Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Amplification and Over-Expression of the MDM2 Gene in Human Soft Tissue Tumours

Amplification and Over-Expression of the MDM2 Gene in Human Soft Tissue Tumours HELEN PATTERSON ,1 D IANA BARS, 2 SAND RA GILL, 1 JAM ES SPICER,2 CYRIL FISHER, 3 M ERION THOM AS,4 ROBERT GRIM ER, 5 CHRIS FLETCHER,6 BARRY GUSTERSON 7 & COLIN COOPER 1 Section of M olecular Carcinogesis, and 7 Section of Cell Biology and Experimental Pathology, Institute of Cancer R esearch, Surrey, 2 Im perial Cancer R esearch Fund Clinical Oncology U nit, Guys H ospital, London, 3 D epartment of H istopathology, and 4D epartm ent of Surgery, Royal M arsden Hospital, London, 5 D epartm ent of Surgery, Royal O rthopaedic H osp ital, Birmingham & 6 D epartment of Histopathology, St T homas’ s Hospital, London, UK A bstract Purpose . Ampli® cation of getic sequences on chromosome 12q13 is frequently found in soft tissue tumours. However, for the M DM 2 ge, over-expression of the M DM 2 protein has not always been shown to accompany ge ampli® cation, raising the possibility that ampli® cation of getic sequences targets alternative ges on chrom osome 12q13 for over-expression. To investigate this discrepancy, we have examid 129 soft tissue tumours for ampli® cation of the M DM 2 ge using Southern analysis, and 39 of these tumours were also examid by immunohistochemical staining for M DM 2 over-expression. Results . Ge ampli® cation was identi® ed in 14/114 (12.3% ) of the malignant tumours, but was not identi® ed in any of the benign tumours; 21/39 (54% ) of the m alignant tumours also demonstrated MDM 2 over-expression. Within this group the M DM 2 ge was over-expressed in every tumour in which the ge ampli® cation was found, and over-expression in the absence of ge ampli® cation was also found in an additional 10 tumours. D iscussion . These data demonstrate a clear correlation between the presence of M DM 2 ampli® cation and M DM 2 over-expression, and provide persuasive evidence therefore that the ampli® cation of getic sequences on chromosome 12q13 in soft tissue sarcomas targets the M DM 2 ge for over-expression. These data also indicate that alternative m echanisms may contribute to M DM 2 over-expression within some tumours. K ey w ords: M DM2 ge, ampli® cation, over-expression, soft tissue sarcoma . Introduction A variety of m am m alian tumour cells contain ampli® ed copies of ges whose transfo rm ing potential is then activated by their concom itant overexpression. In keeping w ith this, double m inute chrom osomes and hom ogeously staining regions have been identi® ed in m alignant ® brous histiocytom as, liposarcom as, and alveolar and embryonal rhabdom yosarcom as. 1± 3 A cluster of ges which m ap to chrom osom e 12q13 have been found to be am pli® ed and over-expressed in soft tissue sarcom as, and include the M D M 2, GLI, C DK4, 4± 8 GA DD153 (C HO P) and SAS ges. The M D M 2 ge was originally identi® ed as a dom inantly transform ing oncoge am pli® ed and over-expressed from double minute chrom osom es in the 3T3D M cell li. 9 M D M 2 was subsequently show n to cooperate with RAS in the transfo rm ation of prim ary rat embryo ® broblasts, and to be am pli® ed in liposarcom as, m alignant ® brous histiocytomas (M FH s) and osteosarcomas. 4 M D M 2 has been show n to bind p53 and inhibit p53-m ediated transactivation 10 and is itself a target for p53m ediated ge regulation via a cis-acting p53 response elem ent in its ® rst intron. 11 Elevated p53 expression results in increased levels of M DM 2 m RN A and protein, which in turn binds to and m odulates p53 function. It has now also been show n that M D M 2 can bind and inhibit the growth regulatory function of RB 1 12 and can activate E2F 1/D P1 transcription factors which are involved in prom oting entry into the S-ph ase. 13 The oncogenic activity of M D M 2 m ay therefore derive from its ability both Correspondence to: H . Patterson, Section of Molecular Carcinogesis, Institute of Cancer Research, Haddow Laboratories, 15 Downs Road, Belm ont, Sutton, Surrey SM2 5N G, U K. Fax: 1 44 181 643 0238. 1357-714 X/97/010017± 06 Ó 1997 Journals Oxford Ltd H. Patterson et al. hyb ridization to the M D M 2 probe, blots were stripped by im m ersion in 0.1% SD S at 95 C and reprobed with pDCC 1.0 as a control to correct for differences betw een tum our sam ples in D N A loading and Southern transfer. 4 The DC C ge probe used was chosen as a control because allele loss at the D CC locus is found in only around 10% of soft tissue sarcom as, and hom ozygous deletion has not been demonstrated. 24 Allele loss without reduplication at the D CC locus in tum our D N A w ould result in a two-fold overestim ation of the degree of M DM 2 am pli® cation, and this error will be introduced in fewer than 10% of the tum ours exam id. T he degree of am pli® cation was quanti® ed with a Joyce-Loebel Chrom oscan 3 scanning densitom eter, using absorbance at 530 nm . To allow for inaccuracies due to the effects of strom al contam ination, and for the non-liar response of the radiographic ® lm , only sam ples showing greater than ® ve-fold am pli® cation w ere considered to be am pli® ed. to repress p53 function and to stim ulate the activity of S-phase prom oting transcription factors. A variety of studies have examid M D M 2 ampli® cation and over-expression in soft tissue tum ours. 14± 17 These studies, together with studies on other tum our types, have demonstrated that M D M 2 over-expression as assessed by im m unohistochemical staining is frequently found in the absence of 14,18,19 and that M D M 2 M DM 2 ge am pli® cation, over-expression may also be absent in tum ours in w hich am pli® cation has been demonstrated by Southern analysis. 