Laser cartilage reshaping (LACR) is a relatively novel technique designed to noninvasively and permanently restructure cartilaginous tissue for applications in otorhinolaryngology, orthopedics, and plastic surgery. Albeit the efficacy of LACR has been empirically demonstrated in animals 1, 2, 3, 4, 5, 6, 7, 8, 9 and humans,10 the specific biophysical and biochemical mechanisms underlying the laser-tissue interactions have remained largely elusive. The current contention is that LACR is mainly governed by laser-induced stress relaxation2, 11, 12 in which a temperature-mediated disruption of binding is associated with conformational alterations in the proteoglycan (PG) and collagen-rich matrix. In a native state, a small percentage of molecules provides structural integrity by facilitating electrostatic interactions between collagen and PGs.13 Laser irradiation causes a dissociation of bound and a corollary reduction in tensile stresses and matrix stability,14, 15, 16, 17 resulting in a shape change of cartilage.
Insofar as the phase transition cannot be solely responsible for the permanent deformation of cartilage, Sobol 18, 19 have postulated that local mineralization may enforce the long-term restructuring process following laser irradiation. In support of this hypothesis, Velegrakis 5 found loci of calcification in laser–irradiated auricular cartilage in vitro, and Sobol 18 reported on the occurrence of submicrometer sodium carbonate crystals in the matrix of laser–irradiated nasal septal cartilage samples. Since crystals are chemically unstable and resolve in time, the authors dismissed the possibility that local mineralization of this type may contribute to the generation of a stable cartilage configuration.18 A later in vivo study20 of porcine ear cartilage evinced the presence of crystals of undetermined composition in the irradiated zone after laser treatment, refuting the notion that laser-induced crystal formation is merely a short-term phenomenon.
In this preliminary study, we aimed to evaluate chronic Er:glass laser–induced mineralization in rabbit auricular cartilage using Raman microspectrometry, an optical technique based on inelastic scattering of light by the molecules in a sample. Provided that each molecular species has a specific set of intramolecular bond vibrations,21 the spectroscopically detected frequency shifts in remitted light function as “optical fingerprints” that can be employed for qualitative analysis of tissue constituents. Raman microspectrometry therefore constitutes a useful technique for studying postinterventional tissular modifications, and in particular for attesting the composition of intra- and extracellular mineral deposits.
Materials and Methods
Two female white New Zealand rabbits weighing 3.1 and were used in this study. The animals were housed under standard laboratory conditions and maintained on standard laboratory food and water ad libitum. All procedures were approved by the Lille University Animal Ethics Committee (protocol 2003-021) and conducted in compliance with the Resolution on the Use of Animals in Research (Ministère de L’Agriculture et de la Forêt, No. 87-848, Agreement No. 4844).
For the experimental procedure, the animals were placed in a restrainer. Rabbit ear lobes were used because of their abundant elastic cartilage content (mean cartilage thickness of ) and accessibility. The target region of the ear was locally anesthetized by topical application of 2% lidocaine, shaved, and disinfected. The upper third portion of the ear was mechanically deformed by the placement of a preperforated reshaping jig (curvature radius of ) prior to lasing. A light source was inserted into the reshaping jig so as to standardize laser irradiation by providing target points through transillumination.8 The animals were sacrificed after the biopsy procedure by intravenous administration of KCl.
Laser irradiation was performed with a Er:glass laser (Aramis, Quantel Medical, Clermont Ferrand, France). The treatment consisted of seven stacked pulses of per pulse, pulse duration, and a repetition rate, yielding an irradiance of and cumulative radiant exposure of with a spot size. An integrated cooling device (Koolburst, Quantel Medical, Clermont Ferrand, France) was placed in direct contact with the skin. The cooling temperature was set to and contact was maintained for approximately before lasing. Eight contiguous parallel rows along the ear consisting of 12 adjacent, nonoverlapping spots were subjected to irradiation (area of ). The irradiated region was delineated by tattooing the margins of the treatment zone. For postoperative curvature maintenance, the reshaping jig was replaced by a holder ( diameter plastic tube) and secured with sutures. The holder was removed after . Curvature radius measurements were performed as reported previously8 before punch biopsies.
Tissue Preparation and Raman Microspectrometry
Samples were excised by punch biopsy at 3 and after laser treatment from the irradiated and nonirradiated (control) sites after application of 2% lidocaine. Six samples were biopsied per site at each time interval and snap frozen by immersion in liquid nitrogen. The excised tissue was cryocut into thick sections and stored at until further processing. Prior to Raman spectroscopic analysis, the samples were fixed in 4% formaldehyde for , washed in deionized water, and passively dried.
