There-fore, in addition to being involved in regulation of purine intracellular pool, the enzyme is also responsible Keywords cN-II active site; cN-II regulatory sites; cN-II structure;
Trang 1Rossana Pesi1, Simone Allegrini2, Maria Giovanna Careddu1,2, Daniela Nicole Filoni1,
Marcella Camici1 and Maria Grazia Tozzi1
1 Dipartimento di Biologia, Unita` di Biochimica, Universita` di Pisa, Pisa, Italy
2 Dipartimento di Scienze del Farmaco, Universita` di Sassari, Sassari, Italy
Introduction
Cytosolic 5’-nucleotidase (cN-II) is a ubiquitous enzyme
that catalyses either the hydrolysis or the transfer of
phosphate esterified in the 5¢ position of
6-hydroxypu-rine monophosphate nucleosides [1] The transfer of
phosphate can lead to phosphorylation of inosine, guanosine and a number of their analogues [2] There-fore, in addition to being involved in regulation of purine intracellular pool, the enzyme is also responsible
Keywords
cN-II active site; cN-II regulatory sites;
cN-II structure; cytosolic 5¢-nucleotidase II
Correspondence
M G Tozzi, Dipartimento di Biologia,
Via S Zeno 51, Pisa, Italy
Fax: +39 502211450
Tel: +39 502211457
E-mail: mtozzi@biologia.unipi.it
(Received 19 July 2010, revised
10 September 2010, accepted
21 September 2010)
doi:10.1111/j.1742-4658.2010.07891.x
Cytosolic 5¢-nucleotidase (cN-II), which acts preferentially on 6-hydroxypu-rine nucleotides, is essential for the survival of several cell types cN-II catalyses both the hydrolysis of nucleotides and transfer of their phosphate moiety to a nucleoside acceptor through formation of a covalent phospho-intermediate Both activities are regulated by a number of phosphorylated compounds, such as diadenosine tetraphosphate (Ap4A), ADP, ATP, 2,3-bisphosphoglycerate (BPG) and phosphate On the basis of a partial crystal structure of cN-II, we mutated two residues located in the active site, Y55 and T56 We ascertained that the ability to catalyse the transfer
of phosphate depends on the presence of a bulky residue in the active site very close to the aspartate residue that forms the covalent phospho-intermediate The molecular model indicates two possible sites at which adenylic compounds may interact We mutated three residues that mediate interaction in the first activation site (R144, N154, I152) and three in the second (F127, M436 and H428), and found that Ap4A and ADP interact with the same site, but the sites for ATP and BPG remain uncertain The structural model indicates that cN-II is a homotetrameric protein that results from interaction through a specific interface B of two identical dimers that have arisen from interaction of two identical subunits through interface A Point mutations in the two interfaces and gel-filtration experi-ments indicated that the dimer is the smallest active oligomerization state Finally, gel-filtration and light-scattering experiments demonstrated that the native enzyme exists as a tetramer, and no further oligomerization is required for enzyme activation
Structured digital abstract
l MINT-8011572 : cN-II (uniprotkb: O46411 ) and cN-II (uniprotkb: O46411 ) bind ( MI:0407 ) by dynamic light scattering ( MI:0038 )
l MINT-8011493 , MINT-8011481 : cN-II (uniprotkb: O46411 ) and cN-II (uniprotkb: O46411 ) bind ( MI:0407 ) by molecular sieving ( MI:0071 )
Abbreviations
cN-II, cytosolic 5¢-nucleotidase; cN-III, cytosolic 5¢-nucleotidase III; cN-IA, cytosolic 5¢-nucleotidase IA; cN-IB, cytosolic 5¢-nucleotidase IB; cdN, cytosolic 5¢(3¢)-deoxyribonucleotidase; mdN, mitochondrial 5¢(3¢)-deoxyribonucleotidase.
