Liang-da Chiu

Structured line illumination Raman microscopy

In the last couple of decades, the spatial resolution in optical microscopy has increased to unprecedented levels by exploiting the fluorescence properties of the probe. At about the same time, Raman imaging techniques have emerged as a way to image inherent chemical information in a sample without using fluorescent probes. However, in many applications, the achievable resolution is limited to about half the wavelength of excitation light. Here we report the use of structured illumination to increase the spatial resolution of label-free spontaneous Raman microscopy, generating highly detailed spatial contrast from the ensemble of molecular information in the sample. Using structured line illumination in slit-scanning Raman microscopy, we demonstrate a marked improvement in spatial resolution and show the applicability to a range of samples, including both biological and inorganic chemical component mapping. This technique is expected to contribute towards greater understanding of chemical component distributions in organic and inorganic materials.

Time-lapse Raman imaging of osteoblast differentiation

Osteoblastic mineralization occurs during the early stages of bone formation. During this mineralization, hydroxyapatite (HA), a major component of bone, is synthesized, generating hard tissue. Many of the mechanisms driving biomineralization remain unclear because the traditional biochemical assays used to investigate them are destructive techniques incompatible with viable cells. To determine the temporal changes in mineralization-related biomolecules at mineralization spots, we performed time-lapse Raman imaging of mouse osteoblasts at a subcellular resolution throughout the mineralization process. Raman imaging enabled us to analyze the dynamics of the related biomolecules at mineralization spots throughout the entire process of mineralization. Here, we stimulated KUSA-A1 cells to differentiate into osteoblasts and conducted time-lapse Raman imaging on them every 4 hours for 24 hours, beginning 5 days after the stimulation. The HA and cytochrome c Raman bands were used as markers for osteoblastic mineralization and apoptosis. From the Raman images successfully acquired throughout the mineralization process, we found that β-carotene acts as a biomarker that indicates the initiation of osteoblastic mineralization. A fluctuation of cytochrome c concentration, which indicates cell apoptosis, was also observed during mineralization. We expect time-lapse Raman imaging to help us to further elucidate osteoblastic mineralization mechanisms that have previously been unobservable.

Dual-polarization Raman spectral imaging to extract overlapping molecular fingerprints of living cells

Raman spectral imaging is gaining more and more attention in biological studies because of its label-free characteristic. However, the discrimination of overlapping chemical contrasts has been a major challenge. In this study, we introduce an optical method to simultaneously obtain two orthogonally polarized Raman images from a single scan of the sample. We demonstrate how this technique can improve the quality and quantity of the hyperspectral Raman dataset and how the technique is expected to further extend the horizons of Raman spectral imaging in biological studies by providing more detailed chemical information. The dual-polarization Raman images of a HeLa cell.

Visualizing the appearance and disappearance of the attractor of differentiation using Raman spectral imaging

Using Raman spectral imaging, we visualized the cell state transition during differentiation and constructed hypothetical potential landscapes for attractors of cellular states on a state space composed of parameters related to the shape of the Raman spectra. As models of differentiation, we used the myogenic C2C12 cell line and mouse embryonic stem cells. Raman spectral imaging can validate the amounts and locations of multiple cellular components that describe the cell state such as proteins, nucleic acids, and lipids; thus, it can report the state of a single cell. Herein, we visualized the cell state transition during differentiation using Raman spectral imaging of cell nuclei in combination with principal component analysis. During differentiation, cell populations with a seemingly homogeneous cell state before differentiation showed heterogeneity at the early stage of differentiation. At later differentiation stages, the cells returned to a homogeneous cell state that was different from the undifferentiated state. Thus, Raman spectral imaging enables us to illustrate the disappearance and reappearance of an attractor in a differentiation landscape, where cells stochastically fluctuate between states at the early stage of differentiation.

Rapid in vivo lipid/carbohydrate quantification of single microalgal cell by Raman spectral imaging to reveal salinity-induced starch-to-lipid shift

BackgroundLipid/carbohydrate content and ratio are extremely important when engineering algal cells for liquid biofuel production. However, conventional methods for such determination and quantification are not only destructive and tedious, but also energy consuming and environment unfriendly. In this study, we first demonstrate that Raman spectroscopy is a clean, fast, and accurate method to simultaneously quantify the lipid/carbohydrate content and ratio in living microalgal cells.ResultsThe quantification results of both lipids and carbohydrates obtained by Raman spectroscopy showed a linear correspondence with that obtained by conventional methods, indicating Raman can provide a similar accuracy to conventional methods, with a significantly shorter detection time. Furthermore, the subcellular resolution of Raman spectroscopy enabled not only the concentration mapping of lipid/carbohydrate content in single living cells, but also the evaluation of standard deviation between the biomass accumulation levels of individual algal cells.ConclusionsIn this study, we first demonstrate that Raman spectroscopy can be used for starch quantification in addition to lipid quantification in algal cells. Due to the easiness and non-destructive nature of Raman spectroscopy, it makes a perfect tool for the further study of starch–lipid shift mechanism.

