Journal of Medical Virology 24275-282 (1988)

Detection of Cytomegalovirus in Clinical Specimens by Virus Isolation and by a Monoclonal Antibody Against the Early Nuclear Antigen Therese Popow-Kraupp and Christian Kunt

lnstitute of Virology, University of Vienna, Austria

A commercially available monoclonal antibody against the 72000 Dalton early nuclear protein (EA) of cytomegalovirus (CMV) strain AD169 was used in an indirect immunofluorescence staining procedure (IF) for rapid detection of CMV-infected cells in tissue cultures inoculated with clinical specimens (200 urines, 22 throat washings, 5 stools, 4 bronchoalveolar lavage fluids). The results obtained by this method were compared with those obtained by virus isolation with and without centrifugal enhancement of viral infectivity.

In 66 (28.6%) of the 231 samples, CMV was detected by at least one of the methods used. Of 59 specimens producing CMV-specific cytopathic effect (CPE) in tissue culture, 46 (78%) were also positive in the EA test 16 hours after inoculation. Seven CPE-negative samples were, however, positive in the EA test. Five (38%) of the false negative EA test results were due to CMV strains that did not react with the monoclonal antibody used.

Key words: cytomegalovim, vim isolation, IF, monoclonal antibody

INTRODUCTION

Cytomegalovirus (CMV) is frequently the cause of severe infections in patients under immunosuppressive therapy, in patients with the acquired immunodeficiency syndrome (AIDS), and in those who have undergone bone marrow and organ transplanta- tion [Duvall et al, 1966; Henson et al, 1972; Fryd et al, 1980; Betts, 1982; Drew et al, 1982; Macher e t al, 1983; Quinnan et al, 1984; Skinhaj et al, 19841. CMV predisposes the

Accepted for publication July 23, 1987.

Address reprint requests to: Therese Popow-Kraupp, Institute of Virology, University of Vienna, Kinderspital- gasse 15, 1090 Vienna, Austria.

0 1988 Alan R. Liss, Inc.

276 Popow-Kraupp and Kunz

patient, through its own immunosuppressive effect, to secondary infections with fungi and bacteria [Rand et al, 1978; Chatterjec et al, 19781. Patients with a CMV infection after transplantation have more frequent rejection reactions or graft versus host disease than those without infection, although the relationship between infection and rejection is still unclear [ Betts et al, 1982; Dunn et al, 1984; Lonnqvist et al, 1984; Meyers et al, 19861.

The rapid confirmation of suspected CMV infection in these risk groups is of particular importance in order to provide the basis for reduction or change of immunosup- pressive therapy or the administration of CMV hyperimmunoglobulin. However, the rapid diagnosis of CMV infection presents a considerable problem in these patients since the virus-specific antibody response develops slowly. Seroconversion and CMV-specific IgM antibodies may remain undetectable for several weeks, even by the most sensitive test systems [Rasmussen et al, 1982; Pass et al, 1983; Dylewsky et al, 19831. Cell culture virus isolation may take up to 5 weeks [Reynolds et al, 19791. It has been shown that the development of CMV-induced cytopathic effect (CPE) in tissue culture cells can be enhanced by centrifugation of the virus onto a fibroblast monolayer [Hudson et aI, 19761. The rapidity of detection of CMV-infected culture cells can be significantly increased further by the detection of early virus-induced proteins by means of immunofluorescence or peroxidase-antiperoxidase staining [Stagno et al, 1980; Goldstein et al, 1982; Griffiths et al, 1984; Hackman et al, 1985; Stirk and Griffiths, 19871. These methods did not find wide application in rapid diagnosis of CMV infection because high-titre antisera against early virus-induced proteins were not generally available.

Recently a monoclonal antibody against the 72000 Dalton early nuclear protein has become available, and its successful application for rapid diagnosis of CMV infection has been reported by several groups [Gleaves et al, 1984; Gleaves et al, 1985; Swenson and Kaplan, 1985; Martin and Smith, 19861.

