Jeurnal.°]'.Ge.nleral. V!~°/°g~ ..!!.99.6).'..77, 22.[.!.C2.2.75.'.....P.[!n!e.d!n .G.[ea!..B[i.ta!n................................................S H O R T C O M M U N I C A T I O N
Quantification of human herpesvirus 6 in immunocompetent persons and post-mortem tissues from AIDS patients by PCR D u n c a n A. Clark, ~ M o u n i r A i t - K h a l e d , ~ A n n e C. W h e e l e r , 1 I. M i c h a e l K i d d , 1 J a m e s E. M c L a u g h l i n , a M a r g a r e t A. J o h n s o n , 3 Paul D. G r i f f i t h s 1 a n d V i n c e n t C. E m e r y ~ Departments of Virology 1, Histopathology2 and Thoracic Medicine3, Royal Free Hospital School of Medicine, Rowland Hill Street, London NW3 2PF, UK
A quantitative competitive PCR assay for human herpesvirus 6 (HHV-6) was developed. Firstly, viral burden was determined in the blood of 25 healthy persons. Using 1 pg of DNA, the prevalence of HHV6 was 36°70 (9/25). Eight persons had viral loads of ~< 32 HHV-6 genomes/~tg DNA. The viral burden in the ninth individual was 1.2 x 106 HHV-6 genome copies/pg DNA, which remained constant over a period of 10 months. This demonstrates the persistence of a high HHV-6 load in the absence of apparent disease. Secondly, HHV-6 burden was determined in 1 O0 post-mortem tissues from seven AIDS patients and three controls. For all tissues combined, there was a statistically significant higher median viral load in AIDS patients (56 copies/pg DNA, range 0 - 4 3 3 2 1 ) compared to controls (10 copies/lzg DNA, range 0 - 4 2 3 ) (P = 0"04). The precision and reproducibility of this assay will allow hypotheses concerning the pathogenic potential of HHV-6 to be tested quantitatively.
Human herpesvirus 6 (HHV-6) is ubiquitous in the population and is a causative agent of febrile illness in children, particularly exanthem subitum (Yamanishi et al., 1988; Hall et al., 1994). The virus has also been associated with disease in immunocompromised persons. In bone marrow transplant patients, HHV-6 has been linked to poor marrow engraftment, encephalitis and pneumonitis (Cone eta]., 1993 b; Drobyski et a]., 1993, 1994). The role of HHV-6 either as a co-factor in human immunodeficiency virus (HIV) disease progression or as an opportunistic infection in AIDS is unknown, although a number of mechanisms have been proposed (Griffiths, 1992; Lusso & Gallo, 1995). PCR provides a highly sensitive and specific technique with which to investigate disease associations. The detection of Authorfor correspondence:Vincent C. EmeW. Fax +44 171 830 2854. e-mail
[email protected]
0001-4015 © 1996 SGM
HHV-6 DNA in serum or plasma by PCR has been used as a marker for active infection in blood (Huang et al., 1992; Secchiero et al., 1995a). Since HHV-6 persists in the host following primary infection, PCR methods which can identify clinically relevant viral replication are required. One approach is to use quantitative PCR (QPCR) to determine the actual number of viral genomes (viral load). Although unable to differentiate active infection from latency, such a measurement is likely to vary between persistence in normal individuals and during primary infection or episodes of reactivation, and may be greater in patients who are immunocompromised. The importance of viral burden in disease pathogenesis has been demonstrated for a number of viruses including human cytomegalovirus (HCMV) and HIV in studies utilizing QPCR assays (Connor et al., 1993; Fox eta]., 1995). We have developed an HHV-6 QPCR based on methodology used previously in our laboratory for HCMV (Fox et al., 1992). This system relies on the co-amplification of a control sequence with the test sample thereby compensating for both the presence of inhibitors and well-to-well variation in the thermal cycler. The control sequence is identical to the wild-type sequence amplified except for specific base changes that create a restriction endonuclease site within the amplicon. The control and target sequences are differentiated in postamplification analysis by digestion of PCR products with the appropriate restriction endonuclease and separation by PAGE. The important advantage of our QPCR assay is that the kinetics of amplification of control and target sequences should not differ since both are similar in size, base composition, and possess identical primer binding sites. All but one HHV-6 QPCR assays reported previously have lacked the advantages of co-amplification-based QPCR. Cone et a]. (1993a, b) and Fairfax et aI. (1994) visually compared autoradiograph bands from the PCR products of the test sample with externally amplified known standards following hybridization with a radiolabelled probe. Limiting dilution of the sample prior to PCR has also been used to estimate viral burden in brain tissue (Challoner et al., 1995). Only Secchiero et al. (1995 b) employed the co-amplification of an internal standard containing identical primer binding sequences, similar percentage G + C composition, but a non-homologous sequence of a different size. ~_27
