Small bowel transplantation (SBT) is the only treatment for end-stage intestinal failure. The patient and graft survival are still dismal, because of the lack of adequate immunosuppression and ischemia-reperfusion injury (IRI).¹ Ischemic-reperfusion injury is one of the obstacles to success in SBT. The destruction of intestinal barrier function induced by IRI may easily stimulate bacterial translocation, which can trigger systemic inflammatory response syndrome with high mortality.^2,3

Two preservation methods exist in SBT: intravascular and luminal preservations. Intravascular preservation cannot solely preserve mucosal damage caused by intestinal IRI for an extended period of storage time.^4,5 Luminal preservation showed better graft integrity in comparison to intravascular preservation.^6-8

Hydroxyl radicals and peroxinitrites released by IRI have strong cytotoxic effect by producing lipid peroxidation, DNA oxidation, and mitochondrial depolarization. It has been reported that hydrogen gas had protective effect against IRI by decreasing hydroxyl radicals and peroxinitrites.⁹ However, hydrogen gas is highly flammable in the air over the concentration of 4.6%. On the other hand, hydrogen-rich solution resolves this problem because it is portable, easily administered and a safe means of delivering hydrogen.¹⁰

Rodent models of intestinal IRI by intravenous administration and SBT by inhalation and using hydrogen-rich solution as a preservation solution are reported.^11-13 It is hypothesized that luminal preservation by using hydrogen-rich solution can be a novel and unique method to attenuate intestinal IRI in the clinical setting of SBT. Our purpose of this study was to evaluate the efficacy of luminal injection with hydrogen-rich solution against intestinal IRI in rat.

RESULTS

Hydrogen Concentration of Small Intestine by Various Administration Methods

Hydrogen concentration of small intestine was significantly higher by oral administration, compared to other administration at 5 and 15 min after injection (Fig. 1A). Hydrogen concentration at 30 and 60 min after injection was no significant difference between the groups.

Time-Dependent Change of Hydrogen Concentration and Hydrogen Consumption in Normal Group and 90 Min Ischemic Group by Luminal Injection

Time-dependent variation of hydrogen concentration by luminal injection in small intestine was shown in Figure 1(B). The mean hydrogen concentration in 90 min ischemic group was significantly lower than that in normal group at 1 and 5 min after injection (1,265 vs. 2,150 ppm; P < 0.05, 232.2 vs. 1395.2 ppm; P < 0.01, respectively).

Effect of Hydrogen-Rich Glucose Saline on Oxidative Stress

The tissue value of malondialdehyde (MDA) was significantly decreased in hydrogen-rich glucose saline (HRGS) group at 1 hr after reperfusion (Fig. 2A). Moreover, 8-hydroxydeoxyguanosine (8-OHdG)–positive cells in the intestinal tissues significantly increased in control and GS group, and suppressed in HRGS group at 1 hr after reperfusion (Fig. 2B). The distribution of 8-OHdG-positive cells was significantly suppressed in HRGS group at 1 hr after reperfusion (Fig. 2C).

Messenger RNA Expression of Small Intestine

Interleukin (IL)-6 and inducible nitric oxide synthase (iNOS) messenger RNA (mRNA) levels in the intestinal tissues were significantly suppressed in HRGS group at 1 hr after reperfusion (Fig. 3). Tumor necrosis factor (TNF)-α and IL-1β mRNA levels in the intestinal tissues tended to decrease in HRGS group at 1 hr after reperfusion with no significant difference. Intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), and high mobility group box chromosomal protein 1 (HMGB1) have a tendency toward lower mRNA levels in HRGS group at 1 hr after reperfusion, although no significant difference were found.

HRGS Attenuates Apoptosis of Crypt Cells After Intestinal IRI

Crypt apoptosis was stimulated at 3 hr after reperfusion in control and GS group. On the other hand, crypt apoptosis was significantly inhibited in HRGS group (Fig. 4A). The distribution of crypt apoptosis at 3 hr after reperfusion was significantly suppressed in HRGS group (Fig. 4B).

Hydrogen-Rich Solution Histologically Reduced IRI

The damage of the intestinal mucosa in control and GS groups was severer than that in HRGS group at 1 and 3 hr after reperfusion (Fig. 5A). Crypt cells revealed to be necrotic in control group. The construction of villi was preserved in HRGS group in spite of the presence of shortened villi and edema. The grade of Park-Chiu classification was significantly lower in HRGS group at 1 and 3 hr after reperfusion (Fig. 5B, C).

