Stephen A. Spector,1'3 Karen HsIa,' Frank Denaro,' and Deborah H.Spector2'3 ...... Pellegrino MG, Lewin M, Meyer WA, et al. A sensitive solution hybridization ...
CLIN.CHEM.35/8, 1581-1587 (1989)
Use of Molecular Probes to Detect Human Cytomegalovirusand Human Immunodeficiency Virus Stephen A. Spector,1’3 Karen HsIa,’ Frank Denaro,’ and Deborah H. Spector2’3
Human cytomegalovirus (HCMV) and human immunodeficiencyvirus(HIV) cause severedisease.The identificationof these viruses in clinical specimens and understandingthe progression of infection and diseases relating to HCMVand HIV are essential to develop effective means for treatment and prevention. Here we describe the application of molec-
ular probes to the diagnosisand pathogenesisof HCMV and HIV. In situ hybridizationand the amplificationprocedureof polymerase chain reaction are used to detect both viruses; these techniques have provided important information regarding the pathogenesis of HCMVand HIV. A new technique,
target cycling, may also prove useful for the detection of viruses by enriching for target sequences. The continued applicationof molecular probes to pathogeneticstudies of HCMV and HIV promisesto further our knowledge of these viruses, and of their interaction. AddItionalKeyphrases:hybridization lion
target cycling
chain reac-
AIDS
The promise of molecular biology to revolutionize viral diagnosis and to supplant tissue culture for routine virus identification has, to date, not been realized. There are several reasons why detection of viral nucleic acids is not routinely performed in most diagnostic laboratories. They include: (a) work-intensive and cumbersome sample preparation; (b) reliance on isotopic detection systems; (c) nonspecific hybridization, which can complicate interpretation of results; (d) difficulty in quantification; and (e) the closeminded nature of the assay, which probes for a specific pathogen but will fail to identify an unsuspected virus. Many of these limitations have been the focus of considerable research, and recent advances promise to make nucleicacid-based diagnostic procedures more widely available (for reviews of rapid viral diagnostic procedures, see 1-3). In contrast to the slow, broad application of hybridization procedures in the area of viral diagnostics, nudeic acid probes have become essential in studies of viral pathogenesis (4). In particular, application of in situ hybridization techniques has contributed greatly to our understanding of viral latency and the progression of viral infections associated with a broad range of human and animal viruses (4-10). Here we will focus on our work relating to the diagnosis and pathogenesis of human cytomegalovirus (HcMv) and human immunodeflciency virus (H1V) infections.4 The ob-
jectives
for the development of nucleic acid hybridization tests must be viewed in the context of other diagnostic procedures available for these two viruses and of the critical questions that relate infection with disease etiology. For example, HCMV is known to cause a broad range of diseases: cytomegalic inclusion disease in newborn infants, perinatal infection, Epstein-Barr-virus-negative infectious mononucleosis, pneumonia, retinitis, colitis, encephalitis, and other diseases in immunocompromised patients, particularly organ-transplant recipients and patients with the acquired immunodeflciency syndrome (AIDS). Establishing HCMV as the etiological agent of specific clinical entities may at times be difficult. The diagnosis of HCMV retinitis is considered straightforward by the experienced ophthalmologist, who identifies an acute retinitis with either a perivascular yellow-white retinal lesion frequently associated with hemorrhage or a focal white granular infiltrate in a patient with AIDS (11). Similarly, the diagnosis of congenital HCMV can be easily established in an infant less than three weeks old who is culturepositive for the virus. However, in many of the other conditions associated with HCMV, documenting that the virus is the etiological agent may be considerably more complicated. Serological tests are frequently not helpful. Additionally, because healthy individuals may shed HCMV in their urine, a positive culture does not document virally associated disease (12-14). Positive viral identification in the blood is more indicative of HCMV-related disease, but frequently direct identification of the virus in biopsy specimens is required for definitive diagnosis. Thus, there is a wide arena in which nucleic acid hybridization procedures can be useful for the diagnosis of HCMV disease. In contrast, the diagnosis of AIDS is considerably less complicated; an individual who is H1V-seropositive and demonstrates the clinical manifestations associated with AIDS has already had the diagnosis established (15). However, nucleic acid hybridization procedures may help to identify persons who are antibody-negative but nonetheless infected with H1V. Therefore, hybridization procedures may be particularly helpful during periods of HIV incubation, or in newborn infants, when serological tests are of limited value. Moreover, the need to identify virological markers to follow patients enrolled in treatment protocols and to help determine those asymptomatic persons at risk for progressive H1V-related disease may be met by a quantitative assay for HW nucleic acid. diagnostic
Materials and Methods Departments of 1Pediatrics and tmBiology, and 3Center for Molecular Genetics, University of California, San Diego. Address for correspondence: UCSD Medical Center, 225 Dickinson St., H-814-H, San Diego, CA 92103. 4Nonstandard abbreviations: HCMV, human cytomegalovirua; HIV, human immunodeficiency virus; AIDS, acquired immunodeficiency syndrome; PCR, polymerase chain reaction; SSC, NaCI 0.15 mol/L and sodium citrate 15 mmolJL; GuSCN, guanine thiocyanate; and bp, base pairs. Received April 11, 1989; accepted May 23, 1989.
