Article (PDF Available)

Structure-function analysis of yeast piD261/Bud32, an atypical protein kinase essential for normal cell life

Dipartimento di Chimica Biologica, Centro Studi delle Biomembrane del Consiglio Nazionale delle Ricerche and Centro Ricerca Interdipartimentale Biotecnologie Innovative, University of Padova, Viale G. Colombo, 3, 35125 Padova, Italy.
Biochemical Journal (Impact Factor: 4.4). 07/2002; 364(Pt 2):457-63. DOI: 10.1042/BJ20011376
Source: PubMed

ABSTRACT

The Saccharomyces cerevisiae YGR262c/BUD32 gene, whose disruption causes a severe pleiotropic phenotype, encodes a 261-residue putative protein kinase, piD261, whose structural homologues have been identified in a variety of organisms, including humans, and whose function is unknown. We have demonstrated previously that piD261, expressed in Escherichia coli as a recombinant protein, is a Ser/Thr kinase, as judged by its ability to autophosphorylate and to phosphorylate casein. Here we describe a mutational analysis showing that, despite low sequence similarity, the invariant residues representing the signature of protein kinases are conserved in piD261 and in its structural homologues, but are embedded in an altered context, suggestive of unique mechanistic properties. Especially noteworthy are: (i) three unique inserts of unknown function within the N-terminal lobe, (ii) the lack of a lysyl residue which in all other Ser/Thr kinases participates in the catalytic event by interacting with the transferred ATP gamma-phosphate, and which in piD261 is replaced by a threonine, and (iii) an exceedingly short activation loop including two serines, Ser-187 and Ser-189, whose autophosphorylation accounts for the appearance of an upshifted band upon SDS/PAGE. A mutant in which these serines are replaced by alanines was devoid of the upshifted band and displayed reduced catalytic activity. This would include piD261 in the category of protein kinases activated by phosphorylation, although it lacks the RD (Arg-Asp) motif which is typical of these enzymes.

Full-text (PDF)

Available from: Sonia Facchin
Biochem. J. (2002) 364, 457–463 (Printed in Great Britain) 457
Structure–function analysis of yeast piD261/Bud32, an atypical protein
kinase essential for normal cell life
Sonia FACCHIN
1
, Raffaele LOPREIATO
1
, Silvia STOCCHETTO, Giorgio ARRIGONI, Luca CESARO, Oriano MARIN,
Giovanna CARIGNANI and Lorenzo A. PINNA
2
Dipartimento di Chimica Biologica, Centro Studi delle Biomembrane del Consiglio Nazionale delle Ricerche and Centro Ricerca Interdipartimentale Biotecnologie
Innovative, University of Padova, Viale G. Colombo, 3, 35125 Padova, Italy
The Saccharomyces cereisiae YGR262c\BUD32 gene, whose
disruption causes a severe pleiotropic phenotype, encodes a 261-
residue putative protein kinase, piD261, whose structural homo-
logues have been identified in a variety of organisms, including
humans, and whose function is unknown. We have demonstrated
previously that piD261, expressed in Escherichia coli as a
recombinant protein, is a Ser\Thr kinase, as judged by its ability
to autophosphorylate and to phosphorylate casein. Here we
describe a mutational analysis showing that, despite low sequence
similarity, the invariant residues representing the signature of
protein kinases are conserved in piD261 and in its structural
homologues, but are embedded in an altered context, suggestive
of unique mechanistic properties. Especially noteworthy are: (i)
three unique inserts of unknown function within the N-terminal
lobe, (ii) the lack of a lysyl residue which in all other Ser\Thr
INTRODUCTION
It has been estimated that more than 30% of the proteins in a
eukaryotic cell undergo phosphorylation, this event representing
the most frequent and general device by which biological
functions are reversibly regulated in higher organisms [1]. The
enzymes responsible for this reaction, protein kinases, make up
a large family of proteins, whose members share a common
catalytic domain composed of about 300–350 residues charac-
terized by a number of conserved features, leading to the
definition of 12 distinct subdomains. Such structural
organization, deduced from the sequence alignment of about 120
protein kinases, either Ser\Thr- or Tyr-specific [2], underlies a
common bi-lobal architecture, with the catalytic site between the
two lobes, that is found in all the protein kinases whose crystal
structures have been solved to date (e.g. [3,4]). This has led to a
general understanding of the relationship existing between con-
served subdomains, three-dimensional structure and catalytic
properties of protein kinases [2].
Sequencing of the genomes of Saccharomyces cereisiae,
Caenorhabditis elegans and Homo sapiens has confirmed the
prediction that in eukaryotes protein kinases will constitute one
of the largest families of enzymes [5–7]. Interestingly, however,
besides a majority of deduced sequences that display the struc-
tural features of protein kinases unambiguously, there are a
number of putative protein kinase genes whose products do not
display all the stigmata of this family of enzymes. Since these
highly conserved features are believed to be essential for correct
folding and\or catalytic competence, a legitimate question is
Abbreviations used: PKA, cAMP-dependent protein kinase; FSBA, fluorylsulphonylbenzoyladenosine.
1
These authors contributed equally to this work.
2
To whom correspondence should be addressed (e-mail pinna!civ.bio.unipd.it).
kinases participates in the catalytic event by interacting with the
transferred ATP γ-phosphate, and which in piD261 is replaced
by a threonine, and (iii) an exceedingly short activation loop
including two serines, Ser-187 and Ser-189, whose auto-
phosphorylation accounts for the appearance of an upshifted
band upon SDS\PAGE. A mutant in which these serines are
replaced by alanines was devoid of the upshifted band and
displayed reduced catalytic activity. This would include piD261
in the category of protein kinases activated by phosphorylation,
although it lacks the RD (Arg-Asp) motif which is typical of
these enzymes.
