Proc. Natl. Acad. Sci. USA
Vol. 92, pp. 5915-5919, June 1995
Cell Biology
Coiled bodies contain U7 small nuclear RNA and associate with
specific DNA sequences in interphase human cells
MARK R. FREY AND A. GREGORY MATERA*
Department of Genetics, Center for Human Genetics and Program in Cell Biology, Case Western Reserve University, 10900 Euclid Avenue,
Cleveland, OH 44106-4955
Communicated by Joseph G. Gall, Carnegie Institution, Baltimore, MD, March 7, 1995 (received for review January 20, 1995)
C snurposomes, was shown to have significant structural
homology to a CB marker protein, p80-coilin (6, 7), and when
Wu et al (8) showed that p80-coilin protein is targeted to
spheres when the human coilin mRNA was injected into
amphibian oocytes. Thus spheres and CBs are homologous
organelles that are likely to have similar functions, but as yet
no functions have been demonstrated (5).
Of particular interest is the fact that some of the spheres
within the amphibian oocyte are attached to the so-called
"sphere-organizer" regions of the lampbrush chromosomes.
These sphere-organizers are located at the histone gene clusters in two classes of amphibia, Anura and Urodela (9, 10). The
fact that spheres and histone genes are associated in two fairly
divergent species suggests a conserved function. Taken together with the fact that U7 small nuclear RNA (snRNA) was
also found to be colocalized within the amphibian C snurposome (11), we decided to examine the CBs of mammalian cell
nuclei for the presence of U7 and for possible associations with
"CB organizer" sequences.
We show that U7 snRNA is distributed throughout the
nucleoplasm, excluding nucleoli, and is concentrated within
CBs in human and murine cells. Furthermore, we find that
subsets of the histone, RNU1, and RNU2 gene loci are
associated with subsets of CBs in HeLa monolayer cells.
Previously (12), we found that in another strain of HeLa cells
(one that grows in suspension) there was no obvious colocalization of the RNU2 gene locus and CBs (as marked by U2
snRNA hybridization). Our interpretation that CBs are not the
site(s) of U2 snRNA synthesis hinged upon the assumption that
U2 loci are equally active, requiring a colocalization of all U2 loci
with CBs. In this paper, we reinvestigate this issue and find that,
in addition to the previously observed differences in snRNP
distribution between the two types of HeLa cells, the cells that
have been adapted for growth in suspension have fewer CBs,
fewer U2 loci, and a much lower fraction of cells showing an
association between the two structures than their monolayergrown counterparts. Furthermore, we find that CBs associate
with Ul and U2 gene loci in a high percentage of HEp-2 cells. In
light of these data, the apparent association of CBs with human
Ul, U2, and histone gene loci must be a very dynamic one.
Coiled bodies (CBs) are nuclear organelles
ABSTRACT
whose structures appear to be highly conserved in evolution.
In rapidly cycling cells, they are typically located in the
nucleoplasm but are often found in contact with the nucleolus.
The CBs in human cells contain a unique protein, called
p80-coilin. Studies on amphibian oocyte nuclei have revealed
a protein within the "sphere" organelle that shares significant
structural similarity to p80-coilin. Spheres and CBs are also
highly enriched in small nuclear ribonucleoproteins and other
RNA-processing components. We present evidence that, like
spheres, CBs contain U7 small nuclear RNA (snRNA) and
associate with specific chromosomal loci. Using biotinylated
2'-O-methyl oligonucleotides complementary to the 5' end of
U7 snRNA and fluorescence in situ hybridization, we show that
U7 is distributed throughout the nucleoplasm, excluding
nucleoli, and is concentrated in CBs. Interestingly, we found
that CBs often associate with subsets of the histone, Ul, and
U2 snRNA gene loci in interphase HeLa-ATCC and HEp-2
monolayer cells. However, in a strain of suspension-grown
HeLa cells, called HeLa-JS1000, we found a much lower rate
of association between CBs and snRNA genes. Possible roles
for CBs in the metabolism of these various histone and
snRNAs are discussed.
