WO2005052559A1

WO2005052559A1 - Method of identifying pigments from a single cell using confocal image spectrophotometry in phototrophic communities

Info

Publication number: WO2005052559A1
Application number: PCT/ES2004/000527
Authority: WO
Inventors: María Concepción HERNÁNDEZ MARINÉ; Mónica ROLDÁN MOLINA; Susana Castel Gil
Original assignee: Universidad De Barcelona
Priority date: 2003-11-27
Filing date: 2004-11-26
Publication date: 2005-06-09
Also published as: ES2237319A1; ES2237319B1

Abstract

The invention relates to a non-invasive method of analysing a sample emitting low-excitation laser fluorescence without any prior labelling, using a confocal laser scanning microscope which is connected to a spectrophotometer detector. The inventive method can be used for in-vivo three-dimensional position finding in relation to each community and for direct in-situ analysis of the fluorescent pigments of an individual cell in intact bulk samples. The relationship between said two determinations enables the pigments to be identified, the groups and species present in the sample to be subsequently identified and the physiological state of each individual organism to be ascertained. The invention can be used to identify problematic proliferations of algae in aquatic and aerophytic ecosystems.

Description

Identification method of single cell pigments by confocal image spectrophotometry in phototrophic communities.

This invention relates to the field of environmental technologies, and particularly to the identification of living phototrophic organisms, that is, organisms that use light as an energy source.

STATE OF THE PREVIOUS TECHNIQUE

Identification of a complex living community and discrimination between phylogenetic groups are particularly useful in the fields of aquaculture management and the sciences involved in the study of ecosystems, such as ecology, ecophysiology, oceanography, and limnology. Such identification and discrimination can be used, for example, to analyze incidence and to develop control strategies against problematic algal blooms that appear within aquatic and aerophytic ecosystems (organisms that live at the air interface and a nonliving structure such as a building or a wall) in natural and artificial conditions. Unfortunately, complex problem communities have been difficult to predict in the past, mainly due to the lack of appropriate methodologies and technologies. The nature of the samples and the need to analyze them without prior isolation from the organisms are other important obstacles to overcome. Current efforts typically depend on microscopic evaluation, but evaluation has limitations such as being laborious, generating variable data, requiring an intrusive sampling program with areas of limited coverage, and lacking near real-time coverage. As an alternative to microscopic evaluation, analyzes based on cellular absorbance and / or fluorescence measurements have been proposed (cf. JJ Cullen et al., "Optical detection and assessment of algal blooms", Limnol. Oceanogr. 1997, vol. 42, pp. 1223-39), but these analyzes are not without limitations. A limitation of discriminating algae by absorbance / fluorescence measurements is the need that the spectra corresponding to the phylogenetic groups or species involved must be sufficiently different. Furthermore, algal pigments are often heterogeneously distributed within each cell, affecting the extent of absorption. Although pigments can be extracted, the extraction procedures carry the risks of the presence of artifacts in the preparation and removal of the original microenvironment from the chromophore. Another known limitation of differentiating algae by absorbance / fluorescence measurements is related to the need for solid empirical approaches to statistically discriminate both the algal component of the composite signal for a given mass of water, and a single algal component by itself (cf. DF Millie et al., "Using absorbance and fluorescence spectra to discriminate microalgae", Eur. J. Phvcol. 2002, vol. 37, pp. 313-22). There have been attempts to improve the use of cellular absorbance / fluorescence techniques, but the improved techniques cannot yet analyze three-dimensional (3D) samples, a fact that prevents discriminating morphologies, measuring emission spectra in a single-pixel area, and analyze 3D samples without altering them.

Phototrophic organisms produce various types of photosynthetic pigments, each of which captures photons in a narrow range of the spectrum. A fraction of the energy absorbed by the pigments can be emitted immediately at a longer wavelength, a phenomenon known as fluorescence. The emitted fluorescence originates primarily from the photosystem II subantenna pigments (PSII) and is the result of the photosystem's inability to use all of the absorbed energy. Since photosynthesis and fluorescence are competitive processes, changes in photosynthetic activity are reflected in variations in fluorescence emission. Thus, fluorescence is an indicator of photosynthetic processes in plants, algae and cyanobacteria, being a powerful analysis tool that allows the description of a complex community in terms of its physiological state, transfer energy, evolution cell and discrimination between phylogenetic groups.

