RCAN1-1L is overexpressed in neurons of Alzheimer’s
disease patients
Cathryn D. Harris, Gennady Ermak and Kelvin J. A. Davies
Ethel Percy Andrus Gerontology Center, and Division of Molecular & Computational Biology, The University of Southern California,
Los Angeles, CA, USA
The RCAN1 gene is located on human chromosome 21
in region q22.12 (Fig. 1) [1]. Initially thought to lie
within the Down’s syndrome critical region, it was sub-
sequently found to lie outside of this region. RCAN1
consists of seven exons, which can undergo alternative
splicing to produce different mRNA isoforms and, con-
sequently, different proteins (Fig. 2) [2]. A cluster of 15
putative nuclear factor of activated T-cells (NFAT)-
binding sites lie in the intron, just 5¢ to exon 4 [3]. All
known mRNA isoforms contain exons 5–7, and the
three isoforms most studied also contain either 29 amino
acids (now RCAN1-1 ‘Short’ or RCAN1-1S), or 55
amino acids (RCAN1-1 ‘Long’ or RCAN1-1L) encoded
by exon 1, or 29 amino acids (RCAN1-4) encoded by
exon 4 (Fig. 2). It has been suggested that isoform 4
may be initiated by an alternative, calcineurin-respon-
sive, promoter, due to the cluster of 15 NFAT-binding
elements 5¢ to exon 5 [4]. A splice variant containing
exon 2 has been reported in fetal liver and brain [2], but
no isoforms containing exon 3 have yet been described.
The RCAN1 protein is able to bind to and inhibit the
catalytic subunit of calcineurin (protein phosphatase 2B)
Keywords
Alzheimer’s disease; calcipressin 1; DSCR1;
Adapt78; RCAN1
Correspondence

cells. This was true in both normal and Alzheimer’s disease brain sections.
We also demonstrate that RCAN1-1 mRNA levels are approximately two-
fold higher in neurons from Alzheimer’s disease patients versus non-Alzhei-
mer’s disease controls. Using western blotting, we now show that there are
three RCAN1 protein isoforms expressed in human brain: RCAN1-1L,
RCAN1-1S, and RCAN1-4. We have determined that RCAN1-1L is
expressed at twice the level of RCAN1-4, and that there is very minor
expression of RCAN1-1S. We also found that the RCAN1-1L protein is
overexpressed in Alzheimer’s disease patients, whereas RCAN1-4 is not.
From these results, we conclude that RCAN1-1 may play a role in Alzhei-
mer’s disease, whereas RCAN1-4 may serve another purpose.
Abbreviations
AD, Alzheimer’s disease; Cb, cerebellum; GAPDH, glyceraldehyde-3-phosphate dehydrogenase gene; GFAP, glial fibrillary acidic protein; Hc,
hippocampus; HLA-DR, human leukocyte antigen-DR; LA, long and accurate; NeuN, neuronal nuclei; NFAT, nuclear factor of activated T-cells.
FEBS Journal 274 (2007) 1715–1724 ª 2007 The Authors Journal compilation ª 2007 FEBS 1715
[3,5]. Calcineurin is a calcium-dependent serine–
threonine protein phosphatase, which has several
known substrates, including the transcription factor
NFAT, which is well characterized, and the tau pro-
tein. We have proposed that RCAN1 may have a role
in the development of Alzheimer’s disease (AD) (and
other ‘tauopathies’), because it inhibits calcineurin
from dephosphorylating the tau protein, resulting in
hyperphosphorylated tau, which may then promote the
formation of paired helical filaments and neurofibril-
lary tangles [6–8]. RCAN1 is chronically overexpressed
in AD, presumably due to to the stress of chronic
inflammation [6–8]. There are data showing decreased
calcineurin activity in AD, and other studies have
shown that calcineurin inhibition results in tau phos-

