Ex vivo intervertebral disc cultures: degeneration-induction methods and their implications for clinical translation.

Because low back pain is frequently a result of intervertebral disc degeneration (IVDD), strategies to regenerate or repair the IVD are currently being investigated. Often, ex vivo disc cultures of non-human IVD organs or tissue explants are used that usually do not exhibit natural IVDD. Therefore, degenerative changes mimicking those reported in human IVDD need to be induced. To support researchers in selecting ex vivo disc cultures, a systematic search was performed for them and their potential use for studying human IVDD reviewed. Five degeneration induction categories (proinflammatory cytokines, injury/damage, degenerative loading, enzyme, and other) were identified in 129 studies across 7 species. Methods to induce degeneration are diverse and can induce mild to severe degenerative changes that progress over time, as described for human IVDD. The induced degenerative changes are model-specific and there is no "one-fits-all" IVDD induction method. Nevertheless, specific aspects of human IVDD can be well mimicked. Currently, spontaneously degenerated disc cultures from large animals capture human IVDD in most aspects. Combinatorial approaches of several induction methods using discs derived from large animals are promising to recapitulate pathological changes on several levels, such as cellular behaviour, extracellular matrix composition, and biomechanical function, and therefore better mimic human IVDD. Future disc culture setups might increase in complexity, and mimic human IVDD even better. As ex vivo disc cultures have the potential to reduce and even replace animal trials, especially during preclinical development, advancement of such models is highly relevant for more efficient and cost-effective clinical translation from bench-to-bedside.

Fibrotic alterations in human annulus fibrosus correlate with progression of intervertebral disc herniation.

BACKGROUND: Intervertebral disc (IVD) herniation is characterized by annulus fibrosus failure (AF) in containing the nucleus pulposus (NP). IVD herniation involves cellular and extracellular matrix (ECM) alterations that have been associated with tissue fibrosis, although still poorly investigated. METHODS: Here, fibrotic alterations in human AF were evaluated, by characterizing the herniated ECM. Human AF samples (herniated lumbar IVD (n = 39, age 24-83) and scoliosis controls (n = 6, age 15-21)) were processed for transmission electron microscopy and histological/immunohistochemical analysis of fibrotic markers. Correlations between the fibrotic markers in AF ECM and the degree of NP containment (protused, contained and uncontained) and patients' age were conducted. RESULTS: Our results demonstrate that with herniation progression, i.e. loss of NP containment, human AF presents less stained area of sulphated glycosaminoglycans and collagen I, being collagen I fibres thinner and disorganized. On the other hand, fibronectin stained area and percentage of α-smooth muscle actin+ cells increase in human AF, while matrix metalloproteinase-12 (MMP12) production and percentage of macrophages (CD68+ cells) remain constant. These structural and biochemical fibrotic alterations observed in human AF with herniation progression occur independently of the age. CONCLUSIONS: The characterization of human AF here conducted evidence the presence of fibrosis in degenerated IVD, while highlighting the importance of considering the herniation progression stage, despite the patients' age, for a better understanding of the mechanisms behind AF failure and IVD herniation.

Mesenchymal stem cell secretome decreases the inflammatory response in annulus fibrosus organ cultures.

