Download Identification of two glutamic acid residues essential for catalysis in

Protein Engineering vol.9 no.12 pp. 1191-1195, 19%
Identification of two glutamic acid residues essential for catalysis
in the {3-glycosidase from the thermoacidophilic archaeon
Sulfolobus solfataricus
Marco Moracci1, Luisa Capalbo2, Maria Ciaramella and
Rossi3
Institute of Protein Biochemistry and Enzymology—CNR,
Via Marconi 10, 80125, Naples, Italy
2
Present address: Biology Department, University of Utah, Salt Lake City,
UT 84112, USA
3
Dipartimento di Chimica Organica e Biologica, Universita di Napoli, Via
Mezzocannone 16, 80134 Naples, Italy
'To whom correspondence should be addressed
The Sulfolobus solfataricus, strain MT4, p-glycosidase (SsPgly) is a thermophilic member of glycohydrolase family 1.
To identify active-site residues, glutamic acids 206 and 387
have been changed to isosteric glutamine by site-directed
mutagenesis. Mutant proteins have been purified to homogeneity using the Schistosoma japonicum glutathione Stransferase (GST) fusion system. The proteolytic cleavage
of the chimeric protein with thrombin was only obtainable
after the introduction of a molecular spacer between the
GST and the SsP-gly domains. The Glu387 -> Gin mutant
showed no detectable activity, as expected for the residue
acting as the nucleophile of the reaction. The Glu206 —»
Gin mutant showed 10- and 60-fold reduced activities
on aryl-galacto and aryl-glucosides, respectively, when
compared with the wild type. Moreover, a significant A^m
decrease with p/o-nitrophenyl-P-n-glucoside was observed.
The residual activity of the Glu206 -> Gin mutant lost the
typical pH dependence shown by the wild type. These data
suggest that Glu206 acts as the general acid/base catalyst
in the hydrolysis reaction.
Keywords: chimeric enzymes/p-glucosidase/glycosyl hydrolase
active site/site-directed mutagenesis/'Sulfolobus solfataricus
Introduction
The Sulfolobus solfataricus P-glycosidase (SsP-gly; EC
3.2.1.x) shows broad specificity against (3-(l-4)-, P-(l-3)- and
p-(l-6)-(9-glucosides, and remarkable exo-glucosidase activity
against oligosaccharides, which are hydrolyzed from the nonreducing end (Nucci et al., 1993). This enzyme hydrolyzes
P-glycosides, maintaining the anomeric configuration of the
substrate; for this reason it can be classified as a retaining
glycosyl hydrolase (Moracci et al., 1994). Like other enzymes
from hyperthermophilic Archaea, Ssp-gly is extremely stable
to heat with a half-life of 48 h at 85CC, and displays optimal
activity at temperatures >85°C (Moracci et al., 1995a). The
amino acid sequence places the enzyme in glycosyl hydrolase
family 1 (Henrissat, 1991), along with archaeal, bacterial and
eukaryal enzymes. Recently the protein has been crystallized
in its native form (Pearl et al., 1993), and resolution of its 3-D
structure is currently under way (C. Aguilar et al., manuscript in
preparation).
Retaining glycosidases operate by means of a two-step
reaction involving a glycosyl-enzyme intermediate, supported
© Oxford University Press
by two residues in the active site (often carboxylic acids),
where one acts as a general acid/base catalyst and the other
as a nucleophile (Sinnot, 1990). In the Agrobacterium
P-glucosidase (Abg), belonging to family 1, the attacking
nucleophile and the general acid/base have been identified as
conserved glutamic acid residues by covalent modification
with a suicide substrate and by site-directed mutagenesis,
respectively (Withers et al., 1990; Wang et al., 1995). 3-D
structure comparisons and hydrophobic cluster analyses have
shown that this system of two catalytic glutamic acid residues
in a conserved a/p-barrel fold is common to several glycosyl
hydrolase families (Henrissat etal, 1995; Jenkins et al, 1995).
More recently, the resolution of the 3-D structure of two
glycohydrolases of family 1 confirmed these observations,
showing that the two catalytic residues are separated by the
distance expected for retaining glycosidases. The 3-D structure
also provided an insight into the active site of the enzymes of
this family (Barret et al., 1995; Wiesmann et al., 1995). In
particular, the analysis of the active centre confirms the
observation of Davies and Henrissat (1995), that small differences in the structure rather than the global fold may be
responsible for the large variety of substrate specificities
observed among glycosidases.
Here we report the construction of SsP-gly mutants which
are severely impeded in their activity. A kinetic analysis
provides evidence that the mutated residues act as the activesite nucleophile and the general acid/base catalyst.
