Journal of Applied Physiology Watch the video to see how APS reaches out to developing nations.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 95: 1352-1360, 2003; doi:10.1152/japplphysiol.00214.2003
8750-7587/03 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (9)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Watters, J. W.
Right arrow Articles by McLeod, H. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Watters, J. W.
Right arrow Articles by McLeod, H. L.

MOUSE PHENOME PROJECT

A mouse-based strategy for cyclophosphamide pharmacogenomic discovery

James W. Watters,1 Ellen F. Kloss,2 Daniel C. Link,1,3 Timothy A. Graubert,1,3 and Howard L. McLeod1,3,4,5

Departments of 1Medicine, 2Dermatology, 5Genetics, and 4Molecular Biology and Pharmacology, Washington University School of Medicine, and 3The Siteman Cancer Center, St. Louis, Missouri 63110

Submitted 3 March 2003 ; accepted in final form 10 April 2003

ABSTRACT

Genome-wide mapping approaches are needed to more fully understand the genetic basis of chemotherapy response. Because of technical and ethical limitations, cancer pharmacogenomics has not yet benefited from traditional robust familial genetic strategies. We have therefore explored the use of the inbred mouse as a genetic model system in which to study response to the cytotoxic agent cyclophosphamide. Multiple phenotypes have been assessed in response to cyclophosphamide in up to 19 inbred mouse strains, including in vitro hematopoietic progenitor cell toxicity and the mobilization of hematopoietic progenitor cells into peripheral blood. Hematopoietic progenitor cell toxicity in vitro varied 2-fold among strains, whereas in vivo progenitor cell mobilization varied almost 75-fold among strains. Males mobilized more hematopoietic progenitor cells than did females, and the low-mobilization phenotype was dominant to the high-mobilization phenotype in F1 hybrid animals. In an initial attempt to analyze candidate genes, genetic variation was assessed in three cytochrome P-450 genes involved in the metabolism of cyclophosphamide. Resequencing of eight strains identified 26 polymorphisms in these genes that may influence response to cyclophosphamide. Distinct regions of high- and low-polymorphism rates were identified, and two common haplotypes were shared among the strains for each gene that exhibited variation. This phenotypic and genotypic variation among inbred strains provides a framework for cyclophosphamide pharmacogenomic discovery.

pharmacogenetics; mobilization

VARIATION IN THE EFFICACY and toxicity of chemotherapeutic agents is consistently observed across human populations (6). Although many clinical variables have been associated with drug response (age, gender, diet, organ function, tumor biology), genetic differences in drug disposition and drug targets can have a great impact on treatment outcome (4-6). Pharmacogenomics is especially important for oncology, as severe systemic toxicity and unpredictable efficacy are hallmarks of cancer therapies. A better understanding of the genetic determinants of chemotherapy response would increase our ability to prospectively identify patients at risk for severe toxicity or those likely to benefit from standard chemotherapy regimens. Bearing in mind the numerous gene products that may interact with a drug as it moves through the body, it is conceivable that virtually any gene in the human genome might be a gene that modifies drug response (17). Therefore, to more fully understand the genetic basis of drug response, genome-wide mapping approaches are needed in which no a priori assumptions are made about genes or genomic regions associated with the drug effect under investigation. However, the genetic study of host factors influencing drug response is made difficult in humans by issues such as environmental variability and the availability of relevant populations (24, 34). Consequently, animal model systems are needed to fully understand the genetic basis of drug response.

The laboratory mouse provides an ideal model system for the study of pharmacogenomics. Well-defined inbred mouse strains, in which genetic factors have segregated and become fixed during inbreeding, facilitate the genome-wide analysis of drug response by reducing genetic complexity (28). Unfortunately, the mouse is currently underutilized in pharmacogenomics research (33). Therefore, we have used resources provided by the Mouse Phenome Database to assess the impact of genetic variation on hematopoietic responses to the cytotoxic chemotherapy drug cyclophosphamide.

Cyclophosphamide is used for the treatment of many types of solid tumors and leukemias, with unpredictable neutropenia being the primary dose-limiting toxicity (2, 14). In addition, cyclophosphamide is used for the mobilization of hematopoietic progenitor cells from the bone marrow into peripheral blood (2, 11, 30). However, mobilizing agents such as cyclophosphamide are not always successful. Therefore, improved understanding of the genetic basis of this variability could lead to more effective strategies for progenitor cell mobilization and cancer treatment. To assess the use of inbred mouse strains as a genetic model system in which to study cyclophosphamide response, we have determined in vitro and in vivo hematopoietic phenotypes in response to this drug in commonly used inbred strains. Our results show interstrain variation in the development of neutropenia and rebound neutrophilia after intraperitoneal injection of cyclophosphamide, as well as in vitro hematopoietic progenitor cell toxicity in response to cyclophosphamide treatment. In addition, the cyclophosphamide-induced mobilization of hematopoietic progenitor cells into peripheral blood varies widely among strains. In an attempt to identify genetic polymorphisms that may influence cyclophosphamide response, interstrain variation in three genes involved in cyclophosphamide metabolism was assessed. These data show that inbred mouse strains are a promising model system for cyclophosphamide pharmacogenomic discovery.

