JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M411104200 on October 28, 2004

J. Biol. Chem., Vol. 280, Issue 6, 4639-4648, February 11, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/6/4639    most recent
M411104200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hosfield, D. J.
Right arrow Articles by Aertgeerts, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hosfield, D. J.
Right arrow Articles by Aertgeerts, K.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Conformational Flexibility in Crystal Structures of Human 11{beta}-Hydroxysteroid Dehydrogenase Type I Provide Insights into Glucocorticoid Interconversion and Enzyme Regulation*

David J. Hosfield{ddagger}, Yiqin Wu, Robert J. Skene, Mark Hilgers, Andy Jennings, Gyorgy P. Snell, and Kathleen Aertgeerts

From the Syrrx Inc., San Diego, California 92121

Received for publication, September 28, 2004 , and in revised form, October 27, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Human 11{beta}-hydroxysteroid dehydrogenase type I (11{beta}-HSD1) is an ER-localized membrane protein that catalyzes the interconversion of cortisone and cortisol. In adipose tissue, excessive cortisol production through 11{beta}-HSD1 activity has been implicated in the pathogenesis of type II diabetes and obesity. We report here biophysical, kinetic, mutagenesis, and structural data on two ternary complexes of 11{beta}-HSD1. The combined results reveal flexible active site interactions relevant to glucocorticoid recognition and demonstrate how four 11{beta}-HSD1 C termini converge to form an as yet uncharacterized tetramerization motif. A C-terminal Pro-Cys motif is localized at the center of the tetramer and forms reversible enzyme disulfides that alter enzyme activity. Conformational flexibility at the tetramerization interface is coupled to structural changes at the enzyme active site suggesting how the central Pro-Cys motif may regulate enzyme activity. Together, the crystallographic and biophysical data provide a structural framework for understanding 11{beta}-HSD1 activities and will ultimately facilitate the development of specific inhibitors.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Obesity is an urgent and growing health problem (1). Because of excessive caloric intake, improper nutrition, and inadequate physical activity, the Center for Disease Control and Prevention has stated that obesity has risen at an epidemic rate during the past 20 years and appears to be worsening (www.cdc.gov/nccdphp/dnpa/obesity). At present, 1 in 4 Americans is clinically obese and often displays symptoms of a condition called the metabolic syndrome (2). Some of these symptoms include congestive heart failure, abdominal obesity, atherogenic dyslipidemia, hypertension, glucose intolerance, insulin resistance, and type II diabetes. Recent advances in our understanding of the molecular mechanisms relevant to the metabolic syndrome suggest that tissue-specific increases in cortisol levels through the action of 11{beta}-hydroxysteroid dehydrogenase activity (11{beta}-HSD)1 could contribute to the pathogenesis of this prevalent disease (3–6).

In humans, two 11{beta}-HSD isoforms mediate the interconversion of cortisone (inactive glucocorticoid) and cortisol (active glucocorticoid). 11{beta}-HSD type 1 (11{beta}-HSD1) is highly expressed in key metabolic tissues including liver, adipose tissue, and the central nervous system (7–9). In these tissues, 11{beta}-HSD1 uses NADPH to reduce cortisone to the active hormone cortisol that activates glucocorticoid receptors (10). 11{beta}-HSD type 2 (11{beta}-HSD2) is an NAD+-dependent dehydrogenase that is expressed in aldosterone-selective tissues (11–13). In these tissues, 11{beta}-HSD2 oxidizes cortisol to cortisone and prevents illicit activation of the mineralocorticoid receptor.

11{beta}-HSD1 belongs to the short-chain dehydrogenase/reductase (SDR) family of enzymes, of which over 60 members are found in humans (14–17). SDRs are reversible NAD(H)/NADP(H)-dependent oxidoreductases containing a structurally conserved {alpha}/{beta} nucleotide-binding Rossman fold. Within the core structure, two conserved motifs are shared among all SDR enzymes (18). A dinucleotide-binding P-loop forms a turn between a {beta}-strand and an {alpha}-helix and directly contacts the ribose sugar and pyrophosphate. A Tyr-X-X-X-Lys motif, often in concert with a conserved Ser that orients substrate, catalyzes proton transfer to and from reduced and oxidized reaction intermediates. A flexible region in SDR enzymes, that often changes conformation upon substrate binding to shield the active site from bulk solvent, mediates enzyme specificity (19–23). SDR oligomerization and intracellular localization is often mediated by extensions at the N and C termini.

11{beta}-HSD1 is a NADPH-dependent enzyme that functions predominantly as a reductase in vivo. In cells, a single N-terminal transmembrane helix and associated linker anchors the C-terminal catalytic domain within the lumen of the endoplasmic reticulum (ER) (24, 25). A growing body of evidence suggests that in the ER 11{beta}-HSD1 colocalizes with NADPH-generating hexose-6-phosphate dehydrogenase (H6PDH), an enzyme that has been genetically linked to cortisone reductase deficiency in people (26). Based on these results, it has been proposed that the 11{beta}-HSD1 reductase activity predominates in metabolic tissues because of an increased NADPH/NADP+ ratio within the ER lumen and/or through direct interactions with H6PDH. On the basis of in vitro assays, it has been suggested that 11{beta}-HSD1 is dimeric, has cooperative enzyme kinetics, and that a C-terminal, primate-specific cysteine residue (Cys272) forms reversible inter-subunit disulfides (27, 51). In contrast to what is observed in vivo, the reductase activity of 11{beta}-HSD1 is unstable in cell homogenates despite high levels of NADPH (28). These results strongly argue that enzyme localization to the ER membrane, and/or post-translational modifications are important for dictating enzyme activity and directionality.

Biochemical, genetic, and clinical data all suggest that selective inhibition of 11{beta}-HSD1 may offer a therapeutic approach for treating metabolic syndrome and other diseases. In mice, adipose-specific overexpression of 11{beta}-HSD1 leads to the development of visceral obesity, insulin-resistant diabetes, and hyperlipidemia (29). In contrast, 11{beta}-HSD1 knockout mice fed high fat diets have low intracellular glucocorticoid levels and are protected from obesity, diabetes, and dyslipidemia (30–32). Further validation of 11{beta}-HSD1 as a therapeutic target comes from experiments that show mice that are exposed to a selective 11{beta}-HSD1 inhibitor have enhanced hepatic insulin sensitivity and lower blood glucose levels (33, 34). 11{beta}-HSD1 inhibitors may also be efficacious as agents to treat cognitive disorders in the elderly and in type II diabetics. Consistent with results from 11{beta}-HSD1 knockout mice showing protection from age-induced glucocorticoid-associated hippocampal dysfunction (35), recent human studies with the nonspecific 11{beta}-HSD1 inhibitor carbenoxelone show a statistically significant improvement in verbal fluency and verbal memory in elderly men and patients with type II diabetes (36).

