Expression of telomerase in normal and malignant rat hepatic ... - Nature

5 downloads 55 Views 503KB Size Report
Vita M Golubovskaya1,3, Sharon C Presnell1, Michelle J Hooth1,2, Gary J Smith1,2 and. William K ..... cells (Tsao et al., 1984; Grisham et al., 1993; Coleman.
Oncogene (1997) 15, 1233 ± 1240  1997 Stockton Press All rights reserved 0950 ± 9232/97 $12.00

Expression of telomerase in normal and malignant rat hepatic epithelia Vita M Golubovskaya1,3, Sharon C Presnell1, Michelle J Hooth1,2, Gary J Smith1,2 and William K Kaufmann1,2,3 1

Department of Pathology and Laboratory Medicine, 2Curriculum in Toxicology, and 3Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-7295, USA

Telomerase is a ribonucleoprotein that synthesizes telomeric DNA repeats onto the ends of chromosomes. More than 85% of human cancers express telomerase activity and a large proportion of human hepatocellular carcinomas are positive. To study the role of telomerase expression in rat hepatocarcinogenesis, telomerase activity was assayed in various rat tissues and in two types of liver epithelial cells: hepatocytes and hepatic epithelial stem-like cells. In the present study, we demonstrate that telomerase activity in rats is tissuespeci®c and stable with animal aging. Liver and testis were found to be telomerase positive, spleen had low or no activity, and kidney was negative. Telomerase activity did not change signi®cantly in 18 month-old rats compared to 2 month-old rats, but was moderately (twofold) increased during liver regeneration induced by a 2/3's partial hepatectomy. Telomerase activity was detected in isolated rat hepatocytes and low passage hepatic epithelial stem-like cells (WB-F344). Telomerase activity displayed signi®cant variations in a propagable clone of WB-F344 cells. At low passage levels after establishment in vitro (passages 4 ± 9) non-tumorigenic WB-F344 cells expressed telomerase activity. During further in vitro passaging these cells lost expression of telomerase. Expression of telomerase in the tumorderived lines of WB-F344 cells but not in the selectively cycled, parental lineages of these cells suggests that there may be a role for telomerase in hepatocarcinogenesis. Keywords: telomerase; rat liver; hepatocyte; stem cell; hepatocarcinogenesis

Introduction Eukaryotic chromosomes are capped with telomeres, structures consisting of highly conserved repeated DNA sequences complexed with specialized proteins (Zakian, 1995). In humans and all other vertebrates, telomeric DNA consists of (TTAGGG)n sequences (Blackburn, 1991). Telomerase is a ribonucleoprotein DNA polymerase that maintains telomeric sequences at chromosome ends (Greider and Blackburn, 1985). The RNA component of telomerase provides the template for de novo telomere repeat synthesis. The RNA component has been cloned and characterized in ciliates, yeast, humans and mice (Blackburn, 1992; Zakian, 1995; Feng et al., 1995; Blasco et al., 1995;

Correspondence: W Kaufmann, CB#7295, Rm 351, Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599 ± 7295 Received 3 March 1997; revised 13 May 1997; accepted 13 May 1997

Singer and Gollschling, 1994; Gilley and Blackburn, 1996). Telomeres protect chromosomes from DNA degradation, end-to-end fusions, and rearrangements (McClintock, 1941; Blackburn, 1991; Zakian, 1995). Thus degradation of telomeres may generate unstable chromosomal termini (McClintock, 1941; Counter et al., 1992). In many types of normal human somatic tissues telomerase activity is low or undetectable (Kim et al., 1994; Broccolli et al., 1995), although, telomerase activity is detectable in stem cells of proliferative tissue compartments (Tahara et al., 1995; Counter et al., 1995; Harle-Bachor and Boukamp, 1996). Telomerase-negative somatic cells of humans lose about 50 ± 200 nucleotides of terminal telomeric DNA per cell division due to the `end-replication problem' (Harley et al., 1990). DNA polymerase a cannot replicate the 5' end of the lagging strand of a linear DNA molecule (Olovnikov, 1971). Telomere shortening during cell division has been proposed to function as a `mitotic clock', limiting cellular proliferative lifespan (Harley et al., 1990). Normal somatic cells with shortened telomeres and active cell cycle checkpoint control systems cease to divide and undergo senescence (Counter et al., 1992; Kaufmann and Paules, 1996). In contrast, germline cells, some stem cells, most human cancers, and most tumor cell lines express telomerase activity (Kim et al., 1994). Cancer cells and immortalized cell lines expressing telomerase have either shortened but stable telomeres or even elongated telomeres (in some immortalized cell lines) (Healy, 1995; Bacchetti and Counter, 1995). Consequently, telomerase expession has been proposed to be required for cellular immortalization (Harley et al., 1994). However, other mechanisms of telomere maintenance are also possible (Bryan et al., 1995; Strahl and Blackburn, 1996), such as recombinational events as seen in yeast (Wang and Zakian, 1990) or transposition of retrotransposons as found in Drosophila (Beissmann and Mason, 1992; Levis et al., 1993). Alterations in the regulation of telomerase expression may play a critical role in the immortalization stage of carcinogenesis. Liver carcinogenesis in the rat has been widely studied. Hepatocellular carcinoma is a common human cancer worldwide, and animal models of human disease allow investigations of mechanisms of pathogenesis. Moreover, one-third to one-half of all chemicals that are carcinogenic in rodents cause liver tumors (Hu€ et al., 1991). Adult liver is a stable, quiescent tissue, but it responds to partial hepatectomy or hepatocyte necrosis with increased cell proliferation. Because of its remarkable regenerative capacity the rat liver has proven to be an excellent model for studies of cell proliferation and carcinogenesis (Cayama et al., 1978; Kaufmann et al., 1981). In spite of an extensive

