In the present study, we have identified lipid-binding protein ACBD3 as a new binding partner and maturation modulator of SREBP1, and demonstrated ACBD3's ability to influence de novo biosynthesis of fatty acids by regulating FASN and ACC. These data significantly broaden our views on the biological roles played by ACBD3 and provide novel insights into the mechanisms underlying how lipid-binding proteins affect lipid homeostasis.
In cells, there are several families of lipid-binding proteins such as ACBDs, FATPs and FABPs, which function coordinately to regulate homeostasis and function of fatty acids. Entry of free fatty acids into the cells can be mediated through either passively diffusing through cell membrane or actively uptaking by FATPs. Intracellular fatty acids are then bound to either FABPs or ACBDs following conversion to fatty acyl-CoA esters by acyl-CoA synthetases 
. In those forms of complexes with chaperone proteins, fatty acids may be transported to various cellular compartments for executing their multifaceted biological functions, such as to the ER for synthesis of biological membrane and more complex lipids, to mitochondria for generating ATP via β-oxidation, to the lumen of secretory pathway for membrane protein lipidation, to the nucleus for signaling lipid-mediated transcription, or to lipid droplets for energy storage as triglycerides 
Considering their amphipathic physicochemical properties and complicated biological functions, fatty acyl-CoA esters need to be tightly regulated inside the cells. Accumulating evidence has demonstrated that the ACBD family proteins, acting as intracellular acyl-CoA pool formers and transporters, play critical roles in lipid metabolism and energy homeostasis 
. ACBD1 (also known as Acyl-CoA Binding Protein or ACBP) is the prototypical ACBD family member and has been investigated most intensively. This 87-amino-acids protein, whose ACB domain covers its entire sequence, was found to bind with acyl-CoA ester with a Kd
at a nanomolar level 
. Homozygous deficiency of this housekeeping gene resulted in embryonic lethality in mice, and siRNA-mediated ACBD1 knockdown in HepG2 cells initiated transcriptional changes on 22 genes related to cellular lipid metabolic process and caused significant reduction of palmitate content 
. Observations that augmentation of muscular fatty acyl-CoA in obese Zucker rats was much higher than that of ACBD1 and that long-chain acyl-CoA esters activated ATP-sensitive potassium channels inhibiting β-cell excitability obviously implicated ACB-containing proteins in the development of obesity and diabetes 
Lipid-regulating function of ACBD3, however, was found so far to be restricted to cellular steroidogenesis that occurred in mitochondria 
. Following hormone stimulation, PKA/cAMP-related signals initiated formation of a protein complex on the mitochondrial membrane, containing ACBD3, mitochondrial translocator protein, protein kinase A regulatory subunit 1a, ACBD1 and steroidogenic acute regulatory protein, and subsequently mediated cholesterol transport into mitochondria, the rate-limiting step along the pathway. According to the findings of the present study, biological functions of ACBD3 are expanded significantly into area of de novo lipogenesis to modulate fatty acid biosynthesis. Due to the embryonic lethality induced by ACBD3 gene knockout in mice and the fundamental nature of fatty acids in cell survival, growth, proliferation and death, the physiological and pathological roles undertaken by ACBD3 appear to be more important and complicated than those currently recognized in the field.
Intracellular SREBPs can be modulated by different nutritional and hormonal cues at several levels, including transcription, translation, processing (maturation) and degradation. Intracellular processing of SREBP1A and SREBP2 could be stimulated by depletion of cellular sterol content 
. On the other hand, maturation of SREBP1A and SREBP1C could be inhibited by incubation with unsaturated fatty acids but promoted by blocking phospholipid (phosphatidylcholine) production 
. Meanwhile, expression and activation of SREBP1C is resistant to sterol deficiency but sensitive to feeding and treatment of insulin and other growth factors (as shown in Figure S5
. Following receptor activation by growth factors, phosphoinositide 3-kinase (PI3K)/Akt (protein kinase B)-mediated signaling pathway seems to play a dominant role in activating SREBP1 (intensively reviewed in 
SREBP processing and activity appears to also be sensitive to a variety of stressful stimuli. NMDA-mediated excitotoxicity in neuronal injury has been also linked to increased activity of SREBP1 by augmented Insig1 degradation 
. Similarly, via depleting Insig1, both ER stress inducers and hypotonic stress inducers were found to promote proteolytic processing of SREBPs 
. Under fasting condition, activation of glucagons/cAMP/PKA signaling cascade caused direct phosphorylation at two consensus PKA recognition sites located in the ligand binding/heterodimerization domain of LXRα and concomitantly suppression of SREBP1C gene transcription 
. Two recent studies demonstrated that Sirtuin 1, a key NAD+
-dependent deacetylase involved in calorie restriction (CR)-mediated lifespan extension, directly deacetylate both SREBP1 and SREBP2, enhancing their ubiquitination and thus decreasing their stability 
. Together, all of these observations indicated that SREBP-regulated lipogenic pathway represents a fundamental adaptation mechanism, which is critical for cells to response to their dynamically-changed living environment.
Our results in and S1A
clearly showed dose-dependent reduction of nuclear SREBP1 (nSREBP1) and the corresponding enhancement of membrane form (fSREBP1) at 48 hours after ACBD3 vector transfection. Moreover, it was displayed that ACBD3 physically bound with fSREBP1 but not nSREBP1 (). Meanwhile, our results excluded possibility that ACBD3-inhibited SREBP1 maturation is due to ACBD3's disturbing effect on the physical interaction among SREBP1-SCAP-Insig1 (Figure S3
). These findings strongly implicated that ACBD3, via binding with a non-N-terminal domain of fSREBP1 on the Golgi membrane (), blocks a process of S1P/S2P-mediated proteolytic cleavage. Considering ACBD3's putative inhibitory function on ER-to-Golgi vesicular trafficking 
, SREBP1 transfer from the ER to Golgi may be also disrupted in the presence of overexpressed ACBD3. Owing to the well-conserved proteolytic processing steps among the three SREBP family members, it is probable that ACBD3 might exert a similar impact on SREBP2. In fact, we have observed that (Figures S1
) that manipulation of ACBD3 induced similar impact on SREBP2 as those of SREBP1. In addition, it is unlikely that ACBD3 directly regulates nSREBP1 post-translational modification, degradation and/or transactivation capacity because of their undetectable physical interaction (), the ACBD3-induced upregulation of fSREBP1 ( and S1A
), and significantly different subcellular localization (cytosol vs. nucleus).
