Large-Scale Low-Cost NGS Library Preparation Using a Robust Tn5 Purification and Tagmentation Protocol

Abstract

Efficient preparation of high-quality sequencing libraries that well represent the biological sample is a key step for using next-generation sequencing in research. Tn5 enables fast, robust, and highly efficient processing of limited input material while scaling to the parallel processing of hundreds of samples. Here, we present a robust Tn5 transposase purification strategy based on an N-terminal His₆-Sumo3 tag. We demonstrate that libraries prepared with our in-house Tn5 are of the same quality as those processed with a commercially available kit (Nextera XT), while they dramatically reduce the cost of large-scale experiments. We introduce improved purification strategies for two versions of the Tn5 enzyme. The first version carries the previously reported point mutations E54K and L372P, and stably produces libraries of constant fragment size distribution, even if the Tn5-to-input molecule ratio varies. The second Tn5 construct carries an additional point mutation (R27S) in the DNA-binding domain. This construct allows for adjustment of the fragment size distribution based on enzyme concentration during tagmentation, a feature that opens new opportunities for use of Tn5 in customized experimental designs. We demonstrate the versatility of our Tn5 enzymes in different experimental settings, including a novel single-cell polyadenylation site mapping protocol as well as ultralow input DNA sequencing.

Keywords: Tn5; single cell; tagmentation.

Figure 1

Purification of the His₆-Sumo3-Tn5 construct. (A) Plasmid map of pETM11-Sumo3-Tn5 encoding the His₆-Sumo3-Tn5 fusion protein. The locations of the point mutations in the Tn5 coding region (R27S, E54K, and L372P) are indicated. (B) SDS-PAGE analysis of the two His₆-Sumo3 purified Tn5 proteins after size exclusion chromatography. The single band runs at a molecular weight of ∼53 kDa, corresponding to the Tn5 monomer. (C) Size exclusion chromatography (Superdex200 Increase 10/300 GL) profile of the purified Tn5_{R27S,E54K,L372P} transposase. The Tn5 dimer peak elutes at 12.9 ml, while the Tn5 aggregate elutes at 8.7 ml. (D) Size exclusion chromatography (Superdex200 Increase 10/300 GL) profile of the purified Tn5_E54K,L372P transposase. The Tn5 dimer peak is the main peak and elutes at 13.2 ml. MW, molecular weight; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Figure 2

NGS-library preparation using the homemade Tn5 constructs. (A) Workflow of Tn5 loading, cDNA tagmentation, and subsequent NGS library preparation for duplex index (Illumina i5/i7) full-length cDNA sequencing. Tn5 molecules are shown as gray hexamers. The double-stranded part of the linker oligonucleotide, the mosaic element, is shown in gray with a yellow circle depicting the phosphorylated 3′ end. The 5′ overhangs as templates for the i5 or i7 index adapter primers are shown in red or blue, respectively. cDNA is shown as two parallel black lines with a 3′ poly(A) tail. The synthesis of the 5′ overhang complementary strand (gap-filling step during PCR amplification) is depicted as a dotted arrow. i5 index adapter primer is shown in dark blue, while i7 index adapter primer is shown in orange. Fragments that are lost during library preparation are transparent. (B) Bioanalyzer traces of NGS libraries processed with different concentrations of the in-house-produced Tn5_{R27S,E54K,L372P} using only homemade or inexpensive commercially available reagents for tagmentation and subsequent PCR reaction. Following the tagmentation protocol presented here, fragmentation of the cDNA works best when using Tn5_{R27S,E54K,L372P} at a concentration of 30 ng/μl. (C) Heat scatter plot showing the correlation of read counts between libraries processed with either in-house-produced Tn5_E54K,L372P or Tn5_{R27S,E54K,L372P}. Data of three technical replicates per condition were pooled for this analysis. (D) Heat scatter plot showing the correlation of gene counts between libraries processed using either in-house-produced Tn5_{R27S,E54K,L372P} and the protocol presented here or the Nextera XT DNA library preparation kit following the manufacturer’s instructions. Data of three technical replicates per condition were pooled for this analysis. (E) Pairwise correlation of read counts between three technical replicates (samples processed from the same cDNA on the same day) when using homemade Tn5_{R27S,E54K,L372P} and the tagmentation protocol presented here. The Pearson correlation of r = 0.99 between all samples demonstrates high reproducibility of both the enzyme and the protocol. (F) Heat map analysis of gene counts in technical replicates processed from the same cDNA but on different days. The color code indicates the Pearson correlation between samples (see legend on the right side). NGS, next-generation sequencing; PCR, polymerase chain reaction; SDS, sodium dodecyl sulfate.

