Molecular and Cellular Probes (1992) 6, 265-270

Detection of HIV1 proviral DNA by PCR and hybridization with digoxigenin labelled probes N . Costa Taveira, M . 0 . Santos Ferreira and J . Moniz Pereira* Faculdade de Farmacia da Universidade de Lisboa . Dep . Microbiologia, Av . das Forcas Armadas, 1699 Lisboa CODEX, Portugal . (Received 18 January 1991, Accepted 25 February 1992)

One of the main obstacles for the introduction of PCR method to identify HIV1 proviral DNA in routine diagnostic laboratories is the use of radiolabelled oligodeoxynucleotide probes . Nonradioactive labelled probes have several advantages over radioactive labelling : they are stable for over 1 year, they can be produced easily in large amounts and they are safe . Polymerase chain reaction is an efficient and simple method to produce vector free inserts to use as probes . In this paper we describe a procedure for labelling DNA probes with digoxigenin-11-dUTP using the polymerase chain reaction . This non-radioactive labelling system was applied to detect HIV proviral sequences, amplified in vitro by PCR, from peripheral blood mononuclear cells DNA of infected subjects . We found identical sensitivities and specificities for probes synthesized with the nonradioactive and radioactive labelling procedures . The digoxigenin-11-dUTP can be efficiently incorporated during amplification of a DNA fragment using the polymerase chain reaction . This labelling and detection method proved to be specific, sensible and simple enough to be used in routine diagnostic laboratories for the detection of HIV1 infected individuals .

KEYWORDS : Digoxigenin-11-dUTP, labelling, polymerase chain reaction, HIV1 .

INTRODUCTION Serological assays for antibodies to human immuno-

hybridization using radiolabelled oligodeoxynucleo-

deficiency virus (HIV) are normally used to identify

tides as probes . The use of radiolabelled probes is one

individuals infected with this virus . However, in some

of the main obstacles to the introduction of PCR in

1-4

situations direct detection methods are desirable . The polymerase chain reaction (PCR) is a powerful

clinical diagnostic laboratories . Usually these labor-

method for the direct demonstration of specific

radioisotopes (security areas, storage and elimination

sequences of HIV1 proviral HIV in peripheral blood mononuclear cells (PBMC) .' By amplifying specific

of radioactive wastes) . In addition, 32 P labelled probes

atories do not have the conditions required to handle

sequences of minute amounts of HIV DNA present in

have a short half-time and are not produced in large quantities .

PBMC of infected individuals, sufficient quantities of DNA are synthesized allowing detection of viral DNA

been developed in order to overcome these inconve-

with hybridization assays .

nients .', ' Non-radioactive labelled probes are stable

Non-radioactive label and detection systems have

PCR may amplify not only the single segment of

for 1 year and can be easily produced in large

interest but also irrelevant segments, which can be

amounts . The main limitation of the general use of

incorrectly identified as HIV related . Usually the de-

non-radioactive molecular hybridization techniques

tection of the PCR-amplified DNA segment is made by

was the detection limit (about 10 pga of DNA target) .

*Author to whom correspondence should be addressed .

0890-8508/92/040265 + 06 $08 .00/0

265

© 1992 Academic Press Limited



266

N . Costa Taveira et al .

The use of PCR for labelling probes offer several advantages :"' (1) It labels only the insert and not the vector ; (2) it labels small DNA fragments ; (3) it requires subnanograms of DNA target ; (4) it is very fast and efficient ; (5) the generated probe can be used immediately without purification ; (6) it allows direct labelling from genomic DNA . In this paper, we describe a procedure to label DNA probes with digoxigenin using PCR .

The slot-blots were prepared using vacuum manifolds from Bethesda Research Laboratories (BRL) as described elsewhere . 7 Briefly : 10 VI of PCR product were heat denatured and applied to pretreated nylon membranes (5 min in distilled water, 15 min in 10 X SSC and 5 min drying) . Once the samples were completely absorbed, membranes were removed and floated for 5 min on 3 M sodium acetate pH 5 . 0, then twice for 5 min each on 2 x SSC . Membranes were dried and fixed by 3 min u .v . exposure .

MATERIALS AND METHODS Labelling of specific HIV1 DNA probes Oligonucleotide • primers The oligonucleotides used in polymerase chain reaction corresponded to positions 2356-2381 (oligonucleotide P3-TGGGAAGTTCAATTAGGAATACC) and 2637-2663 (oligonucleotide P4-CCTACATACAAATCATCCATGTATTG) of HIV1 : isolate LAW) . This pair of primers (P3/P4) defines a 307 base pair conserved nucleotide sequence in the HIV1 pol gene . The oligonucleotides were synthesized on a automated DNA synthesizer system 200 A (Beckman) based on the cyanoethyl-phosphoramidite chemistry according to the instructions of the manufacturer .

