|Home | About | Journals | Submit | Contact Us | Français|
Augmenter of liver regeneration (ALR), a protein synthesized and stored in hepatocytes, is associated with mitochondria, and possesses sulfhydryl oxidase and cytochrome c reductase activities. We sought to determine the effects of ALR depletion in hepatocytes by antisense oligonucleotide transfection.
Rat hepatocytes in primary culture were transfected with antisense oligonucleotide for ALR mRNA (ALR-AS) or scrambled oligonucleotide. Various analyses were performed at times up to 24 h after transfection.
Treatment with ALR-AS caused a decrease in ALR mRNA, cellular depletion of ALR protein primarily from mitochondria, and decreased viability. Flow cytometric analysis of ALR-AS-transfected hepatocytes stained with annexin-Vcy3 and 7-aminoactinomycin D revealed apoptosis as the predominant cause of death up to 6 h; incubation beyond this time resulted in necrosis in addition to apoptosis. ALR-AS-transfection caused release of mitochondrial cytochrome c, activation of caspase-3, profound reduction in the ATP content, and cellular release of LDH. Inhibition of caspase-3 inhibited the early phase of ALR-AS-induced death but not the late phase that included ALR and LDH release.
These results suggest that ALR is critically important for the survival of hepatocytes by its association with mitochondria and regulation of ATP synthesis.
Augmenter of liver regeneration (ALR), originally identified in the soluble fractions of hypertrophic rat livers [1–3], was purified from weanling rat liver [4,5], and its gene cloned in rat, mouse and human [6,7]. The mitogenic and antiatrophic activities of native and cloned ALR were demonstrated in partially hepatectomized rats  and in dogs with portacaval shunt [2–4,6,7]. ALR also inhibits the lytic activity of hepatic natural killer cells [8,9], which is critical for unabated liver regeneration . Presence of hepatotrophic activity in weanling and regenerating [1–3,11] but not in the unmodified adult animal livers [1,3] suggested that ALR is not synthesized by the latter. However, we found that ALR mRNA and protein are expressed similarly in weanling and unmodified adult rat livers in hepatocytes . Together, these observations suggest that ALR may have multiple intracellular and extracellular functions. Indeed, results of experiments with short form (15 kDa) [13–15] differ substantially from that with long form (22 kDa) ALR .
The C-terminus of mammalian ALR is 40% identical with ERV1P (essential for respiration and vegetative growth 1 protein) , product of the Saccharomyces cerevisiae gene ERV1 [17,18]. ERV1p and ALR are flavin-containing sulfhydryl oxidases localized in the mitochondrial intermembrane space [19–21]. ERV1p is necessary for the growth and survival of the yeast as indicated by complete loss of mitochondrial genome and death upon disruption of the ERV1 gene [17,22]. We hypothesized that in hepatocytes, ALR might play role functionally equivalent to ERVlp. Transfection of primary hepatocytes with antisense oligonucleotide for ALR mRNA (ALR-AS) led to mitochondrial and cellular depletion of ALR, profound loss of ATP, mitochondrial release of cytochrome c, cellular release of LDH and apoptotic/necrotic death of hepatocytes.
Hepatocytes were prepared as described previously , and suspended (0.25 × 106 cells/ml) in William’s medium E supplemented with 2 mM l-glutamine, penicillin/streptomycin, 10% FBS, and 10−6 M insulin. Cells were plated at a density of 0.063 × 106/cm2, the medium renewed 3 h later, and the cells used following overnight incubation.
The sequence of phosphorothioated ALR-AS-oligonucleotide complementary to nucleotide sequence 24–43 downstream from the start codon of ALR mRNA was 5′-GACTGCCGCGAGGGAAACCT-3′; the sequence of scrambled oligonucleotide was 5′-ACTGACAGATCGGGCAAGCC-3′. To increase stability , five bases of the oligonucleotides at both 5′ and 3′ ends were replaced with 2′-O-meth RNA, internal dCs were replaced with 5-Me-dC, FAM-labeling was added at the 5′ end and a hairpin loop was introduced at the 3′ end. Lipofectamine-2000 (8 µl/ml; Invitrogen) or the oligonucleotides (final concentration 1.2 µM) in lipofectamine-2000 were incubated separately in 200 µl of Opti-MEM medium (GibcoBRL, Grand Island, NY) for 30 min. Eight hundred microliters of Opti-MEM was added and the combined solution was overlaid onto the cells. After 5 h, 1 ml of William’s medium E containing 20% FBS was added to each well. The efficiency of transfection was determined by fluorescence microscopy.
