Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Neuropeptide F-expressing neurons in Drosophila constitute centrifugal pathway to optic lobes

  • Jing Wang,

    Roles Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft

    Affiliation Department of Animal Physiology, Institute of Biosciences, University of Rostock, Rostock, Germany

    ⨯
  • Fritz-Olaf Lehmann

    Roles Conceptualization, Data curation, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – review & editing

    fritz.lehmann@uni-rostock.de

    Affiliation Department of Animal Physiology, Institute of Biosciences, University of Rostock, Rostock, Germany

    ⨯

Abstract

Neuropeptide F (NPF), the Drosophila homolog of mammalian neuropeptide Y (NPY), plays a central role in the integrative regulation of internal states and behavior by modulation of diverse processes such as feeding, sleep, learning, and stress response. In this study, we systematically map the population of NPF-expressing neurons in the adult Drosophila brain using genetic labeling, microscopic imaging, and morphological analysis. Genetic labeling with GFP reveals ~50 NPF-expressing neurons, which can be grouped into five major anatomical clusters. Each cluster exhibits distinct projection patterns targeting different brain regions, suggesting specialized roles in various behavioral domains. Morphometric analysis indicates that NPF neuronal subtypes vary in soma size, soma location and arborization patterns. Besides P1, P2 and L1 neurons, we identify two ventrolateral NPF-expressing neurons per hemisphere. While the somata of ventrolateral neurons reside in the protocerebrum, their neurites project centrifugally to the optic lobes. This anatomy positions ventrolateral neurons as a potential link between the NPF system and previously reported changes in visual attention. Collectively, the data presented here extend the morphological framework of the NPF circuit in Drosophila and may help to understand the principles by which neuromodulators orchestrate brain-wide regulation of behavior.

Citation: Wang J, Lehmann F-O (2026) Neuropeptide F-expressing neurons in Drosophila constitute centrifugal pathway to optic lobes. PLoS One 21(2): e0343221. https://doi.org/10.1371/journal.pone.0343221

Editor: Carlos Oliva, Pontificia Universidad Catolica de Chile, CHILE

Received: August 25, 2025; Accepted: February 2, 2026; Published: February 27, 2026

Copyright: © 2026 Wang, Lehmann. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The data in figure 1 are available from https://doi.org/10.5281/zenodo.17908600. All other relevant data is within the manuscript and its Supporting Information files.

Funding: Deutsche Forschungsgemeinschaft (DFG) by grant LE 905/18-1 to F.O.L. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Neuropeptides serve as crucial neuromodulators in the animal kingdom, orchestrating a wide spectrum of behavioral and physiological processes by dynamically regulating neural circuit activity [1]. In Drosophila, Neuropeptide F (NPF), the functional homolog of vertebrate neuropeptide Y (NPY) [2–4], has emerged as a key molecular signal that integrates internal states with behavioral outputs [5–9]. NPF signaling influences diverse processes including hunger, arousal, stress [2,10–14], feeding, reward seeking, sleep, learning, memory, and sexual behavior [8,9,12,14–19]. Elevated NPF activity, for example, enhances hunger-driven foraging [12], increases courtship motivation [8,20], and modulates responses to aversive stimuli [9,19]. Some NPF neurons are clock neurons modifying evening locomotor activity and free-running period in Drosophila [5]. This broad functional profile establishes NPF as a behavioral gatekeeper that fine-tunes action selection based on internal need [8,14,18] and external environmental cues [15,16], thereby supporting behavioral flexibility.

Like many neuropeptides in both vertebrates and invertebrates, NPF acts primarily through G-protein-coupled receptors (GPCRs) [21–23], enabling slow, sustained, and context-dependent modulation of neuronal excitability, synaptic plasticity, and network function [3]. In contrast to fast synaptic transmission, neuropeptidergic signaling is often diffuse and modulatory, making it particularly well-suited for encoding global physiological states [2]. However, this diffuseness complicates the dissection of precise circuit mechanisms through which NPF influences specific behaviors [4]. Noteworthy, NPF-mediated behaviors are shaped not solely by peptide levels but also by the anatomical organization and circuit integration of NPF-expressing neurons [5–9]. Therefore, elucidating the anatomical logic of NPF circuits is essential for understanding how internal state are routed through specific neural pathways to generate context-appropriate behavior.

Although previous studies have identified behavioral correlates of NPF signaling and manipulated its activity in specific contexts, the NPF system is often treated as a monolithic modulator [9,24]. In Drosophila, for example, optogenetic activation of NPF neurons promotes feeding [25] and can alter a visual choice behaviors in walking flies [26]. However, NPF neurons likely comprise anatomically and functionally distinct subtypes, each tailored to regulate specific behaviors via specialized projection targets. However, while several previously published studies on NPF cells reported the cell soma of NPF cells, in many cases the arborization patterns of NPF neurons are not sufficiently described including their target neurons [5–9]. This includes possible connections to other parallel neuromodulatory streams, such as dopaminergic and serotonergic circuits with their specialized roles for motivation, motor control, and cognition [27–30].

A primary motivation of this study was to investigate the anatomical link between NPF-expressing neurons and the visual system in Drosophila, as suggested by previous experiments showing NPF-dependent modulation of visual choice behaviors [26]. We present an anatomical characterization of NPF-expressing neurons in the adult female Drosophila brain using NPF-Gal4-driven GFP labeling, confocal microscopy, and quantitative morphometry. We determine subtype identities and projection patterns of these neurons, quantify soma sizes and cell count of each subtype, and eventually discuss their connectivity with respect to various behaviors.

