SYMPOSIUM: REGULATION OF GENE EXPRESSION Acetyl-Coenzyme A Carboxylase Messenger Ribonucleic Acid Metabolism in Liver, Adipose Tissues, and Mammary Glands During Pregnancy and Lactation M. VER6NICA PONCE-CASTANEDA, FERNANDO L6PEZ-CASILLAS, and KI-HAN KIM Department of Biochemistry Purdue University West Lafayette, IN 47907
ful way, it is necessary to examine the metabolism of individual isoforms of acetyl-coenzyme A carboxylase mRNA. (Key words: acetyl-coenzyme A carboxylase, messenger ribonucleic acid, mammary glands)
ABSTRACT
In the rat, the acetylcoenzyme A carboxylase gene exists as a single copy per haploid chromosome set. However, multiple forms of acetyl-coenzyme A carboxylase mRNA exist, the relative abundance of which varies in a tissue-specific manner under different physiological conditions. In the mammary gland, the major acetyl-coenzyme A carboxylase mRNA species are of the class 2 type, which are products of promoter II. In parametrial white adipose tissue, the main form of species of acetyl-coenzyme A carboxylase is of the class 1 type, which are produced by promoter I. Pregnancy and lactation affect the amounts of these acetyl-coenzyme A carboxylase -A. Although the mammary gland acetylcoenzyme A carboxylase mRNA species increase dramatically upon parturition, the parametrial white adipose tissue forms decrease precipitously at the same time and are not expressed at all during the lactation period. In the liver of these animals, the only form of acetylcoenzyme A carboxylase mRNA that is expressed is the mL56 form; this form shows a modest decrease during pregnancy that is slowly reversed during lactation. These observations indicate that the changes in lipogenesis that occur during pregnancy and lactation are determined by the transcriptional activity of the acetylcoenzyme A carboxylase gene. In order to analyze the complex transcriptional activity of this gene in a meaning~
Received August 23, 1990. Accepted December 13, 1990. 1991 J Dairy Sci 7440134X!1
Abbreviation key: ACC = acetylcoenzyme A carboxylase, CoA = coenzyme A. INTRODUCTION
Parturition signals a profound change in fatty acid synthesis in three major lipogenic tissues in mammals: liver, adipose tissues, and mammary glands (3, 4). This change in fatty acid synthesis is brought about as a physiological necessity in providing a large amount of fatty acid and lipid in milk and is under the coordinated control of several hormones, notably, insulin, glucocorticoid, and prolactin (1, 16, 22, 23). The increased demand for fatty acids in milk is met by the epithelial cells in mammary glands in two ways: fatty acids are taken up by the epithelial cells from dietary sources, and fatty acids are synthesized de novo in the cells. The rate of fatty acid synthesis in the mammary glands at parhuition increases to a large extent. This increase is accompanied by an increased amount of two enzymes that are r e q d for fatty acid synthesis: acetyl-coenzyme A carboxylase (ACC) and fatty acid synthase (2). The ACC catalyzes the first committed step in fatty acid synthesis and is considered the rate-limiting step in the process. Thus, changes in ACC activity are closely correlated with the degree of lipogenic activity in mammalian tissues. In this brief review of the metabolism of ACC mRNA in lipogenic tissues during pregnancy and lactation, we will first present some background materials on ACC so that more recent developments in ACC mRNA metabolism can be better understood. 4013
4014
€‘ONCECAST-A
ET AL.
Position of Acetyl-coA Carboxylase (ACC) in Carbon Flow
Citrate 7;~ - - - -
(-)
----
-1
t
I
1
LCF+-CoA
Pyruvate Mito.
- - --- ---
t
Acetyl-coA-
Citlate
t
Acetyl-coA
OAA P i g m 1. Position of acetyl-coenzyme A carboxylase (ACC) in metabolic pathways. Metabolic hterconversion of glucose to the acctyl-cocnymcA (CoA) unit is illustrated. + = Positive effectors, - = negative effectors, LCFA = longchain fatty acid, OAA = oxaloacetic acid.
ACC and Its Regulatlon
Activity of ACC changes rapidly in response to the lipogenic needs of the cell, reflecting its rate-limiting nature in the lipogenic process (8). In addition, malonylcoenzyme A (CoA), the reaction product of ACC, serves as a key regulatory agent in the oxidation of fatty acids by controlling camitine acyl transferase at the mitochondrial transport step of fatty acid oxidation (15). This unique involvement of ACC in both fatty acid synthesis and degradation and the occurrence of fatty acids in multiple cellular components with different half-lives suggests that ACC activity has to be controlled very precisely. Additionally, the position of the ACC c a t a l y d reaction in metabolism suggests that complex and interlocking regulatory mechanisms for this enzyme must exist (Figure 1). Acetyl-coA carboxylase activity is acutely controlled by covalent modification through protein phosphorylation and dephosphorylation (6, 7). This covalent modification mechanism is modulated by various hormones, and the sensitivity of this mode of regulation is further complemented by allosteric control mechanisms involving various cellular metabolites, such as citrate and fatty acyl-CoA. As a result of sequencing of ACC cDNA (9, 21). we are now able to identify the functionally significant phosphorylation site as well as the effect of phosphorylation on the effect of cellular metabolites (7). On the basis of both regulatory Journal of Daiy Science Vol. 74, No. 11, 1991
mechanisms of covalent and allosteric control, we can explain most of the short-term regulatory aspects of ACC under various physiological conditions. However, the long-term regulatory aspects of ACC need to be studied in the light of ACC gene expression. Structure of the ACC Gene and the Yechanlsm for the Generatlon of Multiple Forms of mRNA
There is one copy of the ACC gene per haploid chromosome set (9); however, from this single gene copy, multiple forms of ACC are generated (10, 11). The mechanism &A by which these multiple forms of mRNA arise can be better un&rstood when the structure of the ACC gene is considered. A series of overlapping genomic clones was obtained and assembled into a coherent physical unit based on the information obtained from restriction mapping and DNA sequencing. Four clones (G-3, EM-5,EM-13, and EM-14)that span 52.5 kbp of genomic DNA were shown to contain the five exons that are differentially spliced to constitute the diverse species of ACC mRNA Figure 2). Figure 2 only shows the 5‘ end of ACC gene, and the ACC mRNA heterogeneity is confined to the 5’ untranslated regions of the mRNA (12). The cloned genomic.DNA spans 52.5 kbp and contains five exons, numbered from 1 to 5 in the 5’-+3’ direction of the gene as shown in Figure 2B. These exons contain 242, %, 61, 47, and 250 nucleotides. To determine exact
4015
SYMPOSIUM: REGuLAnoN OF GENE JXPRESSION ?(lib)
2P
'P
30
40
5P
Probe 3
r,
Figure 2. Isolation and restriction map of rat genomic DNA at the 5' region of the acetylcoenzyme A carboxylase (ACC) gme. A. Below the scale. the relative positiom of clones isolated Iiwm the genomic library an shown. The probes used to identify these clones are designated probes 1.2, and 3. B. Ibe restriction map is shown. The positions of five exons are indicated as bladr boxes 1.2, 3.4, and 5. The length of each exon is specified below the boxes. C. The restriction hjpents containing exons have been subclooed into an M13 vector. The sequencing strategy to determine the exon-iulron boundarks and the fhuking region of each exon are indicated with arrows depicting the direction and extent of dideoxyrmclcotide sequencing reactions. The positioas of exons are indicated by the batchtd boxes.