Eukaryotic cells can be viewed as if they were a mixture of
compartments, some of which can be defined physically, some
functionally. In the cytoplasmic compartment, the roles of the
various organelles and the interplay between them have been
extensively studied. In contrast, the domains and their interconnections within the nucleus are poorly understood. However, it is
now clear that the nucleus is a highly organized structure,
permeated by a proteinaceous "matrix" and composed of many
different subdomains. The so-called interchromatin space of
somatic cells can be characterized at the ultrastructural level by
several types of structures. These are perichromatin fibrils, interchromatin granule clusters, and nuclear bodies. Small nuclear
ribonucleoproteins (snRNPs) are major components of each of
these three subcompartments (see refs. 1 and 2 for reviews).
One type of nuclear body, the coiled body (CB), contains
relatively high concentrations of small RNPs as well as other
proteins (see Table 1 for a list of components reported to be
either enriched or not enriched in CBs). Despite (or perhaps
because of) this rather lengthy roster, the list of putative CB
functions is nearly as long (see refs. 3 and 4 for reviews). Early
studies on amphibian oocyte nuclei also revealed morphologically distinct structures, including: nucleoli, lampbrush chromosomes, and "spheres." The subsequent finding that spheres
(which are composed of B- and C-type "snurposomes") contain numerous snRNPs and snRNP proteins caused speculation that CBs and spheres were related organelles (reviewed in
ref. 5). The final pieces to the puzzle linking CBs and spheres
were revealed when SPH-1, an integral protein component of
MATERIALS AND METHODS
DNA Clones, Oligonucleotides, and Antiserum. Clones for
the human RNU1 locus at lp36 and the RNU2 locus at 17q21
were gifts of A. Weiner (13, 14). The 5S rDNA clone was
provided by R. Little (15). Human histone gene subclones from
the loci on human chromosomes 1q21 (16) and 6p2l (17) were
gifts of D. Collart and D. Doenecke, respectively. The cosmid
containing the nerve growth factor receptor (NGFR) gene
from chromosome 17q21-22 was a gift of K. Kidd (see ref. 14).
The human c-myc cosmid at 8q24.1 was a gift of D. Ward (18).
Abbreviations: CB, coiled body; snRNA, small nuclear RNA; snRNP,
small nuclear ribonucleoprotein; NGFR, nerve growth factor receptor; FITC, fluorescein isothiocyanate.
*To whom reprint requests should be addressed.
The publication costs of this article were defrayed in part by page charge
payment. This article must therefore be hereby marked "advertisement" in
accordance with 18 U.S.C. §1734 solely to indicate this fact.
5915
5916
Cell Biology: Frey and Matera
Table 1. CB components
Enriched
Ul
U2
U4
U5
U6
Ull
U12
U3
U7
U8*
p80-coilin
Topoisomerase I
Fibrillarin
U2 auxiliary factor
Trimethylguanosine cap
Sm core proteins
Ribosomal protein S6
U2 B" protein
Not enriched
Poly(A)+ RNA
DNA
28S rRNA
Nucleolin (C23)
Nucleolar protein B23
hnRNPs A, C, I, L, and M
7SK
SR proteins (e.g, SC-35)
5S rRNA/RNP
RNA polymerase I
IG cluster antigen (3C5)
Ribosomal protein S1
PML protein
ND/SP100-Kr antigen
Noppl4O/NAP57
hnRNP, heterogeneous nuclear RNPs; IG, interchromatin granule.
*To date, this RNA has been demonstrated only in CBs formed in vitro
in Xenopus egg extract.
Probes corresponding to the human retinoblastoma gene (no.
