Clinical Cancer Research Bridging the Lab and the Clinic in Cancer Medicine Infection and Cancer: Biology, Therapeutics, and Prevention
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 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 Los, G.
Right arrow Articles by Howell, S. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Los, G.
Right arrow Articles by Howell, S. B.
Clinical Cancer Research Vol. 5, 1610-1618, July 1999
© 1999 American Association for Cancer Research

Advances in Brief

Quantitation of the Change in GADD153 Messenger RNA Level as a Molecular Marker of Tumor Response in Head and Neck Cancer1

Gerrit Los2, Khalid Benbatoul, Dennis P. Gately, Racine Barton, Randolph Christen, K. Thomas Robbins, Daniel Vicario, Saeeda Kirmani, Lisa A. Orloff, Robert Weisman and Stephen B. Howell

UCSD Cancer Center [G. L., K. B., D. P. G., R. B., R. C., D. V., S. K., S. B. H.] and the Division of Otolaryngology [G. L., L. A. O., R. W.], University of California, San Diego, La Jolla, California 92093-0058, and Division of Otolaryngology/Head and Neck Surgery University of Tennessee, Memphis, Tennessee 38163 [K. T. R.]


   ABSTRACT
Top
 ABSTRACT
Introduction
 Materials and Methods
 Results
 Discussion
 REFERENCES
 
Cells injured by exposure to cisplatin (cDDP) undergo a cellular injury response that shares characteristics with responses produced by many other injurious agents. We sought to determine whether the increase of the message of the "growth arrest and DNA damage-inducible" gene, GADD153, could be used to assess the extent of the cellular injury response in model systems and in patients with head and neck cancer after treatment with cDDP. The mRNA levels of GADD153, a gene highly transcriptionally activated by cDDP damage, were increased in a transient, concentration-dependent manner by cDDP when human UMSCC10b head and neck carcinoma cells were treated with cDDP both in vitro and when grown as tumor xenografts in nude mice. There was a good correlation between the change in level of GADD153 mRNA and UMSCC10b cell kill by cDDP in vitro (r = 0.98). The magnitude of the increase was proportionally reduced in UMSCC10b sublines that were 3- or 6-fold resistant to cDDP. GADD153 mRNA levels were measured in biopsies obtained before and 24 h after treatment with cDDP from 32 patients with stage III/IV head and neck cancer. There was a relationship between the increase in GADD153 mRNA levels and the response rate. Seven of the 32 patients had no response and no increase in GADD153 mRNA level. Among the eight patients who attained a partial response, the increase in GADD153 message ranged from 0.7–2.5-fold. In contrast, 17 of 32 patients had a complete response, and this was accompanied by a 2–9-fold induction of GADD153. The mean increase in the complete responders (3.8 ± 2.2-fold) differed significantly from that for the partial responders (1.6 ± 0.9) and nonresponders (0.8 ± 0.5; P <0.05); the difference between the partial responders and nonresponders was also significant (P <0.05). An increase of GADD153 mRNA of 1.75-fold or higher predicted a complete response, with a sensitivity of 94% and a specificity of 87%. We conclude that the magnitude of the increase in GADD153 mRNA is a promising candidate for service as an intermediate marker of head and neck tumor response to cDDP. The fact that the change in GADD153 mRNA reflects the actual extent of injury sustained by the tumor makes it particularly attractive as a potential marker. One strength of this approach is that it can provide a measure of the effectiveness of therapy as early as 24–48 h after the first dose of treatment.


   Introduction
Top
 ABSTRACT
 Introduction
Materials and Methods
 Results
 Discussion
 REFERENCES
 
Cells cope with injury by changing their pattern of gene expression, and some of these changes occur in common in different types of cells. Transcription of the proto-oncogenes c-Fos and c-Jun, for example, is rapidly induced by exposure to a wide variety of exogenous stimuli, including cytotoxic agents (1, 2, 3) . The expression of these "early-response genes" results in the rapid induction of many other genes, primarily regulated at the level of transcription (4 , 5) . GADD153 is one of these genes that has been implicated in the cellular response to stress. Initially isolated as a gene that is induced rapidly by alkylating agents and UV light (6) and was isolated from hamster (7) and human cells (8) , GADD153 has been found to be responsive to other forms of stress and injury (9) , including a broad spectrum of genotoxic agents and metabolic insults (10, 11, 12, 13, 14, 15) . Subsequent work has indicated that the activation of the GADD153 promoter occurred, at least in part, as a direct result of DNA damage (16) . The mechanism responsible for the activation of GADD153 expression after DNA damage is still unclear, but current data suggest that the magnitude of the increase is proportional to the extent of cellular injury with maximal GADD153 promoter activity occurring under circumstances of severe toxicity to the cell (10 ,16, 17, 18, 19) .

The accurate quantitation of the extent of tumor injury in patients as a result of treatment is often a problem. Normally, the effectiveness of chemotherapy or radiation therapy cannot be determined for at least several weeks and often longer, and this is an obstacle to improving the management of cancer patients, particularly when alternative therapeutic modalities are available. Quantitation of molecular events occurring in response to injury in the tumor in vivo after cytotoxic injury may permit more rapid assessment of the likelihood of response (18 , 20) . The fact that GADD153 is induced by DNA and cellular damage in a dose-dependent manner (17 , 18 , 20 , 21) makes the change in the level of GADD153 mRNA an interesting candidate to monitor the extent of the injury in tumors and to serve as an intermediate marker of response. We have investigated the use of GADD153 mRNA measurement for this purpose in head and neck cancer. This tumor type was selected for study because of the ease of obtaining two serial tumor biopsies and the fact that several alternative therapeutic modalities are available to those patients who respond poorly to initial chemotherapy.


