282837 Identifying Transcriptional Phenotypes Associated with Single Cell Variability in Hypertension

Thursday, November 1, 2012: 10:00 AM
Somerset East (Westin )
James Park1,2, Anthony Brureau2, Sonali Gulati2, Carmen Nichols3, Rajanikanth Vadigepalli4, Babatunde A. Ogunnaike1 and James S. Schwaber2, (1)Department of Chemical and Biomolecular Engineering, University of Delaware, Newark, DE, (2)Daniel Baugh Institute for Functional Genomics and Computational Biology, Thomas Jefferson University, Philadelphia, PA, (3)Daniel Baugh Institute or Functional Genomics and Computational Biology, Thomas Jefferson University, Philadelphia, PA, (4)Daniel Baugh Institute for Functional Genomics and Computational Biology, Department of Pathology, Anatomy and Cell Biology, Thomas Jefferson University, Philadelphia, PA

Hypertension is characterized by a sustained elevation of arterial blood pressure; its pathology is multifactorial with many contributing genetic and environmental factors and is not completely understood.  Consequently, various aspects of blood pressure regulation have been studied extensively in order to suggest mechanisms subtending the development of hypertension.  Physiological evidence indicates a strong neurogenic component of hypertension pathology and implicates abnormal neural blood pressure control mechanisms, which involve A2 neurons located within the nucleus tractus solitarius (NTS).  Additional evidence from gene expression studies also show that neuronal adaptations are strongly associated with hypertension development.   These neuronal adaptations are believed to cause heterogeneous gene expression within the NTS and A2 neurons (“extrinsic biological noise”) creating distinct phenotypic subtypes.  Characterizing these phenotypic subtypes may inform strategies for preventative treatment of hypertension.  Our objective is to distinguish these various subtypes though single cell analysis, with a specific focus on A2 neurons and cells responsive to an acute hypertensive challenge in the NTS (as indicated by cFos protein expression, a general marker of neural activation).  Our strategy is to utilize a combination of experimental approaches to acquire gene expression measures, i.e. transcriptional signatures, and computational analysis to identify distinct transcriptional phenotypes.  These phenotypes can be used to classify subgroups within these cell types associated with a hypertensive state. 

In our approach, rats were infused with phenylephrine to induce an acute hypertensive challenge, pharmacologically increasing blood pressure ~40mmHg; corresponding control animals were infused with saline solution.  NTS tissue samples were collected from the rats one hour post infusion.  Single cell samples were categorized within the NTS by several factors including 3-dimensional rostro-caudal localization and immunohistochemistry (IHC) phenotype.  IHC was used to identify the two distinct cell populations of interest:  neurons expressing (1) tyrosine hydroxylase (TH), an enzyme involved in synthesis of catecholaminergic neurotransmitters found exclusively in A2 neurons in the NTS and (2) cFos, a sensitive marker of neural activation identifying cells responsive to the acute hypertensive perturbation.  Laser Capture Microdissection (LCM) was used to lift single cells from the NTS tissue samples for analysis of their respective transcriptional signature. 

We use a high-throughput quantitative PCR platform (BioMarkTM) to analyze gene expression of our single cell samples.  This method enables us to assay 96 genes within 96 samples simultaneously, with low technical variability.  Our 96 gene assays were refined from previous global gene expression studies that provide evidence for their co-regulated and coordinated response to blood pressure changes.  This gene cohort is also relevant to the A2 phenotype and to angiotensin II type 1 receptor (AT1R) pathways, which are mediators of cellular responses associated with hypertension. 

Our analysis involved the use of several multivariate analytical techniques.  Through the use of Principal Component Analysis of single cell gene expression, we were able to identify distinct transcriptional signatures distinguishing the A2 neuronal (Th+) and responsive cell (cFos+) phenotypes within the NTS.  Specifically Th, Dbh, Gal, Rgs4, Slc32a, Rgs1, Npy, Tac1, and Sst transcripts make significant contributions to the variability between these subphenotypes.  Hierarchical clustering of the NTS cell phenotypes within specific regions of the NTS reveals anatomical subgroups within the cFos+ phenotype.  Additionally, the transcriptional signatures of these single cells have been used to identify a subset of single cells that express both Th and cFos, forming a Th+/cFos+ subphenotype.  This group may be used to identify specific A2 neurons that respond to an acute hypertensive challenge and may be a specific focus for future time-series and chronic studies of neurogenic hypertension. 

Taken together, we have demonstrated a powerful methodology for studying mRNA expression variability within in vivo single cell samples in a hypertensive animal model.  Our integrated experimental and computational approach has enabled us to characterize expression heterogeneity and identify specific transcriptional signatures that distinguish subphenotypes within the NTS.  These subpopulations may play distinct roles in the neural pathology of hypertension development.


Extended Abstract: File Not Uploaded