Materials and methods

Fly husbandry and stocks

All Drosophila melanogaster lines were reared on standard cornmeal-based medium at ~25°C under a 12-hour light/dark cycle and at ~65% relative humidity. The following fly stocks were used in this study: NPF-GAL4 (BDSC #25682, y[1] w[*]; P{w[+mC]=NPF-GAL4.1}1) and UAS-2xEGFP (BDSC #6658), both obtained from Bloomington Drosophila Stock Center (BDSC). The GAL4 driver line contains regulatory sequences from the neuropeptide F (NPF) gene inserted on the 3rd chromosome and has used before to label and manipulate NPF-expressing neurons in the central nervous system [8,31]. According to BDSC and the original donor (Jae Park, University of Tennessee, BDSC #6658 description note), this line expresses GAL4 in all neurons that produce NPF, including those in the protocerebral lobe and subesophageal ganglion. All neuronal visualization experiments were performed by crossing NPF-GAL4 flies with UAS-2xEGFP flies to drive GFP expression in NPF-positive cells. A D. melanogaster Canton S line served as a control for genetic background effects and was maintained as an inbred wild-type strain in our laboratory.

Brain dissection and immunostaining

Brain dissection and immunostaining were performed using a modified protocol based on a previous study of Drosophila mushroom body and photoreceptor neurons [32]. Briefly, 3- to 7-day old adult female flies were anesthetized on ice and their brains were dissected in ice-cold phosphate-buffered saline (PBS; pH 7.4). The dissected brains were fixed in 4% paraformaldehyde (PFA; in PBS) for one hour on a rotating platform. After fixation, the tissue was rinsed for 2 minutes in PBS and permeabilized with 0.3% Triton X-100 in PBS (PBT) for 20 minutes. All steps up to this point were carried out at room temperature (~21°C).

To visualize NPF-expressing cells, the brains were immunostained using a primary rabbit anti-GFP antibody (Invitrogen/Thermo Fisher Scientific, #G10362; 1:100) and a secondary goat anti-rabbit IgG (H + L) antibody conjugated to Alexa Fluor 488 (Invitrogen, Thermo Fisher Scientific, #A11008; 1:500). To label synaptic neuropil architecture for reference [33], mouse anti-Bruchpilot (nc82, Developmental Studies Hybridoma Bank [DSHB], #SH30898; 1:38.4) was used as a primary antibody, followed by a goat anti-mouse IgG (H + L) secondary antibody conjugated to Alexa Fluor Plus 405 (Invitrogen, Thermo Fisher Scientific, #A48255; 1:200). All antibodies were diluted in blocking buffer consisting of 5% normal goat serum in PBT. Brain samples were incubated with primary and secondary antibodies for 24h at 4°C. After each incubation step, tissues were washed twice in PBS for 20 minutes per wash. Finally, the stained brains were mounted in Vectashield antifade mounting medium (Vector Laboratories, #H-1000; refractive index = 1.45) and imaged using confocal microscopy.

Image acquisition

Immunestained brains were imaged using a Leica SP8 confocal microscope equipped with a 20x/0.40 DRY objective (HC PL FLUOTAR L). Green fluorescent protein (GFP) was excited with a 488 nm laser (OPSL 488, Leica; ~ 10 mW maximum power) at low intensity (~10% of maximum) to minimize photobleaching, and emission was collected between 495–545 nm. Alexa Fluor Plus 405 was excited using a multiphoton laser at 810 nm (MaiTai, SpectraPhysics, 2.1 W@800 nm) set to 70% intensity with emission detected in the 420–470 nm range.

Image acquisition parameters were optimized across a broad range to maximize the signal-to-noise ratio and neuronal visibility. Final settings included a scan speed of 100 lines per second, 2-line and 4-frame averaging, and a resolution of 1024 × 1024 image pixels. Detector gains were set to 700% for GFP and 1050% for Alexa Fluor Plus 405, with conventional gain maintained at 100% for both channels. The pinhole was set to 1 Airy unit for GFP and 5.68 Airy units for Alexa Fluor Plus 405 to improve optical sectioning. Each brain was scanned as z-stacks comprising 80 optical sections with a step size of 2 µm along the anterior-to-posterior axis of the brain. All images represent transverse cross-sections perpendicular to the sagittal plane. A z-layer at 0 µm shows the anterior and 160 µm the posterior end of the brain. Acquisition parameters were kept consistent across all samples to ensure comparability in subsequent morphometric analyses.

Neuron identification and statistics

Neuronal identification and morphometric analyses were performed on confocal z-stacks using ImageJ (Fiji) software. Stained neurons were classified as NPF-expressing cells (subtypes P1, P2, L1-l, L1-s) using two criteria: in the first approach, cells were counted only if both somatic signal and associated neurite were unambiguously visualized within the same optical z-layer. Since P2 neurites are thin and often difficult to visualize, we also scored cells based only on their presence of their somatic signal for comparison. The difference between two approaches is presented in the results section. Somatic signals and soma counts were validated by confirming their presence across 5–10 consecutive z-slices (10–20 µm depth). As most somata are irregular in shape, we report soma size as the mean soma diameter, calculated as the average of principal and conjugate axes of an oval fitted to the soma in the z-slice where it appeared largest. Soma location along the z-axis was defined as midpoint of the z-range in which the soma was visible. Identification of neuropil substructures, such as the layers of the fan-shaped body, was performed by aligning confocal images with the canonical anatomical maps from previous immunohistochemical studies [34,35].

