In situ ELISA for the evaluation of antiviral compounds effective against human cytomegalovirus

Journal of Virological Methoak, 35 (1991) 207-215 0 1991 Elsevier Science Publishers B.V. / All rights reserved / 0166-0934/91/$03.50

207

VIRMET 01250

In situ ELISA for the evaluation of antiviral compounds effective against human cytomegalovirus Walter

A. Tatarowicz’,

Departments

Nell S. Lurain2 and Kenneth

D. Thompson’*2

of ‘Microbiology and ‘Pathology, Loyola University Medical Center, Maywood, Illinois, U.S.A.

(Accepted 9 July 1991)

Summary An in situ ELISA was developed as an improved procedure over the plaque reduction assay for antiviral susceptibility testing of HCMV. Unlike the plaque reduction assay, the ELISA can be completed at 4-5 days post-infection. The effective dose (ED& of ganciclovir (GCV), acyclovir (ACV), phosphonoacetic acid (PAA), or phosphonoformic acid (PFA), was determined using HCMV strain AD169. The resistance profiles of two laboratory-derived GCV-resistant mutants of HCMV strain AD169 and seven clinical isolates were determined using the ELISA. The ELISA results were confirmed by the plaque reduction assay. The EDso for GCV with the AD169 control ranged from 3.1 to 6.2 PM with a mean inhibitory concentration of 5.4 f 1.4 PM. Six of the clinical isolates were susceptible to GCV (ED50 = 3.1-6.2 PM). The seventh isolate had an EDso of 50 ,uM and was resistant to GCV. This ELISA assay is reproducible and relatively simple to perform. The ELISA endpoints are clearly determined and the assay works well with a variety of antiviral compounds. Human cytomegalovirus; resistance

ELISA;

Antiviral

susceptibility

assay;

Ganciclovir

Correspondence to: Kenneth D. Thompson, Dept. of Pathology, Loyola University Medical Center, 2160 South First Avenue, Maywood, IL 60153, U.S.A.

208

Introduction

Human cytomegalovirus (HCMV) has long been known to cause both congenital and perinatal infections. With the increase in the number of transplant recipients, HCMV has also become an important cause of serious post-transplant infections (Forbes, 1989). Additionally, HCMV is an important pathogen in AIDS patients, a disease that has dramatically escalated in the past few years (Tyms et al., 1989). Antiviral compounds have been developed to treat these infections and some have been found to be effective (Verheyden, 1988). The pyrophosphate analogue phosphonoformic acid (PFA) and the nucleoside analogue ganciclovir (GCV) have been shown to have both in vitro and in vivo activity against HCMV (Biron et al., 1985; Verheyden, 1988). However, we and others have isolated GCV resistant strains of HCMV (Erice et al., 1989). The development of resistance and the toxicity associated with GCV therapy indicate that there is a need for new antiviral compounds to treat HCMV infections. The standard method used for susceptibility testing of HCMV has been the plaque reduction assay. This assay requires 10 to 14 days to reach an endpoint, and the endpoint can be difficult to interpret in the case of abortive infection. With the increasing number of antiviral compounds becoming available for in vitro testing (Andre et al., 1991; Shigeta et al., 1991; Tolman et al., 1985), there is a need for an assay that is faster and that has a more objective interpretation of the endpoint. In situ ELISA techniques have been developed for the determination of antiviral susceptibility for herpes simplex virus (Andre et al., 1988; Rabalais et al., 1987), varicella zoster virus (Berkowitz and Levin, 1985), and respiratory syncytial virus (Kang and Pai, 1989). This report describes an ELISA method for the determination of susceptibility of HCMV to a variety of antiviral compounds. This assay may also be used to investigate possible mechanisms of resistance to these agents.

