Extracellular matrix mimicking scaffold promotes osteogenic stem cell differentiation: A new approach in osteoporosis research

Abstract

Background:

Osteoporosis is a common metabolic disease, with mesenchymal stem cells discussed to play an important role in its pathomechanism. For in vitro osteoporosis studies, selection of adequate culture conditions is mandatory so as to preserve cell properties as far as possible. A suitable cell culture surface would ideally provide reproducible experimental conditions by resembling those in-vivo.

Objective:

Generating an improved growth surface for osteogenic differentiation of human bone marrow derived mesenchymal stem cells (hBMSCs).

Methods:

We modified electrospun gelatine meshes with hydroxyapatite nanopowder. The potential beneficial impact of the ensuing culture conditions were evaluated by cultivating and comparing the growth of cells from osteoporotic and non-osteoporotic donors on either hydroxyapatite-gelatine (HA) meshes, pure gelatine meshes, or 2D standard tissue culture surfaces.

Results:

After 21 days of differentiation, cells grown on pure or HA-gelatine meshes showed significantly higher mineralization levels compared to cells cultured in standard conditions. The amount of mineralization varied considerably in hBMSC cultures of individual patients but showed no significant difference between stem cells obtained from osteoporotic or non-osteoporotic donors.

Conclusions:

Overall, these results indicate that the use of HA-gelatine meshes as growth surfaces may serve as a valuable tool for cultivation and differentiation of mesenchymal stem cells along the osteogenic lineage, facilitating future research on osteoporosis and related issues.

Keywords

Osteoporosis mesenchymal stem cells differentiation electrospun gelatine mesh hydroxyapatite

1. Introduction

The most common chronic metabolic bone disease, osteoporosis, is characterized by low bone mass and low bone mineral density. It causes increased bone fragility, higher fracture incidence, greater morbidity and as a consequence increased mortality [1,2]. The mechanisms underlying this progressive systemic disease appear to involve a disturbed balance of bone resorption and remodelling, the main cellular executioners of which processes are osteoclasts and osteoblasts, respectively. At the molecular level, alterations in signalling pathways like RANK/RANKL/OPG have been identified and the importance of oestrogen-alpha-receptors in osteoblast-like cell activity are well described [3,4]. Further, important roles for Vitamin D and calcium in fracture prevention have been shown in meta-analyses [5] and are still under investigation in a large multi-centric RCT (http://clinicaltrials.gov/show/NCT01745263).

In order to achieve more detailed insight into the cellular aspects of the pathophysiology of osteoporosis, several studies investigated differentiation processes either towards osteoclasts, which descend from hematopoetic stem cells, or towards osteoblasts, which derive from hBMSCs. With regard to the latter, some studies suggested that hBMSCs from osteoporotic patients hold less osteogenic capacity as compared to cells from non-osteoporotic controls and display a tendency towards adipogenesis rather than osteogenesis [6,7]. Other studies, however, indicate that upon chemically induced differentiation hBMSCs from both groups show the same ability to differentiate [8,9].

The majority of these cell culture studies used traditional two-dimensional rigid flat plastic dishes as culture surface, such as tissue culture flasks, micro well plates and Petri dishes. Over the last few years, however, the influence of three dimensional micro- and nanostructure in human tissue on cell behaviour has become increasingly apparent, indicating that structural differences may play a vital role in cell adhesion, proliferation, and differentiation [10–13]. Thus, growth surfaces resembling in vivo tissue structures of extracellular matrix have been developed, and in particular such structures as obtained with electrospun nanofibrous scaffolds have proven to be a relatively inexpensive and easy-to-use cell culture surface suitable for the cultivation of stem cells [6,7,14–16].

It has been previously found that gelatine, with its close structural resemblance to native collagen, appears to be particularly valuable for mesenchymal stem cell cultures [17]. In an own study we have shown that gelatine-based electrospun fibrous meshes do not only support enhanced osteogenic differentiation, but are in addition compatible with the application of state of the art electron microscopy, immunofluorescence and biochemical approaches [18]. Further, it has been reported that the presence of certain inorganic and organic substances may enhance the differentiation of MSCs along the osteogenic cell lineage [19,20]. Noteworthy, hydroxyapatite (HA) added to ceramics or scaffolds even supported increased bone healing properties in experiments with animal models and was thus subsequently implemented to this novel approach of hBMSC culturing [21–24].

In the present study we focussed on the in vitro differentiation of hBMSCs from patients with and without osteoporosis on electrospun gelatine meshes with incorporated HA nano-particles, in this way combining both established osteoinductive methods. This is a novel technique for in vitro osteoanabolic cell studies, providing culture conditions with a close resemblance to in vivo conditions.

2. Materials and methods

2.1. Electrospinning of hydroxyapatite meshes

Electrospinning is a simple method for generating nano- and microfibres. For this purpose, a polymer solution is loaded into a needle which is placed in a defined distance from a grounded metallic collector. By applying high voltage in the range of several kilovolts between the needle and the collector, the polymer is forced towards the collector forming nano- and microscaled fibres which are deposited on the collector. The whole procedure was performed using a specific in-house built electrospinning device. Prior to electrospinning, polymer solutions with hydroxyapatite were prepared. To this end, gelatine (#G2500, Sigma Aldrich, St. Louis, MO, USA) was admixed with hydroxyapatite nanopowder (#677418-5G, Sigma Aldrich, St. Louis, MO, USA) in a ratio of 16:1. The resulting mixture was dissolved in pure acetic acid (#3738.1 Carl Roth, Karlsruhe, Germany) (at a ratio of 16% gelatine, 1% hydroxyapatite and 83% pure acetic acid w/v) and warmed to 80°C for 120 minutes. A homogenous polymer suspension was achieved by continuously swirling with a vortex mixer (Lab dancer, Cole-Parmer, Austria). Subsequently the suspension was cooled down to room temperature and loaded into a clipped-off cannula (BD MicrolanceTM 3 - 21G) with an inner diameter of 0.546 mm. Glass cover slips with a diameter of 12 millimetres (Menzel GmbH, Braunschweig, Germany) were placed on a metallic collector. The distance between needle tip and collector was set to 10 centimetres and the spinning procedure was performed at a voltage of 10 kV.

