Differential regulation of sterol regulatory element-binding protein 1c transcriptional activity by insulin and liver X receptor during liver development.

Sterol regulatory element-binding proteins (SREBPs) are transcription factors involved in the synthesis of cholesterol and fatty acids. In adults, the isoform SREBP-1c is the predominant transcript in the liver of fed animals, and it activates triglyceride production from glucose when diet is enriched in carbohydrates. Studies have shown that SREBP-1c expression is dependent on insulin but also on the availability of oxysterols, ligands of the nuclear liver X receptor (LXR). The aim of this study was to investigate the regulation of the hepatic SREBP-1c expression in vivo in situations where drastic nutritional and hormonal changes occur, from the gestation to the weaning period. In this paper, we report the discovery of LXR-independent SREBP-1c transcriptional activity during late gestation. In utero insulin injection prior to the natural rise in insulin in late gestation triggers SREBP-1c mRNA elevation, nuclear SREBP-1c binding activity, and expression of its target genes independently of LXR transactivation. On the other hand, during suckling, we observed strong SREBP-1c mRNA expression despite very low plasma insulin, an expression that may be due to LXR transactivation. In contrast to insulin, LXR is not sufficient to trigger nuclear SREBP-1c binding activity and target gene induction. This could be due to the concomitant induction of INSIG-2a by LXR and subsequent retention of SREBP-1c in the endoplasmic reticulum.

During development, the organism must continuously adapt its metabolism to the nutritional environment because the availability and quality of the nutrients varies widely throughout the different developmental stages. In rodents, as for most mammals, the fetus receives through the placenta a diet enriched in carbohydrates but very poor in fat (1). Immediately after birth, the maternal supply of substrates ceases abruptly, and the newborn is fed at intervals with milk, a high fat, low carbohydrate diet (2,3). Toward the end of the suckling period the milk is progressively replaced by the solid food diet of the adult, enriched in carbohydrates. The adaptation to these changes in nutrition requires important modifications of glucose and fatty acid metabolism in the liver orchestrated mainly through hormonal variations.
In the adult, the transcription factor sterol regulatory element-binding protein 1c (SREBP-1c) 1 has been shown to play a key role for the control of glucose and fatty acid metabolism in the liver (4,5). The SREBP-1c transcription factor belongs to the basic helix-loop-helix leucine zipper family. It is synthesized as a precursor form anchored in the endoplasmic reticulum (ER). In the ER, SREBPs form a complex with the SREBP cleavage-activating protein. SREBP cleavage-activating protein escorts the SREBPs to the Golgi apparatus where they are processed by two proteases. After proteolytic cleavage, the mature transcriptionally active form of SREBP migrates into the nucleus where it can bind to the sterol regulatory element (SRE) or E boxes of target genes (6 -8). INSIG-1 and INSIG-2 proteins have been identified as proteins that promote SREBP retention in the ER through their interaction with SREBP cleavage-activating protein, thus preventing SREBP translocation to the Golgi apparatus for proteolytic processing (9). It has recently been shown that a liver-specific isoform derived from the INSIG-2 gene called INSIG-2a is selectively down-regulated by insulin (10).
SREBP-1c expression is itself under the control of the nutritional and hormonal status of the animal. Indeed, previous studies have shown that SREBP-1c expression is transcriptionally stimulated by insulin (11,12). In vivo, a reduction of liver SREBP-1c mRNA is observed during fasting, whereas in mice refed a high carbohydrate diet, a strong increase of the SREBP-1c transcript occurs (11). SREBP-1c has been shown to be a major mediator of insulin action on the expression of glucokinase and lipogenesis-related genes in cultured rat hepatocytes (13) and in vivo (14). In the adult liver, insulin induces SREBP-1c transcription, but it has also been suggested that insulin enhances nuclear abundance of SREBP-1c (15,16).