14 These latter observations raise the possibility that ge am pli® cation on chrom osom e 12q13 m ay target a ge or ges other than M DM 2 for over-expression. W e therefore sought to further investigate the role of the M D M 2 ge in soft tissue tumour developm ent by analyzing a large series of m alignant and benign tumours for M DM 2 am pli® cation using Southern analysis and scanning densitom etry, and M D M 2 and p53 over-expression w as exam id in a subgroup of these tum ours using im munohistochemical staining of frozen sections. Immunoh istochemical staining of tum our specimens for M DM 2 and p53 expression T wo anti-M D M 2 m ouse m onoclonal antibodies M D M 2 (clo IF2) (David Hill, O ncoge Science Inc.) and SM P14, 25 and the p53 antibody D O1 (O ncoge Science Inc.) w ere used to stain frozen sections of primary tum our specim ens using a standard peroxidase-conjugated streptavidin-biotin m ethod (ABC). Slides om itting the ® rst antibody were used as gative controls in all cases. T wo tum ours known to be positive for either M D M 2 or p53 were included as positive controls in each batch of staining. T he degree of im m unostaining was graded as follow s: (gative) , 20% of cells staid positively, ( 1 ) 20± 50% of cells staid positively, ( 1 1 ) . 50% of cells stain positively. Results O hundred and twenty-ni tum ours (124 prim ary tumours and ® ve cell lis) including 29 leiom yosarcom as, 17 liposarcom as, 16 M FHs, 16 rhabdom yosarcomas, 12 m alignant peripheral rve sheath tum ours (M PN ST s), 10 synovial sarcom as, 3 osteosarcom as, 2 ® brosarcom as, 2 chondrosarcom as, 7 other sarcom as, 5 ® brom atoses and 10 benign soft tissue oplasm s w ere analyzed. G reater than ® ve-fo ld am pli® cation of M DM 2 was detected in 14/114 (12.3%) of the m alignant tum ours but in no of the ® brom atoses or benign tumours. T he positive sam ples included 3/16 (19% ) M FH s (STS11, ST S41 and ST S140), 4/17 (24%) liposarcom as (STS20, STS44, STS61 and STS131), 2/29 (7% ) leiomyosarcom as (STS87 and STS320), 2/16 (12.5% ) rhab dom yosarcom as (1 embryonal rhabdom yosarcoma STS93 and the alveolar rhabdom yosarcoma cell li RM S), 2/12 (17%) M PN ST s (STS52 and ST S102) and a single M aterials and m ethods C linical samples and cell lis F resh specimens of prim ary soft tissue tumours were obtaid during surgical resection from the Royal M arsden Hospital, London and Surrey, St T hom as’ s H ospital, London and the Royal Orthopaedic Hospital, Birmingham . Sam ples were im m ediately snap frozen in liquid nitrogen and stored at 2 70 C until processed. Hum an cell lis with the exception of RM S 20 were obtaid from the Am erican T ype C ulture Collection (AT CC) and m aintaid as recom m ended by their supplier. South ern analysis D N A was extracted using a m odi® cation of a m ethod described by Steffen and W einberg. 21 G enom ic D N A (10 m g) w as digested with a three- to ® ve-fold excess of HindIII or EcoRI, subjected to electrophoresis in 0.8% agarose gels and transferred to Hybond-N ® lters (Am ersham ) according to the m anufacturer’ s instructions. Filters were hybridized according to publish ed protocols 22 to a - 32P-labelled probes prepared using random oligonucleotide prim ers. 23 Filters w ere w ashed at 65 C to a ® nal stringency of 0.1 3 SSC, 0.1% SD S. Am pli® cation of the M DM 2 ge was assesse d using a 1.5-kb M D M 2 probe prepared by PC R am pli® cation of reverse-transcribed normal ® broblast RN A using the follow ing prim ers: G G G AG C T CC T C G C C AC C A T G G T G A G G AG C AGG CAAATG and GG GG TAC CC TCAT AGAC AG G TCAAC TAGG G G. The PCR product was subclod into the pBluescript vector and characterized by sequencing of its entire length. Following M DM 2 ampli® cation and over-expression in sarcomas T ab le 1. Am pli® cation of the MDM 2 ge in soft tissue tum ours Tumour type Leiomyosarcom a M FH Liposarcoma Rhabdomyosarcoma M PNST Synovial sarcoma Other sarcomas* Fibromatosis Benign tumours** Number examid 29 16 17 16 12 10 14 5 10 Number. ampli® ed 2 3 4 2 2 0 1 0 0 Percentage ampli® ed 7 19 24 12.5 17 0 7 0 0 F ig. 2. Imm unohistochem ical stainin g of primary tum ours for M DM 2 over-expression. Imm unoh istoch em ical staining of STS24, a high-grade M FH, using the mouse m onoclon al M DM 2 antibody M DM 2 (IF2) (O ncoge Science). G reater than 50% of the cells show positive staining for M DM 2. *Includes 3 osteosarcomas, 2 chondrosarcom as, 2 ® brosarcomas (o of which showed M DM 2 ampli® cation), 1 sarcoma NOS, 2 clear cell sarcomas, 1 derm ato® brosarcoma proturberans, 1 post-irradiation spindle cell sarcoma, 1 uroepitheliom a and 1 renal leiomyoblastoma. **Includes 3 haemangiomas, 3 uro® bromas, 2 urilemm omas, 1 paraganglioma and 1 angiolipoma. ® brosarcom a (STS49). T hese results are presented in Table 1, and exam ples of M DM 2 ge ampli® cation are illustrated in F ig. 1. ST S162, a recurrence of the liposarcom a ST S44, was not included in the original group of tum ours but was subsequently analyzed, and also demonstrated M DM 2 ge am pli® cation. Thirty-ni of the tum ours analyzed