ORIGINAL PAPER

Calcitonin Gene-Related Peptide (CGRP) Stimulates PurkinjeCell Dendrite Growth in Culture

Simona D’Antoni • Laura Zambusi • Franca Codazzi •

Daniele Zacchetti • Fabio Grohovaz • Luciano Provini •

Maria Vincenza Catania • Stefano Morara

Accepted: 2 October 2010 / Published online: 20 October 2010

� Springer Science+Business Media, LLC 2010

Abstract Previous reports described the transient

expression during development of Calcitonin Gene-Related

Peptide (CGRP) in rodent cerebellar climbing fibers and

CGRP receptor in astrocytes. Here, mixed cerebellar cul-

tures were used to analyze the effects of CGRP on Purkinje

cells growth. Our results show that CGRP stimulated

Purkinje cell dendrite growth under cell culture conditions

mimicking Purkinje cell development in vivo. The stimu-

lation was not blocked by CGRP8-37, a specific antagonist,

suggesting the activation of other related receptors. CGRP

did not affect survival of Purkinje cells, granule cells or

astrocytes. The selective expression of Receptor Compo-

nent Protein (RCP) (a component of CGRP receptor fam-

ily) in astrocytes points to a role of these cells as mediators

of CGRP effect. Finally, in pure cerebellar astrocyte cul-

tures CGRP induced a transient morphological differenti-

ation from flat, polygonal to stellate form. It is concluded

that CGRP influences Purkinje cell dendrite growth in

vitro, most likely through the involvement of astrocytes.

Keywords Climbing fiber � Astrocyte � Bergmann glia �CGRP receptor � Neuropeptide � Purkinje cells

Introduction

The cerebellar Purkinje cells (Pcs), being the sole output of

the cerebellar cortex, are generally thought to represent the

site of cerebellar input integration and thus a crucial ele-

ment for the main cerebellar functions, sensorimotor inte-

gration and motor learning. Their major integration site

consists of the extensively ramified dendrites where the

two most relevant cerebellar afferent excitatory inputs,

climbing and mossy fibers, converge (directly or through

the interposition of granule cells and their parallel fibers,

respectively) as described starting from the seminal work

of Ramon y Cajal [1]. In light of the relevance of this

anatomical organization, the mechanisms and factors

leading to the development of mature Pc dendritic tree

have been the subject of several studies. As a whole,

maturation of these cells is achieved as a result of a com-

plex sequence of outgrowth and regression events that can

be schematized as follows: elongation (starting at embry-

onic stages) and subsequent retraction of a long smooth

dendrite, emission and retraction of the so called periso-

matic ‘‘disoriented’’ dendrites (around the end of the first

postnatal week) and finally extension of the mature den-

dritic tree (see [2, 3]; for recent reviews).

Among the determinants of this process, both intrinsic

factors such as RORalpha [4] or PKC [2], and extrinsic

factors, mainly consisting of granule cells, have been rec-

ognized. More specifically, it has been proposed that

Special Issue: In Honor of Dr. Abel Lajtha.

S. D’Antoni � M. V. Catania

Institute of Neurological Sciences (ISN), CNR, Catania, Italy

L. Zambusi � S. Morara (&)

Neuroscience Institute, C.N.R., Milan, Italy

e-mail: stefano.morara@in.cnr.it

F. Codazzi � D. Zacchetti � F. Grohovaz

San Raffaele Scient. Inst., Ital. Inst. Technol., Ist. Nazion.

Neurosci, Vita-Salute San Raffaele University, Milan, Italy

L. Provini

DISMAB, Univ. Milano, Milan, Italy

M. V. Catania

IRCCS Oasi Maria SS, Troina, EN, Italy

123

Neurochem Res (2010) 35:2135–2143

DOI 10.1007/s11064-010-0294-0

intrinsic factors control the first step of Pc dendritic

remodeling, whereas extrinsic ones intervene in a second

phase [3]. In this context, we previously hypothesized that

climbing fibers (that we showed to provide the first syn-

aptic input to Pc during embryonic stages) [5] could release

factors influencing the differentiation of the target region,

and, in particular, of their neuronal target, Pc. Interestingly,

during embryonic and postnatal development we revealed

the transient expression of Calcitonin Gene-Related pep-

tide (CGRP) in specific bands of climbing fibers, as well as

in their neurons of origin located in the inferior olivary

complex) [6, 7]. Moreover, we also demonstrated the

expression of CGRP receptor in the developing cerebellar

cortex [8, 9]. Given these premises, in this work we used

various cerebellar culture systems to investigate the pos-

sible role of CGRP on the differentiation of cerebellar cells

with particular regard to Pc dendrite growth.

