Cytokine 67 (2014) 102–106
Contents lists available at ScienceDirect
Cytokine journal homepage: www.journals.elsevier.com/cytokine
Short Communication
Sexual dimorphism, weight gain and glucose intolerance in a B- and T-cell deficient mouse model Zachary A.P. Wintrob a, Emmanuel K. Oppong a, Matthew Foster a, Michael Martorana b, Yu C. Tse a, Li Zhong a, Jeffery M. Welt a, Hans R. Boateng a, Ioana M. Drumea a, John Irlam c, Charles M. LeVea c, Silviu L. Faitar b, Alice C. Ceacareanu a,⇑ a
Department of Pharmacy Practice, School of Pharmacy and Pharmaceutical Sciences, State University of New York at Buffalo, Buffalo, NY, USA Department of Math and Natural Sciences, D’Youville College, Buffalo, NY, USA c Department of Clinical Pathology, Roswell Park Cancer Institute, Buffalo, NY, USA b
a r t i c l e
i n f o
Article history: Received 20 August 2013 Received in revised form 7 January 2014 Accepted 25 February 2014 Available online 29 March 2014 Keywords: Sexual dimorphism Adaptive immune system Insulin resistance Estrogen Adipokines
a b s t r a c t Background: Estrogen is thought to aid maintenance of insulin sensitivity potentially through modulation of a counter-regulatory mechanism that interferes with the contribution of adaptive and innate immune systems to visceral fat deposition. We evaluated the impact of estrogen on long-term high fat diet (HFD) intake in B- and T-cell deficient and immunocompetent animals comparatively. Methods: A total of 16 BALB and 16 SCID mice, 8 of each sex and strain, were randomized to receive low fat diet, 4.1% fat or HFD, 35% fat, such that there was a group of both each sex and each strain receiving each diet. Biweekly levels of adiponectin, leptin and insulin levels were assessed and a glucose tolerance test (GTT) was performed after 13 weeks. Results: Unlike their male counterparts, HFD-fed SCID females neither gained weight, nor became insulin resistant. Meanwhile, in the HFD-fed BALB groups both males and females gained weight similarly, but remarkable sexual dimorphism was nonetheless observed. The females had notable higher adiponectin levels as compared to males (10–60 lg/mL vs. 6–10 lg/mL respectively) causing the adiponectin-toleptin (A/L) ratio to reach 80 one week after HFD initiation. The A/L dropped to 10, still higher than males, by week 13, but dropped to 2 by the end of the study in agreement with inverse insulin trends. None of the HFD-fed female groups developed insulin resistance (IR) by week 13, while all male counterparts had. Similar results were observed in the HFD-fed SCID groups whereby the females did not develop IR and had a higher A/L; however, adiponectin levels were comparable between groups (5–11 lg/mL). Conclusions: The present study provides lacking evidence indicating that estrogen may be sufficient to prevent weight gain and development of glucose intolerance in high-fat fed B- and T-cell deficient mice. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Obesity-associated conditions, such as liver steatosis, chronic low-grade inflammation and loss of insulin sensitivity have a higher prevalence in both adolescent and adult males as compared to females [1–3]. Estrogen is thought to link tissue insulin sensitivity changes and sexual dimorphism. A body of literature supports this hypothesis, such as reports of higher rates of diabetes reported in postmenopausal women as well as the exacerbation of insulin resistance (IR) in ovariectomized high fat diet (HFD)-fed female ⇑ Corresponding author. Address: SUNY School of Pharmacy and Pharmaceutical Sciences, NYS Center of Excellence in Bioinformatics and Life Sciences, 701 Ellicott Street, Buffalo, NY, USA. Tel.: +1 (716)881 7502; fax: +1 (716)849 6651. E-mail address: [email protected] (A.C. Ceacareanu). http://dx.doi.org/10.1016/j.cyto.2014.02.011 1043-4666/Ó 2014 Elsevier Ltd. All rights reserved.
