The Dictyostelium discoideum homologue of Twinkle, Twm1, is a mitochondrial DNA helicase, an active primase and promotes mitochondrial DNA replication

Abstract

Background: DNA replication requires contributions from various proteins, such as DNA helicases; in mitochondria Twinkle is important for maintaining and replicating mitochondrial DNA. Twinkle helicases are predicted to also possess primase activity, as has been shown in plants; however this activity appears to have been lost in metazoans. Given this, the study of Twinkle in other organisms is required to better understand the evolution of this family and the roles it performs within mitochondria.

Results: Here we describe the characterization of a Twinkle homologue, Twm1, in the amoeba Dictyostelium discoideum, a model organism for mitochondrial genetics and disease. We show that Twm1 is important for mitochondrial function as it maintains mitochondrial DNA copy number in vivo. Twm1 is a helicase which unwinds DNA resembling open forks, although it can act upon substrates with a single 3' overhang, albeit less efficiently. Furthermore, unlike human Twinkle, Twm1 has primase activity in vitro. Finally, using a novel in bacterio approach, we demonstrated that Twm1 promotes DNA replication.

Conclusions: We conclude that Twm1 is a replicative mitochondrial DNA helicase which is capable of priming DNA for replication. Our results also suggest that non-metazoan Twinkle could function in the initiation of mitochondrial DNA replication. While further work is required, this study has illuminated several alternative processes of mitochondrial DNA maintenance which might also be performed by the Twinkle family of helicases.

Keywords: DNA helicase; DNA primase; Dictyostelium discoideum; Mitochondrial DNA replication; Twinkle.

Fig. 1

Mitochondrial localization of D. discoideum Twm1. Fluorescence microscopy of D. discoideum cells a stained with Mitotracker Red and b expressing a Twm1-GFP fusion protein c overlayed. Image is representative of the transformant population observed under ×1000 magnification with immersion oil. Scale bar = 5 µm

Fig. 2

Effects of twm1 antisense inhibition on D. discoideum mitochondrial function. a Plaque expansion rates of twm1 antisense transformants on E. coli B2 lawns. Rates are plotted against antisense vector copy number, which was determined through qPCR. b Relative twm1 mRNA levels in antisense transformants, quantified using qRT-PCR, normalized against a structural gene (tubB) and calculated as a percentage of the parental AX2. Values were plotted against antisense vector copy number. The twm1 specific primers (Additional file 3: Table S1B) used recognize downstream of the region targeted by antisense inhibition to quantify only full length twm1 mRNA. c mtDNA copy number of twm1 antisense transformants. mtDNA copy number was determined through qPCR by comparing against a single copy nuclear gene, tubB. Antisense transformants were also compared to the parental AX2, and values plotted against antisense vector copy number. Parental AX2 is depicted in black, with transformants in grey; antisense transformants are shown as squares, while vector controls are diamonds. Vector controls were plotted ignoring vector copy number (102, 142 and 193) given there was no antisense inhibition

Fig. 3

Relative twm1 mRNA levels of D. discoideum AX2 in response to EtBr treatment. Initial AX2 cultures (T0) were first treated with 10 μg/ml EtBr for 24 h (T24), at which point the EtBr-containing medium was removed and replaced. Following this, cells were allowed a further 24 h to recover without EtBr (T48). A duplicate culture without EtBr treatment was used as a control. twm1 mRNA levels were quantified with qRT-PCR and normalized against a structural gene (tubB). Values are relative to initial (T0) mRNA level. Error bars represent the standard error. p values were calculated using Student’s t-test (*< 0.05; **< 0.0001)

Fig. 4

Loss and recovery of twm1 antisense transformant mtDNA following EtBr exposure. Parental AX2 and twm1 antisense transformants (TAS) were exposed to EtBr for 24 h, and subsequently allowed a further 24 h to recover. Relative mtDNA copy number was determined using qPCR and the single copy tubB nuclear gene, with each strain compared to its initial mtDNA copy number (100%). Error bars represent the standard error, while p values were calculated using the Student’s t-test (*< 0.05; **< 0.005; ***< 0.001). p values for TAS samples at T24 were determined compared to AX2 at T24. For T48 samples, the p value was calculated from the proportional increase of mtDNA from T24 to T48, given each strain’s different mtDNA copy number at T24. This data is also presented in Table 1

