Action at a Distance: Systemically Administered Adult Stem/Progenitor Cells (MSCs) Reduce Inflammatory Damage to the Cornea Without Engraftment and Primarily by Secretion of TNF-α Stimulated Gene/Protein 6†‡§
Gavin W. Roddy,
Institute for Regenerative Medicine, Texas A&M Health Science Center, College of Medicine at Scott & White, Temple, Texas, USA
Author contributions: G.W.R. and J.Y.O.: conception and design, provision of study material, collection and assembly of data, data analysis and interpretation, manuscript writing; R.H.L.: conception and design, provision of study material, data analysis and interpretation; T.J.B., J.Y., and K.C.: provision of study material; R.H.R.: data interpretation; D.J.P.: conception and design, data analysis and interpretation, manuscript writing, financial support, final approval of manuscript. G.W.R. and J.Y.O. contributed equally to this article.
Disclosure of potential conflicts of interest is found at the end of this article.
First published online in STEM CELLSEXPRESS August 11, 2011.
Previous reports demonstrated that the deleterious effects of chemical injury to the cornea were ameliorated by local or systemic administration of adult stem/progenitor cells from bone marrow referred to as mesenchymal stem or stromal cells (MSCs). However, the mechanisms for the beneficial effects of MSCs on the injured cornea were not clarified. Herein, we demonstrated that human MSCs (hMSCs) were effective in reducing corneal opacity and inflammation without engraftment after either intraperitoneal (i.p.) or intravenous (i.v.) administration following chemical injury to the rat cornea. A quantitative assay for human mRNA for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) demonstrated that less than 10 hMSCs were present in the corneas of rats 1-day and 3 days after i.v. or i.p. administration of 1 × 107 hMSCs. In vitro experiments using a transwell coculture system demonstrated that chemical injury to corneal epithelial cells activated hMSCs to secrete the multipotent anti-inflammatory protein TNF-α stimulated gene/protein 6 (TSG-6). In vivo, the effects of i.v. injection of hMSCs were largely abrogated by knockdown of TSG-6. Also, the effects of hMSCs were essentially duplicated by either i.v. or topical administration of TSG-6. Therefore, the results demonstrated that systemically administered hMSCs reduce inflammatory damage to the cornea without engraftment and primarily by secretion of the anti-inflammatory protein TSG-6 in response to injury signals from the cornea. STEM CELLS 2011;29:1572–1579
Numerous reports have described therapeutic benefits in various disease models after administration of the adult stem/progenitor cells from bone marrow referred to initially as colony forming units-fibroblastic, then as bone marrow stromal cells, subsequently as mesenchymal stem cells, and most recently as multipotent mesenchymal stromal cells (MSCs) [1–6]. The observations were initially interpreted with the assumption that the cells repaired tissues by engrafting and differentiating to replace injured cells. Long-term engraftment with differentiation was observed in some animal models such as those with severe injuries to tissues, in embryos or with local infusions of the cells. However, repair of tissues and functional improvements were more frequently observed without long-term engraftment of MSCs. Therefore, most of the beneficial effects were explained by paracrine secretions or cell-to-cell contacts that had multiple effects including modulation of inflammatory or immune reactions [6–10]. Although MSCs in culture secreted a large number of cytokines, [11–12] recent reports demonstrated that MSCs were activated by cross-talk with injured cells to express high levels of therapeutic factors [10, 13]. Among the therapeutic factors that MSCs express in response to tissue injury  is the multipotent anti-inflammatory protein tumor necrosis factor (TNF)-α–stimulated gene/protein 6 (TSG-6) [15, 16].
The cornea presents an attractive model to study inflammation because it is an anatomically discrete tissue that can readily be exposed to injury. Moreover, the response to therapeutic agents can be evaluated in terms of both observational improvements and molecular changes. Previous reports demonstrated that consequences of chemical injury to the cornea are ameliorated by MSCs either by intravenous (i.v.) administration , by topical application of MSCs , by topical application of conditioned medium from cultures of MSCs , or by transplantation of MSCs to the cornea [18, 19]. Similar results were also obtained by injection of TSG-6 into the anterior chamber of the eye .
In this report, we demonstrate that both intraperitoneal (i.p.) and i.v. administration of human MSCs (hMSCs) significantly suppressed the development of corneal opacity and inflammation after chemical injury to the cornea without engraftment of the cells in the cornea and primarily by the cells being activated to express TSG-6. We also demonstrate that similar therapeutic effects were reproduced by i.v. or topical administration of TSG-6.
