Many factors have already been identified, mainly by RNA analysis, which regulate cell fate and control the transformation of preadipocytes into mature adipocytes. This process confers massive changes in the structure of the cell, which leads to storage of fat and the formation of fat vacuoles. The orchestrated changes in the structure of the preadipocyte require tight control of the involved factors.

An extensively studied model for adipogenesis in vitro is the mouse embryonic fibroblast cell line 3T3-L1 [2-4]. 3T3-L1 cells are grown to confluency and among stimulation with a hormone cocktail consisting of insulin, dexamethasone and isobutylmethylxanthine (IBMX), cells re-enter cell cycle within 24 to 36 h. Two rounds of cell division are performed after which they permanently withdraw from the cell cycle, begin to accumulate lipid, and undergo terminal differentiation into mature adipocytes. Until recently, regulating mechanisms of the earliest step in the differentiation process are poorly understood in comparison to later stages of adipogenesis.

The orchestrated changes in the expression of structural proteins require changes in the expression and activity of factors controlling the expression of structural proteins. Induction of adipogenesis by hormones leads to the expression of C/EBPβ and C/EBPδ within 2 hours [5], which activate the expression of PPARγ and C/EBPα at low levels. PPARγ and C/EBPα activate each others' expression in a positive feedback loop that promotes and maintains the differentiated state [6]. The latter proteins orchestrate the changes in gene expression typical for mature adipocytes. Besides the C/EBP family and PPARγ, many other factors are known that play a role in adipogenesis. Most of these factors are regulated on the transcriptional level and mRNA steady state levels can be easily measured. However, there is increasing evidence that proteins, which are controlled at the translational level, are of uttermost importance for adipogenesis and other cellular processes. A handful of translationally controlled proteins has been characterized in adipogenesis up to now. The C/EBP factors C/EBPα and C/EBPβ are regulated at the posttranscriptional level, whereupon various isoforms arise from a unique mRNA by differential initiation of translation [7]. In 3T3-L1 cells, high eIF2 and eIF4E activity shifts the ratio of C/EBP isoforms expression toward more truncated isoforms [7]. An important factor that controls adipogenesis at the translational level is HZF [8]. HZF enhances translation of C/EBPα mRNA by binding to its 3´-untranslated region. Knock-down of HZF disrupts adipogenesis.

In most cases, translation efficiency is closely related to ribosomal association of the respective mRNA. The degree of ribosomal association can be easily measured by separation of the mRNAs in a sucrose gradient by velocity sedimentation [9]. The gradient is fractionated, mRNA is isolated from each fraction and the ratio of the 28S/18S rRNA is used to determine the fractions containing free RNA (no ribosomes bound) and polysomal RNA [10]. By use of microarrays, the composition of the RNA pools can be determined easily [11,12].

There are several studies of transcriptional changes during adipogenesis and many factors have been identified that play an important role in this process. Burton et al., 2002 identified 286 clones with a greater than fivefold deviation of expression during the first 24 hours of adipogenesis [5]. In our study, none of these factors showed changes greater than fourfold with the exception of c-Myc. We conclude that analysis of translational activity is important for a full understanding of the processes controlling adipogenesis.

In the last few years, interest increased in the dynamics of the proteome and there is a handful of studies in man and mouse [18-24] about changes of the proteome, comparing preadipocytes and adipocytes in tissues, or later stages of adipogenesis. Here, we report changes as soon as 6 hours after hormonal induction of adipogenesis, a timepoint, at which clonal expansion of 3T3-L1 cells is induced and expression of mitotic and adipogenic genes is initiated.

3T3-L1 adipocytes are grown to confluency and upon stimulation with a hormone cocktail (insulin, dexamethasone and IBMX), post-confluent G₀ cells reenter one to two rounds of cell-cylce called mitotic clonal expansion (MCE) [25]. It has been proposed that MCE might facilitate the DNA remodeling for the adipogenesis gene expression program [2]. As 3T3-L1 cells reenter cell cycle by passing from G₀ to G₁ phase, it might be expected that the translation machinery is activated after hormonal stimulation. In fact, we detected a shift of mRNAs toward the higher molecular weight polysomal fractions, mostly derived from a general activation of gene expression. Expression of ribosomal proteins, ribosome assembly proteins and ribosomal RNA (rRNA) are up-regulated in mitotic active cells. Hence it is not surprising that ribosomal proteins (RP) are prominent among the up-regulated proteins (compare Table ​Table1).1). Additionally, for some RPs, extraribosomal functions have been demonstrated. Many of these extraribosomal functions can be attributed to the regulation of cell cycle and for several RPs a role in cancer, promotion of cell growth or differentiation has been shown [26]. Some of these known extraribosomal functions might well explain the role of the RPs in the early phase of adipogenesis (Figure ​(Figure33).

RPL6 was initially identified as up-regulated in gastric multidrug-resistant cancer cells [27] and was shown to protect gastric cancer cells from drug-induced apoptosis. Furthermore, RPL6 over-expression promotes G1 to S phase transition of gastric cancer cells and promotes cell growth [28] (Figure ​(Figure3).3). RPL7a interacts with the human thyroid hormone receptor and inhibits transactivation. Thyroid hormone signalling stimulates adipogenesis [29,30] and up-regulation of RPL7a might participate in mitotic control. RPL18 was shown to inhibit autophosphorylation of the double-stranded RNA-activated protein kinase (PKR) and PKR mediated phosphorylation of the translation initiation factor eIF2α. Over-expression of RPL18 reduced eIF2α phosphorylation and stimulated translation of a reporter gene in vivo [31]. Over-expression of RPL18 is thought to promote protein synthesis and cell growth through inhibition of PKR activity [31], which also might hold true for adipogenesis. A polymorphism in the promoter region of the RPL27a gene was associated with meat marbling (accumulation of intramuscular fat) in Japanese Black beef cattle [32]. Furthermore, RPL27a is ubiquitinated in a cell-cycle specific manner, leading to increased translational efficiency of the ribosomes [33].

