Maternal selenium status is profoundly involved in metabolic fetal programming by modulating insulin resistance, oxidative balance and energy homeostasis

1 This is an Accepted Manuscript of an article published by Springer in European Journal of 1 nutrition on Epub 2018 Dec 1 available at: doi: 10.1007/s00394-018-1861-4. 2 3 4 MATERNAL SELENIUM STATUS IS PROFOUNDLY INVOLVED IN METABOLIC FETAL 5 PROGRAMMING BY MODULATING INSULIN RESISTANCE, OXIDATIVE BALANCE AND ENERGY 6 HOMEOSTASIS. 7 Ojeda María Luisa, Nogales Fátima*, Membrilla Alba, Carreras Olimpia. 8 Department of Physiology, Faculty of Pharmacy, Seville University, 41012 Seville, Spain. 9 10 *Correspondence to: Dra. Fátima Nogales Bueno 11 Department of Physiology. 12 Faculty of Pharmacy, Seville University. 13 C/ Profesor García González, nº 2. 14 41012. Sevilla. Spain. 15 Tel: +34 954556518 16 Fax: +34 954233765 17 E-mail: [email protected] 18 Running title: Selenium involvement in metabolic fetal programming. 19 20 21 22 23 24 2 25 26 27 28 29 ABSTRACT: 30 Purpose: High and low levels of selenium (Se) have been related to metabolic disorders 31 in dams and in their offspring. Their relationship to oxidative balance and to AMP- 32 activated protein kinase (AMPK), are some of the mechanisms proposed. The aim of 33 this study is to acquire information about how Se is involved in metabolic 34 programming. 35 Methods: three experimental groups of dam rats were used: control (Se: 0.1ppm), Se- 36 supplemented (Se: 0.5ppm) and Se-deficient (Se: 0.01ppm). At the end of lactation, the 37 pups’ metabolic profile, oxidative balance, Se levels, selenoproteins and IRS-1 hepatic 38 expression, as well as hepatic AMPK activation were measured. 39 Results: The experimental groups present deep changes in Se homeostasis, 40 selenoproteins and IRS-1 hepatic expression, oxidative balance, AMPK activation ratio 41 and insulin levels. They do, however, have different metabolic profiles. 42 Conclusions: High- and low- Se diets are linked to insulin resistance, yet the 43 mechanisms involved are completely opposite. 44 45 Key words: Dietary Selenium, metabolic programming, insulin resistance, oxidative balance, energy 46 homeostasis. 47 48 3 49 50 51 52 1. INTRODUCTION: 53 Trace elements, such as Se, Zn, Cr and Li are known to influence the endocrine regulation of 54 energy metabolism in mammals, mainly by enhancing or interfering with components of the 55 insulin signaling cascade, affecting carbohydrate and lipid metabolism and energy homeostasis [1]. 56 Se plays its biological functions by forming part of 25 selenoproteins [2], among which the family 57 of the antioxidant enzymes glutathione peroxidase GPx, and the Se plasma transporter 58 selenoprotein P (SelP) have recently been related to Insulin Resistance (IR) and Metabolic 59 Syndrome (MS). MS is one of the most important metabolic disorders affecting the world 60 population. It is defined as a cluster of risk factors including obesity, IR, raised blood pressure and 61 dyslipidemia, predisposing the sufferer to cardiovascular diseases and diabetes [3]. This syndrome 62 appears among 20% of pregnant women, affecting both mother and offspring [4], even during the 63 latter’s adulthood, through metabolic fetal programming [5]. In an experimental rat MS model, Se 64 body distribution and selenoprotein activities have been found to be altered in both dams and 65 their offspring [6-7] . However, depending on the tissue under study, Se tissue deposits have been 66 both increased and decreased. This makes it difficult to understand whether Se supplementation 67 could be an effective therapy for MS development or, indeed, pose a risk [8]. 