   Materials and Methods
Top
 ABSTRACT
 Introduction
 Materials and Methods
Results
 Discussion
 REFERENCES
 
Cell Lines.
The UMSCC10b cell line was derived from a human head and neck squamous cell carcinoma (22) . Cells were cultured at 37°C under 5% CO2 in 150-cm2 flasks (Corning, Corning, NY) with RPMI 1640 (Mediatech Inc., Herndon, VA) containing 10% fetal bovine serum (Gemini Bioproducts Inc., Calabasas, CA), 2 mML-glutamine, and 100 units/ml penicillin G and 100 µg/ml streptomycin sulfate. Cells were subcultured after reaching confluence (~5 x 107/150 cm2) by trypsinization and replating at a density of ~107 cells/150 cm2.

Selection for cDDP3 Resistance.
UMSCC10b cells were selected for cDDP resistance by chronic exposure to cDDP, which was kindly supplied by Bristol-Myers-Squibb (Princeton, NY). The first three selections were performed at a concentration of 0.5 µM, which allowed ~107 cells to grow to confluence in a 150-cm2 flask within 1 week. For every following set of three selections, the cDDP concentration was increased by ~20% (23 , 24) .

Sensitivity to cDDP in Vitro.
The sensitivity of the UMSCC10b cell line and its cDDP-resistant variants to cDDP was determined by clonogenic assay using a 1-h cDDP exposure. A single-cell suspension was plated into 60-mm tissue culture dishes (Corning) at 300 cells/dish in fresh medium. After incubation for 24 h at 37°C, cells had attached to the plates and cDDP was added to the cells and incubated for 1 h, after which the cells were washed twice with PBS and 5 ml of fresh medium was added. Cultures were incubated for an additional 13–15 days, after which they were fixed with methanol, stained with Giemsa, and clusters of >50 cells were scored as colonies. cDDP-selected cells were cultured in drug-free media for 2 weeks before cDDP sensitivity was determined by clonogenic assay.

Xenografts.
UMSCC10b cells were injected s.c. over the shoulder of athymic (BALB/c nu/nu) female mice, 3–4 weeks of age (Harlan Sprague Dawley, Indianapolis, IN). The xenografts were allowed to grow until they reached ~1 mm3 in volume, at which time cDDP treatment was initiated.

Expression of GADD153 in Vitro.
To determine the time course of the change in the GADD153 mRNA level, UMSCC10b cells were incubated with 100 µM cDDP for 1 h, and total RNA was extracted at 0, 1, 2, 6, 12, 24, 48, 72, and 96 h after treatment. To analyze the effect of cDDP concentration on the expression of GADD153, UMSCC10b cells were incubated for 1 h with various cDDP concentrations (0, 0.1, 1, 10, 100, and 1000 µM), and RNA was extracted 24 h after cDDP exposure.

Expression of GADD153 in Vivo.
One set of nude mice bearing UMSCC10b xenografts were injected i.p. with a single dose of cDDP (15, 30, 50, or 100 mg/kg), and total RNA was isolated from the xenograft excised 24 h later. To analyze the time course of the change in the GADD153 mRNA level, mice bearing UMSCC10b xenografts were treated with a single injection of cDDP (30 mg/kg) and sacrificed at 0, 24, 48, 72, and 96 h after treatment. In both sets of experiments, the excised tumors were minced in guanidine isothiocyanate with a polytron (Biospec Products Inc., Bartlesville, OK) and stored at 4°C until all tumors were processed. The RNA was transcribed to cDNA using random hexamer primers.

Patients and Therapy.
Thirty-two patients with previously untreated stage IV head and neck cancer were treated at the UCSD Cancer Center on a Phase I/II experimental protocol. All patients were considered unresectable. All patients received intra-arterial cDDP (150–200 mg/m2) and i.v. sodium thiosulfate infusion concurrently; the details of this treatment program have been published elsewhere (25, 26, 27) . cDDP was dissolved in 400 ml of 0.9% saline and infused over 3–5 min through a microcatheter placed angiographically to selectively encompass the dominant blood supply of the targeted tumor. Simultaneous with the cDDP infusion, sodium thiosulfate was infused i.v. at a dose of 9 g/m2 dissolved in 300 ml of distilled water over 3 min, followed by 12 g/m2 dissolved in 1 liter of distilled water over 6 h by continuous infusion to prevent severe nephrotoxicity. Treatment was repeated weekly for a total of four doses. In addition, 17 of the 32 patients received radiotherapy in doses of 2 cGy/day (x 35 fractions) starting concurrently with the start of cDDP treatment. Needle aspiration or small-cutting needle biopsies were obtained just before and at 24 h after the first dose of cDDP. Tumor response was assessed at 2 months using criteria based on physical examination, repeat computed tomography/magnetic resonance imaging studies, and repeat endoscopy and biopsy.

RNA Extraction.
RNA was extracted from UMSCC10b cells using the acid guanidinium thiocyanate-phenol-chloroform extraction (28) and from UMSCC10b xenografts and human tumor biopsies in cesium chloride. UMSCC10b cells were harvested in 10 ml of denaturing solution [4 M guanidinium thiocyanate, 25 mM sodium citrate (pH 7), 0.5% sarcosyl, and 0.1 M 2-mercaptoehtanol]. After adding 1 ml of 2 M sodium acetate (pH 4), 10 ml of water-saturated phenol (pH 4), and 2 ml of chloroform isoamyl alcohol (49:1) to the denatured cells, they were mixed thoroughly, placed on ice for 15 min, and centrifuged at 2800 rpm for 20 min at 4°C. The aqueous phase was collected and mixed with 10 ml of isopropyl alcohol, placed for 1 h at -20°C, and centrifuged. The pellet was dissolved in 1.8 ml of denaturing solution and 2 ml of isopropyl alcohol, placed for 1 h at -20°C, and centrifuged. The pellet was washed with 1 ml of 80% ethanol and then centrifuged at 5000 rpm for 10 min at room temperature. The RNA pellet was dried for 20 min in a tissue culture hood and resuspended in 100 µl of diethylpyrocarbonate-treated water. This solution (50 µg) was used immediately for cDNA synthesis.

