Near-Haploidization Significantly Associates with Oncocytic Adrenocortical, Thyroid, and Parathyroid Tumors but not with Mitochondrial DNA Mutations
Willem E. Corver,* Tom van Wezel, Kees Molenaar, Melanie Schrumpf, Brendy van den Akker, Ronald van Eijk, Dina Ruano Neto, Jan Oosting, and Hans Morreau*
Department of Pathology, Leiden University Medical Center, RC, Leiden, Netherlands
Mitochondrial-rich oncocytic thyroid tumors frequently show near-haploidization and endoreduplication (masked haploid- ization), which manifests as a near-homozygous genome (NHG). We now extend this investigation to include adrenocorti- cal cancer and parathyroid carcinoma (PaTC), which we studied for a NHG in association with mitochondrial DNA mutations. Sixty endocrine tumors from 59 patients were studied, including 46 thyroid tumor samples of varying histology, 11 adrenocortical cancers, and 3 PaTCs. Genome-wide SNP array analysis and DNA content analysis were combined to determine the chromosomal dosage (allelic state). The entire mitochondrial genome was also studied for mutations. In addition, tumors were characterized for somatic mutations in a subset of genes that are directly or indirectly implicated in cellular metabolism. In addition to a subset of thyroid cancers (n = 5), a NHG was also observed in | of 3 PaTCs and 6 of Il adrenocortical cancers. All but one of the tumors with a NHG (n = 12) showed oncocytic metaplasia (P = 0.0001, two- tailed Fisher’s exact). One or more damaging or disrupting mtDNA mutations were found in 68% (41/60) of tumor sam- ples. No correlation was found between mtDNA mutations and the oncocytic phenotype or a NHG, and none of the mutations in nuclear encoded genes correlated with the oncocytic phenotype or a NHG. A subset of oncocytic tumors of the thyroid, parathyroid, and adrenocortical carcinomas carries a NHG. Although damaging/disrupting mtDNA mutations are frequently found in oncocytic and nononcocytic endocrine tumors, neither correlates with a NHG phenotype nor with an oncocytic phenotype. @ 2014 Wiley Periodicals, Inc.
INTRODUCTION
The biology of cancer is complex and is perhaps most succinctly described in the now classic review “Hallmarks of cancer” by Hanahan and Weinberg. This article was recently updated and included sev- eral new concepts, one of which was the reprogram- ming of energy metabolism (Hanahan and Weinberg, 2011). This phenomenon was originally described in the 1920s by Otto Warburg (Warburg et al., 1927) when he postulated that cancer cells switch from oxidative phosphorylation to “aerobic glycolysis” even in the presence of oxygen. Recent reports suggest that alterations in cancer cell metab- olism can also effect cells in the tumor microenvir- onment, causing them to display a “reverse Warburg effect” whereby they provide cancer cells with energy-rich metabolites (Pavlides et al., 2009).
Metabolic alterations are associated with activat- ing gene mutations in classic cancer driver genes (RAS genes, PIK3CA-AKT-mTOR pathway; Wullschleger et al., 2006; Gaglio et al., 2011), inac- tivation of tumor suppressor genes (P53, LBK1; Feng and Levine, 2010; Hardie and Alessi, 2013), alterations in nuclear-encoded gene subunits of
mitochondrial enzyme complexes (IDH, FH, and SDH genes; Cervera et al., 2009; Pansuriya et al., 2011) and mutations in mitochondrial DNA (mtDNA; Gasparre et al., 2007) that encodes subu- nits of the respiratory chain. However, the detailed interplay between these factors is not yet fully understood.
Some human tumors show a striking mitochon- drial proliferation that often leads to a swollen appearance, referred to as “oncocytic” metaplasia, which can be observed using microscopy. MtDNA mutations have been associated with oncocytic tumors. In oncocytic thyroid tumors, or so-called
Additional Supporting Information may be found in the online version of this article.
Supported by: the Dutch Cancer Society (Koningin, Wilhelmina Fonds, KWF), UL 2010-4656 (T. van Wezel and H. Morreau). Willem E. Corver and Tom van Wezel contributed equally.
*Correspondence to: Willem E. Corver and Hans Morreau; Department of Pathology, Leiden University Medical Center, P.O. Box 9600, Building 1, L1-Q, 2300 RC Leiden, Netherlands. E-mail: w.e.corver@lumc.nl and j.morreau@lumc.nl Received 12 April 2014; Accepted 28 May 2014 DOI 10.1002/gcc.22194
Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com).
Hürthle cell tumors (Maximo et al., 2012), mito- chondrial proliferation is associated with impaired ATP production (Savagner et al., 2001b) that leads to the upregulation of genes associated with gly- colysis, the tricarboxylic acid cycle, and oxidative phosphorylation (OXPHOS; Baris et al., 2004). The presence of oncocytic metaplasia in thyroid tumors does not seem to be related to IDH1 muta- tions (Hemerly et al., 2010) but has been linked to mitochondrial-encoded DNA mutations (Gasparre et al., 2007).
We recently showed, by integrated DNA con- tent and SNP array analysis, that oncocytic thyroid cancers go through a phase of incomplete genomic haploidization [hyperhaploidy (Mandahl et al., 2012)] that is often followed by endoreduplication [a duplication of the near-haploid genome, also termed masked near-haploid by others (Holmfeldt et al., 2013)], resulting in a near-homozygous genome (NHG; Corver et al., 2012). In combina- tion with the observation of a NHG in a follicular thyroid carcinoma of the oncocytic type (FTC- OV), an association between genomic haploidiza- tion and mtDNA mutations was postulated. How- ever, tumors with clear morphological oncocytic features do not always show a NHG, as seen in several cases in a validation series.
MtDNA mutations have also been found in other types of endocrine tumors such as parathy- roid oncocytoma and adrenal gland oncocytoma (Zimmermann et al., 2011). In chromophobe, renal cell carcinomas show mtDNA mutations (Nagy et al., 2002) and also frequently show a CGH pat- tern of genomic alterations (Speicher et al., 1994) highly comparable to that found in FTC-OV (Wada et al., 2002; Corver et al., 2012), suggesting a correlation between a NHG and mtDNA muta- tions. MtDNA mutations have also been found in renal oncocytoma (Mayr et al., 2008), although a characteristic pattern of genomic alterations is only observed in some of these cases (Kim et al., 2009).
In addition, XTC.UC1 cells (Savagner et al., 2001a)-the only model for oncocytic thyroid can- cer-harbor damaging mtDNA mutations (Bonora et al., 2006) and similarly show a characteristic CGH-pattern of genomic alterations, suggesting a NHG (Ribeiro et al., 2008). These cells grow poorly in glucose-free medium supplemented with galactose and pyruvate and show a strong reduc- tion of ATP production and oxygen consumption under these culture conditions, whereas control cells show adaptation. The mitochondrial dysfunc- tion in these cells is thought to be caused by dam- aging mtDNA mutations in complex I and III,
leading to a dramatic decrease of enzymatic activ- ity of the respiratory chain (Bonora et al., 2006). Taken together, these observations suggest a rela- tion between certain mitochondrial deficiencies, lack of energy and loss of entire chromosomes dur- ing tumor development and progression in a sub- set of human cancers.
Since oncocytic tumors are most prevalent in tumors from the endocrine system, we formulated the following questions: is the NHG phenotype of thyroid tumors also seen in other types of endo- crine tumors? is this genomic phenotype associ- ated with the histological characteristics of oncocytic tumors? and finally, is there a relation between mtDNA mutations and a NHG?
