DUV Autofluorescence for Cell Biology and Tissue Histology Biology of the Cell Biology of the Cell Frederic Jamme, Slávka Kaščáková, Sandrine Villette, Fatma Allouche, Stéphane Pallu, Valerie Rouam, Matthieu Refregiers

To cite this version:

Frederic Jamme, Slávka Kaščáková, Sandrine Villette, Fatma Allouche, Stéphane Pallu, et al.. DUV Autofluorescence Microscopy for Cell Biology and Tissue Histology Biology of the Cell Biology ofthe Cell. Biology of the Cell, Wiley, 2013, 105 (7), pp.277-288. ￿10.1111/boc.201200075￿. ￿hal-01485578￿

HAL Id: hal-01485578 https://hal.archives-ouvertes.fr/hal-01485578 Submitted on 9 Mar 2017

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Biology of the Cell

DUV Autofluorescence Microscopy for Cell Biology and Tissue Histology

Journal:For Biology Peer of the Cell Review Manuscript ID: Draft

Wiley - Manuscript type: Research Article

Date Submitted by the Author: n/a

Complete List of Authors: Jamme, Frédéric; Synchrotron SOLEIL, DISCO Beamline; INRA, UAR 1008 CEPIA Kascakova, Slavka; Synchrotron SOLEIL, DISCO Beamline Villette, Sandrine; CNRS, CBM UPR4301 Allouche, Fatma; INRA, UAR 1008 CEPIA Pallu, Stéphane; Université d'Orléans, Rouam, Valérie; Synchrotron SOLEIL, DISCO Beamline Refregiers, Matthieu; Synchrotron SOLEIL, DISCO Beamline

Keywords: Cellular imaging, Plants, Oncology/cancer, Metabolism

Biology of the Cell Page 1 of 37 Biology of the Cell

1 2 3 DUV Autofluorescence Microscopy for Cell Biology and 4 5 6 Tissue Histology 7 8 9 1, 2, * 1 3 2 10 Frédéric Jamme , Slavka Kascakova , Sandrine Villette , Fatma Allouche , 11 4 1 1 12 Stéphane Pallu , Valérie Rouam , Matthieu Réfrégiers 13 14 15 16 17 18 For Peer Review 19 20 21 1- Synchrotron SOLEIL, L’Orme des Merisiers, Gif sur Yvette, France. 22 23 24 2- INRA, UAR 1008 CEPIA, Rue de la Géraudière, F-44316 Nantes, France. 25 26 27 3- Centre de Biophysique Moléculaire, CNRS UPR4301, Rue Charles Sadron, 45071 28 29 Orléans cedex 2, France. 30 31 32 4- INSERM U-658, Hôpital Porte Madeleine, BP 2439, 45032 Orléans cedex 01, 33 34 35 France. 36 37 38 Corresponding author : Frédéric Jamme 39 40 41 phone : +33-(0)1-69 35 9686 42 43 44 fax : +33-(0)1-69 35 9456 45 46 47 E-mail : [email protected] 48 49 50 Keywords: Autofluorescence, UV, DUV, synchrotron, microspectroscopy 51 52 53 54 55 56 57 58 59 60 Biology of the Cell Biology of the Cell Page 2 of 37

1 2 3 Abbreviations: VUV, vacuum ; DUV, deep ultraviolet; HeLa, Human cervix 4 5 epithelial cell line; FAD, Flavin adenine dinucleotide; FPs, Fluorescent proteins; BPA, 6 7 Bisphenol A; PVC, Polyvinyl chloride. 8 9 10 11 12 13 Acknowledgement 14 15 16 Part of this work was supported by the Région Centre (FRANCE). Synchrotron 17 18 SOLEIL support throughFor projects Peer #20100064, Review 201000181, 20100949 and 20110131 19 20 is acknowledged. 21 22 23 24 25 26 27 Graphical Abstract 28 29 30 Ultraviolet induced autofluorescence of cells is often, if not ever, considered as a 31 32 parasitic signal that must be minimized. Using a tuneable ultraviolet source, it is 33 34 35 possible to exploit the autofluorescence signal for obtaining valuable information on 36 37 selected cells either isolated or in tissular context. 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell 2 Page 3 of 37 Biology of the Cell

1 2 3 Abstract 4 5 6 7 8 Autofluorescence is a powerful tool for molecular histology and follow up of 9 10 metabolic processes in mammal and vegetal samples. However, at the microscopic 11 12 13 scale, it is limited to visible and near infrared excitation of the samples. 14 15 16 Because several interesting naturally occurring fluorochromes absorb in the UV and 17 18 DUV, we have For developed Peer a synchrotron-coupled Review DUV microspectrofluorimeter. 19 20 Available since 2010, this new DUV microscope gives promising results for the study 21 22 of representative samples, cultured cells, bone sections and maize stems. 23 24 25 An extended selection of natural occurring autofluorescent probes is presented with 26 27 all their spectral characteristics; their distribution in various biological samples is also 28 29 30 shown. We demonstrate that DUV autofluorescence is a powerful tool for tissue 31 32 histology and cell biology. 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell 3 Biology of the Cell Page 4 of 37

1 2 3 Introduction 4 5 6 The ultraviolet autofluorescence of cells and tissues has long been evaluated for its 7 8 diagnosis potential. However, since sources and microscopes that allow obtaining 9 10 reliable information in the DUV (bellow 350 nm) are difficult to obtain and couple for 11 12 medical diagnosis imaging of tissues, most of the medical diagnosis concentrated on 13 14 15 the autofluorescence spectroscopy and microscopy of endogenous probes such as 16 17 porphyrins which have emission in the visible range. 18 For Peer Review 19 20 Synchrotron radiation is a broadband light that can be monochromatized in almost 21 22 any energy range. The DISCO beamline at the synchrotron SOLEIL optimizes the 23 24 vacuum ultraviolet (VUV) to the visible range of the spectrum (Giuliani et al., 2009). 25 26 The deep ultraviolet (DUV) energies accessible on two microscopes at atmospheric 27 28 29 pressure are defined by quartz cut-off, starting from 200 nm to longer wavelengths 30 31 (600 nm) (Jamme et al., 2010; Tawil et al., 2011). For biological relevance, we 32 33 specifically focus on the range between 200 and 350 nm that is very rarely covered 34 35 especially under 250 nm. 36 37 38 The deep ultraviolet range is of particular interest for studying biomacromolecules 39 40 41 due to their absorption in this energy range. While spectrometers can often go down 42 43 to this range to study isolated biomacromolecules, microscopes are usually very 44 45 limited in this range with either 350 nm cut-off or access to only a limited number of 46 47 excitation wavelengths. 48 49 50 Köhler developed deep ultraviolet microscopy in 1904 as a bright field transmission 51 52 53 technique (Kohler, 1904). This was implemented for one main reason: to improve the 54 55 spatial resolution of cellular imaging. It passed through the century allowing seeing 56 57 the unseen, namely nucleic acids in cells (Mellors et al., 1950), and to observe colour 58 59 60 Biology of the Cell 4 Page 5 of 37 Biology of the Cell

