Development of cell therapy strategies to overcome copper toxicity in the LEC rat model of Wilson disease

Many genetic disorders are amenable to cell and/or gene therapy, although the required extent of genetic correction or organ replacement with transplanted cells varies. Therefore, it is necessary to combine insights into engraftment and proliferation of transplanted cells with disease-specific mechanisms for phenotypic correction. We have been studying cell therapy for Wilson disease (WD) because this disorder exemplifies the effects of loss-of-function mutations in the ATP7B gene with hepatic and extrahepatic (brain and kidney) damage through impaired biliary copper excretion [1]. Cell and/or gene therapy will be helpful in WD since medical therapy is a lifelong commitment and patients often suffer from noncompliance-related complications.

Recently, excellent animal models have been developed for WD, including Long–Evans Cinnamon (LEC) rats, where the atp7b gene was naturally mutated, and mice with the atp7b gene had inactivation through homologous recombination [2]. The possibility of gene therapy in WD has been explored with adenoviral vector- or lentiviral vector-mediated gene transfer [3,4]. However, it is unresolved whether irreversible damage in native hepatocytes due to prolonged and extensive copper accumulation will pose problems for the gene therapy approach if these cells are prematurely lost. On the other hand, cell therapy studies in the LEC rat model have been instructive since cell transplantation can be effective for WD, although proliferation of transplanted cells without additional perturbations of the recipient liver is extremely gradual, whereas disease correction requires extensive liver repopulation [5,6]. This was initially demonstrated in studies using the chemical retrorsine, a pyrrolizidine alkaloid that synergistically enhanced copper toxicity and permitted replacement of the liver with transplanted cells in healthy rats, as well as LEC rats [5,7], although the oncogenic potential of pyrrolizidine alkaloids prevents their use in humans [8]. More recently, using the principle of oxidative stress-induced genotoxicity, we hypothesized that hepatic preconditioning with radiation (RT) and ischemia reperfusion (IRP) will offer competitive advantages to healthy transplanted cells [9]. Consequently, in rats with healthy liver, the combination of hepatic RT and IRP with temporary occlusion of a portal vein branch shortly before cell transplantation promoted transplanted cell proliferation leading to near-total liver repopulation over several weeks. This suggested that additional RT- and IRP-induced genotoxicity should advance liver repopulation with healthy cells in WD, although uncertainties arise due to pre-existing copper-induced changes that reproduce aspects of RT- or IRP-induced liver damage [10]. Therefore, we examined the role of hepatic RT and/or IRP in LEC rats for inducing proliferation of healthy donor cells from Long–Evans Agouti (LEA) rats [5,6]. We found that while transplanted cells were induced to proliferate by RT with correction of WD phenotype in LEC rats, the kinetics of liver repopulation in LEC rats was slower than in healthy rats [9]. This offers clues for organ repopulation in the setting of pre-existing disease.

Materials & methods

Healthy hepatocytes from donor LEA rats were transplanted into syngeneic LEC rats with copper toxicosis to study the effects of liver preconditioning with RT and IRP on transplanted cell proliferation. Hepatic atp7b mRNA, bile and liver copper, and liver histology were analyzed to determine animal outcomes.

Animals

The Animal Care and Use Committee at Albert Einstein College of Medicine (NY, USA) approved protocols in compliance with National Research Council guidelines (Guide for the Care and Use of Laboratory Animals; United States Public Health Service publication, revised 1996). LEC rats and syngeneic healthy LEA rats were provided by the Special Animal Core of Marion Bessin Liver Research Center. A total of six LEA rats and 90 LEC rats were sacrificed for final studies described here. Rats were 8–10 weeks old when studies were commenced. Animals were housed in 14-h light and 10-h dark cycles with unlimited access to drinking water and pelleted chow containing 11.8 mg copper/kg (Ralston Purina, MO, USA).

Hepatocytes were isolated by two-step collagenase perfusion. Cell viability was assessed by trypan blue dye exclusion and attachment to cell culture plastic in RPMI 1640 medium containing 10% fetal bovine serum and antibiotics. Groups of recipient animals were subjected under anesthesia to transient occlusion for 90 min of the left portal vein branch, as described previously [9]. Cessation and restoration of blood flow were verified by changes in color of the liver. To administer RT to the liver, an orthovoltage X-ray unit was used and 50 Gy external RT was delivered with shielding of other abdominal organs with lead plates, as described previously [9]. For cell transplantation, 1 × 10⁷ viable cells were suspended in serum-free RPMI 1640 medium and cells were rapidly injected into the splenic pulp in 0.5-ml volume within 2 h of cell isolation. To measure copper excretion before sacrificing animals, the common bile duct was cannulated and bile was collected at timed intervals before, and for 60 min after, intrasplenic injection of copper–histidine, as in previous studies [6].

