Abstract
Human cytomegalovirus (HCMV) disease remains a major cause of morbidity and mortality after lung transplantation. Currently, routine diagnostic tests for HCMV are inefficient and insensitive or nonspecific for HCMV disease. We describe an efficient, highly sensitive, quantitative polymerase chain reaction (PCR) assay for HCMV using competitive PCR and fluorescently labeled primers, and we have used this to measure HCMV DNA load in donor and recipient tissues of six lung transplant recipients at the time of transplantation, and 2 wk after transplantation when clinically stable. Total DNA yield was adequate for analysis in transbronchial biopsy, bronchoalveolar lavage, and peripheral blood leukocytes, but the endobronchial biopsy specimens did not consistently produce sufficient DNA for analysis. There was a large intersubject and intrasubject variability between tissues in HCMV DNA load, with a tendency for greater levels in lung tissue compared with BAL or peripheral blood cells. All six HCMV IgG seronegative donors or recipients were found to have HCMV DNA present. One of the three seronegative matched transplant recipients developed histopathologically proven HCMV disease, and HCMV DNA levels were shown to increase at that time point and subsequently decrease with ganciclovir treatment. This assay will allow prospective studies to confirm the predictive value of HCMV DNA load in donor and recipient tissues for HCMV disease. Kotsimbos ATC, Sinickas V, Glare EM, Esmore DS, Snell GI, Walters EH, Williams TJ. Quantitative detection of human cytomegalovirus DNA in lung transplant recipients.
Human cytomegalovirus (HCMV) is an ubiquitous virus with the ability to remain latent in host tissues and become reactivated in those patients that are immunosuppressed (1). It is the most common and often most serious viral infection seen after organ transplantation, and this is particularly so in recipients of lung allografts (2, 3).
Currently, the routine detection of HCMV is relatively slow and inefficient, despite the development of the shell vial culture method of HCMV, and the use of fluorescent antibody staining of HCMV-inoculated cell lines and tissue specimens (4, 5). In addition, the distinction between HCMV infection and HCMV disease is difficult (1, 6, 7).
The previously reported use of the polymerase chain reaction (PCR) has improved the speed and sensitivity of HCMV detection (8-11), but not the specificity for HCMV disease (12-15). Semiquantitative PCR has been used in an attempt to distinguish between HCMV infection and HCMV disease episodes by measuring HCMV DNA load (12). The few reports of semiquantitative PCR in allograft recipients have failed to show a definite correlation between HCMV DNA load and HCMV disease. However, an association cannot be excluded because of the inadequate quantitative nature of the PCR methods used (16-18).
The more recent use of competitive PCR has improved the quantitative capacity of PCR-based assay systems for HCMV, although they have been limited still by relatively complex strategies to detect amplified product that involve hybridization steps and the use of either radioactivity or chemiluminescence labels. These detection systems increase the difficulty of performing the assay and are likely to increase its variability, although reproducibility studies have not been reported (8, 19, 20). Furthermore, most of the reported literature has concentrated on the HCMV load in blood rather than in the more relevant allograft tissue (19-21).
The aims of this work were (1) to develop and evaluate an efficient, quantitative, and user-friendly PCR assay for HCMV based on competitive PCR and fluorescently labeled primers that allowed amplified product to be detected by a laser-based fluorescent scanner (FluorImager; Molecular Dynamics, Melbourne, Australia); (2) to apply this system to compare HCMV DNA load in donor and recipient tissue at the time of transplantation with HCMV serum IgG antibody status, and (3) to compare HCMV DNA load in various samples (transbronchial biopsy, endobronchial biopsy, bronchoalveolar lavage, and peripheral blood) 2 wk after transplantation and report clinical follow-up over 6 mo.
Genomic HCMV, Varicella-Zoster virus (VZV) and Herpes simplex virus (HSV) were provided by the Virology reference laboratory at Fairfield Hospital (Melbourne, Australia). HCMV strain AD169 culture stock was provided by Mr. Eric Uren at the Department of Microbiology, Royal Children's Hospital (Melbourne, Australia), and amplified in MRC-5 cell lines using standard protocols (22).
This institution performs 35 to 45 lung transplants per year. During the first 6 mo of 1994, six consenting lung transplant recipients were recruited into our study with prior approval from the Hospital Ethics Committee. The patients were selected on the basis of providing a spectrum of donor/recipient HCMV IgG status. Serologic testing was performed using an ELISA kit for HCMV IgG (Gull, Salt Lake City, UT).
