- Email: [email protected]
Estrogens induce visfatin expression in 3T3-L1 cells
Peptides 31 (2010) 271–274
Contents lists available at ScienceDirect
Peptides journal homepage: www.elsevier.com/locate/peptides
Estrogens induce visfatin expression in 3T3-L1 cells Jingyu Zhou, Edward R. Seidel * Department of Physiology, Brody School of Medicine At East Carolina University, Greenville, NC 27858, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 1 October 2009 Received in revised form 2 December 2009 Accepted 2 December 2009 Available online 29 December 2009
Visfatin is a 56 kDa protein that is overexpressed in pregnant women. Like insulin, 2 nM visfatin induced GLUT 4 translocation from the cytosolic fraction to the membrane in 3T3-L1 cells. We have previously reported that visfatin induces glucose uptake into 3T3-L1 cells. These two actions define visfatin as an insulinomimetic. Three estrogens are elevated in pregnancy. Estradiol, the predominant estrogen, estriol, produced by the fetal liver and the pro-estrogen progesterone are all higher during pregnancy than in nonparous women. 3T3-L1 cells were treated with 150 ng/ml estriol, 16 ng/ml estradiol or 190 ng/ml progesterone to reflect the circulating concentrations of these steroids during pregnancy. Estriol treatment produced a 2.5-fold increase in visfatin gene expression. Estradiol and progesterone had small but insignificant effects on visfatin gene expression. In a second experiment, cells were treated with a combination of all three steroids together at the same concentrations listed above. The combination treatment produced a 13-fold increase in visfatin gene expression. These data suggest that the estriol, estradiol and progesterone exert a synergistic effect on visfatin gene expression. Taken together these data suggest that visfatin may play a physiological role during pregnancy. Since visfatin potently and efficaciously induced GLUT 4 translocation in a cell culture model, any hypothetical role for visfatin in pregnancy should include the possibility that it may play a role in maternal/fetal glucose metabolism or distribution. Two possibilities present: either maternal visfatin is overexpressed as a protective response in the pregnant female to compensate for the insulin resistance that often accompanies pregnancy or the excess visfatin is a compensatory response to ensure adequate glucose delivery to the growing fetus. ß 2009 Elsevier Inc. All rights reserved.
Keywords: Estradiol Progesterone Estriol Pregnancy GLUT 4 Insulin
1. Introduction The perception of adipose tissue as a cell designed solely to store fat has been transformed. It is a significant endocrine organ— secreting many different compounds with effects on metabolism, such as leptin [20], TNF-a [3], and IL-6 [6]. These molecules, termed adipokines, have been shown to affect insulin resistance and have been implicated in other obesity-associated disease states [15,5]. In fact, many studies have shown differential expression of these molecules between visceral and subcutaneous fat depots [19,16,6]. Visfatin, which is secreted from human fat, is a 52 kDa insulinomimetic peptide (NM_005746) that is identical to pre-B cell colony enhancing factor (PBEF, a cytokine), and nicotinamide phosphoribosyltransferase (Nampt, an enzyme). We have previously demonstrated that visfatin is expressed to high levels in the omental fat of pregnant women [17]. We reported that visfatin mRNA was expressed to a level 8-fold higher
* Corresponding author at: Department of Physiology, Brody School of Medicine At East Carolina University, 600 Moye Blvd, 6N100, Greenville, NC 27858, United States. Tel.: +1 252 744 2775; fax: +1 252 744 3460. E-mail address: [email protected] (E.R. Seidel). 0196-9781/$ – see front matter ß 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.peptides.2009.12.004
in omental fat of pregnant women than in fat from lean controls, obese, and obese diabetic women. Immunoblot analysis suggested that visfatin might even be a peptide specific to pregnancy as little to no visfatin was found in omental fat from obese women whereas visfatin peptide was abundant in fat from pregnant women. Visfatin also induced uptake of 2-deoxyglucose into NIH 3T3 cells at the same concentration, 2 nM, as did insulin [17]. Additionally visfatin was expressed in human placenta. Immunohistochemical analysis indicated that visfatin was isolated almost completely to the fetal placental endothelium (for review see Ref. [10]). If visfatin is an insulinomimetic what better place for it to be found than in endothelial cells of the placental circulation. Three steroids are elevated in pregnancy [8]. Estradiol, the predominant estrogen, estriol, produced by the fetal liver and progesterone are all higher during pregnancy than in nonparous women. The circulating concentrations of these three estrogens during pregnancy are 150 ng/ml for estriol, 16 ng/ml for estradiol and 190 ng/ml for progesterone. Hereafter, the combination of these three hormones will simply be labeled estrogens. These values are strikingly high as compared to those found in nonpregnant women especially for estriol which is for all practical purposes nonexistent in women who are not pregnant. In nonpregnant women, serum estriol is <2 ng/ml, in the early follicular
272
J. Zhou, E.R. Seidel / Peptides 31 (2010) 271–274