68 In this context, both infra- and supra- Selenoprotein regulations have been involved in the 69 development of IR and MS [9-11]. Seale et al., [12], found that upon dietary restriction, 70 selenocysteine lyase knockout mice cannot supply Se for selenoprotein biosynthesis and that they 71 develop fatty liver, hypercholesterolemia, and glucose intolerance; GPx1 and SelP expression 72 4 decreased; oxidative stress appeared and insulin-signaling inhibitor protein-tyrosine phosphatase 73 1B (PTP1B) levels increased. Reddi and Bollineni, [13] found that Se deficiency impaired pancreas 74 islet function and free radical-scavenging systems in rats, resulting in a decreased insulin secretory 75 reserve. If Se levels are below 80 μg/l in the offspring of diabetic patients, Se correlates inversely 76 with insulin resistance [14]. However, no positive effect on diabetes prevention with Se was found 77 and some data even pointed towards supranutritional Se status as an unexpected risk factor in 78 potentiating IR [15]. 79 As an excess of reactive oxygen species (ROS) causes damage to mitochondrial components, which 80 are involved in the pathogenesis and etiology of IR [16], the antioxidant properties of 81 selenoproteins are necessary to prevent metabolic disorders. However, when insulin reaches its 82 receptor, a small amount of ROS, necessary for the insulin to act, are liberated. These species 83 deactivate the insulin-signaling inhibitors’ phosphatase protein (PTEN) and protein tyrosine 84 phosphatase 1B (PTP-1B) which contribute to the insulin signaling process. When GPx1 acts, this 85 process is prevented, thus leading to IR. In agreement with the above, Steinbrenner, [17] showed 86 that GPx1 and/or SelP inhibited phosphorylation (activation) of key mediators, such as protein 87 kinase B (Akt) and adenosine monophosphate-activated protein kinase (AMPK), in energy 88 metabolism in liver and/or skeletal muscle. AMPK is an energy status sensor that controls cellular 89 energy homeostasis and activates energy production processes by the stimulation of catabolic 90 pathways and the inactivation of processes involved in ATP consumption [18]. Supranutritional Se 91 supply also induces alterations in energy-metabolism-related molecular targets in the skeletal 92 muscle and visceral adipose tissue of pigs [19]. Recently Tajima-Shirasaki et al., [20] have found 93 that suppressing SelP may provide a novel therapeutic approach to treating type 2 diabetes in rat 94 hepatoma cell line. Using eicosapentaenoic acid, they suppressed SelP expression by inactivating 95 sterol regulatory element-binding protein-1c and caused a reduction in SelP expression. This 96 reduction, moreover, activates AMPK, preventing IR induction and vascular endothelial growth 97 factor in type 2 diabetes. This is, however, an in vitro study and Se status in different tissues such 98 5 as muscle and heart are also important key factors that need to be analyzed. Zhou et al., [21] 99 found that Se-enriched exopolysaccharides alleviate adipose inflammation in diabetic mice by 100 exerting anti-diabetic effects. GPx1 also plays an important role in pancreas, stimulating different 101 molecules involved in insulin synthesis and secretion [22]. SelP did not, however, show any effect 102 on beta cell mass or insulin synthesis and secretion. 103 Chronic metabolic alterations such as IR, MS or type 2 diabetes leads to cellular oxidative 104 dysfunction, but also to endoplasmic reticulum (ER) stress and inflammation. The ER is an 105 organelle specialized in integrating cellular stress responses, and has seven ER-resident 106 selenoproteins necessary for its correct function [23]. At the present moment, therefore, there is 107 no clear understanding of the relationship between Se status and IR. 108 During healthy pregnancy, maternal organs and placenta adapt to physiological changes 109 related to hormones and energy homeostasis. Therefore, during this period, women have a 110 greater predisposition to suffering metabolic disorders, which could even, in some cases, have 111 repercussion both on them and their progeny. For this reason, the aim of the present study is 112 to analyze the repercussion of Se status in metabolic fetal programing by modulating insulin 113 resistance, oxidative balance and energy homeostasis. 