Differences in immunoreactivities of Ki67 and doublecortin in the adult hippocampus in three strains of mice

ARTICLE IN PRESS

Acta histochemica 111 (2009) 150—156

0065-1281/$ - sdoi:10.1016/j.

�CorrespondE-mail addr

1Also for corE-mail addr

www.elsevier.de/acthis

Differences in immunoreactivities of Ki-67 anddoublecortin in the adult hippocampus in threestrains of mice

Joong-Sun Kima, Jisun Junga, Hae-June Leea, Jong Choon Kimb,Hongbing Wangc, Sung-Ho Kima, Taekyun Shind,1, Changjong Moona,�

aDepartment of Veterinary Anatomy, College of Veterinary Medicine and Veterinary Medical Research Center, ChonnamNational University, Gwangju 500-757, South KoreabDepartment of Veterinary Toxicology, College of Veterinary Medicine and Veterinary Medical Research Center,Chonnam National University, Gwangju 500-757, South KoreacDepartment of Physiology and Neuroscience Program, Michigan State University, MI 48824, USAdDepartment of Veterinary Anatomy, College of Veterinary Medicine and Applied Radiological Science ResearchInstitute, Cheju National University, Jeju 690-756, South Korea

Received 25 March 2008; received in revised form 18 April 2008; accepted 7 May 2008

KEYWORDSNeurogenesis;Hippocampus;Ki-67;Doublecortin;Mouse

ee front matter & 2008acthis.2008.05.002

ing author. Tel.: +82 62ess: [email protected]: Tel.: +82ess: [email protected]

SummaryNeurogenesis in the adult hippocampus is differentially influenced by the geneticbackground. We examined the differences in Ki-67 (a proliferating cell marker) anddoublecortin (DCX; an immature progenitor cell marker) immunolabelling in thedentate gyrus (DG) of the adult hippocampus in three strains of mice (ICR, C57BL/6,and BALB/c) to evaluate the effect of genetic background on adult hippocampalneurogenesis. All strains showed constitutive immunoreactivity of either Ki-67 or DCXin the DG of the adult hippocampus. C57BL/6 mice showed significantly higher levelsof Ki-67-immunopositive cells in the subgranular zone (SGZ) of the DG (approximately2.2-fold) compared to ICR and BALB/c mice. The greatest number of DCX-immunopositive cells was found in C57BL/6 (approximately 1.6-fold), which differedsignificantly from ICR and BALB/c mice. However, there was no significant differencein the number of Ki-67- and DCX-immunopositive cells between BALB/c and ICR mice.Genetic differences with respect to certain aspects of hippocampal neurogenesis inadult mice may influence hippocampal functions, including learning and memory.& 2008 Elsevier GmbH. All rights reserved.

Elsevier GmbH. All rights reserved.

530 2838; fax: +82 62 530 2841.c.kr (C. Moon).64 754 3363; fax: +82 64 756 3354.(T. Shin).

ARTICLE IN PRESS

Difference of Ki-67 and DCX in mouse hippocampi 151

Introduction certain hippocampal functions. However, the use of

In the past, 50-Bromo-20-deoxyuridine (BrdU) hasbeen used as a principal mitotic marker todetermine the rate of cell proliferation in labora-tory mice (Kempermann et al., 1997; Kempermannand Gage, 2002; Schauwecker, 2006). However,BrdU can be incorporated into dividing neurons(Gould et al., 1997, 1998) and numerous other sideeffects have been reported, such as stress duringinjection and mutagenesis following incorporation(Kee et al., 2002). Ki-67, an endogenous marker forproliferating cells, is a very reliable markerbecause it is expressed during mitosis and has ashort half-life (Scholzen and Gerdes, 2000). Inaddition, Ki-67 can be detected easily by immuno-histochemistry. Thus, although the role of Ki-67remains unknown, it appears to be a good markerfor detecting proliferating cells during neurogen-esis of the adult brain.

Doublecortin (DCX) is a microtubule-associatedphospho-protein required for neuronal migrationand differentiation that is expressed in migratingneurons during the development of the centralnervous system (Gleeson et al., 1999; Francis et al.,1999). DCX expression is believed to be specific tonewly generated neurons because virtually all DCX-positive cells express early neuronal antigens (Raoand Shetty, 2004). Therefore, DCX may be a goodmarker for immature progenitor cells in neurogen-esis of the adult brain.

