Abstract Phage satellites–mobile genetic elements which hijack a helper phage to propagate–have great potential to revolutionize synthetic biology applications. In particular, phage satellites have also been employed to deliver genetic circuits to the bacterial chromosome, including CRISPR/Cas9 cassettes that can target virulence genes to weaken pathogens (Ram et al 2018). A novel phage satellite class, Extracellular Prophage-Inducing Particles (EPIPs), can target Mycobacteriaceae and thus have particularly relevant applications for engineering native soil microbial communities to improve environmental health as well as treating common mammalian pathogens causing tuberculosis and leprosy (Qian et al., 2025). Mycobactericae can be particularly tricky to engineer using traditional plasmid techniques due to their mycolic acid walls preventing easy uptake of foreign DNA (Dao et al 2025). Already, mycobacteriophage-based DNA delivery has been found to increase transduction efficiency, including to deliver resolvases that knockout hygromycin antibiotic resistance in M. tuberculosis (Jain et al 2014), and other phage satellite classes infecting different bacterial hosts (such as SaPIs) can deliver circuits in a similar infection and integration manner (Ram et al 2018). These phage and phage-satellite methods are also advantageous in that modifications do not require antibiotic selection to persist, which is important given the rising biosafety concerns of horizontal gene transfer of antibiotic resistance (de Lorenzo and Martinez-Garcia 2025) particularly given a common antibiotic resistance cassette used in plasmid engineering, kanamycin, also confers resistance to amino glycides, a drug class used to treat tuberculosis infections (Yang et al 2015). I thus propose to modify one such EPIP capable of integrating into the M. aichiense genome, the EPIP named Bernie, to investigate its potential use as a microbial community editor. Specificially, I will use Golden Gate Assembly to re-assemble Bernie with the red fluorescence protein mCherry and the strong psmyc promoter optimized for myco, so that engineered Bernie DNA could be electroporated into the M. aichiense host, the phage satellite could reboot and infect, and the bacterial host would then express the red fluorescence gene newly integrated into its genome (as assessed using plate reader techniques).
Phage satellites–mobile genetic elements which hijack a helper phage to propagate–have great potential to revolutionize synthetic biology applications. In particular, phage satellites have also been employed to deliver genetic circuits to the bacterial chromosome, including CRISPR/Cas9 cassettes that can target virulence genes to weaken pathogens (Ram et al 2018). A novel phage satellite class, Extracellular Prophage-Inducing Particles (EPIPs), can target Mycobacteriaceae and thus have particularly relevant applications for engineering native soil microbial communities to improve environmental health as well as treating common mammalian pathogens causing tuberculosis and leprosy (Qian et al., 2025). Mycobactericae can be particularly tricky to engineer using traditional plasmid techniques due to their mycolic acid walls preventing easy uptake of foreign DNA (Dao et al 2025). Already, mycobacteriophage-based DNA delivery has been found to increase transduction efficiency, including to deliver resolvases that knockout hygromycin antibiotic resistance in M. tuberculosis (Jain et al 2014), and other phage satellite classes infecting different bacterial hosts (such as SaPIs) can deliver circuits in a similar infection and integration manner (Ram et al 2018). These phage and phage-satellite methods are also advantageous in that modifications do not require antibiotic selection to persist, which is important given the rising biosafety concerns of horizontal gene transfer of antibiotic resistance (de Lorenzo and Martinez-Garcia 2025) particularly given a common antibiotic resistance cassette used in plasmid engineering, kanamycin, also confers resistance to amino glycides, a drug class used to treat tuberculosis infections (Yang et al 2015). I thus propose to modify one such EPIP capable of integrating into the M. aichiense genome, the EPIP named Bernie, to investigate its potential use as a microbial community editor. Specificially, I will use Golden Gate Assembly to re-assemble Bernie with the red fluorescence protein mCherry and the strong psmyc promoter optimized for myco, so that engineered Bernie DNA could be electroporated into the M. aichiense host, the phage satellite could reboot and infect, and the bacterial host would then express the red fluorescence gene newly integrated into its genome (as assessed using plate reader techniques).
Project Aims
I. Experimental Aims
Use Golden Gate Assembly to reassemble the Bernie phage satellite (that can infect and integrate into Mycolicibacterium aichiense) with the mCherry red fluorescence gene and psmyc strong promoter optimized for mycobacteria. Reboot engineered Bernie into M. aichiense and assess for fluorescence expressed by the bacteria
II. Developmental Aims
Expand this engineering approach to swap in more complex genetic circuits, such as a CRISPR/Cas9 system that could target virulence genes in the host or a pAH (polycyclic aromatic hydrocarbons, a common organic pollutant in environments such as soil) pollutant degradation system.