stage estradiol is 20–100 pg/ml and progesterone is 0.2–1.5 ng/ml [5]. The experiments outlined below were designed to determine if steroids, at the concentrations found in pregnancy, regulate visfatin gene expression. The results support the hypothesis. 2. Methods 2.1. Cells NIH 3T3 cells were cultured in high glucose DMEM and maintained in an humidified 5% CO2 environment at 37 8C. Cells were differentiated by incubating confluent cultures in DMEM differentiation medium containing insulin, dexamethazone and IBMS (ZenBio, RTP, NC) for three days after which they were switched to maintenance medium DMEM (ZenBio, RTP, NC) [18]. Differentiation was judged by observing the appearance of cyotsolic fat droplets. RNA was collected using Trizol according to the manufacturer’s directions. Cells were serum starved overnight before the addition of experimental reagents. In one experiment, cells were treated with either 2 nM visfatin or insulin. After a 5-min incubation, protein was collected and an immunoblot was performed to visualize the GLUT 4 glucose transporter using a GLUT 4 specific antibody (courtesy of Dr. Lynis Dohm, BSOM, East Carolina University) [11]. 2.2. PCR PCR was performed as previously described [17]. The visfatin primers used in these experiments were mouse specific primers of the following sequences: forward CATAGGGGCATCTGCTCATT, and reverse GCTGCTGGAACAGAATAGCC. Beta actin was used as an internal control. Actin expression was not affected by estrogen treatments. 2.3. PCR statistical analysis Cycle threshold (Ct) values were obtained and data analysis utilized these values. Cycle threshold refers to the earliest cycle reading obtained during qPCR and are thus a measure of mRNA abundance. Ct values were averaged for each fat sample and normalized for b-actin. For comparison, the relative expression DDCt was calculated by using the 2 formula. The DCt equals the difference between Ct values for b-actin and visfatin levels. This comparison will be hereafter referred to as Units in the text. For placental qPCR simple Ct values for both b-actin and visfatin are presented.
Fig. 1. GLUT 4 translocation from cytosol to membrane after five minute treatment with 2 nM visfatin or insulin. In the control state (untreated cells) two bands are observed: one representing the cytosolic and the other representing the membrane bound fraction. After treatement with either insulin or visfatin all GLUT 4 transporter has migrated to the cytosol.
GLUT 4, a downstream target of insulin, translocates to the plasma membrane in order to function as a glucose transporter [12]. To determine whether observed insulin-like effects of visfatin is the consequence of activating downstream targets of insulin, we performed a Western immunoblot on GLUT 4. In control cells (untreated cells), the Western blot displayed two bands (Fig. 1). The top band represents membrane-bound GLUT 4 and the lower band is cytosolic GLUT 4. The GLUT 4 protein in cells treated in 2 nM insulin all became attached to the plasma membrane. Cells treated in 2 nM visfatin for 5 min showed a single band identical to cells that received insulin. Thus, visfatin displays characteristics of insulin in that it induced GLUT 4 translocation from the cytosol to the membrane. Since we have reported that visfatin expression is increased in pregnancy, experiments were performed to determine if estrogens regulated visfatin gene expression. Serum starved cells were treated overnight with either estriol (150 ng/ml), progesterone (190 ng/ml) or estradiol (16 ng/ml). Neither estradiol or progesterone had a significant effect on visfatin gene expression although both showed a slight positive trend. On the other hand, the fetal liver derived steroid estriol produced about a 2-fold increase in gene expression (Fig. 2). In a subsequent experiment, cells were treated overnight with a combination of all three estrogens at the concentrations listed above. This treatment induced a 13-fold increase in visfatin gene expression (Fig. 3). This experiment was repeated under similar conditions with n of 6 in each group and a 10-fold increase in visfatin gene expression was calculated. It should be noted that as measured by PCR, undifferentiated 3T3-L1 cells constitutively expressed visfatin in amounts similar to that observed in differentiated cells. However, steroids had no effect on visfatin expression in the undifferentiated cell (data not presented).
2.4. Chemicals Insulin, water soluble estradiol and progesterone were purchased from Sigma Chemical Co. (St Louis, MO). Estradiol, insulin and progesterone were dissolved and used in water. Estriol (MP Biomedicals, Solon, OH) was dissolved in DMSO and delivered in a volume of 20 ml to a well containing 2 ml DMEM. Control wells received 20 ml of DMSO vehicle. 2.5. Statistics A t-test was used to analyze for differences between groups. Significance was set at p < 0.05. 3. Results GLUT 4 is a glucose transport protein that is normally found in vesicles within the cytoplasm. It has been well established that
Fig. 2. Effect of three estrogens, estriol (150 ng/ml), estradiol (16 ng/ml) and progesterone (190 ng/ml) each used individually on the expression of the visfatin mRNA. Data were analyzed as described under Section 2 and presented as mean S.E.M. n = 6 in each group. *p < 0.05.
