- Email: [email protected]
Human Sarcomas Are Mosaic for Telomerase-Dependent and Telomerase-Independent Telomere Maintenance Mechanisms
The American Journal of Pathology, Vol. 182, No. 1, January 2013
ajp.amjpathol.org
SHORT COMMUNICATION Human Sarcomas Are Mosaic for Telomerase-Dependent and Telomerase-Independent Telomere Maintenance Mechanisms Implications for Telomere-Based Therapies April R.S. Gocha,* Gerard Nuovo,y Obiajulu H. Iwenofu,y and Joanna Groden* From the Departments of Molecular Virology, Immunology, and Medical Genetics* and Pathology,y Ohio State University College of Medicine, Columbus, Ohio Accepted for publication October 2, 2012. Address correspondence to Joanna Groden, Ph.D., 986 Biomedical Research Tower, 460 W. 12th Ave, Columbus, OH 43210. E-mail: joanna. [email protected].
Telomere shortening necessitates that tumor cells activate a telomere maintenance mechanism (TMM) to support immortalization. Although most tumor cells activate expression of the enzyme telomerase, some cells elongate telomeres in a telomerase-independent manner, termed alternative lengthening of telomeres (ALT). Previous studies have evaluated the presence of telomerase or ALT mechanisms or both in a variety of tumor types. Our studies also show that TMMs are not mutually exclusive in some tumors. In contrast, our IHC analyses of human sarcomas identified a subset of tumors with some cells containing ALT-associated PML bodies, a hallmark of ALT, and separate cells expressing telomerase in the same tumor. By using a second set of human osteosarcomas, we merged IHC and biochemical analyses to characterize more fully the tumor TMM. The IHC data reveal the presence of both telomerase- and ALT-positive tumor cells in samples that demonstrate characteristics of both telomerase and ALT in biochemical assays. These assays, which measure telomere length and telomerase activity of tumor extracts, are conventionally used to classify tumor TMM. Our results suggest that TMM is not a single or perhaps static characteristic of some tumors and that TMM heterogeneity should be considered in tumor stratification. Furthermore, clinical interest in telomere-based therapies may necessitate accurate characterization of tumor TMM before treatment to maximize therapeutic efficacy. (Am J Pathol 2013, 182: 41e48; http://dx.doi.org/10.1016/j.ajpath.2012.10.001)
Activation of a telomere maintenance mechanism (TMM) is critical for tumor cell immortalization and cancer progression because telomeres shorten with each round of cellular division. Cells can use two known TMMs: catalytic activity of the enzyme telomerase or a telomerase-independent, recombination-based pathway termed alternative lengthening of telomeres (ALT). In contrast to cells with active telomerase, cells that use ALT are characterized by long, heterogeneous telomere lengths, ALT-associated PML bodies (APBs), and extrachromosomal telomeric repeats. Several studies have evaluated TMMs in human tumors1,2 and demonstrated that ALT is overrepresented in mesenchymal tumors,2e12 perhaps reflective of a lack of detectable telomerase expression in mesenchymal stem cells.13 More specifically, almost 60% of osteosarcomas exhibit ALT characteristics,3e6 the highest incidence of any tumor yet evaluated. In contrast, epithelial Copyright ª 2013 American Society for Investigative Pathology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajpath.2012.10.001
tumors more often express telomerase,1,2 suggesting there may be cell-specific preferences for TMM. Studies have demonstrated that either type of TMM can occur within the same cell, but only with experimental manipulation. Exogenous expression of functional telomerase within some ALT cell lines results in telomerase-mediated elongation of short telomeres, although cells still retain ALT characteristics14e16; studies in other ALT cell lines have shown that telomerase expression inhibits ALT characteristics.17 Fusion experiments of telomerase-expressing and ALT cells have shown that the mechanisms governing telomere maintenance are complex. Some telomerase-expressing cells Supported by The Ohio State University Pelotonia Fellowship Program (A.R.S.G.), the Bloom’s Syndrome Foundation (J.G.), and NIH grant CA-117898 (J.G.).
Gocha et al Table 1
Analysis of TMM in Human Sarcoma TMAs
Patient no.
Age (years)
Sex
Tumor
Site
Stage
TNM grading
ALT (%)
Tel (%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
30 39 42 49 28 33 52 42 35 38 46 35 16 42 40 33 38 58 28 40 34 62 65 29 48 24 52 7 34 32 41 18 43 19 40 13 17 64 23 51 13 37 18 15 31 32 19 29 24 32 32 30 14 37 14 17 30
M F F F F M M F F M F M F M M M M F M M M M M M F F F F F M M M M M F M M M M M F M M M M F M M M F F F F M F M M
Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Osteoclastoma Osteoclastoma Osteoclastoma Osteoclastoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma
Leg Scapula ND ND ND ND ND ND ND ND ND Humerus Femur ND ND ND Face ND ND ND Leg Maxilla ND ND ND ND ND Femur ND ND Femur Humerus Costal Femur ND ND ND ND ND Femur Femur ND ND Femur ND ND Leg ND Skull ND ND ND Tibia Femur ND Femur ND
ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND IIb ND ND ND ND ND II ND II II ND IIb IIb ND ND ND ND ND ND IIIb ND ND ND ND ND ND IIb ND IIb ND ND ND IIIb IIb ND IIb ND
ND ND T2N0M0 T2N0M0 T2N0M0 T1N0M0 T1N0M0 T2N0M0 T2N0M0 T2N0M0 T2N0M0 ND ND T2N0M0 ND T2N0M0 ND T2N0M0 T1N0M0 T2N0M0 T2N0M0 ND T2N0M0 T1N0M0 T2N0M0 T1N0M0 T1N0M0 ND T1N0M0 T1N0M0 ND T2N0M0 T2N0M0 ND T2N0M0 T2N0M0 T2N0M0 T2N0M0 T2N0M0 T2N0M0 ND T1N0M0 T2N0M0 ND T2N0M0 T2N0M0 T2N0M0 T1N0M0 T2N0M0 T2N0M0 T2N0M0 T2N0M0 T2N0M0 T1N0M0 T2N0M0 T2N0M0 T2N0M0
0 0 0 0 0 0 0 0 0 0 41 29 5 29 25 12 9 6 5 2 1 1 0 0 0 0 0 11 5 2 0 0 0 0 0 0 0 0 0 0 0 0 20 47 20 17 15 15 9 5 4 4 3 3 2 1 1
95 95 75 75 40 40 40 33 32 30 0 0 0 79 53 33 66 34 85 82 3 2 0 0 0 0 69 32 95 87 95 90 90 90 74 62 50 38 30 16 14 13 0 24 16 38 62 11 2 2 18 10 73 18 48 40 37
42
ajp.amjpathol.org
-
The American Journal of Pathology
Mosaicism of ALT and Telomerase contain ALT inhibitors other than telomerase itself, because only some hybrid cells suppress ALT markers and maintain telomeres exclusively with telomerase.16 Other experiments with hybrid cells demonstrate suppression of telomerase activity and maintenance of telomeres by ALT, suggesting that some ALT cells contain telomerase inhibitors.18 Similar observations have not been made in vivo. Most tumors exhibit characteristics of one TMM. However, most published studies that have classified human tumors have also identified a subset of tumors not definitively ALT or telomerase positive.3e6,9e12,19e23 These tumors display ambiguous characteristics regarding telomere length, telomerase activity, or the presence of APBs. Some tumors have long, heterogeneous telomere lengths, suggestive of ALT, but exhibit telomerase activity; this ambiguity confounds accurate TMM characterization. Furthermore, although previous studies have identified tumors exhibiting both telomerase activity and the presence of APBs,10,11 suggesting that both telomerase and ALT cells are present in the sample, it has not been demonstrated whether these results indicate separate cell populations, infiltrating normal cells/normal surrounding tissue, or the presence of multiple TMMs in one cell. Therefore, we asked whether human sarcomas could demonstrate mosaicism for TMM by evaluating tumors at a cellular level.