114 2. MATERIAL AND METHODS. 115 2.1. Animals. Male and female Wistar rats (Centre of Production and Animal experimentation, 116 Vice-rector’s Office for Scientific Research, University of Seville) weighing approximately 150- 117 200 g, were randomised into three groups: control (C), selenium supplemented (SS) and 118 selenium deficient (SD) groups. Animal care procedures and experimental protocols were 119 performed in accordance with EU regulations (Council Directive 86/609/EEC, November 24th 120 1986) and approved by the Ethics Committee of the University of Seville. All rats received 121 drinking water and diet ad libitum during three week before mate, and then, during gestation 122 (3 weeks) and lactation (3 weeks) periods. C, SS and SD groups received solid diets with 0.1, 123 6 0.5 or 0.01 ppm of Se respectively. Se was supplemented as anhydrous sodium selenite (an 124 inorganic compound; Panreac, Barcelona, Spain). The diets of these rats were prepared 125 according to The Council of the Institute of Laboratory Animal Resources (ILAR, 1979) which 126 details known nutrient requirements for most of the common laboratory animals. 127 In four week, male (n=3) and female (n=6) rats were mated to obtain the first-generation 128 offspring for each group. Pregnant female rats were inspected daily by the presence of the 129 vaginal plug, which indicated day zero of pregnancy; at this moment pregnant rats were 130 housed individually in plastic cages. The day of parturition, which occurs spontaneously three 131 weeks after coitus, was designated as day 1 of lactation. The offspring number was reduced to 132 8 per mother at parturition (four males and four females, when this was possible). The 133 experiments were performed on the offspring of all groups to 21d postpartum. In this study, 134 we have used 8 pups per group to measure all the parameters cited below. These 8 pups 135 represent all the litters, as a maximum of 2 rats per litter, and were allocated to each group 136 taking into account the sex. 137 2.2. Nutritional controls. Body weights of the dam rats were determined once a week while 138 that the amount of food and liquid consumed by rats were monitored daily until the end of the 139 experimental period. Se intake was calculated by multiplying the food consumed by ppm of Se 140 in the diets. Weekly, body weight and cranium-caudal length of pups was controlled, using a 141 metric calliper, until end of the experimental period, to calculate body mass index (BMI) 142 according to the formula: Body weight (g)/length2 (cm2). All measures were taken at 9:00 am 143 to avoid changes due to circadian rhythms. 144 2.3. Samples. The amount of milk consumed by the offspring at the end of the lactation period 145 (days 19 and 20) was estimated by subtracting the weight of the pups obtained immediately 146 prior to returning them to the dam from their weight after 30 minutes of suckling. In order to 147 obtain the maximum amount of milk at day 21 of lactation, 3h after removing the litters from 148 their mothers, the dams were anesthetized with urethane, and milk samples were immediately 149 7 collected. The milk was obtained by gently massaging the area around each of the 12 150 mammary glands and then pressing upward from the base of the gland towards the nipple. 