The development of new 11{beta}-HSD1 inhibitors would clearly benefit from an understanding of the enzyme structural features that mediate substrate and inhibitor binding. Here, we present two high resolution crystal structures of 11{beta}-HSD1 in complex with co-substrate NADP+ and the steroidal detergent molecule CHAPS. When combined with biochemical results, the structural results provide an informed basis for understanding cofactor and glucocorticoid recognition and reveal how four 11{beta}-HSD1 C termini converge to form a tetramerization motif that has not yet been characterized in others proteins. The tetramerization interface is highly malleable and changes conformation to alter the arrangement of active site residues appropriately positioned to modulate substrate binding and product release. Together, these results define the structural basis for 11{beta}-HSD1 enzyme activity and will ultimately aid in the design of therapeutics that have the potential to treat human diseases such as type II diabetes and obesity.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cloning and Expression—The 11{beta}-HSD1 gene was amplified from IMAGE clone 5193867 (ATCC clone 7277078), using PCR with the primers 5'-AACGAGGAATTCAGACCAGAGATG-3' and 5'-TTACTT-GTTTATGAATCTGTCCAT-3'. The resulting PCR product was Topo-cloned into a modified pBAD-ThioE vector (Invitrogen). The final construct contains the coding sequence for the catalytic domain of 11{beta}-HSD1 (residues 24–292) preceded by a MKHQHQHQHQHQHQQPL affinity tag. A single point mutation, C272S, as well as a C-terminal truncation (residues 264–292) were constructed from this plasmid using QuikChange mutagenesis (Stratagene). Recombinant 11{beta}-HSD1 was expressed in Escherichia coli DH10b-Tir cells (Invitrogen) grown overnight at 37 °C, in Luria broth (LB) supplemented to 0.05 mg/ml kanamycin and induced with 0.2% arabinose and 0.25 mM corticosterone. 11{beta}-HSD1 was purified from frozen cell pellets that were thawed and resuspended in lysis buffer (30 mM CHAPS, 50 mM Tris-HCl, pH 7.9, 0.15 M NaCl, 0.5 µl/ml benzonase, 1 µl/ml ReadyLyse), incubated for 30 min at room temperature, and then clarified by centrifugation. The resulting supernatant was loaded on Probond resin (Invitrogen), previously equilibrated with wash buffer (4 mM CHAPS, 50 mM Tris-HCl, pH 7.9, 0.25 M NaCl, 40 mM imidazole), and then washed with 10 column volumes of wash buffer. 11{beta}-HSD1 was eluted with 3 column volumes of wash buffer supplemented to 0.2 M imidazole. The eluate was extensively dialyzed against 4 mM CHAPS, 25 mM Tris-HCl, pH 7.9, 0.25 M NaCl, concentrated to 10 mg/ml, and flash-frozen in 50-µl aliquots for long term storage.

Crystallization and Structure Solution—Two crystal forms of C272S 11{beta}-HSD1 were identified using Nanovolume CrystallizationTM methods (37) and reproduced in larger volumes with a reservoir containing 20% polyethylene glycol 3350 and 100 mM MES buffer, pH 6.2. A single site lutetium derivative was obtained by soaking these crystals for 72 h in reservoir solution containing 1 mM lutetium acetate. Crystals were harvested in reservoir solutions supplemented with 20% ethylene glycol, and flash frozen by direct immersion in liquid nitrogen. For the two native crystals, x-ray diffraction data extending to 1.55 and 1.80 Å resolution were collected at the Advanced Light Source (ALS) Beam Line 5.0.3 using a wavelength of 1.00 Å. Both crystal forms of 11{beta}-HSD1 belong to the monoclinic space group P21 and have four monomers in the asymmetric unit. For the lutetium-derivatized crystal, 4 data sets extending to 2.20 Å resolution were collected at ALS Beam Line 5.0.2 using wavelengths of 1.2782, 1.3404, 1.3408, and 1.3412 Å. All data were integrated and scaled using HKL2000 (38). The position of the lutetium atom in the crystallographic asymmetric unit was determined by SHELXD (39), and MAD phases were calculated with SHARP (40). The initial lutetium-bound structure was built manually using the program Xfit (41) and refined without non-crystallographic symmetry restraints in REFMAC (42). Phases for the two high resolution native structures were obtained by molecular replacement using the program MOLREP, and then improved with Arp/Warp (43) as implemented in CCP4 version 4.0. Iterative cycles of manual model building with Xfit and refinement in REFMAC gave the final refined models. In one of the ternary complexes, electron density consistent with two non-active site steroid molecules was found at the interface between two enzyme dimers. Because of presumed mobility, we could not unambiguously determine the identity of the steroid and thus modeled them as the steroid nucleus of CHAPS. Data collection and refinement statistics are listed in Table I.


View this table:
[in this window]
[in a new window]
  		TABLE I
X-ray data collection and refinement statistics for 11{beta}-HSD1

 
Enzyme Kinetics—The dehydrogenase activity of 11{beta}-HSD1 was measured by incubating 50 nM enzyme with varying amounts of NADP+ (0.2–200 µM) and cortisol (0.2–200 µM) for 60 min in buffer consisting of 50 mM TRIS pH7.5, 150 mM NaCl, 0.5 mM EDTA, and 0.01% Brij35. NADPH production was monitored by measuring fluorescence emission at 460 nm (excitation at 340 nm) using a Molecular Devices SpectraMax Gemini XS fluorescent plate reader. Calibration curves showed a linear increase in fluorescence intensity with increasing NADPH concentration up to 100 µM under these conditions. A disulfide-linked form of wild-type 11{beta}-HSD1 was generated by incubating enzyme (300 µM) with 1 mM K3Fe(CN)6 for 2 h at room temperature followed by dialysis to remove excess oxidizing agent.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Biophysical and Kinetic Characterization of Recombinant 11{beta}-HSD1—To determine the oligomeric solution structure of our recombinantly expressed truncated forms of wild-type (residues 24–292) and C272S 11{beta}-HSD1 enzymes we used static light scattering (SLS) at an enzyme concentration of ~100 µM. The SLS data gave mondisperse molecular masses of 139,106 Da and 136,615 Da for the wild-type and C272S enzymes, respectively. As the theoretical molecular mass of an 11{beta}-HSD1 subunit is ~32,000 Da, these results support tetrameric assembly of enzyme subunits to form the functional enzyme. Since SDR enzymes often form tetramers through complementary interactions of residues C-terminal to the catalytic core, we also determined the molecular mass of a C-terminally truncated enzyme missing the last 28 residues (residues 24–264). The SLS data for this truncated enzyme gave an apparent molecular mass of 63,588 Da. This result is consistent with the truncated enzyme forming dimers in solution and verifies that the enzyme C terminus mediates tetramer assembly. We further explored the possibility of forming enzyme disulfides by oxidizing the wild-type enzyme with K3Fe(CN)6. Using mass spectrometry we verified that this procedure generates a mixture of 31,807 Da monomers and 63,615 Da disulfide-linked dimers (Fig. 1a). This disulfide-linked form of the enzyme was rapidly eliminated by the addition of 1 mM dithiothreitol (Fig. 1a). K3Fe(CN)6 addition to the C272S enzyme did not catalyze disulfide bond formation (not shown) indicating Cys272 mediates intermolecular disulfide formation in the wild-type enzyme.



