![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 12, No. 20, pp. 3145-3148, October 15, 1998
Department of Biochemistry, University of Texas Southwestern Medical Center, Dallas, Texas 75235-9152 USA
Two papers appearing in this issue report the
generation of transgenic mice largely devoid of white adipose tissue
(WAT). In both cases the resulting mice display anatomical and
physiological properties very similar to human patients suffering from
generalized lipodystrophy. Such patients lack WAT, a condition
resulting from either genetic or autoimmune etiologies, and are
severely diabetic (Foster 1994 How was it possible to engineer WAT-free laboratory mice? The two
teams, headed by Brown and Goldstein in Texas (Shimomura et al. 1998 Before delving into the interesting physiological and anatomical
abnormalities observed in WAT-free mice, it is useful to review the
respective experimental approaches in the context of the gene
regulatory pathways that were perturbed. Moitra and colleagues made use
of an artificial transcription factor designated A-ZIP/F. In vitro assays had already demonstrated that the A-ZIP/F
protein functions as an inhibitor of various members of the bZip family of transcription factors (Krylov et al. 1997 Extensive evidence has demonstrated the activities of
C/EBP family members in the growth and differentiation of
WAT. For example, forced expression of either C/EBP The WAT-free phenotype observed by Shimomura and colleagues following
forced, adipocyte-specific expression of a constitutively active form
of SREBP1c was less clearly anticipated. Indeed, as will be discussed
subsequently, it might have been expected that this manipulation would
enhance adipogenesis. This dilemma calls for a brief review of previous
studies on the SREBP family of transcription factors (for more detail,
see Brown and Goldstein 1997). The origin of the SREBPs stems from
studies of the gene encoding the low density lipoprotein (LDL)
receptor. Transcription of the LDL receptor gene is elevated in cells
starved for cholesterol. This observation led to the discovery of a
cis-regulatory DNA sequence, designated the sterol response
element (SRE), that is necessary for transcriptional induction of the
receptor gene upon cholesterol starvation. A DNA-binding activity
designated SRE-binding protein (SREBP) was purified and partially
sequenced, leading to the cloning of its encoding gene. Not unlike many
other mammalian transcription factors, the SREBPs come in a variety of
isoforms encoded by two distinct genes, designated SREBP1 and
SREBP2. The SREBP1 gene directs, by alternative
splicing, isoforms designated SREBP1a and SREBP1c, with the former
having a more potent activation domain than the latter.
Biochemical and molecular biological studies of SREBP have unveiled a
novel pathway by which its activity is regulated in response to the
availability of cholesterol. The mature forms of SREBP, which appear
under conditions of cholesterol starvation, are conventional
transcription factors containing a bHLH-Zip DNA binding domain and an
acidic activation domain. When cholesterol levels are abundant,
however, all SREBP isoforms exist as immature precursors. Latent SREBPs
are much larger polypeptides that attach to the endoplasmic reticulum
via two membrane spanning segments. Upon cholesterol starvation the
immature precursors are cleaved by a sterol-sensitive proteolytic
event, resulting in the release of a mature, amino-terminal fragment
that enters the nucleus and orchestrates the coordinate activation of
genes required for the biosynthesis of cholesterol and fatty acids.
SREBP1c was independently discovered by Spiegelman and colleagues in an
expression screen for proteins that bind to an E-box sequence in the
promoter of the gene encoding fatty acid synthase (Tontonoz et al.
1993 Because SREBP1c/ADD-1 is expressed in the WAT of
laboratory mice, it is most sensible to hypothesize that this
transcription factor should play a stimulatory role in adipogenesis. It
thus comes as some surprise to find that transgenic animals expressing a dominant-positive (mature) form of SREBP1c are compromised with respect to WAT abundance and maturity. When assayed shortly after birth, the adipose tissue of transgenic, experimental animals appeared
similar to control littermates Why does forced expression of a dominant-positive form of SREBP1c
impede formation of mature WAT? Clues to this enigma come from the
extensive series of Northern blot assays conducted by Shimomura and
colleagues (1998) In contrast to the many WAT-specific mRNAs that were observed to be
substantially reduced in transgenic animals expressing the mature form
of SREBP1c, two specific mRNA were expressed at an increased level in
the residual WAT of experimental animals relative to control animals.