To ensure measurement consistency, all measurements were repeated four times, resulting in an elevated intraclass correlation coefficient (ICC) for intra-observer reliability above ~0.95. Statistical analyses were conducted using Origin V9.0 (OriginLab Corporation, Northampton, USA). If not stated otherwise, all data are presented as means ± standard deviation of N = 15 (soma size) and N = 16 (soma count) brains.

Results

Size, location and number of NPF-expressing neurons

Across all brains examined, the size and number of NPF-neuron subtypes were consistent throughout the scanned z-stack (Fig 1). Fig 1A shows a representative GFP-fluorescence image of the entire brain, while Fig 1B displays a brain, stained for both NPF (green) and nc82 (magenta). Neuronal subtypes are classified according to previous descriptions (cf. introduction section).

thumbnail
Download:
Fig 1. Identification and morphometric analysis of NPF-neurons expressing GFP in the adult female Drosophila brain (cross-sectional view from anterior).

(A) Example shows maximum GFP intensity projection (z-stack depth: 0-160 µm) with somata (arrows) of various NPF cells. Abbreviations indicate NPF subtype. Right image side shows left brain hemisphere and vice versa. (B) Example of NPF projection via NPF-Gal4 > UAS-GFP fluorescence (green) and neuropil marker NC82 (magenta). Maximum intensity projection with z-stack depth 0-160 µm. (C) Boxplot of depth-location of NPF-somata of cell subtypes in the cross-sectional image z-stack (0 µm, anterior; 160 µm, posterior). The box is determined by the interquartile range (25th – 75th percentiles) and the whiskers extend to 5th and 95th percentiles. Means, blue; medians, red. (D) Boxplot of soma size in z-direction. For more explanations, see C. (E) Soma diameter of NPF-expressing neurons in the image plane. (F) Number of cell subtypes of NPF-neurons in the entire brain according to the existence of their somata and ignoring the appearance of attached neurites. Counts of each brain are shown by red dots. Means ± standard deviation; N = 15 brains in C – E and N = 16 in F.

https://doi.org/10.1371/journal.pone.0343221.g001

Numerical analysis suggests that each brain hemisphere contains a consistent number of NPF-neurons, i.e., ~ 25 cells per hemisphere or ~50 cells for the entire brain (rounded values, Fig 1F). Variability in cell counts arose from the difficulty in distinguishing closely spaced somata and detecting weakly fluorescence ones. For example, L1-s somata are often obscured by L1-l somata leading to corresponding variations in their counts. In each hemisphere, we typically identified one L1-l and one P1 neuron, one or two L1-s, two ventrolateral (VL) neurons, and approximately 16–23 P2 neurons (Fig 1F, Tables S1 and S2, in S1 File). The two soma identification approaches (cf. materials and methods section) yield similar counts for all subtypes except for P2 neurons. For comparison, cell counts based on both somata and the presence of an attached neurite result in ~10 P2 cells per hemisphere, which is approximately half the above value.

NPF-subtypes are distributed across different brain layers, with somata located from anterior (VL subtype, z-depth, ~ 68 μm) to posterior (P2 subtype, z-depth, ~ 142 μm; Fig 1C). P1 and P2 neurons exhibit the largest (~11.0 ± 0.81 μm xy-diameter, ~ 11.0 ± 0.21 μm z-diameter) and smallest (~4.90 ± 0.70 μm, ~ 4.38 ± 0.62 μm) somata, respectively (Fig 1D and E, Tables S3 and S4 in S1 File). Somata of L1-large (~9.00 ± 1.02 μm, ~ 8.99 ± 2.21 μm), L1-small (~6.38 ± 0.26 μm, ~ 6.66 ± 1.14 μm), and VL neurons (~7.29 ± 0.69 μm, ~ 7.49 ± 1.88 μm) have intermediate size. The similar diameters in all three dimensions suggest nearly spherical somata, with estimated cell volumes of 54.3 ± 18.8 μm3, 723 ± 254 μm3, 396 ± 128 μm3, 151 ± 45 μm3 and 245 ± 125 μm3 for P2, P1, L1-l, L1-s and VL cells, respectively.

Arborization and target specifity of P1 and P2 neurons

The somata of P1 neurons are located in the inferior protocerebrum of the posterior dorsomedial brain. Anatomical mapping based on nc82-labeled neuropil regions suggests that P1 dendrites innervate multiple brain regions (Fig 2). These include (i) the superior lateral protocerebrum (SLP) in the superior protocerebrum, (ii) the superior clamp (SCL), (iii) the inferior clamp (ICL), (iv) the inferior bridge (IB) in the inferior protocerebrum, (v) vest (VES) and superior posterior slope (SPS) in the ventromedial protocerebrum, (vi) the anterior ventrolateral protocerebrum (AVLP), (vii) the posterior ventrolateral protocerebrum (PVLP), (viii) the posterior lateral protocerebrum (PLP) in the ventrolateral protocerebrum, (ix) the cantle (CAN) in the periesophageal ganglion and (x) the gnathal ganglion (GNG). P1 neurons typically show extensive arborization patterns, with varicosity-enriched synaptic sites in the ventromedial protocerebrum (VMPr, Fig. 2C-E). The schematics in Fig 2F outlines the neuroanatomical organization and projection of the P1 neuron subtype.

thumbnail
Download:
Fig 2. Morphology of P1, L1-l (large) and L1-s (small) neurons.

