Design
A randomized, controlled animal study.

Time and setting
The experiments were performed at the Institute of Physiology at Shandong University School of Medicine, China from July 2010 to March 2012.

Materials
A total of 37 adult male Sprague-Dawley rats (140–170 g, 6–7 weeks old) were purchased from the Animal Center of Shandong University of Traditional Chinese Medicine in China (license No. SCXK (Lu) 20110003). Animals were allowed to habituate to the university facility for at least 1 week before experiments began. The experiments were performed according to the Guidance Instructions for the Care and Use of Laboratory Animals, issued by the Ministry of Science and Technology of China^[26].

Methods
Status epilepticus induction

Status epilepticus was induced by intraperitoneal injection of pilocarpine as described previously^[9] with modifications. Rats were injected intraperitoneally with the first dose of pilocarpine hydrochloride (360 mg/kg) 30 minutes after pretreatments with an intraperitoneal injection of scopolamine methyl bromide and terbutaline hemisulfate (both 2 mg/kg). If the first dose of pilocarpine did not induce continuous limbic motor seizures, a second dose of 100 mg/kg was injected 45 minutes later. Animals were accepted for further studies if the continuous limbic motor seizure corresponded to stage 2 or higher according to Racine’s scale and lasted for more than 2 hours (n = 15)^[27]. Status epilepticus was allowed to self-terminate after 6 hours. Control rats (n = 11) were treated with intraperitoneal injection of normal saline following pretreatment with scopolamine methyl bromide and terbutaline hemisulfate (both 2 mg/kg). Pilocarpine hydrochloride, scopolamine methyl bromide, and terbutaline hemisulfate were purchased from Sigma (St. Louis, MO, USA).

Transcardial perfusion and brain tissue preparation
Seven days or at least 10 weeks after the induction of status epilepticus, individual animals were deeply anesthetized with an intraperitoneal injection of sodium pentobarbital (60 mg/kg). After complete paralysis, the animals were perfused through the left ventricle with heparinized saline (20–30 mL), followed by 200–300 mL of 4% paraformaldehyde in 0.1 M PBS (pH 6.8) over a period of 45 minutes. The fixed brain was extracted, the cerebellum and brainstem were trimmed off, and the tissue was post-fixed in the same fixative at 4°C overnight. Thereafter, the tissue block was sequentially transferred into 10% sucrose in 0.1 M PBS for 4 hours, 15% sucrose in 0.1 M PBS for 8 hours, and finally 20% sucrose in 0.1 M PBS at 4 °C overnight. The tissue block was embedded in a medium composed of 30% (w/v) chicken egg albumin (Sigma), 0.5% (w/v) gelatin and 0.9% (v/v) glutaraldehyde in 0.1 M PBS, and serial horizontal slices (60 μm thick) were cut through the hippocampi using a Vibratome 1000 (Vibratome Co., St. Louis, MO, USA). A previous study applying stereological counting methods has shown that the number of granule cells in the hilus is the same in the two hemispheres[14], so hippocampal sections from the left hemisphere were collected into 0.1 M PBS. Horizontal sections containing the ventral hippocampus were used for quantitative comparison of ectopic granule cells, because the ventral hippocampus is more seizure-prone than the dorsal hippocampus^[28-29].

Immunofluorescence and image acquisition
To count ectopic granule cells, the 4^th, 12^th, 20^th, 28^th, 36^th, 44^th, 52^nd and 60^th (if it existed) hippocampal horizontal slices were chosen for prospero homeobox protein 1 and neuron-specific nuclear protein double immunofluorescent staining of free-floating sections. After a thorough wash in 0.1 M PBS, the 7–8 sections were incubated in a blocking buffer consisting of 5% normal goat serum, 2.5% bovine serum albumin and 0.2% Triton X-100 in 0.1 M PBS for 2.5 hours at 4°C to minimize non-specific reactions. Thereafter, the sections were incubated with a cocktail of mouse anti-neuron-specific nuclear protein (1:200) and rabbit anti-prospero homeobox protein 1 (1:6 000) at 4°C overnight. The primary antibodies, both purchased from Millipore (Temecula, CA, USA), were diluted in the blocking buffer. After three 15-minute washes in 0.1 M PBS, the sections were incubated for 2.5 hours at 4°C with a mixture of Alexa Fluor 488-conjugated goat anti-mouse IgG and Alexa Flour 568-conjugated goat anti-rabbit IgG (Invitrogen, San Diego, CA, USA) diluted in the blocking buffer. The final concentration of both secondary antibodies was 1.6 μg/mL. After washing (three times, 15 minutes each), the sections were mounted on standard glass slides and cover-slipped with 75% glycerol in 0.1 M PBS. Sections were visualized with an Olympus fluorescence microscope. Images were captured with a digital camera using 10 × and 20 × objectives.