P5313) and Down syndrome (no. P5321) regions (Oncor) are
localized to 13q14 and 21q22.3, respectively. The specificity of
our genomic DNA hybridization probes was controlled by
routine metaphase fluorescence in situ hybridization analysis,
using the same conditions as for the interphase cells. In each
case the hybridization signals were found exclusively at the
target loci, with the exception of the Ul probe, which very
weakly crosshybridized to the Ul pseudo-array at lql2-q21
(19) on metaphase spreads.
The antisense oligomer U7-(3-20), 5'-CTAAAAGAGCUGUAACACBBBBdC-3', was a gift of E. Scharl and J. Steitz
(Yale University, New Haven, CT). The residue preceded by
a lowercase d is deoxyribose (DNA); A, G, C, and U are
2'-O-methyl RNA. B is a biotin phosphoramidite residue. A
nearly random 22-mer, 5'-(dN)2,dT-3', was used as a nonspecific hybridization competitor. Polyclonal rabbit antiserum R508
was used to identify CBs (7).
Immunofluorescence and in Situ Hybridization. Cells were
grown in monolayers on chambered glass slides (Nunc) and
prepared for hybridization as described (20). The HeLaJS1000 cells, which normally grow in suspension, were allowed
to adhere onto slides for the purposes of the CB association
experiment. Previous studies have shown that multiple passages of adherent growth of these cells do not change the
general nucleoplasmic organization of snRNPs [i.e., the pattern remains diffuse, not speckled (12)]. After equilibrating the
slides in 4x SSC/0.1% Tween 20 for 5-10 min at room
temperature, the cells were hybridized in 20 ,lI of 4x SSC, 10%
dextran sulfate, 1 ,ug of competitor oligonucleotide (a random 22-mer) per ,ul, and 1-2 ng of biotinylated U7 oligomer
per ,ul, as described (20).
Simultaneous detection of genomic DNA sequences and
CBs was performed on slides that were preextracted with
detergent, using a method pioneered by Carter etal (21). After
equilibration in 10% normal goat serum in PBS-A (phosphatebuffered saline, pH 7.2, without Mg2+ or Ca2+), cells were
incubated with anti-p80 coilin (1:200 dilution) for 30 min at
37°C in the same solution. Slides were washed (three times, 5
min each) in PBS-A at room temperature before incubation
with Texas red-conjugated anti-rabbit IgG (Vector Laboratories) and washing in PBS-A. The slides were fixed a second
time in 4% paraformaldehyde, quenched in glycine, and rinsed.
After dehydration through a graded ethanol series, the slides
Proc. Natl. Acad. Sci. USA 92
(1995)
were allowed to air-dry. Hybridization with the appropriate
biotin- or digoxigenin-labeled genomic clone and digital imaging
microscopy were performed essentially as described (20).
A dual-bandpass fluorescein isothiocyanate (FITC)/Texas
red filter set (Chroma Tech) was used for scoring large
numbers of interphase nuclei. The numbers of red and green
signals were counted as well as the number associated (those
whose signals were at least partially overlapping). Care was
taken to examine multiple planes of focus within each cell to
ensure an accurate signal count.
RESULTS
U7 snRNA Is Concentrated in CBs in Mammalian Cells. If
spheres and CBs are truly related organelles, it stands to
reason that they should have similar RNA complements. Wu
and Gall (11) have reported that U7 RNA is present within the
C snurposomes (spheres) in Xenopus. Therefore, we wanted to
see if U7 is also concentrated within mammalian CBs. Using
a biotinylated 2'-O-methyl oligonucleotide complementary to
the 5' end of human U7 snRNA, we hybridized it to fixed,
undenatured HeLa-ATCC and NIH 3T3 cells. Hybridization
was detected throughout the nucleoplasm, excluding nucleoli,
and to a lesser extent in the cytoplasm (Fig. 1). No hybridization signal was detected in cells that were treated with RNase
A. Like the other Sm class snRNPs, U7 also intensely stains
CBs (bright foci in Fig. 1). We have confirmed the identity of
the foci as CBs by double labeling with anti-p80-coilin (data
not shown). The same hybridization pattern was detected in
"normal" diploid lines such as Detroit 551 and HT-1080 cells
FIG. 1. Using a 2'-O-methyl oligonucleotide complementary to the
single-stranded 5' end of U7 RNA, hybridization signals are detected
throughout the nucleoplasm, excluding nucleoli, and to a lesser extent
in the cytoplasm of human (A, HeLa-ATCC) and mouse (B, NIH 3T3)
cells. Like the other Sm class snRNPs, U7 is also enriched in CBs
(bright dots).
