A silkworm infection model to investigate Vibrio vulnificus virulence genes

  • Authors:
    • Mai Yamamoto
    • Takashige Kashimoto
    • Yukihiro Yoshimura
    • Nao Tachibana
    • Shiho Kuroda
    • Yoshiko Miki
    • Sou Kitabayashi
    • Ping Tong
    • Jianbo Xiao
    • Koichi Tanaka
    • Hiroshi Hamamoto
    • Kazuhisa Sekimizu
    • Koichiro Yamamoto
  • View Affiliations

  • Published online on: September 26, 2016     https://doi.org/10.3892/mmr.2016.5782
  • Pages: 4243-4247
Metrics: Total Views: 0 (Spandidos Publications: | PMC Statistics: )
Total PDF Downloads: 0 (Spandidos Publications: | PMC Statistics: )


Abstract

The halophilic marine bacterium, Vibrio vulnificus, occasionally causes fatal septicemia in immunocompromised patients. Mice are commonly used as experimental animals to investigate the virulence of V. vulnificus, however, a large number of mice are generally required for bioassays. The present study examined whether the invertebrate species, silkworms, can be used instead of mice to investigate V. vulnificus virulence. When the silkworms were inoculated with 1.2x107 colony forming units of V. vulnificus OPU1‑Rf, a virulent strain of V. vulnificus, all injected silkworms died within 48 h, however, those injected with culture filtrate or diluent did not. This silkworm infection model was then used to isolate attenuated V. vulnificus mutants from 1,016 transposon‑inserted mutants. Consequently, a harmless mutant, SW998, was isolated. In this strain, the transposon was inserted into the rtxA gene, which is a known V. vulnificus virulence gene. In conclusion, the present study demonstrated that silkworms are useful animals for investigating the virulence of V. vulnificus.

Introduction

Vibrio vulnificus is a Gram-negative, comma-shaped bacterium found in warm seawater and brackish waters. V. vulnificus occasionally causes primary septicemia, gastroenteritis and wound infections in humans. Primary septicemia and gastroenteritis can occur following ingestion of uncooked seafood contaminated with V. vulnificus (1), whereas wound infections are often caused by direct contact of an open wound with seawater during marine activities or in the processing of seafood (2). V. vulnificus rarely infects healthy individuals, however, immunocompromised patients with an underlying illness, including hepatic disorder, diabetes or immunodeficiency, are susceptible to infection by this bacterium (3). Among the aforementioned clinical states of V. vulnificus infection, primary septicemia is the most life-threatening, with a mean mortality rate of >50% worldwide (4). There is no official national system for surveying V. vulnificus infections in Japan, however, it was reported that 117 individuals succumbed to mortality between 1975 and 2005 (5), and the annual number of cases of V. vulnificus-septicemia has been estimated to be >200 (6). Cases of V. vulnificus infection have also been reported annually in other countries. The Centers for Disease Control (Atlanta, GA, USA) reported 32 cases of V. vulnificus infection-associated mortality in the US in 2012 (7). In Europe, V. vulnificus is a known fish pathogen, particularly in eels, and V. vulnificus infection in humans is relatively rare (8). However, due to the global warming, there is concern of an increasing number of cases of V. vulnificus infection as this bacterium grows rapidly in warm seawater (9).

Several studies have reported on V. vulnificus virulence genes (1012), however, the mechanism underlying the infection remains to be fully elucidated. Mice have been used as experimental animals to investigate the host-pathogen interactions of V. vulnificus infections, however, using a large number of mammals is expensive and raises ethical considerations (13,14).

Previously, invertebrate infection models using nematodes, including Caenorhabditis elegans (15), and insects, including Galleria mellonella (16), have being applied to investigate bacterial virulence genes (17). In addition, an infection model using silkworms (Bombyx mori) for the investigation of bacterial pathogenesis has been reported (18). Silkworms are relatively easy to breed in laboratories, and the animal is sufficient large enough for the injection of samples into the hemolymph. The objective of the present study was to use a silkworm model to investigate the virulence of V. vulnificus. This model may be aid in the investigation of V. vulnificus virulence factors and may assist in the development of effective therapies to protect against V. vulnificus infection

