| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
University Of Utah School of Medicine, Department of Pediatrics, Division of Neonatology, Salt Lake City, Utah, USA
1Correspondence: University of Utah School of Medicine, Department of Pediatrics, Division of Neonatology, P.O. Box 581289, Salt Lake City, UT 84158, USA. E-mail: robert.lane{at}hsc.utah.edu
ABSTRACT
Uteroplacental insufficiency leads to intrauterine growth retardation (IUGR) and adult onset insulin resistance in both humans and rats. IUGR rat liver is characterized by persistent changes in histone 3 lysine 9 and lysine 14 acetylation, which may induce postnatal changes in gene expression. We hypothesized that it would be possible to identify hepatic genes whose epigenetic characteristics and mRNA levels are altered due to IUGR using chromatin immunoprecipitation (ChIP) coupled with random primed differential display polymerase chain reaction (PCR). One of the isolated sequences identified contained exon 2 of the dual specificity phosphatase-5 gene (DUSP5). IUGR affected hepatic DUSP5 mRNA levels and exon 2 DNA methylation into adulthood in the rat. DUSP5 dephosphorylates Erk1 and Erk2 within the MAPK signaling cascade, which in turn affects serine 612 phosphorylation of insulin receptor substrate – 1 (p612 IRS-1). In adult rat liver, IUGR increased Erk1/Erk2 phosphorylation and p612 IRS-1 phosphorylation. Increased serine phosphorylation of hepatic IRS-1 may contribute to the insulin resistance that characterizes these animals. We conclude that intrauterine growth retardation induced by uteroplacental insufficiency 1) affects the hepatic epigenetic characteristics and mRNA of the DUSP-5 and 2) increases hepatic insulin receptor substrate-1 phosphorylation at serine 612 in adult rats.—Fu, Q., McKnight, R. A., Yu, X., Callaway, C. W., Lane, R. H. Growth retardation alters the epigenetic characteristics of hepatic dual specificity phosphatase 5.
Key Words: DNA methylation • chromatin • insulin receptor substrate
HYPERTENSIVE DISEASES IN pregnancy are common in both developed and developing countries (1
2
3)
. These disorders cause uteroplacental insufficiency and subsequent intrauterine growth retardation (IUGR) (4)
. IUGR in both humans and rats leads to multiple perinatal and postnatal morbidities, including insulin resistance (5
, 6)
. Though multiple molecular mechanisms are certainly involved, the link between the early event of uteroplacental insufficiency and an adult disease such as diabetes suggests epigenetic phenomena.
Epigenetic phenomena involve covalent modification of histones and DNA, which affect transcription factor complex access to DNA (7
8
9)
. Histone acetylation alters positioning of histone-DNA contacts, as well as histone affinity for DNA. Histone acetylation typically inversely correlates with DNA methylation (7
, 10
, 11)
. Both of these modifications are relatively stable, though not necessarily permanent. Postnatal alterations of hepatic histone 3 lysine 9 (H3/K9) and histone 3 lysine 14 (H3/K14) acetylation characterize a rat model of uteroplacental insufficiency, IUGR, and subsequent adult onset diabetes (12)
. These changes in histone acetylation potentially alter mRNA levels of specific genes over the lifetime of the animal and affect phenotype. We therefore hypothesized that chromatin immunoprecipitation (ChIP) with antibodies against acetylated H3/K9 and H3/K14 coupled with differential display PCR (DD PCR) would identify genes whose mRNA levels are persistently altered by IUGR.
To test this hypothesis, we initially compared day of life 0 (DOL 0) chromatin using these methods from control (CON) and IUGR liver. Our model of bilateral uterine artery ligation and subsequent IUGR has been well characterized by multiple groups (13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32)
. In brief, IUGR pups in this model are 
25% smaller than CON pups. IUGR pups are predisposed to develop insulin resistance early in life and overt diabetes relatively late in life (30
, 31)
. Gender-specific differences become evident by DOL21, with males appearing to be more severely affected (12)
.
