Effects of glucose-dependent insulinotropic peptide on behavior

peptides 27 (2006) 2750–2755 available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/peptides Effects of glucose-dependent insulinotropic peptide on behavior Ke-Hong Ding a, Qing Zhong a, Ding Xie a, Huan-Xin Chen f, Mary Anne Della-Fera e, Roni J. Bollag c, Wendy B. Bollag a, Ravinder Gujral a, Baolin Kang b, Supriya Sridhar a, Clifton Baile e, Walton Curl a,b, Carlos M. lsales a,b,d,* a Medical College of Georgia, Institute of Molecular Medicine and Genetics, Department of Medicine, CB-2803, 1120 15th Street, Augusta, GA 30912, USA b Department of Orthopedic Surgery, Medical College of Georgia, Augusta, GA 30912, USA c Department of Pathology, Medical College of Georgia, Augusta, GA 30912, USA d The Augusta VAMC, Augusta, GA 30912, USA e University of Georgia, Department of Animal and Dairy Science, Athens, GA 30602, USA f Department of Neurological Surgery, University of Florida, Gainesville, FL 32610, USA article info abstract Article history: Glucose-dependent insulinotropic peptide (GIP) is an incretin hormone that rises rapidly in Received 21 March 2006 response to nutrient ingestion. The GIP receptor is widely expressed in the brain including Received in revised form the brain stem, telencephalon, diencephalon, olfactory bulb, pituitary, and cerebellum. Until 8 May 2006 recently it was not clear what the endogenous ligand for this receptor was because no GIP Accepted 9 May 2006 expression had been demonstrated in the brain. GIP synthesis has now been documented in Published on line 5 July 2006 the dentate gyrus of the hippocampus. To define GIP effects on behavior we utilized a mouse model a GIP-overexpressing transgenic mouse (GIP Tg). Specifically, anxiety-related beha- Keywords: vior, exploration, memory, and nociception were examined. Compared to age-matched GIP receptor adult male C57BI/6 controls GIP Tg mice displayed enhanced exploratory behavior in the Brain open-field locomotor activity test. GIP Tg mice also demonstrated increased performance in Transgenic mice some of the motor function tests. These data suggest that the GIP receptor plays a role in the Anxiety regulation of locomotor activity and exploration. To our knowledge, this is the first report of Memory effects of GIP on behavior. # 2006 Elsevier Inc. All rights reserved. Exploration 1. Introduction Glucose-dependent insulinotropic peptide (GIP), a 42 amino acid peptide released from endocrine cells in the small intestine in response to a meal, is known to potentiate nutrient-stimulated insulin release [4]. GIP receptors (GIPR) belong to the seven transmembrane domain G proteincoupled receptor family and are widely distributed in the brain with receptor mRNA being expressed at high levels in the olfactory bulbs, layers 3 and 5 of the cerebral cortex, ventral and dorsal hippocampus and mammilary bodies [11]. There is also moderate GIPR mRNA expression in the anterior and lateral septum, cortical amygdala, claustrum, subthalamic nucleus, substantia nigra, inferior olive and rostral raphe nuclei and choroid plexus [11]. The function of these GIPR was not clear since GIP, in contrast to leptin, does not appear to * Corresponding author. Tel.: +1 706 721 0692; fax: +1 706 721 7915. E-mail address: [email protected] (C.M. lsales). 0196-9781/$ – see front matter # 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.peptides.2006.05.011 peptides 27 (2006) 2750–2755 suppress food intake [5]. However, recently the peptide itself, GIP has been shown to be expressed in the hippocampus and to induce progenitor cell proliferation [7]. In fact GIP receptor knockout mice were demonstrated to have less than half the number of new proliferating cells in the hippocampus than did wild-type mice [7]. Thus, this result would suggest that specific areas in the brain are able to synthesize and respond to GIP. Signals for GIP release in the brain are not well defined but presumably might also respond to changes in nutritional state. As part of our efforts to define GIP’s role as an integrative hormone modulating multiple systems within the body to maximize nutrient absorption [1,2], we utilized a genetic mouse model a GIP-overexpressing transgenic mouse, in which serum