In addition to insulin, hepatic SREBP-1c is also under control of the oxysterols, which are ligands of the nuclear liver X receptor (LXR) (17). LXR directly activates SREBP-1c transcription through two LXRE-binding sites present in the SREBP-1c promoter. ␣ and ␤ isoforms of LXR form heterodimers with the ubiquitous dimerizing partner, RXR (18). Apart from activation of lipogenesis through SREBP-1c, the LXR transcription factors are crucial because they regulate multiple genes involved in cholesterol metabolism and transport, including cholesterol 7␣ hydroxylase, the rate-limiting enzyme in bile acids production (19), and genes such as the ATP-binding cassette transporter A1 (ABC-A1) (20), which is involved in hepatic cholesterol efflux.
The current experiments were designed to explore the in vivo regulation of SREBP-1c levels in the liver during late gestation, suckling, and the weaning period and the respective roles of LXR and insulin on SREBP-1c transcriptional activity throughout these developmental stages.

EXPERIMENTAL PROCEDURES
Animals-C3H mice (Charles River) were used in this study. The animals were housed in a controlled environment (constant temperature and humidity, light from 7 a.m. to 7 p.m.) and fed ad libitum with a standard chow diet: composition (calories) 65% carbohydrates, 11% lipids, and 24% proteins (UAR A 03).
Adult Mice-Adult mice were fasted for 16 h and then refed with standard chow and water containing 20% sucrose for 6 h. After cervical dislocation, the liver was collected for protein preparations.
For LXR agonist gavage experiments, 10-week-old fed male mice were force fed a solution containing the LXR synthetic agonist T0-901317 (Sigma) at a dose of 50 mg/kg in 1% carboxymethylcellulose (Sigma) or with the vehicle for the control mice. After 12 h, the mice were sacrificed, and the livers were collected for RNA isolation Fetuses-Following mating, noon of the day of the appearance of a vaginal plug was taken to be 0.5 day of gestation. Fetuses at stages 16.5, 17.5, and 19.5 days post coitum (pc) were removed from uteri, transferred to dishes containing phosphate-buffered saline, sacrificed, and dissected. The liver was collected for RNA and nuclear protein extraction. For insulin injection (Actrapid®; NovoNordisk), the abdominal wall of pregnant mice was incised under anesthesia, and subcutaneous injections were made through the uterine wall into the fetus with an insulin syringe (U 40 Insulin; Terumo). The incisions in the mothers were then sutured. Four hours later, the mice were sacrificed by cervical dislocation, and the livers of the fetuses were removed for RNA isolation. Plasma was stored at Ϫ80°C.
Pups-After mating C3H mice, noon of the day of birth was considered as time 0. The mice were sacrificed at days 1, 8, 13, 15, 21, and 28 after birth. The mice were weaned at day 20 onto standard chow.
Real Time PCR-RNA were isolated from fresh tissue samples according to Chirgwin et al. (21). Specific primers for each gene (Table I) were designed using primer express software (PerkinElmer Life Science). Real time PCRs were performed starting with 50 ng of reverse transcribed total RNA subsequently mixed with fast start DNA master Sybr Green mix (Roche Applied Science), 3 mM MgCl 2 , and various sets of 525 nM gene-specific forward and reverse primers (Table I) in a final volume of 10 l. The relative amount of each mRNA was calculated using the second derivative comparative C t method. Quantification was obtained after normalization to 18 S rRNA and correction for interplate variation using a reverse transcription calibrator (22).