intmn-exon boundaries, the DNA fkagments containing exons were subcloned into M 1 3 vectors and sequenced as indicated by the arrows in Figure 2C. The five different ACC mRNA that we have characterized are illustrated in Figure 3 (10,11). Exon 5 contains the start of the ACC open reading frame, and exon 5 is present in a l l the forms of ACC mRNA characterized to date. Comparisons of the nucleotide sequences of the five different species of ACC mRNA that we have described previously with those of genomic DNA suggest the following mechanism for generating heterogeneity in the -A: two primary transcripts transcribed, using two merent promoters, one flanking exon 1 and the ather flanking exon 2; these primary transcripts are then differentially spliced to produce all the forms of ACC mRNA. This mechanism for mRNA generation predicts the presence of two promoters, and, indeed, the presence of two distinct promoters has been demonstrated (13). As shown in Figure 3 for the production of FL56 &A,
which is the main species in the lactating rat mammary gland, exons 2,4, and 5 join together. The FL63 species has the same splicing order as FL56, except that exon 3 is inserted between exons 2 and 4. A minor form detected in both the mammary gland and the liver results from the joining of exon 2 directly to exon 5. These three forms of ACC mRNA compose class 2 and are transcribed from promoter II. In contrast, pAU, the pxedominant form in adipose tissues and in the liver upon dietetic induction, directly splices exon 1 to exons 4 and 5, excluding exons 2 and 3 (Figure 3). Band b is generated by splicing exon 1 directly to exon 5 and skipping exons 2, 3, and 4. The latter two forms of ACC mRNA compose class 1 and are transcribed from promoter 1. Approaches to the Study of ACC Gene Actrvlty and ACC mANA Metabollsm
Prior to the demonstration of the heterogeneity of ACC mRNA, the metabolism of Jolnnal of Dairy Science Vol. 74. No. 11, 1991
4016
-
ET AL.
F'ONCE-CAST-A
EXON 1
2
5'uTR
3
4
mRNA NAME
FORMER NAME
AUG
ACC[1:4:5] mRNA
ACC[1:5] mRNA
pAU
band B
]
Class 1
ACC[2:4:5] mRNA
ACC[2:3:4:5] mRNA
FL63 ACC[2:51 mRNA
IJ
c1ass2
Hypothetical
Figure 3. Diagram of the genomic smcture of the 5' end of the acetyl-coenzyme A carboxylase (ACC) gene in rat. PI and PII represent the two carboxylase promoters. The initial five exons and the fmt four introns are indicated. pAU, FL63, and pLs6-type RNA with the various exons used to generate these RNA are also diagramed. Depending upon the use of promoters and the exons, different forms of mRNA arc classified into class 1 and 2 mRNA.
ACC mRNA has been investigated by quantitating the total amount of ACC mRNA through dot blot analysis or northern analysis (5, 17, 18, 19, 20). One example of such studies is shown in Figure 4. In this experiment, the total amount of ACC mRNA in mammary glands and livers was determined through a dot blot analysis using a cDNA probe. during pregnancy and lactation (18). Although this graph reflects the pattern of total mRNA metabolism, it cannot distinguish the true nature of ACC gene activity and the metabolic patterns of individual mRNA species. Clearly, a procedure to identify the activities of promoter I and promoter Il is required. One can identify different ACC mRNA species based on various procedures. However, the simplest procedure among the methods we have used to screen different promoter activities is to carry out the primer extension reaction. Figure 5 shows the primer extension products generated when the total RNA from various samples was primed with a 108-mer (263/ H2/Taq), which contained the most 5' 100 bases of ACC ORF and 8 bases of its flanking untranslated region, all of them e n d e d in exon 5 (13). Because all ACC mRNA contain Journal of Dairy Science Vol. 74, No. 11, 1991
exon 5 , the primer can extend on all ACC mRNA species in the presence of reverse transcriptase. The sizes of the extended products will vary depending on the length of the 5' untranslated region. This point is illustrated in Figure 6. Figure 5 shows the migration patterns of primer extension products and the products of the "linked" primer extension-S1 nuclease treatment reactions (lane s) using primer 263/ H2naq. Mammary gland RNA (lane 2) gave the 269-264 nt long FL56-derived primer extension products (open arrows), whereas liver RNA from animals that had been fasted and refed a fat-free diet generated, in addition to those of the E 5 6 type, products from the class 1 type of ACC mRNA black arrows denote the 415 base long primer extension product derived from pAU ACC mRNA (lane 3). The RNA prepared from livers of rats fed a normal diet produced only the bands derived from the FL56 type in barely detectable amounts (lane 5). However, the RNA prepared from adipose tissue of the same rats resulted in large amounts of products generated from the class 1 ACC mRNA and only trace amounts of FL56's bands (lane 6). The total RNA, which served as a negative control (lane l), and
SYMPOSIUM:REGULATION OF GENE EXPRESSION
60
4017
species of the class 2-type ACC M A , are expressed in all tissues examined, which include liver, adipose tissue, lung, pancreas, and mammary glands. Mammary gland tissue, which is one of the major lipogenic tissues, does not express the class 1-type mRNA at all, even under the stimulated synthesis of fatty acids during lactation (14).
