Biotechnol. Appl. Biochem. (2010) 56, 151–160 (Printed in Great Britain) doi:10.1042/BA20100151
151
A novel virally inactivated human platelet lysate preparation rich in TGF-β, EGF and IGF, and depleted of PDGF and VEGF Pierre-Alain Burnouf*†1 , Po-Kai Juan‡1 , Chen-Yao Su§, Ya-Po Kuo, Ming-Li Chou†, Ching-Hua Su†, Yu-Hung Tseng¶, Che-Tong Lin** and Thierry Burnouf*, **2 *Human Protein Process Sciences (HPPS), 59 800 Lille, France, †Department of Microbiology and Immunology, Taipei Medical University, Taipei, Taiwan, ‡Department of Dentistry, Taipei Medical University, Taipei, Taiwan, §Department of Dentistry, National Yang-Ming University, 11 221 Taipei, Taiwan, Institute of Oral Biology, National Yang-Ming University, 11 221 Taipei, Taiwan, ¶Gwo-Wei Dental Implant Center, 11 221 Taipei, Taiwan, and **College of Oral Medicine, Taipei Medical University, Taipei, Taiwan There is emerging interest in the use of standardized virally inactivated human platelet lysate preparations rich in GFs (growth factors) for cell cultures, cell therapy and clinical applications. In the present paper, we report a simple process to prepare a virally inactivated platelet lysate preparation rich in TGF-β1 (transforming growth factor-β1), EGF (epidermal growth factor) and IGF (insulin-like growth factor) and depleted of PDGF (platelet-derived growth factor) and VEGF (vascular endothelial growth factor). Apheresis platelet concentrates were treated by the S/D (solvent/detergent) viral inactivation procedure, then subjected to an oil extraction followed by adsorption with activated charcoal and finally sterilefiltered. The resulting preparation contained a mean of 368.4, 2.4 and 54.7 ng/ml of TGF-β1, EGF and IGF respectively. PDGF-AB and VEGF were essentially completely removed by the charcoal treatment. The mean albumin, IgG, IgM and IgA and fibrinogen contents were approx. 40.0, 8.5, 0.87, 1.66 and 2.65 mg/ml respectively, cholesterol and triglycerides were at 15 and 20.7 mg/ml respectively and TnBP (trin-butyl phosphate) and Triton X-45 were at 8.7 and 8.8 p.p.m. respectively. Supplementing MEM (minimum essential medium) with 1–10 % of this S/D-treated platelet lysate promoted the proliferation of MG63 and SIRC cell lines as well as, or better than, 10 % (v/v) FBS (fetal bovine serum), as based on the MTS [3-(4, 5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)2-(4-sulfophenyl)-2H-tetrazolium] assay. The process used to prepare such S/D-treated platelet lysates is easily scalable for industrial production. Our results open up the possibility to evaluate the potential of this new preparation for stem cell expansion and/or bone tissue engineering and regeneration.
Introduction As is evident by proteomics and functional analyses, human platelets contain a myriad of molecules exhibiting important
physiological functions [1]. These include the GFs (growth factors) that are stored in the α-granules [2]. PGFs (platelet growth factors) include three PDGF (plateletderived growth factor) isoforms (PDGF-AA, -AB and -BB), VEGF (vascular endothelial growth factor), TGF-β (transforming growth factor-β; TGF-β1 and TGF-β2), EGF (epidermal growth factor), FGF (fibroblast growth factor) and some IGF (insulin-like growth factor). There is increasing interest in the use of human PGFs both as therapeutic biological products in the field of regenerative medicine as well as for various applications in cell cultures and cell therapy as a replacement of FBS (fetal bovine serum). Such preparations need, however, to be standardized [3]. Currently, the major therapeutic applications of platelet lysates rich in GF are to stimulate bone regeneration in oral, maxillofacial, plastic and orthopaedic surgery, or to accelerate wound healing of soft tissues, in particular in the treatment of recalcitrant hard-to-cure leg ulcers [4–6]. For such clinical applications, a single-donor PC (platelet concentrate), or a platelet-rich-plasma donation, of autologous or allogeneic origin, is used as a topical product, as such or after activation by exogenous thrombin to induce the release and temporary entrapment of the GF into a fibrin-rich biomaterial, called platelet gel [7]. The GF-rich fraction can be applied on tissues, either alone or in combination with a carrier, such as collagen or Key words: fetal bovine serum (FBS), growth factor, lysate, platelet, solvent/detergent, stem cell. Abbreviations used: BMP, bone morphogenetic protein; EGF, epidermal growth factor; FBS, fetal bovine serum; GF, growth factor; HAV, hepatitis A virus; HIC, hydrophobic interaction chromatography; IGF, insulin-like growth factor; MEM, minimum essential medium; MSC, mesenchymal stem cell; MTS, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulfophenyl)-2H-tetrazolium; NAT, nucleic acid testing; PC, platelet concentrate; PDGF, platelet-derived growth factor; PGF, platelet growth factor; PMS, phenazine methosulfate; RBC, red blood cell concentrate; rh, recombinant human; S/D, solvent/detergent; S/D-PC, S/D-treated PC; TGF-β, transforming growth factor-β; TnBP, tri-n-butyl phosphate; VEGF, vascular endothelial growth factor. 1 These two authors contributed equally to the experiments. 2 To whom correspondence should be addressed (email