View larger version (27K):
[in this window]
[in a new window]
  		FIG. 1.
Mass spectrometry and kinetic data for oxidized and reduced 11{beta}-HSD1. a, 11{beta}-HSD1 was treated with K3Fe(CN)6 to oxidize enzyme disulfides. The peak at 31,807 corresponds to the reduced monomer and the peak at 63,615 corresponds to a Cys272-mediated disulfide-linked dimer (left panel). Disulfide bond formation in 11{beta}-HSD1 is reversible as addition of dithiothreitol eliminated the disulfide-linked dimers. b, schematic showing the oxidation reaction catalyzed by the dehydrogenase activity of 11{beta}-HSD1. The kinetic constants for this activity using Cys272 disulfide-linked and Cys272 sulfhydryl forms of the enzyme are indicated.

 
The dehydrogenase activity of the Cys272 disulfide-linked, the Cys272 sulfhydryl, and the C-terminally truncated forms of 11{beta}-HSD1 was determined using a fluorescence assay that measures the rate of formation of NADPH. Upon oxidation, the enzyme showed a 2-fold increase in kcat that was coupled to a 4-fold increase in Km for substrate, resulting in a net 2-fold decrease in catalytic efficiency (kcat/Km) for cortisol oxidation (Fig. 1b). The C-terminally truncated enzyme showed almost identical activity to the Cys272 sulfhydryl form of the enzyme (data not shown). These results, when combined with our mass spectrometry data, reveal that while the enzyme C terminus is not crucial for catalysis, intermolecular Cys272 disulfide bond formation must significantly alter its structure to affect the substrate-binding and catalytic properties of the enzyme.

Subunit Structures and Topology of 11{beta}-HSD1—Two crystal forms of 11{beta}-HSD1 grown using identical crystallization reagents were identified while screening heavy atom derivatives. Both crystal forms, constructed from what we herein term "interface-open" and "interface-closed" conformations of the enzyme, belong to similar P21 space groups although the b-axis is ~7-Å longer in the interface-open crystals (Table I). We used the MAD phases derived from the lutetium-derivatized enzyme to solve the structure of both crystal forms. The interface-open structure was solved to a resolution of 1.55 Å and the interface-closed structure was solved to a resolution of 1.80 Å. The overall Rossman fold topology of 11{beta}-HSD1 resembles other SDR enzymes in which a central 6-stranded, all-parallel {beta}-sheet is sandwiched on both sides by 3 {alpha}-helices (Fig. 2a). A conformationally variable {beta}6-{alpha}6 insertion that forms one wall of the steroid binding pocket, an additional {beta}-strand ({beta}7) and two C-terminal {alpha}-helices ({alpha}E and {alpha}F) are appended to the core structure and complete the 11{beta}-HSD1 fold. Aside from large differences in their quaternary structures, (see below) the eight 11{beta}-HSD1 subunits in the two crystal forms, including bound cofactor and detergent are similar, with 0.21 Å2 root mean square (RMS) deviations for all aligned C{alpha} atoms. In the structures, each active site shows well defined electron density for the cofactor NADP+ and the steroidal detergent CHAPS (Fig. 2b).



View larger version (50K):
[in this window]
[in a new window]
  		FIG. 2.
Overall subunit topology and cofactor and steroid binding pockets. a, stereo view of an 11{beta}-HSD1 dimer from the interface-closed structure showing overall topology with one subunit colored green and the other magenta. The active site is located where the NADP+ (yellow) and CHAPS (green) molecules converge. The N terminus is colored yellow, the enzyme C terminus tetramerization motif is colored red, and the conformationally variable {alpha}6-{beta}6 insert is colored blue. The {alpha}-helices and {beta}-strands common to the core SDR fold are labeled with numbers, while those specific to 11{beta}-HSD1 are labeled with letters. Figs. 2, 3, 4, 5 were generated with software from Advanced Visual Systems. b, typical electron density maps with refined model for the 11{beta}-HSD1 active site cofactors. The 2Fo–Fc electron density is contoured at 5{sigma} (red) and 1{sigma} (blue). c, stereo image of the cofactor binding site in 11{beta}-HSD1. In the complex, NADP+ (yellow) is well ordered and bound in a fully extended conformation. Numerous direct and water mediated interactions (see text) anchor the cofactor so its redox active nicotinamide ring is within close proximity to the bound steroid (green). d, stereo image of the steroid binding pocket and interactions with NADP+. In the complex, CHAPS (green) is well ordered and bound with its A-ring deeply penetrating the enzyme active site. Hydrophobic residues (silver) from multiple enzyme loops and helix {alpha}C make {alpha}-face, {beta}-face, and edge-on interactions with the steroidal core of the bound detergent molecule.

 
Cofactor Binding—NADP+ binding to 11{beta}-HSD1 is similar to other SDR enzymes, with the molecule binding in an extended conformation in which both ribose sugars adopt a C2-endo conformation (Fig. 2c). Because of the exceptionally high quality of the experimental electron density in both ternary complexes (Fig. 2, b and c), we could accurately measure the C2N-C3N-C7N-O7N torsion angle. This torsion angle, which defines the planarity of the cofactor carboxamide relative to its attached pyridine ring is indicative of the redox state of the cofactor (22). In each 11{beta}-HSD1 subunit, this angle is very close to 180°, indicating that oxidized NADP+, rather than NADPH is present in the crystal.

The adenine and nicotinamide rings are both well ordered and bind roughly perpendicular to the plane of the ribose sugars with the adenine adopting an anti configuration and the nicotinamide adopting a syn configuration. The adenosine moiety lies in a cleft formed by 4 loops ({beta}1/{alpha}1, {beta}2/{alpha}2, {beta}3/{alpha}3, and {beta}4/{alpha}4) and an {alpha}-helix ({alpha}4). Direct hydrogen bonds between adenines N-1 and extracyclic N-6 atoms, with Thr92 O{gamma} and Met93 NH, as well as stacking interactions with Arg66 on one side and packing interactions with His120 on the other, mediate purine binding. The ribose sugar sits above the backbone of the first P-loop Gly (Gly41) and is secured to the protein by packing interactions with Lys44 and a hydrogen bond between its 3'-OH and Ser43 O{gamma}. 11{beta}-HSD1 specificity for NADP+ over NAD+ is mostly mediated through an electrostatic interaction between the ribose 2'-phosphate and the guanidinium N{epsilon} and N{zeta} atoms of Arg66. Additional contacts to the ribose 2'-phosphate by the backbone amides of Arg66 and Ser67, Ser67 O{gamma}, as well as protein-ligated water molecules, further anchor the NADP+ to the enzyme. The negatively charged pyrophosphate binds at the N terminus of {alpha}1 and makes direct interactions with the backbone amide of Ile46.

