Skip to main content
NCBI home page
iScience logoLink to iScience
. 2019 Aug 29;19:965–975. doi: 10.1016/j.isci.2019.08.046

Geometry-Dependent Spectroscopic Contrast in Deep Tissues

Xin Ge 1,11, Hongying Tang 2,11, Xianghong Wang 1,11, Xinyu Liu 1, Si Chen 1, Nanshuo Wang 1, Guangming Ni 3, Xiaojun Yu 4, Shufen Chen 1, Haitao Liang 1, En Bo 1, Lulu Wang 1, Cilwyn Shalitha Braganza 1, Chenjie Xu 5, Steven M Rowe 6,7,∗, Guillermo J Tearney 8,9,10,∗∗, Linbo Liu 1,5,12,∗∗∗
PMCID: PMC6745491  PMID: 31522119

Summary

Nano-structures of biological systems can produce diverse spectroscopic effects through interactions with broadband light. Although structured coloration at the surface has been extensively studied, natural spectroscopic contrasts in deep tissues are poorly understood, which may carry valuable information for evaluating the anatomy and function of biological systems. Here we investigated the spectroscopic characteristics of an important geometry in deep tissues at the nanometer scale: packed nano-cylinders, in the near-infrared window, numerically predicted and experimentally proved that transversely oriented and regularly arranged nano-cylinders could selectively backscatter light of the long wavelengths. Notably, we found that the spectroscopic contrast of nanoscale fibrous structures was sensitive to the pressure load, possibly owing to the changes in the orientation, the degree of alignment, and the spacing. To explore the underlying physical basis, we further developed an analytical model based on the radial distribution function in terms of their radius, refractive index, and spatial distribution.

Subject Areas: Infrared Optics, Medical Imaging, Optical Imaging, Spectroscopy

Graphical Abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    Long-wavelength shifts of packed nano-cylinders in the backscattered NIR spectra

  • •

    Experimental observations of characterized spectroscopic contrast in deep tissue

  • •

    Physics behind is described by an analytical model using the radial distribution function

  • •

    Spectroscopic contrast of fibrous structures is sensitive to the external pressure

Infrared Optics; Medical Imaging; Optical Imaging; Spectroscopy

Introduction

In contrast to that produced by selective absorption, the spectral modulation of light reflected or scattered by a non-luminous object is mainly due to elastic interactions between the input light and micro- and nano-structures. This spectral modulation may bring about spectroscopic contrasts manifested as spectral centroid shifts in the back-scattered light fields. Since Robert Hooke and Isaac Newton revealed the basis of structural coloration (Hooke, 1665, Newton, 1704), fascinating colors created by natural structures have attracted considerable research interests (Vukusic and Sambles, 2003, Wiersma, 2013). In biology, structural colors are commonly observed under sunlight or white light on the surfaces of animals and plants (Cuthill et al., 2017, Moyroud et al., 2017, Teyssier et al., 2015, Vignolini et al., 2012) (skin, feathers, flowers, and epicarp). However, as turbid tissues are opaque to visible light, the direct observation of structural colors originating from beneath biological tissues is extremely difficult.

The spectroscopic properties of deep biological tissues are commonly investigated within the near-infrared (NIR) window (650–1,350 nm), where the light has its maximum penetration. For more than a decade, an NIR reflectance microscopic technique termed spectroscopic optical coherence tomography (SOCT) has been developed (Leitgeb et al., 2000, Morgner et al., 2000, Oldenburg et al., 2007, Xu et al., 2006). SOCT detects the light signals from a micrometer-scale sample volume while rejecting the light from the background, a unique capability known as optical sectioning, which makes it possible to probe microstructures through 1- to 2-mm opaque tissues. The effect of scatterer size on the back-scattered spectra has led to the use of SOCT to evaluate the cell nuclei (Robles and Wax, 2010, Robles et al., 2010, Wax et al., 2011, Xu et al., 2005) and probe nanoscale information (Azarin et al., 2015, Yi et al., 2013). The spectra of random microspheres packing have been investigated extensively (Tseng et al., 2006); detailed understanding of nanoscale shape on spectroscopic contrast, however, remains elusive.

At the nanometer level, the two fundamental geometries of biological tissues are spherical geometry, such as mitochondria and various intracellular granules, and cylindrical geometry, such as motile cilia, collagen fibrils, and elastic fibers. It is intriguing to know how different nanoscale geometries may alter the spectrum of scattered light and whether it is possible to extract biologically relevant information from their spectroscopic signatures. The answers to these questions are pertinent to a variety of clinical and scientific research fields where optical reflectance imaging techniques are widely used, such as noninvasive anatomical and functional imaging of microstructures in the respiratory mucosa, the blood vessel wall, the posterior segment of the eye, the skin, and the gastrointestinal mucosa (Fujimoto, 2003, Huang et al., 1991, Liu et al., 2011, Tearney et al., 1997, Yelin et al., 2006, Yun et al., 2006), nanoscale mapping of nuclear architecture (Uttam et al., 2015, Wang et al., 2010), or changing of partial wave spectroscopy signals by chromosome condensation (Kim et al., 2011). In this study, we established numerical models of nano-spheres and nano-cylinders and discovered that transversely oriented and regularly arranged nano-cylindrical scatterers are more likely to generate spectral centroid shifts toward the long wavelengths within the spectral window of 700–950 nm than nano-spheres, which tend to exhibit shifts toward the short end. Using a form of SOCT that operates in 700–950 nm, we verified the above-mentioned predictions on a variety of natural nano-cylindrical structures, such as motile cilia and extracellular matrix containing nanoscale fibrous structures, which exhibited a striking and consistent spectroscopic contrast against the surrounding background tissues. To our knowledge, this study is the first to demonstrate geometry-dependent spectroscopic contrasts from nanometer-scale features in deep mammalian tissues in situ and in vivo. Interestingly, we additionally found that this geometry-dependent spectroscopic contrast is sensitive to external stress, probably due to the changes in the orientation, degree of alignment, and spacing of the nano-cylinders.

