Comparing the SHG image formation for thin sections of tail tendon fascia and skeletal muscle fascia, we demonstrate that the forward (F) and backward (B) SHG images are vastly different. In addition, despite the different arrangement of the collagen Type I fibrillar architecture forming these two fascias, their ratios of forward over backward signal (F/B) are nearly equal. SHG images of thick tissue blocks of the fascia-muscle unit show only backward features, as opposed to SHG images of tissue blocks of the fascia-tendon unit. These images are an amalgamation of forward and backward features due to the backscattering of forward components within tendon. These forward features disappear when this tissue block is immersed in glycerol as backscattering is hereby suppressed.
©2007 Optical Society of America
Multiphoton laser scanning microscopy  is an established and powerful technique for cell and thick tissue imaging [2,3]. Generally, multiphoton microscopy uses a femtosecond near-infrared oscillator laser source which is tightly focused and scanned across a biological sample containing, whether natural or by extrinsic labeling [4-6]. Because the interaction of the laser with the sample is nonlinear, fluorescence is only produced at the focus, and therefore inherently providing sub-micron spatial resolution .
Multiphoton fluorescence microscopy can elegantly be combined with second harmonic generation (SHG) imaging providing unique but complimentary information about biological structures of noncentrosymmetric molecular organization [4,8-10]. The application of SHG in microscopy has been successfully used to image structural protein arrays, such as collagen Type I and –III [11-13], myosin [14-16] and tubulin  scaffolds. Unlike fluorescence, SHG is a coherent nonlinear optical process involving no electronic excitation . Due to the coherent nature of the SHG process, light is generated anisotropically [4, 12, 19-20]. The signal generated in the forward direction is stronger than the backward signal. In addition, for collagen arrays, recent studies have shown that forward and backward images contain vastly different features [12, 21].
In this study, we compare the SHG image formation for two fascia types, i.e. the tail tendon fascia and the fascia of skeletal muscle, as well as the corresponding ex vivo thick tissue blocks with their fascias attached. We analyze the thin as well as thick fascia ex vivo tissues and determine the impact of backscattering for the SHG image formation.
2. Experimental methods
2.1 Tissue preparation
Using a Nikon dissecting scope, the skin was removed from the hind legs and the tail of C57/B6 mice. Thick tissue blocks of fascia-muscle and fascia-tail tendon were harvested. The thick tissue blocks were dissected such that the first tissue structure in contact with the coversplip is fascia. For thin samples, the fascia tissues were gained from the same thick tissue blocks. The thin sections were transferred onto coverslips (VWR International, West Chester, USA; 24mmX60mm, No.1) treated with 3-Aminopropyltriethoxysilane (APES) or gelatin-chromium potassium sulfate solution (Gelatin type A- Sigma®, Chromium potassium sulfate, Sigma) for optimal tissue adhesion.
2.2. Imaging system
The laser system, a Ti:Sa oscillator laser (∼140 fs pulse, 76 MHz repetition rate; MIRA 900-F; Coherent radiation, Santa Clara CA) pumped with a Verdi 10W (Coherent radiation, Santa Clara CA), and set at 900 nm wavelength (measured with Wave Scan from APE GmbH, Berlin, Germany). The laser beam exiting the Ti:Sa oscillator is deflected into the NIR port of an inverted Axiovert 200M microscope (Carl Zeiss MicroImaging, Inc., Thornwood, USA), scanned across the sample with the LSM 510 scanning module (Carl Zeiss), and focussed into the sample with a 20X dry objective, plan-apochromat, NA=0.75 (Carl Zeiss).
SHG imaging is performed with 1x physiological salt solution (PBS) added to the sample to prevent drying artifacts. For backward detection mode, the signal is deflected towards the PMT (photomultiplier tube, NDD detector, Carl Zeiss) using a dichroic mirror (long pass 700 nm, Chroma Technology Corp). For the forward detection mode, the signal is collected with a 40X water immersion objective, C-achroplan NIR, NA=0.8 (Carl Zeiss). This setup is virtually identical to the one used by Williams et al. . In forward as well as backward, the SHG signal is filtered with a bandpass filter 445±15 nm combined with a 2 mm thick BG39 (Chroma Technology Corp., Rockingham, USA).
