Comparison of Standard Tube and Shell Vial Cell Culture Techniques for the Detection of Cytomegalovirus in Clinical Specimens

Transcription

1 JOURNAL OF CLINICAL MICROBIOLOGY, Feb. 1985, p /85/ $02.00/0 Copyright 1985, American Society for Microbiology Vol. 21, No. 2 Comparison of Standard Tube and Shell Vial Cell Culture Techniques for the Detection of Cytomegalovirus in Clinical Specimens CURT A. GLEAVES,' THOMAS F. SMITH,'* ELIZABETH A. SHUSTER,2 AND GARY R. PEARSON2 Departments of Laboratory Medicine' and Cell Biology,2 Mayo Clinic and Mayo Foundation, Rochester, Minnesota Received 3 August 1984/Accepted 22 October 1984 A monoclonal antibody was used to detect an early antigen of cytomegalovirus (CMV) by fluorescence 16 h after inoculation of MRC-5 monolayers in 1-dram (ca. 3.7-ml) shell vials and low-speed centrifugation. Of 770 specimens (urine, blood, lung tissue, sputum) processed in shell vials, 124 (16%) were positive for the virus at 16 h postinfection. CMV was isolated in standard tube cell cultures (average time, 9 days) from only 88 specimens, but there were no instances (with the exception of 2 blood specimens) in which CMV was recovered from tube cultures but not from shell vials. Additional specimens from 18 patients were positive in the shell vial assay but negative in the conventional tube cell culture assay. Other specimens from 14 of the 18 patients yielded CMV in conventional tube cell cultures. Of the 4 patients from whom CMV was not recovered from other specimens by conventional tube cell culturing, all had evidence of recent CMV infections, as indicated by a fourfold or greater rise in antibody titer. The specificity of the shell vial assay for the detection of CMV is supported by assays of other specimens from the same patients yielding the virus or serological evidence indicating recent infections, the known enhancement of CMV detection after centrifugation of the shell vials, and the distinct and easily recognizable fluorescence confined to the nuclei of CMV-infected celis. Our data indicate that the shell vial cell culture assay for the detection of CMV is as specific as and more sensitive than conventional tube cell culturing for the diagnosis of CMV infections. Symptomatic cytomegalovirus (CMV) infection is usually manifested in congenitally infected infants, organ transplant recipients, and other patients, such as those who have neoplastic disease and a compromised immune system (1, 9, 12). CMV infection traditionally has been detected in laboratories by the recognition of characteristic cytopathic effects (CPE) in human diploid fibroblast cell cultures (14, 17). However, although this technique is ultimately the most sensitive method for CMV detection, viral isolation is often hampered by slow growth of the virus, generally requiring 7 to 21 days for CPE to occur and in some instances even longer (4, 17). Recently, other methods have been developed to detect CMV more rapidly by using cell cultures followed by detection of the virus or antigenic components by immunological techniques. Stagno et al. (16), using specimens from congenitally infected neonates, detected CMV in cell cultures by using a hyperimmune polyclonal antibody to the early nuclear antigen at 24 h postinfection. Others have used monoclonal antibodies made to CMV antigens and have reported the successful detection of CMV with laboratory strains at 2 to 3 h postinfection (5, 7, 11); however, direct comparisons of this method and standard tube cell culturing were not made with clinical isolates. One general disadvantage is that immune reagents have not been available commercially for evaluation by clinical laboratories. Hybridization techniques have been described for direct CMV DNA detection (2); however, the use of radioactive probes discourages the routine diagnostic use of such techniques. We recently reported the use of centrifugal force to enhance the detection of CMV in shell vials containing circular cover slips with a monolayer of MRC-5 cells. A * Corresponding author. 217 monoclonal antibody to an early nuclear antigen was used to detect CMV in urine specimens at 36 h postinfection (3). In an effort to provide more rapid diagnosis of CMV infections, we used this technique in the present study to detect CMV in clinical specimens at 16 h postinfection and compared the results to those from conventional tube cell culturing. MATERIALS AND METHODS Cells and virus. MRC-5 cells were obtained from Flow Laboratories, Inc., McLean, Va., and Viromed Laboratories, Minneapolis, Minn. Eagle minimal essential medium (MEM) containing 10% fetal bovine serum, penicillin, streptomycin, and gentamicin was used (14). The AD169 strain of CMV (ATCC VR-538) was used as a positive control. CMV monoclonal antibody. A