Investigation of dermal collagen nanostructures in Ehlers-Danlos Syndrome (EDS) patients.

Ehlers-Danlos syndrome (EDS) is a heterogeneous group of inherited connective tissue disorders [ 1 , 2 ]. Approximately 1.5 million individuals have received genetic diagnoses for EDS, with around 10,000 cases reported in Canada alone. However, a significant number of individuals, estimated at 255 million, display the clinical characteristics of EDS but have not been diagnosed due to the absence of identifiable genetic mutations associated with the condition [ 3 – 7 ]. EDS classification has changed over the years based on the inheritance and clinical presentation. In 1988, the "Berlin nosology" defined 11 subtypes of EDS based on their inherent characteristics and clinical manifestations. Subsequently, in 1997, the "Villefranche nosology" was released and used as the classification system for EDS for two decades [ 8 – 10 ]. The current 2017 EDS nosology remains rooted in genetic markers and clinical presentation, which introduce subjective elements into the classification process. Moreover, the classification system does not comprehensively address hypermobile EDS (hEDS), which leads to late or misdiagnosis. It currently takes an average of ten to 16 years from the onset of symptoms before a correct diagnosis of hEDS is made [ 11 ]. The two most prevalent types of EDS are classical EDS (cEDS) and hypermobile EDS (hEDS) [ 12 ]. cEDS is an autosomal dominant connective tissue disorder characterized by skin hyperextensibility, abnormal wound healing, and joint hypermobility. Skin manifestations in cEDS can include easy bruising, atrophic scarring, and fragility, while joint hypermobility may lead to recurrent dislocations or subluxations [ 13 – 15 ]. Mutations in COL5a1 or COL5a2 genes, responsible for encoding type V collagen, are observed in over 90% of cEDS cases. Type V collagen is a crucial component of collagen Type I and III fibrillogenesis nucleation. In rare cases, mutations in the COL1A1 gene, responsible for encoding type I collagen, are also identified [ 10 , 16 – 19 ]. hEDS shares several clinical features with EDS subtypes but lacks confirmed genetic markers. Joint hypermobility and chronic pain are prominent in hEDS, with an increased risk of developing conditions like fibromyalgia [ 13 ] and chronic fatigue syndrome [ 20 ]. The lack of objective markers for hEDS makes the diagnosis difficult, and often, healthcare providers do not have adequate knowledge to make a timely diagnosis of EDS, resulting in perceived trauma for patients who struggle with multisystemic EDS related symptoms. This form of medical trauma stems from perceived hostility and disinterest from healthcare providers, leading to psychological distress and aversive responses toward healthcare encounters [ 21 – 23 ]. Thus, proper diagnosis and comprehensive care for both cEDS and hEDS require a multidisciplinary approach and are crucial for improving the care and well-being of individuals with hEDS. The most recent criteria for hypermobile Ehlers-Danlos syndrome (hEDS) were defined by the 2017 International Classification for the Ehlers-Danlos Syndromes, which the International Consortium published on the Ehlers-Danlos Syndromes [ 10 ]. The 2017 criteria for hypermobile Ehlers-Danlos syndrome (hEDS) consist of three domains: i) assessment of generalized joint hypermobility using the Beighton score, ii) evaluation of skin involvement, musculoskeletal pain, and other systemic characteristics associated with EDS, and iii) exclusion of other conditions related to hypermobility Beighton score involves a series of nine maneuvers tests to evaluate flexibility and hypermobility in various joints. The following criteria are typically used: a) Passive dorsiflexion of each little finger beyond 90 degrees (1 point for each hand); b) Passive apposition of the thumbs to the flexor aspects of the forearm (1 point for each thumb); c) Hyperextension of each elbow beyond 10 degrees (1 point for each elbow); d)Hyperextension of each knee beyond 10 degrees (1 point for each knee); e) Ability to place the palms on the floor with straight legs (1 point). Therefore, the maximum Beighton score is 9 points (5 for the fingers, 2 for the thumbs, and 1 each for the elbows and knees). A higher score signifies a greater degree of joint hypermobility, and generalized joint hypermobility is considered to be present if an individual scores 5 points or more. While the Beighton score is a useful tool, a diagnosis of Ehlers-Danlos syndrome is typically based on a combination of clinical evaluation, family history, and genetic testing. Not everyone with hypermobility has EDS, and other factors are considered in the diagnostic process. We hypothesize that individuals with EDS experience impairments in dermal collagen’s structural and mechanical characteristics at the nanoscale, leading to disruptions in the dermis’s overall structural and mechanical integrity. The current case study involved employing Atomic Force Microscopy (AFM) to image and perform mechanics on ex-vivo skin biopsies sourced from one patient diagnosed with cEDS, one