 |
INTRODUCTION |
The signal recognition particle
(SRP)1 is an evolutionarily
conserved ribonucleoprotein (RNP) that exists in all three kingdoms (1-3). It couples the synthesis of membrane and secretory proteins to
the translocation across the eukaryotic endoplasmic reticulum (ER)
membrane, as well as the bacterial plasma membrane and chloroplast thylakoid membranes (4).
In vitro studies have shown that SRP binds to the
hydrophobic signal peptide as the N-terminal end of the nascent
preprotein emerges from the ribosome, which results in a slowing or
pause of translation, a phenomenon termed "elongation arrest" (5). SRP then targets the nascent chain-ribosome complex to the ER membrane
by interacting with a membrane-bound SRP receptor. After this docking
event that is GTP-dependent, the translocation of the
polypeptide takes place co-translationally through the protein pore,
the translocon (6, 7). The well defined canine SRP complex is composed
of a single RNA, the 7SL RNA, and six proteins. The most studied
protein of the SRP is SRP54, which binds to the signal peptide. SRP54
also binds to the SRP RNA and is a GTPase (8). SRP54 is a highly
conserved protein in nature. Much of what we know about this protein
comes from structural studies performed on the thermopile bacteria
Thermus aquaticus (9) and the Escherichia coli
54-homologue (Ffh) protein (10). The Ffh is composed of three domains:
N-terminal domain (N domain), a Ras-like GTPase domain (G domain), and
a methionine-rich domain (M domain). The N domain is a four-helix
bundle that is closely associated with the G domain. The G domain is a
GTPase that contains a motif unique to the SRP family of GTPases (11,
12). The N and G domains of Ffh mediate the interaction of the SRP with homologous N and G domains of FtsY, the SRP receptor.
The in vivo role of SRP in protein translocation was
elucidated in bacteria and yeast. Disruption of the SRP pathway in
E. coli is lethal (13), most probably because of the defects
in the translocation of inner membrane proteins (14-16). However, the
secretion of signal peptide-containing proteins such as periplasmic or
outer membrane proteins was only slightly affected (13, 14, 16, 17).
Surprisingly, in yeast, different phenotypes were observed when the SRP
pathway was disrupted. In Saccharomyces cerevisiae SRP is
not essential for cell growth, although cells in which SRP was
genetically disrupted grew poorly (18). Moreover, the translocation of
numerous secretory and membrane proteins into the lumen of the ER is
impaired. However, different proteins show translocation defects of
varying severity (18-20). In Schizosaccharomyces pombe, the
SRP pathway is essential, because Srp54p depletion is lethal (21). In
another yeast strain, Yarrowia lipolytica, disruption of the
two 7SL RNA-encoding genes is lethal (22), but SRP54 is important but
not essential for growth. The viability of SRP54-depleted cells
suggests that the SRP-independent pathway functions well enough for
survival (23).
Very little is known about the trypanosomatid SRP pathway. Trypanosomes
diverged very early in the eukaryotic lineage (24); this provides an
interesting system to study the evolution of RNA-mediated processes,
because many exotic and unique processes were first described in
trypanosomes including trans-splicing (25) and RNA editing
(26). The analysis of trypanosomatid 7SL RNA revealed that the RNA can
be folded into the canonical structure of eukaryotic 7SL RNA, except
the Alu domain, which is truncated, missing one of the loops, and most
importantly lacks the potential to form a pseudoknot (27). Most
intriguing is the finding that the trypanosomatid SRP complex is
composed of two RNA molecules: the 7SL RNA and a tRNA-like molecule,
sRNA-76 in Trypanosoma brucei and sRNA-85 in
Leptomonas collosoma
(28).2 Although we do not yet
know the function of this unique tRNA-like molecule, we hypothesize
that because it is a tRNA-like molecule, it may assist the
trypanosomatid 7SL RNA truncated Alu domain in the translational arrest
function of the SRP. Another unusual feature that was discovered for
the L. collosoma 7SL RNA is that the 7SL RNA is found in two
stable conformations: one enriched in the ribosome fraction (7SL I) and
the other, found in free SRP (7SL II). The conversion of 7SL II to 7SL
I is associated with a novel RNA editing event of C to U in domain III
of the RNA (29).
Recently, RNA interference (RNAi) has become a very efficient means for
deciphering the function of trypanosome gene. RNAi is based on the
degradation of mRNA by the production of specific dsRNA (30). The
dsRNA is cleaved to small interfering RNAs, which are 21-23 nt
long that elicit the degradation of the specific mRNA (30). dsRNA
is produced in vivo either from opposing T7 promoters, or in
the form of a stem-loop structure that is transcribed from the EP
promoter (31, 32). In both cases, the production of the dsRNA is
regulated by the tetracycline repressor. RNAi was first reported in
trypanosomes for the silencing of tubulin, which resulted in the "fat
cells" phenotype (31, 33). Subsequently, this methodology was used to
silence different genes (32). Detailed studies were performed on the
T. brucei exosome, mitochondrial topoisomerase II, flagellum
ontogeny, and RNA editing (34-38)
Here, we used RNAi to examine the role of the SRP pathway in protein
sorting in trypanosomes. SRP depletion by RNAi caused lethality,
suggesting that this pathway is essential in trypanosomes. Surprisingly, the signal peptide-containing proteins examined were
translocated to the ER, and properly processed, suggesting that an
alternative translocation pathway must exist in trypanosomes. The
secondary effect on protein sorting observed for procyclin EP,
lysosomal protein p67, and the flagellar pocket protein CRAM may stem
from defects in membrane integrity.