Tumor tissues obtained from either xenografts or tumor biopsies were minced in guanidine isothiocyanate (8 ml) with a polytron (Biospec Products Inc.), added to polyallomer ultracentrifuge tubes containing 4 ml of a 5 M cesium chloride solution, placed in a SW40Ti rotor, and centrifuged at 32,000 rpm for 16 h at 20°C. The pellet was resuspended in 300 µl of sodium acetate (pH 7); ethanol (100%; 600 µl) was added and precipitated overnight at -70°C. The precipitate was resuspended in 50 µl of diethylpyrocarbonate-water.

PCR Quantitation.
GADD153 message levels were quantified using a modification of the technique reported by Horikoshi et al. (29) . Total RNA acquired from the biopsies or xenografts was reverse-transcribed using random hexamers to produce cDNA (28 , 30 , 31) . Serial dilutions of the cDNA were made, and 5 µl of each dilution were placed in a sterile 0.5-ml Eppendorf tube. Taq mix (10 µl) containing 2.5 µl of 10 x Taq buffer [100 mM of Tris hydrochloride (pH 8.3) and 500 mM potassium chloride; Perkin-Elmer Corp., Norwalk, CT], 0.5 µl of 10 mM dNTPs, 1.52 µl or 2.0 µl of a 25 mM magnesium chloride buffer, and 5.48 µl or 4.88 µl of sterile PCR water, respectively, were added. In addition, 10 µl of primer mix were added containing: 1 µl of each of the sense and antisense primers (12.5 µM), 6 µl of sterile PCR water, 1.87 µl of total volume of 1 x Taq buffer and 32P-labeled dCTP (ratio 1 x buffer 23P-dCTP, 15:1), and 0.126 µl of Ampli-Taq DNA polymerase enzyme. Mineral oil was added to each tube. The tubes were placed in the thermocycler and amplified. Each set of PCR reactions contained at least one positive and one negative control (cDNA was substituted by sterile water). Each PCR reaction was performed in triplicate or quadruplicate. PCR products were separated on a 8 M (6%) acrylamide gel and analyzed using a Molecular Imaging System (Bio-Rad, Hercules, CA). For each RNA sample to be analyzed, a series of PCR reactions was performed to generate a graph of the amount of PCR product as a function of input cDNA for both GADD153 and ß-actin. These data were fit to a regression line, and the result was expressed as the ratio of the slope of the regression line for GADD153 to that for ß-actin. The primers listed below were synthesized by the Molecular Core Facility of the UCSD Cancer Center.

GADD153: CATACATCACCACAC (sense);

TGACCACTCTGTTTC (antisense);

ß-ACTIN: GAGCGGGAAATCGTGCGTGACATT (sense);

GATGGAGTTGAAGGTAGTTTCGTG (antisense);

Statistics.
One-way ANOVA (Scheffe’s procedure) was used to evaluate the significance of differences in the magnitude of the change in GADD153 level between patients who achieved either a CR, PR, or NR.


   Results
Top
 ABSTRACT
 Introduction
 Materials and Methods
 Results
Discussion
 REFERENCES
 
Assay Validation.
The PCR technique offers the possibility to quantify mRNA after conversion to cDNA and, therefore, can be a measurement of the expression of specific genes in tumor cells. Fig. 1Citation shows a relative determination of GADD153 expression in UMSCC10b cells exposed to cDDP (10 µM) for 1 h. The key feature is the amount of GADD153 and ß-actin product as a function of the input cDNA over a 20-fold range for both genes. For template amounts up to 2 nl and 5 nl, the increase in ß-actin and GADD153 products were linear, respectively. At higher levels, the PCR products reached a plateau. Because the cellular transcript level of ß-actin, a housekeeping gene, is constant under cDDP damaging conditions (32) , we used ß-actin as a normalization standard, determining the expression of GADD153 to that of ß-actin by calculating the ratio of the slopes of the linear portion of GADD153 and ß-actin curves. In Fig. 1Citation , the slope for GADD153 was 8.3 x 103, and that of ß-actin was 1426 x 103, resulting in a slope ratio of 0.006. This number is an empirical ratio and can be considered as an accurate measure of the relative expression of the target gene within the sample. Comparing the relative expression of the target gene in one sample with that in another sample will provide information about the relative expression of the target gene in the two samples. To further validate the use of RT-PCR, the coefficient of variation of the slope ratio (GADD153:ß-actin) in four different cDNAs obtained from different batches of the UMSCC10b cell line was 16.6%, indicating that the PCR technique can determine relative small changes in the level of RNA messages (data not shown).



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 1. Relationship between the amount of PCR product and cDNA template using UMSCC10b cells exposed to cDDP (10 µM). The linear portions of the curve are visualized in the insets, along with the regression equation for GADD153 and ß-actin cDNA.

