The novel long PDE4A10 cyclic AMP phosphodiesterase shows a pattern

of expression within brain that is distinct from the long PDE4A5 and short

PDE4A1 isoforms

Ian McPheea,1, Susan Cochrana,b, Miles D. Houslaya,*aMolecular Pharmacology Group, Division of Biochemistry and Molecular Biology, University of Glasgow, Wolfson Building, IBLS,

Glasgow G12 8QQ, Scotland, UKbYoshitomi Research Institute for Neuroscience, University of Glasgow, West Medical Building, IBLS, Glasgow G12 8QQ, Scotland, UK

Received 2 April 2001; accepted 4 July 2001

Abstract

In situ hybridisation methods were used to map the distribution of the novel long PDE4A10 isoform in the brain. PDE4A10 distribution

was compared to that of the long PDE4A5 isoform and the short PDE4A1 isoform using probes specific for unique sequences within each

of these isoforms. Coronal sections of the brain, taken at the level of the olfactory bulb, prefrontal cortex, striatum, thalamus, hippocampus

and cerebellum, were analysed. Strongest expression of PDE4A isoforms was found in the olfactory bulb granular layer with high signals

also in the piriform cortex, the dentate gyrus and the CA1 and CA2 pyramidal cells. For the two long forms, level general staining was

noted throughout the striatum, thalamus and hippocampus but no signal was evident in the cerebellum. The long PDE4A10 and PDE4A5

isoforms localised to essentially the same regions throughout the brain, although PDE4A10 was uniquely expressed in the major island of

Calleja. A signal for the short PDE4A1 isoform was found in regions in which the two long isoforms were both expressed, with the

exception of the medial nucleus of the amygdala where weak signals for PDE4A5 and PDE4A10 were detected but PDE4A1 was absent.

Uniquely, strong signals for PDE4A1 were detected in the glomerular layer of the olfactory bulb, the CA3 pyramidal cell region and the

cerebellum; areas where signals for the two long forms were not evident. PDE4A transcripts for both PDE4A5 and PDE4A10 were not

apparent in the brain stem and those for PDE4A1 were low. PDE4A isoforms are present in several key areas of the brain and therefore

present valid targets for therapeutic interventions. Whilst the two long PDE4A isoforms show a remarkably similar distribution, in at least

three regions there is clear segregation between their pattern of expression and that of the PDE4A1 short form. This identifies differential

regulation of the expression of PDE4A long and short isoforms. We suggest that specific PDE4A isoforms may have distinct functional

roles in the brain, indicating that PDE4A isoform-selective inhibitors may have specific therapeutic and pharmacologic properties. D 2001

Elsevier Science Inc. All rights reserved.

Keywords: Cyclic AMP; Phosphodiesterase; PDE4; Brain; Rolipram; CA1; CA2; CA3; Learning; Memory; Depression

1. Introduction

Cyclic AMP (cAMP) has been implicated in the regu-

lation of brain activity in a number of areas such as synaptic

plasticity, learning and memory, emesis and depression.

Until very recently cAMP was thought to exert its effects

solely through the activation of protein kinase A (PKA) and

in so doing modulate signal transduction cascades and

transcription. However, new cAMP effector proteins have

been discovered which, like PKA, bind to and are directly

activated by cAMP. These include cAMP-gated ion channels

[1–3] and GEFs [4–6]. These range of effector molecules

have been implicated in the regulation of neurones [7,8].

cAMP phosphodiesterases provide the sole means

whereby cAMP can be degraded in cells. As such these

enzymes are poised to exert a major regulatory role on

cAMP signalling. A large multigene family expresses a

myriad of enzymes able to hydrolyse cAMP. However,

recently there has been considerable interest in isoforms of

0898-6568/01/$ – see front matter D 2001 Elsevier Science Inc. All rights reserved.

PII: S0898 -6568 (01 )00217 -0

* Corresponding author. Tel.: +44-141-330-5903; fax: +44-141-330-

4365/462.

E-mail address: m.houslay@bio.gla.ac.uk (M.D. Houslay).1 Current address: Scottish Biomedical, Block H, Ground Floor,

Telford Pavilion, Todd Campus West of Scotland Science Park, Glasgow

G20 0XA, Scotland, UK.

