LETTERS

2. Hawthorne VM, Lauder IM. Tuberculosis continuous marine crustacean cell bated in three separate rooms at 37°C, in man, dog, and cat. Am Rev Respir Dis. lines has become an obstacle to con- 35°C, and 33°C, respectively. If a 1962;85:858–69. 3. Foster ES, Scavelli TD, Greenlee PG, ducting research on viral disease in cytopathic effect (CPE) was not evi- Gilbertson SR. Cutaneous lesions caused (2,3). During the last 20 years, dent within 7 days, cell monolayers by Mycobacterium tuberculosis in a dog. J many researchers searched for substi- were washed with Hank’s balanced Am Vet Med Assoc. 1986;188:1188–90. tute cell lines on which to study salt solution (HBSS) six times to 4. Snider WR. Tuberculosis in canine and feline populations. Am Rev Respir Dis. shrimp (4,5). Audelo-del- eliminate viral particles from the pri- 1971;104:877–87. Valle et al. likely chose RD, Hep-2C, mary extract or from infected cells. 5. Liu S, Weitzman I, Johnson GG. Canine and BGM cell lines because TSV was Then cells were lysed in 2 mL HBSS, tuberculosis. J Am Vet Med Assoc. “recently reported to be genomically the lysate was clarified, and a portion 1980;177:164–7. 6. Snider WR, Cohen D, Reif JS, Stein SC, related to the cricket paralysis of of it was used for the first passage. Prier JE. Tuberculosis in canine and feline the genus, family This procedure was repeated three populations. Am Rev Respir Dis. Dicistrovirdae of the ‘ times. The control cell lines were 1971;104:866–76. superfamily’” (1), and these cell lines injected with diluted extract from 7. Michalak K, Austin C, Diesel S, Bacon MJ, Zimmerman P, Maslow JN. Mycobacterium were susceptible to some picor- healthy shrimp, and passage was con- tuberculosis infection as a zoonotic disease: naviruses. If their findings are correct, ducted as described earlier. RNA sam- transmission between human and ele- they may have found substitute cell ples were extracted and purified from phants. Emerg Infect Dis. 1998;4:283–7. lines for isolating and studying TSV. 150-µL lysates of primary cells and 8. Michel AL, Huchzermeyer HF. The zoonot- ic importance of Mycobacterium tuberculo- To confirm their findings, we selected four passage cells and used as tem- sis: transmission from human to monkey. J two mammalian cell lines, Hep-2 and plates for RT-PCR analysis to deter- S Afr Vet Assoc. 1998;69:64–5. Vero, which are highly sensitive to mine the presence of TSV. some (6,7), and tested No CPE was observed in either the Address for correspondence: Paul C. Erwin, them to determine their susceptibility Hep-2 or Vero cell line that had been 1522 Cherokee Trail, Knoxville, TN 37920, to TSV. injected with TSV after 7 days of cul- USA; fax: 865-594-5738; email: paul.erwin@ The TSV extract was prepared ture at any of the three temperatures state.tn.us from frozen cephalothoraxes of tested, and CPE was not found after shrimp, Litopenaeus vananmei, that the fourth passage. The RT-PCR were infected with TSV (confirmed analysis resulted in weak amplifica- by standard reverse transcriptase– tion (positive) from the first lysate, polymerase chain reaction [RT-PCR]) but no amplification was found in (8). To verify the TSV extract’s valid- lysates of four passage cells. Had ity, 50 µL of diluted TSV extract TSV replicated (productive infection) Virus and (approximately 0.8% volume of in either of the two cell lines, RT-PCR shrimp body weight) was injected into would have shown a strong amplifica- Mammalian Cell each of eight healthy shrimp, L. van- tion from each lysate. Such a weak Lines namei. Another eight healthy shrimp amplification may have been the (control group) were injected with a result of residual extracellular viruses To the Editor: Audelo-del-Valle et diluted extract prepared from frozen that remained in the cell culture flask al. concluded that human and monkey cephalothoraxes of healthy shrimp. after washing. However, after first cell lines (rhabdomyosarcoma [RD], All of the TSV-injected shrimp died passage and repeated washing with human larynx carcinoma [Hep-2C], within 6 days and were TSV-positive; HBSS, any remnants of the original and Buffalo green monkey kidney control shrimp did not die and were medium were not likely to have been [BGM]) could be infected by a TSV-negative, which showed that our present. Therefore, our result showed penaeid shrimp virus, Taura syndrome TSV extract was active and viable. that TSV was incapable of infecting virus (TSV) (1). They also concluded The TSV extract was transferred into Hep-2 and Vero cell lines. that spp. could likely be a cell culture flasks according to a Generally, aquatic viruses replicate reservoir of a virus that might become method previously reported (9). The in cells of aquatic animals at pathogenic to humans and other mam- cell monolayers were exposed to 100 20°C–35°C, their natural environ- mals (1). µL of diluted and filtered TSV mental temperature. We incubated Though researchers have tried to extracts for 1 hour; the extracts were cultures as noted earlier, as we did not develop continuous marine crustacean then removed from the flasks, 2 mL of know which temperature was most > cell lines for 30 years, their efforts maintenance medium was added to conducive for viral replication; all have not been successful. The lack of each flask, and the flasks were incu- attempts were unsuccessful. Hep-2

