Process Analytical Utility of Raman Microspectroscopy in the Directed Differentiation of Human Pancreatic Insulin-Positive Cells

Noninvasive and in situ identification of the phenotypes and differentiation stages of individual living cells entrapped within hydrogels

Microscale screening platforms that allow cells to interact in three dimensions (3D) with their microenviroment have been developed as a tool for identifying the extrinsic cues that might stimulate stem cells to replicate without differentiating within artificial cultures. Though these platforms reduce the number of valuable stem cells that must be used for screening, analyzing the fate decisions of cells in these platforms can be challenging. New noninvasive approaches for identifying the lineage-specific differentiation stages of cells while they are entrapped in the hydrogels used for these 3D cultures are especially needed. Here we used Raman spectra acquired from individual, living cells entrapped within a hydrogel matrix and multivariate analysis to identify cell phenotype noninvasively and in situ. We collected a single Raman spectrum from each cell of interest while it was entrapped within a hydrogel matrix and used partial least-squares discriminant analysis (PLS-DA) of the spectra for cell phenotype identification. We first demonstrate that this approach enables identifying the lineages of individual, living cells from different laboratory lines entrapped within two different hydrogels that are used for 3D culture, collagen and gelatin methacrylate (gelMA). Then we use a hematopoietic progenitor cell line that differentiates into different types of macrophages to show that the lineage-specific differentiation stages of individual, living hematopoietic cells entrapped inside of gelMA scaffolds may be identified by PLS-DA of Raman spectra. This ability to noninvasively identify the lineage-specific differentiation stages of cells without removing them from a 3D culture could enable tracking the differentiation of the same cell over time.

Demixing and Analysis of Complex Biological Raman Hyperspectra Based on Peak Fitting, Amplitude Trend Clustering, and Spectrum Reconstruction

To better interpret the Raman spectra from mammalian cells, it is often desirable to reduce their complexity by decomposing them into the spectral contributions from individual macromolecules or types of macromolecules. Diverse methods exist for demixing complex spectra, each with different benefits and drawbacks. However, some methods require a library of component spectra that might not be available, while others are hampered by noise and peak congestion that includes many proximal overlapping peaks. Through rapid fitting of individual peaks in every spectrum of a Raman hyperspectral data set, we have obtained individual peak parameters from which we determined the trends for all the peak amplitudes. We then grouped similar trends with k-means clustering. Then we used the peak parameters of all the peaks in a given cluster to reconstruct a spectrum representative of that cluster. This method produced spectra that were less distorted by unrelated overlapping peaks or noise, were less congested than those in the hyperspectral set, and thereby improved peak identification and macromolecule recognition. We have demonstrated the application of the method with Raman spectra from a perchlorate-polystyrene model system and extended it to complex spectra from methanol-fixed mammalian cells. We were able to recover independent spectra of perchlorate and polystyrene in the model system and spectra pertaining to individual macromolecular types (proteins, nucleic acids, lipids) from the mammalian cell data. We discuss how imperfections in spectral preprocessing and peak fitting can adversely affect the results. In summary, we have provided a proof-of-concept for a novel mixture resolution method with different attributes than extant ones.

Harnessing Raman spectroscopy for cell therapy bioprocessing

Cell therapy manufacturing requires precise monitoring of critical parameters to ensure product quality, consistency and to facilitate the implementation of cost-effective processes. While conventional analytical methods offer limited real-time insights, integration of process analytical technology tools such as Raman spectroscopy in bioprocessing has the potential to drive efficiency and reliability during the manufacture of cell-based therapies while meeting stringent regulatory requirements. The non-destructive nature of Raman spectroscopy, combined with its ability to be integrated on-line with scalable platforms, allows for continuous data acquisition, enabling real-time correlations between process parameters and critical quality attributes. Herein, we review the role of Raman spectroscopy in cell therapy bioprocessing and discuss how simultaneous measurement of distinct parameters and attributes, such as cell density, viability, metabolites and cell identity biomarkers can streamline on-line monitoring and facilitate adaptive process control. This, in turn, enhances productivity and mitigates process-related risks. We focus on recent advances integrating Raman spectroscopy across various manufacturing stages, from optimizing culture media feeds to monitoring bioprocess dynamics, covering downstream applications such as detection of co-isolated contaminating cells, cryopreservation, and quality control of the drug product. Finally, we discuss the potential of Raman spectroscopy to revolutionize current practices and accelerate the development of advanced therapy medicinal products.