Cellular Signalling 13 (2001) 911–918

the multigene PDE4 cAMP-specific phosphodiesterase fam-

ily. This arose originally from the observations that inactiv-

ating mutations in the cognate enzyme in Drosophila

melanogaster caused learning and memory difficulties [9–

12]. More pertinently, however, the archetypal PDE4 select-

ive inhibitor rolipram has been shown to be neuroactive and

to have antidepressant properties shown in both animal

models and in clinical trials [13–16].

There are four PDE4 subfamilies, encoded by separate

genes (A, B, C, D) [17–19]. These share three regions of

homology, namely the catalytic unit and two upstream

conserved regions, called UCR1 and UCR2. Each of these

subfamilies, however, generates distinct isoforms that are

characterised by unique N-terminal regions. The suggested

role [17] of these N-terminal regions appears to be associated

with defining the intracellular localisation of the specific

isoform as well as exerting a regulatory effect on its catalytic

activity and sensitivity to inhibition. Such isoforms, how-

ever, fall into two classes, namely long forms that exhibit

both UCR1 and UCR2 and short forms that lack UCR1. This

confers functionally distinct properties on these two classes

of isoforms. Thus, for example, UCR1 interacts with UCR2

to provide a module [20] that allows for the activation of

long isoforms upon the phosphorylation of UCR1 by PKA

[21,22], thus providing a cellular desensitisation system

[19,23]. Additionally, the paired UCR1/UCR2 module

directs ERK phosphorylation of the PDE4 catalytic unit to

confer inhibition on long isoforms, whilst the lone UCR2 in

short forms programmes resultant inhibition [24].

Recently we have cloned [25] a novel PDE4A long

isoform, called PDE4A10. This is one of a number of

PDE4A long isoforms [17]. Thus three catalytically active

long PDE4A isoforms have been identified in rat; namely

PDE4A5 [26,27], PDE4A8 [28] and PDE4A10 [25]. In

addition to this the short isoform, PDE4A1 has also been

identified [29] and characterised [30,31]. Except for

PDE4A8, all of these isoforms have been shown to have

homologues in man and mouse [17]. Such PDE4A isoforms

differ in relative catalytic activity, sensitivity to inhibition

by rolipram and thermostability [17,25]. In addition, all

these isoforms differ markedly in their intracellular local-

isation [25,31–33] and only the PDE4A5 isoform appears

to become activated through stimulation of the PI-3 kinase

cascade [34]. Interestingly, the PDE4A5 isoform appears to

have a role in the survival of at least certain cells and is the

sole PDE4 isoform that is cleaved by caspase-3 during

apoptosis, causing it to redistribute within the cell [32].

These differences suggest that the various PDE4A isoforms

may have specific functional roles in the cells where they

are expressed.

In situ hybridisation studies have been done previously

using ‘pan’ PDE4A probes in order to gain insight into the

overall pattern of expression of the various PDE4 classes

[35–37]. However, such analyses will be biased by the

level of expression of the predominant isoforms within a

particular PDE4 subclass. No attempt has been made to

compare the expression pattern in brain of isoforms within a

particular PDE4 subfamily. As PDE4A isoforms have been

extensively studied and shown to have properties that

uniquely characterise them [17], we have set out here to

compare the distribution of the novel long PDE4A10 iso-

form with the well-established long PDE4A5 and short

PDE4A1 isoforms. We did not attempt to analyse the

testis-specific PDE4A8 isoform, as it is not expressed in

brain [28]. We show that whilst there is overlap in the

expression profile of these various PDE4A isoforms, espe-

cially of the two long isoforms, there are various regions in

the brain where particular isoforms of the PDE4A subfam-

ily are uniquely expressed.

2. Materials and methods

2.1. Animals

Male Wistar rats (300 g, purchased from Harlen-Olac,

UK) were housed under a 12-h light/dark cycle in a

temperature and humidity controlled environment with free

access to food and water. Animals were killed by cervical

dislocation. The brains were rapidly removed and frozen in

isopentane cooled to � 42�C on solid CO2 and then coated

in embedding medium (Lipshaws) and stored at � 70�Cuntil required.