Trang 2for pro-drug activation and inactivation [3,4] It has
been demonstrated that the catalytic mechanism of
cN-II requires formation of a covalent
phospho-inter-mediate on an aspartate residue located in a conserved
motif (motif I) [5] This motif, together with three
other conserved motifs, is shared among the members
of the haloacid dehalogenase (HAD) superfamily,
including the soluble 5¢-nucleotidase family: cytosolic
5¢-nucleotidases cN-II, cN-III, cN-IA and cN-IB, and
both cytosolic and mitochondrial
5¢(3¢)-deoxyribonu-cleotidases [5] Even though all soluble 5¢-nu5¢(3¢)-deoxyribonu-cleotidases
share the same reaction mechanism and possess
con-served structural motifs in the catalytic site, only two
members of this family possess phosphotransferase
activity, namely cN-II and cN-III
cN-II has several unique aspects, such as its complex
regulation and the very high degree of primary
sequence conservation during evolution [5] These
aspects indicate that this enzyme plays an important
role cN-II knockdown through RNAi causes
apopto-sis in cultured cells [6] Furthermore, over-expression
of cN-II by more than 10-fold in HEK293 cells has
proved impossible to achieve, probably because of an
adverse effect on cell viability [7] The reaction rates
of both nucleotide hydrolysis and phosphate transfer
catalysed by cN-II appear to be regulated by a number
of phosphorylated compounds such as ADP, ATP,
BPG, Ap4A and polyphosphates These compounds
act as allosteric activators ATP causes an increase in
Vmax of approximately 10-fold, with little effect on Km
for the substrate IMP [8,9]
Free inorganic phosphate, on the other hand, acts as
an allosteric inhibitor, causing a 20-fold increase in Km
for the substrate IMP with little effect on Vmax
Inter-estingly, ATP partially counteracts the effect of
phos-phate, by increasing Vmax; however, it is unable to
reverse the increase in Km[9] cN-II has been described
as a homotetramer with the ability to change its
oligo-merization state in response to the presence of
activa-tors or inhibiactiva-tors It has been suggested that the
change in the oligomerization state is accompanied by
a change in specific activity [10] However, this simple
model does not explain the kinetic evidence described
above
Despite its cytosolic location, cN-II has particularly
poor solubility This is why it has been difficult to
obtain the crystal structure of the whole protein A
truncated form of cN-II lacking the last 25 amino
acids is significantly more soluble than the wild-type
enzyme, and was recently crystallized [11] The
crystal-lographic model, constructed by ordering 487 residues
(1-400 and 417-488) out of 561, indicates a
homotetra-meric protein resulting from interaction through a
specific interface (interface B) of two identical dimers that arise from interaction of two identical subunits through interface A Because of the presence of adeno-sine bound to the crystallized protein, Wallde´n et al [11] identified two putative effector sites, and suggested that effector site 1, close to interface A, might interact with BPG and Ap4A, while ATP and ADP might bind
to effector site 2
In a previous paper, two active forms of cN-II were identified in extracts from different organs [12] The two forms purified from calf thymus showed different behav-iour with activators The heavier form (form A) has a high-affinity regulatory site for BPG, while ADP and ATP share a different site The lighter form (form B) has three different sites for the three activators [5,12] Although many papers have been published in the last few years describing structural and functional fea-tures of cN-II, fundamental aspects of how the enzyme functions remain to be unravelled, in particular the number and location of the interaction sites for the activators and inhibitors, the relationship between activity and enzyme oligomerization state, and, finally, the amino acid residue(s) responsible for phospho-transferase ability of cN-II We have utilized a mecha-nistic approach in order to increase knowledge on these topics
Results Active site and phosphotransferase reaction Figure 1 shows the aligned sequences of motif I for the six intracellular 5¢-nucleotidases Of the six enzymes, only cN-II and cN-III possess a Thr instead of a Val (position 56 of cN-II, boxed in Fig 1) [11] These are the only two enzymes for which phosphotransferase activity has been unquestionably ascertained, and Wallde´n et al [11] suggested that the presence of T56
Fig 1 Aligned conserved motif I of the six known intracellular human 5¢-nucleotidases: cN-II, cN-III, cN-IA, cN-IB, cytosolic 5¢(3¢)-deoxyribonucleotidase (cdN) and mitochondrial 5¢(3¢)-deoxyribonu-cleotidase (mdN) Residues 55 and 56 of cN-II are indicated in bold.
