www.elsevier.com/locate/brainres

Brain Research 999 (2004) 9–19

Research report

The epithalamus of the developing and adult frog: calretinin expression

and habenular asymmetry in Rana esculenta

Vittorio Guglielmottia,*, Luigia Cristinoa, Errico Sadaa, Marina Bentivogliob

a Institute of Cybernetics, CNR, ‘‘E. Caianiello’’, I-80078 Pozzuoli, Naples, ItalybDepartment of Morphological and Biomedical Sciences, University of Verona, Verona, Italy

Accepted 14 October 2003

Abstract

Expression of the calcium binding protein (CaBP) calretinin (CR) was studied with immunohistochemistry in the pineal complex and

habenular nuclei (HN) of the developing and adult frog Rana esculenta. The frog pineal complex is a medial structure formed by two

interconnected components, the frontal organ and the pineal organ or epiphysis; the habenular nuclei are bilateral and are asymmetric due

to subdivision of the left dorsal nucleus into medial and lateral components. In the pineal complex, calretinin immunostaining of cells and

fibers was consistently observed in developing and adult frogs. In the habenulae, calretinin immunoreactivity exhibited instead marked

variations during development, and was expressed only in cells of the medial subnucleus of the left dorsal habenula. In particular,

calretinin was detected at larval stages, peaked during metamorphosis, was markedly downregulated at the end of metamorphosis, and was

evident again in adulthood. This sequence of calretinin expression was confirmed by quantitative analysis of immunoreactive cells in the

left habenula. In tadpoles, calretinin-positive cells exhibited a dorsoventral gradient of density, while in adulthood, they were distributed

throughout the dorsoventral extent of the medial subnucleus. The study demonstrates a peculiar developmental pattern, with transient

downregulation, of asymmetric calretinin expression in the frog epithalamus. The findings indicate that calcium and calcium buffering

systems may play critical roles in neurogenetic and neuronal migration processes implicated in the formation of the asymmetric habenular

portion in amphibians. In addition, the reappearance of calretinin expression in the adult frog supports a distinct functional role of the

asymmetric habenular component in amphibians.

D 2004 Elsevier B.V. All rights reserved.

Theme: Development and regeneration

Topic: Cell differentiation and migration

Keywords: Asymmetry; Calretinin; Development; Epithalamus; Frog; Immunocytochemistry

1. Introduction circuits [10]. The organization of the epiphysis exhibits

The habenular nuclei (HN) and the epiphysis are dien-

cephalic structures located medially in the epithalamus of all

vertebrates. The HN are connected with the interpeduncular

nucleus forming the habenulo-interpeduncular system,

which is one of the circuits most conserved in phylogenesis.

In the vertebrate brain, the epithalamus plays a main

functional role in the regulation of cyclic behavior, such

as circadian and reproductive functions, as well as in the

modulation of other systems, and in particular limbic

0006-8993/$ - see front matter D 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.brainres.2003.10.053

* Corresponding author. Tel.: +39-81-8675134; fax: +39-81-8675128.

E-mail address: vigu@biocib.cib.na.cnr.it (V. Guglielmotti).

marked variations in different taxa. Thus, in cyclostomes,

the epiphysis evolves as pineal complex, e.g. pineal organ

proper and parapineal organ, to turn into a single epiphyseal

outgrowth, the pineal gland, in mammals. The HN include

bilaterally two nuclei protruding into the third ventricle and

interconnected by the habenular commissure.

Marked structural asymmetry of the HN represents a

striking feature of the epithalamus in several non-mamma-

lian species [9,16,27,30,31,34,49], and of the pineal com-

plex in most anamniotes and reptiles [17,18,59,60]. In

particular, the comparison between the left and right HN

in the brain of many vertebrates pointed out structural

differences in size, neuronal organization, neurochemistry

and connectivity between the two sides, raising a debate

V. Guglielmotti et al. / Brain Research 999 (2004) 9–1910

about the evolutionary origin of epithalamic structures (see

Ref. [13] for review). Recent data have demonstrated that in

the zebrafish diencephalon, the left–right identity is influ-

enced by the parapineal organ [22].

In the frog, the HN are subdivided into dorsal and ventral

nuclei, homologous to the medial and lateral nuclei, respec-

tively, in mammals. Morphological asymmetries have been

reported in the left dorsal habenular nucleus of Rana

esculenta [32] and they correlate with unilateral projections

of the left habenula to the frontal organ [33], which is the

extracranial component of the pineal complex. The func-

tional significance of such asymmetric organization of the

frog HN is, however, still unclear.

Asymmetric neurochemical features have also been de-

scribed in the frog epithalamus. In particular, selective left–

right differences in the expression of nitric oxide synthase

(NOS), the synthetic enzyme of the free radical nitric oxide,

have been reported in the HN and pineal complex of R.

esculenta during development [26]. Asymmetry of calreti-

nin (CR) immunoreactivity was also mentioned in the HN of

adult frogs [41]. This latter feature indicates differences in

calcium buffering systems between the two sides of the frog

brain, because CR is a calcium binding protein (CaBP)

which acts in the control of the fluctuation of cytosolic

calcium ions. Genes coding for CR are well preserved

during vertebrate evolution [43] and CR is the main EF-

hand CaBP expressed in regions of the frog nervous system

[21]. Fluctuations of calcium ions are involved in many

neuronal functions (see Ref. [39] for review). In particular,

intracellular calcium buffering levels represent a key pa-

rameter during neural development, and a direct relationship

between calcium and neuronal differentiation has been

suggested [2,3,25,40].

Transient expression of CR during development has been

reported to play a primary role in the reshaping of the

habenulo-interpeduncular system in the trout [44]. However,

no data are available on the distribution of CR in the

developing HN of other vertebrates, and, in particular in

the developing and mature pineal complex of amphibians.

