Cryopreserved equine umbilical cord tissue allograft characterization and biocompatibility in vivo in musculoskeletal tissues: a controlled study

Abstract

Background

The use of micro-particulate allografts is rising, but knowledge about the protein characterization and biocompatibility of umbilical cord-derived allografts (UC) in vivo is limited.

Methods

Proteomic analyses using mass spectrometry (MS) determined equine UC protein relative quantification and functions using total spectral counts (TSC). UC cytokines were quantified by enzyme-linked immunosorbent assay (ELISA). Three in vivo studies assessed recipient clinical and tissue biocompatibility in joints and ligaments.

Results

Proteomics revealed 2645 annotated TSCs. Proteins of > 89 TSC were considered abundant and were present in all donors. Proteins within the same donor had a 4.7% mean variation. Inflammatory cytokines were low in UC. In vivo, the prospective randomized, masked, controlled study in carpal joints and ligaments of clinically normal horses had median scores of 0 (none) for lameness and pain for 42 days. Synovial fluid showed a transient transudative synovitis after UC injection that was greater than baseline and control and returned to normal after day 5 (P < 0.001). Synovial fluid inflammatory cytokines were low; however, the anti-inflammatory cytokines Il-1ra, Il-10, and Il-1ra/Il-1 ratio were greater after UC injection than at baseline and control (P < 0.001). Blood hematology, chemistries, and serum amyloid A did not reveal systemic effects. The in vivo study of osteoarthritis and desmitis/tendonitis improved in lameness and pain over a 28-day study and had parallel synovial fluid results to the normal horse study, also without adverse events. The in vivo pathologic study evaluated joint and ligament tissues 2 and 5 days after injection and corresponding lymph nodes for evidence of the allograft or inflammation. The synovial membrane, articular cartilage, and lymph nodes were histologically normal, except for mild inflammation in the injection tracts.

Conclusions

Well-defined proteins were consistently present in different donors and within batches. Proteins included fibrillar and glycan proteins with a variety of roles and regulatory functions in the connective tissue matrix. The rise in Il-1ra and high Il-1ra/Il-1 ratio after UC injection could block the catabolic effect of Il-1. No adverse events were observed. Within the limits of this study, UC was safe for injection into joints and ligaments in clinically normal horses.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12916-025-04231-7.

Background

Musculoskeletal focus

Musculoskeletal (MSK) abnormalities limit mobility and physical capacity in both humans and animals [1, 2]. An abundance of recently published work focuses on the MSK systems of joints, tendons, and ligaments [1]. The rapid advancements of care in this space are, in part, due to the increasing health and mobility span of humans and animals [3], as well as the high expectations for the care of human and equine athletes as a social obligation in connection with sports entertainment [4, 5]. Animal investigations have long been integral to the advancement in the care of MSK tissues [2, 6].

The equine species is particularly recognized for a high frequency of MSK injury [2, 7]. Horses are a validated model for MSK health in humans [1, 2], both in basic science and in clinical applications, due to the early adoption of biological products as well as the substantial overlap in physical activities and medical care [2, 6]. Despite the rapid increase in the use of biologic tissue products for MSK tissues in both human and equine athletes over the past decade, products that offer safety, quality, biocompatibility, convenience, and consistent characterization are still in the early stages of development [8].

Biologic tissue transplants

The study of biologic tissue transplants is extensive and has resulted in the clinical application of many products in humans [9, 10] and horses [11, 12], including various forms and enhancements of plasma, platelet-rich plasma, and bone marrow-derived plasma solutions, primarily for autologous injectable application in humans [13–15] and horses [12, 13, 16]. Autologous flowable products, while unlikely to produce reactions upon reapplication to the donor, have substantial limitations, such as inconvenient and time-consuming processing, donor site morbidity, variable product content [9, 12, 14], and often donor comorbidities that reduce quality and effectiveness of product, such as older age and osteoarthritis (OA) [1, 3, 13].

Tissue transplants, such as grafts of articular cartilage, tendons, and ligaments, have been used extensively for surgical repair of defects in these tissues [10, 17, 18]. The grafts have routinely included both autologous and allogeneic tissue replacement and augmentation. Autologous whole tissue grafts are limited by donor site morbidity with potential dysfunction at the harvest site, a concern particularly in athletes and young patients [10, 17, 18]. Fresh allogeneic grafts offer the advantages of living tissue and minimal degradation or alteration of the proteins, such as may occur with lyophilization, but carry the proven risk of disease transmission [18]. Intact frozen allografts are subject to the effects of freezing tissue in bulk [19, 20], in part due to limited penetration of cryoprotectants [19]. Some great successes have been achieved with autologous and allogeneic grafts, including a long-standing bio-scaffold still in use today, bone grafts [17]. While bone grafts have been quite successful, grafted articular cartilage or articular cartilage constructs have met with greater challenges, as have tendon and ligament defects not amenable to an intact tissue graft [21, 22]. Contributing factors include the slow natural repair of these tissues [14, 21] and the dense matrix required for the physical capacity to return to expected levels of function [23, 24]. Consequently, allograft supplements for articular cartilage, tendon, and ligament have lagged in advancement to credible application. Many applications, however, do not require surgery or complete intact grafts [9]; novel alternatives are needed to expand graft utility and facilitate application in multiple sites.

Microparticulate bio-scaffolds

Innovative processing techniques combined with imaging modalities have created particulate grafts to serve as supplementation to tissue defects and visualize the flow of the graft into anatomic tissue gaps by video endoscopy or ultrasound [2, 8]. The creation of particulate biologic tissue grafts by physical processing in both humans [9, 25, 26] and horses [27] has resulted in the refinement of nonchemical means to create a minimally manipulated biologic milieu while retaining connective tissue structure and functional proteins. Fresh particulate biological tissue grafts can be suspended in a biocompatible cryopreservation solution and cryopreserved for years.

Tissue sources for musculoskeletal supplementation

More recently, science and industry have heavily investigated, and invested in, the development of flowable particulate biologic scaffolds as grafts to supplement tissues in joints, tendons, and ligaments. These biologic scaffolds offer the potential advantages of not being nondegradable or a foreign substance, such as polyacrylamides [28–30] or xenografts [31], providing the ability to select a tissue source that aligns with the structural support and biocompatibility of the intended tissue target [8]. The most investigated tissue sources for microparticulate allografts are birth tissues, including maternal placental tissue (chorioallantois), fetal tissue (amniotic membrane and fluid), and umbilical cord (UC) [9, 32, 33]. In humans, chorioallantois/amniotic membranes and UC have been processed and applied in combination (amnion-umbilical cord), consistent with human placental anatomy [9, 32, 33]. Placentation is diffuse in equine species, and the fetal amniotic membrane and UC can be isolated separately and collected independently from the maternal placenta. Lyophilized equine amniotic membrane and amniotic fluid have been processed for use in equine joints but are no longer commercialized in that form [34]. The amniotic membrane has evolved as a relatively thin, tough membrane designed to resist shear and tension due to amniotic fluid distention and pressure produced by fetal movements in utero. In contrast, UC tissue is designed inherently to provide structural support and cushion to protect the vessels from compression due to external pressure or torsion, a function more aligned with a need for compressive stiffness, such as in articular cartilage [9, 35, 36].

Umbilical cord

Umbilical cord tissue contains connective tissue substances in a loose matrix network that can be both structural and supplemental to natural trophic and anti-inflammatory processes [37]. Wharton’s jelly (WJ) is the term for this umbilical tissue that is neither a vessel nor the encasing external amniotic layer of the umbilical cord. Wharton’s jelly contains a proteinaceous fluid, collagens, and glycosaminoglycans. In humans, WJ has been applied directly at surgery [25, 38–40] and was observed to supplement cartilage defects, proposed, in part, by tessellation into damaged cartilage [25]. Wharton’s jelly has been injected intra-articularly in humans with OA [26] or connective tissue defects of the hip [38], knee [25, 26, 39], talus [40], and shoulder [41]. While further research is needed, WJ has been successfully used in over 180 homologous sites as tissue supplement grafts in humans, showing significant improvements in pain and function scores without reported adverse events [38, 42]. The lack of adverse events has putatively been attributed to the primitive nature of the tissue and immune privilege [38] that is also recognized in human and equine stem cells [31, 43]. This minimally manipulated equine connective tissue was derived from equine UC and has an analogy to WJ in humans.

