Hepatocyte Transplantation through the Hepatic Vein: A New Route of Cell Transplantation to the Liver

Abstract

The efficiency of hepatocyte transplantation into the liver varies with the method of administration. This study investigated whether retrograde infusion via the hepatic vein provides a sufficient number of donor cells for the liver. Donor hepatocytes were isolated from dipeptidyl peptidase IV (DPPIV⁺) rats and transplanted into DPPIV^– rat livers either by antegrade portal vein infusion or retrograde hepatic vein infusion. Hepatocyte engraftment ratios and localization were evaluated by histological DPPIV enzymatic staining at 1 week and 8 weeks after the transplantation. No significant differences in engraftment efficiency were observed at either 1 week or 8 weeks after transplantation by either route. However, the localization of the transplanted hepatocytes differed with the administration route. Portal vein infusion resulted in predominantly periportal engraftment, whereas hepatic vein infusion led to pericentral zone engraftment. Immunohistochemical analysis showed that the transplanted hepatocytes engrafted in the pericentral zone after retrograde infusion displayed intense CYP2E1 staining similar to the surrounding native hepatocytes. CYP2E1 staining was further enhanced by administration of isosafrole, an inducing agent for various cytochrome P450 enzymes, including CYP2E1. This study demonstrates a novel approach of transplanting hepatocytes into the liver through retrograde hepatic vein infusion as the means to target cell implantation to the pericentral zone.

Introduction

Cell-based transplantation using isolated hepatocytes is an emerging field of clinical therapeutics for treating liver diseases and disorders (16,18). Since the first hepatocyte transplantation in 1992 (15), more than 80 patients have been treated using this approach (3,7,17,28). In many cases of acute or terminal liver failure, hepatocyte transplantation has been successfully adopted as a temporary measure until liver transplantation can be performed (20,23,28). Accumulating evidence also suggests that hepatocyte transplantation can be therapeutic in various inherited liver-based metabolic disorders (13,17, 20). In some cases, hepatocyte transplantation provided marked clinical improvement by stabilizing or even correcting the pathological phenotype without the need for subsequent liver transplantation (4,23,27).

In current experimental and clinical applications, hepatocytes are transplanted through the portal circulation (14,17,21,29), which is accessed with a catheter by either a percutaneous transhepatic portal puncture (20,27), a surgical insertion through the middle colic or the inferior mesenteric vein (2,13,14,25), or an interventional insertion through the patent umbilical vein in neonatal cases (7,13). All these approaches allow the cells to flow into the liver in an antegrade direction, leading to the stacking of transplanted donor cells in the portal vein radicles. Some of these cells can subsequently translocate into the liver parenchyma (6,26), and, ultimately, the hepatocytes predominantly engraft in the periportal area known as zone 1 (5,26). The portal circulation approach may carry the risks associated with surgical procedures, and can cause temporary portal hypertension and possible embolization (6,21).

A promising alternative method for transplanting liver cells is retrograde infusion into the hepatic vein, which is widely used in interventional radiological procedures. A balloon catheter is placed into the hepatic vein through the jugular vein in order to allow retrograde perfusion of the contrast media for imaging the liver parenchyma with minimally invasive procedures (11,19). Cell transplantation into the liver by retrograde infusion though the hepatic vein may be a viable and safe strategy. Therefore, the present study used rats to investigate the engraftment efficiency of hepatocytes transplanted via retrograde perfusion.

Materials and Methods

Animals

Male dipeptidyl peptidase IV-positive (DPPIV⁺) F344/NSlc rats (8–12 weeks old) were obtained from Japan SLC (Hamamatsu, Japan) and were used as the hepatocyte donors in this study. Male DPPIV-negative (DPPIV^–) F344/DuCrlCrlj rats (8–12 weeks old) were obtained from Charles River Japan (Yokohama, Japan). These DPPIV^– rats have a syngeneic background to the donor rats, and were used as the recipient animals. All animal procedures were conducted in accordance with the institutional guidelines set by the Tokyo Women's Medical University Animal Care Committee. Rats were kept in cages in a temperature-controlled room with a 12-h light/dark cycle and ad libitum access to food and water.

