cells Article IL-6 Trans-Signaling in the Brain Influences the Metabolic Phenotype of the 3xTg-AD Mouse Model of Alzheimer’s Disease Anna Escrig 1 , Amalia Molinero 1 , Brenda Méndez 1 , Mercedes Giralt 1 , Gemma Comes 1 , Paula Sanchis 1 , Olaya Fernández-Gayol 1,† , Lydia Giménez-Llort 2 , Christoph Becker-Pauly 3 , Stefan Rose-John 3 and Juan Hidalgo 1, * 1 Institute of Neurosciences and Department of Cellular Biology, Physiology and Immunology, Faculty of Biosciences, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain; annesmon@gmail.com (A.E.); Amalia.Molinero@uab.cat (A.M.); bren.mendezm@gmail.com (B.M.); merce.giralt@uab.cat (M.G.); Gemma.comes@uab.cat (G.C.); paula.sanchis@uab.cat (P.S.); of2194@cumc.columbia.edu (O.F.-G.) 2 Institute of Neurosciences and Department of Psychiatry and Forensic Medicine, Faculty of Medicine, Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain; lidia.gimenez@uab.cat 3 Department of Biochemistry, Medical Faculty, Christian-Albrechts-Universität zu Kiel, 24098 Kiel, Germany; cbeckerpauly@biochem.uni-kiel.de (C.B.-P.); rosejohn@biochem.uni-kiel.de (S.R.-J.) * Correspondence: Juan.Hidalgo@uab.es; Tel.: +34-9-3581-2037; Fax: +34-9-3581-2390 † Current affiliation: Department of Pediatrics, Division of Molecular Genetics, Columbia University Irving Medical Center, New York, NY 10032, USA.  Received: 28 May 2020; Accepted: 1 July 2020; Published: 2 July 2020  Abstract: Alzheimer’s disease (AD) is a neurodegenerative disorder that causes the most prevalent dementia in the elderly people. Obesity and insulin resistance, which may cause major health problems per se, are risk factors for AD, and cytokines such as interleukin-6 (IL-6) have a role in these conditions. IL-6 can signal either through a membrane receptor or by trans-signaling, which can be inhibited by the soluble form of the co-receptor gp130 (sgp130). We have addressed the possibility that blocking IL-6 trans-signaling in the brain could have an effect in the triple transgenic 3xTg-AD mouse model of AD and/or in obesity progression, by crossing 3xTg-AD mice with GFAP-sgp130Fc mice. To serve as control groups, GFAP-sgp130Fc mice were also crossed with C57BL/6JOlaHsd mice. Seventeen-month-old mice were fed a control diet (18% kcal from fat) and a high-fat diet (HFD; 58.4% kcal from fat). In our experimental conditions, the 3xTg-AD model showed a mild amyloid phenotype, which nevertheless altered the control of body weight and related endocrine and metabolic factors, suggestive of a hypermetabolic state. The inhibition of IL-6 trans-signaling modulated some of these traits in both 3xTg-AD and control mice, particularly during HFD, and in a sex-dependent manner. These experiments provide evidence of IL-6 trans-signaling playing a role in the CNS of a mouse model of AD. Keywords: Alzheimer’s disease; IL-6; IL-6 trans-signaling; body weight; high-fat diet; insulin; leptin 1. Introduction Alzheimer’s disease (AD) is a neurodegenerative disorder that causes the most prevalent form of dementia in elderly people. Clinically, it courses with impaired cognitive functions and memory loss caused mainly by the accumulation of β-amyloid peptides (Aβ) forming extracellular plaques and the intracellular aggregation of hyperphosphorylated Tau protein forming neurofibrillary tangles [1]. Both features cause a prominent neuroinflammation in the central nervous system (CNS) with marked microglia and astroglia activation and the release of several pro-inflammatory cytokines [2]. Cells 2020, 9, 1605; doi:10.3390/cells9071605 www.mdpi.com/journal/cells Cells 2020, 9, 1605 2 of 15 Multiple risk factors have been associated with AD; as aging is the most important, the accumulation of mtDNA mutations and mitochondrial dysfunction caused by