Detection of human cytomegalovirus, human herpesvirus type 6 and human herpesvirus type 7 in urine specimens by multiplex PCR

Journal of Infection (2003) 47, 59–64

www.elsevierhealth.com/journals/jinf

Detection of human cytomegalovirus, human herpesvirus type 6 and human herpesvirus type 7 in urine specimens by multiplex PCR Ahmed M. Ashshia, Paul E. Klapperb,1, Robert J. Coopera,* a

University Virology, Laboratory Medicine Academic Group, School of Medicine, The University of Manchester, Oxford Road, Manchester M13 9WL, UK b Clinical Virology Laboratory, 3rd Floor Clinical Sciences Building, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK KEYWORDS Neonates; AIDS; In utero infection

Summary Objectives. To develop a sensitive multiplex PCR to detect HCMV, HHV6 and HHV7, to test this PCR on urine specimens sent to the virus diagnostic laboratory and on stored urine samples from HIV-positive patients and their HIV-negative partners and to compare the sensitivity of the multiplex PCR with the diagnostic laboratory’s routine service for the detection of HCMV. Study design. Primers specific for each of the three viruses were combined in a multiplex PCR that was then optimised for sensitivity. This PCR was applied prospectively to 413 unselected routine urine specimens over a 1 year period and retrospectively to 258 urine specimens from 63 HIV-positive patients and 10 HIVnegative partners. Methods. In the prospective study, the multiplex PCR detected 40 specimens positive for HCMV alone, 10 for HHV6, 3 for HHV7 and 3 with a dual infection of HCMV and HHV6. The sensitivity for HCMV was 93.5% by multiplex PCR compared to 28.3% by culture. HHV6 DNA was detected in 6 neonates (2 –21 days) and HHV7 DNA in 2 neonates (4 and 20 days). In the retrospective study of HIV patients, HCMV was the most commonly detected virus (55.6%) compared to HHV6 (7.9%) and HHV7 (4.8%). Conclusions. The multiplex PCR was significantly more sensitive than non-DNA based procedures for the detection of HCMV. Urine may be a useful non-invasive specimen for the detection of HHV6 and HHV7 and their presence in neonates suggest perinatal transmission or the possibility of in utero infection. Q 2003 The British Infection Society. Published by Elsevier Science Ltd. All rights reserved.

Introduction Two new herpesviruses, HHV6 and HHV7 appear to be closely related to HCMV1 and may be found in *Corresponding author. Tel.: þ44-161-276-8844; Fax: þ 44161-276-8840. E-mail address: [email protected] 1 Public Health Laboratory, Bridle Path, Yorks Road, Leeds LS15 7T, UK.

similar clinical situations. HCMV is a common clinical problem in immunocompromised patients and is a well-established cause of congenital infection. HHV6 infects the majority of children by the age of 2 years.2 In some, the primary infection results in exanthem subitum3 but a number of other clinical manifestations including convulsions and encephalitis have been reported.4,5 HHV7 appears to infect children later than HHV66

0163-4453/03/$ - see front matter Q 2003 The British Infection Society. Published by Elsevier Science Ltd. All rights reserved. doi:10.1016/S0163-4453(03)00057-4

60

and has also been reported to cause exanthem subitum.7 The full spectrum of disease in children caused by HHV7 has not yet been established. HHV6 has also been found in a variety of other disease manifestations, particularly in the immunocompromised.8 In this study, a multiplex PCR developed to detect the DNA of all three viruses was used to screen urine samples from a variety of patients but particularly young children. Urine is the most common clinical specimen sent to the virus diagnostic laboratory for HCMV detection by culture or DEAFF test. This has provided the opportunity to explore these specimens for the three viruses.

