Journal $> 1992

of‘ Virologicul Mrthods. 40 (1992) 14% 154 Elsevier Science Publishers B.V. / All rights

VIRMET

145 reserved

/ 0166-0934/92/$05.00

01403

A non-radioisotopic reverse transcriptase assay using biotin- 11-deoxyuridinetriphosphate on primerimmobilized microtiter plates Takeshi Urabe”, Kouichi Sane”, Masashi Tannob, Junzo Mizoguchib, Masaru Otanib, Moon H. Lee’, Tomohiko Takasaki”, Hidenari Kusakabe’, David T. Imagawac and Masuyo Nakai” Toyo “Dr~purtmmt of‘ Microbiology. Osaka Me&al Coliqe, Osuka (Japan) , hRc~srarch Lahorutories. JO;O CO.. Ltd.*. Shizuoka (Japan) and ‘Department of’ Pediatrics, Harbor-UCLA Medical Center, Torrance, CA I USA) (Accepted

5 June

1992)

Summary

We developed a non-radioisotopic (non-RI) reverse transcriptase assay (RTA). The reverse transcriptase (RT) incorporates biotin-1 l-deoxyuridinetriphosphate (bio-dUTP) using a poly(rA) template hybridized with oligo(dT) primer that is immobilized on the surface of a 96-well microtiter plate. This assay is thus semi-automated by adapting it to an ELISA testing format. The incorporation of bio-dUTP was enhanced by adding cold dTTP to the reaction mixture, optimally in a molar ratio 4: 1 (dTTP:bio-dUTP). This non-RI RTA is more sensitive than the conventional RI assay for the detection of purified Rous-associated virus 2 (RAV-2) and of human immunodeficiency virus type 1 (HIV-l) lysate. Because of its simple procedure, higher sensitivity and non-use of RI materials, the assay can be utilized not only for virological studies but also for routine safety screening of biological products for retroviral contamination. Biotin-I I-deoxyuridinetriphosphate; Reverse transcriptase

ciency virus;

Rous-associated assay

Correspondwzce to: K. Sano, Department of Microbiology, Takatsuki-shi, Osaka 569, Japan. *The present name of Toyo Jozo is Asahi Chemical Industry.

Osaka

virus 2; Human immunodefi-

Medical

College,

Daigaku-cho,

2-7,

146

Introduction Reverse transcriptase (RT) is an RNA-dependent DNA polymerase that is essential for retrovirus replication. Thus a reverse transcriptase assay (RTA) to detect this enzyme is important for determining the presence of the virus and quantitating it (Temin et al., 1970; Baltimore, 1970). Recently RT assays have been used to identify RT-inhibiting antibodies to human immunodeficiency virus (HIV) as a clinical marker for the stage of HIV infection (Sano et al.. 1987; Laurence et al., 1987; Advani et al., 1989) and also been used for in vitro screening of RT-inhibitors as anti-HIV agents (Baba et al., 1991). RTAs have been improved in both procedure (Spira et al., 1986; Somogyi et al., 1990) and sensitivity (Tereba et al., 1977; Hoffman et al., 1985; Lee et al., 1987). Two main disadvantages remain: the use of radioisotopic (RI) materials; and a laborious procedure making it difficult to test a large number of samples. A nucleotide analogue, biotin- 11-deoxyuridinetriphosphate (bio-dUTP), has been successfully introduced into many procedures used in recombinant DNA research and has proved a suitable alternative for RI detection and localization of nucleic acid components (Langer et al., 198 1; Binder et al., 1986). The carbodiimide-mediated immobilization of DNA through 5’-terminal phosphate groups on solid supports has been used in molecular biology as a technique for nucleic acid hybridization (Ghosh et al., 1987). To overcome the disadvantages of the present RTA, we combined the use of bio-dUTP as a DNA precursor for RT reaction with a primer-immobilized microtiter plate. We describe a simple non-RI RTA that is more sensitive for the detection of retro-viruses.

Materials and Methods Polymerases

and virus lysates

Two purified RTs of Rous-associated virus 2 (RAV-2; Takara Shuzo) and Avian myeloblastosis virus (AMV, Pharmacia LKB Biotechnology) and one recombinant HIV RT, produced by the expression of the pol gene of HIV-l in Escherichia coli (rHIV-1, Eiken Chemicals), were used for our studies. Amounts of these three RTs commercially available were expressed in ‘units’ (one unit is defined as the enzyme activity that incorporates 1 nmol of [3H]dTMP into an acid insoluble form in 10 min at 37°C using poly(rA) . oligo(dT) as template-primer). HIV-l strain LAV-lBRU was provided by L. Montagnier (Pasteur Institute) and HIV-2 strain GH-1 by M. Hayami (Kyoto University). Molt-4 cells were infected by each HIV separately. and culture supernatants clarified by centrifuging at 1500 x g for 10 min. Each supernatant was passed through a GV Millipore filter and the virus concentrated by pelleting at 218 000 x g for 60 min in an ultracentrifuge (55p-72, Hitachi, Japan, RP-65 rotor). A lysate was made from the virus pellet by solubilizing in 50 mM Tris-HCl (pH 7.8)

147

buffer containing 0.5% Triton X-100, 0.8 M NaCl, 0.5 mM phenylmethylsulphonylfluoride and 20% glycerol, and stored at 4°C. All RTs and viral lysates were diluted with phosphate-buffered saline containing no Ca’+ and Mg2+ (PBS; Dulbecco et al., 1954).

