https://orcid.org/0000-0002-1802-7565
Abstract
Although organ shortage is a rising problem, organs from hepatitis C virus (HCV) ribonucleic acid (RNA)-positive donors are not routinely transplanted in HCV-negative individuals. Because HCV only infects hepatocytes, other organs such as kidneys are merely contaminated with HCV via the blood. In this study, we established a protocol to reduce HCV virions during the cold ischemic time.
Standard virological assays were used to investigate the effect of antivirals, including methylene blue (MB), in different preservation solutions. Kidneys from mini pigs were contaminated with Jc1 or HCV RNA-positive human serum. Afterwards, organs were flushed with MB. Hypothermic machine perfusion was used to optimize reduction of HCV.
Three different antivirals were investigated for their ability to inactivate HCV in vitro. Only MB completely inactivated HCV in the presence of all perfusion solutions. Hepatitis C virus-contaminated kidneys from mini pigs were treated with MB and hypothermic machine perfusion without any negative effect on the graft. Human liver-uPA-SCID mice did not establish HCV infection after inoculation with flow through from these kidneys.
This proof-of-concept study is a first step to reduce transmission of infectious HCV particles in the transplant setting and might serve as a model for other relevant pathogens.
Approximately 71 million individuals are chronically infected with the hepatitis C virus (HCV) worldwide [1]. Today, a complete cure of HCV infection by antiviral treatment with direct-acting antivirals (DAAs) can be achieved in more than 95% of the patients [2]. Hepatitis C virus has a very narrow host range. Naturally, only humans and chimpanzees can be infected. Within the infected host, hepatocytes are the primary reservoir because the viral replication cycle depends on several host factors that are exclusively expressed in the liver. Even if viral ribonucleic acid (RNA) was found in tubular epithelial cells and peripheral blood mononuclear cells, several years ago Marukian et al [3] clearly showed that multiple blocks prevented blood cells from supporting HCV infection. In contrast, Chen et al [4] identified CD86 as a coreceptor of lymphotropic HCV. However, distinguishing RNA association from true HCV replication has been problematic, and multiple artifacts complicate detection of the replicative intermediate minus-strand RNA [3]. Therefore, other organs, such as kidneys and hearts, are contaminated with HCV via blood rather than actively infected.
Kidney transplantation is a standard procedure for several end-stage kidney diseases that lead to dialysis. As a result, the wait time of for transplantation, approximately 4–7 years, increases. This situation has led researchers to consider alternative strategies to overcome organ shortage. One of these strategies is to use grafts from patients with infectious diseases, such as replicative HCV infection in certain cases [5]. However, in the United States, more than 500 kidneys from HCV RNA-positive donors are discarded every year [6–8]. In a recent study, Goldberg et al [9] reported a pilot trial of kidney transplantation from HCV RNA-positive donors into 10 uninfected individuals. All recipients developed an acute HCV infection and were then successfully treated with DAAs. This trial shows that transplantation of HCV genotype 1-positive kidneys into HCV-negative recipients followed by antiviral therapy can provide cure of HCV infection. However, antiviral treatment with DAAs for several weeks is expensive and allograft failure might lead to a limited choice of therapeutics. The optimal way to solve these problems is to avoid HCV infection by reducing the viral load in the graft.
In this study, we were able to show that methylene blue (MB)—a well known and low-priced agent for pathogen inactivation of blood products—is able to inactivate HCV in perfusion solutions (PS) in vitro. In addition, MB treatment in combination with hypothermic machine perfusion (HMP) reduces HCV in contaminated kidneys from mini pigs. The reduced infectivity after the procedure was confirmed using human liver-uPA-SCID mice. Taken together, this approach is a first step to reduce transmission of infectious HCV particles in the kidney transplant setting and might serve as a model for other relevant pathogens.
MATERIAL AND METHODS
Organ Perfusion and Hepatitis C Virus Contamination
Immediately after nephrectomy, cold (4°C) perfusion (~5 mL/gram organ weight) of histidine-tryptophan-ketoglutarate (HTK) was done by isolated arterial perfusion using buttoned cannula fixed to the renal arteries. Afterwards, kidneys were contaminated with either HCVcc or HCV RNA-positive serum (genotype 2b, 6.5 million IU/mL) by arterial perfusion and additional incubation for 1 hour to allow attachment of HCV at 4°C.
