Cultivation for 21 Days Should Be Considered to Isolate Respiratory

Jpn. J. Infect. Dis., 63, 338­341, 2010

Original Article

Cultivation for 21 Days Should Be Considered to Isolate Respiratory Adenoviruses from Samples Containing Small Numbers of Adenoviral Genomes Miki Enomoto1,2õ, Tsuguto Fujimoto1,4õ*, Masami Konagaya1,3, Nozomu Hanaoka1,4, Masatsugu Chikahira2, Kiyosu Taniguchi1, and Nobuhiko Okabe1,4 2Hyogo

1National Institute of Infectious Diseases, Tokyo 162­8640; Prefectural Institute of Public Health and Consumer Sciences, Hyogo 652­0032; 3TRUST Medical Co., Ltd., Hyogo 679­0105; and 4Adenovirus Reference Center of Japan, Tokyo 162–8640, Japan

(Received May 21, 2010. Accepted August 20, 2010) SUMMARY: Adenovirus types 1, 2, and 3 can usually be isolated in only a short time, although oc­ casionally it may take longer. This phenomenon has been explained empirically as being due to the viral load in the sample, although to date there has been no experimental confirmation of this. In this study we therefore tried to establish a correlation between the quantity of respiratory adenovirus genome in the clinical sample and the time required for its isolation. The correct choice of sensitive cell line is im­ portant for this purpose, thus we compared the sensitivity of three different cell lines (HeLa, A549, and RD), and found A549 to be the most sensitive to adenoviruses 1–3. Stored clinical samples (n  21) con­ taining adenoviruses 1–3 were diluted to make solutions containing between 10 and 108 copies/mL of adenovirus genome (n  242). These diluted clinical samples were then inoculated into A549 cells, which were cultivated for 21 days and the results compared to the number of viral genomes in each cul­ tivated sample. Adenoviruses could be isolated from all samples (41/41) containing Æ106 copies/mL within 6 days, whereas samples containing 10 and 102 copies/mL required cultivation for 12.6 } 3.8 and 11.2 } 3.8 days (mean } S.D.), respectively, before adenoviruses could be isolated. A cultivation time of 21 days should therefore be considered for the isolation of respiratory adenoviruses from samples containing º103 adenovirus genome copies/mL. of 6,280 adenovirus isolates were reported between January 2006 and June 2010, of which 2,090 (33.3z) isolates were species B and 3,328 (53.0z) isolates spe­ cies C adenoviruses. Adenovirus types 1 (n  937), 2 (n  1,828), and 3 (n  2,006) were the dominant sero­ types. In this study we used respiratory samples from which these three serotypes were isolated in an attempt to establish a correlation between the number of copies of adenovirus genome in clinical samples and their isola­ tion time.

INTRODUCTION Virus isolation using cultured cell lines is an indis­ pensable method for determining the infectivity and an­ tigenicity of adenoviruses that cause various infectious diseases, such as pharyngoconjunctival fever, pneumo­ nia, and gastroenteritis (1). The time required to isolate the adenovirus can vary depending on the cell line. Furthermore, it is known empirically that low adenovi­ rus titers require a long time to provoke a clear cytopathic effect (CPE) in cultured cells as several rounds of adenovirus replication are required to amplify the amount of virus to an appropriate level (1). However, to date, there has been no study focusing on the relationship between adenoviral genome copy num­ bers in clinical samples and the isolation period. If it were possible to predict the likelihood of being able to isolate an adenovirus from a clinical sample, and the time required to do so, this would reduce unnecessary labor­intensive cell cultures, which often require more than 21 days. According to the National Epidemiologi­ cal Surveillance of Infectious Diseases in Japan, a total

MATERIALS AND METHODS Samples: Of the 430 adenovirus isolates in an infec­ tious disease surveillance program (2004 to 2008) in Hyogo Prefecture, Japan, 408 (94.9z) were species B (types 3, 7, and 11) and species C (types 1, 2, 5, and 6). The most common of these adenoviruses (species B type 3 and species C types 1 and 2) were selected for this study. A total of 21 pharyngeal swabs were selected ran­ domly and used in this study. These samples were col­ lected from 0­ to 10­year­old patients during a disease surveillance program conducted from 2007 to 2008 and stored at |809 C until use. After culture, the isolates were identified as type 1 (n  7), type 2 (n  6), and type 3 (n  8) adenoviruses by a neutralization test. This study was conducted as part of a pathogen surveil­ lance program in Japan. Sensitivity of three different cell lines to adeno­

