Critical Residues for Structure and Catalysis in Short-chainDehydrogenases/Reductases*
Received for publication, March 5, 2002, and in revised form, April 23, 2002Published, JBC Papers in Press, April 25, 2002, DOI 10.1074/jbc.M202160200
Charlotta Filling‡, Kurt D. Berndt§¶, Jordi Benach§�, Stefan Knapp§**, Tim Prozorovski‡,Erik Nordling‡ ‡‡, Rudolf Ladenstein§, Hans Jornvall‡, and Udo Oppermann‡§§
From the ‡Department of Medical Biochemistry and Biophysics, Karolinska Institutet, SE-171 77 Stockholm, Sweden,§Center for Structural Biochemistry, NOVUM, Karolinska Institutet, SE-141 57 Huddinge, Sweden, ¶Department ofNatural Sciences, Sodertorns Hogskola, Box 4101, SE-141 04 Huddinge, Sweden, and ‡‡Stockholm Bioinformatics Center,Karolinska Institutet, SE-171 77 Stockholm, Sweden
Short-chain dehydrogenases/reductases form a large,evolutionarily old family of NAD(P)(H)-dependent en-zymes with over 60 genes found in the human genome.Despite low levels of sequence identity (often 10–30%),the three-dimensional structures display a highly simi-lar �/� folding pattern. We have analyzed the role ofseveral conserved residues regarding folding, stability,steady-state kinetics, and coenzyme binding using bac-terial 3�/17�-hydroxysteroid dehydrogenase and se-lected mutants. Structure determination of the wild-type enzyme at 1.2-Å resolution by x-ray crystallographyand docking analysis was used to interpret the biochem-ical data. Enzyme kinetic data from mutagenetic re-placements emphasize the critical role of residues Thr-12, Asp-60, Asn-86, Asn-87, and Ala-88 in coenzymebinding and catalysis. The data also demonstrate essen-tial interactions of Asn-111 with active site residues. Ageneral role of its side chain interactions for mainte-nance of the active site configuration to build up a pro-ton relay system is proposed. This extends the previ-ously recognized catalytic triad of Ser-Tyr-Lys residuesto form a tetrad of Asn-Ser-Tyr-Lys in the majority ofcharacterized short-chain dehydrogenases/reductaseenzymes.
Since the discovery of fundamental differences between in-sect-type and liver-type alcohol dehydrogenases (1), corre-sponding to the protein families of “short-chain” dehydroge-nases/reductases (SDR)1 and “medium-chain” dehydrogenases/reductases (MDR), respectively, SDR enzymes have receivedmuch attention. They constitute a large protein family withwell over 2000 annotated enzyme and species variant se-quences in databases and are represented in all life forms with
minimally 60 genes found in the human genome (2–4). TheSDR enzymes span several EC classes, from oxidoreductasesand lyases to isomerases, with NAD(P)(H)-dependent oxi-doreductases constituting the majority of forms. In this class,many enzymes with different specificities act on steroids, pros-taglandins, aliphatic alcohols, and xenobiotics.
The pairwise sequence identity between different enzymes islow, typically 10–30%, but all available three-dimensionalstructures (�20) display a highly similar �/� folding pattern(5–11). Most SDR enzymes have a 250–350-residue core struc-ture, frequently with additional N- or C-terminal transmem-brane domains or signal peptides. Conserved sequence regionscover a variable N-terminal Gly-X3-Gly-X-Gly motif as part ofthe nucleotide binding region and the active site with a triad ofcatalytically important Ser, Tyr, and Lys residues, of which Tyris the most conserved residue within the whole family (2, 8, 12)(Fig. 1). The functions of the residues at these particular siteshave been elucidated by a combination of chemical modifica-tions, sequence comparisons, structure analyses, and site-di-rected replacements (2, 13). However, several other conservedbut still variable residues have not been analyzed in detail. Tounderstand their role, we now carried through a mutagenesisstudy using bacterial 3�/17� hydroxysteroid dehydrogenase(3�/17�-HSD) analyzing effects on enzymatic function and thestability of replacements, assisted by x-ray crystallography anddocking analysis. Other residues have been studied before (13),but novel segments investigated include Asp-60 (between �Cand �D), an NNAG motif (Asn-86–Gly-89 in �D), and Asn-111in �E (Fig. 1).
