INVESTIGATION OF NEUROLOGIC ABNORMALITIES IN AN EQUINE MODEL OF EHLERS-DANLOS SYNDROMES By Abigail McElroy A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Comparative Medicine and Integrative Biology—Master of Science 2018 INVESTIGATION OF NEUROLOGIC ABNORMALITIES IN AN EQUINE MODEL OF EHLERS-DANLOS SYNDROMES ABSTRACT By Abigail McElroy Ehlers-Danlos syndromes (EDS) are a group of inherited disorders of connective tissue that affects both humans and animals. Hereditary equine regional dermal asthenia (HERDA), is a form of EDS that occurs in Quarter Horses. Though humans with EDS often present with a wide range of neurologic abnormalities including occult tethered cord syndrome, chronic headache, and craniocervical instability, the neurologic status of HERDA horses has not previously been studied. Five affected and 5 control horses were evaluated in this study. A novel method was developed for ultrasonographic examination of the myodural bridges. The myodural bridges were also assessed via histology, transmission electron microscopy, and tissue biomechanical testing. The filum terminale was assessed via histology and transmission electron microscopy, and results were compared to a large bank of human filum terminale samples at Rhode Island Hospital, Department of Neurosurgery. Preliminary ultrasonographic imaging of the myodural bridges demonstrated evidence of dural infolding in the HERDA population. Preliminary myodural bridge tissue biomechanical testing showed a smaller cross-sectional area in the HERDA population, as well as more ductile MDBs in the HERDA population. The fila terminale and myodural bridges of HERDA horses were found to have inflammatory cells histologically. Transmission electron microscopy showed disorganized collagen fibrils with abnormal fibril orientation in the HERDA myodural bridges and fila terminale. Copyright by ABIGAIL MCELROY 2018 “No other disease in the history of modern medicine, has been neglected in such a way as Ehlers- Danlos syndrome.” – Dr. Rodney Graham iv ACKNOWLEDGEMENTS I would like to thank my mentor, Dr. Ann Rashmir, for her support and guidance during my master’s program. Her expertise on HERDA was invaluable, and I could not have completed this thesis without her support and guidance. I would also like to thank the members of my committee. Dr. Petra Klinge for her wisdom, her boundless enthusiasm, and her expertise on occult tethered cord syndrome. Dr. Jon Patterson and Dr. Dodd Sledge for their support and guidance on equine histopathology, and Dr. Hal Schott for his expertise on equine neurology. I would like to thank the Normal Prince Neurosciences Institute, whose generous financial support helped make this project possible. My sincere thanks to Dr. Alicia Withrow, Michigan State University Center for Advanced Microscopy, for her assistance with transmission electron microscopy. I would also like to thank Dr. Jane Manfredi, Dr. Michael Lavagnino, Dr. Betsy Carr, Dr. Edward Stopa, Dr. John Donahue, and Dr. Kirstin Bubeck. Lastly, I would like to thank Dr. Deborah Kochevar, whose support made this graduate program possible. v TABLE OF CONTENTS LIST OF TABLES ....................................................................................................................... viii LIST OF FIGURES ....................................................................................................................... ix KEY TO ABBREVIATIONS ........................................................................................................ xi CHAPTER 1: REVIEW OF THE LITERATURE ......................................................................... 1 1.1 Project Summary .......................................................................................................................1 1.2 Ehlers-Danlos Syndromes ......................................................................................................... 2 1.2.1 Ehlers-Danlos Syndromes in Humans .......................................................................... 2 1.2.2 Hereditary Equine Regional Dermal Asthenia ............................................................. 4 1.3 Tethered Cord Syndrome ......................................................................................................... 6 1.3.1 Anatomy of the Filum Terminale ................................................................................. 6 1.3.2 Pathogenesis of Tethered Cord Syndrome .................................................................... 7 1.3.3 Pathogenesis of Occult Tethered Cord Syndrome ........................................................ 8 1.3.4 Association of Occult Tethered Cord Syndrome and Ehlers-Danlos Syndromes ........ 9 1.4 Myodural Bridges .................................................................................................................. 10 1.4.1 Structure and Function of the Myodural Bridges ....................................................... 10 1.4.2 Myodural Bridge Dysfunction .................................................................................... 12 1.5 Statement of the Problem ....................................................................................................... 14 REFERENCES ............................................................................................................................. 15 CHAPTER 2: EVALUATION OF THE STRUCTURE AND FUNCTION OF MYODURAL BRIDGES IN AN EQUINE MODEL OF EHLERS-DANLOS SYNDROMES ......................... 21 2.1 Abstract .................................................................................................................................. 21 2.2 Introduction ............................................................................................................................ 22 2.3 Materials and Methods ........................................................................................................... 23 2.3.1 Animals ...................................................................................................................... 23 2.3.2 Neurologic Examinations ........................................................................................... 24 2.3.3 Ultrasonographic Examination .................................................................................. 24 2.3.4 Anatomy of the Equine Myodural Bridge ................................................................. 26 2.3.5 Myodural Bridge Histology ....................................................................................... 26 2.3.6 Myodural Bridge Ultrastructure ................................................................................. 26 2.3.7 Myodural Bridge Tissue Biomechanics .................................................................... 27 2.3.8 Sarcocystis Neurona Testing ..................................................................................... 28 2.4 Statistical Analysis ................................................................................................................. 28 2.5 Results .................................................................................................................................... 28 2.5.1 Neurologic Examinations ........................................................................................... 28 2.5.2 Ultrasonographic Examination .................................................................................. 29 2.5.3 Anatomy of the Equine Myodural Bridge ................................................................. 31 2.5.4 Myodural Bridge Histology ....................................................................................... 33 2.5.5 Myodural Bridge Ultrastructure ................................................................................. 36 2.5.6 Myodural Bridge Tissue Biomechanics ..................................................................... 38 vi 2.6 Conclusion ............................................................................................................................. 39 REFERENCES ............................................................................................................................. 40 CHAPTER 3: EVALUATION OF THE FILUM TERMINALE IN AN EQUINE MODEL OF EHLERS-DANLOS SYNDROMES ............................................................................................ 44 3.1 Abstract .................................................................................................................................. 44 3.2 Introduction ........................................................................................................................... 45 3.3 Materials and Methods ........................................................................................................... 46 3.3.1 Animals ...................................................................................................................... 46 3.3.2 Neurologic Examination ............................................................................................ 47 3.3.3 Necropsy .................................................................................................................... 48 3.3.4 Filum Terminale Histology ........................................................................................ 48 3.3.5 Filum Terminale Ultrastructure ................................................................................. 48 3.3.6 Adenovirus PCR ........................................................................................................ 49 3.4 Results .................................................................................................................................... 49 3.4.1 Neurologic Examination ............................................................................................ 49 3.4.2 Necropsy .................................................................................................................... 50 3.4.3 Filum Terminale Histology ........................................................................................ 50 3.4.4 Filum Terminale Ultrastructure ................................................................................. 53 3.4.5 Adenovirus PCR ........................................................................................................ 58 3.5 Conclusion ............................................................................................................................. 58 REFERENCES ............................................................................................................................. 60 vii LIST OF TABLES Table 1.1: Ehlers-Danlos syndromes subtypes, inheritance pattern (IP), genetic basis, and protein involved. AD= autosomal dominant. AR= autosomal recessive. From Malfait F., et al. (2017) The 2017 International Classification of the Ehlers-Danlos Syndromes. © 2017, John Wiley and Sons. Reproduced with permission from John Wiley and Sons. .................................. 3 Table 2.1: Identities of HERDA and control horses used in MDB study. ................................... 24 Table 2.2: Width of the subarachnoid space in 3 HERDA versus 3 control horses with the head positioned in neutral, flexion, and extension at the AO and AA spaces. Reported as mean with standard deviation. ........................................................................................................................ 30 Table 2.3: Average difference between the width of the subarachnoid space in flexion and neutral in 3 HERDA horses versus 3 control horses at the AO and AA spaces. Reported as mean with standard deviation. ................................................................................................................ 30 Table 2.4: Average difference between the width of the subarachnoid space in extension and neutral in 3 HERDA horses versus 3 controls at the AO and AA spaces. Reported as mean with standard deviation. ........................................................................................................................ 31 Table 2.5: Findings from TEM imaging of the MDB of HERDA and control horses. ............... 37 Table 3.1: Identities of HERDA and control horses used in FT study. ....................................... 47 Table 3.2: Comparison of TEM findings from the FT of HERDA and control horses. .............. 54 viii LIST OF FIGURES Figure 1.1: The human filum terminale internum (red) and filum terminale externum (green) surrounded by the cauda equina. From Saker E., et al. (2017) The Filum Terminale Internum and Externum: A Comprehensive Review. © 2016 Elsevier, Ltd. Reproduced with permission from Elsevier, Ltd. .......................................................................................................................... 7 Figure 1.2: Myodural bridges with connections to both the OCI and RCPma exist bilaterally at the atlanto-axial level. From Pontell M.E., Scali F., Marshall E., Enix D. (2012) The Obliquus Capitis Inferior Myodural Bridge. © 2012 Wiley Periodicals, Inc. Reproduced with permission from Wiley Periodicals, Inc. ......................................................................................................... 11 Figure 2.1: Ultrasonographic anatomy of the MDBs at the AA level. The spinal cord, dura mater, dentate ligaments, MDBs, and subarachnoid space are visible. The width of the subarachnoid space (|) was measured. .......................................................................................... 