| 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
SPECIFIC AIMS
Uteroplacental insufficiency and intrauterine growth retardation (IUGR) alters hepatic histone acetylation at histone 3 lysine 9 (AcH3/K9) and lysine 14 (AcH3/K14) in rats. We hypothesized that these alterations could be used to identify genes whose DNA methylation and mRNA levels are affected in IUGR rat liver.
PRINCIPAL FINDINGS
1. Hepatic chromatin immunoprecipitation (ChIP) and differential display polymerase chain reaction (PCR) identified dual specificity phosphatase 5 (DUSP5) as a gene affected by IUGR
Antibody (Ab) to either AcH3/K14 or AcH3/K9 was used to precipitate control and IUGR hepatic chromatin from pups at day of life 0 (DOL 0). ChIP DNA was amplified using CpG rich primers for DD-polymerase chain reaction and separated. One of the clones isolated from the histone AcH3/K9 ChIP was 611 bp in size and contained of exon 2 of DUSP5, a MAP kinase phosphatase.
2. IUGR affected DUSP5 juvenile and adult CpG DNA methylation in a gender-specific manner
DUSP5 exon 2 includes 6 CpG sites. Bisulfite modification was used to evaluate the methylation status of the 5 downstream sites in liver at DOL 0, 21, and 120. Primer design required that 1 CpG site be included in the 5' primer.
IUGR 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. 1
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 hypomethylated in the DOL 0 liver DNA (Fig. 1A
).
|
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 CpG sites were methylated in from DOL 21 control male livers (**P<0.01) (Fig. 1B
). Two specific CpG sites were hypomethylated in liver from DOL 21 male rats: site 4, IUGR 55 ± 6%* vs. Con 78 ± 5%; and site 6, IUGR 50 ± 19%* vs. Con 83 ± 10% (*P<0.05) (Fig. 1B
). DNA from female livers at DOL 21 did not demonstrate a difference in CpG methylation (Fig. 1C
).
Hepatic DNA from Con and IUGR rats at DOL 120 also revealed gender-specific differences. IUGR decreased overall methylation of the 5 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. 1D
). Among the five sites, DNA methylation at CpG site 3 was decreased to 55.3 ± 9.9%* in IUGR male liver DNA (*P<0.05). Con DNA methylation in DOL 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 hypermethylated at 78 ± 1.5%* in IUGR DOL 120 hepatic female DNA, in contrast to 71 ± 6% in day of lie 120 control hepatic female DNA (*P<0.05) (Fig. 1E
).
3. IUGR affected DUSP5 juvenile and adult mRNA levels in a gender-specific manner
Real-time RT-polymerase chain reaction (RT-PCR) was used to measure DUSP5 mRNA levels at DOL 0, DOL 21, and DOL 120. IUGR decreased hepatic DUSP5 mRNA levels in IUGR pups at day 0 of life 0 (74±5% of Con value; P<0.01). Similarly, DUSP5 mRNA levels continued to be decreased in livers from DOL 21 male (66±5% of Con value; P<0.01) and female rats (76±8% of Con value; P<0.01), as well as DOL 120 IUGR male rats (89±5% of Con value; P<0.05). Hepatic DUSP5 mRNA levels increased in the livers of DOL 120 IUGR female rats (160±11% of control values; P<0.01).
4. IUGR affected juvenile and adult Erk1/2 phosphorylation in a gender-specific manner
Because DUSP5 dephosphorylates both Erk1 and Erk2 kinases, Western blots were used to measure Erk1/Erk2 and phosphorylated Erk1/2 level (pErk). Uteroplacental insufficiency increased pErk1 and pErk2 in 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. At DOL 21, IUGR 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).
In the DOL 120 livers, pErk1 and pErk2 levels were increased in male IUGR livers, without affecting total Erk levels (male IUGR pErk1 665±47%** vs. Con; male IUGR pErk2 378±56%* vs. Con: *P<0.05, **P<0.01). In contrast, total levels of Erk1 and Erk2, as well as pErk1 and pErk2, were 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).
5. IUGR increased phosphorylation of hepatic insulin receptor substrate-1 at serine 612 at DOL 120
A target of Erk1/Erk2 relevant to this model of IUGR and postnatal insulin resistance is serine 612 of insulin receptor substrate 1 (p612-IRS). As expected with increased phosphorylated Erk1/Erk2, IUGR 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. 2
). Total IRS-1 protein levels were decreased in the IUGR male liver and unaffected in the IUGR female livers.
|
CONCLUSIONS AND SIGNIFICANCE
Growing epidemiological evidence indicates that poor fetal growth predicts adult morbidities, such as insulin resistance. "Baker’s fetal origins of adult disease hypothesis" conceptualizes this relationship. 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 phenotype. Our goal, as well the goal of many other investigators, has been to identify a specific gene directly related to glucose (Glc) metabolism, which couples the adaptation to the adult phenotype. The findings in this study suggest a subtle variation in the theme in that IUGR shifts the sensitivity of Erk signaling by affecting DUSP5 epigenetics and subsequent mRNA expression. This shift is upstream of several key cellular processes and thereby potentially initiates a coordinated adaptation.
Our ChIP analysis identified DUSP5 as a gene whose nucleosome position is altered by IUGR. Considering other studies, this is not surprising. Acetylation of H3/K9 typically associates with intergenic regions of genes, and our finding of decreased "downstream" CpG methylation coupled with decreased mRNA levels is a well-established pattern.
A "perfect" direct relationship does not exist between DUSP5 exon 2 CpG methylation and DUSP5 mRNA levels, but it would be naive to expect one aspect of chromatin structure to dominate the regulation of DUSP5 transcription. Histone modifications throughout the length of the gene and transcription factor complex composition are also likely to contribute. Within this context, CpG methylation of exon 2 is likely to either amplify or dampen the signals for DUSP5 transcription.
Liver expresses high amounts of DUSP5 mRNA. DUSP5 dephosphorylates both Erk1/Erk2 kinases. The decrease in DUSP5 mRNA and increase Erk phosphorylation may be an adaptive response to the deprived IUGR intrauterine milieu. Erks mediate survival signals. Because damage to cellular membranes and proteins interferes with signaling pathways, increasing baseline Erk phosphorylation by decreasing DUSP5 mRNA increases the odds of cell survival.
In cell culture, Erk1/2 associate with IRS-1 upon activation and are involved in IRS-1 serine 612 phosphorylation. Phosphorylation of serine 612 inhibits IRS-1 signaling by inhibiting IRS-1 tyrosine phosphorylation by JAK1. As a result, phosphorylation of IRS-1 serine 612 acts as an IRS-1 desensitization mechanism.
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. IRS-1 knockdown experiments demonstrate that decreased IRS-1 leads to hepatic insulin resistance. The present study is 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 influenced by gender. The mechanism through which gender influences the hepatic epigenetics and subsequent gene expression is unknown. However, studies utilizing rats rendered IUGR consistently demonstrate gender-specific differences in gene expression.
In summary, we found that IUGR causes perinatal and postnatal changes in the epigenetic characteristics of the DUSP5 gene, and that these changes associate 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 cell death and become a maladaptation latter in life.
|
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-6179fje
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 |