Materials and methods
Protein expression and purification
Wild-type and mutant enzymes were expressed as fusions with
glutathione S-transferase (GST). Fusion proteins were purified
from Escherichia coli JM 105 transformed with vectors
pGEXGly and pGEX-K-Gly, which were constructed by
inserting the S.solfataricus lacS gene from pDAFl (Moracci
et al., 1995a) into pGEX-2T and pGEX-2TK (Pharmacia,
Uppsala, Sweden), respectively. Bacterial strains expressing
GST-Ssp-gly chimeric proteins were grown in Luria-Bertani
medium with 50 ug/ml ampicillin at 37°C to 0.6 OD at
600 nm, induced by adding 0.1 mM isopropyl-p-D-thiogalactopyranoside and grown overnight. Cells were harvested by
centrifugation at 4000 g at 4°C, resuspended in buffer A
[50 mM sodium phosphate buffer, pH 7.3; 150 mM NaCl; 1%
(v/v) Triton X-100], and broken by three passes through
a French press (American Instrument Company, Travenol
Laboratories Inc., Silver Spring, MD). The extract was clarified
by centrifugation at 17 000 g, passed through a 0.45 (im filter
and loaded onto a glutathione sepharose 4B column equilibrated
with buffer A. Binding of the chimeric proteins was monitored
by measuring the GST activity in the fractions at room
temperature, according to the manufacturer's instructions
(Pharmacia). The column was eluted widi buffer B (0.5 M
Tris-HCl, pH 8.0; 10 mM reduced glutathione), and fractions
displaying GST activity were pooled and incubated with
1191
M.Moracci et aL
thrombin protease (10 U/mg protein) overnight at 25°C to
separate GST and Ssfi-gly. Thrombin cleavage was monitored
by SDS-PAGE. Digested samples were applied to a Superdex
200 HG 26/60 FPLC column (Pharmacia) and eluted in buffer
A without Triton but with 1 mM dithiothreitol. The wild-type
and mutant enzymes displayed the same retention time on this
column. Pooled fractions from gel filtration (which SDSPAGE showed to be >95% pure) were concentrated and
dialyzed against 50 mM sodium phosphate buffer, pH 6.5,
with 50% (v/v) glycerol, and used in every subsequent characterization. When stored at -20°C, the enzyme preparation was
stable for several months. The yield from this procedure was
~ l - 2 mg/1 of bacterial culture.
Protein concentrations were determined by the method of
Bradford (1976), using bovine serum albumin as standard.
Spectrophotometrically, a 10 mg/ml SsP-gly solution in 0.1 M
sodium phosphate buffer, pH 6.5, showed an absorbance of
28.78 OD at 280 nm at room temperature.
Enzyme characterization
Standard assays of Ssp-gly activity against nitrophenyl-glycoside and disaccharide substrates were performed as reported
previously (Nucci et aL, 1993). Kinetic parameters for Ss(3gly were measured at the indicated pHs and temperatures using
aryl-glycoside substrate concentrations ranging from 0.05 to
20 mM. The protein concentrations in the reaction mixture
were 0.2-2.0 (Xg/ml for the wild type and 2-10 (J.g/ml for the
mutants. The effect of pH on activity was studied using 50 mM
sodium citrate (pH 3.0-5.5), 50 mM sodium phosphate (pH 6.08.0) and 50 mM potassium chloride/50 mM boric acid (pH 8.210.3) buffers (pH values were measured at the temperature of
the assay). The stability of the enzyme at each pH was tested
before the assay, and activity values were determined only at
pH values at which the enzyme was stable at least for 5 min.
Thermal activity experiments were performed as reported
previously (Moracci et aL, 1995a). All kinetic data were plotted
and refined using the program GraFit (Leatherbarrow, 1992).
Mutagenesis
The vector pBluMSl used as a template for the site-directed
mutagenesis was prepared by ligating a BamH\-Pst\ fragment
of pDAFl containing the lacS gene into pBluescript n KS(+)
(Stratagene, La Jolla, CA). Site-directed mutagenesis was
performed by the method of Mikaelian and Sergeant (1992)
based on the PCR. The oligonucleotides used were T7 and
'reverse' sequencing primers (Promega and Stratagene) and
the following synthetic oligonucleotide (Primm, Milan, Italy):
5'-TATTATAGCGAGGTCGACGGTATCG-3', where the
bold 5' end represents mismatched nucleotides.
Mutagenic oligonucleotides, designed according to Kuipers
et aL (1991), were purchased from Pharmacia and were as
follows (mismatches are in bold): E206Q, 5'-ACTCAACAATGAATCAACCTAACGTTGTT-3'; E387Q, 5'-CTATATGTACGTTACTC A AA ATGGTATTGCGG A-3'.
Amplified DNA fragments were purified by agarose gel
electrophoresis, digested with Xbal and cloned in vector
pGEM3 (Promega). Mutant clones were identified by direct
sequencing, and restriction fragments containing the mutation
were completely re-sequenced and substituted for the corresponding fragments in the wild-type pGEX-K-Gly.
Results and discussion
Family 1 of glycosyl hydrolases includes enzymes with different substrate specificities: p-glycosidases with broad specificity,
1192
6-phospho-p-gluco and (3-galactosidases, myrosinases and lactases. These retaining enzymes utilize a double displacement
mechanism. During the first step, a carboxylic acid residue
acting as the nucleophile displaces the glycosidic oxygen and
releases the aglycon, in a process that requires the assistance
of a general acid catalyst, resulting in a covalent glycosylenzyme intermediate. During the second step, the general base
catalyst promotes water attack to the anomeric centre of the
glycosyl-enzyme intermediate, releasing the sugar and the free
enzyme. The double role of a general acid/base catalyst is
played by the same carboxyl group.
