- Email: [email protected]
Evaluation and comparison of PCR and hybridization methods for rapid detection of cytomegalovirus in clinical samples
Journal of Virological Methods Journal of Virological Methods 62 (1996) 103- 1I I
Evaluation and comparison of PCR and hybridization methods for rapid detection of cytomegalovirus in clinical samples Rachel
L&y’,
Fatiha
Najioullah, Danikle Thouvenot, Michkle Aymard, Bruno Lina”
Sylvie Bosshard,
Laborutoire de Virologie, Fact&t de MPdecine Lyon Grunge Blanche. Domaine RockefcJller, h937.7 Lyon Cede.\- 08. Frunw
Accepted 3 June 1996
Abstract Rapid diagnosis of cytomegalovirus (CMV) infection may be obtained by molecular techniques. such as the polymerase chain reaction (PCR) and hybridization assays. The optimal technique to detect CMV in clinical samples was assessed. Two different PCR assays were used, targeting either the major immediate early 1 (MIE 1) or the HXLF 4 gene. The PCR products were detected by gel electrophoresis, dot blotting and an easy to use. rapid, solid phase hybridization assay, DNA enzyme immunoassay (DEIA). Standard tissue culture was also used. Cerebrospinal fluids (18) liver biopsies (9) from hepatic transplant recipients, amniotic fluids (7) from mothers with suspected peripartum infection, and samples (6) of miscellaneous origin (brain and fundus biopsy, pericardial and pleural fluid) were tested. Among the 40 samples, CMV was detected in 19 cases. Three were positive by both molecular techniques and tissue culture, 14 by molecular methods and 2 by culture. 16/19 or 9/19 CMV-positive samples were detected by PCR amplification of the HXLF 4 or MIE 1 gene, respectively and 14/16 HXLF 4-positive samples were detected using either dot-blot or DEIA, compared to 9/16 using gel electrophoresis. Thus, the most sensitive assays for the detection of CMV in clinical samples using the methods compared in the current study were PCR amplification of the HXLF 4 gene followed by dot-blot or DEIA hybridization. Kqwovds:
Human
cytomegalovirus;
Polymerase
chain reaction;
* Corresponding author: Tel.: + 33 78 777029; fax: + 33 78 014887; e-mail: [email protected] .fr ’ Laboratoire de Biologie de la Reproduction et du Dtveloppement, Hopital Edouard Herriot, 69487 Lyon Cedex 03, France. 0166.0934,‘96/$15.00
Rapid
detection;
Dot-blot;
DEIA
hybridization
1. Introduction Human cytomegalovirus (CMV) is a member of the herpesvirus group. Infections are common throughout the world. About 50680% of the
Copyright 0 1996 Elsevier Science B.V. All rights reserved /‘II SOl66-0934(96)02091-S
adult population in Central Europe or North America have serologic evidence of prior CMV infection (Onorato et al., 1985). CMV infection is now recognized as an important cause of morbidity and mortality in immunocompromised patients, including transplant recipients and HIV-infected individuals (Drew et al., 1984; Zipeto et al., 1993; Schmidt et al., 1993). CMV may also be responsible for clinically important infections in immunocompetent hosts. Transmission from mother to child may occur both in utero or during the peripartum phase (Huang et al., 1980; Dworsky et al., 1983) resulting in severe, disseminated infection in the fetus (Stagno and Whitley, 1985), or the development of congenital CMV syndrome (Pass et al., 1980). Antiviral angents such as Foscarnet and Ganciclovir are effective for the treatment of acute CMV infection. Thus, rapid and sensitive diagnostic procedures for CMV are needed. Cell culture, serology and histopathologic examination of infected tissues are currently the diagnostic methods of choice. However, culture results may take up to 14 days to become positive and the method is relatively insensitive. Serology is both insensitive and non-specific for acute infection, especially in immunocompromised patients. Invasive procedures are required to obtain biopsy specimens for histopathologic evaluation. CMV nucleic acid amplification by the polymerase chain reaction (PCR) holds promise as a more effective diagnostic modality. The use of nested PCR or hybridization with a specific probe allow an increase in sensitivity and specificity for the detection of CMV. However, nested PCR assays are subject to a high risk of contamination through DNA carry-over (Cimino et al., 1990; Porter-Jordan et al., 1990) and hybridization using a dotblot system is time consuming (Espy and Smith, 1995). The detection of PCR products using DNA enzyme immunoassay (DEIA) may provide the potential for rapid and efficient detection of PCR products (Sakrauski et al., 1994) but requires careful evaluation against other available molecular techniques such as the dot-blot assay. In the present study, two PCR assays targetting the MIE 1 and HXLF 4 gene regions of the CMV genome were evaluated for the detection of CMV
in clinical samples. The MIE region encodes for an immediate early protein (Mach et al., 1989; Stenberg et al., 1984) and the HXLF 4 region for a transmembrdne segment of a late protein (Weston and Barrell, 1986). In addition, gel electrophoresis, dot-blot and DEIA as PCR amplicon detection assays and standard cell culture were evaluated.
2. Material
and methods
2.1. Sumpk
specimens
Forty clinical specimens obtained from three different hospitals in Lyon were investigated retrospectively; 18 cerebrospinal fluids (CSF), 9 liver biopsies (LB), 7 amniotic fluids [AF] and 6 of diverse origins (brain and fundus biopsy, pleura1 and pericardial fluids). All samples were collected and stored at - 20°C for up to 2 months. All biopsy samples were collected in a sterile tube without fixation liquid. 2.2. DNA
extraction
A rapid DNA extraction procedure was used for liquid samples (CSF, AF, pleural and pericardial fluids). Briefly, 100 ,~l of the sample was mixed with 50 ,~l of extraction buffer (10 mM Tris-HCl (Sigma), pH 8.3; 50 mM KC1 (Sigma); 2.5 mM MgCl, (Sigma); 0.45% Tween 20 (Sigma) and 100 pgg/ml proteinase K (Boehringer Mannheim, Germany)). The mixture was vortexed for 10 s, then incubated for 45 min at 56°C followed by 10 min at 95°C. The mixture was centrifugated at IO 000 rpm for 1 min in an Eppendorf centrifuge. The supernatant was collected and directly used for amplification without further purification. Biopsy samples were crushed using a mortar and pestle and diluted in 1 ml of EMEM culture media (Sigma). This mixture (200 ~1) was then transferred in an Eppendorf tube and centrifuged for 10 min at 10 000 rpm in an Eppendorf microfuge. The pellet was resuspended in 500 ,~l of the lysis buffer (10 mM Tris-HCl; 10 mM NaCl (Sigma); 25 mM EDTA (Sigma); 1% SDS (Sigma)
and 1 mg/ml proteinase K) and incubated overnight at 37°C. DNA was extracted on 250 /11 of this sample using a phenol-chloroform procedure (Chomczynski and Sacchi, 1987) followed by precipitation for 1 h at - 80°C with 250 ~1 of S M NaCl in 2.5 vol of 100% ethanol. The DNA extract was pelleted by a brief centrifugation step at 10000 rpm, washed once with 0.5 ml of 70% ethanol, and resuspended in 50 ~1 of TE buffer (10 mM Tris base, 1 mM EDTA). 2.3. DNA ampl$icution Amplification of CMV DNA by PCR was carried out using two oligonucleotide primer sets. Their sequences were as follows: MIE 1 upstream primer, 5’ AGA CCT TCA TGC AGA TCT CC 3’, 20372057, and MIE 1 downstream primer, 5’ GGT GCT CAC GCA CAT TGA TC 3’, 2280.2300 (PCRa; Zipeto et al., 1990); HXLF 4 upstream, 5’ CGG ATC AAC ATA AGG ACT TTT CAC AC 3’, 11 755-11780, and HXLF 3 downstream, 5’ GAA TAC AGA CAC TTA GAG CTC GGG GT 3’, 11 895- 11920 (PCRb; Weston and Barrell, 1986). PCRa and PCRb assays were performed in a 100 111reaction mixture containing 50 mM KCI; 10 mM Tris-HCl (pH 8.3); 1.5 mM M&l,; 10 ,ul of extraction product and 0.5 U of Taq DNA polymerase (Promega). The PCRa mix reaction also contained deoxynucleoside triphosphates (250 /IM each) and 50 pM of each oligonucleotide primer. The PCRb mix reaction contained adenine, cytosine and guanine deoxynucleotide triphosphates (200 PM each); uridine deoxynucleotide triphosphate (400 PM); 40 pM of each oligonucleotide primer; and Uracil DNA Glycosylase (UNG, 1 U; Gibco BRL) was added to prevent DNA carryover contamination (Long0 et al.. 1990). Amplification using PCRa and PCRb involved 50 cycles of denaturation at 94°C (1 min), annealing at 55°C (1 min) and extension at 72°C (2 min) or 35 cycles of denaturation at 94°C (20 s), annealing at 58°C (20 s) and extension at 72°C (20 s), respectively. In addition, PCRb used a hot start (94°C for 5 min) and a final hold temperature at 72°C for 10 min to stop the UNG activity.
