Lucas F.R. Nascimento,1 Gabriela F.P. Souza,1 Joseane Morari,1 Guilherme O. Barbosa,2

Carina Solon,1 Rodrigo F. Moura,1 Sheila C. Victório,1 Letícia M. Ignácio-Souza,1

Daniela S. Razolli,1 Hernandes F. Carvalho,2 and Lício A. Velloso1

n-3 Fatty Acids Induce Neurogenesisof Predominantly POMC-ExpressingCells in the HypothalamusDiabetes 2016;65:673–686 | DOI: 10.2337/db15-0008

Apoptosis of hypothalamic neurons is believed to playan important role in the development and perpetuationof obesity. Similar to the hippocampus, the hypothalamuspresents constitutive and stimulated neurogenesis,suggesting that obesity-associated hypothalamic dys-function can be repaired. Here, we explored the hypoth-esis that n-3 polyunsaturated fatty acids (PUFAs) inducehypothalamic neurogenesis. Both in the diet and injecteddirectly into the hypothalamus, PUFAs were capable ofincreasing hypothalamic neurogenesis to levels similar orsuperior to the effect of brain-derived neurotrophic factor(BDNF). Most of the neurogenic activity induced by PUFAsresulted in increased numbers of proopiomelanocortinbut not NPY neurons and was accompanied by increasedexpression of BDNF and G-protein–coupled receptor 40(GPR40). The inhibition of GPR40 was capable of reducingthe neurogenic effect of a PUFA, while the inhibition ofBDNF resulted in the reduction of global hypothalamiccell. Thus, PUFAs emerge as a potential dietary approachto correct obesity-associated hypothalamic neuronal loss.

The consumption of dietary fats is regarded as one of themost important epidemiological factors leading to theincreased prevalence of obesity in the world (1,2). Owingto their energetic value, dietary fats have a direct impacton overall caloric consumption, which can, per se, favorthe increase in body mass. However, studies performedover the last 10 years have unveiled additional mecha-nisms linking dietary fats to obesity. In rodents, saturatedfatty acids induce hypothalamic inflammation throughthe activation of TLR4 signaling and endoplasmic reticu-lum stress (3,4). In the short term, the hypothalamic neu-rons involved in the control of energy homeostasis areaffected at the functional level and develop resistance to

the adipostatic signals delivered by leptin and insulin(5,6). However, over time, neurons may become perma-nently damaged and undergo apoptosis (7,8). In addition,studies using different neuroimaging methods suggestthat obese humans also present abnormalities in thehypothalamus (8,9).

Neuronal plasticity contributes to the cyclical renewal ofhypothalamic neurons and is known to play an importantrole in the physiological regulation of whole-body energyhomeostasis (10). A part of the renewal process that war-rants the continuous turnover of hypothalamic neuronsdepends on leptin and ciliary neurotrophic factor (CNTF),which induce neurogenesis of the cells that express neuro-transmitters involved in the control of feeding and thermo-genesis, as well as proteins involved in the response toleptin (11,12). However, recent studies have shown thatthe turnover of hypothalamic neurons is disrupted by theconsumption of dietary fats (13,14). In fact, the adult neu-ral stem cells of the mediobasal hypothalamus are damagedby the inflammatory signals induced by dietary fats, placingdiet-induced hypothalamic inflammation in an upstreamposition related to hypothalamic neurogenesis defects inobesity (14).

The beneficial outcomes of the dietary consumption ofpolyunsaturated fatty acids (PUFAs) have been known formany years (15). These outcomes have been shown at theepidemiological level, such as the cardiovascular protectiveeffect of the Mediterranean diet and the consumption offish oil by the Eskimos (15–17), and at the cellular andmolecular levels as illustrated by the regulation of lympho-cyte function and the activation of anti-inflammatory me-diator synthesis (18,19). In a recent study, we showed thatthe dietary substitution of saturated fat by PUFAs reducedobesity-associated hypothalamic inflammation, resulting in

1Laboratory of Cell Signaling, University of Campinas, Campinas, Brazil2Department of Cell Biology, University of Campinas, Campinas, Brazil

Corresponding author: Lício A. Velloso, lavelloso.unicamp@gmail.com.

Received 5 June 2015 and accepted 17 October 2015.

© 2016 by the American Diabetes Association. Readers may use this article aslong as the work is properly cited, the use is educational and not for profit, andthe work is not altered.

See accompanying article, p. 551.

Diabetes Volume 65, March 2016 673

OBESITYSTUDIES

increased responsiveness to leptin and body mass reduc-tion (20). Here, we asked whether PUFAs, and particularlydocosahexaenoic acid (DHA), are capable of inducing hypo-thalamic neurogenesis. Our results show that when admin-istered either via diet or injection directly into thehypothalamus, PUFAs increase the hypothalamic neuro-genesis, which is repressed by the inhibition of G-protein–coupled receptor 40 (GPR40).

RESEARCH DESIGN AND METHODS

Experimental Animals and ProtocolsAll experimental procedures were performed in accor-dance with the guidelines of the Brazilian College for An-imal Experimentation and were approved by the EthicsCommittee at the University of Campinas. Five-week-oldmale Swiss albinus mice were maintained in individualcages at 216 2°C, with a 12–12 h dark–light cycle and dietand water ad libitum. The mice were submitted to sixdifferent experimental protocols. In protocol 1, the micewere fed a high-fat diet (HFD) for 8 weeks and then ran-domly divided into four groups that were fed either anHFD or an HFD with partial (10% or 20%) or total(30%) substitution of the predominantly saturated fat con-tent with flaxseed oil containing ;45% C18:3 (n-3) foranother 8 weeks. In protocol 2, the mice were fed anHFD for 8 weeks and then randomly divided into fourgroups that for another 8 weeks were fed an HFD, anHFD with a partial substitution of the fat content corre-sponding to 20% of the predominantly saturated fat con-tent with flaxseed oil containing ;45% C18:3 (n-3), anormal-fat diet (NFD) (chow), or an NFD with 2.3% (wt/wt)supplementation with flaxseed oil containing ;45% C18:3.During the final 10 days, the mice were treated with a so-lution containing 50 mg/kg of BrdU (no. B5002; Sigma-Aldrich) intraperitoneally twice a day. In protocol 3, themice were fed an HFD for 8 weeks and then randomlydivided into three groups that were treated for 10 dayswith a daily injection of saline (2.0 mL i.c.v.), brain-derivedneurotrophic factor (BDNF) (no. B3795; Sigma-Aldrich)(10 ng in 2.0 mL i.c.v.), or DHA (no. D2534, Sigma-Aldrich)(10 ng in 2.0 mL i.c.v.); additionally, the mice were treatedintraperitoneally with a solution containing 50 mg/kgBrdU twice a day and were then transferred to an NFD(chow) for another 20 days. In protocol 4, the mice werefed an HFD for 8 weeks and then randomly divided intothree groups that were treated for 10 days with saline(200 mL i.p.), DHA (1.0 mg/kg in 200 mL i.p.), or a higherconcentration of DHA (5.0 mg/kg in 200 mL i.p.). In pro-tocol 5, the mice were fed an HFD for 8 weeks and thenrandomly divided into four groups that were treated for anadditional 8 weeks. In two of these groups, the mice weretransferred to an NFD with 2.3% (wt/wt) supplementa-tion with flaxseed oil containing ;45% C18:3 and treat-ed with either an anti-BDNF antibody (no. sc546; SantaCruz Biotechnology, Dallas, TX) (0.8 mg in 100 mL i.p.twice a week) or a similar volume of a preimmune serum(no. R9133; Sigma-Aldrich). In the remaining two groups,

