Hydrogen gas inhalation ameliorates glomerular enlargement after hypoxic-ischemic insult in asphyxiated piglet model

Abstract

Acute kidney injury (AKI) has been reported to occur in 30–70% of asphyxiated neonates. Hydrogen (H₂) gas became a major research focus in neonatal medicine after the identification of its robust antioxidative properties. However, the ability of H₂ gas to ameliorate AKI is unknown. We examined histopathological injuries in the piglet renal cortex on day 5 after a hypoxic-ischemic (HI) insult and if H₂ gas can alleviate kidney injuries. Twenty piglets were divided into three groups: no insult (Control, n = 6), HI insult alone (HI, n = 8), and HI insult with H₂ gas ventilation (HI-H₂, 2.1–2.7% for 24 h, n = 6). The total glomerular cell count was significantly higher in the HI group than in the other groups, with no difference between the HI-H₂ and control groups. Proximal tubular lumen narrowing was significantly increased in the HI group versus control, but not in the HI-H₂ group. In this piglet model, glomerular enlargement with an increase in glomerular cell number due to tubular lumen narrowing was observed on day 5 after HI insult. H₂ gas effectively suppressed this glomerular cell increase and tubular lumen narrowing.

Introduction

Asphyxia accounts for approximately 23% of the estimated 4 million neonatal deaths worldwide each year and is associated with significant morbidities in surviving neonates¹. Asphyxiated neonates commonly have multi-organ dysfunction and/or failure^2,3, and neonatal encephalopathy (NE) is a major issue in such neonates. Due to the preferential perfusion of vital organs such as the heart and brain during hypoxia, the kidney is among the first organs injured by a hypoxic-ischemic (HI) insult due to regional vasoconstriction^4,5. Indeed, acute kidney injury (AKI) has been reported to be present in 30–70% of asphyxiated neonates^6–9. A number of recent studies have documented the association of kidney injury with poor neurologic outcomes and increased risk of mortality¹⁰.

In the absence of a basic understanding of the molecular pathophysiology of human AKI, large animal models such as porcine, simian, ovine, canine, and feline models are attractive for testing novel therapies because analogy and the immune system of these animals more closely resembles that of humans when compared with rodents¹¹. We have already established a NE piglet model that can survive for 5 days after insult with histopathological brain injuries^12–16. Therefore, this model can be considered a translational model similar to the clinical neonatal asphyxiated neonates, and evaluating multiple organs in this model will allow us to understand the characteristics of multiorgan failure in HIE infants in the clinical situation¹⁷, especially in the kidneys.

For infants with NE, mild therapeutic hypothermia (TH) remains the primary treatment option, but death or significant neurological disability still occurs in 40–50% of treated infants¹⁸. In animal translational research, the evidence indicating the efficacy of TH in ameliorating AKI in the asphyxiated piglet is still somewhat limited¹⁹. Assessment of the clinical data of babies who received TH alone following intrapartum asphyxia has concluded that the therapy does not reduce their rate or severity of AKI²⁰. Hence, further prognostic improvement will require the development of new therapies that are more effective for both NE and AKI.

Due to the potent antioxidative properties of hydrogen (H₂) gas, identified both in vivo and in vitro for diseases that affect adults, such as cerebral ischemia, H₂ gas is now a major focus of research in the neonatal medicine field²¹. Critically, H₂ is considered an antioxidant^21,22, anti-inflammatory^23,24, and antiapoptotic²⁴ agent. It acts as a therapeutic and preventive antioxidant by selectively reducing the levels of highly active oxidants, such as hydroxyl radical (^•OH) and peroxynitrite (ONOO⁻), in cultured cells²⁵. We previously reported the neuroprotective potential of combined H₂ gas ventilation therapy and TH by assessing the short-term neurological outcomes of 5-day neonatal NE piglets²⁶. However, there are few reports on the efficacy of H₂ gas for ameliorating AKI.We hypothesized that our NE piglet model would show some injuries in the renal cortex and that H₂ gas would reduce the kidney injury. Thus, in this study, we examined the histopathological injuries in the renal cortex on day 5 after HI insult in our NE piglet model and if H₂ gas can alleviate this damage.

Results

Mean blood pressure and heart rate decreased transiently in the HI and HI-H₂ groups after insult, but recovered gradually. In blood gases, pH, PaO₂, and BE decreased and lactate increased, but recovered to baseline after 3 h, with no difference between the HI and HI-H₂ groups (Table 1).

