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 K. J. A. Davies, Ethel Percy Andrus Gerontology Center, University of Southern California, 3715 McClintock Avenue, Los Angeles, CA 90089-0191, USA Fax: +1 213 740 6462 Tel: +1 213 740 8959 E-mail: kelvin@usc.edu Note The new name RCAN1 (regulator of cal- cineurin) has recently been accepted by the HUGO Gene Nomenclature Committee for the gene previously known as DSCR1 or Adapt78. Similarly, RCAN1 is the new name for its protein product, which was previously know as calcipressin 1 or MCIP1 (Received 24 April 2006, revised 16 Decem- ber 2006, accepted 29 January 2007) doi:10.1111/j.1742-4658.2007.05717.x At least two different isoforms of RCAN1 mRNA are expressed in neuro- nal cells in normal human brain. Although RCAN1 mRNA is elevated in brain regions affected by Alzheimer’s disease, it is not known whether the disease affects neuronal RCAN1, or if other cell types (e.g. astrocytes or microglia) are affected. It is also unknown how many protein isoforms are expressed in human brain and whether RCAN1 protein is overexpressed in Alzheimer’s disease. We explored the expression of both RCAN1-1 and RCAN1-4 mRNA isoforms in various cell types in normal and Alzheimer’s disease postmortem samples, using the combined technique of immunohist- ochemistry and in situ hybridization. We found that both exon 1 and exon 4 are predominantly expressed in neuronal cells, and no significant expression of either of the exons was observed in astocytes or microglial 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- phorylation on serine and threonine residues, consis- tent with those that occur in AD [9–13]. RCAN1 is expressed primarily in neurons in both rat and human brain tissue [6]. Importantly, this complements data from rat tissues showing that calcineurin is also expressed in neurons [14,15]. RCAN1 gene expression is significant in several tis- sues, particularly human brain, spinal cord, kidney, liver, mammary gland, placenta, skeletal muscle, and heart [6]. We have previously found that there are Detection of RCAN1 Isoforms with Various Antibodies Quantification of RCAN1-1L and RCAN1-4 Isoforms RCAN1 Antibody Used Isoform Detected RCAN1 Expression in Human Brain A B Fig. 2. RCAN1 protein expression in human brain. Using an anti- body directed at exon 7, which should recognize all RCAN1 iso- forms, three bands were detected by western blotting (A). These bands appear at 70, 38 and 31 kDa. All three bands were present regardless of the brain region tested or the presence or absence of disease, although the 31 kDa band, in some cases, was very faint. Using antibodies specific to exon 1 or exon 4, we have identified the 70 kDa band as RCAN1-4, the 38 kDa band as RCAN1-1L, and the 31 kDa band as RCAN1-1S. Combined data from western blots from 12 control and 12 AD patients, in all regions tested (A10, A22, Hc and Cb), were quantified by densitometry, and standard errors were calculated (B). Fisher’s test was performed to analyze whe- 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’ 15 NFAT binding sites RCAN1 Genomic DNA RCAN1-1L Protein RCAN1-4 Protein 21.2 2 q 31.22q DSCR Fig. 1. Structure of the RCAN1 gene and the RCAN1 protein. Chro- mosome 21 ) Human RCAN1 is located on chromosome 21 in region q22.12, just outside of the Down’s syndrome candidate region. RCAN1 genomic DNA ) RCAN1 consists of seven exons that are alternatively spliced and vary in their 5¢-exon, but all contain exons 5, 6, and 7. There is a cluster of 15 NFAT-binding sites on the RCAN1 gene, 5¢ to exon 4 which may function as an alternative promoter region for the exon 4 splice variant. RCAN1 protein ) We have found evidence for three RCAN1 protein isoforms in human brain, RCAN1-1S, RCAN1-1L, and RCAN1-4 (see Fig. 2 for these data). All RCAN1 isoforms differ in their initial exon, but share the 168 amino acids encoded by exons 5, 6 and 7, as well as the con- served FLISPP sequence found in all of the RCAN1 family mem- bers. This motif shares homology with the serine–proline (SP) boxes found in NFAT protein family members. All RCAN1 isoforms contain a putative calcineurin-binding motif (PKIIQT). RCAN1 in Alzheimer’s