2.2. Probes and labelling

All probes were 45-mer oligonucleotides purchased from

Interacteva. The probe sequences used were as follows:

PDE4A5 Genbank accession number L27057, probe

4A5AS of sequence cgc tct atg ggc cgg tgc ggt gag cgc tct

gtc tct gag cgc tcg; PDE4A10 Genbank accession number

AF110461, probe 4A10AS of sequence gtc ctc ctc gct gaa

gga gaa atg ggt cag tga ctc tgg tcc tag; PDE4A1 Genbank

accession number M26715, probe 4A1AS of sequence cag

cca ggg ctt gga gca ggt ctc gca gaa gaa gtc aac cag agg.

The probes were 30 end-labelled with terminal deoxyri-

bonucleotidyl transferase (Pharmacia) and with the isotope

5-a-35S-dATP (NEN, specific activity > 1000 Ci/mmol) and

incubated at 37�C for 1 h. The reaction was terminated by

addition of 60 ml diethyl pyrocarbonate (DEPC)-treated

water. The probe was purified using QIAquick Nucleotide

Removal kit (Qiagen). The extent of probe labelling was

assessed using b-scintillation counting, and probes labelled

from 100,000 to 300,000 dpm ml� 1 were used for in situ

hybridisation. One microliter 1 M dithiothreitol (DTT) was

added to the labelled probe in order to prevent oxidation.

Labelled probes were stored at � 20�C until required.

2.3. Dot blot

Probes used in the dot blot were 50 end labelled with T4

polynucleotide kinase (Promega) and g-32P-ATP (Amer-

I. McPhee et al. / Cellular Signalling 13 (2001) 911–918912

sham, specific activity >3000 Ci/mmol). The reaction

proceeded at 37�C for 15 min and was terminated by

adding EDTA to a final concentration of 20 mM. The

probe was purified using QIAquick Nucleotide Removal

kit (Qiagen). Plasmid DNA (approximately 10 ng), con-

taining the genes of interest, were spotted onto hybond –N

membrane, allowed to dry and cross-linked using a trans-

illuminator for 1 min. The membrane was then prehybri-

dised in a solution containing 7% (w/v) SDS, 0.5 M

phosphate, pH 7.2 for 1 h at 65�C in a rotating oven

before freshly boiled radiolabelled probe (10 pmol) was

added. Hybridisation was allowed to proceed overnight

before the probe was poured off and the membrane washed

in SSC buffer. The initial wash was carried out in 3� SSC

for 5 min before washing in 1� SSC, for 30 min each

time, until the background activity was removed from the

membrane. The activity on the filters was imaged using

phosphoimager (Fuji).

2.4. Section collection and preparation

Twenty-micrometer sections were cut on a cryostat at

� 20�C (Leica CM1850) collected onto poly-L-lysine coated

slides from the following bregma level: 6.70 mm (olfactory

bulb), 3.20 mm (prefrontal cortex), 0.20 mm (striatum),

� 2.80 mm (midline thalamus), � 4.80 mm (hippocampus)

and � 10.52 mm (cerebellum). Sections were dried at room

temperature then fixed in freshly prepared 4% (wt/vol)

paraformaldehyde in phosphate-buffered saline (PBS) for

5 min. After rinsing the sections were sequentially dehy-

drated in 70%, 95% and 100% ethanol before being stored

under ethanol at 4�C.

2.5. In situ hybridisation

Sections were hybridised overnight at 42�C in a hybrid-

isation mixture comprising hybridisation buffer (50%

deionised formamide, 20% 20� standard saline citrate

[20� SSC = 3 M sodium chloride; 0.3 M sodium citrate,

pH 7], 5% 0.5 M sodium phosphate [pH 7], 1% 0.1 M

sodium pyrophosphate, 2% 5 mg ml� 1 polyadenylic acid,

10% dextran sulphate, volume adjusted to 50 ml with

DEPC-treated water), 5 ng 5000 ml � 1 labelled probe and

1 M DTT in the proportions 100:1:4. Two hundred micro-

liters of hybridisation mixture was applied to each slide

and the sections were then covered with parafilm. After

overnight hybridisation under humidified conditions, the

parafilm was removed from each slide under 1� SSC,

then washed for 30 min in 1� SSC warmed to 60�C. Thesections were then washed in 1� SSC then 0.1� SSC at

room temperature and dehydrated in 70% then 95%

ethanol. Once dry, the sections were exposed to auto-

radiographic film (Biomax MR, Kodak) for 14 days.