The samples were analyzed with a Raman microspectrometer (Labram, Jobin Yvon, Lille, France). Spectra were measured in intracellular crystal structures ( : control, , LACR, ; : control, , LACR, ) and in the multicellular (Y) and bicellular (I) junctions of the pericellular matrix (Y junction : control, , LACR, ; : control, , LACR, ; I junction : control, , LACR, ; : control, , LACR, ) (see Fig. 1 ). Raman excitation was induced by a HeNe laser ( , , spot size) passed through a narrowband interference filter (FWHM , model 03 FIL 006, Melles Griot, Voisins-le-Bretonneux, France) and focused on the region of interest through the microscope optics (BHM brightfield/darkfield metallurgical microscope, Olympus, Rungis, France). Raman signals were collected by the objective lens [ magnification, numerical aperture (NA) of 0.95] and acquired by a single-grating spectrometer equipped with an air-cooled CCD detector. The overall spectral resolution was in the range with an integration time of . Background fluorescence was subtracted from baseline computations, and polynomial filtering was performed using LabSpec software (Dilor-Jobin Yvon, Lille, France). Raman spectra were normalized to the highest intensity value measured.
The outcome of LACR with respect to treatment-related thermal effects and auricular morphology was similar to previous findings:8 no immediate laser-induced effects such as swelling, erythema, or hypopigmentation were observed macroscopically, and latent sequelae such as crusting or blistering remained absent. At 3 and the curvature radii were 7.5 and , respectively, in comparison to the initial curvature radius of the reshaping jig .
Histologically, intracellular crystalline structures were found in approximately equal numbers in control and laser-irradiated samples at and . Raman microspectrometry of these intracellular structures revealed that the crystals are predominantly composed of calcium carbonate (calcite, ) with distinct peaks at 275, 709, and (Fig. 2 ). In one of the three irradiated samples at we detected calcium sulfate , characterized by spectral peaks at 412, 491, 617, 668, 1006, (Fig. 3 ). was not found in any of the control samples or in the irradiated samples at . In addition, the peaks at ( twisting mode, Fig. 3), ( bending mode, Fig. 3), (symmetrical stretching, results not shown), and (asymmetrical stretching, results not shown) are indicative of phospholipids.22, 23
Figure 4 depicts spectra obtained from a Y junction in control (black) and laser-irradiated (gray) specimens at . The peaks at 426 and from the laser-irradiated cartilage sample correspond to the phosphate and vibration modes of apatite,22, 23, 24, 25, 26, 27 respectively, which was found in two of the 14 analyzed Y junctions at . A reference spectrum of purified carbonated apatite is provided in Fig. 5 . The peak at (gray spectra, Figs. 4 and 5) suggests a type B substitution ( substitution of ) in the crystal lattice,25, 26 as is often the case in biological apatites such as bone.28, 29 The bands at 873 and may be attributed to acidic phosphate ions22, 25 . Since OH stretching modes above (usually present in hydroxyapatite and dental enamel25, 27) fall outside the detection range, it is impossible to draw irrevocable conclusions on the hydroxylation state of the apatite. No apatite was detected in the control samples, the I junctions at and , and the Y junctions examined at .
The principal mechanisms underlying the permanent shape change of cartilaginous tissue following laser irradiation have been elaborately investigated 1, 2, 5, 8, 9, 12, 14, 18, 30, 31, 32, 33, 34, 35, 36, 37 since the inception of LACR in 1993. Currently, it is accepted that laser-induced stress relaxation, i.e., the reduction of tensile stresses in the cartilage extracellular matrix (ECM) as a result of a heat-mediated redistribution and corollary reorganization of intramolecular bonds, lies at the basis of the restructuring machinery. The laser-induced phase transition of is inevitably accompanied by ancillary processes, including thermally mediated de- and subsequent repolymerization of PG aggregates or its glycosaminoglycan subunits, transient disruption of collagen-PG linkages, and thermal denaturation of matrix proteins such as collagen.5, 18, 19, 30
In addition to these acute biochemical events, several recent studies5, 18, 20 have revealed that LACR-induced cartilage mineralization may comprise a chronic component of the permanent restructuring process. Inasmuch as the crystal deposits found in these studies were either uncharacterized5, 20 or of a chemically unstable composition,18 we performed a Raman microspectrometric study to validate the previously obtained results in the context of Er:glass LACR and to identify chemically stable crystal structures in rabbit auricular cartilage 3 and after lasing. Although the results are preliminary due to the limited sample size, we were able to unequivocally identify cytosolic calcium sulfate deposits and extracellular apatite crystals in the laser-treated samples at by their spectral correspondence to purified analogues (black spectra in Figs. 3 and 5, respectively). On top of the fact that healthy auricular cartilage rarely calcifies, the absence of chondrocalcinosis in the untreated specimens strongly suggests that laser-mediated (auricular) cartilage mineralization is a direct consequence of the photothermal interactions with tissue. In this respect, two main questions must be elucidated: what are the exact dynamics of laser-induced cartilage mineralization? and to what extent does crystal formation contribute to the overall conformational stability of laser-reshaped cartilage? Since our findings have the greatest relevance to the former, the discussion predominantly focuses on laser-induced mineralization processes.