Materials and methods

Bacterial strains, plasmids and medium

The V. vulnificus OPU1 strain was clinically isolated, and its rifampicin (Rf)-resistant variant, V. vulnificus OPU1-Rf, was used in the present study. Escherichia coli BW19795 was provided by Dr Barry L. Wanner (Purdue University, West Lafayette, IN, USA (19), which was used as a pUT donor for conjugation. E. coli DH10BTM-competent cells were purchased from Life Technologies; Thermo Fisher Scientific, Inc. (Waltham, MA, USA). The signature-tagged mini-Tn5Km2-y67 transposon in the pUT delivery suicide plasmid pool was provided by Dr David W. Holden (Imperial College London, London, UK) (20). Bacterial cells were grown at 37°C in Luria-Bertani (LB) medium containing 10 g tryptone (BD Biosciences, Tokyo, Japan), 5 g yeast extract (BD Biosciences) and 10 g NaCl/l (21), unless otherwise described. M9 minimal medium without glucose was prepared as previously described (21) and used for bacterial conjugation. The following antibiotics were added to the medium at the indicated concentrations: Rf (100 µg/ml; Wako Pure Chemical Industries, Ltd., Osaka, Japan), kanamycin (Km; 50 µg/ml; Meiji Seika Pharma Co., Ltd., Tokyo, Japan) and ampicillin (Am; 100 µg/ml; Meiji Seika Pharma Co., Ltd.). Bacterial growth was monitored by turbidity using a spectrophotometer (Spectronic 20A; Shimadzu Corporation, Kyoto, Japan).

Silkworm lethality assay

The bacteria were inoculated into silkworms using the protocol described by Hamamoto et al (22). The silkworms were raised from fertilized silkworm eggs (Hu·Yo × Tsukuba·Ne) purchased from Ehime Sansyu Co., Ltd. (Yawatahama, Japan). The eggs were incubated in a clean bench (CCV-800E-AG; Hitachi Koki Co., Ltd., Tokyo, Japan) at 25°C in the dark for 3–5 days, according to the manufacturer's protocol. The hatched larvae were maintained at 27°C, humidity 60–90% and fed artificial food (Silkmate 2S; Nosan Corporation, Yokohama, Japan) for ~3 weeks. The larvae shed their shells four times, and the fifth-instar larvae were fed antibiotic-free artificial food (Silkmate; Katakura Industries, Tokyo, Japan) for 24–26 h prior to inoculation.

The bacteria were cultivated at 37°C in 2 ml of LB medium until the optical density at 600 nm (OD600) reached 1.0. The bacterial cultures were diluted with 10 mM phosphate-buffered saline (pH 7.3), containing 0.01% gelatin (PBSG) to cell densities of 2.4×107 and 2.4×108 cfu/ml, and were maintained for up to 2 h at room temperature until they were inoculated into the silkworms. The culture filtrates were prepared by filtering the diluted cultures through a 0.45-µm filter (Merck Millipore, Tokyo, Japan). A 50-µl aliquot of the diluted bacterial cell suspension or culture filtrate was injected into the hemolymph of the silkworms using a 1-ml plastic disposable tuberculin syringe attached to a 27-gauge needle (Terumo Corporation, Tokyo, Japan). The quantities of bacteria injected were estimated by colony forming units (cfu), determined by inoculation of the diluted bacterial culture suspensions onto LB agar and counting the number of colonies grown following incubation for 18 h at 37°C. The inoculated silkworms were maintained in plastic containers without feeding, and the larval status (dead or alive) was assessed. The silkworms were considered to be dead when they showed no reaction to touch.

Construction of the transposon insertion mutants by conjugation

The transposon insertion mutants were constructed by conjugation, as described previously (23). Briefly, V. vulnificus OPU1-Rf was mated with E. coli BW19795 harboring the pUTy69 conjugative suicide plasmid, which contained the mini-Tn5Km2-y67 Km-resistant transposon. The transposon-inserted mutants were assessed by their growth on Km- and Rf-containing agar medium.

