Department of Cell Biology and Neuroscience, University of California, Riverside, California 92521
Received March 16, 2003; accepted April 16, 2003
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
It is well known that smoking during pregnancy can cause a decrease in fetal weight (Stillman et al., 1986), and while this effect has been observed mainly in active smokers, it has also been reported in fetuses born to women exposed to environmental tobacco smoke during pregnancy (Martin and Bracken, 1986
). We have previously shown that chemicals in both mainstream and sidestream cigarette smoke significantly impair growth of the chick chorioallantoic membrane (CAM) in a dose-dependent manner (Melkonian et al., 2002
). The CAM is an extraembryonic membrane that begins forming between days 4 and 5 of development by fusion of the allantois and chorion (DeFouw et al., 1989
; Hamilton, 1965
). The CAM is of physiological importance to the chick, as it serves as the major respiratory organ for gaseous exchange until hatching, and it provides a "bladder" into which waste products can be delivered (Hamilton, 1965
; Romanoff, 1967
). Because the CAM increases 1020-fold in size between days 5 and 6 of development, it provides a rapid and useful assay for evaluating the effects of chemicals on tissue growth (Melkonian et al., 2001
, 2002
). The CAM is also a widely used model for studying angiogenesis (Melkonian et al., 2001
; Ribatti and Vacca, 1999
).
In the CAM assay, both mainstream and sidestream smoke solutions caused very significant retardation of CAM growth between days 5 and 6 of development and also impaired various aspects of angiogenesis (Melkonian et al., 2002). The active chemicals partitioned mainly in the particulate phase of mainstream smoke and in the gas phase of sidestream smoke (Melkonian et al., 2002
). To identify the chemicals responsible for retarding growth and angiogenesis in CAMs, we previously used a combination of solid phase extraction cartridges and gas chromatography-mass spectrometry to fractionate and identify the chemicals in sidestream gas phase smoke that inhibit CAM growth (Ji et al., 2002
). Using this approach, 12 pyridine and 10 pyrazine derivatives were identified in the active fraction of sidestream gas phase smoke solutions. We previously tested the pyridines using the CAM assay and found that 2- and 3-ethyl pyridine inhibit CAM growth at picomolar doses (Ji et al., 2002
).
The purpose of the present study was to test the pyrazine derivatives identified in the sidestream gas phase smoke solutions in the CAM assay to evaluate the effects of each on CAM growth and angiogenesis. In addition, the embryos of treated CAMs were weighed to establish if treatment with the test chemical inhibited embryo growth. Our data show that some pyrazines in cigarette smoke are able to significantly inhibit CAM and embryo growth and impair angiogenesis at nano and picomolar doses. This finding is significant because several of these pyrazines are generally considered safe and are added to consumer products, including food, for flavoring.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Solid phase extraction of smoke solutions.
Bond-Elut solid phase extraction (SPE) cartridges (3cc with 500 gram capacity) (Phenomenex Torrance, CA) were used to fractionate smoke solutions and concentrate chemicals that inhibit CAM growth. Cartridges which were screened for their ability to bind inhibitory chemicals included a variety of non-polar, polar, and anion and cation exchange cartridges: NH2, 2OH, CN, CBA, SCX, SAX, C18, C8, C2, CH, SI, and PH. The protocol used to screen the cartridges for their ability to bind growth inhibitory toxicants in smoke solutions has been described in detail previously (Ji et al., 2002).
Gas chromatography-mass spectrometry (GC-MS).
The C2 solid phase extraction cartridge retained most of the inhibitory activity in sidestream gas phase smoke solutions as determined in the CAM growth assay. To identify the components in aqueous smoke solutions that inhibit growth, mainstream and sidestream smoke solutions were analyzed with GC-MS after solid phase extraction on a C2 column. The equipment used was a Hewlett Packard 5890 GC interfaced to an HP-5971A MSD (quadropole mass selective detector) with a Zebron ZB1701 cyanopropyl phenyl column 30 m x 0.32 mm and with a 1 µm phase thickness (Phenomenex, Torrance, CA.). The carrier gas was helium and the instrument was operated in the scanning mode (40350 amu). The temperature program was an initial temperature of 45°C for one minute, with an increase of 10°C per minute to a final temperature of 280°C after ten min, with a total run time of 34.5 min. Two microliters of the eluate sample were injected directly into the GC using a Hamilton gas-tight syringe. Identification of compounds was made using the mass spectrometry data matched to mass spectral library entries. Compound identities were confirmed using purified standards and matching both mass spectra and retention times.
