Quantitative micro P30 and reverse transcriptase assays for Moloney murine leukemia virus

Journal of Virological Methods

ELSEVIER

Journal

Quantitative

of Virological

Methods

48 (1994) 109-l 18

micro P30 and reverse transcriptase Moloney murine leukemia virus C.A.

Wilson”, M.V. Eiden”,

J.W.

assays for

Marshb’*

“Laboratory of Cell Biology, Building 36, Room 2005. National Institute of Mental Health, Bethesda, MD 20892, USA; ‘Laboratory of Molecular Biology, Building 36. Room 1003, National Institute of Mental Health, Bethesda, MD 20892, USA Accepted

4 January

1994

Abstiact

An anti-P30 immunohistochemical and a reverse transcriptase assay for Moloney murine leukemia virus (MoMLV) are adapted to 96-well plates. The assay results are shown to be directly proportional to the number of infectious particles, and can therefore be used to estimate the infective titers of a virus preparation. The micro P30 assay yields a direct estimate of infectious centers, and the reverse transcriptase assay quantitates progeny from a single cycle of replication. The semi-automated nature of these assays is well suited to the analysis of a large number of samples and therefore permits the examination of the efficiency of the process of retroviral/MoMLV infection under varied times or conditions. Key words:

Reverse transcriptase;

Retrovirus; MoMLV; Moloney murine leukemia virus; P30

1. Introduction In our studies of the mechanisms underlying the highly variable process of ecotropic retroviral infection, we found a great need to assay large numbers of samples in an expedient and reproducible fashion. In the absence of reporter genes present in many retroviral vectors, quantitation of infection by wild-type retroviruses is technically difficult. Normally it involves titering the viral preparation on susceptible cells in culture followed by detection of virally-induced syncytia (XC plaque assay; Rowe et al., 1970), or immunohistochemical staining to detect plaques or infectious foci (Nexo, 1977). Alternatively, virus particles may be enumerated with electron *Corresponding

author.

0166-0934/94/$07.00 0 1994 Elsevier Science B.V. All rights reserved SSDI 0166-0934(94)00007-4

110

C.A. Wilson et al./Journal

of Virological Methods

48 (1994) 109-118

microscopy, although few laboratories can do this routinely because of the required instrumentation. Finally, although it has not been established as a useful quantitative measure of virus particles, reverse transcriptase activity is commonly used to demonstrate retroviral replication. The semi-automation of the reverse transcriptase assay (Goff et al., 1981; Psira et al., 1987; Gregersen et al., 1988) has permitted adaptation to 96-well dishes, but these methods use the assay in a qualitative way. In this report we describe the adaptation to 96-well plates of an anti-viral P30 plaque assay and of a reverse transcriptase assay for the Moloney murine leukemia virus (MoMLV), and compare the sensitivities and useful ranges of these two assays. We establish that within defined ranges, both assays are highly quantitative and can be correlated with infective titer. To correlate the reverse transcriptase activity of viral progeny and the generated P30 infectious centers with the input number of infectious particles, we first demonstrate that [3H]TTP incorporation into reverse transcribed DNA is directly proportional to input reverse transcriptase concentration. We then demonstrate that the progeny-derived reverse transcriptase levels and the developed P30 foci are directly proportional to the input number of infectious particles.

2. Materials and methods 2.1. Cells and viruses NIH 3T3 murine Iibroblasts (ATCC CRL 1658) were maintained in Dulbecco’s modified Eagle’s medium (DMEM) (Whittaker M.A. Bioproducts, Walkersville, MD) supplemented with 10% fetal bovine serum, 100 U of penicillin per ml, 100 pg/ml of streptomycin and 40 mM glutamine (complete DMEM). Concentrated MoMLV virion particles were obtained from Advanced Biotechnologies, Inc. (Columbia, MD). The particle preparation was concentrated lOOO-foldby high-speed pelleting using continuous-flow ultracentrifugation. The number of virus particles were determined by electron microscopy to be 8 x 10” particles/ml. 2.2. Virus infections MoMLV virions were serially diluted in DMEM supplemented with 3 pg/ml of polybrene over a million-fold range of virus particles. Each dilution was assayed directly for RT activity, in order to determine the value of RT activity relative to a given virion particle number. NIH 3T3 cells were seeded in 96-well plates at a density of approximately 5000 cells/well in complete DMEM 1 day prior to infections. On the day of infections, the media was changed to complete DMEM supplemented with 3 pg/ml of polybrene. The appropriate dilution of virus was then added to each well. 6 h post-exposure to virus, the NIH 3T3 cells were rinsed three times with complete DMEM without polybrene and fed with 200 ~1 of complete DMEM per well. 20 ~1 of cell media from each well was assayed for reverse transcriptase activity as described below at the time points specified.

