ArticlePDF Available

Development and morphology of Class II Kenyon cells in the mushroom bodies of the honey bee,Apis mellifera

Authors:

Abstract and Figures

Class II Kenyon cells, defined by their early birthdate and unique dendritic arborizations, have been observed in the mushroom bodies of evolutionarily divergent insects. In the fruit fly Drosophila melanogaster, Class II (also called clawed) Kenyon cells are well known for their extensive reorganization that occurs during metamorphosis. The present account reports for the first time the occurrence of mushroom body reorganization during metamorphosis in holometabolous insect species outside of the Diptera. In the honey bee, Apis mellifera, Class II Kenyon cells show signs of degeneration and undergo a subtle reshaping of their axons during metamorphosis. Unlike in Drosophila, reorganization of Class II Kenyon cells in the honey bee does not involve the loss of axon branches. In contrast, the mushroom bodies of closely related hymenopteran species, the polistine wasps, undergo a much more dramatic restructuring near the end of metamorphosis. Immunohistochemistry, dextran fills, and Golgi impregnations illuminate the heterogeneous nature of Class II Kenyon cells in the developing and adult honey bee brain, with subpopulations differing in the location of dendritic arbors within the calyx, and branching pattern in the lobes. Furthermore, polyclonal antibodies against the catalytic subunit of Drosophila protein kinase A (anti-DC0) label an unusual and previously undescribed trajectory for these neurons. The observed variations in morphology indicate that subpopulations of Class II Kenyon cells in the honey bee can likely be further defined by significant differences in their specific connections and functions within the mushroom bodies.
Morphology of Class II Kenyon cells in the adult honey bee. A–D: Frontal sections of Golgi impregnations. A: Medial-vertical branch point showing unbranched axons entering the vertical lobe (arrowheads) accompanied by other Class II axons that provide short medial branches (arrow). B,C: Two consecutive sections of the vertical lobe (B) and medial-vertical branch point (C) showing short medial branches of Class II Kenyon cell axons (arrow). D: Characteristic “clawed” dendrites of Class II Kenyon cells. E–G: Selected sagittal sections showing the medial-vertical branch point (E) and, in F and G, two progressively more distal levels through the medial lobe (M). E: The medial lobe's γ layer (arrow) is wedged between the medial and vertical (V) lobes and is delineated from Class I components of the medial lobe by outgoing axons of efferent neurons, here seen as a layer of large translucent profiles (arrowheads). γ, vertical γ division. F: Further distal, the medial lobe's γ division (arrow) is small but still discernable. G: In the distal medial lobe, at the level of the central complex (CC), the medial γ layer is absent (arrow). H–J: Selected serial frontal sections of a group of Class II Kenyon cells filled with dextran conjugated to Texas Red. H: Dye injection site (arrow) providing backfills into Class II Kenyon cells within the γ layer (γ, bracketed) of the vertical lobe (V). I: Vertically oriented axons (arrowhead) and their short medially directed tributaries (arrow) at the medial/vertical branch point of the two lobes (compare with branch point revealed by Golgi impregnation in 5C). J: Clawed dendrites of Class II Kenyon cells belonging to back filled axons from the vertical lobe. All three calycal subdivisions, lip (Li), collar (Co) and basal ring (BR), are represented. Scales bars = 50 μm in A–C; 10 μm in D; 100 μm in E–G; 20 μm in H–J = 100μm.
… 
Development of the mushroom body lobes in the honey bee larva and pupa. Frontal sections unless noted. A–C: Serial sections of DiI-filled tracts in the pedunculus (Ped) and lobes (M, V) of the early (feeding) fifth instar larva reveal branching of larval Class II Kenyon cells. A: Pedunculus (Ped) and vertical lobe (V). B: Medial/vertical lobe branch point (asterisk), with a tract of possibly intrinsic origin (double arrows) emerging from the branch point and describing a partial circle posterior to the pedunculus. Another tract of extrinsic origin is filled laterally (arrow). C: Pedunculus and medial lobe (M), with putative intrinsic tract (double arrows) extending dorsal to the medial lobe. D–E: Anti-DC0 immunostaining of the larval and prepupal pedunculus (Ped) and medial lobe (M) reveal medial branching of Class II Kenyon cells. D: A subset of Class II axons in the spinning fifth instar larva show high affinity for anti-DC0 and reveal medial branches (arrows). The tract encircling the pedunculus, which is seen in the larval DiI fills is also visible (double arrows), its affinity for anti-DC0 indicates its possible intrinsic neuron origin. E: At the prepupal stage, medial branches (arrows) of Class II axons are heavily immunostained. F: Sagittal section of the prepupal pedunculus and vertical lobe. The vertical lobe tip is bent dorsally and has a distinct notch at the ventral surface (arrow). Blebbed axons and debris are seen in the vicinity of axon tips. Inset, F: DiI labeled sagittal section of the medial lobe showing degeneration debris among medial lobe axons. G: Anti-DC0 staining of the tip of the prepupal vertical lobe. Three (numbered) of the four component axon bundles are visible. Vertically projecting axons of Class II Kenyon cells having a high affinity to anti-DC0 comprise the most ventral layer of the vertical lobe (bracketed). H: The tip of the prepupal vertical lobe, Cason's staining, also showing axon bundles in the upturned lobe tip. I: Anti-DC0 staining showing a sagittal section of the pedunculus and vertical lobe in the day 1 pupa. The dorsal curvature of the vertical lobe and the ventral notch (arrow) are less pronounced than in the prepupa. Class II Kenyon cell crossing fibers are visible at the posterior margin of the pedunculus (arrowheads). J: Anti-taurine-stained day 3 pupa. Little specific staining is visible, but the overall structure of the lobe can be discerned. Most axons in the vertical lobe show little or no dorsal extension and the ventral notch (old position indicated by arrow) has at this stage disappeared. Scale bars = 20 μm in A–G; 10 μm in H; 50 μm in I–J.
… 
Axon ingrowth during mushroom body development. A–F: Anti-DC0 (purple) and phalloidin (green) double labeling. A–C are from 3-day-old pupae; D–F compare larva, mid-, and late pupae. A: Heavy phalloidin labeling in the lip (Li) and collar (Co) of the calyces indicates extensive outgrowth of dendritic processes. Axons from newborn Kenyon cells converge from the protocalyces into the necks of the pedunculus (Ped), where they define the ingrowth tracts (arrows). Populations of Class II Kenyon cells continue to show a relatively high affinity for anti-DC0 (arrowheads). B: Ingrowth tracts from each calyx through the necks of the pedunculus (Ped) remain separated then spread as two broad laminar tracts (arrows) in the medial lobe (M). CC, central complex. C: A single ingrowth lamina demarcates the dorsal margin of the vertical lobe (arrows); these axons will at later stages be “displaced” deeper by axons belonging to layers of the lobe representing the basal ring. Strong anti-DC0 affinity is seen in Class II Kenyon cell axons (arrowheads) at the vertical/medial lobe branch point, and in the short medial γ layer (Mγ). D–F: Anti-DC0 and phalloidin double labeling throughout larval and pupal development. D: Spinning fifth instar larva showing anti-DC0 labeling of earlier-born Class II Kenyon cells (arrowheads) and two diffuse phalloidin-labeled ingrowth tracts (arrows). No calycal neuropil has been formed at this stage. E: Ingrowth tracts (arrows) appear more distinct and cohesive in the late prepupa. Calycal (Ca) primordia are visible atop each neck of the pedunculus. F: The ingrowth tracts in the day 7 pupa have dwindled to thin threads (arrows) due to the abatement of MBNB activity 2 days previously. Robust anti-DC0 staining is seen in the majority of Kenyon cell axons. The highest affinities are still seen in Class II Kenyon cells, including those providing cross fibers projections (indicated by arrowheads). G–I: Collateral outgrowths from medial lobe ingrowth tracts. G: Phalloidin staining of a 3-day-old pupa reveals dense arrays of short processes (arrowheads) emerging at an ∼45° angle from the ingrowth tract (arrow). H: DiI labeling of the ingrowth tract (arrow) in a day 1 pupal mushroom body shows an identical arrangement of collaterals (arrowheads). I: Enlarged view of DiI-labeled ingrowth axons and collaterals. Scale bars = 100 μm in A–C, E–F; 20 μm in in D,I; 50 μm in G,H.
… 
Content may be subject to copyright.
Development and Morphology of Class II
Kenyon Cells in the Mushroom Bodies of
the Honey Bee, Apis mellifera
SARAH M. FARRIS,
1
*ANDREW I. ABRAMS,
2
AND NICHOLAS J. STRAUSFELD
2
1
Department of Biology, West Virginia University, Morgantown, West Virginia 26506
2
Division of Neurobiology, Arizona Research Laboratories, University of Arizona,
Tucson, Arizona 85721
ABSTRACT
Class II Kenyon cells, defined by their early birthdate and unique dendritic arboriza-
tions, have been observed in the mushroom bodies of evolutionarily divergent insects. In the
fruit fly Drosophila melanogaster, Class II (also called clawed) Kenyon cells are well known
for their extensive reorganization that occurs during metamorphosis. The present account
reports for the first time the occurrence of mushroom body reorganization during metamor-
phosis in holometabolous insect species outside of the Diptera. In the honey bee, Apis
mellifera, Class II Kenyon cells show signs of degeneration and undergo a subtle reshaping
of their axons during metamorphosis. Unlike in Drosophila, reorganization of Class II
Kenyon cells in the honey bee does not involve the loss of axon branches. In contrast, the
mushroom bodies of closely related hymenopteran species, the polistine wasps, undergo a
much more dramatic restructuring near the end of metamorphosis. Immunohistochemistry,
dextran fills, and Golgi impregnations illuminate the heterogeneous nature of Class II
Kenyon cells in the developing and adult honey bee brain, with subpopulations differing in
the location of dendritic arbors within the calyx, and branching pattern in the lobes. Fur-
thermore, polyclonal antibodies against the catalytic subunit of Drosophila protein kinase A
(anti-DC0) label an unusual and previously undescribed trajectory for these neurons. The
observed variations in morphology indicate that subpopulations of Class II Kenyon cells in
the honey bee can likely be further defined by significant differences in their specific connec-
tions and functions within the mushroom bodies. J. Comp. Neurol. 474:325–339, 2004.
©2004 Wiley-Liss, Inc.
Indexing terms: differentiation; insect; metamorphosis; neuron remodeling; plasticity
Paired neuropils comprising many thousands of approx-
imately parallel axon-like processes are found in the
brains of all insects except those belonging to the basal
order Archaeognatha (Strausfeld et al., 1998). These neu-
ropils, known as the mushroom bodies, have attracted
much attention due to studies that focus on their possible
roles in olfactory learning and memory (Heisenberg et al.,
1985; deBelle and Heisenberg, 1994; Zars et al., 2000;
Cano Lozano et al., 2001; Pascual and Pre´at, 2001), sen-
sory integration (Vowles, 1964; Erber et al., 1980; Schild-
berger, 1984; Laurent and Naraghi, 1994; Li and Straus-
feld, 1997, 1999), and place memory and motor control
(Mizunami et al., 1998a,b).
Mushroom bodies of different taxa share anatomical
and developmental commonalties. Anatomical studies
have demonstrated a suite of conserved morphological
traits, particularly the parallel division of the mushroom
body’s deeper neuropils into discrete layers or lobes, each
defined according to the morphological categories of in-
trinsic cells supplying them (Strausfeld, 2002; Farris and
Sinakevich, 2003; Strausfeld et al., 2003). Likewise, the
recognition that common developmental mechanisms
might generate these structures has begun to be eluci-
dated using insects as widely divergent as cockroaches,
Grant sponsor: National Institutes of Health; Grant number:
P01NS28495.
*Correspondence to: Sarah M. Farris, West Virginia University, Depart-
ment of Biology, Life Sciences Building, 53 Campus Drive, Morgantown,
WV 26506. E-mail: Sarah.Farris@mail.wvu.edu
Received 12 November 2003; Revised 20 January 2004; Accepted 28
January 2004
DOI 10.1002/cne.20146
Published online in Wiley InterScience (www.interscience.wiley.com).
THE JOURNAL OF COMPARATIVE NEUROLOGY 474:325–339 (2004)
©2004 WILEY-LISS, INC.
crickets, honey bees, and fruit flies (Farris and Si-
nakevitch, 2003).
The major constituents of the mushroom bodies are its
intrinsic neurons, known as Kenyon cells (Strausfeld,
1976). Each Kenyon cell consists of a minute basophilic
cell body, one of many thousands clustered posteriorly and
laterally in the protocerebrum. Each perikaryon provides
a slender neurite that then enlarges its diameter to form
a long process, here referred to as an “axon,” although
there is evidence that these cellular structures better re-
semble amacrine processes than axons in the strictest
sense. Electron microscopical observations show that the
Kenyon cell “axons” are equipped with both pre- and
postsynaptic specializations down their length, forming
synapses not only with extrinsic neurons but with one
another (Strausfeld and Li, 1999).
Kenyon cell axons extend anteroventrally, typically bi-
furcating to give rise to a vertical and medial branch.
Clusters of several hundreds or even thousands of simi-
larly branched axons provide the intrinsic constituents of
the vertical and medial lobes of the mushroom bodies. In
neopteran insects, Kenyon cells also give rise to an array
of dendrites in the protocerebrum that lie immediately
beneath their cell bodies of origin; these dendrites form
distinctive neuropils called the calyces.
Kenyon cells are generated during mushroom body de-
velopment by dedicated mushroom body neuroblasts (MB-
NBs) that are typically located at the center of each calyx
(Panov, 1957; Ito et al., 1997; Farris et al., 2001; Malaterre
et al., 2002). As neurogenesis proceeds, early-born Kenyon
cell bodies are gradually pushed outwards to more periph-
eral locations, while late-born cells remain closer to the
central MBNB or MBNB cluster. Axons of newborn Ke-
nyon cells continually enter the lobes from the calyces via
a discrete posteriorly located ingrowth tract. In this man-
ner, maturing axons are displaced relatively more anteri-
orly or peripherally by new ingrowing axons extending
into the lobes (Farris and Strausfeld, 2001; Kurusu et al.,
2002; Malaterre et al., 2002). Due to these developmental
mechanisms, the adult mushroom bodies are character-
ized by an age-dependent organization of Kenyon cell bod-
ies within the calyces and an age-dependent organization
of Kenyon cell axons within the lobes.