MATERIALS AND METHODS
Molecular Cloning, Site-directed Mutagenetic Replacements, and Pu-rification of Wild Type and Mutants of 3�/17�-HSD—Molecular clon-ing of 3�/17�-HSD (EC 1.1.1.51) and purification of recombinant pro-teins from Comamonas testosteroni ATCC 11996 (DSM, DeutscheSammlung fur Mikroorganismen, Braunschweig, Germany) was car-ried out as described (13) using metal-chelate chromatography of His-tagged enzymes. Mutagenetic replacements were performed using themegaprime method (14) with Pfu polymerase (Stratagene). Wild-typesequences used for amplification were: 5�-GGCAGCCATATGACAAAT-CGTTTGCAGG (sense) and 5�-CTAGGGATCCCTATAGCCCCATGCC-CAGAAT (antisense). Mutagenesis primers covering the desired re-placement region were as follows (5�33�): CCTCGCTGCTCACGGCA-TGGCGGAC (D60A), GCCGGCATTGGCGACCAGCACATTG (N86A),CAGCAGGATGCCGGAATTGTTGACCAG (A88S), GAAGACTGACTC-GGTGAGGATCTTGAGCAGG (N111L), GCTGGCGCTAAAGCCGGC-GTATTG (Y151F). The other mutants used were prepared as described(13). All constructs were verified by sequence analysis using ThermoSequenase (Amersham Biosciences) or BigDye Terminator Cycle Se-quencing (PE Applied Biosystems).
SDS-PAGE and Protein Determination—Purity of protein sampleswas assessed by SDS-PAGE on 10% gels. Protein concentration was
* This study was supported by Grant BIO 4 EC 97-2123 from theEuropean Community and by grants from the NOVO Nordisk Founda-tion, Swedish Union of Physicians, Swedish Medical Research Council,and Karolinska Institutet. The costs of publication of this article weredefrayed in part by the payment of page charges. This article musttherefore be hereby marked “advertisement” in accordance with 18U.S.C. Section 1734 solely to indicate this fact.
� Present address: Department of Biological Sciences, Columbia Uni-versity, 702 Fairchild Center, New York, NY 10027.
** Present address: Pharmacia Corporation, Discovery Research On-cology, Dept. of Chemistry, 20014 Nerviano, Italy.
§§ To whom correspondence should be addressed. Fax: 46-8-33-74-62;E-mail: [email protected].
determined by amino acid analysis of pure samples after hydrolysis in6 M HCl followed by ninhydrin-based quantification on an LKB Alphaplus analyzer.
Determination of Kinetic and Binding Constants—Enzyme activitieswere measured as NAD(H)-dependent 3�- and 17�-oxidoreductase ac-tivities by determination of the change of absorbance at 340 nm andusing a molar extinction coefficient for NADH of 6.22 mM�1 cm�1.Recordings were carried out with a Cary 300Bio instrument. Reactionswere performed in 1.0-ml volumes at 298 K. Conditions for dehydro-genase activities were: 20 mM Tris/HCl (pH 8.5) and 250 �M NAD�,varying the amount of iso-ursodeoxycholic acid (3�,12�-dihydroxy-5�-cholan-24-oic acid, isoUDCA) or dehydroepiandrosterone (5-androsten-3�-ol, 17-one, DHEA; 3�-HSD activity) and testosterone (4-androsten-17�ol-3-one; 17�-HSD activity). Conditions for reductase activities
were: 20 mM Tris/HCl (pH 7.0) and 250 �M NADH, varying the amountof 5�-dihydrotestosterone (5�-androstane, 17�-ol-3-one; 3-ketoreduc-tase activity) and androsterone (5�-androstane-3�-ol-17-one; 17-ketore-ductase activity). Kinetic constants were calculated with the EnzPacksoftware (Biosoft). Binding of NADH was determined by monitoringfluorescence energy transfer as a function of nucleotide concentration in20 mM Tris-HCl, pH 7.0, at 298 K using a Shimadzu RF5000 spec-trofluorimeter. Binding of NAD� was assessed in a similar manner bydisplacement of bound NADH in 20 mM Tris/Cl, pH 8.5. Dissociationconstants were obtained by non-linear regression using a 1:1 bindingmodel as described previously (12).
Structure Analysis of 3�/17�-HSD and Comparison with SDRMembers—The structure of the wild-type 3�/17�-HSD apoform wasdetermined by x-ray crystallography to a final resolution of 1.2 Å
FIG. 1. Sequence alignment of SDR enzymes based upon determined three-dimensional structures. The secondary structure elementsof 1hxh are indicated as arrows for �-strands and as cylinders for �-helices. Mutated residues are indicated by black boxes and black downwardarrows. Fabg protein members included in Table III (Protein Data Bank accession numbers 1bvr, 1qsg, and 1d70) were omitted from the alignmentfor clarity. Abbreviations and Protein Data Bank accession numbers are as follows: 1dhr: dihydropteridine reductase; 1a27: Homo sapiens17-�-hydroxysteroid-dehydrogenase; 1e3w: rattus norvegicus short-chain 3-hydroxyacyl-CoA dehydrogenase; 1doh: Magnaporthe grisea trihy-droxynaphthalene reductase; 1geg: Klebsiella pneumoniae acetoin reductase; 1edo: Brassica napus �-keto acyl carrier protein reductase; 1gco:Bacillus megaterium glucose dehydrogenase; 1hdc: Streptomyces hydrogenans 3�,20�-hydroxysteroid dehydrogenase; 1hxh: C. testosteroni 3�/17�-hydroxysteroid dehydrogenase; 1cyd: Mus musculus carbonyl reductase; 1ae1: Datura stramonium tropinone reductase-I; 2ae1: D. stramoniumtropinone reductase-ii; 1bdb: Pseudomonas spec. cis-biphenyl-2,3-dihydrodiol-2,3-dehydrogenase; 1a4u: Drosophila alcohol dehydrogenase; 1nas:m. musculus sepiapterin reductase; 1fjh: C. testosteroni 3�-hydroxysteroid dehydrogenase/carbonyl reductase
Structural and Mechanistic Principles of SDR Enzymes25678
at Karolinska institutet library on February 23, 2015
(Protein Data Bank accession number 1hxh). Crystallization, x-rayanalysis, model building, and refinement have been described indetail elsewhere (15).2 The mean coordinate error for the 1.2-Å struc-ture is 0.06, calculated by block-matrix inversion in SHELXL (16).The structures of known SDR members were retrieved by a FASTAsearch versus PDB. The structures were superimposed, and a struc-tural alignment was created using ICM (Molsoft LLC, San Diego, CA)and edited in BioEdit.
