Pancreatic nerves across species: Anatomical and functional disparities and their metabolic ramifications

Abstract

Pancreatic nerves exhibit significant anatomical and functional disparities across species, with profound implications for metabolism and clinical practices like pancreatic transplantation and disease treatment. In humans, pancreatic nerves feature a complex distribution of sympathetic and parasympathetic fibers, forming intricate plexuses that closely interact with surrounding organs, playing a crucial role in regulating both endocrine (e.g., insulin and glucagon release) and exocrine secretions to maintain metabolic balance. Rodents, such as rats and mice-common experimental models-have relatively simpler pancreatic nerve architectures; for instance, mice show more concentrated nerve fibers around islets, while rats display greater variability in nerve density across pancreatic regions, which influences their responses to metabolic stimuli and makes them valuable but not fully analogous to human models. Canines and felines, often used in translational studies, possess pancreatic nerve systems that share some similarities with humans, such as comparable nerve plexus organization, yet differ in fiber type proportions, affecting their susceptibility to metabolic abnormalities like pancreatitis-related glucose dysregulation. These species-specific differences in pancreatic nerves directly impact metabolic processes: In humans, impaired nerve function is linked to metabolic disorders like diabetes and obesity, whereas in animal models, variations in neural control of hormone release and pancreatic homeostasis lead to differences in disease progression and response to interventions. Understanding these discrepancies is vital: It not only helps interpret findings from animal studies when translating to human therapies but also guides pancreatic transplantation by highlighting the need for species-specific considerations in nerve reconstruction to restore metabolic function, and identifies potential therapeutic targets tailored to the unique neural-metabolic interactions of each species.

Key Words: Pancreatic nerves; Species differences; Autonomic innervation; Sympathetic nervous system; Parasympathetic nervous system; Islets of langerhans; Metabolic regulation; Diabetes; Pancreatic transplantation; Rodent models

Core Tip: This review highlights the significant anatomical and functional disparities in pancreatic innervation across humans, rodents, canines, and felines. These species-specific differences profoundly impact metabolic regulation, disease susceptibility (e.g., diabetes, pancreatitis), and responses to therapeutic interventions. Understanding these variations is crucial for accurately translating findings from animal models to human clinical applications, particularly in areas like pancreatic transplantation where nerve reconstruction strategies must be species-appropriate. Recognizing the unique neuro-metabolic interactions in each species can guide the development of more effective, tailored therapies for metabolic disorders and improve the predictive value of preclinical research.

INTRODUCTION

The pancreas is a dual-function organ with essential roles in digestion and metabolic homeostasis[1,2]. Its neural regulation involves a sophisticated interplay between the sympathetic and parasympathetic branches of the autonomic nervous system, which modulate hormone secretion, blood flow, and exocrine output. However, the anatomical and functional organization of pancreatic nerves is not uniform across species[3,4]. These differences are often overlooked in translational research, leading to misinterpretation of animal data and failed clinical trials.

In humans, pancreatic innervation is highly complex, involving dense plexuses that integrate inputs from multiple sources, including the vagus nerve and splanchnic sympathetic pathways[1]. This network allows for fine-tuned responses to nutrient intake, stress, and other physiological cues. In contrast, rodents-the most commonly used animal models-display a more simplified neural architecture, with clustered nerve trunks and significant regional heterogeneity[5]. Canines and felines, often used in surgical and pharmacological studies, exhibit intermediate complexity, with some structural similarities to humans but distinct functional profiles.

This review aims to provide a comprehensive comparison of pancreatic neuroanatomy and function across humans, rodents, canines, and felines. We will explore how these differences influence metabolic regulation, disease susceptibility, and therapeutic outcomes. By integrating findings from anatomical, electrophysiological, and molecular studies, we seek to bridge the gap between animal models and human physiology, with the goal of informing more effective and species-appropriate research and clinical practices (Figure 1).

Figure 1 Pancreatic nerve variations across species and metabolic implications. This illustration summarizes the significant interspecies disparities in pancreatic neuroanatomy and their metabolic implications. It comparatively depicts the intricate neural plexus in humans, clustered nerve trunks in rodents, and structurally similar yet distinct innervation in canines/felines. The graphic highlights how these anatomical variations influence autonomic regulation of endocrine secretion (insulin/glucagon), contributing to differential disease susceptibilities and therapeutic responses. Understanding these species-specific characteristics is crucial for improving translational research, guiding neural reconstruction in pancreatic transplantation, and developing targeted therapies for metabolic disorders like diabetes and pancreatitis.

ANATOMICAL COMPARISON OF THE PANCREATIC NERVOUS SYSTEM

Human

The human pancreas is characterized by rich and intricate neural innervation that forms a dense, interconnected network to regulate both endocrine and exocrine functions. This innervation primarily originates from two major autonomic plexuses: The celiac plexus and the superior mesenteric plexus, which collectively send branches to infiltrate the entire pancreatic parenchyma and establish functional associations with key pancreatic structures.

Neural plexus origin and intra-pancreatic network formation

Sympathetic innervation pathway: Preganglionic sympathetic neurons, which express acetylcholine (ACh), are localized in the intermediolateral columns of the spinal cord (T6-L2 segments). Their axons travel through the splanchnic nerves to synapse with postganglionic neurons in the celiac ganglia and superior mesenteric ganglia. From these ganglia, postganglionic sympathetic fibers-marked by tyrosine hydroxylase and vesicular monoamine transporter 2-project into the pancreas, where they further branch into smaller fibers to form a diffuse network[1,2].