Hepatocyte Isolation and Purification

Hepatocytes were isolated and purified from the DPPIV⁺ rats by using a modified two-step collagenase perfusion method as previously described (24). Cells were resuspended in DMEM (Sigma-Aldrich, St. Louis, MO) at 2 × 10⁷ cells/ml, and cell viability was determined by trypan blue exclusion. In this study, experiments were conducted only when hepatocyte purity and viability exceeded 98% and 90%, respectively.

Hepatocyte Transplantation

Hepatocytes from donor DPPIV⁺ rats were transplanted into recipient DPPIV^– rat livers (5 × 10⁴ cells/g rat body weight) through two different routes: the portal vein (antegrade) or the hepatic vein (retrograde). For the portal vein method, hepatocytes were infused at the rate of 0.5 ml/min following the insertion of a 27-gauge needle. For the hepatic vein method (Fig. 1A), the infrahepatic vena cava (IHVC) and suprahepatic vena cava (SHVC) were isolated from the surrounding tissues, and rubber bands were placed around the veins. An 18-gauge plastic catheter was inserted into the vena cava from beneath the isolated IHVC, and the catheter tip was placed between the isolated SHVC and IHVC. Prior to cell infusion, the SHVC and IHVC were temporarily clamped by tying the rubber bands to avoid infusion into vessels other than the hepatic veins. Cells were infused through the catheter, followed by an additional infusion of saline (5 ml) for 60 s. Immediately after the saline infusion, the inserted catheter and the rubber bands were removed from the SHVC and the IHVC to allow blood flow the liver. The infusion was done within 60 s to minimize possible liver damage. Sham operations were also performed using 0.5 ml cell-free saline instead of 0.5 ml hepatocyte suspension.

Figure 1.

Hepatocyte transplantation through the hepatic vein in rats. (A) Schematic illustrating hepatic vein transplantation. The suprahepatic vena cava (SHVC) and infrahepatic vena cava (IHVC) were isolated from the surrounding tissue, and rubber bands were placed around the veins (“clamping point”). An 18-gauge angiocatheter was inserted into the vena cava from beneath the IHVC, and the catheter tip was placed between IHVC and SHVC. IHVC and SHVC were temporarily clamped with the rubber bands. Hepatocytes were infused through the catheter for 60 s at a dose of 5 × 10⁴ cells/g body weight (total volume 0.5 ml), followed by an additional infusion of 5 ml of saline for 60 s. (B–E) Live images of the retrograde hepatocyte transplantation procedure. SHVC (B) and IHVC (C) were temporarily clamped by rubber bands after the insertion of the angiocatheter from IHVC (D). (E) Upon cell and saline infusion, the liver gradually swelled and turned slightly red. PV, portal vein. Scale bars: 1 cm.

Biochemical Analysis of Rat Serum

Blood samples were obtained from the retro-orbital plexus of the recipient rats at 3 and 7 days after the transplantation. After centrifugation, serum samples were analyzed for aspartate aminotransferase (AST), alanine aminotransferase (ALT), total bilirubin (T-Bil), and serum albumin (ALB) levels with the clinical analyzer DRI-CHEM 3500i (Fujifilm, Tokyo, Japan) to determine the extent of liver injury related to the cell transplantation.

Histological Analyses of Transplanted Hepatocytes

Recipient rats were sacrificed 24 h, 1 week, or 8 weeks after transplantation. Three separated liver lobes (medial, left lateral, and right lateral lobes) and the lungs were rapidly frozen in OCT compound (Sakura Finetek, Tokyo, Japan). DPPIV enzymatic staining was performed using 5-μm sections from the frozen specimens as previously described (1,6,26). The engraftment ratios of the donor hepatocytes in the recipient liver were determined by counting the number of DPPIV⁺ cells and expressed as the percentage of DPPIV⁺ cells per total number of DPPIV^– and DPPIV cells (approximately 24,000 cells from 30 random areas of the specimens from three liver lobes). The ratio of DPPIV⁺ hepatocytes in a specific area was also obtained. The locations of the engrafted hepatocytes were marked as liver zone 1 (periportal venous area), zone 2 (intermediate area), or zone 3 (pericentral venous area) (21,22).

