Hannu M. Karjalainen

cDNA array reveals mechanosensitive genes in chondrocytic cells under hydrostatic pressure.

Hydrostatic pressure (HP) has a profound effect on cartilage metabolism in normal and pathological conditions, especially in weight-bearing areas of the skeletal system. As an important component of overall load, HP has been shown to affect the synthetic capacity and well-being of chondrocytes, depending on the mode, duration and magnitude of pressure. In this study we examined the effect of continuous HP on the gene expression profile of a chondrocytic cell line (HCS-2/8) using a cDNA array containing 588 well-characterized human genes under tight transcriptional control. A total of 51 affected genes were identified, many of them not previously associated with mechanical stimuli. Among the significantly up-regulated genes were immediate-early genes, and genes involved in heat-shock response (hsp70, hsp40, hsp27), and in growth arrest (GADD45, GADD153, p21(Cip1/Waf1), tob). Markedly down-regulated genes included members of the Id family genes (dominant negative regulators of basic helix-loop-helix transcription factors), and cytoplasmic dynein light chain and apoptosis-related gene NIP3. These alterations in the expression profile induce a transient heat-shock gene response and activation of genes involved in growth arrest and cellular adaptation and/or differentiation.

Neuronal cells show regulatory differences in the hsp70 gene response.

The synthesis of heat shock proteins (Hsps), encoded by heat shock genes, is increased in response to various stress stimuli. Hsps function as molecular chaperones, they dissociate cytotoxic stress-induced protein aggregates within cells and ensure improved survival. Induction of heat shock genes is mainly regulated at the transcriptional level. The stress responsive transcription factor, heat shock factor 1 (HSF1), is involved in the transcriptional induction of the heat shock genes. Our objective was to examine how hsp70 genes are regulated in different transformed and primary neurons upon exposure to elevated temperature. Our findings reveal that the Hsp70 response is regulated at the translational level in Neuro-2a neuroblastoma cells, while the IMR-32 neuroblastoma cells respond to stress by the classical HSF1-driven transcriptional regulatory mechanism. Primary rat hippocampal neurons show a lack of HSF1 and induction of the hsp70 gene. These observations suggest that neuronal cells display different hsp70 gene expression patterns which range from undetected response to transcriptional and posttranscriptional regulation during heat stress.

Computed tomography detects changes in contrast agent diffusion after collagen cross-linking typical to natural aging of articular cartilage.

OBJECTIVE The effect of threose-induced collagen cross-linking on the mechanical and diffusive properties of cartilage was investigated in vitro. In particular, we investigated the potential of Contrast Enhanced Computed Tomography (CECT) to detect changes in articular cartilage after increased collagen cross-linking, which is an age-related phenomenon. METHODS Osteochondral plugs (Ø=6.0 mm, n=28) were prepared from intact bovine patellae (n=7). Two of the four adjacent samples, prepared from each patella, were treated with threose to increase the collagen cross-linking, while the other two specimen served as paired controls. One sample pair was mechanically tested and then mechanically injured using a material testing device. Contrast agent [ioxaglate (Hexabrix™)] diffusion was imaged in the other specimen pair for 25 h using CECT. Water fraction, collagen and proteoglycan content, collagen network architecture and the amount of cross-links [hydroxylysyl pyridinoline (HP), lysyl pyridinoline (LP) and pentosidine (Pent)] of the samples were also determined. RESULTS Cartilage collagen cross-linking, both Pent and LP, were significantly (P<0.001) increased due to threose treatment. CECT could detect the increased cross-links as the contrast agent penetration and the diffusion flux were significantly (P<0.05) lower in the threose treated than in untreated samples. The equilibrium modulus (+164%, P<0.05) and strain dependent dynamic modulus (+47%, P<0.05) were both significantly greater in the threose treated samples than in reference samples, but there was no association between the initial dynamic modulus and the threose treatment. The water fraction, proteoglycan and collagen contents, as well as collagen architecture, were not significantly altered by the threose treatment. CONCLUSIONS To conclude, the CECT technique was found to be sensitive at detecting changes in cartilage tissue due to increased collagen cross-linking. This is important since increased cross-linking has been proposed to be related to the increased injury susceptibility of tissue.

Geldanamycin increases 4-hydroxynonenal (HNE)-induced cell death in human retinal pigment epithelial cells

Development of age-related macular degeneration (AMD) is associated with functional abnormalities and cell death in retinal pigment epithelial (RPE) cells attributable to oxidative stress. To minimize the adverse effects of oxidative stress, cells activate their defence systems, e.g., via increased expression of heat shock protein (Hsp), activation of stress sensitive AP-1 and NF-kappaB transcription factors. In this study, we examined the accumulation of Hsp70 protein, activation of AP-1 and NF-kappaB transcription factors in human ARPE-19 cells subjected to a 4-hydroxynonenal (HNE)-induced oxidative stress. In addition, the influence of Hsp90 inhibitor geldanamycin (GA) was studied in HNE-treated cells. Mitochondrial metabolic activity and apoptosis were determined to evaluate cell death in the ARPE-19 cells. The ARPE-19 cells showed increased accumulation of Hsp70 protein before of the cytotoxic hallmarks appearing in response to HNE. In contrast, increased DNA-binding activities of AP-1 or NF-kappaB transcription factors were not seen under HNE insults. Interestingly, GA significantly increased cell death in the HNE-treated cells, which was involved in caspase-3 independent apoptosis. This study reveals that the Hsps have an important role in the cytoprotection of RPE cells subjected to HNE-derived oxidative stress.