On this basis, the present work was aimed at investigating

the ontogenesis of CR expression in the frog epithalamus,

and at a correlation of such data with previous findings

obtained in this brain region.

2. Materials and methods

2.1. Animals

Tadpoles and adult frogs of the species R. esculenta

were used. Fertilized spawn (available during spring) and

adult animals (25–30 g body weight, corresponding to

about 3 years of age) were collected from the countryside

around Naples. The spawn were bred at room temperature

according to a previously described protocol [24]. The

collection of the material was authorized by the competent

Italian authorities, and the experiments were conducted

under institutional approval. All efforts were made to

minimize animal suffering and the number of animals

used.

Following the staging criteria of Manelli and Margar-

itora [38] for R. esculenta, we used tadpoles of stages 31,

34, 38, 40, 44, 48, and 50 (stage 50 is the last one).

These tables [38] subdivide the development of the frog

in three periods: the embryonic and larval periods, and

metamorphosis. Briefly: (i) the embryonic period includes

stages 1–30, related to the use of deutoplasm as nutritive;

(ii) the larval period, including stages 31 and 32, is

characterized by autonomous feeding and appearance of

the hindlimb buds; (iii) metamorphosis includes stages

33–50: tadpoles develop mainly the body and hindlimbs

until stage 44, followed by the appearance of forelimbs,

atrophy of the tail, and capability also of terrestrial life in

stages 45–50.

Due to difficulties in the supply and breeding of post-

metamorphic frogs, juvenile animals were not available.

2.2. Tissue preparation for immunohistochemistry

For CR imunohistochemistry, tadpoles sampled at the

above-mentioned stages (n = 4 per stage) were anesthetized

in a 0.2 g/l solution of tricaine methanesulfonate (MS 222;

Sigma, St. Louis, MO) and fixed intracardially with 4%

paraformaldehyde and 2% picric acid (salt) in 0.1 M

phosphate buffer, pH 7.4 (PB). The perfusion was per-

formed by injecting the fixative through a glass micropipette

(tip diameter: 60–70 Am), connected with a plastic tube to a

syringe. The viscera were removed after perfusion, and

tadpoles at stages 31–34 were processed as wholemounts.

In animals at stages 38–50, undesired tissue was removed

by transverse cuts through the caudal tip of the spinal cord

and then longitudinally at level of the eyes; the head,

including the frontal organ was thus dissected. Adult frogs

(n = 4) were anesthetized in 0.35 g/l solution of MS 222 and

perfused through the aorta with 10 ml of 0.64% NaCl

followed by 25 ml of the same fixative used for tadpoles.

The brain and the flap of skin containing the frontal organ

were then dissected out.

The tissue samples were postfixed for 4 h at room

temperature in the same fixative used for perfusion. The

excess of picric acid was washed out by several changes in

PB, and the tissue samples were then soaked in 30% sucrose

in PB overnight at 4 jC for cryoprotection. They were

finally embedded in a solution of 6% agar and 30% sucrose

in PB for cutting. Horizontal (n = 3 animals per age group)

and transverse serial sections (n = 1 animal per age group)

were cut on a cryostat at a 14-Am thickness. All sections

were collected, mounted in two adjacent series on glass

slides pretreated with Vectabond (Vector, Burlingame, CA),

and processed for CR immunohistochemistry. After the

immunoreaction, one of the series of horizontal sections

was counterstained with cresyl violet.

V. Guglielmotti et al. / Brain Research 999 (2004) 9–19 11

Due to the anatomical features of the pineal organ, which

is flattened on top of the habenular commissure, the hori-

zontal sections provided the largest surface of epithalamic

structures, allowing also a detailed study of the most dorsal

levels of the HN.

2.3. Tissue preparation for Nissl staining

Adult frogs (n = 2) and tadpoles at the same stages used

for immunohistochemistry (n= 2 per stage) were destined

to the cytoarchitectonic study in cresyl violet-stained

sections. These animals were fixed and dissected using

the same procedure described above. The tissue samples

were postfixed for 3 h at room temperature, washed in PB,

dehydrated, embedded in paraffin and cut at 15 Am in

serial sections, that were collected on slides pretreated with

Vectabond. One animal per age group was cut in the

horizontal plane and the other in the transverse plane.

2.4. CR immunohistochemistry

The sections destined to immunohistochemistry,

mounted on slides, were immersed in 0.75% H2O2 in Tris

buffer saline at pH 7.3 (TBS) to block endogenous

peroxidase activity. After rinsing in TBS, they were

preincubated in 10% normal goat serum (NGS) and

0.4% Triton-X100 in TBS for 1 h, followed by incubation

for 3 days at 4 jC in rabbit anti-CR (Swant, Bellinzona,

Switzerland), diluted 1:2000 in the preincubation solution.

The specificity of this antibody for the frog brain was

previously assessed by Western blot analysis [41].

After incubation in primary antibody, the sections were

rinsed and incubated for 4 h at room temperature in

secondary biotinylated goat anti-rabbit IgG (Vector), dilut-

ed 1:200 in the preincubation solution. After several rinses

in TBS, the sections were incubated for 2 h in avidin–

biotin complex (Vector), diluted in TBS according to the

supplier’s indications, followed by rinses in TBS and by

reaction with 3–3Vdiaminobenzidine (Sigma) and H2O2 for

4–5 min. The sections were then rinsed and, as men-

tioned above, one of the horizontal series per animal was

counterstained with Nissl staining. Finally, the sections

were dehydrated, cleared in xylene and coverslipped with

DPX.

For control, transverse sections randomly collected dur-

ing the cutting procedure were treated with the same

protocol but omitting the primary antibody. No immunopo-

sitivity was found in these sections.

2.5. Nissl staining

For Nissl staining, the sections were stained with 0.1%

cresyl violet (Merck) in 0.1 M acetate–acetic acid buffer,

pH 3.7. The sections were then differentiated in alcohol,

dehydrated, cleared in xylene and coverslipped with

DPX.