for M DM 2 am pli® cation w ere also exam id for M D M 2 overexpression by im m unohistochem ical staining (Table 2). These tum ours were exam id without knowledge of the M DM 2 ge am pli® cation results, but F ig. 1. Am pli® cation of the M DM2 ge in maligna nt soft tissue tum ours. Southe rn blot analysis of Eco RI-digested DNA probed sequentially with p DC C 1.0 and an M DM 2 cD NA probe, dem onstra ting M DM2 ge am pli® cation in several tum ours. The degree of ge am pli® cation was estim ated by using scanning densitom etric analysis of the resultin g autoradiographs. The loading is as follow s: la 1, norm al leucocy te DNA ; la 2, RM S cell li; la 3, STS93; la 4, STS102; la 5, norm al leucocyt e DN A; la 6, STS49; la 7, STS11; la 8, norm al leucocy te DNA ; la 9, STS41; la 10, STS44; la 11, STS61; la 12, norm al leucocy te DN A; la 13, STS140; la 14, STS162; la 15, norm al leucocyt e DNA. had been selected to include 11 of the tum ours with M DM 2 am pli® cation. This group included 8 liposarcom as, 8 M FHs, 6 leiom yosarcomas, 5 M PN STs, 3 rhabdom yosarcom as, 3 synovial sarcom as, 2 sarcom as N O S (not otherwise speci® ed), 1 clear cell sarcom a, 1 ® brosarcom a and 2 ® brom atoses. Of these, 21/39 (54%) including 5 M FH s, 4 liposarcom as (including ST S44 and its recurrence ST S162), 3 M PN STs, 3 rhabdom yosarcom as, 2 leiomyosarcom as, 2 synovial sarcom as, 1 ® brosarcom a and 1 clear cell sarcom a show ed positive im m unohistochem ical staining for M D M 2. An exam ple is show n in Fig. 2. All 11 of the tumours in w hich the M DM 2 ge was am pli® ed staid positively for M DM 2 over-expression, and M D M 2 over-expression was con® rm ed in the RM S cell li by N orthern analysis. C onversely, the M DM 2 ge was am pli® ed in 0/18 (0%) of the gatively staining tum ours. Of the six tum ours with strongly positive staining for M D M 2 ( . 50% of cells positive), 5/6 (83%) demonstrated M D M 2 am pli® cation, w hereas only 6/15 (40% ) of the relatively w eakly staining tum ours (20± 50% cells positive) demonstrated M DM 2 am pli® cation. The level of am pli® cation in weakly staining tumours w as, however, com parable to that seen in the m ore strongly staining tum ours. Am ong ® ve additional tum ours, for which data regarding M DM 2 ge am pli® cation were not availab le, overexpression w as identi® ed in tw o M FH s, ST S7 and STS9. p53 over-expression, as assessed by im m unohistochem ical staining, was identi® ed in four tum ours. In two of these tumours, ST S24 and STS102, the M DM 2 ge was also over-expressed, and in o of these tumours, ST S102, a m alignant peripheral rve sheath tum our, M D M 2 ge am pli® cation was found. H. Patterson et al. Tab le 2. Am pli® cation and/or over-expression of the M DM 2 ge in hum an soft tissue sarcom as M DM 2 ge ampli® cation* Tumour STS7 STS9 STS11 STS20 STS23 STS24 STS32 STS34 STS41 STS44 STS49 STS52 STS56 STS61 STS62 STS69 STS87 STS93 STS102 STS114 STS117 STS125 STS131 STS140 STS162 STS320 RM S Histology MFH MFH MFH Liposarcoma Pleom orphic rhabdom yosarcoma MFH Clear cell sarcoma Leiomyosarcoma MFH Liposarcoma Fibrosarcoma MPN ST MFH Liposarcoma MPN ST Alveolar rhabdom yosarcoma Leiomyosarcoma Embryonal rhabdom yosarcoma MPN ST Synovial sarcoma Liposarcoma Synovial sarcoma Liposarcoma MFH Liposarcoma Leiomyosarcoma Rhabdomyosarcoma Ð Grade High High High Low High High High High Low High Low High High Interm ediate High High Low High Low High High High Low High High High Source Recurrence Recurrence Recurrence Primary Primary Recurrence Primary Primary Recurrence Recurrence Recurrence Primary Recurrence Recurrence Primary Primary Metastasis Metastasis Recurrence Primary Recurrence Primary Primary Primary Recurrence Primary Cell li M DM2 overexpression** 36 7 ND ND ND ND 14 33 . 50 6 ND 44*** ND ND 50 50 20 ND ND ND 48 34 28 . 50 14 *Am pli® cation was detected by Southern blot analysis using an M DM 2 cDN A probe spanning the whole open reading frame. Scanning densitometry was used to compare the signal to that obtaid with the probe p DCC 1 .0. **Detected by immunohistochemical staining of frozen sections with the monoclonal antibody M DM 2 (Ab-1) (Oncoge Science). Staining was graded ( 1 ) 20± 50% cells stain positively, ( 1 1 ) . 50% of cells stain positively. ***Problem s with DN A degradation and digestion with this sample impair the accuracy of this result. 1 Northern analysis. 5 not evaluated; ND 5 not detected. D iscussion T his study con® rm s the observation of am pli® cation of the M DM 2 ge in soft tissue sarcom as which w as ® rst presented by O lir et al., 4 and extends the original observation of M DM 2 am pli® cation in M FH s, liposarcomas and osteosarcom as, to include leiom yosarcom as, M PN ST s, a prim ary embryonal rhabdom yosarcom a and an alveolar rhabdom yosarcom a cell li (RMS). In contrast to O lir et al. 4 w ho detected M DM 2 am pli® cation in 36% (17/47) of their m alignant sarcom as, we have detected ampli® cation in only 12.3% of this sam ple of m alignant tum ours. These results are, however, in keeping w ith the results of Cordon-Cardo et al. 14 who detected a greater than ® ve-fo ld am pli® cation of the M DM 2 ge in only 15% (11/73) of their sarcom as, w ith Forus et al. 7 who detected M DM 2 ampli® cation in 9% (9/98) of their sarcom as, and with the results of Ladanyi et al. 26 who detected M DM 2 am pli® cation in 14% (4/28) of their high-grade osteosarcom