Experimental Procedure

Animals were cared for in accordance with the principles

of the NIH Institutional Animal Care and Use Committee

Guidebook (2002). All the present experiments were car-

ried out in accordance with the European Communities

Council Directive of 24 November 1986 (86 / 609 / EEC).

Mixed Cerebellar Cultures

Mixed cerebellar cultures were prepared from 19–20 day

old Sprague–Dawley rat embryos. The embryos were sac-

rificed by decapitation and their cerebella were dissected

and kept in 10 ml of ice-cold Ca2? and Mg2? free Hank’s

balanced salt solution (Gibco-Invitrogen) containing gen-

tamicin (10 lg/ml) and HEPES (HBSS). Meninges were

removed, cerebella were washed with 10 ml of HBSS once

and digested in 2.5 ml of HBSS containing trypsin (0.1%

w/v; Gibco-Invitrogen) at 33�C for 13–15 min. The

digested cerebella were rinsed with 10 ml of HBSS twice

and then gently triturated into small aggregates in 2 ml of

HBSS supplemented with MgSO4–7H2O (12 mM) and

DNase I (5 U/ml). HBSS was added to the cell suspension

and cells were centrifuged at 1,200 rpm, 4�C for 5 min.

After removal of the supernatant, cell suspension was

plated on an area of approximately 50 mm2 onto 35 mm

cell culture dishes (Nunc) or 14 mm glass coverslips

(Marienfield) coated with poly-L-ornithine (Sigma) at a

density of 450,000 cells in Seeding medium [Dulbecco’s

Modified Eagle’s Medium F12 (D-MEM F12, Gibco-

Invitrogen), containing 10% FBS (Gibco), gentamicin

(10 lg/ml) (Sigma), 0.1 mM putrescine (Sigma), 1.4 mM

Glutamax (Gibco) and 30 nM selenium dioxide (Sigma)].

Three hours later 2 ml of Culture medium prepared

supplementing the seeding medium with 200 mg/ml

transferrin, 40 nM progesterone, 20 mg/ml insulin and

0.5 ng/ml tri-iodothyronine were added. In experiments

with CGRP or CGRP8-37 (both from NeoMPS) cells were

treated from 7 to 11/13 day in vitro (DIV) with CGRP

(using concentrations ranging from 10 nM to 1 lM) and/or

CGRP8-37 (5 lM: in these last experiments CGRP, if

present, was applied at 10 nM). CGRP and/or CGRP8-37

in culture medium (which was also used as vehicle control)

were administered once by ejection from a pipette in the

bulk of the chamber solution. In case of co-administration,

CGRP8-37 was applied 15 min before CGRP application.

Astroglial Cell Cultures

Astroglial cell cultures were prepared from cerebellum of

Sprague–Dawley newborn rats (P2). Meninges were

removed and cerebella were digested in 2.5 ml of HBSS

containing trypsin (0.1% w/v; Gibco-Invitrogen) at 33�C

for 13–15 min. The digested cerebella were rinsed with

HBSS and then gently triturated into small aggregates in

2 ml of HBSS supplemented with MgSO4–7H2O (12 mM)

and DNase I (5 U/ml). HBSS was added to the cell sus-

pension and cells were centrifuged at 1,200 rpm, 4�C for

5 min. After removal of the supernatant, cell suspension

was dissociated by gentle pipetting in Dulbecco’s Modified

Eagle’s Medium (D-MEM, Gibco-Invitrogen), supple-

mented with Glutamax (Gibco-Invitrogen), 1% P/S, 10%

FBS and plated in 75 mm flasks (13 9 105 cells/flask).

After 10–12 days cultures were shaken overnight (o.n.,

180 rpm) to remove, as much as possible, oligodendrocytes

and microglia. At 21 DIV, secondary astrocytic cultures

were established by tripsinizing the primary cultures and

subplating onto 24 multiwell plate (NUNC) (2 9 104 cells/

well) or 35 mm dish (75 9 104 cells/dish) in D-MEM with

10% FBS. At 3 DIV the glial plating medium was either

maintained or substituted with the same medium without

FBS. Cells, in the two different experimental conditions,

were treated with 1 lM CGRP and fixed at different time

(2, 6 or 27 h after treatment).

