Neuropeptides in control of left-right neural circuits

Abstract

Despite extensive research on hemispheric asymmetries, the mechanisms regulating lateralized brain functions are incompletely understood. Growing evidence suggests that lateralized neural circuits are side-specifically controlled in part by neuropeptides acting as neuromodulators, paracrine factors, and neurohormones. This review highlights evidence supporting this concept in the contexts of lateralized pain processing in the amygdala, control of auditory signaling, lateralized interoceptive signaling, and side-specific endocrine regulation. Our focus is primarily on rodent studies, with supporting data from humans and non-mammalian species, including turtles and nematodes. Left-right side-specific control may be rooted in a bipartite, lateralized organization of neuropeptide systems. Neuropeptides with asymmetric actions may act locally within specific brain regions or be coordinated across the neuraxis. These findings converge on a model in which neuropeptides enable lateralized control through interconnected mechanisms spanning gene expression, neural circuits, and behavioral outcomes.

Neuropeptides as regulators of hemispheric lateralization

In bilaterian animals, the body and nervous systems exhibit external symmetry, while many internal organs are asymmetrically positioned and numerous neural functions are lateralized [1–7]. Many of the genetic and cellular mechanisms underlying left–right asymmetry patterning—and the resulting visceral asymmetry—have been characterized, including the identification of key regulatory genes and molecular pathways [1,2,8–13]. In this review, we focus on a related but distinct question: Are there mechanisms that regulate lateralized processes on the left and right sides of the nervous system and body, as well as regulatory systems that maintain balance between functionally symmetric processes across these sides?

Advances in research tools have accelerated the discovery of the mechanisms of left-right side-specific regulation of lateralized functions. Several lines of evidence, both early and recent, suggest that neuropeptides—the largest family of neuro-regulators—may selectively control neural circuits on the left and right sides of the neuraxis and consequently behavioral functions lateralized to the left or right hemisphere. Neuropeptides function as neuromodulators, neurotransmitters, and local mediators of autocrine and paracrine signaling, as well as systemic neurohormones [14–16].

In this review, we discuss studies, mostly in mammals, documenting the roles of neuropeptides in regulating lateralized neural circuits. We highlight four lines of evidence supporting the concept that left-right side-specific neuropeptide systems control lateralized neural functions: (i) the analysis of differential left and right pain processing in amygdala and connected brain areas, (ii) the control of auditory processing during maternal behavior lateralized to the left versus right auditory cortex, (iii) the processing of nutritive reward signaling by the right vagus nerve, and (iv) the contralateral mechanisms of neurological deficits induced by brain lesions. These studies have identified a wide array of neuropeptides including opioid peptides, calcitonin gene related peptide (CGRP), pituitary adenylate cyclase-activating polypeptide (PACAP), cerebellin, tachykinins, oxytocin, cholecystokinin (CCK), and arginine-vasopressin. Mechanistic evidence has been primarily derived from rodent models, while the concept of lateralized neuropeptide signaling has also gained support from human positron emission tomography and functional magnetic resonance imaging, pharmacological treatment and postmortem molecular studies, as well as from analyses in non-mammalian species such as turtles and C. elegans. In rodents, various patterns of lateralized behaviors, physiological functions, and asymmetric structural and molecular features of neural circuits have been identified (for reviews, see [12,17–23]) and some of them may be intertwined with neuropeptide asymmetries.

The curious case of amygdala lateralization and pain – Opioids and CGRP

The amygdala is a critical regulator of pain [24,25]. Curiously, the regulation of pain by the central nucleus of the amygdala (CeA) is lateralized in mice, rats, and humans [26] in ways that cannot be explained solely by anatomic features of ascending sensory or descending modulatory pathways [21,27], although several papers have described functional asymmetry of the brainstem parabrachial nucleus (PB) [28,29]. Here, we present examples of this amygdala asymmetry that highlight the role of neuropeptides (Box 1) in lateralization, especially the dynorphin/κ-opioid receptor (KOR) and CGRP signaling systems in rodents.

Box 1: Neuropeptides as regulators of neural circuits.

Neuropeptides are pervasive modulators of neural circuits, influencing processes ranging from sensorimotor control and neuroendocrine regulation to complex behaviors. Acting primarily via extrasynaptic G protein-coupled receptors, they modulate ion channels, neuronal excitability, and circuit dynamics across diverse spatial and temporal scales [15,16,121,122].

Neuropeptides constitute a “wireless” signaling layer of the brain, complementing the classical “wired” synaptic connectome [15,121,123–131]. In simple nervous systems, neuropeptidergic and synaptic connections overlap in only ~5% of all neuronal interactions, implying that most neuropeptides create new inter-neuronal communication pathways. Nearly all neurons express at least one neuropeptide and one neuropeptide receptor, with many expressing multiple ligands and receptors. This suggests a combinatorial logic of neuropeptide action, enabling the formation of co-regulated neuronal ensembles and layered functional connectivity [127–129].

The selective encoding of numerous neural circuits is based on the vast molecular diversity of the neuropeptide systems: Multiple precursor proteins, differential processing into bioactive fragments, a wide variety of receptor types and subtypes, and biased signaling at these receptors [121,122,126,132]. Neuropeptides are often co-released with traditional small molecule neurotransmitters, allowing fine-tuning of specific synaptic responses and target subsets of neural circuits [16,122]. These mechanisms expand the number of distinct signaling configurations. For example, opioid peptides are differentially processed from their precursor proteins and exhibit different truncation patterns and signaling bias profiles, allowing precise tuning of physiological and behavioral responses [132].

