In the present study, we demonstrate that, in addition to AtPAH1 and AtPAH2, homologous genes from Bras-sica napusalso encode functional PAP enzymes capable of complementing different ph
Trang 1restore lipid synthesis in a pah1D mutant strain of
Saccharomyces cerevisiae
Elzbieta Mietkiewska1, Rodrigo M P Siloto1, Jay Dewald2, Saleh Shah3, David N Brindley2and Randall J Weselake1
1 Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Canada
2 Department of Biochemistry, Signal Transduction Research Group, School of Molecular and Systems Medicine, University of Alberta, Edmonton, Canada
3 Plant Biotechnology, Alberta Innovates-Technology Futures, Vegreville, Canada
Keywords
lipin; phosphatidate phosphatase;
Saccharomyces cerevisiae; subcellular
localization; triacylglycerol
Correspondence
R J Weselake, Agricultural Lipid
Biotechnology Program, Department of
Agricultural, Food and Nutritional Science,
4–10 Agriculture ⁄ Forestry Centre, University
of Alberta, Edmonton, Alberta T6G 2P5,
Canada
Fax: +1 780 492 6739
Tel: +1 780 492 4401
E-mail: randall.weselake@ualberta.ca
(Received 21 September 2010, revised 10
December 2010, accepted 20 December
2010)
doi:10.1111/j.1742-4658.2010.07995.x
The identification of the yeast phosphatidate phosphohydrolase (PAH1) gene encoding an enzyme with phosphatidate phosphatase (PAP; 3-sn-phosphati-date phosphohydrolase, EC 3.1.3.4) activity led to the discovery of mam-malian Lipins and subsequently to homologous genes from plants In the present study, we describe the functional characterization of Arabidopsis and Brassica napus homologs of PAH1 Recombinant expression studies confirmed that homologous PAHs from plants can rescue different pheno-types exhibited by the yeast pah1D strain, such as temperature growth sen-sitivity and atypical neutral lipid composition Using this expression system, we examined the role of the putative catalytic motif DXDXT and other conserved residues by mutational analysis Mutants within the carboxy-terminal lipin domain displayed significantly decreased PAP activity, which was reflected by their limited ability to complement different phenotypes of pah1D Subcellular localization studies using a green fluores-cent protein fusion protein showed that Arabidopsis PAH1 is mostly pres-ent in the cytoplasm of yeast cells However, upon oleic acid stimulation, green fluorescent protein fluorescence was predominantly found in the nucleus, suggesting that plant PAH1 might be involved in the transcrip-tional regulation of gene expression In addition, we demonstrate that mutation of conserved residues that are essential for the PAP activity of the Arabidopsis PAH1 enzyme did not impair its nuclear localization in response to oleic acid In conclusion, the present study provides evidence that Arabidopsis and B napus PAHs restore lipid synthesis in yeast and that DXDXT is a functional enzymic motif within plant PAHs
Database The nucleotide sequence data have been deposited in the GenBank database under accession numbers HQ113853 and HQ113854
Abbreviations
C-LIP, carboxy-terminal lipin domain; DAG, diacylglycerol; DAPI, 4,6-diamidino-2-phenylindole dilactate; DGAT, diacylglycerol acyltransferase; FAME, fatty acid methyl ester; GFP, green fluorescent protein; HAD, haloacid dehalogenase; LPP, lipid phosphate phosphatase; N-LIP, amino-terminal lipin domain; PA, phospatidate; PAH, phosphatidate phosphohydrolase; PAP, phosphatidate phosphatase; PL, phospholipid; TAG, triacylglycerol; WT, wild-type.
Trang 2Phosphatidate phosphatase (PAP; 3-sn-phosphatidate
phosphohydrolase, EC 3.1.3.4) catalyzes the
dephos-phorylation of phospatidate (PA), yielding
diacyl-glycerol (DAG) and inorganic phosphate [1–3] In
eukaryotic cells, PAP activity plays a central role in both
lipid metabolism and intracellular signaling mechanisms
[4,5] Two distinct PAP enzyme activities, referred to as
PAP1 and PAP2, have been described [2,6–8] PAP1,
now known as PAP activity is Mg2+-dependent, utilizes
PA as a unique substrate and localizes in the soluble
fraction, from where it translocates to internal
mem-branes [3,9,10] By contrast, PAP2, currently known as
a family of lipid phosphate phosphatases (LPPs) utilizes
many substrates (e.g PA, lysophosphatidate,
sphingo-sine-1-phosphate and DAG pyrophosphate, amongst
others), does not require Mg2+for activity and is
mem-brane-bound [11] Genes encoding the LPPs have been
identified and extensively studied in both yeast and
plants [12–15]
Studies in Saccharomyces cerevisiae led to the
identifi-cation of the yeast phosphatidate phosphohydrolase
(PAH1) gene encoding an enzyme with PAP activity
[16] The S cerevisiae pah1D strain displays severe
defi-ciency in triacylglycerol (TAG), which is apparently
caused by a decreased level of DAG, reflecting the
importance of this enzyme for de novo synthesis of
neu-tral lipids In addition, the free fatty acid content of
pah1D yeast was significantly increased, presumably as a
result of the decreased ability of cells to utilize fatty
acids for TAG synthesis [16] The PAH1 enzyme was
found in soluble and membrane fractions of the cell and
its association with membranes was shown to be of a
peripheral nature [2] The S cerevisiae PAH1 shares
sequence homology with mammalian Lipin-1 in the
con-served N-terminal and C-terminal domains of the
pro-tein known as N-LIP and C-LIP, respectively [2,3,17]
The C-LIP domain contains a haloacid dehalogenase
(HAD)-like catalytic motif found in the superfamily of
Mg2+-dependent PAPs This motif consists of DXDXT
in which the first aspartate residue is responsible for
binding the phosphate moiety during the catalytic
reac-tion [18,19] In addireac-tion, a conserved glycine within
N-LIP is required for the PAP activity of yeast PAH1
[19] PAH1 is phosphorylated on seven residues
match-ing the minimal Cdk consensus that is required for the
efficient transcriptional derepression of key enzymes