J. Zhou, E.R. Seidel / Peptides 31 (2010) 271–274
Fig. 3. Effect of the combination of three estrogens, estriol (150 ng/ml), estradiol (16 ng/ml) and progesterone (190 ng/ml) on the expression of the visfatin mRNA. Data were analyzed as described under Section 2 and presented as mean S.E.M. n = 6 in each group. *p < 0.05.
4. Discussion At the risk of belaboring the obvious, 3T3-L1 adipocytes express the visfatin gene just as does human omental fat. In addition, visfatin behaves as an insulinomimetic in these cells. It (1) induces GLUT 4 translocation to the membrane and (2) enhances glucose uptake at the same concentration as does insulin [17]. Meeting these two criteria defines visfatin as an insulinomimetic. The fact that visfatin induced glucose uptake into adipocytes relates to a small controversy in this field. Fukuhara et al. [7] were forced to retract their original paper in Science because they could not reproduce their data on visfatin stimulated glucose uptake. Our data support Fukuhara’s [7] original observation that visfatin does regulate glucose transport. Diabetes is the most common complication of pregnancy. Gestational diabetes mellitus is a state of insulin resistance that occurs in as many as 5% of pregnant women. A much higher fraction of women become insulin resistant during pregnancy but are not clinically defined as frankly diabetic. Our previous work [17] and the data presented above indicate that the parous state is accompanied by increases in visfatin. Since visfatin possesses insulinomimetic properties it seems plausible to suggest that the increase in visfatin seen in pregnancy may be a compensatory change in intermediary metabolism to ameliorate the insulin resistance which so often accompanies pregnancy. This becomes an even more plausible suggestion when the fact that estriol, a steroid which is almost exclusively derived from the fetal liver, was capable of increasing visfatin expression when used alone. Although there was a slight positive trend, neither of the other maternal steroids, progesterone nor estradiol, had this effect when used individually. Little is known regarding the control of visfatin gene expression. Choi et al. [2] reported that rosiglitazone or fenofibrate treatments increased visfatin gene expression in diabetic Otsuka Long-Evans Tokushima fatty rats. Dexamethazone increased visfatin gene expression by about 3-fold in both 3T3-L1 pre-adipocytes and adipocytes [13]. These authors also reported that progesterone and testosterone slightly decreased expression of the gene. In the data presented above, we found no significant effect of progesterone on visfatin gene expression when progesterone was used individually. In concordance with our results, they found no effect of estrogens in undifferentiated pre-adipocytes. Taken together the sum of the results reported above suggest that in both whole animals and in cells in culture, expression of the visfatin gene is hormonally regulated. In the current paper we report that neither progesterone nor estradiol had a positive effect on visfatin gene expression. In contrast,
273
estriol doubled expression of the gene. However, when cells were treated with a combination of all three hormones at concentrations found in pregnant women a 13-fold increase in expression of the visfatin gene was observed. The magnitude of this change suggests a synergistic effect occurs when all three steroids are used in combination. There are contradicting reports on serum levels of visfatin in pregnancy. Hu et al. [9] reported very high levels of human serum visfatin in control non-pregnant women (626 45.5 ng/ml), pregnant women (695.9 92.5 ng/ml) and pregnant women with pre-eclimpsia (308.3 ng/ml). These serum visfatin levels are 10– 20-fold higher compared to various other reports dealing with pregnant women (in the range of 9.4–40 ng/ml) [1,4,17,14]. Also, there are contradicting reports on serum/plasma visfatin concentrations in pregnant women with gestational diabetes mellitus. Krzyzanowska et al. [12] reported elevated serum visfatin levels in pregnant women with gestational diabetes mellitus compared to control, which increased during the course of pregnancy as well as after delivery. However, Chan et al. [1] reported slightly lower but probably not significantly different levels of plasma visfatin concentrations in the gestational diabetes mellitus group than in the healthy control group of pregnant women (9.4 3.8 ng/ml vs. 12.6 4.5 ng/ml). Katwa’s [10] suggestion that visfatin acts in a paracrine or autocrine mode of action would explain why there are discrepancies in reports of circulating visfatin in pregnancy. If visfatin acts only locally then measuring circulating blood levels would be ineffective. None of these studies establish any meaningful correlation between the serum