molecular mechanisms associated with leptin resistance: n-3 polyunsaturated fatty acids induce...

A R T I C L E

Molecular mechanisms associated with leptin resistance: n-3polyunsaturated fatty acids induce alterations in the tightjunction of the brain

Shinsuke Oh-I,1 Hiroyuki Shimizu,1,* Tetsurou Sato,1 Yutaka Uehara,1 Shuichi Okada,1 and Masatomo Mori1,2

1Department of Medicine and Molecular Science, Gunma University Graduate School of Medicine, Maebashi 371-8511, Gunma, Japan2 CREST Japan Science and Technology Agency, Gunma University Graduate School of Medicine, Maebashi 371-8511, Gunma, Japan*Correspondence: [email protected]

Summary

High-fat diets cause peripheral leptin resistance, and dietary lipid composition affects sensitivity to leptin. We examinedthe role of n-3 polyunsaturated fatty acid (PUFA) in peripheral leptin resistance. Dietary PUFAs (0.4% wt/wt) caused insen-sitivity to peripherally but not intracerebroventricularly administered leptin. n-3 PUFA increased body weight, associatedwith a significant reduction of leptin concentration in the cerebrospinal fluid. Dietary n-3 PUFA reduced transport ofendogenous or exogenously administered leptin into the brain, associated with increased expression of hypothalamicoccludin, but caused no change in expression of leptin receptors, proteins associated with leptin signaling or other tightjunction proteins. Continuous intracerebroventricular infusion of an antisense morpholino oligonucleotide targeted to oc-cludin mRNA reversed n-3 PUFA-induced insensitivity to peripherally administered leptin. We conclude that n-3 PUFAinduces peripheral leptin resistance via an increase in the expression of hypothalamic occludin, reducing paracellulartransport of leptin into the brain.

Introduction

The ob gene product leptin was discovered through study ofgenetically obese (ob/ob) mice (Zhang et al., 1994). Leptin is a16 kDa adipokine that regulates feeding behavior, repro-duction, and immune function (Campfield et al., 1995; Halaaset al., 1995; Cunningham et al., 1999; Okamoto et al., 2000).Leptin, released from white adipose tissues, acts on the arcu-ate nucleus of the hypothalamus to inhibit appetite, to stimu-late sympathetic nerve activities, and to cause body weightreduction in rodents (Campfield et al., 1995; Halaas et al.,1995). Leptin is also known to stimulate secretion of LH andFSH from the pituitary gland via hypothalamic GnRH releaseand thereby activate the reproductive axis (Cunningham et al.,1999). The central administration of leptin is known to suppressspecific functions of the lymphocytes via the activation of thesympathetic nervous system (Okamoto et al., 2000). Leptin istransported across the blood-brain barrier (BBB) and binds toits specific receptor, which belongs to the class I cytokine re-ceptor family. The long form of the leptin receptor (OB-Rb) isactivated by the formation of a homodimer. The intracellularsignal transduction of leptin is mediated predominantly throughthe phosphorylation of the Janus kinase 2-signal transducerand activator of transcription-3 (STAT-3) pathway (Tartaglia etal., 1995; Vaisse et al., 1996). The SOCS-3 (suppressor of cy-tokine signaling-3) pathway is a leptin-inducible inhibitor of lep-tin signaling and has been suggested as a possible mediatorof leptin resistance in obesity (Bjorbaek et al., 1998).

In humans, the concentration of circulating leptin is wellcorrelated with body mass index (BMI), percentage of body fat,and total body fat weight (Considine et al., 1996; Shimizu etal., 1997) but is not related to body fat distribution (Shimizu et

CELL METABOLISM : MAY 2005 · VOL. 1 · COPYRIGHT © 2005 ELSEVIE

al., 1997). Restriction of dietary intake is also known to resultin a reduction of serum leptin concentration in humans (Shi-mizu et al., 1997). Although obese patients exhibit high con-centrations of circulating leptin, administration of endogenousleptin clearly fails to inhibit appetite in these patients. This indi-cates that these patients were resistant to leptin. A randomizedclinical trial confirmed that obese people are resistant to exog-enously administered leptin and that high doses of leptin arerequired to reduce body weight in obese patients (Heymsfieldet al. 1999). The ratio of cerebrospinal leptin concentration toblood leptin concentration is clearly lower in obese people thanin lean subjects (Caro et al., 1996; Schwartz et al., 1996a), indi-cating that the transport of leptin into the brain might be in-terrupted in human obesity. These clinical data strongly indi-cate that obese patients should exhibit peripheral leptinresistance. However, the existence of peripheral leptin resis-tance has yet to be confirmed, and the mechanisms involvedhave yet to be elucidated. Moreover, high-fat diets are knownto cause peripheral leptin resistance (Van Heek et al., 1997; El-Haschimi et al., 2000), and dietary lipid composition may affectleptin sensitivity. Dietary intake of fish oil appears to influencethe regulation of appetite and obesity (Tisdale and Dhesis,1990; Risica et al., 2000a; Risica et al., 2000b; Bjerregaard etal., 2002). Our recent clinical data have demonstrated that se-rum n-3 polyunsaturated fatty acid (PUFA) concentrations arethree times higher in obese subjects than in lean subjects (BMI,20.5 ± 0.6 kg/m2) 1.36% ± 0.16% (n = 10), obese subjects(BMI, 38.3 ± 1.3 kg/m2); 4.05% ± 0.26% (n = 19) (p < 0.01) (ourunpublished data). In addition, Westernization of the diet of In-uit immigrants resulted in a reduction in the proportion ofobese people within this population (Bjerregaard et al., 2002),

R INC. DOI 10.1016/j.cmet.2005.04.004 331

A R T I C L E

indicating that fish oil may play a role in the development ofhuman obesity.

Leptin is transported to the brain via the arcuate nucleus ofthe hypothalamus (Banks, 2003; Schwartz et al., 1996b). WhileFluoroGold does not penetrate the BBB, it is known to be read-ily taken up by the arcuate nucleus, even when injected periph-erally, suggesting that the arcuate nucleus BBB is poorly devel-oped (Merchenthaler, 1991). Bioactive substances such asleptin can move easily from the general circulation into the neu-ronal space of the arcuate nucleus as compared to other re-gions of the hypothalamus. Accumulating evidence suggeststhat leptin resistance is due to impairment of the transport ofleptin from the blood to the brain (Baskin et al., 1998). Alter-ations in the way that leptin is transported into the brain viathe BBB, along with hypothalamic leptin receptor abnormalitiesand postreceptor signal transduction mechanisms, are pos-sible contributory factors in the pathogenesis of peripheral lep-tin resistance in obese patients. Fasting leads to a reduction inthe rate of leptin transport through the BBB and into the brain(Kastin and Akerstrom, 2000), while α1-adrenergic stimulationwas shown to enhance the rate of leptin transport into the brain(Banks, 2001). However, the critical mechanisms involved inthe regulation of leptin transport in the hypothalamus are notyet fully understood, even in experimental models of obesity. Itwas reported that New Zealand Obese (NZO) mice, an animalmodel of obesity without any obvious genetic defects associ-ated with the leptin receptor, exhibited peripheral leptin resis-tance, as did rats fed upon a high-fat diet (Takahashi et al.,2001). Similar results have also been demonstrated with hu-man obesity. Both mouse and human models associated obe-sity with reduced uptake of leptin by the brain, while neither ofthe models exhibited significant reduction in Ob-R expressionin isolated cerebral microvessels (Hileman et al., 2002). Wehave recently found that NZO mice exhibit a reduced ratio ofarachidonic acid (AA; n-6) to n-3 PUFA in the sera due tochanges associated with lipid metabolism-related enzymeexpression in adipose tissue (Takahashi et al., 2001). Wehave also found a similar change in the ratio of AA to n-3PUFAs in the sera of rats fed upon a high-fat diet (our unpub-lished data).