ROS could be an early event in the disease [3]. Additionally, AD is also linked to some metabolic alterations. Being overweight in middle age (40–45 years) increases the risk of dementia, at older ages (65–75 years) the relation is U shaped, whereas at ≥76 years a higher BMI is related to decreased risk [4]. Accordingly, weight loss is commonly associated with AD patients [5]. Chronic systemic inflammation is also a landmark of obesity with secretion of systemic-acting pro-inflammatory cytokines from the adipose tissue, which can even gain access to the brain through the blood brain barrier [6]. Some studies using diet-induced obesity rodent models demonstrate impairment in memory and learning [7,8], reduced synaptic plasticity [9] and eventually compromised neurogenesis [10]. These effects may be caused by the infiltration of cytokines from the periphery and by local neuroinflammatory processes and subsequent alteration of insulin pathways in the brain [11,12]. Additionally, an altered insulin signaling in the brain contributes to Aβ aggregation and increases hyperphosphorylated tau protein levels [13]. Interleukin-6 (IL-6) is one of the potential cytokines involved in the inflammatory processes previously explained [14,15]. It has a complex signaling pathway mediated via three distinct modes. In the classic signaling, IL-6 binds to a specific membrane receptor (mIL-6R), which activates dimerization of gp130 and subsequent downstream JAK/STAT signaling. In the trans-signaling, IL-6 binds to a soluble form of mIL-6R (sIL-6R) to activate gp130. Due to the ubiquitous expression of gp130, most cells in the body can be stimulated by this process. IL-6 trans-signaling is specifically inhibited by the soluble form of gp130, sgp130 [16]. The third pathway called cluster signaling has been described as an exclusive mechanism for the communication between dendritic and T cells [17]. The implication of IL-6 in the control of body weight and metabolism has been reported previously. Wallenius and colleagues demonstrated that IL-6 KO mice develop mature-onset obesity and decreased glucose tolerance [18]. Accordingly, mice with overexpression of this cytokine in astrocytes (GFAP-IL6 mice) are resistant to high-fat diet (HFD)-induced obesity [19]. The development of conditional KO mice for different brain cells also revealed alterations in the control of body weight [20,21]. Interestingly, IL-6 trans-signaling has also been implicated in obesity, as sIL-6R is a critical chemotactic signal for adipose tissue macrophages recruitment [22]. Classic signaling is crucial for anti-inflammatory IL-6 functions; in contrast, IL-6 trans-signaling is more related to its pro-inflammatory actions, being involved in some neurodegenerative diseases such as AD and Parkinson’s disease, inflammatory colon cancer, arthritis and inflammatory bowel disease [23]. It has been demonstrated that blocking IL-6 trans-signaling with the sgp130Fc protein decreases inflammation-related damage in peripheral tissues [22] and in the CNS (by using the so-called GFAP-sgp130Fc mice, which express sgp130Fc in astrocytes [24]). Additionally, it has been discovered that the common IL-6R variant p.D358A (rs2228145), which increases its proteolysis by ADAM10 and ADAM17 and as a result increases the amount of sIL-6R, may advance the age of onset of AD in APOE ε4 carriers [25]. We hypothesized that blocking IL-6 trans-signaling with sgp130Fc in the brain would impact AD and/or obesity progression. We recently demonstrated the former in two different mouse models of AD, namely the 3xTg-AD mice and the Tg2576 mice, by crossing them with GFAP-sgp130Fc mice in order to obtain 3xTg-AD mice with the IL-6 trans-signaling pathway blocked in the CNS [26]. We herewith present results