Silencer® CellReady™ siRNA Libraries
Identifying Genes Involved in G2/M Cell Cycle Progression: An siRNA Screen Described Step by Step

Gene silencing using libraries of siRNAs is revolutionizing the study of gene function. Ambion’s complete suite of RNAi analysis products makes these experiments easy and inexpensive. The siRNA screening experiment described here demonstrates the ease of designing and carrying out a successful siRNA library screen without overextending one’s budget. Results from this initial screen showed that silencing CDC2 expression caused an increase in the percentage of cells in G2/M phase in cancer and normal cell lines. The screen also identified genes impacting cell cycle progression that have differing levels of importance in cancer versus normal cells.

Introduction

The human genome is comprised of at least 23,000 protein-encoding genes of which only ~15,000 have been functionally annotated; leaving much opportunity for scientific investigation and discovery.  RNAi can provide a wealth of information about mammalian gene function quickly, for little investment. Thanks to the development of new tools and technologies, it is possible to keep costs low and make siRNA screening accessible to any laboratory.

The four main steps in siRNA screening include:

a. Choose siRNA library
b. Choose cell line(s)
c. Optimize siRNA delivery
d. Choose and optimize assay

2. Perform screen

3. Assay phenotype

a. Perform assay
b. Designate hit criteria and identify hits

4. Validate results and perform follow up experiments

Here we show how to apply these steps to an actual siRNA screening experiment. The goal was to identify genes that, upon silencing, induced a G2/M arrest phenotype specifically in cancer-derived A549 cells and not in normal lung cells (WI-38 cells). Genes that control entry and progression of G2/M phase specifically in cancer cells may be good targets for cancer therapy and treatment.

Step 1. Planning and Preparation

a. Choose siRNA Library. For this study, we focused on genes widely recognized for their roles in cell cycle progression. An additional goal was to keep material and labor costs low. Thus, transfections were to be performed with a multi-channel pipettor, so obtaining the library in a ready-to-use format was important for us. Three individual siRNAs to each target were used to avoid the high false positive and false negative hit rates associated with siRNA pools [1], and to provide confidence that the observed hits were due to silencing of the intended genes. Based on these criteria, the Silencer® CellReady™ Popular Gene siRNA Library was used. This siRNA library consists of 3 individual siRNAs to each of 80 different genes pre-aliquotted into 96 well plates in triplicate in ready-to-transfect amounts.

b. Choose cell line(s). Because we wanted to understand how silencing each of 80 different genes affected G2/M transitions in a normal versus a cancer derived cell line, normal diploid lung fibroblast cells (WI-38) and an epithelial lung adenocarcinoma cell line (A549 cells) were selected.

c. Optimizing siRNA Delivery. Using optimal siRNA delivery conditions eliminates the most common causes of unsuccessful gene silencing experiments. Thus optimizing transfection conditions for our two chosen cell lines was a key step in the screening process. Here the Silencer CellReady siRNA Transfection Optimization Kit was used. This kit includes three 96 well plates pre-filled with alternating wells of Silencer GAPDH siRNA and Silencer Negative Control #1 siRNA. The kit was used to test three different transfection reagents (siPORT™ NeoFX™, and two competitors' reagents, Reagent B and Reagent C) with each cell type.  Each reagent was tested at volumes of 0, 0.15, 0.3, and 0.6 µl, with cells plated at two different densities, 4,000 and 8,000 cells/well. Reverse transfection, which involves simultaneous transfection and plating of cells, was used to maximize siRNA delivery and streamline the  delivery procedure by eliminating an entire day from the process [2].

A GAPDH enzymatic assay, Ambion’s rapid fluorescence-based KDalert™ GAPDH Assay Kit, was used to measure GAPDH silencing 48 hours post transfection (Figure 1). This assay was chosen for its ease of use and versatility. By comparing GAPDH enzymatic activity in cultures transfected with GAPDH siRNA versus those transfected with negative control siRNA, the level of gene silencing and thus siRNA delivery efficiency was readily determined. Transfection conditions were also evaluated for induction of cellular toxicity by comparing GAPDH activity in negative control siRNA transfected cells versus nontransfected cells. The optimization experiment resulted in distinct delivery conditions for each cell line. For the A549 cells, optimal conditions were reverse transfection of 8000 cells/well with 0.3 µl of siPORT NeoFX. For WI-38 cells, reverse transfection of 8000 cells/well with 0.15 µl Transfection Agent B was most effective.