Both donor and recipient tissue was obtained at the time of transplantation. The recipient tissue was obtained from the explanted lung, whereas donor tissue obtained was usually splenic tissue, depending upon organ harvesting logistics and surgical constraints.
Bronchoscopy was performed routinely at 2 wk post-transplantation, and at that time at least six transbronchial lung biopsies were taken along with BAL fluid. These were sent for routine histopathologic, cytologic, and microbiologic evaluation. In addition, two transbronchial lung biopsies, two endobronchial biopsies, 30 ml of bronchoalveolar lavage (BAL) fluid, and 10 ml of blood were taken for CMV PCR.
Biopsy samples. The lung and endobronchial biopsies were immediately placed into separate solutions of sterile phosphate-buffered saline (PBS), transported on ice to the laboratory, and centrifuged at 128 × g for 10 min. The supernatant was discarded and the remaining pellet was snap-frozen in liquid nitrogen and stored at −70° C.
BAL samples. BAL was performed by injecting three 60-ml aliquots of sterile, isotonic saline after impacting the bronchoscope in a subsegmental bronchus of the right middle lobe or lingula. Thirty milliliters of lavage fluid aspirated were collected for our study. Total white cells were counted using a Neubaur counting chamber and centrifugated at 128 × g for 10 min; the supernatant was discarded, and the cell pellet was snap-frozen and stored at −70° C.
Blood samples. At the time of bronchoscopy 10 ml of blood were taken (in EDTA vacuum tubes), of which 4 ml were sent for HCMV culture (22). Six milliliters were retained for HCMV PCR and processed within 3 h. The total white blood cell (WBC) fraction (i.e., both polymorphonuclear and mononuclear cells) was isolated using a dex-tran-based density gradient. The washed WBC were then snap-frozen in liquid nitrogen and stored at −70° C.
DNA was extracted with the Progenome II DNA extraction kit (Progen Industries Ltd., Brisbane, Australia). This kit was used according to the manufacturer's instructions, except two ethanol washes were used rather then one in order to remove excess salt contamination from the purified DNA pellet. An RNase step was incorporated into the procedure to remove contaminating RNA species, which could interfere with the accurate measurement of DNA. The extracted DNA was resuspended in 100 μl sterile H2O.
DNA purity was assessed using capillary cuvette microsample spectrophotometry (Capillary adaptor cells; Helix) and absorbance (A) ratio measurements (A260 nm/A280 nm), which ranged between 1.64 and 1.94 (average, 1.85; n = 6; pure DNA = 1.8 to 2.0). DNA concentration was measured with fluorescent dye labeling (Fluorkit DQS assay; Molecular Dynamics) and FluorImager detection as per the manufacturer's instructions. A standard calibration curve was developed (using known amounts of DNA), from which the total DNA concentration of the unknown samples could be determined. The line of best fit for this standard curve was described by a linear regression. By using the measured total DNA as the denominator for the HCMV quantitations, corrections could be made for the variations in DNA extraction yields from different samples. Samples with total extracted DNA levels less than 10 ng were excluded from analysis.
The mean (range) yield of DNA extracted from the parenchymal lung tissue (two combined transbronchial lung biopsies), BAL, and peripheral blood WBC was 1.3 μg (10 to 4,000 ng, n = 14), 6.8 μg (390 to 28,000 ng, n = 6), and 15 μg (10 to 18,000 ng, n = 6), respectively. The yield from endobronchial biopsies was relatively poor, with yields frequently less than 10 ng (i.e., < 10 ng/100 μl), and hence were excluded.
Oligonucleotides. CMV specific primers directed at a conserved region within the DNA polymerase gene of HCMV (HCMV strain AD169) were synthesized (Bresatec Ltd., Melbourne, Australia) as previously reported by Greenfield and coworkers (9). The primers, upstream 5′TCGCTGCTGGACGAGTGCGCCTGCCGCGAT and downstream 5′CCGGCGCCGTGATTGTCGTTGGAAACGCCG, amplify a 314 base pair (bp) fragment, and their specificity for HCMV has been previously determined (9). The upstream primer was labeled during synthesis by the manufacturer with fluorescein to facilitate PCR product detection by fluorescent scanning of agarose gel in the FluorImager.