Fluorescence microscopy techniques have allowed the description of the structure and organization of the samples in two dimensions (2D). These techniques have been used to study photosynthetic microorganisms and various systems have been described both at the microscopic level (cf. L. Ying et al., "Fluorescence emission and absorption spectra of single Anabaena sp. Strain PCC7120 cells", Photochemistrv and Photobiology 2002, vol. 76, pp. 310-3) as at the macroscopic level (cf. HK Lichtenthaler et al., "Detection of vegetation stress via a new high resolution fluorescence imaging system ", __ Plant Phvsiologv 1996. vol. 148, pp. 559-612). But all these methodologies are not useful enough when the samples are made up of complex communities, such as microbial mats or biofilms In these cases, it would be desirable to do a first 3D microstructure survey and a subsequent in vivo location of the organisms of interest within the intact sample.

Thus, it is highly desirable to have a new method for rapid identification and discrimination of complex communities that overcomes some of the limitations of the methods known in the art. In particular, it would be useful if this new method were able to elucidate the 3D location of each community in vivo, and analyze the fluorescent pigments of a single cell in situ. within thick intact samples.

EXPLANATION OF THE INVENTION

To date, the confocal scanning laser microscope (CSLM) has been used primarily to obtain 3D images of samples with minimal preparation or alteration, given the ability of the CSLM to use multiple excitations and detect lengths of wave at different depths of focus. CSLM allows you to target specific elements within the sample, such as molecules, structures (eg surfaces, matrices) and properties (eg cell division phase). Some CSLMs comprise a selective spectrophotometric prism for detecting emission fluorescence. The combination of a CSLM and spectrophotometric detection is referred to herein as "confocal imaging spectrophotometry" (CIS). Overcoming common CSLMs, the CIS apparatus has the additional advantage of determining optimal detection and separating emission spectra from fluorochromes to avoid overlapping (ie, overlapping of signal channels). The present invention provides a new application for the CIS apparatus.

The inventors have surprisingly found a new method for identifying fluorescent signals in individual cells based on the combination of the power of the confocal scanning laser microscope (CSLM). with the capabilities of spectrophotometric methods. The new method allows the unequivocal in vivo identification of the taxonomic group of an individual cell, based on its fluorescence signal, without manipulating the phototrophic communities.

Thus, the invention provides a non-invasive method for analyzing a sample using a CSLM coupled to a spectrophotometer detector, comprising the steps of: (i) selecting a particular area of the sample; (ii) obtaining spectral scanning images of the emission fluorescence from the selected sample area at selected system settings; and (iii) processing said spectral scanning images according to definable algorithms to obtain an emission fluorescence spectrum of said area; where the sample fluoresces under laser excitation, without any previous labeling (that is, without the addition of fluorescent chemicals). Spectral scan images are the result of the spectral scan function ("lambdascan" function) of CIS devices. With the spectral sweep function, the fluorescence signal is captured at a predefined interval for each section, moving along the spectrum. Each individual measurement is based on the detection of real confocal images. In a particular embodiment, the method of the present invention uses the selected variables x-y-λ, that is, the optical plane is recorded at different wavelengths. In other particular embodiments, the method uses the variables x-z-λ, x-y-λ-t (t stands for time), and x-y-λ-z. The term "bandwidth" as used herein means the range of transmitted frequencies of a given signal. By "λ steps" is meant here the number of individual images detected in a specific range of wavelengths, from a single optical section. Images are recorded within a wavelength range, which is limited by their start and end points. The "step size λ" is the magnitude (expressed in nm) between the lower and upper limits of the wavelength range at which an image is recorded.