monoclonal antibody (2H2.4) directed against the 72,000-dalton early nuclear protein of CMV was developed by Gary R. Pearson and Elizabeth A. Shuster (Department of Cell Biology, Mayo Clinic) (3) and can be obtained commercially from Biotech Research Laboratories, Inc., Rockville, Md. Inoculation of clinical specimens. MRC-5 cells were subcultured in culture tubes (16 by 125 mm) and maintained with MEM until inoculated with specimens. The processing and inoculation of specimens were done as previously described (3, 14; C. A. Gleaves, T. F. Smith, A. D. Wold, and W. R. Wilson, Am. J. Clin. Pathol., in press). All urine, blood, and lung tissue specimens submitted to the laboratory were routinely included in this study. Leukocytes from 5 ml of heparinized blood were obtained by Ficoll-Paque/Macrodex (Pharmacia Fine Chemicals, Piscataway, N.J.) separation techniques, and the cells were suspended in 2 ml of Eagle basal medium. A 1-ml portion was inoculated into a conventional cell culture tube, and 0.3 ml was added to each of three shell vials. Lung tissue was suspended in 5 ml of

2 218 GLEAVES ET AL. J. CLIN. MICROBIOL. FIG. 1. Time course detection of CMV strain AD169 in MRC-5 cells after centrifugation with monoclonal antibody. (A) 6-h postinfection; (B) 12-h postinfection; (C) 16-h postinfection; and (D) 36-h postinfection. nutrient broth and homogenized with a Stomacher (Tekmar Co., Cincinnati, Ohio), and the extract was centrifuged at 700 x g for 15 min. Conventional tube cell cultures were inoculated with 0.2 ml, and each shell vial was inoculated with 0.1 ml. After inoculation, culture tubes were placed on a roller drum at 36 C and subsequently examined at a 125 x magnification for typical CPE caused by CMV (14). Rapid detection of CMV in clinical specimens at 16 h postinoculation. MRC-5 cells were seeded (50,000 cells per ml) into 1-dram (ca. 3.7-ml) shell vials each containing a 12-mm round cover slip and were maintained with MEM until used. Before inoculation, MEM was removed from the shell vial cell cultures, and 0.1 ml of the specimen was added to each of the two vessels. Centrifugation, incubation, and staining of the cell cultures after infection were done as previously described (3, 6, 10). Briefly, infected shell vials were centrifuged at 700 x g for 1 h at room temperature (Sorvall RT6000 centrifuge; Du Pont Co., Wilmington, Del.). After centrifugation, 1.0 ml of MEM was added to the shell vials, and the cultures were incubated at 36 C. After 16 h of incubation, cover slips with infected monolayers were washed two times in phosphate-buffered saline and fixed in cold acetone for 20 min. Cell monolayers were then stained by the indirect immunofluorescence technique with a CMV monoclonal antibody (3) and a fluorescein isothiocyanate-labeled goat antimouse immunoglobulin G conjugate (Cooper Biomedical, Malvern, Pa.). CMV detection. The AD169 strain of CMV and two urine specimens positive for CMV were assayed in shell vials to determine the optimal period for assaying with the monoclonal antibody at various times postinfection. The CMV strains were processed as described above, and cover slips were fixed and stained at 2, 6, 12, 16, and 36 h postinoculation. CMV serology. Serum immunoglobulin G titers for CMV were determined by the anticomplement immunofluorescence test (14). RESULTS CMV antigens in infected monolayers in shell vials were detected by the monoclonal antibody after fluorescent-antibody staining as early as 2 h postinfection. Although intense staining was observed at 6 to 12 h postinoculation, we stained the infected cell monolayers at 16 h in this study because most specimens arrived at the laboratory in the early afternoon (Fig. 1). Of 770 specimens processed in shell vials and examined by fluorescence after reaction with the monoclonal antibody to CMV, 124 (16%) were positive for CMV at 16 h postinfection. CMV was isolated in standard tube cell cultures (average, 9 days) from only 88 of these 770 specimens. As expected, the yield of CMV from urine specimens in the 16-h assay was most productive (19.4%), as compared with blood specimens (3.7%) and lung tissue specimens (14.3%). CMV was detected in 36 specimens that were obtained from 23 patients and were processed in shell vial cell TABLE 1. Detection of CMV by the standard tube cell culture assay and by the shell vial assay No. of specimens in which CMV was detected Specimen source (n) Shell vial by: Standard tube cell assay culture assay Urine (555) Blood (133) 5a 2b Ling (56) 8 7 Other (26) 2C 2< Total (16%) 88 (11%) a Positive in the shell vial assay only, negative in the standard tube cell culture assay. b Positive in the standard tube cell culture assay only, negative in the shell vial assay. C Both samples were from sputum.

