CDKN2A (P14arf, P16ink4a), and MTAP Genes in Head and Neck Squamous Cell Carcinoma

Home , CDKN2A, P14arf

ORIGINAL ARTICLE Fine-Mapping Loss of Gene Architecture at the CDKN2B (p15INK4b), CDKN2A (p14ARF, p16INK4a), and MTAP Genes in Head and Neck Squamous Cell Carcinoma

Maria J. Worsham, PhD; Kang Mei Chen, MD; Nivedita Tiwari, MS; Gerard Pals, PhD; Jan P. Schouten, PhD; Seema Sethi, MD; Michael S. Benninger, MD

Objective: To identify the extent and the smallest region Results: Cell line UMSCC-11B retained all 9p loci of loss for CDKN2BINK4b, CDKN2AARF,INK4a, and MTAP. tested in the region. Cell lines UMSCC-17A/B indi- Homozygous deletions of human chromosome 9p21 occur cated homozygous deletion of CDKN2AARF, INK4a start- frequently in malignant cell lines and are common in squa- ing at p16INK4 exon 1␣ to include exons 2 and 3. mous cell carcinoma of the head and neck (HNSCC). This Homozygous loss was indicated for CDKN2BINK4b and complex region encodes the tumor suppressor genes CDKN2AARF,INK4a in UMSCC-11A, and UMSCC-81A. Cell cyclin-dependent kinase 2B (CDKN2B)(p15INK4b) and line UMSCC-81B indicated retention of all 9p loci ex- CDKN2A (p14ARF,p16INK4a) and the housekeeping gene me- cept for exon 1␣ (p16INK4a). Selective loss of the 3Ј end thylthioadenosine phosphorylase (MTAP). of MTAP was observed in UMSCC-11A. Genomic alterations by fine-mapping MLPA were validated at the DNA Design: A targeted probe panel designed to finely map level for CDKN2A. the region of 9p21 loss comprised 3 probes for CDKN2BINK4b, 7 for CDKN2AARF, INK4a, and 3 for MTAP and Conclusions: We identified exon 1␣ (p16INK4a) as the small- was interrogated using the multiplex ligation- est region of loss in the CDKN2AARF, INK4a gene. The fre- dependent probe amplification assay (MLPA). The MLPA quency and precise loss of CDKN2BINK4b, CDKN2AARF, INK4a, genomic copy number alterations for CDKN2A were vali- and MTAP in the prognosis of 9p21-deleted HNSCC may dated using real-time polymerase chain reaction. provide impetus for use of these targets as therapeutic bio- markers in head and neck cancer. Subjects: Six HNSCC primary (A) and recurrent or metastatic (B) cell lines were examined: UMSCC-11A/11B, UMSCC-17A/17B, and UMSCC-81A/81B. Arch Otolaryngol Head Neck Surg. 2006;132:409-415

HE MOST COMMON CHRO- tory function is achieved through the p16 mosome rearrangements protein product, which functions up- associated with squamous stream of Rb, and the p14 protein, which cell carcinomas of the head blocks MDM2 inhibition of p53 activity, and neck (HNSCCs) are preventing p53 degradation, and thus per- unbalanced translocations leading to chro- mitting p53-induced apoptosis or growth T 1 8 mosomal deletions. Chromosome 9p21 arrest. The CDKN2D/p14 gene has a has been reported to be a critical region unique first exon 1␤ that, under the con- of loss in various cancers. Genetic alter- trol of its own promoter, splices onto exon ations at the 9p21 locus have been linked 2ofCDKN2A/p16 in an alternative read- to malignant progression in HNSCC.2,3 The ing frame, allowing production of the 2 cyclin-dependent kinase 2A (CDKN2A) totally unrelated proteins, p14 and p16, Author Affiliations: and CDKN2B genes map to 9p21 and are respectively (Figure 1).8 Department of in tandem, spanning a region of approxi- Whereas CDKN2A/p16 mutations se- Otolaryngology–Head and Neck mately 80 kilobases (kb), with CDKN2B lectively inactivate the Rb pathway, dele- Surgery and Research Division, located 25 kb centromeric to CDKN2A4,5 tion of the CDKN2A locus impairs the Rb Henry Ford Hospital, Detroit, (Figure 1). The CDKN2A locus con- and p53 pathways.6-8 Deletion of the Mich (Drs Worsham, Chen, trols the Rb pathway (which regulates G1/ CDKN2A locus also frequently affects the Sethi, and Benninger and Ms Tiwari); and Department of S-phase transition) and the p53 pathway CDKN2B locus, which encodes p15, an im- Clinical Genetics, VU Medical (which induces growth arrest or apopto- portant mediator of the antiproliferative 9 Center (Dr Pals), and sis in response to either DNA damage or effect of transforming growth factor ␤. In- MRC-Holland (Dr Schouten), inappropriate mitogenic stimuli by gen- activation of the CDKN2B (p15), CDKN2A Amsterdam, the Netherlands. erating 2 gene products).6,7 This regula- (p14), and CDKN2A (p16) genes is a fre-

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 132, APR 2006 WWW.ARCHOTO.COM 409

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 fected by deletion, offering additional therapeutic tar- A gets of consideration in HNSCC. Centromere Chromosome 9p21 Telomere