3. Data analysis

All the material was examined at the microscope under

bright-field illumination. The proportion of CR-immuno-

positive cells was evaluated in the left dorsal habenula in

all the immunoreacted and counterstained sections of

developing and adult animals (n = 3 per age group). All

CR-immunoreactive neurons containing a darkly stained

nucleolus visible in the focal plane, as well as immuno-

negative cells exhibiting darkly stained nucleoli, were

counted, using a 40� objective, in the medial subnucleus

of the left dorsal habenula, in which CR-immunostained

cells were concentrated (see below). In addition, to make

sure that cell counting was not affected by bias in the

delimitation of the subnuclei of the left habenula in

different sections and animals, cells were counted in both

the components (the medial and lateral subnuclei) of the

left dorsal habenula.

4. Results

4.1. Asymmetry of the mature and developing habenulae

The data on habenular asymmetry observed in the

present investigation in cresyl violet-stained sections were

consistent with those described in the adult and developing

frog in previous studies [9,26], and will be briefly recalled

here for clarity. The asymmetry of the adult frog HN is

due to peculiarities of the left dorsal nucleus (Fig. 1A),

which is subdivided into medial and lateral subnuclei,

represented by two fused shells of cells surrounding a

central neuropil, while a single dorsal nucleus is located on

the right side of the HN. In R. esculenta, the rostromedial

part of the medial subnucleus is further subdivided by a

septum in two distinct compartments, named medial and

lateral compartment or neuropil (Fig. 1B). As illustrated in

Fig. 1A, the mediocaudal portion of the medial subnucleus

is formed by a neuropil wrapped in a shell of cells.

A similar cytoarchitectonic organization was observed

in the dorsal HN of developing animals. In the left

nucleus, the different compartments developed sequentially

(Fig. 1C–E). The left medial habenular subnucleus and the

right nucleus were the first structures appearing in the

epithalamus during development. The left lateral subnu-

cleus appeared at stage 30 but became prominent only

from stage 34 onward (Fig. 1C), and developed further

until the end of metamorphosis (Fig. 1E).

4.2. CR immunoreactivity in the epithalamus

As it will be presented in detail below, throughout the

examined developmental stages and in adulthood, CR

immunoreactivity was observed in cells and fibers of the

pineal complex. CR exhibited instead a peculiar pattern of

expression in the HN, since it was detected only in cells of

Fig. 1. Photomicrographs of Nissl-stained transverse (A–D) and horizontal (E) sections through the habenular nuclei at different rostrocaudal levels of adult

(A,B) and developing (C–E) frogs. (A) Section through the middle portion, showing the features of habenular asymmetry in the adult frog: the left dorsal

habenular nucleus comprises the medial (M) and lateral (L) subnuclei, whereas a single nucleus (R) is seen on the right side. (B) Section through the

rostromedial level, showing that a septum subdivides further the medial subnucleus into medial (m) and lateral (l) compartments. C–E, in which the circled

numbers indicate the developmental stage, show the sequence of cytoarchitectural arrangement of the medial subnucleus into medial and lateral compartments,

as well as the development of the lateral subnucleus (arrowhead in C; L in D and E) which starts from stage 31. III, third ventricle; hc, habenular commissure.

Scale bars: 100 Am.

V. Guglielmotti et al. / Brain Research 999 (2004) 9–1912

the left dorsal habenula, and mainly in those of the medial

compartment, with temporal variations.

4.2.1. Stages 31, 34

CR immunoreactivity was found in cells of the medial

subnucleus of the left dorsal habenula since the first

examined stage (Fig. 2A), when the lateral subnucleus

was not yet recognizable. At stage 31, intense CR

immunostaining of cells and fibers was observed in the

pineal organ and in the frontal organ (Fig. 2B, C). In the

latter structure, CR positivity was evident in cells of

different sizes, including relatively large elements (6.5–

11.5 Am in diameter) that exhibited intense immunostain-

ing of cell bodies and their processes (Fig. 2C). The

pineal tract also appeared intensely stained in the epiph-

ysis (Fig. 2B).

At subsequent stages, e.g. stage 34, the organization of

the HN progressed towards a mature configuration, and CR

immunoreactivity was evident in clusters of cells located in

the left habenula and in the adjacent pineal region, where

labeling of cells and of pineal tract fibers was observed in

the epiphysis (Fig. 2D). At stage 34, the percentage

(F standard deviation) of CR-immunoreactive neurons was

4.1F 0.4% of the cells of the medial subnucleus, and

3.6F 0.2% of the total population of cells of the left dorsal

habenula (Fig. 3). It should be reminded that at this stage,

the medial subnucleus occupied most of the extent of the

dorsal habenula.

4.2.2. Stage 38

Compartmentalization of the left dorsal habenula was

well evident at this stage (Fig. 2F). CR immunoreactivity,

clearly asymmetric, was evident in cells of the medial

compartment of the medial subnucleus close to the third

ventricle (Fig. 2E, F). Cells exhibiting various degrees of

intensity of CR immunoreactivity were observed (Fig. 2E,

F). At this stage, the proportion of CR-immunoreactive

neurons was higher than at stage 34, accounting for

10.0F 0.8% of the cells of the medial subnucleus and

for 8.7F 0.6% of the cells of the entire left dorsal

habenula (Fig. 3).

Intense CR immunopositivity of cells and fibers was still

observed in the extra- and intracranial portions of the pineal

complex, including the pineal tract.