In a “wireless” signaling, neuropeptides act as paracrine and endocrine regulators, influencing neurons at considerable distances. Neuropeptides reach these distal targets through volume transmission, cerebrospinal fluid, and the bloodstream [133,134]. Their expression levels can exceed those of their receptors, permitting the formation of diffusible gradients or “peptide clouds” that target receptor-expressing neurons beyond their release area [95]. At distant sites – where their concentrations are low – neuropeptides operate effectively due to their high affinity for receptors. For example, hypothalamic oxytocin and vasopressin released into the CSF modulate distant neural regions and orchestrate complex behaviors [133,135].

At the network level, neuropeptidergic and synaptic connectomes are structurally intertwined. Specific neuron populations act as hubs for neuropeptide signaling, while synaptic hubs also integrate peptide-based “wireless” communication. The mesoscale architecture of the neuropeptidergic connectome is beginning to emerge [124,127–129]. A key question is whether this wireless layer serves to segregate and coordinate functional processes across spatially distributed circuits, such as those in the left and right hemispheres. This would enable the brain to integrate the activities of these circuits and generate coherent, system-level behaviors.

The amygdala, and the CeA in particular, is rich in neuropeptides that have been linked to pain-related neuroplasticity [30]. On the input side, nociceptive information reaches the CeA through a pathway that includes a direct projection from the PB delineated by the neuropeptide CGRP.

Convergence of PB input with polymodal signals from the basolateral amygdala onto chemically and functionally distinct CeA cell types such as corticotropin releasing factor (CRF)-containing neurons that co-express dynorphin, an endogenous KOR ligand, serve major output functions through widespread ascending and descending projections to behavioral control centers.

In the amygdala, dynorphin is synthesized primarily in CeA neurons including those containing CRF [30]. In a stress-induced, priming-based rat migraine model (sumatriptan priming with environmental stressor), a KOR antagonist nor-binaltorphimine blocked cephalic as well as extracephalic hypersensitivity when injected into the right but not left CeA (Figure 1A–B) [31]. Conversely, KOR activation with its agonist U69,593 in the right but not left CeA resulted in cephalic and extracephalic hypersensitivity in primed rats [31].

Figure 1. Overview of the roles of neuropeptides in lateralized response to pain in the central amygdala and parabrachial nucleus.

(A) In rats, activation of the presynaptic κ-opioid receptor (KOR) by dynorphin opioid peptides [30–32] in the left central amygdala does not significantly alter pain-like behaviors in models of functional pain while (B) dynorphins in the right central amygdala inhibit GABA release from parabrachial nucleus pre-synaptic projections, leading to disinhibition and hyperexcitability of corticotropin-releasing factor (CRF)-expressing neurons. This enhances pain-like behaviors in stress-associated functional pain syndromes [30–32]. (C) In the context of bladder pain in mice, the neuropeptide CGRP, acting through the CGPR receptor complex including the Calcrl GPCR, reduces pain-like responses in the left amygdala [42]. (D) In contrast, in the right amygdala, CGRP, acting through Calcrl, and PACAP, acting through the PAC1 receptor, increase bladder pain-like responses [40,42]. (E) The parabrachial nucleus conveys nociceptive sensory input to the central amygdala. In the left parabrachial nucleus of mice, activation of post-synaptic tachykinin receptor 1 (NK1R) by substance P (SP) does not significantly alter pain-like behaviors in models of neuropathic pain [29]. (F) In contrast, the right parabrachial nucleus is preferentially engaged via postsynaptic NK1R activation by SP, contributing to enhanced pain signaling in neuropathic pain [29]. Together, these findings suggest a right-lateralized amplification of pain processing by neuropeptides in the central amygdala and parabrachial nucleus. This asymmetry may stem from lateralized input–output circuit organization or from left–right molecular differences in neuropeptide systems—mechanisms that remain to be fully elucidated. Created with elements from BioRender.com.

In another stress-induced, priming-based rat functional pain model (morphine priming with environmental stressor), nor-binaltorphimine injected into the right but not left CeA restored impaired descending pain inhibition measured as diffuse noxious inhibitory control (DNIC) of hindpaw mechanical sensitivity and activity of nociceptive spinal dorsal horn neurons [32]. Diffuse noxious inhibitory control was induced by capsaicin injection into the forepaw as a conditioning stimulus, and antinociceptive effects of diffuse noxious inhibitory control were assessed on the responses to noxious test stimuli applied to the hindpaw [32]. In a related study, brain slice physiology data identified disinhibition of CRF neurons in the CeA as the neuronal mechanism of KOR signaling in the functional pain [33]. Nor-binaltorphimine increased feedforward inhibitory transmission driven by PB input without affecting monosynaptic excitatory transmission [33]. In the aforementioned studies, both dynorphin and KOR activity were seen to increase in the left and right CeA following either sumatriptan [31] or stress-induced priming [32], but only the right CeA KOR modulation had functional consequences in pain modulation. Taken together, there is evidence to suggest right-hemispheric lateralization of KOR signaling in the CeA enhances descending pain facilitation to mask descending inhibition through the synaptic modulation of the PB to CRF circuitry. An important goal for future work would be to identify dynorphin-KOR neural circuits and to examine whether lateralized peptides or receptors drive asymmetric nociceptive processing.