Hippocampal neurogenesis has been observedthroughout the adult lives of mammals (Kemper-mann et al., 1997). New neurons in the adulthippocampus originate in progenitor cells locatedat the border between the hilus and granular celllayer, which is a region called the subgranular zone(SGZ) of the dentate gyrus (DG). Cell proliferationand survival are influenced differentially by inheri-table traits, and the genetic background deter-mines which regulatory levels of adult hippocampalneurogenesis are preferentially involved in aneurogenic response to environmental stimuli(Kempermann and Gage, 2002).

Differences in hippocampal neurogenesis be-tween laboratory mouse strains can affect thefunctions of the adult hippocampus, includinghippocampus-dependent learning and memory.Hippocampal neurogenesis of in-bred mouse strainssuch as C57BL/6 and BALB/c continues to be ofinterest because these background strains arecommonly used to generate knockout and mutantmouse strains for scientific research. The ICR out-bred strain has attracted less attention becauseit is generally accepted that a variable geneticbackground can complicate the measurements of

out-bred ICR mice has gained popularity in drug andpre-clinical research (Wozniak et al., 1996; Zhaoet al., 2007). Furthermore, it may be advantageousto use out-bred strains because they are morerobust, easy to breed and relatively resistant todiseases. Therefore, a comparison between ICRmice and two in-bred strains, C57BL/6 and BALB/c,will help clarify the effects of the genetic back-ground on hippocampal neurogenesis.

This study examined the differences in Ki-67(a proliferating cell marker) and DCX (an immatureprogenitor cell marker) immunolocalization in theDG of the adult hippocampus in three laboratorymouse strains, ICR, C57BL/6 and BALB/c, toevaluate the effect of the difference in strain onadult hippocampal neurogenesis in mice.

Materials and methods

Animals and tissue preparation

Eight-week-old out-bred ICR as well as in-bredC57BL/6 and BALB/c male mice were obtained fromCharles River (Wilmington, MA, USA). All animalexperiments were carried out according to aprotocol approved by the Committee for AnimalExperimentation at the Chonnam National Univer-sity. The mice were sacrificed and the brain wasdissected from each mouse (n ¼ 3 mice/strain).The brains were processed for embedding inparaffin wax after fixation in 10% neutral bufferedformalin using routine protocols.

Immunohistochemistry

Five-micrometer-thick coronal sections were cutby microtome and deparaffinized by routine proto-cols before being exposed to citrate buffer (0.01M,pH 6.0) and heated in an autoclave for 10min. Allsubsequent steps were performed at room tempera-ture. The sections were treated with 0.3% hydrogenperoxide in methyl alcohol for 20min to blockendogenous peroxidase activity. After three washesin phosphate-buffered saline (PBS), the sectionswere blocked with 10% normal goat serum (VectorABC Elite kit, Burlingame, CA, USA), for 1 h, andthen allowed to react for 2 h with the antibodiesdirected against markers for proliferating cells andimmature progenitor cells: monoclonal rabbit anti-Ki-67 (DRM004; diluted 1:400, Acris AntibodiesGmbH, Hiddenhausen, Germany) and polyclonalrabbit anti-DCX (diluted 1:400, Cell Signaling Tech-nology, Beverly, MA, USA). The sections then were

ARTICLE IN PRESS

J.-S. Kim et al.152

then reacted with biotinylated goat anti-rabbit IgG(Vector), diluted 1:100, for 45min. They were thenincubated with the avidin–biotin peroxidase com-plex (Vector ABC Elite kit), prepared according tothe manufacturer’s instructions, for 45min. Theperoxidase reaction was developed using a diamino-benzidine substrate (DAB kit, SK-4100; Vector),prepared according to the manufacturer’s instruc-tions, for 3min. All dilutions and thorough washesbetween stages were performed using PBS unlessotherwise stated. As a control, the primary anti-bodies were omitted for a few test sections in eachexperiment. After completion of color development,the sections were counterstained with Harris’shematoxylin for 5 s, washed in running tap waterfor 20min, dehydrated through a graded ethanolseries, cleared with xylene and mounted withCanada balsam (Sigma, St. Louis, MO, USA).

Determination of cell number

The number of cells showing the specific char-acteristics of proliferating cells (immunopositivefor Ki-67) and immature progenitor cells (immuno-positive for DCX) in the hippocampus was scoredby an observer blinded to the identity of thesample using a histomorphometric approach (Mizu-matsu et al., 2003; Raber et al., 2004). The brainfrom each mouse was sampled at level �2.12mmbehind the bregma. A standardized countingarea, which contained 5-mm-thick coronal sectionsin a one-in-ten series of sections representingthe rostral/mid-hippocampus, was used. For eachmouse, three non-overlapping sections were ana-lyzed, one each from the three regions of thehippocampus (�50 mm apart). All positively immu-nolabeled cells within the SGZ of the supra- andinfra-pyrimidal blades of the DG were quantified.The number of immunopositive cells was deter-mined from the values obtained from each DGin the three brain sections. The mean numberof immunopositive cells in the three sections ofeach mouse was taken as n ¼ 1. The number ofimmunopositive cells is expressed as the mean7SEM for each strain (n ¼ 3).

