Structured digital abstract l MINT-7909478 : DHH1 uniprotkb: Q4DIE1 physically interacts MI:0915 with PABP2 uni-protkb: Q27335 by anti bait coimmunoprecipitation MI:0006 l MINT-790
Trang 1mRNPs in Trypanosoma cruzi
Fabı´ola B Holetz1, Lysangela R Alves1, Christian M Probst1, Bruno Dallagiovanna1, Fabricio K Marchini1, Patricio Manque2,3, Gregory Buck2, Marco A Krieger1, Alejandro Correa1and Samuel Goldenberg1
1 Instituto Carlos Chagas ⁄ FIOCRUZ, Curitiba, Brazil
2 Center for the Study of Biological Complexity, Virginia Commonwealth University, Richmond, VA, USA
3 Instituto de Biotecnologia, Universidad Mayor, Santiago, Chile
Keywords
differentiation; P-bodies; stress granules;
translation control; Trypanosoma
Correspondence
S Goldenberg, Instituto Carlos
Chagas⁄ FIOCRUZ, Rua Professor Algacyr
Munhoz Mader 3775, Curitiba 81350-010
PR, Brazil
Fax: (55) 41 3316 3230
Tel: (55) 41 3316 3230
E-mail: sgoldenb@fiocruz.br
(Received 29 January 2010, revised 3 June
2010, accepted 23 June 2010)
doi:10.1111/j.1742-4658.2010.07747.x
In trypanosomatids, the regulation of gene expression occurs mainly at the post-transcriptional level Previous studies have revealed nontranslated mRNA in the Trypanosoma cruzi cytoplasm Previously, we have identified and cloned the TcDHH1 protein, a DEAD box RNA helicase It has been reported that Dhh1 is involved in multiple RNA-related processes in vari-ous eukaryotes It has also been reported to accumulate in stress granules and processing bodies of yeast, animal cells, Trypanosoma brucei and
T cruzi TcDHH1 is localized to discrete cytoplasmic foci that vary depending on the life cycle status and nutritional conditions To study the composition of mRNPs containing TcDHH1, we carried out immunopre-cipitation assays with anti-TcDHH1 using epimastigote lysates The protein content of mRNPs was determined by MS and pre-immune serum was used as control We also carried out a ribonomic approach to identify the mRNAs present within the TcDHH1 immunoprecipitated complexes For this purpose, competitive microarray hybridizations were performed against negative controls, the nonprecipitated fraction Our results showed that mRNAs associated with TcDHH1 in the epimastigote stage are those mainly expressed in the other forms of the T cruzi life cycle These data suggest that mRNPs containing TcDHH1 are involved in mRNA metabo-lism, regulating the expression of at least epimastigote-specific genes
Structured digital abstract
l MINT-7909478 : DHH1 (uniprotkb: Q4DIE1 ) physically interacts ( MI:0915 ) with PABP2 (uni-protkb: Q27335 ) by anti bait coimmunoprecipitation ( MI:0006 )
l MINT-7909338 : DHH1 (uniprotkb: Q4DIE1 ) physically interacts ( MI:0914 ) with ATP-depen-dent DEAD ⁄ H RNA helicase, putative (uniprotkb: Q4DIE1 ), Actin, putative (uniprotkb:
Q4D7A6 ), Actin, putative (uniprotkb: Q4CLA9 ), Chaperonin HSP60, mitochondrial (uni-protkb: Q4DYP6 ), ATP-dependent Clp protease subunit, heat shock protein 100 (HSP100), putative (uniprotkb: Q4CNM5 ), Elongation factor 2, putative (uniprotkb: Q4D5X0 ), Elonga-tion factor 1-alpha (EF-1-alpha), putative (uniprotkb: Q4CU73 ), Heat shock protein 85, putative (uniprotkb: Q4CQS6 ), Glutamate dehydrogenase, putative (uniprotkb: Q4DWV8 ), Putative uncharacterized protein (uniprotkb: Q4CNI8 ), 40S ribosomal protein S11, putative (uniprotkb: Q4CRH9 ), Sterol 24-c-methyltransferase, putative (uniprotkb: Q4CMB7 ), Heat shock protein 70 (HSP70), putative (uniprotkb: Q4DTM9 ), Glutamate dehydrogenase, puta-tive (uniprotkb: Q4D5C2 ) and Calpain-like cysteine peptidase, putative (uniprotkb: Q4CYU3 )
by anti bait coimmunoprecipitation ( MI:0006 )
l MINT-7909469 : DHH1 (uniprotkb: Q4DIE1 ) physically interacts ( MI:0915 ) with PABP1 (uniprotkb: Q4E4I9 ) by anti bait coimmunoprecipitation ( MI:0006 )
Abbreviations
eIF, eukaryotic initiation factor; IP, immunoprecipitated; MASP, mucin-associated surface protein; PABP, poly(A)-binding protein; P-bodies, processing bodies; SG, stress granule; SP, supernatant.
