In this study, a panel of biological properties of rose hip powder and its constituent galactolipids, GLGPG, has been described for the first time and provides evidence that cellular parameters related to cartilage destruction and inflammatory responses were modulated by these natural compounds. This feature has been established using two approaches: (1) effects on inflammatory processes including cytokines and chemokines were monitored in macrophages and peripheral blood leukocytes, (2) modulation of catabolic activity and the production of chemokines and cytokines were determined in SW1353 cells and NHAC-kn in vitro.
An adequate homeostasis between anabolic and catabolic events ensures tissue rebuilding and renewal in intact cartilage . Growth factors including insulin-growth factors, connective-tissue growth factors, TGF-β or BMP favor proliferation and differentiation of chondrocytes and the synthesis of collagen and aggrecan. Conversely, inflammatory factors (e.g. pro-inflammatory interleukins, eicosanoids and nitric oxide) and proteinases (e.g. MMPs and ADAMTS) induce tissue erosion and degrade ECM components, respectively. MMP-1 and -13 preferably cleave type II collagen . MMP-3 has broader substrate specificity; ADAMTS-4 and -5 cleave proteoglycans . IL-1β is the most important inducer of catabolic processes in OA. TNF-α, IL-6 and LIF also contribute to tissue erosion in advanced stages of OA, although their implication in rheumatoid arthritis (RA) prevails . IL-1β activated primary chondrocytes express catabolic factors that degrade the ECM. OA chondrocytes were rather refractory to the patho-physiological trigger [7, 32]. Here, we show that RHP and GLGPG impaired IL-1 expression in macrophages, leukocytes and chondrocytes. LIF, which is also involved in OA [14, 28, 29], was reduced by RHP and GLGPG. The data suggest that the substances exert their effects at an early phase of OA development and target different cell populations. Admittedly, this hypothesis needs to be tested in appropriate preclinical models.
While IL-6 is of key importance in RA, it is also involved in OA [33–36]. RHP inhibited IL-6 gene expression in chondrocytes, whereas its impact on cytokine production in macrophages or leukocytes was marginal. Thus, the effect of compounds was confined to cells, where excessive IL-6 production was deleterious; conversely, IL-6 was not modulated in peripheral blood leukocytes where it is required for an efficient humoral immune response. Similarly, TNF-α has multiple actions in the pathogenesis of RA  and may be a contributing factor to OA . The tested natural substances reduced its expression in macrophages and leukocytes, while in chondrosarcoma cells the opposite effect on gene expression was observed. The meaning of this dichotomy is unclear and requires further investigation. Collectively, changes in IL-6 and TNF-α expression by RHP and GLGPG could influence the etiology of OA and RA. Indeed, in a recent clinical study, dietary supplementation of RA patients with RHP alleviated RA symptoms . Obvious limitations in the interpretation of the current in vitro results should be noted: (1) the absence of data on bioavailability and pharmacokinetics of the substances makes it difficult to correlate the described in vitro effects with in vivo efficacy, (2) additional clinical trial to assess efficacy (anti-inflammatory versus pain-relief) of the test substances are warranted, (3) a possible effect of RHP and GLGPG on joint space widening should be investigated.
The involvement of NO and PGE2 in OA has been described in numerous studies [11, 38–43]. The enhanced production of NO in OA joints contributed to a slowly progressing inflammation . IL-1β treatment of chondrocytes induced iNOS and concomitant expression of cartilage-degrading enzymes [41, 42, 44–46]. Conversely, the progression of murine OA was slowed down in iNOS knock-out mice . NO also activated MMPs  and PGE2 production  with concomitant inhibition of proteoglycan and collagen synthesis [43, 49]. Chondrocyte apoptosis was promoted by NO and PGE2 [8, 50–52]. In view of the observation that RHP and GLGPG diminished NO and PGE2 production, they might have anti-apoptotic effects in chondrocytes. The consequence of this inhibition is pivotal, since it affects (1) survival of chondrocytes, (2) production of pro-inflammatory cytokines, and (3) activation of ECM-degrading enzymes. At the molecular level, the expression of prostaglandin E2 synthase (i.e. mPGES) [53–57] rather than COX-2 was altered by RHP (Table 2) and therefore only weakened the production of pro-inflammatory prostaglandins.