MATERIALS AND METHODS

Mice. The following inbred strains and F1 hybrids were used in this study: C3H/HeJ, C57BL/6J, BALB/cByJ, DBA/2J, FVB/nJ, B6D2F1, CByD2F1, C58/J, NOD/LtJ, MOLF/Ei, SM/J, NZB/BINJ, PERA/Ei, SWR/J, C57L/J, SJL/J, A/J, CAST/Ei, SPRET/Ei, 129/SvImJ, PL/J, and AKR/J. All animals were a generous gift from the Mouse Phenome Database at the Jackson Laboratory (Bar Harbor, ME; http://www.jax.org/phenome). All mice used were between 6 and 10 wk old. All mice were housed in a specific pathogen-free environment and examined daily by veterinary staff for signs of illness. The studies presented here were approved by the Washington University Division of Comparative Medicine, protocol no. 20010354.

Baseline hematopoietic assays. Peripheral blood was obtained from the retro-orbital venous plexus and collected in EDTA-containing vacutainer tubes (Becton Dickinson). Twelve males and twelve females were analyzed for each strain. Automated complete blood count with five-part differential was determined with a programmable murine hematology analyzer (Hemavet 950FS, CDC Technologies, Oxford, CT). For flow cytometric analyses, red blood cells were subjected to hypotonic lysis. The remaining leucocytes were washed in FACS buffer (0.2% BSA, 0.01% NaN3 in PBS) and stained with phycoeryrthrin-conjugated anti-CD34 (BD Pharmingen) or isotype control. We collected 10,000 scattergated events for each sample using a FACScan (BD Biosciences) running CellQuest software.

Assessment of neutropenia. Cyclophosphamide was dissolved in PBS and injected intraperitoneally at doses of 10, 75, or 200 mg/kg in a total volume of 200 µl. Two males and two females from each inbred strain were injected at each dose. Blood from the lateral saphenous vein was collected in EDTA-coated tubes (Microvette CB 300, Sarstedt, Newton, NC) three times per week, and automated complete blood counts with five-part differential were determined. The development of neutropenia and neutrophilic recovery were assessed by monitoring the absolute neutrophil count in peripheral blood over a 3-wk period.

Colony-forming unit assay: granulocyte/monocyte. Bone marrow cells were harvested from two males and two females of each inbred strain and enumerated with a hemacytometer. Cells were diluted to 2 x 105 cells/ml in Iscove's MDM medium containing 2% FCS. A dose of 0.3 ml of this dilution (6 x 104 cells) was added to 3.0 ml of methylcellulose medium supplemented with Iscove's MDM, fetal bovine serum, BSA, human recombinant insulin, human transferrin (iron-saturated), 2-mercaptoethanol, L-glutamine, rmIL-3, rhIL-6, and rm stem cell factor (MethoCult M3534, Stem Cell Technologies, Vancouver, BC, Canada) where rm designates recombinant murine and rh recombinant human. This medium supports growth of granulocyte, monocyte, and granulocyte/monocyte precursors [colony-forming unit-granulocyte, -monocyte, and granulocyte/monocyte (CFU-G, CFU-M, and CFU-GM, respectively)]. Duplicate 1.1-ml cultures were plated in 35-mm dishes and incubated at 37°C in a humidified chamber with 5% CO2. Colonies containing at least 50 cells were counted on day 7. The concentration of 4-hydroxycyclophosphamide (4-HC; the active metabolite of cyclophosphamide) that reduced colony number by 50% relative to untreated control (IC50) was determined by interpolation from a four-parameter logistic regression fit of cytotoxicity data (GraphPad Prism 3.0, GraphPad Software, San Diego, CA).

Hematopoietic progenitor cell mobilization. Cyclophosphamide was dissolved in PBS and administered by intraperitoneal injection at a dose of 200 mg/kg in a total volume of 200 µl in two to four males and two to four females from each inbred strain. Peripheral blood was drawn on day 8 after injection by lateral saphenous vein collection. EDTA-anticoagulated whole blood (20 µl) was subjected to hypotonic lysis and centrifuged for 2 min at 5,000 g. The cell pellet was resuspended in 0.3 ml of Iscove's MDM containing 2% FCS and added to 3.0 ml of MethoCult M3534 medium. Duplicate 1.1-ml cultures were plated in 35-mm dishes and incubated at 37°C in a humidified chamber with 5% CO2. Colonies containing at least 50 cells were counted on day 7.