Mesenchymal stem/stromal cell (MSC)-based therapies have been proposed for back pain and disc degeneration, despite limited knowledge on their mechanism of action. The impact of MSCs/their secretome on annulus fibrosus (AF) cells and tissue was analysed in bovine AF organ cultures (AF-OCs) exposed to upper-physiological cyclic tensile strain (CTS, 9 %, 1 Hz, 3 h/d) and interleukin (IL)-1ß in a custom-made device. A 4 d treatment of the CTS + IL-1ß-stimulated AF-OCs with MSC secretome downregulated the expression of inflammation markers [IL-6, IL-8, prostaglandin-endoperoxide synthase 2 (PTGS2)], complement system regulators [cluster of differentiation (CD)46, CD55, CD59] and matrix metalloproteinase 1 but also of tissue inhibitors of metalloproteinases (TIMP-1, TIMP-2) and collagen type I. At the protein level, it was confirmed that IL-6, MMP-3 and collagen content was decreased in AF-OCs treated with the MSC secretome compared to the CTS + IL-1ß stimulation alone. 9 d after treatment, a biomechanical peel-force test showed that the annular adhesive strength was significantly decreased by the MSC secretome treatment. Overall, MSC secretome had a stronger impact on AF tissue than MSCs in co-culture. The secretome contributed to a decrease in the inflammatory and catabolic status of AF cells activated by CTS + IL-1ß and played a role in the regulation of the complement system. However, it also contributed to a decrease in collagen at the gene/protein level and in AF mechanical strength compared to the CTS + IL-1ß stimulation alone. Therefore, the use of MSC secretome requires further investigation regarding its influence on disc matrix properties.

Cyclic tensile stress of human annulus fibrosus cells induces MAPK activation: involvement in proinflammatory gene expression.

OBJECTIVE: To study the role of mitogen-activated protein kinases (MAPKs) in human annulus fibrosus (AF) cells subjected to cyclic tensile stress (CTS). DESIGN: An in vitro system for CTS studies was established using AF cultures on fibronectin-coated silicone dishes. MAPK phosphorylation was studied by western analysis, while gene expression was followed by qRT-PCR. DNA synthesis was assessed by both tritiated thymidine incorporation and flow cytometry, and collagen synthesis using tritiated proline incorporation and the protease-free collagenase method. RESULTS: All three MAPKs studied, i.e., ERK, SAPK/JNK, and p38 were found to be phosphorylated immediately after CTS application within physiological range. A second wave of phosphorylation appeared at later time points. MAPK activation was elevated at higher CTS magnitudes, but independent of the frequency. CTS did not stimulate DNA synthesis neither extracellular matrix turnover, but it stimulated the proinflammatory genes, COX-2, IL-6, and IL-8. This stimulation was more intense at the highest magnitude (8%) tested and at the median frequency (1 Hz) and time interval (12 h). Blocking of ERK, SAPK/JNK, and p38 MAPK inhibited the CTS-induced stimulation of COX-2 and IL-8, while IL-6 expression was mediated only by SAPK/JNK and p38 MAPK. CONCLUSIONS: We have described for the first time the activation of MAPKs in human AF cells in response to CTS and showed that it drives an inflammatory reaction. These observations shed light on the mechanisms of intervertebral disc (IVD) cell responses to mechanical stress, contributing to the understanding of disc pathophysiology and possibly to the design of novel therapeutic interventions.

Influence of extracellular osmolarity and mechanical stimulation on gene expression of intervertebral disc cells.

Intervertebral discs (IVD) have a higher extracellular osmolarity than most other tissues; moreover their osmolarity changes by around 25% during each diurnal cycle. In this study, changes in aggrecan, collagen I and collagen II expression of IVD cells were examined after exposure to osmotic environment alterations or mechanical stimulation under different osmotic conditions. Human and bovine IVD cells seeded in three-dimensional (3D) collagen type I matrices were cultured under hypo-osmotic (300 mOsm), iso-osmotic (400 mOsm), or hyperosmotic (500 mOsm) conditions. Osmolarity-induced changes in gene expression of IVD cells were measured after 5 days. Load-induced changes in gene expression under the different osmotic conditions were measured after application of hydrostatic pressure (0.25 MPa, 0.1 Hz, 30 min) or cyclic strain (4%, 1 Hz, 24 h). The results showed that IVD cells respond strongly to changes in the osmotic environment by altering mRNA expression. Human cells cultured over 5 days increased expression of aggrecan and collagen II in both nucleus and annulus cells under increasing osmolarity. In contrast, collagen I expression was inhibited at high osmolarity in both cell types. Mechanically induced alterations in gene expression appear to have only modest effects on matrix protein expression, but the same stimulus partly resulted in an inhibition or stimulation of gene expression, depending on the osmotic conditions. This study showed that the osmotic environment does not only have an appreciable effect on gene expression but also affects responses to mechanical stimuli. This suggests that the osmotic conditions cannot be ignored when examining physiological and pathological behavior of IVD cells.

