Kalrn promoter usage and isoform expression respond to chronic cocaine exposure.

Kal7 is expressed in the major neuronal cell types in nucleus accumbens and striatum. Male Tg(Drd1a-EGFP)X60Gsat/Mmmh mice were deeply anesthetized, fixed by perfusion with 4% paraformaldehyde, and brains were sectioned as described [17]. Sections were stained for GFP (green) and Kal7 (red) and imaged with a confocal microscope. A-C. NAc core. D-F. Striatum. Scale bar, 20 μm. A collapsed Z-stack (0.5 μm steps) is shown. A total 249 neurons in six sections from two mice (GFP+Kal7, Kal7 only and GFP only) were counted in the striatum: 98% of the GFP-positive neurons contained Kal7 and 2% did not; 33% of the Kal7-positive neurons did not contain GFP. A total of 239 neurons in six sections from two mice were counted in the NAc: 92% of the GFP-positive neurons contained Kal7 and 8% did not; 31% of the Kal7-positive neurons did not contain GFP.

Potential Sp1 sites in proximal regions of ΔKal, KalB and KalC promoters. The 1 kilobase of genomic sequence 5' to the Δ, KalB and KalC transcriptional start sites was analyzed using Transcription Factor Search (http://www.cbrc.jp/research/db/TFSEARCH.html); mouse ΔKal (ENSMUST00000114949), rat ΔKal (AF229255.1), human ΔKal (BM920956.1; NM_001024660.2 starts 291 nt further 3'), mouse KalB (U88156.1) and mouse KalC (AF230644.1; NM_177357). The Sp1 consensus site definition is from [53]. Only one human cDNA corresponding to the mouse and rat ΔKal transcripts was identified; the 5' extent of the human transcript may be longer than depicted here, making the distance from the Sp1 site to the start of transcription shorter.

Structure of the mouse Kalrn gene and major transcripts. Genomic and transcript information was compiled from websites (http://www.ensembl.org/Mus_musculus/Info/;http://genome.ucsc.edu/cgi-bin/) and our previous work [26-28] to create an accurate map of the mouse Kalrn gene and the major brain transcripts, as described in Methods. Protein domains (http://smart.embl-heidelberg.de/) encoded by groups of exons are indicated.

Full-length Kalirin promoter usage varies in different brain regions. A. Putative promoter regions (red line) and initiation exons in the mouse Kalrn gene, which is located on mouse chromosome 16, were identified based on homology to human and rat Kalrn [27,28]; translational start sites are indicated by bent arrows. The introns separating promoters B, C, A and D, which produce full-length Kalirin, are not drawn to scale. The location of each initiation exon is indicated by the nucleotide number below the line; the location of the initiation exon-specific forward primers is indicated, along with the common reverse primer in Exon 3. B. RNA prepared from adult mouse prefrontal cortex (PFC), hippocampus (Hippo) and nucleus accumbens (NAc) was subjected to Q-PCR analysis using these primers; data were normalized to GAPDH. The same data are plotted on a log scale to allow visualization of data for full-length Kalirin promoters A and D, and as a percentage of the total in each brain region.

Expression of transcripts encoding different Kalirin isoforms varies in nucleus accumbens and prefrontal cortex. A. The proteins encoded by the major isoforms of rat and human Kalirin are drawn to scale; although Δ-isoforms of Kal8, -9 and -12 have been identified [26-28], only ΔKal7 is depicted. The ΔKal isoforms are produced when the ΔKal promoter is used instead of the A, B, C, or D promoters diagramed in Figure 2. The sense and anti-sense primers used for real-time Q-PCR are shown in red; 5'- and 3'-untranslated regions of the corresponding transcripts are indicated by red lines. The Kal7-specific and Kal-spectrin antibody epitopes are indicated by blue lines. B. RNA prepared from mouse nucleus accumbens (NAc) and prefrontal cortex (PFC) was subjected to real-time Q-PCR analysis using the ΔKal and full-length primers; three separate preparations of RNA from each tissue were analyzed in triplicate and data were normalized to GAPDH. C. PFC and NAc expression of Kalirin isoforms was examined via Q-PCR analysis of 3'-terminal exons encoding Kal7, Kal8, Kal9 and Kal12. Higher levels of Kal7 transcript were seen in the PFC than the NAc.