The majority of the leptin in the arcuate nucleus of the hypo-thalamus is derived from movement across the BBB and rarelyfrom blood-cerebrospinal fluid traveling via the choroid plexus(Kurrimbux et al., 2004). It has been suggested that the recep-tor-mediated transfer of leptin into the brain appears to occuracross the BBB via a saturable system (Banks et al., 1996).However, the existence of a nonsaturable transport pathwayhas also been proposed (Rubin and Staddon, 1999). In obesity,the movement of leptin into the rat brain and cerebrospinal fluidis not affected by saturation of the transport carriers. In gene-ral, the transport of bioactive substances can be mediatedby both transcellular and paracellular mechanisms. Recentstudies have revealed that transport through the BBB can bemodulated to respond to environmental stimuli (Kastin and Ak-erstrom, 2000; Banks, 2001). Changes of n-3 PUFA composi-tion in brain endothelial cells forming the BBB may affect theability of leptin to move through the BBB. In the present study,we first demonstrate that n-3 PUFAs cause peripheral leptinresistance in rats resulting in obesity and then investigate a

332

mechanism of peripheral leptin resistance induced by an n-3PUFA-rich diet.

Results

n-3 PUFA causes insensitivity to peripherallyadministered leptinRats fed a high-fat (56%) diet for 14 days exhibited a signifi-cant increase in serum levels of n-3 PUFA (high-fat-fed rats:3.2% ± 0.4% [wt/wt], n = 8) when compared with rats fed alow-fat diet (2.4% ± 0.2% [wt/wt], n = 8, p < 0.05). This trendwas not, however, observed with levels of n-6 PUFA. We exam-ined the influence of fatty acid composition in a diet not con-taining fish oil upon leptin activity in rats by measuring the ef-fects of intraperitoneal (i.p.) injections of recombinant leptinduring the dark phase. Effects were monitored for a period of3 hr following injection. Injections were made 7 days after ratswere first provided with the test diets. Body weight gain overa 7 day period from the first provision of each of the test dietdays was significantly higher in rats fed diets containing EPAor DHA than rats fed diets containing other fatty acids (fed non-fish-oil-containing diet, 53.2 ± 1.8 g; stearic acid [18:0]-addeddiet, 53.1 ± 1.2 g; oleic acid [18:1]-added diet, 52.2 ± 1.1 g;linoleic acid [18:2]-added diet, 53.6 ± 1.4 g; docosahexaenoicacid [DHA; 22:5]-added diet, 60.3 ± 2.6 g [p < 0.05]; eicosapen-taenoic acid [EPA; 20:5]-added diet, 62.1 ± 1.5 g [p < 0.01]).As shown in Figure 1, the addition of 0.4% (wt/wt) stearic acid,oleic acid, and linoleic acid did not affect the anorexigenic ef-fects of recombinant leptin on food intake over a period of 3hr. In contrast, the addition of 0.4% (wt/wt) n-3 PUFA, EPA,and DHA completely deleted the anorexigenic effects of re-combinant leptin during this same time period. From these ob-servations, it appears that dietary n-3 PUFAs are involved inthe development of insensitivity to peripherally administeredrecombinant leptin, resulting in significant body weight gainover a 7 day period.

In the next experiment, we determined how n-3 PUFAs de-leted the anorexigenic effects of recombinant leptin by usingEPA. Serum concentration of n-3 PUFA in rats fed a diet con-taining EPA was three times higher than that seen in rats fed acontrol diet (non-n-3-PUFA-containing, 3.60% ± 0.91%; n-3-PUFA-containing diet, 10.56% ± 1.61%, p < 0.05, n = 8 eachgroup). As with obese humans, rats fed a diet containing n-3PUFA exhibited increased levels of n-3 PUFA in their blood.Similar i.p. administration of recombinant leptin (0.5 �g/g bodyweight) significantly inhibited food intake for 1 hr in animalsfed a diet that did not contain fish oil (Figure 2Aa). Significantsuppression of feeding for a period of 3 hr was observed inthose animals receiving recombinant leptin (5.0 �g/g bodyweight) administered i.p. (Figure 2Ab). In contrast, i.p. admin-istration of both 0.5 and 5.0 �g/g body weight of recombinantleptin did not show any significant reduction of food intake inanimals fed a diet containing 0.4% (wt/wt) n-3 PUFA (EPA) (Fig-ures 2Ba and 2Bb). On the other hand, intracerebroventricularadministration of recombinant leptin inhibited food intake in adose-dependent manner in rats fed a diet containing n-3PUFA, in a similar manner to control animals fed a diet notcontaining PUFA (Figure 2C). These data confirmed that n-3PUFA causes peripheral leptin resistance in rats. Evidence insupport of this is that the addition of n-3 PUFA completely de-leted the anorexigenic effects of peripherally administered re-

CELL METABOLISM : MAY 2005

Leptin resistance by n-3 PUFA

Figure 1.

Effects of the addition of 0.4% (wt/wt) saturated(stearic acid, 18:0), monounsaturated (oleic acid,18:1), diunsaturated (linoleic acid, 18:2), or polyun-saturated (EPA, 20:5; DHA, 22:5) fatty acids in dietsnot containing fish oil on the anorexigenic effects ofi.p. administered recombinant leptin (5.0 �g/gbw) inrats. The addition of 0.4% (wt/wt) DHA or EPA sup-pressed the anorexigenic effects of leptin for 3 hr(n = 7–8/group). Data represent mean ± SE.

combinant leptin. Moreover, intracerebroventricular administra-tion of recombinant leptin also inhibited food intake in animalsfed a diet not containing fish oil and also a diet supplementedwith n-3 PUFA.

Leptin insensitivity by n-3 PUFA causes obesityWe next examined whether n-3 PUFA may result in increasedbody weight due to peripheral leptin resistance and a conse-quent increase in the proportion of body fat in treated rats (Ta-ble 1). The rats fed a diet containing n-3 PUFA gained signifi-cantly more body weight over a 2 week period (body weightwas measured on day 7 and day 14). Mesenteric white adiposetissue weight was significantly higher in rats fed a diet contain-ing n-3 PUFA for 3 days. At 14 days, subcutaneous, epidydi-mal, and mesenteric white adipose tissue weight and inter-scapular brown adipose tissue weight were found to besignificantly higher in animals fed a diet containing n-3 PUFA.

n-3 PUFA disturbs the transport of leptin into the brainWe examined changes in the hypothalamic transport of leptinassociated with the diet containing n-3 PUFA. Immunoreactiveleptin (IRL) concentrations in cerebrospinal fluid (CSF) weresignificantly higher in rats fed a diet not containing fish oil thanin rats fed a diet containing n-3 PUFA without exogenous leptinadministration (Figure 2D), thereby indicating that the transportof endogenous leptin may be disturbed by n-3 PUFA. Intraperi-toneal administration of recombinant leptin significantly in-creased CSF-IRL concentration in rats fed a diet not containingfish oil but not in rats fed a diet containing n-3 PUFA. CSFleptin concentration increased to 234 pg/ml in rats fed a dietnot containing fish oil compared to 83 pg/ml in rats fed a dietcontaining n-3 PUFA. As shown in Table 1, n-3 PUFA increasedCSF-IRL concentrations over a 2 week period following initialprovision of the diet and was also associated with an increasein serum IRL concentration at 14 days. In an effort to measurethe transport of exogenously administered leptin across theBBB, we utilized human leptin and observed the transport ofthis into the CSF by using an ELISA system that selectivitymeasured human leptin. The concentration human leptin in rat


CSF 60 min after the i.p. administration of human leptin wassignificantly lower in rats fed a diet containing n-3 PUFA thanrats fed a diet not containing n-3 PUFA; this was in spite ofthese groups of animals exhibiting a similar increase in serumhuman leptin concentration (Figure 2E). The ratio of CSF toserum human leptin was also significantly reduced in rats feda diet containing n-3 PUFA. These data indicated that thetransport of both endogenous and exogenous leptin appearsto be disturbed in animals fed diets containing n-3 PUFA.Therefore, n-3 PUFA-rich diets may cause obesity due to thedevelopment of peripheral leptin resistance.

