(2004) 23, 2582–2586 & 2004 Nature Publishing Group All rights reserved 0950-9232/04 $25.00 www.nature.com/onc

Overexpression of LRP12, a contained within an 8q22 identified by high-resolution array CGH analysis of oral squamous cell carcinomas

Cathie Garnis*,1,2, Bradley P Coe1,2, Lewei Zhang3, Miriam P Rosin1,2 and Wan L Lam1,2

1British Columbia Research Centre, Vancouver, BC, Canada V5Z-1L3; 2Department of Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada V6T-2B5; 3Department of Oral Biological and Medical Sciences, University of British Columbia, Vancouver, BC, Canada V6T-2B5

Chromosome 8q amplification is a common event observed and deletions, few have been associated with oral in cancer. In this study, we used high-resolution array cancer development. comparative genomic hybridization to resolve two neigh- In a previous study, we demonstrated that multiple boring regions on 8q that are both amplified in oral regions of 8q are altered in head and neck cancer. One region (at 8q24) contains the MYC oncogene, squamous cell carcinomas (SCCs) (Garnis et al., 2004). which is frequently overexpressed in many , while In this report, we describe the use of high-resolution the other region (at 8q22) represents a novel amplicon. array comparative genomic hybridization (CGH) that The alignment of array comparative genomic hybridiza- provides near complete coverage of 8q21–24to profile a tion profiles of 20 microdissected oral squamous cell panel of microdissected oral squamous cell carcinomas carcinomas (OSCCs) revealed a B5 Mbp region of (OSCCs) in order to define one of the recurrent frequent copy number alteration. This region harbors 16 amplifications at 8q22. known genes. Gene expression analysis comparing 15 microdissected OSCC with 16 normal epithelium samples revealed overexpression specific to LRP12 but not the Results and discussion neighboring genes, dihydropyrimidinase and FOG2, sug- gesting that LRP12 may function as an oncogene in oral In all, 22 formalin-fixed, paraffin-embedded OSCCs tumors. were microdissected and analysed using a microarray Oncogene (2004) 23, 2582–2586. doi:10.1038/sj.onc.1207367 consisting of 165 human bacterial artificial chromosome Published online 15 December 2003 (BAC) clones spanning 52 Mbp across 8q21 to 8q24. The identity of these clones is available at Keywords: head and neck; array CGH; 8q; http://www.bccrc.ca/cg/OralCancer.html. The array premalignant construction has been described previously (Garnis et al., 2004). Array CGH analysis detected genetic alteration in 18 cases. No alteration in copy number was observed in four of the cases, while five cases showed an increase in copy number for the entire region (all 165 BACs). The remaining 13 cases showed Introduction two distinct regions of amplification at both 8q22 and 8q24. As stated above, the amplification at 8q24 Oral cancer affects over 500 000 people worldwide each contains the MYC oncogene. The alignment of the year with the major cause being related to tobacco and 8q22 alterations in the 13 cases delineated a minimal alcohol exposure (Krajina et al., 1975; Lowry, 1975). region of alteration (MRA) of approximately 5.3 Mbp Oral cancer is believed to arise through the accumula- (Figure 1a and b) with a centromeric boundary at BAC tion of numerous genetic alterations (Vokes et al., 1993). RP11-346H21 and a telomeric boundary at BAC These alterations may activate or disrupt RP11-680F23. Figure 1c shows an array CGH profile tumor suppressor genes that play a crucial role in the of case 574T. development of the disease. While there have been The 8q22 region is gene rich, with 16 RefSeq genes in numerous genetic and gene expression profiling attempts the 5.3 Mbp MRA (summarized in Table 1). Four to identify these key alterations such as amplifications cancer-related genes reside in this region, although, as indicated in Table 1, none have been previously *Correspondence: C Garnis, British Columbia associated with oral cancer. In this study, we focused Centre, 601 West 10th Avenue, Vancouver, BC, Canada V5Z-1L3; E-mail: [email protected] on one of these genes, LRP12, previously called ST7 but Received 24August 2003; revised 21 October 2003; accepted 18 recently renamed by Battle et al. (2003), which encodes a November 2003 low-density lipoprotein receptor-like protein that is LRP12 overexpression in oral squamous cell carcinoma C Garnis et al 2583

