (pH, oxygen saturation, nutrient status, cell density/quorum signaling) found within the host is also accompanied by an ability of the host to monitor the proper bioluminescence response from the symbionts (Visick et al. 2000; Chavez-Dozal et al. 2012; Nourabadi and Nishiguchi 2021; Pipes and Nishiguchi 2022). Inducible CRISPRi repression of bioluminescence provides an experimental tool for over-riding the normal regulation of bioluminescence during colonization; in our proof-of-concept application we were able to repress luminescence 48 h after initial colonization of hatchlings and recapitulate the decline in symbionts seen during initial colonization with mutant dark symbionts (Visick et al. 2000). Currently we are unable to extend the time course of our experiments past 96 h as we have been unsuccessful in raising the juvenile squid past this point. If this husbandry hurdle can be crossed, future studies that follow our CRISPRi repressible luminescence strains past our current 96 h time-course will be able to reveal the effects of reversing the repression and restoring luminescence in the remaining symbionts.

Bioluminescence in V. fischeri is produced by the luciferase enzyme, a heterodimer of LuxA and LuxB proteins. luxA and luxB are members of the lux operon (luxCDABEG). The LuxR-AHL transcriptional activator complex binds to the promoter region of the operon and activates polycistronic transcription of the genes.

CRISPRi repression of any of the genes in the lux operon would lead to loss of bioluminescence resembling the mutations (natural and artificial) in different lux operon genes that result in "dark" mutant strains (O'Grady and Wimpee 2008). We chose to target the luxC locus because it is the first gene in the operon and steric hindrance at the luxC locus would produce a polar repression (Peters et al. 2016) of the remaining genes in the operon (luxDABEG). Functional validation of in vitro bioluminescence repression was conducted using strain ES114:JMP1189 carrying the psgRNA (LC1) plasmid. The maximum level of repression was different than that found in validation tests of the mRFP targeting plasmid psgRNA (RR1)—this is consistent with work showing the level of repression achieved using CRISPRi being gene-specific (Banta et al. 2020a, b), and the time course required for loss of function through transcriptional repression reflecting the stability (half-life) of the protein(s) being repressed, as well as the growth rate of the cells (lowering functional levels of the repressed protein through dilution during cell division) (Qi et al. 2013). The times observed for maximally induced repression, and the lifting of that repression to initial levels in the mRFP test strain could serve to guide gene function experiments completed in similar cell culture conditions but are not as useful to guide research using timed CRISPRi repression of genes during V. fischeri host colonization. The growth pattern and phenotype of cells in an established light organ colonization are notably changed compared to the growth pattern, and phenotype, of the same cells grown in culture (Chavez-Dozal et al. 2014; O'Shea et al. 2005). During the rhythmic diel cycle of the light organ, 95% of the symbionts in the light organ are expelled at dawn. The remaining cells repopulate the light organ crypt microenvironments through rapid growth, reaching maximal cell density levels by noon (Norsworthy and Visick 2013). The rate of growth then diminishes abruptly (generation time increasing from ~ 0.5 h to 10–18 h) and cells can be observed to become reduced in size and non-flagellated (Ruby and Asato 1993). Since accurate estimations of the rate of timing of repression and de-repression of the luxC gene in the light organ microenvironment could not be generated in culture, the timing of induction of CRISPRi repression in the colonized squid experiments had to be determined empirically.

ES114:JMP1189 carrying psgRNA (LC1) was used to inoculate abiotic hatchling squid to demonstrate the ability of inducible timed repression of bioluminescence in vivo to probe the temporal monitoring of symbiont bioluminescence by the host (Fig. 5). The addition of IPTG to the hatchlings seawater at 24 h post inoculation was able to induce and maintain repression of bioluminescence from the symbionts to levels that triggered the same pattern of loss of symbiont numbers seen during early stages of colonization with dark lux mutants (Koch et al. 2014). Notably, the drop in CFU/LO at 55 h post inoculation followed the same pattern as seen at 24 h in "dark" ES114:JMP1189 carrying psgRNA (LC1) that were fully IPTG induced at inoculation. Our results demonstrate the utility of inducible CRISPRi repression in probing the temporal distribution of regulatory conditions (bioluminescence) and responses (symbiont rejection) during the symbiosis.

