- Email: [email protected]
Functionally compromised synovium-derived mesenchymal stem cells in Charcot neuroarthropathy
Accepted Manuscript Functionally compromised synovium-derived mesenchymal stem cells in charcot neuroarthropathy
Reed Mitchell, Jeremy Molligan, Sydney Rooney, Young Cho, Lew Schon, Zijun Zhang PII: DOI: Reference:
S0014-4800(17)30526-9 doi:10.1016/j.yexmp.2018.01.003 YEXMP 4111
To appear in:
Experimental and Molecular Pathology
Received date: Accepted date:
10 October 2017 10 January 2018
Please cite this article as: Reed Mitchell, Jeremy Molligan, Sydney Rooney, Young Cho, Lew Schon, Zijun Zhang , Functionally compromised synovium-derived mesenchymal stem cells in charcot neuroarthropathy. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Yexmp(2018), doi:10.1016/j.yexmp.2018.01.003
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Functionally Compromised Synovium-derived Mesenchymal Stem Cells in Charcot Neuroarthropathy
Reed Mitchell1 , Jeremy Molligan2 , Sydney Rooney3 , Young Cho1 , Lew Schon1 , Zijun Zhang1*
3
Thomas Jefferson University, Philadelphia, Pennsylvania
IP
2
T
Orthobiologic Laboratory, MedStar Union Memorial Hospital, Baltimore, Maryland
Vanderbilt University School of Medicine, Nashville, Tennessee
CR
1
AC
CE
PT
AN M
ED
Zijun Zhang Orthobiologic Lab MedStar Union Memorial Hospital 201 E. University Parkway Bauernschmidt Building, Room 763 Baltimore, MD 21218 Tel: 410 554 2830 Fax: 410 554 2289 Email: [email protected]
US
*Corresponding author:
ACCEPTED MANUSCRIPT Abstract Charcot neuroarthropathy (CNA) often presents as a diabetic foot complication. The role of synovial mesenchymal stem cells (syn-MSCs) in the pathogenesis of CNA is unclear. Synovial samples were collected, for isolation of syn-MSCs, from diabetic patients with CNA (n=7) and
T
non-diabetic patients with intra-articular fracture or normal joints (non-CNA; n=7) during foot
IP
surgery. The syn-MSCs in the CNA and non-CNA groups were characterized comparatively.
CR
The average number of colonies formed in the CNA group was 6 ± 3.5 per half plate (10 mm in diameter), while it was 43 ± 21.6 in the non-CNA group (p < 0.05). The average size (pixels) of
US
the colonies in the CNA group was smaller than that in the non-CNA group. When the colonies
AN
were stratified into high-, medium- and low-density subgroups, colonies in the high-density subgroup of the CNA group were reduced in density. Expression of PPAR-γ, RUNX2, SOX9
M
and type II collagen by syn-MSCs in the CNA group was decreased during adipogenic,
ED
osteogenic and chondrogenic differentiation as compared with the non-CNA group. In conclusion, syn-MSCs in CNA joints were reduced in number, with declined differentiation
CE
pathology of CNA.
PT
potentials. The high-density subpopulation of the syn-MSCs was particularly affected by the
AC
Keywords: mesenchymal stem cells; synovium; neuroarthropathy; diabetes
ACCEPTED MANUSCRIPT 1. Introduction Charcot neuroarthropathy (CNA, also known as neuropathic osteoarthropathy or Charcot joint) was first described by Jean-Martin Charcot in 1868 as a destructive joint condition resulting from neuropathic deficiency [Chisholm and Gilchrist, 2011]. In modern medicine,
T
peripheral neuropathy secondary to diabetes mellitus is the predominant neuropathic condition to
IP
associate with CNA [Rajbhandari el al., 2002]. As a late-stage diabetic complication, CNA
CR
commonly occurs in the foot (also known as Charcot foot). Clinically, CNA displays joint inflammation, with fragmentation of cartilage and periarticular bone, in an insensate foot
US
[Frykberg el al., 2006]. The consequent foot deformity, due to joint dislocation and fracture, in
AN
CNA not only impairs patients’ mobility but also causes foot ulceration and infection [Hastings et al., 2013; Albers and Pop-Busui, 2014]. CNA is difficult to manage and some cases are
M
eventually treated with lower extremity amputation [Frykberg el al., 2006].
ED
It is agreed by the medical community that CNA is often misdiagnosed and non-treated in diabetes care partially because the pathogenesis of CNA remains elusive [Trieb, 2016]. A
PT
hallmark of CNA pathology is that bone and cartilage fragments are engulfed by hyperplastic
CE
synovium [Soudry et al., 1986; Sanders, 2004]. Synovium, which forms the inner lining of the joint capsule, produces fluid to lubricate joint surfaces and nourish chondrocytes. In a state of
AC
“synovitis”, synovium is the driving force behind articular cartilage destruction in inflammatory arthritis, such as rheumatoid arthritis [Fox et al., 2010; Bhattaram and Chandrasekharan, 2016]. A recent study demonstrated that fibroblast- like synoviocytes (FLS) isolated from CNA synovium are more invasive and chondrolytic under inflammatory conditions in vitro [Molligan et al., 2016], suggesting the involvement of synovium in CNA pathogenesis. Also residing in synovium are mesenchymal stromal/stem cells (MSCs), which are
ACCEPTED MANUSCRIPT capable of chondrogenic, osteogenic and adipogenic differentiation [De Bari et al., 2001; Sakaguchi et al., 2005]. These synovium-derived MSCs (syn-MSCs) have unique biological features, which distinguish them from the MSCs of other tissue origins. They express a relatively high level of CD90 [Djouad et al., 2005; Hagmann et al., 2016; Klimczak and Kozlowska, 2016],
T
and are more proliferative and superior in chondrogenesis [Sakaguchi et al., 2005; Yoshimura et
IP
al., 2007]. Joint injury often triggers syn-MSCs to proliferate and differentiate [Kurth et al.,
CR
2011; Fülber et al., 2016].
In addition to tissue homeostasis, syn-MSCs also involve in joint pathology. When co-
US
cultured with chondrocytes, syn-MSCs were able to inhibit their reaction to inflammatory stimuli
AN
[Ryu et al., 2014]. Syn-MSCs also participate in regulation of the proliferation of T cells in inflammatory joints [Eljaafari et al., 2012]. On the other hands, the pathological condition in a
M
joint could affect the functionality of syn-MSCs. In rheumatoid arthritis, syn-MSCs were
ED
suppressed in proliferation and chondrogenic differentiation [Jones et al., 2010]. In osteoarthritis, the chondrogenesis of syn-MSCs was inhibited by the increase of M1-phase macrophages in the
PT
synovium [Fahy et al., 2014]. Synovium is an important part of the CNA pathology [Soudry et
CE
al., 1986; Molligan et al., 2016]. The characteristics of syn-MSCs in CNA and their potential role in CNA pathology, however, are unknown.
AC
In this study, synovial samples were collected from the CNA joints during surgery for isolation of syn-MSCs. The syn-MSCs from the CNA joints were evaluated for clonagenicity, differentiation capacity and proliferation, in comparison with syn-MSCs from non-CNA joints.
2. Materials and Methods Synovial samples were collected from mid-foot, hind-foot and ankle joints during surgery
ACCEPTED MANUSCRIPT for foot reconstruction (approved by MedStar Health Research Institute Institutional Review Board). According to donors’ clinical conditions, the samples were organized in two groups: 1) the CNA group (n = 7) included 4 males and 3 females with an average age of 54.4 years. The clinical and imaging manifestation of CNA varies by the stages of the pathology. A diagnosis of
T
CNA in this study was based on a) foot deformity or swelling; b) loss of sensation or numbness;
IP
c) joint dislocation or fracture without significant injury (Fig 1); d) a medical history of diabetes.
CR
2) The non-CNA group (n = 7), as the controls, included 3 males and 4 females with an average age of 52.6 years. Synovium was collected from either normal or intra-articular- fractured foot
US
joints. None of these patients was diabetic. Small portions of the collected synovium in both
AN
CNA and non-CNA groups were fixed with 4% paraformaldehyde and embedded in O. C. T. compound (Sakura Finetek USA, Torrance, CA). Tissue sections (5 µm) were cut with a cryostat
Isolation of syn-MSCs:
ED
2.1.
M
and stained with hematoxylin and eosin (H&E).
Synovial samples of the CNA and non-CNA groups were washed and kept in Dulbecco's
PT
modified Eagle's medium (DMEM, Life Technologies, Long Island, NY) until processed. They
CE
were minced with scalpels and placed in 0.1% collagenase (Type I, Sigma-Aldrich Co., St. Louis, MO) in DMEM for 4 hours for digestion. The digest was filtered through a 70-μm cell
AC
strainer. After washing with phosphate buffered saline (PBS), cells were counted and cultured in DMEM supplemented with 10% fetal bovine serum (FBS, Rocky Mountain Biologics, Missoula, MT), 1% penicillin/streptomycin and 200mM L-glutamine (Life Technologies), at 37ºC with 5% CO2 in air. Culture medium was changed every 3 days. Cells were passaged with trypsin at 80% confluence. 2.2.
Colony forming units-fibroblast assay and colony characterization:
ACCEPTED MANUSCRIPT After one passage, cells from 4 CNA donors and 4 non-CNA donors were plated on 10cm tissue culture dishes at a density of 3,000 cells per cm2 . Cells from each donor were plated in triplicate. After two weeks, cultures (n = 24) were fixed in cold 100% ethanol and stained with crystal violet. The dishes were imaged, except for one in which staining was not well developed.