patient with hEDS, one with hEDS-scleroderma, and one healthy control. Although the findings presented in this study pertain to the four examined cases, they provide novel perspectives on the nanoscale collagen structure and biomechanics in individuals with EDS. These insights may serve as the foundation for future research into specific biomarkers for the condition. Atomic Force Microscopy is a high-resolution imaging technique used to study a wide range of samples across various scientific disciplines. Unlike optical microscopes that use light, AFM operates by scanning a sharp tip (typically 20 nm in diameter) over the surface of a sample. As the probe moves across the material, it experiences attractive and repulsive forces between the atoms on the tip and those on the sample surface [ 24 – 26 ]. These interactions cause the cantilever to deflect, and the deflections are measured to create a detailed 3-dimensional topographic map of the sample’s surface. In addition, the probe can be used as an indenter and perform localized mechanical measurements directly onto the sample to extract valuable and specific information such as Young’s Modulus [ 27 ]. Unlike other microscopy methods, such as scanning/transmission electron and fluorescence microscopy, AFM excels in providing detailed topographical and mechanical data at the nanoscale without needing sample preparations like labeling or coating, making it an effective tool for examining various collagen structures without altering their fibrillar integrity [ 28 – 30 ]. This makes AFM a very competent technique to perform histological assessment of tissues with unprecedented resolution [ 31 – 33 ]. AFM has previously used for connective tissue morphometry studies such as Study of the extracellular connective tissue matrix in patients with pelvic organ prolapse [ 34 ], nanohistological investigation of scleroderma [ 35 ], oral submucous fibrosis [ 36 ], assessments of extracellular matrix in intervertebral disc and degeneration [ 37 ], and many other studies. In this study we have employed AFM to examine ex-vivo skin biopsies obtained from EDS patients to characterize collagen morphometric differences in cEDS and hEDS. The study was approved by the ethics board at the University Health Network, Toronto, Canada (REB #21–5542), and written consent was obtained by the study participants prior to study commencement. This test includes 22 genes. Details can be found online on the Our present study involved three female patients diagnosed with EDS and one healthy female volunteer of Caucasian ethnicity (mean age: 33± 7 years). Next generation Sequencing (NGS) with duplication and deletion analysis by exon targeted microarray was performed at SickKids hospital for diagnosis. This test includes 22 genes. Details can be found online on the SickKids website [ 38 ]. A 38-year-old woman with a history of multi-joint hypermobility, recurrent subluxation of both hips and knees, stretchy skin with easy bruising, chronic pains, orthostatic intolerance, migraines, and irritable bowel syndrome was referred to the GoodHope EDS clinic in the suspicion of Ehlers-Danlos syndromes (EDS). Family history was negative for genetic inherited connective tissue disorders, including Ehlers-Danlos syndromes. Physical examination revealed a Beighton score of 5/9. In addition, she met the systemic criteria for h-EDS (positive findings for mild skin extensibility, unusually soft/velvety skin, arachnodactyly, peizogenic papules, and a high-arched narrow palate with dental crowding). Genetic testing was ordered to rule out any genetically identifiable EDS subtypes and included a next-generation sequencing panel of more than 30 genes specific to Ehlers-Danlos syndromes. No pathogenic sequence or copy number variants were detected in the Ehlers-Danlos syndrome panel. Therefore, the final diagnosis of hEDS per the 2017 EDS diagnostic criteria was confirmed. A 27-year-old female with a pre-existing diagnosis of scleroderma, previously confirmed based on the clinical skin abnormalities, patient had a 10-year history of inflammatory arthralgias involving PIP and MCP joints, diffuse muscle and joint pains affecting multiple areas of body, secondary diagnosis of fibromyalgia, patchy areas of skin thickening, and Raynaud’s phenomenon. History was negative for uveitis, iritis and psoriasis. Lab testing was positive for Anti-Nuclear Antibody (ANA) and Anti centromere Antibody (ACA) and negative for Anti CCP and rheumatoid factor. The patient also had generalized joint hypermobility, hip subluxations, soft, stretchy skin with easy bruising, fibromyalgia, migraine, IBS irritable bowel syndrome (IBS), endometriosis patella alta, and inflammatory arthritis was referred to rule out a diagnosis of Ehlers-Danlos syndromes. Family history was negative for any subtype of EDS. Physical examination revealed a Beighton score of 9/9. Further, impressive interphalangeal joint (DIP) and proximal interphalangeal (PIP) joint hypermobility of the hands, upper thoracic scoliosis, and joint instability of hips, ankle, and shoulder joints were found. In addition, she had unusually