The appropriate software supplied with the specific CIS apparatus controls the detector during spectral scanning and aids in calculations, using definable algorithms, of the emission spectrum after scanning an image. Thus, the emission fluorescence spectrum of the selected area is obtained by processing said spectral scanning images. In a more particular embodiment of the invention, the selected area corresponds substantially to a single cell. In another particular embodiment, the selected area comprises at least one pixel of phototrophic organisms. The term "pixel", based on the words "picture" and "element", represents the smallest, indivisible element of an image in a two-dimensional system. In this description, both the sample points of a specimen and the points of an image are qualified as pixels. In a more particular embodiment of the invention, phototrophic organisms are selected from the group consisting of plants, algae, cyanobacteria, and mixtures thereof. Plants and algae can be macroscopic or microscopic. In another particular embodiment, the sample has a thickness equal to or less than the detection limit thickness of the used confocal scanning laser microscope.

In another embodiment, the system settings for obtaining the emission fluorescence sample data from the selected area of the sample are set in order to minimize photobleaching. Photobleaching is the loss of emission fluorescence intensity of the sample due to destruction of fluorescent substances by intense illumination. In particular, the step size λ is set between 5 and 40 nm, and the bandwidth is set from 360 to 800 nm. In another particular embodiment, performance and fit are held constant. The performance value ("gain valué") modifies the amplification of the detected signal and, consequently, the brightness and contrast of the image change. The adjustment value ("offset valué") defines a threshold value and, therefore, only those signals that are above the threshold value are detected and represented in the image. Finally, the laser excitation wavelength is set to one or more of the following values: 351nm (UV Ar laser), 364nm (UV Ar laser), 458nm (Ar laser), 476nm (Ar laser) , 488 nm (Ar laser), 514 nm (Ar laser), 543 nm (HeNe laser) and 633 nm (HeNe laser).

In another embodiment, the sample data is processed according to definable algorithms that provide two-dimensional plots of the mean fluorescence intensity versus the emission fluorescence wavelengths. The method of the present invention provides the in vivo and three-dimensional localization of each community and the direct analysis of the fluorescent pigments of a single cell in situ. in coarse intact samples. The relationship between these two determinations allows the identification of the pigments, the subsequent identification of the groups and species present in the sample, and the knowledge of the physiological state of each particular organism. In summary, the main improvements achieved with the new method are: (i) the analysis of single or multiple fluorescent pixels; (ii) 3D localization in vivo: (iii) direct in situ analysis of single cell fluorescent pigments in coarse samples without prior isolation; (iv) establishing the relationship between fluorescent properties and position within the specific microbial assembly; (v) the possible application of eight excitation wavelengths to obtain spectra of a single cell, providing high-resolution detection and detailed sample information; (vi) rapid access to statistical information on the number of cells and spectral properties of a community; (vii) discrimination of cells with particular fluorescent signals within the colony and correlation with individual cell states; and (viii) the free choice of the emission wavelength, which allows the discovery of new pigment signals.

Throughout the description and claims, the word "comprises" and its variants are not intended to exclude other technical characteristics, additives, components or steps. The summary of this application is incorporated herein by reference. For those skilled in the art, other objects, advantages and characteristics of the invention will emerge in part from the description and in part from the practice of the invention. The following particular embodiments and figures are provided by way of illustration, and are not intended to be limiting of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG 1. shows the in vivo spectrophotometric analysis of the BF1 biofilm of the Roman catacomb of St. Callistus. FIG. 1A. shows the extended focus pseudocolor 3D projection in the xy planes and the orthogonal view in the z direction of the biofilm (49 optical sections). In FIGs. 1 BD. I know they represent the in vivo spectral profiles derived from λ _exc of 488, 514 and 543 nm, and the standard error (n = 5 cells).

FIG 2. shows the spectrophotometric analysis of the BF2 biofilm in vivo of the Roman catacomb Domitilla. FIG. 2A. shows the extended focus 3D pseudocolor projection in the xy planes and the orthogonal view in the z direction of the biofilm (66 optical sections). In FIGs. 2B-D. The in vivo spectral profiles derived from λ _exc of 488, 514 and 543 nm, and the standard error (n = 5 cells) are represented.