GIP levels are two to six times higher than in the wild-type mouse. We have demonstrated that this mouse model and a related mouse model, the GIP receptor knockout mice have large differences in bone mass, with GIP receptor knockout mice having a low bone mass compared to wild-type mice [12] and GIP-overexpressing transgenic mice exhibiting a higher bone mass than control mice [13]. Since GIP blood concentrations reflect nutrient ingestion, we studied effects this peptide has on selected behaviors related to anxiety, exploration, memory, and nociception. We report that alterations in GIP levels result in significant changes in mouse behavior. 2. Materials and methods 2.1. Animals 2.1.1. GIP-overexpressing transgenic mice In view of data that the GIP receptor can be downregulated by high doses of the peptide hormone [1], we designed a construct that contained a regulatable promoter. Thus, the regulatory elements associated with the mouse metallothionein promoter, which have been characterized extensively by Palmiter et al. [8,9], were incorporated. The full-length GIP cDNA was inserted into the vector MT2999 (a kind gift of Dr. Richard Palmiter) for generating GIP-overexpressing transgenic mice (Fig. 1). In this construct the GIP coding sequences are flanked by 17 kb of regulatory sequences from the endogenous metallothionein locus (MTI and MTII), leading to an insular construct that retains proper regulation with limited artifactual effects derived from the site of integration. Colonies of transgenic mice were established and characterized by Southern blot and GIP radioimmunoassay. A transgenic mouse line with greater than 10 integrated copies was selected and GIP levels were measured. All experiments were performed in accordance with the guidelines put forth by the Institutional Animal Care and Use Committee at the Medical College of Georgia. Unless otherwise noted, behavioral tests were performed on 11 male transgenic or wild-type mice per group, of 3 months of age. The same animal was used for multiple tests and the order of testing was: (1) hot plate; (2) light/dark preference; (3) tail flick; (4) rotarod; (5) Y-maze; (6) plus maze; (7) two-trial novel object recognition task; and (8) open field locomotor activity. 2751 Fig. 1 – (A and B) Schematic of the MT-GIP construct used to generate GIP-overexpressing transgenic mice. A 500 bp mouse GIP cDNA was inserted into the unique Nrul site of vector 2999. Vector 2999 incorporates the proximal mouse metallothionein I promoter upstream and human growth hormone polyadenylylation signals downstream of the introduced cDNA. This minigene is insulated by 17 kb of 50 and 30 regulatory sequences deriving from the genomic metallothionein locus to ensure tighter transcriptional regulation. In the transgenic group the genetic manipulation has no effect on body weight and growth compared to wild-type mice at 3 months of age (GIP+/+ versus WT body weight (g): 26.3  1.85 versus 25.3  2.14, p = 0.2070; length (cm): 9.6  0.19 versus 9.5  0.24, p = 0.1295). 2.2. Anxiety-related behavior tests 2.2.1. Elevated plus maze Anxiety levels were assessed using an elevated plus-shaped maze consisting of two open arms and two closed arms equipped with rows of infrared photocells interfaced with a computer (Hamilton–Kinder, Poway, CA). Normal rodents avoid the open arms of the plus maze so that decreases in time spent in and entries into the open arms are thought to reflect enhanced anxiety [6]. Mice were placed individually in the center of the maze and allowed free access for 5 min. They could spend their time either in a closed safe area (closed arms), in an open area (open arms), or in the middle, intermediate zone. Recorded beam breaks were used to calculate the time spent and the distance moved in the open and closed arms and the number of times the mice extended over the edges of the open and closed arms. After behavioral 2752 peptides 27 (2006) 2750–2755 testing, the equipment was cleaned with 1 mM acetic acid to remove odors. time that passes before the mouse lifts a hind paw from the surface of the plate was recorded. 2.2.2. 