Collagen-coated (10%) 40-mm coverslips with cultured hepatocytes were inserted in a FCS2 closed heated (37 °C) live cell chamber (Bioptics, Bulter, PA), which was placed on a Nikon TE300 inverted microscope (Melville, NY). Media were perfused through the chamber at a rate of 0.5 ml/h. Using a 40× dry objective, five stage positions were imaged with transmitted light every 3 min. The acquired images were processed with Metamorph software (Universal Imaging Corporation, Downingtown, PA).
The medium containing detached cells was aspirated, and the attached cells were harvested using 0.025% trypsin. The two cell fractions were pooled, washed twice with PBS by centrifugation at 450g (10 min, 4 °C), fixed in ice-cold 70% ethanol for 3 h, and washed with Ca2+-/Mg2+-free HBSS containing 1% BSA. The cells were suspended in 0.5 ml of propidium iodide solution (50 µg/ml propidium iodide, 1 mg/ml sodium citrate, 100 µg/ml RNase I and 0.1% Triton X-100). After 30 min at 37 °C, the cells were analyzed by flow cytometry in a fluorescence-activated cell sorter (Epics XL.MCL, Beckman–Coulter) using EXPO32 software.
To distinguish apoptosis and necrosis, the cells were harvested (see above), washed with PBS, and suspended in buffer A (10 mM Hepes, 140 mM NaCl and 2.5 mM CaCl2, pH 7.4) at 1 × 106 cells/ml. Annexin-Vcy3 (4 µg/ml) and 7-aminoactinomycin D (7-AAD) (5 µg/ml) were added to 100 µl of the cell suspension. After gentle mixing, the suspension was incubated at room temperature for 15 min in dark, followed by the addition of 400 µl of buffer A. Flow cytometry was performed within 1 h.
The viability was determined by the MTT assay . The cells were harvested as described above for ATP determination using Cell Viability Assay Kit-ATP (Sigma Chemical Co., St. Louis, MO). Cytosolic cytochrome c was measured using the Quantikine murine immunoassay kit (R&D Systems, Minneapolis, MN). Caspase-3 activity was determined using caspase fluorescent assay kit (BD Biosciences-Clontech, San Jose, CA). LDH was measured using spectrophotometric assay kit (Stanbeo Laboratory, Boerne, TX).
After treatments, the culture medium was aspirated and centrifuged to separate detached cells. ALR in the medium was measured by ELISA . The attached cells were harvested by trypsin treatment; the two cell fractions were combined and centrifuged (l000g/5 min). Total RNA was extracted from the cells and analyzed for ALR mRNA by semiquantitative RT-PCR as described previously .
For mitochondrial preparation, hepatocytes (pooled attached and detached fractions) were homogenized in 2 mM Hepes, pH 7.4, containing 0.22 M mannitol, 70 mM sucrose, 0.1 mM EDTA, and 1% fatty acid-free BSA. The homogenate was centrifuged at 800g for 10 min, followed by centrifugation of the supernatant at 11,000g for 15 min. The pellet was washed and suspended in RIPA buffer (Santa Cruz Biotechnology, Santa Cruz, CA) containing 25 µl/ml protease inhibitor cocktail (Sigma) and 0.5 mM phenylmethylsulfonylfluoride. After 15 min on ice, the lysate was centrifuged (10,000g/10 min); ALR in the supernatant was determined by ELISA and Western blot analysis .
For cytosolic preparation, hepatocytes were homogenized in 10 mM Tris–HCl, pH 8.0, containing 60 mM KCl, 1 mM EDTA, 100 µM PMSF and 25 µl/ml protease inhibitor cocktail. After 10 min on ice, NP40 (25 µl/ml) was added; the homogenate was vortexed vigorously for 10 s, centrifuged (2500 rpm/4 min) followed by centrifugation of the supernatant at 14,000 rpm for 5 min. ALR in the supernatant was quantified by ELISA .