Biophysical Characterization of 3�/17�-HSD and Mutants—Analyt-ical ultracentrifugation was performed using a Beckman XL-1 ultra-centrifuge. For sedimentation equilibrium and velocity studies, bothabsorbance at 280 nm and interference were recorded at different rotorspeeds and protein concentrations at 20 °C. Sodium phosphate buffer(30 mM, 150 mM NaCl, pH 7.5) was used. Data evaluation was per-formed using the Origin 4.0 software package. Partial specific volumesof wild-type and mutant proteins and buffer densities were calculatedusing the program Sednterp. Double sector cells were used for sedimen-tation velocity studies. Interference scans were recorded in intervals of2 min.
Assessment of wild-type and mutant conformations was achieved bycircular dichroism spectroscopy by recording the ellipticity as a functionof wavelength between 260 and 195 nm using an AVIV Model62 DSspectropolarimeter. Conformational stability was determined by titra-tion of the individual proteins with guanidine-HCl using a titrationrobot and by monitoring CD at 222 nm as described earlier (12, 13).
RESULTS AND DISCUSSION
Overexpression, Purification, Folding, and Stability Analysisof Wild Type and Mutants of 3�/17�-HSD—Wild-type and mu-tant 3�/17�-HSD forms were overexpressed in Escherichia colistrain BL21 and purified by metal-chelate chromatography(Fig. 2). Proteins were analyzed regarding folding, stability,coenzyme binding, and steady-state kinetics. Judged from CDspectroscopy, wild-type and mutant forms displayed essentiallyidentical spectral characteristics, indicating similar secondarystructure properties. Stability measurements were performedby titration with guanidinium hydrochloride, monitored by CDspectroscopy and by determination of the transition tempera-tures using differential scanning calorimetry. Judging fromthese experiments, no significant differences between wild-typeand mutant proteins could be detected, indicating similar foldand conformational stability (data not shown).
Determination of Kinetic and Binding Constants—Steady-state kinetics were determined with different substrates foroxo-reductase and �-dehydrogenase reactions at C3 and C17 ofthe steroid. The kinetic and coenzyme binding data obtainedare summarized in Tables I and II. Bacterial 3�/17�-HSD isable to catalyze specific dehydrogenations/reductions at posi-tions 3 and 17 of steroids with different conformations, i.e.being trans- or cis-configured between ring A and B. First orderconstants kcat/Km for dehydrogenase reactions are an order ofmagnitude higher (1.7 � 106 s�1 M�1 for 17�-HSD activity,1.0 � 106 s�1 M�1 for 3�-HSD activity with isoUDCA) as com-
2 J. Benach, C. Filling, U. Oppermann, P. Roversi, G. Bricogve, K. D.Berndt, H. Jornvall, and R. Ladenstein, submitted for publication.
FIG. 2. SDS-PAGE of wild-type and mutant proteins. Purity wasanalyzed after metal-chelate chromatography and gel filtration. 700 ngof sample was loaded on lanes 1–11, Coomassie staining. Lane M:molecular mass standard (See Blue plus 2, NOVEX); lane 1: wild type;lane 2: T12A; lane 3: T12S; lane 4: D60A; lane 5: N86A; lane 6: N87A;lane 7: A88S; lane 8: N111L; lane 9: S138A; lane 10: S138T; lane 11:Y151F.
TA
BL
EI
Kin
etic
con
stan
tsfo
r3 �
/17
�-H
SD
wil
d-t
ype
and
mu
tan
tsK
mva
lues
are
in10
�6
M,k
cat
valu
esin
103
min
�1,k
cat/K
mva
lues
are
in10
6s�
1M
�1,a
nd
rela
tive
k ca
t/Km
valu
esar
ein
%of
wil
d-ty
peac
tivi
ties
.Nu
mbe
rof
expe
rim
ents
,n�
3–5.
Det
ecti
onli
mit
sfo
rre
lati
vek c
at/K
mva
lues
are
0.00
03%
for
deh
ydro
gen
ase
acti
viti
esan
d0.
003%
for
redu
ctas
eac
tivi
ties
.