CDKN2B CDKN2A MTAP METHODS

HNSCC CELL LINES AND DNA EXTRACTION 121β 1α 2 3 1 2 3 4 5 6 7 8 9 10 11 132 1 Tumor sample acquisition, tissue culture, karyotype analysis, and genomewide mapping for altered individual gene loci meth- B ods have been detailed elsewhere.19-21 Cell lines UMSCC-11A UMSCC-11A and UMSCC-11B were derived from tumor tissue obtained from UMSCC-11B the primary tumor site (larynx) before and after chemotherapy, respectively. Cell lines UMSCC-17A (supraglottis) and UMSCC-17A /B UMSCC-17B (neck soft tissue) were derived from tumor tis- UMSCC-81A sue obtained simultaneously from primary (A) and metastatic UMSCC-81B (B) sites. Cell lines UMSCC-81A (larynx) and UMSCC-81B (ton- sillar pillar) were derived from tumor tissue obtained from pri- Figure 1. Fine mapping of loss of gene architecture at the cyclin-dependent mary (A) and second primary (B) sites. kinase 2A (CDKN2A)(p14 ARF, p16 INK4) and methylthioadenosine DNA from the 6 cell lines was extracted using the QIAamp phosphorylase (MTAP ) loci in squamous cell carcinomas of the head and Kit (Qiagen Inc, Chatsworth, Calif) at passages 83 and 90 for neck. A, Genomic organization of the CDKN2B, CDKN2A, and MTAP genes at UMSCC-11A and UMSCC-11B, respectively; 138 and 184 for 9p21 illustrating 3 multiplex ligation-dependent probe amplification assay UMSCC-17A and UMSCC-17B, respectively; and 24 and 129 probes for CDKN2B, 7 for CDKN2A, and 3 for MTAP. B, Loss of gene segments is illustrated by dashed lines and retention by solid lines. The for UMSCC-81A and UMSCC-81B, respectively. UMSCC-11B cell line retained all 9p21 loci tested. Note the homozygous loss of CDKN2B, CDKN2A, and the 3Ј end of MTAP in UMSCC-11A. The FINE-MAPPING LOSS AT THE 9p21 LOCUS UMSCC-81A cell line had homozygous loss of CDKN2B and CDKN2A but retained MTAP. The UMSCC-81B cell line indicated loss in only exon 1␣ of p16 INK4a. The multiplex ligation-dependent probe amplification assay (MLPA)21,22 provides relative quantification of loss and gain of gene loci using the candidate gene approach. Our group21 quent event in human oral SCCs.10 The main modes of recently showed that for HNSCC cell lines UMSCC-11A/B, p16INK4a inactivation in HNSCC are known to include UMSCC-17A/B, and UMSCC-81A/B, loss and gain of genetic homozygous deletions (40%-60%), mutations (15%- loci using a genomewide probe panel concurred with tumor 20%), and gene hypermethylation events (5%).10,11 karyotypes. The 112–gene probe panel with single probes for the CDKN2A and CDKN2B genes indicated loss Somatic alterations in the CDKN2A gene occur in of CDKN2A in all except UMSCC-11B, with concomitant many cancer types, and germline mutation carriers are loss of CDKN2B in UMSCC-11A, UMSCC-17B, and predisposed to a high risk of pancreatic and breast UMSCC-81A. cancers.12 To examine the extent of deletion at the CDKN2B/ Although inactivation of p16 has been reported as CDKN2A/MTAP locus, a targeted MLPA probe panel was the most common genetic alteration in HNSCC, mak- designed to dissect out the region of 9p21 loss. This panel of ing it an ideal target for gene replacement,13 the extent 38 probes comprises 17 control non–chromosome 9p probes to which p14 is independently altered, or the fre- and 21 chromosome 9p21 probes (Table). The latter start quency with which p14 and p16 are affected, has not 3349 kb from the centromeric region and extend up to been examined thoroughly. Because these proteins are 600 kb to the 9p telomere to include 3 probes for CDKN2BINK4b, 7 for CDKN2AARF, INK4a, and 3 for MTAP (Table completely unrelated, bearing no amino acid homol- and Figure 1A). ogy to one another, each of these targets has the potential to independently serve as therapeutic interventions in HNSCC. INTERPRETATION The human methylthioadenosine phosphorylase Normal tissue from each cancer patient serves as an internal (MTAP) gene at 9p21, approximately 100-kb telomeric reference when available. For cell lines, when normal DNA to CKDN2A, is an essential enzyme for the salvage of ad- samples are not available, control (normal) male and female enine and methionine (Figure 1). Cells that lack this en- DNA samples are run with each probe set. Quantification— zyme become sensitive to purine synthesis inhibitors or loss or gain of gene loci—is determined through a process of methionine starvation and can be therapeutically ex- normalization. This process addresses variations in the sur- ploited for selective therapy.14-16 Deletion of MTAP was face area of a peak (intensity) encountered due to fluctuations the most frequent alteration in genomewide profiling stud- in the assay run, such as amount of DNA, ploidy variations, ies of oral squamous cell carcinoma.17 Because deletion and polymerase chain reaction (PCR) conditions. Briefly, the INK4a ARF peak area for each probe is expressed as a percentage of the total of CDKN2A (p16) and CDKN2A (p14) causes dys- 21 regulation of the 2 pathways, Rb and p53, important in surface area of all peaks of a sample in an assay run (Figure 2). Relative copy number for each probe is obtained as a ratio of most cancers, loss of MTAP activity is thought to be in- 18 the normalized value for each locus (peak) of the sample to cidental and not of pathogenic consequence. that of the control. A difference is significant only if the ratio The goal of this study was to fine map precise archi- is less than 0.7 (loss) or higher than 1.3 (gain). A relative tectural loss of CDKN2A (p14, p16), CDKN2B (p15), and copy number of 2 is considered normal, 1 or 0 is considered MTAP to identify the extent to which this locus is af- loss, and 3 or more is considered gain. Loss of a gene copy is