Statistical Analysis

Statistical analysis was performed by using Wilcoxon

ranksum test (equivalent to Mann–Whitney U test) when

samples did not follow a normal distribution, as for

developmental time course of Pc dendrites in culture (see

Fig. 1) and induction of stellate astrocytes by CGRP (see

Fig. 5), or by using one-way ANOVA when samples did

follow a normal distribution, as for stimulation of Pc

dendrite growth by CGRP (see Fig. 2). Normal distribu-

tion was tested by Lilliefors test (not shown). It should be

2136 Neurochem Res (2010) 35:2135–2143

123

noted that the preparations of Pc culture exhibited both a

normal (in the experiments of dendrite stimulation by

CGRP) or non-normal distribution (in the experiments on

dendrite growth time course): this finding can be

explained by the fact that the number of Pcs/coverslip is

ca. 40, but in the time course experiments this number

was divided in up to four Pc subtypes giving low numbers

that can increase skewness in some Pc subtype distribu-

tion curves.

Immunocytochemistry

Mixed cerebellar cultures or pure cultures of cerebellar

astrocytes were fixed with 4% paraformaldehyde for

10 min at room temperature (R.T.). Cultures were incu-

bated with Phosphate Buffered Saline (PBS) containing

0.2% Triton for 10 min at R.T. and then with the appro-

priate blocking solution [PBS containing 4% normal goat

serum (Vector) and 4% bovine serum albumin (Sigma)] for

Fig. 1 Purkinje (P) cells exhibit

a progressive development of

their dendritic arborization

during growth in culture. a–d:

3D reconstruction of Purkinje

cells following

immunocytochemistry with

anti-calbindin antibody

performed on mixed cerebellar

cultures at 2 DIV (a), 7 DIV

(b) and 13 DIV (c, d). Purkinje

cell dendritic development in

culture displayed four dendritic

stages: embryonic (E-Pc),

disoriented (DD-Pc), definitive

(Def-Pc) and adult (Ad-Pc)

dendrites. Scale bar: 5 lm in

a–c, 10 lm in d. e: histogram

shows percentage of Purkinje

cells with different dendritic

arborization during progressive

maturation in culture at 2, 7 and

13 DIV. n = 5 experiments

performed in 2–4 preparations.

Statistical analysis was

performed by Wilcoxon

ranksum test: significance level

(p) was set at 0.05.

* p = 0.0317 versus E-Pc 2

DIV; ** p = 0.0317 versus

DD-Pc 2 DIV; *** p = 0.0079

versus Def-Pc 7 DIV

Neurochem Res (2010) 35:2135–2143 2137

123

30 min at R.T. with gentle shaking. Subsequently, cultures

were incubated o.n. a 4�C or for 2 h a R.T. with the fol-

lowing primary antibodies: calbindin (1:2,000, Swant),

anti-glial fibrillary acid protein (GFAP) (1:1,000, Dako

Cytomation) or anti-RCP antisera #1,065 (raised in

chicken: directed against RCP fragment 111–121), #1,047

(raised in rabbit: directed against mouse RCP fragment

127–140) or #1,025 (raised in rabbit: directed against

mouse RCP fragment 81–94) (all RCP antisera are kind gift

of Ian M. Dickerson, University of Rochester, Rochester,

NY, USA; in our culture systems, all RCP antisera,

although directed against different RCP fragments, pro-

vided the same labeling: not shown). After washing, cul-

tures were incubated for 2 h at R.T. with corresponding

secondary antibodies. Hoechst 33,258 (0.4 lg/ml, Sigma)

was used for nuclear counterstaining.

Cell cultures were examined using a laser scanning

confocal microscope (Sarastro 2000; Molecular Dynamics)

equipped with a Zeiss Axioskop epifluorescence micro-

scope. Immunolabeled coverslips were examined with 40x

Plan-ApoChromat Zeiss objective using dual excitation

wavelengths (488 nm for Alexa-Fluor 488 and 514 nm for

Alexa-Fluor 594); aperture size of the pinhole was set at

50 lm, laser power was set at 12–24 mW and photomul-

tiplier detector voltage was set in the range 600–800 V.

The scanning mode format was 512 9 512 and the pixel

size was 0.25 lm. For 3D reconstruction of Pc dendritic

tree, Z-series consisting of 30–40 optical sections were

collected with 0.67 lm stepsize, Gaussian filtered (kernel

3 9 3 9 3), and 3D projections obtained by Volume

Workbench tool of ImageSpace software by thresholding

and using Surface Shaded from Depthcode option.

Results

The progression of development of Pc dendrites in our

cerebellar cultures obtained from rat embryos was assessed

during the first two weeks in culture. This analysis (per-

formed by using anti-calbindin immunocytochemistry

which provides staining of all subcellular domains of Pcs)

allowed us to classify Pcs on the basis of their dendritic

morphology. In the range 2 DIV–13 DIV (Fig. 1 a–d), four

main types of Pcs could be identified. A first type was

characterized by the presence of one/two long slender

tapering dendrites showing few ramifications (Fig. 1a).

