摘要目的:探讨改良的双环己酮草酰二腙（CPZ）诱导小鼠多发性硬化（MS）脱髓鞘及髓鞘再生模型的建立方法，并分析小鼠MS模型脱髓鞘及髓鞘再生的影像标记。方法:收集30只C57BL/6雄性小鼠，并以羧甲基纤维素钠（CMCNa）适应灌胃1周后，随机分3组，分别造模成对照组（ n=10）、脱髓鞘组（ n=10）及髓鞘再生组（ n=10）。对照组小鼠立即进行MR扫描及病理取材；脱髓鞘组小鼠每日1次使用CPZ-CMCNa悬浊液连续灌胃诱导急性脱髓鞘6周，然后进行MR扫描及病理取材；髓鞘再生组小鼠每日1次使用CPZ-CMCNa悬浊液连续灌胃6周，停止灌胃并恢复正常饮食4周，最后进行MR扫描及病理取材。以胼胝体喙部（rCC）、喙部两侧正常表现白质（NAWM）及双侧大脑皮层（Cx）作为ROI，采用单因素方差分析比较3组归一化T 2WI（T 2-normalized）、各向异性分数（FA）、平均扩散率（MD）、轴向扩散系数（AD）及径向扩散系数（RD）变化及差异。 结果:成功建立了小鼠MS模型脱髓鞘及髓鞘再生模型。对照组、脱髓鞘组和髓鞘再生组rCC的T 2-normalized值分别为0.47±0.03、0.72±0.04、0.54±0.04，差异具有统计学意义（ F=90.511， P<0.05）；组间两两比较差异具有统计学意义（ P<0.05）。3组NAWM及Cx的T 2-normalized值差异均无统计学意义（ P>0.05）。对照组、脱髓鞘组和髓鞘再生组rCC的FA值（分别为0.36±0.04、0.29±0.03、0.32±0.05），MD值［分别为（0.572±0.015）、（0.598±0.034）、（0.626±0.043）×10 -3 mm 2/s］，AD值［（0.79±0.04）、（0.77±0.06）、（0.83±0.04）×10 -3 mm 2/s］及RD值［（0.46±0.02）、（0.51±0.03）、（0.53±0.05）×10 -3 mm 2/s］差异均有统计学意义（ P<0.05），且两两比较发现脱髓鞘组与对照组FA值差异具有统计学意义（ P<0.05）；髓鞘再生组与对照组MD值差异有统计学意义（ P<0.05）；髓鞘再生组与脱髓鞘组AD值差异具有统计学意义（ P<0.05），脱髓鞘组、髓鞘再生组与对照组RD值差异均有统计学意义（ P<0.05）。3组的NAWM及Cx各项扩散张量成像（DTI）参数差异均无统计学意义（ P>0.05）。劳克坚牢蓝-伊红染色显示脱髓鞘组胼胝体喙部髓鞘丢失，髓鞘再生组胼胝体喙部部分再生修复。 结论:改良CPZ-CMCNa模型能够选择性诱导rCC脱髓鞘及髓鞘再生；利用T 2WI联合DTI可以定量检测改良CPZ-CMCNa模型rCC脱髓鞘和髓鞘再生变化，为研究MS病灶变化提供理论依据。

abstractsObjective:To explore the method of establishing a modified demyelination and myelination regeneration model induced by dicyclohexanone oxalyl dihydrazone (CPZ) in mice with multiple sclerosis (MS), and to analyze the image markers of demyelination and myelination regeneration in mouse MS model.Methods:After the intragastrically administered with sodium carboxymethyl cellulose (CMCNa) for one week, a total of 30 C57BL/6 male mice were randomly divided into the control group ( n=10), the demyelination group ( n=10), and the remyelination group ( n=10). The mice of the control group were immediately performed MR scanning and pathological specimen obtaining; the mice in the demyelination group were administered with intragastrical CPZ-CMCNa once a day for 6 weeks for inducing demyelination, then received MR scanning and specimen obtaining with the same protocols used in control group; the mice in the remyelination group were administered with intragastrical CPZ-CMCNa once a day for six weeks for demyelination, then CPZ was withdrawn and normal diet was given for another four weeks. Then MR scanning and specimen obtaining were performed with the same protocols used in the other two groups. Regions of interest (ROIs) were set at the rostrum of corpus callosum (rCC), the bilateral normal appearing white matters (NAWM) of the rostrum of corpus callosum, and the bilateral cerebral cortex (Cx). The normalized T 2WI (T 2-normalized), fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) values were compared among the three groups by one-way ANOVA. Results:The demyelination and remyelination mice model of MS were successfully established. The T 2-normalized values of rCC in control group, demyelination group and remyelination group were 0.47±0.03, 0.72±0.04, 0.54±0.04, respectively, with statistically significant difference found ( F=90.511, P<0.05). Post-hoc multiple comparisons showed significant differences among those groups ( P<0.05). There was no significant difference of T 2-normalized value in NAWM and Cx among the three groups ( P>0.05). Moreover, there were significant differences in the FA values (0.36±0.04, 0.29±0.03, and 0.32±0.05), the MD values [(0.572±0.015), (0.598±0.034), and (0.626±0.043)×10 -3 mm 2/s], the AD values [(0.79±0.04), (0.77±0.06), and (0.83±0.04)×10 -3 mm 2/s], and the RD values [(0.46±0.02), (0.51±0.03), and (0.53±0.05)×10 -3 mm 2/s] of rCC of the control group, the demyelination group, and the remyelination group (all P<0.05). Significant difference was found in FA values between the demyelination group and the control group ( P<0.05), and in MD values between the remyelination group and the control group ( P<0.05), as well as in AD values between the remyelination group and the demyelination group ( P<0.05). There were also significant differences in RD values between the remyelination group and the control group, and the demyelination group and the control group (all P<0.05). However, no significant difference was found in all diffusion tensor imaging (DTI) metrics of NAWM and Cx among the three groups (all P>0.05). The LFB-eosin staining showed that the myelin sheath of rCC was lost in the demyelination group, and the rCC was partially regenerated and repaired in the remyelination group. Conclusion:The modified CPZ-CMCNa model can selectively induce demyelination and remyelination of rCC, and the changes of demyelination and remyelination of rCC in the modified CPZ-CMCNa model can be quantitatively detected by T 2WI combined with DTI, which might provide related theoretical basis for the study on dynamic changes of MS lesions.