Figure 1 MRA detected at 8q22. (a) Tiling set of BAC clones, selected from the human RPCI-11 BAC library, spanning part of 8q22. Black boxes indicate clones present on the array. (b) Regions of segmental copy increase observed in four samples (566T, 574T, 166T, 573T) aligned with the BAC tiling set. The 5.3 Mbp MRA and the three genes subjected to expression analysis are indicated. All known genes in this region are listed in Table 1. (c) BAC array CGH profile of sample 574T. Data are displayed as a normalized signal ratio between tumor and reference DNA for each BAC clone. Each data point represents the average of three replicate spots on the array and includes the standard deviation. Shading indicates large regions with copy number increase, which contains LRP12 at 8q22 and MYC at 8q24

associated with several cancer-related pathways, and has LRP12) (Figure 2a–d). RNA was extracted from the been localized to BAC clone RP11-437B2 by sequence epithelium of 15 microdissected frozen OSCCs and from alignment. This gene should not be confused with the 16 oral epithelial specimens from individuals without well-studied ST7 gene on 7q31 (Dong and Sidransky, cancer. The normal samples were relatively uniform in 2002). the expression levels of LRP12 with minor fluctuations We assayed for differential expression of LRP12 and observed between samples (Figure 2a and b). In contrast its two neighboring genes, Dihydropyrimidinase to the normal samples, five of the 15 tumor samples (DPYS) (immediately centromeric to LRP12)and showed the overexpression of LRP12 to levels 42 s.d. of Friend of GATA2 (FOG2) (immediately telomeric to the average normal value (Figure 2b).

Oncogene LRP12 overexpression in oral squamous cell carcinoma C Garnis et al 2584 Table 1 RefSeq genes contained within the 5.3 Mbp MRA Gene symbol Accession # Positiona Gene name

DD5 NM_015902 102934923-103093288 Progestin-induced protein ODF1 NM_024410 103232652-103242039 Outer dense fiber of sperm tails 1 TIEG NM_005655 103329813-103336745 Transforming growth factor-b-inducible early growth response OAZIN NM_015878 103508843-103545168 Ornithine decarboxylase antizyme inhibitor ATP6V1C1 NM_001695 103702104-103750586 ARPase, H+ transporting, lysosomal, 42 KDa, V1 subunit C, isoform 1 BAALC NM_024812 103821735-103911325 Brain and acute leukemia gene FZD6 NM_003506 103979911-104013494 Frizzled 6 CTHRC1 NM_138455 104052578-104064008 Collagen triple helix repeat containing 1 MFTC NM_030780 104079660-104096064 Mitochondrial folate transporter RIMS2 NM_014677 104500281-104932880 Protein-regulating synaptic membrane exocytosis 2 DCSTAMP NM_030788 105020847-105037709 Dendritic cell-specific transmembrane protein DPYS NM_001385 105060445-105148070 Dihydropyrimidinase LRP12b NM_013437 105170259-105269961 Suppressor of tumorigenicity 7 FOG2 NM_012082 106471621-106496960 Friend of GATA2 OXR1 NM_018002 107391719-107432882 Oxidation resistance 1 ANGPT1 NM_001146 107930502-108179047 Angiopoietin 1

aBase pair position on corresponds to April 2003 version of the UCSC human draft sequence. bLRP12 is annotated as ST7 in the UCSC Genome Browser.