One morphological type of Kenyon cell has been reliably
identified in the mushroom bodies across a variety of
insect species. This is the Class II Kenyon cell (also
termed clawed Kenyon cell), which can be identified by its
characteristic “clawed” dendritic specializations (Straus-
feld, 1976, 2002; Strausfeld and Li, 1999; Strausfeld et al.,
2003). Class II Kenyon cell bodies are generally located at
the periphery of the Kenyon cell body cluster, and may
even reside outside but adjacent to the calyx neuropil. The
placement of cell bodies distant from the center of the
calyx reflects the fact that Class II Kenyon cells are al-
ways among the first intrinsic neurons produced during
mushroom body development (Farris et al., 1999; Malat-
erre et al., 2002). One consequence of this early specifica-
tion and cellular differentiation is that most of the axons
of Class II Kenyon cells are segregated early in develop-
ment into a distinct anterior layer of the mushroom bodies
that is termed the gamma () lobe or layer (Lee et al.,
1999; Farris and Strausfeld, 2001).
Class II Kenyon cells are of particular functional inter-
est because, as a whole, they represent the entire calycal
neuropil. The importance of this distinction is best illus-
trated in the honey bee and other hymenopteran insects
(Strausfeld, 2002). Sensory interneurons providing axon
collaterals to the calyces segregate according to type
(Gronenberg, 2001; Strausfeld, 2002; Schro¨ter and Men-
zel, 2003). Thus, the axons of Class II Kenyon cells in the
lobe or layer represent all modalities entering the caly-
ces. In contrast, the dendrites of other types of later-
developing (Class I) Kenyon cells are segregated into spe-
cific calycal subdivisions, and thus are also segregated in
their interactions with sensory afferents. As a result, a
lobe or layer containing the axons of one type of Class I
Kenyon cell would represent one modality, whereas an-
other lobe or layer would represent another modality
(Strausfeld, 2002).
In the basal hemimetabolous insect Periplaneta ameri-
cana, the axons of Class II Kenyon cells bifurcate to pro-
duce a single branch each into the medial and vertical
lobes (Strausfeld and Li, 1999). In the holometabolous
insect Drosophila melanogaster, however, the majority of
Class II Kenyon cells in the adult mushroom bodies each
provide an unbranched axon in the medial direction only,
forming an unbranched lobe (Lee et al., 1999; Strausfeld
et al., 2003).
Neuroanatomical descriptions of Class II Kenyon cell
axon morphology in the honey bee are conflicting. An early
study of reduced silver-stained and Golgi-impregnated
honey bee brains by Mobbs (1982) described all Kenyon
cells as having a bifurcated axon with medial and vertical
tributaries of apparently similar length. Using cobalt chlo-
ride fills and Golgi impregnations, Rybak and Menzel
(1993) also reported Class II Kenyon cell (K5) axons with
branches in the medial and vertical lobes. In contrast, a
recent Golgi study by Strausfeld (2002) presented evi-
dence suggesting that honey bee Class II Kenyon cell
axons are unbranched, providing a single long process
through the length of the vertical lobe, where they com-
prise a discrete layer.
As previously mentioned, unbranched Class II Kenyon
cell axons in the adult Drosophila mushroom bodies sup-
ply the medially directed lobe in the adult fly. In the
larva, however, the same Class II Kenyon cells provide
branched axons that supply a larval vertical and medial
lobe. During metamorphosis Class II Kenyon cell axons
degenerate and subsequently regrow in the unbranched,
medially projecting trajectory characteristic of the adult
fly (Technau and Heisenberg, 1982; Lee et al., 1999; Watts
et al., 2003). Aside from other brachyceran Diptera (Gun-
dersen and Larsen, 1978), such reorganization of the
mushroom bodies has not been described in any other
insect. The presence of Drosophila-like unbranched Class
II Kenyon cells in the moth Sphinx ligustri (Pearson,
1971) and in the honey bee (Strausfeld, 2002), however,
may indicate that axon degeneration and regrowth during
metamorphosis may be widespread among the Holo-
metabola.
This study addresses the question of whether Class II
Kenyon cell axons in the honey bee mushroom bodies
undergo reorganization during metamorphosis. We show
compelling evidence for such a reorganization in the honey
bee, and in addition describe extensive developmental re-
structuring of the mushroom bodies in a closely related
taxon, the polistine wasps. Applying specific markers for
Kenyon cells in different stages of differentiation to the
developing honey bee brain provides further evidence that
there are universal mechanisms of mushroom body devel-
326 S.M. FARRIS ET AL.
opment across insects. Finally, a palette of histological
and immunochemical techniques reveals at least four
Class II Kenyon cell subpopulations based on axon projec-
tion patterns in the pedunculus and lobes.
MATERIALS AND METHODS
Insects
Frames of honeycomb containing larval and pupal Eu-
ropean honey bee workers (Apis mellifera) were obtained
from colonies maintained at the USDA-ARS Carl Hayden
Bee Research Center in Tucson, Arizona. Feeding-age lar-
vae (early fifth instar and younger) were removed from the
comb immediately for histology, while late fifth instar and
older bees were collected up to 3 days after the frame was
removed from the colony. When larvae and pupae were not
being collected, the frame was stored in a walk-in incuba-
tor at 30°C. Larval age was determined by measuring
head capsule width (Bertholf, 1925), while that of pupae
was roughly determined by the degree of pigmentation of
the compound eye and degree of sclerotization of the cuti-
cle in older pupae (Michelette and Soares, 1993). One-day-
old adult bees were collected as they emerged from the
brood comb, and adult foragers of unknown age were
collected as they visited flowers.
Nests containing larvae, pupae, and adults of the polis-
tine wasp Polistes apachiensis were obtained from Hatari
Invertebrates (Portal, AZ). Nests were kept in the same
walk-in incubator conditions as described above. Pupae
were collected at intervals from the event of the larval-
pupal molt throughout the 10-day pupal stage. Both
honey bees and wasps were anesthetized with CO
2
or cold
prior to dissection.
Reduced silver, ethyl gallate,
and Cason’s staining
Reduced silver (Bodian) and Cason’s staining was per-
formed on paraffin-embedded sections as described by
Strausfeld and Li (1999) and Farris et al. (1999), respec-
tively. In preparation for ethyl gallate staining, honey bee
brains were dissected in 0.16M cacodylate buffer (pH 7.2)
with 1% sucrose and fixed in the same solution containing
2.5% glutaraldehyde for 4 hours at room temperature.
After fixation, brains were stored in the cacodylate-
sucrose solution at 4°C until staining. Brains were cooled
and placed in cold 0.5% osmium tetroxide in distilled
water with constant but gentle agitation. After 1 hour the
solution was allowed to reach room temperature and the
brains similarly osmicated for an additional 60 minutes.
Then the brains were washed for 1 hour in several
changes of distilled water before cooling to 4°C. Brains
were transferred to a saturated aqueous solution of cold
ethyl gallate in the dark, also with gentle agitation. After
1 hour the solution was allowed to reach room tempera-
ture and after a further hour brains were washed in dis-
tilled water, dehydrated, embedded in Durcupan plastic
(Electron Microscopy Sciences, Fort Washington, PA), po-
lymerized, and serially sectioned at 12–25 m. The results
of this method are shown in Figs. 1A and 5E–G.
Golgi impregnation
In an attempt to stain as many Class II Kenyon cell
subpopulations as possible, two different Golgi impregna-
tion methods were used. The first was a high-osmium
modification of the combined Colonnier/rapid Golgi proto-
col as described by Strausfeld (2002). The second protocol
was a shortened version of the above. Brains were dis-
sected from the head capsules in potassium dichromate
and sucrose solution as previously described. Fixation
time was reduced to 4 days, followed by a thorough wash
in 25% potassium dichromate and then incubation for 3
days in 0.01% OsO
4
and 1% chloral hydrate. Brains were
then removed from the chromating solution and washed
several times in 0.75% AgNO
3
in borax-borate buffer (pH
7.0 –7.3) and then kept in buffered silver nitrate for a
further 3 days. All incubations were carried out at 4°C in
the dark.
DC0 and taurine immunostaining and
phalloidin staining
The anti-DC0 antibody was generated against the cat-
alytic subunit of protein kinase A of Drosophila melano-
gaster. This protein is part of the cAMP-mediated olfac-
tory learning and memory pathway and has previously
been shown to specifically label all Kenyon cell popula-
tions in the adult insect mushroom bodies (Skoulakis et
al., 1993; Farris and Strausfeld, 2003; S.M. Farris, un-
publ. obs.). Antibodies against the amino acid cotransmit-
ter taurine illuminate Kenyon cell subpopulations in de-
veloping and adult mushroom bodies (Farris and
Strausfeld, 2001; Sinakevitch et al., 2001; Strausfeld et
al., 2003). Phalloidin is a fungal-derived toxin that specif-
ically binds to filamentous (f) actin, which is enriched in
the extending processes of newborn Kenyon cells in the
mushroom bodies (Kurusu et al., 2002; Farris and Si-
nakevitch, 2003).
Head capsules of larval and pupal honey bees were
removed from the body in bee saline and fixed in 4%
paraformaldehyde in PBS (phosphate-buffered saline
from tablets, pH 7.4; Sigma Chemical Co., St. Louis, MO).
Adult honey bee brains were removed from the head cap-
sule immediately and processed in the same manner.
Whole heads and brains were stored in fixative until pro-
cessing for immunostaining, at which time they were
washed in PBS and embedded in 7% agarose. Agarose
blocks were then sectioned on a vibratome at 50 –70 m
and the sections washed in PBS containing 0.1% Triton
X-100 (PBST). Sections were blocked for at least 1 hour in
PBST containing 10% normal goat serum (NGS) and in-
cubated overnight in polyclonal anti-DC0 primary anti-
body (a generous gift from Dr. Daniel Kalderon) at a
1:1,000 concentration. Alternatively, sections were incu-
bated in polyclonal anti-taurine antibody (Chemicon, Te-
mecula, CA) at a 1:300 concentration. The following day,
sections were washed in PBST and incubated in Texas
Red-conjugated goat antirabbit secondary antibody (Mo-
lecular Probes, Eugene, OR) at a 1:500 concentration over-
night. For DC0/phalloidin double-staining, Oregon green-
conjugated phalloidin (Molecular Probes) was added at a
1:500 concentration at this time as well. The next day
sections were washed in PBST and cleared in 60% glycerol
for 1 hour, followed by 80% glycerol for 1 hour. After
mounting in 80% glycerol, sections were viewed using a
Zeiss confocal microscope.
DiI labeling
The lipid-soluble cellular membrane tracer DiI (1, 1-
dioctadecyl-3,3,3,3-tetramethylindocarbocyanine perchlor-
ate; Molecular Probes) was used to label ensembles of Ke-
327LOBE DEVELOPMENT IN HONEY BEE MUSHROOM BODIES
nyon cells in the developing mushroom bodies. Larval and
pupal head capsules were removed, fixed, and stored as
described for immunostaining. Brains were dissected from
the head capsules in PBS prior to application of DiI. DiI
crystals were dissolved in methanol and a droplet of this
solution spread on a glass slide, leaving a thin film of ex-
tremely fine crystals. The tip of a glass electrode was rolled
gently over the film until coated with crystals. The coated
electrode was used to deposit DiI into groups of Kenyon cell
bodies or their axons. After the placement of DiI crystals, the
tissue was stored in PBS at 4°C for 1 week. After this time
brains were embedded in agarose, sectioned, and cleared in
glycerol as described above for anti-DC0 immunostaining.
Texas Red-dextran fills
The neuronal tracer dextran was employed to produce
precisely targeted fills of small ensembles of Kenyon cells
in the adult mushroom bodies. Bees were anesthetized by
cooling or with CO
2
gas and secured in a vertical position
in small plastic tubes. On occasion a sucrose/trehalose
solution was fed to bees prior to the injection procedure.
The head capsule was opened to reveal the surface of the
brain and a drop of bee saline (Huang et al., 1991) added
to prevent desiccation. A small part of the neuralemma
was peeled back to expose the anterior surface of the
vertical lobe. A 5% solution of Texas Red (TxR)-labeled
3000MW dextran (Molecular Probes) was loaded into a
1-mm borosilicate electrode that was then placed in a
micromanipulator and connected via 1-mm elastic tubing
toa5ccsyringe filled with sterile water. The electrode tip
was placed into the ventral () division of the vertical lobe
and the dye injected by a hand-operated compression of
the plunger of the syringe. After injection, the head cap-
sule was closed and the dye allowed to diffuse for 45
minutes to 1.5 hours. After the injection procedure whole
heads were fixed in 3.7% formaldehyde in 0.16 M cacody-
late buffer at 4°C overnight. The following day brains were
dissected from the head capsules in cacodylate buffer,
washed, and dehydrated through a series of alcohols. After
clearing in acetone, brains were embedded in Spurr’s low
viscosity plastic (Electron Microscopy Sciences) and sec-
tioned at 17–20 m using a sliding microtome. Sections
were mounted in Fluoromount (Crescent Chemical Co.,
Islandia, NY) and viewed using a Zeiss confocal micro-
scope.
Image processing
Images collected on the confocal and light microscopes
were processed as needed for brightness, contrast, and
sharpness using Adobe PhotoShop 7.0 (Adobe Systems,
San Jose, CA).
