Cancer Research Annual Meeting 2010 Sign up for Cancer Research eTOC's
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 Email this article to a friend
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 Fine, B. M.
Right arrow Articles by Boxer, L. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fine, B. M.
Right arrow Articles by Boxer, L. M.
[Cancer Research 65, 291-299, January 1, 2005]
© 2005 American Association for Cancer Research

Experimental Therapeutics and Molecular Targets, and Chemical Biology

A Genome-Wide View of the In vitro Response to L-Asparaginase in Acute Lymphoblastic Leukemia

Bernard M. Fine1,2, Gertjan J.L. Kaspers3, Minh Ho1, Anne H. Loonen3 and Linda M. Boxer1

1 Center for Molecular Biology in Medicine, Veterans Affairs Palo Alto Health Care System and Department of Medicine; 2 Department of Biochemistry and Howard Hughes Medical Institute, Stanford University School of Medicine, Stanford, California and 3 Pediatric Hematology/Oncology, VU University Medical Center, Amsterdam, the Netherlands

Request for reprints: Linda M. Boxer, Division of Hematology, Stanford University School of Medicine, 269 Campus Drive, CCSR 1155, Stanford, CA 94305-5156. Phone: 650-849-0551; Fax: 650-858-3982; E-mail: lboxer{at}stanford.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
To investigate the effect of L-asparaginase on acute lymphoblastic leukemia (ALL), we used cDNA microarrays to obtain a genome-wide view of gene expression both at baseline and after in vitro exposure to L-asparaginase in cell lines and pediatric ALL samples. In 16 cell lines, a baseline gene expression pattern distinguished L-asparaginase sensitivity from resistance. However, for 28 pediatric ALL samples, no consistent baseline expression pattern was associated with sensitivity to L-asparaginase. In particular, baseline expression of asparagine synthetase (ASNS) was not predictive of response to L-asparaginase. After exposure to L-asparaginase, 5 cell lines and 10 clinical samples exhibited very similar changes in the expression of a large number of genes. However, the gene expression changes occurred more slowly in the clinical samples. These changes included a consistent increase in expression of tRNA synthetases and solute transporters and activating transcription factor and CCAAT/enhancer binding protein family members, a response similar to that observed with amino acid starvation. There was also a consistent decrease in many genes associated with proliferation. Taken together, the changes seem to reflect a consistent coordinated response to asparagine starvation in both cell lines and clinical samples. Importantly, in the clinical samples, increased expression of ASNS after L-asparaginase exposure was not associated with in vitro resistance to L-asparaginase, indicating that ASNS-independent mechanisms of in vitro L-asparaginase resistance are common in ALL. These results suggest that targeting particular genes involved in the response to amino acid starvation in ALL cells may provide a novel way to overcome L-asparaginase resistance.

Key Words: asparaginase • acute lymphoblastic leukemia • expression profiling • microarrays


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Over the past several decades, there has been a steady improvement in the survival of patients with acute lymphoblastic leukemia (ALL; ref. (1). However, a substantial number of both adults and children still die from this disease. A rational approach to improving the outcome in ALL is to gain a better understanding of the mechanisms of response to treatment and to use this understanding to improve its efficacy. Toward this goal, we have used high-density cDNA microarrays to provide a comprehensive, genome-wide view of the expression patterns of both cell lines and pediatric ALL samples at baseline and after invitro exposure to L-asparaginase.

L-Asparaginase is an important component of most treatment regimens for ALL. The importance of L-asparaginase in the treatment of ALL is highlighted by the observation that both invitro and in vivo resistance to L-asparaginase are associated with a poor long-term outcome (2, 3). It has generally been thought that the sensitivity of ALL cells to L-asparaginase was due to relatively low expression of asparagine synthetase (ASNS; ref. (4). This notion is supported by a 1969 study that involved a small number of patients(5). Since then, most studies of the mechanisms of resistance have used cell lines or animal models (6, 7)7. However, it has recently been observed that ASNS activity varies widely in clinical ALL samples (8). Furthermore, the importance of ASNS in L-asparaginase resistance has been brought into question very recently by a study that investigated ASNS mRNA levels in clinical ALL samples (9).

The study reported here is the first description of a comprehensive genome-wide view of the response of ALL cells to L-asparaginase. This study is also the first comprehensive comparison of the L-asparaginase response of cell lines and clinical samples. This comparison is particularly important because, as described above, studies of cell lines are the basis for much of our current understanding of the mechanisms of response to L-asparaginase.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell Lines
The characteristics of the cell lines are shown in Table 1. The cells were grown in complete RPMI 1640 (Cellgro, Herndon, VA); 15% heat-inactivated FCS with amphotericin B (0.125 µg/mL) and gentamicin (200 µg/mL, all three from Sigma, St. Louis, MO); penicillin (100 units/mL), streptomycin (100 µg/mL), and L-glutamine (2 mmol/L, supplied as PSG, all three from Omega Scientific, Tarzana, CA); and insulin (5 µg/mL), transferrin (5µg/mL), and selenium (5 ng/mL, supplied as ITS, Invitrogen, Carlsbad, CA) at 37°C and 5% CO2.


View this table:
[in this window]
[in a new window]

  		Table 1. Cell line characteristics

 
Patient Samples
After obtaining informed consent from their guardians, clinical samples were obtained by peripheral blood draw or bone marrow aspiration from children with ALL before initiation of treatment. Mononuclear cells were isolated from the samples using Ficoll-Paque. The cells were placed in RPMI 1640 with 10% heat-inactivated FCS, penicillin (100 units/mL), streptomycin (100 µg/mL), L-glutamine (2 mmol/L), and 10% DMSO and frozen in liquid nitrogen for long-term storage. An aliquot of each clinical sample was used to determine the concentration of L-asparaginase lethal to 50% of the cells (the LC50) as described below. From the stored samples, we selected samples that contained at least 15 x 106 cells and at least 75% leukemic blasts. Patient characteristics are shown in Table 2.