HeLa

Kidney

Spleen

Testis

Liver

18 mo Kidney

Spleen

2 mo Testis

literature on hepatocarcinogenesis, there are few reports on telomeres and telomerase activity in normal rat liver and liver tumors (Makarov et al., 1993; Yoshimi et al., 1996). To determine whether altered expression of telomerase might play a role in rat hepatocarcinogenesis, we characterized telomerase activity in quiescent and regenerating liver, isolated hepatocytes, cultured hepatic epithelial stem-like cells from normal liver, and in immortal and tumorigenic derivatives of these cells. The results suggest that expression of telomerase may be associated with tumorigenesis of hepatic epithelia.

Liver

1234

Lysis Buffer

Telomerase in hepatic epithelia VM Golubovskaya et al

Results Telomerase activity in di€erent rat tissues A PCR-based protocol, TRAP (Kim et al., 1994), was used to assay for telomerase activity in rat tissue extracts. Telomerase activity was compared in extracts prepared from fresh tissues and from tissues frozen at 7708C. No inhibition of telomerase activity was observed in frozen samples (data not shown). Rat liver and testis expressed high levels of telomerase activity, spleen had weak telomerase activity (found in one of three samples) and kidney was telomerase negative in all animals at 2 months of age (Figure 1). Telomerase-positive extracts produced 6 bp ladders, and pretreatment of the extract with RNase inhibited formation of telomeric repeats (not shown). Lysis bu€er without any cell extract was used as an additional negative control in each reaction. The telomerase-negative samples of spleen and kidney were mixed with a positive HeLa sample. No inhibition of HeLa telomerase activity by the negative samples was found (data not shown). Similarly, telomerase was expressed in liver and testis, but not in kidney, of 18 month-old rats. In the aged rats spleen was again weakly positive in one of three samples (Figure 1). The variation in telomerase activity in spleen may be due to variation in immune stimulation among animals. Regulation of telomerase during liver regeneration The time-course of liver regeneration after a 2/3's partial hepatectomy permits distinction between a prereplicative G1 phase, lasting 12 h after hepatectomy, followed by replicative and mitotic phases (S/G2/M) during the next 12 h. Under the conditions used, an average of 15% of hepatocytes were in S phase and 4% were in mitosis twenty-four hours after the hepatectomy (Kaufmann et al., 1981). Liver that was resected in the partial hepatectomy (0 h) consisted of quiescent hepatocytes. Extracts from rat livers, 0, 12 and 24 h after a 2/3's partial hepatectomy were used for TRAP assay (Figure 2a). Telomerase activity was high in quiescent liver, and nearly doubled in regenerative liver. The 2.0-fold increase at 24 h was signi®cantly di€erent from the 0 h control (P50.05, Figure 2b). Isolated hepatocytes express telomerase activity Although liver expressed telomerase activity which responded to a proliferative stimulus, it was unclear whether mature parenchymal hepatocytes expressed the activity. Therefore, rat hepatocytes were isolated using a two-step collagenase perfusion method (Kaufmann et

