Development of A Fission Yeast Cell-Based Platform for High Throughput Screening of HIV-1 Protease Inhibitors

Abstract

Background: HIV-1 protease inhibitor (PI) is one of the most potent classes of drugs in combinational antiretroviral therapies (cART). When a PI is used in combination with other anti- HIV drugs, cART can often suppress HIV-1 below detection thus prolonging the patient's lives. However, the challenge often faced by patients is the emergence of HIV-1 drug resistance. Thus, PIs with high genetic-barrier to drug-resistance are needed.

Objective: The objective of this study was to develop a novel and simple fission yeast (Schizosaccharomyces pombe) cell-based system that is suitable for high throughput screening (HTS) of small molecules against HIV-1 protease (PR).

Methods: A fission yeast RE294-GFP strain that stably expresses HIV-1 PR and green fluorescence protein (GFP) under the control of an inducible nmt1 promoter was used. Production of HIV-1 PR induces cellular growth arrest, which was used as the primary endpoint for the search of PIs and was quantified by an absorbance-based method. Levels of GFP production were used as a counter-screen control to eliminate potential transcriptional nmt1 inhibitors.

Results: Both the absorbance-based HIV-1 PR assay and the GFP-based fluorescence assay were miniaturized and optimized for HTS. A pilot study was performed using a small drug library mixed with known PI drugs and nmt1 inhibitors. With empirically adjusted and clearly defined double-selection criteria, we were able to correctly identify the PIs and to exclude all hidden nmt1 inhibitors.

Conclusion: We have successfully developed and validated a fission yeast cell-based HTS platform for the future screening and testing of HIV-1 PR inhibitors.

Keywords: HIV-1 protease (PR); HIV-1 protease inhibitor (PI); atazanavir (ATV); fission yeast (Schizosaccharomyces pombe); green fluorescence protein (GFP); high throughput screening (HTS); transcriptional nmt1 inhibitor (TNI)..

Fig. (1).

Absorbance-based measurement of HIV-1 PR-induced growth inhibition and rescue by ATV in the RE294 strain. (A) Inducible production of HIV-1 PR. RE294 cells from the stock culture were cultured in the PMG minimal medium including 1 μM of thiamine to prevent PR gene expression (PR-off), or medium without thiamine to induce the expression of PR (PR-on). The growth curves were generated over time to show the difference between the PR-on and PR-off cultures as part of the quality control of the stock culture. Typically, an nmt1-driven gene is fully expressed at about 16 h after gene induction [39, 40]. (B) Miniaturization and measurements of HIV-1 PR-induced growth arrest and suppression by ATV. RE294 culture from the stock culture was prepared as described in MATERIALS AND METHODS. 40 μl of the cell culture (2 × 10⁴ or 4 × 10⁴ cell/ml) was loaded into a 384-well plate. ATV was added at a final concentration of 150 μM. A value was measured at 620 nm after 72 h post-gene induction. Final A value of each well was normalized by deducting average A values of the media-containing wells. Values shown are means of three independent experiments ± standard errors (SE). Asterisks indicate a statistically significant difference in comparison with PR-off (one-way analysis of variance analysis by Prism 7; ***, p < 0.001).

Fig. (2).

Measurement of nmt1-driven GFP expression by fluorescence intensity in the fission yeast SP223-GFP strain. A start cell concentration of 4 × 10⁴ cells/ml was used in the 384-well microtiter plate as described in MATERIALS AND METHODS. After 48- (2d) and 67-h (3d) post-gene induction, F values (A) and A values (B) were measured. The final F value of each well was normalized by deducting the average F values of the media-containing wells. (A) shows the average F values of GFP-off (n = 176) and GFP-on (n = 176) at two indicated time points. (B) shows A readings of the same plate. The background averages of F and A values (n = 32) were deducted from each of the F and A readings in (A) and (B), respectively. The values are the means of three independent experiments ± SE. Asterisks indicate a statistically significant difference in comparison with GFP-off (one-way analysis of variance analysis by Prism 7; ***, p < 0.001).

Fig. (3).

Pilot runs of HTS of protease inhibitors using the fission yeast RE294-GFP strain. (A) A 384-well microtiter plate map of the small molecule compounds tested. (B) and (C) Results of all 320 compounds tested with A values shown on the left axis and F values shown on the right axis. All compounds were tested with two different drug concentrations, i.e., 75 μM (B) and 150 μM (C). The values of F and A were calculated for each well and the average and standard deviation of three experiments are plotted. The A (●) and F (▲) values of all compounds are shown in descending orders. In each dataset, the data points that meet the predicted selection criteria are shown on the left side of chart and separated by a vertical dashed line. The data points shown in red are the F (top) and A (bottom) values of the protease inhibitor ATV. Localizations of the 4 BT-containing wells are boxed in the L and H plates.

Fig. (4).

Identification of protease inhibitors from false positives by criteria of single (A or F) or double (A and F) selections. (A) shows measurements of A values with 75 μM (a) or 150 μM (b); (B) shows results of F values with 75 μM (a) or 150 μM (b), respectively. A values within the range of 0.8 – 1.2 are considered positive hits; F values within the range of 22.6 – 39.0 are considered positive. Symbols shown in red that are marked by a circle are true PIs. Symbols shown in black circles are TNIs.