| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online March 19, 2004 as doi:10.1096/fj.03-1204fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Division of Gastroenterology and Liver Diseases, USC-UCLA Research Center for Alcoholic Liver and Pancreatic Diseases, USC Liver Disease Research Center, USC School of Medicine, Los Angeles, California, USA;
Department of Pathology, Harbor-UCLA Research and Education Institute, Torrance, California, USA;
Fundación para la Investigación Médica Aplicada (FIMA), University of Navarra, Pamplona, Spain; and
CIC-Biogune, Metabolomics Unit, Technological Park, Derio, Bizkaia, Spain
2Correspondence: Division of Gastrointestinal and Liver Diseases, HMR Bldg., 415, Department of Medicine, USC School of Medicine, 2011 Zonal Ave., Los Angeles, CA 90033, USA. E-mail: shellylu{at}usc.edu
SPECIFIC AIMS
Methionine adenosyltransferase (MAT) is an essential enzyme as it catalyzes the formation of S-adenosylmethionine (SAMe), the principal biological methyl donor. MAT1A is mainly expressed in adult liver whereas MAT2A is expressed in all extrahepatic tissues. Mice lacking MAT1A have reduced hepatic SAMe content and spontaneously develop steatohepatitis and hepatocellular carcinoma. The aims of the current study were to examine the influence of chronic hepatic SAMe deficiency on liver regeneration.
PRINCIPAL FINDINGS
1. Liver regeneration is impaired in the MAT1A knockout mice
Liver regeneration after 2/3 partial hepatectomy (PH) was assessed in knockout and wild-type mice by using proliferating cell nuclear antigen (PCNA) staining and bromodeoxyuridine (BrDU) incorporation. Figure 1
A shows that peak BrDU incorporation occurred 48 h after PH in wild-type and knockout mice. However, BrDU incorporation was much lower in knockout mice at 48 and 72 h. Whereas baseline hepatic PCNA staining was much higher in knockout mice, it fell significantly after PH instead of increasing steadily as in wild-type mice (Fig. 1B
). The impairment in liver regeneration is not due to liver cell death as necrosis was not seen histologically and apoptosis was not detected on TUNEL assay.
|
2. Mechanisms of impaired liver regeneration after 2/3 PH
Immediately after PH, adult hepatocytes enter a state of prereplicative competence before they can fully respond to growth factors. This priming step is thought to be mediated partly by an early release of tumor necrosis 
(TNF) and interleukin-6 (IL-6), resulting in activation of nuclear factor 
B (NFB), c-Jun-N-terminal kinase (JNK), signal transducer and activator or transcription-3 (STAT-3), and entry in G1 phase. These initiated cells progress in early G1 phase but require growth factor stimulation to progress beyond the restriction point in mid- to late G1 and enter S phase. Whereas NFB, JNK, and STAT-3 activation occur in early G1, cyclin D1 is a delayed-early gene induced at the G1-S boundary and required for progression from G1 to S phase. We next examined whether the activation of these pathways may be impaired in the MAT1A knockout mice after PH.
TNF and IL-6 are induced in a similar manner in knockout and wild-type mice. JNK activation as measured by level of phosphorylated c-jun is evident 30 min after PH in wild-type and knockout mice. The baseline level of phospho-c-jun is elevated in knockout mice; whereas the level remained elevated in wild-type mice, it fell progressively after this early increase in knockout mice. By 36 h after PH, the level of phospho-c-jun was lower in knockout than wild-type mice. In contrast, NFB nuclear binding activity appeared comparable between wild-type and knockout mice at baseline and both were induced 24 h after PH. Consistent with intact NFB signaling is the observation that inducible nitric oxide synthase (iNOS), a NFB target gene, is similarly induced in knockout and wild-type mice. STAT-3 is a downstream target of IL-6 that is also critical for normal liver regeneration to occur. This pathway appears to be intact in MAT1A knockout mice (Fig. 2
A).
