Cancer Research The Future of Cancer Research: Science and Patient Impact Translational Medicine Conference in Israel
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dalerba, P.
Right arrow Articles by Castelli, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dalerba, P.
Right arrow Articles by Castelli, C.
[Cancer Research 65, 2321-2329, March 15, 2005]
© 2005 American Association for Cancer Research

Cell and Tumor Biology

Reconstitution of Human Telomerase Reverse Transcriptase Expression Rescues Colorectal Carcinoma Cells from In vitro Senescence: Evidence against Immortality as a Constitutive Trait of Tumor Cells

Piero Dalerba1, Cristiana Guiducci2, Pietro Luigi Poliani3, Ingrid Cifola1, Mariella Parenza2, Milo Frattini4, Gianfrancesco Gallino5, Ileana Carnevali4, Ilvia Di Giulio1, Salvatore Andreola3, Claudia Lombardo6, Licia Rivoltini1, Tamás Schweighoffer7, Filiberto Belli5, Mario P. Colombo2, Giorgio Parmiani1 and Chiara Castelli1

Units of 1 Immunotherapy of Human Tumors, 2 Immunotherapy and Gene Therapy, 3 Pathology B, 4 Pathology C, 5 Colorectal Surgery, and 6 Immunohematology, Istituto Nazionale Tumori, Milan, Italy and 7 Novartis Institutes for BioMedical Research, Vienna, Austria

Requests for reprints: Chiara Castelli, Unit of Immunotherapy of Human Tumors, Department of Experimental Oncology, Istituto Nazionale Tumori, via G. Venezian 1, 20133 Milan, Italy. Phone: 39-02-23903230; Fax: 39-02-23902630; E-mail: chiara.castelli{at}istitutotumori.mi.it.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Although in vitro establishment of new colorectal carcinoma (CRC) cell lines is an infrequent event, we have observed that primary cultures of CRC can be repeatedly and reproducibly initiated following in vitro plating of tumor-derived epithelial cells. These cultures, however, usually display a short life span as they undergo a limited number of cell passages before entering a state of irreversible growth arrest. In this study, we show that short-lived CRC primary cultures lack constitutive telomerase activity and undergo a senescence process characterized by progressive telomere shortening. Moreover, transduction of these cells with a retroviral vector encoding human telomerase reverse transcriptase (hTERT) is sufficient to reconstitute telomerase activity and allow immortalization. Detailed molecular characterization of hTERT-immortalized CRC cell lines confirms their individual tumor origin by showing expression of colonic epithelial differentiation markers, such as cytokeratin-20 (CK20), full match with class I and class II human leukocyte antigen genotyping of autologous B-lymphoblastoid cells, and presence of somatic mutations in key cancer genes (KRAS2, APC) identical to those of the corresponding autologous original tumor tissues. Moreover, functional characterization of hTERT-immortalized CRC cell lines shows that they have a transformed phenotype, being able to form colonies in soft agar and tumors in severe combined immunodeficient mice. Most interestingly, immunohistochemical analysis of original tumor tissues indicates that short-lived CRC primary cultures, although hTERT-negative in vitro, derive from hTERT-positive tumors. Taken together, our data show that, in a least subset of CRC, biochemical pathways involved in maintenance of telomere length, such as telomerase, are not activated in a constitutive way in all tumor cells.

Key Words: hTERT • colorectal cancer • immortality • primary cultures • cell lines


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Colorectal carcinoma (CRC) cell lines are difficult to establish in vitro. Indeed, the low success rate of this process (5-10%) represents a limiting step in experimental studies on CRC (1). Several reasons can account for this observation: (a) microbial contamination of primary tumor surgical specimens; (b) inadequate sampling of tumor tissues either qualitative (e.g., necrotic tissue) or quantitative (e.g., small sample); (c) lack of tumor-cell adhesion to plastic culture flasks (2); and (d) lack of necessary growth factors in standard culture media (3).

We have observed that, in many cases, short-term CRC primary cultures can be easily and reproducibly obtained following in vitro plating of tumor-derived epithelial cells. These cultures, however, display a short life span as they undergo a limited number of cell passages before entering a state of irreversible growth arrest (defined hereafter as "unstable" CRC primary cultures). Unstable cultures can be repeatedly initiated from early-passage frozen aliquots but reproducibly arrest after undergoing a defined number of in vitro cell passages, which is specific to each individual culture. This behavior, which resembles a physiologic senescence process typical of in vitro growing normal cells, is puzzling and unexpected because CRC cells, like all malignant cells, are believed to be constitutively immortal as a consequence of neoplastic transformation (4).

In normal adult somatic cells, in vitro growth is associated with progressive shortening of telomeres, which eventually causes chromosomal instability and activates a senescence process leading to growth arrest. Therefore, as a prerequisite for unlimited in vitro growth (immortality), eukaryotic cells must activate a system for the maintenance of telomere length. In most human tumors (>90%) as well as in normal germ cells, telomere length is maintained by a ribonucleoprotein enzymatic complex, telomerase, whose function is the de novo synthesis of telomeric DNA (5, 6) . In a minority of cases, especially in tumors of mesenchymal origin (osteosarcomas, soft tissue sarcomas, and adrenocortical carcinomas), telomere length is maintained by one or more alternative mechanisms, known together as alternative lengthening of telomeres (ALT) and presumably based on telomere recombination (7, 8).