Trang 2The life cycle of the three trypanosomatid parasites
pathogenic to humans involves the alternation
between a mammal and an arthropod host The two
main developmental stages of Trypanosoma brucei are
extracellular, in both mammalian host blood and the
tsetse fly intestine [1] Trypanosoma cruzi exists in its
amastigote form in the cytoplasm of the mammal
infected cell, or as an epimastigote in the kissing bug
intestine [2] Protozoans of the Leishmania genus, in
turn, multiply in the form of amastigotes within
mac-rophages and as extracellular promastigotes in the
insect vector [3] The regulation of gene expression
thus plays a major role in determining the adaptation
and differentiation of these parasites throughout their
life cycle
Trypanosomatids diverge from other eukaryotes in
several aspects, including the editing of mitochondrial
RNA molecules, trans-splicing, genes almost never
interrupted by introns and a lack of typical promoter
sequences for the transcription of protein-encoding
genes In addition, transcription by RNA
polymer-ase II is constitutive and genes in the same
polycis-tronic unit display different levels of processed mRNA
[4–6] Consequently, the regulation of gene expression
in trypanosomatids seems to occurs mainly
post-trans-criptionally [6,7]
When mRNAs enter the cytoplasm, they can be
translated, stored for later translation or degradation,
degraded or subjected to a combination of these
processes In mammalian cells, mRNAs that are not
translated or destined for degradation are
compart-mentalized in distinct cytoplasmic structures, known as
‘processing bodies’ (‘P-bodies’) and stress granules [8–
10] These RNA granules are key structures in the
reg-ulation of gene expression at the post-transcriptional
level [11–13] Recently, we have identified the protein
TcDHH1, a putative DEAD box RNA helicase, in
T cruzi, homologous to Dhh1 (yeast) and Rck⁄ p54
(mammals) proteins [14] It has been reported that
Dhh1 is involved in multiple RNA-related processes in
various eukaryotes, and accumulates in stress granules
and P-bodies of yeast, animal cells and T brucei
(reviewed in [15–21]) Trypanosoma cruzi DHH1 is
pres-ent in polysome-independpres-ent complexes and is located
diffusely in the cytoplasm and in cytoplasmic granules,
which vary in number when the parasite is subjected to
nutritional stress or conditions interfering with the
translation process, e.g treatment with cycloheximide
or puromycin [14]
In this article, we show that TcDHH1-containing
foci are present throughout the T cruzi life cycle;
how-ever, they are less well defined in the infective metacy-clic trypomastigote form We also show that proteins directly or indirectly associated with TcDHH1 include those of diverse function, e.g translation-related fac-tors, cytoskeleton proteins, heat shock proteins and metabolic proteins We use a ribonomic approach to identify the mRNA content of the TcDHH1-contain-ing complexes Microarray analyses reveal that the mRNAs associated with TcDHH1 in the epimastigote stage are some of those mainly expressed in the other forms of the T cruzi life cycle These data suggest that mRNPs containing TcDHH1 are involved in mRNA metabolism, regulating the expression of, at least, epimastigote-specific genes
Results
TcDHH1-containing granules are present in all developmental forms of T cruzi
We have shown previously that, in epimastigotes, TcDHH1 proteins are localized in discrete cytoplasmic foci during the logarithmic growth phase and increase
in number under nutritional stress [14] In this study,
we confirmed our previous data and also showed that TcDHH1 is localized to cytoplasmic foci in adherent parasites differentiating into metacyclic trypomastig-otes and amastigtrypomastig-otes TcDHH1-containing granules were not readily observed in trypomastigotes This is evident in microscopic fields in which terminally differ-entiated metacyclic trypomastigotes and intermediate forms are present (Fig 1D) However, when the expo-sure time for the photograph was at least doubled, a few TcDHH1-containing granules were observed in metacyclic trypomastigotes (data not shown) To a les-ser extent, TcDHH1 was also obles-served diffusely in the cytoplasm in all forms analyzed (Fig 1) At least 10 random fields of each developmental form were ana-lyzed, and more than 95% of the parasites had the appearance described above
Protein composition of TcDHH1-containing complexes
Cytoplasmic complexes containing the protein TcDHH1 in logarithmic growth phase epimastigotes were immunoprecipitated with specific antiserum As seen by western blot analysis, the antiserum recognizes
a unique band of the expected mass size in a cyto-plasmic extract of T cruzi (Fig 2) Pre-immune serum was used as an experimental control Three different
Trang 3immunoprecipitation experiments were performed.