Nuclear Extracts Preparation-The livers were removed and quickly rinsed in ice-cold phosphate-buffered saline with the protease inhibitors pepstatin (5 g/ml), leupeptin (5 g/ml), and aprotinin (2 g/ml). The livers were then transferred into a beaker, on ice, containing the homogenization buffer (2 M sucrose, 10 mM Hepes, pH 7.6, 25 mM KCl, 0.15 mM spermine, 0.5 mM spermidine, 1 mM EDTA, 10% glycerol, 0.5 mM dithiothreitol) and minced with scissors. After processing in a Dounce apparatus in fresh buffer on ice, the homogenates were layered on 2 M sucrose cushions and centrifuged at 80,000 ϫ g for 35 min at 0°C. The nuclear pellets were then rinsed in a buffer containing 10 mM Tris, pH 7.4, 10 mM NaCl, 3 mM MgCl 2 ; resuspended in a 20 mM Hepes, pH 7.9, 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol buffer; and incubated 15 min on ice. After spinning 5 min at 10,000 rpm, aliquots of the supernatants were stored at Ϫ80°C. The protein content was determined as described by Bradford, using bovine serum albumin as a standard.
Insulin Immunoassay-The blood samples were centrifuged at 1500 ϫ g for 15 min at 4°C, and supernatants were used for insulin immunoassay (Insulin CT; Cis Bio International).
Statistical Analyses-The results are expressed as the means Ϯ S.E. Statistical significance was assessed using a Mann-Whitney nonparametric test.

RESULTS
Liver SREBP-1c Binding Activity Increases with Plasma Insulin during Fetal Development-To investigate the mechanisms involved in the regulation of SREBP-1c expression during development, we first measured hepatic SREBP-1c mRNA in fetal livers from 16.5 days pc to day 1 after birth. Fig. 1A shows that SREBP-1c is expressed in the fetal liver at 16.5 days pc. Earlier stages show very low amounts of SREBP-1c mRNA (results not shown). SREBP-1c mRNA levels increase from 16.5 days pc to 19.5 days pc and then fall dramatically at birth. The level of SREBP-1c mRNA expression at day 1 after birth is the lowest detected throughout our experiments including in the fasted mice (data not shown). These variations in SREBP-1c mRNA expression closely paralleled insulin plasma levels ( Fig. 1A), before and after birth. Two SREBP-1 isoforms are derived from a single gene by the use of two distinct promoters and yield two differently regulated isoforms, SREBP-1a and SREBP-1c. In our experiments, SREBP-1a mRNA levels were low and did not change significantly between 16.5 days pc and 19.5 days pc (Fig. 1A).
We then assessed the potential SREBP-1c binding activity in the liver during development by EMSA using the SRE-binding site of the fatty acid synthase (FAS) gene (Table II). As shown in Fig. 1B, SREBP-1 binding activity was detected from 17.5 days pc. As for SREBP-1c mRNA levels, there is a gradual increase in nuclear binding of SREBP-1 to the SRE probe from 16.5 to 19.5 days pc followed by a dramatic fall at birth. The specificity of the binding activity is confirmed by SREBP-1 antibody competition of the 19.5 days pc binding activity (Fig. GGAGCCATGGATTGCACATT GCTTCCAGAGAGGAGGCCAG Differential Regulation of SREBP-1c Transcriptional Activity 1B). The lowest band (Fig. 1B, band NS) is not displaced by the specific antibody and thus corresponds to a nonspecific binding activity. These experiments provide evidence that SREBP-1c mRNA and nuclear binding to the SRE increase progressively from mid-gestation to late gestation and drop just after birth in parallel with plasma insulin.
In the adult liver, SREBP-1c is a major transcriptional mediator of insulin action on glycolytic and lipogenic enzyme genes such as the glucokinase (GK) and FAS genes (12,13,16). A similar pattern as described for SREBP-1c mRNA and binding activity was observed for GK and FAS mRNA expression, characterized by an increase between 16.5 days pc and 19.5 days pc and a drastic fall at birth (Fig. 1C). These results suggest that GK and FAS are SREBP-1c target genes during late gestation and that SREBP-1c expression is itself related to plasma insulin concentrations.
SREBP-1c Expression Occurs in the Absence of LXRs Binding to LXRE during the Fetal Development in the Liver-In addition to insulin, the nuclear receptor LXR is also able to control SREBP-1c gene expression in the adult liver (24 -26). Two isoforms of LXR are known. LXR␣ is primarily expressed in the liver, whereas LXR␤ expression is ubiquitous with low levels in the adult liver (27). Expression of LXR␣ remains low from day 16.5 to day 19.5 pc when compared with the values observed after birth ( Fig. 2A). LXR␤ expression increases between day 16.5 and 19.5 pc and shows minor variations thereafter.