Fig. 2. CR immunoreactivity in the left dorsal habenula and pineal complex during development in horizontal sections at different stages (indicated by

circled numbers). (A) At stage 31, a few labeled cells (arrow) are clustered in the dorsal region of the medial habenular subnucleus (M). (B) CR-

immunostained cells and fibers are evident in the pineal organ (arrows), from which fibers of the pineal tract depart (arrowhead). (C) CR-

immunostained cells bodies and their processes are also evident in the frontal organ (FO). (D) At stage 34, the pineal organ (arrow), pineal tract

(arrowhead) and cells in the dorsal portion of the medial habenular subnucleus (M) are the only CR-immunopositive structures of the epithalamic

region. (E) At stage 38, cells at the most dorsal level of the medial habenular compartment (m) are intensely immunolabeled. (F) At the ventral level

of the same sample, cells of the medial neuropil (m) lining the third ventricle (III) exhibit moderate immunoreactivity. (G) At stage 40, CR

immunolabeling of cells (arrowhead) is evident in the medial subnucleus (M), grouped near the third ventricle (III), and in the pineal organ (arrow).

(H) Higher power image of intensely (white arrows) and moderately (arrowheads) CR-positive cells in the medial compartment (m). (I) At stage 44,

the low power view of the epithalamus shows CR immunoreactivity on the left side (arrowhead), while the right habenula (R) appears devoid of

labeling; the arrows point to the pineal organ. (J): Intense immunostaining of cells at the most dorsal level of the medial subnucleus. (K) Intense

immunolabeling of cells and fibers in both the medial subnucleus (M; arrowhead) and pineal organ (arrows). Scale bars: 100 Am in A–F; 50 Am in

G–K.

V. Guglielmotti et al. / Brain Research 999 (2004) 9–19 13

4.2.3. Stages 40, 44

The pattern of CR immunoreactivity observed at stage

40 in the HN and pineal organ was similar to that

described above. In the habenular medial compartment,

immunostained cells were still mainly grouped near the

third ventricle (Fig. 2G, H), exhibiting clear asymmetry. In

the pineal complex, immunolabeling was observed in cells

and fibers.

CR labeling of both the left dorsal habenula (Fig. 2J)

and pineal complex (Fig. 2I, K) was also evident at stage

44, so that the features of immunoreactivity of these two

structures, very close one to the other, were still very

similar (Fig. 2I, K).

The relative proportion of CR-immunoreactive neurons

in the left habenula peaked at stage 40 (17.8F 0.5% in the

medial subnucleus, and 9.4F 0.6% in the entire left dorsal

Fig. 3. Bar diagram illustrating the percentage (F standard deviation) of

CR-immunoreactive cells with respect to the total cells population of the

medial subnucleus (black bars) or the entire left dorsal habenula (white

bars) at different stages of development and in adulthood (A).

V. Guglielmotti et al. / Brain Research 999 (2004) 9–1914

habenula), and decreased at stage 44 (12.9F 0.6%, and

6.5F 0.3%, respectively) (Fig. 3).

4.2.4. Stages 48, 50

At stage 48, the HN exhibited mature cytoarchitectonic

features (Fig. 4C). Interestingly, CR immunoreactivity was

found to be drastically downregulated in the left habenula at

this stage (Fig. 4A, C), while it was still present in the pineal

complex (Fig. 4B). In the HN, CR labeling was detected

only in a few cells located in the dorsal region of the medial

III cp

A

M

III

B

CPO

48

Fig. 4. (A–C) CR immunoreactivity at stage 48 in horizontal sections. (A) Sectio

positive cells (arrowheads) in the medial habenular subnucleus (M). Labeled fibe

several cells are intensely labeled. Immunostained fibers of probable commissural

through the pineal organ (large arrow) showing immunostaining of the frontal ne

along the choroid plexus (cp) of the third ventricle (III), and enters the brain at th

indicated by arrowhead). (C) CR-immunostained and Nissl-counterstained section

at the border of the third ventricle (III) are evident in the medial subnucleus (M). No

the boxed regions. Scale bars: 100 Am.

subnucleus (Fig. 4A, C). Fibers located at the border of the

left dorsal habenula, of probable commissural origin, were

also CR-stained (Fig. 4A). The proportion of immunoreac-

tive neurons had decreased further with respect to the

previous stages, accounting at stage 48 for 10.5F 0.7% of

the cells of the medial subnucleus and 6.1F 0.4% of those

of the left habenula (Fig. 3).

At stage 50, a few CR-positive cells were observed in the

most dorsal part of the left habenula (Fig. 5A, B), while no

CR labeling was observed in the other portions of the HN.

CR immunoreactivity was still observed in neurons and

fibers of the pineal complex (Fig. 5A, C). In the frontal

organ (Fig. 5C), some cells were devoid of labeling, while

stained fibers of epiphyseal neurons located at the border of

this structure were oriented toward the adjacent left dorsal

habenula (Fig. 5B). The percentage of CR-immunoreactive

neurons dropped at this stage to 6.8F 0.6% in the medial

subnucleus and to 3.6F 0.3% with respect to the total cell

population of the left dorsal habenula (Fig. 3).

4.2.5. Adulthood

Asymmetric expression of CR was detected in the adult

animals in the medial subnucleus of the left dorsal habenula

(Fig. 5D, F). Most of the immunoreactive cells were intensely

stained. The CR-immunostained cells were mainly clustered

in the habenular region bordering the third ventricle (Fig. 5F),

and they accounted for 14.3F 0.8% of cells of the medial

subnucleus and for 7.8F 0.6% of the total cell population of

the left habenula (Fig. 3). No CR imunoreactivity was found

in the lateral subnucleus and in the right habenula (Fig. 5D,

E). Intensely stained cells and their processes were also

observed in the pineal complex (Fig. 5G, H).

