The availability of potent immunosuppressive drugs has improved outcomes following solid organ and tissue allografts.1 However, commonly used immunosuppressive drugs, such as the calcineurin inhibitors cyclosporine and tacrolimus (Tacro), rapamycin (Rapa), or analogs that bind the mammalian target of rapamycin (mTOR), and mycophenolate mofetil (MMF), an inhibitor of inosine monophosphate dehydrogenase, have multiple systemic effects.2–8 For example, these drugs can either enhance or decrease cell proliferation rates through complex mechanisms,2–8 particularly those involving mTOR.9 Similarly, multiple cell–cell interactions play roles in transplanted cell biology. For instance, studies of cell or tissue transplants showed impairment in cell engraftment and/or graft function after activation of the innate host immune system and release of inflammatory cytokines.10, 11 Perturbations in the fate of transplanted cells will have considerable implications in organ replacement. A significant question of whether immunosuppressive drugs could affect transplanted cell engraftment or proliferation during organ replacement exists. Here, we used the paradigm of liver repopulation with hepatocytes to study the effects of immunosuppressive drugs on transplanted cells. In F344 rats lacking dipeptidyl peptidase IV enzyme activity (DPPIV−), transplanted normal cells can be identified with excellent assays.12 Also, after genotoxic liver damage using retrorsine, a pyrrolizidine alkaloid, and two-thirds partial hepatectomy (PH), the liver of DPPIV− rats can be totally repopulated with transplanted cells.13
Successful grafting of tissues or cells from mismatched donors requires systemic immunosuppression. It is yet to be determined whether immunosuppressive manipulations perturb transplanted cell engraftment or proliferation. We used syngeneic and allogeneic cell transplantation assays based on F344 recipient rats lacking dipeptidyl peptidase IV enzyme activity to identify transplanted hepatocytes. Immunosuppressive drugs used were tacrolimus (a calcineurin inhibitor) and its synergistic partners, rapamycin (a regulator of the mammalian target of rapamycin [mTOR]) and mycophenolate mofetil (an inosine monophosphate dehydrogenase inhibitor). First, suitable drug doses capable of inducing long-term survival of allografted hepatocytes were identified. In pharmacologically effective doses, rapamycin enhanced cell engraftment by downregulating hepatic expression of selected inflammatory cytokines but profoundly impaired proliferation of transplanted cells, which was necessary for liver repopulation. In contrast, tacrolimus and/or mycophenolate mofetil perturbed neither transplanted cell engraftment nor their proliferation. Therefore, mTOR-dependent extracellular and intracellular mechanisms affected liver replacement with transplanted cells. In conclusion, insights into the biological effects of specific drugs on transplanted cells are critical in identifying suitable immunosuppressive strategies for cell therapy. (HEPATOLOGY 2006;44:410–419.)
Materials and Methods
DPPIV− F344 rats were 8 to 10 weeks old. Cell donors were syngeneic F344 rats (National Cancer Institute, Bethesda, MD) or allogeneic Long-Evans Agouti rats (Special Animal Core, Marion Bessin Liver Research Center). Rats were housed under 14:10-hour light/dark cycles with unrestricted chow and water. The Animal Care and Use Committee at Albert Einstein College of Medicine approved the animal use according to the Guide for the Care and Use of Laboratory Animals (United States Public Health Service Publication, 1996).
Tacro (Fujisawa), Rapa (Wyeth), and MMF (Roche) were purchased commercially. The drugs were uniformly suspended in normal saline. Tacro and Rapa were administered via intraperitoneal injection and MMF was administered via gavage daily. The drugs were started 1 day before cell transplantation and continued for predefined periods. The doses used for final experiments were as follows: Rapa, 1 loading dose of 0.8 mg/kg followed by 0.4 mg/kg/d; Tacro, 2 mg/kg/d; and MMF, 100 mg/kg/d.
To demonstrate Rapa and Tacro blood levels in rats, we used the IMx Sirolimus Assay (Axis-Shield Diagnostics Ltd, Dundee, UK) and IMx Tacrolimus Assay (Abbott Laboratories Diagnostics Division, Abbott Park, IL), according to the manufacturers' instructions.