Materials and Methods Tumor Samples Human tumor TMAs (T262, T263, BO241, BO481, and CO483) were obtained from US Biomax (Rockville, MD). De-identified frozen human osteosarcoma samples were obtained from the Human Tissue Resource Network at The Ohio State University, Columbus, under Institutional Review Board protocol 2009 E0409.
TRAP Assay Frozen tumor samples were mechanically homogenized, and telomerase activity was measured from extracts using the TRAPeze kit (Millipore, Billerica, MA), per manufacturer’s directions. Extracts were pre-incubated with template and phenol-chloroform extracted to remove endogenous PCR inhibitors. Products were imaged on a Typhoon Imaging system (GE Healthcare Biosciences, Pittsburgh, PA). Each sample was independently analyzed at least five times. Positive control immortalized cell lines HeLa and WI38
VA-13/2RA were obtained from ATCC (Manassas, VA), and lysates were prepared per manufacturer’s directions.
Telomere Restriction Fragment Southern Blot Analysis Frozen tumor samples were mechanically homogenized and DNA was extracted using the DNeasy kit (Qiagen, Valencia, CA), per manufacturer’s directions. DNA (2 mg) was digested with Hinf I and RsaI restriction enzymes (New England Biolabs, Ipswich, MA), separated by gel electrophoresis, and transferred to a charged nylon membrane (Hybond, Pittsburgh, PA). Cross-linked membrane was hybridized in UltraHyb-oligo (Ambion, Grand Island, NY) with a 32P-labeled (CCCTAA)3 probe, and products were imaged on a Typhoon Imaging system (GE Healthcare Biosciences, Pittsburgh, PA). Each sample was independently analyzed at least three times. Positive control immortalized cell lines HeLa and WI38 VA-13/2RA were obtained from ATCC, and DNA was prepared as previously described.
IHC Analysis Tumor samples were formalin-fixed, embedded in paraffin blocks, and cut into sections (4 mm thick) at The Ohio State University Pathology Core Facility (Columbus, OH). Automated antibody staining was performed on a Ventana Benchmark XT (Roche, Tuscon, AZ) with protease pretreatment. Manual immunohistochemistry (IHC) and combined fluorescent in situ hybridization were performed according to standard procedures after antigen retrieval by steaming in 10 mmol/L sodium citrate buffer (pH 6). Telomeres were hybridized with a Cy3-(TTAGGG) peptide nucleic acid probe (Panagene, Daejeon, South Korea). For IHC, the antibodies used were as follows: anti-telomere repeat binding factor 2 (TRF2) (IMG124A; Imgenex, San Diego, CA), anti-PML (ab53773; Abcam, Cambridge, MA), anti-PML (PG-M3; Santa Cruz Biotechnology, Santa Cruz, CA), and anti-telomerase (Ab-2; Calbiochem, Billerica, MA). Image capture and analysis were performed on a Nuance multispectral imaging system (Cambridge Research and Instrumentation, Hopkinton, MA), as previously described,24 and staining was confirmed on an Olympus (Center Valley, PA) FV 1000 spectral confocal microscope. For analysis, at least 200 cells (when possible) from at least three different geographical positions within each tumor were analyzed. Positive control immortalized cell lines HeLa and WI38 VA-13/2RA were obtained from ATCC and stained with standard immunocytochemical techniques, as previously described.
Patient age, sex, tumor diagnosis, site, stage, and pathological TNM grading were provided with each tissue sample. Four separate human sarcoma TMAs were IHC stained for evidence of TMM by ALT and telomerase. The percentage of ALT-positive cells within each tumor was analyzed by colocalization of telomere/TRF2 and PML, and the percentage of telomerase-positive cells was analyzed by expression of TERT. The percentage of cells was calculated from total cells in a field, in which some cells expressed neither colocalization of telomere/TRF2 and PML nor telomerase staining. F, female; M, male; % ALT, percentage of cells positive for ALT; % Tel, percentage of cells positive for telomerase; ND, no data; TNM, tumor, node, and metastasis.
The American Journal of Pathology
-
ajp.amjpathol.org
43
Gocha et al
Results The observation of tumors with characteristics of both TMMs suggested questions of whether tumors could contain some cells expressing telomerase and other cells expressing characteristics of ALT, or whether the same cell could express both types of telomere-elongating characteristics. We examined TMMs within human sarcomas, because of their high prevalence of ALT, using a computationally enhanced imaging system capable of analyzing three IHC stains per tumor cell. We analyzed four TMAs that included 57 high-grade sarcomas: 26 chondrosarcomas, 4 osteoclastomas, and 27 osteosarcomas (Table 1). The microarrays were analyzed for ALT characteristics using colocalization of PML with the telomere/telomeric proteins to identify APBs. APBs were identified in situ by two techniques: TRF2 and PML chromogenic IHC using an automated stainer and manual PML fluorescent IHC combined with telomere fluorescent in situ hybridization. Microarrays were also stained with antibodies specific to the reverse transcriptase subunit of telomerase (TERT). All staining was validated in human immortalized cell lines with known TMMs: telomerase-positive HeLa cells and ALT-positive WI38 VA-13/2RA cells (Supplemental Figure S1). Spatial localization of all three signals was analyzed using a Nuance multispectral imaging system, which allows management of a threshold to exclude background staining. Cells with TRF2/telomere and PML colocalization excluded telomerase; however, in some tumors, separate populations of cells within the same tumor stained positive for telomerase and negative for TRF2/telomere and PML colocalization. Of 57 tumors, 4 contained only ALT-positive cells, 23 contained only telomerase-positive cells, and 26 contained mixed populations of ALT- and telomerase-positive cells; 4 tumors exhibited neither APBs nor telomerase staining. Those tumors with both TMMs displayed a wide variability in the proportion of each cell type, ranging from 1% to 47% ALT positive and from 2% to 95% telomerase positive (Table 1). Each tumor varied, in that not all cells within a tissue section were classified as ALT or telomerase positive. This might be due to incomplete staining because of the osseous nature of these tumors, methodological detection limits, or varying protein expression levels. Within asynchronous cultures of ALT cells, APBs were not observed in all cells because their presence is cell cycle regulated.25 This could explain the incomplete characterization of all tumor cells in our samples and would underestimate the proportion of ALT-positive cells. Neither ALT- nor telomerase-positive cells were detected in normal bone and cartilage tissues, indicating the staining specificity (data not shown). Because ALT is not detected in colorectal cancers,1,2 a human colon cancer TMA was stained as a negative control. This array included 40 adenocarcinomas, all of which displayed telomerase expression and no colocalization of TRF2 and PML signals (data not shown).
44
Normal thymus tissue also stained positive for telomerase and negative for ALT colocalization (data not shown). Our findings using in situ colocalization and IHC were validated with solution-based biochemical analysis and evaluation of an independent set of five frozen human osteosarcomas. Telomere characteristics of each tumor were assessed using whole tumor lysates and single-cell assays. Whole tumor lysate assays characterized telomere length and telomerase activity as a general representation of many cells within the tumor. Telomere length was measured with TRF Southern blot analysis, and telomerase activity was measured with the telomere repeat amplification protocol (TRAP) assay. The relative average telomere repeat
Figure 1
Characterization of TMM in human osteosarcoma lysates using whole tumor biochemical assays. A: The average telomere length of DNA from five frozen human osteosarcomas is measured by TRF Southern blot analysis. The signal in each lane represents telomeric DNA fragments liberated from digested genomic DNA. HeLa and WI38 VA-13/2RA are positive controls for telomerase and ALT cells, respectively. Telomerase-positive cells have shorter homogeneous telomere lengths, whereas heterogeneous lengths characterize ALT-positive cells. DNA- and l DNA represent negative controls. Telomere length is independently confirmed using quantitative real-time PCR (data not shown). B: Telomerase activity in the same five human osteosarcoma lysates is measured using the TRAP assay. A 36-bp internal control band is present in each lane, with laddered bands beginning from 50 bp upwards signifying positive telomerase activity. The presence of only the internal control band suggests telomerase-negative ALT mechanisms. Heat denaturation serves as a negative control for each extract. HeLa and WI38 VA-13/2RA cell lysates are positive controls for telomerase-positive and ALTpositive cells, respectively; other positive and negative controls are internal controls provided by the manufacturer.
