Tartrate resistant acid phosphatase 5a : a potential regulator of adipocyte cell number and differentiation in white adipose tissue

Abstract

Tartrate-resistant acid phosphatase (TRAP) exists in two isoforms, TRAP 5a which is monomeric and TRAP 5b which is a dimer generated by proteolytic cleavage of TRAP 5a, that exhibit different functions and localizations. TRAP 5a is expressed by adipose tissue macrophages and secreted into the extracellular environment and has been shown to lead to hyperplastic insulin-sensitive obesity when over-expressed in mice. In bone, TRAP is suggested to interact with the heparan sulfate proteoglycan (HSPG) glypican-4. In humans, TRAP 5a serum levels and TRAP in adipose tissue correlate to BMI. TRAP 5a has been shown to have a stimulatory effect on the proliferation and differentiation of pre-adipocytes, preosteoblasts and hematopoietic cells but the mechanism of action remains unknown.

The aims of this thesis were (1) to develop an ELISA for quantification of human TRAP 5a and to evaluate TRAP 5a levels in lean vs. obese individuals and (2) to investigate the mechanism of action of TRAP 5a in pre-adipocytes and the effects on the homeostasis-related events such as cell cycle entry, differentiation and migration using the pre-adipocyte cell line 3T3-L1.

ATRAP 5a ELISA was developed by generation of monoclonal antibodies specific for TRAP 5a and used to measure TRAP 5a in serum from lean and obese females. TRAP 5a in serum of obese females was reduced compared to lean but a positive correlation between BMI >30 and serum levels of TRAP 5a was identified. Regarding effect of TRAP 5a on 3T3-L1 cells, TRAP 5a was shown to bind to the components of extracellular matrix (ECM) heparan sulfate and nidogen-2.

TRAP 5a was shown to bind to the pre-adipocyte cell membrane and to co-localize with HSPG glypican-4. After cell surface binding, TRAP 5a is endocytosed in a caveolaemediated manner and co-localizes with early endosomal marker Rab-7. TRAP 5a was also observed in multivesicular bodies but did not co-localize with lysosomal marker LAMP-1. Mesenchymal stem cells, fibroblasts and pre-osteoblasts were also tested for their ability to bind TRAP 5a and only pre-osteoblasts exhibited TRAP 5a endocytosis. TRAP 5a was shown to enhance the number of cells entering S phase after cell cycle arrest by serum deprivation and to increase cyclin D1 levels.

In a phosphorylation signaling array for proteins of the Akt pathway it was shown that TRAP 5a causes increased inactivation phosphorylation of GSK3β that is most likely not dependent on Akt since Akt activation phosphorylation was reduced after TRAP 5a treatment. Proteins controlling cell cycle arrest and apoptosis p53, p38 and Bad exhibited altered phosphorylation pattern after TRAP 5a treatment that could potentially lead to their inactivation.

TRAP 5a treatment induced changes in morphology of pre-adipocytes andenhanced their migration. Focal adhesion protein (FAK), paxillin and actin were downregulated after TRAP treatment indicating cytoskeletal changes and changes in the adhesion pattern. TRAP 5a treatment lead to enhanced lipid droplet acquisition after 2 days ofdifferentiation induction in pre-adipocytes.

In summary, TRAP 5a interacts with components of the ECM, binds to the surface of pre-adipocytes and co-localizes with glypican-4 and isendocytosed in a caveolae-dependent manner. Treatment of pre-adipocytes with TRAP 5a leads to enhanced cell numbers, increased migration, morphological changes and increased differentiation, all consistent with a growth factor-like role for TRAP 5a in adipose tissue.