In our comparison of DOL 0 CON and IUGR chromatin, the ChIP and DD-polymerase chain reaction technique identified a section of the dual specificity phosphatase 5 (DUSP5; hVH-3/B23) gene as a DNA region whose epigenetic characteristics have been altered. We then set out to determine whether 1) the identified region of hepatic DUSP5 DNA is differentially methylated in juvenile and adult rats; 2) hepatic DUSP5 mRNA levels differ between CON and IUGR rats; and 3) reported targets of DUSP5 are differentially phosphorylated in IUGR liver. Targets of DUSP5 include the extracellular related kinases 1 and 2 (Erk1/Erk2), which are components of the mitogen-activated protein kinase (MAPK) pathway (33
, 34)
. A downstream target of Erk1/Erk2 relevant to this model of IUGR and postnatal insulin resistance is serine 612 of insulin receptor substrate 1 (p612-IRS) (35
, 36)
.
MATERIALS AND METHODS
Animals
All procedures were approved by the University of Utah Animal Care committee and are in accordance with the 3'-amino propyltriethoxy silane (APS Guiding Principles) (37)
. Surgical procedures have been described previously (12
, 38
39
40
41
42
43)
. On day 19 of gestation, maternal rats were anesthetized with intraperitoneal (i.p.) xylazine (8 mg/kg) and ketamine (40 mg/ml), and both uterine arteries were ligated (IUGR) (n=12 litters). Term gestation in the rat is 21.5 days. Sham surgery was performed on control animals who underwent identical anesthetic and surgical procedures except for the uterine artery ligation (n=12 litters). The DOL 0 pups were delivered by caesarian section (n=6 litters Con and IUGR, respectively) at term, 2 days after the bilateral uterine artery ligation. DOL21 animals were separated from their dams for 4 h, anesthetized, and killed (n=6 litters Con and IUGR, respectively). Animals were studied at this age because they have not yet developed overt insulin resistance or dyslipidemia, which may confound our findings (20
, 30
, 31)
. Animals were raised to DOL120, anesthetized, and killed (n=6 litters Con and IUGR, respectively). At the three ages, the liver was quickly removed, flashed frozen in liquid nitrogen, and stored in –80°C.
ChIP assay
ChIP with anti-acetyl H3/K9 and anti-acetyl H3/K14 was performed as described earlier (Cell Signaling Technologies, Beverly, MA, USA) (12)
.
Differential display (DD) and polymerase chain reaction (PCR)
A total of five 16mer CG-rich arbitrary primers were used for differential display (Table 1
). Four samples of ChIP DNA from each control and IUGR group were used for DD-PCR. A DNA sample (50 ng) was used for each PCR reaction. The PCR used 1 x PCR Gold Buffer (Applied Biosystems, Foster City, CA, USA), 2.5 mM MgCl2, 2.5 µCi of [
-33P] dCTP (Perkin Elmer Life Sciences, Boston, MA, USA), 500 nM of primers, and 1.25 U of Ampli Taq Gold (Applied Biosystems). Initial denaturation was performed at 95°C for 10 min, followed by 40 cycles of 94°C for 30 s, 44°C for 60 s and 72°C for 90 s for primers 1–3 (Table 1)
; an annealing temperature of 50°C for primers 4 and 5.5 µl of PCR product was added to 1 µl of loading dye (95% formamide, 20 mM EDTA, 0.05% each bromphenol blue and xylene cyanol). After the samples were heated to 94°C for 3 min and immediately cooled on ice, 4 µl of each was loaded onto a 0.4 mm-thick, 34 cm 5% polyacrylamide sequencing gel under denaturing conditions (7 M urea) and migrated in 1 x Tris borate-EDTA (TBE) for 5 h at 1500V, 60 W (to keep the temperature of the gel at 50°C). After being dried, the gels were exposed to X-ray films at –80°C.
|
Isolation and sequencing of DNA fragments
Candidate bands that were differentially represented between control and IUGR, as shown in Fig. 1
, were excised from dried polyacrylamide gels and placed in a microcentrifuge tube containing 50 µl of sterile H2O. The microcentrifuge tube was then heated to 80°C for 10 min and vortexed to facilitate extraction of DNA. The eluate (2 µl) was then used in PCR with the same primers used in the original DD-PCR to generate sufficient amounts of template for plasmid cloning and sequencing. PCR ingredients and amplification parameters were the same as described above. PCR products were cloned into pCR 2.1 using the TOPO TA Cloning Kit (Invitrogen, San Diego, CA, USA). Multiple colonies derived from the cloning procedure were sequenced according to the manufacturer’s instructions for double-stranded plasmid DNA using the BigDye® Terminator v3.1 Cycle Sequencing kit (Applied Biosystems). The resulting nucleotide sequences were then compared to the GenBank sequences, using the Blast program (44)
.
