Scaffolding function of the Chlamydomonasprocentriole protein CRC70, a member of theconserved Cep70 family

Gen Shiratsuchi, Ritsu Kamiya and Masafumi Hirono*Department of Biological Sciences, University of Tokyo, Tokyo 113-0033, Japan

*Author for correspondence (hirono@biol.s.u-tokyo.ac.jp)

Accepted 17 May 2011Journal of Cell Science 124, 2964–2975� 2011. Published by The Company of Biologists Ltddoi: 10.1242/jcs.084715

SummaryCentriole duplication occurs once per cell cycle through the assembly of daughter centrioles on the side wall of pre-existing centrioles.Little is known about the molecules involved in the assembly of new centrioles. Here, we identify CRC70 as a Chlamydomonas protein

with an important role in the accumulation of centriole proteins at the site of assembly. CRC70 contains a highly conserved ,50-amino-acid sequence shared by mammalian Cep70 and preferentially localizes to immature centrioles (the procentrioles). This localization ismaintained in the mutant bld10, in which centriole formation is blocked before the assembly of centriolar microtubules. RNAinterference (RNAi)-mediated knockdown of CRC70 produces flagella-less cells and inhibits the recruitment of other centriole

components, such as SAS-6 and Bld10p to the centriole. Overexpression of CRC70 induces an accumulation of these proteins in discretespots in the cytoplasm. Overexpression of EGFP-tagged CRC70 in mouse NIH3T3 cells causes the formation of structures apparentlyrelated to centrioles. These findings suggest that CRC70 is a member of a conserved protein family and functions as a scaffold for the

assembly of the centriole precursor.

Key words: Centriole, Basal body, Cartwheel, SAS-6, bld10, bld12

IntroductionThe centriole is a cellular organelle that functions as the core of the

centrosome, as well as the assembly site for the cilium (Azimzadeh

and Bornens, 2007). Centrioles have a characteristic structure

consisting of nine triplet microtubules and their assembly occurs

once per cell cycle through a ‘duplication’ process that has

remained a long-standing puzzle. Various ultrastructural studies

have shown that the first step of assembly is the appearance of an

amorphous structure on the side wall of the mature cylindrical

centriole. This amorphous structure is called the generative disk

in Paramecium (Dippell, 1968) and the amorphous ring in

Chlamydomonas (O’Toole et al., 2003). In the next step, a

cartwheel structure composed of a central hub and nine spokes is

produced on the amorphous ring. Once this complex is formed,

nine triplets assemble at the tips of the cartwheel spokes and

elongate distally to ,400 nm (Dutcher, 2007). The incomplete

centriolar structures that appear next to the mature centriole are

called procentrioles. These findings imply that the characteristic

circular arrangement of triplet microtubules is determined at an

early stage of assembly, probably during the formation of

procentrioles. The identification of the molecules that form the

procentrioles is therefore important to understand the mechanism

of centriole assembly.

Recent studies using Caenorhabditis elegans identified a set of

genes that play a role in the early steps of centriole assembly.

SPD-2 (Kemp et al., 2004; Pelletier et al., 2004) and ZYG-1

(O’Connell et al., 2001) were found to trigger centriole assembly.

SAS-5 (Delattre et al., 2004) and SAS-6 (Dammermann et al.,

2004; Leidel et al., 2005) mediate the formation of the central

tube, which is a centriole precursor uniquely found in C. elegans

(Strnad and Gonczy, 2008). SAS-6 was shown to be a cartwheel

component in Chlamydomonas (Nakazawa et al., 2007), where

another cartwheel protein, Bld10p, was also identified (Matsuura

et al., 2004; Hiraki et al., 2007). Analyses of the null mutants

bld12 and bld10 revealed that the cartwheel serves as the scaffold

for centriolar microtubule assembly and that it stabilizes the

ninefold symmetry of the centriole. In fact, the bld12 mutant

lacked the radial structure of the cartwheel and frequently had

centrioles with aberrant numbers of triplet microtubules, ranging

from seven to 11 (Nakazawa et al., 2007). Despite the defects in

the cartwheel, the majority of the centrioles in bld12 still had nine

microtubule sets, suggesting that the ninefold symmetry is

established by multiple factors (Nakazawa et al., 2007).

Proteins specifically expressed in the procentriole before

cartwheel formation might therefore play important roles in

establishing the ninefold symmetry of the centriole.

Here, we describe the identification of CRC70, a coiled-coil

protein from Chlamydomonas, which functions in the assembly of

the procentriole before the formation of the cartwheel. CRC70

shows a partial similarity to mammalian Cep70. RNA interference

(RNAi)-mediated depletion of CRC70 in Chlamydomonas

prevents the recruitment of SAS-6 and Bld10p to the centriole,

whereas overexpression results in the accumulation of SAS-6 and

Bld10p in multiple foci in the cytoplasm. When expressed in

mouse 3T3 cells, CRC70 induced the formation of multiple

centriole-like structures in the cytoplasm. Hence, CRC70 is a

2964 Research Article

Journ

alof

Cell

Scie

nce

hitherto unidentified protein that participates in the early process of

centriole assembly.

ResultsCRC70 is a coiled-coil protein related to Cep70

Proteomic analyses of centrosomes and basal bodies (Andersenet al., 2003; Li et al., 2004; Keller et al., 2005; Kilburn et al., 2007)

have shown that the centrioles contain many high-molecular-massproteins with extensive coiled-coil domains. Here we identified agene encoding a high-molecular-mass coiled-coil protein in asearch for the mutation responsible for the deficiency in centriole

maturation of the Chlamydomonas mutant uni1 (Huang et al.,1982). Although the identified gene was determined, by a PCR-based method, to be located close to the uni1 locus (Kathir et al.,

2003), no mutation was found in the cDNA sequence for uni1.Although its relationship to the uni1 mutation is not clear, thiscoiled-coil protein was shown to be a centriole protein.

The gene has a single 5592 bp open-reading frame (ORF) that

encodes a protein of 1864 amino acids with a molecular mass of

201 kDa. The program PAIRCOIL (Berger et al., 1995) predicted

that the entire deduced amino acid sequence, except for the C-

terminal 20%, had a high probability of forming coiled-coils

(Fig. 1A). BLAST searches using short segments of the predicted

protein revealed that a sequence of ,50 amino acids in the C-

terminal non-coiled-coil region is highly similar to the partial

sequence of a mammalian centrosomal protein, Cep70 (Fig. 1);

in fact, the predicted protein and Cep70 are mutually ‘best hits’.

It was therefore named CRC70 for Chlamydomonas protein

related to Cep70. Proteins containing the conserved 50-amino-

acid sequence, which was designated ‘the Cep70 motif’, were

found in vertebrates, coelenterates, bryophytes and algae

(Fig. 1B), but not in higher plants, yeast, C. elegans or

Drosophila. Chlamydomonas and most of the other organisms

in which these proteins were present only had a single protein of

Fig. 1. CRC70 is a member of a conserved protein family

related to Cep70. (A) Schematic representation of the proteins

that have the Cep70 motif (GenBank accession numbers are given

unless otherwise noted). Chlamydomonas reinhardtii CRC70

(AB470484); a Volvox carteri predicted protein (JGI protein ID

117327); a Physcomitrella patens predicted protein

(XP_001782614); an Ectocarpus siliculosus predicted protein

(CBN78542); a Trichoplax adhaerens predicted protein

(XP_002109671); a Nematostella vectensis predicted protein

(XP_001629475); Danio rerio Cep70 (NP_001108048) (Wilkinson

et al., 2009); a Xenopus laevis predicted protein (NP_001089214);

Mus musculus Cep70 (NP_076362); and Homo sapiens Cep70

(NP_077817) (Andersen et al., 2003). The Cep70 motif is shown in

red and coiled-coil domains in gray. The boxes with broken lines

indicate predicted amino acid sequences based on incompletely

determined DNA sequences. The C-terminal half of CRC70

displays 51–94% similarity to other Cep70s. The double arrow

indicates the peptide used as the antigen. aa, amino acids.

(B) Alignment of the ‘Cep70 motif’ sequences. The amino acid

sequences were aligned and plotted using ClustalW and

BOXSHADE. Black and gray shadings indicate regions of identity

and conservative amino acid substitutions, respectively. (C) A

phylogenetic tree showing relative distances among Cep70 family

proteins. The distance was visualized using the NJplot program

(Perriere and Gouy, 1996). To assess the confidence level of the

phylogenetic tree, bootstrap probabilities were estimated with 1000

bootstrap replicates, and are shown at the nodes (Felsenstein, 1985).

