Modulation of Cutaneous Fatty Acid-Binding Protein Mrna

Modulation of Cutaneous Fatty Acid-binding Protein mRNA Expression in Rat Adipose Tissues by Hereditary Obesity and Dietary Fats Masayo Kushiro, Yoko Takahashi and Takashi Ide＊ Laboratory of Nutritional Function, Division of Food Functionality, National Food Research Institute (2-1-12 Kannondai, Tsukuba 305-8642, JAPAN)

Abstract: Cutaneous fatty acid-binding protein (C-FABP) is a member of the intracellular lipid-binding protein multigene family expressed in various tissues. A high level of C-FABP mRNA in adipose tissue has been reported, but its physiological significance in regulating adipose tissue function is not clear. To obtain insights into the role of C-FABP in adipose tissue, we studied the obesity-related and dietary fat-related changes of C-FABP mRNA expression in adipose tissues. C-FABP mRNA levels in interscapular brown adipose tissue, and epididymal and perirenal white adipose tissues were higher in Zucker fatty rats than in lean controls despite that the difference in brown adipose tissue was not significant. Fish oil compared to palm and safflower oils significantly reduced the mRNA level of C-FABP in brown adipose tissue and epididymal and perirenal white adipose tissues in Sprague-Dawley rats except for one occasion. Our study demonstrated that C-FABP is a protein whose mRNA expression is easily modified by hereditary obesity and the type of dietary fat. Therefore, C-FABP may play a significant role in regulating adipocyte function in response to changes in nutritional conditions.

Key words: fatty acid-binding protein, Zucker fatty rat, dietary fats, fish oil

1 INTRODUCTION skin but also in a variety of other tissues and cells5,8,9,12-23). Cutaneous fatty acid-binding protein (C-FABP), also These observations indicated that C-FABP is not only known as psoriasis-associated FABP, keratinocyte FABP, involved in maintaining the function of the skin, but also epidermal FABP, DA11 or mal1, is a member of the intra- plays a role in regulating lipid metabolism in the other tis- cellular lipid-binding protein multigene family1). C-FABP sues. Appreciable levels of C-FABP in adipose tissue have has been isolated from human skin, sequenced, and charac- been reported not only in mice5,13,16,17,22) but also in terized2-4). Homologues of its gene have been cloned from humans19,20). Adipocyte lipid-binding protein (ALBP) is the various animal species including mice5,6), rats7), and major cytosolic fatty acid-binding protein in adipose tissue, bovines8). The exact role of C-FABP in skin has not been and C-FABP is much less abundant than ALBP in this tis- clarified. An analysis of the phenotype of C-FABP-gene sue13,16,17,22). However, some studies raised the possibility knockout mice9) indicated that C-FABP is involved in main- that a change in the expression of C-FABP in adipose tis- taining the water permeability barrier function of the epi- sue is associated with altered adipocyte function, and the dermis through participation in the synthesis of epidermal development of obesity and insulin resistance17,19,20,22). How- lipids. Unlike other fatty acid-binding proteins, C-FABP ever, information on the nutritional regulation of the contains a large number of cysteine residues10,11), and it has expression of C-FABP in adipose tissue has been lacking. been suggested that C-FABP functions as an antioxidant in In this context, we examined the nutritional regulation of the skin by scavenging peroxidized lipids through the C-FABP gene expression in adipose tissues. We found that covalent modification of cysteine residues11). the mRNA expression of C-FABP was strongly up-regulat- Studies have shown that this protein occurs not only in ed in adipose tissue of Zucker fatty rats compared to lean

＊ Correspondence to: Takashi Ide, Laboratory of Nutritional Function, Division of Food Functionality, National Food Research Institute, 2-1-12 Kannondai, Tsukuba 305-8642, JAPAN E-mail: [email protected] Accepted June 18, 2007 (received for review May 7, 2007) Journal of Oleo Science ISSN 1345-8957 print / ISSN 1347-3352 online http://jos.jstage.jst.go.jp/en/