Equine micro-particulate allografts

In the equine species, micro-particulate, flowable allografts have been reported and include dental pulp [27] and amniotic membrane sources. [34]. Dental pulp tissue harvested from unerupted neonatal tooth buds of deceased foals and processed by physical means to microparticles demonstrated efficacy in a controlled, prospective clinical trial for the treatment of osteoarthritis and desmitis in lame horses. The product is no longer available. Several amniotic products have been commercialized, in some cases with little characterization or evidence of their safety. Currently, marketed amniotic tissue allografts may differ in formulation and/or route of administration from published work or safety reports.

Contribution to the field

The results of our study will help to fill the gap in peer-reviewed publications on the characterization and biocompatibility of resourced tissue for MSK supplementation. Peer-reviewed studies, human and equine, are lacking or limited, particularly regarding the inclusion of controls for in vivo studies. Human clinical trials for UC allograft injectables for OA are just getting started [10, 39, 40]. Umbilical cord is a unique and potentially preferable connective tissue resource due to the presence of WJ, the immune privilege attributed to tissues of fetal origin and WJ’s inherent purpose to supplement and protect the structure of surrounding tissues.

For the future advancement of efficacious, particulate tissue allografts, it is critical that they be characterized, shown to be safe in target tissues in vivo, and obtained and stored with quality practices that align with the Food and Drug Administration (FDA) Guidelines for Human Cellular and Tissue-Based Products [44]. Potential products should be studied in the target species through prospective, controlled, and randomized studies for biocompatibility and safety as part of due diligence research. Animal studies also serve as preclinical data for human clinical applications. Publication is an important documentation of scientific peer review and disclosure of information that is often omitted prior to marketing and commercialization, particularly for biological tissue grafts [8]. This study aims to fill this gap in scientific knowledge of a UC microparticulate tissue allograft.

Bio-scaffolds from connective tissue can provide structural supportive networks to damaged tissue and allow for an improved tissue microenvironment. These extracellular connective tissue matrices can be applied to fill tissue voids in damaged or torn tissue and potentially serve as a structural foundation and supplement for the endogenous cells to adhere, migrate into, and begin the process of healing and repair [8]. Potential advantages of this particulate, equine fetal tissue allograft for structural supplementation of MSK tissues include a biodegradable, natural source designed to resist tissue compression and torsion, and one that is amenable to injectable delivery and convenient to use. The underlying quality of the product is assured by cryopreservation of fresh tissue, donor health and uniformity, and a meticulously selected tissue source of low immunogenicity.

Goals and aims

The goal of this study was to characterize the protein content of cryopreserved, microparticulate equine umbilical cord-derived (UC) allograft suspension and to evaluate its in vivo biocompatibility for potential supplementation into equine joint, tendon, and ligament tissue. The study aims were to:

1. Report proteomic and enzyme-linked immunosorbent assay (ELISA) data on the UC allograft.

2. Report clinical, clinical pathologic, imaging, and histologic outcomes in vivo of UC suspension administered by injection to equine joints, ligaments, and tendons of normal and diseased horses.

Our hypotheses were that allogeneic umbilical cord-derived particulate suspension for injection would have a consistent protein profile in different donors and within batches and would be clinically and histologically biocompatible when injected into both normal and diseased joints and ligaments, without causing any adverse events.

Methods

Aim 1 design—characterization

Cryopreserved (− 80 °C) equine umbilical cord (UC)-derived tissue suspension (equicentCTM™ Equine Performance Labs, 5540-B N. Lamar Blvd, Austin, TX, USA) from 11 different approved single-sourced donors (Coteau Grove Farms, 270 Borel Rd, Sunset, LA 70584) that had met inclusion criteria, plus 2 additional samples from 1 donor (13 total samples) were thawed and processed by mass spectrometry (MS), tandem mass spectrometry (MS/MS) (Mass Spectrometry and Proteomics Core Facility, The Herbert Wertheim UF Scripps Institute for Biomedical Innovation & Technology; University of Florida, Jupiter, FL 33458, USA (RRID:SCR_023576)) and ELISA (RayBiotech Life, Inc. 3607 Parkway Lane Suite 200 Peachtree Corners, GA 30092, USA). Proteomic analysis included protein identification, relative quantification using total spectral counts (TSC), and function. Using high-resolution MS and MS/MS, proteins and post-translational modifications were identified using a procedure that included tissue processing and mass analysis on Orbitrap Fusion Tribrid mass spectrometer^d coupled to an nEasy-LC1000 nano liquid chromatography. Tandem mass spectra (MS/MS) were searched against the Equus caballus proteome set (https://www.uniprot.org/proteomes/UP000002281) using Thermo Proteome Discoverer 2.5.0.400 (Thermo Fisher Scientific, 355 River Oaks Pkwy, San Jose, CA 95134, USA) and UniProt public databases. Identified proteins (name and accession number) were sorted for protein expression from most abundant to least abundant using a semiquantitative analysis of protein content expressed as total spectral count (TSC). Spectral counts over 100 were designated as high based on all the TSC counts from all the other proteins [45]. Protein TSCs detected in all 11 donors were analyzed further for variability within and among donors and searched for function. Spectral counts over 40 were considered relevantly abundant. Using multiplex ELISA, UC allograft suspensions were analyzed for concentration of interferon-gamma (IFNg), interleukin-1 alpha (Il-1alpha), interleukin-1 receptor antagonist (Il-1ra), interleukin-2 (Il-2), interleukin-4 (Il-4), interleukin-8 (Il-8), interleukin10 (Il-10), interleukin-15 (Il-15), monocyte chemoattractant protein-1 (MCP-1), vascular endothelial growth factor (VEGF), and hyaluronic acid (HA).

Aim 2 design—in vivo studies

Controlled study in clinically normal horses

This was a two-dose (Day 0 and Day 21), masked, randomized, crossover design study with clinical and analytical evaluations of several parameters related to clinical and tissue biocompatibility performed in accordance with prior approval by the research facility’s Institutional Animal Care and Use Committee (IACUC) (ETCR-24–03) (East Tennessee Clinical Research, Inc., Rockwood, TN 37854, USA). Candidate horses underwent a 7-day acclimation period and met the following inclusion criteria: light saddle breeds, > 3 years of age, normal health, hematologic and clinical chemistry parameters (University of Tennessee Veterinary Clinical Pathology Laboratory, 2407 River Dr, Knoxville, TN 37996, USA), including SAA (Serum Amyloid A. Stablelab® EQ-1 Handheld Reader and test packs. Patterson Veterinary Supplies, 1501 Center Park Dr, Charlotte, NC 28217, USA), no apparent lameness and clinically normal carpi and suspensory ligaments based on normal appearance, palpation, range of motion for joints, and pain on pressure for ligaments. Baseline evaluations were documented before initial injections on Day 0.

Concurrently, four horses started the study with the joint administration and four horses started the study with the ligament administration (Additional File 1: Fig. S1). Each horse had a single dose of UC suspension (equicentCTM™ 1.5 mL; as ~ 100-µm-sized particles from 50 mg of tissue) administered into one randomly assigned midcarpal joint or randomly assigned lateral branch of the front suspensory ligament. Twenty-one days later, the alternate joint or ligament was injected with UC suspension (equicentCTM™) resulting in a 42-day study with each horse receiving both a joint and ligament injection. The contralateral joint and ligament were injected contemporaneously with an equivalent volume of sterile saline (0.9% sodium chloride) control (C) solution (0.9% sodium chloride solution injection USP, Hospira Inc., Lake Forest, IL 60045, USA). In a blinded fashion, serial observations, physical and lameness examinations, synovial fluid analyses, pain on flexion or pressure scores, swelling circumference measurements, and clinical pathology analyses were evaluated for effects of UC compared to baseline, controls, and administration of two doses 21 days apart. Adverse events were recorded (Additional File 1: Fig. S1).