Each PCR run included a positive ( 10 ~1 of strain AD 169 at lo4 TICD,,,) and a negative control (DEPC water). To avoid contamination by PCR product carryover, samples were prepared in a separate laboratory and recommended precautions (aliquoted reagents, positive displacement pipettes) were followed. 2.4. Detection
oj ump@ied
DNA
2.4.1. Gel electroplzoresis
PCR products (10 111) were run on a 2% agarose gel (Sea Kern GTG agarose) in TBE buffer, 0.089 M Tris borate, pH 8.3; EDTA 2 mM (Sigma) containing 10 ,ug/ml of ethidium bromide at 200 V for 20 min. A molecular size marker was run in each gel (DNA size marker V, Boehringer Mannheim). The gels were photographed with a Polaroid camera (MP4 + ) under UV light. The size of PCR products were 166 bp for PCRb and 223 bp for PCRa (Fig. 1). 2.4.2. DE/A assu_t The DEIA assay is based on the immobilization of a capture probe/amplicon hybrid on a solid
230 160
Fig. I. Ethidium bromide staining of a 2”/;1agarose gel showing samples positive or negative for CMV DNA. Lane I. molecular weight marker (DNA V); lane 2, PCRb positive control AD 169; lane 3. PCRb positive sample; lane 4, 5 and 8. negative controls: lane 6. PCRa positive control AD 169: lane 7. PCRa positive sample.
106
R. L&y
et al. / Journal of’ Virological
phase using a biotinlavidin bridge. DNA duplexes are detected by an anti-DNA monoclonal antibody that specifically reacts with double-stranded DNA (dsDNA). The specific dsDNA antibody complexes can be visualized by a peroxydase-conjugated anti-mouse antibody and a chromogen/ susbstrate mixture. The detection of amplified CMV DNA segments with DEIA was carried out according to the recommendations of the manufacturer (Sorin Biomedica, Italy). Briefly, streptavidin-coated microtiter plates were incubated with 100 ng of the 5’ biotinylated capture probe overnight at 4°C. After washing four times with phosphate-buffered saline (PBS), the microtiter plate was ready for immediate use or storage for up to 1 month. The procedure was as follows: after denaturdtion at 96°C for 10 min, 20 ~1 of each PCR product was incubated for 1 h at 50°C in one well containing 100 ~1 of hybridization buffer. After each step, the wells were washed five times with the suitable washing buffer. The diluted anti-dsDNA antibody was added and incubated for 60 min at room temperature. Bound anti-dsDNA antibody was detected by adding 100 ~1 of horseradish peroxidase-labeled rabbit anti-mouse IgG antibody. After a last incubation step of 1 h, 100 ~1 of chromogen-substrate mixture was added, and the calorimetric reaction was performed for 30 min. The reaction was stopped with 200 ~1 of 1 M sulfuric acid, and the absorbance was measured at an O.D. of 450 nm against a reference of 630 nm. The assay .;~as performed using two different coated probes: a 5’ CTA GTG TGA TGC TGG CCA AGC GGC CTC TGA 3’, 30-mer MIE 1 probe for PCRa (Zipeto et al., 1990) and a 5’ CAG GAC TGT GTA CCG TGT ATC CAG CGC TTA CTA TC 3’, 35-mer HXLF 4 probe for PCRb (Weston and Barrell, 1986). In order to determine the sensitivity of the assay, serial lo-fold dilutions of a 104TICD,,/ 100 ~1 suspension of the reference strain AD 169 were amplified by PCRa and PCRb and detected by DEIA. The cut-off value was set at 0.5 O.D. at 450 nm with an equivocal zone of &30%, as recommended by the manufacturer.
Methock
62 (1996) IO.?- 1 I I
2nd 1
2
3
4
5
6
2
1st
3 4
5
6
I
Fig. 2. Dot-blot analysis of PCRb products using the tailed oligonucleotide probe and a chemiluminescence detection system: (I, I), positive control; (1,2),negative control; (1,3), (2,2). (3,2), (2,6), (3,6), positive samples; (6,5), ‘indeterminant sample’.
2.4.3. Dot-blot hybridization Hybond N + membranes (Ammersham) were prehybridized for 1 h at 37°C in 20 ml of the hybridization buffer (5 x SSC, 50% formamide, 2% blocking reagent, 0.1% sarcosyl and 0.02% SDS). Hybridization (Fig. 2) with the tailed oligonucleotide probe was carried out overnight with the same buffer and at the same temperature. The tailed probe was labeled with digoxigenindUTP according to the manufacturer’s protocol (DIG Oligonucleotide tailing kit, Boehringer Mannheim). The tailing efficiency was checked by direct immunological detection (DIG Nucleic Acid Detection Kit, Boehringer Mannheim). The membranes were then washed twice (2 x SSC, 0.01% SDS) and the immunological detection of bound probes was performed using a chemiluminescence detection system (DIG Luminescent Detection Kit, Boehringer Mannheim; Zipeto et al., 1993; Wood et al., 1994). 2.5. PCR inhibitor detection The presence of amplification inhibitors was evaluated. All CMV-negative DNA samples were spiked with 10 ~1 of DNA isolated from reference
R. L&v)’ et al. (/IJournal
of Virological
strain AD 169 or diluted 1:5 (fluid samples) or 1:25 and 1:50 (biopsy specimens) followed by amplification and the detection methods described above.
Clinical specimens were inoculated onto human embryonic fibroblast cell cultures (MRCS) using a centrifugation step as described elsewhere (Swenson and Kaplan, 1985). Virus detection and identification were performed at day 2 post-infection by an immunoperoxidase (IP) assay using a commercially available specific monoclonal antibody El3 (Mazeron et al.. 1992). Monolayers were observed daily for 21 days for cytopathic effect on IP negative samples.