In order to increase the efficiency and rapidity of CMV diagnosis for transplant patients, we introduced this method in our laboratory. Above all we were interested in the potential of EA staining for detecting clinically relevant CMV infections, and, therefore, we compared the results obtained with those of cell culture virus isolation with and without centrifugal enhancement of viral infectivity. MATERIALS AND METHODS Patients and Specimens

Clinical material for CMV detection was obtained from 137 patients, ranging in age from 4 days to 75 years. The majority of our patients suffered from the acquired immunodeficiency syndrome (AIDS), leukemia and malignancies (37 or 27%), or were bone marrow or organ transplant recipients (58 or 42%). Fourteen children (10% of our patients) had the clinical diagnosis of suspected congenital CMV infection and 17 (12.4%) patients were diagnosed as having infectious mononucleosis, hepatitis, or fever. For 24 (65%) patients with leukemia, malignancies, or AIDS and 18 (3 1%) transplant patients, additional information was available concerning the presence of acute disease, such as hepatitis or pneumonia, or clinical symptoms such as fever.

A total of 23 1 samples (200 urines, 22 throat washings, 5 stools, 4 bronchoalveolar lavage fluids) were available from these patients for CMV screening. The distribution of the samples among the patients was as follows: single urine specimens from 88 patients, with additional specimens from 4 of these (5 stools, 1 throat washing); from 28 patients, 2 urines with additional throat washings from 5; from 16 patients, more than 2 urines (3 from 10 patients, 4 from 5 patients, and 6 from 1 patient) with additional throat washings

EA Staining for Rapid Detection of CMV Isolates 277

from 5 and bronchoalveolar lavage fluids (BAL) from 2 patients. Single specimens of throat washings were obtained from 4 with an additional BAL from 1 patient, and a single specimen of BAL was obtained from 1 patient.

Detection of CMV in Clinical Specimens

Virus isolation. Specimens were filtered through a filter with a pore size of 0.45 pm and, if necessary, the pH was corrected to neutrality by the addition of sterile 1 M NaOH or 1 N HC1. Virus isolation was carried out by the standard method [Reynolds et al, 19791, whereby 2 ml of the clinical material was inoculated in a 50-ml Roux bottle with primary human foreskin fibroblasts (HFF). After an incubation period of 1 hour at 37°C in 5% CO,, the Roux bottles were washed with minimal essential medium with Earle’s salts (MEM-Earle’s) and 10 ml of maintenance medium (MEM-Earle’s containing 2% heat-inactivated fetal calf serum (FCS), L-glutamin 2 mM, Penicillin 200 IU/ml, Streptomycin 200 mcg/ml, Neomycin 100 mcg/ml, Fungizon 5 mcg/ml) were added to each of them. Infected cultures were kept at 37OC in a 5% CO, atmosphere for a period of 6 weeks and examined daily for cytopathic effect (CPE). This test is further referred to as test A.

In addition, 2 ml of the samples were also inoculated into each of two tubes (Sterilin 129 AX/I; Sterilin Limited, Feltham, England) containing a round coverslip with an HFF monolayer.

The tubes were then centrifuged at 37OC and 2,700 g for 1 hour and, after decanting the supernatant, filled with maintenance medium and incubated at 37OC. One of the two coverslips was removed I6 hours after inoculation, washed with phosphate-buffered saline (PBS), and fixed in acetone -2OOC for 20 minutes for indirect immunofluorescence staining (IF). The second of the two tubes was examined for the development of CPE for a period of 2 weeks. Virus isolation in the tubes after centrifugal enhancement of viral infectivity is further referred to as test B.