DISCUSSION

Our results indicate that luminal injection can reduce oxidative stress, the production of inflammatory cytokines, and apoptotic change in crypt cells in the intestinal tissues, which are compatible with those from the other reports of intestinal IRI.^11-14 Significant suppression of MDA and 8-OHdG in the intestinal tissue may indicate that luminal injection of hydrogen-rich solution neutralizes hydroxyl radicals. Marked suppression of hydroxyl radicals result in reduction of proinflammatory cytokines. Especially, iNOS expression was significantly suppressed in HRGS group. Nitric oxide generated by iNOS reacts with superoxide radicals to form peroxynitrite, which has strong oxidative power. Therefore, hydrogen might reduce not only hydroxyl radicals but also peroxynitrites. Suppression of IL-1β, TNF-α, and IL-6 indicates is compatible with inhibition of macrophage infiltration because these cytokines are produced by macrophages. Moreover, it is known that the Janus kinase-signal transducers and activators of transcription pathway, which plays an essential role in intestinal apoptosis, are activated by IL-6.^15,16 Consequently, apoptosis and intestinal tissue injury after intestinal ischemia reperfusion are suppressed by hydrogen-rich solution. Furthermore, HMGB1, which is known to a crucial mediator of shock and inflammation, is promoted by IL-1β and TNF-α, and passively released by necrotic cells.¹⁷ Stimulation of endothelia with HMGB1 causes increased expression of ICAM-1 and VCAM-1, which are the adhesion molecules expressed on endothelial cells, have a role to mediate adhesion and emigration of activated leukocytes in postcapillary venules.² The expression of HMGB1, ICAM-1, and VCAM-1 may be later than that of IL-1β and TNF-α. Thus, in our study, the decrease of the HMGB1, ICAM-1, and VCAM-1 genes expression in HRGS group is compatible with reduction of proinflammatory cytokines resulting from suppression of hydroxyl radicals.

We showed the variation of hydrogen concentration by various administration methods. Superoxide anion, generated in the mitochondria, stimulates the production of hydrogen peroxide, followed by hydroxyl radicals by means of the Fenton reaction, which has strong oxidative power.¹⁸ In addition, hydroxyl radicals are mainly released at early phase after reperfusion.¹⁹ Ohsawa et al.⁹ reported that molecular hydrogen had antioxidant effect against IRI by selective neutralization of hydroxyl radicals. Chuai et al.²⁰ showed the reduction rate of hydroxyl radicals had strong correlation with hydrogen concentration. Zao et al.²¹ reported the protective effect of hydrogen-rich solution on the abdominal skin flap after ischemia reperfusion by intraperitoneal administration. They demonstrated that survival rate of skin flap was improved hydrogen-dose dependently. Therefore, it is important to maintain high hydrogen concentration within target organ to reduce hydroxyl radicals, especially at early phase after reperfusion.

Previous studies described about the effectiveness of hydrogen in intestinal IRI model and SBT in rats by various administration methods.^11-13 Zheng et al. and Chen et al.^11,12 reported hydrogen-rich saline administered intravenously 10 min and 30 min before reperfusion attenuated IRI in rats, respectively. Buchholz et al.¹³ demonstrated the effect of hydrogen in SBT model in rats, 2% of hydrogen gas inhalation during donor and recipient operation. These papers did not demonstrate tissue hydrogen concentration after reperfusion. Our study indicated these administration methods could not maintain hydrogen concentration in small intestine at early phase of reperfusion. Hydrogen effect of reducing hydroxyl radicals presented with existence of hydrogen in the target organ because hydroxyl radicals have the highest reactivity of all reactive oxygen spices.^9,10 If tissue hydrogen concentration was low at early phase after reperfusion in these studies, we should consider new mechanism of hydrogen cytoprotective effect. Recent article has shown that hydrogen increased the levels of antiapoptotic proteins, such as B-cell lymphoma-2 and B-cell lymphoma-extra large, as a signaling molecules in both the extrinsic and intrinsic apoptotic pathway.²² However, the details of the biologic mechanisms associated with hydrogen were unclear.

In this study, we focused on luminal preservation in SBT. The most important feature of intestinal lumen provides route for absorption of water and nutrients. Luminal preservation reduces mucosal damage caused by intestinal IRI compared to intravascular preservation.^6-8 Luminal injection of hydrogen-rich solution has 2 beneficial effects. First, it is possible to inject tailored hydrogen-rich solution because hydrogen can be easily added to the preservation solution. For example, addition of amino acid or polyethylene glycol to preservation solution by luminal injection has better effect to attenuate intestinal IRI and preserve tight junction.^6,8 Second, luminal injection of hydrogen-rich solution shows the highest hydrogen concentration of small intestine compared to other administration methods. As previously indicated, luminal injection of hydrogen-rich solution can suppress oxidative stress effectively. Thus, luminal injection of hydrogen-rich solution has enormous potential for reducing intestinal IRI. Moreover, it is possible to apply luminal injection of hydrogen-rich solution in the clinical setting, such as abdominal aortic aneurysm surgery, supramesenteric artery occlusion, and SBT. This method ameliorates local injury effectively, followed by bacterial translocation and systemic inflammatory response syndrome, eventually improve mortality rate for intestinal IRI. In case of SBT, hydrogen is expected to prolong preservation time of graft as reported in rodent model of kidney and heart transplantation.^23,24 Large animal study should be necessary to evaluate the effectiveness of hydrogen-rich solution for the clinical application.