mice [4,5]. Furthermore, estrogen has been specifically implicated as a modulator of processes associated with the maintenance of insulin sensitivity, including visceral fat deposition and prevention of adipocyte hypertrophy [4,6]. Of additional interest is the fact that adipocyte hypertrophy results from a very well-orchestrated collaborative contribution of both the adaptive and innate immune systems but as of yet which immune system component is modulated by estrogen and, thereby, prevents the development of insulin resistance remains largely unknown [7–9]. With relation to this particular aspect, Ballak et al. have recently challenged the idea that the adaptive immune system is required for the development of IR [10]. Their study yielded intriguing results, specifically that complete B- and T-cell deficiency neither protects against HFDinduced adipocyte inflammation nor the occurrence of IR [10]. These findings provided immense value in the understanding of
Z.A.P. Wintrob et al. / Cytokine 67 (2014) 102–106
the role for the innate immune system in these processes; however, the study was limited to male mice and the role of estrogen was not evaluated. Using a slightly different study design, the study presented here proceeded to evaluate whether or not estrogen interferes with the HFD-induced IR in B- and T-cell deficient mice. 2. Materials and methods 2.1. Animal study All protocols for animal use were approved by the State University of New York, University at Buffalo and Roswell Park Cancer Institute Institutional Animal Care and Use Committees, protocol #1144M. Seven week old C.B-17/IcrHsd-PrkdcSCID (SCID) and BALB/c AnNHsd (BALB) mice were purchased from Harlan Laboratories. A total of 16 BALB and 16 SCID mice, 8 of each sex and strain, were randomized to receive either low fat diet (LFD) (Cat.#58M1, 4.1% fat, 73% carbohydrate, 13% protein, 5% fiber) or HFD (Cat.#58Y1, 34.9% fat, 25.9% carbohydrate, 23.6% protein, and 6.5% fiber) such that there was a group (n = 4) of both each sex and each strain receiving each diet. Between weeks 11 and 15 all animals were returned to regular fat diet (RFD) (2018S, 6.2% fat, 44.2% carbohydrate, 18.6% protein, 18.2% fiber). All diets were purchased from Test Diet (Richmond, IN) and sterilized by gamma irradiation prior to study initiation. Food and water were provided ad libitum, with the exception of 10 h biweekly fasting periods prior to bodyweight measurement and blood collection. Free access to sterile water was maintained throughout the study. Mice were individually weighed on an Ohaus CS200 digital balance (Ohaus Corporation, Parsippany, NJ), and 75 ll of blood were subsequently collected in sterile uncoated microvettes (Cat# CB300, Sarstedt, Numbrecht, Germany) by saphenous vein venipuncture. Collected samples were labeled and clotted at room temperature (RT) for 5 min. Upon centrifugation, sera from each group were pooled, yielding 150 ll of serum per group. Samples were stored at 80 °C until analysis. A glucose tolerance test (GTT) was performed at week 13, following 10 h of fasting. Blood sugar measurements were performed using a TRUEresult glucometer according to the manufacturer recommendation (NIPRO Diagnostics, Fort Lauderdale, FL; Product #56151-1240-01). Baseline (‘‘time 0’’) glucose values were recorded, then mice received 2.5 g/kg glucose intraperitoneally (TEKnova, Hollister, CA: Cat.#G2020) and subsequent blood glucose measurements were taken 15, 30, 60, 120, and 180 min after glucose injection. 2.2. Enzyme-linked immunosorbent assay (ELISA) Circulating adiponectin (ACRP30) was quantified according to the manufacturer protocol (Millipore, St. Charles, Missouri; Cat.#EZMADP-60 K). This kit did not detect adiponectin globular domain. 2.3. Two-bead array (Luminex) assay Quantification of circulating leptin and insulin was performed by duplex Luminex assay according to the manufacturer protocol (Millipore, Billerica, MA: Cat.#MGT-78K-02). 2.4. Statistical analyses Differences between investigated groups were evaluated by 2-way ANOVA; the level of statistical significance was set at p < 0.05. Results were reported as means ± SD of 4 mice in each