1 2 3 differences in tumours under UV excitation for cancer diagnosis (Ludford et al., 4 5 1948). Thanks to developments in detectors, it became a quantitative method to 6 7 observe protein and DNA contrast at 280 and 260 nm (Zeskind et al., 2007). 8 9 10 However, due to differences of optical density in deep UV transmission images, 11 12 scattering errors are difficult to avoid and interpreting DUV images are more 13 14 challenging than spectral imaging using our confocal DUV microscope. 15 16 17 Nonlinear microscopy is an alternative approach for live tissue visualization of 18 For Peer Review 19 autofluorescent compounds (Zipfel et al., 2003; Diaspro et al., 2005). The infrared 20 21 photons exciting the sample are an asset, allowing deep tissue penetration in the 22 23 24 optical window of living tissue. However this methodology present two inconvenient 25 26 as compared to DUV monophotonic excitation; being a non-linear process relying 27 28 onto infrared photons, its resolution remains in the µm range, second, the two photon 29 30 action spectrum are very broad and do not permit as fine selectivity as DUV 31 32 33 microscopy. 34 35 36 While most biomolecules do present a contrast in DUV transmission microscopy, few 37 38 of them will reemit fluorescence. This autofluorescence is very important in a sense 39 40 that it permits better discrimination of molecules. Moreover, following 41 42 autofluorescence opens label free studies of molecules of interest without any 43 44 45 external probes or radiolabelling that could impair activity of the molecule of interest 46 47 (Tawil et al., 2011; Batard et al., 2011). 48 49 50 While many studies were conducted in DUV fluorescence spectroscopy as a 51 52 diagnosis tool, mainly for cancer [(Wagnieres et al., 1998; Pavlova et al., 2008), very 53 54 few were conducted on DUV fluorescence microscopy or microspectrofluorimetry on 55 56 biopsies and tissues. Most of them were using excitation wavelength longer than 350 57 58 59 60 Biology of the Cell 5 Biology of the Cell Page 6 of 37

1 2 3 nm, mostly due to the lack of laser lines below this value, despite very promising 4 5 sensibility and sensitivity using DUV excitation (Ramanujam et al., 1994). Indeed, 6 7 many endogenous have an absorption / excitability maximum in DUV 8 9 10 below 350 nm, especially (Wagnieres et al., 1998). Signal attributed to 11 12 tryptophan typically exhibits fluorescence intensities orders of magnitude greater than 13 14 those from other endogenous fluorophores. And although its fluorescence can serve 15 16 as an additional marker for monitoring cellular status (Ramanujam, 2000) due to lack 17 18 of DUV fluorescenceFor microscopy, Peer this Review is often ignored. 19 20 21 In addition, considering the recent work by Cox et al, no specific labels are needed 22 23 24 for super resolution microscopy techniques (Cox et al., 2011). Therefore, 25 26 aufluorescent (free labeling) molecules blinking and bleaching could be use to 27 28 improve resolution. 29 30 31 In this study, we first present the main autofluorescent biomolecules spectra detected 32 33 34 in cells or tissues. Then, we demonstrate the usefulness of DUV microscopy for in 35 36 situ studies of cell biology and tissue histology. We choose the most occurring 37 38 natural fluorochromes and representative samples of cells either isolated or in 39 40 various tissular surrounding, namely cultured cancerous cells, osteocytes in a bone 41 42 context, vegetal cells as appropriate illustrative examples. 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell 6 Page 7 of 37 Biology of the Cell

1 2 3 Results 4 5 6 Isolated auto fluorescent compounds 7 8 When measuring endogenous signals, no additional dyes or contrast agents are 9 10 11 necessary to determine native characteristics of cells or tissues. However, when 12 13 considering an autofluorescence signal from biologic material, all the contributing 14 15 fluorophores and chromophores must be accounted for (due to excitation, emission 16 17 and absorption spectra overlapping). To identify the major biological 18 For Peer Review 19 20 autofluorescence compounds in DUV, we selected and recorded their fluorescence 21 22 emission and excitation spectra (Figures 1-4). Most of them have a maximum 23 24 excitation peak below 350 nm. One can notice that each of them have distinct 25 26 spectrum (maximum peak). However, several chromophores, such as , 27 28 and , do present very close emission spectra that become difficult to 29 30 31 discriminate without spectral detection. In those cases, fine-tuning of the excitation 32 33 (Figure 1) may permit better separation of the species by altering their relative 34 35 intensities. 36 37 38 The fluorophores were measured in PBS at pH = 7 (Figure 1), which does not 39 40 represent a realistic in vitro and/or in vivo situation. Indeed, the spectroscopic 41 42 43 characteristics of fluorophores can change due to environment-related effects and 44 45 spectral shifts may occur. A bathochromic shift of absorption maxima of bilirubin is 46 47 observed when bound to HSA, compared to free (and unbound) bilirubin in PBS 48 49 (Kanick et al., 2010). Also coenzymes such as NAD(P)H reveals distinct 50 51 52 spectroscopic characteristics, as shown on Figure 2. Free NADH in PBS has an 53 54 emission maximum located at 460 nm, while, a hypsochromic shift of 10 nm in the 55 56 position of its fluorescence maximum is observed when it is bound to alcohol 57 58 59 60 Biology of the Cell 7 Biology of the Cell Page 8 of 37

1 2 3 dehydrogenase. The knowledge of those spectral differences of NAD(P)H allowed 4 5 previously to envisage the monitoring and quantification of total NAD(P)H 6 7 concentration (free and bound NAD(P)H), which is an important parameter for the 8 9 10 interpretation of cellular metabolism (Paul and Schneckenburger, 1996; Villette et al., 11 12 2006). However, the example of bilirubin or NAD(P)H shows also the difficulty of the 13 14 identification of spectral components within the global autofluorescence signal of cell 15 16 or tissue. It becomes obvious that the detailed knowledge of the influence of factors 17 18 such as pH, binding,For hydrophilicity/lipophilicity Peer Review on the spectral properties of 19 20 21 fluorophores is necessary to identify the chemical composition of biological material 22 23 under investigation. 24 25 26 Grass lignocelluloses are a major resource in the emerging cellulose to ethanol 27 28 strategy for biofuels (Anderson and Akin, 2008). However, the potential 29 30 bioconversion of carbohydrates is limited by the associated aromatic constituents 31 32 33 within the grass fibre. These aromatics include both lignin and low-molecular weight 34 35 phenolic acids. The two main ester linked phenolics (ferulic and para-coumaric acids) 36 37 can be differentiated by DUV absorption maximums near 326 nm and 314 nm, 38 39 respectively. As shown in Figure 4, ferulic acid and para-coumaric acid have an 40 41 emission peak centred at 415 and 400 nm, respectively while lignin shows a 42 43 44 maximum of autofluorescence around 475 nm. Ferulic acid and para-coumaric acid 45 46 can hardly be distinguished from each other using conventional fluorescence 47 48 microscope, infrared or Raman spectroscopies (Saadi et al., 1998; Robert et al., 49 50 2010). 51 52

