Mutational evolution in a lobular breast tumour profiled at single nucleotide resolution.

Abstract

Recent advances in next generation sequencing have made it possible to precisely characterize all somatic coding mutations that occur during the development and progression of individual cancers. Here we used these approaches to sequence the genomes (>43-fold coverage) and transcriptomes of an oestrogen-receptor-alpha-positive metastatic lobular breast cancer at depth. We found 32 somatic non-synonymous coding mutations present in the metastasis, and measured the frequency of these somatic mutations in DNA from the primary tumour of the same patient, which arose 9 years earlier. Five of the 32 mutations (in ABCB11, HAUS3, SLC24A4, SNX4 and PALB2) were prevalent in the DNA of the primary tumour removed at diagnosis 9 years earlier, six (in KIF1C, USP28, MYH8, MORC1, KIAA1468 and RNASEH2A) were present at lower frequencies (1-13%), 19 were not detected in the primary tumour, and two were undetermined. The combined analysis of genome and transcriptome data revealed two new RNA-editing events that recode the amino acid sequence of SRP9 and COG3. Taken together, our data show that single nucleotide mutational heterogeneity can be a property of low or intermediate grade primary breast cancers and that significant evolution can occur with disease progression.

Full text

Page 1

LETTERS
Mutational evolution in a lobular breast tumour
profiled at single nucleotide resolution
Sohrab P. Shah1,2*, Ryan D. Morin3*, Jaswinder Khattra1, Leah Prentice1, Trevor Pugh3, Angela Burleigh1,
Allen Delaney3, Karen Gelmon4, Ryan Guliany1, Janine Senz2, Christian Steidl2,5, Robert A. Holt3, Steven Jones3,
Mark Sun1, Gillian Leung1, Richard Moore3, Tesa Severson3, Greg A. Taylor3, Andrew E. Teschendorff6, Kane Tse1,
Gulisa Turashvili1, Richard Varhol3, Rene ´ L. Warren3, Peter Watson7, Yongjun Zhao3, Carlos Caldas6,
David Huntsman2,5, Martin Hirst3, Marco A. Marra3& Samuel Aparicio1,2,5
Recent advances in next generation sequencing1–4have made it
possible to precisely characterize all somatic coding mutations that
occur during the development and progression of individual can-
cers.Hereweusedtheseapproachestosequencethegenomes(.43-
fold coverage) and transcriptomes of an oestrogen-receptor-a-
positive metastatic lobular breast cancer at depth. We found 32
somatic non-synonymous coding mutations present in the meta-
stasis, and measured the frequency of these somatic mutations in
DNA from the primary tumour of the same patient, which arose
9years earlier. Five of the 32 mutations (in ABCB11, HAUS3,
SLC24A4, SNX4 and PALB2) were prevalent in the DNA of the
primary tumour removed at diagnosis 9years earlier, six (in
KIF1C, USP28, MYH8, MORC1, KIAA1468 and RNASEH2A) were
present at lower frequencies (1–13%), 19 were not detected in the
primary tumour, and two were undetermined. The combined ana-
lysis of genome and transcriptome data revealed two new RNA-
editing events that recode the amino acid sequence of SRP9 and
COG3. Taken together, our data show that single nucleotide muta-
tional heterogeneity can be a property of low or intermediate grade
primary breast cancers and that significant evolution can occur
with disease progression.
Lobularbreastcancer isanoestrogen-receptor-positive (ER1,also
knownasESR11)subtypeofbreastcancer(approximately15%ofall
breast cancers). It is usually of low-intermediate histological grade
and can recur many years after initial diagnosis. To interrogate the
genomic landscape of this class of tumour, we re-sequenced1–4the
DNAfromametastaticlobularbreastcancerspecimen(89%tumour
cellularity;SupplementaryFig.1)atapproximately43.1-foldaligned,