Table 1.

Blood gas and glucose at baseline, at the end of insult (0 h), and at 1, 3, and 6 h after insult in the control, HI and HI-H₂ groups.

	Control (n = 5)	HI group (n = 8)	HI-H2 group (n = 6)
HR
Baseline	230.6 (65.5)	229.9 (29.0)	171.2(7.5)##,§
0 h	236.4 (44.1)	151.3 (18.4)****,####	150.2 (32.5)####
1 h	239.8 (47.6)	228.9 (21.9)	210.1 (18.4)
3 h	235.0 (36.5)	257.5 (15.6)	214.8 (45.5)*
6 h	225.8 (36.6)	223.0 (17.1)	243.8 (28.9)***
MeanBP
Baseline	77.0 (17.0)	78.4 (7.9)	68.0 (7.4)
0 h	75.2 (15.2)	46.9 (5.8)****	43.2 (8.7)****
1 h	74.2 (13.0)	69.4 (8.7)	64.8 (9.8)
3 h	68.8 (8.0)	67.9 (9.0)#	65.0 (9.7)
6 h	58.8 (11.6)**	69.4 (8.9)	70.3 (6.7)
pH
Baseline	7.49 (0.04)	7.46 (0.07)	7.45 (0.07)
0 h	7.45 (0.07)	6.93 (0.15)****, #####	6.99 (0.1)****, #####
1 h	7.47 (0.06)	7.31 (0.06)***, ##	7.35 (0.09)**, #
3 h	7.48 (0.06)	7.50 (0.05)	7.49 (0.04)
6 h	7.50 (0.07)	7.48 (0.1)	7.48 (0.02)
PaCO2(mmHg)
Baseline	36.3 (4.3)	41.7 (7.0)	43.2 (3.9)
0 h	41.2 (4.0)	27.5 (6.5)***, ##	41.7 (9.6)§§§
1 h	39.0 (7.7)	37.8 (8.6)	39.2 (2.5)
3 h	38.5 (6.1)	40.0 (3.6)	40.9 (3.1)
6 h	34.8 (8.6)	44.3 (12.1)#	42.0 (1.8)
PaO2(mmHg)
Baseline	85.4 (17.1)	106.1 (26.6)	102.3 (11.8)
0 h	79.9 (18.2)	16.7 (6.1)****, ####	17.4 (4.0)****, ####
1 h	76.6 (13.2)	96.8 (22.0)	113.2 (22.8)##
3 h	83.3 (24.7)	90.5 (16.2)	103.7 (13.5)
6 h	80.0 (24.9)	96.4 (14.8)	95.5 (18.8)
BE (mmol/L)
Baseline	3.9 (2.7)	6.2 (1.9)	6.0 (3.2)
0 h	4.0 (3.4)	−25.1 (3.7)****, ####	−19.6 (2.9)****, ####, §
1 h	4.2 (3.6)	−6.9 (3.1)****, ####	−3.3 (5.1)****, ##
3 h	4.0 (4.1)	7.7 (3.3)	7.6 (3.6)
6 h	3.4 (4.2)	7.7 (2.5)	7.2 (3.0)
Lactate (mg/dL)
Baseline	26.4 (11.3)	16.6 (6.2)	17.0 (4.2)
0 h	25.2 (14.3)	218.6 (33.2)****, ####	180.8 (15.6)****, ####, §§
1 h	29.8 (23.0)	118.0 (35.9)****, ####	99.8 (18.7)****, ####
3 h	32.8 (24.1)	30.1 (16.2)	28.7 (6.6)
6 h	39.6 (26.5)	25.4 (11.1)	28.2 (5.4)

BE, base excess.^*p < 0.05, ^**p < 0.01, ^***p < 0.005, ^****p < 0.0001 vs. baseline, ^#p < 0.05, ^##p < 0.01, ^####p < 0.0001 vs. control, ^§p < 0.05, ^§§p < 0.01 vs HI by two-way analysis of variance with post hoc Tukey analysis.

Control, one piglet’s data was not available.