Afterwards, the meshes were cross-linked with 2.5% v/V glutaraldehyde (dissolved from glutaraldehyde stock 25% (v/v), #G5882, Sigma) for three hours in order to stabilize the fibres. To remove residues, several washing steps with ddH2O were performed.

Cover slips with the attached meshes were treated with absolute ethanol for five minutes in three cycles and afterwards washed with Dulbecco’s Phosphate Buffered Saline (DPBS) (#BE17-512F, Lonza, Belgium). All electrospun meshes were incubated overnight in growth medium before starting the experiments.

Non-HA-gelatine meshes were made by dissolving 16% gelatine in 84% pure acidic acid. All other consecutive steps were performed as described for the HA meshes.

2.2. Patients

At the Department of Trauma Surgery of the University Hospital Innsbruck, 10 patients between 67 and 97 years (mean 79.8) were included according to the inclusion and exclusion criteria mentioned in Table 1. During a routine surgery, bone marrow was harvested and hBMSCs were isolated according to Wolfe et al. [25]. Patients were admitted to the hospital with an acute traumatic proximal femur fracture that required either intramedullary nailing with the Proximal Femoral Nail Antirotation or hip replacements (total or hemi) with a prosthetic implant. During these procedures bone marrow was harvested from the proximal part of the bone marrow cavity of the femur. Further data on individual patients and both patient groups are summarized in Table 2.

Table 1
Inclusion/Exclusion criteria of patients for the use of BMSCs in this study

Inclusion criteria Exclusion criteria

$Age > 64$ years Malignant tumor

Proximal femoral fracture after trauma with the need of operative treatment Chemotherapy

Osteoporotic group: $BMD < - 2.5$ Non osteoporotic group: $BMD > - 1$ $BMD < 2.5$ Endocrinologic disease: Hyperparathyroidism Hypoparathyroidism Cushing’s syndrome Hyperthyroidism Hypothyroidism Hypogonadism

Written informed consent Renal failure

Anorexia nervosa

Osteogenesis imperfecta

Inflammatory bowel isease

Celiac disease

Longterm use of: Anticonvulsants Long-acting benzodiazepines Cytotoxic drugs Immunosuppressants Heparin

Current systemic infection

Inclusion criteria	Exclusion criteria
$Age > 64$ years	Malignant tumor
Proximal femoral fracture after trauma with the need of operative treatment	Chemotherapy
Osteoporotic group: $BMD < - 2.5$ Non osteoporotic group: $BMD > - 1$ $BMD < 2.5$	Endocrinologic disease: Hyperparathyroidism Hypoparathyroidism Cushing’s syndrome Hyperthyroidism Hypothyroidism Hypogonadism
Written informed consent	Renal failure
Anorexia nervosa
Osteogenesis imperfecta
Inflammatory bowel isease
Celiac disease
Longterm use of: Anticonvulsants Long-acting benzodiazepines Cytotoxic drugs Immunosuppressants Heparin
Current systemic infection

Table 2

Patient characteristics

Age (yrs)	Patient (sex)	T-Score (femoral neck/hip total/lumbar spine)	Group
71	m	$- 1$ /0.3/0.9	Non osteoporotic
67	f	$- 0.2$ / $- 0.4$ / $- 0.9$	Non osteoporotic
78	m	0.5/1.2/1.3	Non osteoporotic
73	m	$- 0.4$ / $- 0.5$ /0.8	Non osteoporotic
69	m	$- 0.2$ /0.3/ $- 0.1$	Non osteoporotic
97	f	$- 2.7$ / $- 2.9$ / $- 1.9$	Osteoporotic
91	f	$- 2.5$ / $- 1.5$ / $- 0.8$	Osteoporotic
75	f	$- 2.5$ / $- 1.9$ / $- 3.5$	Osteoporotic
90	f	$- 3$ / $- 2.7$ / $- 1.5$	Osteoporotic
87	f	prothesis/prothesis/ $- 5.1$	Osteoporotic

Group of patients with osteoprosis: $n = 5$ , mean age 88.0 ± 8.1. Group of patients without osteoprosis: $n = 5$ , mean age 71.6 ± 4.2.

Bone mineral density was measured at the contra lateral hip and/or lumbar spine by dual energy X-ray absorptiometry (DXA; Hologic QDR 4500) within 10 days after the surgical procedure. According to T-Scores, patients were allocated to the osteoporotic ( $T-Score < - 2.5$ ) or non-osteoporotic group ( $T-Score > - 1 < 2.5$ ). Patients with T-Scores between $- 1$ and $- 2.5$ were excluded from this study as considered osteopenic (Table 2). The study was approved by the ethics commission of the Medical University of Innsbruck and all patients gave written consent.

2.3. Cell culture

After harvesting 30 ml of bone marrow from the femoral medullary cavity the biomaterial was transferred to the laboratory of the Department of Trauma Surgery where the isolation process was conducted analogous to an established protocol.

Culturing of pure hBMSCs was performed with growth medium consisting of Dulbecco’s Modified Eagle Medium (DMEM) low glucose with L-glutamine (Life Technologies, Paisley, United Kingdom), supplemented with 10% foetal bovine serum (Biowest, Nuaillé, France), 1% penicillin (50 IU/ml) and 1% streptomycin (50 μg/ml) (PAA Laboratories GmbH, Pasching, Austria). Cells were seeded at density of $2 \times 10^{4}$ per well in 24 well culture plates (Sarstedt, Wiener Neudorf, Austria) on either HA-gelatine meshes, pure gelatine meshes or on well bottoms. Incubation was performed at 37°C with a humidity of 98% and 5% CO₂ admixture in a CO₂-Incubator (Binder, Tuttlingen, Germany).

2.4. hBMSC differentiation

According to a well-recognized differentiation protocol previously established in our laboratory, osteogenic differentiation was triggered 72 h after seeding the cells by adding osteogenic induction medium (OM) consisting of normal growth medium supplemented with glycerol-2-phosphate (10 mM), ascorbic acid-2 phosphate (50 μM) and dexamethason (100 nM) (all from Sigma Aldrich). At the same time normal growth medium was added for the control group (undifferentiated). A complete change of the media was executed three times per week.