The aim of this study was to elucidate the molecular mechanisms responsible for the limited in vitro life span of CRC primary cultures and to devise a strategy for its correction. Using six CRC primary cultures obtained in our laboratories, three of which known to be unstable and three known to progress into "stable" cell lines, we show that "instability" is associated with lack of telomerase activity due to lack of human telomerase reverse transcriptase (hTERT) expression. Moreover, we show that reconstitution of hTERT expression is sufficient to allow conversion of unstable primary cultures into immortal cell lines, which maintain the individual genetic fingerprints of the corresponding original cancer tissues, together with the ability to form tumors in immunodeficient mice. Most interestingly, we show that although unstable primary cultures lack hTERT expression in vitro, their parental tumor tissue samples score as hTERT-positive in vivo. Our findings indicate that not all tumor cells, or at least not all malignant cells within the same tumor, are endowed with a constitutively immortal phenotype. This phenotype, however, is readily acquired following reconstitution of hTERT expression alone, which can be exploited as a simple and effective procedure for the in vitro establishment of new CRC cell lines.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell Lines and Primary Cultures. All CRC primary cultures described in this study are listed in Table 1. They were directly derived from surgical specimens, with the only exception of MICOL-14 cells, which were derived from the first passage of a solid tumor xenograft grown in nude mice. Tumor tissues were mechanically processed and enzymatically digested as described (9). All CRC cells had a clear epithelial morphology, stained positive for CK20, and were serially passaged by trypsin disaggregation. All B-lymphoblastoid cell lines (B-LCL) were originated in our laboratories from peripheral blood mononuclear cells of the corresponding patients. HT29, LoVo, U2OS, and 3T6 cells were obtained from the American Type Culture Collection. GM847 cells were obtained from Coriell Cell Repositories (Camden, NJ). All cell lines and primary cultures were maintained in complete medium (RPMI 1640 supplemented with 10% heat-inactivated FCS, 2 mmol/L L-glutamine, 120 µg/mL penicillin, 40 µg/mL gentamicin, and 20 mmol/L HEPES) and were routinely tested for Mycoplasma contamination.


View this table:
[in this window]
[in a new window]
  		Table 1. Origin, morphology, differentiation markers, and growth behavior of CRC primary culture

 


View larger version (117K):
[in this window]
[in a new window]
  		Figure 6. Epithelial morphology, transformed phenotype, and tumorigenic potential of hTERT-immortalized CRC cell lines. hTERT-immortalized CRC cell lines display a clear epithelial morphology, growing in vitro as adherent monolayers composed of polygonal cells tightly packed in sharp-edged cell nests (A, MICOL-ShTERT). All three cell lines express epithelial differentiation markers (CK20) and stain negative with control antibodies (CD45) in immunocytochemistry (B and C, CG-756hTERT). All three cell lines form multicellular spherical colonies in soft agar (D, MICOL-ShTERT, original magnification x20). S.c. injection of MICOL-14hTERT cells mixed with {gamma}-irradiated 3T6 cells was followed by the growth of solid tumors (E); injection of {gamma}-irradiated 3T6 cells alone on the contralateral side of the same animal (*) did not result in tumor growth. Histologic and immunohistochemical analysis confirmed that all in vivo tumors were adenocarcinomas and expressed human CK20 (F). I.p. injection of MICOL-14hTERT tumor cells lead to the formation of perihepatic solid tumor masses (G, arrows), with the histologic appearance of cystoadenocarcinomas (H). Similarly, i.p. injection of MICOL-ShTERT cells led to the formation of both solid perihepatic tumor masses and widespread peritoneal carcinomatosis (I-L). Solid tumors (I, arrow) formed preferentially along the gastrohepatic ligament, between stomach (s) and liver (arrowheads); in all cases, tumors consisted of poorly differentiated carcinomas and expressed human CK20 (J). Peritoneal carcinomatosis (K) presented as a diffuse thickening of the abdominal wall (a.w.) due to massive mesothelial substitution by CK20+ tumor cell nests (arrows). Peritoneal carcinomatosis also extended along the serosal surface of all major abdominal organs, especially the liver (L), where tumor cells formed a thick coating (arrows) around the liver parenchyma (liver).

 
PCR-Based Telomere Repeat Amplification Protocol Assay. Telomere repeat amplification protocol (TRAP) assay was done on whole cell extract as previously described (10, 11). See Supplementary data for details.

Reverse Transcription-PCR. Total RNA was extracted using the RNAqueous system (Ambion, Austin, TX) and treated with RNase-free DNase I (Ambion). RNA quality was assessed on 1% agarose gel: Samples were defined of good quality when both 28S and 18S rRNA bands where clearly visible. Good quality total RNAs were reverse-transcribed into single-stranded cDNA using random hexamers and SuperScript II M-MLV reverse transcriptase (Life Technologies Italia SRL, San Giuliano Milanese, Italy). For each sample, a mock reaction without reverse transcriptase (RT) was done in parallel and used as a negative control. cDNA quality was checked by RT-PCR with ß-actin-specific oligonucleotide primers. Samples were scored positive when a band of the appropriate size of 615 bp was visible on the gel.

RT-PCR were done using as template 0.25 µg of reverse-transcribed RNA. The amplification reaction included the following steps: denaturation at 95°C for 5 minutes; 38 cycles of denaturation at 95°C for 30 seconds, annealing at 65°C for 45 seconds, and extension at 72°C for 45 seconds; and final extension at 72°C for 15 minutes. To avoid amplification of residual genomic DNA contaminant, oligonucleotide primers were complementary to gene-specific sequences spanning different exons. Primer sequences were TERT.1 (sense) 5'-TTCCTGCACTGGCTGAT GAGTGT-3' and TERT.2 (antisense) 5'-CGCTCGGCCCTCTTTTCTCTG-3'. Expected length of the amplification product is 329 bp. Amplification products obtained with this RT-PCR protocol are specific as previously assessed on a panel of colorectal adenoma and carcinoma tissues (12).