Our observations that LRP12 is overexpressed in oral and d) were similar in both tumor and normal samples, tumors is inconsistent with the work by Qing et al.inthe suggesting that the overexpression is specific to LRP12 fibrosarcoma cell line model, which report a decrease in and not a consequence of regional upregulation. the expression of LRP12 associated with the transfor- However, our study that focuses on LRP12 does not mation of a fibroblast cell line to a tumorigenic rule out the possibility of the presence of other derivative. LRP12 is not expressed in fibrosarcoma cells differentially expressed genes in the 5.3 Mb region. or is greatly downregulated at the mRNA and protein LRP12 belongs to the LDLR superfamily and may levels when compared with the parental fibroblasts or play a role in signal transduction (Qing et al., 1999). other normal human fibroblasts. This difference could Battle et al. (2003) showed that the cytoplasmic C be associated with the usage of a transformed cell line terminus of LRP12 could interact with several proteins, (MSU-1.1) for determining the normal expression base- including the receptor for activated protein C 1 line of LRP12. Our observations indicate that LRP12 (RACK1), a protein that interacts with a myriad of does not behave as a and is likely cellular proteins, muscle integrin binding protein, which a proto-oncogene in the context of oral cancer. plays a role in myogenesis and SMAD anchor receptor However, the overexpression of LRP12 does not rule activation, which interacts with the TGF-b pathway. out the possibility of other overexpressed genes in this In summary, through the use of high-resolution array amplified region. CGH, we have identified a recurrent 5.3 Mbp amplified Archival material from formalin-fixed paraffin blocks region at 8q22 that is present in oral tumors. The LRP12 was available for a subset of the cases used in the gene found in the 8q22 region showed an increase in expression analysis above. This allowed us to test gene expression in oral tumors, while the adjacent genes whether the overexpression of LRP12 is accompanied FOG2 and DPYS exhibited normal expression levels. by genetic alteration. Although the amount of DNA The frequent amplification of the 8q22 genomic region recovered from the microdissected cancer cells was in combination with the overexpression of LRP12 and insufficient for array CGH analysis, we were able to the association of the LRP12 protein to pathways examine these samples for loss of heterozygosity/allelic implicated in suggests its involvement in imbalance (LOH/AI) at D8S1459, a microsatelite head and neck epithelial carcinogenesis. Alteration at marker mapping near LRP12 on BAC RP11-171O13 D8S1459 in oral dysplastic lesions suggests that the role (Figure 1a). Although microsatellite analysis does not of LRP12 in early oral cancer development should be distinguish between amplification and deletions, the further explored. data are consistent with a correlation between genetic alteration and LRP12 overexpression. Four tumors and three normal samples were polymorphic at D8S1459. Materials and methods Three of the four cases overexpressing LRP12 showed LOH/AI, while three cases with normal LRP12 expres- Tissue samples sion exhibited retention of both alleles. Furthermore, we Archival dysplasia and tumor samples for array CGH and observed genetic alteration in oral premalignant lesions, microsatellite analysis (formalin-fixed paraffin-embedded tis- with seven of 21 microdissected dysplastic lesions sue blocks) were obtained from the British Columbia Oral showing LOH/AI (data not shown). Biopsy Service and the diagnoses were confirmed by an oral In contrast to these results, the expression levels of the pathologist. Dysplastic, tumor and adjacent stromal cells were two genes flanking LRP12, DPYS and FOG2 (Figure 2c microdissected from hematoxylin- and eosin-stained sections

Oncogene LRP12 overexpression in oral squamous cell carcinoma C Garnis et al 2585 BAC array construction Array CGH technology has been applied to several different tumor types in order to identify copy number gains and losses (Martinez-Climent et al., 2003; Paris et al., 2003; Veltman et al., 2003; Weiss et al., 2003). In this study, a minimal overlapping tiling set of 165 RPCI-11 BAC clones representing 8q21–24 (B52 Mbp) was selected from the FPC database and the map location of sequenced clones was verified using the UCSC Browser (Karolchik et al., 2003). This clone list is publicly available at http://www.bccrc.ca/cg/OralCan- cer.html. The identity of each BAC DNA sample was confirmed based on its HindIII fingerprint. Mse1-digested BAC DNA was ligated to linkers (50-AGTGGGATTCCGC ATGCTAGT-30 and 50-TAACTAGCATGC-30) and amplified by PCR as described previously (Garnis et al., 2004). An aliquot of each PCR product was further amplified to produce sufficient DNA for spotting. The amplified DNA, dissolved in a 20% DMSO solution, was denatured by boiling for 10 min and rearrayed for robotic printing in triplicate using a VersArray ChipWriter Pro (BioRad, Mississauga, ON, Cana- da) with Stealth Micro Spotting Pins (Telechem/ArrayIT SMP2.5, Sunnyvale, CA, USA). Linker-mediated PCR- amplified male human genomic DNA samples (Novagen, Madison, WI, USA) were spotted on the array in order to allow normalization of the hybridization signal intensities between dyes. The DNA was then covalently bonded to the slides by baking and UV crosslinking. Slides were then washed to remove unbound DNA.

