The HBD of Tma-RNase HI is important not only for substrate binding, but also for Mg2+-dependent activity, probably because the HBD affects the interaction between the substrate and enzy
Trang 1Thermotoga maritima is important for substrate binding
Nujarin Jongruja1, Dong-Ju You1, Eiko Kanaya1, Yuichi Koga1, Kazufumi Takano1,2and
Shigenori Kanaya1
1 Department of Material and Life Science, Graduate School of Engineering, Osaka University, Japan
2 CRESTO, JST, Osaka, Japan
Introduction
Ribonuclease H (RNase H; EC 3.1.26.4) is an enzyme
that specifically cleaves RNA of RNA⁄ DNA hybrids
[1] It requires divalent metal ions, such as Mg2+ and
Mn2+, for activity RNase H is widely present in bac-teria, archaea and eukaryotes These RNase H are involved in DNA replication, repair and transcription
Keywords
cleavage site specificity; hybrid binding
domain; metal preference; RNase H;
substrate binding affinity;
Thermotoga maritima
Correspondence
S Kanaya, Department of Material and Life
Science, Graduate School of Engineering,
Osaka University, 2-1, Yamadaoka, Suita,
Osaka 565-0871, Japan
Fax: +81 6 6879 7938
Tel.: +81 6 6879 7938
E-mail: kanaya@mls.eng.osaka-u.ac.jp
(Received 18 June 2010, revised 6 August
2010, accepted 27 August 2010)
doi:10.1111/j.1742-4658.2010.07834.x
Thermotoga maritima ribonuclease H (RNase H) I (Tma-RNase HI) con-tains a hybrid binding domain (HBD) at the N-terminal region To analyze the role of this HBD, Tma-RNase HI, Tma-W22A with the single mutation
at the HBD, the C-terminal RNase H domain (Tma-CD) and the N-termi-nal domain containing the HBD (Tma-ND) were overproduced in Escheri-chia coli, purified and biochemically characterized Tma-RNase HI prefers
Mg2+to Mn2+for activity, and specifically loses most of the Mg2+ -depen-dent activity on removal of the HBD and 87% of it by the mutation at the HBD Tma-CD lost the ability to suppress the RNase H deficiency of an
E coli rnhA mutant, indicating that the HBD is responsible for in vivo RNase H activity The cleavage-site specificities of Tma-RNase HI are not significantly changed on removal of the HBD, regardless of the metal cofactor Binding analyses of the proteins to the substrate using surface plasmon resonance indicate that the binding affinity of Tma-RNase HI is greatly reduced on removal of the HBD or the mutation These results indicate that there is a correlation between Mg2+-dependent activity and substrate binding affinity Tma-CD was as stable as Tma-RNase HI, indicating that the HBD is not important for stability The HBD of Tma-RNase HI is important not only for substrate binding, but also for
Mg2+-dependent activity, probably because the HBD affects the interaction between the substrate and enzyme at the active site, such that the scissile phosphate group of the substrate and the Mg2+ion are arranged ideally
Abbreviations
Bha-RNase HI, Bacillus halodurans RNase HI; Bst-RNase HIII, Bacillus stearothermophilus RNase HIII; Bsu-RNase HII, Bacillus subtilis RNase HII; D13-R4-D12 ⁄ D29, 29 bp DNA 13 -RNA4-DNA12⁄ DNA duplex; D15-R1-D13 ⁄ D29, 29 bp DNA 15 -RNA1-DNA13⁄ DNA duplex; Eco-RNase HI, Escherichia coli RNase HI; Eco-RNase HII, Escherichia coli RNase HII; GdnHCl, guanidine hydrochloride; HBD, hybrid binding domain; HIV-1 RNase H, RNase H domain of HIV-1 reverse transcriptase; Hsa-RNase H1, human RNase H1; IPTG, isopropyl thio-b- D -galactoside; MMLV RNase H, Moloney murine leukemia virus reverse transcriptase; R12 ⁄ D12, 12 bp RNA ⁄ DNA hybrid; R29 ⁄ D29, 29 bp RNA ⁄ DNA hybrid; R9-D9 ⁄ D18, 18 bp RNA 9 -DNA9⁄ DNA duplex; RNase H, ribonuclease H; Sce-RNase H1, Saccharomyces cerevisiae RNase H1; Sto-RNase HI, Sulfolobus tokodaii RNase HI; Tma-CD, C-terminal catalytic domain (residues 64–223) of Tma-RNase HI; Tma-ND, N-terminal domain (residues 1–63) of RNase HI from Thermotoga maritima containing HBD; Tk-RNase HII, Thermococcus kodakaraensis RNase HII.