Cloning and sequence analysis of the attenuated mutant

The locations of the transposon-inserted regions in the mutant genome were determined by sequencing of the DNA sequence adjacent to the insertion site, as described previously (23). Briefly, whole genome DNA of attenuated mutants were extracted and digested with the restriction enzyme SalI (Nippon Gene Co., Ltd, Tokyo, Japan). The enzyme digested DNA fragments were ligated into pUC18, followed by transformation into E. coli DH10B (Thermo Fisher Scientific, Inc.). The colonies resistant to both Km and Am were selected and the plasmid DNA that ligated with the fragments containing transposon insertion sites of genomic DNA was extracted. The transposon insertion sites were determined by DNA sequencing using the primer P279 (5′-CTAGGTACCTACAACCTC-3′) which anneal to the transposon. DNA sequencing was performed with an Applied biosystems DNA sequencing system and the BigDye terminator cycle sequencing kit (Thermo Fisher Scientific, Inc.). Sequence homologies were searched with the BLAST search algorithm at the National Center for Biotechnology Information (www.ncbi.nlm.nih.gov/)

Results

Lethality of V. vulnificus towards silkworms

To investigate the applicability of silkworms for in vivo V. vulnificus infection experiments, the present study first examined the lethality of V. vulnificus towards silkworms. The spontaneous Rf-resistant V. vulnificus OPU1-Rf was selected for injection into silkworms, as the infectivity and lethality of this strain have been confirmed in mice, and a mutagenesis method using the mini-Tn5Km2 transposon was established (23).

Fresh V. vulnificus OPU1-Rf cultures with an OD600 of 1.0 were diluted with PBSG solution to cell densities of 2.4×107 and 2.4×108 cfu/ml, and were maintained for up to 2 h at room temperature until they were inoculated into the silkworms. A 50-µl aliquot of the diluted culture specimen was injected into the hemolymph of the silkworms from a syringe through the dorsal surface, in a manner similar to that of infection into the human blood stream. The movement of the silkworms reduced immediately following the injection, however, the animals behaved normally at ~1 h. This blunting of silkworm behavior was also observed when the animals were injected with the diluent (PBSG) alone. These findings may have been associated with the injection stimulus, an example being a decrease in body temperature caused by the injected solution. As shown in Fig. 1, the silkworms injected with V. vulnificus OPU1-Rf began to die at 24 h. By contrast, the silkworms injected with PBSG or culture filtrate remained alive at 72 h. In the deceased silkworms, the points of injection turned black in the initial hours, and these black spots gradually spread throughout the entire body. All silkworms injected with 1.2×107 cfu/silkworm died within 48 h. A reduced dose of 1.2×106 cfu/silkworm extended survival rates, and 60% of the worms were alive at 72 h (Fig. 1). Thus, the virulence of V. vulnificus towards silkworms may be dose-dependent at these doses, although only two doses were examined in the present study. These results indicated that V. vulnificus is life-threatening to silkworms, and that silkworms can be used for investigating the virulence of V. vulnificus. As the injection of 1.2×107 cfu/silkworm led to the death of all silkworms within 48 h, 107 cfu/silkworm was used in the subsequent experiments.

Screening of the V. vulnificus-attenuated mutants through the assessment of silkworm lethality

As the silkworms were found to be sensitive to V. vulnificus infection, the present study attempted to use a silkworm infection model to identify the pathogenic genes of V. vulnificus against silkworms. The attenuated mutants, which had lost pathogenicity to silkworms, were screened from the 1,016 transposon insertion mutants.

The transposon insertion mutants were cultured for 4–6 h at 37°C with agitation. Culture fluid with an OD600 of 1.0 was diluted with PBSG, and a 50-µl aliquot was injected into the hemolymph of silkworms. Subsequently, ~3.3×107±2.9×107 cfu of the organism were injected into each silkworm, and silkworm status (dead or alive) was monitored for 5 days. In the process of identifying less virulent mutants, which did not lead to silkworm death within 5 days, 78 candidates were obtained (primary screening). To reduce experimental error, a second inoculation of the first candidate mutant was performed. Of the 78 candidates, 16 mutants did not lead to silkworm death within 5 days of injection (secondary screening). To confirm the avirulent properties of the secondary screened candidates, each of the 16 secondary screened candidates was injected into five silkworms. The candidates, in which all five injected silkworms survived for 5 days, were selected as attenuated mutants. Of the 16 mutants, 15 were not selected as an attenuated mutant as the silkworms died during the observation period. The SW998 mutant was found to be avirulent to all five silkworms (Fig. 2).