Key words : casein, CK2, protein phosphorylation, Saccharo-
myces cereisiae, Ser\Thr kinase.
whether these proteins are actually operating as bona fide protein
kinases.
In S. cereisiae one of these putative atypical protein kinases
is the product of the YGR262c gene [8]; its structural homologues
are present in a variety of organisms, from archaebacteria to
humans. Despite several structural abnormalities, the product of
this gene, originally termed piD261, when expressed in
Escherichia coli with a C-terminal His tag in the presence of Mn
#
+
displays significant ATP-protein phosphotransferase catalytic
activity affecting Ser\Thr residues, as judged from both auto-
phosphorylation and the ability to phosphorylate acidic proteins,
notably casein and osteopontin, in itro [9]. The protein appears
to be implicated in a general cellular mechanism as its deletion
confers to yeast cells a severe slow-growth phenotype [8]. Mutant
cells also display several specific defects [9a], which include
reduced survival of cells in the stationary phase, the inability of
homozygous diploids to enter sporulation, with no visible meiotic
division [9a], and alterations in cell-wall structure, as indicated
by several specific tests [10]. Diploid ygr262c mutants also
exhibit random budding [11]. Due to this phenotype the name
BUD32 has been reserved recently at the Saccharomyces
Genome Database (SGD
TM
; http:\\genome-www.stanford.edu\
Saccharomyces) for the YGR262c gene. The apparent involve-
ment of piD261\Bud32 in different cellular processes might be
explained by its crucial role in a biological pathway endowed
with a general function or, alternatively, by its activity as a
regulator of different substrates by phosphorylation. In order to
define the biochemical characteristics of the protein we started a
mutational study aimed at probing the structure–function
# 2002 Biochemical Society
Page 1
458 S. Facchin and others
relationships of this atypical protein kinase. The results are
described in the present paper.
EXPERIMENTAL
Partially dephosphorylated α-casein was purchased from Sigma.
The CK2β 1-77 peptide reproducing the N-terminal segment of
the protein kinase CK2 β subunit was synthesized as described
previously [12]. The pET-261 plasmid encoding the mutant form
SS187,189AA of piD261\Bud32 was provided kindly by Dr P.
Ghisellini (University of Genoa, Genoa, Italy). The anti-
fluorylsulphonylbenzoyladenosine (FSBA) antibody was a gen-
erous gift from Dr P. J. Parker (ICRF, London, U.K.) [13].
Monoclonal antibody anti-His was from Amersham Pharmacia
Biotech. Mouse antiserum against piD261\Bud32 was raised
using 10 µg of purified recombinant protein dissolved in 100 µlof
Ringer solution: this solution, emulsified previously with
Freund’s adjuvant (1: 1, v\v), was injected intraperitoneally into
mice five times at 1 week intervals.
Strains, media and plasmids
The E. coli strains INVαFh (Invitrogen), used for plasmid
preparations, and BL21(DE3) (Novagen), used for protein
expression, were grown in LB medium (1 % bacto-tryptone,
0.5% yeast extract and 0.5 % NaCl) or LB containing 50 mg\l
ampicillin. The pET-261 plasmid, used for YGR262c expression
in E. coli, has been described in [9].
Mutagenesis of the YGR262c/BUD32 gene
The mutant forms of YGR262c\BUD32 were obtained using the
QuikChange
TM
Site-Directed Mutagenesis Kit (Stratagene, cata-
logue no. 200518). The mutagenic primers were designed ac-
cording to the instruction manual. As double-stranded DNA
templates we used the pET-261 plasmid construct. The resulting
plasmids were controlled by DNA sequencing of the inserts.
Tryptic digestion and MS
After blotting on PVDF membrane (Millipore), the area cor-
responding to piD261\Bud32 was cut out and put into a 200 µl
solution of ammonium bicarbonate [14]. The reaction was started
by addition of trypsin (Sigma; trypsin\protein ratio of 1 :50).
After 2 h an equal amount of fresh trypsin was added and the
digestion stopped at the end of 4 h. The membrane was removed
and the solution lyophilized several times. The residue was
dissolved in water containing 0.1 % trifluoroacetic acid and
subjected to MS (Maldi-1; Kratos-Schimadzu, Manchester,
U.K.) using α-cyano-4-hydroxycinnamic acid as matrix.
Expression and purification of piD261-His
6
The expression of recombinant piD261-His
'
was performed as
described previously [9], with some modifications. The E. coli
strain BL21(DE3), containing plasmid pET-261, was grown in
LB medium at 37 mC until the D
'!!
value reached 0.3 and at room
temperature until D
'!!
reached 0.7–0.8, when transcription of the
YGR262c coding sequence was induced with 0.4 mM isopropyl
β--thiogalactoside. After 16–18 h of incubation at 16 mC bacteria
were harvested and resuspended in 10 ml\g of pellet of puri-
fication buffer (20 mM Tris\HCl, pH 7.5, 0.3 M NaCl, 10 %
glycerol, 1 mM 2-mercapthoethanol and 0.2 mM PMSF).
piD261-His
'
was purified to homogeneity from the bacterial
extract [9] according to the protocol of the manufacturer by an
affinity column containing Ni
#
+
-nitrilotriacetate–agarose gel
Figure 1 Purification of recombinant piD261-His
6
by Ni
2
+
-nitrilotriacetate
affinity chromatography
The soluble bacterial extract was stirred with Ni
2
+
-nitrilotriacetate–agarose according to the
protocol of the manufacturer (Qiagen). The suspension was added to the column and the resin
was subsequently submitted to a two-step elution with purification buffer containing 10 and
100 mM imidazole, respectively. Samples of each eluate (15 µl), as indicated, were subjected
to SDS/PAGE (11% gel) and proteins were detected with Coomassie Brilliant Blue (lanes 1–3).