PARTURITION
50 m
& 40 3
Analysis of ACC Gene Actlvlty During Pregnancy and Lactation
t-
a
2
30
k m a 4
1
20
IO
1
0
"
13 DAYS OF GESTATION
2
0
1
1
3
1
1
5
8
DAYS OF LACTATION
Figure 4. changes in acetyl-coenzyme A carboxylase M A in liver and mammary gland during gesfation and lactation. Female Wistar rats were bred, and the day of conception was determined by dropping of the vagmal plug. Tissues for analysis of preparturition carboxylase mRNA were taken 13 and 20 d afler conception. Total RNA from liver (0)and mammary gland (A) was isolated. and the relative amounts of carboxylase mRNA were determined by dot blot analysis using the cDNA insert of W 1 8 as the probe. Dots in the linear range of the film were quanMied by densitometry. Tssues from three animals were pooled for each point except for 8 d postpartum in which two animals were used.
starved liver RNA (lane 4), which has negligible amounts of ACC mRNA, produced no ACC-specific primer extension products. These results indicate that physiological states play an important role in controlling the tissuespecific levels of different ACC mRNA. In the presence of an excess of the primer over the amount of mRNA in the assay, the intensity of various bands reflects the amounts of different species of ACC mRNA in the total RNA preparation. Using this approach, we have been able to conclude that class 1-type mRNA, such as the pAU type, are specifically expressed in adipose tissues and livers that are stimulated to carry out active lipogenesis. In contrast, FL-type mRNA, which are the major
With this information on tissue-specific and physiologically regulated expression of two promoters, one can critically analyze the activity of the ACC gene. Figure 7 shows changes in the amount of different forms of ACC mRNA or changes in the promoter activity at the transition from pregnancy to the lactation period. This mode of analysis shows that ACC mRNA in mammary gland increases dramatically upon parturition and that all the ACC mRNA are the product of promoter II. However, parametrial adipose tissue primarily expresses the pAU-type ACC mRNA, which is the product of promoter I and very little FXtype ACC mRNA. This promoter I product decreases precipitously up to parturition and could not be detected during the lactation p i od. The only detectable ACC mRNA expressed in livers under these conditions is the FX56 type, which also decreases slightly up to the stage of parturition. This decrease is followed by a modest increase during the lactation period,The fatty acid synthetic activity in these three tissues has been observed to change in a manner similar to the kinetics in ACC mRNA changes presented in Figure 7 (2). Although there are some minor differences, overall metabolic patterns of ACC mRNA in mammary gland and liver are the same whether they are examined by the primer extension or the dot blot analysis, because both tissues primarily express promoter XI product, K56-type mRNA. However, the present studies clarify the promoter activities underlying these changes. Thus, the mechanism underlying repression of the ACC gene in fatty tissues during the lactation period involves the specific suppression of promoter I activity. Also, the same homeostatic signals that inactivate promoter I in adipose tissue must be producing the activation of promoter II in the mammary gland. Investigations needed for further clarifiJournal of Dairy Science Vol. 74, No. 11, 1991
4018
P O N C E C A S T m A ET AL.
S
s m
S
e
S
s m
S
0 -S27
V
I ” Y
e -309
k”
0 1) -201
I) -190 I) -180 -160
9 -141 O
M
$
1)
I
-122
- 110
Figure 5. Rimer exteapion annlysis and linked primer extension41 nuclease analysis (lane s) using primer 2 6 3 m Taq. The lmpg samph ofthc following RNAwae tested m cach experhnent: total RNA (lane 1): total RNA fromrat mamumy glands at d 6 of lactation (lane 2); total RNA fnnn li~ersof rats that before tbtir death nadawent either feeding with a standard laboratory animal diet (lane3 or 3 d of starvation (lane 4) or3 d of starvation followed by 2 d of refeeding a fat-free diet (lane 3); and total RNA from epididymal adipose tissue from a rat fed a standard chow (lane 6). Prominent primer e x t d o n products of tbe -6 type of acetyl^ A carboxylase mRNA (open arrows) and of the pAU type of acetyl-coenzyme A carbxylasc mRNA @lack arrows) are indicated. Lane m shows the molecular weight markers (32P-labclcd H p n fragmmls of pBR322). h e sizcs of which are indicated in nucleotides at the margin of the figure.
Journal of Dairy Science Vol. 74, No. 11, 1991
SYMPOSIUM: REGULATION OF GENE EXPRESSION
4019
M 1
4 1
5
Class 1-type ACC mRNA (pAU type)
-* (..................-....-*.. *
A
Probe 263/H2/Taq (108nt) Extended (4 15nt)
M 2
3 4 1
5
Class 2-type ACC mRNA (FL63type) (.... .................. *
Probe 263/H2/Taq (108nt) Extended (330-325nt)
Class 2-type ACC mRNA (FL56 type) (... ............... *
A
Probe 263/H2/Taq (108nt) Extended (269-264nt)
Figure 6. F‘rimer extension analysis of acetyl-coenzyme A carboxylase (ACC) mRNA. The predicted sizes of the primer ( 2 6 3 W f l a q ) extension products generated from three representaIive types of carboxylase mRNA species are diagramed.The unextended primer 263/H2/Taq is represented as an arrow with an asterisk at its 5’ end and is shown below the sequences in ACC mRNA to which it is complementary. The primer extension products generated from this mRNA are illustrated as broken arrows underneath the mRNA. The open arrowhead signals the pAU primer extension product; the solid arrowhead signals the FL56 primer extension product. The M indicates the position of the o p e are shown as boxes 1 to AUG of ACC open reading frame. Exons forming the 5’ untranslated region of the ACC &A 5 as in Figure 3.
cation of this complex response must be directed to the agents that affect the activity of both ACC gene promoters in such a tissuespecific manner. CONCLUSION
Gene activities are frequently measured by quantitating mRNA using the respective cDNA as a probe. There are many genes whose structural arrangements are very similar to the ACC gene. Using the ACC gene in this review, we have provided evidence indicating that gene activities cannot always be assessed by the total M A popuIation. The rate of transcrip
tion, stability of W A , and other parameters, such as RNA polymerase transition time on the gene, are factors frequently examined to assess the effect of agents on a particular gene activity. However, when multiple promoters control a gene, one has to devise a means of identifying each one of the products of the promoters involved to obtain a clear picture of gene activity. During the transition of the rat from pregnancy to lactation, dramatic shifts in fatty acid synthesis capacities occur in three major lipogenic tissues: liver, fatty tissues, and mammary glands. The ACC genes in each tissue are differentially affected to meet the demand for increased fatty acids during milk production. Journal of Dairy Science Vol. 74, No. 11, 1991
P O N C E C A S T m A ET AL.