MMP-1, MMP-3, MMP-13, ADAMTS-4 and ADAMTS-5 are key catabolic enzymes that degrade collagen and proteoglycan; their sequential expression might occur at, and herald different phases in the progression of OA [28, 32, 58]. Increased expression of tissue inhibitor of matrix metalloproteinase (TIMP) is associated with remodelling of articular tissues . IL-1β induced a high MMP-1 expression in primary chondrocytes, which reflects events related to early OA; SW1353 displayed a pattern of MMP expression that relates to intermediate stages of OA. RHP and GLGPG exerted effects only on a sub-set of these enzymes: MMP-1, -3 and -13 and ADAMTS-4. Other members of the MMP family that have a role in tissue remodeling (e.g. MMP-9)  were not influenced by these treatments.
The impact of chemokines in OA has been substantiated previously : CCL5/RANTES and CXCL8/IL-8 were identified in activated chondrocytes or OA tissue [14–16, 62]. Expression levels of chemokines and their receptors dramatically change in IL-1β activated chondrocytes . This emphasizes the putative role of chemokines in early and intermediate phases of progressing OA. Given the described effects of RHP and GLGPG, it is tempting to hypothesize that RHP components act as biological modifiers on chemotaxis in OA chondrocytes. In accordance with the in vitro study, clinical trials have provided evidence that chemotaxis of leukocytes is reduced after dietary supplementation with RHP .
Biological modulators such as IL-1β or NO eventually activate MAPK that, in turn, leads to the translocation of NF-κB to the nucleus and NF-κB dependent gene activation. The kinetics and extent of RelA and NF-κB1 expression follow similar kinetics and amplitude in IL-1β stimulated SW1353, primary chondrocytes  and RAW264.7 cells (this study). In vitro studies have further demonstrated that this pathway was modulated by various substances contained in the food chain [63–68]. Regulatory motifs identified in chemokine genes include NF-κB . Direct evidence that RHP and GLGPG may act along this pathway is provided by the observed down-regulation of NF-κB1, NF-κB49, NF-κBp65 and - as a consequence of decreased re-synthesis - I-κBα in activated macrophages (Table 2). The analysis of modification of these transcription factors like phosphorylation is required to substantiate this hypothesis. As previously shown [7, 52, 70, 71], binding elements for other transcription factors (MEF-3, AP-1 and CEBPβ) have been mapped to the regulatory region of IL-1β- or LPS-responsive genes  and might also interact with RHP and GLGPG.
The in vitro effects described in this study were elicited at high concentrations of RHP and GLGPG. To date, no bioavailability studies have been reported for RHP; but it is unlikely that IC50 values for RHP and its constituent bioactive components are achieved in the body fluids or tissues as a consequence of dietary uptake. It is possible that after dietary intake RHP constituents accumulate in peripheral blood leukocytes and thus locally achieve threshold concentrations required for biological effects. Assuming that GLGPG is the only bioactive component in RHP powder, it can be deduced from the IC50 values given in Table 1 that RHP needs to contain ~3% of GLGPG. Yet, since the GLGPG contents of the studied RHP preparation does not exceed 0.1% of the dry plant mass (our unpublished results), we hypothesize that other substances contribute to the biological activity of RHP. The observation that substances contained in the food chain alter features of chondrocyte biology has been documented previously: polyphenols, including resveratrol and catechins (epigallocatechin-3-gallate, EGCG) reduced the expression of MMP-1, -3 or -13 and modulated levels of iNOS and COX-2 [63, 65–68, 72–74].