Trang 2[2–6] In antisense therapy, RNase H is involved in the
recognition and cleavage of a disease-causative mRNA
[7] Mutations in human RNase H2, which do not
necessarily significantly affect the activity [8], cause
severe neurological disorder termed Aicardi–Goutieres
syndrome [9] RNase H is also present in retroviruses
as a C-terminal domain of reverse transcriptase
Retro-viral RNases H are required for the conversion of
sin-gle-stranded genomic RNA into double-stranded
DNA, which is an initial step of viral proliferation,
and are therefore regarded as one of the targets for
AIDS therapy [10]
RNases H are classified into two major families
(type 1 and type 2 RNases H) based on the difference
in their amino acid sequences [11] Four acidic
active-site residues are fully conserved in these RNases H,
except that from compost metagenome [12], and their
geometrical configurations are well conserved [13]
According to the crystal structures of the C-terminal
catalytic domains of Bacillus halodurans RNase HI
(Bha-RNase HI) [14] and human RNase H1 [15] in
complex with the RNA⁄ DNA substrate, type 1 RNase
H binds to the minor groove of the substrate, such
that one depression containing the active site interacts
with the RNA backbone and the other depression
con-taining the phosphate-binding pocket interacts with
the DNA backbone These two depressions are
sepa-rated by a ridge, which is composed of three highly
conserved Asn⁄ Gln residues Because two metal ions
are coordinated by the four acidic active site residues,
the scissile phosphate group of the substrate and water
molecules, a two-metal ion catalysis mechanism has
been proposed for RNase H [14,16,17] According to
this mechanism, one metal ion is required for
sub-strate-assisted nucleophile formation and product
release, whereas the other is required to destabilize the
enzyme–substrate complex and thereby promote the
phosphoryl transfer reaction
Thermotoga maritima is a strictly anaerobic,
extre-mely thermophilic eubacterium, isolated from various
geothermally heated locales on the sea floor, and
grows in the temperature range 55–90C with an
opti-mum at 80C [18] Its genome sequence has been
determined previously [19] The genome contains single
rnhA and single rnhB genes, encoding type 1
(Tma-RNase HI; accession no AAD36370) and type 2
(Tma-RNase HII; accession no AAD35996) RNases
H, respectively Tma-RNase HI is composed of 223
amino acid residues and contains a hybrid binding
domain (HBD) at the N-terminal region Without this
domain, Tma-RNase HI shows relatively low (£ 20%)
amino acid sequence identities to any one of the
repre-sentative members of type 1 RNases H, which have
been biochemically characterized Therefore, it would
be informative to characterize Tma-RNase HI and compare its biochemical properties with those of other type 1 RNases H
HBD, previously termed as double-stranded RNA and hybrid binding domain [20], consists of approxi-mately 40 amino acid residues and is commonly present at the N-terminal regions of eukaryotic type 1 RNases H (RNase H1) [21] According to the crystal structure of the HBD of human RNase H1 in complex with the RNA⁄ DNA substrate, HBD consists of a three-stranded anti-parallel b-sheet (b1–b3) and two helices (aA and aB) [22] It binds to the minor groove
of the substrate, such that a loop between aA and b3 interacts with the RNA backbone and a positively-charged depression interacts with the DNA backbone The importance of HBD with respect to substrate binding has been reported for yeast [20,23], mouse [24] and human [22,25,26] RNases H1 The requirement of HBD for processivity [24] and positional preference [25,26] has also been reported for mouse and human RNases H1, respectively