Figure 1. Transfection Optimization Using the Silencer® CellReady™ siRNA Transfection Optimization Kit and KDalert™ GAPDH Assay Kit. Silencer GAPDH siRNA (positive control) and Silencer Negative Control #1 siRNA (negative control) were transfected at 4,000 and 8,000 cells/well into A549 and WI-38 cells using 0, 0.15, 0.3, and 0.6 µl siPORT™ NeoFX™ and two competitor’s transfection agents, Transfection Agent B and C. 48 hr post transfection cells were assayed for GAPDH activity using the KDalert Assay to identify optimal transfection conditions for each cell type. Taking percent remaining gene expression (relative to negative control siRNA) and percent viability into account, 0.3 µl of siPORT NeoFX for A549 cells and 0.15 µl of Transfection Agent B for WI-38 cells transfected at 8000 cells/well gave maximal results (orange).

d. Choose and optimize assay. To ensure proper interpretation of siRNA screening results, it is critical that the selected assay provide the required signal-to-noise ratio and reproducibility. In this screen, cells were to be assayed 96 hours post transfection to determine the percentage of cells in G2/M phase. The Acumen Explorer™ laser-scanning fluorescence microplate cytometer (TTP LabTech) was used to classify cells in G1, S, or G2/M phases by the fluorescence intensity of chromatin stained with propidium iodide. The instrument measures the DNA content of stained nuclei, and its software classifies the cells as to their cell cycle phases. To optimize the assay, cells were treated with vinblastine sulfate, which belongs to the general group of chemotherapy drugs known as plant (vinca) alkaloids, to induce mitotic arrest. Vinblastine treated cell populations have a higher percentage of cells in G2/M phase than untreated cells. Vinblastine concentrations that arrested cells in G2/M and that were accurately identified on the instrument using propidium iodide staining were determined (validation data not shown). These conditions were used to perform the actual siRNA library screen detailed below.

Step 2. Perform Screen

Each siRNA was tested in triplicate in each cell line. Reverse transfection was used to deliver the siRNAs to WI-38 cells (Figure 2A) and A549 cells (Figure 2B). A portion of each cell sample was aliquotted into black-walled clear bottom 96 well plates for analysis on the Acumen Explorer instrument. Using a multichannel pipettor, reverse transfection of the 1584 samples took about 2 hours (for one person).

Figure 2. Effects of Gene Silencing on Cell Cycle Progression. Data resulting from the actual screening experiment in which siRNAs from the Silencer® CellReady™ Popular Gene siRNA Library were delivered to both normal diploid lung fibroblast cells, (WI-38; Panels A & B) and an epithelial lung adenocarcinoma cell line (A549; Panels C & D). The grey horizontal lines represent the criteria we used to define a hit (see article for details). These lines are set 30% above and 30% below the average results from the negative control (NC) siRNAs (shown as 100%). Red boxes denote genes identified as "hits", and blue boxes denote genes identified as "borderline hits".

Step 3. Assay Phenotype

a. Perform assay. 72 hr post transfection, media from three of the “cells alone” wells in each plate was replaced with media + vinblastine sulfate (3 µM) and incubated overnight at 37°C. Cells were then fixed with paraformaldehyde (2% final concentration), washed with 1X PBS, and stained with 3 µM propidium iodide for >1 hr at room temperature. Plates were then individually scanned with the Acumen Explorer, and the percent of cells in G2/M versus other phases of the cell cycle was determined.

b. Designate hit criteria and identify hits. A goal of this screen was to identify genes that, when silenced, resulted in significantly increased or significantly decreased percentages of cells in G2/M phase. The threshold for calling an siRNA a “hit” was set based on comparison of the data to negative control transfected samples. The hit threshold was set at two times the average standard deviation of all of the transfected cell samples (gray horizontal lines, Figure 2); this resulted in the threshold being 30% above and below the average of the negative control transfected samples. Vinblastine treated samples were used as a positive control.