M

LIII

cp

n through the left dorsal habenula and pineal organ showing the paucity of

rs of the frontal nerve (large arrow) enter the pineal organ (PO), in which

origin (small arrows) surround the habenular subnucleus. (B) Dorsal section

rve (small arrow) that runs rostrocaudally (from left to right in the image)

e level of the habenular commissure (which lies under the black skin strip

through the left dorsal habenula: a few positive cells (arrows) located mainly

te the difference in the size of cells of the medial and lateral (L) subnuclei in

Fig. 5. CR-labeling in horizontal (A–G) and transverse (H) sections through the left dorsal habenula and pineal complex at the end of metamorphosis (stage 50;

A–C) and in adulthood (indicated by the letter A; D–H). (A) Dorsal level of the habenula showing a few immunostained cells (white arrows) in the medial

subnucleus (M). The pineal organ (PO) and the frontal nerve (arrow) are intensely positive. (B) Higher magnification of the area boxed in A, showing lightly

stained but well evident processes of pineal cells (arrowheads) oriented toward the left habenula at the border between the two regions. (C) Labeling of cells

and their processes in the frontal organ (FO). (D) Intensely labeled cells and processes in the medial subnucleus (M). (D, E) The lateral subnucleus (L) and the

right nucleus (R) are devoid of labeling. (F) Ventral level of the medial subnucleus (M) of the dorsal left habenula illustrating the intense staining of cells

mainly lining the third ventricle (III). Both D and F show habenular sections at more ventral levels than those shown for tadpoles. (G, H) CR immunoreactivity

in the frontal organ and pineal organ, respectively; the arrowhead in H points to an immunolabeled pineal cell and its process. Scale bars: 100 Am in A, D–F;

20 Am in B; 50 Am in C, G, H.

V. Guglielmotti et al. / Brain Research 999 (2004) 9–19 15

4.2.6. Overall cytoarchitectonic arrangement of CR-immu-

noreactive neurons in the left dorsal habenula

In addition to the developmental sequence of expression

described above, it is noteworthy that the topographical

distribution of the asymmetric CR positivity was different in

the left habenula of developing and adult animals (Fig. 6).

During development, intensely CR-immunostained cells

were packed in the left dorsal habenula along a dorso-

ventral gradient of density and, as described above, they

were mainly grouped close to the third ventricle (Fig.

2D, E, H, J). In the adult frog, intensely CR-immuno-

reactive neurons, and their processes oriented toward the

Fig. 6. Sketch summarizing the distribution of CR-containing cells (dots), concentrated in the left medial subnucleus (M; L indicates the lateral subnucleus) of

the dorsal habenula of developing and adult frogs. (A) At developmental stages 34–48, CR-containing cells are distributed along a dorsoventral gradient of

density and located in the medial compartment (m) and in the habenular region bordering the third ventricle (III). (B) In the adult frog, CR-containing cells are

scattered throghout this nucleus with a prevalence in its periventricular portion.

V. Guglielmotti et al. / Brain Research 999 (2004) 9–1916

central neuropil, were distributed through the dorsoven-

tral axis of the medial subnucleus (Fig. 5D, F). These

differences in the distribution of CR immunoreactivity

were not paralleled by major cytoarchitectonic changes in

the left habenula, in which, in particular, no clustering of

cell subsets was observed in the Nissl-stained sections

(Fig. 4C).

5. Discussion

The present study provides a detailed description of the

developmental pattern of expression of CR in the epitha-

lamus of R. esculenta. The results, summarized in Fig. 6,

demonstrate that CR is expressed in the pineal complex

and asymmetrically in the left dorsal habenular nucleus at

larval and metamorphic stages. CR immunoreactivity was

consistently detected in the pineal complex at subsequent

stages through adulthood, whereas it was strikingly tran-

sient in the left dorsal habenula. In this portion of the HN,

CR immunoreactivity was no longer evident at the final

steps of metamorphosis, while it was again well evident in

the adult animals, with a distribution different from that

observed during development (Fig. 6). The quantitative

analysis confirmed that the population of CR-immunore-

active cells in the left habenula peaked during metamor-

phosis (i.e. at stage 40), dropped at the end of this period

(stage 50), and in adult frogs was similar to the values

detected during metamorphosis. Therefore, the present data

suggest that a marked reshaping, involving selectively the

asymmetric habenular portion, occurs at late metamorphic

stages.

In general, CaBPs play potentially important roles in

neurogenesis [36] can be expressed in neurons during their

differentiation or when they become functional, and may

disappear when a functional system is disconnected [3].

Immunodetection of CR has repeatedly demonstrated tran-

sient features in the developing nervous system of verte-

brates [1,5,7,51,58] including humans [54] implying

selective roles of CR in neural development. In particular,

previous studies have reported the expression of CR in the

developing pineal gland of the rat and other mammals

[6,8,42,45,46], and transient or unusual expression of CR

in pineal cells has suggested an involvement of calcium not

only in mechanisms of signal transduction but also in other

calcium-mediated function(s) that remain to be elucidated.

In addition, CR expression has been used to characterize the

neural circuitry of the medial habenula in the rat, in which

asymmetry of synaptic contacts was described [56].

In the mature animal, our findings provide evidence of

asymmetry of CR expression in the HN of the adult R.

esculenta, extending a previous report in which this feature

was briefly mentioned [41], and adding the novel finding of

CR positivity in the pineal complex.

The present data related to a CaBP suggest in the frog an

intriguing chronology of calcium buffering systems, which

could underlie two sequential distinct physiological pro-

grams, operant during development and in the mature

animals, respectively. The first could be related to neuro-

genesis and neuronal migration, in which CR has been

V. Guglielmotti et al. / Brain Research 999 (2004) 9–19 17

repeatedly implicated [19,20,52,53]. These processes estab-

lish an intimate histogenetic and functional relationship

between the epiphysis and a part of it that migrates into

the habenular region to form the asymmetric portion of the

HN. The second program could be triggered after complete

differentiation of the epithalamus, when the asymmetric

portion of the HN has acquired functional independence.

Altogether the present results in the developing frog

suggest transient functional activity in the left HN, in

contrast with the persistence of parameters related to CaBP

expression in the pineal complex. These findings parallel

those observed in a previous study on NOS positivity in the

HN and pineal complex [26], indicating that the develop-

mental program of the frog epithalamus involves an array of

different molecules. In agreement with the previous results

obtained in the study of NOS, CR was here found to be

expressed during development in the left dorsal habenula

and mainly located in the medial neuropil, a distinct region

of the left medial subnucleus bordering the third ventricle.