|
Another major signaling pathway activated after PH is the mitogen-activated kinase (MAPK) pathway. Extracellular signal-regulated kinases (ERKs) are activated by phosphorylation; two related MAPKs, p44 and p42, also called ERK1 and ERK2, are activated shortly after PH. Two peaks of ERK phosphorylation were observed after PH: one in early G1, within 1 h of PH, and one in mid-late G1, preceding cyclin D1 mRNA induction. Figure 2B
shows that ERKs are hyperphosphorylated at baseline in knockout mice; phosphorylation levels decreased progressively in knockout mice instead of increasing as observed in wild-type mice. At 36 h after PH before peak DNA synthesis, ERK phosphorylation was lower in knockout than wild-type mice (Fig. 2B
).
Cyclin D1 induction is required for the G1/S transition. Upstream signaling pathways involved in cyclin D1 activation include NFB, JNK, ERKs, and STAT-3. Figure 2C
shows that whereas cyclin D1 expression is doubled at baseline in knockout mice, it failed to increase significantly after PH compared with wild-type mice. Cyclin D1 protein levels in knockout livers were 6%, 15%, and 33% of wild-type livers 24, 36, and 48 h after PH, respectively (Fig. 2C
). Early activation of JNK, STAT3, and NFB coupled with the inability of MAT1A knockout hepatocytes to up-regulate cyclin D1 suggests these hepatocytes are arrested in the prereplicative (G1) phase after PH.
Cellular ATP levels are important for cyclin D1/cyclin-dependent kinase activity. UCP2 expression is more than doubled at baseline in knockout mice. UCP2 mRNA levels remained constant during liver regeneration in knockout mice in contrast to the induction that occurred in wild-type mice. Baseline hepatic ATP levels tended to be lower in knockout mice and failed to increase 36 h after PH as in wild-type mice.
3. MAT1A hepatocytes have increased basal proliferation but fail to respond to hepatocyte growth factor (HGF)
To examine whether the impairment in regeneration may be due to a loss of responsiveness to mitogenic signals, DNA synthesis was measured using knockout and wild-type hepatocytes treated with HGF. At baseline, knockout hepatocytes had a 12.7-fold higher incorporation of thymidine than wild-type hepatocytes, which agrees with these cells being in a state of enhanced proliferation in vivo. Upon HGF treatment, wild-type hepatocytes showed a dose-dependent increase in thymidine incorporation into DNA (up to 
4.5-fold), whereas knockout hepatocytes had a minimal response (1.5-fold) and showed no dose-dependency. The response in knockout hepatocytes was not merely delayed, since the same impaired effect of HGF on DNA synthesis was observed after 72 h of treatment.
We had shown that SAMe inhibits the mitogenic effect of HGF on hepatocytes by an iNOS-dependent mechanism. This coupled with the observation that hepatic SAMe levels fall after PH but before peak DNA synthesis led us to speculate that the fall in SAMe may be due to nitric oxide (NO) -mediated inactivation of MAT I/III, which then releases the inhibitory tone of SAMe on HGF. This mechanism would not work in MAT1A knockout mice as MAT II is not subject to inactivation by NO. To examine this, hepatic SAMe levels were measured after PH. In wild-type animals, SAMe levels fell progressively from 24 to 48 h but remained essentially unchanged in knockout mice.
CONCLUSIONS AND SIGNIFICANCE
Adult hepatocytes are normally quiescent (in G0 phase) but retain enormous proliferative capacity when a deficit in hepatic mass occurs, such as after PH. Much has been learned about the mechanisms by which liver regeneration occurs but there are large gaps in many fundamental areas. This paper addresses the effect of chronic hepatic SAMe deficiency on liver regeneration. This is clinically relevant as cirrhotic patients have impaired hepatic SAMe biosynthesis due to decreased MAT1A expression and inactivation of MAT I/III. The MAT1A knockout mouse model exhibits chronic hepatic SAMe deficiency and represents an ideal model to address this question.
