Human bone-marrow mesenchymal stem cells have an important role in the repair of musculoskeletal tissues by migrating from the bone marrow into the injured site and undergoing differentiation. We investigated the use of autologous human serum as a substitute for fetal bovine serum in the ex vivo expansion medium to avoid the transmission of dangerous transfectants during clinical reconstruction procedures.
Autologous human serum was as effective in stimulating growth of bone-marrow stem cells as fetal bovine serum. Furthermore, medium supplemented with autologous human serum was more effective in promoting motility than medium with fetal bovine serum in all cases. Addition of B-fibroblast growth factor to medium with human serum stimulated growth, but not motility. Our results suggest that autologous human serum may provide sufficient ex vivo expansion of human bone-marrow mesenchymal stem cells possessing multidifferentiation potential and may be better than fetal bovine serum in preserving high motility.
Human bone-marrow mesenchymal stem cells (BMSC) are multipotent cells which are present in adult marrow and can replicate as undifferentiated cells. They also have the potential to differentiate with various lineages of mesenchymal tissue, including bone, cartilage, fat, tendon, muscle, ligament and marrow stroma.1 The cells which are expanded in culture have already been used in clinical practice for regeneration of mesenchymal tissue. For example, BMSC combined with porous hydroxyapatite materials have been used in the treatment of patients with large bone defects, and have been shown to be useful for reconstruction of bone.2,3 Alternatively, autologous culture-expanded transplantation of BMSC has been used for patients with defects of cartilage in osteoarthritic knees.4 Hentz and Chang5 stated that surgeons and patients can look forward to new techniques involving BMSC which offer the potential for substantial advances in reconstructive surgery. However, they have also pointed out that tissue engineering in reconstructive surgery is not yet ready for widespread use.5
One of the most important points to be resolved is the development of an ideal culture medium for ex vivo expansion which should not be supplemented with bovine serum. For the ex vivo expansion of BMSC, most previous transplantation studies, both experimental1,6 and clinical2–4 have used cultures in a medium containing fetal bovine serum (FBS). For widespread clinical application, however, the use of FBS must be minimised because it has been implicated as a cause of a delayed hypersensitivity reaction7 and is a potential vector for transmission of prions.8 To solve this problem, autologous human serum could possibly be used instead. It has been used for the treatment of cartilage defects of the knee, with excellent results.9,10 However, there are few reports which compare FBS with human autologous serum in regard to their growth stimulatory abilities and, therefore, the effect of autologous serum on the growth of human BMSC is controversial.11,12 Furthermore, it should be noted that the use of autologous human serum as a substitute for FBS may also influence adhesion and motility.
Cell motility in response to extracellular signals plays a central role in diverse biological phenomena including spermatogenesis, wound healing, inflammation, angiogenesis and metastasis.13,14 BMSC have an important role in the repair of musculoskeletal tissues by migrating from the bone marrow into the injured site and undergoing differentiation.6 Movement may be a fundamental initial step before differentiation into cells suitable for repair of the damaged mesenchymal tissue. Possible factors which induce stem cells to migrate may be tissue-specific or generic to injury.15 Therefore, an ideal culture medium should maintain high cell motility as well as the ability to grow and undergo multidirectional differentiation. We have, therefore, evaluated the growth response of human BMSC and their motility in autologous serum.
Materials and Methods
Human BMSC were obtained from the iliac crest of ten consecutive patients undergoing reconstructive surgery with autologous bone graft. Patients with malignant tumours or severe infectious conditions were excluded. The details of the donors are given in Table I⇓. Cells from donor 1 were used for differentiation of chondrocytes in human autologous serum, as described by Yoo et al.6 Cartilage matrix was found ex vivo around the chondrogenic-differentiated human BMSC (data not shown). Human BMSC derived from donors 8 and 9 were aggregated accidentally at the primary plating and were abandoned. The cells obtained from the remaining seven donors were successfully used for the growth and motility assay (Table II⇓). The local ethics committee approved the study and each individual participating in the study gave informed consent.
Removal of human BMSC and ex vivo expansion.