Single nucleotide polymorphism discovery and genotyping by resequencing. PCR and resequencing of Cyp2c29, Cyp3a13, and Cyp2b10 was performed on the following commonly used inbred strains: C57BL/6J, A/J, AKR/J, FVB/nJ, DBA/2J, C3H/HeJ, BALB/cJ, and C57L/J. PCR and sequencing primers were initially picked with the use of a high-throughput design method, in which we designed two PCR primers and chose one for sequencing. We obtained the genomic sequences with marked intron and exon boundaries (NCBI mouse genome build, November 25, 2002; http://www.ncbi.nlm.nih.gov/mapview/map_search.cgi?chr=mouse_chr.inf) and split the sequence into exons with 600 bases of flanking sequence. If an exon was longer than 800 bases, it was split into fragments shorter than 800 bases. Repetitive and low-complexity sequence was masked with REPEATMASKER, using the rodent-specific and mammalian-wide repeats library (Arian Smit, http://ftp.genome.washington.edu/RM). PCR primers were chosen with PRIMER3 [0.9, Unix version, http://www-genome.wi.mit.edu/genome_software/ (26)]. To avoid regions of sequence known to cause sequencing reaction failure, we determined the optimal PCR primer to use for sequencing with custom software developed in house (31).

To improve the throughput of our reaction, we adopted a new resequencing protocol. This protocol involves adding the primer used for sequencing in 10-fold excess relative to the primer not chosen for sequencing. Primers (Integrated DNA technologies, Coralville, IA) were used at final concentrations of 2.0 and 0.2 µM in the reaction. PCR amplification of DNA was preformed with either a single or pooled sample of 4 ng DNAs. All other steps, including PCR and cycle DNA sequencing, followed the protocol described previously (13). This protocol eliminates the need to add a sequencing primer and remove excess dNTPs after PCR. We identified sequence polymorphisms between strains by aligning analogous sequences using Sequencher 3.0 (GeneCodes, Ann Arbor, MI).

Statistical analyses. Two-way ANOVA and tests for significance were performed with STATISTICA (StatSoft, Tulsa, OK).

RESULTS

Baseline hematology data. To assess pretreatment variation in clinical hematology parameters and provide a framework for interpreting interstrain variation in cyclophosphamide response, levels of pretreatment peripheral white blood cells (WBC) were determined in 19 inbred strains. Analysis of pretreatment total WBC revealed up to threefold variation among strains (Fig. 1). A similar analysis was performed by Peters and Barker of the Jackson Laboratory for the Mouse Phenome Database (data available at http://www.jax.org/phenome). A comparison of the two data sets reveals a high level of consistency. For example, NZB/BINJ has a low WBC count in both data sets (mean WBC ± SD for males of 4.1 1,000/µl ± 1.9 in our study vs. 3.5 1,000/µl ± 0.8), whereas PERA/Ei has a high-WBC count in both data sets (11.8 1,000/µl ± 1.9 in our study vs. 11.5 1,000/µl ± 0.6). A two-way ANOVA revealed a significant strain effect and a significant sex effect for WBC counts (Table 1). A significant strain-by-sex interaction was also detected, in which differences in WBC counts between strains are not the same for males and females. Strain was the variable associated with the greatest influence on WBC count (46.27%).



View larger version (37K):
[in this window]
[in a new window]
  		Fig. 1. Baseline peripheral white blood cell (WBC) counts in multiple inbred strains. WBCs were measured in inbred mice, n = 24 mice per strain, from the Mouse Phenome Database priority A/B strains (n = 19 strains; http://www.jax.org/phenome). Histograms (mean ± SD) are grouped by sex and sorted on the values for the males in each strain. K, thousand.

 

View this table:
[in this window]
[in a new window]
  		Table 1. Two-way ANOVA evaluating resting WBC count among strains

 

In vivo development of neutropenia. Myelosuppression is a common dose-limiting toxicity of cyclophosphamide treatment. We sought to determine whether there were differences in susceptibility to neutropenia after cyclophosphamide treatment between inbred strains. For this purpose, we treated the commonly used strains BALB/cByJ and DBA/2J with varying doses of cyclophosphamide via intraperitoneal injection and monitored the neutrophil response over a 3-wk period in both strains. These two strains were selected because of preliminary reports of differential cyclophosphamide response between these strains in the literature (3, 21, 23, 29). Although no effect was seen with the PBS-only control or at the lowest dose of 10 mg/kg, a single intraperitoneal injection of cyclophosphamide produced the characteristic pattern of neutropenia followed by rebound neutrophilia at doses of 75 and 200 mg/kg (Fig. 2). Interstrain differences were seen in the development of neutropenia, as DBA2/J had a shorter time to nadir (3 days) compared with BALB/cByJ (5 days), at both 75 and 200 mg/kg cyclophosphamide. In addition, DBA/2J had a pronounced increase in the peak and duration of rebound neutrophilia compared with BALB/cByJ, demonstrating that there are considerable strain differences in the in vivo hematopoietic response to cyclophosphamide. To address these issues statistically, we performed unpaired t-tests to assess strain differences in the mean neutrophil counts at day 3 and day 14 after injection of 200 mg/kg cyclophosphamide. There was a significant difference in neutropenia at day 3 (P = 0.014), but the difference in rebound neutrophilia at day 14 did not reach statistical significance (P = 0.056).



View larger version (28K):
[in this window]
[in a new window]
  		Fig. 2. Development of neutropenia and rebound neutrophilia varies between two commonly used inbred strains. Two males and two females from each strain were injected with 200 mg/kg ip cyclophosphamide, and neutrophils were monitored for ~3 wk. Data points represent the average absolute neutrophil count ± SD.

