Title: The relationship between Fibroblast Growth Factor 1 (FGF1) and insulin signaling in the hypothalamic arcuate nucleus
Authors: Jennifer Wong, OMSII; Nicole Richardson, PhD; Jarrad Scarlett, MD, PhD
Introduction
Diabetes is one of the most costly diseases to our healthcare system and recently, members of the fibroblast growth factor (FGF) family of proteins have shown to restore normoglycemia in diabetic rodent models. FGF1 is a unique member of this family, as a single intracerebroventricular (icv) dose of FGF1 to the mediobasal hypothalamus, specifically the hypothalamic arcuate nucleus (ARC), is sufficient to maintain lower glucose levels in diabetic rodents for months. Previous data has shown icv FGF1 improves hepatic insulin sensitivity in type 2 diabetes (T2D) models, but its effect to insulin sensitivity in the brain and in wild-type mice remains unclear. We hypothesized FGF1 would improve insulin sensitivity in the ARC and tested normal, chow-fed mice and mice on a high fat diet.
Methods
We measured insulin sensitivity via phosphorylation of kinase AKT, a downstream readout of the insulin signaling pathway. Male C57Bl/6J mice were either fed a regular chow (n=20) diet or preconditioned on a high fat diet (n=20) for twelve weeks. Each group was split in two by body weight to receive either FGF1 (n=10) or vehicle (n=10) icv injection, then 5 hours later, half of each group was given 4U/kg of insulin or saline via intraperitoneal injection. Mice were sacrificed 15 minutes after insulin administration. Separately, another group of chow fed males (n=20) followed the same experimental paradigm, but were treated with 5U/kg of insulin or vehicle, and sacrificed 30 minutes after insulin administration. Brains were fixed in 4% paraformaldehyde, dehydrated with 30% sucrose solution, processed into coronal sections, then stained for pS473-AKT using immunohistochemistry. Image processing was performed using Fiji to measure the mean fluorescence intensity, and results were collated and visualized using GraphPad Prism.
Results
While we expected to see that FGF1 would influence insulin signaling in the ARC, we did not observe a treatment effect of FGF1 in either the chow fed or high fat diet groups or an effect of insulin to increase pS473-AKT signaling in the ARC.
Discussion
Upon further experimentation, we expect the chow fed group of animals given a higher dose of insulin for a longer duration will be more informative to our initial question and hypothesis. Our data will be informative in determining if FGF1 alters insulin sensitivity or potentiates insulin signaling in the brain. This information is crucial to discovering FGF1’s mechanism of action, leading to the desirable goal of improve human therapeutic treatments for T2D.
Two questions from a judge: 1) You showed two graphs of blood glucose vs. time after intervention, both of which showed average blood glucose of 200-300 mg/dL. In humans, these would be diabetics. In mice, is this normal or expected? 2) In your Results and Discussion panel, several of your bar graphs had only 4 dots (presumably representing individual rats), but the previous discussion suggested there were 5 rats per group. What happened to the data from the missing mice?
Hi Dr. Elliott, thank you for your questions!
1) Yes, I agree that typically in humans, an average blood glucose of 200-300 mg/dL is deemed as diabetic and it’s similar in the mice we used. For healthy, non-diabetic mice, a normal fasting blood glucose level is 80-100 mg/dL when measured after a 4-6 hour fast. Though it was not expected to remain as elevated as it did and was optimistically intended to be closer to the normal range, the novelty of the experiment was more in the success of sustained lowering effects on blood glucose after icv or intra-arcuate injection for days to weeks.
2) When analyzing the data, we excluded mice due: (1) inadequate tissue processing, resulting in damaged coronal sections and (2) mice that had variable background and fluorescent intensities within the same subject, suggesting inconsistent pAKT staining for IHC analysis. Therefore, those mice were not included in our results.
A judge would like to know, given your hypothesis was not supported might there be something unique to wild-type mice that might explain the results (versus mice used in prior studies)? Also, outside of Dr, Scarlett’s lab, what does this study contribute to moving this body of work forward (in relation to other-related studies)?
Hi Dr. Fritz, thank you for your inquires. There could be unique features to this cohort of mice, however, I don’t believe the wild-type mice in this experiment significantly differed from the mice previously used in Dr. Scarlett’s preliminary data. Instead, I believe our hypothesis was not supported more because there may be other avenues that contribute or interfere with the pAKT signaling we quantified. Near the end of my presentation, I mentioned that pAKT is only one of the possible pathways for FGF signaling; therefore, we could explore other pathways, other hormones such as Leptin, or specific neurons in the arcuate nucleus such as POMC or NPY. This study contributes to the scientific field as a whole in understanding the interaction of neuronal pathways and hormones related to diabetes (and obesity) in mice. And if we continue to have success in our exploration, this knowledge can be applied to develop new pharmacologic targets or treatment options in modern medicine for humans with T2DM.
Thank you for your presentation. As a judge, I would like to know more about the FGF1 and the endocrine FGFs. You state the endocrine FGFs have efficacy when injected ICV or administered systemically. The FGF1 was shown to lower blood glucose with ICV injection. What is the advantage (with human trial in mind) of focusing on the hypothalamic arcuate nucleus, even at a lower dose? Would this help to move to systemic delivery or is that not an aim at this point in the research?
Hi Dr. Salido, thank you for your interest in my project. In reviewing previous literature and data, the endocrine FGFs, namely FGF 19, 21, and 23, have an established role in the metabolism of bile acid, energy, and phosphate/Vitamin D, but is less specific to insulin and glucose management, which was the reason for Dr. Scarlett’s exploration of FGF1. The advantage of exploring the hypothalamic arcuate nucleus in mice is with the hope that we can find a potential pathway or receptor as a definitive target in mice first, then use that data to create new pharmacologic agents to better control patients’ blood glucose levels. Dr. Scarlett’s lab has previously explored systemic delivery of FGF1, but the consistency of results, dosing, and metabolism of FGF1 was not as promising. And, an icv dose is drastically smaller than dose needed for systemic administration, which is why Dr. Scarlett has ventured down that path. However, if we find success in narrowing down the specific target, then we can broaden back to systemic administration as this would more accessible for human trials in the future.