Results

We generated numerical models of packed nano-spheres and nano-cylinders in a micrometer-scale focal volume using the finite-difference time-domain (FDTD) method and compared their backscattered spectra detected by a single mode pinhole (Figure 1A). The details on generating packed scatterers and the computational imaging model are described in the Transparent Methods. Our SOCT system obtains cross-sectional images at a rate of up to 60 frames per second, enabling acquisition of three-dimensional (3D) images of biological tissues in situ and in vivo. The spatial resolution of the intensity image is 1.80 μm × 1.80 μm × 1.28 μm (x, y, z) in tissue, which allowed us to identify microscopic anatomic details to match SOCT images with corresponding histology.

Figure 1.

Figure 1

Open in a new tab

FDTD Simulation Reveals Characteristic Spectroscopic Contrast of Motile Cilia

(A) Schematic of light-tissue interactions in the respiratory mucosa. Color bar: light red and dark red color stand for short and long wavelengths in 700–950 nm, respectively.

(B) Backscattered spectra of transversely oriented and regularly arranged nano-cylinders simulating the apical portions of motile cilia at the recovery stroke. These transversely oriented and regularly arranged nano-cylinders were generated by the same strategy but the different seeds.

(C) Backscattered spectra of nano-spheres of the same diameter and refractive index as the cylinders in (B). Each curve in (B) and (C) was normalized to its maximum intensity. Curves with the spectral centroid shift toward the long wavelengths and short wavelengths were denoted with green and red color, respectively. Those without significant shift were denoted with orange color. λc is the center of gravity of the backscattered spectrum, and 825 nm is the center of gravity of the input light spectrum.

Motile Cilia

The motile cilia that line the surface of mammalian respiratory airways, brain ventricles, and fallopian tubes are regularly arranged nano-cylinders. Those in the respiratory airways are ideal samples for our study because they are surrounded by aqueous periciliary liquid in a consistent mucin gel instead of an optically heterogeneous matrix with unknown optical properties (Barrick et al., 2016, Liu et al., 2013, Oldenburg et al., 2012, Sanderson and Sleigh, 1981). According to the electron microscope and digital high-speed imaging studies (Chilvers and O'callaghan, 2000, Sanderson and Sleigh, 1981), at the recovery stroke the motile cilia bend so that the apical portions of cilia are approximately parallel to the cell surface (Figure 1A), whereas at the midpoint of the effective (or power) stroke they are almost straight and perpendicular to the cell surface and stand up above those in the recovery stroke. Therefore, in optical coherence tomography (OCT) images, the scattered signals within 3–5 μm from the apical cell surface are mostly from the cilia in the recovery stroke, whereas those signals within 5–7 μm from the apical cell surface originate from the cilia tips in the effective stroke (Sanderson and Sleigh, 1981).

In our numerical model, we first considered those in the recovery stroke when the closely packed apical portions of the cilia are parallel to the airway surface and perpendicular to the input light beam (Figure 1A and Table S1) (Liu et al., 2013, Sanderson and Sleigh, 1981). The backscattered spectra of 10 groups of packed cylinders mimicking motile cilia exhibit spectral centroid shifts of 23.7 ± 26.6 nm (mean ± SD) toward the long wavelengths (Figures 1B and S1). In contrast, those from spherical scatterers with the same refractive index, diameter, and spatial arrangements tend to shift 16.9 ± 28.7 nm toward the short wavelengths (Figure 1C). Second, we considered those in the effective stroke when the orientation of the cilia varies within 0°–60° to the input beam. Our model predicted that the spectral centroid is more likely to shift toward the short wavelengths during the effective stroke (Figure S2).

We imaged normal ciliated primary human bronchial epithelial (HBE) cultures in vitro and intact sheep trachea ex vivo with functional mucociliary transport using SOCT along the direction of the effective stroke or the mucus transport. The intensity image captured the cross-sectional micro-anatomy of the mucociliary apparatus, including the airway surface liquid (phosphate-buffered saline in the cell culture), cilia, and cell bodies (Figures 2A and 2D). In the cell cultures with mucus replaced by phosphate-buffered saline, cilia in the recovery stroke and cilia in the effective stroke could be clearly differentiated (Figure 2A and Video S1). To visualize the contrast of the spectral centroid shift, we mapped it to grayscale and produced an image we term the spectral centroid shift image (Figures 2B and 2E). In the cell cultures, we observed a mean spectral centroid shift of 20.4 ± 6.3 nm toward the long wavelengths in the back-scattered signals from regions consisting predominantly of cilia in the recovery stroke, whereas the mean spectral centroid shift of signals predominantly from HBE cell bodies, presumably containing more spherical scatterers, shifted 3.3 ± 7.0 nm toward the short wavelengths (Figure 2B and inset). In addition, signals from the effective stroke exhibited a spectral centroid shift of 1.5 ± 8.4 nm toward the short wavelengths (Figure 2B and inset), which is significantly different from that of the recovery stroke (paired signals from the recovery and effective stroke in 50 ciliated cells, p < 0.001 at the 0.05 level, Student's t test). These observations agree with our predictions shown in Figure 1. We also generated a color image, termed SOCT image, by combining the spectral centroid shift (Hue) and the intensity image (Value) using the Hue-Saturation-Value (HSV) color scheme, in which the cilia in the recovery stroke appeared as greenish structures, distinct from the reddish effective stroke and epithelial cell body (Figure 2C). In the cultured full-thickness sheep tracheal mucosa, we could also detect the same spectroscopic contrast as the HBE culture (Figures 2E and 2F). Interestingly, the lamina propria of the sheep trachea demonstrates long-wavelength shifts suggesting spectroscopic contrasts from nanofibrous structures in the extracellular matrices (Figures 2E and 2F).

Figure 2.

Figure 2

Open in a new tab

Geometry-Dependent Spectroscopic Contrast Images of Motile Cilia In Vitro and In Vivo

(A) A representative OCT intensity image of motile cilia in HBE culture.

(B) The corresponding spectral centroid shift image.