Kalirin is not as highly expressed in the striatum as in the cortex. A. Tissue taken from the cortex (Ctx), dorsal striatum (Str) and nucleus accumbens (NAc) of adult C57BL/6 mice was disrupted by sonication in SDS lysis buffer; aliquots containing equal amounts of protein (20 μg) were fractionated by SDS-PAGE and visualized using antibody to the spectrin repeat region of Kalirin or to Kal7 and the indicated proteins: βIII tubulin, NR2B, Cdk5 and Rac1.

Kal7 and Rac1 are enriched in mouse striatal PSDs. Adult mouse striatum homogenized in buffer containing 0.32 M sucrose was subjected to differential centrifugation as described in Methods. After hypotonic lysis, synaptosomal membranes were pelleted, resuspended and subjected to equilibrium sucrose density centrifugation. Fractions were collected from the top (Sample) to bottom [Pellet (mitos)] of the gradient; the fraction taken from the interface of the 1.0/1.2 M sucrose layers was extracted with TX-100, yielding a TX-soluble fraction (TxS) and the PSD fraction. Equal amounts of protein (5 μg) from each fraction were subjected to SDS-PAGE; Kalirin was visualized using the Kal-spectrin antibody and the Kal7 antibody. The NR2B subunit of the NMDA receptor served as a PSD marker; BiP served as an endoplasmic reticulum marker. Rac1, one of the substrates of Kal-GEF1, was also localized to PSDs. Cyt = cytosol; ER/G = endoplasmic reticulum/Golgi-enriched fraction.

Striatal PSDs contain less Kalirin than cortical or hippocampal PSDs. TX-100 extracted PSDs were prepared from adult mouse cortex, hippocampus and striatum as described above. A. The indicated amount of protein was subjected to SDS-PAGE and Kal7 was visualized using the Kal7-specific antibody. B. Known amounts of purified ΔKal7 were analyzed at the same time, making it possible to calculate fmol Kal7/μg PSD protein using Gene Tools; data for two different amounts of protein were averaged; error bars show the range. The cortex and hippocampus numerical data in B are repeated from [17] for comparison.

Chronic exposure to cocaine increases usage of the ΔKal promoter and the Kal7 3'-terminal exon. Samples of RNA were prepared from the nucleus accumbens of three sets of mice treated with cocaine (20 mg/kg for 7 days) or saline. A. Promoter usage was evaluated using the primers described in Figure 1. Cocaine treatment had no effect on usage of full-length Kalrn promoters A, B, C, or D. B. Chronic cocaine treatment increased usage of the ΔKal promoter. C. Usage of the different Kalirin 3'-terminal exons was evaluated as described in Figures 2-3; usage of the Kal7-specific 3'-terminal exon increased following chronic cocaine treatment.

Effect of chronic exposure to cocaine on expression of Kalirin protein. A. Striata harvested from Saline and Cocaine treated mice were subjected to subcellular fractionation; equal amounts (10 μg) of input (S1) and PSD were fractionated by SDS-PAGE. Kalirin proteins were visualized using the Kal-spectrin antibody. DARPP32, a soluble protein highly expressed in the striatum, was excluded from the PSD fraction. B. Striata prepared in a similar fraction to A show increases in Kal7 in the S1 and PSD fractions while demonstrating a lack of Kal7 in the Triton soluble fraction as was expected from previous results [17,18]. C. Levels of Kal7 and Kal9 in SDS lysates prepared from the nucleus accumbens of chronically Saline and Cocaine treated mice were compared to levels of βIII-tubulin; Kal12 levels were not high enough to allow accurate quantification. The Kalirin/Tubulin ratios for Kal7 and Kal9 were calculated for Saline and Cocaine treated mice, with the Saline ratio normalized to 100%; using an unpaired t-Test, Kal7 expression (p < 0.05) differed significantly while Kal9 expression did not.