 

Toxicity of bone marrow-derived CFU-GM to 4-HC. To further explore strain differences in susceptibility to cyclophosphamide-induced neutropenia, we employed an in vitro semisolid culture assay to monitor the growth of granulocyte/monocyte progenitor cells (CFUGM) derived from bone marrow. Previous studies have shown a relationship between reduction in CFU-GM colonies and the decrease in absolute neutrophil count in vivo after administration of antineoplastic agents (18-20). We examined CFU-GM growth-suppressive effects caused by in vitro exposure to 4-HC in 11 inbred strains (Fig. 3). 4-HC reduced the frequency of bone marrow CFU-GM in a dose-dependent manner, and the strains displayed twofold variation in the concentration of 4-HC that was required to reduce the frequency of CFU-GM colonies by 50% compared with untreated controls (IC50).



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 3. Toxicity of colony-forming unit-granulocyte/monocyte (CFU-GM) colonies to incubation with 4-hydroxycyclophosphamide (4-HC) in vitro. Bone marrow cells were harvested from 2 males and 2 females of each inbred strain and plated in duplicate methylcellulose cultures. The toxicity curves of the most resistant strain (DBA/2J) and the most susceptible strain (FVB/nJ) are shown in the graph; each data point represents the mean relative CFU-GM frequency at that dose ± SD. The table (right) shows the interpolated IC50 values for each strain measured.

 

Mobilization of hematopoietic progenitor cells into peripheral blood. We then explored the basis of strain differences in the development of rebound neutrophilia in vivo after cyclophosphamide administration. Mobilization of neutrophils and hematopoietic progenitors in response to a variety of stimuli likely depends on shared molecular mechanisms (9, 27). Indeed, cyclophosphamide (along with hematopoietic growth factors and certain chemokines) can be used to mobilize hematopoietic progenitor cells from the bone marrow into peripheral blood in both humans and mice (11, 12, 30). Although knockout studies have shown that the granulocyte-colony stimulating factor is required for the cyclophosphamide-induced mobilization of hematopoietic progenitor cells from the bone marrow into peripheral blood in mice (11, 12), the molecular mechanisms that control the movement of hematopoietic progenitor cells from the bone marrow to peripheral blood are not clearly understood.

Using the CFU-GM assay, we determined the number of hematopoietic progenitor cells circulating in the peripheral blood 8 days after an intraperitoneal injection of cyclophosphamide in 13 different inbred strains. This has previously been shown as the day of maximum CFU-GM number in the peripheral blood after cyclophosphamide treatment in mice (12). Broad variability in the peak hematopoietic progenitor cell response was observed between strains (Fig. 4). FVB/nJ was the lowest responder, with a CFU-GM colony count of 4 ± 2.5 (SD) for males, and SM/J was the highest responder, with a count of 293 ± 25.1 for males. This represents a 73-fold difference between strains. In addition, there was a significant difference between sexes, with males circulating more progenitor cells than females in every strain analyzed. The strains were concordant with respect to in vivo rebound neutrophilia and peripheral blood progenitor responses (BALB/cByJ low responders, DBA/2J high responders). To establish the dominance relationship of this phenotype, we analyzed two different F1 hybrid strains derived from high- and low-responding parents: B6D2F1 (product of C5BL/6J x DBA/2J), and CByD2F1 (product of BALB/cByJ x DBA/2J). In both cases, low response was the dominant phenotype (Fig. 4). A twoway ANOVA revealed a significant strain effect and a significant sex effect for progenitor cell mobilization (Table 2). A significant strain-by-sex interaction was also detected, in which differences in progenitor cell mobilization between strains are not the same for males and females. Strain was the variable associated with the greatest influence on progenitor cell mobilization (65.7%)



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 4. Response to cyclophosphamide varies widely between strains. Peripheral blood CFU-GM colonies were measured in 2-4 males and 2-4 females from each strain 8 days after injection of 200 mg/kg ip cyclophosphamide. Histograms (mean ± SD) are grouped by sex and sorted on the values for the males in each strain.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Two-way ANOVA evaluating hematopoietic progenitor cell mobilization into peripheral blood in response to cyclophosphamide

 

The high- and low-responding strains did not differ in the baseline hematology parameters that we measured in Fig. 1. There was no significant difference in the resting WBC between the low-efficiency responding strain FVB/nJ male mice (mean ± SD of 6.51 ± 1.9 k/µl) and the high-efficiency responding strain SM/J male mice (mean ± SD of 5.19 ± 1.6 1,000/µl, P = 0.08). In addition, the baseline CD34+ frequency in peripheral blood did not differ significantly between FVB/nJ and SM/J males (data not shown).

