Parkinson’s disease affects more than 270,000 people in France. By the time a diagnosis is made, the destruction of dopamine-producing neurons is too far advanced to consider curative treatments. To identify early markers of the disease and enable preventive treatments, Professor Boulet’s research team at the Grenoble Institute of Neurosciences has developed a metabolomic analysis using Nuclear Magnetic Resonance.
Just as analyzing a car’s exhaust gases during a vehicle inspection—by looking for combustion products—can help detect potential engine malfunctions, nuclear magnetic resonance (NMR) metabolomics enables the identification of all small organic molecules (such as certain amino acids) or their degradation products resulting from the cascades of molecular events occurring within cells. The identification of these substances, or metabolites, within a biological sample—such as blood or other tissues—therefore makes it possible to detect potential markers of disease (biomarkers). This approach follows the study of genes (genomics), the entire set of messenger RNAs (transcriptomics), and the entire set of proteins (proteomics). This innovative approach thus makes it possible to understand the full range of factors—genetic, environmental, nutritional, and others—whose combined effects may lead to a disease.
The researchers compared metabolites from three complementary animal models of Parkinson's disease and subjected blood samples from newly diagnosed but untreated patients ( de novo patients) to the same analyses.
Complementary animal models
In 1959, biologists William Russell and Rex Burch proposed ethical guidelines for animal experimentation known as the “3Rs,” which continue to serve as the consensus standard for the ethical treatment of laboratory animals today. These guidelines aim to replace the use of animals with alternative models whenever possible, reduce the number of animals required, and refine experiments by minimizing the stress imposed on the animals (pain, suffering, distress, etc.).
It seems to run counter to the principle of reduction to use a variety of different animal models in a single study. In this study, using three models made it possible to cross-validate the data and increase the “translational” relevance of the results—that is, their applicability to humans. In fact, each model exhibited complementary characteristics of Parkinson’s disease, with each mimicking different phases of the disease, ranging from the phase in which only motivational deficits are present to the phase with motor symptoms. In addition, the researchers developed a scale for rating disease progression modeled after those used in human clinical practice, assessing motivational and motor performance as well as the extent of neurological damage. This enhanced the translational aspect of these studies and also contributed to refining the experiments.
The first model involves rats treated with a neurotoxin that targets dopaminergic neurons. This makes it possible to create groups of animals that consistently mimic a specific stage of the disease without any changes over time; this represents a simplification of the human disease, in which each patient is a unique, evolving case. In addition, the researchers selected a second model reflecting the progressive nature of the disease, whose biomarker is linked to the production of a harmful protein. Although this model allows researchers to track the progression of the disease in a single animal, it does not encompass all the neuropsychiatric symptoms characteristic of the early stage of the disease (such as apathy).
Finally, the third model, induced by injecting a neurotoxin into macaques, replicates the progression of the disease’s clinical phases and exhibits greater homology with humans. The similarity of a model refers to its ability to mimic the underlying physiological mechanisms responsible for the pathology being studied in humans; this is therefore referred to as identity of causation and the mechanisms involved.
From Animals to Humans
In these animals, metabolic dysregulations led to the identification of six metabolites potentially linked to the neurodegenerative process, whose combined levels constitute a composite metabolic biomarker. This biomarker makes it possible to distinguish animals exhibiting Parkinson’s disease-like symptoms from controls, even in the early stages. This result would not have been achieved if only a single scenario—a single animal model—had been studied.
The results obtained in animals were then compared with those from de novo patients (diagnosed but not yet treated) from two different databases. The same six metabolites that make up the biomarker identified in animals made it possible to distinguish de novo patients from healthy patients with a high level of sensitivity and specificity.
The researchers also demonstrated that the dysregulation of 3 of the 6 metabolites comprising the biomarker can be partially corrected by a drug that mimics the effects of dopamine. This treatment corrects the animals’ motivation deficit in the early phase. This partial improvement was also observed in a subset of patients in the study. This reinforces the value of the identified biomarker and suggests that this marker could be used to monitor treatment progress in a less burdensome manner than current medical imaging methods.
In addition to the diagnostic implications of this study, this research has also identified a mechanism regulating cellular metabolism that could be targeted to counteract the effects of the disease, thereby offering therapeutic prospects. This illustrates, in particular, just how closely fundamental and applied research are intertwined. We have just seen this through this example, but numerous studies demonstrate the value of animal experimentation, as in the case of Pasteur’s work on rabies, the research conducted in Grenoble on deep intracerebral electrical stimulation, which has revolutionized the care of many Parkinson’s patients worldwide, or, closer to home, the research on messenger RNA that led to the development of COVID-19 vaccines and earned its authors the 2023 Nobel Prize in Medicine.
What should we make of transposability?
Beyond these scientific advances, critics of animal testing point to the limited transferability of results obtained in animals to humans; one of their arguments is that“90 percent of treatments successfullytested on animals prove to be ineffective or dangerous for humans.” This, in particular, allows them to justify the use of alternative models to animals, such as in vitro or in silico models.
However, this quote seems to us to be incomplete and biased. It has not been demonstrated that the transferability of in vitro and/or in silico tests is better given the current state of the art.
In fact, these tests serve as the initial screening stage in the entire preclinical testing process, and the role of animals is essential because, by the end of the preclinical phase, they make it possible to rule out 40 percent of drug candidates—particularly based on potential risks to humans.
Furthermore, after highlighting this limited transferability, the authors of the cited text propose instead to emphasize the robustness of experimental approaches, ensuring rigorous research in both animals and humans. They thus suggest “harmonizing the design of experimental protocols conducted in humans (clinical trials) with that of preclinical studies conducted in animals—the use of which remains indispensable—thereby dispelling the black-and-white view that pits a pro-animal experimentation community against another that favors alternative methods.”
Attributing cognitive abilities to others—whether human or animal—provides us with a common foundation of moral rights and gives meaning to our actions. The field of animal experimentation is no exception to this rule.
It has been argued that researchers who use animals for scientific purposes have less empathy for animal suffering than the rest of the population.
This observation seems to us to overlook an important point. Just as a surgeon does not faint at the sight of blood at the crucial moment of an operation, as researchers working with animal models, we reflect on the consequences of our actions. The emotional aspect is managed, but not suppressed: researchers have no interest in neglecting animal suffering. An experiment on an animal model loses its inferential power if some animals are in a state of panic, others are stressed, or others are prostrate. On the contrary, the quality of science in this field goes hand in hand with animal welfare. The refinement process mentioned above also aims to limit symptoms to the absolute minimum necessary for studying the disease, and it is inseparable from a reasoned empathy toward the animals used.
It is also important to note that the quality of our medical advances and our healthcare services is based on the assumption that the average number of animals each French person will “use” over the course of their lifetime is 2.6. This figure should be viewed in the context of the 1,298 animals, on average, that a French person will consume for food over the course of their lifetime. The use of animals in experimental research—as practiced today within a reasoned and controlled framework and with respect for animal welfare—remains, in our view, a necessity for the advancement of our knowledge, even if its “ultimate” ban, as enshrined in European law, should encourage us to limit the number of animals used and to develop alternatives.
Finally, with regard to effective enforcement of laws governing the use of animals in research—including the requirement to apply the 3Rs—departmental veterinary inspectors have at their disposal deterrent penalties that can be severe in cases of animal abuse.
As researchers, we know that animal experimentation is a controversial topic: our critics subscribe to the ideas put forward by anti-animal experimentation movements, in part because there is no public outcry from those who support this practice. That is precisely why it is our responsibility to report to our fellow citizens on what we do in our laboratories.
