Zebrafish as an effective model for evaluating phage therapy in bacterial infections: a promising strategy against human pathogens

ABSTRACT

The escalating prevalence of antibiotic-resistant bacterial infections necessitates urgent alternative therapeutic strategies. Phage therapy, which employs bacteriophages to specifically target pathogenic bacteria, emerges as a promising solution. This review examines the efficacy of phage therapy in zebrafish models, both embryos and adults, which are proven and reliable for simulating human infectious diseases. We synthesize findings from recent studies that utilized these models to assess phage treatments against various bacterial pathogens, including Enterococcus faecalis, Pseudomonas aeruginosa, Mycobacterium abscessus, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, and Escherichia coli. Methods of phage administration, such as circulation injection and bath immersion, are detailed alongside evaluations of survival rates and bacterial load reductions. Notably, combination therapies of phages with antibiotics show enhanced efficacy, as evidenced by improved survival rates and synergistic effects in reducing bacterial loads. We also discuss the transition from zebrafish embryos to adult models, emphasizing the increased complexity of immune responses. This review highlights the valuable contribution of the zebrafish model to advancing phage therapy research, particularly in the face of rising antibiotic resistance and the urgent need for alternative treatments.

INTRODUCTION

The escalating challenge of antibiotic resistance poses a significant threat to global health, necessitating innovative approaches to combat bacterial infections (1). The extensive use and misuse of antibiotics have accelerated the evolution of resistant strains, notably those classified among the ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species). These pathogens are responsible for the majority of hospital-acquired infections worldwide and are particularly alarming due to their frequent resistance to multiple antibiotics, rendering many conventional treatments ineffective (2, 3). This crisis has underscored a critical need for alternative therapeutic strategies. Phage therapy, which utilizes bacteriophages (or phages) to target and eliminate bacterial pathogens, stands out as one of the most promising options. It offers numerous advantages over traditional antibiotics, such as high specificity to target bacteria, minimal disruption to native microbiota, and the ability to evolve in response to bacterial resistance mechanisms (4, 5). However, advancing phage therapy from basic research to a viable clinical option necessitates comprehensive in vivo studies to explore its dynamics, efficacy, and safety (6–8). In this context, the zebrafish (Danio rerio) model has emerged as a valuable tool in preclinical research, offering several unique advantages that make them particularly suitable for these studies (9, 10). Their genetic, physiological, and immunological similarities to humans render them an essential model for studying human diseases (11). Additionally, the high reproductive rate of zebrafish ensures a consistent supply of subjects, thereby facilitating large-scale studies. This is economically advantageous, enabling researchers to conduct extensive testing at a fraction of the cost associated with mouse models, which is particularly critical in the early stages of research where multiple trials are often necessary to optimize therapeutic strategies. Regulatory frameworks mandate data from indispensable mammalian models before progressing to human clinical trials. Therefore, incorporating zebrafish studies allows researchers to refine their hypotheses and treatment protocols, thereby shortening the duration and reducing the scope of subsequent, more costly, and time-consuming mouse studies (12, 13). By utilizing the versatility of zebrafish models, researchers can accelerate the development of phage therapy, gaining crucial insights into its efficacy, mechanisms, and potential limitations in treating bacterial infections. Finally, the zebrafish model complies with the 3Rs (Replacement, Reduction, Refinement) rule, an ethical principle aimed at improving the welfare of animals used in scientific experiments. This review highlights how the integration of the zebrafish model into phage therapy research underscores its valuable contribution to advancing this promising field, particularly in the face of rising antibiotic resistance and the urgent need for alternative treatments. To achieve this, we conducted a comprehensive search on PubMed and other relevant databases using a combination of keywords tailored to our study focus. Specifically, we used terms such as “zebrafish model,” “embryo,” “adult,” “bacterial pathogen,” “human infection,” and “phage therapy.” We included studies that specifically employed either zebrafish embryos or adults as models for human bacterial infections treated with phage therapy and explicitly excluded any studies focusing on pathogens specific to fish. All articles that met these criteria were thoroughly reviewed and are cited in Table 1, ensuring a comprehensive representation of the current research landscape in this field.

TABLE 1.

Main features of studies using zebrafish models (embryo and adult) to assess the efficacy of phage therapy against human pathogenic bacteria^a

