Synergistic Collagenase Expression and Cartilage Collagenolysis Are Phosphatidylinositol 3-Kinase/Akt Signaling-dependent*

The phosphatidylinositol 3-kinase (PI3K) signaling pathway has emerged as a major regulator of cellular functions and has been implicated in several pathologies involving remodeling of extracellular matrix (ECM). The end stage of inflammatory joint diseases is characterized by excessive ECM catabolism, and in this study we assess the role of PI3K signaling in the induction of collagenolytic matrix metalloproteinases (MMPs) in human chondrocytes. We used the most potent cytokine stimulus reported to promote cartilage ECM catabolism, namely interleukin-1 (IL-1) in combination with oncostatin M (OSM). Both OSM and IL-6 (in the presence of its soluble receptor), but not IL-1 nor leukemia inhibitory factor, induced Akt phosphorylation in human chondrocytes. Inhibition of PI3K signaling using LY294002 blocked IL-1+OSM-mediated Akt phosphorylation, induction of MMP-1 and MMP-13, and cartilage collagenolysis. To further explore the role of downstream substrates within the PI3K pathway, complementary use of small molecule inhibitors and specific small interfering RNAs demonstrated that the PI3K subunit p110α and Akt1 were required for MMP-1 mRNA induction. MMP-13 induction was also reduced by loss of function of these molecules and by a lack of p110δ, 3-phosphoinositide-dependent kinase-1 or Akt3. We therefore propose that the activities of specific elements of the PI3K signaling pathway, including Akt, are necessary for the synergistic induction of MMP-1 and MMP-13 and the cartilage breakdown stimulated by IL-1+OSM. Our data provide new insight into the mechanism of synergy between IL-1 and OSM and highlight new therapeutic targets for inflammatory joint diseases that aim to repress the expression of collagenases.

Phosphatidylinositol 3-kinase (PI3K) 3 comprises a group of lipid kinases that catalyze the phosphorylation of the 3-position of the inositol moiety of phosphoinositides at the cell membrane (1). Activation of class I PI3K signaling plays a vital role in the response of many cytokine and growth factor receptors to control cellular processes such as growth, survival, proliferation, differentiation, and migration (1). Activity of the downstream serine/threonine kinase, Akt (protein kinase B), is of key importance for these events. This PI3K subgroup is further subdivided: class IA comprises heterodimers consisting of a p110 catalytic subunit (␣, ␤, or ␦ isoforms), closely associated with a p50, p55, or p85 regulatory subunit, which is responsible for recruitment to tyrosine kinase-associated receptors activated by cytokines and growth factors. In class IB enzymes, usually activated by G-protein-coupled receptors, the catalytic subunit is p110␥, and the regulatory subunit is p84 or p101 (1). Akt is a pivotal kinase in PI3K signaling, and mechanisms exist to limit Akt activation. Positive or negative regulation of PI3K/Akt signaling enables functional differences between receptors which, for example, determine the outcome of T cell activation (2). There are three mammalian Akt family members involved in a plethora of cellular signaling events. These isoforms have central roles in a variety of human cancers, with effects on tumor initiation and progression as well as metastasis. They also contribute to tumorigenesis at multiple levels such as regulation of cell motility and migration, and cellular interactions with the extracellular matrix (ECM) (3). Cross-talk occurs between Akt and other signaling pathways such as the mitogen-activated protein kinases (MAPKs), which are themselves activated by pro-inflammatory stimuli (4). Thus, PI3K pathway activation can influence other signaling pathways; combined with the ability of Akt to regulate the subcellular localization of downstream substrates (and hence their functionality, Ref. 5), this pathway is an important modulator of gene expression in inflammatory milieu where multiple cytokines are present.
Inflammation is a major characteristic of joint diseases such as rheumatoid arthritis (RA) and osteoarthritis (OA). Despite having different aetiologies, inflammatory mediators released by infiltrating immune cells as well as resident joint cells induce alterations in gene expression that can lead to ECM degrada-tion (6,7). Cytokines such as interleukin(IL-)1, IL-17, and tumor necrosis factor ␣ (TNF␣) are the key mediators thought to be involved in promoting inflammatory responses in such destructive joint diseases (8,9). We have shown that the IL-6family cytokine oncostatin M (OSM) and IL-6 markedly exacerbate the catabolic potential of these mediators, synergistically promoting cartilage ECM catabolism both in vitro and in vivo (10 -14).