53 54 Synchrotron flux profile and microscope spectral response 55 56 57 58 59 60 Biology of the Cell 8 Page 9 of 37 Biology of the Cell

1 2 3 The flux was measured (Ph/s) at the entrance of the microscope using a calibrated 4 5 photodiode (AXUV 100, IRD, CA, USA) coupled to a current amplifier (DLPCA-200, 6 7 laser Components, UK). As described previously (Jamme et al., 2010), the excitation 8 9 10 light is provided by a bending magnet emission on the DISCO beamline at 11 12 synchrotron SOLEIL (Giuliani et al., 2009). White beam from 180 to 600 nm is 13 14 monochromatized by an iHR320 (Jobin-Yvon, FR) equipped with a 100 grooves per 15 16 millimetre grating with a 240 nm peak efficiency (Spectrum Scientific, Inc, Irvine, CA, 17 18 USA). Thus the fluxFor variation Peer observed, shown Review in Figure 5, is mainly attributed to the 19 20 21 excitation monochromator grating response convoluted by the transmission view port 22 23 (UV fused silica, MPF, SC, USA). 24 25 26 The spectral response of the microscope was measured using a calibrated lamp 27 28 (QTH 45W, Microcontrole, FR) through the 40X objective (Ultrafluar, Zeiss, D) and 29 30 calculating the ratio between the detected signal and the known emittance of the 31 32 33 lamp (Figure 5). The response variation observed in Figure 5 is mostly attributed to 34 35 the emission spectrometer gratings (T64000, Horiba Jobin-Yvon, FR). 36 37 38 Applications to cell and tissues 39 40 41 Isolated living cells 42 43 44 In order to assess the distribution of endogenous fluorophore in single cell, the whole 45 46 living cell was scanned under 275 nm excitation with 1s exposure time and the 47 48 fluorescence emission spectra of the cell were recorded at each pixel. The Figure 6 49 50 51 represents the spectra of autofluorescence obtained from the highlighted pixels. As it 52 53 can be seen from Figure 6, in a simplified model, the emission spectrum of a living 54 55 cell is mainly tryptophan and fluorescence arising from the proteins. In a 56 57 58 59 60 Biology of the Cell 9 Biology of the Cell Page 10 of 37

1 2 3 cellular context, due to environmental influence, tryptophan fluoresces at 335 nm and 4 5 tyrosine at 306 nm (Villette et al., 2006; Lakowicz, 2006). 6 7 8 9 10 11 Maize stem cell walls 12 13 14 Improving the ease and yield of cell wall saccharification represents the major 15 16 technological hurdle that must be overcome before the full vision of the plant-fuelled 17 18 For Peer Review 19 biorefinery can be realized. Thus, mapping of the spatial distribution of the 20 21 fluorescent species at a micrometric resolution as a function of growth stage allows 22 23 the assessment of the best enzymatic treatments and high yield of recovery of useful 24 25 chemical for biofuel or biomaterials productions can be foreseen. 26 27 28 As shown in Figure 7, the spectral images studied recorded using a 275 nm 29 30 31 excitation, covered a region in the vascular bundle of the maize internodes cross- 32 33 section that contained mainly three cell types: parenchyma, sclerenchyma and 34 35 phloem cells. For clarity, average spectra were calculated for each cell type region as 36 37 shown in Figure 7 (bottom left). In the case of the phloem, a peak around 415 nm 38 39 was observed in the spectra while, in the case of sclerenchyma, the peak was 40 41 42 centred on 420 nm and fluorescence was also detected from 440 nm to 540 nm. This 43 44 confirmed that sclerenchyma cell walls contained lignin while phloem cell walls did 45 46 not (Allouche et al., 2012). In addition, multivariate data analysis reveals that para- 47 48 coumaric acid was found negatively correlated to lignin (Allouche et al., 2012). 49 50 51 Biopsies 52 53 54 The main features observable on tissues under 280 nm excitation are related to 55 56 57 tryptophan, tyrosine and in a lesser extent to constitutive proteins (Petit et al., 2010). 58 59 60 Biology of the Cell 10 Page 11 of 37 Biology of the Cell

1 2 3 However, when shifting the excitation wavelength to 340 nm, the collagen 4 5 fluorescence increases significantly. 6 7 8 Bones 9 10 11 Osteoporotic fractures represent a significant public health concern. Bone quality 12 13 (bone microarchitecture, collagen content) was investigated at subcellular resolution. 14 15 16 Synchrotron DUV spectroscopy allows to characterize and distinguish biochemical 17 18 content of the For osteocytes Peer and their surrounding Review matrix, since osteocytes are 19 20 considered as the orchestrator of the bone remodelling (Rochefort et al., 2010). In a 21 22 rat experimental model of alcoholic induced osteoporosis, sections of cortical bone 23 24 25 were investigated and osteocytes autofluorescence images are presented in Figure 26 27 8. The fluorescence emission spectra present three peaks situated at 305 nm, 333 28 29 nm and 385 nm, attributed to tyrosine, tryptophan and collagen respectively. The 30 31 ratio between tyrosine and tryptophan is dependent of the localisation in the bone 32 33 tissue. Osteocytes do present a higher tyrosine to tryptophan ratio than the matrix. 34 35 36 Moreover, from our knowledge and personal experience, no other tissue does 37 38 present a similar high level of tyrosine fluorescence. 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell 11 Biology of the Cell Page 12 of 37