 
Expression of GADD153 in Human Head and Neck Carcinoma Cells in Vitro.
In anticipation of the fact that the clinical use of GADD153 mRNA measurement will be facilitated if needle aspiration biopsies are sufficient for obtaining tumor tissue, the highly sensitive technique of RT-PCR, rather than the less sensitive technique of Northern blot analysis, was used throughout these studies for quantitation of mRNA. The magnitude of the change in the GADD153 mRNA level as a function of cDDP concentration was determined by treating UMSCC10b cells with 0.1, 1, 10, or 100 µM cDDP for 1 h and quantifying message level in untreated and treated cells harvested at 24 h by RT-PCR. Fig. 2Citation shows that there was a concentration-dependent increase in GADD153 mRNA levels that reached 6-fold at the highest cDDP concentration tested. Fig. 3Citation shows the time course of increase in GADD153 message level after a 1-h exposure to 100 µM cDDP. An increase in the GADD153 mRNA level was evident at 12 h after cDDP exposure, and the peak occurred at 24 h, followed by a decline to basal levels by 96 h.



View larger version (40K):
[in this window]
[in a new window]
  		Fig. 2. The effect of cDDP on GADD153 mRNA levels in UMSCC10b cells. Exponentially growing cultures were treated with various cDDP concentrations for 1 h and then grown in drug-free medium for 24 h. The levels of GADD153 were quantified by RT-PCR. Bars, ± SD.

 


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 3. Time course of cDDP-induced change in GADD153 mRNA level in UMSCC10b cells in vitro. Cells were exposed to 100 µM cDDP for 1 h and incubated in drug-free medium for various periods of time before the harvesting of RNA. Bars, ± SD.

 
Relationship between Extent of Cellular Injury and Change in the GADD153 mRNA Level.
If the magnitude of the increase in GADD153 mRNA reflects the extent of tumor cell injury, then one would expect a good correlation between clonogenic survival and fold increase in the GADD153 mRNA level after treatment of UMSCC10b cells with increasing concentrations of cDDP. Fig. 4Citation shows that this was, in fact, the case; over a 2-log range of tumor cell kill, the correlation coefficient was 0.98.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 4. Correlation between cell kill and the level of GADD153 expression in UMSCC10b cells in vitro. The r value of the best fitting curve was 0.98.

 
Further corroboration that the magnitude of the change in the GADD153 mRNA level reflects the extent of tumor cell damage was obtained using sublines of UMSCC10b selected for cDDP resistance. The parental cells, the 3-fold-resistant subline UMSCC10b-Pt/S6 (23 , 24) , and the 6-fold-resistant subline UMSCC10b-Pt/S15 (23 , 24) were exposed to 2 µM, 7µM, and 12 µM cDDP for 1 h. These levels of exposure correspond to the IC10, IC50, and IC70 for the parental cell line, respectively. Fig. 5Citation (left) shows that the magnitude of the increase in the GADD153 mRNA level in the 3-fold-resistant variant was lower than in the parental cells. In the 6-fold cDDP-resistant variant (Fig. 5Citation , left), the expression of GADD153 did not increase above its basal level at all after exposure to these cDDP concentrations. These results confirm that the increase in GADD153 depends on the extent of injury caused by the drug. A cDDP concentration that was highly toxic to the sensitive parental cells and resulted in a high level of GADD153 expression failed to produce any increase in the GADD153 message level in resistant cells in which there was little toxicity. To assure that the resistant variants had not lost their capacity to activate the GADD153 gene, and to validate the change in the GADD153 message level as a molecular marker of tumor injury in resistant as well as sensitive cells, the effect of cDDP applied at equitoxic concentrations was compared between the parental cells and the two resistant sublines. As shown in Fig. 5Citation (right), no difference between the cell lines in the change in expression of GADD153 was detected at the three different levels of cytotoxicity tested (IC10, IC50, IC70), indicating that the extent of cellular injury is a determinant of the level of GADD153 mRNA expression.



View larger version (28K):
[in this window]
[in a new window]
  		Fig. 5. Magnitude of increase in GADD153 mRNA level in cDDP-sensitive and -resistant variants of UMSCC10b cells after exposure to equimolar concentrations of cDDP (left) and equitoxic concentrations of cDDP (right). RNA was extracted 24 h after a 1 h exposure to CDDP. Error bars, ± SD. {square}, parent line; , 3-fold resistant; , 6-fold resistant.

 
Expression of GADD153 in Xenografts.
Nude mice were inoculated s.c. with human UMSCC10b tumor cells and then treated with cDDP (0, 10, 15, 30, or 50 mg/kg cDDP i.p.) when the tumors reached 1 ml in volume. Fig. 6Citation (left) shows the results obtained when tumors were harvested at 24 h after drug treatment. As was the case for UMSCC10b cells exposed to cDDP in vitro, GADD153 mRNA levels increased as a function of cDDP dose. The time course of the change in message level in the xenografts was similar to that observed with in vitro exposure of the cells (Fig. 6Citation , right). The fact that GADD153 mRNA levels can be increased in vivo by cDDP doses around the maximum tolerated dose in this system (15–20 mg/kg cDDP), and the fact that the increase in GADD153 mRNA level was transient in vivo and returned to toward basal levels within a few days after treatment, suggested that change in the GADD153 mRNA level could be used as a molecular marker of tumor cell injury in patients.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 6. Change in GADD153 mRNA levels in UMSCC10b xenografts growing in nude mice. The effect of cDDP dose on the level of GADD153 mRNA is demonstrated in the left graph, and the effect of time posttreatment is demonstrated in the right graph. Note that the scale of the ordinate differs for the two graphs.