Alterations in the expression of molecules associatedwith the transport of leptinHow does n-3 PUFA manage to disturb the transport of leptinthrough the BBB into the brain? It is possible that leptin mightbe transported into the brain by two different pathways: trans-cellular and paracellular. Possible candidate mechanisms in-volved in peripheral leptin resistance include changes in leptinbinding capacity, leptin transport through the BBB, and postre-ceptor signal transduction, which may modulate transcellularleptin transport. Ob-Ra is capable of transporting intact leptinacross polarized epithelial cells in vitro (Hileman et al., 2000).To investigate alterations in the transcellular transport of leptinmediated by the leptin receptor, we evaluated changes in theexpression of the hypothalamic leptin receptor (Ob-Ra, short-form; Ob-Rb, long-form) by feeding rats diets containing n-3PUFA (Figure 3A). The expression of Ob-Ra and Ob-Rb wasunaffected by feeding rats a diet containing n-3 PUFA over a 7day period. The concentration of the serum soluble leptin re-ceptor (Ob-Re) may also be involved in the development ofperipheral leptin resistance (Huang et al., 2001; Zastrow et al.,2003). However, there were no differences in serum concentra-tions of Ob-Re when compared between rats fed a diet notcontaining fish oil and those fed a diet supplemented with n-3PUFA (Figure 3A). These data indicated that the binding capac-ity of cell membrane-associated leptin and circulating leptin isnot changed by diets containing n-3 PUFA and that bindingcapacities of the leptin receptor are not involved in n-3 PUFA-

333

A R T I C L E

Figure 2.

A−C) Changes in 0–1 hr and 1–3 hr food intake by i.p. ([A], 0.5 �g/gbw; [B], 5.0 �g/gbw body weight) and (C) intracerebroventricular (10–1000 ng/rat) recombinantleptin administration in rats fed a diet not containing fish oil either without (Aa, Ba, Ca) or with 0.4% (wt/wt) n-3 PUFA (EPA) (Ab, Bb, Cb).D) Changes in CSF immunoreactive leptin concentration as a result of an i.p. injection of 5.0 �g/gbw recombinant leptin in rats fed a diet not containing fish oil eitherwithout or with 0.4% (wt/wt) n-3 PUFA (EPA) (n = 8–16 in each group).E) Changes in serum and CSF human immunoreactive leptin concentration and the ratio of CSF to serum human leptin concentration 60 min after an i.p. injection of5.0 �g/gbw of recombinant human leptin in rats fed a non-fish-oil-containing diet either without or with 0.4% (wt/wt) n-3 PUFA (EPA) (n = 6 in each group). Datarepresent mean ± SE.

induced peripheral leptin resistance. There were no obviousdifferences in the expression of STAT-3, the phosphorylatedform of STAT-3 (p-STAT-3), and SOCS-3 prior to the administra-tion of leptin when compared between diets with and withoutn-3 PUFA. While leptin increased the phosphorylation of STAT-3in rats fed diets not containing n-3 PUFA, rats fed EPA exhib-ited no significant changes in the hypothalamic levels of SOCS-3,STAT-3, and p-STAT-3 when measured 60 min after the i.p. ad-ministration of leptin (Figure 3B).

On the other hand, bioactive substances are also trans-ported into the neural spaces through the BBB by paracellularmechanisms. There may be a possibility that the paracellulartransport of leptin may be modified by the presence of dietaryn-3 PUFA. The endothelial tight junctions consist of three inte-gral membrane proteins: claudin, occludin, and junctional ad-hesion molecule (JAM) (Morita et al., 1999; Furuse et al., 1993;Bazzoni and Dejana, 2001; Maness et al., 2000). The develop-ment of intercellular junctions seems to depend upon the ap-pearance of high levels of the tight junction proteins claudin-5,JAM-1, occludin, and ZO (Zonula Occludens)-1. We examinedthe possibility that n-3 PUFA may modify expression of these

334

tight junction proteins in the hypothalamus (Figures 4A–4D).The diet containing n-3 PUFA significantly increased hypothal-amic occludin expression but did not affect expression of hy-pothalamic claudin-5, JAM-1, or ZO-1. Occludin expressionwas investigated in eight separate regions of the brain (Figure4E). The diet containing n-3 PUFA resulted in increased expres-sion of occludin in each region of the brain, indicating that theeffect of n-3 PUFA is widespread across the entire brain. Theincreased expression of occludin was highest in the hypothala-mus, a structure known to regulate feeding behavior. As shownin Figure 4F, hypothalamic occludin expression had increasedby day 7 following the provision of the n-3 PUFA-containingdiet, and there was no difference in occludin expression whencompared between day 7 and day 14. The sensitivity to periph-erally administered leptin was already reduced by day 3. At thistime point, hypothalamic occludin expression was increasing,but circulating leptin levels remained unchanged (Figure 4G).Furthermore, we measured the expression of occludin inmicrovessels isolated from the hypothalamus. As shown in Fig-ure 4H, n-3 PUFA clearly increased the expression of occludinin the microvessels isolated from the hypothalamus. These



Table 1. Chronological changes of body weight gain, adipose tissue weight, and leptin concentration by n-3 PUFA

Days/n-3 PUFA 3 days/(−) 3 days/(+) 7 days/(−) 7 days/(+) 14 days/(−) 14 days/(+)

�BW (mg) 36.2 ± 2.5 43.0 ± 1.2 52.6 ± 7.1 67.4 ± 4.4a 104.3 ± 7.0 140.0 ± 7.5b

Sub (mg) 879.2 ± 53.6 924.5 ± 45.7 887.8 ± 108.9 1058.4 ± 59.3 1368.3 ± 97.4 1786.0 ± 132.6b

Epi (mg) 532.8 ± 41.0 584.0 ± 59.9 561.4 ± 32.2 719.6 ± 39.8b 1113.7 ± 50.9 1409.7 ± 62.5b

Mes (mg) 1234.4 ± 57.3 1580.3 ± 71.8b 1491.0 ± 87.8 1918.4 ± 118.2b 2132.0 ± 115.7 2661.6 ± 45.3b

Ret (mg) 458.4 ± 26.4 452.5 ± 18.5 479.8 ± 75.0 621.6 ± 49.3 987.6 ± 50.5 1307.6 ± 147.7BAT (mg) 288.6 ± 36.0 306.5 ± 23.9 314.8 ± 25.3 377.8 ± 43.0 415.3 ± 40.3 543.6 ± 20.0b

S-Leptin (pg/ml) 2289.2 ± 309.9 2174.4 ± 203.7 2123.8 ± 261.7 2458.5 ± 342.2 2182.0 ± 361.2 2924.0 ± 295.2a

C-Leptin (pg/ml) 192.7 ± 50.3 171.0 ± 35.4 211.7 ± 39.9 146.0 ± 29.2b 209.6 ± 38.2 138.2 ± 31.5b

�BW, body weight gain; Sub, subcutaneous white adipose tissue; Epi, epididymal white adipose tissue; Mes, mesenteric white adipose tissue; Ret, retroperitonealwhite adipose tissue; BAT, brown adipose tissue; S-Leptin, serum leptin concentration; C-Leptin, CSF leptin concentration.a p < 0.05 versus rats fed non-n-3-PUFA-containing diet.b p < 0.005 versus rats fed non-n-3-PUFA-containing diet.

data indicated that an increase of hypothalamic occludin ex-pression might be involved in n-3 PUFA-induced peripheralleptin resistance.