Hypothermic Machine Perfusion
In some experiments, HMP was initiated by connecting the kidneys to the LifePort Kidney Transporter (Organ Recovery Systems). As previously reported, pulsatile flow of HTK or Kidney Preservation Solution 1 at 2–4°C with a fixed systolic perfusion pressure of 30 mmHg was maintained. The intended pump time was 120 minutes.
Perfusion parameters such as flow, renal resistance, and pressure were recorded and analyzed afterwards. Graft integrity was tested thereafter by isolated reperfusion ex vivo via the renal artery as previously described [10].
RESULTS
Methylene Blue Inactivates Hepatitis C Virus in the Presence of All Perfusion Solutions
The most obvious method to inactivate HCV particles in the blood and reduce the risk of HCV transmission in the context of organ transplantation might be the addition of an antiviral agent during the cold ischemia time. Two different drugs targeting the viral particle were chosen for this study because they have previously been reported to have antiviral activity against HCV but are expected to act via different mechanisms. All antivirals were tested in the presence of different perfusion solutions to rule out that these solutions might alter the antiviral effects of the drugs.
First, we infected the cells in the presence of Epigallocatechin gallate (EGCG), a catechin from green tea that has been shown to inhibit HCV attachment [11]. Although 10 µg/mL EGCG diminished HCV infectivity to background levels in the presence of UW Belzer and Celsior, much higher concentrations of EGCG were needed in all other PS (Figure 1A). In some PS (eg, HTK-N), even addition of 50 µg/mL EGCG did not significantly reduce HCV infectivity.
![Antiviral effect of Epigallocatechin gallate (EGCG) (A) and methylene blue (MB) (B) on hepatitis C virus replication in the presence of perfusion solutions. Huh-7.5 cells were infected with Luc-Jc1 in the presence of several perfusion solutions (1:1) and indicated concentrations of the antivirals EGCG and methylene blue for 4 hours at 4°C. Then, the infections mixture was exchanged for standard medium and the cells were incubated at 37°C. Infection was stopped by cell lysis, and luciferase activity (relative luciferase units [RLU]/well) was measured after 48 hours (B) or 72 hours (A). Bars represent the mean ± standard deviation of 3 individual experiments performed in duplicate. In addition, cytotoxicity of EGCG and MB was determined using the Rotitest Vital assay (A and B, right). Abbreviations: DMEM, Dulbecco’s modified Eagle’s medium; HTK, histidine-tryptophan-ketoglutarate; IGL, Institute Goeorges Lopez.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/jid/218/11/10.1093_infdis_jiy386/1/m_jiy38601.jpeg?Expires=1748095977&Signature=stMqFLaG~g3UHCu-8g7pspVgnmKx1ECkvM2mLVGpGnfwD71tkdWS8qRL77u4obML-tH9rIUi~ahPOg~-1J0XTzEF0yfkgJg3DcFBg1lKAMG3CZ6xHqUraNoWzUJ~j8QmgEw4LLgiLBdDQDKznB2k3J24n6UWVe9vE6aZCWOpDTdUQxkD0aelbH1ePtx20ul16uDTMZEYO0dKPLQo4nr6IfD~w7TR5RmqIjXSEQcd~FsOAcx3eAPy4WVyNkMkClWaRKUVYPup2PnBp6nAjUhTHnuGMSbjIiNeEuEPcNVYsOiWzK4QBqym4REbUnv8ZtD57mdn68V0~TlNMMira6zfCg__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Antiviral effect of Epigallocatechin gallate (EGCG) (A) and methylene blue (MB) (B) on hepatitis C virus replication in the presence of perfusion solutions. Huh-7.5 cells were infected with Luc-Jc1 in the presence of several perfusion solutions (1:1) and indicated concentrations of the antivirals EGCG and methylene blue for 4 hours at 4°C. Then, the infections mixture was exchanged for standard medium and the cells were incubated at 37°C. Infection was stopped by cell lysis, and luciferase activity (relative luciferase units [RLU]/well) was measured after 48 hours (B) or 72 hours (A). Bars represent the mean ± standard deviation of 3 individual experiments performed in duplicate. In addition, cytotoxicity of EGCG and MB was determined using the Rotitest Vital assay (A and B, right). Abbreviations: DMEM, Dulbecco’s modified Eagle’s medium; HTK, histidine-tryptophan-ketoglutarate; IGL, Institute Goeorges Lopez.