The SHG images presented have 1024×1024 pixels (92.1 μm×92.1 μm), the scanning time per pixel is 1.6 μs and each line is scanned 16 times. For the background, an image is taken with no sample in focus and without changing the laser parameters. The detector gain is set constant for all our measurements. The laser power is adjusted to avoid saturation of the PMT signal. In addition to the SHG image features, we discuss two signal ratios; F/B and T/B. Using ImageJ software , we measured the average pixel signal in forward (F), backward (B) and for thick tissue block (T) with the average background signal being subtracted (average over five samples). The different signals are calibrated with the square of the laser power. The efficiency ratio between forward and backward detection is obtained using multiphoton fluorescence from coumarin-440. With the assumption that on average, the forward and direct backward signals generated in thin tissue sections equal the signals generated in thick tissue samples, we use the F/B and T/B ratios to determine the contribution of forward signal to the total backward SHG signal from thick fascia tissue blocks.
We focus in this study on SHG imaging of fascia, a sheet like sheath of connective tissue surrounding and connecting organs and tissue groups, such as muscle and tail tendon. Fascia, as one of the four dense connective tissue types, is comprised of highly organized collagen Type I fibres as its main matrix element. Using SHG microscopy, collagen Type I fibers can be imaged elegantly and without the addition of external fluorochromes.
3.1 Imaging thin section of fascia
In Fig. 1, we present SHG forward and backward images from thin (thickness ∼10μm) fascia explants; fascia from skeletal muscle (muscle-fascia) [Figs. 1(a) and 1(b)] and fascia that surrounds tail tendon (tendon-fascia) [Figs. 1(c) and 1(d)]. The observed features of the SHG images obtained from the forward detection scheme [Figs. 1(a) and 1(c)] confirm the histological characteristics; for muscle-fascia, we observe a continuous sheet of parallel collagen Type I fibrils and for tail tendon fascia, we observe a crisscross organization of collagen fibrils. In backward detection, the overall sheet-like collagen architecture remains. However, instead of the continuous fibrillar arrangement, heterogeneous sub-micron features predominate. For tendon-fascia, these heterogeneous features in the backward detected image barely show the crisscross organization of fibrillar collagen. For muscle-fascia, the backward collected image clearly reveals the boundaries of the sheet like organization of the parallel oriented collagen strands.
The striking difference between the features observed in forward and backward detection is a consequence of the coherent nature of the second harmonic generation process. For the SHG process, the generated electromagnetic components are oscillating at twice the phase of the fundamental laser field. In SHG microscopy, those components are produced within the focal volume and add coherently. Within this volume, the architecture of collagen is not uniform and not continuous [23, 24]. Therefore, the phase of the SHG optical field is varying across the images, in forward as well as in backward. This can be measured by performing interferometry with a reference SHG optical field for each focal position within the image .
In a first approximation neglecting the Gouy phase shift, in the forward direction, due to the wavevector mismatch ∣Δk⃗∣ ≈ 0 and the short interaction length (near 2 microns for an objective NA of 0.75), the SHG components generated within the focal volume add constructively, revealing the fibrillar nature of the collagen architecture of fascia [see Figs. 1(a) and 1(c)]. Including the Gouy phase shift to the forward direction for a non-uniform and discontinuous distribution of scatterers, will mostly affects the shape of the SHG radiation pattern, as demonstrated by Moreaux et al. .
For backward SHG, the situation is very different. The intricate sub-micron architecture of fascia collagen arrays can be viewed as a complex interferometer. Due to the coherent nature of the SHG process, depending on the distribution and distance between the scatterers, the interference can change from constructive to destructive. The strength of the detected optical signal is modulated by this complex interference, explaining the heterogeneous sub-micron features seen in Figs. 1(b) and 1(d). These features are intrinsically linked to the complex sub-micron architecture of the fascia collagen matrix.
