Mice and bacteria

Maxim Rousseau, Polit.<url>, based on the materials of Science: Kelly Servick, Mouse microbes may make scientific studies harder to replicate

Scientists note that the results of experiments conducted on laboratory mice are dependent on the species composition of the bacteria that live in these mice in the intestine.

Laura McCabe from the University of Michigan and her colleagues studied the effect of one drug on mouse bone density. After completing the experiment, they found that the mice receiving the drug had decreased bone density compared to the control group. McCabe decided to repeat the experiment for more confidence. She ordered a new batch of laboratory mice of the same genetic line and from the same manufacturer. The mice were kept in the same conditions, even in the same cages as the mice from the first experiment. But, to the surprise of scientists, the drug caused them not loss, but on the contrary, an increase in bone density. Biologists repeated the experiment for the third time, and now found that the drug has no effect on the bones of mice at all.

Trying to find an explanation for what happened, Laura McCabe suggested that three groups of mice differ in the composition of microbes living in their intestines. The scientists took samples of mouse faeces from each of the three groups and determined which bacteria they had. The assumption was confirmed. Indeed, the species composition of the bacteria was different. It is difficult to explain the reasons for this. It is possible that the bacteria got to the mice from the walls of containers during transportation or from the clothes of laboratory staff. But it is very likely that it was the different composition of bacteria that influenced the results of the study.

The problem with reproducibility of research results is one of the most important in modern biology. A few years ago, two pharmaceutical companies reported that they were unable to replicate the vast majority of scientific results in preclinical experiments. The reason for the appearance of non-reproducible results is usually considered an urgent need to be published, which is faced by most scientists, which reduces the quality of work. Another reason is seen in the unwillingness of publications to publish the results of negative experiments. But, as Francis Collins and Lawrence Tabak, head and deputy director of the US National Institutes of Health, wrote in 2014 in the journal Nature, some cases of non-reproducibility of results probably have simple practical explanations: differences in animal genetic strains, laboratory environment and "subtle changes in protocol". In other words, the results of the experiment change because the researcher unconsciously conducts a slightly different experiment.

The microbiome – a collection of bacteria living in the body, which also includes representatives of fungi, archaea, protozoa and even viruses – seems to scientists to be an increasingly significant factor in research conducted on laboratory animals. Moreover, it turns out to be a variable that cannot be controlled. It is very difficult to artificially unify bacterial communities in the organisms of the same mice. Destroying the same intestinal bacteria completely is also not an option, because they are of great importance for the health of the body and its immune reactions. Deprived of symbiont bacteria, mice will no longer be a suitable object for research. In addition, bacteria in the body, as shown by numerous studies in recent years conducted on both animals and humans, can affect susceptibility to various diseases, predisposition to obesity and determine the body's response to medications. More than a decade ago, a Pfizer research team discovered a strange change in rat urine. The relative concentrations of the two substances formed during the splitting of food have changed. Such a change could make the results of clinical studies of the excretion of drugs from the body unclear, since in these studies the concentration of their decomposition products in urine is usually determined. The researchers determined that rats with an unusual composition of urine come from only one room at the enterprise where they are engaged in breeding. As a result, they restored the previous composition of urine in a few weeks, simply seating the rats in other cells. At that time, there was no study of symbiont bacteria in these rats, but scientists believe that they were the cause of all the changes.

Last year, scientists from the University of Missouri, working with a mouse model of multiple sclerosis, unexpectedly achieved the disappearance of symptoms by simply adding an antibiotic to the drinking water of mice. The manifestations of the disease were restored when these mice were transplanted to other neighbors, communicating with whom they restored the composition of their intestinal bacteria.

Craig Franklin and Aaron Ericsson from the same university are trying to explain such effects and influence them by manipulating the composition of intestinal bacteria. According to them, even five years ago, microbiota analysis was rarely carried out in laboratories, unless this was the main purpose of the study. Now more and more scientists are sequencing fecal samples to determine the bacterial genes contained in them, or ordering such an analysis from colleagues. Franklin's lab is conducting this study at a cost of $125 per sample through the Mutant Mouse Resource and Research Center.