The potential binding activity of the LXR nuclear receptors was then assessed in the fetal liver by EMSA, using one functional LXRE-binding site of the SREBP-1c promoter (23) (Table  II). Despite LXR␣ and ␤ mRNA expression, there was no detectable LXR␣ or ␤ binding activity between 16.5 and 19.5 days pc, whereas it was clearly present in the 21-day-old weaned mouse. Thus, SREBP-1c gene expression during fetal life is independent of both LXR␣ and ␤ binding on its promoter at these stages.

In Utero Insulin Injections Can Trigger a Premature Increase of the SREBP-1c mRNAs and Expression of SREBP-1c Target
Genes in the Fetus-To determine the involvement of insulin in the transcriptional regulation of SREBP-1c during fetal life, insulin was injected in utero at 16.5 days pc, a developmental stage characterized by low plasma insulin, low SREBP-1c mRNA level, and the absence of LXRs binding on the LXRE of the SREBP-1c gene. Within 4 h, insulin administration to the 16.5-day-old fetuses induces an increase of SREBP-1c mRNA (Fig. 3A). As shown by EMSA (Fig. 3B), a precocious strong binding activity of SREBP-1 is also triggered by the insulin injection. In contrast, no LXR binding activity was detected upon insulin injection (Fig. 3C). Moreover, no effect on LXR mRNA levels was observed (data not shown). Therefore, insulin is able to trigger a premature increase of SREBP-1c mRNA in the 16.5 day pc fetal liver in the absence of any LXR binding activity (Fig. 3C). This transcriptional activation of the SREBP-1c gene by insulin is followed by an increase in expression of SREBP-1c target genes, FAS and GK (Fig. 3D).
LXR Binding to SREBP-1c LXRE Could Explain High SREBP-1c mRNA Levels during the Suckling Period-Directly after birth, during the suckling period, pups ingest milk, which is a high fat, low carbohydrate diet, inducing low plasma insucific binding. This EMSA is representative of three independent experiments. C, GK and FAS transcript levels were quantified by real time quantitative PCR and expressed in fold increase over the value observed at day 1 after birth (d1). Each value represents the mean Ϯ S.E. of six determinations. *, significant difference versus 16.5 days pc of p Ͻ 0.05; **, significant difference versus 16.5 days pc of p Ͻ 0.01. FIG. 1. SREBP-1c expression levels and binding activity in the fetal liver. Liver RNA and nuclear proteins were extracted from fetuses at stages 16.5, 17.5, and 19.5 days pc and from newborn pup liver at day 1 (d1). A, SREBP-1c and SREBP-1a transcript levels were quantified by real time quantitative PCR. Each value represents the mean Ϯ S.E. of six determinations. **, a significant difference versus 16.5 days pc (p Ͻ 0.01). Plasma insulin (•) from each fetus was quantified by radioimmunological assay. § § indicates a significant difference versus 16.5 days pc (p Ͻ 0.01). B, EMSA with nuclear extracts from fetal livers at different stages of development and from newborn pup liver (d1). 4 g of proteins were incubated in presence of 0.1 ng of the radiolabeled SRE-binding site in absence (Ϫ) or in the presence (ϩ) of a monoclonal antibody to SREBP-1 (SREBP-1 Ab). The position of the SREBP-1-specific complex is shown by an arrow. NS indicates nonspe-lin and high plasma glucagon levels. At weaning, the transition from a milk diet to a high carbohydrate, low fat diet (animal chow) is responsible for a rise in plasma insulin and a decrease in plasma glucagon (28,29).