1 2 3 Discussion 4 5 6 Several autofluorescent biomolecules can be observed inside isolated cells or in a 7 8 tissue. During physiological and/or pathological processes the changes occurring in 9 10 the cell and tissue result in modifications of the amount and distribution of those 11 12 13 endogenous fluorophores and chemical-physical properties of their 14 15 microenvironment. Therefore, monitoring the endogenous fluorophores distribution 16 17 and the intensity of their fluorescence emission can give diagnosis insights and can 18 For Peer Review 19 help to understand biochemical transformations. While the spectral characteristics of 20 21 22 isolated molecules can differ from physiological conditions, it is possible to identify 23 24 them inside our chosen examples. 25 26 27 As for plant tissues, during plant-pathogen infection, the physiological state of the 28 29 invaded tissues is altered. This can be reflected in changes in photosynthesis, 30 31 transpiration or both. Fluorescence imaging is thus useful for visualization of 32 33 34 emerging biotic stresses while detecting the chlorophyll fluorescence (endogenous 35 36 fluorochrome, which mediates photosynthesis) or monitoring metabolic cell activity 37 38 through detection of fluorescent coenzymes, such as NAD(P)H (Chaerle and Van der 39 40 Straeten, 2001; Kasimova, 2006). In addition, the possibility to follow the 41 42 spectroscopic state at micrometer scale of the major compounds without any 43 44 45 markers is of importance for high yield recovery of vegetal wastes. 46 47 48 Nicotinamide adenine dinucleotide (NAD(P)H) and flavin adenine dinucleotide (FAD) 49 50 are metabolic cofactors that act as electron donors and acceptors in the electron 51 52 transport chain of the mitochondria, thus having a major role in metabolism pathways 53 54 of eukaryotic cells. While the reduced form, NAD(P)H, is fluorescent and has an 55 56 57 excitation maximum in the region of 350 nm with emission at 450 nm (Figure 1), the 58 59 60 Biology of the Cell 12 Page 13 of 37 Biology of the Cell

1 2 + 3 oxidized form NAD , does not produce fluorescence (Paul and Schneckenburger, 4 5 1996; Kirkpatrick et al., 2005; Wang and Youle, 2009). Conversely, in the case of 6 7 flavins, the fluorescent form is the oxidized form (excitation maximum at 450 nm with 8 9 10 emission at 535 nm) and the reduced form does not fluoresce. Figure 6 shows the 11 12 possibility of spatial detection of tryptophan and tyrosine-related fluorescence. One 13 14 can envisage following the changes of tyrosine and tryptophan signal to monitor the 15 16 proteins synthesis, degradation or both. Thus, synchrotron DUV microspectroscopy 17 18 opens boundaries,For since understanding Peer of Reviewthe spatial tryptophan and tyrosine signal 19 20 21 changes, in relation to metabolic changes, may have both diagnostic as well as 22 23 prognostic potential when treatment is concerned. 24 25 26 As for bone tissue, synchrotron UV microspectroscopy may give quantitative 27 28 information on the cortical bone content of some autofluorescent molecules both in 29 30 bone cells and their surrounding extracellular matrix. Results have shown that strong 31 32 33 chronic alcohol consumption induced a decrease of the ratio tyrosine/tryptophan both 34 35 in cell and surrounding matrix. This new methodology warrants a cell characterization 36 37 in situ that allows evaluating osteocyte metabolism and gives a better understanding 38 39 of the key role of this cell in the bone remodelling regulation. This local measurement 40 41 of tyrosine and tryptophan levels may give a better understanding of cellular 42 43 44 metabolism according to different physio-pathological conditions such as 45 46 osteoporosis. 47 48 49 Many probes that are usually excited in the visible may be also excited in the UV 50 51 range. For example, fluorescent proteins (FPs) can be excited either directly through 52 53 their chromophore or by exciting the tryptophan residues that transfer their energy to 54 55 56 the chromophore (Visser et al., 2005). This could lead to higher resolution studies of 57 58 FPs by lowering their excitation wavelength. Moreover, this opens the possibility to 59 60 Biology of the Cell 13 Biology of the Cell Page 14 of 37

1 2 3 study simultaneously several different FPs with one single excitation of their 4 5 respective tryptophan. 6 7 8 Moreover, many drugs and toxins do present fluorescence when excited in the DUV; 9 10 it is then possible to follow their pharmacokinetic at tissular, cellular or subcellular 11 12 13 level by directly following their fluorescence distribution without adding external 14 15 probes or radiolabels (Diaspro et al., 2005). As an example in the area of material 16 17 science, phenol- or diphenol-molecules, like Bisphenol A (BPA), exhibit absorption 18 For Peer Review 19 and fluorescence in DUV region (Del Olmo et al., 1999). BPA is used as a 20 21 polymerization inhibitor in PVC while epoxy resins containing BPA are used as 22 23 24 coatings on the inside of almost all food and beverage cans. As such, BPA leaching 25 26 can occur when they are being treated with high temperature or extreme pH (Brede 27 28 et al., 2003; Munguía-López et al., 2005). Because of health risk, one can envisage 29 30 to use synchrotron DUV microspectroscopy to detect its presence at very low 31 32 33 concentrations in food constituents and/or to use this technique to develop and 34 35 validate sample preparations methods, which would lead to heath safe requirements. 36 37 38 Of note, the ideal drug to be followed in UV shall have an emission shifted from the 39 40 tryptophan fluorescence (since tryptophan has strong contribution to 41 42 autofluorescence signal), or to bypass the tryptophan fluorescence by an excitation 43 44 45 outside the 240 to 300 nm range, or if not, then a drug with high quantum yield of 46 47 fluorescence is required (order of higher than the autofluorescence signal). 48 49 Otherwise, multivariate data treatment is required to separate the molecule of interest 50 51 from the proteins fluorescence (Batard et al., 2001). 52 53 54 From the examples discussed above, we clearly showed that DUV 55 56 microspectroscopy could monitor the numerous endogenous fluorophores and their 57 58 59 60 Biology of the Cell 14 Page 15 of 37 Biology of the Cell

1 2 3 distribution on the tissular as well as cellular level. Note, working in monochromatic 4 5 excitation mode, the light power is very low (µW). This allowed us to record the 6 7 spectral map of living cells without detecting any changes in cell morphology during 8 9 10 or after the measurement. As discussed above, measuring of live cell opens up 11 12 possibility to follow the dynamic changes within the specimen. However, in 13 14 experiments were dynamic processes are not issue; one can work with fixed 15 16 specimens. Indeed, fixation with formaldehyde did not reveal any changes in 17 18 autofluorescenceFor of single cellPeer when excited Review with 275 nm; e.g. for fixed as for live 19 20 21 cells tryptophan fluoresced at 335 nm and tyrosine at 306 nm (data not shown). 22 23 Although, the autofluorescence of cells and tissues is often (almost ever) considered 24 25 as a parasitic signal that must be minimized; here, we demonstrate, that using the 26 27 right experimental conditions, this so-called background signal, is rich in information 28 29 30 and can become useful. 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell 15 Biology of the Cell Page 16 of 37