involved in phospholipid (PL) biosynthesis [20,21] In
mammals, highly phosphorylated forms of Lipin-1 are
enriched in the cytoplasm, whereas dephosphorylated
forms are found mostly associated with membranes
[9,10,22,23] Homologous PAH1 genes are also present
in Arabidopsis thaliana and other plants Nakamura
et al.[24] showed that Arabidopsis PAH1and PAH2 are involved in the eukaryotic pathway of galactolipid bio-synthesis Recently, it was demonstrated in Arabidopsis that PAHs regulate PL synthesis in a similar manner to that described for S cerevisiae [25]
In the present study, we demonstrate that, in addition
to AtPAH1 and AtPAH2, homologous genes from Bras-sica napusalso encode functional PAP enzymes capable
of complementing different phenotypes of yeast pah1D strain, including TAG synthesis and temperature-sensi-tive growth We also provide evidence that the con-served aspartate residues in the HAD-like motif of the plant enzymes are required for PAP function, and demonstrate the importance of conserved glycine and serine residues for the activity of the enzyme Finally, we show that, when expressed in yeast, AtPAH1 can be detected in cytosolic and membrane fractions, although
it is recruited to the nucleus when the cells are cultivated
in the presence of oleic acid A preliminary account of some of the results reported in the present study has been presented at the 19th International Symposium on Plant Lipids [26]
Results and Discussion
Arabidopsis and B napus PAHs are orthologs of yeast PAH1 and mammalian lipins
Based on previous studies demonstrating the involve-ment of yeast PAH1 and mammalian lipins in storage lipid accumulation [27], we were initially interested in identifying Arabidopsis genes encoding PAP enzymes that may play a role in de novo DAG synthesis Two PAH1 homologs were found in Arabidopsis, which are designated in the present study as AtPAH1 (At3g09560) and AtPAH2 (At5g42870), and encode proteins with predicted molecular masses of 100.9 and 101.2 kDa, respectively In addition, we identified two isoforms of B napus PAH1 using the expressed sequence tag information available on the Internet (http://brassica.bbsrc.ac.uk) They are designated in the present study as BnPAH1A (GenBank # HQ113853) and BnPAH1B (GenBank # HQ113854), and encode proteins sharing 99% sequence identity and calculated molecular masses of 90.6 kDa Both Arabidopsis and B napus PAHs have the conserved N-LIP and C-LIP domains found in yeast PAH1 and mammalian Lipin-1 AtPAH1 and AtPAH2 share 63% and 58% identity in the conserved N-LIP and C-LIP domain, respectively B napus PAHs display higher
Trang 3sequence identity to AtPAH1 at the level of 95% for
each N-LIP and C-LIP domain The C-LIP domain
harbors the HAD-like catalytic motif (DXDXT) found
in the superfamily of Mg2+-dependent phosphatases
[17,27] The HAD-like motif DVDGT is located at
positions 707–711, 734–738 and 616–620 of AtPAH1,
AtPAH2 and BnPAH1, respectively (Fig 1) By
con-trast to plant PAHs, members of the family of LPPs
that do not require Mg2+ ions for activity in
Arabid-opsis and yeast contain a catalytic motif comprising
the consensus sequences KXXXXXXRP, PSGH and
SRXXXXXHXXXD [2,5,28,29] Analyses made with
several prediction algorithms were unable to detect the
presence of potential transmembrane domains in
Arabidopsis or B napus PAHs This is in agreement
with earlier studies on mammalian and yeast homologs
[3,16,17] Unlike Arabidopsis and B napus PAHs, six
putative transmembrane domains were predicted in
Arabidopsis LPP2 and LPP3 and in yeast
diacylgly-cerol pyrophosphate phosphatase 1 and LPP1 that belong to the family of Mg2+-independent PAP2 enzymes [14,16,30]
Arabidopsis and B napus PAHs complement different phenotypes of pah1-deficient yeast cells
To establish the functional relationship between yeast PAH1 and the corresponding plant homologs, the coding region of Arabidopsis and B napus PAHs were linked to the galactose-inducible GAL1 promoter in the yeast expression vector pYES2⁄ NT, which also provides an N-terminal His-tag The resulting con-structs were transformed into the S cerevisiae pah1D strain, which displays different phenotypes, such as a reduced level of PAP activity, severe growth deficiency
at 37C, elevated levels of PA and decreased levels of DAG and TAG [16,19,31] Transformed yeast cells were cultivated for 16 h after induction with galactose Immunoblot analysis of yeast homogenates showed that the His-tagged Arabidopsis PAH1 and B napus PAHs migrated as 130 and 118 kDa proteins, respectively, upon SDS⁄ PAGE (Fig 2A, B), and these values are higher than the molecular weights of the predicted polypeptides A similar shift in mobility was observed in yeast PAH1 and was attributed to post-translational modification of the protein by phosphor-ylation [16,20] AtPAH2 migrated at a lower molecular weight compared to AtPAH1 and, in this case, a sec-ond band is clearly noticeable on SDS⁄ PAGE (Fig 2A, lane 2) The same pattern has been also observed previously for mammalian Lipin-2 [22] Frac-tionation of yeast homogenate indicated that AtPAH1 was detected in both membrane and soluble fractions
of the yeast cell (Fig 2C) Therefore, to evaluate
Mg2+-dependent PAP activity, we used crude cell homogenates The pah1D strain expressing Arabidopsis
or B napus PAHs displayed a significant increase in
ScPAH1
AtPAH1
AtPAH2
BnPAH1A
BnPAH1B
DIDGT
N-LIP HAD-like
G
DVDGT S G
DVDGT G
DVDGT
A A G
A
DVDGT S G
Fig 1 PAH1 homologs from plants have similar domain
organiza-tion to yeast PAH1 (ScPAH1) polypeptide Arabidopsis PAH1
(At-PAH1), Arabidopsis PAH2 (AtPAH2) and B napus PAHs (BnPAH1A
and BnPAH1B) are members of the lipin family, containing a
conserved N-terminal domain (N-LIP) and a C-terminal catalytic
domain with a HAD-like motif usually found in Mg 2+ -dependent
phosphatidate phosphatases The conserved amino acids that were
mutated in the present study are indicated.
