Human ABAD/HSD10 with Bound Inhibitor, AG10851

Anna Zimmermann '07 and Ann Downer '08


I. Introduction

Amyloid ß-peptide-binding alcohol dehydrogenase or 17ß-hydroxysteroid dehydrogenase type 10 (ABAD/HSD10) is a type of oxidoreductase and is a member of the short-chain hydroxysteroid dehydrogenase (SDR) family.  This enzyme is usually found in mitochondria, particularly in the mitochondria of neurons.  ABAD/HSD10 has been shown to catalyze the oxidation of simple alcohols and hydroxyl steroids, including the estrogen precursor, 17ß-estradiol.  Under normal conditions, ABAD/HSD10 functions as a cytoprotective enzyme involved in maintaining homeostasis. Thus, it influences both a cell’s response to metabolic stress and estrogen levels within the cell. 

Recently, ABAD/HSD10 has been of particular interest due to its ability to bind amyloid ß-peptide (Aß), a “proteolytic fragment of the integral membrane glycoprotein, amyloid-ß precursor protein” (Kissinger et al., 2004).   Recent studies suggest that this protein binding interaction may play a role in the pathogenesis of Alzheimer’s disease (AD).  Accumulations of Aß in plaques in the brain have been correlated to Alzheimer’s disease, a chronic illness which causes memory loss and cognitive impairment and affects more than 15 million worldwide (Xie et al., 2006).  Upon binding Aß, ABAD/HSD10 moves to the plasma membrane, loses its ability to function in many metabolic pathways, and instead promotes free radical formation (Yan and Stern, 2005).  It appears as though binding Aß distorts the structure of ABAD/HSD10, thereby inducing cellular dysfunction, defined by the presence of reactive aldehydes and DNA fragmentation within the cell (Powell et al., 2000).   The inactivation of ABAD/HSD10 by amyloid ß-peptide is therefore thought to impact cells in a way that reflects the onset of AD (Yan and Stern, 2005).

Investigating the structure, function, and binding capabilities of ABAD/HSD10 might help shed light on the causes of and potential treatments for Alzheimer’s disease. Specifically, Kissinger et al. (2004) investigated the mechanism by which ABAD/HSD10 binds the inhibitor molecule, AG18051.  By elucidating how ABAD/HSD10 binds this inhibitor, researchers may improve their ability to effectively design Alzheimer's disease therapeutics which will target and bind ABAD/HSD10, thereby preventing the potentially pathogenic binding of Aß to ABAD/HSD10.

II. General Structure

ABAD/HSD10 exists as a tetramer both in solution and in the asymmetric unit of a crystal.  Each monomer is 261 amino acids in length and is comprised of a single domain with an active site. This structure is characteristic of members of the SDR enzyme family. The four subunits composing ABAD/HSD10 are nearly identical to each other except for a mobile loop (residues 205-220) that is positioned near the substrate binding site of each subunit.

The chain fold present in ABAD/HSD10 is characteristic of the Rossman fold dinucleotide-binding motif, which is seen in other SDR enzymes. This motif consists of a ß-sheet composed of 7 parallel stands and 6 α-helices – 3 helices positioned on each side of the ß-sheet.  

Ser155, Tyr168, and Lys172 form the “catalytic triad” of this enzyme (Kissinger et al., 2004).  These residues exist in an area located at one end of the central ß-sheet.  The location of these three residues does not appear to change when ABAD/HSD10 binds a substrate.

Unlike other SDR enzymes, each ABAD/HSD10 monomer contains two insertions, one composed of residues 102-107 and the other consisting of residues 141-146.   The exact function of these insertions is unclear at present; however, ABAD/HSD10 interaction was blocked by antibodies raised against peptides encompassing the insertion regions, suggesting their potential medical significance. The insertions create two respective loops that project from opposite ends of each of the four subunits .  It has been suggested that these loops aid in the binding of CoA-linked substrates, and that they may have evolved in order to allow ABAD/HSD10 to “accommodate” a wider variety of substrates. As a result, ABAD/HSD10 demonstrates an unusually broad substrate specificity (Powell et al., 2000).

