Biomedical Research 33 (5) 273-279, 2012

Collagen peptides enhance hippocampal neurogenesis and reduce anxiety- related behavior in mice

Chikako Kakoi1, Hiroshi Udo

2, Taiji Matsukawa3, and Koichiro Ohnuki

4

1 College of Physical Education, National Institute of Fitness and Sports in Kanoya, Kagoshima, 891-2393, Japan; 2 Department of Biology, Faculty of Science, Kyushu University, Fukuoka, 812-8581, Japan; 3 Department of Applied Bioscience, Mikakuto Co., Ltd. Osaka, 540-0016, Japan; and 4 Faculty of Food and Nutrition, Kyushu Nutrition Welfare University, Fukuoka, 803-8511, Japan

(Received 31 May 2012; and accepted 28 July 2012)

ABSTRACTThe present study examined the effects of enzymatically hydrolyzed collagen peptides on the level of hippocampal neurogenesis and emotional behavior in adult mice. For this purpose, two kinds of enzymatically hydrolyzed collagen peptides, the lower or higher molecular weight peptides (LP: below 2,000, HP: about 30,000) were administered orally to C57BL/6 mice for 4 weeks. As a re-sult, the density of proliferating cells in subgranular zone of hippocampus showed a 1.2-fold in-crease in LP mice as compared with HP mice. Additionally, LP mice spent less time in closed arms than HP mice in elevated plus maze test to examine anxiety-related behavior. These results suggest that oral administration of the lower molecular weight peptides derived from collagen en-hanced the hippocampal neurogenesis and exerted emotional behavior in adult mice.

A variety of health issues such as neurodegenerative diseases and mood disorders face modern society (16, 22, 26, 27). These neurological disorders can be treated by an enhancement of adult neurogenesis, which helps to improve spatial memory and synap-tic plasticity and moderate depression (3, 5). Inter-estingly, the level of adult neurogenesis depends on environmental conditions such as exercise, learning, enriched environment and chronic stress (6, 12, 13, 17, 20). Therefore, manipulating environmental fac-tors could regulate the level of neurogenesis for maintenance or improvement of human health.　In this study, we set out to examine whether in-gestion of nutrients, considered as one of the major environmental variables, might affect the level of adult neurogenesis and its associated behavior. It has been reported that the dietary restriction enhanc-es hippocampal neurogenesis in adult mice (14), and

we attempted to investigate the effects of food ingestion with no influence of dietary restriction. Several studies have shown that some nutrients in-cluding arachidonic acid and docosahexaenoic acid affect brain functions, such as neurogenesis in the hippocampus or neuronal differentiation of neural stem cells (10, 11, 15). Moreover, it have been re-ported that protein-derived peptides (e.g. soybean peptide) have the beneficial effects on brain func-tions including not only an increase in the number of neural stem cells but also an improvement of memory and emotional state (4, 7, 8).　Among different nutrients available in nature, we focused on peptides from collagen. As well as being the most abundant protein in animal tissues, colla-gen is an extracellular matrix protein and often served as a protein source of human diet. Low mo-lecular weight collagen peptides degraded by en-zymes are more efficiently absorbed through intestinal epithelium than collagen which is a very stable and high molecular weight protein. Although the prima-ry structure of collagen is rather simple and general-ly expressed as tripeptide repeats of Gly-X-Y (where X and Y are any amino acid residues), some colla-gen-derived peptides are known to exert physiologi-

Address correspondence to: Chikako Kakoi, M.P.Ed., College of Physical Education, National Institute of Fit-ness and Sports in Kanoya, Shiromizuchou 1, Kanoya, Kagoshima 891-2393, JapanTel: +81-994-46-1111, Fax: +81-994-46-2831E-mail: chichicaco@ninus.ocn.ne.jp