with 3xTg-AD mice, with and without transgenic sgp130Fc expression by astrocytes, and challenged with a high-fat diet to elicit obesity, which suggest that IL-6 trans-signaling modulates some of the hypermetabolic traits presented by the 3xTg-AD mice in a sex-dependent manner. Cells 2020, 9, 1605 3 of 15 2. Materials and Methods 2.1. Generation of Double Transgenic 3xTg-AD/GFAP-sgp130Fc Mice The parental strains used in this study, GFAP-sgp130Fc and 3xTg-AD mice, were generated and characterized previously [24,27,28]. To obtain the 4 genotypes of interest the following crossing strategy was designed, as in [26]. Homozygous 3xTg-AD (+/+) animals were mated with heterozygous GFAP-sgp130Fc (+/−) to obtain the double transgenic 3xTg-AD/GFAP-sgp130Fc (+/−) (+/−) and 3xTg-AD (+/−) (−/−) mice. In addition, GFAP-sgp130Fc (+/−) mice were also crossed with WT mice (C57BL/6JOlaHsd) to obtain the control groups GFAP-sgp130Fc (+/−) and WT (−/−). Genotype was determined by PCR analysis of tail DNA. (sgp130Fw: 5’-GAGGTGACATGTGTGGTGG TGGATGTTC-3’; sgp130Rv: 5’-GAGAACACGTTGCCCTGCTGCCATCTAG-3’/AppFw: 5’-GCTTGC ACCAGTTCTGGATGG-3’; AppRv: 5’-GAGGTATTCAGTCATGTGCT-3’/TauFw: 5’-GAGGTATT CAGTCATGTGCT-3’; TauRv: 5’-TTCAAAGTTCACCTGATAGT-3’). All mice used in this study were fed ad libitum (unless otherwise stated) and housed in a 12 h dark-light cycle under specific pathogen-free conditions and constant temperature (22 ± 2 ◦ C). All experimental procedures were approved by the Ethics Committee in Human and Animal experimentation (CEEAH 3541 and 3971, valid up to 06/2019 and 05/2023, respectively) from the Autonomous University of Barcelona and were in accordance with Spanish legislation and the EU directive (2010/63/UE) on ‘Protection of Animals Used for Experimental and Other Scientific Purposes’. Animals were sacrificed by decapitation to collect the brain. The right hemisphere was dissected into cortex, hippocampus and hypothalamus and flash-frozen in liquid nitrogen for gene expression and ELISA. Blood was collected from the trunk and centrifuged (9520 g for 10 minutes) at 4 ◦ C to obtain the serum. All samples were stored at −80 ◦ C until processed. Subsets of animals were sacrificed at different ages to check disease progression. 2.2. Diets The HFD (TD 03584, 58.4% kcal from fat), and the control diet (2018, 18% kcal from fat) were obtained from Harlan Iberica (Sant Feliu de Codines, Spain). The experiment had two phases and began when the mice were 17-month-old: they were first fed the control diet (n = 6–11); and then the HFD. Because of aging-related mortality, seven new 3xTg-AD male mice were added in the second phase, which showed similar responses to the other mice. Body weight and food intake were monitored regularly. In most cases mice were not isolated but kept group-housed after weaning (2–3 per cage) according to their genotype and sex. During the control diet phase, animals were subjected sequentially to an insulin tolerance test (ITT), a fasting and refeeding test (F&R), and an oral glucose tolerance test (OGTT), on weeks 3, 4 and 5, respectively. During the HFD phase, mice were subjected to ITT and OGTT only, in the weeks 6 and 8, respectively. Hypothalamus, liver, visceral (gonadal) and subcutaneous (inguinal) fat depots and brown adipose tissue were removed and weighed upon euthanasia of the animals and frozen in liquid nitrogen for further analysis. Body temperature was measured at the beginning of the control diet phase. 