Doublecortin immunostaining performed 7 days after pilocarpine or normal saline injection was used to determine the ectopic distribution of newborn granule cells. The procedures for transcardial perfusion, tissue preparation and immunofluorescence were identical to those described above. Goat anti-doublecortin (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was used as the primary antibody (1:600), whereas Alexa Fluor 488-conjugated donkey anti-goat IgG (Invitrogen) was used as a secondary antibody (1:600). Donkey serum was used to minimize non-specific reactions. Stained sections were mounted on glass slides and scanned using a Carl Zeiss confocal microscope (LSM 780, Jena, Thuringia, Germany).

Cell counting
Paired fluorescent images for prospero homeobox protein 1 and neuron-specific nuclear protein were acquired using the 10 × objective and merged in Adobe Photoshop 7 (Adobe Systems Inc., San Jose, CA, USA). Cells positive for both neuron-specific nuclear protein and prospero homeobox protein 1 were counted by a blinded examiner in the hilus and molecular layer, as defined by the contours illustrated in Figure 4. Granule cells in the hilus were considered to be ectopic when they were located at least two granule cell body widths away from the inner border of the granule cell layer^{[6, 10]}. Ectopic granule cells in the molecular layer were considered ectopic when they were at least three granule cell body widths away from the outer border of the granule cell layer. The number of cells qualified as ectopic granule cells in the hilus or the molecular layer of each section was summed. This number was multiplied by the number of sections (7–8) to approximate the total number of ectopically distributed mature granule cells in the hilus and the molecular layer.

Statistical analysis
Quantitative data are presented as mean ± SEM and statistically compared by Student-Newman-Keuls tests following one-way analysis of variance GraphPad Prism (version 4.0, San Diego, CA, USA). A P-value less than or equal to 0.05 was considered statistically significant.

Acknowledgments: The authors would like to thank Prof. Aijun Hao, Department of Histology and Embryology of Shandong University School of Medicine, for help with epifluorescence microscopy.

Funding: This study was supported by grants from the Self-innovation Programs of Shandong University, No. 1000069961016; and the National Natural Science Foundation of China, No. 81171231.
Author contributions: Zhaoliang Liang, Fei Gao and Fajun Wang conducted the experiments. Xiaochen Wang and Xinyu Song counted cells and did the statistical analysis. Kejing Liu did the confocal imaging. Ren-Zhi Zhan designed the experiments and wrote the manuscript. All authors read and approved the final manuscript submitted.
Conflicts of interest: None declared.
Ethical approval: Animal experiments were approved by the Animal Ethics Committee of Shandong University School of Medicine in China.
Author statements: The manuscript is original, has not been submitted to or is under consideration by another journal, has not been previously published in any language or any form (including electronic), and contains no disclosure of confidential information or authorship/patent application/funding source disputations.

文章设计构思：

癫痫所致的脑损伤发生后，大量颞叶癫痫患者病灶标本和癫痫动物模型海马区均可观察到齿状回颗粒细胞轴突(苔藓纤维)出芽，门区异位颗粒细胞形成等现象。海马CA1、CA3区锥体细胞和门区神经元致痫损伤后，内分子层失神经传入同时苔藓纤维也与靶细胞断离，因而触发苔藓纤维出芽进入内分子层并与颗粒细胞建立突触联系。文章假设癫痫持续状态也增加新生颗粒细胞的分子层异位，并通过实验予以证实，将开拓研究分子层异位颗粒细胞在癫痫发生过程中的作用。