30 Figure 2.2: Atlanto-axial MDB (à) labelled with blue latex dye via ultrasound-guided injection. The AA MDBs spans from C1 (*) to C2 (*). ............................................................................... 31 Figure 2.3: The equine AO MDB with A=RCDmi, B=dense connective tissue MDB, and C=dura mater. ............................................................................................................................... 32 Figure 2.4: Control horse MDB stained with Masson’s trichrome. A=RCDmi, B=connective tissue, C=dura mater (20x). Image courtesy of Dr. Edward Stopa. ............................................. 33 Figure 2.5: Focal aggregate of lymphocytes and plasma cells in a 3-year-old HERDA horse gelding MDB stained with H&E (400x). ...................................................................................... 34 Figure 2.6: Lymphocytic perivascular cuffing in a 3-year-old HERDA horse gelding MDB stained with Masson’s trichrome (200x). ...................................................................................... 34 Figure 2.7: Lymphocytic perivascular cuffing (à) in a 3-year-old HERDA horse gelding MDB. CD3 immunohistochemistry (200x). ............................................................................................ 35 Figure 2.8: Comparison of density of collagen fibril packing in cross-section in age-matched HERDA vs. control horses. A=HERDA horse with loosely packed collagen fibrils, variation in fibril size, variation in fibril shape, and granulo-filamentous deposits in the interfibrillar space (14,000x). B=Control horse with densely packed collagen fibrils and variation in fibril size (14,000x). ...................................................................................................................................... 37 Figure 2.9: Abnormal collagen fibril orientation in the MDB of HERDA horses imaged with TEM. A= Whirling collagen fibrils in a 3-year-old gelding (10,000x). B= Hook shaped (à) and broken (à) collagen fibrils, as well as loosely packed collagen fibrils in a 6-year-old mare (27,000x). C= Granulo-filamentous deposits (à) in the interfibrillar space, as well as loosely ix packed collagen fibrils in a 3-year-old gelding (27,000x). D= Broken (à) and loosely packed collagen fibrils in a 2-year-old gelding (14,000x). ....................................................................... 38 Figure 3.1: The equine FT. The conus medullaris (à) and FT (à) are visible. The conus medullaris terminates at the level of S1 in the horse. ................................................................... 50 Figure 3.2: The overall structure of the HERDA FT is disorganized. A= FT from 21-year-old affected Quarter Horse mare displaying disorganized remnants of the central canal (à) (H&E) (40x). B= FT from 21-year-old affected Quarter Horse mare with disorganized elements of fibrous tissue (à) (Masson’s trichrome) (40x). C=FT from a 4-year-old control Quarter Horse mare with a normal central canal (à) (H&E) (40x). D=FT from a 4-year-old control Quarter Horse mare with normal fibrous tissue (à) (Masson’s trichrome) (40x). ................................... 51 Figure 3.3: Lymphocytic inflammation is present in the HERDA FT. A= FT from 6-year-old affected Quarter Horse mare. Large number of CD3 positive lymphocytes are present. Lymphocytic perivascular cuffing is present (100x). B= FT from 6-year-old control Morgan mare. Few scattered CD3 positive lymphocytes are present (100x). ........................................... 51 Figure 3.4: Common histologic changes in the EDS OTCS FT. A= Increased vascularity (à) (40x). B= Marginating polymorphonuclear cells within a vessel (à) (200x). C= Nerve twigs (à) (100x). D= Neuropil (*) (100x). Photos courtesy of Dr. Petra Klinge and Dr. Edward Stopa. ............................................................................................................................................ 53 Figure 3.5: Collagen fibril size, shape, and packing density in the FT of an affected horse versus a control (27,000x). A= Cross-sectional imaging of a 21-year-old affected Quarter Horse mare. Collagen fibrils are variable in size and shape and are loosely packed. B= Cross-sectional imaging of an 18-year-old control Quarter Horse mare. Collagen fibrils are normal in size with mild variation in shape. Packing density is moderate. C= Longitudinal imaging of a 21-year-old affected Quarter Horse mare. Collagen fibrils display a random, haphazard orientation and are broken. D= Longitudinal imaging of an 18-year-old control Quarter Horse mare. Fibrils are densely packed and display regular orientation. ........................................................................... 55 Figure 3.6: Abnormal TEM findings in a 21-year-old affected Quarter Horse mare. A= Hook shaped (à) and broken (à) collagen fibrils (14,000x). B= Disintegrating collagen fibrils (à) (14,000x). C= Whirling collagen fibrils and elastic fibers (*) (14,000x). D= Viral inclusion (à) (80,000x). ...................................................................................................................................... 56 Figure 3.7: TEM findings in the FT of a 45-year-old female with hypermobile Ehlers-Danlos syndrome. Collagen fibrils are disorganized and swollen. Beading (à) and corkscrewing (à) of the collagen is present (1 µm scale bar). Photo courtesy of Dr. John Donahue. ..................... 57 Figure 3.8: TEM findings in the FT of non-EDS OTCS patients. A= Focally abnormal ultrastructure in the FT of a 17-year-old male non-EDS OTCS patient (1 µm scale bar). Collagen fibrils are disorganized and mildly swollen. Areas of beading (à) and corkscrewing (à) are present. B= Normal ultrastructure in a 22-year-old male non-EDS OTCS patient (1 µm scale bar). Photos courtesy of Dr. John Donahue. ................................................................................ 58 x KEY TO ABBREVIATIONS AA- Atlanto-axial AO- Atlanto-occipital CSF- Cerebrospinal fluid CYP B- Cyclophilin B DPD- Deoxypridinoline EDS- Ehlers-Danlos Syndromes FT- Filum Terminale H&E- Hematoxylin & eosin hEDS- Hypermobile Ehlers-Danlos Syndrome HERDA- Hereditary equine regional dermal asthenia IHC- Immunohistochemistry kEDS- Kyphoscoliotic Ehlers-Danlos Syndrome LRCD- Lateral rectus capitis dorsal MDB/MDBs- Myodural bridge(s) OCA- Obliquus capitis anterior OCI- Obliquus capitis inferior OCP- Obliquus capitis posterior OTCS- Occult tethered cord syndrome PAOM-SDC- Posterior atlanto-occipital membrane-spinal dura complex PPIB- Peptidyl-Prolyl Cis-Trans Isomerase B PYD- Pyridinoline RCDma-Rectus capitis dorsalis major xi RCDmi- Rectus capitis dorsalis minor RCPma- Rectus capitis posterior major RCPmi- Rectus capitis posterior minor spEDS- Spondylodysplastic Ehlers-Danlos syndrome TCS- Tethered cord syndrome TEM- Transmission electron microscopy xii CHAPTER 1: REVIEW OF THE LITERATURE 1.1 Project Summary Ehlers-Danlos syndromes (EDS) are a group of inherited disorders of connective tissue that affect as many as 1 in 5,000 humans [1]. Some forms of EDS are associated with life threatening complications and chronic pain [2]. Despite the frequency with which the disease occurs, EDS is poorly researched. Hereditary equine regional dermal asthenia (HERDA), is a form of EDS that occurs in Quarter Horses. The horse is a naturally occurring model for EDS, as affected horses share similar abnormalities in their skin, joints, eyes, and great vessels [3-6]. Humans with EDS often present with a wide range of neurologic complications, including occult tethered cord syndrome (OTCS), Chiari I malformation, chronic headache, and craniocervical instability [7]. The neurologic status of HERDA horses has not previously been documented. The pathophysiology of OTCS in EDS is poorly understood. Occult tethered cord syndrome is caused by an abnormally inelastic filum terminale (FT), a remnant of embryonic spinal cord formed following caudal regression [8]. Myodural bridges (MDBs) are involved in the control of craniocervical dynamics, and are believed to prevent dural infolding during head and neck extension [9]. Myodural bridge laxity is theorized to contribute to the chronic headache phenotype observed in the EDS population due to disruption of cerebrospinal fluid (CSF) flow [10]. The present study seeks to provide mechanistic evidence of the role of connective tissue abnormalities in OTCS and MDB dysfunction. The aim of this research is to investigate the neurologic status of horses with HERDA, to characterize the normal equine FT and MDBs, to characterize pathologic changes in the FT and MDBs in HERDA horses, and to compare these structures to those in humans with EDS. Tissues collected at necropsy from 4 affected and 5 control horses were compared. In addition, findings 1 from both groups were compared to the large bank of human FT samples at Rhode Island Hospital, Department of Neurosurgery. All FT and atlanto-occipital (AO) MDB specimens underwent histopathologic and transmission electron microscopy (TEM) evaluation. Tissue biomechanical testing was performed on the atlanto-axial (AA) myodural bridge from affected and control horses. Equine CSF was banked for later molecular characterization. 1.2 Ehlers-Danlos Syndromes 1.2.1 Ehlers-Danlos Syndromes in Humans Ehlers-Danlos syndromes are a heterogeneous group of connective tissue disorders caused by mutations in genes encoding collagen or collagen processing [2]. Hippocrates was the first to describe an EDS-like syndrome when he noted nomads and Scythians with joint laxity and scarring in his book On Airs, Waters and Places in 400 BC [11]. Though earlier case reports were published, the disorder was ultimately named after Danish dermatologist Edvard Ehlers and French dermatologist Henri-Alexandre Danlos who published case reports in 1898 and 1908 describing joint hypermobility and skin hyperextensibility [12]. There are currently 13 distinct subtypes of EDS, as defined by the 2017 International Classification [2] [Table 1.1]. While clinical signs vary between the subtypes, all variants of EDS are associated with joint hypermobility, skin hyperextensibility, and tissue fragility which can include vascular and visceral fragility. Depending on the subtype of EDS, the mode of inheritance can be autosomal dominant or recessive. Genetic mutations have been identified for all known subtypes of EDS other than hypermobile EDS (hEDS), which is the most common form of the disorder and is estimated to occur at a rate of 1 in 5,000 [1]. 2 Table 1.1: Ehlers-Danlos syndromes subtypes, inheritance pattern (IP), genetic basis, and protein involved. AD= autosomal dominant. AR= autosomal recessive. From Malfait F., et al. (2017) The 2017 International Classification of the Ehlers-Danlos Syndromes. © 2017 John Wiley and Sons. Reproduced with permission from John Wiley and Sons. 3 Ehlers-Danlos syndromes can lead to a significant amount of pain and disability in affected patients. Chronic pain in EDS has been reported to affect 90% of EDS patients, and is reported to be most severe in patients with hEDS [13]. Not only is EDS associated with joint dislocations and subluxations, but neurologic abnormalities secondary to EDS are also common causes of pain [7, 14-15]. These include headache, OTCS, and Chiari I malformation, among others. Chopra et al. (2017) assessed data from multiple prior studies and concluded that approximately 75% of EDS patients suffer from chronic pain due to headache [15]. There are many proposed causes of headache in EDS, including migraine, temporomandibular joint dysfunction, and craniocervical instability [7]. Myodural bridge dysfunction has recently been proposed as a contributor to the chronic headache phenotype often observed in this patient population [10]. 1.2.2 Hereditary Equine Regional Dermal Asthenia Ehlers-Danlos syndromes have been identified in dogs, mink, cats, horses, cows, sheep, and rabbits [16-20]. Three subtypes of EDS have been identified in horses, though it is likely that other subtypes exist and have yet to be identified. Warmblood Fragile Foal syndrome is an autosomal recessive condition caused by a PLOD1 mutation that occurs in warmblood horses and is analogous to kyphoscoliotic EDS (kEDS) in humans [21]. A B4GALT7 mutation was recently identified in Friesian horses with dwarfism and joint laxity [22]. This condition appears to be similar to spondylodysplastic EDS (spEDS). Hereditary equine regional dermal asthenia (HERDA) is an autosomal recessive form of EDS biochemically and phenotypically similar to human kEDS that occurs in Quarter Horses. Lerner and McCracken first described HERDA in 1978 [23]. The mutation responsible for HERDA was identified in 2007 and is a c.115G>A missense mutation in peptidyl-prolyl cis-trans isomerase B (PPIB) encoding for cyclophilin B (Cyp B) [24]. This PPIB mutation has been 4 identified in Quarter Horses, American Paint Horses, and Appaloosas related to the Quarter Horse sire Poco Bueno [21]. The mutation is thought to have originated with Poco Bueno or his dam Miss Taylor, and was propagated by Poco Bueno’s extreme popularity [4]. Today, carrier frequency is highest among elite cutting and reining horses, where it is estimated to be 28.3%. Carrier frequency in the general Quarter Horse population is estimated to be 3.5% [26]. Horses with HERDA have fragile, hyperextensible skin and often develop large atrophic scars. Scarring is often concentrated over the dorsum presumably due to upregulation of matrix metalloproteinases secondary to ultraviolet irradiation. Lesions associated with HERDA, including seromas, hematomas, and skin sloughing, commonly develop between 18 and 24 months of age. The onset of saddle training may initiate the onset of these lesions or dramatically worsen existing lesions. Similar to their human counterparts, horses with HERDA often display joint hypermobility [4]. Additionally, increased PGE2 and matrix metalloproteinase activity in HERDA chondrocyte cultures suggests that this population is at an increased risk of developing osteoarthritis [27-28]. Many similarities exist between HERDA and kEDS in humans. Both disorders are autosomal recessive and cause skin fragility, atrophic scarring, and ocular fragility. In addition, both conditions cause abnormal ratios of deoxypyridinoline (DPD) and pyridinoline (PYD) crosslinks. The DPD:PYD ratio, a measure of collagen degradation products, is commonly elevated in the urine of kEDS patients, and is elevated in the urine, skin, and blood of HERDA horses [5, 29- 30]. 