Figure 1.
Effects of EB on growth and kDNA.
(A) Effect of EB (2 µg/ml) on growth. Values of parasites/ml on y-axis are measured value times dilution factor. (B) Effect of EB on kinetoplast size. DAPI-stained cells (∼100 per time point) were visually categorized by kinetoplast size. Dk, dyskinetoplastic or no detectable kDNA. (C) Fluorescent images of different-sized kinetoplasts seen in DAPI-stained cells untreated or following EB treatment. Scale bar, 5 µm. (D) Bar graphs showing surface areas, calculated with IPLabs software, of DAPI-stained networks isolated from cells treated for the indicated time with 2 µg/ml EB. Arrows show average size. Insets show a representative field at each time point. Scale bar, 5 µm. (E) Time course of kDNA loss. Total DNA digested with Hind III/Xba I was fractionated on a 1.5% agarose gel (106 cell equivalents/lane). A Southern blot was probed for minicircles (only the 1 kb fragment is shown) and maxicircles (probe recognizes only a 1.4 kb fragment).
Figure 2.
Discovery and characterization of Fraction E.
(A) Effect of EB on free minicircle population. Total DNA (5×106 cell equivalents/lane), from cells treated with 2 µg/ml EB for indicated time, was fractionated on an agarose-EB gel. The autoradiograph is a Southern blot probed for minicircles and hexose transporter gene (load). G, gapped, L, linearized, and CC, covalently-closed minicircles. E is Fraction E. * is probably interlocked minicircle dimer forming a smear like that of Fraction E. We usually see only trace amounts of linearized minicircles in untreated cells (0 time point). (B) Two-dimensional neutral-alkaline gel electrophoresis of free minicircles. The sample was total DNA (5×106 cell equivalents/lane), from untreated or EB-treated (6 h, 2 µg/ml) cells. 32P-labeled oligonucleotides were used for probing minicircle L (left two panels) and H (right two panels) strands. Scales below gels are sizes of linear markers. Abbreviations of minicircle species are same as in Panel A. See [25], [26] for description of other bands in these gels. (C) Treatment with Topo I. Fraction E (10 µl, sucrose gradient purified) was treated with E. coli topo I [51] and fractionated on an agarose-EB gel. Arrow indicates position of covalently-closed relaxed minicircles. (D) EM of fraction E also sucrose gradient purified. Molecule at upper left is a relaxed minicircle. Scale bar, 200 nm. (E) Fraction E contains regions of Z-DNA. Total DNA (T) from 50 ml of culture (0.8×106 cells/ml) of untreated or EB (2 µg/ml)-treated cells was immunoprecipitated with anti-Z DNA antibody and then with protein G-Sepharose [24]. Upon centrifugation, DNA in supernatant (S) and pellet (P) was electrophoresed and probed for minicircles. Increases in linearized minicircles could be due to nuclease contamination of the antibody. Gel conditions and abbreviations of minicircle species are same as in Fig. 2A.
Figure 3.
Effect of EB treatment on replication of minicircles.
(A) Cells (50 ml, 0.7×106 cells/ml, treated with 2 µg/ml EB for 6 h) were incubated with 50 µM BrdU during the last 40 min. Free minicircles were then fractionated on an agarose-EB gel as in Fig. 2A, and a Southern blot was probed for minicircles. (B) The same amount of DNA used for the gel in Panel A was run on a 2D neutral/alkaline gel (as in Fig. 2B) and a blot was probed with anti-BrdU antibody. BrdU label at top of gel is in nuclear DNA. Abbreviations of minicircle species are same as in Fig. 2A.
Figure 4.
(A) Network from untreated cell spread for EM in the absence of EB. Boxed region is enlarged at lower right. (B) Network from cells treated with EB (2 µg/ml, 24 h), but EB was removed prior to EM. Arrows indicate supertwisted minicircles. (C) Network from EB-treated cells (2 µg/ml EB, 6 h) spread for EM in presence of EB. Prior to EM, EB (100 µg/ml) was added to the DNA, the spreading solution, and the hypophase. Arrows indicate rare untwisted minicircles that likely are nicked. Bars, 500 nm.
Figure 5.
Fluorometric measurement of EB-network binding and the effect of p19.
Binding of EB to kDNA was measured by fluorescence spectroscopy using a Fluoromax-3 fluorometer (Horiba). Excitation was at 525 nm and emission was measured from 580 to 680 with a maximum at 604 nm. Titles above each panel show the components of the reaction. Those in parentheses are at constant concentration and those that vary are indicated. Vertical arrows inside each panel show direction of variation. Left panel shows that EB (6.25 µg/ml) emission fluorescence increases markedly on addition of kDNA networks (0, 0.25, 0.5, 1.25, 2.5, 3, 4, 6.5, 25 µg/ml). In middle panel, the same amount of DNA was pre-incubated with 20 µM p19 (room temperature, 5 min) before adding to EB (6.25 µg/ml). Right panel shows displacement of EB (6.25 µg/ml) from DNA (25 µg/ml) by addition of increasing concentrations of p19 (0, 1, 2, 4, 8, 10, 12.5, 15, 20 µM). In this experiment we used Crithidia fasciculata networks (purified as described [54]) as they are much easier to isolate than those from T. brucei. The preparation of recombinant p19 will be published elsewhere.