the mice were transferred to an HFD with a partial sub-stitution of the fat content with 20% flaxseed oil containing;45% C18:3 and treated with an anti-BDNF antibody(0.8 mg in 100 mL i.p. twice a week) or a similar volumeof a preimmune serum. In protocol 6, the mice were fed anHFD for 8 weeks and then randomly divided into fourgroups that were treated for 10 days with a daily injectionof vehicle plus saline (2.0 mL i.c.v.), Gw1100 (100 ng in2.0 mL inhibitor of GPR40 i.c.v.; Cayman Chemical Co.,Ann Arbor, MI) plus saline, vehicle plus DHA (10 ng in2.0 mL i.c.v.), or Gw1100 (100 ng in 2.0 mL i.c.v.) plusDHA (10 ng in 2.0 mL i.c.v.); additionally, the mice weretreated intraperitoneally with a solution containing 50 mg/kgBrdU twice a day and were then transferred to an NFD(chow) for another 20 days.

Intracerebroventricular InstrumentationIn some experiments, mice were stereotaxically instru-mented using a Stoelting stereotaxic apparatus to implant acannula. The stereotaxic coordinates were as follows: ante-roposterior, 0.34 mm; lateral, 1.0 mm; and depth, 2.2 mm tothe lateral ventricle. The cannula efficiency was tested1 week after cannulation by the evaluation of the drinkingresponse elicited by the injection of 1026 mol/L i.c.v. angio-tensin II (no. A9525; Sigma-Aldrich).

Analysis of the Fatty Acid Composition in the Diets andHypothalmiThe methyl esters of fatty acids were prepared as previouslydescribed (21) using a CGC Capillary Gas ChromatographAgilent 6850 Series GC System equipped with a capillary col-umn Agilent DB-23 (50% cyanopropyl-methylpolysiloxane)that was 60 m long with a 0.25-mm internal diameter anda 0.25-mm film.

Glucose Tolerance TestAfter 6 h of fasting, the blood glucose was measured, andthen a glucose solution (2.0 g/kg i.p.) was administered. Theblood glucose was measured after 30, 60, 90, and 120 min.

Insulin Tolerance TestAfter 6 h of fasting, the blood glucose was measured,after which an insulin solution (1 unit/kg i.p.) was admin-istered. The blood glucose was measured after 5, 10, 15, 20,and 25 min.

Blood Glucose and InsulinThe blood glucose was measured using a glucometer fromAbbott (Optium, Abbott Diabetes Care, Alameda, CA). Theinsulin level was determined using ELISA kits (no. EZRMI-13K;Millipore, Billerica, MA).

Leptin Tolerance TestAfter 12 h of fasting, 1026 mol/L i.p. leptin (no. 429705;Calbiochem, Darmstadt, Germany) was administered at6:00 P.M., and the spontaneous food intake was measuredfor 12 and 24 h.

Determination of Spontaneous ActivitySpontaneous activity was determined using an automaticsystem from Harvard Apparatus (LE405) (Panlab-Harvard,Holliston, MA).

674 Nutrient-Induced Hypothalamic Neurogenesis Diabetes Volume 65, March 2016

ImmunoblottingHypothalamus protein extract samples were prepared aspreviously described (20). Doublecortin (no. 4604; Cell Sig-naling, Boston, MA), Bax, Bcl-2, BDNF, GPR40, and b-actin(nos. sc493; sc492, sc546, sc32905, and sc130656; SantaCruz Biotechnology) antibodies were used to detect thetarget proteins. Enhanced chemiluminescence (SuperSignalWest Pico, Pierce) after incubation with a horseradishperoxidase–conjugated secondary antibody was used fordetection by autoradiography. The band intensities were

quantified by optical densitometry (UN-Scan-it Gel 6.1,Silk Scientific, Orem, UT).

RNA Extraction and Real-Time PCRThe mRNA levels of BDNF and GPR40 were measuredin the hypothalamus by real-time PCR (ABI Prism 7500detection system, Applied Biosystems, Grand Island, NY).The intron-skipping primers were obtained from AppliedBiosystems (Mn01334043_m1 and Mm00809442_s1, re-spectively). GAPDH (no. 4352339E; Applied Biosystems)

Figure 1—The lipid composition of flaxseed oil and metabolic outcomes of increased consumption of dietary PUFAs. A sample of flaxseedoil used to prepare the diets was analyzed by gas chromatography. The details of the composition are presented in Table 1. The figuredepicts a representative spectrum obtained from three distinct determinations (A). Swiss mice were submitted to one of the dietaryapproaches as depicted in B. Body mass variation (C ) and mean daily food intake (D) were determined throughout the study. At theend of the experimental period, mice were submitted to a leptin tolerance test (E ), an evaluation of spontaneous activity (F ), measurementof fasting blood glucose levels (G), a glucose tolerance test (mean values [H] and area under the curve [I]), and an insulin tolerance test(mean values for blood glucose [J] and constant for glucose decay [K]). In all experiments, n = 6; *P < 0.05 vs. HFD; §P < 0.05 vs.conditions as depicted in the panels. AU, arbitrary units; AUC, area under the curve; GTT, glucose tolerance test; ITT, insulin tolerance test;KITT, constant for glucose decay during the insulin tolerance test; v3, n-3; w, weeks.

diabetes.diabetesjournals.org Nascimento and Associates 675

was used as the endogenous control. Each PCR contained40 ng reverse-transcribed RNA, 25 mL of each specificprimer, Taqman Universal master mix no. 4369016, andRNase-free water to a 10 mL of final volume.