Materials and Methods Antiviral compounds

Ganciclovir (GCV) was kindly provided by Syntex coloration, Palo Alto, CA. Acyclovir (ACV) was obtained from Burroughs-Wellcome, Research Triangle Park, NC. Aphidicolin (APH), phosphonoacetic acid (PAA) and phosphonoformic acid (PFA), adenine 9-B-D-arabinofuranoside (AraA), and thymine-1-D-arabinofuranoside (AraT) were purchased from Sigma, St. Louis, MO. Cell cultures and virzlses

Maintenance

medium

(VMEM) consisted

of Eagle’s minimal

essential

209

medium supplemented with 2.5 pg/ml amphotericin B, 2 mM L-glutamine, 50 pg/ml of gentamicin, and 1% fetal calf serum (FCS). Growth medium consisted of the same components as VMEM except that it was supplemented with 5% fetal calf serum (FCS) and 5% NuSerum (Collaborative Research, Bedford, MA). For the plaque reduction assay, low passage human foreskin fibroblasts (HFF) were split and 4 x IO4 cells were added to each well of a 24-well plate. For the ELISA procedure, tissue culture treated 96-well plates were inoculated with 1 x lo4 HFF per well. The outer wells of the 96-well plates were not used in these assays but were filled with sterile water to prevent dehydration. Strain AD169 of HCMV was used throughout this study. Clinical isolates of HCMV were obtained from heart, kidney and bone marrow transplant patients from Loyola University Medical Center. These isolates were passaged 5 to 7 times until the virus became extracellular. The titers of stock viruses were determined by plaque assay (Wentworth and French, 1970). GCV-resistant mutants of strain AD169 were selected by exposing the stock to increasing concentrations of GCV. Each of the GCV-resistant strains of HCMV AD169 was plaque-purified three times. ELBA

HFF monolayers were incubated in 96-well plates at 37°C in 5% CO2 for two days until nearly confluent. The configuration of the microtiter plates consisted of six uninfected cell control wells, six virus infected control wells, and six wells per drug concentration (eight concentrations per plate). Dilutions of each virus were prepared in VMEM. The growth medium was removed from all wells and 50 ~1 of VMEM was added to the cell control wells and 50 ~1 virus inoculum was added to the remaining wells. The virus was allowed to adsorb for 2 h at 37°C in 5% COz. The virus suspension was removed and 200 ~1 of VMEM was added to the cell control and virus control wells. Increasing concentrations of the antiviral compound diluted in VMEM were added to the remaining wells. The plates were then incubated at 37°C in 5% COz. At specific times after inoculation, the plates were assayed using the in situ ELISA procedure. The fluid medium was removed from the wells of the mic otiter plate and 100 ~1 of blocking solution (0.5% BSA in phosphate bu $ ered saline, pH 7.2) was added to each well for 30 min. The cells were fixed by adding 100 ~1 of ethanol/acetone (95:5, v/v) to each well and the plates were placed at -20°C for 30 min. -Each well was washed 5 times with 200 ~1 of wash solution (PBS containing O.!?O/BSA and 0.05% Tween 20). 100 ~1 of mouse monoclonal antibody to a CMV late nuclear protein (DuPont, Wilmington, DE) diluted 1:200 in PBS containing 10% normal goat serum was added ‘to each well. The plate was incubated at 37°C for 1 h. The primary antibody solution was removed and the wells were washed five times as before. 100 ~1 of horseradish-peroxidase-conjugated goat anti-mouse IgG (Bio-Rad Laboratories, Richmond, CA) diluted 1:500 was added to the wells and the plates

210

incubated at 37°C for 2 h. The secondary antibody was removed and the wells were again washed five times. The substrate was 1 mg/ml of 2,2’-azino-bis(3ethyl-benzthiazoline-6-sulfonic acid) (AEBS, Sigma) in phosphate-citrate buffer, pH 4.8, containing 0.005% hydrogen peroxide. Alternatively, the substrate used was TMBlue (Transgenic Sciences, Worcester, MA). 100 ~1 of the substrate solution was added to each well and the plates were incubated at room temperature for approx. 15-20 min. The absorbance (OD) of each well was read against a distilled water blank using a Dynatech MR580 Microelisa Auto Reader set at a wavelength of 450 nm (AEBS) or 630 nm (TMBlue). The average OD for the six wells of the uninfected cell control, the virus control and each drug concentration was calculated. The percent change in OD was calculated as follows: (average sample OD-average cell control OD)/(average virus control OD-average cell control OD) x 100. The effective dose (EDSo) is defined as the concentration of antiviral agent which produces a 50% reduction of the OD of the colored substrate product. The actual values that are reported are the lowest concentrations tested resulting in 50% or greater reduction in OD. Plaque reduction assaJ HFF monolayers in 24-well plates were inoculated with virus at an MO1 of 5 x 1o-5 which yielded approx. 20-30 plaques per well. After a 2-h adsorption, the monolayers were overlaid with 1.0 ml of VMEM containing 0.3% agarose and varying concentrations of the antiviral compound. Each concentration was tested in quadruplicate. Control wells received only VMEM with agarose. The assay was read when the plaques in the controls were well developed, usually 10-14 days post-infection. The ED5s for the plaque reduction assay was defined as the concentration of drug which resulted in a 50% reduction in the number of plaques relative to control wells.