|
Bisulfite modification
Genomic DNA was treated with sodium bisulfite to convert unmethylated cytosines to uracil, which then converts to thymine, leaving 5-methylcytosines unchanged using CpGenome DNA Modification Kit (Chemicon International, Temecula, CA, USA) following the manufacturer’s instructions (45)
. Regions of DNA within exon 2 of DUSP5 that contained CpG sites were amplified. The PCR primers are listed in Table 1
. PCR conditions were 95°C for 10 min, followed by 94°C for 30 s, 53°C for 1 min, 72°C for 1 min, 35 cycles. The PCR products were cloned into pCR2.1 as described above.
RNA isolation
Total RNA was extracted from DOL 0, DOL 21, and DOL 120 liver using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA), treated with DNase I (Qiagen), then quantified. Gel electrophoresis confirmed the integrity of the samples.
Real-time RT-polymerase chain reaction (RT-PCR)
cDNA was synthesized using the same method as described previously. mRNA levels of DUSP5 were measured using real-time RT-PCR at DOL 0, DOL 21, and DOL 20 rat liver (12)
. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as internal control. Primer and probe sequences are shown in Table 1
.
Western blot
Cell lysates were used to do Erk1/2, phospho-Erk1/2 MAP-kinase, IRS-1, and phospho-IRS-1 (Ser-612) Western blots. Cell lysates were prepared as described previously (12)
. For Erk1/2 and phospho-Erk1/2, 40 µg of protein from DOL 0, DOL 21, and DOL 120 liver cell lysates was separated by 10% SDS-PAGE gels and transferred to PVDF membrane (Millipore, Billerica, MA, USA). After transfer, membranes were blocked with 3% nonfat milk in Tris-buffered saline (TBS), then incubated in either 1:1000 dilution of total Erk1/2 antibody (Ab) or 1:200 dilution of phospho-Erk1/2 Ab (Cell Signaling Technologies) at 4°C overnight with agitation, washed three times, and incubated with horseradish peroxidase conjugated secondary Ab (rabbit). Signals were detected using enhanced chemiluminescence (ECL) performed according to the manufacturer’s instructions (Amersham, Buckinghamshire, UK). For IRS-1, 200 µg of protein from DOL 120 liver cell lysates was separated on Criterion XT Precast Gels, 4–12%, Bis-Tris (Bio-Rad, Hercules, CA, USA) and transferred to nitrocellulose membrane (Schleicher & Schuell, Dassel, Germany). Membranes were blocked with 1% nonfat milk in TBS, incubated in a 1:100 dilution of either phospho-IRS-1 (Ser-612 Ab) or IRS-1 Ab (Cell Signaling), then processed as above.
Statistics
All data presented are expressed as mean percent of control ± SE. ANOVA was used for real-time RT-PCR; when this was significant, the Fisher least significant difference (LSD) post hoc test was computed. The Student’s 2-tailed t test determined statistical significance for Western blot and DNA methylation. A P value of <0.05 was considered statistically significant.
RESULTS
Hepatic ChIP and differential display
Ab to either acetyl-H3/K14 and acetyl-H3/K9 was used to precipitate control and IUGR hepatic chromatin from pups at DOL 0. ChIP DNA was amplified using five CpG-rich arbitrary primers for DD-PCR and separated by urea PAGE (Fig. 1A
). From the DNA enriched by histone acetyl-H3/K14 ChIP, 25 bands were isolated, cloned, sequenced, and analyzed by Blast. The sequence of six clones was located within either an intron or a 5' untranslated region. Two sequences were homologous to intronic ESTs, and four sequences were repetitive element sequences. Seven sequences were designated "intergenic" by the Blast program, and six sequences did not match any previously characterized DNA.
From the DNA enriched by histone acetyl-H3/K9 ChIP, 11 bands were isolated, cloned, sequenced, and analyzed by Blast. Sequences of three clones were located within a known gene. Four sequences were repetitive elements, and three sequences were within DNA regions designated as intergenic. One sequence did not match any previously characterized DNA. One of the clones isolated from the histone H3/K9 ChIP was 611 bp in size and contained parts of exon 2 and intron 2 of DUSP5, a MAP kinase phosphatase (accession number NM_133578) (Fig. 1B
). Exon 2 of DUSP5 contains 6 CpG sites.