Transposon insertion sites in the attenuated mutant

As SW998 was avirulent in the silkworms (Fig. 2), the virulence-associated gene was expected to have been disrupted by the transposon insertion in SW998. In order to locate the transposon-inserted gene in the SW998 genome, recombinant plasmid DNA which contained the transposon insertion site was prepared as aforementioned and the DNA sequences adjacent to the transposon insertion sites were determined (Fig. 3). The nucleotide sequences of the cloned V. vulnificus DNA fragment adjacent to the transposon insertion site showed a high level of homology to the V. vulnificus rtxA gene, which has been reported as a V. vulnificus cytotoxic factor (24).

Discussion

Certain in vivo experiments designed to investigate the precise mechanisms of microbial infections are difficult to replace using in vitro experiments or models. Mice are often used to investigate V. vulnificus infection; however, using a larger number of animals may not always be economically or ethically feasible. Random transposon insertion mutagenesis and subsequent screening of attenuated mutants have been applied to identify virulence genes in several bacterial pathogens (25). Thousands of mice are required to isolate attenuated mutants. Animal experiments using mammals are required; however, to obtain satisfactory results, the sacrifice of a large number of animals may be required. In vivo infection models using invertebrates instead of mammals have been successfully applied in pathogenic microbe investigations, including Staphylococcus aureus virulence genes identified using a silkworm infection model (18). The purpose of the present study was to examine whether the silkworm infection model can be applied to investigate the virulence of V. vulnificus.

In the present study, silkworms were demonstrated to be sensitive to the V. vulnificus; inoculation of this bacterium into the hemolymph of animals led to death (Fig. 1). Kaito et al (26) reported that several bacterial species, including S. aureus, Streptococcus pyogenes, Pseudomonas aeruginosa and V. cholerae, which are human pathogens, can also infect and induce death in silkworms when injected into the blood stream of silkworms, whereas non-pathogenic laboratory strains of E. coli cannot. Although there is no direct evidence to confirm whether the death of silkworms is caused by the same factors as mammals, silkworms may recognize virulence factors of pathogens as they have the Toll and immune deficiency pathways, which are homologous to the mammalian Toll-like receptor and tumor necrosis factor receptor signaling pathways, respectively (27). Using the silkworm infection model, the SW998 attenuated mutant strain was obtained in the present study, in which the rtxA gene was disrupted by a transposon insertion. The rtxA gene is a member of the rtx gene cluster in the V. vulnificus genome. RtxA has been suggested to be a toxin essential for V. vulnificus virulence, which has been confirmed by experimental infection of a mouse model and in an in vitro tissue culture model (24). In the present study, the injection of culture filtrate did not induce silkworm death (Fig. 1), which suggested that the expression of RtxA increased following interaction of the pathogen with host cells, as reported previously (28). Therefore, the bacteria secreted RtxA in the silkworm in vivo and RtxA may be toxic to silkworms. Complete removal of the genes from the genome and lethality assessment are required to determine whether RtxA is actually toxic to silkworms.

In conclusion, the silkworms examined in the present study died when inoculated experimentally with V. vulnificus and the rtxA gene was identified successfully using this infection model. It is expected this model may be useful for investigating V. vulnificus virulence factors and assist in developing effective therapies to protect against V. vulnificus infection.

Acknowledgements

The authors would like to thank Dr Toru Takahashi (Okayama Prefectural University, Okayama, Japan) for their advice during preparation of the manuscript. The present study was supported in part by JSPS Kakenhi (grant no. 22590400) and the Grant for Special Scientific Research of Okayama Prefectural University, 2012.

References

1 

de Araujo MR, Aquino C, Scaramal E, Ciola CS, Schettino G and Machado MC: Vibrio vulnificus infection in São Paulo, Brazil: Case report and literature review. Braz J Infect Dis. 11:302–305. 2007. View Article : Google Scholar : PubMed/NCBI

2 

Vinh DC, Mubareka S, Fatoye B, Plourde P and Orr P: Vibrio vulnificus septicemia after handling Tilapia species fish: A Canadian case report and review. Can J Infect Dis Med Microbiol. 17:129–132. 2006.PubMed/NCBI

3 

Horseman MA and Surani S: A comprehensive review of Vibrio vulnificus: An important cause of severe sepsis and skin and soft-tissue infection. Int J Infect Dis. 15:e157–e166. 2011. View Article : Google Scholar : PubMed/NCBI