Indirect immunofluorescence staining (IF) for the detection of CMV-induced early nuclear protein (EA). Acetone-fixed coverslips were incubated with 25 p1 of a monoclonal antibody against the 72000 Dalton early nuclear protein (EA) of the CMV strain AD169 (Biotech Research Laboratories, Inc., Rockville, MD) at 37°C for 1 hour in a moist chamber. After washing with PBS, coverslips were overlaid with fluorescein isothiocya- nate (FITC) conjugated goat antimouse immunoglobulin (Jackson Immunores. Lab., Avondale, PA) diluted 1: lO in PBS containing 0.001% Naphthalene black as a counter- stain and incubated for 1 hour at 37OC. After washing in PBS for 15 minutes, the coverslips were air-dried, mounted on a slide with a neutral solution of polystyrene and plasticizers in xylene (De Pex, Serva, Heidelberg, FRG), and examined with a Leitz SM-Lux fluorescence microscope at a 325 x magnification. The microscope was fitted with an HBO 50 W mercury vapor lamp, a BP 450-490 exciter filter, and an LP 515 barrier filter. Acetone-fixed coverslips with CMV (AD 169) infected and uninfected HFF cells served as controls. This assay is hereafter referred to as test C.

In order to determine the optimal time for IF staining of CMV-induced EA, the following experiment was carried out. Tubes containing coverslips with HFF monolayers were infected as described above with the AD169 strain of CMV and with CMV isolates derived from 6 different patients and subsequently fixed and stained at 8, 16, 24, 48, and 72 hours postinoculation. Because identical results were obtained at the various time points and because most of our specimens arrived at noon or in the early afternoon, we decided to stain the coverslips 16 hours after specimen inoculation.

278 Popow-Kraupp and Kunz

Medlan: 12 d . Median: 7 d. 20 h

Ranqe: 1-34 d . Range: 1-14 d.

u n 2 59

Fig. 1. Detection of cytomegalovirus (CMV) in clinical specimens by celi culture isolation (test A), cell culture isolation after centrifugal enhancement of viral infectivity (test B), and by immunofluorescence staining of the early nuclear protein (test C).

T e s t A + T e s t B 5 9

Fig. 2. after centrifugal enhancement of viral infectivity (test B).

Number of cytomegalovirus isolates obtained by cell culture isolation (test A) and cell culture isolation

RESULTS

CMV was detected by at least one of the 3 methods in 66 (28.6%) of the 231 samples. The results are summarized in Figure 1. The CMV-positive specimens (60 urines, 5 throat washings, and 1 bronchoalveolar lavage fluid) were derived from 41 patients. From 18 of these, only a single clinical specimen was obtained (1 7 urines and 1 throat washing). Follow-up urine samples (1 up to 5 per patient) were available from 23 patients. The throat washings and the bronchoalveolar lavage fluid were obtained simultaneously with the urine specimens. CMV was detectable in all the follow-up samples (1 up to 5 per patient) from 11 patients. Fluctuations in virus excretion were observed in the remaining 12 patients.

Using conventional virus isolation (test A), 50 samples were CMV positive. The time from infection to appearance of CPE was 1-34 days, with a median of 12 days. After

EA Staining for Rapid Detection of CMV Isolates 279

l4 1 . . . . . . . A * .. $ 0 .. . . . . .. 0 . . ... . . a.

0 0 . . .. I I I

7 1 4 21 2 8 35

T e s t A / D a y s

7 4 2

Fig. 3. isolation (test A) and cell culture virus isolation after centrifugal enhancement of viral infectivity (test B).

Number of days required for the detection of cytomegalovirus in clinical specimens by cell culture virus

Test A + Test 6 + Test C : 66

Fig. 4. Comparison of the number of cytomegalovirus positive specimens obtained by cell culture virus isolation (test A), cell culture virus isolation after centrifugal enhancement of viral infectivity (test B), and by immunofluorescence staining of the early nuclear protein (test C).

centrifugal enhancement of viral infectivity (test B), CMV was found in 51 samples requiring 1-14 days, with a median of 7 days. Both methods together yielded 59 CMV-positive samples. CMV-EA (test C) was detectable in 53 specimens.