CAM assay and embryo wet weights.
CAM growth was evaluated using a modification of the CAM assay that has been described in detail previously (Melkonian et al., 2001). 200 µl of either EBSS-H or different concentrations of pure chemicals were added to the surface of each CAM at 10 A.M. on the fifth day after fertilization. A sham control group, in which the window was opened and resealed, was also included in each experiment. In all cases, the following five concentrations of test reagents were used 5 x 10-5 M, 5 x 10-7 M, 5 x 10-9 M, 5 x 10-11 M, and 5 x 10-13 M. Development was terminated at 10 AM on the sixth day after fertilization by fixing the embryo and the CAM with 3% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4) for 2 hours at room temperature. The CAMs were later dissected from eggs, post-fixed in the same fixative for 24 hours at room temperature, then thoroughly rinsed in PBS. Embryos were removed from each egg and dissected free of any surrounding tissue before obtaining their wet weights. Each control and treatment group had at least five CAMs and five embryos.
Quantification and imaging of CAMs.
To determine area, each CAM was placed in a Petri dish with PBS and examined with a Wild-M5A dissecting microscope (Max ERB instrument Co., Burbank, CA). The longest and shortest lengths were measured with a ruler to the precision of 0.5 mm, and CAM area was calculated using the formula:
![]() | (1) |
Means and standard deviations were then calculated for each control and treatment group. The treated group means were compared with control means to determine the effect of each chemical on CAM growth. The lowest observed adverse effect level (LOAEL), the no observed adverse effect level (NOAEL), and the maximum percentage of inhibition (efficacy) were determined for each pyrazine derivative.
Measurement of DNA synthesis.
To determine if treatment of CAMs with pyrazine affected synthesis of DNA, day 5 CAMs were exposed to 5 x 10-7 M pyrazine for 21 hours, then 100 µl of H3-thymidine was added to each CAM. After 2 hours, CAMs were dissected from eggs, washed thoroughly several times in EBSS-H, and solubilized overnight in 8 M potassium hydroxide. Aliquots (20 µl) of the CAM lysate were added to 4 ml of scintillation cocktail, and counts were made 24 hours later using a Beckman scintillation counter.
Evaluation of blood vessel pattern formation in developing CAMs.
Blood vessels in day 6 CAMs have a characteristic dendritic branching pattern (Melkonian et al., 2001). To evaluate blood vessel pattern formation, fixed CAMs were further dissected to remove edge tissue and mounted in tissue culture dishes under cover slips to flatten the CAMs. Video images were captured at a magnification of approximately 3x using a Hitachi KP-D50U camera (Hitachi Inc., Torrance, CA). To evaluate the effect of chemicals on pattern formation, blind comparisons of images of control and treated CAMs were made to a representative control image. The images were ranked from "0" to "2," with "0" representing the branching pattern observed in controls and "2" representing the most severe disruption of blood vessel pattern. Images showing examples of disruption of pattern formation have been published previously (Melkonian et al., 2002
).
CAMs that had been treated with pyrazine were embedded in plastic and processed for light microscopy as described previously (Melkonian et al., 2001). To evaluate capillary plexus formation, histological cross sections through CAMs were digitized with a Spot camera (Diagnostic Instruments, Sterling Heights, Michigan) using a 16x objective. For each parameter, one histological section from five different CAMs was examined at each dose group. To determine the percentage of ectoderm subtended by capillary plexus (this is a measure of how much plexus has formed), a 600-µm length of each CAM was marked on the computer monitor, and the length of the plexus that had formed immediately beneath the ectoderm was measured and presented as a percentage of the total (600-µm) projected length. The number of mesodermal blood vessels that did not migrate to the basal lamina beneath the ectoderm was also counted for the same 600-µm region of each CAM.
Statistical analyses.