Figure 1 Telomerase activity in young and old rats. Di€erent tissues from 2-month-old and 18-month-old F344 rats were extracted for TRAP assay (see Materials and methods) with 6 mg of protein being used for detection of teleomerase activity. HeLa cells were used as a telomerase positive control. Lysis bu€er without any protein was used as a negative control. Tissues from three di€erent rats in each age group were analysed. Liver and testis were telomerase positive and kidney was telomerase negative in all tested rats. Spleen was telomerase positive in one of three rats in both age groups. A negative spleen sample is shown for the young age group, and a weakly positive spleen sample is shown for the older age group. Di€erences in the levels of telomerase activity in liver and testis between 2 and 18 month-old rats were due to variations in these two particular animals. These di€erences were not reproduced in the other samples

al., 1986). Isolated rat hepatocytes expressed telomerase activity, that was inactivated with RNase (Figure 3). Activity of telomerase was also seen in hepatocytes that were puri®ed through Percoll to eliminate nonparenchymal cells (Kreamer et al., 1986) (not shown). Additionally, two immortal hepatocyte lines (6/15 hepatocellular carcinoma line, and a cloned 6/27 C1 line), derived from F344 rats also expressed telomerase (Figure 3). Telomerase activity in normal rat hepatic epithelial stemlike cells (WB-F344) and in tumorigenic WB-F344 cell lines Telomerase activity was assayed in the rat hepatic epithelial stem-like cell line, WB-F344 (Tsao et al.,



+



+

Normal Fibroblasts

+

MDAH041



Lysis Buffer

RNase

6/15 HCA

24

6/27 C1

12

Hepatocytes

0

1984), at di€erent passage levels after establishment (one passage was equal to four population doublings) (Figure 4). Cells at low passages (passage 4 ± 9) expressed high levels of telomerase, whereas, cells at passages 19 and 25 expressed telomerase at a very low level or were telomerase-negative (P50.05). Cells with low telomerase activity did not contain TRAP inhibitors, as determined by mixing with telomerasepositive samples (not shown). The TRAP assay was also performed on tumorigenic and non-tumorigenic cell lineages derived from WB-F344 cells by a selective protocol (Lee et al., 1989), and on clonally isolated cell lines derived from tumors arising after transplantation of the tumorigenic lineages into syngeneic animals (see Materials and methods). Both the non-tumorigenic

Liver

Hours Post Hepatectomy

Lysis Buffer

Regenerative Liver

MDAH041

a

MDAH041:10

Telomerase in hepatic epithelia VM Golubovskaya et al









b

Figure 2 Telomerase activity in regenerating liver. (a) Liver samples taken at 0, 12 and 24 h after a 2/3's partial hepatectomy were extracted for TRAP assay. For quantitation purposes, 0.6 mg of protein was used for assay of telomerase (see Materials and methods). MDAH041 was used as a positive control in TRAP reactions (6 mg and 0.6 mg of protein). The image was edited from the same X-ray ®lm. (b) Quantitation of telomerase activity in regenerating rat liver. The data shown on a were analysed with a PhosphorImager. Telomerase activity was expressed as a relative speci®c activity, normalized to MDAH041, which was designated as 100%. The values shown are means plus standard deviations (n=3). Telomerase activity increased 1.5-fold at 12 h (Student's t

Figure 3 Telomerase activity in isolated rat hepatocytes and immortal hepatocyte lines. Hepatocytes were isolated from rat liver by collagenase perfusion, then extracted, and 6 mg of protein was used for TRAP assay. Telomerase assay was done with (+) or without RNase (7) pretreatment. Lysis bu€er and normal human ®broblasts were used as a negative control. MDAH041 was used as a positive control. 6/27 C1 is a cloned immortal hepatocyte line. The data for 6/15 HCA, an immortal rat hepatocellular carcinoma line, was from a separate gel

test, one-sided P=0.041) and 2.0-fold at 24 h post partial hepatectomy (Student's t test, one-sided P=0.024) versus 0 h control

1235

Telomerase in hepatic epithelia VM Golubovskaya et al

4

9

19

25

Tumorigenicity –



WB: L10 L18 P

+

+

L7

L20 T

P

Lysis Buffer

P:

MDAH041

RKO

WB Cells

Lysis Buffer

1236

T

Figure 4 Telomerase activity in rat hepatic epithelial stem-like cells, WB-F344. WB-F344 cells were assayed for telomerase activity at passages 4, 9, 19 and 25. MDAH041 was used as a positive control with 1/10 and 1/50 dilutions