Parasympathetic innervation pathway: Preganglionic parasympathetic neurons arise from the dorsal motor nucleus of the vagus (predominantly the left medial/Lateral columns). Their axons integrate into the vagus nerve, which divides into abdominal branches (hepatic, anterior/posterior gastric, and coeliac branches). Pancreas-targeting parasympathetic fibers primarily travel in the hepatic and anterior gastric branches, synapsing on intrapancreatic ganglia dispersed throughout the pancreas. Postganglionic parasympathetic fibers (cholinergic, marked by vesicular ACh transporter) then extend from these ganglia to form intra-pancreatic connections, further expanding the neural network[2,3].

Network integration: The intra-pancreatic neural network is further reinforced by sensory fibers (spinal and vagal afferents) and intrinsic neurons within intrapancreatic ganglia. These ganglia act as “relay centers” that integrate inputs from sympathetic, parasympathetic, and sensory pathways, ensuring coordinated neural regulation of pancreatic function[1,4]. 3D imaging studies have confirmed that this network exhibits regional heterogeneity: The pancreatic head and body have higher nerve fiber density than the tail, and the endocrine compartment (islets) is more densely innervated than the exocrine compartment[4].

Close spatial relationships between nerve fibers and pancreatic structures

The intra-pancreatic neural network maintains intimate spatial associations with pancreatic islets, blood vessels, acini, and ducts-critical for precise modulation of pancreatic activity.

Islets of langerhans: Sympathetic fibers predominantly contact alpha cells (glucagon-secreting) and delta cells (somatostatin-secreting), while parasympathetic fibers directly innervate beta cells (insulin-secreting). This targeted innervation allows sympathetic activation to inhibit insulin secretion and stimulate glucagon release, and parasympathetic activation to enhance insulin secretion[2,4]. Although only about 4% of human beta cells directly contact nerve fibers, electrical coupling via gap junctions enables neural signals to propagate across the entire islet[4].

Blood vessels: Sympathetic fibers extensively innervate pancreatic capillaries and arterioles. Activation of these fibers induces vasoconstriction, reducing islet blood flow to regulate hormone delivery[1,2]. Parasympathetic fibers, by contrast, weakly innervate vasculature but may indirectly promote vasodilation via nitric oxide release, supporting nutrient and hormone exchange[1].

Acini and ducts: Parasympathetic fibers innervate exocrine acinar cells, stimulating the secretion of digestive enzymes (e.g., amylase) via muscarinic receptors. They also contact ductal epithelial cells, modulating bicarbonate secretion to neutralize digestive enzymes during nutrient absorption[2,3]. Sympathetic fibers have weaker associations with acini and ducts but may inhibit exocrine secretion during stress[1].

Rodents (rats and mice)

Rodent pancreatic innervation is structurally simpler than that of humans but exhibits species-specific and regional characteristics.

Mouse pancreatic nerves are primarily distributed as clustered nerve trunks (rather than single fibers) that localize exclusively around intrapancreatic blood vessels and lymphoid structures. Significant regional differences in nerve density exist: The pancreatic head (0.15% ± 0.08% of tissue area) and anterior surface of the corpus (0.17% ± 0.27%) have significantly higher nerve density than the tail (0.02% ± 0.02%, P = 0.006)[5]. A unique “peri-islet nerve sheath” has been observed in mouse models, forming a protective and regulatory layer around islets[3]. Sympathetic (TH⁺) and parasympathetic (VAChT⁺) fibers innervate islets, with sympathetic fibers inhibiting insulin secretion and parasympathetic fibers stimulating it[4]. Nociceptive fibers [calcitonin gene-related peptide-positive (CGRP⁺)] are enriched in the pancreatic tail but constitute only a small fraction of total innervation[5].

Rat pancreatic innervation shares similarities with mice but lacks the “peri-islet nerve sheath”[3]. Regional heterogeneity in nerve density is evident (head > body > tail), though quantitative data on cholinergic/adrenergic fiber ratios remain limited[4]. Sympathetic fibers from the celiac ganglia innervate islet vasculature and alpha cells, while parasympathetic fibers from the vagus nerve directly contact beta cells[2] (Table 1).

Table 1 Summary of rodent pancreatic innervation features.

Feature	Mouse	Rat
Nerve distribution	Clustered around vessels/lymphoid structures	Diffuse, regional heterogeneity
Key structural feature	Peri-islet nerve sheath	Absent
Nerve density (head/tail)	0.15%/0.02% tissue area	Head > tail (no exact values)
Sympathetic targets	Islet alpha/delta cells, vasculature	Islet alpha cells, vasculature

Canines and felines

Data on pancreatic innervation in canines and felines remain scarce but indicate partial structural similarity to humans.

Canines: Canine pancreatic nerves originate from the celiac and superior mesenteric plexuses, analogous to human innervation[3]. Electrical stimulation studies in beagles show that vagal nerve activation increases insulin secretion, while splanchnic (sympathetic) nerve stimulation enhances glucagon release-suggesting species-specific differences in cholinergic/adrenergic fiber function[3]. These fiber ratio differences may influence pancreatic basal tone and stress-induced metabolic responses.

Felines: Feline pancreatic innervation involves vagal afferents from the dorsal motor nucleus of the vagus and sympathetic postganglionic neurons localized in the thoracic dorsal root ganglia (DRG) (T3-L5)[2-6]. Light microscopy confirms parasympathetic (VAChT⁺) fibers contact islet beta cells, but quantitative data on fiber ratios or regional density are lacking[2].

FUNCTIONAL COMPARISON OF THE PANCREATIC NERVOUS SYSTEM

Regulation of human metabolism

In the human body, during the preprandial period or under hypoglycemic conditions, parasympathetic neurons stimulate α-cells to secrete glucagon[3]. After food intake, parasympathetic neurons are activated, which in turn stimulate acinar cells to secrete digestive enzymes[4] and β-cells to secrete insulin[5]. It has been observed so far that the parasympathetic nervous system promotes the basal secretion and postprandial secretion of insulin in different ways, but the subsequent processes are basically the same. These processes lead to the removal of glucose from the bloodstream and its storage in the liver, thereby restoring glycogen and normal blood glucose levels[6].