The NADP+ ribityl nicotinamide is bound at the enzyme active site and conserved SDR residue Lys187 ligates the 2'- and 3'-ribose hydroxyls (Figs. 1b and 2a). Consistent with recent biochemical and structural results on 3{beta},17{beta}-hydroxysteroid dehydrogenase (18), the Asn143 backbone carbonyl ligates a conserved water that functions as part of a proton relay system with bulk solvent. The oxidized nicotinamide ring is maintained in a syn orientation relative to the ribose by three hydrogen bonds between the carboxamide side chain with the backbone atoms of Ile218 and an oxygen atom of the NADP+ pyrophosphate (Fig. 2c).

Steroid Binding Pocket—In both 11{beta}-HSD1 ternary complexes the steroidal core of the bound detergent is well ordered and in identical conformations in each of the 8 independent copies of the two different crystallographic asymmetric units. The steroid A ring is positioned at the catalytic end of the substrate binding cleft and the zwitterionic tail of the detergent extends toward solvent through a channel formed by residues of the flexible substrate-binding region and the C terminus of two adjacent subunits (Fig. 2, a and d).

The three {alpha}-directed steroidal hydroxyl groups of CHAPS all form bifurcated hydrogen bonds with protein and cofactor atoms in the ternary complexes. The 12-hydroxyl interacts with an active-site water molecule and the oxygen atom of the NADP+ carboxamide. The cis-trans AB ring conformation orients the A-ring C-3 hydroxyl to deeply penetrate the catalytic cleft where it interacts with two NADP+ pyrophosphate oxygens. The C-6 hydroxyl hydrogen bonds to an active site water molecule and the oxygen atom of Tyr183 from the catalytic Tyr-X-X-X-Lys motif. With the exception of these hydrophilic contacts, the rest of the steroid-binding pocket is exclusively hydrophobic. Residues Ile121, Thr124, and Leu126 from the {beta}4-{alpha}4 loop, Val180 from the {beta}5-{alpha}5 loop, and Thr222, Ala226, and Val227 from {alpha}C form the protein surface in close vicinity to the {beta}-face of the steroid. The NADP+ nicotinamide and the {beta}6-{alpha}D loop residues Leu215, Gly216, and Lys217 are within van der Waals contact distance of the {alpha}-face of the steroid, whereas {beta}5-{alpha}5 residues Ala172 and Tyr177, {alpha}5 residue Tyr183, {beta}6 residue Leu217 and {alpha}C residue Ala223 make edge-on interactions.

Crystal structures and biochemical studies of other SDR enzymes have established that substrate binding induces an ordering of the nonconserved and conformationally variable {beta}6-{alpha}6 insert to provide enzyme specificity and protect the active site from bulk solvent to promote hydride transfer (14, 17). In 11{beta}-HSD1, this region comprising residues Ile219 to Pro237 assumes different conformations. In one subunit of the interface-closed structure it adopts a helix-loop-helix motif and in a second subunit it forms a helix extended-strand conformation (Fig. 2a). In two of the interface-closed subunits several residues of this region were disordered and not included in the final model. In the interface-open structure, one subunit forms a helix-loop-helix motif while the other three adopt the helix-extended strand conformation.

11{beta}-HSD1 Subunit Assembly—11{beta}-HSD1 forms tetramers in solution and in the crystal lattice of both ternary complexes. The tetramer is constructed from two dimers, which are assembled through hydrophobic interactions between residues located on {alpha}4 and {alpha}5 (Figs. 2a and 3a), as seen for other SDR enzymes (20–22). To form tetramers, two dimers associate through complementary interactions between pairs of enzyme C termini that are oriented in an antiparallel direction. Each C terminus in the 11{beta}-HSD1 tetramer runs roughly perpendicular to the dimer 2-fold axis. This arrangement of termini generates a ~30-Å long 4-helix bundle-like structure that localizes adjacent enzyme active sites to the top and bottom of the structure (Fig. 3a). In the assembled tetramer, the N terminus of each subunit is close to the N terminus of the subunit from an adjacent dimer (Fig. 3a). The two dimers of the tetramer are not symmetrically related: A twisting-like motion about the center of the tetramer rotates one dimer by ~30° relative to the other.



View larger version (60K):
[in this window]
[in a new window]
  		FIG. 3.
Quaternary association of 11{beta}-HSD1 subunits and associated interactions and conformational changes. a, overall topology of the 11{beta}-HSD1 interface-closed (left) and interface-open (right) tetramers. NADP+ (yellow) and CHAPS (green) bind at two enzyme active sites located at the top and bottom of the enzyme tetramer. Each enzyme subunit is colored differently except for the enzyme N and C termini, which are colored yellow and red, respectively. The conformationally variable {alpha}6-{beta}6 inserts are colored blue. The 3{sigma} electron difference density (blue) at the center of the interface-open tetramer shows the location of the bound steroid molecules. In both structures, all four N and C termini are localized to the cleft between adjacent dimers. In one dimer pair of the interface-closed structure, the C-terminal tail (red) extends into the active site of the adjacent subunit. b, conformation and residues of the interface-closed and interface-open tetramers. In both structures, four enzyme C termini (red) converge to form a 4 helix-bundle like structure that mediates tetramer formation. The tetramer is formed by complementary hydrophobic interactions at both ends of the interface in the closed structure and through steroid-mediated interactions in the open structure. Pro271 and Ser272 (gold) are localized to the center of the enzyme tetramer, which in the closed structure is occupied by solvent molecules (red) and in the open structure is occupied by bound steroid. Trp263 and Tyr280, which change conformation in transitioning between the two structures, are colored purple. The green arrows indicate the paths substrate and product must traverse to access the active site. In the interface-closed structure, movement of Trp263 and Tyropen an enzyme conduit (green arrow) that allow interface-bound steroids to access the active site. c, network of salt bridge interactions links side chains on C-terminal helix {alpha}F to core residues from the same subunit. Ser195, a possible protein kinase C phosphorylation site, participates in this in this network of interactions.

