创新背景

将干细胞转化为其他类型的细胞并不是什么新想法。目前已有几种方法，但结果还有待改进。通常情况下，程序化干细胞在实验室培养时不会正确成熟，因此，为实验寻找成年神经元细胞的研究人员可能最终会得到胚胎神经元，这些神经元将无法模拟迟发性精神疾病和神经退行性疾病。

 

创新过程

利用CRISPR基因编辑技术，由Rooney家族生物医学工程副教授、高级基因组技术中心主任Gersbach领导的实验室创造了一种方法，来识别哪些基因活性的主控因子对生成良好神经元至关重要。

CRISPR技术最常用于编辑DNA序列，被称为“基因组编辑”，即Cas9蛋白质与一个引导RNA结合，引导Cas9在特定位置切割DNA，导致DNA序列的变化。DNA编辑被广泛用于改变基因序列，但在基因被关闭的情况下没有帮助。

然而，一种失活的Cas9 (dCas9)蛋白会附着在DNA上，而不会切断它。事实上，如果没有另一个分子附着或吸收到它上面，它通常不会做任何事情。Gersbach和他的同事们此前报道了多种方法，将不同的分子结构域附加到dCas9蛋白罐上，这将告诉细胞打开一个基因并重塑染色质结构。

 

 

当Black加入Gersbach的实验室时，他对使用这些工具打开基因，将一种细胞类型转化为另一种细胞类型，从而创建更好的疾病模型很感兴趣。

2016年，Black和Gersbach报道了一种使用基于crispr的基因激活剂打开基因网络的方法，该方法将成纤维细胞（构成结缔组织的一种容易接近的细胞类型）转化为神经元细胞。这项研究的目标是已知与神经元规范相关的基因网络，但没有生成具有有效疾病模型所需的所有属性的细胞。然而，生成所需细胞的正确基因网络尚不清楚，人类基因组中编码了数千种可能性。因此，Black和Gersbach设计了一种策略，在一个单一的实验中测试所有的网络。

他们从多能干细胞开始，因为这种细胞类型应该能够变成人体中的任何其他细胞。为了从干细胞中制造成熟的神经元，研究小组对干细胞进行了改造，使其在变成神经元后发出红色荧光。荧光越亮，对神经元命运的推动就越强。然后，他们制作了一个包含数千个导向rna的集合库，针对人类基因组中所有编码转录因子的基因。转录因子是基因网络的主要调节因子，所以要制造所需的神经元，他们必须打开所有正确的转录因子。

他们将CRISPR基因激活子和导向RNA文库引入干细胞，使每个细胞只接收一个导向RNA，从而开启其特定对应的转录因子基因靶点。然后，他们根据细胞变红的程度对细胞进行分类，并对红色最多和最少的细胞中的引导rna进行排序，这就告诉他们，当打开哪些基因时，细胞的神经元数量更多或更少。

当他们分析用导向rna工程的干细胞的基因表达时，结果表明相应的细胞产生了更具体、更成熟的神经元类型。他们还发现，当被同时瞄准时，基因会一起起作用。此外，实验揭示了拮抗干细胞神经元承诺的因素，当他们使用这些基因的基于crispr的阻遏剂时，他们也可以增强神经元的特异性。

 

 

然而，这些结果都只是测量神经元的标记物。为了知道这些经过改造的细胞是否真正再现了更成熟神经元的功能，他们需要测试它们传输电信号的能力。

为此，他们求助于斯科特·索德林教授，他是乔治·巴思·盖勒分子生物学研究杰出教授，也是杜克大学细胞生物学系主任。索德林实验室的研究生Shataakshi Dube使用了一种被称为膜片钳电生理学的技术来测量新形成的神经元内部的电信号。用一个非常小的移液管在细胞上戳一个小洞，她就可以看到神经元内部，看它是否在传输被称为动作电位的电信号。如果是这样，研究小组就知道神经细胞已经适当成熟了。事实上，激活一对特定转录因子基因的神经元在功能上更加成熟，更频繁地释放更多的动作电位。

从干细胞到成熟神经元细胞的过程只需要7天，与其他需要数周或数月的方法相比，时间大大缩短。这一更快的时间表有可能显著加速神经疾病新疗法的开发和测试。

 

