Preclinical Safety Evaluation of Intravenously Administered SAL200 Containing the Recombinant Phage Endolysin SAL-1 as a Pharmaceutical Ingredient

Abstract

Phage endolysins have received increasing attention as potent antibacterial agents. However, although safety evaluation is a prerequisite for the drug development process, a good laboratory practice (GLP)-compliant safety evaluation has not been reported for phage endolysins. A safety evaluation of intravenously administered SAL200 (containing phage endolysin SAL-1) was conducted according to GLP standards. No animals died in any of the safety evaluation studies. In general toxicity studies, intravenously administered SAL200 showed no sign of toxicity in rodent single- and repeated-dose toxicity studies. In the dog repeated-dose toxicity test, there were no abnormal findings, with the exception of transient abnormal clinical signs that were observed in some dogs when daily injection of SAL200 was continued for more than 1 week. In safety pharmacology studies, there were also no signs of toxicity in the central nervous and respiratory system function tests. In the cardiovascular function test, there were no abnormal findings in all tested dogs after the first and second administrations, but transient abnormalities were observed after the third and fourth administrations (2 or 3 weeks after the initial administration). All abnormal findings observed in these safety evaluation studies were slight to mild, were apparent only transiently after injection, and resolved quickly. The safety evaluation results for SAL200 support the implementation of an exploratory phase I clinical trial and underscore the potential of SAL200 as a new drug. We have designed an appropriate phase I clinical trial based on the results of this study.

INTRODUCTION

SAL200 is a new phage endolysin-based candidate drug for the treatment of Staphylococcus aureus infections. SAL200 is formulated for injection and contains the recombinant phage endolysin SAL-1 as its active pharmaceutical ingredient. SAL-1 is derived from the bacteriophage SAP-1, which infects staphylococci, including methicillin-resistant Staphylococcus aureus (MRSA). The in vitro and in vivo antibacterial properties underlying the potency of SAL200 as an anti-MRSA agent have been described previously (1, 2). In our previous study, SAL-1 exhibited rapid and effective bactericidal activity against encapsulated and biofilm-forming S. aureus cells, as well as against planktonic S. aureus cells. SAL-1 also exhibited broad-spectrum lytic activity against S. aureus isolates, and intravenous injection of SAL200 into a mouse model of MRSA infection prolonged the viability of the mice and significantly reduced the bacterial counts in the bloodstream and spleen tissue.

Phage endolysins, also called phage lysins or lysins, are bacteriophage-encoded peptidoglycan-degrading enzymes that have evolved to rapidly break down bacterial cell walls, thereby releasing the phage progeny (3). When phage endolysins are exogenously applied as purified recombinant proteins to Gram-positive bacteria, the endolysins induce the rapid lysis and death of the bacterial cells (4–6). Since their discovery, the use of phage endolysins as antibacterial agents has been proposed due to their distinct mode of action and highly specific antibacterial activity, which is independent of the antibiotic susceptibility pattern of the bacterium (5). A number of studies have applied phage endolysins to target pathogens such as Bacillus anthracis (7), Streptococcus pneumoniae (8), S. aureus (9), and Bacillus thuringiensis (10), and promising results have been observed in animal models of human disease (8, 11–18). Thus, phage endolysins represent a promising line of research for the discovery and development of novel antibacterial therapeutic agents (19).

Although safety evaluation is a prerequisite for the drug development process, few studies of the toxicity and safety pharmacology of phage endolysins have been performed in compliance with good laboratory practice (GLP). In this article, we present a GLP-compliant safety evaluation of a new phage endolysin-based candidate drug, SAL200.

MATERIALS AND METHODS

Test materials and animals.

SAL200 was formulated using SAL-1, which was prepared as described previously (2) using a clinical-grade master cell bank produced in accordance with good manufacturing practice (GMP). The purity of the SAL-1 used was greater than 95%, as confirmed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and size exclusion high-performance liquid chromatography. The endotoxin level of the test article was less than 0.5 endotoxin unit (EU)/mg. In this study, specific-pathogen-free Sprague-Dawley rats (5 to 6 weeks old) and beagle dogs (5 to 6 months old) were used for animal experiments. The animal experiments were reviewed and approved by the Institutional Animal Care and Use Committee and performed at the Korea Institute of Toxicology (KIT), Daejeon, Republic of Korea.

General toxicology tests.