MAT1A knockout mice exhibited an impaired regenerative response after PH. This is documented by a nearly 2/3 reduction in the number of BrDU positive hepatocytes at peak DNA synthesis. There are two critical steps in liver regeneration: the transition of the quiescent hepatocyte into the cell cycle (priming) and the progression beyond the restriction point in the G1 phase of the cell cycle. Priming is believed to be largely attributed to TNF and IL-6, which then activate a number of signal transduction pathways including NFB, JNK, STAT-3 in early G1 phase. Each pathway has been shown to be essential for liver regeneration to occur. Whereas priming is largely under the control of cytokines, progression in the G1 phase and beyond is largely controlled by HGF, transforming growth factor (TGF), and cyclin D1. Our data indicate that priming is intact in knockout mice. TNF and IL-6 responses are similar in knockout and wild-type mice, consistent with early induction of NFB, c-jun, and STAT-3. MAT2A and iNOS, shown to be essential for liver regeneration, are similarly induced in both types of mice. Intact iNOS response suggests NFB signaling is intact in knockout mice.
Our data indicate a defect in the progression in G1 phase in MAT1A knockout hepatocytes. A critical factor in G1 progression and up-regulation of cyclin D1 in mid-late G1 phase is activation of ERK1/2. We found ERK1/2 to be hyperphosphorylated and cyclin D1 expression higher in knockout mice at baseline. After PH, ERK1/2 phosphorylation fell in knockout mice and cyclin D1 failed to increase. This is consistent with arrest of the MAT1A knockout hepatocytes in the G1 phase of the cell cycle. Whereas acute MAPK cascade activation promotes G1 progression and S phase entry, chronic activation of the MAPK cascade inhibits them. JNK is also required for cyclin D1 expression. In knockout mice, baseline JNK activity is enhanced; although it increased shortly after PH, by 36 h after PH it was lower than in wild-type mice. Hepatic ATP levels did not increase in the MAT1A knockout mice after PH, likely due to the impairment in hepatic mitochondrial function of these animals.
Finally, we investigated whether MAT1A hepatocytes are abnormally responsive to mitogenic signals such as HGF. MAT1A hepatocytes have higher baseline DNA synthesis, consistent with a heightened proliferative state in vivo. However, little to no increase in DNA synthesis occurred in response to increasing doses of HGF, supporting the notion that a key abnormality in these hepatocytes is a loss in responsiveness to mitogens. Hepatic SAMe levels fell progressively in wild-type animals but remained unchanged in knockout after PH. This is likely due to the inability of NO to inactivate MAT II in MAT1A knockout livers and may have contributed to the loss of responsiveness to HGF in vivo.
In summary, MAT1A knockout mice exhibit impairment in liver regeneration after PH. Whereas the initial priming event appears to be intact, there is a defect in the progression in G1 phase. This is due largely to the inability of cyclin D1 up-regulation. These findings have direct relevance to human liver cirrhosis, where MAT1A is often silenced and the hepatic SAMe store is depleted.
|
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.03-1204fje; 
This article has been cited by other articles:
|
|
 |
|
  B. J. Spellberg, M. Collins, S. W. French, J. E. Edwards Jr, Y. Fu, and A. S. Ibrahim A phagocytic cell line markedly improves survival of infected neutropenic mice J. Leukoc. Biol., August 1, 2005; 78(2): 338 - 344. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Simile, G. Pagnan, F. Pastorino, C. Brignole, M. R. De Miglio, M. R. Muroni, G. Asara, M. Frau, M. A. Seddaiu, D. F. Calvisi, et al. Chemopreventive N-(4-hydroxyphenyl)retinamide (fenretinide) targets deregulated NF-{kappa}B and Mat1A genes in the early stages of rat liver carcinogenesis Carcinogenesis, February 1, 2005; 26(2): 417 - 427. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