1
105
,/
10 4
8
S
103
r = 0.999 (P 1000. For reactions using < 1000 control sequence copies, a nested PCR was used limiting the second
127;
round to 18 cycles. Depending on whether a nested approach was used, 3 ng of either primer B or primer C radiolabelled with a2p was included. Following Sinai digestion of PCR products the radiolabelled fragment of the control sequence was 131 bp or 67 bp for outer or nested PCR respectively. Subsequent to PAGE, gels were autoradiographed, and the intensity of sample and control bands determined by scanning densitometry (Shimadzu CS-9001PC scanner). Wild-type plasmid ranging from 25 to 50000 copies was co-amplified with different control sequence copy numbers (20, 102, 10 a, 10~) in triplicate. The regression line was calculated by the method of least squares (Fig. 1). Correlation analysis of the relationship between calculated and actual wild-type sequence copy numbers yielded a correlation co-efficient of 0"999 (P < 0"001). These results display the accuracy and suitability of the QPCR over this range of target copy number. The range of accurate quantification is the number of input control sequences-F 1 logarithm, i.e. with 100 control sequences, a test sample with between 10 to 1000 copies can be quantified accurately (data not shown). In each experiment a parallel PCR containing control sequence only was included to confirm complete digestion of amplified product. We have used this QPCR to determine HHV-6 viral load in clinical material, firstly by utilizing blood samples taken from 25 healthy volunteers. Initially, quantities of DNA extracted from whole blood nucleated cells (Wizard Genomic DNA Purification Kit, Promega) ranging from 0"1 lag to I"0 lag were tested by nested HHV-6 qualitative PCR and nine positive persons were found (Table l a). Subsequently, PCR positive samples that were below the threshold of the QPCR assay (estimated at < 10 copies) were given an arbitrary viral load of 5 genome copies. The HHV-6 genome copy numbers in eight of the nine positive blood samples ranged between 5 to 32 genome copies/lag DNA, equivalent to a range of 34 to 214 copies/106 cells. Cone et al. (1993a), using different QPCR methodology, calculated viral load in blood of healthy individuals to be between 4 to 4000 copies/106 cells. However, in that study, cell numbers were estimated by visually comparing the autoradiograph band of test samples against externally amplified known controls following amplification of a fl-globin sequence by PCR and hybridization with a radiolabelled probe. Also, Cone et al. (1993a) separated peripheral blood mononuclear cells from blood prior to DNA extraction which would enrich the cell population for those that harbour HHV-6. The viral load in volunteer 1 (1'2 x 106 HHV-6 genomes per lag DNA) was considerably higher than the other volunteers in our study. The high viral burden was not a transient event since eight blood samples taken over a period of 10 months all contained similar levels of HHV-6 (Table I b). This demonstrates that a high HHV-6 viral load can persist in blood in an apparently healthy individual. High numbers of HHV-6 genomes have been described previously, using Southern blot analysis, in persons with lymphoma and immune disorders and
Table 1. HHV-6 viral load in healthy volunteers positive by qualitative nested PCR (a)
(b)
Volunteer no
Input DNA qualitatively HHV-6 viral load positive (pg)* (9enomeslBgDNA
1
0"1
2
0"1
3
0"5
4 5
0"5 0"5
5% 5
6 7 8 9
1"0 1"0 1"0 1'0
5 5 5 5
Volunteer 1 Sampling time HHV-6 viral load (weeks) (genomes/l~g DNA)
See Table l b 32 13
0 1 4 6 10 14 36 45
1'2 x 0-8 x 1"i x 0"7 x 1'2 x I'0 x 1"4 x 1"5 x
106 106 106 106 105 106 106 106
* Samples tested from 0-1-1-0 gg. $ PCR positive samples below the threshold of quantification were allocated a load of 5 genome copies/gg DNA.
