实验海洋生物学重点实验室知识库

软体动物贝壳形成是重要的科学问题，对动物演化发育、水产育种乃至材料科学都有重要的意义。软体动物的成体贝壳在不同物种中呈现出显著的多样性，但幼虫贝壳的形成过程却非常保守，因而是解析贝壳发生机制的关键阶段。个体发育过程中，贝壳发育的前体组织称为贝壳发育区（shell field）。研究表明大多数软体动物（有壳亚门）的贝壳发育区形态发生（morphogenesis）中都会经历一个共有的内陷（invagination）过程。同时，在贝壳发育区的早期形成阶段即形成了多种细胞类群，它们互相协调，共同保障了初生贝壳的正常形成。贝壳发育区形态发生的机制及其内部细胞类群是贝壳发育研究的重要方面。

本研究以腹足纲软体动物北戴河笠贝（Lottia peitaihoensis）为研究对象，对细胞骨架分子在贝壳发育区形态发生中的功能，贝壳发育区细胞群体的类型、空间关系、发育中的动态变化及其与幼虫贝壳的相关性进行了系统解析，鉴定了一条新的贝壳发育调控通路，揭示了贝壳发育区的精细结构和发育模式，在分子和细胞层面上丰富了贝壳发育机制理论。主要研究结果如下：

1. 实验证实了肌动-肌球蛋白细胞骨架系统（actomyosin networks）及其上游调控基因在贝壳发育区形态发生中的功能。课题组前期研究发现，微丝（F-actin）在贝壳发育区特异性富集，提示肌动-肌球蛋白细胞骨架系统在贝壳发育中发挥功能。本研究中，首先在笠贝的早期贝壳发育过程中鉴定到了肌动-肌球蛋白系统中的关键分子，非肌肉型肌球蛋白II（nonmuscle myosin II, NM II），并通过免疫荧光染色实验发现磷酸化的NM II（pNM II）与F-actin在贝壳发育区呈现共定位关系。在贝壳发育区形态发生的关键阶段（受精后6-8小时，6-8 hpf），用特异的小分子抑制剂阻断NM II的磷酸化，会导致幼虫无法发育出完整的壳板。进一步研究表明，抑制pNM II会导致贝壳发育区形态发生的数个关键事件受到干扰，包括贝壳发育区内陷、细胞形状变化以及细胞重排等，但对该区域细胞的分化状态没有明显影响。之后，为了探索该通路的上游调控分子，继续研究了Rho家族小G蛋白（Rho GTPases）中的3个主要成员RhoA (Ras homolog family member A)、Rac1 (Rac family small GTPase 1)和Cdc42 (Cell division cycle 42)的作用。对每个成员使用2-3种小分子抑制剂来抑制其活性，然后分析不同处理对贝壳发育区和幼虫壳板的影响。结果表明，只有Cdc42可以特异地调控贝壳发育区形态发生的关键事件并导致幼虫壳板败育。免疫荧光染色实验表明，Cdc42蛋白在贝壳发育区细胞的顶端富集，并与此处聚集的F-actin共定位。对Cdc42进行抑制剂处理可以破坏这两种分子在贝壳发育区的聚集。这些研究结果鉴定了Cdc42-F-actin/NM II这一细胞骨架调控通路在早期贝壳发育区形态发生中的作用，是一条全新的贝壳发育调控通路。我们据此提出了细胞骨架分子调控贝壳发育的假说：Cdc42在贝壳发育区细胞的顶端招募F-actin（或actomyosin），然后在此产生机械力，导致一系列形态发生过程（细胞形状变化、内陷和细胞重排），共同保障后续幼虫壳板的形成。