as. M D M 2 am pli® cation was seen in both high- and low-grade tum ours, as well as in prim ary, recurrent and m etastatic tum ours (Table 2). T here did not app ear to be any clear correlation between the degree of ampli® cation and the histological subtype or grade, or whether the source of the tumour specim en w as a prim ary, recurrent or m etastatic tumour. T hese results indicate that am pli® cation of the M DM 2 ge m ay be im portant in the developm ent of a signi® cant m inority of a w ide variety of histological subtypes of m alignant soft tissue tum our, but they provide no support for a role for M DM 2 ge am pli® cation in the developm ent of benign soft tissue oplasm s, or locally aggressive tum ours such as ® brom atoses. W hereas only 11/39 of the tum ours selected for M D M 2 im m unohistochemical staining possessed M DM 2 ge am pli® cation, 21/39 of the tumours in this subgroup analyzed w ere im m unostain positive for M D M 2 over-expression. O ther studies are in broad agreement with these results. Cordon-C ardo M DM 2 ampli® cation and over-expression in sarcomas et al. 14 identi® ed M D M 2 over-expression in 37% (76/207) of their sarcom as, w ith ge am pli® cation present in only 15% (11/73) of a subset of these tum ours, and Lias et al. 18 dem onstrated M D M 2 over-expression in 26/87 human bladder tumours, but only 1/26 of these tum ours dem onstrated M DM 2 ge am pli® cation. These results suggest that m echanism s other than ge am pli® cation m ay be im portant in raising cellular levels of the M D M 2 protein; for exam ple, alterations in the stability of the MD M 2 protein, or in the rate of transcription or translation or of the stability of the M D M2 m RN A. O nly 6/11 of the tum ours with M DM 2 ge ampli® cation reported by Cordon-C ardo et al. 14 showed positive im m unohistochemical staining ( . 20% cells positive) for M D M 2 over-expression, raising the possibility that in som e sarcom as the target for ge am pli® cation on chrom osom e 12q13 m ay not be M DM 2. The results presented here, how ever, have dem onstrated that M DM 2 am pli® cation accurately predicts for M D M 2 overexpression. T his com plete correlation is corroborated by three other studies with soft tissue tumours. Leach et al. 15 identi® ed M D M 2 ge am pli® cation and M D M 2 over-expression in 8/24 M FHs and liposarcom as, F lores et al. 16 dem onstrated M DM 2 ge am pli® cation and over-expression in 10/98 bo and soft tissue tum ours, and N akayam a et al. 17 identi® ed M D M 2 ge am pli® cation, and overexpression by N orthern analysis, in each of 11/48 soft tissue tumours. These observations, the demonstration that M DM 2 inhibits transactivation by the product of the p53 ge (a ge known to contribute to the oplastic process in sarcom as) and the dem onstration that M D M 2 is a dominantly transform ing oncoge 9 provide strong evidence in favour of a role for M DM 2 am pli® cation and over-expression in the gesis of a subgroup of soft tissue sarcom as. W e have identi® ed over-expression of both p53 and M D M 2 in two tum ours, ST S24 and STS102, o of which possessed M D M 2 ge am pli® cation. T he status of the p53 ge in these two tum ours has not been exam id. H ow ever, when sim ilar ® ndings have been demonstrated in other tum ours, 16,18 overexpression of the M DM 2 protein was accom panied by over-expression of wild-type and not m utant p53 protein. For exam ple, Landers et al. 27 exam id choriocarcinom a cell lis and found overexpression of both wild-type p53 and M D M 2. T he over-expression of MD M 2 w as not explaid by ge am pli® cation, elevated transcription or altered protein stability, but appeared to have resulted from increased protein translation. In addition to these observations, the level of M D M 2 transcription is known to be regulated by the expression of wildtype p53. 11 Additional data regarding the status of the p53 ge w ere available for 38 of the tumours exam id for ge am pli® cation MDM2 (29 leiom yosarcom as, 4 cell lis, 1 primary rhab- dom yosarcoma, 2 M FHs, 1 liposarcom a and 1 postirradiation spindle cell tum our), and ni of these had also been exam id for M D M 2 overexpression. Thirteen tumours possessed p53 m utations, seven had undergo hom ozygous rearrangem ent of the p53 ge and six had m is-se nse point m utations. 24,28 Tw o of these tum ours, ST S7, an M FH , and STS117, a liposarcom a, both of which had undergo hom ozygous rearrangement at the p53 locus, also demonstrated M D M 2 overexpression. In STS117 the over-expression was found in the absence of ge am pli® cation. T he status of wild-type p53 protein expression in tum our cells m ay, therefore, m odulate the level of M DM 2 protein expression, thus providing a m echanism for M D M 2 over-expression in the absence of ge am pli® cation in a subgroup of tum our cells. A cknowledgem ents W e would like to thank Professor Bert V ogelstein for providing the probe pDC C1.0, and D r Alasd air Stamps for providing the M D M 2 cDN A probe. T he m ouse m onoclonal antibody M D M 2 (IF2) was a kind gift from D avid Hill, Oncoge Science Inc. T his study was supported by grants from the Cancer Research Cam paign and the M edical Research C ouncil. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Sarcoma Hindawi Publishing Corporation

Amplification and Over-Expression of the MDM2 Gene in Human Soft Tissue Tumours

Download PDF
7 pages

Loading next page...