创新关键点

研究人员从多能干细胞开始，因为这种细胞类型应该能够变成人体中的任何其他细胞。为了从干细胞中制造成熟的神经元，研究小组对干细胞进行了改造，使其在变成神经元后发出红色荧光。荧光越亮，对神经元命运的推动就越强。

 

创新价值

这种新方法有制造成熟成年神经元的方法的潜力，但它可以应用于任何细胞类型的编程。

创造更好的细胞将在许多方面帮助研究人员。阿尔茨海默病、帕金森病和精神分裂症等疾病通常发生在成年人身上，由于在实验室中制造正确的细胞具有挑战性，因此很难进行研究。这种新方法可以让研究人员更好地模拟这些疾病和其他疾病。它还可以帮助药物筛选，因为不同的细胞对药物的反应不同。

更广泛地说，同样的筛选转录因子基因和基因网络的方法可以用来改进制造任何类型细胞的方法，这可能会对再生医学和细胞治疗产生变革。

 

Stem cells can be transformed into desired cell types using CRISPR markers

Using CRISP Gene editing, a laboratory led by Gersbach, Rooney family associate professor of Biomedical Engineering and director of the Center for Advanced Genome Technology, created a method to identify which master factors of gene activity are critical for generating good neurons.

CRISPR technology is most commonly used to edit DNA sequences, known as "genome editing," in which the Cas9 protein binds to a guide RNA that directs Cas9 to cut DNA at specific locations, resulting in changes in the DNA sequence. DNA editing is widely used to change gene sequences, but it doesn't help when genes are turned off.

However, an inactivated Cas9 (dCas9) protein attaches to DNA without cutting it off. In fact, it usually doesn't do anything without another molecule attached to it or absorbed into it. Gersbach and colleagues previously reported a variety of ways to attach different molecular domains to the dCas9 protein canister, which would tell cells to turn on a gene and reshape chromatin structure.

When Black joined Gersbach's lab, he was interested in using these tools to turn on genes that could transform one cell type into another to create better disease models.

In 2016, Black and Gersbach reported a method for turning on gene networks using CRISPR-based gene activators that turn fibroblasts, an easily accessible cell type that makes up connective tissue, into neuronal cells. The study targeted gene networks that are known to be associated with neuronal specification, but did not generate cells with all the attributes required for a valid disease model. However, the correct network of genes to generate the required cells is not known, and thousands of possibilities are encoded in the human genome. Therefore, Black and Gersbach devised a strategy to test all networks in a single experiment.

They started with pluripotent stem cells because these cell types should be able to turn into any other cell in the body. To create mature neurons from stem cells, the team engineered the stem cells so that they fluoresced red after becoming neurons. The brighter the fluorescence, the stronger the push on the neuron's fate. They then made a library of thousands of guide Rnas for all the genes encoding transcription factors in the human genome. Transcription factors are the master regulators of gene networks, so to make the neurons they need, they have to turn on all the right transcription factors.

They introduced CRISPR gene activators and guide RNA libraries into stem cells so that each cell received only one guide RNA to turn on its specific transcription factor gene targets. They then sorted the cells by how red they turned, and ranked the guide Rnas in the cells with the most and least red, which told them which genes had more or fewer neurons when turned on.

When they analyzed gene expression in stem cells engineered with directed RNA, the results showed that the corresponding cells gave rise to more specific, mature types of neurons. They also found that genes act together when targeted simultaneously. In addition, experiments revealed factors that antagonize stem cell neuronal commitment, and when they used CRISPR-based repressors of these genes, they could also enhance neuronal specificity.

However, these results are only markers that measure neurons. To find out whether the engineered cells truly reproduce the functions of more mature neurons, they will need to test their ability to transmit electrical signals.

To do so, they turned to Professor Scott Soderling, the George Barth Geller Distinguished Research Professor of Molecular Biology and chair of Duke's Department of Cell Biology. Shataakshi Dube, a graduate student in Soderling's lab, used a technique known as patch-clamp electrophysiology to measure electrical signals inside newly formed neurons. By poking a small hole in the cell with a very small pipette, she could see inside the neuron to see if it was transmitting electrical signals called action potentials. If so, the team knew the nerve cells had matured properly. In fact, neurons that activate a particular pair of transcription factor genes are functionally more mature and release more action potentials more frequently.

The process from stem cells to mature neuronal cells takes only seven days, a significant reduction compared with other methods that take weeks or months. This faster timeline has the potential to significantly accelerate the development and testing of new therapies for neurological diseases.