General toxicology tests were conducted in compliance with GLP, as outlined in Table 1. In the single-dose rat study, SAL200 was intravenously injected into the tail vein at doses of 0 (vehicle), 25, 50, and 100 mg/kg (SAL-1 dosage). Mortality and clinical signs (general appearance, posture/body position, consciousness/attitude, behavior, breathing, and salivation/vomiting) were recorded every hour for 6 h after injection on the day of administration (day 1) and once daily from day 2 to day 15. Body weight was measured prior to the injection and on days 2, 3, 5, 7, and 15. All animals were macroscopically examined via necropsies performed on day 15.

TABLE 1.

Outline of the general toxicology tests

Toxicity test	Animal	Group	No. of animals/group	Doses	Comments
Single dose	Rat	Main group	5 males + 5 females	0, 25, 50, 100 mg/kg	2-wk observation
Repeated dose	Rat	Main group	10 males + 10 females	0, 4, 10, 25 mg/kg/day	4-wk administration
		Recovery group	6 males + 6 females	0, 25 mg/kg/day	4-wk administration + 2-wk recovery
	Dog	Main group	3 males + 3 females	0, 2.5, 10, 25 mg/kg/day	2-wk administration
		Recovery group	2 males + 2 females	0, 25 mg/kg/day	2-wk administration + 4-wk recovery

In repeated-dose studies, rats and dogs were administered SAL200 once per day for the entire administration period: 4 weeks for rats and 2 weeks for dogs. The dosages were determined based on the results of previous dose range finding studies: 0 (vehicle), 4, 10, and 25 mg/kg for rats and 0 (vehicle), 2.5, 10, and 25 mg/kg for dogs. The recovery periods were 2 weeks for rats and 4 weeks for dogs. After injection, mortality and clinical signs were recorded twice per day for rats and 3 times per day for dogs throughout the administration period and once per day throughout the recovery period. Body weight and food consumption were measured prior to the initial injection and once per week throughout the administration and recovery periods. An ophthalmological examination was performed prior to the injection and within 1 week before necropsy. Urinalysis was performed using the urine collected from each animal on the day of necropsy. The main and recovery group animals were sacrificed at the end of the dosing and recovery periods, respectively. Using the blood samples collected during the necropsy, hematological analyses were conducted using an ADVIA 120 hematology system (Bayer, Tarrytown, NY, USA) and an ACL 9000 coagulation analyzer (Instrumental Laboratory, Milan, Italy). Serum biochemical assays were conducted using a biochemical autoanalyzer, the TBA 200FR NEO (Toshiba, Tokyo, Japan). The necropsy examination included organ weight measurement and a histopathological examination of tissue sample slides using an Olympus X71 optical microscope (Olympus, Tokyo, Japan). In the repeated-dose toxicity tests in dogs, an electrocardiographic examination was also performed. Electrocardiograms (ECGs) were acquired with a Cardiomax FX-3010 (Fukuda Denshi, Tokyo, Japan) prior to injection, on day 13, and during the recovery period (day 41). The following electrocardiography parameters were recorded: heart rate, PR interval, QRS complex, QT interval, and QTc interval.

Safety pharmacology tests.