Virus propagation and clinical specimens. HCMV was isolated from clinical specimens by the Viral Diagnostic Laboratory at the University of California, San Diego Medical Center, as previously described (16, 17). Further propagation of clinical isolates and of the AD169 strain of HCMV was performed in human foreskin fibroblast cells. Strain AD169 was originally obtained from the American Type Culture Collection, Rockville, MD, and maintained in our laboratory. Clinical specimens were obtained from CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989
1581
patients
considered to be at high risk for infection with (or) HIV: bone-marrow or kidney transplant recipients, newborn infants, and persons infected with 11W. Dot/slot blot hybridization. We used cloned EcoRI fragments of HCMV strain AD169 for dotlslot blot hybridization (18, 19). Procedures for preparation of inserts from recombinant plasmids, extraction of sample DNAs, and blot hybridization have been described elsewhere (20,21). Polymerase chain reaction (PCR) for HCMV. For PCR amplification of HCMV, we synthesized four 20-oligonucleotide primers and two 40-oligonucleotide probes (Figure 1) within EcoRI fragment D of strain AD169. These oligomers were synthesized with a DNA synthesizer (Applied Biosystems, Inc., Foster City, CA; Model 380A) and purified by HPLC chromatography (21a). Primer pair 1(461 and 625) is 101 bp (base pairs) from the EcoRI fragment D-Ajunction; primer pair 11(459 and 627) is 160 bp from the EcoRI fragment D-Vjunction. Purified infected and uninfected cell DNA was obtained as previously described (18, 19). Crude cellular lysates were prepared by washing cells in sodium phosphate buffer (50 mmol/L, pH 6.5), resuspending in 50 to 70 zL of doubly distilled water, freeze-thawing three times, and heating to 95#{176}C for 5-10 mm. For the gene amplification reaction, substrates for PCR were added directly into these crude cell lysates. Urine samples used for PCR were first clarified by centrifugation at 900 x g for 20 mm, and virus pelleted either by ultracentrifugation at 90000 x g for 1 h or in a microcentrifuge. Pellets were washed once with cold “STE” buffer (per liter, 50 mmol of NaCl, 10 mmol of Tris hydrochloride, pH 7.5, and 1 mmol of disodium EDTA), and resuspended in sterile water to give a final 200-fold dilution of the original urine sample. The PCR was carried out according to the method of Saiki et al. (22) and as recently described (23). In brief, PCR was performed in 100-FL reaction mixtures consisting of 100 to 150 pmol per 100 zL of each primer, in a solution containing, per liter, 1 mmol each of four deoxynucleotides, 5 mmol of MgC12, 10 mmol of Tris hydrochloride (pH 8.3), 50 mmol of KC1, and 0.1 g of gelatin. The first cycle of each reaction involved denaturation of sample DNA at 94#{176}C for 1.5-2 mm, followed by primer annealing at 37#{176}C for 2 mm. After adding Taq polymerase (2 U/zL), we allowed the nascent DNA chains to extend for 3 mm at 72 #{176}C. For subsequent cycles, DNA was denatured at 94#{176}C for 2 mm, primers were annealed at 37#{176}C for 2 mm, and primer extension was at 72#{176}C for 3 mm. For the last cycle, DNAs were allowed to extend for 6 mm at 72#{176}C. HCMV and
‘ 0 ‘
,,,
WONt
“K’’’S’O’’’ ,, ,,
,
I
U
Y
#{163} ‘M’P’’
I U
A
‘ID
J
‘
W” ,,
V
C
a.
-------
PRIMER PW.,
PAIR
J ‘ t t ,
, ‘ ‘
.0W
NkT
S.-
I
1211 GCCOOATTQTGOATItCOTT
PAIR II ____________
PRIMER
P,Wa., 14111 TCCCAOTCACOATADCIQCC
P.11W
14111 CCGAGGATOCTGATTTOCGT
P,N,,.,
Ilfll
P,.N.