103
group or as a single biomarker result of pooled serum samples from 4 mice belonging to the same study group. SYSTAT version 13.00.05 was used for all the statistical analyses. 3. Results and discussion 3.1. HFD-fed SCID males gain weight and develop IR, whereas females do not Consistent with previously reported data, baseline bodyweight of the SCID mice were lower than BALB [10,11]. SCID females displayed a steady-low weight gain pattern (1 g/week) throughout the study regardless of diet fat content (Fig. 1a and b). All other HFD-fed groups, including the BALB females, gained up to approximately 4-fold that of the SCID females’ by the last week (Fig. 1b). The observations presented here are in congruence with the findings of Ballak et al. and confirm that there is no weight gain difference between BALB and SCID males in response to long-term HFD intake. In contrast, a significantly lower HFD-induced weight gain in SCID females as compared to SCID males was observed as early as week 11 (Fig. 1b, p < 0.05). Given the starting age of the animals and using the corresponding LFD group for comparison (Fig. 1a), it is unlikely that natural age related animal growth interfered with the diet-induced weight gain data. After 13 weeks the GTT confirmed an additional sexual dimorphism in the HFD-fed mice. However, this dimorphism only correlated in part with the observed difference in weight gain. While IR occurred to a similar extent in both HFD-fed male groups neither of the HFD-fed female groups displayed any notable change (Fig. 1b, d, f, and h). Furthermore, there was no significant difference in the GTT between females receiving either the low- or high-fat diets. The most interesting observation was that B- and T-cell deficient females failed to gain weight despite long-term HFD intake. When considered together with the findings of Stubbins et al., the effect of estrogen on the immune system and its role in IR becomes clearer. Stubbins et al. found that female C57BL/6 mice gained weight but did not develop IR; however, upon ovariectomy females both gained weight, approximately equal to male C57BL/6 mice, and developed IR. The effects of ovariectomy were attenuated upon estrogen supplementation. In light of these findings, the results of the present study suggest that the innate immune system is largely responsible for the development of IR and that the process responsible can be attenuated by estrogen in both the presence and absence of B- and T-cells. Estrogen also appears to suppress weight gain given that the ovariectomized mice not receiving estrogen supplementation in the study by Stubbins et al. gained more weight than nonovariectomized females. However, any lean phenotype induced by estrogen seems to eventually become overwhelmed in the presence of B- and T-cells despite being complete in the absence of B- and T-cells. It is also worth noting that the technical notes recorded over the span of this study consistently indicated that HFD-fed SCID females developed an obviously oily coat that was not observed in the other HFD-fed groups. It is possible that in the absence of B- and T-cells estrogen may lead to impaired expression of stearoyl-CoA desaturase (SCD), a central lipogenic enzyme shown to control adipose tissue deposition [12]. However, such a hypothesis requires additional research. 3.2. Hyperinsulinemia and hyperleptinemia occur in all HFD-fed animals with the exception of SCID females Insulin and leptin levels varied dramatically from roughly 0.4 ng/mL each in LFD-fed to > 1 ng/mL and > 4 ng/mL respectively,
104
Z.A.P. Wintrob et al. / Cytokine 67 (2014) 102–106
Fig. 1. Weight gain and glucose tolerance test at week 13 among different study groups, LFD (a and c) and HFD (b and d). Solid line: SCID males, long dashed line: SCID females, short dashed line: BALB males, dotted line: BALB females. Shaded region indicates the duration of feeding with 2018S (6.2%). Results are expressed as mean (n = 4) ± SD and were analyzed by using 2-way ANOVA, *p < 0.05.