Saline dry fixation for improved cell composition analysis using Raman spectroscopy

Raman spectroscopy enables the label-free assessment of cellular composition. While live cell analysis is the most accurate approach for cellular Raman spectroscopy, the analysis of fixed cells has proved to be very useful, particularly in collaborative projects where samples need to be serially examined by different laboratories or stored and reanalyzed at a later date. However, many chemicals that are widely used for cell fixation directly affect cellular biomolecules, yielding Raman spectra with missing or altered information. In this article, we compared the suitability of dry-fixation with saline versus chemical fixatives. We compared the Raman spectroscopy of saline dry-fixed cells with the more commonly used formaldehyde and methanol fixation and found that dry-fixed cell spectra preserved more cellular information than either chemical fixative. We also assessed the stability of dry-fixed cells over time and found that they were stable for at least 5 months. Finally, a comparison of dry-fixed and live cell spectra revealed effects due to the hydration state of the cells since they were recovered upon rehydrating dry-fixed samples. Thus, for fixed cell Raman spectroscopy, we recommend dry-fixation with unbuffered saline as a superior method to formaldehyde or methanol fixation.

Identification of the Differentiation Stages of Living Cells from the Six Most Immature Murine Hematopoietic Cell Populations by Multivariate Analysis of Single-Cell Raman Spectra

Efforts to expand hematopoietic stem and progenitor cells (HSPCs) in vitro are motivated by their use in the treatment of leukemias and other blood and immune system diseases. The combinations of extrinsic cues within the hematopoietic stem cell (HSC) niche that lead to HSC fate decisions remain unknown. New noninvasive and location-specific techniques are needed to enable identification of the differentiation stages of individual hematopoietic cells on biomaterial microarray screening platforms that minimize the usage of rare HSCs. Here, we show that a combination of Raman microspectroscopy and partial least-squares discriminant analysis (PLS-DA) enables the location-specific identification of individual living cells from the six most immature hematopoietic cell populations, HSC, multipotent progenitor (MPP)-1, MPP-2, MPP-3, common myeloid progenitor, and common lymphoid progenitor. Better than 90% accuracy was achieved. We show that the accuracy of this differentiation stage identification was based on spectral features associated with cell biochemistries. This work establishes that PLS-DA can capture the subtle spectral variations between as many as six closely related cell populations in the presence of potentially significant within-population spectral variation. This noninvasive approach can be used to screen HSC fate decisions elicited by extrinsic cues within biomaterial microarray screening platforms.

Spectral Markers for T Cell Death and Apoptosis-A Pilot Study on Cell Therapy Drug Product Characterization Using Raman Spectroscopy

Application of Raman spectroscopy as a T cell characterization tool supporting cell therapy drug product development has been evaluated. Statistically significant correlations between a set of Raman signals and established flow cytometry markers associated with apoptosis of T cells detected during drug product cryopreservation are presented in this study. Our study results demonstrate the potential of Raman spectroscopy for label-free measurements of T cell characteristics relevant to cell therapy product design and process control.

Augmented Two-Dimensional Correlation Spectroscopy for the Joint Analysis of Correlated Changes in Spectroscopic and Disparate Sources

Here, we present an augmented form of two-dimensional correlation spectroscopy, that integrates in a single format data from spectroscopic and multiple non-spectroscopic sources for analysis. The integration is affected by augmenting every spectrum in a hyperspectral data set with relevant non-spectroscopic data to permit two-dimensional correlation analysis^(2D-COS) of the ensemble of augmented spectra. A k-means clustering is then applied to the results of the perturbation domain decomposition to determine which Raman peaks cluster with any of the non-spectroscopic data. We introduce and explain the method with the aid of synthetic spectra and synthetic non-spectroscopic data. We then demonstrate this approach with data using Raman spectra from human embryonic stem cell aggregates undergoing directed differentiation toward pancreatic endocrine cells and parallel bioassays of hormone mRNA expression and C-peptide levels in spent medium. These pancreatic endocrine cells generally contain insulin or glucagon. Insulin has disulfide bonds that produce Raman scattering near 513 cm^-1, but no tryptophan. For insulin-positive cells, we found that the application of multisource correlation analysis revealed a high correlation between insulin mRNA and Raman scattering in the disulfide region. In contrast, glucagon has no disulfide bonds but does contain tryptophan. For glucagon-positive cells, we also observed a high correlation between glucagon mRNA and tryptophan Raman scattering (∼757 cm^-1). We conclude with a discussion of methods to enhance spectral resolution and its effects on the performance of multisource correlation analysis.