III. Visionary Aims
Deploy engineered phage satellites into real-world environments such as soil to rapidly grant restorative capabilities to the microbiome.
Background
Phage satellites–mobile genetic elements which require a helper phage to propagate–have great potential to revolutionize synthetic biology applications. Biomanufacturing efforts often struggle with phage infection killing off bacterial populations (Kaminsky and Paczensky 2024), and phage satellites can defend against phage-induced lysis (Boyd and Seed, 2024). Phage satellites have also been employed to deliver genetic circuits to the bacterial chromosome, including CRISPR/Cas9 cassettes that can target virulence genes to weaken pathogens (Ram et al 2018). Additionally, one ongoing limitation of using phage therapy to treat antibiotic resistant infections is the limited host range of each phage (Cook and Hynes 2025); phage satellites have been identified to expand the host range of certain helper phages (He et al 2025).
Bacteriophage (phage) are viruses that infect bacteria, and many phage are temperate, meaning they can switch between either a lytic life cycle, replicating within and then killing bacteria to release phage particles, or a lysogenic life cycle, integrating into the bacterial chromosome and suppressing expression until a signal to induce (then excising from the bacterial genome and returning to the lytic cycle). Phage satellites are generally smaller than their helper phage and often lack the core machinery necessary for packaging and infection in their own genomes, such as capsid heads or tail fibers. They are not found in plasmids and have little homology to known phage, indicating a distinct evolutionary path (Sousa and Rocha 2021, Penades et. al 2025). Different categories of phage satellites hijack the helper phage in unique ways, but they all propagate by exploiting these phage resources.
Extracellular Prophage-Inducing Particles (EPIPs) are a newly-identified class of phage satellites that target Mycobacteriaceae and thus have particularly relevant applications for engineering native soil microbial communities to improve environmental health as well as treating common mammalian pathogens causing tuberculosis and leprosy (Qian et al., 2025). Like other phage satellites, EPIPs lack foundational proteins such as capsid and tail proteins, however they are genetically distinct from previously-characterized phage satellites, including genes such as Tape Measure Proteins that have not been found in any other phage satellites. These EPIPs are able to induce the prophage HerbertWM in Mycolicibacterium aichiense, and one EPIP, Bernie, has been identified to additionally independently integrate into the M. aichiense bacterial chromosome. My project thus seeks to use Bernie as a delivery tool for genetic circuits into Mycobacteriaceae.
Important ethical considerations for this project include the biosafety risks of deploying engineered bacteria, particularly for long-term goals with field usage, as the engineered circuit release would be difficult to reverse and phage satellite infection and excision could result in horizontal gene transfer. I might apply the ethical principles of Beneficiance, as the benefits of addressing soil pollution or deadly pathogens seems to outweigh the biosafety risks, especially given this is an improvement on the antibiotic-resistance risks existing in the status quo. Additionally, relevant precautions under the principle of non-maleficence would be utilized, such as introducing kill-switches to control circuits and assessing horizontal gene transfer before deployment.
Experimental Design
I. DNA Design of Engineered Bernie Constructs
While several engineering tools could be employed to modify EPIP phage satellites, Golden Gate Assembly has proven recently successful in assembling high GC% mycobacteriophage (Ko et al 2025) and thus would be particularly applicable for Bernie with a GC content of 63.4%. Other alternatives such as Gibson Assembly requires long overlap regions of 30-50 bp between fragments, which can be challenging to synthesize due to the high GC content in mycobacteriophage and EPIP genomes and is thus rarely attempted. The Hatfull lab, a leader in mycobacteriophage engineering efforts due to their work in the HHMI-Sponsored SEA-PHAGES program, has historically developed BRED, and by extension CRISPY-BRED, as the foundation for modifying mycobacteriophage. BRED works by prompting homologous recombination with a phage and synthetic DNA containing the mutation co-electroporated into the host (Marinelli et al 2008). CRISPY-BRED adds a selection mechanism to improve screening given BRED’s limited transformation efficiency by using CRISPR/Cas9 and an sgRNA complementary to the phage gene targeted for deletion/replacement in order to kill any phage that have not undergone recombination (Wetzel et al 2021). That said, CRISPY-BRED is limited by the constraints of PAM sites, and while screening is streamlined, still suffers from the low transformation efficiency of BRED. Likely due to these limitations, Hatfull recently transitioned efforts to focus on Golden Gate engineering of several mycobacteriophages (Ko et al 2025), and thus I selected that approach to test engineering of Bernie.
To achieve a high-fidelity assembly of Bernie with the strong constitutive psmyc promoter and mcherry gene, the Type IIS Restriction Enzyme PaqCI was selected to minimize internal cut sites. The Bernie genome was divided into 5 approximately-equal-sized fragments for scarless assembly using New England Biolabs’ Ligase Fidelity tool NEBridge SplitSet, where the tool predicted the best junction sites to split without overhang scarring when given ranges between gene gaps and Bernie’s two internal PaqCI cut sites. Primers were then designed using the IDT Oligo Analyzer tool to include a region complementary to these recommended junction sites as well as an overhang with the PaqC1 cut site and according filler bases.