The second compound we tested was MB. Methylene blue has antimalarial, antiviral, and antibacterial activity [12–14]. It is or was in clinical use for treatment of methemoglobinemia, cyanide poisoning, urinary tract infection, or ifosfamide toxicity [12]. In addition, it is used to reduce or inactivate pathogens in blood products. In some European countries, inactivation with 1 µM MB is routinely used for single-donor plasma and platelets [15]. When we added increasing amounts of MB during HCV infection, we confirmed that 1 µM MB decreased HCV infectivity to background levels in all PS (Figure 1B). To rule out any cytotoxic effects, a commercial cytotoxicity assay was conducted side by side. As shown in Figure 1, right, both drugs showed no significant cytotoxic effects.
Characterization of the Antiviral Effects of Methylene Blue
As we decided to continue our decontamination experiments with MB, we further characterized the conditions of MB treatment. To exclude that the antiviral effect of MB is different if the organ is heated from 4°C to 37°C, the antiviral effect was investigated side by side at 4°C and 37°C (Figure 2A and B). We confirmed that the concentrations of 1 µM MB effectively inhibited HCV infectivity independent from the temperature. In addition, cytotoxicity of MB at 4°C and 37°C was investigated. Although we were unable to detect any significant cytotoxicity up to 5 µM MB, a concentration of 10 µM was associated with reduced cell viability.
Characterization of the antiviral effects of methylene blue (MB). (A and B) Hepatitis C virus (HCV) infection in the presence of MB at 4°C (A) and 37°C (B). Luc-Jc1 virus was added to complete Dulbecco’s modified Eagle’s medium (DMEM) supplemented with MB to achieve the indicated concentrations. The medium was exchanged after 4 hours. Huh7-5 cells were lysed, and luciferase activity was determined after 48 hours at 37°C. In addition, cytotoxicity of Epigallocatechin gallate and MB was determined using the Rotitest Vital assay (A, right). Mean values ± standard deviation (SD) of 3 independent experiments performed in duplicate are depicted. (C) Hepatitis C virus infection in the presence of MB and histidine-tryptophan-ketoglutarate (HTK). Luc-Jc1 supplemented with MB was then mixed with HTK (1:1). After 10, 30, and 60 minutes, the infection mixture was added to Huh-7.5 cells at an initial dilution of 1:5 and was then serially diluted (1:3). The medium was exchanged for complete DMEM after 4 hours of incubation at 37°C. After 72 hours, Huh-7.5 cells were fixed with methanol and stained with a NS5A primary antibody. Infected wells were counted, and the median tissue culture infective dose (TCID50/mL) was calculated. Mean values ± SD of 1 representative experiment performed in 6 technical replicates are depicted. Experiment was performed 3 times with 6 replicates each. (D) Infectivity of different HCV genotypes in the presence of 1 µM MB and histidine-tryptophan-ketoglutarate (HTK). Huh7-5 cells were used. Data were normalized to the respective HTK control. Mean percentages ± SD are depicted. Data represent 3 independent experiments. * p < 0.05, **p < 0.01, ***p < 0.005. Abbreviations: ns, not significant; RLU, relative luciferase units.
To address the question of whether complete decontamination of HCV can only be reached if MB is present for at least 5 hours or if a shorter incubation time might be sufficient, we added MB for 10, 30, or 60 minutes during infection and determined the HCV infectivity via median tissue culture infective dose (TCID50) endpoint dilution assay (Figure 2C). Although addition of MB for 10 and 30 minutes did not inhibit HCV infectivity completely, 60-minute incubation with MB was effective if more than 1 µM MB was used in the assay.
All previous experiments were performed with Jc1, an intra-genotypic genotype 2a chimera. To ensure that MB has the same antiviral effects against all HCV genotypes, we tested different chimeric reporter viruses with the structural proteins of HCV genotype 1–7. As shown in Figure 2D, MB was able to reduce infectivity of all HCV genotypes (Figure 2D). Taken together, MB proved to be a very effective antiviral against HCV under various conditions.