Exploring the maturation of a monocytic cell line using self-organizing maps of single-cell Raman spectra

Phorbol myristate acetate (PMA)-differentiated THP-1 cells are routinely used in lieu of primary macrophages to study macrophage polarization during host-pathogen interactions and disease progression. The phenotypes of THP-1 macrophages are influenced by the level and duration of PMA stimulation and possibly also by the presence of adhesion factors. Here, we use self-organizing maps (SOMs) of single-cell Raman spectra to probe the effects of PMA stimulation conditions and adhesion factors on THP-1 cell differentiation. Raman spectra encoding for biochemical composition were acquired from individual cells on substrates coated with fibronectin or poly-l-lysine before and after stimulation with 20 or 200 nM PMA for two different time intervals. SOMs constructed from these spectra showed the extent of spectral dissimilarity between different chronological cell populations. For all conditions, the SOMs indicated that the spectra acquired from cells after three-day treatment had diverged from those of untreated cells. The SOMs also showed that the higher PMA concentration produced both fully and partially differentiated cells for both adhesion factors after three days, whereas the outcome of stimulation for three days with the lower PMA concentration depended on the adhesion factor. On poly-l-lysine, treatment with 20 nM PMA for three days induced an intermediate stage of differentiation, but the same treatment produced partially and fully differentiated cells when applied to THP-1 cells on fibronectin. These results are consistent with the modulation of the transition of THP-1 monocytes into macrophage-like cells by integrin-binding interactions. Furthermore, differences in culture and stimulation conditions may confound comparison of results from separate studies.

A single-cell Raman-based platform to identify developmental stages of human pluripotent stem cell-derived neurons

We developed a label-free and noninvasive single-cell Raman microspectroscopy (SCRM)-based platform to identify neural cell lineages derived from clinically relevant human induced pluripotent stem cells (hiPSCs). Through large-scale Raman spectral analysis, we can distinguish hiPSCs and hiPSC-derived neural cells using their intrinsic biochemical profile. We identified glycogen as a Raman biomarker for neuronal differentiation and validated the results using conventional glycogen detection assays. The parameters obtained from SCRM were processed by a novel machine learning method based on t-distributed stochastic neighbor embedding (t-SNE)-enhanced ensemble stacking, enabling highly accurate and robust cell classification. The platform and the proposed biomarker should also be applicable to other cell types and can shed light on developmental biology and glycogen metabolism disorders.

Stem cells with the capability to self-renew and differentiate into multiple cell derivatives provide platforms for drug screening and promising treatment options for a wide variety of neural diseases. Nevertheless, clinical applications of stem cells have been hindered partly owing to a lack of standardized techniques to characterize cell molecular profiles noninvasively and comprehensively. Here, we demonstrate that a label-free and noninvasive single-cell Raman microspectroscopy (SCRM) platform was able to identify neural cell lineages derived from clinically relevant human induced pluripotent stem cells (hiPSCs). By analyzing the intrinsic biochemical profiles of single cells at a large scale (8,774 Raman spectra in total), iPSCs and iPSC-derived neural cells can be distinguished by their intrinsic phenotypic Raman spectra. We identified a Raman biomarker from glycogen to distinguish iPSCs from their neural derivatives, and the result was verified by the conventional glycogen detection assays. Further analysis with a machine learning classification model, utilizing t-distributed stochastic neighbor embedding (t-SNE)-enhanced ensemble stacking, clearly categorized hiPSCs in different developmental stages with 97.5% accuracy. The present study demonstrates the capability of the SCRM-based platform to monitor cell development using high content screening with a noninvasive and label-free approach. This platform as well as our identified biomarker could be extensible to other cell types and can potentially have a high impact on neural stem cell therapy.