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 132, APR 2006 WWW.ARCHOTO.COM 410

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 Table. 9p21 Fine-Mapping Multiplex Ligation-Dependent Probe Amplification Gene Probe Panel CDKN2B : Probes 1, 2, 3, Orange A CDKN2A : p14 Exon: Probes 4, 5, Blue B CDKN2A : p16 Exon: Probes 8, 9, 10, Green C Length CNKN2A : Between p14 and p16: Probes 6, 7, Purple D of PCR Gene Probe Location* MTAP : Probes 11, 12, 13, Red E 130 Control probe 0797-L0463, 5q31 A 139 Control probe 1110-L0172, 1p22 100 150 200 250 300 350 400 450 500 550 600 650 700 750 148 CDKN2A probe 1523-L0957, 9p21 Between p14 and 450C D B p16-probe 7 360C 157 CDKN2A probe 1524-L0962, 9p21 Start p14 exon A 270C A C 1␤-probe 4 C 180C EE 166 MLLT3 probe 1286-L0847, 9p21 C E D B A 90C 175 Control probe 1176-L0736, 7p22 C 184 MLLT3 probe 1287-L0846, 9p21 193 Control probe 0976-L0563, 11p13 B 202 CDKN2B probe 1525-L0960, 9p21 Start p15 exon 0 150 200 250 300 350 400 450 500 550 600 650 700 750 1-probe 1 450C 211 CDKN2B probe 0607-L0591, 9p21 End p15 exon 360C 1-probe 2 270C 220 Control probe 1246-L0794, 14q24 180C 229 Control probe 1304-L0854, 5q35 90C 238 CDKN2A probe 1289-L0834, 9p21 p16 Exon 1␣-probe C 8 C 247 Control probe 1164-L0720, 11q13 0 150 200 250 300 350 400 450 500 550 600 650 700 750 256 CDKN2A probe 1290-L0166, 9p21 p16 Exon 2-probe 9 450C 265 Control probe 0803-L0635, 13q11 360C 274 CDKN2A probe 1291-L0835, 9p21 p16 Exon 3-probe 270C 10 180C 283 Control probe 1326-L0873, 17p13 292 MTAP probe 1292-L0839, 9p21 MTAP Exon 7-probe 90C 11 C 301 Control probe 1041-L0614, 8q24 310 MTAP probe 1293-L0838, 9p21 MTAP Exon 6-probe Figure 2. Fine-mapping multiplex ligation-dependent probe amplification 12 assay illustrated for cell lines UMSCC-81A/B. A, Note 38 individual peaks 319 FLJ00026 probe 1130-L0688, 9p243 ranging from 130 to 472 base pairs in the male control sample (Table). The 3 328 MTAP probe 1294-L0837, 9p21 Start of MTAP-probe cyclin-dependent kinase 2B (CDKN2B) probes are depicted in orange (probes 1, 13 2, and 3 in Figure 1); the 2 CDKN2A p14 probes are in blue (probes 337 Control probe 1310-L0848, 5q35 4 and 5 in Figure 1); the 2 CDKN2A probes between p14 and p16 are in purple 346 Control probe 1335-L0879, 7q11 (probes 6 and 7 in Figure 1); the 3 CDKN2A p16 probes are in green (probes 8, 355 CDKN2A probe 1528-0956, 9p21 Between p14 and 9, and 10 in Figure 1); and the 3 methylthioadenosine phosphorylase (MTAP ) p16-probe 6 probes are in red (probes 11, 12, and 13 in Figure 1). B, Note the complete loss of the CDKN2B/A genes (a total of 10 probes) in UMSCC-81A compared with 364 KIAA1354 probe 1296-L0841, 9p21 the normal male genomic DNA. Note the retention of all 3 MTAP probes. 373 Control probe 1235-L0773, 22q11 C, The UMSCC-81B cell line indicated retention of all 9p21 loci except for exon 382 IFNW1 probe 1297-L0844, 9p22 1␣ (p16INK4a), making the latter the smallest region of overlap for loss in the 400 IFNB1 probe 1298-L0842, 9p21 CDKN2AARF, INK4a region. Note the retention of all 3 MTAP probes. Asterisk 409 Control probe 0963-L0550, 2p14 indicates the minimal region of loss (CDKN2A exon 1␣). 418 ELAVL2 probe 1299-L0843, 9p21 427 CDKN2A probe 1530-L0955, 9p21 p14 Exon 1 ␤-probe 5 436 TEK probe 1300-L0845, 9p21 Cell line DNA, starting with approximately 20 to 50 ng, is in- 445 Control probe 1093-L0661, 22q13 terrogated for the 38 genes in 1 reaction tube. Briefly, DNA is di- 454 CDKN2B probe 1531-L0954, 9p21 End p15 exon luted with water to a total volume of 5 µL and then is denatured 2-probe 3 by heating for 5 minutes at 98°C in a thermocycler. Binary MLPA 463 Control probe 0979-L0568, 10p14 probes are added and allowed to hybridize to their targets dur- 472 Control probe 1243-L0786, 8p23 ing 16 hours of incubation at 60°C. Dilution buffer and a spe- cial Ligase-65 enzyme (MRC-Holland) are added to the vial. Abbreviations: CDKN2A/B, cyclin-dependent kinase 2A/B; ELAVL2, During 15 minutes of incubation at 60°C, the 2 parts of a probe embryonic lethal, abnormal vision, Drosophila, homolog-like 2; FLJ00026, alias, can be ligated to each other and become an amplifiable mol- DOCK8, dedicator of cytokinesis 8; IFNB1, interferon ␤1, fibroblast; IFNW1, ecule provided that the complementary sequence is present in ␻ ␣ interferon 1; 00000000KIAA1354, IFNA5 gene for interferon 5; MLLT3, the sample. This is followed by the addition of PCR primers, myeloid lymphoid or mixed-lineage leukemia (trithorax [Drosophila] homolog), translocated to 3; MTAP, methylthioadenosine phosphorylase; PCR, polymerase deoxynucleotide triphosphates, and Taq polymerase, fol- chain reaction; TEK, endothelium-specific receptor tyrosine kinase. lowed by the following cycles: 1 cycle for 1 minute at 95°C; 10 *See Figure 1. cycles for 30 seconds at 95°C, 30 seconds at 70°C, and 1 minute at 72°C; and 30 cycles for 30 seconds at 95°C, 30 seconds at 60°C, and 1 minute at 72°C. All (ligated) probes are amplified indicated by a reduction in peak height for that particular by the same primer pair, one of which is tagged with a fluo- gene probe, homozygous loss is indicated by the absence of a rescent dye. Amplification products (38 individual probe peaks peak (illustrated for CDKN2A and CDKN2B probes in in the control [Figure 2A]) are analyzed using an ABI 310/ABI UMSCC-81A) (Figure 2B), and gain of a copy number is cor- 3100 DNA sequencer, and the products are quantified, nor- respondingly denoted by an increase in peak height. malized, and interpreted.