CRISPRi systems that target single genes enable relatively rapid and facile transcriptional repression for forward genetics studies, avoiding laborious mutant strain construction methodologies that result in altering the native regulation of the gene (Qi et al. 2013; Peters et al. 2019; Mimee et al. 2015; Rock et al. 2017; Jiang et al. 2024). Simultaneous CRISPRi targeting of multiple genes in the same cell (multiplex CRISPRi) (Larson et al. 2013; Qi et al. 2013; Peters et al. 2016; Zhao et al. 2016) avoids the even more laborious construction of multiple-deletion strains. Multiplex CRISPRi systems have been demonstrated in various bacterial species using a variety of methods to express multiple sgRNAs—including crRNA arrays (Kim et al. 2017), multiple arrayed independent sgRNA cassettes (Qi et al. 2013; Peters et al. 2016), and nonrepetitive extra-long sgRNA arrays (ELSAs; Reis et al. 2019).

Our novel V. fischeri multiplex sgRNA plasmid uses an ELSA cassette for the simultaneous expression of three sgRNAs. Each sgRNA cassette within the multiplex plasmid had a unique strong synthetic constitutive promoter, a 20-nt spacer sequence targeting an independent genetic locus, functional sequence variant sgRNA framework sequences, and unique transcription terminators and spacing sequences. With this design, no sequence part in the three sgRNA cassettes shares more than 17 bp homology with another part. This lack of homology, compared to the > 90% sequence homology between repeated sgRNA cassettes has two advantages: (1) plasmids with repeated sequences (areas of homologous sequence) can recombine in highly recombinogenic bacteria like V. fischeri (Lin et al. 2018), altering the content of the plasmid or rendering it unstable for replication, and (2) Synthetic dsDNA construct services cannot synthesize sequences containing areas with homologous repeats (Reis et al. 2019). Notably, the three unique sgRNA cassettes could not use the same PLlac0-1 inducible promoter used in the single sgRNA cassette plasmids (excessive sequence homology) and having three different inducible promoters would make it logistically difficult to utilize. Inducibility of the multiplexed CRISPRi system comes from the single PLlac0-1 promoter driving the integrated dcas9 cassette in the parent ESS114:JMP1189 strain.

We observed higher mRFP repression from the multiplex plasmid with two non-overlapping mRFP spacer sequences compared to the multiplex plasmid with a single mRFP spacer sequence, demonstrating the ability of a multiplexed sgRNA plasmid to generate increased target gene repression by providing two simultaneous sites of steric hindrance (Qi et al. 2013). The ability of a multiplex sgRNA plasmid to simultaneously repress three separate genes was demonstrated using assays for mRFP fluorescence, bioluminescence, and motility on soft-agar plates. The flrA regulatory gene was chosen as a gene-of-interest in this study because it has also been shown in prior research (Chavez-Dozal et al. 2012; Millikan and Ruby 2003; Shrestha et al. 2022) to be the master transcription activator of the flagella production regulon (which contains other non-flagella associated operons; Millikan and Ruby 2003). In V. fischeri, flagellar gene regulation is controlled in a multilevel cascade (Norsworthy and Visick 2013), with FlrA activating early flagellar genes, including flrBC and fliA. Expression of the late genes, multiple flagellar filament and motor proteins, is regulated in a sequential cascade first by FlrBC and then FliA (Aschtgen et al. 2019). flrA is necessary for initial colonization of the light organ, but its expression is repressed via unknown signals and cues in the light organ where most symbionts appear non-flagellated (Ruby and Asato 1993), only to be re-expressed shortly before symbionts are expelled at dawn (Norsworthy and Visick 2013). When insertional (DM126, DM127, and DM128) and partial deletion (DM159) mutations of the flrA gene were constructed they were shown to be non-motile in soft agar motility plates (Millikan and Ruby 2003). Additional ΔflrA knockout mutants in V. fischeri, generated by targeted homologous recombination, were also found to be non-motile in soft-agar (Visick et al. 2018).