CRC70 is a scaffold protein in centriole assembly 2965

Journ

alof

Cell

Scie

nce

this family. In a structure-based sequence alignment, proteins

containing this motif were found to share a common pattern,

consisting of an extensive coiled-coil domain in the N-terminal

region and the Cep70 motif close to the C-terminus (Fig. 1A).

This structural similarity suggests a common origin (Fig. 1C).

The C-terminal half of CRC70 displays a high degree of

homology (9–69% identity and 51–94% similarity) to other

Cep70 family proteins (Fig. 1).

CRC70 localizes to procentrioles

A polyclonal antibody was raised against a bacterially expressed

peptide corresponding to residues 16–360 of CRC70 (Fig. 1A)

and was affinity-purified with the same peptide. Western blot

analysis using this antibody detected a single band in a protein

sample of the nucleoflagellar apparatus (NFAp), a cytoskeletal

complex containing two mature centrioles, two procentrioles and

two axonemes (Fig. 2A,C). The apparent molecular mass of the

band was as expected from the deduced amino acid sequence.

However, no band was detected in western blot analyses of whole

cell extracts (Fig. 2A), suggesting that CRC70 is present only in

small amounts in the cell but is concentrated in the centriole-

containing cytoskeleton.

Consistent with these results, immunofluorescence microscopy

revealed that CRC70 was localized to discrete spots near the

flagellar-proximal end of the cell (Fig. 2B). The CRC70 signal

was always observed as one or two fluorescent spots, suggesting

that this protein is localized to either centrioles or procentrioles.

In samples double-stained with the anti-CRC70 antibody and an

anti-acetylated a-tubulin antibody, the fluorescent spots for

CRC70 were located beside but not on the flagella–basal body

axis. The positional relationship between centrioles and

procentrioles in the NFAp (Fig. 2C) implied that most CRC70

molecules are localized to procentrioles. This idea was supported

by the analysis of cells double-stained with the anti-CRC70

antibody and an antibody against polyglutamylated tubulin,

which has been shown to specifically localize to mature

centrioles in Chlamydomonas (Lechtreck and Geimer, 2000).

Fluorescent spots corresponding to CRC70 and polyglutamylated

tubulin were located side by side but were not exactly

superimposed (supplementary material Fig. S1).

Immunoelectron microscopy confirmed the procentriolar

localization of CRC70. Cross-sections containing both

centrioles and procentrioles showed that the majority of the

gold particles representing CRC70 were near the procentrioles

and most densely distributed on the side facing the center of the

cruciate rootlet microtubules: the side near triplets 1, 8 and 9 in

the numbering system proposed by Geimer and Melkonian

(Geimer and Melkonian, 2004) (Fig. 2D,E; supplementary

material Fig. S2). In longitudinal sections of the centriole and

nascent procentrioles, gold particles were detected surrounding

the procentrioles and on transitional fibers attached to the distal

part of mature centrioles (Fig. 2D,E; supplementary material Fig.

S2). However, localization to transitional fibers might be non-

specific, on the basis that detection of apparently unrelated

antibodies is often observed surrounding these fibers (our

unpublished results).

Fig. 2. CRC70 preferentially localizes to the

procentrioles. (A) Western blot analysis of the

whole cell extract (10 mg per lane) and NFAp

(4 mg/lane) prepared from Chlamydomonas cells

using the anti-CRC70 antibody. A band of

,200 kDa was detected in the NFAp preparations

but not in the whole cell extract. (B) Indirect

immunofluorescence images of a wild-type cell

(upper panels) and the NFAp (lower panels)

stained for CRC70 (magenta), a-tubulin (green,

upper panels), and acetylated a-tubulin (green,

lower panels). (C) Schematic diagram of the

NFAp. The broken line represents the cell body.

(D) Immunoelectron micrographs of NFAp using

anti-CRC70 antibodies. A cross section (left-hand

panel) and a longitudinal section (right-hand

panel) of the centrioles and procentrioles are

shown. Gold particles show the localization of

CRC70. (E) Diagrams showing the localization of

gold particles. Each diagram represents the

particles found in five immunoelectron

micrographs of cross sections (left-hand panel)

and longitudinal sections (right-hand panel) of

centrioles (supplementary material Fig. S2). c,

centriole; p, procentriole; rt, rootlet microtubules;

cw, cartwheel; tf, transition fiber. Scale bars:

2.5 mm (B); 100 nm (D).

Journal of Cell Science 124 (17)2966

Journ

alof

Cell

Scie

nce

CRC70 appears at an early stage of centriole assembly

The localization pattern of CRC70 suggests that it functions at an

early stage in the centriole assembly process. To establish a

correlation between the stages of centriole assembly and the

recruitment of CRC70, the protein was detected using

immunofluorescence microscopy in the centriole-deficient

mutants bld12 (Nakazawa et al., 2007), bld10 (Matsuura et al.,

2004; Hiraki et al., 2007) and bld2 (Goodenough and St. Clair,

1975; Ehler et al., 1995; Dutcher et al., 2002). These mutants

have defects in the early stages of centriole assembly: bld10 has a

mutation in a component of the cartwheel and lacks centriole

microtubules, bld12 also has a mutation in a cartwheel

component and its centrioles have variable numbers of

microtubule triplets, and bld2 has a mutation in e-tubulin and

fails to produce triplet microtubules. Substantial fractions of

mutant cells displayed fluorescent CRC70 signals at the center of

the radiating cytoplasmic microtubules: ,14% of cells in bld10,

,43% in bld12 and ,58% in bld2, whereas this number is

,91% in wild-type cells under the same conditions (n5200 cells

for each strain) (Fig. 3). These results suggest that CRC70 is

recruited to the centriole assembly site before the assembly of the

cartwheel.

Most CRC70 disappears from the mature centriole beforeor at the onset of the formation of the new centriole

The cell-cycle-dependent changes in CRC70 localization were

examined by co-staining cells with the anti-CRC70 antibody and

an anti-a-tubulin antibody, and sorting images according to cell

cycle stages (supplementary material Fig. S3). CRC70 spots were

observed at the base of the flagella in interphase and at the

spindle poles during the mitotic phase, suggesting that the protein

is associated with the centrioles or procentrioles throughout the

cell cycle.

To determine when CRC70 disappears during the maturation

of procentrioles, changes in the number of fluorescent spots

detected at the spindle poles were assessed. Spindle poles were

examined because centriole duplication in Chlamydomonas

begins in the mitotic metaphase or anaphase, unlike most otherorganisms in which duplication occurs in the S phase (Gaffal,

1988). Strikingly, the number of CRC70 spots at the spindle polesnever exceeded two (supplementary material Figs S1 and S3).This pattern is in contrast with that of proteins that localize to

both the procentriole and centrioles, such as Uni2p, which isdetected as three or four fluorescent spots during the mitoticanaphase (Piasecki et al., 2008). These results suggest thatCRC70 disappears from mature centrioles at the onset of

procentriole formation.

CRC70 expression substantially increases transientlybefore mitosis

Quantitative real-time PCR (qRT-PCR) showed large changes in

the CRC70 mRNA concentration during the cell cycle. WhenChlamydomonas is cultured on a 12-hour-light–12-hour-darkcycle, most cells remain in G1 phase during the light period and

go through two or three consecutive cycles of cell division duringthe dark period (Spudich and Sager, 1980) (supplementarymaterial Fig. S4A). qRT-PCR analysis was performed at varioustimepoints during the light–dark cycle, and the number of mitotic

cells was assessed using fluorescence microscopy. Under theseexperimental conditions, the highest fraction of dividing cells in aculture was ,30%, which occurred 7 hours after the beginning

of the dark period. The concentration of CRC70 mRNAdramatically increased 1 hour into the dark phase, whereas itwas hardly detected during the light phase (supplementary

material Fig. S4B). These findings suggest that CRC70 mRNAexpression, and probably the expression of its protein product,increases before the beginning of the mitotic phase.

RNAi-mediated knockdown of CRC70 impairs thecentriolar localization of SAS-6 and Bld10p

The early recruitment of CRC70 to the centriole assembly site ledto the hypothesis that this protein might play a role in initiatingcentriole assembly. To examine this possibility, CRC70 was

knocked down in Chlamydomonas cells using an artificialmicroRNA (amiRNA) (Molnar et al., 2009; Zhao et al., 2009).Wild-type cells were transformed with plasmids producing

amiRNA precursors and were screened for defective flagellargrowth, a phenotype common to most centriole-deficient mutantsof Chlamydomonas. Two clones (ami1 and ami2) were obtainedthat produced almost no flagellated cells. However, as previously

reported in RNAi experiments with Chlamydomonas (Koblenzet al., 2003; Schroda, 2006), the effect of amiRNA wasapparently unstable and the flagella-less phenotype of the two

clones was gradually lost in subsequent cell division cycles. Atthe time of the analyses, the cultures of ami1 and ami2 contained,10% and ,20% flagellated cells, respectively.