3 VOL. 21, 1985 CELL CULTURING FOR DETECTION OF CYTOMEGALOVIRUS 219 TABLE 2. Additional specimens for culturing and serology from patients from whom a urine specimen(s) was positive for CMV in shell vials at 16 h but negative in standard tube cell cultures Date CMV detected by: Patient (mo/day/yr) Shell vial Conventional Serological specimen was assay at 16 h tube cell titera obtained postinfection culture assay 1 7/8/83 7/18/83 7/25/83 8/1/83 8/8/83 8/15/83 8/18/83 8/19/83 8/23/83 8/24/ /18/83 11/1/83 11/10/83 11/14/83 11/21/83 11/28/83 12/12/83 12/16/83 (Blood) 3 10/20/83 11/22/83 11/28/83 12/12/ /7/83 11/8/83 11/14/ /7/83 11/23/ /19/83 12/27/83 1/3/ /20/83 12/29/83 1/3/84 1/9/84 1/16/84 8 1/17/84 9 1/9/84 1/24/84 2/3/ /19/84 1/27/84 1/27/84 (Lung) 1/31/84 il 1/3/84 2/7/84 2/13/84 2/21/84 2/27/ /2/84 <1:5 <1:5 1:160.1:640.1:640.1:640.1:640.1:640.1:640 1:10 21:640.1:640.1:640.1:640 1:160-1:10-1:10.1:640.1:640 TABLE 2-Continued Date CMV detected by: Patient (mo/day/yr) Shell vial Conventional Serological specimen was assay at 16 h tube cell titer' obtained postinfection culture assay 2/21/84 -.1:640 3/14/84 -.1: /19/84 (Blood) -.1:640 1/20/84.1:640 2/22/84 -.1: /14/ <1:5 3/5/84 (Blood) 3/5/ /26/84 (Lung).1:640 3/26/84 (Lung).1:640 3/27/84 (Blood) -.1:640 3/28/84.1:640 4/12/84 -.1: /7/84.1:640 4/21/84 -.1: /16/84 (Blood) - <1:5 4/23/84-1:160 4/23/84 (Blood) - 1:160 4/27/84.1:640 4/28/84.1: /24/84 <1:5 4/30/84 (Blood) - 1:10 5/1/84 1:10 5/7/84 1:160 a Determined by the anticomplement immunofluorescence test. 1 :10 1:160 cultures and did not produce characteristic CPE in the standard tube cell culture system (Table 1). Sera and addi- - 1:10 tional specimens for the laboratory diagnosis of CMV by - 1:10 culturing were obtained from 18 of these 23 patients (Ta- ble 2). The remaining five positive specimens were submitted from individuals outside the Mayo Clinic (Mayo Re- 1:40 gional Laboratory), and additional specimens were not obtained from these patients. CMV was detected by conven- tional tube cell culturing in specimens from all but 4 of the 18 1:160 patients. Therefore, 14 of the 18 patients had at least one.1:640 other specimen positive for CMV, and 9 of 13 individuals.1:640 excreted CMV on two or more occasions in earlier or 1:40 subsequent samples f the 4 patients (patient 6, 10, 12, and 14) and from whom * CMV was not recovered from other specimens by conven- 1:160 tional tube cell culturing, all had laboratory evidence of - 1:160 CMV infections, as indicated by a fourfold or greater rise in.1:640 antibody titers. Overall, of the 18 patients, 14 had fourfold or greater rises in antibody titers to CMV. Of the four patients - 1:160 without serological conversions, three had antibody titers of - -1:640-1: :640 Thus, the specificity of the monoclonal antibody assay at -.1: h postinfection was supported by the demonstration of CMV infections in other specimens from all 18 patients by 1:40 culturing or by serological determinations. There was complete correlation in the detection of CMV 1:160 from lung tissue specimens between the shell vial and 1:160 conventional tube cell culture assays, with the exception of * 1:160 one specimen (from patient 10; Table 2). There were five <1:5 blood specimens (leukocytes separated from heparinized * blood specimens with Ficoll-Paque/Macrodex) in which