Non-label immune cell state prediction using Raman spectroscopy

The acquired immune system, mainly composed of T and B lymphocytes, plays a key role in protecting the host from infection. It is important and technically challenging to identify cell types and their activation status in living and intact immune cells, without staining or killing the cells. Using Raman spectroscopy, we succeeded in discriminating between living T cells and B cells, and visualized the activation status of living T cells without labeling. Although the Raman spectra of T cells and B cells were similar, they could be distinguished by discriminant analysis of the principal components. Raman spectra of activated T cells with anti-CD3 and anti-CD28 antibodies largely differed compared to that of naïve T cells, enabling the prediction of T cell activation status at a single cell level. Our analysis revealed that the spectra of individual T cells gradually change from the pattern of naïve T cells to that of activated T cells during the first 24 h of activation, indicating that changes in Raman spectra reflect slow changes rather than rapid changes in cell state during activation. Our results indicate that the Raman spectrum enables the detection of dynamic changes in individual cell state scattered in a heterogeneous population.

Use of a white light supercontinuum laser for confocal interference-reflection microscopy

Shortly after its development, the white light supercontinuum laser was applied to confocal scanning microscopy as a more versatile substitute for the multiple monochromatic lasers normally used for the excitation of fluorescence. This light source is now available coupled to commercial confocal fluorescence microscopes. We have evaluated a supercontinuum laser as a source for a different purpose: confocal interferometric imaging of living cells and artificial models by interference reflection. We used light in the range 460–700 nm where this source provides a reasonably flat spectrum, and obtained images free from fringe artefacts caused by the longer coherence length of conventional lasers. We have also obtained images of cytoskeletal detail that is difficult to see with a monochromatic laser.

Polarised Raman Imaging of Living Cells for Chemical Contrast Manipulation

Raman spectral imaging has become a more and more popular technique in biological studies because it can extract chemical information from living cells in a label-free manner. One of the most challenging issues in the label-free Raman imaging of biological samples is to increase the molecular specificity in the spectra for better chemical contrast. Usually, the Raman spectrum from a cell is dominated by a few strong Raman bands such as the amide I band around 1650 cm-1, CH2 bend around 1445 cm-1 or the amide III band around 1300 cm-1 and it is not easy to get chemical contrast from other Raman bands that overlap with them. In this study, we aim to manipulate the chemical contrast in a living cell by exploiting the polarisation effects in Raman spectroscopy. By simply putting an analyser before the spectrometer, we can take the Raman image at the parallel and perpendicular polarisation against the incident light at the sample. The Raman spectra at the two orthogonal polarisations represent the Raman signals with different molecular orientation and symmetry of vibrations. Our experimental results demonstrate that at certain Raman shifts the two orthogonal polarisations indeed present different chemical contrasts. This indicates that polarized Raman imaging can help us visualise the different chemical contrasts that overlap at the same Raman shift and therefore increase the amount of chemical information we can get from cells.

Protein expression guided chemical profiling of living cells by the simultaneous observation of Raman scattering and anti-Stokes fluorescence emission

Our current understanding of molecular biology provides a clear picture of how the genome, transcriptome and proteome regulate each other, but how the chemical environment of the cell plays a role in cellular regulation remains much to be studied. Here we show an imaging method using hybrid fluorescence-Raman microscopy that measures the chemical micro-environment associated with protein expression patterns in a living cell. Simultaneous detection of fluorescence and Raman signals, realised by spectrally separating the two modes through the single photon anti-Stokes fluorescence emission of fluorescent proteins, enables the accurate correlation of the chemical fingerprint of a specimen to its physiological state. Subsequent experiments revealed the slight chemical differences that enabled the chemical profiling of mouse embryonic stem cells with and without Oct4 expression. Furthermore, using the fluorescent probe as localisation guide, we successfully analysed the detailed chemical content of cell nucleus and Golgi body. The technique can be further applied to a wide range of biomedical studies for the better understanding of chemical events during biological processes.

Cell type discrimination based on image features of molecular component distribution

Machine learning-based cell classifiers use cell images to automate cell-type discrimination, which is increasingly becoming beneficial in biological studies and biomedical applications. Brightfield or fluorescence images are generally employed as the classifier input variables. We propose to use Raman spectral images and a method to extract features from these spatial patterns and explore the value of this information for cell discrimination. Raman images provide information regarding distribution of chemical compounds of the considered biological entity. Since each spectral wavelength can be used to reconstruct the distribution of a given compound, spectral images provide multiple channels of information, each representing a different pattern, in contrast to brightfield and fluorescence images. Using a dataset of single living cells, we demonstrate that the spatial information can be ranked by a Fisher discriminant score, and that the top-ranked features can accurately classify cell types. This method is compared with the conventional Raman spectral analysis. We also propose to combine the information from whole spectral analyses and selected spatial features and show that this yields higher classification accuracy. This method provides the basis for a novel and systematic analysis of cell-type investigation using Raman spectral imaging, which may benefit several studies and biomedical applications.