CA.

Wi&wzet al./Journal of Virological Methods 48 (1994) 109-118

111

2.3. Micro reverse ~ra~sc~ip~aseassay

This assay uses modifications of previously reported reverse transcriptase assays (Goff et al., 1981; Spira et al., 1987; Gregersen et al., 1988). To 20 ~1 sample containing virus (in a round bottom 96-well plate or microtube) 5 ~1 of solubilization buffer [120 mM NaCl, 1% Triton X-100, 50 mM Tris (PH 8.3)] was added, followed by a 15 min ambient temperature incubation. This was followed by the addition of 25 ~1 of freshly made 2 x substrate buffer, and incubated for 2 h at 37°C in a non-CO* incubator [2 x substrate buffer: 50 mM Tris, 8.3, 10 mM DTT, 1.2 mM MnC12, 10 pg/ml poly(rA) - p(dT) 12-18 (Pharmacia, Piscataway, NJ), t3H]TTP, 5 pCij25 $1. The incubation was terminated by the addition of 5 ~1 0.25 M EDTA followed by blotting the synthesized DNA on a DEAE filtermat (LKB-Wallac, Gaithersburg, MD} through use of a harvesting apparatus (Skatron, Sterling, VA} with a rinse of 1% Na2HP04 (pH 7.0) containing 1 mM EDTA. Radioactivity was then determined via liquid scintillation. 2.4. Micro P30 plaque assay This assay incorporates previously reported methods (Nexo, 1977; Marsh and Klihman, 1990). Following a 3 day incubation of virally-infected NIH 3T3 cells, as described above, the wells were rinsed briefly with PBS/O.OS% Tween-20 (PBSTween20). The plate was then tapped dry, followed by a 100 ~1 addition of acetone/methanol (1:1, v/v), left at room temperature for 2 min, and followed by a rinse of PBSTween20. (Tapped dry, the plate can be stored at -20°C.) Goat antisera against Rauscher murine leukemia virus P30 core antigen (NC1 Repository, no. 778-195) was diluted into PBS-Tween20 and filtered with a 0.22 pm filter to remove aggregates, then further diluted into prefiltered PBS-Tween20 containing 15% fetal calf serum. The extent of sera dilution will vary from lot to lot, so optimization to minimize background must be performed (dilution typically is lOO-fold). After the plates were rinsed with PBS-Tween20 and tapped dry, 50 ~1 of the diluted anti-P30 sera was added per well, and then incubated at 37°C for 30 min. The plates were then rinsed with PBS-Tween20 three times for 5 min periods at ambient temperatures. This was followed by the addition of 50 ~1 rabbit anti-goat IgG-alkaline phosphatase conjugate (Kirkegaard and Perry, Gaithersburg, MD, no. 05-13-06) that had been filtered and diluted into PBS-Tween20-fetal calf sera as above. Following a 37”C, 30 min incubation, the plate was washed again three times for 5 min, and tapped dry. Foci of cells containing alkaline phosphatase activity were then visualized by addition of a 5-bromo-3-chloroidolyl phosphate solution (Sigma Chemical Co., St. Louis, MO, no. 710-3) in a low melting temperature agarose, as previously described (Marsh and Klinman, 1990). The agarose minimizes diffusion of the developed blue color that appears within hours at room temperature. P30 positive foci can be enumerated with use of a magnifying glass or a low power inverted microscope. Linear regressions and other statistical analyses were carried out with StatView (Abacus Concepts, Berkeley, CA), and Cricket Graph (Cricket Software, Philadel-

112

C.A. Wilson et al./Journal of Virological Methods 48 (1994) 109-118

phia, PA) software. The efficiency of infection was determined from the slope, y/x, of the linear plot in the P30 assay, which defines the relationship of P30 infectious centers with the number of virus particles.