Comparison of Tests A and B (Fig. 2)

Of the 59 samples, 42 were positive in both tests. Discordant results were obtained with a total of 17 samples. Eight of these were positive only by test A (time range 16-25 days) and 9 only by test B (time range 7-13 days). Figure 3 shows a comparison of the number of days until the development of CPE in the 42 samples positive by both methods. As can be seen, centrifugal enhancement significantly reduced the time from inoculation to the appearance of CPE for most of the specimens.

280 Popow-Kraupp and Kunz

Comparison of Tests A and B with EA Staining (Test C ) (Fig. 4)

Of the 59 samples positive by tests A and B, 46 were also positive by test C. To assess the reason for the 13 false negative results in test C, the corresponding virus isolates were examined for their reactivity with the EA-specific monocIona1 antibody by immunofluor- escence. Eight of these samples were positive, whereas 5 isolates, which were derived from 3 different patients, were not stained by this monoclonal antibody. However, all 5 of these isolates reacted with a monoclonal antibody specific for the CMV late nuclear antigen (Biotech Research Laboratories, Inc., Roclcville, MD) as revealed by IF staining. In addition, CMV specific intranuclear inclusion bodies were found in all 5 isolates after IF staining with a polyclonal human CMV hyperimmuneserum.

Of the 7 samples negative by tests A and B and positive in test C, 3 were derived from 2 patients whose specimens induced a CPE 2 weeks before or 3 weeks after the

CPE-negative specimens. The sensitivity of the IF staining of the EA in comparison to the two methods of virus isolation was 78% with a specificity of 96%.

DISCUSSION

Virus isolation using primary human fibroblasts is at present the most sensitive technique for the detection of CMV in clinical specimens. A major drawback of this method, however, is the long time often required for the development of CPE, rendering this technique unsatisfactory for the diagnosis of clinically relevant CMV infection. The aim of the present study was to accelerate the identification of clinical CMV isolates in culture by centrifugal enhancement of viral infectivity and immunofluorescence staining of EA using a commercially available monoclonal antibody. This should provide a rapid, sensitive, and specific technique for the diagnosis of CMV infection.

The comparative analysis of 23 1 samples revealed that 78% of all specimens positive by virus isoIation (with and without centrifugal enhancement of viral infectivity) were also detected by IF staining of EA, I6 hours after inoculation. Similar results were obtained by GriEths et al I19841 using not only a single monoclonal antibody but a mixture of 7 monoclonal antibodies, each directed at a different CMV protein. The majority of the false negative EA test results seems to be due to very low virus titres in these specimens, which are also reflected by the long time required for the appearance of CPE (16-25 days in test A and 7-1 2 days in test B). Analogous observations were also made by Stirk and Griffiths [ 19871 reporting that in the majority of specimens where a false negative EA test result was seen, the viral cytopathic effect was detected in the third week of culture. In addition, a nonhomogeneous distribution of infectious virus in the samples may contribute to false negative results, which is also substantiated by the finding that all except one of the samples false negative by the EA test were found positive by only one of the two virus isolation procedures. The same phenomenon may also be responsible for those cases in which the virus could be detected by only one test system.

Using the same monoclonal antibody for EA detection by immunofluorescence or PAP staining, Gleaves et a1 [ 1984, 19851 reported a higher sensitivity of this rapid method compared with virus isolation. However, in these studies the development of CPE was observed only for a period of 2 weeks-compared to 6 weeks in our study; thus, probably omitting most samples containing only low concentrations of infectious virus.

There is evidence that in certain cases false negative EA test results are due to the presence of CMV strains inducing EA that is not reactive with the monoclonal antibody used in this study. This is based on the finding that virus isolates did not bind the

EA Staining for Rapid Detection of CMV Isolates 281

EA-specific monoclonal antibody, but were reactive with a polyclonal immuneserum as well as with a monoclonal antibody against a CMV late nuclear antigen. Consistent with our results, there are data indicating that the size of the immediate early nuclear protein varies among different CMV strains [Cameron and Preston, 198 1; Gibson, 19811.