The work presented demonstrates proof-of-principle utility of a CRISPR interference suite of vectors and strains for inducible repression of expression of both exogenous and endogenous genes-of interest via simple targeting with target sequence complementary 20-nt spacer sequences cloned into sgRNA expression plasmid vectors. The single IPTG inducible sgRNA expression plasmids, used with a common genomically integrated IPTG inducible dcas9 ES114 strain, can repress V. fischeri gene expression in a titratable and reversable course in culture, as well as in symbionts colonizing juvenile host squid. A multiplex version of the sgRNA plasmid, though not inducible itself, provides the capability for elevated repression of a single gene via multiple independent complementary targeting spacer sequences to that gene, or can simultaneously target up to three genes for repression. CRISPR interference is a powerful, flexible general genetic programming tool for research in the V. fischeri—E. scolopes symbiosis model system, complementing and extending the current toolbox of gene modification methodologies and allowing for determining the exact timing of the genetic cross talk between hosts and their beneficial microbes.

Strains and plasmids used in this study are listed in Table S1. Oligos, and synthetic dsDNA constructs used in this study are listed in Table S2. Escherichia coli strains DH5α and pir + competent GT115 (InvivoGen) were used for plasmid maintenance, cloning, and conjugation. V. fischeri strain ES114 was grown aerobically using various media at 28 °C. For molecular biology applications and fluorescence assays, V. fischeri cells were cultured in Luria–Bertani high salt (LBS) containing 10 g of tryptone, 5 g of yeast extract, 20 g of NaCl, and 20 mM Tris-hydrochloride (Tris–HCl, pH 7.5) per L of sterile water. All squid colonization and bioluminescence assays used seawater tryptone (SWT), which contained 5 g of tryptone, 3 g of yeast extract, and 700 mL of artificial seawater (ASW) per L of sterile water. E. coli, aerobically grown at 37 °C, was cultured in Luria–Bertani (LB) media, which contained 10 g tryptone, 5 g yeast extract, and 10 g NaCl. Agar was added to the media at 15 mg/mL for plating. Where appropriate, the antibiotics chloramphenicol and kanamycin were added to growth media at 1–3 and 100 μg/mL, respectively, for V. fischeri with ampicillin, chloramphenicol and kanamycin added to growth media at 100, 30, and 100 μg/mL respectively, for E. coli. Thymidine (0.3 mM) was used to supplement LB media for culturing of the thy-auxotroph DH5α strain carrying conjugation helper plasmid pEVS104. IPTG (Isopropyl ß-D-1-thiogalactopyranoside) (GoldBio), for CRISPRi induction was added to media when appropriate as indicated.

Artificial seawater for use with E. scolopes hatchlings was made using ddi H₂0 (1L) with 30 g Instant Ocean® marine salt (Instant Ocean Spectrum Brands) and 10 g Marine Mix® marine salt (Bulk Reef Supply). Artificial saltwater was aerated with aquarium air-pumps and air-stones.

All molecular cloning reagents (restriction enzymes) are from NEB. Plasmid extraction Plasmid Miniprep Kits (ZymoPURE™) from Zymogen. Primers were ordered from IDT DNA. Synthetic dsDNA constructs were ordered from Twist Biosciences.

Our experimental goals were to create an inducible CRISPRi system able to target and repress single or multiple genetic loci in V. fischeri, both in culture and in symbiosis with their squid host. These goals lead to several design constraints-The dcas9 and sgRNA components must be stably retained through prolonged growth periods—up to several days of growth in infected squid. The use of antibiotic selection to enforce retention of plasmid vectors is problematic because of potential toxicity of common antibiotic selective agents with the squid host, and incomplete penetration of antibiotics within the light organ crypts (Dunn et al. 2006).The inducible promoter(s) used to control expression of CRISPRi components should have minimal leaky expression when uninduced and the inducer used with the system, when added to sea-water, must be non-toxic to the squid host, and be able to penetrate the crypts of the light organ.Express sgRNA in excess of dcas9 to ensure that the proportionately lower concentration of individual sgRNA transcripts produced from a multiplex (three sgRNA cassette) vector can saturate the expressed dcas9.