Generation of PCR product. PCR was carried out using reagents from Perkin-Elmer (Pomona, CA) in a final volume of 50 μl, consisting of 10 μl of sample DNA template, 50 mM KCl, 10 mM TRIS-HCl at pH 8.3, 1 mM MgCl2, 200 μM nucleotides, 1 unit Taq DNA polymerase, and 0.2 μM oligonucleotide primers. The Taq DNA polymerase was neutralized with TaqStart Antibody (Clontech Laboratories, Palo Alto, CA) before use to facilitate “hot start” PCR. To ensure that all the Taq polymerase was released from the antibody, the first denaturation step was increased from 1 to 3 min. The utilization of a “hot start” protocol with Taq antibody improved the detection threshold by a factor of 10, down to < 100 molecules, with a corresponding decrease in the amount of primer-dimer formation. Master mixes were used whenever possible, and positive and negative PCR controls were used. PCR was conducted, typically for 40 cycles, with denaturation for 1 min of 94° C, annealing for 1 min at 68° C, and extension for 1 min at 72° C in a thermal cycler fitted with a heated lid (MJ Research, MA).
An internal standard was constructed by PCR-mediated deletion mutagenesis (23) using purified human genomic HCMV DNA from strain AD169 (Sigma Aldrich, Sydney, Australia). A second upstream primer site (70 base pairs from the first) internal to PCR primers and compatible with the downstream PCR primer was found using the primer analysis software Oligo 4 (National Biosciences, Plymouth, MN). A 58 base composite primer was synthesized, 5′TCGCTGCTGGAC-GAGTGCGCCTGCCGCGATCGCTGCTATCATCTCTACCGCC-GCTGTG, consisting of the PCR upstream primer sequence (30 bases in bold) and the second upstream site (28 bases). Thus, the composite primer paired with the downstream PCR primer is able to amplify a smaller DNA fragment missing a 70 base pair sequence that intervened between the two upstream primer sites. This PCR product, although smaller than the native product, retains exactly the same primer binding sites as the native HCMV DNA. A 70 bp difference was chosen to allow good resolution of the native and mutant PCR products on agarose gels.
Using the mutant PCR product as a template, a second round of PCR was performed using gene-specific primers to produce a large quantity of this smaller fragment, which would subsequently be used as the internal standard competitor molecule. The internal standard was purified by phenol-chloroform extraction and precipitation and then quantitated spectrophotometrically. A known amount of this internal standard competitor was added to the PCR assay such that any native HCMV DNA present would compete with it for amplification.
Amplified PCR products were analyzed by electrophoresis in 3% agarose gels. The separated native and mutant amplified products were then detected using a fluorescent scanner (FluorImager; Molecular Dynamics). The relative fluorescence of the two fluorescein-labeled bands was then quantitated using ImageQuaNT software (Molecular Dynamics).
The initial number of native HCMV molecules in the sample aliquot was calculated using the known amount of internal standard added to the PCR reaction mix and relative ratio of the two PCR product bands. For each clinical specimen, a number of reactions were performed using different amounts of internal standard. When both native and competitor PCR product bands were quantifiable, the ratio between them could then be measured, and thus the input number of native molecules could be determined. This copy number of HCMV DNA per 10-μl aliquot of sample was then corrected by the amount of total DNA in that aliquot. Hence, HCMV DNA load was expressed as the number of HCMV DNA molecules per microgram of total DNA extracted.
Competitive PCR assays are based on the input ratio of native to mutant molecules remaining unchanged throughout the PCR amplification and the subsequent detection process (24). That is, the output ratio should equal the input ratio. To test this, a changing ratio experiment was performed in which the concentration of native HCMV DNA template in the PCR was kept constant, whereas the amount of competitor internal standard was varied (Figure 1). This confirmed that true competition was occurring. The coefficient of variation for the calculated copy number of HCMV DNA molecules in these reactions was 27%, which was deemed reasonable for the purposes for which the assay system was designed.
Fig. 1. Competitive PCR of HCMV tissue culture lysate. FluorImager scan of a 3% agarose gel detecting fluorescent primer labeled PCR product. Subsequent staining with ethidium bromide allowed marker DNA also to be detected by the FluorImager. In this changing ratio experiment, the native HCMV DNA template in the sample was kept constant, whereas the amount of competitor internal standard in the PCR reaction mix was varied. The fluorescent signals from the specific PCR products was enhanced in Image QuaNT, and the volume (fluorescence times area) of each band was quantitated. The ratio of the competing native and internal standard PCR products was then calculated, and the original number of molecules of HCMV DNA in the aliquot of HCMV tissue culture lysate was determined (mean, 152; standard deviation, 41; coefficient of variation, 27%).
