The Bone-Fat-Pancreas Axis in Children

The overarching hypothesis of this proposal is that obesity and positive energy balance in
children promote both low bone mass accrual and risk for diabetes through events that are
mechanistically associated and that involve bone as an endocrine organ. Recent studies
conducted in mice have uncovered the presence of a unique "bone-fat-pancreas" axis that
regulates energy homeostasis, coordinates energy partitioning between bone and adipose
tissue, and impacts insulin sensitivity. The adipocyte-derived hormone leptin, elevated
levels of which reflect both adiposity and positive energy balance, inhibits bone formation
via sympathetic activation. Decreased bone formation in turn depresses insulin sensitivity
and secretion via decreased production of undercarboxylated osteocalcin (unOC), a novel
bone-derived hormone. Although data from mice are compelling, this novel pathway has not
been widely tested in humans. Sparse data from adult men and women suggest that this axis
is active in humans, and that unOC is regulated in part by exercise. No data are available
regarding the bone-fat-pancreas axis in children. Because the foundations of body
composition trajectories and metabolic "programming" are established early in the life
course, childhood, particularly during early stages of growth and development, is an
especially salient time period for evaluating the bone-fat-pancreas axis. With this pilot
grant, we propose to gather evidence that these interrelationships exist in children. The
data from this project will be used to prepare an NIH R01 proposal to conduct a
lifestyle-based intervention in children aimed both at reducing risk for osteoporosis and
type 2 diabetes, and at identifying the role of unOC in metabolism and tissue partitioning.

Hypothesis 1: Obesity and positive energy balance in children decrease bone mass via
elevated leptin.

Specific Aim 1: Determine the association between bone mass by DXA and serum leptin
concentration in lean and obese children. We predict that body weight will be positively
associated with bone mass, but that at any given body weight, bone mass will be lower in
obese children, and that this difference will be explained by leptin.

Hypothesis 2: Leptin-mediated suppression of unOC decreases insulin secretion through
action on the β-cell, and decreases insulin sensitivity by inhibiting secretion of
adiponectin from adipose tissue.

Specific Aim 2: Determine the association between insulin secretion during oral glucose
tolerance test (OGTT; from C-peptide modeling) and serum unOC in lean and obese children.
Determine the association between insulin sensitivity during OGTT (derived from mathematical
modeling) and serum unOC in lean and obese children. Obese children are less insulin
sensitive, and in an absolute sense, secrete more insulin. However, we predict that at any
given degree of insulin sensitivity, insulin secretion will be lower in obese children, and
that this difference will be explained by unOC. unOC will be inversely associated with
serum leptin, and will be positively associated with adiponectin and insulin sensitivity.

Hypothesis 3: Physical activity prevents leptin suppression of unOC and partitions energy
towards bone mineral at the expense of bone marrow adipose tissue.

Specific Aim 3: Assess the interrelationships among physical activity using accelerometry,
bone mass using DXA and bone marrow adipose tissue using magnetic resonance imaging. We
predict that at any given level of serum leptin, active children will have greater unOC.
Further, we predict that at any given body weight, active children will have greater bone
mass and lesser bone marrow adipose tissue than inactive children.