ADIPONECTIN

Our work on Adiponectin is a collaborative effort with the laboratory of Dr. Philipp Scherer of the Albert Einstein College of Medicine of Yeshiva University. Link to the Scherer Lab      In the course of sequencing a thousand clones from a subtractive library enriched for mRNA species induced in the course of adipocyte differentiation, we found a short cDNA clone that displayed a very interesting expression pattern as judged by Northern blot analysis.  Its expression was limited to adipose tissue and it showed strong induction in the course of adipocyte differentiation in tissue culture.  A full-length cDNA was isolated and sequenced.  The encoded protein, Adiponectin, was novel; it contained 247 amino acids and had a predicted molecular weight of 28 kD.     Adiponectin consists of an amino-terminal signal sequence, a stretch of 27 amino acids that does not show significant homology to any protein in the data base, and then 22 perfect GlyXPro or GlyXX repeats.  This is followed by the carboxy-terminal globular domain exhibits striking homology to a number of proteins, such as the globular domains of type VIII and type X collagens, the subunits of complement factor C1q and hib27, a protein found in the serum of hibernating animals during the summer months.  Structurally, albeit not at the primary sequence level, the protein resembles the lung surfactant protein and the hepatocyte mannan-binding protein, both of which have collagen-like domains and globular domains of similar size.  In a series of experiments described in Scherer et al., 1995, the initial characterization of Adiponectin was described.  It undergoes a lengthy intracellular assembly and maturation process that leads to a half-life of secretion of about three hours.  This is caused by post-translational modifications such as hydroxyprolination within the collagenous domain.  Furthermore, it is able to demonstrate homo-oligomer formation: three Adiponectin monomers assemble to a basic building block.  Four to five of these trimers assemble into a higher order structure to form a structure similar to the one observed for C1q - the first component of the classical complement fixation cascade that has a multimeric structure consisting of a hexamer of hetero-trimers.     We solved the crystal structure of the homotrimeric globular head domain of Adiponectin at a resolution of 2.1 angstroms.  This structure reveals an unexpected homology to the tumor necrosis factor family of cytokines.  Despite the lack of homology at the primary amino acid sequence level, the structural features between TNF and Adiponectin are highly conserved.  Both proteins have a ten-strand jelly-roll folding topology and form bell-shaped homotrimeric oligomers.  TNF family members share many functional similarities, playing roles in immunity as well as in energy homeostasis.  The evolutionary relationship between Adiponectin and TNF family proteins prompts us to look at Adiponectin as a cytokine and suggests that the Acrp30 receptor may also be a member of the TNF receptor superfamily.     Since our collaborators initial description of Adiponectin, two additional groups have reported similar findings: Spiegelman and colleagues reported the cloning of AdipoQ - a protein identical to Adiponectin.  Matsubara and colleagues reported the cloning of apM1, which is also identical to Adiponectin.  Spiegelnans group used differential display to isolate the AdipoQ mRNA and have demonstrated a significant reduction of AdipoQ mRNA in adipose tissue from obese mice and humans.  However, very little functional data is available in the literature to date.  In the preliminary data section of this grant, we will present evidence for a (direct or indirect) regulation of Adiponectin levels by insulin.     The two presently known adipocyte-specific secretory proteins - leptin and adipsin - have both been shown to be of high physiological relevance.  Adiponectin - also being an adipocyte-specific secretory protein - together with its structural and evolutionary relationship to TNF and its regulation by insulin, makes it a prime target for in-depth analysis that has not only the potential to deepen our understanding of the physiology of adipose tissue, but also to be relevant to the pathophysiology of diabetes.   Publications   Curr Biol 1998 Mar 12;8(6):335-8 The crystal structure of a complement-1q family protein suggests an evolutionary link to tumor necrosis factor. Shapiro L, Scherer PE on-line   abstract