3�-H
SD
isou
rsod
eoxy
chol
icac
id3�
-HS
Dde
hyd
roep
ian
dros
tero
ne
3-ke
tore
duct
ase
5�-d
ihyd
rote
stos
tero
ne
17�
-HS
Dte
stos
tero
ne
17-k
etor
edu
ctas
ean
dros
tero
ne
Km
k ca
tk c
at/K
mre
lati
vek c
at/K
mK
mk c
at
k ca
t/Km
rela
tive
k ca
t/Km
Km
k ca
tk c
at/K
mre
lati
vek c
at/K
mK
mk c
at
k ca
t/Km
rela
tive
k ca
t/Km
Km
k ca
tk c
at/K
mre
lati
vek c
at/K
m
Wil
d-ty
pea
22.3
1.4
1.0
100
8.17
0.62
1.3
100
16.1
0.13
0.13
100
11.8
1.2
1.7
100
31.7
0.21
0.11
100
T12
Aa
NA
b32
10.
039
0.00
20.
1515
.80.
100.
1178
NA
bN
Ab
T12
Sa
28.4
1.8
1.1
110
17.7
0.74
0.70
5413
.90.
062
0.07
455
15.1
1.2
1.6
7826
.90.
130.
081
73D
60A
47.3
0.47
0.16
1612
.60.
490.
6550
35.5
0.12
0.05
442
23.9
0.25
0.18
1038
.50.
180.
078
71N
86A
125
0.37
0.04
94.
710
.90.
064
0.09
87.
517
.30.
084
0.08
162
19.5
0.45
0.39
2336
.60.
060.
027
25N
87A
a28
.80.
360.
2120
24.4
0.42
0.29
2218
.60.
092
0.08
261
26.9
0.46
0.29
1724
.60.
120.
081
74A
88S
116
0.17
0.02
42.
319
.70.
300.
2519
27.4
0.24
0.15
120
68.1
0.18
0.04
42.
631
.70.
150.
079
72N
111L
NA
bN
Ab
NA
bN
Ab
NA
b
S13
8Aa
NA
bN
Ab
NA
bN
Ab
NA
b
S13
8Ta
18.8
1.5
1.3
130
ND
c14
.20.
062
0.07
354
15.1
1.1
1.2
7225
.10.
097
0.06
458
Y15
1FN
Ab
NA
bN
Ab
NA
bN
Ab
aC
onst
ants
for
3�-H
SD
(wit
his
ours
odeo
xych
olic
acid
assu
bstr
ate)
and
3-ke
tore
duct
ase
acti
viti
esar
eta
ken
from
Ref
.13
.b
NA
,n
otac
tive
.c
ND
,n
otde
term
ined
.
Structural and Mechanistic Principles of SDR Enzymes 25679
at Karolinska institutet library on February 23, 2015
pared with the reductive reactions (0.13 � 106 s�1 M�1, 3-oxo-reductase, 0.11 � 106 s�1 M�1, 17-oxoreductase).
The mutants compared for dehydrogenase and reductaseactivities have residue exchanges at positions Thr-12, Asp-60,Asn-86, Asn-87, Ala-88, Asn-111, Ser-138, and Tyr-151, sur-rounding the coenzyme binding region or located at the activesite. Depending on the amino acid substitution, differentialeffects on enzymatic constants were observed. Completely in-active enzymes were obtained with the N111L, S138A, andY151F mutants, whereas partially active enzymes with sig-nificant changes were observed for the other substitutions(Table I).
Replacement of Thr-12 by Ala results in an enzyme that is
only able to catalyze the reductive reaction at position C3 andthe corresponding dehydrogenase activity with DHEA as sub-strate. All other activities are not detectable. Exchange to Serresults in an enzyme with enzymatic properties largely similarto those of the wild-type form (Table I). A profound change incofactor binding (Table II) is observed with the T12A mutant;the KD for NADH is altered from 1.0 to 41 �M, whereas the KD
for NAD� changes from 1.5 �M for the wild type to 0.1 �M forT12A. This altered binding is reflected in a lowered Km valuefor NAD� (8.96 versus 29.6 �M for wild type) and an increasedKm value for NADH (36.7 versus 21.6 �M for wild type).
Mutants D60A, N86A, N87A, and A88S display reduced cat-alytic efficiencies over the wild type with the exception of theA88S mutant, which shows a 20% higher relative kcat/Km value
FIG. 3. Details of the central �-sheet (displaying strands �A,�D, and �E) of 3�/17�-HSD in relation to the NAD� molecule.Residues analyzed in this study include Thr-12, Asn-86, Asn-87, andAla-88. Hydrogen bond distances (in Å) between side chain and back-bone atoms are given as dotted lines. The figure was created withRIBBONS (26).