DNA methylation patterns in the DUSP5 exon 2
This 611 bp DUSP5 clone included part of exon 2 (Fig. 1B
). The full length of exon 2 is 149 bp and includes six CpG sites. Bisulfite modification was used to evaluate the methylation status of the five downstream sites in liver from DOL 0, 21, and 120. Primer design required that 1 CpG site be included in the 5' primer, and therefore was not evaluated.
Uteroplacental insufficiency and IUGR significantly decreased DOL 0 CpG methylation within this region. In DOL 0 male and female animals, 56 ± 6%** CpG DUSP5 exon 2 methylation characterized the IUGR livers, whereas 74 ± 5% CpG DUSP5 exon methylation characterized the Con livers (Fig. 2
A) (**P<0.01). No significant difference was noted at this point between male and female pups, which were equally represented. Furthermore, three specific CpG sites were relatively hypomethylated in the DOL 0 liver DNA (male and female): site 2, IUGR 67 ± 10%* vs. Con 88 ± 6%; site 5, IUGR 34 ± 9%* vs. Con 56 ± 6%; and site 6, IUGR 43 ± 9%** vs. Con 78 ± 11% (*P<0.05; **P<0.01) (Fig. 2A
).
|
Similarly, IUGR decreased the overall percent of CpG methylation of exon 2 in the DOL 21 male rats to 61 ± 7%**, whereas 83 ± 4% of the sites were methylated in DNA from the day of life 21 control male liver DNA (**P<0.01) (Fig. 2B
). Two specific CpG sites were relatively hypomethylated in liver from DOL 21 male rats: site 4, IUGR 55 ± 6%* vs. Con 78 ± 5%; site 6, IUGR 50 ± 19%* vs. Con 83 ± 10% (*P<0.05) (Fig. 2B
). DNA from female livers at DOL 21 did not demonstrate a significant difference in CpG methylation at any of the five sites, though a trend existed toward hypomethylation (Fig. 2C
).
Hepatic DNA from Con and IUGR rats at DOL 120 also revealed gender-specific differences. IUGR decreased overall methylation of the five CpG sites in male liver DNA to 84.3 ± 2.7% **; whereas 92.8 ± 4.5% of these sites were methylated in Con male liver DNA (**P<0.01) (Fig. 3
A). Among the five sites, DNA methylation at CpG site 3 was significantly decreased to 55.3 ± 9.9%* in IUGR male liver DNA (*P<0.05). Con DNA methylation in day of life 120 liver DNA at this same site was 91.7 ± 8.3%. Although no significant differences in overall CpG methylation of these sites existed between Con and IUGR DOL 120 hepatic female DNA, CpG site 4 was relatively hypermethylated at 78 ± 1.5%* in IUGR DOL 120 hepatic female DNA, in contrast to 71 ± 6% in day of life 120 control hepatic female DNA (*P<0.05) (Fig. 3B
).
|
Hepatic DUSP5 mRNA levels
Uteroplacental insufficiency decreased hepatic DUSP5 mRNA levels in both male and female IUGR pups at day 0 of life 0 (74±5% of Con value; P<0.01) (Fig. 4
A). Similarly, DUSP5 mRNA levels continued to be decreased in livers from DOL 21 male (66±5% of Con value; P<0.01) (Fig. 4B
) and female rats (76±8% of Con value; P<0.01) (Fig. 4C
), as well as DOL 120 IUGR male rats (89±5% of Con value; P<0.05) (Fig. 4D
). In contrast, hepatic DUSP5 mRNA levels increased in the livers of DOL 120 IUGR female rats (160±11% of control values; P<0.01) (Fig. 4E
).
|
Hepatic Erk 1 and Erk 2 levels
Uteroplacental insufficiency increased the phosphorylated forms of Erk1 and Erk2 in both male and female IUGR livers at DOL 0 (pErk1 150±10%** vs. Con; pErk2 130±7.2%** vs. Con;**P<0.01), without affecting either Erk1 or Erk2 total levels (Fig. 5
A). At DOL 21 IUGR similarly increased pErk1 levels in both male and female livers, as well as pErk2 levels in female livers (male pErk1 144±11%* vs. Con; female pErk1 386±48* vs. Con; female pErk2 235±10%** vs. Con: *P<0.05, **P<0.01) (Fig. 5B
). Again, neither Erk1 nor Erk2 total levels were affected in the livers of either gender at day 21 of life.
