Function of the chemokine receptor CXCR4 in haematopoiesis and ...

25. Connor, R. I., Sheridan, K. E., Ceradini, D., Choe, S. & Landau, N. R. Change in coreceptor use
correlates with disease progression in HIV-1-infected individuals. J. Exp. Med. 185, 621–628 (1997).
26. Donzella, G. A. et al. AMD3100, a small molecule inhibitor of HIV-1 entry via the CXCR4 co-receptor.
Nature Med. 4, 72–77 (1998).
27. Gu, H., Zou, Y.-R. & Rajewsky, K. Independent control of immunogobulin switch recombination at
individual switch regions evidenced through Cre-loxP-mediated gene targeting. Cell 73, 1155–1164 (1993).
28. Schaeren-Wiemers, N. & Gerfin-Moser, A. A single protocol to detect transcripts of various types and
expression levels in neural tissue and cultured cells: in situ hybridization using digoxigenin-labelled
cRNA probes. Histochemistry 100, 431–440 (1993).
29. Landreth, K. S. & Dorshkind, K. Pre-B cell generation potentiated by soluble factors from a bone
marrow stromal cell line. J. Immunol. 140, 845–852 (1988).
Acknowledgements. We thank F. Hatan and M.-J. Sunshine for technical assistance; S. Vukmanovic for
help with the thymic transplant experiments; J. Johnson and N. Heitz for anti-Math1 and anti-BLBP
antibodies; K. Dorshkind and R. R. Hardy for the S17 cell line; and G. Fishell, A. Joyner, M. Chao,
C. Mason, S. Jung, V. KewalRamani and C. Davis for comments on the manuscript; Y.-R.Z. thanks H. Gu
for his continuous support. This work was supported by an NIH grant (to D.R.L.). Y.-R.Z. is the recipient
of a postdoctoral fellowship from the Irvington Institute, D.R.L. is an Investigator of the Howard Hughes
Medical Institute.
Correspondence and requests for materials should be addressed to Y.-R.Z. (e-mail: zou@saturn.med.nyu.edu).
Histone macroH2A1
is concentrated in the
inactive X chromosome
of female mammals
Carl Costanzi & John R. Pehrson
Department of Animal Biology, School of Veterinary Medicine, University of
Pennsylvania, Philadelphia, Pennsylvania 19104, USA
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
In female mammals one of the X chromosomes is rendered almost
completely transcriptionally inactive 1,2 to equalize expression of
X-linked genes in males and females. The inactive X chromosome
is distinguished from its active counterpart by its condensed
appearance in interphase nuclei 3 , late replication 4 , altered DNA
methylation 2 , hypoacetylation of histone H4 (ref. 5), and by
transcription of a large cis-acting nuclear RNA called Xist 6–10 .
Although it is believed that the inactivation process involves the
association of specific protein(s) with the chromatin of the
inactive X, no such proteins have been identified 11 . We discovered
a new gene family encoding a core histone which we called
macroH2A (mH2A) 12,13 . The amino-terminal third of mH2A
proteins is similar to a full-length histone H2A, but the remaining
two-thirds is unrelated to any known histones. Here we show that
an mH2A1 subtype is preferentially concentrated in the inactive X
chromosome of female mammals. Our results link X inactivation
with a major alteration of the nucleosome, the primary structural
unit of chromatin.