Theoretically, all the acute ultrastructural modifications associated with stress relaxation5, 18, 19, 30 could invoke a microscale milieu replete with potential nucleation centers required for crystal formation and growth, provided that certain (patho)physicochemical conditions dominate. The most imperative condition for crystal formation is an exceedance of the critical supersaturation level by the component ions. The critical supersaturation represents a value close to the solubility product above which the component ions of the crystal do not remain in solution but precipitate and form aggregates.38 Cartilage fluids are marginally supersaturated with , the main component ion of biological apatite, but do not crystallize due to the presence of chelators and crystallization inhibitors such as proteoglycan aggregates and chondroitin sulfate.38, 39, 40, 41 A laser-induced alteration in the ECM hydration state could therefore shift the supersaturation level toward a higher ionic disequilibrium, producing an environment that significantly favors crystal formation. A representative shift was recently reported by Youn, 42 who, using polarization-sensitive optical coherence tomography, revealed that laser-induced phase retardation changes observed at temperatures of in superficial cartilage layers are primarily attributable to tissue dehydration. Interestingly, these temperatures overlap the temperatures required for stress relaxation . 2, 12, 14, 30, 31, 34, 35 Similarly, a stress-relaxation-associated redistribution of molecules1, 19, 30, 31, 43 may alter the hydration state locally. The release/relocation of molecules coincides with the exposure of anionic moieties on the sulfate and carboxyl groups of the glycosaminoglycan chain, which may be electrically neutralized by the binding of and ions that are abundantly present in cartilage. Since the phase transition of occurs above the thermal denaturation threshold of collagen and other matrix proteins, cations may also neutralize negatively charged amino acid residues that are exposed during protein unfolding.44, 45 In addition to the probability that thermal denaturation of crystallization inhibitors will impart an agonistic effect on the mineralization process, the cationic substitution of intermolecular bonds and intramolecular hydrogen and sulfur bridges is possibly affiliated with the creation of crystal nuclei.19 Once a crystal is nucleated, lower supersaturation levels are sufficient for crystal growth and proliferation.38
The formation of nucleation centers during and immediately after the laser pulse through cationic substitutions is currently the only putative explanation for postirradiative cartilage mineralization.19 Based on tissue calcification studies in osteology and rheumatology, there is some circumstantial evidence that laser-induced cartilage mineralization in the ECM may be under cellular regulation in addition to the proposed acute biochemical modifications. Nonpathological endochondral ossification, for example, has been mechanistically linked to the release of matrix vesicles (MVs) from apoptotic (hypertrophic) chondrocytes in the epiphyseal growth plate.46, 47, 48, 49 These nanoscale vesicles are extracellular membrane-enclosed particles that are generated by polarized budding from chondrocytes during early apoptosis.46, 47 The unique intravesicular environment makes MVs optimal mineralization units, particularly with respect to (hydroxy)apatite: the MV cytosolic membrane leaflet is replete with acidic phospholipids (predominantly phosphatidylserine) that promote the ingress of ions,47, 50 several types of annexins act as calcium binding proteins and transmembrane calcium channels,46, 47 membrane-embedded phosphohydrolases47, 51, 52, 53 control intra- and perivesicular concentrations, and a sodium-dependent transporter facilitates ionic uptake.47, 54 A pH regulatory mechanism in the form of carbonic anhydrase provides a basic climate most suitable for crystal stabilization and sustained growth of preformed hydroxyapatite templates near the inner surface of the MV membrane.47, 55 An equivalent vesicle-mediated mineralization system can be found in degenerative diseases such as osteoarthritis, where chondrocyte-derived apoptotic bodies comprise56, 57 the functional homologues of MVs. The apoptotic bodies, which can be induced in vitro by nitric oxide (NO) donors or agonistic antibodies to Fas, a chondrocyte-specific proapoptotic signaling molecule, possess a similar calcification machinery to MVs in that they contain alkaline phosphatase and pyrophosphate-generating nucleoside triphosphate pyrophosphohydrolase activity and have the ability to precipitate calcium ions.56, 57