 
/lp/hindawi-publishing-corporation/amplification-and-over-expression-of-the-mdm2-gene-in-human-soft-DUNZJGKcEO

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 1997 Hindawi Publishing Corporation.
ISSN
1357-714X
eISSN
1369-1643
Publisher site
See Article on Publisher Site

Abstract

HELEN PATTERSON ,1 D IANA BARS, 2 SAND RA GILL, 1 JAM ES SPICER,2 CYRIL FISHER, 3 M ERION THOM AS,4 ROBERT GRIM ER, 5 CHRIS FLETCHER,6 BARRY GUSTERSON 7 & COLIN COOPER 1 Section of M olecular Carcinogesis, and 7 Section of Cell Biology and Experimental Pathology, Institute of Cancer R esearch, Surrey, 2 Im perial Cancer R esearch Fund Clinical Oncology U nit, Guys H ospital, London, 3 D epartment of H istopathology, and 4D epartm ent of Surgery, Royal M arsden Hospital, London, 5 D epartm ent of Surgery, Royal O rthopaedic H osp ital, Birmingham & 6 D epartment of Histopathology, St T homas’ s Hospital, London, UK A bstract Purpose . Ampli® cation of getic sequences on chromosome 12q13 is frequently found in soft tissue tumours. However, for the M DM 2 ge, over-expression of the M DM 2 protein has not always been shown to accompany ge ampli® cation, raising the possibility that ampli® cation of getic sequences targets alternative ges on chrom osome 12q13 for over-expression. To investigate this discrepancy, we have examid 129 soft tissue tumours for ampli® cation of the M DM 2 ge using Southern analysis, and 39 of these tumours were also examid by immunohistochemical staining for M DM 2 over-expression. Results . Ge ampli® cation was identi® ed in 14/114 (12.3% ) of the malignant tumours, but was not identi® ed in any of the benign tumours; 21/39 (54% ) of the m alignant tumours also demonstrated MDM 2 over-expression. Within this group the M DM 2 ge was over-expressed in every tumour in which the ge ampli® cation was found, and over-expression in the absence of ge ampli® cation was also found in an additional 10 tumours. D iscussion . These data demonstrate a clear correlation between the presence of M DM 2 ampli® cation and M DM 2 over-expression, and provide persuasive evidence therefore that the ampli® cation of getic sequences on chromosome 12q13 in soft tissue sarcomas targets the M DM 2 ge for over-expression. These data also indicate that alternative m echanisms may contribute to M DM 2 over-expression within some tumours. K ey w ords: M DM2 ge, ampli® cation, over-expression, soft tissue sarcoma . Introduction A variety of m am m alian tumour cells contain ampli® ed copies of ges whose transfo rm ing potential is then activated by their concom itant overexpression. In keeping w ith this, double m inute chrom osomes and hom ogeously staining regions have been identi® ed in m alignant ® brous histiocytom as, liposarcom as, and alveolar and embryonal rhabdom yosarcom as. 1± 3 A cluster of ges which m ap to chrom osom e 12q13 have been found to be am pli® ed and over-expressed in soft tissue sarcom as, and include the M D M 2, GLI, C DK4, 4± 8 GA DD153 (C HO P) and SAS ges. The M D M 2 ge was originally identi® ed as a dom inantly transform ing oncoge am pli® ed and over-expressed from double minute chrom osom es in the 3T3D M cell li. 9 M D M 2 was subsequently show n to cooperate with RAS in the transfo rm ation of prim ary rat embryo ® broblasts, and to be am pli® ed in liposarcom as, m alignant ® brous histiocytomas (M FH s) and osteosarcomas. 4 M D M 2 has been show n to bind p53 and inhibit p53-m ediated transactivation 10 and is itself a target for p53m ediated ge regulation via a cis-acting p53 response elem ent in its ® rst intron. 11 Elevated p53 expression results in increased levels of M DM 2 m RN A and protein, which in turn binds to and m odulates p53 function. It has now also been show n that M D M 2 can bind and inhibit the growth regulatory function of RB 1 12 and can activate E2F 1/D P1 transcription factors which are involved in prom oting entry into the S-ph ase. 13 The oncogenic activity of M D M 2 m ay therefore derive from its ability both Correspondence to: H . Patterson, Section of Molecular Carcinogesis, Institute of Cancer Research, Haddow Laboratories, 15 Downs Road, Belm ont, Sutton, Surrey SM2 5N G, U K. Fax: 1 44 181 643 0238. 1357-714 X/97/010017± 06 Ó 1997 Journals Oxford Ltd H. Patterson et al. hyb ridization to the M D M 2 probe, blots were stripped by im m ersion in 0.1% SD S at 95 C and reprobed with pDCC 1.0 as a control to correct for differences betw een tum our sam ples in D N A loading and Southern transfer. 4 The DC C ge probe used was chosen as a control because allele loss at the D CC locus is found in only around 10% of soft tissue sarcom as, and hom ozygous deletion has not been demonstrated. 24 Allele loss without reduplication at the D CC locus in tum our D N A w ould result in a two-fold overestim ation of the degree of M DM 2 am pli® cation, and this error will be introduced in fewer than 10% of the tum ours exam id. T he degree of am pli® cation was quanti® ed with a Joyce-Loebel Chrom oscan 3 scanning densitom eter, using absorbance at 530 nm . To allow for inaccuracies due to the effects of strom al contam ination, and for the non-liar response of the radiographic ® lm , only sam ples showing greater than ® ve-fold am pli® cation w ere considered to be am pli® ed. to repress p53 function and to stim ulate the activity of S-phase prom oting transcription factors. A variety of studies have examid M D M 2 ampli® cation and over-expression in soft tissue tum ours. 14± 17 These studies, together with studies on other tum our types, have demonstrated that M D M 2 over-expression as assessed by im m unohistochemical staining is frequently found in the absence of 14,18,19 and that M D M 2 M DM 2 ge am pli® cation, over-expression may also be absent in tum ours in w hich am pli® cation has been demonstrated by Southern analysis. 