Expression of hypothalamic occludin by n-3 PUFAcauses insensitivity to leptinTo investigate the possibility that overexpression of hypothala-mic occludin is involved in n-3 PUFA-induced leptin resistance,we used an antisense morpholino nucleotide against the oc-cludin gene (40 �g/day) and investigated its effect upon theexpression of hypothalamic occludin with or without n-3 PUFA(Figure 5A). Chronic intracerebroventricular infusion of an anti-sense morpholino oligonucleotide (40 �g/day) designed againstoccludin in rats fed a diet containing n-3 PUFA resulted in thereduction of occludin expression to the level of animals feda diet not containing fish oil (Figure 5B, right panel). Chronicintracelebroventricular infusion of a 5-missense morpholino oli-gonuleotide failed to exhibit such a change in occudin expres-sion. Body weight gain of rats administered with antisense
that the overexpression of occludin by n-3 PUFA causes insen-
Figure 3.

A) Changes in the expression of hypothalamic Ob-Ra and Ob-Rb along with serum Ob-Re concentration in rats fed a diet not containing fish oil either without or with0.4% (wt/wt) n-3 PUFA (EPA) (n = 8 in each group). No differences in leptin receptor expression were detected between these two groups.B) Expression of hypothalamic SOCS-3, phosphorylated STAT3 (p-STAT3), and total STAT3 when measured 60 min after an i.p. injection of 5.0 �g/gbw recombinantleptin (n = 8 rats in each group). Data represent mean ± SE.


morpholino oligonucleotides was significantly less than thatseen in rats administered with 5-missense morpholino oligonu-leotide over a 7 day period (5-missense-treated rats; 39.2 ±1.5 g, antisense-treated rats; 31.3 ± 2.1 g, p < 0.001). Asshown in the lower panel of Figure 5C, continuous infusion(into the third ventricle) of antisense morpholino oligonucleo-tides against occludin resulted in the reversal of n-3 PUFA-induced leptin insensitivity. However, infusion of the 5-missensemorpholino oligonucleotide did not affect n-3 PUFA-inducedleptin insensitivity. Furthermore, chronic intracerebroventricularinfusion of an antisense morpholino oligonucleotide resulted inreduced expression of hypothalamic occludin in rats fed dietsnot containing n-3 PUFA (Figure 5B, left panel). Body weightgain was not affected by the administration of the antisensemorpholino (5-missense-treated rats; 31.9 ± 2.1 g, antisense-treated rats; 32.9 ± 2.3 g, not significant), and the sensitivityto acute, peripheral leptin administration remained unchanged(Figure 5C, upper panel). This observation strongly indicated

335

A R T I C L E

Figure 4.

A−D) Changes in the expression of hypothalamic (A) claudin-5, (B) JAM-1, (C) occludin, and (D) ZO-1 in rats fed a diet not containing fish oil either without or with0.4% (wt/wt) n-3 PUFA (EPA) (n = 8 in each group). Expression of hypothalamic occludin was markedly elevated in rats fed a diet supplemented with n-3 PUFA.E) Changes in occludin expression in eight separate regions of the brain in rats fed a diet supplemented with n-3 PUFA (n = 5 in each group).F) Chronological change of occludin expression in the hypothalamus in rats fed a diet supplemented with n-3 PUFA (n = 5 in each group).G) Changes in 0–1 hr and 1–3 hr food intake as a result of the i.p. administration of recombinant leptin without or with 0.4% (wt/wt) n-3 PUFA (EPA) at day 3 (n = 4–5in each group).H) Changes in the expression of occludin in microvessels isolated from the hypothalamus in rats fed a diet supplemented with n-3 PUFA (n = 6 in each group). Datarepresent mean ± SE.

sitivity to peripherally administered leptin. The observed in-crease of occludin expression in the hypothalamus of rats feddiets containing n-3 PUFA is expected to contribute tochanges in the transport of leptin through the BBB and into theneural spaces.

Furthermore, we clarified the status of occludin expressionin other models of experimental obesity (Figure 6). Expressionof hypothalamic occludin was clearly increased in rats fed ahigh-fat (56%) diet over a period of 14 days, during which se-rum levels of n-3 PUFA were also significantly increased. Onthe other hand, no such changes were found in genetic modelsof obesity such as genetically obese (ob/ob) mice, geneticallydiabetic (db/db) mice, and fatty Zucker rats.

Discussion

Composition of dietary nutrients may affect appetite. Leptin af-fects preference to these nutrients. It has been established that

336

a high-fat diet can cause leptin resistance. However, at presentwe know nothing about the possible relationship between dif-ferent types of fatty acids and leptin resistance. NZO mice andmice that have become obese as a result of a high-fat diethave been reported to exhibit peripheral leptin resistance (VanHeek et al., 1997; El-Haschimi et al., 2000; Burguera et al.,2000). However, recent data demonstrated that central leptintherapy fails to overcome the leptin resistance associated withdiet-induced obesity (Wilsey et al., 2003) and that obesity-prone rats possess a normal blood brain barrier but defectivecentral leptin signaling pathways prior to the onset of obesity(Levin et al., 2004). SOCS-3 is one of the possible candidatesinvolved in central leptin resistance, and an increase of SOCS-3in the arcuate nucleus of the hypothalamus is known to berelated to region-specific leptin resistance (Munzberg et al.,2004). Our present study first demonstrated that the additionof n-3 PUFA caused peripheral leptin resistance in rats fed adiet not containing fish oil. We also demonstrated that the ob-



Figure 5.

Changes in food intake in rats continuously infused intracerebroventricularly with 40 �g/day of antisense or 5-missense morpholino oligonucleotides directed againstthe rat occludin gene in animals fed a diet not containing fish oil either with or without 0.4% (wt/wt) n-3 PUFA (EPA). (A) Experimental protocol, (B) expression ofhypothalamic occluding, (C) chronological changes in 0–1 hr and 1–3 hr food intake following i.p. administration of recombinant leptin (0.5–5.0 �g/gbw). Continuousintracerebroventricular infusion of antisense morpholino oligonucleotide against occludin restored leptin’s anorexigenic effects in rats fed a diet rich in n-3 PUFA (n =4–14 in each group). Data represent mean ± SE.

served overexpression of hypothalamic occludin in rats feddiets supplemented with n-3 PUFA is very likely to be involvedin the development of peripheral leptin resistance. The additionof n-3 PUFA to diets may affect total caloric intake in the ani-mals, thereby causing leptin resistance. However, an increasein caloric intake caused by the addition of 0.4% n-3 PUFA istrivial in this study (non-fish-oil-containing diet, 4 kcal/g, 0.4%n-3 PUFA, 0.016 kcal/g − total calorie (n-3-PUFA-containingdiet) 4.016 kcal/g). It was noted that both groups of rats con-sumed a very similar number of calories during the experiment,and, therefore, the obesity observed in our model was due notto an increase in total caloric intake but purely to the metaboliceffect of EPA.