CGRP is a well-known modulator of migraine, with over half a dozen new FDA-approved therapies targeting peripheral CGRP receptors [34]. CGRP is also expressed throughout the central nervous system (CNS) [35]. In the amygdala of rats, CGRP modulates neuropathic pain [36] and inflammatory pain [37] with these studies showing CGRP being pronociceptive in the right hemisphere. The impact of CGRP signaling was explored in the right versus left CeA in the context of visceral bladder pain in female mice [38] (Figure 1C–D). CGRP projections to the amygdala arrive from the PB. This PB-to-CeA CGRP pathway was manipulated with excitatory and inhibitory optogenetics as well as pharmacology. When the left hemisphere CGRP projection was activated, this reduced bladder-pain like effects, both under naïve (acute) nociceptive conditions and after bladder injury. In contrast, activation of the right CGRP projection had the opposite effect, causing increases in pain-like behavior similar to what was seen previously in neuropathic and inflammatory conditions [36,37]. Pre-synaptic CGRP levels were reduced in the left amygdala only after injury, suggesting that regulation of CGRP was a driver of this lateralization [38]. These effects highlight the potential for CGRP to be used differentially in the left versus right hemisphere depending on whether there was acute noxious stimulation (e.g. overfilling of the bladder) versus stimulation after bladder injury (e.g. during a urinary tract infection).

In a mouse experiment with bilateral manipulation of CGRP after inflammatory bladder injury, there was no effect of simultaneous left and right CeA stimulation [38]. From a therapeutic perspective, we might hypothesize that bilateral targeting of CGRP with an antagonist may have minimal effect in patients suffering from visceral bladder pain. Another recent study found that CGRP input is anchored to CeA neurons through a transsynaptic signaling complex of cerebellin 1 connecting presynaptic neurexins to postsynaptic delta glutamate receptors GluD1/2 [39]. Downregulation of cerebellin 1 was predominantly found in the right CeA in an inflammatory pain model induced by complete Freund’s adjuvant, and GluD1 was downregulated only in the right CeA in inflammatory and neuropathic pain models. Injection of cerebellin 1 into the right but not left CeA had antinociceptive effects on mechanical hypersensitivity in the inflammatory pain condition by restoring synaptic structure.

A similar unilateral, right hemisphere only, effect was seen with another PB-to-CeA neuropeptide, PACAP, which is pronociceptive in the right CeA but has no effect in the left CeA during bladder pain modeling in mice [40]. Evidence for this right hemisphere dominance of the PB-to-CeA synapse has been strengthened by studies of inflammatory [28] and neuropathic injury [29] rodent models where the side of the body that was injured did not change the right hemisphere dominance. In inflammatory injury in rats, this right PB activation translated into synaptic potentiation of right PB-to-amygdala synapses [28]. Neurons expressing tachykinin receptor 1 (a target for the neuropeptide substance P) in the right PB but not in the left PB have been identified as a critical circuit that mediates early-phase neuropathic pain development [29] (Figure 1E–F) and the right PB to CeA synapse is strengthened in neuropathic injury [41]. Overall, these findings regarding dynorphin, CGRP, PACAP, and substance P reinforce the idea that lateralized neuropeptide signaling in the PB-to-amygdala circuit may underlie the asymmetric processing of acute versus chronic somatic and visceral pain, as supported by both experimental animal and human imaging studies [26].

Lateralization of Oxytocin in the Auditory Cortex Impacts Maternal Behavior

Lateralization of vocal processing is perhaps one of the oldest identified asymmetric organizing principles of brain function, including in humans, with left- and right-hemisphere cortical specializations observed for various aspects of sound, speech, and music processing [43]. Many animals display asymmetry in auditory activation and processing [23,44] driven by changes in development [22] and myelination [45]. Several mammalian species also have a right ear, left hemisphere preference for responding to conspecific vocalizations, as exemplified in early studies in mice [46]. Obviously, lateralized responses to conspecific vocalizations in non-humans is quite different from the computational complexities required for speech and language, but there may be some parallels or evolutionary antecedents in left-lateralization enforced by selective neuropeptide receptor expression.

One of the best studied examples of neuropeptides involved in lateralized auditory functions is oxytocin, particularly in relation to parental behavior (Figure 2). Oxytocin receptor expression has been found to be significantly higher in the left auditory cortex compared to the right auditory cortex of female but not male C57 mice [47–50]. This lateralization appears to be essential for processing socially relevant stimuli, such as pup distress calls, which are key to providing caregiving based on infant needs and urgency. The left auditory cortex is specifically critical for recognizing these calls, and oxytocin enhances this recognition by modulating local cortical microcircuitry and enabling rapid plasticity for behaviorally-relevant stimuli such as infant cries [47–49,51–53].

Figure 2. Oxytocin enables rodent maternal behavior via coordinated excitatory and inhibitory synaptic plasticity in the left auditory cortex.