The coherent nature of the SHG process not only leads to vastly different features between forward and backward images but also to a stronger signal in the forward direction versus the backward direction. For muscle-fascia, we measure a ratio F/B = 4.0 and for tendon-fascia, F/B = 4.6 (for both the error bar is ±0.2). Despite the different histological arrangement, for muscle-fascia, a parallel sheet organization, and for fascia-tendon a crisscross organization, their F/B ratios are very close. Since F/B is directly linked to the distribution of scatterers within the focal volume [12, 26], our observations suggest that the sub-micron collagen architecture and collagen density of fascia-muscle is similar to fascia-tendon despite the different overall histological appearance and arrangement of collagen fibers.
3.2 Imaging fascia of thick ex vivo tissue blocks and the impact of glycerol
Thick tissue imaging requires light collection in the backward detection mode. Therefore, the total backward signal is an amalgamation of the direct backward signal plus a certain fraction of the backscattered forward signal that contributes to the total signal [27, 28]. To investigate the significance of backscattering for fascia imaging is highly important as these observed phenomena apply to any thick or in vivo tissue imaging techniques that rely on coherent nonlinear optical processes. Fascia by itself cannot provide significant backscattering due to its thinness (typically ∼10 microns). As mentioned above, fascia is a thin layer made primarily of collagen Type I molecules that surround organs and tissue structures. Therefore, backscattering of SHG forward components depends strictly on the linear optical properties of the tissue immediately following fascia.
In Fig. 2, we present SHG images of fascia still attached to the excised thick tissue blocks (∼5mm); in Fig. 2(a) the fascia-muscle tissue block and in Fig. 2(b) the fascia-tendon tissue block. The fascia SHG image of the ex-vivo whole fascia-tendon tissue unit clearly shows a combination of forward [see Fig. 1(c)] and backward features [see Fig. 1(d)] in contrast to the fascia image of the fascia-muscle unit that contains only backward features [see Fig. 1(b)]. These results clearly indicate the impact of backscattering for imaging fascia, and hence other tissues using second-harmonic generation microscopy. The results further show that the contribution of backscattering strongly depends on the surrounding tissue. It is possible to separate the backward features from the forward features by performing interferometry with a reference SHG optical field at each focal position within the image as the phase of the forward SHG optical field is scrambled by the multiple scattering events.
To quantify our observations, we measure the ratio of the fascia SHG signal from the thick tissue unit and the thin fascia tissue section only; the T/B ratio. Since the backward signal from a thick tissue block is the summation of direct backward signal plus a fraction of the backscattered forward signal, we can use the F/B and T/B ratios to calculate the fraction of forward signal that contributes to the total collected backward signal. For the fascia-muscle tissue unit, we measure T/B = 1.0 ± 0.2, indicating that the forward signal does not contribute to the total backward signal. For the fascia-tendon tissue block, with T/B = 2.2 ± 0.4 and F/B = 4.6 ± 0.2, we calculate that near 25% of the total forward signal is contributing to the total backward signal due to backscattering within the tendon tissue block, representing more than half of the total backward signal.
This study demonstrates that for SHG imaging of fascia within thick tissue blocks, the observed SHG image features strongly depend on the linear optical properties of the immediate tissue environment as fascia by itself, ∼10 μm of thickness, cannot provide efficient backscattering. To explain our experimental results, we consult light scattering studies for similar tissues, i.e. bovine tendon  and chicken breast tissue for muscle . At 450 nm, for tendon, the reduced scattering coefficient, μs’∼100cm-1, and the absorption coefficient, μa∼1cm-1. For muscle, the coefficients are; μs’∼4.5cm-1 and μa∼1.5cm-1. Qualitatively, our experimental observations can be explained by comparing those coefficients. Absorption coefficients for tendon and muscle are on the same order of magnitude, however, for tendon, μs’ is much larger than μa. This is in stark contrast to muscle. In tendon therefore, light travel within the tissue is dominated by scattering. This explains the strong contribution of backscattered forward components that contribute to the thick fascia-tendon tissue image .