T
To analyze the colonies, all the images were converted to 16-bit images and adjusted to a
IP
standard threshold (8,250). A macro program was written in ImageJ (National Institutes of
CR
Health, Bethesda, MD) that took these processed images and created a map of the colonies, automatically tracing and numbering the colonies using “Analyze Particles” function. The first
US
trials of image processing showed a glare on the right side of each plate due to lighting that
AN
interfered with the program. Thus, the script was rewritten to input only colonies taken from the left semicircle of the plate. A colony (> 50 cells) was defined on image as > 2,000 pixel2 . The
M
number of colonies on each image was recorded. Following the circling of an individual colony,
ED
the Analyze Particles function was used to obtain the “% area” output. This gave the percentage of the area in a colony that had been turned to black in the binary process, as area (pixel2 ) or
PT
density (greyscale) of the colony.
CE
For data analysis, all the individual colonies of the CNA and non-CNA groups were inputted into the R programming language environment (The R Foundation for Statistical
AC
Computing, Vienna, Austria), with the input containing each colony’s size and corresponding density. Using R, the colonies were sorted by size and then stratified by size: for both the CNA and non-CNA groups, a list of colonies was generated ordering by size and the lists were divided into thirds of small, medium and large. Similarly, the colonies in the CNA and non-CNA groups were sorted and ordered by density. According to their density, the colonies of the CNA and nonCNA groups were divided into thirds: low, medium and high, for comparison.
ACCEPTED MANUSCRIPT 2.3.
Flow cytometry of common MSC markers:
Syn-MSCs from both CNA and non-CNA groups, at passage 3, were stained with propidium iodide (PI) for exclusion of dead cells. Cells were then incubated separately with phycoerythrin (PE) or fluorescein isothiocyanate (FITC) conjugated mouse anti-human CD14,
T
CD73, CD90, and CD105 antibody (BD Biosciences, San Jose, CA) at an amount of 0.01μg per
IP
100,000 cells for 30 minutes at room temperature. Fluorochrome conjugated normal mouse IgG
CR
was used as isotype controls. Each sample was prepared in triplicate. The labeled syn-MSCs were analyzed on a flow cytometer (Acurri C6, BD Biosciences). The percentage of positively
US
stained cells for each antigen was calculated using Overton histogram subtraction with FCS
2.4.
Proliferation of syn-MSCs:
AN
Express Analysis Suite (De Novo Software, Glendale, CA).
M
Syn-MSCs isolated from CNA or non-CNA synovial samples, at passage 3-5, were
ED
seeded in 24-well plates at a density of 1000 cells/well. The cells were collected with trypsin and counted with a hemocytometer on days 3, 5, 7, 9, 11, and 13. Proliferation was estimated by
PT
calculating the cell population doubling time.
CE
2.5. Differentiation of syn-MSCs: Syn-MSCs isolated from the CNA and non-CNA synovial samples, after passage 4, were
AC
used for tri-lineage differentiation under the following conditions: 1) for osteogenic differentiation, cells were seeded into T-25 culture flasks at a density of 1x104 /cm2 . Osteogenic media consisted of DMEM, 10% FBS, 10 mM β-glycerophosphate, 100 nM dexamethasone, 50µg/ml L-ascorbic acid 2-phosphate, and 100 ng/mL human recombinant bone morphogenetic protein 2 (BMP-2, Medtronik Sofamer Denek, Memphis, TN). 2) For adipogenic differentiation, syn-MSCs were cultured as above-mentioned but in adipogenic media, which was DMEM
ACCEPTED MANUSCRIPT supplemented with 10% FBS, 10 uM rosiglitazone, 1 uM dexamethasone, and 10 ng/ml bovine insulin [Contador et al., 2015]. 3) Chondrogenic differentiation of syn-MSCs was conducted in pellet culture (3x105 cells/pellet). The pellets were cultured in chondrogenic media that was DMEM supplemented with 10% FBS, 50μ/ml L-ascorbic acid 2-phosphate, 40ug/ml L-proline,
T
0.1 μM dexamethasone and 10 ng/ml recombinant human transforming growth factor β1
IP
(Peprotech, Rocky Hill, NJ). For enhanced chondrogenesis, recombinant human bone
CR
morphogenetic protein-7 (100 ng/ml; Peprotech) was added into the chondrogenic media at days
US
3 and 12 of tissue culture, following a published protocol [Handorf and Li, 2014]. Tissue cultures of osteogenic, adipogenic and chondrogenic differentiation were maintained for three weeks. The
AN
controls of the tri-lineage differentiation were the same syn-MSCs cultured in regular growth media.
M
At the end of chondrogenic, adipogenic and osteogenic differentiation, syn-MSCs in
ED
differentiation cultures and the corresponding controls were lysed with TRIzol (Life Technologies) and RNA was extracted by phase separation. First-strand cDNA synthesis was
PT
performed using the iScript Reverse Transcription Supermix (Bio-Rad Laboratories, Hercules,
CE
CA) with 100 ng of RNA template. For quantitative gene expression, first-strand cDNA was amplified using the SSO Advanced SYBR Green Supermix (Bio-Rad Laboratories) on a CFX-
AC
Connect Real-Time PCR System (Bio-Rad Laboratories). After initial denaturation at 95ºC for 30 seconds, polymerase chain reaction (PCR) was performed with 40 cycles consisting of 10 seconds at 95ºC and 30 seconds at 53-59ºC (depending on the primer pair). An efficiency of >90% was required for each primer pair and annealing temperature. The specificity of each primer pair was verified using a melting curve from 65-95ºC and agarose gel electrophoresis of the PCR product. Genes investigated were type II collagen and Sox9 for chondrogenic
ACCEPTED MANUSCRIPT differentiation, RUNX2 for osteogenic differentiation and peroxisome proliferator-activated receptor gamma (PPAR-γ) for adipogenic differentiation. The expression of glyceraldehyde 3phosphate dehydrogenase (GAPDH) was used as an internal reference. Details of primers (Integrated DNA Technologies, Coralville, IA) are listed in Table 1. Gene expression was
T
expressed as fold-change from undifferentiated syn-MSCs corresponding to each sample using
IP
the ∆∆CT method (User Manual, Bio-Rad Laboratories).
CR
Adipogenic and osteogenic cultures of syn-MSCs isolated from CNA and non-CNA synovial samples, under the above-mentioned conditions, were also conducted in 24-well plates.
US
After three weeks of culture, the cells were fixed with 4% paraformaldehyde. Neutral lipid
AN
droplets in adipogenic cultures were stained with Oil Red-o in polyethylene glycol and matrix mineralization in osteogenic cultures was visualized with Alizarin red staining. To analyze
M
chondrogenic differentiation, frozen sections of chondrogenic pellet cultures of syn-MSCs were
ED
stained with Alcian blue.
Statistical analysis: Data are presented as mean ± standard deviation. The average
PT
number, size and density of colonies were compared between the CNA and non-CNA groups
CE
with t test. The stratified colonies in both CNA and non-CNA groups were compared separately by the averaged area of small-, medium- and large-size colonies, and the averaged greyscale of
AC
low-, medium- and high-density colonies, using unpaired t test. In the proliferation study, the cell numbers of syn-MSCs collected at the same time points were compared between the CNA and non-CNA groups with t test. During the differentiation of syn-MSCs, the expression of genes related to chondrogenesis, adipogenesis and osteogenesis in fold-change in the CNA and nonCNA groups was evaluated with t test. P < 0.05 was set as significant.
ACCEPTED MANUSCRIPT 3. Results 3.1.
Histology of synovium: In the CNA group, hyperplasia of synovium included
multiple layers of cells in the intimal layer (Fig 2A). There were only one or two layers of lining cells in the non-CNA synovium (Fig 2B). There was no significant infiltration of leukocytes in
Colony formation and characterization.
IP
3.2.
T
the subintimal area in both groups.
CR
After two weeks in culture, syn-MSCs in both CNA and non-CNA groups formed colonies with varied size and shape (Fig 3A and B). The average number of colonies in the CNA
US
group was 6 ± 3.5 per half plate, while it was 43 ± 21.6 in the non-CNA group (p < 0.05; Fig
AN
3C). The average colony size (pixels) in the CNA group was smaller than that in the non-CNA group (4780±1232 vs. 7960±1687; p < 0.05) but the average colony densities of the CNA and
M
non-CNA groups were about equal (Fig 3D and E). Based on their sizes, the colonies in both
ED
CNA and non-CNA groups were stratified into thirds, i.e. the small, medium and large subgroups. In all three subcategories, colonies in the CNA group were smaller than those in the
PT
non-CNA group (p <0.05; Fig 3F). When the colonies were similarly stratified into thirds
CE
according to density, the average colony densities in the low- and medium-density subgroups were not significantly different between the CNA and non-CNA groups (Fig 3G). In the high-
AC
density subcategory, however, the CNA group had a lower colony density than the non-CNA group (30±6 vs. 35±10; p < 0.05). 3.3.
Cell surface markers expressed by syn-MSCs.
Syn-MSCs in both CNA and non-CNA groups expressed the common MSC cell surface markers comparably. Syn-MSCs in the CNA group were largely positive for CD90 (99.0%±0.5%), CD73 (98.0%±0.9%) and CD105 (98.6%±0.4%), but negative for CD14
ACCEPTED MANUSCRIPT (16.9%±1.6; Fig 4). 3.4.
Proliferation of syn-MSCs.
The population doubling time of the syn-MSCs was 1.8 days in the CNA group and 1.6 days in the non-CNA group. The average numbers of syn-MSCs sampled at days 3, 5, 7, 11 were
T
not significantly different between the CNA and non-CNA groups (p > 0.05). By day 13, the
group (p < 0.05; Fig 5). 3.5.
Tri-lineage differentiation of syn-MSCs.
CR
IP
number of syn-MSCs in the non-CNA group was significantly greater than that in the CNA
US
After culture in adipogenic media, neutral lipid droplets were developed in the syn-MSCs
AN
of both CNA and non-CNA groups but they were consistently more prevalent in the non-CNA group (Fig 6A). In osteogenic culture, syn-MSCs from the non-CNA group exhibited more
M
extensive matrix mineralization than those in the CNA group. In chondrogenic pellet culture, the
ED
CNA group displayed reduced proteoglycans in cartilage matrix and cell density as compared with the non-CNA group.
