The hUCMSCs were obtained from Shanghai Quansheng Biotechnology Co., Ltd., which were extracted and cultured from the umbilical cords of newborns. The umbilical cords were obtained from Dongguan Chang’an Hospital and approved the inspection of the ethics committee of the hospital. hUCMSCs were routinely tested for surface markers, three-line differentiation function, immunomodulatory function, fungi, bacteria, and other indicators before use. hUCMSCs were approved for use after meeting the certificate of analysis (COA) release criteria.

The use of experimental animals was approved by the Laboratory Animal Ethics Committee of Youji (Tianjin) Pharmaceutical Technology Co., Ltd. In this experiment, the rats were anesthetized with isoflurane and fixed in the supine position. The skin was incised along the median line of the neck to expose the right common carotid artery. The nerves and fascia around the blood vessels from the bifurcation of the common carotid artery to the skull base were carefully removed. A nylon suture (Diameter 0.28 mm) was introduced into the internal carotid artery from the distal end of the external carotid artery and inserted into the circle of Willis at the middle cerebral artery to effectively blocks the middle cerebral artery. The length of the inserted suture was 18 to 20 mm from the bifurcation of the common carotid artery. The free end of the external carotid artery was then ligated with an intraluminal suture to prevent bleeding. The subcutaneous fascia and skin were sutured layer by layer, and penicillin was dripped locally to prevent infection. In the control group, only the internal carotid artery was separated. Two hours after the start of MCAO, the thread in the lumen of the internal carotid artery was carefully extracted to reperfuse the artery, followed by feeding the animals.

The Neurological impairment degree score (NDS) was assessed after 2 hours of ischemia and reperfusion (23). A point above 8 meant a successful model (24). Samples with 10-12 points were selected for downstream experiments. The detailed scoring criteria are indicated in Supplementary Table 1 .

In the mechanism experiment, hUCMSCs were injected into the rat tail vein at a dose of 1 mL/kg based on the standard of 1×10⁷ cells/ml on d1 after modeling. After 7 days, the second administration was carried out. The model group were given the same dose of PBS. Each group consisted of 12 rats, and the serum and brain tissues of MCAO rats on the d1, d4, d7, and d14 were collected after hUCMSCs or PBS infusion. Normal rat blood and brain tissue lysates were used as initial control data before the experiment was carried out. The protein levels of SDF-1α, IL-10, IL-6, TNF-α, BDNF, IL-1β, and VEGF-A in serum and brain tissue lysates were measured using Luminex multi-factor kits (R&D). Meanwhile, the protein levels of CXCR-4, NeuN, and Nestin in the brain tissues of d14 rats were determined by immunofluorescence. The pathological structures in the brain tissues were examined by HE staining ( Figure 1A ).

In the time window experiments, the hUCMSCs were injected into the rat tail vein on d4, d7, and d14 after MCAO, corresponding to the acute, sub-acute, and sequelae phases of IS, respectively. The dose of hUCMSCs was 1×10⁷cells/kg, and the second treatment was performed 7 days after the first administration. Each time point was divided into control group, model group and treatment group with 10 rats in each subgroup. The clinical observation period for the rats lasted until 28 days after the first administration. The balance beam test, forelimb muscle strength test, and limb placement test were all performed on the rats before administration and on the 7d, 14d, 21d, and 28d after administration. The infarct area was detected by triphenyl tetrazolium chloride (TTC, Phygene) staining at the end of the experiment ( Figure 1B ).

Based on the results of the time window experiments, the optimum time point for the injection of hUCMSCs was determined, and the dose-effect relationship experiment was carried out. The low-dose, medium-dose, and high-dose hUCMSCs treatment groups were set up, with doses of 5×10⁶ cells/kg, 1×10⁷ cells/kg, and 2×10⁷ cells/kg, respectively, with 12 rats in each group. The balance beam test, forelimb muscle strength test, and limb placement test were performed on 7d, 14d, 21d, and 28d after the first administration, and the cerebral infarction area was detected by TTC staining at the end of the experiment ( Figure 1C ).