Trang 3might be important for this activity However, we
noted that there is another variable residue near T56
In position 55 of cN-II, a Tyr is present that is
substi-tuted by Met in cN-III and by Ala or Gly in the other
enzymes Therefore, the two phosphotransferases have
a bulkier amino acid in this position compared to the
other four enzymes, which have amino acids with very
small side chains We decided to construct two
mutants of motif I at positions 55 (Y55G) and 56
(T56V) in order to ascertain whether one of them is
responsible for the phosphotransferase activity
Deter-mination of kinetic parameters showed that
substitu-tion of Tyr55 by a smaller residue causes a dramatic
increase in the ratio of nucleotidase to
phosphotrans-ferase activity The affinity for the substrates inosine
and IMP remains unaltered, and the effect of the
enzyme activators ATP and Mg2+ is also unchanged
(Table 1) Furthermore, by simultaneously measuring
the rate of nucleoside and phosphate production from
IMP and monophosphate synthesis from inosine (see
Experimental procedures), we observed that, in the
wild-type, the transfer of phosphate to inosine
accounts for 53% of the phosphate produced from
IMP, but only for 11% with the mutated Y55G
enzyme under the same experimental conditions
(Fig 2) However, substituting Thr56 by Val produced
an enzyme with a lower turnover but the same
func-tional characteristics as the wild-type enzyme, included
the nucleotidase⁄ phosphotransferase ratio (Table 1)
Regulatory sites
Effector site 1 is located near subunit interface A, and
is believed to interact with the activator Ap4A, by
binding one adenosine moiety in each subunit (Fig 3)
We mutated Arg144, which has been proposed to bind
the phosphate moiety of nucleotide activators, Ile152,
which is thought to interact with the adenylic moiety
of the activator, and Asn154, which probably interacts
with the purine ring through a hydrogen bond [11],
substituting these amino acids by negatively charged
residues For putative effector site 2, we mutated
Phe127 and His428, as adenosine is presumably stacked between these two residues, and Met436, which forms a hydrogen bond with the purine ring through its carbonylic group [11] (Fig 3) We substi-tuted the first two residues by a negatively charged res-idue to discourage stacking, and replaced Met436 with
a bulkier amino acid (Trp)
Putative effector site 1 None of the mutations produced had a significant effect on Kmfor the two principal substrates (Table 2)
Table 1 Effect of point mutations on various kinetic parameters of recombinant bovine cN-II Nucleotidase and phosphotransferase activi-ties were measured as described in Experimental procedures Values are means ± SD of at least three independent assays k cat refers to nucleotidase activity and is measured at saturating concentrations of IMP and sub-saturating concentrations of inosine The K50for ATP was measured as phosphotransferase activity, while for Mg 2+ , it was measured as nucleotidase activity.
cN-II
Nucleotidase ⁄
phosphotransferase
Km(inosine) (m M )
Km(IMP) (m M ) k cat (s)1)
K50(Mg 2+ ) (m M )
K50(ATP) (m M )
50
100
0
Ino H2O
E + IMP
E + IMP E-P E + Pi
Ino
Fig 2 Rate of inosine, IMP and P i production catalysed by wild-type cN-II (black bars) or mutant Y55G (white bars) in the presence
of 2 m M IMP and 1.4 m M inosine (Ino) as substrates The assays were performed as described in Experimental procedures For wild-type and Y55G, 100% activity corresponds to 32 and 18 UÆmL)1, respectively The upper scheme indicates the catalytic mechanism
of cN-II E, enzyme; P, phosphate The products measured are boxed.
Trang 4Mutant R144E showed an altered affinity for all the activators tested, N154D was normally activated by ATP and BPG but activation by ADP and Ap4A was completely prevented, as is the case for I152D, which showed a higher K50for ATP and BPG
Putative effector site 2 Study of cN-II crystallized in the presence of adeno-sine indicates that this nucleoside binds to this site, even though only the amino acid residues involved in the binding of the purine base were identified, while the ribose moiety was completely disordered [11] Mutation of two residues involved in the binding of adenine (F127E and M436W) had no effect on the functional characteristics of the enzyme, except for an increase in the K50for Ap4A for mutant M436W The mutations did not alter the activatory capacity of all the compounds tested The mutant H428D was almost insensitive to all known activators of cN-II (Table 2)
CN-II subunit oligomerization Using ‘FirstGlance in Jmol’ (http://molvis.sdsc.edu/ fgij/), we analysed the cN-II crystal structure described
by Wallde´n et al [11] We mutated three amino acids
at interface A (Phe36, Tyr115 and Asp396) and two at interface B (Lys311 and Gly319)
Interface A
In the model proposed on the basis of the crystal struc-ture, interface A is involved in formation of a dimeric structure (Fig 3) Fifty-three amino acids contribute to interaction of the two monomers by forming both hydrogen bonds and salt bridges Some of these residues are near to effector site 1 We designed our mutants in
Fig 3 Model of the homotetrameric quaternary structure of cN-II
showing interfaces A and B and the Mg 2+ site The inset shows
the tertiary structure of each subunit Effector sites 1 and 2 and
the active site are shown.