Polymerase chain reaction Peripheral blood mononuclear cells from seropositive and seronegative patients were separated by means of Ficoll-hypaque gradient. DNA from PBMC was extracted as described by Maniatis et al." The polymerase chain reaction was carried out in a final volume of 100 pl of a reaction mixture containing : 1 lag of genomic DNA, 1 µM of each primer, 0 .75 mm of each of four deoxyribonucleoside triphosphates and 2 . 5 units of Taq DNA polymerase (provided by Cetus, Emeryville, CA) in the buffer recommended by New England Biolabs (67 mm Tris-HCI, pH 8 . 8, 7 mm MgCl 2, 10 mm (3-mercaptoethanol, 16 . 6 mm NH4 SO 4, 170 Rg ml - ' bovine serum albumine, 6 . 7µM EDTA) . The reaction mixture was overlaid with paraffin oil to prevent evaporation . Time and temperature of each step were as follows : denaturation, 2 min at 95 ° C ; reannealing, 2 min at 55 ° C; and chain elongation, 2 min at 72 ° C. This cycle (denaturation, reannealing and elongation) was repeated 40 times in a thermal cycler apparatus . The final chain elongation step of the last cycle was 7 min . The amplified products, 10µI, were separated by 1 . 5% agarose gel electrophoresis, transferred to nylon membranes by Southern blot' and detected by hybridization to digoxigenin and radiolabelled probes .

HIV1 DNA fragments were labelled with digoxigenin11-dUTP by polymerase chain reaction using P3/P4 primer pair and the recombinant plasmid pM 2 containing the Hin, dIll fragment of HIV-1 MAL genome (entire HIV-1 genome), a gift from S . Wain-Hobson of the Pasteur Institut . The amplification was performed in a 100 gl reaction mixture containing : 67 mm TrisHCI pH 8 . 8, 16 . 6 mm ammonium sulphate, 7 mm magnesium chloride, 10 mm (3-mercaptoethanol, 170 µg ml - ' bovine serum albumin, 1 gm of each primer, 40 p.M of dCTP, dATP, dGTP, 26 µM of dTTP, 14 µM of dig-11-dUTP (Boehringer Mannheim), 1 ng of purified plasmid DNA and 2 . 5 units of Taq DNA polymerase (provided by Cetus Emeryville, CA) . The amplification conditions were the same as in PCR previously described . Radioactive HIV1 specific probes were labelled with 32 P dCTP (3000 Ci mmol - ') by the random priming method (Pharmacia) .' 2

DNA hybridization Hybridization with radiolabelled probe (10 9 cpm µg - ') was performed in 50% formamide at 42°C using nylon membrane as described by Maniatis et al ." Hybridization with digoxigenin-d-UTP labelled probe was carried out according to the hybridization protocol of the manufacturer :' briefly, after Southern blot, 100 cm2 nylon membranes were pre-hybridized for 1 hour at 42° C with 50 ml of hybridization solution (50% formamide ; 5% Boehringer blocking reagent ; 5 X SSC; 0. 1% N-lauroylsarcosine sodium salt ; 0 . 02% sodium dodecyl sulphate (SDS) . The hybridization was performed overnight at 42 ° C with 0. 025 ml cm -2 of the hybridization solution added with 16 ng ml"' of freshly denatured digoxigenin labelled probe . Membranes were washed 2 X 15 min at room temperature with 2 SSC, 0. 1% SDS and 2 X 30 min at 68 ° C with 0. 1 SSC, 0-1% SDS .



267

Detection of HIV1 Proviral DNA

ing to labelled DNA fragments developed colour reaction .

Hybridization detection Radiolabelled hybrids were detected by overnight autoradiography at -70 ° C using an intensifying screen ." Digoxigenin labelled probe was detected by specific immunoenzymatic reaction using anti-digoxigenin Fab fragments conjugated to alkaline phosphatase and the substracts 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and nitroblue tetrazolium salts (NBT) as manufacturer instructions .'