The renal cortex of the HI group frequently showed particular histologic features, such as narrowing or disappearance of Bowman’s space with glomerular enlargement, Henle’s coffin luminal structures, and the luminal space of the proximal tubule (Fig. 1b, e). These features were rarely or only occasionally seen in the control and HI-H₂ groups, respectively (Fig. 1a, c, d, f, g, i). The nucleus and eosinophilic cytoplasm of most cells in the glomerulus of the HI group seemed enlarged or swollen. Although it was difficult to accurately identify cell types such as endothelial cells, mesangial cells, and inflammatory cells (Fig. 1e, h), some segmental nucleated cells (indicated by arrows in Fig. 1h), assumed to be neutrophils, were occasionally observed in the HI group. Some αSMA-positive cells (Fig. 1j–l, arrows) and CD31- positive cells (Fig. 1m-o, arrows) were also seen in all groups. The inner diameter of the proximal tubules adjacent to the glomerulus was markedly shorter in the HI group than in the control (Fig. 1p-r).

Fig. 1.

Representative images of PAS staining at low-power (a–c) and high-power (d–f) magnification, H&E staining at high-power magnification (g–i, p–r), αSMA immunostaining (j–l), CD31 immunostaining (m–o) in the renal cortex of the control (a,d, g, j, m, p), HI (b, e, h, k, n, q), and HI-H₂ (c, f, i, l, o, r) groups. Insets (h, j, k, l, m, n, o) show enlarged images of glomeruli in (h, j, k, l, m, n, o). Arrows in (h) of the HI group indicate segmental nucleated cells, whereas arrows in (j, k, l) show αSMA-positive cells and arrows in (m–o) show CD31-positive cells. Proximal tubular inner diameter, Di (short double-headed arrow); outer diameter, Do (long double-headed arrow) (p–r). Scale bar indicates 100 μm.

We morphometrically examined these histologic features. Total numbers of glomeruli were not significantly different among the three groups (Figs. 1a–c and 2a). The percentage of glomeruli showing enlargement with complete loss of Bowman’s space was significantly higher in the HI group than in the other groups but was similar in the HI-H₂ and control groups [Figs. 1d–f and 2b: mean (SD) control, 6.1 (4.4); HI, 65.0 (25.9); HI-H₂,18.4 (16.3)]. The glomerular major axis was significantly longer in the HI group than in the other groups, whereas no difference was found between the HI-H₂ and control groups [Figs. 1d–f and 2c: mean (SD): control, 76.7 (2.5); HI, 85.8 (5.1); HI-H₂, 76.0 (4.1)]. The mean total number of cells in the glomerulus was significantly higher in the HI group than in the other groups but similar in the HI-H₂ and control groups [Fig. 2d: mean (SD): control, 68.2 (5.6); HI, 81.8 (9.4); HI-H₂, 67.6 (6.3)]. A significant positive correlation was found between the percentage of glomeruli showing enlargement and the number of cells in the glomerulus (Fig. 3). On the other hand, the numbers of αSMA-positive cells and CD31-positive cells in the glomerulus did not differ among the three groups (Figs. 1j–l and m-o and 2e and f). The percentage of the Di/Do of the proximal tubule was significantly lower in the HI group than in the control group [Figs. 1p-r and 2g: mean (SD): control, 40.1 (7.6); HI, 23.9 (15.8); HI-H₂, 29.2 (4.4)]. The lower the percentage of the Di/Do, the more enlarged the glomeruli in the HI group (Fig. 4a), although this was not seen in glomerular cells (Fig. 4b). The total duration of low-amplitude-integrated EEG (LAEEG) after the insult was not statistically significant among the two groups: The mean duration ± standard error of the mean (SEM), HI, 18.5 ± 2.8 min, HI-H₂, 15.0 ± 3.0 min. The LAEEG duration after HI insult, which could reflect the severity of brain injuries¹⁴, showed a significant positive correlation with the number of cells in the glomerulus (Fig. 5c) but a tendency for a negative correlation with the Di/Do value (Fig. 5e). However, it showed no correlation or tendency for a correlation with the percentage of glomerular enlargement in both the HI and HI-H₂ groups (Fig. 5a, b). Moreover, the HI-H₂ group did not show any correlation between the number of glomerular cells and the percentage of Di/Do (Fig. 5d, f).

Fig. 2.