4 220 GLEAVES ET AL. CMV was exclusively detected in shell vials and two blood specimens in which CMV produced CPE only in conventional tube cell cultures. DISCUSSION Standard tube cell culturing has been recognized as the most sensitive technique available for CMV detection (4, 17). The detection of CMV infections by cell culturing, however, depends on the recognition of typical CPE, which develop only after several days in most specimens (3, 4, 17). Techniques for direct detection in urine sediment and tissue sections have been used in an effort to decrease the time required for CMV detection; however, these techniques have been shown to be substantially less sensitive than cell culturing (5, 8, 15; Gleaves et al., in press). We have described a rapid and sensitive method for the detection of CMV in clinical specimens at 16 h postinfection. The assay system makes use of cell culturing but is not dependent on the observation of CPE before detection is attempted. With the use of centrifugal enhancement, which increases the sensitivity of CMV detection from 37.5% to 100%, as compared with conventional tube cell culturing (3), and a monoclonal antibody to an early antigen as the immunological probe, we increased our rate of detection of CMV by 5% and decreased the time required for detection from an average of 9 days to 16 h postinfection. With the laboratory-adapted AD169 strain of CMV, fluorescence of infected cells was observed as early as 2 h postinfection. However, fresh clinical isolates of CMV generally had to be subcultured for at least 4 h before CMV could be detected with the monoclonal antibody. Although the fluorescence produced in CMV-infected cells was easily recognizable at 6 h postinfection, other cover slips would probably need to be examined at a later time, such as 16 h, to ensure maximum sensitivity of the method. Because most laboratories receive specimens in the late morning and would inoculate them in the afternoon, the results of a 6-h-postinfection assay would be available only after normal working hours, and so a 6-h postinfection assay would probably not be significantly more advantageous than an assay performed at 16 h postinfection. Of a total of 109 CMV-positive urine specimens, 32 were detected exclusively in shell vials. Although specific toxic effects were not noted in these specimens, urine specimens may possibly impair the development of specific CPE caused by CMV. Because only early antigen formation, rather than CPE, is required to detect CMV in shell vials, the monoclonal antibody assay would seem to obviate this potential problem. Of seven blood specimens positive for CMV, five were detected exclusively in shell vials, and two were detected only in conventional tube cell cultures. For blood specimens only, the inoculum volume was not uniform in both systems in that 0.3 ml of a leukocyte suspension was introduced into each of three shell vials and 1 ml was introduced into a conventional tube cell culture. Perhaps the cytotoxicity of MRC-5 cells occurred in the conventional tube cell cultures because of the large specimen inoculum, which may have ultimately prevented the recognition of specific CPE caused by CMV. Conventional tube cell cultures should be inoculated concurrently with shell vials for blood specimens until better correlation is obtained. CMV was detected in 36 specimens from 23 patients by the shell vial assay but not by conventional tube cell culturing. However, we feel that these results were specific for CMV and that the shell vial assay actually was more J. CLIN. MICROBIOL. sensitive than conventional tube cell culturing, because 18 of the patients showed evidence of CMV infections in other specimens either by the recovery of CMV from conventional tube cell cultures or by serological seroconversion to CMV. Further, in our study 88 specimens yielded CMV in both cell culture systems. Moreover, there were no instances, with the exception of two blood specimens, in which CMV was recovered from conventional tube cell culture but was not detected by monoclonal antibody in shell vials. In addition, centrifugation has been demonstrated to enhance the detection of CMV, as compared with noncentrifuged samples (3, 6, 10). Also, monoclonal antibody combines with an early antigen of CMV