Safety pharmacology tests were conducted in compliance with GLP, as outlined in Table 2. The safety pharmacology tests included an assessment of effects on vital functions, such as cardiovascular, central nervous, and respiratory system functions. A modified Irwin screen test was used for the central nervous system function test (20). Rats were administered a single dose of SAL200 (0 [vehicle], 0.5, 12.5, or 25 mg/kg), and behavioral changes, physiological and neurotoxicological signs, and rectal body temperature were monitored at specific time points (0 [predose], 0.5, 1, 2, 6, 12, and 24 h postdose). Respiratory function was also evaluated in the rats that received a single dose of SAL200 (0 [vehicle], 0.5, 12.5, or 25 mg/kg) by measuring the respiratory rate, tidal volume, and minute volume at 0 (predose), 0.5, 1, 2, 6, and 24 h postdose. For the cardiovascular safety pharmacology evaluation, an in vitro hERG (human Ether-à-go-go-related gene) assay and an in vivo telemetry study were conducted. The hERG activity was measured by a conventional method. Briefly, the hERG assay was performed using a stable Chinese hamster ovary cell line expressing hERG obtained from bSys GmbH (Witterswil, Switzerland). The CHO-hERG cell line was cultured in complete Dulbecco's modified Eagle's medium/nutrient mixture F-12 medium (GIBCO BRL, Grand Island, NY, USA) supplemented with 9% fetal bovine serum (WelGENE, Daegu, Republic of Korea) and 50 μg/ml hygromycin B (Invitrogen, Carlsbad, CA, USA) at 37°C under 5% CO₂. Cells were grown until 80% confluence was reached. The hERG activity under the conditions applied (50, 200, 350, and 500 μg/ml) was measured with the following protocol. Cells were hyperpolarized from a holding potential of −80 to −90 mV for 100 ms, and then the cells were depolarized to +20 mV for 2 s, followed by a repolarization to −30 mV for 3 s. A series of pulses was applied to the cells over a 20-s interval for the formation of tail currents. Positive (100 nM E-4031) and negative (external buffer with 3% [vol/vol] formulation buffer) controls were applied within each plate to evaluate the data quality. A separate cardiovascular telemetry study was performed with 4 male beagle dogs implanted with telemetry transmitters (Data Sciences International, St. Paul, MN, USA) for remote cardiovascular function monitoring. This study was conducted according to a 4-by-4 balanced Latin square design (21, 22). For each administration, blood pressure, heart rate, and ECG parameters were monitored continuously via telemetry for 1 h prior to and 24 h after injection. Clinical signs, including mortality/morbidity, were observed once per day during the safety pharmacology study. The ECG parameters recorded included ECG intervals (PR, QRS, QT, RR, and QTc intervals) and the ECG waveform. Telemetry data were acquired at 0 (predose), 0.5, 1, 2, 6, and 24 h after injection. The cardiovascular parameters were analyzed using NOTOCORD-hem software (Notocord, Paris, France). For the first administration, 4 dogs were given a single dose of SAL200 at one of 4 dose levels (0 [vehicle], 0.5, 12.5, or 25 mg/kg). The next administration was performed after a 1-week washout period. For this administration, the animals were rotated to a different dose. Dosing was continued at 1-week intervals until each animal had been administered all of the different dosages.

TABLE 2.

Outline of the safety pharmacology tests

Test	Animal or cell line	No. of animals/group	Doses	Comment
Central nervous system	Rat	8 males + 8 females	0, 0.5, 12.5, 25 mg/kg
Respiratory system	Rat	8 males	0, 0.5, 12.5, 25 mg/kg
Telemetry study	Dog	4 males	0, 0.5, 12.5, 25 mg/kg	Each dog received all 4 doses in a Latin square design (with 1-wk washout intervals)
hERG assay	CHO-hERG cell line		50, 200, 350, 500 μg/ml

Analysis of anti-SAL-1 antibody and C3 complement levels in the blood.

To evaluate the immunogenicity of SAL200 administration, anti-SAL-1 antibody production and C3 complement levels in the blood were investigated. Samples were collected from the repeated-dose studies. In the rat experiments, immune sera were obtained on day 0 (predose), day 14, day 28, and the final day (day 42) of the recovery period. In the dog study, immune sera were obtained on day 0 (predose), day 14, and the middle and final days of recovery (days 28 and 42). Anti-SAL-1 antibodies in the serum were detected with a conventional bridging enzyme-linked immunosorbent assay (ELISA) with streptavidin-biotin detection.

The C3 complement level in the sera obtained from the dogs was estimated based on a conventional double-antibody sandwich ELISA using a canine complement factor 3 ELISA quantitation kit (GenWay Biotech, San Diego, CA, USA) according to the manufacturer's instructions.

RESULTS

Single-dose toxicity test in rats.

The single-dose toxicity test was performed in Sprague-Dawley rats. No animals died during the test. Drug administration was not associated with abnormal clinical signs or abnormal body weight changes. In addition, there were no abnormal necropsy findings. All dose levels (25, 50, and 100 mg/kg) were well tolerated, with no signs of toxicity.

Repeated-dose toxicity test in rats.

No animal deaths, no clinical signs, and no abnormal changes were noted in the 4-week repeated-dose toxicity study or the subsequent 2-week recovery period in Sprague-Dawley rats. All dose levels were well tolerated. Therefore, the no-observed-adverse-effect level (NOAEL) for the intravenous administration of SAL200 in rats was considered to be higher than 25 mg/kg in both sexes.