RESULTS
Anatomy of the developing and adult honey
bee mushroom bodies
Much of the anatomy of the honey bee mushroom bod-
ies, in terms of their intrinsic neuron morphology as well
as some afferent and efferent connections, has been pre-
viously described (Mobbs, 1982, 1984; Rybak and Menzel,
1993; Strausfeld, 2002). Briefly, the honey bee mushroom
bodies are paired neuropils in the dorsal protocerebrum
made up of more than 200,000 densely packed intrinsic
neurons, the Kenyon cells (Fig. 1A,B). Kenyon cell somata
reside in and around the dorsally situated calyces. Each
mushroom body possesses two calyces made up of Kenyon
cell dendrites that are divided into three major regions
called the lip, collar, and basal ring based on the origins of
their major afferent supply. These gross subdivisions can
be further parsed into smaller zones according to subsets
of afferent supply, intrinsic neuron morphology, and im-
munohistochemical affinities (Strausfeld et al., 2000;
Strausfeld, 2002; Ehmer and Gronenberg, 2002; Schro¨ter
and Menzel, 2003). Ventral to the dendritic branch point,
Kenyon cell axons project into the pedunculus, where they
appear to form both pre- and postsynaptic connections
with extrinsic neurons. With the exception of some Class
II (clawed) Kenyon cells (Strausfeld, 2002; present ac-
count), Kenyon cell axons bifurcate in the pedunculus to
form with their branches the vertical and medial lobes.
The cross-sectional organization of the lobes is into dis-
crete layers, each of which is characterized by its intrinsic
neuron composition as well as the types of afferent end-
ings it receives and efferent dendritic trees it contains
(Mobbs, 1982; Rybak and Menzel, 1993; Strausfeld, 2002).
The most anterior part of the vertical lobe (the division,
which is ventral according to the neuraxis) is composed of
the axons of Class II Kenyon cells. Progressively more
posterior (i.e., dorsal) layers contain the axons of Class I
Kenyon cells, the dendrites of which invade the lip, collar,
and basal ring, respectively (Strausfeld, 2002). The medial
lobe is also composed of layers corresponding to these
three main calycal regions, but has been reported to con-
tain no layer, presumably due to the lack of medial axons
from Class II Kenyon cells (Strausfeld, 2002).
In all insects studied thus far, Kenyon cells are gener-
ated by dedicated progenitor cells (mushroom body neuro-
blasts, MBNBs) that characteristically divide continu-
ously throughout most or all of preadult development
(reviewed in Farris and Sinakevitch, 2003). In the honey
bee, four MBNB clusters, each corresponding to a single
calyx in the adult insect (Fig. 1C), are found in the dorsal
protocerebrum of the newly-hatched larva (Panov, 1957).
The MBNBs at first appear to divide only symmetrically,
thus increasing their numbers nearly 10-fold by the end of
the prepupal stage (Panov, 1957; Farris et al., 1999). The
production of Kenyon cells does not commence until the
mid-larval stage, with the accumulation of neurons occur-
ring with particular rapidity from the mid-fifth instar to
the mid-pupal stage. Newborn Kenyon cells extend axons
before producing dendrites so that while lobe neuropil can
be identified in the fourth instar larva, the calyx itself does
not begin to form until the prepupal stage. Calyx neuropils
grow rapidly during the prepupal and early pupal stage,
with significant expansion and neuronal outgrowth con-
tinuing even in the adult (Withers et al., 1993; Durst et
al., 1994; Farris et al., 2001). Neurogenesis ends at day 5
of the pupal stage, at which time the MBNBs undergo
programmed cell death (Ganeshina et al., 2000). Due to
the central location of the MBNBs within each calyx,
Kenyon cell bodies are arranged concentrically in a
birthdate-dependent manner, with the earliest-born cells
residing outside of the adult calyx, flanking its outer sur-
face even as far as the origin of the pedunculus from the
base of the calyx. The latest-born cells, which supply the
calyx’s basal ring neuropil, take the place of the MBNBs in
the center of each calyx (Panov, 1957; Farris et al., 1999).
A similar age-based organization of Kenyon cell bodies has
been observed in all insect species surveyed to date (Farris
328 S.M. FARRIS ET AL.
and Strausfeld, 2001; Kurusu et al., 2002; Malaterre et al.,
2002). The major events in honey bee mushroom body
development are summarized in Figure 2.
Axon outgrowth and lobe development in
honey bee mushroom bodies
Identifiable lobe neuropil is first resolved at approxi-
mately the fourth larval instar (Farris et al., 1999). DiI
crystals placed in the region of the MBNBs of the feeding
(early) fifth instar larva revealed a thin but recognizable
pedunculus, medial, and vertical lobe (Fig. 3A–C). The
axon bundles of Kenyon cells produced by each of the two
MBNB clusters remained widely separated dorsally, but
appeared to fuse in the region of the lobes. DiI fills at such
early stages of mushroom body development sometimes
revealed additional tracts that comprised bundled neu-
rites in the vicinity of the mushroom bodies, including a
particularly prominent tract that projected from the
medial/vertical lobe junction to a position posterior to the
pedunculus (Fig. 3B,C). Although the nature of the DiI
crystal application technique utilized in this study did not
allow for the determination of the exact origination of
these axons, a similarly located tract was apparent in
anti-DC0 immunostained preparations of larvae of similar
age (Fig. 3D). The localization of high anti-DC0 affinity in
the developing honey bee protocerebrum, which appeared
to be restricted to Class I and II Kenyon cells (see below),
might indicate that this larval tract is composed of mush-
room body intrinsic neurons. Given the early origins dur-
ing mushroom body development of the constituent cells,
and their unusual projection pattern around the peduncu-
lus, these cells could possibly be equivalent to the Class III
cells. Class III Kenyon cells provide the distinctive lobelet
that has been identified in the juvenile and adult mush-
room bodies of certain cockroach and termite species, all
hemimetabolous species (Farris and Strausfeld, 2003).
Class III Kenyon cells in these insects produce axon
branches that encircle but are not integrated into the
vertical lobe, reminiscent of the organization of the early
tract in the honey bee brain seen in Figure 3A–D. Addi-
Fig. 1. Mushroom body anatomy in the adult and developing
honey bee worker brain. A: Ethyl gallate-stained sagittal section of
the adult brain, with Golgi-impregnated neurons overlaid at the same
magnification. White arrow indicates the medial/vertical branch point
of Kenyon cell axons. Dorsal (D), ventral (V), anterior (A), and poste-
rior (P) axes are indicated. B: Reduced silver-stained frontal section of
adult brain, showing mushroom bodies in relation to other structures.
Dorsal, ventral, lateral (L), and medial (M) axes are indicated for one
hemisphere of the brain (midline indicated by white line). C: Cason’s-
stained frontal section of the neuroblast clusters (MBNBs; nb in
figure) and developing calyces (Ca) in the prepupa. Only one mush-
room body shown with its two MBNB clusters representing the two
calyces. Kenyon cell bodies (K cbs) form a cloud encircling each neu-
roblast cluster. L Ca, lateral calyx; M Ca, medial calyx; Li, lip; Co,
collar, BR, basal ring; Ped, pedunculus; M, medial lobe; V, vertical
lobe, , gamma lobe; gl, glia; Ant Lo, antennal lobe, CC, central
complex; oc, ocelli, Op Lo, optic lobe. Scale bars 100 minA;200
minB;20minC.
329LOBE DEVELOPMENT IN HONEY BEE MUSHROOM BODIES
tionally, this tract in the honey bee may be present only
transiently, because both anti-DC0 immunostaining and
DiI fills failed to reveal it in larvae older than the mid-fifth
instar.
Anti-DC0 antibody served as an excellent selective
marker for Kenyon cell processes in the developing honey
bee protocerebrum. As previously reported for various spe-
cies of the hemimetabolous order Dictyoptera (Farris and
Strausfeld, 2003), all Kenyon cell populations in the honey
bee mushroom bodies, with the exception of newly born
cells, showed strong anti-DC0-like immunoreactivity. In
the developing mushroom bodies, particularly high affin-
ity for the antibody was observed for some populations of
Class II Kenyon cells (Fig. 3D,E), as identified by their
projection patterns into the most ventral layer of the ver-
tical lobes (Strausfeld, 2002). As development progressed,
the disparity in strength of anti-DC0 affinity between
Class II Kenyon cells and other Kenyon cell populations
lessened.
Whereas Class II Kenyon cells are the first-born popu-
lation of intrinsic neurons to be produced by the honey bee
MBNBs (Farris et al., 1999; Strausfeld 2002), the first
production of spiny (Class I) Kenyon cells occurs during
the prepupal (late fifth instar) stage. This transition is
obviously denoted by the appearance of the much larger
Class I Kenyon cell bodies (Panov, 1957; Farris et al.,
1999). Prior to this time, the mushroom bodies are made
up entirely of Class II Kenyon cells and, possibly, a short-
lived set of Class III cells (see above). Significantly, how-
ever, DiI fills and anti-DC0 staining of early fifth instar
larvae and of prepupae revealed both a medial and a
vertical lobe resulting from branching Class II Kenyon cell
axons (Fig. 3A–E). This observation indicates that honey
bee Class II Kenyon cells have branched axons early in
their development.
This situation is reminiscent of that described for the
fruit fly D. melanogaster, where the larval mushroom body
comprises Class II Kenyon cells that branch to form a
medial and vertical lobe. Massive reorganization of Class
II (lobe) Kenyon cell axons occurs during metamorphosis
(Technau and Heisenberg, 1982; Lee et al., 1999), result-
ing in most Class II Kenyon cell axons losing their vertical
component. If the mushroom bodies of the fifth instar
honey bee larva comprise two lobes consisting of Class II
Kenyon cell axons, but the adult mushroom body has a
prominent layer in the vertical lobe only, could this
indicate that a similar reorganization is occurring in this
insect?
A subtle reshaping of axons is most evident in the ver-
tical lobe of the honey bee mushroom bodies. In the larval
Fig. 2. Summary diagram schematizing the sequence of observed
events described in this account and elsewhere (Farris et al., 1999).
Postembryonic mushroom body development begins with the initial
presence of four clusters of MBNBs in the newly-hatched first instar
larva (left). Neuroblast proliferation continues until the onset of the
pupal stage (pupal day 1). Class II (clawed) Kenyon cells are gener-
ated from approximately the third instar through to the prepupal
stage, when the neuroblasts switch to the production of Class I Ke-
nyon cells. The calycal neuropil also begins to differentiate at the
prepupal stage. The lobes develop incrementally, with minute but
identifiable medial and vertical lobes first visible at the fourth instar.
Reorganization among Class II Kenyon cells occurs during the prepu-
pal stage. Class II crossing fibers are resolved from the prepupa until
immediately after adult eclosion. A lobed structure satellite to early
developing medial and vertical lobe appears during the early fifth
instar only, and is here suggested to represent the transient appear-
ance of Class III Kenyon cells (transient tract cells).
330 S.M. FARRIS ET AL.
and prepupal mushroom bodies the otherwise anteriorly
projecting vertical lobe axons showed a pronounced dorsal
bend at their tips (Fig. 3F–H). In sagittal sections, a
distinct notch was observed on the ventral surface of the
vertical lobe at the point at which Kenyon cell axons began
to bend dorsally (Fig. 3F). Frontal sections of the upturned
lobe tip revealed four separate fiber bundles making up
the vertical lobe (Fig. 3G,H). These bundles likely repre-
sent cells making up each half of the two calyces. In the
prepupa these axons showed their maximum dorsal exten-
sion, reaching nearly to the level of the calyces. At the
beginning of the pupal stage (pupal day 1, Fig. 3I), this
dorsal extension as well as the ventral notch were less
pronounced, and both characteristics appeared to be ab-
sent from the vertical lobe in the day 3 pupa (Fig. 3J).
DiI fills of prepupal Class II Kenyon cells revealed blebs
and debris around the tips of axons in both the vertical
and medial lobes (Fig. 3F, inset), suggesting that these
axons were undergoing degeneration and reorganization.
A similar phenomenon has been observed associated with
Class II Kenyon cell axons of the fruit fly during mush-
room body reorganization (Watts et al., 2003). Although
these observations indicate that Class II Kenyon cell ax-
ons in the honey bee are undergoing some degree of meta-
morphic reorganization, there was no indication of the
complete degeneration of axon branches, such as has been
documented in the fruit fly.
Oregon green-labeled phalloidin revealed the extending
fiber projections of newborn Kenyon cells into the lobes
and calyces (Kurusu et al., 2002; Farris and Sinakevitch,
2003) (Fig. 4A–F). Throughout mushroom body develop-
ment, axon ingrowth into the lobes occurred via distinct,
posteriorly located “ingrowth” tracts (Farris and Straus-
feld, 2001), equivalent to the mushroom body “core” de-
scribed by Kurusu et al. (2002) from observations of Dro-
sophila. In the honey bee, continuous axon ingrowth into
the lobes during development resulted in the mushroom
body lobes gradually increasing in girth as axons were
added layer-by-layer to the posterior or dorsal surfaces of
the lobes. Again, this incremental addition characterizes
development both in hemimetabolous mushroom bodies
and in the holometabolous fruit fly (Lee et al., 1999; Farris
et al., 2001).