3. Results Sensitivity of the DEIA assay. The lowest detection limit of PCR amplicons was a 10 ~ ’ dilution of AD 169 (0.1 TCID,,/lOO ~1) using PCRa primers versus 10 ’ dilution (0.01 TCID,,),‘lOO ,uI) for PCRb primers. The mean O.D. value of the positive specimens was 1.687 for PCRa and 1.85 I for PCRb. The positive controls of the two PCR assays had an O.D. value 2 2.0. ?. 1. Conzparison
of’ the detection
methods
CMV was detected in 19 of the 40 samples by one or more detection techniques (Table 1). Tissue culture had the lowest sensitivity, with only 5’19 CMV-positive samples detected using this process. Among the 27 fluid samples evaluated, CMV was not detected by cell culture in any of the samples and 10 were positive using PCR and subsequent hybridization. The sensitivity of cell culture was improved when evaluating tissue samples: 5:’13 were culture-positive compared to 7/ 13 using molecular methods. Interestingly, two tissue samples were positive by cell culture but negative by molecular methods. Molecular detection methods remained negative after spiking the negative samples with control CMV DNA.
Methods
62 (1996)
103
III
107
Using molecular techniques, 17/19 CMV positive samples were detected. PCRb was the most sensitive amplification method, 16/ 17 samples with detectable CMV by molecular methods were positive using this amplicon compared to 9/17 using PCRa. Gel electrophoresis and DEIA were equivalent as detection methods for PCRa amplicons. All 9 PCRa-positive samples were detected using both methods. However, dot-blot and DEIA were superior to gel electrophoresis for detection of PCRb amplicons; 14116 PCRb-positive samples were detected by either dot-blot or DEIA compared to 91’16 using gel electrophoresis.
4. Discussion CMV may cause serious disease in both immunocompetent and immunocompromised patients. Although cell culture. serology and histopathologic examination are often the currently used methods for the diagnosis of acute CMV infection, these techniques are limited by several factors. The PCR assay offers promise for routine detection of active CMV infections. In order to be adopted by clinical laboratories, PCR assays must be sensitive, specific, rapid, easy to perform, reproducible and cost-effective. The techniques described in the current study full% these criteria. PCR was shown to be more sensitive than conventional cell culture since among the 19 positive patients, only 5 were detected by cell culture versus 17 by PCR. The specificity was confirmed by attempting to detect CMV DNA in isolates that were culture-positive for other herpesviruses or in samples obtained from seronegative patients. The results were all negative (data not shown). Furthermore, the clinical presentation and/or the results of virological investigations undertaken on samples collected later or from other sites supported the diagnosis of acute CMV infection in the 14 patients with culture-negative, PCR-positive samples. Amniotic fluid was recovered from a mother bearing a fetus with intrauterine growth retardation and hydrocephalus: 4 liver biopsy specimens were collected from hepatic transplant patients that were
108
Table I List of clinical
samples
investigated
by the different
techniques _
Type of samples
PCRa . Gel
AF” AF AF AF AF AF AF CSFb CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF
Gel
+ _
_ +(a)
_ _ _ +(a) _ _ _ _ _ + + _ + + _ _
Dot-blot
DEIA
_ _ + _ _
_ _ _ _ + -
+ _ _ _ _ _ + _ _ + + _
+ _ _ _ _ + +
_ + _ + _ +
_ _
+(a) _ _
+ + + + + _ _
+ +(a) + + + _ -
_ _
_
_ _
-
_ _
_
+(c)
+ Cc)
CMV
_
-
-
_
+ CC)
+(c)
+ Cc)
+(c)
CMV CMV
_
_ _
+(b) _ _
+(b) _
_ _ _ _ -
_ -
_ -
-
+ Cc) +(c)
-
-
_
+(b) _
+(b) -
+(b) -
+(b) -
+(b) _
CMV CMV
i-(c)
+
+(c)
-
_ _
_ -
_ _
_ _
-
-
Fundus Fundus
_ + Cc)
Pericardial fluid Pleural fluid
DEIA
Culture
-
Brain Brain
LB’ LB LB LB LB LB LB LB LB
PCRb
_
PCRa and PCRb procedures and primers are described in Materials and methods. Samples were evaluated at different dilutions: Fluid samples were tested undiluted and at a I:5 dilution (f. positive undiluted; +(a), positive at I:5 dilution). Biopsy specimens were tested undiluted and at I:25 and I:50 dilutions (+, positive undiluted; +(b), positive at I:25 dilution; C(c), positive at I:50 dilution). CMV, positive cell culture of CMV; -. negative culture or test.“Amniotic fluid. %Zerebrospinal fluid. ‘Liver biopsy.
R. LPvy et al. /Journal of’ Virological Methods 62 (1996) 103-l II
seronegative for CMV preoperatively and had either postoperative seroconversion or positive cell culture results from subsequent liver or blood samples. CSF was collected from three HIV-infected patients with encephalitis; one lung and one bone marrow transplant recipient with CMV isolated by cell culture from blood samples one month later; one multitransfused patient with CMV later isolated from the blood; and from a 6-year-old boy with a mononucleosis-like syndrome and blood culture positive for CMV. Clinical information was not available for the two other patients with PCR-positive, culture-negative CSF specimens. The specificity of the PCR assay for the detection of active CMV infection has also been discussed elsewhere (Fillet et al., 1993; Weston and Barrell, 1986; Zipeto et al., 1990). The detection of CMV using PCR and DEIA could be accomplished in 9 h using simple procedures. The use of bulk reagents and limited technician time renders the test cost-effective. PCR appears to be especially important for the diagnosis of CMV-infected fluids. In the present study, all 27 fluid samples evaluated by cell culture were negative, while molecular techniques gave positive results in 10. These results confirm the necessity of developing a more sensitive test than cell culture for fluid samples as discussed by other investigators (Fillet et al., 1993; Mazeron and Alain, 1993; Smith and Dunstan, 1993). However, PCR failed to detect CMV DNA in two of the biopsy samples that had virus isolated in cell culture. As previously described by other authors, these biopsy samples may possess Tuq DNA polymerase inhibitors and/or DNAases (Gass et al., 1993). The results of this study have confirmed these findings, CMV DNA was added to the DNA extracts from these biopsy specimens prior to amplication. They remained negative following PCR and hybridization. Furthermore, CMV was only detected by PCR in biopsy samples that had been diluted prior to amplification. Two PCR assays amplifying different regions of the CMV genome were compared. MIE 1 is the region transcribed in greatest abundance during early infection with CMV. Intermediate Early (IE) gene expression occurs independently of de novo protein synthesis and IE gene products can be detected within l-2 h after infection (Rice et al.,
109
1984). The activation of IE 1 gene expression is one of the first steps of transcriptional interaction between the virus and host cell. Proteins expressed by the HXLF 4 region exhibit close homology with a class of membrane proteins. Markedly hydrophobic sequences close to the C terminus could correspond to transmembrane regions (Tomita and Marchesi, 1975). In addition, all members of the HXLF protein group (HXLF 1, 2, 3 and 4) have a putative signal sequence for membrane insertion (Weston and Barrell, 1986). These differences in both genome position and function may be reflected in different sensitivities for PCR detection of CMV infection in clinical samples (Rice et al., 1984). PCR using the HXLF region of the CMV genome was more sensitive than amplification of the MIE region. This finding was observed both in comparative serial dilution assay (HXLF more sensitive by 1 log,,) and in the evaluation of clinical samples (16 samples positive using HXLF versus 9 using MIE 1). Using HXLF amplicons, dot-blot hybridization and DEIA were equivalent in their detection capabilities. Both were more sensitive than gel electrophoresis. However, DEIA is prefered over dot-blotting because it was easier to perform, less time-consuming and the results were consistently unequivocal. On the other hand, the dot-blot assay required frequent troubleshooting. Difficulties were encountered at three steps in the process: (i) failure of DNA transfer to the nylon membrane, (ii) lack of effective hybridization with the probes, and (iii) difficulty in detection due to high background New hybridization techniques for the detection of PCR amplified DNA, such as DEIA, have considerable potential use in routine diagnostic laboratories since they are characterized by easy handling, rapid results and high sensitivity. DEIA also appears to be superior to conventional techniques such as dot-blot hybridization due to potential automatization and less time-consuming methodology.