DETAILED EXHIBITION OF MODES OF REALIZATION

CIS apparatus operation

Analysis was performed with a CIS microscope, using either the 63x (NA 1.32, oil) or 100x (NA 1.4, oil) objectives (magnification range 1-4). Spectral scanning was performed using the 351 and 364 nm lines of an Ar UV laser; lines 458, 476, 488, and 514 nm from an Ar laser; the 543 nm line from a green HeNe laser and the 633 nm line from a red HeNe laser. The microscope uses spectrophotometric detection that allows the system to perform different scans from 360 to 800 nm of the spectrum using a motorized slit in front of the photomultiplier. Although the CIS apparatus could acquire spectral images of 5nm in size between 360-800nm, each image sequence (that is, the spectral scan or "lambda-scan" function of the system) was obtained by scanning the Same optical section xy using as step size 20nm for detection (λ coordinate of a data set xy-λ) to avoid photobleaching. The emission detection was placed 4-9nm further from the excitation wavelength to avoid reflections from the laser beam.

The scans were performed using the beam filter, the substrate filter (for UV) or the triple dichroic filter (488/543/633). Data series x, y, λ was acquired at the z position where fluorescence was highest. Background noise in areas without a sample was measured and then used to correct the primary spectra in the thin sections. The laser struck the sample perpendicularly and, to avoid interference with background radiation (light from the laboratory or light from excitation sources), the images were captured in the dark. Performance and contrast were the same for each field at each excitation wavelength and were unaltered throughout the scanning process.

Fluorescence analysis

From the above data, the mean fluorescence intensity (MFI) of the xy-λ data series was obtained using the software supplied in conjunction with the microscope. The region of interest (ROÍ) function of the software was used to determine the spectral signal of a selected area of the captured image. An ROI can also be specified to determine the spectrum of each sample and the software displays the average intensity of all pixels within the ROI versus the wavelength. For pigment solutions, ROIs of 1000 µm ² (n = 10 regions) were analyzed. For biofilm analysis, 1 µm ² ROIs, taken from the thylakoid fluorescent region within the cell, were fixed in each series of xy-λ images. Spectral sweeps (lambdascans) of biofilms (n = 5 cells for each species present in the biofilm) were obtained for each excitation wavelength (λ _Θ χc) in at least three independent experiments. Numerical data were processed with Microsoft Excel ^® 97 or 2000. The mean and standard error were calculated for all regions or cells examined in each λ _exc . The maximums of the pigments corresponded to their dispersion interval in the different λ _exc .

Calibrations using pure pigments

The extracted pigments show variations in the fluorescence spectra when compared with the in vivo pigments, therefore a control with pure pigments was performed to compare them with the published studies.

The fat-soluble, chlorophyll (Chis) a and b pigments obtained from Spinacia oralacea and xanthophyll (Xant) from Medicago sativa (Sigma, St. Louis, MO, USA) were dissolved in pure ethanol. Water-soluble pigments such as R-phycoerythrin (R-PE) from Porphyra teñera and C-phycocyanin (C-PE) from Spirulina sp. they were dissolved in filtered distilled water. The standard solution of allophycocyanin- Mastioocladus laminosus XL (APC-XL) was dissolved in ammonium sulfate (60%) and potassium phosphate (pH = 7) (Sigma-Aldrich) with a final concentration of 38 mM. Four hundred µl of each pigment solution at a final concentration of 1 mg / ml was transferred to 8-well glass-bottom chambers (Nunc Laboratory-Tek ™, Nalge Nunc International, Roskilde, Denmark).

The spectra of the pure pigments (TABLE 1) correlated well with the published spectra of extracted pigments. Data are the mean ± standard error (n = 10 regions) from three independent experiments conducted under the same experimental conditions. The asterisk means presence of a shoulder. The maximum emission intervals of Chl a (672.4 ± 2.9 nm) and Chl b (662.4 ± 2.1 nm) partially overlapped. The maximum C-PE fluorescence was located around 577.2 ± 2.2 nm, with a shoulder around 660.1 ± 3 nm. This shoulder could correspond to another contaminating phycobiliprotein present in the standard. The C-PC peak fluorescence peak, located at 656.1 + 4.3 nm, seemed slightly offset from published references. However, changes in C-PC fluorescence maximum are reflected in the literature, depending on the extraction method used. C-PC extracted from fresh biomass showed a large maximum at 615 nm while an additional maximum was observed at 652 nm in dry samples.