Repeated-dose toxicity test in dogs.

No animals died during the 2-week repeated-dose toxicity study in beagles with a 4-week recovery period. Similar to the 4-week rat study, there were no treatment-related changes in body weight, food consumption, ophthalmology, electrocardiography, urinalysis, hematology, serum biochemistry, organ weights, or macroscopic and microscopic examinations. However, clinical signs, including subdued behavior, prone position, irregular respiration, and vomiting, were observed transiently following injections (Table 3) beginning 10 days after the initial administration; these clinical signs resolved after 30 min to 1 h. The abnormal clinical signs were not observed at any other time during the recovery period.

TABLE 3.

Observed clinical signs^a

Sex	Group	Animal no.	Observation	Day(s) of observation
Male	G2	6	Subdued behavior	10
			Vomiting	12
		8	Subdued behavior	11, 13
			Vomiting	7–8
	G3	9	Vomiting	8, 13–14
		10	Subdued behavior	10, 13
			Vomiting	7–9
		11	Vomiting	10
	G4	12	Prone position	9, 12
			Subdued behavior	9–13
			Vomiting	7, 11
		13	Subdued behavior	10–11
			Vomiting	8, 12, 14
		14	Subdued behavior	11, 13
			Vomiting	7
		15	Prone position	14
			Subdued behavior	10–13
			Irregular respiration	14
			Vomiting	11
		16	Prone position	11
			Subdued behavior	10, 12–13
			Irregular respiration	10–14
			Vomiting	7, 9, 12, 14
Female	G2	22	Prone position	11, 13–15
			Subdued behavior	12–13, 15
			Irregular respiration	13–14
			Vomiting	10
		23	Prone position	14–15
			Subdued behavior	13, 15
			Vomiting	15
		24	Subdued behavior	13, 15
	G3	25	Prone position	11–12, 14
			Subdued behavior	10, 12–13, 15
			Irregular respiration	11, 15
			Vomiting	1, 7
		26	Prone position	10
			Subdued behavior	15
			Irregular respiration	11–13, 15
			Vomiting	12–14
		27	Prone position	14–15
			Subdued behavior	10–13, 15
			Vomiting	7, 11–13, 15
	G4	28	Subdued behavior	13, 15
			Vomiting	7
		29	Prone position	9, 11–15
			Subdued behavior	9, 13, 15
			Irregular respiration	10–13
			Vomiting	7–15
		30	Prone position	11, 15
			Subdued behavior	11, 15
			Irregular respiration	10, 14
			Vomiting	7–8, 14
		31	Subdued behavior	10–11
			Irregular respiration	10
			Vomiting	1, 8–12
		32	Subdued behavior	9, 11–12
			Vomiting	8, 11

^aThe animals that did not manifest clinical signs are not presented in this table, and the days on which clinical signs were not manifested are also omitted.

Safety pharmacology. (i) Central nervous and respiratory systems.

In the central nervous and respiratory system function tests, no adverse effects related to the administration of up to 25 mg/kg of SAL200 were observed compared with a vehicle control.

(ii) Cardiovascular system.

An in vitro hERG study was performed to assess the cardiovascular safety of the SAL-1 protein in SAL200. A cardiovascular function inhibition of only 9.8% was observed at 500 μg/ml, the maximum concentration tested. Inhibition of 3.9% was observed for the formulation buffer (vehicle) control, and thus it can be concluded that SAL200 causes negligible inhibition of cardiovascular function.

Cardiovascular function was further assessed by telemetric monitoring in dogs in a Latin square design with 1-week washout intervals. No adverse cardiovascular effects were observed in any of the tested animals after the first or second administrations. However, after the third administration, which was performed 2 weeks after the initial administration, transient changes in cardiovascular function were observed in one dog that was administered 25 mg/kg. These adverse changes included decreases in diastolic, systolic, and mean blood pressure and a PR interval reduction (24%). Following the fourth administration, adverse effects on cardiovascular functions were observed in 2 dogs that were administered 12.5 and 25 mg/kg. These adverse effects included decreased diastolic, systolic, and mean blood pressure, increased heart rate, PR interval reduction (4% and 13%, respectively), and RR interval reduction (23% and 25%, respectively). The effects were accompanied by clinical signs such as subdued behavior and vomiting, similar to the clinical signs observed in the dog repeated-dose toxicity tests, but these signs resolved within 6 h after injection.