In the developing honey bee mushroom bodies, the neu-
rites of newborn Kenyon cells appeared to stream down
the sides of each MBNB cluster, entering the pedunculus
as two separate tracts (Fig. 4A). The two tracts remained
separate into the medial lobe (Fig. 4B) but formed one
wide tract in the vertical lobe (Fig. 4C), where, as a con-
sequence of its horizontal orientation, the ingrowth tract
was located at the dorsal surface of the lobe. Distinct
ingrowth tracts were observed in early lobe development
in the fifth instar larva (Fig. 4D), although the bundles
Fig. 3. Development of the mushroom body lobes in the honey bee
larva and pupa. Frontal sections unless noted. A–C: Serial sections of
DiI-filled tracts in the pedunculus (Ped) and lobes (M, V) of the early
(feeding) fifth instar larva reveal branching of larval Class II Kenyon
cells. A: Pedunculus (Ped) and vertical lobe (V). B: Medial/vertical
lobe branch point (asterisk), with a tract of possibly intrinsic origin
(double arrows) emerging from the branch point and describing a
partial circle posterior to the pedunculus. Another tract of extrinsic
origin is filled laterally (arrow). C: Pedunculus and medial lobe (M),
with putative intrinsic tract (double arrows) extending dorsal to the
medial lobe. D–E: Anti-DC0 immunostaining of the larval and prepu-
pal pedunculus (Ped) and medial lobe (M) reveal medial branching of
Class II Kenyon cells. D: A subset of Class II axons in the spinning
fifth instar larva show high affinity for anti-DC0 and reveal medial
branches (arrows). The tract encircling the pedunculus, which is seen
in the larval DiI fills is also visible (double arrows), its affinity for
anti-DC0 indicates its possible intrinsic neuron origin. E: At the
prepupal stage, medial branches (arrows) of Class II axons are heavily
immunostained. F: Sagittal section of the prepupal pedunculus and
vertical lobe. The vertical lobe tip is bent dorsally and has a distinct
notch at the ventral surface (arrow). Blebbed axons and debris are
seen in the vicinity of axon tips. Inset, F: DiI labeled sagittal section
of the medial lobe showing degeneration debris among medial lobe
axons. G: Anti-DC0 staining of the tip of the prepupal vertical lobe.
Three (numbered) of the four component axon bundles are visible.
Vertically projecting axons of Class II Kenyon cells having a high
affinity to anti-DC0 comprise the most ventral layer of the vertical
lobe (bracketed). H: The tip of the prepupal vertical lobe, Cason’s
staining, also showing axon bundles in the upturned lobe tip. I: Anti-
DC0 staining showing a sagittal section of the pedunculus and verti-
cal lobe in the day 1 pupa. The dorsal curvature of the vertical lobe
and the ventral notch (arrow) are less pronounced than in the pre-
pupa. Class II Kenyon cell crossing fibers are visible at the posterior
margin of the pedunculus (arrowheads). J: Anti-taurine-stained day 3
pupa. Little specific staining is visible, but the overall structure of the
lobe can be discerned. Most axons in the vertical lobe show little or no
dorsal extension and the ventral notch (old position indicated by
arrow) has at this stage disappeared. Scale bars 20 m in A–G; 10
minH;50m in I–J.
331LOBE DEVELOPMENT IN HONEY BEE MUSHROOM BODIES
Fig. 4. Axon ingrowth during mushroom body development.
A–F: Anti-DC0 (purple) and phalloidin (green) double labeling. A–C
are from 3-day-old pupae; D–F compare larva, mid-, and late pupae.
A: Heavy phalloidin labeling in the lip (Li) and collar (Co) of the
calyces indicates extensive outgrowth of dendritic processes. Axons
from newborn Kenyon cells converge from the protocalyces into the
necks of the pedunculus (Ped), where they define the ingrowth tracts
(arrows). Populations of Class II Kenyon cells continue to show a
relatively high affinity for anti-DC0 (arrowheads). B: Ingrowth tracts
from each calyx through the necks of the pedunculus (Ped) remain
separated then spread as two broad laminar tracts (arrows) in the
medial lobe (M). CC, central complex. C: A single ingrowth lamina
demarcates the dorsal margin of the vertical lobe (arrows); these
axons will at later stages be “displaced” deeper by axons belonging to
layers of the lobe representing the basal ring. Strong anti-DC0 affinity
is seen in Class II Kenyon cell axons (arrowheads) at the vertical/
medial lobe branch point, and in the short medial layer (M). D–F:
Anti-DC0 and phalloidin double labeling throughout larval and pupal
development. D: Spinning fifth instar larva showing anti-DC0 label-
ing of earlier-born Class II Kenyon cells (arrowheads) and two diffuse
phalloidin-labeled ingrowth tracts (arrows). No calycal neuropil has
been formed at this stage. E: Ingrowth tracts (arrows) appear more
distinct and cohesive in the late prepupa. Calycal (Ca) primordia are
visible atop each neck of the pedunculus. F: The ingrowth tracts in the
day 7 pupa have dwindled to thin threads (arrows) due to the abate-
ment of MBNB activity 2 days previously. Robust anti-DC0 staining is
seen in the majority of Kenyon cell axons. The highest affinities are
still seen in Class II Kenyon cells, including those providing cross
fibers projections (indicated by arrowheads). G–I: Collateral out-
growths from medial lobe ingrowth tracts. G: Phalloidin staining of a
3-day-old pupa reveals dense arrays of short processes (arrowheads)
emerging at an 45° angle from the ingrowth tract (arrow). H: DiI
labeling of the ingrowth tract (arrow) in a day 1 pupal mushroom body
shows an identical arrangement of collaterals (arrowheads). I: En-
larged view of DiI-labeled ingrowth axons and collaterals. Scale bars
100 m in A–C, E–F; 20 m in in D,I; 50 m in G,H.
332 S.M. FARRIS ET AL.
appeared less cohesive, as indicated by the more diffuse
phalloidin labeling. By the prepupal stage the ingrowth
tracts were well defined (Fig. 4E). Robust ingrowth tracts
were maintained throughout the period of mushroom body
neurogenesis, which ends at about day 5 of the pupal stage
(Farris et al., 1999; Ganeshina et al., 2000). After this time
the tracts became progressively thinner as the number of
newborn, ingrowing Kenyon cells decreased (Fig. 4F).
Phalloidin labeling of the adult honey bee mushroom bod-
ies revealed no ingrowth tract (data not shown). This was
consistent with previous reports of the death of MBNBs in
the pupal stage (Farris et al., 1999; Ganeshina et al.,
2000), and the lack of additional neurogenesis in the adult
honey bee (Fahrbach et al., 1995).
Close examination of the axons making up the ingrowth
tract revealed that they produced numerous fine collater-
als in the medial lobe (Fig. 4G–I). Such collaterals were
not observed in any other population of Kenyon cells. Due
to the transient occupation of the ingrowth tract by newly
extending axons, these collaterals must be transient in
nature. A similar case of extension and retraction of col-
laterals during Kenyon cell maturation has been de-
scribed in the cockroach Periplaneta americana (Farris
and Strausfeld, 2001).
Developmental and adult morphology of
clawed Kenyon cell axons
Golgi impregnations and TxR-dextran labeling demon-
strated that at least four subpopulations of Class II Ke-
nyon cells could be differentiated by their axon projection
pattern. At the region of the medial/vertical lobe branch
point, some Class II Kenyon cell axons were observed to be
unbranched, projecting only into the vertical lobe (Fig.
5A). Other axons bifurcated, however, providing ex-
tremely short medial branches as well as long vertical
branches (Fig. 5A–C). The medial branches of these axons
projected only a short distance along the anterior surface
of the medial lobe. Both types of axons could be traced into
the calyx and definitively identified as Class II Kenyon
cells, which are characterized by the location of their cell
bodies outside of the calyx and their “clawed” dendritic
morphology (Fig. 5D).
Bodian-stained and ethyl gallate-stained preparations
both revealed an anatomically distinct layer in the ante-
rior medial lobe (Fig. 5E–G). Sagittal sections of Golgi-
impregnated preparations showed that this layer con-
tained the short medial axons of Class II Kenyon cells
(data not shown). At the medial/vertical lobe branch point,
this “medial ” layer appeared wedged between the verti-
cal lobe’s layer and the more posterior layers of the
medial lobe (Fig. 5E). The medial lobe’s layer, like the
pronounced layer along the length of the vertical lobe,
was separated from the rest of the lobe neuropil by a layer
of what are interpreted to be profiles belonging to glial
cells and axons of neurons entering and leaving the lobe
(Fig. 5E,F). Being composed of the extremely short medial
axons of Class II Kenyon cells, the medial layer extended
only part of the length of the medial lobe. Serial sections
taken from the origin of the medial lobe, at its branch
point with the vertical lobe, towards its distal tip, showed
that the layer diminishes in size and then ends at the
level of the central complex, approximately halfway down
the medial lobe (Fig. 5F,G).
TxR-dextran fills that targeted the vertical lobe’s layer
confirmed the presence of axons with short medial
Fig. 5. Morphology of Class II Kenyon cells in the adult honey bee.
A–D: Frontal sections of Golgi impregnations. A: Medial-vertical
branch point showing unbranched axons entering the vertical lobe
(arrowheads) accompanied by other Class II axons that provide short
medial branches (arrow). B,C: Two consecutive sections of the vertical
lobe (B) and medial-vertical branch point (C) showing short medial
branches of Class II Kenyon cell axons (arrow). D: Characteristic
“clawed” dendrites of Class II Kenyon cells. E–G: Selected sagittal
sections showing the medial-vertical branch point (E) and, in F and G,
two progressively more distal levels through the medial lobe (M). E:
The medial lobe’s layer (arrow) is wedged between the medial and
vertical (V) lobes and is delineated from Class I components of the
medial lobe by outgoing axons of efferent neurons, here seen as a layer
of large translucent profiles (arrowheads). , vertical division. F:
Further distal, the medial lobe’s division (arrow) is small but still
discernable. G: In the distal medial lobe, at the level of the central
complex (CC), the medial layer is absent (arrow). H–J: Selected
serial frontal sections of a group of Class II Kenyon cells filled with
dextran conjugated to Texas Red. H: Dye injection site (arrow) pro-
viding backfills into Class II Kenyon cells within the layer (,
bracketed) of the vertical lobe (V). I: Vertically oriented axons (arrow-
head) and their short medially directed tributaries (arrow) at the
medial/vertical branch point of the two lobes (compare with branch
point revealed by Golgi impregnation in 5C). J: Clawed dendrites of
Class II Kenyon cells belonging to back filled axons from the vertical
lobe. All three calycal subdivisions, lip (Li), collar (Co) and basal ring
(BR), are represented. Scales bars 50 m in A–C; 10 minD;100
m in E–G; 20 minHJ100m.
333LOBE DEVELOPMENT IN HONEY BEE MUSHROOM BODIES
branches and revealed their identities as Class II Kenyon
cells (Fig. 5H–J). Bundles of branched Class II axons
labeled in this manner were traced back into the calyx,
where they were observed to originate from cell bodies
outside the calyx and to provide dendritic arborizations to
the lip, collar, and basal ring regions (Fig. 5J). No specific
correlation between the location of cell bodies and den-
drites, and the presence or absence of medial axon
branches, could be discerned.
Anti-DC0 immunostaining revealed an additional and
unexpected projection pattern of Class II Kenyon cells
that was particularly evident in the pupal mushroom bod-
ies. In the pupa, a subset of heavily stained processes
provided systems of “crossing fibers” that were observed
emerging dorsally from tracts between the two necks of
the pedunculus, immediately beneath the calyces. These
crossing fibers traversed the posterior surface of the pe-
dunculus and then ventrally entered lateral tracts in the
distal part of the pedunculus, beneath the point of fusion
of the two pedunculus necks (Fig. 6A,B). Certain of these
crossing fibers were grouped into distinct bundles, while
others were more loosely associated and formed a net-like
array across the posterior surface of the pedunculus (Fig.
6B). Due to the heavy staining of other Class II Kenyon
cell subtypes, the cell body locations and the disposition of
the crossing fibers deeper in the lobes could not be deter-
mined from the anti-DC0-immunostained preparations.
However, they were resolved by Golgi impregnations and
TxR-dextran fills, both of which revealed these unusual
Kenyon cell projections (Fig. 6C–G).
At the level of the medial/vertical lobe branch point,
subpopulations of both branched and unbranched Class II
Kenyon cell axons were additionally observed to produce
crossing fibers (Fig. 6C). Class II Kenyon cells in the honey
bee can therefore be subdivided into four subpopulations:
those with medial axon branches, those without medial
axon branches, those with medial axon branches and
crossing fibers, and those without medial branches and
with crossing fibers. The origin of the crossing fibers was
then determined by tracing the fibers retrogradely
through serial sections back to their cell bodies. From the
medial/vertical branch point, crossing fibers could be
traced dorsally and posteriorly (Fig. 6C, curved arrow) to
the posterior surface of the pedunculus, where they turned
abruptly to cross the posterior surface until reaching the
midline separation between the two pedunculus necks
(Fig. 6D, small arrows). At the pedunculus midline, the
fibers again turned dorsally, entering a tract situated at
the innermost surface of the pedunculus neck, and leading
back to one of the two calyces (Fig. 6D,E; large arrow).
This tract was traced back to its origin from neurites
forming clawed dendritic arborizations characteristic of
Class II Kenyon cells (Fig. 6E, arrow; also in basal ring
(BR)).
The cell bodies providing crossing fibers were invariably
located at the very outermost margins of the Kenyon cell
body clusters, outside each calyx near its confluence with
its pedunculus neck, indicating that they are among the
first born Kenyon cells (Fig. 6G). Correspondingly, all
Golgi and TxR-dextran preparations in which crossing
fibers were observed had labeled dendritic arbors in the
basal ring region (Fig. 6F). In contrast, preparations in
which Class II Kenyon cells provided dendrites only to the
lip and collar region did not show labeling of crossing
fibers and must be assumed to provide axons that project
Fig. 6. Crossing fiber trajectories of Class II Kenyon cells, frontal
view. A,B: Anti-DC0 staining in the day 5 pupa reveals crossing fiber
tracts (arrows) extending across the posterior surface of each pedun-
culus neck (Ped). CC, central complex. B: At higher magnification
some crossing fiber axons form bundles, whereas others appear dif-
fuse and isolated. C–E: Serial sections through a Golgi-impregnated
brain of a 1-day-old adult bee tracing the trajectory of a bundle of
crossing fibers derived from a cluster of Class II Kenyon cells from the
lobe branch point in C, through the pedunculus in D, and back to the
calyx in E. C: Short medial tributaries (arrow) arise from the cluster
of stained axons, which also send branches into the division of the
vertical lobe (V). Dorsal to the branchpoint, the axons can be traced
along the lateral surface of the pedunculus (direction indicated by
curved arrow). M, medial lobe. D: Axons of the bundle (curved arrow)
cross the neck of the pedunculus posteriorly (thin arrows) to the
medial surface of the neck, where they join a tract carrying axons of
other Class II Kenyon cells. This composite tract can be traced dor-
sally into the calyx (thick arrow). E: In the calyx, axons forming the
aforementioned composite pedunculus tract (large arrow) are ob-
served to derive from Class II Kenyon cells with dendritic arbors at
the lip (Li)-collar (Co) boundary (small arrow) and basal ring (BR). K
cbs, Kenyon cell bodies. F–G: Serial frontal sections showing ensem-
bles of Class II cell bodies (arrowheads) and their dendrites, shown
mainly clustered in the basal rings (BR) of both calyces (Ca), after
dextran-Texas Red backfills of Class II Kenyon cells, including those
providing cross fibers (arrows). Scales bars 50 m in A,C,D,E; 20
m in B,F,G.