Table 2 Effect of point mutations on various kinetic parameters of recombinant bovine cN-II Nucleotidase and phosphotransferase activi-ties were measured as described in Experimental procedures Values are means ± SD of at least three independent assays K50values for
P i , ATP, ADP and Ap 4 A were measured as phosphotransferase activity, while those for BPG and Mg2+were measured as nucleotidase activ-ity The extent of activation, when present, was between 5- and 10-fold (1) Mutation in putative effector site 1; (2) mutation in putative effector site 2 NA, no activation.
cN-II
Km(IMP)
(m M )
Km(inosine) (m M )
K50(Mg 2+ ) (m M )
K50(Pi) (m M )
K50(BPG) (m M )
K50(ATP) (m M )
K50(ADP) (m M )
K50(Ap4A) (m M )
R144E (1) 0.10 ± 0.05 1.0 ± 0.2 2.0 ± 1.0 6.5 ± 1.0 5.0 ± 1.50 30.0 ± 5.0 33.0 ± 6.0 1.6 ± 1.00
Trang 5an attempt to interfere with monomer aggregation via
interface A by introducing or deleting charged residues
or altering molecular hindrance One of the mutated
amino acids (Tyr115) is located very close to Arg144,
which is part of effector site 1 (Fig S1) The mutant
F36R was inactive, while Y115A behaved normally and
D396A was only activated by BPG (Table 3)
Interface B
The structure described by Wallde´n et al [11] indicates
that the active tetramer is formed by interaction between
two identical dimers via interface B: this interface
con-tains 28 aminoacid residues, of which 8 make hydrogen
bonds (Fig 3) We constructed two mutants by
introdu-cing a negative charged residue or substituting a residue
with a non polar group (K311A), thus forming both
hydrogen and van der Waals interactions The two
mutants showed kinetic characteristics similar to those
of the wild-type (Table 3) FPLC gel-filtration
chroma-tography of purified recombinant wild-type cN-II
indi-cated that this enzyme exists in solution as a tetramer
(molecular mass 260 kDa) (Fig 4) Actually, the
chro-matogram showed two peaks, but the first was inactive
Immunoblotting of both peaks indicated cross-reactivity
with specific antibodies against cN-II, but the first
inac-tive peak disappeared when Escherichia coli cells
expressing cN-II were grown at 20C instead of 37 C
(data not shown) A dimeric active form (130 kDa) was
present in addition to the tetramer for mutants K311A
and G319D (mutations at interface B) (Fig 4)
Finally, we investigated the chromatographic
behav-iour of recombinant wild-type enzyme in the presence of
ATP and Mg2+as an activator and Pias an inhibitor
Figure 5 shows that the enzyme exists and functions as a
tetramer irrespectively of the presence of activator or
inhibitor We also performed light-scattering
measure-ments of the enzyme alone or in the presence of its
effec-tors There was no change in molecular mass in the
presence of enzyme activators or inhibitors (Table 4)
Discussion Active site The soluble 5¢-nucleotidase family shares four con-served motifs with the HAD superfamily that are involved in the reaction mechanism [5] Resolution of the crystal structure of some family members indicated that the active site contains all the conserved motifs, whose role in catalysis was determined through a mechanistic approach [5,11,13] Other than these con-served motifs, cN-IA, cN-IB, cN-II, cN-III and the cytosolic and mitochondrial 5¢(3¢)-deoxyribonucleotid-ases differ considerably in their primary structure Despite this poor similarity, there is much evidence to indicate that all members of soluble 5¢-nucleotidase family share the same reaction mechanism, proceeding through a covalent phospho-enzyme intermediate [13,14] Phospho-cN-II has been isolated, and the phosphate has been found to be localized to a con-served aspartate residue (D52) in the first of the four conserved motifs [13] Formation of a phospho-inter-mediate suggests the possibility that the enzyme cataly-ses transfer of phosphate to a suitable acceptor [15]
A number of HAD superfamily members are able to catalyse a phosphotransferase reaction, including at least two soluble nucleotidases (cN-II and cN-III) The aligned sequence of motif I of soluble nucleotidases indicates that both nucleotidases with phosphotransfer-ase activity had a Thr instead of Val in the fifth posi-tion after the phosphorylated Asp, and a relatively bulky residue in the fourth position instead of Gly, which is present in all the other nucleotidases On this basis, we mutated two residues in motif I, Thr56 and Tyr55 Thr56 was substituted by Val, and Tyr55 was substituted by Gly Our results indicate that the pres-ence of Thr or Val in position 56 is not relevant for the nucleotidase⁄ phosphotransferase ratio However, the increase in flexibility very close to Asp54, which
Table 3 Effect of point mutations on various kinetic parameters of recombinant bovine cN-II Nucleotidase and phosphotransferase activi-ties were measured as described in Experimental procedures K 50 values for P i , ATP, ADP and Ap 4 A were measured as phosphotransferase activity, while those for BPG and Mg 2+ were measured as nucleotidase activity The extent of activation, when present, was between 5- and 10-fold Values are means ± SD of at least three independent assays (A) Mutation in interface A; (B) mutation in interface B NM, not measurable NA, no activation.
