Automated and unbiased methods of non-invasive cell monitoring able to deal with complex biological heterogeneity are fundamentally important for biology and medicine. Label-free cell imaging provides information about endogenous autofluorescent metabolites, enzymes and cofactors in cells. However extracting high content information from autofluorescence imaging has been hitherto impossible. Here, we quantitatively characterise cell populations in different tissue types, live or fixed, by using novel image processing and a simple multispectral upgrade of a wide-field fluorescence microscope. Our optimal discrimination approach enables statistical hypothesis testing and intuitive visualisations where previously undetectable differences become clearly apparent. Label-free classifications are validated by the analysis of Classification Determinant (CD) antigen expression. The versatility of our method is illustrated by detecting genetic mutations in cancer, non-invasive monitoring of CD90 expression, label-free tracking of stem cell differentiation, identifying stem cell subpopulations with varying functional characteristics, tissue diagnostics in diabetes and assessing the condition of preimplantation embryos.

Carbonyl cyanide-p-trifluoro methoxyphenylhydrazone (FCCP) is a well-known mitochondrial uncoupling agent. We examined FCCP-induced fluorescence quenching of reduced nicotinamide adenine dinucleotide / nicotinamide adenine dinucleotide phosphate (NAD(P)H) in solution and in cultured HeLa cells in a wide range of FCCP concentrations from 50 to 1000µM. A non-invasive label-free method of hyperspectral imaging of cell autofluorescence combined with unsupervised unmixing was used to separately isolate the emissions of free and bound NAD(P)H from cell autofluorescence. Hyperspectral image analysis of FCCP-treated HeLa cells confirms that this agent selectively quenches fluorescence of free and bound NAD(P)H in a broad range of concentrations. This is confirmed by the measurements of average NAD/NADH and NADP/NADPH content in cells. FCCP quenching of free NAD(P)H in cells and in solution is found to be similar, but quenching of bound NAD(P)H in cells is attenuated compared to solution quenching possibly due to a contribution from the metabolic and/or antioxidant response in cells. Chemical quenching of NAD(P)H fluorescence by FCCP validates the results of unsupervised unmixing of cell autofluorescence.

Can label-free, non-invasive optical imaging by hyperspectral autofluorescence microscopy discern between euploid and aneuploid cells within the inner cell mass (ICM) of the mouse preimplantation embryo? Hyperspectral autofluorescence microscopy enables discrimination between euploid and aneuploid ICM in mouse embryos. Euploid/aneuploid mosaicism affects up to 17.3% of human blastocyst embryos with trophectoderm biopsy or spent media currently utilized to diagnose aneuploidy and mosaicism in clinical in vitro fertilization. Based on their design, these approaches will fail to diagnose the presence or proportion of aneuploid cells within the foetal lineage ICM of some blastocyst embryos. The impact of aneuploidy on cellular autofluorescence and metabolism of primary human fibroblast cells and mouse embryos was assessed using a fluorescence microscope adapted for imaging with multiple spectral channels (hyperspectral imaging). Primary human fibroblast cells with known ploidy were subjected to hyperspectral imaging to record native cell fluorescence (4–6 independent replicates, euploid n = 467; aneuploid n = 969). For mouse embryos, blastomeres from the eight-cell stage (five independent replicates: control n = 39; reversine n = 44) and chimeric blastocysts (eight independent replicates: control n = 34; reversine n = 34; 1:1 (control:reversine) n = 30 and 1:3 (control:reversine) n = 37) were utilized for hyperspectral imaging. The ICM from control and reversine-treated embryos were mechanically dissected and their karyotype confirmed by whole genome sequencing (n = 13 euploid and n = 9 aneuploid). Two models were employed: (i) primary human fibroblasts with known karyotype and (ii) a mouse model of embryo aneuploidy where mouse embryos were treated with reversine, a reversible spindle assembly checkpoint inhibitor, during the four- to eight-cell division. Individual blastomeres were dissociated from control and reversine-treated eight-cell embryos and either imaged directly or used to generate chimeric blastocysts with differing ratios of control:reversine-treated cells. Individual blastomeres and embryos were interrogated by hyperspectral imaging. Changes in cellular metabolism were determined by quantification of metabolic co-factors (inferred from their autofluorescence signature): NAD(P)H and flavins with the subsequent calculation of the optical redox ratio (ORR: flavins/[NAD(P)H + flavins]). Autofluorescence signals obtained from hyperspectral imaging were examined mathematically to extract features from each cell/blastomere/ICM. This was used to discriminate between different cell populations. An increase in the relative abundance of NAD(P)H and decrease in flavins led to a significant reduction in the ORR for aneuploid cells in primary human fibroblasts and reversine-treated mouse blastomeres (P < 0.05). Mathematical analysis of endogenous cell autofluorescence achieved separation between (i) euploid and aneuploid primary human fibroblast cells, (ii) control and reversine-treated mouse blastomeres cells, (iii) control and reversine-treated chimeric blastocysts, (iv) 1:1 and 1:3 chimeric blastocysts and (v) confirmed euploid and aneuploid ICM from mouse blastocysts. The accuracy of these separations was supported by receiver operating characteristic curves with areas under the curve of 0.97, 0.99, 0.87, 0.88 and 0.93, respectively. We believe that the role of chance is low as mathematical features separated euploid from aneuploid in both human fibroblasts and ICM of mouse blastocysts. N/A. Although we were able to discriminate between euploid and aneuploid ICM in mouse blastocysts, confirmation of this approach in human embryos is required. While we show this approach is safe in mouse, further validation is required in large animal species prior to implementation in a clinical setting. We have developed an original, accurate and non-invasive optical approach to assess aneuploidy within the ICM of mouse embryos in the absence of fluorescent tags. Hyperspectral autofluorescence imaging was able to discriminate between euploid and aneuploid human fibroblast and mouse blastocysts (ICM). This approach may potentially lead to a new diagnostic for embryo analysis. K.R.D. is supported by a Mid-Career Fellowship from the Hospital Research Foundation (C-MCF-58-2019). This study was funded by the Australian Research Council Centre of Excellence for Nanoscale Biophotonics (CE140100003) and the National Health and Medical Research Council (APP2003786). The authors declare that there is no conflict of interest.