Physical examinations, pain, and inflammation evaluations were conducted on experimental days (Additional File 1: Fig. S1); once on Day 0 (before) and on Days 1, 3, 5, 10, and 21 after injections by a blinded clinician to evaluate rectal temperature (T), pulse rate (P), and respiratory rate (R), SSA, synovial fluid analysis for joints, scoring for pain on flexion (joints; 0–4), pain on pressure (ligaments; 0–4), and swelling (circumference [cm]). Scores were 0 = none; 1 = minimal; 2 = mild; 3 = moderate; and 4 = marked. Lameness evaluations (clinical and kinematic (SleipApp, Sleip.com, AI-driven gait analysis, Stockholm, Sweden) were conducted on Days 0, 3, 5, 10, and 21 postinjections using the lameness classification system as the primary lameness outcome. Lameness classification was performed at the trot and graded 0 = no lameness, 1 = intermittent lameness, 2 = consistent lameness under special circumstances, 3 = consistent lameness, 4 lameness at the walk and trot, or 5 nonweight-bearing lameness (https://thehorse.com/199286/the-aaep-horse-lameness-scale-explained/). Kinematic video recordings provided a video record of the horse for use as a reference if needed. Blood samples for hematology and chemistry panels (Days 0 and 21) and serial synovial fluid samples were analyzed (University of Tennessee Veterinary Clinical Pathology Laboratory, 2407 River Dr, Knoxville, TN 37996, USA) for cell counts, differentials, and total protein. Synovial fluid was further analyzed quantitatively for albumin and globulin. Synovial fluid was also analyzed by ELISA (RayBiotech Life, Inc., 3607 Parkway Lane, Suite 200 Peachtree Corners, GA 30092, USA) for the same cytokines as in Aim 1.

Case study in MSK diseased horses

This was a single-dose prospective case series with baseline control and post-treatment clinical and analytical evaluations of several factors related to biocompatibility in three research horses with lameness due to osteoarthritis, desmitis, and tendonitis. The study was conducted in accordance with an approved animal use protocol by the research facility’s IACUC (ETCR-24–03) (East Tennessee Clinical Research, Inc., Rockwood, TN 37854, USA). Horses were acclimated to study conditions of stall confinement for 1 month and enrolled 4 days prior to Day 0 when baseline control data were collected. Horses were healthy, as determined by daily general health observations and physical examinations. Baseline digital radiographic imaging (four standard views) was conducted on two horses confirming joint abnormalities of the carpus (one radiocarpal and one mid-carpal joint) including enthesiophytes, osteophytes, irregular joint surfaces, and evidence of old bone injury. Baseline digital ultrasound imaging (cross-sectional and longitudinal sections of the flexor tendons and suspensory ligament from the carpus to the fetlock) was conducted on one horse confirming desmitis and tendonitis of the accessory ligament of the deep digital flexor tendon and superficial digital flexor tendon. Similar imaging was performed at the end of the study (Pinnacle Equine Hospital, 7001 Ball Rd, Knoxville, TN 37931, USA). Serial outcomes for lameness, pain, enlargement, and inflammation (local and systemic) were completed in all horses using the same methods as in the first in vivo study on clinically normal horses (Additional File 1: Fig. S2).

Two horses with lesions of carpal OA received a single, intra-articular injection of 1.5 mL of UC (equicentCTM™) into the affected joint on Day 0. A single horse with forelimb desmitis and tendonitis received a single intralesional injection of UC (equicentCTM™) on Day 0; 1.5 mL was injected directly into the tendon and ligament lesion. Physical examinations, pain, and inflammation evaluations were conducted once on Days 0, 1, 3, 5, 10, 14, 21, and 28 to evaluate various clinical biocompatibility parameters. Lameness evaluations (clinical and kinematic) were conducted on Days 0, 3, 5, 10, 14, 21, and 28 postinjections. Blood samples for hematology and chemistry panels (University of Tennessee Veterinary Clinical Pathology Laboratory, 2407 River Dr, Knoxville, TN 37996, USA) were collected on Days 0 and 28. Blood samples for SAA measurement and synovial fluid for clinical pathology and cytokine analyses were collected on Days 0, 1, 3, 10, and 28. The trial terminated on Day 28 (Additional File 1: Fig. S2).

Controlled pathologic study for acute tissue biocompatibility

This was an IACUC-24–089 approved (Louisiana State University, Skip Bertman Dr, Baton Rouge, LA 70803, USA), controlled, cross-over design multiple-dose study of the clinical and pathological response of joint tissues (carpus and fetlock), ligament (lateral branch of the suspensory ligament), and pectoral lymph nodes to a similar dose (equicentCTM 1.5 mL) of UC administered by injection. Assigned joints and ligaments were normal on radiography and ultrasound prior to the start of the study. Ultrasound-guided injections confirmed the placement of the UC allograft within the center of the ligament along an ~ 2-cm length. Contralateral joints and ligaments were similarly collected and processed as controls (C). Each of the two horses had two joints in one randomly selected limb (carpus and fetlock, alternated in each horse) injected with UC 2 days or 5 days prior to, and ligament 2 days prior to, euthanasia (Euthaphen® Dechra Veterinary Products, Overland Park, KS) for reasons unrelated to MSK or general health abnormalities. Horses received a nonsteroidal anti-inflammatory drug (phenylbutazone oral paste for horses, Patterson Veterinary Supplies, 1501 Center Park Dr, Charlotte, NC 28217, USA) orally (2 gm BID) for 3 days, physical examinations daily, and comfort assessed twice daily in the 3 × 3 m stall. Synovial fluid gross appearance and total protein were performed at injection and tissue harvest (Additional File 1: Fig. S3). Immediately after euthanasia, synovial membrane, joint capsule, ligament, articular cartilage, and pectoral lymph nodes were harvested by dissection at multiple sites, both near and away from the injection site. Specimens from UC and C tissues were fixed in formalin, embedded in paraffin, and sectioned (5 µm). All tissue sections (UC and C) were stained with hematoxylin and eosin. Articular cartilage was also stained with metachromatic toluidine blue to assess the quantity, location, and pericellular content of aggrecan. The surface of articular cartilage was evaluated for the presence of UC allograft. Histologic sections of UC and C tissues and corresponding lymph nodes were assessed in a blinded fashion by a board-certified veterinary pathologist (DR) for inflammation, tissue architecture, and the potential presence of allograft.

Statistical analyses

Descriptive statistics were performed within and among donors and included mean ± standard deviation (SD) and coefficient of variation (CV) expressed as a % and cutoffs for TSC abundance were used to classify proteins relative quantity. Objective clinical and synovial fluid data were further analyzed by paired t test and two-factor analysis of variance (UC vs C and Day). Ordinal scored data were reported as median and range. Significance was set at P ≤ 0.05. Outcome data from ELISA were graphed (mean ± SD and standard error of the mean (SEM)) and outcomes with statistical significance were additionally graphed as % change from baseline to show trends over time using each horse as its own control.

Results

Proteomics

Two thousand six hundred and forty-five TSCs were identified and annotated with a protein name and unique accession number. Rare, annotated proteins with TSC of 3 or less represented 1182 TSCs, contributed insignificantly to the total protein content, and were present in only 3% of donors; therefore, they were not further analyzed.

Consistency in protein content between different donors was evident even in lower abundance (mean 4–19) TSCs. As mean TSC increased, so did the percentage of donors that expressed that protein. Specifically, at a mean TSC of 4, nearly one-half of the proteins (44%) were expressed in all donors, a large increase from 3% expressed at a mean TSC of 3 or less. At a mean TSC of 5, 6 and 7, 74%, 81%, and 95% of every donor expressed the protein, respectively. Of the 506 proteins in the TSC range of 8–17, 96%–100% were expressed in every donor.

Proteins with a mean TSC of 18 to 39 were considered of lower relative abundance, however, were present in every donor. Even at lower abundance, these proteins may have a highly relevant function. Therefore, proteins with mean TSC of 18–39 were further analyzed and proteins with < 20% CV were identified with consistent presence within the UC allograft. Some proteins in this group include insulin-like growth factor 2 (anabolic particularly to chondrocytes), clusterin (protection from oxidative injury), dimethylarginine dimethylaminohydrolase 2 (nitric oxide synthase inhibitor, anti-inflammatory), cadherin (mediates cell–cell adhesion), many fibrillar collagens, and matrix glycoproteins, to name a few (full proteomics analysis data may be available upon request).