These include the mRNAs for tumor necrosis factor- What consequences result when a laboratory mouse is WAT-free? Not
surprisingly, the answers to this question were fully anticipated by
clinical scientists who have studied lipodystrophic patients for the
past century (for review, see Seip and Trygstad 1996 Now basic scientists can begin to think about a syndrome that clinical
scientists have recognized and studied for decades. Moreover, the field
is now provided with animal models that can be used to extend studies
of lipodystrophic diabetes. For the novices, including me, it comes as
some surprise that WAT-free mammals are diabetic. Having been schooled
to believe that diabetes is a disease of the obese, the uninitiated
might anticipate that the WAT-free condition would correspond to an
antidiabetic state. One simplistic take-home lesson is that type 2 diabetes can result from an imbalance in the abundance of WAT This begs, however, the question of what is the culprit? Several
candidates bear consideration, including TNF Having closed with a listing of two candidate molecules to consider as
diabetes `promoting factors,' it is encouraging to recognize that the
field now has several well-developed animal models that may open new
doors to the study of type 2 diabetes. Indeed, the recent work of
Graves and colleagues (Burant et al. 1997 At the very least, the present work on WAT-free mice should favor a
closer and more productive relationship between basic reserach and
clinical scientists. One can rest assured, for example, that basic
scientists interested in the molecular underpinnings of type 2 diabetes
will keep a close watch on the gene hunt for mutations that cause
congenital lipodystrophy.
Article
Top
Article
References
; Seip and Trygstad 1996). The present
reports provide fresh and interesting observations regarding the
physiological consequences of life without fat and, moreover, establish
animal models that offer new opportunities for the study of type 2 diabetes.
)
and Vinson at the National Institutes of Health (NIH) (Moitra et al.
1998), attacked the problem using similar strategies. The Texans
directed adipocyte-specific expression of a truncated, constitutively
active form of SREBP1c, a transcription factor involved in the
regulation of genes required for the biosynthesis of both cholesterol
and fatty acids. The Federalists likewise directed the expression of a
modified transcription factor in WAT. In the latter case, the team
utilized clever methods to inhibit the function of endogenous members
of the Jun and CCAAT/enhancer-binding protein
(C/EBP) families of transcription factors. Expression of
these dominant-positive and dominant-negative transcription factors
were similarly directed to brown adipose tissue (BAT) and WAT by the
use of regulatory sequences associated with the adipocyte-specific aP2
gene (Ross et al. 1990). The resulting transgenic mice were remarkably
similar
WAT-free, profoundly insulin resistant, and rampantly
diabetic.
). The
A-ZIP/F protein started with the leucine zipper of
C/EBP
, which was modified so as to dimerize
promiscuously with all members of the C/EBP and Jun
families of transcription factors. The protein was further manipulated
to replace the C/EBP basic region, normally employed to establish direct DNA contact, with an acidic, amphipathic
-helix. The resulting protein maintains the ability to dimerize
with the leucine zippers of all C/EBP and Jun family
members. A-ZIP/F further establishes an unusually stable
interaction with unadulterated bZip partners as a result of formation
of a continuous, coiled-coil dimer interface, including the leucine
zipper and a paired basic/acidic alpha helical extension.
(Samuelsson et al. 1991; Umek et al. 1991; Freytag and Geddes 1992;
Freytag et al. 1994) or C/EBP
(Yeh et al. 1995) has
been reported to enhance adipogenesis of cultured fibroblasts.
Likewise, antisense C/EBP RNA has been shown to
suppress coordinate, adipocyte-specific gene expression and
triglyceride accumulation during differentiation of 3T3-L1 preadipocytes (Lin and Lane 1992). Finally, gene knockout studies have
reported that C/EBP-deficient mice are devoid of WAT
(Wang et al. 1995), and that mice doubly deficient in both
C/EBP and C/EBP
display reduced
fat mass (Tanaka et al. 1997). The present observations of Vinson and
colleagues are therefore straightforward and sensible. By directing the
expression of the inhibitory A-ZIP/F protein to WAT, a
gene regulatory cascade essential for terminal adipocyte
differentiation is impeded, resulting in fat-free mice. This approach
offers a valuable step forward from the earlier C/EBP knockout experiments of Darlington and
colleagues (Wang et al. 1995). Because C/EBP is also
expressed in the liver, gut, and other tissues essential for energy
homeostasis, animals fully deficient in this transcription factor die
from hypoglycemia shortly after birth. As such, WAT-free,
C/EBP-deficient animals can not be studied using
traditional physiological approaches.
). Forced expression of SREBP1c in 3T3-L1 cells has been reported
to enhance fatty acid biosynthesis and adipocyte differentiation (Kim
and Spiegelman 1996), leading to the designation of this transcription
factor as adipocyte determination and differentiation factor-1 (ADD-1).
scanty and immature. By 6 weeks of age,
however, the WAT of control animals developed its normal mature
appearance, whereas that of the SREBP1c-expressing, transgenic animals
continued to resemble the immature WAT seen shortly after birth (see
Shimomura et al. 1998). Compared with the
A-ZIP/F-expressing mice, which are born with no metabolic WAT and fail to develop any throughout life, the impediment to WAT
development appears less severe in the case of the Texas mice. It will
become clear, however, that both lines of transgenic mice suffer very
similar physiological perturbations that closely mimic lipodystrophy in
humans.
. Compared with normal mice, transgenic animals forced
to express the mature form of SREBP1c contain substantially reduced
levels of a number of gene products essential for maturation and
function of WAT. Significant reductions were observed for the mRNAs
encoding PPAR
, C/EBP, leptin, insulin receptor,
insulin receptor substrate 1, insulin receptor substrate 2, and the
GLUT4 glucose transporter. Given this wide range of effects it is not surprising that the experimental animals would suffer impediments in
the formation of WAT.