Histology and Cell CountingThe mice were perfused with 4% paraformaldehyde. Thebrains were immersed in 30% sucrose (wt/vol), embeddedin OCT compound (Sakura Finetek, Torrance, CA), and cutinto 12-mm coronal sections using a cryostat. Optimalthickness of the sections was defined after extensive test-ing of the method and according to previous publications(22–24). The sections were washed in PBS and incubatedwith 2.0 N HCl for 10 min at 37°C followed by 0.1 mol/Lsodium borate for 10 min at room temperature. Thereafter,the sections were incubated with a 5% goat serum (no. G9023;Sigma-Aldrich) blocking solution in Tris-phosphate–bufferedsaline (0.2% of Triton X-100) for 30 min at 37°C followedby an overnight incubation with 1:100 primary antibodiesagainst BrdU, neuropeptide Y (NPY), proopiomelanocortin(POMC) (nos. sc32323, sc28943, and sc20148; Santa CruzBiotechnology), neuronal nuclei (NeuN) (no. MAB377C3;Millipore, Temecula, CA), and nestin (no. ab93666; Abcam,Cambridge, U.K.) diluted in the same blocking solution.Then, the sections were incubated with secondary antibodiesconjugated to fluorescein isothiocyanate or rhodamine(nos. sc2777 and sc2092; Santa Cruz Biotechnology) andcovered and mounted with Vectashield Mounting Mediumwith DAPI (no. H-1200; Vector Laboratories, Burlingame,CA). The images were acquired by a confocal laser micro-scope (LSM780; Zeiss, Jena, Germany), using oil immer-sion 403 Plan Apochromatic objective. The tiling functionwas used to cover the whole section, and each image wascollected at 1,0123 1,012 pixels. Three consecutive sectionsfor every three animals were used for quantification. PositiveBrdU cells were considered only when colocalized with DAPIstaining to indicate cell proliferation. In addition, positiveBrdU cells colocalized with NeuN staining were used to in-dicate neurogenesis. For normalization of sections per size,

the results were expressed as the total number of cells givenby DAPI staining. Positive stains were counted by a blindedobserver and automatically using ImageJ (http://rsbweb.nih.gov/ij/), which produced equivalent results. Moreover,ImageJ was used to calculate the colocalization of thedouble stain using the JACoP plugin and the numbersdetermined by Manders coefficient.

Statistical AnalysisThe results are presented as means6 SE. After evaluationof the distribution of all the data, the results were ana-lyzed by Student t test or by one-way ANOVA followed byTukey test or two-way ANOVA followed by the Holm-Sidakmethod to determine the significance of the individual dif-ferences. The level of significance was set at P , 0.05, andthe data were analyzed using the Sigma Stat 3.1 (SystatSoftware, Point Richmond, CA).

RESULTS

Dietary n-3 PUFA Reduces Body Mass and ImprovesMetabolic ParametersSwiss mice were fed an HFD for 8 weeks and then randomlydivided into four groups that for another 8 weeks were fedthe same HFD or an HFD with partial (10% or 20%) or total(30%) substitution of the predominantly saturated fatcontent by flaxseed oil containing ;45% C18:3 (n-3). Thelipid composition of the flaxseed oil was determined by gaschromatography (Fig. 1A and Table 1). The experimentalprotocol is depicted in Fig. 1B. The n-3 substitution resultedin reduced body mass gain (Fig. 1C), increased caloric intake(Fig. 1D), improved responsiveness to leptin (Fig. 1E),

Table 1—Fatty acid composition of the flaxseed oil asanalyzed by gas chromatography

Fatty acid

C16:0 6.34

C16:1 0.10

C17:0 0.07

C18:0 6.66

C18:1 (n-9) 24.05

C18:2 (n-6) 15.36

C18:3 (n-3) 46.64

C20:0 0.21

C20:1 0.11

C22:0 0.21

C24:0 0.13

Data are percent.

Table 2—Fatty acid composition of the hypothalamus asanalyzed by gas chromatography (% of total fat)

Fatty acid NFD HFD HFD n-3

C16:0 22.87 6 1.37 22.30 6 1.30 24.44 6 1.33

C18:0 21.41 6 0.55 20.63 6 0.20 21.41 6 0.93

C22:0 0.24 6 0.04 0.28 6 0.06* 0.23 6 0.13#

SSFA 44.91 6 1.92 44.30 6 1.14 44.66 6 1.12

C16:1 0.80 6 0.03 0.75 6 0.03 0.82 6 0.04

C18:1 (n-9) 25.01 6 1.66 25.20 6 0.55 24.66 6 1.23

SMUFA 26.55 6 0.16 29.44 6 0.45* 26.79 6 0.32#

C18:2 (n-6) 1.93 6 0.77 2.44 6 0.11 1.76 6 0.09

C18:3 (n-3) 0.06 6 0.02 0.08 6 0.02 0.22 6 0.02*#

C20:4 (n-6) 10.17 6 0.80 9.77 6 0.60 10.29 6 0.64

C20:5 (n-3) 0.23 6 0.06 0.31 6 0.03* 0.20 6 0.04#

C22:5 (n-3) 0.28 6 0.17 0.15 6 0.15* 0.23 6 0.12#

C22:6 (n-3) 15.98 6 0.77 15.32 6 0.58 16.93 6 0.54

SPUFA 28.55 6 0.94 25.35 6 0.55* 28.42 6 1.21#

SSFA includes C8:0, C10:0, C12:0, C14:0, C16:0, C18:0, C20:0and C22:0. SMUFA includes C14:1, C16:1, C17:1,C18:1, C20:1and C24:1. SPUFA includes C18:2, C18:3, C20:2, C20:3, C20:4,C20:5, C22:5, C55:6. In all measurements, n = 5. HFD n-3, HFDwith 20% substitution by flaxseed oil; SFA, saturated fatty acid;MUFA monounsaturated fatty acid; PUFA, polyunsaturated fattyacid. *P , 0.05 vs. NFD; #P , 0.05 vs. HFD.