(C) The corresponding HSV-mapped SOCT image. PBS, phosphate-buffered saline; EP, human bronchial epithelial monolayer; M, membrane of the Transwell insert; eff, effective stroke; re, recovery stroke.

(D) A representative OCT intensity image of cultured sheep tracheal mucosa with functional mucociliary clearance in situ. Red arrow stands for the direction of the effective stroke. Note that the epithelium folded in the left side, so that the epithelial surface was parallel to the image plane, which “cut” through motile cilia perpendicularly. The 3D spatial relation between the image plane and the folded epithelium is explained in Figures S3B and S3C. (E) and (F) The corresponding spectral centroid shift image and HSV-mapped SOCT image, respectively. EP, epithelium; LP, lamina propria; M, mucus flow. The mark at 825 nm in the gray scale bars refers to the centroid of the input spectrum. The green hue indicates the long-wavelength shift in NIR, whereas the red hue represents shifts toward the short wavelength for all HSV-mapped images in this paper. Scale bars: 10 μm.

Video S1. Time-Lapsed Spectral Centroid Images of Motile Cilia In Vitro, Related to Figure 2B

32 frames per second. Time in (sec: [1/60 s]).

Download video file (1.2MB, mp4)

Nanoscale Fibrous Structures in Extracellular Matrix

Fibrous structures in the extracellular matrix also assume the cylindrical geometry at the nanometer scale. For initial verification, we established our numerical model (Table S2) based on the collagen fibril organization in the lamina cribrosa and peripapillary sclera since they are simple in geometrical composition: their dry matter is mainly composed of transversely oriented collagen fibrils with respect to the input beam (Komai and Ushiki, 1991, Quigley et al., 1991). Similar to those from motile cilia, the backscattered spectra of 10 groups of packed cylinders mimicking collagen fibrils exhibit spectral centroid shift of 8.5 ± 8.1 nm (mean ± SD) toward the long wavelengths (Figure 3A). In contrast, those in mis-aligned configurations (assigning a random rotation angle around the z axis in [-π, π]) tend to be neutral in spectral centroid shift (Figure 3B). To validate the theoretical prediction, we acquired 3D images of the optic nerve head of freshly enucleated unpressurized swine eyes ex vivo (Figure 3C). Similar to motile cilia, they exhibited strong spectral centroid shifts toward the long wavelengths, whereas the prelamina region and blood vessel remained neutral or exhibited moderate shifts toward the short end (Figures 3D and 3E and Video S2). The SOCT image provides a color contrast similar to the histology image (Figure 3F).

Figure 3.

Figure 3

Open in a new tab

Geometry-Dependent Spectroscopic Scattering of Collagen Fibrils

(A) FDTD predictions of backscattered spectra from densely packed well-aligned nano-cylinders simulating collagen fibrils in lamina cribrosa and sclera.

(B) Backscattered spectra from mis-aligned nano-cylinders with the same diameter and refractive index as those in (A). Each spectrum was from a distinct spatial arrangement of cylinders generated using the same strategy.

(C) A representative intensity cross-sectional image of swine optic nerve head.

(D–F) Corresponding spectral centroid shift image (D), HSV-mapped SOCT image (E), and Masson's trichrome stained histology image (F), respectively. The spectroscopic artifacts at the top of (D) and (E) were the ghost image due to the autocorrelation artifacts of OCT signal. The cyan dashed boxes in (E) and (F) refer to the same region. LCR, lamina cribrosa; S, sclera; R, retina; PLR, prelaminar region; R, retina.

Scale bars: 100 μm.

Video S2. HSV-Coded 3D SOCT Images of Human Skin In Vivo, Related to Figure 3
Download video file (2MB, mp4)

We additionally acquired 3D images of fully expanded human skin in vivo, as well as fully expanded swine esophageal mucosa in vivo. The scatterers in the epidermis (Video S3, and Figure S4) and esophageal epithelium (Figure S5), which presumably contain more microscopic spheres, scattered more light of the short wavelengths than the long wavelengths. Although the compositions of the dermis of human skin and the lamina propria of swine esophageal mucosa are complex, there are abundant aligned nanoscale fibrous structures such as collagen fibrils and elastic fibers. The signals from the reticular dermis (Figures S4D–S4F) and esophageal lamina propria (Figure S5B) exhibited strong spectral centroid shifts toward the long wavelengths, presumably from nanoscale fibrous structures. It is noticed that the keratinized layer of swine esophageal epithelium presented spectral centroid shifts toward the long wavelengths, which was possibly generated by the well-aligned tonofilaments. Of note, in the NIR range, the absorption of water and major chromophores might have contributed to the spectroscopic contrast. However, the results from skin of different phenotypes (Figure S4) do not support a significant contribution from melanin.

Video S3. HSV-Coded 3D SOCT Images of Fresh Swine Optic Nerve head Ex Vivo, Related to Figure 3E
Download video file (7.7MB, mp4)

Thermal Denature of Collagen

The fact that collagen fibrils and fibers denature at high temperature (Abdullahi et al., 2014) provides a convenient model to confirm the geometry dependency of the spectroscopic contrast. To do so, we evaluated the spectroscopic signals from normal and burned ear skin of C57BL/6 mice ex vivo (Figure 4). Although there was no significant change in the intensity of the dermis caused by the thermal damage, the spectral centroid shift in the dermis and the articular cartilage was remarkably reduced in the burned sample (Figures 4E and 4F) compared with that of the normal control (Figures 4B and 4C), presumably due to the thermal damage in the nano-cylindrical structures.

Figure 4.

Figure 4

Open in a new tab

Geometry-Dependent Spectroscopic Contrast Is Negated by Thermal Denaturing of Collagen Fibrils Ex Vivo

(A–F) (A–C) are the representative intensity image, spectral centroid shift image, and HSV-mapped SOCT image from the normal control mouse ear, respectively, whereas (D–F) are those of the burned mouse ear, respectively. EP, epidermis; D, dermis; AC, auricular cartilage. Scale bars: 100 μm.

