**Figure 4.**

*CLSM micrographs of cyanobacterium Synechocystis 1336, visualized with CLSM. Green channel (a) represents fluorescence from phycobilins at 656 nm, red channel (b) from chlorophyll a at 682 nm, (c) represents transmission image and an overlap of all three is shown in (d). Scale bar = 10 μm.*

## **Figure 5.**

*False color overlay images of Phormidium CALU 624 cells at stationary phase of growth are visualized by CLSM simultaneously in three fluorescence detection channels. (a) Channel 1 (light blue): fluorescence signal from PE (575–585 nm), (b) channel 2 (green): PC-APC (650–660 nm), (c) channel 3 (red): Chl a (678–710 nm), and (d) channel 4: transmission microphoto. Scale bar = 25 μm.*

**49**

*Confocal Laser Scanning Microscopy for Spectroscopic Studies of Living Photosynthetic Cells*

fluorescence detection. Fluorescence images (**Figure 5(a)–(c)**) were obtained at different emission wavelengths, excited by a 405 nm laser beam. The fluorescence signal was collected at 647–652, 658–663, and 675–680 nm, and the false colors light blue, green, and red in **Figure 5(a)**–**(c)** represent the fluorescence emitted by PE, PC-APC, and Chl a, respectively. The highest intensity signal of PE-fluorescence is located in the center of the cells (white arrows), and the strongest signal from Chl a fluorescence is located at the cell periphery, the latter provides the evidence of the

*Fluorescence spectra of individual phycobiliproteins, which can be used as reference data for spectral unmixing. Spectra were normalized to their maximum, and all values were adapted to our measurement conditions. The dashed lines indicate the respective fluorescence maxima (PE—580 nm; PC—645 nm; APC—660 nm; Chl* 

Fluorescence spectra of living cyanobacterial cells are composed of overlapping fluorescence emission spectra from several photosynthetic pigments. The application of spectral unmixing with reference spectra for individual pigments (**Figure 6**) can give more pronounced and detailed images of pigment localization. Spectral unmixing can also be used to calculate the relative fluorescence shares of the

Spectral unmixing is a method for the complete separation (unmixing) images with overlapping emission spectra. It is used with specimens labeled with more than one fluorescent dye, exhibiting excitation and emission crosstalk or with self-fluorescent specimens contained of several photosynthetic pigments

If we regard a pixel of a lambda stack that represents a locus in the specimen where several fluorescent pigments with their known reference spectra overlap, the cumulative measured spectrum can be expressed as a linear combination of the reference spectra multiplied by the corresponding intensity for each pigment. By means of known reference spectra, this equation can be solved for the intensities of each pigment. The reference spectra can either be loaded from a spectra database, or from literature [40–44], or directly extracted from the lambda stack. Obtained linear unmixing function will generate a multi-channel image, in which each channel represents only one pigment. The accuracy of the technique allows the complete unmixing even of such dyes or self-fluorescence pigments whose spectra have

Unfortunately, in living cyanobacterial cells, it meets some difficulties. The problems are caused by the fact that the light absorption and emission properties of isolated phycobiliproteins are rather different from those of the intact phycobilisomes in the living cyanobacterial cells. In living cells, the spectral properties

*DOI: http://dx.doi.org/10.5772/intechopen.84825*

thylakoid arrangement.

*(PSII)—682 nm; Chl (PSII)—710 nm).*

**Figure 6.**

(as in our case).

almost identical emission maxima.

individual pigments contributing to the spectrum.

*Confocal Laser Scanning Microscopy for Spectroscopic Studies of Living Photosynthetic Cells DOI: http://dx.doi.org/10.5772/intechopen.84825*

## **Figure 6.**

*Color Detection*

special sample/experiment.

utilization and energy balance [26].

**48**

**Figure 5.**

**Figure 4.**

*CLSM micrographs of cyanobacterium Synechocystis 1336, visualized with CLSM. Green channel (a) represents fluorescence from phycobilins at 656 nm, red channel (b) from chlorophyll a at 682 nm, (c) represents transmission* 

*False color overlay images of Phormidium CALU 624 cells at stationary phase of growth are visualized by CLSM simultaneously in three fluorescence detection channels. (a) Channel 1 (light blue): fluorescence signal from PE (575–585 nm), (b) channel 2 (green): PC-APC (650–660 nm), (c) channel 3 (red): Chl a (678–710 nm), and (d)* 

size, in turn, varies with the objective used, the number of pixels per scan field, and the zoom factor. Pixels that are too large degrade resolution, whereas pixels too small require longer scanning times and thus bleach the specimen. The optimum pixel size for a given objective and a given zoom factor should be obtained for each

In this section, we examined changes in pigments of live, unfixed cells using spectral imaging. This technique captures an entire emission spectrum from every cell. The images corresponding to the four photosynthetic pigments have been pseudocolored for visualization purposes. The pseudocoloring is based on the colors