(A) Left-lateralized oxytocin receptor expression in the female mouse auditory cortex. ~25% of cells express oxytocin receptors (largely inhibitory interneurons) in the left auditory cortex compared to ~20% of cells expressing oxytocin receptors on the right side [48,49]. This was observed for experienced mother mice and nulliparous (virgin) females but not males. (B) Oxytocin receptor activation decreases evoked inhibition in the auditory cortex [48]. Heightened oxytocin modulation increases spontaneous inhibition and might consequently lead to weakened inhibitory response to sensory stimulation such as pup calls (see column ‘Oxytocin-modulated’; note the smaller inhibitory response than during baseline). This disinhibition promotes NMDA receptor-dependent long-term potentiation of excitatory inputs (‘Excitatory LTP’, eLTP) followed by potentiation of inhibitory inputs (‘Inhibitory LTP’, iLTP) [48]. This increases spiking responses to stimuli paired with oxytocin modulation, in a manner regulated by strengthened co-tuned inhibition. (C) Nulliparous mice initially neglect or avoid pups. Cohousing with an experienced mother and pups leads to increased oxytocin release in the nulliparous female, when the mother interacts with her and shepherds her to the nest with the pups [48, 51, 52]. After minutes to days, most nulliparous adults then become effective co-parents. Created with elements from BioRender.com.

Oxytocin works by acting as a neuromodulator of cellular excitability and synaptic transmission, adjusting the balance between excitatory and inhibitory responses in cortical neurons (Figure 2A,B). In the neocortex, oxytocin receptors are largely expressed by interneurons, particularly somatostatin-positive inhibitory neurons in different cortical regions [48,53–55]. This increases the salience of pup distress calls, promotes long-term plasticity of call representations, and ensures more efficient behavioral responses such as pup retrieval [47–49,51–53,56,57]. In the auditory cortex and hippocampus, this occurs by a selective activation of oxytocin receptor-expressing interneurons, depolarizing these cells spontaneously. Direct inhibitory cell depolarization is thought to increase tonic GABAergic inhibition, which is hypothesized to produce a concurrent reduction in stimulus-evoked inhibition, possibly by depleting GABAergic vesicle pools, although the precise mechanism has not yet been discovered [58]. This mechanism both reduces spontaneous activity and increases evoked responses, essentially enhancing the ‘signal-to-noise’ ratio of incoming inputs. After repetitive stimulation, reduction of inhibition allows these changes to persist beyond periods of heightened neuromodulation by inducing NMDA receptor-dependent LTP of incoming inputs [48,49]. However, reducing inhibition while increasing excitation is potentially destabilizing in terms of cellular excitability and network function. More recent work has examined mechanisms of inhibitory plasticity, documenting how the strength of inhibition recovers and is enhanced to a level higher than the baseline in tandem with excitatory potentiation. It remains to be determined to what degree strengthened inhibition is due to more excitatory activity in the network, or if specific forms of inhibitory plasticity are induced downstream of NMDA receptor activation at co-activated synapses [46,53,57].

Oxytocin is synthesized in the hypothalamus, particularly in the paraventricular nucleus, from where it is released both into the brain and the bloodstream [53]. Hypothalamic neurons project directly to the auditory cortex, allowing oxytocin to influence sensory processing by binding to the oxytocin receptor. This direct pathway explains how oxytocin reaches the auditory cortex and fine-tunes the neural circuits that are responsible for processing social auditory signals. Lateralization of the oxytocin system in the mouse left auditory cortex is due to differential oxytocin receptor expression; there are higher levels of oxytocin receptor mRNA and protein in left auditory cortex [47,49,50]. In contrast, oxytocin axon projections to the auditory cortex are sparse but in similar numbers to the left and right hemispheres. In C57 mice, oxytocin receptor expression increases in both hemispheres over postnatal development but then decreases in males and females except for the female left auditory cortex which maintains higher levels of expression [47]. This indicates that lateralization of neuropeptide signaling might be actively constructed by critical period plasticity mechanisms in the auditory cortex; furthermore, receptor expression itself might not be fundamentally lateralized, but the interneuron subtypes that selectively express oxytocin receptors might be the more primary substrate for this process (see Outstanding Questions).

Outstanding Questions.

Arg-vasopressin, CGRP, oxytocin, opioid peptides and other neuropeptides have been identified as left–right side-specific regulators of brain functions. How general is this phenomenon across the broader neuropeptide family and asymmetric processes?

Does lateralized neuropeptide signaling modulate lateralized functions, or more broadly, maintain left–right physiological balance across symmetric areas and organs?

Are these systems plastic and able to adapt to unilateral challenges? Could dysfunction of lateralized neuropeptide signaling contribute to pathology? Might maladaptive changes influence lateralized nociceptive processing or conditions such as hemiparesis and hemiplegia — and could targeted interventions restore bilateral function?

Are neuropeptides acting on the same side functionally coordinated (e.g., CGRP, dynorphin, PACAP and substance P in the right parabrachial–central amygdala circuit)? Are peptides producing contralateral effects (e.g., Arg-vasopressin and Met-enkephalin in the HL-PA model) co-regulated?

Which hypothalamic–pituitary neurohormones released into the bloodstream transmit side-specific signals, and what are their peripheral targets?

What molecular and circuit-level mechanisms support lateralized neuropeptide actions? Do left and right circuits differ in neuropeptide and receptor expression, or in the neuronal subpopulations that carry them? Is lateralized expression of a single peptide sufficient for side-specific regulation — or do such signals act in concert?

Finally, how is neuropeptide asymmetry developmentally established? Are distributions and actions set genetically, or shaped by activity and experience?