HBD is also present in several bacterial type 1 RNases H, including Tma-RNase HI, Bha-RNase HI and RBD-RNase HI from Shewanella sp SIB1 [13] However, it remains to be determined whether these HBDs have a role similar to those of eukaryotic RNases H1, although the isolated HBD from SIB1 RBD-RNases HI, which is renamed as SIB1 HBD-RNase HI in the present study, has been reported to bind to the RNA⁄ DNA substrate [27] Attempts to overproduce the SIB1 HBD-RNase HI derivative lack-ing the HBD have so far been unsuccessful, probably
as a result of the instability of the protein (T Tadok-oro, unpublished data) In the present study, we over-produced, purified and biochemically characterized Tma-RNase HI and its derivatives lacking the HBD or RNase H domain On the basis of the results obtained,
we discuss the role of the HBD from Tma-RNase HI
Results
Protein preparations The amino acid sequence of Tma-RNase HI is com-pared with those of the representative members of type
1 RNases H, Bha-RNase HI, HBD-RNase HI from Shewanella sp SIB1, Saccharomyces cerevisiae RNase H1 (Sce-RNase H1), human RNase H1 (Hsa-RNase H1), E coli RNase HI (Eco-RNase HI) and the RNase H domain of HIV-1 reverse transcriptase
(HIV-1 RNase H) in Fig 1 The HBD of Tma-RNase HI shows relatively high amino acid sequence identities of
Trang 343%, 42%, 33% and 32% with respect to those of
Sce-RNase H1, Bha-RNase HI, SIB1 HBD-RNase HI
and Hsa-RNase H1, whereas the RNase H domain of
Tma-RNase HI shows relatively low amino acid
sequence identities of 20% to Hsa-RNase H1, 19% to
Eco-RNase HI, 18% to SIB1 HBD-RNase HI,
Sce-RNase H1 and Bha-Sce-RNase HI, and 17% to HIV-1
RNase H Nevertheless, all active-site residues (four
acidic and one histidine residues) are fully conserved in
Tma-RNase HI as Asp71, Glu111, Asp135, His179
and Asp189
To analyze the role of the HBD of Tma-RNase HI,
the Tma-RNase HI derivatives lacking either the HBD
(Tma-CD, residues 64–223) or the RNase H domain
(Tma-ND, residues 1–63) were constructed Tma-ND
contains the entire HBD of Tma-RNase HI (Fig 1) Tma-RNase HI, Tma-CD and Tma-ND were over-produced in the rnhA deficient strain E coli MIC3001(DE3) to avoid a contamination of host-derived RNase HI Upon overproduction, these pro-teins accumulated in E coli cells in a soluble form and were purified to give a single band on SDS⁄ PAGE (Fig 2) The amount of the protein purified from 1L culture was approximately 35 mg for Tma-RNase HI,
15 mg for Tma-CD and 30 mg for Tma-ND The molecular masses of these proteins were estimated to be
28 kDa for Tma-RNase HI, 16 kDa for Tma-CD and
8 kDa for Tma-ND by gel filtration column chromato-graphy These values are comparable to those calculated from the amino acid sequences (25 967 for Tma-RNase
Fig 1 Alignment of the amino acid sequences The amino acid sequence of Tma-RNase HI (Tma) is compared with those of Bha-RNase HI (Bha), SIB1 HBD-RNase HI (SIB1HBD), Sce-RNase H1 (Sce), Hsa-RNase H1 (Hsa), Eco-RNase HI (Eco) and HIV-1 RNase H (HIV1) The accession numbers are AAD36370 for Tma-RNase HI, BAF73617 for SIB1 HBD-RNase HI, DAA10134 for Sce-RNase H1, EAX01061 for Hsa-RNase H1, P0A7Y4 for Eco-Hsa-RNase HI and ABU62661 for HIV-1 Hsa-RNase H The ranges of the secondary structures of Hsa-Hsa-RNase H1 are shown above the sequence, based on the crystal structures of its HBD (Protein Data Bank code: 3BSU) and RNase H domain (Protein Data Bank code: 2KQ9), which were independently determined in complex with the substrate The range of HBD is also shown The amino acid residues, which are conserved in at least three (for HBD) or four (for RNase H domain) different proteins, are highlighted in black The five active-site residues are denoted by filled circles above the sequences The amino acid residues that contact the substrate in the co-crystal structure of the HBD of Hsa-RNase H1 with the substrate are also denoted by open circles above the sequence The amino acid residue that
is mutated in the present study is indicated by an arrow Gaps are denoted by dashes The numbers represent the positions of the amino acid residues relative to the initiator methionine for each protein.