cN-II
K m (IMP)
(m M )
K m (inosine) (m M )
K 50 (Mg2+) (m M )
K 50 (P i ) (m M )
K 50 (BPG) (m M )
K 50 (ATP) (m M )
K 50 (ADP) (m M )
K 50 (Ap 4 A) (m M ) Wild-type 0.10 ± 0.02 1.0 ± 0.2 2.0 ± 0.5 2.0 ± 0.3 0.3 ± 0.06 1.0 ± 0.3 2.2 ± 0.5 0.10 ± 0.05
K311A (B) 0.04 ± 0.03 0.2 ± 0.5 0.7 ± 0.5 2.0 ± 0.4 0.5 ± 0.04 7.3 ± 1.5 1.00 ± 0.6 0.25 ± 0.07 G319D (B) 0.05 ± 0.04 0.5 ± 0.5 2.5 ± 0.6 1.6 ± 0.7 0.5 ± 0.03 3.0 ± 1.0 2.00 ± 0.5 0.25 ± 0.05
Trang 6participates in binding of the phosphate [13], obtained
with the Y55G mutant, causes an increase in turnover
for the nucleotidase reaction, and, as a consequence, a
large increase in the nucleotidase⁄ phosphotransferase
ratio
Regulatory sites
cN-II is activated by BPG, a number of triphosphate
and diphosphate nucleosides (ATP and ADP being the
best activators), Ap4A and polyphosphates
Con-versely, orthophosphate has an inhibitory effect
Kinetic studies have indicated that ATP (and presum-ably other phosphorylated compounds) causes stabil-ization of an enzyme form with a high kcat, without substantial alteration of the Km for the substrates, while orthophosphate stabilizes a form at high Kmwith
no effect on kcat If ATP and phosphate are present at the same time, an enzyme form with high Kmand high
kcatis observed [9] Therefore, depending on the effec-tor, cN-II may be present as one of two structures
.
Retention time (min)
A
B
C
.
.
•
Fig 5 FPLC profiles of wild-type cN-II alone (A), or in the presence
of 5 m M ATP and 10 m M MgCl2(B), or in the presence of 5 m M Pi and 10 m M MgCl 2 (C) The column, prepared and eluted as described in Experimental procedures, was loaded with approxi-mately 150 lg of purified protein, and the activity was measured as the rate of IMP production in the presence of inosine (phospho-transferase activity) as described in Experimental procedures Arrows indicate the molecular mass of the marker proteins thyro-globulin (670 kDa), apoferritin (443 kDa), b-amylase (210 kDa) and alcohol dehydrogenase (150 kDa).
.
Retention time (min)
WT
G319D
K311A
.
•
Fig 4 FPLC profiles of wild-type cN-II, and two mutants in
inter-face B: G319D and K311A The column, prepared and eluted as
described in Experimental procedures, was loaded with
approxi-mately 150 lg of purified protein, and the activity was measured as
the rate of phosphate production in the presence of IMP
(phospha-tase) as described in Experimental procedures Arrows indicate the
molecular mass of the marker proteins thyroglobulin (670 kDa),
apoferritin (443 kDa), b-amylase (210 kDa) and alcohol
dehydroge-nase (150 kDa).
Trang 7with high and low Km In addition, a low and high kcat
may be associated with each structure It has also been
suggested, on the basis of kinetic characterization, that
the enzyme has at least three effector sites, one for
ATP, one for ADP and one for BPG [5,12] On the
basis of the results obtained from crystallization of a
truncated form of cN-II [11], we constructed point
mutants for a number of amino acid residues located
in putative effector sites 1 and 2 and involved in
bind-ing of adenylic nucleotides and BPG
Our results partially confirm the molecular
model-ling, suggesting that effector site 1 is the binding site
for Ap4A and that ADP binds this site as well Ap4A
binds between two subunits with one adenosine moiety
in each subunit [11], but ADP may possibly fill the
whole site in each subunit and attain the same result
Mutant N154D shows a decrease in the value of K50
for Mg2+ Moreover, the mutated enzyme showed
some activity even in the absence of added Mg2+,
whereas all the other active mutants, as well as the
wild-type enzyme, are completely dependent on
addi-tion of the metal This behaviour may be due to the
proximity of Asn154 to the active site (Fig S2), and, in
particular, a modification in the interaction of this
resi-due with Asp351 (directly involved in Mg2+ binding)
and His352 both belonging to motif III Remarkably,
mutant D396A, in which the mutation is located in
interface A, very close to effector site 1 and the active
site, shows a similar increase in affinity for Mg2+ Our
results also indicate that ATP and BPG probably bind
to a different site and that this is effector site 2 Point
mutations at this site result in either generalized
impair-ment of enzyme regulation or characteristics very
simi-lar to those of the wild-type As virtually all purine and
pyrimidine triphosphates have some activatory effect