2.5. Memory tests 2.5.1. Y-maze Light/dark preference This apparatus consisted of a Plexiglas rectangular box (45 cm long  27 cm wide  27 cm high) divided into a dark region (15 cm long) and a larger light region (30 cm long) as in Bourin and Hascoet [3]. The light and dark regions were separated by an opening (7.5 cm  7.5 cm) that allowed the animals to move between the two compartments. The dark region was made of black opaque material and covered with a black lid. The light portion was made of Plexiglas, and a 60-W light was positioned directly over it. On the day of testing, each mouse was transported individually from the housing room to the testing room. The mouse was placed in the light compartment and was allowed to move freely between the two compartments. Behavior was video-recorded for a total of 5 min, and the videotapes were scored for latency to return to the light compartment and the number of transitions between the light and dark compartments by an investigator blind to the group assignments of the animals. 2.3. Motor function tests 2.3.1. Open field locomotor activity Mice were placed in a Y-shaped maze for 5–8 min. All arm entries were sequentially scored so that the total number of arm entries, as well as the sequence of entries, were recorded. Data were analyzed to determine the number of arm entries without repetition. Success in this test is indicated by a high rate of alternation in the control groups indicating that the animals can remember which arm was entered last. 2.5.2. 2.6. Automated open field locomotor activity was measured using an Omnitech Digiscan (model CCDIGIO) optical animal activity monitoring system that uses horizontal and vertical banks of photo beam sensors to monitor several categories of animal movement with time. Each animal was placed in a clear test cage (40 cm  40 cm  30 cm), and the trial initiated immediately. Locomotor activity was monitored continuously for 30 min, and the data were accumulated and processed in a spreadsheet format. The following parameters were recorded: horizontal activity, movement time, number of stereotype movements, and vertical activity. A complete description and definition of each parameter has been published [10]. 2.3.2. Rotorod Each animal was placed on an accelerating rotorod (Ugo Basile, Italy) that used a rotating treadmill that accelerated from 4 to 40 rpm over a 5-min period. Animals were given one trial on the apparatus on the day prior to the experiment. On the day of the experiment, the total number of seconds maintained on the rotorod was recorded. 2.4. Tests of nociception 2.4.1. Tail flick Mice were restrained in a conical polypropylene tube with an opening through which its tail was exposed to a heat lamp and the amount of time necessary for the animal to move (flick) its tail away from the heat was recorded. 2.4.2. Thermal sensitivity test In this test the mice were placed on an apparatus that gradually increased in temperature. The temperature of the heating surface was elevated by 3 8C/min from a beginning temperature of 42 8C to a maximum temperature of 49 8C. The Two-trial novel object recognition task With one arm of the Y-maze blocked, mice were placed in one of the arms and allowed to explore the two arms for 15 min. The mice were returned to the maze 1–4 h later with all arms open and scored for 5 min. The first arm entered, the amount of time spent in each arm, and the number of entries into each arm were recorded. Seven to 10 days later the test was repeated with a delay time of only 2 min between the trials. Statistical analyses Results are expressed as mean  S.E.M. Data were analyzed using either ANOVA or unpaired t-tests, where appropriate, with a statistical computer package (Instat, Graphpad Software Inc., San Diego, CA). 3. Results 3.1. GIP-overexpressing transgenic mice had elevated serum levels of GIP Because of concerns about GIP receptor downregulation with high GIP levels, a construct was generated using a heavy metal inducible promoter (Fig. 1A). However, even at baseline prior to heavy metal ingestion, transgenic mice had more than two-fold higher serum levels of GIP than control mice (Fig. 1B). Further, after ingestion of 25 mM zinc sulfate added to the drinking water, the GIP Tg mice had a nearly six-fold rise in serum GIP levels. Sustained elevation of GIP at these high levels however, resulted in GIP receptor downregulation as determined by Western blot analysis (data not shown). Thus, no heavy metal supplement was given to any of the GIP Tg mice for any of the subsequent experiments. Since GIP is a hormone that rises rapidly after a meal and may play a role in satiety, we first examined whether alterations in GIP levels had an impact on anxiety-related behavior. 