 
In both ternary complexes the C-terminal enzyme tetramerization motif adopts similar helix-strand-helix conformations (Figs. 2a and 3, a and b). The first helix ({alpha}E) is comprised of residues Ser261 to Arg269, is largely hydrophobic, and is terminated by Asn270 and helix-destabilizing Pro271. Together, these two residues form an extended structure that interrupts the two helices. Ser272, which was mutated from Cys follows Pro271 and initiates the second {alpha}-helix ({alpha}F), which is highly amphipathic. Charged residues on one helical face (Arg273, Lys274, and Glu277) anchor the enzyme C terminus to charged core residues (Asp191, Arg198, Lys199, Glu254, and Glu255) from the same subunit (Fig. 3c). Ser195, which is located adjacent to a possible protein kinase C phosphorylation site (44), is centrally located in this cluster of charged residues that are unchanged in both structures. Hydrophobic residues on the other face of {alpha}F are directed toward the interior of the tetramerization interface where they mediate intradimer contacts in the interface-open structure and both intra- and interdimer contacts in the interface-closed structures. In two of the four interface-closed 11{beta}-HSD1 subunits, a short ~10 residue C-terminal tail extends from Tyr280 at the end of {alpha}F to the conformationally flexible {beta}6-{alpha}6 insert of an adjacent subunit in the same dimer (Figs. 2a, 3a, and 5b). Direct interactions between C-terminal tail residue Arg288 with the backbone carbonyl of steroid contacting residues Ala-226 in one subunit and Ala-228 in another anchor the C-terminal tail to the conformational-variable {beta}6-{alpha}6 insert. In the other two subunits of the interface-closed structure, and in all 4 subunits of the interface-open structure, the C-terminal tail is disordered.



View larger version (55K):
[in this window]
[in a new window]
  		FIG. 5.
Conformational changes and interactions at the 11{beta}-HSD1 tetramer interface. a, structure and conformation of the Pro-Cys motif (gold) at the center of the 11{beta}-HSD1 tetramer reveal additional conformational changes are needed to form the disulfides characterized in biochemical assays. The Ser O{gamma}-O{gamma} distances in both structures suggests similar S{gamma}-S{gamma} distances in the wild-type enzyme are to long to form either intra- or interdimer disulfides. b, illustration of the interface-closed and interface-open tetramers reveal conformational changes localized at the two ends of the structure. In the closed structure Trp263 (purple) from separate dimers are wedged together, which, in combination with Tyr280, form a continuous groove that leads to the active site (indicated by bound CHAPS, green). Trp263 and Tyr280 conformational changes in transitioning to the open structure disrupt this groove and further open an enzyme conduit allowing steroids to access the active site from the center of the tetramer interface. Interactions between the C-terminal tail (red) and the conformationally variable {alpha}6-{beta}6 insert (blue) are visible in two of the four subunits of the interface-closed tetramer. The 4 enzyme subunits are colored as in Fig. 3a. The view is rotated ~90° to that in Fig. 3a.

 
Despite the highly similar helix-strand-helix conformations of the enzyme C terminus, the 11{beta}-HSD1 tetramer interface is remarkably different in the open and closed ternary complexes. In the closed structure, solvent molecules fill a ~8 Å x 10 Å x 12 Å central chamber located at the middle of the interface (Fig. 3b). In the open structure, two bound steroid molecules occupy a similar position (Fig. 3, a and b). The presence of these steroids forces the two dimers apart, each moving as rigid bodies by ~3 Å from their positions in the interface-closed conformation (Figs. 3, a and b, and 5). This steroid-induced rigid body movement disrupts most of the interdimer contacts observed in the closed structure and opens the ends of the interface to expose a mostly hydrophobic channel that measures ~13 Å x 13 Å x 30 Å. The interface-bound steroid molecules are oriented perpendicular to the channel axis and pack with hydrophobic side chains that line the channel walls (Fig. 3, a and b).

Transition between the interface-open and interface-closed conformations occurs with striking structural rearrangements at both ends of the tetramer interface (Figs. 3b and 5b). In the interface-closed structure, C-terminal residues from each enzyme subunit converge at both ends of the tetramer to form two highly complementary interfaces that are separated by the solvent-filled central chamber. These two interfaces mediate all of the dimer-dimer contacts observed in the structure and include two Trp263 residues from different dimers that are wedged together by packing interactions with Leu262, Leu266, and Leu267 from the same C-terminal helix and Ile275, Leu276, Leu279, and Tyr280 from the two oppositely oriented C termini (Figs. 3b and 5b). At both ends of the tetramer, the juxtaposition of these hydrophobic residues forms a continuous groove that leads to the enzyme active site and positions Tyr280 within the active site of an adjacent dimer subunit (Figs. 3b and 5b). In the open structure, residues at the interface do not form complementary interactions and the two dimers can almost be considered independent since the tetramer is held together solely through interactions mediated by the centrally-bound steroids. The four Trp263 residues in the interface-open complexes are fully exposed to solvent and, as evidenced by their diffuse electron density, are highly mobile. Furthermore, Tyr280 backbone and side chain shifts of up to 13 Å repositions its aromatic side chain more deeply within the enzyme active site where it now packs against the {beta}-face of CHAPS (Figs. 3b and 5b). This repositioning of Tyr280 changes the direction of the backbone atoms C-terminal to it and reveals an enzyme conduit that links the active site with the center of the tetramer interface. The steroids bound at the center of the interface could access the active site through this enzyme conduit.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Glucocorticoid Recognition and Enzyme Catalysis—To better understand substrate binding and catalysis, we have used our 11{beta}-HSD1 ternary complex to model potential interactions between the enzyme and its reaction product cortisol (Fig. 4). Since 11{beta}-HSD1 catalyzes interconversion of C-ring 11-keto/hydroxy groups, we reasoned that the cortisol C11 hydroxyl would closely contact catalytic residues Ser170 and Tyr183. Using this chemical constraint we manually positioned cortisol in the active site and then energy-minimized the system using the MMFF force field (45) with some parts of the active site constrained. The resulting structure-based model predicts that 11{beta}-HSD1 natural substrate binds in an opposite and nearly perpendicular orientation to that of the steroid ring system of CHAPS. Unlike CHAPS in which a cis-trans A/B ring fusion orients the A ring hydroxyl toward the NADP+ nicotinamide, our model suggests that the cortisol D ring converges with the enzyme active site such that its C17 substituents interact with the NADP+ pyrophosphate. The A ring is directed toward the distal end of the binding pocket where its C-3 keto group points to solvent. The perpendicular binding orientation of cortisol relative to CHAPS alters the nature and identity of residues contacting its {alpha} and {beta} substituents. Our structural model predicts Ser170, Leu171, Ala172, Tyr177, and Tyr183 interact with the cortisol {beta}-face, Leu217, Ala223, Ala226, Val227, Val231, and Met233 with the {alpha}-face, while Leu206 makes an edge-on interaction.



