Metabolic signaling under nutrient deprivation

Abstract

1- ROLE OF SIRT1 IN THE REGULATION OF FATTY ACID OXIDATION AND KETOGENESIS UNDER DIFFERENT NUTRIONAL CHANGES.

The homolog of the yeast silencing information regulator2 (SIRT1) has been implicated in several aspects of food limitation and caloric restriction in mammals. We have observed that there were no important changes, between wild type (WT) and SIRT1 liver-specific knockout (LKO) mice subjected to either CR or high fat diet (HFD), in the mRNA expression of Cpt1a, Cpt2 and Hmgcs2.

SIRT1 had been shown to control hepatic glyconeogenic/glycolytic pathways in response to nutrients (Rodgers et al., 2005). So, we have hypothesized that SIRT1 could have a role in the metabolic adaptation to the changes of nutrient of weaning, when milk is replaced by the adult diet which contains less fat and more carbohydrate. Neither fatty acid oxidation (Cpt1a), ketogenesis (Hmgcs2), nor gluconeogenic (Pck1) liver pathways were significantly affected by the liver-specific knockdown of SIRT1, in both suckling and post-weaning conditions.

If SIRT1 is involved in the response to aging, old LKO mice might be more susceptible to age-associated diseases as obesity, type 2 diabetes, hypertension, etc. To test this hypothesis we first weight and performed a glucose tolerance test in 7 months-old mice. There were no differences in weight neither in glucose tolerance between WT and LKO mice.

Using a cell system, PPARα induced the expression of its known target genes CPT1A, HMGCS2, and FGF21 in HepG2 cells. SIRT1 overexpression by itself had almost no effect, although it increased PPARα induction of its target genes. We have interfered SIRT1 in these cells, and PPARα-induced expression of PEPCK and FGF21 was SIRT1-dependent.

We have fasted WT and LKO mice for 15h and we found that the expression of Pck1 in liver was moderately but significantly induced in SIRT1-LKO mice after fasting, consistent with an increase in glucose levels. However, neither G6pase nor Pgc1α mRNA levels were affected. As expected, liver mRNA levels of Cpt1a, Hmgcs2, and Fgf21 were also induced upon fasting. However, Cpt1a and Hmgcs2 mRNA transcripts were comparable in fasted WT versus LKO mice, while Fgf21 expression was reduced around 40% in LKO mice liver. This result was consistent with the fact that fasting induction of FGF21 serum levels was also impaired in LKO mice.

2- ROLE OF SIRT1 IN THE HMGCS2 REGULATION OF FGF21 EXPRESSION. (Article 1: Human HMGCS2 regulates fatty acid oxidation and FGF21 expression in HepG2 cells. Vilà-Brau et al., 2011)

Recently, our group has seen that HMGCS2 expression stimulates FGF21 expression and that these events are dependent on HMGCS2 activity. A catalytic dead mutant (C166A) failed to induce either fatty acid β-oxidation or FGF21 expression, whereas acetoacetate (an oxidized form of ketone bodies) could stimulate FGF21 mRNA expression in a dose-dependent manner.

Because ketone bodies production implies the reduction of acetoacetate to β- hydroxybutyrate with the concomitant generation of NAD+ (Hegardt et al, 1999), and SIRT1 is a NAD+-dependent deacetylase enzyme, this specificity could explain why FGF21 fasting induction was affected in LKO-SIRT1 mice liver, while other PPARα target genes were not. We have treated HepG2 cells with the oxidizing (acetoacetate) partner of ketone bodies, and endogenous SIRT1 was knockdown by a specific siRNA. FGF21 induction was dependent on SIRT1 expression, since knocking down impaired acetoacetate response.

3- ACTIVATING TRANSCRIPTION FACTOR 4-DEPENDENT INDUCTION OF FGF21 DURING AMINO ACID DEPRIVTION (Article 2: De Sousa-Coelho et al., 2012)

Considering the central role of PPARα in the regulation of metabolic homeostasis we sought to investigate how the turnover of PPARα affected the expression of its target genes. HepG2 cells were infected with PPARα and exposed to DMSO or to the 26S proteasome inhibitor MG132. As expected, MG132-treatment blocked the PPARα-dependent expression of HMGCS2, indicating that the transcriptional activity of PPARα is increased by protein degradation (Blanquart et al, 2004). Contrary to what we had predicted, the expression of FGF21 was strongly increased by the MG132 treatment.

We hypothesized that proteasome inhibition in HepG2 could decrease the pool of free amino acids. We treated HepG2 cells with histidinol (HisOH) a potent and reversible inhibitor of protein synthesis, (Hansen et al, 1972). Amino acid deprivation produced a time-dependent induction of FGF21 mRNA. To test whether this induction was due to an increase in the FGF21 gene transcription, we measured the FGF21 primary transcript (hnRNA) levels; HisOH treatment clearly induced FGF21 hnRNA levels in a time-dependent manner.