2. 结合形态学和基因表达数据，初步解析了贝壳发育区的细胞类群组成。前述对细胞骨架分子功能的研究中发现，贝壳发育区包含形态特征和发育模式迥异的多个细胞类群，这也与本课题组前期研究中发现贝壳发育基因存在多种表达模式这一事实相符。然而，目前我们对贝壳发育区中细胞群体的类型、空间关系及其在发育过程中的动态变化了解十分有限。为了鉴定贝壳发育区内的细胞类群，通过多种技术手段，对贝壳发育区细胞学结构的时空变化进行了细致的研究。微分干涉显微镜观察及碳酸钙染色的结果显示，10 hpf的笠贝幼虫中开始出现壳板，至18 hpf时已经发育出具腹足纲特征的幼虫壳板，覆盖在贝壳发育区的表面。为了揭示贝壳发育区细胞变化的细节及其与壳板的相关性，对8-18 hpf的笠贝胚胎或幼虫进行F-actin染色和几丁质染色，在分子和细胞水平揭示了该区域在连续发育过程的变化细节。在短短的10小时内，贝壳发育区经历了内陷、外翻、扩大生长的过程，形成了角质层沟和钙质壳板等重要结构。根据对这些发育事件的观察，初步确定幼虫贝壳的关键发育阶段为8 hpf, 12 hpf, 18 hpf三个时间节点。接下来，为了明确3个关键发育阶段的贝壳发育区的细胞学结构，对样品进行冷冻切片和超薄切片后开展了显微观察。根据贝壳发育区不同区域的形态特征和超微结构，将其大致分为3个区域，分别为中央区域、中间区域（角质层沟处一排细胞）以及边缘区域。之后，将冷冻切片与原位杂交技术结合，基于5个已知贝壳发育基因的表达位置，将其基本定位到3个细胞类群。我们基于上述形态学、细胞学和分子生物学等多个水平的证据，初步揭示了贝壳发育区不同的细胞类群及其发育过程的细节，为后续更高分辨率的研究奠定了坚实基础。

3. 基于单细胞转录组数据，结合双色荧光原位杂交技术，高分辨率揭示了贝壳发育区的精细结构和发育历程。前述研究虽然初步鉴定了3个细胞类群，但受限于已知的贝壳发育基因数量，分辨率不够高。为了解决这一问题，开展了贝壳发育关键时期胚胎的单细胞转录组测序，以鉴定更多的贝壳发育基因。首先对8 hpf笠贝胚胎进行解离，将其制备成高质量的单细胞悬液之后上机测序，共获得13014个细胞的转录组数据，可划分为21个细胞群体。与已知贝壳发育基因对照，鉴定到与贝壳发育相关的有5个细胞群体，其内的大量特征性表达基因提供了全新的贝壳发育候选基因。随后，对新鉴定的10个贝壳发育基因开展了原位杂交实验。结果表明，它们都特异地表达在贝壳发育区内，覆盖了整个贝壳发育区，且不同细胞群内的特征基因表达在不同区域，与预测相符。为了进一步揭示各细胞类群之间的空间关系，对5个细胞类群的特征基因进行两两组合的双色原位杂交实验，成功解析了各细胞类群的空间关系：这5个细胞类群彼此之间界限分明，从内到外依次为cluster5, 9, 7, 15, 11，在贝壳发育区内呈现出“同心圆”的排布模式，覆盖了整个贝壳发育区。接下来，继续追踪了这5个细胞类群在笠贝早期贝壳发育过程中的动态变化，并用小麦胚芽凝集素染色和细胞核染色来辅助定位关键结构的位置。结果表明，从早期贝壳发育区（8 hpf）到基本成熟外套膜（18 hpf）的发育过程中，这5个细胞类群的相对位置关系没有发生改变，但部分细胞群的排布模式由早期的“同心圆”模式转变为后期间隔排列的“镶嵌型”模式。另外，除了中央区域的cluster5在此过程发生显著细胞增殖和面积扩张外，其余4个细胞类群始终仅占据1-2排细胞宽度。综上所述，我们在单细胞分辨率上成功解析了贝壳发育区的精细结构及其复杂的发育模式，创新性地鉴定到5个贝壳发育相关细胞类群。结合其内部表达的基因类型，推测中央区域被壳板覆盖的3个细胞类群（cluster5, 9, 7）可能与钙质壳板的分泌有关，中间及边缘区域的2个细胞类群（cluster15和cluster11）可能与角质层的形成相关。