Cytochrome P450 Induction

To investigate the zone-specific drug-metabolizing function of the engrafted hepatocytes, recipient rats were given IP isosafrole (150 mg/kg body weight; Wako Pure Chemical, Osaka, Japan) dissolved in corn oil as previously described (n = 5) (9). Injections were performed for 3 consecutive days starting 12 days after transplantation. Control rats were also treated with corn oil or isosafrole as described above (n = 5). All the rats were sacrificed 24 h after the final injection. Liver pieces were resected from the three separate liver lobes of the recipient rats for double immunohistochemical analysis of cytochrome P450 subtype 2E1 (CYP2E1) and DPPIV (12). Frozen 5-μm-thick sections were fixed with cold acetone and incubated with rabbit anti-rat CYP2E1 antibody (1:200; Millipore, Billerica, MA, USA) and purified mouse anti-rat CD26 antibody (BD Biosciences, San Jose, CA, USA). For fluorescence immunohistochemistry, Alexa Fluor 488 goat anti-mouse IgG (Invitrogen, Carlsbad, CA, USA) and Alexa Fluor 594 goat anti-rabbit IgG (Invitrogen) were used as the secondary antibodies. Nuclear DNA was stained with 4′,6-diamidino-2-phenylindole (DAPI).

The stained samples were digitally photographed, and the intensity of CYP2E1 staining in the hepatocytes was evaluated by an image analysis software (Scion Image; Scion Corporation, Frederick, MD). Fluorescence intensity was measured in 10 randomly selected fields of zone 1, 2, and 3 in each rat (a total of 30 fields/rat).

X-Ray Computed Tomography

To rule out lung embolism and infarction, contrast-enhanced computed tomography (CT) scanning was performed 24 h after hepatocyte transplantation. A total volume of 3 ml of contrast medium (Iomelon350; Eizai, Tokyo, Japan) was administered by IV injection through the animal's tail vein at a flow rate of 0.2 ml/s. The CT images were obtained on a micro-X-ray CT device used for laboratory animals (R-mCT; Rigaku, Tokyo, Japan) immediately after the contrast injection under inhalation anesthesia.

Statistical Analysis

All values calculated in the present study are mean (SD). The significance of the differences among the groups was tested by one-way or two-way ANOVA followed by either Dunnett's or Tukey-Kramer post hoc test using the Statview 5.0 software (SAS Institute, Cary, NC, USA). A probability value of p < 0.05 was considered to be statistically significant.

Results

Retrograde Hepatocyte Transplantation

As shown in Figure 1A, a catheter was inserted between the SHVC and IHVC, and the vessels were temporarily clamped with rubber bands to selectively infuse the donor hepatocytes into the liver through the hepatic veins in a retrograde manner. As the infusion progressed, the liver swelled and became light red, indicating that the infused solution was entering the liver from the hepatic veins (Fig. 1B–D). Immediately after the clamps were removed, the blood flow to the liver returned to normal (data not shown), and the rats recovered within several minutes after the surgical procedure. None of the rats died during this retrograde infusion procedure.

To determine any adverse effects on the liver from the surgical procedure, the serum levels of AST, ALT, and T-Bil were determined at day 0 (prior to the infusion), and 3 and 7 days after hepatocyte transplantation through both the antegrade and retrograde procedures (Fig. 2). Acute transient injury was detected by elevated AST levels on day 3 regardless of the transplantation procedure, which is consistent with previous studies (31). The AST values returned to normal by day 7 in all the groups. No impairment of liver function, represented by the ALT, ALB, and T-bil levels, was observed in any of the groups (Fig. 2).

Figure 2.