NY.
A . ,
d Figure 7. Changes in the different acetyl-coenzyme A carboxylase (ACC) mRNA during gestation and lactation. Total RNA was prepared from the indicated tissues (MG = mam!muy gland; WAT = white adipose parametrial tissue and liver). Tissues were obtained from female Wistar rats at different stages of gestation and lactation as indicated. The number of animals for which tissues were pooled for the RNA preparation is shown in parentheses. Linked primer extension-Sl nuclease analysis was perfomed on F a l e n t amounts of each pool of RNA using the P-5' end-labeled primer 263/H2flaq as described (11). The extension products were separated in denaturing gels and Quantified by densitometry of the autoradiograms. Plotted in the relative densitometric units are the lwtls of FL56 ACC mRNA in MG (0)and in liver (0) and of pAU ACC mRNA in WAT (A).
ACKNOWLEDGMENTS
Studies from our laboratory have been supported by DK12865 from the National Institutes of Health. This is Journal Paper Number 12653 from the Agricultural Experimental Station, Purdue University. REFERENCES
1 Bauman, D. E., and C. L. Davis. 1974. Page 31 in Lactation: a comprehensive treatise. B. L. Larson and U. R. Smith, ed. Vol. 2. Academic Press, New York, Journal of Dairy Science Vol. 74. No. 11, 1991
2Clegg. R. A. 1988. Regulation of fatty acid up take and synthesis in mammary and adipose tissues: contrasting roles for CAMP. Cur. Top. Cell. Reg. 29:77. 3 Flint, D. J., P. A. Sinnett-Smith, R. A. Clegg. and R. G. Vernon. 1979. Role of insulin receptors in the c h q p g metabolism of adipose tissue dnring pregnancy and lactation in the rat. Biochem. J. 182421. 4Hamosh, M., T. R Clary, S. S. Chernick, and R 0. Scow. 1970. Lipoprotein lipase activity of adipose and mammary tissue and plasma triglyceride in pregnant and lactating rats. Biochim. Biophys. Acta 210473. 5 K a t m d a , A.. N. Iritani, H. Fukuda, Y. Matsumura, N. Nishimoto, T. Noguchi, and T. Tanaka. 1990. Effect of nutrients and hormones on transcriptional and post transcriptional~gulalingacetyl-coA carboxylase in rat liver. Eur. J. Biochem. 190:435. 6 Kim, K.-H. 1983. Regulation of acetyl-coA carboxylase. Cur. Top. Cell Reg. 22:143. 7 Kim, K.-H., F. Lbpe~-Ca~illa~, D. H. Bai, X.LUO.and M.E. Pape. 1989. Physiological sigmficance of covalent phosphorylation-dephosphorylationof acetyl-coA carboxylase in the regulation of long-chain fatty acids. Fed. Am. Soc. Exp. Biol. J. 3:2250. 8Lane. M. D., J. Moss,and S. E. Polakis. 1974. Acetyl coenzyme A carboxylase. Curr. Top. Cell. Reg. 8:139. 9Mpez-Casillas. F..D. H. Bai, X.Luo, I.-S. Kong, M. A. Hermodson, and K.-H. Kim. 1988. Structure of coding sequence and primary amino acid sequence of acetyl-coA carboxylase. Proc. Natl. Acad. Sci. USA 855784. 10Mpez-Casillas, F., and K.-H. Kim. 1989. Heterogeneity at the 5' end of rat acetyl-coA carboxylase mRNA: lipogenic conditions enhance synthesis of unique mRNA in liver. J. Biol. Chem. 2W7176. 11 Mpe~-Casillas, F., X. LUO,I.-S. KOW, a d K.-H. Kim. 1989. Characterization of different forms of rat mammary gland acetyl-coA carboxylase mRNA analysis of heterogeneity in the 5' end. Gene 83:311. 12 b o , X.,and K.-H. Kim. 1990. An enhancer element in the house-keeping promoter for acetyl-coA carboxylase gene. Nucleic Acids Res. 18:3249. 13 Luo, X.,K. Park, F. L6pez-Casillas, and K.-H. Kim. 1989. Structural features of acetyl-coA carboxylase gene: mechanism for the generation of mRNA with 5' end heterogeneity. Proc. NaU. Acad. Sci. USA 8 6 4042. 14MacLall J. C., and M. D. Lane. 1977. Changes in mammary gland acetyl-coA carboxylase associated with lactogenic differentiation. Biochem. J. 162:635. 15 McGany, J. D., G. F. Leatherman, and D. W. Foster. 1978. Carnitine palmitoyl transferase I; the site of inhibition of hepatic fatty acid oxidation by malonyl CoA. J. Biol. Chem. 253:4123. 16 Moore,J. H., and W. W. Chriatic. 1979. Lipid metabolism in the mammary gland of ruminant animals. Prog. Lipid Res. 17:347. 17Pape, M. E., and K.-H. Kim. 1988. Effect of tumor necrosis factor on acetyl-coA carboxylase gene expression and preadipocyte differentiation. Mol. End+ crinol. 2:395. M.E., aOd K.-H. Kim. 1989. Tran~~ripti~oal 18P*, regulation of acetyl-coA carboxylase gene expression
SYMwSlUht REGULATION OF GENE EXPRESSION
by tumor necrosis factor in 30A5 preadipytes. Mol. Cell Biol. 9:974. 19 Pap, M. E., P. Upez-CasiUas, and K.-H. Kim. 1988. Physiological regulation of acetyl-coA carboxylase gene expression: effect of diets, diabetes and lactation on acetyl-coA carboxylase -A. Arch. Biochem. Biophys. 267: 104. ZOTakai, T., Y. Saifo. K. Yamamoto, and T. Tanak. 1988. Developmental changes of the contents of acetyl-coA carboxylase mRNA in chicken liver.
4021
Arch, Biochem. Biophys. 266:313. 21 Takai, T., C. Yokoyama, K. Wada, and T. Tmbe. 1988. Primary struchue of chicken liver acetyl-coA carboxylase deduced form cDNA sequence. J. Biol. Chem. 263:2651. 22 Vernon, R. G. 1980. Lipid metabolism in the adipose tissue of rumiOant animals. hog. Lipid Res. 1923. 23 Williamson, D. H.1980. Integration of metabolism in tissues of the lactating rat. Ped. Eur. Biol. SOC. Lett. 117 (Suppl.):K93.