151 The amount of milk collected was around 1 to 1.5 ml per dam. 152 At the end of the experimental period, dams and their pups were weighed and anesthetized 153 with intraperitoneal 28% w/v urethane (0.5 ml/100 g of body weight). Blood samples were 154 obtained by heart puncture and collected in tubes. The serum was prepared using low-speed 155 centrifugation for 15min. at 1300 x g. The abdomen was opened by a midline incision and 156 pancreas and liver were removed, debrided of adipose and connective tissue in ice-cold saline, 157 weighed and stored at −80◦C prior to biochemical determinations. Hepatic and pancreatic 158 somatic index (HSI and PSI) were calculated as (liver or pancreas weight / total body weight). 159 2.4. Metabolic profile. The metabolic profile was determined in offspring before sacrifice, in 160 blood from their tails. Glucose, triglycerides (TG) and cholesterol were determined using test 161 strips Accutrend (ROCHE, Spain). Serum insulin was determined by using a rat insulin ELISA kit 162 (BioVendor GmbH, Heidelberg, Germany) according to the manufacturer’s instructions. The 163 model homeostasis assessment of insulin resistance index (HOMA-IR) was calculated according 164 to the following formula: (Fasting glucose concentration x Fasting insulin serum 165 concentration)/ 405. Creatinine and urea in serum was determined by colorimetric methods 166 using a commercial kits (BioSystems kit (Barcelona, Spain) and Randox diagnostic kit (Crumlin 167 Co., Antrim, UK) respectively). 168 2.5. Selenium analysis. Selenium levels were determined by graphite-furnace atomic 169 absorption spectrometry, using a PerkinElmer AAnalyst™ 800 high-performance atomic 170 absorption spectrometer with WinLab32 for AA software, equipped with a Transversely 171 Heated Graphite Furnace (THGA) with longitudinal Zeeman-effect background corrector and 172 an AS-furnace autosampler (PerkinElmer, Überlingen, Germany). The source of radiation was a 173 Se electrodeless discharge lamp (EDL). The instrumental operating conditions and the reagents 174 are the same that we have used in the previous paper [24]. Samples: serum samples were 175 8 diluted fivefold in 0.2% v/v HNO3 and 0.2% Triton X-100 solutions and urine samples were 176 diluted 1:2 v/v. After 72h at 100ºC dry temperature, pancreas, liver and milk samples were 177 weighed and digested in a sand bath heater (OVAN, Badalona, Spain) with nitric acid for 72h., 178 and perchloric acid and chlorhydric acid (6N) were added. 179 2.6. Antioxidant enzymes and oxidative stress markers. In order to measure the activity of 180 antioxidant enzymes (SOD, CAT, GPx and GR) as well as lipid and protein oxidation (levels of 181 MDA and carbonyl group (CG) respestively), liver tissue samples were homogenized (100 x g 182 for 1min, 1:4 w/v) using a Potter homogenizer (Pobel 245432, Madrid, Spain) in a sucrose 183 buffer (15 mM Tris/HCl, pH 7.4, 250 mM sucrose, 1 mM EDTA and 1 mM dithiothreitol) in an 184 ice bath. The homogenate was centrifuged at 900 x g for 10min at 4 ºC. The resulting 185 supernatant was employed for the biochemical assay according to techniques described in 186 [25]. 187 2.7. Immunoblotting assays. The expression of total and phosphorylated AMPK (AMPK and 188 pAMPK), IRS-1, SelP and GPx1 were determined in hepatic homogenates (with Protease 189 Inhibitor (ROCHE) and also Phosphatase Inhibitor (ROCHE, Spain) to pAMPK) according to 190 method deeply described in (Nogales et al., (2017). The samples utilized, which contained 100 191 µg of protein, were incubated overnight at 4ºC with specific primary antibodies (rabbit 192 polyclonal IgG, Santa Cruz Biotechnology) in dilutions: AMPK and pAMPK (1:4000), IRS-1 193 (1:1000), SelP and GPx1 (1:20000); and secondary antibody (anti-rabbit IgG HRP conjugate, 194 Santa Cruz Biotechnology) in dilutions: AMPK α½ and pAMPK α½ (1:4000, 1:2000), and IRS-1 195 (1:2500), SelP and GPx1 (1:5000). Monoclonal mouse anti β-actin (IgG1, Sigma-Aldrich, Spain) 196 (1:20000) was used to detect β-actin as a loading control, and a secondary antibody (anti- 197 mouse IgG HRP conjugate, Sigma-Aldrich, Spain) in dilutions 1:8000. The quantification of the 198 blots was performed by densitometry with PCBAS 2.08e software analysis (Raytest Inc, 199 Germany). The results were expressed as percent arbitrary relative units referred to values in 200 control pups which were defined as 100%. The samples from experimental and control animals 201 9 run on the same gel. Activation of AMPK was determined by the ratio pAMPK/ tAMPK, which 202 was calculated dividing pAMPK expression by AMPK expression values. 