in HFD-fed BALB females. BALB and SCID males displayed an increasing trend of insulin and leptin production in perfect correlation with their weight gain, reaching 3 ng/mL insulin and between 3 and 5 ng/mL leptin by the end of the study. SCID females maintained lower levels of insulin and leptin among all groups. 3.3. Low adiponectin-to-leptin ratio (A/L) was observed sooner in HFD-fed males than females In our study the A/L ratio was found to be remarkably low in both HFD-fed male groups (Fig. 2d and h) and, as expected, this has been accompanied by weight gain and IR (Fig. 1b and d). The HFD-fed BALB females’ A/L ratio immediately spiked to the highest observed value of any group upon initiation of HFD and remained elevated throughout the study. This spike resulted from both an increase in adiponectin and a decrease in leptin. The observed biomarker changes in response to HFD may be part of a compensatory and protective mechanism as these initial changes were observed in all HFD-fed male groups as well, but were very quickly overwhelmed. Increased adiponectin levels have been linked to decreased macrophage infiltration of adipocytes and, thereby, less adipocyte hypertrophy and estrogen has been found to increase the levels of adiponectin in obese women [8,13]. However, adiponectin production becomes impaired as adipocytes become hypertrophic from macrophage infiltration [14–16]. The contribution of inflammation from macrophage infiltrated adipocytes and the associated increase in leptin and decrease in adiponectin are all thought to contribute to IR. Since B- and T-cell derived cytokines contribute
to macrophage recruitment into adipocytes it is the adaptive immune system of the HFD-fed BALB female mice that may be responsible for overwhelming what protection estrogen was able to initially provide. This is consistent with the observed slow but continuous decrease in the A/L ratio until week 11 when it plateaued for the duration of the diet change interval (from HFD to RFD), and then continued decreasing once HFD was reinitiated (Fig. 2f). The opposite was observed in the LFD-fed BALB females, with A/L displaying a consistently increasing pattern until the time of diet change (from LFD to RFD), when it decreased, to then remained steadily elevated afterwards upon the reinitiation of the HFD (Fig. 2e). Data presented in Fig. 2 imply that fat intake modulates A/L ratio through changes in adiponectin and leptin production. 3.4. B- and T-cells play a role in the sexual dimorphism of adiponectin production Despite the marked difference in adiponectin production observed in BALB females vs. males at all given time points, such has not been observed in B- and T-cell deficient mice (Fig. 2e–h). This data suggests that higher adiponectin levels are warranted in immunocompetent animals, potentially due to a greater need for the anti-inflammatory properties of adiponectin. The association between obesity-related metabolic disorders, adiponectin, and adaptive immunity was first suggested in 2008, when Okamoto et al. demonstrated that levels of adiponectin P 1 lg/mL could prevent interferon gamma-induced protein 10 (IP-10) secretion and chemokine receptor CXCR3 chemotactic
Z.A.P. Wintrob et al. / Cytokine 67 (2014) 102–106
105
Fig. 2. Adipokines’ levels among different study groups, (LFD a, c, e, g) and (HFD b, d, f, h) respectively. Solid line: adiponectin (ng/mL), long dashed line: leptin (pg/mL), short dashed line: insulin (pg/mL), dotted line: adiponectin/leptin ratio (ng/pg). Shaded region indicates the duration of feeding with 2018S (6.2%). Each biomarker level time point has been determined by analyzing a serum samples pool collected from all the animals of the respective study group (n = 4).
