Ever since Antonie van Leeuwenhoek ﬁrst observed bacteria squirming under a microscope 340 years ago, scientists have been studying the movements of these single-cell organisms. Many (but not all) bacteria are mobile, having evolved mechanisms to actively explore their surroundings. Some bacteria swim with the help of rotating ﬂagella, others secrete slime “jets” on which to glide, and still others extend and retract pili to pull themselves along a surface.
Bacteria make good use of their motility: they navigate toward ideal environmental conditions and food sources, and steer away from harm. Depending
Synechocystis belong to the ancient group of cyanobacteria, found in huge numbers in open water and as slimy bioﬁlms in terrestrial habitats. They are considered to be the ﬁrst organisms to carry out oxygenic photosynthesis, the production of sugars from carbon dioxide and water using light as an energy source, with oxygen released as a by-product. Sunlight is the main source of energy for Synechocystis, but it can also be harmful, as too much radiation damages the bacteria’s photo-synthetic apparatus. Synechocystis cells therefore seek places with optimal light conditions, a behavior termed phototaxis.
Research prior to our own studies of phototaxis in Synechocystis showed that when light is beamed at these bacteria, they move toward it in a fairly straight line. They also possess photoreceptors, a diverse group of proteins—present in animals’ eyes—that will react to the absorption of light of a speciﬁc wavelength. But how does a single cell measuring about three thousandths of a millimeter accurately detect where light is coming from? Although phototaxis is one of the oldest scientiﬁc observations of cyanobacterial behav-ior, no one understood how they did it—until we discovered the mechanism serendipitously.
Our collaboration began when one of our research colleagues, Conrad Mullineaux, professor at Queen Mary University of London, was visiting the working group of Annagret Wilde at the University of Freiburg in Germany. Mullineaux asked Jan Gerrit Korvink, head of the Institute of Microstructure Technology, in Karlsruhe, Germany, whether he knew of a way to measure the refraction index of a small bacterium. The refraction index describes an important optical property of lenses that bend light. At ﬁrst, Korvink believed bacteria are so small that there are no instruments to measure refraction. But on further consideration, he came up with a possible solution.
We ﬁrst studied phototaxis by watching green Synechocystis colo-nies slowly crawling toward a light across an agar plate illuminated from one side. These colonies formed ﬁnger-like projections consisting of numerous cells that always maintained perfect alignment with the light source. To validate that individual cells were orienting their movement toward incoming light, we developed software to track single cells with time-lapse videos, using an upright optical microscope. The microscope was focused on the surface of an agar plate. In addition to the traditional microscope illumination shining vertically through the plate, we pointed light-emitting diodes (LEDs) at the side of the plate to achieve side-on illumination, parallel to the agar surface. Evenly spread cells in the focal plane readily moved towards these LEDs without straying off course. The cells even integrated the signals from two LEDs shone simultaneously from two different corners and moved straight to the midpoint between the two lamps.
A key observation was that when we used a source directly below the plate to establish a steep light gradient across it, so that one edge of the plate was in darkness and the opposite edge was very bright, the cells moved randomly rather than collecting in an area with optimal light intensity. The cells’ ability to ﬁnd the light was therefore not guided by sensing changes in light intensity as they moved, but by an accurate perception of the direction of the light source. In this case, since the light source came from below the plate, they could not pull themselves toward it.
Our ﬁrst hypothesis to explain this phenomenon was that the light-absorbing membranes inside the cells were acting as shades, creating a light gradient between the illuminated side of the cell and the far side. That would tell the bacterium the direction of the light source. However, we calculated that a single cell can absorb at most twenty percent of incoming photons, an amount too low for the light in-tensity gradient across the cell produced by shading to be the physical mechanism of light direction sensing in Synechocystis. The question remained: how do the cells know where the light is coming from?
One day, while exposing the cells to intense LED-light from the side, we switched off the main microscope illumination. To our great surprise, we saw spots at the “back” of the cells, which we previously postulated to be shaded from the lateral LED. Our cells looked like small glass beads, focusing light on the side of the cell farthest from the source. These focused bright spots must have been there all the time, but they had been invisible to us because of the high microscope illumination needed to obtain a high contrast image of the moving cells. We immediately understood that bacteria are optical objects and interact with light!
Excited by our discovery, we used different microscopy techniques and illumination regimes to study the bacteria’s focusing ability and monitor their behavior. We knew that they could focus light; now we had to show that the cells use this ability to “see.”