levels of visfatin and pregnancy or complications of pregnancy. Visfatin is overexpressed in pregnancy [17]. We demonstrate in the current paper that treatment with the three primary estrogens of pregnancy increases visfatin gene expression in a cell culture model. Since visfatin potently and efficaciously induced glucose transport in a cell culture model, any hypothetical role for visfatin in pregnancy should include the possibility that it may play a role in maternal/fetal glucose metabolism or distribution. Two possibilities present: either maternal visfatin is overexpressed as a protective response in the pregnant female to compensate for the insulin resistance that often accompanies pregnancy or the excess visfatin is a compensatory response to ensure adequate glucose delivery to the growing fetus. Acknowledgements The authors wish to express their appreciation to Amanda for her assistance with the GLUT 4 immunoblot and Fatiha Moukdar for her help with the statistical analyses. In particular, thanks to the Polish gentleman who upon viewing one of our posters suggested that it was estrogens that regulated visfatin expression in pregnancy. Without his suggestion these experiments would not have been performed. This work was supported in part by a grant from the Brody Brothers Foundation. References [1] Chan T-F, Chen Y-L, Lee C-H, Chou F-H, Wu L-C, Jong S-B. Decreased plasma visfatin concentrations in women with gestational diabetes mellitus. J Soc Gynecol Investig 2006;13:364–7. [2] Choi KC, Ryu OH, Lee KW, Kim HY, Seo JA, Kim SG, et al. Effect of PPAE alpha and gamma agonist on the expression of visfatin, adiponectin and TNF alpha in visceral fat of OLETF rats. Biochem Biophys Res Commun 2005;336: 747–53. [3] Fain JN, Madan AK, Hiler ML, Cheema P, Bahouth SW. Comparison of the release of adipokines by adipose tissue, adipose tissue matrix, and adipocytes from visceral and subcutaneous abdominal adipose tissues of obese humans. Endocrinology 2004;145(5):2273–82. [4] Fasshauer M, Waldeyer T, Seeger J, Schrey S, Ebert T, Kratzsch J, et al. Serum levels of the adipokine visfatin are increased in pre-eclampsia. Clin Endocrinol (Oxf) 2008;69(1):69–73.
274
J. Zhou, E.R. Seidel / Peptides 31 (2010) 271–274
[5] Fernandez-Real JM, Ricart W. Insulin resistance and chronic cardiovascular inflammatory syndrome. Endocr Rev 2003;24:278–301. [6] Fried SK, Bunkin DA, Greenberg AS. Omental and subcutaneous adipose tissue of obese subjects release interleukin 6: depot difference and regulation by glucocorticoid. J Clin Endocrinol Metab 1998;83:847–50. [7] Fukuhara A, Matsuda M, Nishizawa M, Segawa K, Tanaka M, Kishimoto K, et al. Visfatin: a protein secreted by visceral fat that mimics the effects of insulin. Science 2005;307:426–30 [retracted]. [8] Greenspan FS, Gardner DG. Basic and clinical endocrinology. Lange Medical Books/McGraw Hill Companies; 2004. [9] Hu W, Wang Z, Wang H, Huang H, Dong M. Serum visfatin levels in late pregnancy and pre-eclampsia. Acta Obstet Gynecol Scand 2008;87(4):413–8. [10] Katwa, LC, Seidel, ER. Visfatin in pregnancy: proposed mechanism of peptide delivery (Minireview). Amino Acids; in press, doi:10 1007/s0076008r-r0194-7. [11] Kern M, Wells JA, Stephens JR, Elton CW, Friedman JE, Tapscott EB, et al. Insulin responsiveness in skeletal muscle is determined by glucose transporter (GLUT 4) protein level. Biochem J 1990;270:397–400. [12] Krzyzanowska K, Krugluger W, Mittermayer F, Rahman R, Haider D, Shnawa N, et al. Increased visfatin concentrations in women with gestational diabetes mellitus. Clin Sci (Lond) 2006;110(5):605–9.
[13] MacLaren R, Cui W, Cianflone K. Visfatin expression is hormonally regulated by metabolic and sex hormones in 3T3-L1 pre-adipocytes and adipocytes. Diabetes Obes Metab 2007;9:490–7. [14] Malamitsi-Puchner A, Briana DD, Boutsikou M, Kouskouni E, Hassiakos D, Gourgiotis D. Perinatal circulating visfatin levels in intrauterine growth restriction. Pediatrics 2007;119:1314–8. [15] Moller DE. Potential role of TNF-alpha in the pathogenesis of insulin resistance and type II diabetes. Trends Endocrinol Metab 2000;11(6):212–7. [16] Montague CT, Prins JB, Sanders I, Zhang J, Sewter CP, Digby J, et al. Depotrelated gene expression in human subcutaneous and omental adipocytes. Diabetes 1998;47:1384–91. [17] Morgan SA, Bringolf J, Seidel ER. Visfatin gene expression in human pregnancy. Peptides 2008;29:1382–9. [18] Stephens JM, Pekala PH. Transcriptional repression of GLUT4 and C/EBP genes in 3T3-L1 adipocytes by tumor necrosis factor. J Biol Chem 1991;266:21839–45. [19] Van Harmelen V, Reynisdottir S, Erikkson P, Thorne A, Hoffstedt J, Lonqvist F, et al. Leptin secretion from subcutaneous and visceral adipose tissue in women. Diabetes 1998;47:913–7. [20] Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 1994;372:425–32.