Additional shorter, thin protrusions occasionally arose

from the cell bodies and the axon often originated from the

pole of the cell body opposite to the origin of dendrites.

Since this type of cells resembles that described by

Armengol and Sotelo [10] as ‘‘fusiform’’ stage of Pcs that

starts in the embryonic cerebellum it has been here referred

to as embryonic Pc (E-Pc). A second type of Pc was

characterized by the presence of numerous, thin, twisted

protrusions emanating from all over the cell body (see

Fig. 1b). This type of cells, which resembles the classical

stage described by Ramon y Cajal [1] as cells with ‘‘dis-

oriented dendrons’’ that characterize the most frequent cell

type encountered at the end of the first postnatal week in

the rodent cerebellum, were accordingly referred to as

disoriented dendrite Pcs (DD-Pc). A third cell type was

characterized by the presence of some thick, short stem

dendrites emanating from the cell body. This type of cells

resembles those encountered at the beginning of the second

postnatal week in the rodent cerebellum ([11]; Morara,

unpublished observation): since during this stage it is

thought to begin the construction of the ultimate Pc den-

drite this type of cell has here been referred to as definitive

dendrite Pc (Def-Pc). Finally, a fourth major type of Pc was

characterized by the presence of a well developed, highly

ramified dendritic tree resembling that of the mature Pcs

(even though it may differ from its in vivo counterpart by

the frequent presence of several dendritic stems): this cell

type has been referred to as adult dendrite Pc (Ad-Pc). In

addition to these main forms of Pcs, intermediate forms

were often detectable displaying the simultaneous presence

of two different types of dendrites (not shown).

Cells bearing one of the four cell types of dendrites were

differentially represented during maturation in vitro. In

order to unravel if their progressive appearance in culture

could resembles the stages occurring in the in vivo devel-

opment of Pc, a quantitative analysis was performed by

Fig. 2 Calcitonin gene-related peptide (CGRP) stimulates Purkinje

cell dendrite growth in vitro in mixed cerebellar culture. CGRP-

induced dendritic growth is not blocked by CGRP 8–37 (a peptidergic

antagonist of CGRP receptor). Cells are treated from 7 to 13 DIV with

CGRP 10 nM (with or without 5 lM CGRP 8–37) or with CGRP

8–37 (5 lM) alone, by using a single administration. n = 2–5

experiments performed in 2–4 preparations. Statistical analysis was

performed by one-way ANOVA followed by multcompare test

(MATLAB; The MathWorks, Inc.): significance level (p) was set at

0.05. * p = 0,0357 versus control

2138 Neurochem Res (2010) 35:2135–2143

123

measuring the percentage of each Pc type at 2, 7 and 13

DIV. As it can be seen in Fig. 1e, E-Pc (the earliest form

appearing in vivo) was by far the most prominent type

(92.0% ± 5.0) at the earliest times in culture (2 DIV) and

declined thereafter (40.1% ± 30.2 at 7 DIV, and never

seen at 13 DIV). The second form (DD-Pc, that in vivo

occurs a few days later) showed a significant increase from

2 to 7 DIV (from 8.0% ± 5.0 at 2 DIV to 56.5% ± 28.3 at

7 DIV) and remained at high levels at 13 DIV

(58.9% ± 7.0). The third form (Def-Pc, whose appearance

in vivo is further delayed of a few days) started to be

detected at 7 DIV (3.4% ± 5.5) and increased steadily at

13 DIV (37.7% ± 7.2). Finally, the fourth form (Ad-Pc,

that in vivo appears as the last one during development)

started to be detected only at 13 DIV (3.4% ± 5.8).

The effect of CGRP on Pc dendritic development was

then assessed by incubating cells (from 7 DIV to 13 DIV)

with CGRP, either in the absence or presence of CGRP8-37

(a specific inhibitor of CGRP receptor), and percentages of

E-Pc ? DD-Pc versus Def-Pc ? Ad-Pc analyzed. As it can

be seen in Fig. 2, while Def-Pc ? Ad-Pc (the two latest

forms of dendrites) accounted for only 40.0% ± 4.7% at

13 DIV in control cultures, their percentage increased

significantly to 67.1% ± 14.4 following CGRP (10 nM)

incubation. Notably, CGRP8-37 (5 lM) did not block

CGRP effect (69.5% ± 8.1) and, instead, showed a very

weak agonist activity when incubated alone (57.0% ±

11.7), although its effect was not statistically significant.

CGRP effect was also tested at higher concentrations (up to

1 lM) and found to produce the same effect as lower

concentration, and no significant increase in percentage of

responding Pcs was detected (not shown).