(L10 and L18) and tumorigenic (L7-P and L20-P) cell lineages were telomerase-negative, while the tumor-derived lines (L7-T and L20-T) expressed telomerase activity (Figure 5). The parental lineages, from which three other tumor lines (L2-P, L6-P and L14-P) were derived, also did not express telomerase. The tumor-derived lines (L2-T, L6-T and L14-T) from these lineages expressed telomerase activity (not shown). Levels of telomerase activity in normal rat hepatic epithelial stem-like cells (WB-F344) and in various WB-F344 cell lines were quanti®ed relative to the same positive cell extract, MDAH041 (Figure 6). Telomerase activity in normal WB-F344 cells at passages 19 and 25 was 51% of that seen at passage 4 (Figure 6, left columns). Telomerase activity in the non-tumorigenic L10 and L18 lineages was 5 1% of the positive control. In four tumor-derived lines, telomerase activity ranged from 6 ± 101% of the positive control (Figure 6, right columns). Three of 5 tumor-derived lines (L2, L6 and L14) expressed telomerase activity in the range expressed by low passage WB-F344 cells, while the L20 tumor-derived line expressed telomerase with high speci®c activity equivalent to that seen in MDAH041 (Figure 6, right columns). Telomerase activity in the L7 tumor-derived line was not quanti®ed relative to MDAHO41, but enzyme activity was high (Figure 5). Thus, two nontumorigenic and ®ve tumorigenic lineages were telomerase negative, while ®ve tumor-derived lines were telomerase positive (Figure 7).

Figure 5 Telomerase activity in non-tumorigenic, tumorigenic and tumor-derived WB-F344 cell lines. Non-tumorigenic cell lines, WB L10 and WB L18, and tumorigenic lines, L7 and L20, were assayed by TRAP. L10 and L18 were non-tumorigenic WB-F344 lines that were subjected a selective protocol (Materials and methods). L7 and L20 were tumorigenic WB-F344 cell lines obtained by the same protocol. The parental lineage is designated (P). Following transplantation into syngeneic hosts, tumors were isolated and cloned lines recovered (T). RKO was used as a positive control. The faint signals in telomerase-negative samples are assumed to be non-telomeric, as they did not display 6 bp ladders

Discussion More than 85% of human cancers expresss telomerase activity (Kim et al., 1994) and a similarly large proportion of human hepatocellular carcinomas are positive (Tahara et al., 1995). We have focused on the potential role of telomerase in two rat models of hepatocarcinogenesis. In the present study we found high telomerase activity in rat testis and liver, weak activity in spleen, and no activity in kidney. Telomerase activity did not change in old animals compared to young animals, but telomerase activity was moderately increased in regenerating rat livers. Telomerase activity also was detected in isolated rat hepatocytes, immortal hepatocyte lines and hepatic epithelial stem-like cells.

Telomerase in hepatic epithelia VM Golubovskaya et al

1237

Figure 6 Quantitation of telomerase activity in WB-F344 cells, in non tumorigenic and tumor-derived WB-F344 cell lines. Telomerase activity was expressed as a relative speci®c activity, normalized to MDAH041, which was designated as 100%. The data on telomerase activity in normal WB-F344 cells, shown on Figure 4, were quanti®ed with a PhosphorImager. Telomerase activity in normal WB-F344 cells decreased signi®cantly from passage 4 to pasages 9, 19 and 25 (Student's t test, P50.05, n=3 to 5). Non-tumorigenic cell lines, WB L10 and WB L18, shown on Figure 5 were analysed. For the tumorigenic lineages (L2, L6, L14 and L20), tumor-derived lines were analysed. The data on telomerase activity in non-tumorigenic and tumor-derived WBF344 cell lines were quanti®ed relative to the MDAH041 control. Values shown are means plus standard deviations (n=2 to 6). The L7 tumor-derived line was not plotted, as it was not quanti®ed to the MDAH041 control

Epithelial stem-like cells lost telomerase activity during repeated passaging in vitro. Consequently, expression of the enzyme in tumor-derived lines of these cells suggests a role for telomerase in rat hepatocarcinogenesis. We demonstrate here that telomerase activity in rats is tissue-speci®c and stable with animal aging. These data correspond well to those obtained in murine models (Prowse and Greider, 1995; Chadeneau et al., 1995; Broccoli et al., 1996). Mouse cells have longer telomeres (20 ± 50 kb range) compared to human somatic cells (5 ± 10 kb of TTAGGG repeats) (Prowse and Greider, 1995; Broccoli et al., 1996). High enzyme activity was observed in mouse (Mus spretus) testis and liver, with weaker activity being seen in kidney and spleen. The presence of telomerase activity in M. spretus liver and testis correlated with a longer telomere length in these tissues (Prowse and Greider, 1995). High telomerase activity in adult liver and testis was correlated well with expression of mouse telomerase RNA, which was present in testis, liver and spleen but not in kidney (Blasco et al., 1995). These results suggest that rats and mice di€er from humans in the regulation of telomerase activity (Prowse and Greider, 1995) in that rodent liver is positive and human liver is negative (Tahara et al., 1995). Telomerase was expressed in quiescent hepatocytes and intact liver and activity was moderately increased (1.5 ± 2.0-fold) in regenerating rat liver. The function of telomerase in rat tissues has not been explored yet, but its enhanced activity in regenerating liver may be linked to cellular

