Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2008 Nov;37(6):443-58.
doi: 10.1016/j.asd.2008.03.002. Epub 2008 Jul 17.

Higher order visual input to the mushroom bodies in the bee, Bombus impatiens

Affiliations

Higher order visual input to the mushroom bodies in the bee, Bombus impatiens

Angelique C Paulk et al. Arthropod Struct Dev. 2008 Nov.

Abstract

To produce appropriate behaviors based on biologically relevant associations, sensory pathways conveying different modalities are integrated by higher-order central brain structures, such as insect mushroom bodies. To address this function of sensory integration, we characterized the structure and response of optic lobe (OL) neurons projecting to the calyces of the mushroom bodies in bees. Bees are well known for their visual learning and memory capabilities and their brains possess major direct visual input from the optic lobes to the mushroom bodies. To functionally characterize these visual inputs to the mushroom bodies, we recorded intracellularly from neurons in bumblebees (Apidae: Bombus impatiens) and a single neuron in a honeybee (Apidae: Apis mellifera) while presenting color and motion stimuli. All of the mushroom body input neurons were color sensitive while a subset was motion sensitive. Additionally, most of the mushroom body input neurons would respond to the first, but not to subsequent, presentations of repeated stimuli. In general, the medulla or lobula neurons projecting to the calyx signaled specific chromatic, temporal, and motion features of the visual world to the mushroom bodies, which included sensory information required for the biologically relevant associations bees form during foraging tasks.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
Anatomy of the bumblebee brain. (A) The bumblebee brain structures which are visible superficially are the retina (ret) which sends visual inputs to the optic lobes (ol), the antennal lobes (al), which are the primary olfactory processing centers, and the central brain, which includes the mushroom bodies, the calyces of which (mbc) are visible superficially. (B) frontal brain section (osmium stained) showing the second (medulla, me) and the third optic lobe neuropil (lobula, lo), the central protocerebrum outside the mushroom bodies (pr) and the mushroom bodies (mb) composed of a lateral and medial calyx (mb-lc and mb-mc, respectively) and mushroom body lobes (mb-l). (C) The box in (B) highlights a section of the lateral calyx, which is comprised of the lip (lip), collar (col), and basal ring (br) and the Kenyon cell bodies (KCs) enlarged in this Bodian preparation. (D-F) Fluorescent fills showing visual inputs from the medulla (D), the medulla (black arrowheads) and the lobula (white arrowheads) (E) and olfactory inputs from the antennal lobe (F) to the calyx. Unless otherwise noted, scale bar = 100 μm. All images from bumblebee preparations.
Fig. 2
Fig. 2
Visual stimuli. The layout of the two types of visual stimuli relative to the bee is represented by a sphere around the bees' head. Both the light emitting diodes (LEDs) and motion stimuli using the computer monitor (CRT) were presented on the right side of the bees, but, for the sake of clarity, are placed on either side of the bee in this figure. The array of sixty violet, blue, and green LEDs are oriented to the long axis of the eye subtending 2.9° by 123.3° of the visual field, the computer monitor subtended 114.0° by 103.1°.
Fig. 3
Fig. 3
Mushroom body (MB) visual input neuron projections. (A) Input regions of the MB visual input neurons with characteristic fine processes in the lobula. (B) Outputs in the mushroom body calyx with characteristic blebbed varicosities. (C-G) lobula MB input neurons, showing the large range of differences in the dendritic branching patterns projecting in the lobula (lo) to the MB lateral and medial calyx (mb-lc and mb-mc), and to the superior lateral protocerebrum (s l pr). (C-F) Large field lobula MB input neurons have dendrites that span much of the lobula. However, these neurons do not have processes extending to the other side of the brain (white arrowhead in C). (G) Conversely, this small field lobula MB input neuron with fine dendrites in a small region in the lobula has a projection leaving the medial MB calyx toward the midline (white arrowhead). Boxes in (D) represent areas enlarged in Fig. 4 E and 5E and F, respectively. Boxes in (F) and (G) represent areas enlarged in Fig. 5 A and c, respectively. In (F), two large field lobula MB input neurons are filled and reconstructed in the same brain. For (A) and (B), they are two-dimensionally projected images single 15 μm plastic sections. For (C-G), the entire dendritic projection of each neuron in the lobula and the axons in the protocerebrum outside of the MB calyces are represented here as two-dimensional projections across plastic sections. Scale bar = 100 μm
Fig. 4
Fig. 4