676 Nutrient-Induced Hypothalamic Neurogenesis Diabetes Volume 65, March 2016

Figure 2—Dietary PUFAs induce hypothalamic cell proliferation. Swiss mice were submitted to one of the dietary approaches as depictedin A. At the end of the experimental period, hypothalamic samples were used to determine the expression of doublecortin transcript (B) andprotein (C ). Bax (D) and Bcl-2 (E ) protein levels were also determined and the ratio of Bax to Bcl-2 was obtained (F ). G–I: Representativeimages obtained from the immunofluorescence evaluation of BrdU localization studies in the hypothalamus. G: Representative image ofthe hypothalamus obtained from a Swiss mouse fed chow and illustrating the regions on which BrdU-positive cells were counted.J: Relative number of BrdU-positive cells in the hypothalamus compared with mice fed chow for 16 weeks. K: Representative imagesobtained from the immunofluorescence evaluation of BrdU and NeuN localization studies in the hippocampus. I: Relative number of

diabetes.diabetesjournals.org Nascimento and Associates 677

increased spontaneous activity (Fig. 1F), reduced levels offasting blood glucose (Fig. 1G), reduced area under thecurve during a glucose tolerance test (Fig. 1H and I),and increased responsiveness to insulin during an insulintolerance test (Fig. 1J and K). In addition, the consumptionof an HFD with 20% substitution of the predominantlysaturated fat by flaxseed oil resulted in increased incorpora-tion of C18:3, C20:5, and C22:5 in the hypothalamus, result-ing in a significant increase of total PUFA content comparedwith mice fed an HFD only (Table 2).

Dietary n-3 PUFA Increases Hypothalamic CellProliferationIn the preceding experiment, the substitution of dietaryfat by 20% flaxseed oil presented the most consistentresults regarding the improvement of metabolic parame-ters; therefore, we next evaluated Swiss mice fed an HFDinitially for 8 weeks and then randomly divided into fourgroups that for another 8 weeks were fed an NFD (chow)containing 2.3% fat, an NFD supplemented with 20%flaxseed oil, an HFD, or an HFD supplemented with 20%flaxseed oil (Fig. 2A). The supplementation of flaxseed oilin the HFD resulted in the increased expression of hy-pothalamic doublecortin (Fig. 2B and C), a reduced ex-pression of hypothalamic Bax (Fig. 2D), and a reducedBax-to-Bcl2 ratio (Fig. 2E and F). Moreover, the flax-seed oil supplementation resulted in increased numbersof arcuate and paraventricular nuclei hypothalamic cellsundergoing proliferation in mice fed the HFD (Fig. 2G–J).This was a region-specific effect, since no proliferation couldbe detected in other regions of the hypothalamus (not shown)and only a minor (not significant) increase of neuronal pro-liferation was detected in the hippocampus (Fig. 2K and L).

Intracerebroventricular DHA Increases HypothalamicNeurogenesisBecause the increased dietary content of PUFA was capableof inducing hypothalamic cell proliferation, we decidedto evaluate the impact of intracerebroventricular DHA toinduce neurogenesis. Obese Swiss mice were treatedintracerebroventricularly with DHA or BDNF for 10 daysas shown in Fig. 3A. Despite the fact that it was a short-term treatment, DHA was capable of significantly im-proving insulin sensitivity as determined by an insulintolerance test (Fig. 3B and C). However, there were nosignificant changes in caloric intake and body mass (notshown). Both BDNF and DHA induced cell proliferationand neurogenesis in the hypothalamus (Fig. 3D–F).Interestingly, DHA exerted a more pronounced neurogeniceffect than BDNF, although the difference between the twotreatments was not significant (Fig. 3F). Another interest-ing feature of the treatment with DHA was the presence of

BrdU-positive cells originating from the lateral wall of thethird ventricle (Fig. 3D [arrows]); this effect was virtuallyabsent in the mice treated with BDNF. We evaluated thecapacity of intracerebroventricularly BDNF and DHA toinduce neurogenesis in other regions of the hypothalamusand brain. In the hypothalamus, no signs of neurogenesiswere detected in the preoptic and lateral regions. However,intracerebroventricular DHA induced neurogenesis in thehippocampus (Fig. 3G–I). In the hypothalamus, the ori-gin of DHA-stimulated new cells from the lateral wall ofthe third ventricle was strongly evident when stainingfor nestin (Fig. 4A).

Intracerebroventricular DHA Increases Neurogenesisof POMC but Not NPY NeuronsUsing the same experimental approach as described inFig. 3A, we evaluated the coexpression of BrdU withPOMC or NPY neurons. As depicted in Fig. 4B–D, DHAinduced a significant increase in POMC neurogenesis butnot in NPY neurogenesis.

DHA Induces the Increased Expression of BDNF andGPR40 in the HypothalamusNext, we tested whether n-3 PUFA could induce theincreased expression of GPR40. To this end, we firstevaluated the hypothalami of mice fed according tothe same protocol as shown in Fig. 2A. As depicted inFig. 5A and B, the increased dietary content of fat,irrespective of type, induced the expression of both BDNFand GPR40. There were no differences in either proteinexpression when comparing mice fed the HFD with micefed the HFD supplemented 20% with flaxseed oil. Toevaluate whether the effect of the dietary fats was par-tially due to n-3 PUFA, we treated obese mice with DHAinjected intraperitoneally for 10 days according to theprotocol shown in Fig. 5C. As depicted in Fig. 5D and E,systemic DHA did not induce an increased expressionof BDNF but did increase the hypothalamic expressionof GPR40 in a dose-dependent fashion.

Inhibition of BDNF Reduces the Effect of Dietary n-3PUFA to Increase Hypothalamic NeurogenesisTo evaluate the role of BDNF as a candidate mediator of theeffects of n-3 PUFAs as inducers of hypothalamic neuro-genesis, obese mice were fed an NFD or an HFD with 20%flaxseed oil for 8 weeks and treated (for 8 weeks) withpreimmune serum or anti-BDNF antiserum (Fig. 6A). Asdepicted in Fig. 6B, the use of the anti-BDNF antiserumresulted in a maximum reduction of 40% of hypothalamicBDNF expression. The global count of BrdU-positive cellswas reduced by 30% in the hypothalamus of mice fedcontinuously on an HFD and treated with the anti-BDNFantiserum (Fig. 6E). However, when only BrdU/NeuN

BrdU-positive cells in the hippocampus compared with mice fed chow for 16 weeks. In all experiments, n = 6. *P < 0.05 vs. respectivecontrol on nonsubstituted diet; §P < 0.05 vs. conditions as depicted in the panels. In G–I and K, nuclei were counterstained with DAPI(blue) and BrdU was labeled with fluorescein (green). In K, NeuN is labeled in red. DCX, doublecortin; IB, immunoblot; v3, n-3; PVN,paraventricular; RQ, relative quantification; w, weeks.