According to Franklin, biologists will soon indicate the composition of the microbiome of mice used in the study in the "Materials and Methods" section of their publications. Franklin even considers the practice of "microbiome to order" promising. To do this, his laboratory created four colonies of mice with different variants of the composition of intestinal bacteria found in four suppliers. If scientists want to use a specific type of microbiome in their research, they can transplant mouse embryos from their laboratories to females with the right composition of bacteria. And then compare the resulting mice with the original variants. But so far there have been no customers for a procedure worth several thousand dollars.

Last year, in an attempt to determine the "normal" microbiome of a laboratory mouse, scientists analyzed mouse excrement from two major suppliers. As in all mammalian faeces, there were mainly representatives of two groups of bacteria: Firmicutes, which are believed to play a role in the absorption of dietary fats, and Bacteroidetes, associated with a high fiber content. But the ratio and diversity of specific species varied greatly in mice from different suppliers.

In one group of mice, the absence of segmented filamentous bacteria (SFB) was noticeable. At the same time, recent studies have shown that SFB is associated with the development of an immune response in mice. In mice, these bacteria live mainly in the ileum in close proximity to the intestinal epithelium, where they are believed to help induce the activity of T helper cells 17, producing one of the varieties of interleukin.

The presence of SFT in the intestines of experimental mice now worries many scientists. Jennifer Phelan, a representative of Taconic Biosciences, a supplier of laboratory mice, says that the number of requests for the presence of segmented filamentous bacteria in mice has grown from one per month to several per week in a short time. Now the company has included information about SFT in mice in the standard set of information reported to customers, and offers scientists both mice that have these bacteria and those that do not have them. Phelan also advises researchers to order mice grown in the same room as the mice from the first experiment when they repeat the experiment.

Already this year, a team of scientists led by immunologist David Masopust from the University of Minnesota combined mice obtained from a specialized supplier with "simple" mice from a pet store in the same cells. The store mice had bacteria of diseases that have long been eradicated from laboratory mice, for example, hepatitis or pneumonia. As a result of the outbreak of diseases, almost a quarter of the animals died, but the survivors began to produce memory T-cells, key to the body's fight against infection. The authors of the study believe that such mice will be a more realistic model for studying the immune response in humans. Similarly, the staff of the immunological laboratory of Stephen McSorley from the University of California, Davis, who study the reaction of mice to chlamydia and salmonella, did the same. They settled mice in their vivarium, which they sold as animal feed at the zoo.

A recent paper by Franklin and Erickson indicates that pet store mice have a very rich intestinal microflora. The prospect that they will be able to present researchers with a more realistic model of the human intestine compared to standard laboratory mice seems very likely. But their use is hampered by high requirements for the purity of the experiment, in particular, for the identity of the animals used. If even small changes in the intestinal microflora in mice from special suppliers had a huge impact on the results of experiments, then what can we say about mice whose condition was not particularly controlled by anyone.

Biologists are trying to develop systems to monitor how microbes affect a particular study. First of all, it is necessary to learn how to separate the effects caused by the genes of bacteria from those caused by the genes of their host. To do this, as Herbert Virgin and Thaddeus Stappenbeck from Washington University in St. Louis suggest in the journal Nature, only mice from the same litter should be used as a control group. To determine the role of a particular gene, they write, it is necessary to get mice with and without this gene from one pair of heterozygous parents. Then the researcher will have mice that differ only in the gene he is interested in, and they will have the same microbiome. Vergine also points out the importance of factors affecting the composition of symbiont bacteria, such as feed and the use of antibiotics.

So intestinal bacteria not only complicate the reproduction of research results, but also suggest new ways to scientists. This conclusion was also reached by Laura McCabe, who studied the decrease in bone density in mice. She noted that mice whose bone density dropped in her experiment had more bacteria associated with inflammation in their bodies. This suggested to her the direction of further research.

Portal "Eternal youth" http://vechnayamolodost.ru  26.08.2016