Recipient Organization
UNIVERSITY OF KENTUCKY
500 S LIMESTONE 109 KINKEAD HALL
LEXINGTON,KY 40526-0001
Performing Department
Veterinary Science
Non Technical Summary
Lesions in joint cartilage do not heal completely. This is a primary reason why osteoarthritis is a progressive disease in horsesand other mammals. For athletic horses, it is a major problem. They are elite athletes and during competition approach their physicallimits. Some tissue injury is inevitable, but if the injured tissue can repair following rest and therapy, the problem is less serious.Unfortunately, since articular cartilage has an extremely limited capacity for repair, injuries to the joint surface are very serious andfrequently become a major contributing variable limiting the ability of racehorses to compete successfully. The personal andeconomic consequences for horse owners and equine industries are substantial. Interestingly though, if we look outside of horsesand outside of mammals, there are some vertebrate animals fully capable of repairing even large structural lesions in joint cartilage.We have identified a salamander species that has an amazing ability to fully repair articular cartilage. These animals retain aspecific tissue into adulthood that for mammals is present only during a brief period of fetal limb development. We believe thatcells in this tissue may enable and facilitate articular cartilage repair.Recently, we have been able to characterize the same tissue in early equine fetuses and successfully isolate andculture cells from this tissue. Support is requested to study this developmental tissue to determine if it can improve cell-basedtherapies to repair articular cartilage defects in the horse.
Animal Health Component
(N/A)
Research Effort Categories
Basic
100%
Applied
(N/A)
Developmental
(N/A)
Goals / Objectives
During vertebrate limb development, a morphologically distinct zone of cells in the prechondrogenic mesenchyme initiates synovial joint formation.7,8 This mesenchymal tissue is known as the "interzone" and appears as a flattened layer of cells connected by gap junctions.9 Interzone cells exist during early fetal development in all mammals including horses, but are present only transiently prior to joint space cavitation. In a developmental biology context, the interzone is the normal progenitor of all synovial joint tissues including articular cartilage. As detailed in the Previous Work subsection above, we have demonstrated using an amphibian model system that interzone tissue can also facilitate a remarkable repair of large articular cartilage defects and even generate an entirely new diarthrodial joint de novo. Recently, we have been able to characterize interzone tissue in early equine fetuses and successfully isolate primary horse interzone cells that can readily be expanded in culture. The hypothesis to be tested in this proposal is that equine interzone tissue has an enhanced biological potential relative to other cell types to differentiate into true articular chondrocytes. Support is requested to characterize unique features of the equine mRNA transcriptome in articular cartilage relative to other cartilaginous tissues and to study the differentiation potential of equine interzone tissue on a cellular and molecular level both in vitro and in vivo. Our objective is to further advance emergent cell-based therapeutic strategies that try to achieve the full and complete repair of articular cartilage following injury. Two specific aims are proposed:Aim 1 Identify the unique tissue-restricted mRNA transcripts and exon splice junctions of articular cartilage to define a "transcriptome fingerprint" of articular chondrocytes relative to other cartilaginous tissues.Gene expression profiles of different cell types provide insight into cell- and tissue-specific phenotype and function. Aim 1 will address an important knowledge gap by defining features of the mRNA transcriptome that differentiate chondrocytes in articular cartilage from chondrocytes that produce other types of cartilage. These tissue-restricted biomarkers will be identified through RNA sequencing.Aim 2 Determine if chondrogenic differentiation of equine interzone cells yield "chondrocytes" that have an articular cartilage phenotype.New cell-based strategies for equine orthopaedic injuries are generating a high level of interest. Unfortunately, fibrocartilage formation and structural anchoring of the repair tissue into the surrounding healthy tissue continue to be major problems. Aim 2 will compare "chondrocytes" generated from interzone cells to those that can be generated from other equine cell types currently being studied and even used clinically for joint therapies, including bone marrow mesenchymal stem cells (MSCs), fat-derived MSCs, and primary adult equine articular chondrocytes.