PT
During adipogenic differentiation, PPAR-γ expression by syn-MSCs in the non-CNA
CE
group increased 8-fold while it increased only about 3-fold in the CNA group (p<0.05; Fig 6B). Syn-MSCs subjected to osteogenic differentiation increased in RUNX2 expression by 10-fold in
AC
the non-CNA group and 2-fold in the CNA group (p<0.05), respectively. In chondrogenic differentiation, the expression of Sox9 by syn-MSCs increased by 37-fold in the non-CNA group and by 15-fold in the CNA group (p<0.05). The increased expression of type II collagen by chondrogenic syn-MSCs in the non-CNA group was 3 times of that in the CNA group (p<0.05).
4. Discussion
ACCEPTED MANUSCRIPT The isolated syn-MSCs in both CNA and non-CNA groups expressed the common markers for MSCs: CD14-, CD73+, CD90+, and CD105+ and were capable of tri-lineage differentiation [Dominici et al., 2006]. The syn-MSCs in the CNA group, however, presented changes in quantity, clonogenicity, proliferation and differentiation. The quantity of MSCs in the
T
tissue has been investigated as an indication of healthy conditions [Stolzing et al., 2008]. Syn-
IP
MSCs of both CNA and non-CNA groups formed colonies simultaneously in low-density
CR
cultures but the CNA group produced far fewer colonies than the non-CNA group, indicating a reduced quantity of syn-MSCs in the CNA synovium. It is not uncommon that the number of
US
syn-MSCs is reduced in arthritic conditions. In rheumatoid arthritis, the number of syn-MSCs is
AN
reversely correlated with the severity of joint inflammation [Jones et al., 2010]. The pathology of CNA includes extensive synovial hyperplasia [Soudry et al., 1986]. The significant “synovitis”
M
in CNA could be partially responsible for a reduced number of syn-MSCs. It is noteworthy that
ED
not only the degree but also the stage of the joint inflammation affects the quantity of syn-MSCs. For example, the number of syn-MSCs was increased in acute arthritis induced by carrageenan
PT
injection [Matsukura et al., 2015]. CNA pathology evolves over an extended period clinically, as
CE
the neuropathy progresses, and is generally more chronic. In CNA, disorders of neuroendocrine regulation could also influence the fate of syn-MSCs. In CNA synovium, the number of
AC
sympathetic nerve fibers (labeled with tyrosine hydroxylase) is reduced [Koeck et al., 2009]. The similar neuropathy of sympathetic nerves has been found to be causative of apoptosis of bone marrow MSCs [Arranz et al., 2014]. Syn-MSCs in both CNA and non-CNA groups formed colonies with varied size and density. These colony features are indicative of the proliferation and differentiation potentials of the original MSCs and could be affected by pathological conditions [Mareddy et al., 2007]. The
ACCEPTED MANUSCRIPT colony size of the CNA group, overall and in the three-size subcategories, was smaller than that of the non-CNA group. This suggests that syn-MSCs in the CNA group were less proliferative than that in the non-CNA group, since the size of a colony is largely determined by the proliferation of MSCs [Stolzing et al., 2010]. This was confirmed when syn-MSCs in both CNA
T
and non-CNA groups were plated in parallel for proliferation assay, where the syn-MSCs in the
IP
CNA group grew at a slower rate. A diabetic history of the CNA patients in this study could have
CR
contributed to slow proliferation of the syn-MSCs in the CNA group because bone marrow MSCs showed significant deficiency of proliferation in diabetes [Shin and Peterson, 2012].
US
Colony density is more relevant to the spreading of the proliferating MSCs [Sengers et
AN
al., 2010]. While the average colony densities of the CNA and non-CNA groups were similar, analysis of the stratified colonies revealed a reduced colony density in the high-density
M
subcategory by the CNA group. In the high-density subcategory, syn-MSCs were more
ED
aggregated and less dispersed than those in the low- and medium-density subcategories. Subpopulations of MSCs often distinguish each other with differential expression of cytoskeletal
PT
and structural proteins [Mareddy et al., 2009], which are critical elements of cell mobility or
CE
spreading. Colony density is also influenced by the survival and adaption of the initial MSCs [Barbaric et al., 2014]. It is interesting that the subpopulation of syn-MSCs forming high-density
AC
colonies was selectively affected by the pathology of CNA and warrant further investigation on their specific implication on CNA pathology. Differentiation of the syn-MSCs in CNA was compromised in general. Chondrogenesis of syn-MSCs in the CNA group was suppressed, as demonstrated by gene expression of Sox9 and type II collagen, and Alcian blue staining of proteoglycans in extracellular matrix. Chondrogenesis of syn-MSCs is similarly inhibited in rheumatoid arthritis but the expression of
ACCEPTED MANUSCRIPT Sox9 is unchanged [Jones et al., 2010]. CNA and rheumatoid arthritis may interfere with the chondrogenic differentiation of syn-MSCs through distinct pathways. Adipogenic and osteogenic differentiation of syn-MSCs was inhibited in CNA but it was not the case in rheumatoid arthritis [Jones et al., 2010]. The suppression of tri-lineage differentiation of syn-MSCs in CNA is likely
T
a result of interplay of neuropathy, inflammation and diabetes. Neuropathy alone impacts the
IP
function of MSCs. For example, a reduction of sympathetic nerve fibers enhances osteogenic
CR
differentiation of bone marrow MSCs [Hanoun et al., 2014]. Although detailed neuropathic conditions in the CNA synovium remain to be revealed, it is clear that the neuro-signaling
US
network in synovium varies in different types of arthritic joints [Dirmeier et al., 2008]. In this
AN
study, all the synovium donors of CNA group were diabetic. Hyperglycemia promotes senescence and severely inhibits proliferation and differentiation of MSCs [Chang et al., 2015].
M
It is possible that the CNA group had more syn-MSCs entered senescence than the non-CNA
ED
group, which resulted in decreased clonogenicity and differentiation [Alves et al., 2010]. In conclusion, syn-MSCs in CNA are reduced in number and inhibited in proliferation
PT
and differentiation to chondrogenic, adipogenic, and osteogenic lineages. The altered biological
CE
properties of syn-MSCs are a part of the pathology of CNA.
AC
Acknowledgement: the authors declare no conflicts of interest related to this manuscript.
ACCEPTED MANUSCRIPT References 1. Albers JW, Pop-Busui R. 2014. Diabetic neuropathy: mechanisms, emerging treatments, and subtypes. Curr Neurol Neurosci Rep. 14(8):473. 2. Alves H, Munoz-Najar U, De Wit J et al. 2010. A link between the accumulation of DNA
T
damage and loss of multi-potency of human mesenchymal stromal cells. J Cell Mol Med.
IP
14(12):2729-2738.
CR
3. Arranz L, Sánchez-Aguilera A, Martín-Pérez D et al. 2014. Neuropathy of haematopoietic stem cell niche is essential for myeloproliferative neoplasms. Nature.
US
512(7512):78-81.
AN
4. Barbaric I, Biga V, Gokhale PJ et al. 2014. Time-lapse analysis of human embryonic stem cells reveals multiple bottlenecks restricting colony formation and their relief upon
M
culture adaptation. Stem Cell Reports. 3(1):142-155.
ED
5. Bhattaram P, Chandrasekharan U. 2016. The joint synovium: A critical determinant of articular cartilage fate in inflammatory joint diseases. Semin Cell Dev Biol. S1084-
PT
9521(16)30135-5.
CE
6. Chang TC, Hsu MF, Wu KK. 2015. High glucose induces bone marrow-derived mesenchymal stem cell senescence by upregulating autophagy. PLoS One.
AC
10(5):e0126537.
7. Chisholm, K.A., Gilchrist, J.M. 2011. The Charcot joint: a modern neurologic perspective. J Clin Neuromuscul Dis 13(1):1-13. 8. Contador D, Ezquer F, Espinosa M et al. 2015. Dexamethasone and rosiglitazone are sufficient and necessary for producing functional adipocytes from mesenchymal stem cells. Exp Biol Med (Maywood). 240(9):1235-1246.
ACCEPTED MANUSCRIPT 9. De Bari C, Dell'Accio F, Tylzanowski P et al. 2001. Multipotent mesenchymal stem cells from adult human synovial membrane. Arthritis Rheum. 44(8):1928-1942. 10. Dirmeier M, Capellino S, Schubert T et al. 2008. Lower density of synovial nerve fibres positive for calcitonin gene-related peptide relative to substance P in rheumatoid arthritis
T
but not in osteoarthritis. Rheumatology (Oxford). 47(1):36-40.
IP
11. Djouad F, Bony C, Häupl T et al. 2005. Transcriptional profiles discriminate bone
CR
marrow-derived and synovium-derived mesenchymal stem cells. Arthritis Res Ther. 7(6):R1304-315.
US
12. Dominici M, Le Blanc K, Mueller I et al. 2006. Minimal criteria for defining multipotent
AN
mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy. 8(4):315-317.
M
13. Eljaafari A, Tartelin ML, Aissaoui H et al. 2012. Bone marrow-derived and synovium-
ED
derived mesenchymal cells promote Th17 cell expansion and activation through caspase
64(7):2147-2157.
PT
1 activation: contribution to the chronicity of rheumatoid arthritis. Arthritis Rheum.
