The Effect of Neurodegeneration on Ultrasonic Vocalisations (USV) and Their Neuronal Substrates in Mice and Rats: A Systematic Review

This systematic review synthesizes existing research on how neurodegenerative diseases affect ultrasonic vocalizations and their underlying neuronal substrates in rodent models, aiming to bridge preclinical and clinical findings to identify biomarkers for early diagnosis and therapeutic strategies in human neurodegenerative conditions.

The Gist

🎙️ The Big Picture: Listening to the "Silent" Scream

Imagine you are trying to diagnose a disease in a patient who can't speak. You can't ask them, "How do you feel?" or "Is your voice shaky?"

This paper is about scientists trying to solve that problem by listening to mice and rats. But here's the catch: these animals don't speak human language. They speak in ultrasonic frequencies—sounds so high-pitched that they are like a dog whistle to our ears. We can't hear them, but special microphones can.

The researchers asked a simple question: When a mouse or rat brain starts to break down (like in Parkinson's or Alzheimer's), does their "voice" change before their body does?

They found that, yes, the voice changes first. It's like the "check engine" light on a car dashboard turning on before the engine actually starts smoking.

---

🧠 The "Control Tower" of the Voice

To understand why the voice changes, you have to understand how the animal makes sound. Think of the animal's brain as a busy airport control tower.

1. The Midbrain (PAG): This is the main tower. It decides, "Okay, it's time to talk!"
2. The Brainstem: This is the runway crew. They tell the lungs to push air and the throat muscles to vibrate.
3. The Forebrain (Cortex): This is the air traffic controller deciding what to say based on the situation (e.g., "I'm scared!" or "I want to mate!").

In healthy animals, this system works smoothly. But in neurodegenerative diseases (where brain cells die), the control tower gets foggy, the runway crew gets tired, and the air traffic controller gets confused.

---

🐭 What Happens in the Models? (The "Disease" Scenarios)

The scientists looked at many different "broken brain" models. Here is what they found, translated into everyday terms:

1. Parkinson's Disease (The "Stuttering" Voice)

• The Problem: In Parkinson's, the brain loses dopamine (the chemical that helps us move smoothly).
• The Sound: The animals' voices become quieter, shorter, and less complex.
  • Analogy: Imagine a singer who used to hit high notes and hold them for a long time. Now, they can only whisper short, flat notes. Their voice sounds "flat" and lacks energy.
• The Finding: Even before the animal starts shaking or dragging its feet, its voice has already lost its "spark."

2. Huntington's Disease (The "Silent" Bird)

• The Problem: This is a genetic disease that messes up the brain's wiring from a very young age.
• The Sound: The animals simply stop making as many sounds.
  • Analogy: Imagine a baby bird that usually chirps constantly to get food. In Huntington's models, the baby bird just sits there quietly. It's not that the bird can't chirp; it's that the brain forgot how to start the song.
• The Finding: This happens in baby animals, suggesting the disease disrupts the brain's "music lessons" before the animal is even fully grown.

3. Alzheimer's & Dementia (The "Confused" Socializer)

• The Problem: The brain loses its ability to remember and connect with others.
• The Sound: The animals become less social. They don't call out to their friends or mates as much.
  • Analogy: Imagine a party where everyone is shouting to be heard. In Alzheimer's models, the animal just stands in the corner, ignoring the party. It's not that they are shy; their brain is too busy with internal chaos to notice the social cues.

4. Tauopathy (The "Broken" Wiring)

• The Problem: A specific type of protein (tau) clumps up and clogs the brain's wires.
• The Sound: The results were mixed. Sometimes the voice got quieter, sometimes it got louder, depending on the specific model.
  • Analogy: It's like a radio with a loose wire. Sometimes the signal cuts out completely; other times, it just gets static. This tells scientists that the "clogs" affect different parts of the voice system in different ways.

---

🔍 The "Why" Behind the Changes

The paper digs deep into the brain to find why the voice changes. They found that it's not just one broken part; it's a whole orchestra falling out of tune.