综上所述，本研究从微丝在早期贝壳发育区的富集现象入手，通过功能实验证明了肌动-肌球蛋白细胞骨架系统在上游小G蛋白Cdc42的调控作用下，参与了笠贝的贝壳发育区形态发生过程和幼虫壳板的形成过程。继而通过多种技术手段，联合分析了笠贝早期贝壳发育的复杂过程和贝壳发育区细胞学结构的时空变化。最后，通过将单细胞转录组数据结合双色荧光原位杂交技术，以极高的分辨率创新性地在贝壳发育区鉴定到了5个细胞类群，并且揭示了这些细胞类群的空间关系和在发育过程中的动态变化。这些结果在分子和细胞水平上深度解析了软体动物贝壳发育区的精细结构和发育模式，为今后探究贝壳发育区内不同细胞类群的生物学功能提供了支持，有助于我们深入认识软体动物的发育和演化过程，也可对贝壳相关性状的良种培育提供基础支撑。

Molluscan shell formation is an important research topic, which can provide insights into molluscan development and evolution, aquatic breeding, and even material sciences. Adult shells exhibit remarkable diversity among different molluscan lineages, but the process of larval shell formation is generally conserved, making it a key aspect in understanding shell formation mechanisms. The precursor tissue for shell development is called the shell field. A key feature of shell field morphogenesis refers to the invagination of dorsal epithelial cells, which has been observed among different conchiferans, the major clade of extant mollusks. Additionally, it is revealed that there are various distinct cell populations in the shell field, which coordinate to ensure the correct formation of the nascent shell plate. The mechanisms of shell field morphogenesis and the cell populations within the shell field are important aspects of shell development research.

Here, we comprehensively analyzed the function of cytoskeletal molecules in shell field morphogenesis and identify the types, spatial relationships and the dynamic changes of cell populations within the shell field in the gastropod mollusc Lottia peitaihoensis. Our results identify a new regulatory cascade for shell development and reveal the elaborate structures and complex developmental patterns of the shell field at the molecular and cellular levels, which greatly add to the knowledge of shell development mechanisms. The main results are as follows:

1. We investigated the roles of actomyosin networks and their upstream molecules in shell field morphogenesis based on functional assays. We previously found that F-actin (filamentous actin) is evidently aggregated in the invaginated shell field of L. peitaihoensis, indicating the roles of actomyosin networks. Here, we revealed the aggregation of nonmuscle myosin II (NM II), the key molecule of actomyosin networks, in the shell field of L. peitaihoensis. Immunostaining revealed the colocalization of phosphorylated NM II (pNM II) and F-actin in the shell field. When inhibiting the phosphorylation of NM II using a specific inhibitor during the key stages of shell field morphogenesis (6-8 h post-fertilization, hpf), the manipulated larvae failed to develop shell plates. Further investigations revealed that key events of shell field morphogenesis were prevented by inhibiting pNM II, including invagination of the shell field, cell shape changes and cell rearrangements, but the cell specification in the shell field seems to be unaffected. Then, we investigated the roles of the Rho family of small GTPases, RhoA (Ras homolog family member A), Rac1 (Rac family small GTPase 1) and Cdc42 (Cell division cycle 42), to explore the upstream regulators of actomyosin networks. For each molecule, two or three inhibitors were used to inhibit its activity. Functional assays suggest that Cdc42 is a key modulator of a series of events during shell field morphogenesis, while the roles of RhoA and Rac1 may be nonspecific or negligible. Further investigations revealed that Cdc42 protein was aggregated at the apical side of shell field cells and colocalized with F-actin aggregation. The aggregation of the two molecules could be prevented by Cdc42 inhibitor treatment. These findings emphasize the important roles of Cdc42-actomyosin cytoskeleton in shell field morphogenesis and suggest a new regulatory cascade for shell development. Based on these results, we propose a hypothesis that cytoskeletal molecules regulate shell development: Cdc42 recruits F-actin (or actomyosin) at the apical region of the shell field cells, which then generate mechanical forces that lead to a series of morphogenetic processes (cell shape changes, invagination, and cell rearrangement), collectively ensuring the subsequent formation of larval shell plates.