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 132, APR 2006 WWW.ARCHOTO.COM 411

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 ing a linear temperature transition from 65°C to 97°C at 0.1°C/s. 24 Fluorescence data were converted into melting peaks by using 22 the LightCycler software (version 5.32) to plot the negative derivative of fluorescence over temperature vs temperature. For 20 UMSCC-11A β-Actin PCR products generated on the LightCycler, melting curve analy- 18 UMSCC-11B β-Actin sis was performed immediately after amplification. 16 UMSCC-11B Primers were designed in intron 1␣ of CDKN2A between CDKN2A nucleotides 187602 and 187697 using Oligo 6 (Molecular Bi- 14 UMSCC-11A CDKN2A ology Insights Inc, Cascade, Colo). The primer pairs were as 12 follows: 5Ј nucleotide 187602: 5Ј GGC AAG GAG GAC CAT 10 AAT TC 3Ј;3Ј nucleotide 187697: 5Ј GGA CCA AGA CTT CGC Ј 8 TGAC3 . Detection and quantification of target DNA was per- Fluorsecence (F1) formed using the double-stranded DNA specific dye SYBR green 6 I, which is similar to ethidium bromide in that it binds prefer- 4 entially to double-stranded DNA, emitting a fluorescent sig- 2 nal that is proportional to the amplified target concentration.

0 The signal is measured in channel 1 (at 530 nm) at the end of the elongation phase once per cycle and is monitored in real – 2 ␤ 0 5 10 15 20 25 30 35 40 45 50 55 time. The ubiquitous -globin gene was used as an internal con- Cycle Number trol (housekeeping gene).

Figure 3. DNA amplification by real-time polymerase chain reaction for RESULTS UMSCC-11A and UMSCC-11B with ␤-globin (control) and cyclin-dependent kinase 2A (CDKN2A) primers run in separate reactions but in the same run. Note that the ␤-globin control gene is equally amplified in UMSCC-11A and FINE MAPPING UMSCC-11B (short arrows). Note that the CKDN2A gene is also amplified to the same extent in UMSCC-11B (long arrows). In UMSCC-11A, homozygous loss of CDKN2A was observed, as indicated by a delayed threshold cycle. The UMSCC-11B cell line retained all 9p loci tested in the region. Homozygous loss was indicated for CDKN2BINK4b ARF, INK4a REAL-TIME PCR and CDKN2A in UMSCC-81A and UMSCC-11A (Figure 1B and Figure 2, UMSCC-81A). Cell lines Real-time PCR was used to confirm MLPA DNA copy number UMSCC-17A/B indicated homozygous deletion of loss and retention of CDKN2A in cell lines UMSCC-11A/B, CDKN2AARF, INK4a starting at p16 exon 1␣ to include exons UMSCC-17A/B, and UMSCC-81A/B. In real-time PCR, the 2 and 3. Cell line UMSCC-81B indicated retention of all amount of product formed (detected by binding of the fluo- 9p loci except for exon 1␣ (p16INK4a, probe 8) (Figure 1A), rescent dye SYBR green I) is plotted as a function of the num- making the latter the smallest region of overlap for loss in ber of cycles. Each cycle represents a doubling of DNA (ie, as the CDKN2AARF, INK4a region (Figure 1B and Figure 2). measured by fluorescence of SYBR green I bound to double- In UMSCC-11A, of the 3 MTAP MLPA probes—1 at stranded DNA), so the accumulation of DNA product is expo- the 3Ј end (probe 13, exon 7) (Figure 1A), 1 in exon 6 nential across consecutive cycles (ie, 2n). It is the ability to iden- (probe 12) (Figure 1A), and 1 at the 5Ј end (probe 11)— tify the exponential phase of product formation “in real time” Ј that makes the method quantitative because the amount of prod- homozygous loss was indicated only for the 3 end of the uct is exponentially related to the amount of template (the un- MTAP gene. known quantity) during this phase. Therefore, during the exponential phase of PCR product formation, the amount of VALIDATION OF GENOMIC LOSS product doubles during each cycle. The beginning of the ex- OF THE CDKN2A LOCUS ponential phase of amplification (so-called crossing points or threshold cycle [Ct]) is considered the most reliable point of the PCR related to sample concentration. Two samples that reach Genomic loss and retention of CDKN2A copy number with the threshold within 1 cycle of each other (eg, Ct1=10, Ct2=11) real-time PCR for UMSCC-11A/B, UMSCC-17A/B, and differ by 2-fold in the amount of amplified product (target gene). UMSCC-81A/B was concordant with confirming reten- Melting curve analysis is an exact and fast method for check- tion of the CDKN2A intron 1␣ sequences in UMSCC-11B ing PCR specificity. Every DNA fragment melts at a character- and UMSCC-81B and loss in UMSCC-11A, istic temperature, called the melting temperature, defined as UMSCC-17A/B, and UMSCC-81A (illustrated for the temperature at which 50% of the DNA is single stranded. UMSCC-11A/B in Figure 3). Here, normal copy num- The most important criteria that determine the melting tem- ber was indicated for ␤-globin in UMSCC-11A/B and for ϩ perature are the G C content and the length of the fragment. CDKN2A in UMSCC-11B. In UMSCC-11A, however, loss The LightCycler Instrument (Roche Diagnostic Corp, India- of CDKN2A is indicated by a delayed Ct. Melting curve napolis, Ind) monitors the fluorescence continuously while rais- ing the temperature gradually. When the temperature in the analysis confirmed lack of a specific melting peak for capillary reaches the melting temperature of the fragment un- CDKN2A. The MLPA fine mapping confirmed homozy- der study, there is a sharp decrease in fluorescence because SYBR gous loss of this region (Figure 1B). green I dye is released from the amplicon. Plotting fluorescence vs temperature generates melting curves. When melting COMMENT curves are displayed as the first negative derivative of fluorescence vs temperature, a peak is generated at the melting temperature. DNA melting curves were acquired using the Genetic alterations provide a means of identifying tu- LightCycler by measuring the fluorescence of SYBR green I dur- mor cells and defining changes that presumably deter-

