Our new
MessageAmp RNA Amplification Kits yield sufficient biotinylated antisense mRNA for Affymetrix® GeneChip® Arrays from as little as 20–30 ng input RNA (Figure 1). Both kits incorporate extensive improvements to workflow and sensitivity, but use different technologies for aRNA purification (see sidebar, Choice of aRNA Purification Methods).
Figure 1. Similar aRNA Yield and Size in Half the Time with One-Tenth the Input RNA. The amount and size of labeled aRNA synthesized from HeLa RNA using either the MessageAmp™ Premier RNA Amplification Kit or a competitor’s one cycle amplification/labeling kit were compared. The minimum recommended amount of input RNA (1 μg) and in vitro transcription (IVT) incubation time (16 hr) was used with the competitor’s kit. For the MessageAmp Premier Kit, both the minimum recommended amount of RNA (20 ng) and 100 ng of input RNA were amplified using the indicated IVT reaction times. Amplified RNA product size was comparable among the samples. The amplification efficiency with the MessageAmp Premier Kit was much higher than with the competitor’s kit: 100 ng input RNA processed with the MessageAmp Premier Kit had similar yield to 10-fold the input RNA (1 μg) processed with the competitor kit.
A series of master mixes reduces the number of tubes, pipetting steps, and protocol steps. All reactions take place in a single tube or in a single well of a 96-well plate, with sample transfer required only for the purification of biotin-labeled product at the end of the procedure. Another potential source of variability can be removed by using a thermal cycler for all incubation steps.
The simplified workflow significantly reduces hands-on time, but more importantly, it improves amplification efficiency. This breakthrough in increased amplification efficiency opens the door to gene expression studies from RNA samples that previously were too small for microarray analysis. For researchers with larger RNA samples (e.g., 100–500 ng), the dramatically reduced in vitro transcription (IVT) incubation times (Figure 2) facilitate higher throughput and enable completion of the entire procedure (RNA amplification through hybridization to a GeneChip Array) in a single day.
Figure 2. Recommended In Vitro Transcription (IVT) Incubation Times for MessageAmp™ Premier and MessageAmp III RNA Amplification Kits.
Data from the MicroArray Quality Control Consortium (MAQC) comparing microarray results from samples prepared with the MessageAmp Premier Kit (20 ng input RNA, 14 hr IVT and 100 ng input RNA, 8 hr IVT) or the well-established Affymetrix-recommended kit and method (5000 ng input RNA, 16 hr IVT: AFX Site #3) are presented.
The Venn diagram (Figure 3A) shows that concordance with MAQC AFX Site #3 is high for samples prepared using the MessageAmp Premier Kit from both 100 ng input RNA (95.5%) and 20 ng input RNA (96.3%). This high level of concordance is also seen with the MAQC AFX Sites #1 and #2 (data not shown).
Signal ratios from both total RNA inputs (100 ng and 20 ng) amplified with the MessageAmp Premier Kit show high correlation with the MAQC data (r>0.992) (Figure 3B). Additionally, signal ratios are very similar when comparing the two input levels (r>0.997). This high degree of correlation ensures that low input samples amplified and labeled for gene expression profiling using the MessageAmp Premier Kit will produce data quality comparable to current methods that require much higher total RNA inputs.
Figure 3. Obtain Highly Comparable Gene Expression Data from Affymetrix® GeneChip® Microarrays Using 250-fold Less RNA with the MessageAmp™ Premier RNA Amplification Kit. Samples A and B (Stratagene Universal Human Reference RNA and Ambion® FirstChoice® Human Brain Reference RNA, respectively) were amplified at MAQC AFX Site #3 (5000 ng input, 16 hr IVT) or prepared using the MessageAmp Premier Kit (20 ng input, 14 hr IVT and 100 ng input, 8 hr IVT). (A) Venn diagram representing Present call concordance between samples illustrates the relationships between the Affymetrix GeneChip probe sets that are considered “generally present” (as defined by the MAQC consortium) for sample A and B. (B) Scatter plots comparing Log2 signal ratios from samples A and B (log2(A/B)). Following the guidelines and comparison procedures recommended by the MAQC, only probe sets considered “generally present” in samples A and B are used for this analysis.