Human BMSC were obtained and expanded, as described previously.6 Briefly, an 18-gauge needle was used to penetrate the cortex of the bone, and 5 ml of bone marrow were aspirated into a syringe containing 100 units of heparin. We added 10 ml of Dulbecco’s modified Eagle medium (DMEM; Gibco Brl Co., New York, New York) to the aspirate. A cell pellet was produced by centrifugation, and was cultured with medium supplemented with 10% of FBS or 10% of autologous patient’s human serum (HS) in two 100-mm dishes. They were grown at 37°C in 5% CO2. The human BMSC obtained from patients were identified as ‘STM’ and designated in numerical order. The medium was changed every four days. During primary culture (P1), the cells formed colonies which were removed after they had expanded to cover approximately 90% of the plate. They were then trypsinised, counted and replated at a density of 1 x 105 per 100-mm dish. The secondary culture was designated as P2.
After the serum had been obtained from each patient, it was heat-inactivated immediately and then stored at −40°C. The sera were identified as ‘STS’ and numbered according to the STM number.
Cell-growth ability assay.
An MTT assay was used to investigate cell proliferation as described previously.16 The linearity of the MTT assay had been previously confirmed in murine fibrosarcoma.16 Viable cells of both adherent and floating cells were evaluated in the modified MTT assay. Briefly, cells harvested with trypsin/EDTA (Gibco Brl Co) were plated into flat-bottomed 96-well plates (Corning Glass Works, Corning, New York) in 100 μl of medium containing 10 000 cells per well. At the time indicated, the culture wells were pulsed with 10 μl of MTT solution (2 mg/ml in phosphate-buffered saline (PBS) at pH 7.2) for three hours at 37°C. To each well was added 100 μl of 10% sodium dodecyl sulphate-0.01 N HCI and the cells were then incubated for 12 hours at 37°C. The plates were read on a Tohoh MPR-A4 microplate reader (Tohoh Inc, Tokyo, Japan), using a test wavelength of 570 nm and a reference wavelength of 620 nm. The mean and sd of each group were obtained from four replicates.
Motility was evaluated by a phagokinetic assay, as described previously.17 Briefly, uniform carpets of gold particles were prepared on coverslips coated with bovine serum albumin. Colloidal gold-coated cover-slips were placed in 35 mm tissue-culture dishes containing 10% FBS or HS, and 3000 cells in suspension culture were added to the plates. After 24 hours, the phagokinetic tracks were visualised using dark-field illumination in a Nikon inverted microscope. The area cleared of gold particles by at least 50 cells was measured using NHI image 1.62.
The MTT and phagokinetic assays were analysed by an unpaired Student’s t-test or the Dunn-Bonferroni method when two or more groups were compared respectively. To evaluate the relationship between the response of motility and growth to HS-supplemented cultivation, the ratio of both activities in HS against those in FBS was assessed by linear regression analysis. Significance was set at p < 0.01.
Linearity of the MTT assay with human BMSC.
For the first time, the linearity of the MTT assay used with human BMSC was evaluated. The cells were plated out in dilutions in 0.1 ml growth medium in 96-well flat-bottomed plates, starting at 106 cells/well. The MTT solution was added to all wells 30 minutes after plating. As shown in Figure 1⇓, the absorbance was directly proportional to the number of viable cells, as observed for fibrosarcoma.16
Growth response of FBS-primary cultured human BMSC to autologous human serum.
We isolated human BMSC in the presence of FBS from five samples (3, 5, 6, 7 and 10; Table II⇑). In four of these (3, 6, 7 and 10) P1 cells were used for the MTT assay. Since the experiments could not be performed until a sufficient number of cells had been grown, the time for expansion varied between 23 and 90 days. All cells showed substantial growth at various rates and a representative growth pattern is shown in Figure 2⇓. Autologous HS significantly augmented the growth of cells from sample 3 as compared with FBS by 1.5-fold at day 6. Similar augmentation was observed in the experiment performed one week later with the same cell line (data not shown). This augmentation by autologous serum was not observed in all cells (Table III⇓). In STM cells from sample 6, HS suppressed growth as compared with FBS, although the difference was not significant. In STM sample 5 cells, P2 cells were used for the MTT evaluation. Again autologous HS stimulated the growth of the cells significantly as compared with FBS in cells pre-cultured in both FBS and HS (Table III⇓). Basic fibroblast growth factor (FGF) is a potential growth stimulator of human BMSC, preserving the multidifferentiating ability to various mesenchymal tissues.18 Addition of bFGF to medium with HS stimulated cell growth significantly in the two human BMSC evaluated (STM 3 and 5).