Figure 7 Diagrammatic summary of telomerase expression in non-tumorigenic, tumorigenic and tumor-derived WB-F344 cell lines. Individual lineages of WB-F344 were subjected to 8 ± 10 cycles of selective growth at con¯uence to generate independent, spontaneous transformants of WB-F344 cells (Lee et al., 1989, Materials and methods). Acquisition of tumorigenicity was assessed by transplantation of selectively cycled WB-F344 lineages into animals. Telomerase assay was done on nontumorigenic lineages (L10 and L18), tumorigenic lineages L2, L6, L7, L14 and L20 and on tumor-derived lines obtained after transplantation of the tumorigenic lineages into syngeneic animals. Telomerase-negative lines are designated (7), telomerase-positive lines (+), highly telomerase-positive lines (++)

proliferation. These data are consistent with high telomerase activity in proliferative-phase human endometrium (Kyo et al., 1997). As we were focused on characterization of liver cells and it was unclear what type of liver cell expressed telomerase, we assayed for telomerase activity in two types of hepatic epithelial cells, isolated hepatocytes and the WB-F344 line of hepatic epithelial stem-like cells (Tsao et al., 1984; Grisham et al., 1993; Coleman et al., 1993). We showed that isolated hepatocytes and low passage WB-F344 were telomerase-positive. WBF344 cells at low passage levels are non-tumorigenic in syngeneic hosts but can be transformed to tumorigenicity by several methods (Lee et al., 1989). After repetitive cycles of growth arrest and cell proliferation, tumors that developed from ®ve independently transformed lineages were isolated, and clonally derived lines were established for examination in this study. Expression of telomerase in the tumor-derived lines but not in the selectively cycled, parental lineages of WB-F344 cells suggest that there may be a role for expression of telomerase in tumor outgrowth in vivo. It is possible that during in vitro selection a subfraction of telomerase-positive cells persists within the largely negative population, and it is these persistently expressing cells that lose restriction-point control and transform to tumorigenicity (Greaves, 1996). Alternatively, clonal lineages that have repressed telomerase activity during in vitro propagation may reactivate

Telomerase in hepatic epithelia VM Golubovskaya et al

1238

expression coordinate with or consequent to tumor outgrowth in vivo (Shay and Wright, 1996). The WB-F344 hepatic epithelial cell line has stemcell-like characteristics including the ability to differentiate into hapatocytes in vivo (Grisham et al., 1993; Coleman et al., 1993). Telomerase activity varied considerably in this line, from measurable levels of expression in low passage cultures, to undetectable levels in high passage cultures, and to very high levels in some tumor-derived lines. This pattern of variable expression resembled telomerase biology in humans: telomerase-positive stem cells, loss or repression of telomerase activity in adult, replication-competent, somatic cells, and expression of telomerase activity in cancer cells. It is now necessary to extend these studies to human liver cells and to study enzyme activity in di€erent types of human liver cells. Telomerase activity was shown to be positive in 85% of human hepatocellular carcinoma tissues (Tahara et al., 1995). Interestingly, while 5% of nontumorous control livers was shown to express weak telomerase activity, 50% of hepatitis specimens and 75% of cirrhotic livers from cancer-free patients were found to express telomerase activity (Tahara et al., 1995). Although it is likely that lymphocyte migration in in¯amed hepatic parenchyma will contribute to a positive signal, it will be of interest to determine whether expression in cirrhotic livers re¯ects activation of a proliferative hepatic epithelial stem cell or the generation of telomerase-positive immortal hepatocyes in dysplastic pre-malignant lesions. This hepatic epithelial cell model may be useful for further study or cell aging and immortalization, regulation of telomerase in vitro and mechanisms of carcinogenesis.