Cell
Biology: Frey and Matera
(unpublished results). Thus spheres and CBs contain components of three major RNA-processing pathways: pre-mRNA
(Ul, U2, U4/6, and U5), pre-rRNA (U3), and histone premRNA (U7) (5, 11, 22, 23).
Some snRNA and Histone Gene Loci Are Associated with
CBs in Interphase Human Cells. Since sphere-organizer regions lie adjacent to or within histone gene loci (9, 10) on the
lampbrush chromosomes of amphibian oocyte nuclei, we
wanted to determine whether somatic CBs are associated with
specific DNA sequences in human cells. Using an approach
that allows simultaneous visualization of DNA and protein, we
found that a significant fraction of interphase HeLa-ATCC
and HEp-2 nuclei reveals an association of the tandemly
repeated RNU1 and RNU2 gene loci with CBs (Fig. 2).
Proc. Natl. Acad. Sci. USA 92 (1995)
5917
However, these same sequences were only marginally associated with the CBs of HeLa-JS1000 cells, a strain of HeLa that
has been adapted for growth in spinner flasks. Approximately
72% and 77% of the HeLa-ATCC and HEp-2 cells, respectively, displayed at least one RNU2 locus associated with at
least one CB (Fig. 3), whereas only 22% of the HeLa-JS1000
cells showed this type of colocalization. Interestingly, the
RNU1 locus showed a somewhat lower degree of association
with CBs in HeLa-ATCC and HEp-2 cells (43% and 41%,
respectively), but still more frequently than in HeLa-JS1000
cells (21%). The data for these experiments, including the
controls, are summarized in Fig. 3 and Table 2.
Clearly, only a subset of the snRNA genes is colocalized with
a subset of CBs. We also wanted to determine whether or not
FIG. 2. Paraformaldehyde-fixed HeLa-ATCC cells were incubated with anti-p80-coilin primary antibodies (shown in red). The cells were then
fixed a second time, denatured, and hybridized with biotin- or digoxigenin-labeled DNA and detected with appropriate FITC- (shown in green)
or Cy5- (shown in white) conjugated reporter molecules. Nuclei were counterstained in 4',6-diamidino-2-phenylindole (blue). A subset of the U2
(A) and Ul (B) snRNA gene loci (green) colocalizes with a subset of CBs (red). (C) A four-color hybridization using a digoxigenin-labeled U2
probe (green) and a biotinylated Ul probe (white) reveals that these genes do not associate with the same CBs (see text for additional details).
A genomic clone containing an H2B gene from the histone gene cluster on chromosome 1 (D; His-1, shown in green) and another clone that contains
Hi and H2B gene sequences from chromosome 6 (E; His-6, also in green) associate with CBs (red). For ease of visualization and scoring, the histone
gene probe signals were amplified using biotinylated anti-avidin antibodies and a second round of FITC-conjugated avidin. (F) In contrast with the snRNA
and histone genes, the 5S rDNA sequences (green) do not colocalize with CBs. Images from two different experiments are shown in A, B, D, and E.