The insertion present at residues 102-107 creates a short ß-hairpin near the substrate binding site. It also contacts α-helix E2 of an adjacent monomer, thereby increasing the contact between these particular monomers.

The second insertion is present between α-helix E2 and ß-sheet E at residues 141-146. The insertion also serves to increase “subunit interface” by contacting Phe223 on a neighboring monomer. A notable hydrogen bonding network is present within the loop that forms.  

III. Subunit Interactions

The interactions between the subunits of the ABAD/HSD10 tetramer are consistent with other SDR tetramer enzymes.  Each monomer contacts the other three domains, and the four active sites of the enzyme face toward the outside of the tetramer.   These interactions can be viewed by the three “mutually perpendicular 2-fold axes,” which relate each monomer’s position to the entire tetramer (Kissinger et al., 2004).  In the ABAD/HSD10 rat homologue, type II hydroxyacyl-CoA dehydrogenase/amyloid-ß binding alcohol dehydrogenase (rHADH II/ABAD), the tetramer is formed when two dimers unite in a back-to-back manner.  ABAD/HSD10 has been shown to share 88% sequence identity with rHADH II/ABAD so it may be extrapolated that the ABAD/HSD10 tetramer comes together in much the same way (Powell et al., 2000).  Additionally, as noted with respect to rHADHII/ABAD, the C-terminal region of each ABAD/HSD10 monomer contacts the C-terminal region of an adjacent monomer.

IV. Substrate Binding

An unusual property of this enzyme is its ability to bind a broad range of substrates.  When bound to the inhibitor AG18051 , the substrate binding region of ABAD/HSD10 takes the shape of a long and narrow cavity.  This cleft consists of three main parts.  Two short polypeptide chains, consisting of residues 95-99 and residues 155-168, form one side of the substrate binding cleft.  The other side of the cleft consists of the “substrate binding flap,” residues 205-217.   ABAD/HSD10’s broad substrate specificity may result from the flexibility associated with the this flap.  Finally, the “floor” of the cavity is created by residues 257-261 , located at the C-terminus of the subunit.  As a whole, the substrate binding region is hydrophobic, and a series of lysine residues (Lys99, Lys104, and Lys105) are present directly outside the cleft at the ß-hairpin of the first insertion region (residues 102-107) .

In rHADH II/ABAD, the closure of the substrate binding flap (residues 205-220) has been found to correlate directly to substrate binding and enzymatic activity.  This closure may enhance enzymatic activity by blocking solvent access to the active site.  In some cases this flap assumes a small helical arrangement upon substrate binding.  Yet, this helix formation does not result when rHADH II/ABAD binds all possible substrates, further illustrating the broad substrate specificity of rHADH II/ABAD and ABAD/HSD10.

V. Binding of the Inhibitor, AG18051

Each monomer of the ABAD/HSD10 enzyme is complexed with NAD+ and an inhibitor.   In crystalline form, ABAD/HSD10 identically binds AG18051 at three of the four substrate binding sites of the tetramer.  (When crystallized, it appears as though the substrate binding site of one monomer is altered and narrowed so that it is unable to bind an inhibitor, but still binds NAD+ ).  When bound, AG18051 interacts with the cofactor, NAD+, to form a “covalent adduct” between the N2 atom of AG18051 and the C4N atom of NAD+.  Binding of AG18051 is dependent on the presence of NAD+ and does not occur effectively when NAD+ is absent or when NADH is present.  Binding data indicates that binding of NAD+ and AG10951 is cooperative and that ABAD/HSD10 actively aids in adduct formation.  Though the structure of AG18051 does not resemble that of other known ABAD/HSD10 substrates, this inhibitor interacts closely with the substrate binding cleft at nine specific residues.   In addition to general hydrophobic contacts, two hydrogen bonds are formed, specifically between N1 of AG18051 and Try168 of ABAD/HSD10 and between the carbonyl oxygen of AG18051 and Gln165 of ABAD/HSD10.