The two carboxylic residues essential for catalysis in Sspgly were searched for among the glutamic and aspartic acids
conserved in mesophilic and thermophilic members of family
1. The glutamic acid 206 and 387 residues of SsP-gly, found
in the two fully conserved motifs Asn-Glu-Pro and Glu-AsnGly (Figure 1), respectively, correspond to the general acid/
base catalyst and the nucleophile identified previously in the
active site of Abg (Withers et aL, 1990; Wang et aL, 1995).
Glu206 and Glu387 were changed to glutamine by sitedirected mutagenesis. The substitution of glutamate by the
isosteric glutamine was chosen to delete the charge without
introducing major changes in the local structure of the sites.
To facilitate the purification of proteins expected to have
impaired enzyme activity, wild-type and mutant enzymes
were expressed by fusing their N-termini to the Schistosoma
japonicum GST. Unfortunately, the GST-SsP-gly chimeric
proteins, expressed by the pGEXGly vector, were found to be
resistant to thrombin hydrolysis, thus indicating that the SsPgly N-terminus is not accessible to thrombin. A similar result
was reported for a barley P-glucanase (Chen et aL, 1995).
This problem was overcome by constructing the pGEX-K-Gly
vector in which the recognition sequence of the cAMPdependent protein kinase acts as a spacer between the thrombin
cleavage site and the N-terminus of the SsP-gly (Moracci
et aL, 1995b). Chimeric proteins expressed from this vector
were cleaved efficiently, producing recombinant or 'long'
forms with a seven-residue extension at their N-terminus. The
recombinant enzyme containing the wild-type Ssp-gly coding
sequence did not differ significantly from the native form with
regards to the pH optimum, thermostability, thermophilicity
and kinetic parameters (data not shown). Hence, the SsP-gly
'long' form was used in the subsequent characterization and
will be referred to as 'wild type' throughout this paper. The
use of the recognition sequence of the protein kinase as a
spacer between domains in GST fusions may be applied to
other enzymes with buried N-termini.
Table I compares the specific activity of the wild type and
of the two mutants on several substrates at 65°C. The Glu387 —>
Gin mutant is totally inactive against all the substrates tested.
The complete inactivity showed the absence of contamination
with the wild-type enzyme, and did not allow reliable estimates
of kinetic parameters to be obtained. The mutation did not
affect the overall structure of the molecule, as tested by farUV CD spectra at 20 and 75°C (data not shown). The Glu387
is fully conserved in family 1; the mutation of the corresponding
glutamic acid residue to glutamine in the Pyrococcus furiosus
p-glucosidase led to a reduction in the specific activity of
>103-fold (Voorhost et aL, 1995). These data are consistent
with the function of Glu387 as the attacking nucleophile.
The Glu206 —» Gin mutation strongly affects, but does not
completely abolish, glycosidase activity. Although the mutant
was found to be completely inactive on disaccharides (Table
S.solfataricus [J-glycosidase active site
Sgly 1
Sgol 1
Pfu 1
Tm
Cth
Abg
BpA
BpB
CM
Bet
Cbg
1
1
1
1
1
1
1
1
HYSFP K F R F O S * GFQSEHGTPG SEDPNTD1YK WVKDPBMU GLVSGOLPEN GPGYWGNYKT FHDNAQKKa KIARLNVBTC
HLSFP KGFKFGHSQS GRJSEMGTPG SEDPMSMHV WVHDRBffVS QWSGDLPEN GPGYKNVKR FHDEAEHGL NAVRINVB»S
HKFP KHFHFCYSWS GFQFEMGLPG SE.VESDWV HVHWENIAS GLVSGDLPEH GPAYBfl-YO} DHDIA3XGM DCIRGGIEKA
WWKKFP
HSKITFP
HT DPKTLAARFP
KTIIHJFP
KSEHTFIFP
1O4SFP
HSIHHFP
FKPLP1SFOD FSDURSCFA
EGFLWGVATA
KDFIWGSATA
GDF1.FGVATA
QDFMGTATA
ATFUKTSTS
KGFLHGAATA
SDfWGVATA
PGFVFGTASS
SYQIEGSPU
AYQIEGAYNE
SFQIECSTKA
AYQIEGAYQE
SYQIEGGTDE
SYQIEGASNE
AYQIEGAYNE
AFQYKAAFE
DGAGHSIIWT
DGKGESTJDR
DGRKPSIM*
DGRGLSIMJT
GWH*>HIUI
DGKGESTJOt
DGRGMSIWT
DGKGPSUDT
FSHT
PG WKNGDTGDV
F S H T . . . . P G fflADGHTGDV
FO«
PG HVFGRHHGDI
FAHT....PG KVFHGDNOW
F C Q I . . . . P G KVIGGDCGDV
FTHQ... .KR WLYGHHGDV
FAHT
PG KVKHGDNGMV
F n * . . . Y P E lUKDRTHGDV
ACDHYNRWE
ACDHYHRYEE
ACOHYNRWEE
ACDSYHRYn
ACOHFHHFKE
ACDHYHRfEE
ACDSYHRVEE
AIDEYtRYKE
DIEHEW.GV
DIKIMKHGI
DLDLIKEHGV
OIRLHKELQ
DVqLMKqLGf
DVSUOCELGL
DVQLUUXGV
DIGIWDMM.