CE
14. Fahy N, de Vries-van Melle ML, Lehmann J et al. 2014. Human osteoarthritic synovium impacts chondrogenic differentiation of mesenchymal stem cells via macrophage
AC
polarisation state. Osteoarthritis Cartilage. 22(8):1167-1175. 15. Fox DA, Gizinski A, Morgan R et al. 2010. Cell-cell interactions in rheumatoid arthritis synovium. Rheum Dis Clin North Am. 36(2):311-323. 16. Frykberg RG, Zgonis T, Armstrong DG et al. 2006. Diabetic foot disorders. A clinical practice guideline (2006 revision). J Foot Ankle Surg. 45(5 Suppl):S1-66. 17. Fülber J, Maria DA, da Silva LC et al. 2016. Comparative study of equine mesenchymal
ACCEPTED MANUSCRIPT stem cells from healthy and injured synovial tissues: an in vitro assessment. Stem Cell Res Ther. 7:35. 18. Hagmann S, Rimmele C, Bucur F et al. 2016. Mesenchymal stromal cells from osteoarthritic synovium are a distinct population compared to their bone-marrow
T
counterparts regarding surface marker distribution and immunomodulation of allogeneic
IP
CD4+ T-Cell cultures. Stem Cells Int. 6579463.
CR
19. Handorf AM, Li WJ. 2014. Induction of mesenchymal stem cell chondrogenesis through sequential administration of growth factors within specific temporal windows. J Cell
US
Physiol. 229(2):162-171.
AN
20. Hanoun M, Zhang D, Mizoguchi T et al. 2014. Acute myelogenous leukemia- induced sympathetic neuropathy promotes malignancy in an altered hematopoietic stem cell
M
niche. Cell Stem Cell. 15(3):365-375.
ED
21. Hastings MK, Johnson JE, Strube MJ et al. 2013. Progression of foot deformity in Charcot neuropathic osteoarthropathy. J Bone Joint Surg Am. 95(13):1206-1213.
PT
22. Klimczak A, Kozlowska U. 2016. Mesenchymal stromal cells and tissue-specific
CE
progenitor cells: their role in tissue homeostasis. Stem Cells Int. 2016:4285215. 23. Jones E, Churchman SM, English A et al. 2010. Mesenchymal stem cells in rheumatoid
AC
synovium: enumeration and functional assessment in relation to synovial inflammation level. Ann Rheum Dis. 69(2):450-457. 24. Koeck FX, Bobrik V, Fassold A et al. 2009. Marked loss of sympathetic nerve fibers in chronic Charcot foot of diabetic origin compared to ankle joint osteoarthritis. J Orthop Res. 27(6):736-741. 25. Kurth TB, Dell'accio F, Crouch V et al. 2011. Functional mesenchymal stem cell niches
ACCEPTED MANUSCRIPT in adult mouse knee joint synovium in vivo. Arthritis Rheum. 63(5):1289-1300. 26. Mareddy S, Crawford R, Brooke G et al. 2007. Clonal isolation and characterization of bone marrow stromal cells from patients with osteoarthritis. Tissue Eng. 13(4):819-829. 27. Mareddy S, Broadbent J, Crawford R et al. 2009. Proteomic profiling of distinct clonal
T
populations of bone marrow mesenchymal stem cells. J Cell Biochem. 106:776-786.
IP
28. Matsukura Y, Muneta T, Tsuji K et al. 2015. Mouse synovial mesenchymal stem cells
CR
increase in yield with knee inflammation. J Orthop Res. 33(2):246-253. 29. Molligan J, Barr C, Mitchell R et al. 2016. Pathological role of fibroblast-like
US
synoviocytes in charcot neuroarthropathy. J Orthop Res. 34(2):224-230.
AN
30. Rajbhandari SM, Jenkins RC, Davies C et al. 2002. Charcot neuroarthropathy in diabetes mellitus. Diabetologia. 45(8):1085-1096.
M
31. Ryu JS, Jung YH, Cho MY et al. 2014. Co-culture with human synovium-derived
ED
mesenchymal stem cells inhibits inflammatory activity and increases cell proliferation of
447(4):715-720.
PT
sodium nitroprusside-stimulated chondrocytes. Biochem Biophys Res Commun.
CE
32. Sakaguchi Y, Sekiya I, Yagishita K et al. 2005. Comparison of human stem cells derived from various mesenchymal tissues: superiority of synovium as a cell source. Arthritis
AC
Rheum. 52(8):2521-2529. 33. Sanders LJ. 2004. The Charcot foot: historical perspective 1827-2003. Diabetes Metab Res Rev. 20 Suppl 1:S4-8. 34. Sengers BG, Dawson JI, Oreffo RO. 2010. Characterisation of human bone marrow stromal cell heterogeneity for skeletal regeneration strategies using a two-stage colony assay and computational modelling. Bone. 46(2):496-503.
ACCEPTED MANUSCRIPT 35. Shin L, Peterson DA. 2012. Impaired therapeutic capacity of autologous stem cells in a model of type 2 diabetes. Stem Cells Transl Med. 1(2):125-135. 36. Soudry M, Binazzi R, Johanson NA et al. 1986. Total knee arthroplasty in Charcot and Charcot-like joints. Clin Orthop Relat Res. 208:199-204.
T
37. Stolzing A, Jones E, McGonagle D et al. 2008. Age-related changes in human bone
IP
marrow-derived mesenchymal stem cells: consequences for cell therapies. Mech Ageing
CR
Dev. 129(3):163-173.
38. Stolzing A, Sellers D, Llewelyn O et al. 2010. Diabetes induced changes in rat
US
mesenchymal stem cells. Cells Tissues Organs. 191(6):453-465.
AN
39. Trieb K. 2016. The Charcot foot: pathophysiology, diagnosis and classification. Bone Joint J. 98-B(9):1155-1159.
M
40. Yoshimura H, Muneta T, Nimura A et al. 2007. Comparison of rat mesenchymal stem
ED
cells derived from bone marrow, synovium, periosteum, adipose tissue, and muscle. Cell
AC
CE
PT
Tissue Res. 327(3):449-462.
ACCEPTED MANUSCRIPT
NM_000346.3
COL2A
NM_001844.4
PPAR-γ
AB565476
RUNX2
NM_001024630.3
Reverse Primer Sequence
5' GAC AGT CAG CCG CAT CTT CT 3' 5' AGG AAG TCG GTG AAG AAC GG 3' 5' GCC TGG TGT CAT GGG TTT 3' 5' CCT AAA CTT CGG ATC CCT CC 3'
5' GCG CCC AAT ACG ACC AAA TC 3' 5' CGC CTT GAA GAT GGC GTT G 3' 5' GTC CCT TCT CAC CAG CTT TG 3' 5' TCA AAT CTG GTG TCG TTT GC 3'
5' CTA GTT TGT TCT CTG ACC GC 3'
5' TGG GGT CTG TAA TCT GAC TC 3'
AN M ED PT CE AC
T
SOX9
Forward Primer Sequence
CR
NCBI Accession Number NM_002046.5
US
Gene Name GAPDH
IP
Table 1: A list of primers
ACCEPTED MANUSCRIPT
Figure Legends Figure 1. A radiogram of CNA in a foot. The diabetic patient’s right foot has irregular joint space (arrows) in multiple joints around navicular in the mid-foot, and bony fragmentation of navicular
IP
T
and talar head. It is also notable of the collapse of foot arch.
CR
Figure 2. Histology of CNA and non-CNA synovium. Synovium in the CNA group is more hyperplasic, with an expanded intimal layer (A; arrows). Its subintimal layer is more fibrous,
US
compared with the non-CNA synovium (B). The intimal layer in the non-CNA synovium
AN
consists of 1-2 layers of cells (H&E staining; bar = 100µm).
M
Figure 3. Analyses of colonies formed by syn-MSCs. Colonies are formed by syn-MSCs in both
ED
CNA (A) and non-CNA (B) groups after two weeks of culture. While colonies spread on the non-CNA plate, only a few colonies scatter on the CNA plate (crystal violet staining). On
PT
average, the number of colonies formed by the syn-MSCs in the CNA group is far fewer than
CE
that in the non-CNA group (C). The average size of the colonies in the CNA group is smaller than that in the non-CNA group (D). The average colony density, however, is not different
AC
between the two groups (E). When colonies of the CNA and non-CNA groups are stratified into thirds by the size: small, medium and large, the average colony size of each subgroup of the CNA group is smaller than the corresponding subgroup in the non-CNA group (F). When colonies in the CNA and non-CNA groups are stratified into thirds by density, colony density of the high-density subgroup in the CNA group is lower than the corresponding one in the nonCNA group (G). Colony densities of the low- and medium-density subgroups are not different between the CNA and non-CNA groups (note: * indicates p < 0.05; ** indicates p < 0.001).
ACCEPTED MANUSCRIPT
Figure 4. Flow cytometry for syn-MSCs. Syn-MSCs in the CNA group (top row) express nearly negatively for CD14, but strongly for CD73, CD90 and CD105. The expression profile of the tested cell surface markers is not significantly different between the CNA and non-CNA groups
IP
T
(note: in the representative images, grey line is for isotope control and red line for syn-MSCs).
CR
Figure 5. Proliferation of syn-MSCs. Syn-MSCs from both CNA and non-CNA groups were cultured for 13 days and sampled every other day. The cell number of the non-CNA group
US
increases at a greater pace and becomes significantly greater than the CNA group at day 13
AN
(note: * indicates p < 0.05).
M
Figure 6. Tri-lineage differentiation of syn-MSCs. A: In adipogenic differentiation, formation of
ED
fat droplets is more proliferous in the non-CNA group than the CNA group (Oil Red O staining;). In osteogenic differentiation, matrix mineralization is more intense in the non-CNA
PT
group than in the CNA group (Alizarin red staining). The tissue section of chondrogenic pellets
CE
in the non-CNA group is intensely stained for cartilageous matrix (Alcian blue staining; bar = 50µm). B: Quantitative gene expression during tri-lineage differentiation of syn-MSCs. The
AC
expression of PPAR-γ, Runx2, Sox9 and type II collagen is significantly increased in the non-CNA group that that in the CNA group (note: * indicates p < 0.05).