Contents lists available at ScienceDirect
Peptides journal homepage: www.elsevier.com/locate/peptides
Estrogens induce visfatin expression in 3T3-L1 cells Jingyu Zhou, Edward R. Seidel * Department of Physiology, Brody School of Medicine At East Carolina University, Greenville, NC 27858, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 1 October 2009 Received in revised form 2 December 2009 Accepted 2 December 2009 Available online 29 December 2009
Visfatin is a 56 kDa protein that is overexpressed in pregnant women. Like insulin, 2 nM visfatin induced GLUT 4 translocation from the cytosolic fraction to the membrane in 3T3-L1 cells. We have previously reported that visfatin induces glucose uptake into 3T3-L1 cells. These two actions define visfatin as an insulinomimetic. Three estrogens are elevated in pregnancy. Estradiol, the predominant estrogen, estriol, produced by the fetal liver and the pro-estrogen progesterone are all higher during pregnancy than in nonparous women. 3T3-L1 cells were treated with 150 ng/ml estriol, 16 ng/ml estradiol or 190 ng/ml progesterone to reflect the circulating concentrations of these steroids during pregnancy. Estriol treatment produced a 2.5-fold increase in visfatin gene expression. Estradiol and progesterone had small but insignificant effects on visfatin gene expression. In a second experiment, cells were treated with a combination of all three steroids together at the same concentrations listed above. The combination treatment produced a 13-fold increase in visfatin gene expression. These data suggest that the estriol, estradiol and progesterone exert a synergistic effect on visfatin gene expression. Taken together these data suggest that visfatin may play a physiological role during pregnancy. Since visfatin potently and efficaciously induced GLUT 4 translocation in a cell culture model, any hypothetical role for visfatin in pregnancy should include the possibility that it may play a role in maternal/fetal glucose metabolism or distribution. Two possibilities present: either maternal visfatin is overexpressed as a protective response in the pregnant female to compensate for the insulin resistance that often accompanies pregnancy or the excess visfatin is a compensatory response to ensure adequate glucose delivery to the growing fetus. ß 2009 Elsevier Inc. All rights reserved.
Keywords: Estradiol Progesterone Estriol Pregnancy GLUT 4 Insulin
1. Introduction The perception of adipose tissue as a cell designed solely to store fat has been transformed. It is a significant endocrine organ— secreting many different compounds with effects on metabolism, such as leptin [20], TNF-a [3], and IL-6 [6]. These molecules, termed adipokines, have been shown to affect insulin resistance and have been implicated in other obesity-associated disease states [15,5]. In fact, many studies have shown differential expression of these molecules between visceral and subcutaneous fat depots [19,16,6]. Visfatin, which is secreted from human fat, is a 52 kDa insulinomimetic peptide (NM_005746) that is identical to pre-B cell colony enhancing factor (PBEF, a cytokine), and nicotinamide phosphoribosyltransferase (Nampt, an enzyme). We have previously demonstrated that visfatin is expressed to high levels in the omental fat of pregnant women [17]. We reported that visfatin mRNA was expressed to a level 8-fold higher
* Corresponding author at: Department of Physiology, Brody School of Medicine At East Carolina University, 600 Moye Blvd, 6N100, Greenville, NC 27858, United States. Tel.: +1 252 744 2775; fax: +1 252 744 3460. E-mail address: [email protected] (E.R. Seidel). 0196-9781/$ – see front matter ß 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.peptides.2009.12.004
in omental fat of pregnant women than in fat from lean controls, obese, and obese diabetic women. Immunoblot analysis suggested that visfatin might even be a peptide specific to pregnancy as little to no visfatin was found in omental fat from obese women whereas visfatin peptide was abundant in fat from pregnant women. Visfatin also induced uptake of 2-deoxyglucose into NIH 3T3 cells at the same concentration, 2 nM, as did insulin [17]. Additionally visfatin was expressed in human placenta. Immunohistochemical analysis indicated that visfatin was isolated almost completely to the fetal placental endothelium (for review see Ref. [10]). If visfatin is an insulinomimetic what better place for it to be found than in endothelial cells of the placental circulation. Three steroids are elevated in pregnancy [8]. Estradiol, the predominant estrogen, estriol, produced by the fetal liver and progesterone are all higher during pregnancy than in nonparous women. The circulating concentrations of these three estrogens during pregnancy are 150 ng/ml for estriol, 16 ng/ml for estradiol and 190 ng/ml for progesterone. Hereafter, the combination of these three hormones will simply be labeled estrogens. These values are strikingly high as compared to those found in nonpregnant women especially for estriol which is for all practical purposes nonexistent in women who are not pregnant. In nonpregnant women, serum estriol is <2 ng/ml, in the early follicular
272
J. Zhou, E.R. Seidel / Peptides 31 (2010) 271–274