Number of glomeruli (a) and percentage of glomerular enlargement (b), glomerular major axis (c), number of glomerular cells (d), number of αSMA-positive cells (e), number of CD31-positive cells (f), and percentage of the Di/Do (g) in the proximal tubules in all groups. Total numbers of glomeruli were not significantly different among the three groups (a). The percentage of glomerular enlargement with complete loss of Bowman’s space was significantly higher in the HI group (■, closed square) than in the other groups while that of the HI-H₂ group (△, open triangle) was not different from that of the control group (●, closed circle) (b). The glomerular major axis was significantly longer in the HI group than in the other groups but was similar in the HI-H₂ and control groups (c). Total glomerular cell count was significantly higher in the HI group than in the other groups and was similar in the HI-H₂ and control groups (d). The number of αSMA-positive cells and CD31-positive cells did not differ among the three groups (e, f). The percentage of the Di/Do in the proximal tubule was significantly lower in the HI group than in the control group (g). ^*p < 0.05, ^**p < 0.01, vs. the HI group.

Fig. 3.

Relationship between glomerular cells and the percentage of glomerular enlargement in the HI (closed square) and HI-H₂groups (open triangle). A significant positive correlation was found between the percentage of glomerular enlargement and the number of glomerular cells.

Fig. 4.

Relationship between the percentage of the Di/Do in the proximal tubules and the percentage of glomerular enlargement (a) and number of glomerular cells (b) in the HI (closed square) and HI-H₂ groups (open triangle). A significant negative correlation was found between the percentage of the Di/Do in the proximal tubules and the percentage of glomerular enlargement.

Fig. 5.

Correlation between the duration of LAEEG after HI insult and the percentage of glomerular enlargement (a, b), number of glomerular cells (c, d), and percentage of the Di/Do (e, f) in the HI and HI-H₂ groups. Closed square (a, c, e) indicated HI group and, open triangle (b, d, f) indicated HI-H₂ group. In HI group (closed square) showed the significant positive correlation between the duration of LAEEG after insult and the glomerular cell count and, the tendency of negative correlation between the duration of LAEEG after insult and Di/Do.

Discussion

This is the first report of the ability of H₂ gas inhalation therapy to ameliorate renal injury in the piglet. Our results show that the percentage of glomerular enlargement in the renal cortex due to an elevated total glomerular cell count was increased in piglets subjected to HI but not in those exposed to HI and H₂ gas together. Furthermore, the HI group exhibited significantly more narrowing of the proximal tubular lumens than the control group, but such narrowed lumens were not evident in the HI -H₂ group.

First, this study revealed that glomerular enlargement may be related to an increase in some cell types in the glomeruli. Generally, glomerular enlargement can be caused by two mechanisms. The first mechanism involves an increase in mesangial cells and a concomitant increase in the mesangial matrix. The second mechanism comprises endocapillary proliferation characterized by endothelial cell proliferation and swelling and inflammatory cell infiltration. Here, despite the elevated total glomerular cell count in the HI group, the number of αSMA-positive cells was not significantly different among the three groups, suggesting that the elevated total glomerular cell count was not due to mesangial proliferation. Although the number of endothelial cells and inflammatory cells are generally considered to increase during endocapillary proliferation, the number of CD31-positive cells did not significantly differ among the three groups. CD31 staining alone could not prove the presence of endothelial cell proliferation. However, segmented neutrophils were seen in the HI group. Thus, inflammatory cell infiltration may be associated with the increased glomerular cell counts. In the HI-H₂ group, there was a reduction in glomerular enlargement and total glomerular cell count. Although we could not identify the endothelial cells and inflammatory cells with immunohistochemical staining, H₂ gas may have reduced endocapillary proliferation.

Next, in the HI group, the proximal tubular epithelial cells (PTECs) swelled and the tubular lumen appeared clearly narrowed. But, Di/Do only reflects the degree of lumen of the proximal tubule and was not directly related to the degree of glomerular swelling or the number of glomerular cells. We speculated that the narrowing of the proximal tubule lumen and the increase in luminal pressure may have overloaded the glomeruli, resulting in an increase in the number of swollen glomeruli. This swelling PTEC change is considered to be an acute phase change. In addition, previous reports in adult pig models have shown chronic changes such as tubular atrophy after 16 weeks²⁷, suggesting that the renal tissue we evaluated on day 5 after hypoxia may represent the acute phase. H₂ gas seemed to suppress PTEC swelling to some extent because there were no samples with a markedly lower Di/Do compared with the HI group. It could be that hypoxia causes PTEC swelling, which in turn causes glomerular enlargement in the upstream glomeruli by some mechanism. There are some reports of ischemia-reperfusion injury (IRI)-induced glomerular enlargement. Electron microscopic evaluation of rats showed that the endothelial and mesangial cells in the glomeruli and the tubular cells were swollen²⁸. Another report indicated the possibility of upstream glomerular damage due to tubular stenosis or obstruction²⁹. Moreover, renal injury caused by intermittent hypoxic stimulation of neonatal rats can be alleviated by H₂ inhalation²³. This hypoxic model showed the same histology as ours, with a narrowed tubular lumen and glomerular enlargement. These reports suggest that hypoxic stimulation causes PTEC swelling, which leads to narrowing of the proximal tubular lumen and secondary glomerular enlargement. In our study, H₂ inhalation therapy ameliorated the glomerular enlargement, endocapillary proliferation, and PTEC swelling, which may act to reduce renal injury. However, the definition of a narrowed proximal tubule lumen is unclear, and the data from this study are not sufficient to determine the relationship between the tubule and the glomerulus. Thus, further investigation of this matter is be needed.