2.3. Body Temperature Body temperature measurement was carried out using a lubricated rectal probe (Cibertec, Madrid, Spain). The measurements were taken throughout the day, at 8, 11, 16 and 19 h. 2.4. Fasting and Refeeding (F&R) The animals were fasted overnight for 14 h and posteriorly refed. Body weight and food intake were measured regularly during the first 48 h post-fasting. Using these parameters body weight change post-fasting was calculated as well as the food intake. Cells 2020, 9, 1605 4 of 15 2.5. Insulin Tolerance Test (ITT) and Oral Glucose Tolerance Test (OGTT) The metabolic status of the mice was evaluated by an insulin tolerance test (ITT) and an oral glucose tolerance test (OGTT). For ITT, mice were injected intraperitoneally (i.p.) with 1 U/kg of insulin after a 4 h fast. For OGTT, mice were fasted for 15 h before being administered an oral glucose dose of 2 g/kg of body weight. Blood glucose was measured in tail samples at time points 0, 15, 60 and 120 min for ITT and at 0, 30, 60 and 120 min for OGTT with an ACCU CHECK glucometer (Roche). 2.6. Enzyme-Linked ImmunoSorbent Assays (ELISAs) and Colorimetric Tests ELISA kits were used to measure soluble Aβ40 , Aβ42 and sgp130Fc levels in brain samples (Amyloid beta 40 and 42 Human ELISA Kits, ThermoFisher (Vienna, Austria); Human sgp130 DuoSet ELISA, R&D Systems (Abingdon, UK) and leptin, insulin and FGF-21 in serum (rat/mouse ELISA leptin/insulin kits, Millipore (Billerica, MA, USA), ; Mouse/Rat FGF-21, R&D Systems (Minneapolis, MN, USA). In the case of Aβ measurements, brain samples were homogenized by sonication in a lysis buffer (50 nM TrisHCl, 150 nM NaCl, 2 mM EDTA, 0.1% Igepal CA 630 (Fluka, St. Louis, MO, USA, 3% SDS and 2% sodium deoxycolate) supplemented with a protease inhibitor cocktail and phenylmethylsulfonyl fluoride (Sigma-Aldrich, St. Louis, MO, USA). For sgp130Fc, brain samples were homogenized using a tissue homogenizer (Precellys, Bertin Instruments, Montigny-le-Bretonneux, France) in RIPA buffer supplemented with phosphatase and protease inhibitors. To measure cholesterol, triglycerides and glucose in serum specific colorimetric test kits were used (Cholesterol MonlabTest® /Triglycerides MonlabTests® (Selva de Mar, Spain); and Glucose-TR Spinreact (Sant Esteve de Bas, Spain). 2.7. RT-qPCR For the two-step RT-qPCR, total RNA was isolated using Maxwell simply RNA Tissue Kit (Promega, Madison, WI, USA) from hypothalamus and liver. Then, it was reverse transcribed into cDNA in a reaction using iScript cDNA synthesis kit and iTaq Universal SYBR Green Supermix (Bio-Rad Laboratories, Inc., Hercules, CA, USA). RT-qPCR for selected genes important for this study (Il6 and Agrp in hypothalamus and Fgf21 in liver) was performed in 384-well plates and run in CFX384 real-time PCR detection system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Gapdh was used as the reference gene [29,30]. (Il6Fw: 5’-GCTTA ATTACACATGTTCTCTGGGAAA-3’; Il6Rv: 5’-CAAGTGCATCATCGTTGTTCATAC-3’/AgrpFw: 5’-AGAGTTCCCAGGTCTAAGTCTG-3’; AgrpRv: 5’-GCGGTTCTGTGGATCTAGC-3’/Fgf21Fw: 5’-TTCTTTGCCAACAGCCAGAT-3’; Fgf21Rv: 5’-GTCCTCCAGCAGCAGTTCTC-3’/GapdhFw: 5’-GGCAAATTCAACGGCACA-3’; GapdhRv: 5’-CGGA GATGATGACCCTTT-3’). 2.8. Statistics Statistical calculations were done using the Statistical Package for Social Sciences version 19 (SPSS, Chicago, IL, USA) and RStudio (v 1.2.5033, R version 4.0.1, Boston, MA, USA). Males and females were analyzed separately. Most data were analyzed using the generalized linear model (GLZ) with 3xTg-AD (positive vs. negative) and sgp130Fc (positive vs. negative) as main factors; when there is a significant interaction between the main factors in the GLZ it is stated accordingly. Generalized estimated equations (GEE) were used for repeated measures (i.e., body weight, temperature, ITT and OGTT) including also both genotypes as main factors and time as a within-subject factor. In case of comparisons between two groups, an unpaired t-Student was used. Effect size was calculated using Glass’ delta (with R’s effectsize package, with small