Results Quantitation

of HCMV

antigen by ELISA

Monolayers of HFF in 96-well microtiter plates were inoculated with increasing concentrations of HCMV as described in Materials and Methods. The plates were inspected daily for the extent of CPE and the ELISA was performed when the CPE reached 100% (4 +) in the wells containing the highest concentrations of virus (MO1 = 0.1). At the lowest concentration of virus (MO1 = 4 x 10e4), the CPE involved less than 25% of the monolayer. A low level of viral antigen was detected in monolayers infected with the lowest concentration of virus that exhibited the least CPE. The correlation between MO1 and the 0D450 is shown in Fig. 1. A linear relationship between MO1 and OD450 exists between an MO1 of 0.003 and 0.025. These inoculum

211

Fig. 1. Relationship of MO1 to the absorbance (OD) in the ELISA assay. Monolayers of HFF in 96well microtiter plates were inoculated with increasing concentrations of HCMV. The ELISA was performed when the CPE reached 100% (4+) in the wells inoculated at an MO1 of 0.1.

concentrations resulted in 25-50% CPE. At an MO1 of 0.05 or greater, the monolayer began to disintegrate, which resulted in a loss of viral antigen. This was reflected in a plateau of the OD 450readings. These results indicated that the optimum results were obtained when the CPE involved 25-50% of the monolayer (2-3 -I-). The ED50 of several antiviral compounds was determined using MOIs of 0.005, 0.012, 0.025, and 0.05 with HCMV strain AD169. The ELISA was performed when the CPE in the virus control wells was 2 to 3 + , which ranged from 4 to 7 days post-infection. When an MO1 of 0.012 was used, the ED5e value of HCMV strain AD169 as determined by the ELISA approximated the EDso value in the plaque reduction assay (Table 1). Therefore, in order to TABLE 1 Susceptibility

results using the ELISA assay

Assay and virus tested

EDso WW GCV

ACV

PAA

PFA

AraA

ELISA AD169 D6/3,‘la Dl/3/4”

6.2 200 200

200 150 250

45 90 90

150

loo

150 200

::

Plaque reduction AD169 D6/3/ 1 D1/3/4

6.2 200 200

150 150 150

180 180 180

300 300 300

aLaboratory-derived

GCV-resistant

mutants of strain AD169.

125 :II

AraT 1000 150 150 > 1000 300 100

APh 3.8 3.8 3.8

5 5

212

standardize the ELISA, an MO1 of 0.012 was chosen for use in further studies. The assay at this MO1 was performed at 4-5 days post-infection. Susceptibility

testing of HCMV

using the ELBA

In the plaque reduction assay, GCV-sensitive strains of HCMV produced many single large rounded cells with only rare infection of adjacent cells. This probably represents abortive infection of HCMV in cultures that contain concentrations of GCV just above the ED 50. These single cells present a problem for the plaque reduction assay because quantitation of plaques is very difficult and subjective. The ELISA overcomes this problem, since quantitation is based on production of a late viral antigen which requires active viral replication. The antiviral susceptibility protiles of two laboratory-derived GCV-resistant mutants were determined using the ELISA and the plaque reduction assay (Table 1). Both assays confirm that mutants D6/3/1 and D1/3/4 were resistant to GCV. These mutants were susceptible to PAA, PFA, but were hypersensitive to AraT and AraA. Seven clinical isolates of HCMV were tested for GCV susceptibility using the ELISA assay. Microtiter plates inoculated with strain AD169 were assayed in parallel with each clinical isolate tested. The EDs0 for GCV with the AD169 control ranged from 3.1 to 6.2 PM with a mean inhibitory concentration of 5.2 + 1.6 (Table 2). Six of the clinical isolates were GCV-susceptible when compared to the AD169 control while the seventh, H51856, was resistant (ED50 = 50 PM). This result was confirmed by the plaque reduction assay (data not shown). Fig. 2 shows linear regression analysis of ELISA results for two of these clinical isolates, strain AD169 and two GCV-resistant mutants. This analysis demonstrates that the assay is reliable (r > 0.9) and that resistant strains can easily be identified.