(A) P1 and L1 neurons in a 0–160 μm z-stack. P1 cell morphology suggests widespread axonal projections to multiple brain regions, including the superior lateral protocerebrum (SLP) in the superior protocerebrum, the superior clamp (SCL), the inferior clamp (ICL), the inferior bridge (IB) in the inferior protocerebrum, the vest (VES) and superior posterior slope (SPS) in the ventromedial protocerebrum, the anterior ventrolateral protocerebrum (AVLP), the posterior ventrolateral protocerebrum (PVLP), the posterior lateral protocerebrum (PLP) in the ventrolateral protocerebrum, the cantle (CAN) in the periesophageal ganglion and the gnathal ganglion (GNG). (B) L1-projection via NPF-Gal4 > UAS-GFP fluorescence (green) and neuropil marker NC82 (magenta). (C-E) Magnified examples of synaptic varicosities along P1 projections in the ventromedial protocerebrum (VMPr, region of interest in A). (F) Schematic drawing of P1 neuroarchitecture (orange) including somata localization, projection patterns, and synaptic varicosities. CB, central brain. (G) Schematic drawing of soma and arborization of L1-l (green) and L1-s (blue) neurons in the central brain (white). (H) Magnified view of terminal projections in the inferior protocerebrum of L1 neurons with their somata in or near the brain’s lateral horn (80–120 µm z-depth). Image suggests the soma of a third L1 neuron as found in some brains. (I) Dorsoventral neurite splitting of L1-s neuron (stars).

https://doi.org/10.1371/journal.pone.0343221.g002

P2 neurons have their soma in the posterior-dorsal inferior protocerebrum and extend neurites to the fan-shaped body (FSB), a key structure of the central complex (Figs 2A and 3) [35,36]. Fig 3E shows an example in which P2 somata are circled for counting. A detailed analysis of z-stack images shows that P2 axonal branches target distinct layers of the FSB, as illustrated in the schematic drawing (Fig 3B–D). A comparison with a previous study on the FSB [34] indicates that P2 axons arborize in FSB layers I, IV, and VI, suggesting a potential interaction with integrating sensory and motor pathways [18,36–38].

thumbnail
Download:
Fig 3. Morphology of P2 and VL neurons.

(A) Merged maximum-intensity projections (z-depth, 0–160 μm) NPF-Gal4 > UAS-GFP (green) and NC82 (magenta) fluorescence. Somata of P2 neurons are located in the inferior protocerebrum (posterior dorsal brain) and VL (L2-neuron) somata in the ventrolateral protocerebrum close to the optic lobes. Regions ii and i are shown as close-ups in E, F and G, respectively. (B,C) P2-neurits occur in different layers of the fan-shape body (intensity projection, 0–160 μm). (D) Schematic summary of P2 neuroarchitecture (white) within the fan-shape body (FSB) and its projection (green) in the 9 layers. (E) Example of P2 somata marked with white circles in a z-stack depth of 150–154 µm. (F,G) Close-ups of VL-neurite branching (stars, single confocal z-stack) in the optic lobe with varicosities (fluorescence dots). Z-depth, 50–70 μm. (H) Reconstruction of arborization of VL (L2) neurons within the optic lobes. CB, central brain; OL, optic lobe.

https://doi.org/10.1371/journal.pone.0343221.g003

Subtype-specific organization of L1-l and L1-s neurons

L1-l (l, large) and L1-s (s, small) neurons have their somata near or in the lateral horn and are distinguishable by soma size and projection patterns (Fig 2G,H,I). Neurites of L1-l neurons extend through the lateral horn and branch into varicosity-rich regions of the inferior protocerebrum. By contrast, the smaller L1-s neurons were less consistently detected (2.00 ± 0.42 neurons per hemisphere) and have projections in the ventrolateral and inferior protocerebrum. In some images, we noticed a third small soma near the L1 somata (Fig 2H). The morphological schematic drawing in Fig 2G suggests potential functional overlap between L1-l and L1-s in information processing.

VL neurons and their arborization in the optic lobes

In all examined brains, we found NPF-expressing neurons with their somata located in the ventrolateral protocerebrum (Figs 1A,B; 2A,B; 3A,F-H). We termed these neurons ventrolateral (VL) neurons (see disucssion section). The neurites originating from their somata project centrifugally toward the optic lobes, coursing dorsolaterally through the anterior-ventrolateral protocerebrum before bifurcating into bilateral branches. These branches seem to arborize within visual neuropils, likely including the lobula complex and stratified medulla [39,40]. This laminar specificity aligns with the hierarchical organization of medulla circuits, where distinct layers process sequential visual features such as motion direction (M3) and chromatic signals (M5) [40–42]. Noticeably, VL trajectories appears to mirror the trajectories of visual projection neurons (LPLC2) that potentially suggests a role in modulation of visual-motor circuits.

The strongest fluorescence signaling we found in varicosities of VL terminals in the medulla that is consistent with previously suggested putative neuropeptide release sites [40]. By contrast, VL collaterals in the lobula complex appear to exhibit limited branching, although some varicose terminals appear near the suggested positions of dendrites of lobula columnar neurons (LCNs) [43]. LNCs relay visual information to the central brain [39]. In sum, the morphology and projection pattern of VL neurons suggests that they might modulate layer-specific visual information processing, potentially underlying NPF-mediated changes in visual behaviors (Fig 3H) [26].

