# High-dimensional chemical targeting of multiple neuron types to regulate energy balance

> **NIH NIH DP1** · BETH ISRAEL DEACONESS MEDICAL CENTER · 2024 · $800,000

## Abstract

Summary
Recent drugs for treating obesity and diabetes, while promising, have major side effects. For example, drugs
that reduce food intake by targeting the melanocortin-4 receptor (MC4R) or the glucagon-like peptide-1 receptor
(GLP1R) induce side effects such as nausea and vomiting. Several drugs targeting these receptors likely act in
part via effects on neural activity in the arcuate nucleus (ARC) and paraventricular nucleus of the hypothalamus
(PVH). My collaborator, Brad Lowell, and others have shown that AgRP and POMC neurons in the ARC are
each necessary and sufficient to strongly affect daily food intake, in large part via actions on PVH neurons that
express MC4R (PVHMC4R). Yet, manipulation of these or other cell types one at a time does not accurately mimic
what occurs during a meal, when many neuron types change their activity simultaneously and in different ways.
For example, we and others showed that AgRP neuron activity decreases and POMC neuron activity increases
during feeding. Critically, our lab recently discovered that jointly mimicking these opposing shifts in activity using
a combined optogenetic and chemogenetic approach is far more effective in reducing food intake than
manipulating either type alone. We will test whether coordinated regulation of activity in AgRP, POMC, and
many other feeding-related neuron types can be achieved using AI-guided combinatorial pharmacology,
and whether this more effectively promotes satiety and reduces body weight in a mouse model of high-
fat diet-induced obesity. However, a major overarching challenge in basic and clinical studies of neural control
of energy balance is that effective tools for jointly and selectively activating or inhibiting many specific neuron
types are either very challenging even in mice (in the case of 2 types8) or do not exist in any animal model or in
humans (>2 types). Further, many off-target effects of drugs likely arise from actions on off-target neuron types,
and from the fact that target ‘neuron types’ found to regulate satiety are actually composed of multiple molecularly
and functionally distinct subtypes.
I hypothesize that a novel computational pharmacology approach can
overcome limitations in coordinated optogenetic control of many neuron types and subtypes, while providing a
more direct bridge to novel drug therapies. Our lab is taking a unique approach to this challenge, using machine
learning algorithms together with single-cell sequencing of the unique repertoires of receptor expression across
dozens of satiety-related neuron types in hypothalamus and other brain areas. Using this approach, we have
devised mixtures of 2-10 ligands that can jointly activate or inhibit all or specific transcriptional subtypes of AgRP
or POMC neurons while sparing other neuron types. We will directly test the impact of increasingly high-
dimensional mixtures on various satiety-related neurons using 3D two-photon calcium imaging and
pharmacology in brain slices or in awa...

## Key facts

- **NIH application ID:** 10936558
- **Project number:** 1DP1DK139958-01
- **Recipient organization:** BETH ISRAEL DEACONESS MEDICAL CENTER
- **Principal Investigator:** Mark L Andermann
- **Activity code:** DP1 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $800,000
- **Award type:** 1
- **Project period:** 2024-09-01 → 2029-05-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10936558

## Citation

> US National Institutes of Health, RePORTER application 10936558, High-dimensional chemical targeting of multiple neuron types to regulate energy balance (1DP1DK139958-01). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10936558. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*