Figure 6.
Effect of EB on distribution of gapped circles in isolated networks.
Isolated networks were labeled by fluorescein-12-dUTP using terminal deoxynucleotidyl transferase (TdT) [33]. This procedure adds a fluorescent tag to 3′ OH groups flanking minicircle gaps. Because of the relative abundance of minicircles and the distribution of fluorescein fluorescence within networks, most gaps must be in minicircles rather than maxicircles [33]. This procedure reveals the extent of replication of a network. TdT-positive networks are mostly undergoing replication and some are post-replication with gaps yet to be repaired. In wild type cells, the labeling pattern is usually polar or uniform, representing early and late stages of replication respectively. (A) Fluorescent images of networks isolated from untreated cells stained with 2 µg/ml DAPI (upper-left panel) and labeled with TdT (upper-right). Lower panels are the same except cells were EB-treated (2 µg/ml, 24 h). In upper panel, network (NW) 1 is a TdT-negative pre-replication or post-replication network because it stains with DAPI but not TdT. Networks 2 and 3 are replicating networks with polar labeling. Network 4 is TdT-positive with uniform labeling, appearing double-size and ready to divide. Scale bars, 5 µm. (B) Kinetics of change in TdT labeling pattern during EB treatment. Upper inset shows a uniformly labeled network and lower inset shows one with polar labeling. At least 120 networks were counted at each time point. Scale bars, 1 µm.
Figure 7.
The mechanism of killing BSF trypanosomes by EB.
(A) Effect of EB concentration on killing of wild type 427 and Dk 164 trypanosomes. In a 24 h acid phosphatase-based cytotoxicity assay [38], A405 was measured on a microtiter plate reader (Molecular Devices). Each EB concentration in an experiment was assayed in quadruplicate, and error bars indicate standard deviations for three experiments. Data were fit to the equation for the sigmoidal Emax model [55] using GraphPad Prism software that generated EC50 values of 0.30 µg/ml for Dk cells (R2 = 0.99) and 0.65 µg/ml for wild type (R2 = 0.97). (B) Effect of EB on BrdU incorporation into nuclear DNA of wild type BSF trypanosomes. Cells (105 cells/ml, 10 ml) were incubated with EB (2 µg/ml) and BrdU for up to 6 h. Total DNA at each time point was isolated, fractionated by agarose gel electrophoresis, transferred to a PVDF membrane, probed with anti-BrdU, and labeled DNA was detected by chemiluminescence (quantitated by Image J software; http://rsbweb.nih.gov/ij/). After normalizing for loading using the hexose transporter probe, label in the nuclear DNA band was quantitated by a phosphorimager. Inset graph shows that detection of BrdU was in the linear range (in this experiment, 1, 2.5, 5, 7.5, and 10 µl of the untreated 6 h sample was processed and quantitated with Image J. (C) Effect of EB concentration on kinetoplast size, nuclear replication, and growth of wild type cells. Cells (105 cells/ml, 10 ml) were incubated with the indicated concentrations of EB for 24 h and BrdU was added for the last 2 h. Cytotoxicity (killing) was assayed with 200 µl samples as in Panel A. To measure effect of EB on kinetoplast size, cultures (1 ml) were collected, fixed, stained with DAPI and evaluated by fluorescence microscopy. With regard to kinetoplast size, the reason that some cells retain some kDNA at high EB concentration is not clear. To measure the effect of EB on nuclear DNA replication, total DNA was isolated from remaining samples for electrophoresis as in Fig. 4A; a blot was probed with anti-BrdU and quantitated as in Panel B. Data were fit to the equation for the sigmoidal Emax model as in Panel A. (D) Survival of wild type and Dk trypanosomes in 0.02 µg/ml EB. Cells (5×104/ml) were incubated with 0.02 µg/ml EB and at indicated times were counted by hemocytometer (mean ± standard deviation of 3 independent experiments). Cell densities were maintained between 5×104/ml and 106/ml. (E) Comparison of the effects of EB concentration on free minicircles. The experiment was identical to that in Fig. 2A except that EB was either 2 µg/ml or 0.02 µg/ml. Abbreviations and * are same as in Fig, 2A. This experiment was run 3 times with virtually identical results. The fraction E smear in the second lane is narrower than that in Fig. 2A for unknown reasons. The numbers in each lane below fraction E are relative band intensities of fraction E plus covalently-closed minicircles determined by phosphorimaging and corrected for background and load. (F) Effect of 0.02 µg/ml EB on nuclear DNA replication. A culture was treated with EB and at indicated times, cells (200 µl, 2×106/ml) were incubated with [3H]thymidine (Perkin-Elmer, 300 µCi/ml, 20 Ci/mmole) in thymidine-free HMI-9 for 2 h (the serum may contain low levels of thymidine). At the end of the labeling, radioactivity was measured in 5% TCA-precipitable DNA. Values on Y-axis are percent of incorporation into untreated cells. The latter incorporated 10.5×103±2.8×103 dpm in 2 h. Roughly 5% of the zero time incorporation should be in kDNA. Each experiment was run with duplicate samples, and plotted values are Mean ± S.D. of three independent experiments.