Ongoing transposition in cell culture reveals the phylogeny of diverse Drosophila S2 sublines

Abstract

Cultured cells are widely used in molecular biology despite poor understanding of how cell line genomes change in vitro over time. Previous work has shown that Drosophila cultured cells have a higher transposable element content than whole flies, but whether this increase in transposable element content resulted from an initial burst of transposition during cell line establishment or ongoing transposition in cell culture remains unclear. Here, we sequenced the genomes of 25 sublines of Drosophila S2 cells and show that transposable element insertions provide abundant markers for the phylogenetic reconstruction of diverse sublines in a model animal cell culture system. DNA copy number evolution across S2 sublines revealed dramatically different patterns of genome organization that support the overall evolutionary history reconstructed using transposable element insertions. Analysis of transposable element insertion site occupancy and ancestral states support a model of ongoing transposition dominated by episodic activity of a small number of retrotransposon families. Our work demonstrates that substantial genome evolution occurs during long-term Drosophila cell culture, which may impact the reproducibility of experiments that do not control for subline identity.

Keywords: Drosophila; cell culture; copy number variation; genome evolution; transposable element.

Fig. 1.

TE and CNV profiles reveal the evolutionary relationship among S2 sublines. a) Dollo parsimony tree for a panel of 26 S2 sublines with diverse lab origins, two S1 and S3 sublines to serve as outgroups in the phylogeny, and three mbn2 sublines that were inferred to be misidentified S2 lines by Han et al. (2021). Replicate samples for S2-DRSC were also included. The phylogeny was constructed using genome-wide nonreference TE insertions predicted by TEMP (Zhuang et al. 2014). Percent bootstrap support is annotated below each node. DGRC cell line names are used as taxa labels. Samples obtained from other sources are labeled in the format of “cell line name (source name).” Taxa labels were colorized based on original labs in which cell sublines were developed. b) Copy number profiles separated by chromosome arms for all samples included in panel a. Each data point represents normalized copy number (ratio×ploidy) for a given 10-kb window estimated by Control-FREEC (Boeva et al. 2012). Data points for each window are colorized by CNV status (red: CNV gain; green: no CNV; blue: CNV loss), which are based on the comparison between normalized copy number computed by Control-FREEC and baseline ploidy estimated by Lee et al. (2014). Red shading indicates CNVs that are exclusively shared by all S2 sublines in Clade A. Yellow shading indicates CNVs that are exclusively shared by S2R+ sublines. The red box represents CNVs on chromosome X that are exclusively shared by all S2 sublines in Clade A that are not S2R+. The blue box represents CNVs on chromosome arm 2L that are exclusively shared by S2R+ sublines from the Perrimon lab. Purple shading indicates CNVs that are exclusively shared by a subset of S2 sublines within Clade A or Clade B. Low recombination regions are shaded in gray.

Fig. 2.

TE profiles support ongoing transposition in S2 cell culture. a) Two hypotheses that could explain the mode of TE amplification in Drosophila S2 cell culture and how the resulting TE profiles could help infer the relationship among different cell sublines. Note that the schematic models represent genome-wide TE distributions combining all haplotypes. Therefore, given that S2 cells are tetraploid (Lee et al. 2014), a copy-number-loss event that occurred in one haplotype should only eliminate some TEs that are heterozygous in the affected region. b) Histogram shows the distribution of the number of Drosophila S2 subline samples that share each TE insertion in regions of chromosome X without major shared copy number losses (chrX: 500,000–20,928,973). c) Numbers of TE insertions on branches of the Dollo parsimony tree of 26 Drosophila S2 sublines constructed using nonreference TE predictions made by TEMP (Zhuang et al. 2014). Samples from S1, S3, and mbn2 cell lines were also included. The number of TE insertions estimated using ancestral state reconstruction was annotated in red above each branch. Percent bootstrap support was annotated in black below each node. DGRC cell line names are used as taxa labels. Samples obtained from other sources are labeled in the format of “cell line name (source name).” Taxa labels were colorized based on original labs in which cell sublines were developed.

Fig. 3.

Ongoing transposition in Drosophila S2 culture is contributed by a small subset of LTR retrotransposon families. a) Branch labelled Dollo parsimony tree including 26 Drosophila S2 sublines constructed using nonreference TE predictions made by TEMP (Zhuang et al. 2014). Samples from S1, S3, and mbn2 cell lines were also included. Taxa labels were colorized in the same way as Figs. 1 and 2c. Branch ID is annotated on each branch. b) Heatmap showing the number of estimated family-specific TE insertions on each branch of the tree in panel a. The heatmap is colorized by log-transformed [log10(count + 1)] number of gains per family per branch, sorted top to bottom by overall nonreference TE insertion gains per family across all branches, and sorted left to right into clades representing major clades of S2 phylogeny with major clade color codes indicated at the top of the heatmap. TE family names were colorized by TE type.