Such variations in protein expression are likely to be

involved in the establishment of asymmetrical activity in

the diencephalon of amphibians, and could represent a

phylogenetic step leading to more complex lateralization

phenomena during brain evolution.

Very few data are available on the expression of CaBPs

in the adult and developing epithalamic structures of verte-

brates. In the HN of adult vertebrates, contradictory findings

have been reported on CaBP expression in the different

components of the HN, i.e. in the lateral and/or medial

habenula [11,23,29], but habenular asymmetry was not

reported in these studies. Asymmetric distribution of the

CaBP calbindin was described in the HN of the dogfish by

Rodriguez-Moldes et al. [47] who, however, did not inves-

tigate other epithalamic structures, such as the epiphysis.

Transient CR expression was reported in the habenular

complex of the developing trout, in which habenular asym-

metry was not described [44].

On the other hand, asymmetric transient expression of

neuroactive molecules in the HN is not an event occurring

only during development. Asymmetry of serotonin-like

immunoreactivity was described in the right habenula of

coho salmon during smolt transformation, a life period

linked to naturally occurring endocrine changes [15]. To-

gether with the present findings, these results indicate that

life conditions characterized by major behavioral and phys-

iological changes, e.g. the amphibian metamorphosis or the

smolt transformation in coho salmon, induce unilateral

variations in the habenular complex. It is also interesting

to note that in coho salmon a persistent asymmetric expres-

sion of serotonin was reported in the dorsal subdivision of

the left habenular nucleus and in the pineal organ [16]. The

present findings strikingly parallel such data. Thus, also in

the frog, our neurochemical data support a strict correlation

of the HN with the pineal complex, previously pointed out

on the basis of cytological similarities and topographical

proximity [16]. Similar conclusions have also been reached

in the lizard [17], in which HN asymmetry was hypothe-

sized to represent a specialization associated with unilateral

projection from the parietal eye.

A growing body of anatomical and functional evidence

supports close interactions between the habenular complex

and the pineal gland also in mammals [50]. The region

implicated in such relationship is the medial habenular

nucleus [12,35,48]. Although habenular asymmetry was

not reported in the above studies, it is worth recalling that,

as mentioned above, the medial habenula of mammals is

homologous to the dorsal habenula of the frog [28]. It is

also noteworthy that asymmetry is not an exclusive pecu-

liarity of lower vertebrates, because habenular left–right

differences have been demonstrated in birds and mammals

[27,30,57].

Studies on the developing diencephalon have reported

that the laterality of the asymmetric development of the

HN and pineal complex is regulated by Nodal signaling

pathway genes and by factors released from midline tissue

[14], whose role in left–right axis formation appears to be

conserved from fish to human [4]. In zebrafish, the genes

implicated in such mechanism are expressed on the left

side of the embryonic dorsal diencephalon, in a region

corresponding to the presumptive epiphysis or pineal organ

[14,37]. In zebrafish, the habenular asymmetry is lost

following selective ablation of the parapineal organ [22].

In the sea lamprey, proliferating cell nuclear antigen

immunoreactivity was found in the parapineal organ and

in the asymmetric habenular primordium; moreover, pro-

liferating cells were observed in this species in cells of the

asymmetric right habenula, mostly confined to its ventro-

caudal area [55].

Taken together with the present findings in amphibia, the

above studies indicate that diencephalic asymmetries, very

marked in many species of cyclostomes, fish and reptiles,

can derive from an interplay between the pineal and

habenular primordia. Furthermore, the data recall attention

on the study of cerebral asymmetries in vertebrates.

Acknowledgements

The preparation of this manuscript was supported by

grants from the Italian National Research Council (CNR).

References

[1] A. Ambrus, R. Kraftsik, I. Barakat-Walter, Ontogeny of calretinin

expression in rat dorsal root ganglia, Dev. Brain Res. 106 (1998)

101–108.

[2] L. Anglister, I.C. Farber, A. Shahar, A. Grinvald, Localization of

voltage-sensitive calcium channels along developing neurites: their

possible role in regulating neurite elongation, Dev. Biol. 94 (1982)

351–365.

[3] K.G. Baimbridge, M.R. Celio, J.H. Rogers, Calcium-binding proteins

in the nervous system, Trends Neurosci. 8 (1992) 303–308.

V. Guglielmotti et al. / Brain Research 999 (2004) 9–1918

[4] R.N. Bamford, E. Roessler, R.D. Burdine, U. Saplakoglu, J. de la Cruz,

M. Splitt, J. Towbin, P. Bowers, B. Marino, A.F. Schier, M.M. Shen,

M. Muenke, B. Casey, Loss of function mutations in the EGF–CFC

gene CFC1 are associated with human left – right laterality defects,

Nat. Genet. 26 (2000) 365–369.

[5] E. Bastianelli, R. Pochet, Transient expression of calretinin during

development of chick cerebellum. Comparison with calbindin-D28k,

Neurosci. Res. 17 (1993) 53–61.

[6] E. Bastianelli, R. Pochet, Calbindin-D28k, calretinin, and recoverin

immunoreactivities in developing chick pineal gland, J. Pineal Res. 17

(1994) 103–111.

[7] E. Bastianelli, R. Pochet, Calbindin-D28k, calretinin, and S-100 im-

munoreactivities in rat pineal gland during postnatal development,

J. Pineal Res. 18 (1995) 127–134.

[8] E. Bastianelli, K. Moutairou, M.T. Akele-Akpo, R. Darboux, R. Po-

chet, Calcium binding proteins immunohistochemistry and identifica-

tion of neurons in the mammalian pineal gland of the African giant

rat: Cricetomys gambianus, Gen. Physiol. Biophys. 18 (1999) 5–17.