Bacteria	Route of infection and treatment	Bacterial dose and zebrafish stage	Phage treatment	Combination therapy	Outcomes	Ref.
Zebrafish healthy embryo model
Enterococcus faecalis	Circulation (duct of Cuvier)	30,000 CFU/embryo (30 hpf)	Phage SHEF2, 6 × 105 PFU/embryo (2 hpi)	NA	↗ Survival rate (~57%)	(14)
Staphylococcus aureus	Local (hindbrain ventricle)	15,000 CFU/embryo (30 hpf)	Phage SAVM02, 70 PFU/embryo (2 hpi)	Vancomycin (300 µg/ml)	↗ Survival rate with phage, vancomycin or combination of both (~100%) ↘ Bacterial load with phage (~99%), vancomycin (~97%) or combination of both (~97%)	(15)
Klebsiella pneumoniae	Bath immersion	3.3 × 106 CFU/ml (72 hpf)	Phage UPM2146, 1 × 106 PFU/mL (0.5 hpi)	NA	↘ Bacterial load (~100%)	(16)
Acinetobacter baumannii	Bath immersion	10,000 CFU/ml (48 hpf)	Phages K2, K9, K32 and K45 alone, 109 PFU/mL (12 hpi)	NA	↗ Survival rate (~40%)	(17)
Pseudomonas aeruginosa	Circulation (duct of Cuvier/yolk sac)	30 CFU/embryo (48 hpf)	Phages vB_PaeP_PYO2, vB_PaeP_DEV, vB_PaeM_E215 and vB_PaeM_ E217 in cocktail, 300–500 PFU/embryo (0.5 hpi)	NA	↗ Survival rate with phages with treatment at 0.5 hpi (~28%) and seven hpi (~30%) ↘ Bacterial load with phages (~80%) ↘ Inflammatory response with phages	(18)
Pseudomonas aeruginosa	Circulation (duct of Cuvier)	100–300 CFU/embryo (48 hpf)	Phages vB_PaeP_PYO2, vB_PaeP_DEV, vB_PaeM_E215 and vB_PaeM_ E217 in cocktail, 500 PFU/embryo (3 hpi)	Phosphatidyl serine/ phosphatidic acid liposomes (prophylactic treatment 24 hpf or therapeutic treatment 3pi)	↘ Bacterial load with phages (~80%) ↘ Bacterial load with liposomes in prophylaxis (~95%) and in therapy (~60%) ↘ Bacterial load with phages and liposomes (~95%)	(19)
Mycobacterium abscessus	Circulation (caudal vein)	300 CFU/embryo (30 hpf)	Phage Muddy, 1.5 × 104 PFU/embryo (1 to 5 dpi)	NA	No significant difference in survival rates with phage ↘ Bacterial burden with phage ↘ Pathological signs with phage	(20)
Zebrafish cystic fibrosis embryo model
Pseudomonas aeruginosa	Circulation (duct of Cuvier/yolk)	30 CFU/embryo (48 hpf)	Phages vB_PaeP_PYO2, vB_PaeP_DEV, vB_PaeM_E215 and vB_PaeM_ E217 in cocktail, 300–500 PFU/embryo (0.5 hpi)	Ciprofloxacin immersion (100 µg/ml)	↘ Bacterial load with phages (~80%) ↘ Inflammatory response with phages ↗ Survival rate with phages (~30%), ciprofloxacin (~40%) and combination of both (~45%)	(18)
Pseudomonas aeruginosa	Circulation (duct of Cuvier)	100–300 CFU/embryo (48 hpf)	Phages vB_PaeP_PYO2, vB_PaeP_DEV, vB_PaeM_E215 and vB_PaeM_ E217 in cocktail, 500 PFU/embryo (3 hpi)	Phosphatidyl serine/ phosphatidic acid liposomes (prophylactic treatment 24 hpf or therapeutic treatment 3pi)	↘ Bacterial load with phages (~70%) ↘ Bacterial load with liposomes in prophylaxis (~70%) and in therapy (~40%) ↘ Bacterial load with phages and liposomes in prophylaxis (~90%) and in therapy (~80%)	(19)
Mycobacterium abscessus	Circulation (caudal vein)	300 CFU/embryo (30 hpf)	Phage Muddy, 1.5 × 104 PFU/embryo (1 to 5 dpi)	Rifabutin immersion (50 µg/ml)	↗ Survival rate with phage (~20%), rifabutin (~20%), or combination of both (~40%) ↘ Bacterial burden and pathological signs with phage, rifabutin or combination of both	(20)
Zebrafish adult model
Klebsiella pneumoniae	Local (intramuscular)	105 CFU/ml	Phage KpG, 108 PFU/mL (2 hpi)	Streptomycin	↘ Bacterial load with phage (~77%), streptomycin (~63%), or combination of both (~98%)	(21)
Escherichia coli	Local (intraperitoneal)	4.106 CFU/adult	Phage ΦEcSw, 4.107 PFU/adult (simultaneously)	NA	↗ Survival rate (~30%)	(22)

^a CF, cystic fibrosis; hpf, hours post-fertilization; dpf, days post-fertilization; hpi, hours post-infection; dpi, days post-infection; ↗, increase; ↘, decrease; NA, non-available.

ZEBRAFISH EMBRYO MODEL IN THE DEVELOPMENT OF PHAGE THERAPY TO TREAT BACTERIAL INFECTIONS

The zebrafish embryo model offers distinct advantages compared to mammalian models for the study of infectious diseases and treatment responses, in accordance with the 3Rs. Embryo studies are generally cost effective and not subject to the same ethical regulations as vertebrate animals, as when using embryos under 5 days post-fertilization (dpf), they are not classified as in vivo assays under European Directive 2010/63/EU, thus simplifying experimental approval process. This makes them an excellent replacement for early-stage studies, significantly reducing the number of mammals required for preliminary research (23, 24). One of their most notable features is their transparency, which allows for real-time observation and imaging of cellular behaviors, infection dynamics, and treatment effects. This transparency, combined with the rapid development of embryos, during which major organ systems form within 5 days, provides valuable insights into the dynamics of bacterial infection and phage activity within a living organism (25, 26). The model versatility allows for testing various methods of infection and sites, enhancing its utility and interest for detailed experimental exploration (Fig. 1). Importantly, the zebrafish embryo innate immune system shares significant similarities with the human immune system, making it an excellent model for understanding how treatments interact with immune responses, which is essential for optimizing treatment strategies (27). However, the zebrafish model also has distinct disadvantages when compared to mammalian models. Anatomical and physiological differences, such as the presence of gills instead of lungs and hematopoiesis occurring in the anterior kidney rather than the bone marrow, can limit the applicability of results to mammals. Additionally, the regulatory frameworks currently mandate data from mammalian models before progressing to human clinical trials, meaning zebrafish studies alone are not sufficient for regulatory approval. This situation could change in the future as regulatory perspectives evolve, but at present, it remains a significant limitation. Although the zebrafish model excels in the principles of reduction and refinement by allowing detailed in vivo studies with fewer ethical concerns, it is not yet a complete replacement for mammalian models in the regulatory landscape.

Fig 1.

Administration sites in zebrafish embryos and adults. This illustration details the various administration sites used in both zebrafish embryos and adults as discussed in this review. Injections are categorized as systemic (duct of Cuvier, caudal vein, yolk), localized (hindbrain ventricle, intramuscular, intraperitoneal), or by bath immersion on healthy or tail-injured embryo. Created with BioRender.com.

In this section, we provide a detailed comparison of the zebrafish embryo model, demonstrating its effectiveness in studying methodologies, outcomes, and implications for treating a variety of human bacterial pathogens with phage therapy. Initially, we present a model of non-pathological embryos to establish a baseline for normal responses to bacterial infections and phage treatments. Subsequently, we focus on zebrafish embryos mimicking cystic fibrosis (CF) as a pathological context, providing insights into how this specific disease model can influence the efficacy and dynamics of phage therapy.