| |
MATERIALS AND METHODS |
Cell Growth and Transfection--
Procyclic T. brucei
strain 29-13 (obtained from Paul England's laboratory, a gift from
George Cross laboratory), which carries integrated genes for T7
polymerase and the tetracycline repressor (39), were grown in SDM-79
medium supplemented with 10% fetal calf serum in the presence of 50 µg/ml hygromycin and 15 µg/ml G418. Cell growth was monitored by
hemocytometer. To establish the cell lines with the dsRNA-expressing
construct, transformants were selected with 5 µg/ml phleomycin. For
cloning, individual cells were spread onto agarose plates as described
in Ref. 40, and the plates were incubated at 27 °C in a 5%
CO2 atmosphere. Individual colonies became visible after
5-7 days and were transferred to medium for propagation in microtiter
dishes and expanded in liquid medium. Cells from cultures that showed
the best characteristic phenotype upon tetracycline induction were
grown and frozen. Every 2 weeks a new culture was started from the
original frozen stock.
Oligonucleotides and Plasmids--
9/40,
5'-CGTCTAGAACGTGAAGGAGTTTGTAAAT-3', sense, from position 80 to 99 of the SRP54 coding region, including a
XbaI site (underlined), was used for PCR amplification of
the 535 bp for cloning into the pJM326 and pLew100 vectors,
respectively, to generate the stem-loop construct (32).
10/40, 5'-CCCAAGCTTTCTCTTCGAACAGTGCCGAT-3', antisense, complementary to positions from 594 to 613 of the
SRP54 coding region, including a HindIII site,
was used for PCR amplification of the 535 bp for cloning into the
pJM326 vector to generate the stem-loop construct (32).
3/59, 5'-CGACGCGTTCTCTTCGAACAGTGCCGAT-3', antisense, complementary to positions from 594 to 613 of the
SRP54 coding region, including a MluI, was used
for PCR amplification of 535 bp for cloning into the pLew100 vector to
generate the stem-loop construct (32, 39). 34/251,
5'-ATTCTGGAGGTGCACAACCT-3', sense, positions from 1228 to 1247 of the
SRP54 coding region, was used for cloning and sequencing.
35/251, 5'-TTCCCCATTTGTTACTTTTC-3', antisense, complementary
to positions from 164 to 183 relative to the stop codon was used for
cloning and sequencing.
Northern Analysis--
Total RNA was prepared with Trizol
reagent and 20 µg/lane were fractionated on a 1.2% agarose,
2.2 M formaldehyde gel. The RNA was visualized with
ethidium bromide. The SRP54 mRNA and tubulin mRNA were detected
with [
-32P]-random labeled probe (Random Primer DNA
Labeling Mix, Biological Industries, Co.). For analyzing small RNAs,
total RNA was fractionated on a 10% polyacrylamide gel containing 7 M urea. The RNA was transferred to a nylon membrane
(Hybond; Amersham Biosciences) and probed with
[
-32P]-end-labeled oligonucleotide, as was previously
described (41).
Assembly of T. brucei SRP54 Gene using Bioinformatics--
SRP54
proteins from other organisms were obtained from SwissProt. The
T. brucei data base (Sanger, GenBankTM, and
TIGR) was searched for SRP54 using TBLASNN (42). Ten overlapping fragments were assembled with the GELMERGE program of the Genetics Computer Group (GCG). To confirm the sequence, portions of the gene
were amplified by PCR and sequenced.
In Vivo Labeling and Immunoprecipitation--
Procyclic cells
were grown at 27 °C in SDM-79 medium. The cells were washed twice
with PBS and resuspended to 108 cells/ml in pre-warmed
(27 °C) Met/Cys minus Dulbecco's modified Eagle's medium
(Invitrogen). The cells were labeled with 200 µCi/ml L-[35S]Met/Cys (Hybond; Amersham Biosciences)
for 5 min, and then chased for 5 or 15 min by a 10-fold dilution in
prewarmed complete SDM-79 growth medium. Labeled cells
(107, 1 ml) were washed once with PBS, and solubilized in 1 ml of wash buffer (50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, 0.5%
deoxycholate, 0.1% SDS) containing protease inhibitor mixture (Sigma).
The extract was precleared by centrifugation. Antibodies (BiP, 5 µl/107 cells; p67, 3-5 µl/107 cells) were
bound to the Protein A-Sepharose beads (Santa Cruz Biotech) by rotating
for 1 h at room temperature, washed once with wash buffer, and
resuspended to the original volume in wash buffer. 50 µl of
antibody-protein A-Sepharose suspension was mixed with 0.5 ml of cell
lysate at 4 °C for 1 h. After the supernatant was removed, the
pellet was washed four times with wash buffer and once with TEN buffer
50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 5 mM EDTA. The bound proteins were separated on a 10%
SDS-polyacrylamide gel.
Immunofluoresence and Confocal Microscopy--
The cells were
washed with PBS, mounted on poly-L-lysine-coated slides,
and fixed in 4% formaldehyde, 0.04% glutaraldehyde in PBS at room
temperature for 30 min. Cells were blotted in 10% fetal calf serum in
PBS at room temperature for 30 min and then incubated with the primary
antibodies with 0.1% Nonidet P-40 for 1 h. For the EP surface
staining, the cells were fixed with 4% formaldehyde but not
permeabilized with Nonidet P-40. The BiP antibodies (anti-rabbit) (43)
and monoclonal p67 antibodies (44) were diluted 1:200 and 1:1000,
respectively. The CRAM (45) antibodies and the anti-EP monoclonal
antibodies (Cedarline) were diluted 1:300 and 1:500, respectively.