[email protected]). C
2010 Portland Press Limited
152
P.-A. Burnouf and others
ceramics [8,9]. The therapeutic interest in PGFs is illustrated by the recent development of various rh (recombinant human) PGFs. rh-PDGF-BB expressed by engineered yeast is licensed in some countries for the treatment of chronic neuropathic lower-extremity diabetic ulcers [10,11]. Its efficacy has been established in several clinical studies, but large doses and several weeks of application are often needed to achieve a cure [12,13]. Long-term use of rhPDGF-BB is suspected to increase the risk of cancer development in susceptible patients [14]. rh-BMP (bone morphogenetic protein)-2 and -7, which belong to the TGFβ superfamily, are produced from genetically engineered Chinese-hamster ovary cells. rh-BMP-2 and -7, delivered as absorbable collagen sponges, have been licensed as boneregeneration agents, for instance in the treatment of longbone fractures [15]. Applications for spinal fusion and craniofacial and periodontal surgery are being studied [15]. Interesting new developments in the use of platelet lysates have recently emerged. Platelet lysates can be used as supplement of growth medium for cell cultures, providing a source of GFs and vital nutrients to avoid the addition of FBS. Such an application has been demonstrated in particular for clinical-scale expansion of mesenchymal or haematopoietic stem cells [16–20]. Using human-derived platelet lysate preparations avoids the risks of immunological reactions and infectious disease transmission due to the xenogenic origin of FBS. In addition, as there is extensive demand for FBS, its supply is becoming scarce, supporting the development of alternative materials. At this moment, uncertainties still remain as to which type of GF mixture is best suitable for MSC (mesenchymal stem cell) expansion, depending upon the origin of the stem cells and the type of cell lineage differentiation targeted. For instance, whereas allogeneic pooled human serum and thrombin-activated platelet releasates were shown to be superior adjuvants in isolating and expanding human adipose-tissue-derived MSCs [21], allogeneic pooled freeze–thaw lysed human platelets were most suitable for expansion of MSCs derived from bone marrow [16]. Considering the interest in PGFs for diverse therapeutic applications, and recognizing that pooled industrial human materials used as supplements for MSC expansion requires viral inactivation [22,23], we have initiated a research programme aimed to develop easily scalable processes to prepare a range of virally inactivated PGF preparations. We have first shown that applying an S/D (solvent/detergent) viral inactivation treatment directly to PCs, as the first step of the manufacturing processes, leads to a several-fold enhancement of the release of the GF from the α-granules, as compared with traditional thrombin activation methods or repetitive freeze–thaw cycles, thereby contributing to increasing GF extraction yield [24]. We have also demonstrated that HIC (hydrophobic interaction C
2010 Portland Press Limited
chromatography) of S/D-treated platelets, which is a usual method to remove the toxic S/D agents from therapeutic virally inactivated pooled transfusion plasma [25], allows one to prepare a complex, functionally active, PGF mixture. We have shown that this GF mixture replaces FBS for the cultures of various cell lines [26] and for the expansion of adipose-tissue-derived MSCs [27]. By doing a cationexchange chromatographic fractionation of the SD-treated PC, a purified PDGF/VEGF fraction could be isolated, which could be useful for cell cultures or wound healing [28]. In the present study, we now report a simple and new process where activated charcoal is used instead of HIC to remove the S/D agents, yielding a new PGF mixture enriched in TGF-β, EGF and IGF but depleted of PDGF and VEGF.