533 M. Kushiro, Y. Takahashi and T. Ide

controls. We also showed that mRNA expression of C- ratory animals. FABP in white adipose tissue was modified by the type of dietary fat. These observations suggested that C-FABP in 2.2 RNA analysis adipose tissue plays an important role in regulating Animals were sacrificed by withdrawing blood from the adipocyte function in response to changes in nutritional abdominal aorta under diethyl ether-anesthesia and tissues conditions. were excised. RNA in various tissues was extracted by the acid guanidium thiocyanate-phenol-chloroform method25). RNA was reverse-transcribed with random hexamers using MultiScribe reverse transcriptase (Applied Biosys- 2 EXPERIMENTAL tems) to generate cDNA. mRNA levels were analyzed by a 2.1 Animals and diets quantitative real-time PCR using an Applied Biosystems Male Zucker lean and obese (fa/fa) rats and Sprague- Prism 7000 sequence detection system (Applied Biosys- Dawley rats were purchased from Charles River, Kana- tems, Foster City, CA, USA) as detailed previously26). The gawa, Japan. The animals were housed individually in a nucleotide sequences of sense and antisense primers, and room with controlled temperature (20-22℃), humidity (55- probes designed using Primer Express Software (Applied 65%), and lighting (lights on from 07:00 to 19:00 h), and fed Biosystems) according to the nucleotide sequences avail- a laboratory chow (Type NMF, Oriental Yeast Co., Tokyo, able in the GenBank data base were 5’-TGGCCAAACCA- Japan) for 1 week at least to acclimatize to our housing con- GACTGCAT-3’, 5’-CTCTCCCAAGGTGCAAGAAAA-3’, and ditions. In a nutritional study to examine the effects of 5’-TACCCTCGACGGCAACAACCTCACC-3’ for C-FABP dietary fat types on gene expression, Sprague-Dawley rats (accession number, S69874), and 5’-ACACCGAGATTTC- obtained at 4 weeks of age were divided into groups after 1 CTTCAAACTG-3’, 5’-CCATCCCACTTCTGCACATG-3’, week of acclimatization and were fed purified experimental and 5’-CGTGGAATTCGATGAAATCACCC-3’ for ALBP diets containing 20% of different types of dietary fats for 21 (U75581), respectively. The levels of 18S rRNA in each d. The composition of the experimental diet was (in weight cDNA sample were also analyzed using primers and probes %): casein, 20; corn starch, 15; fat, 20; cellulose, 2; mineral supplied from Applied Biosystems (TaqMan ribosomal RNA mixture,24) 3.5; vitamin mixture,24) 1.0; DL-methionine, 0.3; control reagents). Abundances of mRNAs for C-FABP and choline bitartrate, 0.2 and sucrose, 38. Either palm oil (sat- ALBP were calculated as the ratio to the 18S rRNA level in urated fat), safflower oil rich in linoleic acid or fish oil rich each cDNA sample. in n-3 fatty acids was used as dietary fat sources in this experiment. Fatty acid compositions of these dietary fats 2.3 Analysis of serum levels of lipids, glucose, and insulin, are shown in Table 1. This study was approved by the and fatty acid content and composition in adipose tis- review board of animal ethics of our institute and we fol- sue lowed the institute’s guidelines in the care and use of labo- Serum triacylglycerol, cholesterol, free fatty acid, and glucose levels were analyzed by using commercial enzyme kits (Wako Pure Chemical, Osaka, Japan). The serum Table 1 Fatty Acid Composition of Dietary Fats. insulin concentration was analyzed with a commercial ELISA kit (Morinaga Co., Tokyo, Japan). Lipids in adipose Fatty acids Dietary fats tissues were extracted and purified27). The fatty acid con- (Weight %) Palm oil Safflower oil Fish oil centration and compositions of the lipid extract were ana- 14:0 1.1 0.1 2.2 lyzed by GLC using a 1.4BDS capillary column (40 m × 16:0 45.3 6.9 13.1 0.25 mm, Shimadzu Co., Kyoto, Japan) and heptadecanoic 16:1 (n-7) 0.1 － 6.9 acid was used as an internal standard. 18:0 4.4 2.4 3.0 2.4 Statistical analysis 18:1 (n-9) 39.1 13.0 21.5 The values were expressed as means ± SE. StatView for 18:2 (n-6) 9.4 77.3 2.1 Macintosh (SAS Institute Inc., Cary, NC, USA) was used 18:3 (n-3) 0.3 0.2 0.7 for statistical analyses. Student’s t-test was used to detect 20:4 (n-6) －－ 3.0 significant differences between the means in Zucker fatty 20:5 (n-3) －－ 10.0 and lean rats. In the experiment to examine the effect of 22:4 (n-6) －－ 0.7 dietary fat types on gene expression of C-FABP and ALBP, 22:5 (n-6) －－ 1.7 data were analyzed by one-way ANOVA, and significant 22:5 (n-3) －－ 2.6 differences between the means were inspected using a 22:6 (n-3) －－ 32.6 Tukey-Kramer post-hoc analysis. The differences were considered to be significant when p<0.05.

