Background
The family of particulate guanylate cyclases (GC) is known to play substantial roles in the function of a number of different organs. These membrane receptors include the natriuretic peptide receptors GC-A and GC-B in the cardiovascular system, and GC-E and GC-F in the retinal phototransduction system (reviewed in [
1]). GC-C is most highly expressed in the intestine where it is the receptor for the peptides guanylin (Gn)[
2] and uroguanylin (Ugn)[
3] and contributes to the maintenance of fluid and salt homeostasis via ligand-activated cGMP production. Activation of this receptor by the bacterial heat-stable enterotoxin[
4], which is very similar in sequence and structure to the mammalian peptides Gn and Ugn, results in secretory diarrhea in both animals and man and is a major health problem. Recent advances in our understanding of GC-C mediated signaling in intestinal epithelial cells have suggested a role in basic cellular processes including apoptosis and proliferation ([
5‐
9].
While highest in the intestine, GC-C expression is also found in other tissues including kidney[
10], pancreas[
11], and liver. Similar to its expression in the intestine, GC-C is temporally regulated in the liver of rodents and is expressed at highest levels in the perinatal period [
12,
13]. In the adult liver, GC-C expression is undetectable (by Northern analysis) but is up-regulated in injury/regeneration models, including exposure to the hepatotoxin CCl
4 and by partial hepatectomy [
14,
15]. In mice with a deficiency of GC-C the liver appears normal and no phenotype has been described under specific-pathogen free husbandry[
16,
17].
Administration of CCl4 is a widely used model of necroinflammatory liver injury and regeneration. In centrilobular hepatocytes of the liver, cytochrome P450 enzymes mediate metabolism of CCl
4 into toxic free radicals which cause lipid peroxidation and membrane damage, eventually resulting in necrosis [
18]. Hepatocyte death due to apoptosis via activation of caspase 3, although not as prominent as death due to necrosis, has also been demonstrated to play a role in CCl
4 injury [
19‐
22]. Proliferation of surviving hepatocytes, along with removal of cellular debris and restoration of the extracellular matrix, ultimately results in liver regeneration [
23]. In order to explore the role of GC-C in the liver, we compared the response of wild type (WT) and GC-C null mice to acute CCl
4 injury.
Methods
Mice
GC-C knock out (KO) mice with a targeted disruption of the
Gucy2c gene [
16] were maintained on a C57BL/6J background, following 10 backcross generations. Where indicated, these G-CC null mice were crossed with Black Swiss outbred mice (NTac:NIHBS, Taconic, Hudson, NY) and the resulting heterozygous mice mated to generate homozygous wild type (WT) controls and GC-C null littermates (F2 generation). All mice were genotyped by PCR. Mice were housed under specific-pathogen free conditions and fed food and water
ad libitum. Adult mice of both sexes, aged 8-12 weeks, were used in experiments. Animal protocols were approved by the Institutional Animal Care and Use Committee of the Cincinnati Children's Hospital Medical Center (Cincinnati, OH).
Acute Liver Injury
Mice were exposed to a single intraperitoneal injection of CCl4 (Sigma-Aldrich, St. Louis, MO) at a dose of 1 μl CCl4 per gram body weight as a 50% solution in corn oil. Vehicle control mice were injected with corn oil alone. Injections were performed within a 2 hr window to minimize any variation due to circadian rhythm. Mice were examined daily and sacrificed at the indicated times. The liver was excised, weighed, and divided into aliquots which were snap-frozen for later analysis, or fixed in formalin. After paraffin embedding, sections were either stained with hematoxylin and eosin for light microscopic analysis or reserved for immunohistochemistry.
Hepatocyte Death
DNA fragmentation was examined using TUNEL staining as described [
5]. Briefly, tissue sections were digested with Proteinase K, and then incubated in freshly made reaction buffer containing terminal transferase and biotin-16-dUTP (Roche Diagnostic, Indianapolis, IN) to end-label DNA fragments. Labeled DNA was detected using the Vectastain ABC-alkaline phosphatase system (Vector Laboratories, Burlingame, CA) and Sigma Fast Red (Sigma-Aldrich). As a negative control, terminal transferase was omitted from the reaction buffer. To measure hepatocyte death at 24 hours post- CCl
4 injection, the number of labeled nuclei present in 3 high-power fields per liver was counted by an investigator unaware of the mouse genotype. Intense staining of necrotic areas at later time points prevented similar analyses. Instead, the area of necrotic damage compared to the total area of the field, excluding blood vessels and ducts, was determined using ImageJ version 1.38 (National Institutes of Health, Bethesda, MD), and a percentage of necrotic damage was calculated for each liver sample.