The effects of amiRNA in these clones were examined bywestern blotting and immunofluorescence microscopy(Fig. 4A,C). The amounts of CRC70 within the cell bodies of

the ami1 and ami2 lines were assessed by western blotting ofcytoskeletal fractions, which contain centrioles, and the banddensities were compared between these cells and the cells of

fla10-2, a mutant that retains normal centrioles but lacks flagellaowing to a mutation in flagellar kinesin (Matsuura et al., 2002).Cytoskeletal fractions were used because the whole cell CRC70

concentration is too low for detection by immunoblotting(Fig. 2A). The protein levels of CRC70 in ami1 and ami2 werereduced to 7% and 35% of the CRC70 level in fla10-2,

Fig. 3. CRC70 appears at an early stage of the centriole assembly

process. Indirect immunofluorescence microscopy localization of CRC70 in

bld2, bld10 and bld12 cells. Cells were stained with anti-CRC70 (magenta)

and anti-a-tubulin (green) antibodies. CRC70 is located at the center of the

radiating cytoplasmic microtubules, appearing as a white spot (arrowhead) in

each panel. Scale bar: 5 mm.

CRC70 is a scaffold protein in centriole assembly 2967

Journ

alof

Cell

Scie

nce

respectively (Fig. 4A). These results were confirmed by

immunofluorescence microscopy, which showed greatly

reduced CRC70 fluorescent signals in the flagella-less ami1

and ami2 cells, whereas fluorescent spots were clearly observed

in fla10-2 cells and in the minor populations of ami1 and ami2

cells that had flagella (Fig. 4C).

The ami1 and ami2 clones displayed slow cell growth rates,

similar to the growth rate of bld10 (Fig. 4B), which has mitotic

defects caused by aberrant centriole formation (Matsuura et al.,

2004). Strikingly, the cytoskeletal fraction of the ami1 and ami2

cells contained lower amounts of SAS-6 and Bld10p than the

control. The protein levels of SAS-6 and Bld10p were reduced to

15% and 7% of those of the wild-type cells in ami1, and to 40%

and 30% in ami2, respectively, whereas those in the whole cell

extract from either strain were not reduced (Fig. 4A). The

fluorescent spots corresponding to these proteins showed a

substantial decrease in intensity in non-flagellated ami1 and ami2

cells (Fig. 4D,E). These results suggest that the depletion of

CRC70 interferes with the recruitment of centriole proteins and

the assembly of the centriole structure.

Overexpression of CRC70 induces ectopic localization of

centriole proteins

Strains overexpressing CRC70 were established to examine

further whether CRC70 has a scaffolding function. Wild-type

cells were transformed with an expression vector containing the

full-length CRC70 cDNA tagged with FLAG or a triple

hemagglutinin (HA3) epitope. The transformants were screened

by western blotting using anti-CRC70 or anti-HA antibodies.

Transformants showing the CRC70 band, even in blots of whole

cell extract, were assumed to be overexpressing strains because

CRC70 concentration in control cells is too low to be detected in

whole cells by western blot analysis. Two strains were isolated in

a screen of ,1000 transformants (Fig. 5A). Each isolated strain

contained abnormally large cells that lacked flagella. The number

of those cells were initially ,10% of the total, but gradually

decreased in subsequent cell division cycles. This decrease is

similar to the gradual decrease of the flagella-less cells observed

in the amiRNA experiments. Both phenomena are probably due

to a suppression of exogenous gene expression frequently

observed in Chlamydomonas (Neupert et al., 2009). At the time

Fig. 4. RNAi depletion of CRC70 impairs the localization of centriolar components. (A) Western blot analysis of cytoskeletal fractions and whole cell

extracts prepared from fla10-2, ami1 and ami2 cells using anti-CRC70, anti-Bld10p and anti-SAS-6 antibodies. A Coomassie-Brilliant-Blue-stained tubulin

band serves as the loading control. The samples from the amiRNA clones are compared with the flagella-less mutant fla10-2. The wild-type cell was not used

as the control because its cytoskeletal fraction contains a large amount of flagella-derived tubulin. (B) Growth rates of ami1 and ami2. The growth of the amiRNA

clones is similar to that of bld10, a mutant that lacks the centriole. (C–E) Indirect immunofluorescence microscopy of fla10-2, ami1 and ami2 cells. The

cells were double-stained for a-tubulin (green) and the following centriole proteins (magenta): CRC70 (C), SAS-6 (D) and Bld10p (E). (C) The fluorescence for

CRC70 observed at the MTOC in the control cells is greatly reduced or absent in the amiRNA cells without flagella, but it is not reduced in the amiRNA2

cells that retain flagella. (D,E) The fluorescence for SAS-6 and Bld10p at the MTOC is also greatly reduced in amiRNA cells. Scale bars: 10 mm.

Journal of Cell Science 124 (17)2968

Journ

alof

Cell

Scie

nce

of the analyses, ,5% of the total cells were abnormally large in

the cultures of ov1 and ov2.

The localization of the expressed CRC70 varied in both strains

depending on the expression level in the cell. At relatively high

expression levels, CRC70 formed aggregates in the cytoplasm,

whereas at moderate expression levels, it localized only to the

base of the flagella (Fig. 5B). Most of the CRC70 aggregates

were observed in the abnormally large cells. To assess the effect

of CRC70 overexpression on centriole assembly, the localization

of Bld10p and SAS-6 was examined in the transformants. Like

CRC70, these proteins were found to localize to discrete spots in

the cytoplasm in cells overexpressing CRC70 (Fig. 5C). Double

staining of HA-tagged CRC70 and SAS-6 showed that the spots

described above coincided with the CRC70 aggregates (Fig. 5D).

Strikingly, some of the spots radiated microtubules (Fig. 5C,

arrows). These observations suggest that the overexpression of

CRC70 induces ectopic accumulation of centriole proteins in

discrete spots and some of the spots acquire microtubule-

nucleating activities. However, the tendency of CRC70-

overexpressing cells to lose flagella rather than display an

increase in flagellar number suggests that the microtubule-

nucleating spots found in those cells do not have the activity to

initiate flagellar assembly.

CRC70 localizes to the procentriole in NIH3T3 cells

As shown above, CRC70 is partially similar to Cep70, a

centrosomal protein conserved in a wide range of organisms,

including mammals (Fig. 1). The effect of CRC70 on centriole

assembly in mammalian cells was therefore investigated by

expressing the protein in mouse NIH3T3 cells. A plasmid

designed to express CRC70 fused to enhanced green fluorescent

protein (EGFP) was transfected into mouse 3T3 cells. EGFP

fluorescence was observed in transfected cells as a prominent

spot near the nucleus (Fig. 6A). Immunofluorescence microscopy

using antibodies against a- and c-tubulin showed that the

fluorescent spot coincided with the center of the radiating

microtubules, namely the microtubule-organizing center

(MTOC), and that it overlapped with the c-tubulin signal

(Fig. 6A), suggesting that CRC70 is located at the centrosome.

A more detailed localization of CRC70 was carried out using

antibodies against four centriolar proteins: Odf2, which localizes

to the distal end of the mature centriole (Nakagawa et al., 2001;

Ishikawa et al., 2005); C-Nap1 (also known as centrosome-

associated protein CEP250), which localizes to the proximal end

of the mature centriole (Fry et al., 1998); centrin, which localizes

to the centriolar lumen with a bias toward the distal end (Paoletti

et al., 1996); and SAS-6, which localizes to the proximal end of

the procentriole (Strnad et al., 2007). Images of the centrosome in

the CRC70-overexpressing cells double-stained with anti-Odf-2

and anti-C-Nap1 antibodies showed that the CRC70 signal was

located on the side of the mother centriole axis, just like the SAS-

6 localization observed in HeLa cells (Fig. 6B) (Strnad et al.,

2007). Positional relationships between the CRC70 and centrin

signals also supported the similarity of the CRC70 and SAS-6

localization (Fig. 6C). Indeed, the SAS-6 signal almost

completely overlapped with the CRC70 signal (Fig. 6D). These

Fig. 5. Overexpression of CRC70 induces

ectopic accumulation of centriole proteins

in Chlamydomonas cells. (A) Western blot

analysis of cells overexpressing a FLAG-

tagged CRC70 (Ov1) or a HA-tagged CRC70

(Ov2) using the anti-CRC70 antibody. The

CRC70 protein band (arrowhead) is detected

in the two transformants but not in the wild

type. CBB, Coomassie-Brilliant-Blue-stained

gel. SAS-6 is used as a loading control (lower

panels). (B–D) Indirect immunofluorescence

microscopy of CRC70-overexpressing cells.