3.2. GIP had no effect on levels of anxiety Both the elevated Plus-Maze and the test for light dark preference evaluate an animals’ overall level of anxiety. The elevated Plus-Maze, by measuring the amount of time spent in enclosed (or protected) areas of an elevated platform versus time spent in open areas, monitors the animals’ desire to avoid peptides 27 (2006) 2750–2755 2753 stressful or anxiety-inducing environments. GIP-overexpressing transgenic mice did not differ in the amount of time spent in the open arms of the platform compared to control mice (ratio of time in open arms/total (%): 10.6  1.3 versus 11.7  1.8, control versus Tg; mean  S.E.M., p = 0.627). Similarly, in the test for light–dark preference, there was no difference between GIP Tg mice and controls (ratio of time in light/dark (%): 42.6  4.2 versus 56.3  10.7, control versus Tg; mean  S.E.M., p = 0.254). Increased time spent in the lighted area of the cage is consistent with decreased anxiety. We next examined the effects of GIP on locomotor activity. 3.3. GIP overexpression did not negatively impact locomotor activity In this series of experiments animals were placed in an open cage and motor activity and movements recorded. These tests ascertain whether genetic manipulation negatively impacts on motor, hippocampal or basal ganglia function and thereby mouse behavior. As shown in Fig. 2, GIP-overexpressing transgenic did not have impaired locomotor activity compared to control mice. In fact activity levels were significantly increased: Distance traveled (cm) in total—784.6  67 versus 1212.3  88, control versus Tg; mean  S.E.M., p = 0.001; center—572.8  62 versus 869.4  80, control versus Tg; mean  S.E.M., p = 0.009; periphery—211.7  22 versus 342.9.7  26, control versus Tg; mean  S.E.M., p = 0.001. Any potential impact of genetic manipulation on motor function was further assessed by use of the rotarod test. 3.4. GIP overexpression did not negatively impact motor coordination Sensorimotor coordination of GIP transgenic mice was evaluated using the rotarod test. In this test mice were placed Fig. 2 – GIP transgenic mice did not have impaired locomotor activity. In order to assess whether genetic manipulation negatively affected basal ganglia or hippocampal development, activity levels were measured for the GIP transgenic and control mice. None of the study groups were found to have impaired locomotor activity compared to control animals. Each point represents the mean W S.E.M. of 11 mice per group. There were significant differences in activity level between control and Tg mice (total: **p = 0.001 with 20 degrees of freedom; center: *p = 0.009 with 20 degrees of freedom; periphery: ** p = 0.001 with 20 degrees of freedom). Fig. 3 – (A and B) GIP transgenic mice demonstrated improved sensorimotor coordination. GIP transgenic and control mice were placed on a rotating rod the speed of which was gradually increased and the time spent on the rod before falling off was measured. At two time points, GIP transgenic mice were found to perform significantly better than control animals. Data represent the mean W S.E.M. of 10–12 mice/group. *p < 0.05 GIP transgenic vs. wild-type mice. on a rotating rod whose speed is gradually increased and the time spent on these rods measured. This test evaluates basal ganglia and cerebellar function. As shown in Fig. 3, GIP transgenic mice did not demonstrate decreased motor coordination compared to control animals. In fact GIP transgenic mice show a consistent trend (which was statistically significant on two different trials) to be able to spend more time on the rotating rods compared to control animals (rotarod day 1 trial 3 (s): 27.5  3.7 versus 50.9  8.8, control versus Tg; mean  S.E.M., p = 0.029; rotarod day 2 trial 2 (s): 47.4  7.3 versus 68.8  6.9, control versus Tg; mean  S.E.M., p = 0.0464). To evaluate whether changes in GIP had an impact on pain perception we next performed tests for nociception. 