ALCOHOL USE DISORDER

FROM ADDICTION PHENOTYPES TO THERAPEUTIC SCREENING

Alcohol use disorder needs scalable in vivo models that quantify binge, withdrawal, and chronic-use biology with decision-ready endpoints.Lunai combines zebrafish, controlled alcohol exposure, light-stimulus phenotyping, and machine-vision readouts to enable high-throughput therapeutic discovery.

Alcohol Use Disorder (AUD) affects an estimated 14% of US adults. It is common, morbid, and costly, yet only a small set of approved medications exist and responses vary. Rodent models can capture key biology but are slow and low-throughput, limiting empiric discovery across large chemical space. Lunai’s AUD program builds a scalable zebrafish-based screening platform with automated readouts of addiction-cycle phenotypes. The goal is to de-risk and accelerate novel AUD therapeutic discovery by linking compound effects to reproducible in vivo endpoints and downstream translation.

STEP 1 - DEFINE SCREENABLE AUD-RELEVANT PHENOTYPES AND OPTIMIZE ASSAYS

WHAT WE DID:

We established a light-stimulus phenotyping workflow and created a bespoke suite of AUD-relevant behavioral readouts designed for scalable, automated measurement in living systems.

OUTCOMES:

We established dose-dependent binge/intoxication signatures and quantified statistically significant differences between alcohol-exposed and control animals across multiple behavioral measures.

IMPLICATIONS:

This step establishes reproducible, automatable endpoints suitable for screening and creates a consistent measurement framework for linking compound effects to addiction-cycle biology in living systems.

STEP 2 - CONFIRM PHARMACOLOGIC RESPONSIVENESS AND SEPARATE ADDICTION-CYCLE PHASES

WHAT WE DID:

We tested whether ethanol-driven phenotypes are reversible with a clinically used first-line AUD therapy in acute and recovery/withdrawal paradigms. We also evaluated withdrawal across timepoints and developed chronic exposure paradigms to define a workable window between biological effect and toxicity.

OUTCOMES:

The benchmark AUD therapy partially rescued ethanol-induced behavioral changes, and this effect was reproducible across independent experiments. Across withdrawal paradigms, we identified assay conditions that produce consistent, dose-dependent effects and clearly separate challenge from withdrawal states. In chronic exposure, we observed measurable behavioral changes and a clear toxicity boundary at higher exposure, defining a practical screening window.

IMPACT:

This step anchors the platform to modifiable biology and defines operational windows and failure modes early, improving hit quality and making downstream follow-up more efficient.

STEP 3 - SCALE TO HIGH-THROUGHPUT DISCOVERY + TRANSLATION-READY FOLLOW-UP

CURRENT WORK:

Now that scalable assays are in place, we are initiating screening for compounds that reduce ethanol-driven phenotypes using automated behavioral profile comparisons, then confirm effects with dose-response re-screening and toxicity filtering, followed by integrated phenotyping and translation to mammalian pharmacokinetic studies, to inform dosing and validation in rodent AUD models.

ANTICIPATED OUTCOMES:

The program is designed to support screening at the 1,000-compound scale with built-in safety filters. It also provides a translation path from behavioral hits to integrated systems readouts (cardiac, hepatic, metabolic), and then to mammalian feasibility and efficacy testing.

IMPACT:

This step will enable empiric discovery at scale while preserving in vivo biology and early safety signals, tightening the loop from screening to decision-ready follow-up.