C. Kakoi et al.274

and doublecortin (1/400, rabbit polyclonal IgG; Ab-cam) and then with Cy3- or Cy5-conjugated second-ary antibodies against rat or rabbit IgG (1/200, donkey polyclonal IgG; Jackson ImmunoResearch, Tokyo, Japan). Cell nuclei were stained with Syto13 green (1/2,000; Molecular Probes, CA, USA). Dou-blecortin was used to stain premature neurons. Im-ages were acquired by a LSM510 microscope (Carl Zeiss Microscope, Tokyo, Japan) with ×10 or ×20 objective lens at 40-μm optical thickness under the same settings. The average linear density of BrdU-positive cells (the number of cells per mm of sub-granular zone, averaging of 8 independent slices) was used as an indicator for hippocampal cell gene-sis. For the BrdU-positive cells, we used ImageJ (National Institute of Health, USA) to automatically count the number of particles along subgranular zone (over 2 micrometer in diameter) based on bi-nary images of BrdU-staining (with a threshold set at the half-maximal intensity). The length of subgran-ular zones was measured by LSM Image Browser (Carl Zeiss Microscope) using Syto13 green nuclear staining.

Behavioral analysis. We then performed behavioral studies on 8-week-old male mice after intake of LP or HP for 4 weeks (12 mice per group). All the be-havioral apparatus and data acquisition/analysis soft-ware were designed and manufactured by O’Hara & Co., Tokyo, Japan. The software used for the behav-ioral analysis (Image TS, EP and OF) was based on Image J.

Tail suspension test. The level of depression-like be-havior was examined by tail suspension test (23). Mice were suspended 30 cm above the floor with an adhesive tape placed ~1 cm from the tip of the tail, and their behavior was recorded for 5 min. The pe-riods of immobile status was automatically analyzed by Image TS software.

Elevated plus maze test. In clinical cases of mood disorders, depression often accompanies anxiety or vice versa (19). Anxiety-related behavior was exam-ined by the elevated plus maze test. When mice are placed onto a cross-shaped elevated platform (55 cm above the floor level) under ambient illumination at 300 lx to induce natural aversion to height and open space, mice tend to travel within the safer closed arms (a pair of 5 × 45 cm white plexiglass surround-ed by clear 15-cm high side walls) rather than the riskier open arms (a pair of 5 × 45 cm white plexi-glass, no side wall). Each mouse was placed on the

cal effects such as neural protective effects by Gly-Pro-Glu and anti-inflammatory effect by Pro-Gly-Pro (1). However, the effects of collagen peptides on adult neurogenesis and behavior have been un-known.　In the present study, we examine the effect of the peptide on the hippocampal neurogenesis facilita-tion, antianxiety action and antidepressant activity.

MATERIALS AND METHODS

Peptide. We have prepared two kinds of peptide samples having different molecular weights by the proteolytic digestion of pig skin gelatin with a mix-ture of collagenase I (Washington Biochemical Cor-poration, NJ, USA) and protease N (Amano Enzyme Inc, Aichi, Japan). The mean molecular weights as determined by gel filtration chromatography using AKTA purifier and Superdex Peptide 10/300 GL (Amersham Biosciences, Buckinghamshire, U.K.) were 367.5 (ranging from 300 to 2,000, mostly con-sisted of tripeptides) for the lower molecular weight peptides (LP) and about 30,000 for the higher mo-lecular weight peptides (HP). HP was a less absorb-able peptide sample and used as a control. LP and HP were the same in amino acid composition. We have also analyzed some of the major components of LP by amino acid sequencing, and showed that the highest constituent is Gly-Pro-Hyp which ac-counted for 8 percent in mass.

Mice and feeding procedure. Five-week-old male C57BL/6J mice were divided into two groups, and each received LP or HP-containing water ad libitum for 4 weeks. After performing behavioral studies in 8-week-old mice, we evaluated the level of adult neurogenesis in hippocampus at 9-week old. Mice were housed at 23°C under 12-h light/12-h dark cy-cles, and handled in accordance with the institution-al animal care guidelines.

Immunocytochemistry and measurement of BrdU-positive cells in dentate gyrus. To evaluate the level of adult neurogenesis in hippocampus, the prolifer-ating cells were labeled by the intraperitoneal injec-tion of 5-bromo-2-deoxyuridine (BrdU, 100 mg/kg) for 2 h, and BrdU-positive cells were detected by the immunofluorescence staining, as described pre-viously (25). Briefly, coronal brain sections of 40 μm in thickness were collected every 200 μm in the region of hippocampus, and free-floating sec-tions were reacted with primary antibodies for BrdU (1/400, rat monoclonal IgG; Abcam, Tokyo, Japan)

Effect of collagen peptides in mice 275

(Fig. 2A and 2B). These results suggest that inges-tion of the lower molecular weight collagen peptides for 4 weeks efficiently enhances hippocampal neu-rogenesis in adult mice.