In lane 4 the same sample as in lane 3 was subjected to Western-blot analysis with anti-piD261
antibodies. For details, see the Experimental section.
(Qiagen). As shown in Figure 1, most contaminating proteins
were not retained on the column (Figure 1, lane 1) and\or were
eluted by washing the column with the purification buffer
containing 10 mM imidazole (Figure 1, lane 2) ; piD261-His
'
was
eluted by 100 mM imidazole as a main 29.5 kDa band ac-
companied by two faint, slightly less mobile bands (Figure 1,
lane 3). These were also recognized by anti-piD261 antibodies
(Figure 1, lane 4) consistent with their identification as
phosphorylated forms of piD261-His
'
(see the Results section).
Western-blot analysis
Approx. 200 ng of recombinant piD261-His
'
(wild-type or mu-
tant) were subjected to SDS\PAGE [11% gel ; the protein
concentration was determined by densitometry (Image Master,
Amersham Pharmacia Biotech), using BSA as a standard].
Proteins were blotted on to PVDF membrane and detected by
using the mouse antiserum against piD261-His
'
at a 500-fold
dilution. This reaction was followed by incubation with a
phosphatase-coupled antibody directed against the correspond-
ing IgG type of the first antibody. The same procedure was used
in the case of anti-His (diluted 1000-fold) and anti-FSBA (diluted
2000-fold) antibodies. The experimental procedure used to
examine the binding of FSBA to wild-type or mutant piD261 was
as described in [12], using Mn
#
+
instead of Mg
#
+
.
# 2002 Biochemical Society
Page 2
459Atypical protein kinase piD261/Bud32
Phosphorylation assay
The protein kinase activity of piD261 and of its mutant forms
was assayed routinely by incubating the recombinant His-tagged
protein (0.2 µg\µl) at 37 mC for 15 min in 20 µl of a medium
containing 50 mM Tris\HCl, pH 7.5, 10 mM MnCl
#
,25µM[γ-
$#
P]ATP (Amersham Pharmacia Biotech ; specific radioactivity,
2000–3000 c.p.m.\pmol) and either the CK2β 1-77 peptide (10–
100 µM) or α-casein (0.5–1 µg\µl) as phosphorylatable sub-
strates. The reaction was stopped by adding the gel electro-
phoresis loading buffer and samples were subjected to SDS\
PAGE, with either 11% (casein) or 18 % (CK2β 1-77) gels. The
dried gels were scanned directly using Cyclone apparatus
(Packard).
RESULTS
Mutational analysis of functionally relevant residues
A novel alignment of piD261 with the prototype kinase, mam-
malian cAMP-dependent protein kinase (PKA ; α-catalytic sub-
unit), is presented in Figure 2. The previously adopted alignment
[9] has been modified partially by manual adjustment corrobo-
rated by the mutational analysis presented in this work. To this
effect we introduced in piD261 different amino acid substitutions
Figure 2 Alignment of piD261/Bud32 and PKA sequences
The sequences of yeast piD261/Bud32 (piD261; SWISS-PROT accession no. P53323) and human piD261/Bud32 [piD261(hs), NCB accession no. CAC00561] proteins were aligned with that of
PKA (SWISS-PROT accession no. P05132) by computer analysis and manual adjustment corroborated by mutational analysis (see the text). Asterisks denote PiD261 residues that were mutated
to assess their relevance. Arrowheads denote residues that are highly conserved, as they are found in PKA. Black, identity; grey, similarity. The subdomains of protein kinases and functionally
relevant structural elements are also indicated.
by mutagenizing the YGR262c coding sequence already inserted
in the bacterial pET-261 plasmid. The phosphotransferase ac-
tivities of the different mutant proteins were assayed after
expression in E. coli and purification as described in [9] and the
experimental section. The results are summarized in Table 1.
Starting from the N-terminus, the suspected functional equiv-
alence of Gly-25 with the invariant second glycine of the
phosphate anchor motif, GX
GXXG (equivalent to PKA Gly-
52), was confirmed by mutating it to Val to give a mutant that
was fully inactive (Table 1) and also unable to bind the ATP
analogue FSBA (Figure 3). Gly-25 is therefore the only glycine
of the glycine loop left in piD261. The highly conserved valine
responsible for hydrophobic interaction with the adenine moiety
of ATP, present five residues downstream from the central
glycine of the phosphate anchor in nearly all kinases (PKA Val-
56), is also conserved in piD261 (Val-30). In contrast, the
previously proposed [9] matching of piD261 Lys-57 with the in-
variant lysine essential for the correct positioning of the ATP
triphosphate group (PKA Lys-72), turned out to be incorrect
since Lys-57 could be mutated to alanine without any appreciable
loss of activity (Table 1). Mutagenesis of two lysyl residues
upstream from Lys-57, Lys-48 and Lys-52, demonstrated that
Lys-52 must be the functional equivalent of PKA Lys-72, since
its replacement fully suppressed catalytic activity, whereas mu-
# 2002 Biochemical Society
Page 3
460 S. Facchin and others
Table 1 Mutational analysis of conserved residues in piD261/Bud32
The invariant residues defining protein kinase subdomains are shown (the numbering refers to PKA) together with the putative equivalent in piD261/Bud32. Bold denotes piD261 residues where
the mutation combined with sequence alignment denote functional equivalence with conserved kinase residues. The alignment of Arg-255 with PKA Arg-280 is questionable (see the text). The
mutations performed and their activities relative to the wild type (100 %) are listed in the final two columns. Activity of wild-type and mutant piD261/Bud32 was determined using the CK2β 1-
77 peptide as a phosphorylatable substrate under the conditions specified in the Experimental section. Activities are means from at least three separate determinations with a S.E. of less than
15%.
