Evaluation of Antibacterial Drugs Using Silkworms Infected by Cutibacterium acnes

Simple Summary Cutibacterium acnes, a common bacterium on human skin, is a causative agent of inflammatory skin diseases and systemic infections. Systemic infections caused by C. acnes are difficult to treat because of the drug resistance to typical macrolides. The development of a systemic infection model for C. acnes contributes to searching candidates for drug discovery. Silkworm is a useful model animal for evaluating the efficacy of compounds. Several silkworm infection models have been established for the identification of virulence-related genes and novel antibacterial drugs. However, the systemic infection model of C. acnes using silkworms was not yet established. We established a new silkworm infection model with C. acnes and evaluated the efficacy of antibacterial drugs using the silkworm infection model. The silkworm infection model might be used to identify drug candidates for the treatment against C. acnes infection. Abstract Cutibacterium acnes is a causative agent of inflammatory skin diseases and systemic infections. Systemic infections caused by C. acnes are difficult to treat, and the development of a systemic infection model for C. acnes would be useful for elucidating the mechanisms of infection and searching for therapeutic agents. In this study, we established a silkworm infection model as a new experimental system to evaluate the interaction between C. acnes and the host, and the efficacy of antibacterial drugs. Silkworms infected with C. acnes died when reared at 37 °C. The dose of injected bacterial cells required to kill half of the silkworms (LD50) was determined under rearing conditions at 37 °C. The viable cell number of C. acnes was increased in the hemolymph and fat body of the infected silkworms. Silkworms injected with autoclaved C. acnes cells did not die during the study period. The survival time of silkworms injected with C. acnes was prolonged by the injection of antibacterial drugs such as tetracycline and clindamycin. These findings suggest that the silkworm C. acnes infection model can be used to evaluate host toxicity caused by C. acnes and the in vivo efficacy of antimicrobial drugs.


Introduction
Cutibacterium acnes (formerly Propionibacterium acnes), a common bacterium on human skin, causes inflammatory skin diseases and systemic infections [1,2]. C. acnes is isolated as the predominant species in 34% [3] or 36.2% [4] of intervertebral discs removed from patients with chronic low back pain, such as disc herniation. Biofilm formation by C. acnes on implants and intervertebral discs causes bloodstream infections [5][6][7]. Because C. acnes forms a biofilm and more than 50% of clinically isolated C. acnes are resistant to typical macrolides, systemic infections caused by C. acnes are difficult to treat [8,9]. Therefore, the development of treatments for systemic infections caused by C. acnes is clinically important. Although mammalian animal models of C. acnes infection have been established, their use for the evaluation of antibacterial drugs is difficult due to the long duration of the infection [10,11]. Infection experiments using a large number of mammalian animals are also difficult to perform due to animal welfare issues [12].
Silkworms are useful animals for assessing host-pathogen interactions in systemic infections and for evaluating the therapeutic effects of antimicrobial drugs [12][13][14]. Because silkworms also have advantageous features, such as easy rearing in large numbers in a small space with few ethical issues, experiments with large numbers of silkworms can be performed [15]. Moreover, quantitative drug administration and monitoring of parameters in silkworm blood can be performed due to the ease of sample injection and blood collection [12,16,17]. The use of a silkworm infection model based on these advantageous features led to the discovery of virulence-related genes for pathogenic microorganisms such as Staphylococcus aureus, Candida albicans, and Candida glabrata [18][19][20][21]. Further, exploratory studies of antimicrobial drugs using a silkworm infection model led to the identification of compounds exhibiting therapeutic efficacy in mouse infection experiments [22][23][24]. Therefore, the silkworm infection model is useful for studies aimed at elucidating the infection mechanisms of pathogenic microorganisms and evaluating the efficacy of antimicrobial drugs.
In the present study, we attempted to establish an animal model of systemic infection by C. acnes using silkworms. We found that injection of C. acnes cells killed silkworms. Survival times of the infected silkworms were prolonged by injection of tetracycline and clindamycin. Our findings suggest that the silkworm infection model with C. acnes is useful for evaluating the efficacy of antimicrobial drugs.

Reagents
Tetracycline was purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). Clindamycin was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). These reagents were dissolved in physiologic saline solution (0.9% NaCl). GAM agar was purchased from Nissui Pharmaceutical Co., Ltd. (Tokyo, Japan).