Analysis of candidate genes. In an initial attempt to correlate genotype with phenotype, three candidate genes were resequenced in strains C57BL/6J, A/J, AKR/J, FVB/nJ, DBA/2J, C3H/HeJ, BALB/cJ, and C57L/J to identify single nucleotide polymorphisms (SNPs) and insertions/deletions that may be associated with cyclophosphamide response. A low degree of variation in in vitro bone marrow CFU-GM toxicity was observed, suggesting the possibility that our observed in vivo differences in cyclophosphamide response were due to differential drug metabolism. Therefore, we identified the UniGene mouse homologues of four human cytochrome P-450 enzymes known to be involved in the metabolism of cyclophosphamide: CYP3A4, CYP3A5, CYP2C9, and CYP2B6 (8, 10, 35). The identified mouse homologues were Cyp3a13 (mouse homologue of CYP3A4 and CYP3A5), Cyp2c29 (mouse homologue of CYP2C9), and Cyp2b10 (mouse homologue of CYP2B6). Figure 5A shows the action of these genes in the cyclophosphamide metabolism pathway. All exons, and an average of 200 bp of flanking intronic sequence, were amplified and sequenced. Sequences were aligned, and base pair positions containing polymorphisms were identified. In total, 26 polymorphisms were identified in these three genes, including 18 intronic polymorphisms, 6 synonymous SNPs, and 1 nonsynonymous coding SNP (Table 3). The nonsynonymous coding SNP causes a methionine to isoleucine change at amino acid position 106 in Cyp2c29. No polymorphisms were identified in the exons, introns, or untranslated regions of Cyp2b10. Interestingly, only two distinct haplotypes were observed in these eight strains for both Cyp2c29 and Cyp3a13. Figure 5B shows the haplotype structure for both Cyp2c29 and Cyp3a13. Strains are either homozygous for the rare allele at each SNP position throughout the gene or homozygous for the common allele at each SNP position throughout the gene. This suggests a limited amount of ancestral recombination within these genes. When SNP genotypes were correlated with hematopoietic progenitor cell mobilization response to cyclophosphamide, no correlation between this phenotype and any SNP genotype was observed (data not shown).



View larger version (31K):
[in this window]
[in a new window]
  		Fig. 5. Analysis of candidate genes. A: selection of candidate genes for influencing cyclophosphamide response. Three cytochrome P-450 genes involved in the metabolism of cyclophosphamide (italics) were selected for single nucleotide polymorphism analysis. These genes are involved in the activation of cyclophosphamide to its active metabolite, 4-HC, and conversion to an inactive metabolite, deschloroethylcyclophosphamide. B: haplotype distribution for Cyp2c29 and Cyp3a13. In the 8 strains sequenced, only 2 distinct haplotypes were found across the entire gene in both cases. Rows represent individual strains, and columns represent identified polymorphisms described in Table 3. Gray squares represent homozygosity for the common allele, and white squares represent homozygosity for the rare allele.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Polymorphisms identified in Cyp2b10, Cyp2c29, and Cyp3a13

 

DISCUSSION

Phenotypic variation between inbred mouse strains provides a useful resource for understanding complex genetic phenotypes. Because drug response phenotypes are not amenable to traditional human genetics strategies, differential drug response between inbred strains provides a unique opportunity for pharmacogenomic discovery. By exploiting these phenotypic differences to identify genes involved in drug response, one can begin to understand the molecular mechanisms behind differential drug response and toxicity (33, 34). We have therefore assessed the use of the mouse as a genetic model system to study response to cyclophosphamide.

In vivo response to cyclophosphamide was found to vary between the commonly used inbred strains BALB/cByJ and DBA/2J. DBA/2J was more sensitive to neutropenia induced by cyclophosphamide treatment and also had an increase in the peak and duration of rebound neutrophilia (Fig. 2). To more fully explore genetic variation in neutrophil toxicity and rebound neutrophilia, these two phenotypes were explored separately in more detail. When CFU-GM toxicity in response to 4-HC was assessed in multiple strains, the in vitro strain distribution pattern of CFU-GM toxicity did not reflect in vivo susceptibility to neutropenia. For example, DBA/2J exhibited the most resistance to reduction in CFU-GM colony number in response to treatment with 4-HC (Fig. 3). In addition, only a twofold variation in 4-HC IC50 values was observed. One possible explanation for this result is that differential drug metabolism is at least partially responsible for differential susceptibility to neutropenia in vivo, a component of cyclophosphamide treatment that may not be represented in this experimental system. It should be noted, however, that strain differences do exist, suggesting that these strains could be used to study the direct action of 4-HC on hematopoietic progenitor cells.

The development of rebound neutrophilia was then explored in vivo by the assessment of hematopoietic progenitor response to cyclophosphamide. Very large variations in the ability of inbred strains to respond to this agent were observed. These variations did reflect the development of rebound neutrophilia in vivo, as DBA/2J was a high-response strain and BALB/cJ was a relatively low-response strain in both systems. Importantly, FVB/nJ (lowest responder) and SM/J (highest responder) had statistically similar baseline values for circulating CD34+ cells and resting WBC counts (Fig. 1). Although it is possible that these observations reflect strain-dependent differences in the capacity of bone marrow stem or progenitor cells to recover from cytotoxic chemotherapy, previous studies in C57BL/6 mice have shown that the increase in circulating progenitors after exposure to cyclophosphamide do reflect mobilization from bone marrow to blood (12, 15, 16). Therefore, observed differences in CFU-GM response likely represent genetic differences in the ability to mobilize hematopoietic progenitor cells. This is the first report of mouse inbred strain differences in hematopoietic stem cell mobilization induced by cyclophosphamide.