Proteins associated with immunoprecipitated (IP)
com-plexes and proteins not associated with TcDHH1 from
the supernatant (SP) were resolved by SDS⁄ PAGE and visualized by silver staining (Fig 3A) The IP and SP samples obtained with pre-immune serum were pro-cessed in the same way and are shown in Fig 3C To provide a measure of the efficiency of TcDHH1 immu-noprecipitation, IP and SP samples were loaded onto
an SDS⁄ PAGE gel, transferred to a nitrocellulose membrane, and immunoblotted with the TcDHH1 antibody (Fig 3B) The results showed that different protein profiles were obtained for the IP and SP silver-stained gels (Fig 3A, C) Western blot analysis using the TcDHH1 antiserum allowed the identification of the band corresponding to TcDHH1 (Fig 3B), which was not detected in the IP fraction obtained with the pre-immune serum (Fig 3D) The fact that TcDHH1 could also be detected in the SP fraction is probably a
A
B
Fig 2 Western blot analysis of T cruzi protein extracts using anti-TcDHH1 (A) Western blot analysis of protein extracts from epimastigotes (Epi), epimastigotes under nutritional stress (Stress), differentiating epimastigotes (cells adherent after 24 h) (Ad24h), metacyclic trypomastigotes (Meta) and amastigotes (Ama), probed with antiserum against TcDHH1 protein (1 : 100 dilution) The extracts were standardized and all lanes were loaded with 10 lg of protein (B) SDS ⁄ PAGE stained with Coomassie brilliant blue to control protein load TcDHH1 has an expected molecular mass of 46.7 kDa The molecular mass marker (in kDa) is the Benchmark Protein Ladder (Gibco, Grand Island, NY, USA).
A
B
C
D
E
Fig 1 Localization of TcDHH1 during the life cycle of T cruzi (A) Epimastigotes in logarithmic growth phase (B) Epimastigotes under nutritional stress (C) Differentiating epimastigotes (24 h adherent cells) (D) Metacyclic trypomastigotes (E) Amastigotes Cells were incubated with antiserum against TcDHH1 and the immune com-plexes were reacted with Alexa-labeled goat anti-mouse antibodies Kinetoplasts and nuclei were stained with propidium iodide Open arrowheads indicate an epimastigote form in the metacyclic trypo-mastigote population Filled arrowheads point to a trypotrypo-mastigote among the amastigotes Bar, 10 lm.
Trang 4result of the abundance of TcDHH1, as inferred from
recent data from T brucei [22]
IP proteins were digested with trypsin and
subjected to analysis by two-dimensional nano
LC-MS⁄ MS The mass spectra obtained were compared
with spectra deduced from protein sequences in the
T cruzi data bank We used the following criteria to
reliably select proteins from the database for further
analysis: (a) proteins with probability values equal
to or < 0.001; (b) proteins present in at least two
experiments using TcDHH1 antiserum; and (c)
pro-teins that did not appear in any of the three control
samples Proteins that fulfilled the above criteria are
listed in Table 1 (for a complete list with all proteins
reliably identified in the immunoprecipitation assays, but not meeting the selection criteria, see Table S1) These complexes contain various proteins in addition
to TcDHH1, including heat shock proteins, mRNA-binding proteins, initiation and elongation translation factors, ribosomal proteins and metabolic proteins Abundant proteins, such as heat shock proteins and elongation factor 1a, were present in the TcDHH1 immunoprecipitates, whereas other highly abundant proteins, such as the cysteine proteinase cruzipain, epimastigote-specific mucins and paraflagellar rod proteins, were not detected in any of the experi-ments
TcDHH1 granules contain poly(A)-binding proteins (PABPs)
As described previously, PABP1 is a core component
of stress granules in mammals [23] Two PABPs with a high similarity to other eukaryotic PABPs have been identified previously in T cruzi, TcPABP1 and TcPABP2, with molecular masses of 63.8 and 61.4 kDa, respectively [24,25] To test for the presence
of these proteins in TcDHH1 granules, epimastigotes
in the exponentially growing phase or under nutri-tional stress were used for colocalization analyses PABP2 antiserum showed a punctate distribution in the cytoplasm of epimastigotes and stressed epimastig-otes Most granules in these cells seemed to colocalize with granules containing TcDHH1; those that did not colocalize were grouped in defined regions of the cyto-plasm lying next to each other (Fig 4) Colocalization assays with PABP1 antiserum showed a slightly differ-ent staining pattern, with diffuse fluorescence and some more intense fluorescent foci throughout the cytoplasm These foci partially colocalized with TcDHH1 in both epimastigotes and epimastigotes under nutritional stress (Fig 4) In parallel with the colocalization experiments, we carried out immunopre-cipitation assays using mouse TcDHH1 antiserum, and the presence of TcPABP1 and TcPABP2 proteins in the precipitated complex was determined using western blot analysis with rabbit LmPABP1 and LmPABP2 antisera (Fig 5) Both antibodies specifically recog-nized a single band with the expected molecular mass
in western blots, confirming that TcPABPs are part of the TcDHH1 protein complex
mRNAs present in TcDHH1-containing complexes are regulated in a stage-specific manner
We used a ribonomic approach to identify the mRNAs associated with TcDHH1, and thus to infer the role
Fig 3 Analysis of TcDHH1 immunoprecipitation (A) Epimastigote
proteins immunoprecipitated with TcDHH1 antibody (IP) and
proteins from the supernatant fraction (SP) were resolved by
SDS ⁄ PAGE and visualized by silver staining (B) Western blot
analy-sis of IP and SP samples probed with antiserum against TcDHH1
protein (1 : 100 dilution), showing the efficiency of TcDHH1
immu-noprecipitation (C) Epimastigote proteins immunoprecipitated with
pre-immune serum (IP) and proteins from the supernatant fraction
(SP) were resolved by SDS ⁄ PAGE and visualized by silver staining.