1
kDa
160
kDa
120 100
kDa
120 160
120
Fig 2 Plant PAHs expressed in yeast migrate higher on SDS ⁄ PAGE than the predicted molecular masses of polypeptides and are found in both soluble and membrane fractions Immunoblots of Arabidopsis and B napus PAHs were carried out using anti-HisG-HRP serum and pro-tein extracts from yeast pah1D expressing the recombinant polypeptides after 16 h of induction Propro-teins (40 lg each lane) were run on an 8% SDS ⁄ PAGE gel Protein molecular mass was calculated using BenchMark Protein Ladder (Invitrogen) (A) Crude homogenates from cells expressing AtPAH1 (lane 1) and AtPAH2 (lane 2) (B) Crude homogenates from cells expressing BnPAH1A (lane 1) and BnPAH1B (lane 2) (C) Subcellular fractions from cells expressing AtPAH1: microsomal fraction (lane 1) and soluble fraction (lane 2) Membrane and soluble fractions were obtained using 100 000 g centrifugation of the 15 000 g supernatant of crude homogenate.
Trang 4PAP activity compared to the negative control trans-formed with LacZ (Fig 3A) The highest difference in PAP activity was observed for BnPAH1A, with an increase of 21.4-fold compared to the negative control
In the pah1D mutant, the PAP activity was reduced by 39% compared to the wild-type (WT) (Inv Sc1 strain) control used in the present study This was similar to the difference between pah1D and WT parent strain observed by Han et al [16] The remaining Mg2+ -dependent PAP activity in pah1D has been attributed
to other enzymes with unknown molecular identities [16]
To evaluate the influence of Arabidopsis and B na-pusPAHs in the metabolism of neutral lipids, we ana-lyzed the TAG and PL content of yeast cells expressing recombinant PAHs (Fig 3B) After 48 h of induction, the TAG⁄ PL ratio in pah1D cells bearing recombinant PAHs was considerably higher compared
to the negative control The most pronounced effect was observed for BnPAH1A, with an increase of 40-fold compared to the TAG⁄ PL ratio observed in the negative control (LacZ)
In addition to the effect on lipid composition, yeast pah1D cells also display reduced growth when culti-vated at 37C To determine whether Arabidopsis and
B napus PAHs could rescue this phenotype, we culti-vated several dilutions of cells expressing plant PAHs
at 37C When cells were inoculated in medium supplemented with galactose (induced), lines expressing PAHsfrom Arabidopsis and B napus displayed growth
on dilutions as low as D600= 1.0· 104 (Fig 3C), whereas cells expressing LacZ grew only at
D600= 1.0 In medium without galactose (not induced), only WT cells presented appreciable growth, indicating that complementation of temperature-sensi-tive phenotype resulted from Arabidopsis and B napus PAHsexpression
Taken together, these results show that the previ-ously characterized AtPAH1 and AtPAH2 and the two PAH1 homologs from B napus encode enzymes with PAP activity Previous work on mammalian lipins indicated that the pah1D yeast expression system could
be used as a predictive model for confirming the func-tions of PAH1-homolog genes [22] Arabidopsis and
B napus PAHs complemented the temperature sensi-tive phenotype from the yeast pah1D strain, which was also observed by Nakamura et al [24] for AtPAH1 and AtPAH2 Using an Escherichia coli expression sys-tem, Eastmond et al [25] reported a comparatively higher enzyme activity of AtPAH1 over AtPAH2, which is corroborated by the results obtained in the present study (Fig 3A) The higher enzyme activity of AtPAH1 compared to AtPAH2 is also evident in the
BnPAH1B
AtPAH1
AtPAH2
LacZ
LacZ
BnPAH1A
pah1Δ
Wild type
OD
1 10 10 2 10 3 10 4 Dilutions
C
A
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
–1 ·m
–1 )
0.0 1.0 2.0 3.0 4.0 5.0
B
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Fig 3 Arabidopsis and B napus PAHs complement different
phe-notypes of S cerevisiae pah1D mutant (A) PAP activity of crude
homogenates from yeast pah1D cells bearing AtPAH1, AtPAH2,
BnPAH1A, BnPAH1B or LacZ as a negative control Yeast
homo-genates were prepared from cells induced for 16 h and assayed for
PAP activity in the presence of 1 m M MgCl2 Total PAP activities
were calculated from measurements at three different protein
con-centrations (B) Ratio of TAG to PL of yeast pah1D expressing
recombinant PAHs Total lipids were extracted from yeast cells
after 48 h of induction and separated on TLC plates Spots
corre-sponding to TAG and PL were scraped out and transmethylated.
FAMEs were analyzed by GC Each bar represents the mean ± SD
from three determinations (C) Complementation of temperature
sensitive phenotypes of pah1D by plant PAH homologs Yeast
pah1D expressing AtPAH1, AtPAH2, BnPAH1A, BnPAH1B or LacZ
were cultivated in liquid medium The density of resulting cultures
was adjusted to D 600 = 1 followed by 10-fold serial dilutions Five
microliters of each dilution were spotted onto plates containing 2%
galactose (induced) or 2% raffinose (not induced) and incubated for
3 days at 37 C WT yeast and pah1D strain previously transformed
with LacZ were used as a positive and negative control,
respec-tively.