Materials and methods Platelet material The PCs were obtained as before [26] by using the MCS+ apheresis procedure (Haemonetics Corporation, Braintree, MA, U.S.A.) and were donated by volunteer donors who provided their informed consent. Following the manufacturer’s instructions, whole blood was withdrawn through a venous catheter, with an intermittent flow, and immediately anticoagulated at a 1:9 ratio with citrate dextrose solution Formula A. The platelets were separated from red blood cells by continuous centrifugation, suspended in plasma and collected into a sterile plastic bag. Several separation cycles were performed until the total volume of PC collected reached between 225 and 315 ml and a predefined platelet yield target. PCs were stored under slow mixing at room temperature (22–24 ◦ C) and were processed within 24–72 h of collection. Cell count The content of red blood cells, platelets and leucocytes in the starting PC was determined using an ABC Vet automatic blood counter (ABX Diagnostics, Montpellier, France). Activated charcoal powders Bulk activated charcoal powders were obtained from 3M/Cuno (St. Paul, MN, U.S.A.) and Pall Corporation (SaintGermain-en-Laye, France). Processing of the PCs The processing method of the PC is in Scheme 1. Briefly, the PC (300 ml) was transferred into a plastic bag and incubated with a combination of 1 % (v/v) TnBP (tri-nbutyl phosphate; Merck, Darmstadt, Germany) and 1 % (v/v) Triton X-45 (Sigma–Aldrich), at room temperature for 1 h
Platelet growth factor lysate
Scheme 1
Preparation process of the PGF mixture enriched in TGF-β1 and depleted of PDGF-AB and VEGF
under constant gentle stirring (70 rev./min; FINEPCR, Seoul, South Korea) as described previously [26]. A 10 % (v/v) final concentration of soybean oil (Sigma) was then added and mixed first vigorously for 30 s and then at 70 rev./min with the S/D-PC (S/D-treated PC) for 15 min at room temperature. The oil/S/D-PC mixture was then decanted for 45 min to separate a clear oily phase (top), a turbid phase (intermediate) and a clear aqueous protein phase (bottom). The aqueous protein phase (approx. 270 ml) was recovered by gravity from the bottom of the bag and clarified by centrifugation at 6000 g for 10 min at 20–25 ◦ C to pellet any cell debris or insoluble material. After centrifugation, the supernatant (S/D-PC-O; approx. 265 ml) was recovered and mixed for 30 min at 70 rev./min at room temperature with activated bulk charcoal. The mixture was then centrifuged (6000 g for 10 min at room temperature) to pellet the activated charcoal and recover a clear supernatant (S/DPC-OC). Samples were taken along the process and frozen at − 80 ◦ C until analysis. The results presented are average values from five consecutive experiments.
Determination of the quantity of activated charcoal Preliminary small-scale experiments were performed, first, to determine the optimal quantity and type of activated charcoal required to remove TnBP and Triton X-45 and, secondly, to evaluate the impact of charcoal treatment on the GF and protein content and on cell growth stimulation. One PC was processed as described above. After oil extraction, four aliquots of 20 ml were treated with increasing amounts of activated charcoal (1.5, 3.0, 3.5 and 4 g respectively) for 30 min at room temperature and 70 rev./min using a rotating mixer. The mixtures were then centrifuged (6000 g for 10 min at room temperature) to pellet the activated charcoal and recover the clear supernatant, which was then frozen at − 30 ◦ C until assessment for GF content, S/D residual amounts and cell growth stimulation capacity.