haploid reference genome coverage (120.7gigabases (Gb) aligned
paired-end sequence;Supplementary Fig.2,Table 1andSupplemen-
tary Methods). Deep high-throughput transcriptome sequencing
(RNA-seq)5performed on the same sample generated 160.9-million
reads that could be aligned (Supplementary Table 1, see also
Supplementary Fig. 2 and Supplementary Methods). The saturation
of the genome (Table 1) and RNA-seq (Supplementary Table 1)
libraries for single nucleotide variant (SNV) detection is discussed
in Supplementary Information. The aligned (hg18) reads were used
to identify (Supplementary Fig. 2) the presence of genomic aberra-
tions, including SNVs (Supplementary Table 2), insertions/deletions
(indels), gene fusions, translocations, inversions and copy number
alterations (Supplementary Methods). We examined predicted
coding indels and predicted inversions (coding or non-coding;
Supplementary Methods); however, all of the events that were vali-
dated by Sanger re-sequencing were also present in the germ line
(Supplementary Tables 3 and 4). None of the 12 predicted gene
fusions revalidated. We also computed the segmental copy number
(SupplementaryMethodsandSupplementaryTable5a)fromaligned
reads, and revalidated high level amplicons by fluorescence in situ
hybridization (FISH) (Supplementary Table 5b), revealing the pres-
ence of a new low-level amplicon in the INSR locus (Supplementary
Fig. 3).
We identified coding SNVs from aligned reads, using a Binomial
mixture model, SNVMix (Supplementary Table 2, Methods and
Supplementary Appendix 1). From the RNA-seq (WTSS-PE) and
genome (WGSS-PE) libraries we predicted 1,456 new coding non-
synonymous SNVMix variants (Supplementary Table 2). After the
removalofpseudogeneandHLAsequences(1,178positionsremain-
ing) and after primer design, we re-sequenced (Sanger amplicons)
1,120 non-synonymous coding SNV positions in the tumour DNA
and normal lymphocyte DNA. Some 437 positions (268 unique to
WGSS-PE, 15 unique to WTSS-PE, and 154 in common) were con-
firmed as non-synonymous coding variants. Of these, 405 were new
*These authors contributed equally to this work.
1Molecular Oncology,2Centre for Translational and Applied Genomics,3Michael Smith Genome Sciences Centre, BC Cancer Agency, 675 West 10th Avenue, Vancouver V5Z 1L3,
Canada.4Medical Oncology, BC Cancer Agency, 600 West 10th Avenue, Vancouver V5Z 1L3, Canada.5Department of Pathology, University of British Columbia, G227-2211
Wesbrook Mall, British Columbia, Vancouver V6T 2B5, Canada.6Cancer Research UK, Cambridge Research Institute, Li Ka Shing Centre, Robinson Way, Cambridge CB2 0RE, UK.
7Deeley Research Centre, BC Cancer Agency, Victoria V8R 6V5, Canada.
Table 1 | Summary of sequence library coverage
WGSS-PE WTSS-PE
Total number of reads
Total nucleotides (Gb)
Number of aligned reads
Aligned nucleotides (Gb)
Estimated error rate
Estimated depth (non-gap
regions)
Canonically aligned reads
Exons covered
2,922,713,774
140.991
2,502,465,226
120.718
0.021
43.114
182,532,650
7.108
160,919,484
6.266
0.013
NA
2,294,067,534
93.5 at .10 reads;
95.7 at .5 reads
109,093,616
82,200at10reads(seealso
Supplementary Table 1)
67.79
21,613,166
38.94
Reads aligned canonically (%) 78.49
Unaligned reads
Mean read length (bp)
420,248,548
48.24
The WGSS-PE column shows the genome paired-end read coverage for DNA from the
metastatic pleural effusion sample. The WTSS-PE column shows coverage for the
complementary DNA reads from the matched transcriptome libraries of the metastatic pleural
effusion.Coverageofexonbasesinthereferencegenome(hg18)isshownat5ormorereadsper
position, and 10 or more reads per position for the metastatic genome. bp, base pairs; NA, not
applicable.
Vol 461|8 October 2009|doi:10.1038/nature08489
809