Proteins with a mean TSC of 40–99 were considered of moderate relative abundance and were present in this abundance range in every donor. These proteins have significant biological roles in tissue structure (fibrillar such as collagens, and hydroscopic such as glycans), tissue repair, metabolism, protection to tissue and cellular homeostasis (antioxidant), energy generation and homeostasis, etc. Select and highly relevant, notable proteins are represented within this range (Additional File 2: Table S1).

Proteins with a range mean TSC of 100 or greater (80 proteins) were considered as highly abundant and were present in high abundance in all donors with few exceptions and all proteins of all donors had high-moderate abundance with a median TSC of 89 or greater. The predominant functions of these well-defined proteins included structural support and protection of connective tissue through a variety of roles and regulatory functions in the connective tissue matrix (Additional File 2: Table S2).

In the triplicate samples from one donor, the protein patterns described above were mirrored in all three samples with 80 proteins that had a mean TSC of 100 or greater, the same as the mean of the 11 donors. The CV between the samples for all 245 proteins with a mean TSC of 100 or greater, ranged from 0.6 to 13.3%, with a mean of 4.7%, falling within the strictest standards of variability.

ELISA

Cytokine concentrations in UC suspension (nonparticulate) are reported in Table 1. Values for the inflammatory cytokines (INFg, Il-1, Il-2, Il-4, Il-8, Il-15) were low, near the low end of the standard curve (INFg, Il-2, Il-4, and Il-8), or near the limit of detection (Il-15). Interkeukinl-8, Il-1, and Il-1ra were < 1000 pg/mL.

Table 1.

Umbilical cord allograft suspension mediator content as determined by ELISA

Mediator (pg/mL)	Mean ± SEM	Coefficient of variance within donor (CV %)
IFNg	184.5 ± 32.1	30.1
IL-1a	969.3 ± 5.9	1.1
IL-1ra	317.9 ± 15.6	8.5
IL-1ra/IL-1a	0.3 ± 0.02	9.6
IL-2	839.7 ± 19.9	4.1
IL-4	85.5 ± 21.7	43.9
IL-8	870.8 ± 30.4	6.1
IL-10	95.1 ± 22.5	40.9
IL-15	689.2 ± 20.3	5.1
MCP	1175.8 ± 25.7	3.8
VEGF	833.7 ± 26.0	5.4
HA	42.3 ± 3.6	14.9

N = 3; samples run in triplicate

IFNg interferon gamma,

IL-1 interleukin-1a,

IL-1ra interleukin-1 receptor antagonist,

IL-2, −4, −8, −10, and −15 the corresponding interleukin,

MCP-1 monocyte chemoattractant protein-1,

VEGF vascular endothelial growth factor,

HA hyaluronic acid

Controlled study in clinically normal horses

All eight horses of various breeds completed the study, five male castrates and three females, ranging in age from 8 to 20 years and ranging in body weight from 750 to 1150 lbs. All hematology and chemistry blood parameters were within acceptable limits for the duration of the study. Low SAA values were sustained for the duration of the study with a mean baseline value of 32. SSA values after injection decreased significantly to a mean of 6 for the duration of the study (Table 2).

Table 2.

Mean ± SEM serum amyloid A for umbilical cord (UC) or control (C) for 8 horses

Outcome		Duration after injection (days)
		0	1	3	5	10	21
Serum amyloid A (SAA) 1 st injection		32 ± 14	8 ± 6	8 ± 2	4 ± 1	39 ± 38	24 ± 19
Serum amyloid A (SAA) 2nd injection		7 ± 5	6 ± 3	8 ± 3	4 ± 2	6 ± 2	6 ± 4

SSA values of 100 were considered within normal range for the general horse population, with 300–450 indicating systemic inflammation, and over 500 indicating possible infection

Key outcome parameters for carpal joint and ligament after UC and C injection are listed in Tables 3 and 4, respectively. Horses administered UC or C in either the joint or ligament did not demonstrate clinical signs of lameness, pain, or inflammation after Day 1 for at least 21 days or after the second injection for a total of 42 days. Scores for pain on pressure/flexion and lameness, as well as values for circumference (swelling), were not different between C or UC or from baseline. Joint fluid analyses demonstrated a neutrophilic transudative synovitis that peaked on Day 1 which was greater in UC than in C and had returned to baseline and normal values by Day 10 (Table 3). Horses administered C or UC at a different site 21 days after the first injection also did not have clinical signs of pain, lameness, or inflammation for at least 21 days and showed a similar neutrophilic transudative synovitis that peaked on Day 1 that had returned to baseline and normal values by Day 10. No adverse events attributable to the C or UC administration were observed.

Table 3.

Mean ± SEM or median (range) joint outcomes for umbilical cord (UC) or control (C) for 8 horses/site

		Duration after injection (days)
Outcome	Joint	0	1	3	5	10	21
Circumference (cm)	C	29.2 ± 0.5	29.8 ± 0.4	26.9 ± 1.1	27.2 ± 0.8	28.5 ± 0.8	29.6 ± 0.2
	UC	29.9 ± 0.4	30.1 ± 0.3	28.2 ± 1.1	27.5 ± 0.8	28.6 ± 0.7	29.4 ± 0.4
Pain of flexion (0–4)1	C	0	0	0	0	0	0
	UC	0	0	0	0	0	0
Lameness score (0–4)2	C	0	0	0	0	0	0
	UC	0	0	0	0	0	0
Synovial fluid TNCC3	C	137.5 ± 18.3	1850a ± 594.3	714.3a ± 281.5	1,450a ± 979.1	312.5 ± 100.8	187.5 ± 39.8
	UC	171.4 ± 28.6	21,237.5a,b ± 5,980.7	712.5a ± 344.8	1,237.5a ± 244.2	328.6 ± 74.7	212.5 ± 47.9
Synovial fluid total protein4	C	1.2 ± 0.1	1.7a ± 0.4	1.4 ± 0.2	1.8 ± 0.2	1.3 ± 0.1	1.3 ± 0.1
	UC	1.2 ± 0.1	3.3a,b ± 0.4	1.4 ± 0.2	1.8 ± 0.1	1.5 ± 0.1	1.3 ± 0.1
Synovial fluid albumin4	C	0.7 ± 0.1	1.0a ± 0.2	0.8 ± 0.1	1.0 ± 0.1	0.8 ± 0	0.9 ± 0
	UC	0.7 ± 0.1	1.6a,b ± 0.2	0.8 ± 0.1	1.1 ± 0	0.9 ± 0	0.9 ± 0.1
Synovial fluid globulin5	C	0.5 ± 0	1.1a ± 0.4	0.6 ± 0.1	0.8 ± 0.2	0.6 ± 0.1	0.5 ± 0.1
	UC	0.6 ± 0.1	1.3a ± 0.3	0.7 ± 0.1	0.8 ± 0.1	0.6 ± 0.1	0.6 ± 0.1
Synovial fluid PMN (%)6	C	9.0 ± 7.2	52.6a ± 13.9	16.0a ± 9.6	16.1a ± 5.8	9.3 ± 4.3	2.8 ± 0.9
	UC	4.3 ± 2.6	65.4a,b ± 6.3	18.6a ± 7.8	11.6a ± 6.8	3.7 ± 0.9	5.8 ± 1.4

Synovial Fluid Days 0,1,3 for the first four horses was not analyzed at the laboratory due to an error

¹Pain scores: 0 = none, 1 = minimal, 2 = mild, 3 = moderate, 4 = marked

²Lameness classification: 0 = no lameness; 1 = intermittent lameness at trot; 2 = consistent lameness under special circumstances; 3 = consistent lameness at the trot; 4 = lameness at the walk; 5 = nonweight-bearing lameness

³TNCC = total nucleated cell count (cells/μL); normal range is < 1000 TNCC/μL

⁴Total protein values < 2.0 g/dL are within acceptable limits, however, typically are ≤ 1.5 g/dL in synovial fluid. Values ranging from 2.0 to 4.5 gm/dL reflect a transudate, and over 4.5 g/dL reflect an exudate

⁵Normal synovial fluid values for albumin are < 1.2 g/dL and for globulin are < 3.3 g/dL

⁵Normal PMN percentage is ≤ 10%

^aSignificantly differs from baseline (P < 0.05)

^bSignificantly differs between C and UC (P < 0.05)

Table 4.