Discussion

General remarks

Our analysis of NPF-expressing neurons in the adult female Drosophila brain reveals a diverse population organized into five anatomically distinct subtypes (Fig 4). These include the P1 and P2 clusters with somata in the inferior protocerebrum (medial dorsal brain), L1-l and L1-s neurons with somata near the lateral horn, and VL neurons with somata positioned near the optic lobes. In general, our anatomical reconstruction aligns with previous findings on NPF cell morphology [5–9,44–46].

thumbnail
Download:
Fig 4. Reconstructed neuroarchitecture of NPF neurons in the adult female Drosophila brain.

CB, central brain; FSB, fan-shaped body; IP, inferior protocerebrum; LA, lamina; LH, lateral horn; LO, lobula; LP, lobula plate; ME, medulla; OC, ocelli; OL, optic lobe (lobula, lobula plate); RE, retina. For more details, see legends to Figs 1–3 and main text.

https://doi.org/10.1371/journal.pone.0343221.g004

Notably, we could not detect the existence of a sixth subtype of NPF neurons (DM neurons) that has been described in the medial dorsal brain in a prior study of NPF network function [46]. These small neurons are reported to consists of two sets of four to five neurons that may induce preference changes in a high-throughput two-choice learning assay upon activation [46]. Importantly, the activation of DM neurons alone is sufficient to induce these changes. The absence of DM neurons in our brain samples might reflect differences in genetic drivers, staining sensitivity, or physiological condition of the animals.

P1, P2 and L1 neurons

The existence of five NPF-expressing clusters suggests that NPF signaling is not monolithic but instead operates through different pathways. This is consistent with the function of NPF for diverse behaviors such as feeding, courtship, and stress responses [8,9,12–16,18–20,24,25,44–46]. For example, P1 neurons arborize widely across superior and inferior protocerebral domains (Fig 1B), innervating regions such as the superior lateral protocerebrum and the lateral horn. These areas are further associated with sensory integration centers, and motor coordination areas like the gnathal ganglion [44]. Their widespread projections suggest a prominent role in integrating multimodal inputs to modulate complex behaviors such as feeding or courtship [9]. The distribution of varicosities of their terminals in the anterior ventrolateral protocerebrum further supports this hypothesis [47]. In contrast to P1 neurons, P2 neurons exhibit more targeted projections, restricted to specific layers of the fan-shaped body (FSB), a central complex structure critical for spatial navigation and sensorimotor integration [35,38,48]. To better map the neurit projections of P2 neurons on the layers of the fan-shaped body, we used a previously published work on FSB structure for comparison [34]. According to the imaged FSB details of the latter study, we aligned the fluorescence signals shown in Fig 3B and C with the appropriate layers of the FSB as shown in Fig 3D. The found mapping of FSB-specific layers is consistent with previous research and also suggests that P2 neurons might gate context-dependent transitions between locomotor states such as hunger-driven foraging versus satiety-induced quiescence [11,36,37,49].

Somata of L1-neurons are located near or within the lateral horn and mainly arborize in the dorsal medial protocerebrum. This region plays a crucial role in memory consolidation and sleep regulation. Notably, this area contains dorsal paired medial neurons (DPM), which are involved in long-term memory storage [50,51]. Consistent with a potential memory-related function, a previous behavioral study reported that selective activation of L1-l and P1 neurons impairs associative learning [46].

VL neurons and NPF signaling in optic lobes

Superficially, the somata of ventrolateral neurons identified in this study appear similar to somata of NPF neurons shown by in situ localization of npf RNA in Drosophila [9]. These cells are located in the posterior region of the brain and were termed L2 neurons [9]. However, the authors of the latter study reported only 1.5 ± 0.2 and 1.3 ± 0.3 neurons (means ± standard deviation) for each brain hemisphere in males and females, respectively, whereas we typically found two neurons for each body side in females. In addition, a study on clock neurons reported in total ~6 NPF-positive somata at the transition between ventrolateral protocerebrum and optic lobe (l-LNv, 1.6 ± 0.6; fifth s-LNv, 1.0 ± 0; LNd, 3.5 ± 0.5) [5]. In npf-ablated flies, two of the LNds and two of the l-LNvs clock neurons remain intact, indicating the ablation of 4 LNds and 2–3 l-LNvs NPF-positive clock neurons [5]. As none of these previous findings completely match our data, we introduced the cells in this study as “ventrolateral neurons” but cannot exclude that these cells belong to the group of clock neurons identified earlier. It is possible that at least one VL neuron is similar to the 5th s-LNv clock neuron as shown by cell count [5]. It has been suggested that NPF-positive clock neurons, especially the fifth s-LNv and the LNds are involved in changes of evening locomotor activity and free running period [5]. Fluorescence imaging of sNPF in a study that tackled all peptidergic neurons in Drosophila also shows NPF-GFP signaling of somata at the transition between protocerebrum and optic lobe [6]. The authors of the latter study, however, made no further attempt to identify these signals [6].

Noteworthy, none of the previous studies, reconstructed the arborization pattern of the described neurons in greater detail. The neurites of VL neurons extend sparse collaterals to the lobula complex while forming extensive arborizations within intermediate medulla layers. This stratified projection pattern suggests that NPF release may occur at the level of early visual processing [52,53], potentially explaining the previously observed NPF-dependent changes in visual objects preference [26]. Assuming that VL neurons establish collaterals near lobula columnar neurons (LCNs), this connection could support visual gain modulation during behaviors such as gaze stabilization or object tracking during locomotion [11,40,42,49,52,54–57]. A query of the VirtualFlyBrain connectome [43] further identified a neuron (ID: FBbt_00111763) with a morphology highly reminiscent of our VL neurons, projecting from the ventrolateral protocerebrum to the optic lobes. This aligns with behavioral evidence that links NPF to visual attention and courtship motivation [26,56]. Eventually, the combination of sparse cell number and well-developed medulla arborizations might suggest a “broadcast” modulatory mechanism, whereby VL neurons could amplify NPF signals across retinotopic circuits.