View this table:
[in this window]
[in a new window]

  		Table 2. Patient characteristics

 
Determination of L-Asparaginase LC50
Each clinical sample and cell line was cultured in RPMI 1640 containing FCS and other supplements (10) in the presence of six different concentrations of L-asparaginase as described (11). After 4 days, cell viability was assessed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay as described previously (10). From these data, the concentration of L-asparaginase lethal to 50% of the cells (the LC50) was determined and this was used as the measure of response to L-asparaginase. It has been shown that peripheral blood and bone marrow ALL samples produce the same 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide results, provided they contain at least 70% blasts (11). As in prior studies (2), a sample or cell line was considered to be sensitive if LC50 < 0.03 IU/mL, intermediate if 0.03 IU/mL < LC50 < 0.72 IU/mL, or resistant if LC50>0.72 IU/mL.

RNA Preparation
Baseline Expression in Clinical Samples. Viably frozen vials of individual clinical ALL samples were prepared as described above. Each vial was rapidly thawed at 37°C and then immediately lysed in Trireagent (Molecular Research Center, Cincinnati, OH). Total RNA was isolated according to the manufacturer's protocol. Each total RNA sample was then treated with DNase (DNA-free, Ambion, Austin, TX) and linearly amplified using a protocol based on Wang et al. (12).

Serial Lysis of Cell Lines after Exposure to L-Asparaginase. We examined the changes in gene expression in cell lines at 2, 4, 8, and 12 hours after exposure to 2 IU/mL L-asparaginase. To carry out this study, cell lines were propagated in culture as described above. Twelve hours before exposure to L-asparaginase, the cells were transferred into fresh culture medium at a cell density of 106 cells/mL. Immediately before the addition of L-asparaginase, two aliquots of the cell culture were removed to serve as controls. One of these aliquots was immediately lysed in Trireagent and stored at –80°C. The other aliquot was incubated at 37°C and 5% CO2 for 12 hours to control for any changes in gene expression that might occur over this period. L-Asparaginase was added to the remaining volume of cell culture to achieve a concentration of L-asparaginase of 2 IU/mL. The remaining cell culture was divided into four equal volumes and incubated at 37°C and 5% CO2 until the cells were lysed in Trireagent at 2, 4, 8, and 12 hours and stored at –80°C until total RNA was isolated according to the manufacturer's protocol.

Lysis of Clinical Samples after Exposure to L-Asparaginase. Viably frozen vials of individual clinical ALL samples were prepared as described above. Each vial was rapidly thawed at 37°C, washed, and transferred to the culture medium described above. The sample was then divided into two equal flasks, one of which was exposed to 2 IU/mL L-asparaginase. The flasks were then incubated at 37°C and 5% CO2 for either 8 or 24 hours. The cells were then lysed in Trireagent and stored at –80°C until total RNA was isolated according to the manufacturer's protocol. Each total RNA sample was then treated with DNase (DNA-free) and linearly amplified using MessageAmp (Ambion). We obtained aliquots of the cells immediately after thawing and immediately before lysis. These aliquots were used to inspect the cells using trypan blue exclusion and to prepare cytospins that were stained with Wright-Giemsa. These examinations of the cells confirmed that the majority of cells were viable and that the percentage of blasts did not change significantly between thawing the cells and lysing them.

Gene Expression Measurements
All protocols are posted at http://brownlab.stanford.edu/protocols.html. Spotted cDNA microarrays that contained ~42,000 spots, representing ~30,000 genes (National Center for Biotechnology Information Unigene Build 161), were produced as described previously (13). The sample RNA and reference RNA (Universal Human Reference, Stratagene, La Jolla, CA) were labeled with different fluorescent dyes (Cy5-dUTP and Cy3-dUTP, Amersham, Piscataway, NJ) and comparatively hybridized to an array. For the clinical samples, amplified RNA was comparatively hybridized with reference RNA that was also amplified using the same RNA amplification protocol.

The fluorescence intensities of Cy5 and Cy3 on each array were measured using a GenePix 4000 scanner (Axon Instruments, Foster City, CA). Images were analyzed using GenePix Pro 3.0 software (Axon Instruments) to semiautomatically identify and quantify hybridization to the cDNA spots on the microarrays. Any areas of the microarrays with obvious blemishes were manually omitted from subsequent analysis. Spots were considered well measured only if the reference RNA fluorescent intensity was >1.5 times the local background and the regression correlation was >0.6. Any clone that was not well measured on at least 80% of the arrays was excluded from subsequent analysis. For each array, we used a scaling factor to set the mean sample-to-reference ratio for all well-measured spots to 1. For all subsequent analysis, we used log2 of this normalized sample-to-reference ratio.

Baseline gene expression data were analyzed as follows. The cell line data and the clinical sample data were analyzed as two separate data sets. For each data set, the expression level for each clone was centered by subtracting its mean log2 ratio in the data set from each measurement. We analyzed the data in a supervised manner using agglomerative hierarchical clustering (14). To analyze the data in a supervised manner, we used significance analysis of microarrays (SAM; ref. (15). We also used prediction analysis of microarrays (16) to search for genes that distinguish L-asparaginase-sensitive samples from L-asparaginase-resistant samples. We estimated the misclassification error for this classifier by using 10-fold cross-validation as described (16).

Time course data were analyzed as follows. Initially, each time course was analyzed separately. For the cell lines, measurements at 0 and 12 hours in the absence of L-asparaginase were used to determine the baseline expression level for each clone in each time course. Specifically, for each clone, the expression levels at t = 0 and 12 hours in the absence of L-asparaginase were compared. If the difference between these two measurements differed by >1.0, the data for that clone for that particular time course were omitted from further analysis. Otherwise, these two measurements were averaged. This average was then subtracted from each of the time points in the time course. For the clinical samples, the expression level for each clone in the absence of L-asparaginase was subtracted from the expression level in the presence of L-asparaginase. These transformations provide, for both cell lines and clinical samples, expression levels for each clone in the presence of L-asparaginase relative to its expression level in the absence of L-asparaginase. It was only after these transformations that data from different time courses were compared.