Assessment of liver injury and function after hepatocyte transplantation. Liver injury and function were assessed before (day 0), and 3 and 7 days after hepatocyte transplantation by measuring serum aspartate aminotransferase (AST), alanine aminotransferase (ALT), total bilirubin (T-Bil), and albumin (ALB). Gray bars indicate control sham operation (physiological cell-free saline injected into hepatic vein) group; open bars, antegrade (portal vein) hepatocyte transplantation group; black bars, retrograde (hepatic vein) hepatocyte transplantation group. The number of samples of each group is 6–7. ^*p < 0.05 compared to day 0 by two-way ANOVA followed by multiple comparisons.

Engraftment of Hepatocytes Transplanted by Retrograde Infusion

To determine the feasibility of hepatocyte transplantation using retrograde infusion via the hepatic vein, engraftment of the transplanted cells was evaluated 1 week after transplantation. At that time, histological examination revealed that the donor hepatocytes transplanted via antegrade infusion were predominantly engrafted in the liver parenchyma at the periportal venous area (zone 1) (Fig. 3A–D). In contrast, the hepatocytes transplanted via retrograde infusion were located in the pericentral venous area (zone 3) (Fig. 3A–D). The engraftment ratios of the transplanted hepatocytes were 0.28 ± 0.11% and 0.30 ± 0.13% in the antegrade and retrograde infusion groups, respectively (Fig. 3E); no significant differences were observed between these two groups. The engraftment ratio was calculated as the number of engrafted donor hepatocytes (DPPIV⁺ hepatocytes) divided by the total number of hepatocytes (both recipient and donor, DPPIV^– and DPPIV⁺ hepatocytes) in each rat. A possible procedure-dependent difference in the zonal localization of hepatocyte engraftment was also examined. Figure 3F shows a marked difference in the zonal distribution of engrafted hepatocytes depending on the transplantation procedure. Most (75.8%) of the engrafted hepatocytes infused through the portal vein were found within zone 1. In contrast, 69.0% of the engrafted hepatocytes transplanted through the hepatic vein localized to zone 3.

Figure 3.

Engraftment status of the transplanted hepatocytes in the liver following the antegrade and retrograde hepatocyte transplantation. Livers were examined 1 week (A–F) or 8 weeks (G–J) after hepatocyte transplantation. Engrafted hepatocytes in the recipient livers visualized by DPPIV staining (A–D, G, H). Representative DPPIV enzymatic staining of the recipient livers following antegrade (A, C, G) or retrograde (B, D, H) infusion. Arrows indicate engrafted DPPIV⁺ hepatocytes. (E, I) Engraftment ratios of the transplanted hepatocytes 1 week (E) and 8 weeks (I) after antegrade (open bars) and retrograde (black bars) infusion. Values were determined by counting DPPIV⁺ hepatocytes per 20,000 total hepatocytes from three different liver lobes/rat (n = 8 per group). (F, J) Localization of engrafted hepatocytes in the liver. Transplantation procedure-dependent localization of the engrafted DPPIV⁺ donor hepatocytes was assessed following antegrade (open bars) and retrograde (black bars) after 1 week (F) and 8 weeks (J). The locations of the engrafted hepatocytes were recorded as zones 1, 2, or 3. The majority of DPPIV⁺ hepatocytes localized to the periportal venous area (zone 1) in the antegrade infusion group (A, C, F, G, J), while the majority of DPPIV⁺ hepatocytes were found in the pericentral venous area (zone 3) in the retrograde infusion group (B, D, F, H, J). PV and CV indicate portal and central veins, respectively. Scale bars: 100 μm. ^*p < 0.05 between groups compared by one-way ANOVA with multiple comparisons.

Long-Term Survival of the Hepatocytes Transplanted by Hepatic Vein Infusion

To determine the long-term survival of the engrafted hepatocytes, the recipient animals' livers were dissected 8 weeks following antegrade or retrograde infusion. At week 8, DPPIV⁺ transplanted hepatocytes were detected in the recipient liver parenchyma in both the antegrade and retrograde infusion groups (Fig. 3G, H). The engraftment ratios of the antegrade and retrograde groups were 0.26 ± 0.05% and 0.35 ± 0.15%, respectively (Fig. 3I). The hepatocytes transplanted by antegrade and retrograde infusion were found predominantly in zones 1 and 3, respectively. No apparent alteration of hepatocyte localization was seen 8 weeks after transplantation (Fig. 3J).