Following our observation of the insulin-mediated regulation of SREBP-1c in the fetal liver, we then went on to examine the hepatic expression of SREBP-1c mRNA at days 1, 8, and 13 after birth (suckling period) and after weaning. Surprisingly, hepatic SREBP-1c mRNA levels were high during the suckling period, reaching a level of expression even higher than levels observed in weaned mice (Fig. 4A) or high carbohydrate-refed adult mice (data not shown). The low plasma insulin concentrations during suckling (under 20 microunits/ml) cannot account for this peak of SREBP-1c expression; thus another factor must be responsible for this induction.
Given that LXR is known to increase SREBP-1c transcription in the adult, we then hypothesized that this transcription factor could be responsible for the elevated hepatic SREBP-1c mRNA level during suckling. LXR␣ mRNA expression during suckling is increased 2-fold between days 1 and 8 after birth and peaks at day 13. LXR␣ expression decreases somewhat after weaning (Fig. 5A). At the same time, no statistically significant variation of LXR␤ mRNA expression is observed (Fig. 5A). To assess LXR binding activity during suckling, we performed EMSA with nuclear extracts from days 1 to 15 after birth and from weaned mice (Fig. 5B). Contrary to what we observed during gestation, a strong LXR binding activity is detectable throughout the suckling period. Moreover, LXR target genes such as lipoprotein lipase (LPL) (30) and ABC-A1 (20, 31) have significantly elevated mRNA levels during suckling in comparison with weaned values and show the same expression pattern as SREBP-1c (Fig. 4B). The stronger decrease in the expression of SREBP-1c, LPL, and ABC-A1 mRNA in comparison with the LXR mRNA expression levels at weaning is more likely a reflection of the availability of the LXR agonists decreasing when the animals are weaned from a high fat diet (maternal milk) onto a high carbohydrate diet (standard chow). Thus, during the suckling period and in contrast with the late gestation, LXR appears the more likely activator of SREBP-1c expression.
Despite High SREBP-1c mRNA Expression, Glucokinase and Fatty Acid Synthase Gene Expression Remains Very Low throughout the Suckling Period-To determine whether the high level of SREBP-1c mRNA during the suckling period is followed by the transcriptional activation of its potential target genes, we measured GK and FAS expression at days 1, 8, and 13 after birth and at weaning by real time quantitative PCR (Fig. 6). As previously described (32-34), GK and FAS mRNA Recently, it has been shown by Botolin and Jump (35) that the SREBP-1c precursor protein was present in the ER of suckling rats, indicating that SREBP-1c mRNA is well translated at that time but absent from the nuclei. We have also detected high levels of expression of SREBP-1 precursor protein in the suckling mice (data not shown). In EMSA analysis of SREBP-1c binding activity of nuclear extracts from mouse livers, no SREBP-1 binding is detected during the suckling period, whereas a shift is evident using extracts from weaned mice (Fig. 7). This indicates an absence of nuclear binding of SREBP-1c on SRE probes during suckling. These results reveal that the absence of hepatic SREBP-1c target gene induction during suckling, despite elevated SREBP-1c mRNA and ER precursor protein, is due to the absence of nuclear SREBP-1 binding activity. Thus, these studies indirectly suggest that a specialized mechanism should promote SREBP-1c retention in the ER preventing nuclear translocation. we hypothesized that INSIG-2a expression could be high during suckling and may interfere with SREBP-1c cleavage. We first measured hepatic INSIG-2a mRNA levels from 16.5 days pc to weaned mice (Fig. 8A). INSIG-2a expression is barely detectable during gestation in the liver. INSIG-2a expression increases gradually throughout suckling until day 15, when expression is maximal, and falls at the suckling/weaning transition. Thus, elevated INSIG-2a levels during suckling could be at least partly responsible for the retention of SREBP-1c into the ER compartment. It is interesting to note that INSIG-1 mRNA displays a totally opposite expression profile (Fig. 8A); high during gestation, INSIG-1 mRNA levels drop at birth and remain low throughout suckling until weaning when expression increases again.