Trang 5preferential growth at 37C of yeast expressing
AtPAH1; this was also reported by Nakamura et al
[24] Taken together, these results collectively suggest
that AtPAH1 can complement phenotypes of the
pah1D strain more efficiently than AtPAH2 The
newly-characterized BnPAH1A and BnPAH1B display
relatively stronger complementation over Arabidopsis
PAHs in most aspects Moreover, the restoration of
TAG synthesis in the yeast pah1D strain expressing
plant PAH homologs suggests an evolutionary
conser-vation of the PAP enzyme reaction between yeast and
plants
N-LIP and C-LIP are functional domains in
Arabidopsis and B napus PAHs
To determine the role of the conserved residues within
C-LIP and N-LIP domain of plant PAH1, we
exam-ined the mutational effect of selected residues on PAP
activity using AtPAH1 and BnPAH1A as models
Using site-directed mutagenesis (Fig 1), we
con-structed mutant AtPAH1 alleles (G83A, D707A,
D709A and S752A) and mutant BnPAH1A alleles
(G83A, D616A and D618A) and expressed them in the
pah1D strain Yeast cells were harvested after 16 h of
induction in medium containing galactose Cell
homo-genates were prepared, verified via immunoblotting
and assayed for PAP activity Immunoblot analysis
using anti-HisG serum showed that native and mutant
AtPAH1 and BnPAH1A enzymes were expressed at
comparable levels (Fig 4) Although the expression of
AtPAH1 and BnPAH1A resulted in PAP activity that
was significantly higher compared to LacZ control, the
corresponding mutant alleles did not restore PAP
activity to comparable levels (Fig 4) In particular,
mutations in the predicted catalytic motif of AtPAH1
(D707A and D709A) and BnPAH1A (D616A and
D618A) abolished PAP activity, with levels
compara-ble to the negative control These results are in
agree-ment with mutational analysis of the yeast PAH1
catalytic motif [19] and demonstrate that the conserved
aspartate residues in the plant homologs are required
for their catalytic function In addition, substitution of
the conserved serine 752 with alanine within C-LIP
domain of AtPAH1 had a similar negative effect on
enzyme activity The importance of the equivalent
con-served serine residue for the enzyme activity in human
Lipin-2 and mouse Lipin-1 and Lipin-2 has been
described previously [9] For example, a rare human
mutation S734L in LIPIN-2 gene causes Majeed
syndrome, a human inflammatory disorder Recently,
Majeed syndrome has been linked to the loss of
Lipin-2-mediated PAP activity [9]
Mutation of the conserved glycine (G83) to alanine
in the N-LIP domain of both AtPAH1 and BnPAH1A produced less severe effects on the enzyme activity and resulted in the loss of up to 75% and 54% of the PAP activity of native enzyme, respectively (Fig 4) Inter-estingly, other mutations in the corresponding position
of PAH1 from other organisms appear to have a more
80 100
120
A
B
0 20 40 60
0
130 kDa
80 100 120
20 40 60
0
118 kDa
Fig 4 Mutations within N-LIP and C-LIP domains affect PAP1 activity of plant PAH1 (A) PAP activity of yeast pah1D strain homo-genates bearing AtPAH1 and its mutant alleles: G83A, D707A, D709A and S752A Lower: corresponding western blot of the site-directed mutagenized AtPAH1 (B) PAP activity of yeast pah1D strain homogenates bearing BnPAH1A and its mutant alleles: G83A, D616A, D618A Lower: corresponding western blot of the site-directed mutagenized BnPAH1A Yeast homogenates were prepared from cells induced for 16 h and assayed for PAP activity
in the presence of 1 m M MgCl2 The amount of PAP activity of samples bearing AtPAH1 (34.8 nmolÆmg)1Æmin)1) and BnPAH1A (60.0 nmolÆmg)1Æmin)1) were set at 100% The data shown are the mean ± SD from three determinations Homogenates of yeast cells expressing LacZ were used as negative controls Western blot anal-ysis were carried out using anti-HisG-HRP serum and protein extracts (40 lg each lane) from yeast pah1D strain expressing the recombinant polypeptides separated on an 8% SDS ⁄ PAGE gel.
Trang 6pronounced effect For example, the G80R yeast
PAH1 allele showed essentially no enzyme activity
[19] The same substitution within N-LIP domain was
found to be crucial for the fat-regulating function of
Lipin-1 in mice [32] Mild modifications (such as G to
A) that do not result in dramatic changes in polarity
or steric properties may not be sufficient to completely
hinder enzyme activity
To determine whether the previous results with
respect to enzyme activity were related to other effects
on lipid metabolism, we examined the yeast lipid
com-position in stationary phase of pah1D cells expressing
mutant alleles of AtPAH1 and BnPAH1A (Fig 5) As
demonstrated previously, the TAG⁄ PL ratio in the
pah1D strain was increased by the expression of
AtPAH1 and BnPAH1A genes However, the
expres-sion of their respective mutant alleles within C-LIP
domain did not affect TAG⁄ PL ratio compared to the
negative control Similarly, the expression of mutant
AtPAH1 and BnPAH1A C-LIP mutant alleles did not
complement the temperature sensitivity of pah1D cells
at 37C (Fig 6) The effect of the G83A mutation in
BnPAH1A appears to be comparatively mild and
might be attributed to unique attributes of BnPAH1A
together with a more conservative change as outlined
above (Figs 4B, 5B and 6B) In conclusion, amino acid
substitutions in the conserved C-LIP domain, including
the HAD-like motif of AtPAH1 and BnPAH1A,
resulted in the loss of PAP activity Mutation of G83
in the N-LIP domain also produced a significant
reduction in enzyme activity and was less severe for
BnPAH1A The results obtained from temperature
growth sensitivity analysis correlated with changes in
lipid composition in the pah1D strain and indicate a
close relationship between enzyme activity and the
different phenotypes The findings of the present study
are in agreement with earlier mutational analysis
stud-ies carried out with yeast PAH1 [19] Furthermore, in