Images were obtained via the MCID image analysis system

(M5, Canada).

3. Results and discussion

3.1. Fidelity of the probes

In man it has been shown that the extreme 50 exon that

characterises each of the three isoforms studied here is

unique [25,38]. Thus transcripts from the three PDE4A

isoforms analysed here will each have a distinct 50 region

whose sequence is highly conserved between different

species [17,25]. We have exploited the differences in

sequence of the 50 regions of the PDE4A1, PDE4A5 and

PDE4A10 isoforms to design end-labelled probes that were

specific for each of these isoforms and which could then

be used in the in situ hybridisation experiments to deter-

mine individual expression profiles. We set out to confirm

the specificity of each of these probes. To do this, ‘dot

blots’ were prepared in which plasmid DNA containing the

three different PDE4A isoforms was spotted onto nylon

membranes and then probed with each of the end-labelled

probes (Fig. 1). In each case the end-labelled probes

identified exclusively the appropriate PDE4A isoform they

were designed to interact with. Thus we can expect that

the in situ probes will bind specifically to the appropriate

mRNA targets.

3.2. In situ hybridisation

In these analyses we were able to show that appropriate

sense control probes did not bind to the tissue sections and

the signal obtained for each (antisense) probe could be

competed out with a 10-fold excess of unlabelled probe.

3.3. Olfactory bulb

Transcripts for all three of these PDE4A isoforms

appeared to be expressed to a high degree in the olfactory

bulb (Fig. 2), with transcripts for PDE4A1, PDE4A5 and

PDE4A10 being present in large amounts in the mitral and

granular cell layer. Intriguingly, however, whilst a clear

signal was displayed for PDE4A1 transcripts in the glom-

Fig. 1. Specificity of the in situ probes. Plasmid DNA from PDE4A1,

PDE4A5 and PDE4A10 rat isoforms was spotted onto nylon membranes

and then probed with the rat PDE4A isoform-specific oligonucleotide

sequences used for the in situ studies. Data are typical of experiments done

at least three times.

I. McPhee et al. / Cellular Signalling 13 (2001) 911–918 913

erular layer, only weak signals were seen for PDE4A5 and

PDE4A10. It is interesting to compare these data with the

immunohistochemical study of Cherry and Davis [37] done

on mouse brain. They used an antiserum that, whilst

specific for the PDE4A subclass, was unable to discriminate

between the various isoforms i.e., it was a ‘pan’ PDE4A

antiserum. Using this antiserum they observed a clear signal

for PDE4A protein in the glomerular layer only. This may

reflect a difference between rat and mouse. However,

another possibility is that transcripts present in glomerular

and granular layers are not translated to the same degree. If,

however, PDE4A5 and PDE4A10 transcripts were not

being translated efficiently, then one might have expected

to observe a large excess of PDE4A1 in Western blot

analyses. However, Western blot studies done on both rat

[27] and mouse [37] indicated that bands corresponding to

the similarly sized PDE4A10/PDE4A5 were present at

similar levels to that of PDE4A1 in homogenates of the

olfactory bulb. Of course it may be that the in situ hybrid-

isation method may exhibit a higher degree of sensitivity

than immunohistochemical analyses. Alternatively, the

mRNA detected by the in situ method may be located in

the cell bodies of neurones of the granular layer with the

expressed PDE4A protein translocated along the length of

the axon to the glomerular layer. This might allow the

PDE4A species to be functionally active at the synapses in

this region.

It is clear from these analyses that PDE4A isoforms are

highly expressed in areas that are important to odorant

signal transduction, indicating that they are likely to be

involved in the tuning, processing and possible detection of

odorant signals. PDE4A isoforms in the olfactory bulb may

even have a wider influence on brain function, since the

regions where they are detected have been implicated in

feeding responses to amino acid deficiencies and memory

and where the phosphorylation of cAMP response element

binding protein (CREB) appears to play a role [39–42]. In

addition, abnormalities in these areas are also associated

Fig. 3. Prefrontal cortex sections probed with PDE4 isoform-specific probes. The probes used were for PDE4A1 (A), PDE4A5 (B) and PDE4A10 (C). The

areas highlighted are: prelimbic region (PL); primary motor cortex (PMC); and piriform cortex (PC). Data are typical of experiments done at least three times.