Phage therapy studies in healthy zebrafish embryo model

Enterococcus faecalis

Al-Zubidi et al. (2019) focused on the isolation and characterization of phages against E. faecalis and highlighted the efficacy of the phage SHEF2 in clearing systemic lethal infection in zebrafish (14). E. faecalis is an opportunistic pathogen often implicated in nosocomial infections and exhibits intrinsic resistance to many standard antibiotics. In this study, zebrafish embryos were infected by intravenous injection of E. faecalis OS16 in the duct of Cuvier (30,000 colony-forming units [CFU]/embryo at 30 hours post-fertilization [hpf]), and subsequently treated with either a live phage SHEF2 (6 × 10⁵ plaque-forming units [PFU]/embryo at 2 hours post-infection [hpi]) or a heat-killed SHEF2, and the results were monitored for 72 hpi. The authors found that E. faecalis OS16 caused significant time-dependent lethality in embryos, which was markedly higher than in the control groups treated with phosphate-buffered saline (PBS) or phage alone. Although the heat-killed SHEF2 did not improve survival rates, with a 73% mortality rate identical to that of the infection-only group, the live SHEF2 phage significantly reduced the mortality rate to 16%, demonstrating an 84% survival rate. The zebrafish infected with E. faecalis exhibited severe symptoms such as lack of blood circulation, yolk sac edema and eye abnormalities, pericardial edema, and spinal curvature. Conversely, the majority of phage-treated zebrafish remained healthy, displaying a health status comparable to that of the PBS-injected controls and those injected with phage only. Overall, the study demonstrates that systemic phage treatment following infection significantly reduced mortality rates and enhanced the health of the zebrafish embryos, thus highlighting the phage effectiveness in treating systemic E. faecalis infections. Importantly, the study also confirmed that neither the phage nor the bacterial components released upon bacterial lysis exhibited toxicity to the embryos, suggesting that the phage could be safely used for systemic applications. The zebrafish model proved to be extremely valuable not only for assessing the efficacy of the phages against systemic bacterial infections but also as a dynamic tool for real-time monitoring of infection progression and phage-mediated clearance (14).

Staphylococcus aureus

The study by Plumet et al. (2024) offers significant insights into the use of the zebrafish embryo model for assessing phage therapy efficacy against S. aureus, a pathogenic bacterium responsible for a broad spectrum of infections (15). Zebrafish embryos (30 hpf) were infected with methicillin-resistant S. aureus (MRSA, USA300 and NSA0799 strains) and methicillin-susceptible S. aureus (MSSA, SH1000 strain) with 15,000 CFU/embryo in the hindbrain ventricle and were subsequently treated at 2 hpi with varying doses of the phage SAVM02. Untreated embryos experienced 100% mortality, whereas those receiving phage treatment showed significantly improved survival rates in a concentration-dependent manner, with the highest concentrations achieving 100% survival over 72 hours post-treatment (hpt). Control groups treated with either PBS or phage alone also showed 100% survival, confirming the non-toxicity of the phage. Heat-inactivated SAVM02 phage did not improve survival, verifying that the survival benefit was due to the active phage therapeutic action. Throughout the study, phage-treated embryos remained healthy, underscoring the safety and effectiveness of phage therapy in treating S. aureus infections in this model. Furthermore, phage treatment (7.10⁷ PFU/embryo) significantly reduced the bacterial load in zebrafish embryos infected with MRSA USA300, decreasing it from 15,000 to fewer than 100 CFU/embryo over 72 hpt. Microscopic analysis supported these results, showing a marked reduction in green fluorescence bacteria, which underscores the strong bacteriolytic capability of the SAVM02 phage in effectively diminishing the bacterial burden in the zebrafish infection model. The authors also demonstrated that the phage load increased considerably over 24 hpt before gradually decreasing. Additionally, the study compared the efficacy of phage therapy with vancomycin for treating MRSA infections. It was found that the minimum inhibitory concentration (MIC) for vancomycin against S. aureus strains ranged from 0.5 to 1 µg/mL, confirming their susceptibility to the antibiotic. When administered at different concentrations, vancomycin achieved 100% survival rates in embryos infected with USA300 at the highest concentrations tested, whereas lower concentrations were progressively less effective. In this specific study, a comparison between phage therapy and a single dose of vancomycin (300 µg/mL or 600 × MIC) showed that although both initially reduced bacterial loads, phage therapy demonstrated a more sustained antibacterial effect. However, it should be noted that this is not a head-to-head comparison, as vancomycin in clinical settings is administered in multiple doses due to its limited half-life, which might affect its long-term efficacy. Although vancomycin effectively decreased bacterial counts within the first 24 hpt, its effect diminished over time. In contrast, phage therapy maintained a greater reduction in bacterial load throughout the 72 hpt. Although vancomycin can be effective at high doses for acute infections, its long-term effectiveness diminishes as it is metabolized and cleared. In contrast, phage therapy not only ensures survival but also provides a longer-lasting reduction in bacterial counts, making it potentially more suitable for chronic infections. This study illustrates that the zebrafish embryo model provides crucial insights into the interactions among the host, pathogen, and therapeutic agents (15).

Klebsiella pneumoniae

Assafari et al. (2021) focused on the isolation and characterization of the phage UMP2146, which targets K. pneumoniae, a pathogenic bacterium commonly associated with nosocomial infections (16). In their study, the impact of phage UPM2146 was evaluated using an immersion method with zebrafish embryos (72 hpf). The embryos were exposed to K. pneumoniae 2146 (3.3 × 10⁶ CFU/mL) for either 30 or 90 minutes before direct phage treatment (10⁶ PFU/mL). This study showed that embryos infected for 30 minutes had significantly lower mortality rates, surviving up to 24 hours. Conversely, those infected for 90 minutes reached 100% mortality by 10 hpi. Subsequent treatment with UPM2146 led to a drastic reduction in bacterial load. Specifically, in the phage-treated group, bacterial counts dropped from initial levels to zero by 10 hpt and remained at zero until 20 hours. This contrasts with untreated embryos, where bacterial counts increased sharply. These findings underscore UPM2146 potent bactericidal effect and its potential as a therapeutic agent. Additionally, the study confirmed the safety of UPM2146, positioning this phage as a promising candidate for the safe and effective treatment of K. pneumoniae infections (16).