Stress-Induced Reorganization of Nanoscale Fibers

The above-mentioned static results may provide insights into spectroscopic phenomenon in varieties of complex biological tissues and processes. The following dynamic example is related with the tunica media of the coronary wall, where thin (actin) and thick (myosin) contractile filaments in the smooth muscle cells are stretched and elastic fibers uncoil as the vessel wall expands (Dingemans et al., 2000). When we loaded the coronary artery of a fresh swine heart with a linearly increasing static phosphate-buffered saline pressure from 44.1 to 117.7 mmHg over 10 s ex vivo, the mean spectral centroid shift of the tunica media increased by 14.3 nm toward the long wavelengths (Figures 5A–5D, and blue line in Figure 5E). Clearly, in this dynamic process the spectral centroid shift closely followed the change of the media thickness (red line in Figure 5E), which was correlated with the intracoronary pressure change. Therefore, it is reasonable to attribute the spectral curve shifts (Figure 5F) to the reorganization of the nano-cylinder assembly, which might undergo a transition from a loose and randomly orientated morphology to a straighter, more transversely oriented and more compact morphology with a higher degree of alignment due to the external load (Yang et al., 2015). Interestingly, the fibrous pericardium (FP) fused with the serous pericardium (SP) (Figures 5A and 5B) could be easily distinguished by the spectroscopic contrast of layers of fibrils and elastin fibers in the FP (Figures 5C and 5D).

Figure 5.

Figure 5

Open in a new tab

Changes of the Spectroscopic Contrast of the Swine Coronary Arterial Wall during Pressure Loading

(A) Representative cross-sectional intensity image acquired through the pericardium at the intracoronary hydrostatic pressure of 44.1 mm Hg. (B) Intensity image acquired in the same spot at the intracoronary hydrostatic pressure of 117.7 mm Hg. The bright interface at the bottom in (A) was from the lower surface of an air bubble, which moved out of the recorded view in (B) at high pressure (Video S4).

(C) Spectral centroid shift image corresponding to (A). (D) Spectral centroid shift image corresponding to (B). FP, fibrous pericardium; SP, serous pericardium; A, adventitia; M, media; I, intima. Scale bars: 100 μm.

(E) Dynamic correlation between the relative spectral centroid shift of the tunica media (blue line) and the media thickness reflecting the intracoronary pressure (red line).

(F) Spectral curves at the selected time points. Spectral centroid λc in (E and F) is extracted from the nonframe-averaged spectral centroid shift images.

Video S4. Time-Lapsed Spectral Centroid Images of Swine Coronary Artery Wall Subject to hydrostatic PBS Loading from 44.1 to 117.7 mm Hg Ex Vivo, Related to Figure 5C and 5D

50 frames per second. Time in (sec: [1/60 s]).

Download video file (3.4MB, mp4)

Discussion

Nanometer-scale fibrous structures are ubiquitous in nature. Although their cylindrical geometry is designed to fulfill particular biological functions, it also possesses a unique wavelength-selective backscattering property. The lack of specificity for tissue components and functional states has been the main shortcoming of optical reflectance techniques. The geometry-dependent spectroscopic contrasts may enable us to specifically distinguish nano-cylinders from spherical scatterers or even extract functional information of biological systems. First of all, this geometry-dependent spectroscopic contrast may be particularly useful for imaging applications where the sub-diffraction-limit structures and processes cannot be optimally visualized and differentiated. Analysis of ciliary beat pattern has been proposed as a powerful method to diagnose primary ciliary dyskinesia (PCD) (Chilvers and O'callaghan, 2000, Chilvers et al., 2003), since the current ciliary beat frequency method and ultrastructural analysis may not be accurate (Buchdahl et al., 1988, Rossman and Newhouse, 1988, Santamaria et al., 1999). The most common motility defect of PCD is the inability to bend along the axoneme (Chilvers and O'callaghan, 2000, Chilvers et al., 2003), which may be identified as the lack of the signature spectral centroid shift toward the long wavelengths from the bending cilia in the recovery stroke.

In addition, it is important to understand the effects of the degree of alignment on the spectroscopic contrast. In our simulations, the model collagen fibrils are well aligned and tightly packed and diffract light at roughly the same spacing (Table S2). If the orientation of individual fibrils was random, these mis-aligned nano-cylinders exhibited no significant collective spectral centroid shifts (Figure 3B), which is consistent with the experimental observations in the papillary dermis: loosely arranged collagen fibrils and elastic fibers with random orientations appeared to be neutral and heterogeneous in the spectral centroid shift (Figures S4D–S4F). Even under the aligned condition, the spectral centroid shift of nanoscale fibrous structures is dependent on their spacing (Figure S6). When we increased the interfibrillar spacing from 1D to 1.05D, the peak and valley shifted monotonically; further increase in the interfibrillar spacing may result in an inversion of spectral centroid shift.

There have been theoretical techniques for characterizing the light scattering of the biological tissue that were treated as either a continuously random medium (Rogers et al., 2009, Schmitt and Kumar, 1996, Sheppard, 2007, Xu and Alfano, 2005) or medium having spatial quasi-ordered distribution of scatterers (Hart and Farrell, 1969, Zimm, 1948). In the former treatment, the biological tissue is assumed as a homogeneous and isotropic medium, and a continuous refractive index (RI) fluctuation is dependent on the fractal dimension Df. Based on the previous measurements, the Df values ranged from 2.6 to 3.1 for different types of soft tissues (Schmitt and Kumar, 1996, Yi and Backman, 2012). Since the backscattering coefficient μb∝λDf−4 where a smaller Df corresponds to a smaller length scale RI variation (Yi et al., 2013), backscattered spectra contributed by these nano-structures in the Rayleigh regime must show a blue-shift phenomenon. Therefore, this treatment may not be applied to packed nano-cylinders. The latter treated the spatial inhomogeneity of fibrils as a quasi-crystalline arrangement, described by the radial distribution function. Taking into account the interference effects, the transparency of the cornea (Hart and Farrell, 1969) and the transmission property of the sclera (Tuchin et al., 1997) were well characterized. The latter treatment is more suitable for our study because the cylindrical center positions could be correlated over distances comparable with the wavelength so that the interference between the scattered fields should be accounted for. The radial distribution function is defined by the ratio of a local density of the scatterers' centers within a circular ring ranges from x to x + Δx, to the mean density of the scatterers' centers in that ring. The analytic form of the backscattering coefficient in the far field is given by:

μb=dcπr2∫π−θmaxπ+θmaxσ0(θ){1+2πρ∫0Lxdx[g(x)−1]J0[2k¯xsin(θ2)]}dθ (Equation 1)

where dc is the cross-sectional packing fraction of nano-cylinders, r is the radius of nano-cylinders, sinθmax = N.A. is the numerical aperture of the objective, σ0 is the backscattering cross section of a single nano-cylinder, k¯=2πn¯/λ, λ is the wavelength in the vacuum, m = nc/nm is the ratio of RI of the nano-cylinders nc to that of the surrounding medium nm, average RI n¯=dcnc+(1−dc)nm=nm[dc(m−1)], L should be larger than the correlation distance LR above which the g(x) is essentially unity, and J0 is the zero-order Bessel function of the first kind. Derivation of Equation 1 and variable analysis are provided in detail in the Transparent Methods. Although g(x) can be measured in simulations or electron micrographs, for the sake of analysis, we followed the approach (Yuste and Santos, 1993) that g(x) is generated by the packing fraction dc. Figure S7A shows the undulatory property of g(x) at dc = 0.3, dc = 0.5, and dc = 0.7. A larger dc value corresponds to a longer correlation distance LR of nano-cylinders’ centers. The first peak in g(x) indicates the diameter of nano-cylinders as they cannot overlap. From the Figures S7B–S7D, in terms of radius from 10 to 50 nm, the backscattered intensity always decays with λ when the correlation distance LR is short (i.e., dc is in small value). However, some of cancellation or upward slopes in the spectral curves appear when dc is larger, indicating the possibility of long-wavelength shifts caused by the constructive and destructive interference.

In our FDTD simulations, we were not able to establish theoretical models of all the above-mentioned complex biological systems owing to the lack of reliable data on the optical properties and spatial arrangements of nanoscale cylindrical structures. Even for the simplest systems such as motile cilia and lamina cribrosa/sclera, assumptions were made and much simplified models with estimated parameters were used. For example, as the refractive index of motile cilia and the surrounding liquid is unknown, we used estimated data reported in the previous literature (Schmitt and Kumar, 1998, Welch et al., 2005). Therefore, one should be careful in interpreting the data because the spectroscopic metric is cumulative as the random medium will also alter the spectrum, and interpretation of spectral data in anatomically complex scenarios are prone to the artifacts caused by the cumulative influence of random medium in the light path. These shortcomings of the theoretical approach are primarily compensated by careful selection of simple biological models and SOCT-based experimental verifications. Nevertheless, we tested numerically whether variation in the refractive index of cilia might alter the results. The centroid of the spectrum from a representative motile cilia model shifted 9.3 nm toward the long wavelengths caused by a difference in the refractive index from 1.40 to 1.60 (Figure S8), which indicates that the refractive index of the cilia may not be a critical factor for the spectroscopic contrast.

Unlike previous observations in mouse models that the spectral centroid of the normal dermis shifted to the short (Maher et al., 2014, Zhao et al., 2015), our results showed that the spectral centroid of mouse dermis shifts more toward the long wavelengths than the short. We postulated that it might be due to the differences in axial resolution, spectral data presentation and interpretation, and compositions of the dermis. First of all, Maher and Zhao's works reported the collective spectral change of the whole skin, where contributions from individual skin layers or components were not differentiated or interpreted in the spectral data (Zhao et al., 2015). Actually, with an axial resolution of 60 μm, collagen fibers cannot be differentiated from other tissue components. In contrast, our work aims to specifically investigate the spectral change of collagen fibers, which is the reason why we set the spatial resolution to ∼4 μm in tissue to differentiate the spectroscopic signals of different tissue components and collagen fibers of different orientation/packing state. Second, the compositions of the dermis are quite different between the previous works and our work. Maher and Zhao's works chose dorsal skin that contained a lot of subdermal adipose and muscle tissues. With an axial resolution of 60 μm, the spectroscopic signals from individual collagen fibers might have been averaged with those from other scatterers in the focal volume. Anyway, we can still find some remnant signals of long-wavelength-shifted spectral centroid in the spectroscopic image (Maher et al., 2014). In contrast, in our study we chose ear skin, which does not contain as many adipose and muscle tissues. Third, in our experiment the ear dermis was stretched horizontally by its cartilages, which served our purpose to have more regularly arranged collagen fibrils (Figure S9). We have no way to know if the dorsal skin was stretched or not during the experiments of the previous reports, but from the curvy tissue surfaces in OCT images (Maher et al., 2014) we can tell that they were not stretched as much as the ear dermis in our experiments (Figure 4).

In conclusion, we uncovered the geometry dependency of the reflectance spectroscopic signals of nanoscale scatterers and established a theoretical and experimental approach to understand and characterize the spectroscopic behaviors of mammalian deep tissue models containing nanoscale cylindrical structures. The signature spectral centroid shift from bending cilia may provide a complementary method to noninvasively detect the motility defect of cilia, thereby benefiting the diagnosis PCD. The correlation between the spectral centroid shift and stress may open up a new avenue to qualitatively or semi-quantitatively monitor the localized pressure of body fluid, in particular, the intravascular and intraocular pressure, for the study and diagnosis of vascular and ocular diseases.

Limitations of the Study

As mentioned in the discussion, approximations were made in the FDTD simulation and the theoretical derivation. We considered only the transverse orientation (perpendicular to the input beam) of nano-cylinders based on the fact that the backscattered intensity of the transversely oriented nano-cylinders was significantly greater than that of any other orientation (data not shown). The crimped or helical form of fibrils was approximated by parallel nano-cylinders in this study because the wavelength of a cylindrical helix (10–100 μm) is much larger than the resolution element (Freed and Doehring, 2005). Meanwhile, the absorption property of nanostructures should be taken into account in the absorption-dominant wavelength regions (Robles et al., 2011).