Array CGH Test and reference DNA of 100 ng were separately labeled using Cyanine 3 and Cyanine 5 dCTPs according to a previously described protocol (Garnis et al., 2004). The DNA probes were denatured and annealed with 100 mg human Cot. 1 DNA 25 ml of DIG Easy hybridization solution (Roche, Laval, Quebec, Canada) containing 50 mg of sheared herring sperm DNA (Sigma-Aldrich, Oakville, ON, Canada) and 250 mg of yeast tRNA (Calbiochem, La Jolla, CA, USA). The BAC array slides were prehybridized at 421C for 1 h with DIG Easy hybridiza- tion buffer (Roche, Laval, Quebec, Canada) containing 1% BSA and 2 mg/ml sheared herring sperm DNA. The probe mixture was applied to the slide surface, coverslipped and incubated at 421C for 36 h. Arrays were washed five times for 5 min in 0.1 Â SSC, 0.1% SDS at room temperature with agitation. Each array was then rinsed five times in a clean slide box containing 0.1 Â SSC with agitation and dried by centrifugation. Hybridized slides were imaged using a CCD-based imaging Figure 2 Expression analysis of LRP12 and its neighboring genes. e (a) LRP12 reverse transcription PCR product amplified from system (ArrayWorX Auto, API, Issaquah, WA, USA) and microdissected tumor and normal epithelium resolved by poly- analysed with SoftWoRx Tracker array analysis software. acrylamide gel electrophoresis and stained with SYBR green. (b) Spot signal data for each channel were normalized by applying Histogram of the expression levels of LRP12 relative to that of a scale factor, which balanced the signal intensities of the GAPDH.(c) Levels of DPYS relative to that of GAPDH.(d) Levels human genomic DNA control spots. The average signal ratios of FOG2 relative to that of GAPDH and standard deviations for each triplicate spot set were calculated and displayed as a plot of the normalized Cyanine5/ Cyanine3 log 2 signal ratio versus relative tiling path position. A log 2 signal ratio of 0 at a spot represents equivalent copy number between the sample and reference DNA. Arrays were and DNA was isolated as described previously (Zhang et al., initially subjected to normal versus normal hybridizations for 1997). Samples for gene expression analysis were collected quality control (Veltman et al., 2003). Spots that exhibited within 15 min of surgery and frozen in liquid nitrogen. RNA signal ratios outside of 3 s.d. from 0 in these experiments were was extracted from microdissected OSCCs and a panel of discarded. This resulted in the establishment of a 70.2 log 2 normal oral epithelium samples from individuals without ratio threshold for defining regions of copy number increase cancer (Chomczynski and Sacchi, 1987). and decrease.