Acinetobacter baumanii

The study by Neto et al. (2023) validated the use of a bath infection model with tail-injured zebrafish embryos to evaluate the efficacy of phage treatments against A. baumannii strains, a major bacterial pathogen associated with antibiotic resistance-related deaths (17). Phage treatments were administered at a concentration of 10⁹ PFU/mL at 12 hpi to target bacteria introduced at 10⁴ CFU/mL and 48 hpf, focusing on highly virulent strains such as K2, K9, K32, and K45. Control groups within the study underscored the non-toxicity of the phage treatments and highlighted the need for phage specificity, as demonstrated when the non-specific K1 phage failed to improve survival in embryos infected with the K32 strain. Specific phage treatments markedly enhanced survival rates, boosting survival from an average of 35% to 74% for the K32 strain. The effectiveness of these phages was illustrated in significant improvements in survival, particularly against the most virulent strains (17).

Phage therapy studies in pathological zebrafish embryo model of cystic fibrosis

The significant genetic homology shared between zebrafish embryos and humans, along with the ease of genetic manipulation, through techniques like CRISPR/Cas9, transgenesis, and morpholinos (chemically modified oligonucleotides designed to bind to specific mRNA sequences and inhibit protein translation), allows researchers to dissect gene functions and effectively model human diseases (28). Notably, the morpholino of the cystic fibrosis transmembrane regulator (CFTR) gene in zebrafish embryos is a valuable model for CF research. Embryos share structural similarities with human CFTR channels and exhibit increased susceptibility to infections, mimicking CF patient responses. Despite lacking lungs, zebrafish embryos possess mucins similar to those overexpressed in the lungs of CF patients, enabling relevant studies of infection responses across various organs. Zebrafish embryos also exhibit CF-related symptoms such as pancreatic dysfunction and hematopoietic anomalies, providing insights into aspects of CF not observed in mouse model. Additionally, the absence of an adaptive immune system response during the first weeks of development makes them ideal for exploring exclusively innate immune defenses against infections, which are pertinent to human CF (29, 30). The distinction between wild-type (WT) and CF zebrafish embryos plays a pivotal role in understanding the interaction between genetic backgrounds and treatment outcomes in infectious disease research. WT embryos typically exhibit a robust immune response that can be indicative of general immune capabilities in a non-compromised biological system. In contrast, CF embryos, which genetically mimic the human condition of cystic fibrosis, often display altered immune responses and heightened sensitivity to infections. This sensitivity is primarily due to the mutation in the cftr gene, which affects various physiological responses, including those related to bacterial infections.

Pseudomonas aeruginosa

Cafora et al. conducted two studies, in 2019 and 2022, which provide valuable insights into the development and optimization of phage therapy for treating P. aeruginosa infections in zebrafish embryos in both WT and CF contexts. The studies utilized similar phage cocktails, including two Podoviridae (vB_PaeP_PYO2, vB_PaeP_DEV) and two Myoviridae (vB_PaeM_E215, vB_PaeM_E217). These phages were chosen due to their proven efficacy in preliminary tests on Galleria mellonella larvae against P. aeruginosa, a pathogen that is notably difficult to treat due to its inherent and acquired resistance mechanisms (18, 19).

In the study by Cafora et al. (2019), P. aeruginosa PAO1 strain (30 CFU/embryo) was injected into the duct of Cuvier of WT and CF embryos at 48 hpf to assess their susceptibility, revealing that CF embryos exhibited significantly higher lethality rates compared to WT counterparts (83% vs 66% at 20 hpi) (18). To test the efficacy of phage therapy in zebrafish, they first administered the phage cocktail into the yolk sac of PAO1-infected WT embryos (300–500 PFU/embryo) at either 0.5 or 7 hpi, leading to a notable reduction in lethality from 70% to 42% and 38% for early and late injection, respectively. Further experiments confirmed the efficacy of phage therapy in both WT and CF embryos with earlier infection time, showing a decrease in lethality from 66% to 35% in WT and from 83% to 52% in CF embryos. Additionally, phage treatment significantly reduced the bacterial burden in both embryo types by about 80% at 8 hpi, indicating substantial bacterial clearance. Time-lapse analyses revealed that although untreated WT embryos showed increased fluorescence due to bacterial proliferation of the PAO1 strain labeled with GFP, those treated with phages showed no increase, suggesting effective antibacterial activity by the phages. Similar patterns were observed in CF embryos, though with inherently higher bacterial levels. Interestingly, Cafora et al. (2021) showed that CF zebrafish embryos exhibit significantly higher levels of pro-inflammatory cytokines TNF-α and IL-β even without induced PAO1 infection, indicating that the absence of Cftr function inherently triggers a chronic inflammatory state, similar to human CF patients (31). They also found that administering the phage cocktail to these CF embryos, in the absence of bacterial infection, reduced TNF-α and IL-β levels closer to those observed in wild-type embryos. This implies that phages might have a secondary role as immunomodulators, potentially moderating inflammation rather than exacerbating it, especially in environments with pre-existing inflammation, such as in CF. Moreover, their study also examined the combined effects of phage and antibiotic therapy (ciprofloxacin, 100 µg/mL in fish water). CF embryos treated with both phages and ciprofloxacin exhibited the lowest lethality rates compared to those treated with either treatment alone. Thus, the results demonstrate that although both the phage cocktail and ciprofloxacin effectively combat P. aeruginosa infection, their combination provides a superior benefit. This synergistic effect allows for the same therapeutic outcome with reduced dosages of each agent, potentially lowering the risk of developing bacterial resistance against either treatment. Taken together, this study shows that phage therapy was effectively applied in a CF zebrafish model, marking its first successful in vivo use in such a model. Phage therapy not only combatted bacterial infections but also modulated immune responses, reducing inflammation in CF model while mildly increasing cytokine expression in WT. This immunomodulatory potential of phages could make them valuable as anti-inflammatory agents in CF treatment (18, 31).