To address these questions, we studied a series of 59 patients from whom 60 frozen tissue samples were available, including thyroid tumors, adreno- cortical cancer, and parathyroid cancers. Of the thyroid tumors, two have been described in a pre- vious publication (Corver et al., 2012). We ana- lyzed the current series for NHGs using high- density SNP arrays and DNA content analysis. Using the DNA index (DI), the allelic dosage or allelic state was inferred (Corver et al., 2008). Fur- thermore, the complete mtDNA mitochondrial genomes of these tumors was studied, including the common deletion (Duregon et al., 2011), com- plemented by mutation analysis of molecular com- ponents of the MEK/ERK, PIK3CA/AKT pathways and IDH1/2, all of which are directly or indirectly associated with cellular metabolism.
MATERIALS AND METHODS
Patient Samples
Archival samples from the department of Pathology, Leiden University Medical Center were analyzed and consisted of 60 snap-frozen endocrine tumor samples and matching formalin- fixed paraffin-embedded (FFPE) tissues. These tumor samples comprised 45 thyroid tumors of varying benign or malignant histology, 11 adrenal cortical cancers (ACC), and three PaTCs (Table 1). One thyroid tumor provided two samples (Table 1, No. 55ª and 55b). Samples were handled according to the medical ethical guidelines described in the Code Proper Secondary Use of Human Tissue established by the Dutch Federa- tion of Medical Sciences (www.federa.org). Tis- sues sections were prepared from both frozen and FFPE samples, hematoxylin and eosin stained and then reviewed by an experienced pathologist
| Organ | N Histology | NHG | MT-ATP | MT-CO | MT-ND | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n | N | Y | 6 | 8 | MT-CYB | 1 2 | 3 | 1 | 2 | 3 | 4 | 4L | 5 | ||
| Adrenal | 11 ACA | 1 | 1 | – | – | – | – | – – | – | – | – | – | 1 | – | |
| ACC | 2 | 1 | 1 | 2 | – | – | – – | – | – | – | – | – | – | – | |
| ACC-OV | 8 | 3 | 5 | 1 | – | 2(1/1) | – – | – | – | 1 | – | 2(1/1) | – | 1 | |
| Parathyroid | 3 PaTC-OV | 3 | 2 | 1 | – | – | 1 | – – | 1 | – | – | – | 1 | – | – |
| Thyroid | 46 ATC | 8 | 6 | 2ª | – | – | 3(2/1) | 1 – | – | – | 1 | – | 2(1/1) | – | |
| FA | 4 | 4 | – | – | – | 1 | 1 – | – | – | – | – | – | – | – | |
| FA-OV | 5 | 5 | – | – | – | 1 | 1 – | – | 2 | – | – | 1 | – | 3(2/1) | |
| FTC | 3 | 3 | – | – | – | – | 1 – | – | – | 2(1/1) | – | – | – | ||
| FTC-MI-OV | 3 | 3 | – | – | – | – | – – | – | 1 | – | – | – | – | – | |
| FTC-OV | 4 | 1 | 3 | – | – | 2 | 1 | – | – | 1 | – | – | |||
| FVPTC | 2 | 2 | – | 1 | – | – | 1 | – | – | – | – | – | – | ||
| HP | 1 | 1 | – | – | – | – | – – | – | – | – | – | – | – | – | |
| PTC | 15 | 15 | – | 3 | 1 | 1 | I I | b | 2(1/1) | 3(2/1) | – | 2 | 1 | 1 | |
| PTC-OV | 1 | 1 | – | – | – | – | – – | b | – | – | |||||
| Total | 60 | 48 | 12 | 7 | 1 | 7 1 | 3 | 5 | 6 | 1 | 9 | 1 | 5 | ||
Bold = disruptive mutation, italic = heteroplasmic mutation,
ACA, adrenocortical adenoma; ACC, adrenocortical carcinoma; ACC-OV, ACC-oncocytic variant; PaTh-OV, parathyroid cancer-oncocytic variant; ATC, anaplastic thyroid cancer; FA, thyroid follicular ade- noma; FA-OV, FA-oncocytic variant; FTC, follicular thyroid carcinoma; FTC-OV, FTC-oncocytic variant; PTC, papillary thyroid carcinoma; PTC-OV, PTC-oncocytic variant; FVPTC, follicular variant of a papillary thyroid carcinoma; HP, hyperplasia
ªfrom FTC-OV
“same tumor
6
NEAR-HAPLOIDIZATION IN ONCOCYTIC TUMORS – – – – – – – – – – 1 1
(HM) to confirm the diagnosis and tumor cell per- centage. All sections consisted of at least 70% tumor cells. Representative pictures were taken using a 63X objective and in-house developed software (PathViewer, JO) with a Leica DM600B light microscope equipped with the Leica DFC280 digital camera.
DNA Content Analysis by Flow Cytometry
To enrich for tumor cells, tissue punches (2 mm diameter) were taken from the FFPE tissue blocks using a marked HE slide from the same block. The DNA content was measured by multipara- meter DNA flow cytometry. Cell suspensions were prepared from 2-mm-diameter FFPE punches and stained for vimentin (allophycocyanin (APC) fluo- rescence), keratin (fluorescein isothiocyanate (FITC) fluorescence), and DNA (propidium iodide fluorescence). The protocol has been described in detail elsewhere (Corver et al., 2005) and showed to be robust in the flow cytometric analysis of archival carcinomas. However, anaplastic thyroid carcinomas (ATC) frequently show dedifferentia- tion with decreasing keratin expression. Further- more, most adrenocortical carcinomas are negative for keratin. Both tumor types show high vimentin expression comparable to the stromal cells impair- ing clear discrimination between tumor cells and stromal cells by the method mentioned above. Therefore, we labeled cell suspensions from these tumors for DNA and for CD34 Class II (0.02 µg, clone QBEnd 10, product M7165, Dako, Glostrup) as a marker for DNA diploid endothelial cells.
An LSRII flow cytometer (BD Biosciences, San Jose, CA) was used for data acquisition. The blue 488 nm and the red 635 nm laser were used for excitation. Data files containing information from at least 50,000 single cell events were analyzed using ModFit 3.3, remotely linked to WinList 7.0 (Verity software House, Topsham, ME).
Cell Culture
XTC.UC1 cells, passage p122, were kindly pro- vided by Prof. Orlo Clark, MD, UCSF Compre- hensive Cancer Center, San Francisco, CA. Cells were maintained in our facility under standard cul- ture conditions in DMEM-F12 GlutaMAX™M (Gibco®/Life Technologies, Paisly, UK) culture medium, supplemented with 10% fetal calf serum and 10,000 U/ml Penicillin/Streptomycin (Gibco®/ Life Technologies). Cells were harvested with trypsin/EDTA and washed with HBSS (5 min,
500Xg). The cell pellet was immediately frozen and stored for later use.
DNA Isolation
SNP array and mtDNA mutation analysis was con- ducted using high molecular weight DNA. Three to six 10 um sections were prepared from snap-frozen tissue samples using a cryostat and stored in a pre- cooled micro vessel at -20℃. XTC.UC1 DNA was extracted from frozen cell pellets.