|
In the DOL 120 livers, pErk1 and pErk2 levels were significantly increased in male IUGR livers without affecting the total levels of these proteins (male IUGR pErk1 665±47%** vs. Con; male IUGR pErk2 378±56%* vs. Con: *P<0.05, **P<0.01) (Fig. 5C
). The resultant pErk1/Erk1 and pErk2/Erk2 ratio was 5.68 and 3.05, respectively. In contrast, total levels of Erk1 and Erk2, as well as pErk1 and pEk2 were significantly increased in livers of the IUGR females when compared to control livers (female IUGR Erk1 171±21%*; female IUGR pErk1 177±26*; female Erk2 224±23%*; female pErk2 220±25%*, *P<0.05) (Fig. 5C
). As a result, the ratio of the pErk1/Erl1 and pErk2/Erk2 in the female DOL 120 livers was 1.04 and 0.98, respectively.
DOL 120 hepatic p612 IRS-1 protein levels
In association with the increased levels of pErk1 and pErk2, IUGR significantly increased phosphorylation of IRS-1 at serine 612 in male and female liver to 172.6 ± 14%** and 149.6 ± 13%* of control values, respectively (*P<0.05, **P<0.01) (Fig. 6
A). Total IRS-1 protein levels were significantly decreased in the IUGR male liver and unaffected in the IUGR female livers (Fig. 6B
).
|
DISCUSSION
Growing epidemiological evidence indicates that poor in utero growth predicts adult morbidities, such as insulin resistance and obesity (46
, 47)
. "Baker’s fetal origins of adult disease hypothesis" nicely conceptualizes this relationship between poor in utero growth and adult morbidities (5
, 48
, 49)
. Our interpretation of this hypothesis is that fetal adaptation to a deprived intrauterine environment leads to postnatal changes in cellular biology and predisposes to an altered adult phenotype. Our goal, as well the goal of many other investigators, has been to identify a specific gene that couples the adaptation to the adult phenotype. Many groups focus on a specific effector molecule that directly affects glucose (Glc) or fat metabolism and subsequently suggest that the affected process is essentially on or off, which directly causes the adult morbidity. The findings in this study suggest a subtle variation in the theme in that IUGR potentially shifts the sensitivity of Erk signaling by affecting DUSP5 epigenetics and subsequent mRNA expression (Fig. 7). This shift in a signaling pathway upstream of several key cellular processes may thereby initiate a relatively coordinated adaptation. A component of this adaptation may be DUSP5. The novel findings of our study include 1) the identification of DUSP5 as an hepatic gene whose histone acetylation is altered by uteroplacental insufficiency and IUGR; 2) the identification of specific hepatic DUSP5 CpG sites whose methylation is altered into adulthood; and 3) the determination of hepatic Erk1/Erk2 and p612 IRS-1 phosphorylation status in a model of IUGR and adult onset diabetes.
Our ChIP analysis with acetyl-H3/K9 Ab identified an exon2/intron2 of DUSP5 as a segment of DNA whose position within the nucleosome complex is altered by IUGR. Considering previous studies, this is not surprising. Kurdistani et al. demonstrated that acetylation of H3/K9 associated with intergenic regions of genes (50)
. Similarly, a large-scale study of histone modification patterns in human and mouse cells for chromosomes 21 and 22 found that 58% of the acetylated H3/K9 and H3/K14 sites associated with the 5' of genes (51)
.
Furthermore, our finding of decreased CpG methylation at DOL0 0 and DOL 21 within exon2 of DUSP5, coupled with decreased DUSP5 mRNA levels, is a pattern that is well established. In his review, "The DNA methylation paradox," Peter Jones discusses the observation that whereas CpG methylation within promoters blocks transcription, downstream CpG methylation associates with increased transcription (52)
. This may occur because transcription facilitates CpG methylation in some instances (52)
. It is therefore intriguing that DUSP5 is a direct transcriptional target of the transcription factor p53, whose protein levels are decreased in DOL0 IUGR liver (40)
.
