The expression of the Cln8 gene in mouse tissues was first analyzed by northern blot and real-time quantitative RT-PCR analyses. In northern blot analysis one ~3 kilobase (kb) Cln8 specific transcript was detected in all tissues analyzed including a 14-day embryo (E14) (Fig. 1A). In addition, one ~7 kb transcript was seen in all tissues except testis and heart (Fig. 1A). In spleen, an additional transcript of ~2 kb was detected (Fig. 1A). The ~7 kb and ~2 kb transcripts were notably weaker than the ~3 kb transcript. RT-PCR analysis covering the 867 bp open reading frame of Cln8 resulted in a single fragment of the same size in 12 different mouse tissues (data not shown), suggesting that the different transcripts detected in the northern analysis are not due to alternative splicing in the coding region.

We then estimated the level of Cln8 expression using real-time quantitative RT-PCR analysis. Cln8 showed highest expression in liver, spleen, heart and skeletal muscle, 2.5, 2.2, 1.7 and 1.6 fold higher (respectively) than in the brain (Fig. 1B). Expression was lowest in 11-day embryo and adult kidney, lung and testis, 0.6, 0.9, 0.5 and 0.03 fold lower than in the brain (Fig. 1B). This indicates that Cln8 expression levels in adult mouse tissues as well as in whole mouse embryos are not dramatically different.

The tissue expression of Cln8 during development and brain maturation was characterized using radioactive mRNA in situ hybridization analysis. For this purpose, whole embryo sections from E13, E15.5 and E17 mice and brain sections from postnatal (P) P0, P5, P10 and adult mice were used and the expression of Cln8 was quantitated by a computer based MCID image analysis system.

In fresh-frozen E13 and E17 embryos the overall expression of Cln8 was weak. In E17 the most prominent expression was detected in the developing gastrointestinal tract (Fig. 2A, 2B). In addition, there was high expression in the DRG neurons (Fig. 2C, 2D). Some Cln8-specific signal was also detected in the developing brain (Fig. 2E). The expression of Cln8 in E13 was similar to that in E17 except in the DRG neurons, where no signal was detected (data not shown). In paraffin-embedded E15.5 embryos a Cln8-specific signal was detected in the developing gastrointestinal tract, adrenal glands and the developing brain, especially in the cortical plate (data not shown).

In postnatal mice (P0, P5, P10 and adult) Cln8 expression was quantitated in the cortex, cerebellum and different hippocampal regions. Expression at P0 was low but specific in hippocampal regions CA1, CA3 and the granular cell layer of the dentate gyrus and in the cortex (Fig. 3). At P5 and P10, the Cln8 expression increased in the hippocampal CA1 region (to 115% and 106% of P0) and especially in the CA3 region (to 139% of P0) (Fig. 3). In adult brain the expression levels of Cln8 were lower than at P0, P5 and P10 in every region analyzed (Fig. 3). When compared with adults, a statistically significant difference in Cln8 expression was detected in the cortex at P0, P5 and P10 and in the CA3 region at P10 (Fig. 3). Expression of Cln8 in adult cerebellum was 156% of expression in the cortex (data not shown). In all maturation stages, the expression of Cln8 was highest in hippocampal region CA3 (Fig. 3).

Epileptic seizures are a characteristic feature in human patients with CLN8 mutations. We therefore analyzed, by mRNA in situ hybridization analysis, possible changes in Cln8 expression 2 h, 6 h and 24 h after kindling-induced epileptic seizures. Animals were subjected to 40 rapid kindling stimulations which lead to increased excitability in mice at 4 weeks after the stimulation 25. Each rapid kindling stimulation induced relatively short focal (grade 0–2) or long-lasting generalized (grade 4–5) seizures. The mean duration of afterdischarges was 20 ± 3 s for focal seizures and 76 ± 8 s for generalized seizures. All animals experienced multiple (11 ± 2) grade 4–5 generalized tonic-clonic seizures. The expression of Cln8 in the control, electrode implanted but non-stimulated mice (0 h; Fig. 4) was essentially identical to that in wild-type adult mouse brain (Fig. 3). After kindling stimulations Cln8 was strongly up-regulated in the hippocampus, especially in the CA3 region and most prominently in the granular cell layer of the detate gyrus (Fig. 4). After 2 h, 6 h and 24 h expression of Cln8 in CA3 was 92%, 114% and 134% of that of controls, and in the granular cell layer of the dentate gyrus 128%, 225% and 286% of controls, respectively (Fig. 4). When compared to controls, a statistically significant increase in Cln8 expression was detected in the cortex after 6 h, and in the CA1 and CA3 regions, and the granular cell layer of the dentate gyrus after 2 h, 6 h and 24 h (Fig. 4). All changes in Cln8 mRNA expression after rapid kindling stimulations were bilateral with no differences between the sides.