Conservation of the Drug-Metabolizing Function of the Transplanted Hepatocytes

Metabolic zonation of the liver is characterized by nonuniform distribution of metabolic enzymes along the portal–central axis (12). Xenobiotic metabolism occurs predominantly in the hepatocytes located in zone 3. Cytochrome P450 2E1 (CYP2E1) is an important enzyme involved in xenobiotic metabolism. We first investigated the expression of this enzyme in the livers of animals who did not undergo transplantation. CYP2E1 expression was highest in zone 3 in nontreated rats, and formed a decreasing gradient from zone 3 to zone 1 (Fig. 4A). CYP2E1 expression levels were similar in the livers of rats treated with corn oil and untreated rats (Fig. 4B). In marked contrast, treatment with the CYP2E1-inducing agent isosafrole strongly increased CYP2E1 expression, particularly in zone 3 (Fig. 4C). Quantifying the fluorescence intensity of CYP2E1 with the Scion Image software also revealed a clear increase in intensity in zones 2 and 3 in isosafrole-treated rats compared to ones that did not receive isosafrole (Fig. 4D).

Figure 4.

Induction of cytochrome P450 2E1 (CYP2E1) in the liver and engrafted donor hepatocytes. Animals were treated with corn oil or isosafrole in corn oil for 3 consecutive days, and liver samples were obtained 24 h after the last inoculation. (A–D) CYP2E1 expression in the livers of normal rats (no transplantation). CYP2E1 immunofluorescent staining (red) in the liver of normal rats untreated (A), treated with corn oil (B), or treated with isosafrole (in corn oil) (C). (D) CYP2E1 expression levels of hepatocytes in the three liver zones assessed by measuring the CYP2E1 immunostaining fluorescence intensity using Scion Image. Normal rats treated with corn oil (open bars) or isosafrole (black bars). Ten randomly selected locations of zones 1, 2, and 3 per rat were scanned from individual rats (n = 5). (E–M) CYP2E1 expression of recipient hepatocytes and engrafted donor hepatocytes after hepatocyte transplantation. Treatment with corn oil (E, F, I, J) or isosafrole (G, H, K, L) began 12 days after hepatocyte transplantation. (E–H) Immunofluorescent DPPIV (green) and CYP2E1 (red) staining in the antegrade infusion group. (I–L) Immunofluorescent DPPIV (green) and CYP2E1 (red) staining in the retrograde infusion group. DPPIV staining (E, G, I, K); DPPIV and CYP2E1 staining (F, H, J, L). (M) CYP2E1 expression levels of recipient hepatocytes and engrafted donor hepatocytes in the three liver zones assessed by measuring the CYP2E1 immunostaining fluorescence intensity using Scion Image. Recipient native hepatocytes (open bars), hepatocytes engrafted in zone 1 in the antegrade infusion group (gray bars), and zone 3-engrafted hepatocytes in the retrograde infusion group (hatched bars). Values were expressed by scanning the fluorescence intensity of CYP2E1 at 20 randomly selected locations per rat (10 for engrafted hepatocytes and 10 for native hepatocytes from individual recipient rats in both antegrade and retrograde infusion groups) (n = 5). Oil: corn oil administration group; ISO: isosafrole administration group. PV and CV indicate portal and central veins, respectively. Scale bars: 200 μm (A–C), 50 μm (E–L). ^*p < 0.05 compared between the groups by one-way ANOVA with multiple comparisons.