[More] [Minimize]The within-experiment reproducibility of the assay system was assessed by using 10 replicates at an optimal ratio of native to competitor of 1:1 (mean ratio, 0.99; standard deviation, 0.046; coefficient of variation, 4.6%). The reproducibility of the assay on 5 different days was also assessed as a measure of day-to-day variation in the assay system. Under these conditions, the coefficient of variation for the measured ratios of the two products was 8.3%.
Patient characteristics. The characteristics of the six patients whose tissues were analyzed for HCMV DNA load are shown in Table 1. All patients were treated with a conventional immunosuppressive regime consisting of cyclosporin A, azathioprine, and prednisolone immediately after transplantation. Three of these six patients were donor and recipient HCMV IgG negative. One patient was donor negative and recipient positive for HCMV IgG, and thus received ganciclovir prophylaxis. Two further patients were donor and recipient HCMV IgG positive, and also received ganciclovir prophylaxis. Ganciclovir prophylaxis consisted of 5 mg/kg intravenously twice a day for 2 wk followed by 5 mg/kg three times a week for 8 wk. Subsequent HCMV disease was defined as a clinical HCMV syndrome in association with histopathologic evidence of HCMV disease.
| Patient No. | Indication | Lung Transplant Type | HCMV Serostatus (IgG) | Ganciclovir Prophylaxis | HCMV Clinical Outcome (6 mo) | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Bronchiectasis | Bilateral | D−/ R− | No | Stable | |||||
| 2 | Emphysema | Single | D−/ R− | No | Stable | |||||
| 3 | Cystic fibrosis | Bilateral | D−/ R− | No | HCMV disease | |||||
| 4 | Eisenmenger's | Heart-lung | D−/ R+ | Yes | Stable | |||||
| 5 | Bronchiectasis | Double lung | D+/ R+ | Yes | Stable* | |||||
| 6 | Emphysema | Single lung | D+/ R+ | Yes | HCMV disease |
At 2 wk post-transplantation, all six patients tested were stable, with negative cultures for HCMV and no evidence of clinical HCMV disease. After 6 mo of follow-up (Table 1), four patients had remained free of HCMV disease, but two (Patients 3 and 6) had developed clinical and histopathologic evidence of HCMV pneumonitis. Of particular interest was Patient 3 (a seronegative match for HCMV), who developed HCMV pneumonitis 1 mo after transplantation.
The HCMV DNA load in the donor and recipient tissues was compared with the HCMV IgG antibody serostatus of these patients. For the recipients, lung parenchymal tissue from the explanted lung was tested in all cases. The donor tissue tested in all cases was spleen, except for the donor to Patient 6 where only donor bronchial tissue was available. The results are shown in Figure 2. All seronegative patients had detectable tissue HCMV DNA levels despite the relevant tissue DNA extraction and PCR controls being negative. The DNA extraction control was filtered, sterile water that was subjected to exactly the same DNA extraction procedure as the tissues, and the PCR negative control was also filtered sterile water. Native HCMV DNA was detected in the tissues using 103 molecules of internal standard competitor, but there was no evidence of contaminating HCMV DNA in either of the negative controls. The subsequent detection of HCMV DNA in the tissues of the seronegative matches for HCMV and the development of HCMV disease of Patient 3 strongly suggests that the initial detection of HCMV DNA in tissues of seronegative patients was correct.
Fig. 2. Comparison of serostatus (CMV IgG) and CMV quantitation by competitive PCR in donor and recipient tissue at the time of transplantation. Note that despite the negative serostatus, detectable levels of CMV DNA were found in donors and recipients.
[More] [Minimize]The HCMV DNA load per microgram total DNA was determined for the stable patients' multiple tissues 2 wk post-transplant (first routine bronchoscopy). These results (Figure 3) show that the HCMV DNA load is variable both between individual patients and between each patient's specimen (100 to 100,000 molecules/μg total DNA), with a tendency for the load in the transbronchial lung biopsy specimens to be consistently greater than in the BAL or blood cell samples. Given that the coefficient of variation for the assay system was 27%, it is not likely that this degree of inherent variation in the assay system would dramatically effect the relative magnitude of these results.
Fig. 3. CMV quantitation by competitive PCR in recipient blood, bronchoalveolar lavage (BAL), and transbronchial biopsy (TBB) tissue. The recipients were clinically stable 2 wk after lung transplantation. Consistently high levels of CMV DNA were found in TBB, but more variable levels were found in BAL and blood.