Journal of Dairy Science Vol. 74, No. 11, 1991
ful way, it is necessary to examine the metabolism of individual isoforms of acetyl-coenzyme A carboxylase mRNA. (Key words: acetyl-coenzyme A carboxylase, messenger ribonucleic acid, mammary glands)
ABSTRACT
In the rat, the acetylcoenzyme A carboxylase gene exists as a single copy per haploid chromosome set. However, multiple forms of acetyl-coenzyme A carboxylase mRNA exist, the relative abundance of which varies in a tissue-specific manner under different physiological conditions. In the mammary gland, the major acetyl-coenzyme A carboxylase mRNA species are of the class 2 type, which are products of promoter II. In parametrial white adipose tissue, the main form of species of acetyl-coenzyme A carboxylase is of the class 1 type, which are produced by promoter I. Pregnancy and lactation affect the amounts of these acetyl-coenzyme A carboxylase -A. Although the mammary gland acetylcoenzyme A carboxylase mRNA species increase dramatically upon parturition, the parametrial white adipose tissue forms decrease precipitously at the same time and are not expressed at all during the lactation period. In the liver of these animals, the only form of acetylcoenzyme A carboxylase mRNA that is expressed is the mL56 form; this form shows a modest decrease during pregnancy that is slowly reversed during lactation. These observations indicate that the changes in lipogenesis that occur during pregnancy and lactation are determined by the transcriptional activity of the acetylcoenzyme A carboxylase gene. In order to analyze the complex transcriptional activity of this gene in a meaning~
Received August 23, 1990. Accepted December 13, 1990. 1991 J Dairy Sci 7440134X!1
Abbreviation key: ACC = acetylcoenzyme A carboxylase, CoA = coenzyme A. INTRODUCTION
Parturition signals a profound change in fatty acid synthesis in three major lipogenic tissues in mammals: liver, adipose tissues, and mammary glands (3, 4). This change in fatty acid synthesis is brought about as a physiological necessity in providing a large amount of fatty acid and lipid in milk and is under the coordinated control of several hormones, notably, insulin, glucocorticoid, and prolactin (1, 16, 22, 23). The increased demand for fatty acids in milk is met by the epithelial cells in mammary glands in two ways: fatty acids are taken up by the epithelial cells from dietary sources, and fatty acids are synthesized de novo in the cells. The rate of fatty acid synthesis in the mammary glands at parhuition increases to a large extent. This increase is accompanied by an increased amount of two enzymes that are r e q d for fatty acid synthesis: acetyl-coenzyme A carboxylase (ACC) and fatty acid synthase (2). The ACC catalyzes the first committed step in fatty acid synthesis and is considered the rate-limiting step in the process. Thus, changes in ACC activity are closely correlated with the degree of lipogenic activity in mammalian tissues. In this brief review of the metabolism of ACC mRNA in lipogenic tissues during pregnancy and lactation, we will first present some background materials on ACC so that more recent developments in ACC mRNA metabolism can be better understood. 4013
4014
€‘ONCECAST-A
ET AL.
Position of Acetyl-coA Carboxylase (ACC) in Carbon Flow
Citrate 7;~ - - - -
(-)
----
-1
t
I
1
LCF+-CoA
Pyruvate Mito.
- - --- ---
t
Acetyl-coA-
Citlate
t
Acetyl-coA
OAA P i g m 1. Position of acetyl-coenzyme A carboxylase (ACC) in metabolic pathways. Metabolic hterconversion of glucose to the acctyl-cocnymcA (CoA) unit is illustrated. + = Positive effectors, - = negative effectors, LCFA = longchain fatty acid, OAA = oxaloacetic acid.
ACC and Its Regulatlon
Activity of ACC changes rapidly in response to the lipogenic needs of the cell, reflecting its rate-limiting nature in the lipogenic process (8). In addition, malonylcoenzyme A (CoA), the reaction product of ACC, serves as a key regulatory agent in the oxidation of fatty acids by controlling camitine acyl transferase at the mitochondrial transport step of fatty acid oxidation (15). This unique involvement of ACC in both fatty acid synthesis and degradation and the occurrence of fatty acids in multiple cellular components with different half-lives suggests that ACC activity has to be controlled very precisely. Additionally, the position of the ACC c a t a l y d reaction in metabolism suggests that complex and interlocking regulatory mechanisms for this enzyme must exist (Figure 1). Acetyl-coA carboxylase activity is acutely controlled by covalent modification through protein phosphorylation and dephosphorylation (6, 7). This covalent modification mechanism is modulated by various hormones, and the sensitivity of this mode of regulation is further complemented by allosteric control mechanisms involving various cellular metabolites, such as citrate and fatty acyl-CoA. As a result of sequencing of ACC cDNA (9, 21). we are now able to identify the functionally significant phosphorylation site as well as the effect of phosphorylation on the effect of cellular metabolites (7). On the basis of both regulatory Journal of Daiy Science Vol. 74, No. 11, 1991
mechanisms of covalent and allosteric control, we can explain most of the short-term regulatory aspects of ACC under various physiological conditions. However, the long-term regulatory aspects of ACC need to be studied in the light of ACC gene expression. Structure of the ACC Gene and the Yechanlsm for the Generatlon of Multiple Forms of mRNA
There is one copy of the ACC gene per haploid chromosome set (9); however, from this single gene copy, multiple forms of ACC are generated (10, 11). The mechanism &A by which these multiple forms of mRNA arise can be better un&rstood when the structure of the ACC gene is considered. A series of overlapping genomic clones was obtained and assembled into a coherent physical unit based on the information obtained from restriction mapping and DNA sequencing. Four clones (G-3, EM-5,EM-13, and EM-14)that span 52.5 kbp of genomic DNA were shown to contain the five exons that are differentially spliced to constitute the diverse species of ACC mRNA Figure 2). Figure 2 only shows the 5‘ end of ACC gene, and the ACC mRNA heterogeneity is confined to the 5’ untranslated regions of the mRNA (12). The cloned genomic.DNA spans 52.5 kbp and contains five exons, numbered from 1 to 5 in the 5’-+3’ direction of the gene as shown in Figure 2B. These exons contain 242, %, 61, 47, and 250 nucleotides. To determine exact
4015
SYMPOSIUM: REGuLAnoN OF GENE JXPRESSION ?(lib)
2P
'P
30
40
5P
Probe 3
r,
Figure 2. Isolation and restriction map of rat genomic DNA at the 5' region of the acetylcoenzyme A carboxylase (ACC) gme. A. Below the scale. the relative positiom of clones isolated Iiwm the genomic library an shown. The probes used to identify these clones are designated probes 1.2, and 3. B. Ibe restriction map is shown. The positions of five exons are indicated as bladr boxes 1.2, 3.4, and 5. The length of each exon is specified below the boxes. C. The restriction hjpents containing exons have been subclooed into an M13 vector. The sequencing strategy to determine the exon-iulron boundarks and the fhuking region of each exon are indicated with arrows depicting the direction and extent of dideoxyrmclcotide sequencing reactions. The positioas of exons are indicated by the batchtd boxes.