Table 2. HHV-6 viral load in post-mortem tissues from seven AIDS patients and three controls
HHV-6 viral load (genome copies/IJg DNA) Tissue Lymph node Spleen Brain Lung Heart Kidney Adrenal gland Oesophagus Duodenum Colon Liver
AIDS1
AIDS2
AIDS3
AIDS4
AIDS5
AIDS6
AIDS7
C1 *
C2
C3
1044 145 0 281 0 88 0 0 193 0 1620
310 485 0 165 400 43 321 305 NA 10 000 NA 2200
15 19 55 56 225 224 21 NA Na NA 89
5 0 0 358 0 49 49 251 378 0 0
0 0 0 143 86 25 0 0 99 617 49
248 20 0 118 0 28 8 NA NA NA 121
91 214 NAt 449 97 33 0 16 623 N,¢ 744
165 96 0 44 80 29 43 5 5 193 196
0 0 0 0 157 37 0 0 0 0 99
5 0 0 78 5 10 71 13 328 423 5
* C, control. Jr NA, tissue not available.
were associated with integration of HHV-6 sequences into chromosome 17 (Luppi et al., 1993; Torelli et al., 1995). To assess the association between HHV-6 infection and disease it will be important to monitor viral load longitudinally in individuals under investigation. Secondly, HHV-6 QPCR was used to determine viral burden in multiple post-mortem tissue samples from seven patients with AIDS and three controls (two accidental sudden deaths, one heart attack). Qualitative PCR studies have shown that HHV-6 is widely disseminated in post-mortem samples from patients who died from AIDS and, to a lesser extent, in controls (Corbellino et al., 1993). Active HHV-6 infection has been identified, using immunohistochemical methods, in lung,
liver, lymph node, spleen and kidney from AIDS patients (Knox & Carrigan, 1994). More recently, active infection was also identified in central nervous system tissues from AIDS patients, but not controls (Knox & Carrigan, 1995). D N A was extracted from each tissue b y a standard phenol-chloroform, ethanol precipitation method or with the Wizard Genomic D N A Purification Kit and i ~tg used per PCR reaction. The D N A from each tissue was co-amplified with three different control sequence copy numbers, i.e. 1000, 500, 100 and the HHV-6 genome copy number calculated from the PCR reaction closest to the point of equivalence. The results are shown in Table 2. In the three controls, 2 2 / 3 3 tissues were PCR positive, demonstrating disseminated HHV-6 infection.
227!
iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii i iiiiiiiiiiiiiiiiiii Likewise, the majority of the tissues from AIDS patients (48/67) were qualitatively HHV-6 PCR positive with lymph node, spleen, lung, kidney, and liver frequently positive. However, the difference in HHV-6 median viral load for all tissues between the AIDS patients and controls was statistically significant (P--0"040) using the Mann-Whitney 'U' test (AIDS patients median of 56 copies/,g DNA, range 0-43 321; controls median of 10 copies/l~g DNA, range 0-423). When examining individual tissues, only the lung showed a statistically significant higher median viral load in AIDS patients compared to controls (P < 0"05 ; AIDS patients median of 165 copies/l~g DNA, range 56-449; controls median 44 copies/~g DNA, range 0-78). The prevalence of HHV-6 and viral load has been reported to be low in the blood of HIV-infected persons with CD4 cell counts below 400 (median of 4 HHV-6 genomes/~g DNA) (Fairfax et al., 1994). Thus, the picture in blood contrasts with our findings of widespread HHV-6 prevalence and a higher viral load (median 56 HHV-6 genome copies/l~g DNA) in post-mortem tissues and suggests that results in peripheral blood may underestimate the organ burden of HHV-6. Knox & Carrigan (1994) reported differing levels of HHV6 antigen positive cells, the majority of which were macrophages and lymphocytes, in AIDS post-mortem tissues. Interestingly, a number of tissues that we found to be PCR positive showed histological evidence of infiltrating mononuclear cells. The highest load was identified in the kidney tissue (AIDS2) although