2. Based on morphological and gene expression data, we preliminarily analyzed the cell populations in the shell field. During the research on the roles of actomyosin networks, we found evidence indicating that the shell field contained various cell populations with different morphologies and developmental patterns. This is consistent with our previous findings that some shell development genes exhibit distinct expression patterns. However, it is largely unknown regarding how many cell types in the shell field, their spatial inter-relationships and the dynamic changes during shell development. To identify the cell populations within the shell field, we conducted a detailed investigation of the changes of the shell field using various techniques. Differential interference contrast (DIC) microscopy and calcium carbonate staining showed that the shell plates appeared at 9-10 hpf, which grew rapidly to encompass the larval body, and the 18 hpf-larvae already exhibited the characteristic gastropod larval shell plate. As the shell plate cover on the surface of the shell field in late developmental stages, we performed F-actin and chitin staining to investigate the dynamic changes of the shell field during 8-18 hpf. The results showed that the shell field experienced complex processes including invagination, evagination, and expansion within 10 hours, during which the important structures such as the periostracal groove and calcareous shell plates formed. Based on these observations, the key stages of early shell development were determined to be 8 hpf, 12 hpf and 18 hpf. Then, to reveal the detailed structures of the cell populations within shell field, we prepared frozen sections and ultrathin sections of larvae at 8, 12 and 18 hpf and observed them using microscopes. Based on the morphological characteristics and ultrastructure of different regions, the shell field was roughly divided into three regions: the central region, intermediate region (a row of cells under the periostracal groove) and the peripheral region. We further performed in situ hybridization (ISH) of five potential shell development genes and assign their expression to the three regions. These together identified three cell populations in the shell field. These results revealed many details of the different cell populations in the shell field and their developmental processes, providing fundamental support for research with further higher resolution.

3. Based on single-cell RNA sequencing (scRNA-seq) data and two-color fluorescence ISH, we innovatively revealed the elaborate structures and developmental patterns of the shell field at a very high resolution. Although the previous research preliminarily identified three cell populations, the resolution was limited by the number of known shell development genes. To address this issue, we performed scRNA-seq of samples at the critical stage of shell development (8 hpf). After dissociating 8-hpf embryos, cell suspension was submitted to scRNA-seq. The results identified a total of 13014 single cells, which were divided into 21 cell clusters. By comparing with known shell development genes, we identified five cell clusters related to shell development. Numerous cluster-specific genes provided novel candidate shell development genes. We then performed ISH for 10 of these genes and found that they were all specifically expressed in particular regions of the shell field, together covering the entire shell field. To further explore the spatial inter-relationships between cell populations, we performed two-color ISH and successfully revealed their spatial relationships: the five cell populations had clear boundaries between each other, arranged from inside to outside in the order of cluster5, 9, 7, 15, 11, presenting a “rosette” pattern at 8 hpf. We further tracked the dynamic changes of these five cell populations during the early shell development. Wheat germ agglutinin (WGA) staining and nuclear staining were used to define the position of the periostracal groove, which served as a landmark. The results showed that the spatial inter-relationships of these five cell populations did not change significantly from the early shell field (8 hpf) to the nearly mature mantle (18 hpf), with the major exception that the arrangement patterns of some cell populations changed from the early “rosette” pattern to the subsequent “interclated” pattern. Additionally, we found that only cell cluster 5 underwent evident cell proliferation during this process, while the other four cell populations consistently comprised only 1-2 rows of cells. Together, we for the first time identified five cell populations in the shell field and revealed the elaborate structures and complex developmental patterns of the shell field at a single-cell resolution. Based on their gene expression data, we speculated that three cell populations of them (cluster5, 9, 7), which were located in the central region of the shell field and covered by the shell plate, may be related to the secretion of calcareous parts of the shell plate; the rest two cell populations (cluster 15 and cluster 11), which were located in the intermediate region and the peripheral region, may be related to the formation of the periostracum.