534 J. Oleo Sci. 56, (10) 533-541 (2007) Fatty Acid-binding Protein mRNA Expression in Adipose Tissue

3 RESULTS nal white adipose tissues were found to be appreciably 3.1 Tissue distribution of mRNAs for C-FABP and ALBP high. Although studies indicated that C-FABP mRNA is expressed in various tissues5,8,9,12-23), detailed information on 3.2 mRNA expression of C-FABP and ALBP in Zucker the tissue distribution of the mRNA for this protein has lean and fatty rats been scarce. We therefore analyzed the mRNA abundance To examine the physiological significance of C-FABP in of C-FABP in various tissues of 10-week-old Zucker lean controlling adipocyte metabolism, we analyzed the obesity- rats fed a laboratory chow using the quantitative real-time related changes in the mRNA expression of C-FABP using PCR method (Fig. 1). We also analyzed the abundance of Zucker lean and fatty rats (fa/fa). The rats, maintained on mRNAs for ALBP among tissues. RNAs were extracted a laboratory chow, were sacrificed at 10 weeks of age. Body from various tissues of 5 rats. Equal amounts of RNA from weights at the time of killing were 311 ± 4 and 422 ± 8 g each tissue were combined and cDNA was synthesized. for lean and fatty rats, respectively (p<0.01). The weights mRNA levels in cDNA samples were quantified using spe- of interscapular brown adipose tissue, and epididymal and cific primers and probes. The level of 18S rRNA in each perirenal white adipose tissues were significantly higher in cDNA sample was also analyzed with TaqMan ribosomal fatty than in lean Zucker rats (Table 2). RNA control reagents. mRNA abundances were calculated Serum triacylglycerol, cholesterol, free fatty acid and as the ratio to the 18S rRNA level in each cDNA sample glucose concentrations were significantly (p<0.01) higher and expressed as percentages, assigning the value in epi- in fatty rats (581 ± 68, 283 ± 10, 85.3 ± 6.1 and 966 ± 45 didymal white adipose tissue as 100. C-FABP mRNA was mmol/dl, respectively) than in lean rats (144 ± 15, 165 ± 8, detected in various tissues, the highest level being found in 44.1 ± 3.0 and 714 ± 19 mmol/dl, respectively). The serum skin. Substantial levels were found in brown adipose tissue, insulin concentration was more than 10 times higher in and perirenal and epididymal white adipose tissues, being Zucker fatty rats (3565 ± 935 ng/dl) than in lean rats comparable to those in liver, brain, heart, kidneys, spleen, (241 ± 49 ng/dl) (p<0.01). intestine, lung, and testis. The expression of ALBP was We measured levels of mRNAs for C-FABP and ALBP in mostly confined to interscapular brown adipose tissue, and lean and fatty Zucker rats. mRNA abundances were calcu- epididymal and perirenal white adipose tissues. Our results lated as the ratio to the 18S rRNA level in each cDNA sam- therefore showed that compared to ALBP, C-FABP is more ple and expressed them as percentages assigning the value ubiquitously expressed in various tissues. C-FABP mRNA in epididymal adipose tissue of Zucker lean rats as 100 levels in brown adipose tissue and epididymal and perire- (Table 2). C-FABP mRNA levels in brown adipose tissue, epididymal and perirenal white adipose tissues, and skin were 2.1-, 2.1-, 8.7- and 2.9-times higher in Zucker fatty than lean rats though the difference in brown adipose tissue was not statistically significant. The levels of mRNAs for ALBP in adipose tissues were also higher in obese than in lean rats. Fold-increases in ALBP mRNA levels due to hereditary obesity were 1.4, 1.7 and 5.8 for brown adipose tissue, and epididymal and perirenal white adipose tissues, respectively. Therefore, C-FABP mRNA compared to ALBP mRNA was more affected by hereditary obesity. Correlations between weights of interscapular brown, epididymal white, and perirenal white adipose tissues and the mRNA levels of C-FABP and ALBP are also shown in Table 2. Positive correlations were observed between weights of epididymal and perirenal white adipose tissues, Fig. 1 mRNA Expression of Cutaneous Fatty Acid- and mRNA levels for C-FABP and ALBP. The correlation Binding Protein (C-FABP), and Adipocyte Lipid- between interscapular brown adipose tissue weights and binding Protein (ALBP) in Various Tissues of mRNA levels of C-FABP, but not ALBP, was also significant. Zucker Lean Rats. mRNA levels were quantified by real-time PCR using primers and probes specific to 3.3 Effects of dietary fat on mRNA expression of C-FAB C-FABP and ALBP mRNAs. mRNA abundances and ALBP were calculated as the ratio to the 18S rRNA level It has been demonstrated that types of fat influence gene and expressed as percentages, assigning the value in expression in adipose tissue28,29,30,31). Therefore, in the sec- epididymal white adipose tissue as 100. ond experiment, we examined the role of dietary fat types

535 J. Oleo Sci. 56, (10) 533-541 (2007) M. Kushiro, Y. Takahashi and T. Ide

Table 2 mRNA Levels of Cutaneous Fatty Acid-Binding Protein (C-FABP) and Adipocyte Lipid-Binding Protein (ALBP) in Adipose Tissues and Skin in Zucker Lean and Fa/Fa Rats. Correlations Tissue weight mRNA level (%) Tissue weight vs Tissue weight vs Tissues and animals (g/100 g body C-FABP mRNA ALBP mRNA weight) C-FABP ALBP r p value r p value Interscapular brown adipose tissue Lean 0.16 ± 0.01 36.0 ± 4.0 54.6 ± 5.3 0.621 0.018 0.456 0.100 fa/fa 0.34 ± 0.03＊ 74.2 ± 16.1 75.5 ± 21.2

Epididymal white adipose tissue Lean 1.23 ± 0.03 100 ± 13 100 ± 22 0.808 <0.001 0.588 0.027 fa/fa 2.84 ± 0.17＊ 210 ± 7＊ 172 ± 11＊

Perirenal white adipose tissue Lean 0.77 ± 0.04 27.8 ± 2.6 24.1 ± 5.6 0.902 <0.001 0.926 <0.001 fa/fa 2.88 ± 0.09＊ 243 ± 25＊ 139 ± 10＊