disease C. D. Harris et al. 1716 FEBS Journal 274 (2007) 1715–1724 ª 2007 The Authors Journal compilation ª 2007 FEBS at least two isoforms of RCAN1 expressed at signifi- 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 the RCAN1 gene are expressed in brain tissue, no one has reported any differences in function or localization of these isoforms in the brain. In order to try to understand how these isoforms may differ, we exam- ined the expression of the isoforms known to be tran- scribed in brain tissue, isoforms RCAN1-1 and RCAN1-4. To determine which, if any, protein iso- forms were expressed, brain tissue extracts were pre- pared for western blotting. These blots were first probed with an antibody raised against exon 7, which is a portion of the C-terminus of RCAN1. This region is common to all predicted isoforms, and the antibody should therefore recognize all forms of the RCAN1 protein. The antibodies were first tested on cell extracts to ensure reactivity. After the antibody had been affin- ity purified, it recognized two major bands, and one very light band, in brain lysates by western blot analy- sis (Fig. 2A). These bands resolved at approximately 70, 38 and 31 kDa on denaturing polyacrylamide gels. Next, antibodies specific for exon 1 or exon 4 were tested on adult human brain tissues, again using west- ern blotting, to try to match each band with the unique isoform of the RCAN1 protein to which it cor- responded. The band around 70 kDa was recognized by the exon 4 antibody as RCAN1-4, and the 38 and 31 kDa bands were recognized by the exon 1 antibody as RCAN1-1L and RCAN1-1S isoforms, respectively (Fig. 2A). As this antibody is generated against a pep- tide present in both RCAN1-1 isoforms, it recognizes both bands. RCAN1-1S was the minor band, which 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- ples were normalized to loading controls. We found that RCAN1-1L was upregulated in brain regions affected by AD as compared to control tissues (a rep- resentative western blot is shown in Fig. 3A). As human brain tissue is difficult to obtain, we focused on the most interesting regions for further studies. These regions included the Hc and the Cb (for internal control). We found that there was significant upregulation of RCAN1-1L in the Hc of AD patients, but regulation did not appear to be significant for 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 1717 RCAN1-4 (Fig. 3B). Using Fisher’s protected least significant difference (PLSD) test on RCAN1-1L expression data, AD Hc was significantly different from control Hc (P<0.05). Using Fisher’s PLSD test on RCAN1-4 expression data, no group was signifi- cantly different from any other. As the RCAN1-1S isoform was difficult to detect, and represents only a very minor fraction of RCAN1-1 expression, we have not included it here. Cellular distribution of RCAN1-1 mRNA in human brain As RCAN1-1 protein was upregulated in AD tissues, we wanted to determine if there were any differences in which brain cell types expressed RCAN1. We again focused on RCAN1-1, as it was upregulated in AD, using the combined techniques of in situ hybridization and immunocytochemistry. In this experiment, expres- sion of RCAN1-1 was identified using an antisense 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 PCR, our results showed a clear upregulation of RCAN1-1 mRNA in the primary region that is affected by AD, the Hc (Fisher’s P-value of < 0.05). Expression was not significantly increased in the Cb, as would be expected, as this region is not affected by AD (Fig. 5B). When quantifying mRNA expression in neurons from our combined in situ hybridization–immunocyto- chemistry, we obtained similar results to the RT-PCR data above. By quantifying the grain cluster density associated with a neuron, and subtracting the back- ground expression density, expression of mRNA in AD and control samples can be determined. With this method, it appears that expression of RCAN1 is almost doubled in AD (Fisher’s P-value of < 0.05) compared to control samples (Fig. 5C). The increase in RCAN1-1 mRNA levels seen in the Hc (but not the Cb) of AD patients in Fig. 5B,C is in Fig. 3. RCAN1-1 is overexpressed