1211 YCOGTGTCCTCTTC000CGC
P,.b.
1211 ATTCGTOOTCOTQGCCAACY
GACOQTGGACTCGGCCTTAA
GATCCOACCCATTO1CTAAO
OOTGCTOCCOGTCGCQCTTA
FIg. 1. EcoRlmap of the HCMVgenomestrain AD1 69 wIth sequences of the synthetic oligonucleotide primers and probes and their locations within fragment D 1582
CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989
The oligonucleotide probes were end-labeled with T4 polynucleotide kinase with 100 pCi of gsmnis-emitting-[32P]ATP (Amersham, 10 Ci/L, 3 MCilmol) as previously described. The average efficiency of incorporation of y-[32P]ATP into oligomer probes was about 108 counts/mm per 100 pmol of probe. Polymerase chain reaction for HJV. For amplification of HIV-1 proviral sequences, we have used primer pairs representing the gag region of the HlV strain ARV-2 genome (courtesy of Cetus Corp., Emeryville, CA). The specific primers have been designated SK 38 [sequence: ATAATCCACCTATCCCAGTAGGAGAAAT (gag 155 11578)]; and SK 39 [sequence: ITFGGTCCT’FGTCTFATGTCCAGAATGC (gag 1638-1665)]. The probe used in conjunction with these primers is SK 19 [sequence: ATCCTGGGATTAAATAAAATAGTAAGAATGTATAGCCCTAC (gag 1595-1635)]. The sensitivity and specificity of this primer pair and probe have been described previously (24, 25). We used the procedures described by Ou et al. (25) for amplification of HJV. In situ hybridization. The procedures used for in situ hybridization have been previously described (5,10,19,26) and, briefly, (0.2 mol/L)
were as follows.
Slides
were incubated
in HC1
for 20 mm, rinsed in distilled water, and incubatedin”2 X SSC”(1 x SSCisO.l5mol/LNaClandl5 mmoIJL sodium citrate) for 30 mm at 65#{176}C. Specimens were then protein-digested in Tris (20 mmol/L)-CaCl2 (2 mmol/L) buffer containing proteinase K (EC 3.4.21.14; 1 mg/L) for 15 miii, rinsed in distilled water, and dehydrated in a graduated series of ethanol. For DNA hybridization, slides were treated with pancreatic ribonuclease (EC 3.1.27.5; 100 mg/L) for 30 mm at 37#{176}C and ribonuclease T1 (EC 3.1.27.3; 10 kU/L). RNase activity was inhibited in diethylpyrocarbonate (1 mIJL), and slides were incubated in 50 g/L paraformaldehyde solution for 2 h at room temperature. After rinsing in 2 x SSC, the DNA was denatured in de-ionized formamide at 65 #{176}C for 15 mm, immediately placed in ice-cold 0.1 x SSC for 2 mm, and dehydrated in a graduated series of ethanol solutions. Within 30 miii after the slides were dehydrated, hybridization was carried out for 60 h at room temperature, with use of a 35S-labeled mix of HCMVprobes previously shown not to hybridize to uninfected human DNA (19,27). Slides were covered with GelBond (FMC Bioproducta, Rockland, ME) and sealed with rubber cement. At the end of this hybridization, the slides were washed twice in a solution containing formamide (500 mLIL), Tris hydrochloride buffer (0.1 molIL), NaCl (0.6 mol/L), EDTA (1 mmolIL), and Triton-X (1 mLIL) for 5 mm, and then incubated in 2 x SSC at 70#{176}C three times for 20 miii each. Subsequently, the slides were placed in 1500 mL of the formamide wash solution at 42#{176}C and stirred for 72 h, with daily changes of wash buffer. Slides were dried by dehydration in a graduated series of ethanol solutions, each containing ammonium acetate (0.6 mol/L), dipped in NBT2 nuclear track emulsion (Eastman Kodak Co., Rochester, NY), developed in Kodak D19 developer solution, and fixed in Kodax fixers. Target cycling for HCMV. A 32P-labeled RNA probe and three single-stranded DNA oligomer capture probes complementary to the mRNA encoding the HCMV late matrix protein, pp 65, were used for target cycling. The 1-kb32Plabeled RNA probe (2 x 108 counts/mu per picogram of RNA) was generated from HCMV strain AD169 Hind Ill fragment b, which had been cloned into a riboprobe gemini transcription plasmid by using T7 RNA polymerase and
[32P]UTP. The three oligomer capture probes corresponded to sequences adjacent to Hind ifi fragment b within Hind ifi fragment c. They were covalently linked to oligo d(A) at their 3’ end. DNA samples were resuspended in 20 zL of 5 molfL guanidine thiocyanate (GuSCN) reagent containing 1 mol of EDTA (pH 7) and 100 g of dextran sulfate per liter. The three capture probes (1 pmol of each probe per sample) and labeled RNA probe (3 x iO counts/mn per sample) were resuspended in 50 jL of probe buffer containing 1.5 mol of GUSCN per liter and added to the 20-pL samples. The mixtures were first heated to 95#{176}C for 10 mm to denature the HCMV DNA and then hybridized at 37#{176}C for 4 h. A 50-tL suspension of magnetic particles, to which oligo d(T) was covalently attached was then added to each tube. The tubes’ contents were