Primary data are publicly available through the Stanford Microarray Database (http://genome-www.stanford.edu/microarray). The data have also been deposited at Array Express (http://www.ebi.ac.uk/arrayexpress), with accession nos. E-SMDB-25, E-SMDB-26, and E-SMDB-27. Supplemental Materials are available at http://microarray-pubs.stanford.edu/Lasp/.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Baseline Gene Expression and Response to L-Asparaginase. The baseline gene expression patterns in a collection of ALL cell lines have been published recently and are publicly available (17). For 16 of these cell lines, we used the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay to determine the L-asparaginase LC50. The results are shown in Table 1. Using the L-asparaginase LC50, we classified each of the cell lines as either sensitive or resistant to L-asparaginase. We then used SAM to identify genes whose baseline expression in the cell lines was associated with sensitivity or resistance to L-asparaginase. The results of this analysis are shown in Fig. 1. Among other features, we found that high baseline expression of ASNS was associated with resistance to L-asparaginase in these cell lines. However, one cell line, Nalm27, was resistant to L-asparaginase and had low baseline expression of ASNS.



View larger version (59K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Baseline gene expression in a collection of cell lines. A, variation in expression is displayed as a variation in color (14) for the 331 clones (representing 306 genes) for which the log2 intensity ratio differed by at least 2 from its mean on at least 2 arrays. Upper right corner, the color scale extends from 0.25- to 4.0-fold of the mean (–2 to 2 in log2 space). Gray, data that were omitted because these were not well measured as described in Materials and Methods. The arms of the dendrogram are color coded to indicate the chromosomal translocation associated with each branch: blue, t(9;22); pink, t(1;19); purple, MLL; brown, t(17;19); orange, t(12;21); black, t(5;12). The broad features of the clustering patterns were robust to variations in gene selection criteria. The stripe of color-labeled L-asparaginase LC50 provides a representation of the L-asparaginase LC50 for each cell line: bright red, highest LC50 value (L-asparaginase resistant); bright green, lowest LC50 value (L-asparaginase sensitive). B, variation in expression of the 52 clones (representing 50 genes) identified by SAM (1,000 permutations, median false significant 6) to be different in L-asparaginase-sensitive and L-asparaginase-resistant cell lines.

 
Using prediction analysis of microarrays, we identified a set of classifier genes (adaptive thresholding, threshold = 1.8, 94 clones) that on cross-validation correctly predicted L-asparaginase sensitivity or resistance in 13 of 16 cell lines (81% correct). The gene with the highest magnitude score, in this classifier, was ASNS. The list of genes that comprise the classifier and their associated scores are available in the Web supplement.

We then investigated whether a similar relationship between baseline gene expression and response to L-asparaginase exists for clinical samples. To do this, we obtained the baseline gene expression for a collection of 28 pediatric precursor B (pre-B) ALL samples. As shown in Fig. 2, unsupervised hierarchical clustering of these samples organizes the samples primarily according to the chromosomal translocations that they harbor [i.e., the presence or absence of t(12;21) producing the TEL/AML1 fusion gene]. This is consistent with a recent study of gene expression in pediatric ALL samples (17). Consistent with this recent report, the TEL/AML1-positive samples have a characteristic expression profile, which includes the consistent relatively high expression of many genes, including EPOR, GNG11, TCFL5, and ARHGEF4.



View larger version (40K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Baseline gene expression in clinical samples of pediatric pre-B ALL. A, for each clinical sample, total RNA and reference total RNA (Universal Human Reference) were treated with DNase (DNA-free), linearly amplified, fluorescently labeled (with Cy5-dUTP or Cy3-dUTP), and comparatively hybridized to an array containing ~42,000 features representing ~30,000 unique UniGene clusters. We replaced the gene expression levels obtained from duplicate arrays by their mean expression level. Variation in expression is displayed as a variation in color (14) for the 505 clones (representing 451 genes) for which the log2 intensity ratio differed by at least 2 from its mean on at least 2 arrays. The color scale extends from 0.25- to 4.0-fold of the mean (–2.0 to 2.0 in log2 space). Gray, omitted data. The arms of the dendrogram are color coded to indicate the chromosomal translocation associated with each branch: orange, TEL/AML1 positive; black, TEL/AML1 negative. The stripe of color-labeled L-asparaginase LC50 provides a representation of the L-asparaginase LC50 for each sample: bright red, highest LC50 value (L-asparaginase resistant); bright green, lowest LC50 value (L-asparaginase sensitive). B, clustering of clinical samples using the genes associated with response to L-asparaginase in cell lines. Of the 52 clones displayed in Fig. 1A, 46 clones (representing 45 genes) were well measured in the clinical data set. Only named genes are labeled. C, ASNS expression in the clinical samples studied. Samples are placed in order of increasing L-asparaginase LC50.

 
To see if the genes associated with response to L-asparaginase in the cell lines are also associated with response to L-asparaginase in the clinical samples, we clustered the clinical samples based on their expression of these genes. The results of this clustering are shown in Fig. 2. It is clear that these genes are not consistently associated with response to L-asparaginase in the clinical samples. The prediction analysis of microarray classifier developed for the cell lines also was unable to correctly classify the majority of clinical samples (9/23 = 39% correct).

We used several supervised analyses, including t test and SAM, to search for genes whose baseline expression was correlated with response to L-asparaginase in the clinical samples. None of these statistical approaches identified any genes that had a statistically significant association with response to L-asparaginase. In particular, ASNS was not associated with response to L-asparaginase (t test, P = 0.4). Similarly, prediction analysis of microarray was not able to identify genes that accurately classify samples as sensitive or resistant to L-asparaginase.