View larger version (51K):
[in this window]
[in a new window]
  		FIG. 4.
Structure-based model for cortisol interactions. Illustration showing modeled cortisol (green) interactions with NADP+ (yellow) and the enzyme active site (silver). Our structure-based model predicts that the steroid D-ring deeply penetrates the enzyme active site.

 
Previous studies on SDR enzymes all support a common reaction mechanism in which an active site Tyr functions as a catalytic acid to protonate the reactive keto oxygen of the bound substrate. In these models for substrate interconversion an active site Ser is often found to stabilize the orientation of bound substrate. A conserved Lys forms hydrogen bonds with the nicotinamide ribose and lowers the pKa of the Tyr-OH to promote proton transfer. The arrangement of these catalytic residues in the 11{beta}-HSD1 active site is consistent with a similar mechanism for cortisone reduction. Thus, in our structure-based model for steroid reduction the 11-keto position of bound cortisone is anchored in the enzyme active site through interactions with Ser170 and Tyr183. The pKa of this tyrosine is reduced through electrostatic interactions with buried Lys-187. Together, these two residues increase the electrophilicity of the reactive C-11 atom facilitating proton transfer from Tyr183 to the substrate 11-keto oxygen. Concomitant hydride transfer from NADPH produces the reaction product cortisol and hydrogen exchange with bulk solvent through the conserved relay system (18) reprotonates Tyr183.

Quaternary Structure, C-terminal 11{beta}-HSD1 Disulfides, and Enzyme Regulation—The C-terminal structural motif that mediates the quaternary association of 11{beta}-HSD1 subunits described here has yet to be observed in other protein structures. Although the structure resembles a 4-helix bundle motif, the close localization of the four centrally located helix-disrupting Pro271 residues makes such a structure unattainable in 11{beta}-HSD1 and likely imparts extra flexibility in the tetramer that is important for enzyme activity. The close localization of the 11{beta}-HSD1 C termini also positions the four Ser272 side chains within close proximity to one another at the center of the tetramer. Within the 11{beta}-HSD1 dimer the distance between the two Ser272 O{gamma} atoms is ~7.6 Å. In the interface-closed tetramer the same O{gamma}-O{gamma} distance is reduced to ~6.4 Å for two Ser on similarly oriented helices, and to ~7.4 Å for the Ser-Ser pair of oppositely oriented helices (Fig. 5a). In the interface-open structure these distances are increased significantly (Fig. 5a). The positioning of these Ser272 residues is significant since our combined biophysical and kinetic data on the wild-type enzyme show that while the C terminus is not necessary for catalysis, the Cys272 intermolecular disulfides do in fact modulate enzyme activity (Fig. 1).

Three possible Cys272-Cys272 disulfide pairs (an intradimer disulfide or one of two interdimer disulfides) are consistent with our biochemical data yet the structures indicate significant rearrangements of the enzyme C terminus would be needed to reduce the observed separation of S{gamma} atoms (Fig. 5a) to be within the predicted ~2.0 Å S-S bond distance observed in other proteins. While our structural data does not allow us to distinguish the exact nature of the Cys272 disulfide-linked enzyme, three structural features of our C272S tetramer provide insight into how disulfide bond formation might be facilitated. First, Cys272 would be positioned close to the positively charged helix dipole located at the N terminus of {alpha}F (Figs. 3b and 5a). This positioning would increase the disulfide-forming propensity of Cys272 by lowering the pKA of the reactive sulfhydryl. Second, Cys272 resides within the solvent-filled central chamber of the interface-closed tetramer (Fig. 3b). This chamber is large enough to accommodate disulfide exchange catalysts such as glutathione, suggesting these agents, which are abundant in the ER lumen, could assist disulfide bond formation in cells. Third, Cys272 directly follows Pro271. As recent data show a significant increase in the rate of proline isomerization upon disulfide bond formation in cyclic peptides, (46), and since proline isomerases are frequently localized to the ER lumen (47) isomerization of the protein backbone at Pro271, might help facilitate the conformational changes needed to form the C-terminal 11{beta}-HSD1 disulfides characterized here.

In cells, localization of the 11{beta}-HSD1 catalytic domain within the oxidizing environment of the ER lumen would further facilitate formation of the 11{beta}-HSD1 disulfides. Our structures suggest that formation of interdimer disulfides would stabilize the enzyme tetramer by covalently linking two enzyme dimers. The effect of forming intradimer disulfides, however, is difficult to predict. Intradimer linking might increase the hydrophobicity at the center of interface enhancing tetramer formation. Conversely, intradimer disulfides could induce a conformational change in the enzyme C terminus that is incompatible with tetramer formation and would thus favor discreet 11{beta}-HSD1 dimers. Regardless of the exact nature of the disulfide linkages in cells, cellular stresses that change the redox potential of the ER might change the nature of these disulfides, allowing the enzyme to adopt discreet conformations depending on the exact cellular environment. In this regard it is notable that oxidative damage associated with diabetes mellitus shifts the ER redox environment to a more reducing state (48), suggesting that the oxidation state of the 11{beta}-HSD1 disulfides may be altered as the disease progresses.

Implications for Enzyme Activity in Cells—At both ends of the 11{beta}-HSD1 tetramer, the zwitterionic tail of the bound CHAPS extends out the active site and sits close to the ends of the tetramer interface (Figs. 3a and 5b). Notably, the CHAPS tail is chemically very similar to the head group of phosphatidylcholine, and suggests that the interactions observed here might mimic the binding of membrane-bound phospholipids within the lumen of the ER. Integrating this information with our structural data for the interface-open and interface-closed conformations, we suggest a specific structural basis for enzyme localization within the ER membrane. Docking our two 11{beta}-HSD1 ternary complexes onto the luminal surface of the ER membrane suggests two possible models for membrane association of the 11{beta}-HSD1 tetramer in which four enzyme C termini run roughly perpendicular to the membrane surface (Fig. 6). In the first model all four N-terminal transmembrane helices are localized in the same lipid bilayer (Fig. 6a) and one end of the tetramerization channel is localized close to the membrane surface and the other is open to the lumen of the ER. In the second model (Fig. 6b), two transmembrane helices penetrate separate bilayers in the highly invaginated ER lumen and both ends of the C-terminal tetramerization channel are localized to the membrane surface. Notably, the flexible N-terminal linker region not present in our structures should allow for significant flexibility of the catalytic core relative to the N-terminal transmembrane supporting both modes of membrane attachment. The flexible N terminus would further support a model of membrane attachment in which discreet dimers interact with the lipid bilayer through interactions with the two hydrophobic 11{beta}-HSD1 C termini (Fig. 6). In all our models of membrane attachment Trp263 (Figs. 3b and 5b) is immediately adjacent to the lipid bilayer, consistent with the known preference of tryptophan for residing in the interfacial region of zwitterionic phospholipids membrane (49, 50). A notable feature of the tetramer models is the close proximity of the ends of the 11{beta}-HSD1 tetramer to the ER membrane. This close localization could allow membrane-localized substrates to directly diffuse into the hydrophobic channel observed in the open structure (Fig. 3, a and b) and then to one of the four connected active sites.