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
Fig. 1
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
Fig. 2
ED
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Fig. 3
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
Fig. 4
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
Fig. 5
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
Fig. 6
Reed Mitchell, Jeremy Molligan, Sydney Rooney, Young Cho, Lew Schon, Zijun Zhang PII: DOI: Reference:
S0014-4800(17)30526-9 doi:10.1016/j.yexmp.2018.01.003 YEXMP 4111
To appear in:
Experimental and Molecular Pathology
Received date: Accepted date:
10 October 2017 10 January 2018
Please cite this article as: Reed Mitchell, Jeremy Molligan, Sydney Rooney, Young Cho, Lew Schon, Zijun Zhang , Functionally compromised synovium-derived mesenchymal stem cells in charcot neuroarthropathy. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Yexmp(2018), doi:10.1016/j.yexmp.2018.01.003
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Functionally Compromised Synovium-derived Mesenchymal Stem Cells in Charcot Neuroarthropathy
Reed Mitchell1 , Jeremy Molligan2 , Sydney Rooney3 , Young Cho1 , Lew Schon1 , Zijun Zhang1*
3
Thomas Jefferson University, Philadelphia, Pennsylvania
IP
2
T
Orthobiologic Laboratory, MedStar Union Memorial Hospital, Baltimore, Maryland
Vanderbilt University School of Medicine, Nashville, Tennessee
CR
1
AC
CE
PT
AN M
ED
Zijun Zhang Orthobiologic Lab MedStar Union Memorial Hospital 201 E. University Parkway Bauernschmidt Building, Room 763 Baltimore, MD 21218 Tel: 410 554 2830 Fax: 410 554 2289 Email: [email protected]
US
*Corresponding author:
ACCEPTED MANUSCRIPT Abstract Charcot neuroarthropathy (CNA) often presents as a diabetic foot complication. The role of synovial mesenchymal stem cells (syn-MSCs) in the pathogenesis of CNA is unclear. Synovial samples were collected, for isolation of syn-MSCs, from diabetic patients with CNA (n=7) and
T
non-diabetic patients with intra-articular fracture or normal joints (non-CNA; n=7) during foot
IP
surgery. The syn-MSCs in the CNA and non-CNA groups were characterized comparatively.
CR
The average number of colonies formed in the CNA group was 6 ± 3.5 per half plate (10 mm in diameter), while it was 43 ± 21.6 in the non-CNA group (p < 0.05). The average size (pixels) of
US
the colonies in the CNA group was smaller than that in the non-CNA group. When the colonies
AN
were stratified into high-, medium- and low-density subgroups, colonies in the high-density subgroup of the CNA group were reduced in density. Expression of PPAR-γ, RUNX2, SOX9
M
and type II collagen by syn-MSCs in the CNA group was decreased during adipogenic,
ED
osteogenic and chondrogenic differentiation as compared with the non-CNA group. In conclusion, syn-MSCs in CNA joints were reduced in number, with declined differentiation
CE
pathology of CNA.
PT
potentials. The high-density subpopulation of the syn-MSCs was particularly affected by the
AC
Keywords: mesenchymal stem cells; synovium; neuroarthropathy; diabetes
ACCEPTED MANUSCRIPT 1. Introduction Charcot neuroarthropathy (CNA, also known as neuropathic osteoarthropathy or Charcot joint) was first described by Jean-Martin Charcot in 1868 as a destructive joint condition resulting from neuropathic deficiency [Chisholm and Gilchrist, 2011]. In modern medicine,
T
peripheral neuropathy secondary to diabetes mellitus is the predominant neuropathic condition to
IP
associate with CNA [Rajbhandari el al., 2002]. As a late-stage diabetic complication, CNA
CR
commonly occurs in the foot (also known as Charcot foot). Clinically, CNA displays joint inflammation, with fragmentation of cartilage and periarticular bone, in an insensate foot
US
[Frykberg el al., 2006]. The consequent foot deformity, due to joint dislocation and fracture, in
AN
CNA not only impairs patients’ mobility but also causes foot ulceration and infection [Hastings et al., 2013; Albers and Pop-Busui, 2014]. CNA is difficult to manage and some cases are
M
eventually treated with lower extremity amputation [Frykberg el al., 2006].
ED
It is agreed by the medical community that CNA is often misdiagnosed and non-treated in diabetes care partially because the pathogenesis of CNA remains elusive [Trieb, 2016]. A
PT
hallmark of CNA pathology is that bone and cartilage fragments are engulfed by hyperplastic
CE
synovium [Soudry et al., 1986; Sanders, 2004]. Synovium, which forms the inner lining of the joint capsule, produces fluid to lubricate joint surfaces and nourish chondrocytes. In a state of
AC
“synovitis”, synovium is the driving force behind articular cartilage destruction in inflammatory arthritis, such as rheumatoid arthritis [Fox et al., 2010; Bhattaram and Chandrasekharan, 2016]. A recent study demonstrated that fibroblast- like synoviocytes (FLS) isolated from CNA synovium are more invasive and chondrolytic under inflammatory conditions in vitro [Molligan et al., 2016], suggesting the involvement of synovium in CNA pathogenesis. Also residing in synovium are mesenchymal stromal/stem cells (MSCs), which are
ACCEPTED MANUSCRIPT capable of chondrogenic, osteogenic and adipogenic differentiation [De Bari et al., 2001; Sakaguchi et al., 2005]. These synovium-derived MSCs (syn-MSCs) have unique biological features, which distinguish them from the MSCs of other tissue origins. They express a relatively high level of CD90 [Djouad et al., 2005; Hagmann et al., 2016; Klimczak and Kozlowska, 2016],
T
and are more proliferative and superior in chondrogenesis [Sakaguchi et al., 2005; Yoshimura et
IP
al., 2007]. Joint injury often triggers syn-MSCs to proliferate and differentiate [Kurth et al.,
CR
2011; Fülber et al., 2016].
In addition to tissue homeostasis, syn-MSCs also involve in joint pathology. When co-
US
cultured with chondrocytes, syn-MSCs were able to inhibit their reaction to inflammatory stimuli
AN
[Ryu et al., 2014]. Syn-MSCs also participate in regulation of the proliferation of T cells in inflammatory joints [Eljaafari et al., 2012]. On the other hands, the pathological condition in a
M
joint could affect the functionality of syn-MSCs. In rheumatoid arthritis, syn-MSCs were
ED
suppressed in proliferation and chondrogenic differentiation [Jones et al., 2010]. In osteoarthritis, the chondrogenesis of syn-MSCs was inhibited by the increase of M1-phase macrophages in the
PT
synovium [Fahy et al., 2014]. Synovium is an important part of the CNA pathology [Soudry et
CE
al., 1986; Molligan et al., 2016]. The characteristics of syn-MSCs in CNA and their potential role in CNA pathology, however, are unknown.
AC
In this study, synovial samples were collected from the CNA joints during surgery for isolation of syn-MSCs. The syn-MSCs from the CNA joints were evaluated for clonagenicity, differentiation capacity and proliferation, in comparison with syn-MSCs from non-CNA joints.
2. Materials and Methods Synovial samples were collected from mid-foot, hind-foot and ankle joints during surgery
ACCEPTED MANUSCRIPT for foot reconstruction (approved by MedStar Health Research Institute Institutional Review Board). According to donors’ clinical conditions, the samples were organized in two groups: 1) the CNA group (n = 7) included 4 males and 3 females with an average age of 54.4 years. The clinical and imaging manifestation of CNA varies by the stages of the pathology. A diagnosis of
T
CNA in this study was based on a) foot deformity or swelling; b) loss of sensation or numbness;
IP
c) joint dislocation or fracture without significant injury (Fig 1); d) a medical history of diabetes.
CR
2) The non-CNA group (n = 7), as the controls, included 3 males and 4 females with an average age of 52.6 years. Synovium was collected from either normal or intra-articular- fractured foot
US
joints. None of these patients was diabetic. Small portions of the collected synovium in both
AN
CNA and non-CNA groups were fixed with 4% paraformaldehyde and embedded in O. C. T. compound (Sakura Finetek USA, Torrance, CA). Tissue sections (5 µm) were cut with a cryostat
Isolation of syn-MSCs:
ED
2.1.
M
and stained with hematoxylin and eosin (H&E).
Synovial samples of the CNA and non-CNA groups were washed and kept in Dulbecco's
PT
modified Eagle's medium (DMEM, Life Technologies, Long Island, NY) until processed. They
CE
were minced with scalpels and placed in 0.1% collagenase (Type I, Sigma-Aldrich Co., St. Louis, MO) in DMEM for 4 hours for digestion. The digest was filtered through a 70-μm cell
AC
strainer. After washing with phosphate buffered saline (PBS), cells were counted and cultured in DMEM supplemented with 10% fetal bovine serum (FBS, Rocky Mountain Biologics, Missoula, MT), 1% penicillin/streptomycin and 200mM L-glutamine (Life Technologies), at 37ºC with 5% CO2 in air. Culture medium was changed every 3 days. Cells were passaged with trypsin at 80% confluence. 2.2.
Colony forming units-fibroblast assay and colony characterization:
ACCEPTED MANUSCRIPT After one passage, cells from 4 CNA donors and 4 non-CNA donors were plated on 10cm tissue culture dishes at a density of 3,000 cells per cm2 . Cells from each donor were plated in triplicate. After two weeks, cultures (n = 24) were fixed in cold 100% ethanol and stained with crystal violet. The dishes were imaged, except for one in which staining was not well developed.