TABLE 1. λm _ax and shoulders for different pure pigments by confocal image spectrophotometry in all λ ₀ v-.

Both C-PC and Xant showed weak fluorescence as was already known. While the weak fluorescence of Xant was to be expected because the main path of de-excitation occurs through the transition to the triplet state, which is not fluorescent, the reason for this low fluorescence in pure C-PC could not be explained. The maximum fluorescence for cross-linked allophycocyanin (APC-XL) was 676.2 ± 2.4 nm. The special structure of APC-XL, which prevented dissociation of the molecule in dilute solutions, probably affected the emission maximum and could not be related to the field material. On the other hand, this maximum coincided with a specific type of APC previously described. Other common low-energy forms of APC have the maximum fluorescence near 680 nm, which is similar to the emission of intact phycobilisomes while the APC of cyanobacteria had an emission maximum at 660 nm.

Preparation and analysis of biofilms

Biofilms are made up of populations or communities of microorganisms that adhere to environmental surfaces. These microorganisms are normally immersed in an extracellular polysaccharide that they synthesize themselves. Two aerophytic biofilms were selected in the CSLM observations to test the method with complex natural communities. The biofilms, which were described and identified, contained different phylogenetic groups (Cyanobacteria and Bacillariophyta). The first biofilm (BF1), obtained from the catacomb of St. Callistus (Rome, Italy), was mainly made up of Scvtonema julianum and Leptolvngbva sp. The second biofilm (BF2), obtained from the Domitilla catacomb (Rome, Italy), consisted of the Bacillariophyta Diadesmis gallica and an unidentified cyanobacterium from the Chroococcales group. Both biofilms were obtained from artificially lit surfaces. Fragments of biofilms were separated from their substrates (plaster, mortar or speleothems) or, rarely, were taken together with small pieces of their support. Biofilms were maintained in a 2mm layer of 10% agarose BG11 medium (1%, Merck), and processed for the first week. Biofilms and cultures were mounted in Nunc Lab-Tek ™ glass bottom chambers. The samples were processed at room temperature in the dark.

The extended focus images (that is, the image is divided into three frames representing the maximum fluorescence intensity projection for the xy, xz and yz planes) of the two stratified biofilms showed a differential distribution of the microorganisms in depth in the biofilm (FIG. 1 A and 2A). Emission spectra for 488, 514 and 543 nm of λ _{exc are shown} for each biofilm (FIG. 1 BD and 2B-D).

FIG. 1A shows an extended focus pseudocolor 3D projection in the xy and orthogonal planes in the z direction of the biofilm (49 optical sections). The image represents the maximum autofluorescence emitted in the 590-775 nm range when excited at 543 nm. The volume under observation was 465.03 x 465.03 x 398.73 μm ³ . Z-step: 0.4 μm. Magnification factor: 1. Thickness: 19.54 μm. FIGs. 1B-D show the in vivo spectral profiles derived from 488, 514 and 543 nm λ _exc and the standard error (n = 5 cells). MFI stands for Medium Fluorescence Intensity. The differences obtained in the emission fluorescence profiles obtained in the biofilm indicated the presence of different species of cyanobacteria. Leptolvngbva sp. it is represented with a solid line and Scvtonema ulianum with a dashed line. Magnification factor: 2.

According to the data obtained, in the biofilm BF1, the thin filaments of Leptolvngya sp. they were oriented horizontally above the width Scvtonema iulianum (FIG. 1A). Both cyanobacteria had a broad λ _m at _X at 658.4 ± 3 nm, from the overlap of the Chl a and the phycobiliproteins (FIG. 1B-D). Also, Leptolvngvbva sp., But not S. iulianum. presented an emission peak (579.7 ± 3.8 nm) attributable to the presence of C-PE (FIG. 2B-D).