INSIG-2a but Not INSIG-1 Might Be Implicated in the Lack
LXR Synthetic Agonist T0-901317 Induces INSIG-2a mRNA-Because LXR binding activity and INSIG-2a mRNA are both elevated during the suckling period, we tested the hypothesis that LXR might regulate the INSIG-2a gene expression. We force-fed adult mice a LXR agonist, T0-901317. The results show a 3-fold induction of hepatic INSIG-2a mRNA after 12 h of LXR agonist treatment (Fig. 8B).

DISCUSSION
In this study, we have shown that insulin stimulates SREBP-1c transcriptional activity in the absence of LXR binding activity, and this results in the increased expression of SREBP-1c target genes during the late gestation period. In contrast, the suckling period is a unique situation in which a high LXR␣ transcriptional activity but a low plasma insulin results in the retention of SREBP-1c precursor form into the ER, possibly because of a high INSIG-2a expression.
During late gestation, several lines of evidence suggest that insulin is a major factor triggering SREBP-1c expression. SREBP-1c expression increases in parallel with fetal plasma insulin concentration, and insulin administration is sufficient to stimulate SREBP-1c expression and transcriptional activity. It is interesting that in contrast with what has been observed in the adult liver (17,24,26), SREBP-1c mRNA primary expression occurs in the complete absence of LXR␣ binding activity. It has been recently reported (36,37) that insulin activates SREBP-1c expression through LXR binding on the SREBP-1 promoter. In our experiments, it is clear that during late gestation LXR binding activity is absent, and thus the contribution of LXR on insulin-induced SREBP-1c transcription must be negligible at this stage of development.
The regulation of SREBP-1c after birth displays drastic and unexpected changes throughout the suckling period. Our studies support the recent findings of Botolin and Jump (35), which show that despite high SREBP-1c mRNA content and SREBP-1c precursor protein, the SREBP-1c mature form and SREBP-1c target genes are very low. However, in the previous study, the authors gave no explanation as to why this may occur. During the suckling period, plasma insulin concentration is very low, and thus another factor must be responsible for the induction of SREBP-1c transcription. Here, we propose that LXR, a known activator of SREBP-1c transcription, is responsible for the high SREBP-1c gene expression during the suckling period. This is supported by the fact that we observed high LXR␣ mRNA, high LXR binding activity, and high expression of other known LXR target genes such as ABC-A1 and LPL during that period. Furthermore, milk is rich in cholesterol, which could induce an increased level of oxysterols, the natural LXR ligands. Oxysterols, through binding to the LXR, trigger mainly the expression of enzymes involved in bile acids production. Because the diet of suckling animals is extremely rich in triglycerides, the hepatic production of bile acids is an absolute requirement for complete fat digestion. FIG. 5. Levels of LXR expression and binding activity during the suckling period. Liver RNA and nuclear extracts were prepared from mouse suckling pups at days 1, 8, and 13 (d1, d8, and d13) and from weaned mice at day 28 (W). A, LXR␣ and LXR␤ transcript levels were quantified by real time quantitative PCR and expressed in fold increase over the value observed in the day 1 newborn pups. Each value represents the mean Ϯ S.E. of six determinations. **, significant difference versus day 1 newborn pups (p Ͻ 0.01); §, significant difference versus weaned mice (p Ͻ 0.05). B, EMSA with nuclear extracts from suckling mouse pups at days 1, 8, 13, and 15 (d1, d8, d13, and d15) and from 28-day-old weaned mice (W). 3 g of proteins were incubated in presence of 0.3 ng of the radiolabeled LXRE-binding site. This EMSA is representative of three independent experiments.