Subdomain
Invariant residue
(in PKA) Function
Putative equivalent
residue(s) in
piD261/Bud32
piD261/Bud32
mutant
Specific activity
(% of wild type)
I Gly-52 Provides space for ATP β-phosphate Gly-25 Gly-25 Val 1
II Lys-72 Co-ordinates α- and β-phosphates of ATP Lys-48 Lys-48 Ala 68
Lys-52 Lys-52 Ala 1
Lys-57 Lys-57 Ala 90
III Glu-91 Interacts with Lys-72 Glu-76 Glu-76 Ala 1
VIB Asp-166 Catalytic base Asp-161 Asp-161 Ala 6
VIB Lys-168 Transfer of γ-phosphate Thr-163 Thr-163 Lys 70
Thr-163 Ala 90
VIB Asn-171 Co-ordinates α- and γ-phosphates of ATP Asn-166 ––
VII Asp-184 Positioning of γ-phosphate Asp-182 ––
VIII Glu-208 Correct folding of catalytic core Glu-193 Glu-193 Ala 20
IX Asp-220 Correct folding of catalytic core Asp-198 Asp-198 Ala 10
XI Arg-280 Correct folding of catalytic core Arg-255 ? Arg-255 Ala 90
Figure 3 piD261 mutant Gly-25 Val is unable to bind the ATP analogue
FSBA
piD261 wild-type (WT) and mutant forms [Thr-163 Lys (T163K), Asp-161 Ala (D161A)
and Gly-25 Val (G25V)] were assayed for their ability to bind the ATP analogue FSBA by
Western-blot analysis, as described in the Experimental section.
tation of Lys-48 had only modest effects on activity (Table 1).
Alignment of piD261 Lys-52 with PKA Lys-72 implies the
presence in piD261 of a unique insert of seven amino acids just
downstream from the conserved Val-30. Curiously, this insert
includes a series of four consecutive threonines which are not
found in other protein kinases. Another short insert of six amino
acids is present at the end of putative subdomain II, based on the
unequivocal alignment of the sequence flanking Glu-76, which
matches well the invariant glutamyl residue (PKA Glu-91) that
defines subdomain III. As expected, the mutation of this glutamic
acid to alanine fully suppressed catalytic activity (Table 1).
A third insert of 10 residues must be introduced into the
putative domain V of piD261 to allow the alignment of putative
subdomains VIA and VIB. Subdomain VIA displays also in
piD261 the typical pattern of hydrophobic residues (o) ending
with a histidine (equivalent to PKA His-158), o—o-ooH (where
a dash indicates any residue) [2]; domain VIB includes the highly
conserved catalytic loop, with its invariant motif DXXXXN,
whose actual identification with the 161–166 segment of piD261
(
DLTSSN) was validated by showing that the replacement of
Asp-161 by Ala nearly abolishes activity (Table 1), consistent
with its essential role as a catalytic base. This also assigns piD261
to the small group of non-RD ’ (non-Arg-Asp) protein kinases,
where the catalytic aspartate is not preceded by an arginine [4].
Even more unusual is the lack in piD261 of a lysine (Lys-168 in
PKA) that is present in all Ser\Thr protein kinases two residues
Table 2 Conserved motifs at the catalytic loop of protein kinases
The piD261/Bud32 sequence is aligned and compared with the consensus determined by
Hanks [22]. o denotes hydrophobic consensus. The invariant Asp and Asn residues common
to all the motifs are underlined.
Kinase sequence Subdomain VIB
Ser/Thr kinase general consensus DoKXXN
piD261/Bud32 DLTSSN
Tyr kinase general consensus DLAARN
Tyr kinase Src consensus DLRAAN
downstream from the catalytic base (see Table 2), and which has
been shown to play a crucial role in catalysis by interacting with
the γ-phosphate of ATP while it is transferred to the phospho-
acceptor amino acid [15]. In piD261 this lysine is replaced by
threonine (Thr-163). Mutation of this threonine to lysine does
not increase catalytic activity, but rather decreases it slightly,
whereas its mutation to alanine is not detrimental at all (Table 1).
These data, in conjunction with the finding that, conversely,
mutation of Lys-168 to Ala in yeast PKA is strongly detrimental
[16], suggest that Thr-163 does not play any relevant role in
catalysis, unlike its sequence homologue lysine found in all
Ser\Thr protein kinases.
Alignment downstream from subdomain VIB becomes prob-
lematic due to the reduced size of the C-terminal moiety of
piD261. At variance with the previous alignment [9], however,
the new match, based on the insertion of three series of gaps in
subdomains VII, VIII and IX (Figure 2), would lead to the
conservation in piD261 of the four invariant residues which in
protein kinases specify subdomains VII, VIII, IX and XI. The
reliability of this alignment is supported by mutational analysis
showing that both Glu-193 (equivalent to PKA Glu-208) and
Asp-198 (equivalent to PKA Asp-220) are important for activity
(see Table 1). Both these residues are required for the correct
assembly of the catalytic core. The most C-terminal conserved
residue, defining subdomain XI in nearly all protein kinases, is an
arginine (Arg-280 of PKA) which makes a salt bridge to Glu-208.
# 2002 Biochemical Society
Page 4
461Atypical protein kinase piD261/Bud32
A
B
Figure 4 Western-blot analysis of piD261 wild type and mutants
(A) piD261 wild type and mutants (see the Experimental section and Table 1) were subjected
to SDS/PAGE (11 % gel containing 16 % glycerol), transferred to PVDF membrane and
immunoreacted with anti-piD261 antibodies, as detailed in the Experimental section. Apparent
molecular-mass values were assigned to the immunoreactive bands relative to the migration
position of a kit of eight prestained protein markers (molecular masses, 6500–175 000 Da ;
BioLabs). (B) The slowly migrating bands of wild-type piD261 increase upon incubation with
ATP and MnCl
2
for the indicated periods of time. Experimental conditions were as in (A).