2.5. Quantification of mineralization

At day 21 after the induction of differentiation cells were fixed by adding 4% formaldehyde. Mineralization was shown by Alizarin Red S staining and documented with a Leica DMI 3000 B inverse fluorescence microscope equipped with a DIC 450 C camera and LAS V 4.0 software (all Leica, Microsystems, Wetzlar, Germany). The mineralized compounds of the differentiated cells were extracted by cetylpyridinium chloride (Sigma-Aldrich, Vienna, Austria) and analysed on a spectrophotometer (NanoDrop2000c, Thermo Scientific, USA) at 550 nm. In addition, we applied Hoechst 33258 staining of the nuclei so as to determine the cell count and to obtain a measure of cell proliferation. To visualize Hoechst stained nuclei, a fluorescence filter cube for UV excitation (Leica fluorescence filter cube A4 ET for UV excitation; excitation filter: BP 360/40; dichromatic mirror: 400; suppression filter: BP 470/40; Leica, Microsystems, Wetzlar, Germany) was used. At a magnification of 10× 3 representative images per sample per well (top, centre, bottom) were collected and image analysis and cell counting was then performed with CellProfiler cell image analysis software version 2.1.1 (Scientific Volume Imaging, Hilversum, The Netherlands).

2.6. Fluorescence activated cell sorting analysis of surface markers

Surface markers specific for haemotopoietic lineage, CD 14, CD 34 and CD 45, and for mesenchymal stem cells, CD 90, CD 105 and CD 166, were labelled with specific antibodies (Biozym Scientific GmbH, Oldendorf, Germany) in order to determinate cellular stemness according the gold standard defined by Dominici et al. [26]. A FACScan (BD Biosciences, Schwechat, Austria) was used to detect the phycoerythrin-labelled antibodies. Data were analysed with WinMDI 2.8 freeware.

2.7. Cell proliferation/viability

On day 1, 15 and 21 cell proliferation and viability were assessed using the WST-1 assay (Roche Diagnostics GmbH, Mannheim, Germany). To this end, we compared cells grown on 2D well bottoms with those grown on HA-gelatine meshes. Evaluation was performed according to the instructions of the manufacturer, using a spectrophotometer (NanoDrop2000c, Thermo Scientific, USA) measuring absorption at a wavelength of 437 nm. As a reference point reflecting the absorption signal obtained with 100% dead cells, cells in separate wells were incubated with 30% H₂O₂ before running the assay. The results shown are expressed as fold change of the assay signal relative to the baseline value obtained in these background controls.

2.8. Electron microscopy

hBMSCs grown on HA-gelatine meshes were fixed by high pressure freezing (HPF) followed by freeze-substitution and epoxy resin embedding as described for comparable samples and other cell types by Hess et al. [12]. This procedure achieves optimal ultrastructural resolution and prevents many of the artefacts arising due to conventional chemical fixation procedures.

Samples to be used for standard transmission electron microscopy (TEM) were fixed with freshly made 4% para-formaldehyde, counterstained and post fixed in order to increase the contrast with 1% OsO₄, dehydrated in ascending sequence of ethanol, and finally acetone and epoxy-resin embedded. Ultrathin sections were prepared at a Leica UltracutS and TEM images were generated on a Philips CM120 EM (F.E.I., Eindhoven, Netherlands). Images were recorded with a MORADA digital camera (SIS, Münster, Germany).

For scanning electron microscope (SEM) analysis, cells were fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) followed by 1 h post fixation with 1% OsO₄, dehydration with ethanol, and critical point drying in a Bal-Tec CPD 030 critical point dryer (Balzers, Lichtenstein). Specimens were mounted with conductive carbon cement Leit-C according to Göcke (Plano GMBH, Wetzla, Germany) on aluminium stubs sputtered with 10 nm Au/Pd (Balzers), and examined on a Zeiss scanning electron microscope (Gemini 982).

2.9. Statistical analysis

All data were analysed using the IBM SPSS 24.0 statistics package (IBM Corp., USA). Metric data means were compared using Mann–Whitney-U test or Kruskal–Wallis test with $p < 0.05$ considered as statistically significant. Boxplots show 95% confidence intervals (CI). Means are expressed ± standard deviation (SD).

3. Results

3.1. Stability and general characterization of gelatine meshes with and without HA

The meshes were tested for their stability when submerged into cell culture medium. Microscopic observation revealed that the meshes showed no signs of dissolving, but, due to water uptake, a slight increase of fibre diameter was noticed. Final fibre diameter was 408 nm ± 85 SD ( $n = 32$ ). To establish constant conditions for cell culture experiments, meshes were thus submerged in cell culture medium 48 h prior to cell seeding. An electron microscopic image depicting typical HA gelatine fibres is shown in Fig. 1.

Fig. 1.

Highly magnified hydroxyapatite nanoparticles on the surface and inside HA gelatine fibres.

3.2. Characterization of donor hBMSCs

Analysing the expression of surface markers using fluorescence-activated cell sorting (FACS), we demonstrated that in the cells purified from human donors markers characterizing the haemotopoietic lineage, CD 14, CD 34 and CD 45, were only detectable in low levels, while typical stem cell markers, CD 90, CD 105 and CD 166, showed high expression levels (Fig. 2). Furthermore, three lineage capability of stem cells was shown by inducing hBMSCs to differentiate into adipo-, osteo- and chondrogen lineages (Fig. 3).

Fig. 2.

Results of FACS analysis showing high expression levels of stem cell markers CD 90, CD 105 and CD 166, but low expression levels of hematopoetic lineage markers CD 34, CD 14 and CD 45. Propidium iodide (PI) staining shows a low percentage of dead cells and Neg.KO indicates unstained controls.

Fig. 3.

