Modeling the reactive oxygen species (ROS) wave in Chlamydomonas reinhardtii colonies

Abstract

Reactive oxygen species (ROS) play a crucial role as signaling molecules in both plant and animal cells, enabling rapid responses to various stimuli. Among the many cellular mechanisms used to generate and transduce ROS signals, ROS-induced-ROS release (RIRR) is emerging as an important pathway involved in the responses of various multicellular and unicellular organisms to environmental stresses. In RIRR, one cellular compartment, organelle, or cell generates or releases ROS, triggering an increased ROS production and release by another compartment, organelle, or cell, thereby giving rise to a fast propagating ROS wave. This RIRR-based signal relay has been demonstrated to facilitate mitochondria-to-mitochondria communication in animal cells and long-distance systemic signaling in plants in response to biotic and abiotic stresses. More recently, it has been discovered that different unicellular microorganism communities also exhibit a RIRR cell-to-cell signaling process triggered by a localized stress treatment. However, the precise mechanism underlying the propagation of the ROS signal among cells within these unicellular communities remained elusive. In this study, we employed a reaction-diffusion model incorporating the RIRR mechanism to analyze the propagation of ROS-mediated signals. By effectively balancing production and scavenging processes, our model successfully reproduces the experimental ROS signal velocities observed in unicellular green algae (Chlamydomonas reinhardtii) colonies grown on agar plates, furthering our understanding of intercellular ROS communication.

Keywords: Cell-to-cell signaling; ROS wave; ROS-Induced-ROS release; Reaction-diffusion.

Figure 1.

Measurements of the ROS wave in communities of C. reinhardtii cells grown as a lawn on agar plates at different dilutions. (A) Representative time-lapse images showing ROS accumulation in agar plates containing lawns of C. reinhardtii cells prepared at dilutions of 1.2, 0.12, and 0.012 OD and treated with a focused beam of high light (local treatment; solid circles). (B) Quantification of local (solid circles) and systemic (dashed circles) ROS levels at 0 and 30 minutes post-treatment across different cell densities. Data shown as box plot graphs; X is mean ± S.E., N=30, *P<0.05, Student t-test. Scale bar, 1 cm. (C) Table of experimentally measured ROS wave velocities at each dilutions. All experiments were repeated at least 3 times with 10 agar plates per experiment. (D) Representative snapshots of experimentally observed C. reinhardtii distributions. Top row depicts experimental image snapshots with OD values of 1.2, 0.12, and 0.012 (left to right), bottom row corresponds to the color-mapped snapshots used for estimating the colony coverage $\rho$ via pixel counting. Scale bar, $100\mu \text{m}$. (E) Plot showing experimentally observed ROS wave velocity $v$ versus colony coverage $\rho$. Abbreviations: OD, optical density; TRE, total radiant efficiency; CFU, colony forming unit.

Figure 2.

The conceptual model illustrating the mechanisms underlying the ROS wave propagation. (A) The three key processes within a single C. reinhardtii cell. (B) A simplified model for the RIRR signaling pathway. The ROS produced by NOXs (red circles) undergo three paths: (1) The produced ROS can diffuse along or away from the cell surface, activating ROS-producing NOXs in the same or adjacent cells (red arrows); (2) The diffused ROS can be transported into the cell (blue arrows) through aquaporin channels (yellow ellipses); (3) The transported ROS can be scavenged (green arrows) inside the cell (cytoplasm; gray). The interplay between ROS production, transport, and scavenging gives rise to the self-propagating ROS waves observed in algal communities. Not shown is the potential scavenging of ROS outside of cells.

Figure 3.

Results obtained from the continuum model. (A) Experimentally observed ROS wave velocities (black dots with error bars) and the calculated wave velocity-colony coverage $(v-\rho$) curve (grey). (B) The fitted $\alpha$ and $\gamma$ functions used for deriving velocities in (A). (C) ROS wave front concentration profiles over $\xi$ for different colony coverages. (D) Simulated ROS concentration profiles for a particular colony coverage ($\rho =0.9276$) at different timestamps. The initial activation area is centered at the origin with a radius of 1 mm.

Figure 4.

Example snapshots of computationally sampled C. reinhardtii distributions. (A) Illustration of the sampled C. reinhardtii distribution that satisfies the ratio threshold for an OD value of 0.12 (i.e., $\rho =0.2283$). Blue circle encloses the stimulated area with a radius of 1 mm. 23,062 pseudo colonies are confined in a virtual 2D square plate with a size of 2×2 cm. Each colony in the plate is characterized by a circle with diameter randomly sampled from the Gaussian distribution with mean $\mu$ and standard deviation $\sigma$ set to $75\mu \text{m}$ and $10\mu \text{m}$, respectively. (B) A zoomed-in view of a 2×2 cm sampled colony distributions (colonies are shown in grey color).

Figure 5.

Representative snapshots of simulated ROS wave propagation in communities of C. reinhardtii grown on agar plates at different timestamps and the derived ROS wave velocities for different colony coverages (A,D,G) $\rho =0.1126$, (B,E,H) $\rho =0.2283$, and (C,F,I) $\rho =0.9276$. (A-F) Snapshots of the simulated ROS concentration distributions at different timestamps. (A-C) At the beginning of the simulation (T=10 s), ROS are still confined within the initially stimulated region (blue circle with a radius of 1 mm). (D-F) ROS concentration distributions after simulating the entire system for 80 seconds. The color bar represents the relative strength of the ROS signal (i.e., ROS concentration), where a higher concentration of ROS is depicted by a redder color. (G-I) Calculated ROS wave velocities at different timestamps. The dashed lines (grey) represent the experimentally measured wave velocities. The model parameters obtained from the continuum model are employed for the simulation. The systems were simulated with time step of 10⁻³ s, grid size of $5\mu \text{m}$, and no-flux boundary conditions on the boundary of the (square) system. See Materials and Methods for an explanation of the parameters and variables and technical details of the simulations.