II. Amplification of Bernie Parts for GGA
Bernie DNA was isolated using 1 ml of previously-isolated lysate and Phenol:Chloroform:Isoamyl Alcohol extraction with ethanol precipitation techniques (Phagehunting Program PCI/SDS DNA Extraction). 2 ng of said isolated DNA was utilized for each PCR fragment amplification according to standard NEB protocol alongside 0.5 μl of each forward and reverse primer, 5 μl of Q5 Hot Start 2X Master Mix, and NFW to a total of 10 μl (NEB). The annealing temperature for each reaction was determined using the corresponding primers and the NEB Tm calculator. PCR was performed on a Thermocycler with initial denaturation for 30 seconds at 98℃; 30 cycles of 98℃ for 7 seconds, annealing for 20 seconds, and extending at 72℃ for 40 seconds per amplicon kb length; and a final extension of 72℃ for 2 minutes. Products were run on a 1% agarose gel to confirm successful amplification of the expected length. PCR Products were cleaned up using NEB’s Monarch® PCR & DNA Cleanup Kit (5 μg) (NEB #T1030) following manufacturer’s instructions with 15 μl of elution buffer used.
To clone amplified fragments into plasmid backbones for preservation and use in Golden Gate Assembly, NEB’s PCR Cloning Kit (NEB #E1202S) was used following manufacturer instructions with the exception of scaling down the volume of NEB 10-beta Competent E. coli from 50 μl per reaction to 14 μl per reaction and corresponding NEB 10-beta/Stable Outgrowth Medium from 950 μl to 266 μl per reaction. Transformation steps occurred immediately after ligation. Transformed cells were plated at both 100 and 10-1 dilutions on LB agar plates containing 100 µg/mL ampicillin. Colonies were picked with sterile streaking needles, resuspended in 9 ul of NFW and then immediately dipped into 5 ml of LB media containing 100 µg/mL ampicillin in a glass tube which were placed on a shaking incubator at 37℃ and 250 rpm overnight. 1 ul of the NFW resuspension was used for insert screening PCR according to NEB PCR Cloning Kit protocols, using Q5 as the DNA polymerase and the given manufacturer Cloning Analysis primers designed around the pMiniT 2.0 Toxic Minigene Cloning Site to assess whether amplicons matched expected fragment lengths. The overnight cultures of successfully screened colonies were then miniprepped using the NEB Monarch Plasmid Miniprep Kit following manufacturer instructions. 50 μl of Elution Buffer was used given the pMiniT 2.0 backbone is a high copy plasmid.
III. Golden Gate Assembly Testing
For Golden Gate Assembly with NEB PaqC1 (#R0745S), 75 ng of each miniprepped plasmid, 7.5 U of PaqC1, 5 pmol of PaqC1 Activator, 300 U of T4 DNA Ligase, 2 μl of 10X T4 DNA Ligase Buffer, and NFW to a 20 μl total volume. Reaction was then run for 45 cycles of 37℃ for one minute and 16℃ for one minute, followed by a 5 minute incubation at 60℃. 2 μl of product was run with 2 μl of 6X loading dye and 8 μl NFW on a 1% agarose gel for 40 minutes at 150V and approximately 185 mA to assess assembly success.
IV. List of HTGAA Techniques Relevant To Project
The following HTGAA Techniques were employed:
Pipetting
Lab Safety
Bioethical Considerations
DNA Editing
DNA Construct Design
Gel Electrophoresis
Databases (NCBI)
Use of Benchling
Chassis Selection
Plasmid Preparation
Bacterial Culturing
Primer Design
PCR Reactions
Other Cloning Methods
Primer design for Golden Gate Assembly was fundamental to my project in order to design around each fragment from the phage satellite template DNA as well as the pCherry3 template DNA to include the PaqC1 recognition site and the according overhang. I also had the opportunity to employ real wet lab techniques, particularly gel electrophoresis which was employed at many steps (verifying initial PCR amplification of fragments, screening transformed colonies, running Golden Gate Assembled DNA).
Results and Validation
Project Validation
I chose to validate my project using DNA Design, PCR reactions with primers, and an initial Golden Gate Assembly (protocols explained in above experimental section). Techniques such as Primer Design, PCR Reactions, Golden Gate Assembly (Other cloning methods), and Gel Electrophoresis were used in this Validation step. ‘Validated’ data refers to resulting DNA fragments of according length for each part and the expected assembled lengths.