5 1.3 Tethered Cord Syndrome 1.3.1 Anatomy of the Filum Terminale The filum terminale, historically named the nervus impar, is often described as a fibrous, collagenous band which connects the conus medullaris with the coccyx. The conus medullaris is the caudal termination of the spinal cord. The filum terminale is composed of two segments, the cranial filum terminale internum and the caudal filum terminale externum [Figure 1.1]. The filum terminale internum, located within the dura, is approximately 15 cm long in adult humans [8]. It originates from the conus medullaris and terminates in the dural sac at S2 [7]. The conus medullaris terminates at the level of the middle of the body of L1 in most adults [8]. The filum terminale externum, also called the coccygeal ligament, is approximately 7.5 cm long in adults [31]. The filum terminale externum is extradural and runs from the dural sac to the periosteum of the coccyx [8, 32]. 6 Figure 1.1: The human filum terminale internum (red) and filum terminale externum (green) surrounded by the cauda equina. From Saker E., et al. (2017) The Filum Terminale Internum and Externum: A Comprehensive Review. © 2016 Elsevier, Ltd. Reproduced with permission from Elsevier, Ltd. The entirety of the filum terminale has not previously been described in the horse although it is known to originate from the conus medullaris at the level of S1 and terminate at the level of S4. The filum terminale internum and externum have not been described in horses, though the structure is often depicted as having an intradural and extradural segment [33]. 1.3.2 Pathogenesis of Tethered Cord Syndrome Tethered cord syndrome (TCS) occurs when there is tension on the spinal cord caused by an abnormal structure of the filum terminale within the vertebral column. The spinal cord is stabilized to the level of T12 by the dentate ligaments. Below this, the filum terminale 7 stabilizes the conus medullaris during flexion and extension of the spine and prevents stretch induced injury to the cord. When the filum terminale is abnormally inelastic, stretch-induced injury to the spinal cord can occur. Cellular injury occurs via a decrease in local blood flow and disruption of mitochondrial oxidative metabolism [34]. Spinal dysraphisms such as lipomyelomeningocele and myelomeningocele are often causes of TCS [35]. 1.3.3 Pathogenesis of Occult Tethered Cord Syndrome Unlike in TCS, patients with OTCS do not have any gross abnormalities of the filum terminale or surrounding tissues. Patients with OTCS are theorized to develop tethering after birth. It has been theorized that this patient population has abnormal tissue within the filum terminale at birth and repeated trauma to this abnormal structure is thought to be the eventual cause of tethering. This trauma may occur due to fibrotic changes to the FT, patient growth, repeated micro-trauma secondary to activity, or spinal stenosis [34]. Occult TCS is defined as a syndrome of TCS without radiographic findings of TCS, such as a low-lying conus, a thickened filum, or a fat signal. A low-lying conus ends below L2, and a thickened filum is greater than 2 mm in diameter. As in TCS, patients may have neurologic, orthopedic, urologic or gastrointestinal, and dermatologic symptoms. Neurologic symptoms range from lower motor neuron signs to upper motor neuron signs based on the age of the patient. Infants display decreased lower limb movements and reflexes consistent with lower motor neuron dysfunction. Young adults have spasticity and hyperreflexia consistent with upper motor neuron dysfunction. Other neurologic signs include sensory loss throughout the lumbar and sacral dermatomes. Children often have a history of toe-walking or in-toeing. Urinary symptoms are common and include neurogenic bladder, urinary incontinence, and chronic urinary tract infections. Gastrointestinal symptoms may include fecal incontinence and 8 constipation. Orthopedic abnormalities associated with TCS and OTCS include scoliosis, kyphosis, pes planus, pes cavus, and ankle pronation. In addition to these signs, most patients describe a burning, wandering pain in their low back, legs, and feet. Cutaneous signs commonly observed in TCS include a sacral dimple, cutaneous discoloration, dermal appendages, hypertrichosis, dermal sinus, or asymmetric gluteal cleft. Cutaneous signs are not always apparent in patients with OTCS. While symptomology and diagnostics such as urodynamic testing are highly suggestive of OTCS, the diagnosis cannot be confirmed until the filum terminale is visualized at the time of surgery [7, 36-37]. Urodynamic testing is an important diagnostic tool, as neurogenic bladder is a common finding in both TCS and OTCS. Neurogenic bladder secondary to TCS or OTCS may present as incontinence or urinary retention. Henderson et al. (2017) found that urodynamic findings consistent with TCS or OTCS include a bladder capacity of 800 mL or greater, detrusor sphincter dysynergia, and a large post-void residual volume [7]. Lavallée et al. (2013) found that 85% of children with abnormal urodynamic testing had radiographic evidence of TCS, supporting the use of this test [38]. 1.3.4 Association of Occult Tethered Cord Syndrome and Ehlers-Danlos Syndromes While the incidence of TCS in the general population is 0.1%, the incidence of OTCS in the EDS population is unknown, but is thought to be significantly higher than in the general population [7, 39]. Classical TCS can occur in the EDS population, but approximately 95% of EDS patients present with OTCS. The relationship between EDS and OTCS is not fully understood. Dr. Klinge, a neurosurgeon treating OTCS in EDS patients at Rhode Island Hospital, theorizes that EDS patients undergo abnormal retrogressive differentiation, leaving them with elements of mesoderm in the FT. This is thought to explain the abnormal histologic 9 findings observed in EDS OTCS patients, such as inflammatory cells, vascular lacunae, marginating polymorphonuclear cells, “nerve twigs”, increased elastic fibers, and prominent ependymal tissue. Unlike patients with classical TCS, patients with EDS and OTCS often have little adipose tissue in the filum terminale [40-42]. This abnormal tissue may allow for tethering of the filum terminale in adulthood due to chronic injury to this structure secondary to hypermobility of the spine of EDS patients. 1.4 Myodural Bridges 1.4.1 Structure and Function of the Myodural Bridges In 1995 Hack et al. described bilateral connective tissue bridges spanning from the rectus capitis posterior minor (RCPmi) to the cervical dura mater at the AO interspace in humans. These structures are known as MDBs [9]. Rather than fusing directly to the cervical dura mater, the RCPmi MDBs are thought to attach to the dura mater via the posterior atlanto-occipital membrane-spinal dura complex (PAOM-SDC) [9, 43-44]. Myodural bridges have since been found to exist at both the AO and AA levels [45]. Two additional bilateral MDBs originating from fascia from both the rectus capitis posterior major (RCPma) and the obliquus capitis inferior (OCI) attach to the cervical dura mater at the AA level [Figure 1.2] [46]. In 2006 Zumpano et al. dissected 75 human cadavers at the AO level and found bilateral MDBs spanning from the RCPMi to the PAOM-SDC in 93% of cadaveric specimens [47]. The observation that the posterior dura mater is thicker than the anterior dura mater from C1 to C3 supports the posterior attachment of the MDBs in this region [48]. 10 Figure 1.2: Myodural bridges with connections to both the OCI and RCPma exist bilaterally at the atlanto-axial level. From Pontell M.E., Scali F., Marshall E., Enix D. (2012) The Obliquus Capitis Inferior Myodural Bridge. © 2012 Wiley Periodicals, Inc. Reproduced with permission from Wiley Periodicals, Inc. New evidence suggests that MDBs are conserved across species. A recent study identified these structures in the rhesus macaque, European rabbit, dog, cat, Norway rat, guinea pig, and the Indoasian finless porpoise using plastinated specimens and histologic sections. The suboccipital musculature, termed the post-occipital musculature in animals, was similar to that of humans in all but the Indoasian finless porpoise. In the rhesus macaque, the European rabbit, dog, cat, Norway rat, and guinea pig, the post-occipital musculature consisted of the rectus capitis dorsalis major (RCDma), rectus capitis dorsalis minor (RCDmi), obliquus capitis anterior (OCA), and obliquus capitis posterior (OCP). These muscular connections are similar to the RCPma, RCPmi, and OCI in man. The Indoasian finless porpoise was found to have a RCDma, RCDmi, and lateral rectus capital dorsal (LRCD) muscle. No oblique muscles were present. Additionally, in all but the Indoasian finless porpoise, the RCDmi MDBs were found to insert onto the PAOM-SDC, while the MDBs inserted directly onto the cervical dura mater in the porpoise [49]. No published work examines the presence of MDBs in the horse. 11 The RCPma and OCI MDBs have been evaluated histologically in humans [50]. The RCPma MDBs contain both connective tissue and proprioceptive nerve fibers which span from the RCPma to the dura mater. The OCI MDBs are connective tissue bridges originating from the OCI and inserting on the dura mater. Nerve fibers were also present within the OCI MDBs. The RCDmi MDBs were evaluated histologically in cats, Norway rats, European rabbits, and guinea pigs. The RCDmi MDBs in these animals were also found to be connective tissue structures that originated in the muscle. As was seen in the plastinated specimens, the MDBs connected directly to the PAOM-SDC in some animals, while in others they traversed the PAOM-SDC and connected directly to the dura mater. Animal specimens used in this study were not stained to demonstrate the presence of nervous tissue [49]. 1.4.2 Myodural Bridge Dysfunction Myodural bridges are theorized to stabilize the cervical dura mater during extension of the head in order to prevent infolding of the dura [9]. Myodural bridge dysfunction is theorized to occur via a variety of mechanisms. Hack and Hallgren (2004) described a patient with chronic headaches and RCPmi hypertrophy [51]. They theorized that the patient’s headaches occurred due to chronic tension on the RCPmi MDBs leading to tension on the dura mater, which is innervated by the C1, C2, and C3 spinal nerves [52]. Tension on the dura is thought to have been the source of this patient’s headaches. A “myodural release” was performed in which the MDBs were transected, and the patient reported significant relief following the procedure [51]. This report is supported by numerous reports of chronic headache caused by myodural adhesions following suboccipital craniotomy and craniectomy. Reports of the incidence of headache following suboccipital craniotomy or craniectomy range from 9-64% [53-55]. Soumekh et al. (1996) followed seven patients who underwent cranioplasty for treatment of intractable headache 12 pain following a suboccipital craniotomy or craniectomy. All experienced significant symptom relief. This study also evaluated 56 patients who underwent cranioplasty, which separates the dura from the musculature, at the time of their initial surgery. None of the patients who underwent cranioplasty at the time of their initial surgery complained of post-operative headaches. These results further support the role of myodural adhesions in chronic headache pain [55]. Conversely, atrophy of the RCPmi and RCPma has also been linked to chronic head and neck pain. Patients with chronic neck pain are often reported to have atrophy and fatty infiltration of the RCPmi and RCPma. Weakening and atrophy of the RCPmi and RCPma may lead to laxity of the MDBs and secondary infolding of the dura mater which may lead to a disruption in CSF flow [56]. Interestingly, in addition to its role in cervical pain, RCPmi atrophy has also been linked to increased symptom severity and longer recovery times in patients with mild traumatic brain injuries. This is thought to be due to a smaller RCPmi absorbing less energy during a traumatic event, and is therefore unable to effectively stabilize the head. Increased changes in head velocity during a traumatic brain injury increase the likelihood of post-concussion syndrome. The presence of an atrophied RCPmi at the time of a traumatic brain injury also leads to a greater risk of injury to the MDBs [57]. In patients with EDS, it is theorized that abnormal connective tissue within the MDBs leads to laxity, and ultimately to infolding of the cervical dura mater [10]. This infolding in turn disrupts the flow of CSF through the subarachnoid space [58]. The resulting abnormal CSF dynamics are a potential cause of the chronic cervicogenic headache phenotype observed in the EDS population [10]. Heart rate and respiratory rate have long been known to influence CSF flow dynamics, but recent studies have suggested that head movement can influence CSF dynamics by 13 propelling CSF through the craniocervical junction in a cranial direction. Myodural bridges are hypothesized to play a role in this via their action on the cervical dura mater during head rotation [59]. CSF dynamics are also thought to play a role in the pathogenesis of chronic headache, as increased CSF flow velocity in patients with Chiari I malformation is thought to result in the headaches and syringomyelia often observed in this population [60-61]. 