β 2‐Microglobulin and bone cell metabolism

Dialysis‐related amyloidosis (DRA), also referred to as β2‐microglobulin (β2M) amyloidosis, is a major cause of skeletal morbidity in patients with end‐stage renal disease. The DRA syndrome results in a progressive destructive periarticular osteoarthropathy. The pathological lesions of DRA consist of cystic lesions and localized areas of β2M‐amyloid deposition. Approximately 70% of adult patients who undergo dialysis for more than 10 years develop radiographic evidence and/or symptomatic pathology associated with β2M‐amyloid deposition [1]. With advances in the treatment of the cardiac and cerebrovascular complications associated with end‐stage renal disease, it is anticipated that the life expectancy of dialysis patients will continue to increase. Thus, morbidity from bone disease in general and DRA in particular will become more prevalent. Over the past decade numerous hypotheses have been put forward in an attempt to explain how β2M might affect bone cell metabolism and play a role in the development of DRA. However, much of the reported experimental data has been difficult to interpret, resulting in much controversy regarding the role of β2M in normal and abnormal bone physiology. Although there is substantial evidence suggesting that there is a specific effect of β2M on bone‐cell metabolism, a recurrent argument has been that the effect was the result of other growth factors or undefined contamination of the β2M preparations tested. There are two major facts forming the basis for this argument: a lack of an identified receptor for β2M, and continued disagreement among investigators concerning the mitogenic effect of β2M on bone cells (Table 1). Thus, the question that needs to be addressed is whether β2M plays an active role in bone metabolism or whether β2M is a passive participant, being incidentally deposited in the form of amyloid at sites of bony destruction. Selected effects of β2‐microglobulin on bone and bone‐cell metabolism GF, gel filtration; IEC, ion exchange chromatography; HPLC, reversed phase HPLC, Sigma purified by GF and IEC; ALP, alkaline phosphatase; ↑ increase; ↓ decrease; → no effect; ? unknown. Selected effects of β2‐microglobulin on bone and bone‐cell metabolism GF, gel filtration; IEC, ion exchange chromatography; HPLC, reversed phase HPLC, Sigma purified by GF and IEC; ALP, alkaline phosphatase; ↑ increase; ↓ decrease; → no effect; ? unknown. Although the effect of β2M on osteoblast proliferation is controversial, multiple other effects of β2M on bone‐cell metabolism have been observed. Osteoblasts produce β2M [2]. Subcutaneous injection of β2M induces histological evidence of bone resorption in neonatal mice [3], and purified human β2M induces a dose‐ and time‐dependent net calcium efflux in cultured murine calvariae [4,5]. This calcium efflux is mediated in part by interleukin‐1β (IL‐1β) [6]. β2M also stimulates the synthesis of IL‐6, a potent bone‐resorbing cytokine, leading to an increase in mRNA and protein levels in osteoblasts [7]. β2M has also been shown to stimulate synovial fibroblasts to produce stromelysin, a neutral matrix metalloproteinase (MMP), which is believed to be a key enzyme causing articular destruction in inflammatory joint diseases [8]. The findings that β2M induces the synthesis of collagenase‐1 from rabbit synovial fibroblasts and the preferential collagen binding capacity of β2M also supports the hypothesis that β2M has a principal role in modulating connective‐tissue breakdown [9,10]. Migita and colleagues [11] demonstrated that β2M increases cyclo‐oxygenase‐2 (COX‐2) protein and mRNA expression in a dose‐dependent manner from human synovial cells; however, utilizing the mouse calvarial resorption model, we were unable to demonstrate an effect of β2M on prostaglandin E2 production [6]. Although β2M has a significant bone‐resorbing effect, advanced glycation end‐product (AGE) modification of β2M appears to further increase bone resorption and cytokine production [12,13]. Compared to unmodified β2M, the number of resorption pits formed by isolated osteoclasts are significantly increased by AGE‐modified β2M [13]. AGE modification of β2M seems to alter bone metabolism in a number of ways, not only increasing bone resorption, but also decreasing fibroblastic collagen deposition. AGE modification of β2M compared with unmodified β2M decreases fibroblastic synthesis of type I collagen [14]. Interestingly, the amyloid deposits and surrounding macrophages from patients with DRA react with a monoclonal anti‐AGE antibody [15]. AGE‐modified β2M stimulates chemotaxis of monocytes and macrophages and enhances the secretion of cytokines [16]. The biological effect of AGE‐modified β2M on monocytes and macrophages is thought to be mediated by the receptor for AGE (RAGE). The finding that AGE‐modified β2M further induces bone resorption and osteoblastic cytokine release as well as reducing type I collagen synthesis by fibroblasts compared to unmodified β2M could be the result of cellular RAGE recognition, as RAGE has been described on osteoblasts [17]. It was in the late 1980s when β2M was proposed as a potential bone growth factor. However, the mitogenic effect of β2M continues to be one of the most controversial and hotly debated issues concerning the effect of β2M on bone. As shown in Table 1, studies evaluating the effect of β2M have been performed utilizing different experimental models with cells and tissues obtained