Changes in Gene Expression after In vitro Exposure to L-Asparaginase. To gain greater insight into the mechanisms of response of ALL cells to L-asparaginase, we next investigated the changes in gene expression that occur after exposure to L-asparaginase. We selected several cell lines, both L-asparaginase sensitive and L-asparaginase resistant, for this study. We specifically sought pairs of cell lines with the same chromosomal translocation, one of which was sensitive to L-asparaginase and the other was resistant. However, for t(12;21), we were able to identify only one cell line. The results of this study are shown in Fig. 3. It is clear that all cell lines exhibit very similar global changes in gene expression after exposure to L-asparaginase. In particular, the changes in gene expression are very similar in L-asparaginase-sensitive and L-asparaginase-resistant cell lines. In addition, the changes in expression are very similar in the cell lines harboring different chromosomal translocations. Figure 3B shows that in the cell lines many genes change after exposure to L-asparaginase and that these changes peak at 8 hours after L-asparaginase exposure.



View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Changes in gene expression in cell lines and pediatric ALL samples after exposure to L-asparaginase. A, variation in expression of 972 clones (representing 858 genes) after exposure to L-asparaginase for which the log2 intensity ratio differed by at least 1 from baseline on at least 4 of the cell line arrays. (The clinical sample data were not used to select the genes in this figure. Data for these genes for the clinical samples are displayed for comparison.) The color scale extends from 0.35- to 2.8-fold of the mean (–1.5 to 1.5 in log2 space). Gray, data that were omitted because these did not meet the quality criteria as described in Materials and Methods. The expression of ASNS is shown separately. B, fraction of well-measured clones that differ from the baseline by at least 1.0 at each time point studied. Data are represented in the form of box plots. Upper and lower boundaries of the box, interquartile range (the range between 25th and 75th percentiles); line in the middle of the box, the median. Values (1.5 x interquartile range) above the 75th percentile or below the 25th percentile are plotted individually. C, variation in expression of the 1,015 clones (representing 915 genes) identified by SAM (1,000 permutations, median false significant 26 clones) that had consistent expression in the cell lines at 8 hours and in the clinical samples at 24 hours compared with baseline. Genes are ordered by hierarchical clustering. The names of only some selected genes are displayed. D, legend for A and C. The stripe of color-labeled L-asparaginase LC50 provides a representation of the L-asparaginase LC50 for each sample: bright red, highest LC50 value (L-asparaginase resistant); bright green, lowest LC50 value (L-asparaginase sensitive).

 
Of note, ASNS has an interesting pattern of response in the cell lines after exposure to L-asparaginase. As shown in Fig. 3, cell lines that are resistant to L-asparaginase have an increase in expression of ASNS after exposure to L-asparaginase, whereas cell lines that are sensitive to L-asparaginase do not.

Next, we investigated the changes in gene expression in clinical ALL samples after in vitro exposure to L-asparaginase. Because the number of cells that we could obtain from clinical ALL samples was relatively small, we could not carry out the measurements at multiple time points as we did for the cell lines. Instead, we examined only one time point. Initially, based on our cell line data, we chose to determine gene expression in the clinical samples at 8 hours after exposure to L-asparaginase. However, as shown in Fig. 3, at 8 hours after in vitro exposure to L-asparaginase, there were very minimal differences in gene expression compared with the baseline. We hypothesized that these relatively small differences in gene expression might reflect the relatively slower metabolic and proliferative rates in clinical samples in vitro compared with cell lines. Consequently, for the remaining clinical samples, we chose to determine gene expression at 24 hours after exposure to L-asparaginase. This later time point was generally associated with larger differences in gene expression between L-asparaginase exposed and nonexposed than we observed at 8 hours, as shown in Fig. 3. The number of genes that changed substantially after 24-hour exposure to L-asparaginase in the clinical samples was comparable with the changes we saw in the cell lines after 8-hour exposure. We also investigated whether testing after 40-hour exposure to L-asparaginase would yield even larger magnitude differences in gene expression; however, we found that the differences were not greater than at 24 hours (data not shown).

We systematically searched for genes that were consistently induced or repressed in both cell lines and clinical samples after exposure to L-asparaginase. To do this, we used SAM to identify genes that had consistent expression in the cell lines at 8 hours and in the clinical samples at 24 hours compared with baseline. Using this analysis (1,000 permutations), we found 1,015 clones (median false significant 26 clones). The results of this analysis are shown in Fig. 3C with selected gene names displayed.

Next, we systematically searched for genes that behaved differently in the cell lines and clinical samples after exposure to L-asparaginase. To do this, we again used SAM, but this time to identify genes that had consistently different expression in the cell lines at 8 hours compared with the clinical samples at 24 hours. Using this analysis (1,000 permutations), we found 1,019 clones (median false significant 35 clones). The result of this analysis is contained in the Supplemental Materials and the Web supplement.

Finally, we systematically searched for genes that behaved differently in L-asparaginase-resistant ALL compared with L-asparaginase-sensitive ALL. First, we considered the cell lines alone. SAM identified 157 clones (1,000 permutations, median false significant 10) that were consistently different at 8 hours in the L-asparaginase-resistant cell lines compared with the L-asparaginase-sensitive cell lines. The gene with highest score in this analysis was ASNS, which as we described above was consistently increased in the L-asparaginase-resistant cell lines but not in the L-asparaginase-sensitive cell lines. The full list of genes identified by this analysis is available in the Web supplement. We next considered the data for the clinical samples alone. SAM found no genes to be consistently different between L-asparaginase-resistant and L-asparaginase-sensitive clinical samples at 24 hours. We also asked if there was a difference in the number of genes that changed after exposure to L-asparaginase in the sensitive samples compared with the resistant samples. However, we found that there was no difference (t test, P = 0.2) between sensitive and resistant clinical samples at 24 hours in the fraction of well-measured genes that were substantially different (differed by at least 1.0 in log2 space) from baseline.