Chondrocytes are the only resident cell type in normal articular cartilage and function to preserve homeostasis. This is achieved by regulating the expression of ECM components with catabolic factors such as the matrix metalloproteinases (MMPs), which collectively can degrade all the ECM macromolecules. During the degradative phases of inflammatory joint diseases, chondrocytes are stimulated to secrete elevated levels of MMPs that, once activated, mediate the proteolysis of tendon, bone, and cartilage (15,16). MMP-1 and MMP-13 are collagenolytic MMPs that have been most strongly associated with cartilage collagenolysis, a key proteolytic event in inflammatory joint disease, because it is essentially irreversible (17). A marked synergistic induction of collagenase gene expression occurs in human chondrocytes following stimulation with IL-1ϩOSM (13,18,19), and this has been proposed to be via the interplay of signal transduction pathways whereby signal transducers and activators of transcription (STAT), c-Fos, and alterations in activator protein-1 (AP-1) composition are important (20). MMP gene expression is certainly reliant on transcription factors such as erythroblastosis 26 and nuclear factor-B as well as AP-1 (21), although synergistic MMP gene expression is likely to involve other signaling pathways that impinge on one or more of these factors. Thus, targeting specific pathways such as those activated by OSM may have the potential to suppress collagenolytic MMP gene expression during inflammation.
We propose that elements of the PI3K pathway are important determinants for the synergistic induction of collagenolytic MMPs in cartilage. In this study, we demonstrate activation of Akt by OSM (but not IL-1), and that MMP-1 and MMP-13 gene expression are dependent on distinct sets of signaling molecules within the PI3K pathway. These data therefore further support the differential regulation of these important collagenases (43), confirm that OSM-activated PI3K signaling is an important contributor to MMP synergy, and that specific ablation of PI3K signaling events could function to preserve cartilage homeostasis during inflammation.

EXPERIMENTAL PROCEDURES
Materials-All chemicals, including AS-604850, were obtained from Sigma unless otherwise stated and were of the highest purity available. All cytokines used were recombinant human. IL-1␣ was a generous gift from Dr. Keith Ray (Glaxo-SmithKline, Stevenage, UK). OSM and LIF were kindly donated by Prof. John Heath (Department of Biochemistry, University of Birmingham, Edgbaston, UK), while IL-6 and its soluble receptor (sIL-6R) were purchased from R&D Systems Ltd. (Abingdon, UK). LY294002, and Akt inhibitors IV and VIII were from Merck Chemicals (Nottingham, UK), compound 15e and TGX-221 were from Axxora (Nottingham, UK). All kinase inhibitors and siRNA reagents were screened for toxicity using the XTT assay of metabolic activity (Roche Applied Sciences), and used at concentrations that did not affect cell viability over a 24-h period.
Chondrocytes-Human chondrocytes were obtained by the enzymatic digestion of macroscopically normal articular cartilage from OA patients undergoing joint replacement surgery as described (44). All subjects gave informed consent, and the study was approved by the Newcastle and North Tyneside Joint Ethics Committee. Bovine cartilage was dissected from nasal septi obtained from a local abattoir as described (45). Chondrocytes were only used as primary cultures, or for siRNA transfection, after a single passage.
Cartilage Degradation Assay-Bovine nasal septum cartilage was dissected into ϳ2 ϫ 2 ϫ 2-mm discs, plated into 24-well tissue culture plates (3 discs per well, n ϭ 4) in serum-free medium and incubated for 14 days in the presence of IL-1 Ϯ OSM (ϮLY294002 including a 24-h preincubation), changing medium after 7 days as previously described (45). The cartilage remaining at day 14 was digested with papain (45), and all samples were stored at Ϫ20°C until assay.
Viability of cartilage explants was assessed by screening for the production of lactate dehydrogenase (LDH) using the Cytotox 96 assay (Promega, Southampton, UK). No increase in LDH levels with any of the treatments including inhibitors has been found (data not shown).
Proteoglycan, Collagen, and Collagenolytic Activity Assays-For proteoglycan release, a measurement of sulfated glycosaminoglycans was used as described previously (47). Hydroxyproline measurements (46) were used as an estimate of cartilage collagen, and the cumulative release calculated and expressed as a percentage of the total for each well (47). Collagenolytic activity present in the culture medium from cartilage explants was determined using a diffuse fibril assay with 3 H-acetylated collagen (48). One unit of collagenase activity degrades 1 g of collagen per min at 37°C.