5. Conclusion The region
PCRb system amplifying the of the CMV genome followed
HXLF 4 by DEIA
110
R. LCvy et al. /Journal
of’ Virological
hybridization has proven to be a very-‘sensitive method for the diagnosis of CMV infection. The method is efficient, fast and practical, providing potential for use in routine diagnostic clinical laboratories. However, the presence of Tuq DNA polymerase inhibitors in tissues may limit its usefulness for these samples.
Methods
Longo, M.C., Berninger, M.S. and Hartley, J.L. (1990) Use of uracil DNA glycosylase to control carry-over contamination
in polymerase
We are grateful to Sorin Biomedica for providing DEIA kits. Thanks to Marc Carteron and Olivia RCny for their excellent technical assistance, and to Darryl See for careful reading of the manuscript.
chain
reactions.
Gene 93, 125- 128.
Mach, M., Stamminger, T. and Jahn G. (1989) Human c’ytomegalovirus: Recent aspects from molecular biology. J. Gen. Virol. 70, 3117-3146. Mazeron,
M.C.
and
tomegalovirus nosis. Pathol. Mazeron,
Acknowledgements
62 (1996) 10% I II
Alain-Albertini,
S. (1993)
infection: new methods Biol., 41, 487-494.
M.C., Jahn,
G. and Platcher
Human
S.K.
I.M., Morens, (1985)
recommendations Dis. 7. 479-497. Pass,
R.F.,
sults
S., Myers,
of symptomatic
of long
Martone,
term
infections:
and control.
G.J.
Rev. Infect.
and Alford,
congenital
longitudinal
recogimmedi-
W.J. and Stansfield,
of cytomegaloviral
for prevention
Stagno,
Outcome
D.M.,
Epidemiology
diag-
B. (1992) Monoclonal
Antibody El3 (M810) to human cytomegalovirus nize an epitope encoded by exon 2 of the major ate early gene. J. Gen. Virol. 73, 2699 -2703. Onorato,
cy-
for virological
C.A. (1980)
CMV
follow-up.
infection:
re-
Pediatrics
66,
758-762.
References Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidium thiocyanate phenol chloroform extraction. Anal. Biochem. 162, 156- 159. Cimino, G.D., Metchette, K.C., Tessman, J.W., Hearst, J.E. and Isaaca ST. (1990) Post PCR sterilization: a method to control carry-over contamination for the polymerase chain reaction. Nucleic Acids Res. 19, 99-107. Drew, W.L., Sweet, E.S., Miner, R.C. and Mocarski, E.S. (1984) Multiple infections by cytomegalovirus in patients with acquired immunodeficiency syndrome: documentation by Southern blot hybridization. J. Infect. Dis. 6, 952-953. Dworsky, M.E., Yow, M., Stagno, S., Pass, R.F. and Alford, C.A. (1983) Cytomegalovirus infection of breast milk and transmission in infancy. Pediatrics 72, 295-299. Espy, M.J. and Smith, T.F. (1995) Comparison of SHARP signal system and Southern blot hybridization analysis for detection of cytomegalovirus in clinical specimens by PCR. J. Clin. Microbial. 33, 3028-3030. Fillet, A.M., Katlama, C., Visse, B., Camilleri, S., Rogeaux, 0. and Huraux, J.M. (1993) Human CMV infection of the CNS: concordance between PCR detection in CSF and pathological examination. AIDS, 7, 1016-1018. Gass, P., Kiessling, M., SchBfer, P., Mester, C., Schmitt, H.P. and Kiihn, J.E. (1993) Detection of human cytomegalovirus DNA in paraffin sections of human brain by polymerase chain reaction and the occurrence of false negative results. J. Neural. Neurosurg. Psychiatry 56, 211-214. Huang, E.S., Alford, C.A., Reynolds, D.W., Stagno, S. and Pass R.F. (1980) Molecular epidemiology of cytomegalovirus infections in women and their infants. N. Engl. J. Med. 303, 958-962.
Porter-Jordan, K., Rosenberg, El. and Keiser, W.F. (1990) Nested polymerase chain reaction assay for the detection of cytomegalovirus contamination
overcomes
false
with fragmented
positives
DNA.
caused
by
J. Med. Virol.
30,
85.-91. Rice, C.P.A., Schrier, R.O. and Oldstone, M.B.A. Cytomegalovirus infects human lymphocytes and cytes:
virus
gene
products.
expression Proc.
is restricted Natl.
Acad.
(1984) mono-
to immediate Sci. USA.
early
81, 6134
6138. Sakrauski, Doerr,
A.. Weber, B., Kessler, H.W. (1994) Comparison
H.H., Pierer, K. and of two hybridization
assays for the rapid detection of PCR amplified genome sequences from cerebrospinal fluid. J. Methods 50, l75- 184. Schmidt, C.A.. Oettle, H., Neuhaus, stration
of cytomegalovirus
after liver transplantation. Smith, K.L. and Dunstan,
HSV Virol.
P. et al. (1993) Demon-
by polymerase
chain
reaction
Transplantation 56, 872-874. R.A. (1993). PCR detection of
cytomegalovirus: a review. Br. J. Haematol. Stagno, S. and Whitley, R.J. (1985) Herpesvirus
84, 187.. 190. infections of
pregnancy. Part I: Cytomegalovirus and Epstein-Barr virus infections. N. Engl. J. Med. 313, 1270-1274. Stenberg,
R.M..
Structural
Thomsen,
analysis
D.R.
of the major
human cytomegalovirus. Swenson, P.D. and Kaplan,
and
Stinski,
immediate
M.F. early
(1984) gene of
J. Virol. 49, 190- 199. M.H. (1985). Rapid detection
of
cytomegalovirus in cell culture by indirect immunoperoxydase staining with monoclonal antibody to an early nuclear antigen. J. Clin. Microbial. 21, 669-673. Tomita, M. and Marchesi, V.T. (1975) Amino-acid sequence and oligosaccharide attachment sites of human erythrocyte glycophorin. Proc. Natl. Acad. Sci. USA 72, 29642968.
R. LPty er al. / Journal qf Virological Methods 62 (1996) 103- 111 Weston, K. and Barrell, B.C. (1986) Sequence of the short unique region. short repeats, and part of the long repeats of human cytomegalovirus. J. Mol. Biol. 192, 177-208. Wood, N.B., Sheikholeslami, M., Pool, M. and Coon, J.S. (1994) PCR production of a digoxigenin-labeled probe for the detection of human cytomegalovirus in tissue sections. Diagn. Mol. Pathol. 3, 200-208.
111
Zipeto, D., Silini, E. and Parea M. (1990) Identification ofhuman cytomegalovirus isolates by the polymerase chain reaction. Microbiologica 13, 2977304. Zipeto, D., Baldanti, F., Zella, D. et al. (1993) Quantification of human cytomegalovirus DNA in peripheral blood polymorphonuclear leukocytes of immunocompromised patients bythepolymerasechainreaction. J.Virol. Methods44,45 -55.