To illustrate some of the themes that emerge from this study, in Fig. 3C, we include the names of some selected genes. The data in this figure reveal some recognizable and coordinated biological responses to L-asparaginase exposure. In particular, a large number of tRNA synthetase genes and solute carrier family member genes exhibit consistently increased expression. In addition, we find that members of the CCAAT/enhancer binding protein and activating transcription factor families exhibit consistent changes in gene expression. Several genes related to proliferation have decreased expression, including MCM3, MCM4, MCM5, CHEK2, and TERF2. Genes involved in folate metabolism display an interesting pattern of expression: DHFR and MTHFD1 decrease after exposure to L-asparaginase, whereas MTHFD2 (NMDMC) increases.

An important difference between cell lines and clinical samples is revealed in the pattern of expression of ASNS, as shown in Fig. 3A. In contrast to the cell lines, for the clinical samples, there is no correlation between increased expression of ASNS after exposure to L-asparaginase and sensitivity or resistance to L-asparaginase.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
This study provides a rich overview of the relationship between gene expression patterns in ALL and in vitro response to L-asparaginase.

Our examination of the association between baseline gene expression and in vitro response to L-asparaginase reveals important differences between the clinical samples and the cell lines studied. In particular, in the cell lines, a gene expression signature could be identified that distinguished in vitro L-asparaginase sensitivity from L-asparaginase resistance. This signature included an association between high baseline expression of ASNS and resistance to L-asparaginase. This observation is consistent with previous studies in other cell lines (7, 18). However, in the clinical samples, this gene expression signature was not associated with response to L-asparaginase. Specifically, baseline expression of ASNS in the clinical samples did not correlate with in vitro response to L-asparaginase. This observation is consistent with another recent study (9). Furthermore, in the clinical samples, no simple relationship could be identified between baseline gene expression and in vitro response to L-asparaginase. This suggests that no single consistent mechanism explains in vitro L-asparaginase resistance in clinical samples. Instead, it is likely that multiple different mechanisms of resistance are clinically relevant.

To gain further insight into the response to L-asparaginase, we obtained a genome-wide view of the changes in gene expression that occur after in vitro exposure to L-asparaginase. When we considered the cell lines alone, a very consistent pattern of change in gene expression was obtained. Specifically, we found that the predominant changes in gene expression did not differ between cell lines with different chromosomal translocation or between cell lines that were L-asparaginase sensitive or resistant. Interestingly, we did find that the expression ASNS increased after exposure to L-asparaginase in the L-asparaginase-resistant cell lines but not in the L-asparaginase-sensitive cell lines.

When the changes in gene expression after exposure to L-asparaginase in the cell lines were compared with the changes in clinical samples, several important features became apparent. First, we found that the changes in gene expression occurred more slowly in the clinical samples compared with the cell lines. As judged by the fraction of genes substantially changed, we found that 8 hours in the cell lines were roughly equivalent to 24 hours in the clinical samples. This likely reflects the slower metabolic and proliferative rates of clinical samples in vitro compared with cell lines.

By treating the 8-hour time point in the cell lines and the 24-hour time point in the clinical samples as equivalent, we identified a large number of genes that exhibited consistent behavior, indicating that in many respects cell lines and clinical ALL samples share a common response to L-asparaginase. However, we also found that a similar number of genes exhibited consistently different changes after exposure to L-asparaginase in cell lines compared with clinical samples. This suggests that great care is required in using cell lines as a model system to investigate features of ALL. Some of these differences in gene expression may be related to the fact that primary ALL cells do not survive in culture for prolonged periods of time, whereas cell lines continue to proliferate in culture.

To our knowledge, a genome-wide investigation of the changes that occur after amino acid starvation in human cells has not been reported previously. However, such studies have been reported in yeast (19, 20). These studies identified changes in gene expression that overlap with the changes observed in our study. For example, as in our study, in these studies, it was found that many tRNA synthetases and amino acid transporters had increased expression with amino acid starvation.

Studies have been reported of the expression of individual genes in mammalian model systems following amino acid starvation. For example, in human fibroblasts, amino acid starvation leads to increased expression of system A (e.g., SLC38A1 and SLC38A2) and system ASC (e.g., SLC1A4 and SLC1A5) amino acid transporters (21). Amino acid starvation also induces the expression of DDIT3 (CHOP) and CCAAT/enhancer binding protein-ß in rat hepatoma cell lines (22) and DDIT3 in several different human cell lines (23). Similar changes were observed in our study.

In addition, studies have been reported of the changes in expression of individual genes in human cell lines following in vitro exposure to L-asparaginase. In vitro L-asparaginase results in undetectable levels of both extracellular and intracellular asparagine (24). After in vitro exposure to L-asparaginase, it has been found that members of the amino transporter families (systems A and ASC) have increased expression in cell lines (24). This observation is clearly borne out in our study in both cell lines and clinical samples.

Finally, in our study, several genes associated with proliferation have decreased expression after exposure to L-asparaginase. These include MCM3, MCM4, MCM5, CDC23, and TERF2. Although this would be expected following amino acid deprivation, to our knowledge, the pattern of expression of these genes has not been reported under such conditions.

Thus, the study reported here identifies a consistent gene expression response in both cell lines and clinical ALL samples that overlaps with previous studies of amino acid starvation or L-asparaginase exposure in model systems. This overlap suggests that there is a consistent response to amino acid starvation exhibited in a wide variety of eukaryotic cell types. Our study provides a comprehensive characterization of this response in clinical ALL samples.

Of course, in addition to confirming the expected behavior of several genes after in vitro exposure to L-asparaginase, our study also identifies a large number of genes whose expression was previously not known to be affected by L-asparaginase exposure or amino acid starvation. As an example, MTHFD2 (NMDMC), the human homologue of yeast MIS1 (25), was among the genes with the most consistently increased expression after exposure to L-asparaginase in both cell lines and clinical samples. MTHFD2 knockout mice die at embryonic day 12.5 with impaired establishment of erythropoiesis in the fetal liver (26). We found that other genes involved in folate metabolism (DHFR and MTHFD1) decreased after exposure to L-asparaginase. Interestingly, it has been observed in a mouse lymphoma model that pretreatment with L-asparaginase reduces the efficacy of methotrexate (27). Our results raise the possibility that this interaction reflects the effect of L-asparaginase on the folate pathway as reflected by the consistent induction of MTHFD2 and decreased expression of DHFR and MTHFD1. These results also lead us to speculate that targeted inhibition of critical enzymes in the folate pathway, such as MTHFD2, may enhance the efficacy of L-asparaginase.