Isolated hBMSCs can differentiate into three lineages. (A) Alizarin Red S stained hBMSCs differentiated towards osteogenesis. (B) Alcian blue stained proteoglycans built form cells differentiated along the chondrogenic pathway. (C) hBMSCs differentiated towards adipogenesis stained with Oil Red O.

3.3. Influence of growth substrate on proliferation/viability of hBMSCs (WST-1 assay)

Proliferation/viability measurements were performed on day 1, 15 and 21 after seeding of the cells. Measurements were normalized to the baseline of H₂O₂ treated dead cells mean optical density. As summarized in Table 3, no significant difference in proliferation/viability was found when comparing cells cultured on HA-gelatine meshes to cells grown on the well bottom.

3.4. Cell proliferation, differentiation and mineralization

For the assessment of cell differentiation and mineralization, 14,000 hBMSCs were plated onto each well and cell attachment was observed on all surfaces microscopically. Following 21 days in culture, cell proliferation was assessed by counting nuclei and mineralization was measured by Alizarin Red S staining. In line with the results of the WST-1 assay, cell proliferation analysis by counting Hoechst stained cell nuclei with cell profiler yielded no significant differences between groups. Mean cell count per field of view was 1731 ± 902 in the HA-mesh group, 1976 ± 980 in the pure gelatine mesh group, and 1723 ± 1078 in the well bottom group.

Alizarin Red S staining showed that, as expected, staining was positive on stimulated cells and negative on control cells, regardless of the growth substrate used. However, mineralization levels varied between groups, the highest mineralization levels being observed in cells cultured on HA gelatine meshes as growth surfaces, followed by pure gelatine meshes and flat bottom control wells (Fig. 4(a)).

Table 3
Cell proliferation of hBMSCs cultured in standard flat bottom wells or on HA gelatine meshes

Day 1 Day 15 Day 21

$4.826 \pm 0.656$ $7.576 \pm 1.032$ $6.494 \pm 0.140$ Well bottom

$4.978 \pm 0.584$ $7.795 \pm 1.090$ $6.839 \pm 1.320$ HA gelatine mesh

$p > 0.05$ $p > 0.05$ $p > 0.05$ P

Day 1	Day 15	Day 21
$4.826 \pm 0.656$	$7.576 \pm 1.032$	$6.494 \pm 0.140$	Well bottom
$4.978 \pm 0.584$	$7.795 \pm 1.090$	$6.839 \pm 1.320$	HA gelatine mesh
$p > 0.05$	$p > 0.05$	$p > 0.05$	P

Data are absorption values expressed relative to a dead cell background and are means ± SD, $n = 20$ .

Fig. 4.

(a) Alizarin Red staining of differentiated cells (left) and non-differentiated controls (right). (A) and (B): HA gelatine mesh, (C) and (D): pure gelatine mesh, (E) and (F): well bottoms. (b) Boxplots showing relative Alizarin Red S staining levels in control cultures grown without differentiation medium (left) and in differentiated hBMSCs (right). Cells cultured on HA-gelatine meshes revealed significantly increased mineralization compared to those cultured on pure gelatine meshes or well bottom surfaces ( $p < 0.001$ ), $n = 24$ cultures.

Spectrophotometric quantification of the grade of mineralization, by measuring absorption of extracted Alizarin Red S stained products, confirmed the visual impression. Expressed relative to the absorption signal observed in differentiated cells grown on conventional plastic wells, the increase in mineralization on the HA-gelatine meshes amounted to 4.375 ± 1.559 fold, compared to the 2.295 ± 1.172 fold increase measured on the pure gelatine-mesh (1 ± 0.679 fold for conventional plastic surfaces), with significantly differing means ( $p < 0.001$ ). In the respective undifferentiated controls, the fold change means, compared to differentiated cells on conventional plastic surfaces, were 0.8, 0.9 and 0.6 (Fig. 4(b)).

A comparison of the differentiation from hBMSCs from individual patients showed that the fold change ranged from 2.032–6.554 on HA-gelatine meshes, 0.589–4.585 on pure gelatine meshes and 0.177–2.872 on conventional wells. However, despite this variability, cells from each patient showed the highest increase on HA-gelatine meshes (Fig. 5). Noteworthy, a comparison between cells from osteoporotic and non-osteoporotic donors revealed that there was no significant difference between both groups ( $p > 0.05$ ), the mean of the relative increase in the osteoporotic group amounting to 2.613 (95% CI 2.006–3.220) and that in the non-osteoporotic group to 2.501 (95% CI 1.863–3.138).

Fig. 5.

Fold change of mineralization in cells obtained from osteoporotic and non-osteoporotic patients and grown on HA-gelatine meshes, pure gelatine meshes and well bottom surfaces. Mineralization was measured by spectrophotometric quantification of Alizarin Red S levels. For all patients, HA-gelatine meshes reached highest mineralization levels compared to pure gelatine and well bottom.

SEM analysis showed the fine structure of the gelatine mesh with HA inclusions. The diameter of the fibres was 408 nm ± 85 SD, this has been shown as an ideal diameter for cell growth in other publications [18].

We further assessed the morphology of non-differentiated and differentiated cells cultured on the different growth substrates using scanning (SEM) and transmission electron microscopy (TEM). Observing the surface morphology by SEM of stem cells growing on the HA-mesh we found undifferentiated cells being extremely flat with long protrusions as well as filopodia and few surface structures (microvilli) (Fig. 6(b)). Further, we frequently observed established cell to cell contacts, whereas, typically (Fig. 6(b)) no ingrowth into the mesh was observed. In contrast, cells differentiated for 21 days showed a clear tendency for ingrowth into the mesh. Furthermore, the formation of heterogeneous crumble-like structures, so called bone nodules [27], in combination with filamentous secreted matrix could be observed. The cell surface of differentiated cells showed a tendency for microvilli formation (Fig. 6(a)).

Fig. 6.

(a) Ingrown differentiated cells on the HA gelatine mesh. Small heterogeneous crumble-like structures, so called bone nodules, in combination with filamentous secreted matrix are resembling mineralization. (b) Showing interaction of undifferentiated cells with the HA gelatine mesh fibres without ingrowth.

