Placenta Derived Adherent Cell (PDAC) Interaction and Response on Extracellular Matrix Isolated from Human Placenta

The placenta is a rich source for proteins such as albumin,¹ growth factors,^2,3 and extracellular matrix proteins such as collagen⁴ and stem cells.⁵ This article describes the methods for the isolation of the whole placenta-derived extracellular matrix (pECM) tissue from the human placenta. The composition of the isolated matrix can be modified with slight changes in isolation methods.As the pECM is insoluble in water, this pECM material can be molded into water and cell culture media stable sheets and 3-dimensional (3-D) structures. Further, pECM-cell binding studies have been conducted to examine the cellular response of an adherent, pluripotent progenitor cell (PDACs) population isolated from human placenta (described as CD200+, CD105+, CD73+, CD34-, and CD45-).⁵
Based on the results presented in this work, one could envision a combination product composed of the pECM with PDACs. Such a 3-D mold may be useful for the treatment of nonhealing deep wounds, such as diabetic ulcers. The pECM would provide a support matrix for cells to attach to, proliferate and initiate cellular signaling and the PDACs would deliver key cytokines and chemokines to re-invigorate the wound healing process in nonhealing wounds.⁵

Methods
Isolation of pECM. The pECM was isolated from frozen placentas, qualified as described previously.^7,8 A frozen human placenta was thawed in 0.5 M sodium chloride, ground in an Anvil meat-grinder (Model MIN0012), and repeatedly washed in 0.5 M sodium chloride and water mixture using a New Brunswick Scientific (Model C25; Edison, NJ) incubator shaker at 23^°C. This was followed by a detergent wash with either 1% SDS or deoxycholic acid. The blood-free placental tissue was treated with 0.1–0.5 N sodium hydroxide (3–24 hours), followed by rinsing with PBS to neutralize the pH. The material produced was a stable paste and stored at 4^°C.
Biochemical analysis. To determine the biochemical composition of the isolated pECM,a sample of known dry weight was freeze-dried and the dry weight was determined.The pECM was solubilized by either dissolving in 10 mM HCl at 100^°C or by treating with pepsin (1 mg/g) in 10 mM HCl at 23^°C for 18 hours.The tissue dissolved in 10 mM HCl was used to determine content of fibronectin, laminin, glycosaminoglycans (GAGs), and elastin. The pepsin-solubilized tissue was used to determine collagen content using the SIRCOL dye based assay kits from Accurate Chemical and Scientific (Westbury, NY). Fibronectin and laminin concentrations were determined using a sandwich ELISA obtained from Takara Bio USA (Madison,Wis). Elastin and GAG content were determined using the FASTIN and BLYSCAN dye based assay kits (Accurate Chemical and Scientific).For collagen I, III, and IV, sandwich ELISAs were developed using primary and HRP conjugated secondary antibodies from Rockland Immunochemicals, Inc. (Gilbertsville, Pa).
Preparation of pECM constructs. To prepare sheets of this pECM, an even layer of hydrated pECM paste was sandwiched between 2 medical grade Tyvek sheets from Tolas Health Care (Feasterville, Pa).This construct was loaded into a gel drier (BioRad Model 583) and vacuum was applied overnight at 23^°C until the pECM film was dry. Sheets were cut to an appropriate size for cell culture studies.To prepare 3-D structures of the pECM, the pECM paste was filled into various molds and freeze-dried using a benchtop Labconco Freezone 4.5 freeze dryer (Labconco,Kansas City,Mo).To study the stability of the pECM sheets and 3-D molds in cell culture media or water.The constructs were incubated at 37^°C up to 1 week in water, saline, or cell culture media.
Cell culture. Human placentas, following birth of normal full term infants,were procured under informed consent of donors. Pluripotent progenitor cells were isolated from the placenta by digestion of the tissue with trypsin and collagenase and placed in growth media containing FCS,growth factors,and 1% gentamicin in 5% CO₂ in a humidified incubator at 37^°C. Cells were detached with 0.25% trypsin-EDTA replated at 4 x 103 cells/cm². PDACs derived by the methods described above showed a phenotype CD200+, CD105+, CD10+, CD73+, CD34, CD45-. For cell culture studies, the placental- derived adherent cells or PDACs were subcultured in 60% lowglucose DMEM (Invitrogen, Carlsbad, Calif), 40% MCDB- 201 (Sigma, St. Louis, Mo), 2% fetal bovine serum (Hyclone, Logan, Utah), 1x insulin-transferrin-selenium supplement (Invitrogen), 0.02% linoleic acid/bovine serum albumin (Sigma, St. Louis, Mo), 10 ng/mL epidermal growth factor (Sigma, St. Louis, Mo), 10 ng/mL platelet-derived growth factor (R&D Systems, Minneapolis, Minn), 0.05 mM dexamethasone (Sigma, St. Louis, Mo), 0.1 mM ascorbic acid 2-phosphate (Sigma, St. Louis, Mo), and 100U/penicillin 1000U/streptomycin (Invitrogen, Carlsbad, Calif). Pluripotent progenitor cells (30,000 per well) were seeded onto pECM films that had been positioned into 24 multi-well cluster plates. Pluripotent progenitor cells were also seeded at equivalent density on Labtek chamber slides (Nalgene Nunc International, Rochester, NY) that had been pre-coated with collagen (Inamed, Fremont, Calif). Cells were incubated at 37^°C for 3 and 48 hours and processed for immunofluorescence microscopy.
Immunofluorescence microscopy. After 3- or 48- hour incubation with pECM films, PDAC-pECM constructs were fixed with 3.7% formaldehyde for 10 minutes and permeabilized with 0.5% Triton-X 100 for 20 minutes. Pluripotent progenitor cells were incubated with AlexaFluor 488-conjugated phalloidin to visualize Factin. For fibronectin staining, samples incubated with a rabbit, anti-human fibronectin antibody (Sigma, St. Louis, Mo) in blocking buffer (3% bovine serum albumin/1x phosphate-buffered saline) for 1 hour,washed with phosphate- buffered saline, and further incubated with the AlexaFluor 594-conjugated anti-rabbit antibody in a blocking buffer for 30 minutes. Samples were again washed with phosphate-buffered saline, mounted on slides, and observed with a fluorescent microscope.
Cytokine secretion analysis. Media samples (100 uL) were sampled at 4 time-points: 0, 3, 24, and 48 hours. Samples were diluted into 1-mL PBS and analyzed for the presence of cytokines using a 25-plex antibody array from Biosource (Camarillo, Calif) on a Luminex-100 instrument (Luminex Corp, Austin, Tex). The concentration of each cytokine was calculated from a standard plot of known concentrations of cytokines.

