Fig 1.
Assessing benznidazole treatment conditions to optimise the generation of T. cruzi persisters.
(a) MA104 cells were infected with T. cruzi CL Luc::mNeon [22] in 24-well plates containing glass coverslips, incubated for 3 days, and then treated with benznidazole for 8 days at a range of concentrations (5–30 μM) (Materials and Methods). Each day, selected cultures were exposed to 10 μM EdU for 6 hours, fixed on coverslips, developed and scanned by fluorescence microscopy. The data are presented as % infected host cells containing at least one EdU+ve amastigote. (b) Upper image; an infected cell prior to drug exposure, with parasites that are in S-phase during the 6 hour exposure period shown in red (EdU incorporation). DAPI staining (blue) identifies host cell nuclei and amastigotes, which are recognisable by their distinctive disc-like kinetoplast genome (orange arrow). Lower image; an infected cell after 8 days exposure to 10 μM benznidazole. The two amastigotes highlighted by arrows did not undergo DNA replication during the 6 hour period of EdU exposure. The adjacent red stained host cell nucleus serves as a positive control for EdU labelling. White scale bars = 10 μm. (c) Infected cell monolayers in 24-well plates (as above) were treated with 20 μM benznidazole. On the days indicated, a coverslip was fixed and the amastigote content of randomly selected infected cells recorded. By day 5, all imaged infected cells contained a single amastigote. Average intracellular burden indicated by red line. (d) Infected monolayers in 24-well plates were treated with benznidazole for 8 days at the concentrations indicated. After washing, cultures were maintained in MEM for 32 days and monitored for the appearance of extracellular differentiated trypomastigotes. This assay period (a total of ~50 days) is the limit attainable with monolayers of MA104 cells.
Fig 2.
Isolation of intracellular T. cruzi persisters by live cell sorting following in vitro benznidazole treatment.
(a) Experimental outline. Cultures of MA104 cells were infected with T. cruzi CL Luc::mNeon parasites. After 72 hours, they were treated with 20 μM benznidazole for a further 8 days. Cellular suspensions were then generated for analysis by live cell sorting (Materials and Methods). (b) Live fluorescence images of a cellular suspension showing DNA (Hoechst staining, blue), non-viable cells (propidium iodide (PI) staining, red) and parasite-infected cells (green fluorescence). In the merged image (bottom right), the white arrow indicates a T. cruzi infected MA104 cell. Yellow scale bar = 50 μm. (c) Fractionation of infected cell suspension, following PI staining, using an BD FACSAria Cell Sorter. The small percentage of non-viable cells can be separated on the basis of acquired PI fluorescence. (d) Fractionation of live host cells into infected (within red circle) and non-infected sub-populations based on the green fluorescence of parasite persisters. 0, background fluorescence; G, lowest green fluorescence gating. (e) Image of an infected cell sub-population suspension post-sorting. The number of parasites in each cell is indicated (1 or 2). (f) Sorted infected cells 18 hours post-plating (Materials and Methods). DNA, (blue); amastigotes, (green). Scale bar = 25 μM.
Fig 3.
Assessing the replicative status of persister parasites in vitro following benznidazole treatment.
MA104 cells were infected with T. cruzi CL Luc::mNeon in 24-well plates as indicated, and after 3 days growth, they were treated with 20 μM benznidazole for 8 days. When benznidazole was removed, the cells were maintained in complete MEM and monitored by fluorescence microscopy. To identify parasites undergoing DNA replication, cultures were exposed to 10 μM EdU for 6 hours and the coverslips processed for analysis using a Zeiss LSM880 confocal microscope (Materials and Methods). (a) Amastigote numbers in infected cells immediately after drug removal, and at day 11 post-treatment. The number of EdU+ve parasites is shown in red. (b) Timeline of an independent experiment in which the number of EdU+ve amastigotes was assessed periodically after the cessation of treatment. (c) A pre-treatment parasite nest in which ~50% of the amastigotes are in S-phase during the period of EdU exposure (red). DAPI staining (blue) identifies host cell nuclei (large) and the parasite kinetoplast DNA (small, intense blue discs). (d) At 6 days post-wash, amastigotes (green) are in a non-replicative state. An MA104 cell in S-phase is identified by nuclear EdU staining (red). (e) Images showing parasites that have re-entered the cell cycle (day 11) and are undergoing asynchronous DNA replication [24]. White scale bars = 10 μm.
Fig 4.
Fragmentation of T. cruzi DNA following benznidazole treatment.