DPPIV⁺ donor hepatocytes that localized to zone 1 after infusion through the portal vein were weakly positive for CYP2E1, similar to the surrounding native hepatocytes regardless of the isosafrole treatment (Fig. 4E–H). In contrast, DPPIV⁺ hepatocytes engrafted in zone 3 showed stronger CYP2E1 staining (Fig. 4I, J) compared to the donor hepatocytes in zone 1 (Fig. 4E, F). To evaluate the functionality of hepatocytes engrafted in zone 3, which takes up circulating xenobiotics, we injected isosafrole into the IP space of recipient rats 2 weeks after transplantation. The hepatocytes engrafted in zone 3 demonstrated strong CYP2E1 staining similar to that of the surrounding native hepatocytes (Fig. 4K, L). Analyzing the fluorescence intensity of CYP2E1 in livers of recipient rats with Scion Image revealed that hepatocytes engrafted in zone 3 by retrograde infusion had stronger CYP2E1 staining and a higher degree of isosafrole induction than donor hepatocytes transplanted via antegrade infusion (Fig. 4M). Furthermore, fluorescence intensity analysis showed that CYP2E1 expression in hepatocytes transplanted by retrograde infusion was equivalent to that of the surrounding native hepatocytes in the liver of rats treated with isosafrole (Fig. 4M).

These results indicate that donor hepatocytes engrafted in zone 3 after retrograde infusion were able to survive long term and to maintain drug-metabolizing activity similar to that of the native hepatocytes.

Complications and Side Effects of Retrograde Infusion

Some donor hepatocytes transplanted via retrograde infusion are likely to flow into the lungs through the inferior vena cava (IVC). Therefore, we investigated whether pulmonary embolism or infarction occurred 24 h after retrograde infusion. Histological staining of the sagittal sections of the lung for DPPIV after retrograde infusion showed approximately 20 donor hepatocytes in the lung (Fig. 5A, B). However, H&E staining revealed no karyolysis around the transplanted hepatocytes, indicating that no pulmonary infarction occurred (Fig. 5C, D). The number of donor hepatocytes in the lung decreased to four to five 1 week after retrograde infusion (Fig. 5E, F).

Figure 5.

Detection of hepatocyte migration into the lung after retrograde hepatocyte infusion. (A–D) Histological images of the lung 24 h after retrograde hepatocyte transplantation. DPPIV staining for donor hepatocytes (A, B) and H&E staining (C, D). (E, F) Histological images of the lung 1 week after retrograde infusion. DPPIV staining for donor hepatocytes (E) and H&E staining (F). Arrows indicate engrafted hepatocytes (D). Scale bars (black): 100 μm (A–F). Contrast-enhanced CT image of a normal rat (no transplantation) (G), and 24 h after retrograde infusion (H). L.F., B, H, S, and V indicate lung field, bronchus, heart, spine and vena cava, respectively. Scale bars (white): 1 cm.

We also investigated the possible occurrence of embolism or infarction with contrast-enhanced computed tomography (CT) 24 h after hepatocyte transplantation. Compared with control rats, no circulation abnormalities and no pulmonary embolism or infarction were found in the lung 24 h after retrograde hepatocyte infusion (Fig. 5G, H).

Discussion

The present study analyzed a novel cell transplantation approach that allowed hepatocytes to be infused in a retrograde manner through the hepatic vein to the liver parenchyma. Our results show that retrograde infusion of hepatocytes resulted in cellular engraftment predominantly in zone 3 of the liver, an area of robust xenobiotic metabolism. The hepatocytes engrafted in zone 3 survived for up to 8 weeks after transplantation, and were functionally capable of xenobiotic uptake and CYP2E1 expression.

Previous hepatocyte transplantations for experimental and clinical applications have mostly been delivered by antegrade infusion through the portal vein (7,14,17,21,25,29). Antegrade infusion results in hepatocyte accumulation at the periphery of the portal pedicles, and some of the hepatocytes enter the hepatic sinusoids and integrate with the parenchymal plates (5,26). The results of our study are consistent with these previously published studies documenting donor hepatocyte engraftment predominantly in zone 1 (5,26).