ajp.amjpathol.org
-
The American Journal of Pathology
Mosaicism of ALT and Telomerase
Figure 2
In situ analyses of human osteosarcoma sections. A: High-power magnification confocal micrographs demonstrate PML (red; top panels), TRF2 signals (green; top panels), and TERT signals (white; bottom panels) in DAPI-stained (blue) nuclei of representative tumor sections. ALT-positive tumors exhibit colocalization (yellow) of PML and TRF2 signals in discrete nuclear foci (top left panel) and a lack of TERT staining (bottom left panel). Telomerase-positive tumors exhibit both PML and TRF2 nuclear signals without colocalization (top right panel) and TERT staining (bottom right panel). B: Slides are prepared from five frozen human osteosarcomas after formalin fixation and paraffin embedding and analyzed for TRF2/PML colocalization and for TERT expression using in situ IHC. H&E-stained reference sections are shown (top panels), along with an analysis of PML/TRF2 colocalization (pseudocolored yellow; middle panels) and telomerase expression (pseudocolored white; bottom panels). Cells are counterstained with hematoxylin for visualization. Original magnification, 100 (A); 40 (B). For PML/TRF2 colocalization and telomerase expression, images are pseudocolored and analyzed using Nuance software version 2.4 (Hopkinton, MA).
length of each tumor was independently confirmed using a quantitative real-time PCR26 (data not shown). Long heterogeneous telomeres, identified by TRF Southern blot analysis, and a lack of telomerase activity by TRAP suggest
The American Journal of Pathology
-
ajp.amjpathol.org
ALT mechanisms; conversely, telomerase-positive staining and shorter, more homogeneous telomere lengths suggest telomerase-associated mechanisms. The telomerase-positive human immortalized cell line HeLa and the ALT-positive
45
Gocha et al human immortalized cell line WI38 VA-13/2RA were used as positive controls for qualitative characteristics of each TMM. Each of the five osteosarcomas contained telomeres that closely corresponded with characteristics of either HeLa or WI38 VA-13/2RA cells. One of five human osteosarcomas (tumor A) displayed short and homogeneous telomere lengths via TRF Southern blot analysis (Figure 1A) and telomerase activity via TRAP assay (Figure 1B). Tumors C and D exhibited only ALT phenotypes (Figure 1, A and B). Tumors B and E provided ambiguous results with weak and variable telomerase activity (Figure 1B) and telomere lengths (Figure 1A) that clearly suggested telomerase- (tumor B) and ALT- (tumor E) maintained telomeres. Single cells within each frozen tumor sample were then examined for ALT-associated protein colocalization and telomerase by in situ IHC. After fixing tumor sections in 10% buffered formalin, in situebased co-expression analyses were performed blinded to the solution-phase biochemistry data. Tumor sections were stained with antibodies to PML, TRF2, and telomerase, and analyzed using the Nuance multispectral imaging system. Colocalization of TRF2 and PML, denoting an ALT cell, appeared as distinct nuclear foci consistent with the appearance of APBs. Colocalization was confirmed using confocal microscopy to ensure that signals were overlapping in the same focal plane. Examples of confocal micrographs of individual representative tumor cells are shown in Figure 2A. Cells scored as ALT-positive demonstrated colocalization of PML and TRF2 and minimal or no TERT staining; cells scored as telomerase-positive demonstrated no colocalization of PML and TRF2 but were positive for TERT staining. Of cells from tumor A, 48% were telomerase positive, whereas none exhibited TRF2/PML colocalization (Figure 2B); these results corroborated those from TRF Southern blot analysis and the TRAP assay. Conversely, tumors C and D contained 31% and 65% of cells, respectively, with colocalization of TRF2/ PML, corroborating whole tumor lysate assays. Tumor D completely lacked cells with telomerase expression, whereas tumor C contained some telomerase-positive cells (Figure 2B). However, tumor C was geographically heterogeneous, containing distinct areas of weak or robust staining; this dimorphism could explain the negative TRAP results for this tumor, even though some telomerase-positive cells were present. Finally, tumors B and E, which had mixed results in whole tumor assays, displayed two separate cell populations: one staining positive for ALT protein colocalization (32% and 34%, respectively) and one staining positive for telomerase (2% and 6%, respectively) (Figure 2B). The TMM results from all three assays are summarized in Table 2. Independent and blinded histopathological analyses were performed to confirm that ALT- and telomerase-positive cells likely represented two distinct populations of tumor cells, rather than infiltrating normal cells.
46
Discussion Our studies with an enhanced imaging system explicitly demonstrate that some human sarcomas are mosaic for two types of cells: those that use telomerase and those that use ALT. The presence of both cell types highlights a novel type of tumor heterogeneity and suggests that telomere maintenance may not be a static characteristic of all tumors. In addition, the presence of more than one TMM confounds accurate tumor assessment using a single method and indicates a need for precise tumor characterization. Our results also highlight the question of whether individual cells can switch TMM or whether a cell commits to telomerasedependent or telomerase-independent telomere maintenance. Furthermore, we speculate that tumor types, in addition to sarcomas, may also contain mosaic populations of cells with regard to TMM. Previous studies have reported a lower percentage of tumors that display characteristics of both TMMs than determined in our studies, which is 50%. Although those studies have used an either-or approach for the determination of TMM, we have combined both IHC and biochemical analyses to gain a more comprehensive view of the TMM status of some of these tumors. Some tumors contain a small percentage of ALT-positive or telomerase-positive cells and would likely be classified into one category or another with assays that analyze a tumor extract, such as the TRF Southern blot analysis or the TRAP assay. In these solutionphase assays, tumor cells are combined into a quasihomogeneous solution, in which a small percentage of one particular cell type would not be measured. Combining the results of these assays with a cell-by-cell analysis by IHC has identified multiple TMM cell types within one tumor. The IHC staining of TERT protein does not confirm telomerase activity, although it is highly correlated with telomerase activity27,28 and implies that those cells are not using ALT. TERT staining correlates well with TRAP activity in the subset of four homogeneously stained osteosarcomas for which IHC and biochemical assays were possible. Table 2 Analysis of Telomere Maintenance in Frozen Human Osteosarcomas Age TRF Southern TRAP Tumor Sex (years) blot analysis assay In situ IHC (%) A B C D E
M M F M F
53 21 81 69 22
Telomerase Telomerase ALT ALT ALT
þ þ/ þ/
Telomerase Telomerase Telomerase ALT (65) Telomerase
(48) (2)/ALT (32) (7)/ALT (31) (6)/ALT (34)
Summary of TMM characteristics in five frozen human osteosarcomas. Telomere length was analyzed by TRF Southern blot analysis, telomerase activity was analyzed by TRAP assay, and single-cell telomerase and ALT expression was analyzed by in situ IHC. The percentage of cells was calculated from total cells in a field, in which some cells expressed neither colocalization of TRF2/PML nor telomerase. F, female; M, male; þ, positive result; , negative result.
ajp.amjpathol.org
-
The American Journal of Pathology
Mosaicism of ALT and Telomerase TMM can be a prognosticator for patient survival: ALTpositive glioblastoma multiforme tumors are associated with a better prognosis,6,22 whereas ALT-positive liposarcomas are associated with a worse prognosis11 than telomerasepositive tumors of each type. Recent clinical trials of telomerase-targeted therapies also underline the clinical importance of telomere maintenance. Mosaicism of TMM within human tumors may affect the choice of therapeutic approaches, because tumors with cells that maintain telomeres via more than one mechanism may require combination therapies to target each type of cell. In addition, recent work in mouse models demonstrates that inhibition of telomerase activity after tumor formation promotes the survival of tumors via the ALT pathway.29 A similar scenario is likely to occur in mosaic human tumors after telomerase inhibitor treatment, stressing the value of precise tumor TMM characterization before treatment and/or treatment with combination therapies that target both telomerase and ALT.