Diurnal fluid expression and activity of intervertebral disc cells.

The intervertebral discs are large cartilaginous structures situated between the vertebral bodies, occupying around one third of the length of the spinal column. They act as the joints of the spine and carry mechanical load arising from body weight and muscle activity. Loads change with every alteration of posture and activity and the discs thus undergo a diurnal loading pattern with high loads on the discs during the day's activity and low loads on it at night during rest. As the disc is an osmotic system, around 25% of the disc's fluid is expressed and re-imbibed during each diurnal cycle with consequent changes in the osmotic environment of the disc cells. Here, present information on the effect of osmotic changes in disc cell metabolism is reviewed; results indicate that prevailing osmolarity is a powerful regulator of disc cell activity.

Tissue engineering of bone: effects of mechanical strain on osteoblastic cells in type I collagen matrices.

The aim of the present study was to investigate the effect of cyclic uniaxial mechanical strain on a human osteoblastic precursor cell line (hFOB 1.19) in three-dimensional type I collagen matrices. Cell seeded collagen constructs were mechanically stretched by a daily application of cyclic uniaxial strain using a special motor-driven apparatus and compared to unstretched controls. Expression of genes involved in cell proliferation and osteoblastic differentiation as well as matrix production were investigated by analyzing the mRNA of histone H4, core binding factor 1, alkaline phosphatase, osteopontin, osteocalcin, and collagen type I (Col I) up to a cultivation period of 3 weeks using real-time PCR. Cyclic stretching of cell seeded Col I matrices at a magnitude occurring in healing bone increased cell proliferation and slightly elevated the expression of nearly all investigated genes over unstrained controls at various time points. It was concluded that mechanical load promotes the proliferation and differentiation of osteoblastic precursor cells in a Col I matrix and that the application of mechanical stimuli may have a beneficial effect on in vitro tissue formation.

Fibroblast orientation to stretch begins within three hours.

Most connective tissue cells align in response to stretch. Previous studies have shown these responses occur within 12-14 h of initiation of stretch, but do not identify the time at which this orientation occurs, nor whether the orientation continues after cessation of stretch. To ascertain the earliest times at which fibroblast orientation occurs, we cultured primary human fibroblasts on deformable culture dishes and stretched (1 Hz, 8% uniaxial strain) them for up to 24 h. We photographed the cells at 0.5, 1-6, 8, 10, 12, 14, 16, and 24 h. Similarly cells were photographed at 1-3, or 4 h after cessation of stretch for stretch durations of 1, 2, and 3 h. Orientation of cells were ascertained by an interactive computer program. The fibroblasts began to orient by 2-3 h and orientation appeared nearly complete by 24 h. Cultures stretched for 2 or 3 h continued to exhibit greater degrees of orientation (compared to controls) for 2 or 3 h respectively after cessation of stretch. We conclude fibroblasts begin to orient within 3 h of initiation of stretch, and that they continue to orient for several hours after cessation of stretch.

Cell alignment is induced by cyclic changes in cell length: studies of cells grown in cyclically stretched substrates.

Many types of cells, when grown on the surface of a cyclically stretched substrate, align away from the stretch direction. Although cell alignment has been described as an avoidance response to stretch, the specific deformation signal that causes a cell population to become aligned has not been identified. Planar surface deformation is characterized by three strains: two normal strains describe the length changes of two initially perpendicular lines and one shear strain describes the change in the angle between the two lines. The present study was designed to determine which, if any, of the three strains was the signal for cell alignment. Human fibroblasts and osteoblasts were grown in deformable, rectangular, silicone culture dishes coated with ProNectin, a biosynthetic polymer containing the RGD ligand of fibronectin. 24 h after plating the cells, the dishes were cyclically stretched at 1 Hz to peak dish stretches of 0% (control), 4%, 8%, and 12%. After 24 h of stretching, the cells were fixed, stained, and their orientations measured. The cell orientation distribution was determined by calculating the percent of cells whose orientation was within each of eighteen 5 degrees angular intervals. We found that the alignment response was primarily driven by the substrate strain which tended to lengthen the cell (axial strain). We also found that for each cell type there was an axial strain limit above which few cells were found. The axial strain limit for fibroblasts, 4.2 +/- 0.4%, (mean +/- 95% confidence), was lower than for osteoblasts, 6.4 +/- 0.6%. We suggest that the fibroblasts are more responsive to stretch because of their more highly developed actin cytoskeleton.