 
Expression of GADD153 in Human Tumor Biopsies.
If quantitation of message levels is to be used successfully to assess the extent of tumor injury, then the coefficient of variation between samples taken from the same tumor must be small enough that the changes detected in any one biopsy are reasonably representative of changes occurring throughout the tumor. A needle biopsy samples only a limited portion of a tumor nodule, and regional variation in histology, drug delivery, and cellular response is to be expected. The challenge is to demonstrate that, despite these potential sources of variance, the induction of the genes of interest can be quantitated with a sufficiently small coefficient of variation to have predictive value. We analyzed the level of GADD153 mRNA in three to five biopsies obtained from different portions of the same tumor mass in four cancer patients 24 h after administration of cytotoxic therapy, and did the same thing with an UMSCC10b xenograft growing in nude mice after treatment with cDDP. The data presented in Table 1Citation indicate that the coefficient of variation ranged between 20% and 29%, which demonstrates a relatively high reproducibility of the assay itself and a relatively small effect of potential regional heterogeneity within the tumor tissue.


View this table:
[in this window]
[in a new window]
  		Table 1 Heterogeneity between biopsies

 
Having demonstrated that, despite potential sources of variation, message levels can be quantitated with a sufficiently small coefficient of variation to make this approach of interest, we examined the relationship between the change in the GADD153 mRNA level at 24 h after the start of the first course of treatment with cDDP or cDDP plus radiation therapy and subsequent clinical response to this therapy in 32 patients with stage IV head and neck cancer. The data for each patient is presented in Table 2Citation and summarized in Fig. 7Citation . Seventeen of the 32 patients attained a CR; in this group, the increase in the GADD153 message level ranged from 2–9-fold, with a mean of 3.8 ± 2.2 (SD)-fold. Eight of the 32 patients achieved a PR. Only four of these patients had an increase in GADD153 mRNA that ranged from 1.3–3.5-fold; the mean for all eight patients was 1.6 ± 0.9 (SD)-fold. Seven patients had NR, and their mean change in GADD153 mRNA was 0.8 ± 0.5 (SD)-fold. The magnitude of the induction for the complete responders was statistically significantly different from that observed for both the partial responders (P < 0.05) and nonresponders (P < 0.005), and the value for the partial responders differed from that of the nonresponders (P < 0.05). An increase in GADD153 mRNA of 1.75-fold or higher predicted a CR with a sensitivity of 94% and a specificity of 87%.


View this table:
[in this window]
[in a new window]
  		Table 2 Expression of GADD153 in tumor tissue of head and neck patients before and after cDDP treatment

 


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 7. Change in GADD153 mRNA levels in patients attaining either a CR, PR, or NR to cDDP-based treatment. The increase in the GADD153 message level was significantly higher in the complete responders compared with either the partial responders or nonresponders (P < 0.05 and P < 0.005, respectively, determined by one-way ANOVA, Scheffe’s procedure).

 

   Discussion
Top
 ABSTRACT
 Introduction
 Materials and Methods
 Results
 Discussion
REFERENCES
 
Elucidation of the molecular processes involved in the cellular injury response is yielding opportunities for the identification of novel markers that reflect the extent of injury produced in tumors by treatment with chemotherapeutic agents (33) . Work by this and other laboratories has identified GADD153 as an important player in the cellular injury response and potentially a useful marker for response (9 , 10 , 14 , 17 , 18 , 20,) . The work reported here documents that changes in GADD153 mRNA are quantitatively linked to the extent of tumor cell kill in vitro and to the dose of drug when the same cells are grown as xenografts in vivo. Most important, we demonstrate that change in the GADD153 mRNA level measured at 24 h is linked to the clinical response in patients with far advanced head and neck carcinomas.

GADD153 is a gene that is strongly transcriptionally activated by cDDP (17 , 31) , as well as by other types of cellular stress (11, 12, 13 , 34) , including other classes of chemotherapeutic drugs (10 , 35) . Among the genes, the message levels of which are known to increase after cellular injury, GADD153 is of particular interest as a potential marker because of the magnitude of its induction. The results reported here further document the linkage between the extent of cellular injury as quantitated by clonogenic survival and change in mRNA level. Not only was there an excellent correlation between clonogenic survival and fold increase in GADD153 mRNA in the parental UMSCC10b cells, but the UMSCC10b-Pt/S6 and UMSCC10b-Pt/S15 sublines, which sustain less lethal injury than the parental cells at a given concentration of cDDP, had proportionately smaller increases in GADD153 message. When the cells lines were compared at concentrations of cDDP that produced the same clonogenic survival, the magnitude of GADD153 message increase was the same for all three lines. This linkage seems to extend to the in vivo situation, as well. The magnitude of the change in the GADD153 mRNA level in the xenografts increased with increasing cDDP dose over a range in which UMSCC10b tumors are known to be responsive.

Little is known about the role or function of GADD153 in the cellular injury response. Several DNA-damage inducible genes have been shown to be regulated transcriptionally via the phorbol ester-responsive element within the promoter region (1 , 36) . Both the collagenase and c-Jun genes contain AP-1 sites that are critical for both UV- and 12-O-tetradecanoylphorbol-13-acetate-induced expression (1) . The GADD153 gene is unique in that it is not responsive to 12-O-tetradecanoylphorbol-13-acetate despite the presence of an AP-1 binding site, distinguishing itself from other response genes such as the early-response gene c-Jun and heat shock family genes. However, GADD153 can undergo inducible phosphorylation on two adjacent serine residues by a specific stress-activated MAP kinase (37) , which enhances the ability of GADD153 to function as a transcriptional activator (36) . Our observations indicated that GADD153 is functionally located downstream of a hypothetical injury detection site, as shown by the induction kinetics of GADD153 (Fig. 3)Citation , but upstream of the cell cycle control and cell growth events [Fig. 4Citation , (32 , 36) ]. The latter suggests that GADD153 may serve as a link between early and late events in the response to cellular damage. This hypothetical chain of events is further based on findings by Wang et al. (15) , who established the GADD153 protein as a stress-inducible transcription factor that activates a novel set of target genes in response to stress. In this role, GADD153 possibly can serve as an excellent marker for the extent of the cellular response to injury.