Cell Isolation and Animal Procedures.
Hepatocytes were isolated via collagenase perfusion. Cells were used only when >80% excluded trypan blue dye. Freshly isolated 1 × 107 viable hepatocytes in 0.5 mL RPMI 1640 medium were injected intrasplenically via a left subcostal incision under ether anesthesia. Animals subjected to two-thirds PH according to the Higgins and Anderson method received 0.5 × 107 cells in 0.25 mL RPMI 1640 medium.
To identify transplanted cells, 5-μm cryosections were obtained from tissue frozen to −70°C and stained for DPPIV activity as described previously.11 To identify tissue integrity and inflammatory cell infiltrates, sections were stained with hematoxylin-eosin.
Transplanted cell numbers in tissues were measured per liver lobule or unit liver volume (cubic millimeters) according to validated methods.12 Briefly, 3 to 4 sections were prepared at depths of >20 μm from each liver lobe, and 100 consecutive liver lobules per section were examined under ×400 magnification in multiple animals per condition. To analyze liver repopulation, 3 to 4 sections from different liver lobes per rat were stained for DPPIV activity. Microphotographs were obtained from adjacent areas under ×40 magnification using a Spot RT digital camera (Diagnostic Instruments Inc., Sterling Heights, MI) and the area occupied by transplanted cells was quantitated using J Image software (National Cancer Institute).12 The extent of liver repopulation was expressed as follows: difference in liver area repopulated between 2 and 3 weeks ÷ difference in liver area repopulated in controls between 2 and 3 weeks × 100.
Phagocytic Capacity of Kupffer Cells.
Rats received 1 × 107 syngeneic hepatocytes with or without prior administration of Tacro plus Rapa and were divided into groups (n = 3 rats each) for sacrifice 6, 12, and 24 hours later. Thirty minutes before sacrifice, 0.1 mL Pelican no. 17 India ink (Hannover, Germany) was injected intrasplenically, and carbon incorporation was analyzed in samples from multiple liver lobes with grading of Kupffer cell phagocytosis in 100 liver lobules per tissue as previously described11: grade 1, minimal carbon; grade 2, moderate carbon; grade 3, maximal carbon incorporation.
DPPIV− rats received 1 × 107 syngeneic hepatocytes with or without prior administration of Tacro plus Rapa. Animals were sacrificed 24 hours after cell transplantation with perfusion-fixation of the liver using glutaraldehyde for electron microscopy with examination of endothelial integrity in 50 consecutive liver lobules as previously described.14
RNA Extraction and Reverse-Transcription Polymerase Chain Reaction.
Liver RNA was extracted with Trizol reagent (Life Technologies) from 2 to 3 DPPIV− rats each 6, 12, and 24 hours after transplantation of F344 cells with unmanipulated rats and rats subjected to laparotomy alone serving as controls. Polymerase chain reaction primers have been published previously for vascular endothelial growth factor, Tie-2, angiopoietin 1, angiopoietin 2, β-actin,15 macrophage inflammatory protein 2, cytokine-induced neutrophil chemoattractant 1, tumor necrosis factor α,16 monocyte chemotactic protein 1,17 and interferon-inducible protein 10.18 RNAs (1 μg) were reverse-transcribed (Omniscript RT PCR kit, Qiagen) and amplified for 35 cycles in 50 μL using 1× polymerase chain reaction buffer, 10 U RT, 100 U RNase inhibitor, 1.25 μg oligo-dT primer, and dNTPs (Platinum PCR kit, Invitrogen). The products were resolved in agarose containing ethidium bromide.
Cryosections of 5 μm thickness were fixed in ice-cold acetone for 10 minutes, blocked in 3% goat serum for 1 hour at room temperature, and incubated with human-specific mouse monoclonal Ki-67 antibody (1:50) (Pharmingen) at 4°C overnight. Color development used goat anti-mouse peroxidase-conjugated immunoglobulin G (1:100) (Sigma) at room temperature for 1 hour followed by DAB substrate (DAKO Plus) for 3 minutes. The fraction of stained hepatocytes was determined in 8 to 10 microphotographs per animal under ×400 magnification.