1 2 3 Material and Methods 4 5 6 Isolated auto fluorescent compounds 7 8 9 Chemicals: Pyridoxin, N-acetyl-L-tryptophanamide, N-acetyl-L-tyrosinamide, •- 10 11 Nicotinamide adenine dinucleotide, reduced dipotassium salt (NADH), Alcohol 12 13 14 dehydrogenase from Saccharomyces cerevisiae (ADH), Collagen type I, Elastin, 15 16 phosphate-buffered saline (PBS), sodium hydroxide (NaOH) and TRIS were 17 18 purchased from For Sigma-Aldrich Peer (Lyon, France). Review Acetic acid was purchased from 19 20 Merck (Fontenay sous-Bois, France) and hydrogen chloride (HCl) from Carl Roth 21 22 GmbH (Germany). 23 24 25 Preparation of solutions: Stock solution of Pyridoxine (c = 8.75x10-3M) was 26 27 28 prepared by dissolving pyridoxine powder in 1 M HCl. For recording the pyridoxine 29 30 absorption and fluorescence spectra, the stock solution was diluted with PBS pH = 31 32 7.4 to a final concentration of 5x10-6M. Collagen type I was first dissolved in 0.1 M 33 34 acetic acid to a concentration of 1 mg/ml. The stock of collagen was then diluted with 35 36 37 PBS pH = 7.4 to a final concentration of 0.074 mg/ml. Stock solution of Elastin (c = 1 38 39 mg/ml) was prepared by dissolving elastin powder in 1M NaOH, followed by dilution 40 41 in PBS pH = 7.4 to a final concentration of 0.027 mg/ml. The N-acetyl-L- 42 43 tryptophanamide and N-acetyl-L-tyrosinamide were both dissolved in 25x10-3M Tris- 44 45 HCl pH = 7.8 to obtain stock solutions concentrations of 8.15x10-3M for N-acetyl-L- 46 47 -3 48 tryptophanamide and 6.39x10 M for N-acetyl-L-tyrosinamide respectively. Both stock 49 -6 50 solutions were then diluted to a concentration of 10x10 M for N-acetyl-L- 51 52 tryptophanamide and 21.3x10-6 M for N-acetyl-L-tyrosinamide. Absorption and 53 54 fluorescence spectra were measured immediately after dilution. 55 56 57 58 59 60 Biology of the Cell 16 Page 17 of 37 Biology of the Cell

1 2 3 Preparation of NADH: ADH complexes: Stock solutions of NADH (c = 1.85x10-2 M) 4 5 and stock solution of ADH (c = 2.83x10-3M) were prepared in 0.1M Tris-HCl pH = 8. 6 7 Solutions of complexes were obtained by dilution of stock solutions of NADH and 8 9 10 ADH with 0.1M Tris-HCl pH = 8. Three different complexes of NADH:ADH have been 11 -6 12 prepared at the following concentrations for both components: NADH (10x10 M) : 13 14 ADH (1x10-4 M); NADH (8.2x10-6 M) : ADH (5x10-4 M) and NADH (6.5x10-6 M) : ADH 15 16 (1x10-3 M). The absorption and fluorescence spectra of thus prepared solutions were 17 18 measured immediatelyFor after Peer mixing. In orderReview to distinguish between the spectral 19 20 21 characteristics of these components, the NADH and ADH were measured also 22 23 individually: NADH in 0.1M Tris-HCl pH = 8 at 10x10-6 M concentration and ADH at 24 25 1.62x10-4 M concentration. 26 27 28 Absorption: To avoid inner filter effect in our fluorescence emission experiment, for 29 30 each sample absorption spectrum was also measured. Absorption spectra were 31 32 33 recorded at 20° C using UV-VIS spectrophotometer (Specord 210, Analytic Jena AG, 34 35 Jena, Germany) in the 190-900 nm range with slits width corresponding to a 36 37 resolution of 2 nm. For measurements, we used a 1 cm path length quartz cell. 38 39 40 Fluorescence: Fluorescence spectra were measured on optically diluted samples 41 42 (i.e. absorption < 0.1 at the excitation wavelength) recorded at 20° C using a 43 44 45 FluoroMax-4 (HORIBA Jobin Yvon INC, Chilly Mazarin, France) spectrofluorimeter. 46 47 To record the excitation and emission spectra of individual compounds, the best 48 49 excitation and emission wavelengths were chosen after optimisation (see Figure 2). 50 51 52 Para-coumaric acid, ferulic acid and lignin 53 54 55 Para-coumaric acid and ferulic acid were purchased from Sigma-Aldrich (Lyon, 56 57 France). Powder was first dissolved in water at a concentration of 3.2 mg/mL and 2.6 58 59 60 Biology of the Cell 17 Biology of the Cell Page 18 of 37

1 2 3 mg/mL respectively. Samples were further diluted in water in order to obtain an 4 5 absorbance value of about 0.03 when measured in 1 cm path length cuvettes. 6 7 Fluorescence excitation and emission spectra were recorded on a Fluorolog-3 8 9 10 spectrophotometer (HORIBA Jobin Yvon INC, Chilly Mazarin, France), using 3 mm 11 12 path length cuvettes (Hellma France, Paris). Lignin was observed as a powder in a 13 14 special holder (HORIBA Jobin Yvon INC, Chilly Mazarin, France) in a front-face 15 16 configuration. A long pass filter with a cut-off at 351 nm was inserted between 17 18 sample and detectorFor for ferulic Peer acid measurement. Review 19 20 21 22 DUV microspectrometer 23 24 25 Microspectrofluorescence spectra were recorded on Polypheme, the DUV inverted 26 27 microspectrofluorimeter installed at DISCO Beamline (Jamme et al., 2010). Excitation 28 29 was provided by the continuous emittance from the DISCO beamline bending 30 31 magnet at Synchrotron SOLEIL (Giuliani et al., 2009). While the system was already 32 33 described elsewhere (Jamme et al., 2010), it is constructed around an Olympus IX71 34 35 36 inverted microscope stand with homemade replacement of the intermediate lenses 37 38 that were not transparent in UV. Light detection is collected through a DUV lens and 39 40 an adjustable pinhole. In order to suppress the Rayleigh band whatever the excitation 41 42 wavelength, the detection system (T6400, Jobin-Yvon, Fr) uses a triple 43 44 45 monochromator in a substractive mode. Thereafter, the fluorescence emission 46 47 spectrum of the studied voxel is projected onto a -70°C peltier cooled iDus CCD 48 49 (Andor) of 1024x256 pixels with a 26x26 µm pixel size and a 26.6x6.7 mm detector 50 51 size and a dark current of less than 0.002e/pixel/sec. Spectra are recorded at a 1024 52 53 pixels depth and a 16 bits dynamic range. 54 55 56 57 58 59 60 Biology of the Cell 18 Page 19 of 37 Biology of the Cell