ther differences were statistically significant. In these samples,
RCAN1-1L was expressed at a level approximately two-fold higher
than RCAN1-4 (P<0.05), a significant difference.
RCAN1 Structure
4 5 6 7
1 5 6 7
1 5 6 7
FLISPP
RCAN1-1S Protein
197 Amino Acids29
CaN binding motif (PKIIQT)
197 Amino Acids29
252 Amino Acids55
Chromosome 21
31p
21p
2.11p
1.11p
11q
12q
11
.
2
2q
2.22q
3.22q
1NACR
1 2 3 4 5 6 7
5’
3’

cant levels in human brain (RCAN1-1 and RCAN1-4),
and that, in general, RCAN1 is overexpressed in AD
only in regions actually affected by the disease [6].
RCAN1 has also been shown to be upregulated in
Down’s syndrome postmortem brain tissue [5,6], and it
is interesting to note that Down’s syndrome patients
also suffer from an early-onset form of AD. It is poss-
ible that RCAN1 may be protective when expressed
transiently, but may be part of a maladaptive response
if its expression fails to be turned off, resulting in dis-
ease conditions.
There is, as yet, no explanation for why cells have
multiple isoforms of this gene and protein, or what the
differences in function of each form of the gene and
protein may be. We hypothesized that there might be
differences either in the levels of RCAN1 isoform
expression, or in the cellular localization of expression,
in brain regions affected by AD. We therefore felt that
it was first important to test for the expression of dif-
ferent RCAN1 mRNAs and proteins in AD human
brain tissue as compared to that of age-matched
controls. Second, we felt that it was important to investi-
gate the cel lular di stribu tion of the isoforms in br ain-
specific cell types: neurons, microglia, and astrocytes.
Results
RCAN1 isoform expression in human brain
Previous work from our laboratory has shown that
RCAN1 mRNA is significantly expressed in adult
human brain, and upregulated in those brain areas
affected by AD. Although both isoforms 1 and 4 of

was very weak and difficult to detect and quantify.
The densities of the RCAN1-4 and RCAN1-1L bands
recognized by the common antibody were quanti-
fied using ipgel software (Scanalytics, Vienna, VA)
(Fig. 2B). In good agreement with our previous work on
RCAN1 mRNA isoforms in brain [6], the RCAN1-1L
protein was expressed at a much higher level than was
the RCAN1-4 protein. The RCAN1-1L protein concen-
tration was approximately double that of RCAN1-4 in
whole brain homogenates (combined regions). However,
our antibody specific for exon 4 binds with much greater
affinity to the RCAN1-4 protein, and produces a pro-
portionately stronger signal, than does our RCAN1-1
antibody (specific for exon 1), even though there is a
greater amount of RCAN1-1L. Thus, the actual quanti-
ties of RCAN1-1 and RCAN1-4 can only be directly
compared in western blots using the common antibody,
containing the exon 4 sequence.
RCAN1-1L is overexpressed in AD
Northern blots show that RCAN1 mRNA is upregulat-
ed in regions of the brain that are affected by AD, as
well as in a non-AD patients with neurofibrillary tan-
gles [6]. In this study, protein extracts originate from
regions of the brain including the cerebellum (Cb),
which should not be affected by AD and therefore can
serve as an internal control, and regions that are affec-
ted by AD, including the cerebral cortex (regions A10
and A22) and the hippocampus (Hc). To ensure that
effects were due to actual differences, and not loading,
membranes were stained with Ponceau S, and all sam-

