1. A comparative study of ex vivo skin optical clearing using two-photon and Raman microscopy
Anton Sdobnov1,*, Maxim E. Darvin2, Juergen Lademann2, Valery Tuchin3
1Optoelectronics and Measurement Techniques Laboratory, University of Oulu, Finland
2Center of Experimental and Applied Cutaneous Physiology, Department of Dermatology,
Venerology and Allergology, Charité – Universitätsmedizin Berlin, Germany
3Research-Education Institute of Optics and Biophotonics,
Saratov National Research State University, Russian Federation
* Corresponding author:

2. Presentation content Optical clearing 3
Main mechanisms of tissue optical immersion clearing
Materials and methods: Multiphoton tomography
Materials and methods: Confocal Raman microscopy
Materials and methods: Skin Sample Preparation
Materials and methods: Optical Clearing Agents
Results: optical clearing using two-photon microscopy
Conclusions for two-photon measurements
Results: optical clearing usingConfocal Raman microscopy
Conclusions for Raman measurements
References

3. Optical clearing
Large variety of noninvasive optical skin diagnostic techniques has been developed for clinical applications, such as optical coherence tomography (OCT), laser speckle contrast imaging (LSCI), confocal laser scanning microscopy, Raman and coherent anti-Stokes Raman spectroscopy (CARS), confocal Raman microscopy, multiphoton tomography (MPT), including CARS tomography and second harmonic generation (SHG) imaging. However, because of strong light scattering of the stratum corneum and living epidermis the penetration depth of the light beam and its focusing accuracy are strongly limited. The optical clearing (OC) technique, which was developed and has been intensively studied since the 1990s, allows to effectively reduce light scattering in tissue, reach the maximum probing depth, improve contrast, provide enhanced light focusing ability and the spatial resolution of optical diagnostic methods. 3

4. Main mechanisms of tissue optical immersion clearing
Three hypothesized mechanisms of tissue OC were suggested:
Matching of refractive indices (n) between tissue components and interstitial fluid (ISF) modified by an OCA diffused into the tissue.
Reversible dissociation of collagen fibers
Tissue dehydration induced by hyperosmolarity of the applied agent.
These and possibly other not known OC mechanisms usually works not independently but simultaneously with different relative contributions dependent on tissue and OCA and delivery method. 4

5. Materials and methods: Multiphoton tomography
Multiphoton tomography for application in dermatology is an imaging technique based on the two photon excitation of cutaneous fluorophores. The Two-Photon Excited Auto Fluorescence (TPEAF) originates from NAD(P)H, keratin, elastin and melanin, while the second harmonic generation (SHG) is due to collagen I molecule response.
The investigations were carried out using a commercially available two-photon tomograph (DermaInspect, Jenlab GmbH, Jena, Germany) equipped with a tunable femtosecond titanium sapphire laser (Mai Tai XF, Spectra Physics, USA, 710–920 nm, 100-fs pulses at a repetition rate of 80 MHz). An objective lens with 40× magnification and NA = 1.3 has been used. The lateral and axial resolutions in the skin are 0.5 ± 0.1 μm and 1.6 ± 0.4 μm, respectively. 5

6. Materials and methods: Confocal Raman microscopy
The CRM measurements were performed using a skin composition analyzer appropriated for in vivo/ex vivo skin measurements (River Diagnostics, Model SCA; Rotterdam, The Netherlands). The following settings were used: excitation wavelength of 785 nm for the fingerprint region (400–2000 cm−1), oil objective of ×50 with spot of 5 μm,laser power of 20 mW on the sample and an exposure time of 5 s.
Such a wavelength allows a deep penetration into theskin sample due to reduced absorption and scattering, and the utilized doses of excitation light (approx. 1 J cm−2) are not sufficient to damage the skin components. The spatial axial
and spectral resolutions of the instrument were ⩽5 μm and2 cm−1, respectively. 6

7. Materials and methods: Skin Sample Preparation
Measurements were performed on fresh porcine ear skin. Porcine ear skin was obtained on the day of sacrifice, cleaned with cold running water, dried using a paper towel, and stored in a refrigerator at +5 °C not more than 2 days. To standardize initial condition of skin samples in order to get reproducible results on optical clearing efficiency, and reduce the influence of artifacts, before OCA application, skin samples were left for 30 min at room temperature (+20 °C). In order to increase OCA penetration into the skin, the stratum corneum was partially removed using the following procedure:
Hair together with superficial SC was removed from the skin by shaving with a disposable razor.
The tape stripping procedure was performed (15 strips).
The skin surface was defatted with ethanol for 5 s. 7