RT-PCR for hepatic atp7b mRNA

Liver samples were frozen in liquid nitrogen for RNA extraction with Trizol™ Reagent (Life Technologies, NY, USA). Atp7b and β-actin mRNAs were co-amplified by reverse transcription PCR (RT-PCR) with a commercial kit (Access RT-PCR, Promega Corp., WI, USA). Atp7b primers spanned a region lacking in LEC rats: 5´CCATCTCCAGTGACATCAG3´ (forward) and 5´AGTCCCAATAGCAATGCC3´ (reverse) and β-actin primers spanned an intron: 5´AGGCATACAGGGACAACAC3´ (forward) and 5´GGAGAAGATTTGGCACCAC3´ (reverse). Single-step reverse transcription occured for 45 min at 37°C with heat denaturation of reverse transcriptase. cDNA amplification used 40 PCR cycles at 94°C × 1 min, 60°C × 1 min and 72°C × 2 min. PCR products were resolved in 1.8% agarose gels containing ethidium bromide. The expected PCR products were 380 and 200 bp in size for atp7b and β-actin, respectively.

Serum ceruloplasmin assay

Total oxidase activity was measured with 1 mg/ml dimethoxybenzidine dihydrochloride (o-dianisidine; Sigma Chemical Co., MO, USA) in 0.1 M sodium acetate buffer, pH 5.6. We added 20µg of o-dianisidine in sodium acetate buffer to 10µl of serum and adjusted the total reaction volume with buffer to 100µl. For negative controls, 0.1 mg sodium azide was added to another aliquot of each serum sample for inhibiting oxidase reaction. After adding 100 µl of 9 M sulfuric acid, reaction tubes were incubated for 90 min at 37°C. Absorbance was measured at 540 nm with corresponding negative controls. Purified ceruloplasmin protein was used to generate standard curves for converting oxidase activity measurements to represent ceruloplasmin protein.

Copper measurement

Livers were desiccated at 65°C under vacuum for 12 h. Dried liver and bile samples were stored at -20°C. Tissue samples were solubilized in nitric acid and copper was measured by graphite furnace atomic absorption spectroscopy.

Histological analysis

Liver samples were fixed in 10% buffered formalin. Paraffin-embedded sections were stained with hematoxylin and eosin. Two experienced observers assessed blinded tissue sections for macro- and microvesicular steatosis, polyploidy, apoptosis and mitosis, according to a scale described previously [11]. Any discrepancies in grading were resolved by reassessment. The grades were then summed and averaged. The maximum possible histological score in the presence of advanced liver damage was 13. The minimum histological score in healthy livers was 2.

Statistical analysis

Data are shown as means ± standard deviation. The significances were analyzed by χ² test with Yates correction, Student’s t-test and Kruskall–Wallis analysis of variance (ANOVA using Dunn’s test for multiple pairwise comparison of group ranks). SigmaStat 3.1 software was used (Jandel Scientific, CA, USA); p < 0.05 was considered significant.

Results

Healthy hepatocytes from donor LEA rats were transplanted into syngeneic LEC rats with copper toxicosis to study the effects of liver preconditioning with RT and IRP on transplanted cell proliferation. Hepatic atp7b mRNA, bile and liver copper, and liver histology were analyzed to determine animal outcomes.

Experimental design

We grouped animals for studies, including unperturbed LEA and LEC rats, as well as animals after various experimental perturbations, as indicated (Figure 1A). Each animal group used between six and 12 rats. LEC rats were subjected to cell transplantation without prior manipulation, or following hepatic RT, hepatic IRP or both hepatic RT and IRP. Hepatic preconditioning in LEC rats was performed 1–2 days before transplanting LEA hepatocytes via the spleen (Figure 1B), which deposits cells into liver sinusoids [1]. This was followed by the study of WD-specific parameters to establish the success of cell therapy in LEC rats after 3 or 6 months following cell transplantation. A positive therapeutic response in LEC rats was defined as detection of atp7b mRNA in the liver, decrease in hepatic copper content below 350 µg/g dry liver weight and improvement in liver histology. Levels of serum ceruloplasmin and biliary copper excretion were also measured.

Untreated LEC rats showed progressive disease

To verify that untreated LEC rats accumulated copper with onset of liver disease, we examined groups of LEC rats alongside healthy LEA rats. In LEA rats, serum ceruloplasmin levels were normal,atp7b mRNA was present in the liver, hepatic copper levels were low, bile copper excretion increased in response to a bolus of copper–histidine and liver histology showed absence of injury (Table 1). By contrast, LEC rats demonstrated multiple abnormalities in these parameters at 5–6 months, as well as at 8–9 months of age, which corresponded to the 3-month and 6-month time periods, respectively, of planned observation in rats subjected to cell transplantation. These findings established that our animal groups were appropriate for further studies.