Since backscattering of forward components within the tendon tissue block is a major contributor for the fascia SHG image of the fascia-tendon tissue unit, we investigate the impact of glycerol on the SHG image formation. Glycerol greatly reduces tissue scattering . In Fig. 3, we show the resulting SHG image after immersing the fascia-tendon tissue block in glycerol. Forward features are absent in the image since backscattering is greatly reduced by immersing the thick tissue block in glycerol. Under glycerol treatment, the T/B ratio drops to 1.2 ± 0.2. Immersing thin tissue sections in glycerol shows no impact on the F/B ratio. Therefore, with F/B = 4.6, we estimate that the fraction of the forward signal that contributes to the total backward signal has dropped to 4%. Since F/B for fascia is small, 4% of the forward signal has no significant impact on the SHG image features. For tissue with a higher F/B ratio, 4% will significantly contribute to the image formation.
Due to the coherent nature of the SHG process, the forward signal is stronger than the backward signal. In addition, the image features of the forward detection scheme are vastly different and show the fibrillar nature of fascia collagen matrix, in stark contrast to the backward features which reflect more the sub-micron architecture. The F/B ratios for the two fascias investigated do not appear to depend on the macroscopic arrangement of the collagen fiber matrix given that both fascias have the same thickness (10μm). In addition, for both fascias, backward imaging of thin sections exhibits very similar sub-micron features. Finally, this study shows that for SHG imaging of fascia within thick tissue blocks, the observed SHG image features depend strongly on the immediate tissue environment.
This work was supported by National Institutes of Health grants R01 AR 36819 and R21 AR053143. Dr. François Légaré acknowledges the financial support from INRS-EMT and the Canada’s Natural Science and Engineering Research Council. We thank Dr. William Stoothoff and Brian Bacskai at the Harvard Martinos center for providing the condenser. We thank the Center for Nanoscale Systems CNS) and Dr. Martin Vogel, Harvard University for providing the imaging facility.
References and links
02. V. E. Centonze and J. G. White, “Multiphoton excitation provides optical sections from deeper within scattering specimens than confocal imaging,” Biophys. J. 75, 2015–2024 (1998). [CrossRef] [PubMed]
03. F. Helmchen and W. Denk, “Deep tissue two-photon microscopy,” Nat. Meth. 2, 932–940 (2005). [CrossRef]
04. W. R. Zipfel, R. M. Williams, R. Christie, A. Y. Nikitin, B. T. Hyman, and W. W. Webb, “Live tissue intrinsic emission microscopy using multiphoton-excited native fluorescence and second harmonic generation,” Proc. Natl. Acad. Sci. USA 100, 7075–7080 (2003). [CrossRef] [PubMed]
05. C. Stociek, O. Garaschuk, K. Holthoff, and A. Konnerth, “In vivo two-photon calcium imaging of neuronal networks,” Proc. Natl. Acad. Sci. USA 100, 7319–7324 (2003). [CrossRef]
07. K. W. Dunn and P. A. Young, “Principles of multiphoton microscopy,” Nephron Exp Nephrol. 103, 33–40 (2006). [CrossRef]
08. S. Roth and I. Freund, “Second harmonic generation in collagen,” J. Chem. Phys. 70, 1637–1643 (1979). [CrossRef]
09. P. J. Campagnola and L. M. Loew, “Second-harmonic imaging microscopy for visualizing biomolecular arrays in cells, tissues and organisms,” Nat. Biotechnol. 21, 1356–1360 (2003). [CrossRef] [PubMed]
10. A. Zoumi, A. Yeh, and B. J. Tromberg, “Imaging cells and extracellular matrix in vivo by using second-harmonic generation and two-photon excited fluorescence,” Proc. Natl. Acad. Sci. USA 99, 11014–11019 (2002). [CrossRef] [PubMed]