Fig. 2. Olfactory bulb sections probed with PDE4 isoform-specific probes. The probes used were for PDE4A1 (A), PDE4A5 (B) and PDE4A10 (C). The areas

shown are: external plexiform layer (ExP); glomerular layer (GL); granular layer (GR); and olfactory ventricle (OV). Data are typical of experiments done at

least three times.

I. McPhee et al. / Cellular Signalling 13 (2001) 911–918914

with limbic kindling and seizures such as those seen in

epilepsy [43–45].

3.4. Prefrontal cortex and striatum

Signals for mRNA transcripts of all three of these

PDE4A isoforms were seen in the piriform cortex, the

prefrontal cortex (Fig. 3) and striatum sections (Fig. 4).

These structures have been implicated as being involved in a

number of processes, such as olfactory signal processing,

memory, depression and seizures [43,45–47]. However, the

signal for each of these isoforms was more diffuse in the

primary motor cortex prelimbic region of the prefrontal

cortex, the anterior cingulate cortex and somatosensory

cortex of the striatum (Figs. 3 and 4). The deep layers of

the somatosensory cortex showed a relatively stronger

signal than the superficial layers. PDE4A1 appeared to be

uniformly distributed through the anterior cingulated cortex

and somatosensory cortex, whilst levels of PDE4A5 and

PDE4A10 appeared to be relatively higher in the anterior

cingulated cortex (Fig. 4). Interestingly, however, PDE4A10

uniquely showed a strong and well-defined signal in the

major island of Calleja (Fig. 4), a region of the brain that has

been implicated in depression [48]. This was the clearest

indication that we observed in this study of a difference

in the expression patterns of the PDE4A10 and PDE4A5

long isoforms.

3.5. Hippocampus and hypothalamus

Strong, clear signals for transcripts of all three PDE4A

isoforms were seen in the dentate gyrus, the CA1 and CA2

regions of the hippocampus (Figs. 5 and 6). However, we

noted that PDE4A1 uniquely provided a strong signal in the

CA3 region (Figs. 5 and 6). Stronger expression of PDE4A1

was also seen in the thalamus (Fig. 5), the superficial layer

Fig. 4. Striatum sections probed with PDE4 isoform-specific probes. The probes used were for PDE4A1 (A), PDE4A5 (B) and PDE4A10 (C). The areas shown

are: anterior cingulate cortex (AC); somatosensory cortex (SC); striatum (St); major island of Calleja (MC); olfactory tubercle (OT); and piriform cortex (PC).

Data are typical of experiments done at least three times.

Fig. 5. Thalamus sections probed with PDE4 isoform-specific probes. The probes used were for PDE4A1 (A), PDE4A5 (B) and PDE4A10 (C). The areas

shown are: the CA1, CA2 and CA3 areas of the pyramidal cell layer; the dentate gyrus (DG); and medial nucleus of the amygdala (MN). Data are typical of

experiments done at least three times.

I. McPhee et al. / Cellular Signalling 13 (2001) 911–918 915

of the retrosplenial cortex and the lateral and medial layer of

the mammillary body (Fig. 6). The CA1 and dentate gyrus

regions are very important for cognitive function, such as

memory, with modulation of cAMP in these regions affect-

ing long-term potentiation [9,12,29,49] and altering the

activity of presynaptic neurones [50]. In addition to this,

expression of the CREB in the hippocampus produces an

antidepressant effect [51]. Thus, PDE4A isoforms may have

particular importance in regulating plasticity processes

associated with these brain areas. Consistent with this, the

PDE4 selective inhibitor rolipram has been shown to facil-

itate long-term potentiation and enhance memory [52–54].

In this regard it also has been shown to induce the

expression of BNDF in the CA1, CA3 and the dentate

gyrus [55], a process associated with antidepression. It is

thus very intriguing to see such profound differences in the

distribution of these various PDE4A isoforms. This may

signify particular PDE4A isoforms as providing appropriate

targets for the development of cognitive enhancers and also

antidepressive therapeutics.

