Makiko Ohshima

Comparison of Angiogenic, Cytoprotective, and Immunosuppressive Properties of Human Amnion- and Chorion-Derived Mesenchymal Stem Cells

Although mesenchymal stem cells (MSCs) can be obtained from the fetal membrane (FM), little information is available regarding biological differences in MSCs derived from different layers of the FM or their therapeutic potential. Isolated MSCs from both amnion and chorion layers of FM showed similar morphological appearance, multipotency, and cell-surface antigen expression. Conditioned media obtained from amnion- and chorion-derived MSCs inhibited cell death caused by serum starvation or hypoxia in endothelial cells and cardiomyocytes. Amnion and chorion MSCs secreted significant amounts of angiogenic factors including HGF, IGF-1, VEGF, and bFGF, although differences in the cellular expression profile of these soluble factors were observed. Transplantation of human amnion or chorion MSCs significantly increased blood flow and capillary density in a murine hindlimb ischemia model. In addition, compared to human chorion MSCs, human amnion MSCs markedly reduced T-lymphocyte proliferation with the enhanced secretion of PGE2, and improved the pathological situation of a mouse model of acute graft-versus-host disease. Our results highlight that human amnion- and chorion-derived MSCs, which showed differences in their soluble factor secretion and angiogenic/immuno-suppressive function, could be ideal cell sources for regenerative medicine.

Effects of intravenous administration of umbilical cord blood CD34+ cells in a mouse model of neonatal stroke

Neonatal stroke occurs in approximately 1/4000 live births and results in life-long neurological impairments: e.g., cerebral palsy. Currently, there is no evidence-based specific treatment for neonates with stroke. Several studies have reported the benefits of umbilical cord blood (UCB) cell treatment in rodent models of neonatal brain injury. However, all of the studies examined the effects of administering either the UCB mononuclear cell fraction or UCB-derived mesenchymal stem cells in neonatal rat models. The objective of this study was to examine the effects of human UCB CD34(+) cells (hematopoietic stem cell/endothelial progenitor cells) in a mouse model of neonatal stroke, which we recently developed. On postnatal day 12, immunocompromized (SCID) mice underwent permanent occlusion of the left middle cerebral artery (MCAO). Forty-eight hours after MCAO, human UCB CD34(+) cells (1×10(5)cells) were injected intravenously into the mice. The area in which cerebral blood flow (CBF) was maintained was temporarily larger in the cell-treated group than in the phosphate-buffered saline (PBS)-treated group at 24h after treatment. With cell treatment, the percent loss of ipsilateral hemispheric volume was significantly ameliorated (21.5±1.9%) compared with the PBS group (25.6±5.1%) when assessed at 7weeks after MCAO. The cell-treated group did not exhibit significant differences from the PBS group in either rotarod (238±46s in the sham-surgery group, 175±49s in the PBS group, 203±54s in the cell-treated group) or open-field tests. The intravenous administration of human UCB CD34(+) cells modestly reduced histological ischemic brain damage after neonatal stroke in mice, with a transient augmentation of CBF in the peri-infarct area.

Cerebral blood flow during reperfusion predicts later brain damage in a mouse and a rat model of neonatal hypoxic-ischemic encephalopathy

Children with severe neonatal hypoxic-ischemic encephalopathy (HIE) die or develop life-long neurological impairments such as cerebral palsy and mental retardation. Decreased regional cerebral blood flow (CBF) is believed to be the predominant factor that determines the level of tissue injury in the immature brain. However, the spatio-temporal profiles of CBF after neonatal HIE are not well understood. CB17 mouse and Wistar rat pups were exposed to a unilateral hypoxic-ischemic (HI) insult at eight or seven days of age. Laser speckle imaging sequentially measured the cortical surface CBF before the hypoxic exposure and until 24h after the hypoxic exposure. Seven days after the HI insult, brain damage was morphologically assessed by measuring the hemispheric volumes and by semi-quantitative scoring for neuropathologic injury. The mean CBF on the ipsilateral hemisphere in mice decreased after carotid artery ligation. After the end of hypoxic insult (i.e., the reperfusion phase), the mean CBF level gradually rose and nearly attained its pre-surgery level by 9h of reperfusion. It then decreased. The degree of reduced CBF during reperfusion was well correlated with the degree of later morphological brain damage. The correlation was the strongest when the CBF was measured in the ischemic core region at 24h of reperfusion in mice (R²=0.89). A similar trend in results was found in rats. These results suggest that the CBF level during reperfusion may be a useful predictive factor for later brain damage in immature mice. This may enable optimizing brain damage for detail analyses.