Retroviral Vectors and Transduction Protocols. The full-length hTERT cDNA (13) was cloned into EcoRI and XhoI site of pLXSN retroviral vector (BD Biosciences, Franklin Lakes, NJ). hTERT-pLXSN retroviral vector was transfected into the ecotropic gp+E86 packaging cell line by standard calcium phosphate coprecipitation, and the 48-hour culture supernatant was used to infect the amphotropic Am12 packaging cell line. Infected Am12 were selected with neomycin and used to generate helper-free virus-containing supernatants. Tumor cells were infected by three cycles of overnight exposure to undiluted supernatant in the presence of polybrene (8 µg/mL) and 48 hours after infection tumor cells were selected in 1 mg/mL of G418. The pLBSN retroviral vector encoding -galactosidase (-gal) was a gift of Dr. Giuliana Ferrari (Telethon Institute for Gene Therapy, San Raffaele, Milan, Italy).

Calculation of Population Doublings. Calculation of population doublings was done at each cell passage, assuming exponential growth of tumor cells, according to the following formula:


where N0 is the number of cells at the time of plating in culture flask (beginning of the growth period), Nx is the number of cells at the time of harvest (end of the growth period), and x is the number of population doublings between N0 and Nx.

Analysis of Terminal Restriction Fragment Length. Analysis of terminal restriction fragment (TRF) length was done as described (14). A detailed description of the protocol used in this study for the analysis of TRF length is provided as Supplementary data.

Immunocytochemistry and Immunohistochemistry. Immunocytochemical analysis of in vitro cultured cells was done on cytospins following either acetone or formalin fixation, whereas immunohistochemical analysis of tumor tissues was done on formalin-fixed, paraffin-embedded tissue sections. Primary monoclonal antibodies (mAb) used in this study include anti-human CD45 monoclonal antibody (1:200; clones 2B11 + PD7/26, DakoCytomation, Glostrup, Denmark), anti-human CK20 monoclonal antibody (1:50; clone Ks20.8; NeoMarkers, Fremont, CA), and anti-hTERT monoclonal antibody (NCL-hTERT, clone 44F12, Novocastra, Newcastle upon Tyne, United Kingdom). Immunohistochemistry for hTERT was done following the protocol recently set-up by Yan et al. (15), which allows reliable and specific hTERT-specific staining under very strict binding conditions. For each experiment, a negative control for primary antibody staining was done in parallel by using a pool of two different mouse monoclonal antibodies of undefined specificity (see Supplementary data for details).

Molecular Class I and Class II Human Leukocyte Antigen Genotyping. Molecular class I and class II human leukocyte antigen (HLA) typing was done on genomic DNA by PCR with sequence-specific primers, using the Olerup SSP HLA typing system (Olerup SSP AB, Saltsjöbaden, Sweden).

Analysis of APC and KRAS2 Mutations. Analysis of APC and KRAS2 mutations was done as described (16). Genomic DNA was extracted from paraffin-embedded tissues using the QIAamp Tissue Extraction kit (Qiagen, Chatsworth, CA), according to manufacturer's instructions. All samples were screened for mutations in APC (exon 15) or KRAS2 (codons 12, 13, and 61). Samples identified as mutated were subjected to automated sequencing with ABI Prism 377 DNA Sequencer (Applied Biosystems, Foster City, CA) and analyzed with Sequencing Analysis and Sequence Navigator software programs (Applied Biosystems). Each sequence reaction was done at least twice, and each mutation was confirmed both on sense and antisense strands.

Soft Agar Colony Formation Assay. Soft agar colony formation assays were done according to a protocol kindly provided by Prof. Rob Ramsay (Peter MacCallum Cancer Center, Melbourne, Australia). Shortly after trypsinization, tumor cells were resuspended in 1.5 mL liquid (40°C) top agarose mixture (RPMI 1640, 10% FCS, 0.3% w/v agarose) and immediately added in 35-mm-diameter wells with a basal layer of 1.5 mL of a previously solidified agar (RPMI 1640, 10% FCS, 0.5% w/v agar). After top agarose was solidified, plates were incubated at 37°C in 7.5% CO2. For each cell line, the experiment was done twice at two different cellular densities (5 x 103 and 50 x 103 cells/well) and in triplicates; only spherical, multicellular aggregates were scored as positive colonies.

In vivo Tumorigenicity Assays. To assess their in vivo tumorigenicity, CRC cells were injected in 6- to 8-week-old female severe combined immunodeficient mice (Charles River Laboratories Italia S.p.A., Calco, Italy), in escalating doses (range 2-25 x 106). All experiments were done in accordance with institutional and international guidelines (17). Tumor cells were harvested after trypsin detachment of subconfluent monolayers, counted, washed once in PBS, and readily inoculated either by s.c. or i.p. injection. Mice health and tumor diameter were controlled every 7 days; s.c. tumors were allowed to grow up to 1 cm of maximum diameter, and tumor take was considered as having occurred only when tumors grew progressively up to this size. To increase chances of tumor engraftment, and to accelerate in vivo tumor growth kinetics, all experiments were done both with tumor cells alone and with tumor cells admixed with -irradiated (100 Gy) 3T6 murine embryo fibroblasts at 1:1 ratio (e.g., 10 x 106 tumor cells + 10 x 106 -irradiated 3T6 cells). To exclude the possibility that tumor formation could be attributed to 3T6 cells rather than to CRC cells, all mice were injected with an equal inoculum of 3T6 cells alone on the contralateral side. No growth of -irradiated 3T6 cells alone was observed. Moreover, all tumors grown in severe combined immunodeficient mice were confirmed to be adenocarcinomas by the pathologist (P.L. Poliani) and be composed by epithelial cells by immunohistochemical staining for human CK20.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Lack of Telomerase Activity and hTERT Messenger RNA in Short-Lived (Unstable) Primary Cultures. To evaluate whether lack of long-term stabilization was associated with lack of telomerase enzymatic activity, we did TRAP assays on the six primary cultures included in our study panel. Whereas all three stable primary cultures (CG-705, CG-758, and MICOL-29), as well as "classic" stable CRC cell lines (HT29, LoVo), resulted strongly telomerase positivfe, all three unstable ones (CG-756, MICOL-S, and MICOL-14) were telomerase-negative (Fig. 1 and data not shown for CG-705 stable primary cells and LoVo cell line). To shed light on the molecular mechanism underlying lack of telomerase activity in unstable primary cultures, we next analyzed hTERT expression by RT-PCR. We focused on hTERT because, unlike other components of the telomerase enzymatic complex, such as human telomerase RNA (hTR) or telomerase-associated protein (TP1), hTERT is the main determinant of telomerase activity due to its nonubiquitous and highly regulated expression (12, 18). So RT-PCR results would not be influenced by contamination of primary cultures by normal cells known to be telomerase-positive (e.g., intratumor T lymphocytes; ref. 19), the analysis was done on cells between the 7th and the 13th in vitro passage (p7-p13). Whereas all three stable primary cultures (CG-705, CG-758, and MICOL-29), as well as classic stable CRC cell lines (HT29 and LoVo), scored hTERT-positive, no amplification of hTERT cDNA was obtained from the three unstable ones (CG-756, MICOL-S, and MICOL-14), suggesting that lack of telomerase activity was due to lack of hTERT transcription (Fig. 2).



