Well it has been a little while since I wrote a post, and a lot has happened since then, which I am sure you read about in our blogs and tweets. We changed our company name, partnered with Novartis and started partnering with patient groups to engineer our organisms to carry disease mutations and screen for potential therapeutic candidates. Things are moving really quickly, and we are looking forward to expanding our disease pipeline.
A lot of exciting things are happening but we still have to characterize our lead compound PERL101 to determine possible mechanism of action and see if it will correct the disease in Niemann Pick Type C mouse models. As mentioned in a previous post, we run our mouse studies at Vium. And we were lucky to get some excellent disease characterization data using their sensored cages, such as what happens to motion, breathing rate, weight loss and more. They also conducted some behavioral studies on the animals to capture neurodegeneration. This included rotarod, gait in an open field, and walking on a beam. With rotarod we looked at how long animals were able to stay on the rotating rod, in the gait assay we looked at the change in animal movement or if the mouse was shaking, stumbling or hunched and with the beam test we looked to see how far the mice could walk on a straight beam. These tests were extremely valuable, and based on the results, we felt that the neurodegenerative effect in the disease state was best captured by the rotarod assay. So going forward, we will most likely add that to future studies.
The reason why I am mentioning all of this is because while these metrics are really informative, we also wanted to be able to add biochemical measurements to our suite of metrics to determine if PERL101 is rescuing disease. So not only should we see if physical neurobehavioral metrics change upon treatment, but we should also look to see if the brain itself is healthier.
In order to better understand the neurodegenerative properties of NPC disease, we characterized the cerebellum (the part of the brain that is responsible for coordinated muscular activity and degenerates in NPC disease). The cerebellum is full of neurons called Purkinje neurons, named after the person who discovered them, Jan Evangelista Purkyně. Purkinje neurons are highly branched (many dendritic spines) and are the largest neurons in the brain. I really like the illustration below.
Figure 1. Illustration of Purkinje neurons
In a healthy brain, these Purkinje neurons are present throughout the cerebellum, however in a diseased brain, these neurons begin to die. In the NPC KO mice we use, which display a severe form of the disease, there is almost complete Purkinje neuron loss by mouse age day 63 (D63). It has also been shown that there is some neuron depletion in these animals at D42. This mouse age corresponds to approximately a 20year old human. What we wanted to determine is, at what age is Purkinje neuron loss evident? Knowing this will help us determine at what age we should treat the NPC KO animals, as well as at what age should we look for correction of the Purkinje neuron loss (correction of neurodegeneration). What is really great, is that we can stain for the presence of Purkinje neurons by detecting the Calbindin protein. This protein is a calcium binding protein and is responsible for buffering calcium in these neurons in response to stimulation of glutamate receptors (responsible for neural communication, memory formation, learning, and regulation).
Not only did we stain for Calbindin, but we also stained for the presence of CD68. CD68 is used as a marker of inflammation in the brain. It is a glycoprotein that is expressed in monocytes, macrophages and microglia. Microglia cells are the macrophages (immune cells) of the brain, and are responsible for clearing damaged cells or infections. The presence of CD68 is often used to measure the inflammatory response in the brain and is expressed more highly in many neurodegenerative diseases, including in the brains of Niemann Pick C mice and humans (brown areas of stain in panel (C) and (D) are of CD68).
Figure 2. CD68 in the brain of NPC1 patient samples
So what did we do to confirm the presence and onset of neurodegeneration in the NPC KO (NPCm1n) NIH mouse model line we are using for studying disease progression and potential correction? We designed a study with 24 mice total, no WT controls, just NPC KO animals. Luckily we were able to capture data in both male and female mice. Our groups were divided as indicated in the table below. The Group category is at what age the animals were sacrificed for brain characterization. We decided to choose these mouse ages based on previous data we collected on these animals and also the literature. We observed a decline in weight and motion around day 45 in the past, so we want to see if Purkinje neuron death and brain inflammation correspond to the onset of these observed characteristics. This will help us gauge when we need to treat animals in order to prevent neurodegeneration. If we treat once Purkinje neurons begin to die, we might be able to prevent further loss but there is also a chance too many neurons were already lost to see rescue of disease. Therefore, it would be best to treat animals prior to observed Purkinje neuron loss for maximum prevention of neurodegeneration.