FIG. 4. Close-up view of active-site residues in 3�/17�-HSDdetermined by x-ray crystallography in relation to modeledNAD� (red) and steroid substrate (3�-hydroxy-5-androsten-17-one, blue) molecules. Tyr-151 is in catalytic distance to the substratehydroxyl (2.7 Å), further stabilized by Ser-138 interaction to the sub-strate hydroxyl (2.6 Å). A network of residues and interactions compris-ing Lys-155, Ser-154, Asn-111, and Ile-90 via a conserved water mole-cule is shown. For clarity, NAD� contacts are omitted (cf. “Results andDiscussion”). The figure was created with RIBBONS (26).
TABLE IIKm and KD values for NADH and NAD� with 3�/17�-HSD wild-type and mutant forms
for 3-oxoreductase activity. The drop in kcat/Km values is morepronounced for dehydrogenase activities (ranging from 2.3 to50% versus wild type) than for reductase values (ranging from25 to 120% versus wild type). The relative change in the ratio ofdehydrogenase/reductase reaction for the different comparablesubstrate pairs (i.e. similar ring configurations and substitu-ents) at positions C3 (DHEA/5�-dihydrotestosterone) and C17(testosterone/androsterone) ranges from about 3-fold (N87A,position C3) to over 25-fold for the A88S form. Here the relativekcat/Km for reactions at C17 is shifted from 72% for the reduc-tase reaction to 2.6% for dehydrogenase activity as comparedwith the corresponding wild-type values. Significant changes inkcat/Km values are also observed with the 5�-reduced steroidisoUDCA as substrate for 3�-HSD activity, supporting theconclusion that Asn-86, Asn-87, and Ala-88 mutants signifi-cantly attenuate oxidative kinetic constants and show a lesspronounced effect in the reductive reactions catalyzed.
NAD� binding constants for the N86A and N87A mutantsare in the same range as the wild-type value (1.5 and 1.1 �M,respectively, versus 1.5 �M). Slightly lowered values are ob-served for the D60A and A88S forms (0.3 �M). Depending on thesubstrate for determination of 3�-HSD activity (isoUDCA orDHEA), significant increases (up to 1550 �M, A88S, isoUDCAsubstrate) in Km for NAD� are observed except for N87A (81.2versus 83.2, isoUDCA substrate). Weaker NADH binding isobserved for Asp-60, Asn-86, Asn-87, and Ala-88 mutants,ranging from 3.1 to 19 �M, which is accompanied by a moderateincrease in Km for NADH (range from 25.5 to 52.3 �M) exceptfor the N86A mutant displaying a Km of 7.5.
Taken together, the drastic reduction in kcat/Km values forthe oxidative reaction of the Asp-60, Asn-86, Asn-87, andAla-88 forms is mainly due to increased Km for NAD� withoutsignificant decrease in binding affinities, indicating inhibitionof substrate formation without affecting association of the en-zyme-substrate complex. The kcat/Km values for the reductivereactions are slightly lowered, accompanied by a decrease inNADH binding. Steroid structure is a further contributing fac-tor to the differential decrease in oxidation since differencesbetween isoUDCA and DHEA are observed.
Determination of Kinetic and Binding Constants of ActiveSite Mutants—Mutations performed within or close to the ac-tive site were residues Ser-138 and Tyr-151, recognized previ-ously to be part of a catalytic triad with Lys-155, and theconserved Asn-111, located within helix �E, which forms themain subunit interaction surface. Mutants N111L, S138A, andY151F are enzymatically inactive. Coenzyme binding constantsof these mutants are changed but within the range observedwith the mutants described above (from 0.2 �M for NAD� for
Y151F to 12 �M for NADH for S138A). As described earlier (13),substitution of Ser for Thr at position 138 yields an activeenzyme. Loss of activity observed for S138A and Y151F mu-tants thus underscores the critical role of these residues incatalysis.
Crystallographic Analysis of 3�/17�-HSD and Comparisonwith High Resolution SDR Structures—The apo structure ofwild-type 3�/17�-HSD was determined by x-ray crystallogra-phy to a resolution of 1.2 Å.2 The enzyme forms a tetramer withsubunit interactions similar to those of tetrameric SDR struc-tures (9, 10, 17). The geometry of coenzyme binding and activesites was sufficiently similar to allow docking analysis of a3�/17�-HSD ternary complex with NAD� and a steroid sub-strate (3�-OH-5-androsten-17-one) (Figs. 3 and 4).
Details of side chain and backbone interactions of residuesThr-12, Asp-60, Asn-86, Asn-87, and Ala-88 within the central�-sheet formed by strands �A, �D, and �E are given in Fig. 3and illustrate the essential nature of these residues for correctcoenzyme positioning and binding (Tables I and II). No directinteractions of Thr-12, Asn-86, and Asn-87 with the coenzymeare observed in our model and in most other coenzyme com-plexes (9, 10, 17–19). An exception is the recently determinedporcine carbonyl reductase, where backbone carbonyl interac-tions of Asn-89 to the 3�-OH of the nicotinamide ribose arefound (20). Based on structural and kinetic data, we concludethat these residues (Thr-12, Asn-86, Asn-87) supply a frame-work essential for keeping the strands oriented within thecentral �-sheet, important for coenzyme positioning (Fig. 3).Multiple and critical side chain and backbone interactions areformed through Thr-12 to strand �D (ND2Asn-87-OThr-12:3.05 Å; OG1Thr-12-NAsn-87: 3.03 Å; OG1Thr-12-NAla-88:2.98Å) and between strands �D and �E through Asn-86 andAsn-87 (OAsn-86-NMet-136: 2.93Å; NAsn-86-OIle-134: 2.83Å;OD1Asn-135-ND2Asn-86: 2.87Å).