Methods Primers: Three sets of primer pairs were used. The first set, specific for HCMV9 , was 233C – 50 TGCAGTTTGGTCCCTTAAAG 3 0 and 724C – 5 0 AAGAATCCTCACCTGGCTTA 3 0 that amplify a 171 bp sequence of UL32, the large structural phosphoprotein. The second set, specific for HHV6,10 was H6-6-50 AAGCTTGCACAATGCCAAAAAACAG 30 and H6-7-50 CTCGAGTATGCCGAGACCCCTAATC 30 that amplify a 223 bp sequence from U67. The final set, specific for HHV71, was HV10-50 CAGAAATGATAGACAGATGTTGG 30 and HV11-50 TAGATTTTTTGAAAAAGATTTAATAAC 30 that amplify a 124 bp sequence from U10. Multiplex PCR: The specific primers for HCMV, HHV6 and HHV7 were combined in a PCR using Amplitaq Golde (Perkin-Elmer). The multiplex mixture contained 10 mM tris/HCl pH 8.3, 50 mM KCl, 2.2 mM MgCl2, 200 mM of each dNTP, 0.1 mM of each HHV6 primer (H6-6, H6-7), 0.2 mM of each HCMV primer (233C, 724C), 0.3 mM of each HHV7 primer (HV10, HV11), 2.5 U of Amplitaq Gold DNA polymerase and 5 ml of sample DNA or sterile distilled water, as contamination control, in a final volume of 50 ml. One cycle of 94 8C/9.9 min, 55 8C/1.0 min and 72 8C/1.0 min was followed by 40 cycles at the same temperatures for 1.0 min each using a Techne Ori-Block PHC-1 thermal cycler. Reaction products were analysed on 6% polyacrylamide gels and stained with ethidium bromide before viewing on a transilluminator. Anti-contamination measures were as previously described11,12 and included separate rooms for preparation of the reaction mixtures, for preparation and addition of DNA extracts and for product analysis. Plugged pipette tips were used throughout. Specimens: Urine specimens (413) in an equal volume of transport medium (50% sorbitol, 5% fetal

A.M. Ashshi et al.

calf serum, 4.4% NaHCO3, 50 mg/ml gentamycin, 5 mg fungizone) sent for routine diagnostic virology over a 1 year period were assayed by multiplex PCR. All were from patients with various clinical disorders including bone marrow and renal transplant recipients and babies with developmental problems. A further 239 urine specimens were from HIV-positive patients or their HIV-negative partners which had been collected for an unrelated study over a period of 27 months and stored frozen at 2 20 8C since 1993. DNA extraction for PCR: All urine specimens were extracted with Hybaid Recoverye Amplification Reagent (Hybaid Ltd, Middlesex, UK). A 5 ml sample of urine was mixed with 15 ml of reagent and heated in a microwave oven for 7 min. A 5 ml aliquot of this extraction mixture was then added directly to the multiplex PCR. Control DNA: The HindIII restriction fragment J of HCMV (AD169) DNA cloned in pAT15313 and containing the viral phosphoprotein gene, and the HindIII restriction fragment pHD5 of HHV6 DNA14 were propagated and purified with the Qiagen Plasmid Maxi Kit (Qiagen Ltd, Crawley, UK) as described by the manufacturer. The plasmid concentrations were estimated spectrophotometrically. HHV7 infected cell DNA was purchased from Autogen Bioclear UK Ltd, Wiltshire, UK.

Results Urine specimens. The multiplex PCR was applied prospectively to 413 routine urine specimens and retrospectively to 258 urines from 63 HIV-positive patients and 10 HIV-negative partners (1–9 samples/ patient). An example of the resulting gels is shown in Fig. 1, where, 8 specimens are from HIV-positive patients and 6 specimens are from the prospective study. The gel includes positive examples of HHV7, HHV6, HCMV and a dual infection with both HHV6 and HCMV. Overall, in the prospective study, the multiplex PCR detected 40 specimens positive for HCMV alone, 10 for HHV6, 3 for HHV7 and 3 samples with a dual infection of HCMV and HHV6 from the total of 413 urines tested. In the retrospective study of HIV-positive patients, HCMV was the most commonly detected virus (35/63: 55.6%) with HHV6 (5/63: 7.9%) and HHV7 (3/63: 4.8%) much less frequently encountered. Most of the HCMV was detected in those patients with a CD4þ count of , 200/mm3 (Table I) and were also the only category of patients to show evidence of dual infections. Sensitivity and specificity of the multiplex PCR. The primers are highly specific and in single PCRs do

Detection of HCMV, HHV6 & HHV7 by multiplex PCR

61

Figure 1 Multiplex PCR of urine specimens. Lanes 1– 8 are from HIV seropositive patients and lanes 9–14 are routine diagnostic specimens. Lane 1 is positive for HHV7 (124 bp), lane 5 for HHV6 (223 bp) and lane 6 for HCMV (171 bp). Routine urine samples show a positive HCMV in lane 10, HHV6 in lane 12 and dual infection with HHV6 and HCMV in lane 14. The remaining numbered lanes are negative for all three viral DNAs. L ¼ 1 kb ladder, P ¼ positive control (103 copies each of HCMV and HHV6 plasmid DNA and 1023 dilution of HHV7-infected cell DNA), C ¼ contamination control.