1 2 3 Isolated cells 4 5 6 Cell line and culture condition: HeLa cells, a human cervix epithelial cell line, were 7 8 a generous gift from Dr. Denis Biard from CEA-DSV-iRCM/INSERM U935, Institut A. 9 10 Lwoff-CNRS, Villejuif Cedex, France. Cells were routinely cultured as monolayer and 11 12 13 were grown in Dulbecco’s modified Eagle’s medium (DMEM) containing L-glutamine 14 -1 -1 -1 15 (862 mg L ), sodium pyruvate (110 mg L ) and glucose (4500 mg L ), 16 17 supplemented with 10% fetal calf serum (FCS), penicillin (50 µg mL-1) and 18 For Peer Review 19 streptomycin (50 µg mL-1). All chemicals were obtained from Gibco Invitrogen 20 21 (Villebon sur Yvette, France). The cells were maintained at 37°C in a humidified 22 23 24 atmosphere of 5% CO2. For the microscopic detection of endogenous fluorophores 25 26 from HeLa cells, the cells were plated in plastic Petri dishes (35x10 mm) containing 27 28 25 mm round quartz coverslip (Cening, Cn) and incubated for 2 days. Cell 29 30 attachment and growth was monitored by visible inspection using a microscope. 31 32 33 Coverslip with desired cell growth confluence was rinsed in DMEM in absence of 34 35 FCS or antibiotics for 1-2 minutes. Thereafter, the living cells were directly mounted 36 37 on the microscope and spectra of cells in PBS were taken. 38 39 40 Maize stem cell walls 41 42 43 Maize from a reference genotype F2 were grown at INRA Lusignan and stems were 44 45 harvested at the female flowering stage. The internodes under the ear were collected 46 47 48 and stored in 70% (v/v) ethanol/water. 49 50 51 Samples were taken from the middle of maize internodes, embedded into paraffin 52 53 and sectioned using a microtome (Jamme et al., 2008). Proteins were removed using 54 55 a protease enzyme named subtilisin A type VIII, bacterial, from bacillus licheniformis. 56 57 Preliminary results showed that 10 µm thick sections were adapted for fluorescence 58 59 60 Biology of the Cell 19 Biology of the Cell Page 20 of 37

1 2 3 spectral acquisitions. Sections were collected on ZnS windows. A region 4 5 corresponding to a vascular bundle near the epidermis was selected as showing a 6 7 chemical variability according to cell types (Figure 7) (Jung and Casler, 2006). 8 9 10 Bones 11 12 13 These studies were approved by the National Institute for Health and Medical 14 15 16 Research (INSERM). The procedure for the care and killing of the animals was in 17 18 accordance with For the European Peer Community Review standards on the care and use of 19 20 laboratory animals (Ministère de l’agriculture, France, Authorisation INSERM45-001). 21 22 Wistar male rats (8 weeks old at baseline) were chosen at random and assigned to 23 24 25 one of the three following groups: controls and two different percentages ethanol 26 27 beverage (25% v/v and 35% v/v) during 19 weeks. After the sacrifice, tibias were 28 29 dissected free of connective and fat tissue. They were fixed in a 4% v/v formalin 30 31 solution and kept at + 4°C. As required the tibias were cut transversally in slices 32 33 (thickness 300 µm) in the superior third part of the diaphysis, with a high speed rotary 34 35 36 tool (Dremel 300, Dremel, USA). 37 38 39 UV monochromatised light (typically between 270 and 330 nm) was used to excite 40 41 cortical bone sections through a 40 ultrafluar objective (Zeiss, Germany). The 42 43 fluorescence emission spectrum arising from each excited pixel is recorded. 44 45 Rastering of the sample allows one to record x, y, lambda, I maps of interest [2]. 46 47 2 48 Mapping of 20  20 µm was performed with a 2•2 µm step size, chosen by 49 50 considering the relative size of the cells and the area to be scanned, with a 5 s 51 52 acquisition time per spectrum. The Region of Interest (ROI) was centred on the 53 54 osteocyte. For each rat sample, 3 acquisitions were achieved both in osteocytes ROI 55 56 and surrounding matrix ROI. 57 58 59 60 Biology of the Cell 20 Page 21 of 37 Biology of the Cell

1 2 3 4 5 6 7 8 9 10 Author contributions 11 12 13 F.J., S.V. and M.R. conceived and designed the research; F.J., S.K., S.V., F.A. and 14 15 V.R. carried out the experiments; F.A. and S.P. provided some of the samples; F.J., 16 17 S.K, S.V. and M.R analysed the data; F.J., S.K., S.V. and M.R. wrote the article. 18 For Peer Review 19 20 21 References 22 23 24 25 Allouche, F., Hanafi, M., Jamme, F., Robert, P., Guillon, F., Devaux, M.F. (2012) 26 27 Coupling hyperspectral image data having different spatial resolutions by 28 29 extending multivariate inter-battery Tucker analysis. Chemometr. Intell. Lab. 30 31 113(1), 43-51 32 33 34 35 Anderson, W., Akin, D. (2008) Structural and chemical properties of grass 36 37 lignocelluloses related to conversion for biofuels. J. Ind. Microbiol. Biot. 35, 38 39 355–366 40 41 42 Batard, E., Jamme, F., Villette, S., Jacqueline, C., de la Cochetière, M. F., Caillon, J., 43 44 Réfregiers, M. (2011) Diffusion of ofloxacin in the endocarditis vegetation 45 46 47 assessed with synchrotron radiation UV fluorescence microspectrocopy. PLoS 48 49 ONE 6 50 51 52 Brede, C., Fjeldal, P., Skjevrak, I., Herikstad, H. (2003) Increased migration levels of 53 54 bisphenol A from polycarbonate baby bottles after dishwashing, boiling and 55 56 57 brushing. Food Addit. Contam. 20, 684–689 58 59 60 Biology of the Cell 21 Biology of the Cell Page 22 of 37