RESULTS Digoxigenin DNA labelling using PCR To verify if dig-11-dUTP was incorporated during the DNA amplification, the electrophoretic mobility of the PCR products generated in mixtures with and without digoxigenin were compared in 1 . 5% agarose gel electrophoresis (Fig . 1a) . We can see that the 307 by labelled product (lane 3) migrates more slowly than the unlabelled (lane 2) which suggests that the hapten was incorporated during the amplification reaction . To verify this assumption Southern blot of the gel and the immunological detection reaction were carried out (Fig . 1b) . Only the lanes correspond-

I

2

3

I

2

Sensitivity and specificity of the dig-11-dUTP DNA probe To examine the sensitivity and specificity of the digoxigenin labelled probe generated by PCR, serial dilutions of a mixture composed of the pM z recombinant plasmid cleaved by Hin dill (to release the HIV DNA) and the 307 base pairs pol fragment were carried out and run on a 1 . 5% agarose gel, (Fig 2a) . After Southern blot and hybridization using 50 ng of digoxigenin labelled probe, we detected 10 pg of HIV DNA and 10 pg of pol fragment, that is 3 . 10 7 copies, after 16 h of colour reaction, (Fig 2b) . No colour development was observed with the vector (pIJC19) indicating that the probe generated by PCR from 1 ng of DNA target (pMz) is specific and ready to be used . We have observed that the use of concentrations higher than 1 ng/100 µl of reaction mixture generates probes with a weak specificity able to react with the vector DNA (data not shown) . The sensitivity depends on the concentration of labelled DNA . To test the efficiency of the labelling by PCR, dilutions of the probe were performed and used to detect 100 ng of the 307 by poi fragment, fixed to a

3 (b) M

Bp

2

3

4

5

307-

Fig . 1 . (a) Electrophoretic analysis of the digoxigenin labelled DNA . A 307 bases fragment of the HIV1 poi gene defined by primers P3/P4, was amplified from 1 ng of DNA target (pM z) using the polymerase chain reaction in a reaction mixture without (lane 2) and with digoxigenin-11-dUTP (lane 3) . After 40 cycles of amplification 10 pl of each PCR product was electrophoresed through a 1 .5% agarose gel . Lane 1, 0 X174 Rf DNA Hae III digest (DNA molecular weight standards) . (b) Southern blot of the PCR products and detection of digoxigenin labelling using an immunoenzymatic colour reaction .

Fig . 2 . Sensitivity and specificity of the PCR-generated probe . Serial dilutions of a mixture, composed by pMz digested with Hin dill (to release HIV1 proviral DNA 9229 base pairs) and the 307 base pair poi fragment generated by PCR, were carried out and electrophoresed through a 1 . 5% agarose gel . (a) 1 : mixture not diluted ; H=HIV DNA (9229 base pairs), V=vector (pUC19), P=poi fragment (307 base pairs) . M : 0 X174 RF DNA Hae III digest (DNA molecular weight standards) . (b) Southern transfer and detection of HIV DNA sequences by hybridization and immunoenzymatic colour reaction using 50 ng of probe (the 307 base pair poi fragment labelled with digoxigenin by PCR) . 2 : H = 5 ng, P=1 ng ; 3 : H=500 pg, P=100 pg; 4 : H=50 pg, P=10 pg; 5 : H=5pg, P=1 pg .



268

N . Costa Taveira et al .

strip of nylon membrane, in 1 ml of hybridization solution (Fig. 3) . After 16 h of colour reaction, 1 ng of digoxigenin labelled probe generated by PCR was sufficient to detect 100 ng the pol fragment . Using the conditions described in Materials and Methods one can obtain more than 1 .tg of digoxigenin labelled probe . This amount of probe is sufficient to perform more than' 40 assays using 2 . 5 ml of hybridizatik solution and nylon membranes of 10 X 10 cm in size. (24 samples per assay) . With the random priming method, 100 ng of template DNA allowed the performance of five assays .

I

2

3

4

5

6

Fig. 3 . Minimal concentration of probe able to detect a PCR product . Serial dilutions of the PCR generated probe were used to detect 100 ng of the 307 base pair pol fragment fixed to strips of nylon membrane using 1 ml of hybridization solution . (1) 100 ng (2) 50 ng, (3) 25 ng, (4) 10 ng, (5) 1 ng, (6) 0. 1 ng . These results were obtained after 16 h of colour reaction .