not amplify DNA from heterologous human herpesviruses. With all three target DNAs present the multiplex PCR could detect between 60 and 600 copies of the HHV6 plasmid, between 40 and 400 copies of the HCMV plasmid and a dilution of 1026.5 for HHV7-infected cell DNA. This compares with 10 copies, 10 copies and a 1026.5 dilution, respectively, when a single target DNA was included in the multiplex PCR. HHV6 and HHV7 are not part of the diagnostic laboratory’s routine service but the performance of the multiplex PCR in detecting HCMV was compared to virus isolation and/or DEAFF test. In comparison to virus isolation, 10 samples were positive by both tests, 33 were positive by multiplex PCR alone, three were positive by virus isolation alone and 367 were negative by both tests. Assuming all positive tests are true positives, the sensitivity of the multiplex PCR for

HCMV was 93.5% (43/46) whereas culture was only 28.3% (13/46) ( p , 0.00001). Of the 413 specimens, only 134 were tested by DEAFF. Of these, three were positive by both tests, 14 by multiplex PCR alone, none by DEAFF alone and 117 were negative by both tests. Again, assuming all positives were true positive, multiplex PCR was 100% (17/17) sensitive compared to 17.6% (3/17) for DEAFF ( p , 0.001). Clinical information. Clinical information is only available for some of the patients positive for any of the three viruses and all were less than 2 years old. Table II summarises the clinical details of these patients with the results of the multiplex PCR. A significant proportion (13/22) of these patients are neonates. HCMV was detected in 6 neonates (2–19 days) with most having episodic staring, fitting or growth retardation. HHV6 DNA was also detected in 6 neonates (2–21 days). One, (patient 9), had

Table I Prevalence of HCMV, HHV6, and HHV7 DNA in HIV-positive and HIV-negative partners HIV status

CD4þ/mm3

No. of patients

No. of HCMV þ (%)

No. of HHV6 þ (%)

No. of HHV7 þ (%)

No. of dual HCMV/HHV6 þ (%)

No. of dual HCMV/HHV7 þ (%)

HIV þ HIV þ HIV 2

, 200 . 200 ND

31 32 10

22(71.0) 9(28.1) 1(10.0)

3(9.7) 1(3.1) 1(10.0)

0(0.0) 1(3.1) 0(0.0)

1(3.2) 0(0.0) 0(0.0)

2(6.5) 0(0.0) 0(0.0)

62

A.M. Ashshi et al.

Table II Clinical information on patients aged less than 2 years Patient

Age

Sex

Virus

Clinical information

1 2 3 4

7 days 14 days 15 days 16 days

M F M F

HCMV HCMV HCMV HCMV

5 6 7

19 days 8 months 5 days

F F M

HCMV HCMV HHV6

8 9

13 days 14 days

F M

HHV6 HHV6

10 11 12 13

15 days 21 days 5 weeks 5 months

F M F M

HHV6 HHV6 HHV6 HHV6

14

11 months

M

HHV6

15 16 17 18

18 months 19 months 2 days 12 months

M M F M

HHV6 HHV6 HCMV and HHV6 HCMV and HHV6

19

12 months

M

HCMV and HHV6

20 21 22

4 days 20 days 7 months

F F F

HHV7 HHV7 HHV7

Special care baby unit Episodic staring, fitting and twitching Retarded growth Episodes of staring and fitting, neurological abnormality. Not feeding well No information Fever, enlarged spleen, staring episodes Premature (29 weeks). Small for dates. Discharged after 9 days Jaundice Emergency admission, extremely jittery and tremulous for 7 days, afebrile, no rash No information No information Erythematous macular rash, afebrile on admission Admission with wheezy symptoms, bulging fontanelle, fever of 38 8C, no rash, premature delivery, head growth normal Raised temperature with slight rash on chest (persisted for 24 h), described as viral exanthem similar to measles, screaming and irritable on admission Febrile convulsions Fit, high temperature, febrile convulsion, cough Premature (26 weeks). Small for dates. Died, no post mortem Congenital CMV. Regular detection of HCMV in urine. No symptoms associated with HHV6 Congenital CMV. Small head, flattish occipitally, prominent eyes. No symptoms associated with HHV6 High temperature Neuroblastoma, died at 20 day Cardiology case, persistent airways obstruction, adenovirus type 2 isolated