The structural equivalent of this arginine in piD261 would be
Arg-255. Its mutation to alanine only marginally affects catalytic
activity (Table 1), suggesting that the functional equivalent of
PKA Arg-280 might be one of the other three arginines clustered
together with Arg-255 at the very C-terminal end of piD261.
Taken together, the results of the mutational analysis and
computer-assisted alignment, which was refined manually in
critical regions, are consistent with a scenario in which the
invariant residues of protein kinases are conserved in piD261,
although in an altered context.
Autophosphorylation of piD261 at its activation loop correlates
with increased catalytic activity
Recombinant piD261-His
'
expressed in E. coli and purified by
Ni
#
+
-nitrilotriacetate affinity chromatography gave rise to two
faint protein bands that were slightly less mobile than the main
band upon SDS\PAGE (see Figure 1). Better resolution was
achieved by including 16% glycerol in the gel, giving rise to four
bands having apparent molecular masses of 29.5, 30, 30.5 and
31 kDa upon SDS\PAGE (see Figure 4A). All bands were
recognized by anti-piD261 antibodies as well as by antibodies
raised against the C-terminal histidines (results not shown) and
against the ATP analogue FSBA covalently bound to the protein
(Figure 3). Given the C-terminal location of the His
'
tag and the
vicinity of the ATP\FSBA-binding motif to the N-terminus (see
Figure 2), these data argue against the possibility that het-
erogeneity on SDS\PAGE is due to proteolytic degradation.
Conversely, the possibility that heterogeneity is due to variable
extents of autophosphorylation occurring in E. coli is suggested
by the observation that heterogeneity on SDS\PAGE disappears
in the case of inactive mutants, notably Gly-25 Val, Lys-
52 Ala and Glu-76 Ala ; these give rise only to the 29.5 kDa
band (Figure 4A). In contrast, heterogeneity is preserved with
active mutants, e.g. Thr-163 Lys and Lys-57 Ala. These
Figure 5 Effect of mutating Ser-187 and Ser-189 on piD261 SDS/PAGE
pattern and catalytic activity
The Western blots of the wild type (WT) and the SS187,189AA mutant are shown, with specific
activity expressed as pmol of P : min
1
: mg
1
, determined using the CK2β 1-77 peptide as
a phosphorylatable substrate (see the Experimental section).
data are consistent with the concept that the upper bands with
apparent molecular masses of 29.5 kDa are indeed gener-
ated by multiple autophosphorylation of piD261 expressed in
bacteria, by analogy with other kinases, primarily PKA
[17,18].
Autophosphorylation of PKA occurs at a threonyl residue
(Thr-197) located in the activation loop (also known as the T-
loop) [18,19] and is required for full activity ; phosphorylation
gives rise to an upshifted band on SDS\PAGE (the slow form’)
whose catalytic efficiency is much higher than that of the
unphosphorylated form (the fast form ’) [20]. Likewise, many
other protein kinases have been shown to undergo a similar
activation mechanism based on the phosphorylation of seryl,
threonyl or tyrosyl residues of their activation loops, either by
autocatalysis or by heterologous kinases [4]. Similar to PKA,
whose isoforms cannot be readily interconverted in itro by
incubation with either MgATP
#
or phosphatases [20], suggesting
that once the mature protein is assembled the phosphoacceptor
site is hardly accessible, the upper band (31 kDa) of piD261
could also not be converted to more mobile band(s) by incubation
with either acidic or alkaline phosphatases (results not shown).
However, the fast-migrating band of piD261 is significantly,
albeit slowly, converted into the upper bands by incubation with
MnATP
#
(Figure 4B).
The activation loop of piD261 is abnormally short (see above) ;
however, it includes two phosphorylatable residues, Ser-187 and
Ser-189, the latter possibly representing the sequence homologue
of PKA Thr-197 (see Figure 2). To check whether these residues
might undergo autophosphorylation, which would account for
the heterogeneity of piD261, a mutant in which both Ser-187 and
Ser-189 had been replaced by Ala (SS187,189AA) was generated
and expressed in E. coli. As shown in Figure 5 this mutant,
similar to the inactive mutants, gives rise almost exclusively to
the more mobile 29.5 kDa band, the most upshifted band of
31 kDa being fully undetectable. These data support the concept
that Ser-187 and\or Ser-189 do indeed undergo auto-
phosphorylation accounting at least partially for the upshift of
piD261 on SDS\PAGE.
The demonstration that in the case of the main upshifted band
(31 kDa) this inference is correct was provided by MS analysis of
the tryptic fragments obtained from this band and from the
29.5 kDa band of piD261. As shown in Figure 6, the 29.5 kDa
band gives rise to a fragment of 2228.4 Da, corresponding
exactly to the expected tryptic peptide, including the catalytic
loop, encompassing residues between Trp-176 and Lys-195
(theoretical molecular mass, 2228.5 Da). This individual peak is
absent in the tryptic digest of the 31 kDa band, where instead a
# 2002 Biochemical Society
Page 5
462 S. Facchin and others
Figure 6 Matrix-assisted laser-desorption ionization–time-of-flight
(MALDI-TOF) mass spectrum of piD261 after tryptic digestion
(A) Analysis of the 29.5 kDa band. The mass spectrum region including the signal (a) of the
tryptic peptide between residues 176 and 195 (theoretical m/z, 2228.51 Da) is shown. (B)
Analysis of the 31 kDa band. The same mass spectrum region as in (A) is shown. Signal (a)
disappears while a novel signal (b) appears corresponding to the bisphosphorylated tryptic
fragment (theoretical m/z, 2388.47 Da).