sample correction) for unpaired t-tests. For GLZ data, we indicate R2 (Equation. (1), [31]) as a measure of the variance explained by the model. Representative % of changes between 3xTg-AD and WT groups (unless otherwise stated) will be stated in some cases. Statistical significance was defined as p ≤ 0.05. Cells 2020, 9, 1605 5 of 15 Equation (1). R2 , or coefficient of determination. The numerator, σ2ε , indicates the variance of the residuals, and the denominator is the variance of the dependent variable: σ2ε R2 = 1 − P (1) n i = 1 ( yi − y)2 Cells 2020, 9, x FOR PEER REVIEW 5 of 15 3. Results 3.1.3.1. sgp130Fc sgp130Fc Influenced Influenced BodyBody Weight Weight andand FoodFood Intake Intake in ain a Sex-Dependent Sex-Dependent Manner Manner in Response to a High- in Response to aFat Diet Diet High-Fat Having Having in mind in mind that that during during thethe controldiet control dietphase phasethe themice mice were were subjected to an an ITT, ITT,aaF&R F&Rand andananOGTT, OGTT,thetheresults indicate results that indicate 3xTg-AD that 3xTg-AD animals hadhad animals lower bodybody lower weight (decreasing weight for instance (decreasing for instance a ~13.5% and ~21.3% in 3xTg-AD vs WT in male and female mice, respectively). This decreasewas a ~13.5% and ~21.3% in 3xTg-AD vs WT in male and female mice, respectively). This decrease wassignificant significantwhen when comparing comparing thethe four groups four groupsin both sexes in both (males sexes p = 0.039; p = 0.039; (males females p < 0.001), p < 0.001), females whereas the inhibition of IL-6 trans-signaling did not have a clear effect (Figure 1A). Despite their whereas the inhibition of IL-6 trans-signaling did not have a clear effect (Figure 1A). Despite their lower lower body bodyweight, absolute weight, absolute(g/day) food (g/day) intake food waswas intake significantly increased significantly in 3xTg-AD increased in 3xTg-ADmicemice (increasing ~25.6% (increasing and and ~25.6% 12.8% in 3xTg-AD 12.8% in 3xTg-AD mice vs WT mice vs WTmice in in mice male and male female and femalemice, mice,respectively) (malespp<<0.001 respectively)(males 0.001R2 = R2 =0.54; females 0.54; females p =p =0.002 0.002 = 0.33) R2R=2 0.33) (Figure (Figure1B). NoNo 1B). significant significanteffect ofofthe effect thesgp130Fc sgp130Fcprotein proteinwas wasobserved. observed. Figure 1. Body weight and food intake alterations in control diet. (A) During the control diet phase, Figure 1. Body weight and food intake alterations in control diet. (A) During the control diet phase, 3xTg-AD mice presented lower body weight. The inhibition of the IL-6 trans-signaling did not have 3xTg-AD mice presented lower body weight. The inhibition of the IL-6 trans-signaling did not have a significant effect. (B) 3xTg-AD mice presented increased food intake in both sexes, with no effect of a significant effect. (B) 3xTg-AD mice presented increased food intake in both sexes, with no effect of the sgp130Fc protein. (C) Sgp130Fc quantification by ELISA in brain cortex of AD mice. Results are the sgp130Fc protein. (C) Sgp130Fc quantification by ELISA in brain cortex of AD mice. Results are MEAN ± SEM. F p ≤ 0.05 vs. 3xTg-AD negative; N p ≤ 0.05 vs. sgp130Fc negative. MEAN ± SEM. ★ p≤ 0.05 vs. 3xTg-AD negative; ▲ p≤ 0.05 vs. sgp130Fc negative. Despite their advanced age, in general mice gained weight during the HFD phase, and as seen Despite their advanced age, in general mice gained weight during the HFD phase, and as seen with the control diet, with HFD the 3xTg-AD mice again had lower body weight (decreasing a ~18.5% with the control diet, with HFD the 3xTg-AD mice again had lower body weight (decreasing a ~18.5% and ~16.0% in 3xTg-AD vs WT in male and female mice, respectively). This decrease