There might be a limitation in this study, warm ischemia is caused during ischemic period; thus, this may not apply equally to clinical SBT. Further investigation should be required to examine the efficacy of luminal injection of hydrogen-rich solution in SBT model.

In conclusion, we demonstrated the effectiveness of hydrogen-rich solution by luminal injection, which luminal injection was specific method for small intestine. We suggest luminal injection of tailored hydrogen-rich solution can be a breakthrough of intestinal graft preservation method for SBT.

MATERIALS AND METHODS

Animals

Inbred male Lewis rats, weighing 170 to 200 g, were purchased from SLC Japan (Hamamatsu, Japan). The animals were maintained in an environment with controlled temperature and light and allowed free access to a standard diet and water throughout the experimental period. All studies were performed in accordance with the principles of the Guidelines for Animal Experimentation at the National Research Institute for Child Health and Development.

Preparation of HRGS and the Measurement of Hydrogen Concentration in Small Intestine

High concentration of hydrogen-rich solution (5 ppm) was employed, which was described by Ishibashi et al.²⁵ Briefly, 450 mL of saline and 50 mL of 50% glucose were mixed into 530 mL plastic bottle (5% GS solution) to provide a carbonated drink. The HRGS was prepared using hydrogen generating agent (MiZ Co., Ltd., Kanagawa, Japan) including metal aluminium grains and calcium hydroxide; 0.5 g of the materials was enclosed and heat-sealed within a nonwoven fabric. Hydrogen-generating agent was inserted into small bottle, and then added 0.8 mL of pure water, a cap with a check valve was tightly closed. Small bottle was inserted into 530 mL bottle filled with 5% GS.

Approximately 1 g of tissue sample was obtained and put into a small container (24 mL) for the measurement of hydrogen concentration. The hydrogen concentration was measured using a sensor gas chromatograph, SGHA-PA (FIS Inc., Hyogo, Japan) after pulverizing the tissue using gentleMACS Octo Dissociator (Miltenyi biotec GmbH, Bergisch Gladbach, Germany). Hydrogen concentration of tissue samples was measured after cleaning tissue samples with saline to avoid mixture of hydrogen rich solution into a small container. Hydrogen concentration in 24 mL container was converted into concentration per 1 mL.

Distribution of Hydrogen in Small Intestine by Oral, Intraperitoneal, Intravenous, and Inhalation Administration

Hydrogen concentration in small intestine was measured by various administration methods, such as oral, intraperitoneal, and intravenous administration of HRGS and 4% of hydrogen gas inhalation. Two milliliters of HRGS was injected by oral route, abdominal puncture, and dorsal vein of penis. Hydrogen gas was given until sacrificed. Rats were sacrificed to obtain tissue samples (oral and intraperitoneal administration: before injection, 5, 15, 30, and 60 min after injection, intravenous administration; before injection and 5 min after injection, inhalation; before injection, 5, 30, and 60 min after injection, n = 3, respectively).

Experimental Protocols

Experimental protocols in this study were divided into four groups; sham group, no treatment group (control), GS and HRGS treatment group. Ischemic reperfusion was induced by occlusion of the supramesenteric artery and vein for 90 min by a microvascular clamp. Two points of small intestine were ligated at the points of 5 cm and 10 cm from treitz ligament to make injection space. Two milliliters of GS or HRGS was injected into injection space using 24G catheter (Surflo; Terumo Corporation, Tokyo, Japan), and then vascular clamp was removed. Tissue samples were obtained at 1 and 3 hr after reperfusion.

The Measurement of Hydrogen Consumption After 90 Min IRI

Hydrogen concentration of small intestine was measured before injection, at 1, 5, 15, 30, and 60 min after luminal injection of HRGS in 90 min ischemic group (n = 3, respectively) and no ischemic group (normal group, n = 3 respectively). Hydrogen consumption was defined as difference between mean hydrogen concentration of normal group and 90 min ischemic group.

