Viewpoint pubs.acs.org/est
Time Matters in Ecotoxicological Sampling Due to Circadian Rhythm Yanbin Zhao† and Karl Fent*,†,‡ †
University of Applied Sciences and Arts Northwestern Switzerland (FHNW), School of Life Sciences, Gründenstrasse 40, CH−4132 Muttenz, Switzerland ‡ Swiss Federal Institute of Technology (ETH Zürich), Institute of Biogeochemistry and Pollution Dynamics, Department of Environmental System Sciences, CH−8092 Zürich, Switzerland
ABSTRACT: As in general, technological inventions also drive the development in the field of toxicology and ecotoxicology. In the past decade, gene expression analysis has become a universally applied technology allowing many insights into toxicological pathways of environmental contaminants. Due to the novel technologies, including quantitative determination of mRNA by quantitative reverse transcription analysis (qRT-PCR), and semiquantitative methods, such as microarrays and RNA-sequencing technologies, toxicological profiles of contaminants could be identified. For instance, gene expression analysis of genes associated with the hypothalamic−pituitary−gonadal axis (HPG axis) in fish had become a conventional end point for endocrine disrupting chemicals. While these gene expression data provide novel insights into identifying potential toxicological end points and molecular mechanisms, often not enough attention is given to the question of mRNA stabilities and reliabilities of transcriptional data, in particular when links to physiological effects are difficult to make. A crucial factor in this issue is the endogenous circadian oscillations of genes during sampling.
A
Normally no attention is being paid to the stability of mRNA expression, and time is not considered as a parameter important for mRNA quantification. However, serious artifacts can result, when time is not appropriately considered in the sampling. This is mainly due to potential transient transcriptional levels of genes, which may display 24 h circadian oscillations. Endogenous circadian oscillation is driven by a master pacemaker located in the brain in the suprachiasmatic nuclei (SCN) of the hypothalamus, which further influences a variety of biochemical pathways and metabolic processes, such as steroidogenesis and catabolism.1,2 In rodents, there are
crucial factor often neglected in sampling of probes is the time of sampling. This may result in potential erroneous gene expression data. In toxicological experiments, multipledose exposure is always employed to demonstrate the dosedependency of transcriptional and physiological effects. This implies a significant amount of samples for each sampling. Due to practical constraints, the time differences between sampling of different groups usually spans up to a few hours. Especially for adult fish experiments, besides tens of exposure tanks for sampling, different types of samples, such as blood for hormone levels, gonads for histology, and other organs for gene expression analysis have to be collected. Generally, this requires not only a team of co-workers, but a significant amount of time. Sampling usually lasts for hours. © 2016 American Chemical Society
Received: March 13, 2016 Published: March 24, 2016 3287
DOI: 10.1021/acs.est.6b01166 Environ. Sci. Technol. 2016, 50, 3287−3289
Viewpoint
Environmental Science & Technology
Figure 1. Schematic diagrams depict the circadian rhythm genes involved in classic pathways for toxicological experiments in embryonic/larval zebrafish (left) and approaches that help to reduce the sampling duration effect (right). Key: CT: circadian time; white bar: light/day period; black bar: dark/night period. HPG-L, hypothalamic−pituitary−gonadal-liver. Raw data taken from refs 4 and 5.
involved in metabolism (Figure 1); hsd17b3, hsd11b2, and cyp17a1 involved in steroidogenesis (Figure 1); atf6, atf4b1, and xbp1 involved in ER stress; and ddb2, caspase9, and afap1 involved in DNA damage/apoptosis showed a circadian rhythm. Specifically, the key genes of AhR signaling, ahr1a, cyp1a and cyp1b1, all displayed 24 h circadian oscillations (Figure 1). These genes serve as crucial molecular biomarkers in response to a variety of environmental pollutions, like PAHs, PCBs, and PCDDs. Obviously, a sampling duration spanning several hours may result in transcriptional alterations that are not compound-related, but are due to time-related changes regulated by circadian rhythm. Consequently, this may result in potential false interpretation of gene expression data. Many circadian oscillating genes, accompanied by daily light transitions, show peak transcriptional levels either immediately before dawn or dusk. A schematic diagram in Figure 1 depicts this classic time-dependent gene expression pattern that increase at day time and decrease at night, which actually represents the pattern of cyp17a1 and caspase9 in zebrafish.4 Obviously, false positive or negative interpretation may arise, when sampling time of controls and exposed fish did not closely match. This occurred no matter whether tissue samples were taken in the order of control, low dose and high dose, or