Distinctive branching patterns of visual mushroom body input neurons. (A) Frontal view of an unstained bumblebee brain (top; broken line indicates position of section shown below) and horizontal section (bottom) of an osmium-stained preparation; inset enlarged in (B) to show the six layers in the lobula (Ribi and Scheel, 1981). (C) Dendritic arborizations of recorded and labeled lobula MB input neurons (white) and lobula non-MB input neurons (magenta) can be assigned to the different lobula layers shown in (B) in this horizontal section. Notably, the MB input neurons branch in layers 5 and 6 while the non-MB input neurons send their branches into layers 1-4. (any overlap of the magenta branching with layer 5 is due to the three-dimensional structure of the lobula layers and the neurons, which is difficult to depict two-dimensionally) (D) Lobula specializations of the neuron shown in Fig. 3D two-dimensionally projected within a single 15 μm plastic section. Note the close vicinity of spines (blue arrows) and bleb-like structures (red arrows) in the lobula (D). (E) The medulla neuron with input to the MBs projects through the medulla (me), around the lobula, to the MB lateral and medial calyx; cell body located below the lateral calyx. (F) A closer view of the medulla dendritic arborizations. Species represented: Bombus impatiens: A,B,D-F; Apis mellifera: C. Boxes in (C) and (E) represent areas enlarged in Fig. 5 B and D, respectively. Unless otherwise noted, scale bar = 100 μm.
Fig. 5
Fig. 5
Terminal fibers of visual input neurons in the calyces. (A-C) The lobula MB input neurons feature blebbed terminals in the inner rim of the MB calyces (arrows). The large field lobula MB input neurons projected into the innermost collar region (A, B) while the small-field lobula MB input neuron projected closer to the outer collar region (C). (A, C) frontal sections; the dorsal view (horizontal section) in (B) reveals that the lobula MB input neurons input branches project all around the inner rim of the calyx. (D) Medulla MB input neuron sends blebbed varicosities into the outer rim of the collar in the calyces (arrows). (E, F) Specializations of the neuron shown in Fig. 3D. Bleb-like terminals (arrows) near the lip of the calyx (E) and in superior lateral protocerebrum (s l pr) below the lateral MB calyx (F). MB pedunculus (ped); In all cases, the images are two-dimensionally projected within a single 15 μm plastic section. Species represented: Bombus impatiens: A, C, D-F; Apis mellifera: B. A-D: Scale bar = 100 μm. E, F: Scale bar = 10 μm.
Fig. 6
Fig. 6
Chromatic response characteristics of MB input neurons. (A) The medulla MB input neuron responded phasically to a flash of violet light. (B) The medulla neuron preferentially responded to the presence of violet light over green or blue light as seen in a raster plot and peristimulus time histogram (PSTH; bin-width 5 ms) of the cell's responses to five consecutive light flashes to stimuli including violet (black bars) or not including violet light (grey bars). Abbreviations: blue (b), green (g), blue-green (b-g), violet (v), and violet-green (v-g) light. (C) A large field lobula MB input neuron is phasically inhibited by green, excited by blue, and has no obvious response for violet LED light. (D) Another large field lobula MB input neuron had no response to violet light, had a phasic response for green light which was suppressed by the simultaneous presence of violet light. The response presented here is to the first of five light flashes in all three cases. (E) The response of a large field lobula MB input neuron to blue and green light was inhibited by the concurrent presence of violet light (arrows). The black bars signify the duration of the light flash stimulus, which is 500 ms in all cases.
Fig. 7
Fig. 7
Motion response characteristics of MB input neurons. (A) A large field lobula MB input neuron responded to directional motion of a moving edge. Traces (Ai) and (Aii) show different trials. (B) A small field lobula MB input neuron responded differently to a leftward moving black edge and a rightward moving white edge (Bi) compared to a rightward moving black edge and a leftward moving white edge (Bii), but was not sensitive to directional motion. (C) A large field lobula MB input neuron responded to a directionally moving grating with inhibition of different strength in response to different directions (Ci- Civ). Arrows indicate direction of movement. Black bars signify stimulus duration in all cases.
Fig. 8
Fig. 8
Temporal response characteristics of individual MB input neurons. (A) One small field lobula MB input neuron responded with a phasic, highly precise spike at the beginning of the light flash to simultaneously presented blue-green-violet stimuli over fifteen trials, as illustrated in the raster plot and peristimulus time histogram (PSTH). Highly precise spikes responses is also observed in two large field lobula MB input neurons, with an example in (B) and in the medulla MB input neuron (Fig. 6A,B). (C) A small field lobula neuron increases its spike rate over five sequential light flashes for blue and somewhat for green light, but not for violet light. (D) In addition, the normalized spike rate for each light flash increases with the progress of light flashes for blue-green-violet, blue, and green stimuli (black) but not for violet (grey) (one trial for each light flash color). A-C: Black bars signify the duration of the light flash stimulus, which is 500 ms in all cases.
Fig. 9
Fig. 9
