Time to ban antibiotic use in animal husbandry?

As a microbiologist, I sometimes worry that we are killing the golden goose in terms of our ability to combat human disease.  As I wrote earlier, the cost of developing and bringing to market new drugs is astronomical and many companies are shunning new drug development in favour of more lucrative endeavours, such as development of molecular-based test kits for human disease diagnosis.

In this light, it is disturbing to read of the amount of antibiotics used in animal production.  In 2001, Kristin Leutwyler wrote in Scientific American that a study by the Union for Concerned
Scientists had shown meat producers in the USA feed around 25 million pounds of antibiotics to chickens, pigs and cows for non-therapeutic purposes each year (the antibiotics suppress microbial populations and lead to increased growth rates of the animals).  Compared with this, only 3 million pounds per year are used for treating human disease.

In 2009, the House Committee on Rules held a hearing: H.R. 1549 - Preservation of Antibiotics for Medical Treatment Act of 2009, which would have amended the Federal Food, Drug, and Cosmetic Act to require the Secretary of Health and Human Services to deny an application for a new animal drug that is a "critical antimicrobial animal drug" (CAAD) unless the applicant demonstrated a reasonable certainty of no harm to human health due to the development of antimicrobial resistance attributable to the non-therapeutic use of the drug.

The CAAD is a drug intended for use in food-producing animals that contains specified antibiotics, or other drugs used in humans to treat or prevent disease or infection caused by microorganisms. The Rule would also require the Secretary to withdraw approval of a non-therapeutic use of such drugs in food-producing animals two years after the date of enactment of the Act unless certain safety requirements are met.

That rule died in Committee.  It has been reintroduced in the 2011-2012 Congress, and went to Committee on 9th March, 2011. 

It has been argued that there are no good studies linking antibiotic use in animal rearing to the development of antibiotic-resistant diseases.

While this may be so, and I'm not sure that it is, it is hard to imagine how else multiple antibiotic resistant strains of bacteria could be found in animals and ultimately in humans.  A classic experiment shown to most undergraduate students of microbiology over the last 50 years is the selection of antibiotic-resistant bacteria.  All that is required is the production of a concentration gradient of antibiotic across a Petri dish.  The agar is seeded with the target bacteria and those that grow at the most concentrated side are picked off and recultured on another gradient.  Very quickly, a resistant strain can be isolated.

This laboratory procedure is effectively repeated in animals fed sub-therapeutic levels of antibiotics in their diet or drinking water - the survivors of the first dose are constantly exposed and eventually dominate the population.  As microorganisms transfer genetic information among themselves, the number of different strains resistant to the antibiotics also increases.

I would be less concerned if the antibiotics used for treating animals were no longer in use for human therapy.  Then the benefits could perhaps be justified, though even then, the class of antibiotic is important, as bacteria becoming resistant to one member of the class might also develop resistance to others.  If humans become infected by strains resistant to antibiotics used in human therapy, there may be no cure available for the infection.  These infections could easily be transferred by handling raw meats from treated animals, though there is again a lack of sound scientific evidence.

The view is clouded by the indiscriminate use of antibiotics by humans.  It's not just a case of doctors prescribing antibiotics when they are not justified, though I have seen antibiotics prescribed for viral diseases, which generally can't be treated with these chemicals.  Look up "availability of antibiotics over the counter" and you will get pages telling you that there is none.  However, it is well known that antibiotics can be bought freely over the counter in some countries.  Even the antibiotics of last resort - the only ones that can treat certain diseases - can be bought in pharmacies in some Asian countries and there are papers in the scientific literature showing the levels of resistance in bacterial populations.

I truly hope that President Obama eventually signs this legislation and that leaders of other countries do likewise.  I would hate for us to descend to the state before the introduction of penicillin, when even a scratch from a rose thorn could result in painful death.

Those dreadful scientists - do they ever do any good?

 If you read the popular press, you could be excused for thinking that scientists, particularly the genetic engineers, are a pretty bad bunch.  Wandering round spooky laboratories wearing white coats, these bearded boffins invented preservatives that are now widely used in processed foods, they developed milk pasteurisation and homogenisation, and the genetic engineers made corn with built-in insecticide.  Food technologists are employed by food manufacturers to make food keep forever and fool the consumers.

My perhaps somewhat biased view is that this is absolute rubbish.  To name just a few good things, we have safe food and effective vaccines, a whole range of new fruit cultivars and wonderful test kits that can quickly diagnose all kinds of disease.  All these things were developed by scientists, technologists and engineers.

A recent report from Cornell University describes a new test to trace and identify outbreaks of foodborne disease.  So far this is applied only to a common Salmonella subspecies, but the principle can be applied to many other foodborne disease bacteria.

One way of recognising a specific bacterial strain, such as one suspected of causing a food poisoning outbreak, is to chop up the DNA into bits with enzymes and to amplify the bits, followed by separation of the fragments by gel electrophoresis, which is a way of creating a band pattern or fingerprint.  Alternatively, parts of the DNA can be amplified with random primers, or starting sequences, that will also produce a fingerprint.  You will have seen DNA fingerprints used for crime detection in various television series.  However, it is never as quick and simple as portrayed in these cop shows.  Unfortunately, closely related bacteria may produce band patterns that can't be distinguished, making it impossible to differentiate one strain from another.

With the development of very rapid sequencing techniques, it is now possible to determine the nucleotide sequence of the full bacterial genome.  In other words, we can read the whole genetic code of the bacterium.  Very closely related strains may differ by only a few nucleotides, or code letters.  By looking at these very small differences, we can tell if a particular strain was responsible for apparently linked illnesses in Germany, United Kingdom, New York, and France.  The technique is called Single Nucleotide Polymorphism (SNP) test.

The technique is still quite expensive, but as rapid sequencing is developed further, the cost is likely to decrease.  Being able to track an infecting bacterium, such as the Escherichia coli that caused so much disease and death in Europe earlier this year, is a valuable tool in fighting such outbreaks.  In the face of such devastating outbreaks of foodborne disease, the cost of full sequencing is insignificant.

The researchers, led by Martin Wiedmann, who developed the technique intend to continue perfecting the method and to apply it to other bacteria.


Of course, the description above is a gross simplification of the SNP test.  Wiedmann's original paper is highly technical.  You can read it in:
Applied and Environmental Microbiology, 2011; DOI: 10.1128/AEM.06538-11

Alternatively, you can read a press release from Cornell University at:
http://www.sciencedaily.com/releases/2011/10/111025113540.htm#.TqoxsV5BB1U.email