Under stress conditions, sympathetic neurons in the mature pancreas cause pancreatic vasoconstriction[10]. They also regulate glucose homeostasis by inhibiting insulin secretion from β-cells[11,12], upregulate glucagon release from α-cells[13], and convert glycogen stores into blood glucose to maintain blood glucose stability and cope with stress.

Functional manifestations in animal models

In vitro experiments on the effects of substance P (SP) on insulin in rats have yielded inconsistent results, but it has been shown that SP inhibits glucagon secretion[14].

In rat experiments, the response of the pancreas to load tests in nerve reconstruction and implantation (NRI) or aTx models is characterized by a slightly delayed onset, excessive intensity development, and prolonged duration. This response is most prominent in a series of experiments where fasted and non-fasted animals were stimulated with epinephrine[15].

TRPM3 channels are expressed on the nerve endings of primary pancreatic afferents containing CGRP. Activation of TRPM3 in the pancreas by injection of its specific agonist CIM0216 (10 μM) induces pain, release of CGRP and SP, and neurogenic inflammation, as evidenced by edema, plasma extravasation, and accumulation of pancreatic inflammatory cells. Increased functional expression of TRPM3 has been detected in DRG neurons of the pancreas in rats with acute pancreatitis (AP). Blocking TRPM3 activity with its antagonist (primidone, 5 mg/kg, i.p.) alleviates AP-related pain behavior and pancreatic inflammation. Pre-incubation of pancreatic DRG neurons with nerve growth factor enhances the TRPM3 agonist-induced increase in intracellular Ca²⁺ (CIM0216, 1 μM)[16].

The regulation of insulin secretion by the sympathetic nervous system differs among humans, mice, and dogs. In mice and humans, sympathetic nerves primarily stimulate α-cells to secrete glucagon and reduce insulin secretion from β-cells by regulating the islet vascular system[17]. However, it has been reported that insulin levels increase slowly when the visceral sympathetic nerves of anesthetized dogs are stimulated[14].

The nervous system is difficult to analyze in mouse models, which complicates efforts to simulate human pancreatic diseases[14]. AP is an inflammatory disease that begins in the exocrine pancreas and can progress to peripancreatic and distant tissues. Numerous studies conducted in different tissues have shown that PAKs are important general mediators of inflammation[18,19], so it may be suspected that their activation may play an important role in AP, which has a prominent inflammatory component. Results from multiple recent studies have demonstrated that activation of p21-activated kinases may play an important role in the pathophysiology of AP. In a recent study[20], using a widely used experimental model of AP induced by supramaximal concentrations of CCK in rat pancreatic acinar cells, many results supported the conclusion that activation of group I PAKs (PAK2) plays an important role in the early mediation of AP[21].

In humans, the release of head insulin depends on neural pathways, and its response is important for postprandial glucose tolerance[22]. However, the nerve density in the head and body of the mouse pancreas is higher than in other regions, which may contribute to the differences in postprandial glucose tolerance responses between humans and mice.

Peri-arteriolar aggregation and network formation of islets and ganglia have been found in mice. The neuro-insular network is characterized by TUJ¹⁺ nerve plexuses coupling islets and ganglia (both periloobular) to establish islet-ganglion associations[23]. However, the small arteries of human islets and ganglia do not form obvious networks; instead, they are relatively scattered. Therefore, compared with humans, the interaction between islets and nerves in mice is stronger, which also affects the transmission speed of relevant neurotransmitters, insulin, and other substances.

Sympathetic nerves and intrapancreatic neural networks in pancreatic neural input indicate potential pathways by which external (brain-to-pancreas) and internal (pancreatic ganglia-to-islet) neural inputs regulate hormone secretion[24-27]. Although islets are scattered throughout the pancreas, they are coupled with intrapancreatic ganglia to coordinate their insulin secretion activities. The intrapancreatic neural network may have hotspots involved in local neural activity rather than uniformly innervating the pancreas for signal transmission[23].

METABOLIC IMPLICATIONS OF INTERSPECIES VARIATIONS AND THEIR DISEASE ASSOCIATIONS

Mechanisms: The precise operation of the nervous system in metabolic regulation

The maintenance of energy homeostasis is a complex neuroendocrine process meticulously coordinated by the brain. The hypothalamus acts as the metabolic “command center”, receiving hormonal signals (e.g., leptin, insulin) from adipose tissue, the pancreas, and the gastrointestinal tract. It then issues instructions to peripheral organs (e.g., pancreas, liver) via the autonomic nervous system to precisely regulate energy intake and expenditure[28]. Studies have shown that the paraventricular nucleus and ventromedial nucleus of the hypothalamus are involved in inhibiting food intake and weight gain in the energy regulation center; damage to both leads to increased food intake, hyperglycemia, and obesity in animals.

However, when this sophisticated regulatory network is disrupted by genetic susceptibility and environmental factors (e.g., a high-fat diet), it can trigger a series of metabolic disorders, potentially resulting in metabolic diseases such as obesity and diabetes. Accumulating evidence indicates that neural dysfunction is not merely a passive consequence but an active driver of metabolic diseases such as obesity and type 2 diabetes mellitus (T2DM)[29]. This review will systematically elaborate on the central role of the nervous system in metabolic regulation, analyze its dysregulation in disease, and specifically discuss the insights revealed by animal models, alongside the limitations arising from species differences.

Hypothalamus: The central hub in metabolic regulation

The hypothalamus, particularly the arcuate nucleus, is the key central hub for sensing systemic energy status and the neural regulation center for feeding behavior. It receives and integrates circulating hormonal signals such as leptin (primarily from adipose tissue) and insulin (from the pancreas) by expressing specific long-form receptors[29], while also regulating vital activities of the organism, including the secretion of α-cells and β-cells in the islets.