intmn-exon boundaries, the DNA fkagments containing exons were subcloned into M 1 3 vectors and sequenced as indicated by the arrows in Figure 2C. The five different ACC mRNA that we have characterized are illustrated in Figure 3 (10,11). Exon 5 contains the start of the ACC open reading frame, and exon 5 is present in a l l the forms of ACC mRNA characterized to date. Comparisons of the nucleotide sequences of the five different species of ACC mRNA that we have described previously with those of genomic DNA suggest the following mechanism for generating heterogeneity in the -A: two primary transcripts transcribed, using two merent promoters, one flanking exon 1 and the ather flanking exon 2; these primary transcripts are then differentially spliced to produce all the forms of ACC mRNA. This mechanism for mRNA generation predicts the presence of two promoters, and, indeed, the presence of two distinct promoters has been demonstrated (13). As shown in Figure 3 for the production of FL56 &A,
which is the main species in the lactating rat mammary gland, exons 2,4, and 5 join together. The FL63 species has the same splicing order as FL56, except that exon 3 is inserted between exons 2 and 4. A minor form detected in both the mammary gland and the liver results from the joining of exon 2 directly to exon 5. These three forms of ACC mRNA compose class 2 and are transcribed from promoter II. In contrast, pAU, the pxedominant form in adipose tissues and in the liver upon dietetic induction, directly splices exon 1 to exons 4 and 5, excluding exons 2 and 3 (Figure 3). Band b is generated by splicing exon 1 directly to exon 5 and skipping exons 2, 3, and 4. The latter two forms of ACC mRNA compose class 1 and are transcribed from promoter 1. Approaches to the Study of ACC Gene Actrvlty and ACC mANA Metabollsm
Prior to the demonstration of the heterogeneity of ACC mRNA, the metabolism of Jolnnal of Dairy Science Vol. 74. No. 11, 1991
4016
-
ET AL.
F'ONCE-CAST-A
EXON 1
2
5'uTR
3
4
mRNA NAME
FORMER NAME
AUG
ACC[1:4:5] mRNA
ACC[1:5] mRNA
pAU
band B
]
Class 1
ACC[2:4:5] mRNA
ACC[2:3:4:5] mRNA
FL63 ACC[2:51 mRNA
IJ
c1ass2
Hypothetical
Figure 3. Diagram of the genomic smcture of the 5' end of the acetyl-coenzyme A carboxylase (ACC) gene in rat. PI and PII represent the two carboxylase promoters. The initial five exons and the fmt four introns are indicated. pAU, FL63, and pLs6-type RNA with the various exons used to generate these RNA are also diagramed. Depending upon the use of promoters and the exons, different forms of mRNA arc classified into class 1 and 2 mRNA.
ACC mRNA has been investigated by quantitating the total amount of ACC mRNA through dot blot analysis or northern analysis (5, 17, 18, 19, 20). One example of such studies is shown in Figure 4. In this experiment, the total amount of ACC mRNA in mammary glands and livers was determined through a dot blot analysis using a cDNA probe. during pregnancy and lactation (18). Although this graph reflects the pattern of total mRNA metabolism, it cannot distinguish the true nature of ACC gene activity and the metabolic patterns of individual mRNA species. Clearly, a procedure to identify the activities of promoter I and promoter Il is required. One can identify different ACC mRNA species based on various procedures. However, the simplest procedure among the methods we have used to screen different promoter activities is to carry out the primer extension reaction. Figure 5 shows the primer extension products generated when the total RNA from various samples was primed with a 108-mer (263/ H2/Taq), which contained the most 5' 100 bases of ACC ORF and 8 bases of its flanking untranslated region, all of them e n d e d in exon 5 (13). Because all ACC mRNA contain Journal of Dairy Science Vol. 74, No. 11, 1991
exon 5 , the primer can extend on all ACC mRNA species in the presence of reverse transcriptase. The sizes of the extended products will vary depending on the length of the 5' untranslated region. This point is illustrated in Figure 6. Figure 5 shows the migration patterns of primer extension products and the products of the "linked" primer extension-S1 nuclease treatment reactions (lane s) using primer 263/ H2naq. Mammary gland RNA (lane 2) gave the 269-264 nt long FL56-derived primer extension products (open arrows), whereas liver RNA from animals that had been fasted and refed a fat-free diet generated, in addition to those of the E 5 6 type, products from the class 1 type of ACC mRNA black arrows denote the 415 base long primer extension product derived from pAU ACC mRNA (lane 3). The RNA prepared from livers of rats fed a normal diet produced only the bands derived from the FL56 type in barely detectable amounts (lane 5). However, the RNA prepared from adipose tissue of the same rats resulted in large amounts of products generated from the class 1 ACC mRNA and only trace amounts of FL56's bands (lane 6). The total RNA, which served as a negative control (lane l), and
SYMPOSIUM:REGULATION OF GENE EXPRESSION
60
4017
species of the class 2-type ACC M A , are expressed in all tissues examined, which include liver, adipose tissue, lung, pancreas, and mammary glands. Mammary gland tissue, which is one of the major lipogenic tissues, does not express the class 1-type mRNA at all, even under the stimulated synthesis of fatty acids during lactation (14).