Conclusions

Collectively, our findings present a careful anatomical map of NPF-expressing neurons in the adult female Drosophila brain. We characterized soma size, location, and arborization patterns of five NPF neuronal subtypes across both the central brain and optic lobes. This work was initially motivated by behavioral evidence that NPF activation modulates visual attention [26], while previous anatomical research on NPF was often limited to central brain regions and the proof of cell somata [46]. The finding of bilateral VL (possibly L2) neurons, each comprising ~2 neurons with arborizations in the optic lobes, might provide an anatomical substrate for direct modulation of visual circuits by NPF. However, our morphological analysis does not explain how the NPF system is activated in behaving animals or its precise mode of action at target synapses. Elucidating the functional relationship between NPF neuron architecture and circuit-level mechanisms thus remains an essential direction for future research to uncover how this neuropeptide translates internal brain states into adaptive behaviors.

Supporting information

S1 File.

S1 Table. Data as shown in figure 1F of the main text. Cell counts are shown for both each hemisphere and the entire brain. N = 15 brains for soma size and location; N = 16 brains for cell counts; n.c., not calculated; 1, cell counts based on the presence of somatic signal only; 2, cell counts based on somatic signal and associated neurite within the same optical z-layer; mean ± standard deviation. S2 Table. Cell counts for each brain. Values in parenthesis show cell counts for left and right hemisphere (left, right), respectively. If values in parenthesis are not shown, cell counts are the same for each brain hemisphere or could not be estimated. 1, cell counts based on the presence of somatic signal only; 2, cell counts based on somatic signal and associated neurite within the same optical z-layer. S3 Table. Cell soma size. Data are means of each cell type in each brain. S4 Table. Cell soma location in z-direction (z-depth). Data are means of each cell type.

https://doi.org/10.1371/journal.pone.0343221.s001

(ZIP)

Acknowledgments

We thank Bärbel Redlich-Witt for her help on brain staining and microscopy.