in a humidi®ed atmosphere of 5% CO2 at 378C. Phenobarbital (2 mM) was added to the medium of the phenobarbital-dependent 6/15 HCA and 6/27 C1 immortal rat hepatocytes, as described previously (Chiao et al., 1995). WB-F344 cells were passaged every week by 1 : 12 split ratio. With a 60% plating eciency, each passage was estimated to be approximately equal to four population doublings. Spontaneously transformed derivatives of WBF344 cells were obtained following multiple cycles of selection at con¯uent cell density in Richter's improved minimal essential medium supplemented as described (Lee et al., 1989). Each cycle of selection growth consisted of one week for population expansion to con¯uence followed by three weeks maintenance at con¯uent cell density. Two non-tumorigenic cell lineages, WB-L10C10 and WBL18C10, and ®ve tumorigenic cell lineages, WB-L2C10, WB-L6C10, WB-L7C10, WB-L14C8, and WB-L20C10, were used in this study. Lineages determined to be tumorigenic at selection cycle 8 or 10 were transplanted into animals. Tumor-derived clones of WB-F344 cells were obtained from these lineages after subsequent tumor outgrowth in vivo. The nomenclature for the spontaneously transformed derivatives of WB-F344 cells incorporates the name of the parental cell line, the lineage identi®cation number and the selection cycle number. WB-L2C10, for example, corresponds to cell lineage number 2 that has undergone 10 cycles of selective growth. (Hooth MJ, Coleman WB, Grisham JW and Smith GJ. submitted, 1997). In this report, we have used an abbreviated nomenclature, L10, L18, L2, L6, L7, L14, and L20 for these WB-F344 cell lineages. Cultured cells were harvested when in logarithmic growth phase and frozen before cell extraction for assay of telomerase. Tissue and cell extracts

Adult male Fischer 344 rats obtained from the Charles River Breeding Laboratories (Raleigh, NC) were used for studies involving hepatocyte isolation and partial hepatectomy. Rats were subjected to a two-thirds partial hepatectomy at 7 weeks of age, and livers at 0, 12 and 24 h after hepatectomy were used for preparation of cell extracts. Age-controlled male Fischer 344 rats were obtained from the Aged Rat Colony of the National Institute on Aging (Harlan Sprague-Dawley, Indianapolis, IN). Animals at 2 and 18 months of age were used for the harvest of various tissues (liver, spleen, kidney, testis). Cell extracts for telomerase assays were prepared from fresh or frozen (7708C) rat tissues.

Selected rat tissues were rinsed with ice-cold phosphatebu€ered saline, followed by cold 26 Hypo bu€er (16Hypo bu€er contained: 10 mM Tris-HCl pH 8.5, 3 mM KCl, 1 mM MgCl, 1 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl ¯uoride, 1 mM Leupeptin, 10 mM Pepstatin and 10 mM RNasin) (Counter et al., 1992). Tissue fragments were homogenized in one volume of icecold 16Hypo bu€er, incubated on ice for 15 min, homogenized further with ®ve passages through a 5/8' 25 gauge needle, and incubated on ice for an additional 15 min. Samples were centrifuged at 16 0006g for 20 min at 48C. This step of centrifugation was repeated in some experiments to clarify the extract. Glycerol was added to the extract supernatants to a ®nal concentration of 11%. Cell extracts were kept frozen at 7708C prior to assay of telomerase. To prepare cell extracts from cultivated cells, the same procedure was used but cells were homogenized in 16Hypo bu€er at 1 ± 56107 cells/ml. The concentration of protein was measured for each extract with the bicinchroninic acid protein assay kit (Pierce Chemical Co).

Isolation of hepatocytes

Telomerase assay

Livers were perfused through the vena cava with 0.1% collagenase to dissociate and isolate hepatocytes (Kaufmann et al., 1986). Further puri®cation of viable hepatocyte preparations was done by sedimentation through Percoll (Kreamer et al., 1986).

Telomerase activity was detected by the PCR-mediated, telomeric-repeat ampli®cation protocol (TRAP) described by Kim et al., 1994. The oligonucleotides, TS: 5'AATCCGTCGAGCAGAGTT-3', and CX: 5'-CCCTTACCCTTACCCTTACCCTAA-3' were used as primers. The initial telomerase extension reaction and the secondary PCR reaction were performed in di€erent tubes. A phenolchloroform extraction of the telomerase-extended product was used to eliminate potential inhibitors of the secondary PCR reaction (Bryan et al., 1995). The telomerase extension reaction was performed at room temperature for 30 min. To eliminate proteins, TE bu€er (10 mM TrisHCl, pH 7.5, 1 mM EDTA) with proteinase K at 1 mg/ml

Materials and methods Animals

Cell culture Immortal hepatocyte lines, 6/15 HCA (hepatocellular carcinoma) and 6/27 C1, described in (Chiao et al., 1995; Zhang et al., 1995), and the normal hepatic epithelial stemlike cell line, WB-F344 (Tsao et al., 1984), were cultivated in minimal essential medium with 10% fetal bovine serum