small yet significant peak of 2389 Da, with the expected molecular
mass of the bisphosphorylated tryptic fragment (theoretical
molecular mass, 2388.4 Da), is detectable. A similar analysis
with the two faint intermediate bands (30.0 and 30.5 kDa) did
not provide reliable results due to their very small amounts. It
is not clear therefore whether these bands are generated through
the individual phosphorylation of Ser-187 and\or Ser-189 or are
Table 3 Kinetic constants of wild-type piD261 and the SS187,189AA
mutant
Catalytic activity was determined at variable concentrations of ATP (10–300 µM) using the
CK2β 1-77 peptide as a phosphoacceptor substrate. The peptide concentration (20 µM) was
far above its K
m
value (approx. 1 µM for both the wild-type and mutant proteins). For additional
details, see the Experimental section. The data presented are means from three separate
determinations with S.E. values of less than 20%.
piD261 V
max
(pmol : min
1
: mg
1
) K
m
(ATP; µM) Efficiency (V
max
/K
m
)
Wild type 1500 25 60
SS187,189AA 460 75 6.1
accounted for by phosphorylation occurring at residue(s) outside
the activation loop. This latter hypothesis would be more
consistent with the observation that only the two intermediate
bands, but neither the 29.5 nor the 31.0 kDa bands, immunoreact
with anti-phosphothreonine antibodies (results not shown).
The reduced catalytic activity of mutant SS187,189AA (see
Figure 5) supports the idea that the phosphorylation of both or
one of these residues is required for optimal activity. As shown
in Table 3 the mutant lacking the two serines in the activation
loop exhibits both a higher K
m
for ATP and a reduced V
max
, with
an overall phosphorylation efficiency 10-fold lower than that of
the wild type. Assuming that, as in the case of PKA [18,20], the
activity of the SS187,189AA mutant is comparable with that of
the unphosphorylated form of the wild type, one would conclude
that the experimentally determined catalytic efficiency of wild-
type piD261 is an average value, resulting from highly active
phosphorylated molecules and less-active non-phosphorylated
molecules. Moreover, if it is taken into account that the un-
phosphorylated 29.5 kDa band accounts for about 80% of
piD261 expressed in E. coli, whereas the doubly phosphorylated
31 kDa band accounts for just 15% of the whole protein, and
considering the negligible contribution of the two intermediate
bands (see Figure 5), it can be argued from the data of Table 3
that phosphorylation promotes an up to 60-fold increase in
catalytic efficiency. This figure compares quite well with the
increase in catalytic activity undergone by PKA upon auto-
phosphorylation [20].
DISCUSSION
The data presented indicate that, despite its small size and low
overall similarity with the other members of the protein kinase
family, piD261\Bud32 displays all the main signatures of a
protein kinase catalytic domain, with special reference to the
conservation of invariant residues, whose relevance to piD261
catalytic activity has been confirmed by mutational analysis.
These canonical features, however, are dispersed in a deeply
altered context, suggesting that in several respects the properties
of piD261 are unique. Especially remarkable features are dis-
cussed below.
(i) The lack of the N-terminal helix A and C-terminal extension,
which are present in PKA and in many other protein kinases.
Moreover, the C-terminal stretch of piD261 is unusually basic
and includes a seryl residue which displays the consensus sequence
for several basophilic protein kinases (e.g. PKA, p70
rsk
and
p90
rsk
) [21]. (ii) A striking paucity of residues in the region
encompassing subdomains VII and VIII, where only 16 residues
connect the end of the putative β8 strand to helix F, as compared
with 35–40 residues in PKA (see Figure 1) and most protein
kinases. This is especially remarkable if it is considered that this
region includes crucial functional elements, notably the activation
# 2002 Biochemical Society
Page 6
463Atypical protein kinase piD261/Bud32
segment and the pj1 loop ’, which in piD261 are shortened
drastically (see Figure 2). (iii) The replacement of a lysyl residue
(PKA Lys-168), which is strictly conserved in all Ser\Thr protein
kinases (see Table 2), where it plays a critical role in the
phosphotransferase reaction [15]. In piD261 this lysine is replaced
by a threonine (Thr-163) that does not appear to play any critical
role in catalysis since its replacement with either Lys (as in all
Ser\Thr kinases) or Ala does not significantly affect activity
(Table 1). It has to be concluded therefore that Thr-163 is
structurally, not functionally, the homologue of conserved PKA
Lys-168. This in turn raises the possibility that the mechanism of
the phosphotransferase reaction is different from that generally
adopted by other Ser\Thr kinases. (iv) Activation through
phosphorylation of one or two seryl residues in its abnormally
short activation loop, although piD261 does not belong to the
category of RD kinases. In these enzymes this kind of regulatory
mechanism is considered to be mandatory in order to neutralize
the positive charge of the arginine side chain, which otherwise
hampers the efficiency of the catalysis accomplished by the
adjacent aspartate. It is possible that, as in the case of other
kinases [4], the residue(s) of the activation loop phosphorylated
in piD261 also interact with basic residues other than the arginine
of the RD motif, thus stabilizing the active conformation of the
catalytic site. (v) The presence in the piD261 N-terminal moiety
of three unique inserts in sharp contrast with the extremely
reduced size of the C-terminal lobe. The function of these inserts
could be to interact with subunits and\or regulatory proteins
that are presently unknown but which are likely to exist,
considering the very small size of piD261, shorter than any other
protein kinase known.