was significant and ~16.0% in 3xTg-AD vs WT in male and female mice, respectively). This decrease was significant when comparing the four groups (males p = 0.009; females p < 0.001) (Figure 2A). Additionally, when comparing the four groups (males p = 0.009; females p < 0.001) (Figure 2A). Additionally, 3xTg- 3xTg-AD male mice presented higher food intake (~18.2% in 3xTg-AD vs WT) (p < 0.001 R2 = 0.50) AD male mice presented higher food intake (~18.2% in 3xTg-AD vs WT) (p < 0.001 R2 = 0.50) (Figure 2B). The inhibition of IL-6 trans-signaling did not have a significant effect on the body weight in male mice, although, it is likely that this rather indicates that the 3xTg-AD/GFAP-sgp130Fc male mice have recovered better from the different experimental procedures they were subjected to. Thus, at the end of the HFD phase, a similar body weight difference (compared to 3xTg-AD mice) to that observed at Cells 2020, 9, 1605 6 of 15 (Figure 2B). The inhibition of IL-6 trans-signaling did not have a significant effect on the body weight in male mice, although, it is likely that this rather indicates that the 3xTg-AD/GFAP-sgp130Fc male Cellshave mice 2020, recovered 9, x FOR PEER REVIEW better 6 of 15 from the different experimental procedures they were subjected to. Thus, at the end of the HFD phase, a similar body weight difference (compared to 3xTg-AD mice) to that it was increased in 3xTg-AD/GFAP-sgp130Fc (~17.8% vs GFAP-sgp130Fc) but not in 3xTg-AD (vs observed at the beginning of the control phase is now apparent. In contrast, in females the inhibition WT) females (p = 0.049 R2 = 0.50). of IL-6 trans-signaling decreased HFD-induced obesity of both 3xTg-AD and control mice (p < 0.001). The weights of liver and fat tissues (Figure 2C) were in concordance with those of body weight. Results for food intake were similar to those of the control diet phase, but significant interaction Thus, all of them were decreased (from ~29 to 74% in 3xTg-AD vs WT) significantly in 3xTg-AD male emerged since it was increased in 3xTg-AD/GFAP-sgp130Fc (~17.8% vs GFAP-sgp130Fc) but not in (WATSc p < 0.001 R2 = 0.50; WATGon p = 0.032 R2 = 0.27; BAT p = 0.001 R2 = 0.38; Liver p = 0.002 R2 = 3xTg-AD (vs WT) females (p = 0.049 R2 = 0.50). 0.31) and female (from ~23 to 56% in 3xTg-AD vs WT) mice (WATSc p < 0.001 R2 = 0.69; WATGon p = The weights of liver and fat tissues (Figure 2C) were in concordance with those of body weight. 0.002 R2 = 0.50; BAT p < 0.001 R2 = 0.64; Liver p = 0.008 R2 = 0.37), and the inhibition of IL-6 trans- Thus, all of them were decreased (from ~29 to 74% in 3xTg-AD vs WT) significantly in 3xTg-AD male signaling further decreased them in all females (WATSc p < 0.005 ; WATGon p = 0.005; BAT p = 0.001) (WATSc p < 0.001 R2 = 0.50; WATGon p = 0.032 R2 = 0.27; BAT p = 0.001 R2 = 0.38; Liver p = 0.002 The presence of sgp130Fc was measured in brain cortex from 3xTg-AD and 3xTg-AD/GFAP- R2 = 0.31) and female (from ~23 to 56% in 3xTg-AD vs. WT) mice (WATSc p < 0.001 R2 = 0.69; WATGon sgp130Fc2mice at 10 months of age2by ELISA (Figure 1C). As 2expected, a clear increase in sgp130Fc p = 0.002 R = 0.50; BAT p < 0.001 R = 0.64; Liver p = 0.008 R = 0.37), and the inhibition of IL-6 immunostaining was observed in 3xTg-AD/GFAP-sgp130Fc mice (males p = 0.001 Glass’ delta =33.42; trans-signaling further decreased them in all females (WATSc p < 0.005 ; WATGon p = 0.005; BAT females p = 0.001 Glass’ delta =34.42), with no differences between males and females. p = 0.001). Figure 2. Body weight and food intake alterations in high-fat diet. (A) HFD caused obesity, and again Figure 2. 