Adhesion of Crithidia fasciculata promotes a rapid change in developmental fate driven by cAMP signaling

Abstract

Trypanosomatids are single-celled parasites responsible for human and animal disease. Typically, colonization of an insect host is required for transmission. Stable attachment of parasites to insect tissues via their single flagellum coincides with differentiation and morphological changes. Although attachment is a conserved stage in trypanosomatid life cycles, the molecular mechanisms are not well understood. To study this process, we elaborate upon an in vitro model in which the swimming form of the trypanosomatid Crithidia fasciculata rapidly differentiates following adhesion to artificial substrates. Live imaging of cells transitioning from swimming to attached shows parasites undergoing a defined sequence of events, including an initial adhesion near the base of the flagellum immediately followed by flagellar shortening, cell rounding, and the formation of a hemidesmosome-like attachment plaque between the tip of the shortened flagellum and the substrate. Quantitative proteomics of swimming versus attached parasites suggests differential regulation of cyclic adenosine monophosphate (cAMP)-based signaling proteins. We have localized two of these proteins to the flagellum of swimming C. fasciculata; however, both are absent from the shortened flagellum of attached cells. Pharmacological inhibition of cAMP phosphodiesterases increased cAMP levels in the cell and prevented attachment. Further, treatment with inhibitor did not affect the growth rate of either swimming or established attached cells, indicating that its effect is limited to a critical window during the early stages of adhesion. These data suggest that cAMP signaling is required for attachment of C. fasciculata and that flagellar signaling domains may be reorganized during differentiation and attachment.IMPORTANCETrypanosomatid parasites cause significant disease burden worldwide and require insect vectors for transmission. In the insect, parasites attach to tissues, sometimes dividing as attached cells or producing motile, infectious forms. The significance and cellular mechanisms of attachment are relatively unexplored. Here, we exploit a model trypanosomatid that attaches robustly to artificial surfaces to better understand this process. This attachment recapitulates that observed in vivo and can be used to define the stages and morphological features of attachment as well as conditions that impact attachment efficiency. We have identified proteins that are enriched in either swimming or attached parasites, supporting a role for the cyclic AMP signaling pathway in the transition from swimming to attached. As this pathway has already been implicated in environmental sensing and developmental transitions in trypanosomatids, our data provide new insights into activities required for parasite survival in their insect hosts.

Keywords: Crithidia fasciculata; adhesion; attachment; cyclic AMP; differentiation; kinetoplastid; signaling.

Fig 1

An adherence assay reveals features of C. fasciculata attachment. (A) Schematic showing steps of the assay, including the variables tested. (B) Representative images showing attached rosettes produced by varying the initial plating time. The dashed red lines were drawn to indicate what was counted manually as individual rosettes, defined as clusters containing four or more cells. (C) Quantitation of the experiment shown in (B). This experiment was performed in triplicate. Error bars show standard error. (D) The effect of initial cell concentration on adherence. 10⁷ cells/mL from cultures in different phases of growth were adhered for each sample. Quantitation of a representative experiment of three replicates is shown with median values indicated by red lines. (E) Rosettes resulting from adherence of different numbers of cells, all derived from a log phase culture (10⁷ cells/mL). A representative experiment of three replicates is shown. Red lines indicate median values, and the blue dotted line represents the proportion of plated cells that were able to adhere.

Fig 2

Culture conditions can impact attachment. (A) Log-phase cells were allowed to attach for 2 hours in BHI medium adjusted to the indicated pH levels. Following washing with PBS, plates were imaged for single attached cells. *P < 0.05; **P < 0.01. Kruskal–Wallis. (B) Standard attachment assay performed in non-treated flasks or (C) tissue culture-treated flasks. Log phase cells were allowed to attach for 2 hours, followed by washing and imaging for quantitation. The mean of two replicates is shown. Error bars are standard error.

Fig 3

Flagellar shortening occurs at a relatively constant rate following an initial adhesion event. (A) Following 5 minutes of plating and subsequent washing, cells were observed at the indicated times and the length of the flagellum was measured in 20 representative cells (different cells were measured at each time point). (B) An example of a single cell undergoing flagellar shortening. The numbers indicate time in minutes. The schematic on the right shows a representation of the length of the flagellum at the beginning (red) and end (blue) of the time course.