Telomerase in hepatic epithelia VM Golubovskaya et al

and 0.5% sodim dodecyl sulfate was added. Samples were then incubated 15 min at 378C. Following extraction with phenol-chloroform and precipitation with ethanol the telomerase-extended product was dissolved in TE bu€er. Telomeric DNA product (2 ml) was added to 18 ml of PCRmix, containing 16Taq polymerase bu€er (Gibco, BRL), 1.5 mM MgCl2, 0.2 mM each of four dNTP's, 0.1 mg TS and 0.1 mg CX primers, 0.8 units of Taq polymerase and 1 ± 2 mCi [a-32P]dCTP. PCR was performed with an initial denaturation step at 908C for 2 min, followed by 31 cycles of: 948C for 30 s, 508C for 30 s and 728C for 1 min. A ®nal extension step of 5 min at 728C completed the TRAP reaction. A cell extract with known high telomerase activity, HeLa, RKO or MDAH041 (an immortal Li ± Fraumeni syndrome ®broblast line), was used as a positive control in each experiment. Lysis bu€er without cell extract and RNase treated samples were used as a negative controls. For RNase treatment, cell extract with 6 mg protein was incubated with 0.5 mg RNase before the telomerase extension reaction. Negative samples without telomerase activity were mixed with a positive sample (9 : 1 ratio) and incubated for 10 min at room temperature before reaction to check for inhibitory activities. Di€erent dilutions of protein (1/10, 1/50) were also tested in the TRAP assay to determine that telomerase activity was proportional to the protein concentration in reaction mixtures. Electrophoresis and phosphorImager quantitation of telomerase activity Telomerase-extended 6 bp ladder products were analysed on an 8% denaturing polyacrylamide gel in 16Tris-borate-

EDTA bu€er. The gel was exposed to X-ray ®lm at 7708C or dried and used for quantitation with a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). In the various experiments, 6 mg of protein was used for detection of telomerase activity and 0.6 mg of protein was used for quantitation, as this lower amount was found to be in the linear range for telomerase activity versus protein concentration (Broccoli et al., 1995; 1996). The total PhosphorImage units (pixels) for each lane were measured in the central region of the gel. The mean number of pixels in control lanes (lysis bu€er reaction) were subtracted from the mean number in each lane as a background. The relative speci®c activity of each extract was determined as a percentage of the speci®c activity of telomerase-positive cell extract of MDAH041 (with high telomerase activity, considered to be 100%) as described in (Broccoli et al., 1995; 1996). Statistical signi®cance of relative telomerase activities was determined by Student's t-test.

Acknowledgements We are grateful to Drs Craig Albright, Lola Reid, Karen McCullough, William Coleman and to Laura Byrd, Cynthia Irvin and Kristin Borchert for providing the various cells and tissues used in this analysis. We wish to thank Betty Lloyd and members from Medical Illustrations and Photography Department for expert assistance in preparation of the ®gures. This work was supported by Public Health Service grants CA 59495 (WKK), CA 59486 (GJS), CA 64340 (GJS) and ES 07126 (MJH).

References Bacchetti S and Counter CM. (1995). Int. J. Oncol., 7, 423 ± 432. Biessmann H and Mason JM. (1992). Adv. Genet., 30, 185 ± 249. Blackburn EH. (1991). Nature, 350, 569 ± 573. Blackburn EH. (1992). Annu. Rev. Biochem., 61, 113 ± 129. Blasco M, Funk WD, Villeponteau B and Greider CW. (1995). Science 269, 1267 ± 1270. Broccoli D, Young JW and de Lange T. (1995). Proc. Natl. Acad. Sci. USA, 92, 9082 ± 9086. Broccoli D, Gogley LA, Donehower LA, Varmus HE and de Lange T. (1996). Mol. Cell. Biol., 16, 3765 ± 3772. Bryan TM, Englezou A, Gupta J, Bacchetti S and Redell RR. (1995). EMBO J., 14, 4240 ± 4248. Cayama E, Tsuda H, Sarma DSR and Farber E. (1978). 275, 60 ± 62. Chadeneau C, Siegel P, Harley CH, Muller WJ and Bacchetti S. (1995). Oncogene, 11, 893 ± 898. Chiao C, Zhang Y, Kaufman DG and Kaufmann WK. (1995). Am. J. Path., 146, 1248 ± 1259. Coleman WB, Wennerberg AE, Smith GJ and Grisham JW. (1993). Am. J. Pathol., 142, 1373 ± 1382. Counter CM, Avilion AA, LeFeuvre CE, Stewart NG, Greider CW, Harley CB and Bacchetti S. (1992). EMBO J., 11, 1921 ± 1929. Counter CM, Gupta J, Harley CB, Leber B and Bacchetti S. (1995). Blood, 9, 2315 ± 2320. Feng J, Funk WD, Wang SS, Weinrich SL, Avilion AA, Chiu CP, Adams RR, Chang E, Allsopp RC, Yu J, Le S, West MD, Harley CB, Andrews WH, Greider CW and Villeponteau B. (1995). Science, 269, 1236 ± 1241. Gilley D and Blackburn EH. (1996). Mol. Cell. Biol., 16, 66 ± 75. Greaves M. (1996). Trends in Genetics, 12, 127 ± 128. Greider CW and Blackburn EH. (1985). Cell, 43, 405 ± 413.

