Adipocyte Biology from the Perspective of In Vivo Research: Review of Key Transcription Factors

Abstract

Obesity and type 2 diabetes are both significant contributors to the contemporary pandemic of non-communicable diseases. Both disorders are interconnected and associated with the disruption of normal homeostasis in adipose tissue. Consequently, exploring adipose tissue differentiation and homeostasis is important for the treatment and prevention of metabolic disorders. The aim of this work is to review the consecutive steps in the postnatal development of adipocytes, with a special emphasis on in vivo studies. We gave particular attention to well-known transcription factors that had been thoroughly described in vitro, and showed that the in vivo research of adipogenic differentiation can lead to surprising findings.

Keywords: CEBP/α; CEBP/β; CEBP/δ; CREB; PPARγ; adipogenesis; knockout; obesity; transcription factor.

Figure 1

An illustration of the phenotypes of systemic C/EBPβ (−/−), C/EBPδ (−/−), and C/EBPβ (−/−)∙C/EBPδ (−/−) transgenic mice. (A) C/EBPβ knockout: 35% of newborn C/EBPβ (−/−) mice died within 24 h after birth. The surviving mice had small lipid droplets in isBAT (interscapular BAT) with a normal expression of C/EBPα and PPARγ. Adult mice had normal eWAT (epididymal WAT) and a decreased expression of UCP1 (the functional marker of terminally differentiated BAT). The differentiation of mouse embryonic fibroblasts (MEFs) from C/EBPβ (−/−) mice was significantly impaired. (B) C/EBPδ knockout: small lipid droplets in isBAT, normal eWAT in adults, and MEF differentiation were normal or slightly impaired. (C) C/EBPβ + C/EBPδ knockout: 85% of newborn C/EBPβ (−/−)∙C/EBPδ (−/−) mice died within 24 h after birth. The surviving mice had no lipid droplets in isBAT, and UCP1 expression was markedly reduced, with a normal expression of C/EBPα and PPARγ. Adult mice had a reduced volume of eWAT with a normal expression of C/EBPα and PPARγ. MEFs did not differentiate into mature adipocytes.

Figure 2

Embryonic and neonatal adipose tissue development was C/EBPα-independent. (A) Pregnant female control mice (*mice contain only Adn-rtTA and C/EBPα flox/flox) were given doxycycline-supplemented chow during the E11-E18 embryonic days (the period of subcutaneous WAT development), which induced the C/EBPα knockout in the embryos. After E18 doxycycline supplementation was stopped, C/EBPα expression was restored. (B) Neonatal C/EBPα flox/flox pups were put on doxycycline-supplemented chow from P0 (postnatal day 0), which induced C/EBPα knockout, until P42 (the period of epididymal WAT development). Both epididymal and subcutaneous WAT were comparable to the control. (C) Pregnant female control mice were given doxycycline-supplemented chow from E From P0 until P42, newborn mice continued to receive doxycycline-supplemented chow. Thus, C/EBPα was knocked out during both critical periods of WAT development (embryonic period, critical for subcutaneous WAT, and postnatal period, for epididymal WAT). Either subcutaneous or epididymal WAT were comparable to the control.

Figure 3

C/EBPα is indispensable for de novo adipogenesis in adults. The transgenic line was derived by crossing the inducible C/EBPα floxed/floxed mice with FAT-ATTAC mice (FAT apoptosis through triggered activation of caspase-8). FAT-ATTAC mice expressed an inactive form of caspase-8 under the aP2 promoter (thus, in adult adipocytes). A single treatment with a dimerizer activates caspase-8 and induces apoptosis in mature adipocytes. A week after the dimerizer treatment, the fat depots were significantly reduced. These mice were then put on a doxycycline chow diet (for C/EBPα knockout) or on a chow diet (C/EBPα was expressed). In mice on the chow diet, their fat depots recovered to approximately 50% of the original tissue, while in mice on the doxycycline chow diet, their fat pads were still reduced.

Figure 4

Metabolic changes in mice with a C/EBPα knockout in adipose tissue. Four weeks under normal chow feeding led to a normal phenotype of mature adipocytes, but impaired the insulin-stimulated phosphorylation of Akt and ERK1/2 in WAT depots and a decrease in the systemic adiponectin level. Four weeks under HFD conditions induced a slower weight gain (compared to C/EBPα-expressing mice), which is thought to be due to adipocyte hypertrophy. After another one to two weeks of HFD feeding, C/EBPα −/− mice begin to lose weight moderately (presumably due to the termination of hypertrophic adipogenesis), while the normal adipocyte morphology in WAT and BAT depots was retained.

Figure 5

PPARγ is essential in embryonic adipose tissue development and adult adipogenesis. (A) Pregnant female control mice (*mice contain only Adn-rtTA and C/EBPαflox/flox) were given doxycycline-supplemented chow from day E11 until birth. Newborn pups were given doxycycline-supplemented chow from P0 until PThus, PPARγ was knocked out in adipose tissue from E11 until PThese mice had miniscule subcutaneous WAT and increased epididymal WAT (due to compensatory overgrowth). (B) Under normal chow diet conditions, the adipose tissue of adult mice with the PPARγ knockout had normal morphology, though these mice had serious metabolic abnormalities: insulin resistance, increased VLDL (very low-density lipoprotein), and decreased adiponectin levels. In contrast, PPARγ −/− adult mice on a doxycycline-containing a high-fat diet developed severe weight loss and insulin resistance.

Figure 6

Overlapping functions of PPARγ and C/EBPα in maintaining the survival of mature adipocytes. The double knockout of PPARγ and C/EBPα was induced by doxycycline-containing chow diet supplementation for 10 days. The weight of the transgenic mice has not changed, while almost all adipocytes of subcutaneous WAT were disrupted and dead. Epididymal WAT size was comparable to control.