(B) Ov2 cells were double-stained for HA–

CRC70 (magenta) and a-tubulin (green).

Anti-HA (upper panels) or anti-CRC70

antibodies (lower panels) were used for

CRC70 detection. The upper panels show a

cell with a moderate expression level, and the

lower panels show a cell with a high

expression level. Arrowheads indicate

CRC70-containing aggregates in the

cytoplasm. (C) Ov1 cells double-stained for

centriole proteins (SAS-6 or Bld10p,

magenta) and a-tubulin (green). The centriole

proteins are localized to discrete spots in the

cytoplasm. Some of the fluorescent spots

(arrows) radiate cytoplasmic microtubules but

some do not (arrowheads). (D) Ov2 cells

double-stained for HA-tagged CRC70 and

SAS-6. The two proteins colocalize at the

centriole and in the cytoplasmic aggregates.

Scale bars: 5 mm.

CRC70 is a scaffold protein in centriole assembly 2969

Journ

alof

Cell

Scie

nce

observations suggest that CRC70 localizes to the procentriole in

mouse cells, as well as in Chlamydomonas cells.

CRC70 contains two regions that direct its localization tothe centrosome

To identify the CRC70 sequence that determines the centrosomallocalization of the protein in mouse cells, various CRC70fragments were fused with EGFP at the N-terminus and their

location within the cell was assessed (Fig. 7). Of the six fusionproteins generated (F1–F6) (Fig. 7A), F3 and F6 showed diffuselocalization throughout the cytoplasm, whereas F1, F2, F4 and F5were, at least in part, localized to centrosomes (Fig. 7B). These

observations clearly show that CRC70 has two distinct regionsthat enable centrosomal localization. Interestingly, the sequencealignment shows that one of them, the region F4–F5, corresponds

to a coiled-coil region in zebrafish Cep70 that has been shown tobe crucial for its centrosomal localization (Wilkinson et al., 2009)(Fig. 7).

Overexpression of CRC70 induces assembly of centriole-like structures in mouse cells

To investigate whether CRC70 overexpression in mouse cells

induces the formation of centriole-related structures, as observedin Chlamydomonas cells, the localization of mammalian SAS-6and c-tubulin was examined in NIH3T3 cells expressing EGFP-tagged CRC70 at relatively high expression levels (which

account for ,30% of the transfected cells). In most of thesecells, EGFP fluorescence was observed to form clusters of spots,most of which, but not all, overlapped with the signals of a-

tubulin, c-tubulin, centrin and SAS-6 (Fig. 8A–D). Theseobservations suggest that when CRC70 is overexpressed inmouse cells it forms aggregates by itself, and that some of the

aggregates contain centriolar and centrosomal proteins.

To examine whether these aggregates contain centrioles or

centriole-like structures, the CRC70-overexpressing cells wereobserved by electron microscopy. Strikingly, multiple centriole-like structures were observed in distinctive areas near the nuclei;

the areas contained numerous electron-dense clumps withradiating microtubules and densely packed microtubules(Fig. 8E–J). Some of the structures had the characteristic

cylindrical structure of the centriole with a diameter of,200 nm and a length of ,400 nm, whereas others had

defective structures such as open cylinders, which are similarto the regenerating centrioles observed after laser ablation inHeLa cells (Khodjakov et al., 2002). Although we cannot rule out

the possibility that some of these structures are endogenouscentrioles formed during multiple rounds of cell cycles without

cytokinesis, the CRC70 aggregates might well have inducedectopic formation of centriole-like structures and centrosomes.

DiscussionThe present study identified CRC70 as a Chlamydomonas proteinwith a partial sequence similarity to the mammalian centrosomal

protein Cep70. CRC70 probably appears at an early stage ofcentriole assembly and is absent from mature centrioles. RNAi-

mediated CRC70 silencing resulted in cells that lacked flagellaand grew slowly, which are phenotypes characteristic ofcentriole-deficient mutants. The centriole components SAS-6

and Bld10p were greatly reduced in these cells, indicating thatCRC70 is crucial for the recruitment of centriole proteins to the

assembly site. This concept is supported by the finding that theoverexpression of CRC70 causes the ectopic accumulation ofcentriole proteins in both Chlamydomonas and mammalian cells

and the formation of centriole-like structures in mammalian cells.Hence, CRC70 is likely to function as a scaffold for centriole

assembly, a function that might be shared with other Cep70family proteins.

CRC70 is a possible component of a pre-cartwheel structure

Immunofluorescence microscopy showed that CRC70 localizes

to MTOC-like spots in the mutants bld2, bld10 and bld12, whichare deficient at specific stages in the centriole assembly process.

The mutant bld2 has centrioles with singlet microtubules(Goodenough and St. Clair, 1975), bld10 has centriole-associated structures with no centriolar microtubules (Matsuura

et al., 2004) and bld12 has fragmented centrioles or centrioleswith variable numbers of triplet microtubules (Nakazawa et al.,

2007). In bld10, the mutant with the most severe phenotype, theassembly process probably stops at the stage of cartwheel

Fig. 6. CRC70 expressed in NIH 3T3 cells

localizes to the proximal end of the centrioles.

(A) Cells expressing EGFP–CRC70 (green) were

stained for a-tubulin (upper panels, magenta) or

c-tubulin (lower panels, magenta). CRC70 is

located at the MTOC of the cell and colocalized

with c-tubulin. (B) Images of the centrosomes in

cells expressing EGFP–CRC70 (green) stained

for Odf2 (magenta) and C-Nap1 (Blue). The left

panel is a magnified image of the centrosome

area (arrow) in the middle panel. (C,D) Images of

the centrosomes in cells expressing EGFP–

CRC70 (green) stained for centrin (magenta, C)

or SAS-6 (magenta, D). When the orientations of

centrioles and procentrioles can be detected,

those are indicated by rectangles. Scale bars:

10 mm (A); 1 mm (B–D).

Journal of Cell Science 124 (17)2970

Journ

alof

Cell

Scie

nce

formation because the product of the BLD10 gene is a component

of the cartwheel spoke tip to which a microtubule attaches

(Hiraki et al., 2007). Thus, the detection of a single CRC70 dot in

bld10 cells suggests that CRC70 localizes to the presumptive

centriole assembly site during or before cartwheel formation

(Fig. 3).

In these centriole-deficient mutants, however, the number of

the cells having the CRC70 dot was much lower than that in wild

type. This is probably because efficient formation of centriole

assembly sites requires pre-existing centrioles or centriole-like

structures, as shown by previous studies (Marshall et al., 2001;

Khodjakov et al., 2002). In support of this idea, the percentage of

CRC70-positive cells appears to correlate with the degree of the

centriole defects in the mutants: only ,14% of the bld10 cells

were positive for CRC70, whereas ,58% and ,43% were

positive for CRC70 in bld2 and bld12, which retain some

centrioles or centriole-like structures. Thus, these observations

suggest that the CRC70 fluorescent dot represents the centriole

assembly site.

In the Chlamydomonas centriole assembly process, an

amorphous ring structure is formed before the appearance of

the cartwheel (O’Toole et al., 2003). The structure, which is

difficult to observe by conventional electron microscopy but is

detectable by cryoelectron microscopy (Dutcher, 2007), is seen

around the proximal end of both procentrioles and mature

centrioles. Despite its apparent importance in the centriole

assembly process, the molecular components of the amorphous

ring have not been identified. CRC70 could be one of these

components because it is localized to the outer side of the

microtubule wall of the procentriole before cartwheel formation,

in accordance with the timing and location of the amorphous ring

(Figs 2 and 3) (O’Toole et al., 2003). However, whereas the ring

structure remains attached to mature centrioles (O’Toole et al.,

2003), CRC70 is absent in mature centrioles (Fig. 2;

supplementary material Fig. S3). Hence, CRC70 could be a

protein that transiently localizes to the amorphous ring during the

initial assembly process.

Scaffolding function of CRC70

RNAi-depletion of CRC70 impaired the localization of SAS-6

and Bld10p to the centriole and produced aflagellate cells that

grew as slowly as the centriole-deficient mutant bld10 (Fig. 4B).

Because Bld10p is required for the formation of centriolar

microtubules (Matsuura et al., 2004), these results suggest that

flagella-less cells depleted of CRC70 do not form centrioles. In

contrast to the RNAi experiments, overexpression of CRC70 in

mouse cells resulted in the assembly of multiple centriole-like

structures in the cytoplasm (Fig. 8). Overexpression of CRC70 in

Chlamydomonas cells also caused centriole proteins to assemble

into discrete spots in the cytoplasm, some of which appeared to

function as cytoplasmic microtubule foci (Fig. 5). Although these

spots have not yet been observed by electron microscopy, it is

possible that the CRC70 aggregates in the Chlamydomonas

cytoplasm trigger the assembly of ectopic centrioles.