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 132, APR 2006 WWW.ARCHOTO.COM 412

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 mine biological differences from their normal counter- The latter would explain the observation of an intact 9p21 parts. The underlying hypothesis is that behavior of tumor region in UMSCC-11B and offers a rationale, presumably, cells is determined by genetic changes that alter cell for selection of a tumor subpopulation that escapes che- growth, cell differentiation, programmed cell death, and motherapy likely aided by an intact 9p21 DNA repair and cell migration. Knowledge of the genetic mechanisms that survival repertoire. drive cancer growth and development can provide bet- In UMSCC-11A, of the 3 MTAP MLPA probes (1 at ter diagnostic and prognostic information and more ap- the 3Ј end, 1 in intron 6, and 1 at the 5Ј end), homozy- propriate selection of therapy. gous loss was indicated for only the 3Ј end of the MTAP The precise role of the clinical application of mo- gene, pointing to selective loss of the carboxy terminal lecular prognostic markers in HNSCC remains elusive. region of the gene (Figure 1B). Studies of consistent genetic changes in HNSCC have Toohey,28 a quarter of a century ago, first recognized been instrumental in the initiation of gene therapy trials that certain murine malignant hematopoietic cell lines in vitro and in experimental animals. The transfer of lacked MTAP29 activity. The function of MTAP is to wild-type p53 into HNSCC tumor cells with mutant cleave methylthioadenosine, a by-product of polyamine p53 was shown to induce growth arrest and tumor re- metabolism, to adenine and 5Ј-methylthioribose1- gression.23 Other genes being targeted for gene therapy phosphate, which are recycled to adenine nucleotides include the B-cell leukemia/lymphoma 2 (BCL2) gene and methionine, respectively. The MTAP, the first en- homologue BCL2XS, which inhibits BCL2 function and zyme in the pathway, seems to be expressed in all nor- has been shown to have altered the sensitivity of tumor mal human tissues,30,31 which suggests that the entire cells to chemotherapeutic agents by restoring sensitiv- salvage pathway is present in all of the cells in the hu- ity to apoptosis.24 Several strategies directed against epi- man body, designating MTAP an important housekeep- dermal growth factor receptor overexpression have in- ing gene in malignancy. cluded the use of monoclonal antibodies targeted Genetic studies show high rates of MTAP loss in against epidermal growth factor receptor.25 A chimeric non–small cell lung cancer, melanoma, bladder cancer, anti–epidermal growth factor receptor monoclonal an- pancreatic cancer, osteosarcoma, and endometrial can- tibody (C225) has been tested in combination with cer.32-35 Tumors that lack MTAP are expected to be sen- either cisplatin26 or radiation,27 and the results are sitive to inhibitors of purine synthesis or methionine promising. starvation. Genomewide profiling studies of oral SCC17 Genetic alterations at the 9p21 locus have been linked found frequent deletion of MTAP; however, mapping of to malignant progression in HNSCC.2,3 Currently, re- precise loss of MTAP and its association with concomi- combinant adenovirus capable of directing a high level tant or independent loss with CDKN2A/B genes was not of p16 protein expression (Ad5-p16) demonstrated a sig- addressed. nificant antitumor effect of Ad5-p16 against human Because deletion of the CDKN2AINK4a (p16) and HNSCC in vivo.13 Other proteins produced at this 9p21 CDKN2AARF (p14) genes causes dysregulation of the 2 locus—p15 (CDKN2B), p14 (CDKN2A), and MTAP— pathways important in most cancers (Rb and p53), loss each have the potential to independently serve as thera- of MTAP activity is thought to be incidental and not of peutic targets in HNSCC. pathogenic consequence.18 However, there are several rea- In this study, 7 HNSCC cell lines, including 6 that in- sons proposed to suggest that this may not be the case.14 dicated loss and retention of the CDKN2A/B locus by an First, homozygous deletion is an unusual mechanism for MLPA genomewide 112-probe panel,21 were examined inactivation of a tumor suppressor gene. Most tumor sup- using a fine-mapping 9p21 probe set. We found homo- pressor genes are inactivated by point mutation of one zygous loss at the CKDN2A locus in 5 of 6 HNSCC cell allele followed by loss of the other allele (loss of hetero- lines: UMSCC-11A, UMSCC-17A/B, UMSCC-81A, and zygosity). This is rarely observed for CDKN2A.36 A likely UMSCC-81B. The UMSCC-81B cell line indicated reten- hypothesis for this observation is that homozygous de- tion of all 9p loci except for exon 1␣ (p16INK4a), making letion can remove more than 1 gene from the region, the latter the smallest region of overlap for loss in the whereas point mutation followed by loss of heterozygos- CDKN2AARF, INK4a region (Figure 1B and Figure 2). Stud- ity cannot.14 Second, in certain cancers, loss of MTAP has ies reporting homozygous deletions of CDKN2A in been observed in cells that retain p16. Schmid et al32 found HNSCC may, therefore, underestimate the frequency of in a study of non–small cell lung cancer that homozy- this region depending on the extent to which deletion gous deletion of MTAP occurred in 38% (19 of 50) of the mapping was performed. samples compared with only 18% (9 of 50) for p16. In HomozygouslossforCDKN2BINK4b andCDKN2AARF, INK4a another study37 it was found that 3 of 7 primary astro- wasobservedinUMSCC-81A(Figure1B).ForUMSCC-11A, cytomas were deleted for MTAP, but only 2 of 7 were de- homozygous loss of 9p21 observed by MLPA fine-map- leted for p16. The fact that MTAP is lost independently ping analysis was also supported by real-time quantitative of p16 hints that loss of MTAP may have some func- PCR for the CDKN2A gene. In addition, the inherent sen- tional basis in tumor biology.14 In UMSCC-11A, despite sitivity of real-time PCR revealed the late appearance of a homozygous loss of CDKN2A/B genes, only the 3Ј end product (Figure 3), where this late product seems to rep- of MTAP was lost. The latter may point to the 3Ј end as resent a minor population of cells (many logs fewer than a selective target for gene deletion, suggesting the car- the predominant population) from the original tumor boxy terminal of the gene as a critical region. that may correspond to the population that survived che- Tumors that lack MTAP are expected to be sensitive motherapy to predominate in the UMSCC-11B cell line. to inhibitors of purine synthesis or methionine starva-