Evaluation and comparison of PCR and hybridization methods for rapid detection of cytomegalovirus in clinical samples Rachel
L&y’,
Fatiha
Najioullah, Danikle Thouvenot, Michkle Aymard, Bruno Lina”
Sylvie Bosshard,
Laborutoire de Virologie, Fact&t de MPdecine Lyon Grunge Blanche. Domaine RockefcJller, h937.7 Lyon Cede.\- 08. Frunw
Accepted 3 June 1996
Abstract Rapid diagnosis of cytomegalovirus (CMV) infection may be obtained by molecular techniques. such as the polymerase chain reaction (PCR) and hybridization assays. The optimal technique to detect CMV in clinical samples was assessed. Two different PCR assays were used, targeting either the major immediate early 1 (MIE 1) or the HXLF 4 gene. The PCR products were detected by gel electrophoresis, dot blotting and an easy to use. rapid, solid phase hybridization assay, DNA enzyme immunoassay (DEIA). Standard tissue culture was also used. Cerebrospinal fluids (18) liver biopsies (9) from hepatic transplant recipients, amniotic fluids (7) from mothers with suspected peripartum infection, and samples (6) of miscellaneous origin (brain and fundus biopsy, pericardial and pleural fluid) were tested. Among the 40 samples, CMV was detected in 19 cases. Three were positive by both molecular techniques and tissue culture, 14 by molecular methods and 2 by culture. 16/19 or 9/19 CMV-positive samples were detected by PCR amplification of the HXLF 4 or MIE 1 gene, respectively and 14/16 HXLF 4-positive samples were detected using either dot-blot or DEIA, compared to 9/16 using gel electrophoresis. Thus, the most sensitive assays for the detection of CMV in clinical samples using the methods compared in the current study were PCR amplification of the HXLF 4 gene followed by dot-blot or DEIA hybridization. Kqwovds:
Human
cytomegalovirus;
Polymerase
chain reaction;
* Corresponding author: Tel.: + 33 78 777029; fax: + 33 78 014887; e-mail: [email protected] .fr ’ Laboratoire de Biologie de la Reproduction et du Dtveloppement, Hopital Edouard Herriot, 69487 Lyon Cedex 03, France. 0166.0934,‘96/$15.00
Rapid
detection;
Dot-blot;
DEIA
hybridization
1. Introduction Human cytomegalovirus (CMV) is a member of the herpesvirus group. Infections are common throughout the world. About 50680% of the
Copyright 0 1996 Elsevier Science B.V. All rights reserved /‘II SOl66-0934(96)02091-S
adult population in Central Europe or North America have serologic evidence of prior CMV infection (Onorato et al., 1985). CMV infection is now recognized as an important cause of morbidity and mortality in immunocompromised patients, including transplant recipients and HIV-infected individuals (Drew et al., 1984; Zipeto et al., 1993; Schmidt et al., 1993). CMV may also be responsible for clinically important infections in immunocompetent hosts. Transmission from mother to child may occur both in utero or during the peripartum phase (Huang et al., 1980; Dworsky et al., 1983) resulting in severe, disseminated infection in the fetus (Stagno and Whitley, 1985), or the development of congenital CMV syndrome (Pass et al., 1980). Antiviral angents such as Foscarnet and Ganciclovir are effective for the treatment of acute CMV infection. Thus, rapid and sensitive diagnostic procedures for CMV are needed. Cell culture, serology and histopathologic examination of infected tissues are currently the diagnostic methods of choice. However, culture results may take up to 14 days to become positive and the method is relatively insensitive. Serology is both insensitive and non-specific for acute infection, especially in immunocompromised patients. Invasive procedures are required to obtain biopsy specimens for histopathologic evaluation. CMV nucleic acid amplification by the polymerase chain reaction (PCR) holds promise as a more effective diagnostic modality. The use of nested PCR or hybridization with a specific probe allow an increase in sensitivity and specificity for the detection of CMV. However, nested PCR assays are subject to a high risk of contamination through DNA carry-over (Cimino et al., 1990; Porter-Jordan et al., 1990) and hybridization using a dotblot system is time consuming (Espy and Smith, 1995). The detection of PCR products using DNA enzyme immunoassay (DEIA) may provide the potential for rapid and efficient detection of PCR products (Sakrauski et al., 1994) but requires careful evaluation against other available molecular techniques such as the dot-blot assay. In the present study, two PCR assays targetting the MIE 1 and HXLF 4 gene regions of the CMV genome were evaluated for the detection of CMV
in clinical samples. The MIE region encodes for an immediate early protein (Mach et al., 1989; Stenberg et al., 1984) and the HXLF 4 region for a transmembrdne segment of a late protein (Weston and Barrell, 1986). In addition, gel electrophoresis, dot-blot and DEIA as PCR amplicon detection assays and standard cell culture were evaluated.
2. Material
and methods
2.1. Sumpk
specimens
Forty clinical specimens obtained from three different hospitals in Lyon were investigated retrospectively; 18 cerebrospinal fluids (CSF), 9 liver biopsies (LB), 7 amniotic fluids [AF] and 6 of diverse origins (brain and fundus biopsy, pleura1 and pericardial fluids). All samples were collected and stored at - 20°C for up to 2 months. All biopsy samples were collected in a sterile tube without fixation liquid. 2.2. DNA
extraction
A rapid DNA extraction procedure was used for liquid samples (CSF, AF, pleural and pericardial fluids). Briefly, 100 ,~l of the sample was mixed with 50 ,~l of extraction buffer (10 mM Tris-HCl (Sigma), pH 8.3; 50 mM KC1 (Sigma); 2.5 mM MgCl, (Sigma); 0.45% Tween 20 (Sigma) and 100 pgg/ml proteinase K (Boehringer Mannheim, Germany)). The mixture was vortexed for 10 s, then incubated for 45 min at 56°C followed by 10 min at 95°C. The mixture was centrifugated at IO 000 rpm for 1 min in an Eppendorf centrifuge. The supernatant was collected and directly used for amplification without further purification. Biopsy samples were crushed using a mortar and pestle and diluted in 1 ml of EMEM culture media (Sigma). This mixture (200 ~1) was then transferred in an Eppendorf tube and centrifuged for 10 min at 10 000 rpm in an Eppendorf microfuge. The pellet was resuspended in 500 ,~l of the lysis buffer (10 mM Tris-HCl; 10 mM NaCl (Sigma); 25 mM EDTA (Sigma); 1% SDS (Sigma)
and 1 mg/ml proteinase K) and incubated overnight at 37°C. DNA was extracted on 250 /11 of this sample using a phenol-chloroform procedure (Chomczynski and Sacchi, 1987) followed by precipitation for 1 h at - 80°C with 250 ~1 of S M NaCl in 2.5 vol of 100% ethanol. The DNA extract was pelleted by a brief centrifugation step at 10000 rpm, washed once with 0.5 ml of 70% ethanol, and resuspended in 50 ~1 of TE buffer (10 mM Tris base, 1 mM EDTA). 2.3. DNA ampl$icution Amplification of CMV DNA by PCR was carried out using two oligonucleotide primer sets. Their sequences were as follows: MIE 1 upstream primer, 5’ AGA CCT TCA TGC AGA TCT CC 3’, 20372057, and MIE 1 downstream primer, 5’ GGT GCT CAC GCA CAT TGA TC 3’, 2280.2300 (PCRa; Zipeto et al., 1990); HXLF 4 upstream, 5’ CGG ATC AAC ATA AGG ACT TTT CAC AC 3’, 11 755-11780, and HXLF 3 downstream, 5’ GAA TAC AGA CAC TTA GAG CTC GGG GT 3’, 11 895- 11920 (PCRb; Weston and Barrell, 1986). PCRa and PCRb assays were performed in a 100 111reaction mixture containing 50 mM KCI; 10 mM Tris-HCl (pH 8.3); 1.5 mM M&l,; 10 ,ul of extraction product and 0.5 U of Taq DNA polymerase (Promega). The PCRa mix reaction also contained deoxynucleoside triphosphates (250 /IM each) and 50 pM of each oligonucleotide primer. The PCRb mix reaction contained adenine, cytosine and guanine deoxynucleotide triphosphates (200 PM each); uridine deoxynucleotide triphosphate (400 PM); 40 pM of each oligonucleotide primer; and Uracil DNA Glycosylase (UNG, 1 U; Gibco BRL) was added to prevent DNA carryover contamination (Long0 et al.. 1990). Amplification using PCRa and PCRb involved 50 cycles of denaturation at 94°C (1 min), annealing at 55°C (1 min) and extension at 72°C (2 min) or 35 cycles of denaturation at 94°C (20 s), annealing at 58°C (20 s) and extension at 72°C (20 s), respectively. In addition, PCRb used a hot start (94°C for 5 min) and a final hold temperature at 72°C for 10 min to stop the UNG activity.