Successful CRISPRi systems have been developed that genomically integrate both dcas9 and sgRNA (Liu et al. 2017; Peters et al. 2016), as well as systems that contain dcas9 and sgRNA on dual or single stable plasmid vectors (Rachwalski et al. 2024; Tan et al. 2018). For our goals though, heterologous systems, with an integrated dcas9 and plasmid expressed sgRNA (Rousset et al. 2018) fit all of the design constraints.dcas9 cassettes have been previously integrated into the V. fischeri genome, at the intergenic attTn7 Tn7 transposon site, via conjugation of an integrating plasmid vector(Geyman et al. 2024). Tn7 integrated dcas9 cassettes have been shown to be stably retained through > 50 generations of antibiotic selection-free growth in culture while maintaining dcas9 expression (Peters et al. 2019). Stable V. fischeri plasmids have been developed, derived using the native pES213 plasmid origin of replication, which are stable (Dunn et al. 2005) during colonization of the squid. pVSV105 is a widely utilized conjugatable pES213 derived plasmid (copy number of 9.4 ± 2.4) which has been found to be ≥ 99% retained in V. fischeri after nonselective growth in culture for 60 generations, or 72 h of squid colonization (Dunn et al. 2006).IPTG inducible promoters, such as the Plac, Pma, and PLlac0-1 promoters, have been successfully used in V. fischeri, both in culture and in squid hosts (Geyman et al. 2024; Visick et al. 2018). IPTG is non-toxic to the squid host and penetrates the light organ crypts (Visick et al. 2018). Dual PLlac0-1 promoters controlling both dcas9 and sgRNA expression were recently used in the attTn7 integrating Mobile-CRISPRi system to repress the rpoB essential gene in V. fischeri. The system exhibited negligible non-induced dcas9 expression, while showing significant induced rpoB repression (Geyman et al. 2024).The multicopy stable pVSV105 plasmid provides ~ 10 sgRNA expression cassettes for every single genomic dcas9 expression cassette. Using the same PLlac0-1 inducible promoter for both provides coordinated induction of both components, with an excess of sgRNA. This excess of sgRNA transcripts, due to using the same promoters for dcas9 and sgRNA while having the dcas9 integrated and the sgRNA expressed from multiple plasmids, is an elegant way to ensure that in the multiplex version of the sgRNA plasmid there is enough of each individual sgRNA transcript to saturate the expressed dcas9 (Luo et al. 2015; Reis et al. 2019; Vigouroux and Bikard 2020). In a CRISPRi study in Zymomonas mobilis using independently inducible promoters for dcas9 and sgRNA expression, it was noted that the concentration of sgRNA relative to the concentration of dcas9 was a limiting factor in CRISPRi repression (Banta et al. 2020a, b).

A Mobile-CRISPRi Tn7 transformation donor plasmid developed for conjugational transfer into γ-proteobacteria (pJMP1183-Addgene, #119254; (Peters et al. 2019)) was used to transfer a mRFP (monomeric red fluorescent protein) expression cassette (for use as a repression target in testing), the lacI gene needed for IPTG induction, an IPTG inducible (PLlacO-1 promoter) dcas9 cassette, an IPTG inducible (PLlacO-1 promoter) sgRNA (RR1) expression cassette targeting the introduced mRFP gene for repression, and a KanR selection marker into V. fischeri ES114 at the attTn7 locus to generate strain ES114:pJMP1183. We also separately transferred pJMP1189 (Addgene, #119257; Peters et al. 2019) to introduce an IPTG inducible (PLlacO-1 promoter) dcas9 cassette, the lacI gene needed for IPTG induction, a KanR selection marker, and an mRFP expression cassette (for use as a repression target in testing, and later as a fluorescent cell tag) into V. fischeri ES114 at the attTn7 locus to generate strain ES114:pJMP1189. These were transferred to V. fischeri ES114 by tetra-parental conjugation with the helper E. coli strains CC118 λpir + /pEVS104 thy- (pESV104 contains conjugative transfer genes and is a thymidine auxotroph) and BW25141/pJMP1039 (expresses Tn7 transposase; see Table S1). (Christensen et al. 2020). Briefly, E. coli donor strains (carrying Tn7 transposon plasmids pJMP1183, and pJMP1189, respectively) and E. coli helper strain Dh5a thy- carrying pEVS104 (Eric Stabb and Ruby 2002) along with a second E. coli helper strain carrying transposase (pJMP1039, Addgene # 119239; Peters et al. 2019) were cultured individually overnight in 5 mL LB media at 37 °C with appropriate antibiotics and additives. V. fischeri strain ES114 was cultured overnight (14 h) in SWT media at 28 °C with shaking. All cultures were then sub-cultured (100 μL aliquots into 5 mL appropriate media) and grown at their respective temperatures with shaking for 3–4 h. 250 μL of each of the four cultures were combined into a 1.5 mL microcentrifuge tube (Eppendorf). A 250 μL donor only control from each of the three E. coli cultures was placed in 1.5 mL microcentrifuge tubes, and a 250 μL recipient-only (ES114) control was placed in a 1.5 mL microcentrifuge tube. Cells were then pelleted at 8000 rpm for 5 min in a microcentrifuge (Eppendorf), and the supernatant discarded leaving 10–20 μL liquid in the tube. The pellets were resuspended in this liquid, and 100 μL of each tube was spotted in the middle of a pre-warmed LBS agar plate. After the liquid had absorbed into the plates, they were inverted and incubated overnight in a 28 °C incubator. In the morning, the colony was scraped off with a sterile pipette tip and the cells resuspended in 1 mL LBS by vortexing briefly. 100 μL aliquots of each cell culture were spread onto antibiotic containing LBS agar plates for selection of attTn7-site integrated transformed colonies (in the case of pJMP1183 and pJMP1189 this is Kanamycin, 100 μg/μL). Plates were incubated at 18 °C to repress growth of donor E. coli, and colonies of transformed V. fischeri selected when they appeared after 1–3 days. 20 of these colonies were isolated onto master LBS Kan 100 μg/mL agar plates and tested for the presence of integrated CRISPRi transposon construct by patching the master plate colonies on LBS with ampicillin agar and LBS kanamycin agar and incubating the plates at 28 °C overnight. Strains that have integrated the transposon segment and no longer retain the donor plasmid are AmpR-, KanR+. Further confirmation of genomic integration at the attTn7 site is completed colony PCR using primers flanking the attTn7 insertion site.