KAYRFSISKP
KSYRFSISWP
EAYRFSLAWP
RTYRFSVSKP
LHYRFSVAKP
KAYRFSIAWT
ICVYRFSISUP
DAYRFSISWP
Sgly 86
Sgal 86
Pfu 8 4
RIFFHPLPRP Q. ..HFDESK CPVTEVHHE NEURLDEYA KKDAUffffiE IFKDLXSRGL YFILNKYWP LPLWLHDPIR VRR.GDFT6P SGWLSTRTVY
RIFPRPLPKP EHQTGTDKEN SPYISVDUrt SKLREHONYA ItfEALSHYRH ILEOLRKRGF HIVLNHYWrT LPIWLHDPIR VRR.GDfTGP TGWLNSRTVY
RIFPKPTFW KVDVEKDEE. QCIISVDVPE STIKaEKIA WEALEHYRX IYSOKERGK TFILNLYWP LPUflHDPIA VRKLGPORAP AGWLDaCTW
Tm
Cth
Abg
BpA
BpB
CM
Bd
Cbg
RILPEG..TG
RIFPEG. .TG
RIIPDG..FG
RIFPNG. .DG
RIWAA...G
RIFPOG. .FG
RVLPQG. -TG
RVLPKQCLSG
84
84
89
84
86
83
84
98
RV
KL
PI
EV
II
TV
EV
GV
NQKGLDFYHR
HQKGLDFYKR
NEKOLDFYDR
NQEGLDYYHR
NEEO.LFYEH
KJCGLEFYDR
NRAGLDYYHR
NREGIKITYNN
IIDTLLEKGI
LTKLLLEHGI
LVDGCKARGI
WDLUCNGI
LLDHEUGL
LINKLVEMGI
LVDELLANGI
LIKEVLANGM
S g l y 1 8 2 EFARFSAYIA HKFDOLVDEY STHMEEHWG GLGYVGVKSG FPPGYLSFEL S
S g a l 1 8 5 EFARFSAYVA IKLDOLASEY ATWE0WVW GAGYAFPRAG FPPMITLSFRL S
P f u 1 8 3 EFVKFAAFVA YHLDDLVDW SDfJEENWY NOGYINLRSG FPPGYLSFEA A
Tun
Cth
Abg
BpA
BpB
CM
Bet
Cbg
1 4 2 WFAEYSRVLF
1 4 2 YFTEYSEVIF
1 4 7 AFQRYAKTVM
1 4 2 AFVOFAETW
1 4 3 HFKTYASVI1I
1 4 1 YYFDYANLVI
1 4 2 AFAEYAEU4F
1 5 9 DFRDYA£LCF
S g l y 2 6 6 SF
S g a l 2 6 9 SY
Pfu 2 6 7 WH
Tin
Cth
Abg
BpA
BpB
CM
Bet
Cbg
ENfGDRVXNi
KNLGDIVPIW
ARLGDRLDAV
REFHOaQH*
DRFGERIMMI
mYKOCVWOI
KELGGKIKQW
KEFGDRVKHi
nLME0»WA
FTHiEeGWS
ATFME0KAV
LTFHE0KIA
NTINiEYCAS
nFHEEYCIA
nFffiPJICm
ITI MFPWCTrt
•• T
HAPGHRDIYV
HAPaKDLRT
HAPGERtMA
HAPGLTN1.QT
HAPGHEHWE
HAPGIKDFKV
HAPGKKDLQl
FAPGRCSOTL
2 2 5 YF
EP ASEKEEDIRA
2 2 5 YH
YP ASEKAEDIEA
2 3 8 SA
I P ASDGEAOLKA
2 2 5 WA
VP YSTSEEDKAA
2 2 6 HV
DA ASERPEDVAA
2 2 4 PVYLQTERLG YKVSEIEREH
2 2 S WA
VP YRRTKEDffiA
2 5 7 WF
EP ASKEKADVDA
VRFMHQFNNY
AELSFSLAG.
AERAFQFHHG
CARTISLHS.
AIRRDGRN.
VSLSSQLDN.
ORVNGSSG.
AKRGLDFH.L
3 9 9 KVS
FVE R
3 1 9 GFSPA.NSIL. E
3 1 3 EFPATWAPA V
3 9 8 GFLQS.EEIN M
3 1 9 SLLQV.EflVH H
3 1 8 WIFPI.RWEH P
3 1 9 GHLSS.EAIS M
3 4 9 ARPAIQTDSL I
PLFLNPIYRG
RWYLDPVLKG
AFF.OPVFKG
DWFLQPTYQG
RWFAEPLFNG
QLFLDPVUG
DWLDPIYFG
GWFWPLTKG
DYPELVLEFA
RYPENAULY
EYPAEWEAL
SYPOFLVDWF
KYPSXVBIY
SYPOKLLDYL
EYPKFWJ3WY
RYPESWYLV
AFR
AVrtfl.
SU
VSttiL
ALA
UQVI
AID
VOtHL
AFT
AAttfl
A»
WHSL
AID
VStffl.