In Vitro Thymidine Incorporation Assay.
Primary rat hepatocytes at 1 × 105 per 12-well plastic dish were cultured in Dulbecco's modified Eagle medium containing 2% fetal bovine serum and antibiotics. DNA synthesis was stimulated with 20 ng/mL epidermal growth factor (Sigma). Cells in triplicate conditions were incubated 6 hours later with Tacro or Rapa using concentrations previously shown to inhibit DNA synthesis in mixed lymphocyte cultures. After 72 hours, 1 μCi 3H-thymidine was added for 1 hour, and incorporation of thymidine into DNA was analyzed as previously described.19
Data are presented as the mean ± SD. The significance of differences was analyzed via 1-way ANOVA using Scheffe's post hoc multiple comparison method, the Mann-Whitney method for nonparametric data, and the Fisher exact probability test using SPSS software (SPSS Inc., Chicago, IL). A P value of <.05 was considered statistically significant.
We developed experimental protocols to select effective immunosuppressive regimens for promoting survival of allografted hepatocytes and to address whether immunosuppressive drugs would alter transplanted cell engraftment and/or proliferation, as shown in Fig. 1.
An initial goal was to obtain immunosuppressive regimens capable of supporting survival of allogeneic Long-Evans Agouti cells similar to that of syngeneic F344 hepatocytes. Tacro (2 and 4 mg/kg/d), Rapa (0.4 and 0.8 mg/kg/d), and MMF (25, 50, and 100 mg/kg/d) were used either alone or in combination (n = 3 DPPIV− rats each). By themselves, Tacro, Rapa, or MMF in these dosages did not yield allograft survival to 7 days. However, 2 mg/kg Tacro plus 0.4 mg/kg Rapa (after 1 loading dose of 0.8 mg/kg), or 2 mg/kg Tacro plus 100 mg/kg MMF produced allograft survival for over 7 days (Fig. 2). Therefore, we used these conditions to address drug effects in syngeneic F344 cell transplants to avoid allograft-related interference.
To establish the blood levels of drugs, we studied additional DPPIV− rats treated with Rapa alone (0.8 mg/kg loading followed by 0.4 mg/kg/d), Tacro alone (2 mg/kg/d), and Rapa plus Tacro. Blood was obtained for trough levels immediately before and for peak levels 60 to 90 minutes after the fourth drug doses. The peak levels of Rapa and Tacro were similar in rats treated with drugs singly or in combination and were 17.7 ± 3.2 ng/mL (Rapa) and 58.0 ± 12.9 ng/mL (Tacro). The trough levels of the drugs were in the 5- to 10-ng/mL range with no difference in animal groups (Table 1).
|Treatment (n = 3 Rats Each)||Rapa (ng/mL)||Tacro (ng/mL)|
|Rapa alone||8.5 ± 3.2||—|
|Tacro alone||—||5.1 ± 0.9|
|Rapa + Tacro||7.5 ± 3.1||4.9 ± 1.6|
Cell Engraftment Improved in Rats Treated With Rapa Alone or Rapa Plus Tacro.
In untreated controls, we observed 134 ± 25 transplanted cells per 100 liver lobules 3 days after cell transplantation. The transplanted cell numbers increased by 2.1 ± 0.2-fold and 2.4 ± 0.1-fold versus controls in animals treated daily with Rapa or Rapa plus Tacro, respectively (ANOVA with Scheffe post hoc multiple comparison test; P < .05) (Fig. 3). Rats treated with Tacro alone or Tacro plus MMF showed some (albeit nonsignificant) increases in transplanted cell numbers (1.4 ± 0.1-fold and 1.5 ± 0.1-fold vs. controls, respectively; P value not significant).