The transport rates of intravenously administered leptin inboth leptin-receptor-deficient db/db mice and leptin-deficientob/ob mice do not differ from those of wild mice (Maness etal., 2000). The CSF/plasma ratio of leptin was decreased by10-fold in obese Koletsky rats, which have a mutation in the


leptin receptor gene (Wu-Peng et al., 1997). Furthermore,transport of leptin across the BBB of the Koletsky rat is knownnot to be mediated by a product of the leptin receptor gene(Banks et al., 2002). These previous observations strongly indi-cate the existence of a leptin transport system that is indepen-dent of the leptin receptor. In the present study, feeding rats adiet rich in n-3 PUFA for 14 days resulted in increased concen-trations of serum leptin concomitant with significant reductionin the concentration of CSF leptin. As a result of this, there wasalso a decrease in the CSF/serum ratio of leptin. A similarchange in ratio has been reported for obese humans. This find-ing indicates that dietary n-3 PUFA is associated with de-creased leptin uptake in the hypothalamus. However, the ex-pression of both short and long forms of the leptin receptor inthe hypothalamus remained unchanged during the administra-tion of a diet rich in n-3 PUFA. Therefore, it is possible that anon-leptin-receptor-mediated transport system may play a role

337

A R T I C L E

Figure 6.

Changes in the expression of hypothalamic oc-cludin in rats fed diets containing 10% fat and 56%fat diet, ob/ob mice, db/db mice, Zucker fatty rats,and lean littermates. The high-fat diet increased hy-pothalamic occludin expression. In contrast, therewere no significant changes in the expression of hy-pothalamic occludin when compared between leanand obese mice and rats. Data represent mean ± SE.

in the n-3-PUFA-induced disturbance of leptin transport intothe brain.

It has been reported that some proteins can be transportedby two different types of pathways: transcellular and paracellu-lar (Broadwell et al., 1988; Quagliarello et al., 1991; Van Breeet al., 1989). The present data indicate that a paracellular trans-port system in brain endothelial cells may be important interms of leptin transport through the BBB into the brain. Twomembrane proteins, occludin and claudin, are mainly responsi-ble for BBB function. Occludin consists of four loops throughthe membrane and forms a homodimer with occludin from ad-jacent cells. Polymers of occludin may form junctional fibrils(Gumbiner, 1993), and it is known that occludin contributes toincreased cell-cell adhesion (Van Itallie and Anderson, 1997).An increase in paracellular permeability to macromolecules isassociated with loss of total cellular occludin, indicating thatoccludin plays a critical role in the tight junction permeabilitybarrier (Wong and Gumbiner, 1997). Furthermore, occludin isknown to regulate the paracellular transport of some proteinsin brain endothelial cells. The present study demonstrated asignificant increase of occludin expression in microvessels iso-lated from the hypothalamus of rats fed a diet rich in n-3 PUFA.Treatment of the human vascular endothelial cell line ECV304with n-3 PUFA resulted in increased transendothelial cell resis-tance, upregulated occludin expression, and reduced paracel-lular permeability to large molecules (Jiang et al., 1998). EPAexerted an upregulatory effect on occludin expression in thiscell line. Brain capillary endothelial cells cultured in EPA exhib-ited an increase in trans-epithelial electrical resistance. Further-more, expression levels of occludin messenger RNA increasedmarkedly after exposure to EPA, indicating that n-3 PUFA fromastrocytes may contribute to the formation of the BBB (Yama-gata et al., 2003). Feeding rats a diet rich in n-3 PUFA resultedin increased concentrations of circulating n-3 PUFA, which islikely to contribute to increased expression of occludin in thehypothalamus.

Dietary supplements of n-3 PUFA result in reduced expres-sion of vascular endothelial growth factor (VEGF) in an animalmodel of progressive malignancy (Tevar et al., 2002). VEGF in-creases the permeability of a cultured brain microvessel endo-thelial cell monolayer by influencing occludin expression (Wanget al., 2001). Therefore, it is possible that n-3 PUFA may stimu-late occludin expression via an increase in VEGF production inthe hypothalamus. However, the involvement of VEGF in theobserved increase of hypothalamic occludin expression in re-sponse to n-3 PUFA can be ruled out, since hypothalamicVEGF expression was not reduced in the present study byfeeding rats a diet rich in n-3 PUFA (data not shown).

Recently, occludin knockout mice have been reported to ex-hibit delayed development and a significantly lower body

338

weight than wild mice (Saitou et al., 2000). These observationsare compatible with our own findings in that occludin is in-volved in the regulation of body weight. It was not possiblefor us to evaluate whether occludin knockout mice also exhibitreduced food intake, reduced body fat, and increased CSF/plasma ratio of leptin, since these factors have yet to bestudied in occludin knockout mice. It is known that the expres-sion of truncated occludin increased both trans-epithelialelectrical resistance and flux of nontransported solutes withmolecular masses from 4 to 40 kDa but to a greater degreethan wild-type occludin (Balda et al., 1996). It is possible,therefore, that leptin (16 kDa) may be easily transportedthrough the BBB into the brain in occludin knockout mice,thereby delaying body development due to appetite reduction.An increase of hypothalamic occludin expression by the addi-tion of n-3 PUFA in the diet may induce leptin insensitivity byreducing leptin transport to the feeding centers in the hypo-thalamus.

It was demonstrated by clinical data that EPA administrationcauses significant weight gain and an increase in dietary intakein cachexic patients suffering from malignant tumors (Barber etal., 2000; Wigmore et al., 2000). Westernization of diets alsoreduced the proportion of obese people amongst Inuit immi-grants (Risica et al., 2000a). These clinical observations sug-gest that diet rich in n-3 PUFA causes increased body weightin humans. In addition, our unpublished clinical data demon-strated that n-3 PUFA concentrations were significantly higherin the sera of massively obese Japanese subjects than leansubjects, indicating that the n-3 PUFA component of brain en-dothelial cell membrane may also be increased in these sub-jects. These data raised the possibility that the observedchanges in hypothalamic tight junction-related peptide expres-sion in response to supplementary dietary n-3 PUFA shouldalso be important in the regulation of appetite and metabolicrates in obese people.

The present observations indicate that a diet rich in n-3PUFA induces peripheral leptin insensitivity and that hypothala-mic tight junction proteins are likely to be very important in thedevelopment of peripheral leptin resistance caused by the n-3PUFA-rich diet.

Experimental procedures

AnimalsMale Wistar rats weighing approximately 150 g were used throughout thisstudy. For studies involving a high-fat diet, the animals ate either a semisyn-thetic low-fat (10% of Kcal as fat) or high-fat (56% of Kcal) diet (Shimizu etal., 1994). Ob/ob mice, db/db mice, and their controls were used at an ageof 5 weeks. Zucker fa/fa rats and their lean littermates were used at an ageof 7weeks. All animals were housed individually in a temperature-controlledroom with a 12 hr light and 12 hr dark cycle (illumination from 6:00 to 18:00).Food and water were available ad libitum. All experimental protocols were



approved by the Animal Use and Care Committee of the Gunma UniversityGraduate School of Medicine.