Exclusion of Methylene Blue Toxicity Under Preservation Conditions
Although MB is widely applied clinically and known to be largely nontoxic [12–14], it is also known to be a redox-cycling compound and, in conjunction with light, a singlet oxygen-sensitizer [16–18]. Thus, it might also exert pro-oxidative effects and cell injury. Because cell injury elicited by hypothermia is mediated by an iron-dependent formation of reactive oxygen species [19–21], which might be enhanced by redox-cycling/pro-oxidative compounds, we opted to exclude MB toxicity under hypothermic conditions. To this end, porcine kidney epithelial cells of the cell line LLC-PK1 were exposed to 1 µM and/or 10 µM MB under normothermic control conditions (Figure 3A) as well as during cold incubation in Belzer UW solution (followed by rewarming in cell culture medium, simulating reperfusion; Figure 3B, C, and E) and in HTK solution (Figure 3D). As expected, the application of MB at concentrations of 1 µM (data not shown) and 10 µM (Figure 3A) was not toxic under control conditions (for a positive control for toxicity, cells were treated with 500 µM hydrogen peroxide). Under hypothermic conditions, which itself elicited injury in both cold storage solutions (for a time course of the injury in UW solution, see Figure 3B), no toxicity of MB was observed (Figure 3B–D). In contrast, MB even proved to exert protection against hypothermic injury with slight, nonsignificant protection at a concentration of 1 µM and significant protection at a concentration of 10 µM (Figure 3C and D).
![Exclusion of toxic effects of methylene blue (MB) under preservation conditions. LLC-PK1 porcine kidney cells were exposed to 1 and/or 10 µM MB during warm control incubation (A) (5 hours in Krebs-Henseleit [KH] buffer supplemented with 5 mM glucose [Glu] at 37°C) or during cold (4°C) incubation in Belzer UW solution (B, time course; C, concentration dependence; E, 18 hours cold incubation in Belzer UW solution followed by 3 hours rewarming in cell culture medium) or histidine-tryptophan-ketoglutarate (HTK) solution (D). Cell injury was determined by assessing the release of the cytosolic enzyme lactate dehydrogenase (LDH) into the cellular supernatant (A–D) and by assessing nuclear staining with propidium iodide (E). Released LDH activity is given in percentage of total LDH activity (for details, refer to Methods); as a positive control for toxicity, some wells were exposed to hydrogen peroxide ([H2O2] 500 µM at 37°C; A). Bars represent mean ± standard deviation of 4 individual experiments performed in duplicate. *, P < .05 against the respective incubation without MB. (E) After 5 hours of warm incubation in KH buffer with 5 mM glucose (control) or after 18 hours of cold (4°C) incubation in Belzer UW solution with or without 10 µM MB followed by 3 hours rewarming in cell culture medium, cells were stained with the fluorescent dyes Hoechst 33342 (staining nuclei of all cells) and propidium iodide (staining only nuclei of dead cells); fluorescence was assessed by microscopy at λexc. = 359 ± 24 nm/λem. 445 ± 25 nm (Hoechst 33342, top row) and λexc. = 546 ± 6 nm/λem. ≥ 590 nm (propidium iodide, bottom row); images represent 3 experiments performed in duplicate.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/jid/218/11/10.1093_infdis_jiy386/1/m_jiy38603.jpeg?Expires=1748095977&Signature=W-2fzN6j6vQ~8OmsFWNcR6tzQPoCktAS8FsmyZp-xb9Hl8rGUl49npjvoWJjAlSDm1QpPKZsTaMZpZyLGrCma7S0411DB3~qtAIVyLLj0tNcXL8mp1Xh03eh2DiNcV37~c6wSebuAFwLWNN76bHbFgElV8Uxfy3q0Ex~nQ1oewQ3jPrfatGigAhEsZWYr55F8oiqkmRZx-Hb1AMJ9Bf1i5hQwP2sHSFXTNfJcfnktxRi94pLp0wTiJm1ABWMszTucCWXou~-Jq6t9ta2gC9nzmTqyARof12~D7JJ5W8KWQmB5Z4hlcNuAPgABcqyrGq13mJ~JbEgb~xLNxu8h1pXug__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Exclusion of toxic effects of methylene blue (MB) under preservation conditions. LLC-PK1 porcine kidney cells were exposed to 1 and/or 10 µM MB during warm control incubation (A) (5 hours in Krebs-Henseleit [KH] buffer supplemented with 5 mM glucose [Glu] at 37°C) or during cold (4°C) incubation in Belzer UW solution (B, time course; C, concentration dependence; E, 18 hours cold incubation in Belzer UW solution followed by 3 