In order to reveal whether CGRP had additional effects

on this or other cell types present in the cerebellar mixed

cultures, an extensive analysis of survival (or proliferative)

effects on the main cell types, Pcs, granule cells and

astrocytes, has been conducted. As it can be seen in Figs. 3,

1 lM CGRP did not influence cell number of any of these

cell types.

With the aim to reveal the actual site of action of CGRP,

we performed immunocytochemistry for Receptor Com-

ponent Protein (RCP, a cytoplasmic protein associated with

receptors of the CGRP family, in particular CGRP and

adrenomedullin receptors) has been performed: the analy-

sis revealed that RCP is expressed in astrocytes (Fig. 4),

but undetectable in Pcs or any other cell type (not shown).

Finally, a possible direct action of CGRP on astrocytes

was investigated by using cultured cerebellar astrocytes. In

the presence of serum these cells acquire the widely

described flat, poligonal shape (Fig. 5a), whereas a stellate

morphology can be acquired in the absence of serum

(Fig. 5c), although by a minority, as expected. In order to

test the morphological alterations induced by CGRP,

cerebellar astrocytes were incubated under different con-

ditions (with or without serum, or with CGRP in the

presence or absence of serum). The results showed that

following 2 h of incubation serum depletion induce a small

(not significant) increase of stellate astrocytes from 0.5%

(± 0.7) to 2.8% (± 2.0). CGRP significantly increased the

number of stellate astrocytes when added to serum-deple-

ted medium (7.3% ± 0.4), whereas the peptide was inef-

fective in the presence of serum (0.7% ± 0.5). As for the

molecular mediators of CGRP action, it should be noted

that under control (serum-containing) conditions RCP was

expressed in a vast majority (63.9% ± 18.3) of astrocytes

(Fig. 5b) and, interestingly, 100% of stellate astrocytes

were found to express it (Fig. 5c, d). A time course of the

above effects (2, 6 and 27 h) was then performed. Cere-

bellar astrocytes were incubated under different conditions

(with or without serum, or with CGRP in the presence or

absence of serum) for 2, 6 or 27 h. As it can be seen in

Fig. 5e, the results show that after 6 h of incubation under

serum depletion the percentage of stellate astrocyte was

increased with respect to 2 h (4.60% ± 5.8), but under

these conditions CGRP further increased this percentage

(15.5% ± 5.3). However, with or without CGRP after 27 h

of incubation no significant difference to control was

found. It should be noted that, at this time point, astrocytes

still expressed RCP in a vast majority (not shown), a

finding that must be considered for the identification of the

mechanisms underlying the transient effect of CGRP itself.

Fig. 3 CGRP does not modify the number of Purkinje cells, granule

cells and astrocytes in mixed cerebellar cultures. The cultures were

treated for 4 days (from 7 to 11 DIV) with CGRP 1 lM (one

administration). The number of astrocytes, Purkinje cells and granule

cells after CGRP treatment is calculated as percentage of respective

control. Data represent mean ± standard deviation of three experi-

ments, each performed in duplicate or triplicate

Neurochem Res (2010) 35:2135–2143 2139

123

Finally, the presence of serum blocked CGRP effect at all

time points (Fig. 5e).

Discussion

The main finding of the present paper is that CGRP stim-

ulates the growth of Pc dendrites in culture. This action is

likely to be mediated by astrocytes on which CGRP is able

to exert a morphological differentiation effect.

In our in vitro model of cerebellar development, Pc

growth appears to largely mimic the progression of stages

described in vivo, with Pcs switching from embryonic

features (2 days cultures) to the phenotype characterized

by disoriented dendrites (a few days later). Subsequently

cells with definitive dendrites become a major component

whereas adult type of cells start to be detected at 13 DIV. It

is worth mentioning that this timing is quite comparable to

the one occurring in vivo, in spite of the disruption of tissue

parameters that provide many extrinsic factors for the

differentiation, as it was already found in a previous dis-

sociated cerebellar culture [12]: however, it must be

acknowledged that medium components can deeply influ-

ence Pc dendrite growth in vitro (see e.g. [13]). In this in

vitro model CGRP induces a significant increase of late

stages Pcs, an effect which is not mediated or accompanied

by effects on survival (and/or proliferation) of this or other

cell types. CGRP receptor specific antagonist (CGRP8-37)

did not block CGRP stimulation of Pc dendrite growth. In a

few cases, CGRP8-37 was described as being unable to

inhibit CGRP effects and, instead, to exert an agonist

action, such as in spinal cord [14], subfornical organ neu-

rons [15] and hypothalamus [16]: if present, such an effect

could mask its antagonist activity. In our experimental

conditions CGRP8-37 alone produced no statistically sig-

nificant effect even at concentration (5 lM) higher than the

one showing agonist activity (1 lM; 15). The absence of an

agonist activity of CGRP8-37 under our experimental

condition rules out the possibility that its failure to act as

antagonist depends on its agonist activity and corroborates

the hypothesis that the stimulatory effect of CGRP on Pc

dendrites is not mediated by CGRP receptor. However,

further investigation is needed to fully rule out the possi-

bility that CGRP receptor does not play a role.