Cafora et al. (2022) subsequently explored the efficacy of prophylactic and therapeutic treatments of phosphatidylserine/phosphatidic acid (PS/PA) liposomes and the phage cocktail mentioned above on PAO1-infected WT or CF embryos (19). The rationale behind this combination is that although phages target extracellular bacteria, PS/PA liposomes can potentially address intracellular bacteria and stimulate macrophage activity, which is crucial as P. aeruginosa can reside inside macrophages where phages cannot reach (32). The research initially focused on the prophylactic administration of PS/PA liposomes (28 hpf) followed by a PAO1 infection (100–300 CFU/embryo, 48 hpf) and subsequent phage cocktail treatment (500 PFU/embryo, 3 hpi) into the duct of Cuvier. In WT embryos, which exhibit a natural immune response, PS/PA liposomes alone were surprisingly more effective than phage treatment in reducing PAO1 infection, and the combination of PS/PA liposomes with the phage cocktail did not enhance the antimicrobial effect beyond that achieved by the liposomes alone. However, in CF embryos, both PS/PA liposomes and phages showed comparable effectiveness when used individually but acted synergistically when combined, significantly lowering the bacterial burden. The therapeutic potential of PS/PA liposomes and phage cocktail was then assessed under conditions that mimic a clinical scenario where embryos where first infected with PAO1 and then treated. Although both treatments reduced the bacterial load in infected embryos, the therapeutic application of PS/PA liposomes was notably less effective than the prophylactic approach, likely due to the shorter time available for full antimicrobial macrophage activation. Phage cocktail was effective in significantly reducing bacterial burdens in both WT and CF embryos to a similar extent. When used together, PS/PA liposomes and phage cocktail only further decreased the bacterial load in WT embryos, with no significant difference observed in CF embryos compared to phage treatment alone. The studies also examined the effect of these treatments on a phage-resistant PAO1 strain. Treatment with either phage cocktail or PS/PA liposomes reduced the overall bacterial burden by about 40%–50% in both WT and CF embryos. However, the combination therapy was more effective than either treatment alone, suggesting a synergistic effect. Interestingly, the proportion of phage-resistant bacteria was significantly reduced by the combination treatment in both embryo types, indicating that PS/PA liposomes effectively helped to control the infection, including the one caused by the phage-resistant strain. These findings underscored the potential of using PS/PA liposomes in conjunction with phage therapy to treat PAO1 infections effectively but also to enhance outcomes in settings where phage resistance is a concern. The ability of PS/PA liposomes to augment the effects of phage therapy, particularly against resistant strains, provides a promising avenue for the development of comprehensive treatment strategies for bacterial infections in CF and other clinical contexts (19).

These studies collectively highlight the versatility and potential of the zebrafish embryo model for evaluating phage therapy in treating P. aeruginosa infections, particularly in a CF context. Although zebrafish embryos do not replicate the chronic P. aeruginosa infections typical in CF patients, they offer a rapid, cost-effective system for in vivo testing. The conservation of the innate immune response between fish and humans allows for insights into how macrophages, which are often defective in CF, can be activated by liposomes. Despite lacking lungs, zebrafish offers a valuable genetic model for studying CF due to their optical transparency and the ease of genetic manipulation. This model helps in understanding CF pathophysiology and developing new treatments. Additionally, these studies emphasize the need for innovative approaches that combine phages with other therapeutic modalities to optimize treatment efficacy and enhance patient outcomes (18, 19).

Mycobacterium abscessus

The study by Johansen et al. (2021) focused on the application of phage therapy to treat M. abscessus infections in zebrafish embryos in WT and CF contexts (20). Two phages were used in this research, Muddy, which efficiently kills the M. abscessus GD01 strain, and Gabriela, which infects and kills M. abscessus very inefficiently compared to Muddy. Following initial confirmation that M. abscessus infection could be established via caudal vein injection (300 CFU/embryo, 30 hpf), the application of Muddy (1.5 × 10⁴ PFU/embryo, 1 to 5 days post-infection, dpi) showed no significant difference in survival rates of infected WT embryos. However, a notable reduction in bacterial burden was achieved with Muddy, as measured by fluorescent pixel count (FPC), contrasting with the lesser effect observed with Gabriela. This reduction in bacterial load by Muddy did not translate into increased survival but did significantly decrease pathological signs associated with infection, such as cord formation and abscess development. Specifically, Muddy treatment resulted in about a twofold decrease in cord formation, a hallmark of acute infection in zebrafish, at 2, 4, and 6 dpi. In contrast, Gabriela alone did not produce a significant change in cord formation, and combining Muddy with Gabriela did not enhance the effect over Muddy alone. Regarding abscess formation, which indicates severe infection, both Muddy alone and in combination with Gabriela significantly reduced the occurrence of abscesses at 4 and 6 dpi, whereas Gabriela alone did not affect abscess formation. These findings underscore Muddy’s ability to effectively lyse M. abscessus and reduce severe infection markers in zebrafish embryos, although it did not extend larval survival within the study’s 12-day observation period. This suggests that although phage therapy with Muddy can reduce the bacterial load and pathology, extending treatment duration might be necessary to observe potential improvements in survival.

The authors then assessed the role of functional innate immunity, particularly the involvement of macrophages, in supporting the efficacy of the phage Muddy in treating M. abscessus infections in zebrafish embryos (20). Macrophages are crucial for controlling this infection, as they are integral to granuloma formation and the containment of mycobacterial growth. Depletion of macrophages using liposomal clodronate led to rapid larval death within days, emphasizing the vital role these cells play in managing mycobacterial infections. When macrophages were depleted, infection with M. abscessus resulted in 100% mortality by 8 dpi, a significantly more severe outcome compared to infections in WT embryos or those treated with liposomal PBS. Despite the administration of phages (Muddy, Gabriela, or both), no improvement in survival rates was observed in macrophage-depleted embryos, with median times to 50% mortality remaining at 6 and 7 days, respectively. Furthermore, phage therapy did not alter the bacterial burden in these embryos, nor did it reduce the incidence of cords and abscesses, which were drastically increased in macrophage-depleted larvae compared to controls. This was further evidenced by the presence of large infection foci in phage-treated embryos compared to untreated ones, as demonstrated by FPC analysis and whole-embryo imaging. These findings underscore the indispensable role of macrophages in the effectiveness of phage therapy against M. abscessus, highlighting the necessity of a functional innate immune system to achieve successful outcomes in phage therapy.