Growth response of HS-primary-cultured human BMSC to autologous HS.
Samples 2 and 4 were cultured in autologous HS at primary culture. The growth pattern of the sample 4 cells with pre-culture is shown in Figure 3⇓. Of cells pre-cultured in both FBS (Fig. 3a⇓) and HS (Fig. 3b⇓) those in HS grew more rapidly than the cells in FBS. However, in sample 2 no significant difference in growth occurred between cells grown in HS and FBS (data not shown). The addition of bFGF significantly stimulated the two human BMSC cultured in autologous HS.
In vitro morphology of human BMSC.
Cell morphology was investigated in two stem-cell lines (STM 2 and 3). Both showed similar findings, although they were prepared with different serum types at preparation of the primary culture (Fig. 4⇓). Cells in the presence of FCS showed large, well-spread cells with multiple cytoplasmic processes (Fig. 4a⇓). By contrast, those cultured in the presence of autologous HS had a different morphology and were smaller and more spindle-shaped (Fig. 4b⇓). Addition of bFGF to the HS-supplemented culture did not change the cell morphology substantially (Fig. 4c⇓), although the cells seemed to overlay each other.
Motility of human BMSC in autologous HS.
A morphological change in HS, revealing a more spindle-shaped morphology, prompted us to investigate changes in motility since cell shape is closely correlated with cell motility.19 All the cells showed a higher motility in HS than in FBS, with a significant difference regardless of the pre-culture medium(Fig. 5⇓). As shown in Figure 6⇓, simple STM 2 cells moving in the presence of autologous HS had an 11.0-fold higher locomotor activity than those moving in the presence of FBS, with a highly significant difference (p > 0.001). By contrast with growth response, addition of FGF did not significantly stimulate the motility of this cell line (Fig. 6c⇓). Similarly none of the six cell lines of human BMSC examined STM (2, 3, 4, 5, 6 and 7) showed significant stimulation of motility when the cells were examined in the presence of HS (data not shown).
To evaluate the relationship between the response of motility and growth to HS-supplemented cultivation, the ratio of both activities in HS was plotted against that in FBS (Fig. 7⇓). There was no significant correlation between the motility response to HS and the growth response (r = 0.479, p = 0.168), suggesting that there was no linear association between the responses.
Effects of human allosera on the growth and motility of human BMSC.
The finding that human autologous sera induced higher motility allowed us to investigate whether allosera may also elicit similar effects to those observed in autoserum. In FBS-precultured STM sample 5 cells in which growth was augmented significantly by autologous serum (Table III⇑), an alloserum (sample 4), did not show significant augmentation of growth as compared with FBS (od570/620 at day 6: 40.2 + 0.50 vs 36.5 + 1.00, p = 0.024). In sample STM 10 cells, one of three allosera, sample STS 7, enhanced growth significantly as compared with auto-serum sample 10 (Fig. 8a⇓). On the other hand, three of four allosera showed similar stimulation of motility to that induced by autoserum (Fig. 8b⇓). It is interesting that the effect on motility of the serum from sample 7, which showed more stimulation of growth, was reduced as compared with that of autoserum (Fig. 8⇓), suggesting the discrepancy between growth and motility stimulatory activities.
In our study, medium with autologous HS was no less effective in stimulating human BMSC growth than that with FBS, but in some cases medium with autologous HS was more effective, although the individual growth rate varied among cases, probably reflecting the wide range of the period between the day when culture started and that when examination was started. Furthermore, medium with autologous HS was more effective in promoting motility than that with FBS in all cases, although the degree of stimulatory effects varied among cases. To our knowledge, this is the first demonstration of the effect of autologous serum on the motility of human BMSC.