3.5. GIP overexpression had no impact on nociception In this series of experiments the sensitivity to spinal perception of pain was assessed in the GIP transgenic mice. Mice were restrained in a conical tube and the amount of time taken for moving their tail from under a heat lamp was measured. GIP-overexpressing transgenic mice did not differ in tail flick latency compared to control mice (Tail flick latency 2754 peptides 27 (2006) 2750–2755 in seconds: 7  0.4 versus 6.7  0.3, control versus Tg; mean  S.E.M., p = 0.566). In contrast, to assess supraspinal nociception, experiments using a thermal sensitivity test were performed. In this series of experiments, mice were placed on a hot plate that gradually increased in temperature (38 min 1 from 42 to 49 8C) and the time taken for the animals to lift their hindpaws from the hot plate was measured. There was no significant difference between the test groups animals (hot plate average time (s): 79.5  3.1 versus 80.6  4.6, control versus Tg; mean  S.E.M., p = 0.845; hot plate average temperature (8C): 46.1  0.1 versus 46.0  0.2, control versus Tg; mean  S.E.M., p = 0.661). 3.6. GIP-overexpressing transgenic mice demonstrated improved performance in the Y-maze In order to assess potential hippocampal damage and as a way to evaluate any impact on memory, mice were placed in a Y-maze for 8 min and then the number of entries into each arm measured. A high number of alternations is consistent with the mice remembering the arm that they previously entered and subsequently testing the other arm. As can be seen in Fig. 4A, there was no significant difference between the Fig. 4 – GIP transgenic mice perform better than control mice in the Y-maze. GIP transgenic or control mice were placed in a Y-maze and the number of alternations (entering the different arms) an index of how well they remember the arm previously entered (panel A) was measured. Alternatively, animals were placed in a Y-maze, removed and subsequently again placed in the maze and the number of entries into each arm measured as an index of how well they remembered their previous time in the maze (panel B). Data represent the mean W S.E.M. of 10–12 mice per group. **p < 0.01 GIP transgenic vs. wild-type mice. different mouse groups tested. However, in the two-trial recognition memory test, mice were permitted to explore the Y-maze, in which one of the arms was blocked, for fifteen minutes, removed from the maze for 1–4 h and then again placed in the maze and the arm of the maze entered and the number of entries recorded. This test evaluates longer term memory and as can be seen in Fig. 4B, GIP transgenic mice had a significantly greater number of total entries than control mice (total entries: 23.9  1.5 versus 34.9  2.0, control versus Tg; mean  S.E.M., p = 0.0003) although there were no significant differences in the % novel entries between the Tg and control mice. This would suggest that differences in performance may be related to the increased activity observed in the transgenic mice. 4. Discussion The present results demonstrate that GIP-overexpressing transgenic mice have better sensorimotor coordination and memory recognition compared to control mice. The widespread distribution of the GIP receptor in the brain has been known for some time [11]; however the role these receptors play in normal brain function has been unclear, since GIP synthesis has been well documented to occur almost exclusively in the endocrine (K) cells of the small intestine. Nyberg et al. [7] have recently shown however that GIP mRNA and immunoreactivity is present in the rat hippocampus. These investigators also demonstrated that the GIP receptor is expressed in hippocampal progenitor cells and in vivo and in vitro experiments showed that GIP increased hippocampal cell proliferation. Interestingly, these investigators also examined GIP receptor knockout mice and found that these mice had fewer than half the number of proliferating cells in the hippocampus than did wild-type mice. However, no behavioral testing was performed on the GIPR knockout mice [7]. The hippocampus is important in a number of brain processes including recent memory and