Antibodies from a rabbit chronically infected with trypanosomes (CI
serum, kindly provided by R. G. Nelson) were diluted 1:400. After
the cells were washed with PBS, they were reacted with fluorescein
isothiocyanate (Jackson ImmunoResearch). To stain the nucleus and
kinetoplast, the cells were incubated with propidium iodide (PI) (10 µg/ml) for 5 min. The cells were imaged by a Bio-Rad MRC 1024 upright
confocal microscope with a krypton-argon ion laser. The confocal
microscope image analysis was performed using the Laser Sharp 3.0 application.
Western Analysis--
The whole cell extract (106
cell equivalent per lane) was fractionated on a SDS-PAGE (10% gel),
transferred to NitroBind (Micron Separations, Inc.), and probed with
the antibodies. Anti-EP monoclonal antibodies (Cedarline), and
anti-CRAM antibodies (kindly provided by Mary Lee) were diluted 1:2000
and 1:1500, respectively. The anti-hnRNP D0 monoclonal antibodies
(kindly provided by Gideon Dreyfuss) were diluted 1:1000. The bound
antibodies were detected with goat anti-rabbit or goat anti-mouse IgG
bound to horseradish peroxidase and visualized by ECL (Amersham Biosciences).
| |
RESULTS |
Identification and Characterization of Trypanosomatid
SRP54--
We searched the T. brucei genome project for the
SRP54 homologue using the protein sequence data from human,
E. coli, and S. cerevisiae. The gene sequence was
derived from four different pieces that were clustered and the sequence
was confirmed by PCR and sequencing. The alignment of the T. brucei protein with its homologues is presented in Fig.
1. The BLAST search revealed a close
identity to mammalian, S. cerevisiae, and E. coli
proteins, 51.3, 44.8, and 30.8%, respectively. The protein
sequence was also compared with the SRP54 from Leishmania
major that we have recently cloned and sequenced
(GenBankTM accession number AY064402). As expected, the
protein is mostly related to the L. major protein (identity
of 77.5% and similarity of 83.3%). The protein carries the conserved
functional N, G, and M domains. Most striking are the three amino acid
changes (circled in Fig. 1). Interestingly, the changes are
trypanosomatid-specific and are not found in any other SRP54 protein
homologues that exist in the data base. Moreover, these changes are
found in the most conserved positions of the protein (position 82 in
the N domain, position 222 in the G domain, and position 335 in the M
domain). Surprisingly, the C terminus of the L. major
protein is longer by 32 amino acids compared with the T. brucei protein. In addition, the L. major C-terminal
domain carries a high glycine and methionine content compared with
other homologous domains, for instance, 15 glycines and 15 methionines
in L. major, compared with 9 glycines and 11 methionines in
Drosophila. The proximity of the C-terminal extension to the
hydrophobic groove of the M domain, and the high content of methionine
in this portion of the protein suggest that this domain perhaps also
participates in signal sequence binding. In fact, this hypothesis is
supported by data demonstrating that the deletion of this extension
from the mammalian SRP abolishes cross-linking to the signal sequence
in vitro, suggesting the involvement of the C-terminal
sequence in signal peptide recognition (46). It is reasonable to
speculate that the longer Leishmania C terminus, highly
enriched in methionines expands the hydrophobic surface area, which
functions in recognizing the trypanosomatid signal peptides. The
distinct amino acid changes in the three conserved positions found in
the two trypanosomatid species as well as the changes in the C terminus
of the protein may explain the failure of the mammalian translocation
machinery to recognize and properly translocate trypanosomatid signal
peptide-containing proteins (47). However, the shorter C terminus
in the T. brucei protein, compared with that of
L. major, is intriguing and may affect the ability of these
related proteins to function interchangeably.
View larger version (107K):
[in this window]
[in a new window]
|
Fig. 1.
Sequence alignment of T. brucei SRP54 with its homologues. The alignment
was performed using the ClustalW multiple sequence alignment program
and is illustrated using the program SeqVu 1.0.1. Identity is
represented by the gray background and the structural
domains are indicated. Secondary structure elements are shown
above the aligned sequences. The deviated residues are
circled. PGB, the putative guanine nucleotide
dissociation stimulator-binding site, and PRBM, the putative
RNA binding motif, are indicated. The sequences were obtained from the
following sources: T. b., T. brucei,
GenBankTM accession number AY1 35212; L. m., L. major, GenBankTM accession
number AY064402; H. sap., Homo
sapiens, GenBankTM accession number U51920; S. cer., S. cerevisiae, GenBankTM accession
number 20424; E. coli, GenBankTM accession
number P07019.
|
|
Silencing of SRP54 Expression by RNAi--
As a first step toward
understanding the essentiality of the SRP pathway for trypanosomes and
developing criteria for translocation defects because of depletion of
the SRP pathway, we alleviated this pathway by silencing
SRP54 using the RNAi methodology.
We used a construct that expresses the dsRNA in the form of a stem-loop
structure (32). After linearizing the construct, the DNA was used to
transfect T. brucei 29-13 cells that express the
tetracycline repressor as well as the T7 polymerase (39). The stem-loop
RNA (dsRNA) was designed to contain 1620 nt composed of 535 nt sense, a
550-nt loop derived from the Pex11
gene, and 535-nt
antisense RNA. The EP promoter, which is under the control of the
tetracycline repressor, is used to transcribe this RNA. In addition,
the construct also carries the phleomycin resistance gene, which is
under the control of the T7 promoter and the rRNA spacer regions, which
enable its integration to the nontranscribed rDNA spacer (32). Because
the population of transformants is heterologous due to integration into
different rDNA loci, the cells were cloned by selection on agar plates.
Different colonies were selected and the growth was monitored after the
addition of tetracycline. For further analysis, we chose the clones
whose growth was severely affected upon the addition of tetracycline.