As expected, HisOH induced an increase in the ATF4 protein levels after 2h treatment. By analyzing the sequence of the 5’-flanking region of the human FGF21 gene, we found two putative ATF4 response elements (AARE) starting at positions -152 and -610 upstream of the transcription start site. HepG2 cells were transfected with pGL3b-hFGF21 promoter-luciferase constructs and an expression vector for human ATF4. The expression of ATF4 induced the WT reporter in a concentration dependent manner. This induction was totally obliterated either when the AARE1 was mutated or when both elements were deleted. Induction was diminished when AARE2 was mutated.

To further analyze the functionality of this sequence we tested the binding of ATF4 by an EMSA, where ATF4 bound as a C/EBPβ heterodimer to both AARE sequence elements. We also confirmed the in vivo binding by ChIP experiments. The chromatin binding of ATF4 was greatly increased in both ATF4 responsive sequences in HisOH treated cells.

To confirm if the induction of FGF21 produced by amino acid starvation was mediated by ATF4, we treated siCtl and siATF4 HepG2 cells with HisOH. FGF21 mRNA levels after HisOH treatment were significantly lower when ATF4 was depleted.

To analyze the effect of amino acid deprivation on FGF21 expression in vivo, we fed mice with a leucine-deficient [(-)leu] diet or a control (Ctl, nutrionally complete) diet for 7 days. Fgf21 mRNA levels were greatly increased in liver from mice fed a (-)leu diet compared to control. The circulating FGF21 levels were also increased in the serum of leucine deprived animals, paralleling hepatic gene expression.

4- LEUCINE DEPRIVATION SIGNALING UNDER FASTING CONDITIONS.

We were interested to know whether FGF21 induction by a (-)leu diet would affect, or be affected by, the adaptive fasting response. We have fed mice for 7 days within a Ctl or (-)leu diet. Weights and food intake were recorded daily. Then, mice were randomly separated in a total of 4 groups, where 2 groups (one from each diet) were fasted overnight.

Leucine deprivation affected the levels of free fatty acids and ketone bodies in serum in the fed state, while it does not upon fasting. Although no changes between groups were observed in the fed state, after fasting the β-oxidation, ketogenesis and gluconeogenesis keygenes were further up-regulated in the (-)leu diet group compared to control. The highest induction was seen in the Pgc1α gene, a known coactivator on these processes. The fasting activation of FGF21 was impaired in mice fed with (-)leu diet, underlying a crosstalk between the fasting and amino acid deprivation signalling.

5- ROLE OF FGF21 IN THE LEUCINE DEPRIVATION PHENOTYPE IN MICE (Article 3: De Sousa-Coelho et al., in preparation).

According with our previously reported results (De Sousa-Coelho et al., 2012) mice maintained on a leucine-deficient [(-)leu] diet show a dramatic increase in FGF21 circulating levels. To check its origin we analyzed the Fgf21 gene expression in liver, where Fgf21 mRNA levels paralleled those in serum; brown adipose tissue (BAT), where it were unchanged; and in epididymal white adipose tissue (eWAT), where unexpectedly it were significantly decreased in wild type mice maintained in (-)leu diet.

Because upon (-)leu diet, mice undergo rapid weight loss (Cheng et al., 2010), we wanted to investigate whether this phenotype is FGF21-dependent. For this purpose, WT and FGF21-KO mice were fed a Ctl or (-)leu diet for 7 days. Weight loss was diminished in FGF21-KO, while food intake decrease by (-)leu was unchanged between genotypes. Histological analysis of WAT showed that leucine deprivation resulted in a reduction in adipocyte volume compared with mice fed a control diet, while it was only slightly reduced in (-)leu FGF21-KO mice. It has been previously described that leucine deprivation increases lipolysis in WAT (Cheng et al., 2010). Consistent with changes in body weight, lack of FGF21 significantly decreased levels of phosphorylated (P)-HSL in WAT, indicating an impaired lipolysis. Gene expression analysis revealed reduction in the mRNA levels of the lipogenic genes Fas, Srebp1c and Acc1 in the WAT of mice maintained in (-)leu diet. These changes were impaired in FGF21-KO.

Consistent with previous results (Cheng et al., 2010), leucine deprivation increased levels of Ucp1 mRNA in WT mice BAT. This increase was not observed in the FGF21-KO mice. mRNA levels of Pgc1α, which regulates the expression of Ucp1 (Handschin and Spiegelman, 2006), were also increased, although did not differ between WT and FGF21-KO mice under either control or (-)leu diet conditions.

It has been recently proposed a link between FGF21 and SREBP1c during lipogenesis in HepG2 cells (Zhang et al., 2011). We examined levels of Fas, Srebp1c and Acc1 mRNA in liver of WT and FGF21-KO. As expected (Guo and Cavener, 2007), lipogenic program was decreased upon (-)leu diet, but this reduction was blocked in FGF21-KO mice. However, the amino acid response program was correctly initiated in these mice as shown by the increased levels of ATF4 protein and the increase in mRNA levels of Asns, a prototypical ATF4 target gene. The liver staining showed a decreased lipid accumulation under (-)leu in WT animals that is not produced in the FGF21-KO mice.