[More] [Minimize]Patients 3 and 6 went on to develop CMV disease (clinical and histopathologic evidence), and stored specimens from these patients were tested using our assay. Patient 3 was a negative/ negative CMV seromatch, but the donor's blood tested positive for CMV PCR at the hospital where the organs were harvested, thus raising the possibility of a CMV-positive blood transfusion having been given to this patient prior to transplantation. Nevertheless, no Ganciclovir prophylaxis was administered. The absolute level of CMV DNA at the CMV disease time point was no greater than that in the stable patients, but the levels for all three specimens tested (TBX, BAL, WBC) all increased by a factor of approximately 10 to 100 above baseline, and then decreased with Ganciclovir treatment, but not to baseline levels. The CMV DNA load was highest in the TBX followed by BAL and WBC. For Patient 6, who had received Ganciclovir prophylaxis, the amount of donor tissue CMV DNA was very high, but the tissue CMV DNA levels fell to relatively low levels after the institution of a 10-wk course of Ganciclovir prophylaxis post-transplantation (decreased by a factor of approximately 1,000). Nevertheless, this patient developed CMV disease several months after Ganciclovir was ceased. Unfortunately, no specimens were obtained for testing at this time point because of clinical contraindications to obtaining extra bronchoscopy samples.
We have developed an effective and practical quantitative PCR assay for HCMV DNA in human tissue, which was used to quantitate the amount of HCMV DNA in the tissues of six stable lung transplant recipients. The system required the development of an internal standard molecule, with utilization of fluorescent-labeled primers allowing quantitation with a FluorImager laser scanner. Sample preparation with adequate DNA extraction and purification was crucial for high sensitivity PCR. The accurate measurement of the total DNA being used per PCR reaction provided an excellent denominator for the HCMV PCR quantitation results, and it was essential for a quantitative assay system given the large variation in the yield of DNA that was shown to occur.
The data show that HCMV DNA appears more sensitive than serology as a marker of HCMV infection, and that there is a large intersubject and intrasubject variability between different tissues in HCMV DNA load. Our data support existing evidence that the amount of HCMV virus in the donor organ at the time of transplantation may be an important predictor of HCMV disease (25), and CMV serology can be quite misleading, confirming reported high rates of positive samples using conventional PCR in seronegative immunocompromised patients (15) and even blood donors (26, 27).
Although all the seronegative tissues tested were positive for HCMV DNA, the small number of samples tested needs to be borne in mind. Nevertheless, these results raise several issues. Firstly, there is the possibility of contamination with CMV DNA, which seems unlikely given that the negative controls in the PCR runs remained negative. Secondly, it is possible that our serologic testing for CMV may not be very sensitive as serologic testing for CMV is not a well-standardized test. Thirdly, the donor tissue tested in five of the six donors was splenic tissue and this tissue site may be more sensitive for latent CMV. And finally, both the donor and the recipient tissues were tested at the time that these subjects were under a major physiologic stress. The question of a gold standard is always difficult when introducing a test that is likely to be more sensitive than the existing tests available. In such a situation one must rely on other parameters, which include: consistency with known pathobiologic principles, correlation with disease development, and response to therapy.
HCMV DNA was measured at 2 wk post-transplantation in lung tissue and BAL of the six stable patients selected. This was unique for lung transplant recipients as most published studies using PCR to detect HCMV in these patients have used either peripheral blood white cells or, occasionally, BAL cells. For these stable patients, the HCMV DNA levels in their tissues was quite variable, ranging from 100 to 100,000 genome equivalents per microgram of total DNA extracted (Figure 3). Given that there are approximately 7.2 pg of DNA per cell (28), then these numbers are approximately equivalent to a range of one copy of HCMV per 10 to 1,000 cells. These levels are in agreement with those determined by a relatively more complex PCR-based assay system based on nested PCR, solution hybridization with radioactive DNA probe, and temperature gradient gel electrophoresis (20), where the average level of HCMV DNA detected in the peripheral blood white cells of six renal allograft recipients who were HCMV PCR positive, but asymptomatic, was approximately one genome equivalent/1,000 cells and one genome equivalent/100 cells in three patients who had HCMV disease. The tissue HCMV DNA levels demonstrated in the lung transplant recipients in this study are also of the same order as those published for normal healthy patients using in situ hybridization techniques, of approximately one genome per 1,000 cells (29).