Trang 4HI, 18 860 for Tma-CD and 7107 for Tma-ND),
sug-gesting that all proteins exist as a monomer in solution
CD spectra
The far- and near-UV CD spectra of Tma-RNase HI,
Tma-CD and Tma-ND were measured at 20C and
pH 9.0, and comparisons are shown inFig 3 The
far-and near-UV CD spectra of Tma-CD are similar to
those of Tma-RNase HI, suggesting that removal
of the HBD does not significantly affect the structure
of the RNase H domain of Tma-RNase HI The
far- and near-UV CD spectra of Tma-ND were
sig-nificantly different from those of Tma-RNase HI,
probably because the secondary structure contents and
environment of the aromatic residues are different in
these proteins According to the crystal structures of
the HBD [22] and RNase H domain [15] of
Hsa-RNase HI, the b-strand contents are 37% for HBD
and 21% for RNase H domain, whereas the a-helix
contents of these domains are similar to each other
(39% for HBD and 38% for RNase H domain)
Enzymatic activity
The dependencies of the RNase HI and
Tma-CD activities on pH, salt and metal ion were
analyzed at 30C by changing one of the conditions used for assay [10 mm Tris⁄ HCl, 1 mm MgCl2,
50 mm KCl (pH 9.0) for Tma-RNase HI, and 10 mm Tris⁄ HCl, 1 mm MnCl2, 10 mm KCl (pH 9.0) for Tma-CD] The M13 DNA⁄ RNA hybrid was used as
a substrate The enzymatic activities of these proteins were determined at the temperature (30C), which could be much lower than the optimum one because the substrate used for assay is not fully stable at
‡ 60 C When the enzymatic activity was determined over the range pH 5–12, both proteins exhibited the highest activities at around pH 9.0 (data not shown) They exhibited approximately 50% of the maximal activities at pH 7.0 and 11.0 When the enzymatic activity was determined in the presence of various concentrations of NaCl or KCl, Tma-RNase HI exhibited the highest activity in the presence of
50 mm KCl, whereas Tma-CD exhibited it in the presence of 10 mm KCl (Fig 4) Their enzymatic
Fig 2 SDS ⁄ PAGE of Tma-RNase HI and its derivatives The
purified proteins of Tma-RNase HI (lane 1), Tma-CD (lane 2) and
Tma-ND (lane 3) were subjected to electrophoresis on a 15%
poly-acrylamide gel in the presence of SDS After electrophoresis, the
gel was stained with Coomassie Brilliant Blue Lane M, a
low-molecular-weight marker kit (GE Healthcare, Tokyo, Japan).
Fig 3 CD spectra of Tma-RNase HI and its derivatives Far-UV (A) and near-UV (B) CD spectra of Tma-RNase HI (thick solid dark line), Tma-W22A (thin solid dark line), Tma-CD (dashed dark line) and Tma-ND (thick solid gray line) are shown These spectra were mea-sured at pH 9.0 and 20 C, as described in the Experimental proce-dures.
Trang 5activities decreased to a large extent at higher
(‡ 0.2 m) salt concentrations When the enzymatic
activity was determined in the presence of various
concentrations of MgCl2, MnCl2, NiCl2, ZnCl2,
CoCl2 or CaCl2, both Tma-RNase HI and Tma-CD
exhibited the highest activities in the presence of
1 mm MgCl2 and 0.1–5 mm MnCl2 (Fig 5) Both
proteins exhibited little activity (less than 0.01% of
the maximal activity) in the presence of NiCl2, ZnCl2,
CoCl2 or CaCl2 The maximal Mg2+- and Mn2+
-dependent activities of these proteins are summarized
in Table 1 Tma-RNase HI prefers Mg2+ to Mn2+
because its maximal Mg2+-dependent activity is
higher than its maximal Mn2+-dependent activity by
16-fold By contrast, Tma-CD prefers Mn2+ to Mg2+
because its maximal Mn2+-dependent activity is
higher than its maximal Mg2+-dependent activity by
69-fold Interestingly, the maximal Mn2+-dependent
activity of CD is comparable to that of
Tma-RNase HI These results indicate that removal of the HBD severely reduces the Mg2+-dependent activity of Tma-RNase HI without significantly affecting its
Mn2+-dependent activity
The kinetic parameters of Tma-CD were determined
at 30C in the presence of 1 mm MgCl2or MnCl2and compared with those of Tma-RNase HI (Table 1) The
Vmax values of Tma-CD determined in the presence of
1 mm MgCl2 and MnCl2 were 410-fold lower and 1.6-fold higher than those of Tma-RNase HI The Km values of Tma-CD determined in the presence of 1 mm MgCl2 and MnCl2 were 5.1- and 6.8-fold higher than those of Tma-RNase HI These results indicate that the substrate binding affinity of Tma-RNase HI is reduced by five- to seven-fold on removal of the HBD, regardless of the metal cofactor, and the large reduc-tion in Mg2+-dependent activity on removal of the HBD is not a result of the marked decrease in substrate binding affinity
Fig 4 Salt dependencies of Tma-RNase HI and Tma-CD The
enzy-matic activities of Tma-RNase HI (A) and Tma-CD (B) were
deter-mined at 30 C in 10 m M Tris ⁄ HCl (pH 9.0) containing 1 m M MgCl2
(Tma-RNase HI) or 1 m M MnCl2 (Tma-CD), 1 m M
b-mercaptoe-thanol, 50 lgÆmL)1BSA, and various concentrations of NaCl (open
circle) or KCl (closed circle), using M13 DNA ⁄ RNA hybrid as a
sub-strate Experiments were carried out at least twice and the average
values are shown together with the errors.