Group means in each of the assays were compared by analysis of variance (ANOVA). When significance was found (p < 0.05), Dunnetts post hoc test was used to identify significantly affected groups. Dunnetts test compares treated group means to the EBSS-H control group. Data were checked to determine if they satisfied the assumptions of ANOVA (homogeneity of variances and normal distribution). If the assumptions were not satisfied, the Kruskal-Wallis non-parametric test was used followed by Dunns post hoc test. Analyses were done using GraphPad (Instat, San Diego, CA).
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
A Single Ethyl Substitution to the Pyrazine Ring Slightly Decreased Its Potency
One pyrazine compound with a single ethyl substitution (2-ethylpyrazine) was identified in the active fraction of sidestream smoke. Like pyrazine alone, it was highly inhibitory in both the CAM and embryo growth assays (Figs. 2A,B). A dose dependent decrease in CAM and embryo growth was observed over the concentration range of 5 x 10-11 to 5 x 10-5 M. Significant inhibition of CAM and embryo growth was observed at nanomolar doses (5 x 10-9 M). The maximum percentage of inhibition was 51% and 26% for the CAM and embryo, respectively.
|
|
|
|
|
Pyrazine Treatment Inhibits DNA Synthesis
Since pyrazine had the greatest potency and efficacy of the chemicals tested, it was studied further. To determine if pyrazine treatment inhibited CAM growth by inhibiting DNA synthesis, day 5 CAMs were treated with 5 x 10-7 M pyrazine, then 21 h later were labeled with H3-thymidine as described in Materials and Methods. CAMs that were treated with pyrazine incorporated significantly less H3-thymidine than the EBSS controls (Fig. 8), indicating that DNA synthesis was reduced by the treatment.
|
Treatment of CAMs with Pyrazine Disrupts Formation of the Capillary Plexus
The capillary plexus normally begins forming during days 5 and 6 of chick development. By day 6, a discrete plexus underlies most of the ectoderm (Fig. 9A, arrows). This plexus originates mainly from migration of mesodermal blood vessels to the ectoderm, with some contribution from vascularization or formation of new vessels from angiogenic clusters just beneath the ectoderm (Melkonian et al., 2001
). In histological sections of control CAMs treated with EBSS-H, the plexus was well formed, and most of the ectoderm had plexus vessels directly touching it (Fig. 9A
). In CAMs treated with pyrazine, the plexus usually did not appear well developed and many mesodermal vessels (arrowheads) were still evident (Fig. 9B
). To determine if there existed a quantitative difference in the amount of plexus formed in treated versus control CAMs, the percentage of ectoderm with plexus immediately beneath it was computed for the pyrazine experiment (Fig. 10A
). There was a dose-dependent decrease in the amount of plexus formed. Concentrations as low as 5 x 10-9 M significantly inhibited formation of the plexus. To determine if this inhibition of plexus formation was related to failure of mesodermal blood vessels to migrate to the ectoderm, the number of mesodermal vessels was counted for each CAM (Fig. 10B
). A dose-dependent inhibition of vessel migration by pyrazine was observed. Significant inhibition of migration occurred at 5 x 10-7 M pyrazine. While the 5 x 10-9 M treatment group likewise appeared to show inhibition, it was not significantly different than the control group.
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The pyrazine ring inhibited both CAM and embryo growth more effectively than any of its derivatives. Substitution of an ethyl functional group on pyrazine decreased its potency 100-fold, while substitution of one or more methyl groups further decreased its potency by 1000- to 100,000-fold in the CAM/embryo growth assay. There was not a clear correlation with the type of substitution and the level of inhibition as was seen previously for the pyridine derivatives in smoke (Ji et al., 2002), nor was there a correlation with lipophilicity (not shown). In a previous study in which pyridine derivatives were screened in the CAM growth assay, pyridine itself was not strongly inhibitory. However, addition of a single methyl group increased its potency by 10 thousand-fold and addition of a single ethyl group increased its potency by 10 million-fold (Ji et al., 2002
). In contrast to the pyridines, the data with the pyrazines indicate that the effect of pyrazines on growth cannot necessarily be predicted from their structure and that each chemical needs to be evaluated individually to determine its effect on tissue growth. It is interesting that pyrazine differs from pyridine only by the presence of a second nitrogen atom in the fourth position of the ring. This simple modification to the ring resulted in a 1 million-fold increase in the potency of pyrazine (LOAEL = 5 x 10-11 M) versus pyridine (LOAEL = 5 x 10-5 M) in the CAM growth assay.