Applications of Raman spectroscopy in the development of cell therapies: state of the art and future perspectives

This comprehensive review article discusses current and future perspectives of Raman spectroscopy-based analyses of cell therapy processes and products.

Raman and infrared spectroscopy reveal that proliferating and quiescent human fibroblast cells age by biochemically similar but not identical processes

Dermal fibroblast cells can adopt different cell states such as proliferation, quiescence, apoptosis or senescence, in order to ensure tissue homeostasis. Proliferating (dividing) cells pass through the phases of the cell cycle, while quiescent and senescent cells exist in a non-proliferating cell cycle-arrested state. However, the reversible quiescence state is in contrast to the irreversible senescence state. Long-term quiescent cells transit into senescence indicating that cells age also when not passing through the cell cycle. Here, by label-free in vitro vibrational spectroscopy, we studied the biomolecular composition of quiescent dermal fibroblast cells and compared them with those of proliferating and senescent cells. Spectra were examined by multivariate statistical analysis using a PLS-LDA classification model, revealing differences in the biomolecular composition between the cell states mainly associated with protein alterations (variations in the side chain residues of amino acids and protein secondary structure), but also within nucleic acids and lipids. We observed spectral changes in quiescent compared to proliferating cells, which increased with quiescence cultivation time. Raman and infrared spectroscopy, which yield complementary biochemical information, clearly distinguished contact-inhibited from serum-starved quiescent cells. Furthermore, the spectra displayed spectral differences between quiescent cells and proliferating cells, which had recovered from quiescence. This became more distinct with increasing quiescence cultivation time. When comparing proliferating, (in particular long-term) quiescent and senescent cells, we found that Raman as well as infrared spectroscopy can separate these three cellular states from each other due to differences in their biomolecular composition. Our spectroscopic analysis shows that proliferating and quiescent fibroblast cells age by similar but biochemically not identical processes. Despite their aging induced changes, over long time periods quiescent cells can return into the cell cycle. Finally however, the cell cycle arrest becomes irreversible indicating senescence.

Developing Fully Automated Quality Control Methods for Preprocessing Raman Spectra of Biomedical and Biological Samples

Spectral preprocessing is frequently required to render Raman spectra useful for further processing and analyses. The various preprocessing steps, individually and sequentially, are increasingly being automated to cope with large volumes of data from, for example, hyperspectral imaging studies. Full automation of preprocessing is especially desirable when it produces consistent results and requires minimal user input. It is therefore essential to evaluate the "quality" of such preprocessed spectra. However, relatively few methods exist to evaluate preprocessing quality, and fully automated methods for doing so are virtually non-existent. Here we provide a brief overview of fully automated spectral preprocessing and fully automated quality assessment of preprocessed spectra. We follow this with the introduction of fully automated methods to establish figures-of-merit that encapsulate preprocessing quality. By way of illustration, these quantitative methods are applied to simulated and real Raman spectra. Quality factor and quality parameter figures-of-merit resulting from individual preprocessing step quality tests, as well as overall figures-of-merit, were found to be consistent with the quality of preprocessed spectra.

Empirical Factors Affecting the Quality of Non-Negative Matrix Factorization of Mammalian Cell Raman Spectra

Mammalian cells contain various macromolecules that can be investigated non-invasively with Raman spectroscopy. The particular mixture of major macromolecules present in a cell being probed are reflected in the measured Raman spectra. Determining macromolecular identities and estimating their concentrations from these mixture Raman spectra can distinguish cell types and otherwise enable biological research. However, the application of canonical multivariate methods, such as principal component analysis (PCA), to perform spectral unmixing yields mathematical solutions that can be difficult to interpret. Non-negative matrix factorization (NNMF) improves the interpretability of unmixed macromolecular components, but can be difficult to apply because ambiguities produced by overlapping Raman bands permit multiple solutions. Furthermore, theoretically sound methods can be difficult to implement in practice. Here we examined the effects of a number of empirical approaches on the quality of NNMF results. These approaches were evaluated on simulated mammalian cell Raman hyperspectra and the results were used to develop an enhanced procedure for implementing NNMF. We demonstrated the utility of this procedure using a Raman hyperspectral data set measured from human islet cells to recover the spectra of insulin and glucagon. This was compared to the relatively inferior PCA of these data.