For somatic mutation analysis in FFPE tissues, three 0.6-mm-diameter tissue punches (Beecher Instruments, USA) were taken from selected tumor areas (HM) to enrich for tumor DNA. Tissue punches were then dewaxed prior to DNA isola- tion, and tissue treated with overnight digestion with proteinase K at 56℃. DNA from frozen and FFPE samples was extracted using the NucleoSpin purification kit (Macherey-Nagel GmbH & Co. KG, Düren, BRD) according to the manufacturer’s instructions. DNA concentrations were determined using the Picogreen method (Invitrogen).
Genome-Wide SNP Array Analysis
Two types of SNP array were used for genomic profiling: 23 samples were hybridized to ICOG iSelect genotyping arrays (Illumina, San Diego, CA) and 37 to the HumanCytoSNP-12 BeadChip (Illumina, San Diego, CA)
HumanCytoSNP-12 BeadChip incorporates over 220,000 single nucleotide polymorphisms (SNPs) across the genome. The ICOG arrays are custom Illumina iSelect genotyping arrays that comprise over 200,000 SNPs, drawn from previous genome-wide association studies by the Collabora- tive Oncological Gene-environment Study (Michailidou et al., 2013). The two platforms were evaluated in-house, showed similar results and did not impair the clear discrimination between a NHG and a genome containing more complex alterations. All arrays were run by a genomics serv- ice provider (ServiceXS, Leiden, Netherlands). The arrays were hybridized according to the man- ufacturers recommendations with high molecular weight DNA extracted from snap-frozen samples. The arrays were analyzed as described previously (Corver et al., 2008), using the beadarraySNP package (Oosting et al., 2007).
Mitochondrial DNA (mtDNA) Sequencing
For Sanger sequencing of mtDNA, 24 primer pairs (Eurofins MWG Operon, Ebersberg, Germany)
were chosen from the MitoAll application note (Life Technologies), based on the work of (Bonora et al., 2006). The presence or absence of a previously reported “common mtDNA deletion of 4977 bp” was also studied using earlier described primer pairs; that is, one pair outside and one pair amplifying this deletion (Duregon et al., 2011). All primers (Sup- porting Information Table S1) were blasted against the National Centre for Biotechnology Information (NCBI) mitochondrial reference genome NC_012920 and a 100% match was seen for all pri- mers. M13 tails were added to the primers for uni- versal sequencing. The amplicons ranged between 763 and 1117 bp, except for the control amplicon for the common deletion (474 bp). Each successive PCR product overlapped the previous amplicon by approximately 150 bp. PCR was performed in a Real-Time PCR Detection System (CFX96, Bio- Rad, Veenendaal, the Netherlands) in 20 ul reactions with 10 ng DNA, iQ Supermix (Bio-Rad) and 2 pmol primers, as described (van Eijk et al., 2011). The PCR conditions were 5 min at 95℃, 40 cycles of 10 sec at 95℃, 10 sec at 60℃, and 90 sec at 72°℃, with a final elongation step of 10 min at 72°C. After this final elongation step, a melt curve was obtained to evaluate the quality of the PCR prod- ucts. Purified PCR products were Sanger sequenced in both directions at MacroGen (Amsterdam, Neth- erlands) and the mtDNA sequences were analyzed using Mutation Surveyor™M version 4.0.8 (MS) (Soft- genetics, State College, PA). Variants were called against a GenBank reference sequence of the mito- chondrial genome, NC_012920, and checked man- ually. Next, coding missense mutations were selected and analyzed in silico using the Polyphen V2 package (Adzhubei et al., 2010). PolyPhen V2 predicts the possible impact of amino acid substitu- tions on the stability and function of human pro- teins. The software provides an arbitrary value between 0 and 1. A value <0.5 was predicted to be benign, 0.5 ≥ x ≤ 0.9 was predicted as possibly dam- aging and >0.9 was probably damaging. We com- bined the possibly damaging and probably damaging into a single category of “damaging variants.” The static correlation between the benign OV/damaging OV versus benign non-OV/damaging non-OV was calculated using the two-tailed Fisher exact test (GraphPad Prism 6).
Somatic Mutation Analysis
PCR assays were used to detect seven different KRAS substitutions in codons 12 and 13 (p.G12S, p.G12R, p.G12C, p.G12D, p.G12A, p.G12V, and
p.G13D), NRAS (p.G12D, p.Q61H, p.Q61K, p.Q61L, and p.Q61R), three PIK3CA variants (p.E542K, p.E545K, and p.H1047R), one BRAF (p.V600E) and two EGFR variants (p.L858R) and a deletion in exon 19.
The total reaction volume of the PCR was 10 ul, including 5 ul of FastStart Universal Probe Master (Roche Applied Science), 1 ul of 10X primer and hydrolysis probe solutions and 10 ng of DNA. The PCR was performed in a real-time PCR detection system (CFX384, Bio-Rad) as fol- lows: 10 min 95℃, followed by 40 cycles of 10 sec at 92℃ and 30 sec at 60℃.
PCR amplification and subsequent Sanger sequencing was performed on HRAS (exon 1), IDH1 (exon 4 and 6), and IDH2 (exon 4) as described above but with adapted PCR conditions due to the smaller size of the PCR products: 32 cycles of 10 sec at 95°℃, 10 sec at 60℃, and 10 sec at 72°C.
RESULTS
Identification of Thyroid, Adrenal Cortical and Parathyroid Tumors with a NHG.
Sixty endocrine tumor samples (thyroid-, parathyroid-, and adrenal cortical carcinomas) were studied, for which snap-frozen and FFPE tissue was available. Twelve tumors, including six ACC (Nos. 3, 7-11), one parathyroid cancer (PaTC; No. 14), and five thyroid cancers (Nos. 15, 22, 39-41) showed a clear NHG (Table 1), with many entire chromosomes in a homozygous state (allelic state [A], [AA], or even [AAA], depending on the DI). Of the 12 tumors with a NHG, 92% (11/12) also showed an oncocytic phenotype (Nos. 7-11, 14, 39-41) or originated from an oncocytic tumor (Nos. 15 and 22). The presence of a NHG corre- lated significantly with an oncocytic phenotype (P=0.0001, two-tailed Fisher’s exact, Supporting Information Table S3). The one nononcocytic tumor with a NHG was a pancreatic metastasis of an ACC (No. 3).