hours rewarming in cell culture medium) or histidine-tryptophan-ketoglutarate (HTK) solution (D). Cell injury was determined by assessing the release of the cytosolic enzyme lactate dehydrogenase (LDH) into the cellular supernatant (A–D) and by assessing nuclear staining with propidium iodide (E). Released LDH activity is given in percentage of total LDH activity (for details, refer to Methods); as a positive control for toxicity, some wells were exposed to hydrogen peroxide ([H2O2] 500 µM at 37°C; A). Bars represent mean ± standard deviation of 4 individual experiments performed in duplicate. *, P < .05 against the respective incubation without MB. (E) After 5 hours of warm incubation in KH buffer with 5 mM glucose (control) or after 18 hours of cold (4°C) incubation in Belzer UW solution with or without 10 µM MB followed by 3 hours rewarming in cell culture medium, cells were stained with the fluorescent dyes Hoechst 33342 (staining nuclei of all cells) and propidium iodide (staining only nuclei of dead cells); fluorescence was assessed by microscopy at λexc. = 359 ± 24 nm/λem. 445 ± 25 nm (Hoechst 33342, top row) and λexc. = 546 ± 6 nm/λem. ≥ 590 nm (propidium iodide, bottom row); images represent 3 experiments performed in duplicate.
The lack of toxicity of MB under warm control conditions was confirmed by Hoechst 33342/propidium iodide staining (1% ± 1% of propidium iodide-positive, ie, dead cells after 5-hour incubation in KH + 5 mM glucose + 10 µM MB vs 3% ± 2% in KH + 5 mM glucose at 37°C; n = 3 in duplicate, 3 random fields per sample). Likewise, the protection by MB against the injury elicited by hypothermia/rewarming was confirmed by Hoechst 33342/propidium iodide staining, showing that the (vast) majority of (detached and still attached) cells cold incubated in Belzer UW or HTK solution (18 hours) in the absence of MB were propidium iodide positive after rewarming, whereas after cold incubation in the presence of 10 µM MB, the majority of cells were propidium iodide negative (Figure 3E for Belzer UW; data not shown for HTK solution; n = 3). In these cold stored or rewarmed cells, partial detachment of cells occurred, especially in the absence of MB (Figure 3E, middle panels; even more marked after cold incubation in HTK solution in the absence of MB [data not shown]; higher detachment in the absence of MB confirmed by phase contrast microscopy [data not shown]); this detachment only allowed semiquantitative assessment of propidium iodide staining in cold-storedor rewarmed cells.
Hepatitis C Virus Decontamination of Mini Pig Kidneys by Flushing With Methylene Blue
So far, our in vitro data indicated that 1 µM MB inactivates HCV within 1 hour. To mimic conditions more closely related to the transplant setting, we removed kidneys from mini pigs and contaminated them with either cell culture-derived virus or patient-derived HCV-positive serum (Figure 4E). For contamination, the renal artery was rinsed with a suspension containing HCV and HTK and incubated for 1 hour at 4°C to allow attachment of viral particles. Afterwards, kidneys were flushed either with HTK supplemented with 1 µM MB (Figure 4A, B, E, F, and G) or only with HTK without MB supplement (Figure 4C and D). Several fractions (each 50 mL) were collected. Next, we infected Huh-7.5 cells with each individual flow and quantified HCV RNA from each fraction. As shown in Figure 4A and B, flushing the organs with MB after contamination reduced (1) HCV infectivity to background levels and (2) HCV RNA copies to an average of 2.300 IU/mL. Flushing the organ with HTK alone (without MB) reduces HCV infectivity and HCV RNA; however, in comparison to HTK supplemented with MB, the effect was less (Figure 4C and D). We observed a similar decrease of HCV infectivity and the number of RNA copies after flushing of kidneys contaminated with human serum from a patient with chronic HCV genotype 2 infection (Figure 4E). In addition to simple flushing of the kidneys, we incubated them with HTK containing 1 µM MB for 1 hour to determine whether RNA copies could be further reduced. However, we could not detect any significant differences in the amount of remaining HCV RNA (Figure 4F and G).