334 S.M. FARRIS ET AL.
via a more direct route down the pedunculus into the
layers of the lobes.
Significantly, crossing fibers were not observed in DiI
filled mushroom bodies prior to the early prepupal stage,
and did not display an affinity for anti-DC0 until the late
prepupal stage (Fig. 7A,B). Prepupal crossing fiber bun-
dles were thin and appeared to be composed of fewer fibers
than those of the pupa. This indicated that, although the
location of cell bodies providing crossing fibers places
them among the first born Kenyon cells, their fiber out-
growth may lag behind that of other Class II Kenyon cells.
Crossing fibers displayed a particularly high affinity to
anti-DC0 antiserum in the late prepupa and throughout
the pupal stage (Figs. 6A,B, 7B). In the 1-day-old adult,
bundles of crossing fibers were still visible, but did not
appear as distinct from the rest of the mushroom body
neuropil (Fig. 7C). This could have resulted from a relative
decrease in anti-DC0 affinity in the cross fibers, or a
relative increase in anti-DC0 affinity in the axons of other
Kenyon cells. Anti-DC0 immunostaining of crossing fibers
in the brains of honey bee foragers, which are typically 3
or more weeks of age (Winston, 1987) was extremely faint
(Fig. 7D) or entirely absent. It was not possible to deter-
mine from immunocytology whether this was solely due to
a loss of anti-DC0 immunoreactivity, or if the cross fibers
physically degenerated as well. However, examination of
forager brains stained by a variety of histological tech-
niques including TUNEL staining showed no evidence for
degenerating Class II Kenyon cell bodies (S.E. Fahrbach,
J.E. Mehren, unpubl. obs.) and it must be assumed that
those providing crossing fibers are maintained throughout
adult life.
Massive reorganization of mushroom bodies
in a polistine wasp
Extensive remodeling of the mushroom bodies involving
degeneration and regrowth of Class II Kenyon cell pro-
cesses during metamorphosis has been well documented
in the fly D. melanogaster (Technau and Heisenberg, 1982;
Lee et al., 1999; Watts et al., 2003). In contrast, the
present account has shown that a subtle reshaping of the
lobes characterizes mushroom body reorganization in the
honey bee. Does this suggest that large-scale remodeling
of the mushroom bodies, which is characteristic of the
Diptera, is not a general feature of hymenopteran brains?
The morphology of the adult mushroom body lobes in
most hymenopteran species is similar to that of Apis mel-
lifera, with an anteriorly projecting vertical lobe and me-
dially projecting medial lobe. However, one exceptional
group has been described. These are the vespid wasps
(Hymenoptera, Vespidae), in which mushroom bodies
have a very different adult morphology (Jawlowski, 1959;
Ehmer and Hoy, 2000). Vespid species of the subfamily
Polistinae display the most extreme example: an attenu-
ated vertical lobe that is oriented ventrally, with the me-
dial lobe split into two parallel components that protrude
anteriorly from the top of the pedunculus and in some
species even appear to “herniate” through the calyces.
Kenyon cell axons of polistine wasps do not attain this
adult projection pattern until late in pupal development
(B. Ehmer, unpubl. obs.; this account). In the early pupal
stage, polistine mushroom bodies, here represented by the
species Polistes apachiensis, exhibit a morphology that is
nearly identical to that of adult mushroom bodies of other
hymenopteran species, including the honey bee (Fig. 8).
However, this morphology is transient, because during the
mid-pupal stage the vertical lobe appears to retract,
shorten, and broaden, while changing from an anterior to
a ventral orientation (Fig. 8A–C). The pedunculus also
thickens, and by the end of the pupal stage the short
vertical lobe is oriented directly beneath the pedunculus
(Fig. 8D–F), similar to the position of the medial lobe in
the honey bee. The reorganization of the medial lobe is
even more dramatic, with an apparently complete retrac-
tion and then regrowth retrogradely into the pedunculus
followed by a posteriorly directed reextension of two sep-
arated axon bundles (Fig. 8G–I). Golgi impregnations of
the mid- to late pupa reveal axon debris (not shown) in the
vicinity of the larval medial lobe. As in the honey bee this
suggests that medial axon “retraction” is a consequence of
axon degeneration and regrowth reminiscent of that ob-
served in Drosophila, but in the case of the wasp involving
not only Class II Kenyon cells, but all classes of intrinsic
neurons.
DISCUSSION
Conserved sequential cell generation during
mushroom body development
Mushroom body development has been studied in detail
in four divergent taxa: two hemimetabolous species, the
Fig. 7. Frontal sections showing juvenile and adult development of
Class II Kenyon cell cross fibers. A: DiI labeling first reveals sparse
cross fiber tracts (arrows) traversing the posterior pedunculus (Ped)
in the early prepupa. Ca, calyx. B: DC0-like immunoreactivity in thin
cross fiber tracts is first apparent in the late prepupa. M, medial lobe,
nb, mushroom body neuroblasts. C: Anti-DC0 labels cross fibers (ar-
rows) in the 1-day-old adult, although the affinity of the antibody for
the fibers does not appear as strong as in the developing brain.
D: Cross fibers (arrows) in the adult honey bee forager (3 weeks old)
have little or no affinity for the anti-DC0 antibody. Abbreviations as
in previous figures. Scale bars 20 minA;50minB;100min
C,D.
335LOBE DEVELOPMENT IN HONEY BEE MUSHROOM BODIES
American cockroach Periplaneta americana and the house
cricket Acheta domestica; and two holometabolous species,
the fruit fly Drosophila melanogaster and the (worker)
honey bee Apis mellifera. The following common develop-
mental events are shared among these taxa:
1) There is a continuous generation of Kenyon cells by
mushroom body neuroblasts during development, re-
sulting in the sequential production of different classes
of neurons, the cell bodies of which in the adult have
specific positional relationships with respect to the ca-
lyx. The oldest cells lie farthest from the proliferative
center and the youngest cells reside closest (Panov,
1957; Ito and Hotta, 1992; Farris et al., 1999; Lee et al.,
1999; Cayre et al., 2000; Farris and Strausfeld, 2001;
Kurusu et al., 2002; Malaterre et al., 2002).
2) Axons of newborn Kenyon cells enter the lobes via an
ingrowth tract or ingrowth core that lies at the back of
the pedunculus and lobes. As newly ingrowing axons
are added, older axons are shifted to a relatively more
anterior position across the pedunculus and lobes. This
incremental addition of axons results in the peduncu-
lus and its deeper neuropils being composed of discrete
lamination and layers that are organized from anterior
to posterior in gradually decreasing age (Armstrong et
al., 1998; Farris and Strausfeld, 2001; Kurusu et al.,
2002; Malaterre et al., 2002).
The development and organization of the adult honey
bee mushroom bodies, as described in this account and
others, provides little exception to these general rules. The
ontogeny of the crossing fibers of Class II Kenyon cells,
however, appears to reveal the sole discontinuity between
cell birth and axon extension. Specifically, delayed axon
outgrowth by these populations of Class II neurons may be
associated with the late development of the basal ring
neuropil in which their dendrites reside. Since all Class II
neurons supply the layer, any late developing axons
would have to maneuver across already established layers
formed by Class II Kenyon cells and the initial popula-
tions of Class I cells to reach the ventrally located layer
of the lobes.
In conclusion, conserved mechanisms of Kenyon cell
organization during development result in a generally
age-based ordering of intrinsic neurons that appears to be
a universal characteristic of insect mushroom bodies (re-
viewed in Farris and Sinakevitch, 2003). Since Kenyon
cell subtypes are produced sequentially during develop-
ment (Farris et al., 1999; Lee et al., 1999; Farris and
Strausfeld, 2001), the adult mushroom bodies are com-
posed of distinct “modules” of Kenyon cells that can be
differentiated by morphology, gene expression pattern,
connectivity, and, presumably, function. This age-based
organization of neuronal subtypes into layered neuropils
is not restricted to the insect mushroom bodies, but is also
a well-known characteristic of the vertebrate brain, such
as the neocortex (reviewed in McConnell, 1995). The se-
quential generation of cell types and resulting age-based
organization of distinct cellular and functional subcompo-
nents is likely to be a highly conserved or even convergent
strategy for the generation of complex brain neuropils
across phyla.
Conserved remodeling of intrinsic neurons
during mushroom body development
More surprisingly, another aspect of Kenyon cell devel-
opment that appears to be common to both hemimetabo-
lous and holometabolous taxa is the transient production
of collaterals by axons of the ingrowth tract. The first
report of this apparent transdifferentiation during Ke-
nyon cell development was from observations of ingrowing
Class I Kenyon cell axons in Periplaneta americana, which
produce an elaborate collateral system that is lost as the
cells mature (Farris and Strausfeld, 2001). In Periplaneta
these newborn Kenyon cells also show transiently high
affinities for anti-glutamate antibodies. As maturation
proceeds, the Kenyon cells lose their affinity for antiglu-
tamate and gain affinity for antibodies against other mol-
ecules such as taurine and aspartate (Sinakevitch et al.,
2001). Ingrowing axons in the honey bee also give rise to
transient collaterals (Farris and Sinakevitch, 2003; this
account). In the case of the cockroach, it was proposed that
the collateral system might provide guidance cues for the
ingrowth of extrinsic neuron processes (terminals of affer-
ents to the lobes, and dendrites of efferent from the lobes)
that must continually establish synapses with new Ke-
Fig. 8. Metamorphosis of the mushroom body lobes in a polistine
wasp. A–C: Frontal view of the anterior mushroom bodies in the early
(A), middle (B), and late pupa (C). The anteriorly oriented vertical lobe
(V) of the early pupa appears to migrate ventrally during development
and is no longer visible in anterior sections of the late pupal brain. Ca,
calyx. D–F: Frontal section midway through the mushroom bodies in
the early (D), middle (E), and late pupa (F). During development, the
vertical lobe shortens and in the late pupa resides directly ventral to
the pedunculus (Ped). The pedunculus appears much thicker in the
late pupa due to the insertion of medial lobe Kenyon cell axons
through each pedunculus neck (see 8I). G–H: Frontal sections of the
mushroom bodies at a posterior level of the brain (arrows) in the early
(G), middle (H), and late pupa (I). The medial lobes (M) appear to
retract from the midline, and in the late pupa they project posteriorly
through each neck of the pedunculus. Scale bars 100 m.
336 S.M. FARRIS ET AL.
nyon cell axons during the months-long juvenile develop-
ment of this hemimetabolous insect. The collaterals of the
honey bee ingrowth tract may have a similar function,
although the developmental period is far shorter (10
days as opposed to as long as a year in the cockroach).
Short collaterals have also been observed from the Kenyon
cell axons that belong to the core of the vertical and medial
lobes of Drosophila. These axons are also glutamate-
immunoreactive and are recognized by phalloidin (Kurusu
et al., 2002; Strausfeld et al., 2003), and thus share both
morphological and immunochemical characteristics of
newly extending Kenyon cell axons with other taxa. Fur-
ther studies will be necessary in order to determine the
function of transient axon collaterals in these species.
Metamorphic reorganization of mushroom
bodies in holometabolous insects
Although the reorganization of mushroom body intrin-
sic neurons during metamorphosis has been extensively
characterized in Drosophila, in particular with respect to
Class II Kenyon cells (Lee et al., 1999), it was not previ-
ously known whether this process occurred in other in-
sects. Apart from the production and loss of collaterals
described above, neither Periplaneta nor Acheta show any
evidence for Kenyon cell reorganization reminiscent of
that described in Drosophila (Farris and Strausfeld, 2001;
Malaterre et al., 2002). Possibly no reorganization might
be expected due to their hemimetabolous mode of devel-
opment, which by definition lacks complete metamorpho-
sis via a pupal stage.
Holometabolous insects, by contrast, lead “double lives.”
The functional significance of mushroom body metamor-
phosis is not currently known, but as with other instances
of central nervous system metamorphosis, is likely to re-
flect the vast disparity in larval and adult morphology and
lifestyle. The flightless larvae of Drosophila and Apis are
feeding stages, displaying a relatively limited capacity for
movement and behavior. The winged adults, however,
navigate extensively in the environment using odor and
visual cues and perform many additional complex behav-
iors. Accordingly, the larva and adult differ enormously in
the morphology of sensory and motor apparatus. The tran-
sition between the two life stages is made possible by
metamorphosis of the central nervous system, allowing
neurons innervating larval-specific structures to acquire
new connections with adult-specific structures and neural
circuits (Consoulas et al., 2000; Libersat and Duch, 2002).
In the brain, the optic and antennal lobes are primarily
adult-specific structures (Oland et al., 1999). Larval MBs
necessarily lack these sources of afferent input, and per-
haps must dismantle or otherwise rearrange larval MB
circuits to accommodate those of the adult. Larval neurons
may also serve transiently as guidance cues for adult-
specific neuronal projections (Williams and Shepherd,
2002). Studies of Drosophila mushroom bodies suggest
that early-born Kenyon cells provide guidance cues for
later-born neurons entering the lobes (Wang et al., 2002).
In particular, the thin tract of ␣⬘/␤⬘ axons that persist
during Drosophila mushroom body metamorphosis are
proposed to guide axons of the adult-specific /Kenyon
cells into the adult lobes (Lee et al., 1999).