Comparable to our earlier findings in FTC-OV, two of the six ACCs (Nos. 7 and 8) with a NHG showed a bimodel DNA histogram with two tumor populations, one representing a clear near-haploid population (with DI’s of 0.7) and a second repre- senting a population with DI’s of 1.2 and 1.3, respectively, that appears to have undergone endoreduplication. Two other ACC’s with a NHG [Nos. 3 and 10 (No. 10, Fig. 1A)] showed a single tumor population (with DI’s of 1.3), most likely due to complete endoreduplication of a previous
A: No 10, ACC-OV
100 200 300 400 500 600 700
DNA diploid
Number
DI = 1.3
50 um
0
0
500 1000 1500 2000 2500
DNA PI
(x 100)
AAª
A AA
AABB AABB
AA
AA
¡¡
iii
AA
AA
1
2
3
22
AA
AABB
AA
AA
AA
AA
5
6
7
8
9
AAABB
AA
AA
AA
AABB
AABB
AA
AA
AA AABB
10
11
12
13
14
15
21
AABB
AABB
AA
AA
AA
AABB
AAA
4
16
17
18
19
20
x ☒
Y ☒
1
0
1
0
1
0
1
0
B: No 14, PaTC-OV
50 100 150 200 250 300 350
DNA diploid
Number
DI = 1.2
DI = 2.3
50 um
0
0
500 1000 1500 2000 2500
DNA PI
(x 100)
AAİ
AA
AA
A
ii
iii
1
2
3
22
AABB
AA
AABB
AA
AABB
5
6
7
8
9
AAAA
AA
AAA
AA
AA
AA
AA
AA
10
11
12
13
14
15
21
AA
AA
AA
AA
AAB
AAAB
AAA
Y ☒
4
16
17
18
19
20
x ☒
near-haploid chromosomal state. Strikingly, the remaining two ACC-OV (Nos. 9 and 11) with NHG showed very high DI’s of 3.9 and 2.6,
population), respectively. Green: DNA diploid stromal cells (reference peak), red: tumor cells. Lower right genome-view: The annotation of the chromosomes is based on the major population. Only chromo- somes 5, 7, and 9 are in a heterozygous state. Chromosomes 19 and 20 are imbalanced. All other chromosomes are strict homozygous, except for a small fragment on chromosome 17, which is heterozy- gous. i = allelic state, ii = copy number, iii = 0 <allelic ratio < 1; A = homozygous, one copy; AA = homozygous, two copies; and so forth; AB = heterozygous, two copies; AABB = heterozygous, four copies; AAB = imbalance, three copies; AAAB = imbalance, four cop- ies. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
respectively, which can only be explained by sev- eral rounds of endoreduplication of an entire near- haploid genome.
A: No 39, FTC-OV
B: No 15, ATC out of FTC-OV
250
DNA near-haploid
0
500
DNA near-haploid
2000
DI = 0.7
DI = 0.6
400
Number
1500
DNA diploid
Number
DNA diploid
300
1000
DI = 1.4
200
DI = 1.1
500
50 um
clumps
50 pim
100
0
0
0
500
1000
1500
2000
2500
0
500
1000
1500
2500
DNA PI
DNA PI
2000
(x 100)
(x 100)
Aª
A
A
AA
AAAª
AA
AA
AA
ii
ii
ili
iii
1
0
1
2
3
22
1
2
3
22
AB
A
AB
A
A
AAAA
AA
AA
AABB
AAB
AA
AA
1
0
5
6
7
8
9
5
6
7
8
9
AB
A
AB
AB
A
A
AA
AA
AA
AA
AA
AA
AA
AA
1
0
10
11
12
13
14
15
21
10
11
12
13
14
15
21
A
A
AB
AA
AB
AB
AB
A
AA
AA
AA
AAA
AA
AA
AA
AAA
Y, ☒
1
0
4
16
17
18
19
20
x ☒
4
16
17
18
19
20
x ☒
Y ☒
The only PaTC-OV with a NHG (No. 14, HRPT2|CDC73 wild type) showed a major G1 pop- ulation with DI of 1.2, exactly in the range of an
right picture, magnification 63X), a NHG and a DNA histogram show- ing two tumor populations with DI’s 0.6 (minor population) and I.I (major population), respectively. Green: DNA diploid stromal cells (ref- erence peak), red: tumor cells. Lower right genome-view: the annota- tion of the chromosomes is based on the major population. Only chromosome 7p is in a heterozygous state. Chromosome 7q shows an imbalance. All other chromosomes are strict homozygous. i = allelic state, ii = copy number, iii = 0 < allelic ratio < 1; A = homozygous, one copy; AA = homozygous, two copies; and so forth; AB = heterozygous, two copies; AABB = heterozygous, four copies; AAB = imbalance, three copies; AAAB = imbalance, four copies. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
endoreduplicated near-haploid genome (Fig. 1B). In this case, chromosomes 5, 7, 9, and a small seg- ment of 17p were retained (heterozygous, allelic
59 tumors 1569 variants
50.6% non-coding variants (n = 794)
49.4% coding variants (n = 775)
7.7% tRNA (n = 61) 27.6% RNR1/RNR2 (n = 219) 63.9% D-loop (n = 507) 0.9% VUS (n = 7)
55.1% synonymous (n = 427)
43.5% non-synonymous (n = 336)
1.4% Disruptive (n = 12) In 10 tumors
damaging (n = 46) in 41 tumors
benign (n = 290)
state [AABB]), while chromosomes 19 and 20 showed an imbalance. All other chromosomes were homozygous ([A] or [AA]).
NHG was also observed in FTC-OV. Three FTC-OVs with NHG (Nos. 39-41) showed a DNA histogram with a near-haploid population (DI’s between 0.6 and 0.7). The histograms of FTC-OV No. 39 also showed a second population with a DI of 1.4 (as a result of endoreduplication; Fig. 2A). Two anaplastic thyroid cancers (ATC, Nos. 15 and 22) that derived from an FTC-OV also showed a clear NHG, with two subpopulations with a DI of 0.6 and 1.1 (Fig. 2B), and 0.5 and 1.0, respectively, again pointing to endoreduplication.
Highly reminiscent of our previous study, NHG-thyroid tumors (Nos. 15, 22, 39-41) showed retention of chromosome 7 in the allelic states [AB] or [AABB], or [AAAABB] for 7q (No 15 only), respectively. Chromosome 12 was the sec- ond most commonly retained chromosome and 12p and 12q were seen in a heterozygous state (allelic states [AB] or [AABB]) in four and three of the five cases, respectively. Chromosome five was retained in three cases (Nos. 39-41). In one FTC- OV (No 38), a NHG was not detected and showed a pseudo-diploid DNA histogram and a corre- sponding genome with very limited number of chromosomal aberrations. No unambiguous NHG was detected in any of the other thyroid tumors.
In contrast to FTC-OV, ACC-OV with NHG commonly showed loss of chromosome 7 (with allelic states [A] or [AA]; Nos. 7-11) but chromo- somes 4, 5, and 12 were mostly retained. Overall, ACC-OV clearly showed more chromosomal
used as mitochondrial reference sequence. Coding missense mutations were analyzed in silico using Polyphen V2, which assigns an arbitrary value between 0 and I. The value <0.5 equals benign, 0.5 >x≤0.9 equals possibly damaging and >0.9 equals probably damaging. In this study, possibly damaging and probably damaging were defined as dam- aging variants.
breakpoints than FTC-OV. Consequently, the NHG pattern was sometimes less pronounced. NHGs were not found in any of the remaining 48 samples.
Mitochondrial DNA Mutation Analysis
We initially sequenced the mitochondrial DNA (mtDNA) of XTC.UC1 cells and could confirm the reported mtDNA mutations and variants, including the 3507Cins in MT-ND1 and the dam- aging mutation in MT-CYB (15557G>A, E271K; Bonora et al., 2006).
We then sequenced the entire mtDNA of all 60 samples under study using Sanger sequencing. In total, 58 damaging/disruptive mtDNA mutations were identified of which 47 (81%) were homoplas- mic. The remaining 290 (86.3%) nonsynonymous coding mtDNA variants were benign (PolyPhen-2 prediction <0.5). Figure 3 shows the division of all mtDNA mutations found in our cohort. The distri- bution of the damaging/disruptive mutations found in the different histological subtypes is shown in Table 1. No significant correlation was found between mtDNA damaging and/or disrup- tive mutations (insertions/deletions and stop-gains mutations) and oncocytic tumors, histological sub- types, or NHG tumors (Supporting Information Table S4). Finally, we studied a previously reported common mtDNA deletion (Duregon et al., 2011), first confirming the deletion by real- time PCR. There was also no correlation between the presence of the common mtDNA deletion and any of the histological parameters.
