Within the last year, workers in the Section on Neuroendocrinology and their collaborators have advanced our understanding of the regulation of melatonin production. Figure 1 summarizes the dynamics of melatonin production, showing that large daily changes in melatonin production are regulated by changes in the activity of the next-to-last enzyme in melatonin synthesis, arylalkylamine N-acetyltransferase (AANAT). This enzyme has a unique photochemical transduction role in mammalian biology as the interface between regulation and synthesis of melatonin. It was known that large changes in AANAT protein were regulated through cyclic AMP. However, the mechanism through which cyclic AMP acted was not entirely known. We have uncovered a previously unrecognized binding partner for AANAT that regulates the activity and stability of the enzyme. Cyclic AMP-dependent phosphorylation switches on binding.

Figure 26

Dynamics of melatonin production. Large changes in the rate of melatonin production are driven by changes in the rate at which serotonin is acetylated.

Identification of AANAT Binding Partners
Ganguly, Kim, Morin, Weller, Klein in collaboration with Hickman,a Obsil,a Dyda,a Namboodirib
Using affinity chromatography with immobilized AANAT and phosphorylated AANAT, Jonathan Gastel initiated an analysis of the proteins binding to AANAT, revealing that phosphorylation caused binding of a pair of proteins (~ 29 and 31 kDa; Figure 2). Using mass spectrometry, Howard Jaffe (NINDS) identified the proteins as members of the 14-3-3 family of proteins. Our subsequent studies showed that expressed pure phospho-AANAT and 14-3-3 bound to each other and coeluted as a single complex, demonstrating that binding is encoded in the molecules and that other molecules are not required for complex formation. Analysis of homogenates of sheep pineal glands of cells expressing AANAT revealed that most of the AANAT in cells exists in a phosphorylated form complexed to 14-3-3, thereby establishing the physiological relevance of the AANAT/14-3-3 complex.

Figure 27

SDS-PAGE analysis of proteins bound to immobilized AANAT (C ) and phospho-AANAT. (P).

Figure 28

3-D structure of the AANAT/14-3-3 complex.

Structure of the AANAT/14-3-3 Complex
Ganguly, Coon, Schwartz, Klein in collaboration with Obsil,a Dyda,a Steinbachc
As shown in Figure 3, the tight binding of AANAT and 14-3-3 proteins made it possible for us to determine the 3-D structure of AANAT, thus representing the first crystal structure of a 14-3-3 complex to be solved. AANAT was found to bind to an amphipathic binding groove in 14-3-3, and multiple interactions were found to stabilize the complex. However, the most crucial are a set of interactions between the phosphate of the phosphothreonine (pT in Figure 3) in the N-terminal region of phospho-AANAT. These interactions are essential for formation of the complex. As predicted, based on studies with phosphorylated binding peptides, binding occurs in the site in 14-3-3.
A second important interaction between AANAT and 14-3-3 proteins involves a loop of the enzyme that forms the binding pocket for serotonin. Crystallographic studies have suggested that, in the absence of substrates, the loop is floppy and that, in the presence of serotonin, it wraps around the substrate, binding it tightly. Examination of the configuration of the loop in the 14-3-3 complex revealed that binding constricts movement of the loop, thereby stabilizing the flexible loop in a conformation that promotes binding to serotonin. The effect was seen in experiments using both expressed proteins and intact cells, indicating that complex formation increases the activity of AANAT in the physiological range of serotonin concentrations.

Functional Effects of AANAT/14-3-3 Complexing
Ganguly, Coon, Gaildrat, Weller, Klein
We discovered another result of binding: when phospho-AANAT is bound to 14-3-3, it is protected against proteolytic destruction. It was already known that AANAT is rapidly destroyed upon termination of adrenergic cyclic AMP stimulation of the gland. The studies in Figure 4 demonstrated that 14-3-3 proteins prevent the proteolysis of phospho-AANAT.

Figure 29

14-3-3 proteins protect AANAT against proteoloytic attack. The image was developed by using an anti-ovine AANAT(1-205) serum. Thrombin was used as a model protease.

The recent developments summarized here clearly indicate that the AANAT/14-3-3 complex plays a pivotal role in regulating melatonin synthesis. Cyclic AMP-dependent phosphorylation of newly formed molecules of AANAT lead to binding of the protein to 14-3-3 proteins, which prevents proteolytic destruction and activates the enzyme by increasing affinity for substrates. A decline in cyclic AMP levels, which occurs during the day, leads to dephosphorylation of AANAT, which reduces association with 14-3-3 proteins. The result is destruction of AANAT and the termination of melatonin synthesis. Our work provides an excellent example of how cyclic AMP regulated protein-protein interactions play a critical role in physiological regulation of biological processes.