MATERIALS AND METHODS
Animals and Reagents
The experimental protocols were approved by the Institutional Animal Care and Use Committee of Texas A&M Health Science Center. Six-week-old male Lewis rats (LEW/Crl; Charles River Laboratories International, Wilmington, MA) weighing 180–200 g or 8-week-old male BALB/c mice (BALB/cAnNCrl; Charles River Laboratories International) weighing 18–20 g were used in all experiments.
Vials of frozen passage 1 hMSCs were obtained from the Center for the Preparation and Distribution of Adult Stem Cells (http://medicine.tamhsc.edu/irm/msc-distribution.html) that supplies standardized preparations of MSCs enriched for early progenitor cells to over 300 laboratories under the auspices of an NIH/NCRR grant (P40 RR 17447-06). All of the experiments were performed with hMSCs from one donor (7071L). The cells consistently differentiated into three lineages in culture, were negative for hematopoietic markers (CD34, CD36, CD117, and CD45), and positive for CD29 (95%), CD44 (>93%), CD49c (99%), CD49f (>70%), CD59 (>99%), CD90 (>99%), CD105 (>99%), and CD166 (>99%) . Following culture at high density for 24 hours to recover viable cells, hMSCs were plated at low density (100 cells per square centimeter), incubated in complete culture medium (CCM) with 16% fetal bovine serum (FBS) for 8 days until approximately 70% confluence was reached, and harvested with 0.25% trypsin/1 mM EDTA at 37°C for 2 minutes. Thereafter, the trypsin was inactivated by adding the CCM to the cells, and the cells were washed with phosphate-buffered saline (PBS) by centrifugation at 1,200 rpm for 5 minutes. The cells were frozen in α–minimum essential medium with 30% FBS and 5% dimethyl sulfoxide at a concentration of 1 × 106 cells per milliliter. Passage 2 cells were expanded with the same protocol, and passage 3 cells were used for all experiments. Following lifting the cells prior to injection, a final wash was performed using Hank's balanced saline solution (HBSS; BioWhittaker, Walkersville, MD). After washing by centrifugation, the cells were suspended in HBSS at a concentration of 10,000 cells per microliter for injection. Recombinant human TSG-6 (rhTSG-6) was purchased from R&D Systems (Minneapolis, MN).
For small interfering RNA (siRNA) experiments, hMSCs were transfected with siRNA for TSG-6 (sc-39819; Santa Cruz Biotechnology, Santa Cruz, CA) or scrambled siRNA (StealthTM RNAi Negative Control; Invitrogen) with a commercial kit (Lipofectamine RNAiMAX reagent; Invitrogen, Carlsbad, CA). To confirm knockdown of TSG-6, RNA was extracted from aliquots of the cells (RNeasy Mini kit; Qiagen, Valencia, CA) 15 hours after the start of transfection (the same time point when TSG-6-siRNA or scrambled siRNA MSCs were injected into the rats), and assayed for TSG-6 by real-time reverse-transcriptase polymerase chain reaction (RT-PCR).
Animal Model of Injury and Treatment
Following anesthesia by isoflurane inhalation, absolute ethanol was applied to the corneal and limbal surface for 30 seconds followed by rinsing with balanced salt solution and mechanical debridement of the corneal and limbal epithelium with a surgical blade as described previously . Following epithelium removal, tarsorrhaphy was performed using 8-0 nylon suture. Immediately following injury, the treatment was applied as follows: for systemic MSC experiments, rats received 1 × 107 hMSCs in 1 mL HBSS (i.v. or i.p.) or HBSS 1 mL (i.v. or i.p.); for i.v. TSG-6 experiments, mice received 30 μg of rhTSG-6 in 150 μL PBS or PBS 150 μL via tail vein; for siRNA experiments, mice received 1 × 106 hMSCs treated with siRNA for TSG-6 in 150 μL HBSS, 1 × 106 hMSCs treated with scrambled siRNA control, untreated control hMSCs, or HBSS 150 μL; for topical TSG-6 experiments, rats received topical application of 8 μg of rhTSG-6 in 20 μL PBS or PBS once immediately after injury. After application, the tarsorrhaphy was performed.
Ocular Surface Evaluation
At 1 or 3 days following injury and treatment, rat corneas were photographed and graded clinically as described previously [20, 22]: grade 0, completely transparent cornea; grade 1, minimal corneal opacity, but iris clearly visible; grade 2, moderate corneal opacity, iris vessels still visible; grade 3, moderate corneal opacity, pupil margin but not iris vessels visible; and grade 4, complete corneal opacity, pupil not visible.
