Septic shock- replacing sick mouse ‘models’ with healthy human volunteers makes for better science.

The Oxford Medical dictionary defines sepsis as “the putrefactive destruction of tissues by disease-causing bacteria or their toxins” and there is no argument that sepsis, or ‘blood poisoning,’ is a serious health problem. Around the world, sepsis kills more people than AIDS, breast cancer and prostate cancer combined. In people, sepsis can affect anyone but is more dangerous in the very young or the very old, and it’s usually triggered by an infection, often in the lungs, abdomen, pelvis or urinary tract.

Sepsis can occur when an infection, like pneumonia, spills over into the blood and the resulting, widespread inflammatory response causes extensive damage throughout the body, resulting in death for up to 50 percent of people with the severe form of sepsis.

Image copyright ttsz

Sepsis treatment has remained the same for the last 30 years, despite the best efforts of researchers and the hundreds of thousands of animals killed in attempts to find a better treatment or to work out the mechanism(s) of sepsis in humans. Mice are the most frequently, but by no means the only, animals used. Rabbits have bacteria injected into their bellies, along with other compounds known to ‘enhance lethality’. Beagle dogs – purpose-bred to die of sepsis – have bacterial-infected clots inserted into their abdomens. Baboons are infused with huge numbers of live bacteria so that the levels of bacteria in their circulatory systems are way beyond those found in people with the condition. And yet, none of these animal models predict the human disease or accurately reflect the effects of treatments. For example, tifacogin, a drug that was effective when given to the rabbits, failed to have a positive effect in human clinical trials.  Antibodies that had improved human survival did not save dogs.  Interventions that helped baboons did not have such a good effect in people.

So if rabbits, dogs and baboons are not good as ‘models’, then that must be why we use mice – performing caecal ligation and puncture, or simply CLP, on them. CLP is considered the ‘gold standard’ for animal models of sepsis and it is as ominous as it sounds. The operation creates a deliberate hole in the part of the small intestine known as the caecum, so that the gut contents will leak out into the abdominal cavity. This intentional destruction of the intestinal barrier has devastating effects- the gut contents are full of dangerous bacteria that, once ‘freed’ from the confines of the intestines, can run rampant in the body and produce a widespread, life-threatening infection, or sepsis.

It’s easy enough to paraphrase the reasons given by the people doing this work as to why mice are used. Mice are very small, readily available, cheap to buy and maintain, not thought of as pets so their use is apparently more ethical than using dogs or cats, and mice can be easily genetically manipulated. There are also many laboratory reagents available to measure mouse responses.

There’s not much here about the science, however. Important details are missing concerning the similarity of responses in mice and in people, the precise recapitulation of human symptoms, the mechanisms of disease as they operate in mice and people, and the ability of mouse models to predict the likelihood of success of a possible treatment, for example.

Looking more closely at the science, in fact, there are many significant differences between the mouse response to deliberate sepsis and the human disease. For example, it is well known that mice are highly resistant to endotoxin (one of the causative agents of sepsis and the factor behind most experimental sepsis). It takes around 2,000 times MORE endotoxin to give mice the low blood pressure, high heart rate and signs of inflammation that resemble the symptoms of sepsis in a person. One reason for this possible difference is that mice usually inhabit an entirely different, less hygienic environment than people, and so they need more resilience against dirt and bugs. Unsurprisingly then, mice have different defences to people.  In fact, a component was recently discovered in mouse blood that suppresses the damaging response to endotoxin.  Mice are also very different to people genetically. A study in 2013 showed that the genes activated in the human response to burns, trauma or endotoxin were very closely related to each other, but bore little resemblance to the mouse genes switched on or off under similar circumstances. This tells us that the pathways involved in the response to endotoxin are likely to be quite different in mice compared to people – further questioning the use of mice as ‘models’ of human disease.

It’s natural enough to hear people say, “We can’t just test on humans, so we have to use mice.”  Obviously, we cannot give people increasing doses of endotoxin to see how much will kill them, but we can (and we have) give healthy consenting volunteers very small amounts of endotoxin (about ten thousand times less than is given to the mice) to gauge the effects.  Then, in carefully controlled conditions, we can monitor their symptoms, measure the changes in the different components in their blood, and apply the resulting insights to test possible new drugs. In fact, this very approach has resulted in several clinical trials of much-needed new therapies. These studies have helped us to work out when it is important to treat, as we can figure out when those vital changes in the blood occur, and we can see that endotoxin has wider effects on the body beyond an inflammatory response – influencing stress hormone release.

Recently the Netherlands National Committee for the Protection of Animals in Scientific Research (NCad) produced its vision of the transition to non-animal research. The committee members acknowledge that effective and innovative use of technologies, including the generation of data from human volunteers, will be an important step in phasing out animal experiments and will aid in better understanding of the human disease. That’s something we all want, as better understanding of what happens when bacteria or endotoxin overwhelms the human system is the path to better treatments for sepsis.