【Background】
　Cyanobacteria (blue-green algae), a group of oxygenic photosynthetic bacteria, inhabit a wide range of environments--oceans, rivers, lakes, terrestrial habitats, hot springs and deserts--and supply oxygen and organic matter. They adapt to their environments by forming diverse biofilms such as blooms and microbial mats, which in turn serve as habitats and nutrient sources for other organisms. Thus, elucidating the structure and formation mechanisms of cyanobacterial biofilms is important not only for cyanobacterial physiology but also from an ecological perspective. The main structural components of cyanobacterial biofilms are extracellular polysaccharides produced outside the cells. Cyanobacteria are known to synthesize highly diverse extracellular polysaccharides in a species-specific manner. Among these, sulfated polysaccharides, in which the sugar chains are modified with sulfate groups, are produced by only a limited number of bacterial species. In contrast, many cyanobacterial species have been shown to produce a wide variety of sulfated polysaccharides. We therefore consider sulfated polysaccharides to be one of the key factors that define the structure and function of cyanobacterial biofilms, and we have previously identified cyanobacterial sulfated polysaccharides and elucidated their biosynthetic and regulatory pathways and functions¹. In reality, cyanobacterial biofilms contain not only sulfated polysaccharides but also various other polysaccharides, proteins, nucleic acids and other biomolecules. To understand the functions of these components, it is important not only to perform genetic analyses but also to directly capture the spatial distributions of molecules and cells within the biofilm. To visualize extracellular sulfated polysaccharides, imaging with basic dyes such as Alcian blue has been widely used. However, such staining can markedly perturb the distribution of polysaccharides and the structure of cyanobacterial biofilms, making it difficult to observe the distribution of sulfated polysaccharides in a state close to their native one (Fig. 1).

Figure 1.	A (A) Culture of Synechocystis sp. PCC 6803 (S.6803) during the process of forming a bloom-like biofilm. (B) Alcian blue-stained image of the culture shown in (A). Sulfated polysaccharides are stained blue. In addition to the distribution of cells, characteristic distributions of sulfated polysaccharides are observe. However, because Alcian blue itself causes sulfated polysaccharides to aggregate and can alter their distribution and the cyanobacterial biofilm structure as a whole, it is highly likely that this image does not faithfully reflect the actual distribution of sulfated polysaccharides.

　

　Vibrational spectroscopic techniques that detect molecular Raman scattering or infrared absorption can be used to visualize the molecular composition of samples without staining, and Raman spectroscopy in particular has been widely applied to compositional analysis of living cells, bacteria and biofilms. However, cyanobacteria exhibit strong autofluorescence derived from photosynthetic pigments, and the weak Raman scattering signals are buried under this autofluorescence, making conventional Raman microscopy difficult to apply. In contrast, infrared absorption spectroscopy detects light signals in the mid-infrared region and is therefore free from the influence of autofluorescence, making it suitable for measuring cyanobacterial biofilms. Nevertheless, because the spatial resolution of conventional infrared spectroscopic imaging is on the order of 10 µm, it has been difficult to visualize the spatial distributions of cells and molecules within biofilms.

　To overcome these limitations, we used a super-resolution mid-infrared spectroscopic imaging technique recently developed by Assistant Professor Ryo Kato at Osaka University and colleagues, which addresses the weaknesses of these conventional methods. By applying this technique, we succeeded in achieving label-free, high-resolution visualization of sulfated polysaccharides and cyanobacterial cells within cyanobacterial biofilms^{2, 3}. In this article, we outline this approach and findings.

【Results】
　A major factor limiting the spatial resolution of microscopy is the wavelength of the light used to read out the signal. The mid-infrared photothermal microscope used in this study irradiates the sample with pulsed light in the mid-infrared region and detects the transient photothermal effects induced by molecular infrared absorption (changes in refractive index and volume expansion) using visible light whose wavelength is ten times shorter than that of the mid-infrared light. In this way, it enables infrared absorption analysis with submicron spatial resolution comparable to that of visible-light microscopy (Fig. 2). Another major advantage of this method is that, simply by inserting fluorescence filters, we can separate and detect the fluorescence signal from the sample and the mid-infrared signal (light at the same wavelength as the incident mid-infrared beam), allowing us to image infrared absorption information and fluorescence simultaneously and at the same spatial resolution. We therefore focused on the infrared absorption signal characteristic of sulfate groups in sulfated polysaccharides and on the chlorophyll autofluorescence characteristic of cyanobacteria.

Figure 2.	(A) Schematic of the mid-infrared photothermal microscope. (B) Infrared absorption image of polymer particles (diameter 3.5 µm) acquired with a quantum-cascade-laser-based infrared spectroscopic microscope (conventional method), and infrared intensity image of polymer particles (diameter 0.5 µm) acquired with the mid-infrared photothermal microscope (super-resolution). Images were obtained using the intensity of the vibrational mode of the aromatic ring C=C bond at 1494 cm⁻¹. (C) Comparison of intensity line profiles along the line indicated in (B). The method used in this study achieves more than tenfold higher spatial resolution than the conventional method.