Oreffo, Virdi and Triffitt20 compared the effects of FBS and HS on the proliferation of human BMSC pre-cultured in FBS by determining the incorporation of radiolabelled thymidine as indicative of DNA synthesis, and showed that HS stimulated thymidine incorporation in a dose-dependent manner. Recently, the number and area of colony-forming units produced by primary cultures of human BMSC in media supplemented with autologous HS have been shown to be larger than or similar to those supplemented with FBS, depending on the case.12 These data correspond well to our present findings although in some studies differentiation agents such as dexamethasone, beta-glycerophosphate, and 1,25 dihydroxyvitamin D3 were added to the culture medium within two days.12,20 Human serum, even alloserum, has been shown to be better at inducing mesenchymal differentiation than FBS.11,12,20 These results suggest that autologous HS may provide sufficient ex vivo expansion of human BMSC with multidifferentiation potential, and more importantly provide avoidance of contact of human BMSC with dangerous exogenous FBS.
In contrast to cell growth, cell motility was higher in autologous HS than in FBS in all cases, although the increased rate varied among patients. This was observed also in cells prepared with HS in primary culture suggesting that autologous HS is better for ex vivo expansion than FBS in regard to the preservation of motility in vitro. At present, BMSC in combination with biomaterials such as porous hydroxyapatite are used in the treatment of large bone defects.2,3 The incorporation of the biomaterials requires the cells responding to bone formation to penetrate the materials.21 High motility may thus be necessary for human BMSC when the cells are utilised for reconstruction of bone in combination with biomaterials. Furthermore, the propensity of chondrogenic precursor cells to migrate and proliferate in an injured area plays an important role in the repair of cartilage tissue.22 Hyaluronic-acid-stimulated cell motility of chondrocytes is thought to be related to the repair of cartilage tissue.23 The regeneration of ligaments and menisci has been shown to be related to cell migration occurring within four months.24 Thus, the preservation of high motility of human BMSC plays an important role in the application of ex vivo expanded stem cells to musculoskeletal tissue repair. Therefore, autologous HS may be better than FBS during ex vivo expansion of human BMSC possessing multidirectional differentiation with high motility.
In our study, cell growth was not always stimulated more significantly by allosera as compared with FBS. Kuznetsov et al11 demonstrated that medium with FBS remained more effective than human allosera for ex vivo growth of human BMSC. By contrast, all allosera evaluated in our study stimulated cell motility as compared with FBS. Furthermore, there was no significant correlation between the ratio of HS-supplemented to FBS-supplemented motility and the ratio of HS-supplemented to FBS-supplemented growth, suggesting the discrepancy of the molecules responding for the two phenotypes. These results indicated that cells from the different donors were more or less responsive to the growth and motility stimulation effects, and suggested that this could be in response to the same factor or to different factors present in the serum. Alternatively, motility factors may not be identical growth factors, but at least in part species specific.
Previously, FBS has been shown to contain motility factor(s), although the biochemical features are uncertain.25 At present, it is not known which factor(s) respond to the stimulation of human BMSC motility. Several, including growth factors are known to be associated with the motility of marrow stem cells. A recent study has shown that osteoprogenitor cells express osteoblast stimulating factor-1, more commonly known as pleiotrophin, a 136 amino-acid bone growth factor which drives primed human osteoprogenitor cells, suggesting that it may act as feedback on the activity of osteoprogenitor cells.26 Chondrocytes and synovial cells show chemokinesis to bFGF and hyaluronic acid, respectively.23 In our study, bFGF stimulated cell growth of human BMSC in the presence of autologous HS cell motility. It is unlikely that bFGF is a key motility factor involved in HS, since the cells with responding capacity to the cytokine did not show motility to the cytokine. Further study will be necessary to elucidate the molecules preserving high motility of human BMSC involved in HS.
In summary, autologous HS may provide sufficient ex vivo expansion of human BMSC possessing multidifferentiation potentials, and be better than FBS in the preservation of high motility, although further studies involving a larger number of patients will be necessary to confirm the effect of bFGF and the effect of allosera on cell morphology and growth response using HS. Diseases for which mesenchymal tissue reconstruction is applied are mainly non-lethal. Thus the possibility of being lethal, such as a potential vector for prion transmission, should be avoided. There is also a possibility of encountering new unknown lethal transfectants in allografts as well as in xenografts. For clinical application, we strongly recommend the use of autologous human serum as a supplementation for ex vivo culture of BMSC.
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
- Received November 19, 2004.
- Accepted May 4, 2005.
- © 2005 British Editorial Society of Bone and Joint Surgery