CGRP stimulatory effect can be explained likely by the

involvement of other receptor types of the same family,

such as adrenomedullin, amylin or calcitonin receptors,

whose structure is based on the association of one of the two

related molecules Calcitonin Receptor or Calcitonin-Like

Receptor with one of the three related molecules called

Receptor Activity Modifying Proteins 1–3, giving rise to

receptors some of which can be activated by CGRP, but are

insensitive to CGRP8-37 action [17]. Astrocytes were found

to express different receptor types of the family: in partic-

ular they encode Calcitonin-Like Receptor and Receptor

Activity Modifying Proteins 1, 2, 3 [18] and hence poten-

tially express CGRP, AM1 and AM2 receptors, respec-

tively. It is worth mentioning that RCP can interact with

CLR [19], potentially taking part in all these receptors (in

addition to CGRP receptor, indeed, it was shown to form a

functional AM1 receptor; [20]). Although AM1 receptors

show very low sensitivity to CGRP AM2 receptors can

respond to 10 nM CGRP, but are insensitive to CGRP8-37

antagonist action [17]. Thus, AM2 receptors are likely

candidate for mediating CGRP stimulatory activity on Pc

dendrites. Finally, our results show that RCP is present in

the vast majority of astrocytes even following 27 h incu-

bation with CGRP, when CGRP effect is already exhausted.

Thus, the fact that CGRP effect is transient does not seem to

be caused by CGRP receptor component down-regulation.

The present in vitro action of CGRP is likely to be

exerted in vivo as well. In rat cerebellum, an analysis

conducted at the optical level showed that CGRP is tran-

siently and selectively expressed during development in

Fig. 4 Receptor Component

Protein (RCP) is expressed in

astrocytes in mixed cerebellar

cultures. Immunocytochemistry

with anti-GFAP a and anti-RCP

b antibodies performed at 7

DIV. Arrows indicate the

regions of co-localization of the

two antibodies. Scale bar: 5 lm

2140 Neurochem Res (2010) 35:2135–2143

123

Fig. 5 RCP is expressed in

astrocytes with flat a–b and

stellate c–d morphology in

cytoplasmic and membrane

localization b, d. CGRP

transiently induces stellate

morphology in astrocytes e.

a–d: Immunocytochemistry

with anti-GFAP a, c and anti-

RCP b, d antibodies performed

on cerebellar pure astrocytic

cultures at 3 DIV in absence

a–b or presence (6 h

incubation) c–d of 1 lM CGRP.

e: histogram shows the

percentage of astrocytes with

stellate morphology in four

experimental conditions

(?FCS - CGRP; ?FCS

?CGRP; -FCS - CGRP;

-FCS ?CGRP) following

treatments with CGRP 1 lM for

2, 6, 27 h. n = 2–4 experiments

performed in 2–3 preparations.

Statistical analysis was

performed by Wilcoxon

ranksum test: significance level

(p) was set at 0.05. * p = 0.029

versus ?FCS - CGRP at 2 h;

** p = 0.029 versus ?FCS

- CGRP at 6 h

Neurochem Res (2010) 35:2135–2143 2141

123

specific compartments of the rat olivocerebellar system [6],

a finding later confirmed by CGRP mRNA expression

analysis in the inferior olivary complex [7]. An ultra-

structural analysis showed that CGRP is expressed in

climbing fiber synapses apposed to Pc dendrites or cell

bodies starting from embryonic day 19 and have been

described as the first synaptic input to Pcs [5]. Moreover,

the peptide is localized to vesicular structures, i.e. in a

compartment from which it can be released [5].

The site responsible for the stimulatory activity of

CGRP on Pc dendrites does not seem to be, however, the

climbing fiber-Pc synapse, but, instead, the cerebellar

astrocyte that is known to enwrap this synapse, in particular

Bergmann glia. Indeed, by using RCP as marker of the

presence of receptors of the CGRP family only astrocytes

were labeled in our mixed cerebellar culture, a results

confirmed by using different RCP antisera. This finding

reflect the results obtained by a previous confocal analyses

on the developing cerebellar cortex using a monoclonal

antibody directed against a putative purified cerebellar

CGRP receptor [8] or anti RCP antisera [9]: in these studies

CGRP receptors were primarily found in Bergmann glia

cells, where they attain a presumptive membrane locali-

zation during the time period of CGRP expression in

climbing fibers, whereas during the same period Pcs

express the receptor in a cytoplasmic localization [8, 9].