*Corresponding author: Mailing address: Infectious Dis­ ease Surveillance Center, National Institute of Infectious Diseases, Toyama 1­23­1, Shinjuku­ku, Tokyo 162­8640, Japan. Tel: {81­3­5285­1111, Fax: {81­3­5285­1129, E­mail: fujimo­t—nih.go.jp õThese two authors contributed equally to this study. 338

vector and used as a standard. Isolation of adenoviruses from dilute clinical sam­ ples: On the basis of the adenoviral genome copy num­ bers in the 21 clinical samples obtained by real­time PCR, we diluted each sample to contain 104 (n  4), 105 (n  4), 106 (n  7), 107 (n  4), and 108 (n  2) copies/mL of adenoviral genomes with serum­free D­ MEM medium. Ten­fold serial dilutions of these clinical samples were then performed with the same medium to obtain a total of 121 diluted samples containing 10–108 copies/mL of the adenoviral genome. A 50­mL aliquot of each diluted sample was inoculated into two wells on a monolayer of A549 cells in a 24­well microplate. Each well was considered to be one sample, therefore a total of 242 wells were observed for 21 days. D­MEM con­ taining 2z FBS was used as maintenance medium, and the medium was changed every 2–3 days after inocula­ tion. Viral culture was continued without blind passage. Positive cell culture was defined as observation of a CPE in about 25z or more of the area in a well. The results of the virus isolation study were then compared with the number of copies of the adenoviral genome ini­ tially present.

viruses: The pharyngeal swabs (n  21) were cen­ trifuged at 3,000 rpm for 20 min and the resulting super­ natants used for isolation of the virus in HeLa, A549, and RD cell lines. Each cell line was cultured in a monolayer on 24­well microplates (1.88 cm2 well). The culture medium (D­MEM) was removed from the wells and 50 mL of sample inoculated in each well. After incu­ bation for 30 min in a 5z CO2 incubator (359C), the in­ oculated samples were removed, 1 mL of D­MEM cul­ ture medium containing 2z FBS was added to each well, and culture continued for a further 7 days. The TCID50/mL of adenovirus type 2 isolate was also com­ pared in the A549, HeLa, and RD cell lines. Thus, cells were cultured in 96­well microplates (seeded 1–2 ~ 103 cells in 140 mL of D­MEM { 2z FBS the day before in­ oculation). Adenovirus type 2 was diluted 10|2 to 10|12 with D­MEM (without serum) and a 50­mL aliquot of diluted virus solution inoculated in each of eight wells then incubated in a CO2 incubator (359C). The microplates were checked every day for CPE, and the adenovirus culture was considered positive when microscopic observation showed a CPE of about 25z or more of the well area for each cell line (Fig. 1). Measurement of the number of adenoviral genome copies: The number of adenoviral genome copies per microliter of clinical sample was determined by real­ time PCR following the method described by Watanabe et al. (2) with a slight modification (3). The primers 5?­GACATGACTTTCGAGGTCGATCCCATGGA­3? (Hex3) and 5?­CCGGCTGAGAAGGGTGTGCGCAG GTA­3? (Hex4) were derived from highly conserved adenovirus hexon genes as described by Echavarria et al. (4). PCR was carried out using a Rotor­Gene RG­3000 (Qiagen, Hilden, Germany) with a cycle of 959 C for 5 s, followed by 40 cycles at 609 C for 20 s after initial denaturation at 959C for 10 s. The DNA of the prototype 2 adenovirus was amplified following the PCR method of Echavarria et al. (4). The product of this amplification (140 bp) was cloned into a pCR2.1

RESULTS Selection of optimal cell line for adenovirus isolation: In this study, positive cell culture was defined as obser­ vation of a CPE in about 25z or more of the area in a well. The virus isolation rates were 95.2z (20/21) in A549 and HeLa cells and 81.0z (17/21) in RD cells. The A549 and HeLa cells therefore had higher virus iso­ lation rates than RD cells. Furthermore, the average number of days required for the samples to become cul­ ture­positive was shortest for A549 cells (2.8 days), fol­ lowed by 4.3 days for HeLa cells and 5.5 days for RD cells. We therefore used A549 cells to detect adenoviruses subsequent experiments. The sensitivity (TCID50/mL) of the adenovirus type 2 isolate was different for the three different cell lines (HeLa, A549, and RD) tested, with the highest sensitivity being ob­ served for A549 cells (Table 1). Number of copies of the adenovirus genome in dilut­ ed clinical samples and adenovirus isolation rate: The 242 diluted adenovirus samples contained 10 (n  14), 102 (n  14), 103 (n  14), 104 (n  14), 105 (n  8), 106 (n  6), and 107 (n  2) copies/mL of type 1 adenovirus genome, 10 (n  12), 102 (n  12), 103 (n  12), 104 (n  12), 105 (n  9), 106 (n  6), 107 (n  2), and 108 (n  2) copies/mL of type 2 adenovirus genome, and 10 (n  16), 102 (n  16), 103 (n  16), 104 (n  16), 105 (n  16), 106 (n  14), 107 (n  7), Table 1. Comparison of sensitivity of three different cells (HeLa, A549, and RD) to adenovirus type 2 isolate Cell line