Grisham JW, Coleman WB and Smith GJ. (1993). Proc. Soc. Exp. Biol. Med., 204, 270 ± 279. Harle-Bachor C and Boukamp P. (1996). Proc. Natl. Acad. Sci. USA, 93, 6476 ± 6481. Harley CB, Futcher AB and Greider CW. (1990). Nature, 345, 458 ± 460. Harley CB, Kim NW, Prowse KR, Weinrich SL, Hirsch KS, West MD, Bacchetti S, Hirte HW, Counter CM, Greider CW, Piatyzek MA, Wright WE and Shay JW. (1994). Cold Spring Harbor Symp. Quant. Biol., LIX, 307 ± 315, Cold Spring Harbor, NY. Healy KC. (1995). Oncology Res., 7, 121 ± 130. Hu€ J, Cirvello J, Haseman J and Butcher J. (1991). Environ. Health Perspect., 93, 247 ± 270. Kaufmann WK, Kaufman DG, Rice JM and Wenk ML. (1981). Cancer Res., 41, 4653 ± 4660. Kaufmann WK, Tsao MS and Novicki DL. (1986). Carcinog., 7, 669 ± 671. Kaufmann WK and Paules RS. (1996). FASEB J., 10, 238 ± 247. Kreamer BL, Staecker JL, Sawadw N, Sattler GL, Hsia MTS, Pitot HC. (1986). In vitro 201 ± 211. Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PLC, Coviello GM, Wright WE, Weinrich SL and Shay JW. (1994). Science, 266, 2011 ± 2015. Kyo S, Takahura M, Kohama T and Inoue M. (1997). Cancer Res., 57, 610 ± 614. Lee LW, Tsao MS, Grisham JW and Smith GJ. (1989). Am. J. Pathol., 135, 63 ± 71. Levis RW, Canesan R, Houtchens K, Tolar LA and Sheen F. (1993). Cell, 75, 1083 ± 1093. Makarov VL, Lejnine S, Bedoyan J and Langmore JP. (1993). Cell, 73, 775 ± 787. McClintock B. (1941). Genetics, 26, 234 ± 282. Olovnikov AM. (1971). Doklady Biochem., 201, 394 ± 397.

1239

Telomerase in hepatic epithelia VM Golubovskaya et al

1240

Prowse K and Greider CW. (1995). Proc. Natl. Acad. Sci. USA, 201, 4818 ± 4822. Shay JW and Wright WE. (1996). Trends in Genetics, 12, 129 ± 131. Singer MS and Gollschling DE. (1994). Science, 266, 404 ± 409. Strahl C and Blackburn EH. (1996). Mol. Cell. Biol., 16, 53 ± 65. Tahara H, Nakanishi T, Kitamoto M, Nakashio R, Shay JW, Tahara E, Kajiyama G and Ide T. (1995). Cancer Res., 55, 2734 ± 2736.

Tsao MS, Smith JD, Nelson KG and Grisham JW. (1984). Exp. Cell Res., 154, 38 ± 52. Wang SS and Zakian VA. (1990). Nature, 345, 456 ± 458. Yoshimi N, Ino N, Suzui M, Hara A, Nakatani K, Sato S and Mori H. (1996). Mol. Carcinog., 16, 1 ± 5. Zakian VA. (1995). Science, 270, 1601 ± 1606. Zhang Y, Chiao C, Byrd LL, Kaufman DG and Kaufmann WK. (1995). Liver regeneration and carcinogenesis: Molecular and cellular mechanisms, Jirtle RL (ed). Academic Press: San Diego, CA, pp. 179 ± 197.