Three-Dimensional HIV RNA Genome Targeted Drug Leads

描述(由申请人提供)：该项目的总体目标是利用一种新的药物设计范例开发有前途的抗艾滋病药物。使用基本的RNA元素作为开发小型治疗药物的靶点是一种很少使用的策略，尽管许多抗生素是通过结合rRNA上的特定位点来发挥作用的先例。我们的实验室和其他实验室已经确定了一些关键的HIV-1RNA基因组元素的3D结构，这些结构可以作为计算筛选可结合化合物数据库的目标。对于RNA靶向，我们继续开发两个对接程序，DOCK和Mordor，后者相当新颖，因为它允许假定的配体和靶的灵活性，以实现真正的诱导配合；这种灵活的配合对于蛋白质靶的虚拟筛选是有用的，但对RNA靶非常重要，因为它们通常具有很强的可塑性。候选化合物是水溶性、非肽、非核苷酸有机化合物，分子量一般为500道尔顿，电荷不超过+2(一般为0或+1)。将使用核磁共振和表面等离子激元共振对计算的“点击”进行实验测试，以结合到RNA目标，但不是目标外的RNA控制。那些结合的将作为相似性搜索的模型，以确定其他潜在的结合剂。核磁共振将被用来绘制RNA上的结合位点，并使用两个对照RNA来测量最佳命中的Kd值，以确定选择性。核磁共振将被用来模拟最有希望的粘结剂的结构；这种结构将定义随后的有机合成(这里没有明确建议)的支架，以提高亲和力和选择性。在这个简短的研究项目中，我们建议继续进行所需的方法开发。对一个新目标的应用将说明该方法的实用性，并有望产生有希望的命中结果。主要靶点是HIV-1RNA基因组5‘-UTR区包装序列中茎环1(SL1)的富含G的内环(GRIL)，我们发现它基本上是100%保守的(这表明有可能避免耐药性)。更重要的是，人类rRNA和tRNAs没有任何类似GRIL的结构基序。GRIL对于将基因组包装成新的感染性病毒粒子是至关重要的，但考虑到5‘UTR的其他部分缺乏序列保守性，它可能还发挥着另一个未知的关键作用。我们已经确定了SL1的结构，它将用于靶向。随着我们已经确定了另一个与包装部位结合的配体，专注于我们对该配体的了解的研究将试图确定HIV-1复制的抑制剂。 与公共健康相关：通过计算机筛选与HIV-1活性重要的3D RNA或蛋白质结构结合的大型类药物化合物数据库，然后对预测最多的化合物进行实验验证，应该是找到可能成为药物先导的化合物的一种快速、经济高效的方法。我们已经开发了用于筛选的软件，这是对现有软件的重大改进，特别是针对RNA上的关键位置，与典型的蛋白质靶点相比，RNA在很大程度上被忽视了。我们建议改进该软件，并使用它来寻找与HIV-1 RNA基因组的一个重要保守部分结合的有希望的化合物，其中的序列保守表明耐药性的进化将被最小化，并最终抑制HIV-1的复制。

DESCRIPTION (provided by applicant): The overall goal of this project is to develop promising anti-AIDS drug leads using a new drug design paradigm. Using essential RNA elements as targets for developing small therapeutic drugs is a little-used strategy, despite the precedent that many antibiotics work by binding specific sites on rRNA. Our lab and others have determined 3D structures of some crucial HIV-1 RNA genome elements that can serve as targets for computational screening of databases of available compounds for binding. For RNA targeting, we continue to develop two docking programs, DOCK and MORDOR, the latter quite novel in that it permits flexibility in both putative ligand and target for a true induced fit; such flexible fitting is useful for virtual screening of protein targets but is very important for RNA targets, as they are typically quite malleable. Candidate compounds are water-soluble, nonpeptide, nonnucleotide organic compounds generally with molecular weight <500 daltons and a charge of no more than +2 (generally 0 or +1). Computational "hits" will be tested experimentally using NMR and surface plasmon resonance for binding to the RNA target but not off-target RNA controls. Those that bind will serve as models for similarity searches to identify other potential binders. NMR will be used to map the binding site on the RNA and to measure Kd values for the best hits using two control RNAs to determine selectivity. NMR will be used to model the structure of the most promising binders; such structures will define scaffolds for subsequent organic synthesis (not explicitly proposed here) to enhance affinity and selectivity. In this short research project, we propose a continuation of needed methodology development. Application to a novel target will illustrate the utility of the methodology with the prospect that promising hits will result. The primary target is the G-rich internal loop (GRIL) of stem-loop 1 (SL1) in the packaging sequence in the 5'-UTR of the HIV-1 RNA genome, which we discovered to be essentially 100% conserved (suggesting the possibility to avert drug resistance). More importantly, human rRNA and tRNAs do not have any structural motifs similar to GRIL. GRIL is critical for packaging of the genome into new infectious virions but, considering the lack of sequence conservation in other parts of the 5'UTR, it may play another unknown crucial role as well. We have determined the structure of SL1, which will be used for targeting. As we have identified another ligand that binds to the packaging site, studies focused on our knowledge of that ligand will attempt to identify inhibitors of HIV-1 replication. PUBLIC HEALTH RELEVANCE: Screening a large database of drug-like compounds by computer for binding to a 3D RNA or protein structure important for HIV-1 activity followed by experimental verification of the top predicted compounds should be a fast, cost effective way to find compounds that might become drug leads. We have developed software for screening that is a significant improvement over existing software, especially for targeting crucial sites on RNA, which has largely been ignored compared with the typical protein targets. We propose to improve the software and to use it to find promising compounds that bind to an important conserved portion of the RNA genome of HIV-1, where the sequence conservation suggests evolution of drug resistance will be minimized, and ultimately to inhibit HIV-1 replication.