Neuropeptide Lateralization in Ascending Function of the Vagus Nerve

The vagus nerve is a major carrier of visceral sensory information to the CNS, conveying mechanical and chemical signals from most internal organ systems, including the cardiac, respiratory, and digestive systems. As most visceral organs in mammals exhibit left-right asymmetry, it is perhaps unsurprising that vagal innervation of the viscera also exhibits anatomical and functional asymmetry [59–66]. The extensive branching and crossing of vagal fibers within the periphery present a particular challenge to the characterization of functional lateralization of parasympathetic sensorimotor signaling. Using traditional techniques, functional left-right asymmetry has nonetheless been observed for most visceral organ systems examined (Figure 3A). As an example of such laterality, the right vagus nerve, which has been shown to preferentially innervate the atria and sinoatrial node, is often seen to produce greater effects on heart rate than the left vagus nerve, an effect driven by acetylcholine release and modulated by neuropeptides including vasoactive intestinal peptide and neuropeptide Y. By contrast, the left vagus nerve preferentially innervates the ventricles and atrioventricular node and generally produces stronger effects on cardiac contractility compared to the right vagus nerve [67–69].

Figure 3. Lateralized peptide receptor expression in the vagus nerve contributes to asymmetric gut-brain sensory signaling.

(A) The left and right vagus nerves branch extensively to innervate nearly all visceral organs. Anatomical left-right asymmetry is seen in the branching patterns of afferent vagal fibers [59], whose cell bodies lie in the nodose ganglia (NG). Notable side-biased innervation of cardiopulmonary [60,61] and gastrointestinal [62,63] systems are indicated for left-nodose ganglia (blue) and right-nodose ganglia (red) neurons. (B) Digestive signaling is conducted in the upper gut by largely non-overlapping populations of mechanosensors (green; defined by expression of the mechanosensitive ion channel, Piezo2) and chemosensors [74,75], which include peptide receptor-expressing chemosensors (purple) and other chemo- and thermo-sensors (yellow). Recent studies have shown that many of these digestive peptide receptors, including CCK type 1 (Cckar), glucagon-like peptide type 1 (Glp1r), secretin (Sctr), and peptide YY (Npy2r) receptors, which are known to play a role in signaling satiety, reward, and the nutrient content of food, are often co-expressed in the same nodose ganglia neurons [62,64–66,76,77]. (C) In mice, the population of peptide-sensing nodose ganglia neurons (purple) is 1.5 to 2 times more numerous in the right nodose ganglia than the left nodose ganglia [62], contributing to right-biased processing of satiety and reward signals from the upper gut [64–66]. Created with elements from BioRender.com.

The enhanced spatiotemporal precision of genetic techniques has revealed previously unknown laterality including a gut-brain reward pathway conducted by the right vagus nerve [64,70]. Gut-innervating right (but not left) nodose ganglia neurons (NG) in mice preferentially innervate the nucleus of the solitary tract (NTS), and activation of this right NG-to-NTS pathway induces appetitive behavior via a multisynaptic network that additionally includes the PB nucleus and mesolimbic dopamine neurons, ultimately increasing striatal dopamine release [64]. Similar laterality is seen in rats: Female rats self-administer right, but not left, cervical vagus nerve stimulation, and right cervical vagus nerve stimulation self-administration is accompanied by enhanced cFos expression in the ventral tegmental area (VTA) and substantia nigra pars compacta (SNc) [70].

Neuropeptides are likely to play an important role in this lateralized vagal reward signaling (Figure 3B,C)[64]. The satiety-inducing effects of the digestive peptide CCK in mice are abolished following ablation of right, but not left, vagal neurons [64]. In the PB, distinct patterns of cFos labeling are seen following CCK versus lithium chloride treatment [64]. While administration of malaise-inducing lithium chloride activates avoidance-associated CGRP-positive neurons in the externolateral PB, CCK administration activates CGRP-negative, vesicular glutamate transporter 2-positive neurons in the dorsolateral PB that projected directly to the dopaminergic midbrain nuclei. Combined, these results suggest that distinct peptidergic signaling pathways through the PB likely mediate appetitive versus aversive consummatory behaviors [64].

Subsequent studies have identified differential right > left expression of multiple neuropeptide receptors in mice and rats, including CCK type 1 (Cckar), glucagon-like peptide type 1 (Glp1r), secretin (Sctr), and peptide YY (Npy2r) receptors (Figure 3C) [62,66]. The corresponding neuropeptides are released upon ingestion of sugars and fats, making them strong candidates for conveying the rewarding properties of food to the CNS via the vagus nerve. These neuropeptides play a role in regulating satiety and glucose metabolism, though behaviorally their role in lateralized vagal reward-related signaling remains to be demonstrated. Nor is it known whether other vagal branches innervating, for example, liver, spleen, pancreas, or other organs may contribute to reward-related nutritive or metabolic signaling from the periphery to the CNS. In mice, an increase in activity of the ventral tegmental area has been reported following stimulation of left nodose ganglia neurons [71], suggesting that multiple vagal reward pathways may exist. As left cervical vagus nerve stimulation is already widely used to treat neurological disorders including epilepsy [72] and stroke [73], additional research is needed to fully understand the functional and clinical relevance of neuropeptide left-right asymmetry within vagal reward pathways.