no histopathologica] abnormalities were noted. Most other tissues harboured considerably lower HHV-6 viral loads. The viral loads detected in lymph nodes were similar to those reported recently in a study that used QPCR to determine HHV-6 burden in lymph nodes from different origins including AIDS patients (Secchiero et al., 1995 b). Cone et al. (1993 b) using QPCR found increased levels of HHV-6 DNA in lung tissue from bone marrow transplant patients which correlated with a decreased risk of fatal pneumonitis, increased severity of graft versus host disease and idiopathic pneumonitis. Knox & Carrigan (1994) suggested that the high density of HHV-6-infected cells in the lung of one AIDS patient could be responsible for fatal pneumonitis. However, if HHV-6 interstitial pneumonitis is an immunopathological condition, as has been proposed for HCMV (Grundy et al., 1987), it would require a functional immune system, which is defective in AIDS patients with low CD4 cell counts. This has to be taken into consideration when analysing the relationship of HHV-6 viral load and pneumonitis in such a patient group (Griffiths et al., 1994). Although we found higher HHV-6 loads in the lungs of AIDS patients compared to controls, no cases of interstitial pneumonitis occurred. In conclusion, we have developed a sensitive and accurate HHV-6 QPCR assay and shown that the majority of healthy individuals have low levels of HHV-6 in the blood. In addition, although HHV-6 infection was highly prevalent in organs of
~_27~
both AIDS patients and controls at death, there was a significantly higher HHV-6 load in the AIDS patients. Extension of these studies will enable the pathological consequences of HHV-6 infection to be assessed. We thank Dr M. C. Atkins for help in processing post-mortem samples. This work was funded by the National Institutes of Health USA (Grant number AI33389).
References Challoner, P. B., Smith, K. T., Parker, J. D., MacLeod, D. L., Coulter, S. N., Rose, T. H., Schultz, E. R., Bennett, J. L., Garber, R. L., Chang, M., Schad, P.A., Stewart, P.M., Nowinski, R.C., Brown, J.P. & Burmer, G. C. (1995). Plaque-associated expression of human herpes-
virus 6 in multiple sclerosis, Proceedings of the National Academy of Sciences, USA 92, 7440--7444. Cone, R. W., Huang, M.-L. W., Ashley, R. & Corey, L. (1993o). Human
herpesvirus 6 DNA in peripheral blood cells and saliva from immunocompetent individuals.Journal of Clinical Microbiology 31, 1262-1267. Cone, R. W., Hackman, R. C., Huang, H.-L. W., Bowden, R. A., Meyers, J. D., Hetcaff, M., Zeh, J., Ashley, R. & Corey, L. (1993b). Human
herpesvirus 6 in lung tissue from patients with pneumonitis after bone marrow transplantation. New England Journal of Medicine 329, 156-161. Connor, R. I., Mohri, H., Cao, Y. & Ho, D. D. (1993). Increased viral
burden and cytopathicity correlate temporally with CD4+ T-lymphocyte decline and clinical progression in human immunodeficiency virus type 1-infectedindividuals.Journal of Virology 67, 1772-1777. Corbellino, M., Lusso, P., Gallo, R.C., Parravicini, C., Galli, H. & Moroni, M. (1993). Disseminated human herpesvirus 6 infection in
AIDS. Lancet 342, 1242. Drobyski, W. R., Dunne, W. M., Burd, E. M., Knox, K. K., ASh, R. C., Horowitz, M.M., FIomenberg, N. & Carrigan, D. R. (1993). Human
herpesvirus-6 (HHV-6) infection in allogeneic bone marrow transplant recipients: evidence of a marrow-suppressive role for HHV-6 in vivo. Journal of Infectious Diseases 167, 735-739. Drobyski, W. R., Knox, K. K., Majewski, D. & Carrigan, D. R. (1994).
Brief report: fatal encephalitis due to variant B human herpesvirus-6 infection in a bone marrow-transplant recipient. New England Journal of Medicine 330, 1356--1360. Fairfax, M. R., Schacker, T., Cone, R.W., Collier, A.C. & Corey, L.