stage estradiol is 20–100 pg/ml and progesterone is 0.2–1.5 ng/ml [5]. The experiments outlined below were designed to determine if steroids, at the concentrations found in pregnancy, regulate visfatin gene expression. The results support the hypothesis. 2. Methods 2.1. Cells NIH 3T3 cells were cultured in high glucose DMEM and maintained in an humidified 5% CO2 environment at 37 8C. Cells were differentiated by incubating confluent cultures in DMEM differentiation medium containing insulin, dexamethazone and IBMS (ZenBio, RTP, NC) for three days after which they were switched to maintenance medium DMEM (ZenBio, RTP, NC) [18]. Differentiation was judged by observing the appearance of cyotsolic fat droplets. RNA was collected using Trizol according to the manufacturer’s directions. Cells were serum starved overnight before the addition of experimental reagents. In one experiment, cells were treated with either 2 nM visfatin or insulin. After a 5-min incubation, protein was collected and an immunoblot was performed to visualize the GLUT 4 glucose transporter using a GLUT 4 specific antibody (courtesy of Dr. Lynis Dohm, BSOM, East Carolina University) [11]. 2.2. PCR PCR was performed as previously described [17]. The visfatin primers used in these experiments were mouse specific primers of the following sequences: forward CATAGGGGCATCTGCTCATT, and reverse GCTGCTGGAACAGAATAGCC. Beta actin was used as an internal control. Actin expression was not affected by estrogen treatments. 2.3. PCR statistical analysis Cycle threshold (Ct) values were obtained and data analysis utilized these values. Cycle threshold refers to the earliest cycle reading obtained during qPCR and are thus a measure of mRNA abundance. Ct values were averaged for each fat sample and normalized for b-actin. For comparison, the relative expression DDCt was calculated by using the 2 formula. The DCt equals the difference between Ct values for b-actin and visfatin levels. This comparison will be hereafter referred to as Units in the text. For placental qPCR simple Ct values for both b-actin and visfatin are presented.
Fig. 1. GLUT 4 translocation from cytosol to membrane after five minute treatment with 2 nM visfatin or insulin. In the control state (untreated cells) two bands are observed: one representing the cytosolic and the other representing the membrane bound fraction. After treatement with either insulin or visfatin all GLUT 4 transporter has migrated to the cytosol.
GLUT 4, a downstream target of insulin, translocates to the plasma membrane in order to function as a glucose transporter [12]. To determine whether observed insulin-like effects of visfatin is the consequence of activating downstream targets of insulin, we performed a Western immunoblot on GLUT 4. In control cells (untreated cells), the Western blot displayed two bands (Fig. 1). The top band represents membrane-bound GLUT 4 and the lower band is cytosolic GLUT 4. The GLUT 4 protein in cells treated in 2 nM insulin all became attached to the plasma membrane. Cells treated in 2 nM visfatin for 5 min showed a single band identical to cells that received insulin. Thus, visfatin displays characteristics of insulin in that it induced GLUT 4 translocation from the cytosol to the membrane. Since we have reported that visfatin expression is increased in pregnancy, experiments were performed to determine if estrogens regulated visfatin gene expression. Serum starved cells were treated overnight with either estriol (150 ng/ml), progesterone (190 ng/ml) or estradiol (16 ng/ml). Neither estradiol or progesterone had a significant effect on visfatin gene expression although both showed a slight positive trend. On the other hand, the fetal liver derived steroid estriol produced about a 2-fold increase in gene expression (Fig. 2). In a subsequent experiment, cells were treated overnight with a combination of all three estrogens at the concentrations listed above. This treatment induced a 13-fold increase in visfatin gene expression (Fig. 3). This experiment was repeated under similar conditions with n of 6 in each group and a 10-fold increase in visfatin gene expression was calculated. It should be noted that as measured by PCR, undifferentiated 3T3-L1 cells constitutively expressed visfatin in amounts similar to that observed in differentiated cells. However, steroids had no effect on visfatin expression in the undifferentiated cell (data not presented).
2.4. Chemicals Insulin, water soluble estradiol and progesterone were purchased from Sigma Chemical Co. (St Louis, MO). Estradiol, insulin and progesterone were dissolved and used in water. Estriol (MP Biomedicals, Solon, OH) was dissolved in DMSO and delivered in a volume of 20 ml to a well containing 2 ml DMEM. Control wells received 20 ml of DMSO vehicle. 2.5. Statistics A t-test was used to analyze for differences between groups. Significance was set at p < 0.05. 3. Results GLUT 4 is a glucose transport protein that is normally found in vesicles within the cytoplasm. It has been well established that
Fig. 2. Effect of three estrogens, estriol (150 ng/ml), estradiol (16 ng/ml) and progesterone (190 ng/ml) each used individually on the expression of the visfatin mRNA. Data were analyzed as described under Section 2 and presented as mean S.E.M. n = 6 in each group. *p < 0.05.