Apoptosis Detection
Paraffin-embedded liver sections were stained with antibody against activated caspase 3 (1:250 dilution, Cleaved Caspase 3 (Asp175) Antibody, Cell Signaling Technology, Inc., Danvers, MA) and counterstained with hematoxylin as described [
5]. To quantify, cleaved caspase 3 -positive cells were counted in 10 randomly selected fields per mouse liver (200× magnification) by an investigator unaware of the genotype. Caspase 3 activity was measured using a colorimetric substrate (Ac-DEVD-pNA, Enzo Life Sciences, Plymouth Meeting, PA). An aliquot of frozen liver was homogenized in lysis buffer containing 50 mM HEPES, pH 7.4, 5 mM CHAPS, and 5 mM DTT, and centrifuged at 20,000 ×
g for 30 min at 4°C. Cytosolic protein (100 μg as determined by the Bradford assay, Bio-Rad, Hercules, CA) was incubated overnight at 37°C with 20 mM Ac-DEVD-pNA in assay buffer (20 mM HEPES, pH 7.4, 2 mM EDTA, 0.1% CHAPS, and 5 mM DTT). The amount of pNA released from the substrate was measured by determining the absorbance value at 405 nm. The activity in control livers from mice injected with vehicle alone did not differ between genotypes so the combined average value was used for normalization.
Hepatocyte Proliferation
Proliferating cell nuclear antigen (PCNA) staining was used as a marker for hepatocyte proliferation that occurs in response to CCl4 injury. Tissue sections were deparaffinized, rehydrated and endogenous peroxidase activity quenched with 1% hydrogen peroxide for 15 min. After washing, tissues were heated in 0.01 M sodium citrate, pH 6, to promote antigen exposure. Sections were then washed, blocked in 3% normal goat serum, and incubated with PCNA polyclonal antibody (1:500, Santa Cruz Biotechnology, Santa Cruz, CA) overnight at 4°C. Following washes, sections were incubated with biotinylated anti-rabbit secondary antibody and labeled nuclei detected using the Vectastain ABC-horseradish peroxidase kit and 3,3-diaminobenzidine (Vector Laboratories). Sections were counterstained with hematoxylin (Fisher Scientific, Pittsburgh, PA). As a negative control, primary antibody was omitted. For each tissue, the hepatocyte labeling index (percent of nuclei that stain positive for PCNA) was determined. Both PCNA-labeled and unlabeled hepatocyte nuclei were counted in 3 high-power fields (≥ 1100 nuclei per tissue) in a blinded fashion.
Quantitative Real-Time PCR
Total RNA was isolated from frozen liver using Tri Reagent (Molecular Research Center, Inc., Cincinnati, OH) according to the manufacturer's protocol. RNA samples were treated with DNase I (DNA-
free, Ambion, Austin, TX) and reverse-transcribed (2 μg) using random decamers (RETROscript, Ambion). PCR reactions using specific gene primers were performed with Brilliant II SYBR Green QPCR mix (Stratagene, La Jolla, CA) in the Mx3000p thermocycler (Stratagene). A relative amount for each gene examined was obtained from a standard curve generated by plotting the cycle threshold value against the concentration of a serially diluted RNA sample expressing the gene of interest. This amount was normalized to the level of 18 S RNA, which did not vary with CCl
4 treatment. Primer sequences used were as follows: GCC, forward 5'-CGAAAGCGCCTGCGTGAAGC and reverse 5'-TTCACAGGTGCTGCTCCGGC; 18 S, forward 5'-GATCCGAGGGCCTCACTAAAC and reverse 5'-AGTCCCTGCCCTTTGTACACA; Ugn, forward 5'-TGAGTTGGAGGAGAAGGAGATGTC and reverse 5'-AAGGGCAAGGCTGGGTTATG; Gn, forward 5'-GAGTGACATCGCTTGCCTTTC and reverse 5'-TGAGTTTGTTAGCCTCGTGACTTC; and Cyp2e1, forward 5'-TCAAAAAGACCAAAGGCCAGC and reverse 5'-TCCGCAATGACATTGCAGG [
24].