Oxidative Damage Measurement

Tissue MDA concentration and 8-OHdG immunostaining were performed for the analysis of oxidative stress. As for MDA measurement, tissue samples were frozen immediately in liquid nitrogen, then, stored at −80°C. Tissue samples were thawed on ice and divided 50 μg of samples each after all samples were collected. Divided samples were homogenized in RIPA buffer (Wako Pure Chemical Industries, Ltd., Osaka, Japan) to prevent sample oxidation. All samples were centrifuged (10,000×g, 10 min, 4°C) to collect supernatant. Tissue MDA levels were assessed using an OxiSelect TBARS Assay kit (Cell Biolabs, Inc., San Diego, CA) according to manufacturing protocol. The absorbance at 532 nm was measured using a NanoDrop (Thermo Fisher Scientific, Yokohama, Japan).

Subsequently, in 8-OHdG immunostaining, tissue samples were soaked in Bouin solution for 3 to 5 days, and then embedded in paraffin. The specimens were cut into 3-μm tissue sections, deparaffinized with xylene and alcohol. The sections were treated with autoclave for 10 min, 120°C in citric buffer for antigen retrieval. The primary antibody (anti-8-hydroxy-2’-deoxyguanosine monoclonal antibody N45.1; Japan Institute for the Control of Aging, Nikken Seil Co., Ltd., Shizuoka, Japan) was applied overnight at 4°C after incubation with 8% skimmed milk for 30 min to block nonspecific reaction. They were then incubated using LSABTM2 kits/HRP (Dako, Tokyo, Japan) and streptavidin biotin complex (Dako) according to manufacturing protocol as the secondary antibody, and dyed with 3,3′-diaminobenzidine,tetrahydrochloride. As for 8-OHdG immunostaining, 10 viewing fields randomly selected on each slide section were examined at ×400 magnification, and the rate of 8-OHdG-positive cells was calculated in the crypt.

RNA Isolation and Quantitative Real-Time Reverse-Transcriptase Polymerase Chain Reaction

Total RNA was extracted from the small intestine using a sepasol RNA 1 super G (Nacalai Tesque, Kyoto, Japan) and DNA-free kit (Life technologies, Carlsbad, CA) according to manufacturing protocol. Aliquots of 800 ng of RNA were reverse transcripted to complementary DNA using PrimeScript RT reagent kit (Perfect Real Time) (Takara Bio Inc., Shiga, Japan) according to manufacturing protocol. Quantitative real-time reverse-transcriptase polymerase chain reaction was performed using TaqMan on an Applied Biosystems 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA). Primers amplifying the rat mRNA regions and a specific TaqMan probe were designed using the Primer Express software package (Applied Biosystems). Data are expressed as the comparative cycle threshold. The normalized cycle threshold value of each gene was obtained by subtracting the cycle threshold value of 18S ribosomal RNA.

Analysis of Crypt Apoptosis

Terminal deoxynucleotudyl transferase-mediated dUTP nick end-labeling (TUNEL) staining was performed using a Peroxidase in situ Apoptosis Detection Kit S7100 (Chemicon International, Inc. CA), according to the manufacturer’s instructions. Briefly, the sections were deparaffinized and treated with proteinase K (20 μg/mL) for 15 min at room temperature. The slides were incubated with 3% hydrogen peroxide in PBS for 10 min to block endogenous peroxidase activity, followed by incubation in Working Strength TdT enzyme solution for 60 min at 37°C. The reaction was terminated by incubation in a Working Strength Stop/Wash Buffer for 10 min. The slides were incubated with antidigoxigenin peroxidase for 30 min at room temperature. Chromogenic color was developed with 3,3′-diaminobenzidine, and nuclei were counterstained with hematoxylin. Tissue samples for TUNEL staining were obtained at 3 hr after reperfusion. The rate of TUNEL-positive cells in the crypt was examined at ×400 magnification in the 10 viewing fields randomly selected on each slide.

Morphologic Analysis

Tissue samples were fixed in 10% formalin, embedded in paraffin. Three-micron tissue sections were mounted on slides, deparaffinized with xylene and alcohol, and stained with hematoxylin-eosin. Histologic damage was assessed using Park-Chiu classification.^26,27 Park-Chiu classification was graded by a skillful pathologist.

Statistical Analysis

The SPSS software program was used for the statistical analysis (SPSS version 18.0; SPSS Inc. Chicago, IL). All results are given as the means ± standard deviation. Statistical analyses were performed using the Tukey test for parametric multiple comparisons. P values less than 0.05 were considered to be significant.

ACKNOWLEDGMENT

The authors are deeply grateful to Bunpei Sato (MiZ Co., Ltd., Kanagawa, Japan) who supplied hydrogen generating agent to prepare hydrogen-rich solution and supported their experiment.