In Vivo Targeting of Clostridioides difficile Using Phage-Delivered CRISPR-Cas3 Antimicrobials

Abstract

Clostridioides difficile is an important nosocomial pathogen that causes approximately 500,000 cases of C. difficile infection (CDI) and 29,000 deaths annually in the United States. Antibiotic use is a major risk factor for CDI because broad-spectrum antimicrobials disrupt the indigenous gut microbiota, decreasing colonization resistance against C. difficile Vancomycin is the standard of care for the treatment of CDI, likely contributing to the high recurrence rates due to the continued disruption of the gut microbiota. Thus, there is an urgent need for the development of novel therapeutics that can prevent and treat CDI and precisely target the pathogen without disrupting the gut microbiota. Here, we show that the endogenous type I-B CRISPR-Cas system in C. difficile can be repurposed as an antimicrobial agent by the expression of a self-targeting CRISPR that redirects endogenous CRISPR-Cas3 activity against the bacterial chromosome. We demonstrate that a recombinant bacteriophage expressing bacterial genome-targeting CRISPR RNAs is significantly more effective than its wild-type parent bacteriophage at killing C. difficile both in vitro and in a mouse model of CDI. We also report that conversion of the phage from temperate to obligately lytic is feasible and contributes to the therapeutic suitability of intrinsic C. difficile phages, despite the specific challenges encountered in the disease phenotypes of phage-treated animals. Our findings suggest that phage-delivered programmable CRISPR therapeutics have the potential to leverage the specificity and apparent safety of phage therapies and improve their potency and reliability for eradicating specific bacterial species within complex communities, offering a novel mechanism to treat pathogenic and/or multidrug-resistant organisms.IMPORTANCEClostridioides difficile is a bacterial pathogen responsible for significant morbidity and mortality across the globe. Current therapies based on broad-spectrum antibiotics have some clinical success, but approximately 30% of patients have relapses, presumably due to the continued perturbation to the gut microbiota. Here, we show that phages can be engineered with type I CRISPR-Cas systems and modified to reduce lysogeny and to enable the specific and efficient targeting and killing of C. difficilein vitro and in vivo. Additional genetic engineering to disrupt phage modulation of toxin expression by lysogeny or other mechanisms would be required to advance a CRISPR-enhanced phage antimicrobial for C. difficile toward clinical application. These findings provide evidence into how phage can be combined with CRISPR-based targeting to develop novel therapies and modulate microbiomes associated with health and disease.

Keywords: CRISPR; CRISPR-Cas; Cas3; Clostridioides difficile; lysogeny; phage.

FIG 1

Overview of CRISPR phage engineering and mechanism of action. The genome of phage ϕCD24-2 was modified to encode a bacterial genome-targeting CRISPR array composed of a repeat-spacer-repeat meeting the requirements of the conserved C. difficile type I-B system. The genome-targeting CRISPR array is transduced into the bacterial cell during infection and is expressed concurrently with the lytic genes of the bacteriophage. Cell death occurs by two independent mechanisms of action: irreparable genome damage by the natively expressed type I-B Cas effector proteins directed by the CRISPR RNA and cell lysis by the holin and endolysin expressed during lytic replication.

FIG 2

In vitro comparison of the activity and lysogen formation rates of wild-type and engineered phage variants. (a) Time course of the reduction in the numbers of CFU during in vitro infection by bacteriophage at an input MOI of 0.1. CRISPR-engineered phage offers a clear improvement in the reduction in the numbers of CFU between 2 and 6 h, but by 24 h, all phage-treated cultures recover. Interestingly, there was no observable effect of lysogeny gene knockout on the activity of the phage. Data are presented as the mean ± standard error of the mean. BHIS, brain heart infusion supplemented with 0.1% taurocholate. (b) Time course of PCR-based detection of lysogeny in surviving bacterial colonies after phage infection. The CRISPR-enhanced phage exhibits impaired lysogen formation. Phage variants lacking key lysogeny genes exhibit no detectable lysogeny in vitro. Two biological replicates were performed per treatment per experiment.

FIG 3

A bacteriophage encoding a CRISPR targeting the C. difficile genome reduces the C. difficile burden and clinical signs of disease in vivo. (a) Schematic depicting the experimental design, timeline, and treatment groups. For the vehicle-, wtPhage-, and crPhage-treated groups, 20 mice were used in each group over three experiments; for the wtPhage Δlys- and crPhage Δlys-treated groups, 8 mice were used in each group. (b) Numbers of fecal C. difficile vegetative CFU from mice in each treatment group at days 2 and 4 postchallenge. For day 2, the vehicle-treated group contained 13 mice, the wtPhage-treated group contained 12 mice, the crPhage-treated group contained 13 mice, and the wtPhage Δlys- and crPhage Δlys-treated groups contained 8 mice each. For day 4, the vehicle-treated group contained 12 mice, the wtPhage-treated group contained 10 mice, the wtPhage Δlys-treated group contained 2 mice (6/8 mice treated with the wtPhage Δlys failed to defecate after three handling attempts), and the crPhage Δlys-treated group contained 8 mice. (c) Total histological scores of colons harvested at day 4 postchallenge. An uninfected, vehicle-treated group was included to control for the effects of the vehicle on colonic tissue. The uninfected vehicle-, wtPhage Δlys-, and crPhage Δlys-treated groups contained 8 mice each; the vehicle-, wtPhage-, and crPhage groups-treated contained 4 mice each. Each variable was scored on a scale of from 0 to 4; thus, the total summary score for maximum tissue damage is 12. (d) Representative images of hematoxylin- and eosin-stained colonic tissue from day 4 postchallenge for which the results are shown in panel c. Bars, 500 μm. (e) Percentage of lysogens isolated from the feces of mice treated with either wtPhage Δlys or crPhage Δlys over the course of the experiment (n = 8 mice per day, with 6 colonies per mouse being screened by PCR). For panels b and c, data are presented as the mean ± standard error of the mean. P values were determined by the Kruskal-Wallis one-way ANOVA with Dunn’s correction for multiple comparisons for panels b and c. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.