View larger version (58K):
[in this window]
[in a new window]
  		FIG. 6.
Structure-based model for membrane interactions and subunit assembly in the ER lumen. a, schematic illustrating interactions with a single membrane within the ER lumen. Four N-terminal transmembrane helices (orange) anchor the interface-closed and interface-open tetramers (blue subunits, black C-terminal helices) to the membrane. One end of the tetramerization interface faces the membrane and one the interior of the ER lumen. The hydrophobic channel observed in the interface open structure would facilitate substrate diffusion directly from the membrane to the active sites. Trp263 and Tyr280 are represented as purple and yellow hexagons, respectively. In our dimer model for membrane association, the enzyme is anchored to the bilayer by its N-terminal transmembrane helices as well as through hydrophobic residues on the enzyme C terminus. b, schematic illustrating interactions with two closely spaced membranes within the ER lumen. The highly invaginated membrane structure of this cellular compartment suggest such interactions may be relevant in cells.

 
In animals, it is noteworthy that Cys272 is found only in higher primates. In many other species this Cys is replaced with glycine or serine. When considered in the structural context presented here, this genetic change argues that primates may have evolved an additional regulatory mechanism to control 11{beta}-HSD1 activity through formation of enzyme disulfides. It will be interesting to determine if this primate-specific change was introduced to enable the enzyme to directly sense the redox potential of the ER and then modulate its activity through conformational change. Regardless of the precise molecular mechanisms that regulate enzyme activity in cells, the close localization of four Pro-Cys motifs at the center of the enzyme tetramer, the structurally defined C-terminal and {alpha}6-{beta}6 conformational changes, the location of the protein kinase phosphorylation site, and the C-terminal tail interactions that physically connect neighboring monomers, all support the dynamic assembly of enzyme subunits in the cellular functions of 11{beta}-HSD1.


   FOOTNOTES
 
* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and structure factors (codes 1XU7 and 1XU9) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/). Back

{ddagger} To whom correspondence should be addressed: Dept. of Structural Chemistry, Syrrx Inc. 10410 Science Center Drive, San Diego, CA 92121. Tel.: 858-731-3574; E-mail: david.hosfield{at}syrrx.com.

1 The abbreviations used are: 11{beta}-HSD, 11{beta}-hydroxysteroid dehydrogenase type I; SDR, short-chain dehydrogenase/reductase; CHAPS, 3-[(3-cholamidopropyl)-dimethylammonio]-1-propane sulfonate; MES, 4-morpholineethanesulfonic acid; RMS, root mean-squared; ER, endoplasmic reticulum. Back