Practically, to use GADD153 as a marker for the extent of the cellular injury response in human tumor biopsies, the within-tumor heterogeneity must be small enough that a single biopsy would be sufficient to provide a reliable indication of the response of the whole tumor nodule. Our results establish that it is possible to make GADD153 mRNA measurements on the very small amounts of tissue typically obtained from a needle aspiration or cutting biopsy, and that the within-tumor coefficient of variance seems to be acceptably small, ranging from 26–29% in the patient tumors examined, indicating that the expression of GADD153 in a single tumor biopsy is sufficiently representative of the whole tumor that major changes in mRNA level can be detected. This is in agreement with results reported by two other investigators. Horikoshi et al. (29) demonstrated that thymidylate synthase mRNA levels in three different parts of the same tumor varied by only 21%. In a second study, in which bcl-2 expression was measured immunohistochemically, heterogeneous staining of the bcl-2 protein in paraffin sections of a subgroup of head and neck tumors (stages II-IV) with weak bcl-2 expression was found to be <25% (38) , indicating that locally obtained molecular information can be representative for the whole tumor. Irrespective of whether a single biopsy or a paraffin section can be used as a representative sample of a rather heterogeneous tumor, our data indicated that the measurement of the GADD153 message level has clinical relevance.

Another issue of great concern with this approach is the fact that an aspiration biopsy inevitably contains some normal tissue elements, as well as tumor cells. The actual fraction of tumor cells in the sample cannot practically be determined without sacrificing most of the sample, and it is not known how GADD153 induction differs in the normal and malignant elements of the biopsy. However, analyzing the cDDP-DNA adduct formation and increase in GADD153 protein level in both tumor and stromal cells in a xenograft model, we recently estimated that ~25-fold more GADD153 protein per cDDP-DNA adduct was formed in tumor cells than in stromal cells (39) . The latter indicates that the contribution of nontumor cells to the level of GADD153 expression within biopsies is minimal, at least in cases where >30% of the cells biopsied are tumor cells. Although more information is needed to resolve this issue, the good correlation between the measured increase in GADD153 mRNA and clinical response in this study argues that operationally useful information on the extent of tumor injury can be obtained.

As was expected from our in vitro and in vivo xenograft experiments, GADD153 mRNA was detectable in all tumor samples obtained from patients. The important observation that emerged from analysis of this group of 32 patients was that the level of GADD153 mRNA after treatment increased in patients who went on to attain a CR, but not in patients who failed to respond. Thus, it seems that the linkage between actual cellular injury and increase in the GADD153 message level that was evident in the in vitro and xenograft studies of UMSCC10b, and in similar types of studies that we have previously reported on a human melanoma and ovarian carcinoma (17 , 31) , does translate to the clinical setting in the case of head and neck cancers treated with cDDP. It is of some concern that the magnitude of the increase in message level averaged only 3.8-fold in the patients who went on to attain a CR and that there was some overlap in the fold increase between complete and partial responders and partial and nonresponders. Additional clinical studies will be required to determine whether greater separation between responders and nonresponders can be obtained in other tumor types and with other chemotherapeutic regimens. It seems quite likely that it will be possible to identify other genes for which the magnitude of the change in mRNA is even greater than for GADD153 and that assessment of the extent of tumor injury can be refined by combining information obtained from several genes, the induction of which reflects different elements of the cellular injury response.

It was the goal of the clinical study reported here to determine the feasibility of making GADD153 measurements in the clinical setting. The fact that the magnitude of the increase in the GADD153 message in patients with far-advanced head and neck carcinoma differed as a function of whether or not the patient had a clinical response or not and that an increase of GADD153 mRNA of 1.75-fold or greater predicted a CR with a sensitivity of 94% and a specificity of 87%, indicates that the measurement of a molecular marker of cellular injury does provide prognostic information. However, there are a number of limitations to this approach, and its validation will eventually require direct comparison with other indicators of tumor response in the same group of patients. Operationally, facile clinical use of this approach requires that it be possible to make the GADD153 mRNA measurement on very small tissue samples, that the within-tumor variance be relatively small, and that the magnitude of the increase in mRNA level be large enough relative to the within-tumor variance to detect clinically significant degrees of tumor injury.

Despite the limitations mentioned above, this study has confirmed the validity of this novel approach to determine the response of tumors to treatment by analyzing the increase of GADD153 mRNA levels after cDDP treatment. It provides the basis for similar studies examining a diverse group of gene products, including other cellular injury-response gene products involved in cell cycle control and apoptosis.


   FOOTNOTES
 
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.

1 This study was supported in part by NIH Grant CA 67269 and was conducted in part by the Clayton Foundation for Research—California Division. G. L., R. C., and S. B. H. are Clayton Foundation Researchers. Back

2 To whom requests for reprints should be addressed, at UCSD Cancer Center, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0058. Back

3 The abbreviations used are: cDDP, cisplatin; NR, no response; PR, partial response; CR, complete response; RT-PCR, reverse transcription-PCR. Back

Received 2/ 9/99; revised 3/22/99; accepted 3/29/99.