Fig 5 Metal ion dependencies of Tma-RNase HI and Tma-CD The enzymatic activities of Tma-RNase HI (A) and Tma-CD (B) were determined at 30 C in 10 m M Tris ⁄ HCl (pH 9.0) containing
50 m M KCl (Tma-RNase HI) or 10 m M KCl (Tma-CD), 1 m M
b-mercaptoethanol, 50 lgÆmL)1BSA, and various concentrations of MgCl2(open circle) or MnCl2(closed circle), using M13 DNA ⁄ RNA hybrid as a substrate Experiments were carried out at least twice and the average values are shown together with the errors.
Trang 6Complementation assay
E coli MIC3001 shows an RNase H-dependent
temperature-sensitive growth phenotype [28] E coli
MIC3001(DE3) also displays this phenotype To
exam-ine whether the genes encoding Tma-RNase HI and
Tma-CD complement the temperature-sensitive growth
phenotype of MIC3001(DE3), E coli MIC3001(DE3)
transformants for overproduction of these proteins
were grown in the absence of isopropyl
thio-b-d-galac-toside (IPTG) at permissive (30C) and nonpermissive
(42C) temperatures The results showed that the
Tma-RNase HI gene complements the
temperature-sensitive growth phenotype of MIC3001(DE3),
whereas the Tma-CD gene does not (data not shown)
These results suggest that HBD is required for in vivo
function of RNase HI It is unlikely that
Tma-CD is not produced or produced in a nonfunctional
form in E coli cells in the absence of IPTG because
the protein is overproduced in a soluble and functional
form upon overproduction, as noted above
Cleavage-site specificity
The cleavage-site specificities of Tma-RNase HI and
Tma-CD were analyzed by using 12 bp RNA⁄ DNA
hybrid (R12⁄ D12), 29 bp DNA13-RNA4-DNA12⁄
DNA duplex (D13-R4-D12⁄ D29), 29 bp DNA15
-RNA1-DNA13⁄ DNA duplex (D15-R1-D13 ⁄ D29) and
18 bp RNA9-DNA9⁄ DNA duplex (R9-D9 ⁄ D18) For
comparative purposes, these substrates were cleaved
by Eco-RNase HI, Sulfolobus tokodaii RNase HI
(Sto-RNase HI) and Thermococcus kodakaraensis
RNase HII (Tk-RNase HII) as well D13-R4-D12
and D15-R1-D13 are the chimeric oligonucleotides,
in which four and single ribonucleotides are flanked
by 12–15 bp of DNA at both sides R9-D9⁄ D18 is a
Okazaki fragment-like substrate, in which the 18 base
chimeric oligonucleotide (RNA9-DNA9) is hybridized
to the 18 base complementary DNA
Cleavage of the R12⁄ D12 substrate with various RNase H enzymes is summarized in Fig 6A,B Tma-RNase HI, Eco-Tma-RNase HI, Sto-Tma-RNase HI and Tk-RNase HII cleaved this substrate at multiple sites, although with different site specificities Tma-RNase
HI cleaved this substrate slightly more efficiently in the presence of Mg2+than in the presence of Mn2+
Tma-CD cleaved this substrate with much less and compa-rable efficiencies compared to those of Tma-RNase HI
in the presence of Mg2+ and Mn2+, respectively These results are consistent with those obtained by using M13 DNA⁄ RNA as a substrate The cleavage sites of the R12⁄ D12 substrate with Tma-CD are simi-lar to those with Tma-RNase HI, regardless of the metal cofactor, although their preferable cleavage sites are slightly different with each other The cleavage sites of this substrate and their susceptibilities to cleav-age with Eco-RNase HI, Sto-RNase HI, and Tk-RNase HII are essentially the same as those reported previously [27–30]
Cleavage of the D13-R4-D12⁄ D29 substrate with various RNase H enzymes is summarized in Fig 6C,D Tma-RNase HI, Eco-RNase HI, Sto-RNase HI and Tk-RNase HII cleaved this substrate most preferably
at a16-a17, a15-a16, a14-a15 and a16-a17, respectively The cleavage sites of this substrate with Eco-RNase
HI and Tk-RNase HII are the same as those reported previously [30] The a16-a17 site has been reported to
be exclusively cleaved only by type 2 RNases H, except for bacterial RNases HIII [31,32] Therefore, Tma-RNase HI is the first type 1 Tma-RNase H enzyme that exclusively cleaves this substrate at this site Tma-CD also cleaved this substrate at a16-a17 with a similar efficiency to that of Tma-RNase HI However, these enzymes cleaved this substrate only in the presence of
Mn2+
Table 1 Specific activities and kinetic parameters of Tma-RNase HI and its derivatives Hydrolysis of the M13 DNA ⁄ RNA hybrid by the enzyme was carried out at 30 C under the conditions described in the Experimental procedures ND, not determined.