1 2 3 Chaerle, L., Van der Straeten, D. (2001) Seeing is believing: imaging techniques to 4 5 monitor plant health. B.B.A. Gene Struct. Expr. 1519, 153-166 6 7 8 Cox, S., Rosten, E., Monypenny, J., Jovanovic-Talisman, T., Burnette, D.T., 9 10 11 Lippincott-Schwartz, J., Jones, G.E., Heintzmann, R. (2011) Bayesian 12 13 localization microscopy reveals nanoscale podosome dynamics. Nat Meth 9, 14 15 195–2001. 16 17 18 Del Olmo, M., Zafra,For A., Navas, Peer N. (1999) TraceReview determination of phenol, bisphenol A 19 20 and bisphenol A diglycidyl ether in mixtures by excitation fluorescence 21 22 23 following micro liquid–liquid extraction using partial least squares regression. 24 25 Analyst 124, 385-390 26 27 28 Diaspro, A., Chirico, G., Collini, M. (2005) Two-photon fluorescence excitation and 29 30 related techniques in biological microscopy, Q. Rev. Biophys 38(2), 97-166 31 32 33 Giuliani, A., Jamme, F., Rouam, V., Wien, F., Giorgetta, J. L., Lagarde, B., Chubar, 34 35 36 O., Bac, S., Yao, I., Rey, S., Herbeaux , C., Marlats, J. L., Zerbib, D., Polack, 37 38 F., Réfregiers, M. (2009) DISCO: a low-energy multipurpose beamline at 39 40 synchrotron SOLEIL. J. Sync. Rad. 16 835–841 41 42 43 Jamme, F., Robert, P., Bouchet, B., Saulnier, L., Dumas, P., Guillon, F. (2008) 44 45 46 Aleurone cell walls of wheat grain: high spatial resolution investigation using 47 48 synchrotron infrared microspectroscopy. Appl. Spectrosc. 62, 895–900 49 50 51 Jamme, F., Villette, S., Giuliani, A., Rouam, V., Wien, F., Lagarde, B., Réfrégiers, M. 52 53 (2010) Synchrotron UV fluorescence microscopy uncovers new probes in cells 54 55 and tissues. Microsc. Microanal. 16, 507–514 56 57 58 59 60 Biology of the Cell 22 Page 23 of 37 Biology of the Cell

1 2 3 Jung, H.G., Casler, M.D. (2006) Maize Stem Tissues. Crop Science 46, 1793-1800 4 5 6 Kanick, S., Van der Leest, C., Aerts, J. (2010) Integration of single-fiber reflectance 7 8 spectroscopy into ultrasound-guided endoscopic lung cancer staging of 9 10 11 mediastinal lymph nodes. J. Biomed. Opt. 15, 017004 12 13 14 Kasimova, M.R. (2006) The free NADH concentration is kept constant in plant 15 16 mitochondria under different metabolic conditions. Plant Cell 18, 688–698 17 18 For Peer Review 19 Kirkpatrick, N., Zou, C., Brewer, M. (2005) Endogenous Fluorescence Spectroscopy 20 21 of Cell Suspensions for Chemopreventive Drug Monitoring. Photochem. 22 23 24 Photobiol. 81, 125-134 25 26 27 Kohler, A. (1904) A microphotographic device for ultraviolet light (lambda=275 mu) 28 29 and investigations into organic tissues using this device. Phys. Z. 5, 666–673 30 31 32 Lakowicz, J.R. (2006) Principles of Fluorescence Spectroscopy. Boston: Springer US 33 34 35 Ludford, R.J., Smiles, J., Welch, F.V. (1948) Ultra-Violet microscopy of living 36 37 38 malignant cells. Nature 162, 650–651 39 40 41 Mellors, R.C., Berger, R.E., Streim, H.G. (1950) Ultraviolet Microscopy and 42 43 Microspectroscopy of Resting and Dividing Cells: Studies with a Reflecting 44 45 Microscope. Science 111, 627–632 46 47 48 Munguía-López, E.M., Gerardo-Lugo, S., Peralta, E., Bolumen, S., Soto-Valdez, H. 49 50 51 (2005) Migration of bisphenol A (BPA) from can coatings into a fatty-food 52 53 simulant and tuna fish. Food Addit. Contam. 22, 892–898 54 55 56 Paul, R.J., Schneckenburger, H. (1996) Oxygen concentration and the oxidation- 57 58 59 60 Biology of the Cell 23 Biology of the Cell Page 24 of 37

1 2 3 reduction state of yeast: determination of free/bound NADH and flavins by 4 5 time-resolved spectroscopy. Naturwissenschaften 83, 32–35 6 7 8 Pavlova, I., Williams, M., El-Naggar, A., Richards-Kortum, R., Gillenwater, A. (2008) 9 10 11 Understanding the biological basis of autofluorescence imaging for oral Cancer 12 13 detection: high-resolution fluorescence microscopy in viable tissue. Clin. 14 15 Cancer Res. 14, 2396–2404 16 17 18 Petit, V.W., Réfregiers,For M., Guettier,Peer C., Jamme, Review F., Sebanayakam, K., Brunelle, A., 19 20 Laprévote, O., Dumas, P., Le Naour, F. (2010) Multimodal spectroscopy 21 22 23 combining time-of-flight-secondary ion mass spectrometry, synchrotron-FT-IR, 24 25 and synchrotron-UV microspectroscopies on the same tissue section. Anal. 26 27 Chem. 82, 3963–3968 28 29 30 Ramanujam, N. (2000) Fluorescence spectroscopy of neoplastic and non-neoplastic 31 32 tissues. Neoplasia (New York) 33 34 35 36 Ramanujam, N., Mitchell, M.F., Mahadevan, A., Warren, S., Thomsen, S., Silva, E., 37 38 Richards-Kortum, R. (1994) In vivo diagnosis of cervical intraepithelial 39 40 neoplasia using 337-nm-excited laser-induced fluorescence. Proc. Natl. Acad. 41 42 Sci. USA 91, 10193–10197 43 44 45 Robert, P., Jamme, F., Barron, C., Bouchet, B., Saulnier, L., Dumas, P., Guillon, F. 46 47 (2010) Change in wall composition of transfer and aleurone cells during wheat 48 49 grain development. Planta 223, 393-406 50 51 52 Rochefort, G.Y., Pallu, S., Benhamou, C.L. (2010) Osteocyte: the unrecognized side 53 54 of bone tissue. Osteoporos. Int. 21, 1457–1469 55 56 57 58 Saadi, A., Lempereur, I., Sharonov, S., Autran, J. C., Manfait, M. (1998) Spatial 59 60 Biology of the Cell 24 Page 25 of 37 Biology of the Cell