   REFERENCES
Top
 ABSTRACT
 Introduction
 Materials and Methods
 Results
 Discussion
 REFERENCES
 

  1. Büscher M., Rahmsdorf H. J., Litfin M., Karin M., Herrlich P. The activation of the c-fos gene by UV and phorbol ester: different signal transduction pathways converge to the same enhancer element. Oncogene, 3: 301-311, 1988.[Medline]
  2. Futscher B. W., Erickson L. C. Changes in c-myc and c-fos expression in a human tumor cell line following exposure to bifunctional alkylating agents. Cancer Res., 50: 62-66, 1990.[Abstract/Free Full Text]
  3. Rubin E., Kharbanda S., Gunji H., Weichselbaum R., Kufe D. cis-diamminedichloroplatinum(II) induces c-jun expression in human myeloid leukemia cells: potential involvement of a protein kinase C-dependent signaling pathway. Cancer Res., 52: 878-882, 1992.[Abstract/Free Full Text]
  4. Karin M., Herrlich P. Cis- and trans-acting genetic elements responsible for induction of specific genes by tumor promoters, serum factors, and stress Colburn N. H. eds. . Genes and Signal Transduction in Multistage Carcinogenesis, : 425-440, Marcel Dekker, Inc. New York 1989.
  5. Holbrook N. J., Fornace A. J. Response to adversity: molecular control of gene activation following genotoxic stress. New Biol., 3: 825-833, 1991.[Medline]
  6. Fornace A. J., Nebert D. W., Hollander M. C., Luethy J. D., Papathanasiou M., Fragoli J., Holbrook N. J. Mammalian genes coordinately regulated by growth arrest signals and DNA-damaging agents. Mol. Cell. Biol., 9: 4196-4203, 1989.[Abstract/Free Full Text]
  7. Luethy J. D., Fragnoli J., Park J. S., Fornace A. J., Jr., Holbrook N. J. Isolation and characterization of the hamster gadd153 gene. J. Biol. Chem., 265: 16521-16526, 1990.[Abstract/Free Full Text]
  8. Park J. S., Luethy J. D., Wang M. G., Fargnoli J., Fornace A. J., Jr., McBride O. W., Holbrook N. J. Isolation, characterization and chromosomal localization of the human gadd153 gene. Gene, 116: 259-267, 1992.[Medline]
  9. Fornace A. J., Jr., Nebert D. W., Hollander C., Luethy J. D., Papathanasiou M., Fargnoli J., Holbrook N. Mammalian genes coordinately regulated by growth arrest signals and DNA-damaging agents. Mol. Cell. Biol., 10: 4196-4203, 1989.
  10. Luethy J. D., Holbrook N. J. Activation of the gadd153 promoter by genotoxic agents: a rapid and specific response to DNA damage. Cancer Res., 52: 5-10, 1992.[Abstract/Free Full Text]
  11. Price B., Calderwood S. Gadd45 and Gadd153 messenger RNA levels are increased during hypoxia and after exposure of cells to agents which elevate the levels of glucoes regulated proteins. Cancer Res., 52: 33814-3817, 1992.
  12. Wang X-Z., Ron D. Stress-induced phosphorylation and activation of the transcription factor CHOP (GADD153) by p38 MAP kinase. Science (Washington DC), 272: 1347-1349, 1996.[Abstract]
  13. Wang X-Z., Lawson B., Brewer J. W., Zinszner H., Sanjay A., Mi L-J., Boorstein R., Kreibich G., Hendershot L. M., Ron D. Signals from the stressed endoplasmic reticulum induce C/EBP-homologous protein (CHOP/GADD153). Mol. Cell. Biol., 16: 4273-4280, 1996.[Abstract]
  14. Eymin B., Dubrez L., Allouche M., Solary A. Increased gadd153 messenger RNA level is associated with apoptosis in human leukemic cells treated with etoposide. Cancer Res., 57: 686-695, 1997.[Abstract/Free Full Text]
  15. Wang X-Z., Kuroda M., Sok J., Batchvarova N., Kimmel R., Chung P., Zinszner H., Ron D. Identification of novel stress-induced genes downstream of chop. EMBO J., 17: 3619-3630, 1998.[Medline]
  16. Gately D. P., Sharma A., Christen R. D., Howell S. B. Cisplatin and taxol activate different signal pathways regulating cellular injury-induced expression of GADD153. Br. J. Cancer, 73: 18-23, 1996.[Medline]
  17. Gately D. P., Jones J. A., Christen R., Barton R., Los G., Howell S. B. Induction of growth arrest and DNA damage inducible gene gadd153 by cisplatin in vitro and in vivo. Br. J. Cancer, 70: 1102-1106, 1994.[Medline]
  18. Los G., Barton R. M., Nakata B., Lee L. K., van Veelen L. R., Howell S. B. DNA damage and early-response genes as a tool to monitor tumor injury following chemotherapy. Proc. Am. Assoc. Cancer Res., 34: 433 1993.
  19. Friedman A. D. GADD153/CHOP, a DNA damage-inducible protein, reduced CAAT/enhancer binding protein activities and increased apoptosis in 32D cl3 myeloid cells. Cancer Res., 56: 3250-3256, 1996.[Abstract/Free Full Text]
  20. Los G., Barton R., van Veelen L., Woods A., Vicario D., Robbins K. T., Howell S. B. Association between DNA damage-inducible gene expression and tumor response. Proc. Am. Assoc. Cancer Res., 35: 550 1994.
  21. Ou Y. C., Thompson S. A., Kirchner S. C., Kavanagh T. J., Faustman E. M. Induction of growth arrest and DNA damage-inducible genes Gadd45 and Gadd153 in primary rodent embryonic cells following exposure to methylmercury. Toxicol. Appl. Pharmacol., 147: 31-38, 1997.[Medline]