Statistical analysis

The data are reported as the mean7SEM. Thedata were analyzed using a one-way analysis ofvariance (ANOVA) followed by a Student–Newman–

Keuls post hoc test for multiple comparisons. In allcases, Po0.05 was considered significant.

Results

In the adult hippocampus of all three mousestrains, both Ki-67 and DCX immunoreactivitieswere seen only in the DG.

Immunohistochemical analysis of Ki-67 in theDG of adult hippocampus

Ki-67 immunoreactivity was apparent in thenuclei of cells located at the border between thegranular cell layer (GCL) and hilus, which is a regioncalled the SGZ, of the DG of the adult hippocampusin all three mouse strains (Figure 1).

The number of Ki-67-immunopositive cells wassimilar (P ¼ 0.842) in ICR (770.6 cells/DG, n ¼ 3;Figure 1A, B, and G) and BALB/c (7.371.5 cells/DG, n ¼ 3; Figure 1E, F, and G). Only C57BL/6showed a significantly higher number of Ki-67-immunopositive cells (1671.7 cells/DG, n ¼ 3;Po0.01 vs. ICR; Po0.05 vs. BALB/c; Figure 1C, D,and G; F (2, 6) ¼ 14.347) than the other twostrains. No nuclear labeling was observed in anytest section in which the primary antibodies hadbeen omitted (Figure 1H).

Immunohistochemical analysis of DCX in theDG of the adult hippocampus

DCX immunoreactivity was observed in thecytoplasm of cells located in the GCL adjacent tothe hilus of the DG from the adult hippocampus ofthe three mouse strains (Figure 2). The immunor-eaction revealed the ramified body shape of thecells in the GCL (Figure 2). In order to count theDCX-immunopositive cells, the number of cells withimmunopositively labeled cytoplasm and nucleicounterstained with hematoxylin were determined.

Compared with the frequency of Ki-67-immuno-positive cells, all strains showed a very highfrequency of DCX-immunopositive cells in theSGZ. The largest number of DCX-immunopositivecells was found in C57BL/6 mice (79.773.8 cells/DG, n ¼ 3; Figure 2C, D, and G; F (2, 6) ¼ 34.564),which was significantly different from thoseof both ICR (48.372.2 cells/DG, n ¼ 3; Po0.01vs. C57BL/6; Figure 2A, B, and G) and BALB/c(46.373.3 cells/DG, n ¼ 3; Po0.01 vs. C57BL/6;Figure 2E, F, and G) strains. However, there was nosignificant difference in the number of DCX-immunopositive cells between ICR and BALB/c mice(Figure 2G). No cytoplasmic labeling (ramified cellbody) was found in any test section in which theprimary antibodies had been omitted (Figure 2H).

ARTICLE IN PRESS

Figure 1. Immunoreactivity of Ki-67 in the DG of adult hippocampus in three strains of mice: ICR (A, B), C57BL/6 (C, D),and BALB/c (E, F). Most Ki-67-immunopositive cells were found in the SGZ of the DG. The cells were frequentlyclustered and their nuclei were dark and had an irregular shape. A, C, and E show low-magnification images of Ki-67immunoreactivity in the DG of ICR, C57BL/6, and BALB/c mice, respectively. B, D, and F show a high-magnificationimage of Ki-67 immunopositive cells in the SGZ of ICR, C57BL/6, and BALB/c mice, respectively. The bar graphs in Gshow the number of Ki-67-immunopositive cells/DG (mean7SEM). The arrows indicate Ki-67-immunopositive nuclei. Noreaction product was observed in the sections incubated with non-immune sera (H). GCL, granular cell layer; SGZ,subgranular zone. Scale bars ¼ 200 mm (A, C, and E), 50 mm (B, D, and F).

Difference of Ki-67 and DCX in mouse hippocampi 153

Discussion

This is the first study to demonstrate thedifferences in the number of Ki-67-immunopositiveproliferating new neural cells and DCX-immunopo-sitive immature progenitor neurons in the DG ofadult hippocampus in the three mouse strains, ICR,C57BL/6, and BALB/c. These results suggest thatthe proliferating and migrating ability of newneural cells in the DG was significantly higher inthe C57BL/6 strain than in the other two strains.