TEM analysis showed the fine structure of the gelatine mesh with HA nanoparticle inclusions as electron dense fibres with less electron dense inclusions. Interestingly, in presence of differentiated cells only, calcite spikes formed on the fibre surface (Fig. 7(A) and (B)). Furthermore, filamentous extracellular material resembling collagen was detected. As shown in Fig. 7(C) these structures were missing in undifferentiated controls, were cells were grown for the same time in presence of standard cultivation medium (Fig. 7).

Fig. 7.

Gelatine mesh with HA inclusions (arrows). (A) and (B) show the mesh in the presence of differentiating cells, (C) presents the same mesh with cells cultured in standard medium. Arrowhead in (A) and (B) indicates surface calcite spikes. Secreted collagen fibres are indicated by asterisks. Cellular vacuoles observed at the fibre site are indicated with V, the cell nucleus with N.

In order to analyse the ultrastructure of hBMSCs grown on HA-gelatine meshes cryo fixation by means of HPF was performed to avoid artefacts caused by standard chemical fixation. These cells showed perinuclear, tubular, healthy looking mitochondria as described previously for other differentiated cell types [28–30], a bean shaped nucleus and numerous heterogeneous lyso- and autophagosomes in the cell periphery (Fig. 8). Furthermore, even with the sophisticated method of HPF the abundantly present rough endoplasmic reticulum (rER) appears swollen, with abundant ribosomes as described also by Pietilä et al. for chemically fixed cells [31]. This suggests active protein synthesis and high membrane turnover.

Fig. 8.

Cryo fixed differentiated hBMSC grown on an HA-gelatine mesh showing tubular, perinuclear mitochondria (arrow), abundant lyso-autophagosomes (L) and a swollen rER (asterisk). N indicates the nucleus.

4. Discussion

The ideal way of culturing cells requires an environment that reproduces in vivo conditions. Especially the differentiation of stem cells is known to be profoundly influenced by surrounding factors and may thus highly benefit from the improvement of culturing techniques. In mesenchymal cell culture investigations and osteoporosis research, the ideal cell culture base would therefore be a bone-like structure, which in humans consists of collagen fibres and hydroxyapatite in a trabecular setup.

Here we tested meshes based on gelatine, presuming good compatibility of gelatine to mesenchymal stem cells, as it is of a similar structure and closely related to native collagen. Further, Zhang et al. already demonstrated that bone marrow stromal cells of rabbits showed better attachment and growth on electrospun gelatine/polycaprolaconte (PCL) meshes than on pure electrospun PCL meshes [32]. In another study Alvarez Perez et al. successfully differentiated a human mesenchymal stem cell line on gelatine/PCL nanofiber scaffolds along the osteogenic linage [33]. However, despite these positive results, it has been questioned that due to the impact of the electrospinning process on collagen the resulting fibres actually match the composition of native collagen [34].

By mimicking the structure of native extra cellular matrix, a three dimensional micro architecture of electrospun meshes of PCL and collagen (COL) had already been shown to increase cell proliferation and to facilitate an even distribution of hMSCs inside the scaffolds [16]. Gelatine with its biocompatibilty and complete biodegradability is superior to poly-L-lactide acide (PLLA), that although often used for the fabrication of electrospun meshes, releases acidic degradation products which in consequence increase acidification of the surrounding tissue [19]. In a previous study, Schmiedinger et al. have shown that electrospun gelatine nanofiber meshes are compatible with a great number of biochemical and microscopical techniques in addition of being an ideal substrate for a multitude of cells [18]. These results have also revealed that the scaffolds used allowed good nourishment of cells as culture media could penetrate the mesh and reach the basic cell surface which is especially important for polarized cells. Despite the numerous advantages of 3D culture surfaces, their availability is limited and most laboratory protocols are established and optimized for 2D surface cultures only.

In the present study we found that, to adjust the solubility and stability of gelatine meshes, glutaraldehyde was required to provide the crosslinking step. Despite of this addition, both a metabolic activity based assay (WST-1) and a simple cell counting approach (Hoechst staining) indicated that cell proliferation and viability were unaffected in gelatine meshes with or without HA as compared to conventional two-dimensional growth surfaces, eliminating any concerns regarding short term toxicity.

A number of previous studies has demonstrated the beneficial effects of hydroxyapatite on osteoblast function, bone tissue engineering and osteogenic differentiation [19,21,23,35–37]. Thus, to further enhance the positive features of three dimensional growth surfaces, we also added hydroxyapatite nanopowder to the gelatine polymer solution. Noteworthy, it has been observed that when fibres were uniformly coated with HA powder this limited the access of cells to the underlying material [36,38]. In contrast, as verified by SEM imaging, with the protocol applied here we were able to introduce hydroxyapatite particles both inside the fibres and on the fibre surface, leaving ample space for cell attachment.

Presumably as a result of these improved culturing conditions, after 21 days in culture, cells seeded on the HA-gelatine mesh showed significantly higher mineralization than cells grown on pure gelatine meshes and on well bottoms. The non-differentiated control group revealed only weak signs of the basic osteoinductive effect of HA, with even lower mean mineralization levels on HA-gelatine meshes than on pure gelatine meshes. This result is in contrast to findings obtained with polyester compounds, where the pure presence of HA was described to be slightly osteoinductive [24].

In order to ascertain that the use of herein improved cell culture conditions would not only facilitate the study of hBMSCs in general, but also that obtained in osteoporotic patients, we further examined, in a comparative fashion, their impact on mesenchymal stem cells of osteoporotic donors to those of non-osteoporotic ones. Some authors claim that hBMSCs from osteoporotic patients hold less osteogenic capacity than those of non-osteoporotic controls, and they also showed a shift towards adipogenesis rather than osteogenesis upon induction of differentiation [6,7,39]. This is controversial to findings of others, which state that mesenchymal stem cells of osteoporotic and non-osteoporotic patients have the same potential to differentiate if chemically induced [8,9]. Our results here confirm the latter, showing no apparent difference in proliferation or osteogenic differentiation capacity between the two groups. This difference could also in part be attributed to the diverging experimental details of the above mentioned studies. Compared to the present study, the donors of the former studies were of a younger age group and exclusively female [7,39], or cells were used postmortem and induced with a different osteogenic medium for a longer period [6]. Admittedly, we observed rather high variability between individual patients regarding these parameters. A possible explanation for this finding is, that despite relatively narrow exclusion and inclusion criteria in this study, a multitude of factors could still alter the ability of proliferation and the osteogenic capacity of hBMSC, such as patient mobility, comorbidities or the overall state of health. However, we consistently saw in all ours donor that their cells showed the highest differentiation rate on HA-gelatine meshes, followed by pure gelatine meshes and well bottoms, regardless of the donor’s osteoporosis diagnosis. To the best of our knowledge, this is the first direct demonstration of this kind, since electrospun HA-gelatine meshes have never been used as a base for culturing and differentiating hBMSCs from osteoporotic patients before.