[In vitro cell behavior of human osteoblasts after physiological dynamic stretching]. / In-vitro-Zellverhalten humaner Osteoblasten nach physiologischen dynamischen Dehnungen.

The cell activity of human bone derived cell cultures was studied after mechanical stimulation by cyclic strain at a magnitude occurring in physiologically loaded bone tissue. Monolayers of subconfluently grown human bone derived cells were stretched in rectangular silicone dishes with cyclic uniaxial movement along their longitudinal axes. Strain was applied over two days for 30 min per day with a frequency of 1 Hz and a strain magnitude of 1000 mustrain. Cyclic stretching of the cells resulted in an increased proliferation (10-48%) and carboxyterminal collagen type I propeptide release (7-49%) of human cancellous bone derived osteoblasts while alkaline phosphatase activity and osteocalcin release were significantly reduced by 9-25% and 5-32% respectively. These results demonstrate that cyclic strain at physiologic magnitude leads to an increase of osteoblast activities related to matrix production while those activities which are characteristic for the differentiated osteoblast and relevant for matrix mineralization are decreased.

In vitro effects of dynamic strain on the proliferative and metabolic activity of human osteoblasts.

AIM OF THE STUDY: It has been well shown by human and animal studies that mechanical load is an important regulator of skeletal mass and architecture. However, cellular reactions which adapt bone tissue to the mechanical environment are not definitively determined. For this purpose we studied the cell activity of human bone derived cell cultures after mechanical stimulation by cyclic, uniaxial strain at a magnitude occurring in normal loaded bone tissue. MATERIALS AND METHODS: Human osteoblasts were isolated from cancellous bone biopsies of 5 different donors. Cell seeding was made in DMEM in a density of 10.000 cells/cm(2) on deformable culture dishes for three days prior to initiating cell stretching at 1000 microstrain, 1Hz for 1800 cycles for two subsequent days with an especially developed cell stretching device. 48h after the second stimulation cells were harvested and cell number was determined with a Coulter Counter. Cell bound alkaline phosphatase activity was analyzed in cell lysates by a colorimetric assay, osteocalcin and CICP (procollagen I propeptide) production were analyzed in cell supernatants with ELISAs. Three parallel cultures were tested. STATISTICS: Wilcoxon. RESULTS: In all experiments mechanical stimulation resulted in a significant increase in cell number (10-48%) and CICP release (7-49%). Simultaneously a significant decrease in alkaline phosphatase activity (9-25%) and osteocalcin release (5-32%) could be observed. CONCLUSIONS: The results demonstrate that cyclic strain at physiologic magnitude leads to an increase of early osteoblast activities related to matrix production while those activities which are characteristic for the differentiated osteoblast and relevant for matrix mineralization are decreased. These new findings confirm in vivo observations about the importance of dynamic strain for bone formation during fracture healing and bone remodeling and could contribute to the optimization of fracture healing.

Dynamic cell stretching increases human osteoblast proliferation and CICP synthesis but decreases osteocalcin synthesis and alkaline phosphatase activity.