RNA probe against exon 1. Expression was examined
in neurons, astrocytes and microglia, by labeling cells
with antibodies against each of these specific cell types.
We first created a construct that could produce both
an RCAN1-1 antisense and sense (control) transcript
for use as a radiolabeled probe (Fig. 4A). Our anti-
sense probe hybridized to tissue sample, as shown by
clusters of black grains, whereas our control, sense,
probe did not hybridize and only showed scattered
background grains (Fig. 4B). This indicates that our
system was working correctly.
Next we tested samples by labeling neurons, astro-
cytes, or microglia. We found that in both control and
AD postmortem samples, expression of RCAN1-1,as
shown by clusters of grains, highly colocalized with
neuronal cells and not with astrocytes or microglia
(Fig. 4C). The clusters were also larger and denser in
AD samples as compared with control samples. This is
in good agreement with our previously reported nor-
thern blot data, showing that RCAN1 mRNA expres-
sion is greater in AD than in age-matched control
samples [6]. Expression of RCAN1-4 also localized to
neurons, although, as it is expressed at low levels, its
concentration was still not dramatically higher than
background levels (Fig. 4C).
RCAN1-1 mRNA is overexpressed in neuronal
cells of AD patients
We examined mRNA expression of RCAN1 in brain
tissue from AD and age-matched control samples by
RT-PCR of cDNA (Fig. 5A). Upon quantification of

P-value of < 0.05) between the control and AD samples found was
in the RCAN1-1 protein in the Hc (marked with an asterisk). As
RCAN1-1 protein expression was approximately double that of
RCAN1-4 (Fig. 2B), the signal strength of the two isoforms has
been adjusted accordingly in this figure.
RCAN1 in Alzheimer’s disease C. D. Harris et al.
1718 FEBS Journal 274 (2007) 1715–1724 ª 2007 The Authors Journal compilation ª 2007 FEBS
good agreement with the increase in hippocampal
RCAN1-1 protein levels reported for AD patients in
Fig. 3C. Thus, it is possible that elevated RCAN1-1
protein concentrations in AD are the result of tran-
scriptional upregulation; this possibility will now have
to be rigorously tested.
Discussion
RCAN1 has been shown to bind to and inhibit the ser-
ine–threonine protein phosphatase calcineurin [5]. The
brain is an especially interesting organ in which to
examine RCAN1 expression, because calcineurin is
highly expressed in this organ, comprising approxi-
mately 1% of total protein. We have hypothesized that
a role for RCAN1 in the development of neurodegen-
erative ‘tauopathies’, such as AD, is that it may inhibit
calcineurin from dephosphorylating the tau protein,
resulting in hyperphosphorylated tau, which may then
promote the formation of paired helical filaments and
neurofibrillary tangles [6–8]. This fits nicely with data
from other studies showing decreased calcineurin activ-
ity in AD, and other data showing that calcineurin
inhibition results in tau phosphorylation on serine and
threonine residues consistent with those that occur in

neurons, because these clusters are denser.
C. D. Harris et al. RCAN1 in Alzheimer’s disease
FEBS Journal 274 (2007) 1715–1724 ª 2007 The Authors Journal compilation ª 2007 FEBS 1719
protein isoforms in human brain (Fig. 2A). We now
demonstrate that two of the possible protein isoforms,
RCAN1-1L and RCAN1-4, appear to be highly
expressed in brain, whereas RCAN1-1S is expressed at
very low levels (Fig. 2A). Our antibodies detect
RCAN1-4 at approximately 70 kDa, which is about
twice as large as the RCAN1-1S protein. This is also
much larger that has been described in other tissues
(25–29 kDa). There are several possible explanations
for this. First, there are additional stop codons located
in exon 7. One of these would produce a peptide con-
taining 595 amino acids, which would produce a protein
with a predicted size of 67 kDa, and another would pro-
duce a peptide containing 632 amino acids, which
would have a predicted size of 71.7 kDa. Another
explanation is that the protein may form a covalent
dimer (not a disulfide-linked dimer) that is not separ-
ated by SDS ⁄ PAGE. The expression of RCAN1-1L
protein was approximately double that of RCAN1-4 in
general, as determined by quantifying the densities of
bands detected with a common antibody that recognizes
all isoforms of the RCAN1 protein, in all regions of the
brain, and in samples from both AD and control
patients (Fig. 2B). Northern blots show that RCAN1
expression is upregulated in regions of the brain affec-
ted by AD, as well as in a non-AD patient exhibiting
neurofibrillary tangles [6]. Therefore, RCAN1-1 may be