FirstChoice™ Northern Blot Mouse Blot I (Ambion, Austin, TX, USA) was used to analyze the distribution of Cln8 expression in various tissues. An 872 base pair (bp) deoxyribonucleic acid (DNA) fragment including a Cln8 open reading frame of 867 bp, digested from the SvPoly-Cln8 expression vector with restriction enzyme BamHI, was used as a probe. The hybridization was performed according to the manufacturer's instructions with Ultra sensitive Hybridization buffer (Ambion), and salmon sperm DNA and Cot-1 DNA for blocking.

Real-time quantitative RT-PCR analysis of Cln8 was performed in an ABI PRISM^® 7000 SDS thermal cycler (Applied Biosystems, Foster City, CA, USA) using 5 μl of each tissue from Mouse Multiple Tissue cDNA (MTC™) Panel II (BD Biosciences) as template. Each 25 μl reaction contained 300 nM of reverse primer 5'-CCACTGGTTGGCCTTCCA-3' and 900 nM of forward primer 5'-GCCCTTCACCTGCATTTCC-3' as well as 200 nM of Cln8 specific probe 6FAM-TGACCACCCAGCCTTCAGGAGCAT-TAMRA in TaqMan^® Universal PCR Master Mix (Applied Biosystems). The primers were optimized for analysis and the PCR fragment of 76 bp covered the region from 510 to 585 bp of the Cln8 open reading frame (GenBank sequence accession no AF125307). The Cln8 specific probe included a reporter dye 6-carboxyfluorescein (FAM) and a quencher dye 6-carboxy-tetramethyl-rhodamine (TAMRA). Endogenous control reactions were performed using Assays-on-Demand Mm00446973_m1 TATA-box binding protein amplification according to the manufacturer's instructions (Applied Biosystems). Each reaction was performed in triplicate. Standard complementary DNA (cDNA) solutions, generated from C57/BL mice cortex mRNA, and standard curve method were used (User Bulletin #2: ABI PRISM(R) 7700 Sequence Detection System. P/N 4303859B (2001) Applied Biosystems. 36 s.). Statistical analyses were performed using Microsoft Excel.

The DNA fragment covering the whole open reading frame of Cln8 was PCR amplified using primers 3'-GTGATTTCTCCGGTGCTAGG-5' and 3'-GCAACCACCATTTCTCAGGT-5'.

Paraffin and cryosections of NMRI or C57/BL mice were prepared and analyzed by mRNA in situ hybridization as previously described 434445 with slight modifications. The isolated tissues from postnatal wild-type and kindling mice were fresh-frozen and sectioned. Embryos were either fixed in 4% paraformaldehyde and embedded in paraffin or fresh-frozen and sectioned. For all serial sections of 8–14 μm, a single ³⁵S-labeled 370 bp Cln8 specific complementary RNA (cRNA) probe (from bp 29 to 398; GeneBank AF125307) was used. For the generation of the antisense and control sense cRNAs, the vector (pBluescript vector; Stratagene, La Jolla, CA, USA) containing the PCR-amplified Cln8 fragment was linearized and used as a template for RNA polymerases T3 and T7, respectively, using α-³⁵S-labeled UTP (Amersham Biosciences, Uppsala, Sweden).

The paraffin sections were treated as cryosections, after first being deparaffinized in xylene and rehydrated in a descending ethanol series. All the slides were fixed in 4% paraformaldehyde, rinsed twice in phosphate-buffered saline and treated with proteinase K (20 μg/ml for paraffin sections, 1 μg/ml for cryosections). Next, the slides were postfixed in 4% paraformaldehyde, and the cryosections rinsed in 50% deionized formamide in 2 × standard sodium citrate (SSC). After being washed in distilled water, all slides were acetylated for 10 minutes with 0.25% acetic anhydride in 0.1 M triethanolamine. The paraffin sections were then dehydrated in an ascending ethanol series, while the cryosections were rinsed again with 50% formamide in 2 × SSC. Prehybridization for 1 h was subsequently performed at +52°C with hybridization buffer containing 60% formamide, 10% dextran sulphate, 1 × Denhardt's solution, 0.5 mg/ml yeast transfer RNA (tRNA), and 100 mM dithiothreitol. After prehybridization, cRNA probes were added to the hybridization buffer at a concentration of 35 000 cpm/μl, and hybridization was carried out overnight at +52°C in a chamber humidified with 60% formamide and 5 × SSC.