ItLNaGCDSG REPYLAAHYQ
RE
«KGI
GOR
AEQGA
GT
YLN
VOXDLLDSQK
EHLGY
RKR
LPfALQL
LPIHLQD
LPLTLHG
LP(>LQD
LPTJIED
LPqCLOD
LPQ4LQD
VPQALED
K. GGWANRE1AD
K. GGWKNROTTD
D. GOUSRSTAH
A. GGWGNRRTT.Q
E. GGWTORETIQ
I . GGWAHPEIVN
Q. GGWGSRITID
EY RGFLGRNIVD
RRHHYHI KJWARAYD5 IKSVSWCP.. . . V Q I Y A N S
EIAMNI IQAHARAYDA IKSVSKKS.. ..VGIIYANT
EKAKFNL IQAHIGAYDA IKEYSEKS.. . .VGVIYAFA
QPLTDKOeA VEHA.EfCNR WKFFOAIIRG ETJRGteaV ROD
YPLRPQDNEA VEIA.ERLW WSFFDSIIKG ETrSEGQN.V RED
DPUEEYKDE VE...EIRHC DYEFVTILHS
S g l y 3 3 6 SLGGYGHaE RNSVSUGLP TSDFGWE
S g a l 3 3 8 TLPGYGDRCE RNSLSLANLP TSDFGWE
P f u 3 2 1 PLPGYGFMSE RGGFAKSGRP ASDFGWE
Tna
Cth
Abg
BpA
BpB
CM
Bet
Cbg
IVGHLY. .GV
LLGHR..GI
WLSHIY. .GV
FLSIM...GV
ILGYGT..GE
FLGYFH..GI
FLSNYL. .GV
HNAYAY..GT
TPFVTIYMD
WAIUYWO
(CTYATLYrtH)
EPFCTLYHID
IPMLTLYHWD
EPWTLYWD
EPFCTLYHWD
(FYVTLFhM)
•T
LRAHARAVXV
LLSHGKAVH.
NUHGFGVEA
LVAHaSVRR
LHCHGIASNL
MLSHFKWKA
LVAHGRAVTL
UAHAAAARL
F R . . .ETVXD
FR.. .EHNID
S R . . .HVAPK
F R . . .ELGTS
HK.. .EKGLT
VK...EIHID
FR.. .ELGIS
YICTXYVSON
GHGIVFNNG
AQIGIALNLS
VPVaVLHAH
GQIGIAPNVS
GKIGITLWE
VEVGITLNLT
GHGIAPHTS
GIIGITLVSH
T
UGRLD WIGVNYYTRT WtCRTEKGYV
LRNRLD WIGVNYYTRT WTKAESGYL
KGKLD WIGVNYYSRL VY6AKTXH.V
YLPEHYKD
&SFPE. . . D
MPWEAE
TVPK3D...G
GLDFVQ. .PG
ALSMQQEVKE
KPPIVD...G
LPKFSTE
DMSEIQBaD
DUOJSOPID
DLGIISQKLD
CHDIIGEPID
WEUQQPGD
NFIF....PD
DXEUHQPID
ESKELTGSFD
FF PEGLYDVLTK YWRYH..LY HYVTfliGJA
FF PFGIYDVUK YIWRYG..IP 1YVHFMGIA
Hi PKLEMJJCY UfWYE. .LP W n f l f i i W
DLP laAHGWE
IV PEGIYfflUK VKEEYNP.PE VYITQifi.AA
KFE KTDHGH
IY TCGLYDLLHL LDROYGK.PN IVISfiifi.AA
SOV KTDIGWE
VY APALHTLVET LYERYD.LPE CYTTafi.AC
GLP VTDIGWP
VE SRGLYEVLHY .LQKYGN.ID IYITfHS..AC
EEP VTDHGWE
IH PESFYKLLTR IEKDFSKGLP ILTTHK-AA
AGE YTEHGWE
VF PQGLFDLLIW IKESYPQ.IP IYnfiJfi.AA
GAP KTDIGW
IY AEGLYOLLRY TADKYGN.PT LYTTHK.AC
KAT FE)«GKPLGP HAASSWLCIY PQGIRXLaY VMWYNN.PV IYnfJfiiRNE
»••
FVaKYYSGH
FIAFNKYSSE
WWGLNYYTPH
WGIHYYSHS
FLQHYYTRS
FLGINYYTRA
FIMNYYTSS
FLGLNtTSSY
LVKFDPDAPA
FIKYDPSSES
RVADOATPGV
VNRFNPE..A
IIR..STNDA
VRLYD.ENSS
HNRYNPGEAG
YAAKAPRIPN
DDADYQRPYY LVSHVYQVHR
DDADYQRPYY LVSHIYQVHR
DAADRYRPHY LVSHLKAVYN
FDO.WSEOG
FKD.HGSHG
YIW.GV.ENG
IKD.EV.VMG
WD.a.VHG
YND.IVTEDG
YND.GLSLDG
FNDPTLSLQE
RVHJQNRIDY LKAHIGQAKK
KIEDTKRIQY LKDYLTQAHR
QVNDQPRLDY YAEHLGIVAD
KVQODRRISY MQQHLVQVHR
QIEDTGRHGY IEEHLKACHR
KVWSKRIEY LKQHFEAARK
RimQRRIDY LAMHLIO/kSR
SLLDTPRIDY YYRHLYYVLT
S g l y 4 1 2 AINSGADVRG YUKSLADNY EWASGFSHRF GLUCVDY.NT O0.YWRPSAL VYReATNGA TTDEIEHmS VPPVKPtRH
S g a l 4 1 4 ALNEGVDVRG YLHTSLADMY BISSGFSHRF GU-KVDY.LT KRLYWRPSAL VYREITRSKG IPEELEHLKR VPP1KPLRH
P f u 3 9 7 AWCEGADVRG YLH«SLTDMY EWAQGFRHRF GLVYVDF. ET WCRYLRPSAL VFREIATQKE IPEELAHLAD LKFVTRK. .