678 Nutrient-Induced Hypothalamic Neurogenesis Diabetes Volume 65, March 2016

Figure 3—Intracerebroventricular DHA induces hypothalamic neurogenesis. Swiss mice were submitted to one of the approaches depicted in A.At the end of the experimental period, mice were submitted to an insulin tolerance test and the blood glucose levels were determined (B) andused to calculate the constant for glucose decay (C). D: Representative images obtained from the immunofluorescence evaluation of NeuN(rhodamine [red]) and BrdU (fluorescein [green]) colocalization studies in the hypothalamus. The yellow arrows indicate DHA-induced neuro-genesis in the wall of the third ventricle. E: Quantitative evaluation of BrdU-positive cells in the hypothalamus. F: Quantitative evaluation of NeuNand BrdU double-positive cells in the hypothalamus. G: Representative images obtained from the immunofluorescence evaluation of NeuN(rhodamine [red]) and BrdU (fluorescein [green]) colocalization studies in the hippocampus. H: Quantitative evaluation of BrdU-positive cells in thehippocampus. I: Quantitative evaluation of NeuN and BrdU double-positive cells in the hippocampus. In all experiments, n = 6. *P < 0.05 vs.

diabetes.diabetesjournals.org Nascimento and Associates 679

double-positive cells were considered, the anti-BDNF treat-ment was capable of inhibiting neurogenesis only in micefed on the NFD (Fig. 6F). The intracerebroventricular in-jection of anti-BDNF antiserum had no effect on n-3–induced hippocampal neurogenesis (not shown).

Chemical Inhibition of GPR40 Reduces the Effectof Dietary n-3 PUFA to Increase HypothalamicNeurogenesisGPR40 is one of the receptors for n-3 PUFAs and in brainregions other than the hypothalamus it has been relatedto neurogenesis (25). For evaluation of its potential rolein hypothalamic neurogenesis in response to an n-3 PUFA,obese mice were treated with DHA in the presence or notof a chemical inhibitor of GPR40, GW1100, as depicted inFig. 7A. As expected, DHA significantly increased hypotha-lamic cell proliferation (Fig. 7B) and neurogenesis (Fig. 7C),effects that were completely abolished by GW1100 (Fig.7B–E). The GW1100 had no effect on DHA-induced hippo-campal neurogenesis (not shown).

DISCUSSION

One of the most important problems during the treatmentof obesity is the high rate of recurrence that follows aninitial period of body mass reduction (26–28). Even patientswho undergo bariatric surgery experience body mass regainafter some time (29). During the development of obesity, acontinuous and progressive resetting of the hypothalamicadipostat is thought to take place, which is partially dueto the development of hypothalamic resistance to leptinand insulin (30,31). At first, adipostatic hormone resistancewas regarded as a molecular phenomenon; however, re-cent studies have shown that hypothalamic neuronsmay undergo apoptosis during diet-induced obesity(7,8), which provides a cellular basis for the loss ofthe coordinated control of caloric intake and energyexpenditure.

Apparently, hypothalamic neurons involved in thecontrol of body mass are differentially affected by obe-sity. In at least three studies, preferential reductions ofPOMC compared with NPY/AgRP neurons occurred in thehypothalami of mice with hypothalamic inflammationtriggered by increased fat intake or induced by geneticapproaches (7,8,14). Thus, the relative numbers of POMCneurons in the medium-basal hypothalamus appear toplay an important role in the control of body adiposity.Accordingly, a recent study provided a major advance inthe field using a reversible rodent model of obesity inwhich POMC expression was blocked in neurons of thehypothalamus. POMC reactivation during early obesitycaused a complete reversal of the phenotype; however,late reactivation was insufficient to promote the normali-zation of the body weight (32).

If the high rate of recurrence associated with obesity ispartially a result of the loss of certain hypothalamicneuronal populations, a definitive therapeutic solution forobesity may only be achieved by the restoration of thecorrect neural circuitries. In this context, stimulating theneurogenesis of hypothalamic neurons emerges as a poten-tial mechanism to be explored for the development of moreefficient approaches to treat obesity.

BDNF was the first factor shown to promote neuro-genesis of hypothalamic neurons (33); however, hypotha-lamic neurogenesis associated with changes in whole-bodyenergy homeostasis was first demonstrated in response toCNTF (11). The study was designed to investigate the mech-anisms behind the clinical and experimental evidence forsustained body mass reduction after the transient useof CNTF or its analog, Axokine (34,35). CNTF promotedthe neurogenesis of leptin-responsive neurons in the hypo-thalamus, and the sustained action of CNTF, which could lastfor weeks after the interruption of its use, was proposed tobe due to the rewiring of the hypothalamic circuitries in-volved in whole-body energy homeostasis (11). In additionto BDNF and CNTF, other biological factors and drugs induceadult hypothalamic neurogenesis, such as estrogens (36),IGF-1 (37), ethanol (38), nicotine (39), and fluoxetine (40).

In addition to neurogenesis induced by different typesof stimuli, the hypothalamus also presents constitutiveneurogenesis (41), similar to the subventricular zone of thelateral ventricles and the subgranular zone of the dentategyrus (42,43), which places this region among the few ana-tomical sites capable of continuously producing new neuronsduring adult life (43). Two recent studies have shown thattanycytes of the median eminence are the source of at leastsome of the newborn hypothalamic cells, and, interestingly,these cells are responsive to dietary factors (44,45).

An important aspect of diet-induced hypothalamicneurogenesis involves the simultaneous induction of theapoptosis of neurons and neurogenesis in animals fed anHFD (7,13,14); however, in opposition to an apparentlybeneficial compensation for apoptotic loss, the neurogen-esis induced by an HFD results in increased production ofNPY/AgRP neurons (13), therefore widening the imbal-ance between orexigenic and anorexigenic neuronal pop-ulations. A recent study has probed this issue by showingthat the defect in neurogenesis associated with the con-sumption of an HFD is due to the activation of inflam-mation, specifically through inhibitor of kB kinase (14).

In the current study, we evaluated whether n-3 PUFAsare capable of inducing hypothalamic neurogenesis. Recentstudies have unveiled new anti-inflammatory mechanismslinked to the beneficial effects of PUFAs in metabolicconditions (46,47). Defective neurogenesis generated by anHFD is due to the activation of hypothalamic inflammation

saline. In D and G, nuclei were counterstained with DAPI (blue). Scale white bars = 100 mm. ARC, arcuate; ITT, insulin tolerance test; KITT,constant for glucose decay during the insulin tolerance test; PVN, paraventricular; w, weeks.