The balance beam, forelimb placement, and forelimb muscle strength of MCAO rats were tested before the first administration and on 7d, 14d, 21d, and 28d after administration to evaluate the recovery of neurological/motor function of MCAO rats (25).

Following the motor function experiments, the rats were sacrificed by cervical vertebra detachment after isoflurane gas anesthesia and bleeding. The brain tissues were harvested, stored in a -80°C freezer, and then sectioned from front to back with a thickness of 2 mm. Brain tissue sections were placed in 2% red tetrazolium solution and incubated at 37 °C for 5 minutes. Infarct tissues were white and non-infarct tissues were red. The cerebral infarction area (%) was measured using Image J software. Bias due to brain edema or brain atrophy was excluded using a modified formula as follows:

Cerebral infarction area (%) = (area of the hemisphere on the non-infarct side – area of normal brain tissue on infarct side)/(area of the hemisphere on non-infarct side × 2)× 100% (26, 27).

The SPSS 22 software was used for statistical analysis. One-way analysis of variance (ANOVA) was used for multi-group comparison. Meanwhile, the independent data of the two groups were analyzed using a t-test. A P<0.05 was considered statistically significant. The graphical representations of the results were constructed using the Graphpad Prism 8 software.

Cell surface markers of hUCMSCs were detected by flow cytometry assay, and the result indicates that they are CD14, CD19, CD31, CD34, CD45, HLA-DR negative and CD73, CD90, CD105 positive ( Figure 2A ). hUCMSCs were detected by immunocytochemistry for their lipogenic, osteogenic, and chondrogenic differentiation potential in the different induction differentiation media. The results showed that hUCMSCs could also successfully transdifferentiate into adipocytes (analyzed by Oil Red O staining), osteoblasts (analyzed by Alizarin Red staining), and chondrocytes (analyzed by Arsenic Blue staining) ( Figures 2B–D ). Together, these results indicate that hUCMSCs are characterized by high purity and with multidirectional differentiation capabilities.

The administration of hUCMSCs led to a significant reduction in the levels of the proinflammatory factors IL-1β, IL-6, and TNF-α and an increase in the levels of the anti-inflammatory factor IL-10 in serum ( Figures 3D–G ), but no significant difference in the levels of these factors has been observed in the brain tissue samples ( Figures 4C–E ). Furthermore, the hUCMSCs effectively promoted the secretion of the SDF-α, BDNF, and VEGF in the serum ( Figures 3A–C ). In addition, the levels of SDF-α and BDNF in the brain tissues also showed an increasing trend ( Figures 4A, B ); particularly, these factors were relatively high in the sub-acute phase of MCAO. Unfortunately, the levels of IL-6 and VEGF-A were lower than the sensitivity limit of the Luminex multi-factor kits. Based on these results, the hUCMSCs administration showed a good immunomodulatory function in the treatment of IS, improved the secretion of chemokines and neurotrophic factors in the sub-acute phase.

The pathological and structural analysis of rat brain tissues in each group by HE showed that the normal control group had an intact brain structure with clear borders and abundant cell numbers ( Figure 5A ), while the model group had increased neuronal death and inflammatory cells, as well as vacuolar degeneration and edema formation ( Figure 5B ). Meanwhile, the treatment group was characterized by relatively more neuronal cells, regular brain tissue and cell arrangement, and a more complete structure ( Figure 5C ). Compared with the model group, the striatum of the treatment group showed a smaller number of apoptotic cells ( Figures 5D–F ). These results indicate that hUCMSCs administration via the tail vein could effectively inhibit pathological damage, reduce inflammatory infiltration, and promote tissue repair to some extent in the brain tissues of MCAO rats.

Immunofluorescence was used to evaluate the protein expression of CXCR4, Nestin, and NeuN in the brain tissue of MCAO rats in the sequelae phase ( Figure 6 ). Upon hUCMSCs administration in sequelae phase MCAO rats, an apparent upregulation of CXCR4 was observed in the treatment group compared with the model group ( Figures 6A–C ). The expression of Nestin, a neural stem cell marker (28), was significantly increased after the injection of hUCMSCs ( Figures 6D–F ). Consistently, the expression of NeuN, a neuronal marker (29), also showed an upward trend ( Figures 6G–I ). These findings suggest that the hUCMSCs administration could promote the expression of chemokines in the brain tissue of MCAO rats at the sequelae phase and promote the repair, proliferation, and maturation of neurons.