203 2.8. Statistical Analysis. The results are expressed as means  standard error of the mean 204 (SEM). The data were analysed using a statistical program (GraphPad InStat 3, CA, USA) by 205 analysis of variance (one-way ANOVA). The statistical significance was established at p<0.05. 206 When ANOVA resulted in differences, multiple comparisons between means were studied by 207 the Tukey-Kramer test. 208 3. RESULTS: 209 3.1. Effects of selenium status on food intake, morphological changes and Se body 210 distribution. Table 1 shows that SS dams intake a higher amount of solid diet than 211 controls, reporting an increase in weight gain during lactation. SD dams, unexpectedly, 212 present the highest solid intake figures; however, these dams have lower increase in body 213 weight, which is consistent with the lowest Se intake. The same pattern appears in their 214 offspring, SD pups have lower body weight gain and cranium-caudal length than controls, 215 but their BMI is not altered. On the contrary, SS pups have higher body weight increase and 216 BMI than control ones. 217 Milk Se concentration decreases in SD dams; their pups, however, present normal Se 218 serum values, yet drastically low Se levels in liver and pancreas, presenting this last tissue 219 a poor development. SS offspring receive higher amount of Se via milk, they have higher 220 serum and hepatic Se levels than control pups, along with hepatomegaly. However 221 pancreas Se deposits are not altered. 222 3.2. Selenium status and hepatic oxidative balance. With respect to control animals, the 223 pups which received a Se-deficient diet have higher SOD activity, lower CAT and GPx 224 activity and a lower antioxidant enzymes ratio, consistent with protein oxidation. SS 225 16 Conflict of interest: On behalf of all authors, the corresponding author states that there is no 375 conflict of interest. 376 377 378 REFERENCES. 379 [1] Wiernsperger N, Rapin J (2010) Trace elements in glucometabolic disorders: an update. 380 Diabetol Metab Syndr 2:70. 381 [2] Carreras O, Ojeda ML,Nogales F (2016) Selenium Dietary Supplementation and 382 Oxidative Balance in Alcoholism In: Patel V (ed) Molecular Aspects of Alcohol and 383 Nutrition, 1st edn. Elsevier, UK, pp. 133–142. 384 [3] Day C (2007) Metabolic syndrome, or What you will: definitions and epidemiology. Diab 385 Vasc Dis Res 4:32. 386 [4] Zou M,Arentson EJ, Teegarden D et al (2012) Fructose consumption during pregnancy 387 and lactation induces fatty liver and glucose intolerance in rats. Nutr Res 32:588–598. 388 [5] Fowden AL, Forhead AJ (2004) Endocrine mechanisms of intrauterine programming. 389 Reproduction 127: 515–526. 390 [6] Nogales F, Ojeda ML, Muñoz del Valle P, Serrano A, Murillo ML, Carreras O (2017) 391 Metabolic syndrome and selenium during gestation and lactation. Eur J Nutr 56:819– 392 830. 393 [7] Ojeda ML, Nogales F, Muñoz Del Valle P, Díaz-Castro j, Murillo ML,Carreras O (2016) 394 Metabolic syndrome and selenium in fetal programming: Gender differences. Food 395 Funct 7: 3031-3038. 396 [8] Serrano A, Nogales F, Sobrino P, Murillo ML, Carreras O, Ojeda ML (2016) Heart 397 selenoproteins status of metabolic syndrome-exposed pups: A potential target for 398 attenuating cardiac damage. Mol Nutr Food Res 60: 2633-2641. 399 17 [9] Zhou J, Huang K, Lei XG (2013) Selenium and diabetes—Evidence from animal studies. 400 Free Radic Biol Med 65:1548–1556. 