(1994). Human herpesvirus 6 DNA in blood cells of human immunodeficiencyvirus-infectedmen: correlationof high levels with high CD4 cells counts. Journal of Infectious Diseases 169, 1342-1345. Fox, J. C., Griffiths, P. D. & Emery, V. C. (1992). Quantification of human cytomegalovirus DNA using the polymerase chain reaction. Journal of General Virology 73, 2405-2408. Fox, J.C., Kidd, |. M., Griffiths, P.D., Sweny, P. & Emery, V.C. (1995). Longitudinalanalysisof cytomegalovirusload in renal transplant recipients using a quantitative polymerase chain reaction: correlation with disease. Journal of General Virology 76, 309-319. Griffiths, P.D. (1992). Studies to define viral cofactors for human immunodeficiencyvirus. Infectious Agents and Disease 1, 237-244.
Griffiths, P. D., Emery, V. C., Lee, C., Johnson, H. A. & McLaughlin, J. (1994). HHV-6 in AIDS. Lancet 343, 1104-I105.
Grundy, J.E., Shanley, J.D. & Griffiths, P.D. (1987). Is cyto-
megalovirus interstitial pneumonitis in transplant recipients an immunopathological condition? Lancet it, 996--999.
Hall, C. B., Long, C. E., Schnabel, K. C., Caserta, H. T., Mclntyre, K. M., Costanzo, M. A., Knott, A., Dewhurst, S., Insel, R. A. & Epstein, L. G. (1994). Human herpesvirus-6 infection in children. A prospective study of complications and reactivation. New EnglandJournal of Medicine 331, 432-438.
Secchiero, P., Carrigan, D. R., Asano, Y., Benedetti, L., Crowley, R. W., Komaroff, A. L., Gallo, R. C. & Lusso, P. (19950). Detection of human herpesvirus 6 in plasma of children with primary infection and immunosuppressed patients by polymerase chain reaction. Journal of Infectious Diseases 171, 273-280.
Huang, L.-M., Kuo, P.-F., Lee, C.-Y., Chen, J.-Y., Liu, H.-Y. & Yang, C.-S. (1992). Detection of human herpesvirus-6 DNA by polymerase chain reaction in serum or plasma. Journal of Medical Virology 38, 7-10.
Secchiero, P., Zella, D., Crowley, R.W., Gallo, R.C. & Lusso, P.
Knox, K.K. & Carrigan, D.R. (1994). Disseminated active HHV-6 infections in patients with AIDS. Lancet 343, 577-578.
Torelli, G., Barozzi, P., Marasca, R., Cocconcelli, P., Herelli, E., Ceccherini-Nelli, L., Ferrari, S. & Luppi, H. (1995). Targeted integration
Knox, K. K. & Carrigan, D. R. (1995). Active human herpesvirus (HHV6) infection of the central nervous system in patients with AIDS. Journal
of Acquired Immune Deficiency Syndromes and Human Retrovirology 9, 69-73.
Luppi, M., Barozzi, P., Marasca, R., Ceccherini-Nelli, L. & Torelli, G. (1993). Characterization of human herpesvirus 6 genomes from cases of latent infection in human lymphomas and immune disorders. Journal of Infectious Diseases 168, 1074-1075.
Lusso, P. & Gallo, R.C. (1995). Human herpesvirus 6 in AIDS. Immunology Today 16, 67-71. Murphy, G. & Kavanagh, T. (1988). Speeding up the sequencing of double-stranded DNA. Nucleic Acids Research 16, 5198.
(1995 b). Quantitative PCR for human herpesviruses 6 and 7. Journal of CJinicaIMicrobioJogy 33, 2124-2130.
of human herpesvirus 6 in the p arm of chromosome 17 of human peripheral blood mononuclear cells in vivo. Journal of Medical Virology 46, 178-188.
Wakefield, A. J., Fox, J. D., Sawyerr, A. M., Taylor, J. E., Sweenie, C. H., Smith, H., Emery, V. C., Hudson, H., Tedder, R. S. & Pounder, R. E. (1992). Detection of herpesvirus DNA in the large intestine of patients with ulcerative colitis and Crohn's disease using the nested polymerase chain reaction. Journal of Medical Virology 38, 183-190.
Yamanishi, K., Okuno, T., Shiraki, K., Takahashi, H., Kondo, T., Asano, Y. & Kurata, T. (1988). Identification of human herpesvirus-6 as a causal agent for exanthem subitum, Lancet i, 1063-1067. Received 20 March 1996; Accepted 20 May 1996
!27~