, cerebral cortex area. RNA
was amplified using LA RT-PCR for 30 cycles: 98 °C for 20 s,
followed by 68 °C for 3 min. (B) The amount of input cDNA in
each sample was equalized by amplification of the GAPDH gene.
To ensure that GAPDH amplification was quantitative, we ran
serially diluted cDNA samples for different numbers of cycles.
Typically, it took about 25 cycles to achieve a linear dependency
between the amount of input DNA and the resulting PCR prod-
ucts. Then, equal amounts of the cDNA (according to amplifica-
tion of control GAPDH fragment) were used to estimate the
amount of RCAN1-1 mRNA. As with GAPDH amplification, seri-
ally diluted cDNA samples were run for different numbers of
cycles to find conditions in which the amount of amplified
RCAN1-1 fragments was proportional to the amount of the input
cDNA in the reactions. (C) Radioative In situ hybridization was
performed to label either RCAN1-1 or RCAN1-4 expression. This
technique was combined with immunocytochemistry to label spe-
cific cell types with antibodies. Anti-NeuN mAb was used to
label neurons, anti-GFAP was used to label astrocytes, and anti-
HLA-DR was used to label microglia. In this experiment, each
slide contained a set of one AD patient and one control patient
section, in triplicate. The hybridization signal of RCAN1-1 was
quantified in neurons by counting grain density on neurons and
subtracting background grain density levels. Standard errors were
calculated, and Fischer’s test was performed to analyze signifi-
cance. This shows approximately a two-fold increase in RCAN1-1
mRNA in the four Alzheimer’s disease versus four control patient
tissue samples.
RCAN1 in Alzheimer’s disease C. D. Harris et al.
1720 FEBS Journal 274 (2007) 1715–1724 ª 2007 The Authors Journal compilation ª 2007 FEBS

preferentially expressed in neurons, rather than astro-
cytes or microglia, in both normal brain tissue and
brain samples from AD patients. Therefore, there
are differences in levels of RCAN1-1 expression, but
there do not appear to be differences in the cell type
in which the different isoforms are expressed.
RCAN1-1 is upregulated not only in AD, but also
in non-AD brain tissue that exhibits one of the AD
hallmarks: neurofibrillary tangles. Chronically eleva-
ted RCAN1-1 levels may, thus, cause an increase in
phosphorylation of the tau protein, leading to the
formation of neurofibrillary tangles in a variety of
neurodegenerative tauopathies.
Experimental procedures
Postmortem human brain tissue
The brain samples used in this project were graciously pro-
vided by the Alzheimer’s Disease Research Center at the
University of Southern California’s Keck School of Medi-
cine, Los Angeles, CA. Brain tissues, with a postmortem
interval of less than 6 h, were fresh frozen at ) 70 °C until
use. Samples analyzed in this study originated from the Hc,
cerebral cortex region A10, cerebral cortex region A22, and
the Cb. All samples were accompanied by Alzheimer’s Dis-
ease Research Center neuropathology summaries and AD
samples, and all displayed between moderate and severe
disease pathology.
Antibodies
Antibodies to exon 7 (the common C-terminal region),
exon 1 and exon 4 of the RCAN1 gene were custom pro-
duced against peptides injected into rabbits, and affinity