3. Results 3.1. Reverse transcriptase input concentration correlates linearly with (3H]TTP incorporation into reverse transcribed DNA

Wild-type Moloney murine leukemia virus (MoMLV) was concentrated by ultrafiltration as a single preparation. The whole viral particle count was achieved by electron microscopy, and the suspension was frozen as aliquots. The 96-well microplate reverse transcriptase assay was performed as described in Section 2 over a range of 3 x lo* to 1.6 x lo8 virus particles/well. As shown in Fig. 1, the assay displays saturation (non-linearity) at 80 million viral particles per well, but is remarkably linear with the added viral particle count (R’ = 0.99) through a range of 0 to 20

0

25

50

75

ViWS

100

125

150

175

2 10

Particles (millions)

Fig. 1. Analysis of reverse transcriptase activity associated with MoMLV virus particle number. Reverse transcriptase activity is represented as cpm of [3H]TTP incorporated into nucleic acid. Each sample was performed in duplicate as described in Section 2. Inset graph shows portion of curve with linear relationship between reverse transcriptase activity and number of virus particles b= 2.58 + 15.87x; R’=0.994).

million virus particles (Fig. 1, inset). The rationale for this experiment was to determine the upper limit of the reverse transcriptase enzyme activity (as measured by cpm [3H]TTP incorporated into reverse transcribed DNA/2 h incubation) which was no longer first order with respect to the reverse transcriptase enzyme input (as measured by viral particle input). These results imply that input reverse transcriptase concentrations which generate greater than 3 x 10’ cpm/2 h will not correlate linearly with the experimentally derived cpm values. The linear relationship below this cpm level can be used to quantitate reverse transcriptase input and thus, viral input. It does not, however, describe the number of infective particles, nor should it be assumed that the reverse transc~ptase activities wiil be identical relative to particle number from preparation to preparation. To correlate reverse transcriptase activity with infectivity of a viral stock suspension, we first examined the kinetics of reverse transcriptase appearance following viral infection of cell culture. NIH 3T3 tibroblasts were incubated with a dilution of the stock MoMLV at time zero. Incubation proceeded over 3 days during which

60000

Time

(hours)

Fig. 2. Kinetics of reverse transcriptase activity generated at various time points post-exposure of NIH 3T3 cells to MoMLV virions. NIH 3T3 fibroblasts were infected with approximately 3 x IO6 viral particles. 20 ~1 of cell media was assessed from four wells at every time point. Data presented are the mean value of these four determinations + S.E.

114

C.A. Wilson et al.lJournal

of Virological Methods

48 (1994) 109-118

aliquots of cell media were examined for reverse transcriptase activity. A relatively constant level of reverse transcriptase is generated between 24 and 32 h of incubation following addition of the virus to the culture (Fig. 2). Subsequent time points yielded an exponential growth in reverse transcriptase activity, consistent with a secondary, spreading infection. We thus chose the 30 h time point to establish the correlation of the viral particle number added to culture with the resulting (progenyderived) reverse transcriptase activity. 3.2. Input virus particle number correlates linearly with progeny-derived

reverse

transcriptase activity at 30 h

Serial dilutions of the viral stock were added to NIH 3T3 cultures and following a 30 h incubation, reverse transcriptase activities were determined. As shown in Fig. 3 the lowest infecting viral level that is differentiated from background in this 30 h incubation is 1.25 million particles added. There is a relatively linear response between virus particle input at and below 20 million and the reverse transcriptase

20

Virus

40

60

Particles Added (milllons)

60

To

Cells

Fig. 3. Reverse transcriptase activity generated from NIH 3T3 cells exposed to various dilutions of MoMLV virions. Reverse transcriptase activity was assayed from 20 ~1 of cell media 30 h post-exposure to virus. RT activity is represented as cpm of [3H]TTP incorporated. The data plotted are the mean of triplicate determinations + SE.

C.A. Wilson et aLlJournal of Virological Methods 48 (1994) 109-118

0

Fig. 4. Detection Values represent cells exposed to tions f S.E. Least

2

4 Virus

6 8 10 12 14 16 18 Particles Added to Cells (thousands)

115

:

of MoMLV P30 antigen positive foci associated with MoMLV virus particle number. the number of foci staining positive for MoMLV gag antigen, P30, per well of NIH 3T3 various dilutions of MoMLV. The data plotted are the mean of triplicate determinasquares lit: y = 0.260 +0.0029x, R* = 0.997.

activity generated by the infection (I?* = 0.97). Above 20 million input particles/well the resulting progeny-derived reverse transcriptase activity loses linearity with a plateau forming at approximately 9000 cpm/well. As the direct reverse transcriptase enzymatic assay of virus particle (Fig. 1) does not show saturation at lo-fold higher reverse transcriptase activities, the plateau in Fig. 3 presumably represents a loss in the efficiency of infection. 3.3. Input virus particle number correlates linearly with progeny-derived P30 infectious center number

We then evaluated serial dilutions of the viral stock for their ability to produce P30 positive infectious centers. The relationship between virus particle number added and the number of infectious centers was also linear, with the upper limit being the number of plaques discernible in the wells of the 96-well plate (approximately 40/well). This assay was considerably more sensitive than the reverse transcriptase assay, in that infection by an input of 1000 particles was discernible. The slope of the fitted line in Fig. 4 defines the efficiency of infection of this viral preparation as 0.29% (see Section 2).