Endocrine signals mediating contralateral effects of brain injury

The cross-association concept states that each cerebral hemisphere is functionally connected to the contralateral side of the body via decussating tracts [78–82]. This concept derives from longstanding observations that motor deficits following brain lesions manifest mostly on the opposite side of the body [78–84]. Recent evidence indicates that contralateral effects of brain injury are mediated also through a humoral mechanism involving neurohormones (Figure 4) [85–89]. This was demonstrated in rats with complete spinal cord transection, which eliminates descending neural input. Subsequent unilateral lesions to the sensorimotor cortex produced hindlimb postural asymmetry characterized by contralateral limb flexion, asymmetric withdrawal reflexes, and lateralized gene expression in the lumbar spinal cord [86,89]. Left-hemisphere injuries induced right hindlimb flexion, and right-hemisphere injuries triggered left-sided effects. Hypophysectomy abolished these responses, while serum transfer from animals with unilateral brain injury to naïve rats replicated the side-specific effects—indicating the involvement of circulating hormonal factors [86].

Figure 4. The topographical neuroendocrine system (T-NES) mediating contralateral effects of unilateral brain injury.

(A) In the classical model, each brain hemisphere exerts control over contralateral body functions via decussating neural pathways. (B,C) In addition to neural transmission, unilateral brain injury may also engage blood-borne signaling mechanisms, whereby neural activity is encoded into lateralized neurohormonal messages in the hypothalamic–pituitary axis [85–89]. These neurohormones are released into the circulation and activate their receptors in the spinal cord or neurons projecting to the hindlimbs, producing side-specific motor effects. (D) The T-NES appears to be bipartite, with side-specific receptor modulation: In rats, hindlimb postural asymmetry induced by left cortical injury is blocked by the vasopressin V1B antagonist SSR-149415 and the DOR antagonist naltrindole [86,87,89]. In contrast, hindlimb postural asymmetry following right-side injury is inhibited by β-Funaltrexamine, a MOR antagonist [87–89]. (E) Intravenous transfer of serum from donor rats with left-sided brain injury into recipient rats with intact brains induces hindlimb postural asymmetry with right hindlimb flexion, demonstrating blood-borne transmission of lateralized signals [86]. (F,G) Several neuropeptides mimic injury-induced asymmetries. Intravenous or intrathecal administration of arginine-vasopressin, β-endorphin, or the DOR agonist Leu-enkephalin induces hindlimb postural asymmetry with right hindlimb flexion, while the DOR/MOR agonist Met-enkephalin and the KOR agonists dynorphin and U-50,488 induce left-side flexion [85–89]. (H) These asymmetric effects may be mediated by a lateralized spinal opioid system: The prodynorphin marker Leu-enkephalin–Arg and DOR mRNA are left-lateralized, while the proenkephalin marker Met-enkephalin–Arg–Phe and the KOR/DOR receptor ratio are elevated on the right side [85–89].

Search for the blood-borne factors that induce postural asymmetry identified arginine-vasopressin and β-endorphin as contralateral humoral messengers of the unilateral brain injury (Figure 4D–F) [86]. Intravenous administration of these pituitary-derived neuropeptides - neurohormones induced hindlimb postural asymmetry in animals with intact brains and fully transected spinal cords. The right hindlimb was flexed, which mirrored the effect of a left-sided brain injury and suggested that these peptides can reproduce lateralized motor outcomes. The postural effects of left cortical injury were blocked by SSR-149415, a V1B vasopressin antagonist, naloxone, a non-selective opioid antagonist, and naltrindole, a DOR antagonist [86,89].

These findings suggest the existence of a previously unrecognized topographic neuroendocrine system (T-NES)—a hemisphere-specific neuroendocrine axis relaying left-right side-specific signals to the contralateral body side. The T-NES may function via a three-step process: (i) Encoding hemisphere-specific neural activity into side-specific neurohormonal messages in the hypothalamus and pituitary; (ii) releasing these hormones into the bloodstream; and (iii) decoding them into spatially specific physiological responses at distant targets such as the spinal circuits or their projection to peripheral tissues [85,86,89].

Transmission of side-specific neurohormonal signals from anatomically symmetric hemispheres implies a bipartite and lateralized organization of the T-NES. Pharmacological evidence supports a dual-hemisphere structure of the T-NES, with distinct left- and right-sided subsystems operating through different sets of neuropeptides (Figure 4D,F,G) [86–89]. The postural effects of an ablation injury to the left or right sensorimotor cortex were differentially modulated by selective opioid receptor antagonists. The μ-opioid receptor (MOR) antagonist β-Funaltrexamine blocked hindlimb postural asymmetry induced by right-sided brain injury, whereas the δ-opioid receptor (DOR) antagonist naltrindole was effective only after left-sided injury [86,89]. Furthermore, both intrathecal and intravenous administration of endogenous opioid peptides and synthetic agonists in rats with intact brains reproduced the side-specific motor responses observed after unilateral brain injury. Specifically, the DOR/MOR agonist Met-enkephalin and KOR agonists dynorphin and U-50,488 induced left-sided responses, while the DOR agonist Leu-enkephalin triggered right hindlimb flexion [87,90–93].