Genetic factors can influence adult hippocampalneurogenesis. Previous studies have suggested that

different strains have different rates of neuralcell generation in the DG of adult hippocampus(Kempermann et al., 1997; Perfilieva et al., 2001;Hayes and Nowakowski, 2002; Kempermann andGage, 2002). Kempermann et al. (1997) reportedthat among four mouse strains, C57BL/6, BALB/c,CD1/J and 129/Sv, the C57BL/6 strain showed thehighest number of newly proliferating cells, stainedwith the thymidine analog BrdU, in the SGZ of theDG. On the other hand, the survival of BrdU-labeledprogenitor cells was greatest in the CD1/J strain,but the differentiation of surviving cells observedwas similar in these four strains. In this study, the

ARTICLE IN PRESS

Figure 2. Immunoreactivity of DCX in the DG of the adult hippocampus in three strains of mice: ICR (A, B), C57BL/6 (C,D), and BALB/c (E, F). The DCX-immunopositive cells are darkly labeled in ramified granular cells in the DG. Most cellbodies immunopositive for DCX are located at the SGZ of the DG. A, C, and E show low-magnification images of DCXimmunoreactivity in the DG of ICR, C57BL/6, and BALB/c mice, respectively. B, D, and F show a high-magnificationimage of DCX-immunopositive cells in the GCL of ICR, C57BL/6, and BALB/c mice, respectively. The bar graphs in Gindicate the number of DCX-immunopositive cells/DG (mean7SEM). The arrows indicate DCX-immunopositive cells. Noreaction product was observed in the sections incubated with non-immune sera (H). GCL, granular cell layer; SGZ,subgranular zone. Scale bars ¼ 200 mm (A, C, and E), 50 mm (B, D, and F).

J.-S. Kim et al.154

number of Ki-67-immunopositive cells in the SGZ ofthe DG in C57BL/6 was significantly higher thanthat in the BALB/C and ICR mice. The highernumber of Ki-67-immunopositive cells suggests alarger population of dividing progenitor cells, aslower cell cycle, or a combination of both(Kempermann and Gage, 2002). Therefore, theproliferating ability of new neural cells in adulthippocampus of the C57BL/6 strain might besignificantly higher than in ICR and BALB/c mice.

Whether the frequency of DCX-immunopositivecells in the adult DG differs between mousestrains was also examined in this study. DCX is an

immunohistochemical marker for the migration anddifferentiation of newly generated neurons in theDG and has been used to evaluate adult hippocam-pal neurogenesis (Gleeson et al., 1999; Franciset al., 1999). It has been suggested that DCXexpression in post-mitotic neurons is quite robustduring migration, differentiation and the earlygrowth period (Rao and Shetty, 2004). In this study,the C57BL/6 mice showed significantly higherfrequencies of DCX-immunopositive cells in theDG than ICR and BALB/c mice. This suggests thatthe migrating and differentiating ability of newlygenerated neurons in adult DG of the C57BL/6

ARTICLE IN PRESS

Difference of Ki-67 and DCX in mouse hippocampi 155

strain are significantly higher than those of theother two strains.

The function of the hippocampus (i.e. learningand memory ability) may be related to hippocampalneurogenesis in adult mice. Previous studiesshowed that C57BL/6 mice, which are goodlearners (Montkowski et al., 1997; Owen et al.,1997), also have high rates of neurogenesis in theadult DG (Kempermann et al., 1997) and this isconsistent with the results of the current study. Onthe other hand, 129/svj mice, which are consideredpoor learners (Montkowski et al., 1997; Owenet al., 1997), have a low rate of adult hippocampalneurogenesis (Kempermann et al., 1997) and showsignificant defects in the maintenance of long-termpotentiation (LTP; Nguyen et al., 2000). Differ-ences in hippocampal neurogenesis may influencehippocampal neuronal function and memory. Inaddition, there is stronger evidence for a funda-mental connection from the finding that activity-induced increases in adult hippocampal neurogen-esis are paralleled by the increased LTP in the GCL(van Praag et al., 1999). Significant differences inadult hippocampal neurogenesis were found in thedifferent strains of mice. However, this does notimply that the genes in question are limited tothose that directly affect the regulation of neuro-genesis. Further studies on the relationship be-tween adult hippocampal neurogenesis, geneticstrain and hippocampal function including LTP,learning, and memory are needed.

In summary, the number of proliferating andmigrating cells in the adult hippocampus was higherin the C57BL/6 mice than in ICR and BALB/c mice.Such genetic differences in hippocampal neurogen-esis in adult mice may influence the hippocampalfunctions including learning and memory.