Culture of C. acnes
C. acnes ATCC6919 strain was used in this study. The C. acnes ATCC6919 strain was grown on GAM agar plates at 37 • C for 3 days under anaerobic conditions. Physiologic saline (2 mL) was added to the agar plate and suspended using a cell spreader. The C. acnes cell suspension was collected from the agar plate and measured absorbance at 600 nm using a spectrophotometer (U-1100; Hitachi Ltd., Tokyo, Japan). The C. acnes cells were suspended in physiologic saline to 5-10 of absorbance at 600 nm and the cell suspension was used in the experiments. For calculating the cell number in the cell suspension, the suspension of C. acnes cells was 10 7 -fold diluted with saline, and a 100 µL aliquot was spread on a GAM agar plate. After incubation at 37 • C for 3 days under anaerobic conditions, the number of colonies was counted for determining the colony-forming unit (CFU).

Silkworm Infection Experiments
The silkworm infection experiments were performed as previously described [25]. Silkworm fifth instar larvae were fed an artificial diet (1.5 g; Silkmate 2S; Ehime-Sanshu Co., Ltd., Ehime, Japan) overnight. A suspension (50 µL) of the C. acnes cells was injected into the silkworm hemolymph with a 1 mL tuberculin syringe (Terumo Medical Corporation, Tokyo, Japan). Silkworms injected with the C. acnes cells were placed in an incubator and survival was monitored.

Measurement of Viable Cell Number of C. acnes in Silkworms
Silkworms were injected with C. acnes cell suspension (3.3 × 10 8 cells/50 µL) and incubated at 37 • C. Hemolymph and fat body were harvested from the silkworms at 1 or 24 h after injection. Hemolymph was collected from the larva through a cut on the first proleg according to the previous report [16]. For calculating the C. acnes cell number in the hemolymph, the hemolymph was 10 3 -or 10 4 -fold diluted with saline, and a 100 µL aliquot was spread on GAM agar plates. The number of colonies was counted after incubation at 37 • C for 3 days under anaerobic conditions. The viable cell number of C. acnes in the hemolymph was calculated as the colony-forming unit (CFU) per 1 mL.
Fat body isolation was performed as previously described [26]. Silkworms were placed on ice for 15 min. The fat body was isolated from the dorsolateral region of each silkworm and rinsed in saline. The wet weight of the fat body was measured. The fat body was homogenized in 1 mL of saline, and the lysate was obtained. The lysate was 10 2 or 10 3 -diluted with saline, and a 100 µL aliquot was spread on GAM agar plates. The number of colonies was counted after incubation at 37 • C for 3 days under anaerobic conditions. The viable cell number of C. acnes per wet weight of the fat body was determined.

Evaluation of Therapeutic Activities of Antibacterial Drugs Using Silkworms
Therapeutic activity tests using silkworms were performed according to a previous report with slight modification [25]. Saline (50 µL) or C. acnes suspension (3.6 × 10 8 cells/50 µL) was injected into the hemolymph of silkworms. Immediately after inoculation with the C. acnes suspension, 50 µL of an antibacterial drug (50 µg/g larva or 10 µg/g larva) was injected into the hemolymph of the silkworms.

Statistical Analysis
The significance of differences between groups on survival time of silkworms was calculated using a log-rank test based on the Kaplan-Meier method using Prism 9 (GraphPad Software, LLC, San Diego, CA, USA, https://www.graphpad.com/scientific-software/ prism/, accessed on 25 April 2021). The statistically significant differences between groups on viable cell number of C. acnes in silkworm body were evaluated using the Student t-test. p < 0.05 was considered a statistically significant difference. Raw data are shown in Table S1.

Pathogenicity of C. acnes against Silkworms
Silkworm models of infection by various microorganisms have been established [15]. The body temperature of silkworms, which can be regulated by changing the rearing temperature, is important for bacterial pathogenicity against silkworms [15]. In the case of Aspergillus fumigatus and dermatophytes, silkworms do not die at 37 • C by depending on their infection [24,28]. Therefore, it is important to consider the temperature at which the experiment is performed for each pathogen. C. acnes grows at approximately 32 • C on the skin and 37 • C in the human body. We examined the rearing temperatures that caused death in C. acnes-infected silkworms. Silkworms injected with C. acnes (1.6 × 10 9 cells) reared at 37 • C after injection died within 40 h, whereas infected silkworms reared at 32 • C survived longer (Figure 1). The time required for half of the infected silkworms to die (LT 50 ) was 48 h for silkworms reared at 32 • C and 27 h for those reared at 37 • C (Figure 1). The LD 50 value, which is the bacterial number required to kill half of the silkworms, was 1-3 × 10 8 cells when infected silkworms were reared at 37 • C ( Figure 2). These results suggest that rearing C. acnes-infected silkworms at 37 • C decreased survival and that the pathogenicity of C. acnes can be quantitatively assessed based on the LD 50 value.