TABLE 2 GCV susceptibility Clinical isolate

S43096 H46893 S42651 W66730 M25527 H51853 H51856

testing of clinical isolates using the ELISA EDso OLM) Experimental

AD169 control

6.2 1.6 1.6 1.6 3.1 3.1 50

6.2 3.1 6.2 3.1 3.1 6.2 6.2

213

0.5

1.0

10.0

100

IGCVI /.JM Fig. 2. Linear regression analysis of ELBA results: two GCV-susceptible clinical isolates (......),(- - -), wild-type AD I69 (p), and two GCV-resistant mutants of ADl69(- - - - - -), (-.-,-.). The r value was greater than 0.9 for all five strains.

Discussion The in situ ELISA was developed to determine the susceptibility of HCMV to several antiviral compounds. The assay is reproducible, easy to perform and more rapid than the plaque reduction assay. The production of the late viral antigen is measured calorimetrically. This measurement is proportional to the actual amount of virus replication within each cell because the production of late antigen requires viral DNA replication (Shuster et al., 1985). The plaque reduction assay measures the ability of an infectious particle to produce additional infectious progeny, which in turn infect adjacent cells. However, GCV-susceptible strains of HCMV tend to form single enlarged cells in the presence of GCV without spread of the virus to adjacent cells. This likely represents abortive infection. Therefore, it becomes difficult to read the endpoint of the plaque reduction assay, because the amount of viral replication within single cells cannot be determined. In the case of GCV, the ELISA appears to be the more accurate assay for the measurement of inhibition of viral replication. The ELISA is more rapid than the standard plaque reduction assay, which usually requires up to two weeks incubation. Similar EDse values can be obtained with the ELISA in 4-5 days post-inoculation. The ELISA can also be read in less time, 3-4 days, if the MO1 of both test and controls is increased. The EDse values increase slightly with the shortened assay; however, this does not change the interpretation of the assay for either susceptible or resistant strains. A subsequent improvement in the ELISA has been the substitution of the substrate TMBlue for AEBS. The change in substrate does not affect the

214

ELISA assay results. In comparison to AEBS, TMBlue substrate requires no preparation and is non-mutagenic. In addition, the color change of this substrate occurs faster than that of AEBS. Other assays have been described for the determination of susceptibility of viruses to antiviral compounds. The neutral red dye uptake assay has been used for the measurement of susceptibility of HSV to antiviral compounds (McLaren et al., 1983). This indirect assay measures susceptibility to antiviral compounds by cell viability rather than the actual amount of viral replication. Nucleic acid hybridization assays have been reported for antiviral and susceptibility testing (Dankner et al., 1990; Gadler, 1983; Stanberry Myers, 1988). The advantage of these procedures is that they measure virus DNA replication. However, these procedures require the transfer of samples onto membranes and the use of radioactive probes. A rapid method to screen for GCV-resistance in clinical isolates of HCMV was reported by Telenti and Smith (1989). Shell vials containing MRC-5 cells are infected with HCMV in the presence of a single concentration of drug and incubated for 4 days. An early HCMV antigen is detected by an indirect fluorescent antibody procedure using a mouse monoclonal antibody. Although this method is rapid, the results are not quantitative. This method is primarily used to screen clinical HCMV isolates for resistance to antiviral compounds. Recently, an enzyme immunocytochemical staining assay to determine HCMV infectivity has been reported (Musiani et al., 1988). HCMV infected HFF in 24-well cell culture plates are stained with an alkaline phosphataseconjugated antibody to late viral antigen. The infected cells are counted visually. This assay could be adapted for measuring susceptibility to antiviral compounds. However, reading this assay for multiple drug dilutions would be cumbersome and subjective. The ELISA described in this report is a useful procedure for testing new antiviral compounds for anti-HCMV activity and to characterize drug-resistant strains of HCMV. It is more rapid than the plaque reduction assay and quantitation of virus replication can be measured objectively. Finally, the results from the ELISA reflect more accurately the replication characteristics of HCMV in the presence of various antiviral compounds.