We also propose that the absence of SREBP-1c nuclear form during the suckling period is linked to the induction of INSIG-2a through LXR stimulation (Fig. 8B). During suckling and at weaning, the pattern of INSIG-2a expression is similar to the expression of known LXR target genes, including SREBP-1c. The presence of INSIG-2a during the suckling period would then explain the absence of the nuclear mature form of SREBP-1c and thus the absence of SREBP-1c target gene expression. At weaning, the decrease in the diet cholesterol content would reduce the synthesis of LXR ligands, and the increase in plasma insulin would lead to a rapid decrease in INSIG-2a (Fig. 8A) and thus the appearance of a SREBP-1c binding activity in the nucleus (Fig. 7). Moreover, our observations have shown an opposite expression profile for INSIG-1 and INSIG-2a mRNA. This suggests that in contrast with INSIG-2a, INSIG-1 does not interact with SREBP-1c to induce its retention in the ER. In fact, SREBP-1c is present in liver nuclei during late gestation and at weaning despite elevated FIG. 6. GK and FAS mRNA expression levels during the suckling period. Liver RNA was extracted from mouse suckling pups at days 1, 8, and 13 (d1, d8, and d13) and from 28-day-old weaned mice (W). GK and FAS transcript levels were quantified by real time quantitative PCR and expressed in fold increase over the value observed in the day 1 newborn pups. Each value represents the mean Ϯ S.E. of six determinations. **, significant difference versus day 1 newborn pups (p Ͻ 0.01); § §, significant difference versus day 13 suckling pups (p Ͻ 0.01).  (16.5, 17.5, and 19.5) and from suckling pups at days 1, 8, 13, and 15 (d1, d8, d13, and d15) and from 28-day-old weaned mice (W). The INSIG-1 and INSIG-2a transcripts were quantified by real time quantitative PCR. Each value represents the mean Ϯ S.E. of six independent determinations. **, significant difference versus day 1 newborn pups and weaned mice (p Ͻ 0.01); † †, significant difference versus day 15 suckling pups (p Ͻ 0.01). B, RNAs were isolated from adult mouse liver after force-feeding with either the vehicle or T0-901317. The INSIG-2a transcripts were quantified by real time quantitative PCR. Each value represents the mean Ϯ S.E. of six independent determinations. **, significant difference versus vehicle treated mice (p Ͻ 0.01).
INSIG-1 mRNA level. Conversely, because mature SREBP-2 is present in the liver nuclei during suckling (35), this tends to indicate that INSIG-2a does not interact with SREBP-2.
What could be the physiological significance of this unique regulation of SREBP-1c during suckling? SREBP-1c can be considered as a transcription factor reflecting carbohydrate availability. Its transcription is induced by insulin in the liver (13), and its target genes include GK, the first enzyme of glucose utilization necessary for both glycogen and lipid synthesis. SREBP-1c together with the glucose-responsive carbohydrate-responsive element-binding protein transcription factor induces also lipogenic genes such as FAS, acetyl CoA carboxylase, and stearoyl CoA desaturase 1 (38 -40). Finally, SREBP-1c is also able to repress genes involved in hepatic glucose production such as the phosphoenolpyruvate carboxykinase (41,42). SREBP-1c expression is also inducible in the presence of a high cholesterol diet through LXR activation. This link between cholesterol and lipid metabolism could allow achievement of a proper cellular cholesterol/fatty acid ratio, an important membrane parameter. However, when a high cholesterol/triglyceride, low carbohydrate diet is absorbed (suckling period), fatty acid synthesis is not anymore necessary, and low glucose availability precludes lipid synthesis from this substrate. Induction of SREBP-1c transcriptional activity in these conditions would be deleterious because it would favor glucose utilization and repress glucose production by the liver in a period of glucose shortage, leading to hypoglycemia. The mechanism that we are describing with a potential role for INSIG-2a would then be a safety mechanism allowing escape from such an adverse situation. One would predict that this mechanism is switched on each time a high cholesterol, triglyceride/low carbohydrate diet is consumed. At weaning, when the plasma insulin concentration increases, this leads to a decrease of INSIG-2a, then allowing the maturation and translocation of SREBP-1c to the nucleus, where it activates its target genes.