View larger version (54K):
[in this window]
[in a new window]
  		Figure 1. TRAP assay on stable and unstable CRC primary cultures. CRC primary cultures known to progress spontaneously into long-term stable cell lines (stable primary cultures) scored positive for telomerase activity, as shown for MICOL-29 and CG-758 cells collected, respectively, at the 6th (p6) and 26th (p26) in vitro passage. Identical results were obtained with the third stable primary culture included in this study (CG-705, data not shown). Telomerase levels seemed comparable with those of a classic CRC cell line, used as a positive control (HT29). Primary cultures known to grow only for a limited number of cell passages and unable to progress into stable cell lines (unstable primary cultures), scored negative as shown for MICOL-S, CG-756, and MICOL-14 cells analyzed at the 13th, 21st, and 6th in vitro passage, respectively. All experiments were done using 5 µg of cell-extract proteins. TRAP assay sensitivity was controlled on decreasing quantities of HT29 cell-extract proteins, in an interval between 1 and 4 µg: No substantial modification of test sensitivity was observed within this range. As a negative control, the assay was done on lysis buffer alone in the absence of any cell-extract.

 


View larger version (52K):
[in this window]
[in a new window]
  		Figure 2. RT-PCR analysis of hTERT mRNA expression in early passage stable and unstable CRC primary cultures. Top, RT-PCR analysis shows that hTERT gene expression is detectable in all three CRC stable primary cultures (CG-705, CG-758, and MICOL-29), whereas it is absent in all three unstable ones (CG-756, MICOL-S, and MICOL-14). RT-PCR analysis was done on cell lines collected before the 12th in vitro passage. Negative controls are represented by the GM847 fibroblast cell line, which has an ALT phenotype and is telomerase negative, and by a sample where no cDNA was added to the reaction. Two classic stable CRC cell lines (HT29 and LoVo) and GM847 cells transfected with an hTERT-encoding plasmid (GM847-hTERT) were used as positive controls. Bottom, equal amount of cDNA used for hTERT analysis was checked for ß-actin expression. Amplification of hTERT and ß-actin cDNAs gives unique bands of 329 and 615 bp, respectively. The molecular weight marker (m.w.) is the 100 bp DNA ladder (Invitrogen SRL, San Giuliano Milanese, Italy).

 
Reconstitution of hTERT Expression by Retroviral Gene Transfer Restores Telomerase Activity and Allows Immortalization of Short-Lived Primary Cultures. In the attempt to reconstitute telomerase activity, we produced a retroviral vector encoding the full hTERT cDNA and used it to transduce the three unstable primary CRC cultures. As a negative control, all three primary cultures were transduced in parallel with a retroviral vector encoding ß-gal. In all three cases, retroviral transduction of hTERT allowed reconstitution of telomerase activity, as assessed by the TRAP assay, whereas transduction of ß-gal had no effect (Fig. 3). To evaluate whether reconstitution of telomerase activity was sufficient to allow immortalization, the three pairs of primary cultures (transduced with hTERT or ß-gal, respectively) were maintained in vitro and followed over time, calculating the number of population doublings for each cell passage. In all three cases, cells transduced with ß-gal progressively reduced their growth rate and eventually arrested, reproducing the growth behavior previously observed for parental nontransduced cells. On the contrary, all three hTERT-transduced CRC primary cultures continued to grow well beyond the critical passage where parental cultures usually arrested. All three hTERT-immortalized CRC cultures underwent >100 population doublings after retroviral transduction (Fig. 4).



View larger version (55K):
[in this window]
[in a new window]
  		Figure 3. Reconstitution of telomerase enzymatic activity in unstable CRC primary cultures by transduction with an hTERT-encoding retroviral vector. TRAP assay shows that transduction with the pLXSN-hTERT retroviral vector is sufficient to achieve full reconstitution of telomerase activity both in control GM847 fibroblasts and in MICOL-14 unstable primary cells. Identical results have been obtained with CG-756 and MICOL-S unstable primary cultures (data not shown). MICOL-14 cells transduced with a mock construct encoding ß-gal remain telomerase negative. Negative controls for TRAP assay include the U2OS osteosarcoma cell line, which has an ALT phenotype and a sample of lysis buffer alone in the absence of any cell extract.