VI. Medical Implications:

Despite widespread interest in Alzheimer’s disease, no cure or effective treatment has yet been developed to address this illness (Xie et al., 2006).  The onset of Alzheimer’s disease is characterized by the build up of plaques, consisting primarily of Aß, within and surrounding nerve cells.  The presence of increased cellular levels of Aß is thought to result in neurotoxicity, likely through a mechanism involving the binding of ABAD/HSD10.  Normally, ABAD/HSD10 plays a role in metabolic homeostasis by helping to metabolize energy sources and to break down isoleucine and branched-chain fatty acids (Yan and Stern, 2005).   However, when Aß is present in high concentrations, it is more likely to bind ABAD/HSD10, resulting in the loss of function of ABAD/HSD10 due to structural changes within the enzyme.  When inhibited by Aß, ABAD/HSD10 is not able to bind and process other substrates, and the subsequent build up of these metabolites may have adverse effects within the cell (Yan and Stern, 2005).  Recent studies also suggest that the binding of Aß to ABAD/HSD10 induces a signal cascade that leads to apoptosis (Xie et al., 2006).   The neuronal dysfunction associated with ABAD/ Aß binding and Aß build-up can be suppressed by interrupting the interaction between ABAD/HSD10 and Aß, a discovery which has vast clinical possibilities.  Following the release of human ABAD/HSD10’s crystal structure and information on its binding properties (Kissinger et al., 2004), a class of benzothiazole ureas have been shown to inhibit ABAD/ Aß binding in an ELISA- based assay (Xie et al., 2006).   

In order to develop effective AD therapeutics, more research needs to be undertaken to determine how exactly Aß binding results in the amyloid-mediated cell stress that is thought to cause Alzheimer’s disease and how ABAD/HSD10 functions in vivo.  Because rHADH II/ABAD is very closely related to ABAD/HSD10, one may utilize the rat model in investigating AD and potential treatments.  The molecular inhibitors of ABAD/ Aß binding identified by Xie et al. (2006) will soon be investigated in animal models of Alzheimer’s disease.  However, because both rHADH II/ABAD and ABAD/HSD10 play a critical role in many metabolic pathways, difficulties may arise when trying to engineer potential therapeutics that target this particular enzyme (Powell et al., 2000). 

VII. References

Kissinger, Charles R., Paul A. Rejto, Laura A. Pelletier, James A. Thomson, Richard E. Showalter, Melwyn A. Abreo, Charles S. Agree, Stephen Margosiak, Jerry J. Meng, Robert M. Aust, Darin Vanderpool, Bin Li, Anna Tempczyk-Russell, and J. Ernest Villafranca. 2004. Crystal Structure of Human ABAD/HSD10 with a Bound Inhibitor: Implications for Design of Alzheimer’s Disease Therapeutics. J. Mol. Biol. 342: 943-952.

Powell, A. J., J. A. Read, M. J. Banfield, F. Gunn-Moore, S. D. Yan, J. Lustbader, A. R. Stern, D. M. Stern, and R. L. Brady. 2000. Recognition of Structurally Diverse Substrates by Type II 3-Hydroxyacyl-CoA Dehydrogenase (HADH II)/ Amyloid-ß Binding Alcohol Dehydrogenase (ABAD). J. Mol. Biol. 303: 311-327.

Reddy, P. Hemachandra and M. Flint Beal. 2005. Are mitochondria critical in the pathogenesis of Alzheimer’s disease? Brain Research Reviews 49: 618-632.

Xie, Yuli, Shixian Deng, Zhenzhang Chen, Shidu Yan, and Donald W. Landry. 2006. Identification of small-molecule inhibitors of the Aß-ABAD interaction. Bioorganic & Medicinal Chemistry Letters 16: 4657-4660.

Yan, Shi Du and David M. Stern. 2005. Mitochondrial dysfunction and Alzheimer’s disease: role of amyloid-ß peptide alcohol dehydrogenase (ABAD). Int. J. Exp. Path. 86: 161-171.

Back to Top

Biomolecules at Kenyon    HHMI at Kenyon    Jmol Home    Biology Dept    COMMENTS and CORRECTIONS