Somatic Nuclear Gene Mutation Analysis
Certain subsets of nuclear encoded genes that code for proteins prominently involved in signal transduction pathways have been implicated in metabolic switching. We studied components of the MAPK/ERK and PIK3CA/AKT pathways, and IDH1/2 mutations. All mutations found are sum- marized in Supporting Information Table S5. BRAF p.V600E mutations were found in 22% (12/ 59) of the tumors and, as expected, were most prevalent in papillary thyroid cancer (PTC; 56%, 10/16) (Nos. 47-49, 51-54, 55, 56, and 59). No mutations were found in EGFR (exons 18-21) or in PIK3CA (p.E542K, p.E545K, and p.H1047R). RAS/RAF mutations did not correlate with onco- cytic tumors or NHG tumors. A detailed overview of all individual patients and the corresponding findings are shown in Supporting Information Table S6.
DISCUSSION
A NHG is a relatively rare phenomenon in can- cer (Mandahl et al., 2012) and has been described in chondrosarcoma (Bovee et al., 2000), acute lym- phoblastic leukemia (Safavi et al., 2013; Holmfeldt et al., 2013), chronic myeloid leukemia (Andersson et al., 1987; Gancberg et al., 2001) and has occa- sionally been found in plasmocytoma (Kristoffers- son et al., 1986), malignant fibrous histiocytoma (Aspberg et al., 1995) and peritoneal mesothelioma (Sukov et al., 2010). More recently, we showed that a NHG is a common feature of the oncocytic variant of follicular thyroid carcinoma (FTC-OV; Corver et al., 2012). Here we show that a NHG can also be found in Pa’TC and in ACC. Strikingly, almost all endocrine tumors with a NHG show an oncocytic morphology. Just recently, similar results were observed in >30% of ACC, although different terminology was used and the phenotype of the tumors was not further specified (Assie et al., 2014). However, not all oncocytic tumors show a NHG. Apparently, a subset of oncocytic tumors exists in which a currently unexplained underlying biology leads to a NHG phenotype. In addition, it was remarkable that the only PaTC with a NHG was the one lacking a HRPT2|CDC73 mutation (Carpten et al., 2002), the gene impli- cated in sporadic and familial PaTC, although this isolated example precludes any definitive conclusions.
XTC.UC1 cells, a cell line obtained from an oncocytic follicular thyroid carcinoma (Zielke et al., 1998), are known to harbor a homoplastic
disruptive mtDNA mutation in MT-ND1 (3507Cins, stop codon) and a damaging mutation in MT-CYB (m.15557G>A, p.E271K; Bonora et al., 2006). Probably as a result of these mutations in complex I and III, respectively, XTC.UC1 cells show metabolic alterations that include impaired ATP production and glucose-dependent growth due to aerobic glycolysis [the Warburg effect (Warburg et al., 1927)]. This lack of ATP might cause a coupling defect (Savagner et al., 2001b), triggering a mitochondrial proliferation that results in the typical cytoplasmic accumulation of mito- chondria responsible for the oncocytic phenotype. This observation suggests the possibility of an association between mtDNA mutations and the oncocytic phenotype in clinical samples. Gasparre et al. (2007) performed a large study on oncocytic versus nononcocytic tumors and reported that the oncocytic phenotype indeed significantly corre- lates with disruptive mtDNA mutations, mainly in subunit MT-ND1 of complex I.
In the present study, we wished to extend this investigation to include our observation of a NHG in predominantly endocrine tumors of the onco- cytic type and a possible relation between damag- ing/disruptive mtDNA mutations and NHG. However, we were not able to confirm earlier find- ings (Maximo et al., 2002; Gasparre et al., 2007). Furthermore, mtDNA mutations did not correlate with NHG. In addition, we studied D-loop muta- tions in relation to the oncocytic phenotype and found results comparable to those of Maximo et al. (2005), who showed that D-loop mutations are not a marker for malignancy in thyroid tumors. Over- all, damaging and/or disruptive mtDNA mutations in coding regions were approximately equally dis- tributed in oncocytic and nononcocytic thyroid tumors (Supporting Information Table S4), further strengthening the argument for no direct relation between mtDNA mutation and the oncocytic phe- notype. While these findings might differ from the results reported by Gasparre et al. (2007), they are in line with results reported by other groups. A recent study described the spectrum of mtDNA mutations in five nonendocrine cancers (Larman et al., 2012). Functionally damaging and disruptive mutations were mainly found in colon adenocarci- noma and rectal adenocarcinoma. However, onco- cytic adenocarcinoma of the colon or rectum are rarely found (Chetty et al., 2009), though these may be observed following preoperative radiation (Ambrosini-Spaltro et al., 2006; Rouzbahman et al., 2006). Taken together, these observations contradict the conclusions of Gasparre et al.
(2007). In the latter study, in addition to the 66 thyroid tumors (45 oncocytic, 21 nononcocytic) and 20 breast cancer tumors (five oncocytic, 15 nononcocytic), 16 nononcocytic gliomas were included. In a detailed reexamination of the data reported by Gasparre et al., excluding the breast tumors and gliomas, we were unable to find a sig- nificant correlation between the oncocytic pheno- type and disruptive or damaging/disruptive mtDNA mutations in the remaining 66 thyroid tumors (P= 0.1949 and 0.1876, respectively, two- tailed Fisher’s exact, Supporting Information Table S7). We now conclude that our data con- firms and supports the actual data in the Gasparre et al. study (although not their interpretation of it) that there is no direct correlation between damag- ing/disruptive mtDNA mutations and the onco- cytic phenotype in thyroid tumors. The work of Zimmermann et al., in which they sequenced the mitochondrially encoded subunits of complex I and found no potentially damaging mutations in eight of 19 oncocytic thyroid tumors, also supports this conclusion (Zimmermann et al., 2009).
Interestingly, a significant correlation between the strongly downregulated expression of the nuclear-encoded complex I subunit NDUFS4 (5q11.1) and oncocytic tumors has been reported, leading to the proposal that complex I is a tumor suppressor in oncocytic tumors (Zimmermann et al., 2011). This may indicate that additional fac- tors, for example, nuclear-encoded subunits of the respiratory chain, determine the oncocytic pheno- type, either with or without a role for mtDNA mutations.
In addition to detection of damaging/disruptive mtDNA mutations, we also studied a previously identified “common mtDNA4977 deletion” in our cohort. However, our Cq results in real-time PCR (Supporting Information Table S6) indicate that the detection of this deletion is either a technical artifact or that the deletion is only present in a minor fraction of the mitochondria. No correlation was found between the oncocytic phenotype as the mutation was approximately equally distrib- uted between oncocytic and nononcocytic tumors, comparable to the study by Savagner et al. in which the mtDNA4977 deletion was only found in two of the 22 oncocytic thyroid tumors and was simultaneously observed in controls (Savagner et al., 2001b). Taken together, these results indi- cate that the presence of this deletion cannot explain the oncocytic phenotype or a NHG.