2260 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 10, No. 12, December 2004 LETTERS and Hep-2C derive from the same tis- areas in the world, and L. vannamei 9. Yin Z. Animal virology. Beijing: Science sue (human Caucasian larynx carcino- (principal host for TSV) are eaten by Press; 1985. 10. Mari J, Poulos BT, Lightner DV, Bonami ma), while Vero and BGM derive people worldwide (8). In , some JR. Shrimp Taura syndrome virus: genomic from another organ (Africa green persons eat fresh shrimp without dis- characterization and similarity with mem- monkey kidney). Thus, Hep-2 and infecting them; however, no evidence bers of the genus Cricket paralysis-like Vero cell lines that we used are likely shows that TSV can infect humans. viruses. J Gen Virol. 2002;83:915–26. 11. Mayo MA. A summary of taxonomic susceptible to TSV if the virus can The results of our study show that changes recently approved by ICTV. Arch infect Hep-2C and BGM as reported TSV cannot infect mammalian cell Virol. 2002;147:1655–63. by Audelo-del-Valle et al. lines or cells. The difference between our meth- Address for correspondence: Peng Luo, South ods and those used by Audelo-del- This work was supported by innova- China Sea Institute of Oceanology, The Chinese Valle et al. may explain the discrepant tion-projects funds provided by The Academy of Sciences, 164 Xingang Xi Road, result. If CPE occurred “usually from Chinese Academy of Sciences (Projects Guangzhou, 510301, People’s Republic of 19–23 hours” and “cells were then No. ZKCX2-211). China; fax: 8620-8445-1672; email: harvested and lysed” for next injec- [email protected] tion, TSV most likely persisted in the Peng Luo,* Chao-Qun Hu,* lysate after the third passage because Chun-Hua Ren,* the cell monolayers had not been and Zhao-Feng Sun* washed as they were in our method. *The Chinese Academy of Sciences, According to the time the CPE was Guangzhou, People’s Republic of China Bartonella observed and the methods of Audelo- del-Valle et al., we assumed that, in References clarridgeiae and their study, cells might be passaged at B. henselae in least three times within 1 week, 1. Audelo-del-Valle J, Clement-Mellado O, Magaña-Hernández A, Flisser A, Montiel- Dogs, Gabon whereas TSV might remain viable and Aguirre F, Briseño-García B. Infection of infective for 1 week. Additionally, the cultured human and monkey cell lines with To the Editor: The genus Barton- Office International des Epizooties extract of penaeid shrimp infected with ella contains several recently recommended an injection volume of Taura syndrome virus. Emerg Infect Dis. 2003;9:265–6. described species, many of which are 1% of shrimp body weight (8); 2 Toullec JY. Crustacean primary cell culture: emerging human pathogens. Human Audelo-del-Valle et al. used 10%. a technical approach. Methods Cell Sci. infections are mostly due to Bartonella With such a large dose, shrimp could 1999;21:193–8. henselae and B. quintana. Like many be infected easily with TSV from the 3. Crane MS. Mutagenesis and cell transfor- mation in cell culture. Methods Cell Sci. vectorborne disease agents, Bartonella initial medium and die suddenly. 1999;21:245–53. species have a natural cycle. This cycle Moreover, the evidence of successful 4. Philip CL, Lu YA, James AB. Growth of the contains a reservoir host, in which infection from photos of CPE only is penaeid shrimp virus infectious hypoder- Bartonella species cause an intraery- not sufficient; Audelo-del-Valle et al. mal and hematopoietic necrosis virus in a fish cell line. J Virol Methods. throcytic bacteremia, and a vector, should offer more convincing evi- 1990;28:273–80. which transmits the bacteria from the dence from images of viral particles in 5. Yue YL, Li CY, Yang SH, Lu Q, Tao ZS, reservoir host to a new susceptible host cells by electron microscope or in situ Wang WD, et al. Cell isolation and culture (usually the uninfected reservoir host) hybridization. Therefore, we think the of penaied shrimp paramyxovirus-like virus. Chinese Journal of Veterinary (1). In the case of B. quintana and B. CPE that Audelo-del-Valle et al. Science. 1997;17:306–7. bacilliformis, the natural host is reported was not caused by TSV but 6. Johnston SLG, Siegel CS. Presumptive human. In Bartonella diseases, by a virus contaminant or some harm- identification of enteroviruses with RD, humans act as accidental hosts. Among ful component from shrimp extract. Hep-2, and RMK cell lines. J Clin Microbiol. 1990;28:1049–50. the nonhuman Bartonella species that The structure of the TSV 7. Kok T, Pryor T, Payne L. Comparison of infect humans, B. henselae is most is similar to that of small rhabdomyosarcoma, Buffalo green monkey commonly encountered and usually insect–infecting RNA viruses (10), kidney epithelial, A549 (human lung causes cat-scratch disease. However, which belong to a renamed virus epithelial) cells and human embryonic lung fibroblasts for isolation of enteroviruses for several cases of infections in humans genus, Cripavirus (11). No published clinical specimens. J Clin Virol. attributable to other Bartonella reports have shown that other viruses 1998;24:61–5. species, including B. elizabethae, B. in this genus are able to infect mam- 8. Office International des Epizooties. grahamii, B. vinsonii arupensis, B. malian cells or cell lines. Moreover, Diagnostic manual for aquatic animal health diseases. 3rd ed. Paris: The Office; vinsonii berkhoffii, and possibly B. TSV is prevalent in 2000. clarridgeiae, have been reported (1).

Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 10, No. 12, December 2004 2261