Previous studies of honey bee mushroom body develop-
ment have made no note of axon reorganization (Farris et
al., 1999; Schro¨ter and Malun, 2000). However, closer
scrutiny afforded by the present study reveals a subtle
reshaping of the mushroom body vertical lobe without
large-scale loss of axon branches. Axon reshaping in the
honey bee is accompanied by the apparent degeneration of
parts of Class II Kenyon cell axons, as evidenced by the
accumulation of cellular debris around their tips. The
cellular debris is reminiscent of that produced during the
initial degeneration of Class II Kenyon cell axons in the
mushroom bodies of the Drosophila pupa prior to their
regrowth into their unbranched adult configuration in the
medial lobe (Watts et al., 2003). In that study, the debris
was shown to be the result of local degeneration of axons
rather than retraction, and was mediated by the
ubiquitin-proteasome system.
In contrast to the minor reshaping observed in the
honey bee, mushroom bodies of polistine wasps undergo a
massive reorganization of their lobes. There are, however,
many differences between this reorganization and that
observed in the honey bee and fruit fly. In the polistine
wasp Polistes apachiensis, reorganization occurs late in
development relative to that of the other insects, taking
place in mid- to late pupal stages rather than in the
prepupal or early pupal stage. Assuming that the timeline
of Kenyon cell production is similar to that described for
the honey bee, metamorphosis of the polistine mushroom
bodies may also involve other cell types besides Class II
Kenyon cells. The late occurrence of axon reorganization
during the pupal stage likely occurs after Class II Kenyon
cell production has ended and the generation of other cell
types has begun. Despite these differences, it is possible
that further investigation into development of the mush-
room bodies of polistine wasps will reveal similar mecha-
nisms at work both in axon outgrowth and reorganization.
Comparative studies of adult mushroom body anatomy
of a variety of hymenopteran species by Ehmer and Hoy
(2000) indicate that the adult polistine-type mushroom
body morphology is a derived trait, even within the Vespi-
dae (which contains the polistine wasps). Other species of
the Hymenoptera, as well as some vespid wasps, have
Apis-like mushroom bodies, and some vespids even appear
to have morphologies intermediate between the two. The
polistine mushroom bodies thus pass through a transient
morphological stage in development that resembles the
adult mushroom bodies of other hymenopteran species.
Subsequent reorganization of this presumably basal mor-
phology produces the highly derived lobe configuration of
the adult wasp. Early development of a neuronal structure
reminiscent of that found in basal species, followed by late
developmental reorganization into a derived configura-
tion, has also been described in the interaural time differ-
ence coding circuit of the barn owl Tyto alba (Kubke et al.,
2002). The first phase of development produces a circuit
similar to that seen in adult brains of more basal bird
species, such as the chicken. A subsequent wave of mor-
phogenesis results in the reorganization of the circuit into
the distinct configuration characteristic of the adult owl.
Axon morphologies of honey bee Class II
Kenyon cells
Past investigations of the projection patterns of Class II
Kenyon cells have starkly disagreed, with two articles
reporting that Class II Kenyon cells branch the full length
of the vertical and medial lobes (Mobbs, 1982; Rybak and
Menzel, 1993), and another showing unbranched Class II
Kenyon cell axons supplying only the layer of the verti-
cal lobe (Strausfeld, 2002). The present account shows
337LOBE DEVELOPMENT IN HONEY BEE MUSHROOM BODIES
that a complete Class II axon bifurcation into the full
length of the two lobes is incorrect, but that a complete
restriction of Class II axons to the vertical lobe’s layer
also needs revision. Far from simplifying matters, how-
ever, the present results suggest that there are at least
four categories of Class II Kenyon cells based on their
axon projection pattern alone. The recognition here of
Class II Kenyon cells having short branches into the me-
dial lobe suggests that early in development the organi-
zation of Class II Kenyon cells is no different from that in
Drosophila, where they provide a medial and vertical lobe
in the larva. In the honey bee, however, vertical branches
undergo partial pruning by degeneration and the medial
branches apparently cease to grow further than approxi-
mately half the length of the adult medial lobe. This is also
evidenced by the tiny medial layer arising from the root
of the vertical lobe’s layer.
The trajectories of certain Class II Kenyon cells, which
form the distinct crossing fiber tracts fasciculating over
the posterior surface of the pedunculus, have not been
previously described and suggest a special case of late
development of a subset of this cell type. Although in-
tensely stained in the developing brain using the anti-DC0
antibody, the crossing fibers are virtually invisible using
conventional histological techniques such as Cason’s or
ethyl gallate staining (S.M. Farris, unpubl. obs.) although
the axons can be identified in Golgi impregnations. The
most parsimonious explanation for these elaborate trajec-
tories has been discussed above (a discontinuity between
cell birth and axon outgrowth). Anti-DC0 antibody has
also revealed a previously undescribed population of in-
trinsic neurons in the cockroach (the Class III “lobelet”
Kenyon cells), and was subsequently used to survey other
members of the order Dictyoptera for the presence of these
cells (Farris and Strausfeld, 2003). The novelty of these
structures suggests that despite decades of research into
the honey bee and cockroach mushroom bodies, much
remains to be learned about their cellular composition and
commonalties, which may provide important insights into
the ancestral and present functions of these structures.
Strategies for studying the evolution of
insect mushroom bodies
The present study suggests two central strategies for
further comparative studies. First, the use of molecular
markers such as anti-DC0 and phalloidin may be crucial
for identifying and differentiating between small popula-
tions of Kenyon cell subtypes in different insect species.
Staining patterns can then be interpreted in the light of
other evidence, such as morphology and developmental
history, to determine the presence or absence of specific
cell types across taxa (Farris and Strausfeld, 2003).
Second, observations of both developing and adult in-
sects are critical to forming a more complete picture of the
evolution of mushroom bodies. Some apparent loss of
staining of cellular components, such as the Class II Ke-
nyon cell crossing fibers or the medial/vertical branch
point tract in the late larval honey bee, may suggest
degeneration or loss of affinity for the marker. In any
event, such structures would certainly be missed in stud-
ies performed at only a single time point in the life of the
insect. Other mushroom body components may be orga-
nized in such a way in the adult that it is initially difficult
to compare their morphology with that of other species, as
is the case with polistine wasps (see Ehmer and Hoy,
2000). Yet an understanding of earlier developmental
stages can reveal that even such a derived structure in the
adult wasp shares principles of organization with other
taxa. Finally, the widespread occurrence of mushroom
body metamorphosis in holometabolous species may pro-
vide useful insights into different requirements by larvae
and adults for higher sensory integration neuropils.
ACKNOWLEDGMENTS
The authors are indebted to Dr. Gloria DeGrandi-
Hoffman, Mr. Thomas Deeby, and others at the USDA-
ARS Carl Hayden Bee Research Center at the University
of Arizona, who kindly provided many frames of honeybee
brood for this study. Dr. Irina Sinakevitch provided assis-
tance with immunostaining protocols. We also thank Dr.
Daniel Kalderon for generously providing the anti-DC0
antibody.
LITERATURE CITED
Armstrong JD, deBelle JS, Wang Z, Kaiser K. 1998. Metamorphosis of the
mushroom bodies; large-scale rearrangements of the neural substrates
for associative learning and memory in Drosophila. Learn Mem 5:102–
114.
Bertholf LM. 1925. The moults of the honeybee. J Econ Entomol 18:380
384.
Cano Lozano V, Armengaud C, Gauthier M. 2001. Memory impairment
induced by cholinergic antagonists injected into the mushroom bodies
of the honeybee. J Comp Physiol A 187:249 –254.
Cayre M, Malaterre J, Charpin P, Strambi C, Strambi A. 2000. Fate of
neuroblast progeny during postembryonic development of mushroom
bodies in the house cricket, Acheta domesticus. J Insect Physiol 46:313–
319.
Consoulas C, Duch C, Bayline R, Levine R. 2000. Behavioral transforma-
tions during metamorphosis: remodeling of neural and motor systems.
Brain Res Bull 53:571–583.
deBelle JS, Heisenberg M. 1994. Associative odor learning in Drosophila
abolished by chemical ablation of mushroom bodies. Science 263:692–
695.
Durst C, Eichmuller S, Menzel R. 1994. Development and experience lead
to increased volume of subcompartments of the honeybee mushroom
body. Behav Neural Biol 62:259 –263.
Ehmer B, Hoy R. 2000. Mushroom bodies of vespid wasps. J Comp Neurol
416:93–100.
Ehmer B, Gronenberg W. 2002. Segregation of visual input to the mush-
room bodies in the honey bee (Apis mellifera). J Comp Neurol 451:362–
373.
Erber J, Masuhr T, Menzel R. 1980. Localization of short-term memory in
the brain of the bee, Apis mellifera. Physiol Entomol 5:343–358.
Fahrbach SE, Strande JL, Robinson GE. 1995. Neurogenesis is absent in
the brains of adult honey bees and does not explain behavioural neu-
roplasticity. Neurosci Lett 197:145–148.
Farris SM, Sinakevitch I. 2003. Development and evolution of the insect
mushroom bodies: Towards the understanding of conserved develop-
mental mechanisms in a higher brain center. Arthropod Struct Devel
32:79 –101.
Farris SM, Strausfeld NJ. 2001. Development of laminar organization in
the mushroom bodies of the cockroach: Kenyon cell proliferation, out-
growth, and maturation. J Comp Neurol 439 :331–351.
Farris SM, Strausfeld NJ. 2003. A unique mushroom body substructure
common to both basal cockroaches and to termites. J Comp Neurol
456:305–320.
Farris SM, Robinson GE, Davis RL, Fahrbach SE. 1999. Larval and pupal
development of the mushroom bodies in the honey bee, Apis mellifera.
J Comp Neurol 414:97–113.
Farris SM, Robinson GE, Fahrbach SE. 2001. Experience- and age-related
outgrowth of intrinsic neurons in the mushroom bodies of the adult
worker honey bee. J Neurosci 21:6395– 6404.
Ganeshina O, Schafer S, Malun D. 2000. Proliferation and programmed
338 S.M. FARRIS ET AL.
cell death of neuronal precursors in the mushroom bodies of the hon-
eybee. J Comp Neurol 417:349 –365.
Gronenberg W. 2001. Subdivisions of hymenopteran mushroom body caly-
ces by their afferent supply. J Comp Neurol 436:474 489.
Gundersen R, Larsen J. 1978. Postembryonic development of the lateral
protocerebral lobes, corpora pedunculata, deutocerebrum and tritoce-
rebrum of Phormia regina Meigen (Diptera: Calliphoridae). J Insect
Morphol Embryol 7:467– 477.
Heisenberg M, Borst A, Wagner S, Byers D. 1985. Drosophila mushroom
body mutants are deficient in olfactory learning. J Neurogenet 2:1–30.
Huang Z-Y, Robinson GE, Tobe SS, Yagi KJ, Strambi C, Strambi A, Stay B.
1991. Hormonal regulation of behavioural development in the honey
bee is based on changes in the rate of juvenile hormone biosynthesis.
J Insect Physiol 37:733–741.
Ito K, Hotta Y. 1992. Proliferation pattern of postembryonic neuroblasts in
the brain of Drosophila melanogaster. Dev Biol 149:134 –148.
Ito K, Awano W, Suzuki K, Hiromi Y, Yamamoto D. 1997. The Drosophila
mushroom body is a quadruple structure of clonal units each of which
contains a virtually identical set of neurones and glial cells. Develop-
ment 124:761–771.
Jawlowski H. 1959. The structure of corpora pedunculata in Aculeata
(Hymenoptera). Folia Biol 7:61–70.
Kubke MF, Massoglia DP, Carr CE. 2002. Developmental changes under-
lying the formation of the specialized time coding circuits in barn owns
(Tyto alba ). J Neurosci 22:7671–7679.
Kurusu M, Awasaki T, Masuda-Nakagawa LM, Kawauchi H, Ito K,
Furukubo-Tokunaga K. 2002. Embryonic and larval development of the
Drosophila mushroom bodies: concentric layer subdivisions and the
role of fasciclin II. Development 129:409 419.
Laurent G, Naraghi M. 1994. Odorant-induced oscillations in the mush-
room bodies of the locust. J Neurosci 14:2993–3004.
Lee T, Lee A, Luo L. 1999. Development of the Drosophila mushroom
bodies: sequential generation of three distinct types of neurons from a
neuroblast. Development 126:4065– 4076.
Li Y-S, Strausfeld NJ. 1997. Morphology and sensory modality of mush-
room body extrinsic neurons in the brain of the cockroach, Periplaneta
americana. J Comp Neurol 387:631– 650.
Li Y-S, Strausfeld NJ. 1999. Multimodal efferent and recurrent neurons in
the medial lobes of cockroach mushroom bodies. J Comp Neurol 409:
647– 663.
Libersat F, Duch C. 2002. Morphometric analysis of dendritic remodeling
in an identified motoneruon during postembryonic development.
J Comp Neurol 450:153–166.
Malaterre J, Strambi C, Chiang A-S, Aouane A, Strambi A, Cayre M. 2002.
Development of cricket mushroom bodies. J Comp Neurol 452:215–227.
McConnell SK. 1995. Strategies for the generation of neuronal diversity in
the developing central nervous system. J Neurosci 15:6987– 6998.
Michelette ER de F, Soares AEE. 1993. Characterization of preimaginal
developmental stages in Africanized honey bee workers (Apis mellifera
L). Apidologie 24:431– 440.
Mizunami M, Okada R, Li Y-S, Strausfeld NJ. 1998a. Mushroom bodies of
the cockroach: activity and identities of neurons recorded in freely
moving animals. J Comp Neurol 402:501–519.
Mizunami M, Weibrecht J, Strausfeld NJ. 1998b. Mushroom bodies of the
cockroach: their participation in place memory. J Comp Neurol 402:
520 –537.