Decontamination of kidneys by flushing with methylene blue (MB). Kidneys from mini pigs were contaminated with HCVcc (A–D) or hepatitis C virus (HCV) genotype 2-positive patient serum (E) in the presence of histidine-tryptophan-ketoglutarate (HTK) via the Arteria renalis (HCVcc:HTK 1:3 = 1 part of virus was mixed with 3 parts of HTK). (A) After 1 hour, the organs were flushed with HTK + 1 µM MB, and flow through was collected in 5 fractions (each 50 mL). Huh-7.5 cells were infected with each fraction for 4 hours at 37 °C. After 48 hours, the cells were lysed and luciferase activity was measured. In addition, the number of ribonucleic acid (RNA) copies in the flow through was determined via quantitative polymerase chain reaction (qPCR) (B). Bars represent mean values ± standard deviation (SD) of 3 individual experiments with 2 replicates. (C) After 1 hour, the organs were flushed with HTK without MB, and flow through was collected as stated above. After 48 hours, the cells were lysed and luciferase activity was measured. In addition, the number of RNA copies in the flow through was determined via qPCR (D). Bars represent mean values ± SD of 3 individual experiments with 2 replicates. (E) After 1 hour, the organs were flushed with HTK + 1 µM MB, and flow through was collected in 5 fractions (each 50 mL). The number of RNA copies in the flow through was determined via qPCR (B). Bars represent mean values ± SD of 3 individual experiments with 2 replicates. (F) After 1 hour, the organs were flushed and incubated for 1 hour with HTK + 1 µM MB; flow through was collected in 6 fractions (F0–F5, each 50 mL). Huh-7.5 cells were infected with each fraction for 4 hours at 37°C. After 48 hours, the cells were lysed and luciferase activity was measured. In addition, the number of RNA copies in the flow through was determined via qPCR (G). Bars represent mean values ± SD of 3 individual experiments with 2 replicates. (H) After contamination with HCVcc, the organs were flushed with HTK + 1 µM MB and fractions were collected. Afterwards, kidneys were flushed with trypsine for 10 minutes, and afterwards HCV RNA copies in the flow through were determined via qPCR. * p < 0.05, **p < 0.01, ***p < 0.005. Abbreviations: ns, not significant; RLU, relative luciferase units.
Because MB does not interfere with HCV binding, it is possible that the majority of HCV virions were still retained in the kidney after perfusion by remaining attached to the cells. Therefore, HCV-contaminated porcine kidneys were treated for 10 minutes with trypsine. Afterwards, HCV RNA was quantified and compared with the MB-perfused fractions of the HCV-contaminated kidneys. As shown in Figure 4H, we were unable to detect any significant differences in HCV RNA amount after trypsine treatment, indicating that HCV did not remain attached to the cells. Taken together, we were unable to detect HCV infectivity after MB treatment, whereas only low levels of HCV RNA remained detectable.
Pulsatile Hypothermic Machine Perfusion Optimizes Kidney Decontamination
Preservation by continuous HMP compared with cold storage significantly improves graft survival in kidneys [22]. Therefore, it is very likely that this technology will be more frequently used in the future. Moreover, a protocol as described above might be even more effective and standardized by using pulsatile HMP.
To assess whether the use of HMP has similar or even better results in HCV decontamination of kidney grafts, we contaminated kidneys from mini pigs with HCV as described above. Afterwards, pulsatile HMP with PS containing 1 µM MB was initiated for 2 hours. Then, the flow through was recovered and analyzed for the residual amount of HCV RNA. Only very low copy numbers of HCV RNA remain detectable (range, 15–1.228 IU/mL), whereas we were unable to detect any HCV RNA in the kidneys itself. In addition, HCV core protein was not detectable in the last flow throughs (data not shown).
To assess the infectivity of the perfusate after MB treatment in vivo, we used a human liver-uPA-SCID mouse model that was infected with HBV and HCV. The chimeric mice carrying human hepatocytes in their liver received either a 10× concentrate of the flow through after perfusion with MB, without MB, or the solution used for kidney contamination as a positive control and were monitored for several weeks. Although the positive control developed viremia of up to 2.57 × 105 IU/mL, the mice given injections with the perfusate with MB were negative for viral RNA according to polymerase chain reaction results (Figure 5B) and did not establish acute HCV infection. The mice given injections with the perfusate without MB established HCV infection. However, 1 mouse died during follow up. This verifies that only fragments of HCV RNA and no viral particles remained after MB treatment.