the study performed by Han et al [19], the lack of
complementation by the D398E and D400E mutant
PAH1 alleles was linked to the specific loss of
PAH1-encoded PAP activity
Oleic acid stimulates translocation of GFP-AtPAH1
from the cytosol to the nucleus in the yeast cell
We have previously determined that plant PAH1
homologs are present in both soluble and membrane
fractions of yeast cells (Fig 2C) To obtain a more
com-prehensive understanding of the subcellular localization
of these enzymes, we prepared a construct encoding an
N-terminal green fluorescent protein (GFP)-fusion with
AtPAH1, and expressed this construct in pah1D under
the control of the GAL1 promoter As shown in Fig 7A, the GFP-AtPAH1 fusion was present through-out the cytoplasm as a soluble protein, in agreement with the immunoblot on Fig 2C, and apparently absent
in the nucleus Previous localization studies using yeast PAH1-GFP fusions also indicated that PAH1 was present throughout the cytoplasm [16,33] Confocal microscopy of GFP-AtPAH1 fusion expressed in Nicotiana benthamiana agrobacterium-infiltrated leaves indicated that the fusion protein was located predomi-nantly in the cytosol [25] In the case of mammalian PAH1 homologs, both mouse Lipin-1 isoforms can localize to either the cytosol or nucleus The majority of Lipin-1B is present in the cytosol and the remaining Lipin-1A is prevalent in the nucleus of mature adipo-cytes [3] PAP activity is primarily cytosolic but, after fatty acid stimulation, it can be largely detected in the
0.0 0.1 0.2 0.3 0.4 0.5
A
0 0.04 0.08
0.0 0.1 0.2 0.3 0.4 0.5
B
0 0.04 0.08
Fig 5 Mutations within N-LIP and C-LIP domains influence the ability of plant PAH1 to restore TAG synthesis in yeast pah1D strain (A) TAG ⁄ PL ratio of the S cerevisiae pah1D transformed with AtPAH1 and its mutant alleles: G83A, D707A, D709A and S752A (B) TAG ⁄ PL ratio of S cerevisiae pah1D transformed with BnPAH1A and its mutant alleles: G83A, D616A, D618A Lipids were extracted from yeast cells after 48 h of induction and separated on TLC plates Spots corresponding to TAG and PL were scraped out and transmethylated FAMEs were analyzed by GC Each bar represents the mean ± SD from three determinations.
WT yeast and pah1D strain (LacZ) previously transformed with LacZ were used as a positive and negative control, respectively.
Trang 7endoplasmic reticulum, as described previously in rat
hepatocytes [34,35], as well as in the developing seeds of
safflower, following stimulation with oleic acid [36] The
yeast expression system used in the present study offers
the versatility of controlling environmental conditions
and stimuli Therefore, we aimed to determine whether
oleic acid supplementation could affect the subcellular
localization of GFP-AtPAH1 When cells were
cultivated in the presence of 125 lm oleic acid,
fluores-cence was found almost exclusively in the nucleus
(Fig 7B) This suggests that AtPAH1 might be involved
in transcriptional gene regulation, although additional
studies are required to address this hypothesis We have
shown that conserved residues G83, D707, D709 and
S752 are essential for PAP activity of Arabidopsis
PAH1 (Fig 4A) To determine the significance of these
conserved residues for nuclear localization, we
intro-duced point mutations into a GFP-AtPAH1 fusion at
the respective sites of AtPAH1 and investigated whether
they exhibited oleate-induced nuclear localization As
shown in Fig 7C–F, GFP-AtPAH1 mutant alleles localized to the nucleus of yeast cells cultivated in the presence of 125 lm oleic acid These results demonstrated that conserved amino acid residues G83, D707, D709 and S752 are required for PAP activity of AtPAH1, although they are not required for nuclear localization Previously, Santos-Rosa et al [20] indicated that yeast PAH1 could also play a role in transcriptional regulation of PL synthesis In addition, mammalian Lipin-1 has been also suggested to act as a transcriptional co-activator in the regulation of lipid
DIC
A
B
C
E D
F
Fig 7 Localization of GFP-AtPAH1 fusion in yeast cells is influ-enced by oleic acid supplementation Yeast pah1D cells expressing GFP-AtPAH1 and the ensuing mutants were cultivated for 16 h in induction medium containing 2% galactose without or with supple-mentation with 125 l M oleic acid (A) Cells expressing GFP-AtPAH1 cultivated without oleic acid (B) Cells expressing GFP-AtPAH1 culti-vated with oleic acid (C–F) Cells expressing GFP-AtPAH1 G83A, GFP-AtPAH1 D707A, GFP-AtPAH1 D709A and GFP-AtPAH1 S752A cultivated with oleic acid, respectively Nuclei were detected by DNA staining with DAPI in the blue channel and recombinant AtPAH1 was detected with GFP in the green channel Fluorescence signals were examined using a Leica TCS-SP5 multiphoton confo-cal laser scanning microscope Arrows indicate the position of a representative nucleus in each micrograph Scale bar = 2 lm DIC, Differential interference contrast.
AtPAH1
G83A
D707A
D709A
S752A
LacZ
LacZ
pah1Δ
Wild type
Induced Not induced
OD
1 10 10 2 10 3 10 4 Dilutions
BnPAH1A
G83A
D616A
D618A
LacZ
LacZ
Wild type
pah1Δ
A
B
Fig 6 Plant PAH1 homologs containing mutations in the N-LIP and
C-LIP domains fail to rescue the temperature sensitivity of pah1D
cells (A) Yeast pah1D expressing AtPAH1 and its mutant alleles:
G83A, D707A, D709A and S752A (B) Yeast pah1D expressing
BnPAH1A and its mutant alleles: G83A, D616A, D618A Yeast pah1D
expressing AtPAH1, BnPAH1A and their respective mutants were
cultivated in liquid medium The density of resulting cultures was
adjusted to D 600 = 1 followed by 10-fold serial dilutions Five
microli-ters of each dilution were spotted onto plates containing 2%
galac-tose (induced) or 2% raffinose (not induced) and incubated for 3 days
at 37 C WT yeast and pah1D strain previously transformed with
LacZ were used as a positive and negative control, respectively.
