Pin-hole array correlation imaging: highly parallel fluorescence correlation spectroscopy

Abstract

In this work, we describe pin-hole array correlation imaging, a multipoint version of fluorescence correlation spectroscopy, based upon a stationary Nipkow disk and a high-speed electron multiplying charged coupled detector. We characterize the system and test its performance on a variety of samples, including 40 nm colloids, a fluorescent protein complex, a membrane dye, and a fluorescence fusion protein. Our results demonstrate that pin-hole array correlation imaging is capable of simultaneously performing tens or hundreds of fluorescence correlation spectroscopy-style measurements in cells, with sufficient sensitivity and temporal resolution to study the behaviors of membrane-bound and soluble molecules labeled with conventional chemical dyes or fluorescent proteins.

Figure 1

Schematic of the experimental setup. The filter cube and microlense and pinhole arrays (which have been modified to allow laser illumination while not rotating) are inside the spinning disk confocal head. Demagnifying optics were placed between the confocal head and the camera to allow more pinholes to be imaged.

Figure 2

Solution with Alexa 488 (100 nM) imaged with the Nipkow disk spinning (left) and stationary (right). The same laser power, camera settings, and gray scale were used in both pictures, illustrating the higher measured intensities when the disk is stopped (see text). The exposure time was 30 ms.

Figure 3

Measurements on a solution of RPE in 32% sucrose. (A, left) 40 correlation curves measured from different pinholes in 3.8 s (blue), with associated best fits (red). (A, right) correlation curves from the same 40 pinhole locations measured in 19 s (blue), with associated best fits (red). (B) The variance between pinholes (divided by the mean squared) of measured particle number (N, green) and diffusion time (τ_D, blue) as a function of acquisition time. Both curves where fit to the function $a/T+E$ (red), where E is the intrinsic variance between pinholes, T is the acquisition time, and a is a constant (see text for details).

Figure 4

Color-scale maps showing the values of diffusion time (left, with scale in ms), particle number (center), and particle brightness (right, with scale in arbitrary units), at different locations in a homogeneous solution of 40 nm colloids.

Figure 5

PACI measurements obtained by averaging data from different pinholes for solutions of RPE with 0%, 16%, 32%, 40%, 48%, and 56% sucrose. (A) Normalized correlation curves (blue) with best fits to Eq. 1 (red). (B) As expected, the measured diffusion time, τ_D, increases linearly with the solutions viscosity, η, (where η₀ is the viscosity of water). Viscosities for sucrose solutions were obtained from (24). Errors in viscosity were assumed to arise from a 1% error in determining the sucrose concentration (with a linear extrapolation of the viscosities from (24)).

Figure 6

(A, upper) A correlation curve from a solution of 40 nm colloids, obtained in 5 min by averaging data from ∼380 pinholes (blue) with a best fit to Eq. 1 (red). (A, lower) The residuals to the fit are of order 10⁻³. (B) A dilution series for 40 nm colloids showing how the measured number of particles changes (FCS in blue, PACI in red). Lines are best fits to the expected linear trend.

Figure 7

Dynamics of labeled diI in the plasma membrane of U2OS cells. (A) 16 representative correlation curves (blue) with best fits to Eq. 2 (red). (B) Histograms of the measured values of diffusion time, (τ_D, left), and particle number (right) obtained from 44 locations in ∼90 s.

Figure 8

Behavior of FOXO-GFP in tissue culture cells studied with PACI. (A) 15 representative correlation curves (blue) with best fits to Eq. 1 (red). (B) Histograms of the measured values of diffusion time, (τ_D, left), and particle number (right) obtained from 89 locations in ∼161 s. (C) Spatial maps showing the measured mean fluorescence intensity (left) and particle number (right) at different pinholes. The focus is a few microns above the coverslip. Under the experimental conditions FOXO-GFP is predominantly cytoplasmic, and consistent correlation curves could not be obtained from the nucleus. Scale bar = 10 μm.

Figure 9

FOXO-GFP at the edge of a cell. (A) An image of the cell edge obtained with the Nipkow disk spinning. (B) The disk was stopped and a PACI measurement was performed. The displayed map shows the locations of the pinholes with their color indicating the measured number of particles at that location (scale to right). (C) The measured number of particles at each location as a function of distance along the cell. (D) The measured diffusion time as a function of distance along the cell. (E) Two represented correlation curves from the indicated locations (blue) with the associated best fits to Eq. 1 (red).