Skin Lean ND 134 ± 25 ND ―― fa/fa ND 392 ± 54＊ ND Values are means ± SE for 8 rats. mRNA levels were expressed as percentages assigning the value for the epididymal white adipose tissue of lean rats as 100. ＊Values are significantly different from those in lean Zucker rats at p <0.05. ND = not determined.

in regulating the mRNA expression of C-FABP in adipose not only reduced white adipose tissue weights but also sig- tissues. Sprague-Dawley rats were fed diets containing nificantly reduced fatty acid concentrations in the tissues. 20% palm oil (saturated fat), safflower oil (a fat rich in As a consequence, fatty acid contents in these white adi- linoleic acid), or fish oil (a fat rich in n-3 fatty acids) for 21 pose tissues of rats fed fish oil were approximately one-half d. No significant differences were observed in weight gain those observed in the animals fed palm and safflower oils. among the groups (168 ± 6, 161 ± 7, and 165 ± 4 g/21 d Compositions of adipose tissue fatty acids generally for rats fed palm, safflower, and fish oil diets, respectively). reflected those in the diets (data not shown). Types of Daily energy intake was the same among the groups dietary fat also modified fatty acid compositions (data not (82.7 ± 1.9, 76.8 ± 2.5, and 76.1 ± 1.2 kcal/d for the ani- shown) but did not affect the concentrations in skin. mals fed palm, safflower, and fish oil diets, respectively). mRNA levels of C-FABP and ALBP calculated as the There were no significant differences in the weight of ratio to the 18S rRNA level in each cDNA sample were interscapular brown adipose tissue among the groups expressed as percentages assigning the value in epididy- (Table 3). However, fish oil compared to palm and safflower mal white adipose tissue of rats fed a palm oil diet as 100. oils significantly reduced the fatty acid concentration in Safflower oil compared to palm oil did not significantly brown adipose tissue. Fatty acid content expressed as modify the mRNA expression of C-FABP in various adipose mg/100 g body weight in this tissue became significantly tissues despite that the values in perirenal white adipose lower in rats fed fish oil than in those fed palm oil. Epididy- tissue and skin were considerably lower in rats fed safflow- mal and perirenal white adipose tissues weighed less in er oil than in those fed palm oil. Fish oil compared to palm rats fed fish oil than in those fed palm oil. Also, these val- and safflower oils significantly decreased mRNA levels of ues tended to be lower in rats fed fish oil than in those fed C-FABP in brown adipose tissue, epididymal white adipose safflower oil. Fish oil compared to palm and safflower oils tissue, and skin. The mRNA level of C-FABP in perirenal

536 J. Oleo Sci. 56, (10) 533-541 (2007) Fatty Acid-binding Protein mRNA Expression in Adipose Tissue alue v ALBP mRNA rp issue fatty acid content vs issue fatty T us Fatty Acid-Binding us Fatty alue v rp C-FABP mRNA C-FABP issue fatty acid content vs issue fatty Correlations T as 100. alue v rp issue weight vs ALBP mRNA T alue v －－－－ rp issue weight vs 0.144 0.524 0.017 0.9400.389 0.163 0.074 0.001 0.467 0.9960.612 0.161 0.460 0.002 0.473 0.289 0.031 0.192 0.037 0.543 0.870 0.009 0.263 0.236 C-FABP mRNA C-FABP T a a a a a a a a a <0.05. 2.73 2.3 2.58 26 22 12.1 3.6 6.1 6.4 p ND ND ND ± ± ± ± ± ± ± ± ± 100 103 8.73 12.0 9.56 66.4 24.6 23.8 17.1 a b b b b a ab a a b b b 4.2 2.7 0.61 6 19 6.3 12.8 8.1 7.8 11.0 4.8 23 mRNA level (%) level mRNA ± ± ± ± ± ± ± ± ± ± ± ± C-FABP ALBP 100 110 108 14.5 13.8 4.65 57.2 61.6 35.0 25.9 69.7 32.6 b b b b ab a b a a 6.6 8.1 6.0 121 107 67 31 183 44 ± ± ± ± ± ± ± ± ± ND ND ND 945 446 982 595 87.9 86.2 67.5 1011 1253 issue fatty acid issue fatty weight) T content (mg/100 g body a c a a a a a b b b b b 17 18 37 19 36 15 12 17 35 24 30 45 ± ± ± ± ± ± ± ± ± ± ± ± 164 367 934 555 704 161 145 495 498 729 841 847 issue fatty acid issue fatty T concentration (mg/g tissue) a a a b a b a a a 0.01 0.01 0.01 0.14 0.10 0.0 0.11 0.14 0.07 SE for 7-8 rats. ± ± ± ± ± ± ± ± ± ND ND ND ± issue weight weight) 0.18 0.17 0.19 1.37 1.03 0.80 1.49 1.12 0.86 T (g/100 g body Effect of Dietary Fat Types on Weights, and Fatty Acid Concentration and Content of Adipose Tissues, and mRNA Levels of Cutaneo Levels and mRNA Acid Concentration and Content of Adipose Tissues, and Fatty on Weights, Types of Dietary Fat Effect Protein (C-FABP) and Adipocyte Lipid-Binding Protein (ALBP) in Adipose Tissues and Skin. Lipid-Binding Protein (ALBP) in Adipose Tissues and Adipocyte Protein (C-FABP) alues in a column in each tissue with different superscripts differ significantly at significantly superscripts differ alues in a column each tissue with different issues and alm oil alm oil alm oil alm oil V able 3 alues are means Interscapular brown adipose tissue Interscapular brown P oil Safflower Fish oil T dietary fats Epididymal white adipose tissue P oil Safflower Fish oil Perirenal white adipose tissue P oil Safflower Fish oil Skin P oil Safflower Fish oil T V for the epididymal white adipose tissue of rats fed palm oil diet as percentages assigning the value were expressed levels mRNA ab ND = not determined.