3xTg-AD Body mice weight and presented foodbody lower intake alterations weight. The in high-fat diet. inhibition (A)trans-signaling of IL-6 HFD caused obesity, did not and again have a significant effect in male mice, whereas in females it decreased HFD-induced obesity of both 3xTg-AD a 3xTg-AD mice presented lower body weight. The inhibition of IL-6 trans-signaling did not have andsignificant effect(B) control mice. in 3xTg-AD male mice, whereas male in females mice had it decreased also higher HFD-induced food intake obesity in this phase. Thisofeffect bothwas 3xTg- notAD and control observed mice. (B) in females. The3xTg-AD inhibitionmale mice of IL-6 had also higher trans-signaling hadfood intakeon no effects in males, this phase. This effect in females, it was not food decreased observed in females. intake, The inhibition more markedly in the WTof IL-6 thantrans-signaling had no effects in 3xTg-AD animals. (C) Theonabsolute males, inweight females, it decreased from the adiposefood intake, tissue more and depots markedly in the liver was WT than inin3xTg-AD diminished 3xTg-AD animals. mice. (C) TheThe absolute blocking of weight IL-6 from the adipose tissue depots and liver was diminished in 3xTg-AD mice. The blocking trans-signaling further diminished the absolute tissue weights in females. Results are MEAN ± SEM. of IL-6 trans- F signaling further p ≤ 0.05 vs. diminished 3xTg-AD negative;the N pabsolute ≤ 0.05 vs.tissue weights sgp130Fc in females. negative; Resultsaare  Indicates ± SEM. ★ p ≤ MEAN interaction significant 0.05 vs. between factors.negative; ▲ p ≤ 0.05 vs. sgp130Fc negative; ◆ Indicates a significant interaction 3xTg-AD both between both factors. 3.2. 3xTg-AD Mice Recovered the Body Weight Quicker and Had a Higher Food Intake after an Overnight Fasting. The Inhibition of IL-6 Trans-Signaling Attenuated These Effects Especially in 3xTg-AD Males As expected, all mice lost weight after an overnight fasting (Figure 3A). In females, this loss was higher in 3xTg-AD mice regardless of sgp130Fc expression (p = 0.015 R2=0.25). No differences were Cells 2020, 9, 1605 7 of 15 The presence of sgp130Fc was measured in brain cortex from 3xTg-AD and 3xTg-AD/GFAP-sgp130Fc mice at 10 months of age by ELISA (Figure 1C). As expected, a clear increase in sgp130Fc immunostaining was observed in 3xTg-AD/GFAP-sgp130Fc mice (males p = 0.001 Glass’ delta = 33.42; females p = 0.001 Glass’ delta = 34.42), with no differences between males and females. 3.2. 3xTg-AD Mice Recovered the Body Weight Quicker and Had a Higher Food Intake after an Overnight Fasting. The Inhibition of IL-6 Trans-Signaling Attenuated These Effects Especially in 3xTg-AD Males Cells 2020, 9, x FOR PEER As expected, REVIEW all mice lost weight after an overnight fasting (Figure 3A). In females, this loss7 was of 15 higher in 3xTg-AD mice regardless of sgp130Fc expression (p = 0.015 R = 0.25). No differences were 2 observed in observed in males. males. Once Once the the animals animals werewere re-fed re-fed their their body body weight weight increased; increased; the the 3xTg-AD 3xTg-AD mice mice recovered faster in both sexes (males t4 p = 0.001 R2 = 0.45, t24 p = 0.029 R 2 = 0.42, recovered faster in both sexes (males t4 p = 0.001 R = 0.45, t24 p = 0.029 R = 0.42, t48 p = 0.026 2 2 t48 p = 0.026 R2 = 0.32; Rfemales 2 = 0.32;t4females p = 0.002t4Rp2 = = 0.33, 0.002t24 R2 p==0.33, 0.032t24R2 p= = 0.28), 0.032 a response R2 = 0.28),blunted by the a response blocking blunted of IL-6 by the trans- blocking signaling, of especially in IL-6 trans-signaling, males. inAbsolute especially food intake males. Absolute measurements food intake measurements (Figure 3B)3B) (Figure showed showeda compensatory hyperphagia in 3xTg-AD mice after fasting, evident in both a compensatory hyperphagia in 3xTg-AD mice after fasting, evident in both sexes (males p < 0.001 sexes (males p < 0.001 R2 = R0.66; females p < 0.001 p