4 

Hlady WG and Klontz KC: The epidemiology of Vibrio infections in Florida, 1981–1993. J Infect Dis. 173:1176–1183. 1996. View Article : Google Scholar : PubMed/NCBI

5 

Miyasaka J, Yahiro S, Arahira Y, Tokunaga H, Katsuki K and Hara-Kudo Y: Isolation of Vibrio parahaemolyticus and Vibrio vulnificus from wild aquatic birds in Japan. Epidemiol Infect. 134:780–785. 2006. View Article : Google Scholar : PubMed/NCBI

6 

Osaka K, Komatsuzaki M, Takahashi H, Sakano S and Okabe N: Vibrio vulnificus septicaemia in Japan: An estimated number of infections and physicians' knowledge of the syndrome. Epidemiol Infect. 132:993–996. 2004. View Article : Google Scholar : PubMed/NCBI

7 

National Enteric Disease Surveillance, . Centers for Disease Control and Prevention. Atlanta, Georgia, USA: COVIS Annual Summary; 2012, https://www.cdc.gov/ncezid/dfwed/pdfs/covis-annual-report-2012-508c.pdf

8 

Fouz B, Roig FJ and Amaro C: Phenotypic and genotypic characterization of a new fish-virulent Vibrio vulnificus serovar that lacks potential to infect humans. Microbiology. 153:1926–1934. 2007. View Article : Google Scholar : PubMed/NCBI

9 

Bier N, Bechlars S, Diescher S, Klein F, Hauk G, Duty O, Strauch E and Dieckmann R: Genotypic diversity and virulence characteristics of clinical and environmental Vibrio vulnificus isolates from the Baltic Sea region. Appl Environ Microbiol. 79:3570–3581. 2013. View Article : Google Scholar : PubMed/NCBI

10 

Kreger A..Lockwood D: Detection of extracellular toxin(s) produced by Vibrio vulnificus. Infect. Immun. 33:583–590. 1981.

11 

Miyoshi S, Oh EG, Hirata K and Shinoda S: Exocellular toxic factors produced by Vibrio vulnificus. J Toxicol Rev. 12:253–288. 1993.

12 

Powell JL, Wright AC, Wasserman SS, Hone DM and Morris JG Jr: Release of tumor necrosis factor alpha in response to Vibrio vulnificus capsular polysaccharide in vivo and in vitro models. Infect Immun. 65:3713–3718. 1997.PubMed/NCBI

13 

Jeong HG, Oh MH, Kim BS, Lee MY, Han HJ and Choi SH: The capability of catabolic utilization of N-acetylneuraminic acid, a sialic acid, is essential for Vibrio vulnificus pathogenesis. Infect Immun. 77:3209–3217. 2009. View Article : Google Scholar : PubMed/NCBI

14 

Hor LI, Chang YK, Chang CC, Lei HY and Ou JT: Mechanism of high susceptibility of iron-overloaded mouse to Vibrio vulnificus infection. Microbiol Immunol. 44:871–878. 2000. View Article : Google Scholar : PubMed/NCBI

15 

Bae T, Banger AK, Wallace A, Glass EM, Aslund F, Schneewind O and Missiakas DM: Staphylococcus aureus virulence genes identified by bursa aurealis mutagenesis and nematode killing. Proc Natl Acad Sci USA. 101:12312–12317. 2004. View Article : Google Scholar : PubMed/NCBI

16 

Seed KD and Dennis JJ: Development of Galleria mellonella as an alternative infection model for the Burkholderia cepacia complex. Infect Immun. 76:1267–1275. 2008. View Article : Google Scholar : PubMed/NCBI

17 

Mylonakis E and Aballay A: Worms and flies as genetically tractable animal models to study host-pathogen interactions. Infect Immun. 73:3833–3841. 2005. View Article : Google Scholar : PubMed/NCBI

18 

Kaito C, Kurokawa K, Matsumoto Y, Terao Y, Kawabata S, Hamada S and Sekimizu K: Silkworm pathogenic bacteria infection model for identification of novel virulence genes. Mol Microbiol. 56:934–944. 2005. View Article : Google Scholar : PubMed/NCBI

19 

Metcalf WW, Jiang W and Wanner BL: Use of the rep technique for allele replacement to construct new Escherichia coli hosts for maintenance of R6K gamma origin plasmids at different copy numbers. Gene. 138:1–7. 1994. View Article : Google Scholar : PubMed/NCBI