537 J. Oleo Sci. 56, (10) 533-541 (2007) M. Kushiro, Y. Takahashi and T. Ide

Table 4 Effect of Dietary Fat Types on Serum Lipid, Glucose and Insulin Concentrations. Dietary fats Serum components Palm oil Safflower oil Fish oil

Triacylglycerol (mmol/dl) 347 ± 48b 157 ± 26a 96.7 ± 22.5a Cholesterol (mmol/dl) 272 ± 22b 283 ± 11b 147 ± 5a Phospholipid (mmol/dl) 315 ± 23b 275 ± 13b 167 ± 7a Free fatty acid (mmol/dl) 187 ± 40b 113 ± 22ab 52.0 ± 6.9a Glucose (mmol/dl) 731 ± 33 790 ± 74 748 ± 53 Insulin (ng/dl) 146 ± 30ab 213 ± 48b 70.1 ± 19.5a Values are means ± SE for 7-8 rats. abValues in a line with different superscripts differ significantly at p < 0.05. white adipose tissue was significantly lower in rats fed fish ization of C-FABP may exist between mice and rats. Krieg oil than in the animals fed palm oil but not safflower oil. et al.5) indicated that C-FABP mRNA was relatively abun- However, types of dietary fats did not modify the mRNA dant in adipose tissue, but did not supply information on levels of ALBP in various adipose tissues. the location of the adipose tissue they analyzed. We found Significant correlations were observed between perire- that the level of C-FABP mRNA depends on the location of nal white adipose tissue weight and the mRNA level of C- the fat pads. The level was high in the epididymal depot, FABP in this tissue. Correlations between tissue fatty acid being more than 70% of the level observed in skin. Also, content in epididymal and perirenal white adipose tissues relatively high levels were detected in perirenal white adi- and the mRNA level of C-FABP in respective tissues were pose tissue and interscapular brown adipose tissue indicat- also significant. However, no significant correlations ing that C-FABP plays a crucial role in regulating the func- between weights of tissue or their fatty acid contents and tion of these fat deposits. It has been demonstrated that ALBP mRNA levels were observed. the expression of ALBP is mostly confined to adipose tis- The serum triacylglycerol concentration was significant- sues32) as confirmed in the present study. ly higher in rats fed palm oil than in the animals fed saf- ALBP is the most abundant fatty acid-binding protein in flower and fish oil (Table 4). The difference was not signifi- adipose tissue1,13,16,17,22). C-FABP has also been detected in cant between the latter two groups. The serum cholesterol adipose tissue in mice13,16,17,22), though at much lower levels concentration was significantly lower in rats fed fish oil than ALBP. It has been demonstrated that targeted disrup- than in the animals fed palm and safflower oils. The serum tion of the ALBP gene in mice provides significant protec- free fatty acid concentration was lowest in rats fed fish oil tion from hyperinsulinemia and insulin resistance in the among the groups, and a significant difference was seen context of both dietary33) and genetic34) obesity. However, in relative to the rats fed palm oil but not safflower oil. There ALBP knockout mice, a huge compensatory increase in C- were no significant differences in serum glucose concen- FABP expression occurs in adipose tissue17,33-35). Therefore, trations among groups. The serum insulin concentration there is the possibility that the phenotype observed in was significantly lower in rats fed fish oil than in the ani- ALBP-null mice results from the increased expression of mals fed safflower oil, but not in those fed palm oil. C-FABP rather than the deficiency of ALBP. With regard to this, knockout of the C-FABP-gene resulted in increased systemic insulin sensitivity in mice, while overexpression of C-FABP had the opposite effect22). Therefore, both ALBP 4 DISCUSSION and C-FABP in adipose tissue may be involved in the regu- C-FABP was first isolated from human skin2-4). However, lation of whole body glucose metabolism and insulin sensi- its mRNA was later found to be expressed in various tis- tivity. sues including adipose tissue5,8,9,12-23). Krieg et al.5) examined It has been well demonstrated that hereditary obesity levels of C-FABP mRNA in mouse tissues, and found that results in impaired glucose tolerance and insulin sensitivi- the mRNA was most abundant in stratified epithelia of the ty in experimental animals36). In fact we confirmed that epidermis and tongue. Relatively large amounts were found serum glucose and insulin concentrations were higher in in adipose and mammary tissues, and small amounts in Zucker fatty rats than in lean rats. We observed, in the pre- heart, brain, liver, spleen, muscle, and lung. Our observa- sent study, that mRNA levels not only of ALBP but also of tion indicated a more widespread distribution of the mRNA C-FABP in white adipose tissue were much higher in obese among tissues in rats. Therefore, differences in the local- Zucker rats than in lean controls. This observation may be