Increase in Viable Cell Number of C. acnes in Silkworms
We next investigated whether C. acnes cells can grow in the silkworm body. The scheme of the experiment is shown in Figure 3a. The viable cell number of C. acnes was increased in hemolymph, the blood of silkworm (Figure 3b). Silkworm has a fat body, which is a soft organ that has a similar function to the liver and adipose tissue in mammals. The viable cell number of C. acnes was also increased in the fat body of silkworms (Figure 3c). The result suggests that C. acnes can proliferate in the silkworm body. Silkworms were injected with saline or C. acnes cell suspension (3.3 × 10 8 cells/50 µL) and incubated at 37 • C. Hemolymph (b) and fat body (c) were harvested from the silkworms at 1 or 24 h after injection. Viable cell number of C. acnes in the samples was measured by counting the colony-forming unit (CFU). Statistically significant differences between groups were evaluated using Student t-test. n = 5 / group.

Effect of Heat-Killed C. acnes Cells on Silkworms
Bacterial components such as peptidoglycans of Porphyromonas gingivalis lead to shock in silkworms, resulting in their death [29]. Injection of viable or heat-killed P. gingivalis cells causes silkworm death [29]. Under conditions in which shock occurs, silkworms cannot be treated with antibiotics [29]. Therefore, it is important to determine whether the death of silkworms by injection of a pathogen is shock by the bacterial components. We next examined the heat-killed cells to evaluate whether C. acnes cells must be alive to exert pathogenicity against silkworms. Injection of C. acnes cells killed silkworms, but injection of autoclaved C. acnes cells did not (Figure 4). These results suggest that the heat-stable components of C. acnes do not cause acute silkworm death in the experimental condition.

Therapeutic Effects of Antibacterial Drugs against Silkworms Infected with C. acnes
Tetracycline and clindamycin are used for the treatment of dermatologic disorders including acne vulgaris and are often administrated orally [30][31][32]. We next examined the efficacy of these antibacterial drugs in the silkworm C. acnes infection model. In the clinic, tetracycline and clindamycin are administered 250-500 mg/person/day and 600 mg/person/day, respectively [30][31][32]. Based on the information, we determined injection amounts of the antibacterial drugs in the experiments. Administration of tetracycline and clindamycin to silkworms infected with C. acnes prolonged the survival time ( Figure 5). These results suggest that the efficacy of antibacterial drugs can be evaluated using the silkworm infection model with C. acnes. Figure 5. Effect of antibacterial drugs in silkworms infected by C. acnes. Silkworms were injected with saline (50 µL) or C. acnes cell suspension (3.6 × 10 8 cells/50 µL), and then with tetracycline (50 µg/g larva) (a), tetracycline (10 µg/g larva) (b), clindamycin (50 µg/g larva) (c), or clindamycin (10 µg/g larva) (d). The number of surviving silkworms under incubation at 37 • C was measured for 48 h. Statistically significant differences between groups were evaluated using a log-rank test. n = 8/group. (e) Silkworms were injected with saline (50 µL), and then with tetracycline (50 µg/g larva), tetracycline (10 µg/g larva), clindamycin (50 µg/g larva), or clindamycin (10 µg/g larva). The number of surviving silkworms under incubation at 37 • C was measured for 54 h. n = 6/group.