Under healthy conditions, elevated leptin and insulin levels convey a signal of “energy sufficiency” to the brain. Specifically, after food intake, blood glucose levels rise, and insulin secreted by the pancreas as well as leptin secreted by white adipocytes increase accordingly; leptin and insulin, as important signaling molecules, transmit the “energy sufficiency” signal to the brain. This activates anorexigenic pro-opiomelanocortin (POMC) neurons and inhibits orexigenic agouti-related peptide (AgRP) neurons in the hypothalamus, ultimately reducing food intake and increasing energy expenditure to avoid excessive energy intake[28,30].

Autonomic nervous system

Central commands are relayed to the periphery via the autonomic nervous system. The parasympathetic nerve, specifically the vagus nerve, is active in the postprandial state. It releases ACh to directly stimulate insulin secretion from pancreatic β-cells, while suppressing hepatic glucose output, thereby promoting nutrient storage and maintaining stable blood glucose levels[31]. Conversely, the sympathetic nervous system is activated during stress or fasting. It releases norepinephrine to inhibit insulin secretion and promote the release of glucagon, hepatic gluconeogenesis, and lipolysis, mobilizing energy reserves[32]. This system maintains dynamic balance in healthy individuals, forming the cornerstone of metabolic homeostasis.

Neurological dysfunction in metabolic diseases

The progression of metabolic diseases forms a vicious cycle with nervous system dysfunction. When the body is chronically exposed to unhealthy lifestyles or genetic factors, metabolic diseases may develop subtly; in this process, neural dysfunction plays a crucial role and forms a vicious cycle with the occurrence and progression of metabolic diseases. A high-fat diet can rapidly induce hypothalamic microinflammation, inhibition of neuronal regeneration, and neuronal damage characterized by elevated levels of pro-inflammatory cytokines (e.g., tumor necrosis factor-alpha, interleukin-6) and activation of inflammatory pathways such as IKKβ/NF-κB[29]. This inflammatory environment leads to central leptin and insulin resistance, rendering the hypothalamus incapable of correctly interpreting peripheral energy signals and erroneously issuing instructions to “promote feeding and reduce energy consumption,” constituting an early event in obesity and T2DM[29,33].

In obese individuals, overall sympathetic nerve activity, which should suppress appetite, is often reduced-a sympathetic tone abnormality-while sympathetic nerves innervating the kidneys and cardiovascular system become overactive, closely associated with the development of hypertension[34]. Simultaneously, parasympathetic function declines. Reduced vagus nerve function governing the pancreas leads to the loss of the “first-phase” insulin secretion after meals, a critical early characteristic of diabetes[35]. Furthermore, the disease exacerbates neural damage. Persistent hyperglycemia and dyslipidemia exhibit neurotoxicity, directly injuring the small nerve fibers that innervate the pancreas and peripheral tissues, thereby worsening autonomic neuropathy and forming a vicious cycle[31,36,37]. The degeneration of the intrapancreatic neural plexus causes pancreatic α-cells to lose normal inhibitory neural signals, resulting in inappropriate glucagon secretion that further aggravates hyperglycemia[38]. Additionally, a high-fat diet inactivates Rb protein in anorexigenic POMC neurons, causing POMC neurons to re-enter the cell cycle from a quiescent state and subsequently undergo rapid apoptosis.

Insights from specific neuromodulators: Vasoactive intestinal peptide, CGRP, and SP

In the pancreas of cystic fibrosis (CF) mice, reduced vasoactive intestinal peptide (VIP) levels, decreased innervation, decreased insulin secretion, upregulated glucagon production, and elevated random blood glucose levels have been observed. We propose that due to the reduced innervation of the CF pancreas, which begins in the early stages of the disease, low VIP levels lead to changes in insulin and glucagon secretion, thereby contributing to the development of CF-related diabetes (CFRD)[39]. Therefore, the level of VIP has a greater impact on CF in mice.

In mice, researchers have not observed an association between VIP-stimulated insulin release and increased cAMP levels[40], suggesting that signaling pathways other than the cAMP pathway may also be involved in this process[14].

In addition to parasympathetic and sympathetic nerves, islets are innervated by a variety of sensory nerves that secrete CGRP and SP. The CGRP receptor complex is expressed in mouse β-cells, while the SP receptor NK-1R is expressed in mouse α-cells. CGRP stimulates the release of somatostatin in the gastrointestinal tract and inhibits insulin release, and this pathway may be the mechanism by which CGRP inhibits insulin release. CGRP increases blood flow to the pancreas, and when elevated insulin concentrations are detected, the increased blood flow leads to a reduction in insulin concentration[41]. Multiple experiments have shown that CGRP inhibits glucose-induced insulin release, but the exact mechanism involved remains to be further studied[14].