Acknowledgments We thank past and current members of the Groden laboratory and Dr. Samir Acharya for helpful discussions and technical advice, and The Ohio State University Core Facilities: Nucleic Acids Shared Resource, Campus Microscopy and Imaging Facility, and Pathology Core Facility.
Supplemental Data Supplemental material for this article can be found at http:// dx.doi.org/10.1016/j.ajpath.2012.10.001.
References 1. Henson JD, Reddel RR: Assaying and investigating alternative lengthening of telomeres activity in human cells and cancers. FEBS Lett 2008, 584:3800e3811 2. Heaphy CM, Subhawang AP, Hong SM, Goggins MG, Montgomery EA, Gabrielson E, Netto GJ, Epstein JI, Lotan TL, Westra WH, Shih IeM, Iacobuzio-Donahue CA, Maitra A, Li QK, Eberhart CG, Taube JM, Rakheja D, Kurman RJ, Wu TC, Roden RB, Argani P, De Marzo AM, Terracciano L, Torbenson M, Meeker AK: Prevalence of alternative lengthening of telomeres telomere maintenance mechanism in human cancer subtypes. Am J Pathol 2011, 179: 1608e1615 3. Bryan TM, Englezou A, Dalla-Pozza L, Dunham MA, Reddel RR: Evidence for an alternative mechanism for maintaining telomere length in human tumors and tumor-derived cell lines. Nat Med 1997, 3: 1271e1274 4. Ulaner GA, Huang HY, Otero J, Zhao Z, Ben-Porat L, Satagopan JM, Gorlick R, Meyers P, Healey JH, Huvos AG, Hoffman AR, Ladanyi M: Absence of a telomere maintenance mechanism as a favorable prognostic factor in patients with osteosarcoma. Cancer Res 2003, 63:1759e1763 5. Sanders RP, Drissi R, Billups CA, Daw NC, Valentine MB, Dome JS: Telomerase expression predicts unfavorable outcome in osteosarcoma. J Clin Oncol 2004, 22:3790e3797
The American Journal of Pathology
-
ajp.amjpathol.org
6. Henson JD, Hannay JA, McCarthy SW, Royds JA, Yeager TR, Robinson RA, Wharton SB, Jellinek DA, Arbuckle SM, Yoo J, Robinson BG, Learoyd DL, Stalley PD, Bonar SF, Yu D, Pollock RE, Reddel RR: A robust assay for alternative lengthening of telomeres in tumors shows the significance of alternative lengthening of telomeres in sarcomas and astrocytomas. Clin Cancer Res 2005, 11:217e225 7. Guilleret I, Yan P, Guillou L, Braunschweig R, Coindre JM, Benhattar J: The human telomerase RNA gene (hTERC) is regulated during carcinogenesis but is not dependent on DNA methylation. Carcinogenesis 2002, 23:2025e2030 8. Montgomery E, Argani P, Hicks JL, DeMarzo AM, Meeker AK: Telomere lengths of translocation-associated and nontranslocationassociated sarcomas differ dramatically. Am J Pathol 2004, 164: 1523e1529 9. Yan P, Benhattar J, Coindre J-M, Guillou L: Telomerase activity and hTERT mRNA expression can be heterogeneous and does not correlate with telomere length in soft tissue sarcomas. Int J Cancer 2002, 98: 851e856 10. Johnson JE, Varkonyi RJ, Schwalm J, Cragle R, Klein-Szanto A, Patchefsky A, Cukierman E, von Mehren M, Broccoli D: Multiple mechanisms of telomere maintenance exist in liposarcomas. Clin Cancer Res 2005, 11:5347e5355 11. Costa A, Daidone MG, Daprai L, Villa R, Cantu S, Pilotti S, Mariani L, Gronchi A, Henson JD, Reddel RR, Zaffaroni N: Telomere maintenance mechanisms in liposarcomas: association with histologic subtypes and disease progression. Cancer Res 2006, 66: 8918e8924 12. Matsuo T, Shay JW, Wright WE, Hiyama E, Shimose S, Kubo T, Sugita T, Yasunaga Y, Ochi M: Telomere-maintenance mechanisms in soft-tissue malignant fibrous histiocytomas. J Bone Joint Surg Am 2009, 91:928e937 13. Zimmermann S, Voss M, Kaiser S, Kapp U, Waller CF, Martens UM: Lack of telomerase activity in human mesenchymal stem cells. Leukemia 2003, 17:1146e1149 14. Cerone MA, Londono-Vallejo JA, Bacchetti S: Telomere maintenance by telomerase and by recombination can coexist in human cells. Hum Mol Genet 2001, 10:1945e1952 15. Grobelny JV, Kulp-McEliece M, Broccoli D: Effects of reconstitution of telomerase activity on telomere maintenance by the alternative lengthening of telomeres (ALT) pathway. Hum Mol Genet 2001, 10: 1953e1961 16. Perrem K, Colgin LM, Neumann AA, Yeager TR, Reddel RR: Coexistence of alternative lengthening of telomeres and telomerase in hTERT-transfected GM847 cells. Mol Cell Biol 2001, 21: 3862e3875 17. Ford LP, Zou Y, Pongracz K, Gryaznov SM, Shay JW, Wright WE: Telomerase can inhibit the recombination-based pathway of telomere maintenance in human cells. J Biol Chem 2001, 276: 32198e32203 18. Katoh M, Katoh M, Kameyama M, Kugoh H, Shimizu M, Oshimura M: A repressor function for telomerase activity in telomerase-negative immortal cells. Mol Carcin 1998, 21:17e25 19. Villa R, Daidone MG, Motta R, Venturini L, De Marco C, Vannelli A, Kusamura S, Baratti D, Deraco M, Costa A, Reddel RR, Zaffaroni N: Multiple mechanisms of telomere maintenance exist and differentially affect clinical outcome in diffuse malignant peritoneal mesothelioma. Clin Cancer Res 2008, 14:4134e4140 20. Else T, Giordano TJ, Hammer GD: Evaluation of telomere length maintenance mechanisms in adrenocortical carcinoma. J Clin Endocrinol Metab 2008, 93:1442e1449 21. Gupta J, Han L-P, Wang P, Gallie BL, Bacchetti S: Development of retinoblastoma in the absence of telomerase activity. J Natl Cancer Inst 1996, 88:1152e1157 22. Hakin-Smith V, Jellinek DA, Levy D, Carroll T, Teo M, Timperley WR, McKay MJ, Reddel RR, Royds JA: Alternative lengthening of telomeres and survival in patients with glioblastoma multiforme. Lancet 2003, 361:836e838
47
Gocha et al 23. Omori Y, Nakayama F, Li D, Kanemitsu K, Semba S, Ito A, Yokozaki H: Alternative lengthening of telomeres frequently occurs in mismatch repair system-deficient gastric carcinoma. Cancer Sci 2008, 100:413e418 24. Nuovo GJ: In situ detection of microRNAs in paraffin embedded, formalin fixed tissues and the co-localization of their putative targets. Methods 2010, 52:307e315 25. Grobelny JV, Godwin AK, Broccoli D: ALT-associated PML bodies are present in viable cells and are enriched in cells in the G(2)/M phase of the cell cycle. J Cell Sci 2000, 113:4577e4585 26. Cawthon RM: Telomere length measurement by a novel monochrome multiplex quantitative PCR method. Nucleic Acids Res 2009, 37:e21
48
27. Kanaya T, Kyo S, Takakura M, Ito H, Namiki M, Inoue M: hTERT is a critical determinant of telomerase activity in renal-cell carcinoma. Int J Cancer 1998, 78:539e543 28. Wu A, Ichihashi M, Ueda M: Correlation of the expression of human telomerase subunits with telomerase activity in normal skin and skin tumors. Cancer 1999, 86:2038e2044 29. Hu J, Hwang SS, Liesa M, Gan B, Sahin E, Jaskelioff M, Ding Z, Ying H, Boutin AT, Zhang H, Johnson S, Ivanova E, KostAlimova M, Protopopov A, Wang YA, Shirihai OS, Chin L, DePinho RA: Antitelomerase therapy provokes ALT and mitochondrial adaptive mechanisms in cancer. Cell 2012, 148:651e663
ajp.amjpathol.org
-
The American Journal of Pathology
ajp.amjpathol.org
SHORT COMMUNICATION Human Sarcomas Are Mosaic for Telomerase-Dependent and Telomerase-Independent Telomere Maintenance Mechanisms Implications for Telomere-Based Therapies April R.S. Gocha,* Gerard Nuovo,y Obiajulu H. Iwenofu,y and Joanna Groden* From the Departments of Molecular Virology, Immunology, and Medical Genetics* and Pathology,y Ohio State University College of Medicine, Columbus, Ohio Accepted for publication October 2, 2012. Address correspondence to Joanna Groden, Ph.D., 986 Biomedical Research Tower, 460 W. 12th Ave, Columbus, OH 43210. E-mail: joanna. [email protected].