Statistical significance was determined by ANOVA-Duncan analysis using SPSS program. A p value of 0.05 was considered statistically significant.
Fig. 1A shows purity of the rrALR by Coomassie blue staining and Western blot analysis . The rrALR did not affect the DNA synthesis in hepatocytes at either 24 h or 48 h (Fig. 1B); in contrast, TGF-α caused a robust increase in the DNA synthesis at both points. The lack of response of rat hepatocytes to 22 kDa-rrALR is consistent with absence of ALR-specific receptors in them .
ALR-AS-treatment caused time-dependent loss of ALR from cells with concomitant increase in the extra-cellular medium (Fig. 2A and B). This effect was associated with progressive loss of viability (Fig. 2C). The relatively high basal level of ALR release without apparent loss from cultured hepatocytes is consistent with our previous report . Phase contrast microscopy showed rounding and detachment of hepatocytes by ALR-AS but not scrambled-ODN treatment (Fig. 3A). Flow cytometric cell cycle analysis (Fig. 3B) showed that 35% and 33% of ALR-AS-transfected hepatocytes were in G0/G1 and G2/M phases, respectively, as compared with 20% and 70% control cells, indicating strong growth arrest and increased apoptosis . Scrambled ODN-treatment induced only marginal changes. Continuous acquisition of the live cell images of ALR-AS-transfected hepatocytes showed beginning of their shrinking and detachment within 2–3 h. In the image captured at 5.5 h (Fig. 3C), several hepatocytes exhibited apoptotic characteristics (rounding, loose contact with neighboring cells and surface, and cytoplasmic blebs). Only few scrambled-ODN-treated hepatocytes showed evidence of morphological changes indicative of apoptosis at 12 h. Together these results suggest that ALR is essential for physiological function and survival of hepatocytes.
We determined ALR mRNA and protein levels at 12 h time point considering near-maximal loss of ALR and viability (Fig. 2). ALR-AS-treatment caused nearly complete loss of ALR mRNA, (Fig. 4A). There was substantial loss of ALR from ALR-AS-treated cells when the whole cell homogenate was analyzed (Fig. 4B); tracking the subcellular distribution showed that cellular ALR loss was predominantly from the mitochondrial compartment. The decrease in the cytosolic ALR was modest, but the loss of mitochondrial ALR was very strong both by ELISA (Fig. 4B) and Western blot analysis (Fig. 4C).
Whether ALR-AS causes necrosis secondary to apoptosis was determined in hepatocytes labeled with annexin-Vcy3 [marker of phosphatidylserine externalization during apoptosis [27,28], and 7-AAD [permeant marker of presumptively necrotic cells . At 24 h of ALR-AS treatment, apoptotic cells (Fig. 5A, lower right quadrant) increased threefold (Fig. 5B), while the number of cells in late apoptosis/necrosis phase (upper right quadrant) (Fig. 5A) increased fourfold (Fig. 5B). Relatively small changes were seen in scrambled ODN-transfected cells (Fig. 5A and B).
To evaluate more precisely the role of apoptosis in ALR-AS-treated hepatocytes, we quantified cytosolic cytochrome c and caspase-3-like activity. Both cytosolic cytochrome c and caspase-3-like activity increased robustly between 6 and 12 h in ALR-AS-treated cells, with peak values at 12 h (Fig. 6). Scrambled ODN-treated hepatocytes showed a smaller progressive increase in cytosolic cytochrome c between 6 and 24 h, and relatively small increase in the caspase-3-1ike activity. The ATP content of ALR-AS-treated cells decreased progressively between 3 h and 24 h, with the sharpest decrease between 3 and 6 h (Fig. 7A). Scrambled ODN-treated cells showed a small progressive decrease, while in control cells ATP content was unchanged. LDH was released at 12 h and later from ALR-AS-transfected cells (Fig. 7B), with no change in scrambled ODN-treated cells. Together these results confirm that both apoptosis and necrosis contribute to the loss of ALR-AS-treated hepatocytes.
