175

Journal of Immunological Methods, 137 (1991) 175-180 © 1991 Elsevier Science Publishers B.V. 0022-1759/91/$03.50 ADONIS 0022175991001077

JIM 05855

The detection of human cytomegalovirus immediate early antigen in peripheral blood leucocytes G. Bein 1, A. Bitsch ~, J. Hoyer 2 and H. Kirchner 1 1 Institute oflmmunology and Transfusion Medicine, and 2 Transplantation Unit, Unioersity of Liibeck, Medical School, Liibeck, F,R.G.

(Received 22 October 1990, accepted 14 November 1990)

Recently, Van der Bij et al. (1988) reported that active human cytomegalovirus (HCMV) infection could be diagnosed by the detection of HCMV immediate early antigen (IEA) directly in the peripheral blood leucocytes of renal transplant recipients. However, the indirect peroxidase technique used resulted in high background staining due to endogenous peroxidase activity and thus the detection of HCMV-IEA positive leucocytes, which are sometimes present in extremely low numbers, was not always reliable. In an attempt to solve this problem, we have evaluated the alkaline phosphatase-anti-alkaline phosphatase (APAAP) technique, immunogold-silver staining (IGSS), and several fixatives. Fixation with acetone : methanol 1 : 1 in conjunction with the APAAP technique proved to be the most successful method. In 155 blood samples obtained from 44 patients following renal transplantation and from three AIDS patients, the number of positive cells ranged between 1 and 700 out of 400,000 (median 2). In 23 samples from 11 patients (one AIDS patient) at least one positive cell was found. In this series there were no problems with the evaluation since strong positive signals were obtained without any background staining. We therefore recommend the use of this protocol for the rapid and reliable detection of HCMV-IEA in peripheral blood leucocytes. Key words: Cytomegalovirus, human; Immediate early antigen; Alkaline phosphatase-anti-alkaline phosphatase technique; Immunogold-silver staining

Introduction

Human cytomegalovirus (HCMV) is the single most important infectious agent influencing the

Correspondence to: G. Bein, Institute of Immunology and Transfusion Medicine, University of Liibeck, Ratzeburger A1lee 160, W-2400 Liibeck, F.R.G. Abbreoiations: APAAP, alkaline phosphatase-anti-alkaline phosphatase; HCMV, human cytomegalovirus; HEL, human embryonic lung fibroblasts; IEA, immediate early antigen; IGSS, immunogold-silver staining; PBL, peripheral blood leucocytes; PBS, phosphate-buffered saline; RT, room temperature; TBS, Tris-buffered saline.

clinical course of patients undergoing renal transplantation (see review by Van Son and The, 1989) contributing significantly to morbidity, organ loss and mortality. The increasing availability of antiviral agents with proven efficacy in HCMV infection, e.g., ganciclovir, has emphasized the need for a method by which patients with an active infection can be rapidly identified. Recently Van der Bij and coworkers (Van der Bij et al., 1988; 1989; Van den Berg et al., 1989) reported the diagnosis of active HCMV infection by the detection of HCMV immediate early antigen (IEA) in peripheral blood leucocytes (PBL). This antigen was detected very early during the course of an active

176 infection even before seroconversion, positive virus culture or overt disease. However, the indirect immunoperoxidase staining of nuclear HCMVIEA in blood leucocytes with monoclonal antibodies was found to be associated with some major technical drawbacks in our hands and we concluded that the method was not suitable for routine screening purposes. A major cause of concern was that the number of IEA-positive cells early in the course of infection may be extremely low not exceeding 1 out of 105 leucocytes. Moreover the endogenous peroxidase activity in blood cells cannot be inactivated by conventional methods, since the antigen is vulnerable to these procedures (Jiwa et al., 1989) and the identification of single positive cells in the presence of high background staining due to endogenous peroxidase is time consuming and not always reliable. On the other hand, for exact quantification, each 'positive' cell has to be examined under high magnification in order to distinguish microscopically nuclear IEA positive cells from endogenous cytoplasmic peroxidase positive cells. Revello et al. (1989a,b) proposed the application of the indirect immunofluorescent method including fluorescence nuclear counterstaining to overcome this problem. However, fading of the fluorescent dye may be a problem without the use of anti-fading reagents as recently described (Krenik et al., 1989). More recently Jiwa et al. (1989) reported an improved indirect peroxidase method for the detection of HCMV-IEA consisting of m e t h a n o l / a cetic acid fixation, postfixation after the first layer of monoclonal antibody with formaldehyde and glucose/glucose oxidase peroxidase inactivation. In our hands HCMV-IEA was not resistant to methanol/acetic acid fixation, neither in HCMV infected fibroblasts nor in leucocytes from antigen positive patients. The aim of this present study was to identify an immunocytochemical method which would permit the sensitive and reliable detection of an extremly low number of HCMV-IEA positive cells in peripheral blood leucocyte cytospin preparations. We tested several fixatives and compared the immunogold-silver staining (IGSS) method with the alkaline phosphatase-anti-alkaline phosphatase (APAAP) technique.