T
To analyze the colonies, all the images were converted to 16-bit images and adjusted to a
IP
standard threshold (8,250). A macro program was written in ImageJ (National Institutes of
CR
Health, Bethesda, MD) that took these processed images and created a map of the colonies, automatically tracing and numbering the colonies using “Analyze Particles” function. The first
US
trials of image processing showed a glare on the right side of each plate due to lighting that
AN
interfered with the program. Thus, the script was rewritten to input only colonies taken from the left semicircle of the plate. A colony (> 50 cells) was defined on image as > 2,000 pixel2 . The
M
number of colonies on each image was recorded. Following the circling of an individual colony,
ED
the Analyze Particles function was used to obtain the “% area” output. This gave the percentage of the area in a colony that had been turned to black in the binary process, as area (pixel2 ) or
PT
density (greyscale) of the colony.
CE
For data analysis, all the individual colonies of the CNA and non-CNA groups were inputted into the R programming language environment (The R Foundation for Statistical
AC
Computing, Vienna, Austria), with the input containing each colony’s size and corresponding density. Using R, the colonies were sorted by size and then stratified by size: for both the CNA and non-CNA groups, a list of colonies was generated ordering by size and the lists were divided into thirds of small, medium and large. Similarly, the colonies in the CNA and non-CNA groups were sorted and ordered by density. According to their density, the colonies of the CNA and nonCNA groups were divided into thirds: low, medium and high, for comparison.
ACCEPTED MANUSCRIPT 2.3.
Flow cytometry of common MSC markers:
Syn-MSCs from both CNA and non-CNA groups, at passage 3, were stained with propidium iodide (PI) for exclusion of dead cells. Cells were then incubated separately with phycoerythrin (PE) or fluorescein isothiocyanate (FITC) conjugated mouse anti-human CD14,
T
CD73, CD90, and CD105 antibody (BD Biosciences, San Jose, CA) at an amount of 0.01μg per
IP
100,000 cells for 30 minutes at room temperature. Fluorochrome conjugated normal mouse IgG
CR
was used as isotype controls. Each sample was prepared in triplicate. The labeled syn-MSCs were analyzed on a flow cytometer (Acurri C6, BD Biosciences). The percentage of positively
US
stained cells for each antigen was calculated using Overton histogram subtraction with FCS
2.4.
Proliferation of syn-MSCs:
AN
Express Analysis Suite (De Novo Software, Glendale, CA).
M
Syn-MSCs isolated from CNA or non-CNA synovial samples, at passage 3-5, were
ED
seeded in 24-well plates at a density of 1000 cells/well. The cells were collected with trypsin and counted with a hemocytometer on days 3, 5, 7, 9, 11, and 13. Proliferation was estimated by
PT
calculating the cell population doubling time.
CE
2.5. Differentiation of syn-MSCs: Syn-MSCs isolated from the CNA and non-CNA synovial samples, after passage 4, were
AC
used for tri-lineage differentiation under the following conditions: 1) for osteogenic differentiation, cells were seeded into T-25 culture flasks at a density of 1x104 /cm2 . Osteogenic media consisted of DMEM, 10% FBS, 10 mM β-glycerophosphate, 100 nM dexamethasone, 50µg/ml L-ascorbic acid 2-phosphate, and 100 ng/mL human recombinant bone morphogenetic protein 2 (BMP-2, Medtronik Sofamer Denek, Memphis, TN). 2) For adipogenic differentiation, syn-MSCs were cultured as above-mentioned but in adipogenic media, which was DMEM
ACCEPTED MANUSCRIPT supplemented with 10% FBS, 10 uM rosiglitazone, 1 uM dexamethasone, and 10 ng/ml bovine insulin [Contador et al., 2015]. 3) Chondrogenic differentiation of syn-MSCs was conducted in pellet culture (3x105 cells/pellet). The pellets were cultured in chondrogenic media that was DMEM supplemented with 10% FBS, 50μ/ml L-ascorbic acid 2-phosphate, 40ug/ml L-proline,
T
0.1 μM dexamethasone and 10 ng/ml recombinant human transforming growth factor β1
IP
(Peprotech, Rocky Hill, NJ). For enhanced chondrogenesis, recombinant human bone
CR
morphogenetic protein-7 (100 ng/ml; Peprotech) was added into the chondrogenic media at days
US
3 and 12 of tissue culture, following a published protocol [Handorf and Li, 2014]. Tissue cultures of osteogenic, adipogenic and chondrogenic differentiation were maintained for three weeks. The
AN
controls of the tri-lineage differentiation were the same syn-MSCs cultured in regular growth media.
M
At the end of chondrogenic, adipogenic and osteogenic differentiation, syn-MSCs in
ED
differentiation cultures and the corresponding controls were lysed with TRIzol (Life Technologies) and RNA was extracted by phase separation. First-strand cDNA synthesis was
PT
performed using the iScript Reverse Transcription Supermix (Bio-Rad Laboratories, Hercules,
CE
CA) with 100 ng of RNA template. For quantitative gene expression, first-strand cDNA was amplified using the SSO Advanced SYBR Green Supermix (Bio-Rad Laboratories) on a CFX-
AC
Connect Real-Time PCR System (Bio-Rad Laboratories). After initial denaturation at 95ºC for 30 seconds, polymerase chain reaction (PCR) was performed with 40 cycles consisting of 10 seconds at 95ºC and 30 seconds at 53-59ºC (depending on the primer pair). An efficiency of >90% was required for each primer pair and annealing temperature. The specificity of each primer pair was verified using a melting curve from 65-95ºC and agarose gel electrophoresis of the PCR product. Genes investigated were type II collagen and Sox9 for chondrogenic
ACCEPTED MANUSCRIPT differentiation, RUNX2 for osteogenic differentiation and peroxisome proliferator-activated receptor gamma (PPAR-γ) for adipogenic differentiation. The expression of glyceraldehyde 3phosphate dehydrogenase (GAPDH) was used as an internal reference. Details of primers (Integrated DNA Technologies, Coralville, IA) are listed in Table 1. Gene expression was
T
expressed as fold-change from undifferentiated syn-MSCs corresponding to each sample using
IP
the ∆∆CT method (User Manual, Bio-Rad Laboratories).
CR
Adipogenic and osteogenic cultures of syn-MSCs isolated from CNA and non-CNA synovial samples, under the above-mentioned conditions, were also conducted in 24-well plates.
US
After three weeks of culture, the cells were fixed with 4% paraformaldehyde. Neutral lipid
AN
droplets in adipogenic cultures were stained with Oil Red-o in polyethylene glycol and matrix mineralization in osteogenic cultures was visualized with Alizarin red staining. To analyze
M
chondrogenic differentiation, frozen sections of chondrogenic pellet cultures of syn-MSCs were
ED
stained with Alcian blue.
Statistical analysis: Data are presented as mean ± standard deviation. The average
PT
number, size and density of colonies were compared between the CNA and non-CNA groups
CE
with t test. The stratified colonies in both CNA and non-CNA groups were compared separately by the averaged area of small-, medium- and large-size colonies, and the averaged greyscale of
AC
low-, medium- and high-density colonies, using unpaired t test. In the proliferation study, the cell numbers of syn-MSCs collected at the same time points were compared between the CNA and non-CNA groups with t test. During the differentiation of syn-MSCs, the expression of genes related to chondrogenesis, adipogenesis and osteogenesis in fold-change in the CNA and nonCNA groups was evaluated with t test. P < 0.05 was set as significant.
ACCEPTED MANUSCRIPT 3. Results 3.1.
Histology of synovium: In the CNA group, hyperplasia of synovium included
multiple layers of cells in the intimal layer (Fig 2A). There were only one or two layers of lining cells in the non-CNA synovium (Fig 2B). There was no significant infiltration of leukocytes in
Colony formation and characterization.
IP
3.2.
T
the subintimal area in both groups.
CR
After two weeks in culture, syn-MSCs in both CNA and non-CNA groups formed colonies with varied size and shape (Fig 3A and B). The average number of colonies in the CNA
US
group was 6 ± 3.5 per half plate, while it was 43 ± 21.6 in the non-CNA group (p < 0.05; Fig
AN
3C). The average colony size (pixels) in the CNA group was smaller than that in the non-CNA group (4780±1232 vs. 7960±1687; p < 0.05) but the average colony densities of the CNA and
M
non-CNA groups were about equal (Fig 3D and E). Based on their sizes, the colonies in both
ED
CNA and non-CNA groups were stratified into thirds, i.e. the small, medium and large subgroups. In all three subcategories, colonies in the CNA group were smaller than those in the
PT
non-CNA group (p <0.05; Fig 3F). When the colonies were similarly stratified into thirds
CE
according to density, the average colony densities in the low- and medium-density subgroups were not significantly different between the CNA and non-CNA groups (Fig 3G). In the high-
AC
density subcategory, however, the CNA group had a lower colony density than the non-CNA group (30±6 vs. 35±10; p < 0.05). 3.3.
Cell surface markers expressed by syn-MSCs.
Syn-MSCs in both CNA and non-CNA groups expressed the common MSC cell surface markers comparably. Syn-MSCs in the CNA group were largely positive for CD90 (99.0%±0.5%), CD73 (98.0%±0.9%) and CD105 (98.6%±0.4%), but negative for CD14
ACCEPTED MANUSCRIPT (16.9%±1.6; Fig 4). 3.4.
Proliferation of syn-MSCs.
The population doubling time of the syn-MSCs was 1.8 days in the CNA group and 1.6 days in the non-CNA group. The average numbers of syn-MSCs sampled at days 3, 5, 7, 11 were
T
not significantly different between the CNA and non-CNA groups (p > 0.05). By day 13, the
group (p < 0.05; Fig 5). 3.5.
Tri-lineage differentiation of syn-MSCs.
CR
IP
number of syn-MSCs in the non-CNA group was significantly greater than that in the CNA
US
After culture in adipogenic media, neutral lipid droplets were developed in the syn-MSCs
AN
of both CNA and non-CNA groups but they were consistently more prevalent in the non-CNA group (Fig 6A). In osteogenic culture, syn-MSCs from the non-CNA group exhibited more
M
extensive matrix mineralization than those in the CNA group. In chondrogenic pellet culture, the
ED
CNA group displayed reduced proteoglycans in cartilage matrix and cell density as compared with the non-CNA group.