Specific activity (UÆmg)1)
Relative activity a (%) Km(l M ) Vmax(UÆmg)1)
a The specific activities of the proteins relative to that of Tma-RNase HI determined in the presence of 1 m M MgCl2and 50 m M KCl.
Trang 8The D15-R1-D13⁄ D29 substrate was used to
con-firm that Tma-RNase HI and Tma-CD do not cleave
the DNA-RNA-DNA⁄ DNA substrate containing a
single ribonucleotide This substrate is not cleaved by
type 1 RNases H but is cleaved by type 2 RNases H,
except for bacterial RNases HIII, at the DNA-RNA
junction [21,27,32] As expected, this substrate was not
cleaved with Tma-RNase HI, Sto-RNase HI and
Eco-RNase HI, although it was cleaved with Tk-Eco-RNase
HII at the DNA-RNA junction (data not shown)
These results exclude the possibility that the cleavage
of the D13-R4-D12⁄ D29 substrate with Tma-RNase
HI at a16-a17 is caused by the contamination of a type
2 RNase H enzyme
Cleavage of the R9-D9⁄ D18 substrate with various
RNase H enzymes is summarized in Fig 6E,F
Tma-RNase HI and Tma-CD cleaved this substrate most
preferably at g7-c8 and c8-c9, and much less preferably
at u6-g7 and c9-T10 in the presence of Mn2+ They
cleaved this substrate with similar site specificities in
the presence of Mg2+ However, their abilities to
cleave this substrate are greatly reduced in the presence
of Mg2+ by more than 100-fold Eco-RNase HI and
Sto-RNase HI cleaved this substrate at all sites
between a5 and c9 and between a5 and T10,
respec-tively, as reported previously [33] However, both
enzymes showed a preference for the sites far from the
RNA-DNA junction (a5-u6, u6-g7 and g7-c8 for
Eco-RNase HI, and u5-a6 and u6-g7 for Sto-Eco-RNase HI)
Tk-RNase HII cleaved this substrate almost exclusively
at c8-c9 Eco-RNase HI and Tk-RNase HII cleaved
the RNA-DNA junction (c9-T10) as well, although
with very poor efficiency
It has been demonstrated for mouse RNase H1 that
the HBD is required for processivity of the enzyme
[24] Tma-RNase HI did not show the processivity
for cleavage of the R12⁄ D12 substrate (Fig 6)
How-ever, this result does not necessarily indicate that
Tma-RNase HI shows no processivity because mouse
RNase H1 shows the processivity only for long
RNA⁄ DNA substrates Therefore, it would be
infor-mative to examine whether Tma-RNase HI shows processivity for long RNA⁄ DNA substrates and loses this processivity on removal of the HBD
Binding to substrate
To examine whether the HBD of Tma-RNase HI is important for substrate binding, the binding affinities
of Tma-RNase HI, Tma-CD and Tma-ND to the
29 bp RNA⁄ DNA hybrid (R29 ⁄ D29) were analyzed in the absence of the metal cofactor using surface plas-mon resonance These proteins were injected onto the sensor chip, on which the R29⁄ D29 substrate was immobilized The sensorgrams obtained by injecting
1 lm of these proteins are shown in Fig 7 The disso-ciation constants, KD, of the proteins for binding to the R29⁄ D29 substrate, which were determined by measuring the equilibrium-binding responses at various concentrations of the proteins, are summarized in
Table 2 The KD value of Tma-ND was higher than (although comparable to) that of Tma-RNase HI By
Fig 7 Binding of Tma-RNase HI and its derivatives to the sub-strate Sensorgrams from Biacore X showing binding of 1 l M of Tma-RNase HI (thick solid dark line), Tma-W22A (thin solid dark line), Tma-CD (dashed dark line) and Tma-ND (thick solid gray line)
to the immobilized R29 ⁄ D29 substrate are shown Injections were performed at time zero for 60 s.