Mean ± SEM or median (range) ligament outcomes for umbilical cord (UC) or control (C) for 8 horses/site

		Duration after injection (days)
Outcome	Ligament	0	1	3	5	10	21
Circumference (cm)	C	21.4 ± 0.4	21.6 ± 0.3	21.6 ± 0.6	21.9 ± 0.6	22.1 ± 0.8	21.6 ± 0.7
	UC	21.3 ± 0.3	21.5 ± 0.3	21.7 ± 0.4	22.0 ± 0.6	21.7 ± 0.6	21.5 ± 0.6
Pain on pressure (0–4)a	C	0	0	0	0	0	0
	UC	0	0	0	0	0	0
Lameness score (0–4)b	C	0	0	0	0	0	0
	UC	0	0	0	0	0	0

^a Pain scores: 0 = none, 1 = minimal, 2 = mild, 3 = moderate, 4 = marked

^bAmerican Association of Equine Practitioners Lameness Grades: 0 = no lameness; 1 = intermittent lameness at trot; 2 = consistent lameness under special circumstances; 3 = consistent lameness at the trot; 4 = lameness at the walk; 5 = nonweight-bearing lameness

Synovial fluid total protein (g/dL) analysis was completed on four horses for Days 0, 1, and 3 and all eight horses for other days. Total protein synovial fluid values were within the acceptable range (< 2.0 g/dL) and near-normal typical range (< 1.5 g/dL) at most time points, with the greatest increase at Day 1 in both C and UC (i.e., a transudate). Synovial fluid albumin (g/dL) increased on Day 1, was greater in UC than C, above the normal range, and quickly returned to and sustained normal values on Day 3. Synovial fluid globulin (g/dL) did not increase above the normal range but was greater than baseline in both C and UC. Horses administered C or UC at a different site 21 days after the first injection did not differ in synovial fluid protein values (Table 3).

Synovial fluid mediator concentrations were low and did not significantly change for the inflammatory cytokines Il-2, Il-4, Il-8, and Il-15 (Table 5). Il-1ra, however, increased significantly (P < 0.001) on Days 1, 3, and 10 after UC injection and were greater than saline control and baseline (Fig. 1). In all horses, synovial fluid Il-1ra was increased over the paired baseline value for the duration of the study (seven of eight horses) or until Day 10 (1 horse) after injection with UC that did not occur similarly after injection of saline. The statistically significant increase of Il-1ra with UC administration resulted in an increase of the Il-1ra to Il-1 ratio (Il-1ra/Il-1) several-fold greater than baseline and C (P < 0.001). This statistically significant increase was also seen at Day 5 indicating a UC effect that persisted longer than the transient neutrophilic response. The ratio remained > 1.0 for all horses to Day 10 or longer (Fig. 1; Table 5). Il-10 was significantly greater (P < 0.01) than both baseline and C after UC injection (Fig. 1). Monocyte chemoattractant protein (MCP)−1 significantly increased after UC and C injection compared to baseline on Day 1, and UC was greater than C (P < 0.05) on Day 1 (Fig. 1). MCP-1 values returned to baseline values without any other increase in the subsequent 20 days of the study.

Table 5.

Mean ± SEM cytokine concentration (pg/mL) in synovial fluid

		Duration after injection (days)
Cytokine	Joint	0	1	3	5	10	21
IFNg	C	426.6 ± 245.1	607 ± 421.6	661 ± 455.2	548.6 ± 308.3	600 ± 397.9	507.1 ± 343.2
	UC	207.1 ± 41.4	515.4 ± 305.6	349.3 ± 133.7	292.8 ± 102.8	313.5 ± 92.1	307.8 ± 57.2
IL-1a	C	947 ± 78.7	1015.5 ± 106.9	1119.9 ± 103.5	1168.3 ± 201.2	1160.6 ± 226.9	1266.9 ± 170.8
	UC	973.7 ± 118.9	969.7 ± 130.2	1114.9 ± 172.4	969.9 ± 78.6	1106.5 ± 185.8	1086.5 ± 158.8
IL-1ra	C	657.6 ± 225.1	8785.7 ± 2567.3	1767.6 ± 332.9	1294.1 ± 293.9	829.8 ± 322.2	687.9 ± 305.2
	UC	437.8 ± 72.4	15,635.2 ± 5565.6	1625.6 ± 332.3	2418.1 ± 1260.6	729.9 ± 141.9	525.6 ± 91.9
IL-1ra/IL-1a	C	0.7 ± 0.2	10.1 ± 3.6	1.6 ± 0.2	1.2 ± 0.3	0.6 ± 0.1	0.5 ± 0.1
	UC	0.5 ± 0.1	16.4 ± 4.3	1.9 ± 0.5	3.2 ± 2.0	0.7 ± 0.1	0.5 ± 0.1
IL-2	C	1089.1 ± 116.3	1218.3 ± 201.3	1483.4 ± 254.3	1711.5 ± 337.5	1500.9 ± 282.4	1270.5 ± 202.5
	UC	1072.9 ± 113.2	1125.8 ± 160.1	1317.6 ± 148.6	1295.6 ± 122.6	1501.6 ± 349.3	1250.5 ± 166.1
IL-4	C	272.4 ± 160.3	401.8 ± 290.0	477.3 ± 351.5	347.66 ± 199.7	309.4 ± 198.6	343.0 ± 247.0
	UC	155.5 ± 47.9	339.1 ± 203.6	253.5 ± 133.7	191.9 ± 72.0	213.6 ± 75.1	203.5 ± 63.6
IL-8	C	867.5 ± 63.5	1099.3 ± 98.5	1046.9 ± 81.5	1483.8 ± 522.4	1484.8 ± 455.5	1150.4 ± 145.0
	UC	872 ± 128.9	1169.8 ± 245.1	1439.1 ± 435.9	1131.3 ± 299.0	2934.6 ± 2061.1	1056.4 ± 162.8
IL-10	C	526.8 ± 166.8	824.7 ± 258.4	784.2 ± 334.8	764.8 ± 277.7	664.1 ± 297.8	643.8 ± 274.9
	UC	411.3 ± 51.9	943.7 ± 266.0	546.3 ± 117.6	498.6 ± 83.3	515.0 ± 84.3	461.5 ± 61.4
IL-15	C	823.5 ± 104.1	959.8 ± 207.2	1169.8 ± 220.0	1294.4 ± 218.6	1192.4 ± 265.0	1049.6 ± 203.4
	UC	759.8 ± 97.8	949.4 ± 194.0	1028.0 ± 142.1	1043.3 ± 107.9	1185.3 ± 238.6	938.3 ± 103.4
MCP-1	C	4874.9 ± 2119.5	10,265.3 ± 4536.8	7212.0 ± 3223.2	6810.6 ± 2438.1	4967.0 ± 1278.1	3629.6 ± 598.8
	UC	5479.0 ± 2259.0	14,610.8 ± 4737.8	6471.8 ± 1933.9	7635.8 ± 2405.7	6312.1 ± 3016.4	3896.0 ± 1009.2
VEGF	C	1608.5 ± 222.5	1617.9 ± 105.9	1630.2 ± 133.7	1583.4 ± 136.6	1747.3 ± 200.8	2001.3 ± 158.1
	UC	1523.6 ± 259.9	1929.2 ± 252.7	1860.9 ± 197.3	1278.1 ± 116.3	1642.4 ± 281.5	1842.0 ± 238.0

Cytokine abbreviations are the same as Table 1

C control,

UC umbilical cord

Fig. 1.