FIG. 2A shows optical sections at 66 xy of the laminated biofilm, which consists of two layers, the upper layer consisting of Diadesmis gallica colonies and the lower layer consisting of Chroococcales colonies. The volume under observation was 75.82 x 75.82 x 98.51 μm ³ . Z-step: 0.1 μm. Magnification factor: 3.86. Thickness: 6.6 μm. The image represents the maximum autofluorescence emitted in the 590-775 nm range when excited at 543 nm. FIGs. 2B-D show the in vivo spectral profiles derived from λ _exc of 488, 514 and 543 nm, and the standard error (n = 5 cells). The difference in the emission profiles obtained in the biofilm indicate the presence of different groups of algae and cyanobacteria. Chroococcal is represented with a solid line and Diadesmis gallica with a dashed line. A large decrease in fluorescence was observed in this biofilm at λ _exc 543 nm, attributable to photobleaching after successive spectral scans. Magnification factor: 2.

CSLM analysis revealed two layers in the BF2 biofilm: Diadesmis gallica (Bacillariophyta) was mainly concentrated at the top of the biofilm while the unidentified Chroococcal formed a discontinuous layer at the bottom (FIG. 2A). D. gallica, 3 μm in diameter, showed less fluorescence than the cyanobacterium. Their λ _ma χ, at 676.2 ± 5 nm (FIG. 2B-D), did not coincide with the λ _max of the other groups due to the presence of Chl c. To avoid photobleaching, caused when these cells were consecutively excited with different λ _exc , different optical fields were used to obtain the emission spectra in all the λ _exc . The unidentified Chroococcal presented the same spectral shape as Leptolvngbva sp.

TABLE 2 shows the results of the fluorescence emission acquired from biofilms after excitation at four wavelengths (λ _e ∞) (351, 488, 514 and 543 nm). Each λ _max value was obtained from 5 cells. TABLE 2. Comparison of λ _ma χ for each λ_ _* _ of the two aerophytic biofilms.

Differences in mean fluorescence intensity (MFD

In the biofilms, the cyanobacteria showed the highest MFI in the 640-740 nm range at any of the λ _exc , when compared with the Bacillariophyta (FIG. 1 BD and 2B-D). Cyanobacteria also had a high MFI at 577-580 nm λ _max , due to C-PE.

Each photosynthetic pigment present in species absorbs light of a certain wavelength, but in general, pigments pick up photons from a wide range of wavelengths when excited at wavelengths of 351-633 nm. Only at 364 nm (UV) and 458 nm (blue) λ _exc was a smaller emission response observed in all microorganisms. In the case of cyanobacteria, this indicates that the minimum efficiency of the photoreceptors is 430-460 nm.

Claims

1. Non-invasive method for analyzing a sample using a scanning confocal laser microscope coupled to a spectrophotometer detector comprising the steps of:

(i) select a particular area of the sample;

(¡I) obtain spectral scan images of the emission fluorescence from the selected sample area at selected system settings; Y

(iii) process said spectral scan images according to definable algorithms to obtain an emission fluorescence spectrum of said area;

where the sample emits fluorescence under a laser excitation, without any previous marking.

2. Method according to claim 1, wherein the spectral scan images are x-y-λ images.

3. Method according to any of claims 1-2, wherein the selected area corresponds substantially to a single cell.

4. Method according to any of claims 1-2, wherein the selected area comprises at least one pixel of phototrophic organisms.

5. Method according to claim 4, wherein the phototrophic organisms are selected from the group consisting of plants, algae, cyanobacteria and mixtures thereof.

6. Method according to any of claims 1-2, wherein the system settings are selected to minimize photobleaching.

7. Method according to claim 6, wherein the pitch size λ is set between 5 and 40 nm, and the bandwidth is set from 360 to 800 nm.

8. Method according to any of claims 1-2, wherein the performance and contrast are kept constant.

9. Method according to any of claims 1-2, wherein the laser excitation wavelength is set to one or more of the following values: 351 nm (UV laser Ar), 364 nm (UV laser Ar), 458 nm ( Ar laser), 476 nm (Ar laser), 488 nm (Ar laser), 514 nm (Ar laser), 543 nm (HeNe laser) and 633 nm (HeNe laser).

10. Method according to any of claims 1-2, wherein the algorithms of step (iii) provide two-dimensional graphs of the average fluorescence intensity versus the emission fluorescence wavelengths.

11. A method according to any of claims 1-2, wherein the sample has a thickness equal to or less than the detection thickness of the confocal scanning laser microscope used.