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Acknowledgments

This research was supported by Agency for Science, Technology and Research (A*STAR) under its Industrial Alignment Fund (Pre-positioning) (H17/01/a0/008), National Natural Science Foundation of China (Grant No. 61540042 & No. 61705184), Natural Science Basic Research Plan in Shaanxi Province of China (Grant No. 2018JQ6014), Ministry of Education - Singapore (MOE2013-T2-2-107 & RG 83/18 & RG 131/15), National Research Foundation Singapore (NRF-CRP13-2014-05), National Medical Research Council Singapore (NMRC/CBRG/0036/2013), NTU-Northwestern Institute for Nanomedicine, and the Fundamental Research Funds for the Central Universities (Grant No. G2018KY0308).

Author Contributions

X.G., S.M.R., G.J.T., and L.L. conceived and designed the research. X.G., H.T., X.W., S.C., and L.L. conducted experiments. X.G. conducted theoretical analysis and numerical simulation; X.L., X.W., and H.T. contributed to the SOCT algorithm; all authors contributed to the imaging studies and data analysis. X.G. and L.L. co-wrote the manuscripts. S.M.R., G.J.T., and L.L. supervised the overall project. All authors read and edited the manuscript.

Declaration of Interests

The authors declare no competing interests.

Published: September 27, 2019

Footnotes

Supplemental Information can be found online at https://doi.org/10.1016/j.isci.2019.08.046.

Contributor Information

Steven M. Rowe, Email: smrowe@uab.edu.

Guillermo J. Tearney, Email: gtearney@partners.org.

Linbo Liu, Email: liulinbo@ntu.edu.sg.

Supplemental Information

Document S1. Transparent Methods, Figures S1–S17, and Tables S1–S4
mmc1.pdf (9.6MB, pdf)