The correlation between the growth defects and the level of
SRP54 mRNA was examined. Briefly, RNA was extracted from
the parental 29-13 cells and from the strain carrying the SRP54
RNAi construct, before and after induction with tetracycline, and
was subjected to Northern analysis with a probe derived from the coding
region of the SRP54. The results are presented in Fig.
2A and indicate that 1-2 days
after induction, the mRNA was reduced by ~90% (determined by
densitometric analysis). The production of dsRNA was also observed in
the absence of tetracycline, and its production was slightly reduced
during the course of silencing. However, reduction in the level of
dsRNA did not change the reduction in the level of SRP54
mRNA. The degradation of SRP54 was specific because no reduction was observed for tubulin mRNA (Fig. 2A).
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 2.
Northern and Western analyses of
SRP54 mRNA and protein upon SRP54
silencing. A, Northern analysis of SRP54 mRNA. RNA
was prepared from the parental strain (PS), and cells
carrying the RNAi construct, uninduced ( Tet), and after
1-8 days of induction (+Tet). Total RNA (20 µg) was
subjected to Northern analysis with a random-labeled SRP54 and tubulin
probes. The transcripts of SRP54 mRNA, dsRNA, and tubulin are
indicated by arrows. B, Western analysis of SRP54 protein.
Whole cell extract was fractionated on a 10% SDS-polyacrylamide gel
and subjected to Western analysis with the anti-yeast Srp54p antibodies
(18), and antibodies specific for mammalian cytoplasmic RNP protein
hnRNP D0 (48). SRP54 and hnRNP D0 proteins are indicated by
arrows. PS, parental stain, Tet and +Tet, uninduced and
induced cells carrying the RNAi construct, respectively.
1-8 are the dates after induction.
|
|
To examine whether the reduction at the level of SRP54
mRNA was accompanied by a reduction at the level of the protein,
extracts from parental strain and cells carrying the RNAi construct,
induced and uninduced, were subjected to Western analysis with
antibodies raised against the yeast SRP54 homologue. The results (Fig.
2B) indicate that there was almost a 95% reduction at the
level of the protein from the second day of silencing. The level of
SRP54 did not change during the course of silencing (8 days).
Antibodies to the mammalian cytoplasmic hnRNP D0 protein (48) that
cross-reacts with the T. brucei protein were used as a
control to demonstrate the specific reduction at the level of SRP54.
The Effects of SRP54 Silencing on Cell Growth, Cytokinesis, and the
Kinetoplast Content--
The growth of cells carrying the RNAi
construct upon tetracycline induction was compared with those cells
carrying the RNAi construct (uninduced) and with the parental strain.
The results, presented in Fig.
3A, a, indicate
that the growth of the induced cells was severely affected. Three days
after tetracycline induction the growth of the silenced cells ceased,
whereas the uninduced cells grew at much a higher rate than the induced
cells but were slightly inhibited compared with the parental strain.
The growth inhibition of the uninduced cells originates from
"leaky" dsRNA production. In other experiments (data not shown)
massive death was observed after 10 days. Interestingly, the inhibition
of growth was reversible; if induction was for 3 days and tetracycline
was removed after that time, the cells continued to grow (Fig.
3A, b). However, if silencing was for 6 days, the
cells never grew again after tetracycline removal and eventually died.
These results indicate that SRP54 is essential for parasite
growth.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 3.
A, growth curves of
cells carrying the SRP54 RNAi construct. a, the
growth of parental stain (PS) was compared with the silenced
cells. The number of cells of uninduced (Tet) is
designated by filled triangles, of induced cells
(+Tet) by filled squares, and the PS by
circles. The arrow indicates the time when
tetracycline was added. b, the reversibility of cell growth
after the removal of tetracycline. Growth was monitored as in
A. The addition of tetracycline is indicated by an
arrow. Cell growth in the presence of tetracycline is
designated by filled triangles. After 3 or 6 days the
tetracycline was removed (Tet) (indicated in
circles and squares, respectively) and growth was
monitored thereafter. B, changes in cell shape and
kinetoplast during SRP54 silencing. Uninduced cells
(Tet) and induced cells (+Tet) 1-6, 7-8, and
10 days after tetracycline induction were fixed, permeabilized, and
reacted with CI serum as described under "Materials and Methods."
The nucleus and kinetoplast were stained with PI. The cells were
incubated with antisera obtained from a rabbit that was chromically
infected with T. brucei. The fluorescence images were
visualized in a confocal microscope. Detached flagellums
(DF) are indicated. C, the content of kinetoplast
and nuclei upon SRP54 silencing. Uninduced
(Tet) and induced cells (+Tet) 4 days after
tetracycline induction were stained with PI. The numbers of
kinetoplast (K) and nuclei (N) per cell are
indicated. Scale bar, 5 µm.
|
|
The most remarkable phenotype of the SRP54-silenced cells
was visualized by confocal microscopy. Three parameters were examined: nuclear content, number of kinetoplasts, and the shape of the cells.
The cells were subjected to immunofluorescence with CI antibodies
described under "Materials and Methods." The nucleus and
kinetoplast were stained with PI. The results are presented in Fig. 3,
B and C. Major changes in the shape of the cells
were evident already 1 to 3 days upon the addition of tetracycline. The
different phenotypes observed were: 1) cells lost their normal shape;
2) cells possessing detached flagella; and 3) multiple conjoined cells,
two or more cells that were physically connected and moving in
different directions. After 4 days of SRP54 silencing, most
of the cells contained more than a single nucleus. These phenotypes
suggest that the silencing disrupted normal cytokinesis. At the later
stages of silencing (10 days or more), the cells were rounded and
contained numerous nuclei. Anucleate cells were also observed and many
of the cells lost their flagellum. At this point these "monstrous"
cells stopped moving and died. Interestingly, during later stages of
induction (day 8 and onward), the staining with CI serum and PI was
yellow because of the merging between the red PI staining and the green
fluorescence of the antibodies (Fig. 3B). The yellow
staining suggests that either the nuclear envelope became more
permeable to the antibody or that during silencing, epitopes, which are
normally internal, were exposed, like in apoptotic cells. To examine
whether the cells undergo apoptosis, we analyzed the DNA from induced
and uninduced cells, but no signs for apoptotic DNA degradation were
observed in the silenced cells.