Macmillan Publishers Limited. All rights reserved
©2009

Page 2

Table 2 | Somatic coding sequence SNVs validated by Sanger sequencing
GeneDescriptionPositionSource Allele
change
Amino
acid
change
Protein domain
affected
Expression
(sequenced
bases per
exonic base)
Allelic
expression
bias (R, NR
allele)
Copy number
classification
(HMM state)
ABCB11
Bile salt export pump
(ATP-binding cassette
sub-family B member 11)
HAUS3 coiled-coil
protein (C4orf15)
Cell division control
protein 6 homologue
Chromodomain-
helicase-DNA-binding
protein 3
Disks large
homologue 4
Receptor tyrosine-
protein kinase erb-b2
Fibrinogen alpha chain
2:169497197
WGSSC.TR.HTransmembrane
helix 3
0.31, 1
Amplification
(4)
HAUS34:2203607
WGSS, WTSSC.TV.M Unknown
14.14, 23
Neutral (2)
CDC6 17:35701114
WGSS, WTSSG.AE.KN-terminal,
unknown
Unknown,
C-terminal
2.73, 3
Amplification
(4)
Neutral (2)
CHD3
17:7751231
WGSSG.AE.K
3.9 41, 11
(Q,0.01)
DLG4 17:7052251
WGSSG.AP.LUnknown,
N-terminal
Kinase domain
5.57, 1
Neutral (2)
ERBB2 17:35133783
WGSS, WTSSC.GI.M
67.162, 35
Amplification
(4)
Gain (3) FGA
4:155726802
WGSSC.TW.stop Fibrinogen a/b/c
domain
Unknown,
N-terminal
Unknown,
C-terminal
0.01
NA
GOLGA4*
Golgin subfamily
A member 4
Glutathione S-transferase
C-terminal domain-
containing protein
3:37267947
WGSS, WTSSG.CE.Q
111.8 37, 12
Gain (3)
GSTCD
4:106982671
WGSS, WTSSG.CE.Q
23.2 23, 8
Neutral (2)
KIAA1468* LisH domain and HEAT
repeat-containing protein
KIF1C Kinesin-like protein
KIF1C
KLHL4
Kelch-like protein 4
18:58076768
WGSS, WTSSG.CR.TARM type fold
36.123, 11
Neutral (2)
17:4848025
WGSS, WTSSG.CK.NKinesin motor
domain
Unknown,
N-terminal
Actin-interacting
protein domain
N-terminal
prefolding
Unknown
28.5 16, 13
Neutral (2)
X:86659878
WGSSC.TS.L
1.71, 0
Neutral (2)
MYH8
Myosin8 (myosin heavy
chain 8)
Partner and localizer
of BRCA2
Polycystic kidney
disease and receptor for
egg-jelly-related protein
RAS and EF-hand
domain-containing
protein
17:10248420
WGSSC.GM.I
0
NANeutral (2)
PALB216:23559936
WGSS T.GE.A
13.0
NAAmplification
(4)
Gain (3)
PKDREJ
22:45035285
WGSSC.GE.Q
0.1
NA
RASEF
9:84867250
WTSSG.AS.LEF-hand Ca21-
binding motif
65.03, 2
Gain (3)
RNASEH2A Ribonuclease H2
subunit A (EC 3.1.26.4)
RNF220
RING finger protein
C1orf164
SP1
Transcription factor Sp1
19:12785252
WGSS, WTSSG.AR.H Unknown,
C-terminal
Unknown,
N-terminal
Glu-rich
N-terminal domain
Unknown
5.32, 2
Neutral (2)
1:44650831
WGSSG.AD.N
16.1
NA Neutral (2)
12:52063157
WGSSG.CE.Q
57.340, 10
(Q,0.01)
3, 7
Amplification
(4)
Gain (3) USP28