Treatment of LEC rats with cell transplantation

The major goal of these studies was to establish whether liver repopulation and the time frame for correction of the WD phenotype could be accelerated in animals. Our expectation was that transplanted cells would proliferate in response to hepatic injury induced by RT and/or IRP. If proliferation of transplanted cells were to result in significant liver repopulation, atp7b mRNA should have become detectable in the liver and excretion of copper into bile should lower hepatic copper levels, as well as improvement of histological abnormalities. Since use of RT and IRP in healthy F344 rats resulted in extensive liver repopulation of transplanted hepatocytes within 3 months [9], we initially examined outcomes in LEC rats treated with cells at that time point.

After 3 months, analysis of LEC rats receiving healthy LEA rat hepatocytes, without any hepatic conditioning, did not show atp7b mRNA, lowering of hepatic copper or histologic improvement, indicating absence of therapeutic correction (Table 2). Similarly, no improvement was apparent in LEC rats treated with hepatic IRP followed by cell transplantation. In some LEC rats, where cells were transplanted subsequent to hepatic RT or RT plus IRP, atp7b mRNA was detected in the liver, indicating presence of transplanted cells. It should be noted that detection of atp7b mRNA meant more than 10% liver was repopulated with transplanted cells [5]. However, in these rats, hepatic copper content remained abnormally elevated and liver histology did not improve.

To determine whether more time was needed for transplanted cell proliferation in the diseased liver of LEC rats, we examined rats 6 months after cell transplantation under various hepatic preconditioning manipulations. After 6 months, in LEC rats treated with cells without any liver preconditioning, we again found that therapeutic responses were lacking (Figure 2).

On the other hand, animals subjected to preconditioning manipulations began to show evidence of disease correction. Regimens incorporating RT were most effective in this regard. In response to cell transplantation after hepatic RT, six out of 12 rats (50%) showed improvement, whereas only three out of 11 (27%) rats improved following conditioning with IRP alone (p < 0.05, ANOVA with Dunn’s test). Similarly, hepatic conditioning with RT plus IRP produced therapeutic responses in seven out of 12 rats (58%), which was again superior to conditioning with IRP alone (p < 0.05, ANOVA with Dunn’s test). RT-PCR for atp7b mRNA verified that transplanted cells were present in the liver even after 3 months (Figure 3A). Moreover, all animals showing a therapeutic response were positive for hepatic atp7b mRNA expression. Additionally, liver histology significantly improved in animals showing atp7bmRNA and lowering of hepatic copper content compared with either untreated LEC rats or LEC rats showing absence of therapeutic responses within various groups (Table 3; Figure 3B–G).

Cumulative analysis of outcomes in LEC rats 6 months after cell transplantation (Table 3) demonstrated excellent correlation between presence of hepatic atp7b mRNA, lowering of hepatic copper content and histological improvement, as well as increased biliary excretion of copper and serum ceruloplasmin activity. However, several animals in each group showed no improvement in any of these parameters, indicating absence of universal therapeutic responses after cell transplantation and the possibility of variations in fundamental processes regulating transplanted cell engraftment and proliferation under our experimental conditions.

Discussion

These studies further established that in the LEC rat model of WD, cell transplantation can restore hepatic atp7b expression with excretion of copper in bile and reversal of copper-induced liver damage. Transplantation of healthy cells alone into diseased LEC livers was insufficient for that purpose during the period of these studies, and additional manipulations were required for altering the hepatic microenvironment in LEC rats, such that transplanted cells could acquire competitive advantages for liver repopulation. In our studies, RT-induced liver conditioning prior to cell transplantation appeared to be the most effective manipulation. Provision of additional oxidative stress through IRP alone was far less effective, although the combination of IRP and RT was beneficial.

These findings will be helpful for developing suitable strategies for cell transplant therapy in WD and for defining additional mechanisms for augmenting or accelerating liver repopulation with transplanted cells in this disorder, as well as in other conditions with underlying liver injury. It should be noteworthy that our studies were restricted to the use of hepatic atp7b mRNA reconstitution to assess the extent of liver repopulation, as described previously [5], since reliable anti-rat atp7b antibodies are not available for immunohistochemical localization of transplanted cells. Also, it should be noted that as LEA and LEC rats are inbred substrains, our studies constituted syngeneic cell transplants, which would be different from additional mechanisms related to prevention of rejection following allografts in the clinical setting. On the other hand, the inbred and syngeneic nature of LEA and LEC rat avoids confounding by host immune responses, for example, antibody responses against atp7b, as indicated by permanent cures in LEC rats when the liver was appropriately repopulated by healthy cells in our studies here and previously [5,6].