Real-time PCR of Relative mRNA Levels-RNA was stabilized in cell lysates in a 96-well format and cDNA synthesized using the Cells-to-Signal TM kit as directed. For TaqMan or SYBR Green PCR, mRNA levels for each gene were obtained from standard curves and corrected using 18 S ribosomal RNA levels. Cycling conditions (7900HT system, Applied Biosystems, Foster City, CA) for SYBR Green PCR (using Takara SYBR ExTaq premix; Lonza Biologics, Cambridge, UK) were 95°C 10 s, then 40 cycles of 95°C 5 s, then 60°C 30 s, followed by a standard dissociation curve analysis. Cycling conditions for Taqman PCR (Jumpstart Taq Readymix; Sigma) were 2 min at 50°C, 10 min at 95°C, then 40 cycles of 15 s at 95°C, and 1 min at 60°C. Primers and probe sequences for Taqman PCR were: MMP-1:

OSM Stimulates Akt Phosphorylation in Human
Chondrocytes-Stimulation of primary human articular chondrocytes with OSM and IL-6 (ϩsIL-6R), but not LIF, led to the phosphorylation of Akt at both Thr-308 and Ser-473 (Fig. 1A) indicating Akt activation. LIF treatment did induce STAT3 phosphorylation in these chondrocytes (data not shown). IL-1 stimulation did not result in Akt activation, and inclusion of LY294002 (5 M) prevented OSMmediated Akt phosphorylation (Fig.  1B). LY294002 similarly blocked IL-6-mediated activation of Akt (data not shown).

PI3K Inhibition Blocks IL-1ϩOSMmediated Cartilage Degradation-
We examined the effect of inhibiting the PI3K/Akt pathway in an established model of cartilage degradation. The IL-1ϩOSM combination (0.02 ng/ml and 10 ng/ml, respectively) was used at pathologically relevant concentrations, and induced a reproducible and synergistic release of collagen from bovine cartilage as shown previously (47), while inclusion of LY294002 throughout the 14-day culture resulted in a concentrationdependent inhibition of this release (Fig. 2). There was no increase in LDH release from inhibitor-treated cartilage (not shown) indicating that the observed effects were not due to toxicity of the inhibitor. A concentration-dependent reduction in the levels of active collagenolytic activity was also seen in the presence of LY294002 (Fig. 2). Although collagenase activity was significantly reduced, no inhibition of collagenolysis was observed at low inhibitor levels, consistent with previous findings when collagenase levels exceed 1 unit; such remaining collagenase activity is sufficient to catalyze maximum collagen release (52). The most marked reduction in activity correlated with the inhibition of collagenolysis (10 M LY294002) (Fig. 2). LY294002 did not affect IL-1ϩOSM-mediated proteoglycan release (Fig. 2, inset).
PI3K Inhibition Blocks IL-1ϩOSM-induced Collagenase Expression from Human Chondrocytes-Because the inclusion of LY294002 resulted in reduced collagenase activity and collagen degradation in cartilage, we examined the effect of PI3K/ Akt pathway inhibition on collagenase expression in primary human articular chondrocytes. We have previously demonstrated a synergistic induction of MMP-1 and MMP-13 in these cells when treated with IL-1ϩOSM (13). Addition of LY294002 resulted in concentration-dependent inhibition of MMP-1 and MMP-13 secretion into the culture medium (Fig. 3A) that mirrored the collagenolytic effects in cartilage (see Fig. 2). Furthermore, real-time RT-PCR showed that LY294002 (5 M) significantly inhibited IL-1ϩOSM-stimulated induction of both MMP-1 (Fig. 3B) and MMP-13 mRNA (Fig. 3C) in human chondrocytes. Inclusion of LY294002 had no effect on TIMP-1 levels (data not shown). Furthermore, no toxicity (as measured by XTT and LDH assays) was observed, nor any affect on IL-1- induced IL-6 production (53) was seen in the presence of LY294002 (data not shown), indicating no loss of normal cellular responses.