A novel reproducible model of neonatal stroke in mice: Comparison with a hypoxia–ischemia model

Neonatal stroke occurs in 1/4000 live births and leaves life-long neurological impairments, such as cerebral palsy and epilepsy. Currently, the rodent models of neonatal stroke that are available exhibit significant inter-animal variability, which makes it difficult to accurately assess the mechanisms of brain injury and the efficacy of candidate treatments. We aimed to introduce a novel, highly reproducible model of stroke, middle cerebral artery occlusion (MCAO), in immature mice, and to evaluate the reproducibility of this model compared with a conventional hypoxia-ischemia (HI) model. Postnatal day 12 CB-17 mice underwent left MCAO by direct electrocoagulation. The MCAO model exhibited excellent long-term survival; 85% up to 8 weeks after the insult. Infarct was evident in every animal with MCAO (n=27) and was confined to the cortex, with the exception of some mild thalamic injury. While the % stroke volume 48 h after the insult was consistent in the MCAO group, range: 17.8-30.4% (minimum-maximum), it was substantially less consistent in the HI group, range: 3.0-70.1%. This contrasting variability between the two models was also evident in the cerebral blood flow, 24h after the insult, and in the ipsilateral hemispheric volume, as assessed at 8 weeks after the insult. Mice with MCAO exhibited significant neurofunctional deficits in the rotarod and open-field tests. Preclinical studies for neonatal stroke could become more reliable using this model, with even a potential reduction in the number of pups required for statistical significance. The contrasting variability between the two models may provide insights into the factors that contribute to inter-animal variability in brain injury.

Progesterone and allopregnanolone exacerbate hypoxic-ischemic brain injury in immature rats.

Progesterone and its metabolite, allopregnanolone, are neurosteroids that are present at high concentrations in fetal brains that decrease right after birth. Allopregnanolone is a potent positive modulator of γ-aminobutyric acid A (GABA(A)) receptor function. We examined the effect of exogenous administration of these steroids on hypoxic-ischemic encephalopathy in immature rats. Progesterone (10mg/kg), allopregnanolone (10mg/kg), or vehicle alone was intraperitoneally administered immediately before and then subcutaneously 6h and 24h after hypoxia-ischemia to postnatal day 7 (P7), day 14 (P14), and day 21 (P21) rats. The effects of the treatments were evaluated using histological analyses (hemispheric volumes and semi-quantitative scoring for neuropathologic injury). Both progesterone and allopregnanolone significantly exacerbated brain injury in P7 and P14 rats, but not in P21 rats. This detrimental effect was similar across the examined brain regions (the cortex, striatum, hippocampus, and thalamus) and showed no sex differences. Co-administration of the GABA(A) receptor antagonist, bicuculline, partially mitigated the exacerbating effect of allopregnanolone. Based on the similarity of the effects of these neurosteroids, we speculate that progesterone accentuates neuronal injury mainly via the activity of allopregnanolone. The present study indicates that the detrimental effects of allopregnanolone were, at least in part, mediated via GABAergic neuroexcitability. This is in line with the notion that GABA is excitatory for immature neurons, while it is inhibitory for mature neurons.

Systemic transplantation of allogenic fetal membrane-derived mesenchymal stem cells suppresses Th1 and Th17 T cell responses in experimental autoimmune myocarditis.