[9] V. Braitenberg, M. Kemali, Exceptions to bilateral symmetry in

the epithalamus of lower vertebrates, J. Comp. Neurol. 138 (1970)

137–146.

[10] A.B. Butler, W. Hodos, Comparative Vertebrate Neuroatomy. Evolu-

tion and Adaptation, Wiley-Liss, 1996.

[11] M.R. Celio, Calbindin D-28k and parvalbumin in the rat nervous

system, Neuroscience 35 (1990) 375–475.

[12] M.D. Chafetz, F.H. Gage, Pineal and habenula innervation altered by

septal lesions, Brain Res. Bull. 10 (1983) 27–31.

[13] M.L. Concha, S.W. Wilson, Asymmetry in the epithalamus of verte-

brates, J. Anat. 199 (2001) 63–84.

[14] M.L. Concha, R.D. Burdine, C. Russell, A.F. Schier, S.W. Wilson, A

Nodal signaling pathway regulates the laterality of neuroanatomical

asymmetries in the zebrafish forebrain, Neuron 28 (2000) 399–409.

[15] L.O.E. Ebbesson, B. Holmqvist, T. Ostholm, P. Ekstrom, Transient

serotonin-immunoreactive neurons coincide with a critical period of

neural development in coho salmon (Oncorhynchus kisutch), Cell

Tissue Res. 268 (1992) 389–392.

[16] P. Ekstrom, S.O.E. Ebbesson, The left habenular nucleus contains a

discrete serotonin-immunoreactive subnucleus in the coho salmon

(Oncorhynchus kisutch), Neurosci. Lett. 91 (1988) 121–125.

[17] G.A. Engbretson, A. Reiner, N. Brecha, Habenular asymmetry and

the central connections of the parietal eye of the lizard, J. Comp.

Neurol. 198 (1981) 155–165.

[18] W.D. Eldred, T.E. Finger, J. Nolte, Central projections of the frontal

organ of Rana pipiens, as demonstrated by the anterograde transport

of horseradish peroxidase, Cell Tissue Res. 211 (1980) 215–222.

[19] P.S. Eriksson, E. Perfilieva, T. Bjork-Eriksson, A.M. Alborn, C. Nord-

borg, D.A. Peterson, F.H. Gage, Neurogenesis in the adult human

hippocampus, Nat. Med. 4 (1998) 1313–1317.

[20] A.V. Faure, D. Grunwald, M.J. Mountin, M. Hilly, J.P. Mauger, I.

Marty, M. De Waard, M. Villaz, M. Albrieux, Developmental ex-

pression of the calcium release channels during early neurogenesis of

the mouse cerebral cortex, Eur. J. Neurosci. 14 (2001) 1613–1622.

[21] R. Gabriel, B. Volgyi, E. Pollak, Calretinin-immunoreactive elements

in the retina and optic tectum of the frog, Rana esculenta, Brain Res.

782 (1998) 53–62.

[22] T.J. Gamse, C. Thisse, B. Thisse, M.E. Halpern, The parapineal me-

diates left – right asymmetry in zebrafish diencephalon, Development

130 (2003) 1059–1068.

[23] L.M. Garcia-Segura, D. Baetens, J. Roth, A.W. Norman, L. Orci,

Immunohistochemical mapping of calcium-binding protein immu-

noreactivity in the rat central nervous system, Brain Res. 296 (1984)

75–86.

[24] D. Gioffre, Un acquario per la Rana, Aquarium 6 (1976) 393–396.

[25] R. Guglielmone, G. Corvetti, First appearance and distribution of cal-

retinin-immunoreactive neurons in the early development of the chick

central nervous system, Cell Tissue Res. 300 (2000) 21–28.

[26] V. Guglielmotti, L. Fiorino, Nitric oxide synthase activity reveals an

asymmetrical organization of the frog habenulae during development:

a histochemical and cytoarchitectonic study from tadpoles to the ma-

ture Rana esculenta, with notes on the pineal complex, J. Comp.

Neurol. 411 (1999) 441–454.

[27] C.J. Gurusinghe, D. Erhlich, Sex-dependent structural asymmetry of

the medial habenular nucleus of the chicken brain, Cell Tissue Res.

240 (1985) 149–152.

[28] J.A. Harris, V. Guglielmotti, M. Bentivoglio, Diencephalic asymme-

tries, Neurosci. Biobehav. Rev. 20 (1996) 637–643.

[29] S.S. Jande, L. Maler, D.E.M. Lawson, Immunohistochemical map-

ping of vitamin D-dependent calcium-binding protein in brain, Nature

294 (1981) 765–767.

[30] M. Kemali, Morphological asymmetry of the habenulae of a macros-

matic mammal, the mole, Z. Mikrosk.-Anat. Forsch. 98 (1984)

951–954.

[31] M. Kemali, I. Agrelli, The habenulo-interpeduncular nuclear system

of a reptilian representative Lacerta sicula, Z. Mikrosk.-Anat. Forsch.

85 (1972) 325–333.

[32] M. Kemali, V. Braitenberg, Atlas of the Frog’s Brain, Springer, Hei-

delberg, 1969.

[33] M. Kemali, A. De Santis, The extracranial portion of the pineal com-

plex of the frog (frontal organ) is connected to the pineal, the hypo-

thalamus, the brain stem and the retina, Exp. Brain Res. 53 (1983)

193–196.

[34] M. Kemali, A. Miralto, E. Sada, Asymmetry of the habenulae in the

elasmobranch Scyllium stellare. Light microscopy, Z. Mikrosk.-Anat.

Forsch. 94 (1980) 794–800.

[35] H.W. Korf, T. Sato, A. Oksche, Complex relationships between the

pineal organ and the medial habenular nucleus-pretectal region of the

mouse as revealed by S-antigen immunocytochemistry, Cell Tissue

Res. 261 (1990) 493–500.