TBO
Cth
Abg
BpA
BpB
Cto
Bet
Cbg
3 8 5 AIQEGVPUG
3 8 9 AIQDGVNLKA
3 9 2 LIRDGYPWG
3 8 5 TIHXLHVKG
3 8 9 FIEEGGQLKG
3 9 7 AIENGVDLRG
3 8 9 AIEDGINUCG
4 3 3 AIGDGVNVKG
YFVWSLLDHF
YYIWSLLDNF
YFAWSLHDHF
YHAHSLLDHF
YFVWSFLDNF
YFVWSLMDNF
YMEWSLHDNF
YFAWSLFDW
BMEGYSKRF
EHAYGYHCRF
BUEGYRHRF
BlAEGYtMRF
EXAWCYSKRF
BIAHGYTKRF
BMEGYGWIF
EIDSGYTVRF
GWYVDY.ST
GIVHVHF.DT
GLVHVDY.QT
G«HVDF.RT
GIVHINY. ET
GIIYVDY. ET
GLVHVDY.DT
GLVFVDFKtM
QKRIVXDSGY
LERHICDSGY
QVRTVKHSa
QVRTP1CESYY
QERTPKQSAL
QKRIKKDSFY
LVRTPKDSfY
LMWKLSAH
WYSHWKNNG
WYKEVIIOtW
WYSALASGFP
WYRNWSMH
WFKQXMKNG
FYQQYIKEHS
WYKGV1SRGW
WFKSFLKK
LED
F
ICGNHGVAKG
LETRR
F
LOL
Fig. 1. Amino acid sequence alignment of enzymes of the glycohydrolase family 1. Alignment was performed using the program PILEUP. Numbers on the
left represent the residue numbers of the first amino acid of each line. The glutamic acid residues changed to glutamine by site-directed mutagenesis in
SsfJ-gly are indicated by an asterisk. The conserved motives Asn-Glu-Pro and Glu-Asn-Gly are underlined. Residues that make up the active centre in Cbg
of T.repens and are invariant in all sequences are labeled by a circle. Trpl38, Glyl86 and Val254 in the Cbg of T.repens are marked by arrows. Abbreviations
used and Swiss-Prot Data Bank accession numbers are as follows: Abg, Agrobaclerium sp. fi-glucosidase (P12614); Bci, B.circulans p-glucosidase (Q03506);
BpA, B.polymixa fi-glucosidase A (P22073); BpB, B.polymixa p^glucosidase B (P22505); Cbg, T.repens cyanogenic p^-glucosidase; Csa, Caldocellum
sacchawlyiicum fi-glucosidase A (P10482); Cth, Clostridium thermoceUum p^-glucosidase A (P26208); Pfu, Rfuriosus P-glucosidase (U37557); Tma,
Thermotoga maritima P-glucosidase A (Q08638); Sgal, S.solfaiaricus (strain DSM1616) (5-galactosidase (P14288); Sgly, S.solfataricus (strain MT4)
fi-glycosidase (P22498). The amino acid sequence of T.repens Cbg was obtained from the Brookhaven Protein Data Bank because the sequence in the
Swiss-Prot Data Bank did not correspond to that reported by Barret et al. (1995).
1193
M.Moracci et al.
I), and lacked any transglycosylating activity (M.Moracci and
A.Trincone, unpublished results), it showed a residual activity
on p/o-nitrophenyl-P-D-glycosides (Table I). Kinetic constants
were determined for the Glu206 —> Gin mutant at 65°C and
compared with values obtained for the wild type (Table II). A
10- to 30-fold decrease in the Km value and a 60-fold
decrease in the k^ value on p/oNpGlu occurred upon mutation,
suggesting that the glycosyl-enzyme intermediate accumulates
during the reaction, k,.^ values on p/oNp-fJ-D-galacto and
fucoside (/j/oNpGal and Fuc) were found to have lower
reduction levels. Furthermore, the tccJKm ratio decreased to
different extents using the different substrates; whereas values
obtained with aryl-gluco- and aryl-fucosides were barely affected (1.6- and 6.0-fold reductions, respectively), a 14- and 33fold reduction occurred respectively with aryl-galactosides.
These results suggest that p/oNp-gluco- and -galactoside substrates make different interactions with the Glu206 residue in
the Ssp-gly active site. This may reflect the differences in
enantioselectivity towards secondary hydroxyl groups of 1,2diols shown by Ssp-gly in the transglycosylation reaction
using phenyl-p-D-gluco- and -galactosides as glycoside donors
(Moracci et al, 1994).
It has been reported that adding small nucleophiles such as
azide partially restores the activity of the Abg mutant in which
the general acid/base residue has been changed to glycine
(Wang et al., 1995). In contrast, the addition of azide and
other nucleophiles to the reaction mixture did not significantly
affect the activity of the Glu206 -> Gin mutant (data not
shown). This difference may be ascribed to the different nature
of the substituting residues, because the glycine may permit
the access of small nucleophiles to the active site whereas the
larger glutamine may not.