1 2 3 Distribution of Phenolic Materials in Durum Wheat Grain as Probed by 4 5 Confocal Fluorescence Spectral Imaging. J. Cereal Sci. 28, 107–114 6 7 8 Tawil, G., Jamme, F., Réfregiers, M., Viksø-Nielsen, A., Colonna, P., Buléon, A. 9 10 11 (2011) In situ tracking of enzymatic breakdown of starch granules by 12 13 synchrotron UV fluorescence microscopy. Anal. Chem. 83, 989–993 14 15 16 Villette, S., Pigaglio-Deshayes, S., Vever-Bizet, C., Validire, P., Bourg-Heckly, G.V. 17 18 (2006) Ultraviolet-inducedFor Peer autofluorescence Review characterization of normal and 19 20 tumoral esophageal epithelium cells with quantitation of NAD(P)H. Photochem. 21 22 23 Photobiol. Sci. 5,483-492 24 25 26 Visser, N.V., Borst, J.W., Hink, M.A., van Hoek, A., Visser, A.J.W.G. (2005) Direct 27 28 observation of resonance tryptophan-to-chromophore energy transfer in visible 29 30 fluorescent proteins. Biophys. Chem. 116, 207–212 31 32 33 Wagnieres, G., Star, W., Wilson, B. (1998) In vivo fluorescence spectroscopy and 34 35 36 imaging for oncological applications. Photochem. Photobiol. 68, 603–632 37 38 39 Wang, C., Youle, R.J. (2009) The role of mitochondria in apoptosis. Annu. rev. genet. 40 41 43, 95-118 42 43 44 Zeskind, B.J., Jordan, C.D., Timp, W., Trapani, L., Waller, G., Horodincu, V., Ehrlich, 45 46 D.J., Matsudaira, P. (2007) Nucleic acid and protein mass mapping by live-cell 47 48 49 deep-ultraviolet microscopy. Nat. Meth. 4, 567–569 50 51 Zipfel, W.R., Williams, R.M., Christie, R., Nikitin, A.Y., Hyman, B.T., Webb, W.W. 52 53 (2003) Live tissue intrinsic emission microscopy using multiphoton-excited 54 55 native fluorescence and second harmonic generation. Proc. Natl. Acad. Sci. 56 57 58 USA 100, 7075–7080 59 60 Biology of the Cell 25 Biology of the Cell Page 26 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell 26 Page 27 of 37 Biology of the Cell

1 2 3 Legend 4

5 6 Figure 1: Fluorescence excitation spectra of the main autofluorescent contributors 7 8 found in mammalian cells and tissues. Maxima of excitation are presented following 9 10 the name of the compound. Within brackets is indicated the emission wavelength at 11 12 which was recorded the excitation spectrum. 13 14 Figure 2: Fluorescence emission spectra of the main autofluorescent contributors 15 16 17 observable in mammalian cells and tissues. Maxima of emission are presented 18 For Peer Review 19 following the name of the compound. Within brackets is indicated the excitation 20 21 wavelength at which was recorded the emission spectrum. 22 23 Figure 3: Fluorescence excitation spectra of the main autofluorescent contributors 24 25 26 found in plant tissues: lignin, para-coumaric and ferulic acids. 27 28 Figure 4: Fluorescence emission spectra of the main autofluorescent contributors 29 30 found in plant tissues: lignin, para-coumaric and ferulic acids. 31 32 33 Figure 5: (a) Spectral distribution of the photon flux delivered to the sample and (b) 34 35 relative response of the microscope over the whole spectral range. 36 37 38 Figure 6: (a) Transmission image of a living single Hela cell, (b) corresponding sum 39 40 intensity spectral image and (c) spectrum of the grey pixel. The gaussian 41 42 deconvolution presents two main components: tryrosine and tryptophan. Pixel size 43 44 2 45 1.5 x 1.5 µm 46 47 48 Figure 7: (a) Transmission image of a maize stem section, (b) corresponding sum 49 50 intensity spectral image and (c) average spectra of the three cell types. Scale bar 4 51 52 microns. 53 54 55 56 57 58 59 60 Biology of the Cell 27 Biology of the Cell Page 28 of 37

1 2 3 Figure 8: (a) Mouse bone tissue transmission image showing the localization of the 4 5 osteocyte inside the bone matrix and (b) typical spectra of matrix and osteocyte 6 7 2 recorded from 1x1 µm pixels (bottom). Scale bar 8 microns. 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell 28 Page 29 of 37 Biology of the Cell

1 2 3 4 5 6 7 8 9 10 For Peer Review 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 Biology of the Cell 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell Page 30 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Figure 1: Fluorescence excitation spectra of the main autofluorescent contributors found in mammalian cells 39 and tissues. Maxima of excitation are presented following the name of the compound. Within brackets is 40 indicated the emission wavelength at which was recorded the excitation spectrum. 41 296x278mm (72 x 72 DPI) 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell Page 31 of 37 Biology of the Cell

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Figure 2: Fluorescence emission spectra of the main autofluorescent contributors observable in mammalian 39 cells and tissues. Maxima of emission are presented following the name of the compound. Within brackets is indicated the excitation wavelength at which was recorded the emission spectrum. 40 298x275mm (72 x 72 DPI) 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell Biology of the Cell Page 32 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Figure 3: Fluorescence excitation spectra of the main autofluorescent contributors found in plant tissues: 39 lignin, para-coumaric and ferulic acids. 300x277mm (72 x 72 DPI) 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell Page 33 of 37 Biology of the Cell

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Figure 4: Fluorescence emission spectra of the main autofluorescent contributors found in plant tissues: 38 lignin, para-coumaric and ferulic acids. 39 306x275mm (72 x 72 DPI) 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell Biology of the Cell Page 34 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 Figure 5: (a) Spectral distribution of the photon flux delivered to the sample and (b) relative response of the microscope over the whole spectral range. 48 215x279mm (72 x 72 DPI) 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell Page 35 of 37 Biology of the Cell

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 Figure 6: (a) Transmission image of a living single Hela cell, (b) corresponding sum intensity spectral image and (c) spectrum of the grey pixel. The gaussian deconvolution presents two main components: tryrosine 48 and tryptophan. Pixel size 1.5 x 1.5 µm2 49 209x297mm (72 x 72 DPI) 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell Biology of the Cell Page 36 of 37

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 Figure 7: (a) Transmission image of a maize stem section, (b) corresponding sum intensity spectral image and (c) average spectra of the three cell types. Scale bar 4 microns. 48 209x297mm (72 x 72 DPI) 49 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell Page 37 of 37 Biology of the Cell

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For Peer Review 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 Figure 8: (a) Mouse bone tissue transmission image showing the localization of the osteocyte inside the bone matrix and (b) typical spectra of matrix and osteocyte recorded from 1x1 µm2 pixels (bottom). Scale 48 bar 8 microns. 49 210x297mm (72 x 72 DPI) 50 51 52 53 54 55 56 57 58 59 60 Biology of the Cell