1.5 Statement of the Problem There is a gap in knowledge in our understanding of the relationship between OTCS and EDS. Little is known about why tethering occurs in the EDS population, and what the true effects of long-term tethering are. Additionally, there is a gap in knowledge in our understanding of the role of MDB dysfunction in the role of chronic headache in the EDS population. This project is a pilot study to evaluate the usefulness of an equine model of EDS in the study of OTCS, and to evaluate the structure and function of MDBs in an equine model of EDS. 14 REFERENCES 15 REFERENCES 1. Levy H.P. Ehlers-danlos syndrome, hypermobility type. 2004 Oct 22 [Updated 2016 Mar 31]. In: Adam M.P., Ardinger H.H., Pagon R.A., et al., editors. GeneReviews ® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2018. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1279/ 2. Malfait F., et al. (2017) The 2017 international classification of the Ehlers-Danlos syndromes. Am J Med Genet C Semin Med Genet. 175(1): 8-26. 3. White, S.D., et al. (2004) Hereditary equine regional dermal asthenia (‘hyperelastosis cutis’) in 50 horses: clinical, histologic, immunohistologic and ultrastructural findings. Vet Dermatol. 15: 207-217. 4. Rashmir-Raven A. (2013). Heritable equine regional dermal asthenia. Vet Clin North Am Equine Pract. 29(3): 689-702. 5. Mochal, C.A., et al. (2010) Ocular findings in Quarter Horses with hereditary equine regional dermal asthenia. J Am Vet Med Ass. 237: 304-310. 6. Brinkman E.L., et al. (2017) Cardiac findings in quarter horses with heritable equine regional dermal asthenia. JAVMA. 250(5): 538-547. 7. Henderson F.C., et al. (2017) Neurologic and spinal manifestations of the Ehlers-Danlos syndromes. Am J Med Genet C Semin Med Genet. 175(1): 195-211. 8. Saker E., et al. (2017) The filum terminale internum and externum: a comprehensive review. J Clin Neurosci. 40: 6-13. 9. Hack, G.D., Koritzer, R.T., Robinson, W.L., Hallgren, R.C., Greenman, P.E. (1995) Anatomic relation between the rectus capitis posterior minor muscle and the dura mater. Spine. 20: 2484–2486. 10. Castori M., et al. (2015) Connective tissue, Ehlers-Danlos syndrome(s), and head and cervical pain. Am J Med Genet C Semin Med Genet.169C: 84-96. 11. Hippocrates (400BC) On Airs, Waters, and Places (F. Adams, Trans.). Retrieved from: http://classics.mit.edu/Hippocrates/airwatpl.mb.txt 12. Gazit Y., Jacob G., Graham R. (2016) Ehlers-Danlos syndrome-hypermobility type: a much neglected multisystemic disorder. Rambam Maimonides Med J. 7(4): e0034. 13. Voermans N.C., et al. (2009) Neuromuscular involvement in various types of Ehlers-Danlos syndrome. Ann Neurol. 65(6): 687-697. 16 14. Cazzato D., et al. (2016) Small fiber neuropathy is a common feature of Ehlers-Danlos syndromes. Neurology. 87(2): 155-159. 15. Chopra P., et al. (2017) Pain management in Ehlers-Danlos syndomes. Am J Med Genet C Semin Med Genet. 175(1): 212-219. 16. Hegreberg G.A. (1975) Animal model of human disease: Ehlers-Danlos syndrome. Am J Path. 79(2): 383-386. 17. Weingart C., Haußer I., Kershaw O., Kohn B. (2014) Ehlers-Danlos-like syndrome in a cat. Schweiz Arch Tierheilkd. 156(11): 543-548. 18. Carty C.I., et al. (2016) Dermatosparaxis in two Limousin calves. Ir Vet J. 69: 15. 19. Monteagudo L.V., et al. (2015) In silico identification and three-diminsional modelling of the missense mutation in ADAMTS2 in a sheep flock with dermatosparaxis. Vet Dermatol. 26(1): 49-52, e15-6. 20. Iglauer F., Wilmering G., Huisinga E., Wölm M., Lorke D.E. (1999) Cutaneous asthenia (Ehlers-Danlos syndrome) in a domestic rabbit. Dtsch Tierarztl Wochenschr. 106(12): 500- 505. 21. Monthoux C., de Brot S., Jackson M., Bleul U., Walter J. (2015) Skin malformations in a neonatal foal tested homozygoes positive for warmblood fragile foal syndrome. BMC Vet Res. 11(12): 1-8. 22. Leegwater P.A., et al. (2016) Dwarfism with joint laxity in Friesian horses is associated with a splice site mutation in B4GALT7. BMC Genomics. 17(1): 839. 23. Lerner D.J., McCracken M.D. (1978) Hyperelastosis cutis information from vetstream equis. J Eq Med Surg. 2: 350–352. 24. Tryon, R.C., White, S.D., Bannasch, D.L. (2007) Homozygosity mapping approach identifies a missense mutation in equine cyclophilin B (PPIB) associated with HERDA in American Quarter Horses. Genomics. 90: 93-102. 25. White, S.D., et al. (2004) Hereditary equine regional dermal asthenia (‘hyperelastosis cutis’) in 50 horses: clinical, histologic, immunohistologic and ultrastructural findings. Vet Dermatol. 15: 207-217. 26. Tryon, R.C., et al. (2009) Evaluation of allele frequencies of inherited disease genes in subgroups of American Quarter Horses. J Am Vet Med Ass. 234: 120-125. 27. Gardner L.C., et al. (2011) Articular cartilage from horses afflicted with hereditary equine regional dermal asthenia (HERDA) exhibits osteoarthritic lesions and isolated chondrocytes 17 produce high levels of the pro-inflammatory mediator prostaglandin E2. J Vet Intern Med. 25: 632–767. 28. Dorris M.L., et al. (2011) Increased metalloproteinase activity in chondrocytes from articular cartilage of horses with hereditary equine regional dermal asthenia. J Vet Intern Med. 25: 632–767. 29. Yeowell H.N., Steinmann B. Ehlers-Danlos Syndrome, Kyphoscoliotic Form. 2000 Feb 2 [Updated 2013 Jan 24]. In: Adam M.P., Ardinger H.H., Pagon R.A., et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993- 2018. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1462/ 30. Rashmir-Raven A., Spier S.J. (2015) Hereditary equine regional dermal asthenia (HERDA) in quarter horses: a review of clinical signs, genetics and research. Equine Vet Educ. 27(11): 604- 611. 31. Tubbs R.S., et al. (2005) The filum terminale externum. J Neurosurg Spine. 3(2): 149-152. 32. De Vloo P., Monea A.G., Sciot R., van Loon J., Van Calenbergh F. (2012) The filum terminale: a cadaver study of anatomy, histology, and elastic properties. World Neurosurg. 90: 565-573. 33. Budras K.D., Sack W.O., Rock S. (2003) Anatomy of the horse: an illustrated text (4th ed.). Hannover, Germany: Schlutersche. 34. Yamada S., et al. (2007) Pathophysiology of tethered cord syndrome and similar complex disorders. Neurosurg Focus. 23(2): E6. 35. Khoshhal K.I., Murshid W.R., Elgamal E.A., Salih M.A.M. (2012) Tethered cord syndrome: a study of 35 patients. J Taibah Univ Sci. 7(1): 23-28. 36. Fabiano A.J., Khan M.F., Rozzelle C.J., Li V. (2009) Preoperative predictors for improvement after surgical untethering in occult tight filum terminale syndrome. Pediatr Neurosurg. 45: 256-261. 37. Klinge, P. (2015) Tethered cord syndrome in Ehlers-Danlos- diagnostic criteria, surgical indication and filum pathology. Presented at the Ehlers-Danlos National Foundation Learning Conference, Baltimore, MD. 38. Lavallée L.T., Leonard M.P., Dubois C., Guerra L.A. (2013) Urodynamic testing—is it a useful tool in the management of children with cutaneous stigmata of occult spinal dysraphism? J Urol. 189(2): 678-683. 39. Bademci G., et al. (2006) Prevalence of primary tethered cord syndrome associated with occult spinal dysraphism in primary school children in Turkey. Pediatr Neurosurg. 42(1): 4- 13. 18 40. Astafiev S., Donahue J., Stopa E., Klinge P. (2017) Occult tethered cord syndrome: a histopathological characterization of the filum terminale in Ehlers-Danlos and non-Ehlers- Danlos patients. The Warren Alpert Medical School of Brown University Sumer Research Showcase, Providence, RI. 41. Klinge P. (2015) Histopathological evidence of clinically relevant filum pathology in tethered cord syndromes with little radiographic evidence including hypermobility syndromes. Presented at the Chiari & Syringomyelia Foundation Research Colloquium, New Orleans, LA. 42. Klinge P., personal communication, January 28, 2018. 43. Mitchell B.S., Humphreys B.K., O’Sullivan E. (1998) Attachments of the ligamentum nuchae to cervical posterior spinal dura and the lateral part of the occipital bone. J Manipulative Physiol Ther. 21(3): 145-148. 44. Kahkeshani K., Ward P.J. (2012) Connection between the spinal dura mater and the cord syndrome produced by slow traction. Sci Rep. 5: 9116. 45. Yuan X.Y., et al. (2016) Patterns of attachment of the myodural bridge by the rectus capitis posterior minor muscle. Anat Sci Int. 91(2): 175-179. 46. Pontell M.E., Scali F., Marshall E., Enix D. (2012) The obliquus capitis inferior myodural bridge. Clin Anat. 26(4): 450-454. 47. Zumpano M.P., Hartwell S., Jagos C.S. (2006) Soft tissue connection between rectus capitus posterior minor and the posterior atlanto-occipital membrane: a cadaveric study. Clin Anat. 19(6):522-527. 48. Taylor J.R., Taylor M.M., Twomey L.T. (1996) Posterior cervical dura is much thicker than the anterior cervical dura [letter]. Spine. 21(19):2300. 49. Zheng N., et al. (2017) The universal existence of myodural bridge in mammals: an indication of a necessary function. Sci Rep. 6:8248. 50. Scali F., Pontel M.E., Enix D.E., Marshall E. (2013) Histological analysis of the rectus capitis posterior major’s myodural bridge. Spine J. 13(5):558-563. 51. Hack G.D., Hallgren R.C. (2004) Chronic headache relief after section of suboccipital muscle dural connections: a case report. Headache. 44(1):84-89. 52. Alix M.E., Bates D.K. (1999) A proposed etiology of cervicogenic headache: the neurophysiologic basis and anatomic relationship between the dura mater and the rectus posterior capitis minor muscle. J Manipulative Physiol Ther. 22(8):534-539. 19 53. Schaller B., Baumann A. (2003) Headache after removal of vestibular schwannoma via the retrosigmoid approach: a long-term follow-up-study. Otolaryngol Head Neck Surg. 128(3):387-395. 54. Schessel D.A., Rowed D.W., Nedzelski J.M., Feghali J.G. (1993) Postoperative pain following excision of acoustic neuroma by the suboccipital approach: observations on possible cause and potential amelioration. Am J Otol. 14(5):491-494. 55. Soumekh B., Levine S.C., Haines S.J., Wulf J.A. (1996) Retrospective study of postcraniotomy headaches in suboccipital approach: diagnosis and management. Am J Otol. 17(4):617-619. 56. McPartland J.M., Brodeur R.R., Hallgren R.C. (1997) Chronic neck pain, standing balance, and suboccipital muscle atrophy—a pilot study. J Manipulative Physiol Ther. 20(1):24-29. 57. Fakhran S., Qu C., Alhilali L.M. (2016) Effect of the suboccipital musculature on symptom severity and recovery after mild traumatic brain injury. AJNR Am J Neuroradiol. 37(8):1556- 1560. 58. Hallgren R.C., Hack G.D., Lipton J.A. (1997) Clinical implications of a cervical myodural bridge. AAO J. 7:30–34. 59. Xu Q., et al. (2016) Head movement, an important contributor to human cerebrospinal fluid circulation. Sci Rep. 6:31787. 60. Hentschel S., Mardal K.A., Løvgren A.E., Linge S., Haughton V. (2010) Characterization of cyclic CSF flow in the foramen magnum and upper cervical spinal canal with MR flow imaging and computational fluid dynamics. AJNR Am J Neuroradiol. 31(6):997-1002. 61. Wolpert S.M., Bhadelia R.A., Bogdan A.R., Cohen A.R. (1994) Chiari I malformations: assessment with phase-contrast velocity MR. AJNR Am J Neuroradiol. 15:1299-1308. 20 CHAPTER 2: EVALUATION OF THE STRUCTURE AND FUNCTION OF MYODURAL BRIDGES IN AN EQUINE MODEL OF EHLERS-DANLOS SYNDROMES 2.1 Abstract Our objective was to determine if myodural bridges are present in horses, and to determine if myodural bridge dysfunction exists in horses with hereditary equine regional dermal asthenia (HERDA). Horses with HERDA are a model of Ehlers-Danlos syndromes in humans. Neurologic examinations were performed on 3 HERDA and 5 control horses. Ultrasonographic examination of the myodural bridges was performed on 3 HERDA and 4 control horses. Histology and transmission electron microscopy was performed on the myodural bridges from 4 HERDA and 5 control horses. CD3 and CD20 immunohistochemistry was performed on 4 HERDA and 5 control horse myodural bridges. Tissue biomechanical testing was performed on myodural bridges from 2 HERDA and 4 control horses. Neurologic examination was abnormal in 1 HERDA horse. Evidence of dural infolding in HERDA horses was present on ultrasonographic imaging. Lymphocytic perivascular cuffing was observed in 1 HERDA horse, while myofiber degeneration was observed in 3 HERDA and 1 control horse. Ultrastructural examination revealed loosely packed collagen fibrils with abnormal orientation in 3 HERDA horse myodural bridges, and mild abnormalities in 2 control horse myodural bridges. Tissue biomechanical testing found that the cross-sectional area of the myodural bridges was reduced in HERDA horses, and that HERDA horses had more strain at complete pull-apart than control horses. The HERDA myodural bridge is abnormal. This supports the theory that myodural bridge dysfunction occurs in the human Ehlers-Danlos syndromes population. 