(D) Western blot analysis of IP and SP samples, obtained with
pre-immune serum, and probed with serum against TcDHH1
pro-tein (1 : 100 dilution) It should be noted that tracks in (C) and (D)
have a higher IgG background when compared with those in (A)
and (B) as pre-immune serum was used at a lower dilution in order
to discard the spurious recognition of TcDHH1 The molecular mass
marker (in kDa) is the Benchmark Protein Ladder (Gibco).
Trang 5played by this protein in the regulation of gene
expres-sion in T cruzi The mRNAs associated with IP
com-plexes from epimastigotes were compared with the
mRNAs not associated with TcDHH1 from SP IP
and SP samples were obtained from six experiments,
and were compared using competitive hybridization
assays in oligo-DNA microarrays The amount of
RNA extracted from IP samples was generally smaller
than that from SP samples This was expected, as the
mRNA population associated directly or indirectly
with TcDHH1 is likely to represent only a fraction of
the total RNA
We identified mRNAs that were present in IP and
SP samples in different amounts, based on a two-fold
difference in mRNA levels and a 5% false discovery
rate For epimastigote forms in the logarithmic growth
phase, 203 distinct mRNAs displayed higher levels in
the IP than SP fraction, with 265 mRNAs present at
lower levels in the SP sample Most of the mRNAs
associated with TcDHH1 were from mucin-associated
surface protein (MASP) and mucin protein families,
with others corresponding to several hypothetical and
hypothetical conserved proteins We also observed
mRNAs corresponding to metabolic proteins,
mRNA-binding proteins, amastin and cyclin, among others
(Fig 6A, Table 2) (for a detailed list of mRNAs, see
Table S2) A semi-quantitative approach, RT-PCR
and densitometry analysis of gel bands for five
mRNAs with different distribution patterns between
IP and SP samples confirmed these results (Fig 6B)
Surface proteins (MASP and mucins) are encoded by
multigene families Although these gene families were
the largest in the T cruzi genome, we did not identify
any of the corresponding mRNAs in the SP sample, demonstrating the specific presence of these mRNAs in the IP sample Moreover, mRNAs from other large gene families in T cruzi, e.g those encoding trans-siali-dase and cysteine protease, were not found in the IP sample
Discussion
Recently, we have identified a putative RNA helicase (TcDHH1), which is similar to its eukaryotic orthologs [14] Members of this protein family are involved in several aspects of mRNA metabolism and localize to distinct foci in the cytoplasm [26] We compared the replicative, nutritionally stressed, differentiating and infective forms of T cruzi, and showed that these
T cruzi life cycle stages display distinct granular pat-terns of cytoplasmic TcDHH1-containing foci (Fig 1) These granules were not as clearly visible in trypom-astigotes as in the other forms, but TcDHH1 protein levels remained similar for all T cruzi developmental stages [14] Thus, the variation in the number of TcDHH1-containing granules does not seem to be related to changes in gene expression levels, but is probably related to diffuse or foci-like distributions In
a recent study, Cassola et al [20] showed that epim-astigotes in culture and parasites at different time points during in vitro differentiation did not display mRNA granules Although, at first, these findings seem to disagree with our work, it is not possible to compare these studies because different experimental approaches were used to investigate the presence of mRNA granules in developmental forms of T cruzi
Table 1 Protein composition of TcDHH1-containing complexes IP, complexes immunoprecipitated with anti-TcDHH1 antibody; C, complexes immunoprecipitated with pre-immune serum; 1, 2, 3, biological replicates; •, indicates the presence of a protein in that fraction.