Cellular metabolic mechanisms underlying neural regulation

When blood glucose increases, glucose is transported into β-cells and phosphorylated by glucokinase, a low-affinity hexokinase. This step is the rate-limiting step in glucose metabolism[42]. Because the proportion of β-cells in mouse islets is higher than that in human islets, more glucose can be transported into β-cells in mice, which also accelerates the rate of glucose metabolism to a certain extent. An increase in the ATP/ADP ratio inhibits ATP-sensitive K+ (KATP) channels, leading to membrane depolarization, activation of voltage-gated calcium channels, and the generation of bursts of action potentials. Oscillatory increases in intracellular free calcium activity (Ca²⁺)i trigger pulses of insulin granule exocytosis[43-45]. In the absence of elevated ATP/ADP, KATP channels remain open, hyperpolarizing the cell, preventing the activation of voltage-gated calcium channels, and inhibiting (Ca²⁺)i. β-cells in the islets are electrically coupled through Connexin 36 gap junction channels[46-48]. Due to this electrical coupling, β-cells exhibit synchronized (Ca²⁺)i oscillations under elevated glucose conditions and are uniformly silent under basal glucose conditions[46,48-51]. β-cells are functionally heterogeneous, showing variations in electrical activity, insulin secretion[52-55], and gene expression profiles[56-58]. However, due to electrical coupling, β-cells within an islet show nearly identical (Ca²⁺)i responses and oscillatory kinetics[48]. Nevertheless, the overall (Ca²⁺)i response of the entire islet may be influenced by the β-cell population. In normal islets, almost all β-cells can respond to elevated glucose. However, a small number of non-excitable cells that remain hyperpolarized when blood glucose increases can reverse this response and silence the entire islet[59-61]. Conversely, only a small number of highly excitable cells can support uniformly elevated (Ca²⁺)i kinetics[62]. In the presence of high levels of metabolic heterogeneity, gap junction electrical coupling can robustly synchronize (Ca²⁺)i responses. When the number of metabolically defective β-cells is small, electrical coupling promotes (Ca²⁺)i elevation in the islet; however, when the number of metabolically defective β-cells is large, electrical coupling inhibits (Ca²⁺)i elevation in the islet[63].

SPECIES DIFFERENCES

Important contributions of animal models

Animal models, particularly rodents, have played an indispensable role in revealing the causality within these neuro-metabolic pathways. For instance, specific manipulation of hypothalamic AgRP or POMC neuron activity using optogenetics or chemogenetics directly demonstrates their role in regulating feeding and energy expenditure. When AgRP neurons are activated, mice show a significant increase in appetite; in contrast, inhibiting AgRP neurons leads to a marked decrease in mice’s appetite-this experimental result directly confirms the regulatory role of hypothalamic AgRP neurons in feeding behavior. Activating POMC neurons can significantly increase energy expenditure in mice and enhance their activity, while inhibiting POMC neurons reduces energy expenditure in mice and causes gradual weight gain-this study clearly reveals the key role of POMC neurons in regulating energy expenditure[30].

Unignorable species differences

However, inherent species differences present a core challenge for translational medicine, see Table 2.

Table 2 Differences in the innervation of the pancreas among humans, mice, cats and dogs.

Species category	Origin and pattern of pancreatic innervation	Islet architecture	Innervation of individual cells within the pancreatic islets	Disease-associated regulatory pathways or mechanisms of action
Humans	Derived from celiac and superior mesenteric plexuses; sympathetic (T6-L2) and parasympathetic (vagus) postganglionic fibers form networks, with sensory fibers and intrinsic neurons; denser innervation in pancreatic head/body and endocrine region, forming intrapancreatic neural network	Mixed distribution of α/β/δ cells	Sympathetic fibers contact α/δ cells; parasympathetic directly innervates a small number of β cells (signal transmission via electrocoupling). Sympathetic causes pancreatic vasoconstriction and inhibits exocrine secretion under stress; parasympathetic induces indirect vasodilation and promotes exocrine secretion	Parasympathetic: Preprandial glucagon secretion promotion, postprandial digestive enzyme and insulin secretion promotion; Sympathetic (stress): Insulin secretion inhibition and glucagon release promotion. Neural dysfunction + hypothalamic microinflammation → obesity and diabetes; sustained hyperglycemia/dyslipidemia aggravates nerve injury, forming a vicious cycle
Mice	Pancreatic nerves aggregate along blood vessels/Lymphatics, denser in pancreatic head/body; nerve sheaths around islets. Sympathetic/parasympathetic innervate islets; CGRP+ nociceptive fibers enriched in pancreatic tail	Pancreatic islets: Α/δ cells at periphery, β cells in core; nerves distribute as aggregated trunks with perislet nerve sheaths	Sympathetic innervates islet α/δ cells and blood vessels; parasympathetic innervates islets	Parasympathetic promotes postprandial insulin secretion; weak α-cell inhibition on β-cells. Reduced VIPergic fibers cause glucose metabolism disorders in CF mice; CGRP inhibits glucose-induced insulin release
Rats	Nerves diffusely distributed with regional heterogeneity (pancreatic head > body > tail); no perislet nerve sheaths. Sympathetic innervates islet blood vessels and α-cells; parasympathetic contacts β-cells	Islet β-cells in core, surrounded by α/δ/PP cells; nerves diffusely distributed	Sympathetic innervates islet blood vessels and α-cells; parasympathetic contacts β-cells	SP inhibits glucagon secretion in rats; TRPM3/PAK2. Excessive vagal nerve activity enhances glucose-stimulated insulin secretion by directly acting on pancreatic β-cells through the release of acetylcholine from its nerve terminals, leading to pathological hyperinsulinemia. The vagus nerve regulates the local immune environment of the pancreas via the spleen-dependent cholinergic anti-inflammatory pathway. Spleen deficiency impairs this pathway, thereby reducing the responsiveness of pancreatic islets to cholinergic stimulation and exacerbating pancreatic fat infiltration and inflammatory states[7]. Activation induces acute pancreatitis in rats, NRI/aTx model causes abnormal pancreatic load test responses in rats
Cat	Vagal afferent fibers originate from its dorsal motor nucleus; sympathetic postganglionic neurons in thoracolumbar DRG (T3-L5); VAChT+ parasympathetic fibers contact islet β-cells	Regional specificity: Α-cell enrichment in splenic region, PP-cell enrichment in duodenal region[8]	VAChT+ parasympathetic fibers contact islet β-cells; sympathetic postganglionic and sensory neurons innervate the more cranial duodenal segments densely, and the splenic segments sparsely and centrally[8]	Pancreatitis is often complicated with inflammatory bowel disease, cholangitis (triad); obesity induces insulin resistance, increasing the β-cell load; chronic hyperglycemia generates toxicity, impairing β-cell function and reducing their quantity, thereby forming a vicious cycle of “resistance-secretion imbalance”[9]. NRI + SDR regimen restores pancreatic nerve conduction function in cat models
Dog	Derived from celiac and superior mesenteric plexuses; vagal activation increases insulin secretion, visceral sympathetic stimulation enhances glucagon release	Mixed distribution of α/β/δ cells (similar to humans)	Vagal activation promotes insulin secretion in β-cells; sympathetic stimulation enhances glucagon release in α-cells	Sympathetic-parasympathetic fiber ratio affects pancreatic tone & stress metabolism; SP regulates insulin/glucagon secretion depending on concentration; pancreatitis is mostly associated with high fat, diet and drugs