The cell activity of human-bone-derived cell cultures was studied after mechanical stimulation by cyclic strain at a magnitude occurring in physiologically loaded bone tissue. Monolayers of subconfluently grown human-bone-derived cells were stretched in rectangular silicone dishes with cyclic predominantly uniaxial movement along their longitudinal axes. Strain was applied over two days for 30 min per day with a frequency of 1 Hz and a strain magnitude of 1000 microstrain. Cyclic stretching of the cells resulted in an increased proliferation (10-48%) and carboxyterminal collagen type I propeptide release (7-49%) of human-cancellous bone-derived osteoblasts while alkaline phosphatase activity and osteocalcin release were significantly reduced by 9-25 and 5-32%, respectively. These results demonstrate that cyclic strain at physiologic magnitude leads to an increase of osteoblast activities related to matrix production while those activities which are characteristic for the differentiated osteoblast and relevant for matrix mineralization are decreased.

Cell orientation induced by extracellular signals.

Cells like fibroblasts and osteoblasts are oriented by different extracellular guiding signals like an electric field, a bent surface, and a periodically stretched surface. An automatic controller is responsible for the cell alignment. The controller contains both a deterministic and a stochastic signal. The following machine properties were determined: (1) The angle dependence of the cellular signal transformer is cos 2(psi 0 - psi). (2) The set point of the automatic controller is psi 0 = +/- 90 degrees. The cells like to orient their long axis perpendicular to the direction of the applied guiding signal. (3) The signal transformer measures the extracellular signal in a quadratic fashion. The cells cannot register the sign of the guiding field. (4) The stochastic signal in the automatic controller can be quantified by a characteristic time (approximately 130 min for fibroblasts). (5) The extracellular signal is registered in cell-made standards (ratio of the deterministic and stochastic signal equals one): 0.3 +/- 0.05 V/mm for human fibroblasts (electric field) and 85 +/- 3 microns for human fibroblasts and osteoblasts (cyclindrically bent surface). (6) The lag-time in the signal transduction system of fibroblasts is approximately 4 min.

Effects of mechanical factors on the fracture healing process.

An interdisciplinary study based on animal experiments, cell culture studies, and finite element models is presented. In a sheep model, the influence of the osteotomy gap size and interfragmentary motion on the healing success was investigated. Increasing gap sizes delayed the healing process. Increasing movement stimulated callus formation but not tissue quality. Typical distributions of intramembranous bone, endochondral ossification, and connective tissue in the fracture gap are quantified. The comparison of the mechanical data determined by a finite element model with the histologic images allowed the attribution of certain mechanical conditions to the type of tissue differentiation. Intramembranous bone formation was found for strains smaller than approximately 5% and small hydrostatic pressure (< 0.15 MPa). Strains less than 15% and hydrostatic pressure more than 0.15 MPa stimulated endochondral ossification. Larger strains led to connective tissue. Cell culture studies on the influence of strain on osteoblasts supported these findings. Proliferation and transforming growth factor beta production was increased for strains up to 5% but decreased for larger strains. Osteoblasts under larger strains (> 4%) turned away from the principal strain axis and avoided larger deformations. It is hypothesized that gap size and the amount of strain and hydrostatic pressure along the calcified surface in the fracture gap are the fundamental mechanical factors involved in bone healing.

Human osteoblasts from younger normal and osteoporotic donors show differences in proliferation and TGF beta-release in response to cyclic strain.