Interestingly, similar results are obtained when hematopoietic progenitor cells are mobilized with G-CSF. Inbred strains vary in their ability to mobilize hematopoietic progenitor cells in response to G-CSF, with DBA/2J being the highest responder (25). These differences have recently been exploited to identify two regions of the genome influencing G-CSF-induced progenitor cell mobilization (7). None of the cyclophosphamide metabolism genes that we analyzed maps to these regions. Although our results are consistent with the notion that cyclophosphamide induces hematopoietic progenitor cell mobilization through G-CSF receptor signaling (12), different factors are likely to be operating in the initial phases of progenitor cell mobilization when induced by cyclophosphamide. For example, it has previously been shown that the number of hematopoietic progenitor cells in the bone marrow of G-CSF receptor knockout mice increases after cyclophosphamide treatment (12). Therefore, the defect in cyclophosphamide-induced mobilization in G-CSF receptor knockout mice appears to be a failure to release progenitors from the bone marrow, rather than an inability to cause the proliferation of progenitors in the bone marrow. In addition, it has been shown that cyclophosphamide and G-CSF act in an additive fashion to mobilize hematopoietic progenitor cells in mice (16). It is therefore possible that cyclophosphamide provides a major signal for progenitor cell proliferation in the bone marrow, and G-CSF provides a major signal for progenitor cell migration out of the bone marrow. Nevertheless, our results provide a resource for the discovery of genes involved in hematopoietic progenitor cell mobilization. The identification of these genetic factors will contribute to our understanding of this complex process and may offer opportunities to improve current clinical strategies for progenitor cell mobilization.

In an initial attempt to identify candidate SNPs that may be influencing these cyclophosphamide phenotypes, three cytochrome P-450 genes involved in cyclophosphamide metabolism (Cyp2b10, Cyp2c29, and Cyp3a13) were resequenced in eight inbred strains, and 26 polymorphisms were identified. Interestingly, no polymorphisms were identified in the exons, introns, or untranslated regions of Cyp2b10, but numerous polymorphisms were identified in Cyp2c29 and Cyp3a13. Additionally, only two distinct haplotypes were observed in Cyp2c29 and Cyp3a13 for the eight strains that we resequenced (Fig. 5B). These data confirm similar observations made in a recent landmark study (32), which demonstrated that the mouse genome consists of long segments of either extremely high or extremely low rates of sequence polymorphism and that many loci in the mouse genome consist of only two shared haplotypes among strains. These results suggest that the founder populations from which these strains were derived were quite limited, an observation consistent with genetic evidence from the Y chromosome (1). If this pattern of polymorphism is extended through the whole genome, it is likely that relatively few SNPs will need to be genotyped to infer the genotype of an entire haplotype blocks; therefore, relatively few genotypes will be required to infer the genotypes of all haplotype blocks in the genome. This pattern of genomic variation between strains may facilitate the genetic analysis of complex traits by associating haplotype blocks with particular phenotypes among inbred strains.

Because differential drug metabolism between individuals has a significant impact on differential drug response observed in the clinic (22), we analyzed our identified polymorphisms in Cyp2b10, Cyp2c29, and Cyp3a13 in an attempt to correlate genotype with phenotype. No definite correlation was found between the genotype at any position and cyclophosphamide response. This result is not surprising, as it is likely that a complex phenotype, such as the cyclophosphamide response, is the result of many genes contributing with varying strengths to the overall phenotype (33). Therefore, quantitative trait locus analysis and more complex analysis of candidate genes will be needed to fully understand cyclophosphamide response. Because these drug-metabolizing enzymes participate in the metabolism of many drugs, the polymorphisms that we identified will be of use in other studies of drug response and possibly also as genetic markers in future mouse genetic mapping projects.

The genetic mapping of phenotypic differences between inbred mouse strains is an important approach for the dissection of complex traits. Because the Mouse Phenome Database is a publicly available database of phenotypic variation between inbred mouse strains, the Mouse Phenome Database represents a unique and valuable resource to the scientific community for the identification of genes regulating complex phenotypes. In addition, just as the construction of a SNP map in humans is expected to increase our ability to analyze complex traits, future pharmacogenomic studies in the mouse will benefit greatly by large-scale SNP discovery efforts. However, even with necessary improvements in SNP discovery and genotyping technology, issues such as environmental variability and availability of relevant populations will continue to complicate human genetic studies of drug response. Therefore, our data provide a useful framework for the identification of genes and relevant polymorphisms regulating cyclophosphamide response.

DISCLOSURES

This work was supported in part by a fellowship from the W. M. Keck Foundation and a grant from the National Institutes of Health Pharmacogenetics Research Network (GM-63340) (http://pharmacogenetics.wustl.edu).

ACKNOWLEDGMENTS

We thank Dr. Molly Bogue and the Jackson Laboratory Mouse Phenome Database for the generous contribution of the animals used in this project. The authors also thank Melissa Meucci and Piia Hanson for technical assistance. The assistance of Dr. Bill Shannon, Division of Biostatistics, is greatly appreciated.