CGRP: Calcitonin gene-related peptide; VAChT: Vesicular acetylcholine transporter; SP: Substance P; VIP: Vasoactive intestinal peptide; NK-1R: Neurokinin-1 receptor; TRPM3: Transient receptor potential melastatin type 3 channel; PAK2: P21-activated kinase 2; CF: Cystic fibrosis; CFRD: Cystic fibrosis-related diabetes; AP: Acute pancreatitis; NRI: Nerve reconstruction and implantation; aTx: Axotomy; SDR: Selective dorsal root rhizotomy; PP cells: Pancreatic polypeptide cells; DRG: Dorsal root ganglion.

Compared with human islets, mouse islets contain a higher proportion of β-cells and a lower proportion of α-cells[64]. Based on the way the pancreatic nervous system regulates metabolism, it can be inferred that compared with humans, the parasympathetic nervous system of mice exerts a weaker promoting effect on the basal secretion of insulin, but a stronger promoting effect on the postprandial secretion of insulin. They also have different cellular structures. In humans, α-cells, β-cells, and δ-cells are mixed throughout the islets, while mouse islets are more organized, with α-cells and δ-cells in the periphery and β-cells in the core[64].

In rodents, β-cells are located in the center of the islets, surrounded by α-cells (which secrete glucagon), δ-cells (which secrete somatostatin), and PP cells (which secrete pancreatic polypeptide). In humans, islets are composed of three main cell types (α-cells, β-cells, and δ-cells), allowing for more local interactions compared to rodents because these three types of cells are mixed throughout the islets. For example, adjacent α-cells can directly transmit inhibitory signals to β-cells (e.g., through interactions between specific membrane proteins) to regulate insulin release from β-cells and prevent excessive insulin secretion when blood glucose is low[14]. In contrast, rodents exhibit weaker mutual regulatory effects in this regard, and their ability to inhibit excessive insulin secretion during hypoglycemia is also less effective.

The nerve density in the head and body of the mouse pancreas is higher than in other regions, and the pancreatic head is significantly enriched with nociceptive nerve fibers that transmit pain. Among the subtypes of pain-transmitting nerve fibers, nerves containing pain-related neuropeptides such as SP, CGRP, VIP, or nitric oxide synthase (NOS) show no significant difference in distribution across different regions of the mouse pancreas, accounting for approximately 10% of the total innervation. VIPergic nerve fibers are distributed relatively evenly across the three main regions of the pancreas. Additionally, the loss of VIPergic fibers in the pancreas appears to be accompanied by an increase in nNOS-containing nitrergic nerve fibers in the pancreatic head and body[65].

As a neuromodulator, VIP exerts an insulinotropic effect on pancreatic β-cells. Combined with the above context, compared with humans, mouse islets contain a higher proportion of β-cells; therefore, the insulinotropic effect of VIP on mice in this regard may be stronger.

The most significant differences in the pancreas between humans, dogs, and cats are reflected in the pancreatic duct system.

Humans generally have a dual duct system. The main pancreatic duct runs the entire length of the pancreas, converges with the common bile duct, and jointly opens into the ampulla of Vater in the duodenum. The accessory pancreatic duct opens directly into another minor papilla of the duodenum. This makes it easy for stones at the end of the common bile duct (gallstones) to block the ampulla, leading to obstruction of the common excretion of bile and pancreatic juice, thereby causing biliary pancreatitis.

Dogs almost exclusively have a single pancreatic duct system. The main duct flows from the left lobe to the right lobe and opens independently into the duodenum. The common bile duct opens independently into the duodenum proximal to the opening of the main duct (on the side closer to the stomach). Since the bile duct and pancreatic duct are separate, dogs rarely develop pancreatitis caused by direct obstruction of the pancreatic duct by gallstones. Their pancreatitis is more related to hyperlipidemia, improper diet, or drugs. In dogs, SP stimulates the secretion of insulin and glucagon in a concentration-dependent manner[14]. As previously mentioned, SP inhibits glucagon secretion in rats.

Cats, similar to dogs, have a single pancreatic duct system. The main duct opens into the duodenum, and the common bile duct opens independently near it. The susceptibility of cats and dogs to AP is not determined solely by differences in pancreatic duct anatomy but is the result of multifactorial interactions. Although ductal anatomy may influence the disease mechanism-for example, humans are prone to biliary pancreatitis due to a common channel for the bile and pancreatic ducts, whereas the separated ducts in cats and dogs make this etiology less common-interspecies differences require a comprehensive analysis from the following aspects.

Differences in pathophysiological mechanisms: Canine pancreatitis is often associated with hyperlipidemia, dietary indiscretion (e.g., high-fat diets), or certain drugs (e.g., L-asparaginase). Canine pancreatic acinar cells exhibit higher sensitivity to CCK, making them prone to excessive activation of zymogens under stress, leading to autodigestion and inflammatory responses[15,66].

Feline pancreatitis frequently co-occurs with inflammatory bowel disease and cholangitis (the “triaditis” complex), suggesting a dominant role for immune-inflammatory pathways. Differences in calcium signaling regulation within feline pancreatic acinar cells, compared to humans, predispose them to aberrant zymogen activation[67].