J. Zhou, E.R. Seidel / Peptides 31 (2010) 271–274
Fig. 3. Effect of the combination of three estrogens, estriol (150 ng/ml), estradiol (16 ng/ml) and progesterone (190 ng/ml) on the expression of the visfatin mRNA. Data were analyzed as described under Section 2 and presented as mean S.E.M. n = 6 in each group. *p < 0.05.
4. Discussion At the risk of belaboring the obvious, 3T3-L1 adipocytes express the visfatin gene just as does human omental fat. In addition, visfatin behaves as an insulinomimetic in these cells. It (1) induces GLUT 4 translocation to the membrane and (2) enhances glucose uptake at the same concentration as does insulin [17]. Meeting these two criteria defines visfatin as an insulinomimetic. The fact that visfatin induced glucose uptake into adipocytes relates to a small controversy in this field. Fukuhara et al. [7] were forced to retract their original paper in Science because they could not reproduce their data on visfatin stimulated glucose uptake. Our data support Fukuhara’s [7] original observation that visfatin does regulate glucose transport. Diabetes is the most common complication of pregnancy. Gestational diabetes mellitus is a state of insulin resistance that occurs in as many as 5% of pregnant women. A much higher fraction of women become insulin resistant during pregnancy but are not clinically defined as frankly diabetic. Our previous work [17] and the data presented above indicate that the parous state is accompanied by increases in visfatin. Since visfatin possesses insulinomimetic properties it seems plausible to suggest that the increase in visfatin seen in pregnancy may be a compensatory change in intermediary metabolism to ameliorate the insulin resistance which so often accompanies pregnancy. This becomes an even more plausible suggestion when the fact that estriol, a steroid which is almost exclusively derived from the fetal liver, was capable of increasing visfatin expression when used alone. Although there was a slight positive trend, neither of the other maternal steroids, progesterone nor estradiol, had this effect when used individually. Little is known regarding the control of visfatin gene expression. Choi et al. [2] reported that rosiglitazone or fenofibrate treatments increased visfatin gene expression in diabetic Otsuka Long-Evans Tokushima fatty rats. Dexamethazone increased visfatin gene expression by about 3-fold in both 3T3-L1 pre-adipocytes and adipocytes [13]. These authors also reported that progesterone and testosterone slightly decreased expression of the gene. In the data presented above, we found no significant effect of progesterone on visfatin gene expression when progesterone was used individually. In concordance with our results, they found no effect of estrogens in undifferentiated pre-adipocytes. Taken together the sum of the results reported above suggest that in both whole animals and in cells in culture, expression of the visfatin gene is hormonally regulated. In the current paper we report that neither progesterone nor estradiol had a positive effect on visfatin gene expression. In contrast,
273
estriol doubled expression of the gene. However, when cells were treated with a combination of all three hormones at concentrations found in pregnant women a 13-fold increase in expression of the visfatin gene was observed. The magnitude of this change suggests a synergistic effect occurs when all three steroids are used in combination. There are contradicting reports on serum levels of visfatin in pregnancy. Hu et al. [9] reported very high levels of human serum visfatin in control non-pregnant women (626 45.5 ng/ml), pregnant women (695.9 92.5 ng/ml) and pregnant women with pre-eclimpsia (308.3 ng/ml). These serum visfatin levels are 10– 20-fold higher compared to various other reports dealing with pregnant women (in the range of 9.4–40 ng/ml) [1,4,17,14]. Also, there are contradicting reports on serum/plasma visfatin concentrations in pregnant women with gestational diabetes mellitus. Krzyzanowska et al. [12] reported elevated serum visfatin levels in pregnant women with gestational diabetes mellitus compared to control, which increased during the course of pregnancy as well as after delivery. However, Chan et al. [1] reported slightly lower but probably not significantly different levels of plasma visfatin concentrations in the gestational diabetes mellitus group than in the healthy control group of pregnant women (9.4 3.8 ng/ml vs. 12.6 4.5 ng/ml). Katwa’s [10] suggestion that visfatin acts in a paracrine or autocrine mode of action would explain why there are discrepancies in reports of circulating visfatin in pregnancy. If visfatin acts only locally then measuring circulating blood levels would be ineffective. None of these studies establish any meaningful correlation between the serum