Tc00.1047053507641.280 Cnp60 chaperonin HSP60, mitochondrial precursor; groELprotein; heats • •
Tc00.1047053508665.14 ATP-dependent Clp protease subunit, heat shock protein 100,putative • •
Trang 6In yeast, the Dhh1 protein interacts with decapping
and deadenylation complexes, stimulating mRNA
decapping [15–18] Dhh1p homologs from
Caenorhabd-itis elegans (Cgh1), Drosophila (Me31b) and Xenopus
(Xp54) are involved in the storage of translationally
repressed maternal mRNAs [16,17] In addition, the
RNA helicase DOZI, involved in the storage and
silencing of certain mRNA species, has been identified
recently in Plasmodium berghei and is localized to
cyto-plasmic granules in female gametocytes [19]
Dhh1⁄ Rckp54 is common to both P-bodies and stress
granules [10–13]; both of these types of granule may
therefore exist in T cruzi Therefore, we studied
mRNPs containing TcDHH1 and investigated their
potential similarity with mRNPs in other organisms
We have shown previously that TcDHH1 is present in cytoplasmic complexes containing mRNA and pro-teins TcDHH1-containing complexes have been puri-fied previously from polysome and polysome-free fractions [27] TcDHH1 must therefore be, at least partly, associated with mRNAs that are independent
of the translation machinery
Our analysis of IP complexes showed that TcDHH1 interacts with proteins that are described as stress granule components (SGs) We identified TcDHH1, as expected, and proteins previously identified in stress granules, such as heat shock proteins and 40S ribo-somal subunit proteins Translation initiation factors (eukaryotic initiation factors 3 and 4, eIF3 and eIF4), also typical of SGs, were observed in one of the three
A
B
Fig 4 Colocalization assays of PABP proteins with TcDHH1 (A) Logarithmic growth phase epimastigotes (B) Epimastig-otes under nutritional stress Antibodies against PABP1 and PABP2 were produced
in rabbit and tested at 1 : 100 dilutions Antibodies against TcDHH1 protein were produced in mouse and used at 1 : 100 dilution Immune complexes reacted with Alexa-labeled 546 goat anti-rabbit and Alexa-labeled 488 goat anti-mouse antibod-ies (1 : 400) Bar, 10 lm.
Trang 7biological replicates (Table S1) Although abundant
proteins, such as heat shock proteins and eIF1a, are
present in these complexes, other proteins highly
abun-dant in epimastigotes, such as the cysteine proteinase
cruzipain, the family of GP63 metalloproteases, many
ribosomal proteins, epimastigote-specific mucins and
the epimastigote-specific metabolic protein histidine
ammonia-lyase, were not detected in any of the
experi-ments [28–30] Nonetheless, it remains to be elucidated
whether these proteins indeed interact with TcDHH1
or are fraction contaminants Interestingly, Pare et al
[31] have revealed recently that a heat shock protein
(Hsp90) is important for recruiting the argonaute
protein (hAgo2) to SGs and for efficient biogenesis and⁄
or stability of P-bodies in mammals Thus, it seems probable that the heat shock proteins identified in our study might interact with TcDHH1 We also identified proteins that are not characteristic of these structures, such as translation elongation factors, metabolic proteins and actin (Table 1) Although actin is not a
Fig 6 mRNAs present in TcDHH1-containing complexes in epimastigotes (A) Pie chart diagram displaying the percentage representation of the most abundant mRNAs present in TcDhh1-containing complexes (B) RT-PCR analysis of five mRNAs with different distribution patterns between IP and SP fractions Putative glucose-regulated protein 78, more represented in the SP fraction, showed an IP⁄ SP ratio of 0.51 Putative (H+)-ATPase G subunit, equally represented in both fractions, resulted in an IP ⁄ SP ratio of 0.98 Putative cyclin, putative mucin TcMUCII and hypothetical protein were more represented in the IP fraction, with IP ⁄ SP ratios of 2.0, 6.7 and 1.8, respectively.
Fig 5 Co-immunoprecipitation assays of TcDHH1 with PABP
pro-teins Epimastigote proteins were immunoprecipitated with mouse
TcDHH1 antibody (I) or pre-immune serum (PI), resolved by
SDS ⁄ PAGE, electrotransferred onto Hybond-C membranes and
probed with rabbit LmPABP1 (A) and LmPABP2 (B)
anti-sera (1 : 100 dilution), showing that PABPs are part of the TcDHH1
protein complex The molecular mass marker (in kDa) is the
Bench-mark Protein Ladder (Gibco).
Table 2 mRNAs present in TcDHH1-containing complexes in epimastigotes.