on cN-II, the interaction site is likely to be more
selec-tive for the phosphorylated sugar moiety than for the
base Therefore, binding of adenosine to effector site 2,
as demonstrated by the cN-II crystal structure [11], may not be indicative of the location of the nucleoside triphosphate effector site
CN-II subunit oligomerization CN-II has been purified from various sources and has always been described as a homotetramer [16,17] The crystal structure suggests that the tetramer arises from interaction of two dimers through interface B Muta-tions in interface A, through which two monomers interact, and which is very close to the effector site 1, either resulted in a completely inactive enzyme or strongly interfered with activation by ADP and Ap4A This indirectly confirms that effector site 1 is specific for these compounds Mutation of amino acid residues located in interface B generated proteins for which an active dimeric form was detected in addition to the tet-ramer However, stabilization of the dimeric form had
no effect on the catalytic capacity of the enzyme Our results show that the monomer is probably inactive, and the dimer is the smallest active cN-II quaternary structure FPLC analysis of purified recombinant enzymes shows a heavier protein (720 kDa) in addition
to the proteins at the expected molecular mass This protein was an unusual oligomerization state of a pro-tein that was identified by immunoblotting as cN-II but completely inactive E coli produces a small amount of cN-II that is correctly folded and a large amount of incorrectly folded and insoluble protein It
is conceivable that a cN-II protein that is incorrectly folded but still soluble is also produced A decrease in growth temperature resulted in disappearance of the heavier inactive peak Furthermore, FPLC analysis of the recombinant wild-type cN-II showed the presence
of the tetrameric active form irrespective of the pres-ence of effectors This result disagrees with other results obtained on recombinant human cN-II [10], which is almost identical to our bovine enzyme To support our finding, we used light scattering, a tech-nique that, unlike gel-filtration chromatography, does not require protein dilution However, light-scattering experiments confirmed that enzyme activation or inhi-bition is not followed by a change in enzyme subunit oligomerization
In conclusion, our results confirm that there is indeed
an activatory site specific for Ap4A located at the inter-face between two subunits (interinter-face A) This site may also accommodate ADP at lower affinity, but resulting
in the same level of activation as Ap4A ATP binds to
a different site; however, it was not possible to confirm this as effector site 2 indicated by the crystal structure obtained in the presence of adenosine In contrast to
Table 4 Variation of the protein molecular mass in the presence
of specific effectors as estimated by the ratio of the scattered light
intensity Values are means ± SD of at least three measurements.
cN-II in 20 m M Tris ⁄ HCl, pH 8, +0.2 M NaCl (control)
(control), +5 m M ⁄ 10 m M ATP ⁄ Mg 2+ 1.16 ± 0.06
(control), +10 m M ⁄ 10 m M ATP ⁄ Mg 2+
1.15 ± 0.08
(control), +5 m M ⁄ 10 m M P i ⁄ Mg 2+ 0.95 ± 0.07
a
Ratio of molecular mass in the presence of specific effectors to
that of the free protein.
Trang 8the suggestion by Wallde´n [11], we demonstrated that
BPG does not bind effector site 1 Finally, mutations
at the interfaces and gel-filtration experiments indicated
that the tetramer is the major quaternary structure of
cN-II, irrespectively of the presence of effectors, but
that the dimer may also be stable and active
Experimental procedures
Materials
DpnI was provided by New England Biolabs (Ipswich, MA,
USA) and Pfu DNA polymerase by Promega (Madison,
WI, USA) Ni-NTA agars was purchased from Qiagen
(Valencia, CA, USA) [8-14C]-inosine and [8-14C]-IMP were
obtained from Sigma-Aldrich (St Louis, MO, USA)
Opti-phase ‘HiSafe’ 3 scintillation liquid was obtained from
Per-kin-Elmer (Waltham, MA, USA) Superdex-200 was
purchased from GE Healthcare (Piscataway, NJ, USA) All
other chemicals were of reagent grade
Site-directed mutagenesis
The point mutants F36R, Y115A, F127E, I152D, R144E,
N154D, K311A, G319D, D396A and M436W were
obtained using the PCR-based site-directed mutagenesis
method described by Fisher and Pei [18], and the point
mutants Y55G, T56V and H428D were produced as
