Modulation of host cell mechanics by Trypanosoma cruzi

Abstract

To investigate the effects of Trypanosoma cruzi on the mechanical properties of infected host cells, cytoskeletal stiffness and remodeling dynamics were measured in parasite-infected fibroblasts. We find that cell stiffness decreases in a time-dependent fashion in T. cruzi-infected human foreskin fibroblasts without a significant change in the dynamics of cytoskeletal remodeling. In contrast, cells exposed to T. cruzi secreted/released components become significantly stiffer within 2 h of exposure and exhibit increased remodeling dynamics. These findings represent the first direct mechanical data to suggest a physical picture in which an intact, stiff, and rapidly remodeling cytoskeleton facilitates early stages of T. cruzi invasion and parasite retention, followed by subsequent softening and disassembly of the cytoskeleton to accommodate intracellular replication of parasites. We further suggest that these changes occur through protein kinase A and inhibition of the Rho/Rho kinase signaling pathway. In the context of tissue infection, changes in host cell mechanics could adversely affect the function of the infected organs, and may play an important role on the pathophysiology of Chagas' disease.

Figure 1. Reduction in the F-actin content in host fibroblasts

A. Alexafluor 488-phalloidin staining of T. cruzi-infected HFF at 72 hours post-infection (left panel). Parasitized cells, as shown in the phase contrast image (center panel), lacked the actin filaments present in uninfected cells. The arrows highlight three infected cells. B. Biochemical quantification of F-actin to G-actin ratio in HFF treated with Jasplakinolide (Jas), cytochalasin D (CytD) or following T. cruzi infection for 72 hours. C. Quantification of relative F-actin content (% of total actin) following infection. Results from 4 independent experiments were quantified using densitometry of western blots revealing the reduction in F-actin in T. cruzi-infected cells (*p <0.05, **p<0.01, ***p<0.005 Student's t test).

Figure 2. Storage modulus g′(f) and loss modulus g″(f) under baseline condition vs. frequency

Solid line: best fit by Eq. (1). Median values over n = 1381 measurements.

Figure 3. Changes in mechanical properties in fibroblasts infected with T. cruzi or treated with PCM

T. cruzi-infected HFF at 2, 24, 48 and 72 hr post-infection (open symbols) or PCM-treated cells after 2 hr (closed symbols) were subjected to OMTC and analyzed as follows: A. g₀ vs. time. Solid line: best fit by linear function of infection data. B. Power law exponent α vs. time. Dashed line: best fit by linear function of infection data. C. Newtonian viscosity μ vs. time. Solid line: best fit by linear function of infection data. All three parameters are obtained from fit of g′ and g″ with Eq. (1). Numbers of measurements are: for PCM data 582, and for infection data 1381, 681, 1126, 616, and 572 for 0, 2, 24, 48, and 72 hr respectively.

Figure 4. Changes in mechanical properties after challenge with jasplakinolide and cytochalasinD

Mock- and T. cruzi-infected HFF (72 hr) were treated with Jasplakinolide (Jas) or cytochalasin-D (CytD) and analyzed for (A) stiffness, g₀, (B) power law exponent, α, and (C) Newtonian viscosity, μ, are obtained from fit with Eq. (1). Error bars are standard errors. Numbers of measurements taken are between 200 and 800.

Figure 5. Changes in mean square displacements, MSD, vs. time lag, Δt, in T. cruzi infected cells

HFF, at different time points after infection (0, 2, 24, 48, and 72 hr) and after 2 hr incubation with PCM were analyzed for changes in MSD. Inset: MSD taken at Δt = 100 s vs. time after infection; error bars are standard errors. Baseline, n = 1623; PCM 2h, n = 876; TC 2h, n = 727; TC 24h, n = 1131; TC 48h, n = 607; TC 72h, n = 622.

Figure 6. T. cruzi elicits changes in Rho kinase and PKA-dependent substrate phosphorylation

Lysates prepared from mock- or T. cruzi-infected HFF (C or Inf) were blotted and probed with antibodies to phospho-MLC(19-20), phospho-MYPT1-Thr853, phospho-MYPT1-Ser695, stripped and reprobed with MLC or MYPT1 as appropriate (A) and densitometric analysis of the bands was carried out to determine the average relative phosphorylation levels from 2 independent experiments (B). Mock- and T. cruzi-infected (C; Inf) or PCM-treated (PCM) HFF lysates were subjected to western blotting and probed with antiphospho-PKA substrate antibody (C,D). A mixture of isolated parasites 2 × 10⁶trypomastigote/amastigote mix was included as a control for phosphorylated parasite proteins (Tc). Arrow indicates differentially phosphorylated protein band that is not present in isolated parasites.