Blood chemistrySerum n-3 PUFA concentration in human and rats was measured by gascolumn chromatography. After transmethylation, fatty acid methyl esterswere extracted and quantified with a gas column chromatography system(Hewlett-Packard Model HP6890, USA). The results are expressed as a per-centage (weight/weight) and can be defined as the relative percentage ofeach fatty acid concentration within a given serum lipid concentration.Commercially available kits were used to measure immunoreactive leptinconcentrations in serum (Pharmacia, Linco Research, St. Charles, USA) andCSF (Immuno-Biological Laboratories, Inc. Ltd., Fujioka, Gunma, Japan).

Feeding experimentsAll animals were habituated to a powdered chow diet that did not containfish oil (Funabashi Farm, Chiba, Japan) with or without 0.4% wt/wt stearicacid, oleic acid, linoleic acid (Sigma-Aldrich Chem. Co., St. Louis, MO,USA), DHA (WAKO Pure Chemical Industries, Osaka, Japan), and EPA(Mochida Pharmaceuticals, Co., Tokyo, Japan) for 7 days prior to the startof each experiment.

For the acute experiments, recombinant leptin (0.5 or 5.0 �g/g bodyweight, R&D Systems, Minneapolis, MN, U.S.A.) dissolved in distilled wateror a vehicle solution was injected by i.p. or intracerebroventricular (ICV)routes at 18:00 hr, and subsequent changes in the animals’ food consump-tion were measured to the nearest 0.1 g. ICV infusion of recombinant leptin,or vehicle, was performed through a 29-gauge injection cannula, which wasinserted through a 23-gauge guide cannula stereotaxically implanted intothe third ventricle of rats 7 days earlier (Uehara et al., 1998). Recombinantleptin (5–500 ng/rat, dissolved in 5 �l of distilled water) or a vehicle solutionwas infused for 2–3 min through a polyethylene tube connected to a 25 �lmicrosyringe. Following the infusion, the injection cannula was maintainedin the guide cannula for at least 3 min to prevent backflow, after whicheach animal was returned to its own cage. Food intake was then measuredchronologically for all rats in this study.

CSF collectionRats were anesthetized by the administration of 40 mg/kg pentobarbital(i.p.) by cisternal puncture with a 28-gauge needle. CSF was then collectedfrom anesthetized rats that did not receive leptin and those that were in-jected with leptin (i.p.) 1 hr earlier (Seki et al., 1994). CSF was slowly col-lected into the syringe, and the collection was immediately terminated whenblood appeared. On average, 70–80 �l of CSF was obtained per rat. Centri-fuged samples were frozen at −20°C until use.

Microvessel isolationCerebral microvessels were isolated from rat brain by the method describedpreviously (Nonaka et al., 2004). Two brain hypothalami were collected frommale Wister rats (n = 6) and homogenized gently with a Dounce tissuegrinder in ice-cold stock buffer (pH 7.4; 1% dextran in minimam essentialmedium [GIBCO-BRL, Grand Island, NY, USA]) with 25 mM HEPES on ice.The homogenized tissue was filtered through a nylon mesh net (308 �mfollowed by 108 �m two times), mixed with an equal volume of 40% dextranin cold stock buffer, and centrifuged at 5000 × g for 15 min at 4°C. Thepellet was then resuspended in stock buffer, filtered through a 25 �m nylonmesh membrane, washed from the surface of the membrane with stockbuffer four times, collected, and centrifuged at 5000 × g for 15 min at 4°C.Isolated microvessels were then stored at −80°C until the assay.

Leptin transport assayRats were anesthetized by i.p. administration of 40 mg/kg pentobarbital.Blood was sampled from the right jugular vein 60 min after the i.p. admin-istration of human immunoreactive leptin and CSF collected by cisternalpuncture at the same time point. A commercially available human leptinELISA kit (Immuno-Biological Laboratories, Inc. Ltd., Fujioka, Gunma, Ja-pan) was used to measure immunoreactive human leptin concentrations inserum and CSF. In this assay kit, crossreactivity with rat leptin was lessthan 0.1%.


Western blot analysesProteins were extracted for immunoblot analysis. Dissected rat hypothalamiwere lysed in ice-cold Tris-HCl buffer containing NP-40, phenylmethylsulfo-nylfluoride (PMSF), leupeptin, pepstatin, and sodium vanadate. Followingprotein determination using the method described by Lowry (Lowry et al.,1951), 20 �g of protein was separated using SDS-PAGE electrophoresis ona 12.5% gel. Separated proteins were then transferred onto nitrocellulosemembranes. Membranes were first preblocked with nonfat milk at roomtemperature for 1 hr and were then incubated with primary antibodies over-night at 4°C. Primary antibodies against the three rat leptin receptor forms(Ob-Ra, Ob-Rb, Ob-Re) as well as occludin, SOCS-3, phosphorylatedSTAT-3, and STAT-3 were obtained from Santa Cruz Biotechnology Inc.(Santa Cruz, CA, USA). Primary antibodies against rat claudin-5 and ratZO-1 were obtained from Zymed (South San Francisco, CA, USA). An anti-body against rat JAM-1, the second antibody, and goat anti-rabbit IgG wereobtained from Jackson ImmunoResearch Laboratories, Inc. (West Balti-more, PA, USA). Finally, membranes were washed in Tris-buffered salinecontaining Tween 20 and positive antibody binding detected usingchemifluoresence ECL plus (Amersham Biosciences, UK).

Antisense morpholino oligonucleotide against occludinThe antisense and 5-missense morpholino oligonucleotides against rat oc-cludin were obtained from Gene Tools, LLC (Philomath, OR, USA). The se-quences of the antisense and 5-missense morpholino oligonucleotideswere CATGGCTGATGTCACTGGTCACCTG and CAAGGGTGATGTGACTGGTGACGTGA, respectively. These oligos were continuously infused (40 �g/day) into the third ventricle using Alzet osmotic mini-infusion pumps (Cuper-tino, CA, USA) via a cannula that was implanted 7 days prior to the start ofthe infusion. The infusion period used for the oligonucleotide was deter-mined in preliminary experiments (data not shown). Food intake in responseto leptin administration (5.0 �g/g body weight) following the continuousinfusion of each morpholino oligonucleotide over a 7 day period was mea-sured, as described earlier.

Data analysisResults are expressed as mean ± SEM. Significant differences betweengroups were evaluated using analysis of variance (ANOVA) followed byNewman-Keuls test.

Acknowledgments

This study was supported by the funds from Core Research for EvolutionScience and Technology (Japan).

Received: November 1, 2004Revised: February 12, 2005Accepted: April 20, 2005Published: May 10, 2005

References

Balda, M., Whitney, J.A., Flores, C., Gonzalez, S., Cereijido, M., and Matter,K. (1996). Functional dissociation of paracellular permeability and transepi-thelial electrical resistance and disruption of the apical-basolateral intra-membrane diffusion barrier by expression of a mutant tight junction mem-brane protein. J. Cell Biol. 134, 1031–1049.

Banks, W.A. (2001). Enhanced leptin transport across the blood-brain bar-rier by α-adrenergic agents. Brain Res. 899, 209–217.

Banks, W.A. (2003). Is obesity a disease of the blood-brain barrier? Physio-logical, pathological, and evolutional considerations. Curr. Pharm. Des. 9,801–809.

Banks, W.A., Kastin, A.J., Huang, W., Huang, W., Jaspan, J.B., and Maness,L.M. (1996). Leptin enters the brain by a saturable system independent ofinsulin. Peptides 2, 305–311.