vortex-mixed and incubated for 10 mm at 37#{176}C, washed in 200 L of buffer containing 0.5 mol of GuSCN per liter, vortex-mixed, and placed in a magnetic separator (Gene-Trak Systems, Framingham, MA) to pull the magnetic beads with the attached hybrids to the sides of the tubes. The supernates were aspirated from each tube, wash buffer was again added, and the process of magnetic separation repeated. After removal of the wash solution, 50 zL of release buffer containing 3 mol of GuSCN per liter was added to each tube. These samples were vortex-mixed and placed in the magnetic separator. Using a guided pipettor, we removed 50 j.tL of buffer containing released hybrids and placed this in a new tube. To the solution we added 50 L of magnetic particles, repeating the capture and release procedures two additional times. After a final capture and wash procedure, 200 z.L of the magnetic particle suspension from each tube was filtered onto a nitrocellulose filter through a Minifold apparatus (from Schleicher & Schuell, Keene, NH). Filters were dried and exposed to autoradiography as previously described.
Results Detection of HCMV DNA Dot/slot blot hybridization. We have applied DNA-DNA hybridization extensively for the identification of HCMV in clinical specimens, including urine, peripheral blood leukocytes (buffy coats), bronchial alveolar lavage, and bonemarrow, lung, kidney, and other organ tissues (10, 18, 19). In our experience, this procedure can routinely detect the equivalent of 1 pg of homologous EcoRl fragment-D DNA. Although dotlslot blot hybridization offers the advantages of high sensitivity and specificity, broad applicability to most clinical specimens, and relative quickness to complete, it remains cumbersome, requires 32P as a detector system to attain the desired sensitivity, and is occasionally complicated by nonspecific hybridization. The recent development of the amplification procedure of PCR promises to help resolve many of these problems. Amplification procedure of PCR. In the PCR, target DNA sequences are selectively amplified through repeated cycles of denaturation, annealing with oligomer primers complementary to the flanking region of the target sequence, and primer extension with DNA polymerase. During this procedure, the amount of amplified DNA theoretically increases exponentially as a function of cycles (2”). The PCR procedure has been the subject of several recent reports (28-30).
For performing
PCR, we chose two fragments lying EcoRI fragment D, which is transcribed early and in abundance late in infection. This region was selected because of our extensive experience in within
HCMV AD169
using full-length fragment D as a probe. Also, in previous experiments we had found fragment D to be highly conserved among all clinical HCMV isolates examined (18, 19, and unpublished data). The sequences of the four synthetic primers (461, 625, 459, and 627) and two synthetic probes
(626 and 628) are indicated in Figure 1. Because we found that amplification with primer pair 11(459/627) was consistently more efficient than that achieved with primer pair I (461/625), we report here only our findings with primer pair 11. Before using our primer pairs for the detection of HCMV in clinical specimens, we had to ascertain that they were conserved among different clinical HCMV isolates. In >100 strains of HCMV evaluated, we have always been able to amplify the authentic 152 bp fragment (23, and unpublished data). Additionally, there is no specific amplification with DNA from HSV-1, HSV-2, Epstein-Barr virus, vancelia zoster virus, adenovirus, or HIV-1. We next examined the sensitivity of our amplification procedure. In reconstruction experiments, using purified EcoRI fragment D (17 kb) and 35 cycles of PCR, we could detect 0.01 pg by slot blot hybridization and at least 0.1 pg by Southern hybridization (23). Thus, with 35 cycles of PCR, we have attained 100-fold greater sensitivity than by direct slot blot hybridization with fragment D as probe with hybridization to homologous DNA. In other reconstruction experiments, we can detect one foreskin fibroblast cell infected with AD 169 (Figure 2). Having established the sensitivity and specificity of the PCR procedure, we next examined its ability to detect HCMV in clinical specimens. Applying the procedure to urine specimens obtained from premature infants, transplant recipients, and patients with ms, we have, to date, been able to detect HCMV DNA by PCR in all urines tested that had been identified by culture to contain infectious HCMV. An example of detection of urine specimens by PCR is shown in Figure 2. However, in our studies, two urines positive for viral DNA by PCR were culture-negative. Both 1
