Heresy about Ethidium Bromide

Recently I've received several copies of an email warning of the dangers of ethidium bromide (a chemical used to visualize DNA in agarose gels) and advertising a seminar about a safer alternative. The email originated from Invitrogen, the company that sells the alternative DNA stain ("SYBR safe"*) and is sponsoring the seminar. Here's the first two paragraphs of the email:

A real risk of lab science is accidental exposure to hazardous chemicals, especially ones that may be fatal if inhaled, harmful if swallowed or absorbed through skin, causing irritation to skin, eyes and the respiratory tract and causing heritable damage.

Ethidium bromide has just such characteristics and is frequently in use at many UBC labs. A safer alternative is available that is non-toxic, non-mutagenic and is not classified as cytotoxic waste when disposed. To alert the UBC research community to this safer alternative called SYBRSafe Gel Stain, the Department of Health Safety and Environment and the UBC Sustainability Office are sponsoring a seminar presentation by Invitrogen on the product comparison and practical application of SYBRSafe.

This would be fine if ethidium bromide (EthBR) really was a serious hazard. But it's not. And it's actually less toxic than this alternative.

Ethidium (also called homidium) was developed as a treatment for trypanosomiasis (African sleeping sickness), and is still used in Africa where resistance to it is not a problem. The following quote is taken from the Encyclopedic Reference of Parasitology (2004): "...homidium is generally well tolerated at recommended dose and also at higher dose levels (no systemic toxicity)..." (source). The recommended dose for cattle is 1mg/kg body weight (up to 50mg/kg has been used in mice). Compare this with the 0.25 - 1 microgram/ml used in molecular biology (previous error corrected - thanks, anonymous commenter). A 50kg researcher would need to drink 50 liters of gel-staining solution to get even the non-toxic dose used in cattle.

Here's information from a materials safety data sheet (MSDS) provided by Fisher Scientific:

Ethidium Bromide: (inhalation, rat): 0.0118 - 0.1340 mg/L/6H. (oral,rat): 1503 mg/kg. sts determined to be toxic based on: LC50 (inhalation rat) 0.0472mg/L/1H * 100 (dilution rate)=approximately 4.72mg/L/1H.
Carcinogenicity:
CAS# 7732-18-5: Not listed by ACGIH, IARC, NTP, or CA Prop 65.
CAS# 1239-45-8: Not listed by ACGIH, IARC, NTP, or CA Prop 65.

Epidemiology: No information found
Teratogenicity: No information found
Reproductive Effects: No information found
Mutagenicity: Possible mutagenic effect in humans. The suspicion is based on proven damage to the genetic material in the somatic cells of man and animals and requires further clarification.
Neurotoxicity: No information found

Translating as best I can, half of mice given 1.5g /kg body weight will die, as will half of mice that spend 6 hours breathing air containing about 0.12mg EthBr dust/liter. There is no evidence of neurotoxicity or of birth defects or reduced fertility even at the high doses routinely used in cattle.

Despite the evidence, the "EthBR is very dangerous" meme has become dogma in molecular biology. This is because EthBR can increase the frequency of mutations, a topic on which most people are very irrational.

Ethidium is a flat molecule, just the right size to get in between the stacked base-pairs of double-stranded DNA. This, and its fluorescence, are what makes it a sensitive dye for detecting DNA. The presence of ethidium when DNA is being replicated can cause DNA polymerase to slip, creating short insertion or deletion mutations. Here's a 1999 reference. However there's no direct evidence that exposure to EthBr causes mutations, tumors or birth defects in any animal, and its routine use at high doses in cattle suggests that it doesn't.

Excessive concern about mutagenicity can make us overlook short-term toxic effects, and here EthBr is the safer dye. The reference above found that the SYBRsafe alternative was actually much more toxic than EthBr to the bacterial cells used in the mutagenicity tests. SYBRsafe was toxic at concentrations as low as 1 microgram/ml, whereas EthBr toxicity was not observed until 250micrograms/ml. The authors suggest that this is because living cells are much more permeable to SYBR green than to EthBr. But a MSDS for SYBR safe reports a LD50 for rats of >5g/kg, which is higher than that of EthBr (1.5g/kg). As both these LD50s are many orders of magnitude higher than the concentrations used in molecular biology, toxicity of gel staining solutions is trivial compared to the risks of of burns from melted agarose or slipping on spilled gel buffer.

Perhaps the largest real hazards associated with use of EthBr in molecular biology are the methods used to inactivate it. Some labs now incinerate all waste containing even a trace of EthBr, and others absorb it onto activated charcoal. Harsher methods involve use of bleach and sodium hydroxide, or hydrophosphorous acid and sodium nitrite, all much more dangerous than EthBr.

A final issue is cost, and here SYBR dyes may indeed be better, even though they cost a lot more per use than EthBr. This note compares the costs of different DNA stains used in agarose gels, and concludes that the new SYBR stains may save money because their greater sensitivity allows use of smaller amounts of the costly size standards. For example, this photo shows that SYBR gold can reveal bands containing as little as 100pg of DNA. (It also shows how big a problem fluorescent dust flecks can be when using SYBR stains.)

Despite the evidence of its safety, all the scientists I know continue to believe that EthBr is dangerous. It's not that we differ on how the evidence should be interpreted - they have never examined the evidence and see no point in doing so, even when I tell them that it contradicts their beliefs. I find this scary.

*I haven't been able to find out what SYBR safe is, nor the actual concentrations of the products on the market. It's all "proprietary information".