Banks, W.A., Niehoff, M.L., Martin, D., and Farrell, C.L. (2002). Leptin trans-

339

A R T I C L E

port across the blood-brain barrier of the Koletsky rat is not mediated by aproduct of the leptin receptor gene. Brain Res. 950, 130–136.

Barber, M.D., McMillan, D.C., Preston, T., Ross, J.A., and Fearon, K.C.(2000). Metabolic response to feeding in weight-losing pancreatic cancerpatients and its modulation by a fish-oil-enriched nutritional supplement.Clin. Sci. (Lond.) 98, 389–399.

Baskin, D.G., Seeley, R.J., Kuijper, J.L., Lok, S., Weigle, D.S., Erickson, J.C.,Palmiter, R.D., and Schwartz, M.W. (1998). Increased expression of mRNAfor the long form of the leptin receptor in the hypothalamus is associatedwith leptin hypersensitivity and fasting. Diabetes 47, 538–543.

Bazzoni, G., and Dejana, E. (2001). Pores in the sieve and channels in thewall: control of paracellular permeability by junctional proteins in endothelialcells. Microcirculation 8, 143–152.

Bjerregaard, P., Jorgensen, M.E., Andersen, S., Mulvad, G., and Borch-Johnsen, K. (2002). The Greenland Population Study. Decreasing over-weight and central fat patterning with Westernization among the Inuit inGreenland and Inuit migrants. Int. J. Obes. Relat. Metab. Disord. 26,1503–1510.

Bjorbaek, C., Elmquist, J.K., Frantz, J.D., Shoelson, S.E., and Flier, J.S.(1998). Identification of SOCS-3 as a potential mediator of central leptinresistance. Mol. Cell 1, 619–625.

Broadwell, R.D., Balin, B.J., and Salcman, M. (1988). Transcytotic pathwayfor blood-borne protein through the blood-brain barrier. Proc. Natl. Acad.Sci. USA 85, 632–636.

Burguera, B., Couce, M.E., Curran, G.L., Jensen, M.D., Lloyd, R.V., Cleary,M.P., and Poduslo, J.F. (2000). Obesity is associated with a decreased leptintransport across the blood-brain barrier in rats. Diabetes 49, 1219–1223.

Campfield, L.A., Smith, F.J., Guisez, Y., Devos, R., and Burn, P. (1995). Re-combinant mouse OB protein: evidence for a peripheral signal linking adi-posity and central neural networks. Science 269, 546–549.

Caro, J.F., Kolaczynski, J.W., Nyce, M.R., Ohannesian, J.P., Opentanova, I.,Goldman, W.H., Lynn, R.B., Zhang, P.-L., Sinha, M.K., and Considine, R.V.(1996). Decreased cerebrospinal-fluid/serum leptin ratio in obesity: a pos-sible mechanism for leptin resistance. Lancet 348, 159–161.

Considine, R.V., Sinha, M.K., Heiman, M.L., Kriauciunas, A., Stephens, T.W.,Nyce, M.R., Ohannesian, J.P., Marco, C.C., McKee, L.J., Bauer, T.L., andCaro, J.F. (1996). Serum immunoreactive-leptin concentrations in normal-weight and obese humans. N. Engl. J. Med. 334, 292–295.

Cunningham, M.J., Clifton, D.K., and Steiner, R.A. (1999). Leptin’s actionson the reproductive axis: perspectives and mechanisms. Biol. Reprod. 60,216–222.

El-Haschimi, K., Pierroz, D.D., Hileman, S.M., Bjorbaek, C., and Flier, J.S.(2000). Two defects contribute to hypothalamic leptin resistance in micewith diet-induced obesity. J. Clin. Invest. 105, 1827–1832.

Furuse, M., Hirase, T., Itoh, M., Nagafuchi, A., Yonemura, S., and Tsukita,S. (1993). Occludin: a novel integral membrane protein localizing at tightjunctions. J. Cell Biol. 123, 1777–1788.

Gumbiner, B.M. (1993). Breaking through the tight junction barrier. J. CellBiol. 123, 1631–1633.

Halaas, J.L., Gajiwala, K.S., Maffei, M., Cohen, S.L., Chait, B.T., Rabinowitz,D., Lallone, R.L., Burley, S.K., and Friedman, J.M. (1995). Weight-reducingeffects of the plasma protein encoded by the ob gene. Science 269, 543–546.

Heymsfield, S.B., Greenberg, A.S., Fujioka, K., Dixon, R.M., Kushner, R.,Hunt, T., Lubina, J.A., Patane, J., Self, B., Hunt, P., and McCamish, M.(1999). Recombinant leptin for weight loss in obese and lean adults. A ran-domized, controlled, dose-escalation trial. JAMA 282, 1568–1575.

Hileman, S.M., Tornoe, J., Flier, J.S., and Bjorbaek, C. (2000). Transcellulartransport of leptin by the short leptin receptor isoforms ObRa in Madin-Darby canine kidney cells. Endocrinology 141, 1955–1961.

Hileman, S.M., Pierroz, D.D., Masuzaki, H., Bjorbaek, C., El-Haschimi, K.,Banks, W., and Flier, J.S. (2002). Characterization of short isoforms of the

340

leptin receptor in rat cerebral microvessels and of brain uptake of leptin inmouse models of obesity. Endocrinology 143, 775–783.

Huang, L., Wang, Z., and Li, C. (2001). Modulation of circulating leptin levelsby its soluble receptors. J. Biol. Chem. 276, 6343–6349.

Jiang, W.G., Bryce, R.P., Horrobin, D.F., and Mansel, R.E. (1998). Regulationof tight junction permeability and occludin expression by polyunsaturatedfatty acids. Biochem. Biophys. Res. Commun. 244, 414–420.

Kastin, A.J., and Akerstrom, V. (2000). Fasting, but not adrenalectomy, re-duces transport of leptin into the brain. Peptides 21, 679–682.

Kurrimbux, D., Gaffen, Z., Farrell, C.L., Martin, D., and Thomas, S.A. (2004).The involvement of the blood-brain and the blood-cerebrospinal fluid barri-ers in the distribution of leptin into and out of the rat brain. Neuroscience123, 527–536.

Levin, B.E., Dunn-Meynell, A.A., and Banks, W.A. (2004). Obesity-prone ratshave normal blood-brain barrier transport but defective central leptin signal-ing before obesity onset. Am. J. Physiol. Regul. Integr. Comp. Physiol. 286,R143–R150.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951). Proteinmeasurement with the Folin phenol reagent. J. Biol. Chem. 193, 265–275.

Maness, L.M., Banks, W.A., and Kastin, A.J. (2000). Persistance of blood tobrain transport of leptin in obese leptin-deficient and leptin receptor-defi-cient mice. Brain Res. 873, 165–167.

Merchenthaler, I. (1991). Neurons with access to the general circulation inthe central nervous system of the rat: a retrograde tracing study with fluoro-gold. Neuroscience 44, 655–662.

Morita, K., Sasaki, H., Furuse, M., and Tsukita, S. (1999). Endothelial clau-din: claudin 5 TMVCF, constitutes tight junction strands in endothelial cells.J. Cell Biol. 147, 185–194.

Munzberg, H., Flier, J.S., and Bjorbaek, C. (2004). Region-specific leptinresistance within the hypothalamus of diet-induced obese mice. Endocri-nology 145, 4880–4889.

Nonaka, N., Hileman, S.M., Shioda, S., Vo, T.Q., and Banks, W.A. (2004).Effects of lipopolysaccharide on leptin transport across the blood-brain bar-rier. Brain Res. 1016, 58–65.