Telomere shortening necessitates that tumor cells activate a telomere maintenance mechanism (TMM) to support immortalization. Although most tumor cells activate expression of the enzyme telomerase, some cells elongate telomeres in a telomerase-independent manner, termed alternative lengthening of telomeres (ALT). Previous studies have evaluated the presence of telomerase or ALT mechanisms or both in a variety of tumor types. Our studies also show that TMMs are not mutually exclusive in some tumors. In contrast, our IHC analyses of human sarcomas identified a subset of tumors with some cells containing ALT-associated PML bodies, a hallmark of ALT, and separate cells expressing telomerase in the same tumor. By using a second set of human osteosarcomas, we merged IHC and biochemical analyses to characterize more fully the tumor TMM. The IHC data reveal the presence of both telomerase- and ALT-positive tumor cells in samples that demonstrate characteristics of both telomerase and ALT in biochemical assays. These assays, which measure telomere length and telomerase activity of tumor extracts, are conventionally used to classify tumor TMM. Our results suggest that TMM is not a single or perhaps static characteristic of some tumors and that TMM heterogeneity should be considered in tumor stratification. Furthermore, clinical interest in telomere-based therapies may necessitate accurate characterization of tumor TMM before treatment to maximize therapeutic efficacy. (Am J Pathol 2013, 182: 41e48; http://dx.doi.org/10.1016/j.ajpath.2012.10.001)
Activation of a telomere maintenance mechanism (TMM) is critical for tumor cell immortalization and cancer progression because telomeres shorten with each round of cellular division. Cells can use two known TMMs: catalytic activity of the enzyme telomerase or a telomerase-independent, recombination-based pathway termed alternative lengthening of telomeres (ALT). In contrast to cells with active telomerase, cells that use ALT are characterized by long, heterogeneous telomere lengths, ALT-associated PML bodies (APBs), and extrachromosomal telomeric repeats. Several studies have evaluated TMMs in human tumors1,2 and demonstrated that ALT is overrepresented in mesenchymal tumors,2e12 perhaps reflective of a lack of detectable telomerase expression in mesenchymal stem cells.13 More specifically, almost 60% of osteosarcomas exhibit ALT characteristics,3e6 the highest incidence of any tumor yet evaluated. In contrast, epithelial Copyright ª 2013 American Society for Investigative Pathology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajpath.2012.10.001
tumors more often express telomerase,1,2 suggesting there may be cell-specific preferences for TMM. Studies have demonstrated that either type of TMM can occur within the same cell, but only with experimental manipulation. Exogenous expression of functional telomerase within some ALT cell lines results in telomerase-mediated elongation of short telomeres, although cells still retain ALT characteristics14e16; studies in other ALT cell lines have shown that telomerase expression inhibits ALT characteristics.17 Fusion experiments of telomerase-expressing and ALT cells have shown that the mechanisms governing telomere maintenance are complex. Some telomerase-expressing cells Supported by The Ohio State University Pelotonia Fellowship Program (A.R.S.G.), the Bloom’s Syndrome Foundation (J.G.), and NIH grant CA-117898 (J.G.).
Gocha et al Table 1
Analysis of TMM in Human Sarcoma TMAs
Patient no.
Age (years)
Sex
Tumor
Site
Stage
TNM grading
ALT (%)
Tel (%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
30 39 42 49 28 33 52 42 35 38 46 35 16 42 40 33 38 58 28 40 34 62 65 29 48 24 52 7 34 32 41 18 43 19 40 13 17 64 23 51 13 37 18 15 31 32 19 29 24 32 32 30 14 37 14 17 30
M F F F F M M F F M F M F M M M M F M M M M M M F F F F F M M M M M F M M M M M F M M M M F M M M F F F F M F M M
Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Chondrosarcoma Osteoclastoma Osteoclastoma Osteoclastoma Osteoclastoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma Osteosarcoma
Leg Scapula ND ND ND ND ND ND ND ND ND Humerus Femur ND ND ND Face ND ND ND Leg Maxilla ND ND ND ND ND Femur ND ND Femur Humerus Costal Femur ND ND ND ND ND Femur Femur ND ND Femur ND ND Leg ND Skull ND ND ND Tibia Femur ND Femur ND
ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND IIb ND ND ND ND ND II ND II II ND IIb IIb ND ND ND ND ND ND IIIb ND ND ND ND ND ND IIb ND IIb ND ND ND IIIb IIb ND IIb ND
ND ND T2N0M0 T2N0M0 T2N0M0 T1N0M0 T1N0M0 T2N0M0 T2N0M0 T2N0M0 T2N0M0 ND ND T2N0M0 ND T2N0M0 ND T2N0M0 T1N0M0 T2N0M0 T2N0M0 ND T2N0M0 T1N0M0 T2N0M0 T1N0M0 T1N0M0 ND T1N0M0 T1N0M0 ND T2N0M0 T2N0M0 ND T2N0M0 T2N0M0 T2N0M0 T2N0M0 T2N0M0 T2N0M0 ND T1N0M0 T2N0M0 ND T2N0M0 T2N0M0 T2N0M0 T1N0M0 T2N0M0 T2N0M0 T2N0M0 T2N0M0 T2N0M0 T1N0M0 T2N0M0 T2N0M0 T2N0M0
0 0 0 0 0 0 0 0 0 0 41 29 5 29 25 12 9 6 5 2 1 1 0 0 0 0 0 11 5 2 0 0 0 0 0 0 0 0 0 0 0 0 20 47 20 17 15 15 9 5 4 4 3 3 2 1 1
95 95 75 75 40 40 40 33 32 30 0 0 0 79 53 33 66 34 85 82 3 2 0 0 0 0 69 32 95 87 95 90 90 90 74 62 50 38 30 16 14 13 0 24 16 38 62 11 2 2 18 10 73 18 48 40 37
42
ajp.amjpathol.org
-
The American Journal of Pathology
Mosaicism of ALT and Telomerase contain ALT inhibitors other than telomerase itself, because only some hybrid cells suppress ALT markers and maintain telomeres exclusively with telomerase.16 Other experiments with hybrid cells demonstrate suppression of telomerase activity and maintenance of telomeres by ALT, suggesting that some ALT cells contain telomerase inhibitors.18 Similar observations have not been made in vivo. Most tumors exhibit characteristics of one TMM. However, most published studies that have classified human tumors have also identified a subset of tumors not definitively ALT or telomerase positive.3e6,9e12,19e23 These tumors display ambiguous characteristics regarding telomere length, telomerase activity, or the presence of APBs. Some tumors have long, heterogeneous telomere lengths, suggestive of ALT, but exhibit telomerase activity; this ambiguity confounds accurate TMM characterization. Furthermore, although previous studies have identified tumors exhibiting both telomerase activity and the presence of APBs,10,11 suggesting that both telomerase and ALT cells are present in the sample, it has not been demonstrated whether these results indicate separate cell populations, infiltrating normal cells/normal surrounding tissue, or the presence of multiple TMMs in one cell. Therefore, we asked whether human sarcomas could demonstrate mosaicism for TMM by evaluating tumors at a cellular level.
Materials and Methods Tumor Samples Human tumor TMAs (T262, T263, BO241, BO481, and CO483) were obtained from US Biomax (Rockville, MD). De-identified frozen human osteosarcoma samples were obtained from the Human Tissue Resource Network at The Ohio State University, Columbus, under Institutional Review Board protocol 2009 E0409.