and fluorescent conjugate to produce distinct and easily recognizable fluorescence confined to the regular outlined shape of the nucleus of the infected cell. We feel that the shell vial cell culture assay for the detection of CMV by a monoclonal antibody at 16 h postinfection is as specific as and more sensitive than the conventional tube cell culture assay for the diagnosis of CMV infections. LITERATURE CITED 1. Alford, C. A., Jr., P. D. Griffiths, S. Stagno, R. F. Pass, and R. J. Smith Congenital cytomegalovirus infection: diagnostic and prognostic significance of the detection of specific immunoglobulin M antibodies in cord serum. Pediatrics 69: Chou, S., and T. C. Merigan Rapid detection and quantitation of human cytomegalovirus in urine through DNA hybridization. N. Engl. J. Med. 308: Gleaves, C. A., T. F. Smith, E. A. Shuster, and G. R. Pearson Rapid detection of cytomegalovirus in MRC-5 cells inoculated with urine specimens by using low-speed centrifugation and monoclonal antibody to an early antigen. J. Clin. Microbiol. 19: Goldstein, L. C., J. McDougall, R. Hackman, J. D. Myers, E. D. Thomas, and R. C. Nowinski Monoclonal antibodies to cytomegalovirus: rapid identification of clinical isolates and preliminary use in diagnosis of cytomegalovirus pneumonia. Infect. Immun. 38: Ho, M Human cytomegalovirus infections in cells and tissues, p In W. B. Greenough III and T. C. Merigan (ed.), Cytomegalovirus: biology and infection. Plenum Publishing Corp., New York. 6. Hudson, J. B., V. Misra, and T. R. Mosmann Cytomegalovirus infectivity: analysis of the phenomenon of centrifugal enhancement of infectivity. Virology 72: Kim, K. S., V. J. Sapienza, M. J. Chen, and K. Wisniewski Production and characterization of monoclonal antibodies specific for a glycosylated polypeptide of human cytomegalovirus. J. Clin. Microbiol. 18: Macasaet, F. F., K. E. Holley, T. F. Smith, and T. F. Keys Cytomegalovirus studies of autopsy tissue. Il. Incidence of inclusion bodies and related pathologic data. Am. J. Clin. Pathol. 63: Myers, J. D., J. Leszczynski, J. A. Zaia, N. Flournoy, B. Newton, D. R. Snydman, G. G. Wright, M. J. Levin, and E. D. Thomas Prevention of cytomegalovirus infection by cytomegalovirus immune globulin after marrow transplantation. Ann. Intern. Med. 98: Osborn, J. E., and D. L. Walker Enhancement of infectivity of murine cytomegalovirus in vitro by centrifugal inoculation. J. Virol. 2: Pereira, L., M. Hoffman, D. Gallo, and N. Cremer Monoclonal antibodies to human cytomegalovirus: three surface membrane proteins with unique immunological and electrophoretic properties specify cross-reactive determinants. Infect. Immun. 36: Peterson, P. K., H. H. Balfour, Jr., S. C. Marker, D. S. Fryd, R. J. Howard, and R. L. Simmons Cytomegalovirus

5 VOL. 21, 1985 CELL CULTURING FOR DETECTION OF CYTOMEGALOVIRUS 221 disease in renal allograft recipients: a prospective study of the clinical features, risk factors and impact on renal transplantation. Medicine 59: Smith, M. J., R. W. I. Cooke, N. Hood, C. A. Hart, and P. L. Yap High-titre anti-cmv plasma in treatment of neonatal cytomegalovirus infection. Lancet i: Smith, T. F Viruses, p In J. A. Washington II (ed.), Laboratory procedures in clinical microbiology. Springer-Verlag New York, Inc., New York. 15. Smith, T. F., K. E. Holley, T. F. Keys, and F. F. Macasaet Cytomegalovirus studies of autopsy tissue. I. Virus isolation. Am. J. Clin. Pathol. 63: Stagno, S., R. F. Pass, D. W. Reynolds, M. A. Moore, A. J. Nahmias, and C. A. Alford Comparative study of diagnostic procedures for congenital cytomegalovirus infection. Pediatrics 65: Sullivan, J. L., and J. B. Hanshaw Human cytomegalovirus infections, p In R. Glaser and T. Gotlieb-Stematsky (ed.), Human herpes virus infection: clinical aspects. Marcel Dekker, Inc., New York.