To ensure that the hits reported were due to target specific silencing, the definition of a “hit” was refined to include only those genes where at least two of the three targeting siRNAs resulted in G2/M phase cell percentages that fell outside of the set thresholds. Using these stringent hit criteria, four genes were identified for the A549 cells and four genes for the WI-38 cells that when silenced, increased or decreased the percentage of cells in G2/M phase above the set threshold. Those target genes are listed in Figure 3 (Venn diagram). Silencing of the CDC2 gene was found to induce apparent G2/M arrest in both cell lines while the rest of the hits were cell line specific. The WI-38 specific hits included BRAF, ESR1, and MAPK3K14, while A549 specific hits included CDC42, CDKN1A, and ESRRA.

Figure 3. Pathways in which Hits Were Found. Silencing of certain genes in WI-38 and A549 cells led to misregulation of the cell cycle. These results are presented in a genetic network (Ingenuity Pathway Analysis Software, Ingenuity Systems) to illustrate the many different genes that may be involved in cell cycle progression. The hits identified in the screen are shaded (red for "true hits" and blue for "borderline hits"–see text). Locations of the hits in the cell compartments are listed along with other components that are found in this signaling pathway. These charts can be used to help map genes involved in the specific pathways in specific cells and to design follow up experiments.

Step 4. Validate Results and Perform Follow Up Experiments

Once the screening data are acquired, many follow-up experiments can be designed. Interesting targets need to be confirmed, and their roles in the pathways need to be better understood. Confirmation of the data in a second experiment and verification of gene knockdown at the mRNA level by real-time PCR or other technique is a common first step. Experiments can also be designed to help independently confirm and expand upon the results found in data published in the literature.

Conclusions and Discussion

To learn about the functions and putative functions of the genes identified in our screen, genes designated as hits were analyzed using Ingenuity Systems’ software for pathway analysis. It was observed that the pathways to which these genes belong are very closely related (Figure 3, red boxes). ESR1 (estrogen receptor 1) and ESRRA (estrogen-related receptor alpha) are both hormone binding transcription activators. Since A549 cells rely on ESRRA and normal lung cells (WI-38) rely on ESR1 for normal G2/M progression, it might be inferred that these two cell lines have adapted two distinctly different methods to regulate cell growth and that these differences may be exploited to target cancer or disease specific cells.

Gene pathway networks generated from screening data are powerful tools to help determine which experiments to perform next and to help elucidate similarities and differences between cell lines. Looking more closely at the screening data, it was noticed that siRNAs to TP53 (commonly known as p53) and EGFR, both found in the pathway maps in Figure 3 (blue boxes), yielded borderline hits in WI-38 cells, but did not yield hits in A549 cells. Although the data require follow-up, they suggest that TP53 and EGFR may be involved in cell cycle progression through G2/M phase in WI-38 cells but not in A549 cells. Intriguingly, the p53 gene is often found mutated or lost in cancer cells and thus the A549 lung cancer cells may have found a network of genes that allow for these cells to circumvent TP53’s regulation of the cell cycle. This hypothesis is currently being tested.

The experiments described above highlight how easy it is to use siRNA libraries to help ascertain gene function. In only about 3 weeks from start to finish (including transfection optimization), the roles of several genes in cell cycle progression were better understood.

The Acumen Explorer™ is a laser-scanning fluorescence microplate cytometer from TTP Labtech that can be used for high content screening.

Ingenuity Pathways Analysis software (Ingenuity Systems) dynamically computes a set of relevant biological pathways for a presented set of genes or proteins, and was used here to create Figure 3.

Scientific Contributors
Angie Cheng, Lesslie Beauchamp, Ann Hartman, and Lance Ford • Ambion, Inc.

E-mail:
lford@ambion.com

back to top

Ordering Information for Ambion Products: