Troubleshooting

Does Mix of RNA/DNA Interfere with Qubit Measurement?

Hypothesis:

The measured RNA in the sample could be due to lack of specificity of Qubit dye eg. RNA dye actually has affinity to DNA and therefore shows RNA in our samples.

Experiment:

Samples containing only RNA or DNA in concentration of 10 ng/µl were prepared in triplicate and each sample was measured with two different Qubit kits (RNA HS Kit, DNA HS Kit, Thermo Fisher).

Results:

Measuring RNA (or DNA) with corresponding Kit resulted in expected results eg. around 10 ng of RNA and very low amount of DNA as shown in Table 1. Similar results were seen for DNA sample. When DNA was added into RNA, the measured RNA amount decreased, which was the only measuring bias we have observed.

Table 1. The average of measured values for each of the samples. Measuring RNA (or DNA) with corresponding Kit resulted in expected results eg. around 10 ng of RNA and very low amount of DNA.

Conclusion

The experiment only investigated interactions between dsDNA and ssRNA, which are not supposed to introduce any bias into the measurement as per manufacturer's information [2]. The remaining question is how would a RNA:DNA hybrid be treated by the dye. It is possible that RNA was not properly digested and therefore remains in the hybrid form. A hybrid could hypothetically be detected by both RNA and DNA specific kit. Unfortunately no proven hybrid sample was available and therefore this hypothesis could not be tested.

It was concluded that having DNA - RNA mixture does not influence the measurement in a significant way. RNA amount is moderately decreased after addition of DNA into the sample, which would not explain the presence of RNA in cDNA samples. What remains to be investigated is how a RNA:DNA hybrid would influence the measurement. We therefore conclude that the RNA measurement in our samples is accurate and there is RNA present, either as a ssRNA or RNA:DNA hybrid.

Is RNA Template Properly Digested? Is It Carried Over with the AMPure XP Beads?

Hypothesis:

RNA is found in the final cDNA product as it is not digested properly. The RNase Cocktail contains RNAse A and T1, which are only supposed to introduce nicks into the RNA. Provided that LongAmp Taq polymerase does not degrade these fragment and that the purification beads (AMPure XP, Agencourt) do have affinity to both RNA and DNA, the RNA would be found in the sample.

Experiment:

This troubleshooting experiment had three major parts: i) optimizing the protocol for the use of RNase Cocktail using longer incubation time and/or larger quantity, ii) use of RNAse H for digestion and iii) clean-up step with purification beads to determine whether the beads have affinity for RNA.

1. Total of 8 samples with 125 ng of RNA were prepared (from previous cDNA samples that shown RNA contamination, eg. there was DNA present in the samples as well). The experiment was optimized for time and amount of RNase Cocktail.

2. RNA ladder was digested using both RNase H, RNAse Cocktail or both. Also each of the tests was prepared in either RNase-free water or buffer solution used during the cDNA synthesis. Digested DNA was visualised on a gel.

3. 1000 ng of RNA ladder were purified using AMX beads and shown on a gel.

Results:

Overall, it was seen that subsequent digestion of RNA contaminated samples was successful regardless of used concentration as shown below. Digestion of RNA ladder using RNase H and/or RNase Cocktail shows complete digestion for RNase Cocktail while RNase H seems to not be working. The beads do carry over most of the RNA including rather small fragments (200 bp). The detailed results of each experiments are shown below:

1.

Table 2. The table shows treatment for 9 samples of RNA and DNA mixture, each containing 125 ng of RNA. *Corresponds to experimental set-up used during the actual cDNA synthesis.

According to measurement with Qubit RNA HS, all samples regardless of treatment show all RNA being degraded including sample, where treatment is identical to the one used in the actual experiment.

2. The gel below in Figure 2 shows results of the digestion of RNA ladder with the available RNases, RNase H (line 5 and 6) and RNase Cocktail (line 7 and 8) or both (line 8 and 9).

From this experiment we can conclude that the buffer has no effect on digestion since sample in water and in reaction buffer appear the same. It can further be said that RNase Cocktail efficiently degrades the ladder, which is composed of ssRNA. RNase H seems to not digest the ladder at all.

This corresponds with described ability of RNase H to preferentially digest RNA:DNA hybrids.

Figure 2. Results after digestion with RNases.

3. The gel showed very significant carry over of RNA during the purification with AMPure beads (image not shown), which corresponds to manufacturer's information. If RNA was not properly digested, it would therefore be carried together with the cDNA during the entire library prep process.

Conclusion

Overall, the results of this troubleshooting procedure present results conflicting with result of the actual experiment. Here the RNases are shown to work very efficiently. Most likely, RNA used in troubleshooting has different properties that the RNA template in the cDNA synthesis (could be in a form of hybrid or some other unusual form) which leads to different digestion results.

Discussion

During the cDNA synthesis it was possible to achieve sufficiently high yields, usually exceeding double of the input mRNA amount. This cDNA was used to prepare sequencing libraries. As was shown later, it unfortunately contained undigested RNA, which significantly decrease the quality of sequencing results. We have therefore spend significant amount of time searching for the source of RNA contamination as reported in this section.

The experiments above show very contradicting results. In section i) it can be seen that treatment of cDNA with RNase Cocktail after synthesis resulted in complete clearance of RNA from the sample. RNase Cocktail has also been shown to digest RNA ladder as visualized on the gel in Figure 1. The same enzyme has always been used during the cDNA synthesis procedure and it remains unclear why does it efficiently digest RNA after synthesis or RNA ladder and would not work during the actual synthesis.

One hypothesis that we considered was the presence of RNA:DNA hybrids, which would decrease efficiency of RNase Cocktail, which is more efficient in digesting ssDNA. RNase H was therefore added to address this issue but clearance of RNA from samples did not significantly increase. Even more confusingly, digestion did work in some of the samples despite the content/treatment being identical.

Conclusion

We have managed to successfully synthesize complementary DNA to our mRNA samples, which unfortunately did contain undigested RNA. A protocol needs to be developed that assures all of the RNA has been removed from the sample prior to the preparation of the library. Moreover, additional troubleshooting needs to be performed to determine why the digestion is not efficient.

References

[1] IDT, Use of template switching oligos (TS oligos, TSOs) for efficient cDNA library construction, [online], 2018 https://eu.idtdna.com/pages/education/decoded/article/use-of-template-switching-oligos-(ts-oligos-tsos)-for-efficient-cdna-library-construction

[2] Thermo Fisher Scientific Inc, User Guide: Qubit RNA HS Assay Kits, [online], 2015 https://www.thermofisher.com/order/catalog/product/Q32852