Microorganisms respond to environmental physicochemical conditions, such as chemical concentration (Colombetti and Diehn, 1978), gravity (Lebert and Häder, 1996), and light intensity (Diehn, 1969; Diehn, 1973; Creutz et al., 1978; Hader, 1987). Swimming microorganisms spontaneously generate macroscopic convection patterns originating from a response called “taxis” (Pedley et al., 1988; Hill and Pedley, 2005; Bees, 2020). When the environmental conditions prompt the organisms to swim upward, the cell density near the water surface increases and a top-heavy condition is generated. Once this unstable distribution is broken owing to fluctuation, the microorganisms settle locally and sink down owing to their own weight. Thereafter, upward swimming recommences, and the collective behavior of the population results in a spontaneous convection pattern. This phenomenon is known as “bioconvection.”

Euglena is a microorganism that responds to light intensity (Ogawa et al., 2016; Tsang et al., 2018; Noselli et al., 2019; Ozasa et al., 2019; Muku et al., 2023) and is known to form a localized bioconvection pattern under strong illumination from the bottom of the culture medium (Suematsu et al., 2011; Shoji et al., 2014). Usually, light strength is constant in time and homogeneous in space in experiments. However, in nature, the environmental conditions are neither constant nor homogeneous. Therefore, the response of bioconvection patterns to dynamic non-uniform illumination will yield a greater understanding of how microorganisms adapt to periodical changes in environmental conditions.

Here, we suggest temporally periodic illumination as the next simplest condition after constant uniform illumination. The bioconvection of Euglena was observed under spatially homogeneous illumination, whose light intensity changed periodically over time. With a long period rather than forming a pattern, bioconvection alternates formation and decomposition. By contrast, a complex spatiotemporal behavior was observed under a short period of flashing illumination. Based on our experimental observations, we suggest one of the response mechanisms of microorganisms to the periodically changing environmental conditions.

Euglena gracilis was cultured in Koren–Hutner (KH) solution for 1 week (Koren and Hutner, 1967) and then inoculated into HYPONeX aqueous solution to observe bioconvection. The culture was diluted to the desired cell density and then placed in the dark for 1 h before use for bioconvection observation. The cell density was adjusted to 1.0 × 10⁶ cells/mL. All incubation processes were performed under 28°C and light conditions with 14 h bright and 10 h dark.

The HYPONeX culture was soaked in a custom-made container in which a silicone sheet with a circular hole was placed on a glass plate and covered with another glass plate (Figure 1A). As a result, the culture solution was shielded into the hole of the silicone sheet with a thickness of 3 mm and diameter of 40 mm.

A bioconvection pattern was observed under temporally changing illumination intensity. A flat-light panel (LED Flat Light, Aitec System Co., Ltd., Japan) was connected to a controller (Arduino UNO). The light intensity changed between 0 lx (ϕ _min) and 3000 lx (ϕ _max) with periods (T) of 8, 20, 40, 80, and 200 s, where the period of bright was 3T/4 and of dark was T/4 (Figure 1B). The experiments curried out under room temperature, which was 24°C. Snapshots were captured using a digital camera (Eos Kiss X8, Canon, Japan). Images were analyzed using ImageJ (NIH, United States) on PC.

The convection pattern originated from the negative phototaxis of Euglena, which induced upward swimming of cells. The gathering of cells at the top of the culture lead to a density instability. Consequently, downward flows occurred at these times. Therefore, the downward flow appeared as a green spot when viewed from above. Once the illumination from the bottom stopped, the driving force was lost, and the pattern disappeared during a few tens of seconds. The wavelength of bioconvection pattern once decreased and then increased with increasing the period of the light intensity oscillation (Figure 1E). In addition, the temporal evolution of the pattern strongly depended on the period of external oscillatory illumination (Figure 2).

The swimming speed of the cells was approximately 100 μm/s and the depth of the container was 3 mm. Therefore, more than 30 s was required to reach the top. However, cells often change their direction. Therefore, it might take several hundred seconds to gather the cells at the top and induce convection. The period of 200 s of external light was comparable to the period of upward swimming of the cell. Based on this speculation, the disappearance of the spots in the initial stage (Figure 2A) was induced due to the lack of a driving force for the long period case.

In the initial two or three cycles, the cells were sensitive to illumination because they were cultured in the dark before starting the experiments. Therefore, convection spots were generated in a short time. However, illumination for several hundred seconds induced adaptation of the cells; thus, the sensitivity was slightly weakened. Owing to the weakness of the light sensitivity, the gathering at the top also slowed. As a result, the cells might fail to fulfill the convection-inducing condition for 150 s illumination. Accordingly, the convection pattern disappeared. On the other hand, the lateral gathering of the cells occurred slowly, as could be seen in the space-time diagram (Figure 2A). By increasing the local cell density, gathering at the top could be easily overcome. Hence, after several cycles, a convection pattern appeared again in the localized region. In this time, the wavelength of the pattern became longer than that in constant illumination. This result is expected because of diffusion during long dark period, which decreased the local cell density resulting in long wavelength pattern.

In the case of the shorter period of 20 s, the cells experienced both high and low relative intensity during upward swimming. Although the detailed mechanism is still under investigation, the characteristic behavior observed in the convection patterns may originate from the alternation between upward swimming and random motion. It is known that Euglena shows tumbling behavior when the light intensity suddenly increases. Therefore, the upward swimming cells start to swim in a random direction while the light intensity is decreasing, and then tumble in response to a sudden increase in light intensity. On the other hand, downward moving cells experience gradual increase in intensity owing to the shading effect (Vincent and Hill, 1996; Ghorai and Hill, 2005). Notably, some of the cells can respond to strong light and others fail to respond to it and show tumbling. This complicated situation generates a characteristic time evolution of convection pattern, as shown in Figure 2D. Furthermore, the wavelength became slightly shorter than that observed under constant light illumination (Figure 1E). It might originate from failure to deformation of the pattern because of short dark period (5 s), which was shorter than the period to disappearing the pattern (Figure 1C). Uncovering the detailed mechanism of those characteristic observations is a topic for the future works. In addition, with period 8 s, a similar complicated pattern formation was observed on one of the three experiments. Thus, the threshold value of the period was not clear for observing the complicated pattern formation.

Finally, under short-period light, the convection pattern is expected to approach to that observed under constant illumination, in which the convection spots monotonically gathered and formed localized pattern. Because the cells cannot perceive the oscillation but feel an effective constant illumination with a time-averaged intensity under fast oscillation. This is also the future works.

In this study, we demonstrated one of the simplest setups for self-organized behavior in a temporally changing environment. If the oscillation period was similar in order of time scale for pattern formation behavior, non-trivial characteristic patterns appeared. In our study, the characteristic time was the upward swimming period, which comprised several tens of seconds. Taking a scientific approach to dissect a complex dynamical system behavior, the setup and environmental conditions are typically simplified. As such, it remains important to devise more complicated experimental models to understand these underlying mechanisms. Nevertheless, our work provides a first step toward understanding the spatiotemporal pattern formation of biological systems exposed to periodical environmental fluctuations. Our preliminary experiments clearly showed that a slight change in environmental conditions often induces unexpected and interesting behaviors. We should consider the effect of such a minimally complicated condition as a diorama of more notably complicated natural dynamical systems.