21 2.2 Introduction Myodural bridges (MDBs) are bilateral dense bands of connective tissue that connect the suboccipital musculature with the dura mater at the atlanto-occipital (AO) and atlanto-axial (AA) levels. These structures were first described by Hack et al. (1995) in humans [1]. They have since been described in several animal species, including rhesus macaques, European rabbits, dogs, cats, Norway rats, guinea pigs, and the Indoasian finless porpoise [2]. Myodural bridges have not previously been identified in horses. The MDBs are theorized to stabilize the dura mater during extension of the head and neck, thus preventing infolding of the dura and disruption of the flow of cerebrospinal fluid (CSF) [3]. The MDBs are also thought to function as a pump in conjunction with the suboccipital musculature in order to power CSF flow [4]. Humans with Ehlers-Danlos syndromes (EDS) have a high incidence of chronic headache, with as many as 75% of patients reporting some degree of chronic head pain [5]. Headache secondary to MDB dysfunction has been described in a patient with rectus capitis posterior minor muscle (RCPmi) hypertrophy, as well as due to myodural adhesions following suboccipital craniotomy or craniectomy [6-10]. Additionally, chronic neck pain has been reported in patients with RCPmi and rectus capitis posterior major muscle (RCPma) atrophy and fatty infiltration [11]. Dysfunction of the MDBs is theorized to lead to cervicogenic headache in EDS due laxity of connective tissue within the MDBs leading to infolding of the dura mater into the subarachnoid space and disruption of cerebrospinal fluid (CSF) flow [12]. The purpose of this study is to identify MDBs in horses, and to examine the function of this structure in an equine model of EDS. Horses with hereditary equine regional dermal asthenia (HERDA) were used as a naturally occurring animal model of EDS due to their similarities to humans with kyphoscoliotic EDS (kEDS) [13-14]. Hereditary equine regional dermal asthenia is 22 an autosomal recessive disorder that occurs in Quarter Horses and is caused by a mutation in peptidyl-prolyl cis-trans isomerase B (PPIB) encoding for cyclophilin B (Cyp B) [15]. Horses with HERDA have fragile, hyperextensible skin as well as joint hypermobility, ocular fragility, and abnormalities within their great vessels [14, 16-19]. The neurologic status of HERDA horses has not previously been documented. 2.3 Materials and Methods 2.3.1 Animals A total of 10 horses (5 HERDA and 5 control) were evaluated [Table 2.1]. The HERDA group consisted of 4 donated Quarter Horses and 1 client-owned Paint Horse. The control horses consisted of 3 Quarter Horses, 1 Arabian, and 1 Morgan, all of which were donated. Genetic testing and physical examination were required for study enrollment for Quarter Horses and Paint Horses. Genetic testing was performed on 3 HERDA horses prior to study enrollment. All were confirmed to be homozygous positive, H/H, for the PPIB mutation via DNA testing conducted on hair root samples at UC Davis Veterinary Genetics Laboratory. The remaining 2 HERDA horses were tested via PCR testing on hair root samples by the master’s candidate (AM) at Michigan State University and were also homozygous positive, H/H, for the PPIB mutation. All HERDA horses displayed a phenotype consistent with the disorder. All 3 control Quarter Horses were confirmed to be homozygous negative, N/N, for the PPIB mutation via PCR testing on hair root samples performed by the master’s candidate (AM) at Michigan State University, and were phenotypically normal. Carriers of the PPIB mutation, H/N, were not evaluated. Owner consent was obtained prior to study enrollment for the privately-owned horse. Experimental protocols were approved by the Michigan State University institutional animal care and use committee. 23 HORSE ID BREED HERDA 1 Quarter Horse SEX Mare AGE 21 yrs HERDA 2 Quarter Horse Mare 6 yrs HERDA 3 Quarter Horse HERDA 4 Quarter Horse HERDA 5 Paint Control 1 Quarter Horse Control 2 Morgan Gelding Gelding Mare Mare Mare Control 3 Quarter Horse Gelding Control 4 Arabian Mare Control 5 Quarter Horse Mare 3 yrs 2 yrs 10 yrs 18 yrs 6 yrs 3 yrs 2 yrs 2 yrs Study Role Neurologic exam, ultrasonographic exam, necropsy, histopathology, TEM, tissue biomechanical testing Neurologic exam, ultrasonographic exam, necropsy, histopathology, TEM, tissue biomechanical testing Necropsy, histopathology, TEM Necropsy, histopathology, TEM Neurologic exam, ultrasonographic exam Neurologic exam, ultrasonographic exam, necropsy, histopathology, TEM, tissue biomechanical testing Necropsy, histopathology, TEM Neurologic exam, ultrasonographic exam, necropsy, histopathology, TEM, tissue biomechanical testing Neurologic exam, ultrasonographic exam, necropsy, histopathology, TEM, tissue biomechanical testing Neurologic exam, ultrasonographic exam, necropsy, histopathology, TEM, tissue biomechanical testing Table 2.1: Identities of HERDA and control horses used in MDB study. 2.3.2 Neurologic Examinations Static and dynamic neurologic examinations were performed on 3 HERDA horses and 4 control horses. Examinations were performed by a boarded large animal internist (EC) in conjunction with the master’s candidate (AM) on all but 1 horse, who was examined by the master’s candidate (AM) and a boarded large animal surgeon (AR). 2.3.3 Ultrasonographic Examination A novel method for ultrasonographic examination of the MDBs was developed. Past studies in humans have attempted to use MRI imaging to visualize the MDBs; however, ultrasound provides a more dynamic means of imaging. Prior MRI and ultrasonographic studies in horses 24 have identified the AO membrane, but did not identify MDBs [20-21]. Ultrasonographic imaging allows for visualization of motion of both the MDBs and the dura mater. Three HERDA horses and 4 controls were imaged [Table 2.1]. For this study, 6 horses were imaged with a Mindray TE7 ultrasound and a 2-4 MHz cardiac probe. These horses were imaged by both a large animal surgeon (JM) and the master’s candidate (AM). One horse was imaged with an Esaote ultrasound and a convex 1-8 MHz probe. This horse was imaged by the master’s candidate (AM). A dorsolateral approach was used to image the attachment of the MDBs to the dura mater at the AO level, and a lateral approach was used to image this attachment at the AA level. Horses were imaged with their heads positioned in neutral, flexion, and extension. Angulation was measured using a HALO Next Gen digital goniometer, and was measured at the angle between the bottom of the mandible and the jugular groove. Neutral was defined as 105º to 110º ±2º, flexion as 85º to 95º ±2º, and extension as 115º to 120º ±2º [22]. Results from 1 control, an Arabian, were excluded due to significant breed related differences in head and neck conformation. The width of the subarachnoid space was measured from the dorsal spinal cord to the dorsal dura mater at the bifurcation of the MDBs. Measurements taken at the subarachnoid space were compared between HERDA and control horses. Each measurement was repeated 3 times to ensure reliability. In order to confirm that the structures viewed on ultrasound were in fact the MDBs, ultrasound guided injection of latex dye was performed at the AO and AA levels immediately following euthanasia in 2 control horses. An 18-gauge spinal needle was used to inject 2-3 mL of blue latex dye. 25 2.3.4 Anatomy of the Equine Myodural Bridge Anatomic examination of the MDBs was performed at necropsy in 4 HERDA and 5 control horses [Table 2.1]. Examination was performed by the master’s candidate (AM), a neurosurgeon (PK), and a boarded veterinary pathologist (DS). Necropsy was performed within 1 hour of euthanasia in all but 1 control horse. Bilateral AO and AA MDBs were successfully identified in all horses. 2.3.5 Myodural Bridge Histology The AO MDBs were collected at necropsy from 4 HERDA horses and 5 control horses, and were submitted for histopathologic examination [Table 2.1]. Specimens were fixed for at least 24 hours in 10% neutral buffered formalin prior to sectioning. All specimens were stained with hematoxylin and eosin (H&E), as well as Masson’s trichrome. CD3 and CD20 immunohistochemistry (IHC) was performed on 4 HERDA and 5 control MDBs. Efforts were made to section all specimens in the same orientation, and to include the RCDmi, MDB, and dura mater in each section. Slides were processed at both the Michigan State University Veterinary Diagnostics Laboratory and Rhode Island Hospital. 2.3.6 Myodural Bridge Ultrastructure The AO MDBs from 4 HERDA and 5 control horses were submitted for TEM. Connective tissue from the cranioventral aspect of the bridge, adjacent to the junction of the dense connective tissue and dura mater, was submitted for processing. Samples were formalin fixed for 24 hours prior to transfer to glutaraldehyde in preparation for TEM processing. After primary fixation in glutaraldehyde, samples were washed with 0.1M cacodylate buffer. Samples were then postfixed with 1% osmium tetroxide in 0.1M cacodylate buffer, dehydrated in a gradient series of acetone, and infiltrated and embedded in Spurr. A Power Tome 26 Ultramicrotome (RMC, Boeckeler Instruments. Tucson, AZ) was used to cut 70 nm thin sections and post stained with uranyl acetate and lead citrate. Sample processing was conducted at Rhode Island Hospital and Michigan State University Center for Advanced Microscopy. The MDBs were imaged at 14,000x, 27,000x, and 40,000x in longitudinal and cross-section using a JEOL 100CX Transmission Electron Microscope (Japan Electron Optics Laboratory, Japan) at an accelerating voltage of 100kV [23]. Imaging at higher or lower magnification was performed when indicated. 2.3.7 Myodural Bridge Tissue Biomechanics A pilot study on AA MDB tissue biomechanics was conducted using 2 HERDA and 4 control MDBs. The AA MDBs were removed at necropsy and frozen at -80°C until the time of use. At the time of testing, samples were thawed and the dense connective tissue was dissected out. A double-bladed scalpel was used to take four 3 mm samples from each horse. Two samples were taken from the medial aspect of each of the 2 AA MDBs from each horse. Sample thickness in the dorso-ventral plane was measured using a calibrated micrometer. The samples were loaded into custom-made saw-tooth clamps, and pieces of emory board were used to secure the sample ends between the clamps. Samples were then loaded onto a custom-made material testing system with a 25-lb load cell (Sensotec, Columbus, OH), a linear variable differential transformer (Lucas Schaevitz, Pennsauken, NJ), and a motion controller (Newport, Fountain Valley, CA). Samples were submerged in 0.9% saline solution for the duration of testing. The initial sample length was measured under a 50-gram tare load. Uniaxial tension was applied at a constant speed of 0.41 mm/s until failure. Force values were recorded at 10 Hz throughout the test. An analogue-to-digital computer data acquisition system was used to measure load values and rate of displacement. Values for load and rate of displacement were then used to calculate 27 stress and strain based on measurements of cross-sectional area and initial MDB sample length. To determine tensile modulus, a stress-strain curve was constructed and a best-fit line was applied to the most linear portion of the curve [24]. Cross-sectional area, maximum stress, strain at maximum stress, strain at complete pull-apart, and tensile modulus were then compared between HERDA and control horses. 2.3.8 Sarcocystis Neurona Testing Cerebrospinal fluid from 1 HERDA horse, a 21-year-old mare, was submitted for a western blot for antibody for Sarcocystis neurona due to the presence of sarcocysts in the RCDmi. Cerebrospinal fluid was obtained via centesis of the cerebellomedullary cistern immediately following euthanasia, and was frozen at -80°C prior to testing. 2.4 Statistical Analysis An unpaired Student t test was used to determine whether there was a significant difference between HERDA and control horse ultrasonographic measurements of the subarachnoid space taken in neutral, flexion, and extension. An unpaired Student t test was also used to determine whether there was a significant difference in the cross-sectional area, maximum stress, strain at maximum stress, strain at complete-pull apart, and modulus between HERDA and control horses. Significance was set at p<0.05. 2.5 Results 2.5.1 Neurologic Examinations All control horse examinations were normal. Abnormalities were noted in 1 HERDA horse, a 21-year-old Quarter Horse mare, that displayed mild truncal sway when blindfolded and a base- narrow, single-track gait in the hind limbs when blindfolded with the head elevated. This horse had cervical spine osteoarthritis, and this gait may be attributed to cervical spine pathology; 28 however, a lumbosacral lesion such as occult tethered cord cannot be ruled out. Equine protozoal myeloencephalits due to S. neurona is unlikely as a western blot performed on CSF was negative; however, Neospora hughesi should still be considered. 