NEWS AND VIEWS
822 VOLUME 42 | NUMBER 10 | OCTOBER 2010 | NATURE GENETICS
human hotspots, other gene products are likely
to play hotspot-specific roles.
Berg et al.
6
confirm that PRDM9 plays a key
role in both allelic recombination and certain
forms of genome instability and demonstrate the
remarkable effect that variation in one gene can
have on specific recombination and mutation
events. Furthermore, they raise the question of
which, if any, recurrent genomic mutations are
activated by individuals lacking the common
PRDM9 allele and whether, because there are
strong population differences in PRDM9 alleles,
NAHR disorders may also have large differ-
ences in frequency between populations. Future
studies are needed to search for the molecular
partners of PRDM9 in recruiting recombina-
tion and to characterize which of these partners
play general versus hotspot-specific roles.
COMPETING FINANCIAL INTERESTS
The authors declare no competing financial interests.
1. Ptak, S.E. et al. Nat. Genet. 37, 429–434 (2005).
2. Winckler, W. et al. Science 308, 107–111 (2005).
3. Baudat, F. et al. Science 327, 836–840 (2010).
4. Parvanov, E.D., Petkov, P.M. & Paigen, K. Science 327,
835 (2010).
5. Myers, S. et al. Science 327, 876–879 (2010).
6. Berg, I.L. et al. Nat. Genet. 42, 859–863 (2010).
7. Myers, S., Freeman, C., Auton, A., Donnelly, P. &
McVean, G. Nat. Genet. 40, 1124–1129 (2008).
8. Borde, V. et al. EMBO J. 28, 99–111 (2009).
9. Buard, J., Barthes, P., Grey, C. & de Massy, B. EMBO J.
28, 2616–2624 (2009).
10. Lindsay, S.J., Khajavi, M., Lupski, J.R. & Hurles, M.E.
Am. J. Hum. Genet. 79, 890–902 (2006).
11. Myers, S., Bottolo, L., Freeman, C., McVean, G. &
Donnelly, P. Science 310, 321–324 (2005).
12. Kong, A. et al. Science 319, 1398–1401 (2008).
13. Chowdhury, R., Bois, P.R., Feingold, E., Sherman, S.L.
& Cheung, V.G. PLoS Genet. 5, e1000648 (2009).
14. Broman, K.W., Murray, J.C., Sheffield, V.C., White, R.L.
& Weber, J.L. Am. J. Hum. Genet. 63, 861–869
(1998).
15. Jeffreys, A.J. & Neumann, R. Hum. Mol. Genet. 14,
2277–2287 (2005).
inactivity for almost all non-A PRDM9
alleles. More likely, PRDM9 may be capable of
binding highly degenerate copies of the motif,
and it is possible that a favorable flanking
sequence
7
strengthens binding to highly degen-
erate motifs. Consistent with this idea, sequences
containing three mismatches to the eight non-
degenerate bases in the 13-bp motif occur near
the center of each of the five non-motif hotspots
examined by Berg et al.
6
Another possibility is
that PRDM9 does bind in a sequence-specific
manner but also exerts H3K4me3 activity in cis
at some distance from the binding location.
Although PRDM9 is an important player
in recombination, additional factors influ-
ence binding and hotspot activity. Modifiers
of recombination rate have been identified
at the genome-wide level
12,13
, and there are
differences between recombination rates in
males and females at megabase scales
14
. Within
several hotspots, specific SNPs influencing
recombination activity have been identified
15
,
and these SNPs often occur in sequences lack-
ing the 13-bp motif. Further, previous statisti-
cal analysis
7
has identified multiple additional
motifs enriched in recombination hotspots
but which bear no homology to the predicted
PRDM9 binding sequence. For example, one of
the motif-containing hotspots studied by Berg
et al.
6
is centered within a THE1B retrotrans-
poson. On this specific repeat background, the
presence of an 8-bp motif, 129 bp upstream
from the 13-bp motif and therefore outside
the region likely bound by PRDM9, leads to a
twofold increase in the average recombination
rate
11
. The apparent background specificity of
this motif and others suggests that although
PRDM9 binding is a shared feature across
to the motif. The results of Berg et al. imply
that PRDM9 is involved in a much higher
fraction of hotspots than the 40% previously
estimated to be activated by the motif, and
that PRDM9 may be involved in all hotspots.
Berg et al. further found that these hotspots are
primarily activated by the common reference
PRDM9 allele (referred to as allele A). However,
in an additional hotspot cluster, one hotspot was
activated more strongly by a non-A allele and
the other hotspot was activated only by non-A
alleles, with subtle amino acid changes within
the array strongly altering hotspot activity.
The connections between recombination,
minisatellite and NAHR rearrangement events
suggest that PRDM9 variation is likely to
influence rearrangement frequencies, at least
where the 13-bp motif is present. Berg et al.
6
directly tested this prediction, finding that for
three hypervariable minisatellites where the
repeated element contains a close match to
the hotspot motif
7
, and for a recurrent NAHR
rearrangement where there is also a crossover
hotspot likely driven by the hotspot motif
7,10
,
variation in PRDM9 has a profound effect on
the rate of mutation. In contrast, at a recurrent
translocation site where there was no previous
evidence for the involvement of recombina-
tion or the hotspot motif, PRDM9 variation
had no influence on the mutation rate.
Motifs and modifiers
Berg et al.
6
demonstrate that in some cases,
PRDM9 may define hotspot location without
binding to the known 13-bp hotspot motif.
This may suggest that different zinc fingers are
required for binding to different hotspots
11
.
However, this seems unlikely given the consistent
Harvesting the apple genome
James Giovannoni
The genome sequence of the domesticated apple has been assembled and compared to previously sequenced plant
genomes. The genetic sequence of the 17 apple chromosomes shows evidence of a recent genome duplication that
may have spawned the additional gene family members needed for the evolution and development of the unique fruit
structure of the apple termed the pome.
James Giovannoni is at the United States
Department of Agriculture, Agricultural
Research Service and the Boyce Thompson
Institute for Plant Research, Cornell University,
Ithaca, New York, USA.
e-mail: jjg3@cornell.edu
On page 833 of this issue, an international
consortium of plant scientists led by the
Istituto Agrario di San Michele all’Adige
(IASMA) Research and Innovation Center
in Trento, Italy report the genome sequence
of the cultivated apple (Malus × domestica)
1
.
Apples are among the most widely grown
and consumed fruits in temperate regions
of the world. This is in part due to years of
extensive worldwide breeding and selec-
tion resulting in a treasure trove of apple
colors, flavors and textures with broad ver-
satility for the creation of numerous fresh
and processed foods. Equally important to
the apple’s prominence in the marketplace
(though less appreciated) is the fact that its
unique fruit structure, termed a pome, has
proven amenable to long-term controlled-
atmosphere storage, facilitating year-round
availability of high quality fruit from a crop
© 2010 Nature America, Inc. All rights reserved.