The data for 2,5-dimethylpyrazine are interesting in that they show a very significant decrease (over 50% inhibition) in CAM growth at 5 x 10-7 and 5 x 10-5 M, while the same doses produced no significant effect on embryo growth. These data indicate that a smaller CAM does not necessarily produce a smaller embryo. 2,5-Dimethylpyrazine was the only chemical that specifically inhibited CAM growth without affecting embryo growth, which further shows that the effects of these chemicals are complex and may depend on the type of tissue being studied. These data also suggest that the chemicals that inhibited both CAM and embryo growth probably inhibited embryo growth by acting directly on the embryos, as decreased CAM size did not necessarily correlate with decreased embryo size.
Angiogenesis is complex, involving numerous cell processes (Folkman and Shing, 1992; Risau, 1997
). In this study, pyrazine and 2,3-dimethylpyrazine altered normal branching of blood vessels in the developing CAMs. However, the other pyrazines were without significant effect in this assay at the highest dose tested (5 x 10-5 M). Since whole smoke produces a strong effect on vessel branching (Melkonian et al., 2000
, 2002
), the data obtained with pyrazine and 2,3-dimethylpyrazine suggest that there may be other chemicals in smoke that are more active in this assay. Pyrazine also inhibited the migration of mesodermal blood vessels to the ectoderm and the subsequent formation of the capillary plexus at nanomolar doses. This observation shows that pyrazine can retard at least one process important in angiogenesis at low doses. Failure of the capillary plexus to form properly in pyrazine-treated CAMs could compromise oxygen and CO2 exchange and may, like inhibition of DNA synthesis, also be a factor contributing to retarded growth of CAMs and embryos.
The biological effects of pyrazine and its derivatives are not well understood. In fact, pyrazine plus several of its derivatives appear on a list of chemicals published by the Flavor and Extract Manufacturers Association (FEMA) that are generally recognized as safe (GRAS) and are often added to consumer products such as food, tobacco, cosmetics, and fragrances (Adams et al., 2002; Smith et al., 2001
). Pyrazine, which was added to this list in December 2001, may be the most significant with respect to our data. Of the compounds we tested, pyrazine was the most potent at inhibiting CAM and embryo growth, having a LOAEL of 5 x 10-11 M in both the CAM and embryo assays. In the CAM growth assay, the LOAEL and estimated ED50 (5 x 10-7 M) doses are equivalent to 4 ppb and 0.4 ppm, respectively. The ED50 dose in ppm is within the range of concentrations for the average usual (0.31 ppm) and average maximum (1.55 ppm) concentration of pyrazine added to various types of food (Smith et al., 2001
). The LOAEL dose (4 ppb) is, of course, well below the concentrations of pyrazine that are added to food products.
In our study of developing tissues, pyrazine had a profound effect on growth in both the CAM and embryo assay. Most toxicological studies on pyrazine and its derivatives have involved adults, and effective doses may be different for adult tissue and developing tissue.
However, there is evidence from both in vitro and in vivo models that supports the idea that pyrazine and its derivatives can inhibit growth in adults. In vitro studies with cultured cells have shown that growth can be inhibited by pyrazine derivatives (Shang et al., 1998; Zurbonsen et al., 1999
). Moreover, in some prior studies done in vivo using adult rodent models, the weight of the animals or of specific organs decreased following oral exposure to a pyrazine derivative (Adams et al., 2002
; Posternak et al., 1969
, 1975
). In some instances, this was attributed to test animals eating less and therefore gaining less weight (Posternak et al., 1969
, 1975
), although intake of food was not actually quantified in all studies. In some in vivo studies, body weight or specific organ weight was compromised by exposure to pyrazines that were administered as vapors (Katz et al., 1999
) or by subcutaneous injection (Yamada et al., 1996
). Reproductive organs can be targets of growth inhibition, as administration of 2,5-dimethylpyrazine to female rats caused a significant reduction in uterine, but not ovarian, weight, perhaps by affecting uptake of estradiol by the uterus (Yamada et al., 1992
). Reproductive processes appear also to be sensitive to pryrazine derivatives, as 2,5-dimethylpyrazine decreased the overall success of reproduction for rodents that were housed together, although high doses (>70 mg/kg body weight/day) were required to observe this effect (Novotny et al., 1986
). Moreover, first vaginal opening in rats, which is an indicator of the onset of puberty, was also inhibited by tetramethyl and 2,5-dimethylpyrazine (Yamada et al., 1989
). However, in one study on reproduction, administration of tetramethylpyrazine produced no significant effect on any reproductive parameter that was monitored (Vollmuth et al., 1990
). In our study, the LOAELs of the tested pyrazines were quite different, and the derivative with the most substituted groups (2,3,5-trimethylpyrazine) was not as effective at inhibiting CAM growth as pyrazine and 2-ethylpyrazine.