The EA detection is technically easy to accomplish and represents a feasible procedure that does not require expensive laboratory equipment. Through the use of a monoclonal reagent, there is no background fluorescence, and cells containing EA are easily recognized. This assay, therefore, represents an efficient means of providing the clinician with a precise diagnosis of CMV infection the day after sample collection in 78% of all samples where virus can be detected by prolonged virus isolation procedures.

ACKNOWLEDGMENTS

We wish to thank Heide Dippe and Karin Festl for their excellent technical assistance, and Susanne Pfauser for typing the manuscript.

REFERENCES

Betts RF (1982): Cytomegalovirus infection in transplantant patients. In Melnick JL (ed): “Progress in Medical Virology,” Vol. 28, Basel: S. Karger, pp 4 6 6 4 .

Cameron JM, Preston CM (1 981): Comparison of the immediate early polypeptides of human cytomegalovirus isolates. Journal of General Virology 54:42 1424.

Chatterjee SN, Fiala M, Werner J, Stewart JA, Stacey B, Warner N (1978): Primary cytomegalwirus and opportunistic infections. Incidence in renal transplant recipients. Journal of the American Medical Association 240:244&2449.

Drew WL, Conant MA, Miner RC, Huang ES, Ziegler JL, Groundwater JR, Gullett JH, Volberding P, Abrams DI, Mintz L (1982): Cytomegalovirus and Kaposi’s sarcoma in young homosexual men. Lancet 2.1 25-1 27.

Dunn DL, Malas AJ, Fryd DS, Simmons RL, Najarian JS (1984): Association of concurrent herpes simplex virus and cytomegalovirus with detrimental effects after renal transplantation. Archives of Surgery

Duvall CP, Casazza AR, Grimley PM, Carbone PP, Rowe WP (1 966): Recovery of cytomegalovirus from adults with neoplastic disease. Annals of Internal Medicine 64:53 1-541.

Dylewsky J, Chou S, Merigan T C (1983): Absence of detectable IgM antibody during cytomegalovirus disease in patients with AIDS. New England Journal of Medicine 309:493497.

Fryd DS, Peterson PK, Ferguson RM, Simmons RL, Balfour HH, Najarian J S (1980): Cytomegalo-virus as a risk factor in renal transplantation. Transplantation 30:436439.

Gibson W (1981): Immediate early nuclear protein of human cytomegalovirus strain AD 169, Davis, and Towne differ in electrophoretic mobility. Virology 1 12:35&354.

Gleaves CA, Smith TF, Shuster EA, Pearson G (1984): Rapid detection of cytomegalovirus in MRC-5 cells inoculated with urine specimens by using low-speed centrifugation and monoclonal antibody to an early antigen. Journal of Clinical Microbiology 19:917-919.

Cleaves CA, Smith TF, Shuster EA, Pearson GR (1985): Comparison of standard tube and shell vial cell culture techniques for the detection of cytomegalo-virus in clinical specimens. Journal of Clinical Microbiology 2 1 :2 17-221,

Goldstein LC, McDougall J, Hackman R, Meyers JD, Donnall Thomas E, Nowinski RC (1982): Monoclonal antibodies to cytomegalovirus: Rapid identification of clinical isolates and preliminary use in diagnosis of cytomegalovirus pneumonia. Infection and Immunity 38:273-281.

Griffiths PD, Stirk PR, Ganczakowski M, Panjwani DD, Ball MG, Blacklock HA, Prentice HG (1984): Rapid diagnosis of cytomegalovirus infection in immunocompromised patients by detection of early antigen fluorescent foci. Lancet 2: 1242-1 245.