Statistics
Kaplan-Meier survival curves were generated for each genotype and the log-rank test used for comparison (Prism, GraphPad, San Diego, CA). The Mann-Whitney test was used to evaluate the statistical significance between 2 groups (Prism). A P value less than or equal to 0.05 was considered significant.
Discussion
This study demonstrates that GC-C, known primarily for its role in intestinal epithelial cell function, also plays an important role in the response to liver injury. GC-C null mice do not survive the acute liver injury mediated by a dose of CCl4 that causes 35% lethality in WT mice. The median length of survival is 5 days with almost all GC-C KO mice succumbing by post-injury day 9.
There was no effect on survival in GC-C deficient mice on the C57Bl/6J inbred background following toxic liver injury (Figure
3). Strain-specific differences in the phenotype of gene mutations are found in the literature, especially in the study of multigenic diseases such as cancer and diabetes [
28]. For example, the C57BL/6J strain is known to be resistant to liver tumorigenesis in response to the chemical diethylnitrosamine and thus was used to elucidate the role of p27Kip1 as a tumor suppressor [
29]. This strain is also resistant to liver fibrosis following chronic CCl
4 administration, most probably due to a predominantly Th1 cytokine response compared to a Th2 response in susceptible strains [
30]. At this time the C57BL/6J-specific factors that are responsible for the successful recovery of GC-C KO mice after liver injury have not been identified. Our use of the outbred Black Swiss mouse was crucial in uncovering the lack of survival following liver injury in the GC-C KO [
24] that was not seen on the C57Bl/6J inbred background. However, the concomitant individual variability also complicates the study of the mechanism (s) of GC-C action. We focused our research on comparing the early hallmarks of hepatocyte apoptosis/necrosis and subsequent proliferation in both genotypes. Possible causes of deaths occurring later in the regeneration process have not been explored.
GC-C has been localized to non-parenchymal cells and hepatocytes [
15,
31] and although the cellular localization of Gn and Ugn is not known their detection and increased expression following CCl
4 exposure (Figure
1) is consistent with a role for the GC-C receptor signalling system during liver repair. In addition, both Gn [
32,
33]and Ugn [
34,
35]are known to circulate in the plasma. In response to CCl
4 injury GC-C expression is also up-regulated in the mouse liver, as has previously been shown in rats [
14]. In primary rat hepatocyte cultures, the addition of glucocorticoids and IL-6 [
31], and both insulin and heregulin-β1[
36], have been shown to increase GC-C levels. Therefore, we speculate that IL-6 [
19], a key coordinator in the response to liver injury, plays a role in mediating the increased expression of GC-C that we and others have observed.
While the level of hepatocyte proliferation after injury is comparable to WT (Figure
8), GC-C deficiency results in increased hepatocyte cell death (4 times the level in WT liver) 1 day after CCl
4 administration, as well as an increased extent of necrotic areas at later time points (Figures
4 &
6). Our Caspase 3 studies (Figure
7) demonstrate that increased apoptosis also contributes to the overall increased hepatocyte death in GC-C null mice. This suggests that the mortality that occurs within a few days of injury may be due to an imbalance in the rates of hepatocyte loss and repopulation. Increased levels of apoptosis have been demonstrated in GC-C KO intestine in a non-lethal radiation model and appear to be mediated by a cGMP-dependent mechanism [
5] while one study also demonstrated an increase in crypt cell apoptosis in the GC-C KO intestine at baseline compared to WT mice [
6]. It is not known if the protective mechanisms of GC-C signalling in the intestine and liver are similar and further exploration is warranted.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
EAM designed and performed the experiments, analyzed the data, and wrote the paper. KS participated in experimental design and in editing of the paper. MBC designed experimental strategy, interpreted the data, and critically edited the paper. All authors read and approved the final manuscript.