A-
2
3
4j
B-
C-
E Fig. 2. Slot blotanalysisof amplification by PCR of HCMV DNA8 extracted from human foreskin fibroblast (FF) cells infected with strain AD169 and urine from infected and uninfectedindividuals PCR was carried out In 100-jL reaction mixtures, and 125 pmol each of the pflmer ollgomers 459 and 627 with probe 628. Columns D and E are DNAs from the equivalent of 1 (lane 1), 10 (lane 2),and 100 (lane InfectedFF cells. Column D is before 35 cycles of PCR and column E Is after RCA. Column C is uninfected FF cells, 100 (lane 1), 1000 (lane 2), and 10000 (lane . ColumnsA and Bare urine samples before (column and after (column A) 35 cycles of PCR. Urinesin lanes 1 and 2 were culture positive; urinein lane3 was culture negative
CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989 1583
of these urines were obtained from patients with Ams who previously had been identified by culture to be positive. Thus, PCR may be more sensitive than viral isolation for detection of HCMV in urine specimens. In further studies, we have applied the PCR procedure for detection of HCMV in peripheral blood leukocytes of patients with active HCMV disease. Viral DNA was consistently identified in all specimens ultimately found by culture to contain infectious virus. Additionally, when repeat specimens have been obtained over time in patients with HCMV disease, the PCR procedure has been repeatedly positive, whereas viral isolation is frequently only intermittently positive. We have also followed patients with Ams who were being treated with ganciclovir for lifethreatening or sight-threatening HCMV retinitis. Figure 3 shows the results of applying PCR to a sample from such a patient being treated for retinitis with ganciclovir. Although the viral culture was positive only before treatment was begun, HCMV nucleic acid could be detected during the treatment with ganciclovir, progressively declining while on treatment. However, there is an apparent increase in viral DNA at week 16, at which time the patient was developing reactivation of his retinitis. PCR has also been applied for the detection of viral DNA in tissue specimens. Again, HCMV has been consistently detected in positive specimens, but not in tissue from individuals seronegative for HCMV (data not shown). In preliminary experiments, we have used an alkaline phosphatase-labeled oligonucleotide probe to detect HCMV after amplification by PCR. Figure 4 shows a slot blot hybridization obtained when we used this probe after amplification of viral DNA contained in 1, 10, and 100 AD169-infected foreskin fibroblast cells. Again, as little as one infected cell can be detected after 35 cycles of PCR. This entire procedure takes less than 8 h and promises to be as sensitive and specific as viral isolation. Application of target cycling to the identification of HCMV. The technique of target cycling has been applied by others to detect HIV mRNA. In this procedure, cells are first dissolved in the chaotropic salt, GuSCN, which releases mRNA and inhibits RNase activity (31, 32). To this solution is added a P- or ‘I-labeled complementary RNA probe and unlabeled complementary DNA oligonucleotides linked at their 3’ end to poly d(A). After hybridization, the hybrids are captured from solution by specific hybridization
Fig. 4. Slot blot analysis of PCR-amplified HCMV DNA from the equivalent of 1, 10, and 100 cells infected with HCMVAD 169, and detected with probe628 labeled withalkalinephosphatase(see text for details) between the poly d(A) tail sequences on the capture probe and the poly d(T) sequences, which have been covalently attached to magnetic beads. The captured hybrids can then be cycled (i.e., washed, released, and recaptured), effectively increasing the relative amount of signal while decreasing background noise. Recently, we have begun to apply target cycling for the detection of HCMV DNA and RNA. Figure 5 shows a reconstruction experiment with different quantities of HCMV EcoRI fragment D, to which the target recycling procedure was applied. The sensitivity of this procedure appears to approximate standard dotlslot hybridization when probe length is considered. The entire process takes approximately 8 h, can be done with minimal training of technicians, and can be performed on multiple specimens. This procedure may also be useful as a first step to “clean up” sample nucleic acid before applying PCR. Application olin situ hybridization for detection of HCMV. We and others have applied in situ hybridization in studies on detection of HCMV in clinical specimens and on identifying mechanisms of pathogenesis. Identification of }ICMVas the etiological agent of interstitial pneumonia in lung specimens is important if appropriate therapy is to be instituted and also for documenting clinical disease. Procedures have been developed for both nonisotopic and isotopic detection systems. Figure 6 shows the application of in situ hybridization for detection of HCMV in the retinal cells of a