Detection of HIV1 sequences in PBMC DNA of infected subjects Dig-11-dUTP DNA labelled probe produced, as described in Materials and Methods, was used to detect HIV1 proviral sequences in PBMC DNA of seropositive subjects after specific amplification using PCR . Using the same primer pair (P3/P4) as in the digoxigenin DNA labelling procedure, a 307 by sequence of the HIV1 pol gene was selectively amplified from the peripheral blood mononuclear cell DNA . We tested by Southern blot hybridization analysis 45 samples from seropositive subjects and 25 from seronegative blood donors (see Fig. 4) . All DNA samples from seropositive subjects were positive by DNA hybridization using both digoxigenin and 32 P labelled probes, after 'in vitro' amplification by PCR . Seronegative samples were all found to be devoid of HIV1 sequences in our test (Fig. 4) . The same results were obtained using a radiolabelled 32 P probe (Fig . 4a) . In order to eliminate electrophoresis and Southern transfer, slot-blot analysis was performed (Fig . 5) . Special care was taken in post-hybridization washes to avoid false-positive results due to the reaction of unincorporated primers with the PCR labelled probe,

I

2

3

4

5

6

7

8

9

10

II

(a )

(b)

Fig. 4. Southern hybridization analysis of P3/P4 PCR product using a radiolabelled probe (a) or, a PCR-generated digoxigenin-dUTP labelled probe (b) . Lanes 1 and 2 are respectively the positive and negative controls . Lanes : 3, 5-11, are samples from HIV1 seropositive subjects, lanes 4 and 12 from seronegative samples . Nylon membranes were first hybridized with the 32P labelled probe and then dehybridized (30 min at 42°C in 0. 4 M NaOH ; 30 min at 42°C in 15 mm NaCl, 1 . 5 mm trisodium citrate,,0 . 1 % SDS, 0. 2 M Tris HCI (pH 7 .5)) and rehybridized with the digoxigenin labelled probe .

(a)

( b)

logo on . logo I

Fig . 5 . Slot-blot analysis using PCR-generated digoxigenin-dUTP labelled probe of the 307 base pair HIV1 pol fragment amplified by PCR. Rows (a) and (b) are duplicates of the sample . Lane 1 is the positive control (pM2 recombinant plasmid), lane 2 is the primer P3 (10 gm), lane 3 is the primer P4 (10 gm), lane 4 is the mixture of primer P3 and P4 (10 lam of each), lanes 5, 8, 9 and 10 are samples from seronegative subjects, lanes 6 and 7 are samples from seropositive subjects .

since the primers used in the synthesis of the probe were the same as those used in the DNA amplification of the samples . The results obtained show that the unincorporated primers do not result in nonspecific background hybridization of the probe.

DISCUSSION The data presented in this report demonstrate the feasibility of the use of the polymerase chain reaction to label HIV DNA fragments with digoxigenini1-dUTP and the application of this non-radioactive probe to detect the presence of HIV1 proviral nucleic acid sequences in the DNA of peripheral blood mononuclear cells of seropositive subjects .

12

Detection of HIV1 Proviral DNA With the presented protocol we succeeded in incorporating a dNTP labelled with a steroid derived from digitalis plants, the digoxigenin-11-dUTP, into the DNA synthesized using the polymerase chain reaction . The method is very efficient, as more than 1 pg of probe can be produced in 4 h . One of the advantages of the PCR generated probes is the possibility of their direct use without purification . However, the concentration of the starting DNA target is a very important parameter . When more than 1 ng of the pM 2 recombinant plasmid was used as DNA template, the specificity of the probe decreased in the sense that sequences of the vector were also labelled . This was revealed by hybridization of PCR-labelled probe with the vector (results not shown) . The explanation of this may be linked to the production of longer DNA chains in the first cycles of amplification which overcome the limits of the HIV DNA fragment and extend through the vector sequences . If the concentration of the starting DNA increases, the amount of those longer DNA chains increases as well . These observations were also described by Lo et a! .9 This probe labelled with digoxigenin and generated by PCR is able to detect 10 pg of a 307 by HIV1 pol fragment either by dot- or Southern blot after overnight enzymatic colour reaction . This sensitivity is similar to that reported by Boehringer Mannheim' using a different labelling technique, the random priming method developed by Feinberg and Vogelstein . 12 Some investigators described better sensitivities, but used longer DNA hybrids .", " The sensitivity of a PCR-generated biotinylated probe is of the same level .' However, biotinylated probes develop a non-specific background colour reaction ." This did not occur with digoxigenin probe which could be incubated for over 24 h in the detection solution without appearance of significant non-specific background colour. This method was used to synthesize a labelled HIV1 pol fragment for probing PCR products, following the amplification of the genomic DNA of peripheral blood mononuclear cells from seropositive and seronegative subjects . We used the same pair of primers either in the preparation of the probe or in the amplification of the genomic DNA . Our results show, without ambiguities, that it is not necessary to use internal primers to synthesize the probe . None of the samples from seronegative subjects reacted either with the radiolabelled probe produced by random primer method nor with digoxigenin labelled probe generated by PCR. Other approaches have been used to detect and quantify PCR products which eliminate electrophoresis, Southern transfer, and use of radiolabelled probes . They also provide a rapid procedure for

2 69

diagnosis of genetic, malignant and infectious diseases .' 5-18 In conclusion this labelling and detection method is specific, sensitive, safe, economical and simple enough to be used in routine diagnostic laboratories for the detection of HIV-1 and other microbial pathogens .