In the subjects who developed CMV disease (clinical and histopathologic evidence), the changes observed in CMV DNA viral load before and after disease development suggests that CMV viral load monitoring may be useful for individual patients. In Patient 3, who was a seronegative match for CMV and who developed CMV disease, CMV viral load increased prior to disease and decreased with ganciclovir treatment. In Patient 6, CMV disease developed within 1 mo of ceasing ganciclovir prophylaxis. Interestingly, the amount of donor tissue CMV DNA was very high for this subject, although the tissue CMV DNA levels fell to relatively low levels after the institution of a 10-wk course of Ganciclovir post-transplantation. Hence, these results suggest that direct viral load measurements of CMV are more meaningful than indirect, “all or none” measurements such as serology, even though CMV prophylaxis based on serologic testing has proven to be very useful (i.e., ganciclovir may be even better targeted using viral load measurements).
Semiquantitative PCR techniques have been used to distinguish between large differences in HCMV DNA load. Lee and coworkers (17) used semiquantitative PCR based on a visual scoring of PCR product band intensity, and they found that there was no level of HCMV DNA in the blood that predicted for HCMV disease in a group of renal transplant recipients, and that rises in HCMV DNA levels in any one individual predicted for the development of HCMV in only 50% of the patients. Other studies using only qualitative or semiquantitative PCR have also failed to show any clear difference between HCMV DNA levels in patients with HCMV disease and in stable patients (8, 16, 18). Reproducibility tests in such studies are often not reported, and it is therefore not clear whether random variation and noise in the detection system used may have confounded these results.
Gerna and coworkers (19) used a competitive PCR-based assay system to determine that HCMV DNA levels in the blood did not correlate with episodes of HCMV disease in lung transplant recipients. However, this assay included slot blot hybridization after PCR amplification, which increased the quantitative range of the assay system at the cost of making the assay more difficult and unable to discriminate less than 100-fold differences in HCMV levels. Relatively smaller differences like those demonstrated on TBB at 2 wk post-transplant may be important in predicting CMV disease. The failure of previous studies to identify a link between CMV load and disease may be due to the, at best, semiquantitative nature of the detection systems utilized.
The quantitative range for the assay system developed was of the order of 1,000 molecules, and although the internal standard competitor molecule concentration often needed to be titrated to find the range within which both PCR product bands could be detected (usually at least three PCR reactions for each quantitation), this was estimated to be less time-consuming and more accurate than the use of quantitative PCR techniques, which need post-PCR hybridization steps in their assay protocol (19).
The possibility that a gene sequence similar to HCMV, but from a different source (e.g., another Herpes virus), was being detected is unlikely given the specificity of the HCMV primer set and the lack of any significant binding of this primer set with the DNA polymerase gene sequences of the other Herpes viruses. Nevertheless, human Herpes virus 6 (HHV6) is one possibility that cannot be completely discounted given that the HCMV primers were not directly tested against this virus and that it also is a β-Herpes virus. Furthermore, HHV6 has been reported as a cause of pneumonitis in lung transplant recipients (30). The presence of HHV6 is also a possible explanation for the reported incidence of detectable HCMV DNA in the blood of 30 to 50% of HCMV seropositive patients despite being culture negative, although the presence of a transcription block prior to full virion assembly and release is more likely (31). Importantly, the increases in CMV DNA load that were observed in association with CMV disease (Patient 3), and the decreases in CMV viral loads associated with ganciclovir therapy (Patient 3 and 6) suggest that the CMV PCR signal that was amplified was a real signal. These results are also in agreement with recent reports that a correlation between CMV DNA levels in the blood and CMV disease is likely in subsets of immunocompromised subjects (10, 14).
There was a tendency for the HCMV DNA levels to be much greater in lung tissue compared with either the BAL cells or the white blood cells. We speculate that the lung allograft is a relatively privileged site of virus replication because virus-specific cytotoxic T-cells critical for virus elimination are MHC-restricted, and thus elimination of the virus from the allograft may be difficult (32). Larger numbers of patient specimens, however, will need to be tested to confirm this finding.
DNA polymerase is a product of early gene transcription in HCMV infection that occurs in the absence of prior protein synthesis. Hence, it is found in abortive as well as permissive infections (33). The antiviral effect of several cytokines is most pronounced on late gene transcription and nucleocapsid formation, with no effect on early gene transcription (34). Both these factors make HCMV DNA measurements a better marker for the early effects of HCMV on the host cell, particularly as early viral events may alter the intracellular environment and disturb cell function (35).
The availability of this assay will allow prospective, longitudinal studies to confirm the relevance of early detection of high HCMV DNA levels in lung tissue and whether subsequent rises in levels may be a means of detecting potential cases of clinical disease at an early stage. Such data may allow better definition of the indications of expensive and toxic ganciclovir prophylaxis after lung transplantation.
The writers would like to thank Professor Graham Brown, Dr. John Reeder, and the Walter and Eliza Hall Institute for the generous provision of practical assistance.