Our ﬁrst task was to establish that the focal length of the cells is suitable for “bacterial vision;” that is, that they are not so long– or short-sighted so as to be effectively blind. For this experiment we employed a trick commonly used in laser ﬂuorescence microscopy to increase con-trast of objects close to the coverslip. Usually, microscope illumination is aligned with the central axis of the objective lens, but by positioning the excitation laser beam to the side of this axis, the beam emerges from the lens at an angle. When that angle is just below a certain critical angle at the coverslip-sample interface, the excitation beam is
refracted to such a degree that it effectively illuminates the sample from the side.
We used Synechocystis cells that had been modiﬁed to contain a ﬂuorescent protein that was evenly distributed between the cell membranes. This enabled us to visualize the distribution of excitation light at the cell periphery: wherever light fell on the cell membrane, these proteins would glow bright green. When illuminated from the top (or bottom), an even green ring of ﬂuorescence was visible, but when illuminated from the side, a focused spot of ﬂuorescence appeared on the side of the cell furthest from the illumination source. Thus our cells had a length appropriate for focusing light. Moreover, the light intensity at the focus spot at the rear of the cell was approximately four times higher than the light intensity on the opposite side. We therefore established a spatial illumination pattern within the cell that could signal the direction of the light source.
To capture a high-resolution image of the light pattern produced by Synechocystis cells, we turned to a technique commonly used in the production of computer chips. This process, called photolithography, involves ﬁrst creating a plate with holes or transparencies in the desired pattern, and then shining ultraviolet light through this plate onto a light-sensitive material. Inﬁnitesimally small patterns can be produced on the light-sensitive material. Instead of using a plate with holes, we shone light through single Synechocystis cells and imprinted the near-ﬁeld light scattering onto the light-sensitive material. Using an atomic force microscope, we measured the surface relief produced by the refracted light beneath and around each cell. The resulting pattern of ridges and valleys was an accurate reproduction of the light ﬁeld created by interference, refraction, and internal reﬂection in the bacterial cells. We also found that below the center of each cell, the light was focused to a remarkably sharp spot with a diameter less than the wavelength of the light used to draw it.
So far we had shown that Synechocystis focuses light at a scale that should be usable by the small cells. Next we tested whether the light intensity difference across the cells created by focusing is, in fact, the signal that the cells use to perceive direction. We again imaged Synechocystis cells moving toward an LED at the side of an agar plate. This time, we focused a weak laser beam in the path of some of the cells on the agar surface, providing a second, highly-localized light source. As before, the cells moved in a straight line toward the LED, and a focused spot from the LED illumination was visible on the back edge of the cells. However, while most cells car-ried on their steady march toward the LED, the cells that encountered the laser spot—experiencing a sudden increased light intensity at their leading edge—stopped and reversed direction! This ﬁnding conﬁrmed that the cells compare the intensity of light at their back and front edges and use the difference to direct their movement. When this polarity was reversed artiﬁcially, the cells were fooled into going in the opposite direction. A surprising implication is that the cells are moving away from the side detecting higher light intensity to reach the light source.
Our results show that the whole spherical cell body of Synechocystis acts like a lens. It focuses light basically the same way as the lenses of our eyes, even though the biological structures are totally different. As with the retina in the human eye, the upside-down, reversed image on the rear of the cell is detected by light receptors. Therefore, we can think of Synechocystis cells as the simplest possible camera eye. Because each bacterium is approximately half a billion times smaller than the human eye, we estimate that their angular resolution is only about twenty degrees, roughly 1,000 times lower than the sharp resolution of the human eye. Synechocystis might not have a full picture of their surroundings, but their “vision” gives them a clear sense of where light is coming from, and they may even perceive the blurred outlines of a person watching them.
Determining how a single spherical cell detects light direction has raised entirely new questions about how bacteria perceive the visual world. For example, some cyanobacteria are rod-shaped. Is it possible that they refract light and concentrate it at their poles for guidance? Can colonial species, in which cells are arranged in ﬁlaments like a string of pearls, guide light, like an optical ﬁber, deep into the shaded parts of a bioﬁlm? Or consider non-photosynthetic bacteria that share many physical properties with cyanobacteria: maybe they have adopted a simple vision for reasons still unknown? We suspect there are many fascinating discoveries yet to be made in this new ﬁeld of bacterial micro-optics research.
Knowing that bacteria make sense of their world by making an image of it and interpreting that image, without the neural pathways and processing used by organisms with eyes, adds a new dimension to philosophical discussions about what it means to see, perceive, and understand. With such discoveries as ours, perhaps humans will now see the world differently. --NS, TL