Stimulus entrainment. (A) After two to three presentations of violet light, the spiking ‘response’ of the large field lobula MB input neuron actually precedes the light flashes (black arrows); this effect is much less pronounced or absent in responses to the blue light stimulus. Only after the end of violet stimulus series, the neuron produced a burst of activity (white arrow) which coincided with the ‘expected’ timing of a sixth light flash which was not present (colored in gray). (B) The presence of violet light with or without blue or green (white) resulted in a shift in the activity ahead of the light flashes and an additional burst after the last light flash, which was not observed in presentation of blue or green light only (black), as seen in smoothed PSTHs of spike times binned over 11 trials. Grey shading indicates the black curve ‘hidden’ behind (= smaller than) the white curve. (C, D) The binned spike times were averaged to get the phase of the spike times relative to the light flashes, where half of the plot is when the light flashes are on (white bar), and the second half is when the light flashes are off (black bar). (C) Phase plots of the responses to all five consecutive light flashes. (D) Phase plots of the responses to the last four consecutive light flashes. The responses to the blue and green LED light flashes without violet light (black arrows) followed the onset of the light flash. When violet light was present in the stimulus (grey arrows), the mean phase of the response was shifted ahead of the light flash. The phase of the responses was significantly different between the non-violet and violet responses. (For all five sequential light flashes across trials: Mann-Whitney U-test; rank sum statistic = 316480; z-statistic = 9.5093; p <0.00001; for the last four light flashes: Mann-Whitney U-test; rank sum statistic = 138539; z-statistic = 8.0390; p <0.00001). A-B: Black bars signify the duration of the light flash stimulus.
Fig. 10
Fig. 10
Habituation responses. (A) The response strength to the first light flash of the optic lobe neurons projecting to the MB calyces decreased considerably to the subsequent light flashes in terms of both excitatory (Ai) and inhibitory (Aii) responses to blue and blue-green light, respectively. (B) In contrast, a lobula neuron not projecting to the mushroom bodies (non-MB neuron) responds similarly to blue (b), green (g), violet (v), and to combinations of the colors across the five subsequent light flashes. (C) Measurement of the habituation responses between the MB input neurons and the non-MB input neurons showed the MB input neurons had significantly higher habituation response values compared to 21 non-MB input neurons (Mann-Whitney U-test; rank sum statistic = −9.2801; z-statistic = 3.5199; p < 0.00001). A, B: Black bars signify the duration of the light flash stimulus.
Fig. 11
Fig. 11
Summary diagram of the sensory inputs to MB calyces examined in this study. (A) Bumblebee brain with different input pathways color coded. Dark gray: Olfactory input from the antennal lobe (two separate pathways; unilateral projections); Green: an input pathways from the medulla (bilateral projections); Magenta: small field lobula MB input neurons (bilateral projections with additional projections into the superior lateral protocerebrum (s l pr; arrow)); Blue: large field lobula MB input neurons (unilateral with additional projections into the superior lateral protocerebrum (arrow)). The large field and small field lobula MB input neuron schematics are presented on either side for visualization purposes only. Abbreviations: mushroom body (B) A closer view of half of a calyx shows general sensory inputs segregate in the calyx with the olfactory inputs in the inner basal ring and lip (dark gray shading) and visual inputs in the collar and outer basal ring (light gray and white shading). In addition to schematic drawings of the blebbed optic lobe visual inputs to the MB calyces, the images of the optic lobe projections into the MB calyces have been combined across preparations and false-colored to indicate the relative depths of the visual MB input within the lip (top), collar (col; middle), and basal ring (br; bottom). Visual input from the medulla occupies the outer collar and basal ring (area: light gray; individual projections: green in the schematic and in the insets). The small field lobula MB inputs generally terminate on the inner collar and basal ring relative to the medulla inputs (area: white; individual projections: magenta). The large field lobula MB inputs generally terminate on the inner-most collar and basal ring, and outer lip (area: white; individual projections: blue). Additional abbreviations: antennal lobe, al; medulla, me; mushroom body lateral calyx, mb-lc; mushroom body medial calyx, mb-mc; mushroom body vertical lobe, mbvl; lobula, lo.

Similar articles

Cited by

References

    1. Backhaus W, Werner A, Menzel R. Color vision in honeybees: metric, dimensions, constancy, and ecological aspects. In: Menzel R, Mercer A, editors. Neurobiology and Behavior of Honeybees. Springer-Verlag; New York: 1987. pp. 172–190.
    1. Baumann F, Hadjilazaro B. A depolarizing aftereffect of intense light in the drone visual system. Vision Research. 1972;12:17–31. - PubMed
    1. Cassenaer S, Laurent G. Hebbian STDP in mushroom bodies facilitates the synchronous flow of olfactory information in locusts. Nature. 2007;448:709–713. - PubMed
    1. Douglass JK, Strausfeld NJ. Visual Motion-Detection Circuits in Flies: Parallel Direction- and Non-Direction-Sensitive Pathways between the Medulla and Lobula Plate. Journal of Neuroscience. 1996;16:4551–4562. - PMC - PubMed
    1. Ehmer B, Gronenberg W. Segregation of visual input to the mushroom bodies in the honeybee (Apis mellifera) Journal of Comparative Neurology. 2002;451:362–373. - PubMed

Publication types

LinkOut - more resources