Molecular components of the PIK3CA-RAS/ RAF pathways are upstream of the m’TOR path-
way, which is involved in energy metabolism (Laplante and Sabatini, 2012). DNA mutations in components of the PIK3CA-RAS/RAF pathways might, therefore, play a role in phenotypes related to metabolic alterations. For instance, a significant proportion of the p.V600E mutated BRAF translo- cate to the mitochondria and have been implicated in a reduction in oxidative phosphorylation (Lee et al., 2011). As previously reported, we found the BRAF p.V600E mutation mainly in PTC, but we also found this mutation in one ACC and two ATCs without oncocytic features. Thus, despite the mitochondrial binding properties of mutated BRAF it did not correlate with oncocytic or NHG phenotypes, even in the presence of disruptive mtDNA mutations. Similar observations were made for RAS mutations. Activating IDH1/2 muta- tions lead to metabolic alterations in subsets of gliomas and chondrosarcomas. Only two endocrine tumors in our cohort carried an activating IDH mutation (one FVPTC without oncocytic metapla- sia, carrying also a damaging MT-CO1 mutation; and an ATC with oncocytic features) and neither showed a NHG.
In conclusion, we demonstrate for the first time that a NHG, previously observed in a subset of oncocytic thyroid tumors (FTC-OV), is also com- monly found in oncocytic ACC and oncocytic parathyroid cancer. Like FTC-OV, ACC-OV, and PaTC-OV frequently show DNA near-haploidy and/or show endoreduplication (masked near-hap- loidy) of the entire near-haploid genome. Overall, the correlation between a NHG and the oncocytic phenotype was highly significant for these endo- crine tumors, although not all oncocytic tumors show a NHG. In a previous publication, we dis- cussed the possibility of a relationship between mtDNA mutations, decreased ATP production and a NHG (Corver et al., 2012). However, we found no evidence for this hypothesis in the pres- ent study. No association was found between a NHG and damaging and/or disruptive mtDNA mutations, even in the presence of RAS/RAF mutations known to be involved in tumor metabo- lism. A lack of ATP might still underlie the pro- gression to a NHG but other or additional factors are likely to also play a role. XTC.UC1 cells as well as other cell lines can serve as a model for NHG (manuscript in preparation). These cell line models might be used to study lagging chromo- somes and missegregation errors, which can under- lie the molecular mechanism of the near- haploidization process leading to a NHG with respect to the oncocytic phenotype.
REFERENCES
Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P, Kondrashov AS, Sunyaev SR. 2010. A method and server for predicting damaging missense mutations. Nat Meth- ods 7:248-249.
Ambrosini-Spaltro A, Salvi F, Betts CM, Frezza GP, Piemontese A, Del PP, Baldoni C, Foschini MP, Viale G. 2006. Oncocytic modifications in rectal adenocarcinomas after radio and chemo- therapy. Virchows Arch 448:442-448.
Andersson BS, Beran M, Pathak S, Goodacre A, Barlogie B, McCredie KB. 1987. Ph-positive chronic myeloid leukemia with near-haploid conversion in vivo and establishment of a continuously growing cell line with similar cytogenetic pattern. Cancer Genet Cytogenet 24:335-343.
Aspberg F, Mertens F, Bauer HC, Lindholm J, Mitelman F, Mandahl N. 1995. Near-haploidy in two malignant fibrous his- tiocytomas. Cancer Genet Cytogenet 79:119-122.
Assie G, Letouze E, Fassnacht M, Jouinot A, Luscap W, Barreau O, Omeiri H, Rodriguez S, Perlemoine K, Rene-Corail F, Elarouci N, Sbiera S, Kroiss M, Allolio B, Waldmann J, Quinkler M, Mannelli M, Mantero F, Papathomas T, de Krijger R, Tabarin A, Kerlan V, Baudin E, Tissier F, Dousset B, Groussin L, Amar L, Clauser E, Bertagna X, Ragazzon B, Beuschlein F, Libe R, de Reynies A, Bertherat J. 2014. Inte- grated genomic characterization of adrenocortical carcinoma. Nat Genet 46:607-612.
Baris O, Savagner F, Nasser V, Loriod B, Granjeaud S, Guyetant S, Franc B, Rodien P, Rohmer V, Bertucci F, Birnbaum D, Malthiery Y, Reynier P, Houlgatte R. 2004. Transcriptional profiling reveals coordinated up-regulation of oxidative metabo- lism genes in thyroid oncocytic tumors. J Clin Endocrinol Metab 89:994-1005.
Bonora E, Porcelli AM, Gasparre G, Biondi A, Ghelli A, Carelli V, Baracca A, Tallini G, Martinuzzi A, Lenaz G, Rugolo M, Romeo G. 2006. Defective oxidative phosphorylation in thyroid oncocytic carcinoma is associated with pathogenic mitochondrial DNA mutations affecting complexes I and III. Cancer Research 66:6087-6096.
Bovee JV, van Royen M, Bardoel AF, Rosenberg C, Cornelisse CJ, Cleton-Jansen AM, Hogendoorn PC. 2000. Near-haploidy and subsequent polyploidization characterize the progression of peripheral chondrosarcoma. Am J Pathol 157:1587-1595.
Carpten JD, Robbins CM, Villablanca A, Forsberg L, Presciuttini S, Bailey-Wilson J, Simonds WF, Gillanders EM, Kennedy AM, Chen JD, Agarwal SK, Sood R, Jones MP, Moses TY, Haven C, Petillo D, Leotlela PD, Harding B, Cameron D, Pannett AA, Hoog A, Heath H, III, James-Newton LA, Robinson B, Zarbo RJ, Cavaco BM, Wassif W, Perrier ND, Rosen IB, Kristoffersson U, Turnpenny PD, Farnebo LO, Besser GM, Jackson CE, Morreau H, Trent JM, Thakker RV, Marx SJ, Teh BT, Larsson C, Hobbs MR. 2002. HRPT2, encoding parafibro- min, is mutated in hyperparathyroidism-jaw tumor syndrome. Nat Genet 32:676-680.
Cervera AM, Bayley JP, Devilee P, McCreath KJ. 2009. Inhibition of succinate dehydrogenase dysregulates histone modification in mammalian cells. Mol Cancer 8:89.
Chetty R, Serra S, Kennedy E, Govender D. 2009. Oncocytic rec- tal adenocarcinomas. Hum Pathol 40:478-483.
Corver WE, Ter Haar NT, Dreef EJ, Miranda NF, Prins FA, Jordanova ES, Cornelisse CJ, Fleuren GJ. 2005. High-resolution multi-parameter DNA flow cytometry enables detection of tumour and stromal cell subpopulations in paraffin-embedded tissues. J Pathol 206:233-241.
Corver WE, Middeldorp A, Ter Haar NT, Jordanova ES, van Puijenbroek M, van Eijk R, Cornelisse CJ, Fleuren GJ, Morreau H, Oosting J, van Wezel T. 2008. Genome-wide allelic state analysis on flow-sorted tumor fractions provides an accurate mea- sure of chromosomal aberrations. Cancer Res 68:10333-10340.
Corver WE, Ruano D, Weijers K, den Hartog WC, van Nieuwenhuizen MP, de Miranda N, van Eijk R, Middeldorp A, Jordanova ES, Oosting J, Kapiteijn E, Hovens G, Smit J, van Wezel T, Morreau H. 2012. Genome haploidisation with chro- mosome 7 retention in oncocytic follicular thyroid carcinoma. PLoS ONE 7:e38287.
Duregon E, Volante M, Cappia S, Cuccurullo A, Bisceglia M, Wong DD, Spagnolo DV, Szpak-Ulczok S, Bollito E, Daffara F, Berruti A, Terzolo M, Papotti M. 2011. Oncocytic adrenocorti- cal tumors: Diagnostic algorithm and mitochondrial DNA pro- file in 27 cases. Am J Surg Pathol 35:1882-1893.