Proof that the general acid/base catalyst of a glycohydrolase
has been removed is indicated by the change in activity
versus pH upon mutation. Therefore we determined the pH
dependence of the hydrolysis of pNpGal substrate for the wild
type and the Glu206 -> Gin mutant at 65°C (Figure 2). The
wild-type SsP-gly showed a typical bell-shaped curve, with
maximum activity at pH 6.5, suggesting that two amino acid
side chains with pKa values of 6.0 and 7.6 are involved in the
catalysis. This result is in agreement with the abnormally high
pKa value (6.8) found for the glutamic acid acting as the
general acid/base in the active site of the Bacillus circulans
xylanase (Davoodi et al., 1995). The Glu206 -» Gin mutation
strongly affected the activity dependence on pH, confirming
that the residual activity measured is due to the mutant itself
and not to contamination with the wild type. The low activity
of Glu206 —» Gin on pNpGal showed the same thermal
activation of the wild type. In addition, no anomalous Arrhenius
behavior was detected over the whole temperature range for
both enzymes (Figure 3). These results, and far-UV CD spectra
at 20 and 75°C (data not shown), suggest that the Glu206 -»
Gin mutation specifically affects catalysis with minor changes
to the general 3-D structure.
The results presented strongly suggest that glutamic acid
206 in Ssp-gly works as the general acid/base catalyst in the
hydrolysis reaction. The studies reported so far on glycohydrolases mutated in the general acid/base catalyst demonstrate that
the kinetic parameters can be affected to different extents. In
Abg, the mutation Glu -» Gly resulted in a lO^fold reduction
in the value of kml of the hydrolysis reaction on pNpGlu
(Wang et al., 1995); in the Staphylococcus aureus 6-phosphop-galactosidase, the corresponding mutation to glutamine
Table I. Specific activity of wild-type and mutant SsfJ-gly*
pNpGal
oNpGal
pNpGlu
oNpGlu
pNpFuc
pNp-tio-Gal
pNp-tio-Glu
Cellobiose
Laminartibiose
Gentiobiose
Wild type
Glu206 -> Gin
Glu387 -> Gin
55 6
60.1
54.8
48.0
58.2
NA
3.1
27.1
37.9
36.3
3.9
6.9
0.97
0.94
8.6
NA
0.4
NA
NA
NA
NAb
NA
MA
NA
MA
HA
MA
UA
NA
NA
10
•Values are shown as U/mg at 65°C.
b
NA (no measurable activity) means that, using concentrations of enzyme of
10 Hg/ml in the assay, the rates of change in absorbance did not vary by
altering the substrate concentrations, and were approximately the same as in
the control without substrate (typically <0.001
pH
Fig. 2. Comparison of the pH dependence on pNpGal substrate of wild-type
(O) and Glu206 -> Gin mutant ( • ) SsP-gly at 65°C.
Table II. Kinetic constants of hydrolysis at 65°C by wild-type and mutant Glu206 -» Gin
Substrate
oNpGal
pNpGal
oNpGlu
pNpGlu
pNpFuc
1194
Afm(mM)
Wild type
Mutant
Wild type
Mutant
Wild type
Mutant
0.95
1.17
1.01
0.30
0.45
1.53
4.79
0.03
0.03
0.09
295
275
252
240
263
34.0 ± 0.8
31.0 ± 0.4
4.46 ± 0.13
4.40 ± 0.06
35.0 ± 0.4
310
234
251
830
584
22
7
160
132
382
± 0.08
± 0.06
± 0.24
± 0.04
±0.11
±0.10
±0.17
± 0.006
± 0.003
± 0.005
±
±
±
±
±
6
4
12
7
12
S.solfataricus fj-glycosidase active site
400
1
1
1
1
1
i
1
r
•
[•
i
•-
- 100
1° 300
'_
t
Jr
,
£> '
'"-Si
-V
t
t
- 60
I 200
<
Zt
J
I
11
- 40
<s 10
I °
- 20
40
60
Temneralure (°C)
Fig. 3. Companson of the temperature dependence of wild-type (O, D) and
Glu206 —> Gin mutant ( • , • ) Ssp"-gly. Activities are shown as U/mg ( • ,
0 ) and as percentage relative to maximum activity at 85°C ( • , • ) . Inset:
Arrhenius plot of wild-type (O) and Glu206 -» Gin mutant ( • )
resulted in a 103-fold reduction of the km value on
o-nitrophenyl-p-r>galactoside-6-phosphate (Witt et al, 1993).
However, in the exoglucanase/xylanase from Cellulomonas
fimi, the mutation of the general acid/base catalyst to glycine
resulted in only a 200- and 20-fold reduction for the kcat and
KJKm values, respectively, on p-nitrophenyl-p-D-cellobioside
(MacLeod et al, 1994). The reasons for these differences are
not clear; it is possible that enzymes with different substrate
specificities require the assistance of the general acid/base
catalyst to varying extents (Wang et al, 1995).