References Andre, P.M., Narbonne, C.H., Donnio, P.Y., Ruffault, A. and Fauconnier, B. (1988) Evaluation of herpes simplex virus susceptibility to acyclovir using an enzyme-linked immunosorbent assay. Ann. Inst. Pasteur/Viral. 139, 185-195. Berkowitz, F.E. and Levin, M.J. (1985) Use of and enzyme-linked immunosorbent assay performed directly on fixed infected cell monolayers for evaluating drugs against varicella-zoster virus. Antimicrob. Agents Chemother. 28, 207-210. Biron, K.K., Stanat, SC., Sorrell, J.B., Fyfe, J.A., Keller, P.M., Lambe, C.U. and Nelson, D.J. (1985) Metabolic activation of the nucleoside analog 9-{[2-hydroxy-l-(hydroxymethyl)ethoxy]methyl}guanine in human diploid libroblasts infected with human cytomegalovirus. Proc. Natl. Acad. Sci. USA 82, 2473-2477.

215 Dankner, W.M., Scholl, D., Stanat, S.C., Martin, M., Sonke, R.L. and Spector, S.A. (1990) Rapid antiviral DNA-DNA hybridization assay for human cytomegalovirus. J. Virol. Methods 28, 293-298. Erice, A., Chou, S., Biron, K.K., Stanat, S.C., Balfour, H.H. and Jordan, MC. (1989) Progressive disease due to ganciclovir-resistant cytomegalovirus in immunocompromised patients. N. Eng. J. Med. 320, 289-293. Forbes, B.A. (1989) Acquisition of cytomegalovirus infection: an update. Microbial. Rev. 2, 204216. Gadler, H. (1983) Nucleic acid hybridization for measurement of effects of antiviral compounds on human cytomegalovirus DNA replication. Antimicrob. Agents Chemother. 24, 37&374. Kang, J. and Pai, C.H. (1989) In situ immunoassay for antiviral susceptibility testing of respiratory syncytial virus. Am. J. Clin. Pathol. 91, 323-326. McLaren, C., Ellis, M.N. and Hunter, G.A. (1983) A calorimetric assay for the measurement of the sensitivity of herpes simplex viruses to antiviral agents. Antiviral Res. 3, 223-234. Musiani, M., Zerbini, M., Gentilomi, G. and La Placa, M. (1988) Rapid quantitative assay of cytomegalovirus infectivity. J. Virol. Methods 20, 333-340. Rabalais, G.P., Leviln, M.J. and Berkowitz, F.E. (1987) Rapid herpes simplex virus susceptibility testing using an enzyme-linked immunosorbent assay performed in situ on fixed virus-infected monolayers. Antimicrob. Agents Chemother. 31, 946-948. Shigeta, S., Konno, K., Baba, M., Yokota, T. and de Klerq, E. (1991) Comparative inhibitory effects of nucleoside analogues on different clinical isolates of human cytomegalovirus in vitro. J. Infect. Dis. 163, 270-275. Shuster, E.A., Beneke, J.S., Tegtmeier, G.E., Pearson, G.R., Gleaves, C.A., Wold, A.D. and Smith, T.F. (1985) Monoclonal antibody for rapid laboratory detection of cytomegalovirus infections: characterization and diagnostic application. Mayo Clin. Proc. 60, 577-585. Stanberry, L.R. and Myers, M.G. (1988) Evaluation of varicella-zoster antiviral drugs by a nucleic acid hybridization assay. Antiviral Res. 9, 367-377. Telenti, A. and Smith, T.F. (1989) Screening with a shell vial assay for antiviral activity against cytomegalovirus. Diagn. Microbial. Infect. Dis. 12, 5-8. Tolman, R.L., Field, A.K., Karkas, J.D., Wagner, A.F., Germershausen, J., Crumpacker, C. and Scolnick, E.M. (1985) 2’-nor-cGMP: A seco cyclic nucleoside with powerful anti-DNA-viral activity. Biochem. Biophys. Res. Commun. 128, 1329-1335. Tyms, A.S., Taylor, D.L. and Parkin, J.M. (1989) Cytomegalovirus and the acquired immunodeficiency syndrome. J. Antimicrob. Chemother. 23, 89-105. Verheyden, J.P.H. (1988) Evolution of therapy for cytomegalovirus infection. Rev. Infect. Dis. 10, 477489. Wentworth, B.B. and French, L. (1970) Plaque assay of cytomegalovirus of human origin. Proc. Nat]. Acad. Sci. USA 135, 253-258.