Real-Time RT-PCR Analysis
For RNA extraction, the cornea was minced into small pieces, lysed in RNA isolation reagent (RNA Bee, Tel-Test, Friendswood, TX) and homogenized. Total RNA was extracted, (RNeasy Mini kit; Qiagen) and cDNA was generated by reverse transcription (SuperScript III; Invitrogen) using 1 μg total RNA. Real-time amplification was performed using TaqMan Universal PCR Master Mix (Applied Biosystems, Carlsbad, CA). An 18s ribosomal RNA (rRNA) probe (Taqman Gene Expression Assays ID, Hs03003631_g1) was used for normalization of gene expression. For all the PCR probe sets, Taqman Gene Expression Assay kits were purchased from Applied Biosystems. The assay was performed in triplicate technical replicates for each biological sample.
For protein extraction, the cornea was minced into small pieces and lysed in tissue extraction reagent (Invitrogen) containing protease inhibitor cocktail (Roche, Indianapolis, IN). The samples were sonicated on ice (Ultrasonic Processor, Cole Parmer Instruments, Vernon Hills, IL). After centrifugation at 12,000 rpm at 4°C for 20 minutes, the supernatant was collected and assayed by enzyme-linked immunosorbent assay (ELISA) for interleukin 1β (IL-1β), CXCL1/CINC-1, IL-6 (Mouse Quantikine or Rat Duoset kit; R&D Systems, Minneapolis, MN), and myeloperoxidase (MPO) (Mouse and Rat MPO ELISA kit; HyCult biotech, Plymouth Meeting, PA).
Real-Time RT-PCR–Based Standard Curve for Human-Specific Glyceraldehyde 3-Phosphate Dehydrogenase (hGAPDH)
A standard curve was generated by adding serial dilutions of hMSCs to rat tissue as described previously . Briefly, 10–100,000 hMSCs were added to a deepithelialized rat cornea. Following RNA extraction (RNeasy Mini kit; Qiagen), cDNA was generated by reverse transcription (SuperScript III; Invitrogen) using 1 μg total RNA. hGAPDH primer and probe (TaqMan Gene Expression Assays ID, GAPDH HS99999905_05) were used. The standard curve was made based on hGAPDH expression from a known number of hMSCs added to one rat cornea and the values were normalized to total eukaryotic 18s rRNA (TaqMan Gene Expression Assays, 4352930E).
Coculture of hMSCs and Human Corneal Epithelial Cells
Primary human corneal epithelial progenitor cells (hCECs) were purchased from CELLnTEC (HSCEP05; Bern, Switzerland). The cells at passage 3 were plated at 4,000 cells per square centimeter in the bottom well of six-well transwell coculture system (Corning Incorporated, Corning, NY), and cultured in CnT-20 PCT Corneal Epithelium Medium (CELLnTEC, Bern, Switzerland) for 3 days. The identity of the corneal epithelial cells was confirmed with immunocytochemistry for p63 and CK3 using goat polyclonal anti-human p63 (sc-25039, Santa Cruz Biotechnology) and goat polyclonal anti-human cytokeratin 3 (sc-49179, Santa Cruz Biotechnology, Santa Cruz, CA) as primary antibodies. Either hMSCs or human adult dermal fibroblasts (hFbs; ScienCell Research Laboratories, Carlsbad, CA) were separately plated at 1,000 cells per square centimeter on the upper inserts of six-well transwell coculture system and cultured in CCM with 16% FBS separately from hCECs for 3 days. The hCECs were injured with 20% ethanol for 60 seconds and washed with PBS three times. The insets with hMSCs or hFbs were then inserted for transwell culture with hCECs and cocultured for 1-day. RNA was extracted from hMSCs or hFbs (RNeasy Mini Kit; Qiagen, Valencia, CA) and assayed for TSG-6 by real-time RT-PCR. The assay was performed in three independent biological samples per each group. For one biological sample, the real-time RT-PCR was performed in triplicate technical replicates.
Comparisons of parameters among the groups were made using the one-way analysis of variance or two-tailed Student's t test using SPSS software (SPSS 12.0, Chicago, IL). Differences were considered significant at p < .05.