The effects induced by CGRP on astrocytes can be

rather complex: the peptide was shown to increase cAMP

[21], c-Fos [22] and to induce calcium transients in Berg-

mann glia in cerebellar slices [9]. It is worth mentioning

that in the experiments with cerebellar slices, CGRP did

not elicit calcium responses in Pcs [9]. An additional effect

that has been described for CGRP on cortical astrocytes is

the induction of morphological differentiation, the forma-

tion of so-called stellate cells [21]. Our present findings

extend this effect also to cerebellar astrocytes. This new

finding can be relevant in view of the experimental evi-

dence that cerebellar neurons induce a morphological

transition in cerebellar astrocytes from a flat, polygonal

shape to a stellate shape, a form which is associated with

higher cerebellar neuronal survival and neurite extension

[23] even though it may not be necessary for pontine

neuron axonal extension [24].

Although during the complex developmental processes

occurring in the cerebellum Bergmann glia has been mainly

considered a riding trail for radial migration of granule cells

[1, 25–27], its involvement in Purkinje cell differentiation is

receiving increasing attention ([28, 29]; for reviews). In

addition, postnatal ablation of cerebellar astrocytes causes

severe disruption of cerebellar development including

marked secondary effects on Pc and granule cell differen-

tiation [30], suggesting crucial developmental roles for

cerebellar astrocytes, in particular Bergmann glia.

Altogether these results strongly support the hypothesis

that, among other neuropeptides described to influence

cerebellar development and physiology [31], CGRP,

released from climbing fibers, may influence Pc dendritic

growth by activating a receptor of its family on adjacent

astrocytes during cerebellar development.

Acknowledgments S. M. is supported by the FIRB and FISR pro-

jects of the Italian Ministry of University and Research, and by RSTL

projects of CNR; F. G. is supported by the Italian Ministry of

Research (PRIN, Progetti di Ricerca di Interesse Nazionale, project

2006054051) and the Italian Telethon Foundation (GGP05141 grant).

M. V. C: is supported by a FIRB projects of the Italian Ministry of

University and Research and by Oasi Maria SS, Troina. The authors

wish to thank Mr. Francesco Marino (Institute of Neurological

Sciences-CNR, Catania) for his technical assistance in the preparation

of figures. I would like to thank prof. A.M. Giuffrida-Stella and

Dr. M.V. Catania for the proposal to give a contribution to this issue.

Indeed, even if I’ve never had the chance to meet Prof. Lajtha per-

sonally his pioneering work on protein metabolism and turnover in

brain was a hallmark for a young student wishing to move its

scientific interests from biochemistry to neuroscience, thus repre-

senting an exciting bridge for my scientific career.

References

1. Ramon y Cajal S (1911) Histologie du systeme nerveux de

l’homme et des vertebres, vol II. Maloine, Paris

2. Kapfhammer JP (2004) Cellular and molecular control of den-

dritic growth and development of cerebellar Purkinje cells. Prog

Histochem Cytochem 39:131–182

3. Sotelo C, Dusart I (2009) Intrisic versus extrinsic determinants

during the development of Purkinje cell dendrites. Neuroscience

162:589–600

4. Boukhtouche F, Brugg B, Wehrle R et al (2010) Induction of

early Purkinje cell dendritic differentiation by thyroid hormone

requires RORa. Neural Dev 5:18–28

5. Morara S, van der Want JJL, De Weerd H et al (2001) Ultra-

structural analysis of climbing fiber-Purkinje cell synaptogenesis

in the rat cerebellum. Neuroscience 108:655–671

6. Morara S, Provini L, Rosina A (1989) CGRP expression in the rat

olivocerebellar system during postnatal development. Brain Res

504:315–319

7. Morara S, Sternini C, Provini L et al (1995) Developmentally

regulated expression of a- and b-CGRP mRNA and CGRP-

immunoreactivity in the rat inferior olive. J Comp Neurol 354:

27–38

8. Morara S, Rosina A, Provini L et al (2000) Calcitonin gene-

related peptide receptor expression in the neurons and glia of

developing rat cerebellum: an autoradiographic and immunohis-

tochemical analysis. Neuroscience 100:381–391

9. Morara S, Wang L-P, Filippov V et al (2008) Calcitonin Gene-

Related Peptide (CGRP) triggers Ca2 ? responses in cultured

astrocytes and in Bergmann glial cells from cerebellar slices Eur.