(a) Experimental outline. MA104 cells were infected with T. cruzi CL Luc::mNeon in 24-well plates. They were treated with either benznidazole (BZ) (200 μM) for 24 hours, or tert-butyl hydroperoxide (TBHP) (50 μM) for 3 days. Post-fixation, as a control group, untreated cells were treated with DNase. TUNEL assays were then performed and cells imaged using a Nikon Ti-2 E inverted microscope (Materials and Methods). (b) Representative images showing infected cells following each of the treatments. Parasites (green fluorescence), DNA (blue, DAPI), TUNEL (red). The enlarged inset (left) highlights replicating kinetoplast DNA (kDNA). The white arrows in the non-treated images show the location of a highly infected cell in which all the parasites have differentiated into TUNEL-negative non-replicating trypomastigotes. White scale bars = 10 μm.
Fig 5.
Monitoring the tissue-specific impact of non-curative benznidazole treatment by bioluminescence imaging.
(a) Representative in vivo images of T. cruzi infected BALB/c mice after once daily oral treatment with 25 mg/kg benznidazole for 5 days (Materials and Methods). Treatment was initiated in the acute stage either 9 (pre-peak) or 14 (peak) days post-infection (dpi). The percentage drop in whole-body bioluminescence is indicated (n = 6 in each cohort). (b) Schematic showing the arrangement of tissues, organs and carcass used for ex vivo imaging. (c) Ex vivo images of non-treated and treated infected mice. The heat-map for both ex vivo and in vivo imaging is on a log10 scale and indicates the intensity of bioluminescence from low (blue) to high (red) with minimum and maximum radiance values as indicated. (d) Fluorescent detection of parasites in the bladder and cardiac muscle of non-treated and treated mice during the acute stage of infection, using a Zeiss LSM880 confocal laser scanning microscope. DNA (red, DAPI); parasites (green fluorescence; yellow, if on a red background). White scale bars = 20 μm. Lower images show expanded view of single amastigote infections.
Fig 6.
In benznidazole treated mice, the majority of cells that remain infected contain only a single parasite.
BALB/c mice in the pre-peak (9 dpi) and peak (14 dpi) stages of infection were treated once daily with 25 mg/kg benznidazole for 5 days (as in Fig 5). Serial sections (10 μm) from a range of tissues were prepared and examined in 3-dimensions by z-stacking to determine the precise number of amastigotes in each infected cell (Materials and Methods) (see also S1 Video). (a) Illustrative 0.7 μm serial images across a section of cardiac tissue from a non-treated mouse. Amastigotes (green); DNA (blue, DAPI). Parasite numbers can be determined with precision by counting the distinctive intensely stained kinetoplast (mitochondrial) DNA that co-localises with green fluorescence across serial sections. White scale bar = 5 μm. (b) Parasite numbers per infected cell determined by exhaustive screening of multiple sections obtained from a range of tissues from treated (orange bars) and non-treated (blue bars) mice.
Fig 7.
Parasites that persist in mouse cardiac tissue after benznidazole treatment are in a non-replicative state.
CB17 SCID mice were infected with T. cruzi and 10 days post-infection they were treated once daily with 25 mg/kg benznidazole for 5 days. EdU labelling was then carried out as described previously (Materials and Methods). Mice were euthanised (16 dpi), and cardiac sections prepared and imaged using a Nikon Ti-2 E inverted microscope (9 mice per group, 3 randomly selected sections from each mouse) (a) Infection burden per infected cardiac cell (nest), with the number of EdU+ve/-ve parasites indicated (see also S3 Fig). (b) Images showing an infected cardiomyocyte from a non-treated mouse that contains both replicating and non-replicating amastigotes. The inset shows an example of a single serial cross section derived by 3-dimensional confocal laser scanning microscopy (z-stacking), which was used to determine the precise number of amastigotes in the infected cell (white arrow indicates intensely stained kinetoplast DNA) (see S1 Video for 3-dimensional image of this infected cell). DNA (blue, DAPI); EdU+ve amastigotes (red); parasites (green fluorescence). White scale bar = 10 μm. (c) Images of infected cardiomyocytes from benznidazole-treated mice. None of the amastigotes detected were EdU+ve. A purple arrow indicates two host cells that were in S-phase during the period of EdU exposure (upper image). White scale bars = 20 μM. The insets (right) show enlarged images of DAPI stained host cell nuclei and single infecting amastigotes (green). Full 3-dimensional images of parasites that persist after benznidazole treatment are shown in S2 and S3 Videos.