Physico-electrochemical Characterization of Pluripotent Stem Cells during Self-Renewal or Differentiation by a Multi-modal Monitoring System

Monitoring pluripotent stem cell behaviors (self-renewal and differentiation to specific lineages/phenotypes) is critical for a fundamental understanding of stem cell biology and their translational applications. In this study, a multi-modal stem cell monitoring system was developed to quantitatively characterize physico-electrochemical changes of the cells in real time, in relation to cellular activities during self-renewal or lineage-specific differentiation, in a non-destructive, label-free manner. The system was validated by measuring physical (mass) and electrochemical (impedance) changes in human induced pluripotent stem cells undergoing self-renewal, or subjected to mesendodermal or ectodermal differentiation, and correlating them to morphological (size, shape) and biochemical changes (gene/protein expression). An equivalent circuit model was used to further dissect the electrochemical (resistive and capacitive) contributions of distinctive cellular features. Overall, the combination of the physico-electrochemical measurements and electrical circuit modeling collectively offers a means to longitudinally quantify the states of stem cell self-renewal and differentiation.

The quest to make fully functional human pancreatic beta cells from embryonic stem cells: climbing a mountain in the clouds

The production of fully functional insulin-secreting cells to treat diabetes is a major goal of regenerative medicine. In this article, I review progress towards this goal over the last 15 years from the perspective of a beta cell biologist. I describe the current state-of-the-art, and speculate on the general approaches that will be required to identify and achieve our ultimate goal of producing functional beta cells. The need for deeper phenotyping of heterogeneous cultures of stem cell derived islet-like cells in parallel with a better understanding of the heterogeneity of the target cell type(s) is emphasised. This deep phenotyping should include high-throughput single-cell analysis, as well as comprehensive 'omics technologies to provide unbiased characterisation of cell products and human beta cells. There are justified calls for more detailed and well-powered studies of primary human pancreatic beta cell physiology, and I propose online databases of standardised human beta cell responses to physiological stimuli, including both functional and metabolomic/proteomic/transcriptomic profiles. With a concerted, community-wide effort, including both basic and applied scientists, beta cell replacement will become a clinical reality for patients with diabetes.

Identifying States along the Hematopoietic Stem Cell Differentiation Hierarchy with Single Cell Specificity via Raman Spectroscopy

A major challenge for expanding specific types of hematopoietic cells ex vivo for the treatment of blood cell pathologies is identifying the combinations of cellular and matrix cues that direct hematopoietic stem cells (HSC) to self-renew or differentiate into cell populations ex vivo. Microscale screening platforms enable minimizing the number of rare HSCs required to screen the effects of numerous cues on HSC fate decisions. These platforms create a strong demand for label-free methods that accurately identify the fate decisions of individual hematopoietic cells at specific locations on the platform. We demonstrate the capacity to identify discrete cells along the HSC differentiation hierarchy via multivariate analysis of Raman spectra. Notably, cell state identification is accurate for individual cells and independent of the biophysical properties of the functionalized polyacrylamide gels upon which these cells are cultured. We report partial least-squares discriminant analysis (PLS-DA) models of single cell Raman spectra enable identifying four dissimilar hematopoietic cell populations across the HSC lineage specification. Successful discrimination was obtained for a population enriched for long-term repopulating HSCs (LT-HSCs) versus their more differentiated progeny, including closely related short-term repopulating HSCs (ST-HSCs) and fully differentiated lymphoid (B cells) and myeloid (granulocytes) cells. The lineage-specific differentiation states of cells from these four subpopulations were accurately identified independent of the stiffness of the underlying biomaterial substrate, indicating subtle spectral variations that discriminated these populations were not masked by features from the culture substrate. This approach enables identifying the lineage-specific differentiation stages of hematopoietic cells on biomaterial substrates of differing composition and may facilitate correlating hematopoietic cell fate decisions with the extrinsic cues that elicited them.