Compound 15e (2 M), like LY294002, significantly inhibited IL-1ϩOSM-stimulated induction of both MMP-1 and MMP-13 mRNA expression (Fig. 4B). Treatment of cells with TGX-221 (1 M) or AS-604850 (2 M) did not affect the induction of either MMP gene. At higher concentrations, AS-604850 (10 M) did inhibit MMP-13 induction partially (data not shown). However, it is important to note that this compound is only moderately selective for p110␥ and is reported to also inhibit p110␣ at this higher concentration (24).
We complemented the above studies with siRNA-mediated, isoform-specific p110 gene silencing. Because we failed to detect p110␥ transcripts in human chondrocytes, and AS-604850 did not affect Akt phosphorylation or collagenase induction at p110␥-selective concentrations, we chose the ␣, ␤, and ␦ isoforms for silencing, which was both effective (Ն75% mRNA reduction) and selective (data not shown). Combined silencing of the Class IA p110 isoforms reduced both p110 protein levels and OSM-stimulated Akt phosphorylation (Fig. 4C), and concomitantly caused a significant loss of the synergistic induction of both MMP-1 and MMP-13 (Fig. 4D). Isoformspecific silencing of p110␣ inhibited the induction of both MMP-1 and MMP-13, while p110␤ silencing failed to significantly affect the expression of either MMP gene. MMP-13 induction was significantly diminished when the p110␦ gene was silenced, although MMP-1 expression was unaffected (Fig. 4D).
We then performed Akt isoformspecific gene silencing experiments to determine which Akt isoform(s) may play a role in IL-1ϩOSM-mediated collagenase expression. Chondrocyte Akt1 protein was readily detectable by immunoblotting of cell lysates, but Akt2 and Akt3 were much less readily detected (data not shown). However, because transcripts encoding all three Akt isoforms were detected by real-time PCR, we performed gene silencing for each Akt isoform, which resulted in effective and selective silencing of each Akt mRNA (data not shown). Fig. 5C shows that combined gene silencing of all three Akt isoforms effectively inhibited OSM-stimulated Akt phosphorylation, as well as Akt protein levels. Assessment of IL-1ϩOSM-induced collagenase expression following combined Akt silencing revealed significantly reduced induction of both MMP-1 and MMP-13 (Fig.  5D). Selective silencing of individual Akt isoforms showed that silencing of Akt1, but not Akt2 or Akt3, resulted in a significant loss of MMP-1 expression, while MMP-13 mRNA induction was inhibited following silencing of Akt1 as well as Akt3 (Fig. 5D).
PDK1 Is Required for IL-1ϩOSMstimulated MMP-13 Induction in Human Chondrocytes-To gain further insight into the mechanism of collagenase induction by IL-1ϩOSM, we used the benzimidazole compound, Akt inhibitor IV, which indirectly inhibits Akt activity by targeting a kinase downstream of PI3K but upstream of Akt (56), and is therefore likely to be PDK1. This compound significantly inhibited Akt phosphorylation at both Thr-308 and Ser-473 in a concentration-dependent manner (Fig. 6A). Moreover, Akt inhibitor IV (3 M) significantly inhibited IL-1ϩOSM-stimulated induction of MMP-1 and MMP-13 (Fig. 6B).
We next investigated the role of PDK1, an essential link between PI3K and Akt activities (1) in the collagenase induction. Specific siRNA treatment of human chondrocytes caused a loss of PDK1 protein and hence loss of PDK1 phosphorylation at Ser-241; a necessary modification for kinase activity toward Akt (57), which resulted in a marked loss of dual phosphorylation of Akt by OSM although levels of Akt protein were unaffected (Fig. 6C). Furthermore, this PDK1 . The data are presented as a percentage of the expression for IL-1ϩOSM (DMSO included) (mean Ϯ S.E., n ϭ 4; ***, p Ͻ 0.001, inhibitor-treated compared with IL-1ϩOSM; *, p Ͻ 0.05, ANOVA). C, chondrocytes were stimulated with OSM (10 ng/ml) for 20 min after transfection with combined siRNAs specific for PI3K␣, -␤, and -␦ (33 nM each) or a non-targeting siRNA (siCon). Cell lysates were Western-blotted 48 h after the start of transfection. D, chondrocytes were stimulated with IL-1 (0.02 ng/ml) and OSM (10 ng/ml) for 24 h following transfection with siRNA specific to p110␣, p110␤, p110␦, or siCon (100 nM), or pooled siRNAs to silence all three p110 isoforms (33 nM each). Real-time RT-PCR of the isolated RNA was performed for MMP-1 (black bars) and MMP-13 (gray hatched bars) 72 h after the start of transfection. Data are presented as a percentage of the IL-1ϩOSM-induced expression (mean Ϯ S.E., n ϭ 4; ***, p Ͻ 0.001, specific siRNA-transfected compared with siCon-transfected; **, p Ͻ 0.01). Data are representative of (A and C), or pooled from (B and D), at least three separate chondrocyte populations. depletion resulted in a significant inhibition of IL-1ϩOSMstimulated MMP-13 mRNA expression, but had no effect on MMP-1 induction (Fig. 6D).