The highest ranking candidate sgRNA spacer 20-nt sequences targeting the mRFP1, mRFP2, luxC, and flrA genes were found with the CRISPy-web on-line sgRNA software (CRISPy-web; secondarymetabolites.org), using the V. fischeri ES114 reference genome (GenBank: GCA_000011805.1) to identify optimal dcas9 targeting sites that have no potential off-target sites (sequences with more than 9/20 homologous bases to the spacer sequence). Candidate spacers closest to the 5' end of the ORFs were always chosen since studies of the efficacy of sgRNA targeting locations along several gene sequences suggest that achieving high efficacy gene knockdown in Vibrio species necessitates targeting toward the 5′ end of the open reading frame (Geyman et al. 2024). Additional sequence was added to the 5' end of these sequences to provide compatible single strand overhangs that match the restriction sites of the sgRNA plasmid spacer cloning site (BsaI for the single sgRNA plasmids, and BbsI, BsaI, and BaeI for the multiplex plasmid spacer cloning sites. Complementary pairs of annealed 5'-phosphorylated oligos with these 20-nt spacer + single strand overhangs were ordered from IDT DNA. See Table S2 for sequences of oligos.

The annealed oligo spacer sequences targeting the mRFP, luxC, and flrA genes were cloned into the triple-BsaI targeting sites of psgRNA plasmids via ligation of annealed oligos (Brooks et al. 2014). 2 μL of a 1:20 dilution of the annealed oligos was ligated to 100 ng BsaI-digested psgRNA for 30 min at room temperature following the manufacturers protocol (NEB.T4 Ligase, M0202). The ligation mixture was used to transform ChemiComp GT115 Competent E. coli (Invivogen, Cat. Code gt115-11, 0.1 mL aliquots) following the manufacturers protocol. Briefly, 100 μL of frozen GT115 cells were thawed on ice and then transferred into a chilled 1.5 mL tube. 5 μl of the ligation reaction was added to the cells, mixed, and returned to ice for 30 min. The cells were heat shocked in a 42 °C water bath for 30 s and then returned to ice for two minutes. 900 μL of room temperature SOC medium was added to the reaction and the cells were incubated for one hour at 37 °C with shaking (250 rpm). 200 μL of the reaction was spread on LB agar plates with 50 μg/mL chloramphenicol and incubated overnight at 37 °C. 20 isolated colonies were collected the next day and spotted with a sterile pipette tip on a master plate (LB agar plates with 50 ug/mL chloramphenicol) which was grown overnight at 37 °C. When spotting the master plate, the pipette tip was also used to transfer some of the colony to a 50 μL PCR reaction mix in a 0.2 mL PCR tube. Colony PCR was performed using Q5 Hi-Fi 2X MasterMix (NEB, Cat. # M0492S) per the manufacturers protocol. The thermocycling conditions for the reaction were: Initial denaturization 98 °C 30 s, 25 cycles @ 98*C denaturation 10 s, annealing 66–70 °C (primer dependent, see Table S2 for primer sequences) 30 s, extension 72*C 30 s. A final extension at 72*C 2 min, and then held at 4 °C. 10 μL of each colony PCR reaction was mixed with 2 μL 6X Purple Loading Dye (NEB; Cat. #B7024S) and run on a 1% agarose gel in TAE buffer at 75 V until the loading dye front was ~ ¾ down the gel. 1 μL of 1 kb Plus DNA Ladder (NEB; Cat. #N3200S) mixed with 1 μL 6X Purple Loading Dye (NEB; Cat. B7024S) and 4 μL TAE buffer was added to the gel as a size standard.