activity in lipopolysaccharide-stimulated macrophages [14,15]. These in vitro findings support the hypothesis that low levels of adiponectin associated with obesity favor recruitment and accumulation of T lymphocytes. Observed adiponectin levels in the study presented here range from roughly 10 lg/mL in all SCID females to 10–40 lg/mL in LFD-fed BALB females and 10–60 lg/mL in HFD-fed BALB females (Fig. 2a, b, e, and f). Adiponectin levels recorded during this study in all male mice, regardless of strain or dietary fat intake, ranged from 6 to 10 lg/mL (Fig. 2c, d, g, and h). 3.5. B- and T-cell deficient females have lower leptin levels despite HFD intake Unlike HFD-fed immunocompetent females, where elevated A/L ratio is determined by high adiponectin levels, HFD-fed SCID females elevated A/L ratio is determined by lowering leptin
levels. This suggests that the adaptive immune system may play a role in the stimulation of leptin production in high-fat fed immunocompetent animals, perhaps through stimulating macrophage recruitment into adipocytes. The elevated A/L ratio – although in decline – prevented weight gain in HFD-fed BALB females, but only until week 9 of the study when it reached a range similar to the male groups (
Contents lists available at ScienceDirect
Cytokine journal homepage: www.journals.elsevier.com/cytokine
Short Communication
Sexual dimorphism, weight gain and glucose intolerance in a B- and T-cell deficient mouse model Zachary A.P. Wintrob a, Emmanuel K. Oppong a, Matthew Foster a, Michael Martorana b, Yu C. Tse a, Li Zhong a, Jeffery M. Welt a, Hans R. Boateng a, Ioana M. Drumea a, John Irlam c, Charles M. LeVea c, Silviu L. Faitar b, Alice C. Ceacareanu a,⇑ a
Department of Pharmacy Practice, School of Pharmacy and Pharmaceutical Sciences, State University of New York at Buffalo, Buffalo, NY, USA Department of Math and Natural Sciences, D’Youville College, Buffalo, NY, USA c Department of Clinical Pathology, Roswell Park Cancer Institute, Buffalo, NY, USA b
a r t i c l e
i n f o
Article history: Received 20 August 2013 Received in revised form 7 January 2014 Accepted 25 February 2014 Available online 29 March 2014 Keywords: Sexual dimorphism Adaptive immune system Insulin resistance Estrogen Adipokines
a b s t r a c t Background: Estrogen is thought to aid maintenance of insulin sensitivity potentially through modulation of a counter-regulatory mechanism that interferes with the contribution of adaptive and innate immune systems to visceral fat deposition. We evaluated the impact of estrogen on long-term high fat diet (HFD) intake in B- and T-cell deficient and immunocompetent animals comparatively. Methods: A total of 16 BALB and 16 SCID mice, 8 of each sex and strain, were randomized to receive low fat diet, 4.1% fat or HFD, 35% fat, such that there was a group of both each sex and each strain receiving each diet. Biweekly levels of adiponectin, leptin and insulin levels were assessed and a glucose tolerance test (GTT) was performed after 13 weeks. Results: Unlike their male counterparts, HFD-fed SCID females neither gained weight, nor became insulin resistant. Meanwhile, in the HFD-fed BALB groups both males and females gained weight similarly, but remarkable sexual dimorphism was nonetheless observed. The females had notable higher adiponectin levels as compared to males (10–60 lg/mL vs. 6–10 lg/mL respectively) causing the adiponectin-toleptin (A/L) ratio to reach 80 one week after HFD initiation. The A/L dropped to 10, still higher than males, by week 13, but dropped to 2 by the end of the study in agreement with inverse insulin trends. None of the HFD-fed female groups developed insulin resistance (IR) by week 13, while all male counterparts had. Similar results were observed in the HFD-fed SCID groups whereby the females did not develop IR and had a higher A/L; however, adiponectin levels were comparable between groups (5–11 lg/mL). Conclusions: The present study provides lacking evidence indicating that estrogen may be sufficient to prevent weight gain and development of glucose intolerance in high-fat fed B- and T-cell deficient mice. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Obesity-associated conditions, such as liver steatosis, chronic low-grade inflammation and loss of insulin sensitivity have a higher prevalence in both adolescent and adult males as compared to females [1–3]. Estrogen is thought to link tissue insulin sensitivity changes and sexual dimorphism. A body of literature supports this hypothesis, such as reports of higher rates of diabetes reported in postmenopausal women as well as the exacerbation of insulin resistance (IR) in ovariectomized high fat diet (HFD)-fed female ⇑ Corresponding author. Address: SUNY School of Pharmacy and Pharmaceutical Sciences, NYS Center of Excellence in Bioinformatics and Life Sciences, 701 Ellicott Street, Buffalo, NY, USA. Tel.: +1 (716)881 7502; fax: +1 (716)849 6651. E-mail address: [email protected] (A.C. Ceacareanu). http://dx.doi.org/10.1016/j.cyto.2014.02.011 1043-4666/Ó 2014 Elsevier Ltd. All rights reserved.