14 These latter observations raise the possibility that ge am pli® cation on chrom osom e 12q13 m ay target a ge or ges other than M DM 2 for over-expression. W e therefore sought to further investigate the role of the M D M 2 ge in soft tissue tumour developm ent by analyzing a large series of m alignant and benign tumours for M DM 2 am pli® cation using Southern analysis and scanning densitom etry, and M D M 2 and p53 over-expression w as exam id in a subgroup of these tum ours using im munohistochemical staining of frozen sections. Immunoh istochemical staining of tum our specimens for M DM 2 and p53 expression T wo anti-M D M 2 m ouse m onoclonal antibodies M D M 2 (clo IF2) (David Hill, O ncoge Science Inc.) and SM P14, 25 and the p53 antibody D O1 (O ncoge Science Inc.) w ere used to stain frozen sections of primary tum our specim ens using a standard peroxidase-conjugated streptavidin-biotin m ethod (ABC). Slides om itting the ® rst antibody were used as gative controls in all cases. T wo tum ours known to be positive for either M D M 2 or p53 were included as positive controls in each batch of staining. T he degree of im m unostaining was graded as follow s: (gative) , 20% of cells staid positively, ( 1 ) 20± 50% of cells staid positively, ( 1 1 ) . 50% of cells stain positively. Results O hundred and twenty-ni tum ours (124 prim ary tumours and ® ve cell lis) including 29 leiom yosarcom as, 17 liposarcom as, 16 M FHs, 16 rhabdom yosarcomas, 12 m alignant peripheral rve sheath tum ours (M PN ST s), 10 synovial sarcom as, 3 osteosarcom as, 2 ® brosarcom as, 2 chondrosarcom as, 7 other sarcom as, 5 ® brom atoses and 10 benign soft tissue oplasm s w ere analyzed. G reater than ® ve-fo ld am pli® cation of M DM 2 was detected in 14/114 (12.3%) of the m alignant tum ours but in no of the ® brom atoses or benign tumours. T he positive sam ples included 3/16 (19% ) M FH s (STS11, ST S41 and ST S140), 4/17 (24%) liposarcom as (STS20, STS44, STS61 and STS131), 2/29 (7% ) leiomyosarcom as (STS87 and STS320), 2/16 (12.5% ) rhab dom yosarcom as (1 embryonal rhabdom yosarcoma STS93 and the alveolar rhabdom yosarcoma cell li RM S), 2/12 (17%) M PN ST s (STS52 and ST S102) and a single M aterials and m ethods C linical samples and cell lis F resh specimens of prim ary soft tissue tumours were obtaid during surgical resection from the Royal M arsden Hospital, London and Surrey, St T hom as’ s H ospital, London and the Royal Orthopaedic Hospital, Birmingham . Sam ples were im m ediately snap frozen in liquid nitrogen and stored at 2 70 C until processed. Hum an cell lis with the exception of RM S 20 were obtaid from the Am erican T ype C ulture Collection (AT CC) and m aintaid as recom m ended by their supplier. South ern analysis D N A was extracted using a m odi® cation of a m ethod described by Steffen and W einberg. 21 G enom ic D N A (10 m g) w as digested with a three- to ® ve-fold excess of HindIII or EcoRI, subjected to electrophoresis in 0.8% agarose gels and transferred to Hybond-N ® lters (Am ersham ) according to the m anufacturer’ s instructions. Filters were hybridized according to publish ed protocols 22 to a - 32P-labelled probes prepared using random oligonucleotide prim ers. 23 Filters w ere w ashed at 65 C to a ® nal stringency of 0.1 3 SSC, 0.1% SD S. Am pli® cation of the M DM 2 ge was assesse d using a 1.5-kb M D M 2 probe prepared by PC R am pli® cation of reverse-transcribed normal ® broblast RN A using the follow ing prim ers: G G G AG C T CC T C G C C AC C A T G G T G A G G AG C AGG CAAATG and GG GG TAC CC TCAT AGAC AG G TCAAC TAGG G G. The PCR product was subclod into the pBluescript vector and characterized by sequencing of its entire length. Following M DM 2 ampli® cation and over-expression in sarcomas T ab le 1. Am pli® cation of the MDM 2 ge in soft tissue tum ours Tumour type Leiomyosarcom a M FH Liposarcoma Rhabdomyosarcoma M PNST Synovial sarcoma Other sarcomas* Fibromatosis Benign tumours** Number examid 29 16 17 16 12 10 14 5 10 Number. ampli® ed 2 3 4 2 2 0 1 0 0 Percentage ampli® ed 7 19 24 12.5 17 0 7 0 0 F ig. 2. Imm unohistochem ical stainin g of primary tum ours for M DM 2 over-expression. Imm unoh istoch em ical staining of STS24, a high-grade M FH, using the mouse m onoclon al M DM 2 antibody M DM 2 (IF2) (O ncoge Science). G reater than 50% of the cells show positive staining for M DM 2. *Includes 3 osteosarcomas, 2 chondrosarcom as, 2 ® brosarcomas (o of which showed M DM 2 ampli® cation), 1 sarcoma NOS, 2 clear cell sarcomas, 1 derm ato® brosarcoma proturberans, 1 post-irradiation spindle cell sarcoma, 1 uroepitheliom a and 1 renal leiomyoblastoma. **Includes 3 haemangiomas, 3 uro® bromas, 2 urilemm omas, 1 paraganglioma and 1 angiolipoma. ® brosarcom a (STS49). T hese results are presented in Table 1, and exam ples of M DM 2 ge ampli® cation are illustrated in F ig. 1. ST S162, a recurrence of the liposarcom a ST S44, was not included in the original group of tum ours but was subsequently