680 Nutrient-Induced Hypothalamic Neurogenesis Diabetes Volume 65, March 2016

Figure 4—Intracerebroventricular DHA induces neurogenesis in the wall of the third ventricle and preferential proliferation of POMC neurons. TheSwiss mice were submitted to one of the dietary approaches as depicted in Fig. 3A. A: Representative images obtained from the immunoflu-orescence evaluation of nestin and BrdU colocalization studies in the hypothalamus. The red arrows indicate DHA-induced neurogenesis in thewall of the third ventricle. B: Representative images obtained from the immunofluorescence evaluation of NPY and BrdU (left-hand column) andPOMC and BrdU (right-hand column) colocalization studies in the arcuate nucleus of the hypothalamus. C: Quantitative evaluation of NPY andBrdU double-positive cells in the hypothalamus. D: Quantitative evaluation of POMC and BrdU double-positive cells in the hypothalamus. In allexperiments, n = 6. In A, nestin is stained with rhodamine (red) and BrdU is stained with fluorescein (green). Scale white bars = 200 mm. In A andB, nuclei are counterstained with DAPI (blue). In B, scale white bars = 100 mm. ARC, arcuate; PVN, paraventricular.

diabetes.diabetesjournals.org Nascimento and Associates 681

(14); therefore, we reasoned that the anti-inflammatoryproperties of PUFAs could correct the defect. In fact, in arecent study we showed that PUFAs in the diet and in-jected directly in the hypothalamus reduce HFD-inducedinflammation, resulting in body mass reduction (20).

In the first part of this study, we show that either whenpresent in the diet or acting directly in the hypothalamus,n-3 PUFAs increase neurogenesis in the hypothalamus. Thiseffect is accompanied by the reduction of apoptosis markers,increased responsiveness to leptin, and reduced body massgain as previously reported (20). Moreover, the consumptiona diet rich in n-3 PUFAs changes the fatty acid compositionof the hypothalamus, increasing the total content of PUFAs.

When injected directly into the hypothalamus, DHAis capable of inducing more pronounced neurogenesisthan the level induced by BDNF, which is an importantoutcome considering that, in parallel with CNTF, BDNFis one of the most potent inducers of neurogenesis inthe brain, particularly in the hypothalamus (48,49).

Interestingly, DHA produced a neurogenic response inthe hypothalamus that affected the numbers and the dis-tribution and subtypes of cells differently than BDNF. Incontrast to previous reports concerning the neurogen-esis induced by HFD, which induces the generation ofnewborn cells particularly from the tanycytes of themedium eminence (44), DHA but not BDNF inducedthe generation of new cells from the wall of the thirdventricle. Additionally, unlike BDNF and, to our knowl-edge, all previously reported factors capable of inducingneurogenesis in the hypothalamus (11,13,14,36–40),DHA induced the preferential neurogenesis of POMCneurons.

Owing to the well-known role of BDNF as an endogenousinducer of neurogenesis and because dietary PUFAs werecapable of increasing BDNF, we hypothesized that BDNFcould be the mediator of the neurogenic activity of PUFAs.To test this hypothesis, we immunoneutralized hypotha-lamic BDNF using an antiserum. Our approach resulted in

Figure 5—DHA increases hypothalamic GPR40. Swiss mice were submitted to one of the dietary approaches as depicted in Fig. 2A, exceptthat no BrdU was administered. At the end of the experimental period, hypothalamus samples were obtained and used in immunoblottingexperiments to determine the expression of BDNF (A) and GPR40 (B). In another set of experiments, the obese Swiss mice (fed for 8 weekson an HFD) were treated for 10 days with intraperitoneal injections of DHA according to the protocol as depicted in C. At the end ofthe experimental period, the hypothalamus was obtained to measure the transcript expression of BDNF (D) and GPR40 (E). In all experiments,n = 6. In A and B, *P < 0.05 vs. NFD. In D and E, *P < 0.05 vs. saline. IB, immunoblot; v3, n-3; RQ, relative quantification; w, weeks.

682 Nutrient-Induced Hypothalamic Neurogenesis Diabetes Volume 65, March 2016

Figure 6—The inhibition of BDNF reduces hypothalamic cell proliferation induced by dietary PUFAs. Swiss mice were submitted to one ofthe dietary approaches as depicted in A. At the end of the experimental period, the samples of the hypothalamus were obtained and used inimmunoblot experiments to determine the expression of BDNF (B). C and D: Representative images obtained from the immunofluorescenceevaluation of NeuN and BrdU colocalization studies in the arcuate nucleus and paraventricular nucleus, respectively. E: Quantitativeevaluation of BrdU-positive cells in the hypothalamus. F: Quantitative evaluation of NeuN and BrdU double-positive cells in the hypothal-amus. In all experiments, n = 6. *P < 0.05 vs. respective control treated with preimmune serum only. In C and D, BrdU is labeled withfluorescein (green) and NeuN is labeled with rhodamine (red); nuclei are counterstained with DAPI (blue). Scale white bars = 100 mm. ARC,arcuate; IB, immunoblot; PVN, paraventricular; v3, n-3; w, weeks.

diabetes.diabetesjournals.org Nascimento and Associates 683

Figure 7—GPR40 chemical inhibition dampens DHA-induced neurogenesis in the hypothalamus. Swiss mice were submitted to one of theapproaches depicted in A. B: Quantitative evaluation of BrdU-positive cells in the hypothalamus. C: Quantitative evaluation of NeuN and BrdUdouble-positive cells in the hypothalamus. D and E: Representative images obtained from the immunofluorescence evaluation of NeuN and BrdUcolocalization studies in the arcuate nucleus and paraventricular nucleus, respectively. In all experiments, n = 6. *P < 0.05 vs. respective controltreated with saline only; §P< 0.05 vs. conditions as depicted in the panels. InD and E, BrdU is labeled with fluorescein (green) and NeuN is labeledwith rhodamine (red); nuclei are counterstained with DAPI (blue). Scale white bars = 100 mm. ARC, arcuate; GW1100, chemical inhibitor of GPR40;PVN, paraventricular; w, weeks.

684 Nutrient-Induced Hypothalamic Neurogenesis Diabetes Volume 65, March 2016

up to 40% reduction of hypothalamic BDNF. This wasaccompanied by a global reduction of cell proliferation.However, when counting new neurons, BDNF inhibitionwas capable of reducing the effect of PUFAs in mice fed onthe NFD only.

In the final part of the study, we asked whether GPR40could act as the mediator of the neurogenic effects of n-3PUFAs. GPR40 is a class A G-protein–coupled receptorexpressed at the highest levels in the pancreas and severalregions of the brain, including the hypothalamus (50,51).Studies have shown that free fatty acids in the brain couldact through GPR40 to control memory-stimulating pro-genitor cell proliferation (52). However, no previous stud-ies have explored the potential neurogenic actions ofGPR40 in the hypothalamus. Therefore, we targeted hypo-thalamic GPR40 using a chemical inhibitor, GW1100. This isa selective inhibitor with almost no cross-reactivity withGPR120 (53). Upon inhibition of GPR40, we obtained acomplete reversal of the neurogenic effect of DHA, suggest-ing that most of the effect of n-3 PUFAs in inducingneurogenesis depends on this receptor.