soft/velvety skin, mild skin hyperextensibility (1.5 cm at forearm), unusual striae in unusual locations, arachnodactyly, and peizogenic papules were identified. A next-generation sequencing panel of more than 30 genes specific to Ehlers-Danlos syndromes was ordered to rule out any genetically identifiable EDS subtypes [ 38 ]. No pathogenic sequence or copy number variants were detected in the Ehlers-Danlos syndrome panel. Therefore, the final diagnosis of hEDS per the 2017 EDS diagnostic criteria was confirmed. A 24-year-old woman with complaints of significant skin fragility, easy bruising, and skin splitting requiring multiple stitches as a child on several occasions, multi-joint hypermobility and chronic pains, and scoliosis, retinal tears as a kid with a previously given diagnosis of classical EDS was referred for diagnostic confirmation and treatment planning. She had no significant family history of EDS, aneurysms, vascular ruptures, or sudden death under 40 yr. in the family. Physical examination revealed a Beighton score of 9/9. In addition, significant skin features, including soft/velvety skin, significant skin hyperextensibility at multiple spots in keeping with classical EDS, and significant atrophic scarring with cigarette paper scaling were identified. Genetic testing revealed a COL1A1 mutation (c.1053delG in exon7), confirming the diagnosis of classical EDS.

In conclusion, histology image analysis combined with nanoscale investigations using Atomic Force Microscopy (AFM) has provided valuable insights into the structural alterations within the dermal collagen matrix in individuals with Ehlers-Danlos Syndrome (EDS). These findings could revolutionize the diagnosis and understanding of EDS subtypes, particularly hypermobile EDS (hEDS), which often lacks specific diagnostic markers. Histological analysis revealed distinct differences in collagen organization and density between healthy individuals and those with EDS, with a notable impact on the mechanical behavior of the skin. Polarized light microscopy further emphasized these differences, highlighting the presence of disrupted collagen in EDS cases. Additionally, the use of Sirius Red staining and polarization allowed for quantifying collagen disruption, providing valuable information for assessing EDS severity. The correlation between nanoscale AFM images and histological staining patterns demonstrated unique structural characteristics within the collagen scaffold of EDS patients. These characteristics, such as disrupted collagen fibrils and amorphous collagen, could be potential biomarkers for EDS subtypes. The reliability and repeatability of identifying disrupted collagen regions by non-specialists through AFM-based nanoscale assessments further support the potential for developing diagnostic tools based on collagen structural variations. Furthermore, evaluating collagen fibril elasticity using AFM nanoindentation revealed distinct mechanical properties in healthy and disrupted collagen regions. These findings can potentially contribute to developing specific biomechanical markers for EDS subtypes, particularly hEDS, which currently lacks diagnostic criteria.

Two skin punch biopsies, three millimeters in diameter, were obtained from the medial surface of both arms above the elbow at UHN GoodHope EDS clinic. These biopsies were immediately transferred to the Faculty of Dentistry at the University of Toronto on ice in Dulbecco’s Modified Eagle Medium (DMEM) (Sigma-Aldrich, USA) supplemented with 2% Penicillin-Streptomycin (Sigma-Aldrich, USA), and 2% fungizone (Sigma-Aldrich, USA) The tissue biopsies were frozen in optimal cutting temperature resin (OCT) (VWR- Richmond, IL. USA). Serial tissue sections, 7μm thick, were made longitudinally through the epidermis and papillary dermis. Slides were divided into sets, each consisting of four serial sections. The first slide of each set was stained with routine H&E, the second was stained with picrosirius red, and the third and fourth were used for atomic force microscopy. The slides stained with H&E were kept at room temperature for 10 minutes to thaw the OCT before being immersed in alcohol 95%, followed by Formalin 10% for 20 seconds each. The slides were washed with MilliQ water, stained with hematoxylin for 3 minutes, washed, and passed through acid alcohol, sodium bicarbonate, wash, eosin, upgraded alcohol, and xylene for 20 seconds each. Finally, these slides were covered with mounting media and cover glass. To stain slides with PS, nuclei were stained first with hematoxylin as described in H&E stain. Slides were then immersed in PS solution for 1 hour Sirius red (Sigma-Aldrich, USA) diluted in picric acid (Sigma-Aldrich, USA)). The slides were washed in acidified water for 30 seconds (0.5% glacial acetic acid diluted in distilled water) before being dehydrated in alcohol and xylene and finally covered with mounting media and cover glass. Picrosirius red stained histological sections were imaged using a polarized light microscope (Leica DM500 microscope, USA). Images of specific areas were captured before and after a 90° polarization