Improved cell engraftment in Rapa-treated rats was observed within 1 day after cell transplantation without further change subsequently. Because 70% to 80% of transplanted cells are normally phagocytosed within 1 day after cell transplantation,10, 14 Rapa likely interfered with this mechanism, which was also suggested by the greater occupancy of portal vein radicles by transplanted cells. For instance, in untreated control rats, transplanted cells were observed in only 3 ± 3% portal vein radicles after 1 day, and no portal vein radicles contained transplanted cells after 3 days. In contrast, 1 day and 3 days after cell transplantation, rats treated with Rapa plus Tacro showed transplanted cells in 10 ± 8% and 8 ± 6% portal vein radicles, respectively, which was significantly different from the corresponding untreated controls both 1 day and 3 days after cell transplantation (Mann-Whitney U test; P < .05) (Fig. 4).
To identify the mechanisms regulating cell engraftment, we examined hepatic cell–cell interactions. Previous studies indicated that Kupffer cells are activated by cell transplantation and contribute in the early clearance of transplanted cells.11 Analysis of carbon incorporation in Kupffer cells showed that the phagocytotic activity of Kupffer cells in animals treated with Rapa plus Tacro was not perturbed (Fig. 5).
To examine whether drug treatments affected specific cytokine effectors, we performed additional studies. Reverse-transcription polymerase chain reaction revealed alterations in the expression of messenger RNA (mRNA) in inflammatory cytokines following cell transplantation (Fig. 6).
Rapa and Tacro downregulated cell transplantation–induced mRNA-level expression of monocyte chemotactic protein 1, macrophage inflammatory protein 2, and cytokine-induced neutrophil chemoattractant, all of which are released by monocytes or phagocytes, but not tumor necrosis factor α or interferon-inducible protein 10, both of which are major Kupffer cell effectors. The mRNA expression of vascular endothelial growth factor—and its receptors, angiopoietin 1, angiopoietin 2, and Tie-2, which contribute in endothelial remodeling during entry of transplanted hepatocytes in the liver parenchyma—was also not perturbed.14 Electron microscopy established that Rapa and Tacro did not cause hepatic endothelial injury (data not shown), which was consistent with these findings.
Rapa Impairs Transplanted Cell Proliferation.
Retrorsine–PH-conditioned rats provided an excellent assay for studying perturbations in liver repopulation. In these rats, 8 ± 0.8% of the liver was repopulated 2 weeks after cell transplantation. Subsequently, the transplanted cell mass doubled twice between 2 to 3 weeks, with another doubling between 3 to 4 weeks, leading to 52 ± 8% liver repopulation after 4 weeks (Fig. 7).
We treated retrorsine–PH-conditioned rats with immunosuppressive drugs in the 2- to 3-week window after cell transplantation, when transplanted cells underwent exponential proliferation. This strategy helped avoid confounding by perturbation of initial cell engraftment or potential alterations in drug handling, because the pharmacokinetics of immunosuppressive drugs was normal 2 weeks after PH.20
Liver repopulation in rats treated with Tacro alone, MMF alone, or Tacro plus MMF was 28 ± 6%, 25 ± 3%, and 24 ± 5%, respectively, and this was similar to untreated controls (Fig. 7). The sizes of individual transplanted cell foci were also comparable in these conditions. However, liver repopulation was markedly impaired in rats treated with Rapa alone or Rapa plus Tacro (9 ± 3% and 10 ± 2%, respectively) and was significantly less than that in untreated controls (ANOVA using Scheffe test; P < .05). Moreover, the mean size of transplanted cell foci after 3 weeks in Rapa-treated rats approximated that of transplanted cell foci in drug-untreated rats after 2 weeks, which further indicated that transplanted cell proliferation was largely abrogated.
Similarly, 30 hours after PH in DPPIV− rats, 76 ± 5% hepatocytes expressed Ki-67, a nonhistone nuclear protein expressed in cells other than in G0 or early G1 (Fig. 8). In MMF-treated rats, 72 ± 2% hepatocytes expressed Ki-67. However, after Rapa treatment, the Ki-67-expressing cell fraction declined significantly to only 21 ± 2% (Fisher exact test; P < .05).
We further established an effect of Rapa on hepatocellular proliferation. In vitro, DNA synthesis significantly increased in rat hepatocytes stimulated with epidermal growth factor (Fig. 8), whereas this mitogenic response was inhibited by Rapa but not Tacro.