(TNF) and
preadipocyte factor 1 (Pref-1). It is unclear how either of these gene
products might functionally impede adipocyte maturation. It is
interesting, however, that elevated levels of TNF have been
implicated in the genesis of insulin resistance (Hotamisligil and
Spiegelman 1994). Likewise, elevated Pref-1 levels point to the
possibility that forced expression of mature SREBP1c may block
adipocytes at an immature stage of differentiation. Intriguingly,
forced expression of Pref-1, a cell surface protein that contains
epidermal growth factor-like repeats and is expressed exclusively in
immature adipocytes, has been reported to impede adipocyte maturation
in cell culture assays (Smas and Sul 1993).
). Humans suffering
from congenital or acquired lipodystrophy become severely diabetic,
have voracious appetites, are hypermetabolic, and display an anabolic
syndrome that includes organomegaly of the liver, spleen, pancreas, and
kidney. These conditions perfectly describe the WAT-free mice developed
respectively in Texas and the NIH. Both papers provide compelling
evidence that WAT-free mice are rampantly diabetic as judged by a
thorough series of physiological assays. Moreover, both models show
organomegalyincluding fatty livers. Indeed, the anatomical and
physiological properties of the two lines of mice differ in but one
notable aspect. Even by birth the Texas mice contain a massive
enlargement of BAT, with continued hyperplastic growth of this tissue
during postnatal and adult development. This was not the case with the
NIH mice. Although residual adipose tissue present in the latter
animals resembled BAT, it was not hyperplastic. Finally, the BAT of
both animals appeared to be metabolically inactive with respect to thermogenesis. BAT coloration was faded in a manner consistent with a
reduced density of mitochondria. Moreover, Northern blot assays
conducted in both studies revealed very marked reductions in the
expression of the UCP1 mRNA.
too
little, just as too much, is unhealthy.
and free fatty acids
(FFAs). Spiegelman and colleagues have contributed extensive evidence
that TNF can impede signaling by the insulin receptor (Hotamisligil and Spiegelman 1994). Whether the modest increase in
TNF expression observed at the mRNA level in WAT-free mice is
sufficient to account for their diabetic state remains open to
question. McGarry and colleagues have developed the concept that type 2 diabetes may result from an increase in circulating levels of FFAs (for
review, see McGarry 1994). The mice studied by Moitre et al. show
modest elevations in FFAs in the fed state, but those studied by
Shimomura and colleagues do not. Similarly, some human patients
suffering from lipodystrophy have elevated FFAs, whereas other equally
insulin resistant patients have normal FFA levels (Seip and Trygstad
1996).
) has begun to show the
utility of this pathway. Transgenic mice expressing the A chain of
diptheria toxin in BAT and WAT lose adipose tissue as they age,
resulting in similar, if less dramatic, anatomical and physiological
perturbations as reported by Moitra et al. (1998) and Shimomura et al.
(1998). Graves and colleagues used these adipose attenuated mice to ask
whether insulin sensitivity could be restored in response to
troglitazone. Torglitazone is a member of the thiazolidinedione (TZD)
class of compounds that represent a new and exciting class of
antidiabetic drugs. TZDs are specific and potent ligands for PPAR
and have been found to promote both weight gain in rodents and
adipocyte differentiation in cultured cell models. It was therefore
reasonable to anticipate that TZDs might not be effective in treating
the diabetic symptoms of WAT-depleted mice. Remarkably, Graves and
colleagues have demonstrated that troglitazone restores insulin
sensitivity, alleviates hyperglycemia, and lowers serum cholesterol,
triglycerides, and FFAs in WAT-depleted mice. If correct, these
observations indicate that adipose tissue is not the target of TZDs.
| |
Footnotes |
|---|
1 E-MAIL smckni{at}biochem.swmed.edu; FAX (214) 648-3346.
| |
References |
|---|
|
Top
Article
References
|
|---|
.
Science
256:
379-382 and/or C/EBP gene.
EMBO J.
16:
7432-7443[CrossRef][Medline].
CiteULike 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Genetically Modified Animals in Endocrinology Endocr. Rev., June 1, 2004; 25(3): 512 - 519. [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Xia, K. L. Stanhope, E. Digitale, O.-M. Simion, L. Chen, P. Havel, and K. Cianflone Acylation-stimulating Protein (ASP)/Complement C3adesArg Deficiency Results in Increased Energy Expenditure in Mice J. Biol. Chem., February 6, 2004; 279(6): 4051 - 4057. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Murray, A. D. Sniderman, P. J. Havel, and K. Cianflone Acylation Stimulating Protein (ASP) Deficiency Alters Postprandial and Adipose Tissue Metabolism in Male Mice J. Biol. Chem., December 17, 1999; 274(51): 36219 - 36225. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