stage rotation, ensuring consistent imaging conditions, including exposure time and gain. PSR image quantification was performed by analyzing the color composition of each image using RGB values. The identification of different colors was based on specific ranges of RGB combinations. Red pixels were defined as having RGB values between (50,0,0) and (255,0,0), yellow pixels were defined as having RGB values between (50,50,0) and (255,255,0), green pixels were defined as having RGB values between (0,50,0) and (0,255,0). Blue pixels were defined as having RGB values between (0,0,50) and (0,0,255). It is important to note that the selection of red and yellow colors was not deliberate but based on the colors in the images. Consequently, only red and yellow pixel values were displayed, while green and blue values were observed to be zero. To determine the color ratios, the number of pixels corresponding to each color (red, yellow, green, and blue) was divided by the total sum of red, yellow, green, and blue pixels. This calculation allowed for the calculation of the percentage color ratio. Each image was processed using Python 3.10 within the PyCharm community version virtual environment. The analysis utilized various modules, including Skimage, NumPy, Pandas, and OpenCV. Two unfixed and unstained sections were allocated for AFM-based quantitative nanohistology (QNH). The AFM images and force-distance curves (nano-indentation) were acquired using a Nano-wizard 4 Bioscience instrument (Bruker, Germany) operating in contact mode under ambient conditions. The imaging and force spectroscopy measurements were performed directly on histological sections within the reticular dermis area of the skin biopsies in ambient conditions. The slides were not rehydrated post-sectioning to avoid further alterations in the collagen structure. MSLN-10-C (k = 0.01N/m) and RTESPA-150 (k = 6N/m) probes (Bruker, Germany) were used for imaging and force spectroscopy, respectively. To correlate the AFM data with the picrosirius red stain images, seven healthy areas marked as red and seven damaged regions marked as greenish yellow in SR stained were selected for imaging and force spectroscopy analysis. The imaging was captured using a size of 5×5 μm, and the settings were optimized for the best results at a frequency of 2.5 Hz. As discussed previously, seven healthy and seven disrupted areas were selected for the indentation measurements. An image was acquired in contact mode at a frequency of 2.5 Hz. Subsequently, a grid of 10×10 was defined on acquired AFM images, and a total of 700 force curves was recorded for each of the healthy and damaged areas of the skin biopsies, providing a comprehensive dataset for analysis. The Young Modulus of the collagen fibrils (E), measured here as the compressive elastic modulus, was determined by fitting the Hertzian model to the force-distance curve acquired from all measurements. The obtained modulus values were then plotted as a distribution to calculate the median and error for all groups. All images were processed, and data analysis was conducted using the JPK data processing software (version 6.3.5). It is important to note that to accurately position the AFM probe on desired areas, two adjacent sections from skin biopsies were used. One was stained with PSR, and the other was unstained and unfixed. Tissue gross morphology was used to align the PSR and optical images on the AFM instrument. Zooming in on the respective images enhanced the geolocation accuracy for the probe placement within the desired regions of interest, however small these may be, even though the AFM data acquisition was performed on an unstained and unfixed section. Since the AFM is equipped with x-y scanner that can place the probe anywhere on the sample surface with high-accuracy (<110nm), it is therefore possible to target the area to be imaged or mechanically probe with a high degree of confidence. To investigate whether nanoscale changes in collagen structure could serve as a distinctive marker or fingerprint for EDS, we performed Kappa analysis using nine raters who did not have prior knowledge of AFM imaging or collagen fibril anisotropy interpretation. The assessors were first trained to recognize collagen fibril clarity, D-banding, orientation, and linearity on 3 AFM images of healthy skin. The training emphasized the structure of healthy collagenous tissue involving the four mentioned parameters. The same procedures were followed for 3 AFM images of disrupted collagen. After training, nine assessors were presented with 32 randomized AFM images (16 from each healthy and diseased group). We performed reliability and repeatability tests using Fleiss’s and Cohen’s kappa score analyses. Kolmogorov–Smirnov test at 0.05 level was used to find significance between all groups. All statistical analyses and data plotting were performed with OriginPro 2021 software.

(DOCX) A unimodal distribution of Young’s moduli was observed for both normal and disrupted collagen across all three regions. The red dotted line represents the distribution of healthy collagen in the control group. (TIF)