The RANKL/OPG system and bone mineral density in patients with chronic liver disease

Background/Aims

Osteopenia and osteoporosis are common complications of chronic liver disease (CLD). Receptor activator of nuclear factor κB ligand (RANKL) and osteoprotegerin (OPG) regulate osteoclastogenesis and bone remodelling, and are involved in several inflammatory diseases. This study investigated the activation state of the RANKL/OPG system and its association with bone loss in CLD.

Methods

Serum levels of OPG and sRANKL were determined in 193 patients with CLD and 56 age- and gender-matched healthy controls. Cellular sources of OPG and RANKL were determined immunohistochemically. Dual-energy x-ray absorptiometry was performed to determine bone mineral density (BMD) of lumbar spine and femoral neck.

Results

sRANKL serum levels were significantly elevated in non-cirrhotic, but not cirrhotic patients compared to healthy controls. OPG serum levels were elevated 1.6-fold in non-cirrhotic and 2.8-fold in cirrhotic CLD patients. RANKL⁺ cells were mainly confined to portal fields, while OPG was broadly expressed. In the cirrhotic subgroup (87 patients) we observed a significantly higher OPG/sRANKL ratio in patients with osteopenia or osteoporosis of the lumbar spine and femoral neck region (T-score <−1) compared to those with normal BMD (T-score≧−1).

Conclusions

CLD is associated with alterations in RANKL/OPG serum levels, which could modulate bone loss in CLD.

Introduction

Chronic liver diseases (CLD) represent a group of disorders commonly associated with a dysregulation of inflammatory pathways and tissue damage in the liver [1]. A variety of etiological agents such as viral infections, chronic alcohol ingestion, drugs, autoimmune hepatitis, metabolic and hereditary causes may result in a chronic inflammatory process which is characterized by inflammatory infiltration, necrotic and apoptotic tissue damage followed or accompanied by regeneration with liver fibrosis and finally liver cirrhosis. In human CLD and various mouse models, several cytokines have been linked to both pathological damage and the regeneration of liver tissue after injury with tumor necrosis factor (TNF) α, interferon (IFN) γ and interleukin- (IL-) 18 playing important roles in CLD [2], [3], [4].

Bone disease – osteopenia as well as osteoporosis – is a common complication of CLD and can result in spontaneous or low-trauma fracture with significant impact on morbidity, quality of life and even survival. The pathogenesis of hepatic osteodystrophy is thought to comprise multiple factors including the genetic background, nutritional deficiency, low vitamin D, Ca²⁺ deficiency, low insulin-like growth factor 1, and excessive alcohol intake (reviewed by Leslie et al. [5]).

It is currently unclear whether the underlying inflammatory process per se might also contribute to CLD-associated bone loss. Gonzalez-Calvin et al. demonstrated that high levels of soluble TNF receptor p55 (sTNFRp55) are inverselycorrelated with BMD and positively with deoxypyridinoline – a marker of bone resorption – in patients with viral cirrhosis [6]. Soluble TNF receptors have been shown to correlate well with TNFα and therefore possibly reflect the activation state of TNFα/TNFR system as well as the clinical activity in alcoholic and viral cirrhosis [7], [8], [9]. Another report that connects bone loss with the inflammatory cytokine IL-6 comes from Dai et al. who demonstrated that bone loss due to chronic alcohol ingestion in mice is dependent on IL-6 and abrogated in IL-6^−/− mice. Notably, these authors suggest that IL-6 achieves its bone resorbing effect by inducing receptor activator of nuclear factor κB (RANKL) [10].

RANKL – a member of the TNF family – is a ligand for RANK and together with OPG – a natural decoy receptor for RANKL – they have been shown to be key regulators of bone metabolism. Via binding to its receptor RANK, RANKL is essential for both osteoclastogenesis and activation of bone resorbing osteoclasts [11]. This process is counteracted by OPG – the natural decoy receptor for RANKL – and actually a fine tuned balance between RANKL and OPG is required for bone homeostasis [12], [13]. Activated T cells produce membrane-anchored RANKL that is shedded by metalloprotease-disintegrin TNFα convertase (TACE) to form soluble RANKL (sRANKL) [14]. Activated T cells have been shown to be critically involved in bone loss through osteoclastogenesis in rheumatoid arthritis and IL-2 deficient colitic mice [15], [16]. Besides its outstanding role in the regulation of osteoclastogenesis, the RANKL/OPG system plays another important role in immune regulation. RANKL enhances activation and survival of dendritic cells (DC) and monocytes/macrophages and is a potential chemotactic factor for mononuclear cells [17], [18], [19]. OPG exerts anti-apoptotic properties by binding TNF related activation induced ligand (TRAIL) which in turn is able to block the anti-osteoclastogenic effect of OPG [20].

As inflammatory cytokines – involved in several etiologies of human CLD – can trigger bone loss and as OPG and RANKL mRNA have been detected in the liver, we hypothesized that the RANKL/OPG system might be involved in the regulation of bone loss in CLD [21], [22].