Human pancreatitis is primarily of biliary or alcoholic origin, which is distinct from the metabolic or immune-mediated etiologies prominent in cats and dogs[68].

Data supporting species-specific risk factors: (1) Dogs: Epidemiological studies indicate that approximately 2% of canine AP cases are triggered by high-fat dietary intake, with certain breeds like Miniature Schnauzers and Poodles showing genetic predispositions[69]; (2) Cats: Approximately 1.5% of cats are diagnosed with pancreatitis, although the true incidence is likely higher due to often subtle clinical signs. Risks are significantly elevated in senior cats (> 10 years old) and obese individuals[70]; and (3) Comparative data: The estimated annual incidence in humans ranges from 13 to 45 per 100000. In contrast, incidence rates in high-risk canine and feline populations can be orders of magnitude higher[71]. Direct comparisons, however, require caution due to differences in diagnostic criteria and surveillance mechanisms.

Potential influence of neuro-immune interactions: As noted in the document, differences in pancreatic innervation between dogs/cats and humans (e.g., the ratio of cholinergic to adrenergic fibers) may modulate local inflammatory responses and influence susceptibility. For instance, overexpression of TRPM3 channels in sensory nerve endings within the canine pancreas may exacerbate neurogenic inflammation, a pathway whose role in human pancreatitis remains less defined[16].

Research limitations and inferential cautions: Current conclusions are largely based on experimental models (e.g., induced pancreatitis in dogs and cats) or retrospective clinical studies. There is a lack of direct, prospective comparative data between humans, dogs, and cats. Therefore, the original assertion of “greater susceptibility” should be revised to state that “there are species-specific differences in etiology and susceptibility”, emphasizing the limitations of cross-species extrapolation. However, the etiology of pancreatitis in cats is complex, and it is often complicated by inflammatory bowel disease and cholangitis, forming the so-called “triaditis”. Although the pancreatic duct and bile duct are separate, inflammation can spread between them. Electrophysiological studies of NRI and NRI + selective dorsal root (SDR) pancreas in cat models showed that the proposed SDR protocol can restore nerve conduction function. By comparing the results of load tests (glucose, epinephrine, and insulin) and amylase/lipase detection in rat and canine models, the effectiveness of pancreatic tail nerve retransplantation or SDR technology implemented after transplantation was confirmed[15]. However, considering the differences in the pancreas between humans, cats, and dogs, whether this technology is effective in humans still requires verification from multiple aspects.

Anatomical and functional comparisons reveal fundamental disparities: In human islets, β-cells and α-cells are intermingled, creating a more complex neuroregulatory network. In contrast, rodent islets have β-cells located in the core and α-cells distributed in the periphery with a relatively simpler innervation pattern. This may lead to therapies targeting neural regulation that are effective in animal models but exhibit altered or diminished efficacy in the more complex human microenvironment. While vagal control of postprandial insulin secretion is highly significant in rodents, its relative contribution in humans is smaller, relying more on hormonal pathways [e.g., glucose-dependent insulinotropic polypeptide (GIP), glucagon-like peptide-1 (GLP-1)[32]. Consequently, drugs that show strong neurally-mediated effects in animal models (e.g., some GLP-1 receptor agonists) may act primarily via hormonal pathways in humans.

Human metabolic diseases progress slowly and exhibit high genetic diversity, whereas animal models (e.g., those induced by gene knockout or high-fat diet) progress rapidly and have a uniform genetic background. This makes it difficult for animal models to fully replicate the complex pathological network and individualized treatment responses seen in human disease.

Translational implications

In summary, the functional integrity of the nervous system is the cornerstone of metabolic homeostasis. From central resistance in the hypothalamus to peripheral dysregulation of the autonomic nerves, neural impairment permeates the entire course of obesity and T2DM, forming a self-reinforcing vicious cycle. Animal models provide an unparalleled window for dissecting the causal relationships within this complex network, but their inherent species differences necessitate caution when translating basic research findings to the clinic.

Future research should focus on methodological innovations, integrating more advanced human organoids, in vivo neural functional imaging technologies, and animal models for cross-validation. As an emerging research tool, human organoids can simulate the structure and function of human organs in vitro, providing a model closer to the actual human situation for the study of metabolic diseases. Efforts must continue to promote clinical translation by incorporating neural function assessment into the routine diagnosis and management of metabolic diseases. A paradigm shift is also needed-from the traditional “organ-centric” view towards a “neuro-endocrine network-centric” systems biology perspective. Only through such approaches can we more deeply unravel the complexity of metabolic diseases and develop more effective prevention and treatment strategies.

IMPLICATIONS FOR CLINICAL AND TRANSLATIONAL MEDICINE

Given the significant interspecies differences in pancreatic neuroanatomy and function (Parazacco spilurus subsp. spilurus), particular caution is warranted when translating animal experimental data (especially rodent and canine/feline models) to Homo sapiens clinical applications. Researchers must incorporate cross-species comparative analyses to appropriately adjust experimental designs and extrapolation strategies, thereby avoiding misinterpretations of animal models’ limitations in metabolic regulation or disease mechanism studies.

These disparities have profound implications for translational medicine and clinical practice, primarily manifested in the following aspects.

Re-evaluation of the translational validity of animal models and strategic adjustments

Animal models, particularly rodents, are indispensable tools for studying pancreatic physiology and the mechanisms of metabolic diseases. However, as previously indicated, human pancreatic innervation is highly complex, forming dense plexiform networks that interact closely with surrounding organs. In contrast, nerve fibers in mice are distributed more as clustered nerve trunks and exhibit significant regional heterogeneity; for instance, nerve density in the pancreatic head and body is significantly higher than in the tail. This structural simplification, coupled with fundamental differences in islet cell architecturehuman islets feature a intermingled distribution of α and β cells, whereas mouse islets have β-cells located in the core, implies essential differences in the neural regulation of hormone secretion across species.