GF and P-selectin assays PDGF-AB, TGF-β1, EGF, VEGF and IGF-1 were measured by ELISA using Quantikine® kits (R&D Systems, Minneapolis, C
2010 Portland Press Limited
153
154
P.-A. Burnouf and others
MN, U.S.A.) as described previously [24,26]. P-selectin was determined by ELISA using Quantikine kits (R&D Systems) as described in the manufacturer’s instructions. Proteins and lipids assays Total protein, albumin, total cholesterol and triglycerides were determined in the starting PC, after the oil extraction and after activated charcoal adsorption, by using a Roche P800 analyser. IgG, IgM and IgA were measured using fully automatic latex-enhanced immunoturbidimetric assays on a Cobas Integra 800 instrument (Roche Diagnostics, Mannheim, Germany). SDS/PAGE SDS/PAGE was performed under non-reduced and reduced conditions as described previously [29,30] using 4– 12 % gradient gels, reagents and electrophoretic systems from Invitrogen (Carlsbad, CA, U.S.A.). Molecular mass assessment was done using Novex® Sharp Pre-Stained Protein Molecular Weight Standard (Invitrogen). TnBP and Triton X-45 determination The extraction of the samples for determination of residual TnBP and Triton X-45 was done as described previously [26]. The quantification of TnBP was performed by GLC (GC-17A; Shimadzu, Tokyo, Japan) and flame ionization detection and that of Triton X-45 by HPLC (Shimadzu) using a LiChrospher® 100 RP-18 column (interior diameter, 4 mm; length, 250 mm; part no. 724964; Merck) respectively. Cell cultures The capacity of the S/D-PC-OC to stimulate in vitro cell growth and replace FBS was evaluated using the following cell lines: (i) human osteoblasts MG63 [ATCC CRL 1427, BCRC (Bioresource Collection and Research Center), Hsing-chu, Taiwan] and (ii) Statens Serum Institute rabbit corneal epithelial cells (SIRC; ATCC CCL-60, BCRC). Maintenance of the cell lines was at 37 ◦ C in a controlled atmosphere containing 5 % CO2 . They were cultured at a density of 2 × 103 cells per well in flatbottomed 96-well plates (Greiner bio-one, Tokyo, Japan) as described previously [26] and using MEM (minimum essential medium; Gibco, Invitrogen) containing 2.2 g of NaHCO3 , 0.1 mM non-essential amino acids, 1 mM sodium pyruvate, 2 mM L-glutamine, 100 units/ml penicillin and 100 μg/ml streptomycin, and supplemented with 10 % (v/v) FBS (Gibco, Invitrogen). For the experiments, cells were first allowed to adhere for 18 h, and were then starved for 6 h in a serum-free medium before testing the different GF supplements. After two washes with the serum-free medium, cells were cultured for up to 5 days in the medium C
2010 Portland Press Limited
supplemented with various concentrations (1–10 %) of the S/D-PC-OC (corresponding to a final concentration range in TGF-β between 0.31 and 3.1 ng/ml respectively). As controls, cell cultures without any protein/GF supplement or supplemented with 10 % FBS were also carried out in each run. MTS assay Viable cells in proliferation were determined at 6 days of culture using the MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium], inner salt (Promega, Madison, WI, U.S.A.), according to the manufacturer’s instructions. This assay uses solutions of MTS and an electron coupling reagent PMS (phenazine methosulfate). MTS is bioreduced by cells into a formazan product that is soluble in tissue culture medium. The conversion of MTS into formazan is induced by dehydrogenase enzymes present in metabolically active cells. The attenuance of formazan at 490 nm was measured from the 96-well assay plates and is directly proportional to the number of living cells in culture. After preparation, the solutions were kept in light-protected tubes at − 20 ◦ C. MTS and PMS detection reagents were mixed, using a ratio of 20:1 (MTS/PMS), immediately prior to addition to the cell culture at a ratio of 1:5 (MTS+PMS/medium). Statistical analysis Results are reported as means + − S.D. Statistical comparisons were made with a two-tailed paired Student’s t test. Differences were considered significant if P < 0.05.
Results Cell count 6 The starting PC had a mean of (1216 + − 188.6) × 10 6 platelets/ml [range: (896–1388) × 10 /ml]. The mean leucocyte and red blood cell counts were (0.32 + − 0.10) (range: 0.20–0.4) × 106 cells/ml and (0.04 + − 0.01) (range: 0.02–0.06) × 109 cells/ml respectively. Blood cells were undetectable in all fractions subsequent to the S/D treatment. Determination of the optimal ratio of activated charcoal Various activated charcoals were first evaluated for their capacity to remove the S/D agents. Good preliminary results were obtained using 3M/Cuno activated charcoal porosity 5/grade 5 (Cuno R55) and with Pall activated charcoal AKS grade 6. The treatment also dramatically reduced the content of PDGF-AB and VEGF. Further tests were
Platelet growth factor lysate
Figure 1 Growth factor content: impact of charcoal content Content of PDGF-AB (A), VEGF (B), EGF (C), IGF-1 (D) and TGF-β1 (E) in S/D-PC-O (a) and after treatment with 1.5 g (b), 3 g (c), 3.5 g (d) and 4 g (e) of activated charcoal per 20 ml of S/D-PC-O. Results are means + − S. D. for four independent experiments. N.D, not detectable.
performed where increasing amounts of Cuno R55 activated charcoal (1.5, 3, 3.5 or 4 g) were mixed with 20 ml of S/DPC-O. Residual TnBP was