These findings suggest that neuropeptides may regulate the activity balance of mirror-symmetric spinal neural circuits controlling the left and right hind limbs. In this mechanism, contralateral effects of unilateral brain injury mediated by neuropeptides may be counteracted by their ipsilateral actions [87,94]. This is illustrated by the observation that, following right-sided cortical impact, the KOR antagonists nor-binaltorphimine and LY2444296 did not reduce the magnitude of asymmetry but reversed its direction: Instead of contralesional (left) hindlimb flexion, the ipsilesional (right) limb flexed [87].

Transcriptomic analyses suggest that neurohormones in the hypothalamus – the putative encoding area – are organized into left- and right-dominant gene co-expression networks [89]. The hypothalamic networks exhibit asymmetric coordination with spinal gene expression—even in spinalized rats—supporting the existence of a multi-level, laterally organized endocrine communication system along the neuraxis. This coordination is ipsilateral; it differs between the left and right networks on each side and for both networks between the sides. Left-sided brain injury perturbed these patterns by affecting the left network [89]. These lateralized hypothalamic networks may serve as a molecular foundation for encoding of hemisphere-specific neural signals into neurohormonal messages, by amplifying subtle asymmetries in individual neurohormone expression.

T-NES signals are likely translated into lateralized motor outputs in the spinal cord or at peripheral nerve terminals. In support of this, opioid receptor expression in the spinal cord is lateralized, and levels of endogenous opioid peptides differ between the left and right sides (Figure 4H) [88,95]. These features may enable spatially selective interpretation of humoral signals and the generation of side-specific motor responses. It would be crucial to pinpoint spinal neural circuits that interpret these signals, leading to asymmetric functional outcomes.

While morphogenetic programs establish bilateral anatomical symmetry, the T-NES may serve as an adaptive mechanism for regulating left–right physiological balance (see Outstanding Questions). Its response to unilateral perturbations—such as brain injury—implicates it in the emergence of pathological asymmetries, including hemiparesis and hemiplegia.

Neuropeptides in other asymmetry scenarios

Beyond the core findings reviewed above, other studies support the existence of lateralized neuropeptide functions [94,96–114]. Early rodent work revealed that the neuroendocrine hypothalamus is asymmetrically organized, with right-side dominance in gonadotropin-releasing hormone expression [97–103]. This asymmetry may contribute to ipsilateral control of gonadal functions via neural or endocrine pathways. In humans, postmortem molecular analyses and positron emission tomography imaging have shown lateralization of opioid peptides and their receptors, potentially linked to hemisphere-specific processing of pain and emotions [104,105]. Neuropeptide expression also correlates with hemispheric differences in cortical architecture, and intranasal administration of vasopressin induces hemisphere-specific behavioral effects [106–108]. In rodent models, ghrelin asymmetrically modulates brain metabolism, and CCK-8 produces opposite behavioral responses when microinjected into the left or right hemisphere [109,110]. Developmental studies show left biased galanin expression in the embryonic heart and lateralized opioid effects in the mouse embryonic cortex [111,112]. In C. elegans, asymmetric insulin signaling drives associative learning (see below) [113]. In turtles, KOR and DOR are enriched in the left and right visual cortices, respectively, where they modulate contralateral visual input [94,96]. Taken together, these findings suggest that left–right side-specific neuropeptide signaling is a broadly conserved and functionally significant phenomenon across species, peptides, and physiological domains.

Neural basis of asymmetric neuropeptide functions

The relationship between macroscopic, circuit-level, and molecular asymmetries on one hand, and lateralized physiological or behavioral functions on the other, is an emerging focus in brain lateralization research [8,9,19,113,115]. While causal links across these levels remain to be defined in mammals, studies in simpler model systems—such as C. elegans, Drosophila, and Danio rerio—provide compelling evidence that functional lateralization is rooted in anatomical and molecular asymmetries. In the zebrafish habenula, structural and circuit-level asymmetries underlie lateralized decision-making behaviors [8,115]. In C. elegans, salt conditioning induces a shift in network bias from a left-sided sensory circuit to a right-dominant one, accompanied by asymmetric synaptogenesis and circuit remodeling [113]. This experience-dependent reorganization is mediated by paracrine insulin signaling, establishing a causal link between neurohormonal signaling, connectome asymmetry, and behavior.

Neural circuits may encode functional asymmetries through quantitative differences (e.g., in neuron number) or qualitative differences (e.g., connectivity patterns) between hemispheres [10]. Lateralized neuropeptide systems may be embedded within—or interact with—asymmetric neural circuits, jointly contributing to the generation and regulation of lateralized physiological and behavioral functions. Several hypothetical models can be envisioned (Figure 5A–D). Model 1: Symmetrically located left and right circuits have unequal numbers of neuropeptide-expressing neurons; Model 2: Structurally asymmetric circuits with lateralized connectivity and functions are controlled through symmetrical neuropeptide systems; Model 3: Structurally symmetric circuits with symmetrical connectivity and functions are differentially modulated by lateralized neuropeptide systems; Model 4: There are structurally symmetric circuits with symmetrical connectivity and functions in which a neuropeptide system operates exclusively on one side. Either asymmetric neuropeptide release or lateralized receptor distribution, or both, may be required for differential control of bilateral circuits. Some configurations may enable neuropeptides to selectively modulate left- and right-sided processes even within structurally symmetric neuronal assemblies (Figure 5C–D)—for example, for fine-tuning the left–right balance of otherwise symmetric functions such as limb coordination. Within these models, individual neuropeptides can operate in a “unilateral mode” with zero effects in one hemisphere or in a “bilateral mode” where opposite effects are seen in the two hemispheres (Figure 5E–F).