All the main features of piD261, with the exception of the two
inserts in subdomains I and V, are conserved in the structural
homologues of higher eukaryotes, notably the human one (see
Figure 2), suggesting that they underlie common biological
function(s) in distantly related organisms. Apparently these must
be related to a very general cellular mechanism, or an early step
of a ramified pathway, since in yeast cells disruption of the gene
encoding piD261 causes a pleiotropic phenotype that is mani-
fested as severely impaired vegetative growth.
We are indebted to Dr P. Ghisellini for the pET-261 plasmid encoding the mutant form
SS187,189AA and to Dr P. J. Parker for the generous gift of anti-FSBA antibody. This
work was supported by grants to L. A. P. from the Armenise-Harvard Foundation,
Associazione Italiana per la Ricerca sul Cancro, the Italian Ministry of Health (Project
AIDS), the Italian M. U. R. S.T. (COFIN 2000) and C.N. R (Target Project on
Biotechnology) ; to G. C. from Italian M. U.R. S.T. (COFIN 1999) and the EEC
(EUROFAN II project).
REFERENCES
1 Marks, F. (1996) The brain of the cell. In Protein Phosphorylation (Marks, F., ed.),
pp. 1–35, VCH, Weinheim
Received 27 September 2001/2 January 2002; accepted 19 March 2002
2 Hanks, S. K. and Hunter, T. (1995) The eukaryotic protein kinase superfamily: kinase
(catalytic) domain structure and classification. FASEB J. 9, 576–596
3 Taylor, S. S. and Radzio-Andzelm, E. (1994) Three protein kinase structures define a
common motif. Structure 2, 345–355
4 Johnson, L. N., Noble, M. E. M. and Owen, D. J. (1996) Active and inactive protein
kinase: structural basis for regulation. Cell 85, 149–158
5 Hunter, T. and Plowman, G. D. (1997) The protein kinase of budding yeast: six score
and more. Trends Biochem. Sci. 22, 18–22
6 Plowman, G. D., Sudarsanam, S., Bingham, J., Whyte, D. and Hunter, T. (1999) The
protein kinase of Caenorhabditis elegans : a model for signal trasduction in
multicellular organism. Proc. Natl. Acad. Sci. U.S.A. 96, 13603–13610
7 Venter, J. C., Adams, M. D., Myers, E. W., Li, P. W., Mural, R. J., Sutton, G. G.,
Smith, H. O., Yandell, M., Evans, C. A., Holt, R. A. et al. (2001) The sequence of the
human genome. Science 291, 1304–1351
8 Sartori, G., Mazzotta, G., Stocchetto, S., Pavanello, A. and Carignani, G. (2000)
Inactivation of six genes from chromosomes VII of Saccharomyces cerevisiae and
basic phenotypic analysis of the mutant strains. Yeast 16, 255–265
9 Stocchetto, S., Marin, O., Carignani, G. and Pinna, L. A. (1997) Biochemical evidence
that Saccharomyces cerevisiae YGR262c gene required for normal growth, encodes a
novel Ser/Thr-specific protein kinase. FEBS Lett. 414, 171–175
9a Briza, P., Bogengruber, E., Thur, A., Rutzler, M., Munsterkotter, M., Dawes, I. W. and
Breitenbach, M. (2002) Systematic analysis of sporulation phenotypes in 624 non-
lethal homozygous deletion strains of Saccharomyces cerevisiae. Yeast 19, 403–422
10 De Groot, P. W. J., Ruiz, C., Va
!
zquez De Aldana, C. R., Duen
4
as, E., Cid, V. J., Del
Rey, F., Rodrı
!
guez-Pen
4
a, J. M., Pe
!
rez, P., Andel, A., Caubı
!
n, J. et al. (2001) A
genomic approach for the identification and classification of genes involved in cell
wall formation and its regulation in Saccharomyces cerevisiae. Comp. Funct. Genom.
2, 124–142
11 Ni, L. and Snyder, M. (2001) A genomic study of the bipolar bud site selection
pattern in Saccharomyces cerevisiae. Mol. Biol. Cell. 12, 2147–2170
12 Marin, O., Meggio, F., Sarno, S. and Pinna, L. A. (1997) Physical dissection of the
structural elements responsible for regulatory properties and intersubunit interactions
of protein kinase CK2 β-subunit. Biochemistry 36, 7192–7198
13 Parker, P. J. (1993) Antibodies to fluorylsulfonylbenzoyladenosine permit identification
of protein kinases. FEBS Lett. 334, 347–350
14 Boyle, J. W., Van der Green, P. and Hunter, T. (1991) Phosphopeptide mapping and
phosphoamino acid analysis by two-dimensional separation on thin-layer cellulose
plates. Methods Enzymol. 201, 111–115
15 Taylor, S. S., Radzio-Andzelm, E. and Hunter, T. (1995) How do protein kinases
discriminate between serine/threonine and tyrosine? Structural insights from the
insuline receptor protein-tyrosine kinase. FASEB J. 9, 1255–1266
16 Gibbs, C. S. and Zoller, M. (1991) Rational scanning mutagenesis of a protein kinase
identifies functional regions involved in catalysis and substrate interactions. J. Biol.