5918
Proc. Natl. Acad. Sci. USA 92
Cell Biology: Frey and Matera
ml)
I
0
ct 60
0
co*c
ux
-
Al
40-
on
cn
20n
ua
_
C
a)
C)
CZ
z
Zd
z
O
M
C6
v
C0
.cn
Z
ZJZ0
ZpHEp-2
Z
J~~~S1000
FIG. 3. CB association data. The results from multiple scoring
experiments for each probe were pooled. The experiments were
reproducible, varying no more than ±5% from the average. Each of
the probes listed below was hybridized to detergent-preextracted
HeLa-ATCC cells (solid bars): RNU1 (Ul genes), RNU2 (U2 genes),
NGFR (NGFR gene), His-1 (histone gene on 1q21), His-6 (histone
gene on 6p2l), pH5S (5S genes), c-myc (c-myc gene), pRb (retinoblastoma gene), Down (Down syndrome region probe). Additionally,
the RNU1 and RNU2 probes were each scored for CB colocalizations
in HeLa-JS1000 cells (striped bars) and HEp-2 cells (stippled bars).
See Table 2 for more details. Of the HeLa-ATCC cells that had
CB/U2 gene colocalizations, 46% had one association, 42% had two,
and 12% had three or more. For Ut genes in HeLa-ATCC cells the
corresponding numbers were 81%, 17%, and 2%. Of the HEp-2 cells
that displayed CB/U2 gene associations, 44% had one, 44% had two,
and 12% had three or more. The Ul genes in HEp-2 cells showed 97%
with one association, 3% with two, and none with three or more.
Ut and U2 genes associated with the same subset of CBs. In
a four-color hybridization experiment (Fig. 2C), we showed
that in cells that displayed associations of Ut and U2 gene loci
with CBs, they did not colocalize with the same ones. Since
four-color experiments are not amenable to visual scoring of
large numbers of cells (one of the colors is from an infrared
dye), we have also confirmed this result by a simple cohybridization of differentially labeled Ut and U2 genes. Slides were
scored as described above with a dual-bandpass (FITC/Texas
red) filter set. We found that >90% of the HeLa-ATCC cells
had nonoverlapping Ut and U2 hybridization patterns (i.e.,
(1995)
about 8% of the cells had one signal overlap; none of the cells
had two or more overlapping signals).
The cell-cycle-regulated histone genes are clustered on
human chromosomes 1 and 6 (17). Using single-copy probes
from each of these clusters, we found that 44% (His-1) and
32% (His-6) of HeLa-ATCC cells had at least one CB association (see Fig. 3). Additional control hybridizations were
performed using the tandemly repeated 5S rDNA locus from
human chromosome 1, the NGFR gene (which is located near
the RNU2 locus on chromosome 17), the c-myc gene on
chromosome 8, and cosmid contigs spanning the retinoblastoma gene on chromosome 13 and the Down syndrome
"critical region" on chromosome 21. We found that these
sequences did not colocalize with CBs (the hybridization data
can be found in Fig. 3 and Table 2). The NGFR gene is located
about 8 centimorgans distal to RNU2 and is resolvable on
metaphase chromosomes (14). Approximately 8% of the
NGFR signals were overlapping with at least one CB, whereas
our other control probes (which are not near the RNU2,
RNU1, or histone gene loci) were completely negative (Fig. 3).
We believe that the slightly higher level of signal overlap seen
with the NGFR cosmid is the artefactual result of a large,
brightly fluorescent object (the CB) in the general vicinity
("70% of the time) of the RNU2 locus.
DISCUSSION
Histone protein synthesis is coordinated with DNA synthesis
(reviewed in ref. 24). There are two major groups of histone
genes: (i) most of the core histones, which are replication dependent and therefore cell-cycle regulated, and (ii) the variant
histones, which are expressed constitutively. The genes for the
replication-dependent histones do not contain introns and are not
polyadenylylated. Instead, the mRNAs possess a 3' terminal loop
structure that is formed by an endonucleolytic cleavage event that
requires the U7 snRNP (24). Thus it would not have been
unreasonable to expect that the localization of the machinery
required for proper 3' end maturation of histone mRNAs (i.e.,
the U7 snRNP) would also be cell-cycle regulated and localized.