4. Discussion The micro P30 and reverse transcriptase assays developed and compared in this report make use of aspects from numerous previous assays (Nexo, 1977; Goff et al., 1981; Spira et al., 1987; Gregersen et al., 1988) which have been altered to permit semi-automated examination of multiple conditions. The linear correlations between viral particle number and reverse transcriptase activity and between particle number

116

C.A. Wilson et al./Journal of Virological Methods 48 (1994) 109-118

and the number of P30 infectious centers, as established in this report, imply that within a defined range reverse transcriptase activity can be used to estimate infectious particles in concert with the P30 assay. Additionally, we define the limits within which these assays can be used quantitatively. We have used a concentrated, frozen MoMLV preparation to permit examination of viral function in multiple analytical methodologies. Although much viability is lost in the processing (as demonstrated by the 0.29% infection efficiency calculated from the P30 assay), this preparation permitted day-to-day reproducibility within our developing assay systems. The upper limit of the P30 assay is largely defined by the area of the wells. In the case of the flat-bottom 96-well plate, this is approximately 40 P30 positive infectious centers or foci/well. As a single P30 foci can be detected, there is no theoretical lower limit. This type of assay has the advantage that samples can be examined at various dilutions with each dilution being done repetitively in multiple wells. The micro reverse transcriptase assay has both upper and lower limits. Data from the lower limit must be sufficiently above background to be statistically significant, and data reaches the upper limit as saturation (non-linearity) of the assay is reached. Appropriate dilutions can extend the useful range. The first virus concentration from Fig. 3 that gave above background progeny-derived reverse transcriptase readings was 1.25 x 1O6input particles. At the 0.29% infection efficiency defined by the P30 assay, this would imply that approximately 4000 infectious particles are required to reach this detectable level of reverse transcriptase activity, and that this reverse transcriptase activity is linear up to 60 000 infectious particles. In the absence of sample dilution, both assays are quantitative over a range greater than one order of magnitude, and within these linear portions the two assays can be correlated directly. As the P30 assay can detect a single infectious particle, there is no lower limit, and in theory, appropriate dilution of the sample should bring any concentration of infectious particles into the range of the assay. Relative to our P30 assay, the reverse transcriptase assay described here is more rapidly completed, and by using a cell harvester, quantitation can be achieved instrumentally. As noted previously (Gregersen et al., 1988), the viruses are lysed by detergent prior to the initiation of this assay, and thus, reverse transcriptase activity is a safe means of quantitating potentially hazardous retroviruses.

5. References Goff, S., Traktman, P. and Baltimore, D. (1981) Isolation and properties of Moloney murine leukemia virus mutants: use of a rapid assay for release of virion reverse transcriptase. J. Virol. 38, 239-248. Gregersen, G.P., Wege, H., Preiss, L. and Jentsch, K.D. (1988) Detection of human immunodeficiency virus and other retroviruses in cell culture supematants by reverse transcriptase microassay. J. Virol. Methods 19, 161-168. Marsh, J.W. and Khnman, D.M. (1990) Development of a diphtheria toxin mutant conjugate directed against antigen-specific B cells expressing high affinity surface Ig. J. Immunol. 144, 10461051. Nexo, B.A. (1977) A plaque assay for murine leukemia virus using enzyme-coupled antibodies. Virology 77, 8499852.

C.A. Wilson et al./Journal of Virological Methods 48 (1994) 109-118

117

Rowe, W.P., Pugh, W.E., and Hartley, J.W. (1970) Plaque assay techniques for murine leukemia viruses. Virology 42, 1136-l 139. Spira, T.J., Bozeman, L.H., Holman, R.C., Warfield, D.T., Phillips, S.K., and Feorino, P.M. (1987) Micromethod for assaying reverse transcriptase of human T-cell lymphotropic virus type III/ lymphadenopathy-associated virus. J. Clin. Microbial. 25, 97-99.