The implications of our data for human reproduction are at present not known, but well-established information on smoking and growth during reproduction indicate that further work should be done to address this topic. Establishing and maintaining pregnancies in mammals requires extensive growth in the embryo/fetus, corpus luteum, and the placenta (Findlay, 1986), and this growth can be compromised by exposure to cigarette smoke (Stillman et al., 1986
). Both active (Stillman et al., 1986
) and passive (Martin and Bracken, 1986
) smokers have fetuses with lower than normal birth weights. This reduction in birth weight is thought to be due to fetal hypoxia caused by nicotine-induced vasoconstriction and decreased placental blood flow and by increased carboxyhemoglobin leading to decreased fetal oxygenation (Tourmaa, 1995
). Given the complexity of cigarette smoke, it is likely that multiple mechanisms affect prenatal growth in smoke-exposed fetuses. We have found that both ethyl and methyl substituted pyridines (Ji et al., 2002
), pyrazine, and 2-ethylpyrazine are very effective at retarding growth in the CAM and embryo assay at very low doses and that, for pyrazine, this inhibition correlates with inhibition of DNA synthesis. Since pyrazine and several of its derivatives are present in cigarette smoke and are added to consumer products, further work on the toxicity of these pyrazines, especially in developing organisms, should be undertaken to obtain a more complete understanding of their effect on human health and reproduction.
![]() |
NOTES |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
DeFouw, D. O., Rizzo, V. J., Steinfeld, R., and Feinberg, R. N. (1989). Mapping of the microcirculation in the chick chorioallantoic membrane during normal angiogenesis. Microvasc. Res. 38, 136147.[ISI][Medline]
Environmental Protection Agency (EPA) (1992). EPA Report/600/690/006F: Respiratory health effects of passive smoking: Lung cancer and other disorders, Washington, DC.
Findlay, J. K. (1986). Angiogenesis in reproductive tissues. J. Endocrinol. 111, 357366.[Abstract]
Folkman, J., and Shing, J. (1992). Angiogenesis. J. Biol. Chem. 267, 1093110934.
Hamilton, H. L. (1965). Lillies Development of the Chick. Holt, Rinehart, and Winston, New York.
Ji, L., Melkonian, G., Riveles, K., and Talbot, P. (2002). Identification of pyridine compounds in cigarette smoke solution that inhibit growth of the chick chorioallantoic membrane. Toxicol. Sci. 69, 217225.
Katz, A., Gvaryahu, G., Robinzon, B., Snapir, N., and Barnea, A. (1999). The effect of pyrazine odor on body weight and the weight of various organs in chicks (Gallus gallus domesticus). Poult. Sci. 78, 17861789.[ISI][Medline]
Knoll, M., and Talbot, P. (1998). Cigarette smoke inhibits oocyte cumulus complex pick-up by the oviduct independent of ciliary beat frequency. Reprod. Toxicol. 12, 5768.[CrossRef][ISI][Medline]
Martin, T. R., and Bracken, M. B. (1986). Association of low birth weight with passive smoke exposure in pregnancy. Am. J. Epidem. 124, 633642.[Abstract]
Melkonian, G., Cheung, L., Marr, R., Tong, C., and Talbot, P. (2002). Mainstream and sidestream cigarette smoke inhibit growth and angiogenesis in the day 5 chick chorioallantoic membrane. Toxicol. Sci. 68, 237248.