 


View larger version (20K):
[in this window]
[in a new window]
  		Figure 4. Reconstitution of hTERT expression by retroviral gene transfer allows immortalization of unstable CRC primary cultures. A frozen aliquot of MICOL-S, CG-756, and MICOL-14 cells was thawed, grown for one passage ({lozenge}), and then split and transduced in parallel with ß-gal and hTERT-encoding retroviral vectors. All cell cultures were subsequently followed over time, recording the number of population doublings for each cell passage. In all three cases, cells transduced with ß-gal ({triangleup}) progressively reduced their growth rate and eventually entered a state of growth arrest ({dagger}), reproducing the behavior of parental cells. On the contrary, cells transduced with hTERT ({blacksquare}) kept growing in vitro, well beyond the critical passage where parental cells usually arrest (compare with Table 1). All three hTERT-transduced unstable CRC primary cultures underwent >50 serial in vitro passages and >100 population doublings because the critical passage where their parental and ß-gal counterparts have arrested. MICOL-S, CG-756, and MICOL-14 cells were retrovirally transduced at the 12th (p12), 21st (p21), and 5th (p5) in vitro passage, respectively. Cells transduced with ß-gal arrested as expected at the 14th (p14), 25th (p25), and 8th (p8) in vitro passage, respectively, whereas cells transduced with hTERT continued to grow beyond the 64th (p64), 81st (p81), and 71st (p71) in vitro passage, respectively.

 
Progressive Telomere Shortening Is Associated with Growth Arrest of Unstable Primary Cultures and Is Reversed by hTERT Immortalization. To gain insight into the molecular mechanisms underlying growth arrest of unstable primary cultures and their rescue by hTERT transduction, we did a prospective evaluation of telomere length in CG-756 cells, comparing the progressive change in TRF length of untreated cells with that of hTERT-immortalized ones. CG-756 cells were chosen for this analysis on the basis of the relatively long in vitro life span of untreated cells (~28 cell passages). As shown in Fig. 5, untreated cells displayed progressive shortening of TRF length, therefore confirming that growth arrest was associated with replicative senescence, characterized by telomere erosion and dysfunction. On the contrary, their hTERT-transduced counterparts showed reconstitution and maintenance of telomere sequences at a level comparable with that of a stable CRC cell line, such as HT29.



View larger version (30K):
[in this window]
[in a new window]
  		Figure 5. Prospective evaluation of TRF length in parental and hTERT-immortalized CG-756 cells. Unstable CG-756 cells (wt, black column, {triangleup}) underwent progressive shortening of TRF length during in vitro culture, from 3,800 bp at the 13th passage (p13) down to 2,300 bp at the 28th passage (p28, where wt CG-756 cells usually arrest; see Table 1). On the contrary, their hTERT-transduced counterparts (hTERT, hatched columns, {blacksquare}) showed progressive reconstitution of telomere sequences and TRF lengthening, from 2,800 bp at the 21st passage (p21, when transduction was done, see Fig. 4) up to 5,000 bp at the 28th passage (p28) and 9,000 bp at the 63rd passage (p63). TRF length in CG-756hTERT cells is comparable with that of classic stable CRC cell lines, such as HT29 (white column).

 
Molecular and Functional Characterization of hTERT-Immortalized Colorectal Carcinoma Lines. Because the observation that in vitro cultured CRC cells could spontaneously undergo a senescence process was unexpected and contradicts a basic assumption of cancer biology (i.e., that malignant tumor cells are constitutively immortal), we decided to perform a detailed characterization of our hTERT-immortalized cell lines to rule out the possibility that our results could be explained by trivial experimental mistakes, such as immortalization of normal cells or in vitro outgrowth of contaminating cells from other tumor cell lines.

To rule out the possibility that the three unstable primary cultures were originated by the outgrowth of normal fibroblasts, we evaluated expression of CK20, an epithelial differentiation marker preferentially expressed by colonic epithelium and CRC (20). Moreover, all three hTERT-immortalized cell lines maintained the characteristic epithelial morphology of their parental primary cultures (Fig. 6A). All three hTERT-immortalized cell lines displayed strong immunoreactivity for CK20 and scored negative for an unrelated lineage marker such as CD45 (Fig. 6B and C).

Cross-contamination of the three hTERT-immortalized cell lines with one of the three stable ones, or with other CRC cell lines grown in our laboratories, was excluded by molecular class I and class II HLA typing of the six cell lines included in our study panel, and comparison of the typing results with those obtained in six autologous B-LCL originated from the corresponding patients (Table 2). In all cases, molecular HLA typing of tumor cells perfectly matched that of autologous B-LCL, indicating that all cell lines indeed originated from the patients whose tissue samples were used as cellular sources for primary cultures.


View this table:
[in this window]
[in a new window]
  		Table 2. Molecular genotyping of cell lines derived from both stable and unstable CRC primary culture

 
To obtain confirmation that all our primary cultures were indeed of tumor origin, we searched for somatic mutations in key cancer genes, such as the KRAS2 oncogene and the APC tumor-suppressor gene, all known to be mutated in a high percentage of human CRC (~50% and 70%, respectively). All six cell lines contained mutations in KRAS2 or APC, and each cell line was characterized by an individual set of mutations (Table 2). All mutations were absent in autologous B-LCL, thus indicating that they were all somatically acquired. Most importantly, in the three cases where comparison with the original tumor tissue was possible (MICOL-29, stable; MICOL-14, unstable; MICOL-S, unstable), the set of mutations detected in cell lines was identical to that of the corresponding original tumor tissues. This observation formally proved that at least two of three unstable primary cultures were indeed representative of the tumor cell population that formed the original metastatic tumor mass in vivo.