described in the QuikChangesite-directed mutagenesis kit
manual (Stratagene, La Jolla, CA, USA) The primers used
are listed inTable 5
Expression and purification of the recombinant
proteins
Expression of the recombinant mutants was performed as
previously described [19] The 6· His-tagged proteins were
purified using the Ni-NTA agar method as described in the QIAexpressionist handbook (Qiagen) The protein con-centration was determined using the Bradford method [20], with BSA as the standard
Enzyme assays
The nucleotidase activity of cN-II and its mutants was mea-sured as the rate of [8-14C]-inosine formation from 2 mm [8-14C]-IMP in the presence of 1.4 mm inosine, 20 mm MgCl2, 4.5 mm ATP and 5 mm dithiothreitol, as previously described [9] Phosphotransferase activity was measured as the rate of [8-14C]-IMP formation from 1.4 mm [8-14 C]-ino-sine, in the presence of 2 mm IMP, 20 mm MgCl2, 4.5 mm ATP and 5 mm dithiothreitol, as previously described [9] For determination of kinetic parameters (Km and kcat), the concentration of the labelled substrates ranged from 0.02 to
4 mm A plot of the dependence of the rate of phospho-transferase activity on MgCl2, ATP, ADP and BPG con-centration was used to determine the value of K50for these compounds Under these experimental conditions, the accu-mulation of radiolabelled inosine (nucleotidase activity) rep-resents the sum of the phosphatase and the phosphotransferase activities It has been previously reported that, at a concentration close to the Km value (1.4 mm), inosine reduces phosphatase activity to 50% without affecting the Vmaxfor both reactions [9] Thus, the expected value of 2 was determined for the ratio between nucleotidase and phosphotransferase activities under the experimental conditions used for the wild-type recombinant cN-II assay Accordingly, an alteration of this ratio for a mutant was considered as caused either by an alteration of the Kmvalue for one of the two substrates or by a variation
of the kcat value for one of the two activities When required, the rate of phosphate formation (phosphatase activity) was measured as described by Chifflet et al [21] One unit of enzyme activity is the amount of enzyme
Table 5 Primers used for site-directed mutagenesis.
Trang 9required to convert 1 lmol of substrate to product per
min-ute under the assay conditions
Gel-filtration chromatography
The gel-filtration chromatography was performed on a
FPLC system utilizing a Superdex-200 column
(1.2· 32 cm) Purified wild-type or mutant cN-II (150 lg)
was loaded onto the column, and the chromatography was
performed at a flow rate of 0.3 mLÆmin)1 using 50 mm
Tris⁄ HCl, pH 7.4, with addition of 200 mm NaCl
Frac-tions of 0.1 mL were collected
Light scattering
The intensity of light scattered at 90 from the incident
beam was measured using a spectrofluorometer
(Fluoro-max-4; Horiba, Edison, NJ, USA) The wavelength of the
incident light was 350 nm, and the band-pass used in the
excitation and emission monochromators was 1.0 nm The
sample cell (fluorescence cell 1· 1 cm2 cross-section) was
cleaned using water filtered through 0.22 lm pore
mem-brane Protein and additive solution were passed through
identical filters All reagents were dissolved in 20 mm
Tris⁄ HCl + 0.2 m NaCl, pH 8, and the enzyme
concentra-tion ranged between 0.09 and 0.14 mgÆmL)1 Scattering
intensities were compared between samples that had the
same protein concentration and refractive index Under
these conditions, their ratio, according to Parr and
Ham-mes [22], is proportional to the ratio of molecular mass
Acknowledgements
We would like to thank Dr Giovanni Strambini and
Dr Margherita Gonelli of the Institute of Biophysics
(National Research Centre, Pisa, Italy) for the
light-scattering analysis We would also like to thank Dr
Adrian Wallwork for careful language revision of the
manuscript This work was supported by a grant from
the Ministero dell’Istruzione, dell’Universita` e della
Ricerca and by local funds from the University of
Pisa
References
1 Tozzi MG, Camici M, Pesi R, Allegrini S, Sgarrella F
& Ipata PL (1991) Nucleoside phosphotransferase
activ-ity of human colon carcinoma cytosolic 5¢-nucleotidase
Arch Biochem Biophys 291, 212–217
2 Banditelli S, Baiocchi C, Pesi R, Allegrini S, Turriani
M, Ipata PL, Camici M & Tozzi MG (1996) The
phosphotransferase activity of cytosolic 5¢-nucleotidase;
a purine analog phosphorylating enzyme Int J Biochem
Cell Biol 2, 711–720