Although there was a reduction in tubular swelling and a decrease in the glomerular cell count after administration of hydrogen gas in this study, H₂ inhalation therapy has been reported to protect multiple organs via antioxidant^21,22, anti-inflammatory^23,24, and antiapoptotic²⁴ effects on various cells throughout the body. These effects may act in combination on glomeruli and tubules to reduce ischemic renal injury in the neonatal period. In Table 2, we list five reports of H₂ therapy for IRI models, all of which were performed in rats. Although the H₂ administration conditions varied, four of the five studies showed elevated serum blood urea nitrogen and creatinine and injury to the tubulointerstitial area due to IRI, which was alleviated by H₂ therapy^22,24,30,31. There was no description of glomerular changes. On the other hand, a report by Nishida et al., who used the same H₂ gas inhalation therapy as we did²³, identified glomerular changes such as glomerular adhesion to Bowman’s capsule in addition to damage to the tubulointerstitial area caused by IRI, which were alleviated by H₂ gas inhalation. Antioxidant, anti-inflammatory, and antiapoptotic effects were demonstrated as the mechanisms of action of H₂. In the NE piglet model, these mechanisms may act in combination on glomeruli and tubules to ameliorate renal damage.

Table 2.

Previous studies of H₂ therapy for ischemia-reperfusion injury models.

Animal	Method	Outcomes of H2 treatment			References
		Renal function	Histology	Immunohistochemistry and other assays
Rat	Continuous intravenous administration of HRS	Elevated Cr and BUN were significantly reduced	Decreased score for tubulointerstitial damage	Decreased 8-OHdG	Shingu et al. 30
Rat	Intraperitoneal administration of HRS	Elevated Cr and BUN were significantly reduced	Ameliorated tubulointerstitial damage	Decreased MDA, 8-OHdG, TNFα, Il-1β, IL-6, TUNEL-positive cells, and MPO Increased SOD and CAT activity	Wang et al. 31
Rat	Intraperitoneal administration of HRS	Elevated Cr and BUN were significantly reduced	Decreased score for tubulointerstitial damage	Decreased TNFα, IL-6, caspase activity, and TUNEL-positive cells Increased Bcl-2/Bax ratio in kidney	Li et al. 24
Rat	Gas treatment with H2	Not analyzed	Decreased damaged area for tubulointerstitial and adhesion between capillary tuft and Bowman capsule	Decreased dihydroethidium and TUNEL-positive cells	Nishida et al. 23
Rat	Intraperitoneal administration of HRS	Elevated Cr and BUN were significantly reduced	Decreased score for tubulointerstitial damage	Decreased MDA and 8-OHdG Increased heme oxygenase-1	Xu et al. 22

HRS, hydrogen-rich saline; Cr, creatinine; BUN, blood urea nitrogen; MDA, malondialdehyde; MPO, myeloperoxidase; SOD, superoxide dismutase; CAT, catalase; N/A, not applicable.