Supported by Alfred Hospital Heart–Lung transplant Service and Department of Respiratory Medicine Research Scholarship, the Australian Lung Foundation, Roche Australia, and the Rebecca L. Cooper Medical Research Foundation Ltd.
| 1. | Ho M.Epidemiology of cytomegalovirus infections. Rev. Infect. Dis.121990S701710 |
| 2. | Smyth R. L., Scott J. P., Borysiewicz L. K., Sharples L. D., Steward S., Wreghitt T. G., Gray J. J., Higenbottam T. W., Wallwork J.Cytomegalovirus infection in heart-lung transplant recipients: risk factors, clinical associations, and response to treatment. J. Infect. Dis.164199110451050 |
| 3. | Duncan, A. J., J. S. Dummer, I. L. Paradis, J. H. Dauber, S. A. Yousem, M. A. Zenati, R. L. Kormos, and B. P. Griffith. 1991. Cytomegalovirus infection and survival in lung transplant recipients. J. Heart Lung Transplant. 10(5, Pt. 1):638–644. |
| 4. | Matthey S., Nicholson D., Ruhs S., Alden B., Knock M., Schultz K., Schmuecker A.Rapid detection of respiratory viruses by shell vial culture and direct staining by using pooled and individual monoclonal antibodies. J. Clin. Microbiol.301992540544 |
| 5. | Emanuel D., Peppard J., Stover D., Gold J., Armstrong D., Hammerling U.Rapid immunodiagnosis of cytomegalovirus pneumonia by bronchoalveolar lavage using human and murine monoclonal antibodies. Ann. Intern. Med.1041986476481 |
| 6. | Flint, A., and T. S. Frank. 1994. Cytomegalovirus detection in lung transplant biopsy samples by polymerase chain reaction. J. Heart Lung Transplant. 13(1, Pt. 1):38–42. |
| 7. | Maurer, J. R., D. E. Tullis, M. Scavuzzo, and G. A. Patterson. 1991. Cytomegalovirus infection in isolated lung transplantations. J. Heart Lung Transplant. 10(5, Pt. 1):647–649. |
| 8. | Zipeto D., Revello M. G., Silini E., Parea M., Percivalle E., Zavattoni M., Milanesi G., Gerna G.Development and clinical significance of a diagnostic assay based on the polymerase chain reaction for detection of human cytomegalovirus DNA in blood samples from immunocompromised patients. J. Clin. Microbiol.301992527530 |
| 9. | Greenfield C., Sinickas V., Harrison L. C.Detection of cytomegalovirus by the polymerase chain reaction: a simple, rapid and sensitive non-radioactive method. Med. J. Aust.1541991383385 |
| 10. | Peiris J. S., Taylor C. E., Main J., Graham K., Madeley C. R.Diagnosis of cytomegalovirus (CMV) disease in renal allograft recipients: the role of semiquantitative polymerase chain reaction (PCR). Nephrol. Dial. Transplant.10199511981205 |
| 11. | Cunningham R., Harris A., Frankton A., Irving W.Detection of cytomegalovirus using PCR in serum from renal transplant recipients. J. Clin. Pathol.481995575577 |
| 12. | Wolf D. G., Spector S. A.Early diagnosis of human cytomegalovirus disease in transplant recipients by DNA amplification in plasma. Transplantation561993330334 |
| 13. | Buffone G. J., Frost A., Samo T., Demmler G. J., Cagle P. T., Lawrence E. C.The diagnosis of CMV pneumonitis in lung and heart/lung transplant patients by PCR compared with traditional laboratory criteria. Transplantation561993342347 |
| 14. | Drouet E., Coliman R., Michelson S., Fourcade N., Niveleau A., Ducerf C., Biobeaux A., Chevallier M., Denoyel G.Monitoring levels of human cytomegalovirus in blood after liver transplantation. J. Clin. Microbiol.331995389394 |
| 15. | Velzing J., Rothbarth P. H., Kroes A. C., Quint W. G.Detection of cytomegalovirus mRNA and DNA encoding the immediate early gene in peripheral blood leucocytes from immunocompromised patients. J. Med. Virol.421994164169 |
| 16. | Cagle P. T., Buffone G., Holland V. A., Samo T., Demmler G. J., Noon G. P., Lawrence E. L.Semi-quantitative measurement of cytomegalovirus DNA in lung and heart-lung transplant patients by in vitro DNA amplification. Chest10119929396 |
