High throughput mutagenesis for identification of residues regulating human prostacyclin (hIP) receptor expression and function

Abstract

The human prostacyclin receptor (hIP receptor) is a seven-transmembrane G protein-coupled receptor (GPCR) that plays a critical role in vascular smooth muscle relaxation and platelet aggregation. hIP receptor dysfunction has been implicated in numerous cardiovascular abnormalities, including myocardial infarction, hypertension, thrombosis and atherosclerosis. Genomic sequencing has discovered several genetic variations in the PTGIR gene coding for hIP receptor, however, its structure-function relationship has not been sufficiently explored. Here we set out to investigate the applicability of high throughput random mutagenesis to study the structure-function relationship of hIP receptor. While chemical mutagenesis was not suitable to generate a mutagenesis library with sufficient coverage, our data demonstrate error-prone PCR (epPCR) mediated mutagenesis as a valuable method for the unbiased screening of residues regulating hIP receptor function and expression. Here we describe the generation and functional characterization of an epPCR derived mutagenesis library compromising >4000 mutants of the hIP receptor. We introduce next generation sequencing as a useful tool to validate the quality of mutagenesis libraries by providing information about the coverage, mutation rate and mutational bias. We identified 18 mutants of the hIP receptor that were expressed at the cell surface, but demonstrated impaired receptor function. A total of 38 non-synonymous mutations were identified within the coding region of the hIP receptor, mapping to 36 distinct residues, including several mutations previously reported to affect the signaling of the hIP receptor. Thus, our data demonstrates epPCR mediated random mutagenesis as a valuable and practical method to study the structure-function relationship of GPCRs.

Figure 1. Error-prone PCR mediated mutagenesis of hIP receptor.

(A, B) Number and distribution of mutations per clone after error-prone PCR mediated mutagenesis of hIP receptor using 750 ng (A) or 500 ng (B) parental PTGIR plasmid as input. Grey bars represent the expected mutational spectrum at a mutation rate of 0.8 mut/kb (750 ng) or 1.5 mut/kb (500 ng), respectively. Black bars represent the experimentally observed number of mutations after sequencing a small subset of clones. (C, D) Expected number and distribution of amino acid changes per clone at a mutation rate of 0.8 mut/kb (750 ng, C) or 1.5 mut/kb (500 ng, D) (E) Expected and observed number and distribution of mutations after pooling both libraries (500+750 ng) depicted as described in (A,B). (F) Expected number and distribution of amino acid changes per clone in our library with an estimated mutation rate of 1.3 mut/kb. (G) Scatter blot showing the mutation rate for each nucleotide position of the coding region of PTGIR as determined by next generation sequencing. The overall mutation rate was 1.48 mutations/kb. (H) Comparison of the observed codon changes in our library (bars) with the expected codon changes (considering an unbiased mutation rate and one nucleotide substitution per codon, red symbol).

Figure 2. Error-prone PCR mediated mutagenesis of hIP receptor identified 32 mutants with reduced activity.

(A) Scatter blot showing activity data of the hIP receptor library in the primary screen. Cells stably expressing the GloSensor cAMP plasmid were transfected with our hIP receptor constructs and stimulated with 1 µM MRE-269. The luminescent signal was normalized to the signal obtained in hIP receptor wild-type transfected cells. The experiment was performed in duplicates, shown on separate axis. Red box indicates clones with <25% activity as compared to wild-type hIP receptor. (B) Exemplary images of vector or hIP receptor expressing cells, stained with Alexa-647 anti HA-antibody. (C) Bar chart showing the relative expression level (as assessed by staining with Alexa-647 anti-HA-antibody) and the relative activity (as assessed by GloSensor cAMP assay) for 34 selected constructs. Asterisk indicates mutants classified as hits. Data were normalized to hIP receptor wild-type expressing cells and represents the mean ± SEM of three independent experiments.

Figure 3. Functional characterization of hIP receptor identified inactive mutants of hIP.

(A) Bar chart depicting the number and distribution of amino acid changes found in the 18 inactive mutants. Black bars represent the total number of amino acid changes; grey bars represent non-synonymous mutations only. (B) Bar chart showing the number and position of all 38 identified non-synonymous mutations in hIP receptor. Black boxes represent the α-helical, membrane-spanning domains of hIP receptor.