Materials and methods

Monoclonal antibodies The mouse monoclonal antibodies C10 and C l l (Van der Bij et al., 1988a) raised against HCMVIEA (both of IgG1 subclass) were obtained from Biotest (Dreieich, F.R.G.). Con trols The positive controls used consisted of HCMV strain AD 169 infected human embryonic lung fibroblasts (HEL) and PBL from an H C M V infected renal transplant recipient. As a negative control an IgG isotype matched irrelevant mouse monoclonal antibody (anti-human cytokeratin, Dakopatts, Hamburg, F.R.G.) was used diluted 1 / 1 0 (16/~g/ml). Sample preparation Leucocytes were separated using the method described by Van der Bij et al. (1988a) with minor modifications. 7 ml of citrate-anticoagulated venous blood were gently mixed with 2 ml of 6% poly(O-2-hydroxyethyl)starch MW 450,000 in 0.15 M NaCI (Plasmasteril, Fresenius, Bad Homburg, F.R.G.). After 30 rain incubation at 3 7 ° C the supernatant was collected and centrifuged at 200 × g for 8 min. The pellet was resuspended in phosphate-buffered saline (PBS) 0.01 M, p H 7.2 and washed twice (500 × g , 8 rain). 1 × 105 leucocytes were centrifuged on acetone cleaned glass slides using a cytocentrifuge (Cytospin-2, Shandon, Astmoor, U.K.) at 550 rpm for 5 min. Fixation The following fixation protocols were investigated: (a) 5% p-formaldehyde for 90 s at room temperature (RT) followed by washing in H20; (b) acetone for 10 min at 4 ° C; (c) methanol : acetic acid 20 : 1 for 15 min at RT; (d) acetone : methanol 1 : 1 for 90 s at 4 ° C ; and (e) buffered formol acetone (20 mg N a 2 H P O 4, 100 mg K H z P O 4, 45 ml acetone, 25 ml formaldehyde 37%, 30 ml H20 ) for 30 s at RT followed by washing in H20. Immunostaining The determination of optimal dilutions for all primary and secondary antibodies used in the

177 following immunostaining protocols was undertaken by checkerboard titration.

tions of the manufacturer (Janssen, Olen, Belgium).

Indirect peroxidase staining

Alkaline phosphatase-anti-alkaline phosphatase staining

All antibodies were diluted in PBS 0.01 M, pH 7.2 diluted 8/1 in HCMV seronegative inactivated (30 min, 56 o C) pooled human AB serum. Acetone fixed slides were rinsed briefly in PBS, incubated with 2"5/zl of the primary antibody (30 min, RT), washed three times in PBS, incubated with 25/~1 peroxidase-labelled rabbit anti-mouse IgG (Dianova, Hamburg, F.R.G.) diluted 1/100 (30 min, RT), washed three times in PBS, and subsequently incubated with 25 /xl peroxidase-labelled goat anti-rabbit IgG (Dianova, Hamburg, F.R.G.) diluted 1/100 (30 min, RT). After washing three times with PBS the substrate reaction was carried out with 3,3'-diaminobenzidine tetrahydrochloride as substrate according to standard protocols. Slides were mounted with Immunomount (Shandon, Astmoor, U.K.).

Immunogold-silver staining Immunogold-silver staining with immunogold probes was carried out with minor modifications according to the recommendations of the manufacturer (Janssen, Olen, Belgium). Acetone-fixed slides were incubated for 5 min with blocking solution (PBS 0.01 M, pH 7.4, 1% inactivated HCMV seronegative human AB serum, 0.8% bovine serum albumin). Monoclonal antibodies were diluted in blocking solution, secondary antibodies in blocking solution diluted 1/1 with inactivated HCMV seronegative human AB serum. Slides were incubated with 25 btl of the primary antibody (30 rain, RT), washed three times in PBS, incubated with 25 /~1 rabbit anti-mouse Immunoglobulin (Dakopatts, Hamburg, F.R.G.) diluted 1/20 (30 min, RT), washed three times in PBS, and subsequently incubated with 25/L1 1 nm gold-labelled goat anti-rabbit antibody (AuroProbeOne, Janssen, Olen, Belgium) diluted 1/50 (60 min, RT). After washing three times with PBS containing 0.1% gelatin, 0.8% bovine serum albumin, and twice in PBS the slides were incubated with 50 t~l 2% glutaraldehyde in PBS for 10 rain at RT and then washed twice with glass distilled water. The silver enhancement step with the IntenSE II kit was performed following the instruc-