Each PCR run included a positive ( 10 ~1 of strain AD 169 at lo4 TICD,,,) and a negative control (DEPC water). To avoid contamination by PCR product carryover, samples were prepared in a separate laboratory and recommended precautions (aliquoted reagents, positive displacement pipettes) were followed. 2.4. Detection
oj ump@ied
DNA
2.4.1. Gel electroplzoresis
PCR products (10 111) were run on a 2% agarose gel (Sea Kern GTG agarose) in TBE buffer, 0.089 M Tris borate, pH 8.3; EDTA 2 mM (Sigma) containing 10 ,ug/ml of ethidium bromide at 200 V for 20 min. A molecular size marker was run in each gel (DNA size marker V, Boehringer Mannheim). The gels were photographed with a Polaroid camera (MP4 + ) under UV light. The size of PCR products were 166 bp for PCRb and 223 bp for PCRa (Fig. 1). 2.4.2. DE/A assu_t The DEIA assay is based on the immobilization of a capture probe/amplicon hybrid on a solid
230 160
Fig. I. Ethidium bromide staining of a 2”/;1agarose gel showing samples positive or negative for CMV DNA. Lane I. molecular weight marker (DNA V); lane 2, PCRb positive control AD 169; lane 3. PCRb positive sample; lane 4, 5 and 8. negative controls: lane 6. PCRa positive control AD 169: lane 7. PCRa positive sample.
106
R. L&y
et al. / Journal of’ Virological
phase using a biotinlavidin bridge. DNA duplexes are detected by an anti-DNA monoclonal antibody that specifically reacts with double-stranded DNA (dsDNA). The specific dsDNA antibody complexes can be visualized by a peroxydase-conjugated anti-mouse antibody and a chromogen/ susbstrate mixture. The detection of amplified CMV DNA segments with DEIA was carried out according to the recommendations of the manufacturer (Sorin Biomedica, Italy). Briefly, streptavidin-coated microtiter plates were incubated with 100 ng of the 5’ biotinylated capture probe overnight at 4°C. After washing four times with phosphate-buffered saline (PBS), the microtiter plate was ready for immediate use or storage for up to 1 month. The procedure was as follows: after denaturdtion at 96°C for 10 min, 20 ~1 of each PCR product was incubated for 1 h at 50°C in one well containing 100 ~1 of hybridization buffer. After each step, the wells were washed five times with the suitable washing buffer. The diluted anti-dsDNA antibody was added and incubated for 60 min at room temperature. Bound anti-dsDNA antibody was detected by adding 100 ~1 of horseradish peroxidase-labeled rabbit anti-mouse IgG antibody. After a last incubation step of 1 h, 100 ~1 of chromogen-substrate mixture was added, and the calorimetric reaction was performed for 30 min. The reaction was stopped with 200 ~1 of 1 M sulfuric acid, and the absorbance was measured at an O.D. of 450 nm against a reference of 630 nm. The assay .;~as performed using two different coated probes: a 5’ CTA GTG TGA TGC TGG CCA AGC GGC CTC TGA 3’, 30-mer MIE 1 probe for PCRa (Zipeto et al., 1990) and a 5’ CAG GAC TGT GTA CCG TGT ATC CAG CGC TTA CTA TC 3’, 35-mer HXLF 4 probe for PCRb (Weston and Barrell, 1986). In order to determine the sensitivity of the assay, serial lo-fold dilutions of a 104TICD,,/ 100 ~1 suspension of the reference strain AD 169 were amplified by PCRa and PCRb and detected by DEIA. The cut-off value was set at 0.5 O.D. at 450 nm with an equivocal zone of &30%, as recommended by the manufacturer.
Methock
62 (1996) IO.?- 1 I I
2nd 1
2
3
4
5
6
2
1st
3 4
5
6
I
Fig. 2. Dot-blot analysis of PCRb products using the tailed oligonucleotide probe and a chemiluminescence detection system: (I, I), positive control; (1,2),negative control; (1,3), (2,2). (3,2), (2,6), (3,6), positive samples; (6,5), ‘indeterminant sample’.
2.4.3. Dot-blot hybridization Hybond N + membranes (Ammersham) were prehybridized for 1 h at 37°C in 20 ml of the hybridization buffer (5 x SSC, 50% formamide, 2% blocking reagent, 0.1% sarcosyl and 0.02% SDS). Hybridization (Fig. 2) with the tailed oligonucleotide probe was carried out overnight with the same buffer and at the same temperature. The tailed probe was labeled with digoxigenindUTP according to the manufacturer’s protocol (DIG Oligonucleotide tailing kit, Boehringer Mannheim). The tailing efficiency was checked by direct immunological detection (DIG Nucleic Acid Detection Kit, Boehringer Mannheim). The membranes were then washed twice (2 x SSC, 0.01% SDS) and the immunological detection of bound probes was performed using a chemiluminescence detection system (DIG Luminescent Detection Kit, Boehringer Mannheim; Zipeto et al., 1993; Wood et al., 1994). 2.5. PCR inhibitor detection The presence of amplification inhibitors was evaluated. All CMV-negative DNA samples were spiked with 10 ~1 of DNA isolated from reference
R. L&v)’ et al. (/IJournal
of Virological
strain AD 169 or diluted 1:5 (fluid samples) or 1:25 and 1:50 (biopsy specimens) followed by amplification and the detection methods described above.
Clinical specimens were inoculated onto human embryonic fibroblast cell cultures (MRCS) using a centrifugation step as described elsewhere (Swenson and Kaplan, 1985). Virus detection and identification were performed at day 2 post-infection by an immunoperoxidase (IP) assay using a commercially available specific monoclonal antibody El3 (Mazeron et al.. 1992). Monolayers were observed daily for 21 days for cytopathic effect on IP negative samples.