538 J. Oleo Sci. 56, (10) 533-541 (2007) Fatty Acid-binding Protein mRNA Expression in Adipose Tissue

consistent with a role for C-FABP22) and ALBP33-35) in regu- level was not significant, correlations between fatty acid lating insulin resistance. Meanwhile, it has been demon- contents and C-FABP mRNA levels in both perirenal and strated that dietary fish oil improves glucose tolerance and epididymal white adipose tissues were found to be signifi- stimulates insulin-dependent glucose transport and cant. Therefore, it is possible that C-FABP but not ALBP is metabolism in isolated adipocytes37,38). In the present study, involved in dietary fat-dependent changes in lipid accumu- we did not find any significant alteration by dietary fish oil lation in white adipose tissues. of the serum glucose concentration, but found that the It is rather unexpected that the changes in C-FABP serum insulin concentration was considerably lower in rats mRNA expression in skin caused by hereditary obesity and fed fish oil than in the animals fed palm and safflower oils. dietary fats generally paralleled those in adipose tissues, In this situation, fish oil compared to palm oil and safflow- because the role of each of these tissues in maintaining er oil effectively reduced C-FABP mRNA expression in adi- physiology differs greatly. This observation at least indi- pose tissues. Therefore, it is possible that down-regulation cates the existence of a common signaling pathway regu- of C-FABP expression contributes to the fish oil-mediated lating C-FABP expression in these tissues. improvement of glucose tolerance. As fish oil failed to mod- We showed in the present study that hereditary obesity ify the ALBP mRNA level, this fatty acid-binding protein and types of dietary fat caused significant changes in may not be involved in this outcome. It has been also mRNA levels of C-FABP in adipose tissues. Although the demonstrated that dietary fish oil did not modify the ALBP physiological significance of these changes is not necessar- mRNA level in mouse epididymal white adipose tissue26). ily clear at present, it seems that C-FABP is involved in There is the possibility that changes in the expression of regulating adipocyte function in response to changes in C-FABP and ALBP caused by nutritional and physiological nutritional conditions. conditions alter fat pad mass despite that neither knockout of the ALBP17,34,35) or C-FABP22) gene nor overexpression of C-FABP22) in mice is associated with any alteration in fat pad mass. We observed in the present study that mRNA ACKNOWLEDGEMENT levels of C-FABP and ALBP in epididymal and perirenal This study was supported by a grant from the Ministry white adipose tissues were much higher in fatty than in of Agriculture, Forestry and Fisheries (MAFF)’s Food lean Zucker rats. Also, strong positive correlations were Research Project “Integrated Research on Safety and observed between mRNA levels of these proteins and Physiological Function of Food” and by a grant from the weights of epididymal and perirenal white adipose tissues. Ministry of Education, Science, Sports and Culture, Grant- Therefore, it is likely that these fatty acid-binding proteins in-Aid for Scientific Research Young Scientists (B), no. are involved in the development of obesity. However, the 16780104. observations made in rats fed different types of fats did not support the notion that ALBP at least is involved in the regulation of fat pad mass. In the present study, we confirmed a previous finding30) that fish oil compared to palm References oil, reduced the weight of both epididymal and perirenal 1. Zimmerman, A.W.; Veerkamp, J.H. New insights into white adipose tissue. In addition, we showed that fish oil, the structure and function of fatty acid-binding pro- compared to palm and safflower oils, significantly lowered teins. Cell Mol. Life Sci. 59, 1096-1116 (2002). the fatty acid concentrations and contents in these white 2. Madsen, P.; Rasmussen, H.H.; Leffers, H.; Honore, B.; adipose tissues. Therefore, it is apparent that fish oil rich Celis, J.E. Molecular cloning and expression of a novel in n-3 polyunsaturated fatty acids, compared to both satu- keratinocyte protein (Psoriasis-associated fatty acid- rated fat and a fat rich in n-6 polyunsaturated fatty acids, binding protein ([PA-FABP]) that is highly up-regulated attenuates lipid accumulation in white adipose tissues. in psoriatic skin and that shares similarity to fatty However, fish oil compared to other fats failed to modify acid-binding proteins. J. Invest. Dermatol. 99, 299-305 ALBP mRNA expression in various adipose tissues, and no (1992). significant correlations were observed between the weight 3. Siegenthaler, G.; Hotz, R.; Chatellard-Gruaz, D.; Jaconi, of white adipose tissues or their fatty acid content and S.; Saurat, J.H. Characterization and expression of a ALBP mRNA levels. In contrast, we showed that fish oil novel human fatty acid-binding protein: the epidermal compared to palm and safflower oils down-regulated the type (E-FABP). Biochem. Biophys. Res. Commun. 190, mRNA expression of C-FABP in brown and white adipose 482-487 (1993). tissues. We observed that a correlation between perirenal 4. Siegenthaler, G.; Hotz, R.; Chatellard-Gruaz, D.; Didier- white adipose tissue weight and the C-FABP mRNA level jean, L.; Hellman, U.; Saurat, J.H. Purification and char- was significant. Although a correlation between epididy- acterization of the human epidermal fatty acid-binding mal white adipose tissue weight and the C-FABP mRNA protein: Localization during epidermal cell differentia-

