MUTATIONAL MODULATION OF SUBSTRATE BOND-TYPE SPECIFICITY AND THERMOSTABILITY OF GLUCOAMYLASE FROM ASPERGILLUS-AWAMORI BY REPLACEMENT WITH SHORT HOMOLOG ACTIVE-SITE SEQUENCES AND THIOL DISULFIDE ENGINEERING/
Hp. Fierobe et al., MUTATIONAL MODULATION OF SUBSTRATE BOND-TYPE SPECIFICITY AND THERMOSTABILITY OF GLUCOAMYLASE FROM ASPERGILLUS-AWAMORI BY REPLACEMENT WITH SHORT HOMOLOG ACTIVE-SITE SEQUENCES AND THIOL DISULFIDE ENGINEERING/, Biochemistry, 35(26), 1996, pp. 8696-8704
Rational protein engineering based on three-dimensional structure, seq
uence alignment, and previous mutational analysis served to increase t
hermostability and modulate bond-type specificity in glucoamylase from
Aspergillus awamori. The single free cysteine, Cys320, became disulfi
de bonded in the Ala246-->Cys mutant, thus enhancing T-50 by 4 degrees
C to 73 degrees C. Compared to wild-type, Ala246-->Cys was roughly tw
ice as active at 66 degrees C, but half as active at 45 degrees C. The
alternative, elimination of the thiol group in Cys320-->Ala, barely i
mproved thermostability or altered activity. Secondly, to acquire exce
ptionally high specificity toward alpha-1,6 glucosidic linkages, chara
cteristic of Hormoconis resinae glucoamylase, two short sequential mut
ants, Val181-->Thr/Asn182-->Tyr/Gly183-->Ala (L3 glucoamylase) and ro3
07-->Ala/Thr310-->Val/Tyr312-->Met/Asn313-->Gly (L5 glucoamylase), wer
e made. These homologue mutants are located in the (alpha/alpha)(6)-fo
ld of the catalytic domain in segments that connect alpha-helices 5 an
d 6 and alpha-helices 9 and 10, respectively. The kinetics of malto- a
nd isomaltooligosaccharides hydrolysis clearly demonstrated that combi
nation of the mutations in L3L5 compensated adverse effects of the sin
gle replacements in L3 or L5 glucoamylases to yield wild-type or highe
r activity. On alpha-1,4-linked substrates, typically K-m increased 2-
fold for L3, and k(cat) decreased up to 15-fold for L5 glucoamylase. I
n contrast, on alpha-1,6-linked substrates L3 showed both a 2-fold inc
rease in K-m and a 3-fold decrease in k(cat), while L5 GA caused a sim
ilar k(cat) reduction, but up to 9-fold increase in k(m). L3L5 glucoam
ylase had remarkably Low K-m for isomaltotriose through isomaltoheptao
se and elevated k(cat) on isomaltose, resulting in an approximately 2-
fold improved catalytic efficiency (k(cat)/K-m). Rational loop replace
ment thus proved powerful in achieving variants with enhanced properti
es of a highly evolved enzyme.