All antibodies were diluted in Tris-buffered saline (TBS) 0.05 M, pH 7.6 diluted 8/1 in inactivated HCMV seronegative human AB serum. Slides were rinsed briefly in TBS and were then incubated with 25 #1 primary antibody (30 min, RT). After three washings with TBS the slides were incubated with 25 /zl rabbit anti-mouse immunoglobulin (Dakopatts, Hamburg, F.R.G.) diluted 1/50 (30 min, RT), washed three times in TBS and subsequently incubated with APAAP complex (Dakopatts, Hamburg, F.R.G.) diluted 1/50 (30 rain, RT). These two steps were repeated twice (each 15 rain, RT). The enzyme label was developed with naphthol AS-BI phosphate as the coupling agent and hexazotized new fuchsin as capture agent according to standard procedures (Cordell et al., 1984). Slides were mounted in Immunomount (Shandon, Astmoor, U.K.).

Results

Indirect peroxidase staining Using the above protocol, which was derived from the original method for the detection of HCMV-IEA positive leucocytes in cytospin preparations (Van der Bij et al., 1988), the nuclei of HCMV infected HEL cells and of granulocytes from patients with an active HCMV infection showed a dark-brown colour. However, due to the cytoplasmic endogenous peroxidase activity in granulocytes (particularly eosinophils) there was considerable background staining with extensive variation from sample to sample. At low magnification ( X 10 objective) the identification of nuclear staining was not possible. Only the examination of each 'positive' cell at high magnification ( x 40 or x l00 oil immersion objective) permitted discrimination between nuclear staining and cytoplasmic background staining. In many cases definite conclusions could not be drawn, because the background staining led to stronger signals than immunostaining.

178 development step of each slide after 20 min was not possible if large sample numbers were being processed. When screening large sample numbers, we were not able to obtain reproducible results.

Alkaline phosphatase-anti-alkaline technique

phosphatase

Using the APAAP method, nuclei of H C M V infected H E L cells and of granulocytes from patients with active H C M V infection showed a strong red staining virtually without any background (Fig. 1B). If the slides were not counterstained, H C M V - I E A positive cells could be easily recognized even by evaluating the full cytospin preparation at low magnification ( x 10 objective).

A

|

Fig. 1. A: three HCMV-1EA positive blood leucocytes. Immunogold-silver staining. Original magnification ×800. B: two HCMV-IEA positive blood leucocytes.APAAP procedure; counterstain hemalum. Original magnification × 800.

Immunogold-siloer staining Positive signals appeared as black granules in the nuclei of H C M V infected H E L cells or granulocytes of patients with an active H C M V infection (Fig. 1A). Without post-fixation with glutaraldehyde before the silver enhancement step, the immunostaining was considerably weaker and the morphological differentiation of the granulocytes was insufficient. The most critical step with regard to background staining was the silver development time with the IntenSE II silver enhancement kit. In our hands, 20 min at RT was optimum. Silver development for a prolonged time led to background staining which appeared as black spots, sometimes with crystalline structure. However, the time after which background staining appeared was not always reproducible. On the other hand, the exact termination of the silver

Comparison of the sensitivity between the staining procedures Serial two-fold dilutions of the primary monoclonal antibody were incubated with cytospin preparations from a patient with H C M V antigenemia and H C M V infected H E L cells. The highest titer that gave visible immunostaining was found to be 1/256 with the indirect peroxidase method and 1 / 5 1 2 with the A P A A P and IGSS method respectively. For further experiments the A P A A P method was used with the primary antibody diluted 1/10.

Fixation The strongest immunostaining, but associated with the worst morphology, was observed after fixation with acetone for 10 min at 4 ° C. Fixation with a c e t o n e : m e t h a n o l (90 s 4 ° C) yielded excellent immunostaining and sufficient preservation of all cell types in peripheral blood (Fig. 1B). Fixation with the other protocols ((a), (c), (e)) gave good morphology but unsatisfactory immunostaining.

Enumeration of HCMV-IEA positive leucocytes in cytospin preparations from clinical samples Using the APAAP protocol described above we investigated 155 blood samples from 44 patients after renal transplantation and three A I D S patients. 400,000 leucocytes (four cytospin preparations) were screened from each patient. The number of H C M V - I E A positive leucocytes ranged between 1 and 700 (median 2) in total. In 23 samples from 11 patients (one A I D S patient) at least one

179

positive cell was found. Almost invariably the immunostaining was restricted to the nuclei of granulocytes, whereas immunostaining in the nuclei of lymphocytes was extremely rare. In this series no background staining occurred, and morphological identification of cells carrying positive signals was always possible.

