Two-color STED microscopy reveals different degrees of colocalization between hexokinase-I and the three human VDAC isoforms

Abstract

The voltage-dependent anion channel (VDAC, also known as mitochondrial porin) is the major transport channel mediating the transport of metabolites, including ATP, across the mitochondrial outer membrane. Biochemical data demonstrate the binding of the cytosolic protein hexokinase-I to VDAC, facilitating the direct access of hexokinase-I to the transported ATP. In human cells, three hVDAC isoforms have been identified. However, little is known on the distribution of these isoforms within the outer membrane of mitochondria and to what extent they colocalize with hexokinase-I. In this study we show that whereas hVDAC1 and hVDAC2 are localized predominantly within the same distinct domains in the outer membrane, hVDAC3 is mostly uniformly distributed over the surface of the mitochondrion. We used two-color stimulated emission depletion (STED) microscopy enabling a lateral resolution of ~40 nm to determine the detailed sub-mitochondrial distribution of the three hVDAC isoforms and hexokinase-I. Individual hVDAC and hexokinase-I clusters could thus be resolved which were concealed in the confocal images. Quantitative colocalization analysis of two-color STED images demonstrates that within the attained resolution, hexokinase-I and hVDAC3 exhibit a higher degree of colocalization than hexokinase-I with either hVDAC1 or hVDAC2. Furthermore, a substantial fraction of the mitochondria-bound hexokinase-I pool does not colocalize with any of the three hVDAC isoforms, suggesting a more complex interplay of these proteins than previously anticipated. This study demonstrates that two-color STED microscopy in conjunction with quantitative colocalization analysis is a powerful tool to study the complex distribution of membrane proteins in organelles such as mitochondria.PACS: 87.16.Tb, 87.85.Rs.

Figure 1

Image analysis. The two-color STED images were collected with an interleaved pulse scheme. The first pair of excitation/STED pulses (at wavelengths of 650 and 755 nm) was followed by the second pair (at wavelengths of 570 and 710 nm) with a 40 ns delay. Using two detectors and a time-gated detection scheme, it was thus possible to detect the fluorescence emission following each pulse in two different wavelength bands, resulting in a set of four recordings. The four raw images were linearly unmixed to correct for channel crosstalk and were normalized. Segmentation was then performed with an iterative seeded region growing algorithm which extracted the regions of the image containing mitochondria. Only those regions were used in the subsequent calculation of the correlation coefficients. For better visibility the images are shown with a logarithmic color map.

Figure 2

Amino acid sequence alignment of the three hVDAC isoforms. Identical amino acids are colored in yellow, conserved substitutions in green and semi-conserved substitutions in turquoise. Substitutions with no similarity are red colored. hVDAC1 and hVDAC2 exhibit 72% identity and 91% similarity, hVDAC2 and hVDAC3 share 71% identical and 91% similar amino acids and hVDAC1 and hVDAC3 have 67% identical and 94% similar amino acids.

Figure 3

Distinct sub-mitochondrial distributions of the three hVDAC isoforms. Confocal microscopy of immunolabeled cultured human U2OS cells. Green: Different hVDAC isoforms, as indicated. For visualization, the respective Flag-tag fusion proteins were decorated with an antiserum against the Flag-peptide. Red: Mitochondria, highlighted with antibodies against the mitochondrial protein Tom20. Blue: Nucleus, stained with DAPI. The close-ups of the overlays show that hVDAC1 and hVDAC2 are localized in distinct domains, whereas hVDAC3 is uniformly distributed along the mitochondrial tubules.

Figure 4

hVDAC1 and hVDAC2 colocalize in distinct domains in the mitochondrial outer membrane. Confocal imaging of cells expressing two tagged hVDAC isoforms. U2OS cells were decorated with antibodies against the Flag and the V5 tag. The fluorescence signals of hVDAC1-V5 and hVDAC2-Flag overlap to a large extent in confined domains, whereas hVDAC3-Flag has a distinct distribution. For a quantitative analysis of the colocalization see Additional file 3.

Figure 5

Two-color STED microscopy of the three hVDAC isoforms and hexokinase-I reveals individual protein clusters. U2OS cells expressing one of the three Flag-tagged hVDAC isoforms were labeled with antibodies directed against the Flag-tag and hexokinase-I. With confocal microscopy, neither individual hVDAC3 nor individual hexokinase-I clusters could be resolved (bottom row). In contrast, the corresponding STED images clearly resolved hVDAC3 and hexokinase-I clusters (third row). Two-color STED images of hVDAC1 or hVDAC2, imaged together with hexokinase-I, further shows that the hVDAC1/hVDAC2 domains are frequently composed out of smaller clusters. In the overlay the green color corresponds to the hVDAC signal and the red color to the hexokinase-I signal.

Figure 6

The hVDAC isoforms exhibit different degrees of colocalization with hexokinase-I. Quantitative colocalization analysis of any of the three hVDAC isoforms with hexokinase-I based on two-color STED images. Shown are the Pearson's correlation coefficients r_p (A), the overlap coefficient r (B) as well as the colocalization coefficients M₁ and M₂ (C, D). As positive controls for full colocalization, we analyzed cells doubly labeled for hexokinase-I as well as cells co-expressing hVDAC1-Flag and hVDAC1-V5. Error bars: Standard deviation. The numbers on the columns represent the number of analyzed cells. The detailed values for all columns are given in Additional file 4. M₁ and M₂ may overestimate the level of colocalization because of the required thresholding. To demonstrate this effect, an intensity profile through a two-color STED image, where the cell was labeled for hVDAC2 and hexokinase-I, is shown (E). The fluorescence intensity was determined at the indicated line in the images. For details see main text.