  22. Grenman R., Burk D., Virolainen E., Buick R. N., Church J., Schwartz D. R., Carey T. E. Clonogenic cell assay for anchorage-dependent squamous carcinoma cell lines using limiting dilution. Int. J. Cancer, 44: 131-136, 1989.[Medline]
  23. Nakata B., Barton R. M., Robbins T. K., Howell S. B., Los G. Association between hsp60 mRNA levels and cisplatin resistance in human head and neck cancer cell lines. Int. J. Oncol., 6: 1425-1432, 1994.
  24. Nakata B., Albright K. D., Barton R. M., Howell S. B., Los G. Synergistic interaction between cisplatin and tamoxifen and the delay of cisplatin resistance in head and neck carcinoma cell lines. Cancer Chemother. Pharmacol., 35: 511-518, 1995.[Medline]
  25. Robbins R. T., Vicario D., Seagren S., Weisman R., Pelliterri P., Kerber C., Orloff L., Los G., Howell S. B. A targeted supradose cisplatin chemoradiation protocol for advanced head and neck cancer. Am. J. Surg., 168: 419-422, 1994.[Medline]
  26. Robbins K. T., Storniolo A. M., Kerber C., Vicario D., Seagren S., Shea M., Hanchett C., Los G., Howell S. B. Phase I study of highly selective supradose cisplatin infusion for advanced head and neck cancer. J. Clin. Oncol., 12: 2113-2120, 1994.[Abstract/Free Full Text]
  27. Weisman R. A., Christen R., Los G., Jones V., Kerber C., Seagren S., Glassmeyer S., Orloff L. A., Wong W., Kirmani S., Howell S. Phase I trial of retinoic acid and cis-platinum for advanced squamous cell cancer of the head and neck based on experimental evidence of drug synergism. Otolaryngol. Head Neck Surg., 118: 597-602, 1998.[Medline]
  28. Chomczynski P., Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162: 156-159, 1987.[Medline]
  29. Horikoshi T., Danenberg K. D., Stadlbauuer T. H. W., Volkenandt M., Shea L. C. C., Aigner K., Gustavsson B., Leichman L., Frösing R., Ray M., Gibson N. W., Spears P. C., Danenberg P. V. Quantitation of thymidylate synthase, dihydrofolate reductase, and DT-diaphorase gene expression in human tumors using the polymerase chain reaction. Cancer Res., 52: 108-116, 1992.[Abstract/Free Full Text]
  30. Sambrook J., Fritch E. F., Maniatis T. Molecular Cloning: A Laboratory Manual Cold Spring Harbor Laboratory Press Cold Spring Harbor, NY 1989.
  31. Davis L. G., Bidner M. D., Battey J. F. Basic Methods in Molecular Biology Elsevier New York 1986.
  32. Jones J. A., Gately D. P., Barton R. C., Albright K. D., Christen R. D., McClay E. F., Los G., Howell S. B. Induction of gadd153 in human melanoma xenografts as an indicator of genotoxic injury. Cell. Pharmacol., 1: 233-237, 1994.
  33. Howell S. B., Gately D. P., Christen R. D., Los G. The cisplatin induced cellular injury response Pinedo H. M. Schornagel J. H. eds. . Platinum and Other Metal Coordination Compounds in Cancer Chemotherapy 2, : 269-282, Plenum Press New York and London 1996.
  34. Fornace A. J., Jackman J., Hollander M. C., Hoffman-Lieberman B., Lieberman D. H. Genotoxic-stress-response genes and growth-arrest-genes. Anal. NY Acad. Sci., 663: 139-153, 1992.[Abstract]
  35. Luethy J. D., Holbrook N. J. The pathway regulating gadd153 induction in response to DNA damage is independent of protein kinase C of tyrosine kinases. Cancer Res., 54: 1902-1906, 1994.
  36. Devary Y., Gottlieb R. A., Lau L. F., Karin M. Rapid and preferential activation of the c-jun gene during the mammalian UV response. Mol. Cell. Biol., 11: 2804-2811, 1991.[Abstract/Free Full Text]
  37. Wang X-Z., Ron D. Stress-induced phosphorylation and activation of the transcription factor CHOP (GADD153) by p38 MAP kinase. Science (Washington DC), : 1347-1349, 1996.
  38. Gasparini G., Bevilacque P., Bonoldi E., Testolin A., Galassi A., Verderio P., Boracchi P., Guglielmi R. B., Pezzella F. Predictive and prognostic markers in a series of patients with head and neck squamous cell invasive carcinoma treated with concurrent chemoradiation therapy. Clin. Cancer Res., 1: 1375-1383, 1995.[Abstract]
  39. Johnsson A., Strand C., Los G. Expression of GADD153 in tumor cells and stromal cells from xenografted tumors in nude mice treated with cisplatin: correlation with cisplatin-DNA adducts. Cancer Chemother. Pharmacol., 43: 348-352, 1999.[Medline]



This article has been cited by other articles:


Home page
 Toxicol SciHome page
Q. Huang, R. T. Dunn II, S. Jayadev, O. DiSorbo, F. D. Pack, S. B. Farr, R. E. Stoll, and K. T. Blanchard
Assessment of Cisplatin-Induced Nephrotoxicity by Microarray Technology
Toxicol. Sci., October 1, 2001; 63(2): 196 - 207.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 Los, G.
Right arrow Articles by Howell, S. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Los, G.
Right arrow Articles by Howell, S. B.

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