mice [4,5]. Furthermore, estrogen has been specifically implicated as a modulator of processes associated with the maintenance of insulin sensitivity, including visceral fat deposition and prevention of adipocyte hypertrophy [4,6]. Of additional interest is the fact that adipocyte hypertrophy results from a very well-orchestrated collaborative contribution of both the adaptive and innate immune systems but as of yet which immune system component is modulated by estrogen and, thereby, prevents the development of insulin resistance remains largely unknown [7–9]. With relation to this particular aspect, Ballak et al. have recently challenged the idea that the adaptive immune system is required for the development of IR [10]. Their study yielded intriguing results, specifically that complete B- and T-cell deficiency neither protects against HFDinduced adipocyte inflammation nor the occurrence of IR [10]. These findings provided immense value in the understanding of
Z.A.P. Wintrob et al. / Cytokine 67 (2014) 102–106
the role for the innate immune system in these processes; however, the study was limited to male mice and the role of estrogen was not evaluated. Using a slightly different study design, the study presented here proceeded to evaluate whether or not estrogen interferes with the HFD-induced IR in B- and T-cell deficient mice. 2. Materials and methods 2.1. Animal study All protocols for animal use were approved by the State University of New York, University at Buffalo and Roswell Park Cancer Institute Institutional Animal Care and Use Committees, protocol #1144M. Seven week old C.B-17/IcrHsd-PrkdcSCID (SCID) and BALB/c AnNHsd (BALB) mice were purchased from Harlan Laboratories. A total of 16 BALB and 16 SCID mice, 8 of each sex and strain, were randomized to receive either low fat diet (LFD) (Cat.#58M1, 4.1% fat, 73% carbohydrate, 13% protein, 5% fiber) or HFD (Cat.#58Y1, 34.9% fat, 25.9% carbohydrate, 23.6% protein, and 6.5% fiber) such that there was a group (n = 4) of both each sex and each strain receiving each diet. Between weeks 11 and 15 all animals were returned to regular fat diet (RFD) (2018S, 6.2% fat, 44.2% carbohydrate, 18.6% protein, 18.2% fiber). All diets were purchased from Test Diet (Richmond, IN) and sterilized by gamma irradiation prior to study initiation. Food and water were provided ad libitum, with the exception of 10 h biweekly fasting periods prior to bodyweight measurement and blood collection. Free access to sterile water was maintained throughout the study. Mice were individually weighed on an Ohaus CS200 digital balance (Ohaus Corporation, Parsippany, NJ), and 75 ll of blood were subsequently collected in sterile uncoated microvettes (Cat# CB300, Sarstedt, Numbrecht, Germany) by saphenous vein venipuncture. Collected samples were labeled and clotted at room temperature (RT) for 5 min. Upon centrifugation, sera from each group were pooled, yielding 150 ll of serum per group. Samples were stored at 80 °C until analysis. A glucose tolerance test (GTT) was performed at week 13, following 10 h of fasting. Blood sugar measurements were performed using a TRUEresult glucometer according to the manufacturer recommendation (NIPRO Diagnostics, Fort Lauderdale, FL; Product #56151-1240-01). Baseline (‘‘time 0’’) glucose values were recorded, then mice received 2.5 g/kg glucose intraperitoneally (TEKnova, Hollister, CA: Cat.#G2020) and subsequent blood glucose measurements were taken 15, 30, 60, 120, and 180 min after glucose injection. 2.2. Enzyme-linked immunosorbent assay (ELISA) Circulating adiponectin (ACRP30) was quantified according to the manufacturer protocol (Millipore, St. Charles, Missouri; Cat.#EZMADP-60 K). This kit did not detect adiponectin globular domain. 2.3. Two-bead array (Luminex) assay Quantification of circulating leptin and insulin was performed by duplex Luminex assay according to the manufacturer protocol (Millipore, Billerica, MA: Cat.#MGT-78K-02). 2.4. Statistical analyses Differences between investigated groups were evaluated by 2-way ANOVA; the level of statistical significance was set at p < 0.05. Results were reported as means ± SD of 4 mice in each