Yet, we did not detect any obvious local concentrations of U7
snRNA in the nucleoplasm, other than in CBs. This result was
somewhat surprising in view of the fact that overall histone
mRNA levels can vary up to 35-fold during the cell cycle (25).
More specifically, Mannironi et at (26) have shown that the levels
of individual human H2A genes (including those from clusters on
the same chromosome) varied by at least 8-fold and that these
differences in expression levels were similar in various types of
proliferating cells. While it is possible that only active histone
genes associate with CBs (during S phase), it is clear that more
in-depth and controlled cell-cycle studies need to be performed
before any conclusions are drawn.
Table 2. Hybridization data
Probe
a-p80-coilin
RNUl
RNU2
NGFR
His-1
His-6
pH5S
Location
CB
lp36
17q21-22
17q21-22
1q21
6p2l
c-myc
lq42.1
8q24.1
pRb
Down
13q14
Organization
Tandem
Tandem
Single-copy
Clustered
Clustered
Tandem
Single-copy
Single-copy
Single-copy
21q22.3
*Approximate target size.
tAverage number ± 1 SD (n 2 100).
4:Size expressed in ,um.
Insert size, kb
30
6
30
18
10
2.2
37
4 x 30
4 x 30
Target size,* kb
0.5-1.0t
900
100
30
18
10
200
37
100
100
Average no. of signalst
HeLa-ATCC
HeLa-JS1000
5.1 ± 1.7
1.9 ± 0.8
3.0 ± 0.8
2.1 ± 0.9
2.8 ± 0.8
1.5 ± 0.6
2.8 ± 1.2
2.2 ± 0.8
1.4 ± 0.5
3.7 ± 0.9
1.3 ± 0.5
1.7 ± 0.7
1.3 ± 0.5
HEp-2
2.5 ± 1.4
2.0 ± 1.0
2.9 ± 1.0
Cell
Proc. Natl. Acad. Sci. USA 92
Biology: Frey and Matera
In analogy with the rRNA genes, Matera and Ward (12)
postulated that CBs might be a kind of "snRNP nucleolus," a
place where snRNA genes from different chromosomes come
together to perform snRNA transcription. Our in situ hybrid-
ization data reveal that Ul and U2 gene loci do not colocalize
with the same CBs or with each other (Fig. 2C). Therefore,
CBs are probably not the snRNA equivalent of the nucleolus.
The hypothesis that CBs are not the sites of U2 RNA synthesis
(12) was based on the assumption that all U2 loci were equally
active, requiring a complete colocalization of U2 genes and
CBs. Since ultrastructural analyses of CBs reveal that CBs are
neither efficiently nor rapidly labeled when cells are pulsed
with [3H]uridine (1, 27) we maintain that they are not major
sites of RNA synthesis and do not accumulate nascent RNAs.
Several lines of evidence (e.g., the presence of trimethylguanosine cap and Sm protein antigens within CBs) suggest
that mature rather than nascent snRNPs are concentrated
within CBs (see ref. 28 for review). Restriction analysis of the
tandemly repeated U2 genes from human cells shows that the
repeat units are indistinguishable and that efficient gene
correction events have homogenized the U2 arrays in other
primate lineages (29). Thus the individual gene loci are
apparently equivalent, yet some of them colocalize with CBs
while others do not. However, the high percentage of cells
having at least one CB-snRNA gene colocalization is not likely
due to random chance or we would have detected a higher
association rate with our other control gene probes. Our data
suggest a situation that is much more dynamic than previously
envisioned. Formally, there are at least three models that
explain the data: (i) CBs fortuitously associate with sequences
that are near the histone, U1, and U2 gene loci; (ii) CBs denote
sites of snRNA transcription, but only some of the loci are
active; (iii) CBs are involved in some post-transcriptional
aspect of histone and snRNA metabolism. In attempting to
experimentally discriminate between these three scenarios, we
offer the following comments:
(i) The chance occurrence of four different CB-organizer
regions being located adjacent to the Ul, U2, and the two histone
gene clusters seems unlikely. However, it is possible that structural rearrangements between the notoriously aneuploid HeLaJS1000 and HeLa-ATCC cells could cause CB-organizer regions
to be juxtaposed to these loci in one strain and not in the other.