401 [10] Labunskyy VM, Lee BC, Handy DE et al (2011) Both maximal expression of 402 selenoproteins and selenoprotein deficiency can promote development of type 2 403 diabetes-like phenotype in mice. Antioxid Redox Signal 14:2327–2336. 404 [11] Wang X, Zhang W, Chen H et al (2014) High selenium impairs hepatic insulin sensitivity 405 through opposite regulation of ROS. Toxicol Lett 224:16–23. 406 [12] Seale LA, Hashimoto AC, Kurokawa S et al (2012) Disruption of the Selenocysteine 407 Lyase-Mediated Selenium Recycling Pathway Leads to Metabolic Syndrome in Mice. 408 Mol Cell Biol 32:4141–4154. 409 [13] Reddi AS Bollineni JS (2001) Selenium-deficient diet induces renal oxidative stress and 410 injury via TGF-β1 in normal and diabetic rats. Kidney Int 59:1342–1353. 411 [14] Ozkaya M, Sahin M, Cakal E et al (2009) Selenium Levels in First-Degree Relatives of 412 Diabetic Patients. Biol Trace Elem Res 128:144–151. 413 [15] Stranges S, Marshall JR, Natarajan R (2007) Effects of long-term selenium 414 supplementation on the incidence of type 2 diabetes: a randomized trial. Ann Intern 415 Med 147:217–223. 416 [16] Houstis N, Rosen ED, Lander ES (2006) Reactive oxygen species have a causal role in 417 multiple forms of insulin resistance. Nat 440: 944–948. 418 [17] Steinbrenner H (2013) Interference of selenium and selenoproteins with the insulin- 419 regulated carbohydrate and lipid metabolism. Free Radic Biol Med 65:1538–1547. 420 [18] Hardie DG (2015) AMPK: positive and negative regulation, and its role in whole-body 421 energy homeostasis. Curr Opin Cell Biol 33:1–7. 422 [19] Pinto A, Juniper DT, Sanil M (2012) Supranutritional selenium induces alterations in 423 molecular targets related to energy metabolism in skeletal muscle and visceral adipose 424 tissue of pigs. J Inorg Biochem 114:47–54. 425 18 [20] Tajima-Shirasaki N, Ishii KA, Takayama H et al (2017) Eicosapentaenoic acid down- 426 regulates expression of the selenoprotein P gene by inhibiting SREBP-1c protein 427 independently of the AMP-activated protein kinase pathway in H4IIEC3 hepatocytes. J 428 Biol Chem 292:10791–10800. 429 [21] Zhou X, Wang F, Yang H et al (2014) Selenium-enriched exopolysaccharides produced 430 by Enterobacter cloacae Z0206 alleviate adipose inflammation in diabetic KKAy mice 431 through the AMPK/SirT1 pathway. Mol Med Rep 9:683–688. 432 [22] Pepper MP, Vatamaniuk MZ, Yan X et al (2011) Impacts of Dietary Selenium Deficiency 433 on Me tabolic Phenotypes of Diet-Restricted GPX1-Overexpressing Mice. Antioxid 434 Redox Signal 14:383–390. 435 [23] Addinsall AB, Wright CR, Andrikopoulos S, van der Poel C, Stupka N (2018) Emerging 436 roles of endoplasmic reticulum-resident selenoproteins in the regulation of cellular 437 stress responses and the implications for metabolic disease. Biochem J 475:1037-1057. 438 [24] Ojeda ML, Nogales F, Vázquez B et al (2009)Alcohol, Gestation and Breastfeeding: 439 Selenium as an Antioxidant Therapy,” Alcohol Alcohol. 44:272–277. 440 [25] Nogales F, Ojeda ML, Fenutría M et al (2013) Role of selenium and glutathione 441 peroxidase on development, growth, and oxidative balance in rat offspring 442 Reproduction 146: 659-667. 443 [26] Zhao H, Li K, Tang JY et al (2015) Expression of Selenoprotein Genes Is Affected by 444 Obesity of Pigs Fed a High-Fat Diet. J Nutr 145:1394–1401. 445 [27] Metzger BE, Buchanan TA, Coustan DR et al (2007) Summary and Recommendations of 446 the Fifth International Workshop-Conference on Gestational Diabetes Mellitus. 447 Diabetes Care 30:S251–S260. 448 [28] Ojeda ML, Jotty K, Nogales F et al (2010) Selenium or selenium plus folic acid intake 449 improves the detrimental effects of ethanol on pups’ Selenium balance. Food Chem 450 Toxicol 48: 3486-3491. 451 19 [29] Feoli AM, Macagnan FE, Piovesan CH (2014) Xanthine Oxidase Activity Is Associated 452 with Risk Factors for Cardiovascular Disease and Inflammatory and Oxidative Status 453 Markers in Metabolic Syndrome: Effects of a Single Exercise Session. Oxid Med Cell 454 Longev 2014:587083. 