   ACKNOWLEDGMENTS
 
We thank D. Dougan and L. Tari and for assistance with data collection, D. Scheibe for Nanovolume CrystallizationTM experiments, E. Chien, K. Lim, O. Brodsky, E. Sims, and E. Chi for assistance with protein purification and mass spectrometry, and E. D. Garcin, C. Grimshaw, C. D. Mol, and C. M. Hosfield for careful reading of the manuscript. We also thank the Advanced Light Source, which is supported by the Director, Office of Science, Office of Basic Energy Sciences, Materials Sciences Division of the United States Department of Energy under Contract No. DE-AC03-76SF00098 at Lawrence Berkeley National Laboratory.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Lemonick, M. D. (2004) Time 163, 57–71
  2. Zimmet, P., Alberti, K. G., and Shaw, J. (2001) Nature 414, 782–787[CrossRef][Medline] [Order article via Infotrieve]
  3. Engeli, S., Bohnke, J., Feldpausch, M., Gorzelniak, K., Heintze, U., Janke, J., Luft, F. C., and Sharma, A. M. (2004) Obes. Res. 12, 9–17[Medline] [Order article via Infotrieve]
  4. Seckl, J. R., Morton, N. M., Chapman, K. E., and Walker, B. R. (2004) Recent Prog. Horm. Res. 59, 359–393[Abstract/Free Full Text]
  5. Wake, D. J., and Walker, B. R. (2004) Mol. Cell. Endocrinol. 215, 45–54[CrossRef][Medline] [Order article via Infotrieve]
  6. Walker, B. R. (2004) Obes. Res. 12, 1–3[Medline] [Order article via Infotrieve]
  7. Stewart, P. M., and Krozowski, Z. S. (1999) Vitam. Horm. 57, 249–324[Medline] [Order article via Infotrieve]
  8. Walker, E. A., and Stewart, P. M. (2003) Trends Endocrinol. Metab. 14, 334–339[CrossRef][Medline] [Order article via Infotrieve]
  9. Jamieson, P. M., Chapman, K. E., Edwards, C. R., and Seckl, J. R. (1995) Endocrinology 136, 4754–4761[Abstract]
  10. Seckl, J. R., and Walker, B. R. (2001) Endocrinology 142, 1371–1376[Abstract/Free Full Text]
  11. Agarwal, A. K., Mune, T., Monder, C., and White, P. C. (1994) J. Biol. Chem. 269, 25959–25962[Abstract/Free Full Text]
  12. Yang, K., Smith, C. L., Dales, D., Hammond, G. L., and Challis, J. R. (1992) Endocrinology 131, 2120–2126[Abstract]
  13. Rusvai, E., and Naray-Fejes-Toth, A. (1993) J. Biol. Chem. 268, 10717–10720[Abstract/Free Full Text]
  14. Oppermann, U., Filling, C., Hult, M., Shafqat, N., Wu, X., Lindh, M., Shafqat, J., Nordling, E., Kallberg, Y., Persson, B., and Jornvall, H. (2003) Chem. Biol. Interact 143–144, 247–253
  15. Kallberg, Y., Oppermann, U., Jornvall, H., and Persson, B. (2002) Protein Sci. 11, 636–641[Abstract/Free Full Text]
  16. Kallberg, Y., Oppermann, U., Jornvall, H., and Persson, B. (2002) Eur. J. Biochem. 269, 4409–4417[Medline] [Order article via Infotrieve]
  17. Persson, B., Kallberg, Y., Oppermann, U., and Jornvall, H. (2003) Chem. Biol. Interact. 143–144, 271–278
  18. Filling, C., Berndt, K. D., Benach, J., Knapp, S., Prozorovski, T., Nordling, E., Ladenstein, R., Jornvall, H., and Oppermann, U. (2002) J. Biol. Chem. 277, 25677–25684[Abstract/Free Full Text]
  19. Powell, A. J., Read, J. A., Banfield, M. J., Gunn-Moore, F., Yan, S. D., Lustbader, J., Stern, A. R., Stern, D. M., and Brady, R. L. (2000) J. Mol. Biol. 303, 311–327[CrossRef][Medline] [Order article via Infotrieve]
  20. Sawicki, M. W., Erman, M., Puranen, T., Vihko, P., and Ghosh, D. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 840–845[Abstract/Free Full Text]
  21. Grimm, C., Maser, E., Mobus, E., Klebe, G., Reuter, K., and Ficner, R. (2000) J. Biol. Chem. 275, 41333–41339[Abstract/Free Full Text]
  22. Tanaka, N., Nonaka, T., Tanabe, T., Yoshimoto, T., Tsuru, D., and Mitsui, Y. (1996) Biochemistry 35, 7715–7730[CrossRef][Medline] [Order article via Infotrieve]
  23. Price, A. C., Zhang, Y. M., Rock, C. O., and White, S. W. (2001) Biochemistry 40, 12772–12781[CrossRef][Medline] [Order article via Infotrieve]
  24. Odermatt, A., Arnold, P., Stauffer, A., Frey, B. M., and Frey, F. J. (1999) J. Biol. Chem. 274, 28762–28770[Abstract/Free Full Text]
  25. Ozols, J. (1995) J. Biol. Chem. 270, 2305–2312[Abstract/Free Full Text]
  26. Draper, N., Walker, E. A., Bujalska, I. J., Tomlinson, J. W., Chalder, S. M., Arlt, W., Lavery, G. G., Bedendo, O., Ray, D. W., Laing, I., Malunowicz, E., White, P. C., Hewison, M., Mason, P. J., Connell, J. M., Shackleton, C. H., and Stewart, P. M. (2003) Nat. Genet. 34, 434–439[CrossRef][Medline] [Order article via Infotrieve]
  27. Walker, E. A., Clark, A. M., Hewison, M., Ride, J. P., and Stewart, P. M. (2001) J. Biol. Chem. 276, 21343–21350[Abstract/Free Full Text]
  28. Jamieson, P. M., Walker, B. R., Chapman, K. E., Andrew, R., Rossiter, S., and Seckl, J. R. (2000) J. Endocrinol. 165, 685–692[Abstract]
  29. Masuzaki, H., Paterson, J., Shinyama, H., Morton, N. M., Mullins, J. J., Seckl, J. R., and Flier, J. S. (2001) Science 294, 2166–2170[Abstract/Free Full Text]
  30. Morton, N. M., Paterson, J. M., Masuzaki, H., Holmes, M. C., Staels, B., Fievet, C., Walker, B. R., Flier, J. S., Mullins, J. J., and Seckl, J. R. (2004) Diabetes 53, 931–938[Abstract/Free Full Text]
  31. Morton, N. M., Holmes, M. C., Fievet, C., Staels, B., Tailleux, A., Mullins, J. J., and Seckl, J. R. (2001) J. Biol. Chem. 276, 41293–41300[Abstract/Free Full Text]
  32. Kotelevtsev, Y., Holmes, M. C., Burchell, A., Houston, P. M., Schmoll, D., Jamieson, P., Best, R., Brown, R., Edwards, C. R., Seckl, J. R., and Mullins, J. J. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 14924–14929[Abstract/Free Full Text]
  33. Alberts, P., Nilsson, C., Selen, G., Engblom, L. O., Edling, N. H., Norling, S., Klingstrom, G., Larsson, C., Forsgren, M., Ashkzari, M., Nilsson, C. E., Fiedler, M., Bergqvist, E., Ohman, B., Bjorkstrand, E., and Abrahmsen, L. B. (2003) Endocrinology 144, 4755–4762[Abstract/Free Full Text]
  34. Barf, T., Vallgarda, J., Emond, R., Haggstrom, C., Kurz, G., Nygren, A., Larwood, V., Mosialou, E., Axelsson, K., Olsson, R., Engblom, L., Edling, N., Ronquist-Nii, Y., Ohman, B., Alberts, P., and Abrahmsen, L. (2002) J. Med. Chem. 45, 3813–3815[CrossRef][Medline] [Order article via Infotrieve]
  35. Yau, J. L., Noble, J., Kenyon, C. J., Hibberd, C., Kotelevtsev, Y., Mullins, J. J., and Seckl, J. R. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 4716–4721[Abstract/Free Full Text]
  36. Sandeep, T. C., Yau, J. L., MacLullich, A. M., Noble, J., Deary, I. J., Walker, B. R., and Seckl, J. R. (2004) Proc. Natl. Acad. Sci. U. S. A.
  37. Hosfield, D., Palan, J., Hilgers, M., Scheibe, D., McRee, D. E., and Stevens, R. C. (2003) J. Struct. Biol. 142, 207–217[CrossRef][Medline] [Order article via Infotrieve]
  38. Otwinowski, Z., and Minor, W. (1997) Met. Enzymol. 276, 307–326
  39. Schneider, T. R., and Sheldrick, G. M. (2002) Acta Crystallogr. D Biol. Crystallogr. 58, 1772–1779[CrossRef][Medline] [Order article via Infotrieve]
  40. La Fortelle, E., and Bricogne, G. (1997) Met. Enzymol. 276, 472
  41. McRee, D. E. (1999) J. Struct. Biol. 125, 156–165[CrossRef][Medline] [Order article via Infotrieve]
  42. Winn, M. D., Murshudov, G. N., and Papiz, M. Z. (2003) Methods Enzymol. 374, 300–321[Medline] [Order article via Infotrieve]
  43. Morris, R. J., Perrakis, A., and Lamzin, V. S. (2003) Methods Enzymol. 374, 229–244[Medline] [Order article via Infotrieve]
  44. Alfaidy, N., Xiong, Z. G., Myatt, L., Lye, S. J., MacDonald, J. F., and Challis, J. R. (2001) J. Clin. Endocrinol. Metab. 86, 5585–5592[Abstract/Free Full Text]
  45. Halgren, T. A. (1992) J. Am. Chem. Soc 114, 7827–7843[CrossRef]
  46. Shi, T., Spain, S. M., and Rabenstein, D. L. (2004) J. Am. Chem. Soc. 126, 790–796[Medline] [Order article via Infotrieve]
  47. Bush, K. T., Hendrickson, B. A., and Nigam, S. K. (1994) Biochem. J. 303, 705–708
  48. Nardai, G., Korcsmaros, T., Papp, E., and Csermely, P. (2003) Biofactors 17, 259–267[Medline] [Order article via Infotrieve]