   FOOTNOTES
 

Address for reprint requests and other correspondence: H. L. McLeod, Washington Univ. School of Medicine, Dept. of Medicine, Campus Box 8069, St. Louis, MO 63110 (E-mail: hmcleod{at}im.wustl.edu).

FOOTNOTES

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

REFERENCES

  1. Bishop CE, Boursot P, Baron B, Bonhomme F, and Hatat D. Most classical Mus musculus domesticus laboratory mouse strains carry a Mus musculus musculus Y chromosome. Nature 315: 70-72, 1985.[Medline]
  2. Colvin OM. An overview of cyclophosphamide development and clinical applications. Curr Pharm Des 5: 555-560, 1999.[Web of Science][Medline]
  3. Dragani TA, Zunino A, and Sozzi G. Differences in sister chromatid exchange (SCE)-induction in vivo by cyclophosphamide in murine strains. Carcinogenesis 2: 219-222, 1981.[Abstract/Free Full Text]
  4. Evans WE and Johnson JA. Pharmacogenomics: the inherited basis for interindividual differences in drug response. Annu Rev Genom Human Genet 2: 9-39, 2001.
  5. Evans WE and McLeod HL. Pharmacogenomics—drug disposition, drug targets, and side effects. N Engl J Med 348: 538-549, 2003.[Free Full Text]
  6. Evans WE and Relling MV. Pharmacogenomics: translating functional genomics into rational therapeutics. Science 286: 487-491, 1999.[Abstract/Free Full Text]
  7. Hasegawa M, Baldwin TM, Metcalf D, and Foote SJ. Progenitor cell mobilization by granulocyte colony-stimulating factor controlled by loci on chromosomes 2 and 11. Blood 95: 1872-1874, 2000.[Abstract/Free Full Text]
  8. Huang Z, Roy P, and Waxman DJ. Role of human liver microsomal CYP3A4 and CYP2B6 in catalyzing N-dechloroethylation of cyclophosphamide and ifosfamide. Biochem Pharmacol 59: 961-972, 2000.[Web of Science][Medline]
  9. Levesque JP, Hendy J, Takamatsu Y, Simmons PJ, and Bendall LJ. Disruption of the CXCR4/CXCL12 chemotactic interaction during hematopoietic stem cell mobilization induced by GCSF or cyclophosphamide. J Clin Invest 111: 187-196, 2003.[Web of Science][Medline]
  10. Lindley C, Hamilton G, McCune JS, Faucette S, Shord SS, Hawke RL, Wang H, Gilbert D, Jolley S, Yan B, and Le-Cluyse EL. The effect of cyclophosphamide with and without dexamethasone on cytochrome P450 3A4 and 2B6 in human hepatocytes. Drug Metab Dispos 30: 814-822, 2002.[Abstract/Free Full Text]
  11. Link DC. Mechanisms of granulocyte colony-stimulating factor-induced hematopoietic progenitor-cell mobilization. Semin Hematol 37: 25-32, 2000.[Web of Science][Medline]
  12. Liu F, Poursine-Laurent J, and Link DC. The granulocyte colony-stimulating factor receptor is required for the mobilization of murine hematopoietic progenitors into peripheral blood by cyclophosphamide or interleukin-8 but not flt-3 ligand. Blood 90: 2522-2528, 1997.[Abstract/Free Full Text]
  13. Miller RD, Duan S, Lovins EG, Kloss EF, and Kwok PY. Efficient high-throughput resequencing of genomic DNA. Genome Res 13: 717-720, 2003.[Abstract/Free Full Text]
  14. Montillo M, Tedeschi A, O'Brien S, Di Raimondo F, Lerner S, Ferrajoli A, Morra E, and Keating MJ. Phase II study of cladribine and cyclophosphamide in patients with chronic lymphocytic leukemia and prolymphocytic leukemia. Cancer 97: 114-120, 2003.[Web of Science][Medline]
  15. Morrison SJ, Wright DE, and Weissman IL. Cyclophosphamide/granulocyte colony-stimulating factor induces hematopoietic stem cells to proliferate prior to mobilization. Proc Natl Acad Sci USA 94: 1908-1913, 1997.[Abstract/Free Full Text]
  16. Neben S, Marcus K, and Mauch P. Mobilization of hematopoietic stem and progenitor cell subpopulations from the marrow to the blood of mice following cyclophosphamide and/or granulocyte colony-stimulating factor. Blood 81: 1960-1967, 1993.[Abstract/Free Full Text]
  17. Nebert DW. Extreme discordant phenotype methodology: an intuitive approach to clinical pharmacogenetics. Eur J Pharmacol 410: 107-120, 2000.[Web of Science][Medline]
  18. Parchment RE, Huang M, and Erickson-Miller CL. Roles for in vitro myelotoxicity tests in preclinical drug development and clinical trial planning. Toxicol Pathol 21: 241-250, 1993.[Abstract/Free Full Text]
  19. Pessina A, Albella B, Bayo M, Bueren J, Brantom P, Casati S, Croera C, Parchment R, Parent-Massin D, Schoeters G, Sibiri Y, Van Den Heuvel R, and Gribaldo L. In vitro tests for haematotoxicity: prediction of drug-induced myelosuppression by the CFU-GM assay. Altern Lab Anim 30: 75-79, 2002.