Mucin-associated surface protein (MASP), putative 61 Mucin-associated surface protein (MASP,
pseudogene), putative
4
ADP-ribosylation factor family, putative 1
Expression site-associated gene (ESAG-like) protein, putative
1 Meiotic recombination protein SPO11, putative 1 Mitochondrial carrier protein, putative 1
Phosphatidic acid phosphatase, putative 1
Ras-related GTP-binding protein, putative 1
Serine ⁄ threonine protein phosphatase, putative 1 Serine ⁄ threonine protein phosphatase 2A,
catalytic subunit, putative
1 Serine-, alanine- and proline-rich protein, putative 1
Trang 8typical SG, there seems to be a close interaction
between cytoplasmic mRNA granules and the
cytoskel-eton Indeed, granules remaining motionless are
associ-ated with actin filaments, whereas those that move in
the cytoplasm remain associated with microtubules
[32] Thus, it is possible that actin is indeed a
compo-nent of TcDHH1-containing complexes and not a
putative contaminant The fact that other abundant
proteins were not detected in this fraction would be
consistent with this Interestingly, we did not identify
P-body core proteins, such as LSMs and exonuclease
5¢–3¢ XRN1 These findings suggest that
TcDHH1-containing complexes are more likely to be
compo-nents of SGs than of P-bodies We cannot rule out the
possibility that the identified proteins of
TcDHH1-con-taining complexes also correspond to the cytoplasmic
granule-free fraction of TcDHH1
We tested for the colocalization of TcDHH1 with
PABP present in stress granules to extend our findings
from the immunoprecipitation experiments and to gain
further insight into the function of
contain-ing granules Our results suggest that
TcDHH1-containing granules contain PABPs Indeed, PABP2
seemed to colocalize with most granules containing
TcDHH1 in both unstressed epimastigotes and
epi-mastigotes subjected to nutritional stress Granules
appearing to contain only PABPs remained adjacent
to these TcDHH1-containing granules (Fig 4)
Our findings are in agreement with the data
pub-lished by Cassola et al [20], who demonstrated that, in
starved parasites, TcPABP1 and TcPABP2 showed
strong accumulation in mRNA granules However, in
contrast with our study, these authors showed that
TcPABP1 and TcPABP2 were not recruited to mRNA
granules in parasites not subjected to starvation One
possible explanation for this difference is the fact that
these authors used parasites overexpressing green
fluo-rescent protein fusions, in contrast with the native
pro-teins evaluated here PABP1 is used as a marker for
stress granules and is absent from P-bodies in
mam-mals However, Brengues and Parker [33] showed that
PABP1 is present in the P-bodies of Saccharomyces
cerevisiae, and that mRNPs containing poly(A)+
mRNA, PABP1, eIF4E and eIF4G enter these
struc-tures, possibly representing a transitional state during
mRNA exchange between P-bodies and the translation
machinery These authors also demonstrated that
PABP1 may be present in P-bodies even in the absence
of stress Another study demonstrated that granules
containing PABP1 in S cerevisiae were distinct from
the P-bodies formed specifically under stress caused by
glucose deprivation, and that these two types of
gran-ule partially colocalize These grangran-ules may function as
mRNA storage compartments, called EGP-bodies, allowing mRNA translation to resume when cell growth conditions are restored, and are analogous to the stress granules observed in mammals [34]
The characterization of mRNAs from mRNP com-plexes immunoprecipitated with TcDHH1 antiserum revealed the presence of several mRNA species, with overrepresentation of those encoding MASP, mucins, hypothetical proteins and amastin In general, mRNAs associated with TcDHH1 are not translated into pro-teins in epimastigotes, but are predominantly trans-lated during other stages of the parasitic life cycle For example, TcMUCII proteins are specific to, and MASP proteins are mostly produced in, the trypomas-tigote forms [35,36] The mRNA encoding amastin, a protein predominantly produced in amastigotes [37], was also present in the TcDHH1-containing complexes from epimastigotes These findings suggest that epi-mastigote TcDHH1-associated mRNAs are either stored for later use or are present in the initial steps to degradation, given that their poly(A) tails are mostly intact Accordingly, recent work using an ATPase-defi-cient dhh1 mutant provided evidence that a pathway including Dhh1 has a selective role in the destabiliza-tion of many regulated mRNAs in procyclic forms of
T brucei[22]
It should be noted that the ribonomic approach used
in this study favors the identification of polyadenylated mRNAs present in IP complexes; deadenylated mRNAs destined for, or in the initial stages of, degradation would not be detected by these microarray analyses
Materials and methods
Parasites
The T cruzi clone Dm28c [38] was maintained at 28C in liver infusion tryptose medium supplemented with 10% heat-inactivated fetal bovine serum Epimastigotes under nutritional stress, metacyclic trypomastigotes and amastig-otes were obtained in vitro as described previously [39,40]
Immunofluorescence and imaging
Immunofluorescence assays were carried out using a proto-col described previously [14], which was modified slightly to ensure that parasites were resuspended and washed in NaCl⁄ Pifor only 5 min before being fixed
Mouse polyclonal anti-TcDHH1 antibody was produced
as described previously [14]; the antiserum was affinity puri-fied and stored in aliquots at )20 C prior to use Rabbit polyclonal antibodies against PABP1 and PABP2 were kindly provided by Dr Osvaldo P de Melo Neto (Centro
Trang 9de Pesquisas Aggeu Magalha˜es, Fiocruz, Recife, Brazil).