In contrast, retrograde infusion of hepatocytes through the hepatic vein resulted in engraftment predominantly in zone 3. The infused hepatocytes were speculated to enter into the periphery of the hepatic vein and subsequently migrate into the hepatic sinusoid for transport into the parenchymal plates. Engrafted hepatocytes transplanted by retrograde infusion survived for up to 8 weeks, indicating that our procedure may be a viable clinical cell transplantation approach. Moreover, zone 3 of the liver is an active site for xenobiotic metabolism (12), and zone 3 hepatocytes have higher levels of many members of cytochrome P450 compared to the hepatocytes in zones 1 and 2 (9,12). Although the mechanisms underlying the differences in xenobiotics metabolism among the 3 zones are unknown, the histological structure and different arrangement of the liver sinusoidal endothelial cells could account for this disparity (10). Zone 3 sinusoidal endothelial cells have larger and more numerous fenestrae than do cells in zones 1 and 2 (10,30). These zone-dependent characteristics of the fenestrae allow the zone 3 hepatocytes to contact and uptake xenobiotic compounds at higher rates than their counterparts in zones 1 and 2. The periportal-to-perivenous gradients of oxygen, hormones, and nutrients carried by the blood are also associated with the functional differences of the hepatocytes in the zones of the liver (10). In this study, the hepatocytes engrafted in zone 3 (transplanted through the hepatic vein) expressed levels of CYP2E1 similar to the levels expressed by surrounding native hepatocytes in response to isosafrole. On the other hand, hepatocytes engrafted in zone 1 (transplanted through the portal vein) expressed much lower levels of CYP2E1. These results show that the target localization of hepatocyte engraftment should be considered according to the function desired from the engrafted cells. Our findings indicate that retrograde infusion of hepatocytes could have an advantage over antegrade infusion in treating liver diseases or disorders that require high rates of drug metabolism.

Some transplanted hepatocytes could flow into the IVC and the lungs after retrograde infusion. However, histological examination and contrast-enhanced CT revealed no pulmonary embolism or infarction in the lungs. Since blood from the bronchial artery can often sustain the lung parenchyma despite obstructions in the pulmonary arterial system, blocking the pulmonary artery with hepatocytes or small blood clots is unlikely to cause an infarction (8). We found that a small numbers of transplanted hepatocytes migrated to the lung, but that no significant complications, such as pulmonary infarction, occurred. Large animal studies are needed to evaluate the safety of retrograde infusion in clinical applications.

The present study proposes a new route for hepatocyte transplantation though the hepatic vein. We consider an occlusion balloon catheter to be a valuable and feasible addition to the hepatic vein transplantation procedure, because the catheter can be inserted into the hepatic vein through the internal jugular vein and the superior vena cava. The temporary occlusion created by the catheter allows for the retrograde infusion of donor cells from the hepatic veins to the liver parenchyma without leakage, minimizing cell migration into the lungs (11). In fact, Yoshino et al. (32) successfully delivered naked plasmid DNA via retrograde infusion with a balloon catheter to promote selective liver transfection in pigs. Further studies and improvements, and careful risk assessments, are required before our retrograde infusion technique can be used in clinical applications. The number of hepatocytes to be used in infusion, the flow rate and pressure of the injected solution, and the method for regulating portal vein pressure all require careful optimization. Currently, cell transplantation into the liver has gained attention, and such techniques are being used to transplant islet cells to treat type 1 diabetes mellitus. These procedures are generally conducted using antegrade infusion through the portal vein. Our findings will be helpful in future efforts to transplant islet cells via retrograde infusion to minimize liver injury and preserve the function of the donor cells.

In conclusion, ours is the first report to successfully demonstrate that retrograde infusion into the liver can be a viable method of cell transplantation. We demonstrate that retrograde infusion provides efficient hepatocyte engraftment into zone 3 of the liver parenchyma. In current liver-targeted cell therapies, including islet and hepatocyte transplantations, cells are traditionally infused into the liver through the portal vein, but the findings of this study show that similarly effective cell transplantation can be achieved in a less invasive manner. Although further studies will be required to address the therapeutic potential of retrograde infusion in various diseases, this new transplantation approach is an important step in the further development of innovative strategies for hepatocyte and islet cell transplantation therapies.