Ubiquitin carboxyl-
terminal hydrolase 28
UPF0197
transmembrane
protein C11orf10
Thyroid hormone-
inducible hepatic protein
Sciellin
Na1/K1/Ca21-
exchange protein 4
Collagen alpha-1(I)
chain precursor
GREB1-like protein
Coiled-coil domain-
containing protein 117
Novel protein
11:113185109
WGSS, WTSSC.TD.N
12.5
C11orf10
11:61313958
WGSSG.AT.ITransmembrane
domain
28.9 13, 3
Amplification
(4)
THRSP
11:77452594
WGSSC.TR.C Unknown
0.3
NA Gain (3)
SCEL
SLC24A4
13:77076497
14:92018836
WGSS
WGSS
A.G
G.A
K.R
V.I
Unknown
Transmembrane
domain
Pro-rich domain
0.3
1.2
1, 0
1, 0
Gain (3)
Amplification
(4)
Amplification
(4)
Neutral (2)
Neutral (2)
COL1A117:45625043
WGSSC.TG.D
80.024, 0
(Q,0.01)
4, 1
2, 0
KIAA1772
CCDC117
18:17278222
22:27506951
WGSS
WGSS
A.G
G.C
D.G
K.N
Unknown
Unknown
2.8
12.9
RP1-
32I10.10
MORC1
22:43140252
WGSSG.CE.QUnknown
0
NAGain (3)
MORC family CW-type
zinc finger protein 1
Sorting nexin 4
3:110271286
WGSSG.AA.VCoiled-coil
0.1
NA Gain (3)
SNX43:126721688
WGSSC.TD.NUnknown,
N-terminal
Hydroxylase
domain
Unknown,
C-terminal
43.4
NAGain (3)
LEPREL1
Prolyl 3-hydroxylase 2
precursor (EC 1.14.11.7)
WD repeat-containing
protein 59
3:191172415
WGSST.CE.G
1.1
NAGain (3)
WDR59*
16:73500342
WTSSC.TM.I
17.36, 5
Neutral (2)
Omnibustableshowingthefeaturesassociatedwiththe32Sangeramplicon-validatednon-synonymoussomaticmutationsfromtheWGSS-PEandWTSS-PElibraries.Mutationpositionsareonthe
basis of reference genome hg18. The nucleotide substitutions are shown as reference.variant. The amino acid change is shown as reference.variant amino acid. If the mutation occurs in a
recognized protein domain or motif this is shown. The transcript expression level in WTSS-PE reads is shown as the mean number of reads supporting each position in the transcript. The allelic
expression bias column shows the number of reference (R), non-reference (NR) reads in the WTSS-PE library at the mutated position. Three transcripts (CHD3, SP1 and COL1A1) show significant
expression bias (annotated with Q,0.01, Supplementary Methods) in favour of the reference allele; however, none of the heterozygous somatic mutations were biased in favour of the non-
referenceallele.TheexpressionofHAUS3ispredominantlynon-referenceasexpectedforahomozygousallele.TheHMMstateclassifierofcopynumberforthegenomicregionencompassingeach
mutation position is shown in the last column, as state (state number). C-terminal, carboxy-terminal; N-terminal, amino-terminal.
*Genes showing alternative splicing.
LETTERS
NATURE|Vol 461|8 October 2009
810