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 132, APR 2006 WWW.ARCHOTO.COM 413

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 tion. It has been demonstrated that the antipurine- There is a large gap in our knowledge and understand- related growth-inhibitory action of the antifolate agent ing of the frequency and precise loss of p15, p14, p16, methotrexate was more pronounced in subsets of pan- and MTAP in HNSCC tumorigenesis. Better understand- creatic carcinoma cell lines that were p16−, MTAP− than ing of the contribution of these 4 gene products in the in pancreatic carcinomas or normal keratinocyte epithe- prognosis of 9p21-altered HNSCC can provide impetus lial cells that were p16ϩ, MTAPϩ.38 Furthermore, it was for exploitation of these targets as therapeutic biomar- shown that the co-addition of an inhibitor of MTAP en- kers in head and neck cancer. hanced the potency and efficacy of the antipurine- related growth-inhibitory actions of methotrexate in Submitted for Publication: April 30, 2005; final revi- MTAP− but not MTAPϩ cell lines.38 Thus, MTAPϩ cells sion received September 1, 2005; accepted October 10, are less sensitive to the inhibitory effects that antifolates 2005. such as methotrexate have on purine de novo synthesis. Correspondence: Maria J. Worsham, PhD, Department In contrast, because MTAP-deficient malignant cells can- of Otolaryngology–Head and Neck Surgery, Henry Ford not recycle the purine moiety of methylthioadenosine, Hospital, 1 Ford Pl, 1D, Detroit, MI 48202 (mworsha1 tumor cells are more dependent on purine de novo bio- @hfhs.org). synthesis and are more sensitive than MTAP-containing Author Contributions: Dr Worsham had full access to cells to the antipurine actions of antifolates. all the data in the study and takes responsibility for To examine the effects of MTAP in tumorigenesis, Chris- the integrity of the data and the accuracy of the data topher et al14 reintroduced MTAP into MCF-7 breast ad- analysis. enocarcinoma cells. Although MTAP expression does not Financial Disclosure: None. affect the growth rate of cells in standard tissue culture con- Funding/Support: This study was supported by grant 1 ditions, it severely inhibits their ability to form colonies in R01 DE15990 from the National Institutes of Health, soft agar or collagen. In addition, the study showed that Bethesda, Md (Dr Worsham). MTAP-expressing cells are suppressed for tumor forma- Previous Presentation: This study was presented as a late- tion when implanted into severe combined immunodefi- breaking abstract at the annual meeting of the American cient (SCID) mice. Because MTAP expression causes a sig- Society of Investigative Pathology; April 15, 2003; San nificant decrease in intracellular polyamine levels and alters Diego, Calif. the ratio of putrescine to total polyamines, consistent with this observation, the study also showed that the polyamine biosynthesis inhibitor ␣-difluoromethylornithine in- REFERENCES hibited the ability of MTAP-deficient cells to form colonies in soft agar, whereas addition of the polyamine 1. Van Dyke DL, Worsham MJ, Benninger MS, et al. Recurrent cytogenetic abnor- malities in squamous cell carcinomas of the head and neck. Genes Chromo- putrescine-stimulated colony formation in MTAP- somes Cancer. 1994;9:192-206. expressing cells, suggesting that MTAP-negative tumors may 2. Danahey DG, Tobin EJ, Schuller DE, Bier-Laning CM, Weghorst CM, Lang JC. be particularly sensitive to ␣-difluoromethylornithine. The p16 Mutation frequency and clinical correlation in head and neck cancer. Acta results of this study indicate that MTAP has tumor sup- Otolaryngol. 