The coenzyme is bound through few specific contacts, per-formed through other residues than Thr-12, Asn-86, and withindicated exceptions, Asn-87. Among the residues investigated,Asp-60 and Ala-88 contribute directly to coenzyme binding, aview supported by the significantly altered coenzyme con-stants. In the modeled structure, the carboxyl side chainof Asp-60 is in weak H-bonding distance to the adenine ring(OD1 Asp-60-N6A 3.99 Å) (Fig. 2), and similar interactionsbetween Asp-60 and coenzyme are observed in, for example,MLCR, 7�-HSD, or CR (9, 10, 20). Side chain to backboneinteractions are also observed (OD2Asp-60-OGSer-62: 2.7 Å),and the main role of Asp-60 appears to be in the stabilization ofthe turn between �C and �D as part of the adenine ring bindingpocket (10). Ala-88 can make hydrophobic contacts to the ade-
FIG. 5. Postulated reductive reac-tion mechanism of 3�/17�-HSD in-volving NADH and steroid substrate(5�-androstane, 3-one, 17ol). Catalysisis initiated by proton transfer from Tyr-151 hydroxyl to the substrate carbonylfollowed by hydrid transfer to C3 of thesteroid. A proton relay is formed and in-volves the 2�OH of the ribose, the Lys-155side chain, and a water molecule bound tothe backbone carbonyl of Asn-111. ARPP,the adenosine ribose pyrophosphate moi-ety of NADH.
Structural and Mechanistic Principles of SDR Enzymes 25681
at Karolinska institutet library on February 23, 2015
nine ring and thus contributes to binding of the coenzyme.These hydrophobic interactions of Ala-88 appear to be an im-portant characteristic since mutation to Ser significantlychanges activities.
Reaction Mechanism of 3�/17�-HSD—The previously deter-mined triad of Ser-Tyr-Lys residues (positions 138, 151, and155, respectively in 3�/17�-HSD) constitutes the active site (2)(Fig. 4). Our data extend this concept by addition of an essen-tial Asn-111 to this triad to form an active site tetrad. Previousstudies support the concept that Tyr-151 functions as the cat-alytic base (2), whereas Ser-138 stabilizes the substrate, andLys-155 forms hydrogen bonds with the nicotinamide ribosemoiety and lowers the pKa of the Tyr-OH to promote protontransfer (Fig. 5). In the 3�/17�-HSD apo structure, water mol-ecules are bound to the Tyr-OH and Lys side chain, thusmimicking substrate and ribose hydroxyl group positions. De-termination of the Drosophila alcohol dehydrogenase structure(21) revealed interaction of the conserved Asn-111 via a watermolecule, binding to the active-site Lys-155, and this interac-tion is also observed in the present structure (OAsn-111-H2O:2.76 Å). In the Drosophila alcohol dehydrogenase structure, alarge hydrogen-bonded solvent network including the watermolecule bound by Asn-111 and Lys-155 was found (21),thereby substantiating our assumption of a proton relay withaccess to bulk solvent molecules (cf. below). Inspection of avail-able three-dimensional SDR structures reveals interactions of
Asn-111 (Table III) similar to those in 3�/17�-HSD and Dro-sophila alcohol dehydrogenase, indicating a homologous role ofAsn-111 in all these cases. Out of 20 SDR structures retrievedfrom the Protein Data Bank, 16 contain Asn at the positionhomologous to the one in 3�/17�-HSD, show similar side chain/backbone interactions, and display the feature of having aconnecting water molecule to the active site lysine. Moreover,the four structures without a homologous Asn-111 contain aSer residue, which is connected through a water molecule tothe active site Lys. An extended network is created in mousesepiapterin reductase built (Protein Data Bank accession num-ber 1nas) with an additional Arg residue involved. Based onthis general configuration, we conclude that Asn-111 is impor-tant to stabilize the position of Lys-155, and furthermore, thata proton relay is formed in most if not all SDR structures at theactive site, including coenzyme, substrate, Tyr-151, ribose2�OH, Lys-155, water, and Asn-111 or a corresponding Ser(Figs. 4–6). A proton relay system involving water and essen-tial ribose contacts to the catalytic base similar to that found inhorse liver alcohol dehydrogenase (22) has been postulatedearlier also for SDR (7); however, here we provide direct evi-dence for a critical involvement of Asn-111 in this process. Thestabilization of the active site geometry is thus achievedthrough maintaining the Lys-155 position and furthermorethrough Asn side chain interactions with main chain atoms
TABLE IIIObserved side-chain and main-chain interactions of Asn-111 and Thr-12 in 3�/17�-HSD (Protein Data Bank accession number 1hxh) and
homologous interactions in available SDR structures
a Predicted to be 2.7 Å by comparisons to water-containing crystal structures of other SDR members.b Lys-175 NZ–Water 2.93 Å, Water–Arg-178 ND2 2.90 Å.