Selected synovial fluid cytokine concentrations (pg/mL) in normal horses injected into the carpal joint with umbilical cord (UC) allograft or saline control (C) expressed as a percentage of baseline. A A dramatic increase in Il-1ra after UC injection was significantly greater than Day 0 baseline and C (P < 0.001) and remained greater than baseline values in seven of eight horses until the end of the study, a finding that did not occur in control. B Il-1concentrations were low and did not significantly increase after UC injection. C The Il-1/Il-1ra ratio was an even greater difference from C due to the very low Il-1 that exaggerated the ratio change (P < 0.001). D VEGF was significantly greater after UC injection than baseline Day 0 or C (P < 0.01). VEGF growth factor is a biologic potent promoter of blood vessel synthesis, particularly during development and tissue differentiation. In damaged and degenerate tissues, vascularization is a key first step to repair. E Il-10 is a potent anti-inflammatory cytokine with reported chondroprotective effects and was significantly increased to greater (P < 0.01) than both baseline and C after UC injection. F MCP-1 is a chemokine, a group of secreted proteins within the cytokine family, whose generic function is to induce cell migration. The significant (P < 0.05) increase in MCP-1 after UC injection compared to baseline and C, along with rapid return to baseline values without any other increase in 20 days suggests lack of continued immune stimulation and inflammation in vivo

Case study in MSK diseased horses

Three horses completed the study and ranged in age from 12 to 18 years, two females and one male castrate. No adverse events attributable to the UC administration were observed. All horses showed at least one grade of improvement in lameness and/or pain compared to baseline values (Table 6).

Table 6.

Outcome scores for lameness and pain for horses with osteoarthritis or desmitis and tendonitis

	Evaluation day
Score	0	1	3	5	10	14	21	28
Mean lameness score1	3	*	2	2	1	1	2	2
Ligament pressure score2	3	2	0	1	1	2	0	0
Mean carpal flexion score2	1	1	0	1	0	0	0	0

^* Horses not trotted for lameness on Day 1

¹Lameness classification: 0 = no lameness; 1 = intermittent lameness at trot; 2 = consistent lameness under special circumstances; 3 = consistent lameness at the trot; 4 = lameness at the walk; 5 = nonweight-bearing lameness

²Pain scores: 0 = none; 1 = minimal; 2 = mild; 3 = moderate; 4 = marked

None of the horses developed clinical signs of increased pain, lameness, or swelling after injection at any time point for the 28 days of the study. Values for circumference (swelling) were variable and did not change consistently in any horse. Joint fluid analyses demonstrated a mild to moderate transudative synovitis that returned to within normal baseline values after Day 3 (Table 7). Little alteration occurred in synovial fluid albumin and globulin, and values remained below reported normal for joint fluid (1.2 and < 3.3 g/dL, respectively). Synovial fluid inflammatory cytokine Il-1 values were greater at baseline in OA horses than in our normal horses (Table 2), such that the starting average Il-1ra/Il-1 ratio was 0.3 (fractional and undesirable), and lower than the approximately even ratio (0.95) in our normal horses. After injection of UC in these OA joints, synovial fluid Il-1ra spiked as in the normal horses, such that the Il-1ra/Il-1 ratio was converted to > fourfold and was greater than C. (Table 7). Similarly, anti-inflammatory mediators Il-10, HA, and VEGF were increased and greater than baseline and controls. Serum amyloid A did not detect systemic inflammation caused by UC in horses with OA or desmitis and tendonitis. Digital radiography for carpi and digital ultrasound of the affected forelimb flexor tendons and suspensory ligaments were compared before treatment to study termination (Day 28) without changes noted over the course of the study. Hematology and serum chemistry were within acceptable limits at baseline and at the end of the study.

Table 7.

Trends in mean synovial fluid values in osteoarthritic joints injected with umbilical cord allograft

	Day 0	Day 1	Day 3	Day 10	Day 28
TNCC1 (cells/μL)	150.0	9,000	2,400	250.0	200.0
Albumin (g/dL)	0.8	1.4	1.2	0.8	1.0
Globulin (g/dL)	0.4	1.2	0.7	0.6	0.4
IL-1a	1386.5	1204.0	1160.0	1271.0	1314.5
IL-1ra	429.0	4926.0	2086.0	447.5	636.5
IL-1ra/IL-1a	0.3	4.1	1.8	0.4	0.5
IL-10	280.0	529.5	493.5	328.0	353.5
VEGF	1535.5	1905.5	2253.0	1641.5	1850.5
HA	127.7	301.2	303.4	175.0	122.7
IFNg	215.0	209.5	200.5	205.5	309.0
IL-2	1095.0	1131.0	1161.0	1198.0	1202.5
IL-4	147.0	142.0	125.0	117.5	462.5
IL-8	1039.0	2017.0	1105.5	1017.5	1173.0
IL-15	882.5	863.5	766.5	842.5	1028.0
MCP-1	2082.5	2514.0	5071.0	1887.5	1965.0

¹TNCC total nucleated cell count (cells/μL); normal range is < 1000 TNCC/μL; see abbreviations in Table 3 for cytokines

Total protein values < 2.0 g/dL are within acceptable limits, however, typically are ≤ 1.5 g/dL in synovial fluid. Values ranging from 2.0 to 4.5 gm/dL reflect a transudate, and over 4.5 g/dL reflect an exudate

Normal synovial fluid values for albumin are < 1.2 g/dL and for globulin are < 3.3 g/dL

Controlled pathologic study for acute tissue biocompatibility

At tissue harvest, the synovial membrane appeared grossly normal, with one of the four joints showing minor hemorrhage at the site of injection. Other than at the site of injection, the synovium and cartilage were grossly normal, indistinguishable from the contralateral control joints. All joints had clear, viscous, yellow synovial fluid with a refractometer protein of < 2.5 gm/dL in injected joints and < 2.0 gm/dL in uninjected control joints. Ligaments had no gross evidence of the injection site, including on cut surfaces, both longitudinal and cross-sectional through the site of the injection. Pectoral lymph nodes were grossly normal and of similar size and appearance between the injected and the uninjected control limb. Similar to our findings in the single-dose biocompatibility study, these horses with multiple UC allograft doses injected over 3 days (one carpus, one fetlock, and one suspensory ligament) were normal on physical examinations and general health observations. Horses were not lame in the stall and had no visible swelling after injection.

Histologic evaluation of 90 glass slides from injected and control joints, ligaments, and corresponding lymph nodes was imaged in a blinded fashion using a Motic digital scanner and sent to a veterinary pathologist (DR) for evaluation using the proprietary Motic viewing software. After unblinding, all intra-articular and extraneous materials at the injection sites were further examined and compared with control tissues for the presence of the experimental material. No distinct signature of the experimental product was established in the tissues examined. This suggested the injected material was readily accepted over the acute (2–5 days) period. Except for the needle tract, the subintimal synovium of none of the injected joints had evidence of reaction to foreign material. Two focal areas of acute mild inflammation were identified, and both were associated with adjacent needle tracts. In one well-defined needle tract in the ligament, eosinophilic amorphous tissue was filling the tract with mild multifocal areas of inflammation visible. There was no evidence of a strong antigenic response, and no foreign material transiting the regional nodes examined. The nodes examined were minimally reactive in controls and treated alike.

Discussion

The use of biologics in the management of MSK abnormalities is currently prevalent in sports medicine, orthopedics, and wound care, and numerous studies have been reported [9, 11, 46]. Given the variety in tissue source, processing, and formulation, results are difficult to compare and are often inconclusive or a low level of evidence. In many cases, commercialization has proceeded with insufficient characterization of the biologic product [8, 25]. We are the first to focus on a novel umbilical cord-derived equine product that is analogous to human WJ products [25, 40, 41]. Our results will serve as an essential first step to characterize and determine in vivo safety of the first reported use of allogeneic WJ in equine, as well as serve as a pre-clinical study for humans.

Our studies purposefully evaluated both the connective tissue matrix particulate via proteomics and the suspension diluent via ELISA, to identify proteins in the bio scaffold and to determine if inflammatory cytokines leached into the diluent during processing or storage. The proteomics data supported our hypotheses by finding a strikingly similar rich and robust variety of relevant proteins in the UC product from each donor, supporting a consistent, stable characterization of the allograft. Variability among the more abundant proteins from the same batch (donor) was low, demonstrating the high fidelity of the processing specifically for the characterization of UC product. ELISA data was also consistent for the presence of VEGF protein and HA in the product and confirmed that the UC suspension was low in cytokines; at or near the limit of detection, similar to values reported in normal equine joints [47], or similar to our normal synovial fluid reported in Table 5 as Day 0. MCP-1 and the growth factor VEGF were higher in the suspension diluent than other mediators but less than found in normal synovial fluid on Day 0 (Table 5). Importantly, UC processing did not elicit a significant release of inflammatory mediators. These data supported that the product is not delivering inflammatory cytokines upon administration. The particulate UC proteins would be expected to be released into the joint by degradation of the protein scaffold in vivo. The proteins of potential value were found in the particulate bio scaffold as indicated by the proteomics data (Additional File 2: Tables S1 and S2). In addition to structural components, many proteins necessary for the regulation and control of inflammation, protection of tissues from oxidants, as well as support for cellular homeostasis necessary in extracellular matrix turnover, are consistently present in abundance in the bio scaffold particles. The presence of these proteins supports the use of this UC extracellular connective tissue matrix as a tissue graft to be used along with medical procedures as a supplement to aid in tissue repair. In concert, these data showed a high and consistent tissue protein content with every protein defined by functions that are relevant to the health of connective tissues supporting this UC allograft as a homologous tissue for transfer to connective tissues.