Behavioral analysisTail suspension test. The total moved distance had no significant difference between the two groups (LP mice: 799.4 ± 212.9 mm, HP mice: 805.6 ± 188.0 mm, P = 0.94) (Fig. 3). Although both group moved at a similar extent during the initial 2 min, we found LP mice continue to move more frequent-ly than HP mice (F(4,84) = 3.119, P = 0.02) (Fig. 4). The total freeze time (percent) had no significant difference between the two groups (LP mice: 34.2 ± 22.0 %, HP mice: 31.8 ± 18.7 %, P=0.80).

Elevated plus maze test. LP mice spent less time in closed arms than HP mice (LP mice: 264.3 ± 37.8 s, HP mice: 312.6 ± 55.3 s, P = 0.03) (Fig. 5), whereas the total moved distance during the 10-min test period was similar (LP mice: 2013.8 ± 238.8 cm, HP mice: 2127.4 ± 325.2 cm, P = 0.36) (Fig. 6). Thus, this result suggests that LP mice reduced anxiety-like behavior than HP mice.

Open field test. When mice were placed in the appa-ratus, LP mice more frequently traveled between the center and peripheral area than HP mice (LP: 197.3 ± 11.2 times, HP: 185.4 ± 12.6 times, P = 0.03) while the time spent in the center area was similar (LP mice: 111.5 ± 18.3 s, HP mice: 104.7 ± 13.9 s,

central platform facing one of open arms and the activities were recorded for 10 min. The number of entries into open and closed arms, the stay time on the open and closed arms (s), and total distance traveled (cm) were automatically analyzed using Im-age EP software.

Open field test. Anxiety-related behavior was also analyzed by the open field test. Mice were placed in the defined corner of the open-field apparatus (white plexiglass: 50 × 50 × 30 cm) under illumination at 500 lx, and the activities were recorded with a vid-eo camera for 10 min. Total distance (cm) and time spent in the central part of the open field (s) and move episode in the open field (number of times) were automatically analyzed using Image OF soft-ware.

Statistical analysis. All the presented data are ex-pressed as mean ± standard deviation. Statistical analyses were performed with unpaired Student’s t-test unless otherwise specified, where a value of P < 0.05 was considered to be significant.

RESULTS

The consumption of peptides (amount of drinking), food intake and the body weight of animals were similar between LP and HP ingested groups during 3 weeks of ingestion before performing behavioral studies (amount of drinking per day: 6.7 ± 0.7 g in LP mice and 6.7 ± 0.7 g in HP mice, P = 0.99; food intake per day: 3.4 ± 1.3 g in LP mice and 3.6 ± 1.4 g in HP mice, P = 0.65; the body weight of ani-mals: 23.2 ± 0.9 g in LP mice and 23.9 ± 1.3 g in HP mice, P = 0.09).　In contrast, the brain weight of LP mice was in-creased by 3% compared with that of HP mice when the brain was removed to evaluate the level of adult neurogenesis in hippocampus (LP mice: 454.9 ± 11.6 mg, HP mice: 440.1 ± 3.9 mg, P = 0.01).

Measurement of BrdU-positive cells in dentate gyrusWe observed most of BrdU-positive cells in sub-granular zone (SGZ) of dentate gyrus where neural progenitor cells (NPCs) are enriched. We counted the number of BrdU-positive cells along SGZ, and found that the density of proliferating cells in LP mice was 1.2-fold higher than that in HP mice (LP mice: 8.7 ± 1.1 cells/mm SGZ, HP mice: 7.4 ± 0.7 cells/mm SGZ, P = 0.04) (Fig. 1, 2C and 2D). In ad-dition, doublecortin-staining for premature neurons in LP mice were consistently stronger than HP mice