Fig. 1. Cytopathic effects (CPE) induced by adenovirus type 1. Respiratory adenoviruses types 1–7 generally produce similar CPE. In this study, we considered the adenovirus culture to be culture­positive when microscopic observation showed a CPE about 25z or more. The scale bar indicates a distance of 100 mm.

HeLa A549 RD 1):

339

Infectivity (log TCID50/mL) Day 7

Day 14

8.0 8.7 6.8

9.6 10.0 8.1

Day 21 Not available1) 10.2 9.0

HeLa cell was not possible to maintain over 14 days.

Table 2. Adenovirus types 1, 2, and 3 isolation from diluted samples in A549 cells, which were cultivated for 21 days (n  242) Serotype

Virus isolation

Copy number of adenoviral genome/mL 10

102

103

104

105

106

107

108

Total

Type 1

positive negative

2 12

4 10

8 6

10 4

6 2

6 0

2 0

38 34

Type 2

positive negative

1 11

5 7

8 4

9 3

9 0

6 0

2 0

2 0

42 25

Type 3

positive negative

6 10

16 0

16 0

16 0

16 0

14 0

7 0

2 0

93 10

Total

positive negative positive rate (z)

9 33 (21)

25 17 (60)

32 10 (76)

35 7 (83)

31 2 (94)

26 0 (100)

11 0 (100)

4 0 (100)

173 69 (71)

Table 3. Days required for adenovirus isolation shown in Table 1 (n  1731)) Days required for adenovirus isolation Type 1 Type 2 Type 3

mean } standard deviation (days)

Copy number of adenoviral genome/mL 10 (n  9)

102 (n  25)

8.5 } 0.5 9.5 } 3.6 7.1 } 1.5 14.0 } 0 10.8 } 3.4 11.5 } 4.4 13.7 } 3.8 11.8 } 3.9 7.3 } 1.4

Total 12.6 } 3.8 11.2 } 3.8 95z confidence interval (days) 10.1–16.9 9.6–12.8 1):

103 (n  32)

8.3 } 3.1 7.2–9.5

108 (n  4)

Total (n  173)

104 (n  35)

105 (n  31)

106 (n  26)

107 (n  11)

6.0 } 2.9 8.4 } 3.1 5.1 } 0.6

6.8 } 2.4 7.3 } 3.0 4.1 } 0.3

3.3 } 1.9 4.5 } 0.8 3.0 } 0.9

2.0 } 0 5.0 } 0 1.9 } 1.0

4.0 } 0 1.0 } 0

6.2 } 2.6 5.3–7.1

5.6 } 2.5 4.7–6.5

3.4 } 1.3 2.9–4.0

2.5 } 1.4 1.4–3.5

2.5 } 1.5 6.9 } 4.2 0–5.3 0–16.9

6.6 } 3.9 8.3 } 4.1 6.4 } 4.2

Among 242 samples, 69 samples were negative by cell culture.

and 108 (n  2) copies/mL of type 3 adenovirus genome. The adenovirus isolation rate increased with the number of genome copies present, reaching 100z for more than 106 copies/mL (Table 2). Samples containing 101 and 102 copies/mL of adenoviral genome had low virus isolation rates of 14z (6/42) and 45z (19/42), respectively, on day 14, and 21z (9/42) and 60z (25/42), respectively, on day 21. Nine of the 11 samples (82z) which became culture­positive between days 14 and 21 had genome concentrations of 102 copies/mL or less. The adenoviral copy numbers per microliter for TCID50 were in the range, 102–103 for types 1 and 2 and 10–102 for type 3. Number of copies of the adenovirus genome and the number of days required for isolation: Our study of the time required to isolate the 173 viruses that were culture­ positive by day 21 showed that, as would be expected, lower concentrations of genome copies required longer culture periods (Table 3). In particular, 10 and 102 copies/mL required long culture periods of 12.6 } 3.8 and 11.2 } 3.8 days (mean } S.D.), respectively.