With cell death at the basis of physiological and pathological cartilage mineralization, it is not unlikely that laser-induced chondrocalcinosis in the ECM is mediated via common molecular pathways and crystallization processes. Several flow cytometric and histological studies5, 9, 58, 59, 60, 61 have demonstrated that chondrocyte viability is partially compromised following subablative laser irradiation, with a decrease in cell viability occurring near a temperature threshold of .59, 62 Although the mechanisms of laser-induced chondrocyte death are yet to be irrefutably established, membrane permeability–based live/dead assays of isolated laser-irradiated chondrocytes suggest a necrotic (cell membrane thermolysis and possibly thermal denaturation of cytoplasmic elements) and apoptotic origin.59, 61, 62 Wong 61 recently identified a subpopulation of cultured chondrocytes 2 and after Nd:YAG laser irradiation that, based on flow cytometric light scattering properties, displayed an apoptotic phenotype. The extent of apoptosis seemed to be positively correlated to irradiation times, i.e., the volumetric heat production. Inasmuch as heat-induced disintegration of the cell membrane causes efflux of cellular contents into the ECM, the release of physiologically active compounds such as NO by dead or injured cells may induce a host of deleterious cellular responses, including oxidative stress, oxidant injury, and apoptosis in neighboring, intact chondrocytes.56, 63, 64, 65 Hyperthermic damage to cytosolic polypeptides and proteins may lead to activation of apoptotic signal transduction pathways even in the absence of membrane disruption (the differential thermodestructive criteria between proteins, lipoproteins, and membranes emanate from the thermodynamically stabilizing properties of phospholipids66). During severe heat stress , apoptogenic nuclear transcription factors are upregulated by two subfamilies of mitogen-activated protein kinases (MAPKs), namely, p38MAPK and SAPK/JNK, in response to the stress-induced dissociation of apoptosis signal-regulating kinase-1 (Ask1) from its inhibitor GSTM1-1 (glutathione S-transferase Mul-1). 67, 68, 69, 70 At the level of the unfolded protein response (UPR), a ubiquitously conserved countermechanism against unfolded or misfolded protein accumulation in the endoplasmic reticulum, three known apoptotic cascades governed by IRE1, caspase-12, and PERK/CHOP are initiated when the protein overload cannot be resolved.71 Excessive thermal denaturation of nascent polypeptides and preproteins could therefore trigger apoptosis if the UPR, in cooperation with stress-inducible heat shock proteins (HSPs), is ultimately ineffective. Other molecular triggers leading to programmed cell death, including HSP-facilitated degradation of death regulatory proteins,72 may also play a role in the pathophysiological fate of laser-irradiated chondrocytes. On initiation of the apoptotic program, cells undergo nuclear and cytoplasmic condensation, blebbing, and the formation of apoptotic bodies. The exocytosolic membrane surface of apoptotic bodies becomes replete with target receptors for phagocytosis.73 Cartilage is unique in the sense of being avascular and lacking phagocytes responsible for engulfing cellular debris and apoptotic bodies,56 which implies that apoptotic bodies enjoy a relatively long life span in this tissue with ample time to develop mineral templates. Ultimately, in the absence of phagocytosis, apoptotic bodies may proceed to secondary or apoptotic necrosis characterized by a loss of membrane integrity,73 but probably not before having fulfilled their function as mineralization units.
Taken all together, our Raman microspectrometric characterization of extracellular apatite in the Y junctions of laser-irradiated cartilage specimens fits the mineralization paradigm just outlined. Apatite is a geographically foreign material in the context of auricular cartilage; only 13 clinical cases on auricular cartilage calcification have been reported since 1890, none of which had an alleged etiology associated with severe hyperthermia.74 The absence of biomineralization in the ECM (Y and I junctions) of control samples therefore indicates that mineralization occurred in response to laser irradiation. Additional support for our hypothesis is derived from the possible connection between laser-induced chondrocyte damage and chondrocalcinosis, with necrosis and apoptosis as the common denominators. Moreover, cartilage tissue constitutes an optimal niche for apoptotic bodies, which can thrive in the ECM without being attacked by inflammatory scavengers of the reticuloendothelial system. The formation of biological apatite in laser-reshaped cartilage may be pertinent to LACR insofar as the sole physiological purpose of crystallization is to impose structure and rigidity during the development of the skeletal framework and dental enamel. These properties could potentially be extrapolated to the chronic shape retention of laser-irradiated cartilage, where local tissue petrification could deter a conformational relapse to the original shape over time.