TRAP Assay Frozen tumor samples were mechanically homogenized, and telomerase activity was measured from extracts using the TRAPeze kit (Millipore, Billerica, MA), per manufacturer’s directions. Extracts were pre-incubated with template and phenol-chloroform extracted to remove endogenous PCR inhibitors. Products were imaged on a Typhoon Imaging system (GE Healthcare Biosciences, Pittsburgh, PA). Each sample was independently analyzed at least five times. Positive control immortalized cell lines HeLa and WI38
VA-13/2RA were obtained from ATCC (Manassas, VA), and lysates were prepared per manufacturer’s directions.
Telomere Restriction Fragment Southern Blot Analysis Frozen tumor samples were mechanically homogenized and DNA was extracted using the DNeasy kit (Qiagen, Valencia, CA), per manufacturer’s directions. DNA (2 mg) was digested with Hinf I and RsaI restriction enzymes (New England Biolabs, Ipswich, MA), separated by gel electrophoresis, and transferred to a charged nylon membrane (Hybond, Pittsburgh, PA). Cross-linked membrane was hybridized in UltraHyb-oligo (Ambion, Grand Island, NY) with a 32P-labeled (CCCTAA)3 probe, and products were imaged on a Typhoon Imaging system (GE Healthcare Biosciences, Pittsburgh, PA). Each sample was independently analyzed at least three times. Positive control immortalized cell lines HeLa and WI38 VA-13/2RA were obtained from ATCC, and DNA was prepared as previously described.
IHC Analysis Tumor samples were formalin-fixed, embedded in paraffin blocks, and cut into sections (4 mm thick) at The Ohio State University Pathology Core Facility (Columbus, OH). Automated antibody staining was performed on a Ventana Benchmark XT (Roche, Tuscon, AZ) with protease pretreatment. Manual immunohistochemistry (IHC) and combined fluorescent in situ hybridization were performed according to standard procedures after antigen retrieval by steaming in 10 mmol/L sodium citrate buffer (pH 6). Telomeres were hybridized with a Cy3-(TTAGGG) peptide nucleic acid probe (Panagene, Daejeon, South Korea). For IHC, the antibodies used were as follows: anti-telomere repeat binding factor 2 (TRF2) (IMG124A; Imgenex, San Diego, CA), anti-PML (ab53773; Abcam, Cambridge, MA), anti-PML (PG-M3; Santa Cruz Biotechnology, Santa Cruz, CA), and anti-telomerase (Ab-2; Calbiochem, Billerica, MA). Image capture and analysis were performed on a Nuance multispectral imaging system (Cambridge Research and Instrumentation, Hopkinton, MA), as previously described,24 and staining was confirmed on an Olympus (Center Valley, PA) FV 1000 spectral confocal microscope. For analysis, at least 200 cells (when possible) from at least three different geographical positions within each tumor were analyzed. Positive control immortalized cell lines HeLa and WI38 VA-13/2RA were obtained from ATCC and stained with standard immunocytochemical techniques, as previously described.
Patient age, sex, tumor diagnosis, site, stage, and pathological TNM grading were provided with each tissue sample. Four separate human sarcoma TMAs were IHC stained for evidence of TMM by ALT and telomerase. The percentage of ALT-positive cells within each tumor was analyzed by colocalization of telomere/TRF2 and PML, and the percentage of telomerase-positive cells was analyzed by expression of TERT. The percentage of cells was calculated from total cells in a field, in which some cells expressed neither colocalization of telomere/TRF2 and PML nor telomerase staining. F, female; M, male; % ALT, percentage of cells positive for ALT; % Tel, percentage of cells positive for telomerase; ND, no data; TNM, tumor, node, and metastasis.
The American Journal of Pathology
-
ajp.amjpathol.org
43
Gocha et al
Results The observation of tumors with characteristics of both TMMs suggested questions of whether tumors could contain some cells expressing telomerase and other cells expressing characteristics of ALT, or whether the same cell could express both types of telomere-elongating characteristics. We examined TMMs within human sarcomas, because of their high prevalence of ALT, using a computationally enhanced imaging system capable of analyzing three IHC stains per tumor cell. We analyzed four TMAs that included 57 high-grade sarcomas: 26 chondrosarcomas, 4 osteoclastomas, and 27 osteosarcomas (Table 1). The microarrays were analyzed for ALT characteristics using colocalization of PML with the telomere/telomeric proteins to identify APBs. APBs were identified in situ by two techniques: TRF2 and PML chromogenic IHC using an automated stainer and manual PML fluorescent IHC combined with telomere fluorescent in situ hybridization. Microarrays were also stained with antibodies specific to the reverse transcriptase subunit of telomerase (TERT). All staining was validated in human immortalized cell lines with known TMMs: telomerase-positive HeLa cells and ALT-positive WI38 VA-13/2RA cells (Supplemental Figure S1). Spatial localization of all three signals was analyzed using a Nuance multispectral imaging system, which allows management of a threshold to exclude background staining. Cells with TRF2/telomere and PML colocalization excluded telomerase; however, in some tumors, separate populations of cells within the same tumor stained positive for telomerase and negative for TRF2/telomere and PML colocalization. Of 57 tumors, 4 contained only ALT-positive cells, 23 contained only telomerase-positive cells, and 26 contained mixed populations of ALT- and telomerase-positive cells; 4 tumors exhibited neither APBs nor telomerase staining. Those tumors with both TMMs displayed a wide variability in the proportion of each cell type, ranging from 1% to 47% ALT positive and from 2% to 95% telomerase positive (Table 1). Each tumor varied, in that not all cells within a tissue section were classified as ALT or telomerase positive. This might be due to incomplete staining because of the osseous nature of these tumors, methodological detection limits, or varying protein expression levels. Within asynchronous cultures of ALT cells, APBs were not observed in all cells because their presence is cell cycle regulated.25 This could explain the incomplete characterization of all tumor cells in our samples and would underestimate the proportion of ALT-positive cells. Neither ALT- nor telomerase-positive cells were detected in normal bone and cartilage tissues, indicating the staining specificity (data not shown). Because ALT is not detected in colorectal cancers,1,2 a human colon cancer TMA was stained as a negative control. This array included 40 adenocarcinomas, all of which displayed telomerase expression and no colocalization of TRF2 and PML signals (data not shown).
44
Normal thymus tissue also stained positive for telomerase and negative for ALT colocalization (data not shown). Our findings using in situ colocalization and IHC were validated with solution-based biochemical analysis and evaluation of an independent set of five frozen human osteosarcomas. Telomere characteristics of each tumor were assessed using whole tumor lysates and single-cell assays. Whole tumor lysate assays characterized telomere length and telomerase activity as a general representation of many cells within the tumor. Telomere length was measured with TRF Southern blot analysis, and telomerase activity was measured with the telomere repeat amplification protocol (TRAP) assay. The relative average telomere repeat
Figure 1
Characterization of TMM in human osteosarcoma lysates using whole tumor biochemical assays. A: The average telomere length of DNA from five frozen human osteosarcomas is measured by TRF Southern blot analysis. The signal in each lane represents telomeric DNA fragments liberated from digested genomic DNA. HeLa and WI38 VA-13/2RA are positive controls for telomerase and ALT cells, respectively. Telomerase-positive cells have shorter homogeneous telomere lengths, whereas heterogeneous lengths characterize ALT-positive cells. DNA- and l DNA represent negative controls. Telomere length is independently confirmed using quantitative real-time PCR (data not shown). B: Telomerase activity in the same five human osteosarcoma lysates is measured using the TRAP assay. A 36-bp internal control band is present in each lane, with laddered bands beginning from 50 bp upwards signifying positive telomerase activity. The presence of only the internal control band suggests telomerase-negative ALT mechanisms. Heat denaturation serves as a negative control for each extract. HeLa and WI38 VA-13/2RA cell lysates are positive controls for telomerase-positive and ALTpositive cells, respectively; other positive and negative controls are internal controls provided by the manufacturer.