Fig. 1　The quantification of proliferating cells in dentate gy-rus. The linear density of BrdU-positive cells along SGZ is almost 1.2-fold higher in LP mice of 9-week-old-mice. BrdU-positive cell densities in LP mice and HP mice: 8.7 ± 1.1 and 7.4 ± 0.7 cells/mm SGZ, P = 0.04 by Student’s t-test (gray: LP mice, white: HP mice). n = 8 (LP mice), n = 6 (HP mice)

C. Kakoi et al.276

adult mice. Since the level of adult neurogenesis, especially in hippocampus, is susceptible to various environmental stimuli (17), we have expected that intake of different nutrients under no calorie restric-tion might also affect the degree of hippocampal neurogenesis.　We used collagen as an sample which was the most abundant protein in animal tissues and served as protein source for human diet. A mixture of collagen peptide we tested has a mean molecular weight of 367.5, mainly consisted of tri-peptides (Gly-Pro-Hyp is the major component accounted for 8% in mass). Part of the proline in collagen is hy-droxylated by the enzyme reaction in vivo. Proteins

P = 0.84). The total distance traveled during the 10-min test period was not significantly different be-tween groups (LP mice: 4829.8 ± 488.1 cm, HP mice: 4596.2 ± 673.2 cm, P = 0.36). These results suggest that although the differences turned out to be small in this behavioral paradigm, LP-ingested mice showed a tendency of reduced anxiety-like behavior.

DISCUSSION

In the present study, we found that oral administra-tion of the lower molecular weight peptides derived from collagen enhanced hippocampal neurogenesis and exerted anxiolytic and antidepressant effects in

Fig. 2　A, B Fluorescent images of nuclear-staining (green), doublecortin-staining (blue) and BrdU-positive cells-staining (red) in dentate gyrus of hippocampus (left: LP mice, right: HP mice). Especially within the circles of figures, doublecortin-staining for premature neurons in LP mice (A) are stronger than HP mice (B). C, D Distribution of BrdU-positive cells (white) in the same area. Most of BrdU-positive cells were found in subgranular zone (SGZ). The linear density of BrdU-positive cells along SGZ (cells/mm SGZ) is higher in LP mice (C) as compared with HP mice (D). In the presented images, there are twenty BrdU-positive cells per 1858.2 μm subgranular zone in LP, and seven BrdU-positive cells per 1748.8 μm sub-granular zone in HP. Scale bar: 200 μm.

Effect of collagen peptides in mice 277

are absorbed in the intestinal tract after they are degraded to peptides or amino acids. It has been demonstrated in the animal experiment that only lower molecular weight collagen peptides (less than 15,000) can be absorbed (18). Therefore, we consid-ered LP used in this study (mean molecular weights of 367.5) served as efficient substrates compared with LP (mean molecular weights of about 30,000). In general, small molecules less than 500 tend to cross blood-brain barrier (BBB) (24). For example,

peripheral administration of Gly-Pro-Glu is known to cross BBB in adult rats (2). Thus, it suggests that these peptides may be able to permeate into the brain tissue to take effect.　However, we still do not know whether these peptides affect the neural progenitor cells (NPCs) in a direct or indirect manner. For example, some growth factors like brain-derived neurotrophic factor (BDNF) and vascular endothelial growth factor (VEGF) directly act on NPCs to promote its prolif-eration for adult neurogenesis (21, 25). Therefore, in

Fig. 3　Total moved distance (mm) in tail suspension test during the 5-min test period. The total moved distance has no significant difference between the two groups (LP mice: 799.4 ± 212.9 mm, HP mice: 805.6 ± 188.0 mm, P = 0.94, by Student’s t-test, n = 11 and 12). The behavioral tests were performed in 8-week-old male mice. HP littermates were used as controls (gray: LP mice, white: HP mice).

Fig. 4　Distance (mm) measured along motion in tail sus-pension test per one-minute bin (s). Both group move at a similar extent during the initial 2 min, but we found that LP mice continue to move more frequently than HP mice (F(4,84) = 3.119, P = 0.02, by two-way ANOVA, n = 11 and 12).