455 [30] Misu H, Takamura T, Takayama H et al (2010) A Liver-Derived Secretory Protein, 456 Selenoprotein P, Causes Insulin Resistance. Cell Metab 12:483–495. 457 [31] Tanti JF, Jager J (2009) Cellular mechanisms of insulin resistance: role of stress- 458 regulated serine kinases and insulin receptor substrates (IRS) serine phosphorylation. 459 Curr Opin Pharmacol 9:753–762. 460 [32] Sonntag AG, Dalle Pezze P, Shanley DP, Thedieck K (2012) A modelling-experimental 461 approach reveals insulin receptor substrate (IRS)-dependent regulation of adenosine 462 monosphosphate-dependent kinase (AMPK) by insulin. FEBS J: 279:3314–3328. 463 [33] Bijland S, Mancini SJ, Salt IP (2013) Role of AMP-activated protein kinase in adipose 464 tissue metabolism and inflammation. Clin Sci 124:491–507. 465 [34] Ruderman NB, Carling D, Prentki M, Cacicedo JM (2013) AMPK, insulin resistance, and 466 the metabolic syndrome. J Clin Invest 123:2764-2772. 467 [35] Mistry HD, Broughton Pipkin F et al (2012) Selenium in reproductive health. Am J 468 Obstet Gynecol 206:21–30. 469 [36] Hamieh A, Cartier D, Abid H, et al. (2017) Selenoprotein T is a novel OST subunit that 470 regulates UPR signaling and hormone secretion. EMBO Rep 18:1935-1946. 471 [37] Imai H, Hirao F, Sakamoto T et al (2003) Early embryonic lethality caused by targeted 472 disruption of the mouse PHGPx gene. Biochem Biophys Res Commun 305:278–86. 473 [38] Shi D, Guo S, Liao S et al (2012) Influence of Selenium on Hepatic Mitochondrial 474 Antioxidant Capacity in Ducklings Intoxicated with Aflatoxin B1. Biol Trace Elem Res 475 145:325–329. 476 [39] He S, Guo X, Tan W et al (2016) Effect of Selenium Deficiency on Phosphorylation of the 477 20 AMPK Pathway in Rats. Biol Trace Elem Res. 169:254–260. 478 [40] Yoon MS (2017) mTOR as a Key Regulator in Maintaining Skeletal Muscle Mass. Front 479 Physiol 8:788. 480 [41] Gong T, Torres DJ, Berry MJ, Pitts MW (2018) Hypothalamic redox balance and leptin 481 signaling - Emerging role of selenoproteins. Free Radic Biol Med S0891-5849:30103- 482 30105. 483 [42] Yang J, Hamid S, Cai J et al (2017) Selenium deficiency-induced thioredoxin suppression 484 and thioredoxin knock down disbalanced insulin responsiveness in chicken 485 cardiomyocytes through PI3K/Akt pathway inhibition. Cell Signal 38:192–200. 486 [43] Xu J, Wang L, Tang J et al (2017) Pancreatic atrophy caused by dietary selenium 487 deficiency induces hypoinsulinemic hyperglycemia via global down-regulation of 488 selenoprotein encoding genes in broilers. PLoS One 12:e0182079. 489 490 491 492 493 494 495 496 497 498 499 500 501 21 502 503 504 505 Table 1. Nutritional parameters in dams and offspring at end of lactation 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 The results are expressed as mean ± SEM and analysed by a multifactorial analysis of variance 526 (one-way ANOVA) followed by the Tukey´s test. The number of animals in each group is 8. 527 C SS SD DAMS Solid intake (g/day) 31.24 ± 3.2 47.8 ± 4.0 cc 52.93 ± 3.4 ccc Se intake (μg/day) 3.12 ± 0.32 23.9 ± 2.0 ccc 0.53 ± 0.03 c, sss Weight gain (g) 42.5 ± 1.06 53.6 ± 3.49 cc 36.48±1.85 c, sss Se in milk (μg/ml) 0.124 ± 0.005 0.126 ± 0.003 0.102 ± 0.003 cc, sss OFFSPRING Milk intake (g/30 min sucklig) 0.39 ± 0.03 0.61 ± 0.02 cc 0.37 ± 0.03 ss Milk intake/body weight 1.2 ± 0.09 1.7 ± 0.12 cc 1.8 ± 0.10 cc Weight gain (g) 26.3± 0.9 28.2±0.9 c 16±0.9 ccc, sss Cranium-caudal length (cm) 10.95± 0.122 10.86 ± 0.204 8.95±0.204 ccc, sss Body Mass Index (BMI) (kg/m2) 2.66 ± 0.047 2.9 ± 0.05 cc 2.6 ± 0.03 sss Se in serum (ng/mL) 117 ± 4.1 217 ± 3.7 ccc 109± 5.1 sss Se in liver (µg/g dry weight) 0.38 ± 0.03 0.46 ± 0.02 c 0.05 ± 0.003 ccc, sss HSI (g/g body weight (%)) 3.3 ± 0.1 3.94 ± 0.05 ccc 3.4 ± 0.05 sss Se in pancreas (µg/g dry weight) 0.225 ± 0.01 0.23 ± 0.01 No detected PSI (g/g body weight (%)) 0.41 ± 0.02 0.41 ± 0.03 0.34 ± 0.02 c, s 22 Groups: C: control group, SS: selenium supplemented group, and SD: selenium deficient group. 