PT
During adipogenic differentiation, PPAR-γ expression by syn-MSCs in the non-CNA
CE
group increased 8-fold while it increased only about 3-fold in the CNA group (p<0.05; Fig 6B). Syn-MSCs subjected to osteogenic differentiation increased in RUNX2 expression by 10-fold in
AC
the non-CNA group and 2-fold in the CNA group (p<0.05), respectively. In chondrogenic differentiation, the expression of Sox9 by syn-MSCs increased by 37-fold in the non-CNA group and by 15-fold in the CNA group (p<0.05). The increased expression of type II collagen by chondrogenic syn-MSCs in the non-CNA group was 3 times of that in the CNA group (p<0.05).
4. Discussion
ACCEPTED MANUSCRIPT The isolated syn-MSCs in both CNA and non-CNA groups expressed the common markers for MSCs: CD14-, CD73+, CD90+, and CD105+ and were capable of tri-lineage differentiation [Dominici et al., 2006]. The syn-MSCs in the CNA group, however, presented changes in quantity, clonogenicity, proliferation and differentiation. The quantity of MSCs in the
T
tissue has been investigated as an indication of healthy conditions [Stolzing et al., 2008]. Syn-
IP
MSCs of both CNA and non-CNA groups formed colonies simultaneously in low-density
CR
cultures but the CNA group produced far fewer colonies than the non-CNA group, indicating a reduced quantity of syn-MSCs in the CNA synovium. It is not uncommon that the number of
US
syn-MSCs is reduced in arthritic conditions. In rheumatoid arthritis, the number of syn-MSCs is
AN
reversely correlated with the severity of joint inflammation [Jones et al., 2010]. The pathology of CNA includes extensive synovial hyperplasia [Soudry et al., 1986]. The significant “synovitis”
M
in CNA could be partially responsible for a reduced number of syn-MSCs. It is noteworthy that
ED
not only the degree but also the stage of the joint inflammation affects the quantity of syn-MSCs. For example, the number of syn-MSCs was increased in acute arthritis induced by carrageenan
PT
injection [Matsukura et al., 2015]. CNA pathology evolves over an extended period clinically, as
CE
the neuropathy progresses, and is generally more chronic. In CNA, disorders of neuroendocrine regulation could also influence the fate of syn-MSCs. In CNA synovium, the number of
AC
sympathetic nerve fibers (labeled with tyrosine hydroxylase) is reduced [Koeck et al., 2009]. The similar neuropathy of sympathetic nerves has been found to be causative of apoptosis of bone marrow MSCs [Arranz et al., 2014]. Syn-MSCs in both CNA and non-CNA groups formed colonies with varied size and density. These colony features are indicative of the proliferation and differentiation potentials of the original MSCs and could be affected by pathological conditions [Mareddy et al., 2007]. The
ACCEPTED MANUSCRIPT colony size of the CNA group, overall and in the three-size subcategories, was smaller than that of the non-CNA group. This suggests that syn-MSCs in the CNA group were less proliferative than that in the non-CNA group, since the size of a colony is largely determined by the proliferation of MSCs [Stolzing et al., 2010]. This was confirmed when syn-MSCs in both CNA
T
and non-CNA groups were plated in parallel for proliferation assay, where the syn-MSCs in the
IP
CNA group grew at a slower rate. A diabetic history of the CNA patients in this study could have
CR
contributed to slow proliferation of the syn-MSCs in the CNA group because bone marrow MSCs showed significant deficiency of proliferation in diabetes [Shin and Peterson, 2012].
US
Colony density is more relevant to the spreading of the proliferating MSCs [Sengers et
AN
al., 2010]. While the average colony densities of the CNA and non-CNA groups were similar, analysis of the stratified colonies revealed a reduced colony density in the high-density
M
subcategory by the CNA group. In the high-density subcategory, syn-MSCs were more
ED
aggregated and less dispersed than those in the low- and medium-density subcategories. Subpopulations of MSCs often distinguish each other with differential expression of cytoskeletal
PT
and structural proteins [Mareddy et al., 2009], which are critical elements of cell mobility or
CE
spreading. Colony density is also influenced by the survival and adaption of the initial MSCs [Barbaric et al., 2014]. It is interesting that the subpopulation of syn-MSCs forming high-density
AC
colonies was selectively affected by the pathology of CNA and warrant further investigation on their specific implication on CNA pathology. Differentiation of the syn-MSCs in CNA was compromised in general. Chondrogenesis of syn-MSCs in the CNA group was suppressed, as demonstrated by gene expression of Sox9 and type II collagen, and Alcian blue staining of proteoglycans in extracellular matrix. Chondrogenesis of syn-MSCs is similarly inhibited in rheumatoid arthritis but the expression of
ACCEPTED MANUSCRIPT Sox9 is unchanged [Jones et al., 2010]. CNA and rheumatoid arthritis may interfere with the chondrogenic differentiation of syn-MSCs through distinct pathways. Adipogenic and osteogenic differentiation of syn-MSCs was inhibited in CNA but it was not the case in rheumatoid arthritis [Jones et al., 2010]. The suppression of tri-lineage differentiation of syn-MSCs in CNA is likely
T
a result of interplay of neuropathy, inflammation and diabetes. Neuropathy alone impacts the
IP
function of MSCs. For example, a reduction of sympathetic nerve fibers enhances osteogenic
CR
differentiation of bone marrow MSCs [Hanoun et al., 2014]. Although detailed neuropathic conditions in the CNA synovium remain to be revealed, it is clear that the neuro-signaling
US
network in synovium varies in different types of arthritic joints [Dirmeier et al., 2008]. In this
AN
study, all the synovium donors of CNA group were diabetic. Hyperglycemia promotes senescence and severely inhibits proliferation and differentiation of MSCs [Chang et al., 2015].
M
It is possible that the CNA group had more syn-MSCs entered senescence than the non-CNA
ED
group, which resulted in decreased clonogenicity and differentiation [Alves et al., 2010]. In conclusion, syn-MSCs in CNA are reduced in number and inhibited in proliferation
PT
and differentiation to chondrogenic, adipogenic, and osteogenic lineages. The altered biological
CE
properties of syn-MSCs are a part of the pathology of CNA.
AC
Acknowledgement: the authors declare no conflicts of interest related to this manuscript.
ACCEPTED MANUSCRIPT References 1. Albers JW, Pop-Busui R. 2014. Diabetic neuropathy: mechanisms, emerging treatments, and subtypes. Curr Neurol Neurosci Rep. 14(8):473. 2. Alves H, Munoz-Najar U, De Wit J et al. 2010. A link between the accumulation of DNA
T
damage and loss of multi-potency of human mesenchymal stromal cells. J Cell Mol Med.
IP
14(12):2729-2738.
CR
3. Arranz L, Sánchez-Aguilera A, Martín-Pérez D et al. 2014. Neuropathy of haematopoietic stem cell niche is essential for myeloproliferative neoplasms. Nature.
US
512(7512):78-81.
AN
4. Barbaric I, Biga V, Gokhale PJ et al. 2014. Time-lapse analysis of human embryonic stem cells reveals multiple bottlenecks restricting colony formation and their relief upon
M
culture adaptation. Stem Cell Reports. 3(1):142-155.
ED
5. Bhattaram P, Chandrasekharan U. 2016. The joint synovium: A critical determinant of articular cartilage fate in inflammatory joint diseases. Semin Cell Dev Biol. S1084-
PT
9521(16)30135-5.
CE
6. Chang TC, Hsu MF, Wu KK. 2015. High glucose induces bone marrow-derived mesenchymal stem cell senescence by upregulating autophagy. PLoS One.
AC
10(5):e0126537.
7. Chisholm, K.A., Gilchrist, J.M. 2011. The Charcot joint: a modern neurologic perspective. J Clin Neuromuscul Dis 13(1):1-13. 8. Contador D, Ezquer F, Espinosa M et al. 2015. Dexamethasone and rosiglitazone are sufficient and necessary for producing functional adipocytes from mesenchymal stem cells. Exp Biol Med (Maywood). 240(9):1235-1246.
ACCEPTED MANUSCRIPT 9. De Bari C, Dell'Accio F, Tylzanowski P et al. 2001. Multipotent mesenchymal stem cells from adult human synovial membrane. Arthritis Rheum. 44(8):1928-1942. 10. Dirmeier M, Capellino S, Schubert T et al. 2008. Lower density of synovial nerve fibres positive for calcitonin gene-related peptide relative to substance P in rheumatoid arthritis
T
but not in osteoarthritis. Rheumatology (Oxford). 47(1):36-40.
IP
11. Djouad F, Bony C, Häupl T et al. 2005. Transcriptional profiles discriminate bone
CR
marrow-derived and synovium-derived mesenchymal stem cells. Arthritis Res Ther. 7(6):R1304-315.
US
12. Dominici M, Le Blanc K, Mueller I et al. 2006. Minimal criteria for defining multipotent
AN
mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy. 8(4):315-317.
M
13. Eljaafari A, Tartelin ML, Aissaoui H et al. 2012. Bone marrow-derived and synovium-
ED
derived mesenchymal cells promote Th17 cell expansion and activation through caspase
64(7):2147-2157.
PT
1 activation: contribution to the chronicity of rheumatoid arthritis. Arthritis Rheum.