(ii) One intriguing possibility that has not yet been postulated for CBs would be that they represent sites where snRNA
transcription is being repressed. It should be noted that we do
not actually know the relative gene dosage in the various cell
lines (i.e., the number of tandem repeats); rather we can only
assess the average number of loci. Perhaps the active U2 loci
are the ones without associated CBs; the presence of the
mature snRNPs within these structures could be a form of
autologous regulation. This scenario would potentially explain
our results with the suspension-grown HeLa-JS1000 cells in
which there are fewer U2/CB associations than in the HeLaATCC strain (20% vs. 70%). On average, the suspension cells
also have fewer U2 loci and fewer, but larger, CBs (see Table
2). Mechanisms for regulating the relative amounts of the
various spliceosomal RNAs must certainly exist. Given that
there are, on average, three copies of the RNU2 locus in
HeLa-ATCC and HEp-2 cells, and only half that number in
HeLa-JS1000 cells, it is likely that there are "20 fewer U2
genes in JS1000 cells (29). Coupled with the faster doubling
time (unpublished observations) and the more diffuse nucleoplasmic snRNP distribution in the suspension cells (12), it
would seem that they are also more metabolically (i.e., transcriptionally) active than the monolayer version. Taken together, the CBs in JS1000 cells may not associate as often with
U2 loci because of the increased transcriptional demand per
U2 transcription unit.
(iii) One common thread that links the major U snRNAs and
histone mRNAs is that they are all nonpolyadenylylated poly-
(1995)
5919
merase II transcripts with conserved 3' terminal stem-loop
structures. The association we noted of human CBs to histone,
Ul, and U2 genes and the well-documented association of
spheres to histone loci on amphibian lampbrush chromosomes
may be the result of a common mechanism of 3' end maturation.
Perhaps the most plausible "post-transcriptional" role for CBs
is in the preassembly/transport of snRNP complexes. Recent
evidence demonstrates that, like fibrillarin, two proteins called
Noppl40 and NAP57 are located in the nucleolus and in CBs (30)
(see Table 1). The latter two proteins have been implicated in
nucleolo-cytoplasmic transport. Our data suggest that the associations of CBs with the individual chromosomal loci are not
stable ones. Hence, it is also possible that CBs are motile,
although the data supporting this hypothesis are indirect.
We thank E. Chan, D. Collart, D. Crumpler, D. Doenecke, J.
Jacobberger, K Kidd, R. Little, E. Scharl, S. Schwartz, J. Stein, J.
Steitz, D. Ward, and A. Weiner for gifts of materials used in this work.
We are also grateful to C. O'Keefe for preparation of metaphase
spreads, to A. Weiner for helpful discussions, and to H. Willard for
critical reading of the manuscript. Special thanks go to K. Carter and
J. Lawrence for pointing out the discrepancy between our two strains
of HeLa cells and for communicating results prior to publication. This
work was supported by a start-up grant from Case Western Reserve
University School of Medicine (to A.G.M.).
Fakan, S. (1994) Trends Cell Biol. 4, 86-90.
Spector, D. L. (1993) Annu. Rev. Cell BioL. 9, 265-315.
Lamond, A. I. & Carmo-Fonseca, M. (1993) Trends CellBiol. 3,198-204.
Brasch, K. & Ochs, R. L. (1992) Exp. Cell Res. 202, 211-223.
5. Gall, J. G., Tsvetkov, A., Wu, Z. & Murphy, C. (1995) Dev. Genet. 16,
1.
2.
3.
4.
25-35.