Analysis of anti-SAL-1 antibody and C3 complement levels in blood.

We analyzed anti-SAL-1 antibody levels and C3 complement levels in blood samples to determine if the SAL200-dependent adverse effects observed in this study were due to an immune response to SAL-1. Anti-SAL-1 antibodies were not detected in the blood samples collected on day 14 but were detected in day 28 samples in the rat repeated-dose study. The levels of anti-SAL-1 antibodies in the blood obtained from rats on the final day (day 42) of the recovery period were slightly greater. In the dog repeated-dose study, anti-SAL-1 antibodies were detected beginning on day 14 in all treated groups, and the amount of antibody increased slightly during the recovery period. Although the levels of anti-SAL-1 antibodies varied greatly among animals of the same dose group in both rats and dogs, these results indicate that anti-SAL-1 antibody production was induced by the repeated administration of SAL200 for longer than 1 week.

We measured C3 complement levels in dog blood because SAL200-dependent adverse effects were observed in the dog experiments. C3 complement is a representative member of the complement system, and complement system activation may affect the level of C3 complement in the blood. As expected, changes in C3 complement levels were observed in the dog blood samples (Fig. 1). Specifically, C3 complement levels in the blood were significantly decreased compared to predose levels on day 14 in all test groups except the vehicle control group. The C3 complement levels in the blood samples obtained during the recovery period were similar to predose levels and those of the vehicle control group.

FIG 1.

Blood C3 complement levels. G1, 0 mg/kg (vehicle control group); G2, 2.5 mg/kg; G3, 10 mg/kg; G4, 25 mg/kg.

DISCUSSION

Phage endolysins have recently been explored as promising antibacterial agents. However, although some reports on the safety of phage endolysins have been published (13, 17), no GLP-compliant safety evaluation, including general toxicology and safety pharmacology tests, has been reported. Our study is the first GLP-compliant case report describing the general toxicology and safety pharmacology of a phage endolysin.

Intravenously administered SAL200 showed no signs of toxicity in the rodent single-dose and repeated-dose toxicity studies or in the central nervous system and respiratory function tests. Some abnormal findings and adverse effects were observed when SAL200 injections were continued for more than 1 week after initial administration in dogs. These abnormal findings and adverse effects were slight to mild, were apparent only transiently after injection, and resolved quickly.

The transient SAL200-dependent abnormal findings and adverse effects were likely due to an immune response to SAL-1 that includes the activation of the complement system by anti-SAL-1 antibody production following the formation of SAL-1 and anti-SAL-1 antibody complexes. This model is supported by the following observations: SAL200 is a protein-based candidate drug, the abnormal findings and adverse effects were observed only when SAL200 injection was performed more than 1 week after the initial administration, anti-SAL-1 antibodies were detected more than 1 week after the initial administration, and the decrease in C3 complement levels in the blood was largely correlated with the appearance of clinical signs.

The decrease in C3 complement levels in the blood indicates complement system activation, because this decrease was correlated with the generation of small complement fragments (complement split products), such as C3a (23, 24). These small complement fragments may have caused the SAL200-dependent abnormal findings and adverse effects observed in our experiments. The circulating immune complexes may be deposited in small vessels, causing complement system activation and subsequent inflammation (25–27). This type of hypersensitivity is called type III immune complex hypersensitivity and is also referred to as “serum sickness.” The SAL200-dependent abnormal findings and adverse effects observed in our experiments are not sufficiently explained by serum sickness, because there was no evidence of an inflammatory response in the blood vessels of the animals used in our experiments.

The SAL200-dependent abnormal findings and adverse effects were observed only when SAL200 injection was performed more than 1 week after initial administration. The effects were slight to mild and were only transiently apparent after injection; thus, the SAL200 safety evaluation results support the initiation of an exploratory phase I clinical trial and suggest that an appropriate study design would be able to minimize the adverse effects induced by the immune response to SAL-1. We plan to circumvent these side effects by using a shorter dosing period (<1 week) and by using a considerably lower dosage in the clinical trial than what was used in this safety evaluation. In fact, we assume that the clinically effective dose is likely much lower than that used in the toxicity tests. In addition, intravenous infusion will be used instead of a bolus injection; this change might reduce the degree of immune complex formation and/or rapid complement system activation by decreasing the immune complex load.