We next evaluated whether hTERT-immortalized cell lines not only contained the genetic fingerprints of original tumor cells but were also endowed with functional features of malignancy. First, we assessed their capacity for anchorage-independent growth in a soft-agar colony formation assay. All three hTERT-immortalized cell lines formed round-shaped multicellular colonies in soft agar, thus indicating that they had a transformed phenotype (Fig. 6D). As a second step, we evaluated whether the three hTERT-immortalized cell lines were able to form tumors in vivo when injected in immunodeficient severe combined immunodeficient mice. All cell lines were injected both s.c. and i.p. To increase the chances of tumor take in vivo, s.c. injections were done either with tumor cells alone or admixed with {gamma}-irradiated 3T6 murine fibroblasts (21, 22). Two of three hTERT-immortalized cell lines (MICOL-14hTERT and MICOL-ShTERT) proved tumorigenic. Injection of MICOL-14hTERT cells was followed by the formation of tumors both s.c. and i.p. Tumor take after s.c. injection was greatly improved by addition of irradiated feeder cells (Fig. 6E); i.p. injection lead to the formation of multiple solid tumor masses in the perihepatic and perisplenic fat (Fig. 6G). MICOL-ShTERT cells were nontumorigenic s.c. but readily engrafted i.p. (~3 months) forming both solid perihepatic tumor masses and widespread peritoneal carcinomatosis (Fig. 6I-L). All tumors were confirmed to be adenocarcinomas by an experienced pathologist (P.L. Poliani) and resulted CK20 positive by immunohistochemistry (Fig. 6F, H, and J-L). On the contrary, CG-756hTERT cells failed to grow as solid tumors in vivo. This observation, however, was not completely unexpected because it is well known that many human stable tumor cell lines, including CRC cell lines, are nontumorigenic in immunodeficient mice (21–24).

Immunohistochemical Analysis of hTERT Expression in Original Tumor Tissue Samples. To evaluate whether lack of hTERT expression in unstable primary cultures was due to lack of hTERT expression in the parental tumor in vivo, or to loss of hTERT expression as a result of in vitro culture, we analyzed hTERT expression in original tissue samples by immunohistochemistry, taking advantage of a recently developed staining protocol (15). In all three cases where original tumor tissue was available (MICOL-29, stable; MICOL-14, unstable; MICOL-S, unstable), staining for hTERT showed diffuse tumor cell nuclear positivity with no substantial difference between stable and unstable cases (Fig. 7). This observation shows that the short-lived CRC primary cultures derive from hTERT-positive CRC metastases. Interestingly, and in accordance with data reported by Yan et al. (15), hTERT-specific staining is frequently heterogeneous within the tumor cell population, with positive and negative cells intermingled inside the same tumor cell nests.



View larger version (144K):
[in this window]
[in a new window]
  		Figure 7. Immunohistochemical analysis of hTERT expression in original tumor tissues. Tumor tissues used as sources of living cells for the initiation of in vitro CRC primary cultures scored positive for hTERT protein expression [A and B, MICOL-29; C and D, MICOL-14; original magnification x20 (A and C); original magnification x40 (B and D)]. Staining for hTERT is restricted to tumor cells, is mainly nuclear, and is frequently heterogeneous within the tumor cell population, with some cells scoring negative (arrows). Stromal cells (S) are completely negative. Similar findings were obtained with MICOL-S (data not shown).

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
According to current models, one of the hallmarks of cancer is the acquisition by tumor cells of a limitless replicative potential, a condition also known as immortality (4). To become immortal, cancer cells must be endowed with a system for the maintenance of telomere length, either based on telomerase or ALT, which counteracts the progressive shortening of telomere repeat sequences that is usually observed during in vitro proliferation of normal somatic cells (5, 8). However, the molecular mechanisms underlying either telomerase or ALT activation in human cancer remain poorly understood (18, 25).

Our findings indicate that, in a subset of cases, in vitro cultured CRC cells display only a limited growth potential due to lack of telomerase activity and that retroviral transduction of the hTERT gene allows immortalization of telomerase-negative short-lived CRC primary cultures. Moreover, immunohistochemical analysis of parental tumor tissues shows that these short-lived CRC cells derive from hTERT-positive tumors.

A possible explanation of our findings is that although we can formally exclude that our unstable primary cultures are composed of normal cells, they could nonetheless be derived from cells still in the very early stage of tumor transformation, which have not yet fully undergone the process of telomere erosion, proliferative crisis, and selection of hTERT-positive clones (i.e., tumor cells from benign adenoma tissues adjacent to invasive tumor tissues sampled at the moment of specimen collection). Indeed, it is known that in CRC, telomerase activation is usually associated with the adenoma-to-carcinoma transition, and is absent in most low-grade adenomas (11, 12, 26). This explanation, however, is unlikely because at least two of three of our unstable primary cultures (MICOL-14 and MICOL-S) have been derived from metastatic lesions, and thus from the terminal stages of the corresponding tumor natural history (Table 1). From this point of view, our study substantially differs from those of Yasunaga et al. (27, 28), which successfully exploited an hTERT-based approach for the in vitro establishment of prostatic-epithelium cell lines from primary tumors.

Therefore, the main biological implication of our findings is that tumor cells from fully malignant, metastatic tumors are not necessarily endowed with a constitutively active biochemical pathway for the maintenance of telomere length. This observation is apparently in contrast with data collected on classic stable tumor cell lines, which all possess mechanisms for telomere maintenance, either based on telomerase or ALT. However, data collected from cell lines are of limited value because all stable cell lines, whether tumor derived or not, are immortal by definition. Our study, therefore, suggests that classic stable tumor cell lines may represent exceptional cases of constitutively immortal tumor cells, which have been artificially selected for the property of autonomous and unlimited in vitro growth by the very same procedure of in vitro culture itself. Indeed, it is well known that in vitro behavior of stable tumor cell lines is only partially representative of in vivo tumor cell biology and that long-term stable cancer cell lines can be established from surgical specimens with only very low success rates especially in the case of epithelial tumors (1, 29).

