Options for High Throughput Sample Homogenization for RNA Isolation
David W. Burden, Ph.D., President, OPS Diagnostics, LLC
Responses to this commentary are encouraged and will be posted as an addendum. Email comments to firstname.lastname@example.org.
As a scientist who has been involved in molecular biology since the early 1980s, working with RNA has always been associated with a sense of dread. From my early training it was stressed that RNA, and especially mRNA, rapidly turns over in cells and is extremely susceptible to degradation once liberated. Working with RNA was akin to a child and a kitchen stove; if you're not careful, you'll get burned. The prospect of having RNA "disappear" after a series of manipulations as a consequence of using materials contaminated with RNase was very real. None of today's niceties, such as nuclease-free tips and tubes, molecular biology grade buffers, and RNA isolation kits existed. Countless hours were spent treating water with DEPC, baking glassware, and fending off associates seeking my stash of pristine reagents.
That was then, and this is now. By the 1990s many of the home grown methods had been replaced with commercially available reagents and kits specifically designed for isolating and manipulating RNA. Working with RNA is no longer so tenuous. Virtually all aspects of isolating and subsequently manipulating RNA to a set of desired goals are now routinely performed. However there are still some unknown variables in getting from tissue to a qRT-PCR reaction or microarray. The effect of high throughput homogenization methods on RNA quality and yields is a major one. This is not to say that high throughput processing of tissues is a "black box", rather, methods used by researchers are very divergent. To my knowledge, no published studies have been made comparing processes. In lieu of a lack of published data, bits and pieces have been gathered by our team to get a better picture of the options concerning high throughput homogenization for RNA isolation.
Many of the original notions of RNA handling impressed upon us "experienced" scientists (a.k.a., old) appear to be in question. RNA is apparently not as fragile a molecule as once believed. Sharova et al. (2009) examined a wide spectrum of mouse embryonic stem cell mRNAs for their half-life and found all but a few to have a half-life of over 7 hours. This type of stability is in stark contrast to the notion that RNA turnover in cells should be measured in minutes. RNA levels in porcine retinal pigment cells (Malik et al., 2003) were relatively stable up to five hours after extracting intact eyes from sacrificed animals. Another eye opener (no pun intended) is that RNA in harvested tissues and deceased animals, including cadavers, may be relatively stable post-mortem for several hours (Lee et al., 2005), though there is significant variability between tissues and individuals. This is as long as the tissue remains intact. Once tissues are dissected, the rate of RNA degradation increases (Fajardy et al., 2009).
This last point is significant to high throughput work as sectioning tissues to load a deep well plate promotes differential treatment to the samples. The samples first placed in wells may incubate for an hour or more compared to the last samples. Though RNA may be stable in harvested tissues, the manipulation of those tissues during sample preparation does speed degradation.
In constant discussion with scientists, I have seen three distinct approaches to homogenizing tissues for high throughput processing of RNA. These are cryogenic grinding, homogenizing with denaturing solutions, and processing of tissue previously treated with RNA stabilizer solution. These approaches all involve bead beating which has evolved as the most common and universal approach for high throughput homogenization.
The first approach uses cryogenics: freezing tissues in liquid nitrogen and subsequently bead beating while the tissues are frozen. This parallels the traditional use of liquid nitrogen chilled mortar and pestles. Cryogenic grinding not only protects the RNA, but also pulverizes the samples well, given that they are brittle. In a high throughput format, deep well plates or vials, tissues and grinding ball are all prepared while in a liquid nitrogen bath (or vapor phase CryoCooler). Once the grinding vessels are loaded, they are quickly placed in the homogenizer and processed. Warming can occur thus some researchers opt for processing samples in vials held in cryoblocks (essentially aluminum heat blocks) which are pre-chilled as well.
The second strategy starts with either fresh or frozen tissue which is placed into TRIzol®, TRI Reagent® or some other chaotropic solution followed by homogenization at room temperature. Denaturing solutions can diffuse and even dissolve finely diced tissues. Dropping tissue into a denaturant followed by homogenizing may provide sufficient protection for RNA isolation. However as noted above, the preparation of the tissue (e.g., weighing and cutting), along with the time of processing multiple samples, may significantly impact the yields and quality of RNA. Preparing 96 or 192 samples may generate results very different than those yielded while processing a limited number of samples. Hence, the third option of using RNA stabilizers may address this issue.
The third method is not much different than the second except tissues are first treated with a RNA stabilizer solution, such as RNAlater™. This treatment completely negates the need for any cryogenic handling of the tissues before and during homogenization. It also protects RNA during tissue cutting, weighing and plate loading. Dicing fresh tissues and submersing them in the RNAlater™ serves to preserve the RNA for short durations at room temperature, for prolonged periods at 4°C and indefinitely when frozen at -80°C. The value of RNA stabilizers is well accepted, but its value as compared to cryogenic grinding or simply submersing samples in denaturants, in a high throughput format, is unknown.
In all three methods described above, the final fate of the tissue is the same. A solid mass of tissue is converted to a puree. However, the underlying unknown is whether the route travelled to get the tissue homogenized has a potential impact on the subsequent experimental results. The question which needs to be asked is whether the methods for handling and homogenizing multiple samples of tissue affect the quantity and quality of RNA species isolated. Though this is a topic we will be addressing in our lab, at this time most of our information is anecdotal.
As OPS Diagnostics relies heavily on providing the tools to researchers for homogenizing tissues for RNA isolation, the question of whether one high throughput method is better than another is important. At this time, I have no answer as to what is best, or whether it matters. Indeed, a major objective of starting this discussion (please comment!) is to better compile a cross section of experiences to help better understand the variables. We are looking forward to your comments which we will post with permission.
Fajardy, I., E. Moitrot, A. Vambergue, M. Vandersippe-Millot, P. Deruelle and J. Rousseaux. 2009. Time course analysis of RNA stability in human placenta. BMC Molecular Biology 2009, 10:21
Lee J., A. Hever, D. Willhite, A. Zlotnik, and P. Hevezi. 2005. Effects of RNA degradation on gene expression analysis of human postmortem tissues.FASEB Journal 10.1096/fj.04-3552fje. Published online June 13, 2005.
Malik, K., C. Chen, and T. Olsen. 2003. Stability of RNA from the Retina and Retinal Pigment Epithelium in a Porcine Model Simulating Human Eye Bank Conditions. IOVS 44(6): 2730-2735.
Sharov, L., A. Sharov, T. Nedorezov, Y. Piao, N. Shaik, and M. Ko. 2009. Database for mRNA Half-Life of 19 977 Genes Obtained by DNA Microarray Analysis of Pluripotent and Differentiating Mouse Embryonic Stem Cells. DNA Research 16, 45-58.