These studies indicated that the mTOR inhibitor, Rapa, had profound effects on transplanted cells. These effects were both positive and negative and led to the promotion of transplanted cell engraftment on the one hand but to the inhibition of proliferation on the other hand. By contrast, calcineurin inhibition using Tacro or inosine monophosphate dehydrogenase inhibition using MMF did not affect the fate of transplanted cells. The combination of immunosuppressive drugs required for preventing rejection of allogeneic hepatocytes in our studies was similar to that required previously in solid organ allografts,21 despite the differences in immune responses elicited by solid organ or cell allografts.22
Although most current immunosuppressive drugs do not specifically target cell subsets regulating the adoptive immune system,23, 24 we found that pharmacological regulation of the innate immune system benefited transplanted cell engraftment. It is noteworthy that the major fraction of transplanted hepatocytes is cleared intravascularly even before cells enter the liver parenchyma.14 The monocyte/phagocyte component of the innate immune system mediates this transplanted cell clearance, which can be decreased by depleting the resident hepatic macrophage (Kupffer cell) pool.11 In this context, Rapa can inhibit dendritic cell activity as well as chemokine-dependent leukocyte migration, both of which are major components of the inflammatory response.25, 26 We found that Rapa impaired hepatic cytokine expression at the mRNA level after cell transplantation, including that of monocyte chemotactic protein 1 mRNA, which mediates monocyte/macrophage chemotaxis, as well as mRNAs of macrophage inflammatory protein 2 and cytokine-induced neutrophil chemoattractant, both of which are potent neutrophil chemoattractants. Cell engraftment in Rapa-treated rats most likely benefited from these perturbations, similar to the depletion of Kupffer cells in earlier studies,11 although our studies did not verify changes in cytokines at the protein level.
Disruption of the hepatic endothelial barrier promotes cell engraftment in the liver, because transplanted cells must migrate across the hepatic endothelium to enter the liver parenchyma.14, 27 Although immunosuppressive drugs may perturb endothelial cell activation and proliferation,8 we found that Rapa and/or Tacro neither injured the hepatic endothelium nor indirectly affected angiogenic activity (e.g., expression of vascular endothelial growth factor and its receptors). This indicated that improved cell engraftment in Rapa-treated rats was independent of endothelial perturbation.
The prominent proliferation–inhibitory effect of Rapa in our in vivo and in vitro assays was in agreement with this property of mTOR inhibitors in various cell types.6–9, 28 The downstream targets of the mTOR complex 1, which is sensitive to Rapa and affects cellular transcription and translation—includes cell cycle regulators such as p21, p27, Rb, and cyclin D, although these interactions are complex.9 Nonetheless, this proliferation-inhibiting effect of mTOR inhibition has been useful in preventing coronary restenosis after stenting and will likely be effective in the treatment of cancer.29, 30 On the other hand, in the context of cell therapy, interference with transplanted cell proliferation during organ replacement, as shown here, will have profound implications. In many situations, large portions of organs must be replaced for optimal therapeutic effects; for example, individuals with liver failure and various metabolic deficiency states will require extensive liver repopulation.
Similarly, our findings should be applicable to additional cell types, where graft function may be regulated by inflammatory cytokine release or where transplanted cell proliferation will be required. This should be especially relevant in cell therapy using allogeneic stem cells, which are likely to be transplanted in small numbers with the expectation that transplanted cells will proliferate soon after transplantation or at some other stage dictated by the underlying conditions capable of initiating, promoting, and sustaining transplanted cell proliferation in given organs.
In conclusion, these considerations suggest that Rapa could be a part of immunosuppression early in cell therapy, because mTOR inhibitors will be beneficial during the phase of transplanted cell engraftment. However, it should be appropriate to avoid Rapa or other mTOR inhibitors during the anticipated period of transplanted cell proliferation, where Tacro and MMF could serve as alternatives. Later, when proliferation in transplanted cells has been completed, Rapa could possibly be used again, if required.
The authors thank Ms. Chaoying Zhang for providing technical assistance.