Both i.p. and i.v. Injection of hMSCs Suppressed Corneal Inflammation in Rats
To test the therapeutic effects of systemically administered hMSCs, we induced corneal inflammation by applying 100% ethanol for 30 seconds to the rat corneal surface followed by removal of both the corneal epithelium and limbal epithelium containing the limbal stem cells. Immediately after injury, hMSCs (1 × 107 cells in 1 mL HBSS) were administered by either i.p. or i.v. injection. Both i.p. and i.v. injection of hMSCs significantly reduced the development of corneal opacity 1 or 3 days after the injury as scored by clinical grade [20, 22] (Figs. 1A, 1B, 2A, 2B). In addition, infiltration of neutrophils into the cornea was markedly decreased as measured by the amount of MPO (Figs. 1C, 2C), an enzyme stored in cytoplasmic azurophilic granules of neutrophils and released extracellularly by activated neutrophils . We recently demonstrated that the amount of MPO in the cornea following chemical and mechanical injury correlates with clinical grade of corneal opacity  and provides a quantitative and reproducible indication of corneal inflammation. The production of proinflammatory cytokines and chemokines was also significantly decreased in the cornea as measured by RT-PCR and ELISA (Figs. 1D–1F, 2D, 2E). The expression of the proangiogenic and profibrotic molecule, tenascin C, was also significantly reduced in the MSC-treated group (Figs. 1G, 2F).
Systemically Administered MSCs Did Not Engraft into the Cornea
To test whether hMSCs reduced corneal inflammation by engrafting in the cornea, we assayed for hMSCs in the rat cornea. We capitalized upon the species differences in our model to detect hGAPDH gene expression as mRNA by real-time RT-PCR as described previously . A standard curve was prepared by adding hMSCs to one rat cornea that was excised following injury. The standard curve was linear with the sensitivity to detect 10 hMSCs per one rat cornea which amounted to 0.0001% of injected cells (1 × 107 hMSCs) (Fig. 3; Table 1). We assayed the rat cornea 1-day and 3 days after injury and treatment with hMSCs (i.p. or i.v.). Expression of hGAPDH within the detection limits of the standard curve was not found at either 1-day or 3 days following either i.p. or i.v. injection of hMSCs (Fig. 3; Table 1). Therefore, the results suggested that the MSCs reduced inflammation in the cornea from a distance without reaching the cornea by being activated to secrete therapeutic factors such as TSG-6 .
Table 1. The standard curve for the number of human mesenchymal stem cells in one rat cornea based on the expression of hGAPDH relative to total eukaryotic 18s ribosomal RNA
MSCs Were Activated In Vitro to Secrete TSG-6 in Response to Injury Signals from Corneal Cells
We next tested the hypothesis that signals from injured corneal cells would activate hMSCs to express TSG-6. We cocultured hMSCs with human corneal epithelial cells (hCECs, Fig. 4A) injured with ethanol (20% ethanol, 1 minute) in a transwell system (Fig. 4B). Human skin fibroblasts were used as control cells. TSG-6 expression was increased approximately 13-fold in hMSCs cocultured with injured hCECs compared with hMSCs cultured alone, whereas TSG-6 was increased about sixfold in hMSCs cocultured with uninjured hCECs (Fig. 4C). TSG-6 expression was significantly higher in hMSCs cocultured with injured hCECs than hMSCs cocultured with undamaged hCECs. In contrast, there was no significant upregulation of TSG-6 expression in human skin fibroblasts by coculture under the same conditions with damaged hCECs.
MSCs with TSG-6 siRNA Knockdown Did Not Reduce Corneal Inflammation
To further evaluate the role of TSG-6 in the action of hMSCs, we knocked down the expression of TSG-6 in hMSCs by transient transfection with a TSG-6 siRNA. The knockdown efficiency was approximately 85% by RT-PCR (Fig. 5C). There was no significant reduction in corneal opacity and inflammation in the mice receiving hMSCs with TSG-6 siRNA (Fig. 5A, 5B). However, both inflammation and opacity were significantly decreased in the mice receiving hMSCs with scrambled siRNA controls (Fig. 5A, 5B). The data indicated that TSG-6 was an important mediator for the action of MSCs, and the results were consistent with the hypothesis that TSG-6 secretion by hMSCs was required for hMSCs to reduce corneal inflammation.
Both i.v. Injection and Topical Administration of TSG-6 Reduced Corneal Inflammation
Next, we tested the hypothesis that systemically administered rhTSG-6 could reproduce the effects of systemically administered hMSCs by reducing inflammation in the cornea. We administered 30 μg of rhTSG-6 into a mouse via tail vein injection immediately after chemical and mechanical injury to the cornea. Expression of MPO and proinflammatory cytokines in the cornea was significantly reduced following rhTSG-6 injection (Fig. 5D–5F). Topical application of rhTSG-6 as eye drops was also effective in decreasing corneal opacity and inflammation (Fig. 5G), apparently because the loss of epithelial cells in the model facilitated the diffusion of the protein into the cornea.