The MCAO rat models were monitored for totally 28 days from the first injection of the hUCMSCs. Each experimental group showed motor function, balance beam crawling disorder, gait side-slip, forelimb placement rate, and forelimb muscle strength reduction ( Figures 7 ). The d4 and d7 hUCMSCs treatment could significantly improve the motor function and forelimb placement rate from d21 to d28 after administration, and the gait side-slip was significantly reduced ( Figures 7A, B, D, E ). The d14 hUCMSCs treatment had a relatively weak lower trend in improving the balance beam test scores and forelimb placement rate ( Figures 7C, F ). The d4, d7, and d14 hUCMSCs treatment had significantly improved forelimb muscle strength ( Figures 7G–I ). Based on the results of the balance beam test, forelimb placement test, and forelimb muscle strength test, the d4 hUCMSCs treatment group showed a greater trend of score improvement in each group. Meanwhile, the d7 hUCMSCs treatment group had a slight score improvement, but the difference was relatively lower than that of the d4 hUCMSCs treatment group. In contrast, there was only a statistically significant difference in the forelimb muscle strength scores of the d14 hUCMSCs treatment group. These findings suggest that the hUCMSCs administration can promote the recovery of motor function in the MCAO model, with earlier intervention treatment time leading to a better therapeutic effect.

The brain tissues of rats in each group were stained with TTC 28 days after hUCMSCs administration ( Figures 8 ). In the model establishment, the right cerebral infarction area of the MCAO rats was larger by inserting the suture from the right common carotid artery. The brain tissues in the model group and the treatment group showed a white cerebral infarction area, which was analyzed by the Image J software ( Figures 8A–I ). Compared with the model group, The d4 ( Figures 8B, C ), d7 ( Figures 8E, F ), and d14 ( Figures 6H, I ) hUCMSCs treatments showed less white areas after modeling. No white necrosis area in the control group at each time point has been observed ( Figures 8A, D, G ). In contrast with the model control group, the cerebral infarction area was reduced in the hUCMSCs treatment group on d4, d7, and d14 post-modeling, and the improvement rates were 35%, 30.4%, and 14.1%, respectively ( Figures 8J–L ). These findings indicate that within a certain time limit, early hUCMSCs treatment could reduce the cerebral infarction area and improve the therapeutic effect.

The time windows experiment showed that the hUCMSCs had a better therapeutic effect in acute IS. Based on this observation, the dose effect of the hUCMSCs was performed 4 days after modeling. In the balance beam test, the low-dose hUCMSCs treatment had no significant improvement effect ( Figure 9A ). Furthermore, the low-dose, medium-dose, and high-dose stem cell treatment could significantly improve the motor function of rat models on d14 to d21 after administration, while the gait side-slip was significantly reduced ( Figure 9A ). Additionally, the frequency of forelimb placement and forelimb muscle strength was significantly increased in all treatment groups ( Figures 9B, C ). There was a significant dose-dependent improvement in the comprehensive scores of neurological/motor functions in the high-dose treatment group compared with the medium-dose and low-dose treatment groups.

In the dose-effect relationship experiments, the brain tissues of rats in each group were stained with TTC 28 days after hUCMSCs administration ( Figures 10A–E ). The low-dose treatment group exhibited a slight reduction in the cerebral infarction area; however, no statistically significant value was observed. Meanwhile, the medium-dose treatment group and the high-dose treatment group could significantly reduce the cerebral infarction area. The three treatment groups improvement rates of cerebral infarction area were 15.8%, 21.7%, and 22.9%, respectively ( Figure 10F ). These results show that the middle and high doses of hUCMSCs had a higher effect on reducing the cerebral infarction area in MCAO rats.