Roberta Poceviciute, Kenneth Mitchell, Angeliki Maria Nikolakopoulou, Suehyun K. Cho, Xiaobo Ma, Phillip Chen, Samantha Figueroa, Ethan J. Sarmiento, Aryan Singh, Oren Hartstein, William G. Loudon, Florent Cros, Alexander S. Kiselyov
{"title":"Quantitative 3D reconstruction of viral vector distribution in rodent and ovine brain following local delivery","authors":"Roberta Poceviciute, Kenneth Mitchell, Angeliki Maria Nikolakopoulou, Suehyun K. Cho, Xiaobo Ma, Phillip Chen, Samantha Figueroa, Ethan J. Sarmiento, Aryan Singh, Oren Hartstein, William G. Loudon, Florent Cros, Alexander S. Kiselyov","doi":"10.1016/j.ynirp.2024.100218","DOIUrl":null,"url":null,"abstract":"<div><div>Viral vectors are an active area of research and development to treat diseases of the central nervous system (CNS). However, systemic delivery of large-molecular weight biologics is complicated by limited crossing of the blood-brain barrier, immunological clearance from the circulation, off-target effects, and systemic or organ toxicity. Local drug delivery can mitigate these obstacles, however, the drug must still be distributed over sufficiently large tissue volume to achieve the desired efficacy. In the field of drug delivery, quantitative, high resolution spatial analysis of drug distribution in the brain and other organs poses a challenge. To address this issue, we introduce a computational pipeline to reconstruct and quantify 3D distribution of locally delivered viral vectors from 2D microscopy images of subsampled brain sections. This pipeline, which combined existing and newly developed machine-learning and other computational tools, effectively removed false positive artifacts abundant in large-scale images of uncleared tissue sections, and subsampling adequately predicted the dispersion of model viral vectors from the point of local drug delivery. Furthermore, the pipeline successfully captured differences in the distribution of adeno virus (AdV) and adeno-associated virus (AAV) vectors exhibiting varying sizes and transport properties. Notably, the proposed method is directly applicable to the distribution studies of therapeutics in large animal models. Thus, our developed pipeline is an accessible tool that can aid the research and development of local drug delivery strategies for the treatment of CNS diseases with viral vectors and potentially other therapeutics.</div></div>","PeriodicalId":74277,"journal":{"name":"Neuroimage. Reports","volume":"4 4","pages":"Article 100218"},"PeriodicalIF":0.0000,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2666956024000242/pdfft?md5=649e0ec0565896c07b30e30068195caa&pid=1-s2.0-S2666956024000242-main.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Neuroimage. Reports","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2666956024000242","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Neuroscience","Score":null,"Total":0}
引用次数: 0
Abstract
Viral vectors are an active area of research and development to treat diseases of the central nervous system (CNS). However, systemic delivery of large-molecular weight biologics is complicated by limited crossing of the blood-brain barrier, immunological clearance from the circulation, off-target effects, and systemic or organ toxicity. Local drug delivery can mitigate these obstacles, however, the drug must still be distributed over sufficiently large tissue volume to achieve the desired efficacy. In the field of drug delivery, quantitative, high resolution spatial analysis of drug distribution in the brain and other organs poses a challenge. To address this issue, we introduce a computational pipeline to reconstruct and quantify 3D distribution of locally delivered viral vectors from 2D microscopy images of subsampled brain sections. This pipeline, which combined existing and newly developed machine-learning and other computational tools, effectively removed false positive artifacts abundant in large-scale images of uncleared tissue sections, and subsampling adequately predicted the dispersion of model viral vectors from the point of local drug delivery. Furthermore, the pipeline successfully captured differences in the distribution of adeno virus (AdV) and adeno-associated virus (AAV) vectors exhibiting varying sizes and transport properties. Notably, the proposed method is directly applicable to the distribution studies of therapeutics in large animal models. Thus, our developed pipeline is an accessible tool that can aid the research and development of local drug delivery strategies for the treatment of CNS diseases with viral vectors and potentially other therapeutics.