Macmillan Publishers Limited. All rights reserved
©2009

Page 3

germline alleles and 32 were revealed as non-synonymous coding
somatic point mutations (Table 2). Of the 32 somatic mutations,
30 were present in WGSS-PE and/or WTSS-PE, whereas two were
detected from the WTSS library sequence alone (Table 2). None of
the 32 genes were found in common with the CAN breast genes6,
which were discovered from ER2cell lines. Eleven genes appear in
thecurrentreleaseofCOSMIC7(CHD3,SP1,PALB2,ERBB2,USP28,
KLHL4, CDC6, KIAA1468, RNF220, COL1A1 and SNX4) but with
mutations at different positions. We examined the population
frequency of the somatic mutation positions for PALB2, ERBB2,
USP28, CDC6, CHD3, HAUS3 (previously known as C4orf15), SP1,
KIAA1468 and DLG4 in a further 192 breast cancers (Supplemen-
tary Methods; 112 lobular, 80 ductal). None of these 192 breast
cancersshowedidenticalmutationstothosedescribedhere;however,
3 out of 192 cases (2 lobular, 1 ductal) contained neighbouring non-
synonymous variants/deletions affecting the ERBB2 kinase domain
(Supplementary Fig. 4). Interestingly, 2 out of 192 cases (both lobu-
lar) contained two different heterozygous truncating variants in
HAUS3: chr4:2203685 G.T on minus strand, GAG.TAG
(Glu.stop), and chr4:2203483 C.G on minus strand, TCA.TGA
(Ser.stop) (Supplementary Fig. 5). Notably, HAUS3 is a member of
therecentlydescribed8–10multiproteinaugmincomplex,thefunction
of which is required for genome stability mediated by appropriate
kinetochore attachment and centrosome morphogenesis.
To determine how many of the somatic non-synonymous coding
sequence mutations were already present at diagnosis 9years earlier,
we next examined genomic DNA from the primary tumour directly,
byasinglemoleculefrequencycountingexperiment(Supplementary
Methods)4. Twenty-eight of the 32 mutations yielded amplicons
compatible with Illumina sequencing (Supplementary Methods),
and two extra mutations were sampled by Sanger sequencing
(Supplementary Fig. 5). As controls we selected 36 heterozygous
germline SNVs at random. The PCR amplicons for known germline
and somatic mutations were sequenced on an Illumina device. After
alignment, the observed counts of reference and non-reference bases
at the target position were compared using the Binomial exact test.
TocalibratetheexpectedmeanoftheBinomialdistribution, weused
the non-reference allele frequency from positions 25 to 15 sur-
rounding (but not including) the target position (Supplementary
Table 6a, b), where only reference bases should be called. Unequal
segmental amplification/deletion in the genome may contribute to a
departure from the theoretical ratio of 0.5 for a heterozygous allele.
As a result, amplicons from heterozygous germline alleles showed
occasional measured frequencies of between 0.2 and 0.8 in both the
primary and metastatic tumour DNA (Table 3 and Supplementary
Table 7), but with a modal frequency around 0.5, as expected. In the
metastatic genomic DNA the somatic mutations showed frequencies
of between 0.2 and 0.79 (Table 3). Notably, the somatic coding
mutation positions examined in the primary tumour showed three
patterns of abundance: prevalent, rare and undetectable (Table 3).
Mutations in ABCB11, PALB2 and SLC24A4 were detected at preval-
ent frequencies for heterozygous mutations ($0.2, the lowest value
seen for known germline alleles) given a 73% tumour content. The
frequency of the mutation in HAUS3 was 0.79, consistent with it
being a prevalent homozygous mutation, also confirmed by Sanger
sequencing (Supplementary Fig. 5). Sanger amplicon sequencing
showed that the SNX4 somatic mutation was also present in the