In asynchronous trypanosome culture, there are either cells that carry
one nucleus and one kinetoplast (1K1N), or cells carrying 2N2K and
therefore have completed mitosis and are ready for cytokinesis. Whereas
staining the cells with PI (Fig. 3C), we noticed that during
silencing (starting from day 4) multiple nuclei started to accumulate,
which was not accompanied by kinetoplast multiplication, suggesting
that uncoupling takes place between nuclear and kinetoplast duplication
and segregation. At later stages of induction we could observe cells
carrying four, even up to nine nuclei per cell with a single
kinetoplast. These results suggest that during silencing, both
cytokinesis and kinetoplast multiplication were severely affected.
To correlate the deleterious effects of SRP54 silencing with
specific defects in protein translocation, we examined the processing of four different proteins harboring the signal peptide. We chose for
most of the analyses cells on the second day after tetracycline induction, because this is the first day that the level of SRP54 protein was minimal and no dramatic effect on cell shape was yet observed.
Changes in the Distribution and Content of the Procyclin Surface
Protein EP during SRP54 Silencing--
The first protein examined was
the glycosylphosphatidylinositol (GPI)-anchored surface glycoprotein,
the procyclin EP. EP contains up to 30 tandem repeats of glutamic acid
and proline. The EP protein is attached to the membrane by a complex
glycosylated GPI anchor that is capped with sialic acid (49).
The EP gene encodes for a 15-kDa protein. The protein
undergoes different modifications including GPI anchoring and
glycosylation. The fully glycosylated protein migrates as a
heterogeneous population of the 45-50-kDa protein on a 12%
polyacrylamide gel. The heterogeneity is derived from the addition of
branched poly-N-acetyllactosamine to the GPI anchor. Further
contributing to heterogeneity is the sialylation of the GPI branches by
trans-sialidase when EP arrives at the cell surface (50). A
monoclonal antibody that recognizes the EP by immunofluorescence and
Western analysis was used to examine the effects of silencing on EP
maturation and cellular localization. To this end, proteins prepared
from uninduced and induced cells were subjected to Western analysis
(Fig. 4A). No changes were
observed between induced and uninduced cells. The fully processed
proteins formed during silencing can represent protein translocated by
the residual level of SRP (less than 10%) or most likely proteins that
are translocated by an alternative chaperone pathway as in yeast (18,
21). The putative precursor, 35-kDa polypeptide (designated PEP in Fig.
4A) was also present in uninduced cells. Recent studies on
EP processing suggest that the EP precursor carrying the GPI anchor is
already modified with a short carbohydrate side chain, suggesting that
PEP is an ER-precursor (51).
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 4.
The effect of SRP54 silencing on
EP translocation. A, Western analysis with anti-EP
antibodies. Extracts (106 cells per lane) were prepared
from uninduced cells (Tet) and induced cells 1, 3, and 6 days upon tetracycline induction (+Tet). After separating
the proteins in a 12% SDS-polyacrylamide gel, the blot was reacted
with monoclonal antibodies against EP. The size of the marker proteins
is indicated in kDa. The positions of the PEP (a putative ER precursor)
and the mature glycosylated protein (MEP) are indicated.
B, immunofluorescence assay (IFA) with anti-EP
(surface staining). Cells were fixed with 4% formaldehyde for 20 min.
The cells were visualized by indirect immunofluorescence after
incubation with anti-EP antibodies. Uninduced cells (Tet)
and induced cells (+Tet) on the second day after induction
are indicated. C, immunofluorescence assay with anti-EP
(internal staining). Cells were stained as described in A,
except that the cells were permeabilized with 0.1% Nonidet P-40.
Scale bar, 5 µm.
|
|
To further explore the processing of EP under SRP54
silencing, we examined the cellular distribution of the protein.
Briefly, cells were fixed by paraformaldehyde and reacted with anti-EP antibodies. This type of staining detects only the protein present on
the surface of the parasite. The results, presented in Fig. 4B, indicate that while in the wild-type cells the EP was
uniformly distributed on the surface, forming a tight and heavy coat,
upon silencing, the coat became much thinner. In the presence of
detergent (Nonidet P-40) the antibodies can enter the cell and stain
the EP internally. Indeed, staining in the presence of detergent
revealed very strong punctated staining of the induced cells compared
with uninduced cells that showed only staining around the flagellar pocket (Fig. 4C), suggesting that the protein is synthesized
but cannot be properly translocated to the surface of the parasite. The
punctated staining observed during silencing is reminiscent of staining
observed during overproduction of the T. brucei TbRab2p protein that regulates intracellular membrane trafficking (52). If the
phenotype observed in SRP54 silenced cells is similar to overexpression of TbRab2p, this suggests that the mislocalization observed for EP is because of a secondary defect that may result from
augmented production of ER transport vesicles.