Kidney decontamination using pulsatile machine perfusion. Kidneys from mini pigs were contaminated with hepatitis C virus (HCV) in the presence of histidine-tryptophan-ketoglutarate (HTK). The organs were flushed via pulsatile hypothermic machine perfusion with perfusion solutions + 1 µM methylene blue (MB) for 2 hours. Afterwards, the flow through was recovered and analyzed. (A) Hepatitis C virus (HCV) ribonucleic acid (RNA) copies were detected via quantitative polymerase chain reaction. (B) Human liver cell xenograft mice (uPA-SCID) were infected with a 10× concentrate of the perfusate or the contamination solution (HTK + HCV) as a positive control. To assess the progress of infection, HCV RNA copies were determined over the course of several weeks. One mouse per condition was analyzed; limit of detection (LOD) = 750 IU/mL. (C) Effect of machine perfusion on kidney function was evaluated after 90 minutes of HTK perfusion with or without 10 µM MB. Perfusate flow, renal clearance, urine flow, fractional excretion (FE) of glucose and sodium, as well as malondialdehyde as a marker for oxidative stress were used as read out. Bars represent mean values ± standard deviation of 4 individual experiments. * p < 0.05, **p < 0.01, ***p < 0.005. Abbreviations: FE, fractional extraction; FENa, fractional extraction of natrium; ns, not significant.
To ensure that MB treatment of kidney grafts had no negative effects on organ function, we next performed pulsatile HMP with PS containing 10 µM MB for 90 minutes. Afterwards, perfusion parameters and graft integrity were recorded. More importantly, we could not detect any differences in perfusate flow, renal clearance, urine flow, as well as fractional excretion of glucose and sodium or malondialdehyde as a marker for oxidative stress of kidneys treated with 10 µM MB in comparison to kidneys that were not treated with MB (Figure 5C). In addition, a biopsy was taken from the kidneys and examined after hematoxylin and eosin staining. Pathological evaluation for MB treatment-associated changes did not reveal any differences in comparison to healthy renal tissue (Supplementary Figure 1). Overall, HMP of contaminated kidneys with HTK supplemented with MB effectively reduced viral particles and diminished HCV RNA while maintaining renal function.
DISCUSSION
The underutilization of kidneys from HCV-positive donors is still an ongoing controversial debate among transplant surgeons, nephrologists, and infectious diseases specialists [6, 23]. Although previous studies primarily considered transplantation into HCV-positive recipients, with the recent advances in antiviral treatment, HCV-negative patients can also benefit from HCV-contaminated kidneys. In a pilot trial at the University of Pennsylvania, successful transplantation from an HCV-positive donor into a negative recipient was performed [9]. Subsequent antiviral treatment with DAAs led to complete cure of the infection without affecting graft function. Nonetheless, considering the cost-intensive antiviral therapy and stigma related to HCV infection, there is still room for improvement. For this reason, we established a novel approach to reduce the amount of infectious particles in kidney grafts to reduce the risk of HCV transmission.
First, we tested 2 compounds with known effect against HCV but different antiviral mechanisms. Epigallocatechin gallate inhibited attachment and consequently HCV entry, as proposed in a recent study [11]. However, it became unstable in cell culture and could not decrease HCV infection effectively in the presence of all preservation solutions. In contrast, HCV infection was significantly decreased in the presence of all tested PS when supplementing them with 1 µM MB, an established and low-priced agent for virus inactivation. The same concentration of MB was shown to effectively rid blood products from viral contaminations, including human immunodeficiency virus and HCV [24, 25]. As a result, it has been mostly used in transfusion medicine over the last decades [26]. In our study, virus inactivation with 1 µM MB was successful after 60 minutes and for 7 different HCV genotypes. Methylene blue is activated by visible light, leading to the release of singlet oxygen, which attacks the viral lipid envelope. Moreover, viral RNA is destroyed by photo degeneration [14]. However, clinical applications of MB extend beyond its antiviral activity: for instance, it is used for visualization during surgical procedures as an antimalarial drug and to treat methemoglobinemia [12]. Although MB is generally known as a safe drug at therapeutic doses of <2 mg/kg, higher concentrations are known to induce tissue necrosis [12]. Moreover, some recent reports have demonstrated allergic reactions towards MB-treated blood products [15]. Nevertheless, the experimental concentration of 1 µM in 250 mL PS, used in our experiments, corresponds to a dose of <1 mg/kg for a kidney weighing approximately 100 grams and therefore does not exceed the recommended therapeutic dose. Although MB is also known to be a redox-cycling compound with potential pro-oxidative effects [16–18], the previously described antioxidative effects of MB appeared to prevail under hypothermic conditions, yielding protection against cold-induced injury. Although the extent of this protection against hypothermic injury was not as pronounced as known from iron chelators [19, 21], these results emphasize that the use of MB should also be safe under hypothermic conditions.