PARTURITION
50 m
& 40 3
Analysis of ACC Gene Actlvlty During Pregnancy and Lactation
t-
a
2
30
k m a 4
1
20
IO
1
0
"
13 DAYS OF GESTATION
2
0
1
1
3
1
1
5
8
DAYS OF LACTATION
Figure 4. changes in acetyl-coenzyme A carboxylase M A in liver and mammary gland during gesfation and lactation. Female Wistar rats were bred, and the day of conception was determined by dropping of the vagmal plug. Tissues for analysis of preparturition carboxylase mRNA were taken 13 and 20 d afler conception. Total RNA from liver (0)and mammary gland (A) was isolated. and the relative amounts of carboxylase mRNA were determined by dot blot analysis using the cDNA insert of W 1 8 as the probe. Dots in the linear range of the film were quanMied by densitometry. Tssues from three animals were pooled for each point except for 8 d postpartum in which two animals were used.
starved liver RNA (lane 4), which has negligible amounts of ACC mRNA, produced no ACC-specific primer extension products. These results indicate that physiological states play an important role in controlling the tissuespecific levels of different ACC mRNA. In the presence of an excess of the primer over the amount of mRNA in the assay, the intensity of various bands reflects the amounts of different species of ACC mRNA in the total RNA preparation. Using this approach, we have been able to conclude that class 1-type mRNA, such as the pAU type, are specifically expressed in adipose tissues and livers that are stimulated to carry out active lipogenesis. In contrast, FL-type mRNA, which are the major
With this information on tissue-specific and physiologically regulated expression of two promoters, one can critically analyze the activity of the ACC gene. Figure 7 shows changes in the amount of different forms of ACC mRNA or changes in the promoter activity at the transition from pregnancy to the lactation period. This mode of analysis shows that ACC mRNA in mammary gland increases dramatically upon parturition and that all the ACC mRNA are the product of promoter II. However, parametrial adipose tissue primarily expresses the pAU-type ACC mRNA, which is the product of promoter I and very little FXtype ACC mRNA. This promoter I product decreases precipitously up to parturition and could not be detected during the lactation p i od. The only detectable ACC mRNA expressed in livers under these conditions is the FX56 type, which also decreases slightly up to the stage of parturition. This decrease is followed by a modest increase during the lactation period,The fatty acid synthetic activity in these three tissues has been observed to change in a manner similar to the kinetics in ACC mRNA changes presented in Figure 7 (2). Although there are some minor differences, overall metabolic patterns of ACC mRNA in mammary gland and liver are the same whether they are examined by the primer extension or the dot blot analysis, because both tissues primarily express promoter XI product, K56-type mRNA. However, the present studies clarify the promoter activities underlying these changes. Thus, the mechanism underlying repression of the ACC gene in fatty tissues during the lactation period involves the specific suppression of promoter I activity. Also, the same homeostatic signals that inactivate promoter I in adipose tissue must be producing the activation of promoter II in the mammary gland. Investigations needed for further clarifiJournal of Dairy Science Vol. 74, No. 11, 1991
4018
P O N C E C A S T m A ET AL.
S
s m
S
e
S
s m
S
0 -S27
V
I ” Y
e -309
k”
0 1) -201
I) -190 I) -180 -160
9 -141 O
M
$
1)
I
-122
- 110
Figure 5. Rimer exteapion annlysis and linked primer extension41 nuclease analysis (lane s) using primer 2 6 3 m Taq. The lmpg samph ofthc following RNAwae tested m cach experhnent: total RNA (lane 1): total RNA fromrat mamumy glands at d 6 of lactation (lane 2); total RNA fnnn li~ersof rats that before tbtir death nadawent either feeding with a standard laboratory animal diet (lane3 or 3 d of starvation (lane 4) or3 d of starvation followed by 2 d of refeeding a fat-free diet (lane 3); and total RNA from epididymal adipose tissue from a rat fed a standard chow (lane 6). Prominent primer e x t d o n products of tbe -6 type of acetyl^ A carboxylase mRNA (open arrows) and of the pAU type of acetyl-coenzyme A carbxylasc mRNA @lack arrows) are indicated. Lane m shows the molecular weight markers (32P-labclcd H p n fragmmls of pBR322). h e sizcs of which are indicated in nucleotides at the margin of the figure.
Journal of Dairy Science Vol. 74, No. 11, 1991
SYMPOSIUM: REGULATION OF GENE EXPRESSION
4019
M 1
4 1
5
Class 1-type ACC mRNA (pAU type)
-* (..................-....-*.. *
A
Probe 263/H2/Taq (108nt) Extended (4 15nt)
M 2
3 4 1
5
Class 2-type ACC mRNA (FL63type) (.... .................. *
Probe 263/H2/Taq (108nt) Extended (330-325nt)
Class 2-type ACC mRNA (FL56 type) (... ............... *
A
Probe 263/H2/Taq (108nt) Extended (269-264nt)
Figure 6. F‘rimer extension analysis of acetyl-coenzyme A carboxylase (ACC) mRNA. The predicted sizes of the primer ( 2 6 3 W f l a q ) extension products generated from three representaIive types of carboxylase mRNA species are diagramed.The unextended primer 263/H2/Taq is represented as an arrow with an asterisk at its 5’ end and is shown below the sequences in ACC mRNA to which it is complementary. The primer extension products generated from this mRNA are illustrated as broken arrows underneath the mRNA. The open arrowhead signals the pAU primer extension product; the solid arrowhead signals the FL56 primer extension product. The M indicates the position of the o p e are shown as boxes 1 to AUG of ACC open reading frame. Exons forming the 5’ untranslated region of the ACC &A 5 as in Figure 3.
cation of this complex response must be directed to the agents that affect the activity of both ACC gene promoters in such a tissuespecific manner. CONCLUSION
Gene activities are frequently measured by quantitating mRNA using the respective cDNA as a probe. There are many genes whose structural arrangements are very similar to the ACC gene. Using the ACC gene in this review, we have provided evidence indicating that gene activities cannot always be assessed by the total M A popuIation. The rate of transcrip
tion, stability of W A , and other parameters, such as RNA polymerase transition time on the gene, are factors frequently examined to assess the effect of agents on a particular gene activity. However, when multiple promoters control a gene, one has to devise a means of identifying each one of the products of the promoters involved to obtain a clear picture of gene activity. During the transition of the rat from pregnancy to lactation, dramatic shifts in fatty acid synthesis capacities occur in three major lipogenic tissues: liver, fatty tissues, and mammary glands. The ACC genes in each tissue are differentially affected to meet the demand for increased fatty acids during milk production. Journal of Dairy Science Vol. 74, No. 11, 1991
P O N C E C A S T m A ET AL.
NY.