Feng Z, Levine AJ. 2010. The regulation of energy metabolism and the IGF-1/mTOR pathways by the p53 protein. Trends Cell Biol 20:427-434.
Gaglio D, Metallo CM, Gameiro PA, Hiller K, Danna LS, Balestrieri C, Alberghina L, Stephanopoulos G, Chiaradonna F. 2011. Oncogenic K-Ras decouples glucose and glutamine metabolism to support cancer cell growth. Mol Syst Biol 7:523. Gancberg D, Dargent JL, Verhest A, Kentos A, Feremans W, Lohrisch C, Kornreich A. 2001. Near haploid blast phase in a chronic myeloid leukemia detected by fluorescence in situ hybridization using a BCR-ABL probe. Cancer Genet Cytoge- net 128:172-174.
Gasparre G, Porcelli AM, Bonora E, Pennisi LF, Toller M, Iommarini L, Ghelli A, Moretti M, Betts CM, Martinelli GN, Ceroni AR, Curcio F, Carelli V, Rugolo M, Tallini G, Romeo G. 2007. Disruptive mitochondrial DNA mutations in complex I subunits are markers of oncocytic phenotype in thyroid tumors. Proc Natl Acad Sci USA 104:9001-9006.
Hanahan D, Weinberg RA. 2011. Hallmarks of cancer: The next generation. Cell 144:646-674.
Hardie DG, Alessi DR. 2013. LKB1 and AMPK and the cancer- metabolism link-Ten years after. BMC Biol 11:36.
Hemerly JP, Bastos AU, Cerutti JM. 2010. Identification of several novel non-p.R132 IDH1 variants in thyroid carcinomas. Eur J Endocrinol 163:747-755.
Holmfeldt L, Wei L, Diaz-Flores E, Walsh M, Zhang J, Ding L, Payne-Turner D, Churchman M, Andersson A, Chen SC, McCastlain K, Becksfort J, Ma J, Wu G, Patel SN, Heatley SL, Phillips LA, Song G, Easton J, Parker M, Chen X, Rusch M, Boggs K, Vadodaria B, Hedlund E, Drenberg C, Baker S, Pei D, Cheng C, Huether R, Lu C, Fulton RS, Fulton LL, Tabib Y, Dooling DJ, Ochoa K, Minden M, Lewis ID, To LB, Marlton P, Roberts AW, Raca G, Stock W, Neale G, Drexler HG, Dickins RA, Ellison DW, Shurtleff SA, Pui CH, Ribeiro RC, Devidas M, Carroll AJ, Heerema NA, Wood B, Borowitz MJ, Gastier-Foster JM, Raimondi SC, Mardis ER, Wilson RK, Downing JR, Hunger SP, Loh ML, Mullighan CG. 2013. The genomic landscape of hypodiploid acute lymphoblastic leuke- mia. Nat Genet 45:242-252.
Kim HJ, Shen SS, Ayala AG, Ro JY, Truong LD, Alvarez K, Bridge JA, Gatalica Z, Hagenkord JM, Gonzalez-Berjon JM, Monzon FA. 2009. Virtual-karyotyping with SNP microarrays in morphologically challenging renal cell neoplasms: A practical and useful diagnostic modality. Am J Surg Pathol 33:1276-1286. Kristoffersson U, Olsson H, Kelly D, Akerman M, Mitelman F. 1986. Near-haploidy in a case of plasmocytoma. Cancer Genet Cytogenet 19:239-243.
Laplante M, Sabatini DM. 2012. mTOR signaling in growth con- trol and disease. Cell 149:274-293.
Larman TC, DePalma SR, Hadjipanayis AG, Protopopov A, Zhang J, Gabriel SB, Chin L, Seidman CE, Kucherlapati R, Seidman JG. 2012. Spectrum of somatic mitochondrial muta- tions in five cancers. Proc Natl Acad Sci USA 109:14087-14091. Lee MH, Lee SE, Kim DW, Ryu MJ, Kim SJ, Kim SJ, Kim YK, Park JH, Kweon GR, Kim JM, Lee JU, De Falco V, Jo YS, Shong M. 2011. Mitochondrial localization and regulation of BRAFV600E in thyroid cancer: A clinically used RAF inhibitor is unable to block the mitochondrial activities of BRAFV600E. J Clin Endocrinol Metab 96:E19-E30.
Mandahl N, Johansson B, Mertens F, Mitelman F. 2012. Disease- associated patterns of disomic chromosomes in hyperhaploid neoplasms. Genes Chromosomes Cancer 51:536-544.
Maximo V, Lima J, Prazeres H, Soares P, Sobrinho-Simoes M. 2012. The biology and the genetics of Hurthle cell tumors of the thyroid. Endocr Relat Cancer 19:R131-R147.
Maximo V, Soares P, Lima J, Cameselle-Teijeiro J, Sobrinho- Simoes M. 2002. Mitochondrial DNA somatic mutations (point mutations and large deletions) and mitochondrial DNA variants in human thyroid pathology: A study with emphasis on Hürthle cell tumors. Am J Pathol 160:1857-1865.
Maximo V, Lima J, Soares P, Botelho T, Gomes L, Sobrinho- Simoes M. 2005. Mitochondrial D-Loop instability in thyroid tumours is not a marker of malignancy. Mitochondrion 5:333-340. Mayr JA, Meierhofer D, Zimmermann F, Feichtinger R, Kogler C, Ratschek M, Schmeller N, Sperl W, Kofler B. 2008. Loss of complex I due to mitochondrial DNA mutations in renal onco- cytoma. Clin Cancer Res 14:2270-2275.
