New Technologies Applied to the Study of an Old Toxin – Arsenic

Home / In the Spotlight / New Technologies Applied to the Study of an Old Toxin – Arsenic

In the Spotlight

New Technologies Applied to the Study of an Old Toxin – Arsenic

Published: February 4, 2008
A recent study found that human maternal exposure to arsenic results in substantial gene expression changes in the newborns (Fry, et al., 2007). Cells from the umbilical cords were evaluated in this study. Hundreds of gene transcripts were altered in the exposed population, resulting in the identification of 105 modulated proteins (Figure 1). Eleven of these transcripts have been identified as a set of highly predictive biomarkers for prenatal arsenic exposure. The Fry et al. study is the first to show specific changes in human gene expression following prenatal arsenic exposure.

Epidemiological/biomonitoring studies are important components in assessing the extent of human exposure to chemicals found in the environment. Using techniques that document human chemical exposure effects at the molecular level, such as those of Fry et al., can also contribute to the understanding of the mechanism(s) of toxicity of the substance being assessed.

For example, pathway analysis of the arsenic-modulated transcripts in human newborns revealed that the affected biological processes include genes involved in cellular stress responses, transcription, signal transduction, inflammation, cell proliferation and adhesion, and apoptosis.

Three sub-networks were identified within this larger interaction network. The sub-networks involve transcripts of the following protein families:

  • The transcription factor NF-κb and the proinflammatory cytokine IL-1
  • Two stress-activated transcription factors: STAT1 and HIF-1α
  • Proteins involved in cell cycle regulation (JUN and FOS) and stress-response (IL-8)

Eleven potential arsenic biomarker gene products were identified. Network analysis was used to identify interactions among the 11 arsenic biomarker genes, and 8/11 of these genes are highly integrated with the proinflammatory cytokine TNF-α.

TNF-α and NF-κb pathways have been shown to be involved in pathways that contribute to oxidant production found in chronic inflammation (Gauss, et al., 2007).

Studies at the National Institute of Environmental Health Sciences (NIEHS) indicate that inorganic arsenic is a transplacental carcinogen in mice (Liu, et al., 2007). Prenatal arsenic exposure in mice resulted in changes in the expression of about 187 genes, and produced hepatocellular carcinoma in the adult male mice. It is thought that the types of genetic alterations found in the human and mouse prenatal studies could lead to tumor formation in the adult offspring.

Since the responses to arsenic exposure by most animal species are significantly different in degree and/or kind from those of humans (Agency for Toxic Substances and Disease Registry (ATSDR), 2007), the availability of human data to corroborate relevant biomarkers of human arsenic toxicity is essential.

Newborns from the Fry et al. study will be followed to determine how long the gene activation persists and its potential health effects.

Figure 1. Arsenic differentially modulated the transcripts of 105 known human proteins (red = up-regulated; green = down-regulated) in infants of arsenic-exposed mothers. Source: Fry, et al., 2007. PLoS Genet. 3(11), e207 [Open Access License].

Arsenic Toxicity

Arsenic is a naturally occurring element found in the soil and water in many parts of the world. It is also used in man-made substances such as pressure-treated lumber and pesticides. Human exposure to arsenic occurs primarily from drinking water or occupational exposure by inhalation, ingestion, or dermal absorption. Arsenic compounds are also a component of some chemotherapy treatments.

The toxicokinetics (absorption, distribution, metabolism, and excretion (ADME)) of arsenic have been documented, primarily in animals (ATSDR, 2007). Arsenic absorption depends on its chemical form; it becomes widely distributed in the body including the brain, and most of it is rapidly excreted.

The metabolism of arsenic has been summarized as follows (ATSDR, 2007):

  • Arsenites are oxidized to arsenates and methylated
  • Enzymatic methylation of arsenates occurs primarily in the liver, forming two metabolites (MMA and DMA)
  • The rate of methylated product formation varies among tissues
  • The proportion of methylated products and their rate or production varies among species

The US Department of Health and Human Services (DHHS), the US Environmental Protection Agency (EPA), and the International Agency for Research on Cancer (IARC) have determined that inorganic arsenic is carcinogenic to humans (ATSDR, 2007). Forms of organic arsenic are thought to be less toxic, although little is known about their human health effects. In animal studies, arsenic has been shown to cross the placenta and enter fetal tissues, and maternal exposures have resulted in malformed offspring and stillbirths (ATSDR, 2007).

Other chronic human diseases, including cardiovascular disease, diabetes, and neurologic damage (peripheral neuropathy, encephalopathy, and intellectual deficits in children) have also been associated with arsenic exposure (ATSDR, 2007). Developmental toxicity studies of arsenic, based on small studies of female employees or families living near smelter plants, have shown an increase in spontaneous abortion and congenital malformations, and a decrease in average birth weights. In vitro human genotoxicity studies of arsenic are generally positive, showing chromosomal aberrations, DNA damage and repair, or inhibition of DNA synthesis. Limited in vivo human genotoxicity studies have shown a “higher-than-average incidence of chromosomal aberrations in peripheral lymphocytes.”