Oncogene LRP12 overexpression in oral squamous cell carcinoma C Garnis et al 2586 Semiquantitative analysis of LRP12, DPYS and FOG2 Microsatellite analysis expression Microsatellite analysis was performed to detect LOH/AI in cDNA was synthesized using the Superscript II RNAse H specific chromosomal regions according to published protocols reverse transcriptase system (Invitrogen, Burlington, ON, (Zhang et al., 1997). For comparison, the underlying stroma Canada). GAPDH levels were used to normalize samples. was used as a source of matched control DNA. Signal ratios Gene-specific primers were used to assay the quantity of for microsatellite analysis were assessed using previously LRP12 (50-ACTGTAGTTCACTGT-30,50-AGTTCAAGATG described criteria (Ng et al., 2000; Rosin et al., 2000). TGCT-30), DPYS (50-CACAGCCTTTCCATC-30,50-ACATT GTCAACCCAT-30), FOG2 (50-CTGATGCAGCTCTGTCT Acknowledgements AA-30,50-TACTTGCAGCATTGAGTTTA-30) and GAPDH We thank Sandra Henderson, Spencer Watson and Chad (50-ACCTACCAAATATGATGACATCA-30,50-CGCTGTT Malloff for technical assistance, Dr Calum MacAulay for GAAGTCAGAGGA-30). PCR cycle conditions were as helpful discussion and Dr Marco Marra for providing BAC follows: one cycle of 951C, 1 min; 30 cycles of (951C, 30 s; clones. This work was supported by scholarships from the 551C, 30 s; 721C, 1 min) and a 10 min extension at 721C. PCR Natural Sciences and Engineering Research Council and products were resolved by polyacrylamide gel electrophoresis, Michael Smith Foundation for health research to CG and imaged by SYBR green staining (Roche, Laval, Quebec, BPC, and grants from the Canadian Institute of Health Canada) on a Storm phosphoimager, and quantified using Research, the National Cancer Institute of Canada and ImageQuant software (Molecular Dynamics, Piscataway, NJ, Genome British Columbia. Tissue DNA was archived using USA). funds from Grant R01 DE13124from NIDCR.

References

Battle MA, Maher VM and McCormick JJ. (2003). Biochem- Paris PL, Albertson DG, Alers JC, Andaya A, Carroll P, istry, 24, 7270–7282. Fridlyand J, Jain AN, Kamkar S, Kowbel D, Krijtenburg Chomczynski P and Sacchi N. (1987). Anal. Biochem., 162, PJ, Pinkel D, Schroder FH, Vissers KJ, Watson VJ, 156–159. Wildhagen MF, Collins C and Van Dekken H. (2003). Dong SM and Sidransky D. (2002). Clin. Cancer Res., 8, Am. J. Pathol., 162, 763–770. 2939–2941. Qing J, Wei D, Maher VM and McCormick JJ. (1999). Garnis C, Coe BP, Ishkanian A, Zhang L, Rosin MP and Lam Oncogene, 18, 335–342. WL. (2004). Genes Cancer, 39, 93–98. Rosin MP, Cheng X, Poh C, Lam WL, Huang Y, Lovas J, Karolchik D, Baertsch R, Diekhans M, Furey TS, Hinrichs A, Berean K, Epstein JB, Priddy R, Le ND and Zhang L. Lu YT, Roskin KM, Schwartz M, Sugnet CW, Thomas DJ, (2000). Clin. Cancer Res., 6, 357–362. Weber RJ, Haussler D and Kent WJ. (2003). Nucleic Acids Veltman JA, Fridlyand J, Pejavar S, Olshen AB, Korkola JE, Res., 31, 51–54. DeVries S, Carroll P, Kuo W, Pinkel D, Albertson D, Krajina Z, Kulcar Z and Konic-Carnelutti V. (1975). Cordon-Cardo C, Jain AN and Waldman FM. (2003). Laryngoscope, 7, 1155–1161. Cancer Res., 63, 2872–2880. Lowry S. (1975). Laryngoscope, 8, 1275–1280. Weiss MM, Kuipers EJ, Postma C, Snijders AM, Siccama I, Martinez-Climent JA, Alizadeh AA, Segraves R, Blesa D, Pinkel D, Westerga J, Meuwissen SG, Albertson DG and Rubio-Moscardo F, Albertson DG, Garcia-Conde J, Dyer Meijer GA. (2003). Oncogene, 22, 1872–1879. MJ, Levy R, Pinkel D and Lossos IS. (2003). Blood, 101, Zhang L, Michelsen C, Cheng X, Zeng T, Priddy R and Rosin 3109–3117. MP. (1997). Am. J. Pathol., 151, 323–327. Ng I, Xiao L, Lam KY, Yuen PW and Ng M. (2000). Oral Vokes EE, Weichselbaum RR, Lippman SM and Hong WK. Oncol., 36, 484–490. (1993). N. Engl. J. Med., 328, 184–194.

Oncogene