Next we determined the time-course of the loss of ALR and ATP from mitochondria (Fig. 8). Within 1 h after ALR-AS treatment, there was substantial loss of mitochondrial ALR (maximal at 6 h), followed by ATP loss with significant decrease at 3 h and maximal between 12 and 24 h.
Finally, we determined if caspase-3 inhibition blocks ALR-induced apoptosis/necrosis of ALR-treated hepatocytes. Pretreatment and presence of DEVD-fmk during ALR-AS-treatment inhibited depletion of ALR, and LDH release till 6 h (Fig. 9A–C) and inhibited substantially the loss of viability till 12 h (Fig. 9D). But after this, the loss of ALR, LDH and viability was nearly similar in incubations without and with DEVD-fmk (Fig. 9). These results suggest that the initial phase of ALR-AS-induced cells death (apoptotic) is inhibited by caspase-3 inhibition, but not the continued ALR loss leading to necrosis.
In this report we present evidence supporting ALR’s role as an intracellular survival factor for hepatocytes. The 22 kDa rrALR did not stimulate DNA synthesis in rat hepatocytes unlike in human hepatocytes with recombinant human ALR dimer (rhALR) [13,15]. On a molar basis hepatocyte DNA synthesis stimulated by rhALR was much stronger than that by the powerful hepatocyte mitogens EGF and TGF-α . The effects of rhALR and the most potent hepatocyte mitogen HGF on the DNA synthesis in human hepatocytes were also comparable . The discrepancy may be because we used the entire rrALR molecule (198 amino acids; starting sequence MAAPS–) and not the short ALR molecule (125 amino acids) with starting sequence of MRTQ– (“M” representing amino acid “74” in native ALR) . In previous studies [13,15], dimer of the human ALR C-terminus sequence starting with MRTQ– (“M” representing amino acid “71” in native ALR ) expressed in the bacterial system was used. The monomer (short ALR) has mw of about 15 kDa while the cDNA for native ALR encodes a protein with mw of about 22 kDa [6,7]. The crystal structure  and the FAD-linked sulfhydryl oxidase and cytochrome c reductase activities of ALR  were deduced based on the assumption that it is present as a dimer of the short ALR. Unlike the rat and mouse ALR monomers, each 15 kDa subunit of human ALR contains two additional nonconserved free cysteine residues (C74 and C85) . Again whether the long human ALR performs the same function is unclear . However, we note that both the full-length 22 kDa and 15 kDa carboxy terminal portion of the ERV1 protein, which contains the conserved CXXC motif, were found to bind FAD and perform the sulfhydryl oxidase function . Interestingly, the full-length ERV1 protein (22 kDa/189 amino acid residues), but not its carboxy terminal 15 kDa domain that forms dimer , contains leader sequence for import into mitochondria, and thus the N-terminal domain is clearly required for mitochondrial translocation of ERV1p . Putative targeting sequences for mitochondria have also been found in the homologous mammalian proteins . ALR is colocalized with GFAP in neurons in all brain regions, specifically in the nucleus and the external envelope of mitochondria; Western blot analysis demonstrated at least 2 dimers, which under reducing condition migrated as two bands with MW of 21 and 23 kDa . The rat ALR was found to exist as high molecular weight species (about 38–42 kDa) and can very well be the dimers of the full-length ALR . A 15 kDa ALR isoform was reported to be present in the human cancer cell line HepG2 , but whether it dimerises before eliciting biological actions is not known.
Transfection of rat hepatocytes with ALR-AS decreased their viability profoundly due to apoptosis and necrosis. Significant release of cytochrome c, but not of LDH, at 6 h indicates that apoptosis is a primary mechanism of cell death at this time point. Thereafter, the release of LDH increased suggesting that necrosis was a major cause of cell death after 6 h. In the late stages of apoptosis also cell membrane may become permeable thus allowing the release of LDH [34–36]. The involvement of both cell death processes was confirmed by the presence of annexin-5cy3-positive/7-AAD-negative and annexin-5cy3-positive/7-AAD-positive cells after ALR-AS treatment. While ALR depletion caused apoptosis and necrosis of hepatocytes, interestingly, only the apoptotic phase of cell death was strongly but not completely inhibited by caspase-3 inhibition. DEVD-fmk inhibited but did not prevent ALR loss as well as LDH release from ALR-AS-treated cells. Thus the necrotic phase of ALR depletion-induced cell death seems to be independent of prior apoptosis. Also caspase-independent apoptosis at early times (compare Figs. 6B and and9D)9D)  may not be ruled out. A small effect of the scrambled-ODN observed at late time points might be nonspecific as these effects are not accompanied by any cellular or mitochondrial ALR loss compared to the control.