After hybridization the sections were first washed in 5 × SSC, 10 mM dithiothreitol at +50°C for 30 minutes, followed by 30 minutes in 50% formamide, 2 × SSC, 30 mM dithiothreitol at +65°C, and 3 × 10 minutes in NTE-buffer (0.5 M NaCl, 10 mM Tris-HCl, pH 8.0, 5 mM EDTA) at +37°C. Next, the slides were treated with RNase A (20 μg/ml) in NTE-buffer at +37°C for 30 min and washed for 15 min in NTE-buffer at +37°C. Washing with 50% formamide, 2 × SSC, 30 mM dithiothreitol was then repeated, after which the sections were washed in 2 × SSC and 0.1 × SSC, each for 15 min at +37°C. The slides were dehydrated at room temperature with ethanol containing 0.3 M ammonium acetate, air-dried, and exposed to BioMax MR Film (Kodak) together with ¹⁴C radioactive standards (Amersham Biosciences) for 5 days. Finally, the sections were dipped in NTB-2 emulsion (Kodak), developed after 3 to 4 weeks using D-19 developer (Kodak), fixed with Kodak Unifix, counterstained with hematoxylin (Shandon Inc./Thermoshandon, Pittsburgh, PA, USA), and mounted with Permount (Fisher Scientific International Inc., Hampton, NH, USA). Mice were bred in the Viikki Laboratory Animal Center, University of Helsinki. The experiments were conducted according to the "European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific purposes" with the approval of the institutional ethics committee.

Adult male C57/BL mice weighing 22–25 g (n = 16) were anaesthetized with sodium pentobarbital (60 mg/kg i.p.), and bipolar stainless steel electrodes (Plastics One) were implanted in the left ventral hippocampus using a Kopf stereotaxic frame as previously described 46. One week later, the mice were electrically stimulated with forty threshold stimulations with 5 min intervals (10 Hz frequency, 1 ms square-wave pulses for 10 s). The electroencephalogram was continuously recorded during the whole rapid kindling procedure using the MacLab/4e system (ADInstruments). To determine the threshold, current intensity was increased stepwise (10 μA increments, starting from 10 μA) every 5 min until focal epileptiform activity (afterdischarge) of at least 5 s duration was elicited. Rapid kindling stimulations were then given on the same day. Behavioral convulsions were scored according to a modification 25 of the scale of Racine 47: grade 0, arrest, normal behavior; grade 1, facial twitches; grade 2, chewing and nodding; grade 3, forelimb clonus; grade 4, rearing, body jerks, tail upholding; and grade 5, imbalance, hind-limb clonus and vocalization. At 2, 6 and 24 h after the last stimulus-evoked seizure, animals were decapitated (four animals in each group) and brains were processed for in situ hybridization. Four non-stimulated, electrode-implanted mice served as controls and were killed together with the experimental animals from different groups. The correct localization of the electrode was controlled for each animal in the coronal sections of the brain used for in situ hybridization. All experimental procedures were approved by the Research Ethics Committee at the Medical Faculty of the University of Lund.

Quantification of radioactive in situ hybridization films was done by digitizing the film images with a computer-based MCID image analysis system (Imaging Research). Gray levels from ¹⁴C radioactive standards were used in a third-degree polynomial calibration to obtain equivalent values of tissue radioactivity (nCi/g) for different brain regions. Measurements for each structure were carried out in several sections per animal. Since there were no significant differences between the two sides, values were pooled to obtain the mean value for each animal and brain region. Data from four control and four experimental animals for every probe and time point were used for analysis. Statistical analysis was performed using Student's t-tests. In the figures, the densities of hybridization signals of sense and antisense probes are shown separately. Sense values were not subtracted from antisense values because of different probes used. Error bars represent standard error of the mean; symbols *, ** and *** represent p < 0.05, 0.01 and 0.001, respectively.