In at least one important respect, our study reveals that the response to L-asparaginase in ALL is considerably more complex than previous studies have suggested. Although our results for the cell lines show that induction of ASNS after exposure to L-asparaginase is associated with resistance to L-asparaginase, this is not the case for the clinical samples. Instead, we found that some of the L-asparaginase-sensitive samples exhibited significant induction of ASNS, whereas some of the L-asparaginase-resistant samples did not. This is in contrast to previous studies, most of which were in cell lines, which have generally emphasized the role of ASNS induction in L-asparaginase resistance (6, 7) . Specifically, our study shows that the induction of ASNS expression does not explain L-asparaginase resistance in the clinical samples. A recent report came to a similar conclusion (9).

It is interesting to note that the changes in gene expression after exposure to L-asparaginase are very similar in all of the clinical samples regardless of whether they were L-asparaginase sensitive or resistant. In particular, all of the clinical samples exhibit gene expression changes associated with amino acid starvation. This indicates that over the time scale studied, both L-asparaginase-sensitive and L-asparaginase-resistant samples are responding to a similar stress due to L-asparaginase. This suggests that mechanisms of resistance to L-asparaginase do not take effect immediately but require some time to be induced.

In summary, the study presented here provides a comprehensive view of the relationship between in vitro L-asparaginase response and genome-wide gene expression patterns in ALL. This study identifies a consistent pattern of gene expression changes in cell lines and clinical samples that overlaps with the response to amino acid starvation in many model systems. However, these changes occur more slowly in clinical samples than in cell lines. This study also identifies several gene expression features in clinical samples that were not predicted by studies in model systems. Importantly, this study shows that mechanisms of resistance, other than ASNS induction, are common and clinically important. In addition, there may be important post-transcriptional mechanisms involved in resistance to L-asparaginase, and our study would not have identified these. Because the L-asparaginase-resistant clinical samples also exhibit gene expression changes that seem to reflect a response to amino acid starvation, targeting particular genes within this response may provide a novel approach to overcome L-asparaginaseresistance.


   Acknowledgments
 
Grant support: NIH grants CA92326 (L.M. Boxer) and CA85129 (Patrick O. Brown), Howard Hughes Medical Institute (Patrick O. Brown), and Howard Hughes Postdoctoral Fellowship for Physicians and NIH Mentored Clinical Scientist Award grant K08CA095563 (B.M. Fine).

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 Patrick O. Brown for support and advice, Drs. Michael Cleary and Yoshinobu Matsuo for gifts of cell lines, the members of the Brown and Boxer laboratories for helpful discussions, the Stanford Functional Genomics Facility for production of microarrays, and the Stanford Microarray Database, especially Janos Demeter, for providing a repository for the primary data and for hosting the companion Web site for this article.