DISCUSSION
Although there is mounting evidence for the involvement of PI3K␥ in inflammatory diseases, this appears to be confined to a role in immune modulation (22,24) and there is still little known about the possible function(s) of PI3K/Akt signaling in chondrocytes and cartilage ECM catabolism. The combination of IL-1ϩOSM represents the most potent inducer of human cartilage collagenolysis to be reported (47), and the novel data presented in the present study demonstrate a requirement for PI3K/Akt signaling in such catabolism via an essential role in IL-1ϩOSM-stimulated collagenase gene expression in human articular chondrocytes. The results of both pharmacological inhibition and gene silencing of class I PI3K isoforms and Akt isoforms consistently indicated a role for p110␣ and Akt1 in the induction of both MMP-1 and MMP-13 by IL-1ϩOSM. Furthermore, siRNA transfections indicated involvement of p110␦, Akt3, and a requirement for PDK1 activity in IL-1ϩOSM-stimulated MMP-13 induction, confirming differential regulatory mechanisms exist in chondrocytes for these collagenase genes as previously reported for IL-1 stimulation (43).
Strong evidence now implicates both IL-6 and OSM in pathological cartilage ECM destruction (10,13,14,47,58,59), especially in combination with other major pro-inflammatory mediators (10 -14, 18). Indeed, IL-6-family cytokines are known to be present in the rheumatoid joint (47,60), as are elevated levels of both MMP-1 and MMP-13 (61), and our data strongly indicate a role for OSM-mediated PI3K/Akt signaling in the synergistic induction of these collagenase genes. This is also highly likely for IL-6-activated PI3K signaling in chondrocytes. PI3K␣ and PI3K␤ are ubiquitously expressed in mammalian tissues, whereas PI3K␦ and PI3K␥ expression is mainly associated with leukocytes (22). Interestingly, we found no evidence for the involvement of PI3K␤ or PI3K␥ in collagenase induction in chondrocytes although the use of selective inhibitors of PI3K␥ ameliorates collagen II-induced arthritis in mice (24). PI3K␦ is considered to be particularly important for B cell function (22); to our knowledge this is the first report of PI3K␦ expression in primary human chondrocytes, although expression of all four class I PI3K isoforms has been detected in the pre-chondrogenic cell line ATDC5 (35). This suggests modulation of immune functions may not be the sole mechanism of action of PI3K inhibitors against inflammatory joint diseases, and it will be valuable to ascertain the effect of PI3K␦-specific inhibitors in animal models of joint destruction.