Melkonian, G., Le, C., Zheng, W., Talbot, P., and Martins-Green, M. (2000). Normal patterns of angiogenesis and extracellular matrix deposition in chick chorioallantoic membranes are disrupted by mainstream and sidestream cigarette smoke. Toxicol. Appl. Pharmacol. 163, 2637.[CrossRef][ISI][Medline]
Melkonian, G., Munoz, N., Chung, J., Tong, C., Marr, R., and Talbot, P. (2001). Capillary plexus development in the day 5day 6 chick chorioallantoic membrane is inhibited by cytochalasin D and suramin. J. Exp. Zool. 292, 241254.[ISI]
Novotny, M., Jemiolo, B., Harvey, S., Wiesler, D., and Marchlewska-Koj, A. (1986). Adrenal-mediated endogenous metabolites inhibit puberty in female mice. Science 231, 722725.[ISI][Medline]
Posternak, J. M., Dufour, J. J., Rogg, C., and Vodoz, C. A. (1975). Toxicological tests on flavouring matters. II. Pyrazines and other compounds. Food Cosmet. Toxicol. 13, 487490.[ISI][Medline]
Posternak, J. M., Linder, A., and Vodoz, C. A. (1969). Toxicological tests on flavouring matters. Food Cosmet. Toxicol. 7, 405407.[ISI][Medline]
Ribatti, D., and Vacca, A. (1999). Models for studying angiogenesis in vivo. Int. J. Biol. Markers 14, 207213.[ISI][Medline]
Risau, W. (1997). Mechanisms of angiogenesis. Nature 386, 671674.[CrossRef][ISI][Medline]
Romanoff, A. L. (1967). Biochemistry of the Avian Embryo. Wiley, New York.
Shang, Q., Zhang, D., and Guan, W. (1998). [The inhibitory effect in vitro of salvia miltiorrhiza and tetramethyl pyrazine on the growth of fibroblasts]. Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi 12, 321324.[Medline]
Smith, R. L., Doull, J., Feron, V. J., Goodman, J. I., Munro, I. C., Newberne, P. M., Portoghese, P. S., Waddell, W. J., Wagner, B. M., Adams, T. B., and McGowen, M. M. (2001). GRAS Flavoring Substances. Food Technol. 55, 3455.[ISI]
Stillman, R. J., Rosenberg, M. J., and Sachs, B. P. (1986). Smoking and Reproduction. Fertil. Steril. 46, 545566.[ISI][Medline]
Talbot, P., DiCarlantonio, G., and Knoll, M. (1998). Identification of cigarette smoke components that alter functioning of hamster (Mesocricetus auratus) oviducts in vitro. Biol. Reprod. 58, 10471053.[Abstract]
Tourmaa, T. E. (1995). The adverse effects of tobacco smoking on reproduction and health: A review from the literature. Nutr. Health 10, 105120.[Medline]
Vollmuth, T. A., Bennett, M. B., and Hoberman, A. M. (1990). An evaluation of food flavoring ingredients using an in vivo reproductive and developmental toxicity screening test. Teratology 41, 597.
Yamada, K., Itoh, R., and Ohta, A. (1989). Influence of pyrazine derivatives on the day of vaginal opening in juvenile female rats. Jpn. J. Pharmacol. 49, 529530.[ISI][Medline]
Yamada, K., Komatsu, H., and Ohta, A. (1996). Effects of 3,6-dimethylpyrazine-2-thiol on reproductive and accessory reproductive organs in male rats. Biol. Pharm. Bull. 19, 360363.[ISI][Medline]
Yamada, K., Takahashi, H., and Ohta, A. (1992). Effects of 2,5-dimethylpyrazine on reproductive and accessory reproductive organs in female rats. Res. Commun. Chem. Pathol. Pharmacol. 75, 99107.[ISI][Medline]
Zurbonsen, K., Michel, A., Bonnet, P. A., Mathieu, M. N., and Chevillard, C. (1999). Antiproliferative, differentiating and apoptotic effects elicited by imidazo[1,2-a]pyrazine derivatives. Gen. Pharmacol. 32, 135141.[CrossRef][Medline]
|