Mechanical stimulation of bone tissue by physical activity stimulates bone formation in normal bone and may attenuate bone loss of osteoporotic patients. However, altered responsiveness of osteoblasts in osteoporotic bone to mechanical stimuli may contribute to osteoporotic bone involution. The purpose of the present study was to investigate whether osteoblasts from osteoporotic patients and normal donors show differences in proliferation and TGF beta production in responses to cyclic strain. Human osteoblasts isolated from collagenase-treated bone explants of 10 osteoporotic patients (average age 70 +/- 6 yr) and 8 normal donors (average age 54 +/- 10 yr) were plated into elastic rectangular silicone dishes. Subconfluent cultures were stimulated by cyclic strain (1%, 1 Hz) in electromechanical cell stretching apparatus at three consecutive days for each 30 min. The cultures were assayed for proliferation, alkaline phosphatase activity and TGF beta release in each three parallel cultures. In all experiments, osteoblasts grown in the same elastic dishes but without mechanical stimulation served as controls. Significant differences between stimulated cultures and unstimulated controls were determined by a paired two-tailed Wilcoxon test. In comparison to the unstimulated controls, osteoblasts from normal donors significantly increased proliferation (p = 0.025) and TGF beta secretion (p = 0.009) into the conditioned culture medium. In contrast, osteoblasts from osteoporotic donors failed to increase both proliferation (p > 0.05) and TGF beta release (p > 0.05) in response to cyclic strain. Alkaline phosphatase activity was not significantly affected (p > 0.05) in normal as well as osteoporotic bone derived osteoblasts. These findings suggest a different responsiveness to 1% cyclic strain of osteoblasts isolated from normal and osteoporotic bone that could be influenced by both the disease of osteoporosis and the higher average age of the osteoporotic patient group. While osteoblasts from osteoporotic donors failed to increase proliferation and TGF beta release under the chosen mechanical strain regimen that stimulated both parameters in normal osteoblasts, it is possible that some other strain regimen would provide more effective stimulation of osteoporotic cells.

Ilizarov callus distraction produces systemic bone cell mitogens.

We tested the hypothesis that the application of strain during callus distraction induces systemic osteoblast stimulating factors that enhance osteoblast activity both locally and systemically. To study the systemic occurence of strain-induced osteoblast stimulating factor during callus distraction, we investigated the mitogenic capacity of sera from 12 patients who had undergone callus distraction on the osteoblastic cell line SaOS-2 (part I). Serum samples from six patients who had undergone rigidly fixed high tibial osteotomy (i.e., without distraction) served as controls. The sera were assayed for platelet-derived growth factor and transforming growth factor-beta. In part II of the study, the in vitro effects of mechanical strain were investigated in a simplified model by cyclic stimulation of osteoblast cultures isolated from cortical bone explants from the same patients; a specially developed apparatus was used for cell-stretching. Sera taken during the third to fourth week of callus distraction demonstrated a significant increase in proliferation of SaOS-2 cells (p < 0.005). In contrast, sera from patients who had had an osteotomy failed to induce or decreased the mitogenic capacity of SaOS-2 cells. The concentration of platelet-derived growth factor increased significantly (p < 0.01) in sera from both the patients who had undergone callus distraction and the controls who had had osteotomy. However, the level of transforming growth factor-beta was increased (p < 0.05) in the sera from the patients who had distraction (sera that stimulated proliferation of SaOS-2 cells), but the level was not increased in the sera from patients who had osteotomy.(ABSTRACT TRUNCATED AT 250 WORDS)

Cyclic stretching of human osteoblasts affects proliferation and metabolism: a new experimental method and its application.

We developed an experimental system to stimulate cell cultures by uniform and cyclic biaxial strain of the cell culture surface. The studies reported here were designed to determine the uniformity of the strain distribution, the suitability of the surface for the growth of human osteoblasts, and the effects of strain magnitude on cell proliferation and alkaline phosphatase (AP) activity. Subconfluent cell cultures were grown in rectangular silicone dishes that were stretched cyclically (1 Hz) in the long axis by an electromechanical apparatus that controlled peak stretch and cycle frequency. We applied cyclic strains (1.0, 2.4, 5.3, and 8.8% surface strains) for 15 minutes per day on 3 consecutive days. Phase contrast microscopy confirmed the transfer of dish surface strain to the cells. Stretching of the dish resulted in a homogeneous strain distribution that deviated approximately 0.05% from the applied strain. In comparison with plastic dishes, there was a 20% reduction of cell proliferation on the silicone substrate whereas morphology, AP activity, and total protein content of the cells were similar. The proliferation of human osteoblasts was increased significantly (16.4-100%) by 1% strains, although higher strain magnitudes had lesser (nonsignificant) effects or decreased the mitotic activity of the cells. AP and lactate dehydrogenase activities were not influenced significantly by cyclic strains. This study demonstrates that the cell stretching system is suitable for the investigation of the effects of well defined cyclic strains.