Project Methods
Specific Aim 1.Experimental SamplesAll of the samples required for Aim 1 are already in hand.Transcriptional Profiling At least 30 million paired-end and strand-specific 2x100bp mRNA reads have already been generated from each tissue sample. The reads are high quality (Figure 5) and will be aligned to the equine reference genome using MapSplice (http://www.netlab.uky.edu/p/bioinfo/MapSplice2), a leading software program for this purpose that was developed by Dr. Jinze Liu in collaboration with Dr. MacLeod (PI) and Dr. Jan Prins.10Data analysisRelative expression levels for the gene locus and individual splice junctions will be determined by direct quantitation of RNA-seq read coverage using the Pileup Format from SAMtools software (http://samtools.sourceforge.net). Equine gene structure annotation for the alignment of RNA-seq reads will be based on our analyses of Ensembl and NCBI in silico gene predictions refined by RNA-seq data.11 These gene structure annotations have recently been further improved through our collaborative efforts to generate a third assembly of the equine reference genome.12-14The expression level at each nucleotide coordinate within a sample will be normalized using the total number of aligned reads for that specific sample divided by the total number of aligned reads in all samples. For individual genes, the data will also be normalized by the gene's cDNA length. Gene loci and splice junctions will be categorized as tissue-specific when expression above background is present in only one of the tissue sample groups.Specific Aim 2.Equine interzone samplesInterzone tissue will be collected from the developing appendicular joints of fetal horses at 45-60 days of gestational age. Pregnancy termination will be achieved through an IACUC-approved non-surgical extraction procedure of the early fetus from sedated mares after prostaglandin-induced (misoprostal) relaxation of the cervix. Samples used for RNA isolation will be snap-frozen in liquid nitrogen until further processing. Samples used for immunohistochemistry will be fixed in 4% paraformaldehyde. For tissue culture experiments, cells will be enzymatically released from their extracellular matrix under sterile conditions using commercial protease cocktails that are optimized for primary cell isolation.Chondrocyte differentiationThe chondrogenic differentiation potential of equine interzone cells with comparisons to adult MSCs isolated from bone marrow and fat, as well as to culture-expanded primary adult chondrocytes will be assessed both in vitro and in vivo.Cells in 2- and 3-dimensional culture systems will be expanded in high-glucose DMEM (4.5 g/L) supplemented fetal bovine serum 30% (v/v), penicillin/streptomycin 1% (v/v), and 1 mg/ml amphotericin. For the in vitro induction of chondrogenic differentiation, the DMEM basal medium will be supplemented with recombinant BMP-2 (500 ng/ml, InductOs; Wyeth), TGF-β1 (10 ng/ml, R&D Systems), 0.1 μM dexamethasone, 1 mM sodium pyruvate, 0.1 mM ascorbic acid-2-phosphate, 1% ITS (insulin 25 μg/mL, transferrin 25 μg/mL, and sodium selenite 25 ng/mL), and 1.25 mg/mL human serum albumin (Octapharma).15 Morphological and gene expression analyses will be completed as detailed below on samples collected at weekly intervals to 12 weeks. To complete a comparative analysis of each cell type's tri-lineage differentiation potential, induction of osteogenic (assayed by positive Alizarin Red staining and biomarker genes like alkaline phosphatase) and adipogenic (assayed by positive Oil Red O staining, lipid droplet formation, and biomarker genes like adipocyte fatty acid-binding protein 2) differentiation will be applied using standard protocols.16,17 All cells will be maintained at 37°C, 95% air/5% CO2 in a humidified incubator with medium changes every 2 or 3 days.For in vivo analyses, tissue transplant experiments will be conducted in SCID mice on a NOD (non obese /diabetic) background (NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ) available commercially (http://jaxmice.jax.org/cancer/xenoHosts.html#005557). Subcutaneous, renal capsule, and skeletal transplant sites will be compared.RNA isolationTotal RNA for gene expression analyses will be isolated using variations of acid guanidinium thiocyanate/phenol/chloroform extraction with alcohol and salt precipitations routinely used in my laboratory.23Transcriptional Profiling Whole transcriptome level analyses will be performed through direct sequencing of mRNA (http://www.illumina.com/technology/paired_end_sequencing_assay.ilmn) using Illumina's paired-end mRNA-seq protocol. The protocol to be used for the preparation of cDNA libraries will be modified to generate strand-specific reads (http://www.illumina.com/products/truseq_stranded_total_rna_sample_prep_kit.ilmn). This is important in that it enables better resolution of overlapping transcripts. The analysis of mRNA transcriptome data will be done as described under Aim 1 above.RT/qPCRTo confirm quantitation of specific mRNA transcripts, real time, reverse transcription-qPCR (RT-qPCR) assays will be used with consideration of MIQE guidelines.24Histology and ImmunohistochemistryHematoxylin and Eosin (H&E) staining will be performed using established protocols.5,6 Safranin-O/Fast Green staining will be optimized for each sample type using aqueous 0.001% Fast Green and 0.05% Safranin-O solutions. Immunostaining will be performed using standard avidin biotin complex reagents and methods according to the manufacturer's protocol (Santa Cruz Biotechnology, Cat. #sc2017, sc2019, Santa Cruz, CA). Antigen retrieval methods, either 0.2M sodium citrate buffer, 1% pepsin in 10mM HCl, or another strategy, will be optimized for each primary antibody starting with recommendations in the published literature. Similar anecdotal methods will be used to optimize dilutions of primary and secondary antibodies. All sections will be counterstained with Gill No.3 hematoxylin (Sigma, Cat#HS316). Positive control samples will be identified from the literature and analyzed using the same hybridization reagents and protocols. Negative controls for each assay will include omission of the primary or secondary antibody from the immunostaining protocol.In situ hybridizationIn situ hybridizations will be performed to detect and localize the expression of transcripts from genes of interest.Statistical analysisDifferential expression of individual genes and splice sites will be compared using the Mixed Procedure for SAS (Statistical Analysis System, SAS Institute, Cary, NC, USA) as described in recent papers from the MacLeod laboratory.4,11 A pooled variance, two-tailed Student's t-test will be produced to compare expression between experimental groups. An unprotected Least Significant Difference test will be used to analyze pair-wise multiple comparisons of gene expression levels across different cartilaginous tissues.Gene and splice site expression patterns in cell lines before and after the induction of "chondrogenic" differentiation (from Aim 2) will also be compared relative to the biomarker profiles defined for articular cartilage and the other tissues (from Aim 1). It is anticipated that the panel will contain ~200 tissue-restricted transcripts or splice junctions. We will compute the Pearson correlations among the expression profiles. A dendrogram, which visualizes the closeness of the expression profiles will then be generated through hierarchical clustering.27,28 In this way, we will be able to determine which induced chondrocyte sample most closely resembles normal articular cartilage, as well as the relative distance among the other cell line samples.