HAS3-induced accumulation of hyaluronan in 3D MDCK cultures results in mitotic spindle misorientation and disturbed organization of epithelium

The amount of hyaluronan (HA) is low in simple epithelia under normal conditions, but during tumorigenesis, trauma or inflammation HA is increased on the epithelial cells and surrounding stroma. Excessive HA in epithelia is suggested to interfere with cell–cell adhesions, resulting in disruption of the epithelial barrier function. In addition, stimulated HA synthesis has been correlated with epithelial-to-mesenchymal transition and invasion of cancer cells. However, the effects of HA overload on normal epithelial morphogenesis have not been characterized in detail. Madin-Darby canine kidney (MDCK) cells form polarized epithelial cysts, when grown in a 3-dimensional (3D) matrix. These cells were used to investigate whether stimulated HA synthesis, induced by stable overexpression of GFP-HAS3, influences cell polarization and epithelial morphogenesis. GFP-HAS3 expression in polarized MDCK cells resulted in active HA secretion at apical and basolateral membrane domains. HA-deposits interfered with the formation of cell–cell junctions, resulting in impaired barrier function. In 3D cyst cultures, HA accumulated into apical lumina and was also secreted from the basal side. The HAS3-expressing cysts failed to form a single lumen and instead displayed multiple small lumina. This phenotype was correlated with aberrant mitotic spindle orientation in dividing cells. The results of this study indicate that excess pericellular HA disturbs the normal cell–cell and cell–ECM interactions in simple epithelia, leading to aberrant epithelial morphogenesis. The morphological abnormalities observed in 3D epithelial cultures upon stimulated HAS3 expression may be related to premalignant changes, including intraluminal invasion and deregulated epithelialization, probably mediated by the mitotic spindle orientation defects.

Diffusion of ionic and non-ionic contrast agents in articular cartilage with increased cross-linking : contribution of steric and electrostatic effects

OBJECTIVE To investigate the effect of threose-induced collagen cross-linking on diffusion of ionic and non-ionic contrast agents in articular cartilage. DESIGN Osteochondral plugs (Ø=6mm) were prepared from bovine patellae and divided into two groups according to the contrast agent to be used in contrast enhanced computed tomography (CECT) imaging: (I) anionic ioxaglate and (II) non-ionic iodixanol. The groups I and II contained 7 and 6 sample pairs, respectively. One of the paired samples served as a reference while the other was treated with threose to induce collagen cross-linking. The equilibrium partitioning of the contrast agents was imaged after 24h of immersion. Fixed charge density (FCD), water content, contents of proteoglycans, total collagen, hydroxylysyl pyridinoline (HP), lysyl pyridinoline (LP) and pentosidine (Pent) cross-links were determined as a reference. RESULTS The equilibrium partitioning of ioxaglate (group I) was significantly (p=0.018) lower (-23.4%) in threose-treated than control samples while the equilibrium partitioning of iodixanol (group II) was unaffected by the threose-treatment. FCD in the middle and deep zones of the cartilage (p<0.05) and contents of Pent and LP (p=0.001) increased significantly due to the treatment. However, the proteoglycan concentration was not systematically altered after the treatment. Water content was significantly (-3.5%, p=0.007) lower after the treatment. CONCLUSIONS Since non-ionic iodixanol showed no changes in partition after cross-linking, in contrast to anionic ioxaglate, we conclude that the cross-linking induced changes in charge distribution have greater effect on diffusion compared to the cross-linking induced changes in steric hindrance.