Received 7/17/04. Revised 10/23/04. Accepted 10/31/04.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Pui CH, Evans WE. Acute lymphoblastic leukemia. N Engl J Med 1998;339:605–15.[Free Full Text]
  2. Kaspers GJ, Veerman AJ, Pieters R, et al. In vitro cellular drug resistance and prognosis in newly diagnosed childhood acute lymphoblastic leukemia. Blood 1997;90:2723–9.[Abstract/Free Full Text]
  3. Asselin BL, Kreissman S, Coppola, DJ, et al. Prognostic significance of early response to a single dose of asparaginase in childhood acute lymphoblastic leukemia. J Pediatr Hematol Oncol 1999;21:6–12.[CrossRef][Medline]
  4. Chesson BD. Miscellaneous Chemotherapeutic agents. In: DeVita VT, Hellman S, Rosenberg SA, editors. Cancer: principles and practice of oncology. 6th ed. Philadelphia: Lippincott Williams & Wilkins; 2001. p. 452.
  5. Haskell CM, Canellos GP. L-Asparaginase resistance in human leukemia-asparagine synthetase. Biochem Pharmacol 1969;18:2578–80.[CrossRef][Medline]
  6. Aslanian AM, Fletcher BS, Kilberg MS. Asparagine synthetase expression alone is sufficient to induce L-asparaginase resistance in MOLT-4 human leukaemia cells. Biochem J 2001;357:321–8.[CrossRef][Medline]
  7. Peng H, Shen N, Qian L, et al. Hypermethylation of CpG islands in the mouse asparagine synthetase gene: relationship to asparaginase sensitivity in lymphoma cells. Partial methylation in normal cells. Br J Cancer 2001;85:930–5.[CrossRef][Medline]
  8. Dubbers A, Wurthwein G, Muller HJ, et al. Asparagine synthetase activity in paediatric acute leukaemias: AML-M5 subtype shows lowest activity. Br J Haematol 2000;109:427–9.[CrossRef][Medline]
  9. Stams WA, den Boer ML, Beverloo HB, et al. Sensitivity to L-asparaginase is not associated with expression levels of asparagine synthetase in t(12;21)+ pediatric ALL. Blood 2003;101:2743–7.[Abstract/Free Full Text]
  10. Pieters R, Loonen AH, Huismans DR, et al. In vitro drug sensitivity of cells from children with leukemia using the MTT assay with improved culture conditions. Blood 1990;76:2327–36.[Abstract/Free Full Text]
  11. Kaspers GJ, Kardos G, Pieters R, et al. Different cellular drug resistance profiles in childhood lymphoblastic and non-lymphoblastic leukemia: a preliminary report. Leukemia 1994;8:1224–9.[Medline]
  12. Wang E, Miller LD, Ohnmacht GA, Liu ET, Marincola FM. High-fidelity mRNA amplification for gene profiling. Nat Biotechnol 2000;18:457–9.[CrossRef][Medline]
  13. Schena M, Shalon D, Davis RW, Brown PO. Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 1995;270:467–70.[Abstract/Free Full Text]
  14. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci U S A 1998;95:14863–8.[Abstract/Free Full Text]
  15. Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci U S A 2001;98:5116–21.[Abstract/Free Full Text]
  16. Tibshirani R, Hastie T, Narasimhan B, Chu G. Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proc Natl Acad Sci U S A 2002;99:6567–72.[Abstract/Free Full Text]
  17. Fine BM, Stanulla M, Schrappe M, et al. Gene expression patterns associated with recurrent chromosomal translocations in acute lymphoblastic leukemia. Blood 2004;103:1043–9.[Abstract/Free Full Text]
  18. Hutson RG, Kitoh T, Moraga Amador DA, et al. Amino acid control of asparagine synthetase: relation to asparaginase resistance in human leukemia cells. Am J Physiol 1997;272:C1691–9.
  19. Gasch AP, Spellman PT, Kao CM, et al. Genomic expression programs in the response of yeast cells to environmental changes. Mol Biol Cell 2000;11:4241–57.[Abstract/Free Full Text]
  20. Natarajan K, Meyer MR, Jackson BM, et al. Transcriptional profiling shows that Gcn4p is a master regulator of gene expression during amino acid starvation in yeast. Mol Cell Biol 2001;21:4347–68.[Abstract/Free Full Text]
  21. Gazzola GC, Dall'Asta V, Guidotti GG. Adaptive regulation of amino acid transport in cultured human fibroblasts. Sites and mechanism of action. J Biol Chem 1981;256:3191–8.[Abstract/Free Full Text]
  22. Marten NW, Burke EJ, Hayden JM, Straus DS. Effect of amino acid limitation on the expression of 19 genes in rat hepatoma cells. FASEB J 1994;8:538–44.[Abstract]
  23. Bruhat A, Jousse C, Wang XZ, et al. Amino acid limitation induces expression of CHOP, a CCAAT/enhancer binding protein-related gene, at both transcriptional and post-transcriptional levels. J Biol Chem 1997;272:17588–93.[Abstract/Free Full Text]
  24. Aslanian AM, Kilberg MS. Multiple adaptive mechanisms affect asparagine synthetase substrate availability in asparaginase-resistant MOLT-4 human leukaemia cells. Biochem J 2001;358:59–67.[CrossRef][Medline]
  25. Yang XM, MacKenzie RE. NAD-dependent methylenetetrahydrofolate dehydrogenase-methenyltetrahydrofolate cyclohydrolase is the mammalian homolog of the mitochondrial enzyme encoded by the yeast MIS1 gene. Biochemistry 1993;32:11118–23.[CrossRef][Medline]
  26. Di Pietro E, Sirois J, Tremblay ML, MacKenzie RE. Mitochondrial NAD-dependent methylenetetrahydrofolate dehydrogenase-methenyltetrahydrofolate cyclohydrolase is essential for embryonic development. Mol Cell Biol 2002;22:4158–66.[Abstract/Free Full Text]
  27. Capizzi RL, Summers WP, Bertino JR. L-Asparaginase induced alteration of amethopterin (methotrexate) activity in mouse leukemia L5178Y. Ann N Y Acad Sci 1971;186:302–11.[Medline]
  28. Hurwitz R, Hozier J, LeBien T, et al. Characterization of a leukemic cell line of the pre-B phenotype. Int J Cancer 1979;23:174–80.[Medline]
  29. Wlodarska I, Aventin A, Ingles-Esteve J, et al. A new subtype of pre-B acute lymphoblastic leukemia with t(5;12)(q31q33;p12), molecularly and cytogenetically distinct from t(5;12) in chronic myelomonocytic leukemia. Blood 1997;89:1716–22.[Abstract/Free Full Text]
  30. Rosenfeld C, Goutner A, Choquet C, et al. Phenotypic characterisation of a unique non-T, non-B acute lymphoblastic leukaemia cell line. Nature 1977;267:841–3.[CrossRef][Medline]
  31. Uphoff CC, MacLeod RA, Denkmann SA, et al. Occurrence of TEL-AML1 fusion resulting from (12;21) translocation in human early B-lineage leukemia cell lines. Leukemia 1997;11:441–7.[CrossRef][Medline]
  32. Findley HW Jr, Cooper MD, Kim TH, Alvarado C, Ragab AH. Two new acute lymphoblastic leukemia cell lines with early B-cell phenotypes. Blood 1982;60:1305–9.[Abstract/Free Full Text]
  33. Barker PE, Carroll AJ, Cooper MD. t(1;19)(q23;p13) in pre-B acute lymphocytic leukemia cell line 697. Cancer Genet Cytogenet 1987;25:379–80.[CrossRef][Medline]
  34. Nourse J, Mellentin JD, Galili N, et al. Chromosomal translocation t(1;19) results in synthesis of a homeobox fusion mRNA that codes for a potential chimeric transcription factor. Cell 1990;60:535–45.[CrossRef][Medline]
  35. Kamps MP, Murre C, Sun XH, Baltimore D. A new homeobox gene contributes the DNA binding domain of the t(1;19) translocation protein in pre-B ALL. Cell 1990;60:547–55.[CrossRef][Medline]
  36. Jack I, Seshadri R, Garson M, et al. RCH-ACV: a lymphoblastic leukemia cell line with chromosome translocation 1;19 and trisomy 8. Cancer Genet Cytogenet 1986;19:261–9.[CrossRef][Medline]
  37. Ohyashiki K, Fujieda H, Miyauchi J, et al. Establishment of a novel heterotransplantable acute lymphoblastic leukemia cell line with a t(17;19) chromosomal translocation the growth of which is inhibited by interleukin-3. Leukemia 1991;5:322–31.[Medline]
  38. Hunger SP, Ohyashiki K, Toyama K, Cleary ML. Hlf, a novel hepatic bZIP protein, shows altered DNA-binding properties following fusion to E2A in t(17;19) acute lymphoblastic leukemia. Genes Dev 1992;6:1608–20.[Abstract/Free Full Text]
  39. Inaba T, Roberts WM, Shapiro LH, et al. Fusion of the leucine zipper gene HLF to the E2A gene in human acute B-lineage leukemia. Science 1992;257:531–4.[Abstract/Free Full Text]
  40. Naumovski L, Morgan R, Hecht F, et al. Philadelphia chromosome-positive acute lymphoblastic leukemia cell lines without classical breakpoint cluster region rearrangement. Cancer Res 1988;48:2876–9.[Abstract/Free Full Text]
  41. Matsuo Y, Ariyasu T, Ohmoto E, Kimura I, Minowada J. Bi-phenotypic t(9;22)-positive leukemia cell lines from a patient with acute leukemia: NALM-20, established at the onset; and NALM-21, NALM-22 and NALM-23, established after relapse. Hum Cell 1991;4:335–8.[Medline]
  42. Matsuo Y, Drexler HG. Establishment and characterization of human B cell precursor-leukemia cell lines. Leuk Res 1998;22:567–79.[CrossRef][Medline]
  43. Pegoraro L, Matera L, Ritz J, et al. Establishment of a Ph1-positive human cell line (BV173). J Natl Cancer Inst 1983;70:447–53.
  44. Hermans A, Gow J, Selleri L, et al. bcr-abl oncogene activation in Philadelphia chromosome-positive acute lymphoblastic leukemia. Leukemia 1988;2:628–33.[Medline]
  45. Stong RC, Korsmeyer SJ, Parkin JL, Arthur DC, Kersey JH. Human acute leukemia cell line with the t(4;11) chromosomal rearrangement exhibits B lineage and monocytic characteristics. Blood 1985;65:21–31.[Abstract/Free Full Text]
  46. Gu Y, Nakamura T, Alder H, et al. The t(4;11) chromosome translocation of human acute leukemias fuses the ALL-1 gene, related to Drosophila trithorax, to the AF-4 gene. Cell 1992;71:701–8.[CrossRef][Medline]
  47. Lange B, Valtieri M, Santoli D, et al. Growth factor requirements of childhood acute leukemia: establishment of GM-CSF-dependent cell lines. Blood 1987;70:192–9.[Abstract/Free Full Text]
  48. Tkachuk DC, Kohler S, Cleary ML. Involvement of a homolog of Drosophila trithorax by 11q23 chromosomal translocations in acute leukemias. Cell 1992;71:691–700.[CrossRef][Medline]
  49. Bene MC, Castoldi G, Knapp W, et al. Proposals for the immunological classification of acute leukemias. European Group for the Immunological Characterization of Leukemias (EGIL). Leukemia 1995;9:1783–6.[Medline]