The medial nucleus of the amygdala provided an area of

the brain where both PDE4A5 and PDE4A10 were clearly

expressed (Fig. 5). This region is typically associated with

emotional responses, learning and social awareness and has

been implicated in depression and autism [56–59].

3.6. Cerebellum and paraflocculus

Little or no expression of either PDE4A10 or PDE4A5

was evident in the cerebellum. In profound contrast to this,

however, PDE4A1 appeared to be strongly expressed in

both the cerebellum, where it was exclusively localised to

the grey matter, and also in the paraflocculus (Fig. 7).

Indeed, the signal achieved by PDE4A1 was at a compar-

ably high level to that seen for PDE4A1 in the olfactory

bulb regions. This is consistent with Western blotting

Fig. 6. Hippocampus sections probed with PDE4 isoform-specific probes. The probes used were for PDE4A1 (A), PDE4A5 (B) and PDE4A10 (C). The areas

shown are: the CA1, CA2 and CA3 areas of the pyramidal cell layer; the dentate gyrus (DG); mammillary body medial nuclei (MN); mammillary body lateral

nuclei (LN); auditory cortex (AC); visual cortex (VC); and retrosplenial cortex (RC). Data are typical of experiments done at least three times.

Fig. 7. Cerebral sections probed with PDE4 isoform-specific probes. The probes used were for PDE4A1 (A), PDE4A5 (B) and PDE4A10 (C). The areas shown

are: grey matter (GM); white matter (WM); and paraflocculus (Pf). Data are typical of experiments done at least three times.

I. McPhee et al. / Cellular Signalling 13 (2001) 911–918916

studies done on rat cerebellum fractions, which only iden-

tified a single PDE4A immunoreactive species of the size of

PDE4A1 [27]. This may indicate a specific role for the

PDE4A1 isoform in some of the functions associated with

these structures such as motor control, temporal sensing and

occular signal processing.

In the brain stem, expression of PDE4A1 appeared to be

low and diffuse whilst transcripts for PDE4A5 and

PDE4A10 were not evident. This corroborates Western

blotting studies done on membrane and cytosol extracts

from rat brain stem that identified a single immunoreactive

species of the size of PDE4A1 in brain stem [27].

4. Conclusion

We have designed probes that can be used to detect

specifically transcripts for the three major PDE4A isoforms

expressed in brain. Their use has allowed us to show that

these isoforms from the PDE4A gene are differentially

expressed in various brain areas. Studies using inhibitors

that are selective for PDE4 cAMP-specific phosphodies-

terases have led to the suggestion that PDE4 enzymes have a

major role in the regulation of learning, memory and the

psychological state of an individual. Here we have shown

that PDE4A isoforms are present in areas of the brain

closely associated these processes such as the piriform

cortex, amygdala and hippocampus.

Expression of the two long PDE4A isoforms was highest

in the olfactory bulb and for the short PDE4A1 isoform

highest in the olfactory bulb, the cerebellum and the

paraflocculus. It is interesting to note that the two long

isoforms PDE4A5 and PDE4A10 displayed very similar

distribution throughout the brain except in the major island

of Calleja where only PDE4A10 was seen. This suggests

that the promoter regions controlling expression of these

two long isoforms are likely to be rather similar. Notwith-

standing the fact that PDE4A5 and the recently discovered

PDE4A10 are similar in size, they appear to differ in a

number of properties [25], implying that they may have

distinct functional roles in cells where they are both

expressed together.

With the exception of the major island of Calleja, where

PDE4A10 was uniquely expressed (Fig. 4), the short

PDE4A1 isoform was present in every area where the

PDE4A5 and PDE4A10 long isoforms were expressed.

However, PDE4A1 was uniquely found in several areas

where PDE4A5 and PDE4A10 were absent, such as the

cerebellum (Fig. 7), the CA3 region of the hippocampus

(Figs. 5 and 6), and the glomerular layer of the olfactory

bulb (Fig. 2). PDE4A1 differs markedly from the long

PDE4A forms. For example, it lacks UCR1 [17,38] and

so is insensitive to stimulatory phosphorylation by PKA. It

is also the only PDE4 isoform that is entirely membrane-

associated [31] and this is attributable to its unique 25 amino

acid N-terminal region [31,60,61]. These distinct attributes

may be pivotal to the particular functioning of PDE4A1 in

cells and serve to distinguish it dramatically from the

PDE4A5 and PDE4A10 long forms.