2.5.2 Ultrasonographic Examination Bilateral MDBs were visualized ultrasonographically at both the AO and AA spaces. The MDBs, dura mater, dentate ligaments, subarachnoid space, and spinal cord were easily visible [Figure 2.1]. The 3 measurements taken in each position were used to calculate a mean for each horse. Mean HERDA and mean control measurements were then calculated for each position, along with the standard deviation [Table 2.2]. With the exception of AA extension, the mean width of the HERDA subarachnoid space was larger at every position than the mean width of the control horse subarachnoid space. The difference between the mean width of the subarachnoid space in flexion and in neutral position was calculated, and was greater at both the AO and AA spaces in HERDA horses [Table 2.3]. The difference between the mean width of the subarachnoid space in extension and neutral position was calculated, and was smaller at both the AO (p=0.21) and AA (p=0.08) levels in control horses, resulting in negative values [Table 2.4]. This suggests some degree of MDB dysfunction, and dural infolding, in the HERDA population during head extension. A larger sample size is needed in order to determine significance. 29 Figure 2.1: Ultrasonographic anatomy of the MDBs at the AA level. The spinal cord, dura mater, dentate ligaments, MDBs, and subarachnoid space are visible. The width of the subarachnoid space (|) was measured. AO Neutral AO Flexion AO AA Neutral AA Flexion AA Extension Extension HERDA 1.060 ±0.318 1.147 ±0.424 0.953 ±0.124 0.590 ±0.115 0.670 ±0.215 0.483 ±0.105 Control 0.740 ±0.125 0.747 ±0.168 0.760 ±0.166 0.527 ±0.038 0.633 ±0.114 0.563 ±0.072 Table 2.2: Width of the subarachnoid space in 3 HERDA versus 3 control horses with the head positioned in neutral, flexion, and extension at the AO and AA spaces. Reported as mean with standard deviation. HERDA Control AO Flexed-Neutral 0.087 ±0.146 0.007 ±0.050 AA Flexed-Neutral 0.080 ±0.101 -0.067 ±0.178 Table 2.3: Average difference between the width of the subarachnoid space in flexion and neutral in 3 HERDA horses versus 3 control horses at the AO and AA spaces. Reported as mean with standard deviation. 30 HERDA Control AO Extended-Neutral AA Extended-Neutral -0.107 ±0.206 0.020 ±0.101 -0.107 ±0.012 0.037 ±0.110 Table 2.4: Average difference between the width of the subarachnoid space in extension and neutral in 3 HERDA horses versus 3 controls at the AO and AA spaces. Reported as mean with standard deviation. At necropsy, the presence of latex dye within the AO and AA MDBs following ultrasound- guided injection confirmed the identity of the structures identified as the MDBs on ultrasound [Figure 2.2]. Figure 2.2: Atlanto-axial MDB (à) labelled with blue latex dye via ultrasound-guided injection. The AA MDBs spans from C1 (*) to C2 (*). 2.5.3 Anatomy of the Equine Myodural Bridge Myodural bridges were discovered in the horse at both the AO and AA levels. At the AO level, the MDBs originated from the rectus capitis dorsalis minor (RCDmi), a muscle analogous 31 to the RCPmi in humans [Figure 2.3]. As in humans, the MDBs attached via the posterior atlanto-occipital membrane-spinal dura complex, which should be termed the dorsal atlanto- occipital membrane-spinal dura complex in horses [1]. At the AA level, the MDBs originated from the rectus capitis dorsalis major (RCDma), a muscle analogous to the RCPma in humans. The obliquus capitis caudalis, analogous to the OCI in humans, is also thought to make a contribution to the MDB in the horse [25]. The equine nuchal ligament was also not observed to make a contribution to the MDB. No difference in MDB anatomy was observed between horses. Manipulation of the MDBs via traction on the RCDmi and RCDma was observed to tent the dura, providing evidence for the role of the MDBs in dural stabilization. Figure 2.3: The equine AO MDB with A=RCDmi, B=dense connective tissue MDB, and C=dura mater. 32 2.5.4 Myodural Bridge Histology The MDB is a heterogeneous structure composed of areas of dense connective tissue interspersed with loose connective tissue, nervous tissue, adipose tissue, and vasculature [Figure 2.4]. Synovium is present along the outer edges of the MDB in equine specimens, and likely facilitates the ability of the MDB to glide against C1 and the occiput. Past human histologic studies did not note the presence of synovium in the myodural bridge. Figure 2.4: Control horse MDB stained with Masson’s trichrome. A=RCDmi, B=connective tissue, C=dura mater (20x). Image courtesy of Dr. Edward Stopa. Prominent inflammatory changes were present in 1 HERDA horse MDB. This horse, a 3- year-old, displayed large aggregates of lymphocytes, as well as lymphocytic perivascular cuffing [Figure 2.5] [Figure 2.6]. The presence of both B and T lymphocytes was confirmed via CD3 and CD20 IHC [Figure 2.7]. A second HERDA horse, a 2-year-old, had no evidence of inflammation detected via IHC. The remaining 7 horses, a 2 HERDA horses and 5 controls, displayed a few scattered T lymphocytes in the MDB vessel walls, changes which were not felt to be indicative of inflammation. 33 Figure 2.5: Focal aggregate of lymphocytes and plasma cells in a 3-year-old HERDA horse gelding MDB stained with H&E (400x). Figure 2.6: Lymphocytic perivascular cuffing in a 3-year-old HERDA horse gelding MDB stained with Masson’s trichrome (200x). 34 Figure 2.7: Lymphocytic perivascular cuffing (à) in a 3-year-old HERDA horse gelding MDB. CD3 immunohistochemistry (200x). Myofiber degeneration was observed in the RCDmi in 3 HERDA horses and 1 control. A 21- year-old HERDA mare had myofiber degeneration with secondary macrophage cleanup as well as multiple sarcocysts within the muscle. Due to the presence of sarcocysts, CSF was submitted for a western blot to test for S. neurona antibodies; results were negative. The sarcocysts were suspected to be Sarcocystis fayeri, a commonly occurring sarcocyst found in the muscle of horses. No evidence of sarcocyst rupture was observed, and the myofiber degeneration in this horse was not suspected to be related to the presence of sarcocysts. Lack of periodic acid-Schiff staining ruled out polysaccharide storage myopathy as a cause of myofiber degeneration. A 6- year-old HERDA mare displayed rare degenerate myofibers with macrophage and eosinophil cleanup in the RCDmi. A single sarcocyst was observed within the RCDmi of this horse. A third HERDA horse, a 3-year-old gelding, was noted to have a focal area of myofiber degeneration with macrophage infiltration. The musculature was normal in all but one control, a 3-year-old Quarter Horse gelding that had a large accumulation of macrophages within the 35 RCDmi. Additionally, it was observed at necropsy that this horse had a grossly visible area of necrosis in the superficial cervical musculature. On histology, this unidentified muscle was found to have a focal area myofiber necrosis and macrophage cleanup. The damage was theorized to be secondary to trauma sustained during trailering. 2.5.5 Myodural Bridge Ultrastructure Differences in collagen fibril size, density, and orientation were compared between HERDA and control horses. Humans with EDS often display collagen fibrils described as whirling within bundles, as well as hook shaped or twisted fibrils. Collagen fibrils with irregular edges in cross- section, as well as loosely packed collagen fibrils, are also common findings in EDS [26]. Similar findings have been described in the skin of HERDA horses [17]. The presence of granulo-filamentous deposits in the interfibrillar space, which are associated with kEDS in humans, was also assessed [26]. Horses with HERDA were found to have more collagen abnormalities [Table 2.5]. In cross- section, 3 out of 4 HERDA horses were found to have large areas of loosely packed collagen fibrils. None of the control horses displayed these areas, rather collagen in the control horses was significantly more densely packed [Figure 2.8]. In cross-section, variation in fibril size and shape was found among HERDA and control horses of all ages, suggesting that this is a normal finding in the equine MDB. Abnormal collagen fibril orientation was more common in the HERDA horses, with 3 out of 4 HERDA horses displaying abnormal fibril orientation [Figure 2.9]. Granulo-filamentous deposits were observed in 2 HERDA horses and no controls. 36 Collagen Fibril Size and Shape in Cross-Section, Fibril Packing Density Abnormal Collagen Fibril Orientation Granulo-filamentous Deposits HERDA 1 HERDA 2 HERDA 3 HERDA 4 Control 1 Control 2 Control 3 Control 4 Control 5 Variation in collagen fibril size and shape, densely packed collagen fibrils Variation in collagen fibril size and shape, loosely packed collagen fibrils Normal collagen fibril size with mild variation in shape, loosely packed collagen fibrils Normal collagen fibril size with mild variation in shape. Areas of loosely packed collagen fibrils Normal collagen fibril size and shape. Densely packed with a few focal areas of loosely packed collagen fibrils Variation in collagen fibril size and shape, densely packed collagen fibrils Variation in collagen fibril size and shape. Densely packed with few focal areas of loosely packed collagen fibrils Variation in collagen fibril size and shape, densely packed collagen fibrils Variation in collagen fibril size and shape, densely packed collagen fibrils No Hook shaped collagen fibrils, broken collagen fibrils Whirling collagen fibrils Hook shaped collagen fibrils, broken collagen fibrils Few hook shaped and broken collagen fibrils No Hook shaped and collagen broken fibrils No No No Yes Yes No No No No No No Table 2.5: Findings from TEM imaging of the MDB of HERDA and control horses. Figure 2.8: Comparison of density of collagen fibril packing in cross-section in age-matched HERDA vs. control horses. A=HERDA horse with loosely packed collagen fibrils, variation in fibril size, variation in fibril shape, and granulo-filamentous deposits in the interfibrillar space (14,000x). B=Control horse with densely packed collagen fibrils and variation in fibril size (14,000x). 37 Figure 2.9: Abnormal collagen fibril orientation in the MDB of HERDA horses imaged with TEM. A= Whirling collagen fibrils in a 3-year-old gelding (10,000x). B= Hook shaped (à) and broken (à) collagen fibrils, as well as loosely packed collagen fibrils in a 6-year-old mare (27,000x). C= Granulo-filamentous deposits (à) in the interfibrillar space, as well as loosely packed collagen fibrils in a 3-year-old gelding (27,000x). D= Broken (à) and loosely packed collagen fibrils in a 2-year-old gelding (14,000x). 2.5.6 Myodural Bridge Tissue Biomechanics The average HERDA horse cross-sectional area of the MDBs (4.8 mm2 ± 0.7 mm2) was less than controls (6.8 mm2 ± 0.9 mm2) due to smaller MDB thickness. A t test performed on all values, rather than the averaged values, for strain at complete pull-apart demonstrated 38 significantly (p=0.04) more strain at complete pull-apart for HERDA horses (1.72 ± 0.80) versus control horses (1.19 ± 0.29). Control horse MDBs pulled apart at 119% of their original length, while HERDA horse MDBs pulled apart at 172% of their original length. This data is indicative of more ductile MDBs in the HERDA population as a result of their PPIB mutation [27]. Average strain at maximum stress, maximum stress, and tensile modulus were less in HERDA horses; however, additional data points are needed in order to determine significance. 2.6 Conclusion The results of this study support that MDBs are a conserved anatomic structure across mammalian species [2]. Horses with HERDA, a known animal model of EDS in humans, were found to have histologic and ultrastructural abnormalities in the MDBs. Additionally, preliminary ultrasonographic data showed evidence of dural infolding in the HERDA population. Preliminary tissue biomechanical testing showing HERDA horses had a smaller average MDB cross-sectional area, average strain at maximum stress, maximum stress, and tensile modulus. Additionally, HERDA horses displayed more strain at complete pull-apart than controls. This data supports the theory that MDB dysfunction exists in the EDS population. Future ultrasonographic studies are now encouraged to examine the human MDBs in the EDS population. 39 REFERENCES 40 REFERENCES 1. Hack, G.D., Koritzer, R.T., Robinson, W.L., Hallgren, R.C., Greenman, P.E. (1995) Anatomic relation between the rectus capitis posterior minor muscle and the dura mater. Spine. 20: 2484–2486. 2. Zheng N., et al. (2017) The universal existence of myodural bridge in mammals: an indication of a necessary function. Sci Rep. 6: 8248. 3. Hallgren R.C., Hack G.D., Lipton J.A. (1997) Clinical implications of a cervical myodural bridge. AAO J. 7: 30–34. 4. Xu Q., et al. (2016) Head movement, an important contributor to human cerebrospinal fluid circulation. Sci Rep. 6: 31787. 5. Chopra P., et al. (2017) Pain management in Ehlers-Danlos syndomes. Am J Med Genet C Semin Med Genet. 175(1): 212-219. 6. Hack G.D., Hallgren R.C. (2004) Chronic headache relief after section of suboccipital muscle dural connections: a case report. Headache. 44(1): 84-89. 7. Alix M.E., Bates D.K. (1999) A proposed etiology of cervicogenic headache: the neurophysiologic basis and anatomic relationship between the dura mater and the rectus posterior capitis minor muscle. J Manipulative Physiol Ther. 22(8): 534-539. 8. Schaller B., Baumann A. (2003) Headache after removal of vestibular schwannoma via the retrosigmoid approach: a long-term follow-up-study. Otolaryngol Head Neck Surg. 128(3): 387-395. 9. Schessel D.A., Rowed D.W., Nedzelski J.M., Feghali J.G. (1993) Postoperative pain following excision of acoustic neuroma by the suboccipital approach: observations on possible cause and potential amelioration. Am J Otol. 14(5): 491-494. 10. Soumekh B., Levine S.C., Haines S.J., Wulf J.A. (1996) Retrospective study of postcraniotomy headaches in suboccipital approach: diagnosis and management. Am J Otol. 17(4): 617-619. 11. McPartland J.M., Brodeur R.R., Hallgren R.C. (1997) Chronic neck pain, standing balance, and suboccipital muscle atrophy—a pilot study. J Manipulative Physiol Ther. 20(1): 24-29. 