Figure 5. Putative models of the interactions between neuropeptide systems and lateralized neural circuits.

The schematics illustrate four possible models for how lateralized neuropeptides could influence lateralized physiological and behavioral function, and summarize the respective patterns of lateralized neuropeptide effects, as well as their putative coordination within the left and right side and across the two sides. In the first model, (A) symmetrically located left and right circuits differ in the numbers of neuropeptide-expressing neurons. In the second model, (B) structurally asymmetric circuits with lateralized connectivity and functions are controlled through symmetrical neuropeptide systems. In the third model, (C) structurally symmetric circuits with symmetrical connectivity and functions are differentially modulated by lateralized neuropeptide systems. In the related fourth model, (D) there are structurally symmetric circuits with symmetrical connectivity and functions in which a neuropeptide system operates exclusively on one side. (E-F) Neuropeptides may activate or inhibit neural circuits asymmetrically—acting selectively in the left or right hemisphere (unilateral mode) or producing opposing effects in the two hemispheres (bilateral contra-directional mode). We illustrate four situations that may involve different types of neuropeptide lateralization (see main-text for related discussion and references): (G) Schematic representation of lateralized neuropeptide modulation of pain processing in the central amygdala and parabrachial nucleus, with a putative coordination of neuropeptide effects in the right hemisphere. Abbreviation: Dyn-Kor, the dynorphin–κ-opioid receptor system. (H) Left-sided modulation of rodent auditory cortex activity by oxytocin, which can modulate dam-pup behavior. (I) Lateralized VNS effects on reward, associated with lateralized cholecystokinin expression in nodose ganglia. (J) Left–right side-specific hindlimb responses, elicited by the opioid peptides Met-enkephalin (Met-enk), dynorphin (Dyn) and Leu-enkephalin (Leu-enk), and the neurohormone arginine-vasopressin (AVP), that may be coordinated across peptides and the sides of the spinal cord. Created with elements from BioRender.com.

Concluding remarks

As seen in this review, neuropeptide asymmetric effects follow three main patterns (Figure 5G–J). First, some neuropeptides exert opposing effects on each side—for example, CGRP suppresses pain via the left amygdala but facilitates it on the right [38]. Second, certain neuropeptides act preferentially on circuits located on one side of the midline, as shown for cholecystokinin and oxytocin. Third, functionally symmetric left- and right-sided responses may be differentially regulated by neuropeptides whose receptors are lateralized to opposite sides. For example, KOR agonists and arginine-vasopressin elicit left and right hindlimb flexion, respectively [86,89]. These patterns may reflect a monopartite mechanism, which operates on one side, and a bipartite system, which balances processes across hemispheres. Lateralized neuropeptide systems can operate both locally and across brain and spinal cord regions along the neuroaxis. In these roles, neuropeptides may act as both paracrine modulators and systemic neurohormones. Notably, arginine-vasopressin and β-endorphin can transmit signals from the injured left hemisphere brain to the spinal cord via the bloodstream [85–89].

Lateralized neuropeptide regulation has been observed in both the amygdala and the PB—two regions with dense reciprocal connections and strong anatomical links to the hypothalamus [116,117]. The hypothalamus is a likely source of asymmetrically acting neurohormones such as oxytocin, arginine-vasopressin, and β-endorphin [48,51,56,85,86]. This anatomical and functional connectivity raises an important question: Are the lateralized neuropeptide systems coordinated across these brain areas to produce side-specific physiological responses? Furthermore, several studies have reported molecular alterations in neuropeptide systems that are left–right side-specific rather than simply contralateral–ipsilateral relative to external impact [29–32,40,42,47,48,51,52,118–120]. These findings suggest the presence of intrinsic side-specific molecular plasticity. A critical avenue for future research is to determine whether this plasticity serves an adaptive role in maintaining left–right physiological homeostasis—and whether its disruption by unilateral tissue damage can lead to maladaptive allostatic states that underlie pathological asymmetries.

In summary, while developmental programs establish a largely symmetrical body plan, lateralized neuropeptide systems appear to maintain a dynamic functional balance between the left and right sides of the body across the lifespan. This system’s plasticity may fine-tune hemisphere-specific brain functions and behaviors, while its dysregulation could influence lateralized pain processing and contribute to conditions such as hemiparesis or hemiplegia. These insights provide a compelling rationale for further exploring neuropeptides as lateralized regulators of left–right brain and body functions in both health and disease.

Highlights.

Neuropeptides are emerging as key regulators of lateralized neural circuits, acting through side-specific neuromodulatory, paracrine, and hormonal pathways.

Recent studies reveal that calcitonin gene related peptide, dynorphin, pituitary adenylate cyclase activating polypeptide, and substance P drive lateralized pain responses via asymmetric signaling in the parabrachial-amygdala circuit.

Similarly, oxytocin and vagal neuropeptides modulate social, sensory, and reward behaviors through lateralized cortical and brainstem circuits.

A newly proposed topographic neuroendocrine system may encode hemisphere-specific brain activity into hormonal signals targeting contralateral spinal and peripheral systems.

Lateralized neuropeptide signaling is evolutionarily conserved and may operate through distinct, coordinated mechanisms to maintain functional left–right asymmetry and interhemispheric balance.