8. Materials and methods: Optical Clearing Agents
For the investigation of the optical clearing effect, two OCAs were chosen:
Glycerol which is the most frequently used OCA as its biocompatibility and pharmacokinetics renders it suitable for skin. The refractive index for 100% solution was n = Viscosity and osmolarity was 1410 cp and Osm l−1, respectively.
Iohexol,N,N′-bis(2,3-dihydroxypropyl)-5-[N-(2,3-dihydroxypropyl)-acetamido]- 2,4,6-triiodoisophthalamide, a non-ionic, water-soluble radiographic contrast medium with a molecular weight of g mol−1 manufactured by GE Healthcare Ireland, Cork, Ireland, known by the trademark Omnipaque™ (300). The refractive index for 100% solution was n = Viscosity and osmolarity was 11.8 cp and Osm l−1, respectively. 8

9. Results: optical clearing using two-photon microscopy
Figure 1. Structural images of different skin layers obtained ex vivo for porcine ear skin samples at application of different OmnipaqueTM and glycerol solutions. Red color corresponds to TPEAF signal channel. Green color corresponds to SHG signal channel. 9

10. Figure 2. Averaged depth-dependent intensity profiles of TPEAF (a) and SHG (b) signals obtained ex vivo for porcine ear skin samples at application of different OmnipaqueTM and glycerol solutions; SD –standard deviations of TPEAF and SHG signals; depth-dependent indices of optical clearing efficiency (OCE) calculated for signal intensities of TPEAF (c) and SHG (d) signals. Index of the optical clearing efficiency has been calculated as a ratio between intensity after and before optical clearing. 10

11. Figure 3. Comparison of skin epidermal layer structure on 35 μmdepth: without (a) and with [(b)–(d)] OCA treatment; 100% OmnipaqueTM (b); 40% glycerol (c);100% glycerol (d).
It is clearly seen that glycerol causes greater cells shrinkage due strong dehydration. 11

12. Conclusions for two-photon measurements
The results show that a topical application of glycerol or OmnipaqueTM solutions onto the skin for 60 min significantly improved the depth and contrast of the MPT signals. By utilizing 40%, 60% and 100% glycerol, and 60% and 100% OmnipaqueTM it was demonstrated that both agents improve autofluorescence and SHG (second harmonic generation) signals from the skin. At the applied concentrations and agent time exposure, glycerol is more effective than OmnipaqueTM. However, tissue shrinkage and cell morphology changes were found for highly concentrated glycerol solutions. OmnipaqueTM, on the contrary, increases the safety and has no or minimal tissue shrinkage during the optical clearing process. Moreover OmnipaqueTM allows for robust multimodal optical/X-ray imaging with automatically matched optically cleared and X-ray contrasted tissue volumes. These findings make OmnipaqueTM more prospective than glycerol for some particular application.
112

13. Results: optical clearing using Confocal Raman microscopy
Figure 4. Depth-dependent Raman spectra of porcine skin without OCA treatment (a), after 30 min (b) and 60 min (c) of 100% Omnipaque™ treatment and after 30 min (d) and 60 min (e) of 70% glycerol in water solution treatment. The dotted vertical lines indicate Omnipaque™-related Raman peaks at 774 and 1516 cm−1 (b) and (c) and glycerol-related Raman peaks at 486 and 1056 cm−1 (d) and (e). 13

14. 14

15. To show the collagen hydration, the 938/922 cm−1 peak ratio was calculated as a potential spectroscopic marker of collagen hydration. The 922 cm−1 and 938 cm−1 peaks correspond to the stretching vibration of C–C bond (νC–C) and vibration of skeletal C–C stretching mode of collagen chains, respectively. The 938/922 cm−1 peak ratio increases with the increasing collagen hydration. 15

16. Conclusions for Raman measurements
The intensity of the skin-related Raman peaks significantly increased starting from the depth 160 μm for Omnipaque™ and 40 μm for glycerol (p ⩽ 0.05) after 60 min of treatment. The OCAs’ influence on the collagen hydration in the deep- located dermis was investigated. Both OCAs induce skin dehydration, but the effect of glycerol treatment (30 min and 60 min) is stronger. The obtained results demonstrate that with increasing the treatment time, both glycerol and Omnipaque™ solutions improve the optical clearing of porcine skin making the deep-located dermal regions able for investigations. At the used concentrations and time intervals, glycerol is more effective than Omnipaque™. However, Omnipaque™ is more promising than glycerol for future in vivo applications as it is an already approved pharmaceutic substance without any known impact on the skin structure. 16

17. The results of presented studies has been published in the next articles:
Sdobnov, A. Y., Tuchin, V. V., Lademann, J., & Darvin, M. E. (2017). Confocal Raman microscopy supported by optical clearing treatment of the skin- influence on collagen hydration. JOURNAL OF PHYSICS D-APPLIED PHYSICS, 50(28).
Sdobnov, A., Darvin, M. E., Lademann, J., & Tuchin, V. (2017). A comparative study of ex vivo skin optical clearing using two‐photon microscopy. Journal of Biophotonics. 17

18. Thank you for your attention18