Chem. 266, 8923–8931
17 Herberg, F. W., Bell, S. M. and Taylor, S. S. (1993) Expression of the catalytic
subunit of cAMP-dependent protein kinase in E. coli : multiple isozymes reflect
different phosphorylation states. Protein Eng. 6, 771–777
18 Yonemoto, W., Garrod, S. M., Bell, S. M. and Taylor, S. S. (1993) Identification of
phosphorylation sites in the recombinant catalytic subunit of cAMP-dependent protein
kinase. J. Biol. Chem. 268, 18626–18632
19 Shoji, S., Titani, K., Demaille, J. G. and Fischer, E. H. (1979) Sequence of two
phosphorylated sites in the catalytic subunit of bovine cardiac muscle adenosine 3h :
5h-monophosphate-dependent protein kinase. J. Biol. Chem. 254, 6211–6214
20 Steinberg, R. A., Cauthron, R. D., Symcox, M. M. and Shontoh, H. (1993)
Autoactivation of catalytic (Cα) subunit of cyclic AMP-dependent protein kinase by
phosphorylation of threonine 197. Mol. Cell. Biol. 13, 2332–2341
21 Pinna, L. A. and Ruzzene, M. (1996) How do protein kinases recognize their
substrates? Biochim. Biophys. Acta 1314, 191–225
22 Hanks, S. K. (1991) Eukaryotic protein kinases. Curr. Opin. Struct. Biol. 1, 369–383
# 2002 Biochemical Society
Page 7
  • Source
    • "Archaeal PaKae1 has been shown to have DNA-binding and apurinic site endonuclease activities (Hecker et al., 2007). Bud32p is an atypical serine/threonine protein kinase (Facchin et al., 2002) that is homologous to human TP53RK/PRPK, which has been shown to phosphorylate p53 (Abe et al., 2001). In the KEOPS/EKC complex , Kae1p and Bud32p are closely associated (Mao et al., 2008). "
    [Show abstract] [Hide abstract] ABSTRACT: The kaeA(KAE1) (suDpro) gene, which was identified in Aspergillus nidulans as a suppressor of proline auxotrophic mutations, encodes the orthologue of Saccharomyces cerevisiae Kae1p, a member of the evolutionarily conserved KEOPS/EKC (Kinase, Endopeptidase and Other Proteins of Small size/Endopeptidase-like and Kinase associated to transcribed Chromatin) complex. In yeast, this complex has been shown to be involved in tRNA modification, transcription, and genome maintenance. In A. nidulans, mutations in kaeA result in several phenotypic effects, the derepression of arginine catabolism genes, and changes in the expression levels of several others, including genes involved in amino acid and siderophore metabolism, sulfate transport, carbon/energy metabolism, and translation and transcription regulation, such as rcoA(TUP1), which encodes the global transcriptional corepressor. Copyright © 2015. Published by Elsevier B.V.
    Full-text · Article · Jul 2015 · Gene
  • Source
    • "Bud32 was initially described as an RIO-type serine/threonine protein kinase and its phosphotransferase activity was demonstrated in vitro and in vivo (30). We therefore tested if the Bud32 ATPase activity corresponded to the protein kinase activity under the conditions used for the synthesis of t6A in vitro. "
    [Show abstract] [Hide abstract] ABSTRACT: N6-threonylcarbamoyladenosine (t6A) is a universal tRNA modification essential for normal cell growth and accurate translation. In Archaea and Eukarya, the universal protein Sua5 and the conserved KEOPS/EKC complex together catalyze t6A biosynthesis. The KEOPS/EKC complex is composed of Kae1, a universal metalloprotein belonging to the ASHKA superfamily of ATPases; Bud32, an atypical protein kinase and two small proteins, Cgi121 and Pcc1. In this study, we investigated the requirement and functional role of KEOPS/EKC subunits for biosynthesis of t6A. We demonstrated that Pcc1, Kae1 and Bud32 form a minimal functional unit, whereas Cgi121 acts as an allosteric regulator. We confirmed that Pcc1 promotes dimerization of the KEOPS/EKC complex and uncovered that together with Kae1, it forms the tRNA binding core of the complex. Kae1 binds l-threonyl-carbamoyl-AMP intermediate in a metal-dependent fashion and transfers the l-threonyl-carbamoyl moiety to substrate tRNA. Surprisingly, we found that Bud32 is regulated by Kae1 and does not function as a protein kinase but as a P-loop ATPase possibly involved in tRNA dissociation. Overall, our data support a mechanistic model in which the final step in the biosynthesis of t6A relies on a strictly catalytic component, Kae1, and three partner proteins necessary for dimerization, tRNA binding and regulation.
    Full-text · Article · Aug 2013 · Nucleic Acids Research
  • Source
    • "Bud32p is an atypical kinase that lacks some of the canonical domains required for substrate recognition (5,6). In vitro phosphorylation assays have demonstrated that this protein is a functional kinase that is also responsible for its autophosphorylation (8). It has been suggested that the association of Kae1p with Bud32p inhibits the phosphorylation activity of the latter (6). "
    [Show abstract] [Hide abstract] ABSTRACT: The EKC/KEOPS complex is universally conserved in Archaea and Eukarya and has been implicated in several cellular processes, including transcription, telomere homeostasis and genomic instability. However, the molecular function of the complex has remained elusive so far. We analyzed the transcriptome of EKC/KEOPS mutants and observed a specific profile that is highly enriched in targets of the Gcn4p transcriptional activator. GCN4 expression was found to be activated at the translational level in mutants via the defective recognition of the inhibitory upstream ORFs (uORFs) present in its leader. We show that EKC/KEOPS mutants are defective for the N6-threonylcarbamoyl adenosine modification at position 37 (t(6)A(37)) of tRNAs decoding ANN codons, which affects initiation at the inhibitory uORFs and provokes Gcn4 de-repression. Structural modeling reveals similarities between Kae1 and bacterial enzymes involved in carbamoylation reactions analogous to t(6)A(37) formation, supporting a direct role for the EKC in tRNA modification. These findings are further supported by strong genetic interactions of EKC mutants with a translation initiation factor and with threonine biosynthesis genes. Overall, our data provide a novel twist to understanding the primary function of the EKC/KEOPS and its impact on several essential cellular functions like transcription and telomere homeostasis.
    Full-text · Article · Apr 2011 · Nucleic Acids Research
Show more