12. Castori M., et al. (2015) Connective tissue, Ehlers-Danlos syndrome(s), and head and cervical pain. Am J Med Genet C Semin Med Genet.169C: 84-96. 13. Yeowell H.N., Steinmann B. Ehlers-Danlos Syndrome, Kyphoscoliotic Form. 2000 Feb 2 [Updated 2013 Jan 24]. In: Adam M.P., Ardinger H.H., Pagon R.A., et al., editors. 41 GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993- 2018. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1462/ 14. Rashmir-Raven A., Spier S.J. (2015) Hereditary equine regional dermal asthenia (HERDA) in quarter horses: a review of clinical signs, genetics and research. Equine Vet Educ. 27(11): 604- 611. 15. Tryon, R.C., White, S.D., Bannasch, D.L. (2007) Homozygosity mapping approach identifies a missense mutation in equine cyclophilin B (PPIB) associated with HERDA in American Quarter Horses. Genomics. 90: 93-102. 16. Lerner D.J., McCracken M.D. (1978) Hyperelastosis cutis information from vetstream equis. J Eq Med Surg. 2: 350–352. 17. White, S.D., et al. (2004) Hereditary equine regional dermal asthenia (‘hyperelastosis cutis’) in 50 horses: clinical, histologic, immunohistologic and ultrastructural findings. Vet Dermatol. 15: 207-217. 18. Brinkman E.L., et al. (2017) Cardiac findings in quarter horses with heritable equine regional dermal asthenia. JAVMA. 250(5): 538-547. 19. Mochal, C.A., et al. (2010) Ocular findings in Quarter Horses with hereditary equine regional dermal asthenia. J Am Vet Med Ass. 237: 304-310. 20. Gutiérrez-Crespo B., Kircher P.R., Carrera I. (2014) 3 tesla magnetic resonance imaging of the occipitoatlantoaxial region in the normal horse. Vet Radiol Ultrasound. 55(3): 278-285. 21. Audigié F., Tapprest J., Didierlaurent D., Denoix J.-M. (2004) Ultrasound-guided atlanto- occipital puncture for myelography in the horse. Vet Radiol Ultrasound. 45(4): 340-344. 22. McCluskie L.K., Franklin S.H., Lane J.G., Tremaine W.H., Allen K.J. (2008) Effect of head position on radiographic assessment of laryngeal tie-forward procedure in horses. Vet Surg. 37(7): 608-612. 23. Withrow, A., personal communication, July 18, 2018. 24. Lavagnino M., Gardner K., Arnoczky S.P. (2013) Age-related changes in the cellular, mechanical, and contractile properties of rat tail tendons. Connect Tissue Res. 54(1): 70-75. 25. Budras K.D., Sack W.O., Rock S. (2011) Anatomy of the horse: an illustrated text (6th ed.). Hannover, Germany: Schlutersche. 26. Hermanns-Lê T., Reginster M., Piérard-Franchimont C., Piérard G.E. (2013) Ehlers-Danlos syndrome. In Stirling J.W., Curry A., Eyden B. (Eds.), Diagnostic electron microscopy: a practical guide to interpretation and technique (pp. 309-320). Chichester, England: John Wiley & Sons Ltd. 42 27. Lavagnino, M., personal communication, August 31, 2018. 43 CHAPTER 3: EVALUATION OF THE FILUM TERMINALE IN AN EQUINE MODEL OF EHLERS-DANLOS SYNDROMES 3.1 Abstract The objective of this portion of the study was to determine the relevance of horses with hereditary equine regional dermal asthenia (HERDA) as a model of occult tethered cord syndrome in Ehlers-Danlos syndromes. Neurologic examinations were performed on 2 HERDA and 4 control horses. Histologic and ultrastructural examinations of the fila terminale of 4 HERDA and 5 control horses were performed following sample collection at necropsy. CD3 and CD20 immunohistochemistry was performed on 2 HERDA and 5 control horses. Adenovirus PCR was performed on the filum terminale of 1 HERDA horse. Neurologic examination was abnormal in 1 HERDA horse, that displayed truncal sway and a base-narrow, single-track gait. The overall structure of the fila terminale of HERDA horses was disorganized. Severe lymphocytic inflammation was observed in 1 HERDA horse, while moderate lymphocytic inflammation was observed in 1 HERDA horse and 2 controls. Abnormalities in collagen fibril orientation were observed in 4 HERDA horses, while mild abnormalities were observed in 2 controls. Variations in fibril size and shape were seen in 3 HERDA horses, while 3 control horses had mild variation in fibril shape only. Results were compared to human specimens collected at the time of occult tethered cord release surgery at Rhode Island Hospital, Department of Neurosurgery. Although the filum terminale is abnormal in HERDA horses, further studies are needed to determine if occult tethered cord syndrome exists in the HERDA population. 44 3.2 Introduction The filum terminale (FT) is a dense, fibrous band of connective tissue that spans from the conus medullaris at the level of L1 or L2 to the level of the coccyx. The FT is formed from embryologic spinal cord which has undergone retrogressive differentiation. In humans, the FT has an intradural section, the filum terminale internum, and an extradural section, the filum terminale externum [1-3]. Filum terminale dysfunction may result in tethered cord syndrome (TCS) or occult tethered cord (OTCS) syndrome, which leads to the same symptoms as TCS but does not have radiographic evidence of tethering. These conditions occur due to mechanical tethering or abnormally inelastic tissue within the FT, leading to tension on the spinal cord [4-5]. Symptoms of TCS/OTCS may include neurologic, urologic, gastrointestinal, orthopedic, or dermatologic dysfunction. Occult tethered cord syndrome commonly occurs in patients with Ehlers-Danlos syndromes (EDS), a group of genetic connective tissue disorders [6]. It is theorized that OTCS may occur in the EDS population due to abnormal elements of mesoderm left during embryologic development, leading to an abnormally inelastic FT [7]. The incidence of OTCS in the EDS population is unknown, but is thought to be much greater than in the general population. While the incidence of OTCS in the general population is unknown, the rate of TCS in the general population is 0.1% [8]. Both histopathologic and ultrastructural abnormalities have been noted in the FT of human patients with both EDS and OTCS. Common findings in EDS OTCS FT include increased vasculature as well as the presence of marginating polymorphonuclear cells in the vessels. Other common findings include inflammatory cells, excessive numbers of elastic fibers, and prominent ependyma. Unlike patients with classical TCS, patients with EDS and OTCS often have little adipose tissue in the FT [9-11]. On TEM imaging, the collagen of the EDS OTCS FT is 45 disorganized and displays corkscrewing or beading; however, these changes are also observed in the non-EDS OTCS population and may be related to tethering rather than EDS [10]. These changes are not observed in humans without connective tissue disease or OTCS/TCS, though it has been noted that the collagen fibrils in the filum terminale are not always tightly packed in normal individuals [12]. The anatomy of the FT has been described in horses; however, histopathologic and ultrastructural studies have not previously been conducted. Tethered cord syndrome has been described in humans, dogs, and cats; however, it has not been described in horses [13-14]. In this study, horses with hereditary equine regional dermal asthenia (HERDA) were used as a naturally occurring model of EDS. Hereditary equine regional dermal asthenia is caused by a mutation in peptidyl-prolyl cis-trans isomerase B (PPIB), encoding for cyclophilin B, and occurs in Quarter Horses [15]. 3.3 Materials and Methods 3.3.1 Animals A total of 9 horses (4 HERDA and 5 control) were evaluated [Table 3.1]. All horses were donated to the study by their owners. The control group consisted 3 Quarter Horses, 1 Arabian, and 1 Morgan. Controls ranged in age from 2 to 18 years and included 4 mares and 1 gelding. The 4 HERDA horses were all Quarter Horses, and ranged in age from 2 to 21 years old. The affected group included 2 mares and 2 geldings. All Quarter Horses were confirmed homozygous positive, H/H, for the PPIB mutation. Testing on 2 HERDA horses was performed at the UC Davis Veterinary Genetics Laboratory, while the other 2 HERDA horses were tested at Michigan State University by the master’s candidate (AM). All HERDA horses displayed a phenotype consistent with the disorder. All 3 control Quarter Horses were confirmed to be 46 homozygous negative, N/N, for the PPIB mutation via PCR testing at Michigan State University by the master’s candidate (AM), and were phenotypically normal. Carriers, H/N, for the PPIB mutation were not evaluated in this study. Experimental protocols were approved by the Michigan State University institutional animal care and use committee. HORSE ID BREED HERDA 1 Quarter Horse HERDA 2 Quarter Horse SEX Mare Mare HERDA 3 Quarter Horse Gelding HERDA 4 Quarter Horse Gelding Control 1 Quarter Horse Control 2 Morgan Mare Mare Control 3 Quarter Horse Gelding Control 4 Arabian Control 5 Quarter Horse Mare Mare AGE 21 yrs 6 yrs 3 yrs 2 yrs 18 yrs 6 yrs 3 yrs 2 yrs 2 yrs Study Role Neurologic exam, necropsy, histopathology, TEM Neurologic exam, necropsy, histopathology, TEM Necropsy, histopathology, TEM Necropsy, histopathology, TEM Neurologic exam, necropsy, histopathology, TEM Necropsy, histopathology, TEM Neurologic exam, necropsy, histopathology, TEM Neurologic exam, necropsy, histopathology, TEM Neurologic exam, necropsy, histopathology, TEM Table 3.1: Identities of HERDA and control horses used in FT study. 3.3.2 Neurologic Examination In order to assess for deficits consistent with TCS or OTCS, static and dynamic neurologic examinations were performed on 2 HERDA horses and 4 control horses. Two HERDA horses were not examined due to the early point at which they were enrolled in the study, which was prior to the current iteration of the study design. One control horse was unable to be examined due to a severe P2 fracture. All examinations were performed by two veterinarians, a boarded large animal internist (EC) and the master’s candidate (AM). 47 3.3.3 Necropsy The FT were collected at necropsy from 4 HERDA horses and 5 control horses. Samples were collected within 1 hour of euthanasia in all but 1 control horse. The anatomy of the equine FT was assessed by the master’s candidate (AM), a neurosurgeon (PK), and a boarded veterinary pathologist (DS). An approximately 15 cm section of FT was removed from the termination of the conus medullaris until the point at which it became confluent with the cauda equina. 3.3.4 Filum Terminale Histology Samples were fixed in 10% neutral buffered formalin for at least 24 hours prior to sectioning. An approximately 1.5 cm sample was sectioned from 3 cm caudal to the termination of the conus medullaris. All specimens were stained with hematoxylin and eosin (H&E), as well as Masson’s trichrome. CD3 and CD20 immunohistochemistry (IHC) was performed on 2 HERDA and 5 control specimens. Sample processing was conducted at both Rhode Island Hospital and the Michigan State University Veterinary Diagnostic Laboratory. 3.3.5 Filum Terminale Ultrastructure The FT from 4 HERDA and 5 control horses were submitted for TEM. An approximately 1.5 cm sample was submitted from 4.5 cm distal to the termination of the conus medullaris from each specimen. Samples were fixed in 10% neutral buffered formalin for at 24 hours prior to transfer to glutaraldehyde for TEM processing. Samples were processed according to standard TEM procedures. Sample processing was conducted at both Rhode Island Hospital and the Michigan State University Center for Advanced Microscopy. A Power Tome Ultramicrotome (RMC, Boeckeler Instruments. Tucson, AZ) was used to cut 70 nm sections. The FT were imaged at 14,000x, 27,000x, and 40,000x in longitudinal and cross-section using a JEOL 100CX 48 Transmission Electron Microscope (Japan Electron Optics Laboratory, Japan) at an accelerating voltage of 100kV [16]. Imaging at higher or lower magnification was performed when indicated. 3.3.6 Adenovirus PCR Adenovirus PCR was performed after a viral particle with a structure consistent with adenovirus was observed on TEM imaging from a 21-year-old HERDA mare. Paraffin block shavings of the FT were submitted for PCR. Sample processing and testing was conducted at the Michigan State University Veterinary Diagnostics Laboratory. 3.4 Results 3.4.1 Neurologic Examination Mild neurologic deficits were observed in 1 HERDA horse, a 21-year-old Quarter Horse mare, during the dynamic phase of the neurologic examination. Mild truncal sway during blindfolding and head elevation was observed. A base-narrow single-track gait in the hind limbs, described as “tight-rope walking”, during blindfolding and head elevation was also noted. These findings may be secondary to this mare’s cervical spine osteoarthritis, which was visible on ultrasound. While cervical spine pathology is the most likely cause of these symptoms, other causes cannot be ruled out. A lumbosacral lesion such as TCS/OTCS could account for these symptoms. Urinary incontinence was not observed; however, urodynamic testing was not performed. Equine protozoal myeloencephalitis must also be considered. A western blot performed on CSF was negative for antibodies to Sarcocystis neurona; however, Neospora hughesi was not ruled out. All control horse examinations were within normal limits. 49 3.4.2 Necropsy The equine FT has previously been described as running from S1 to S4 [17]. Our necropsy examination confirmed these findings [Figure 3.1]. The equine FT was also confirmed to have an intradural and extradural section, similar to those found in the human FT. Figure 3.1: The equine FT. The conus medullaris (à) and FT (à) are visible. The conus medullaris terminates at the level of S1 in the horse. 