RPSa, which was up-regulated in our study, was previously known as 37-kDa laminin receptor precursor/67-kDa laminin receptor (LRP/LR). It has a number of functions depending on its subcellular localisation. In the nucleus, RPSa binds to DNA by histone binding, in the cytoplasm it is associated with the 40S small ribosomal subunit and at the cell surface it acts as a receptor for various components [34]. It confers an anti-obesity effect when stimulated by the green tea catechin EGCG [35]. Interestingly, RPSa inhibits insulin stimulation of 3T3-L1 mitogenesis and EGCG inhibited differentiation of preadipocytes to adipocytes [36].

Most of the RPs up-regulated immediately after hormonal induction were shown to stimulate cell cycle which is concordant with the fact that 3T3-L1 cells undergo mitosis after stimulation. Translational control allows for rapid changes of protein redundancy and it may be speculated that proteins that initiate MCE and reprogramming of gene expression are regulated at the translational level. Therefore we suggest that the rapid increase of L6, L7a, L18, L27a, Sa, and S18 may reflect their importance of adipogenesis control in 3T3-L1 cells.

Higher translation rates require transport of amino acids, and we detected up-regulation of the amino acid transporters SLC25a5 and SLC25a30. Higher translation rates also lead to increased misfolding of nascent polypeptide chains. Up-regulation of chaperones in translation promoting conditions has been described before and was also observed in our study (BAG3, HSPA8, HSP90ab1).

The PI3K-AKT-mTOR pathway, which is stimulated by insulin, has been identified to be essential for many cellular processes (reviewed in [37]). mTORC1 is a protein complex containing mTOR (mammalian target of rapamycin) and raptor. mTORC1 activates S6K1, a kinase that promotes protein synthesis and cell growth by phosphorylation of multiple substrates including components of translation initiation or elongation such as ribosomal protein S6, eIF4B and eukaryotic elongation factor 2 kinase [38]. One target of this pathway is eIF4B (Figure ​(Figure4)4) (reviewed in [39]). It was suggested that phosphorylation of eIF4B by S6 kinases, which are regulated by mTOR, stimulates its function. Indeed, this phosphorylation event favors recruitment of eIF4B into complexes with eIF3, which promotes the recruitment of ribosomes to the 5´end of the message (reviewed in [40]). eIF4B, which was up-regulated in our study, stimulates the RNA helicase activity of eIF4A in unwinding secondary structures in the 5´-untranslated regions (5´-UTR) of mRNAs [41,42]. By knock-down of eIF4B, selective reduction of translation was observed for mRNAs harboring strong to moderate secondary structures in their 5´-UTRs. These mRNAs code for proteins that function in cell proliferation (e.g. CDC25C, c-MYC) or cell survival (e.g. BCL-2). Silencing of eIF4B also leads to decreased proliferation rates and caspase-dependent apoptosis: eIF4B is required for cell proliferation and survival by regulating the translation of proliferative and prosurvival mRNAs [43]. PPARγ expression is stimulated in response to mTORC1 [44]. PPARγ is a key adipogenic factor and exogenous expression is sufficient to induce adipogenesis. Zhang et al., 2009 discuss the possibility that AKT and mTORC1 facilitate adipogenesis by up-regulation of PPARγ via regulation of FOXO1 [44]. However, they do not discuss the activation of eIF4B upon mTORC1 activation with subsequent changes in the preference of ribosomes for certain mRNAs. We think that regulation of C/EBPα could possibly be explained by up-regulation of eIF4b activity, as members of the C/EBP family are regulated at the translational level (Figure ​(Figure44).

Microarray results of polysomal fractions from 3T3-L1 cell lysate (6 h after hormonal induction) show up-regulation of eIF4B and MYC (arrow head on top) and down-regulation of Ghrelin (arrow head below). IR, insulin receptor; Pol I/II/III, RNA polymerase I/II/III

c-MYC over-expression in cycling cells has been reported to block exit from the cell cycle, accelerate cell division, and increase cell size (reviewed in [45]). When c-MYC levels are high, 3T3-L1 adipoblasts are locked in a proliferation-competent state and normal differentiation can not be activated. Persisting high levels of c-MYC can inhibit the expression of genes that promote adipogenesis namely C/EBPα and PPARγ2 and therefore prevent terminal differentiation of preadipocytes to mature adipocytes [46,47]. In microarray analysis, c-MYC was up-regulated in 3T3-L1 cells several hours after hormonal induction in a study by Burton et al., 2002 [5] and at day 2 of differentiation in a study by Kim et al., 2007 [48]. c-MYC is an important regulator of ribogenesis, as it activates Pol I, Pol II and Pol III leading to activation of expression of rRNA, tRNA, ribosomal proteins, initiation factors of translation and other cell cycle relevant genes [49]. Therefore, c-MYC activation might be important for activation of the translation apparatus at the entry of 3T3-L1 cells into G₁.

In our study, two genes were down-regulated: IFIT1 and ghrelin/obestatin prepropeptide (referred to as Ghrelin). The members of the IFIT gene family are silent in most cell types, but are activated by e.g. interferons [50]. IFIT proteins are considered silencers of translation and down-regulation might be another factor of translation stimulation.