These results demonstrate an important role of FGF21 in the regulation of lipid metabolism during amino acid starvation. References:

Blanquart C, Mansouri R, Fruchart JC, Staels B, & Glineur C (2004) Different ways to regulate the PPARalpha stability. Biochem Biophys Res Commun 319, 663-70.

Cheng, Y., Meng, Q., Wang, C., Li, H., Huang, Z., Chen, S., Xiao, F., and Guo, F. (2010). Leucine deprivation decreases fat mass by stimulation of lipolysis in white adipose tissue and upregulation of uncoupling protein 1 (UCP1) in brown adipose tissue. Diabetes 59, 17-25.

Guo, F., and Cavener, D.R. (2007). The GCN2 eIF2alpha kinase regulates fatty-acid homeostasis in the liver during deprivation of an essential amino acid. Cell Metab 5, 103-14.

Handschin, C., and Spiegelman, B.M. (2006). Peroxisome proliferator-activated receptor gamma coactivator 1 coactivators, energy homeostasis, and metabolism. Endocr Rev 27, 728-735.

Hansen BS, Vaughan MH, & Wang L (1972) Reversible inhibition by histidinol of protein synthesis in human cells at the activation of histidine. J Biol Chem 247, 3854-3857.

Hegardt FG (1999) Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase: a control enzyme in ketogenesis. Biochem J. 338, 569-582.

Hotta, Y., Nakamura, H., Konishi, M., Murata, Y., Takagi, H., Matsumura, S., Inoue, K., Fushiki, T., and Itoh, N. (2009). Fibroblast growth factor 21 regulates lipolysis in white adipose tissue but is not required for ketogenesis and triglyceride clearance in liver. Endocrinology 150, 4625-4633.

Rodgers JT, Lerin C, Haas W, Gygi SP, Spiegelman BM, Puigserver P (2005) Nutrient control of glucose homeostasis through a complex of PGC-1alpha and SIRT1. Nature 434, 113-8

Zhang, Y., Lei, T., Huang, J.F., Wang, S.B., Zhou, L.L., Yang, Z.Q., and Chen, X.D. (2011). The link between fibroblast growth factor 21 and sterol regulatory element binding protein 1c during lipogenesis in hepatocytes. Mol Cell Endocrinol 342, 41-47.

Durante su vida un individuo se somete a diversos cambios nutricionales. La capacidad de detectar la disponibilidad de nutrientes y regular la homeostasis energética es un proceso fundamental. SIRT1 es un regulador clave en el metabolismo energético. SIRT1 puede modular la expresión génica en tejidos metabólicamente activos en respuesta a la restricción calórica o el ayuno. La dependencia de la actividad deacetilasa de SIRT1 en los niveles de NAD+ constituye un vínculo fundamental entre el estado metabólico celular y la regulación de genes. FGF21 es una hormona que se induce en el ayuno y que afecta al metabolismo de los carbohidratos y de los lípidos. Recientemente se han demostrado sus efectos benéficos en la protección de la obesidad inducida por la dieta y en la mejoría en la resistencia a la insulina.

En este trabajo hemos demostrado que en células hepáticas en cultivo, SIRT1 desempeña un papel importante en la activación por PPARα de la expresión de FGF21, CPT1A, HMGCS2, y PEPCK. También, que la actividad de SIRT1 regula los niveles de glicemia y la expresión de PCK1 en hígado, en la respuesta al ayuno. Aún así, hemos visto que la actividad de SIRT1 no afecta la expresión de los genes de la oxidación de los ácidos grasos o la cetogénesis en el hígado, en respuesta a diferentes cambios nutricionales, como la restricción calórica, la transición de la lactancia/destete, y el ayuno. Adicionalmente, hemos demostrado que la activación de FGF21 por SIRT1 depende de la actividad HMGCS2.

También hemos descrito que FGF21 se induce por la privación de aminoácidos, de manera dependiente de ATF4, y hemos identificado dos elementos de respuesta funcionales en la región promotora del gene humano, altamente conservados entre las especies.

Además, hemos demostrado que FGF21 interviene en la regulación del metabolismo de los lípidos en el hígado y el tejido adiposo blanco, y de la termogénesis en el tejido adiposo marrón, durante la privación de aminoácidos. De todas formas, hemos visto que la privación de leucina afecta a los niveles de ácidos grasos libres y cuerpos cetónicos en suero en el estado de alimentación, mientras que no lo hace en el ayuno; y la activación de FGF21 en el ayuno está afectada en los ratones alimentados con esta dieta, desvelando un “crosstalk” entre la señalización del ayuno y la privación de aminoácidos.