We have reported that systemic administration of autologous bone marrow or allogenic fetal membrane (FM)-derived mesenchymal stem cells (MSCs) similarly attenuated myocardial injury in rats with experimental autoimmune myocarditis (EAM). Since rat EAM is a T-helper (Th) cell-mediated autoimmune disease, and recent evidence has indicated that both autologous and allogenic MSCs exert an immunosuppressive effect on Th cell activity, we focused on Th cell differentiation in allogenic FM-MSC administered EAM rats. EAM was induced in Lewis rats by injecting porcine cardiac myosin (day 0). Allogenic FM-MSCs, obtained from major histocompatibility complex mismatched ACI rats, were intravenously injected (5 × 10(5)cells/rat) on days 7, 10, or 14 (MSCd7, MSCd10, or MSCd14 groups, respectively). At day 21, echocardiography confirmed that reduced ejection fraction in the untreated EAM group (63 ± 2%) was significantly improved in the MSCd10 and MSCd14 groups (74 ± 1 and 75 ± 2%, respectively, P<0.01). CD68 immunostaining revealed that prominent macrophage infiltration in the myocardium of the EAM group (1466 ± 93 cells/mm(2)) was significantly decreased in the MSCd10 group (958 ± 139 cells/mm(2), P<0.05). To evaluate Th cell differentiation, we used flow cytometry to determine the percentage of interferon (IFN)-γ positive Th1 and interleukin (IL)-17 positive Th17 cells in peripheral CD4-positive Th cells. The percentage of Th1 cells at day 16 was significantly lower in the MSCd10 (1.3 ± 0.2%) and MSCd14 (1.6 ± 0.3%) groups compared to the EAM group (2.4 ± 0.3%, P<0.05), as was the percentage of Th17 cells in the MSCd10 group (1.9 ± 0.5%) compared to the EAM group (2.2 ± 0.9%, P<0.05). At day 21, infiltrating Th17 cells in myocardium were significantly decreased in the MSCd10 group (501 ± 132 cells/mm(2), P<0.05) compared to EAM (921 ± 109 cells/mm(2)). In addition, human CD4+ Th cells co-cultured with human FM-MSCs exhibited reduced Th1 and Th17 cell-differentiation and proliferation, with increased expression of immunosuppressive molecules including indoleamine 2,3-dioxygenase 2 and IL-6 in co-cultured FM-MSCs. These results suggest that intravenous administration of allogenic FM-MSCs ameliorates EAM via the suppression of Th1/Th17 immunity.

Intraperitoneal and intravenous deliveries are not comparable in terms of drug efficacy and cell distribution in neonatal mice with hypoxia-ischemia

BACKGROUND AND PURPOSE Most therapeutic agents are administered intravenously (IV) in clinical settings and intraperitoneally (IP) in preclinical studies with neonatal rodents; however, it remains unclear whether intraperitoneal (IP) injection is truly an acceptable alternative for intravenous (IV) injection in preclinical studies. The objective of our study is to clarify the differences in the therapeutic effects of drugs and in the distribution of infused cells after an IP or IV injection in animals with brain injury. METHODS Dexamethasone or MK-801, an N-methyl-d-aspartate receptor antagonist was administered either IP or IV in a mouse model of neonatal hypoxic-ischemic encephalopathy. Green fluorescent protein-expressing mesenchymal stem cells (MSCs) or mononuclear cells (MNCs) were injected IP or IV in the mouse model. Two hours and 24h after the administration of the cells, we investigated the cell distributions by immunohistochemical staining. We also investigated distribution of IV administered MNCs labeled with 2-[18F]fluoro-2-deoxy-d-glucose in a juvenile primate, a macaque with stroke 1h after the administration. RESULTS IP and IV administration of dexamethasone attenuated the brain injury to a similar degree. IP administration of MK-801 attenuated brain injury, whereas IV administration of MK-801 did not. The IV group showed a significantly greater number of infused cells in the lungs and brains in the MSC cohort and in the spleen, liver, and lung in the MNC cohort compared to the IP group. In the macaque, MNCs were detected in the spleen and liver in large amounts, but not in the brain and lungs. CONCLUSIONS This study demonstrated that the administration route influences the effects of drugs and cell distribution. Therefore, a preclinical study may need to be performed using the optimal administration route used in a clinical setting.