One important aspect of the neurogenic effect of n-3PUFAs in the brain is its apparent anatomic specificity.We evaluated other regions of the hypothalamus and thebrain, and the hippocampus was the only other site present-ing some neurogenic activity in response to n-3 PUFAs. Thereare previous reports describing the neurogenic potential ofPUFAs in the hippocampus (54). In fact, at least part of thebeneficial effects of PUFAs in age-related brain disorders havebeen attributed to their capacity to induce neurogenenesis inthe hippocampus (55,56). However, in the current study,the inhibitions of either BDNF or GPR40 were capableof mitigating PUFA-induced neurogenesis in the hypo-thalamus only.

In conclusion, dietary PUFAs are capable of inducinghypothalamic neurogenesis. Most of the effect is mediatedby GPR40, and BDNF could be at least one of the effectorsof this process. Importantly, the effect appears to be directedtoward the POMC neuronal population, providing a dietaryapproach to potentially correct the imbalance in hypo-thalamic neuronal subpopulations, which is a hallmarkof experimental obesity.

Acknowledgments. The authors thank Erika Roman, Gerson Ferraz, andMarcio Cruz, from the University of Campinas, for technical assistance.Funding. Support for the study was provided by Fundação de Amparo aPesquisa do Estado de São Paulo and Conselho Nacional de DesenvolvimentoCientifico e Tecnologico. The Laboratory of Cell Signaling belongs to the Obesityand Comorbidities Research Center and National Institute of Science andTechnology–Diabetes and Obesity. H.F.C. belongs to the National Instituteof Science and Technology–Photonics Applied to Cell Biology.Duality of Interest. No potential conflicts of interest relevant to this articlewere reported.Author Contributions. L.F.R.N. performed most of the experiments,which are part of his PhD thesis. Some of the experiments were performed undersupervision of other researchers; G.F.P.S., J.M., G.O.B., C.S., R.F.M., S.C.V., L.M.I.-S.,and D.S.R. performed some of the experiments. H.F.C. provided the expertise in the

microscopy experiments and helped write the paper. L.A.V. is the mentor of thestudy, planned most experiments, and wrote the paper. L.A.V. is the guarantor of thiswork and, as such, had full access to all the data in the study and takes responsibilityfor the integrity of the data and the accuracy of the data analysis.

References1. Moussavi N, Gavino V, Receveur O. Could the quality of dietary fat, and not

just its quantity, be related to risk of obesity? Obesity (Silver Spring) 2008;16:7–152. Goris AH, Westerterp KR. Physical activity, fat intake and body fat. Physiol

Behav 2008;94:164–1683. Zhang X, Zhang G, Zhang H, Karin M, Bai H, Cai D. Hypothalamic IKKbeta/

NF-kappaB and ER stress link overnutrition to energy imbalance and obesity. Cell

2008;135:61–734. Milanski M, Degasperi G, Coope A, et al. Saturated fatty acids produce an

inflammatory response predominantly through the activation of TLR4 signaling in

hypothalamus: implications for the pathogenesis of obesity. J Neurosci 2009;29:

359–3705. De Souza CT, Araujo EP, Bordin S, et al. Consumption of a fat-rich diet

activates a proinflammatory response and induces insulin resistance in the

hypothalamus. Endocrinology 2005;146:4192–41996. El-Haschimi K, Pierroz DD, Hileman SM, Bjørbaek C, Flier JS. Two defects

contribute to hypothalamic leptin resistance in mice with diet-induced obesity.

J Clin Invest 2000;105:1827–18327. Moraes JC, Coope A, Morari J, et al. High-fat diet induces apoptosis of

hypothalamic neurons. PLoS One 2009;4:e50458. Thaler JP, Yi CX, Schur EA, et al. Obesity is associated with hypothalamic

injury in rodents and humans. J Clin Invest 2012;122:153–1629. van de Sande-Lee S, Pereira FR, Cintra DE, et al. Partial reversibility of

hypothalamic dysfunction and changes in brain activity after body mass reduction

in obese subjects. Diabetes 2011;60:1699–170410. Dietrich MO, Horvath TL. Feeding signals and brain circuitry. Eur J Neurosci

2009;30:1688–169611. Kokoeva MV, Yin H, Flier JS. Neurogenesis in the hypothalamus of adult

mice: potential role in energy balance. Science 2005;310:679–68312. Pinto S, Roseberry AG, Liu H, et al. Rapid rewiring of arcuate nucleus

feeding circuits by leptin. Science 2004;304:110–11513. McNay DE, Briançon N, Kokoeva MV, Maratos-Flier E, Flier JS. Remodeling

of the arcuate nucleus energy-balance circuit is inhibited in obese mice. J Clin

Invest 2012;122:142–15214. Li J, Tang Y, Cai D. IKKb/NF-kB disrupts adult hypothalamic neural stem cells

to mediate a neurodegenerative mechanism of dietary obesity and pre-diabetes.

Nat Cell Biol 2012;14:999–101215. Knoops KT, de Groot LC, Kromhout D, et al. Mediterranean diet, lifestyle

factors, and 10-year mortality in elderly European men and women: the HALE

project. JAMA 2004;292:1433–143916. Estruch R, Ros E, Salas-Salvadó J, et al.; PREDIMED Study Investigators.

Primary prevention of cardiovascular disease with a Mediterranean diet. N Engl J

Med 2013;368:1279–129017. Bjerregaard P, Mulvad G, Pedersen HS. Cardiovascular risk factors in Inuit of

Greenland. Int J Epidemiol 1997;26:1182–119018. Fischer S, Weber PC. Prostaglandin I3 is formed in vivo in man after dietary

eicosapentaenoic acid. Nature 1984;307:165–16819. Fowler KH, Chapkin RS, McMurray DN. Effects of purified dietary n-3

ethyl esters on murine T lymphocyte function. J Immunol 1993;151:5186–519720. Cintra DE, Ropelle ER, Moraes JC, et al. Unsaturated fatty acids revert

diet-induced hypothalamic inflammation in obesity. PLoS One 2012;7:

e3057121. Hartman L, Lago RC. Rapid preparation of fatty acid methyl esters from

lipids. Lab Pract 1973;22:475–476, passim22. Kempermann G, Kuhn HG, Gage FH. Experience-induced neurogenesis in