Contributing factors that likely promoted the uniformity of results in our study included the careful and meticulous dissection of the optimal portion of the UC to maximize the desirable tissue and remove tissue that may be traumatized or influenced by outside factors prior to processing. External variables were minimized by following aseptic technique in a closed tissue processing facility under controlled standardized conditions and processing techniques (Equine Performance Labs, 5540-B N. Lamar Blvd, Austin, TX, USA) Additionally, the single source participating farm (Coteau Grove Farms, 270 Borel Rd, Sunset, LA 70584) has a high-quality, closed breeding operation with valuable mares and foals. Stability and quality of the thoroughbred donor herd were maintained under standardized policies and procedures under veterinary supervision, including the collection of the UC at foaling. Each mare was fully vaccinated on a schedule for optimal transfer of immunoglobulin to the foal, tested for pathogens, and comprehensively evaluated for health throughout the gestational and parturition periods. Only umbilical cords from healthy, vigorous foals were used for the final product.

Unique to the equine studies reported here, the source of the tissue is pure fetal UC without contamination of amnion or chorionic maternal tissues. Our tissue source has been processed to optimize the product, devoid of the dense vascular vessel lining and outer amnion membrane. Importantly, these abundant proteins are fetal in origin and populated the UC during gestational development. These young abundant proteins are perceived by longevity science as a “holy grail” for rejuvenating connective tissues [3], and some in UC are particularly associated with repair and regeneration, such as the smaller collagens [23], and in early development, such as TGF-β proteins. These proteins are not only structural, such as the fibrillar smaller collagens and filamentous proteins, but also the critical smaller proteoglycans and proteins that affect the quality of the extracellular matrix [24, 48]. These protein composites assist with extracellular matrix integrity, composition, interaction with cells, tissue permeability, and fluid flow that can affect tissue strength in tension and turgidity in compression [24]. This study confirmed that the diversity and abundance of relevant connective tissue proteins in particulate UC connective tissue matrix further support its use as an allograft [48].

Transfer of the particulate allograft also delivered tissue elements with mechanical properties. These particles are anticipated to serve as a scaffold with a strong affinity for local cell migration and adhesion, based on protein composition (fibrillar collagens) [23]. The scaffold structure likely can provide a filler for tissue tears and support vulnerable early fibroblasts in a more mature tissue than is provided by a fibrin clot. In soft connective tissues, such as synovial membrane or early granulation tissue during healing, mechanical properties are defined by permeability and kinetic (k) coefficients of fluid flux through the extracellular matrix (i.e., trans synovial fluid flow) [49]. Importantly, the small proteins, glycans, chondroitin, keratan, decorin, and proteoglycans found in abundance in UC (Additional File 2: Tables S1 and S2) are integral to forming networks that retain fluid both osmotically, chemically, and ionically (hydroscopic properties), properties attributed to WJ and HA. Hyaluronic acid was found in UC allograft suspension by ELISA. These proteins define the interstitial micro pressure and permeability (pore size) that is so important for compliance (flexibility) and turgidity, properties that resist compression and protect embedded cells. These are best described for the joint; however, devices that can measure interstitial fluid pressure, flow, and content are rapidly advancing this field to better understand the impact of extracellular matrix quality on tissue integrity and healing [50]. These potential properties of the extracellular matrix may provide supplementation that could enhance the acceleration of healing. UC connective tissue matrix delivers these proteins, in part as a suspension, and as a microparticle, functioning as an allograft, that can be targeted to the site of interest, including the soft connective tissues in joints and early granulation tissue.

Mediator analyses of synovial fluid in our in vivo studies showed that inflammatory mediators (Il-1, Il-2, Il-4, Il-8, Il-15) did not change from baseline, and UC was not different than control. The anti-inflammatory mediators (Il-1ra and Il-10 [51, 52]) increased several-fold in joints injected with UC compared to baseline and control joints (P < 0.01) in both the in vivo clinically normal joints and the in vivo diseased joints. This effect was substantial and produced a robust positive IL-1ra/Il-1 ratio (> fourfold) in both the clinically normal joints and OA joints. In the case study, the OA joint fluid had greater Il-1 at baseline than the clinically normal joints (P < 0.05). As a result, the fractional Il-1ra/Il-1 ratio (0.3) was low at baseline and reflected a profile out of balance with Il-1ra [53, 54]. A single injection of UC rapidly converted the ratio to a positive Il-1ra/Il-1 ratio for greater than 5 days that remained > 1 for a longer term (21 days) balance of these receptor competing proteins and slightly greater than our clinically normal joints (Il-1ra/Il-1 = 0.95). Although plasma and platelet products may contain greater Il-1ra than Il-1 [13, 53–55], information on the effect on the joint environment after injection is limited with only a few reports in horses [12, 55, 56]. One short-term study [56] was unable to show a difference in joint fluid Il-ra 48 h after platelet-rich plasma injection in lipo-polysaccharide-induced synovitis compared to no treatment. In another report [57], Il-1 was not measured such that the Il-1ra/Il-1 ratio could not be assessed. In a study of naturally occurring OA [12], a high Il-1ra/Il-1 ratio was reported but, importantly, was unable to show a statistical difference between the control and treated joints or between the before and after values. Earlier time points comparable to our study were not reported. Regardless, the consensus of the literature is that a greater amount of IL-1ra compared to Il-1 is desirable, and the presumption has been that the greater the ratio, the better [53]. The high Il-1ra to Il-1 provides the balance that gives the competitive edge to Il-1ra for binding to Il-1 sites on chondrocytes and synovial cells, potentially inhibiting the release of destructive enzymes that degrade cartilage [53, 54]. Conclusions on relevant concentrations of Il-1 and Il-1ra in joints in vivo are elusive and limited due to the variability of these cytokine concentrations in native OA, as well as dilution by joint fluid and the inflammation associated with the injection. Therefore, the ratio of Il-1ra to Il-1 in the joint fluid continues to be given greater credence over individual values [53, 55]. Plasma products are mostly water, albumin, globulin, and fibrinogen (92% or more) [58] such that growth factor proteins like insulin-like growth factor and TGF-beta, as well as IL-1ra, are present in picogram concentrations [55] unless the plasma is concentrated [12, 13]. These picogram administered doses are lower than the measured nanogram concentrations in the joint fluid after UC administration (3–4 ng/mL) or helper-dependent adenoviral-Il-1ra gene therapy (20 ng/mL) [59]. Gene therapy offers only the Il-1ra protein unlike plasma, platelets, or allografts. UC allograft offers the potential advantage of vast numbers of directly relevant connective tissue proteins with the additional potential advantage of being a scaffold and of fetal in origin.

Il-10 is a potent anti-inflammatory cytokine with reported chondroprotective effects and was significantly greater (P < 0.01) than both baseline and C after UC injection. The potential effect of this finding is unknown; however, the scientific evaluation of Il-10 as a therapeutic gene therapy for OA has been evaluated in horses with beneficial results [51, 52]. Synovial fluid Il-10 was also increased after injection of isolated amnion mesenchymal stem cells in dogs with OA and was directly correlated with improvement in joint range of motion [60]. MCP-1 is a chemokine, a group of secreted proteins within the cytokine family, whose generic function is to induce cell migration. Both chemokines and cytokines have essential roles in maintaining immune homeostasis. The early (Day 1) significant (P < 0.05) increase in MCP-1 after UC injection compared to baseline and C, along with rapid return to baseline values without any other increase in 20 days in our study (Fig. 1), supports the lack of continued immune stimulation of UC in vivo. Synovial fluid MCP-1 was decreased in dogs with OA after injection of isolated amnion-derived mesenchymal stem cells and inversely correlated to an increased joint range of motion [60, 61]. MCP-1 has been proposed as a biomarker of joint inflammation [60, 61], and our results also showed an increase on Day 1 during the peak of joint inflammation. Further work evaluating MCP-1 in horses with OA is warranted.