Trang 8metabolism gene expression [27,37] Similar to the
find-ings of the present study, Donkor et al [9] have shown
that the equivalent serine residue in mouse Lipin-1 and
Lipin-2 was required for PAP activity, although it was
not required for transcriptional co-activator function
The results obtained in the present study suggest that
similar mechanisms of PAH1 trafficking to the nucleus
might also be present in plants Eastmond et al [25]
determined that the expression of AtPAH1 and
AtPAH2 is considerably higher in developing seeds,
where a substantial flux of FAs to TAG occurs Similar
analyses carried out in our laboratory corroborate these
findings (Fig 8) However, analysis of lipids from seeds
of an Arabidopsis Atpah1⁄ Atpah2 double knockout
showed that TAG content was not substantially reduced
relative to WT [25] Therefore, although mRNA
accu-mulation suggests that PAHs might have a key function
in seeds, their role in storage lipid metabolism remains
unclear
Materials and methods
Plant material
A thaliana plants (Columbia-O) were cultivated in a
growth chamber at 22C with an 18 h photoperiod
(120 lEÆm)2Æs)1)
Cloning and expression of Arabidopsis and
B napus PAHs
A thaliana PAH1ORF was amplified from a cDNA clone
obtained from the Arabidopsis Biological Resource Center
using primers: F1: 5¢-ATAGGTACCTATGGGGTTGGTT GGAAGAG-3¢ (KpnI site is underlined) and R1: 5¢-CGC GCGGCCGCTCATTCTACCTCTTCTATTGGCA-3¢ (NotI site is underlined) and ligated into the pYES2⁄ NT (Invitro-gen, Burlington, ON, Canada) yeast expression vector at the KpnI and NotI restriction sites A thaliana PAH2 ORF was amplified using a cDNA preparation from developing seeds with primers: F2: 5¢-ATAGGATCCAGATGAATG CCGTCGGTAGG-3¢ (BamHI site is underlined) and R2: 5¢-CGCGCGGCCGCTCACATAAGCGATGGAGGAG-3¢ (NotI site is underlined) and then ligated into the pYES2⁄ NT vector at the BamHI and NotI sites Under the control of the GAL1 promoter, the PAH1 genes in the pYES2⁄ NT yeast vector were expressed as an N-terminal fusion protein to the Xpress epitope and polyhistidine (6· His) tag B napus PAHs were isolated using sequence information identified in ESTs database (http://brassica bbsrc.ac.uk) These partial B napus PAH1 homolog sequences were used to design primers to amplify the 5¢ and 3¢ ends of the cDNA using the SMART RACE cDNA Amplification kit (Clontech, Palo Alto, CA, USA) and a cDNA preparation from B napus developing seeds After sequence assembly to determine the full-length sequence of the cDNA, the ORF was amplified using the primer F3: 5¢-ATAGGTACCTATGAGTTTGGTCGGAAG-3¢ (KpnI site is underlined) and R3: 5¢-CGCGCGGCCGCTCAGT-CAACCTCTTCTACCG-3¢ (NotI site is underlined), and subsequently cloned into the KpnI and NotI sites of pYES2.1⁄ NT expression vector The Arabidopsis and
B napus PAHs in pYES2.1⁄ NT were transformed into
S cerevisiae mutant strain pah1D [19] using the S c EasyComp transformation kit (Invitrogen) For yeast cells, the pah1D mutant and WT (Inv Sc1 strain; Invitrogen) transformed with pYES2.1⁄ NT ⁄ lacZ plasmid (Invitrogen), designated as LacZ and WT in the present study, were used
as controls The transformants were selected and grown as described previously [38] Briefly, yeast cultures were culti-vated in minimal medium containing 0.67% (w⁄ v) yeast nitrogen base, 2% (w⁄ v) raffinose, 20 mgÆL)1adenine, argi-nine, tryptophan, methionine, histidine and tyrosine,
30 mgÆL)1lysine and 100 mgÆL)1leucine The cultures were grown at 30C in a rotary shaker at 250 r.p.m Expression
of the recombinant genes was induced using minimal medium containing 2% (w⁄ v) galactose and 1% (w ⁄ v) raffinose
Site-directed mutagenesis studies
To introduce point mutations into the Arabidopsis PAH1 coding region, a QuikChange Site-Directed Mutagenesis kit (Stratagene, Mississauga, ON, Canada) was used The primers used were: G83A (F4: 5¢-ATGTATCTTGA TAATTCTGCTGAAGCATATTTCATCAGG-3¢ and R4: 5¢-CCTGATGAAATATGCTTCAGCAGAATTATCAAG ATACAT-3¢); D707A (F5: 5¢- ACCAAGATAGTGATTT
0
1
2
3
4
5
6
7
8
Leaves Flowers Buds Roots Stems Siliques
Fig 8 Expression profile of Arabidopsis PAHs Total RNA was
obtained from tissues of mature Arabidopsis plants as well as from
developing green siliques Equal amounts of total RNA were used
for cDNA synthesis and serial dilutions of the resulting reaction
were used for quantitative RT-PCR Each bar represents the
mean ± SD from three determinations with individual reference
genes (At4g34270, At4g33380 and At1g58050).