The Translocation of BiP Takes Place during SRP54
Silencing--
The BiP is an ER resident chaperone. SRP depletion in
S. cerevisiae but not in S. pombe affected the
translocation of BiP (18, 21). The distribution of BiP was examined
upon silencing by immunoprecipitation and confocal microscopy. The
uninduced and SRP54-silenced cells on the second day after
induction were subjected to pulse-chase labeling with a
[35S]methionine/cysteine mixture. The extracts were
subjected to immunoprecipitation with anti-BiP antibodies. Even during
a short pulse-chase time, no preprotein was detected (Fig.
5A), suggesting that the BiP
is normally translocated during SRP54 silencing, like in
S. pombe (21). To verify that BiP is normally translocated during SRP54 depletion, the uninduced and silenced cells on the second
day after induction were stained with anti-BiP antibodies. The results,
presented in Fig. 5B, demonstrate that in induced cells, BiP
is still distributed within the ER, suggesting that despite the SRP54
depletion, BiP is normally translocated to the ER. It should be noted
that exactly the same results were obtained when the cells were
analyzed on later days of induction, indicating that the ability to
translocate the protein to the ER was maintained through the course of
silencing.
View larger version (52K):
[in this window]
[in a new window]
|
Fig. 5.
Distribution of BiP in uninduced and induced
cells. A, immunoprecipitation of in vivo labeled
BiP protein. Uninduced (Tet) and induced cells
(+Tet) on the second day after tetracycline induction were
labeled with [35S]methionine/cysteine mixture (5-min
pulse and 5- or 15-min chase). Extracts were prepared and
immunoprecipitation was performed as described under "Materials and
Methods." The total cell extracts and the immunoprecipitated products
were analyzed on a 8% SDS-polyacrylamide gel. The size of the marker
proteins is indicated in kDa. The specific immunoprecipitated product
is indicated by an arrow. B, immunofluorescence with
anti-BiP antibodies. Uninduced cells (Tet) and induced
cells (+Tet) on the second day after induction were fixed
and reacted with anti-BiP antibodies. The images were visualized using
the confocal microscope. Scale bar, 5 µm.
|
|
The Translocation of Lysosomal Protein p67 during SRP54
Silencing--
To examine the defects of SRP 54 depletion on integral
membrane protein, we examined the localization of p67, which is a
lysosomal protein that is highly glycosylated (44). The protein is a
type I transmembrane protein that harbors a luminal domain and short cytoplasmic domain. In bloodstream trypanosomes, p67 is synthesized as
a 100-kDa glycoform (gp100) harboring an estimated 14 N-linked oligosaccharides. It is then transported to
the Golgi where elongation of the N-glycans takes place. In
procyclic form, p67 is synthesized exclusively as the gp100 glycoform,
which is transported directly to the lysosome (44).
To follow the translocation of p67 during silencing, we monitored the
synthesis and processing of nascent p67. Cells on the second day of
silencing were subjected to pulse-chase with
[35S]methionine/cysteine. Extracts were prepared and
subjected to immunoprecipitation. A major 100-kDa protein was observed
but the p67 precursor was hardly detected (Fig.
6A). However, two additional
immunoprecipitated products, 50- and 37-kDa polypeptides, were observed
upon silencing. Interestingly, a mature glycosylated protein appeared
in the induced cells in amounts identical to the uninduced cells. The
synthesis of the fully glycosylated form under the silencing condition
may reflect either the residual activity of the SRP or the activity of
an alternative translocation pathway as already discussed. The 50- and
37-kDa polypeptides are strikingly similar to the standard p67
degradation products that can be observed in normal cells (44).
However, these degradation products are normally seen only after
several hours of chase in procyclic cells, whereas the duration of
chase was only for few minutes in the experiment presented in Fig.
6A. The degradation of p67 is mediated by trypanopain, the
major lysosomal thiol protease, which also has to enter the secretory
pathway. Perhaps trypanopain is also mislocalized to the same aberrant
secretory compartment discussed above and cleavage takes place in the
same locale prior to its entry to the lysosome. Interestingly, however,
the majority of p67 is properly translocated to the ER and glycosylated
during SRP54 silencing, suggesting that also p67 can utilize
an alternative pathway for translocation.
View larger version (55K):
[in this window]
[in a new window]
|
Fig. 6.
Distribution of the lysosomal protein p67
upon SRP54 silencing. A, immunoprecipitation of p67.
Uninduced cells (Tet) and induced cells (+Tet)
on the second day after induction were labeled with
[35S]methionine/cysteine mixture (5-min pulse and 5-min
chase). The procedure is as in Fig. 5A. The extracts and the
immunoprecipitated products were analyzed on a 10% SDS-polyacrylamide
gel. The molecular mass of marker proteins is indicated in kDa. The
specific immunoprecipitated products of the fully glycosylated
glycoprotein 100, and two additional degradation products are marked by
arrows. Lanes 1 and 2 are extracts
from the uninduced and induced cells, respectively; lanes 3 and 4 are immunoprecipitated products from uninduced and
induced cells, respectively. B, immunofluorescence with
anti-p67 antibodies. Uninduced cells (Tet) and induced
cells (+Tet) on the second day after tetracycline induction
were fixed and reacted with anti-p67 antibodies (44). The images were
visualized using a confocal microscope. Kinetoplast (K),
nucleus (N), and lysosome (L) are indicated.
Scale bar, 5 µm.
|
|
To investigate the effect of silencing on the processing of
p67, SRP54-silenced cells on the second day of induction
were stained with antibodies. The results indicate that in uninduced cells staining was confined to the lysosome localized near the nucleus.
However, in induced cells, staining was not confined to a single site
but gave punctated staining, like in the case of EP (Fig.
4C). The data therefore suggest that despite proper translocation to the ER, the p67 protein is mislocalized and is also
found in these aberrant megavesicular structures. Also in the case of
p67, the same results were observed in the later days of silencing.