In **Figure 4**, images illustrated differences in fluorescence between diverse physiological states of Microcystis CALU 398 cells are presented. Two channels for fluorescence detection and one transmission channel were used. Channel 1 (a), colored false green, represents averaged fluorescence of PC and APC in the range 650–660 nm, channel 2 (b), colored false red, shows Chl a fluorescence at 678–710 nm, channel 3 (c) is a transmission microphoto, and in (d) all three channels are superimposed. Here small signal in both fluorescent channels corresponds to healthy cells, cells in bad physiological state (dying cells) have a high PC-fluorescence (green arrows), and dead cells have no fluorescence (white arrows). The high intensity in the shorter wavelength interval reflects that fluorescence in dying cells mainly originates from phycobilisomes at ca 650–660 nm. Healthy cells displayed weaker emission intensity over the whole spectral range, located at the periphery of the cells. This response correlates with effective light

**Figure 5** represents confocal laser scanning photomicrographs illustrating photosynthetic pigment localization in filamentous cyanobacterium Phormidium CALU 624. False color overlay images were obtained by simultaneous three channel

shown in **Figure 5**: Chl a is red, PC-APC is green, and PE is light blue.

*image and an overlap of all three is shown in (d). Scale bar = 10 μm.*

*channel 4: transmission microphoto. Scale bar = 25 μm.*

*Fluorescence spectra of individual phycobiliproteins, which can be used as reference data for spectral unmixing. Spectra were normalized to their maximum, and all values were adapted to our measurement conditions. The dashed lines indicate the respective fluorescence maxima (PE—580 nm; PC—645 nm; APC—660 nm; Chl (PSII)—682 nm; Chl (PSII)—710 nm).*

fluorescence detection. Fluorescence images (**Figure 5(a)–(c)**) were obtained at different emission wavelengths, excited by a 405 nm laser beam. The fluorescence signal was collected at 647–652, 658–663, and 675–680 nm, and the false colors light blue, green, and red in **Figure 5(a)**–**(c)** represent the fluorescence emitted by PE, PC-APC, and Chl a, respectively. The highest intensity signal of PE-fluorescence is located in the center of the cells (white arrows), and the strongest signal from Chl a fluorescence is located at the cell periphery, the latter provides the evidence of the thylakoid arrangement.

Fluorescence spectra of living cyanobacterial cells are composed of overlapping fluorescence emission spectra from several photosynthetic pigments. The application of spectral unmixing with reference spectra for individual pigments (**Figure 6**) can give more pronounced and detailed images of pigment localization. Spectral unmixing can also be used to calculate the relative fluorescence shares of the individual pigments contributing to the spectrum.

Spectral unmixing is a method for the complete separation (unmixing) images with overlapping emission spectra. It is used with specimens labeled with more than one fluorescent dye, exhibiting excitation and emission crosstalk or with self-fluorescent specimens contained of several photosynthetic pigments (as in our case).

If we regard a pixel of a lambda stack that represents a locus in the specimen where several fluorescent pigments with their known reference spectra overlap, the cumulative measured spectrum can be expressed as a linear combination of the reference spectra multiplied by the corresponding intensity for each pigment. By means of known reference spectra, this equation can be solved for the intensities of each pigment. The reference spectra can either be loaded from a spectra database, or from literature [40–44], or directly extracted from the lambda stack. Obtained linear unmixing function will generate a multi-channel image, in which each channel represents only one pigment. The accuracy of the technique allows the complete unmixing even of such dyes or self-fluorescence pigments whose spectra have almost identical emission maxima.

Unfortunately, in living cyanobacterial cells, it meets some difficulties. The problems are caused by the fact that the light absorption and emission properties of isolated phycobiliproteins are rather different from those of the intact phycobilisomes in the living cyanobacterial cells. In living cells, the spectral properties

### **Figure 7.**

*Simultaneous observation of self-fluorescent photosynthetic cells of Microcystis CALU 398 and labeled by quantum dots extracellular polysaccharide. (a) Channel 1: quantum dots fluorescence (green), (b) channel 2: natural pigment fluorescence (red), (d) channel 3: corresponding transmission image, (d) all three channels superimposed. Fluorescence spectra in (e) have the colors corresponding to (a) and (b). Scale bar = 5 μm.*

of pigments from certain organisms may differ crucially from the properties of the dissolved ones, for example, spectra of the components can vary in peak widths and may be shifted in wavelength due to the different pigment-protein and linker connections. In this case, only hyperspectral CLSM measurements [42] can give suitable reference spectra for decomposition.

If the overlapping of the emission spectra of the fluorescent components is not very much, then the image spectral unmixing can be easily obtained without any calculations. The example of such unmixing is presented in **Figure 7**. In **Figure 7**, the usage of combination of natural pigment fluorescence and fluorescence probes for imaging of cyanobacteria and their extracellular polymeric substances (polysaccharides) is demonstrated. To target extracellular substances, fluorescent labels should be used. Here, nonfluorescent polysaccharides were labeled by quantum dots CdSe/ZnS. Quantum dots have a fluorescence in green (520–580 nm) when excited by ultraviolet light (405 nm) and photosynthetic pigments of Microcystis CALU 398 have a fluorescence in red (620–780 nm). Thus the fluorescent signals will not overlap. The reference spectra for quantum dots and Microcystis CALU 398 cells were obtained by means of lambda scanning (**Figure 7(e)**).