levels of visfatin and pregnancy or complications of pregnancy. Visfatin is overexpressed in pregnancy [17]. We demonstrate in the current paper that treatment with the three primary estrogens of pregnancy increases visfatin gene expression in a cell culture model. Since visfatin potently and efficaciously induced glucose transport in a cell culture model, any hypothetical role for visfatin in pregnancy should include the possibility that it may play a role in maternal/fetal glucose metabolism or distribution. Two possibilities present: either maternal visfatin is overexpressed as a protective response in the pregnant female to compensate for the insulin resistance that often accompanies pregnancy or the excess visfatin is a compensatory response to ensure adequate glucose delivery to the growing fetus. Acknowledgements The authors wish to express their appreciation to Amanda for her assistance with the GLUT 4 immunoblot and Fatiha Moukdar for her help with the statistical analyses. In particular, thanks to the Polish gentleman who upon viewing one of our posters suggested that it was estrogens that regulated visfatin expression in pregnancy. Without his suggestion these experiments would not have been performed. This work was supported in part by a grant from the Brody Brothers Foundation. References [1] Chan T-F, Chen Y-L, Lee C-H, Chou F-H, Wu L-C, Jong S-B. Decreased plasma visfatin concentrations in women with gestational diabetes mellitus. J Soc Gynecol Investig 2006;13:364–7. [2] Choi KC, Ryu OH, Lee KW, Kim HY, Seo JA, Kim SG, et al. Effect of PPAE alpha and gamma agonist on the expression of visfatin, adiponectin and TNF alpha in visceral fat of OLETF rats. Biochem Biophys Res Commun 2005;336: 747–53. [3] Fain JN, Madan AK, Hiler ML, Cheema P, Bahouth SW. Comparison of the release of adipokines by adipose tissue, adipose tissue matrix, and adipocytes from visceral and subcutaneous abdominal adipose tissues of obese humans. Endocrinology 2004;145(5):2273–82. [4] Fasshauer M, Waldeyer T, Seeger J, Schrey S, Ebert T, Kratzsch J, et al. Serum levels of the adipokine visfatin are increased in pre-eclampsia. Clin Endocrinol (Oxf) 2008;69(1):69–73.
274
J. Zhou, E.R. Seidel / Peptides 31 (2010) 271–274
[5] Fernandez-Real JM, Ricart W. Insulin resistance and chronic cardiovascular inflammatory syndrome. Endocr Rev 2003;24:278–301. [6] Fried SK, Bunkin DA, Greenberg AS. Omental and subcutaneous adipose tissue of obese subjects release interleukin 6: depot difference and regulation by glucocorticoid. J Clin Endocrinol Metab 1998;83:847–50. [7] Fukuhara A, Matsuda M, Nishizawa M, Segawa K, Tanaka M, Kishimoto K, et al. Visfatin: a protein secreted by visceral fat that mimics the effects of insulin. Science 2005;307:426–30 [retracted]. [8] Greenspan FS, Gardner DG. Basic and clinical endocrinology. Lange Medical Books/McGraw Hill Companies; 2004. [9] Hu W, Wang Z, Wang H, Huang H, Dong M. Serum visfatin levels in late pregnancy and pre-eclampsia. Acta Obstet Gynecol Scand 2008;87(4):413–8. [10] Katwa, LC, Seidel, ER. Visfatin in pregnancy: proposed mechanism of peptide delivery (Minireview). Amino Acids; in press, doi:10 1007/s0076008r-r0194-7. [11] Kern M, Wells JA, Stephens JR, Elton CW, Friedman JE, Tapscott EB, et al. Insulin responsiveness in skeletal muscle is determined by glucose transporter (GLUT 4) protein level. Biochem J 1990;270:397–400. [12] Krzyzanowska K, Krugluger W, Mittermayer F, Rahman R, Haider D, Shnawa N, et al. Increased visfatin concentrations in women with gestational diabetes mellitus. Clin Sci (Lond) 2006;110(5):605–9.
[13] MacLaren R, Cui W, Cianflone K. Visfatin expression is hormonally regulated by metabolic and sex hormones in 3T3-L1 pre-adipocytes and adipocytes. Diabetes Obes Metab 2007;9:490–7. [14] Malamitsi-Puchner A, Briana DD, Boutsikou M, Kouskouni E, Hassiakos D, Gourgiotis D. Perinatal circulating visfatin levels in intrauterine growth restriction. Pediatrics 2007;119:1314–8. [15] Moller DE. Potential role of TNF-alpha in the pathogenesis of insulin resistance and type II diabetes. Trends Endocrinol Metab 2000;11(6):212–7. [16] Montague CT, Prins JB, Sanders I, Zhang J, Sewter CP, Digby J, et al. Depotrelated gene expression in human subcutaneous and omental adipocytes. Diabetes 1998;47:1384–91. [17] Morgan SA, Bringolf J, Seidel ER. Visfatin gene expression in human pregnancy. Peptides 2008;29:1382–9. [18] Stephens JM, Pekala PH. Transcriptional repression of GLUT4 and C/EBP genes in 3T3-L1 adipocytes by tumor necrosis factor. J Biol Chem 1991;266:21839–45. [19] Van Harmelen V, Reynisdottir S, Erikkson P, Thorne A, Hoffstedt J, Lonqvist F, et al. Leptin secretion from subcutaneous and visceral adipose tissue in women. Diabetes 1998;47:913–7. [20] Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 1994;372:425–32.