References

  1. 1. van den Pol AN. Neuropeptide transmission in brain circuits. Neuron. 2012;76(1):98–115. pmid:23040809
  2. 2. Miyamoto T, Amrein H. Neuronal Gluconeogenesis Regulates Systemic Glucose Homeostasis in Drosophila melanogaster. Curr Biol. 2019;29(8):1263-1272.e5. pmid:30930040
  3. 3. Nässel DR, Wegener C. A comparative review of short and long neuropeptide F signaling in invertebrates: Any similarities to vertebrate neuropeptide Y signaling?. Peptides. 2011;32(6):1335–55. pmid:21440021
  4. 4. Brown MR, Crim JW, Arata RC, Cai HN, Chun C, Shen P. Identification of a Drosophila brain-gut peptide related to the neuropeptide Y family. Peptides. 1999;20(9):1035–42. pmid:10499420
  5. 5. Hermann C, Yoshii T, Dusik V, Helfrich-Förster C. Neuropeptide F immunoreactive clock neurons modify evening locomotor activity and free-running period in Drosophila melanogaster. J Comp Neurol. 2012;520(5):970–87. pmid:21826659
  6. 6. Deng B, Li Q, Liu X, Cao Y, Li B, Qian Y, et al. Chemoconnectomics: Mapping Chemical Transmission in Drosophila. Neuron. 2019;101(5):876-893.e4. pmid:30799021
  7. 7. Bozler J, Kacsoh BZ, Bosco G. Transgeneratonal inheritance of ethanol preference is caused by maternal NPF repression. Elife. 2019;8:e45391. pmid:31287057
  8. 8. Liu W, Ganguly A, Huang J, Wang Y, Ni JD, Gurav AS, et al. Neuropeptide F regulates courtship in Drosophila through a male-specific neuronal circuit. Elife. 2019;8:e49574. pmid:31403399
  9. 9. Lee G, Bahn JH, Park JH. Sex- and clock-controlled expression of the neuropeptide F gene in Drosophila. Proc Natl Acad Sci U S A. 2006;103(33):12580–5. pmid:16894172
  10. 10. Bidaye SS, Machacek C, Wu Y, Dickson BJ. Neuronal control of Drosophila walking direction. Science. 2014;344(6179):97–101. pmid:24700860
  11. 11. Liu L, Wolf R, Ernst R, Heisenberg M. Context generalization in Drosophila visual learning requires the mushroom bodies. Nature. 1999;400(6746):753–6. pmid:10466722
  12. 12. Wu Q, Wen T, Lee G, Park JH, Cai HN, Shen P. Developmental control of foraging and social behavior by the Drosophila neuropeptide Y-like system. Neuron. 2003;39(1):147–61. pmid:12848939
  13. 13. Yoshinari Y, Kosakamoto H, Kamiyama T, Hoshino R, Matsuoka R, Kondo S, et al. The sugar-responsive enteroendocrine neuropeptide F regulates lipid metabolism through glucagon-like and insulin-like hormones in Drosophila melanogaster. Nat Commun. 2021;12(1):4818. pmid:34376687
  14. 14. Chung BY, Ro J, Hutter SA, Miller KM, Guduguntla LS, Kondo S, et al. Drosophila Neuropeptide F Signaling Independently Regulates Feeding and Sleep-Wake Behavior. Cell Rep. 2017;19(12):2441–50. pmid:28636933
  15. 15. Kim WJ, Jan LY, Jan YN. A PDF/NPF neuropeptide signaling circuitry of male Drosophila melanogaster controls rival-induced prolonged mating. Neuron. 2013;80(5):1190–205. pmid:24314729
  16. 16. Krashes MJ, DasGupta S, Vreede A, White B, Armstrong JD, Waddell S. A neural circuit mechanism integrating motivational state with memory expression in Drosophila. Cell. 2009;139(2):416–27. pmid:19837040
  17. 17. Li W, Pan Y, Wang Z, Gong H, Gong Z, Liu L. Morphological characterization of single fan-shaped body neurons in Drosophila melanogaster. Cell Tissue Res. 2009;336(3):509–19. pmid:19357870
  18. 18. Quinn WG, Harris WA, Benzer S. Conditioned behavior in Drosophila melanogaster. Proc Natl Acad Sci U S A. 1974;71(3):708–12. pmid:4207071
  19. 19. Shohat-Ophir G, Kaun KR, Azanchi R, Mohammed H, Heberlein U. Sexual deprivation increases ethanol intake in Drosophila. Science. 2012;335(6074):1351–5. pmid:22422983
  20. 20. Gao XJ, Riabinina O, Li J, Potter CJ, Clandinin TR, Luo L. A transcriptional reporter of intracellular Ca(2+) in Drosophila. Nat Neurosci. 2015;18(6):917–25. pmid:25961791
  21. 21. Du P, Salon JA, Tamm JA, Hou C, Cui W, Walker MW, et al. Modeling the G-protein-coupled neuropeptide Y Y1 receptor agonist and antagonist binding sites. Protein Eng. 1997;10(2):109–17. pmid:9089810
  22. 22. Singh G, Davenport AP. Neuropeptide B and W: neurotransmitters in an emerging G-protein-coupled receptor system. Br J Pharmacol. 2006;148(8):1033–41. pmid:16847439
  23. 23. Xu J, Li M, Shen P. A G-protein-coupled neuropeptide Y-like receptor suppresses behavioral and sensory response to multiple stressful stimuli in Drosophila. J Neurosci. 2010;30(7):2504–12. pmid:20164335
  24. 24. Li W, Wang Z, Syed S, Lyu C, Lincoln S, O’Neil J, et al. Chronic social isolation signals starvation and reduces sleep in Drosophila. Nature. 2021;597(7875):239–44. pmid:34408325
  25. 25. Williams MJ, Akram M, Barkauskaite D, Patil S, Kotsidou E, Kheder S, et al. CCAP regulates feeding behavior via the NPF pathway in Drosophila adults. Proc Natl Acad Sci U S A. 2020;117(13):7401–8. pmid:32179671
  26. 26. Grabowska MJ, Steeves J, Alpay J, Van De Poll M, Ertekin D, van Swinderen B. Innate visual preferences and behavioral flexibility in Drosophila. J Exp Biol. 2018;221(Pt 23):jeb185918. pmid:30322983
  27. 27. Lundell MJ, Hirsh J. Temporal and spatial development of serotonin and dopamine neurons in the Drosophila CNS. Dev Biol. 1994;165(2):385–96. pmid:7958407
  28. 28. Niens J, Reh F, Çoban B, Cichewicz K, Eckardt J, Liu Y-T, et al. Dopamine Modulates Serotonin Innervation in the Drosophila Brain. Front Syst Neurosci. 2017;11:76. pmid:29085286
  29. 29. Kasture AS, Hummel T, Sucic S, Freissmuth M. Big Lessons from Tiny Flies: Drosophila melanogaster as a Model to Explore Dysfunction of Dopaminergic and Serotonergic Neurotransmitter Systems. Int J Mol Sci. 2018;19(6):1788. pmid:29914172