Fig. 5　Stay time in closed arm (s) during the 10-min test period in elevated plus maze test. LP mice spent less time in closed arms than HP mice (LP mice: 264.3 ± 37.8 s, HP mice: 312.6 ± 55.3 s, P = 0.03, by Student’s t-test, n = 12 and 12).

Fig. 6　The total moved distance (mm) during the 10-min test period in elevated plus maze test. The total moved dis-tance during the 10-min test period is similar (LP: 2013.8 ± 238.8 cm, HP: 2127.4 ± 325.2 cm, P = 0.36, by Student’s t-test, n = 12 and 12).

C. Kakoi et al.278

vitro analysis using NPC cell culture would be nec-essary to address this issue.　Moreover, it remains to be clarified which specif-ic peptides are required for adult neurogenesis. Our preliminary study in cell culture suggests that the application of synthetic peptides of GPVGAR in the presence of 50 ng/mL NGF enhances the differentia-tion of PC12 cells into neurite-bearing cells. There-fore, these peptides might enhance hippocampal neurogenesis and produce antidepressant- and anx-iolyitic-like effects that we have observed with lower molecular weight collagen peptides. Further investi-gation is required to address this issue.　This study also provided novel evidence that in-gestion of the lower molecular weight peptides derived from collagen influenced mood-related be-havior. The correlation of hippocampal neurogenesis and clinical depression has been suggested, in which a reduced level of neurogenesis may provide a ground for the progress of this mental illness. As we expected, antidepressant trend was observed from the result of the tail suspension test. Interestingly, we also found that anxiety-related behavior was im-proved in the elevated plus maze test and the open field test, consistent with the fact that clinical de-pression often accompanies an increased level of anxiety (9). However, it still remains unclear wheth-er there is a common neural pathway for depression and anxiety, and how hippocampal neurogenesis in-fluences these emotional behavior.　In conclusion, we demonstrated that intake of the lower molecular weight peptides derived from colla-gen was able to influence the level of hippocampal neurogenesis and emotional behavior in a correlative manner. Our data indicate that food intake without dietary restriction also influences the level of adult neurogenesis and emotional behavior in a correlated manner. There may be a variety of natural substanc-es yet to be discovered that contributes to the im-provement of human health and prevention of neurological disorders.

Acknowledgement

Financial support for this study was provided by Grant-in-Aid for Scientific Research 20780097 and Asahi Breweries, Japan Research Foundation Re-search Grant.

REFERENCES 1. Ashmarin LP (2007) Glyprolines in regulatory tripeptides.

Neurochem J 1, 173–175. 2. Baker AM, Batchelor DC, Thomas GB, Wen JY, Rafiee M,

Lin H and Guan J (2005) Central penetration and stability of N-terminal tripeputide of insulin-like growth factor-I, gly-cine-proline-glutamate in adult rat. Neuropeptides 39, 81–87.

3. Eriksson PS, Perfilieva E, Bjork-Eriksson T, Alborn A, Norfborg C, Peterson DA and Gage FH (1998) Neurogenesis in the adult human hippocampus. Nat Med 4, 1313–1317.

4. Furuya S (2010) Effect of soy peptide ingestion on the free amino acid contents in adult brain of genetic serine deficien-cy model mice. Soy Protein Res 13, 109–115.

5. Gould E, Tanapat P, McEwen BS, Flugge G and Fuchs E (1998) Proliferation of granule cell precursors in the dentate gyrus of adult monkeys is diminished by stress. Proc Natl Acad Sci USA 95, 3168–3171.

6. Gould E and Tanapat P(1999) Stress and hippocampal neuro-genesis. Biol Psychiatry 46, 1472–1479.

7. Hatakeyama E, Yamaguchi M, Muramoto K, Ito G, Motohashi Y amd Higuchi S (2003) Modulating effects of soy protein isolate and soy protein hydrolysate on human brain function. Soy Protein Res 6, 147–152.

8. Hisatsune T (2003) Regeneration of neural circuit: Role of soy-protein. Soy Protein Res 6, 100–103.

9. Kaiya H (2010) Anxiety and depression: a reapprisal. Jpn J Clin Psychiatry 39, 403–409.

10. Katakura M, Hashimoto M, Shahdat HM, Gamoh S, Okui T, Matsuzaki K and Shido O (2009) Docosahexaenoic aid pro-motes neuronal differentiation by regulating basic helix-loop-helix transcription factors and cell cycle in neural stem cells. Neuroscience 160, 651–660.