CE
14. Fahy N, de Vries-van Melle ML, Lehmann J et al. 2014. Human osteoarthritic synovium impacts chondrogenic differentiation of mesenchymal stem cells via macrophage
AC
polarisation state. Osteoarthritis Cartilage. 22(8):1167-1175. 15. Fox DA, Gizinski A, Morgan R et al. 2010. Cell-cell interactions in rheumatoid arthritis synovium. Rheum Dis Clin North Am. 36(2):311-323. 16. Frykberg RG, Zgonis T, Armstrong DG et al. 2006. Diabetic foot disorders. A clinical practice guideline (2006 revision). J Foot Ankle Surg. 45(5 Suppl):S1-66. 17. Fülber J, Maria DA, da Silva LC et al. 2016. Comparative study of equine mesenchymal
ACCEPTED MANUSCRIPT stem cells from healthy and injured synovial tissues: an in vitro assessment. Stem Cell Res Ther. 7:35. 18. Hagmann S, Rimmele C, Bucur F et al. 2016. Mesenchymal stromal cells from osteoarthritic synovium are a distinct population compared to their bone-marrow
T
counterparts regarding surface marker distribution and immunomodulation of allogeneic
IP
CD4+ T-Cell cultures. Stem Cells Int. 6579463.
CR
19. Handorf AM, Li WJ. 2014. Induction of mesenchymal stem cell chondrogenesis through sequential administration of growth factors within specific temporal windows. J Cell
US
Physiol. 229(2):162-171.
AN
20. Hanoun M, Zhang D, Mizoguchi T et al. 2014. Acute myelogenous leukemia- induced sympathetic neuropathy promotes malignancy in an altered hematopoietic stem cell
M
niche. Cell Stem Cell. 15(3):365-375.
ED
21. Hastings MK, Johnson JE, Strube MJ et al. 2013. Progression of foot deformity in Charcot neuropathic osteoarthropathy. J Bone Joint Surg Am. 95(13):1206-1213.
PT
22. Klimczak A, Kozlowska U. 2016. Mesenchymal stromal cells and tissue-specific
CE
progenitor cells: their role in tissue homeostasis. Stem Cells Int. 2016:4285215. 23. Jones E, Churchman SM, English A et al. 2010. Mesenchymal stem cells in rheumatoid
AC
synovium: enumeration and functional assessment in relation to synovial inflammation level. Ann Rheum Dis. 69(2):450-457. 24. Koeck FX, Bobrik V, Fassold A et al. 2009. Marked loss of sympathetic nerve fibers in chronic Charcot foot of diabetic origin compared to ankle joint osteoarthritis. J Orthop Res. 27(6):736-741. 25. Kurth TB, Dell'accio F, Crouch V et al. 2011. Functional mesenchymal stem cell niches
ACCEPTED MANUSCRIPT in adult mouse knee joint synovium in vivo. Arthritis Rheum. 63(5):1289-1300. 26. Mareddy S, Crawford R, Brooke G et al. 2007. Clonal isolation and characterization of bone marrow stromal cells from patients with osteoarthritis. Tissue Eng. 13(4):819-829. 27. Mareddy S, Broadbent J, Crawford R et al. 2009. Proteomic profiling of distinct clonal
T
populations of bone marrow mesenchymal stem cells. J Cell Biochem. 106:776-786.
IP
28. Matsukura Y, Muneta T, Tsuji K et al. 2015. Mouse synovial mesenchymal stem cells
CR
increase in yield with knee inflammation. J Orthop Res. 33(2):246-253. 29. Molligan J, Barr C, Mitchell R et al. 2016. Pathological role of fibroblast-like
US
synoviocytes in charcot neuroarthropathy. J Orthop Res. 34(2):224-230.
AN
30. Rajbhandari SM, Jenkins RC, Davies C et al. 2002. Charcot neuroarthropathy in diabetes mellitus. Diabetologia. 45(8):1085-1096.
M
31. Ryu JS, Jung YH, Cho MY et al. 2014. Co-culture with human synovium-derived
ED
mesenchymal stem cells inhibits inflammatory activity and increases cell proliferation of
447(4):715-720.
PT
sodium nitroprusside-stimulated chondrocytes. Biochem Biophys Res Commun.
CE
32. Sakaguchi Y, Sekiya I, Yagishita K et al. 2005. Comparison of human stem cells derived from various mesenchymal tissues: superiority of synovium as a cell source. Arthritis
AC
Rheum. 52(8):2521-2529. 33. Sanders LJ. 2004. The Charcot foot: historical perspective 1827-2003. Diabetes Metab Res Rev. 20 Suppl 1:S4-8. 34. Sengers BG, Dawson JI, Oreffo RO. 2010. Characterisation of human bone marrow stromal cell heterogeneity for skeletal regeneration strategies using a two-stage colony assay and computational modelling. Bone. 46(2):496-503.
ACCEPTED MANUSCRIPT 35. Shin L, Peterson DA. 2012. Impaired therapeutic capacity of autologous stem cells in a model of type 2 diabetes. Stem Cells Transl Med. 1(2):125-135. 36. Soudry M, Binazzi R, Johanson NA et al. 1986. Total knee arthroplasty in Charcot and Charcot-like joints. Clin Orthop Relat Res. 208:199-204.
T
37. Stolzing A, Jones E, McGonagle D et al. 2008. Age-related changes in human bone
IP
marrow-derived mesenchymal stem cells: consequences for cell therapies. Mech Ageing
CR
Dev. 129(3):163-173.
38. Stolzing A, Sellers D, Llewelyn O et al. 2010. Diabetes induced changes in rat
US
mesenchymal stem cells. Cells Tissues Organs. 191(6):453-465.
AN
39. Trieb K. 2016. The Charcot foot: pathophysiology, diagnosis and classification. Bone Joint J. 98-B(9):1155-1159.
M
40. Yoshimura H, Muneta T, Nimura A et al. 2007. Comparison of rat mesenchymal stem
ED
cells derived from bone marrow, synovium, periosteum, adipose tissue, and muscle. Cell
AC
CE
PT
Tissue Res. 327(3):449-462.
ACCEPTED MANUSCRIPT
NM_000346.3
COL2A
NM_001844.4
PPAR-γ
AB565476
RUNX2
NM_001024630.3
Reverse Primer Sequence
5' GAC AGT CAG CCG CAT CTT CT 3' 5' AGG AAG TCG GTG AAG AAC GG 3' 5' GCC TGG TGT CAT GGG TTT 3' 5' CCT AAA CTT CGG ATC CCT CC 3'
5' GCG CCC AAT ACG ACC AAA TC 3' 5' CGC CTT GAA GAT GGC GTT G 3' 5' GTC CCT TCT CAC CAG CTT TG 3' 5' TCA AAT CTG GTG TCG TTT GC 3'
5' CTA GTT TGT TCT CTG ACC GC 3'
5' TGG GGT CTG TAA TCT GAC TC 3'
AN M ED PT CE AC
T
SOX9
Forward Primer Sequence
CR
NCBI Accession Number NM_002046.5
US
Gene Name GAPDH
IP
Table 1: A list of primers
ACCEPTED MANUSCRIPT
Figure Legends Figure 1. A radiogram of CNA in a foot. The diabetic patient’s right foot has irregular joint space (arrows) in multiple joints around navicular in the mid-foot, and bony fragmentation of navicular
IP
T
and talar head. It is also notable of the collapse of foot arch.
CR
Figure 2. Histology of CNA and non-CNA synovium. Synovium in the CNA group is more hyperplasic, with an expanded intimal layer (A; arrows). Its subintimal layer is more fibrous,
US
compared with the non-CNA synovium (B). The intimal layer in the non-CNA synovium
AN
consists of 1-2 layers of cells (H&E staining; bar = 100µm).
M
Figure 3. Analyses of colonies formed by syn-MSCs. Colonies are formed by syn-MSCs in both
ED
CNA (A) and non-CNA (B) groups after two weeks of culture. While colonies spread on the non-CNA plate, only a few colonies scatter on the CNA plate (crystal violet staining). On
PT
average, the number of colonies formed by the syn-MSCs in the CNA group is far fewer than
CE
that in the non-CNA group (C). The average size of the colonies in the CNA group is smaller than that in the non-CNA group (D). The average colony density, however, is not different
AC
between the two groups (E). When colonies of the CNA and non-CNA groups are stratified into thirds by the size: small, medium and large, the average colony size of each subgroup of the CNA group is smaller than the corresponding subgroup in the non-CNA group (F). When colonies in the CNA and non-CNA groups are stratified into thirds by density, colony density of the high-density subgroup in the CNA group is lower than the corresponding one in the nonCNA group (G). Colony densities of the low- and medium-density subgroups are not different between the CNA and non-CNA groups (note: * indicates p < 0.05; ** indicates p < 0.001).
ACCEPTED MANUSCRIPT
Figure 4. Flow cytometry for syn-MSCs. Syn-MSCs in the CNA group (top row) express nearly negatively for CD14, but strongly for CD73, CD90 and CD105. The expression profile of the tested cell surface markers is not significantly different between the CNA and non-CNA groups
IP
T
(note: in the representative images, grey line is for isotope control and red line for syn-MSCs).
CR
Figure 5. Proliferation of syn-MSCs. Syn-MSCs from both CNA and non-CNA groups were cultured for 13 days and sampled every other day. The cell number of the non-CNA group
US
increases at a greater pace and becomes significantly greater than the CNA group at day 13
AN
(note: * indicates p < 0.05).
M
Figure 6. Tri-lineage differentiation of syn-MSCs. A: In adipogenic differentiation, formation of
ED
fat droplets is more proliferous in the non-CNA group than the CNA group (Oil Red O staining;). In osteogenic differentiation, matrix mineralization is more intense in the non-CNA
PT
group than in the CNA group (Alizarin red staining). The tissue section of chondrogenic pellets
CE
in the non-CNA group is intensely stained for cartilageous matrix (Alcian blue staining; bar = 50µm). B: Quantitative gene expression during tri-lineage differentiation of syn-MSCs. The
AC
expression of PPAR-γ, Runx2, Sox9 and type II collagen is significantly increased in the non-CNA group that that in the CNA group (note: * indicates p < 0.05).
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
Fig. 1
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
Fig. 2
ED
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Fig. 3
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
Fig. 4
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
Fig. 5
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN
US
CR
Fig. 6