The multiplex plasmid pMMsgRNA (3TIIS) expression cassette, with an array of three separate sgRNA expression cassettes, was designed using the Extra Long sgRNA Array (ELSA) software (De Nova DNA; https://salislab.net/software/design_elsa_calculator↗) to contain three programable sgRNA cassettes (BbsI, BsaI, and BaeI spacer cloning sites) with non-homologous sequences. The insert was ordered as a synthetic dsDNA construct with 30 bp homologous ends (TWIST Bioscience, CA, USA) and cloned into a PCR amplified (primers pVSV105 cassette/MMsgRNA F/R (Table S2)) pVSV105 backbone using the NEBuilder® HiFi DNA Assembly Cloning Kit (NEB Cat. #E5520S) following the NEB protocol. The resulting plasmid, pMMsgRNA(3TIIS), is an empty cloning vector which can express three independent targeted sgRNAs from three different constitutive promoters. Cloning a 20-nt targeting spacer into a multiplexed sgRNA plasmid follows the same protocol described above for designing and cloning the spacer, and follows the same protocol for transforming the targeted multiplex vector into a V. fischeri. Strain with integrated dcas9. While the three sgRNA cassettes are not IPTG inducible, the dcas9 is—so the multiplex CRISPRi system is inducible, but not independent. Four additional multiplex plasmid were constructed using a different protocol (—for these plasmids the synthetic dsDNA construct was designed to already contain 20-nt spacer sequences in the three sgRNA cassettes. All that was needed to construct these plasmids was to use HiFi Assembly (see above) and transform into V. fischeri (see below).pMMsgRNA(RR1)—targets site 1 in mRFP.pMMsgRNA(RR1:RR2)—targets sites 1 & 2 in mRFP.pMMsgRNA(RR1:LC1)—targets site 1 in mRFP and site 1 in luxC.pMMsgRNA(RR1:LC1:FA1)—targets site 1 in mRFP, site 1 in luxC, and site 1 in flrA.