in AD. RCAN1-1L and RCAN1-4 protein expression was measured via western blot [a representa- tive blot is shown in (A)] in controls and AD patients. Blots contain- ing control and AD patient samples were probed with antibody against exon 1, stripped, and then probed with antibody against exon 4. Ponceau S staining of membranes, and probing of blots with b-tubulin antibody, were used to control loading levels. In (B), densities of the bands from 12 control and 12 AD patient samples were quantified using IPGEL Laboratory software, and normalized to a b-tubulin loading control, and standard errors were calculated. Fisher’s test was performed to analyze whether differences were statistically significant. The only significant difference (producing a 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 AD [9–13]. In the studies presented in this article, we provide evi- dence for the presence of at least three distinct RCAN1 C A B Fig. 4. Analysis of RCAN1 mRNA expres- sion in human brain. Probes for in situ hybridization were created by cloning RCAN1-1 into the multiple cloning site of the pBluescript II SK(+ ⁄ –) vector (A). Use of this vector allowed for both sense (control) and antisense probes to be produced from a single clone. The sense probe did not hybridize to the sample, whereas the anti- sense probe did (B). In all slides, specific cell types (either neurons, astrocytes or microglia) are immunochemically stained with diaminobenzidine and appear brown. Cell type-specific antibodies used were: anti-NeuN mAb for neurons, anti-GFAP for astrocytes, and anti-HLA-DR for microglia, and are shown at a magnification of 200·. Expression of RCAN1-1 mRNA was detec- ted by in situ hybridization, in which hybrid- ization produces clusters of black grains. Representative samples show that clusters align with neurons in both control and AD samples, but not with astrocytes or micro- glia (C). Expression is clearly higher in AD 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 related to this particular AD pathology. Expression of the RCAN1-1L protein was greater in AD patients as compared to age-matched control patients. We found that, regardless of the isoform, RCAN1 was expressed in each region of brain tested, in both AD and control samples (Fig. 3A). We found, however, that RCAN1-1L was the only isoform clearly upregulated in AD, as compared to age-matched con- trol samples (Fig. 3B). Thus, RCAN1-1L may play a role in AD, whereas RCAN1-4 does not appear to be involved in this pathology. As RCAN1-1 protein is overexpressed in AD, we next examined its mRNA transcript expression at the cellular level, to see if there were any differences in localization between AD and control samples. As RCAN1-1S represents a minor proportion of total A B C Fig. 5. RCAN1 mRNA is overexpressed in AD. (A) RCAN1-1 mRNA expression was detected using RT-PCR in AD and control samples. Amplification of GAPDH was used as a loading control. A 10 –A 10 , cerebral cortex area; A 22 –A 22 , 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 RCAN1 expression, we reasoned that signals detected by our probe against exon 1 would be predominantly due to expression of RCAN1-1L. We found that RCAN1-1 is expressed in neurons in both AD and control samples as detected by in situ hybridization (Fig. 4C). It does not appear to be expressed in micro- glia or astrocytes. RCAN1-4 is expressed at a much lower level, and therefore difficult to detect by this method, but also appears to be expressed in neurons. When RCAN1-1 expression is measured by RT-PCR, it is also seen to be upregulated selectively in brain regions affected by AD, as compared to (age-matched) control patients (Fig. 5A). Quantification of RCAN1-1 mRNA expression in neurons also shows that it is upregulated in AD (Fig. 5B). Both RCAN1-1 and RCAN1-4 are expressed in brain tissue in both control and AD patients. RCAN1-4 is expressed at a much lower level, how- ever, and it does not appear to play a role in this disease. RCAN1-4 is under the control of an alter- native promoter and is also feedback regulated, which may account for, at least in part, differences in regulation of the different RCAN1 isoforms [4]. It has been shown that RCAN1-4 is expressed as a stress-protective protein [16], which can arrest cell growth, whereas RCAN1-1 can induce cellular growth [7,17]. The data presented in this article show that RCAN1-1 is upregulated at both the mRNA and protein levels in AD, and therefore may contrib- ute to disease pathology. RCAN1-1 appears to be 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 purified by ProSci Incorporated (Poway, CA). An exon 1 antibody was generated against the peptide NH 2 - MEEVDLQDLPSAT-OH, and an exon 4 antibody was produced against the peptide NH 2 -VANSDIFSESETR- OH. The antibody against exon 7 was created as previ- ously described [7]. After production, sera, purified anti- bodies and flow-through were tested, along with competitive binding assays. Commercially produced b-tub- ulin and secondary antibodies were purchased from Santa Cruz Biotech (Santa Cruz, CA). Experimental animals were handled according to NIH guidelines for the care and use of laboratory animals. Western blotting Extracts were prepared by homogenization in cell lysis buffer (1 · NaCl ⁄ P i , 1% Igepal, 0.1% SDS, 0.1 mgÆmL )1 phenylmethanesulfonyl fluoride, 1 lgÆmL )1 leupeptin, 1 lgÆmL )1 pepstatin A, 1 lgÆmL )1 antipain, 10 lgÆmL )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 RNA isolation Total RNA was extracted using the TRIzol reagent (Life Technologies, Gaithersburg, MD). The RNA concentration was quantified spectrophotometrically, and relative content was further confirmed with ethidium bromide-stained gels. Integrity of the RNA was estimated by agarose gel electrophoresis. Only RNA samples displaying discrete 28S and 18S bands were used in experiments. Northern hybridization Samples containing 10 lg of total RNA were subjected to electrophoresis through 1% agarose formaldehyde gels, blotted onto nylon membranes (Oncor, Gaithersburg, MD) with HETS (CINNA ⁄ BIOTECX, Houston, TX), and cross- linked by ultraviolet radiation. The membranes were then prehybridized for 4 h and hybridized for 15 h in Hybrizol I (Oncor) at 42 °C. They were washed with 2 · NaCl ⁄ Cit + 0.1% SDS at room temperature for 1 and 10 min, and then with 0.1 · NaCl ⁄ Cit + 0.1% SDS at 60 ° C for 10 and 30 min. The membranes were exposed, developed, and scanned using the PhosphoImager system (Molecular Dynamics, Sunnyvale, CA). To rehybridize blots, probes were removed by washing membranes in a solution contain- ing 0.1 · NaCl ⁄ Cit + 0.1% SDS and 10 mm Tris ⁄ HCL (pH 7.0) at 90 °C for 10 min. To quantify levels of RCAN1 mRNA, the membranes were scanned, and the hybridiza- tion signal was measured using imagequant software (Molecular Dynamics). Each signal was recalculated according to the amount of RNA actually loaded onto the gels. The amount of the loaded RNA was controlled using a glyceraldehyde-3-phosphate dehydrogenase gene (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 sheared salmon sperm DNA] for 30 min at 53 °C in humidified chambers. Prehybridization solution was then removed, and slides were hybridized to either antisense or sense (control) 35 S-labeled probes, cover-slipped, and incubated at 53 °C for 3 h in hybridization solution (prehybridization solution plus 10% dextran sulfate). Slides were soaked in 4 · NaCl ⁄ Cit and 100 mm b-merca- ptoethanol to remove coverslips. After coverslips were removed, and slides were soaked in 0.5 m sodium chloride and 0.05 m phosphate buffer pH 7.4 for 10 min at room temperature; this was followed by incubation with 0.025 mgÆmL )1 RNaseA in 0.5 m sodium chloride and 0.05 m PB, for 30 min at 37 °C. The slides were then washed in a criterion wash solution, containing 50% formamide, 0.5 m sodium chloride, 0.05 m PB and 100 mm b-mercapto- ethanol, for 30 min at 50 °C, and then finally washed over- night in 0.5 · NaCl ⁄ Cit and 20 mm b-mercaptoethanol. RNA probe preparation Exon 1 and exon 4 sequences of RCAN1 were amplified from human cDNA by RT-PCR, using the LA-PCR kit (TaKaRa Bio Inc., Kusatsu, Japan). Primers used to amplify exon 1 consisted of the first 25 bases of exon 1 (5¢-GACTGGAGCTTCATTGACTGCGAGA-3¢) and the last 24 bases of exon 1 (5¢-CCGGCACAGGCCGTCCACG AACAC-3¢); primers for