A recent report has suggested a role for Akt in OSM-induced MMP-13 expression in chondrocytes (42), although the compound used also inhibits protein kinase A (62). Our isoformspecific Akt gene silencing revealed a role for Akt1 in the induction of both MMP-1 and MMP-13, while Akt3 was only associated with MMP-13 expression. Akt1 was the most readily detected isoform in human chondrocytes (not shown), and has been shown to promote invasion of cancer cells, partly through increased MMP production (63). While Akt3 is largely regarded to have a neuronal-specific expression (64), all three Akt isoforms have been detected in ATDC5 cells (35), and we have found them to be readily detectable in SW1353 chondrosarcoma cells (data not shown). This is to our knowledge the first report of Akt3 in primary chondrocytes and the first report FIGURE 5. Differential IL-1؉OSM-induced collagenase expression in human chondrocytes is revealed by specific Akt isoform blockade. A, primary human articular chondrocytes were stimulated with OSM (10 ng/ml) for 20 min after a 20-min pretreatment with Akt inhibitor VIII at the concentrations indicated. Cell lysates were Western-blotted using the antibodies indicated. B, chondrocytes were stimulated with IL-1 (0.02 ng/ml) and OSM (10 ng/ml) for 24 h Ϯ Akt inhibitor VIII (3 M) and real-time PCR performed on the harvested RNA for MMP-1 (black bars) and MMP-13 (gray hatched bars). Data are presented as a percentage of the IL-1ϩOSMinduced expression (DMSO included) (mean Ϯ S.E., n ϭ 4; ***, p Ͻ 0.001, inhibitor-treated compared with IL-1ϩOSM; ANOVA). C, chondrocytes were stimulated with OSM (10 ng/ml) for 20 min after transfection with combined siRNAs specific for Akt1, -2, and -3 (33 nM each) or a non-targeting siRNA (siCon). Cell lysates were Western-blotted 48 h after the start of transfection. D, chondrocytes were stimulated with IL-1 (0.02 ng/ml) and OSM (10 ng/ml) for 24 h following transfection with siRNA specific to Akt1, Akt2, Akt3, or siCon (100 nM), or pooled siRNAs to silence all three Akt isoforms (33 nM each). Real-time PCR of the isolated RNA was performed for MMP-1 (black bars) and MMP-13 (gray hatched bars) 72 h after the start of transfection. The data are presented as a percentage of the expression for IL-1ϩOSM (mean Ϯ S.E., n ϭ 4; ***, p Ͻ 0.001, specific siRNAtransfected compared with siCon-transfected; *, p Ͻ 0.05). Data are representative of (A and C) or pooled from (B and D), in at least three separate chondrocyte populations. of a role for this kinase in differential MMP gene expression by IL-1ϩOSM. This divergence of mechanisms may provide future therapeutic opportunities to specifically intervene in the expression of specific collagenase genes in an inflammatory setting.
Because PDK1 is an important link between PI3K and Akt (1), it was intriguing to note that depletion of PDK1 significantly inhibited IL-1ϩOSM-stimulated MMP-13 induction but not MMP-1. This could be explained by a more dominant role for the PI3K/Akt pathway in IL-1ϩOSM-induced MMP-13 expression compared with MMP-1, although the higher levels of MMP-1 expression (18) may indicate that sufficient PDK1 activity remains to contribute normally to MMP-1 induction after siRNA treatment, while MMP-13 expression is much more sensitive to such depletion. Indeed, this is consistent with other data reported here, wherein a greater impact on MMP-13 compared with MMP-1 expression is observed as a result of blockade of other specific elements of the PI3K/Akt pathway.
We have previously shown that in addition to IL-1, OSM (and IL-6) also synergizes with TNF␣ and IL-17 to similarly promote the synergistic release of cartilage collagen (11,14), and it is most likely that the PI3K/Akt pathway also contributes to the catabolism mediated by these cytokine combinations. The points of interaction between these cytokine signaling pathways are unknown. However, it was interesting to note that OSM-stimulated Akt phosphorylation was consistently reduced in the presence of IL-1 in this study; this may indicate that Akt is a point of interaction with IL-1 signaling. Similarly, in chondrocytes, human cartilage glycoprotein 39-induced Akt suppresses p38 and c-Jun N-terminal MAPK activation induced by either IL-1 or TNF␣; a consequence of this is a reduction in MMP levels (65). Insulin-like growth factor-1 also binds a specific receptor and activates the PI3K/Akt pathway in chondrocytes (66), and we have previously shown that this growth factor protects cytokinestimulated cartilage by reducing MMP levels (12). Our data would therefore suggest that Akt activation in chondrocytes may mediate different downstream phosphorylation events, leading to either catabolic or anabolic effects, depending on the nature of the stimulus. Because several reports indicate a role for Akt in normal chondrocyte functions (33-35, 39, 66), it may well be that when cytokines and growth factors interact with OSM or IL-6 (with subsequent Akt activation), abnormal chondrocyte proliferation and MMP production (both characteristics of OA) may occur even as a consequence of potentially reparative mechanisms (19).
In conclusion, this study highlights a novel role for the PI3K/ Akt pathway in the pathological cartilage ECM catabolism mediated by members of the IL-6-type cytokine family when in combination with other pro-inflammatory cytokines (10 -14, 18, 47). Therapeutic intervention targeted at this pathway may provide a means of differentially regulating collagenase gene expression, which could be highly beneficial in the treatment of inflammatory joint diseases as well as other pathologies associated with PI3K/Akt-dependent collagenase expression.