primary tumour, whereas the KIAA1772 (also known as GREB1L)
mutation was not. Six mutations (KIF1C, USP28, MORC1, MYH8,
KIAA1468 andRNASEH2A)
(P,0.01, Binomial exact test) intermediate frequencies of between
1% and 13% (Table 3), suggesting that these mutations were
showedstatisticallysignificant
Table 3 | Frequency of germline and somatic alleles in the metastatic and primary genomes
PositionLocusR NRPrimary
depth
Primary
NR ratio
Primary
Pvalue
Primary statusMetastasis
depth
Metastasis
NR ratio
M Copy number
classification(HMMstate)
4:2203607
16:23559936
2:169497197
14:92018836
17:10248420
3:110271286
17:4848025
11:113185109
18:58076768
19:12785252
4:106982671
17:35701114
17:7751231
4:155726802
17:7052251
3:37267947
9:84867250
17:35133783
X:86659878
3:191172415
16:73500342
1:44650831
22:45035285
11:61313958
12:52063157
11:77452594
17:45625043
13:77076497
19:9314428
4:130144460
8:27835012
6:32908543
20:43363061
4:8672089
16:1331138
HAUS3
PALB2
ABCB11
SLC24A4
MYH8
MORC1
KIF1C
USP28
KIAA1468
RNASEH2A
GSTCD
CDC6
CHD3
FGA
DLG4
GOLGA4
RASEF
ERBB2
KLHL4
LPREL1
WDR59
RNF220
PKDREJ
C11ORF10
SP1
THRSP
COL1A1
SCEL
2
2
2
2
2
2
2
C
T
C
G
C
G
G
C
G
G
G
G
G
C
G
G
G
C
C
T
C
G
C
G
G
C
C
A
A
A
G
C
G
G
C
T
G
T
A
G
A
A
T
A
A
T
T
A
T
A
T
T
A
T
C
T
A
T
A
T
T
A
G
G
T
A
T
A
A
T
5700
115
506
13347
10657
24572
8587
6654
719
6537
7273
4894
9665
5756
4383
13051
1690
3736
6561
11963
4846
8160
6674
116705
7732
24219
26343
49
176
2020
13587
4718
5950
381
677
0.5472
0.4957
0.3261
0.2341
0.1353
0.0468
0.0107
0.0095
0.0083
0.0029
0.0008
0.0008
0.0007
0.0007
0.0007
0.0006
0.0006
0.0005
0.0005
0.0004
0.0004
0.0004
0.0003
0.0003
0.0003
0.0002
0.0001
0.0000
0.5057
0.2188
0.8602
0.7484
0.5249
1.0000
0.4963
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0020
0.0276
0.9885
0.9733
0.9981
0.9911
0.9835
0.9999
0.9500
0.9899
0.9993
1.0000
0.9982
0.9999
0.9999
1.0000
1.0000
1.0000
1.0000
1.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
Dominant
Dominant
Dominant
Dominant
Subdominant
Subdominant
Subdominant
Subdominant
Subdominant
Subdominant
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Absent
Present
Present
Present
Present
Present
Present
Present
762
669
959
13670
1797
32273
2272
1387
1056
1497
2208
4208
1737
2287
706
3262
796
1722
977
8381
1396
967
1230
14354
2011
40652
32259
187
321
2081
10781
16370
5540
2850
554
0.7874
0.4350
0.3691
0.3518
0.5932
0.4107
0.3077
0.4484
0.3059
0.2806
0.2174
0.3577
0.2671
0.2755
0.3272
0.2235
0.3656
0.3612
0.3153
0.2148
0.2629
0.2203
0.3366
0.4651
0.2193
0.4750
0.2543
0.5722
0.4953
0.3099
0.6667
0.4897
0.5049
0.8032
0.6245
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
G
G
G
G
G
G
G
Neutral (2)
Amplification (4)
Amplification (4)
Amplification (4)
Neutral (2)
Gain (3)
Neutral (2)
Gain (3)
Neutral (2)
Neutral (2)
Neutral (2)
Amplification (4)
Neutral (2)
Gain (3)
Neutral (2)
Gain (3)
Gain (3)
Amplification (4)
Neutral (2)
Gain (3)
Neutral (2)
Neutral (2)
Gain (3)
Amplification (4)
Amplification (4)
Gain (3)
Amplification (4)
Gain (3)
Neutral (2)
Neutral (2)
Deletion (1)
Amplification (4)
Amplification (4)
Gain (3)
High-levelamplicon(5)
Only7germlineallelesareshown,thefulllistisinSupplementaryTable7.Thegenomepositionsareshownaschr:coordinate.Theprimaryreaddepthrepresentsthenumberofreads.Binomialexact
Pvalues were calculated using a Binomial exact test. R, reference base; NR, non-reference base. Primary status indicates whether the variant was present, subdominant or absent in the primary
tumour. Column M denotes somatic (S) or germline (G) single nucleotide variants in the metastasis. HMM state refers to the metastasis.
NATURE|Vol 461|8 October 2009
LETTERS
811

