Proximity ligation assays with peptide conjugate 'burrs' for the sensitive detection of spores

Abstract

The proximity ligation assay (PLA) has previously been used for the sensitive and specific detection of single proteins. In order to adapt PLA methods for the detection of cell surfaces, we have generated multivalent peptide-oligonucleotide-phycoerythrin conjugates ('burrs') that can bind adjacent to one another on a cell surface and be ligated together to form unique amplicons. Real-time PCR detection of burr ligation events specifically identified as few as 100 Bacillus anthracis, 10 Bacillus subtilis and 1 Bacillus cereus spore. Burrs should prove to be generally useful for detecting and mapping interactions and distances between cell surface proteins.

Figure 1

Specificity of monovalent and polyvalent probes. Fluorescent probes were constructed using the NH-peptide (BS-specific). BS and BC spores were incubated with either (A) monovalent NH-peptide–fluorescein conjugates or (B) polyvalent NH-peptide–PE conjugates. Specific binding was only observed when the polyvalent NH-peptide–PE probes were used. Spores were visualized using differential interference microscopy (DIC) and fluorescence microscopy with either fluorescein (FITC) or Texas Red filter sets (TR).

Figure 2

Construction and ligation of burrs. (A) Burrs: oligonucleotides and peptides are separately conjugated to PE. There are two distinct oligonucleotide conjugates, one linked through its 5′ end and one linked through its 3′ end. (B) Burr ligation and amplification. When simultaneously bound to a spore target, burrs can be aligned by a splint oligonucleotide and ligated to generate a unique amplicon. The sequences of the 5′ and 3′ oligonucleotide probes as well as the splint oligonucleotide are given in Materials and Methods.

Figure 3

Optimization of PLA probe concentration for the detection of B.cereus spores. The real-time PCR data represent a single dataset in which the probe concentration was varied from 0.1 to 100 pM. PLA reactions conducted in the presence of 100 BC spores are indicated by a solid line and those conducted in the absence of spores by dashed lines. A positive, spore-dependent signal was only observed when reactions were conducted using 10 pM probe (bold red). Reactions were assembled as described in Materials and Methods.

Figure 4

Optimization of PLA probe concentration for 100 B.subtilis and B.cereus spores. A splint concentration of 10 pM was used. The cycle difference represents the difference between the C[T] value of the background amplification reaction (no spores) and amplification in the presence of spores. Reactions containing BS spores and BC spores were carried out with burrs that presented either the NH- (BS-specific) or S-peptides (BC-specific). Reactions containing BA (Sterne) spores were carried out with burrs presenting either the NH-, S- or the ATY-peptides (BA-specific).

Figure 5

Splint optimization for 100 spores. A probe concentration of 1 pM was used. Reactions contained burrs as described in Figure 4. Cycle difference is as in Figure 4.

Figure 6

Specificity of spore detection assays. Reactions were carried out with 10 pM probe and 10 pM splint, and contained burrs bearing one of the three spore-specific peptides. Cycle difference is as in Figure 4.

Figure 7

Optimization of the limits of detection for BC spores. Reactions were carried out with 10 pM probe and 1 pM splint, and contained burrs bearing one of the three spore-specific peptides. Cycle difference is as in Figure 4.