References

  1. Abdullahi A., Amini-Nik S., Jeschke M. Animal models in burn research. Cell. Mol. Life Sci. 2014;71:3241–3255. doi: 10.1007/s00018-014-1612-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Azarin S.M., Yi J., Gower R.M., Aguado B.A., Sullivan M.E., Goodman A.G., Jiang E.J., Rao S.S., Ren Y., Tucker S.L. In vivo capture and label-free detection of early metastatic cells. Nat. Commun. 2015;6:8094. doi: 10.1038/ncomms9094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barrick J., Doblas A., Gardner M.R., Sears P.R., Ostrowski L.E., Oldenburg A.L. High-speed and high-sensitivity parallel spectral-domain optical coherence tomography using a supercontinuum light source. Opt. Lett. 2016;41:5620–5623. doi: 10.1364/OL.41.005620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Buchdahl R., Reiser J., Ingram D., Rutman A., Cole P., Warner J. Ciliary abnormalities in respiratory disease. Arch. Dis. Child. 1988;63:238–243. doi: 10.1136/adc.63.3.238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chilvers M.A., O'callaghan C. Analysis of ciliary beat pattern and beat frequency using digital high speed imaging: comparison with the photomultiplier and photodiode methods. Thorax. 2000;55:314–317. doi: 10.1136/thorax.55.4.314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chilvers M.A., Rutman A., O'callaghan C. Ciliary beat pattern is associated with specific ultrastructural defects in primary ciliary dyskinesia. J. Allergy Clin. Immunol. Pract. 2003;112:518–524. doi: 10.1016/S0091-6749(03)01799-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cuthill I.C., Allen W.L., Arbuckle K., Caspers B., Chaplin G., Hauber M.E., Hill G.E., Jablonski N.G., Jiggins C.D., Kelber A. The biology of color. Science. 2017;357:eaan0221. doi: 10.1126/science.aan0221. [DOI] [PubMed] [Google Scholar]
  8. Dingemans K.P., Teeling P., Lagendijk J.H., Becker A.E. Extracellular matrix of the human aortic media: an ultrastructural histochemical and immunohistochemical study of the adult aortic media. Anat. Rec. 2000;258:1–14. doi: 10.1002/(SICI)1097-0185(20000101)258:1<1::AID-AR1>3.0.CO;2-7. [DOI] [PubMed] [Google Scholar]
  9. Freed A.D., Doehring T.C. Elastic model for crimped collagen fibrils. J. Biomech. Eng. 2005;127:587–593. doi: 10.1115/1.1934145. [DOI] [PubMed] [Google Scholar]
  10. Fujimoto J.G. Optical coherence tomography for ultrahigh resolution in vivo imaging. Nat. Biotechnol. 2003;21:1361. doi: 10.1038/nbt892. [DOI] [PubMed] [Google Scholar]
  11. Hart R.W., Farrell R.A. Light scattering in the cornea. JOSA. 1969;59:766–774. doi: 10.1364/josa.59.000766. [DOI] [PubMed] [Google Scholar]
  12. Hooke R. John Martyn and James Allestry; 1665. Micrographia: Or Some Physiological Descriptions of Minute Bodies Made by Magnifying Glasses, with Observations and Inquiries Thereupon. [Google Scholar]
  13. Huang D., Swanson E.A., Lin C.P., Schuman J.S., Stinson W.G., Chang W., Hee M.R., Flotte T., Gregory K., Puliafito C.A. Optical coherence tomography. Science. 1991;254:1178–1181. doi: 10.1126/science.1957169. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kim J.S., Pradhan P., Backman V., Szleifer I. The influence of chromosome density variations on the increase in nuclear disorder strength in carcinogenesis. Phys. Biol. 2011;8:015004. doi: 10.1088/1478-3975/8/1/015004. [DOI] [PubMed] [Google Scholar]
  15. Komai Y., Ushiki T. The three-dimensional organization of collagen fibrils in the human cornea and sclera. Investig. Ophthalmol. Vis. Sci. 1991;32:2244–2258. [PubMed] [Google Scholar]
  16. Leitgeb R., Wojtkowski M., Kowalczyk A., Hitzenberger C., Sticker M., Fercher A. Spectral measurement of absorption by spectroscopic frequency-domain optical coherence tomography. Opt. Lett. 2000;25:820–822. doi: 10.1364/ol.25.000820. [DOI] [PubMed] [Google Scholar]
  17. Liu L., Chu K.K., Houser G.H., Diephuis B.J., Li Y., Wilsterman E.J., Shastry S., Dierksen G., Birket S.E., Mazur M. Method for quantitative study of airway functional microanatomy using micro-optical coherence tomography. PLoS One. 2013;8:e54473. doi: 10.1371/journal.pone.0054473. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Liu L., Gardecki J.A., Nadkarni S.K., Toussaint J.D., Yagi Y., Bouma B.E., Tearney G.J. Imaging the subcellular structure of human coronary atherosclerosis using micro–optical coherence tomography. Nat. Med. 2011;17:1010. doi: 10.1038/nm.2409. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Maher J.R., Jaedicke V., Medina M., Levinson H., Selim M.A., Brown W.J., Wax A. In vivo analysis of burns in a mouse model using spectroscopic optical coherence tomography. Opt. Lett. 2014;39:5594–5597. doi: 10.1364/OL.39.005594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Morgner U., Drexler W., Kärtner F., Li X., Pitris C., Ippen E., Fujimoto J. Spectroscopic optical coherence tomography. Opt. Lett. 2000;25:111–113. doi: 10.1364/ol.25.000111. [DOI] [PubMed] [Google Scholar]
  21. Moyroud E., Wenzel T., Middleton R., Rudall P.J., Banks H., Reed A., Mellers G., Killoran P., Westwood M.M., Steiner U. Disorder in convergent floral nanostructures enhances signalling to bees. Nature. 2017;550:469. doi: 10.1038/nature24285. [DOI] [PubMed] [Google Scholar]
  22. Newton I. Reprinted by Dover Publications; 1704. Opticks, or, a Treatise of the Reflections, Refractions, Inflections & Colours of Light. [Google Scholar]
  23. Oldenburg A.L., Chhetri R.K., Hill D.B., Button B. Monitoring airway mucus flow and ciliary activity with optical coherence tomography. Biomed. Opt. Express. 2012;3:1978–1992. doi: 10.1364/BOE.3.001978. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Oldenburg A.L., Xu C., Boppart S.A. Spectroscopic optical coherence tomography and microscopy. IEEE J. Sel. Top. Quant. Electron. 2007;13:1629–1640. [Google Scholar]
  25. Quigley H.A., Dorman-Pease M.E., Brown A.E. Quantitative study of collagen and elastin of the optic nerve head and sclera in human and experimental monkey glaucoma. Curr. Eye Res. 1991;10:877–888. doi: 10.3109/02713689109013884. [DOI] [PubMed] [Google Scholar]
  26. Robles F.E., Wax A. Measuring morphological features using light-scattering spectroscopy and Fourier-domain low-coherence interferometry. Opt. Lett. 2010;35:360–362. doi: 10.1364/OL.35.000360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Robles F.E., Wilson C., Grant G., Wax A. Molecular imaging true-colour spectroscopic optical coherence tomography. Nat. Photonics. 2011;5:744–747. doi: 10.1038/nphoton.2011.257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Robles F.E., Zhu Y., Lee J., Sharma S., Wax A. Detection of early colorectal cancer development in the azoxymethane rat carcinogenesis model with Fourier domain low coherence interferometry. Biomed. Opt. Express. 2010;1:736–745. doi: 10.1364/BOE.1.000736. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rogers J.D., Çapoğlu İ.R., Backman V. Nonscalar elastic light scattering from continuous random media in the Born approximation. Opt. Lett. 2009;34:1891–1893. doi: 10.1364/ol.34.001891. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Rossman C.M., Newhouse M.T. Primary ciliary dyskinesia: evaluation and management. Pediatr. Pulmonol. 1988;5:36–50. doi: 10.1002/ppul.1950050109. [DOI] [PubMed] [Google Scholar]