CRC70 is a member of the Cep70 protein family

Cep70 is a centrosomal protein that was identified in a proteomic

analysis of human centrosomes (Andersen et al., 2003). Although

many mammalian homologs of this protein are found in DNA

sequence databases, its sequence conservation in organisms other

Fig. 7. Centrosomal localization of CRC70 depends on one of

two distinct regions in the protein. (A) Schematic diagrams of

truncated CRC70 products and their localization to the centrosome

(++, strong centrosomal localization; +, some centrosomal

localization; –, no centrosomal localization). A diagram of zebrafish

Cep70 with the portion required for centrosomal localization

(Wilkinson et al., 2009), indicated by a double arrow, is also shown.

aa, amino acids. (B) NIH3T3 cells expressing the EGFP-tagged

CRC70 fragments (green). Cells were stained with the anti-a-tubulin

antibody (magenta). Centrosomal localizations of the EGFP-tagged

fragments are indicated by arrowheads. Scale bar: 10 mm.

CRC70 is a scaffold protein in centriole assembly 2971

Journ

alof

Cell

Scie

nce

than vertebrates is not known. BLAST searches with the CRC70

sequence identified a highly conserved short sequence motif

shared by Cep70-homologous proteins and showed that proteins

containing this motif are present in various eukaryotic organisms,

including vertebrates, bryophytes and algae, but not in organisms

that do not have centrioles, such as higher plants and fungi, or

organisms that have non-canonical centrioles such as C. elegans

and Drosophila (Fig. 1). We suggest therefore that the group

formed by these proteins should be called the Cep70 family. The

structural features shared by these proteins, namely a large

coiled-coil domain in the N-terminal region and the Cep70 motifin the C-terminal region, imply that these proteins originate from

a common ancestor and retain common functions. In fact, when

expressed in mouse cells, partial constructs of CRC70 thatinclude a region homologous to the N-terminal sequence of

Cep70 also localize to the procentriole.

We note, however, that some Cep70 family members have

structural domains shared by only a limited group of organisms.For example, a tetratricopeptide repeat domain is found only in

mammals (see the GeneCards database, http://www.genecards.org/cgi-bin/carddisp.pl?gene5CEP70) (Rebhan et al., 1998), a

histone-deacetylase-interacting domain only in zebrafish(Wilkinson et al., 2009) and an additional long coiled-coil

domain only in algae (the present study). These structuraldeviations suggest that Cep70 family proteins have some species-

specific functions in addition to common functions in the

centriole or centrosome. The long coiled-coil domain of thealgal protein is particularly interesting because it contains a

region that functions as a second locus for centriole localization(Fig. 7). The CRC70-mediated induction of centriole-like

structures in mammalian cells might depend on this region, asoverexpression of Cep70 does not induce ectopic centriole

formation (see below). However, the two centriole-localizingdomains could function synergistically to produce centriolar

precursors. The precise role of the two domains in CRC70

remains to be studied. Expression of these domains, singly ortogether, in Drosophila cells, which have no Cep70-homologous

proteins, might also provide some clues as to their otherfunctions.

Functions of Cep70 family proteins

Despite the close relationship between CRC70 and vertebrate

Cep70, previous studies have not suggested the involvement ofCep70 in centriole formation in vertebrates. Genome-wide

analyses using RNAi showed that depletion of Cep70 does notdisturb mitosis in human cells (Neumann et al., 2010).

Overexpression of Cep70 in human cells results in the

formation of cytoplasmic aggregates but these aggregates donot colocalize with c-tubulin or cause overduplication of

centrosomes (Nigg, 2004). Depletion of Cep70 in zebrafishusing antisense morpholino oligonucleotides impairs ciliogenesis

but does not prevent centriole formation (Wilkinson et al., 2009).Thus, Cep70 might not be involved in centriole formation in

these cells. However, our finding that CRC70 localizes to theprocentriole in Chlamydomonas, together with the observation

that Cep70 localizes to the centrosome in mammals (Nigg, 2004)

Fig. 8. Overexpression of CRC70 induces formation of centriole-like

structures in NIH 3T3 cells. (A–D) Ectopic accumulation of centrosomal

and centriolar proteins in NIH 3T3 cells expressing EGFP–CRC70 (green).

Cells were stained for a-tubulin (A), c-tubulin (B), centrin (C) or SAS-6 (D).

The a-tubulin, c-tubulin, centrin and SAS-6 signals (magenta) overlap with

the CRC70 signals. The CRC70 signals from the intrinsic centrosomes are not

detected because of their low intensity compared with those from the

cytoplasmic aggregates. (E–J) Electron microscopy of cells overexpressing

CRC70. (E) A region near the nucleus showing numerous densely packed

microtubules. Multiple centrioles (arrows) and electron-dense clumps

(arrowheads) are seen. Magnified images of the boxed regions showing a

centriole-like structure (F) and an electron-dense clump (G) are also shown.

(H–J) Other examples of the centriole-like structures. Centrioles (arrows) are

accompanied by electron-dense clumps (arrowheads). Scale bars: 10 mm

(A–D); 1 mm (E); 200 nm (F–J).

Journal of Cell Science 124 (17)2972

Journ

alof

Cell

Scie

nce

and to the centrosome and ciliary base in zebrafish (Wilkinsonet al., 2009), suggests that these proteins perform some unknown

common functions related to the centriole. Detailed localizationof Cep70 in vertebrate cells will be necessary to understand thefunction of these proteins and the significance of the highly

conserved Cep70 motif.

Materials and MethodsStrains and culture conditions

Chlamydomonas reinhardtii CC-124 [wild type; mating type (mt–)], CC-125 (wildtype; mt+), CC-1926 (uni1-1; mt+), CC-478 (bld2-1; mt+), CC-503 (cw92; mt+),and CC-2290 (S1-D2; mt–) were obtained from the Chlamydomonas GeneticsCenter. A null allele of fla10 (fla10-2), and the bld10-1 and bld12-1 mutants wereisolated in our laboratory (Matsuura et al., 2002; Matsuura et al., 2004; Nakazawaet al., 2007). For assessment of the cellular levels of CRC70 mRNA by RT-PCR,cells were grown in liquid M medium (Harris, 1989) at 24 C with aeration on a 12-hour-light–12-hour-dark cycle. For immunofluorescence and immunoelectronmicroscopy, cells were grown in Tris acetate phosphate (TAP) medium (Gormanand Levine, 1965) under constant illumination.

Mouse NIH3T3 cells were cultured in Dulbecco’s modified Eagle’s medium(DMEM) supplemented with 10% fetal bovine serum at 37 C. For microscopicobservation, the cells were transferred to eight-well glass slides and cultured in amoisture chamber for 24 hours at 37 C.

Cloning and expression of CRC70 cDNA

Total RNA was isolated from wild-type cells using the TRIzol reagent (Invitrogen)and was used for cDNA synthesis after treatment with DNase I (Invitrogen). ThecDNA was synthesized using Superscript III (Invitrogen) and Tth polymerase(Promega) with oligo(dT) primers (Myers and Gelfand, 1991). The cDNAfragments of 2.4-kb and 3.2-kb that cover the 59- and 39-halves of CRC70 cDNA,with an overlap of ,130 bp containing a unique XhoI site, were amplified by PCRusing PrimeSTAR GXL polymerase (TaKaRa). The primers used were as follows:59-GGAATTCCATATGCGCGCTGCACGCGCAGC-39 and 59-GCTCGCCGCCTCGGCCTCGCTCTGC-39; and 59-AGCGGCTCGTGCAGGCAGAAGGCCTGC-39 and 59-CGCGGATCCGTACAGCGCCTGCTGTGCCG-39. The underlinedsequences in the primers are NdeI and BamHI sites. The 2.4-kb and 3.2-kbfragments amplified were digested with NdeI and XhoI, and XhoI and BamHI,and cloned into pBluescript KS+ (Stratagene) and pCold I (TaKaRa). Afterdetermining the sequences of the inserts, the 2.4-kb fragment was cut out andligated to the 3.2-kb fragment in pCold I digested with NdeI and XhoI. Thecombined insert was used as the full-length cDNA of CRC70.

