Team:Munich/Testing/celllysis

Cell Lysis

For our decision on how to extract RNA from our disease targets, we looked at several possible lysis methods. Since our goal is to make the use of our designed device as easy as possible and accessible to everyone, we were trying out techniques differing in required equipment and chemicals and compared their feasability .

The standard approach for RNA extraction, Guanidinium-Phenol-Chlorophorm extraction, uses Guanidinium salts as a lytic agent. Since it is a very strong chaotrope, however, the lysis product would have to be thoroughly purified for our downstream proteins to work, and our waste would be environmentally problematic for nearly any kind of guanidine salt (The most potent guanidinium salt for lysis, Gu-Thiocyanide, might even release cyanide fumes when disposed together with acidic waste, which is unacceptable in an easy-to-use device).

Another commonly used lysis process is incubation at high temperature with detergents, which we often used in the form of SDS for our benchtop analyses. Again, SDS would have to be seperated from the RNA afterwards in our device, which would complicate our procedure. While we investigated RNA-silica binding properties for such a case (see labbook Sept. 1st to 5th, section "other"), we decided against adding unnecessary complexity for now.

Alkaline lysis is a well known method to extract DNA from cells, but not used for RNA due to rapid degradation thereof under alkalic conditions. Since our protein responds to a very short target sequence, we wanted to test if we might use it in spite of that due to some significant advantages. First, after the alkaline lysis reactin only salts and acetate remain, which interfere significantly less with downstream processing than SDS or GuSCN and therefor can be easily cleaned up. Secondly, the waste would consists only of harmless salts and some acetate, scoring points for environmental friendliness.

As a first test, we compared the general lysis / RNA-release efficiency to the aforementioned SDS-method. Even though the incubation time for the alkaline lysis is seconds instead of a quarter hour (SDS-lysis), the measured RNA concentration was at least equal (10^8 cells/ml assays) to that of SDS-lysis. We therefore proceeded to investigate the more problematic aspect of RNA degradation.

From our Urea page gel, we conclude that alkaline lysis is a very fast way of extracting RNA. It is commonly known that RNA (unlike DNA) swiftly self-degredates at high pH, and this also showed in our experiments. Comparison between different incubation times shows clearly the continuing degredation over time. Within only a few seconds of inverting tubes, RNA is already cut and thus shortened, as the two "NaOH, 0 min" bars show. However, the target sequence of Cas13a is very short (~ 20 base pairs) and taking this observations into account we deem it unlikely that a significant percentage of target-sequences is destroyed by a very fast alkaline lysis. Most of the cut strands are still well above 200 bp for a "seconds-long" lysis, with the peak of brightness between the 500 bp and 1000 bp mark, as the gradient of band intensity shows. With higher incubation times, the peak of brightness goes steadily lower, and even under 200 bp for a incubation time of 10 minutes. It is thus of utmost importance to keep the lysis time to a few seconds, which makes a proper and fast mixing of the sample with alkalic lysis buffer and seconds later with acidic equilibration buffer necessary.

Since microfluidic mixing of liquids is a rather complex process, we chose a tecnically simpler approach to extract pathogen RNA in our first prototype: Isothermal PCR.

[1] 65 °C, [2] log2 DNA ladder, [3] 70 °C, [4] 75 °C, [5] 80 °C, [6] 85 °C, [7] 90 °C, [8] empty °C, [9] positive control (DNA sequence diluted), [10] log2 DNA ladder
2 uL of lysed E. Coli suspension with density of 10⁶ cells/ml used for PCR-Mix.

PCR is a common method in analytics to achieve huge amplification of genetic material. It allows us to use a relatively inefficient lysis method, lysis by heat only, and increases the resulting low concentration of target sequence to above a detectable threshold. With Reverse Transcription before, it is also possible to amplify an RNA sequence as DNA.
To increase the detection threshold of our system with PCR, it is however still important to optimize the heat lysis parameters, most prominently the temperature at which to incubate. It has been shown repeatedly (e.g. Ren et. al.,J Biotechnol. May 2007), that higher temperatures not necessarily lead to better lysis efficiency. Thats why we tested isothermal PCR (RPA) on samples lysed with different temperatures (65 °C, 70 °C, 75 °C, 80 °C, 85 °C, 90 °C) and concluded from our RPA gel that any temperature above 75 °C degree works at least as good as 90 °C, if not better. This is consistent to mentioned literature and allows us to use lower temperatures in our device, which reduces issues with evaporation and build-up pressure. Still, a higher temperature of over 90 °C might be advantageous to reduce RNAse activity.

[1] log2 DNA ladder, [2] 10⁶ cells/ml, [3] 10⁵ cells/ml, [4] 5*10⁴ cells/ml, [5] 10⁴ cells/ml,[6] 5*10³ cells/ml, [7] 10³ cells/ml, [8] 10² cells/ml,[9] 10⁶ cells/ml (no heat-lysis step. only PCR at 37 °C)

To assess the theoretical sensibility of our PCR-enhanced lysis, we did tests on a dilution series of a E. coli with TEV-plasmid (psB1C3-His-008-TEV) culture with a cell density range from 10^6 to 10^2 cells/ml. After lysis at 80 °C, the isothermal PCR was performed on 2 uL of lysed sample. The gel shows a clearly visible band of PCR product down to a cell density of 5*10^4 cells/ml. Image processing even reveals low PCR-product concentration at 10^4 cell/ml, which corresponds to on average 20 cells in the PCR reaction volume. This fits nicely with our findings that heat-only lysis is less efficient by a factor of 5 to 10 when compared to lysis with detergents and heat (see labbook Sept. 14th). Additionally, we expect gram positive bacteria to not deviate significantly from the E. coli heat-lysis efficiency, as B. subtilis showed very similar yields (while SDS-lysis performed far worse, see labbook Sept. 19th).