In the LEC rat liver burdened by copper toxicosis, several pathophysiological mechanisms would be different from the healthy liver. For instance, copper contributes to oxidative stress, DNA damage and other types of cellular injury that may modify responses to conditioning mechanisms. The major mechanism by which RT and IRP affect the healthy liver in rats involved DNA damage in hepatocytes, followed by attrition of damaged cells, resulting in their replacement by healthy transplanted cells [9]. In this situation, hepatic conditioning with RT and IRP resulted in virtually total liver repopulation over 3 months; requiring only six to eight doublings of the transplanted cell mass. In LEC rats, major dissimilarities in RT plus IRP-induced responses concerned relatively slower kinetics of liver repopulation, requiring at least twice as long for therapeutic correction, as well as the failure of some rats to exhibit evidence of transplanted cell survival and/or proliferation and lack of phenotypic correction. Potential reasons for these differences in LEC rats should include copper-induced perturbations of transplanted cell engraftment, the less effective nature of RT- or IRP-induced changes in already perturbed hepatocytes, lower rates of transplanted cell proliferation due to increased intracellular shunting of copper during the early phases of its removal in the face of large cellular stores, or through additional unknown mechanisms.

In recent studies of the hepatic effects of copper, we observed that oxidation of extracellular matrix components perturbed hepatocyte survival, including through altered outside–in cell signaling involving nuclear factor-κB and other transcriptional activators [12]. Therefore, this altered microenvironment may impair transplanted cell engraftment, since interactions between integrin domains in transplanted hepatocytes and extracellular matrix receptors in the liver play critical roles during this process [13]. Failure of cell engraftment will probably alter cell therapy results. Moreover, if cell engraftment was suboptimal or limited, this would have major effects on the kinetics of transplanted cell proliferation and liver repopulation.

After hepatic ischemic preconditioning, some hepatocytes do acquire resistance to further injury and show prolonged survival in vivo over several months [14]. Therefore, attrition rates of native hepatocytes in LEC rats might have been different from healthy rats subjected to RT and/or IRP. Gaining insights into additional cell survival and proliferation mechanisms will be helpful for liver repopulation in WD. In particular, strategies that could be used for clinical applications will be necessary. If transplanted cells failed to engraft and proliferate or were to be subsequently cleared, various preconditioning manipulations could possibly leave behind greater damage in the native liver, perhaps including oncogenic perturbations [7]. This area needs further study, as in some situations (e.g., after exposure to pyrrolizidine alkaloids), damaged hepatocytes are eventually replaced by endogenously emerging healthy hepatocytes, suggesting that the regenerative capacity of the liver is not extinguished.

Whether genetic manipulation of cells to provide more copies of the atp7b gene, such as through lentiviral vectors that permit permanent integration of transgenes [4], could be useful needs further examination. Recent studies of genetically modified autologous LEC rat hepatocytes demonstrated that such cells showed limited survival, and proliferation of transplanted cells was not observed. These studies used partial hepatectomy as a growth stimulus for transplanted cells prior to their implantation, which by itself was not a potent stimulus for inducing transplanted cell proliferation in healthy rats [15]. Study of genetically modified healthy donor cells to express multiple atp7b copies, along with suitable liver preconditioning to promote transplanted cell proliferation, may be more effective.

Conclusion

Cell therapy can cure genetic disorders with liver damage due to toxins, as exemplified by studies in the LEC rat model of WD, where pathophysiological processes reversed after liver repopulation with healthy cells. However, inducing an effective mass of transplanted cells in LEC rats required manipulations in the form of liver RT and IRP. Without these manipulations, the liver was insufficiently repopulated and disease regression was not observed. Also, despite RT-based liver conditioning, cell therapy was less effective compared with previous studies in healthy rats, suggesting that disease-specific perturbations affected liver repopulation. This should strongly emphasize the role of organ- and disease-specific mechanisms in the development of cell therapy approaches.

Future perspective

Identification of suitable mechanisms to promote replacement of organs with transplanted cells will remain a major goal of translational studies in the areas of cell and gene therapy. Considerable investment in investigations will be necessary to establish how the liver will be best conditioned, for example, by specific drugs, RT or other manipulations capable of producing hepatic genotoxicity, oxidative stress and other cellular perturbations. The probability that suitable clinical strategies to precondition the liver prior to cell transplantation will be developed within the near future is very high, although disease-specific studies will also be needed to define the limits of cell and gene therapy efforts. Eventually, the potential of liver-directed cell and gene therapy will extend to numerous genetic and acquired disorders. The paradigms of hepatic preconditioning should be applicable to other cell and tissue compartments or organs, where proliferation of transplanted cells will offer therapeutic benefits.