Acknowledgements This work was funded by a grant from 'Junta Nacional de Investigacao Cientifica e Tecnologica' contract no . 87679 . N . Costa Taveira is the recipient of a research fellowship from 'Instituto Nacional de lnvestigacao Cientifica', Portugal .

References 1 . Laure, F ., Courgnaud, V., Rouzioux, C . et al. (1988) . Detection of HIV1 DNA in infants and children by means of the polymerase chain reaction . The Lancet ii, 538-41 .

Loche, M . & Mach, B . (1988) . Identification of HIVinfected seronegative individuals by a direct diagnostic test based on hybridization to amplified viral DNA . The Lancet ii, 418-21 . 3 . Imagawa, D . T ., Lee, M . H ., Wolinsky, S . M . et al . (1989) . Human immunodeficiency virus type 1 in homosexual men who remain seronegative for prolonged periods . The New England Journal of Medicine 320, 1458--62 . 4 . Taylor, G . R . (1988) . HIV detection by amplification . 2.

Journal of Clinical Pathology 41, 142-3 . 5 . Ou, C . Y ., Kwok, S ., Mitchell, S . W ., Mack, D . H . et al . (1988) . DNA amplification for direct detection of HIV-1 DNA of peripheral blood mononuclear cells . Science 239, 295-7 . 6.

Matthews, J . A . & Kricka, L . J . (1988). Analytical strategies for the use of DNA probes . Analytical Bio-

chemistry 7.

8.

9.

169, 1 .

Boehringer Mannheim (1989) . Biochemica-applications Manual : DNA labeling and Detection nonradioactive. Germany, Boehringer Mannheim GmbH Biochem i ca . Saiki, R . K ., Gelfand, D . H ., Stoffel, S . et al. (1988) . Primerdirected enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, 487-91 . Lo, Y-M .D ., Mehal, W . Z . & Fleming, K . A . (1988) . Rapid production of vector-free biotinylated probes using the polymerase chain reaction . Nucleic Acids Research 16, 8719 .

10 . Lion, T. & Haas, O . A . (1990) . Nonradioactive labeling of probe with digoxigenin by polymerase chain reaction .

Analytical Biochemistry

188, 335-7 .

11 . Maniatis, T ., Frisch, E . F . & Sambrook, J . (1982). Molecular Cloning: A Laboratory Manual . New York : Cold Spring Harbor Laboratory . 12 . Feinberg, A . P . & Vogelstein, B . (1983) . A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Analytical Biochemistry 132, 6-13 . 13 .

Azzi, A ., Zakrzewska, K ., Gentilomi, G ., Musiani, M . &



270

N . Costa Taveira et al .

Zerbini, M . (1990) . Detection of B19 parvovirus infections by dot-blot hybridization assay using a digoxigenin labeled probe . Journal of Virological Methods 27, 125-34 . 14 . Kimpton, C . P ., Corbitt, G. & Morris, D . J . (1989) . Detection of cytomegalovirus DNA using probes labeled with digoxigenin . journal of Virological Methods 24, 335-46. 15 . Nickerson, D . A ., Kaiser, R ., Lappin, S . et al . (1990). Automated DNA diagnostics using an ELISA-based oligonucleotide ligation assay . Proceedings of the National Academy of Science, U.S .A . 87, 8923-27.

16 . Wahlberg, I ., Lundeberg, J ., Hultman, T . & Uhlen, M . (1990) . General colorimetric method for DNA diagnostics allowing direct solid-fase genomic sequencing of the positive samples . Proceedings of the National Academy of Science, U .S .A . 87, 6569-73 . 17. Keller, G . H ., Huang, D . & Manak, M . M. (1989) . A sensitive nonisotopic hybridization assay for HIV-1 DNA . Analytical Biochemistry 177, 27-32 . 18. Coutlee, F ., Yang, B ., Bobo, L . et al. (1990) . Enzyme immunoassay for detection of hybrids between PCRamplified HIV-1 DNA and a RNA probe : PCR-EIA. AIDS Research of Human Retroviruses 6, 775-784 .