from different animal species, and using varying doses of β2M. In most of the studies utilizing bone cells, the developmental stage of the bone cells was not defined. Bone cells undergo a series of developmental stages, such as proliferation, differentiation, and apoptosis, and each of them involves induction and suppression of various genes. The response of osteoblasts to various factors has been shown to be dependent on their developmental stage [18]. This may shed some insight as to why different experimental systems yield various results depending on the maturational stage of the osteoblasts. The variability in bone‐cell response has led some investigators to suggest that β2M might not be a typical growth factor, but a regulator of the growth‐promoting effects of other growth factors [19]. To further complicate the interpretation of the various results, the β2M utilized in these studies was isolated by different techniques and from different sources. The possible impurity of some preparations of β2M has resulted in at least one group proposing that the mitogenic effect of β2M is the result of growth factor contamination. Jennings et al. [20] further purified β2M and observed that the mitogenic activity of the original protein was diminished after purification by reverse‐phase high‐performance liquid chromatography (RP‐HPLC). Maximal mitogenic activity was detected in fractions different from their ultra‐purified β2M; thus they concluded that the original mitogenic activity was the result of growth factor contamination [20]. Although RP‐HPLC ensures analytical purity, it may cause denaturation and loss of protein function. This denaturation might explain loss of mitogenic activity in the β2M fraction. Furthermore, the β‐sheet structure of β2M favours amyloid fibril formation and spontaneous precipitation [21], thus high concentrations of β2M in a salty environment, as occurs during RP‐HPLC, may result in precipitation and dimer formation, with a consequent decrease in the amount of bioavailable β2M. This could serve as an alternative interpretation as to why RP‐HPLC leads to a loss of mitogenic activity of β2M. Since growth‐factor contamination has been suggested as being responsible for some of the mitogenic effects of β2M, the purification method has become a crucial issue. The majority of studies, as listed in Table 1, have been performed using β2M purified by gel filtration and ion exchange chromatography. Thus the possibility of growth factor contamination raised serious concerns as to whether the accumulated data concerning β2M is reliable. Although the majority of the experimental data supports a role of β2M in both normal and abnormal bone metabolism, it is not surprising that defining this role has been elusive. β2M is the small extracellular subunit of the MHC Class I molecule, and is present on the surface of all nucleated cells. As molecules of the MHC complex commonly do, β2M has been suggested to possibly interact with hormonal and/or growth factor receptors [22]. This theory is further supported by the fact that despite the intensive research and focus on β2M, a receptor for it has not been identified. Interaction of β2M with various receptors may potentially induce various signal transduction pathways and genes depending on the experimental system utilized and the differentiation stage of osteoblasts and could explain some of the controversial findings (Table 1). Initial work on β2M offered a promising role in cell regulation, until questions were raised about possible contamination of β2M preparations. This cloud of suspicion shed significant doubt primarily on the mitogenic effect of β2M. Since the amyloid deposits contain up to 95% β2M, it is tempting to blame β2M for the altered bone metabolism observed in DRA. The presence of β2M in the immediate environment throughout the development of the osteoblast might alter maturation and differentiation. The cumulative data demonstrate that β2M has biological activity far beyond that which could be explained by growth factor contamination. Nonetheless, the proposed roles of β2M are so numerous that is difficult to imagine one molecule having such a variety of specific effects, unless β2M acts through a variety of receptors in diverse ways depending on the developmental phase of cells and the availability of receptors. The current theory about the mechanism of β2M causing DRA invokes the release of β2M and the activation of monocytes by AGE‐modified β2M. After being shed from cell surfaces, the free AGE–β2M could trigger cell migration and release of bone‐resorbing cytokines at osteoarticular sites, and the free heavy chains remaining at the cell surfaces may also activate monocytes. Clearly, the β2M‐mediated alteration of bone cell metabolism is very complex and could not be explained by a single metabolic pathway. Despite the accumulating knowledge about β2M toxicity, strategies for prevention and therapy for DRA have been unsuccessful, and many issues concerning β2M amyloidosis remain unresolved. Notwithstanding the controversies, determination of the true significance and role of β2M in bone metabolism is imperative and is critical in regard to preventing long‐term skeletal morbidity in dialysis patients. Correspondence and offprint requests to: Dr Stuart M. Sprague, Associate Professor of Medicine, Department of Medicine, Northwestern University Medical School, Evanston Northwestern Healthcare, 2650 Ridge Avenue, Evanston IL 60201, USA.