This technique offers a unique research to for proving your cell's identity or physiological characteristics at the single-cell level without the need for invasive tagging. The main advantages of this technique are that it facilitates single-cell level spatial resolution for your analysis, and that it allows the distinction between and background process. So this technique can potentially contribute to the identification and the phenotypic analysis of pathogenic microbes.
This technique can also be applied to the study of the phenotypic heterogeneity, or the monitoring of the physiological status of microbial population of interests. Demonstrating the procedure will be Kyosuke Takabe, an assistant professor from my laboratory. The confocal reflection microscopy and multichannel confocal microspectroscopy setup.
Connect the confocal microscope with descanned spectral channels to a photomultiplier tube. Equip the microscope with a high numerical aperture objective, with an adequate magnification. And equip the microscope with a half-reflection mirror to accommodate confocal reflection microscopy, which relies on the cellular scatter of incident light to visualize cell morphology.
For multichannel confocal microspectroscopy, equip the microscope with dichroic mirrors and use a laser power meter to adjust the illumination intensity for each excitation wavelength. Then set the output under the microscope to be constant through the excitation wavelengths. To image the bacteria through the microscope, set the pinhole size to 1.0 area unit in the microscope software and set the pixel dwell time for each excitation wavelength.
Set the scanning resolution. For small cells such as bacteria, use a scanning area of 1024 by 1024. Set the Z-Scanning range so that the region of interest is covered.
And using a spectral window of eight to 10 nanometers, set the D-Scan detector to capture the visible wavelength range. Then acquire multichannel confocal microspectroscopy images in a sequence from longest to shortest excitation wavelengths to create Z-Stacks of fluorescence images and confocal reflection microscopy images, and save the images in 16-bit tiff format. To perform cell segmentation and the reconstruction of single cell innate fluorescent signatures, open an appropriate image analysis software program and double click to open one of the provided scripts.
Under the editor tab, click run. A folder selection window will appear. Select the directory in which the Z-Stack images were saved and click open.
A dialogue box prompting the input of the segmentation parameter will automatically appear. Set the threshold of image binarization to 0-1, the image binarization to 0.45, the upper threshold for a cell region to 200, the lower threshold for a cell region to 10, and the number of detectors to 32. A dialogue box requesting input for the number of excitation wavelengths will appear.
Input the number of wavelengths used for the image acquisition and click okay. A dialogue box that is requesting input for the excitation wavelengths will appear. Enter the excitation wavelengths in sequence from shortest to longest and click okay.
A new image window presenting a confocal reflection microscopy image will appear. Select an arbitrary background region to use for the background subtraction and draw a rectangle within the confocal reflection microscopy image. Double click within the selected region to confirm the selection and locate a new directory named, signature.
To perform dimensional reduction techniques, create an empty directory and name the directory, parent_directory. Store the fluorescent signatures of each of the two cell populations into two separate directories and open the command line interface of the workstation. Enter the command.
When select target directory is displayed, select the parent_directory. Then in the parent_directory folder, locate the PCA. png file, which will contain the resulting principle component analysis plot.
Here, a typical single cell fluorescent signature of a bacterial cell presented as a traditional spectrum plot and as a heat map is shown. In this figure, inaccurate 2D cell segmentation superimposed over the original confocal reflection microscopy image of a population of soil bacteria can be observed. With the resulting innate fluorescent signatures for the population presented as a heat map.
Note that the intra population variability was relatively minor following successful cell segmentation. Here, an example of inaccurate cell segmentation is shown super imposed onto the same population of P.putida as previously shown. The impact of the inaccurate cell segmentation on the innate fluorescent signatures of the population is readily apparent from the considerable number of outliers observed in the corresponding heat map.
Inaccurate cell segmentation results in a looser cluster after principle component analysis, compared to the type cluster obtained following accurate cell segmentation. Despite the minor variabilities of innate fluorescent signature observed within individual bacterial strains, each population forms a distinct cluster on the principle component analysis plot. In this figure, inaccurate 3D cell segmentation superimposed over the original confocal reflection microscopy image of a population of budding yeast Saccharomyces cerevisiae YM4271, can be observed.
Note the lack of outliers and the resulting innate fluorescent signatures for the population. It is essential to acquire as clean as possible, an image by confocal microscopy and to avoid a signal intensity saturation as noise can spoil accuracy of subsequent analysis. You can train machine learning models with the innate Western signature dataset for use in classification or prediction tasks, offering a tag-free population analysis and phenotypic prediction.
