Student Theses and Dissertations

Date of Award


Document Type


RU Laboratory

Gilbert Laboratory


RNA interference, brain derived neurotrophic factor, cortical plasticity, microRNA's, in situ hybridization, small hairpin RNA's


During the course of adult cortical plasticity, a number of signal transduction mechanisms are brought into play. To study regulatory genes implicated in this process, we inhibited gene expression by harnessing the machinery of RNA interference (RNAi) pathway via small hairpin RNAs (shRNA) delivered by viral vectors. Using this technology we sought to influence plasticity of the mouse vibrissal barrel cortex. In this model system, chronic whisker plucking reliably leads to the expansion of the cortical representation of the adjacent non-deprived whiskers. The mechanism underlying this process involves changes in synaptic efficacy and sprouting of axon collaterals. Initial studies describing the molecular events leading to synaptogenesis and remapping have implicated a number of signal transduction pathways, including neurotrophins such as brain derived neurotrophic factor (BDNF). Though previous experiments have shown up-regulation of neurotrophins and their receptors in reorganized cortex, it is necessary to remove these factors to definitively prove their involvement in adult cortical plasticity. We therefore repressed neurotrophic gene expression in the mouse somatosensory cortex to determine if these trophic factors are essential for cortical remodeling after sensory deprivation. We blocked gene expression in vivo using a non-replicative adeno-associated virus bearing genes encoding shRNA constructs. The shRNA nucleotide sequences were designed to target and destroy selected neurotrophin messenger RNAs by triggering the RNAi pathway, resulting in a reduction of BDNF protein levels by up to 80% in vivo. This knockdown of BDNF expression effectively repressed functional cortical reorganization induced by chronic whisker plucking. Unexpectedly, when transducing the somatosensory cortex with a viral vector carrying a control shRNA that does not target a gene in mouse, we saw no cortical reorganization after sensory deprivation. These results suggest that a non-specific blockade of cortical plasticity occurs when using shRNAs in the brain. Our results suggest that manipulating gene expression via virally delivered shRNAs may have limited value for blocking expression of specific genes involved in cortical plasticity. However, the results suggest a potential role for RNA interference itself in regulating cortical plasticity. In the second part of this thesis, we study methods to visualize microRNAs in tissue sections. MicroRNAs (miRNAs) are small regulatory RNAs with many biological functions and disease associations. Association studies are rapidly linking miRNAs with cancers and neurological disorders. miRNAs have specific expression and function in specialized cell types, emphasizing the need to define cell–type–specific miRNA expression patterns. For pathologists, the most common method for visualizing gene expression in specific cell types is in situ hybridization (ISH). In our laboratory, conventional ISH worked for highly abundant miRNAs, however examining less abundantly expressed miRNAs often yielded inconsistent or negative results. A review of the literature indicates that zebrafish whole–mount ISH were robust, whereas ISH in D. melanogaster embryos were mostly unsuccessful. In tissue sections, several studies set the foundation for establishing miRNA ISH,and were used to support association to disease, or defined expression patterns in multiple cell types present in the brain and eye. The technical difficulties in miRNA ISH also led to development of transgenic or cell sorting methods that monitor cell–type–specific miRNA expression. To better understand the technical challenges associated with miRNA ISH, we investigated the importance of miRNA fixation and probe hybridization. For fixation of proteins and nucleic acids in tissues, a solution containing 3.7% formaldehyde (10% formalin) is commonly used. Formaldehyde crosslinks are reverted by incubation at elevated temperature and this process is facilitated by proteinase K treatment. Reversal of the formaldehyde–based nucleic acid base modifications is also necessary for probe hybridization, but it creates the problem of miRNA release and diffusion out of the tissue sections. Therefore, we examined the extent of miRNA escape from tissue during ISH. We conducted a mock ISH for conventionally fixed brain sections, isolated RNA from the tissue sections and the ISH buffer, and probed both fractions for the highly expressed neuronal miRNAs. We showed that in situ hybridization (ISH) using conventional formaldehyde fixation results in significant miRNA loss from mouse tissue sections, which can be prevented by fixation with 1–ethyl–3–(3– dimethylaminopropyl) carbodiimide (EDC) that irreversibly immobilizes the miRNA at its 5' phosphate. Eliminating the possibility of miRNA diffusion by introducing an irreversible crosslink between the 5’ phosphate of the miRNA and protein side chains significantly improves miRNA retention in tissues. The advantage of EDC–based phosphoramidate–linked miRNA is that the sample can be exposed to higher temperatures for a longer time, resulting in a more complete reversion of the formaldehyde induced nucleobase modification of the miRNAs and less interference with probe hybridization. At present, we showed that different tissues and detection systems are compatible with our approach. This method paves the way for reliable disease association studies and the potential use of miRNA in situ hybridization expression analysis as a diagnostic tool for biopsy material.


A thesis presented to the faculty of The Rockefeller University in partial fulfillment of the requirements for the degree of Doctor of Philosophy.

Permanent URL

Included in

Life Sciences Commons