ajp.amjpathol.org
-
The American Journal of Pathology
Mosaicism of ALT and Telomerase
Figure 2
In situ analyses of human osteosarcoma sections. A: High-power magnification confocal micrographs demonstrate PML (red; top panels), TRF2 signals (green; top panels), and TERT signals (white; bottom panels) in DAPI-stained (blue) nuclei of representative tumor sections. ALT-positive tumors exhibit colocalization (yellow) of PML and TRF2 signals in discrete nuclear foci (top left panel) and a lack of TERT staining (bottom left panel). Telomerase-positive tumors exhibit both PML and TRF2 nuclear signals without colocalization (top right panel) and TERT staining (bottom right panel). B: Slides are prepared from five frozen human osteosarcomas after formalin fixation and paraffin embedding and analyzed for TRF2/PML colocalization and for TERT expression using in situ IHC. H&E-stained reference sections are shown (top panels), along with an analysis of PML/TRF2 colocalization (pseudocolored yellow; middle panels) and telomerase expression (pseudocolored white; bottom panels). Cells are counterstained with hematoxylin for visualization. Original magnification, 100 (A); 40 (B). For PML/TRF2 colocalization and telomerase expression, images are pseudocolored and analyzed using Nuance software version 2.4 (Hopkinton, MA).
length of each tumor was independently confirmed using a quantitative real-time PCR26 (data not shown). Long heterogeneous telomeres, identified by TRF Southern blot analysis, and a lack of telomerase activity by TRAP suggest
The American Journal of Pathology
-
ajp.amjpathol.org
ALT mechanisms; conversely, telomerase-positive staining and shorter, more homogeneous telomere lengths suggest telomerase-associated mechanisms. The telomerase-positive human immortalized cell line HeLa and the ALT-positive
45
Gocha et al human immortalized cell line WI38 VA-13/2RA were used as positive controls for qualitative characteristics of each TMM. Each of the five osteosarcomas contained telomeres that closely corresponded with characteristics of either HeLa or WI38 VA-13/2RA cells. One of five human osteosarcomas (tumor A) displayed short and homogeneous telomere lengths via TRF Southern blot analysis (Figure 1A) and telomerase activity via TRAP assay (Figure 1B). Tumors C and D exhibited only ALT phenotypes (Figure 1, A and B). Tumors B and E provided ambiguous results with weak and variable telomerase activity (Figure 1B) and telomere lengths (Figure 1A) that clearly suggested telomerase- (tumor B) and ALT- (tumor E) maintained telomeres. Single cells within each frozen tumor sample were then examined for ALT-associated protein colocalization and telomerase by in situ IHC. After fixing tumor sections in 10% buffered formalin, in situebased co-expression analyses were performed blinded to the solution-phase biochemistry data. Tumor sections were stained with antibodies to PML, TRF2, and telomerase, and analyzed using the Nuance multispectral imaging system. Colocalization of TRF2 and PML, denoting an ALT cell, appeared as distinct nuclear foci consistent with the appearance of APBs. Colocalization was confirmed using confocal microscopy to ensure that signals were overlapping in the same focal plane. Examples of confocal micrographs of individual representative tumor cells are shown in Figure 2A. Cells scored as ALT-positive demonstrated colocalization of PML and TRF2 and minimal or no TERT staining; cells scored as telomerase-positive demonstrated no colocalization of PML and TRF2 but were positive for TERT staining. Of cells from tumor A, 48% were telomerase positive, whereas none exhibited TRF2/PML colocalization (Figure 2B); these results corroborated those from TRF Southern blot analysis and the TRAP assay. Conversely, tumors C and D contained 31% and 65% of cells, respectively, with colocalization of TRF2/ PML, corroborating whole tumor lysate assays. Tumor D completely lacked cells with telomerase expression, whereas tumor C contained some telomerase-positive cells (Figure 2B). However, tumor C was geographically heterogeneous, containing distinct areas of weak or robust staining; this dimorphism could explain the negative TRAP results for this tumor, even though some telomerase-positive cells were present. Finally, tumors B and E, which had mixed results in whole tumor assays, displayed two separate cell populations: one staining positive for ALT protein colocalization (32% and 34%, respectively) and one staining positive for telomerase (2% and 6%, respectively) (Figure 2B). The TMM results from all three assays are summarized in Table 2. Independent and blinded histopathological analyses were performed to confirm that ALT- and telomerase-positive cells likely represented two distinct populations of tumor cells, rather than infiltrating normal cells.
46
Discussion Our studies with an enhanced imaging system explicitly demonstrate that some human sarcomas are mosaic for two types of cells: those that use telomerase and those that use ALT. The presence of both cell types highlights a novel type of tumor heterogeneity and suggests that telomere maintenance may not be a static characteristic of all tumors. In addition, the presence of more than one TMM confounds accurate tumor assessment using a single method and indicates a need for precise tumor characterization. Our results also highlight the question of whether individual cells can switch TMM or whether a cell commits to telomerasedependent or telomerase-independent telomere maintenance. Furthermore, we speculate that tumor types, in addition to sarcomas, may also contain mosaic populations of cells with regard to TMM. Previous studies have reported a lower percentage of tumors that display characteristics of both TMMs than determined in our studies, which is 50%. Although those studies have used an either-or approach for the determination of TMM, we have combined both IHC and biochemical analyses to gain a more comprehensive view of the TMM status of some of these tumors. Some tumors contain a small percentage of ALT-positive or telomerase-positive cells and would likely be classified into one category or another with assays that analyze a tumor extract, such as the TRF Southern blot analysis or the TRAP assay. In these solutionphase assays, tumor cells are combined into a quasihomogeneous solution, in which a small percentage of one particular cell type would not be measured. Combining the results of these assays with a cell-by-cell analysis by IHC has identified multiple TMM cell types within one tumor. The IHC staining of TERT protein does not confirm telomerase activity, although it is highly correlated with telomerase activity27,28 and implies that those cells are not using ALT. TERT staining correlates well with TRAP activity in the subset of four homogeneously stained osteosarcomas for which IHC and biochemical assays were possible. Table 2 Analysis of Telomere Maintenance in Frozen Human Osteosarcomas Age TRF Southern TRAP Tumor Sex (years) blot analysis assay In situ IHC (%) A B C D E
M M F M F
53 21 81 69 22
Telomerase Telomerase ALT ALT ALT
þ þ/ þ/
Telomerase Telomerase Telomerase ALT (65) Telomerase
(48) (2)/ALT (32) (7)/ALT (31) (6)/ALT (34)
Summary of TMM characteristics in five frozen human osteosarcomas. Telomere length was analyzed by TRF Southern blot analysis, telomerase activity was analyzed by TRAP assay, and single-cell telomerase and ALT expression was analyzed by in situ IHC. The percentage of cells was calculated from total cells in a field, in which some cells expressed neither colocalization of TRF2/PML nor telomerase. F, female; M, male; þ, positive result; , negative result.
ajp.amjpathol.org
-
The American Journal of Pathology
Mosaicism of ALT and Telomerase TMM can be a prognosticator for patient survival: ALTpositive glioblastoma multiforme tumors are associated with a better prognosis,6,22 whereas ALT-positive liposarcomas are associated with a worse prognosis11 than telomerasepositive tumors of each type. Recent clinical trials of telomerase-targeted therapies also underline the clinical importance of telomere maintenance. Mosaicism of TMM within human tumors may affect the choice of therapeutic approaches, because tumors with cells that maintain telomeres via more than one mechanism may require combination therapies to target each type of cell. In addition, recent work in mouse models demonstrates that inhibition of telomerase activity after tumor formation promotes the survival of tumors via the ALT pathway.29 A similar scenario is likely to occur in mosaic human tumors after telomerase inhibitor treatment, stressing the value of precise tumor TMM characterization before treatment and/or treatment with combination therapies that target both telomerase and ALT.
Acknowledgments We thank past and current members of the Groden laboratory and Dr. Samir Acharya for helpful discussions and technical advice, and The Ohio State University Core Facilities: Nucleic Acids Shared Resource, Campus Microscopy and Imaging Facility, and Pathology Core Facility.