Contrary to apatite, the biochemical origin and clinical relevance of intracellular calcium sulfate deposits are elusive, particularly because has no apparent intrinsic physiological or structural value. The negative results from extensive Boolean searches using the major search engines (PubMed, ISI Web of Science, and Google) hence suggest that crystal formation in laser-treated chondrocytes is an end result of severely disrupted cellular homeostasis. A condition of cytoplasmic hypercalcemia, concurrently accounting for a local exceedance of the critical supersaturation level, can be justified by the abundant cytosolic and intraorganellar presence of calcium ions and their liberation from intracellular stores following a laser-induced stimulus. However, it is unclear as to how a similar condition in regard to sulfate is established. Individual cells acquire this highly dissociated, divalent hydrophilic anion predominantly from serum via cell-specific transmembrane transport systems and, to a lesser extent, by hydrolysis of sulfoconjugates and oxidation of reduced organic sulfur.75 The available inorganic sulfate is subsequently metabolized into intracellular sulfate-containing nucleotides, e.g., adenosine -phosphate -sulfatophosphate (PAPS), that act as sulfate donors in various types of sulfation reactions required for the synthesis of numerous biologically active compounds and for posttranslational modifications of structural components of membranes and tissues75 (i.e., sulfated glycosaminoglycans, the main constituents of cartilage). Its availability as a component ion for the mineralization process is therefore restricted, unless photothermal damage induces an increased rate of hydrolysis of conjugated sulfate groups, oxidation of organic sulfur, or accelerated influx of sulfate ions from the extracellular environment. It is known that membrane fluidity, i.e., membrane motional order or lipid packing order, affects the passive permeability properties of membranes and the activity and kinetics of membrane-embedded transport carriers.76, 77 Since one of the affectors of membrane fluidity is temperature, the generation of isotherms as a result of light absorption by water might have altered the transmembrane transport kinetics of sulfate ions. Note that these speculations are based on very few related studies. Unfortunately, sulfate is the least-studied serum ion, and little is known about the factors that govern sulfate homeostasis in animals.78
In this preliminary study, Raman microspectrometry was employed to qualify biominerals in laser-reshaped rabbit auricular cartilage 3 and after Er:glass laser irradiation in an attempt to elucidate the potential role of chondrocalcinosis in LACR. In addition to the previously reported occurrence of sodium carbonate crystals in laser-irradiated porcine nasal septal cartilage,18 our report presents evidence that laser-induced cartilage mineralization also includes intracellular calcium sulfate and extracellular calcium phosphate crystallization. We were also able to localize calcium carbonate deposits in both control and laser-irradiated samples, which, due to their equiquantitative distribution and intracellular occurrence, probably do not contribute to the permanent shape change of laser-reshaped cartilage. The limited knowledge on laser-induced cartilage mineralization prompted us to explore potential biochemical and molecular origins of chondrocalcinosis outside the thermal stress relaxation framework, also with the intent to pave the way for additional research in this field. We proposed a novel hypothesis on the formation of apatite that is based on photothermally induced necrosis and apoptosis and apoptotic body-mediated mineralization. Conversely, the exact mechanism of intracellular calcium sulfate crystallization remains elusive, and warrants further validation and investigation. Most importantly, the efficacy of LACR as a function of laser-induced biomineralization must be clarified, particularly in ratio to the discussed acute biochemical effects associated with stress relaxation. Once a positive correlation has been established, testing of the hypotheses formulated in this paper should be initiated to devise LACR-enhancing therapeutic modalities.
M. Heger was in part supported by the Technological Collaboration Grant No. TSGE 1048 of the Dutch Ministry of Economic Affairs. The authors wish to thank Drs. S. Vilette (Laser Center, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands) and G. Penel (LBM Raman, Department of Odontology, Lille University Hospital, Lille, France) for careful review of the manuscript and Prof. A. Boskey (Hospital for Special Surgery, New York) for constructive input on cartilage mineralization.