[36] E.D. Lephart, H. Taylor, N.A. Jacobson, M.A. Watson, Calretinin and

calbindin-D28K in male rats during postnatal development, Neuro-

biol. Aging 19 (1998) 253–257.

[37] J.O. Liang, A. Etheridge, L. Hantsoo, A.L. Rubinstein, S.J. Nowak,

J.C. Izpisua Belmonte, M.E. Halpern, Asymmetric Nodal signaling in

the zebrafish diencefhalon positions the pineal organ, Development

127 (2000) 5101–5112.

[38] H. Manelli, F. Margaritora, Tavole cronologiche dello sviluppo di

Rana esculenta, Rendiconti Accademia Nazionale dei XL. XII

(1961) 1–15 + 12 Tables.

[39] M.P. Mattson, Calcium as sculptor and destroyer of neural circuitry,

Exp. Gerontol. 27 (1992) 29–49.

[40] M.P. Mattson, A. Taylor-Hunter, S.B. Kater, Neurite outgrowth in

individual neurons of a neuronal population is differentially regulated

by calcium and cyclic AMP, J. Neurosci. 8 (1988) 1704–1711.

[41] D. Necchi, C. Soldani, M.B. Pisu, G. Bernocchi, E. Scherini, Distri-

bution of calretinin-like immunoreactivity in the brain of Rana es-

culenta, J. Chem. Neuroanat. 16 (1999) 233–243.

[42] A. Novier, D. Nicolas, R. Krstic, Calretinin immunoreactivity in pineal

gland of different mammals including man, J. Pineal Res. 21 (1996)

121–130.

[43] M. Parmentier, Structure of the human cDNAs and genes coding for

calbindin D28k and calretinin, in: R. Pochet, D.E.M. Lawson, C.W.

Heizmann (Eds.), Calcium Binding Proteins in Normal and Trans-

formed Cells, Plenum, New York, 1990, pp. 27–34.

[44] A. Porteros, J.G. Brinon, R. Arevalo, C. Crespo, J. Aijon, J.R.

Alonso, Transient expression of calretinin in the trout habenulo-in-

terpeducular system during development, Neurosci. Lett. 254 (1998)

9–12.

[45] P. Redecker, Developmental pattern of cell type-specific calretinin

immunoreactivity in the postnatal gerbil pineal gland, Brain Res.

Dev. Brain Res. 105 (1998) 43–50.

[46] P. Redecker, Y. Cetin, H.W. Korf, Differential immunocytochemical

localization of calretinin in the pineal gland of three mammalian

species 25 (1996) 9–18.

[47] I. Rodriguez-Moldes, J.P. Timmermans, D. Adriaensen, M.H.A. De

V. Guglielmotti et al. / Brain Research 999 (2004) 9–19 19

Groodt-Lasseel, D.W. Scheuermann, R. Anadon, Asymmetric distri-

bution of calbindin-D28k in the ganglia habenulae of an elasmobranch

fish, Anat. Embryol. 181 (1990) 389–391.

[48] O.K. Ronnekleiv, M. Moller, Brain–pineal nervous connections in

the rat: an ultrastructure study following habenular lesion, Exp. Brain

Res. 37 (1979) 551–562.

[49] C.M. Rovainen, Neurobiology of lampreys, Physiol. Rev. 59 (1979)

1007–1077.

[50] R. Sandyk, Relevance of the habenular complex to neuropsychiatry: a

review and hypothesis, Int. J. Neurosci. 61 (1991) 189–219.

[51] B. Schlosser, G. Klausa, G. Prime, G. Ten Bruggencate, Postnatal

development of calretinin- and parvalbumin-positive interneurons in

the rat neostriatum: an immunohistochemical study, J. Comp. Neurol.

405 (1999) 185–198.

[52] M. Setzer, N. Ulfig, Differential expression of calbindin and calretinin

in the human fetal amygdala, Microsc. Res. Tech. 46 (1999) 1–17.

[53] Y. Toba, K. Ajiki, M. Horie, K. Sango, H. Kawano, Immunohistochem-

ical localization of calbindin D-28k in the migratory pathway from the

rat olfactory placode, J. Neuroendocrinol. 13 (2001) 683–694.

[54] N. Ulfig, Calcium-binding proteins in the human developing brain,

Adv. Anat. Embryol. Cell Biol. 165 (2002) 1–92.

[55] B. Villar-Cheda, E. Perez-Costas, M. Melendez-Ferro, X.M. Abalo,

R. Rodrıguez-Munoz, R. Anadon, M.C. Rodicio, Proliferating cell

nuclear antigen (PCNA) immunoreactivity and development of the

pineal complex and habenula of sea lamprey, Brain Res. Bull. 57

(2002) 285–287.

[56] J.A. Wilson, M.D. Kawaja, Distribution of calretinin-immunoreactive

septal axons in the normal and deafferented medial habenula of adult

rats, J. Comp. Neurol. 374 (1996) 593–606.

[57] A. Wree, K. Zilles, A. Schleicher, Growth of fresh volumes and

spontaneous cell death in the nuclei habenulae of albino rats during

ontogenesis, Anat. Embryol. 161 (1981) 419–431.

[58] Y.H. Yan, J.F. Van Brederode, A.E. Hendrickson, Transient co-local-

ization of calretinin, parvalbumin, and calbindin-D28K in developing

visual cortex of monkey, J. Neurocytol. 24 (1995) 825–837.

[59] J. Yanez, H. Meissl, R. Anadon, Central projections of the parapineal

organ of the adult rainbow trout (Oncorhynchus mykiss), Cell Tissue

Res. 285 (1996) 69–74.

[60] J. Yanez, M.A. Pombal, R. Anadon, Afferent and efferent connections

of the parapineal organ in lampreys: a tract tracing and immunocy-

tochemical study, J. Comp. Neurol. 403 (1999) 171–189.