103
group or as a single biomarker result of pooled serum samples from 4 mice belonging to the same study group. SYSTAT version 13.00.05 was used for all the statistical analyses. 3. Results and discussion 3.1. HFD-fed SCID males gain weight and develop IR, whereas females do not Consistent with previously reported data, baseline bodyweight of the SCID mice were lower than BALB [10,11]. SCID females displayed a steady-low weight gain pattern (1 g/week) throughout the study regardless of diet fat content (Fig. 1a and b). All other HFD-fed groups, including the BALB females, gained up to approximately 4-fold that of the SCID females’ by the last week (Fig. 1b). The observations presented here are in congruence with the findings of Ballak et al. and confirm that there is no weight gain difference between BALB and SCID males in response to long-term HFD intake. In contrast, a significantly lower HFD-induced weight gain in SCID females as compared to SCID males was observed as early as week 11 (Fig. 1b, p < 0.05). Given the starting age of the animals and using the corresponding LFD group for comparison (Fig. 1a), it is unlikely that natural age related animal growth interfered with the diet-induced weight gain data. After 13 weeks the GTT confirmed an additional sexual dimorphism in the HFD-fed mice. However, this dimorphism only correlated in part with the observed difference in weight gain. While IR occurred to a similar extent in both HFD-fed male groups neither of the HFD-fed female groups displayed any notable change (Fig. 1b, d, f, and h). Furthermore, there was no significant difference in the GTT between females receiving either the low- or high-fat diets. The most interesting observation was that B- and T-cell deficient females failed to gain weight despite long-term HFD intake. When considered together with the findings of Stubbins et al., the effect of estrogen on the immune system and its role in IR becomes clearer. Stubbins et al. found that female C57BL/6 mice gained weight but did not develop IR; however, upon ovariectomy females both gained weight, approximately equal to male C57BL/6 mice, and developed IR. The effects of ovariectomy were attenuated upon estrogen supplementation. In light of these findings, the results of the present study suggest that the innate immune system is largely responsible for the development of IR and that the process responsible can be attenuated by estrogen in both the presence and absence of B- and T-cells. Estrogen also appears to suppress weight gain given that the ovariectomized mice not receiving estrogen supplementation in the study by Stubbins et al. gained more weight than nonovariectomized females. However, any lean phenotype induced by estrogen seems to eventually become overwhelmed in the presence of B- and T-cells despite being complete in the absence of B- and T-cells. It is also worth noting that the technical notes recorded over the span of this study consistently indicated that HFD-fed SCID females developed an obviously oily coat that was not observed in the other HFD-fed groups. It is possible that in the absence of B- and T-cells estrogen may lead to impaired expression of stearoyl-CoA desaturase (SCD), a central lipogenic enzyme shown to control adipose tissue deposition [12]. However, such a hypothesis requires additional research. 3.2. Hyperinsulinemia and hyperleptinemia occur in all HFD-fed animals with the exception of SCID females Insulin and leptin levels varied dramatically from roughly 0.4 ng/mL each in LFD-fed to > 1 ng/mL and > 4 ng/mL respectively,
104
Z.A.P. Wintrob et al. / Cytokine 67 (2014) 102–106
Fig. 1. Weight gain and glucose tolerance test at week 13 among different study groups, LFD (a and c) and HFD (b and d). Solid line: SCID males, long dashed line: SCID females, short dashed line: BALB males, dotted line: BALB females. Shaded region indicates the duration of feeding with 2018S (6.2%). Results are expressed as mean (n = 4) ± SD and were analyzed by using 2-way ANOVA, *p < 0.05.