orientation. GIP-overexpressing transgenic animals did significantly better than wild-type mice in the Y-maze (Fig. 4) and coordination (Figs. 2 and 3) as might be predicted from the effect of GIP on hippocampal cell proliferation. Although part of the increase in total entries in the two-trial recognition memory test may relate to the increased locomotor activity. The role of GIP in normal physiology has become increasingly relevant as new therapies for treatment of diabetes mellitus become available. GIP and glucagon-like peptide-1 (GLP-1) are well known ‘‘incretin’’ hormones, that is, they potentiate glucose-induced insulin secretion. Both GIP and GLP-1 are rapidly broken down by the enzyme dipeptidyl peptidase IV (DPP IV) and inhibitors of this enzyme are now in clinical trials for treatment of diabetes mellitus type II. The impact these medications may have on these neuropeptides and brain function is not clear. Our data suggest that GIP overexpression in our transgenic mice can have significant effects on a number of behavioral tests. Whether the observed behavioral changes are due only to changes in circulating GIP levels or there may be other developmental differences between control and transgenic mice is not completely clear thus our findings must be interpreted with caution. However, peptides 27 (2006) 2750–2755 only a relatively narrow spectrum of behavioral tests were performed and how these changes in behavior in mice may translate to effects on human brain function is also unclear. This suggests that human testing on behavior on patients receiving medication that can alter the levels of these neuropeptides is warranted and that until these test results become available that caution may be justified. Acknowledgements This study was funded by a grant from the National Institutes of Health (R01DK058680, CMI). We thank Dr. Alvin Terry and the small animal behavior core for behavioral testing. reference [1] Bollag RJ, Zhong Q, Ding KH, Phillips P, Zhong L, Qin F, et al. Glucose-dependent insulinotropic peptide is an integrative hormone with osteotropic effects. Mol Cell Endocrinol 2001;177:35–41. [2] Bollag RJ, Zhong Q, Phillips P, Min L, Zhong L, Cameron R, et al. Osteoblast-derived cells express functional glucosedependent insulinotropic peptide receptors. Endocrinology 2000;141:1228–35. [3] Bourin M, Hascoet M. The mouse light/dark box test. Eur J Pharmacol 2003;463:55–65. [4] Fehmann HC, Goke R, Goke B. Cell and molecular biology of the incretin hormones glucagon-like peptide-I and glucosedependent insulin releasing polypeptide. Endocr Rev 1995;16:390–410. 2755 [5] Li Q, Choi Y-H, Hartzell DL, Della-Fera MA, Isales CM, Hamrick MW, et al. Effects of Intracerebroventricular administration of GIP and leptin on food intake, body weight and spontaneous activity in rats. In: XXXIV Society of Neurosciences Meeting, Washington, DC; 2005. [6] Lister RG. Ethologically-based animal models of anxiety disorders. Pharmacol Ther 1990;46:321–40. [7] Nyberg J, Anderson MF, Meister B, Alborn AM, Strom AK, Brederlau A, et al. Glucose-dependent insulinotropic polypeptide is expressed in adult hippocampus and induces progenitor cell proliferation. J Neurosci 2005;25:1816–25. [8] Palmiter RD. Transgenic mice—the early days. Int J Dev Biol 1998;42:847–54. [9] Palmiter RD, Sandgren EP, Koeller DM, Brinster RL. Distal regulatory elements from the mouse metallothionein locus stimulate gene expression in transgenic mice. Mol Cell Biol 1993;13:5266–75. [10] Sanberg PR, Hagenmeyer SH, Henault MA. Automated measurement of multivariate locomotor behavior in rodents. Neurobehav Toxicol Teratol 1985;7:87–94. [11] Usdin TB, Mezey E, Button DC, Brownstein MJ, Bonner TI. Gastric inhibitory polypeptide receptor, a member of the secretin-vasoactive intestinal peptide receptor family, is widely distributed in peripheral organs and the brain. Endocrinology 1993;133:2861–70. [12] Xie D, Cheng H, Hamrick M, Zhong Q, Ding KH, Correa D, et al. Glucose-dependent insulinotropic polypeptide receptor knockout mice have altered bone turnover. Bone 2005;37:759–69. [13] Xie D, Ding K, Hamrick M, Mulloy A, Insogna K, Isales C. Glucose-dependent insulinotropic peptide regulates bone turnover in vivo. J Bone Miner Res 2004;19:S462.