528 Statistic difference between groups was expressed as p value; and was indicated as c (p<0.05), 529 cc (p<0.01) and ccc (p<0.001) to control group, and s (p<0.05), ss (p<0.01) and sss (p<0.001) to 530 Se supplemented group. 531 532 533 Table 2. Hepatic oxidative balance in offspring. 534 535 The results are expressed as mean ± SEM and analysed by a multifactorial analysis of variance 536 (one-way ANOVA) followed by the Tukey´s test. The number of animals in each group is 8. 537 Groups: C: control group, SS: selenium supplemented group, and SD: selenium deficient group. 538 Statistic difference between groups was expressed as p value; and was indicated as c (p<0.05), 539 cc (p<0.01) and ccc (p<0.001) to control group, and s (p<0.05), ss (p<0.01) and sss (p<0.001) to 540 Se supplemented group. 541 542 543 544 545 546 547 C SS SD OFFSPRING SOD (U/mg protein) 1.96 ± 0.16 2.75 ± 0.20 c 4.08 ± 0.16 ccc, sss CAT (U/mg protein) 187.1 ± 6.6 251.3 ± 8.9 ccc 125.4± 4.0 ccc, sss GPx (mU/mg protein) 114.9 ± 4.5 144.2 ± 6.1 ccc 45.5 ± 2.7 ccc, sss MDA (mol/mg protein) 0.42 ± 0.008 0.38 ± 0.017 0.32 ± 0.015 ccc (CAT+GPx1)/SOD 117.1 ± 6.4 143.3 ± 8.4 41.9 ± 2.9 ccc, sss CG (nmol/mg protein) 4.19 ± 0.21 4.19 ± 0.22 5.39 ± 0.27 cc, ss 23 548 549 550 551 552 FIGURE LEGENDS. 553 Figure 1. Metabolic parameters in offspring. The results are expressed as mean ± SEM and 554 analysed by a multifactorial analysis of variance (one-way ANOVA) followed by the Tukey´s 555 test. The number of animals in each group is 8. Groups: C: control group, SS: selenium 556 supplemented group, and SD: selenium deficient group. Statistic difference between groups 557 was expressed as p value; and was indicated as c (p<0.05) and ccc (p<0.001) to control group, 558 and s (p<0.05) and s (p<0.05), ss (p<0.01) and sss (p<0.001) to Se supplemented group. 559 Figure 2. Expression of GPx1 (A) and SelP (B) in liver of offspring. Representative western 560 blots of proteins (normalized to β-actin) (C). The results are expressed as mean ± SEM and 561 analysed by a multifactorial analysis of variance (one-way ANOVA) followed by the Tukey´s 562 test. The number of animals in each group is 8. Groups: C: control group, SS: selenium 563 supplemented group, and SD: selenium deficient group. Statistic difference between groups 564 was expressed as p value; and was indicated as c (p<0.05) and ccc (p<0.001) to control group, 565 and s (p<0.05) and sss (p<0.001) to Se supplemented group. 566 Figure 3. Expression of AMPK (A), p-AMPK (B) and its ratio (C) in liver of offspring. 567 Representative western blots of proteins (normalized to β-actin) (D). The results are 568 expressed as mean ± SEM and analysed by a multifactorial analysis of variance (one-way 569 ANOVA) followed by the Tukey´s test. The number of animals in each group is 8. Groups: C: 570 control group, SS: selenium supplemented group, and SD: selenium deficient group. Statistic 571 difference between groups was expressed as p value; and was indicated as c (p<0.05) and ccc 572 (p<0.001) to control group, and sss (p<0.001) to Se supplemented group. 573 Figure 4. Expression of IRS-1 (A) in liver of offspring. Representative western blots of 574 proteins (normalized to β-actin) (B). Serum insulin levels (C) and HOMA-IR Index (D). The 575 results are expressed as mean ± SEM and analysed by a multifactorial analysis of variance 576 (one-way ANOVA) followed by the Tukey´s test. The number of animals in each group is 8. 577 Groups: C: control group, SS: selenium supplemented group, and SD: selenium deficient group. 578 24 Statistic difference between groups was expressed as p value; and was indicated as c (p<0.05), 579 cc (p<0.01) and ccc (p<0.001) to control group, and sss (p<0.001) to Se supplemented group. 580 581