For expression of EGFP-tagged CRC70 in mouse cells, the whole insert inpCold I was transferred into the SalI-BamHI site of the expression vector pEGFP-C1 (Clontech). For expression of CRC70 fragments, the corresponding cDNAsequences were amplified by PCR using the following primer sets: 59-CGGAATTCTATGCGCGCTGCACGCGC-39 and 59-CGGGATCCTGGCTGCCAGGTCGTCA-39; 59-CGGAATTCTGCCACCATGCGTGACGA-39 and 59-CGGGATCCCTCGAAGTCGGAGGTCTTCG-39; 59-CGGAATTCTGCTGCGAAGACCTCCGACTT-39 and 59-CGGGATCCCGATCTCGGCGGTCTTGCTC-39; 59-CGGAATTCTAGCAAGACCGCCGAGA-39 and 59-CGGGATCCTGTGCTCCAGCTGTGAGT-39; 59-CGGAATTCTGTGCGGCAGAAGTACTCAC-39

and 59-CGGGATCCCGACAGGTCGAGCTCTATGC-39; and 59-CGGAATTCTGTTTGCATAGAGCTCGACC-39 and 59-CGGGATCCGTACAGCGCCTGCTGTG-39.The amplified fragments were digested at BamHI and EcoRI sites(underlined), and cloned into pEGFP-C1.

Antibodies

The anti-CRC70 antibody was produced as follows. A cDNA fragment encod-ing residues 16–360 was amplified by PCR using the primers 59-CGCGGATCCCAGCCCTCTGTCAACACCGCA-39 and 59-CGGAATTCCTCGGAGTCTGCGCGCCGCTT-39. The fragment was digested at the BamHI andEcoRI sites in the primer sequences (underlined), and cloned into the bacterialexpression vector pGEX-2T (GE Healthcare). The expressed GST-tagged peptidewas loaded onto a glutathione–Sepharose 4B column (GE Healthcare) and elutedby applying thrombin to the column, as per the manufacturer’s instructions. Elutedproteins were separated by SDS-PAGE, and a band of the expected size was cutout and used for immunizing rabbits. The antiserum was affinity-purified using theantigen protein blotted onto a PVDF membrane strip (Olmsted, 1981). Forelimination of non-specific reactions to a protein of ,170 kDa inChlamydomonas, the antibody was absorbed with a membrane strip blotted withthe cell body extract. The pre-immune serum used in control experiments wastreated in the same manner.

Rabbit polyclonal antibodies against Chlamydomonas Bld10p and SAS-6 havebeen described previously (Matsuura et al., 2004; Nakazawa et al., 2007). Rabbitpolyclonal antibodies against human SAS-6 and Odf2 were gifts from PierreGonczy (Swiss Institute for Experimental Cancer Research, Lausanne,

Switzerland) and Sachiko Tsukita (Osaka University, Osaka, Japan),respectively. Monoclonal antibodies against a-tubulin (B-5-1-2), acetylated a-tubulin (6-11B-1) and polyglutamylated tubulin (B3), and polyclonal antibodiesagainst c-tubulin (T3559) and centrin (C7736) were purchased from Sigma-Aldrich. Monoclonal antibody against C-Nap1 (Clone 42) was purchased from BDBiosciences. The secondary antibodies used were goat anti-mouse-IgG conjugatedto fluorescein isothiocyanate (FITC) (Sigma-Aldrich), goat anti-rabbit-IgGconjugated to rhodamine (Sigma-Aldrich), and goat anti-rabbit-IgG conjugatedto 10-nm gold particles (British Biocell International).

Western blot analysis

For immunodetection of CRC70 in Chlamydomonas cell extracts, a detergent-insoluble fraction was prepared as follows. Wild-type and bld10 cells were treatedwith autolysin to remove the cell wall, and washed with MT buffer [30 mM Tris-acetate pH 7.3, 5 mM MgSO4, 5 mM ethylene glycol tetraacetic acid (EGTA),25 mM KCl, 0.01% aprotinin, 0.4 mM Pefabloc (Roche) and 5 mg/ml leupeptin)]containing 25% sucrose. The cells were washed again with MT buffer alone forhypotonic exposure, and then lysed with 1% NP-40 in the buffer. The lysates werecentrifuged at 13,000 g for 20 minutes at 4 C. The pellets were suspended withMT buffer, transferred onto Immobilon-P membranes (Millipore), and analyzed bySDS-PAGE. Immunoreactive proteins were detected using ECL Advance reagents(GE Healthcare) and the Light-Capture System AE-6972 (ATTO).

Immunofluorescence microscopy

Chlamydomonas cells were fixed and processed for antibody staining as describedby Holmes and Dutcher (Holmes and Dutcher, 1989). NFAp was prepared andobserved by immunofluorescence microscopy according to the method of Wrightet al. (Wright et al., 1985). Images were recorded using an Axioplan fluorescencemicroscope (Carl Zeiss MicroImaging) with a 636 1.4 NA plan-APOCHROMATobjective and a CoolSNAP CCD camera (Roper Scientific).

Mouse cells were washed with PBS and fixed with –20 C methanol for10 minutes or with 4% formaldehyde in PBS for 20 minutes. The fixed cells werewashed with PBS for 5 minutes and permeabilized by treating three times with0.1% Triton X-100 in PBS for 5 minutes each time. After three 5 minute washeswith PBS, cells were incubated with blocking buffer at 37 C for 30 minutes. Theblocking buffers used were 1% BSA in PBS for anti-human SAS-6, anti-C-Nap1,and anti-Odf2 antibodies, 10% goat serum in PBS for anti-centrin antibody, andthe blocking solution used by Holmes and Dutcher (Holmes and Dutcher, 1989)(10 mM sodium phosphate pH 7.2, 5% goat serum, 5% glycerol, 1% cold fishgelatin and 0.004% sodium azide) for anti-a-tubulin antibody B-5-1-2 or anti-c-tubulin antibody. Images were recorded using an LSM710 confocal microscope(Carl Zeiss MicroImaging).

Quantitative RT-PCR

The relative expression level of the CRC70 mRNA was quantified by qRT-PCRanalyses (Higuchi et al., 1992; Higuchi et al., 1993). Approximately 26107 cellswere collected, flash-frozen in liquid nitrogen and stored at –80 C. RNA wasextracted from the cells and reverse transcribed (Myers and Gelfand, 1991). Thesynthesized cDNA was treated with 0.02 units per ml of RNase H (Invitrogen) at37 C for 20 minutes and analyzed by real-time PCR using a LightCycler (Roche)and TaKaRa R-PCR kit version 2.1 (TaKaRa). The primer sets used for amplifyingcDNA fragments of CRC70 and elongation factor 1 alpha (EF1a; a controlgene) were: 59-CATGCGCCTGTTCGACTGCC-39 and 59-TCCGCACCGTGAAGACAGCC-39; and 59-CGCGTGGGCTG-GAAGGACGA-39 and 59-CTT-GCCGGTGCAGGGGTTGG-39. The sets of fluorescence-labeled oligonucleotideDNA probes used for the detection of the PCR products were 59-GCGATCGGC-TGCTGCCAGACTACATCAAGG-39 –FITC and LCRed640–59-CAACCATGC-GTCCGCCGCCAGCC-39; and 59-GAAGATCCAGGTGCACACCCTGCTGGA-CGC-39–FITC and LCRed640–59-CTGAACAGCTTCGTGGTCGTGCCCGAGC-39. The program used for amplification was as follows: 95 C for 30 seconds, 50cycles of 95 C for 5 seconds, 64 C for 15 seconds, and 70 C for 20 seconds. Eachreal-time PCR assay was performed in duplicate, and the quantificationexperiments were repeated three times using RNA isolated from independentcultures.

Gene silencing using artificial micro RNA (amiRNA)

Gene silencing using amiRNA was performed as described previously (Molnaret al., 2009). pChlamiRNA3int, an expression vector for the miRNA precursor,was obtained from the Chlamydomonas Resource Center (University of Minnesota,St. Paul, MN). Two amiRNA sequences, each consisting of ,90 bases, weredesigned using the web MicroRNA designer platform (WMD3; http://wmd3.weigelworld.org/cgi-bin/webapp.cgi). One sequence was designed to target thesequence 59-AAGGGCGAAGAGATTCGCTC-39 in the coding region of theCRC70 cDNA, and the other was designed to target the sequence 59-TTGGCTTGTTAGCACAGCTA-39 in the 39-untranslated region (UTR). Afterannealing with the complementary oligonucleotides, the amiRNAs were digestedwith SpeI and ligated into SpeI-digested pChlamiRNA3int. Wild-type cells were

CRC70 is a scaffold protein in centriole assembly 2973

Journ

alof

Cell

Scie

nce

transformed with the constructed plasmids using electroporation (Shimogawaraet al., 1998). The transformed cells were cultivated on TAP agar plates thatcontained 10 mg/ml paromomycin. Next, ,500 colonies of drug-resistanttransformants were transferred into liquid TAP medium in 96-well plates andobserved under a microscope. Two clones that contained no flagellated cells weresaved and used as candidates for knockdown strains. One clone, ami1, wasobtained by targeting the coding region of the CRC70 gene, whereas the other,ami2, was obtained by targeting the 39-UTR fragment.