This article has been cited by other articles:


Home page
 BloodHome page
G. J. L. Kaspers
Old drug, new lessons
Blood, May 7, 2009; 113(19): 4480 - 4481.
[Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
P. L. Lorenzi, J. Llamas, M. Gunsior, L. Ozbun, W. C. Reinhold, S. Varma, H. Ji, H. Kim, A. A. Hutchinson, E. C. Kohn, et al.
Asparagine synthetase is a predictive biomarker of L-asparaginase activity in ovarian cancer cell lines
Mol. Cancer Ther., October 1, 2008; 7(10): 3123 - 3128.
[Abstract] [Full Text] [PDF]

Home page
 ASH Education BookHome page
D. Campana
Molecular Determinants of Treatment Response in Acute Lymphoblastic Leukemia
Hematology, January 1, 2008; 2008(1): 366 - 373.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
M. H. Kang, Y. H. Kang, B. Szymanska, U. Wilczynska-Kalak, M. A. Sheard, T. M. Harned, R. B. Lock, and C. P. Reynolds
Activity of vincristine, L-ASP, and dexamethasone against acute lymphoblastic leukemia is enhanced by the BH3-mimetic ABT-737 in vitro and in vivo
Blood, September 15, 2007; 110(6): 2057 - 2066.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
S. S. Winter, Z. Jiang, H. M. Khawaja, T. Griffin, M. Devidas, B. L. Asselin, and R. S. Larson
Identification of genomic classifiers that distinguish induction failure in T-lineage acute lymphoblastic leukemia: a report from the Children's Oncology Group
Blood, September 1, 2007; 110(5): 1429 - 1438.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
Y. Ge, C. L. Haska, K. LaFiura, M. Devidas, S. B. Linda, M. Liu, R. Thomas, J. W. Taub, and L. H. Matherly
Prognostic Role of the Reduced Folate Carrier, the Major Membrane Transporter for Methotrexate, in Childhood Acute Lymphoblastic Leukemia: A Report from the Children's Oncology Group
Clin. Cancer Res., January 15, 2007; 13(2): 451 - 457.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
R. B. Reinert, L. M. Oberle, S. A. Wek, P. Bunpo, X. P. Wang, I. Mileva, L. O. Goodwin, C. J. Aldrich, D. L. Durden, M. A. McNurlan, et al.
Role of Glutamine Depletion in Directing Tissue-specific Nutrient Stress Responses to L-Asparaginase
J. Biol. Chem., October 20, 2006; 281(42): 31222 - 31233.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
I. M. Appel, M. L. den Boer, J. P. P. Meijerink, A. J. P. Veerman, N. C. M. Reniers, and R. Pieters
Up-regulation of asparagine synthetase expression is not linked to the clinical response L-asparaginase in pediatric acute lymphoblastic leukemia
Blood, June 1, 2006; 107(11): 4244 - 4249.
[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 Email this article to a friend
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 Fine, B. M.
Right arrow Articles by Boxer, L. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fine, B. M.
Right arrow Articles by Boxer, L. M.

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