| Group | Animals | Tissue and blood collected | Cerebellum stain |
| Day 35 | 3M and 3F | Brain, liver, muscle, spleen and terminal plasma | Calbindin and CD68 |
| Day 45 | 3M and 3F | Brain, liver, muscle, spleen and terminal plasma | Calbindin and CD68 |
| Day 55 | 3M and 3F | Brain, liver, muscle, spleen and terminal plasma | Calbindin and CD68 |
| Day 65 | 3M and 3F | Brain, liver, muscle, spleen and terminal plasma | Calbindin and CD68 |
Most animals in the Day 65 Group did not make it to Day 65 since there is a decline in health prior to that age in this disease model, therefore we sacrifice those animals as soon as we observe signs that it is no longer humane for the mice to remain in the study.
We fixed the brain tissue in 4% paraformaldehyde so that all of the structures and proteins are preserved and sent those brains to Histowiz for sectioning and immunohistochemistry (staining of the tissue with antibodies specific to the protein markers we are looking for). Histowiz was a great find because they automate the sectioning, staining and data recording process for pathology, which greatly complements our research pipeline. An example of the saved digital slide is shown below:
Figure 3. Example image of Histowiz digital slide
You can view these slides online and zoom in to look at specific regions of the tissue as well as the stain in detail.
So what did our data look like? We submitted brain tissue for both male and female mice from the groups in the previously discussed table. I will show you the results from the female animals since both male and female mice displayed similar neurodegenerative results based on our timepoints and antibody stains.
Below in Figure 4a and b are images of cerebellar slices from four female NPC KO mice at various days post birth, D35, D45, D55 and D63. In 4a you are looking at slice the entire cerebellum, more or less, and in 4b magnified regions from the cerebellum of each animal. The purple stain is haematoxylin, which stains nuclei. It is widely used in histology to identify cells. The brown stain in Figure 4 is CD68.
Figure 4: Timecourse of CD68 protein expression in the cerebellum of female NPC1 KO mice
a.
b.
From the data presented you can see that D35 animals have low levels of CD68 in the cerebellum, meaning minimal inflammation in the brain. However, as the NPC KO animals age, there is a marked increase in CD68 in the cerebella of these animals, meaning neuroinflammation increases as the disease progresses. By the time animals are D45, there is a significant amount of CD68 in the brain.
Below in Figure 5a and b are images of cerebellar slices from four female NPC KO mice at various days post birth, D35, D45, D55 and D63. The brown stain, brown circles and projections (Figure 5b) are the Purkinje neuron cell bodies and dendritic spines.
Figure 5: Timecourse of Calbindin protein expression in the cerebellum of female NPC1 KO mice
a.
b.
What we observed is that Purkinje neurons are present in high amounts in the cerebellum of D35 animals, which corresponds with the low CD68 levels. But by D45, there is a decrease in the number of Purkinje neurons, which corresponds with the presence of neuroinflammation, CD68. By D63, almost all Purkinje neurons are lost. From these data we are were able to mark at what age we observe neuroinflammation and neuron loss. This will greatly help us decide at what age we should expose animals to treatment to prevent neurodegeneration. Interestingly we captured some other metrics on these animals while they were housed at Vium, since the cages are sensored.
We noticed that breathing rate and body weight both start to decrease by D45. This corresponds with when we start to observe signification neuroinflammation based on the CD68 stain and Purkinje neuron loss based on the Calbindin stain. By better understanding the onset of these disease metrics, we can design optimal treatment studies to capture potential rescue of disease progression.
