ToxSci Advance Access originally published online on May 12, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Toxicological Sciences 80, 1-2 (2004)
Toxicological Sciences vol. 80 no. 1 © Society of Toxicology 2004; all rights reserved.
TOXICOLOGICAL HIGHLIGHT |
Nature Is Complex: Our Orchestra Seats at the Most Wonderful Show on Earth
Pfizer Global Res. and Devolpment, MS 8274-1336, Eastern Point Road, Groton, CT 06340
Received May 6, 2004; accepted May 6, 2004
Any student of Nature's complexity is constantly rewarded as science slowly uncovers hints about the processes of life and the controls embedded in it. Life scientists, of course, have a front row seat to absolutely the most wonderful show on earth, and if we're smart, lucky, and tenacious, we get to help discover little pieces of this puzzle. One of the next pieces presented for us in this issue is the article by Bob Markelewicz, Susan Hall, and Kim Boekelheide, "2,5-Hexanedione and Carbendazim Co-exposure Synergistically Disrupts Rat Spermatogenesis despite Opposing Molecular Effects on Microtubules" (pp. 92–100).
Toxicologists shoulder a challenge, along with our awe, and that is to use science to serve public health by putting our knowledge to work by recommending things like allowable exposure levels and recommended tolerances. This process is fraught with uncertainties in the best of situations, but the particular situation of regulating mixtures exposures, or understanding the biologic processes involved in responding to mixtures, is at once the most frustrating and most rewarding. Most rewarding, because mixtures really encapsulate the complexities of real world: no one has a single exposure at a time. And the most frustrating because the depth of our ignorance is profound, and it is frequently these knowledge chasms that encompass the very information that the regulator or the scientist most desperately needs.
Mixtures science has been making steady progress over the past decade, and several recent reviews (Cassee et al., 1998
; Feron and Groten, 2002; Pohl et al., 2003) do an excellent job of capturing the critical issues facing regulators (who must make policy decisions based on imperfect information) and scientists (who must allocate increasingly scarce resources to answer the most pressing or illuminating questions in a complex field). They also review some of the literature in this area (Carpy et al., 2000), which provides an excellent opportunity for reacquainting oneself with this complicated and critical area.
The authors of one of these reviews (Feron and Groten, 2002, pp. 836) presciently described the Markelewicz et al. article when they wrote, "Well-thought-out, tailor-made mechanistic and statistical designs to study the toxicity and the mechanism of toxicity of mixtures have been developed and successfully applied." In this article, the Boekelheide lab turns its long experience with the toxicity and mechanism of 2,5-hexanedione (HD) to prime advantage.
Three to four weeks after the start of exposure, HD produces basal vacuoles in the Sertoli cell, with a subsequent loss of germ cells that then happens over the course of one to two weeks. In the 1980's, Dr. Boekelheide showed that HD stabilized microtubules, and concluded that disrupted microtubule dynamics is responsible for the effects seen in the testis. (reviewed in Boekelheide et al., 1989). This theory has significant appeal: the Sertoli cell is loaded with microtubules; we know that the seminiferous epithelium is astoundingly active, with the Sertoli cells moving the germ cells around, controlling access to the privileged space distal to the blood-tubule barrier and shoving the late spermatids out of the way to make room for the space-hungry and dazzlingly beautiful meiotic divisions that mark the transition from pachytene to spermatid. It seems intuitively obvious that much of this activity would depend on normal cytoskeletal activity, and stabilizing microtubules by HD should logically disrupt this delicately balanced dynamic equilibrium.
Enter carbendazim (CBZ), the active metabolite of benomyl, another testicular toxicant. Within 24 h of administration, CBZ causes basal vacuoles and the wholesale sloughing of the adluminal one-half to one-third of the epithelium (germ cells and Sertoli cells), with the result that the efferent ducts become clogged with this debris, while the persistent fluid production by the seminiferous tubules causes those tubules to enlarge in the testis. Carbendazim was found to block microtubule polymerization (Howard and Aist, 1980; Quinlan et al., 1980), which was thought to contribute to the sloughing of the epithelium. It makes some sense: if you rapidly deploymerize the microtubules, and if those microtubules are responsible for the integrity of the Sertoli cell (and if the Sertoli cell controls the structure of the adluminal 7/8ths of the epithelium), then you might, indeed, cause the wholesale detachment of some part of that epithelium.
As a Commentary writer, I will employ my poetic license and cherish the image of a late night in the Boekelheide lab, with someone slaving away over some 12 h timepoint, when the metaphoric light-bulb goes off: ‘Hey, we should try to coadminister these two compounds, and see if they'll cancel each other out!’ The sympathetic observer can imagine (and vicariously enjoy) the ensuing nikhedonia: if tubulin disruption is the prime mechanism of toxicity for these two compounds, there should, indeed, be a lessening of the toxicity seen by coadministration. And if they don't cancel each other out, then we've learned something new. Good mentors always teach their students to construct win-win experiments precisely such as this.
Ah, but Nature is complex, and surprises are wonderful.
Forced into a complicated dosing regimen by the nature of the lesions' pathogenesis, Markelewicz et al. found that coadministration of modest-effect doses of these two compounds resulted in more severe toxicity than was seen in animals given either compound alone. Satisfactorily, they added significant context by showing that CBZ did, indeed, delay tubulin polymerization kinetics in vitro, and HD speeded these up. Adding CBZ to HD-treated tubulin in vitro moved the polymerization curve more towards that of untreated tubulin. They also showed that HD administration did not confound CBZ pharmacokinetics, and testis levels of CBZ were unaltered by HD. Delightfully, they performed the in vivo experiment twice, and obtained similar results each time.