Trang 9CAGCTGTTGATGGAACTATAAC-3¢, R5: 5¢-GTTATA
GTTCCATCAACAGCTGAAATCACTATCTTGGT-3¢);
D709A (F6: 5¢-TAGTGATTTCAGATGTTGCTGGAACT
ATAACTAAATC-3¢, R6: 5¢-GATTTAGTTATAGTTCC
AGCAACATCTGAAATCACTA-3¢); and S752A (F7:
5¢-CAGTTACTGTTTTTGGCCGCTCGTGCCATCGTTC-3¢
and R7: 5¢-GAACGATGGCACGAGCGGCCAAAAA
CAGTAACTG-3¢) The primers used to introduce a point
mutation into BnPAH1A were: G83A (F8: 5¢-TATCTAGA
CAATTCCGCGGAAGCGTATTTCATC-3¢ and R8:
5¢-GATGAAATACGCTTCCGCGGAATTGTCTAGATA-3¢);
D6161A (F9: 5¢-GATTGTAATTTCAGCTGTTGATGGA
ACTATA-3¢ and R9: 5¢-TATAGTTCCATCAACAGCTG
AAATTACAATC-3¢); and D618A (F10: 5¢-GTAATTTCA
GATGTTGCTGGAACTATAACTAAA-3¢ and R10: 5¢-TTT
AGTTATAGTTCCAGCAACATCTGAAATTAC-3¢)
Primers were complementary to opposite strands of
pYES2.1⁄ NT (Invitrogen) yeast expression vector
contain-ing either the Arabidopsis PAH1 or B napus PAH1A gene
The presence of the desired mutation was confirmed by
DNA sequencing
Preparation of the GFP-AtPAH1 fusion construct
To prepare the GFP-AtPAH1 fusion, the coding sequences
of GFP and Arabidopsis PAH1 were PCR amplified
sepa-rately using Pfx Platinum polymerase (Invitrogen), which
was used for all PCR reactions in the present study
The GFP fragment was generated by PCR with primers F11:
5¢-ATAGGTACCTATGACGCACAATCCCACTATC-3¢ (KpnI
site is underlined) and R11: 5¢-CCAACTCTTCCAACCAACCC
CATTTTGTATAGTTCATCCATGCCATG-3¢ (the sequence
found in AtPAH1 is in italics) Arabidopsis PAH1 fragment
was amplified with primers: F12: 5¢-CATGGCATGGATG
AACTATACAAAATGGGGTTGGTTGGAAGAGTTGG-3¢ (the sequence found in GFP is in bold) and R12:
5¢-CGCGCGGCCGCTCATTCTACCTCTTCTATTGGCA-3¢
(NotI site is underlined) The resulting amplicons were
combined, re-amplified with primers F11 and R12 and
then cloned into the KpnI and NotI sites of pYES2.1⁄ NT
(Invitrogen)
Point mutations: G83A, D707A, D709A and S752A at the
corresponding sites of Arabidopsis PAH1 coding region in
GFP-PAH1fusion construct were introduced with primers as
described above using QuikChange Site-Directed
Mutagenesis kit (Stratagene) and their presence was
con-firmed by DNA sequencing
Immunodetection
Total protein (40 lg) was separated onto an 8%
SDS⁄ PAGE gel using standard protocols [39] After
elec-trophoresis, proteins were electrotransferred (1.5 h at
180 mA and 4C) to poly(vinylidene difluoride) membrane
(GE Healthcare, Baie d’Urfe, Canada) using a Mini
Trans-blot (Bio-Rad, Mississauga, ON, Canada) apparatus and transfer buffer [190 mm glycine, 25 mm Tris, 0.1% SDS, 20% (v⁄ v) methanol] Anti-His G-HRP serum (Invitrogen) was used at a dilution of 1 : 10 000 The proteins were detected using the Amersham ECL Plus Western Blotting Detection kit (GE Healthcare) The fluorescent signal was detected with the Tyhoon Imaging System (GE Health-care)
Gene expression analysis Total RNA was isolated from Arabidopsis tissues with the RNeasy kit (Qiagen, Mississauga, ON, Canada) and used
to synthesize single-stranded cDNA with the Superscript II reverse transcriptase followed by RNAse H treatment (both obtained from Invitrogen) The product of these reactions was used for quantitative RT-PCR using the Platinum SYBR Green qPCR (Invitrogen) in accordance with the manufacturer’s instructions PCR was performed in a 7900HT Fast Real-Time PCR System (Applied Biosystems, Carlsbad, CA, USA) and efficiency was calculated through serial dilutions of the initial amount of RNA The relative expression level was calculated using the comparative
Ctmethod after normalizing to controls using three reference genes (At4g34270, At4g33380 and At1g58050) with stable expression levels in Arabidopsis [40] The pair of primers used for each reference gene was: At4g34270 Ref1Fwd 5¢-CATACTGTGGAAGTGAAGTAGTTGAGAA-3¢ and Ref1Rev 5¢-CTTCCCCCTTTGGATTAGCTTT-3¢; At4g33380 Ref2Fwd 5¢-TTTGAAAAGCTTTGAGGACAAATCT-3¢ and Ref2Rev 5¢-TTCTCATTGCGCCACGTTT-3¢; At1g58050 Ref3Fwd 5¢-GAATTGCCAGTGAACTTTTCTAACG-3¢ and Ref3Rev 5¢-TCAGCAGACACATTCCAATCTTTC-3¢; AtPAH1 AtPAH1Fwd 5¢-TCACCAGATGGCCTATTTC CA-3¢ and AtPAH1Rev 5¢-GATCTTGAACTCATGAGG TGCTCTT-3¢; and AtPAH2 AtPAH2Fwd 5¢-GCCTCAGT CACAAGACAATTTCTAGT-3¢ and AtPAH2Rev 5¢-AGG CCCATCCGGCAAT-3¢
Lipid analysis Total lipids were extracted from induced yeast cells by the method of Bligh and Dyer [41] The internal standards
of 1,2-dinonadecanoyl-sn-glycero-3-phosphocholine (19 : 0-phosphatidylcholine; 100 lg in methanol) and triheptad-ecanoin (17 : 0-TAG; 50 lg in chloroform) were added to each sample to permit quantitative fatty acid analysis Lipid extracts were separated by 1D TLC on silica gel plates (SIL G25, 0.25 mm; Macherey-Nagel, Du¨ren, Germany) using the solvent system hexane⁄ diethyl ether ⁄ glacial acetic acid (70 : 30 : 1 v⁄ v) Lipid classes were visualized under UV after spraying with 0.05% primuline solution Spots corre-sponding to TAG and PL were scraped out and transme-thylated with 3 m methanolic HCl at 80C for 1 h The fatty acid methyl esters (FAMEs) were extracted with
Trang 10hexane and dried under N2 Finally, FAMEs were
resus-pended in 1 mL of iso-octane with an internal standard
(21 : 0, methyl heneicosanoin, 0.1 mgÆmL)1) FAMEs were
analyzed on an Agilent6890N Gas Chromatograph (Agilent