Mild intrauterine hypoperfusion reproduces neurodevelopmental disorders observed in prematurity OPEN

Severe intrauterine ischemia is detrimental to the developing brain. The impact of mild intrauterine hypoperfusion on neurological development, however, is still unclear. We induced mild intrauterine hypoperfusion in rats on embryonic day 17 via arterial stenosis with metal microcoils wrapped around the uterine and ovarian arteries. All pups were born with significantly decreased birth weights. Decreased gray and white matter areas were observed without obvious tissue damage. Pups presented delayed newborn reflexes, muscle weakness, and altered spontaneous activity. The levels of proteins indicative of inflammation and stress in the vasculature, i.e., RANTES, vWF, VEGF, and adiponectin, were upregulated in the placenta. The levels of mRNA for proteins associated with axon and astrocyte development were downregulated in fetal brains. The present study demonstrates that even mild intrauterine hypoperfusion can alter neurological development, which mimics the clinical signs and symptoms of children with neurodevelopmental disorders born prematurely or with intrauterine growth restriction.

Evaluations of Intravenous Administration of CD34+ Human Umbilical Cord Blood Cells in a Mouse Model of Neonatal Hypoxic-Ischemic Encephalopathy

Several cell therapies have been explored as novel therapeutic strategies for neonatal encephalopathy because the benefits of current treatments are limited. We previously reported that intravenous administration of human umbilical cord blood (hUCB) CD34+ cells (hematopoietic stem cells/endothelial progenitor cells) at 48 h after insult exerts therapeutic effects in neonatal mice with stroke, i.e., permanent middle cerebral artery occlusion. Although neonatal stroke and hypoxic-ischemic encephalopathy (HIE) are grouped under the term “neonatal encephalopathy,” their pathogenesis differs. However, little is known about the differences in the effects of the same treatment between these 2 diseases. In this study, we investigated whether the same treatment protocol exerts therapeutic effects in neonatal mice with HIE. The treatment significantly ameliorated the decreased cerebral blood flow in the ischemic penumbra. Although the cylinder and rotarod tests showed a trend of amelioration of behavioral impairments from the treatment, these were not statistically significant. Morphological brain injuries were not altered by treatment. The cell administration did not cause any adverse effects apart from hyperactivity in the open-field test. Some of these findings are consistent with the results obtained in our previous study using a stroke model, but others are not. This study suggests that the treatment protocol needs to be optimized for each pathological condition.

A Rat Model of Mild Intrauterine Hypoperfusion with Microcoil Stenosis

Intrauterine hypoperfusion/ischemia is one of the major causes of intrauterine/fetal growth restriction, preterm birth, and low birth weight. Most studies of this phenomenon have been performed in either models with severe intrauterine ischemia or models with gradient degree of intrauterine hypoperfusion. No study has been performed in a model on uniform mild intrauterine hypoperfusion (MIUH). Two models have been used for studies of MIUH: a model based on suture ligation of either side of the arterial arcade formed with the uterine and ovarian arteries, and a transient model based on clipping the bilateral ovarian arteries and aorta having patency. Those two rodent models of MIUH have some limitations, e.g., not all fetuses are subjected to MIUH, depending on their position in the uterine horn. In our MIUH model, all fetuses are subjected to a comparable level of intrauterine hypoperfusion. MIUH was achieved by mild stenosis of all four arteries feeding the uterus, i.e., the bilateral uterine and ovarian arteries. Arterial stenosis was induced by metal microcoils wrapped around the feeding arteries. Producing arterial stenosis with microcoils allowed us to control, optimize, and reproduce decreased blood flow with very little inter-animal variability and a low mortality rate, thus enabling accurate evaluation. When microcoils with an inner diameter of 0.24 mm were used, the blood flow in both the placenta and fetus was mildly decreased (approximately 30% from the pre-stenosis level in the placenta). The offspring of our MIUH model clearly demonstrates long-lasting alterations in neurological, neuroanatomical and behavioral test results.