in HFD-fed BALB females. BALB and SCID males displayed an increasing trend of insulin and leptin production in perfect correlation with their weight gain, reaching 3 ng/mL insulin and between 3 and 5 ng/mL leptin by the end of the study. SCID females maintained lower levels of insulin and leptin among all groups. 3.3. Low adiponectin-to-leptin ratio (A/L) was observed sooner in HFD-fed males than females In our study the A/L ratio was found to be remarkably low in both HFD-fed male groups (Fig. 2d and h) and, as expected, this has been accompanied by weight gain and IR (Fig. 1b and d). The HFD-fed BALB females’ A/L ratio immediately spiked to the highest observed value of any group upon initiation of HFD and remained elevated throughout the study. This spike resulted from both an increase in adiponectin and a decrease in leptin. The observed biomarker changes in response to HFD may be part of a compensatory and protective mechanism as these initial changes were observed in all HFD-fed male groups as well, but were very quickly overwhelmed. Increased adiponectin levels have been linked to decreased macrophage infiltration of adipocytes and, thereby, less adipocyte hypertrophy and estrogen has been found to increase the levels of adiponectin in obese women [8,13]. However, adiponectin production becomes impaired as adipocytes become hypertrophic from macrophage infiltration [14–16]. The contribution of inflammation from macrophage infiltrated adipocytes and the associated increase in leptin and decrease in adiponectin are all thought to contribute to IR. Since B- and T-cell derived cytokines contribute
to macrophage recruitment into adipocytes it is the adaptive immune system of the HFD-fed BALB female mice that may be responsible for overwhelming what protection estrogen was able to initially provide. This is consistent with the observed slow but continuous decrease in the A/L ratio until week 11 when it plateaued for the duration of the diet change interval (from HFD to RFD), and then continued decreasing once HFD was reinitiated (Fig. 2f). The opposite was observed in the LFD-fed BALB females, with A/L displaying a consistently increasing pattern until the time of diet change (from LFD to RFD), when it decreased, to then remained steadily elevated afterwards upon the reinitiation of the HFD (Fig. 2e). Data presented in Fig. 2 imply that fat intake modulates A/L ratio through changes in adiponectin and leptin production. 3.4. B- and T-cells play a role in the sexual dimorphism of adiponectin production Despite the marked difference in adiponectin production observed in BALB females vs. males at all given time points, such has not been observed in B- and T-cell deficient mice (Fig. 2e–h). This data suggests that higher adiponectin levels are warranted in immunocompetent animals, potentially due to a greater need for the anti-inflammatory properties of adiponectin. The association between obesity-related metabolic disorders, adiponectin, and adaptive immunity was first suggested in 2008, when Okamoto et al. demonstrated that levels of adiponectin P 1 lg/mL could prevent interferon gamma-induced protein 10 (IP-10) secretion and chemokine receptor CXCR3 chemotactic
Z.A.P. Wintrob et al. / Cytokine 67 (2014) 102–106
105
Fig. 2. Adipokines’ levels among different study groups, (LFD a, c, e, g) and (HFD b, d, f, h) respectively. Solid line: adiponectin (ng/mL), long dashed line: leptin (pg/mL), short dashed line: insulin (pg/mL), dotted line: adiponectin/leptin ratio (ng/pg). Shaded region indicates the duration of feeding with 2018S (6.2%). Each biomarker level time point has been determined by analyzing a serum samples pool collected from all the animals of the respective study group (n = 4).
activity in lipopolysaccharide-stimulated macrophages [14,15]. These in vitro findings support the hypothesis that low levels of adiponectin associated with obesity favor recruitment and accumulation of T lymphocytes. Observed adiponectin levels in the study presented here range from roughly 10 lg/mL in all SCID females to 10–40 lg/mL in LFD-fed BALB females and 10–60 lg/mL in HFD-fed BALB females (Fig. 2a, b, e, and f). Adiponectin levels recorded during this study in all male mice, regardless of strain or dietary fat intake, ranged from 6 to 10 lg/mL (Fig. 2c, d, g, and h). 3.5. B- and T-cell deficient females have lower leptin levels despite HFD intake Unlike HFD-fed immunocompetent females, where elevated A/L ratio is determined by high adiponectin levels, HFD-fed SCID females elevated A/L ratio is determined by lowering leptin
levels. This suggests that the adaptive immune system may play a role in the stimulation of leptin production in high-fat fed immunocompetent animals, perhaps through stimulating macrophage recruitment into adipocytes. The elevated A/L ratio – although in decline – prevented weight gain in HFD-fed BALB females, but only until week 9 of the study when it reached a range similar to the male groups (