Michailidou K, Hall P, Gonzalez-Neira A, Ghoussaini M, Dennis J, Milne RL, Schmidt MK, Chang-Claude J, Bojesen SE, Bolla MK, Wang Q, Dicks E, Lee A, Turnbull C, Rahman N,
Fletcher O, Peto J, Gibson L, Dos SS, I, Nevanlinna H, Muranen TA, Aittomaki K, Blomqvist C, Czene K, Irwanto A, Liu J, Waisfisz Q, Meijers-Heijboer H, Adank M, van der Luijt RB, Hein R, Dahmen N, Beckman L, Meindl A, Schmutzler RK, Muller-Myhsok B, Lichtner P, Hopper JL, Southey MC, Makalic E, Schmidt DF, Uitterlinden AG, Hofman A, Hunter DJ, Chanock SJ, Vincent D, Bacot F, Tessier DC, Canisius S, Wessels LF, Haiman CA, Shah M, Luben R, Brown J, Luccarini C, Schoof N, Humphreys K, Li J, Nordestgaard BG, Nielsen SF, Flyger H, Couch FJ, Wang X, Vachon C, Stevens KN, Lambrechts D, Moisse M, Paridaens R, Christiaens MR, Rudolph A, Nickels S, Flesch-Janys D, Johnson N, Aitken Z, Aaltonen K, Heikkinen T, Broeks A, Veer LJ, Van der Schoot CE, Guenel P, Truong T, Laurent-Puig P, Menegaux F, Marme F, Schneeweiss A, Sohn C, Burwinkel B, Zamora MP, Perez JI, Pita G, Alonso MR, Cox A, Brock IW, Cross SS, Reed MW, Sawyer EJ, Tomlinson I, Kerin MJ, Miller N, Henderson BE, Schumacher F, Le ML, Andrulis IL, Knight JA, Glendon G, Mulligan AM, Lindblom A, Margolin S, Hooning MJ, Hollestelle A, van den Ouweland AM, Jager A, Bui QM, Stone J, Dite GS, Apicella C, Tsimiklis H, Giles GG, Severi G, Baglietto L, Fasching PA, Haeberle L, Ekici AB, Beckmann MW, Brenner H, Muller H, Arndt V, Stegmaier C, Swerdlow A, Ashworth A, Orr N, Jones M, Figueroa J, Lissowska J, Brinton L, Goldberg MS, Labreche F, Dumont M, Winqvist R, Pylkas K, Jukkola-Vuorinen A, Grip M, Brauch H, Hamann U, Bruning T, Radice P, Peterlongo P, Manoukian S, Bonanni B, Devilee P, Tollenaar RA, Seynaeve C, van Asperen CJ, Jakubowska A, Lubinski J, Jaworska K, Durda K, Mannermaa A, Kataja V, Kosma VM, Hartikainen JM, Bogdanova NV, Antonenkova NN, Dork T, Kristensen VN, Anton-Culver H, Slager S, Toland AE, Edge S, Fostira F, Kang D, Yoo KY, Noh DY, Matsuo K, Ito H, Iwata H, Sueta A, Wu AH, Tseng CC, Van Den Berg D, Stram DO, Shu XO, Lu W, Gao YT, Cai H, Teo SH, Yip CH, Phuah SY, Cornes BK, Hartman M, Miao H, Lim WY, Sng JH, Muir K, Lophatananon A, Stewart-Brown S, Siriwanarangsan P, Shen CY, Hsiung CN, Wu PE, Ding SL, Sangrajrang S, Gaborieau V, Brennan P, Mckay J, Blot WJ, Signorello LB, Cai Q, Zheng W, Deming-Halverson S, Shrubsole M, Long J, Simard J, Garcia-Closas M, Pharoah PD, Chenevix-Trench G, Dunning AM, Benitez J, Easton DF. 2013. Large-scale genotyping identifies 41 new loci associated with breast cancer risk. Nat Genet 45:353-361.
Nagy A, Wilhelm M, Sukosd F, Ljungberg B, Kovacs G. 2002. Somatic mitochondrial DNA mutations in human chromo- phobe renal cell carcinomas. Genes Chromosomes Cancer 35: 256-260.
Oosting J, Lips EH, van Eijk R, Eilers PH, Szuhai K, Wijmenga C, Morreau H, van Wezel T. 2007. High-resolution copy num- ber analysis of paraffin-embedded archival tissue using SNP BeadArrays. Genome Res 17:368-376.
Pansuriya TC, van Eijk R, d’Adamo P, van Ruler MA, Kuijjer ML, Oosting J, Cleton-Jansen AM, van Oosterwijk JG, Verbeke SL, Meijer D, van Wezel T., Nord KH, Sangiorgi L, Toker B, Liegl-Atzwanger B, San-Julian M, Sciot R, Limaye N, Kindblom LG, Daugaard S, Godfraind C, Boon LM, Vikkula M, Kurek KC, Szuhai K, French PJ, Bovee JV. 2011. Somatic mosaic IDH1 and IDH2 mutations are associated with enchon- droma and spindle cell hemangioma in Ollier disease and Maf- fucci syndrome. Nat Genet 43:1256-1261.
Pavlides S, Whitaker-Menezes D, Castello-Cros R, Flomenberg N, Witkiewicz AK, Frank PG, Casimiro MC, Wang C, Fortina P, Addya S, Pestell RG, Martinez-Outschoorn UE, Sotgia F, Lisanti MP. 2009. The reverse Warburg effect: Aerobic glycoly- sis in cancer associated fibroblasts and the tumor stroma. Cell Cycle 8:3984-4001.
Ribeiro FR, Meireles AM, Rocha AS, Teixeira MR. 2008. Conven- tional and molecular cytogenetics of human non-medullary thy- roid carcinoma: Characterization of eight cell line models and review of the literature on clinical samples. BMC Cancer 8:371. Rouzbahman M, Serra S, Chetty R. 2006. Rectal adenocarcinoma with oncocytic features: Possible relationship with preoperative chemoradiotherapy. J Clin Pathol 59:1039-1043.
Safavi S, Forestier E, Golovleva I, Barbany G, Nord KH, Moorman AV, Harrison CJ, Johansson B, Paulsson K. 2013. Loss of chromo- somes is the primary event in near-haploid and low-hypodiploid acute lymphoblastic leukemia. Leukemia 27:248-250.
Savagner F, Chevrollier A, Loiseau D, Morgan C, Reynier P, Clark O, Stepien G, Malthiery Y. 2001a. Mitochondrial activity in XTC.UCI cells derived from thyroid oncocytoma. Thyroid 11:327-333.
Savagner F, Franc B, Guyetant S, Rodien P, Reynier P, Malthiery Y. 2001b. Defective mitochondrial ATP synthesis in oxyphilic thyroid tumors. J Clin Endocrinol Metab 86:4920-4925.
Speicher MR, Schoell B, Du Manoir S, Schrock E, Ried T, Cremer T, Storkel S, Kovacs A, Kovacs G. 1994. Specific loss of chromosomes 1, 2, 6, 10, 13, 17, and 21 in chromophobe renal cell carcinomas revealed by comparative genomic hybridization. Am J Pathol 145:356-364.
Sukov WR, Ketterling RP, Wei S, Monaghan K, Blunden P, Mazzara P, Raghavan R, Oliviera AM, Wiktor AE, Keeney GL, Van Dyke DL. 2010. Nearly identical near-haploid karyotype in a peritoneal mesothelioma and a retroperitoneal malignant peripheral nerve sheath tumor. Cancer Genet Cytogenet 202: 123-128.
van Eijk R, Licht J, Schrumpf M, Talebian YM, Ruano D, Forte GI, Nederlof PM, Veselic M, Rabe KF, Annema JT, Smit V, Morreau H, van Wezel T. 2011. Rapid KRAS, EGFR, BRAF and PIK3CA mutation analysis of fine needle aspirates from non-small-cell lung cancer using allele-specific qPCR. PLOS ONE 6:e17791.
Wada N, Duh QY, Miura D, Brunaud L, Wong MG, Clark OH. 2002. Chromosomal aberrations by comparative genomic hybrid- ization in Hürthle cell thyroid carcinomas are associated with tumor recurrence. J Clin Endocrinol Metab 87:4595-4601.
Warburg O, Wind F, Negelein E. 1927. The metabolism of tumors in the body. J Gen Physiol 8:519-530.
Wullschleger S, Loewith R, Hall MN. 2006. TOR signaling in growth and metabolism. Cell 124:471-484.
Zielke A, Tezelman S, Jossart GH, Wong M, Siperstein AE, Duh QY, Clark OH. 1998. Establishment of a highly differentiated thyroid cancer cell line of Hürthle cell origin. Thyroid 8:475-483. Zimmermann FA, Mayr JA, Neureiter D, Feichtinger R, Alinger B, Jones ND, Eder W, Sperl W, Kofler B. 2009. Lack of com- plex I is associated with oncocytic thyroid tumours. Br J Cancer 100:1434-1437.
Zimmermann FA, Mayr JA, Feichtinger R, Neureiter D, Lechner R, Koegler C, Ratschek M, Rusmir H, Sargsyan K, Sperl W, Kofler B. 2011. Respiratory chain complex I is a mitochondrial tumor suppressor of oncocytic tumors. Front Biosci 3:315-325.