neurological symptoms but was afebrile and had no rash. Two patients (7 and 17) were premature (29 and 26 weeks) and small for their dates but had very different outcomes. Patient 7 was well enough to be discharged after 9 days but patient 17 with a dual infection of HHV6 and HCMV died. HHV7 was found in 2 neonates, one with a high temperature (patient 20; 4 days) and the other (patient 21; 20 days) with neuroblastoma who died at 20 days. Among the older babies, HHV6 was detected in 5 patients (12 – 16; 5 weeks – 19 months) and all but one (12) had a fever, two (15 and 16) had convulsions and two (12 and 14) had a rash. Two 12-month-old patients (18 and 19) had dual infections with HCMV and HHV6. Both were known to have congenital CMV that was regularly detected in urine but there was no evidence of symptoms associated with the transient HHV6 detection.

Discussion Relatively few studies have tested urine samples for the presence of HHV6 and HHV7. Of these, the data

has been conflicting with some studies suggesting the viruses are not present15 – 17 whereas others have reported their detection.18 HHV6 has also been detected in the urine of immunocompromised patients e.g. transplant recipients19,20 or HIV seropositive individuals.21 The usual sample used for the detection of HHV6 and 7 is peripheral blood, but the most appropriate specimen has yet to be established. Although only 13/413 (3.1%) of the routine urine samples were positive for HHV6 and 3/413 (0.7%) for HHV7, the specimens were unselected and many would come from conditions unconnected to these viruses. Urine is a noninvasive specimen and its collection is much less traumatic than blood for very young children. Although the majority of HHV6 infections are thought to occur between 6 months and 2 years, in this study, 8/13 were younger than 6 months and six were less than 28 days including one patient 5 days old and 1 of only 2 days. It is also worth noting that 2/3 HHV7-positive cases were aged 4 days and 20 days. Whether any of these originated as intrauterine infections has not been determined. HHV6 DNA in cord blood of 5/305 (1.6%) of babies born to

Detection of HCMV, HHV6 & HHV7 by multiplex PCR ostensibly healthy mothers has been detected,22 suggesting intrauterine infection had occurred. Also,23 detected HHV6 antigen in abortive villous tissue from two patients with spontaneous abortion. In our laboratory, HHV6 has been detected in two cases of fetal hydrops24 and the possibility of in utero infection is therefore an area where, further investigations are warranted. Only two patients (12 and 14) presented with a rash, one of who may have been in the later stages of exanthem subitum as this patient was afebrile at the time of admission. Two patients had neurological symptoms, complications known to be linked to HHV6 infections.5 Two patients (5 and 6) have congenital CMV and HHV6 was detected in their urine during the routine monitoring for HCMV. This could indicate primary asymptomatic infection with HHV6 that, under normal circumstances, would have gone unnoticed. In urine from HIV seropositive patients, HCMV was much more common than either HHV6 or HHV7, a result confirming that in a previous report.21 The multiplex PCR for HCMV, HHV6 and HHV7 is a useful tool for detecting the viral DNA in urine samples. It is significantly more sensitive for the detection of HCMV than the more traditional isolation or DEAFF tests. The test has demonstrated the presence of HHV6 and HHV7 in neonates. Although this may be the result of perinatal transmission, it also raises the possibility that in utero infection by these viruses may be more common than previously thought. The test has detected three cases of dual infections with HCMV and HHV6. This is not due to cross over amplification between these herpesviruses because the primers are highly specific. One of the three cases was a neonate who died, although the relevance of the dual infection to this outcome is unknown.

Acknowledgements We are grateful to Dr S. Khoo for access to his collection of urine specimens from HIV-positive patients and their HIV-negative partners.

References 1. Berneman ZN, Ablashi DV, Li G, et al. Human herpesvirus 7 is a T-lymphotropic virus and is related to, but significantly different from, human herpesvirus 6 and human cytomegalovirus. Proc Natl Acad Sci USA 1992;89: 10552—10556. 2. Lusso P. Human herpesvirus 6. Antiviral Res 1996;31:1—21.