reverse.5 Therefore, care has to be taken to control for correct sampling time and interpretation of transcriptional responses. To avoid the false interpretations, quick sampling is certainly the best approach, while it is usually difficult to be achieved due to practical constraints. An alternative approach is to add unexposed fish as negative controls at each sampling time point.5 The compound-related transcriptional alterations can be verified by comparison of transcriptional levels between exposed and unexposed fish at each time point. Alternatively, sampling of one replicate of each treatment and control group can be performed at the same time point, and subsequently the average and standard deviation can be assessed (Figure 1). These two approaches would be accurate to reduce false negative/positive results in experiments. In conclusion, sampling duration is often overlooked in experimental sampling, while in fact, it is a crucial factor that can result in artifacts in the transcriptional responses in
approximately 3%−10% genomic genes displaying robust circadian cycling in regard to different organs.2,3 Of which, certain biomarker genes involved in several classic pathways for toxicological experiments were also included, such as enzymes involved in metabolism, such as cyp1a and gstα, and casp6 and casp9 involved in apoptosis.2,3 Similarly, in zebrafish, about 2860 circadian oscillating genes accounting for over 17% of expressed genes occur in larval zebrafish.4 To better understand the role of circadian oscillating genes in toxicological experiments, here we further investigated raw data provided by Li et al. 2013.4 First, eight target signaling pathways were selected based on their widespread utilization in toxicological transcriptional analysis, including phase I and phase II metabolism enzymes, cell cycle regulation, DNA damage and apoptosis, HPG-Liver axis, nuclear receptors, endoplasmic reticulum (ER) stress and arylhydrocarbon receptor (AhR) signaling. Considering that multiple gene copies of most of these genes exist in zebrafish due to genome duplication as compared to mammals, we performed homologous alignments in the zebrafish genome (Database version 79.9) based on Ensembl database. In total, 294 core genes were identified. We further verified their 24 h gene expression levels. The data demonstrated that about 8 to 50% of genes associated with the eight selected target signaling pathways displayed circadian changes. Of which, genes involved in AhR signaling were the most significant ones (50%, 7 out of 14 genes), followed by phase I and phase II metabolism enzymes (44%, 19 out of 43, and 24%, 13 out of 55 genes, respectively), ER stress (13%, 3 out of 24 genes), DNA damage and apoptosis (12%, 5 out of 41 genes), HPG-L axis (14%, 5 our of 36 genes) and nuclear receptors (9%, 3 out of 33 genes). The least of these analyzed pathways showing circadian rhythm were cell cycle regulation, with 8% genes (4 out of 48 genes) displaying circadian changes. These results were consistent with the phenomenon observed in mouse, of which the metabolic process was also one of the most significant peripheral pathways regulated by circadian rhythm.2,3 Our detailed analysis demonstrated that some biomarker genes usually employed in effects analysis of environmental contaminants, including cyp1a, cyp2p6, cyp3c4, and mgst1 3288
DOI: 10.1021/acs.est.6b01166 Environ. Sci. Technol. 2016, 50, 3287−3289
Viewpoint
Environmental Science & Technology (eco)toxicological experiments. Adopting appropriate strategies to reduce or eliminate the effect of sampling time would enhance the data reliability and reproducibility, and further improve the ecotoxicological assessment of contaminants.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: +41 61 467 4571; fax: +41 61 467 47 84; e-mail: karl.
[email protected] or
[email protected]. Notes
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS The research is funded by the Swiss National Science Foundation (contract no. 310030_141040 to K.F.). REFERENCES
(1) Panda, S.; Antoch, M. P.; Miller, B. H.; Su, A. I.; Schook, A. B.; Straume, M.; Schultz, P. G.; Kay, S. A.; Takahashi, J. S.; Hogenesch, J. B. Coordinated transcription of key pathways in the mouse by the circadian clock. Cell 2002, 109 (3), 307−320. (2) Takahashi, J. S.; Hong, H. K.; Ko, C. H.; McDearmon, E. L. The genetics of mammalian circadian order and disorder: implications for physiology and disease. Nat. Rev. Genet. 2008, 9 (10), 764−775. (3) Storch, K. F.; Lipan, O.; Leykin, I.; Viswanathan, N.; Davis, F. C.; Wong, W. H.; Weitz, C. J. Extensive and divergent circadian gene expression in liver and heart. Nature 2002, 417 (6884), 78−83. (4) Li, Y.; Li, G.; Wang, H.; Du, J.; Yan, J. Analysis of a gene regulatory cascade mediating circadian rhythm in zebrafish. PLoS Comput. Biol. 2013, 9 (2), e1002940. (5) Zhao, Y. B.; Fent, K. Progestins alter photo-transduction cascade and circadian rhythm network in eyes of zebrafish (Danio rerio). Sci. Rep. 2016, 6, 21559.
3289
DOI: 10.1021/acs.est.6b01166 Environ. Sci. Technol. 2016, 50, 3287−3289