At 72 hours after transfection (Figure S1A
), ACBD3 overexpressed from the lowest-dose transfection apparently boosted the fSREBP1, while those from higher-dose transfection largely downregulated the full-length protein. This discrepancy could be explained by SREBP1's auto-regulatory function and lag time between gene transcription, protein translation and degradation. We might expect to see the reduced fSREBP1 if cells were allow to overexpress ACBD3 at the lowest level for longer duration than 72 hours. This expectation was further supported by the observation that all three different levels of overexpressed ACBD3 promoted the expression of fSREBP1 at 48-hours time point.
Additionally, the different responsiveness of FASN and ACC to ACBD3 overexpression at 48 and 72 hours following transfection () could also be attributed to the lag time between transcription inhibition and protein turnover of FASN and ACC. At 48 hours after transfection, pre-existing FASN and ACC seemed to contribute to the irresponsiveness even though nSREBP1 was down-regulated (Figure S1A
) and transcription of the two target genes was inhibited () at this point. With the prolonged ACBD3 overexpression (72 hr after transfection), significant amounts of the pre-existing proteins were degraded and there was no enough newly-synthesized proteins to compensate in the system. As a result, dose-dependent decreases of steady-state FASN and ACC were displayed under this prolonged expression condition. These findings were in accord with the results from experiments quantifying de novo palmitate synthesis (). Palmitate production was significantly reduced only when ACC and FASN protein levels were decreased.
Utilizing molecular and biochemical approaches (), we found that N-terminal domain (residues 1–171) of ACBD3 played an important role in its inhibitory function on the intracellular processing and transcriptional activity of SREBP1. This observation implied that the ACB domain (residues 80–171), accounting for more-than-half portion of the deleted N-terminal sequence, might be required for ACBD3 to modulate SREBP1 activation to highest extent. Moreover, confocal imaging results showed that the ΔN mutant still localized mainly on the Golgi just like the full-length wildtype, indicating that Golgi association was not sufficient for its maximum inhibitory function. This observation, from a different perspective, supported the concept that ACB domain might involve in ACBD3-mediated SREBP1 regulation. On the other hand, deletion of C-terminal GOLD domain did not disrupt ACBD3's function to inhibit SREBP1 activity, revealing that these residues (381–526) were not required for its effect. Interestingly, unlike full-length and ΔN ACBD3, this ΔC mutant is localized more diffusely and only a small portion appeared to be located on the Golgi. Based on this finding, we still could not exclude the possibility that Golgi association contributes to ACBD3's effects on SREBP1.
As one master lipid regulator, Akt/mTORC1-regulated SREBP function is apparently fundamental for maintaining structure and energy homeostasis of cells, determining cell survival and growth. Accordingly, it is not surprising that altered SREBP expression and activity have been implicated in imbalanced lipid metabolism during the development of many dyslipidemic diseases such as obesity, diabetes, hepatic steatosis, cancer, neurodegeneration, and schizophrenia 
. Thus, measures to curtail lipogenic function of SREBPs have displayed therapeutic potentials on different disorders involving imbalanced lipid metabolism. Metformin, a widely-used AMPK activator in clinics for type 2 diabetes, was reported to ameliorate fatty liver by reducing mRNA and nuclear protein of SREBP1 
. Through facilitating physical SCAP-Insig1 interaction, small molecule betulin inhibited SREBP maturation and subsequently attenuated target gene transcription. Betulin administration improved the lipid profiles and insulin resistance in mice fed with Western-diet and antagonized atherosclerotic lesion formation in LDLR-deficient mice 
. Due to established roles of FASN overexpression in tumor growth and malignancy, targeting on fatty acid de novo synthesis for cancer therapy had been intensively explored. Direct FASN inhibition by its selective inhibitors was one of the most popular interventions to be tested under various in vitro and in vivo settings, and many supportive results were reported 
. Apparently, discoveries of novel signaling pathways that regulate lipid metabolism via controlling SREBP transcription, expression and maturation are desirable from not only basic but also clinical perspectives.
The present study undoubtedly revealed that ACBD3 overexpression reduced SREBP1 maturation, transactivation ability, and ultimately, fatty acid de novo synthesis. Furthermore, with an extensively-employed cell model of SREBP regulation, ACBD3 demonstrated its potential to control cellular lipid homeostasis (). Considering the multifaceted cellular functions of lipids, these observations suggest that ACBD3 may have many unappreciated lipid-related functions in various cells and that it would be a feasible measure to manipulate ACBD3 expression and function for modulating many SREBP1-involved physiopathological conditions.
To the best of our knowledge, the current study is the first report showing a physical and functional connection between a lipid-binding protein (ACBD3) and a master lipid regulator (SREBPs). Our findings provide evidence that ACBD3 is involved in lipid homeostasis through influencing cellular lipogenic pathway by regulating intracellular maturation of SREBP1. More studies are warranted to clarify ACBD3's biological roles in maintaining lipid homeostasis and therapeutic potentials for diseases attributed to imbalanced lipid metabolism such as obesity, diabetes, cancer and neurodegeneration.