)1
soy-
bean trypsin inhibitor) and were cleared by centrifugation
at 16 000 g after incubation on ice for 30 min. Protein con-
centrations were determined using the BCA protein assay
kit (Pierce, Rockford, IL), and equal amounts (20 lgof
each sample) were loaded onto SDS polyacrylamide gels
for fractionation. The samples were electrophoretically
transferred onto poly(vinylidene difluoride) membranes and
stained with Ponceau S to verify loading. The membranes
were then blocked in 5% nonfat dry milk (Bio-Rad, Hercu-
les, CA) with 0.1% Tween-20, and washed three times in
wash solution (NaCl ⁄ P
i
with 0.1% Tween-20). The mem-
branes were then probed with primary antibody at a dilu-
tion of 1 : 1000, washed in washing solution three times,
and then probed with a horseradish peroxidase-conjugated
secondary antibody at a dilution of 1 : 10 000 (Santa Cruz
Biotech). Membranes were washed three more times in
wash solution, and then visualized by use of the enhanced
chemiluminescent reagent (ECL kit; Amersham, Piscata-
way, NJ) and autoradiograpy. Films were scanned, and
expression was quantified using ipgel software. Bands were
normalized to b-tubulin expression. Membranes were
stripped in Pierce strip buffer and reprobed. Statistical ana-
lysis of western blot data was performed using statview
software, using Fisher’s PLSD test for significance.
C. D. Harris et al. RCAN1 in Alzheimer’s disease
FEBS Journal 274 (2007) 1715–1724 ª 2007 The Authors Journal compilation ª 2007 FEBS 1721

(GAPDH) probe. Probes containing [
32
P]dCTP[aP]-labeled
DNA were prepared using the High Prime system (Boehrin-
ger Mannheim, Mannheim, Germany). A PCR fragment
corresponding to RCAN1 isoform 1 was used to prepare
the RCAN1 probe, and a PCR fragment consisting of
GAPDH exons 7 and 8 was used to prepare GAPDH
probes.
In situ hybridization
Brain samples were sectioned and mounted onto positively
charged slides. Each slide contained samples from one spe-
cific brain region, with alternating AD and control samples.
Immediately prior to use, sections were air-dried and fixed
in freshly prepared 4% buffered paraformaldehyde. The
samples were then treated in acetic anhydride with 0.1 m
triethanolamine, and then rinsed and dehydrated in an
ethanol series and dried. Slides were incubated in prehy-
bridization solution [50% formamide, 0.75 m sodium chlor-
ide, 0.05 m sodium phosphate buffer (PB, pH 7.4), 0.01 m
EDTA, 0.15 mm dithiothreitol, 1% SDS, 5 · Denhardt’s
solution, 0.2 mgÆmL
)1
heparin, 0.5 mgÆmL
)1
tRNA,
0.05 mgÆmL
)1
polyA and polyC, and 0.25 mgÆmL
)1

(5¢-CCTGGTTTCACTTTCGCTGAAGATA-3¢). Amplified
fragments were then sequenced, and correct sequences were
cloned into the SmaI site of the pBluescript II SK vector,
between the recognition sites for the T3 and T7 polymeras-
es, so that both antisense and sense (control) RNA probes
could be produced from the same plasmid. To verify that
the correct sequence was inserted, and to determine the
orientation of the insert, all clones were sequenced.
These plasmids were transfected into Epicurian Coli
XL2-Blue ultracompetent cells (Stratagene, La Jolla, CA),
and grown. Plasmids were collected using the Wizard
Plus Miniprep kit (Promega, Madison, WI), and digested
with the appropriate restriction enzyme. Digestion of the
template was confirmed by resolution on an agarose
gel. Probes were produced using the Riboprobe in vitro
Transcription System (Promega), labeled with
35
S accord-
ing to the manufacturer’s protocol, and purified using
Mini Quick Spin columns (Qiagen, Valencia, CA). Probes
were then precipitated and dissolved in hybridization
solution.
Immunocytochemistry
Immediately following in situ hybridization, samples were
rinsed twice in NaCl ⁄ P
i
, and endogenous peroxidases were
blocked in NaCl ⁄ P
i
containing 10% methanol and 0.3%