Hackman RC, Myerson D, Meyers JD, Shulman HM, Sale GE, Goldstein LC, Rastetter M, Flournoy N, Donne11 Thomas E ( 1985): Rapid diagnosis of cytomegaloviral pneumonia by tissue immunofluorescence with a murine monoclonal antibody. Journal of Infectious Diseases 151 :325-329.

119:812-817.

282 Popow-Krauppand Kunz

Henson D, Siege1 SE, Fuccillo DA, Mathew E, Levine AS (1972): Cytomegalovirus infections during acute childhood leukemia. Journal of Infectious Diseases 126:46948 1.

Hudson JB, Misra V, Mosmann TR (1976): Cytomegalovirus infectivity: Analysis of the phenomenon of centrifugal enhancement of infectivity. Virology 72:235-243.

Lonnqvist B, Ringden 0, Wahren B, Gahrton G, Lundgren G (1984): Cytomegalovirus infection associated with and preceding chronic graft-versus-host disease. Transplantation 38:465468.

Macher AM, Reichert CM, Straus S E (1983): Death in the AIDS patients: Role of CMV. New England Journal of Medicine 309:1454-1457.

Martin WJ, Smith T F (1986): Rapid detection of cytomegalovirus in bronchoalveolar lavage specimens by a monoclonal antibody method. Journal of Clinical Microbiology 23: 1 oOf%loO8.

Meyers J, Flournoy N, Donell Thomas E (1986): Risk factors for cytomegalovirus infection after human marrow transplantation. Journal of Infectious Diseases 153:478488.

Pass RF, Griffiths PD, August AM (1983): Antibody response to cytomegalovirus after renal transplantation: Comparison of patients with primary and recurrent infections. Journal of Infectious Diseases 147:4& 46.

Quinnan GV, Masur H, Rook AH, Armstrong G, Fredrick WR, Epstein J, Manischewitz JF, Macker AM, Jackson L, Ames J , Smith HA, Parker M, Pearson GR, Parrillo J, Mitchell C, Straus S E (1984): Herpes virus infections in the acquired immune deficiency syndrome. Journal of the American Medical Association 252:72-77.

Rand KH, Pollard RB, Merigan T C (1978): Increased pulmonary superinfections in cardiac transplant patients undergoing primary cytomegalovirus infection. New England Journal of Medicine 298:95 1-953.

Rasmussen L, Kelsall D, Nelson R, Carney W, Hirsch M, Winston D, Preiksaitis J, Merigan T C (1982): Virus-specific IgG and IgM antibodies in normal and immunocompromised subjects infected with cytomegalo-virus. Journal of Infectious Diseases 145: 191-199.

Reynolds DW, Stagno S, Alford CA (1 979): Laboratory diagnosis of cytomegalovirus infections. In Lennette EH and Schmidt NJ (eds): “Diagnostic Procedures for Viral, Rickettsia1 and Chlamydia1 Infections.” Washington: American Public Health Association Inc., pp 399439.

Skinhaj P, Anderson HK, Mailer J, Jacobsen N (1984) Cytomegalovirus infection after bone marrow transplantation: Relation of pneumonia to postgrafting immunosuppressive treatment. Journal of Medical Virology 14:91-99.

Stagno S, Pass RF, Reynolds DW, Moore MA, Nahmias AJ, Alford CA (1980): Comparative study of diagnostic procedures for congenital cytomegalovirus infection. Pediatrics 65:25 1-257.

Stirk PR, Griffiths PD (1987): Use of monoclonal antibodies for the diagnosis of cytomegalovirus infection by the detection of early antigen fluorescent foci (DEAFF) in cell culture. Journal of Medical Virology 2 1329-337.

Swenson PD, Kaplan MH (1985): Rapid detection of cytomegalovirus in cell culture by indirect immunoperoxi- dase staining with monoclonal antibody to an early nuclear antigen. Journal of Clinical Microbiology 21 :669473.