Mobbs PG. 1982. The brain of the honeybee Apis mellifera L. The connec-
tions and spatial organization of the mushroom bodies. Philos Trans R
Soc Lond B 298:309 –354.
Mobbs PG. 1984. Neural networks in the mushroom bodies of the honey-
bee. J Insect Physiol 30:43–58.
Oland L, Marrero H, Burger I. 1999. Glial cells in the developing and adult
olfactory lobe of the moth Manduca sexta. Cell Tissue Res 297:527–545.
Panov AA. 1957. The structure of the brain in insects in successive stages
of postembryonic development. Rev Entom USSR 36:269 –284.
Pascual A, Pre´at T. 2001. Localization of long-term memory within the
Drosophila mushroom body. Science 294:1115–1117.
Pearson L. 1971. The corpora pedunculata of Sphinx ligustri L. and other
Lepidoptera: an anatomical study. Philos Trans R Soc Lond B 259:477–
516.
Rybak J, Menzel R. 1993. Anatomy of the mushroom bodies in the honey
bee brain: The neuronal connections of the alpha-lobe. J Comp Neurol
334:444 – 465.
Schildberger K. 1984. Multimodal interneurons in the cricket brain: prop-
erties of identified extrinsic mushroom body cells. J Comp Physiol
154:71–79.
Schro¨ter U, Malun D. 2000. Formation of antennal lobe and mushroom
body neuropils during metamorphosis in the honeybee, Apis mellifera.
J Comp Neurol 422:229 –245.
Schro¨ter U, Menzel R. 2003. A new ascending sensory tract to the calyces
of the honeybee mushroom body, the subesophageal-calycal tract.
J Comp Neurol 465:168 –178.
Sinakevitch I, Farris SM, Strausfeld NJ. 2001. Taurine-, aspartate- and
glutamate-like immunoreactivity identifies chemically distinct subdi-
visions of Kenyon cells in the cockroach mushroom body. J Comp
Neurol 439:352–367.
Skoulakis EMC, Kalderon D, Davis RL. 1993. Preferential expression in
mushroom bodies of the catalytic subunit of protein kinase A and its
role in learning and memory. Neuron 11:197–208.
Strausfeld NJ. 1976. Atlas of an insect brain. Berlin: Springer.
Strausfeld NJ. 2002. Organization of the honey bee mushroom body: rep-
resentation of the calyx within the vertical and gamma lobes. J Comp
Neurol 450:4 –33.
Strausfeld NJ, Li Y-S. 1999. Representation of the calyces in the medial
and vertical lobes of cockroach mushroom bodies. J Comp Neurol 409:
626 – 646.
Strausfeld NJ, Hansen L, Li Y, Gomez RS, Ito K. 1998. Evolution, discov-
ery, and interpretations of arthropod mushroom bodies. Learn Mem
5:11–37.
Strausfeld NJ, Homberg U, Kloppenburg P. 2000. Parallel organization in
honey bee mushroom bodies by peptidergic Kenyon cells. J Comp Neu-
rol 424:179 –195.
Strausfeld NJ, Sinakevitch I, Vilinsky I. 2003. The mushroom bodies of
Drosophila melanogaster: an immunocytological and Golgi study of
Kenyon cell organization in the calyces and lobes. Microsc Res Tech
62:151–169.
Technau GM, Heisenberg M. 1982. Neural reorganization during meta-
morphosis of the corpora pedunculata in Drosophila melanogaster.
Nature 295:405– 407.
Vowles D. 1964. Olfactory learning and brain lesions in the wood ant
(Formica rufa). J Comp Physiol Psychol 5:105–111.
Wang J, Zugates CT, Liang IH, Lee C-HJ, Lee T. 2002. Drosophila Dscam
is required for divergent segregation of sister branches and suppresses
ectopic bifurcation of axons. Neuron 33:559 –571.
Watts R, Hoopfer E, Luo L. 2003. Axon pruning during Drosophila meta-
morphosis: evidence for local degeneration and requirement of the
ubiquitin-proteasome system. Neuron 38:871– 885.
Williams D, Shepherd D. 2002. Persistent larval sensory neurones are
required for the normal development of the adult sensory afferent
projections in Drosophila. Development 129:617– 624.
Winston ML. 1987. The Biology of the honey bee. Cambridge, MA: Harvard
University Press.
Withers GS, Fahrbach SE, Robinson GE. 1993. Selective neuroanatomical
plasticity and division of labour in the honeybee. Nature 364:238 –240.
Zars T, Fischer M, Schulz R, Heisenberg M. 2000. Localization of a short-
term memory in Drosophila. Science 288:672– 675.
339LOBE DEVELOPMENT IN HONEY BEE MUSHROOM BODIES
... In addition, we dare to speculate that there might be a mechanism to convert a neuroblast from type 1 to type 2, generating a large "clan" of intermediate progenitors from one ancestral type 1 neuroblast. The description of hundreds of proliferating neural progenitors (originally described as neuroblasts) in the large bee MB (Farris et al., 2004) could represent a case of such "type 1>type2" transformation. ...
Article
Full-text available
Insect brains are formed by conserved sets of neural lineages whose fibers form cohesive bundles with characteristic projection patterns. Within the brain neuropil, these bundles establish a system of fascicles constituting the macrocircuitry of the brain. The overall architecture of the neuropils and the macrocircuitry appear to be conserved. However, variation is observed, for example, in size, shape, and timing of development. Unfortunately, the developmental and genetic basis of this variation is poorly understood, although the rise of new genetically tractable model organisms such as the red flour beetle Tribolium castaneum allows the possibility to gain mechanistic insights. To facilitate such work, we present an atlas of the developing brain of T. castaneum, covering the first larval instar, the prepupal stage, and the adult, by combining wholemount immunohistochemical labeling of fiber bundles (acetylated tubulin) and neuropils (synapsin) with digital 3D reconstruction using the TrakEM2 software package. Upon comparing this anatomical dataset with the published work in Drosophila melanogaster, we confirm an overall high degree of conservation. Fiber tracts and neuropil fascicles, which can be visualized by global neuronal antibodies like antiacetylated tubulin in all invertebrate brains, create a rich anatomical framework to which individual neurons or other regions of interest can be referred to. The framework of a largely conserved pattern allowed us to describe differences between the two species with respect to parameters such as timing of neuron proliferation and maturation. These features likely reflect adaptive changes in developmental timing that govern the change from larval to adult brain.
... In addition, we dare to speculate that there might be a mechanism to convert a neuroblast from type 1 to type 2, generating a large "clan" of intermediate progenitors from one ancestral type 2 neuroblast. The description of hundreds of proliferating neural progenitors (originally described as neuroblasts) in the large bee mushroom body (Farris et al., 2004) could represent a case of such "type 1>type2" transformation. ...
Preprint
Full-text available
Insect brains are formed by conserved sets of neural lineages whose fibres form cohesive bundles with characteristic projection patterns. Within the brain neuropil these bundles establish a system of fascicles constituting the macrocircuitry of the brain. The overall architecture of the neuropils and the macrocircuitry appear to be conserved. However, variation is observed e.g., in size and shape and timing of development. Unfortunately, the developmental and genetic basis of this variation is poorly understood although the rise of new genetically tractable model organisms such as the red flour beetle Tribolium castaneum allows the possibility to gain mechanistic insights. To facilitate such work, we present an atlas of the developing brain of T. castaneum , covering the first larval instar, the prepupal stage and the adult, by combining wholemount immunohistochemical labelling of fibre bundles (acetylated tubulin) and neuropils (synapsin) with digital 3D reconstruction using the TrakEM2 software package. Upon comparing this anatomical dataset with the published work in D. melanogaster , we confirm an overall high degree of conservation. Fibre tracts and neuropil fascicles, which can be visualized by global neuronal antibodies like anti-acetylated tubulin in all invertebrate brains, create a rich anatomical framework to which individual neurons or other regions of interest can be referred to. The framework of a largely conserved pattern allowed us to describe differences between the two species with respect to parameters such as timing of neuron proliferation and maturation. These features likely reflect adaptive changes in developmental timing that govern the change from larval to adult brain.
... In the stomatopod lateral protocerebrum there are four adjacent columns, each associated with a distinct ensemble of globuli cells and hence a calyx. That organization has been compared to the Drosophila mushroom body, which is a composite neuropil of four hemi-mushroom bodies derived from four clonal lineages, as it is in the honey bee (Ito et al., 1997;Farris et al., 2004;Wolff et al., 2017). Consequently, the mushroom bodies in an insect's lateral protocerebrum are paired but completely fused. ...
Article
Full-text available
Neural organization of mushroom bodies is largely consistent across insects, whereas the ancestral ground pattern diverges broadly across crustacean lineages resulting in successive loss of columns and the acquisition of domed centers retaining ancestral Hebbian-like networks and aminergic connections. We demonstrate here a major departure from this evolutionary trend in Brachyura, the most recent malacostracan lineage. In the shore crab Hemigrapsus nudus , instead of occupying the rostral surface of the lateral protocerebrum, mushroom body calyces are buried deep within it with their columns extending outwards to an expansive system of gyri on the brain’s surface. The organization amongst mushroom body neurons reaches extreme elaboration throughout its constituent neuropils. The calyces, columns, and especially the gyri show DC0 immunoreactivity, an indicator of extensive circuits involved in learning and memory.
... Anti-DCO and phalloidin. Polyclonal antibody (generously provided by Dr. Daniel Kalderon; see Lane and Kalderon, 1993) raised against DCO (the catalytic subunit of Drosophila cAMP-dependent protein kinase) reliably labels all Kenyon cell subpopulations in the mushroom bodies of neopteran insects (Farris and Sinakevitch, 2003;Farris and Strausfeld, 2003;Farris et al., 2004). The anti-DCO polyclonal antibody was generated by cloning the DCO cDNA sequence into a pAR3040 T7 expression vector (Studier and Moffatt, 1986); then, after purification of the protein from sedimentation of inclusion bodies, this was used to immunize rabbits (Lane and Kalderon, 1993) and the final antibody was purified using DCO protein immobilized on nitrocellulose (Harlow and Lane, 1988). ...
Article
Full-text available
In most insects with olfactory glomeruli, each side of the brain possesses a mushroom body equipped with calyces supplied by olfactory projection neurons. Kenyon cells providing dendrites to the calyces supply a pedunculus and lobes divided into subdivisions supplying outputs to other brain areas. It is with reference to these components that most functional studies are interpreted. However, mushroom body structures are diverse, adapted to different ecologies, and likely to serve various functions. In insects whose derived life styles preclude the detection of airborne odorants, there is a loss of the antennal lobes and attenuation or loss of the calyces. Such taxa retain mushroom body lobes that are as elaborate as those of mushroom bodies equipped with calyces. Antennal lobe loss and calycal regression also typify taxa with short nonfeeding adults, in which olfaction is redundant. Examples are cicadas and mayflies, the latter representing the most basal lineage of winged insects. Mushroom bodies of another basal taxon, the Odonata, possess a remnant calyx that may reflect the visual ecology of this group. That mushroom bodies persist in brains of secondarily anosmic insects suggests that they play roles in higher functions other than olfaction. Mushroom bodies are not ubiquitous: the most basal living insects, the wingless Archaeognatha, possess glomerular antennal lobes but lack mushroom bodies, suggesting that the ability to process airborne odorants preceded the acquisition of mushroom bodies. Archaeognathan brains are like those of higher malacostracans, which lack mushroom bodies but have elaborate olfactory centers laterally in the brain.
... However, we want to point out here that a systematic, quantitative account of the dendritic branching patterns within and across the two types of KCs is still missing and requires future attention. Earlier investigations have also categorized KCs depending on the position and size of their cell bodies, which relates to the developmental trajectory of the different MB neuroblasts [27,61,62]. The outermost KCs (called outer compact cells) are born first, whereas the innermost ones (inner compact cells) are born last. ...
Article
Full-text available
Mushroom bodies (MBs) are multisensory integration centers in the insect brain involved in learning and memory formation. In the honeybee, the main sensory input region (calyx) of MBs is comparatively large and receives input from mainly olfactory and visual senses, but also from gustatory/tactile modalities. Behavioral plasticity following differential brood care, changes in sensory exposure or the formation of associative long-term memory (LTM) was shown to be associated with structural plasticity in synaptic microcircuits (microglomeruli) within olfactory and visual compartments of the MB calyx. In the same line, physiological studies have demonstrated that MB-calyx microcircuits change response properties after associative learning. The aim of this review is to provide an update and synthesis of recent research on the plasticity of microcircuits in the MB calyx of the honeybee, specifically looking at the synaptic connectivity between sensory projection neurons (PNs) and MB intrinsic neurons (Kenyon cells). We focus on the honeybee as a favorable experimental insect for studying neuronal mechanisms underlying complex social behavior, but also compare it with other insect species for certain aspects. This review concludes by highlighting open questions and promising routes for future research aimed at understanding the causal relationships between neuronal and behavioral plasticity in this charismatic social insect.
... While the question remains open whether the observed differences between young and older honeybees are a result of a predetermined maturation process or of sensory experience, these findings are in line with patterns of development in other structures [67,68] and indicate a refinement of connectivity and cellular response properties for reliable encoding of the information conveyed in the waggle dance. ...
Article
Full-text available
Since the honeybee possesses eusociality, advanced learning, memory ability, and information sharing through the use of various pheromones and sophisticated symbol communication (i.e., the “waggle dance”), this remarkable social animal has been one of the model symbolic animals for biological studies, animal ecology, ethology, and neuroethology. Karl von Frisch discovered the meanings of the waggle dance and called the communication a “dance language.” Subsequent to this discovery, it has been extensively studied how effectively recruits translate the code in the dance to reach the advertised destination and how the waggle dance information conflicts with the information based on their own foraging experience. The dance followers, mostly foragers, detect and interact with the waggle dancer, and are finally recruited to the food source. In this review, we summarize the current state of knowledge on the neural processing underlying this fascinating behavior.