6. Tuma, R. S., Stolk, J. A. & Roth, M. B. (1993) J. Cell Biol. 122,767-773.
7.
8.
9.
10.
11.
12.
13.
Chan, E. K. L., Takano, S., Andrade, L. E. C., Hamel, J. C. & Matera,
A. G. (1994) Nucleic Acids Res. 22, 4462-4469.
Wu, Z., Murphy, C. & Gall, J. G. (1994) Mo!. BioL Cell 5, 1119-1127.
Gall, J. G., Stephenson, E. C., Erba, H. P., Diaz, M. 0. & BarsacchiPilone, G. (1981) Chromosoma 84, 159-171.
Callan, H. G., Gall, J. G. & Murphy, C. (1991) Chromosoma 101,
245-251.
Wu, C.-H. H. & Gall, J. G. (1993) Proc. Nat!. Acad. Sci. USA9,4
6257-6259.
Matera, A. G. & Ward, D. C. (1993) J. Cell Bio. 121, 715-727.
Bernstein, L. B., Manser, T. & Weiner, A. M. (1985) Mo!. Cell. Biol.
5,2159-2171.
14. Pavelitz, T., Rush6, L., Matera, A. G., Scharf, J. & Weiner, A. M.
(1995) EMBO J. 14, 169-177.
15. Little, R. D. & Braaten, D. C. (1989) Genomics 4, 376-383.
16. Coliart, D., Romain, P. L, Huebner, K., Pockwinse, S., Pilapil, S.,
Cannizzaro, L A., Croce, C. M., Stein, J. L & Stein, G. S. (1992) J. Cell.
Biochem. 50, 374-385.
17. Albig, W., Drabent, B., Kunz, J., Kaiff-Suske, M., Grzeschik, K.-H. &
Doenecke, D. (1993) Genomics 16, 649-654.
18. Lux, S. E., Tse, W. T., Menninger, J. C., John, K. M., Harris, P.,
Shalev, 0., Chilcote, R. R., Marchesi, S. L., Watkins, P. C., Bennett,
V., McIntosh, S., Collins, F. S., Francke, U., Ward, D. C. & Forget,
B. G. (1990) Nature (London) 345, 736-739.
19. Lindgren, V., Bernstein, L. B., Weiner, A. M. & Francke, U. (1985)
Mo!. Cell. Biol. 5, 2159-2171.
20. Matera, A. G., Frey, M. R., Margelot, K. & Wolin, S. L. (1995)1J. Cell
Biol., in press.
21. Carter, K. C., Taneja, K. L. & Lawrence, J. B. (1991) J Cell Biol. 115,
1191-1202.
22. Carmo-Fonseca, M., Pepperkok, R., Carvalho, M. T. & Lamond, A. I.
(1992)1. Cell Biol. 117, 1-14.
23. Jim6nez-Garcia, L. F., Segura-Valdez, M. d. L., Ochs, R. L. & Spector, D. L. (1994) MoL Biol. Cell 5, 955-966.
24. Marzluff, W. F. (1992) Gene Expression 2, 93-97.
25. Harris, M. E., B6hni, R., Schneiderman, M. H., Ramamurthy, L,
Schuimperli, D. & Marzluff, W. F. (1991) Mo!. Cell. Biol. 11, 2416-2424.
26. Mannironi, C., Orr, A., Hatch, C., Pilch, D., Ivanova, V. & Bonner,
W. (1994) DNA Cell Biol. 13, 161-170.
27. Fakan, S. & Puvion, E. (1980) Int. Rev. Cytol. 65, 255-287.
28. Lamond, A. I. & Carmo-Fonseca, M. (1993) Mol. Biol. Rep. 18,127-133.
29. Matera, A. G., Weiner, A. M. & Schmid, C. W. (1990) Mo!. Cell. Biol.
10, 5876-5882.
30. Meier, U. T. & Blobel, G. (1994) J. Cell Biol. 127, 15051514.