Although there are many reports of rodent studies of NE, it is considered appropriate to use piglets for the perinatal period because they are developmentally similar to humans³². It is not clear whether rodents or piglets are more similar to humans in the perinatal period in terms of the kidney. However, numerous AKI studies, including IRI studies, have demonstrated that the anatomical and physiological characteristics, genetic background, and immune response of piglet are more similar to those of humans than to those of rodents³³. Our findings on the characteristics of the histopathological renal injury in the NE piglet model in this study have considerable significance. In addition, it may be technically difficult to create a situation similar to that of resuscitated human neonates in small animals such as rodents, but it is possible in piglets, which means that the findings may be more clinically relevant. We previously reported renal fibrosis after HI insult in the asphyxiated piglet, although TH was unable to ameliorate the damage¹⁹. Renal hypoxia is pathophysiologically involved in the AKI-to-chronic kidney disease (CKD) transition. After AKI occurs, capillary rarefaction, which is probably induced by decreased expression of vascular factors, such as vascular endothelial growth factor in tubular cells, or is associated with pericyte detachment, causes renal hypoxia. Hypoxia damages tubular epithelial cells, activates fibroblasts, and induces inflammatory reactions, all of which lead to tubulointerstitial fibrosis. Hypoxia and tubulointerstitial fibrosis form a vicious cycle, resulting in progression to CKD³⁴. We suggest that CKD may be caused by this renal fibrosis in later life. On the other hand, the present study focused on elucidating the pathogenesis of AKI. Therefore, in this study, we concentrated on the glomerulus and proximal tubules because we need to look for other findings that are not fibrosis and that contribute to AKI. Hence, in contrast to previous work, we did not examine renal fibrosis.

Our neonatal piglet model exhibits histopathological brain damage and can survive for 5 days¹². This is due to a uniform insult strength based on the duration of LAEEG and the use of cerebral blood volume during the insult as a parameter to control insult severity. However, the brain damage still varied, and the duration of LAEEG after insult was correlated with the severity of the brain damage^13,14. In the present study, glomerular cell count and the duration of LAEEG after insult were positively correlated in the HI group, similar to brain damage, and the duration LAEEG after insult reflected the severity of the glomerular damage. However, this relationship was not observed in the hydrogen group, which may indicate that hydrogen gas reduced the glomerular cell count and glomerular damage.

We have previously reported the neuroprotective effects of hydrogen gas inhalation combined with therapeutic hypothermia on short-term motor function²⁶, reduction of seizure burden³⁵, and improvement of cerebral hemodynamics and oxygen metabolic changes³⁶. Although some therapeutic agents with such neuroprotective effects like Xe and Ar do not have similar organ-protective effects on other organs, it is clear that hydrogen gas has organ-protective effects, at least on the kidneys in this study. It can be verified that hydrogen gas is a medical gas that can be safely used on newborns³⁷. Although no clinical trials of hydrogen gas inhalation therapy for neonates have been reporterd, in adult patients with post-cardiac arrest syndrome, a mixture of 4% H₂ and 96% O₂ inhalation was safe, but did not affect outcomes^38,39. As long as the concentration is below 4%, ignition is almost impossible⁴⁰. H₂ is cheaper than other medical gases such as Xe and Ar and is thus cost-effective in clinical situations²⁵.

Our study has several limitations. The immunostaining reagents for piglets were not well prepared and could not stain glomerular inflammatory cells, which would have provided an indirect assessment of the mechanism of cell proliferation in the glomerulus. Second, we have not been able to evaluate the progression of the proximal tubules and enlarged glomeruli more than 5 days after resuscitation. Further studies are warranted to determine the long-term renal prognosis.

Conclusion

In this translational asphyxiated piglet model, glomerular enlargement with an increase in glomerular cells due to tubular lumen narrowing was observed on day 5 after HI insult. H₂ gas inhalation was able to suppress this increase in glomerular cells and narrowing of the tubular lumen. These results suggest that H₂ gas inhalation therapy may have the potential to ameliorate renal damage in asphyxiated neonates.

Methods

Ethical approval and animal preparation

The study protocol was approved by the Animal Care and Use Committee for Kagawa University (15070-1) and was conducted in accordance with the Animal Research: Reporting In Vivo Experiments (ARRIVE) guidelines and all other applicable guidelines and regulations, as in previous research⁴¹.