• The Dopamine Connection: In Parkinson's, the "fuel" (dopamine) for the voice engine is gone.
• The Serotonin Connection: Surprisingly, they found that serotonin (a mood chemical) is also messed up in the brainstem. This is huge because it means voice problems might not just be about movement; they might be about the brain's emotional and sensory wiring too.
• The "Check Engine" Light: In many cases, the voice changed before the animal showed physical signs of disease. This is the most exciting part. It means we could potentially detect these diseases in humans by analyzing their speech patterns long before they lose their ability to walk or think clearly.

---

⚠️ The "Messy Lab" Problem

The authors admit that comparing these studies is like trying to compare apples and oranges because:

• Different Tests: Some scientists tested the animals while they were courting a mate; others tested them when they were scared.
• Different Mics: Some used high-tech microphones; others used older ones.
• Different Animals: Some used male rats, some used female mice.

Because of this, the results aren't always perfectly identical. It's like trying to judge a singing competition where some judges are listening in a quiet room and others are listening in a noisy stadium.

---

🚀 The Takeaway: Why This Matters

This paper is a roadmap for the future. It tells us that speech is a powerful window into the brain.

If we can learn to "speak mouse" (or rat), we can:

1. Catch diseases early: Detect Parkinson's or Alzheimer's years before the patient feels sick.
2. Test medicines faster: Instead of waiting years to see if a drug helps a patient walk, we can listen to the mice. If their "ultrasonic songs" get better, the drug is working!
3. Understand the human brain: Since the brain circuits for voice are similar in mice and humans, fixing the "broken radio" in a mouse might help us fix the "broken radio" in a human.

In short: The next time you hear a dog whistle, remember that inside that high-pitched sound is a secret code that might one day help us cure some of the world's most devastating diseases.

Technical Summary

1. Problem Statement

Neurodegenerative diseases (NDs) such as Parkinson's (PD), Huntington's (HD), Alzheimer's (AD), and Frontotemporal Dementia (FTD) are characterized by early and progressive impairments in speech and communication. While speech deficits are prominent clinical markers in humans, the underlying neurobiological mechanisms and the potential for early detection via vocal biomarkers in preclinical models remain underexplored.

• The Gap: There is a lack of a comprehensive synthesis linking specific neurodegenerative pathologies in rodent models to alterations in Ultrasonic Vocalizations (USVs) and their corresponding neuronal substrates.
• The Objective: To systematically review existing literature on how neurodegeneration affects USV production in rats and mice and to correlate these behavioral changes with specific neural circuit dysfunctions.

2. Methodology

• Design: A systematic review following PRISMA 2020 guidelines, prospectively registered in PROSPERO (ID: CRD420250615249).
• Data Sources: Searches were conducted on PubMed, Web of Science, and Embase (as of January 16, 2025) without date or language filters.
• Selection Criteria:
  • Inclusion: Laboratory rats and mice; models of neurodegeneration (genetic, neurotoxin, overexpression, drug-induced); age/gender-matched wild-type controls.
  • Exclusion: Non-rodent models, in vitro/in silico studies, humans, therapeutic intervention controls, reviews, and conference abstracts.
• Screening Process: From an initial pool of 4,690 studies, 3649 were screened by title/abstract, 47 by full text, resulting in 33 final included papers (21 rat studies, 14 mouse studies).
• Data Extraction: Extracted data included animal models, USV elicitation paradigms, acoustic parameters (e.g., call number, duration, frequency, bandwidth), and neural findings (neurochemistry, histology, imaging).

3. Key Contributions

• Synthesis of Heterogeneous Data: The review consolidates findings across diverse disease models (PD, HD, AD, Tauopathy, FTD-ALS) and species, identifying consistent patterns despite methodological variability.
• Neural-Behavioral Correlation: It explicitly maps USV deficits to specific neural circuits (e.g., nigrostriatal dopamine, brainstem serotonergic systems, periaqueductal gray), moving beyond simple behavioral observation to mechanistic understanding.
• Identification of Biomarkers: It highlights specific USV features (e.g., reduced call number, bandwidth, and peak frequency) that serve as sensitive, early biomarkers for neurodegeneration, often preceding overt motor symptoms.
• Methodological Critique: The paper critically analyzes the lack of standardization in USV elicitation paradigms (e.g., courtship vs. maternal separation) and reporting (e.g., sex differences, age specificity), offering recommendations for future research.