analyzed, and also demonstrated M DM 2 ge am pli® cation. Thirty-ni of the tum ours analyzed for M DM 2 am pli® cation w ere also exam id for M D M 2 overexpression by im m unohistochem ical staining (Table 2). These tum ours were exam id without knowledge of the M DM 2 ge am pli® cation results, but F ig. 1. Am pli® cation of the M DM2 ge in maligna nt soft tissue tum ours. Southe rn blot analysis of Eco RI-digested DNA probed sequentially with p DC C 1.0 and an M DM 2 cD NA probe, dem onstra ting M DM2 ge am pli® cation in several tum ours. The degree of ge am pli® cation was estim ated by using scanning densitom etric analysis of the resultin g autoradiographs. The loading is as follow s: la 1, norm al leucocy te DNA ; la 2, RM S cell li; la 3, STS93; la 4, STS102; la 5, norm al leucocyt e DN A; la 6, STS49; la 7, STS11; la 8, norm al leucocy te DNA ; la 9, STS41; la 10, STS44; la 11, STS61; la 12, norm al leucocy te DN A; la 13, STS140; la 14, STS162; la 15, norm al leucocyt e DNA. had been selected to include 11 of the tum ours with M DM 2 am pli® cation. This group included 8 liposarcom as, 8 M FHs, 6 leiom yosarcomas, 5 M PN STs, 3 rhabdom yosarcom as, 3 synovial sarcom as, 2 sarcom as N O S (not otherwise speci® ed), 1 clear cell sarcom a, 1 ® brosarcom a and 2 ® brom atoses. Of these, 21/39 (54%) including 5 M FH s, 4 liposarcom as (including ST S44 and its recurrence ST S162), 3 M PN STs, 3 rhabdom yosarcom as, 2 leiomyosarcom as, 2 synovial sarcom as, 1 ® brosarcom a and 1 clear cell sarcom a show ed positive im m unohistochem ical staining for M D M 2. An exam ple is show n in Fig. 2. All 11 of the tumours in w hich the M DM 2 ge was am pli® ed staid positively for M DM 2 over-expression, and M D M 2 over-expression was con® rm ed in the RM S cell li by N orthern analysis. C onversely, the M DM 2 ge was am pli® ed in 0/18 (0%) of the gatively staining tum ours. Of the six tum ours with strongly positive staining for M D M 2 ( . 50% of cells positive), 5/6 (83%) demonstrated M D M 2 am pli® cation, w hereas only 6/15 (40% ) of the relatively w eakly staining tum ours (20± 50% cells positive) demonstrated M DM 2 am pli® cation. The level of am pli® cation in weakly staining tumours w as, however, com parable to that seen in the m ore strongly staining tum ours. Am ong ® ve additional tum ours, for which data regarding M DM 2 ge am pli® cation were not availab le, overexpression w as identi® ed in tw o M FH s, ST S7 and STS9. p53 over-expression, as assessed by im m unohistochem ical staining, was identi® ed in four tum ours. In two of these tumours, ST S24 and STS102, the M DM 2 ge was also over-expressed, and in o of these tumours, ST S102, a m alignant peripheral rve sheath tum our, M D M 2 ge am pli® cation was found. H. Patterson et al. Tab le 2. Am pli® cation and/or over-expression of the M DM 2 ge in hum an soft tissue sarcom as M DM 2 ge ampli® cation* Tumour STS7 STS9 STS11 STS20 STS23 STS24 STS32 STS34 STS41 STS44 STS49 STS52 STS56 STS61 STS62 STS69 STS87 STS93 STS102 STS114 STS117 STS125 STS131 STS140 STS162 STS320 RM S Histology MFH MFH MFH Liposarcoma Pleom orphic rhabdom yosarcoma MFH Clear cell sarcoma Leiomyosarcoma MFH Liposarcoma Fibrosarcoma MPN ST MFH Liposarcoma MPN ST Alveolar rhabdom yosarcoma Leiomyosarcoma Embryonal rhabdom yosarcoma MPN ST Synovial sarcoma Liposarcoma Synovial sarcoma Liposarcoma MFH Liposarcoma Leiomyosarcoma Rhabdomyosarcoma Ð Grade High High High Low High High High High Low High Low High High Interm ediate High High Low High Low High High High Low High High High Source Recurrence Recurrence Recurrence Primary Primary Recurrence Primary Primary Recurrence Recurrence Recurrence Primary Recurrence Recurrence Primary Primary Metastasis Metastasis Recurrence Primary Recurrence Primary Primary Primary Recurrence Primary Cell li M DM2 overexpression** 36 7 ND ND ND ND 14 33 . 50 6 ND 44*** ND ND 50 50 20 ND ND ND 48 34 28 . 50 14 *Am pli® cation was detected by Southern blot analysis using an M DM 2 cDN A probe spanning the whole open reading frame. Scanning densitometry was used to compare the signal to that obtaid with the probe p DCC 1 .0. **Detected by immunohistochemical staining of frozen sections with the monoclonal antibody M DM 2 (Ab-1) (Oncoge Science). Staining was graded ( 1 ) 20± 50% cells stain positively, ( 1 1 ) . 50% of cells stain positively. ***Problem s with DN A degradation and digestion with this sample impair the accuracy of this result. 1 Northern analysis. 5 not evaluated; ND 5 not detected. D iscussion T his study con® rm s the observation of am pli® cation of the M DM 2 ge in soft tissue sarcom as which w as ® rst presented by O lir et al., 4 and extends the original observation of M DM 2 am pli® cation in M FH s, liposarcomas and osteosarcom as, to include leiom yosarcom as, M PN ST s, a prim ary embryonal rhabdom yosarcom a and an alveolar rhabdom yosarcom a cell li (RMS). In contrast to O lir et al. 4 w ho detected M DM 2 am pli® cation in 36% (17/47) of their m alignant sarcom as, we have detected ampli® cation in only 12.3% of this sam ple of m alignant tum ours. These results are, however, in keeping w ith the results of Cordon-Cardo et al. 14 who detected a greater than ® ve-fo ld am pli® cation of the M DM 2 ge in only 15% (11/73) of their sarcom as, w ith Forus et al. 7 who detected M DM 2 ampli® cation in 9% (9/98) of their sarcom as, and with the results of Ladanyi