Moreover, the study validated phage treatment against M. abscessus in CF zebrafish embryos (20). The results showed that M. abscessus-infected CF embryos were highly susceptible to infection, displaying rapid and severe symptoms including increased bacterial burden, pronounced cording, and abscess formation. Phage treatment, using either Muddy alone or in combination with Gabriela, significantly improved outcomes in CF embryos. Specifically, treatment extended survival (median time to 50% mortality increased from 8 to 10 days) and enhanced overall survival rates (20% in untreated vs approximately 40% in Muddy-treated embryos at 12 dpi). Moreover, phage therapy markedly reduced the physical manifestations of infection such as cords and abscesses at various days post-infection, as well as the bacterial load, compared to untreated controls. These findings suggest that CF zebrafish, like young CF patients, are hypersensitive to M. abscessus infection, but respond favorably to phage therapy. The effective reduction in infection severity and improvement in survival rates highlight the potential of phage therapy as a treatment option for CF-related infections, underscoring the utility of the zebrafish CF model in studying the effectiveness of therapeutic interventions in a context mimicking human disease. In addition, this study also investigated the efficacy of combining phage therapy (Muddy) with the antibiotic rifabutin in CF embryos with M. abscessus infection. The results demonstrated that although treatment with rifabutin alone or Muddy alone significantly improves survival rates compared to untreated CF embryos, the combination of both treatments substantially enhances the therapeutic outcomes. Specifically, the combination of Muddy and rifabutin not only increases the survival rate of infected CF zebrafish to 70% at 12 dpi, comparable to survival rates in WT zebrafish, but also extends the median time to 50% mortality beyond the calculation threshold, indicating a significant reduction in mortality. The results underscore the complex interactions between phages, the host immune system, and bacteria, highlighting the need to consider a patient’s immunological status before implementing phage therapy. Overall, the zebrafish embryo model supports the development of phage therapies and could serve as a preclinical platform for assessing effective phage and antibiotic combinations, providing crucial insights before clinical application in M. abscessus patients (20).

THE ZEBRAFISH ADULT MODEL FOR ADVANCED PHAGE THERAPY

The adult zebrafish model provides complementary advantages to the embryo model, making it a valuable tool for preclinical research. Its fully developed immune system, which includes both innate and adaptive immunity, allows for studies that closely mimic human immune responses, offering relevant insights into disease mechanisms and treatment outcomes. Additionally, adult zebrafish facilitate long-term studies essential for chronic diseases research and sustained treatment evaluations. With fully formed and matured organs, they are pivotal for investigating organ-specific diseases and functions. Moreover, the increased size of adult zebrafish allows for the administration of larger quantities of bacteria, phages, or active substances, which is advantageous when studying the effects of these agents at realistic physiological levels. However, this can sometimes pose a problem when compound is precious or limited in quantity. Unlike their embryonic counterparts, adult zebrafish lose the transparency that facilitates direct observation, necessitating the fixation of tissues and the production of histological sections for detailed study. This requirement can complicate the research process and may limit the immediate visual assessment of internal processes (9).

In this section, we highlight the adult zebrafish model’s effectiveness in validating the phages’ efficacy against human bacterial pathogens.

Klebsiella pneumoniae

In their 2021 study, Sundaramoorthy et al. explored the safety and therapeutic potential of phage KpG in treating K. pneumoniae infections within adult zebrafish model (21). They monitored liver and brain enzyme profiles for potential toxicity and assessed the therapeutic impact through phage administration for treating K. pneumoniae infections. Zebrafish injected with phage KpG showed no significant variations in liver enzyme profiles, and a slight but statistically insignificant increase in brain acetylcholine esterase levels. Histological analysis of muscle and liver tissues, stained with hematoxylin and eosin, confirmed the absence of morphological or biochemical aberrations, suggesting that phage treatment does not induce adverse effects in zebrafish. To assess the in vivo efficacy of the phages, they infected zebrafish intramuscularly with the MTCC 432 strain of K. pneumoniae (10⁵ CFU/mL). Treatments included streptomycin, phage KpG (10⁸ PFU/mL), or a combination of both administered intramuscularly at 2 hpi. Results indicated that phage KpG alone reduced bacterial counts in muscle tissue by 77%, whereas streptomycin alone achieved a 62.8% reduction. The combination of both treatments significantly enhanced efficacy, leading to a 98% decrease in bacterial counts. These findings underscore the utility of the adult zebrafish model in advancing our understanding of infectious diseases and treatment strategies (21).

Escherichia coli

In 2020, Easwaran et al. explored the therapeutic potential of phage ΦEcSw in treating E. coli bacterial infections in adult zebrafish that were infected intraperitoneally with the E. coli Sw1 strain (4.10⁶ CFU/fish) and simultaneously treated with phage ΦEcSw (4.10⁷ CFU/fish) (22). Zebrafish treated with ΦEcSw exhibited a 100% survival rate up to 84 hpi, markedly higher than the 70% survival rate observed in the E. coli Sw1-infected group without phage treatment. The findings from the Easwaran et al. (2020) study demonstrate ΦEcSw potential as an effective therapeutic agent against E. coli Sw1 infections in adult zebrafish, reinforcing the relevance of using adult zebrafish to study specific phage therapy (22).