Technologies, Wilmington, DE, USA) with a 5975 inert XL
Mass Selective Detector equipped with an auto sampler
FAMEs were separated using a DB-23 capillary column
(30 m· 0.25 mm · 0.25 lm) with a constant helium flow
of 1.2 mLÆmin)1and the temperature program: 165C hold
for 4 min, 10CÆmin)1 to 180C, hold 5 min and
10CÆmin)1 to 230C hold 5 min Integration events were
detected and identified between 2 and 19.5 min, and
com-pared against a Nu-Chek 463 gas-liquid chromatography
standard (Nu-Chek Prep, Inc., Elysian, MN, USA)
Preparation of yeast homogenates and PAP
enzyme assay
Yeast homogenates were prepared essentially as described
by Han et al [16] Briefly, cells were harvested and washed
with 5 mL of ice-cold isolation buffer (50 mm Tris⁄ HCl,
pH 7.5, 300 mm sucrose, 2 mm dithiothreithol and 0.5 mm
phenylmethylsulfonyl fluoride), pelleted by centrifugation
and resuspended in 500 lL of isolation buffer All buffers
were pre-treated with AG 50W-X8 (Bio-Rad) ion exchange
resin Na+ salt form to minimize the presence of Mg2+
Cells were broken using three 60-s pulses with a
Mini-BeadbeaterTM (BioSpec Products, Bartlesville, OK, USA)
using 0.5 mm glass beads The homogenate was collected
and briefly centrifuged to remove unbroken cells The
protein concentration of each lysate preparation was
determined using the Bio-Rad method [42]
For the PAP enzyme assay, initially, we used the procedure
described by Han et al [16] Essentially, this procedure was
designed to study the kinetic of purified PAP using a surface
dilution kinetic model in which PA is dispersed in micelles of
Triton X100 The Mg2+-dependent activity, which
distin-guishes PAP from LPP activity, was approximately half of
the total activity and therefore our differential assay was
subject to a larger error than anticipated We then compared
this assay with one that we had designed to measure PAP in
homogenates of mammalian cells [43,44]
This latter assay maximizes the level of PAP activity and
decreases that of LPP activity Our optimized assay system
contained in a final volume of 0.1 mL: 100 mm Tris buffer,
pH 7.5, 1 mm MgCl2, 200 lm tetrahydrolipstatin (to inhibit
diacylglycerol lipases), 2 mgÆmL)1 fatty acid-poor bovine
serum albumin and 0.6 mm PA labeled with [3H]palmitate
( 1 · 105
d.p.m.⁄ assay), which was dispersed in 0.4 mm
phosphatidylcholine, and 1 mm EDTA plus 1 mm EGTA
that was used to prepare the lipid substrate Mg2+ was
removed from all buffers by treating with AG 50W-X8
(Bio-Rad) ion exchange resin Na+salt form [44] Reactions
were stopped after incubation at 30C with 2.2 mL of
chloroform containing 0.08% olive oil as a carrier for
neu-tral lipids Next, 0.8 g of basic alumina was added to absorb the PA and any [3H]palmitate formed by phospholi-pase A type activities [44] The tubes were centrifuged and
1 mL of the chloroform, which contained the [3H]DAG product, was dried and quantified by scintillation counting The times of incubation (normally 30 min) were adjusted so that < 15% of the PA was consumed during the incuba-tion Total PAP activities were calculated from measure-ments at three different protein concentrations to ensure the proportionality of the assay Parallel incubations were performed in the absence of Mg2+ to block PAP activity and to measure the LPP activity, which had to be sub-tracted from the total to give the PAP activity This method gave 10-fold greater total activity than the Triton X-100 micelle assay The Mg2+-independent LPP activity was only 10% of the total activity Therefore, this assay provided us with a more accurate method of determining PAP activity in homogenates where the measurement of kinetic constants was not required
Confocal microscopy Yeast pah1D cells expressing the GFP-AtPAH1 fusion were induced for 16 h using minimal medium containing 2% (w⁄ v) galactose, 1% (w ⁄ v) raffinose, 0.6% ethanol ⁄ tylox-apol (5 : 1, v⁄ v) without or with supplementation with
125 lm oleic acid A Leica TCS-SP5 multiphoton confocal laser scanning microscope (Leica Microsystems, Wetzlar, Germany) was used to examine the subcellular localization
of GFP fusions in yeast cells For the imaging of GFP, a
488 nm laser excitation was used at 30% and 520–570 nm emission Nuclei were identified by DNA staining with 4,6-diamidino-2-phenylindole dilactate (DAPI; Sigma-Aldrich, Oakville, ON, Canada) Briefly, 5 lL of fresh cell suspension were mixed with 5 lL of 80% glycerol contain-ing 50 ngÆmL)1 of DAPI The mix was placed onto speci-men slides, covered with a cover glass and visualized immediately Imaging of DAPI was conducted using a
405 nm laser excitation at 10% and 420–450 nm emission Data were acquired using a· 63 ⁄ 1.2 HCX PL APO objec-tive
Acknowledgements
We are grateful to Dr S Siniossoglou for providing the pah1D mutant yeast strains D.N.B is a Senior Scientist for the Alberta Heritage Foundation for Medical Research This work was supported by Alberta Innovates Bio Solutions, the Natural Sciences and Engi-neering Research Council of Canada, the Canada Research Chairs Program, the Canada Foundation for Innovation and the University of Alberta We also thank Crystal Snyder for her critical assessment of the manuscript