3. Results Sensitivity of the DEIA assay. The lowest detection limit of PCR amplicons was a 10 ~ ’ dilution of AD 169 (0.1 TCID,,/lOO ~1) using PCRa primers versus 10 ’ dilution (0.01 TCID,,),‘lOO ,uI) for PCRb primers. The mean O.D. value of the positive specimens was 1.687 for PCRa and 1.85 I for PCRb. The positive controls of the two PCR assays had an O.D. value 2 2.0. ?. 1. Conzparison
of’ the detection
methods
CMV was detected in 19 of the 40 samples by one or more detection techniques (Table 1). Tissue culture had the lowest sensitivity, with only 5’19 CMV-positive samples detected using this process. Among the 27 fluid samples evaluated, CMV was not detected by cell culture in any of the samples and 10 were positive using PCR and subsequent hybridization. The sensitivity of cell culture was improved when evaluating tissue samples: 5:’13 were culture-positive compared to 7/ 13 using molecular methods. Interestingly, two tissue samples were positive by cell culture but negative by molecular methods. Molecular detection methods remained negative after spiking the negative samples with control CMV DNA.
Methods
62 (1996)
103
III
107
Using molecular techniques, 17/19 CMV positive samples were detected. PCRb was the most sensitive amplification method, 16/ 17 samples with detectable CMV by molecular methods were positive using this amplicon compared to 9/17 using PCRa. Gel electrophoresis and DEIA were equivalent as detection methods for PCRa amplicons. All 9 PCRa-positive samples were detected using both methods. However, dot-blot and DEIA were superior to gel electrophoresis for detection of PCRb amplicons; 14116 PCRb-positive samples were detected by either dot-blot or DEIA compared to 91’16 using gel electrophoresis.
4. Discussion CMV may cause serious disease in both immunocompetent and immunocompromised patients. Although cell culture. serology and histopathologic examination are often the currently used methods for the diagnosis of acute CMV infection, these techniques are limited by several factors. The PCR assay offers promise for routine detection of active CMV infections. In order to be adopted by clinical laboratories, PCR assays must be sensitive, specific, rapid, easy to perform, reproducible and cost-effective. The techniques described in the current study full% these criteria. PCR was shown to be more sensitive than conventional cell culture since among the 19 positive patients, only 5 were detected by cell culture versus 17 by PCR. The specificity was confirmed by attempting to detect CMV DNA in isolates that were culture-positive for other herpesviruses or in samples obtained from seronegative patients. The results were all negative (data not shown). Furthermore, the clinical presentation and/or the results of virological investigations undertaken on samples collected later or from other sites supported the diagnosis of acute CMV infection in the 14 patients with culture-negative, PCR-positive samples. Amniotic fluid was recovered from a mother bearing a fetus with intrauterine growth retardation and hydrocephalus: 4 liver biopsy specimens were collected from hepatic transplant patients that were
108
Table I List of clinical
samples
investigated
by the different
techniques _
Type of samples
PCRa . Gel
AF” AF AF AF AF AF AF CSFb CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF CSF
Gel
+ _
_ +(a)
_ _ _ +(a) _ _ _ _ _ + + _ + + _ _
Dot-blot
DEIA
_ _ + _ _
_ _ _ _ + -
+ _ _ _ _ _ + _ _ + + _
+ _ _ _ _ + +
_ + _ + _ +
_ _
+(a) _ _
+ + + + + _ _
+ +(a) + + + _ -
_ _
_
_ _
-
_ _
_
+(c)
+ Cc)
CMV
_
-
-
_
+ CC)
+(c)
+ Cc)
+(c)
CMV CMV
_
_ _
+(b) _ _
+(b) _
_ _ _ _ -
_ -
_ -
-
+ Cc) +(c)
-
-
_
+(b) _
+(b) -
+(b) -
+(b) -
+(b) _
CMV CMV
i-(c)
+
+(c)
-
_ _
_ -
_ _
_ _
-
-
Fundus Fundus
_ + Cc)
Pericardial fluid Pleural fluid
DEIA
Culture
-
Brain Brain
LB’ LB LB LB LB LB LB LB LB
PCRb
_
PCRa and PCRb procedures and primers are described in Materials and methods. Samples were evaluated at different dilutions: Fluid samples were tested undiluted and at a I:5 dilution (f. positive undiluted; +(a), positive at I:5 dilution). Biopsy specimens were tested undiluted and at I:25 and I:50 dilutions (+, positive undiluted; +(b), positive at I:25 dilution; C(c), positive at I:50 dilution). CMV, positive cell culture of CMV; -. negative culture or test.“Amniotic fluid. %Zerebrospinal fluid. ‘Liver biopsy.
R. LPvy et al. /Journal of’ Virological Methods 62 (1996) 103-l II
seronegative for CMV preoperatively and had either postoperative seroconversion or positive cell culture results from subsequent liver or blood samples. CSF was collected from three HIV-infected patients with encephalitis; one lung and one bone marrow transplant recipient with CMV isolated by cell culture from blood samples one month later; one multitransfused patient with CMV later isolated from the blood; and from a 6-year-old boy with a mononucleosis-like syndrome and blood culture positive for CMV. Clinical information was not available for the two other patients with PCR-positive, culture-negative CSF specimens. The specificity of the PCR assay for the detection of active CMV infection has also been discussed elsewhere (Fillet et al., 1993; Weston and Barrell, 1986; Zipeto et al., 1990). The detection of CMV using PCR and DEIA could be accomplished in 9 h using simple procedures. The use of bulk reagents and limited technician time renders the test cost-effective. PCR appears to be especially important for the diagnosis of CMV-infected fluids. In the present study, all 27 fluid samples evaluated by cell culture were negative, while molecular techniques gave positive results in 10. These results confirm the necessity of developing a more sensitive test than cell culture for fluid samples as discussed by other investigators (Fillet et al., 1993; Mazeron and Alain, 1993; Smith and Dunstan, 1993). However, PCR failed to detect CMV DNA in two of the biopsy samples that had virus isolated in cell culture. As previously described by other authors, these biopsy samples may possess Tuq DNA polymerase inhibitors and/or DNAases (Gass et al., 1993). The results of this study have confirmed these findings, CMV DNA was added to the DNA extracts from these biopsy specimens prior to amplication. They remained negative following PCR and hybridization. Furthermore, CMV was only detected by PCR in biopsy samples that had been diluted prior to amplification. Two PCR assays amplifying different regions of the CMV genome were compared. MIE 1 is the region transcribed in greatest abundance during early infection with CMV. Intermediate Early (IE) gene expression occurs independently of de novo protein synthesis and IE gene products can be detected within l-2 h after infection (Rice et al.,
109
1984). The activation of IE 1 gene expression is one of the first steps of transcriptional interaction between the virus and host cell. Proteins expressed by the HXLF 4 region exhibit close homology with a class of membrane proteins. Markedly hydrophobic sequences close to the C terminus could correspond to transmembrane regions (Tomita and Marchesi, 1975). In addition, all members of the HXLF protein group (HXLF 1, 2, 3 and 4) have a putative signal sequence for membrane insertion (Weston and Barrell, 1986). These differences in both genome position and function may be reflected in different sensitivities for PCR detection of CMV infection in clinical samples (Rice et al., 1984). PCR using the HXLF region of the CMV genome was more sensitive than amplification of the MIE region. This finding was observed both in comparative serial dilution assay (HXLF more sensitive by 1 log,,) and in the evaluation of clinical samples (16 samples positive using HXLF versus 9 using MIE 1). Using HXLF amplicons, dot-blot hybridization and DEIA were equivalent in their detection capabilities. Both were more sensitive than gel electrophoresis. However, DEIA is prefered over dot-blotting because it was easier to perform, less time-consuming and the results were consistently unequivocal. On the other hand, the dot-blot assay required frequent troubleshooting. Difficulties were encountered at three steps in the process: (i) failure of DNA transfer to the nylon membrane, (ii) lack of effective hybridization with the probes, and (iii) difficulty in detection due to high background New hybridization techniques for the detection of PCR amplified DNA, such as DEIA, have considerable potential use in routine diagnostic laboratories since they are characterized by easy handling, rapid results and high sensitivity. DEIA also appears to be superior to conventional techniques such as dot-blot hybridization due to potential automatization and less time-consuming methodology.