3.4.3 Filum Terminale Histology The overall structure of the FT of all HERDA horses was disorganized compared to controls [Figure 3.2]. Additionally, inflammation was observed in HERDA horses. CD3 and CD20 IHC was performed on 2 HERDA horses and 5 controls. Subjectively, one HERDA horse, a 6-year- old mare, had severe lymphocytic inflammation with lymphocytic perivascular cuffing [Figure 3.3]. A second HERDA horse, a 21-year-old mare, had moderate lymphocytic inflammation. Two controls, a 2-year-old mare and an 18-year-old mare, had moderate lymphocytic inflammation. Three control horses had few scattered lymphocytes, classified as mild lymphocytic inflammation. Moderate inflammatory changes in the 2 control horses may be secondary to undiagnosed neurologic disease, or may be secondary to age-related changes in the FT in the 18-year-old. Few studies have utilized IHC in the FT, and little is known about the presence of lymphocytes in the normal FT. 50 Figure 3.2: The overall structure of the HERDA FT is disorganized. A= FT from 21-year-old affected Quarter Horse mare displaying disorganized remnants of the central canal (à) (H&E) (40x). B= FT from 21-year-old affected Quarter Horse mare with disorganized elements of fibrous tissue (à) (Masson’s trichrome) (40x). C=FT from a 4-year-old control Quarter Horse mare with a normal central canal (à) (H&E) (40x). D=FT from a 4-year-old control Quarter Horse mare with normal fibrous tissue (à) (Masson’s trichrome) (40x). Figure 3.3: Lymphocytic inflammation is present in the HERDA FT. A= FT from 6-year-old affected Quarter Horse mare. Large number of CD3 positive lymphocytes are present. Lymphocytic perivascular cuffing is present (100x). B= FT from 6-year-old control Morgan mare. Few scattered CD3 positive lymphocytes are present (100x). 51 Results from equine FT histology were compared to human FT histology results from Rhode Island Hospital, Department of Neurosurgery from 66 patients with EDS OTCS and 24 patients non-EDS OTCS patients. Specimens were collected at the time of tethered cord release surgery. Based on H&E staining, no non-EDS OTCS patients had inflammation, while 1.51% of EDS OTCS FT had inflammatory cells. Heavy mast cell infiltration was observed in EDS OTCS FT, while non-EDS OTCS FT had few scattered mast cells [9]. In humans, EDS OTCS FT had a larger amount of vasculature, nerve twigs, and ependyma compared to non-EDS TCS FT, while EDS OTCS FT had less adipose and arachnoid tissue than non-EDS OTCS FT [Figure 3.4] [10]. No difference in the amount of vasculature, ependyma, adipose or arachnoid tissue was observed in HERDA horses. Neuropil was observed in 2 HERDA horses and no controls, while the number of nerve twigs was not appreciably different in the HERDA population. Melanocytes were observed in both HERDA and control horses, and have been observed in the human population. This is likely an incidental finding. Several findings were observed in the equine FT that were not observed in the human population. Rosenthal fibers, which are indicative of chronic gliosis, were observed in 1 HERDA horse and 1 control. Mild mineralization of collagen within the FT was observed in 3 HERDA horses and 2 controls. Marginating eosinophils were observed in 1 HERDA horse and 1 control. 52 Figure 3.4: Common histologic changes in the EDS OTCS FT. A= Prominent vascularity (à) (40x). B= Marginating polymorphonuclear cells within a vessel (à) (200x). C= Nerve twigs (à) (100x). D= Neuropil (*) (100x). Photos courtesy of Dr. Petra Klinge and Dr. Edward Stopa. 3.4.4 Filum Terminale Ultrastructure The collagen of the FT of HERDA horses displayed more abnormalities in collagen fibril size, shape, packing density, and orientation than control horses [Table 3.2]. Collagen fibrils were more loosely packed in HERDA horses than controls, a finding consistent with other HERDA tissues [18-19]. Variation in collagen fibril size and shape in cross-section was observed in 3 of the 4 HERDA horses [Figure 3.5]. All control horses had normal collagen fibril size; however, mild variation in fibril shape was observed in 3 of the 5 control horses. While a few minor abnormalities in collagen fibril orientation were observed in 2 control horses, all HERDA horses had moderate to severe abnormalities in collagen fibril orientation. These 53 abnormalities included whirling collagen fibrils, broken collagen fibrils, disintegrating collagen fibrils, and hook shaped collagen fibrils [Figure 3.6]. These abnormalities in collagen fibril orientation are consistent with changes observed in the skin of human EDS patients [20]. Additionally, a viral particle consistent with adenovirus was observed in 1 HERDA horse, a 21- year-old mare [Figure 3.6]. Collagen Fibril Size and Shape in Cross-Section, Fibril Packing Abnormal Collagen Fibril Orientation Whirling collagen fibrils, broken and disintegrating collagen fibrils, hook shaped collagen fibrils Broken and disintegrating collagen Hook shaped collagen fibrils, broken collagen fibrils fibrils Hook shaped fibrils, broken and disintegrating fibrils Few broken fibrils No Few hook shaped and broken collagen fibrils No No Other Findings Viral inclusion No No No No No No No No Density HERDA 1 Moderate variation in collagen fibril size and shape, areas of loosely packed collagen fibrils HERDA 2 Normal collagen fibril size and shape, areas of loosely collagen packed fibrils HERDA 3 Mild variation in collagen fibril size and shape, areas of loosely packed collagen fibrils HERDA 4 Mild variation in collagen fibril size and shape. Areas of loosely packed collagen fibrils Control 1 Control 2 Control 3 Control 4 Control 5 Normal collagen fibril size with mild variation in shape, few areas of loosely packed fibrils Normal collagen fibril size with mild variation in shape. Densely packed Normal collagen fibril size with mild variation in shape. Areas of loosely packed collagen fibrils Normal collagen fibril size and shape, few areas of loosely packed collagen fibrils Normal collagen fibril size and shape, densely packed Table 3.2: Comparison of TEM findings from the FT of HERDA and control horses. 54 Figure 3.5: Collagen fibril size, shape, and packing density in the FT of an affected horse versus a control (27,000x). A= Cross-sectional imaging of a 21-year-old affected Quarter Horse mare. Collagen fibrils are variable in size and shape and are loosely packed. B= Cross-sectional imaging of an 18-year-old control Quarter Horse mare. Collagen fibrils are normal in size with mild variation in shape. Packing density is moderate. C= Longitudinal imaging of a 21-year-old affected Quarter Horse mare. Collagen fibrils display a random, haphazard orientation and are broken. D= Longitudinal imaging of an 18-year-old control Quarter Horse mare. Fibrils are densely packed and display regular orientation. 55 Figure 3.6: Abnormal TEM findings in a 21-year-old affected Quarter Horse mare. A= Hook shaped (à) and broken (à) collagen fibrils (14,000x). B= Disintegrating collagen fibrils (à) (14,000x). C= Whirling collagen fibrils and elastic fibers (*) (14,000x). D= Viral inclusion (à) (80,000x). Results from equine FT TEM imaging were compared to human FT TEM imaging conducted at Rhode Island Hospital, Department of Neurosurgery on a segment of FT removed at the time of surgery from 66 EDS OTCS patients and 15 non-EDS OTCS patients. Disorganization as 56 well as beading and corkscrewing of the collagen fibrils are common findings in the EDS OTCS FT (57.57%) [Figure 3.7]. Variability in fibril size is occasionally observed in the EDS OTCS population. Focal areas of abnormal collagen, including beading and corkscrewing, are also observed in the non-EDS OTCS filum (46.67%) [Figure 3.8]. This suggests that tethering itself can lead to ultrastructural abnormalities, though the possibility that some patients in the non-EDS OTCS population have an undiagnosed connective tissue disorder cannot be ruled out. Alternatively, ultrastructural abnormalities of the FT may lead to eventual tethering due to an inelastic FT. Patients with normal collagen in the FT are found within both the EDS OTCS and non-EDS OTCS population. Subtype of EDS or degree of tethering could account for this [10, 21]. Beading and corkscrewing was not observed in the equine FT. This may be due to the nature of the PPIB mutation or because tethering was not present. Figure 3.7: TEM findings in the FT of a 45-year-old female with hypermobile Ehlers-Danlos syndrome. Collagen fibrils are disorganized and swollen. Beading (à) and corkscrewing (à) of the collagen is present (1 µm scale bar). Photo courtesy of Dr. John Donahue. 57 Figure 3.8: TEM findings in the FT of non-EDS OTCS patients. A= Focally abnormal ultrastructure in the FT of a 17-year-old male non-EDS OTCS patient (1 µm scale bar). Collagen fibrils are disorganized and mildly swollen. Areas of beading (à) and corkscrewing (à) are present. B= Normal ultrastructure in a 22-year-old male non-EDS OTCS patient (1 µm scale 3.4.5 Adenovirus PCR bar). Photos courtesy of Dr. John Donahue. The FT of a 21-year-old HERDA mare was submitted for adenovirus PCR testing after a viral particle consistent with adenovirus was observed on TEM [Figure 3.4]. There are 2 case reports of isolation of adenovirus-1 from the spinal cord and lumbar dorsal root ganglia of horses with polyneuritis equi [22]. Adenovirus-1 is not a proven cause of neurologic disease in horses, and its association with polyneuritis equi is unproven at this time. The results of this test were negative, a result thought to be due to a low viral load. 3.5 Conclusion The gross anatomy of the equine and human FT displays many similarities. Histologically, HERDA horse FT are abnormal and display disorganization as well as lymphocytic inflammation and neural elements. Ultrastructural abnormalities of the collagen are present in the HERDA population; however, the corkscrewing and beading of the collagen fibrils observed in the human OTCS population was not observed in horses. Further studies, such as urodynamic 58 testing, are needed to determine if true tethering is present in the HERDA population; however, the results of this study suggest that the HERDA horse is an appropriate animal model for the study of OTCS in EDS. 59 REFERENCES 60 REFERENCES 1. Saker E., et al. (2017) The filum terminale internum and externum: a comprehensive review. J Clin Neurosci. 40: 6-13. 2. Tubbs R.S., et al. (2005) The filum terminale externum. J Neurosurg Spine. 3(2): 149-152. 3. De Vloo P., Monea A.G., Sciot R., van Loon J., Van Calenbergh F. (2012) The filum terminale: a cadaver study of anatomy, histology, and elastic properties. World Neurosurg. 90: 565-573. 4. Yamada S., et al. (2007) Pathophysiology of tethered cord syndrome and similar complex disorders. Neurosurg Focus. 23(2): E6. 5. Khoshhal K.I., Murshid W.R., Elgamal E.A., Salih M.A.M. (2012) Tethered cord syndrome: a study of 35 patients. J Taibah Univ Sci. 7(1): 23-28. 6. Henderson F.C., et al. (2017) Neurologic and spinal manifestations of the Ehlers-Danlos syndromes. Am J Med Genet C Semin Med Genet. 175(1): 195-211. 7. Klinge P., personal communication, January 28, 2018. 8. Bademci G., et al. (2006) Prevalence of primary tethered cord syndrome associated with occult spinal dysraphism in primary school children in Turkey. Pediatr Neurosurg. 42(1): 4- 13. 9. Klinge P. (2015) Histopathological evidence of clinically relevant filum pathology in tethered cord syndromes with little radiographic evidence including hypermobility syndromes. Presented at the Chiari & Syringomyelia Foundation Research Colloquium, New Orleans, LA. 10. Astafiev S., Donahue J., Stopa E., Klinge P. (2017) Occult tethered cord syndrome: a histopathological characterization of the filum terminale in Ehlers-Danlos and non-Ehlers- Danlos patients. The Warren Alpert Medical School of Brown University Sumer Research Showcase, Providence, RI. 11. Bouffard J., Rushing E.J. (2005) The neuropathology of “tethered cord”. Semin Spine Surg. 17(1): 19-22. 12. Ricardo B.V., et al. (2006) Ultrastructural study of the filum terminale and its elastic fibers. Neurosurgery. 58(5): 978-984. 13. de Decker S. Gregori T., Kenny P.J., Erles K., Volk H.A. (2015) Tethered cord syndrome associated with a thickened filum terminale in a dog. J Vet Intern Med. 29(1): 405-409. 61 14. Tamura M., et al. (2017) Tethered cord syndrome with spina bifida aperta in cats: two case reports of different types. JFMS Open Rep. 3(1): 2055116917708060. 15. Rashmir-Raven A., Spier S.J. (2015) Hereditary equine regional dermal asthenia (HERDA) in quarter horses: a review of clinical signs, genetics and research. Equine Vet Educ. 27(11): 604- 611. 16. Withrow, A., personal communication, July 18, 2018. 17. Budras K.D., Sack W.O., Rock S. (2011) Anatomy of the horse: an illustrated text (6th ed.). Hannover, Germany: Schlutersche. 18. White, S.D., et al. (2004) Hereditary equine regional dermal asthenia (‘hyperelastosis cutis’) in 50 horses: clinical, histologic, immunohistologic and ultrastructural findings. Vet Dermatol. 15: 207-217. 19. Mochal, C.A., et al. (2010) Ocular findings in Quarter Horses with hereditary equine regional dermal asthenia. J Am Vet Med Ass. 237: 304-310. 20. Hermanns-Lê T., Reginster M., Piérard-Franchimont C., Piérard G.E. (2013) Ehlers-Danlos syndrome. In Stirling J.W., Curry A., Eyden B. (Eds.), Diagnostic electron microscopy: a practical guide to interpretation and technique (pp. 309-320). Chichester, England: John Wiley & Sons Ltd. 21. Donahue, J., personal communication, July 27, 2018. 22. Edington N., Wright J.A., Patel J.R., Edwards G.B., Griffiths L. (1984) Equine adenovirus 1 isolated from cauda equina neuritis. Res Vet Sci. 37(2): 252-254. 62