with 1% Tween-20
three times for 5 min, and then incubated in preadsorbed
mouse secondary antibody for 1 h. Cell types were detected
using the Vectastain ABC kit (Vector Laboratories, Burlin-
game, CA), using diaminobenzidine as a substrate, accord-
ing to the manufacturer’s protocols. Immediately following
immunocytochemistry, slides were dehydrated in a 0.3 m
ammonium acetate series, and then dried and exposed to
film to estimate signal strength. Slides were then dipped in
NTB2 autoradiography emulsion (Kodak, Rochester, NY),
and incubated at 4 °C until development. In situ hybridiza-
tion was quantified on each specific cell type by counting
grain density on cells and subtracting background grain
density.
Long and accurate (LA) RT-PCR
The synthesis of first-strand cDNA was performed using
the SuperScript preamplification system from Life Technol-
ogies. One to three micrograms of total RNA per reaction
was reverse transcribed using oligo(dT) as the primer.
About 2 lL of the 20 lL total volume of cDNA was used
per PCR reaction. The LA RT-PCR method utilizes a mix-
ture of Taq polymerase and a small amount of a proofread-
ing polymerase, producing a reaction mixture with greatly
increased product fidelity, yield, length and reproducibility
over either enzyme alone. LA RT-PCR was performed
using a kit from Tamara Shuzo (TaKaRa Bio Inc.) and
conditions had been adjusted to ensure that results were in
a linear range and that a plateau had not been reached.
Primers used were as follows: (a) human RCAN1 mRNA
isoform 1, consisting of exons 4, 5, 6, and 7 ) the forward

3 Rothermel B, Vega RB, Yang J, Wu H, Bassel-Duby R
& Williams RS (2000) A protein encoded within the
Down syndrome critical region is enriched in striated
muscles and inhibits calcineurin signaling. J Biol Chem
275, 8719–8725.
4 Y ang J, Rothermel B, Vega RB, Frey N, McKinsey TA,
Olson EN, Bassel-Duby R & Williams RS (2000) Inde-
pendent signals control expression of the calcineurin
inhibitory proteins MCIP1 and MCIP2 in striated
muscles. Circulation Res 87, E61–E68.
5 Fuentes JJ, Genesca L, Kingsbury TJ, Cunningham KW,
Perez-Riba M, Estivill X & de la Luna S (2000)
DSCR1, overexpressed in Down syndrome, is an inhibi-
tor of calcineurin-mediated signaling pathways. Human
Mol Genet 9, 1681–1690.
6 Ermak G, Morgan TE & Davies KJ (2001) Chronic
overexpression of the calcineurin inhibitory gene
DSCR1 (Adapt78) is associated with Alzheimer’s
disease. J Biol Chem 276, 38787–38794.
7 Ermak G, Harris CD & Davies KJ (2002) The DSCR1
(Adapt78) isoform 1 protein calcipressin 1 inhibits calci-
neurin and protects against acute calcium-mediated
stress damage, including transient oxidative stress.
FASEB J 16, 814–824.
8 Ermak G & Davies KJ (2003) DSCR1 (Adapt78) ) a
Janus gene providing stress protection but causing
Alzheimer’s disease? IUBMB Life 55, 29–31.
9 Lian Q, Ladner CJ, Magnuson D & Lee JM (2001)
Selective changes of calcineurin (protein phosphatase
2B) activity in Alzheimer’s disease cerebral cortex. Exp

& Davies KJ (1997) Hamster adapt78 mRNA is a
Down syndrome critical region homologue that is indu-
cible by oxidative stress. Arch Biochem Biophys 342,
6–12.
17 Leahy KP & Crawford DR (2000) adapt78 protects cells
against stress damage and suppresses cell growth. Arch
Biochem Biophys 379, 221–228.
RCAN1 in Alzheimer’s disease C. D. Harris et al.
1724 FEBS Journal 274 (2007) 1715–1724 ª 2007 The Authors Journal compilation ª 2007 FEBS