The anabolic mediator, vascular endothelial growth factor (VEGF), was significantly increased several folds after injection with UC, compared to control and baseline (P < 0.05). VEGF belongs to a subfamily of platelet-derived growth factors, specifically the cystine-knot growth factors. These are important signaling proteins involved in both vasculogenesis (de novo vessel formation) and angiogenesis (the growth of blood vessels from preexisting vasculature) [62, 63]. The angiogenesis and granulation tissue stage of tendon and ligament healing may benefit from increased VEGF [59, 63]. VEGF has an important role in the early healing of these tissues [62, 64] and may serve to supplement and accelerate tendon and ligament repair. In concert, these data report low synovial fluid inflammatory cytokines, except Il-1 in OA joints, and significantly increased anti-inflammatory (Il-1ra, Il-10) and growth factor mediators (VEGF) with a conversion from a fractional to a positive integer > 4 for the Il-1ra/Il-1 ratio in normal and OA joints.

Our study is the first to report on the safety and clinical biocompatibility of UC particulate allograft administered into clinically normal joints and ligaments in a prospective, controlled, randomized in vivo study. Biocompatibility was monitored for up to 42 days in normal horses and for 28 days in lame horses with OA and in lesions of desmitis and tendonitis. Studies in humans have focused on applications of amniotic/UC particles during surgery for OA pain (symptoms) or cartilage defects. However, significant limitations exist regarding comparative controls. We found, in a well-controlled study, that equine UC allograft was safe for use in equine joints and ligaments in the conditions of our study. No adverse events were attributed to the product or the administration of the product. These data support the future use of UC allograft in a clinical safety and effectiveness study in horses with clinical MSK disease of joints, tendons, and ligaments.

The MSK diseased horses in this case study were owned by a research facility which permitted a prospective protocol for quantitative outcomes, baseline values, and a controlled environment. Horses with MSK disease meeting inclusion criteria were brought into box stalls one month before enrollment which minimized the effect of rest on the outcomes (Additional File 1: Fig. S2). The lameness values, including kinematic data (not shown), were similar before and after the month of stall rest. Based on these assessments in a limited number of horses, the administration of microparticulate UC allograft into OA joints or diseased tendon/ligament was associated with an improvement in pain and lameness scores without evidence of systemic inflammation. Natural case studies, including in humans, often lack matched untreated controls due to the variation in clinical presentation; however, improvement of at least 1 grade in lameness and pain in the target species was observed. Although inconclusive due to the limited number of horses, the UC allograft did not cause them discomfort and may have supported comfort in horses with diseased connective tissues. Further research is warranted.

The pathology study provided both clinical and histologic evidence of short-term safety for dose escalation and repeat dosing (Additional File 1: Fig. S3). Systemically and clinically, the horses were comfortable. Synovial fluid protein values were like those reported in our clinically normal horse clinical study. Histologically, synovial membrane and ligament were normal, showing a lack of response to the allograft. Injection tracts were located and had some minor inflammation. In one injection tract in a ligament, the eosinophilic amorphous tissue filling the tract appeared to be compatible with UC allograft and showed some minor inflammation, biointegration, and degradative characteristics. This histopathologic finding could not be confirmed with certainty as the UC allograft. Future studies to identify the material and examine the tissue over a more extended period are warranted.

Following joint administration, a synovitis was anticipated and confirmed to be transient and without clinical signs of pain or outward inflammation and without histologic evidence of synovial membrane reaction. Within the measurements of this study, an immune response to a second administration was not detected in joints or systemically after 3 weeks of exposure from an earlier injection. A transient inflammatory response to pharmacologic or particulate injections into equine joints has been repeatedly reported as a rapid influx of neutrophils that clear unless there is residence and a foreign body reaction [27, 34, 65]. Salt solutions produce an acute neutrophilic synovitis with synovial fluid cell counts to ~ 8000 cells/µL and have served as a comparative control, including for HA products [65], dental pulp tissue allografts [27], and amnion allografts [34]. Substances that contain proteins incite a greater neutrophilic reaction [27, 34]. Importantly, the attraction of neutrophils should be transient and not be associated with infiltration of the synovial membrane by neutrophils or macrophages, indicating a synovial reaction to a substance [28–30]. In combination, our work with single and multiple joint injections that included synovial fluid analyses and acute histopathology confirms that the neutrophils were passing through from the synovial vasculature to the fluid as a response to the perturbation in the joint fluid and not setting up residency in the synovial interstitium. In this early phase, synovial lining cells appeared healthy and normal without a histopathologic transition to macrophage phenotype or lymphocyte accumulation. Studies evaluating the joint response to injection with biologics, such as plasma and platelet products [12, 55, 56], hyaluronic acid [65], and Il-1ra gene therapy [59], have also reported a transient inflammation in the joint fluid. Many studies have not evaluated such an early phase as day 1 after joint injection or compared to saline. In a study with synovial fluid analyses, equine isolated mesenchymal stem cells did not increase synovial fluid cell counts over placebo controls or normal ranges; however, baseline values were not reported [66]. In our report on UC allograft, the joint fluid cellular inflammation had returned to normal after Day 5 and would have been missed in the other reports due to their later days of sampling.

It is recognized that the studies in this report are limited by the study durations and timelines, selection of specific outcome parameters that were not all inclusive, and the limited number of horses particularly for the diseased clinical study, such that the results are inconclusive in diseased joints or ligaments. Additional studies are underway to further evaluate these findings.

Conclusions

Umbilical cord allograft can be characterized as a cryopreserved, off-the-shelf, protein-rich suspension with a well-defined and consistent protein profile both among donors and within samples from a single donor. Many proteins were identified with a structural and protective function in connective tissue, supporting their use as a connective tissue allograft. Umbilical cord allografts may have the potential to supplement tissue repair by providing biologic scaffold proteins (fibrillar and gelatinous) to optimize the endogenous tissue environment. In concert, these data provide evidence to support a clinical trial in client-owned horses.

Abbreviations

BID

“Bis in die” or “twice a day”

C

Control

CV

Coefficient of variance

ELISA

Enzyme-linked immunosorbent assay

FDA

Food and Drug Administration

HA

Hyaluronic acid

IACUC

Institutional Animal Care and Use Committee

IFNg

Interferon gamma

Il-1a

Interleukin-1 alpha

Il-1ra

Interleukin-1 receptor antagonist

Il-2

Interleukin-2

Il-4

Interleukin-4

Il-8

Interleukin-8

Il-15

Interleukin-15

MCP-1

Monocyte chemoattractant protein-1

MS

Mass spectrometry

MSK

Musculoskeletal

PMN

Polymorphonuclear

OA

Osteoarthritis

SD

Standard deviation

SEM

Standard error of the mean

SSA

Serum amyloid A

TNCC

Total nucleated cell count

TSC

Total spectral count

UC

Umbilical cord

VEGF

Vascular endothelial growth factor

WJ

Wharton’s jelly

Authors' contributions

ALB made substantial contributions to the conception, design, and interpretation of the work, analyzed and interpreted data, collated data from the various contributors, and was the major contributor in writing and revising the manuscript. CR performed the safety clinical and case studies as a third party commercial research organization, including data collection and recording as per regulatory standards. GT performed mass spectrometry, acquired proteomics data, and contributed to the interpretation of the work. DR performed the histological examination of the joint and ligament tissues and contributed to the interpretation of the work BL performed the pathologic study and collected the speciments.

Funding

Equine performance labs funded the research and consultant contracts.

Data availability

All data generated or analyzed during this study are included in this published article; supplementary information may be available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Animal studies were performed in accordance with and by approval of Institutional Animal Care and Use Committees (ETCR-24-03; LSU-24-089). ARRIVE checklist completed.

Consent for publication

NA.

Competing interests

ALB was an external consultant for EPL charged to design and manage the research for this study as a third-party contractor due to her publishing and science background in regenerative medicine. She is not an employee and had no equity or other arrangement to benefit from any data outcome.