Animal model and experimental protocol

Twenty newborn piglets within 24 h after birth were anesthetized and surgically prepared. Anesthesia was induced with 1–2% isoflurane (Forane^® inhalant liquid; Abbott Co., Tokyo, Japan) in air using a facemask. Each piglet was then intubated and mechanically ventilated with an infant ventilator. The umbilical vein and artery were cannulated with a neonatal umbilical catheter (Atom Indwelling Feeding Tube for Infants; Atom Medical Co., Tokyo, Japan); the umbilical vein catheter was used for drip infusion and for blood pressure monitoring/sampling. After cannulation, the piglets were anesthetized with fentanyl citrate at an initial dose of 10 µg/kg followed by continuous infusion at 5 µg/kg/h and were paralyzed with pancuronium bromide at an initial dose of 100 µg/kg followed by continuous infusion at 100 µg/kg/h. Maintenance solution (electrolytes plus 2.7% glucose [KN3B]; Otsuka Pharmaceutical Co., Tokyo, Japan) was infused continuously at a rate of 4 mL/kg/h via the umbilical vein (glucose was infused at a rate of 2 mg/kg/min). Each piglet was then placed in a copper mesh-shielded cage under a radiant warmer to maintain a rectal temperature of 38.0 °C ± 0.5 °C. Inspired gas was prepared by mixing O₂ and N₂ gases to obtain the oxygen concentrations required for the experiment. Ventilation was adjusted to maintain PaO₂ and PaCO₂ within their normal ranges. Arterial blood pressures were measured and recorded via the umbilical arterial catheter.

Hypoxic-ischemic insult protocol

Because the protocol was detailed in our previous work^12,26, only an outline of the HI insult protocol is provided here. Briefly, hypoxia was induced by reducing the inspired oxygen concentration of the ventilator to 4% after at least 120 min of stabilization from the initial anesthetic induction. To obtain a low-amplitude electroencephalography (LAEEG) pattern (< 5 µV), the inspired oxygen concentration was reduced further if required, adjusting it so as to not cause cardiopulmonary arrest. From the beginning of the LAEEG, the insult was continued for 30 min. FiO₂ was decreased (1% decrements) or increased (1% increments) during the insult to maintain the LAEEG, heart rate (> 130 beats/min), and mean arterial blood pressure (MABP) (> 70% of baseline). LAEEG was maintained for 20 min. For the final 10 min of the 30-min insult, if the MABP exceeded 70% of the baseline, hypotension was induced by decreasing the FiO₂. Resuscitation was performed when the cerebral blood volume value dropped below 30% and/or the MABP declined below 70% of baseline. Hypoxia was terminated by resuscitation with 100% oxygen. NaHCO₃ was used to correct a base deficit (base excess below − 5.0 mEq/L) to maintain a pH of 7.3–7.5. After 10 min of 100% FiO₂, the ventilator rate and FiO₂ were gradually reduced to maintain an SpO₂ of 95–98%.

Hypoxic-ischemic insult protocol

Because the protocol was detailed in our previous work^12,26, only an outline of the HI insult protocol is provided here. Briefly, hypoxia was induced by reducing the inspired oxygen concentration of the ventilator to 4% after at least 120 min of stabilization from the initial anesthetic induction. To obtain a low-amplitude electroencephalography (LAEEG) pattern (< 5 µV), the inspired oxygen concentration was reduced further if required, adjusting it so as to not cause cardiopulmonary arrest. From the beginning of the LAEEG, the insult was continued for 30 min. FiO₂ was decreased (1% decrements) or increased (1% increments) during the insult to maintain the LAEEG, heart rate (> 130 beats/min), and mean arterial blood pressure (MABP) (> 70% of baseline). LAEEG was maintained for 20 min. For the final 10 min of the 30-min insult, if the MABP exceeded 70% of the baseline, hypotension was induced by decreasing the FiO₂. Resuscitation was performed when the cerebral blood volume value dropped below 30% and/or the MABP declined below 70% of baseline. Hypoxia was terminated by resuscitation with 100% oxygen. NaHCO₃ was used to correct a base deficit (base excess below − 5.0 mEq/L) to maintain a pH of 7.3–7.5. After 10 min of 100% FiO₂, the ventilator rate and FiO₂ were gradually reduced to maintain an SpO₂ of 95–98%.

Post-insult Treatment

After the HI insult, 20 piglets were randomized into three groups: Control (n = 6), HI insult with normothermia (HI, n = 8), and HI insult with H₂ gas ventilation (H₂, 2.1–2.7% HI-H₂, n = 6). Once the piglets were weaned off the anesthesia and ventilator and extubated, they were allowed to recover and were maintained for 5 days in the incubator. The piglets were fed 50–100 mL artificial animal milk via a nasogastric tube every 6 h. The presence of seizures was recognized clinically as rhythmic pathologic movements (cycling) and tonic postures sustained between cycling episodes. If seizures occurred, the piglet was treated with phenobarbital (20 mg/kg) via intramuscular injection. If seizures persisted, the piglet was treated with two successive anticonvulsant doses. If seizures persisted after two successive anticonvulsant doses, the piglet was euthanized. For H₂ inhalation, two types of cylinders were used: one contained a gas mixture containing 3.8% H₂ and 96.2% N₂; the other contained 100% O₂. The H₂ concentration depended on the oxygen requirement of each piglet. Therefore, the H₂ concentration was usually between 2.1 and 2.7 (FiO₂ range, 0.21–0.4) during the therapy. H₂ gas was delivered via the ventilator for 24 h. The concentration of H₂ gas was measured using a portable gas monitor (TP-70D; Riken Keiki Co., Ltd., Tokyo, Japan). After 24 h of treatment, the H₂-N₂ gas mixture was again replaced with compressed air.