Our studies suggest that subfamily- and even isoform-

selective PDE4 inhibitors may have distinct pharmacological

properties and particular therapeutic usage. Also that disrup-

tion of the exons encoding the unique N-terminal regions of

these PDE4A isoforms may generate distinct phenotypes.

Acknowledgments

We thank the Medical Research Council (UK) for

financial support and Professor Brian Morris (YRING,

IBLS, University of Glasgow, Glasgow, UK) for helpful

discussion.

References

[1] Finn JT, Grunwald ME, Yau KW. Annu Rev Physiol 1996;58:

395–426.

[2] Kaupp UB. Curr Opin Neurobiol 1995;5(4):434–42.

[3] Zagotta WN, Siegelbaum SA. Annu Rev Neurosci 1996;19:235–63.

[4] de Rooij J, Zwartkruis FJ, Verheijen MH, Cool RH, Nijman SM,

Wittinghofer A, Bos JL. Nature 1998;396(6710):474–7.

[5] Kawasaki H, Springett GM, Mochizuki N, Toki M, Nakaya M,

Matsuda M, Housman DE, Graybiel AM. Science 1998;282

(5397):2275–9.

[6] Kawasaki H, Springett GM, Toki S, Canales JJ, Harlan P, Blumenstiel

JP, Chen EJ, Bany IA, Mochizuki N, Ashbacher A, Matsuda M, Hous-

man DE, Graybiel AM. Proc Natl Acad Sci USA 1998;95(22):

13278–83.

[7] He Q, Wu G, Lapointe MC. Am J Physiol: Endocrinol Metab

2000;278(6):E1115–23.

[8] Strijbos PJ, Pratt GD, Khan S, Charles IG, Garthwaite J. Eur J Neuro-

sci 1999;11(12):4463–7.

[9] Davis RL. Physiol Rev 1996;76:299–317.

[10] Qiu Y, Chen CN, Malone T, Richter L, Beckendorf SK, Davis RL.

J Mol Biol 1991;222(3):553–65.

[11] Qiu Y, Davis RL. Genes Dev 1993;7(7 B):1447–58.

[12] Davis RL, Cherry J, Dauwalder B, Han PL, Skoulakis E. Mol Cell

Biochem 1995;149–150:271–8.

[13] Wachtel H. Psychopharmacology 1982;77:309–14.

[14] Wachtel H. Neuropharmacology 1983;22:267–72.

[15] O’Donnell JM. J Pharmacol Exp Ther 1993;264(3):1168–78.

[16] O’Donnell JM, Frith S. Pharmacol, Biochem Behav 1999;63(1):

185–92.

[17] Houslay MD, Sullivan M, Bolger GB. Adv Pharmacol 1998;44:

225–342.

[18] Bolger G. Cell Signalling 1994;6:851–9.

[19] Conti M, Jin SLC. Prog Nucleic Acid Res 1999;63:1–38.

[20] Beard MB, Olsen AE, Jones RE, Erdogan S, Houslay MD, Bolger

GB. J Biol Chem 2000;275(14):10349–58.

[21] Sette C, Conti M. J Biol Chem 1996;271(28):16526–34.

[22] Sette C, Iona S, Conti M. J Biol Chem 1994;269(12):9245–52.

[23] Oki N, Takahashi SI, Hidaka H, Conti M. J Biol Chem 2000;275(15):

10831–7.

[24] MacKenzie SJ, Baillie GS, McPhee I, Bolger GB, Houslay MD. J Biol

Chem 2000;275:16609–17.

[25] Rena G, Begg F, Ross A, MacKenzie C, McPhee I, Campbell L,

Huston E, Sullivan M, Houslay MD. Mol Pharmacol 2001;59:

996–1011.

I. McPhee et al. / Cellular Signalling 13 (2001) 911–918 917

[26] Bolger GB, Rodgers L, Riggs M. Gene 1994;149(2):237–44.

[27] McPhee I, Pooley L, Lobban M, Bolger G, Houslay MD. Biochem J

1995;310(3):965–74.

[28] Bolger GB, McPhee I, Houslay MD. J Biol Chem 1996;271(2):

1065–71.