1999;119:285-288. pressor activity and suggest that its effects may be medi- 3. Lydiatt WM, Davidson BJ, Schantz SP, Caruana SC, Chaganti RS. 9p21 Deletion cor- relates with recurrence in head and neck cancer. Head Neck. 1998;20:113-118. ated by altering intracellular polyamine pools. 4. Sherr CJ. Cancer cell cycles. Science. 1996;274:1672-1677. There are a variety of potential implications of these find- 5. Johnson DG, Walker CL. Cyclins and cell cycle checkpoints. Annu Rev Pharma- ings for clinical cancer chemotherapy, including HNSCC. col Toxicol. 1999;39:295-312. If normal cells, which are uniformly MTAPϩ,39 are intrin- 6. Haber DA. Splicing into senescence: the curious case of p16 and p19ARF. Cell. 1997;91:555-558. sically less sensitive to the antipurine actions of antifo- 7. Xing EP, Nie Y, Song Y, et al. Mechanisms of inactivation of p14ARF, p15INK4b lates such as methotrexate, then this might in part be the and p16INK4a genes in human esophageal squamous cell carcinoma. Clin Can- basis for the relatively selective action that these agents have cer Res. 1999;5:2704-2713. against certain malignancies. If the corollary is correct that 8. Sherr CJ. The Pezcoller Lecture: cancer cell cycles revisited. Cancer Res. 2000; MTAP-deficient malignant cells are especially sensitive to 60:3689-3695. 9. Hannon GJ, Beach D. p15INK4B Is a potential effector of TGF-␤–induced cell cycle certain antifolates that act in whole or in part by inhibit- arrest. Nature. 1994;371:257-261. ing purine de novo synthesis, one would theoretically ob- 10. Shintani S, Nakahara Y, Mihara M, Ueyama Y, Matsumura T. Inactivation of the tain an improved therapeutic index by identifying the sub- p14(ARF), p15(INK4B) and p16(INK4A) genes is a frequent event in human oral set of patients with MTAP-deficient malignancies (including squamous cell carcinomas. Oral Oncol. 2001;37:498-504. ␣ 11. Rocco JW, Sidransky D. p16 (MTS-1/CDKN2/INK4a) in cancer progression. Exp HNSCC) and treating them with such agents. -Difluoro- Cell Res. 2001;264:42-55. methylornithine is currently undergoing clinical trials for 12. Borg A, Sandberg T, Nilsson K, et al. High frequency of multiple melanomas and use as adjunct therapy for a variety of different can- breast and pancreas carcinomas in CDKN2A mutation-positive melanoma families. cers.40,41 Genomewide profiling studies of oral squamous J Natl Cancer Inst. 2000;92:1260-1266. cell carcinoma,17 which found frequent deletions of MTAP 13. Rocco JW, Li D, Liggett WH, et al. p16INK4A adenovirus-mediated gene therapy for human head and neck squamous cell cancer. Clin Cancer Res. 1998;4:1697- and loss of MTAP in 1 of 6 cell lines in this study, provide 1704. a strengthening rationale for a role of MTAP in HNSCC. 14. Christopher SA, Diegelman P, Porter CW, Kruger WD. Methylthioadenosine phos- Stratifying the prognosis of 9p21-deleted HNSCC for MTAP phorylase, a gene frequently codeleted with p16(cdkN2a/ARF), acts as a tumor status may provide a rationale to examine this locus in larger suppressor in a breast cancer cell line. Cancer Res. 2002;62:6639-6644. 15. Harasawa H, Yamada Y, Kudoh M, et al. Chemotherapy targeting methylthioad- studies to assess the feasibility of polyamine biosynthesis enosine phosphorylase (MTAP) deficiency in adult T cell leukemia (ATL). Leukemia. inhibitors as novel treatment interventions in MTAP- 2002;16:1799-1807. deleted HNSCC tumors. 16. Efferth T, Miyachi H, Drexler HG, Gebhart E. Methylthioadenosine phosphory-

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 132, APR 2006 WWW.ARCHOTO.COM 414