In conclusion, we demonstrated that the actomyosin networks and the upstream regulator Cdc42 were required for shell field morphogenesis and larval shell formation based on the results of functional assays. Then, we revealed many details of the complex processes of early shell development by means of multiple techniques. Finally, combining scRNA-seq and two-color fluorescence ISH, we identified five cell populations in the shell field and revealed their spatial relationships as well as dynamic changes during shell development at a very high resolution. Together, these results reveal the elaborate structures and the developmental patterns of the shell field at the molecular and cellular levels, which will provide fundamental support for exploring the biological functions of different cell populations within the shell field in future. These findings greatly enrich our knowledge of molluscan development and evolution, which may also provide support for aquatic breeding focusing on shell-related traits.

第1章 绪论... 1

1.1 软体动物成体贝壳的特征及形成... 1

1.1.1 成体贝壳的多样性... 1

1.1.2 成体贝壳的结构与形成... 2

1.2 软体动物幼虫贝壳的发育... 5

1.2.1 贝壳发育区的形态发生过程... 5

1.2.2 幼虫贝壳的形成... 7

1.3 软体动物早期贝壳发育机制的研究进展... 8

1.3.1 早期贝壳发育的分子机制... 8

1.3.2 早期贝壳发育的细胞机制... 11

1.4 细胞骨架及其调控分子的简介... 14

1.4.1 肌动-肌球蛋白网络的作用机制... 14

1.4.2 肌动蛋白细胞骨架上游的Rho家族小G蛋白的简介... 16

1.5 无脊椎动物单细胞转录组测序的研究进展... 17

1.5.1 10×Genomics单细胞转录组测序的原理... 17

1.5.2 单细胞转录组测序在无脊椎动物发育生物学研究中的应用... 18

1.6 本研究的内容和意义... 19

第2章 肌动-肌球蛋白及其上游分子在早期贝壳发育的功能研究... 23

2.1 研究背景... 23

2.2 材料与方法... 24

2.2.1 实验材料的收集与固定... 24

2.2.2 针对NM II, RhoA, Rac1和Cdc42的小分子抑制剂处理... 24

2.2.3 扫描电镜观察... 26

2.2.4 微丝染色和细胞核染色... 26

2.2.5 Western blot分析... 26

2.2.6 pNM II, Cdc42, α-tubulin的免疫染色... 27

2.2.7 荧光信号的半定量分析... 28

2.2.8 探针制备... 28

2.2.9 整装原位杂交实验... 29

2.2.10 基因表达范围的半定量分析... 29

2.2.11 可重复性... 30

2.3 研究结果... 30

2.3.1 磷酸化非肌肉型肌球蛋白II与微丝在贝壳发育区共定位... 30