... In the present study, we investigated the mechanism of action of the mixture composed of a pyrethroid, permethrin and a neonicotinoid, dinotefuran on cockroach cholinergic synaptic transmission and isolated neurons expressing nAChR subtypes. More precisely, we used dorsal unpaired median (DUM) neurons which are neurosecretory cells generating endogenous spontaneous action potentials (Grolleau and Lapied, 2000) and mushroom body Kenyon cells involved in learning and memory processes (Farris et al., 2004). We found that the mixture increases ganglionic depolarization and ionic currents, suggesting that permethrin enhances the excitatory activity and the agonist effect of dinotefuran. ...
Thesis
L’utilisation intensive des insecticides pour lutter contre les insectes ravageurs de culture et vecteurs demaladies, a conduit à des polémiques sur le mode d’action des insecticides. Ces polémiques sont liéesau fait que le mode d’action des insecticides, notamment des néonicotinoïdes est mal connu. Ils agissentprincipalement sur les récepteurs à l’acétylcholine (ACh) de type nicotinique (nAChR) qui jouent un rôlefondamental dans la transmission synaptique cholinergique. Bien que ces récepteurs soient bien décritschez les mammifères, très peu d’études ont évalué l’effet des néoniotinoïdes sur un récepteur natifd’insecte.Au cours de cette thèse, nous avons pour la première fois exprimé en ovocytes de xénope un récepteurhomomérique ⍺7 de blatte et étudié ces propriétés pharmacologiques vis-à-vis des néonicotinoïdes,comparativement au récepteur a7 de rat. Nos résultats révèlent un récepteur atypique qui est insensibleà l’⍺-bungarotoxine et qui n’est pas activé par les néonicotinoïdes. Ainsi, bien que les gènes codantpour les sous-unités α7 de blatte et de rat forment un groupe monophylétique distinct des autres sousunitésd’insectes et de mammifères, les récepteurs homomériques obtenus semblent avoir despropriétés pharmacologiques différentes. Parallèlement, nous avons étudié les propriétéspharmacologiques des nAChR natifs et notamment l’effet modulateur d’un pyréthrinoïde, la permethrine,sur les courants induits par le dinotefurane. Ce travail a permis d’évaluer le mode d’action d’unantiparasitaire, le Vectra 3D. Enfin, nous avons également entrepris de développer la techniqued’extraction membranaire afin de l’utiliser comme alternative pour étudier le mode d’action desinsecticides.Pour conclure, cette thèse a permis une avancée sur l’étude de la caractérisation des propriétéspharmacologiques des récepteurs nicotiniques neuronaux des insectes et sur l’étude du mode d’actiondes insecticides néonicotinoïdes.
Preprint
Full-text available
Neural organization of mushroom bodies is largely consistent across insects, whereas the ancestral ground pattern diverges broadly across crustacean lineages, resulting in successive loss of columns and the acquisition of domed centers retaining ancestral Hebbian-like networks and aminergic connections. We demonstrate here a major departure from this evolutionary trend in Brachyura, the most recent malacostracan lineage. Instead of occupying the rostral surface of the lateral protocerebrum, mushroom body calyces are buried deep within it, with their columns extending outwards to an expansive system of gyri on the brain’s surface. The organization amongst mushroom body neurons reaches extreme elaboration throughout its constituent neuropils. The calyces, columns, and especially the gyri show DC0 immunoreactivity, an indicator of extensive circuits involved in learning and memory.
Article
Honeybees are social insects, and individual bees take on different social roles as they mature, performing a multitude of tasks that involve multi-modal sensory integration. Several activities vital for foraging, like flight and waggle dance communication, involve sensing air vibrations through their antennae. We investigated changes in the identified vibration-sensitive interneuron DL-Int-1 in the honeybee Apis mellifera during maturation by comparing properties of neurons from newly emerged adult and forager honeybees. Although comparison of morphological reconstructions of the neurons revealed no significant changes in gross dendritic features, consistent and region-dependent changes were found in dendritic density. Comparison of electrophysiological properties showed an increase in the firing rate differences between stimulus and nonstimulus periods in foragers compared with newly emerged adult bees. The observed differences in neurons of foragers compared with newly emerged adult honeybees suggest refined connectivity, improved signal propagation, and enhancement of response features possibly important for the network processing of air vibration signals relevant for the waggle dance communication of honeybees.
Preprint
Full-text available
Honeybees are social insects, and individual bees take on different social roles as they mature, performing a multitude of tasks that involve multi-modal sensory integration. Several activities vital for foraging, like flight and waggle dance communication, involve sensing air vibrations using antennae. We investigated changes in the identified vibration-sensitive interneuron DL-Int-1 in the honeybee Apis mellifera during maturation by comparing properties of neurons from newly emerged adult and forager honeybees. Comparison of morphological reconstructions of the neurons revealed minor changes in gross dendritic features and consistent, region dependent and spatially localized changes in dendritic density. Comparison of electrophysiological properties showed an increase in the firing rate differences between stimulus and non-stimulus periods in foragers compared to newly emerged adult bees. The observed differences in neurons of foragers as compared to newly emerged adult honeybees indicate refined connectivity, improved signal propagation, and enhancement of response features important for the network processing of air vibration signals relevant for the waggle-dance communication of honeybees.
Article
Antisera against the neuromodulatory peptides, Phe-Met-Arg-Phe-NH2-amide (FMRF-amide) and gastrin cholecystokinin, demonstrate that the mushroom bodies of honey bees are subdivided longitudinally into strata. Three-dimensional reconstructions demonstrate that these strata project in parallel through the entire pedunculus and through the medial and vertical lobes. Immunostaining reveals clusters of immunoreactive cell bodies within the calyx cups and immunoreactive bundles of axone that line the inside of the calyx cup and lead to strata. Together, these features reveal that immunoreactive strata are composed of Kenyon cell axons rather than extrinsic elements, as suggested previously by some authors. Sorting amongst Kenyon cell axons into their appropriate strata already begins in the calyx before these axons enter the pedunculus. The three main concentric divisions of each calyx (the lip, collar, and basal ring) are divided further into immunoreactive and immunonegative zones. The lip neuropil is divided into two discrete zones, the collar neuropil is divided into five zones, and the basal ring neuropil is divided into four zones. Earlier studies proposed that the lip, collar, and basal ring are represented by three broad bands in the lobes: axons from adjacent Kenyon cell dendrites in the calyces are adjacent in the lobes even after their polar arrangements in the calyces have been transformed to rectilinear arrangements in the lobes. The universality of this arrangement is not supported by the present results. Although immunoreactive zones are found in all three calycal regions, immunoreactive strata in the lobes occur mainly in the two bands that were ascribed previously to the collar and the basal ring. In the lobes, immunoreactive strata are visited by the dendrites of efferent neurons that carry information from the mushroom bodies to other parts of the brain. Morphologically and chemically distinct subdivisions through the pedunculus and lobes of honey bees are comparable to longitudinal subdivisions demonstrated in the mushroom bodies of other insects, such as the cockroach Periplaneta americana. The functional and evolutionary significance of the results is discussed. J. Comp. Neurol. 424:179-195, 2000. (C) 2000 Wiley-Liss, Inc.
Article
Le developpement preimaginal d'Apis mellifera a deja ete decrit par plusieurs chercheurs et l'existence de 5 stades larvaires consacres a l'alimentation et a la croissance est bien etablie. Le dernier stade larvaire peut etre subdivise en diverses periodes, qui impliquent une serie de modifications morphologiques et physiologiques en liaison avec la metamorphose. Plusieurs stades ont egalement ete observes pendant la phase nymphale. Diverses etudes ont montre que la duree des stades des phases larvaire et nymphale varie d'une race d'abeilles a l'autre. Nous avons etudie en detail ces 2 phases du developpement preimaginal chez des abeilles africanisees recoltees a Ribeirao Preto, SP, Bresil, et caracterise les divers stades ainsi que leur duree
Article
The corpora pedunculata, or mushroom bodies, are paired lobes of neuropile present in the protocerebrum or dorsal brain of all insects. They are divisible into three parts: calyx, stalk and roots. The latter usually comprise two simple lobes, the alpha and beta lobes. The corpora pedunculata of a variety of Lepidoptera were examined. All had a double calyx-cup. Each 'cup-cavity' is composed of 'globuli' cell bodies. The broad stalk, a tract of fibres and neuropile, leads from the calyx to the complex 'roots'-alpha, beta and gamma lobes. A third group of globuli cells near the calyx gives rise to a tract leading to a second lobe-system-the tripartite Y-lobe-in the roots. As neither the Y tract nor the Y lobe has been described before in any insect, their possible homologues are unknown. The two lobe systems in the roots are closely intertwined, yet have no interaction except in the gamma lobe. A number of different neuron types with branches in the mushroom bodies has been described from Golgi preparations. Some (intrinsic cells) divide in the calyx and again in the roots, but do not pass out of the mushroom bodies. Others (extrinsic cells) branch in the mushroom bodies and in other areas of the brain, thus connecting two regions. Intrinsic cells arise from cell bodies in the calyx-cups or posterior to them. There are two types: one has extensive spine-covered branches in the calyx, while the second has claw-like terminals covering a narrow cylindrical field. Processes from these cells run to the alpha, beta and gamma lobes via the stalk. A wide-field accessory cell, which arises from the third group of globuli cell bodies, also has claw-like endings in the calyx. A process of this cell runs in the Y-tract to the Y-lobe. Extrinsic terminals in the calyx arise from cells branching in the antennal lobe, in an accessory optic area in the protocerebrum, in the 'undifferentiated' protocerebral neuropile, or in the suboesophageal lobes. The antennal terminals in the calyx are knob-like. It is proposed that they form the centre of the 'glomeruli' typically present in calycal neuropile. The claws of the bunched intrinsic and accessory cells probably fit around these knobs. Within the stalk, different subvarieties of intrinsic cells have been distinguished on the basis of the distribution of the side-branches and spines which they bear. The stalk is thought to be the site of extensive postsynaptic interaction between intrinsic cells. Fibres in the stalk run in bundles or groups. All the fibres in one bundle are of the same subvariety. In the roots, the subvarieties of intrinsic cells have different branching patterns. The alpha and beta lobes are not homogeneous, but are divided into sublobes. Extrinsic fibres ramify only within one sublobe generally, though some have very large fields. The connexions of the roots are obscure. Some extrinsic fibres branch again in the 'undifferentiated' protocerebral neuropile; others, from the beta lobe, may run to the suboesophageal lobes. There are profound differences between the internal organization of the mushroom bodies in Hymenoptera (Kenyon 1896; Goll 1967) and Lepidoptera. The functional implications of the Lepidopteran form are discussed.
Article
The mushroom bodies of the bee are paired neuropils in the dorsal part of the brain. Each is composed of the arborizations of over 17 10^4 small interneurons of similar architecture called Kenyon cells. Golgi staining demonstrates that these neurons can be divided into five groups distinguished on the basis of their dendritic specializations and geometry. The mushroom body neuropils each consist of a pair of cup-shaped structures, the calyces, connected by two short fused stalks, the pedunculus, to two lobes, the α- and β-lobes. Each calyx is formed from three concentric neuropil zones, the basal ring, the collar and the lip. The calyces are organized in a polar fashion; within the calyces each of the five categories of Kenyon cell has a distribution limited to particular polar contours. The dendritic volumes of neighbouring Kenyon cells arborizing within each individual contour are greatly overlapped. Fibres from groups of neighbouring cells within a calycal contour are gathered into bundles that project into the pedunculus, each fibre dividing to enter both the α- and β-lobes. The pedunculus and the lobes are conspicuously layered. Kenyon cells with neighbouring dendritic fields within the same calycal contour occupy a single layer in the pedunculus and lobes. Thus the two-polar organization of the calyces is transformed into a Cartesian map within the pedunculus, which continues into the α- and β-lobes. The calyx receives input fibres from both the antennal lobes and the optic neuropils. The branching patterns of these cells reflect the polar organization of the calyces as their terminals are restricted to one or more of the three gross compartments of the calycal neuropil. The course of these tracts and the morphologies of the fibres that they contain are described. Cells considered to represent outputs from the mushroom bodies arborize in the pedunculus and α- and β-lobes. Generally the arborizations of the output neurons reflect the layered organization of these neuropils. Fibres from the two lobes run to the anterior median and lateral protocerebral neuropil, and the anterior optic tubercle. Additionally there is an extensive network of feedback interneurons that inter-connect the α- and β-lobes with the ipsi- and contralateral calyces. Many individual neurons have branches in both the α- and the β-lobes and in the pedunculus. The pathways and geometries of the fibres subserving the two lobes are described. The hypothesis of Vowles (1955) that the individual lobes represent a separation of sensory and motor output areas is shown to be incorrect. The anatomy of the bee's mushroom bodies suggests that they process second-order antennal and fourth- and higher-order visual information. The feedback pathways are discussed as possible means of creating long-lasting after-effects which may be important in complex timing processes and possibly the formation of short-term memory.
The postembryonic development of the corpora pedunculata, deutocerebrum and tritocerebrum of Phormia regina Meigen was studied using reduced silver stains to provide detailed observations and determine the relationships between larval and imaginal neurones. The imaginal ganglion cells of these areas are formed from isolated neuroblasts by a series of asymmetric and symmetric divisions. The imaginal corpora pedunculata are formed by 10 isolated neuroblasts located dorsal to the larval calyces. The imaginal antennal ganglia are formed by 2 isolated neuroblasts. These ganglia lie dorsal to the larval antennal ganglia which degenerate during the late larval and early pupal stages. The glomerular structure of the imaginal ganglia develops during the pupal stage after receiving antennal sensory fibers. The tritocerebrum of the imaginal brain is formed by isolated neuroblasts.The cellular and neuropilar components of the larval brain degenerate during the late larval and early pupal stages. The degenerating larval ganglion cells are characterized by pycnotic nuclei and the neuropile by a ragged, darkly stained appearance.The larval and adult brains of Phormia are discrete morphological entities as revealed by the degeneration of the larval brain, concomitant with its replacement by corresponding imaginal elements.