Discussion With the advent of antiviral drug therapy against HCMV there is a clinical need for a rapid and sensitive method for the diagnosis of active H C M V infection in renal transplant recipients. Traditional methods including serology, virus culture, detection of early antigen positive foci in culture, and DNA-hybridisation techniques are of limited value in determining clinical intervention (Jiwa et al., 1989). Viremia is supposed to be a hallmark of active HCMV infection (Saltzman et al., 1988). Thus, the recently introduced procedure for the detection of HCMV-IEA in peripheral blood leucocytes is one of the most promising methods for the rapid and sensitive diagnosis of active HCMV infection. The assay has previously been shown to have a sensitivity and specificity of at least 90% (Van den Berg et al., 1989) and can be performed within 6 h. As we have discussed above, the peroxidase method originally described is hampered by interpretation problems due to background staining. Under experimental conditions the IGSS method gave very clearcut results with a high signal-to-noise ratio. The exact timing of the silver development step could, however, not be maintained if large sample numbers were screened. For this reason we could not obtain uniform, comparable staining intensities and unwanted background staining often occurred. The APAAP procedure proved to be a highly successful method. In terms of the 'technical' sensitivity the identification of one single HCMVlEA positive cell against a background of 400,000 antigen negative leucocytes was always unequivocally possible in our series. This was due to the complete absence of background staining, and the well preserved morphology. Nevertheless it should be recognised that the question of whether the

epitopes recognized by the two monoclonal antibodies are conserved within all H C M V strains is one which remains to be elucidated. We would encourage the use of the APAAP procedure for rapid, sensitive and convenient detection of HCMV-IEA in peripheral blood leucocytes. Although we are aware that the biological significance of the presence of HCMV-IEA in granulocytes is still unclear, this assay seems to be one of the most reliable available tests for the rapid diagnosis of active H C M V infection.

Acknowledgements We acknowledge gratefully the advice of Drs. Arndt and Gerdes and the excellent technical assistance of Mrs. Karow and Mrs. Bentfeld. Parts of this work are included in the doctoral thesis of A. Bitsch.

References Cordell, J.L., Falini, B., Erber, W.N., Ghosh, A.K., Abdulaziz, Z., Macdonald, S., Pulford, K.A.F., Stein, H. and Mason, D.Y. (1984) Immunoenzymatic labeling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes). J. Histochem. Cytochem. 32, 219. Jiwa, N.M., Van de Rijke, F.M., Mulder, A., Van der Bij, W., The, T.H., Rothbarth, Ph.H., Velzing, J., Van der Ploeg, M. and Raap, A.K. (1989) An improved immunocytochemical method for the detection of human cytomegalovirus antigens in peripheral blood leucocytes. Histochemistry 91, 345. Krenik, K.D., Kephart, G.M., Offord, K.P., Dunnette, S.L. and Gleich, G.J. (1989) Comparison of antifading agents used in immunofluorescence. J. Immunol. Methods 117, 91. Revello, M.G., Zavattoni, M., Percivalle, E., Grossi, P. and Gerna, G. (1989a) Correlation between immunofluorescent detection of human cytomegalovirus immediate early antigens in polymorphonuclear leukocytes and viremia. J. Infect. Dis. 160, 159. Revello, M.G., Percivalle, E., Zavattoni, M., Parea, M., Grossi, P. and Gerna, G. (1989b) Detection of human cytomegalovirus immediate early antigen in leukocytes as a marker of viremia in immunocompromised patients. J. Med. Virol. 29, 88. Saltzman, R.L., Quirk, M.R. and Jordan, M.C. (1988) Disseminated cytomegalovirus infection. Molecular analysis of virus and leucocyte interactions in viremia. J. Clin. Invest. 81, 75, Van den Berg, A.P., Van der Bij, W., Van Son, W.J., Anema, J., Van der Giessen, M., Schirm, J., Tegzess, A.M. and The,

180 T.H. (1989) Cytomegalovirus antigenemia as a useful marker of symptomatic cytomegalovirus infection after renal transplantation - a report of 130 consecutive patients. Transplantation 48, 991. Van der Bij, W., Torensma, R., Van Son, W.J., Anema, J., Schirm, J., Tegzess, A.M. and The, T.H. (1988) Rapid immunodiagnosis of active cytomegalovirus infection by monoclonal antibody staining of blood leucocytes. J. Med. Virol. 25, 179.

Van der Bij, W., Van Son, W.J., Van der Berg, A.P.M., Tegzess, A.M., Torensma, R. and The, T.H. (1989) Cytomegalovirus (CMV) antigenemia: rapid diagnosis and relationship with CMV-associated clinical syndromes in renal allograft recipients. Transplant. Proc. 21, 2061. Van Son, W.J. and The, T.H. (1989) Cytomegalovirus infection after organ transplantation: an update with special emphasis on renal transplantation. Transplant. Int. 2, 147.