Depletion of ATP in ALR-AS-transfected hepatocytes suggests possible association of ALR with mitochondrial function. Mitochondria playa critical role in apoptosis, which cannot occur in the setting of cellular ATP depletion [38–43]. Both an apoptosis-dependent increase in ATP consumption and a separate decrease in ATP synthesis may contribute to the lethal ATP depletion, global loss of cellular integrity, and necrosis  of ALR-AS-treated hepatocytes. In support of this supposition, ERV1P is shown to be essential for the biogenesis of mitochondria [17,22,32], normal mitochondrial morphology and stable maintenance of mitochondria in S. cerevisiae [16,45]. Disruption of the ERV-1 gene results in complete loss of mitochondrial genome and death of the yeast following a few replications . Thus the significant association of ALR with mitochondria and its progressive loss that preceded mitochondrial ALR depletion and cytochrome c release during treatment with ALR-AS are consistent with the notion that ALR, like ERV-1P, might regulate oxidative phosphorylation, which is critical for the function and survival of the cells. In intact rats, ALR modulates hepatocyte mitochondrial function by increasing both nuclear-encoded mitochondrial transcription factor A and the oxidative phosphorylation capacity of the liver . It is well known that a number of hepatically active mediators have no distinct effect in hepatocytes, but elicit the release of secondary mediators from nonparenchymal cells. Indeed, our preliminary data (not shown) indicate that rat Kupffer cells specifically bind radioiodinated 22 kDa-rrALR. These data provide both a convincing positive control for the undetectable binding seen in hepatocytes, and a mechanism for ALR-elicited hepatocyte-targeted mediator release from this nonparenchymal population.
Although ALR and ERV-1 are essential for the survival of hepatocytes and S. cerevisiae, respectively, a major difference is that ALR is abundantly expressed in hepatocytes , but the expression of ERV-1 is very low in the yeast . Despite the low level of ERV-1 expression, null mutation in the gene in the haploid yeast strain stopped their replication after 3–4 days ; it was suggested that this effect was due to high in vivo stability of the ERV-1 mRNA and protein. Lisowsky , using expression of the ERV1 gene under the GAL1-10 promoter, found complete loss of mitochondrial genome after the promoter was shut off. The cells grew for a limited time under these conditions, but showed arrest in cell-division cycle as the terminal phenotype suggesting the essential requirement but stability of the ERV1 mRNA and protein. The ALR mRNA levels were almost completely reduced and cellular and the mitochondrial release of ALR protein was significantly increased in ALR-AS-treated cells suggesting that the depletion of ALR from mitochondria is associated with hepatocyte death.
In summary, our results indicate a fundamental role of ALR in the survival of primary hepatocytes. Whether the death caused by the inhibition of its synthesis and specific release from the mitochondria has clinical relevance needs further investigation. Based on the results presented herein, we propose that there is a dynamic exchange between the mitochondrial and newly synthesized and post-translationally modified cytosolic ALR (Fig. 10). Inhibition of ALR synthesis leads to decreased uptake but does not stop its mitochondrial release. This results in reduced ATP synthesis, mitochondrial injury, release of cytochrome c and apoptosis. Progressive depletion of ALR and further decrease in ATP causes necrosis, which is independent of caspase-mediated apoptotic phase.
This work was supported by a Merit Review Award from the Veterans Hospital Medical Center to CRG.
The authors who have taken part in the research of this paper declared that they do not have a relationship with the manufacturers of the materials involved either in the past or present and they did not receive funding from the manufacturers to carry out their research.