DISCUSSION

The zebrafish embryo model has become a useful tool for studying the dynamics of infectious diseases and evaluating treatment responses in a controlled environment (Fig. 2; Table 2). Some of the research mentioned above has focused on comparing WT and CF zebrafish embryos to understand how genetic variations influence their responses to bacterial infections and phage therapy treatments. Studies have shown that CF embryos exhibit higher susceptibility to severe infections and increased mortality rates compared to WT embryos when exposed to the same pathogenic challenges. For instance, CF embryos infected with the same concentration of P. aeruginosa show significantly higher mortality rates due to the impact of the cftr mutation on their ability to manage bacterial threats (18, 19). Similar vulnerability is observed in CF embryos infected with M. abscessus, resulting in bacterial burdens and more severe clinical manifestations (such as cording and abscess formation) compared to WT embryos (20). Phage therapy efficacy also varies between the two genetic types, with WT embryos responding better to phage treatments due to their more robust immune system synergizing with phage bactericidal actions. These findings from the described studies illustrate that genetic differences between WT and CF zebrafish significantly influence their response to infections and treatments. Additionally, in CF embryos, the mutation contributes to other complications, such as pancreatic dysfunction, which can affect the overall health and treatment outcomes in these models. Therefore, understanding these differences is crucial for tailoring phage therapy strategies to enhance efficacy and reduce mortality, particularly in vulnerable populations, and CF zebrafish embryos represent a valuable tool in achieving these goals. Currently, only the CF model has been tested for phage therapy treatment; however, other pathological models could also prove highly useful for testing various infections.

Fig 2.

Workflow for zebrafish model in phage therapy. Adult zebrafish are maintained, crossed, and raised according to general guidelines provided by the Zebrafish Information Network (ZFIN). Embryos are collected, and when necessary, morpholino microinjection is performed for gene knockdown studies. The embryos are then manually or enzymatically dechorionated to remove the chorion. In the infection stage, adult zebrafish are injected with a bacterial inoculum, and embryos can be either microinjected or bathed with bacterial inoculum. During the treatment phase, adults receive an injection of phage or antibiotic, whereas embryos undergo administration of phage, antibiotic, or liposomes through microinjection or balneation. The analysis phase includes various methods to evaluate the outcomes, such as histological examination, survival rate assessment, bacterial burden measurement, observation of pathological signs, live imaging, phage load determination, and inflammatory response analysis. The color-coded legend indicates the different statuses of the zebrafish during these procedures: yellow for anesthetized, and red for anesthetized and sacrificed.

TABLE 2.

Comparative summary of advantages and limitations of zebrafish embryo vs adult models in phage therapy research

Aspect	Zebrafish embryos	Adult zebrafish
Immune system	Simplified immune system; ideal for studying innate immunity	Fully developed immune system; includes both innate and adaptive responses
Real-time imaging	Transparency allows for real-time observation of infection dynamics and phage activity	Less transparency, making real-time imaging more challenging
Ethical considerations	Less regulated for embryos up to 5 days post-fertilization	More stringent ethical regulations due to vertebrate status
High-throughput screening	Suitable for large-scale studies due to high reproductive rate and lower cost	Lower throughput due to size and cost constraints
Genetic manipulation	Easier and quicker genetic manipulations (e.g., CRISPR/Cas9, morpholinos)	Genetic manipulations possible but more complex and time consuming
Number of individuals	Large number of individuals can be tested in a single experiment	Fewer individuals can be tested; more labor intensive to raise in large numbers
Complexity of organs	Less complex organ systems; may not fully mimic human physiology	Fully developed and more complex organ systems, better mimicking human physiology
Disease modeling	Limited to early developmental stages; not suitable for chronic or long-term disease studies	Suitable for chronic and long-term disease studies due to longer lifespan
Anatomical differences	Differences in hematopoiesis (anterior kidney vs bone marrow) and respiratory system (gills vs lungs)	Closer to human anatomy in terms of organ systems and physiological functions
Phage therapy specifics	Allows rapid screening of phage efficacy and dynamics in a controlled environment	Provides insights into long-term phage therapy effects and interactions with a fully developed immune system

In addition, it is also important to discuss various parameters that are essential for comparing the models, including experimental variables such as the timing, methods of administration, dosages used, and how these correlate with mortality rates. The age of zebrafish embryos during experimentation impacts study results on infectious diseases and treatments. The physiological characteristics and immune system capabilities of embryos undergo rapid developmental changes within the first few days post-fertilization, which can influence their response to infections and treatments. Innate immune response begins emerging around 24 hpf, but complete functional maturity occurs around 72 hpf with the presence of functional macrophages and neutrophils (33–35). Consequently, embryos at different developmental stages may exhibit varied susceptibility to pathogens and differing efficacies of therapeutic interventions. For example, younger embryos (less than 24 hpf) may be more susceptible to bacterial infections due to their less mature immune system and the absence of neutrophil population, potentially leading to higher mortality rates. Conversely, older zebrafish embryos and larvae demonstrate more robust immune responses and, therefore, higher survival rates when exposed to the same pathogenic load.

The method of administering treatments in zebrafish embryos is crucial for optimizing therapeutic efficacy and understanding the dynamics of infection control. Various administration routes can impact the distribution, absorption, and effectiveness of treatments, particularly in the context of phage therapy. Intravenous injection, typically into the duct of Cuvier or the caudal vein, is a common method for introducing phages directly into the bloodstream of zebrafish embryos. This route ensures rapid dissemination of the treatment throughout the organism, providing immediate therapeutic effects. For instance, intravenous administration of phage cocktails in CF embryos significantly reduced bacterial loads and improved survival rates, demonstrating the efficacy of this approach for systemic infections (14, 18–20). Intraperitoneal injections which potentially exposes multiple organs to phages, also demonstrated no adverse effects suggesting that phages may be safely used without triggering harmful systemic immune responses in adult zebrafish (22). Administering treatments intramuscularly or into the hindbrain ventricle target specific regions for localized infections. These methods were used to treat E. coli in adult zebrafish and S. aureus infections in zebrafish embryos, resulting in significant reductions in mortality rates. The localized administration helps concentrate the phage at the site of infection, enhancing the localized immune response and therapeutic effectiveness (15, 21). The immersion method involves exposing zebrafish embryos to a solution containing phages or antibiotics, allowing for a more generalized treatment application. This is particularly useful for treating external infections or when a less invasive method is preferred (16, 17). Studies using the immersion approach with embryos exposed to K. pneumoniae demonstrated its potential, with treated embryos showing drastically lower mortality rates and reduced bacterial loads compared to controls. While each administration route has its advantages, the choice of method depends on the type of infection, the desired speed of therapeutic action, and the specific physiological targets. Therefore, intravenous and hindbrain ventricular injections provide controlled and immediate effects, ideal for acute or localized infections. In contrast, immersion method is more suitable for external infections or when a broader treatment distribution is needed. The efficacy of these methods depends on the pathogen involved and the health status of the embryos. For example, the effectiveness of immersion can be influenced by the water solubility and stability of the phage or drug ability to penetrate biological tissues. Carefully selecting the method of treatment administration can significantly influence the outcome of therapeutic interventions. Understanding the nuances of each method allows for the development of more tailored and effective treatment strategies, thereby improving overall success rates in combating infectious diseases.