Fig 6 Cleavage of various oligomeric substrates with various RNases H (A, C, E) Separation of the hydrolysates by urea gel The 5¢-end labeled R12 ⁄ D12 (A), 5¢-end labeled D13-R4-D12 ⁄ D29 (C) and 3¢-end labeled R9-D9 ⁄ D18 (E) were hydrolyzed by the enzyme at 30 C for
15 min and the hydrolysates were separated on a 20% polyacrylamide gel containing 7 M urea, as described in the Experimental procedures The concentration of the substrate was 1.0 l M The amount of the enzyme added to the reaction mixture (10 lL) is indicated above each lane The metal cofactors used to cleave these substrates with Tma-RNase HI and Tma-CD are also shown above the gel together with their concentrations The complete sequence of R12 (A) as well as the partial sequences of D13-R4-D12 (C) and R9-D9 (E) are indicated along the gel (B, D, F) Schematic representation of the sites and extents of cleavage by various RNases H Cleavage sites of R12 ⁄ D12 (B), D13-R4-D12 ⁄ D29 (D) and R9-D9 ⁄ D18 (F) by the enzyme are shown by arrows In these panels, only the sequences of the oligonucleotides cleaved
by the enzyme are shown The differences in the lengths of the arrows reflect relative cleavage intensities at the position indicated These lengths do not necessarily reflect the amount of the products accumulated upon complete hydrolysis of the substrate Deoxyribonucleotides are indicated by capital letters and ribonucleotides are indicated by lowercase letters.
Trang 9contrast, the KD value of Tma-CD was considerably
higher than that of Tma-RNase HI by 49-fold These
results indicate that the HBD of Tma-RNase HI is
important for substrate binding When binding of
Tma-RNase HI to the R29⁄ D29 substrate was
analyzed in the presence of 0.5 m NaCl, only a small
positive signal was observed, even when 4 lm of the
protein was injected, indicating that the binding
affin-ity of Tma-RNase HI to the substrate is severely
decreased at high salt concentration
Thermal stability
To examine whether removal of the N- or C-terminal
domain affects the stability of Tma-RNase HI, the
thermal stabilities of Tma-RNase HI, Tma-CD and
Tma-ND were determined by monitoring changes of
the CD values at 222 nm In the presence of 3 m
guanidine hydrochloride (GdnHCl) and 10 mm
dith-iothreitol at pH 9, all proteins unfolded in a single
cooperative fashion in a reversible manner The
ther-mal denaturation curves of these proteins are
com-pared with one another in Fig 8 The parameters
characterizing the thermal denaturation of these
pro-teins are summarized in Table 2 A comparison of
these parameters indicates that Tma-CD and Tma-ND
are less stable than Tma-RNase HI by 1.3 and 10.8C
in Tm, respectively These results suggest that the
inter-actions between the N- and C-terminal domains of
Tma-RNase HI do not significantly contribute to the
stabilization of its C-terminal domain but contribute
to the stabilization of its N-terminal domain
Tma-RNase HI is thermally denatured in a single
coopera-tive fashion, probably because its N-terminal domain
is denatured immediately after its C-terminal RNase H
domain is denatured It is noted that the DHm and
DSm values of Tma-CD are considerably higher than
those of Tma-RNase HI and Tma-ND, which are
comparable to each other The reason why the DHm and DSm values of Tma-RNase HI increase on removal of the N-terminal domain remains to be clarified
Analysis for interaction between two domains
To examine whether the HBD of Tma-RNase HI strongly interacts with the RNase H domain, Tma-ND was mixed with Tma-CD in a 1 : 1 molar ratio and applied to gel filtration column chromatography Both proteins were eluted from the column as two inde-pendent peaks (data not shown), indicating that Tma-ND does not form a stable complex with
Table 2 Dissociation constants and parameters characterizing thermal denaturation of Tma-RNase HI and its derivatives ND, not determined.
a Dissociation constant of the protein for binding to the R29 ⁄ D29 substrate was determined by measuring equilibrium-binding responses at various concentrations of the protein using surface plasmon resonance (Biacore) as described in the Experimental procedures b Parameters characterizing thermal denaturation of the proteins were determined from the thermal denaturation curves shown in Fig 8 The melting tem-perature (Tm) is temperature of the midpoint of the thermal denaturation transition DTmis the difference in Tmbetween the intact and trun-cated proteins and is calculated as: Tm(truncated) ) T m (intact) DHmand DSm are the enthalpy and entropy changes of unfolding at Tm calculated by van’t Hoff analysis.
Fig 8 Thermal denaturation curves Thermal denaturation curves
of Tma-RNase HI (closed circl), Tma-CD (open square) and Tma-ND (closed triangle) are shown These curves were obtained at pH 9.0
in the presence of 3 M GdnHCl and 10 m M dithiothreitol by monitor-ing the change in the CD value at 222 nm, as described in the Experimental procedures The theoretical curves are drawn on the assumption that the proteins are denatured via a two-state mechanism.