Results

Isolation of pECM. The dry weight of a typical placenta was about 30 g, corresponding to a wet weight of about 300 g per placenta. As shown in Scheme 1, the alternating salt and water wash steps (osmotic shock) can be used to remove a considerable amount of nonextracellular matrix tissue; the final residual weight was about 10 g. The use of a combination of solubilization using sodium hydroxide (NaOH) and detergent resulted in a final residual weight to 6 g. These studies indicate that time of exposure to NaOH and concentration affected the total mass of pECM isolated from the placenta (data not shown). As shown in Scheme 1, variations of the detergent and NaOH wash steps were utilized to generate 5 different forms of the pECM.Typically,between 6- g and 10-g of pECM were isolated from a single placenta. This material could easily be filled into syringes as a paste or molded into a variety of shapes as shown in Figure 1.

Biochemical composition of pECM.Biochemical analysis of the 5 variations of pECM showed that they were composed essentially of collagens;Type I is the major collagen component (~73 %); Type III (~3%) and Type IV (~6%) are minor components.The other important protein found in the pECM was elastin. Elastin represented 4% of the total dry weight (Table 1). In the case of pECM- 5,which was generated without the use of NaOH, a higher elastin content of 12% was observed. Glycosaminoglycans were identified in all 5 pECMs; the percent dry weight was unaffected by the use of NaOH in the isolation methods.The presence of fibronectin and laminin was highly sensitive to the use of NaOH. The important adhesion proteins fibronectin and laminin did not survive the NaOH treatment and could not be found in pECMs 1 through 4. However, pECM-5, which was isolated without the use of NaOH, had a composition that was richer in the adhesion proteins (Table 1).

Cell binding studies. After seeding for 3 hours, similar levels of PDAC attachment were observed on all pECMs (numbers 1–5); the levels of PDAC binding to pECMs were slightly less than that observed on purified collagen. Immunostaining of PDACs for fibronectin at this time revealed abundant intracellular staining, with no detectable extracellular fibronectin (data not shown). As shown in Figure 2, by hour 48 of culture, PDACs increased in number and adopted similar well-spread morphologies on purified collagen,pECM-2,and pECM-4. In contrast, PDACs did not thrive on pECM-1 or pECM-3.Not only were fewer cells observed, but also their morphologies were rounded and not well spread. Pluripotent progenitor cells on pECM-5 appeared more elongated and polarized than PDACs on other pECMs and collagen. Immunostaining for fibronectin at the 48-hour time point revealed an extensive network of extracellular fibronectin matrix fibrils on pECMs numbers 1–4.These fibronectin matrix fibrils were assembled by PDACs; the control pECM-only samples (ie, samples in which PDACs were not cultured on pECM) did not show evidence of fibronectin fibrils. It is also known from the literature that fibronectin deposition into the matrix is a celldependent process and does not occur spontaneously.⁹ In contrast to pECMs numbers 1–4, pECM-5 and collagen did not support fibronectin matrix assembly by PDACs and no extracellular fibrillar fibronectin was detected on these surfaces.