4. Key Results

A. Disease-Specific Findings

• Parkinson's Disease (PD):
  • Models: Most studied in Pink1-/- rats and 6-OHDA neurotoxin models.
  • USV Deficits: Consistent reductions in the average bandwidth and peak frequency of 50-kHz frequency-modulated (FM) calls. Reduced call number and intensity were also observed, though intensity results were contradictory across studies (likely due to sex/age differences).
  • Neural Substrates: Deficits correlate with dopamine depletion in the substantia nigra pars compacta (SNc) and striatum. Crucially, serotonergic (5-HT) and noradrenergic alterations in the brainstem (hypoglossal nucleus, dorsal raphe) were identified as early drivers of vocal dysfunction, potentially explaining why vocal deficits do not always respond to dopaminergic therapy.
• Huntington's Disease (HD):
  • Models: R6/1 and BACHD mice; tgHD rats.
  • USV Deficits: Significant reductions in call number and total calling time observed in both pup (developmental) and adult stages.
  • Neural Substrates: Early dysregulation of dopaminergic and glutamatergic signaling in the striatum. Imbalances in oxytocin and vasopressin systems in the hypothalamus were linked to social vocal deficits.
• Alzheimer's Disease (AD):
  • Models: TgF344-AD, McGill-R-Thy1-APP rats; APP/PS1 mice.
  • USV Deficits: Reduced call number and altered acoustic structure (e.g., increased duration of FM calls in late stages). Cholinergic lesions reduced fear-associated 22-kHz calls without affecting freezing behavior.
  • Neural Substrates: Linked to cholinergic depletion and circadian clock gene dysregulation.
• Tauopathies (FTD/ALS):
  • Models: Tau.P301L and Tau.P301S mice; C9BAC mice.
  • USV Deficits: Tau.P301L showed age-dependent reductions in call complexity and number. Tau.P301S pups showed increased vocal output early on. C9BAC mice showed no USV changes despite significant neural pathology (synaptic loss).
  • Neural Substrates: Hyperphosphorylated tau accumulation in the Periaqueductal Gray (PAG) and brainstem nuclei (KF, NRA) was observed, though the direct correlation with vocal phenotype was inconsistent.

B. General Trends

• Early Onset: Vocal deficits often appear in prodromal stages, sometimes before motor symptoms or significant neuronal loss.
• Common Metrics: Across almost all models, reduced call number and reduced call duration were the most robust and consistent findings.
• Paradigm Variability: The "courtship" paradigm was the most common elicitation method, but variations in female estrus phase control and social context led to significant heterogeneity in results.

5. Significance and Implications

• Translational Bridge: The review establishes USVs as a quantitative, cross-species biomarker that bridges preclinical rodent models and human speech deficits (e.g., hypokinetic dysarthria in PD, aprosody in HD).
• Early Diagnosis: Since vocal changes often precede motor symptoms, USV monitoring could facilitate earlier diagnosis and intervention in human NDs.
• Therapeutic Targets: The identification of non-dopaminergic systems (serotonin, norepinephrine, oxytocin) as critical for vocal control suggests new therapeutic targets for speech and swallowing disorders in NDs that are currently unresponsive to standard dopaminergic treatments.
• Future Directions: The authors call for:
  1. Standardization: Unified protocols for USV elicitation and acoustic analysis.
  2. Inclusion: Systematic inclusion of female subjects and longitudinal tracking across the lifespan.
  3. Circuit-Level Analysis: Investigating the earliest neural events underlying vocal dysfunction, particularly in brainstem circuits.

In conclusion, this systematic review validates USVs as a sensitive readout of neurodegeneration, providing a roadmap for using rodent vocalizations to decode the neural mechanisms of human speech disorders and accelerate the development of early diagnostic tools.