Complete inducible CRISPRi V. fischeri strains were generated by introducing an inducible sgRNA expression plasmid, such as psgRNA (this study), from an E. coli cloning host by triparental conjugation with an E.coli thymidine- auxotroph containing conjugative helper plasmid pEVS104 (Stabb and Ruby 2002) and the appropriate V. fischeri recipient (ES114:pJMP1183 and ES114:pJMP1189) as described previously (DeLoney et al. 2002). Briefly, E. coli donor strains (i.e. psgRNA) and E. coli helper strain DH5α thy- carrying pEVS104 (Stabb and Ruby 2002) were cultured individually overnight in 5 mL LB media at 37 °C with appropriate antibiotics and additives. A recipient V. fischeri strain (e.g. ES114:pJMP1183; mFRP + , KanR) was cultured overnight (14 h) in SWT media at 28 °C with shaking. All cultures were then sub-cultured (100 μL aliquots into 5 mL appropriate media) and grown at their respective temperatures with shaking for 3–4 h. 250 μL of each of the three cultures were combined into a 1.5 mL microcentrifuge tube (Eppendorf). A 250 μL donor only control from each of the two E. coli cultures was placed in 1.5 mL microcentrifuge tubes, and a 250 μL recipient-only control was placed in a 1.5 mL microcentrifuge tube. The cells were pelleted at 8,000 rpm for 5 min in a microcentrifuge (Eppendorf), and the supernatant discarded leaving 10–20 μL liquid in the tube. The pellets were resuspended in this liquid, and 100 μL of each tube was spotted in the middle of a pre-warmed LBS agar plate. After the liquid had absorbed into the plates, they were inverted and incubated overnight in a 28 °C incubator. In the morning, the colony was scraped off with a sterile pipette tip and the cells resuspended in 1 mL LBS by vortexing briefly. 100 μL aliquots of each cell culture were spread onto LBS agar plates containing the appropriate antibiotic to select for plasmid transformed colonies (Chloramphenicol, 1–3 μg/μL). The plates are incubated at 18 °C to repress growth of donor E. coli, and colonies of plasmid bearing V. fischeri can be selected when they appear after 1–3 days. 20 of these colonies are isolated onto master LBS Kan 100 μg/mL agar plates and tested for the presence of the plasmid using plasmid extraction kits (Zymogen) and restriction digest analysis. Strains that have the sgRNA plasmid along with the chromosomally integrated dcas9 cassette will be CamR+, KanR+.

GraphPad Prism Version 10.1 for Windows, GraphPad Software, (www.graphpad.com↗). One-way analysis of variance (ANOVA) was performed on multiple groups comparisons. Significance in the ANOVA analysis was followed by the use of between-group comparisons calculated using the Tukey's Post Hoc comparison test. One-way or two-way ANOVAs and unpaired t-tests were used to analyze data for each graph as noted. For two-way ANOVA analyses, Šídák's multiple comparisons test was used.

E. scolopes husbandry and hatchling inoculation protocols are detailed in Supplementary File 1. The Nishiguchi laboratory's squid husbandry room at UC Merced is IACUC certified for all related experiments.

Luminescence produced by V. fischeri colonized hatchlings was measured with a Turner Designs TD20/20 luminometer (Turner Design). Juveniles were placed in 5 mL of sterile artificial seawater in 30 mL glass scintillation vials, and light produced was recorded for 15 s. Triplicate measurements (arbitrary light units, LU, per animal) were obtained and averaged.

The number of V. fischeri cells colonizing the squid light organ were assayed using LBS agar plates. Individual hatchlings to be assayed (typically after having their luminescence reading taken) were rinsed three times in 5 mL of sterilized artificial seawater. Juveniles were homogenized in a 1.5 mL Eppendorf microtubes using sterile 1.5 mL plastic pestles (Bel-Art) with 0.5 mL of sterile artificial seawater. 100 μL of the homogenate was used after serial dilution to plate in triplicate on LBS agar plates incubated at 28 °C. At 24 h colonies were counted to determine the colony forming units/light organ (CFU/LO).

Strains were grown with shaking at 28 °C in SWT, with and without CRISPRi induction. At regular intervals, 1 mL aliquots of each culture were assayed for cell density (Optical density (OD₆₀₀) via absorbance at 600 nm) with 1 mL cuvettes in a UV-3100PC spectrophotometer (VWR) and 5 mL aliquots in 30 mL glass scintillation vials were measured for luminescence (relative light units, RLU) in a Turner Designs TD-20/20 luminometer. A SpectraMax iD5 multimodal platereader was also used for measuring OD₆₀₀, and for bioluminescence production. Fluorescence of mRFP tagged CRISPRi strains could also be quantified (ext: 590, emm: 612), and read in conjunction with OD₆₀₀ and luminescence.

For motility repression experiments, V. fischeri ES114 (control) and V. fischeri CRISPRi strains targeting flrA were grown overnight in TBS broth (1% tryptone, 2% sodium chloride) at 28 °C, with and without IPTG inducer supplementation as indicated, then sub-cultured in the same medium for 2 h at 28 °C (static). Cells to be assayed were normalized to an OD₆₀₀ of 0.2 and 10 μL were drop spotted onto TBS soft agar plates with and without IPTG inducer supplementation as indicated, and incubated at 28 °C. Migration distance was measured hourly (as diameter to outer motility ring, mm) and plotted versus time.