| 17. | Lee W. T., Antoszewska H., Powell K. F., Collins J., Doak P. B., Williams L. C., Munn S., Verran D., Croxson M. C.Polymerase chain reaction in detection of CMV DNA in renal allograft recipients. Aust. N.Z. J. Med.221992249255 |
| 18. | Gerna G., Zipeto D., Parea M., Revello M. G., Silini E., Percivalle E., Zavattoni M., Grossi P., Milanesi G.Monitoring of human cytomegalovirus infections in heart transplant recipients by determination of viraemia, antigenemia, and DNAemia. J. Infect. Dis.1641991488498 |
| 19. | Gerna G., Furione M., Baldanti F., Sarasini A.Comparative quantitation of human cytomegalovirus DNA in blood leukocytes and plasma of transplant and AIDS patients. J. Clin. Microbiol.32199427092717 |
| 20. | Schafer, P., R. W. Braun, K. Mohring, K. Henco, J. Kang, J. Wendland, and J. E. Kuhn. 1993. Quantitative determination of human cytomegalovirus target sequences in peripheral blood leukocytes by nested polymerase chain reaction and temperature gradient gel electrophoresis. J. Gen. Virol. 74(Pt. 12):2699–2707. |
| 21. | Zipeto D., Baldanti F., Zella D., Furione M., Cavicchini A., Milanesi G., Gerna G.Quantification of human cytomegalovirus DNA in peripheral blood polymorphonuclear leukocytes of immunocompromised patients by the polymerase chain reaction. J. Virol. Methods4419934555 |
| 22. | Hodinka, R. L., and H. M. Friedman. 1991. Human Cytomegalovirus. In A. Balows, W. J. Hausler, Jr., K. L. Herrmann, H. D. Isenberg, and H. J. Shadomy, editors. Manual of Clinical Microbiology, 5th ed. American Society for Microbiology, Washington, DC. 829–837. |
| 23. | Celi F. S., Zenilman M. E., Shuldiner A. R.A rapid and versatile method to synthesize internal standards for competitive PCR. Nucleic Acids Res.2119931047 |
| 24. | Siebert P. D., Larrick J. W.PCR MIMICS: competitive DNA fragments for use as internal standards in quantitative PCR. Biotechniques141993244249 |
| 25. | Lowry, R. W., E. Adam, C. Hu, N. S. Kleiman, B. Cocanougher, N. Windsor, J. N. Bitar, J. L. Melnick, and J. B. Young. 1994. What are the implications of cardiac infection with cytomegalovirus before heart transplantation? J. Heart Lung Transplant. 13(1, Pt. 1):122–128. |
| 26. | Smith K. L., Kulski J. K., Cobain T., Dunstan R. A.Detection of cytomegalovirus in blood donors by the polymerase chain reaction. Transfusion331993497503 |
| 27. | Barbara J. A., Garson J. A.Polymerase chain reaction and transfusion microbiology. Vox Sang.6419937381 |
| 28. | Trapnell, B. C. 1993. Quantitative evaluation of gene expression in freshly isolated human respiratory epithelial cells. Am. J. Physiol. 264 (3, Pt. 1):L199–L212. |
| 29. | Hendrix M. G., Daemen M., Bruggeman C. A.Cytomegalo-virus nucleic acid distribution within the human vascular tree. Am. J. Pathol.1381991563567 |
| 30. | Cone R. W., Hackman R. C., Huang M. L., Bowden R. A., Meyers J. D., Metcalf M., Zeh J., Ashley R., Corey L.Human herpes virus 6 in lung tissue from patients with pneumonitis after bone marrow transplantation. N. Engl. J. Med.3291993156161 |
| 31. | Benson J. D., Huang E. S.Recent progress in cytomegalovirus research. Virus Genes31990263276 |
| 32. | Rubin R. H.The indirect effects of cytomegalovirus infection on the outcome of organ transplantation. J.A.M.A.261198936073609 |
| 33. | Stamminger T., Fleckenstein B.Immediate-early transcription regulation of human cytomegalovirus. Curr. Top. Microbiol. Immunol.1541990319 |
| 34. | Lucin, P., S. Jonjic, M. Messerle, B. Polic, H. Hengel, and U. H. Koszinowski. 1994. Late phase inhibition of murine cytomegalovirus replication by synergistic action of interferon-gamma and tumour necrosis factor. J. Gen. Virol. 75(Pt. 1):101–110. |
| 35. | Grundy, J. E. 1991. Alterations of cellular proteins in human cytomegalovirus infection: potential for disease pathogenesis. Transplant. Proc. 23(Suppl. 3):38–41. |