  31. Sanderson M., Sleigh M. Ciliary activity of cultured rabbit tracheal epithelium: beat pattern and metachrony. J. Cell Sci. 1981;47:331–347. doi: 10.1242/jcs.47.1.331. [DOI] [PubMed] [Google Scholar]
  32. Santamaria F., De Santi M., Grillo G., Sarnelli P., Caterino M., Greco L. Ciliary motility at light microscopy: a screening technique for ciliary defects? Acta Paediatr. 1999;88:853–857. doi: 10.1080/08035259950168784. [DOI] [PubMed] [Google Scholar]
  33. Schmitt J., Kumar G. Turbulent nature of refractive-index variations in biological tissue. Opt. Lett. 1996;21:1310–1312. doi: 10.1364/ol.21.001310. [DOI] [PubMed] [Google Scholar]
  34. Schmitt J.M., Kumar G. Optical scattering properties of soft tissue: a discrete particle model. Appl. Opt. 1998;37:2788–2797. doi: 10.1364/ao.37.002788. [DOI] [PubMed] [Google Scholar]
  35. Sheppard C.J. Fractal model of light scattering in biological tissue and cells. Opt. Lett. 2007;32:142–144. doi: 10.1364/ol.32.000142. [DOI] [PubMed] [Google Scholar]
  36. Tearney G.J., Brezinski M.E., Bouma B.E., Boppart S.A., Pitris C., Southern J.F., Fujimoto J.G. In vivo endoscopic optical biopsy with optical coherence tomography. Science. 1997;276:2037–2039. doi: 10.1126/science.276.5321.2037. [DOI] [PubMed] [Google Scholar]
  37. Teyssier J., Saenko S.V., Van Der Marel D., Milinkovitch M.C. Photonic crystals cause active colour change in chameleons. Nat. Commun. 2015;6:6368. doi: 10.1038/ncomms7368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Tseng S.H., Taflove A., Maitland D., Backman V. Pseudospectral time domain simulations of multiple light scattering in three-dimensional macroscopic random media. Radio Sci. 2006;41:1–5. [Google Scholar]
  39. Tuchin V.V., Maksimova I.L., Zimnyakov D.A., Kon I.L., Mavlyutov A.H., Mishin A.A. Light propagation in tissues with controlled optical properties. J. Biomed. Opt. 1997;2:401–418. doi: 10.1117/12.281502. [DOI] [PubMed] [Google Scholar]
  40. Uttam S., Pham H.V., LaFace J., Leibowitz B., Yu J., Brand R.E., Hartman D.J., Liu Y. Early prediction of cancer progression by depth-resolved nanoscale mapping of nuclear architecture from unstained tissue specimens. Cancer Res. 2015;75:4718–4727. doi: 10.1158/0008-5472.CAN-15-1274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Vignolini S., Rudall P.J., Rowland A.V., Reed A., Moyroud E., Faden R.B., Baumberg J.J., Glover B.J., Steiner U. Pointillist structural color in Pollia fruit. Proc. Natl. Acad. Sci. U S A. 2012;109:15712–15715. doi: 10.1073/pnas.1210105109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Vukusic P., Sambles J.R. Photonic structures in biology. Nature. 2003;424:852–855. doi: 10.1038/nature01941. [DOI] [PubMed] [Google Scholar]
  43. Wang P., Bista R.K., Khalbuss W.E., Qiu W., Uttam S., Staton K.D., Zhang L., Brentnall T.A., Brand R.E., Liu Y. Nanoscale nuclear architecture for cancer diagnosis beyond pathology via spatial-domain low-coherence quantitative phase microscopy. J. Biomed. Opt. 2010;15:066028. doi: 10.1117/1.3523618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wax A., Terry N.G., Dellon E.S., Shaheen N.J. Angle-resolved low coherence interferometry for detection of dysplasia in Barrett's esophagus. Gastroenterology. 2011;141:443–447. doi: 10.1053/j.gastro.2011.06.020. 447.e1–e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Welch V., Vigneron J., Parker A. The cause of colouration in the ctenophore Beroe cucumis. Curr. Biol. 2005;15:R985–R986. doi: 10.1016/j.cub.2005.11.060. [DOI] [PubMed] [Google Scholar]
  46. Wiersma D.S. Disordered photonics. Nat. Photonics. 2013;7:188–196. [Google Scholar]
  47. Xu C., Carney P.S., Boppart S.A. Wavelength-dependent scattering in spectroscopic optical coherence tomography. Opt. Express. 2005;13:5450–5462. doi: 10.1364/opex.13.005450. [DOI] [PubMed] [Google Scholar]
  48. Xu C., Vinegoni C., Ralston T.S., Luo W., Tan W., Boppart S.A. Spectroscopic spectral-domain optical coherence microscopy. Opt. Lett. 2006;31:1079–1081. doi: 10.1364/ol.31.001079. [DOI] [PubMed] [Google Scholar]
  49. Xu M., Alfano R.R. Fractal mechanisms of light scattering in biological tissue and cells. Opt. Lett. 2005;30:3051–3053. doi: 10.1364/ol.30.003051. [DOI] [PubMed] [Google Scholar]
  50. Yang W., Sherman V.R., Gludovatz B., Schaible E., Stewart P., Ritchie R.O., Meyers M.A. On the tear resistance of skin. Nat. Commun. 2015;6:6649. doi: 10.1038/ncomms7649. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Yelin D., Rizvi I., White W., Motz J., Hasan T., Bouma B., Tearney G. Three-dimensional miniature endoscopy. Nature. 2006;443:765. doi: 10.1038/443765a. [DOI] [PubMed] [Google Scholar]
  52. Yi J., Backman V. Imaging a full set of optical scattering properties of biological tissue by inverse spectroscopic optical coherence tomography. Opt. Lett. 2012;37:4443–4445. doi: 10.1364/OL.37.004443. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Yi J., Radosevich A.J., Rogers J.D., Norris S.C., Çapoğlu İ.R., Taflove A., Backman V. Can OCT be sensitive to nanoscale structural alterations in biological tissue? Opt. Express. 2013;21:9043–9059. doi: 10.1364/OE.21.009043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Yun S.H., Tearney G.J., Vakoc B.J., Shishkov M., Oh W.Y., Desjardins A.E., Suter M.J., Chan R.C., Evans J.A., Jang I.-K. Comprehensive volumetric optical microscopy in vivo. Nat. Med. 2006;12:1429. doi: 10.1038/nm1450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Yuste S.B., Santos A. Radial distribution function for sticky hard-core fluids. J. Stat. Phys. 1993;72:703–720. [Google Scholar]
  56. Zhao Y., Maher J.R., Kim J., Selim M.A., Levinson H., Wax A. Evaluation of burn severity in vivo in a mouse model using spectroscopic optical coherence tomography. Biomed. Opt. Express. 2015;6:3339–3345. doi: 10.1364/BOE.6.003339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Zimm B.H. The scattering of light and the radial distribution function of high polymer solutions. J. Chem. Phys. 1948;16:1093–1099. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Video S1. Time-Lapsed Spectral Centroid Images of Motile Cilia In Vitro, Related to Figure 2B

32 frames per second. Time in (sec: [1/60 s]).

Download video file (1.2MB, mp4)
Video S2. HSV-Coded 3D SOCT Images of Human Skin In Vivo, Related to Figure 3
Download video file (2MB, mp4)
Video S3. HSV-Coded 3D SOCT Images of Fresh Swine Optic Nerve head Ex Vivo, Related to Figure 3E
Download video file (7.7MB, mp4)
Video S4. Time-Lapsed Spectral Centroid Images of Swine Coronary Artery Wall Subject to hydrostatic PBS Loading from 44.1 to 117.7 mm Hg Ex Vivo, Related to Figure 5C and 5D

50 frames per second. Time in (sec: [1/60 s]).

Download video file (3.4MB, mp4)
Document S1. Transparent Methods, Figures S1–S17, and Tables S1–S4
mmc1.pdf (9.6MB, pdf)

Articles from iScience are provided here courtesy of Elsevier

RESOURCES