Histopathological analysis

For the euthanasia of piglets on day 5 after the insult, their face was inserted into a mask and inhalation anesthesia was administered. The anesthetic agent isoflurane was introduced via a vaporizer. The vapor was inhaled until respiration ceased and death ensued. The kidney of each animal was perfused with 0.9% saline and 4% phosphate-buffered paraformaldehyde. Kidneys were embedded in paraffin and sectioned. The Sect. (4 μm thick) were stained with hematoxylin and eosin (H&E) and Periodic acid-Schiff (PAS). In addition, the sections were boiled in a citrate buffer for 20 min to retrieve antigens and immunostained with anti-alpha smooth muscle actin (αSMA) antibody (1:200, Novus Biologicals, NBP2-34522, Centennial, USA) and CD31 (1:50, Abcam, ab28364, Cambridge, UK) followed by nuclear staining with hematoxylin respectively. αSMA was used to evaluate mesangial cell proliferation and CD31 was used to evaluate endothelial cell proliferation. Total numbers of glomeruli in sections stained with PAS were counted in the three experimental groups at 100× magnification. The ratio of glomeruli showing enlargement with complete loss of Bowman’s space to glomeruli showing normal histologic appearance was evaluated in 10 randomly selected fields of view in each sample. The average major axis length was measured for 50 randomly selected glomeruli in each sample. To evaluate the degree of proximal tubular luminal obstruction, the percentage of the proximal tubular inner diameter to outer diameter (Di/Do) adjacent to Bowman’s capsule was calculated for 50 randomly selected proximal tubules in each sample in H&E-stained sections. The total numbers of glomerular cells and αSMA-positive cells were counted for 10 randomly selected glomeruli in each sample in αSMA-immunostained sections. And CD31- positive cells were counted for 10 other randomly selected glomeruli in each sample.

Statistical analysis

GraphPad Prism 7.02 (GraphPad Software, La Jolla, CA, USA) was used for all statistical analyses. The mean values for the percentage of glomeruli with complete loss of Bowman’s space, the glomerular major axis length, the total number of glomerular cells, the total numbers of cells, αSMA-positive cells and CD31, and the percentage of Di/Do were calculated for each sample, and the mean values were analyzed by one-way ANOVA followed by Tukey multiple comparisons test. The correlations between the number of glomerular cells and the percentage of glomerular enlargement, between the percentage of Di/Do and the percentage of glomerular enlargement, between the percentage of Di/Do and the number of glomerular cells, the correlations of the duration of LAEEG after insult with the percentage of glomerular enlargement, number of glomerular cells and Di/Do were calculated by using Pearson’s analysis. A p value < 0.05 was considered significant.

Author contributions

T.Iwaki, S.N., and T.K. were involved in the initial study design and wrote the main text. K.K., Y.N., Y.H., T.Mitsuie, T.T., A.M., N.F., Y.K. carried out the animal experiments and recorded the blood gas and physiological data. T.Iwaki, T.W.,M.U. and T.Miki worked on and scored the histopathology. K.K., S.K., T.Y. and T.Iwase contributed to the data analysis and performed the statistical analysis. All members drafted the article and critically revised it.

Funding

This study was financially supported by JSPS KAKENHI grants (19K08253 (S.N), 19K08349 (K.K), 20K08159(S.K), 20H00102 (T.K), 22K15923 (Y.N), and 22K07822 (T.K), 22K15899 (T.W), 22K15922 (A.M), 22H04922 (AdAMS),23K07332(S.N)) and, Kagawa University School of Medicine Alumni Association. the Sanjukai R1-1, R5-1 (S.N).　

Data availability

All data generated or analyzed during this study are included in this article. Further enquiries can be directed to the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.