Trang 10Tma-CD In addition, the Mg2+-dependent activity of
Tma-CD was not significantly changed in the presence
of a 10–1000 molar excess of Tma-ND, indicating that
the Mg2+-dependent activity of Tma-CD is not
restored in the presence of Tma-ND It has been
pro-posed for eukaryotic type 1 RNases H that the HBD
and RNase H domain are separated by a flexible linker
and move rather freely [21] The HBD of Tma-RNase
HI also may not strongly interact with the RNase H
domain
Biochemical properties of Tma-W22A
According to the crystal structure of the HBD of
Hsa-RNase H1 in complex with the substrate, Tyr29,
Trp43, Phe58, Lys59 and Lys60 interact with the DNA
strand of the substrate [22] These residues, except for
Lys60, are well conserved in various HBDs, suggesting
that the HBDs of other type 1 RNases H bind to the
substrate in a manner similar to the interaction of the
HBD of Hsa-RNase H1 The mutation of Trp43 to
Ala greatly reduces both the substrate binding affinities
and enzymatic activities of Hsa-RNase H1 [22,26] and
mouse RNase H1 [24] To examine whether the
corre-sponding tryptophan residue (Trp22) is important for
substrate binding and enzymatic activity of
Tma-RNase HI, the mutant protein, Tma-W22A, was
con-structed, overproduced and purified The production
level and purification yield of Tma-W22A were
compa-rable to those of Tma-RNase HI The far- and
near-UV CD spectra of Tma-W22A are similar to those of
Tma-RNase HI (Fig 3), suggesting that the mutation
at Trp22 does not significantly affect the structure of
Tma-RNase HI
The pH, salt and metal ion dependencies of
Tma-W22A were similar to those of Tma-RNase HI (data
not shown) Its maximal Mn2+-dependent activity was
also similar to that of Tma-RNase HI (Table 1)
How-ever, its maximal Mg2+-dependent activity was lower
than that of Tma-RNase HI by 7.5-fold (Table 1),
indicating that the mutation of Trp22 to Ala
con-siderably reduces the Mg2+-dependent activity of
Tma-RNase HI without significantly affecting the
Mn2+-dependent activity The binding affinity of
Tma-W22A to the R29⁄ D29 substrate was analyzed in
the absence of the metal cofactor using surface
plas-mon resonance and compared with that of Tma-RNase
HI The KDvalue of Tma-W22A was higher than that
of Tma-RNase HI by 21-fold, suggesting that Trp22 of
Tma-RNase HI is involved in substrate binding These
results suggest that the HBD of Tma-RNase HI
acts with the substrate in a manner similar to the
inter-action of the HBD of Hsa-RNase H1
The cleavage site specificities of Tma-W22A were not analyzed because the cleavage site specificities of Tma-RNase HI are not significantly changed even when its N-terminal domain is removed, and therefore
it is highly likely that the cleavage site specificities of Tma-W22A are similar to those of Tma-RNase HI Likewise, the stability of Tma-W22A was not analyzed because the stability of Tma-W22A is not significantly changed even when the HBD is completely removed, and therefore it is highly likely that Tma-W22A is as stable as Tma-RNase HI
Discussion
Importance of HBD for substrate binding
In the present study, we showed that the HBD of Tma-RNase HI is important for substrate binding However, on removal of the HBD, the Km value of Tma-RNase HI increases by 5–7-fold, whereas its KD value increases by 49-fold Because the Kmand KD val-ues are determined in the presence and absence of the metal cofactor, these results suggest that the difference
in substrate binding affinity between Tma-RNase HI and Tma-CD determined in the presence of the metal cofactor is smaller than that determined in its absence Presumably, the HBD governs binding of Tma-RNase
HI to the substrate and its substrate binding affinity is not significantly changed either in the presence or absence of the metal cofactor By contrast, the sub-strate binding affinity of the RNase H domain proba-bly increases in the presence of the metal cofactor compared to that in its absence The cleavage-site spec-ificity of Tma-RNase HI is not significantly changed
on removal of the HBD, probably because the HBD
of Tma-RNase HI facilitates initial nonsite-specific interactions with the substrate and promotes the site-specific interactions between the RNase H domain of Tma-RNase HI and substrate
Importance of HBD for Mg2+-dependent activity Removal of the HBD severely reduces the Mg2+ -dependent activity of Tma-RNase HI by 750-fold without significantly affecting the Mn2+-dependent activity Similarly, single mutation at the HBD (Trp22 to Ala) reduces the Mg2+-dependent activity
of Tma-RNase HI by 7.5-fold without significantly affecting the Mn2+-dependent activity Removal of the HBD and single mutation at the HBD reduces the binding affinity of Tma-RNase HI by 49- and 21-fold, respectively Thus, there is a correlation between the Mg2+-dependent activity of Tma-RNase HI and