63

3. Yamanishi K, Okuno T, Shiraki K, et al. Identification of human herpesvirus 6 as a causal agent for exanthem subitum. Lancet 1998;1:1065—1067. 4. Kimberlin DW, Whitley RJ. Human herpesvirus-6: neurologic implications of a newly described viral pathogen. J Neurovirol 1998;4:474—485. 5. Yoshikawa T, Asano Y. Central nervous system complications in human herpesvirus-6 infection. Brain Develop 2000;22: 307—314. 6. Tanaka-Taya K, Kondo T, Mukai T, et al. Seroepidemiology study of human herpesvirus-6 and 7 in children of different ages and detection of these two viruses in throat swabs by polymerase chain reaction. J Med Virol 1996;48: 88—94. 7. Tanaka K, Kondo T, Torigoe S, Okada S, Mukai T, Yamanishi K. Human herpesvirus-7: another causal agent for roseola (exanthem subitum). J Pediatr 1994;125:1—5. 8. Clark DA. Human herpesvirus 6. Rev Med Virol 2000;10: 155—173. 9. McElhinney LM, Cooper RJ, Morris DJ. Multiplex polymerase chain reaction for human herpesvirus 6, human cytomegalovirus and human b-globin DNA. J Virol Meth 1995;53: 223—233. 10. Gopal MR, Thomson BJ, Fox J, Tedder RS, Honess RW. Detection by PCR of HHV-6 and EBV DNA in blood and oropharynx of healthy adults and HIV-seropositives. Lancet 1990;335:1598—1599. 11. Morris DJ, Bailey AS, Cooper RJ, et al. Polymerase chain reaction for rapid detection of ocular adenovirus infection. J Med Virol 1995;46:126—132. 12. Turner PC, Bailey AS, Cooper RJ, Morris DJ. The polymerase chain reaction for detecting adenovirus DNA in formalinfixed, paraffin-embedded tissue obtained post mortem. J Inf 1993;27:43—46. 13. Oram JD, Downing RG, Akrigg A, et al. Use of recombinant plasmids to investigate the structure of the human cytomegalovirus genome. J Gen Virol 1982;59:111—129. 14. Martin ME, Thomson BJ, Honess RW, et al. The genome of human herpesvirus 6: maps of unit-length and concatemeric genomes for nine restriction endonucleases. J Gen Virol 1991;72:157—168. 15. Suga S, Yazaki T, Kajita Y, Ozaki T, Asano Y. Detection of human herpesvirus 6 DNAs in samples from several body sites of patients with exanthem subitum and their mothers by polymerase chain reaction assay. J Med Virol 1995;46: 52—55. 16. Asano Y, Suga S, Yoshikawa T, Yazaki T, Uchikawa T. Clinical features and viral excretion in infant with primary human herpesvirus 7 infection. Pediatrics 1995;95: 187—190. 17. Pozo F, Tenorio A. Detection and typing of lymphotropic herpesviruses by multiplex polymerase chain reaction. J Virol Meth 1999;79:9—19. 18. Suga S, Yoshikawa T, Kajita Y, Ozaki T, Asano Y. Prospective study of persistence and excretion of human herpesvirus-6 in patients with exanthem subitum and their parents. Pediatrics 1998;102:900—904. 19. Wilborn F, Brinkmann V, Schmidt CA, Neipel F, Gelderblom H, Siegert W. Herpesvirus type 6 in patients undergoing bone marrow transplantation: serologic features and detection by polymerase chain reaction. Blood 1994;83: 3052—3058. 20. Ratnamohan VM, Chapman J, Howse H, et al. Cytomegalovirus and human herpesvirus 6 both cause viral disease after renal transplantation. Transplantation 1998;66: 877—882. 21. Gautheret-Dejean A, Aubin JT, Poirel L, et al. Detection of

64

human betaherpesvirinae in saliva and urine from immunocompromised and immunocompetent subjects. J Clin Microbiol 1997;35:1600—1603. 22. Adams O, Krempe C, Ko ¨gler G, Wernet P, Scheid A. Congenital infections with human herpesvirus 6. J Inf Dis 1998;178:544—546.

A.M. Ashshi et al.

23. Ando Y, Kakimoto K, Ekuni Y, Ichijo M. HHV-6 infection during pregnancy and spontaneous abortion. Lancet 1992; 340:1289. 24. Ashshi AM, Cooper RJ, Klapper PE, Al-Jiffri O, Moore L. Detection of human herpes virus 6 in fetal hydrops. Lancet 2000;355:1519—1520.