11. E. Brown, T. McKee, E. diTomaso, A. Pluen, B. Seed, Y. Boucher, and R.K. Jain, “Dynamic imaging of collagen and its modulation in tumors in vivo using second-harmonic generation,” Nat. Med. 9, 796–800 (2003). [CrossRef] [PubMed]
12. R. M. Williams, W. R. Zipfel, and W. W. Webb, “Interpreting second-harmonic generation images of collagen I fibrils,” Biophys. J. 88, 1377–1386 (2005). [CrossRef]
13. S. Zhuo, J. Chen, T. Luo, and D. Zou, “Multimode nonlinear optical imaging of the dermis in ex vivo human skin based on the combination of multichannel mode and Lambda mode,” Opt. Express 14, 7810–7820 (2006). [CrossRef] [PubMed]
14. T. Boulesteix, E. Beaurepaire, M.-P. Sauviat, and M.-C. Schanne-Klein, “Second-harmonic microscopy of unstained living cardiac myocytes: measurements of sarcomere length with 20-nm accuracy,” Opt. Lett. 29, 2031–2033 (2004). [CrossRef] [PubMed]
15. S.-W. Chu, S.-Y. Chen, G.-W. Chern, T. H. Tsai, Y. C. Chen, B.-L. Lin, and C.-K. Sun, “Studies of χ(2)/χ(3) tensors in submicron-scaled bio-tissues by polarization harmonics optical microscopy,” Biophys. J. 86, 3914–3922 (2004). [CrossRef] [PubMed]
16. S. V. Plotnikov, A. C. Millard, P. J. Campagnola, and W. A. Mohler, “Characterization of the myosin-based source for second-harmonic generation from muscle sarcomeres”, Biophys. J. 90, 693–703 (2006). [CrossRef]
17. P. J. Campagnola, A. C. Millard, M. Terasaki, P. E. Hoppe, C. J. Malone, and W. A. Mohler, “Three-dimensional high-resolution second-harmonic generation imaging of endogenous structural proteins in biological tissues,” Biophys. J. 82, 493–508 (2002). [CrossRef]
18. R. W. Boyd, Nonlinear optics (Academic Press, London, 2003), pp. 87–99.
19. L. Moreaux, O. Sandre, M. Blanchard-Desce, and J. Mertz, “Membrane imaging by simultaneous second-harmonic generation and two-photon microscopy,” Opt. Lett. 25, 320–322 (2000). [CrossRef]
21. T. A. Theodossiou, C. Thrasivoulou, C. Ekwobi, and D. L. Becker, “Second harmonic generation confocal microscopy of collagen type I from rat tendon cryosection,” Biophys. J. 91, 4665–4677 (2006). [CrossRef] [PubMed]
22. W. S. Rasband ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, h7ttp://rsb.info.nih.gov/ij/ (1997–2007).
23. P. P. Provenzano and R. Jr. Vanderby, “Collagen fibril morphology and organization: implications for force transmission in ligaments and tendon,” Matrix Biol. 25, 71–84 (2006). [CrossRef]
26. J. Mertz and L. Moreaux, “Second-harmonic generation by focused excitation of inhomogeneously distributed scatterers,” Opt. Comm. 196, 325–330 (2001). [CrossRef]
27. D. Débarre, W. Suppato, A. M. Pena, A. Fabre, T. Tordjmann, L. Combettes, M.-C. Schanne-Klein, and E. Beaurepaire, “Imaging lipid bodies in cells and tissues using third-harmonic generation microscopy,” Nat. Meth. 3, 47–53 (2006). [CrossRef]
28. C. L. Evans, E. O. Potma, M. Puoris’haag, D. Coté, C. P. Lin, and X. S. Xie, “Chemical imaging of tissue in vivo with video-rate coherent anti-Stokes Raman scattering microscopy,” Proc. Natl. Acad. Sci. USA 102, 16807–16812 (2005). [CrossRef] [PubMed]
29. B. Chan, C. Amann, A. Yaroslavsky, C. Title, D. Smink, B. Zarins, I. Kochevar, and R. Redmond, “Photochemical repair of Achilles tendon rupture in a rat model,” J. Surg. Res. 124, 274–279 (2005). [CrossRef] [PubMed]
30. G. Marquez, L. V. Wang, S.-P. Lin, J. A. Schwartz, and S. L. Thomsen, “Anisotropy in the absorption and scattering spectra of chicken breast tissue,” App. Opt. 37, 798–804 (1998). [CrossRef]
31. I. S. Saidi, S. L. Jacques, and F. K. Tittel, “Mie and Rayleigh modeling of visible-light scattering in neonatal skin,” App. Opt. 34, 7410–7418 (1995). [CrossRef]