amplifying exon 4 consisted of the first 25 bases of exon 4 and the last 25 bases of exon 4 (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% RCAN1 in Alzheimer’s disease C. D. Harris et al. 1722 FEBS Journal 274 (2007) 1715–1724 ª 2007 The Authors Journal compilation ª 2007 FEBS hydrogen peroxide. After being washed in NaCl ⁄ P i , slides were treated with 1% NP-40 in NaCl ⁄ P i , and then washed again in NaCl ⁄ P i . After blocking for 30 min in blocking solution (NaCl ⁄ P i , 0.01 mgÆmL )1 heparin, 10 lm dithiothre- itol, 100 unitsÆmL )1 RNase inhibitor, and 3 lLÆmL )1 sera), samples were incubated with primary antibody for 90 min. Cell type-specific antibodies used were: anti-neuronal nuclei (NeuN) IgG from Chemicon (Temecula, CA) for neurons (1 : 500), anti-(glial fibrillary acidic protein) (GFAP) from Chemicon for astrocytes (1 : 30), and anti-(human leuko- cyte antigen-DR) (HLA-DR) from Dako for microglia (1 : 500). Slides were then rinsed in NaCl ⁄ P i 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 primer was 5¢-GACTGGAGCTTCATTGACTGCGAGA-3¢, corresponding to bases 79–103 of exon 1 (bases 1–25 of the short exon 1-containing isoform), and the reverse primer was 5¢-ACCACGCTGGGAGTGGTGTCAGTCG-3¢, cor- responding to bases 1–25 of exon 7; (b) human RCAN1 mRNA isoform 4, consisting of exons 1, 5, 6, and 7 ) the forward primer was 5¢-AAGCAACCTACAGCCTCTTGG AAAG-3¢, corresponding to bases 1–25 of exon 4, and the reverse primer was the same primer used to amplify iso- form 1; and (c) human GAPDH, for which the primers and conditions were the same as previously described [8]. All DNA fragments produced by LA RT-PCR were veri- fied by sequencing, using an ABI Prism377 DNA sequencer (Perkin-Elmer, Waltham, MA) in our core facility. Acknowledgements The authors wish to acknowledge the generous support of NIH ⁄ NIA grant no. AG 16256. Tissue for this study was obtained from the Alzheimer’s Disease Center Neuropathology Core, Keck School of Medicine, Uni- versity of Southern California, Los Angeles, CA, which is funded by P59-AG05142, National Institute of Aging. References 1 Hattori M, Fujiyama A, Taylor TD, Watanabe H, Yada T, Park HS, Toyoda A, Ishii K, Totoki Y, Choi DK et al. (2000) The DNA sequence of human chromo- some 21. Nature 405, 311–319. 2 Fuentes JJ, Pritchard MA & Estivill X (1997) Genomic organization, alternative splicing, and expression pat- terns of the DSCR1 (Down syndrome candidate region 1) gene. Genomics 44, 358–361. 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 Neurol 167, 158–165. 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 1723 10 Ladner CJ, Czech J, Maurice J, Lorens SA & Lee JM (1996) Reduction of calcineurin enzymatic activity in Alzheimer’s disease: correlation with neuropathologic changes. J Neuropathol Exp Neurol 55, 924–931. 11 Gong CX, Singh TJ, Grundke-Iqbal I & Iqbal K (1994) Alzheimer’s disease abnormally phosphorylated tau is dephosphorylated by protein phosphatase-2B (calcineu- rin). J Neurochem 62, 803–806. 12 Drewes G, Mandelkow EM, Baumann K, Goris J, Mer- levede W & Mandelkow E (1993) Dephosphorylation of tau protein and Alzheimer paired helical filaments by calcineurin and phosphatase-2A. FEBS Lett 336, 425–432. 13 Kayyali US, Zhang W, Yee AG, Seidman JG & Potter H (1997) Cytoskeletal changes in the brains of mice lacking calcineurin A alpha. J Neurochem 68, 1668– 1678. 14 Goto S, Matsukado Y, Mihara Y, Inoue N & Miya- moto E (1986) The distribution of calcineurin in rat brain by light and electron microscopic immunohisto- chemistry and enzyme-immunoassay. Brain Res 397, 161–172. 15 Kuno T, Mukai H, Ito A, C hang CD, Kishima K, Saito N & Tanaka C (1992) Distinct cellular expression of calci- neurin A alpha and A beta in rat brain. J Neurochem 58, 1643–1651. 16 Crawford DR, Leahy KP, Abramova N, Lan L, Wang Y & 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 . many protein isoforms are expressed in human brain and whether RCAN1 protein is overexpressed in Alzheimer’s disease. We explored the expression of both. mRNA isoforms in various cell types in normal and Alzheimer’s disease postmortem samples, using the combined technique of immunohist- ochemistry and in situ