Direct Evidence of Apex-Hypha Interactions During Vegetative Growth of Fungal Thallus via Comprehensive Network and Trajectory Extraction

Abstract

The mycelium of a filamentous fungus is a growing, branching network of numerous entangled hyphae exhibiting polarised apical growth. Expansion occurs during the vegetative phase from a single ascospore, driven by the need to explore and occupy surrounding space-limiting competitors, enhancing nutrient uptake, and promoting spore dispersal. Radial, rapid, and rectilinear growth combined with frequent branching appears adaptive. However, passive growth without interactions or feedback may produce suboptimal networks, as neither local density nor potential connectivity is considered. Reorientations of the apex near existing hyphae suggest apex-hypha feedback. Yet, the diversity of behaviours, spontaneous fluctuations, and limited apical trajectories studied leave open the question of active regulation. To investigate possible apex-hypha interactions, we analyse a dataset of Podospora anserina thallus growth by reconstructing all apical trajectories post-branching and fitting them with a classical Langevin model that incorporates potential interactions. Comparing isolated and non-isolated hyphae trajectories allows to identify a clear signature of interaction composed of abrupt deceleration and reorientation. This work opens the path towards a systematic exploration of hyphal interactions.

Figures (5)

• Figure 1: Examples of possible interactions within a small portion of the mycelium of P. anserina. From top to bottom and left to right, the reader can observe a probable anastomosis, a meeting followed by a branching event, a stop-and-go, an avoidance and two successive overlaps. The supplementary materials offer the opportunity to view the movie of the growth of this section of the mycelium.
• Figure 2: A) Set of aligned trajectory of free branches (initial growth vector aligned to the right) and obstructed branches (initial growth vector aligned to the left).Insert A-i) and A-ii) are example of respectively free (i) and obstructed (ii) interaction area on top of an apex.B) distribution of tortuosity $\tau$ of free branches (in green) and obstructed branches (in red) after 1h, 3h and 5h of branch growth. The boxes represent the range of distribution between the first and third quartiles, with an orange line indicating the median. The whiskers show the range extending to the distribution extremes, or to 1.5 times the interquartile range, whichever is closer to the median. The dots correspond to measurements that fall outside this range. C) density function estimate of distance $r$ between starting point and the apex of free branches (in green) and obstructed branches (in red) after 1h, 3h and 5h of branch growth. Kolmogorov-Smirnov tests have confirmed that the distributions of free and obstructed branches are statistically different at any time and in both cases (part B and C of the figure) (p-value always lower than $3\,10^{-4}$).
• Figure 3: A) scatter plot of elongation speed as a function of time since the start of the growth of unobstructed apical branches for the network B, estimated to be representative of the five networks. Green dashed line correspond to the model evaluated on this network with yellow shade indicating three standard deviation estimated via bootstrap. Black line show the mean model over the five networks with the gray area showing three standard deviation. B) example of reconstruction of the network C showing, in black, the lateral branches, in green, the free apical branches and in red, the obstructed apical branches (pink portions indicates growth under a free interaction area while darker red indicate growth occurring under obstructed interaction area). Cyan dots represent successive apexes positions. Scale bar: 200.
• Figure 4: Example of 2D histograms $\boldsymbol{\eta}+\boldsymbol{I}$ estimations made from the model for free branches (A), where $\boldsymbol{I} = \boldsymbol{0}$, and obstructed branches (B) for the network A. Green dashed lines correspond to normal law centered at zero and of same standard deviation as the empirical distribution of free branches $\Sigma$ ; orange dashed lines correspond to numerical error estimations $\epsilon$ ; black dashed lines correspond to the normal law centered at zero and of standard deviation $\sigma_X = \sqrt{\Sigma_X^2-\epsilon_X^2}$ where $X$ corresponds to either $\theta$ or $v$, depending on the projection to which it refers. Dashed circles centred at $\boldsymbol{0}$ are of radius 1;2;3 $\sigma_\theta$ All theses lines are reported on the (B) side for comparison between free and obstructed branches.
• Figure 5: Characteristic (means $\mu$ and standard deviations $\Sigma$) of the distributions of $\eta_v$ and $\eta_\theta$ depending on the delay $T$ between two consecutive images.