20 

Hensel M, Shea JE, Gleeson C, Jones MD, Dalton E and Holden DW: Simultaneous identification of bacterial virulence genes by negative selection. Science. 269:400–403. 1995. View Article : Google Scholar : PubMed/NCBI

21 

Sambrook J and Russell DW: Molecular Cloning: A laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor; New York: 3. Appendix 2.2. 2001

22 

Hamamoto H, Kurokawa K, Kaito C, Kamura K, Razanajatovo I Manitra, Kusuhara H, Santa T and Sekimizu K: Quantitative evaluation of the therapeutic effects of antibiotics using silkworms infected with human pathogenic microorganisms. Antimicrob Agents Chemother. 48:774–779. 2004. View Article : Google Scholar : PubMed/NCBI

23 

Yamamoto M, Kashimoto T, Tong P, Xiao J, Sugiyama M, Inoue M, Matsunaga R, Hosohara K, Nakata K, Yokota K, et al: Signature-tagged mutagenesis of Vibrio vulnificus. J Vet Med Sci. 77:823–828. 2015. View Article : Google Scholar : PubMed/NCBI

24 

Lee JH, Kim MW, Kim BS, Kim SM, Lee BC, Kim TS and Choi SH: Identification and characterization of the Vibrio vulnificus rtxA essential for cytotoxicity in vitro and virulence in mice. J Microbiol. 45:146–152. 2007.PubMed/NCBI

25 

Saenz HL and Dehio C: Signature-tagged mutagenesis: Technical advances in a negative selection method for virulence gene identification. Curr Opin Microbiol. 8:612–619. 2005. View Article : Google Scholar : PubMed/NCBI

26 

Kaito C, Akimitsu N, Watanabe H and Sekimizu K: Silkworm larvae as an animal model of bacterial infection pathogenic to humans. Microb Pathog. 32:183–190. 2002. View Article : Google Scholar : PubMed/NCBI

27 

Tanaka H and Yamakawa M: Regulation of the innate immune responses in the silkworm, Bombyx mori. ISJ. 8:59–69. 2011.

28 

Kim YR, Lee SE, Kook H, Yeom JA, Na HS, Kim SY, Chung SS, Choy HE and Rhee JH: Vibrio vulnificus RTX toxin kills host cells only after contact of the bacteria with host cells. Cell Microbiol. 10:848–862. 2008. View Article : Google Scholar : PubMed/NCBI

Related Articles

Journal Cover

November-2016
Volume 14 Issue 5

Print ISSN: 1791-2997
Online ISSN:1791-3004

Sign up for eToc alerts

Recommend to Library

Copy and paste a formatted citation
x
Spandidos Publications style
Yamamoto M, Kashimoto T, Yoshimura Y, Tachibana N, Kuroda S, Miki Y, Kitabayashi S, Tong P, Xiao J, Tanaka K, Tanaka K, et al: A silkworm infection model to investigate Vibrio vulnificus virulence genes. Mol Med Rep 14: 4243-4247, 2016
APA
Yamamoto, M., Kashimoto, T., Yoshimura, Y., Tachibana, N., Kuroda, S., Miki, Y. ... Yamamoto, K. (2016). A silkworm infection model to investigate Vibrio vulnificus virulence genes. Molecular Medicine Reports, 14, 4243-4247. https://doi.org/10.3892/mmr.2016.5782
MLA
Yamamoto, M., Kashimoto, T., Yoshimura, Y., Tachibana, N., Kuroda, S., Miki, Y., Kitabayashi, S., Tong, P., Xiao, J., Tanaka, K., Hamamoto, H., Sekimizu, K., Yamamoto, K."A silkworm infection model to investigate Vibrio vulnificus virulence genes". Molecular Medicine Reports 14.5 (2016): 4243-4247.
Chicago
Yamamoto, M., Kashimoto, T., Yoshimura, Y., Tachibana, N., Kuroda, S., Miki, Y., Kitabayashi, S., Tong, P., Xiao, J., Tanaka, K., Hamamoto, H., Sekimizu, K., Yamamoto, K."A silkworm infection model to investigate Vibrio vulnificus virulence genes". Molecular Medicine Reports 14, no. 5 (2016): 4243-4247. https://doi.org/10.3892/mmr.2016.5782