Obesity and Bone

Introduction

Both obesity and osteoporosis are common conditions leading to increased morbidity and mortality. Cardiovascular disease, cancer, type 2 diabetes, obstructive sleep apnoea, osteoarthritis, psychosocial problems, hypertension, fatty liver, and infertility are also all caused by or aggravated by obesity. Of these, it has been shown that cardiovascular disease and cancer account for the highest mortality¹. Osteoporosis increases the risk of fractures and results in increased mortality² and morbidity, including increased pain and reduced mobility, giving a reduction in quality of life³. The very first link between obesity and bone was the common belief that an individual with obesity was heavier because he was “big boned”⁴. Although this is a common myth, science has revealed a surprising bidirectional relationship between excess fat and bone. Furthermore, Ilich and colleagues have proposed the existence of osteosarcopenic obesity linking obesity and osteoporosis to muscle weakness⁵. In their formulation, they discuss possible mechanisms, such as mesenchymal cell development, linking these three conditions. Osteosarcopenic obesity occurs in old age, but it has been shown that even preschool obesity is associated with an increase in fracture risk in teenage children⁶. This review will briefly discuss how excess fat damages bone and how bone contributes to the regulation of body weight.

How does obesity influence bone?

Weight-bearing exercise increases bone density⁷, so pressure from excessive weight may strengthen bone⁸. Consistent with this, significant weight loss achieved with bariatric surgery results in bone loss⁹, which can be mitigated by exercise¹⁰. However, despite more bone mass, obesity confers an increased risk of osteoporosis¹¹ and bone fractures¹², at least in some sites. Compston and colleagues¹³ followed 52,939 post-menopausal women for 3 years. During that time, 3,628 reported a fracture. BMI was protective for hip with a hazard ratio (HR) of 0.8 (95% confidence interval [CI] 0.71–0.90), spine with a HR of 0.83 (95% CI 0.76–0.92), and for wrist with a HR of 0.88 (95% CI 0.83–0.94) (HR with 95% CI for each increase in BMI of 5 kg/m²). However, excess weight increased the risk of ankle fractures (HR 1.05 [1.02–1.07]) (P <0.01). For pelvic and rib fractures, they found a non-linear relationship with inverse associations at low BMI/body weight (P = 0.05) and positive associations at high values (P = 0.03). Interestingly, Liu and colleagues in a cross-sectional study involving 471 women divided into normal, overweight, and obese groups found that there is a threshold effect of fat on bone density. They concluded that while body fat below 33% has a positive relationship with bone density, for body fat above 33% the relationship between fat mass and bone is negative for most skeletal sites¹⁴. So extra weight builds bone and extra fat padding protects the bone during a fall. So why is there an increased rate of fracture in some sites? It appears that obesity causes bone to be of poor quality.

How can obesity damage bone?

Several mechanisms have been proposed for the deleterious effects of obesity on bone:

Replacement of osteoblasts by fat cells in bone marrow

Clinically, it has been found that situations leading to bone loss such as old age, thiazolidinedione treatment, and prolonged corticosterone treatment are associated with increased fat in bone marrow¹⁵. Adipocytes and osteoblasts are both derived from the same bone marrow-derived mesenchymal cell precursor (Figure 1). Increasing adipocyte differentiation decreases osteoblast differentiation and vice versa^16,17. The signals that cause the stem cells to differentiate into one or the other include the transcriptional activator PPARγ, which can move the differentiation towards the adipocyte lineage¹⁸. Other triggers include leptin¹⁹ and the Wingless-related Integration site (Wnt) signalling glycoproteins. Wnt signalling is an ancient controller of cell fate determination, cell migration, and more during embryogenesis²⁰. It has been found that Wnt activation stimulates osteoblast formation and inhibits adipocyte differentiation. Secreted frizzled-related protein 1 (sFRP-1), an inhibitor of Wnt signalling, is increased in adipocyte stem cells²¹. sFRP-1 has been shown to be increased in mild obesity, leading to increased adipocyte formation, but interestingly is reduced in severe obesity, possibly to limit further adipocyte formation²².

Figure 1. Ontogeny of fat, osteoblast, and osteoclast cells in bone marrow.

How does bone influence body weight?

It appears that bone can assess body weight. Jansson and colleagues³² placed fluid-filled capsules into the abdomens of Sprague-Dawley rats and C57BL6 mice with diet-induced obesity to equal approximately 15% of their body weight. Control rodents had empty capsules inserted into their abdominal cavities (3% of their body weight). Increased loading suppressed body weight in both rats and mice with the difference in weight compared to control rodents already present from day 2 post-surgery. After 14 days of loading, 80% of the increased load was counteracted by reduced body weight. At the end of the experiment, body weight plus capsule weight was similar in the load and control rodents.

MRI imaging on day 14 showed that increased loading reduced the amount of white adipose tissue with no difference in brown adipose tissue. Not surprisingly, leptin levels were lower, consistent with the reduced adipose tissue mass. Weight loss was due to reduced food intake in both rats and mice. There were no changes in VO₂ max, respiratory quotient, or muscle activity, suggesting that all of the effects of capsule loading were due to a reduction in energy intake. This was further confirmed by the fact that pair fed rodents decreased their body weight to the same extent as load rodents.

The effects of loading to suppress weight were shown to be independent of leptin when they showed that leptin-deficient ob/ob mice respond similarly to wild-type mice to capsule load. In any case, the reduction of leptin should have led to an increase in energy intake, not in the observed reduction in intake, so what is the signal?

It was known that osteocytes can sense dynamic short-term high-impact bone loading, so the authors hypothesised that osteocytes may also sense chronic static bone loading due to increased body weight. To test this hypothesis, mice deficient in osteocytes were loaded with the capsules. It was found that the control mice lost weight as shown before but the equally loaded osteocyte-deficient mice did not. So, load sensing is mitigated by the osteocyte.

The following question then arose: how does the osteocyte send the signal of increased load to the hypothalamus? In previous studies, dynamic short-term high-impact loading was shown to reduce the expression of Sost, the gene coding for sclerostin. However, there was no difference in sclerostin levels. There was also no change in carboxylated and undercarboxylated forms of osteocalcin, fibroblast growth factor 23, and lipocalin 2. In addition, deficiency of ghrelin, glucagon-like peptide 1, the melanocortin 4 receptor, or the oestrogen receptor did not remove the effect. This leaves the possibility of a neuronal signal from osteocytes and the hypothalamus. To begin to investigate this, the authors measured urinary noradrenaline/creatinine ratio and serum choline level but found no differences. It was concluded that the osteocyte can sense chronic weight load and can send an unknown signal that can fairly accurately correct body weight.

Conclusion

There is a bidirectional complex relationship between obesity and bone. The advances in understanding how obesity leads to increased fragility in bone may lead to new treatments to improve bone strength. Finally, finding the signal that osteocytes send to the hypothalamus may lead to new approaches in the treatment of obesity.