Consequently, the interpretation of animal data must incorporate cross-species comparative analysis to adjust experimental designs and extrapolation strategies. For example, therapies that demonstrate significant efficacy in rodent models by modulating vagal nerve activity to improve insulin secretion may have limited effects in humans, as human insulin secretion relies more heavily on humoral pathways (e.g., GIP, GLP-1) than neural innervation. Similarly, drug targets validated in animal models based on specific nerve fiber ratios (e.g., cholinergic/adrenergic) must be applied to humans with consideration for the differences in fiber type proportions. Neglecting these disparities and directly applying conclusions from animal models is a significant reason for the failure of many preclinical studies in clinical trials.

Guiding neural functional reconstruction in pancreatic transplantation

Pancreatic transplantation is an effective treatment for type 1 diabetes, and the restoration of neural function post-transplantation is crucial for long-term metabolic stability. Previous sections emphasize that successful neural regeneration must account for species-specific innervation patterns. Human pancreatic nerves primarily originate from the celiac plexus and superior mesenteric plexus, forming a sophisticated regulatory network. Although similar plexus organizations are observed in large animal models like canines, their cholinergic-to-adrenergic fiber ratios differ from those in humans, which could lead to suboptimal neural reinnervation post-transplantation, affecting the long-term stability of graft function.

Clinical practice should emphasize the concept of “species-specific neural reinnervation”. During transplant surgery, efforts should be made to preserve or reconstruct neural pathways homologous to humans. For instance, meticulous handling of the anatomy around the celiac plexus, preserving the neural pedicle of the donor pancreas as much as possible, or exploring the possibility of neural anastomosis during transplantation. This aids in restoring physiological rhythmic secretion of insulin and glucagon, particularly the first-phase insulin secretion after meals, which is decisive for achieving precise blood glucose control and enhancing the long-term survival of the graft. Research in animal models (e.g., dogs, cats) provides technical validation for this; for example, specific nerve retransplantation or selective DRG techniques can partially restore nerve conduction function and the pancreatic response to load tests.

Providing new perspectives for developing personalized therapies

The unique neuro-metabolic interactions in different species offer novel perspectives for developing personalized therapies for human metabolic diseases. For example, the attenuated response to certain neurotransmitters (such as VIP) observed in rodent models may reveal the potential role of receptors like NPY receptors in regulating insulin secretion. In mouse models of CF, the reduction of VIPergic nerve fibers in the pancreas is closely associated with decreased insulin secretion and the development of CFRD, suggesting that interventions targeting the VIP pathway could be a potential therapeutic strategy for CFRD.

On the other hand, the association between neurogenic inflammation and susceptibility to pancreatitis in canine/feline models hints that targeting specific neural signaling pathways (e.g., the cholinergic anti-inflammatory pathway) might offer new avenues for treating human pancreatitis and its metabolic complications. For example, in rat models of AP, increased functional expression of the TRPM3 channel on pancreatic sensory nerve endings was found; its activation exacerbates pain and inflammation, while blocking TRPM3 with an antagonist alleviates symptoms. This provides a direct target for developing new analgesic and anti-inflammatory drugs.

Through cross-species comparisons, conserved or species-specific neuroregulatory mechanisms can be identified. For instance, despite structural differences, the fundamental principle that sympathetic nerves inhibit insulin secretion and parasympathetic nerves stimulate it is conserved between humans and many animals. Interventions targeting these conserved mechanisms may have broader applicability. For human-specific complex neural networks and cellular arrangements, more precise intervention strategies that mimic the human microenvironment need to be developed, such as using human islet organoids for high-throughput drug screening. This research paradigm, based on a deep understanding of species differences, will contribute to designing more precise interventions tailored to human disease phenotypes.

Future directions for clinical research

Future clinical research should place greater emphasis on integrating neurological assessments. Autonomic nerve function tests (e.g., heart rate variability analysis) should be incorporated into the routine diagnosis and management of patients with diabetes and obesity to early detect dysregulation of neural modulation. Furthermore, in clinical trials of novel therapies for metabolic diseases (e.g., GLP-1 receptor agonists, immunosuppressive regimens for pancreatic transplantation), subgroup analyses should be established to explore the association between treatment response and the patient’s autonomic nerve status, thereby achieving truly individualized treatment. In summary, a deep understanding of the species differences in pancreatic nerves is a bridge connecting basic research and clinical application. It requires a shift from a simplistic linear extrapolation mindset of “animal model to human” towards a systematic and precision-oriented paradigm based on comparative biology and emphasizing species specificity, ultimately advancing the diagnosis and treatment of metabolic diseases.

CONCLUSION

In summary, the anatomical and functional divergences in pancreatic innervation across species underscore the complexity of metabolic regulation and disease manifestation. In humans, impaired neural function is closely associated with metabolic disorders such as diabetes and obesity, whereas in animal models, variations in neural control over hormone release and pancreatic homeostasis lead to divergent disease progression and therapeutic responses. A nuanced understanding of these differences is indispensable for bridging the gap between animal studies and human applications, as emphasized in recent PubMed-indexed reviews. Moreover, this knowledge directly informs clinical practices, particularly in pancreatic transplantation, where successful neuroregeneration must account for species-specific innervation patterns.

Looking ahead, future research should prioritize multi-omics approaches and advanced imaging techniques to elucidate conserved and species-specific neural pathways. As noted in CNKI-indexed studies, integrating data from human and animal models will facilitate the identification of novel therapeutic targets tailored to unique neuro-metabolic interactions. Furthermore, standardized cross-species comparative frameworks are needed to enhance the predictive value of animal models in metabolic disease research. Ultimately, a deeper comprehension of pancreatic neural diversity will accelerate the development of precision medicine strategies, fostering improved outcomes in both basic science and clinical translation.