©2006 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 lase as target for chemoselective treatment of T-cell acute lymphoblastic leuke- 30. Kamatani N, Nelson-Rees WA, Carson DA. Selective killing of human malignant mic cells. Blood Cells Mol Dis. 2002;28:47-56. cell lines deficient in methylthioadenosine phosphorylase, a purine metabolic 17. Chen YJ, Lin SC, Kao T, et al. Genome-wide profiling of oral squamous cell enzyme. Proc Natl Acad Sci U S A. 1981;78:1219-1223. carcinoma. J Pathol. 2004;204:326-332. 31. Olopade OI, Pomykala HM, Hagos F, et al. Construction of a 2.8-megabase yeast 18. Nobori T, Takabayashi K, Tran P, et al. Genomic cloning of methylthioadenosine artificial chromosome contig and cloning of the human methylthioadenosine phos- phosphorylase: a purine metabolic enzyme deficient in multiple different cancers. phorylase gene from the tumor suppressor region on 9p21. Proc Natl Acad Sci Proc Natl Acad Sci U S A. 1996;93:6203-6208. USA. 1995;92:6489-6493. 19. Carey TE, Van Dyke DL, Worsham MJ, et al. Characterization of human laryn- 32. Schmid M, Malicki D, Nobori T, et al. Homozygous deletions of methylthioad- geal primary and metastatic squamous cell carcinoma cell lines UM-SCC-17A enosine phosphorylase (MTAP) are more frequent than p16INK4A (CDKN2) ho- and UM-SCC-17B. Cancer Res. 1989;49:6098-6107. mozygous deletions in primary non-small cell lung cancers (NSCLC). Oncogene. 20. Bradford CR, Kimmel KA, Van Dyke DL, et al. 11p deletions and breakpoints in 1998;17:2669-2675. squamous cell carcinoma: association with altered reactivity with the UM-E7 33. Della Ragione F, Russo G, Oliva A, et al. 5Ј-Deoxy-5Ј-methylthioadenosine phos- antibody. Genes Chromosomes Cancer. 1991;3:272-282. phorylase and p16INK4 deficiency in multiple tumor cell lines. Oncogene. 1995; 21. Worsham MJ, Pals G, Schouten J, et al. Delineating genetic pathways of disease 10:827-833. progression in head and neck squamous cell carcinoma. Arch Otolaryngol Head 34. Stadler WM, Sherman J, Bohlander SK, et al. Homozygous deletions within chro- Neck Surg. 2003;129:702-708. mosomal bands 9p21–22 in bladder cancer. Cancer Res. 1994;54:2060-2063. 22. Schouten JP, McElgunn CJ, Waaijer R, Zwijnenburg D, Diepvens F, Pals G. 35. Garcia-Castellano JM, Villanueva A, Healey JH, et al. Methylthioadenosine phosphory- Relative quantification of 40 nucleic acid sequences by multiplex ligation- lase gene deletions are common in osteosarcoma. Clin Cancer Res. 2002;8:782- dependent probe amplification. Nucleic Acids Res. 2002;30:e57. 787. 23. Liu TJ, Zhang WW, Taylor DL, et al. Growth suppression of tumor cells of hu- 36. Cairns P, Polascik TJ, Eby Y, et al. Frequency of homozygous deletion at p16/ man head and neck cancer cells by the introduction of wild-type p53 gene via a CDKN2 in primary human tumours. Nat Genet. 1995;11:210-212. recombinant adenovirus. Cancer Res. 1994;54:3662-3667. 37. Brat DJ, James CD, Jedlicka AE, et al. Molecular genetic alterations in radiation- 24. Sumantran VN, Ealovega MW, Nunez G, et al. Overexpression of Bcl-XS sensi- induced astrocytomas. Am J Pathol. 1999;154:1431-1438. tizes MCF-7 cells to chemotherapy-induced apoptosis. Cancer Res. 1995;55: 38. Chen ZH, Olopade OI, Savarese TM. Expression of methylthioadenosine phos- 2507-2510. phorylase cDNA in p16−, MTAP− malignant cells: restoration of methylthio- 25. Hoffmann T, Hafner D, Ballo H, et al. Antitumor activity of anti-epidermal growth adenosine phosphorylase-dependent salvage pathways and alterations of sensi- factor receptor monoclonal antibodies and cisplatin in ten human head and neck tivity to inhibitors of purine de novo synthesis. Mol Pharmacol. 1997;52:903- squamous cell carcinoma lines. Anticancer Res. 1997;17:4419-4425. 911. 26. Shin DM, Donato NJ, Perez-Soler R, et al. Epidermal growth factor receptor– 39. Myers RW, Abeles RH. Conversion of 5-S-methyl-5-thio-D-ribose to methio- targeted therapy with C225 and cisplatin in patients with head and neck cancer. nine in Klebsiella pneumoniae: stable isotope incorporation studies of the ter- Clin Cancer Res. 2001;7:1204-1213. minal enzymatic reactions in the pathway. J Biol Chem. 1990;265:16913- 27. Bonner JA, Raisch KP, Trummell HQ, et al. Enhanced apoptosis with combina- 16921. tion C225/radiation treatment serves as the impetus for clinical investigation in 40. O’Shaughnessy JA, Demers LM, Jones SE, et al. ␣-Difluoromethylornithine as head and neck cancers. J Clin Oncol. 2000;18:47S-53S. treatment for metastatic breast cancer patients. Clin Cancer Res. 1999;5:3438- 28. Toohey JI. Methylthioadenosine nucleoside phosphorylase deficiency in methylthio- 3444. dependent cancer cells. Biochem Biophys Res Commun. 1978;83:27-35. 41. Levin VA, Uhm JH, Jaeckle KA, et al. Phase III randomized study of postradio- 29. Trackman PC, Abeles RH. Methionine synthesis from 5Ј-S-methylthioadeno- therapy chemotherapy with ␣-difluoromethylornithine-procarbazine: N-(2- sine: resolution of enzyme activities and identification of 1-phospho-5-S chloroethyl)-N’-cyclohexyl-N-nitrosurea, vincristine (DFMO-PCV) versus PCV for methylthioribulose. J Biol Chem. 1983;258:6717-6720. glioblastoma multiforme. Clin Cancer Res. 2000;6:3878-3884.

(REPRINTED) ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 132, APR 2006 WWW.ARCHOTO.COM 415