Serum dilutions of antibodies were as follows: rabbit
LmPABP1 and LmPABP2, 1 : 100; mouse
anti-TcDHH1, 1 : 100 Alexa Fluor-488 and Alexa Fluor-546
were used as conjugated secondary antibodies (1 : 400;
Molecular Probes, Invitrogen, Eugene, OR, USA) Images
showing subcellular localization were acquired using a
Nikon Eclipse E600 microscope (Nikon Corporation,
Tokyo, Japan) coupled to a Cool SNAP-PRO color camera
(Media Cybernetics, Bethesda, MD, USA) Merged images
were obtained by superimposing image files with image-pro
plussoftware (Media Cybernetics)
DHH1 immunoprecipitation assays: protein
content
Immunoprecipitation assays with the TcDHH1
anti-body were carried out in cytoplasmic extracts from
loga-rithmic growth phase epimastigotes Mouse anti-TcDHH1
(50 lL) was incubated with 150 lL of resin containing
stirring Pre-immune serum was incubated with resin under
the same conditions and used as a control for
immunopre-cipitation reaction specificity After incubation, the resin
was collected by centrifugation at 600 g for 2 min, the
supernatant was discarded and the resin was incubated with
5% nonfat milk in NaCl⁄ Pifor 30 min The resin was then
washed twice with NaCl⁄ Pi
To obtain T cruzi cytoplasmic extracts, 2· 109parasites
were washed with NaCl⁄ Piand incubated in 2 mL of IMP1
200 UÆmL)1; Nonidet P40, 0.5%) for 2 h on ice, with
mod-erate agitation Parasite lysis was monitored with a light
microscope Cytoplasmic extracts were obtained by
centri-fugation at 7000 g for 20 min at 4C; 1 mL of this extract,
corresponding to the lysis of 1· 109
parasites, was incu-bated with resin previously coupled to TcDHH1
anti-body or to the pre-immune serum, as described above, for
16 h at 4C, with moderate agitation
The IP complexes were collected by centrifugation at
600 g for 2 min and SPs were saved The resin was washed
three times with IMP2 buffer (KCl, 100 mm; MgCl2, 5 mm;
Hepes, 10 mm, pH 7.0; protease inhibitor, 1 : 100; RNase
same centrifugation step Proteins linked to the resin were
eluted with 150 lL of glycine (0.1 m, pH 2.0) and the pH
was adjusted to 7.5–8.0 Next, 150 lL of buffer (urea, 7 m;
thiourea, 2 m; Chaps, 2%; Triton, 2%; dithiothreitol, 1%;
nuclease, 1 : 100; protease inhibitor, 1 : 100) was added to
the sample and proteins were analyzed by two-dimensional
puri-fied before analysis using a two-dimensional clean-up kit,
following the manufacturer’s instructions (GE Healthcare,
Buckinghamshire, UK) Proteins were reduced with
dith-iothreitol, alkylated with iodoacetamide and digested over-night with trypsin The resulting tryptic peptides were desalted on C8 cartridges (Michrom BioResources, Auburn,
LC⁄ MS ⁄ MS analyses on a Michrom BioResources Paradigm MS4 Multi-Dimensional Separations Module, a Michrom NanoTrap Platform and an LCQ Deca XP plus ion trap mass spectrometer The mass spectrometer was used in data-depen-dent mode, and the four most abundant ions in each mass spectrum were selected and fragmented to produce tandem mass spectra The MS⁄ MS spectra were recorded in the pro-file mode Proteins were identified by comparing the MS⁄ MS spectra obtained with our T cruzi database and its reversed complement using Bioworks v3.2 Peptide and protein hits were scored and ranked using the probability-based scoring algorithm incorporated in Bioworks v3.2 and adjusted to a false positive rate of less than 1% Only peptides showing fully tryptic termini, with cross-correlation scores (Xcorr) greater than 1.9 for single-charged peptides, 2.3 for double-charged peptides and 3.75 for triple-double-charged peptides, were used for peptide identification In addition, delta correlation scores (DCn) were set to be > 0.1 and, for increased strin-gency, proteins were accepted only if their probability score was less than 0.001 The following search parameters were used: taxonomy, eukaryota; monoisotopic mass tolerance, 0.1 Da; partial methionine oxidation; and one missed tryptic cleavage allowed Criteria for positive protein identification included Mascot scores and sequence coverage
Parallel to MS, IP and SP samples were analyzed by
following protocol: incubation in fixing solution (50% ethanol, 12% acetic acid, 0.02% formaldehyde) for 1 h, three washes in 50% ethanol for 15 min and sensibilization
in 0.02% sodium thiosulfate for 1 min, followed by an extensive wash in distilled water Staining was performed
by incubating the gel for 30 min in silver nitrate solution (0.2% silver nitrate, 0.02% formaldehyde), followed by washing three times in distilled water for 1 min The gel was developed in 3% sodium carbonate and 0.05% formal-dehyde Staining was stopped in 50% ethanol and 12% ace-tic acid for 5 min For western blot analysis, IP and SP
transferred to a nitrocellulose membrane Nonspecific bind-ing sites were blocked by incubatbind-ing the membrane with
for 30 min For analysis of the efficiency of TcDHH1 immunoprecipitation, membranes were incubated for 1 h with anti-TcDHH1 antibody (1 : 100 dilution) or with pre-immune serum For co-immunoprecipitation analysis of TcDHH1 and TcPABPs, membranes were incubated for