  20. Pessina A, Albella B, Bueren J, Brantom P, Casati S, Gribaldo L, Croera C, Gagliardi G, Foti P, Parchment R, Parent-Massin D, Sibiril Y, and Van Den Heuvel R. Pre-validation of a model for predicting acute neutropenia by colony forming unit granulocyte/macrophage (CFU-GM) assay. Toxicol In Vitro 15: 729-740, 2001.[Web of Science][Medline]
  21. Pevnitsky LA, Telegin L, Zhirnov GF, Mazurov AV, and Viktorov VV. Sensitivity to immunodepressant action of cyclophosphamide: analysis of interstrain differences in mice. Int J Immunopharmacol 7: 875-880, 1985.[Web of Science][Medline]
  22. Phillips K, Veenstra D, Oren E, Lee J, and Sadee W. Potential role of pharmacogenomics in reducing adverse drug reactions. JAMA 286: 2270-2279, 2001.[Abstract/Free Full Text]
  23. Reimer DL and Singh SM. Cyclophosphamide-induced in vivo sister chromatid exchanges (SCE) in Mus musculus. III. Quantitative genetic analysis. Genetics 105: 169-179, 1983.[Abstract/Free Full Text]
  24. Risch N and Merikangas K. The future of genetic studies of complex human diseases. Science 273: 1516-1517, 1996.[Abstract/Free Full Text]
  25. Roberts AW, Foote S, Alexander WS, Scott C, Robb L, and Metcalf D. Genetic influences determining progenitor cell mobilization and leukocytosis induced by granulocyte colony-stimulating factor. Blood 89: 2736-2744, 1997.[Abstract/Free Full Text]
  26. Rozen S and Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol 132: 365-386, 2000.[Medline]
  27. Semerad CL, Liu F, Gregory AD, Stumpf K, and Link DC. G-CSF is an essential regulator of neutrophil trafficking from the bone marrow to the blood. Immunity 17: 413-423, 2002.[Web of Science][Medline]
  28. Silver LM. Mouse Genetics. New York: Oxford Univ. Press, 1995.
  29. Telegin L, Zhirnov GF, Mazurov AV, and Pevnitskii LA. Immunosuppressant effect of cyclophosphamide activated in vitro by liver microsomes from different strains of mice. Biull Eksp Biol Med 92: 57-60, 1981.[Medline]
  30. To L, Haylock D, Simmons P, and Juttner C. The biology and clinical uses of blood stem cells. Blood 89: 2233-2258, 1997.[Free Full Text]
  31. Vieux EF, Kwok PY, and Miller RD. Primer design for PCR and sequencing in high-throughput analysis of SNPs. Biotechniques Suppl: 28-30, 32, 2002.
  32. Wade CM, Kulbokas EJ, Kirby AW III, Zody MC, Mullikin JC, Lander ES, Lindblad-Toh K, and Daly MJ. The mosaic structure of variation in the laboratory mouse genome. Nature 420: 574-578, 2002.[Medline]
  33. Watters JW and McLeod HL. Murine pharmacogenomics: using the mouse to understand the genetics of drug therapy. Pharmacogenomics 3: 781-790, 2002.[Web of Science][Medline]
  34. Watters JW and McLeod HL. Using genome-wide mapping in the mouse to identify genes that influence drug response. Trends Pharmacol Sci 24: 55-58, 2003.[Medline]
  35. Xie HJ, Lundgren S, Broberg U, Finnstrom N, Rane A, and Hassan M. Effect of cyclophosphamide on gene expression of cytochromes p450 and {beta}-actin in the HL-60 cell line. Eur J Pharmacol 449: 197-205, 2002.[Web of Science][Medline]



This article has been cited by other articles:


Home page
 Molecular Cancer TherapeuticsHome page
U. Emmenegger, Y. Shaked, S. Man, G. Bocci, I. Spasojevic, G. Francia, A. Kouri, R. Coke, W. Cruz-Munoz, S. M. Ludeman, et al.
Pharmacodynamic and pharmacokinetic study of chronic low-dose metronomic cyclophosphamide therapy in mice
Mol. Cancer Ther., August 1, 2007; 6(8): 2280 - 2289.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
V. Y. Gorbacheva, R. V. Kondratov, R. Zhang, S. Cherukuri, A. V. Gudkov, J. S. Takahashi, and M. P. Antoch
From The Cover: Circadian sensitivity to the chemotherapeutic agent cyclophosphamide depends on the functional status of the CLOCK/BMAL1 transactivation complex
PNAS, March 1, 2005; 102(9): 3407 - 3412.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
M. Bogue
Mouse Phenome Project: understanding human biology through mouse genetics and genomics
J Appl Physiol, October 1, 2003; 95(4): 1335 - 1337.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (9)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Watters, J. W.
Right arrow Articles by McLeod, H. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Watters, J. W.
Right arrow Articles by McLeod, H. L.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2003 by the American Physiological Society.