#{149}.1000 pg
100 pg
#{149}XADFFO 4 5 16 10 pg 152 bp 4
1 pg
no DNA Fig. 3. Southern blotanalysisof PCR-amplifiedHCMVDNA obtained
fromperipheral blood leukocytes of a patientwithAIDS and HCMV retinitis,beingtreatedwith ganciclovir Has Ill-digested .X 174 RE DNA was used as sue markers.AD = HCMvstrain AD169 DNA; FF = foreskin flbroblast DNA; 0, 4, 5, 18 weeks of ganciclovir therapy (see text for details) 1584
CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989
no capture probe Fig. 5. HybrIdization of HCMV AD169 EcoRI fragment D to 32Plabeled oligonucleotideprobes,with use of target cycling
+PCR
-PCR
1 2 3 4 5
4, FIg. 6. in situ hybridization of retina tissue obtained postmortem from a patient with AIDS and HCMVretinitis Hybridization was performed with a mix of MS-labeled HCMv subgenomlc probes. Arrows Indicate the areas of retina exhibiting strong hybridization for HCMV nucleicacid
patient with AIDS, who during life was identified to have a diffuse fluffy hemorrhagic retinitis consistent with HCMV infection. The major advantage of the in situ form of hybridization over other hybridization procedures is its ability to localize specific infected areas and cells. Although this ability to localize infection is shared by immunocytochemical procedures, in situ hybridization can be applied without difficulty to formalin-fixed tissue, whereas many antigens detected by specific antibodies are destroyed. This is particularly a problem with monoclonal antibodies, which are frequently unusable in fixed tissue. The power of both in situ hybridization and immunocytochemistry can be used in double-labeling studies. We have used this procedure frequently to identify a specific cell population infected by HCMV or HIV (10, and unpublished data).
J1t
Fig. 7. Slot blot analysis of amplification by PCR of DNA obtained from three individuals infected with HIV land 2are 10% ofthe amplificationproductof 1 zgof DNA extractedfrom the peripheral blood mononuclearcells. 3, 4, and 5 are 10% of the amplIfication productof 102,iO, and iO peripheralblood mononuclearcells, respectively. -PCR = beforePCR; +PCR = after35 cyclesof PCR
(EC 3.1.3.1)-labeled oligonucleotides for the detection of H1V by in situ hybridization. Figure 8 shows a lymph node (obtained at autopsy) hybridized with the 35S-labeled HIlT probe. In our experience, lymph nodes are almost always positive for HIV in seropositive patients. In further studies, we have used alkaline phosphataselabeled oligonucleotide probes to detect HilT nucleic acid. These probes have the advantages of shorter hybridization times and less nonspecific hybridization than larger probes. These procedures can be completed within 8 h and have the same sensitivity as the full-length probes (data not shown). DIscussion We plied HilT, these
have reported how molecular probes have been apto improve methods for the diagnosis of HCMV and and to extend our knowledge of the pathogenesis of two viruses. Particular emphasis has been placed on
Detection of HIV Polymerase
chain reaction. With the development of enables the amplification of specific gene sequences, the developers at Cetus and their colleagues quickly appreciated the potential for this procedure to identify H1V in clinical specimens. Using a set of two 20-oligonucleotide primers and a probe from the gag region (obtained courtesy of Cetus Corp.), we have consistently identified the presence of HIV DNA in clinical specimens obtained from persons known to be seropositive for HJV. We have been particularly interested in using PCR to determine the approximate percentage of peripheral blood mononuclear cells infected with HIV in seropositive individuals. In preliminary findings, we have evidence that as many as 1% of peripheral blood mononuclear cells in seropositive persons contain H1V DNA (Figure 7). This proportion of infected cells is substantially greater than the 0.01% to 0.001% determined to be infected cells by in situ hybridization (33), and probably reflects the greater sensitivity of the PCR procedure. Application of in situ hybridization for HI V detection and studies of pathogenesis. We have applied a full-length 35S-labeled HIV probe as well as alkaline phosphatase
PCR, which