Acknowledgments

This study was supported by the Ministry ofEducation, Science and Technology of Koreathrough the Nuclear R&D Program and the RegionalResearch Centers Program (Bio-housing ResearchInstitute).

References

Francis F, Koulakoff A, Boucher D, Chafey P, Schaar B,Vinet MC, et al. Doublecortin is a developmentallyregulated, microtubule-associated protein expressedin migrating and differentiating neurons. Neuron1999;23:247–56.

Gleeson JG, Lin PT, Flanagan LA, Walsh CA. Doublecortinis a microtubule-associated protein and is expressedwidely by migrating neurons. Neuron 1999;23:257–71.

Gould E, McEwen BS, Tanapat P, Galea LA, FuchsE. Neurogenesis in the dentate gyrus of the adult treeshrew is regulated by psychosocial stress and NMDAreceptor activation. J Neurosci 1997;17:2492–8.

Gould E, Tanapat P, McEwen BS, Flugge G, FuchsE. Proliferation of granule cell precursors in thedentate gyrus of adult monkeys is diminished bystress. Proc Natl Acad Sci USA 1998;17:95:3168–71.

Hayes NL, Nowakowski RS. Dynamics of cell proliferationin the adult dentate gyrus of two inbred strains ofmice. Brain Res Dev Brain Res 2002;134:77–85.

Kee N, Sivalingam S, Boonstra R, Wojtowicz JM. Theutility of Ki-67 and BrdU as proliferative markersof adult neurogenesis. J Neurosci Methods 2002;115:97–105.

Kempermann G, Gage FH. Genetic influence on pheno-typic differentiation in adult hippocampal neurogen-esis. Brain Res Dev Brain Res 2002;134:1–12.

Kempermann G, Kuhn HG, Gage FH. Genetic influence onneurogenesis in the dentate gyrus of adult mice. ProcNatl Acad Sci USA 1997;94:10409–14.

Mizumatsu S, Monje ML, Morhardt DR, Rola R, Palmer TD,Fike JR. Extreme sensitivity of adult neurogenesisto low doses of X-irradiation. Cancer Res 2003;63:4021–7.

Montkowski A, Poettig M, Mederer A, Holsboer F.Behavioural performance in three substrains of mousestrain 129. Brain Res 1997;762:12–8.

Nguyen PV, Abel T, Kandel ER, Bourtchouladze R. Strain-dependent differences in LTP and hippocampus-de-pendent memory in inbred mice. Learn Mem 2000;7:170–9.

Owen EH, Logue SF, Rasmussen DL, Wehner JM. Assess-ment of learning by the Morris water task and fearconditioning in inbred mouse strains and F1 hybrids:implications of genetic background for single genemutations and quantitative trait loci analyses. Neu-roscience 1997;80:1087–99.

Perfilieva E, Risedal A, Nyberg J, Johansson BB, ErikssonPS. Gender and strain influence on neurogenesis indentate gyrus of young rats. J Cereb Blood Flow Metab2001;21:211–7.

Raber J, Rola R, LeFevour A, Morhardt D, Curley J,Mizumatsu S, et al. Radiation-induced cognitive im-pairments are associated with changes in indicators ofhippocampal neurogenesis. Radiat Res 2004;162:39–47.

Rao MS, Shetty AK. Efficacy of doublecortin as a markerto analyse the absolute number and dendritic growthof newly generated neurons in the adult dentategyrus. Eur J Neurosci 2004;19:234–46.

Schauwecker PE. Genetic influence on neurogenesis inthe dentate gyrus of two strains of adult mice. BrainRes 2006;1120:83–92.

Scholzen T, Gerdes J. The Ki-67 protein: from the knownand the unknown. J Cell Physiol 2000;182:311–22.

van Praag H, Christie BR, Sejnowski TJ, Gage FH. Runningenhances neurogenesis, learning, and long-term

ARTICLE IN PRESS

J.-S. Kim et al.156

potentiation in mice. Proc Natl Acad Sci USA 1999;96:13427–31.

Wozniak DF, Brosnan-Watters G, Nardi A, McEwen M,Corso TD, Olney JW, et al. MK-801 neurotoxicity inmale mice: histologic effects and chronic impairmentin spatial learning. Brain Res 1996;707:165–79.

Zhao Q, Murakami Y, Tohda M, Obi R, Shimada Y,Matsumoto K. Chotosan, a kampo formula, amelio-rates chronic cerebral hypoperfusion-induced deficitsin object recognition behaviors and central choliner-gic systems in mice. J Pharmacol Sci 2007;103:360–73.