Macmillan Publishers Limited. All rights reserved
©2009

Page 4

restricted to minor subclones of tumour cells. The remaining 19 out
of 30 of the somatic coding mutations were not detected in the
primary tumour DNA. Thus, significant heterogeneity in tumour
somatic mutation content existed in the primary tumour at dia-
gnosis. In contrast with the recently reported sequence of cytogen-
etically normalacutemyeloid leukaemia(AML)tumour4,significant
evolution of coding mutational content occurred between primary
and metastasis. It is unknown whether the 19 mutations present in
themetastasis,butnotdetectedintheprimary,wereaconsequenceof
radiation therapy or innate tumour progression.
We also examined how the transfer of information from the nuc-
lear genome to proteins was modified by alternative splicing
(Supplementary Table 8 and Supplementary Fig. 6), biased allelic
expression (Supplementary Table 9) and RNA editing. At the single
nucleotide level, RNA-editing enzymes (which can be regulated by
oestrogens11) may also recode transcripts resulting in a proteome
divergent from the genome12–15. Interestingly, the ADAR enzyme—
one of the principal RNA-editing enzymes that mediates ARI(G)
edits—was one of the top 5% of genes expressed (145.6 reads per
base, Supplementary Table 10), and the only editing enzyme
expressed at a high level. We searched for potential editing events
(Methods) and found 3,122 candidate edits in 1,637 gene loci
(Supplementary Table 11). Some 526 out of 3,122 candidate edits
are non-synonymous changes and 232 are synonymous changes
(with the remainder affecting untranslated regions). We revalidated
independently (Supplementary Methods) by Sanger sequencing 75
editing events in 12 gene loci from the lobular metastasis
(Supplementary Table12and
molonc.bccrc.ca/). Two genes, COG3 and SRP9 (Fig. 1), showed
confirmed high frequency non-synonymous transcript editing,
resultinginvariantproteinsequences.Theseobservations emphasize
theimportanceofintegratingRNA-seqdatawithtumourgenomesin
assessing protein variation.
The coding mutation landscape of breast cancers has, so far, been
mostly determined from ER2metastatic cell lines/samples6,16, and
has suggested the presence of large numbers of passenger events as
well as drivers. Our results show the importance of sequencing sam-
plesoftumourcellpopulationsearlyaswellaslateintheevolutionof
tumours,andofestimatingallelefrequencyintumourgenomes.Our
observations suggest that the sequencing of primary breast cancers
and pre-invasive malignancy may reveal significantly fewer candi-
dates for tumour initiating mutations.
see tracedataat http://
METHODS SUMMARY
Paired-end reads were assigned quality scores and aligned to the reference gen-
ome (hg18) using Maq17(Supplementary Methods and Supplementary Fig. 2).
For identification of SNVs we used a simple Binomial mixture model, SNVMix
(Supplementary Appendix 1), which assigns a probability to each base position
as homozygous reference (aa), heterozygous non-reference (ab) and homo-
zygous non-reference (bb), based on the occurrence of reference (hg18) and
non-referencebasesateachalignedposition.Thismodelwascalibratedinitially,
using high confidence allele calls from Affymetrix SNP6.0 hybridization of
tumour and normal DNA. We estimated the receiver operating characteristic
(ROC) performance (Supplementary Fig. 8) and determined that an SNVMix
threshold of P50.77 for (ab) or (bb) for a non-reference call wouldyield a false
discoveryrate(FDR)of1%.FortheRNA-seqlibrary,athresholdofP50.53was
used (Supplementary Fig. 8; FDR50.01) to call non-reference positions. Non-
reference positions were then filtered for known variants against the sources of
germline variation, the single nucleotide polymorphism database (dbSNP) and
the completed individual genomes18,19(Supplementary Table 2). Saturation of
the libraries for SNV discovery was determined by random re-sampling
(Supplementary Fig. 9 and Supplementary Methods). Segmental copy number
was inferred with a hidden Markov model (HMM) method (Supplementary
Table 4a, b and Supplementary Methods).
We searched for RNA-editing events by examining all very high confidence
(P(ab)1P(bb).0.9) SNVMix predictions from the RNA-seq library of the
metastatic tumour, that were not found with extreme confidence (P(aa).0.99,
derived from the SNVMix receiver operating curve at FDR50.01) at the same
positions in the metastatic tumour genome library.
Received 4 September; accepted 10 September 2009.
1. Campbell, P. J. et al. Identification of somatically acquired rearrangements in
cancer using genome-wide massively parallel paired-end sequencing. Nature
Genet. 40, 722–729 (2008).
Bentley, D. R. et al. Accurate whole human genome sequencing using reversible
terminator chemistry. Nature 456, 53–59 (2008).