Proteomics of chondrocytes with special reference to phosphorylation changes of proteins in stretched human chondrosarcoma cells

For proteomic analysis, cartilage molecular composition is a challenging mixture of highly glycosylated proteoglycans and triple-helical collagens, which constitute the major part of cartilage macromolecules. Selective separation of these molecules from the minor components is generally needed before mass spectrometry-based identification of lower-abundancy proteins is possible. The cell density of cartilage is also very low, therefore, cell cultures offer an easier approach to study cellular responses of chondrocytic cells, e.g., to mechanical stimuli. In this study, we investigated the phosphorylation events in human chondrosarcoma cells during cellular stretching. Human chondrosarcoma cells were stretched to 8% strain at a frequency of 1 Hz. One set of experiments included cellular stretching which lasted 2 hours, and the other one included experiments of 2 hours daily treatment for up to 3 days. Two-dimensional polyacrylamide gel electrophoresis combined with chromatographic phosphoprotein pre-enrichment and electrospray ionization mass spectrometry-based protein identification was used to reveal changes of phosphoproteins in cells exposed to cyclic stretching. We discovered that 2 hours cyclic stretching increased the phosphorylation of moesin, elongation factor eEF1D and leprecan, while the phosphorylation of elongation factor eEF1B decreased after cellular stretching. Western blot analyses with phospho-specific antibodies suggested that stretching induces phosphorylation of ERK of the MAP kinase pathway, but did not induce phosphorylation of phosphatidylinositol 3-kinase. In conclusion, the proteomic approach revealed that cellular stretching induced specific phosphorylation changes in chondrosarcoma cells.

High hydrostatic pressure inhibits the biosynthesis of eukaryotic elongation factor-2

High continuous hydrostatic pressure is known to inhibit the total cellular protein synthesis. In this study, our goal was to identify pressure‐regulated proteins by using two dimensional gel electrophoresis and mass spectrometry. This analysis showed that under 30 MPa continuous hydrostatic pressure the biosynthesis of eukaryotic elongation factor‐2 (eEF‐2) was inhibited both in HeLa carcinoma and T/C28a4 chondrocytic cell lines. Western blot analysis of HeLa cells revealed that the cellular protein level of eEF‐2 decreased by 40%–50% within 12 h of the pressure treatment. However, the steady‐state mRNA level of eEF‐2 was not affected by the pressure. Cycloheximide addition after 4 h‐pressure treatment suggested that the half‐life of eEF‐2 protein was shorter in pressurized cells. eEF‐2 is responsible for the translocation of ribosome along the specific mRNA during translation, and its phosphorylation prevents the ribosomal translocation. Therefore, increased phosphorylation of eEF‐2 was considered as one mechanism that could explain the reduced level of protein synthesis in pressurized HeLa cell cultures. However, Western blot analysis with an antibody recognizing the Thr56‐phosphorylated form of eEF‐2 showed that phosphorylation of eEF‐2 was not elevated in pressurized samples. In conclusion, the inhibition of protein synthesis under high pressure occurs independent of the phosphorylation of eEF‐2. However, this inhibition may result from the decrease of cellular eEF‐2 protein.

Genome-wide microarray analysis of MG-63 osteoblastic cells exposed to ultrasound

It is well documented that low intensity pulsed ultrasound can be clinically used to accelerate bone fracture healing. Additionally, in vitro studies have shown that ultrasound can, for instance, increase mineralization, collagen production and alkaline phosphatase activity in osteoblasts. Despite the extensive research on the subject, the exact mechanism of ultrasound effect on bone cell gene regulation has not yet been deduced. In this study, we made an effort to reveal the features of genome-wide transcriptional response of osteoblast-type cells to ultrasound. MG-63 osteoblastic cell transcriptome was analyzed with whole genome microarray either 6 or 24 h after 30 min long exposure to 1.035 MHz pulsed ultrasound with three different acoustic pressures. Special attention was paid to the experimental design to minimize thermal effects and unwanted reflections of ultrasound. Microarray analysis suggested that ultrasound affects the genes involved with cellular membranes, and regulation of transcription as well. Several plasma membrane solute carriers were also regulated by ultrasound. It also changed the transcript level of several transcription factors belonging to the zinc finger proteins. However, ultrasound did not clearly promote genes involved with osteoblast differentiation.

Chondrocytic cells express the taurine transporter on their plasma membrane and regulate its expression under anisotonic conditions

Taurine is a small organic osmolyte which participates in cell volume regulation. Chondrocytes have been shown to accumulate and release taurine; in bone, taurine participates in bone metabolism. However, its role in skeletal cells is poorly understood, especially in chondrocytes. This study investigated the regulation of taurine transporter in chondrocytic cells. We examined the transcriptional regulation of the taurine transporter under anisotonia by reporter gene and real-time RT-PCR assays. The effect of providing supplementary taurine on cell viability was evaluated with the lactate dehydrogenase release assay. The localization of the taurine transporter in human chondrosarcoma cells was studied by overexpressing a taurine transporter-enhanced green fluorescent protein. We observed that the transcription of the taurine transporter gene was up-regulated in hypertonic conditions. Hyperosmolarity-related cell death could be partly abolished by taurine supplementation in the medium. As expected, the fluorescently labeled taurine transporter localized at the plasma membrane. In polarized epithelial MDCK cells, the strongest fluorescence signal was located in the lateral cell membrane area. We also observed that the taurine transporter gene was expressed in several human tissues and malignant cell lines. This is the first study to present information on the transcriptional regulation of taurine transporter gene and the localization of the taurine transporter protein in chondrocytic cells.