As a final aspect of this study, ultrastructural and morphological features of cultured cells were examined. Importantly, the secretion of collagen fibres and the formation of bone nodules from differentiated hBMSCs were demonstrated, these being the basal components needed for new bone formation. At the same time a perinuclear localization of small groups of mitochondria was noted, previously considered an attribute of non-differentiated stem cells [40,41]. The conspicuously abundant presence of lyso- and autophagosoms is usually regarded as indicator for stressed cells. However, this can also be interpreted as a sign for high membrane turnover, which would appear more reasonable in our differentiating cells [42,43]. The same holds true for the swollen rER as this feature of differentiating stem cells was recently described elsewhere [31] and is not attributable to a fixation artefact.

Our HA-gelatine mesh seemed to change significantly in the presence of differentiating cells as calcite spikes were deposited on the surface and the material appeared more electron dense. These features were never seen in undifferentiated samples which suggests an agglomeration of secreted material on the mesh and further supports the presence of HA as being beneficial for cell mineralization.

It should finally be pointed out that we are clearly aware of limitations of this study. Specifically, a relatively low patient number was used, but it appears to us that despite of this the beneficial effect of the HA-mesh on osteogenesis in stem cells from osteoporotic and non-osteoporotic donors is nonetheless evident. Moreover, we have a clearly uneven sex distribution among the study groups. However, given the large individual variability observed in terms of cell proliferation or mineralization, a potential gender bias would have to be immensely to obscure any difference in osteogenic activity between the groups. Similarly, in a recent study on hBMSC energetics and ultrastructure we also failed to observe significant differences between cells from osteoporotic and non-osteoporotic donors, with values of a single male osteoporotic falling well into the range of other non-osteoporotic males and that of a female control corresponding to female osteoporosis patients [44].

5. Conclusion

The mechanisms underlying the development of osteoporosis are still not sufficiently understood. We believe that new approaches for cell culture work are thus needed to obtain more detailed results under more in vivo-like conditions. Therefore, our aim was to develop a more natural cell culture surface, where cells of osteoporotic and non-osteoporotic donors can adhere, proliferate and differentiate better than on conventional two-dimensional plastic surfaces. Our study clearly shows that electrospun HA-gelatine meshes may serve this purpose, as cells cultured on them display significantly better mineralization levels, irrespective of whether they were obtained from controls or osteoporotic donors, while proliferation and cell adhesion were at par with conventional cell culture surfaces. Overall, HA-gelatine meshes present a valuable tool on which cells can be cultivated and differentiated along the osteogenic lineage and simultaneously remain easy to handle for in vitro studies. The HA-gelatine mesh thereby provides a remarkably improved culture tool for osteogenic differentiation of hBMSC.

Conflict of interest

The authors have no conflict of interest to report.

References

Cooper

Campion

and Melton 3rd

L.J.

, Hip fractures in the elderly: A world-wide projection, Osteoporos Int. 2(6) (1992), 285–289. doi:10.1007/BF01623184.

Lane

N.E.

, Epidemiology, etiology, and diagnosis of osteoporosis, American Journal of Obstetrics and Gynecology 194(2) (2006), S3–S11. doi:10.1016/j.ajog.2005.08.047.

Lee

Jessop

Suswillo

Zaman

and Lanyon

, Endocrinology bone adaptation requires oestrogen receptor-alpha, Nature 424(6947) (2003), 389. doi:10.1038/424389a.

Silva

and Branco

J.C.

, RANK/RANKL/OPG: Literature review, Acta Reumatol. Port. 36(3) (2011), 209–218.

Bischoff-Ferrari

H.A.

Willett

W.C.

Orav

E.J.

Lips

Meunier

P.J.

et al., A pooled analysis of vitamin D dose requirements for fracture prevention, N. Engl. J. Med. 367(1) (2012), 40–49. doi:10.1056/NEJMoa1109617.

D’Ippolito

Schiller

P.C.

Ricordi

Roos

B.A.

and Howard

G.A.

, Age-related osteogenic potential of mesenchymal stromal stem cells from human vertebral bone marrow, J. Bone Miner. Res. 14(7) (1999), 1115–1122. doi:10.1359/jbmr.1999.14.7.1115.

Rodríguez

J.P.

Montecinos

Ríos

Reyes

and Martínez

, Mesenchymal stem cells from osteoporotic patients produce a type I collagen-deficient extracellular matrix favoring adipogenic differentiation, J. Cell. Biochem. 79(4) (2000), 557–565. doi:10.1002/1097-4644(20001215)79:4<557::AID-JCB40>3.0.CO;2-H.

Charoenpanich

Wall

M.E.

Tucker

C.J.

Andrews

D.M.K.

Lalush

D.S.

Dirschl

D.R.

et al., Cyclic tensile strain enhances osteogenesis and angiogenesis in mesenchymal stem cells from osteoporotic donors, Tissue Engineering A 20(1–2) (2014), 67–78. doi:10.1089/ten.tea.2013.0006.

Gao

Usas

Tang

Tan

Schneppendahl

et al., A comparison of bone regeneration with human mesenchymal stem cells and muscle-derived stem cells and the critical role of BMP, Biomaterials 35(25) (2014), 6859–6870. doi:10.1016/j.biomaterials.2014.04.113.

10.