The relationship between phage dosages and mortality rates in zebrafish embryos infected with various bacterial species is pivotal in optimizing phage therapy. The studies reviewed demonstrate how dosage levels can significantly affect the survival outcomes, emphasizing the importance of precision in phage application. The research on S. aureus indicated a concentration-dependent survival benefit, where higher doses of phages correlated with better survival rates. For instance, untreated embryos experienced 100% mortality, whereas those receiving the highest phage concentration had a 100% survival rate over 72 hpt, highlighting the need for adequate phage dosing against aggressive strains like MRSA (15). Moreover, consistently between the S. aureus and E. faecalis studies, the authors showed that live phage significantly reduced mortality rates compared to heat-killed phages, emphazing the importance of phage viability alongside dosing (14, 15). Finally, study on treatments against A. baumannii requiring high doses of specific phages to improve survival rates underscored the need for phage specificity in addition to dose optimization (17). These examples illustrate the critical role of phage dosage in determining the success of infection control. Optimal dosages not only ensure the reduction of bacterial loads but also contribute to the health and survival of the host.

Moreover, combining phage therapy with antibiotics has shown promise in enhancing therapeutic efficacy and addressing complex bacterial infections in adult zebrafish and zebrafish embryos. However, it is essential to evaluate the effectiveness of each combination independently due to varying synergistic effects. In the study involving S. aureus by Plumet et al. (2024), phage SAVM02 was tested against MRSA infections in zebrafish embryos. The study found that high doses of SAVM02 alone were as effective as the combination of the phage with vancomycin, indicating that in some cases, phage therapy alone may be sufficient to effectively manage infections (15). In addition, a study by Cafora et al. (2022) demonstrated that liposomes combined with phages effectively helped to control the infection, including those caused by resistant strains (19). Furthermore, when phage therapy was combined with the rifabutin in CF embryos, the therapeutic outcomes were substantially enhanced, showing a greater reduction in mortality and pathology than when either treatment was used alone. This synergistic effect underscores the potential for combined treatments in managing severe infections, especially in genetically predisposed populations (20).

Zebrafish models could also be used to gain valuable insights into several other critical aspects of phage therapy research, beyond just assessing therapy efficacy. For example, zebrafish models could be used to study the pharmacodynamics and pharmacokinetics (PD/PK) of phage therapy. It would be very interesting to test if the zebrafish model could allow real-time tracking of phages within a living organism, potentially providing critical data on absorption, distribution, metabolism, and excretion (ADME) of phages, though this remains to be demonstrated. This could be particularly useful in determining optimal dosing and understanding the interaction between phages and the host biological systems. The timing of phage administration is another crucial aspect for maximizing therapeutic efficacy. While our review primarily focuses on efficacy assessments, it is important to note that zebrafish models can be used to investigate the critical aspect of therapy timing. For instance, Cafora et al. (2019) tested different times of phage administration post-infection, finding similar results for both early (about 30 minutes) and late (7 hours) injections (18). This suggests that the timing of phage administration may not significantly impact efficacy, though this remains to be confirmed by further studies as it is the only known study addressing this aspect. Additionally, the innate immune response to phages is a critical area of study, as it can influence the efficacy and safety of phage therapy. Zebrafish, with their well-characterized innate immune system, provide an excellent model for studying these interactions as studies have shown that phages can impact or require the host immune response (20, 31).

CONCLUSIONS

In conclusion, zebrafish models have significantly advanced our understanding of phage therapy in the context of bacterial infections, making a rapidly evolving field bolstered by recent publications. These models are essential for elucidating bacterial pathogenesis and gauging the efficacy of treatments, particularly in the exploration of diverse therapeutic combinations. This research complements broader studies that demonstrate the efficacy of phage-antibiotic synergies against multidrug-resistant organisms, thereby establishing phage therapy as a vital element of comprehensive antimicrobial strategies. Despite anatomical and physiological distinctions from mammals, such as possessing gills rather than lungs and conducting hematopoiesis within the anterior kidney instead of bone marrow, zebrafish models provide critical preclinical insights. They expedite the drug discovery and development trajectory by swiftly identifying compounds that are either ineffective or unsafe, thereby conserving both time and resources. This initial screening is pivotal, ensuring that only the most promising therapeutic candidates progress to more intricate animal models and, ultimately, to human clinical trials. Zebrafish models offer a rapid, cost-effective, and ethically responsible approach for assessing novel therapeutic agents, thereby serving an intermediary role in the translation of new treatments from laboratory research to clinical practice. However, it is important to recognize several limitations of the zebrafish model. There is significant variability in experimental protocols across different laboratories, such as variations in inoculums and administration times, leading to inconsistent results. Anatomically, zebrafish have gills instead of lungs and undergo hematopoiesis in the anterior kidney rather than the bone marrow, which can affect the generalization of results to mammals. The smaller size of zebrafish limits complex surgical interventions, and the reduced complexity of some organs compared to mammals may impact the study of certain diseases. Additionally, while the zebrafish immune system is similar to that of mammals, there are key differences that can influence the outcomes of phage therapy studies, particularly in embryos that lack a fully developed adaptive immune system. Finally, some human pathologies, such as chronic or long-term conditions, may not be perfectly mimicked in zebrafish. Despite these limitations, the zebrafish model remains a valuable tool in preclinical research, providing critical insights towards the development of new therapeutic strategies.