Restore Western blot stripping reagent (Pierce) was used to remove bound antibodies from immunoblots to allow for reprobing of membranes. Densitometry and calculations Densitometry of Coomassie blue-stained protein bands and Western blot signals acquired with a Fuji LAS-4000 fluorescence imager with a linearity of 4 orders of magnitude was done using the Image J image analysis software http://rsb.info.nih.gov/ij/. The percentage of surface-localized protein was calculated using the following formula: % surface = 100 – [(mRFP1+pK x FlaB-pK) selleck products ÷ (mRFP1-pK x FlaB+pK)] × 100, where
mRFP1 and FlaB indicate the raw Western immunoblot densitometry data in absence (-pK) or presence (+pK) of proteolysis. Negative % surface values obtained for four mutants (ED, SK, TR and GR) were set to zero. The OM/PC distribution ratio using the following formula: ratioOM/PC = (mRFP1OM ÷ mRFP1PC) ÷ [(OspAOM ÷ OspAPC) - (OppAIVOM ÷ OppAIVPC)], where mRFP1, OspA and OppAIV represent the raw Western immunoblot
densitometry data in either the OM or PC fractions. Genomic B. burgdorferi strain B31 (GenBank Accession # NC_001318) www.selleckchem.com/products/cb-839.html codon usage data were acquired from the Georgia Tech Codon Usage Database http://exon.gatech.edu/GeneMark/metagenome/CodonUsageDatabase/ and compared to detected protein levels. Codon usage-to-protein level correlation coefficients were calculated using Microsoft Excel for Mac 2008. Results & Discussion Design of a fluorescence-based screen for lipoprotein localization in B. burgdorferi In our recent studies, the use of fusions of red fluorescent mRFP1 to various N-terminal fragments and point mutants of B. burgdorferi surface lipoprotein OspA led to an Tolmetin initial assessment of the sequence requirements for proper surface display [4, 21]. To complement this step-wise, targeted mutagenesis approach, we set out to develop a random mutagenesis screen. Our LB-100 in vitro starting point was a previously described OspA-mRFP1 fusion, OspA20:mRFP1, which could be redirected from
the IM to the bacterial surface by mutagenesis of two adjacent negatively charged amino acids (Glu-Asp) at the N-terminus of mRFP1 to two Ala residues. We therefore hypothesized that (i) additional mutagenesis in this OspA20:mRFP1 dipeptide would reveal the specificity of periplasmic, particularly IM retention signals in this model lipoprotein, and that (ii) periplasmically localized fusion protein mutants could be enriched by a combination of in situ surface proteolysis and fluorescence-activated cell sorting (FACS). The approach is detailed in the Materials & Methods section and shown in Figure 1. Two plasmid libraries were generated from two different starting materials, pRJS1009 and pRJS1016 [4]. pRJS1009 carried a fusion of the full-length signal peptide and tether of OspA to mRFP1 (OspA28:mRFP1), which was targeted to the bacterial surface.