We examined the distribution of mH2A in mouse liver nuclei by
immunofluorescence using antibodies against the non-histone
region of one of the mH2A1 subtypes, mH2A1.2 (Fig. 1a). Most
hepatocyte nuclei were brightly stained by these antibodies (Fig. 2a),
although the nuclei of bile duct cells, endothelial cells and connective
tissue showed less staining (data not shown). Speckled
staining was present through most of the nuclei of both males
and females. The nuclei of females, however, also had large, distinct
mH2A-dense regions, which we name macrochromatin bodies
(MCBs) (Fig. 2a, b). We found this sex difference in all mice we
examined, including several sets of siblings, as well as in dog liver
sections and human primary skin fibroblasts (data not shown). We
detected MCBs using several methods of tissue fixation and a variety
of polyclonal antibodies, including ones raised in rabbits or
chickens and against the non-histone region of mH2A1.1 (data
not shown). MCBs were usually perinucleolar, but little or no
mH2A1.2 staining was detected in nucleoli, as assessed by doublelabel
immunofluorescence with a monoclonal antibody against the
nucleolar protein fibrillarin 14 (data not shown).
To quantify the sex specificity of MCBs in liver parenchyma, we
letters to nature
identified complete nuclei in liver sections by confocal microscopy
and assessed their MCB content (Fig. 2c). In mice, 85% of nuclei
from females contained MCBs, compared with less than 1% of
nuclei from males. A similar distribution was observed in dogs: 90%
of female nuclei contained MCBs compared with less than 1% of
male nuclei. The relative mH2A1 protein content of female and
male mouse livers was similar (Fig. 1b), and previously estimated in
rat liver to be one mH2A per 30 nucleosomes 12 .
The female specificity of MCBs suggested a relationship to Xchromosome
inactivation. We therefore localized MCBs relative to
X chromosomes by staining MCBs in female mouse liver sections by
immunofluorescence and then localizing the X chromosomes in
these sections by fluorescent in situ hybridization (FISH) with a
DNA probe that ‘paints’ mouse X chromosomes (X-paint) 15
(Fig. 3a). Using confocal microscopy, we found that 99% of
MCBs colocalized to an X chromosome and 43% of X chromosomes
colocalized to an MCB (Fig. 3b). In a control experiment, MCBs
never colocalized with chromosome 4 (data not shown). The Xpaint
results also confirmed that hepatocyte nuclei with more than
one MCB (Fig. 2a, c) were polyploid (Fig. 3a). The incidence of
polyploid nuclei was higher in older mice (data not shown).
These results indicate that the female-specific MCBs involve one
of the two X chromosomes. To determine which X chromosome is
involved, we examined the nuclear distribution of mH2A1 proteins
in female mice with one X-chromosome, male mice with two and
human Klinefelter fibroblasts with four. X/0 mice are females with
just one X chromosome (the active X), as is found in human
Turner’s syndrome. The mH2A1.2 staining pattern of X/0 liver
sections was identical to that of normal males (data not shown).
Sex-reversed mice have one normal X and one (designated Xsxr) that
carries a translocated piece of the Y chromosome including the sexdetermining
locus16 . XXsxr mice are phenotypically male but have
one active and one inactive X chromosome. The mH2A1.2 staining
of liver sections of XXsxr mice was identical to that of normal female
mice (data not shown). Finally, we analysed a human skin cell line
derived from a boy with Klinefelter’s syndrome17 . The sex chromosome
complement of this cell line is XXXXY (one active and three
inactive X chromosomes). Of these nuclei, 63% had three MCBs
each (Fig. 4a, b). We also observed preferential mH2A1.2 staining of
three chromosomes in metaphase spreads prepared from these cells
8
Figure 1 Specificity of mH2A antibodies. a, A diagram of mH2A1 subtypes.
mH2A1.1 and 1.2 are identical apart from a segment generated by alternative
splicing (cross-hatched). The fragment used to generate the antibodies is
indicated by the solid bar. The H2A region is shaded; the segment rich in basic
amino acids is indicated by plus signs. b, Western blot analysis of mouse liver
nuclear extracts using antibodies raised against mH2A1.1 (anti-1.1) and mH2A1.2
(anti-1.2). The mH2A1.2 blot shows extracts from six littermates. Gel loading was
normalized against core-histone content 12 .
Nature © Macmillan Publishers Ltd 1998
NATURE | VOL 393 | 11 JUNE 1998 599