[29] Davis RL, Takayasu H, Eberwine M, Myres J. Proc Natl Acad Sci

USA 1989;86:3604–8.

[30] Shakur Y, Pryde JG, Houslay MD. Biochem J 1993;292(3):677–86.

[31] Shakur Y, Wilson M, Pooley L, Lobban M, Griffiths SL, Campbell

AM, Beattie J, Daly C, Houslay MD. Biochem J 1995;306(3):801–9.

[32] Huston E, Beard M, McCallum F, Pyne NJ, Vandenabeele P, Scotland

G, Houslay MD. J Biol Chem 2000;275(36):28063–74.

[33] Pooley L, Shakur Y, Rena G, Houslay MD. Biochem J 1997;271:

177–85.

[34] MacKenzie S, Fleming I, Houslay MD, Anderson NG, Kilgour E.

Biochem J 1997;324:159–65.

[35] Engels P, AbdelAl S, Hulley P, Lubbert H. J Neurosci Res 1995;41:

169–78.

[36] Cherry JA, Davis RL. J Neurobiol 1995;28(1):102–13.

[37] Cherry JA, Davis RL. J Comp Neurol 1999;407(2):287–301.

[38] Sullivan M, Rena G, Begg F, Gordon L, Olsen AS, Houslay MD.

Biochem J 1998;333(Pt 3):693–703.

[39] Wilson DA. J Neurophysiol 2000;84(6):3036–42.

[40] Gheusi G, Cremer H, McLean H, Chazal G, Vincent JD, Lledo PM.

Proc Natl Acad Sci USA 2000;97(4):1823–8.

[41] Coronas V, Krantic S, Jourdan F, Moyse E. Neuroscience 1999;90(1):

69–78.

[42] Yuan Q, Harley CW, Bruce JC, Darby-King A, McLean JH. Learn

Mem 2000;7(6):413–21.

[43] Loscher W, Ebert U. Prog Neurobiol 1996;50(5–6):427–81.

[44] Woldbye DP, Bolwig TG, Kragh J, Jorgensen OS. Neurochem Res

1996;21(5):585–93.

[45] Ferland RJ, Applegate CD. Epilepsy Res 1998;30(1):49–62.

[46] Barkai E, Hasselmo MH. Mol Neurobiol 1997;15(1):17–29.

[47] Litaudon P, Mouly AM, Sullivan R, Gervais R, Cattarelli M. Eur J

Neurosci 1997;9(8):1593–602.

[48] Maj J, Dziedzicka-Wasylewska M, Rogoz R, Rogoz Z. Eur J Pharma-

col 1998;351(1):31–7.

[49] Schulz S, Siemer H, Krug M, Hollt V. J Neurosci 1999;19(13):

5683–92.

[50] Ma L, Zablow L, Kandel ER, Siegelbaum SA. Nat Neurosci 1999;

2(1):24–30.

[51] Chen AC, Shirayama Y, Shin K, Neve RL, Duman RS. Biol Psychia-

try 2001;49(9):753–62.

[52] Barad M, Bourtchouladze R, Winder DG, Golan H, Kandel E. Proc

Natl Acad Sci USA 1998;95(25):15020–5.

[53] Zhang HT, Crissman AM, Dorairaj NR, Chandler LJ, O’Donnell JM.

Neuropsychopharmacology 2000;23(2):198–204.

[54] Zhang HT, O’Donnell JM. Psychopharmacology (Berlin) 2000;

150(3):311–6.

[55] Fujimaki K, Morinobu S, Duman RS. Neuropsychopharmacology

2000;22(1):42–51.

[56] Drevets WC. Ann NY Acad Sci 1999;877:614–37.

[57] McKenna KE. Int J Impotence Res 1998;10(Suppl 1):S25–34.

[58] Newman SW. Ann NY Acad Sci 1999;877:242–57.

[59] Bachevalier J, Beauregard M, Alvarado MC. Behav Neurosci 1999;

113(6):1127–51.

[60] Scotland G, Houslay MD. Biochem J 1995;308(2):673–81.

[61] Smith KJ, Scotland G, Beattie J, Trayer IP, Houslay MD. J Biol Chem

1996;271:16703–11.

I. McPhee et al. / Cellular Signalling 13 (2001) 911–918918