A few of the confounding factors in the study of arsenic toxicity are its many varied chemical forms and the lack of toxicity and human exposure data for many of these chemcial forms, as well as the fact that “most laboratory animals appear to be substantially less susceptible to inorganic arsenic than humans” (ATSDR, 2007). Insufficient data are available to know whether significant species differences will be found with organic arsenic compounds.

In Vitro Methods for Elucidating Mechanism(s) of Arsenic Toxicity

Despite the fact that arsenic has been known as a poison to humans for centuries, and its toxicity has been evaluated in animal studies for decades, the mechanism(s) of action of arsenic toxicity is still being deciphered.

Epidemiological studies are extremely valuable, especially for substances like arsenic where the animal data may not be predictive of human toxicity. However, retrospective human data has its limitations, such as potential co-exposure to other substances, varied routes of exposure, unsubstantiated amounts of exposure, limitations on tissues for analysis, and so on.

In vitro human cell-based assays are therefore being used to elucidate the mechanisms of human arsenic toxicity. Recent studies using immortalized human uroepithelial (Huang, et al., 2007), bronchial epithelial (Chang, et al., 2007), and prostate epithelial (Benbrahim-Tallaa, et al., 2007) cells have identified genes/biomarkers associated with in vitro arsenic exposure. These are only a few examples of the many published in vitro studies.

Many of the in vitro studies report alterations in the expression of genes and/or proteins that have also been observed to be modulated in in vivo studies.

Experiments to assess biomarkers of arsenic toxicity vary from identifying the changes in several genes/proteins to assessing global changes in gene expression using microarray technology. Most genes will be modulated in response to multiple types of stimuli. Therefore, to be useful, biomarkers of a toxic effect must be specific for the substance being assessed. Biomarkers identified by in vitro methods must also be confirmed as having relevance to the in vivo response. The simultaneous assessment of many genes is useful because a set of genes that are modulated in response to a particular toxicant or chemical class can then be identified and used as whole as a “specific biomarker.”


Gene expression following arsenic exposure can be used to identify specific biomarkers of arsenic exposure, and to indicate potential mechanisms of arsenic toxicity. Arsenic-modulated gene expression can vary depending on the species evaluated, the cell/organ type, the chemical form of arsenic, metabolism of the compound within the body, and the experimental conditions.

Research results need to be carefully scrutinized. In spite of the many known species differences in responses to arsenic exposure, reports on potential molecular mechanisms of action of arsenic in humans are still being conducted using animal cells. Additionally, studies exposing extra-hepatic tissues to arsenic are commonly conducted without consideration of the effects of metabolism, including which metabolites will actually reach the cell-type being studied.

Some of the arsenic-modulated genes identified with in vitro methods are the same as those identified in the prenatal arsenic studies. Our hypothesis from looking across these different studies is that a comparative analysis of in vivo and in vitro human biomarkers might be useful in identifying biomarkers that can be used in in vitro assays to predict human toxicity to heavy metals such as arsenic. The human epidemiological data will also be useful in validating the biological relevance of in vitro biomarkers.

AltTox Toxicity Testing Resource Center (TTRC):

Applications of Toxicogenomic Technologies to Predictive Toxicology and Risk Assessment. The National Academies Press. 2007.

Agency for Toxic Substances and Disease Registry (ATSDR). (2007). Toxicological Profile for Arsenic (August 2007 Update). Atlanta, GA. US Department of Health and Human Services, Public Health Service. Available here.

Benbrahim-Tallaa, L., Webber, M.M. & Waalkes, M.P. (2007). Mechanisms of Acquired Androgen Independence During Arsenic-induced Malignant Transformation of Human Prostate Epithelial Cells. Environ. Health. Perspect. 115, 243-247.

Chang, Q., Bhatia, D., Zhang, Y., et al. (2007). Incorporation of an Internal Ribosome Entry Site-dependent Mechanism in Arsenic-induced GADD45 Alpha Expression. Cancer Res. 67, 6146-6154.

Fry, R.C., Navasumrit, P., Valiathan, C., et al. (2007) Activation of Inflammation/NF-κB Signaling in Infants Born to Arsenic-exposed Mothers. PLoS Genet. 3(11), e207. Available here.

Gauss, K.A., Nelson-Overton, L.K., Siemsen, D.W., et al. (2007). Role of NF-κB in Transcriptional Regulation of the Phagocyte NADPH Oxidase by Tumor Necrosis Factor-alpha. J. Leukoc. Biol. 82, 729-741.

Huang, Y.C., Hung, W.C., Kang, W.Y., et al. (2007). Expression of STAT3 and Bcl-6 Oncoprotein in Sodium Arsenite-treated SV-40 Imortalized Hman Uoepithelial Cells. Toxicol. Lett. 173, 57-65.

Liu, J., Xie, Y., Cooper, R., et al. (2007). Transplacental Exposure to Inorganic Arsenic at a Hepatocarcinogenic Dose Induces Fetal Gene Expression Changes in Mice Indicative of Aberrant Estrogen Signaling and Disrupted Steroid Metabolism. Toxicol. Appl. Pharmacol. 220, 284-291.