Figure 1.
Procyclic 927 cells harboring c-Myc-tagged TbPIF5 were fixed with 3% paraformaldehyde and then adhered to poly-L-lysine-treated slides. Immunostaining for TbPIF5-Myc used 1∶100 rabbit anti-Myc polyclonal antibody (Santa Cruz) and 1∶600 Alexa Fluor 568-conjugated goat anti-rabbit IgG (Molecular Probes). Conditions for fixing, permeabilizing, and staining cells were described [52]. In the merged image, anti-Myc is in red and DAPI in green. Arrows in panel A point out two cells with an enlarged magnification in panel B. Bar, 5 µm.
Figure 2.
ATPase and helicase assays of recombinant TbPIF5.
(A) Coomassie-stained SDS-PAGE gel and Western blot of purified recombinant TbPIF5. (B) Assay of TbPIF5 ATPase activity. The substrates and products were separated by polyethyleneimine thin layer chromatography; arrow shows origin. The [32P]Pi standard in the left-hand lane was prepared from [γ-32P] ATP by boiling 5 min in 1 M HCl. (C) Assay of TbPIF5 helicase activity. Substrates and products were separated by 12% polyacrylamide gel electrophoresis. (D) Helicase activity was measured at various concentrations of Mg++-ATP. (E) Assay of polarity of TbPIF5 helicase activity. Helicase substrates are diagrammed in Panels C (strand lengths are not to scale) and E (strand lengths for oligonucleotides a, b and c are 21, 21 and 90 nucleotides). * indicates 5′ 32P end label.
Figure 3.
Effects of TbPIF5 overexpression.
(A) Effect of TbPIF5 overexpression on cell growth. Overexpression was induced by addition of tetracycline (1 µg/ml) at day 0. The value of parasites/ml on the y-axis is the measured value times the dilution factor. Inset, Northern blot of mRNA level without or with overexpression. (B) Effect of overexpression on kinetoplast size as visualized by fluorescence microscopy of cells stained with DAPI (5 µg/ml). K, kinetoplast; N, nucleus. Bar, 5 µm. (C) Kinetics of kDNA loss as determined by visual analysis of images (>200 randomly-selected DAPI-stained cells for each time point). Inset images are examples of a cell with normal kinetoplast, small kinetoplast and no kinetoplast (kinetoplast is marked by arrow). (D) Effect of TbPIF5 overexpression on minicircle and maxicircle abundance. Total maxicircles (Maxi) and minicircles (Mini) were detected by probing a Southern blot after the total DNA (106 cell equivalents/lane) was digested with Hind III/XbaI and fractionated onto an agarose gel. The maxicircle probe detects only the 1.4 kb fragment, and only the 1 kb fragment derived from the heterogeneous minicircle population is shown. A hexose transporter fragment was probed as a loading control (Load). (E) Quantitation of the Southern blot in Fig. 3D showing maxicircle and minicircle species as indicated. Values represent the abundance of minicircle/maxicircle relative to its abundance in the uninduced cells. Values were normalized to load control.
Figure 4.
Electron micrographs of kDNA networks from TbPIF5 overexpression cells.
(A) and (B), kDNA isolated from wild-type cells. (C–E), kDNA isolated from TbPIF5 overexpression cells six days after induction. Arrow, maxicircle loops. Bar, 500 nm.
Figure 5.
Effect of TbPIF5 overexpression on free minicircle intermediates.
(A) Total DNA (106 cell equivalents/lane) was fractionated on a 1.5% agarose gel in TBE buffer (both the gel and running buffer contained 1 µg/ml ethidium bromide). A southern blot was probed for minicircles and hexose transporter (Load). (B) Neutral/alkaline two-dimensional gel electrophoresis. Total DNA from 3×107 wild type or induced cells (1 day overexpression) was fractionated on a two-dimensional gel. Strand-specific hybridizations were conducted with synthetic oligonucleotide probes. The upper panel shows a longer exposure version of the same 2-D gel used in Fig. 3F of our previous paper [20]. The scales below the panels indicate the sizes of linear markers in the second dimension. (C) Sedimentation of free minicircle intermediates in a 5–20% sucrose gradient. Fractions were collected from the top (1 ml fractions), subjected to electrophoresis, and assayed by probing a Southern blot [20]. (D) Gel electrophoresis (using conditions described for Panel A) of free minicircles treated with various enzymes. Total free minicircles were purified on the sucrose gradient in Panel C by pooling fractions 8 to 16 and ethanol precipitating the DNA. The free minicircles were then treated with T4 DNA polymerase (0.6 U, 1 U, 2 U and 3 U, New England Biolabs) and/or T4 DNA ligase (400 U, New England Biolabs). CM, catenated minicircles; N/G, nicked/gapped minicircles; CC, covalently-closed minicircles; θ, theta-structure; k, knotted minicircle; H, fraction H; ccD, covalently-closed dimer; ccT, covalently-closed trimer; nD, nicked dimer; L, linearized minicircle; MG, multiply-gapped minicircle; OF, Okazaki fragments.
Figure 6.
Effect of TbPIF5 overexpression on replication primers.
See [11] for experimental details of this experiment. (A) Analysis of 5′ ribonucleotides on the first Okazaki fragment. kDNA networks and free minicircle intermediates were isolated from cells without TbPIF5 overexpression or after overexpression for 1 day. DNA was digested with TaqI and treated with 0.3 M NaOH as indicated; alkali treatment would remove ribonucleotides and alter fragment mobility. After fractionation on a denaturing 9% polyacrylamide gel, a Southern blot was probed for the first Okazaki fragment with a 32P-labeled oligonucleotide. (B) Analysis of primers on the leading strand. kDNA networks and free minicircle intermediates were digested with HpyCH4V, and after electrophoresis a Southern blot was probed with a 32P-labeled oligonucleotide complementary to the 5′ end of the leading strand. M, size marker; OE, overexpression.
Figure 7.
Comparison of the normal free minicircle replication mechanism with that in cells overexpressing TbPIF5.
(A) Replication scheme showing (line 1) conversion, via theta structures (θ), of covalently-closed parental minicircles to gapped (G) and multiply-gapped (MG) progeny. The MG molecules are then converted (line 2) to nicked minicircles (N). The thick strand is H, which is synthesized discontinuously by Okazaki fragments, and the thin strand is L, which is synthesized continuously. * is an RNA primer and horizontal lines linking two circles in the θ-structures represent base pairs in the unreplicated portion. (B) When TbPIF5 is overexpressed, it binds to θ-structures and triggers primer removal. Removal of primers between the newly-synthesized Okazaki fragments generates gaps which can then be repaired. A similar removal of primers that have not yet been used for initiation blocks subsequent lagging strand replication. Leading strand synthesis is essentially unaffected and proceeds to completion allowing segregation. This process generates fraction H, a family of minicircles with a circular parental L-strand and increasing numbers of joined Okazaki fragments (ranging in size from 73 nt, the first Okazaki fragment, to 1 kb). This diagram shows the generation of three different species of fraction H, in which one, two or three Okazaki fragments are synthesized and then subjected to primer removal and joining. For simplicity, this diagram was not drawn to scale. We speculate that reactions in line 1 of panel A occur in the KFZ and those in line 2 take place in the antipodal sites. We further speculate that all reactions in Panel B occur in the KFZ. An alternative explanation for the existence of fraction H is that TbPIF5 overexpression somehow causes failure of the coordination of leading and lagging strand replication, so that the lagging strand is now synthesized continuously. Further studies are needed to test this possibility.