Supplemental Data Supplemental material for this article can be found at http:// dx.doi.org/10.1016/j.ajpath.2012.10.001.
References 1. Henson JD, Reddel RR: Assaying and investigating alternative lengthening of telomeres activity in human cells and cancers. FEBS Lett 2008, 584:3800e3811 2. Heaphy CM, Subhawang AP, Hong SM, Goggins MG, Montgomery EA, Gabrielson E, Netto GJ, Epstein JI, Lotan TL, Westra WH, Shih IeM, Iacobuzio-Donahue CA, Maitra A, Li QK, Eberhart CG, Taube JM, Rakheja D, Kurman RJ, Wu TC, Roden RB, Argani P, De Marzo AM, Terracciano L, Torbenson M, Meeker AK: Prevalence of alternative lengthening of telomeres telomere maintenance mechanism in human cancer subtypes. Am J Pathol 2011, 179: 1608e1615 3. Bryan TM, Englezou A, Dalla-Pozza L, Dunham MA, Reddel RR: Evidence for an alternative mechanism for maintaining telomere length in human tumors and tumor-derived cell lines. Nat Med 1997, 3: 1271e1274 4. Ulaner GA, Huang HY, Otero J, Zhao Z, Ben-Porat L, Satagopan JM, Gorlick R, Meyers P, Healey JH, Huvos AG, Hoffman AR, Ladanyi M: Absence of a telomere maintenance mechanism as a favorable prognostic factor in patients with osteosarcoma. Cancer Res 2003, 63:1759e1763 5. Sanders RP, Drissi R, Billups CA, Daw NC, Valentine MB, Dome JS: Telomerase expression predicts unfavorable outcome in osteosarcoma. J Clin Oncol 2004, 22:3790e3797
The American Journal of Pathology
-
ajp.amjpathol.org
6. Henson JD, Hannay JA, McCarthy SW, Royds JA, Yeager TR, Robinson RA, Wharton SB, Jellinek DA, Arbuckle SM, Yoo J, Robinson BG, Learoyd DL, Stalley PD, Bonar SF, Yu D, Pollock RE, Reddel RR: A robust assay for alternative lengthening of telomeres in tumors shows the significance of alternative lengthening of telomeres in sarcomas and astrocytomas. Clin Cancer Res 2005, 11:217e225 7. Guilleret I, Yan P, Guillou L, Braunschweig R, Coindre JM, Benhattar J: The human telomerase RNA gene (hTERC) is regulated during carcinogenesis but is not dependent on DNA methylation. Carcinogenesis 2002, 23:2025e2030 8. Montgomery E, Argani P, Hicks JL, DeMarzo AM, Meeker AK: Telomere lengths of translocation-associated and nontranslocationassociated sarcomas differ dramatically. Am J Pathol 2004, 164: 1523e1529 9. Yan P, Benhattar J, Coindre J-M, Guillou L: Telomerase activity and hTERT mRNA expression can be heterogeneous and does not correlate with telomere length in soft tissue sarcomas. Int J Cancer 2002, 98: 851e856 10. Johnson JE, Varkonyi RJ, Schwalm J, Cragle R, Klein-Szanto A, Patchefsky A, Cukierman E, von Mehren M, Broccoli D: Multiple mechanisms of telomere maintenance exist in liposarcomas. Clin Cancer Res 2005, 11:5347e5355 11. Costa A, Daidone MG, Daprai L, Villa R, Cantu S, Pilotti S, Mariani L, Gronchi A, Henson JD, Reddel RR, Zaffaroni N: Telomere maintenance mechanisms in liposarcomas: association with histologic subtypes and disease progression. Cancer Res 2006, 66: 8918e8924 12. Matsuo T, Shay JW, Wright WE, Hiyama E, Shimose S, Kubo T, Sugita T, Yasunaga Y, Ochi M: Telomere-maintenance mechanisms in soft-tissue malignant fibrous histiocytomas. J Bone Joint Surg Am 2009, 91:928e937 13. Zimmermann S, Voss M, Kaiser S, Kapp U, Waller CF, Martens UM: Lack of telomerase activity in human mesenchymal stem cells. Leukemia 2003, 17:1146e1149 14. Cerone MA, Londono-Vallejo JA, Bacchetti S: Telomere maintenance by telomerase and by recombination can coexist in human cells. Hum Mol Genet 2001, 10:1945e1952 15. Grobelny JV, Kulp-McEliece M, Broccoli D: Effects of reconstitution of telomerase activity on telomere maintenance by the alternative lengthening of telomeres (ALT) pathway. Hum Mol Genet 2001, 10: 1953e1961 16. Perrem K, Colgin LM, Neumann AA, Yeager TR, Reddel RR: Coexistence of alternative lengthening of telomeres and telomerase in hTERT-transfected GM847 cells. Mol Cell Biol 2001, 21: 3862e3875 17. Ford LP, Zou Y, Pongracz K, Gryaznov SM, Shay JW, Wright WE: Telomerase can inhibit the recombination-based pathway of telomere maintenance in human cells. J Biol Chem 2001, 276: 32198e32203 18. Katoh M, Katoh M, Kameyama M, Kugoh H, Shimizu M, Oshimura M: A repressor function for telomerase activity in telomerase-negative immortal cells. Mol Carcin 1998, 21:17e25 19. Villa R, Daidone MG, Motta R, Venturini L, De Marco C, Vannelli A, Kusamura S, Baratti D, Deraco M, Costa A, Reddel RR, Zaffaroni N: Multiple mechanisms of telomere maintenance exist and differentially affect clinical outcome in diffuse malignant peritoneal mesothelioma. Clin Cancer Res 2008, 14:4134e4140 20. Else T, Giordano TJ, Hammer GD: Evaluation of telomere length maintenance mechanisms in adrenocortical carcinoma. J Clin Endocrinol Metab 2008, 93:1442e1449 21. Gupta J, Han L-P, Wang P, Gallie BL, Bacchetti S: Development of retinoblastoma in the absence of telomerase activity. J Natl Cancer Inst 1996, 88:1152e1157 22. Hakin-Smith V, Jellinek DA, Levy D, Carroll T, Teo M, Timperley WR, McKay MJ, Reddel RR, Royds JA: Alternative lengthening of telomeres and survival in patients with glioblastoma multiforme. Lancet 2003, 361:836e838
47
Gocha et al 23. Omori Y, Nakayama F, Li D, Kanemitsu K, Semba S, Ito A, Yokozaki H: Alternative lengthening of telomeres frequently occurs in mismatch repair system-deficient gastric carcinoma. Cancer Sci 2008, 100:413e418 24. Nuovo GJ: In situ detection of microRNAs in paraffin embedded, formalin fixed tissues and the co-localization of their putative targets. Methods 2010, 52:307e315 25. Grobelny JV, Godwin AK, Broccoli D: ALT-associated PML bodies are present in viable cells and are enriched in cells in the G(2)/M phase of the cell cycle. J Cell Sci 2000, 113:4577e4585 26. Cawthon RM: Telomere length measurement by a novel monochrome multiplex quantitative PCR method. Nucleic Acids Res 2009, 37:e21
48
27. Kanaya T, Kyo S, Takakura M, Ito H, Namiki M, Inoue M: hTERT is a critical determinant of telomerase activity in renal-cell carcinoma. Int J Cancer 1998, 78:539e543 28. Wu A, Ichihashi M, Ueda M: Correlation of the expression of human telomerase subunits with telomerase activity in normal skin and skin tumors. Cancer 1999, 86:2038e2044 29. Hu J, Hwang SS, Liesa M, Gan B, Sahin E, Jaskelioff M, Ding Z, Ying H, Boutin AT, Zhang H, Johnson S, Ivanova E, KostAlimova M, Protopopov A, Wang YA, Shirihai OS, Chin L, DePinho RA: Antitelomerase therapy provokes ALT and mitochondrial adaptive mechanisms in cancer. Cell 2012, 148:651e663
ajp.amjpathol.org
-
The American Journal of Pathology