We do appreciate that a "perfect" direct relationship does not exist between DUSP5 exon 2 CpG methylation and DUSP5 mRNA levels at DOL 120, particularly when considering the relatively small differences in methylation in relation to the large increase in DOL 120 female DUSP5 mRNA levels. However, it would be naive to expect one aspect of chromatin structure to dominate the regulation of DUSP5 transcription when histone modifications and transcription factor complex composition are also likely to contribute. This allows for CpG methylation in specific regions to either amplify or dampen the signals for transcription, a "fine-tuning" if you will. Again, this represents a subtle variation on the theme of strict "on or off," the latter of which would be potentially dangerous, considering that environmental conditions are often outside the control of the individual. Furthermore, when epigenetics is being used to adapt to environmental clues such as nutrition, we expect that the histone code of epigenetically regulated genes will vary as one moves from nucleosome to nucleosome to endow chromatin structure with a relatively fine control of transcription. As a result, to fully understand transcriptional regulation of an entire gene via epigenetics, one would need to map the histone code from promoter regions to 3' untranslated regions. In contrast, the epigenetic phenomena surrounding events such as early embryonic imprinting appear to be triggered by more reliable clues, such as the gender of DNA’s parenteral origin. As a result, it makes sense that these events involve promoters that have a more definitive effect on transcription. We speculate the changes we observe in DNA methylation play a partial role in the regulation of DUSP5 transcription and, what is more important, serve as a marker for DUSP5 epigenetic regulation.
Ishibashi et al. initially cloned the human DUSP5 cDNA from a B5/589 human mammary epithelial cell cDNA library (34)
. The rat DUSP5 cDNA was identified via a screen of candidate plasticity-related genes from hippocampal neurons stimulated with the glutamate analog kainite (53)
. Liver and placenta express the highest amounts of DUSP5 mRNA, though it is also found in brain, heart, and kidney (34)
. DUSP5 belongs to the family of VH-1 like phosphatases based on the presence of the canonical protein-tyrosine phosphatase motif [HCXAGXXR(S/T)] (54)
. DUSP differs from other members of this family in that the transcript half-life is relatively stable after induction (54)
. In cell culture, heat shock and growth factors (e.g., epidermal growth factor and insulin) induce DUSP5 expression (34
, 54)
. Like other VH-1 phosphatases, DUSP5 exhibits catalytic activity toward phosphotyrosine and phosphothreonine residues. Early reports demonstrated that DUSP5 exerted significant activity against Erk-1 (33
, 54)
. A recent report by Mandl et al. found that DUSP5 binds specifically to both Erk-1 and Erk-2 kinases, which is in contrast to DUSP1, DUSP2, and DUSP4 in that DUSP5 does not inactivate Jun N-terminal protein kinase or p38 MAP kinase. From their seminal studies, Mandl et al. concluded that DUSP5 is a distinct inducible, nuclear Erk-specific phosphatase.
The decrease in DUSP5 expression and increase Erk phosphorylation may be an adaptive response to the deprived intrauterine milieu that characterizes the IUGR fetus. IUGR in our model increases oxidative stress in the DOL 0 rat liver. Erks mediate both cytoplasmic and nuclear survival signals in primary hepatocyte cultures (33
, 55)
. When Erk activation is prevented, cell survival is reduced (56
, 57)
. Damage to cellular membranes and protein interferes with signaling pathways and intracellular transport. Therefore, anticipatorily increasing Erk phosphorylation by decreasing DUSP5 mRNA may increase the odds toward cell survival by shifting the threshold toward increased MAPK signaling. We do recognize that the regulation of Erk phosphorylation is complex and that the activities of other systems confound any research done on in vivo tissue. For example, 17-ß-estradiol also protects injured hepatocytes by increasing Erk phosphorylation (58)
.
Similarly, we recognize that the regulation of serine 612 IRS-1 phosphorylation is likely to be complex. However, the following three things have been demonstrated in cell culture: 1) conversion of IRS-1 serine 612 to an alanine prevents phosphorylation mediated inhibition of IRS-1 signaling; 2) phosphorylation of IRS-1 serine 612 requires activation of the MAPK pathway and involves Erk1/Erk2; and 3) Erk1/Erk2 associate with IRS-1 on activation (59
60
61)
. Phosphorylation of serine 612 inhibits IRS-1 signaling by inhibiting IRS-1 tyrosine phosphorylation by JAK1 and the subsequent association with PI 3-kinase (36
, 62)
. As a result, phosphorylation of IRS-1 serine 612 may play counter regulatory role for pathways outside the insulin signaling system and, in effect, act as an IRS-1 desensitization mechanism.