5. Conclusion The region
PCRb system amplifying the of the CMV genome followed
HXLF 4 by DEIA
110
R. LCvy et al. /Journal
of’ Virological
hybridization has proven to be a very-‘sensitive method for the diagnosis of CMV infection. The method is efficient, fast and practical, providing potential for use in routine diagnostic clinical laboratories. However, the presence of Tuq DNA polymerase inhibitors in tissues may limit its usefulness for these samples.
Methods
Longo, M.C., Berninger, M.S. and Hartley, J.L. (1990) Use of uracil DNA glycosylase to control carry-over contamination
in polymerase
We are grateful to Sorin Biomedica for providing DEIA kits. Thanks to Marc Carteron and Olivia RCny for their excellent technical assistance, and to Darryl See for careful reading of the manuscript.
chain
reactions.
Gene 93, 125- 128.
Mach, M., Stamminger, T. and Jahn G. (1989) Human c’ytomegalovirus: Recent aspects from molecular biology. J. Gen. Virol. 70, 3117-3146. Mazeron,
M.C.
and
tomegalovirus nosis. Pathol. Mazeron,
Acknowledgements
62 (1996) 10% I II
Alain-Albertini,
S. (1993)
infection: new methods Biol., 41, 487-494.
M.C., Jahn,
G. and Platcher
Human
S.K.
I.M., Morens, (1985)
recommendations Dis. 7. 479-497. Pass,
R.F.,
sults
S., Myers,
of symptomatic
of long
Martone,
term
infections:
and control.
G.J.
Rev. Infect.
and Alford,
congenital
longitudinal
recogimmedi-
W.J. and Stansfield,
of cytomegaloviral
for prevention
Stagno,
Outcome
D.M.,
Epidemiology
diag-
B. (1992) Monoclonal
Antibody El3 (M810) to human cytomegalovirus nize an epitope encoded by exon 2 of the major ate early gene. J. Gen. Virol. 73, 2699 -2703. Onorato,
cy-
for virological
C.A. (1980)
CMV
follow-up.
infection:
re-
Pediatrics
66,
758-762.
References Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidium thiocyanate phenol chloroform extraction. Anal. Biochem. 162, 156- 159. Cimino, G.D., Metchette, K.C., Tessman, J.W., Hearst, J.E. and Isaaca ST. (1990) Post PCR sterilization: a method to control carry-over contamination for the polymerase chain reaction. Nucleic Acids Res. 19, 99-107. Drew, W.L., Sweet, E.S., Miner, R.C. and Mocarski, E.S. (1984) Multiple infections by cytomegalovirus in patients with acquired immunodeficiency syndrome: documentation by Southern blot hybridization. J. Infect. Dis. 6, 952-953. Dworsky, M.E., Yow, M., Stagno, S., Pass, R.F. and Alford, C.A. (1983) Cytomegalovirus infection of breast milk and transmission in infancy. Pediatrics 72, 295-299. Espy, M.J. and Smith, T.F. (1995) Comparison of SHARP signal system and Southern blot hybridization analysis for detection of cytomegalovirus in clinical specimens by PCR. J. Clin. Microbial. 33, 3028-3030. Fillet, A.M., Katlama, C., Visse, B., Camilleri, S., Rogeaux, 0. and Huraux, J.M. (1993) Human CMV infection of the CNS: concordance between PCR detection in CSF and pathological examination. AIDS, 7, 1016-1018. Gass, P., Kiessling, M., SchBfer, P., Mester, C., Schmitt, H.P. and Kiihn, J.E. (1993) Detection of human cytomegalovirus DNA in paraffin sections of human brain by polymerase chain reaction and the occurrence of false negative results. J. Neural. Neurosurg. Psychiatry 56, 211-214. Huang, E.S., Alford, C.A., Reynolds, D.W., Stagno, S. and Pass R.F. (1980) Molecular epidemiology of cytomegalovirus infections in women and their infants. N. Engl. J. Med. 303, 958-962.
Porter-Jordan, K., Rosenberg, El. and Keiser, W.F. (1990) Nested polymerase chain reaction assay for the detection of cytomegalovirus contamination
overcomes
false
with fragmented
positives
DNA.
caused
by
J. Med. Virol.
30,
85.-91. Rice, C.P.A., Schrier, R.O. and Oldstone, M.B.A. Cytomegalovirus infects human lymphocytes and cytes:
virus
gene
products.
expression Proc.
is restricted Natl.
Acad.
(1984) mono-
to immediate Sci. USA.
early
81, 6134
6138. Sakrauski, Doerr,
A.. Weber, B., Kessler, H.W. (1994) Comparison
H.H., Pierer, K. and of two hybridization
assays for the rapid detection of PCR amplified genome sequences from cerebrospinal fluid. J. Methods 50, l75- 184. Schmidt, C.A.. Oettle, H., Neuhaus, stration
of cytomegalovirus
after liver transplantation. Smith, K.L. and Dunstan,
HSV Virol.
P. et al. (1993) Demon-
by polymerase
chain
reaction
Transplantation 56, 872-874. R.A. (1993). PCR detection of
cytomegalovirus: a review. Br. J. Haematol. Stagno, S. and Whitley, R.J. (1985) Herpesvirus
84, 187.. 190. infections of
pregnancy. Part I: Cytomegalovirus and Epstein-Barr virus infections. N. Engl. J. Med. 313, 1270-1274. Stenberg,
R.M..
Structural
Thomsen,
analysis
D.R.
of the major
human cytomegalovirus. Swenson, P.D. and Kaplan,
and
Stinski,
immediate
M.F. early
(1984) gene of
J. Virol. 49, 190- 199. M.H. (1985). Rapid detection
of
cytomegalovirus in cell culture by indirect immunoperoxydase staining with monoclonal antibody to an early nuclear antigen. J. Clin. Microbial. 21, 669-673. Tomita, M. and Marchesi, V.T. (1975) Amino-acid sequence and oligosaccharide attachment sites of human erythrocyte glycophorin. Proc. Natl. Acad. Sci. USA 72, 29642968.
R. LPty er al. / Journal qf Virological Methods 62 (1996) 103- 111 Weston, K. and Barrell, B.C. (1986) Sequence of the short unique region. short repeats, and part of the long repeats of human cytomegalovirus. J. Mol. Biol. 192, 177-208. Wood, N.B., Sheikholeslami, M., Pool, M. and Coon, J.S. (1994) PCR production of a digoxigenin-labeled probe for the detection of human cytomegalovirus in tissue sections. Diagn. Mol. Pathol. 3, 200-208.
111
Zipeto, D., Silini, E. and Parea M. (1990) Identification ofhuman cytomegalovirus isolates by the polymerase chain reaction. Microbiologica 13, 2977304. Zipeto, D., Baldanti, F., Zella, D. et al. (1993) Quantification of human cytomegalovirus DNA in peripheral blood polymorphonuclear leukocytes of immunocompromised patients bythepolymerasechainreaction. J.Virol. Methods44,45 -55.