adenovirus isolate was also compared and A549 found to have a higher log TCID50 than both HeLa and RD cells. The A549 cell line was therefore considered to be optimal for the isolation of adenoviruses. These results are consistent with a previous study by Woods and Young (5), which found that the adenovirus isolation rate is higher in A549 cells than in HEp­2, MRC­5, and PMK cells. The isolation of viruses, including adenoviruses, from clinical samples is important to determine their infectivi­ ty and other properties, although occasionally it takes up to a month to determine whether this isolation has been successful or not (1). Viral culture is traditionally performed for only 2 weeks as this period of time is usu­ ally sufficient to isolate enteroviruses and influenza viruses. Adenovirus types 1, 2, and 3 (respiratory adenoviruses) are also usually easy to isolate within 2 weeks, although our results show that longer cultivation times are required when the sample contains fewer cop­ ies of the adenoviral genome. Watanabe et al. (2) used adenovirus genome concentrations of between 102 and 108 copies/mL in pharyngeal samples and 1 and 102 copies/mL in blood samples. In our routine diagnosis, genome copies as low as 102 copies/mL were also found in pharyngeal samples, with less than 103 copies/mL in blood samples. Adenovirus type 3 could be isolated from 100z (16/16) of A549 cells to which a solution containing 500 copies of adenoviral DNA (50 mL ~ 102 copies/mL), whereas only 6 out of 16 (37.5z) adenovirus type 3 could be isolated from solutions containing 50 copies of adenoviral DNA. Samples containing 10 and 102

DISCUSSION The majority of patients in the virus surveillance had species B (particularly type 3) and C (types 1, 2, 5, and 6) adenoviruses, therefore samples positive for species B (type 3) and C (types 1 and 2) were used to investigate the sensitivity of three different cell lines (A549, HeLa, and RD) to adenoviruses. The results showed that the virus isolation rates in A549 and HeLa cells were higher than in RD and that the isolation time was lower in A549 cells than in HeLa cells. The cell sensitivity to 340

copies/mL of adenoviral genome required 12.6 } 3.8 and 11.2 } 3.8 days, respectively, for isolation of adenovirus types 1, 2, and 3. The adenovirus replication process was thought to be inefficient (6). Thus, although we counted the genome copy numbers of adenoviruses in this study, only a small portion of the adenoviral genome detected is thought to represent in­ tact adenovirus particles. In summary, a cultivation time of 21 days should be considered for the isolation of respiratory adenoviruses from samples containing º103 adenovirus genome copies/mL as these samples are like­ ly to contain only a small number of intact adenovirus.

REFERENCES 1. Wold, W.S.M. and Marshall S.H. (2007): Adenoviruses. p. 2395–2436. In Knipe, P.M. and Howly, P.M. (ed.), Fields Virolo­ gy. 5th ed. Lippincott Williams and Wilkins, a Wolters Kluwer Business, Philadelphia. 2. Watanabe, M., Kohdera, U., Kino, M. et al. (2005): Detection of adenovirus DNA in clinical samples by SYBR Green real­time polymerase chain reaction assay. Pediatr. Int., 47, 286–291. 3. Fujimoto, T., Konagaya, M., Enomoto, M. et al. (2010): Novel high­speed real­time PCR method (Hyper­PCR): results from its application to adenovirus diagnosis. Jpn. J. Infect. Dis., 63, 31–35. 4. Echavarria, M., Forman, M., Ticehurst, J., et al. (1998): PCR method for detection of adenovirus in urine of healthy and human immunodeficiency virus­infected individuals. J. Clin. Microbiol., 36, 3323–3326. 5. Woods, G.L. and Young, A. (1988): Use of A­549 cells in a clinical virology laboratory. J. Clin. Microbiol., 26, 1026–1028. 6. Hierholzer, J.C. (1995): Adenoviruses. p. 169–188. In Lennette, E.H., Lennette, D.A., Lennette, E.T. (ed.), Diagnostic Procedures for Viral, Rickettsial, and Chlamydial Infections. 7th ed. Ameri­ can Public Health Association, Washington, D.C.

Acknowledgments We want to thank sentinel hospitals and

Government of Hyogo Prefecture for providing clinical samples. We also thank Dr. Arun Kumar Adhikary for critical review. This study was supported in part by a grant from Daido Life Wel­ fare Foundation, Osaka, Japan. This study was also supported in part by grant for Research Promotion of Emerging and Re­emerging Infec­ tious Diseases from the Ministry of Health, Labour and Welfare of Japan.

Conflict of interest None to declare.

341