　

　When we measured the infrared absorption spectra of cyanobacterial cells and purified target sulfated polysaccharides, the cells showed signals mainly originating from proteins, whereas the sulfated polysaccharides exhibited signals that included a peak at 1494 cm⁻¹ attributable to sulfate groups (Fig. 3A). Based on these results, we prepared cyanobacterial biofilm samples on transparent substrates, measured the infrared absorption intensity of the sulfated polysaccharides and the autofluorescence intensity of the bacteria, and imaged these together with bright-field images (Fig. 3B, C).As a result, we succeeded in label-free super-resolution visualization of cyanobacteria and sulfated polysaccharides within the biofilm, and observed that the sulfated polysaccharides formed fibrous structures along which the cyanobacterial cells were aligned. At the same time, we also found regions within the biofilm where only linear structures of sulfated polysaccharides were observed (Fig. 4).

Figure 3.	(A) Infrared absorption spectra of cyanobacteria and sulfated polysaccharides. (B) Bright-field image of a cyanobacterial biofilm. (C) Autofluorescence and mid-infrared absorption intensity images of the same field of view as in (B). Cyanobacterial cells, imaged in green based on chlorophyll fluorescence intensity, are observed to be distributed along linearly distributed sulfated polysaccharides.
Figure 4.	(A) Super-resolution infrared spectroscopic imaging of a cyanobacterial biofilm. (B) Intensity line profile along the white line segment in (A). These data demonstrate that imaging of cyanobacterial biofilm samples is achieved with a spatial resolution of 500 nm.

 

【Discussion and Outlook】
　From our previous macroscopic observations of bloom-like biofilms formed by this cyanobacterial species, we had hypothesized that cyanobacterial cells first attach to sulfated polysaccharides that have spread in a fibrous manner throughout the culture medium, and that these cell-polysaccharide complexes then self-assemble, driven for example by the buoyancy of gas bubbles, to form dense biofilms¹. Alcian blue staining-based imaging of sulfated polysaccharides had revealed fibrous and aggregated sulfated polysaccharides colocalized with cells; however, we could not rule out the possibility that these structures were artifacts introduced by the staining procedure. In this study, under unstained conditions, we successfully observed sulfated polysaccharides distributed in linear structures with cells attached along them. We also observed regions in which only linear distributions of sulfated polysaccharides were present without attached cells. These findings support our model in which sulfated polysaccharides first spread out in the medium, cells then attach to them, and finally biofilms are formed through self-assembly. Moreover, this technique can also be used to visualize the distributions of other molecular species, such as proteins and lipids, by exploiting signals characteristic of those components. In infrared absorption intensity images based on amide bonds in proteins within the cyanobacterial biofilm, we observed line-like structures similar to those of the sulfated polysaccharides, suggesting that proteins may also be involved in biofilm formation and may interact with polysaccharide components to help construct the biofilm architecture.

　To our knowledge, this is the first demonstration of label-free, high-resolution visualization of sulfated polysaccharides, the main components of cyanobacterial biofilms--together with the cyanobacteria themselves. In the future, combining this approach with data analysis methods such as principal component analysis and machine learning should make it possible to visualize the localization of a wide variety of polysaccharides, including sulfated polysaccharides, as well as enzymes, extracellular DNA and other molecules present in cyanobacterial biofilms, thereby further deepening our understanding of biofilm formation and function.

　In addition, sulfated polysaccharides constitute a group of high-value, useful polysaccharides that are used as ingredients in cosmetics and pharmaceuticals, and cyanobacteria are attracting attention as producers of such sulfated polysaccharides. If this technique can be used to easily evaluate the quality and quantity of sulfated polysaccharides produced by different cyanobacterial strains, it will also be valuable for applied research.

　Furthermore, biofilms themselves are formed in nature by highly diverse microorganisms, not only cyanobacteria, and are extremely important subjects of study in fields such as ecology, medicine and agriculture. For example, biofilm formation is crucial during infections by pathogenic bacteria, and microbial biofilms in the environment are sites of information transfer mediated by biomolecules. We therefore expect that the microspectroscopic techniques used in this study will also become powerful tools in these broader areas of biofilm research.

【References】

1.	Kaisei Maeda, Yukiko Okuda, Gen Enomoto, Satoru Watanabe, Masahiko Ikeuchi. Biosynthesis of a sulfated exopolysaccharide, synechan, and bloom formation in the model cyanobacterium Synechocystis sp. strain PCC 6803. eLife, 2021, 10:e66538.
2.	Ryo Kato*, Kaisei Maeda*, Taka-aki Yano, Kan Tanaka and Takuo Tanaka. (*Co-corresponding author), Label-Free Visualization of Photosynthetic Microbial Biofilms using Mid-Infrared Photothermal and Autofluorescence Imaging, Analyst, 2023, 148, 6241-6247, cover art.
3.	Ryo Kato; Kaisei Maeda. Super-resolution mid-infrared spectroscopic imaging to reveal formation of bacterial biofilms. Biophysics, 2024, 64, 321-323.