1 h with rabbit anti-LmPABP1 and anti-LmPABP2 (1 : 100 dilution) The membranes were then extensively washed in NaCl⁄ Pi and incubated with goat phosphatase-conjugated anti-rabbit IgG (Sigma) diluted 1 : 10 000 The color reaction was developed with 5-bromo-4-chloro-3-indolyl
Trang 10phosphate and nitroblue tetrazolium (Promega, Fitchburg,
WI, USA)
DHH1 immunoprecipitation assays: ribonomics
To determine which mRNAs were associated with the
with resin containing protein A Sepharose (Sigma) for
mixed with 150 lL of resin and 1 lL of RNAse OUT
(Invitrogen) After incubation, the resin was collected by
centrifugation and processed as described above
Cyto-plasmic extract, corresponding to 2· 109
cells, was
antibodies for 2 h in an ice bath, with agitation The
resin was then collected by centrifugation and SP was
retained for a control Resin was washed three times with
IMP2 buffer The RNAs from the SP or resin (IP) were
using the ‘Animal Cells I’ protocol in the manufacturer’s
manual, with the additional step of DNase treatment in
a column Linearly amplified RNA was generated from
1 lg of total RNA (single round) using a MessageAmp
amplified RNA kit (Ambion, Austin, TX, USA),
follow-ing the manufacturer’s instructions cDNA was
synthe-sized from 1 lg of total or affinity-purified RNA using
an oligo(dT) primer (US Biochemical Corporation,
Cleve-land, OH, USA) and reverse transcriptase (IMPROM II;
Promega), as recommended
Oligonucleotide DNA microarrays
The microarray was constructed with 70-mer
oligonucleo-tides As a result of the hybrid and repetitive nature of the
sequenced T cruzi strain (CL Brener), all coding regions
(CDS) identified in the genome (version 3) were retrieved
and clustered using the blastclust program, with 40%
coverage and 75% identity For the probe design,
content; 10 359 probes were designed to the longest T cruzi
CDS of each cluster; 393 probes correspond to genes of an
external group (Cryptosporidium hominis) and 64 spots
con-tain only spotting solution (NaCl⁄ Cit, 3 ·), totaling 10 816
spots These oligonucleotides were spotted from a 50 lm
solution onto poly-l-lysine-coated slides and cross-linked
with 600 mJ UV Each probe corresponding to T cruzi
genes was identified following the T cruzi Genome
Consor-tium annotation (http://www.genedb.org) The microarray
slides were produced at Virginia Commonwealth
Univer-sity, Richmond, VA, USA
Microarray hybridization and analysis
Fluorescent cyanin (Cy) dyes, Cy3 or Cy5 as appropriate,
were incorporated into second-strand cDNA synthesis using
2 lg of amplified RNA as the starting material for each sample Labeled cDNA was purified with a Microcom 30
hybridizations and washes were carried out in a GeneTac
Chelmsford, MA, USA) The Cy3- and Cy5-labeled cDNAs were mixed and added to 120 lL of hybridization solution and allowed to hybridize for 14–16 h at 42C The micro-array slides were then washed in buffer of increasing
centrifugation at 280 g for 5 min The dried slides were scanned in a 428 Array Scanner (Affymetrix, Santa Clara,
CA, USA) The images were analyzed with spot software The resulting data were corrected for background and nor-malized, using the normexp and PrintTip-Loess methods, respectively, within the Limma package [41]
A total of six individual IP and SP pairs was hybridized
in a semi-balanced dye design; overrepresented genes from both fractions were selected using sam software [42] Genes were thus selected on the basis of at least a two-fold differ-ence in mRNA levels and a 5% false discovery rate Micro-array data were submitted to ArrayExpress accession number E-MEXP-2448
RT-PCR
cDNA was synthesized from 1 lg of total RNA using 1 lL
of 10 lm random primers (USB Corporation, Cleveland,
OH, USA) and 1 lL of reverse transcriptase (IMPROM II; Promega), according to the manufacturer’s instructions PCR was carried out with 20 ng of cDNA as template,
20 mm Tris-HCl (pH 8.4), 10 pmol of primers, 2.5 mm
(Invitrogen) The oligonucleotide primer sets used for PCR were as follows: putative cyclin (Tc00.1047053506945.270),
F, 5¢-TGGGGAGGATTATAGCGATG-3¢; R, 5¢-ACTTC
(Tc00.1047053506131.20), F, 5¢-GCGGAGAACAAGATG
hypothetical protein (Tc00.1047053509891.40), F, 5¢-GCCG TCATGCAAAAATATCC-3¢; R, 5¢-CCTTTTCAGCCAA
(Tc00.1047053506585.40), F, 5¢-TGGCGGTAAGAAGAA
putative (H+)-ATPase G subunit (Tc00.1047053510993 10), F, 5¢-ACAACGTGCAAAGGCTTCTT-3¢; R, 5¢-CTC GTGCCAACTCCAAGTTT-3¢
PCR, using a Bio-Cycler II thermocycler (Peltier Thermal Cycler; Bio-Rad, Hercules, CA, USA), included heating at
94C for 2 min, followed by 25 cycles of 94 C for 15 s,
58C for 30 s and 72 C for 30 s, with a final extension of
72C for 3 min Ten microliters of RT-PCR products were resolved by 2% agarose gel electrophoresis, visualized by ethidium bromide staining Gel photographs were taken using a UVP Bioimaging System (UVP, Upland, CA,