Cytokine array studies. The secretion of key cytokines/chemokines from PDACs resulting from the binding and proliferation to the pECM was investigated. Cytokine secretion on pECM was compared to that from PDACs incubated on tissue culture treated cell culture plates using a 25-multiplex cytokine array, which includes several interleukins and cytokines (Table 2). Of the 25 cytokines studied, increased secretion of 3 cytokines when on the pECM sheets (compared to tissue culture treated plates or TCP) was observed. These include IL-6, IL-8,and monocyte chemoattractant protein- 1 (MCP-1). Figure 3 (A-C), shows a time-dependent increase in cytokine secretion (IL-6, IL-8, and MCP-1) by PDACs on the 5 pECM constructs; data are normalized to cell numbers present on pECM sheets. Interestingly, pECM-5 was determined to be an anomaly in that there was no increase in MCP-1 secretion. As previously shown, pECM-5 did not support the expression of fibronectin, quite unlike pECM-1 through pECM-4.Taken together, these data suggest a possible change in cellular behavior or signaling when situated on this particular pECM. It is interesting to note that pECM-5 was the only matrix generated without the use of NaOH and had a biochemical composition that maintained the 2 key cell adhesion proteins—fibronectin and laminin.

Discussion
These studies were conducted with the intention of designing a cell-matrix combination product with variable 3-D space filling characteristics for deep wounds such as diabetic ulcers. Such a product would be a tissue replacement product that could potentially assist in restoring the normal wound healing process in such wounds.¹⁰ Diabetic and venous ulcers are characterized by a dysfunctional extracellular matrix that does not support normal wound healing.¹¹ As demonstrated here, the pECM may act not only as a structural support for cells but also as a source of biochemical signals that are critical to the normal wound healing.¹² As a combination product, the PDAC cellular component would provide a newly secreted fibronectin matrix that is capable of guiding cell migration, proliferation, and differentiation.¹³ In addition, the PDAC would also deliver key “wound-healing” cytokines at the wound site.¹⁴
The methods described for isolating the pECM are scalable and can operate in an aseptic mode providing a raw material that is both sterile and endotoxin free (data not shown here). More importantly, the amount of extracellular matrix isolated from a single placenta is substantial (6 g–10 g) allowing manufacturing of reasonable quantities of constructs (or units) from a single placenta.
The isolated pECM contains key structural proteins such as collagens and elastin and cell adhesive proteins such as fibronectin and laminin. Results described here show that the isolated pECM allows cellular attachment and proliferation. PDACs show cytoskeletal spreading and an ability to assemble a fibronectin matrix. In a wound environment such material would be able to provide a conduit for infiltrating cells as well as provide the biochemical cues, which induce signal transduction events resulting in the deposition of a matrix in the wound space.
On the pECM, the PDACs secrete key cytokines including IL-6, IL-8, and MCP-1.These cytokines play intricate roles in wound healing processes as MCP-1 and IL-8 belong to a family of chemotactic cytokines or chemokines, which are involved in recruiting leukocytes to the wound site, promoting production of extracellular matrix and regulating repair processes.¹⁵ In rat models, MCP-1 has been examined for its ability to accelerate wound healing processes by improved re-epithilialization.^16–17 IL-8 has also been shown to enhance the overall re-epithelialization process of wound healing in a guinea pig model.¹⁸ IL-6 is a pro-inflammatory cytokine that participates in early immune response by recruiting white blood cells to wound sites.¹⁹ Cutaneous wound healing is significantly delayed in IL-6 knockout mice²⁰—evidence that argues for the importance of IL-6 in accelerating wound healing. By secreting these cytokines at a wound site, the PDACs might accelerate wound healing in a nonhealing wound.

Conclusion
To date, there are no wound-filling 3-D matrix products with therapeutic cells for the repair of nonhealing and deep diabetic wounds. The product conformation described in this study is unique in that it is composed of a human matrix with human cells. More detailed in-vivo studies are needed in order to better understand the wound healing properties of the pECM-PDAC combination product.