Okamoto, S., Irie, Y., Ishikawa, I., Kimura, K., and Masayuki, S. (2000).Central leptin suppresses splenic lymphocyte functions through activationof the corticotropin-releasing hormone-sympathetic nervous system. BrainRes. 855, 192–197.

Quagliarello, V.J., Ma, A., Stukenbrok, H., and Palade, G.E. (1991).Ultrastructual localization of albumin transport across the cerebral micro-vasculature during experimental meningitis in the rat. J. Exp. Med. 174,657–672.

Risica, P.M., Schraer, C., Ebbesson, S.O.E., Nobmann, E.D., and Caballero,B.H. (2000a). Body fat distribution in Alaskan Eskimos of the Bering Straitsregion: the Alaskan Siberia Project. Int. J. Obes. Relat. Metab. Disord. 24,171–179.

Risica, P.M., Schraer, C., Ebbesson, S.O.E., Nobmann, E.D., and Caballero,B.H. (2000b). Overweight and obesity among Alaskan Eskimos of the BeringStraits region: the Alaska Siberia Project. Int. J. Obes. Relat. Metab. Disord.24, 939–944.

Rubin, L.L., and Staddon, J.M. (1999). The cell biology of the blood-brainbarrier. Annu. Rev. Neurosci. 22, 11–28.

Saitou, M., Furuse, M., Sasaki, H., Schulzke, J.-D., Fromm, M., Takano,H., Noda, T., and Tsukita, S. (2000). Complex phenotype of mice lackingoccluding, a component of tight junction strands. Mol. Biol. Cell 11, 4131–4142.

Seki, T., Sato, N., Hasegawa, T., Kawaguchi, T., and Juni, K. (1994). Nasalabsorption of zidovudine and its transport to cerebrospinal fluid in rats. Biol.Pharm. Bull. 17, 1135–1137.

Shimizu, H., Fisler, J.S., and Bray, G.A. (1994). Extracellular hypothalamicmonoamines measured by in vivo microdialysis in a rat model of dietary fat-induced obesity. Obes. Res. 2, 100–109.



Shimizu, H., Shimomura, Y., Hayashi, R., Ohtani, K., Sato, N., Futawatari,T., and Mori, M. (1997). Serum leptin concentration is associated with totalbody fat mass, but not abdominal fat distribution. Int. J. Obes. Relat. Metab.Disord. 21, 536–541.

Schwartz, M.W., Peskind, E., Raskind, M., Boyko, E.J., and Porte, D., Jr.(1996a). Cerebrospinal fluid leptin levels: relationship to plasma levels andto adiposity in humans. Nat. Med. 2, 589–593.

Schwartz, M.W., Seeley, R., Campfield, L.A., Burn, P., and Baskin, D.G.(1996b). Identification of targets of leptin action in rat hypothalamus. J. Clin.Invest. 98, 1101–1106.

Takahashi, H., Oh-Ishi, M., Shimizu, H., and Mori, M. (2001). Detection andidentification of subcutaneous adipose tissue protein related to obesity inNew Zealand Obese mouse. Endocr. J. 48, 205–211.

Tartaglia, L.A., Dembski, M., Weng, X., Deng, N., Culpepper, J., Devos, R.,Richards, G.J., Campfield, L.A., Clark, F.T., and Deeds, J. (1995). Identifica-tion and expression cloning of a leptin receptor, OB-R. Cell 83, 263–271.

Tevar, R., Jho, D.H., Babcock, T., Helton, W.S., and Espat, N.J. (2002).Omega-3 fatty acid supplement reduces tumor growth and vascular endo-thelial growth factor expression in a model of progressive non-metastasiz-ing malignancy. JPEN J. Parenter Enteral. Nutr. 26, 285–289.

Tisdale, M.J., and Dhesis, J.K. (1990). Inhibition of weight loss by omega-3fatty acids in an experimental cachexia model. Cancer Res. 50, 5022–5026.

Uehara, Y., Shimizu, H., Ohtani, K., Sato, N., and Mori, M. (1998). Hypothal-amic corticotropin releasing hormone (CRH) is a mediator of the anorexi-genic effect of leptin. Diabetes 47, 890–893.

Vaisse, C., Halaas, J.L., Horvath, C.M., Darnell, J.E., Jr., Stoffel, M., andFriedman, J.M. (1996). Leptin activation of Stat3 in the hypothalamus ofwild-type and ob/ob mice but not db/db mice. Nat. Genet. 14, 95–97.

Van Bree, J.B., de Boer, A.G., Verhoef, J.C., Danhof, M., and Breimer, D.D.(1989). Transport of vasopressin fragments across the blood-brain barrier:in vitro studies using monolayer cultures of bovine brain endothelial cells.J. Pharmacol. Exp. Ther. 249, 901–905.


Van Heek, M., Compton, D.S., France, C.F., Tedesco, R.P., Fawzi, A.B.,Graziano, M.P., Sybertz, E.J., Strader, C.D., and Davis, H.R., Jr. (1997). Diet-induced obese mice develop peripheral, but not central, resistance to leptin.J. Clin. Invest. 99, 385–390.

Van Itallie, C.M., and Anderson, J.M. (1997). Occludin confers adhesivenesswhen expressed in fibroblasts. J. Cell Sci. 110, 1113–1121.

Wang, W., Dentler, W.L., and Borchardt, R.T. (2001). VEGF increases BMECmonolayer permeability by affecting occludin expression and tight junctionassembly. Am. J. Physiol. Heart Circ. Physiol. 280, H434–H440.

Wigmore, S.J., Barber, M.D., Ross, J.A., Tisdale, M.J., and Fearon, K.C.(2000). Effect of oral eicosapentaenoic acid on weight loss in patients withpancreatic cancer. Nutr. Cancer 36, 177–184.

Wilsey, J., Sergei, Z., Victor, P., and Philip, J.S. (2003). Central leptin genetherapy fails to overcome leptin resistance associated with diet-inducedobesity. Am. J. Physiol. Regul. Integr. Comp. Physiol. 285, 1011–1020.

Wong, V., and Gumbiner, B.M. (1997). A synthetic peptide corresponding tothe extracellular domain of occludin perturbs the tight junction permeabilitybarrier. J. Cell Biol. 136, 399–409.

Wu-Peng, X.S., Chua, S.C., Jr., Okada, N., Liu, S.M., Nicolson, M., andLeibel, R.L. (1997). Phenotype of the obese Koletsky (f) rat due to Tyr796Stop mutation in the extracellular domain of the leptin receptor (Lepr): evi-dence for deficient plasma-to-CSF transport of leptin in both the Zuckerand Koletsky obese rat. Diabetes 46, 513–518.

Yamagata, K., Tagami, M., Takenaga, F., Yamori, Y., Nara, Y., and Itoh, S.(2003). Polyunsaturated fatty acids induce tight junctions to form in braincapillary endothelial cells. Neuroscience 116, 649–656.

Zastrow, O., Seidel, B., Kiess, W., Thiery, J., Keller, E., Bottner, A., andKratzsch, J. (2003). The soluble leptin receptor is crucial for leptin action:evidence from clinical and experimental data. Int. J. Obes. Relat. Metab.Disord. 27, 1472–1478.

Zhang, Y., Proenca, R., Maffei, M., Barone, M., Leopold, L., and Friedman,J.M. (1994). Positional cloning of the mouse obese gene and its humanhomologue. Nature 372, 425–432.

341

molecular mechanisms associated with leptin resistance: n-3 polyunsaturated fatty acids induce...

Documents