Discussion
In this study, we demonstrated that C. acnes kills silkworms reared at 37 • C and that the silkworm infection model can be used to evaluate the efficacy of antibacterial drugs. Our findings suggest that the silkworm infection model is useful for assessing pathogenicity and the efficacy of antimicrobial drugs against systemic infection by C. acnes.
Because C. acnes causes bloodstream infections, we selected injection the C. acnes into the hemolymph of the silkworms. C. acnes-infected silkworms under the rearing condition at 37 • C were more sensitive than that at 32 • C. We assumed that this is due to two factors: the effect of high-temperature stress on the silkworm and the optimal temperature at which the pathogens have high virulence. The silkworms injected with the autoclaved C. acnes cells seem to be unhealthy compared with those in the saline group (Figure 4b). We assumed the reason is that the hemolymph of silkworms was melanized and the melanization might affect the silkworm condition. The C. acnes cells grow in the hemolymph and on the tissues of the silkworms. C. acnes use the nutrition of the silkworms and might cause toxicity to the tissues. The melanization of the hemolymph of silkworms caused by C. acnes might affect tissue toxicity. We assumed that the silkworms infected with C. acnes cause sepsis by the proliferation of C. acnes cells in the hemolymph and on tissues of the silkworms.
C. acnes ATCC6919 used in this study is a type strain whose genome is publicly available and can invade osteoclasts and osteoblasts [33]. Therefore, we used the C. acnes ATCC6919 for systemic infection experiments using silkworms. The viable cell number of C. acnes was increased in the hemolymph and fat body of silkworms. The results suggest that C. acnes causes bacteremia by growth in the silkworm body. C. acnes forms a biofilm on the intervertebral discs [5]. We assumed that C. acnes might form a biofilm on the fat body of silkworms. Future work will be to evaluate the difference in pathogenicity and drugresistant in clinical isolates from patients with disc herniation. Moreover, identification of high pathogenic C. acnes clinical strain using the silkworm infection model and genetic study of the high pathogenic strain and drug-resistant strain will be an important subject.
C. acnes causes superficial inflammatory diseases and systemic infections. In this study, we focused on the systemic infections caused by C. acnes and established the systemic infection model using silkworms. We assumed that the silkworm C. acnes infection model established in this study is not a superficial inflammatory disease model. The establishment of a superficial inflammatory disease model using silkworms is a future subject for the evaluation of topical drugs.
In a previous study using nucleus pulposus-derived disc cells in vitro, C. acnes cells (10 7 -10 8 cells/mL) were co-cultured with the disc cells and proinflammatory responses of the disc cells were induced by the addition of the C. acnes cells [34]. We injected C. acnes cells (3.3 × 10 8 cells/larva), and the range of viable cell number of C. acnes in silkworm hemolymph at 1 h after the injection was 3.2-4.8 × 10 7 cells/mL ( Figure 3). Moreover, the LD 50 was 1-3 × 10 8 cells when infected silkworms were reared at 37 • C ( Figure 2). Therefore, we used C. acnes cells (10 8 -10 9 cells/larva) in this study, because the amounts are needed for silkworm death and the similar range of C. acnes cells in previous in vitro study.
In the silkworm C. acnes infection model established in this study, administration of heat-killed bacteria did not kill the silkworm. Further, C. acnes-infected silkworms could be effectively treated with antibacterial drugs, suggesting that the growth of C. acnes in the body of silkworms is important for its pathogenicity. In the clinic, tetracycline and clindamycin are administered 250-500 mg/person/day and 600 mg/person/day, respectively [30][31][32]. Given that the average weight of humans is 60 kg, the calculated values of tetracycline and clindamycin are 4.2-8.3 and 10 mg/kg, respectively, for humans. The bodyweight of the silkworm used in this study was approximately 2 g. One-hundred micrograms (50 µg/g larva) and 20 µg (10 µg/g larva) of tetracycline and clindamycin were administered to each silkworm ( Figure 5). Therefore, doses of tetracycline and clindamycin were 50 mg/kg or 10 mg/kg. The 10 mg/kg dose is a similar dose to human clinical use. The results support that the silkworm infection model is useful for the eval-uation of antibacterial drugs against systemic infection by C. acnes. Silkworm infection models are useful for identifying virulence factors of pathogenic microorganisms [13,15]. Further studies are needed to determine which factors in C. acnes are responsible for the pathogenicity against silkworms.
The pharmacokinetics of antimicrobial agents are similar between silkworms and mammals, and antimicrobial drug efficacy in the silkworm infection model can be evaluated based on the pharmacokinetics [23,[35][36][37]. A previous study demonstrated that total body clearance (CL tot ), distribution volume in steady state (Vd ss ), and the elimination half-life (t 1/2 ) of antibiotics in silkworm were calculated [37]. For example, CL tot , Vd ss , and t 1/2 of tetracycline in silkworm were 0.07 mL/g/h, 0.6 mL/g, and 5.5h, respectively. In rabbit, these values of tetracycline were 0.35 mL/g/h, 1.0 mL/g, and 2.0 h, respectively. The pharmacokinetic parameters of tetracycline in silkworms and rabbits are not different at least 10-fold [37]. Moreover, silkworm infection models are useful to develop for in vivo screening of new antimicrobial agents [22,24]. The silkworm C. acnes infection model may be useful for identifying effective antibacterial compounds against systemic infection by C. acnes. Azole antifungals exhibit antimicrobial activity against C. acnes in vitro, and ketoconazole inhibits the lipase activity of C. acnes [38,39]. Further studies are needed to identify effective compounds against systemic C. acnes infections from among clinically applied drugs using the silkworm infection model.
Recently, an infection model with C. acnes using a nematode, Caenorhabditis elegans, was reported [40]. C. elegans is useful for identifying host factors against C. acnes infection based on genetic approaches [40]. The differences between the silkworm system and the C. elegans system are that the silkworm blood can be directly injected with C. acnes and its pathogenicity at 37 • C, the same temperature as the human body, can be verified. C. elegans is difficult to inject quantitatively into body fluids and cannot grow at 37 • C [41]. The silkworm infection model might allow us to identify virulence factors of C. acnes at the body temperature of humans.

Conclusions
We established a silkworm infection model with C. acnes and found the system to be useful for evaluating antibacterial drug efficacy. Further studies are needed to determine the clinical applicability of research using the silkworm C. acnes infection model.