J Neurosci 28:2213–2220

10. Armengol JA, Sotelo C (1991) Early dendritic development of

Purkinje cells in the rat cerebellum. A light and electron micro-

scopic study using axonal tracing in ‘in vitro’ slices. Brain Res

Dev Brain Res 64(1–2):95–114

11. Crepel F, Delhaye-Bouchaud N, Dupont JL et al (1980) Dendritic

and axonic fields of Purkinje cells in developing and x-irradiated

2142 Neurochem Res (2010) 35:2135–2143

123

rat cerebellum. A comparative study using intracellular staining

with horseradish peroxidase. Neuroscience 5(2):333–347

12. Dunn ME, Schilling K, Mugnaini E (1998) Development and fine

structure of murine Purkinje cells in dissociated cerebellar cul-

tures: dendritic differentiation, synaptic maturation, and forma-

tion of cell-class specific features. Anat Embryol 197:31–50

13. Kimura-Kuroda J, Nagata I, Negishi-Kato M et al (2002) Thyroid

hormone-dependent development of mouse cerebellar Purkinje

cells in vitro. Dev Brain Res 137:55–65

14. Xu XJ, Wiesenfeld-Hallin Z (1996) Calcitonin gene-related

peptide (8–37) does not antagonize calcitonin gene-related pep-

tide in rat spinal cord. Neurosci Lett 204:185–188

15. Riediger T, Schmid HA, Young AA et al (1999) Pharmacological

characterisation of amylin-related peptides activating subfornical

organ neurones. Brain Res 837:161–168

16. Kobayashi A, Osaka T, Namba Y et al (1999) CGRP microin-

jection into the ventromedial or dorsomedial hypothalamic

nucleus activates heat production. Brain Res 827:176–184

17. Poyner DR, Sexton PM, Marshall I et al (2002) International

union of pharmacology. XXXII. The mammalian calcitonin gene-

related peptides, Adrenomedullin, Amylin, and Calcitonin

receptors. Pharmacol Rev 54:233–246

18. Moreno MJ, Terron JA, Stanimirovic DB et al (2002) Charac-

terization of calcitonin gene-related peptide (CGRP) receptors

and their receptor-activity-modifying proteins (RAMPs) in

human brain microvascular and astroglial cells in culture. Neu-

ropharmacol 42:270–280

19. Evans BN, Rosenblatt MI, Mnayer LO et al (2000) CGRP-RCP, a

novel protein required for signal transduction at Calcitonin Gene-

related peptide and adrenomedullin receptors. J Biol Chem

275:31438–31443

20. Prado MA, Evans-Bain B, Oliver KR et al (2001) The role of the

CGRP-receptor component protein (RCP) in adrenomedullin

receptor signal transduction. Peptides 22:1773–1781

21. Lazar P, Reddington M, Streit W et al (1991) The action of

calcitonin gene-related peptide on astrocyte morphology and

cyclic AMP accumulation in astrocyte cultures from neonatal rat

brain. Neurosci Lett 130:99–102

22. Haas CA, Reddington M, Kreutzberg GW (1991) Calcitonin

Gene-related peptide stimulates the induction of c-fos gene

expression in rat Astrocyte cultures. Eur J Neurosci 3:708–712

23. Hatten ME (1985) Neuronal regulation of astroglial morphology

and proliferation in vitro. J Cell Biol 100(2):384–396

24. Wang L-C, Baird DH, Hatten ME et al (1994) Astroglial differ-

entiation is required for support of neurite outgrowth. J Neurosci

74:3195–3207

25. Rakic P (1971) Neuron-glia relationship during granule cell

migration in developing cerebellar cortex. A Golgi and elec-

tronmicroscopic study in Macacus Rhesus. J Comp Neurol

141(3):283–312

26. Altman J (1982) Morphological development of the rat cerebel-

lum and some of its mechanisms. In: Palay SL, Chan-Palay V

(eds) The cerebellum. New vistas. Springer, Berlin, pp 8–46

27. Komuro H, Yacubova E (2003) Recent advances in cerebellar

granule cell migration. Cell Mol Life Sci 60(6):1084–1098

Review

28. Yamada K, Watanabe M (2002) Cytodifferentiation of Bergmann

glia and its relationship with Purkinje cells. Anat Sci Int 77(2):

94–108 Review

29. Lordkipanidze T, Dunaevsky A (2005) Purkinje cell dendrites

grow in alignment with Bergmann glia. Glia 51(3):229–234

30. Delaney CL, Brenner M, Messing A (1996) Conditional ablation

of cerebellar astrocytes in postnatal transgenic mice. J Neurosci

16(21):6908–6918

31. Ito M (2009) Functional roles of neuropeptides in cerebellar

circuits. Neuroscience 162:666–672

Neurochem Res (2010) 35:2135–2143 2143

123