Fig. 5

Nucleotide hydrolysis by purified Twm1 in vitro. a Hydrolysis of both rNTPs (grey bars) and dNTPs (white bars) by Twm1 was performed at 21 °C. b NTPase activity of Twm1 was also measured in the presence of linear dsDNA (FHA0; black bars), circular ssDNA (M13mp18; grey bars) or linear ssDNA (FHA3.1; hatched bars) at 21 °C. Error bars represent the standard error. p values were calculated using Student’s t-test (*< 0.05; **< 0.005; ***< 0.001). All values were normalized against no protein negative controls (empty vector purification)

Fig. 6

Helicase activity and substrate preference of D. discoideum Twm1. In vitro helicase activity of Twm1 was determined at 21 °C using various fluorescently labelled dsDNA templates (Additional file 3: Table S1B). Each DNA template was heated to 100 °C (H; first lane) and assayed using a no protein negative control (N; empty vector purification; second lane) in addition to Twm1 (T; third lane). Substrate (S) and final product (P) are indicated. Overhang polarities and FAM labels (red dots) of substrates are also indicated. a Helicase assay using strict dsDNA (FHA0) or open fork-like dsDNA (5′ and 3′ overhangs; FHAOF). b Determination of Twm1 directionality using open fork-like dsDNA with one duplex overhang (FHAOF5 or FHAOF3). c Overhang requirements of Twm1 were determined using dsDNA with a single ssDNA overhang (5′ or 3′; FHA5 or FHA3, respectively). Directionality of Twm1 was reconfirmed by using a duplex 3′ overhang (FHA3D)

Fig. 7

Alignment of primase motifs from bacteriophage T7 gp4 and various Twinkle helicase homologues. Conserved primase motifs (I-VI and RNA polymerase basic; RNA Pol) are those observed by Ilyina, Gorbalenya [64] and Shutt and Gray [10], while the critical residues for T7 gp4 primase activity [36, 37] are shown (arrows). Identical and similar residues (compared to the T7 gp4 sequence) are shaded in black and grey, respectively. Parentheses indicate the number of residues flanking each motif. Proteins from the following sources were used: Arabidopsis thaliana (At; NP_849735), Dictyostelium discoideum (Dd; XP_636842), Homo sapiens (Hs; NP_068602), Mus musculus (Mm; NP_722491), Plasmodium falciparum (Pf; XP_001348285) and bacteriophage T7 (NP_041975)

Fig. 8

In vitro primase activity of purified D. discoideum Twm1. a Recombinant Twm1 was tested for primase activity in vitro using radiolabelled nucleotides and a circular ssDNA template (M13mp18). Following incubation, samples were run under non-denaturing conditions in order to visualise any synthesized RNA primers annealed to the template (P). Disassociated primers were too small to be differentiated from unincorporated nucleotides (U) under denaturing conditions. b The primase assay was repeated with the addition of Klenow DNA polymerase to determine whether any generated primer could be used for DNA synthesis. Signals were observed above the unincorporated nucleotides (U) when both enzymes (Twm1 and Klenow) were included, indicating incorporation of the radiolabelled primer into larger synthesized DNA (SD). All reactions lacking Twm1 were prepared using the empty vector purification as a negative control

Fig. 9

In bacterio replication of pZErO-2 by D. discoideum Twm1. E. coli BL21 strains each possessed two vectors; pET-23a:tetA either as an empty vector (−) or encoding Twm1 (+), and pZErO-2 with (+) or without (−) the NCRrnl or rnl region of the D. discoideum mitochondrial genome. Twm1 expression was induced by the addition of 1 mM IPTG. Relative pZErO-2 vector copy number (using primers for kanR) was quantified using qPCR against a single copy number bacterial gene, talB. Error bars represent the standard error. p values were calculated using Student’s t-test (*< 0.05; **< 0.001)