We demonstrated that both i.v. and i.p. administration of hMSCs significantly suppressed inflammation in the cornea without the evidence of significant engraftment of the hMSCs. Also, we found that the anti-inflammatory protein TSG-6 was one of the therapeutic factors hMSCs secreted in response to signals from injured corneal epithelial cells. As rhTSG-6 was able to recapitulate the effects observed with hMSC treatment, and siRNA TSG-6 MSCs were not able to reduce corneal inflammation, we concluded that TSG-6 secretion by hMSCs was both necessary and sufficient to reduce corneal inflammation. These results are consistent with recent findings that i.v. injection of hMSCs secreted TSG-6 to improve cardiac function without long-term engraftment by decreasing inflammation in a mouse model of myocardial infarction . Our findings may explain the therapeutic effects of MSCs in the injured cornea observed by previous groups [17–19].
TSG-6 is a 35-kDa secreted protein composed mainly of contiguous hyaluronan-binding Link and CUB modules [24, 25]. The protein is not expressed in normal cells or tissues, but is expressed in many cells after exposure to TNF-α and other proinflammatory cytokines. TSG-6 was shown to have anti-inflammatory activities in several different models including arthritis, myocardial infarction, and chemical injury to the cornea [14, 16, 20, 26]. The anti-inflammatory activity of TSG-6 was largely attributed to its ability to bind to the fragments of hyaluronan, to inhibit components in the inflammatory network of proteases, and to suppress neutrophil migration into the site of inflammation [26, 27]. More recently, our group demonstrated  that in a mouse model for zymosan-induced peritonitis, hMSCs were activated by inflammatory signals to secrete the anti-inflammatory protein TSG-6 which interacts through the CD44 receptor on resident macrophages to decrease zymosan/TLR2-mediated nuclear translocation of the NF-κB complex. The negative feed-back loop created by hMSCs and TSG-6 attenuated the inflammatory cascade that was initiated by resident macrophages and then amplified by mesothelial cells and probably other cells of the peritoneum.
However, other anti-inflammatory actions of MSCs have also been observed. Experiments with a cecal ligation and puncture model  indicated that MSCs attenuated sepsis by MSCs being stimulated to secrete prostaglandin E2, and the prostaglandin E2 then reprogrammed resident macrophages to increase production of IL-10. Other reports suggested that the anti-inflammatory activity of MSCs was explained by the cells expressing IL-1 receptor antagonist  or the soluble TNF receptor 1 . Therefore, it is apparent that MSCs can produce a variety of anti-inflammatory factors in addition to TSG-6. The results here do not exclude the possibility that one or more of these factors augmented the effects of the TSG-6 produced by systemically infused hMSCs.
The anti-inflammatory actions of TSG-6 in the cornea can likely be attributed to one of the two mechanisms. One possibility is that TSG-6 reached the corneal surface to locally modulate inflammatory responses in a manner similar to that observed after an intracameral injection of TSG-6 . The limbus is a highly vascular area; therefore, TSG-6 secreted by hMSCs into the bloodstream or i.v. administration of rhTSG-6 could have delivered sufficient amounts of TSG-6 to the corneal and limbal area to reduce inflammation. The beneficial effects of topical TSG-6 that we observed in this study are also consistent with this possibility. Another possibility is that TSG-6 did not reach the cornea, but rather acted at sites distant from the cornea to decrease inflammatory cell recruitment to the cornea. This possibility is consistent with previous reports, which demonstrated that TSG-6 in the circulation suppressed neutrophil extravasation by inhibiting neutrophil migration or neutrophil-endothelial cell interaction [27, 32-34].
In conclusion, we observed that systemically administered hMSCs suppressed inflammation in the cornea without engraftment primarily by the cells being activated to secrete TSG-6. The findings that hMSCs reduced inflammation distant from the site of injury may have important implications for treating various diseases characterized by excessive inflammation including inflammatory corneal surface diseases [17–19], diabetes mellitus [35, 36], multiple sclerosis , acute lung injury [38, 39], and myocardial infarction [14, 40].
Systemically administered hMSCs, both i.v. and i.p., significantly reduced inflammatory damage to the cornea without engraftment and primarily by secretion of TSG-6.
We thank support from Laura Quinlivan for assistance with animal experiments. This work was supported in part by NIH grant R21EY020962.
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
D.J.P. is a member of the scientific advisory board of Temple Therapeutics LLC. The other authors indicate no potential conflicts of interest.