A . ,
d Figure 7. Changes in the different acetyl-coenzyme A carboxylase (ACC) mRNA during gestation and lactation. Total RNA was prepared from the indicated tissues (MG = mam!muy gland; WAT = white adipose parametrial tissue and liver). Tissues were obtained from female Wistar rats at different stages of gestation and lactation as indicated. The number of animals for which tissues were pooled for the RNA preparation is shown in parentheses. Linked primer extension-Sl nuclease analysis was perfomed on F a l e n t amounts of each pool of RNA using the P-5' end-labeled primer 263/H2flaq as described (11). The extension products were separated in denaturing gels and Quantified by densitometry of the autoradiograms. Plotted in the relative densitometric units are the lwtls of FL56 ACC mRNA in MG (0)and in liver (0) and of pAU ACC mRNA in WAT (A).
ACKNOWLEDGMENTS
Studies from our laboratory have been supported by DK12865 from the National Institutes of Health. This is Journal Paper Number 12653 from the Agricultural Experimental Station, Purdue University. REFERENCES
1 Bauman, D. E., and C. L. Davis. 1974. Page 31 in Lactation: a comprehensive treatise. B. L. Larson and U. R. Smith, ed. Vol. 2. Academic Press, New York, Journal of Dairy Science Vol. 74. No. 11, 1991
2Clegg. R. A. 1988. Regulation of fatty acid up take and synthesis in mammary and adipose tissues: contrasting roles for CAMP. Cur. Top. Cell. Reg. 29:77. 3 Flint, D. J., P. A. Sinnett-Smith, R. A. Clegg. and R. G. Vernon. 1979. Role of insulin receptors in the c h q p g metabolism of adipose tissue dnring pregnancy and lactation in the rat. Biochem. J. 182421. 4Hamosh, M., T. R Clary, S. S. Chernick, and R 0. Scow. 1970. Lipoprotein lipase activity of adipose and mammary tissue and plasma triglyceride in pregnant and lactating rats. Biochim. Biophys. Acta 210473. 5 K a t m d a , A.. N. Iritani, H. Fukuda, Y. Matsumura, N. Nishimoto, T. Noguchi, and T. Tanaka. 1990. Effect of nutrients and hormones on transcriptional and post transcriptional~gulalingacetyl-coA carboxylase in rat liver. Eur. J. Biochem. 190:435. 6 Kim, K.-H. 1983. Regulation of acetyl-coA carboxylase. Cur. Top. Cell Reg. 22:143. 7 Kim, K.-H., F. Lbpe~-Ca~illa~, D. H. Bai, X.LUO.and M.E. Pape. 1989. Physiological sigmficance of covalent phosphorylation-dephosphorylationof acetyl-coA carboxylase in the regulation of long-chain fatty acids. Fed. Am. Soc. Exp. Biol. J. 3:2250. 8Lane. M. D., J. Moss,and S. E. Polakis. 1974. Acetyl coenzyme A carboxylase. Curr. Top. Cell. Reg. 8:139. 9Mpez-Casillas. F..D. H. Bai, X.Luo, I.-S. Kong, M. A. Hermodson, and K.-H. Kim. 1988. Structure of coding sequence and primary amino acid sequence of acetyl-coA carboxylase. Proc. Natl. Acad. Sci. USA 855784. 10Mpez-Casillas, F., and K.-H. Kim. 1989. Heterogeneity at the 5' end of rat acetyl-coA carboxylase mRNA: lipogenic conditions enhance synthesis of unique mRNA in liver. J. Biol. Chem. 2W7176. 11 Mpe~-Casillas, F., X. LUO,I.-S. KOW, a d K.-H. Kim. 1989. Characterization of different forms of rat mammary gland acetyl-coA carboxylase mRNA analysis of heterogeneity in the 5' end. Gene 83:311. 12 b o , X.,and K.-H. Kim. 1990. An enhancer element in the house-keeping promoter for acetyl-coA carboxylase gene. Nucleic Acids Res. 18:3249. 13 Luo, X.,K. Park, F. L6pez-Casillas, and K.-H. Kim. 1989. Structural features of acetyl-coA carboxylase gene: mechanism for the generation of mRNA with 5' end heterogeneity. Proc. NaU. Acad. Sci. USA 8 6 4042. 14MacLall J. C., and M. D. Lane. 1977. Changes in mammary gland acetyl-coA carboxylase associated with lactogenic differentiation. Biochem. J. 162:635. 15 McGany, J. D., G. F. Leatherman, and D. W. Foster. 1978. Carnitine palmitoyl transferase I; the site of inhibition of hepatic fatty acid oxidation by malonyl CoA. J. Biol. Chem. 253:4123. 16 Moore,J. H., and W. W. Chriatic. 1979. Lipid metabolism in the mammary gland of ruminant animals. Prog. Lipid Res. 17:347. 17Pape, M. E., and K.-H. Kim. 1988. Effect of tumor necrosis factor on acetyl-coA carboxylase gene expression and preadipocyte differentiation. Mol. End+ crinol. 2:395. M.E., aOd K.-H. Kim. 1989. Tran~~ripti~oal 18P*, regulation of acetyl-coA carboxylase gene expression
SYMwSlUht REGULATION OF GENE EXPRESSION
by tumor necrosis factor in 30A5 preadipytes. Mol. Cell Biol. 9:974. 19 Pap, M. E., P. Upez-CasiUas, and K.-H. Kim. 1988. Physiological regulation of acetyl-coA carboxylase gene expression: effect of diets, diabetes and lactation on acetyl-coA carboxylase -A. Arch. Biochem. Biophys. 267: 104. ZOTakai, T., Y. Saifo. K. Yamamoto, and T. Tanak. 1988. Developmental changes of the contents of acetyl-coA carboxylase mRNA in chicken liver.
4021
Arch, Biochem. Biophys. 266:313. 21 Takai, T., C. Yokoyama, K. Wada, and T. Tmbe. 1988. Primary struchue of chicken liver acetyl-coA carboxylase deduced form cDNA sequence. J. Biol. Chem. 263:2651. 22 Vernon, R. G. 1980. Lipid metabolism in the adipose tissue of rumiOant animals. hog. Lipid Res. 1923. 23 Williamson, D. H.1980. Integration of metabolism in tissues of the lactating rat. Ped. Eur. Biol. SOC. Lett. 117 (Suppl.):K93.
Journal of Dairy Science Vol. 74, No. 11, 1991