the senescent dentate gyrus. J Neurosci 1998;18:3206–3212

diabetes.diabetesjournals.org Nascimento and Associates 685

23. Antle MC, LeSauter J, Silver R. Neurogenesis and ontogeny of specific cellphenotypes within the hamster suprachiasmatic nucleus. Brain Res Dev BrainRes 2005;157:8–1824. Wojtowicz JM, Kee N. BrdU assay for neurogenesis in rodents. Nat Protoc2006;1:1399–140525. Boneva NB, Yamashima T. New insights into “GPR40-CREB interaction inadult neurogenesis” specific for primates. Hippocampus 2012;22:896–90526. Haber S. Effective treatment of obesity produces remission, not cure. Int JObes 1980;4:265–26727. Goodrick GK, Foreyt JP. Why treatments for obesity don’t last. J Am DietAssoc 1991;91:1243–124728. Kraschnewski JL, Boan J, Esposito J, et al. Long-term weight loss main-tenance in the United States. Int J Obes 2010;34:1644–165429. Kalarchian MA, Marcus MD, Courcoulas AP, Cheng Y, Levine MD, JosbenoD. Optimizing long-term weight control after bariatric surgery: a pilot study. SurgObes Relat Dis 2012;8:710–71530. Münzberg H, Myers MG Jr. Molecular and anatomical determinants ofcentral leptin resistance. Nat Neurosci 2005;8:566–57031. Morton GJ, Cummings DE, Baskin DG, Barsh GS, Schwartz MW. Centralnervous system control of food intake and body weight. Nature 2006;443:289–29532. Bumaschny VF, Yamashita M, Casas-Cordero R, et al. Obesity-programmedmice are rescued by early genetic intervention. J Clin Invest 2012;122:4203–421233. Pencea V, Bingaman KD, Wiegand SJ, Luskin MB. Infusion of brain-derivedneurotrophic factor into the lateral ventricle of the adult rat leads to new neuronsin the parenchyma of the striatum, septum, thalamus, and hypothalamus. J Neurosci2001;21:6706–671734. Ettinger MP, Littlejohn TW, Schwartz SL, et al. Recombinant variant of ciliaryneurotrophic factor for weight loss in obese adults: a randomized, dose-rangingstudy. JAMA 2003;289:1826–183235. Lambert PD, Anderson KD, Sleeman MW, et al. Ciliary neurotrophic factoractivates leptin-like pathways and reduces body fat, without cachexia or reboundweight gain, even in leptin-resistant obesity. Proc Natl Acad Sci U S A 2001;98:4652–465736. Fowler CD, Liu Y, Wang Z. Estrogen and adult neurogenesis in the amygdalaand hypothalamus. Brain Res Brain Res Rev 2008;57:342–35137. Pérez-Martín M, Cifuentes M, Grondona JM, et al. IGF-I stimulates neurogenesisin the hypothalamus of adult rats. Eur J Neurosci 2010;31:1533–154838. Chang GQ, Karatayev O, Liang SC, Barson JR, Leibowitz SF. Prenatal ethanolexposure stimulates neurogenesis in hypothalamic and limbic peptide systems:possible mechanism for offspring ethanol overconsumption. Neuroscience 2012;222:417–42839. Chang GQ, Karatayev O, Leibowitz SF. Prenatal exposure to nicotine stim-ulates neurogenesis of orexigenic peptide-expressing neurons in hypothalamusand amygdala. J Neurosci 2013;33:13600–13611

40. David DJ, Samuels BA, Rainer Q, et al. Neurogenesis-dependent and-independent effects of fluoxetine in an animal model of anxiety/depression.Neuron 2009;62:479–49341. Kokoeva MV, Yin H, Flier JS. Evidence for constitutive neural cell pro-liferation in the adult murine hypothalamus. J Comp Neurol 2007;505:209–22042. Emsley JG, Mitchell BD, Kempermann G, Macklis JD. Adult neurogenesisand repair of the adult CNS with neural progenitors, precursors, and stem cells.Prog Neurobiol 2005;75:321–34143. Ming GL, Song H. Adult neurogenesis in the mammalian central nervoussystem. Annu Rev Neurosci 2005;28:223–25044. Lee DA, Bedont JL, Pak T, et al. Tanycytes of the hypothalamic medianeminence form a diet-responsive neurogenic niche. Nat Neurosci 2012;15:700–70245. Haan N, Goodman T, Najdi-Samiei A, et al. Fgf10-expressing tanycytes addnew neurons to the appetite/energy-balance regulating centers of the postnataland adult hypothalamus. J Neurosci 2013;33:6170–618046. Oh DY, Talukdar S, Bae EJ, et al. GPR120 is an omega-3 fatty acid receptormediating potent anti-inflammatory and insulin-sensitizing effects. Cell 2010;142:687–69847. Li P, Spann NJ, Kaikkonen MU, et al. NCoR repression of LXRs restrictsmacrophage biosynthesis of insulin-sensitizing omega 3 fatty acids. Cell 2013;155:200–21448. Noble EE, Billington CJ, Kotz CM, Wang C. The lighter side of BDNF. Am JPhysiol Regul Integr Comp Physiol 2011;300:R1053–R106949. Sousa-Ferreira L, de Almeida LP, Cavadas C. Role of hypothalamicneurogenesis in feeding regulation. Trends Endocrinol Metab 2014;25:80–8850. Brown AJ, Jupe S, Briscoe CP. A family of fatty acid binding receptors. DNACell Biol 2005;24:54–6151. Ma D, Tao B, Warashina S, et al. Expression of free fatty acid receptorGPR40 in the central nervous system of adult monkeys. Neurosci Res 2007;58:394–40152. Yamashima T. A putative link of PUFA, GPR40 and adult-born hippocampalneurons for memory. Prog Neurobiol 2008;84:105–11553. Briscoe CP, Peat AJ, McKeown SC, et al. Pharmacological regulation ofinsulin secretion in MIN6 cells through the fatty acid receptor GPR40: identifi-cation of agonist and antagonist small molecules. Br J Pharmacol 2006;148:619–62854. He C, Qu X, Cui L, Wang J, Kang JX. Improved spatial learning performanceof fat-1 mice is associated with enhanced neurogenesis and neuritogenesis bydocosahexaenoic acid. Proc Natl Acad Sci U S A 2009;106:11370–1137555. Janssen CI, Zerbi V, Mutsaers MP, et al. Impact of dietary n-3 polyunsaturatedfatty acids on cognition, motor skills and hippocampal neurogenesis in de-veloping C57BL/6J mice. J Nutr Biochem 2015;26:24–3556. Denis I, Potier B, Vancassel S, Heberden C, Lavialle M. Omega-3 fatty acidsand brain resistance to ageing and stress: body of evidence and possiblemechanisms. Ageing Res Rev 2013;12:579–594

686 Nutrient-Induced Hypothalamic Neurogenesis Diabetes Volume 65, March 2016