3 Hunsucker SA, Mitchell BS & Spychala J (2005) The 5¢-nucleotidases as regulators of nucleotide and drug metabolism Pharmacol Ther 107, 1–30
4 Galmarini CM, Jordheim L & Dumontet C (2003) Role
of IMP-selective 5¢-nucleotidase (cN-II) in haematologi-cal malignancies Leuk Lymphoma 44, 1105–1111
5 Allegrini S, Scaloni A, Careddu MG, Cuccu G, D’Am-brosio C, Pesi R, Camici M, Ferrara L & Tozzi MG (2004) Mechanistic studies on bovine cytosolic 5¢-nucle-otidase II, an enzyme belonging to the HAD superfam-ily Eur J Biochem 271, 4881–4891
6 Careddu MG, Allegrini S, Pesi R, Camici M, Garcia-Gil M & Tozzi MG (2008) Knockdown of cytosolic 5¢-nucleotidase II (cN-II) reveals that its activity is essential for survival in astrocytoma cells Biochim Biophys Acta 1783, 1529–1535
7 Rampazzo C, Gazziola C, Ferraro P, Gallinaro L, Johansson M, Reichard P & Bianchi V (1999) Human high-Km5¢-nucleotidase: effects of overexpression of the cloned cDNA in cultured human cells Eur J Biochem 261, 689–697
8 Spychala J, Madrid-Marina V & Fox IH (1988) High
Kmsoluble 5¢-nucleotidase from human placenta Prop-erties and allosteric regulation by IMP and ATP J Biol Chem 263, 18759–18765
9 Pesi R, Turriani M, Allegrini S, Scolozzi C, Camici M, Ipata PL & Tozzi MG (1994) The bifunctional cytosolic 5¢-nucleotidase: regulation of the phosphotransferase and nucleotidase activities Arch Biochem Biophys 312, 75–80
10 Spychala J, Chen V, Oka J & Mitchell BS (1999) ATP and phosphate reciprocally affect subunit association of human recombinant high Km 5¢-nucleotidase Role for the C-terminal polyglutamic acid tract in subunit associ-ation and catalytic activity Eur J Biochem 259, 851– 858
11 Wallden K, Stenmark P, Nyman T, Flodin S, Graslund
S, Loppnau P, Bianchi V & Nordlund P (2007) Crystal structure of human cytosolic 5¢-nucleotidase II: insights into allosteric regulation and substrate recognition
J Biol Chem 282, 17828–17836
12 Pesi R, Baiocchi C, Allegrini S, Moretti E, Sgarrella F, Camici M & Tozzi MG (1998) Identification, separation and characterisation of two forms of cytosolic 5¢-nucle-otidase⁄ nucleoside phosphotransferase in calf thymus Biol Chem 379, 699–704
13 Allegrini S, Scaloni A, Ferrara L, Pesi R, Pinna P, Sgarrella F, Camici M, Eriksson S & Tozzi MG (2001) Bovine cytosolic 5¢-nucleotidase acts through the forma-tion of an aspartate 52-phosphoenzyme intermediate
J Biol Chem 276, 33526–33532
14 Baiocchi C, Pesi R, Camici M, Itoh R & Tozzi MG (1996) Mechanism of the reaction catalysed by cytosolic 5¢-nucleotidase ⁄ phosphotransferase: formation of a phosphorylated intermediate Biochem J 317, 797–801
Trang 1015 Fersht A (1999) Structure and Mechanism in Protein
Science – A Guide to Enzyme Catalysis and Protein
Folding, 2nd edn W.H Freeman and Company, New
York, NY
16 Zimmermann H (1992) 5¢-nucleotidase: molecular
struc-ture and functional aspects Biochem J 285, 345–365
17 Itoh R (1993) IMP–GMP 5¢-nucleotidase Comp
Biochem Physiol B 105, 13–19
18 Fisher CL & Pei GK (1997) Modification of a
PCR-based site-directed mutagenesis method BioTechniques
23, 570–571
19 Allegrini S, Pesi R, Tozzi MG, Fiol CJ, Johnson RB &
Eriksson S (1997) Bovine cytosolic IMP⁄ GMP-specific
5¢-nucleotidase: cloning and expression of active enzyme
in Escherichia coli Biochem J 328, 483–487
20 Bradford MM (1976) A rapid and sensitive method for
the quantitation of microgram quantities of protein
uti-lizing the principle of protein–dye binding Anal
Bio-chem 72, 248–254
21 Chifflet S, Torriglia A, Chiesa R & Tolosa S (1988)
A method for the determination of inorganic phosphate
in the presence of labile organic phosphate and high
concentrations of protein: application to lens ATPases
Anal Biochem 168, 1–4
22 Parr GR & Hammes GG (1975) Subunit dissociation and unfolding of rabbit muscle phosphofructokinase by guanidine by hydrochloride Biochemistry 14, 1600– 1605
Supporting information The following supplementary material is available: Fig S1 Proximity between effector site 1 and inter-face A
Fig S2 Proximity between effector site 1 and the active site
This supplementary material can be found in the online version of this article
Please note: As a service to our authors and readers, this journal provides supporting information supplied
by the authors Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset Technical support issues arising from supporting information (other than missing files) should be addressed to the authors