I, Fig. 8. In situ hybridization of a lymph node obtained from patient with AIDS Hybridization was performed by using a full-length HIV probe, pHXB-2D. Arrows indicate cells exhibiting strong hybridization for HIV nucleic acid CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989
1585
the application of the amplification procedure of PCR. This procedure shows great potential for the identification of both RNA and DNA viruses, and will be particularly useful where serology or culture are difficult or when rapid identification is required. For HCMV, nucleic acid hybridization procedures may redefine our existing understanding of infection. In particular, application of the PCR for detection of HCMV is more sensitive than tissue-culture procedures, which are currently considered the “gold standard.” These findings with PCR confirm previous observations with dotislot blot hybridization (18, 19). However, application of this information to improve clinical care for patients requires further investigation. Additionally, the application of molecular hybridization procedures, including in situ hybridization, double-labeling procedures, and target cycling as well as PCR, promises to contribute to our understanding of the pathogenesis of HCMV infection, particularly in defining the site(s) of viral latency/persistence and reactivation. The application of in situ hybridization procedures and PCR will also provide a better understanding of the pathogenesis of HIlT infection. In preliminary work on application of PCR for the detection of HIV in peripheral blood mononuclear cells, we have found that HilT DNA can be detected as much as 100 times more frequently than the estimated 1 in 10 to i0 cells previously found to contain virus by in situ hybridization; PCR is also more sensitive than direct detection of HilT RNA (34). Additional uses of PCR for HilT are still being defined. Of particular importance are detection of HIV DNA in mononuclear cells in the peripheral blood of high-risk individuals six to 48 months before seroconversion (35); early identification of infection in infants born to H1V-seropositive women (36); and in monitoring the viral load of seropositive patients receiving antiretroviral treatment. In summary, we have described the application of molecular probes to investigations of HCMV and HIV. The severe diseases related to these viruses both individually and together stress the importance of developing sensitive and specific procedures to follow the course of infection. Molecular probes will play a critical role in furthering our knowledge of the pathogenesis of these viruses. This research was supported in part by NIH grants HB-6-7019, MH 43298, and AI-27670, and by AmFAR.
References 1. Richman DD. Developments in rapid viral diagnosis [Review]. Infect Dis Clin North Am 1987;1:311-22. 2. Norval M, Bingham RW. Advances in the use of nucleic acid probes in diagnosis of viral diseases of man [Review]. Brief Rev Arch Virol 1987;97:151-65. 3. Spector SA, Danker WM. Rapid viral diagnostic techniques. In: Aronoff SC, ed. Advances in pediatric infectious diseases. Chicago: Year Book Medical Publishers, Inc., 1986:37-59. 4. Tenover FC. DNA probes for infectious diseases. Boca Raton: CRC Press, Inc., 1989:1-286. 5. Haase A, Brahic M, Stowring L, Blum H. Detection of viral nucleic acids by in situ hybridization. Methods Virol 1984;7:189226. 6. Myerson
D, Hackman RC, Meyers JD. Diagnosis of cytomega-
loviral pneumonia 1984;150:272-7.
by in situ hybridization.
J Infect
Die
7. Myerson D, Hackman RC, Nelson JA, Ward DC, McDougall JK. Widespread presence of histologically occult cytomegalovirus. Hum Pathol 1984;15:430-9. 8. McDougall JK, Myerson D, Beckmann AM. Detection of viral 1586
CLINICALCHEMISTRY,Vol.35, No.8, 1989
DNA and RNA by in situ hybridization. J Histochem Cytochem 1986;34:35-8. 9. Sklar J. DNA hybridization in diagnostic pathology. Hum Pathol 1985;16:654-8. 10. Spector SA, Dankner WM, Denaro F, Spector DH. DNA probes for the detection of human cytomegalovirus. In: Tenover Fe, ed. DNA probes for infectious diseases. Boca Raton: CRC Press, Inc., 1989:135-40.
11. Bloom JN, Palestine AG. The diagnosis of cytomegalovirus retinitis. Ann Intern Med 1988;12:963-9. 12. Stagno S, R,eynold 5, Tsiantos A, et al. Cervical cytomegalovirus excretion in pregnant and nonpregnant women: suppression in early gestation. J Infect Die 1975;131:522-7. 13. Drew WL, Mintz L, Miner RC, Sands M, Ketterer B. Preyalance of cytomegalovirus infection in homosexual men. J Infect Dis 1981;143:188-92. 14. Lange M, Klein EB, Konnlleld H, Cooper LZ, Grieco MH. Cytomegalovirus isolation from healthy homosexual men. J Am Med Assoc 1984;252:1908-10.