Fig 4

Transmission electron microscopy reveals structural differences between attached and swimming cells. (A) Cross-section through C. fasciculata parasites attached to tissue culture plastic (Fig. S2, yellow plane). Asterisks indicate attachment plaques at the distal tip of the shortened flagella. Scale bar is 1 µm. The dotted box indicates the region enlarged in panel (A’). (A’) Close up of an attachment plaque (ap) from one of the cells in panel A. FAZ structures, connecting the flagellum to the cell body, are also observed (fz). Scale bar is 200 nm. (B) and (C) Additional examples of attachment plaques (ap) and FAZ (fz). The cell in (C) was scraped from the dish prior to fixation. Scale bar is 200 nm. (D) Sectioning en face shows a region near the end of the flagellum in an adherent cell (Fig. S2, blue planes). The flagellar membrane is enlarged, and no flagellar axoneme can be observed. Instead, diffuse filaments probably comprising the attachment plaque are seen. Scale bar is 200 nm. (E) Another example of an attachment plaque sectioned en face. Scale bar is 200 nm. (F) Section of a swimming cell shows the flagellum extending from the flagellar pocket. Dotted box indicates the region enlarged in (F’). Scale bar is 1 µm. (F’) Higher magnification view of the flagellar pocket of a swimming cell. FAZ structures link the flagellar membrane to the flagellar pocket membrane, but no hemidesmosome is seen. The slight bulge in the flagellum where it exits the pocket may be a physical correlate to the site of initial adhesion. Filamentous material can be observed in the flagellar pocket of the swimming cell but was never observed on the surface. In contrast, the surface of attached cells (G) was covered with this material. Some of the same material is also visible in panels (A–E). Scale bar of F’ and G is 400 nm.

Fig 5

Proteomics analysis of swimming and attached C. fasciculata. (A) Volcano plot highlighting differentially abundant proteins upregulated in attached forms (right side) versus swimming forms (left side). Four cAMP phosphodiesterases are indicated. (B) Trimmed gene ontology (GO) terms associated with proteins upregulated in swimming (orange) or attached (blue) parasites.

Fig 6

Components of cAMP signaling localize to different domains of the flagellum in C. fasciculata swimmers. confocal images of (A) CfPDEA::GFP, (B) CfRAC1::GFP, and (C) CfPF16::YFP fusion proteins with DAPI staining of nuclear and kDNA of fixed swimmer and attached cell cultures. A Z-stack of the entire cell or rosette was captured. Maximum projections of individual deconvolved images are shown. A brightfield image for each field is included for reference and was used to generate the white dashed outline of the flagellum in the merged and brightfield images. Scale bar 5 µm. (D–F) Western blots of extracts of swimmers (Sw) and attached (Att) cells of each line probed with an antibody to GFP (α-GFP) or BiP (α-BiP) as a loading control.

Fig 7

Phosphodiesterase inhibitors block C. fasciculata adherence. (A) Attachment assays performed in the presence of 10 µM of either NPD-001 (NPD1) or NPD-008 (NPD8). The number of single attached cells at 2 h (open circles) and the number of rosettes (≥4 cells, gray squares) at 24 h were counted and compared with those from a vehicle-treated control sample (Veh) and an untreated sample (Con). Treatments were present in the media both during the 2-h plating time and the subsequent 24 h of culture. All comparisons were made. Asterisks indicate Kruskal-– test with Dunn’s correction P < 0.0001 (****) and P < 0.05 (*). (B) Mass spectrometry-based detection of NPD-001. Individual dots show the mass spectrometry (MS) peak area for NPD-001-treated (NPD1) that was consistent across all 15 samples, and DMSO-treated vehicle control that had five samples with very low peaks and 10 samples with no detectable signal. The red line indicates the mean. Peaks for cAMP (red bars) and AMP (blue bars) were converted to a fold-change between DMSO and NPD1 treatments. There was an average 3.6-fold increase in cAMP and no difference in AMP. Metabolomics with performed with three biological replicates each with five technical replicates. Error bars indicate the standard error. (C) Graph of C. fasciculata growth in shaking cultures assayed over approximately 31 h in standard medium (Control; gray circles), medium supplemented containing 10 µM NPD-001 in vehicle (NPD1; orange diamonds), or medium containing vehicle only (Vehicle, black squares). Mean of three replicates. Error bars show standard deviation. (D) Time-encoded movies were used to measure the doubling time of established attached cultures after the addition of NPD-001 in vehicle or in vehicle alone. The red line indicates the mean. (E) Representative images and (F) quantitation of attachment assays comparing vehicle and untreated controls to assays performed in the presence of different concentrations of NPD-001 or (G) NPD-226. The number of single adhered cells at 2 h (open circles) and the number of rosettes (≥4 cells, gray squares) at 24 h were counted. All comparisons were made, but only the groups that are significantly different from the adjacent treatment are shown. Asterisks indicate Kruskal–Wallis test with Dunn’s correction P < 0.0001 (****), and P < 0.01 (**).