( I ) Reugen Is 01igo(dT)19__24, poly(rA), and poly(dA) were purchased from Pharmacia, and dissolved at 1 ,ug per ~1 in TE buffer (10 mM Tris-HCl (pH 7.5), 1 mM EDTA). dTTP was from Boehringer Mannheim Biochemicals and prepared at 0.3 mM in TE buffer, and bio-dUTP (0.3 mM in 50 mM Tris-HCl, pH 7.5) was from Enzo Biochemicals. A concentrated (2 x ) reaction buffer previously reported (RB; Lee et al., 1987) was used for the RT reaction. The buffer for washing the microtiter plates consisted of 0.1 M Tris-HCl (pH 9.5), 0.5 M NaCl, and 0.05 M MgClz. (2) Procedure fbr non- RI RTA The principle of this procedure is schematically summarized in Fig. 1. Oligo(dT) was immobilized via its 5’-terminal phosphate group to the surface of aminated microtiter plates (Sumitomo Bakelite) by a modified method (Fig. la; Ghosh et al., 1987). In brief, 50 ~1 of 100 mM l-methyl-imidazole-HCl buffer (pH 7.0) containing 100 mM l-ethyl-3-dimethylaminopropyl carbodiimide and 4 pg per ml of oligo(dT)t(, _24was placed into each well of the plates, and the plates incubated at room temperature for 24 h. Then the wells were washed three times with 200 ~1 of 0.1 M Tris-HCl (pH 7.5) saline buffer to remove free primers, the plates were dried, sealed with vinyl sheets, and stored at 4°C until use. Forty-five ,ul of the 2 x RB, 1 ~1 of poly(rA), 0.8 ~1 of biodUTP, and 3.2 ~1 of dTTP were placed into the well of the primerimmobilized microtiter plates and mixed with 50 ~1 of purified RT or viral lysate. Poly(dA) was used as template at the same concentration in certain experiments in place of poly(rA). The plates with reaction mixture (100 ~1) were sealed with vinyl sheets to prevent drying and incubated at 37°C for O-18 h. The reaction was stopped by the addition of 10 1.11of 5 M NaCl and the wells were washed five times with 200 ~1 of the washing buffer (Fig. lb). 100 ~1 of 3% bovine serum albumin in the washing buffer were placed into each well in order to block the nonspeci~c binding of streptavidin-conjugated alkaline phosphatase @A-ALP; streptavidin 0.088 mg/ml; alkaline phosphatase 2034 U/ml; Bethesda Research Laboratory), and the plates were sealed again with vinyl sheets and incubated at 37*C for 1 h. After aspirating the blocking solution, 50 ill of SA-ALP, diluted IOOO-fold with the washing buffer, was placed into each well for the detection of the bio-dUTP DNA product. After 1 h at room temperature, each well was washed five times with 200 /ll of the washing buffer to remove any unbound SA-ALP, and then 50 ~1 of the substrate, paranitrophenyl phosphate (PNPP; I mg/ml) in 50 mM diethanol-

148

I-

RT /VIRUS RB BiodUTP ($) dTTP (t) poWA)

b

b

SA-ALP

(>ALP)

i*r

Fig. I. The principle of a non-RI RT reaction employing

RTA.

bio-dUTP.

(a) Immobilization (c) DetectIon

of primers to the well of a microtltcr

of hiotinylated

DNA

plate. (b)

by the binding of SA-ALP.

amine-MC1 buffer (pH 9.5) containing 1 mM MgC%, was placed into each well. The plates were incubated for IO-30 min at 37 C, and 50 ~1 of 0.5 N NaOH was added into each well to stop further color development (Fig. Ic). The absorbance (A) of each well at 405 nm was measured by a microplate reader (NJ-2001; Nippon InterMed). For the blank (A = 0) distilled water was substituted for both RT and SA-ALP solution. Radioiso topic reverse transcriptux~ ama?, (RI RTA / RI RTA was performed

by the procedure

previously

reported

(Lee et al.,

149

1987) by polymerizing (dT) template-primer.

a [‘HI-TTP

precursor

into DNA on a poly(rA).

oligo

Results

In order to determine the optimal conditions for this assay, we first examined the effect of dTTP addition to bio-dUTP at a constant concentration of total deoxynucleotides (12 ,uM). The incorporation of bio-dUTP into DNA was enhanced by adding dTTP, although the concentration of bio-dUTP itself was decreased. The incorporation of bio-dUTP was highest when the molar ratio of dTTP to bio-dUTP was 4:l with 1.O mU of RAV-2 RT (data not shown). Therefore, the ratio of 4: 1 (dTTP 9.6 pM/bio-dUTP 2.4 PM) was considered to be appropriate under these assay conditions, and was used in the subsequent experiments. To examine the influence of incubation time on this RTA, the RT reaction was allowed to proceed over a period of time. The kinetics of the RT reaction for various amounts of RAV-2 RT at 37°C in an 18-h period is shown in Fig. 2. With a smaller amount of RT (2.5 mU of RAV-2 RT), the A values were proportionally related to reaction time for 18 h. As the amount of RT was increased, however, the reaction had linear kinetics for only an initial shorter period and its A value exceeded the quantitative A range (62.000) earlier. To determine the appropriate reaction time for quantitating RT, we next quantitated various amounts of RT by non-RI RTA in two different reaction times and compared the results with those of RI RTA (Fig. 3). The 18-h nonRI assay was linearly related to only a low range of up to 1 mU of RAV-2 RT, while the 2-h non-RI assay was linearly related up to at least 16 mU of RT (r =0.96, P ~0.01). On the contrary, the 18-h RI RTA was linearly related up to at least 16 mU of RT (v= 0.97, P