539 J. Oleo Sci. 56, (10) 533-541 (2007) M. Kushiro, Y. Takahashi and T. Ide

tion in vivo and in vitro. Biochem. J. 302, 363-371 lar lipid binding proteins of adipocytes. Mol. Cell (1994). Biochem. 192, 33-40 (1999). 5. Krieg, P.; Feil, S.; Furstenberger, G.; Bowden, G.T. 17. Coe, N.R.; Simpson, M.A.; Bernlohr, D.A. Targeted dis- Tumor-specific overexpression of a novel keratinocyte ruption of the adipocyte lipid-binding protein (aP2 Pro- lipid-binding protein. Identification and characteriza- tein) gene impairs fat cell lipolysis and increases cellu- tion of a cloned sequence activated during multistage lar fatty acid levels. J. Lipid Res. 40, 967-972 (1999). carcinogenesis in mouse skin. J. Biol. Chem. 268, 18. Allen, G.W.; Liu, J.; Kirby, M.A.; De Leon, M. Induction 17362-17369 (1993). and axonal localization of epithelial/epidermal fatty 6. Hertzel, A.V.; Bernlohr, D.A. Cloning and chromosomal acid-binding protein in retinal ganglion cells are asso- location of the murine keratinocyte lipid-binding pro- ciated with axon development and regeneration. J. tein gene. Gene 221, 235-243 (1998). Neurosci. Res. 66, 396-405 (2001). 7. Watanabe, R.; Fujii, H.; Odani, S.; Sakakibara, J.; 19. Fisher, R.M.; Eriksson, P.; Hoffstedt, J.; Hotamisligil, Yamamoto, A.; Ito, M.; Ono, T. Molecular cloning of a G.S.; Thorne, A.; Ryden, M.; Hamsten, A.; Arner, P. cDNA encoding a novel fatty acid-binding protein from Fatty acid binding protein expression in different adi- rat skin. Biochem. Biophys. Res. Commun. 200, 253- pose tissue depots from lean and obese individuals. 259 (1994). Diabetologia 44, 1268-1273 (2001). 8. Kingma, P.B.; Bok, D.; Ong, D.E. Bovine epidermal 20. Fisher, R.M.; Hoffstedt, J.; Hotamisligil, G.S.; Thorne, fatty acid-binding protein: Determination of ligand A.; Ryden, M. Effects of obesity and weight loss on the specificity and cellular localization in retina and testis. expression of proteins involved in fatty acid Biochemistry 37, 3250-3257 (1998). metabolism in human adipose tissue. Int. J. Obes. 9. Owada, Y.; Suzuki, I.; Noda, T.; Kondo, H. Analysis on Relat. Metab. Disord. 26, 1379-1385 (2002). the phenotype of E-FABP-gene knockout mice. Mol. 21. Kitanaka, N.; Owada, Y.; Abdelwahab, S.A.; Iwasa, H.; Cell Biochem. 239, 83-86 (2002). Sakagami, H.; Watanabe, M.; Spener, F.; Kondo, H. Spe- 10. Odani, S.; Namba, Y.; Ishii, A.; Ono, T.; Fujii, H. Disul- cific localization of epidermal-type fatty acid binding fide bonds in rat cutaneous fatty acid-binding protein. protein in dendritic cells of splenic white pulp. His- J. Biochem. 128, 355-361 (2000). tochem. Cell Biol. 120, 465-473 (2003). 11. Bennaars-Eiden, A.; Higgins, L.; Hertzel, A.V.; Kap- 22. Maeda, K.; Uysal, K.T.; Makowski, L.; Gorgun, C.Z.; phahn, R.J.; Ferrington, D.A.; Bernlohr, D.A. Covalent Atsumi, G.; Parker, R.A.; Bruning, J.; Hertzel, A.V.; modification of epithelial fatty acid-binding protein by Bernlohr, D.A.; Hotamisligil, G.S. Role of the fatty acid 4-hydroxynonenal in vitro and in vivo. Evidence for a binding protein mal1 in obesity and insulin resistance. role in antioxidant biology. J. Biol. Chem. 277, 50693- Diabetes 52, 300-307 (2003). 50702 (2002). 23. Yun, X.; Nourani, M.R.; Abdelwahab, S.A.; Kitanaka, N.; 12. De Leon, M.; Welcher, A.A.; Nahin, R.H.; Liu, Y.; Ruda, Owada, Y.; Spener, F.; Iwasa, H.; Takahashi, A.; Kondo, M.A.; Shooter, E.M.; Molina, C.A. Fatty acid binding H. Differential localization of brain-type and epidermal- protein is induced in neurons of the dorsal root ganglia type fatty acid binding proteins in the adrenal gland of after peripheral nerve injury. J. Neurosci. Res. 44, 283- mice. Tohoku J. Exp. Med. 203, 77-86 (2004). 292 (1996). 24. Am. Inst. Nutr. Report of the american institute of 13. Bernlohr, D.A.; Coe, N.R.; Simpson, M.A.; Hertzel, A.V. nutrition Ad Hoc committee on standards for nutrition- Regulation of gene expression in adipose cells by al studies. J. Nutr. 107, 1340-1348 (1977). polyunsaturated fatty acids. Adv. Exp. Med. Biol. 422, 25. Chomczynski, P.; Sacchi, N. Single-step method of 145-156 (1997). RNA isolation by acid guanidium thiocyanate-phenol- 14. Liu, Y.; Molina, C.A.; Welcher, A.A.; Longo, L.D.; De chloroform extraction. Anal. Biochem. 162, 156-159 Leon, M. Expression of DA11, a neuronal-injury- (1987). induced fatty acid binding protein, coincides with axon 26. Ide, T. Interaction of fish oil and conjugated linoleic growth and neuronal differentiation during central ner- acid in affecting hepatic activity of lipogenic enzymes vous system development. J. Neurosci. Res. 48, 551- and gene expression in liver and adipose tissue. Dia- 562 (1997). betes 54, 412-423 (2005). 15. Masouye, I.; Hagens, G.; Van Kuppevelt, T.H.; Madsen, 27. Folch, J.; Lees, M.; Sloane-Stanley, G.H. A simple P.; Saurat, J.H.; Veerkamp, J.H.; Pepper, M.S.; Siegen- method for the isolation and purification of total lipids thaler, G. Endothelial cells of the human microvascula- from animal tissues. J. Biol. Chem. 226, 497-509 (1957). ture express epidermal fatty acid-binding protein. Circ. 28. Okuno, M.; Kajiwara, K.; Imai, S.; Kobayashi, T.; Res. 81, 297-303 (1997). Honma, N.; Maki, T.; Suruga, K., Goda, T.; Takase, S.; 16. Simpson, M.A.; LiCata, V.J.; Coe, N.R.; Bernlohr, D.A. Muto, Y.; Moriwaki, H. Perilla oil prevents the exces- Biochemical and biophysical analysis of the intracellu- sive growth of visceral adipose tissue in rats by down-