Structural and Mechanistic Principles of SDR Enzymes25682
at Karolinska institutet library on February 23, 2015
of residues located within the segment preceding helix �E(Table III).
Asn-111 is located within �E, the main dimerization inter-face in oligomeric SDRs. Notably, at this position, the helixforms a sharp kink. This motif created by the side chain ofAsn-111 presumably forces its backbone carbonyl group to binda water molecule instead of the amide group (Val-115) thatwould have been expected in an �-helical structure. Moreover,we found other water molecules in the same cavity, forming asmall water-rich enclosure inside the protein fold. In 3�/17�-HSD, we observed four water molecules, and in other SDRstructures, the number ranges from two to five. This water-filled hydrophilic cavity is lined by other well conserved aminoacids: the side chains of Thr-12 (for Protein Data Bank acces-sion number 1a4u, Thr-114; for 1dhr, Thr-110; and for 1a27,Thr-118), Asn-86, and Ser-114 and the main chain carbonylgroup of Ala-88. This hydrophilic pocket serves as a protonrelay system by apparently acting as a proton bridge betweenLys-155 and the bulk solvent, similar to a temporary protonreservoir during enzymatic catalysis (Fig. 6), and it can pre-sumably stabilize a positively charged Lys-155. The protontransfer in 3�/17�-HSD between W1 and W2 can be mediatedvia a carbonyl group (Fig. 6), proceeding through an enolicintermediate (Fig. 7). This mechanism explains the necessityfor a small amino acid (Gly-89) within the NNAG motif sincethe rotational freedom of a small residue at that position willmore easily allow a positive charge to be temporarily localizedon the amide group (Fig. 7, step 2). The amino acids (Thr-12,Asn-86, Ala-88, and Ser-114) lining the hydrophilic pocket ap-pear to have a secondary role in stabilizing these water mole-cules and in allowing proton transfer among them, inside anotherwise hydrophobic environment (Fig. 6).
To exclude involvement of Asn-111 in oligomerization pro-cesses, we performed sedimentation analysis and found nochanges between wild type and N111L. Thus Asn-111 interac-tions are not essential to subunit associations. We postulatethat the Asn-111 side chain interactions to Ile-90 and Ser-154(OD1Asn-111-NI90: 2.80 Å; ND2Asn-111-OIle-90 3.03 Å;ND2Asn-111-OGSer-154: 3.06 Å) and homologous interactions
(Table III) are necessary to position the Asn-111 backbonecarbonyl for binding of the water molecule, which participatesin the postulated proton relay and stabilizes the geometry ofactive site residues. This view is supported by mutational anal-ysis of a homologous Ser (Ser-154 in 3�/17�-HSD) in type 111�-hydroxysteroid dehydrogenase, showing a critical involve-ment of this residue in catalysis (23).
Role of Conserved Residues in SDR Enzymes—With over2000 sequences annotated in databases and about 20 crystalstructures determined, a picture of the general SDR architec-ture and mechanism emerges. Considering all sequences, nostrict positional conservation is noted. However, multiple se-quence alignments revealed several consensus motifs, the mostconserved being the N-terminal TGX3GXG around Thr-12, aspart of the nucleotide binding fold, and the active site SYKtriad, now shown to form a tetrad with the conserved Asn-111.The NNAG motif around residue 86, the conserved Asn-111,defined in this study, and further motifs (comprising a con-served Asn-179 in strand �F, the PG motif, and the conservedThr-188; Fig. 1), identified through structural alignments andfunctional analyses (6, 12, 24) reveal the critical involvement ofconserved elements for coenzyme binding, maintenance of theSDR scaffold, and catalysis. Notably, the recent structure de-termination of sequence-unrelated proteins displaying the SDRfold considerably extend structure-activity relationships (25).Thus the SDR domain structure appears to be a generic scaf-fold not only including dehydrogenases/reductase, lyase, epi-merase, and hydratase activities but also comprising RNAbinding proteins, kinases, and transcription factors (25).Greater understanding of the mechanistic and structural prin-ciples governing the SDR architecture will reveal novel sub-strate and protein-protein interactions and will facilitate thedevelopment of inhibitors directed against biologically relevantSDR targets. These efforts constitute avenues currently pur-sued at several pharmaceutical sites.
Acknowledgment—Expert technical assistance by Eva Mårtensson isgratefully acknowledged.
REFERENCES
1. Jornvall, H., Persson, M., and Jeffery, J. (1981) Proc. Natl. Acad. Sci. U. S. A.78, 4226–4230
2. Jornvall, H., Persson, B., Krook, M., Atrian, S., Gonzalez-Duarte, R., Jeffery,J., and Ghosh, D. (1995) Biochemistry 34, 6003–6013
FIG. 6. Coordination distances (in Å) of the water clusterfound in 3�/17�-HSD and its connection to Lys-155. RIBBONS(26) was used to plot this figure.