Though the focus of this manuscript is the differences in DUSP5 epigenetic characteristics in Con vs. IUGR rats, our findings of increased IRS-1 serine phosphorylation are intriguing, considering that uteroplacental insufficiency and subsequent IUGR is a model of adult onset insulin resistance and diabetes. IRS-1 Recent knock down experiments demonstrate that decreased IRS-1 leads to hepatic insulin resistance, and the IRS-1 is closely linked to Glc homeostasis (63
, 64)
. Using in vivo rodent models of endoplasmic reticulum stress, Özcan et al. demonstrated that hepatic IRS-1 serine phosphorylation at serine 307 leads to insulin resistance (65)
. In addition, rat pups that experience maternal diabetes in utero show significantly increased IRS-1 serine 307 and 636 phosphorylation in skeletal muscle (35)
. Neither study measured phosphorylation of hepatic IRS-1 serine 612, however, despite the previous cell culture data discussed above. Thus, the present studies are among the first to demonstrate increased hepatic IRS-1 serine 612 phosphorylation in an animal of an early perinatal insult and late onset insulin resistance.
Though IRS-1 serine phosphorylation was increased in both male and female livers relative to controls, DUSP5 methylation and mRNA levels are obviously influenced by gender. The mechanism through which gender influences the hepatic epigenetics and subsequent gene expression is unknown. However, studies utilizing rats rendered IUGR through uteroplacental insufficiency consistently demonstrate gender-specific differences in growth, serum triglycerides, and gene expression (20
, 22
, 25)
. In liver, IUGR rats of both sexes showed decreased liver expression of CPTI at day 21 of postnatal life, but only male IUGR rats showed decreased expression of CPTI at day 120 of postnatal life relative the controls (20)
. Furthermore, histone H3 is hyperacetylated in male IUGR rat livers at DOL 21, whereas the livers of female IUGR rats are relatively hypoacetylated at histone H3 (12)
. We speculate that these differences between tissues explain, at least in part, why the general IUGR phenotype is similar between the genders, but many specific characteristics appear to differ.
Caution is always necessary when attempting to apply data from a rat model to human pathophysiology. Fetal and juvenile rats are physiologically immature relative to the human, and the insult imposed on the fetal rat in this model of uteroplacental insufficiency is severe. In contrast, the effect of uteroplacental insufficiency experienced by humans ranges across a continuum. Furthermore, we acknowledge that early postnatal nutrition is important. We previously characterized breast milk from our Con and IUGR dams after 21 days of supporting their respective pups, and no significant differences in K CAL, protein, fat, or zinc were noted between the two groups (41)
. Differences may have existed earlier, and other components of the breast milk that were not assessed earlier could be important. With these data in hand, we chose not to cross foster in this study, because this rarely occurs in the human situation and is unlikely to be recommended secondary to infectious issues.
In summary, we found that uteroplacental insufficiency and subsequent IUGR cause perinatal and postnatal changes in the epigenetic characteristics of the DUSP5 gene, and that these changes are associated with predictable alterations in DUSP5 mRNA levels, Erk1/Erk2 phosphorylation, and IRS-1 phosphorylation. We speculate that the changes are an adaptation to the prenatal insult that minimize perinatal hepatic cell death, and become a maladaptation latter in life that contributes to postnatal insulin resistance.
ACKNOWLEDGMENTS
We want to thank J. R. Milley and the Division of Neonatology at the University of Utah for its gracious and generous support. This research was funded by HD41075 (R.H.L.) and the March of Dimes (R.H.L.).
Received for publication April 3, 2006. Accepted for publication May 15, 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  K. M Aagaard-Tillery, K. Grove, J. Bishop, X. Ke, Q. Fu, R. McKnight, and R. H Lane Developmental origins of disease and determinants of chromatin structure: maternal diet modifies the primate fetal epigenome J. Mol. Endocrinol., August 1, 2008; 41(2): 91 - 102. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