2.3.2 抑制非肌肉型肌球蛋白II的功能对幼虫贝壳的影响... 32

2.3.3 抑制非肌肉型肌球蛋白II的功能后贝壳发育区的变化... 34

2.3.4 Rho家族小G蛋白在早期贝壳发育中发挥不同的作用... 39

2.3.5 Cdc42在正常发育和抑制剂处理后的动态变化... 43

2.3.6 抑制Cdc42的功能后对贝壳发育区内细胞重排的影响... 48

2.4 讨论... 50

2.4.1 Cdc42和细胞骨架系统：笠贝早期贝壳发育的一条新调控通路... 51

2.4.2 细胞骨架系统对幼虫贝壳形成至关重要... 53

2.4.3 细胞骨架参与贝壳发育区细胞重排，但不影响细胞分化... 54

2.4.4 本部分研究结果对贝壳演化机制的启示... 55

2.4.5 不同区域的贝壳发育细胞类群... 56

2.5 小结... 57

第3章 笠贝贝壳发育区细胞学结构的时空变化... 58

3.1 研究背景... 58

3.2 材料与方法... 58

3.2.1 实验材料的收集与固定... 58

3.2.2 微丝染色... 58

3.2.3 几丁质染色... 58

3.2.4 碳酸钙染色... 59

3.2.5 扫描电镜观察... 59

3.2.6 透射电镜观察... 59

3.2.7 探针制备... 60

3.2.8 整装原位杂交实验... 60

3.2.9 冷冻切片及切片后染色... 60

3.3 研究结果... 61

3.3.1 幼虫贝壳发育过程中的外部形态变化... 61

3.3.2 幼虫贝壳发育过程的细胞学变化及其与贝壳的相关性分析... 63

3.3.3 关键发育阶段的贝壳发育区的细胞学结构... 66

3.3.4 关键发育阶段的贝壳发育区的基因表达情况... 75

3.4 讨论... 82

3.4.1 贝壳发育区中央区域的细胞表面突起的性质与功能... 82

3.4.2 贝壳发育区内不同细胞类群与贝壳形成的相关性推测... 85

3.4.3 贝壳发育区内存在不只3个细胞类群... 86

3.5 小结... 87

第4章 基于单细胞转录组的贝壳发育区解析... 88

4.1 研究背景... 88

4.2 材料与方法... 88

4.2.1 实验材料的收集与固定... 88

4.2.2 胚胎的单细胞悬液的制备... 88

4.2.3 单细胞文库构建及测序... 88

4.2.4 测序数据的分析... 89

4.2.5 细胞群体标记基因分析... 89

4.2.6 细胞群体特征基因的KEGG富集分析... 89

4.2.7 不同标记的探针制备... 89

4.2.8 双色原位杂交实验... 90

4.2.9 小麦胚芽凝集素染色... 91

4.3 研究结果... 91

4.3.1 笠贝胚胎单细胞转录组数据的基本情况... 91

4.3.2 基于单细胞转录组数据的细胞群体的鉴定... 92

4.3.3 贝壳发育相关细胞类群的特征基因及KEGG富集分析... 94

4.3.4 贝壳发育相关细胞类群的特异性验证... 96

4.3.5 早期贝壳发育区内各细胞类群的空间关系解析... 97

4.3.6 细胞类群在贝壳早期发育过程中动态变化的追踪... 101

4.4 讨论... 110

4.4.1 细胞类群在其他软体动物的贝壳发育区中的研究情况... 110

4.4.2 贝壳相关基因的快速演化... 111

4.4.3 五个细胞类群在贝壳发育过程的功能推测... 112

4.5 小结... 113

第5章 总结与展望... 115

5.1 结论... 115

5.2 创新性... 116

5.3 展望... 116

参考文献... 117

附录一  Cdc42抑制剂处理样品的基因表达范围半定量分析数据... 127

附录二  本研究使用的抑制剂的特异性分析... 128

附录三  早期贝壳发育区内细胞群体的空间关系... 131

附录四  早期贝壳发育区内细胞群体的精确关系... 132

附录五  早期外套膜内细胞群体的空间关系... 133

附录六  基本成熟的外套膜内细胞群体的空间关系... 135

附录七  单细胞转录组数据中21个细胞群体特征基因表达丰度展示... 136

致谢... 137

作者简历及攻读学位期间发表的学术论文与其他相关学术成果...139