So we have a significant surprise: despite the fact that they documented opposing effects on microtubule polymerization, Boekelheide's group found that HD administration appears to sensitize the rats to the effects of CBZ, and the toxicity in animals receiving both compounds is greater than that seen in animals given CBZ alone or HD alone. The mind spins: how might something like this work and does this change the state of play in mixtures science?
Markelewicz et al. briefly review the formation of active pyrrole intermediates after HD treatment, and remind us that they bind to and modify (indeed, even cross-link) biological molecules. It requires no huge imagination to see that molecules other than tubulin could well be affected by pyrrole crosslinking or modification. Which molecules these might be, which ones are really involved in the observed toxicity, and where the pathways of pathogenesis for these two compounds intersect, is no doubt the subject of intense activity even now, and probably for some years to come.
Another likelihood is that the disruption of microtubule dynamics and, therefore, function, whether by stabilization or by depolymerization, adversely affects the cellular processes that depend on normal microtubules. It's possible that we're witnessing the additive effects of disrupted function, albeit by two opposing molecular actions. In this scenario, the emphasis is on an integrated change in a functional endpoint, rather than focusing on the molecular means by which that function is disrupted.
I will yield to temptation, and take this opportunity to remind the Gentle Reader of a favorite refrain: One molecule, many targets. As scientists, we all struggle daily against a desire to focus on a single known molecular interaction or effect, and forget that the cell is positively packed with potential targets. We focus on the known because it lets us create testable hypotheses that might explain most of what we observe. But it bears repeating that we simplify Nature's complexities at our peril. It is to the credit of Markelewicz, Hall, and Boekelheide that they used this opportunity to test their previous conclusions, and found it more complex than much previous data would imply. They acknowledged the possibility of complexity, and were proven right (it is complex).
Mixtures science is, perhaps sadly, acclimated to such surprises (Hooth et al., 2002; Narotsky et al., 1995; Zeliger, 2003). The new article from the Boekelheide lab adds another specific example of the perils of simplification, and the benefits of testing our assumptions. The science of studying mixtures, and the risk assessors who peer closely over the shoulders of the toxicologists, are immensely edified by such an approach. Risk assessors and model-makers are moved incrementally closer to a real understanding of the best way to regulate mixtures by having another example of an important surprise that can function as a model-tester at some time in the future. (It's along the lines of: "If your model can correctly predict this interaction, it can handle anything.")
I'm loving my ringside seat. Popcorn, anyone?
NOTES
1 For correspondence via fax: (860) 715-3577. E-mail: robert_e_chapin{at}groton.pfizer.com.
REFERENCES
Boekelheide, K., Neely, M. D., and Sioussat, M. D. (1989). The Sertoli cell cytoskeleton: Target for toxicant-induced germ cell loss. Toxicol. Appl. Pharmacol. 101, 373–389.[CrossRef][ISI][Medline]
Carpy, S. A., Kobel, W., and Doe, J. (2000). Health risk of low-dose pesticide mixtures: A review of the 1985–1998 literature on combination toxicology and health risk assessment. J. Toxicol. Environ. Health, Part B 3, 1–25.
Cassee, F. R., Groten, J. P., van Bladeren, P. J., and Feron, V. J. (1998). Toxicological evaluation and risk assessment of chemical mixtures. Crit. Rev. Toxicol. 28, 73–101.[CrossRef][ISI][Medline]
Feron, V. J., and Groten, J. P. (2002). Toxicological evaluation of chemical mixtures. Food Chemical Toxicol. 40, 825–839.
Hooth, M. J., McDorman, K. S., Hester, S. D., George, M. H., Brooks, L. R., Swank, A. E., and Wolf, D. C. (2002). The carcinogenic response of Tsc2 mutant Long-Evans (Eker) rats to a mixture of drinking water disinfection by-products was less than additive. Toxicol. Sci. 69, 322–331.
Howard, R. J., and Aist J. R. (1980). Cytoplasmic microtubules and fungal morphogenesis: Ultrastructural effects of methyl benzimidazole-2-yl carbamate determined by freeze substitution of hyphal tip cells. J. Cell Biol. 87, 55–64.
Narotsky, M. G., Weller, E. A., Chinchilli, V. M., and Kavlock, R. J. (1995). Nonadditive developmental toxicity in mixtureds of trichloryethylene, di(2-ethylhexyl) phthalate, and heptachlor in a 5 x 5 x 5 design. Fundam. Appl. Toxicol. 27, 203–216.[CrossRef][ISI][Medline]
Pohl, H. R., Roney, N., Wilbur, S., Hansen, H., and de Rosa, C. T. (2003). Six interaction profiles for simple mixtures. Chemosphere 53, 183–197.[Medline]
Quinlan, R. A., Pogson, C. I., and Gull, K. (1980). The influence of the microtubule inhibitor methyl benzimidazole-2-yl carbamate on nuclear division and the cell cycle in Saccharomyces cerevisiae. J. Cell Biol. 46, 341–352.
Zeliger, H. I. (2003). Toxic effects of chemical mixtures. Arch. Environ. Health 58, 23–29.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||