FIG. 7. Extended proton relay system involving the backbonecarbonyl group of Ala-88.
Structural and Mechanistic Principles of SDR Enzymes 25683
at Karolinska institutet library on February 23, 2015
3. Oppermann, U. C., Filling, C., and Jornvall, H. (2001) Chem. Biol. Interact.130–132, 699–705
4. Kallberg, Y., Oppermann, U., Jornvall, H., and Persson, B. (2002) Protein Sci.11, 636–641
5. Ghosh, D., Weeks, C. M., Grochulski, P., Duax, W. L., Erman, M., Rimsay,R. L., and Orr, J. C. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 10064–10068
6. Ghosh, D., and Vihko, P. (2001) Chem. Biol. Interact. 130–132, 637–6507. Liao, D. I., Thompson, J. E., Fahnestock, S., Valent, B., and Jordan, D. B.
(2001) Biochemistry 40, 8696–87048. Oppermann, U. C., Persson, B., Filling, C., and Jornvall, H. (1997) Adv. Exp.
Med. Biol. 414, 403–4159. Tanaka, N., Nonaka, T., Nakanishi, M., Deyashiki, Y., Hara, A., and Mitsui, Y.
(1996) Structure 4, 33–4510. Tanaka, N., Nonaka, T., Tanabe, T., Yoshimoto, T., Tsuru, D., and Mitsui, Y.
(1996) Biochemistry 35, 7715–773011. Varughese, K. I., Skinner, M. M., Whiteley, J. M., Matthews, D. A., and Xuong,
N. H. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 6080–608412. Filling, C., Nordling, E., Benach, J., Berndt, K. D., Ladenstein, R., Jornvall, H.,
and Oppermann, U. (2001) Biochem. Biophys. Res. Commun. 289, 712–71713. Oppermann, U. C., Filling, C., Berndt, K. D., Persson, B., Benach, J.,
Ladenstein, R., and Jornvall, H. (1997) Biochemistry 36, 34–4014. Sarkar, G., and Sommer, S. S. (1990) BioTechniques 8, 404–407
15. Benach, J., Knapp, S., Oppermann, U. C., Hagglund, O., Jornvall, H., andLadenstein, R. (1996) Eur. J. Biochem. 236, 144–148
16. Sheldrick, G. M., and Schneider, T. R. (1997) Methods Enzymol. 277, 319–34317. Ghosh, D., Wawrzak, Z., Weeks, C. M., Duax, W. L., and Erman, M. (1994)
Structure 2, 629–64018. Andersson, A., Jordan, D., Schneider, G., and Lindqvist, Y. (1996) Structure 4,
1161–117019. Auerbach, G., Herrmann, A., Gutlich, M., Fischer, M., Jacob, U., Bacher, A.,
and Huber, R. (1997) EMBO J. 16, 7219–723020. Ghosh, D., Sawicki, M., Pletnev, V., Erman, M., Ohno, S., Nakajin, S., and
Duax, W. L. (2001) J. Biol. Chem. 276, 18457–1846321. Benach, J., Atrian, S., Gonzalez-Duarte, R., and Ladenstein, R. (1998) J. Mol.
Biol. 282, 383–39922. Eklund, H., Plapp, B. V., Samama, J. P., and Branden, C. I. (1982) J. Biol.
Chem. 257, 14349–1435823. Obeyesekere, V. R., Trzeciak, W. H., Li, K. X., and Krozowski, Z. S. (1998)
Biochem. Biophys. Res. Commun. 250, 469–47324. Zhou, H., and Tai, H. H. (1999) Biochem. Biophys. Res. Commun. 257,
414–41725. Stammers, D. K., Ren, J., Leslie, K., Nichols, C. E., Lamb, H. K., Cocklin, S.,
Dodds, A., and Hawkins, A. R. (2001) EMBO J. 20, 6619–662626. Carson, M. (1987) J. Mol. Graph. 5, 103–106
Structural and Mechanistic Principles of SDR Enzymes25684
at Karolinska institutet library on February 23, 2015
and Udo OppermannNordling, Rudolf Ladenstein, Hans JörnvallBenach, Stefan Knapp, Tim Prozorovski, Erik Charlotta Filling, Kurt D. Berndt, Jordi Dehydrogenases/ReductasesCatalysis in Short-chain Critical Residues for Structure andPROTEIN STRUCTURE AND FOLDING:
doi: 10.1074/jbc.M202160200 originally published online April 25, 20022002, 277:25677-25684.J. Biol. Chem.
10.1074/jbc.M202160200Access the most updated version of this article at doi:
.JBC Affinity SitesFind articles, minireviews, Reflections and Classics on similar topics on the
Alerts:
When a correction for this article is posted•
When this article is cited•
to choose from all of JBC's e-mail alertsClick here