Processing and Localization of the Flagellar Pocket Protein CRAM
during SRP54 Silencing--
The CRAM protein is the trypanosome
homologue to the human low density lipoprotein receptor (45). CRAM is a
type I membrane protein that is concentrated in the flagellar pocket,
an invagination of the cell surface of trypanosomes where endocytosis
and exocytosis take place. CRAM has a predicted molecular mass of 130 kDa. The glycosylated form of the protein migrates as a 200-kDa
protein. The protein harbors a transmembrane domain and a 41-amino acid cytoplasmic extension. CRAM also possesses a large intracellular domain
with a cysteine-rich repeat.
To investigate the effect of silencing on the processing of CRAM,
Western analysis was performed on whole cell lysates from induced and
uninduced cells. The results presented in Fig.
7A indicate that the protein
was synthesized at the level identical to the level in the uninduced
cells and the protein was perfectly glycosylated. This result resembles
the data obtained for EP. We further examined the localization of CRAM
by immunofluorescence. The results, presented in Fig. 7B,
reveal major differences between the uninduced and induced cells.
Whereas in uninduced cells, the CRAM was confined to the flagellar
pocket, in the induced cells the staining was punctated and not
concentrated in the flagellar pocket similar to the staining observed
for p67 and EP, indicating defects in localization but not in
translocation and glycosylation.
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 7.
Distribution of CRAM upon SRP54
silencing. A, Western analysis was as described in the
legend to Fig. 4A. The anti-hnRNPD0 was used as a control
for equal loading. B, immunofluorescence of cells with
anti-CRAM antibodies. Uninduced cells (Tet) and induced
cells (+Tet) on the second day after induction are
indicated. The position of the flagellar pocket is marked. Scale
bar, 5 µm.
|
|
| |
DISCUSSION |
The signal peptide-binding protein SRP54 was identified in
trypanosomatids and compared with its homologues from E. coli to mammals. RNAi of the SRP54 gene was utilized to
elucidate the essentiality of the SRP pathway and its role in protein
translocation in T. brucei. SRP54 depletion elicits dramatic
effects on cell shape, cytokinesis, and kinetoplast duplication.
Surprisingly, all the signal peptide-containing proteins examined were
translocated to the ER during silencing. However, a secondary effect on
protein sorting was observed for the GPI-anchored surface protein (EP) and two type I integral membrane proteins (p67 and CRAM). These proteins were translocated to the ER and properly glycosylated but
mislocalized. The translocation of EP, BiP, p67, and CRAM to the ER
upon depletion of SRP54 indicates that an alternative protein
translocation pathway exists in trypanosomes. However, the
mislocalization of these proteins suggests that the proper sorting of
these proteins was not compromised by this alternative pathway. The
secondary effect on protein sorting may stem from defects in the
translocation of polytopic membrane proteins. In the absence of
antibody reagents to T. brucei multispanning membrane proteins, we could not examine the effect of SRP54 silencing
on the translocation of such proteins.
The T. brucei SRP54 is highly related to its homologous
proteins in other organisms. It is more related to its mammalian
counterpart than to yeast, as was observed for the 7SL RNA (53). Like
all SRP54 proteins, the trypanosome protein carries the conserved N, G,
and M domains. However, three amino acid changes were observed in the
most conserved positions (circled in Fig. 1). Interestingly, all these changes convert negatively charged amino acids to uncharged amino acids. Functional data are necessary to determine whether these
changes contribute to the peculiar behavior of the signal peptide-containing proteins in the in vitro mammalian
translocation system (47).
The Cellular Implications of SRP54 Depletion in T. brucei--
This work joins an increasing number of studies that have
utilized RNAi in trypanosomes to elucidate the function of essential genes. In all cases, when dsRNA production was induced, the target mRNA decreased after 24 h and the protein levels dropped to
~10% after 24 to 48 h of induction (35, 36). In this study, the mRNA dropped to less than 10% after 2 days. This reduction in mRNA level was accompanied by a similar reduction in the level of
the protein. The depletion of SRP54 resulted in the inhibition of
growth, similar to growth defects that were observed for other essential genes (35, 36). In all cases reported, growth continued for
the first few days upon tetracycline induction. However, when essential
genes were silenced, as in this study, the cells stopped growing (after
3 days) and the number of cells either remained constant or was
reduced thereafter. Microscopic examination of cells after 10 days of
induction (as shown in Fig. 3B) indicated that all cells had
aberrant morphology, either multinucleated or anucleated and eventually
died. These results suggest that like topoisomerase II (36) and
many of the exosome genes (35), SRP54 is also an essential gene.
Note that the dramatic morphological changes observed here were not
observed in any other organisms where SRP function was previously
disrupted. However, this is the first study that was carried out in an
organism that lacks a tight cell wall. In the absence of cell wall
depletion of the surface coat, GPI-anchored or polytopic membrane
proteins that interact with the subpellicular microtubules may change
the shape of the cell and lead to the production of monster and huge
round cells observed in this study.