et al. 26 who detected M DM 2 am pli® cation in 14% (4/28) of their high-grade osteosarcom as. M D M 2 am pli® cation was seen in both high- and low-grade tum ours, as well as in prim ary, recurrent and m etastatic tum ours (Table 2). T here did not app ear to be any clear correlation between the degree of ampli® cation and the histological subtype or grade, or whether the source of the tumour specim en w as a prim ary, recurrent or m etastatic tumour. T hese results indicate that am pli® cation of the M DM 2 ge m ay be im portant in the developm ent of a signi® cant m inority of a w ide variety of histological subtypes of m alignant soft tissue tum our, but they provide no support for a role for M DM 2 ge am pli® cation in the developm ent of benign soft tissue oplasm s, or locally aggressive tum ours such as ® brom atoses. W hereas only 11/39 of the tum ours selected for M D M 2 im m unohistochemical staining possessed M DM 2 ge am pli® cation, 21/39 of the tumours in this subgroup analyzed w ere im m unostain positive for M D M 2 over-expression. O ther studies are in broad agreement with these results. Cordon-C ardo M DM 2 ampli® cation and over-expression in sarcomas et al. 14 identi® ed M D M 2 over-expression in 37% (76/207) of their sarcom as, w ith ge am pli® cation present in only 15% (11/73) of a subset of these tum ours, and Lias et al. 18 dem onstrated M D M 2 over-expression in 26/87 human bladder tumours, but only 1/26 of these tum ours dem onstrated M DM 2 ge am pli® cation. These results suggest that m echanism s other than ge am pli® cation m ay be im portant in raising cellular levels of the M D M 2 protein; for exam ple, alterations in the stability of the MD M 2 protein, or in the rate of transcription or translation or of the stability of the M D M2 m RN A. O nly 6/11 of the tum ours with M DM 2 ge ampli® cation reported by Cordon-C ardo et al. 14 showed positive im m unohistochemical staining ( . 20% cells positive) for M D M 2 over-expression, raising the possibility that in som e sarcom as the target for ge am pli® cation on chrom osom e 12q13 m ay not be M DM 2. The results presented here, how ever, have dem onstrated that M DM 2 am pli® cation accurately predicts for M D M 2 overexpression. T his com plete correlation is corroborated by three other studies with soft tissue tumours. Leach et al. 15 identi® ed M D M 2 ge am pli® cation and M D M 2 over-expression in 8/24 M FHs and liposarcom as, F lores et al. 16 dem onstrated M DM 2 ge am pli® cation and over-expression in 10/98 bo and soft tissue tum ours, and N akayam a et al. 17 identi® ed M D M 2 ge am pli® cation, and overexpression by N orthern analysis, in each of 11/48 soft tissue tumours. These observations, the demonstration that M DM 2 inhibits transactivation by the product of the p53 ge (a ge known to contribute to the oplastic process in sarcom as) and the dem onstration that M D M 2 is a dominantly transform ing oncoge 9 provide strong evidence in favour of a role for M DM 2 am pli® cation and over-expression in the gesis of a subgroup of soft tissue sarcom as. W e have identi® ed over-expression of both p53 and M D M 2 in two tum ours, ST S24 and STS102, o of which possessed M D M 2 ge am pli® cation. T he status of the p53 ge in these two tum ours has not been exam id. H ow ever, when sim ilar ® ndings have been demonstrated in other tum ours, 16,18 overexpression of the M DM 2 protein was accom panied by over-expression of wild-type and not m utant p53 protein. For exam ple, Landers et al. 27 exam id choriocarcinom a cell lis and found overexpression of both wild-type p53 and M D M 2. T he over-expression of MD M 2 w as not explaid by ge am pli® cation, elevated transcription or altered protein stability, but appeared to have resulted from increased protein translation. In addition to these observations, the level of M D M 2 transcription is known to be regulated by the expression of wildtype p53. 11 Additional data regarding the status of the p53 ge w ere available for 38 of the tumours exam id for ge am pli® cation MDM2 (29 leiom yosarcom as, 4 cell lis, 1 primary rhab- dom yosarcoma, 2 M FHs, 1 liposarcom a and 1 postirradiation spindle cell tum our), and ni of these had also been exam id for M D M 2 overexpression. Thirteen tumours possessed p53 m utations, seven had undergo hom ozygous rearrangem ent of the p53 ge and six had m is-se nse point m utations. 24,28 Tw o of these tum ours, ST S7, an M FH , and STS117, a liposarcom a, both of which had undergo hom ozygous rearrangement at the p53 locus, also demonstrated M D M 2 overexpression. In STS117 the over-expression was found in the absence of ge am pli® cation. T he status of wild-type p53 protein expression in tum our cells m ay, therefore, m odulate the level of M DM 2 protein expression, thus providing a m echanism for M D M 2 over-expression in the absence of ge am pli® cation in a subgroup of tum our cells. A cknowledgem ents W e would like to thank Professor Bert V ogelstein for providing the probe pDC C1.0, and D r Alasd air Stamps for providing the M D M 2 cDN A probe. T he m ouse m onoclonal antibody M D M 2 (IF2) was a kind gift from D avid Hill, Oncoge Science Inc. T his study was supported by grants from the Cancer Research Cam paign and the M edical Research C ouncil.

Journal

Sarcoma – Hindawi Publishing Corporation

Published: Jul 15, 2015

There are no references for this article.