540 J. Oleo Sci. 56, (10) 533-541 (2007) Fatty Acid-binding Protein mRNA Expression in Adipose Tissue

regulating adipocyte differentiation. J. Nutr. 127, 1752- tein. Science 274, 1377-1379 (1996). 1757 (1997). 34. Uysal, K.T.; Scheja, L.; Wiesbrock, S.M.; Bonner-Weir, 29. Raclot, T.; Groscolas, R.; Langin, D.; Ferre, P. Site-spe- S.; Hotamisligil, G.S. Improved glucose and lipid cific regulation of gene expression by n-3 polyunsatu- metabolism in genetically obese mice lacking aP2. rated fatty acids in rat white adipose tissues. J. Lipid Endocrinology 141, 3388-3396 (2000). Res. 38, 1963-1972 (1997). 35. Shaughnessy, S.; Smith, E.R.; Kodukula, S.; Storch, J.; 30. Takahashi, Y.; Ide, T. Dietary n-3 fatty acids affect Fried, S.K. Adipocyte metabolism in adipocyte fatty mRNA levels of brown adipose uncoupling protein 1, acid binding protein knockout mice (aP2-/-) after short- and white adipose tissue leptin and glucose transporter term high-fat feeding: Functional compensation by the 4 in the rat. Br. J. Nutr. 84, 175-184 (2000). keratinocyte fatty acid binding protein. Diabetes 49, 31. Han, J.; Hamilton, A.; Kirkland, J.L.; Corkey, B.E.; Guo, 904-911 (2000). W. Medium-chain oil reduces fat mass and down-regu- 36. Mathe, D. Dyslipidemia and diabetes: Animal models. lates expression of adipogenic genes in rats. Obes. Res. Diabete Metab. 21, 106-111 (1995). 11, 734-744 (2003). 37. Ezaki, O.; Tsuji, E.; Momomura, K.; Kasuga, M.; Itaku- 32. Bernlohr D.A.; Doering T.L.; Kelly T.J. Jr; Lane M.D. ra, H. Effect of fish and safflower oil feeding on subcel- Tissue specific expression of p422 protein, a putative lular glucose transporter distributions in rat lipid carrier, in mouse adipocytes. Biochem. Biophys. adipocytes. Am. J. Physiol. 263, E94-E101 (1992). Res. Commun. 132, 850-855 (1985). 38. Huang, Y.J.; Fang, V.S.; Juan, C.C.; Chou, Y.C.; Kwok, 33. Hotamisligil, G.S.; Johnson, R.S.; Distel, R.J.; Ellis, R.; C.F.; Ho, L.T. Amelioration of insulin resistance and Papaioannou, V.E.; Spiegelman, B.M. Uncoupling of hypertension in a fructose-fed rat model with fish oil obesity from insulin resistance through a targeted supplementation. Metabolism 46, 1252-1258 (1997). mutation in aP2, the adipocyte fatty acid binding pro-

541 J. Oleo Sci. 56, (10) 533-541 (2007)