Morin, R. et al. Profiling the HeLa S3 transcriptome using randomly primed cDNA
and massively parallel short-read sequencing. Biotechniques 45, 81–94 (2008).
Ley, T. J. et al. DNA sequencing of a cytogenetically normal acute myeloid
leukaemia genome. Nature 456, 66–72 (2008).
Mortazavi, A., Williams, B. A., McCue, K., Schaeffer, L. & Wold, B. Mapping and
quantifying mammalian transcripomes by RNA-Seq. Nature Methods 5, 621–628
(2008).
Wood, L. D. et al. The genomic landscapes of human breast and colorectal
cancers. Science 318, 1108–1113 (2007).
Forbes,S.A.etal.TheCatalogueofSomaticMutationsinCancer(COSMIC).Curr.
Protoc. Hum. Genet. Unit 10.11, doi:10.1002/0471142905.hg1011s57 (2008).
Goshima, G., Mayer, M., Zhang, N., Stuurman, N. & Vale, R. D. Augmin: a protein
complexrequiredforcentrosome-independentmicrotubulegenerationwithinthe
spindle. J. Cell Biol. 181, 421–429 (2008).
Meireles, A. M., Fisher, K. H., Colombie, N., Wakefield, J. G. & Ohkura, H. Wac: a
newAugminsubunitrequiredforchromosomealignmentbutnotforacentrosomal
microtubule assembly in female meiosis. J. Cell Biol. 184, 777–784 (2009).
10. Lawo, S. et al. HAUS, the 8-subunit human Augmin complex, regulates
centrosome and spindle integrity. Curr. Biol. 19, 816–826 (2009).
11.Pauklin, S., Sernandez, I. V., Bachmann, G., Ramiro, A. R. & Petersen-Mahrt, S. K.
Estrogen directly activates AID transcription and function. J. Exp. Med. 206,
99–111 (2009).
12. Blow, M., Futreal, P. A., Wooster, R. & Stratton, M. R. A survey of RNA editing in
human brain. Genome Res. 14, 2379–2387 (2004).
13. Athanasiadis, A., Rich, A. & Maas, S. Widespread A-to-I RNA editing of Alu-
containing mRNAs in the human transcriptome. PLoS Biol. 2, e391 (2004).
14. Maas, S., Kawahara, Y., Tamburro, K. M. & Nishikura, K. A-to-I RNA editing and
human disease. RNA Biol. 3, 1–9 (2006).
15. Li, J. B. et al. Genome-wide identification of human RNA editing sites by parallel
DNA capturing and sequencing. Science 324, 1210–1213 (2009).
16. Greenman, C. et al. Patterns of somatic mutation in human cancer genomes.
Nature 446, 153–158 (2007).
17. Li, H., Ruan, J. & Durbin, R. Mapping short DNA sequencing reads and calling
variants using mapping quality scores. Genome Res. 18, 1851–1858 (2008).
18. Wang, J. et al. The diploid genome sequence of an Asian individual. Nature 456,
60–65 (2008).
19. Wheeler, D. A. et al. The complete genome of an individual by massively parallel
DNA sequencing. Nature 452, 872–876 (2008).
2.
3.
4.
5.
6.
7.
8.
9.
Supplementary Information is linked to the online version of the paper at
www.nature.com/nature.
Acknowledgements We thank C. Eaves and M. Pollak for comments on earlier
versions of the manuscript. We thank and acknowledge the patients of the BC
Cancer Agency for donations of tumour tissues to the TTR-BREAST tumour
banking program. S.A. is supported by a Canada Research Chair in Molecular
Oncology,S.P.S.,J.K.,L.P.,A.B.andT.P.aresupportedbyMichaelSmithFoundation
for Health Research awards. R.D.M. is a Vanier scholar (CIHR). A.B. is also
supported by an NSERC award, and L.P. by a CIHR award. We are grateful for
platformsupportfromCIHR,GenomeCanada,GenomeBC,CanadaFoundationfor
Innovation and the Michael Smith Foundation for Health Research. The work was
funded by the BC Cancer Foundation and the CBCF BC/Yukon chapter.
Author Contributions S.P.S. and R.D.M.: led the data analysis and wrote the
manuscript. M.H.: oversaw the sequencing efforts. J.K., L.P., T.P., J.S., C.S., A.B.,
Tumour RNA
Tumour DNA
COG3SRP9
Figure 1 | RNA editing in COG3 and SRP9. Sanger sequence traces from the
non-synonymouseditingpositionsinCOG3andSRP9.Theeditingposition
is arrowed. Top trace is tumour RNA, bottom trace tumour DNA. The
editing positions were confirmed with reverse strand reads (not shown).
LETTERS
NATURE|Vol 461|8 October 2009
812

Macmillan Publishers Limited. All rights reserved
©2009

Page 5

R.M. and T.S.: validation of variants. A.D.: primer design. K.G. and P.W.:
establishment of TTR-BREAST tumour bank. K.T., R.G., R.A.H., S.J., M.S., G.L.,
A.E.T., R.V., G.A.T. and R.L.W.: bioinformatic analysis. G.T., D.H. and P.W.: sample
selection and histological grading. Y.Z.: Illumina sequencing library preparation.
C.C. and D.H.: data analysis and interpretation. S.A. and M.A.M.: conceived and
oversaw the study and wrote the manuscript.
Author Information Genome sequence data have been deposited at the European
Genotype Phenotype Archive (http://www.ebi.ac.uk/ega), which
ishostedbytheEBI, under accession number EGAS00000000054. Reprints and
permissions information is available at www.nature.com/reprints.
Correspondence and requests for materials should be addressed to M.A.M.
(mmarra@bcgsc.ca) or S.A. (saparicio@bccrc.ca).
NATURE|Vol 461|8 October 2009
LETTERS
813

Macmillan Publishers Limited. All rights reserved
©2009