Genome-wide analysis of LXXLL-mediated DAX1/SHP-nuclear receptor interaction network and rational design of stapled LXXLL-based peptides to target the specific network profile.
Genome-wide analysis of LXXLL-mediated DAX1/SHP-nuclear receptor interaction network and rational design of stapled LXXLL-based peptides to target the specific network profile.
复制标题
DOI：
10.1016/j.ijbiomac.2019.02.014
复制
发表时间：
2019-05
影响因子：
8.2
通讯作者：
Haihong Qian;P. He;F. Lv;Wei Wu
中科院分区：
文献类型：
--
作者：
Haihong Qian;P. He;F. Lv;Wei Wu
非典型孤儿受体DAX1和SHP是人类核受体(HNR)家族的NR0B亚类，它们在新陈代谢、生殖、营养和类固醇合成中发挥关键作用，并参与多种疾病的发病，如癌症和肾上腺发育不良。这两个受体缺乏经典的DNA结合域，并作为其他hNRs的辅助抑制物。DAX1和SHP分别含有三个和两个保守的LxxLL基序，可被激动剂构象中HNR蛋白的激活功能-2(AF-2)结构域识别和结合。在这里，我们试图探索五个DAX1/SHP LxxLL基序与人类基因组中发现的所有48个HNR AF-2结构域之间的系统相互作用谱，分析这些基序对完整结构域阵列的结合亲和力和特异性，并设计能够针对每个基序的特定相互作用谱的基于LxxLL的碳氢化合物结合多肽。基于模型化的结构域-基序复杂结构和计算的结合势，建立了从基序到结构域的加权源-目标网络，根据该网络描述并聚类了每个基序与整个HNR阵列的特定相互作用轮廓，以衡量这些基序之间的结合相似性和关系。动力学模拟表明，基于LxxLL的多肽在自由非结合状态下具有高度的柔性，不利于被AF-2结构域识别和结合。碳氢化合物钉扎技术被用来帮助这些非结构多肽约束到活性螺旋构象，从而极大地提高了它们与HNR阵列的结合亲和力。碳氢桥的设计是为了指出结构域的活性口袋，这不会破坏结构域和多肽之间的直接相互作用。能分解结果表明，装订对结构域-多肽结合的相互作用热和脱溶效应的影响很小，但可以显著降低对结合的熵惩罚。对于一个多肽配体，熵降低可以大致视为一个常数，它只提高了对整个结构域阵列的(绝对)多肽结合亲和力，而不改变阵列中不同结构域的(相对)多肽结合专一性。总体而言，在DAX1和SHP蛋白中，被装订的多肽可以被认为是选择性地靶向它们的母体LxxLL基序介导的特定相互作用网络的强有力的竞争对手。
The atypical orphan receptors DAX1 and SHP constitute the NR0B subgroup of human nuclear receptor (hNR) family; they play key roles in metabolism, reproduction, nutrition and steroidogenesis, and are involved in the pathogenesis of a variety of diseases such as cancer and adrenal hypoplasia. The two receptors lack the classical DNA-binding domain and act as the corepressors of other hNRs. The DAX1 and SHP contains three and two conserved LXXLL motifs, respectively, which can be recognized and bound by the activation function-2 (AF-2) domain of hNR proteins in agonist conformation. Here, we attempt to explore the systematic interaction profile between the five DAX1/SHP LXXLL motifs and all the 48 hNR AF-2 domains found in the human genome, to analyze the binding affinity and specificity of these motifs towards the complete domain array, and to design LXXLL-based, hydrocarbon-stapled peptides that can target the specific interaction profile for each motif. A weighted source–target network from motifs to domains is created based on the modeled domain–motif complex structures and calculated binding potencies, from which the specific interaction profile of each motif against the whole hNR array is depicted and clustered to measure the binding similarity and relationship among these motifs. Dynamics simulations reveal that the LXXLL-based peptides are highly flexible in free unbound state, thus unfavorable to be recognized and bound by AF-2 domains. Hydrocarbon-stapling technique is employed to help the constraint of these unstructured peptides to active helical conformation, thus largely improving their binding affinity to the hNR array. The hydrocarbon bridge is designed to point out of the domain's active pocket, which would not disrupt the direct interaction between the domain and peptide. Energetic decomposition imparts that the stapling has only a very modest influence on the interaction enthalpy and desolvation effect of domain–peptide binding, but can substantially reduce entropy penalty upon the binding. For a peptide ligand, the entropic reduction can be roughly regarded as a constant, which only improves (absolute) peptide binding affinity towards the whole domain array, but does not alter (relative) peptide binding specificity over different domains in the array. Overall, the stapled peptides can be considered as potent competitors to selectively target the specific interaction networks mediated by their parent LXXLL motifs in DAX1 and SHP proteins.