Overexpression of CRC70Two plasmid vectors were used for the production of Chlamydomonas strains thatoverexpress CRC70 tagged with HA3 or the FLAG sequence. For expression of theHA-tagged protein, the plasmid pHY1 carrying the PsaD promoter for cDNAexpression and the aphVIII gene for paromomycin-resistance were used(Nakazawa et al., 2007). For expression of the FLAG-tagged proteins, anotherplasmid, pHY2, was constructed from pHY1 by replacing the HA3 sequence withthe FLAG sequence. The full-length CRC70 cDNA was ligated into the NdeI andEcoRI sites of pHY1 and pHY2, and used for the transformation of wild-type cells.The transformed cells were cultivated on agar plates that contained 10 mg/mlparomomycin; ,1000 colonies of drug-resistant transformants were saved andanalyzed by western blotting using the anti-CRC70 antibody.

Transfection of NIH3T3 cells

NIH3T3 cells on glass slides or plastic dishes were transfected with plasmid DNAusing Lipofectamine 2000 (Invitrogen) according to the manufacturer’sinstructions. Immunofluorescence microscopy was performed using the cellsimmobilized on glass slides 24 hours after transfection. Electron microscopy wasperformed using the cells on plastic dishes 72 hours after transfection.

Electron microscopyImmunoelectron microscopy was performed mostly as described previously(Silflow et al., 2001). All procedures were carried out at 4 C unless statedotherwise. For NFAp observations, samples were treated with the primary antibodyin HMT-BSA solution (30 mM Hepes pH 7.0, 5 mM MgSO4, 5 mM EGTA,25 mM KCl, 1% BSA, 0.01% aprotinin, 0.4 mM Pefabloc (Roche) and 5 mg/mlleupeptin) for 90 minutes. Following three washes with HMT-BSA, the sampleswere treated for 90 minutes with goat anti-rabbit-IgG antibody conjugated to10-nm gold particles diluted 1:20 in HMT-BSA. The NFAp preparations werewashed once with HMT-BSA and fixed, first with 2% paraformaldehyde and 2.5%glutaraldehyde in 0.1 M sodium-phosphate buffer (pH 7.4) for 1 hour, and thenwith 1% glutaraldehyde in 0.1 M sodium-phosphate buffer (pH 7.4) overnight. Forobservation of cells on dishes, the samples were washed with PBS and fixed with2.5% glutaraldehyde in PBS for 1 hour. Both the NFAp preparations and themouse cells were post-fixed with 1% OsO4 for 1 hour on ice, and stained en blocwith 1% uranyl acetate for 30 minutes on ice. The samples were then dehydratedand embedded in EPON812 (Shell Chemical Company). Ultrathin sections werepost-stained with 7% aqueous uranyl acetate for 20 minutes and 0.8% lead citratefor 2 minutes.

We thank Pierre Gonczy (Swiss Institute for Experimental CancerResearch) for providing the anti-hsSAS-6 antibody, Sachiko Tsukita(Osaka University) for the anti-Odf2 antibody and Minkung Park(University of Tokyo) for the NIH 3T3 cells. We also thank TakeoKubo and Hideaki Takeuchi, and Hiroyuki Takeda (University ofTokyo) for allowing us to use their equipment. This study has beensupported by Grants-in-Aid for Scientific Research from the Ministryof Education, Culture, Science and Technology of Japan (21370088).

Supplementary material available online at

http://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.084715/-/DC1

ReferencesAndersen, J. S., Wilkinson, C. J., Mayor, T., Mortensen, P., Nigg, E. A. and Mann,

M. (2003). Proteomic characterization of the human centrosome by proteincorrelation profiling. Nature 426, 570-574.

Azimzadeh, J. and Bornens, M. (2007). Structure and duplication of the centrosome.J. Cell Sci. 120, 2139-2142.

Berger, B., Wilson, D. B., Wolf, E., Tonchev, T., Milla, M. and Kim, P. S. (1995).Predicting coiled coils by use of pairwise residue correlations. Proc. Natl. Acad. Sci.

USA 92, 8259-8263.

Dammermann, A., Muller-Reichert, T., Pelletier, L., Habermann, B., Desai, A. andOegema, K. (2004). Centriole assembly requires both centriolar and pericentriolarmaterial proteins. Dev. Cell 7, 815-829.

Delattre, M., Leidel, S., Wani, K., Baumer, K., Bamat, J., Schnabel, H., Feichtinger,

R., Schnabel, R. and Gonczy, P. (2004). Centriolar SAS-5 is required for centrosomeduplication in C. elegans. Nat. Cell Biol. 6, 656-664.

Dippell, R. V. (1968). The development of basal bodies in paramecium. Proc. Natl.

Acad. Sci. USA 61, 461-468.

Dutcher, S. K. (2007). Finding treasures in frozen cells: new centriole intermediates.BioEssays 29, 630-634.

Dutcher, S. K., Morrissette, N. S., Preble, A. M., Rackley, C. and Stanga, J. (2002).e-tubulin is an essential component of the centriole. Mol. Biol. Cell 13, 3859-3869.

Ehler, L. L., Holmes, J. A. and Dutcher, S. K. (1995). Loss of spatial control of themitotic spindle apparatus in a Chlamydomonas reinhardtii mutant strain lacking basalbodies. Genetics 141, 945-960.

Felsenstein, J. (1985). Confidence limits on phylogenies: an approach using thebootstrap. Evolution 39, 783-791.

Fry, A. M., Mayor, T., Meraldi, P., Stierhof, Y. D., Tanaka, K. and Nigg, E. A.

(1998). C-Nap1, a novel centrosomal coiled-coil protein and candidate substrate ofthe cell cycle-regulated protein kinase Nek2. J. Cell. Biol. 141, 1563-1574.

Gaffal, K. P. (1988). The basal body-root complex of Chlamydomonas reinhardtii

during mitosis. Protoplasma 143, 118-129.

Geimer, S. and Melkonian, M. (2004). The ultrastructure of the Chlamydomonas

reinhardtii basal apparatus: identification of an early marker of radial asymmetryinherent in the basal body. J. Cell Sci. 117, 2663-2674.

Goodenough, U. W. and St. Clair, H. S. (1975). BALD-2: a mutation affecting theformation of doublet and triplet sets of microtubules in Chlamydomonas reinhardtii.J. Cell Biol. 66, 480-491.

Gorman, D. S. and Levine, R. P. (1965). Cytochrome f and plastocyanin: theirsequence in the photosynthetic electron transport chain of Chlamydomonas

reinhardtii. Proc. Natl. Acad. Sci. USA 54, 1665-1669.

Harris, E. H. (1989). The Chlamydomonas Sourcebook. San Diego: Academic Press,780.

Higuchi, R., Dollinger, G., Walsh, P. S. and Griffith, R. (1992). Simultaneousamplification and detection of specific DNA sequences. Biotechnology 10, 413-417.

Higuchi, R., Fockler, C., Dollinger, G. and Watson, R. (1993). Kinetic PCR analysis:real-time monitoring of DNA amplification reactions. Biotechnology 11, 1026-1030.

Hiraki, M., Nakazawa, Y., Kamiya, R. and Hirono, M. (2007). Bld10p constitutes theCartwheel-spoke tip and stabilizes the 9-fold symmetry of the centriole. Curr. Biol.

17, 1778-1783.

Holmes, J. A. and Dutcher, S. K. (1989). Cellular asymmetry in Chlamydomonas

reinhardtii. J. Cell Sci. 94, 273-285.

Huang, B., Ramanis, Z., Dutcher, S. K. and Luck, D. J. (1982). Uniflagellar mutantsof Chlamydomonas: evidence for the role of basal bodies in transmission of positionalinformation. Cell 29, 745-753.

Ishikawa, H., Kubo, A., Tsukita, S. and Tsukita, S. (2005). Odf2-deficient mothercentrioles lack distal/subdistal appendages and the ability to generate primary cilia.Nat. Cell Biol. 7, 517-524.

Kathir, P., LaVoie, M., Brazelton, W. J., Haas, N. A., Lefebvre, P. A. and Silflow, C.D. (2003). Molecular map of the Chlamydomonas reinhardtii nuclear genome.Eukaryot. Cell 2, 362-379.