Thus, the most likely explanation of our findings is that unstable CRC primary cultures, although derived from telomerase-positive tumors, lack telomerase activity as a consequence of hTERT down-regulation. We propose two hypotheses to explain this phenomenon: a "tissue microenvironment" hypothesis and a "cancer stem cell" hypothesis. In the first case, hTERT expression is postulated to be inducible in tumor cells as a consequence of their interaction with the tissue microenvironment in vivo (e.g., fibroblasts within the stromal tissue, macrophages of tumor inflammatory infiltrates). This interpretation is consistent with the notion that hTERT expression is under the control of several tissue environmental factors, including hormones, growth factors, and cytokines (18), and especially with the fact that no mutations able to determine the constitutive activation of telomerase have hitherto been described in human tumors (25). According to this hypothesis, unstable CRC primary cultures lose hTERT expression as a consequence of being cultured in vitro alone.

In the second scenario, which takes into account the concept of cancer stem cell and its recent extension to solid tumors (30, 31), hTERT is postulated to be expressed and/or constitutively activated only in the cancer stem cell subpopulation. Indeed, it has recently been shown that one of the genes involved in the control of stem-cell self-renewal, Bmi-1, is also able to up-regulate hTERT in epithelial cells (32). According to this second hypothesis, unstable CRC primary cultures lack the cancer stem cell fraction of the original tumor and contain only differentiated and short-lived committed progenitors. It must be pointed out, however, that the tissue microenvironment and the cancer stem cell hypotheses are not mutually exclusive and that, interestingly, both predict that hTERT expression in vivo would be diffusely heterogeneous within the tumor cell population. Indeed, our present histochemical data and the observation previously published (15) confirm this prediction.

In conclusion, our data show that immortality is not necessarily an intrinsic trait of all tumor cells or at least of all malignant cells composing a tumor mass. This observation provides a partial explanation for why it is so difficult to obtain stable long-term cell lines from surgically resected tumor tissues and points to cancer stem cell and to their interaction with the tumor stroma as key emerging concepts for the understanding of this disease.

A final, purely methodologic, implication of our results is that retroviral transduction of hTERT can be exploited as a tool to obtain stabilization of new CRC cell lines. Indeed, the low success rate in establishment of stable long-term CRC cell lines still represents a limiting step for several experimental studies on CRC, especially for immunologic studies dealing with antitumor T-cell responses (33). In this setting, autologous pairs of tumor cell lines and patient-derived T-lymphocytes provide still unique, key experimental reagents both for the discovery of new tumor-associated antigens and for the monitoring of vaccine-induced antitumor T-cell responses in cancer patients enrolled in experimental clinical trials (34).


   Acknowledgments
 
Grant support: Consiglio Nazionale delle Ricerche (Progetto Strategico, Ministero dell'Istruzione, Università e Ricerca-Consiglio Nazionale delle Ricerche, Rome, Italy), Italian Ministry of Health (4AB/F6, Rome, Italy), Associazione Italiana per la Ricerca sul Cancro (Milan, Italy), and a fellowship from Fondazione Italiana per la Ricerca sul Cancro (Milan, Italy; P. Dalerba).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Ivano Arioli and Claudio Stucchi for expert technical help with in vivo experiments in immunodeficient mice, Elena Tamborini for useful advice on DNA extraction from paraffin-embedded tissues, Silvia Bacchetti (McMaster University, Hamilton, Ontario, Canada) and Michael Clarke (University of Michigan, Ann Arbor, MI) for helpful discussion and critical reading of the manuscript, and Grazia Barp for her skillful help in editing the manuscript.