Data Analysis
The data I generated includes DNA fragment lengths after Golden Gate Assembly as visualized on a gel (after gel electrophoresis and EtBr staining). I found partial assembly from the full reaction, hypothesized to be fragments 2-5 as the band length was approximately 8000 bp and pairwise reactions (i.e. just 2 + 3, 3 + 4, and 4 + 5) revealed bands of the expected sum of the length of the two fragments.
Overcoming Challenges
I did not see full assembly from my Golden Gate reaction, which I suspect based on my data analysis is due to complications from the 1st Fragment, which is 4000 bp and nearly twice the length of the other fragments (due to how I had to split the genome based on internal PaqC1 cut sites and avoiding splitting genes). I plan to continue to test my validation by increasing the incubation time for Golden Gate as well as testing PCRs on the Golden Gate reaction to see whether there was some assembly with fragment 1 but just not at a high enough efficiency to visualize on the gel.
Additional Material
References
Boyd, C. M., & Seed, K. D. (2024). A phage satellite manipulates the viral DNA packaging motor to inhibit phage and promote satellite spread. Nucleic Acids Research, 52(17), 10431–10446. doi:10.1093/nar/gkae675
Cook, B. W. M., & Hynes, A. P. (2025). Re-evaluating what makes a phage unsuitable for therapy. Npj Antimicrobials and Resistance, 3(1), 45. doi:10.1038/s44259-025-00117-z
de Lorenzo, V., & Martínez‐García, E. (2025). On the choice of the right plasmid vector(s) in the times of synthetic biology. Microbial Biotechnology, 18(12), e70273. doi:10.1111/1751-7915.70273
Dao TO, Park HE, Lee JH, Kim KM, Trinh MP, Kang HL, Yoo HS, Shin MK. Advances and Challenges in Mycobacterial Genetic Engineering: Techniques for Knockout, Knockdown and Overexpression. J Microbiol Biotechnol. 2025 Nov 27;35:e2507051. doi: 10.4014/jmb.2507.07051. PMID: 41309382; PMCID: PMC12685580.
He, L., Patkowski, J. B., Wang, J., Miguel-Romero, L., Aylett, C. H. S., Fillol-Salom, A., . . . Penadés, J. R. (2025). Chimeric infective particles expand species boundaries in phage-inducible chromosomal island mobilization. Cell, 188(23), 6636–6653.e17. doi:10.1016/j.cell.2025.08.019
Jain, P., Hsu, T., Arai, M., Biermann, K., Thaler, D. S., Nguyen, A., . . . Jacobs, W. R. (2014). Specialized transduction designed for precise high-throughput unmarked deletions in mycobacterium tuberculosis. mBio, 5(3), 1245. doi:10.1128/mBio.01245-14
Kamiński, B., & Paczesny, J. (2024). Bacteriophage challenges in industrial processes: A historical unveiling and future outlook. Pathogens, 13(2), 152. doi:10.3390/pathogens13020152
Ko, C., Sikkema, A. P., Lauer, M. J., Amarh, E. D., Garlena, R. A., Russell, D. A., . . . Lohman, G. J. S. (2025). Genome synthesis, assembly, and rebooting of therapeutically useful high G+C% mycobacteriophages. Proceedings of the National Academy of Sciences of the United States of America, 122(46), e2523871122. doi:10.1073/pnas.2523871122
Penadés, J. R., Seed, K. D., Chen, J., Bikard, D., & Rocha, E. P. C. (2025). Genetics, ecology and evolution of phage satellites. Nature Reviews Microbiology, 23(7), 410–422. doi:10.1038/s41579-025-01156-z
Qian, H. L., Roman, A. N., Paudel, S., Hussey, G. E., Carline, K. B. R., & Saha, M. S. (2025). A novel phage satellite class induces prophage excision in mycolicibacterium aichiense openRxiv. doi:10.64898/2025.12.02.691453
Ram, G., Ross, H. F., Novick, R. P., Rodriguez-Pagan, I., & Jiang, D. (2018). Conversion of staphylococcal pathogenicity islands to CRISPR-carrying antibacterial agents that cure infections in mice. Nature Biotechnology, 36(10), 971–976. doi:10.1038/nbt.4203
Sousa, J. A., & Rocha, E. P. C. (2021). To catch a hijacker: Abundance, evolution and genetic diversity of P4-like bacteriophage satellites. Philosophical Transactions of the Royal Society B: Biological Sciences, 377(1842), 20200475. doi:10.1098/rstb.2020.0475
Yang, F., Njire, M. M., Liu, J., Wu, T., Wang, B., Liu, T., . . . Zhang, T. (2015). Engineering more stable, selectable marker-free autoluminescent mycobacteria by one step. Plos One, 10(3), e0119341. doi:10.1371/journal.pone.0119341