15. Centers for Disease Control. Human iminunodeficiency virus infection in the United States: a review of current knowledge. MMWR 1987;36:1-48. 16. Spector SA, Schmidt K, Ticknor W, Grossman M. Cytomegaloviruria in older infants in intensive care nurseries. J Pediatr 1979;95:444-6. 17. Spector SA, Spector DH. Molecular epidemiology of cytomegalovirus infections in premature twin infants and their mother. Pediatr Infect Die 1982;1:405-9. 18. Spector SA, Rua JA, Spector DH, McMillan R. Detection of human cytomegalovirus in clinical specimens by DNA-DNA hy.. bridization. J Infect Die 1984;150:121-6. 19. Spector SA, Spector DH. The use of DNA in probes in studies of human cytomegalovirus. Clin Chem 1985;31:1514-20. 20. TamashiroJC, Hock LI, Spector DH. Construction of a cloned library of the EcoRI fragments from the human cytomegalovirus genome (strain AD169). J Virol 1982;42:547-57. 21. Spector DH, Hock L, Tamashiro JC. Cleavage maps for human cytomegalovirus DNA strain AD169 for restriction endonucleases EcoRI, Bgl H, and Hind ifi. J Virol 1982;42:558-82. 21a. Beaucage SL, Caruthers MI!. Deoxynucleoside phosphoramidites: a new class of key intermediates for deoxypolynucleotide synthesis. Tetrahedron Lett 1981;22:1859-62. 22. Saiki RK, ScharfS, Faloona F, et al. Enzymatic amplification of $-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 1985;230:1350-4. 23. Hsia K, Spector DH, Lawrie J, Spector SA. Enzymatic amplification of human cytomegalovirus sequences by polymerase chain reaction. J Cliii Microbiol (in press).
24. Kwok 5, Mack DH, Mullis KB, et al. Identification of human immunodeficiency virus sequences by using in vitro enzymatic amplification and oligomer cleavage detection. J Virol 1987;61:1690-4. 25. Ou C-Y, Kwok 5, Mitchell SW, et al. DNA amplification for direct detection of HIV-1 in DNA of peripheral blood mononuclear
cells. Science 1988;230:295-7. 26. Dankner WM, McCutchan JA, Richman DD, Hirata KK, Spector SA. Localization of human cytomegalovirus in peripheral blood leukocytea by in situ hybridization. (Submitted for publication). 27. Shaw SB, Rasmussen RD, McDonough SH, Staprans SI, Vacquier JA, Spector DH. Cell-related sequences in the DNA genome
of human cytomegalovirus strain AD169. J Virol 1985;55:843-8. 28. Saiki RK, Annheim N, Erlich HA. A novel method for the detection of polymorphic restriction sites by cleavage of oligonucleotide probes: application to sickle-cell anemia. Biotechniques 1985;3:1008-12. 29. Oste C. Polymerase chain reaction. Biotechniques 1988;6:1627.
30. Schochetman G, Ou C-Y, Jones WK. Polymerase chain reaction. J Infect Die 1988;158:1154-7. 31. Thompson J, Gillespie D. Molecular hybridization with RNA probes in concentrated solutions of guanidine thiocyanate. Anal Biochem 1987;163:281-91. 32. Pellegrino MG, Lewin M, Meyer WA, et al. A sensitive solution hybridization technique for detecting RNA in cells: application to H1V in blood cells. Biotechniques 1987;5:153-9.
33. Harper ME, Marselle LM, Gallo RT, Wong-Staal F. Detection of lymphocytes expressing human T-lymphotropic virus type ifi in lymph nodes and peripheral blood from infected individuals by in situ hybridization. Proc Natl Acad Sci USA 1986;83:772-6. 34. Richman DD, McCutchan JA, Spector SA. Detecting human immunodeficiency virus RNA in peripheral blood mononuclear cells by nucleic acid hybridization. J Infect Die 1987;156:823-7.
35. Wolinsky 5, Runaldo C, Farzedegan H, et al. Polymerase chain reaction (PCR) detection of HIV provirus before HilT seroconversion. Abstract #1099, Program and Abstracts of the IV International Conference on AIDS, June 1988, Stockholm, Sweden. 36. Laure F, Rouzioux C, Veber F, et al. Detection of HIV1 DNA in infants and children by means of the polyinerase chain reaction. Lancet 1988;ii:538-41.
CLINICAL CHEMISTRY, Vol. 35, No. 8, 1989 1587