Badami

A.S.

Kreke

M.R.

Thompson

M.S.

Riffle

J.S.

and Goldstein

A.S.

, Effect of fiber diameter on spreading, proliferation, and differentiation of osteoblastic cells on electrospun poly(lactic acid) substrates, Biomaterials 27(4) (2006), 596–606. doi:10.1016/j.biomaterials.2005.05.084.

11.

Blakeney

B.A.

Tambralli

Anderson

J.M.

Andukuri

Lim

D.-J.

Dean

D.R.

et al., Cell infiltration and growth in a low density, uncompressed three-dimensional electrospun nanofibrous scaffold, Biomaterials 32(6) (2020), 1583–1590. doi:10.1016/j.biomaterials.2010.10.056.

12.

Hess

M.W.

Pfaller

Ebner

H.L.

Beer

Hekl

and Seppi

, 3D versus 2D cell culture: Implications for electron microscopy, Methods Cell Biol. 96 (2010), 649–670, http://linkinghub.elsevier.com/retrieve/pii/S0091679X10960275.

13.

Zhong

Zhang

and Lim

C.T.

, Fabrication of large pores in electrospun nanofibrous scaffolds for cellular infiltration: A review, Tissue Engineering B 18(2) (2012), 77–87.

14.

Jakob

Ebert

Ignatius

Matsushita

Watanabe

Groll

et al., Bone tissue engineering in osteoporosis, Maturitas 75(2) (2013), 118–124. doi:10.1016/j.maturitas.2013.03.004.

15.

Ocarino

N.deM.

Boeloni

J.N.

Jorgetti

Gomes

D.A.

Goes

A.M.

and Serakides

, Intra-bone marrow injection of mesenchymal stem cells improves the femur bone mass of osteoporotic female rats, Connect. Tissue Res. 51(6) (2010), 426–433. doi:10.3109/03008201003597049.

16.

Srouji

Kizhner

Suss-Tobi

Livne

and Zussman

, 3-D nanofibrous electrospun multilayered construct is an alternative ECM mimicking scaffold, J. Mater. Sci. Mater. Med. 19(3) (2008), 1249–1255. doi:10.1007/s10856-007-3218-z.

17.

Chang

Lei

Liu

Qin

Luo

et al., Optimization of culture of mesenchymal stem cells: A comparison of conventional plate and microcarrier cultures, Cell Prolif. 45(5) (2012), 430–437. doi:10.1111/j.1365-2184.2012.00836.x.

18.

Schmiedinger

Vogel

G.F.

Eiter

Pfaller

Kaufmann

W.A.

Floerl

et al., Cryo-immunoelectron microscopy of adherent cells improved by the use of electrospun cell culture substrates, Traffic 14(8) (2013), 886–894.

19.

Anderson

J.M.

Patterson

J.L.

Vines

J.B.

Javed

Gilbert

S.R.

and Jun

H.-W.

, Biphasic peptide amphiphile nanomatrix embedded with hydroxyapatite nanoparticles for stimulated osteoinductive response, ACS Nano 5(12) (2011), 9463–9479. doi:10.1021/nn203247m.

20.

Zhang

and Ma

P.X.

, Nanofiber-based delivery of bioactive agents and stem cells to bone sites, Advanced Drug Delivery Reviews 64(12) (2012), 1129–1141. doi:10.1016/j.addr.2012.04.008.

21.

Dinarvand

Seyedjafari

Shafiee

Babaei Jandaghi

Doostmohammadi

Fathi

M.H.

et al., New approach to bone tissue engineering: Simultaneous application of hydroxyapatite and bioactive glass coated on a poly(l-lactic acid) scaffold, ACS Appl. Mater. Interfaces 3(11) (2011), 4518–4524. doi:10.1021/am201212u.

22.

Kim

S.-S.

Sun Park

Jeon

Yong Choi

and Kim

B.-S.

, Poly(lactide-co-glycolide)/hydroxyapatite composite scaffolds for bone tissue engineering, Biomaterials 27(8) (2006), 1399–1409. doi:10.1016/j.biomaterials.2005.08.016.

23.

Nguyen

N.K.

Leoni

Maniglio

and Migliaresi

, Hydroxyapatite nanorods: Soft-template synthesis, characterization and preliminary in vitro tests, J. Biomater. Appl. 28(1) (2013), 49–61. doi:10.1177/0885328212437065.

24.

Lü

L.-X.

Zhang

X.-F.

Wang

Y.-Y.

Ortiz

Mao

Jiang

Z.-L.

et al., Effects of hydroxyapatite-containing composite nanofibers on osteogenesis of mesenchymal stem cells in vitro and bone regeneration in vivo, ACS Appl. Mater. Interfaces 5(2) (2013), 319–330.

25.

Wolfe

Pochampally

Swaney

and Reger

R.L.

, Isolation and culture of bone marrow-derived human multipotent stromal cells (hMSCs), Methods Mol. Biol. 449 (2008), 3–25.

26.

Dominici

Le Blanc

Mueller

Slaper-Cortenbach

Marini

F.C.

Krause

D.S.

et al., Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement, Cytotherapy 8(4) (2006), 315–317. doi:10.1080/14653240600855905.

27.

Joubert

L.M.

, Scanning electron microscopy: Bridging the gap from stem cells to hydrogels, Microscopy and Microanalysis 16(S2) (2010), 596–597.

28.

Bruce

J.I.E.

Giovannucci

D.R.

Blinder

Shuttleworth

T.J.

and Yule

D.I.

, Modulation of [Ca2+]i signaling dynamics and metabolism by perinuclear mitochondria in mouse parotid acinar cells, J. Biol. Chem. 279(13) (2004), 12909–12917. doi:10.1074/jbc.M309070200.

29.

Park

M.K.

Ashby

M.C.

and Erdemli

, Perinuclear, perigranular and sub-plasmalemmal mitochondria have distinct functions in the regulation of cellular calcium transport, EMBO J. 20 (2001), 1863–1874.

30.

Yaffe

M.P.

, The machinery of mitochondrial inheritance and behavior, Science 283(5407) (1999), 1493–1497. doi:10.1126/science.283.5407.1493.