  30. 30. Neckameyer WS, Bhatt P. Neurotrophic actions of dopamine on the development of a serotonergic feeding circuit in Drosophila melanogaster. BMC Neurosci. 2012;13:26. pmid:22413901
  31. 31. Munneke AS, Chakraborty TS, Porter SS, Gendron CM, Pletcher SD. The serotonin receptor 5-HT2A modulates lifespan and protein feeding in Drosophila melanogaster. Front Aging. 2022;3:1068455. pmid:36531741
  32. 32. Kelly SM, Elchert A, Kahl M. Dissection and immunofluorescent staining of mushroom body and photoreceptor neurons in adult Drosophila melanogaster brains. JoVE. 2017;129:56174.
  33. 33. Nieratschker V, Schubert A, Jauch M, Bock N, Bucher D, Dippacher S, et al. Bruchpilot in ribbon-like axonal agglomerates, behavioral defects, and early death in SRPK79D kinase mutants of Drosophila. PLoS Genet. 2009;5(10):e1000700. pmid:19851455
  34. 34. Kahsai L, Winther AME. Chemical neuroanatomy of the Drosophila central complex: distribution of multiple neuropeptides in relation to neurotransmitters. J Comp Neurol. 2011;519(2):290–315. pmid:21165976
  35. 35. Hu W, Peng Y, Sun J, Zhang F, Zhang X, Wang L, et al. Fan-Shaped Body Neurons in the Drosophila Brain Regulate Both Innate and Conditioned Nociceptive Avoidance. Cell Rep. 2018;24(6):1573–84. pmid:30089267
  36. 36. Jones JD, Holder BL, Montgomery AC, McAdams CV, He E, Burns AE, et al. The dorsal fan-shaped body is a neurochemically heterogeneous sleep-regulating center in Drosophila. PLoS Biol. 2025;23(3):e3003014. pmid:40138668
  37. 37. Pan Y, Zhou Y, Guo C, Gong H, Gong Z, Liu L. Differential roles of the fan-shaped body and the ellipsoid body in Drosophila visual pattern memory. Learn Mem. 2009;16(5):289–95. pmid:19389914
  38. 38. Currier TA, Matheson AM, Nagel KI. Encoding and control of orientation to airflow by a set of Drosophila fan-shaped body neurons. Elife. 2020;9:e61510. pmid:33377868
  39. 39. Meinertzhagen IA, Takemura S, Lu Z, Huang S, Gao S, Ting C-Y, et al. From form to function: the ways to know a neuron. J Neurogenet. 2009;23(1–2):68–77. pmid:19132600
  40. 40. Currier TA, Pang MM, Clandinin TR. Visual processing in the fly, from photoreceptors to behavior. Genetics. 2023;224(2):iyad064. pmid:37128740
  41. 41. Borst A, Haag J, Reiff DF. Fly motion vision. Annu Rev Neurosci. 2010;33:49–70. pmid:20225934
  42. 42. Cheong HS, Siwanowicz I, Card GM. Multi-regional circuits underlying visually guided decision-making in Drosophila. Curr Opin Neurobiol. 2020;65:77–87. pmid:33217639
  43. 43. Klapoetke NC, Murata Y, Kim SS, Pulver SR, Birdsey-Benson A, Cho YK, et al. Independent optical excitation of distinct neural populations. Nat Methods. 2014;11(3):338–46. pmid:24509633
  44. 44. Lin A, Yang R, Dorkenwald S, Matsliah A, Sterling AR, Schlegel P, et al. Network statistics of the whole-brain connectome of Drosophila. Nature. 2024;634(8032):153–65. pmid:39358527
  45. 45. Schlegel P, Yin Y, Bates AS, Dorkenwald S, Eichler K, Brooks P, et al. Whole-brain annotation and multi-connectome cell typing of Drosophila. Nature. 2024;634(8032):139–52. pmid:39358521
  46. 46. Shao L, Saver M, Chung P, Ren Q, Lee T, Kent CF, et al. Dissection of the Drosophila neuropeptide F circuit using a high-throughput two-choice assay. Proc Natl Acad Sci U S A. 2017;114(38):E8091–9. pmid:28874527
  47. 47. Cardona A, Saalfeld S, Preibisch S, Schmid B, Cheng A, Pulokas J, et al. An integrated micro- and macroarchitectural analysis of the Drosophila brain by computer-assisted serial section electron microscopy. PLoS Biol. 2010;8(10):e1000502. pmid:20957184
  48. 48. Liu Q, Liu S, Kodama L, Driscoll MR, Wu MN. Two dopaminergic neurons signal to the dorsal fan-shaped body to promote wakefulness in Drosophila. Curr Biol. 2012;22(22):2114–23. pmid:23022067
  49. 49. Vogt K, Aso Y, Hige T, Knapek S, Ichinose T, Friedrich AB, et al. Direct neural pathways convey distinct visual information to Drosophila mushroom bodies. Elife. 2016;5:e14009. pmid:27083044
  50. 50. Haynes PR, Christmann BL, Griffith LC. A single pair of neurons links sleep to memory consolidation in Drosophila melanogaster. Elife. 2015;4:e03868. pmid:25564731
  51. 51. Keene AC, Krashes MJ, Leung B, Bernard JA, Waddell S. Drosophila dorsal paired medial neurons provide a general mechanism for memory consolidation. Curr Biol. 2006;16(15):1524–30. pmid:16890528
  52. 52. Schretter CE, Hindmarsh Sten T, Klapoetke N, Shao M, Nern A, Dreher M, et al. Social state alters vision using three circuit mechanisms in Drosophila. Nature. 2025;637(8046):646–53. pmid:39567699
  53. 53. Sun Y, Nern A, Franconville R, Dana H, Schreiter ER, Looger LL, et al. Neural signatures of dynamic stimulus selection in Drosophila. Nat Neurosci. 2017;20(8):1104–13. pmid:28604683
  54. 54. Koenig S, Wolf R, Heisenberg M. Visual Attention in Flies-Dopamine in the Mushroom Bodies Mediates the After-Effect of Cueing. PLoS One. 2016;11(8):e0161412. pmid:27571359
  55. 55. Malin J, Desplan C. Neural specification, targeting, and circuit formation during visual system assembly. Proc Natl Acad Sci U S A. 2021;118(28):e2101823118. pmid:34183440
  56. 56. Shinomiya K, Nern A, Meinertzhagen IA, Plaza SM, Reiser MB. Neuronal circuits integrating visual motion information in Drosophila melanogaster. Curr Biol. 2022;32(16):3529-3544.e2. pmid:35839763
  57. 57. Vogt K, Schnaitmann C, Dylla KV, Knapek S, Aso Y, Rubin GM, et al. Shared mushroom body circuits underlie visual and olfactory memories in Drosophila. Elife. 2014;3:e02395. pmid:25139953