Base preferences in non-templated nucleotide incorporation by MMLV-derived reverse transcriptases

Abstract

Reverse transcriptases derived from Moloney Murine Leukemia Virus (MMLV) have an intrinsic terminal transferase activity, which causes the addition of a few non-templated nucleotides at the 3' end of cDNA, with a preference for cytosine. This mechanism can be exploited to make the reverse transcriptase switch template from the RNA molecule to a secondary oligonucleotide during first-strand cDNA synthesis, and thereby to introduce arbitrary barcode or adaptor sequences in the cDNA. Because the mechanism is relatively efficient and occurs in a single reaction, it has recently found use in several protocols for single-cell RNA sequencing. However, the base preference of the terminal transferase activity is not known in detail, which may lead to inefficiencies in template switching when starting from tiny amounts of mRNA. Here, we used fully degenerate oligos to determine the exact base preference at the template switching site up to a distance of ten nucleotides. We found a strong preference for guanosine at the first non-templated nucleotide, with a greatly reduced bias at progressively more distant positions. Based on this result, and a number of careful optimizations, we report conditions for efficient template switching for cDNA amplification from single cells.

Figure 1. | Optimal conditions for template switching.

The qPCR-derived Ct values for the investigated conditions are shown. Please note that the y-axis scales are broken and that the graphs have different Ct value scales. Also, no error bars are shown for the no template controls (NTC) as these reactions were performed in only one tube. In contrast, the actual experiments were performed in triplicate with the standard deviation error bars displayed in the graphs. The central part of the figure illustrates the template-switching process in the setting of our STRT method. (a) The optimal TSO concentration was 1 μM and (b) shorter TSOs showed a tendency for better performance (see also Table S2 and Figure S1 in File S1). The green datapoint in the TSO length graph represents the published version of the TSO [10] that is 40 bases in length. (c) shows analysis of different RT enzymes. SSII was a better choice than SSIII and, a cycled SSIII protocol (8 cycles of 50°C for 5 min and 60°C for 1 min; [19]) did not work, possibly due to the elevated temperature periods inactivating the enzyme (see also Table S2 in File S1). (b) The optimal SSII amount was 10 units per 10 μl reaction.

Figure 2. | Optimal TSO sequences for template switching.

The center panel shows the template-switching process. The template-switching event occurs in the middle of the grey circle. The investigated positions of the TSO are numbered from the template-switching site. The first three positions correspond to ribo bases and are shown in grey. The other analyzed positions are DNA bases and are depicted in black. In (a)) nucleotide preferences for all positions for RPLP1 in the RNA10N3 sample are shown. Corresponding graphs for the other transcripts and RNA spikes are shown in Figures S3 and S4 in File S1. (b)shows the distribution of the number of guanidines seen in the sequencing output for RPLP1 in the three performed reactions. Corresponding graphs for the other transcripts and RNA spike molecules are shown in Figure S5 in File S1. The numbers of reads for the analyzed RNA spikes and transcripts are shown in Table S6 and Table S7 in File S1, respectively.

Figure 3. | UMI length.

(a) shows the UMI principle and the output depending on reaction efficiency. If three out of the ten transcript molecules in total are labeled with the unique identifier (i.e. barcoded), only three barcodes will be observed in the sequencing data, translating into a transcript count of three for this particular transcript. Converting eight out of the ten molecules leads to the identification of eight barcodes in the sequencing reads. A UMI can become saturated if the number of transcripts copies exceeds the number of possible UMI combinations.(b) shows the distribution of barcodes for RPLP1 and spike MC28 in the performed reactions. Corresponding graphs for the other investigated transcripts are shown in Figure S6 and Figure S7 in File S1 for the analyzed RNA spikes. The reactions with the N10rG3 TSOs exhibited the highest complexity, followed by the reactions employing the N12rG3 and, lastly, the N10rN3 oligonucleotides. In (c) 6-base and 4-base barcodes were extracted from the MALAT1, RPLP1, MT2A, AHSG and CNIH4 reads. The red numbers indicate that the UMI has become saturated.