31.

Pietilä

Palomäki

Lehtonen

Ritamo

Valmu

Nystedt

et al., Mitochondrial function and energy metabolism in umbilical cord blood- and bone marrow-derived mesenchymal stem cells, Stem Cells Dev. 21(4) (2012), 575–588. doi:10.1089/scd.2011.0023.

32.

Zhang

Ouyang

Lim

C.T.

Ramakrishna

and Huang

Z.-M.

, Electrospinning of gelatin fibers and gelatin/PCL composite fibrous scaffolds, J. Biomed. Mater. Res. B 72(1) (2005), 156–165.

33.

Alvarez Perez

M.A.

Guarino

Cirillo

and Ambrosio

, In vitro mineralization and bone osteogenesis in poly(ε-caprolactone)/gelatin nanofibers, J. Biomed. Mater. Res. A 100(11) (2012), 3008–3019.

34.

Zeugolis

D.I.

Khew

S.T.

Yew

E.S.Y.

Ekaputra

A.K.

Tong

Y.W.

Yung

L.-Y.L.

et al., Electro-spinning of pure collagen nano-fibres – Just an expensive way to make gelatin?, Biomaterials 29(15) (2008), 2293–2305. doi:10.1016/j.biomaterials.2008.02.009.

35.

Gupta

Venugopal

Mitra

Dev

V.R.G.

and Ramakrishna

, Nanostructured biocomposite substrates by electrospinning and electrospraying for the mineralization of osteoblasts, Biomaterials 30(11) (2009), 2085–2094. doi:10.1016/j.biomaterials.2008.12.079.

36.

Sato

Sambito

M.A.

Aslani

Kalkhoran

N.M.

Slamovich

E.B.

and Webster

T.J.

, Increased osteoblast functions on undoped and yttrium-doped nanocrystalline hydroxyapatite coatings on titanium, Biomaterials 27(11) (2006), 2358–2369. doi:10.1016/j.biomaterials.2005.10.041.

37.

Venugopal

Rajeswari

Shayanti

Low

Bongso

Dev

V.R.

Deepika

Choon

A.T.

and Ramakrishna

, Electrosprayed hydroxyapatite on polymer nanofibers to differentiate mesenchymal stem cells to osteogenesis, J. Biomater. Sci. Polym. Ed. 24(2) (2013), 170–184. doi:10.1163/156856212X629845.

38.

Zandi

Mirzadeh

Mayer

Urch

Eslaminejad

M.B.

Bagheri

et al., Biocompatibility evaluation of nano-rod hydroxyapatite/gelatin coated with nano-HAp as a novel scaffold using mesenchymal stem cells, J. Biomed. Mater. Res. A 92(4) (2010), 1244–1255.

39.

Rodríguez

J.P.

Garat

Gajardo

Pino

A.M.

and Seitz

, Abnormal osteogenesis in osteoporotic patients is reflected by altered mesenchymal stem cells dynamics, J. Cell. Biochem. 75(3) (1999), 414–423. doi:10.1002/(SICI)1097-4644(19991201)75:3<414::AID-JCB7>3.0.CO;2-C.

40.

Lonergan

Bavister

and Brenner

, Mitochondria in stem cells, Mitochondrion 7(5) (2007), 289–296. doi:10.1016/j.mito.2007.05.002.

41.

Collins

T.J.

Berridge

M.J.

Lipp

and Bootman

M.D.

, Mitochondria are morphologically and functionally heterogeneous within cells, EMBO J. 21(7) (2002), 1616–1627. doi:10.1093/emboj/21.7.1616.

42.

Ravikumar

Moreau

and Rubinsztein

D.C.

, Plasma membrane helps autophagosomes grow, Autophagy 6(8) (2010), 1184–1186. doi:10.4161/auto.6.8.13428.

43.

Vogel

G.F.

Ebner

H.L.

de Araujo

M.E.G.

Schmiedinger

Eiter

Pircher

et al., Ultrastructural morphometry points to a new role for LAMTOR2 in regulating the endo/lysosomal system, Traffic 16(6) (2015), 617–634. doi:10.1111/tra.12271.

44.

Moll

C.W.

Lindtner

R.A.

Vogel

G.F.

Brunner

Gutleben

Blauth

et al., Bone marrow mesenchymal stem cells from osteoporotic patients do not show altered mitochondrial energetics or ultrastructure, 2016, http://biorxiv.org/content/early/2016/03/30/046300. doi:10.1101/046300.

Extracellular matrix mimicking scaffold promotes osteogenic stem cell differentiation: A new approach in osteoporosis research

Abstract

Background:

Objective:

Methods:

Results:

Conclusions:

Keywords

1. Introduction

2. Materials and methods

2.1. Electrospinning of hydroxyapatite meshes

2.2. Patients

2.4. hBMSC differentiation

2.5. Quantification of mineralization

2.6. Fluorescence activated cell sorting analysis of surface markers

2.7. Cell proliferation/viability

2.8. Electron microscopy

2.9. Statistical analysis

3. Results

3.1. Stability and general characterization of gelatine meshes with and without HA

3.4. Cell proliferation, differentiation and mineralization

Table 3 Cell proliferation of hBMSCs cultured in standard flat bottom wells or on HA gelatine meshes Day 1 Day 15 Day 21 4.826 ± 0.656 7.576 ± 1.032 6.494 ± 0.140 Well bottom 4.978 ± 0.584 7.795 ± 1.090 6.839 ± 1.320 HA gelatine mesh p > 0.05 p > 0.05 p > 0.05 P

5. Conclusion

Conflict of interest

References

Table 3
Cell proliferation of hBMSCs cultured in standard flat bottom wells or on HA gelatine meshes

Day 1 Day 15 Day 21

$4.826 \pm 0.656$ $7.576 \pm 1.032$ $6.494 \pm 0.140$ Well bottom

$4.978 \pm 0.584$ $7.795 \pm 1.090$ $6.839 \pm 1.320$ HA gelatine mesh

$p > 0.05$ $p > 0.05$ $p > 0.05$ P