Keller, L. C., Romijn, E. P., Zamora, I., Yates, J. R., 3rd and Marshall, W. F.

(2005). Proteomic analysis of isolated Chlamydomonas centrioles reveals orthologs ofciliary-disease genes. Curr. Biol. 15, 1090-1098.

Kemp, C. A., Kopish, K. R., Zipperlen, P., Ahringer, J. and O’Connell, K. F. (2004).Centrosome maturation and duplication in C. elegans require the coiled-coil proteinSPD-2. Dev. Cell 6, 511-523.

Khodjakov, A., Rieder, C. L., Sluder, G., Cassels, G., Sibon, O. and Wang, C. L.

(2002). De novo formation of centrosomes in vertebrate cells arrested during S phase.J. Cell Biol. 158, 1171-1181.

Kilburn, C. L., Pearson, C. G., Romijn, E. P., Meehl, J. B., Giddings, T. H., Jr,Culver, B. P., Yates, J. R., 3rd and Winey, M. (2007). New Tetrahymena basalbody protein components identify basal body domain structure. J. Cell Biol. 178, 905-912.

Koblenz, B., Schoppmeier, J., Grunow, A. and Lechtreck, K. F. (2003). Centrindeficiency in Chlamydomonas causes defects in basal body replication, segregationand maturation. J. Cell Sci. 116, 2635-2646.

Lechtreck, K. F. and Geimer, S. (2000). Distribution of polyglutamylated tubulin inthe flagellar apparatus of green flagellates. Cell Motil. Cytoskeleton 47, 219-235.

Leidel, S., Delattre, M., Cerutti, L., Baumer, K. and Gonczy, P. (2005). SAS-6defines a protein family required for centrosome duplication in C. elegans and inhuman cells. Nat. Cell Biol. 7, 115-125.

Li, J. B., Gerdes, J. M., Haycraft, C. J., Fan, Y., Teslovich, T. M., May-Simera, H.,Li, H., Blacque, O. E., Li, L., Leitch, C. C. et al. (2004). Comparative genomicsidentifies a flagellar and basal body proteome that includes the BBS5 human diseasegene. Cell 117, 541-552.

Marshall, W. F., Vucica, Y. and Rosenbaum, J. L. (2001). Kinetics and regulation ofde novo centriole assembly. Implications for the mechanism of centriole duplication.Curr. Biol. 11, 308-317.

Matsuura, K., Lefebvre, P. A., Kamiya, R. and Hirono, M. (2002). Kinesin-II is notessential for mitosis and cell growth in Chlamydomonas. Cell Motil. Cytoskeleton 52,195-201.

Matsuura, K., Lefebvre, P. A., Kamiya, R. and Hirono, M. (2004). Bld10p, a novelprotein essential for basal body assembly in Chlamydomonas: localization to thecartwheel, the first ninefold symmetrical structure appearing during assembly. J. Cell

Biol. 165, 663-671.

Molnar, A., Bassett, A., Thuenemann, E., Schwach, F., Karkare, S., Ossowski, S.,

Weigel, D. and Baulcombe, D. (2009). Highly specific gene silencing by artificialmicroRNAs in the unicellular alga Chlamydomonas reinhardtii. Plant J. 58, 165-174.

Journal of Cell Science 124 (17)2974

Journ

alof

Cell

Scie

nce

Myers, T. W. and Gelfand, D. H. (1991). Reverse transcription and DNA amplificationby a Thermus thermophilus DNA polymerase. Biochemistry 30, 7661-7666.

Nakagawa, Y., Yamane, Y., Okanoue, T., Tsukita, S. and Tsukita, S. (2001). Outerdense fiber 2 is a widespread centrosome scaffold component preferentiallyassociated with mother centrioles: its identification from isolated centrosomes. Mol.

Biol. Cell 12, 1687-1697.Nakazawa, Y., Hiraki, M., Kamiya, R. and Hirono, M. (2007). SAS-6 is a cartwheel

protein that establishes the 9-fold symmetry of the centriole. Curr. Biol. 17, 2169-2174.Neumann, B., Walter, T., Heriche, J. K., Bulkescher, J., Erfle, H., Conrad, C.,

Rogers, P., Poser, I., Held, M., Liebel, U. et al. (2010). Phenotypic profiling of thehuman genome by time-lapse microscopy reveals cell division genes. Nature 464,721-727.

Neupert, J., Karcher, D. and Bock, R. (2009). Generation of Chlamydomonas strainsthat efficiently express nuclear transgenes. Plant J. 57, 1140-1150.

Nigg, E. A. (2004). A proteomic approach to the inventory of the human centrosome. InCentrosomes in Development and Disease (ed. E. Nigg), p. 125. Weinheim: Wiley-VCH.

O’Connell, K. F., Caron, C., Kopish, K. R., Hurd, D. D., Kemphues, K. J., Li, Y.

and White, J. G. (2001). The C. elegans zyg-1 gene encodes a regulator ofcentrosome duplication with distinct maternal and paternal roles in the embryo. Cell

105, 547-558.O’Toole, E. T., Giddings, T. H., McIntosh, J. R. and Dutcher, S. K. (2003). Three-

dimensional organization of basal bodies from wild-type and d-tubulin deletionstrains of Chlamydomonas reinhardtii. Mol. Biol. Cell 14, 2999-3012.

Olmsted, J. B. (1981). Affinity purification of antibodies from diazotized paper bots ofheterogeneous protein samples. J. Biol. Chem. 256, 11955-11957.

Paoletti, A., Moudjou, M., Paintrand, M., Salisbury, J. L. and Bornens, M. (1996).Most of centrin in animal cells is not centrosome-associated and centrosomal centrinis confined to the distal lumen of centrioles. J. Cell Sci. 109, 3089-3102.

Pelletier, L., Ozlu, N., Hannak, E., Cowan, C., Habermann, B., Ruer, M., Muller-

Reichert, T. and Hyman, A. A. (2004). The Caenorhabditis elegans centrosomalprotein SPD-2 is required for both pericentriolar material recruitment and centrioleduplication. Curr. Biol. 14, 863-873.

Perriere, G. and Gouy, M. (1996). WWW-Query: an on-line retrieval system for

biological sequence banks. Biochimie 78, 364-369.

Piasecki, B. P., LaVoie, M., Tam, L. W., Lefebvre, P. A. and Silflow, C. D. (2008).

The Uni2 phosphoprotein is a cell cycle regulated component of the basal

body maturation pathway in Chlamydomonas reinhardtii. Mol. Biol. Cell 19, 262-273.

Rebhan, M., Chalifa-Caspi, V., Prilusky, J. and Lancet, D. (1998). GeneCards: a

novel functional genomics compendium with automated data mining and query

reformulation support. Bioinformatics 14, 656-664.

Schroda, M. (2006). RNA silencing in Chlamydomonas: mechanisms and tools. Curr.

Genet. 49, 69-84.

Shimogawara, K., Fujiwara, S., Grossman, A. and Usuda, H. (1998). High-efficiency

transformation of Chlamydomonas reinhardtii by electroporation. Genetics 148,

1821-1828.

Silflow, C. D., LaVoie, M., Tam, L. W., Tousey, S., Sanders, M., Wu, W.,

Borodovsky, M. and Lefebvre, P. A. (2001). The Vfl1 protein in Chlamydomonas

localizes in a rotationally asymmetric pattern at the distal ends of the basal bodies.

J. Cell Biol. 153, 63-74.

Spudich, J. L. and Sager, R. (1980). Regulation of the Chlamydomonas cell cycle by

light and dark. J. Cell Biol. 85, 136-145.

Strnad, P. and Gonczy, P. (2008). Mechanisms of procentriole formation. Trends Cell

Biol. 18, 389-396.

Strnad, P., Leidel, S., Vinogradova, T., Euteneuer, U., Khodjakov, A. and Gonczy,

P. (2007). Regulated HsSAS-6 levels ensure formation of a single procentriole per

centriole during the centrosome duplication cycle. Dev. Cell 13, 203-213.

Wilkinson, C. J., Carl, M. and Harris, W. A. (2009). Cep70 and Cep131 contribute to

ciliogenesis in zebrafish embryos. BMC Cell Biol. 10, 17.

Wright, R. L., Salisbury, J. and Jarvlk, J. W. (1985). A nucleus-basal body connector

in Chlamydomonas reinhardtii that may function in basal body localization or

segregation. J. Cell Biol. 101, 1903-1912.

Zhao, T., Wang, W., Bai, X. and Qi, Y. (2009). Gene silencing by artificial microRNAs

in Chlamydomonas. Plant J. 58, 157-164.

CRC70 is a scaffold protein in centriole assembly 2975

Journ

alof

Cell

Scie

nce