11. Kawakita E, Hashimoto M and Shido O (2006) Docosa-hexaenoic acid promotes neurogenesis in vitro and in vivo. Neuroscience 139, 991–997.

12. Kempermann G, Kuhn HG and Gage FH (1998) Experience-induced neurogenesis in the senescent dentate gyrus. J Neu-rosci 19, 3206–3212.

13. Kitamura T, Mishina M and Sugiyama H (2003) Enhance-ment of neurogenesis by running wheel excercises is sup-pressed in mice lacking NMDA receptor 1 subunit. Neurosci Res 47, 55–63.

14. Kitamura T, Mishina M and Sugiyama H (2006) Dietary re-striction increases hippocampal neurogenesis by molecular mechanisms independent of NMDA receptors. Neurosci Lett 393, 94–96.

15. Maekawa M, Takashima N, Matsumura M,Ikegami S, Kontani M, Hara Y, Kawashima H, Ogawa Y, Kiso Y, Yoshikawa T, Inokuchi K and Osumi N (2009) [cited 2012 May 1] Arachi-donic acid drives postnatal neurogenesis and elicits a benefi-cial effect on prepulse inhibition, a biological trait of psychiatric illnesses. PLoS One[serial online] 4, e5085.

16. Mogi M and Horiuchi M (2008) Aging brain and RAS. An-giotensin Res 5, 19–23.

17. Nilsson M, Perfilieva E, Johansson U, Orwar O and Eriksson PS (1999) Enriched environment increases neurogenesis in the adult rat dentate gyrus and improves spatial memory. J Neurobiol 9, 569–578.

18. Oesser S, Adam M, Babe W and Seifert J (1999) Oral ad-ministration of 14C labeled gelatin hydrolysate leads to an accumulation of radioactivity in cartilage of mice (C57/BL). J Nutr 129, 1891–1895.

19. Paul SM (1988) Anxietry and depression: a common neuro-biological substrate? J Clin Psychiatry 49 Suppl, 13–16.

20. Praag H, Kempermann G and Gage FH (1999) Running in-creases cell proliferation and neurogenesis in the adult mouse dentate gyrus. Nat Neurosci 2, 266–270.

21. Rossi C, Angelucci A, Costantin L, Braschi C, Mazzantini M,

Effect of collagen peptides in mice 279

Babbini F, Fabbri ME, Tessarollo L, Maffei L, Berardi N and Caleo M (2006) Brain-derived neurotrophic factor (BDNF) is required for the enhancement of hippocampal neurogenesis following environmental enrichment. Eur J Neurosci 24, 1850–1856.

22. Sapolsky RM, Krey LC and McEwen BS (1986) The neuro-endocrinology of stress and aging: the glucocorticoid cascade hypothesis. Endocr Rev 7, 284–301.

23. Steru L, Chermat R, Thierry B and Siomo P (1985) The tail suspension test: a new method for screening antidepressants in mice. Psychopharmacology (Berl) 85, 367–370.

24. Tanobe K, Nishikawa K, Hinohara H, Okamoto T, Saito S and Goto F (2003) Blood-brain barrier and general anesthet-

ics. Jpn J Anesthesil 52, 840–845.25. Udo H, Yoshida Y, Kino T, Ohnuki K, Mizunoya W, Mukuda

T and Sugiyama H (2008) Enhanced adult neurogenesis and angiogenesis and alterd affective behavior in mice overex-pressing vascular endothelial growth factor 120. J Neurosci 28, 14522–14536.

26. Watanabe Y, Gould E and McEwen BS (1992) Stress induces atrophy of apical dendrites of hippocampal CA3 pyramidal neurons. Brain Res 588, 341–345.

27. Woolley CS, Gould E and McEwen BS (1990) Exposure to excess glucocorticoids alters dendritic morphology of adult hippocampal pyramidal neurons. Brain Res 531, 225–231.