How Does SRP Depletion Affect Protein Translocation in Trypanosomes
Compared with Other Organisms--
The translocation defects observed
during the silencing of the SRP54 in T. brucei
are different from those observed during SRP depletion in S. cerevisiae. Four proteins were chosen to examine the translocation
defects: 1) DPAP-B, a type II integral membrane protein; 2) the Kar2p,
the ER resident chaperone (Kar2p = BiP); 3) invertase; and 4) the
secreted protein carboxypeptidase. The translocation of DPAP and Kar2p
was severely affected by SRP depletion but carboxypeptidase was
normally translocated to the ER. These data suggest that the proteins
differ in their dependence on the SRP pathway and some proteins can
utilize the alternative chaperone pathway (18, 20). Indeed, the
translocation of carboxypeptidase was completely dependent on the
chaperone pathway, because it was blocked in a chaperone pathway
mutant, Sec63 (20). Likewise, in S. pombe, SRP54
depletion severely affected the translocation of acid phosphatase but
had only minor effects on the secretion of BiP (21). These results
suggest that yeast process their proteins by two active pathways: the
SRP pathway and the chaperone pathway. The choice between these
pathways for any given protein is unpredictable because the same
protein BiP is translocated in S. pombe by the chaperone
pathway (21) and in S. cerevisiae (18) and Y. lypolytica (23) by the SRP pathway. The pathway specificity is
conferred by the hydrophobic core of signal sequences (54).
The presence of these two pathways for protein translocation explains
why depletion of the SRP pathway is not lethal in S. cerevisiae. However, structural features of some preproteins may preclude efficient translocation using a post-translational route. These proteins must therefore depend on the SRP pathway for
translocation. In S. cerevisiae, all essential proteins can
be translocated via the chaperone pathway; however, the cells grow
slowly and are defective. In contrast, in S. pombe the
pathway is essential, because essential proteins are dependent on the
SRP pathway for translocation.
The intriguing finding that the translocation of four proteins from
different locales (ER, surface of the parasite, lysosome, and flagellar
pocket) may suggest that in trypanosomes all signal peptide-containing
proteins do not utilize the SRP pathway. The SRP pathway may
exclusively translocate polytopic membrane proteins. This is indeed the
case in E. coli. In E. coli the essentiality of
the SRP pathway stems from the absolute dependence on the SRP pathway
for processing of polytopic membrane proteins (15, 16). If this is also
the case in trypanosomes and because trypanosomes diverged very early
from the eukaryotic lineage (24) this may suggest that the primordial
function of the SRP particle was to translocate polytopic membrane
proteins and only later in the course of SRP evolution was it adjusted
to translocate signal peptide-containing proteins. It will therefore be
of great interest to describe all the proteins that are severely
affected during SRP54 silencing by comparing the nascent
translated proteins in uninduced and SRP54-silenced cells by
two-dimensional gels. Such a proteomic approach will define all the
trypanosome proteins that are dependent on the SRP pathway, and that
their translocation cannot be compromised by the chaperone alternative pathway.
We cannot exclude the possibility that all signal peptide-containing
proteins are properly translocated to the ER during SRP54 silencing. However, the fact that four different signal
peptide-containing proteins were translocated to the ER during
silencing may suggest that the major deficiency in the silenced cells
originates from defects in polytopic membrane protein biogenesis.
Defects in the processing of such membrane proteins could change the
composition of both intracellular membranes such as the ER and Golgi
apparatus and the plasma membrane and affect the intracellular
trafficking. Such secondary defects may cause the accumulation of the
post-ER membranous structures observed for p67, EP, and CRAM. The
mislocalization observed for these proteins is therefore a secondary
effect that may result from sorting defects. Interestingly defects in
secretion of signal peptide-containing proteins (-lactamase) were
also observed in E. coli SRP receptor,
FtsY-depleted cells but these effects were secondary
and appeared only after the clear effects on the membrane proteins (16,
55).
At present there are no antibody reagents to examine the processing of
polytopic membrane proteins in T. brucei and this analysis awaits the development of a reporter fusion protein carrying
multispanning membrane segments. Interestingly, however, membrane
proteins carrying an N-terminal signal peptide and a single TM, such as
the CRAM and p67, were translocated to the ER under SRP54 depletion but were mislocalized.
The Trypanosome SRP Complex--
At present, we know very little
about the T. brucei SRP protein composition. We have
recently identified three additional SRP protein homologues, SRP19,
SRP72, and SRP68 in the T. brucei genome
project.3 It is currently
unknown whether the trypanosome 7SL RNA, like in yeast and mammals, is
first assembled with five of the SRP proteins in the nucleolus (56,
57). Experiments are in progress to silence the genes encoding for
SRP19 and SRP72. These experiments will enable us to determine whether,
like in yeast, the assembly of the trypanosome particle takes place in
the nucleolus (56). Interestingly, the silencing of SRP54
did not affect the level of 7SL RNA. In addition, when SRP complex from
the silenced cells was analyzed on sucrose gradient only a marginal
difference was observed in its S value compared with uninduced cells
(data not shown). These results indicate that the SRP subparticle can
be stably formed in the absence of SRP54. Srp54p depletion in S. cerevisiae also did not significantly change the S value of the particle. However, no SRP complex was observed upon depletion of
Srp68p, Srp72p, Srp21p, and Srp14p (20). These differences stem from
the fact that these four proteins are essential for the assembly of the
preparticle in the nucleolus, whereas SRP54 joins the preparticle in
the cytoplasm (56).
The most intriguing property of the trypanosome particle is its RNA
composition. The trypanosome SRP complex carries a tRNA-like molecule
that is a genuine constituent of the particle (28).2 The
function of this unique tRNA-like molecule is unknown. We proposed that
it may carry the arrest function of the particle and substitute for the
truncated domain I of the trypanosomatid 7SL RNA. However, we cannot
exclude the possibility that this RNA may also be involved in the
transport of the particle to the cytoplasm, because domain I of 7SL RNA
was shown to be involved in the transport of SRP from the nucleolus to
the cytoplasm (58).
This is the first study to decipher the in vivo function of
SRP in trypanosomes. The unique RNA composition of the SRP and the
essentiality of the SRP translocation pathway make it a very attractive
drug target to fight the diseases caused by this important family of parasites.
, and
TOP
ABSTRACT
INTRODUCTION