Non-covalent interactions play a crucial role in the catalysis
of glycohydrolases (Namchuk and Withers, 1995); therefore,
residues surrounding the fully conserved active-site amino
acids may affect the catalytic mechanism. The analysis of the
3-D structure reported recently for a phospho-fi-galactosidase
from Lactococcus lactis and a cyanogenic (3-glucosidase from
Trifolium repens (Cbg) provided more information on the
amino acids involved in the catalysis of family 1 enzymes
(Barret et al, 1995; Wiesmann et al, 1995). In Cbg, three
residues (Trp, Gly and Val, indicated by an arrow in Figure
1) are thought to contribute to the increase in pKa of the
general acid/base catalyst Glul83 (Barret et al, 1995). We
noticed that the glycine and valine are not perfectly conserved
among family 1 enzymes, and in SsP-gly they are substituted
by valine and alanine, respectively. Interestingly, Val209 is
invariant in archaeal and bacterial thermophilic (3-glycosidases,
and AIa263 is fully conserved among archaeal thermophilic
enzymes. It is tempting to speculate that in active sites evolved
to promote activity at high temperatures a particular amino
acid environment is required to maintain the appropriate pKa
value of the general acid/base catalyst. Furthermore, at high
temperatures the substrate may assume conformations that
need a structural environment different from that of mesophilic
enzymes. Studies are in progress to identify additional residues
involved in the activity of SsP-gly and to clarify their role in
the catalytic mechanism.
by CNR Progetto Finalizzato Ingegneria Genetica and by EC project 'Biotechnology of Extremophiles' contract no. BIO2-CT93-0274.
References
Barret/T., Suresh.C.G., Tolley.S.P, Dodson.E.J. and Hughes.M.A. (1995)
Structure, 3, 951-960.
Bradford,M.M. (1976) Anal. Biochem., 72, 248-254.
Chen.U Garret.T.PJ., Fincher.G.B and H0J.P.B. (1995) J. Biol. Chem., 270,
8093-8101.
Davies.G. and Henrissat,B. (1995) Structure, 3, 853-859.
DavoodiJ., Wakarchuk.W.W., Campbell.R.L., Carey,P.R. and Surewicz,W.K.
(1995) Eur. J. Biochem., ZSl, 839-843.
Henrissat,B (1991) Biochem. J., 280, 309-316.
Hennssat,B., Callebaut,I., Fabrega,S., Lehn,P., MomonJ.P. and Davies.G.
(1995) Proc. Nail Acad. Set. USA, 92, 7090-7094.
JenkinsJ., Lo Leggio.L., Harris.G. and Pickersgill.R. (1995) FEBS Lett., 362,
281-285.
Kuipers,O.P., Boot,HJ. and de Vos,W.M. (1991) Nucleic Acids Res., 19, 4558
Leatherbarrow.RJ. (1992) GraFit. Version 3.0. Erithacus Software Ltd,
Staines, UK.
MacLeod,A.M., LindhorstX, Withers.S.G. and Warren.R.AJ. (1994)
Biochemistry; 33, 6371-6376.
MikaelianJ. and Sergeant,A. (1992) Nucleic Acids Res., 20, 376.
Moracci,M., Ciaramella.M., Nucci.R., Pearl.L.H., Sanderson.I., Tnncone.A.
and Rossi,M. (1994) Biocatalysis, 11, 89-103
Moracci.M., Nucci.R., Febbraio.F., Vaccaro.C, Vespa,N., La Cara,F. and
Rossijvl. (1995a) Enzyme Microbiol. Techno!., 17, 992-997.
Moracci.M., Capalbo.L., De Rosa.M., La Montagna,R., Morana.A., Nucci,R.,
Ciaramella,M. and Rossi.M. (1995b) In Petersen.S.B., Svensson.B. and
Pedersen,S. (eds), Carbohydrate Bioengineenng. Elsevier Science BV,
Amsterdam, The Netherlands, pp. 77-84.
Namchuk,M.N. and Withers.S.G. (1995) Biochemistry, 34, 16194-16202.
Nucci,R., Moracci.M., Vaccaro,C, Vespa,N. and Rossi.M. (1993) Biolechnol.
Appl. Biochem., 17, 239-250.
Pearl.L.H., Hemmmgs.A.M., Nucci.R and Rossi.M. (1993) J. Mol. Biol., 229,
558-560.
Sinnot.M.L. (1990) Chem. Rev., 90, 1171-1202.
Voorhost,W.G.B., Eggen.R.I.L., Luesink,E.J. and deVos.W.M. (1995) J.
Bacterioi, 1T7, 7105-7111.
Wang.Q., Trimbur.D., Graham.R., Warren.R.AJ. and Withers.S.G (1995)
Biochemistry, 34, 14554-14562.
Wiesmann.C, Beste.G, Hengstenberg.W. and Schulz.G.E. (1995) Structure,
3,961-968.
Withers.S.G., Warrcn.RAJ., Street,I.R, Rupitz,K., KemptonJ.B. and
Aebersold.R. (1990) J. Am. Chem Soc., 112, 5887-5889.
Witt,E., Frank.R. and Hengstenberg.W. (1993) Protein Engng, 6, 913-920.
Received April 24, 1996; revised July 24, 1996; accepted July 25, 1996
Acknowledgements
We thank Mr Giovanni Imperato for his technical assistance, and the helpful
critical reading of Dr Carlo A.Raia is gratefully acknowledged. This work
was supported by CNR Target Project Biotechnology and Bioinstrumentation,
1195