   Footnotes
 
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Received 10/13/04. Revised 11/23/04. Accepted 12/15/04.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Leibovitz A, Stinson JC, McCombs WB III, McCoy CE, Mazur KC, Mabry ND. Classification of human colorectal adenocarcinoma cell lines. Cancer Res 1976;36:4562–9.[Abstract/Free Full Text]
  2. Brattain MG, Brattain DE, Fine WD, et al. Initiation and characterization of cultures of human colonic carcinoma with different biological characteristics utilizing feeder layers of confluent fibroblasts. Tumour Biol 1981;2:355–66.
  3. Park JG, Oie HK, Sugarbaker PH, et al. Characteristics of cell lines established from human colorectal carcinoma. Cancer Res 1987;47:6710–8.
  4. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100:57–70.[CrossRef][Medline]
  5. Shay JW, Bacchetti S. A survey of telomerase activity in human cancer. Eur J Cancer 1997;33:787–91.
  6. McEachern MJ, Krauskopf A, Blackburn EH. Telomeres and their control. Annu Rev Genet 2000;34:331–58.[CrossRef][Medline]
  7. 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;3:1271–4.[CrossRef][Medline]
  8. Reddel RR. Alternative lengthening of telomeres, telomerase, and cancer. Cancer Lett 2003;194:155–62.[CrossRef][Medline]
  9. Ebert T, Bander NH, Finstad CL, Ramsawak RD, Old LJ. Establishment and characterization of human renal cancer and normal kidney cell lines. Cancer Res 1990;50:5531–6.[Abstract/Free Full Text]
  10. Kim NW, Piatyszek MA, Prowse KR, et al. Specific association of human telomerase activity with immortal cells and cancer. Science 1994;266:2011–5.[Abstract/Free Full Text]
  11. Chadeneau C, Hay K, Hirte HW, Gallinger S, Bacchetti S. Telomerase activity associated with acquisition of malignancy in human colorectal cancer. Cancer Res 1995;55:2533–6.[Abstract/Free Full Text]
  12. Yan P, Saraga EP, Bouzourene H, Bosman FT, Benhattar J. Expression of telomerase genes correlates with telomerase activity in human colorectal carcinogenesis. J Pathol 2001;193:21–6.[CrossRef][Medline]
  13. Guiducci C, Anglana M, Wang A, Bacchetti S. Transient expression of wild-type or biologically inactive telomerase allows the formation of artificial telomeres in mortal human cells. Exp Cell Res 2001;265:304–11.[CrossRef][Medline]
  14. Bryan TM, Englezou A, Gupta J, Bacchetti S, Reddel RR. Telomere elongation in immortal human cells without detectable telomerase activity. EMBO J 1995;14:4240–8.[Medline]
  15. Yan P, Benhattar J, Seelentag W, Stehle JC, Bosman FT. Immunohistochemical localization of hTERT protein in human tissues. Histochem Cell Biol 2004;121:391–7.[CrossRef][Medline]
  16. Frattini M, Balestra D, Suardi S, et al. Different genetic features associated with colon and rectal carcinogenesis. Clin Cancer Res 2004;10:4015–21.[Abstract/Free Full Text]
  17. United Kingdom Co-ordinating Committee on Cancer Research, United Kingdom Co-ordinating Committee on Cancer Research (UKCCCR) guidelines for the welfare of animals in experimental neoplasia (second edition). Br J Cancer 1998;77:1–10.
  18. Kyo S, Inoue M. Complex regulatory mechanisms of telomerase activity in normal and cancer cells: how can we apply them for cancer therapy? Oncogene 2002;21:688–97.[CrossRef][Medline]
  19. Roth A, Yssel H, Pene J, et al. Telomerase levels control the lifespan of human T lymphocytes. Blood 2003;102:849–57.[Abstract/Free Full Text]
  20. DeYoung BR, Wick MR. Immunohistologic evaluation of metastatic carcinomas of unknown origin: an algorithmic approach. Semin Diagn Pathol 2000;17:184–93.[Medline]
  21. Picard O, Rolland Y, Poupon MF. Fibroblast-dependent tumorigenicity of cells in nude mice: implication for implantation of metastases. Cancer Res 1986;46:3290–4.[Abstract/Free Full Text]
  22. Noel A, De Pauw-Gillet MC, Purnell G, Nusgens B, Lapiere CM, Foidart JM. Enhancement of tumorigenicity of human breast adenocarcinoma cells in nude mice by matrigel and fibroblasts. Br J Cancer 1993;68:909–15.[Medline]
  23. Brattain MG, Levine AE, Chakrabarty S, Yeoman LC, Willson JK, Long B. Heterogeneity of human colon carcinoma. Cancer Metastasis Rev 1984;3:177–91.[CrossRef][Medline]
  24. Fidler IJ. Orthotopic implantation of human colon carcinomas into nude mice provides a valuable model for the biology and therapy of metastasis. Cancer Metastasis Rev 1991;10:229–43.[CrossRef][Medline]
  25. Zhang A, Zheng C, Lindvall C, et al. Frequent amplification of the telomerase reverse transcriptase gene in human tumors. Cancer Res 2000;60:6230–5.[Abstract/Free Full Text]
  26. Plentz RR, Wiemann SU, Flemming P, et al. Telomere shortening of epithelial cells characterises the adenoma-carcinoma transition of human colorectal cancer. Gut 2003;52:1304–7.[Abstract/Free Full Text]
  27. Yasunaga Y, Nakamura K, Ewing CM, Isaacs WB, Hukku B, Rhim JS. A novel human cell culture model for the study of familial prostate cancer. Cancer Res 2001;61:5969–73.[Abstract/Free Full Text]
  28. Yasunaga Y, Nakamura K, Ko D, et al. A novel human cancer culture model for the study of prostate cancer. Oncogene 2001;20:8036–41.[CrossRef][Medline]
  29. Giard DJ, Aaronson SA, Todaro GJ, et al. In vitro cultivation of human tumors: establishment of cell lines derived from a series of solid tumors. J Natl Cancer Inst 1973;51:1417–23.
  30. Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem cells, cancer, and cancer stem cells. Nature 2001;414:105–11.[CrossRef][Medline]
  31. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 2003;100:3983–8.[Abstract/Free Full Text]
  32. Dimri GP, Martinez JL, Jacobs JJ, et al. The Bmi-1 oncogene induces telomerase activity and immortalizes human mammary epithelial cells. Cancer Res 2002;62:4736–45.[Abstract/Free Full Text]
  33. Dalerba P, Maccalli C, Casati C, Castelli C, Parmiani G. Immunology and immunotherapy of colorectal cancer. Crit Rev Oncol Hematol 2003;46:33–57.[Medline]
  34. Parmiani G, Castelli C, Dalerba P, et al. Cancer immunotherapy with peptide-based vaccines: what have we achieved? Where are we going? J Natl Cancer Inst U S A 2002;94:805–18.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Proc. Natl. Acad. Sci. USAHome page
P. Dalerba, S. J. Dylla, I.-K. Park, R. Liu, X. Wang, R. W. Cho, T. Hoey, A. Gurney, E. H. Huang, D. M. Simeone, et al.
Phenotypic characterization of human colorectal cancer stem cells
PNAS, June 12, 2007; 104(24): 10158 - 10163.
[Abstract] [Full Text] [PDF]

Home page
 J. Cell Sci.Home page
Y.-L. Wu, C. Dudognon, E. Nguyen, J. Hillion, F. Pendino, I. Tarkanyi, J. Aradi, M. Lanotte, J.-H. Tong, G.-Q. Chen, et al.
Immunodetection of human telomerase reverse-transcriptase (hTERT) re-appraised: nucleolin and telomerase cross paths
J. Cell Sci., July 1, 2006; 119(13): 2797 - 2806.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dalerba, P.
Right arrow Articles by Castelli, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dalerba, P.
Right arrow Articles by Castelli, C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online