Second, we showed that perfusion of HCV-contaminated kidneys with PS supplemented with 1 µM MB, manually or via HMP, reduced HCV RNA to very low levels and undetectable infectious viral particles. Furthermore, uPA-SCID mice with humanized liver did not establish HCV infection after inoculation with concentrated flow through from HCV-contaminated kidneys. The reduction of the viral load of HCV-contaminated kidneys by machine perfusion alone was demonstrated as early 1995 by Roth et al [27]. Merely flushing the organ for 20 hours reduced virus titers by 75%, whereas additional flushes and perfusion resulted in a 99% decrease. Even though the use of HCV-contaminated kidneys for transplantation has been debated for years, and despite its success in inactivating viruses in blood products, MB has not yet been used for reduction of HCV in kidney grafts. It has to be noted that very low counts of viral RNA were still detectable after flushing kidneys with MB in HTK. However, the amount of RNA was insufficient to effectively replicate in Huh-7.5 cell culture as well as in uPA-SCID mouse chimera carrying human liver xenografts. Because the Abbott RealTime HCV RNA Kit used in this study detected a small section of 96 base pairs, the results suggest that only small fragments of viral RNA remain after MB perfusion. In addition, we have previously shown that the quantity of HCV RNA does not necessarily correlate with viral infectivity [28]. At present, we cannot rule out that kidneys from HCV-infected individuals may be permeated with HCV virions, HCV-immune complexes, or infectious microparticles, and future studies addressing these questions are needed.
Due to the limited host range of HCV, there is a lack of animal models for HCV infection [29]. Studies in human kidneys—even if they were discarded—are not allowed in Germany. As a result, immune compromised mouse models, such as the aforementioned uPA-SCID mice with humanized livers, have been the most suitable model at present.
Finally, in regard to the clinical application and considering the adverse effects resulting from MB treatment in some studies [12], it is still important to estimate how the substance affects organ function. For this reason, we showed that even MB concentrations 10-fold higher than the sufficient dose of 1 µM did not result in significant differences of common parameters of kidney function. Methylene blue treatment during cold ischemia time might even have other benefits: it has been proposed that infections occurring soon after kidney transplant and manifesting due to immunosuppression might be transmitted from the graft, in some cases [30–32]. Consequently, treatment with MB before transplantation might even decrease the risk for other infections.
CONCLUSIONS
Taken together, our results suggest a novel approach for reduction of infectious HCV particles and transmission during transplantation. The procedure is based on MB treatment and HMP, 2 medical applications that are already well established. Additional data suggest effective suppression of HCV infection without compromising organ function. This novel approach might also serve as a model for reducing the risk of pathogen transmission in the context of transplantation.
Supplementary Data
Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Notes
Acknowledgments. We thank Takaji Wakita for gift of the JFH1 strain, Charles Rice for provision of Huh-7.5 cells and the 9E10 monoclonal antibody, and Jens Bukh for providing chimeric hepatitis C virus constructs.
Disclaimer. The design, performance, data interpretation, and manuscript writing was under the complete control of the authors and has never been influenced by Dr. Franz Köhler Chemie GmbH.
Finanical support. This work was funded by the German Center for Infection Research (DZIF) (to S. C.) and the Research Foundation - Flanders (FWO Vlaanderen) (to P. M.).
Potential conflicts of interest. U. R. is a consultant of Dr. Franz Köhler Chemie, the manufacturer of 2 of the preservation solutions used in the current study. All other authors have nothing to disclose. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