Figure 3

NGS library preparation using different batches of in-house-produced Tn5 and reproducibility of the Tn5 purification protocol across institutes. (A) Left panel: comparison of two technical replicates for each of the three Tn5 batches. The color code indicates the Pearson correlation between samples (see legend on the right side). Right panel: pairwise correlation of read counts between NGS-libraries processed with three different Tn5 batches. Two technical replicates were pooled for each condition. (B) SDS-PAGE analysis of His₆-Sumo3 Tn5 enzyme after size exclusion chromatography, purified at BZH. The single band runs at a molecular weight of ∼53 kDa, corresponding to the Tn5 monomer. (C) Size exclusion chromatography (Superdex200 10/300 GL) profile of Tn5_{R27S,E54K,L372P} transposase purified at BZH. The Tn5 dimer peak elutes at 13.88 ml, while the Tn5 aggregate elutes at 8.93 ml. (D) Bioanalyzer traces of NGS-libraries processed with different concentrations of Tn5_{R27S,E54K,L372P} produced at BZH (see Figure 2B for libraries processed with Tn5 produced at EMBL). BZH, Heidelberg University Biochemistry Center; EMBL, European Molecular Biology Laboratory; MW, molecular weight; NGS, next-generation sequencing; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Figure 4

Polyadenylation site mapping in single cells using a customized tagmentation protocol. (A) Workflow of Tn5 loading, cDNA tagmentation, and subsequent NGS library preparation for the mapping of polyadenylation sites in single cells. Tn5 molecules are shown as gray hexamers. The double-stranded part of the linker oligo, the mosaic element, is shown in gray with a yellow circle depicting the phosphorylated 3′ end. The 5′ overhang template for the i7 index adapter primer is shown in blue. cDNA is shown as two parallel black lines with a 3′ poly(A) tail followed by a 6 nt cell barcode (light green) and the template sequence for the customized P5 adapter primer (green, introduced during the priming of the mRNA). In contrast to the duplex index full-length NGS libraries, Tn5 is loaded solely with the linker oligonucleotide for the i7 index adapter primer (blue, i7 index is dark blue). In the subsequent PCR, the initial gap-filling step is omitted so that only the customized i5 adapter primer (dark green) can bind to the cDNA fragments, resulting in amplification from the 3′ end of the cDNA during the first PCR cycle. In the following cycles, PCR enrichment only works for terminal cDNA fragments harboring both adaptors (corresponding to the RNA 3′ end) while the other fragments are lost during library preparation (shown as transparent). (B) Schematic representation of the percentage of reads mapping to the annotated TTS or within the surrounding 50 bp. (C) Distribution of poly(A) reads mapping to different genomic features: TTS, poly(A) signal maps to the TTS of a gene ± 10 or ± 200 bp (TTS10 or TTS200, respectively), poly(A) maps to the terminal exon (term.Exon), poly(A) signal found DS of the TTS (TTS-DS), poly(A) signal is found in the exon before the terminal exon (US-exon), antisense, or intergenic. (D) Pairwise correlation of gene counts between three technical replicates when using in-house-produced Tn5_{R27S,E54K,L372P} and the tagmentation protocol for the mapping of polyadenylation sites in single cells. DS, downstream; NGS, next-generation sequencing; PCR, polymerase chain reaction; SDS, sodium dodecyl sulfate; TTS, transcription termination site; US, upstream.