Optimally preserved urinary exfoliated renal proximal tubule cells were assessed by multispectral imaging of cell autofluorescence. We demonstrated different multispectral autofluorescence signals in such cells extracted from the urine of patients with healthy or diseased kidneys. Using up to 10 features, we were able to differentiate cells from individuals with heathy kidneys and impaired renal function (indicated by estimated glomerular filtration rate (eGFR) values) with the receiver operating characteristic area under the curve (AUC) of 0.99. Using the same method, we were also able to discriminate such urine cells from patients with and without renal fibrosis on biopsy, where significant differences in multispectral autofluorescence signals (AUC = 0.90) were demonstrated between healthy and diseased patients (p < 0.05). These findings show that multispectral assessment of the cell autofluorescence in urine exfoliated proximal tubule kidney cells has the potential to be developed as a sensitive, non-invasive diagnostic method for CKD.

Bioimaging using endogenous cell fluorescence, without any external biomarkers makes it possible to explore cells and tissues in their original native state, alsoin vivo. In order to be informative, this label-free method requires careful multispectral or hyperspectral recording of autofluorescence images followed by unsupervised extraction (unmixing) of biochemical signatures. The unmixing is difficult due to the scarcity of biochemically pure regions in cells and also because autofluorescence is weak compared with signals from labelled cells, typically leading to low signal to noise ratio. Here, we solve the problem of unsupervised hyperspectral unmixing of cellular autofluorescence by introducing the Robust Dependent Component Analysis (RoDECA). This approach provides sophisticated and statistically robust quantitative biochemical analysis of cellular autofluorescence images. We validate our method on artificial images, where the addition of varying known level of noise has allowed us to quantify the accuracy of our RoDECA analysis in a way that can be applied to real biological datasets. The same unsupervised statistical minimisation is then applied to imaging of mouse retinal photoreceptor cells where we establish the identity of key endogenous fluorophores (free NADH, FAD and lipofuscin) and derive the corresponding molecular abundance maps. The pre-processing methodology of image datasets is also presented, which is essential for the spectral unmixing analysis, but mostly overlooked in the previous studies.

Damage and degradation of articular cartilage leads to severe pain and loss of mobility. The development of new therapies for cartilage regeneration for monitoring their effect requires further study of cartilage, ideally at a molecular level and in a minimally invasive way. Hyperspectral microscopy is a novel technology which utilises endogenous fluorophores to non-invasively assess the molecular composition of cells and tissue. In this study, we applied hyperspectral microscopy to healthy bovine articular cartilage and osteoarthritic human articular cartilage to investigate its capacity to generate informative molecular data and characterise disease state and treatment effects. We successfully demonstrated label-free fluorescence identification of collagen type I and II – isolated in cartilage here for the first time and the co-enzymes free NADH and FAD which together give the optical redox ratio that is an important measure of metabolic activity. The intracellular composition of chondrocytes was also examined. Differences were observed in the molecular ratios within the superficial and transitional zones of the articular cartilage which appeared to be influenced by disease state and treatment. These findings show that hyperspectral microscopy could be useful for investigating the molecular underpinnings of articular cartilage degradation and repair. As it is non-invasive and non-destructive, samples can be repeatedly assessed over time, enabling true time-course experiments with in-depth molecular data. Additionally, there is potential for the hyperspectral approach to be adapted for patient examination to allow the investigation of cartilage state. This could be of advantage for assessment and diagnosis as well as treatment monitoring.