Department of Cellular and Molecular Medicine, University of Bristol, UK
Murillo, I., Virji, M. Visualisation and Quantification of Intracellular Interactions of Neisseria meningitidis and Human α-actinin by Confocal Imaging. J. Vis. Exp. (44), e2045, doi:10.3791/2045 (2010).
The Opc protein of Neisseria meningitidis (Nm, meningococcus) is a surface-expressed integral outer membrane protein, which can act as an adhesin and an effective invasin for human epithelial and endothelial cells. We have identified endothelial surface-located integrins as major receptors for Opc, a process which requires Opc to first bind to integrin ligands such as vitronectin and via these to the cell-expressed receptors1. This process leads to bacterial invasion of endothelial cells2. More recently, we observed an interaction of Opc with a 100kDa protein found in whole cell lysates of human cells3. We initially observed this interaction when host cell proteins separated by electrophoresis and blotted on to nitrocellulose were overlaid with Opc-expressing Nm. The interaction was direct and did not involve intermediate molecules. By mass spectrometry, we established the identity of the protein as α-actinin. As no surface expressed α-actinin was found on any of the eight cell lines examined, and as Opc interactions with endothelial cells in the presence of serum lead to bacterial entry into the target cells, we examined the possibility of the two proteins interacting intracellularly. For this, cultured human brain microvascular endothelial cells (HBMECs) were infected with Opc-expressing Nm for extended periods and the locations of internalised bacteria and α-actinin were examined by confocal microscopy. We observed time-dependent increase in colocalisation of Nm with the cytoskeletal protein, which was considerable after an eight hour period of bacterial internalisation. In addition, the use of quantitative imaging software enabled us to obtain a relative measure of the colocalisation of Nm with α-actinin and other cytoskeletal proteins. Here we present a protocol for visualisation and quantification of the colocalisation of the bacterium with intracellular proteins after bacterial entry into human endothelial cells, although the procedure is also applicable to human epithelial cells.
1. Immunofluorescence Protocol
Seeding, Infection & immuno-staining
The following procedures require suitable safety level tissue culture and microbiological laboratory facilities.
A.Preparation of target cells for infection
B. Bacterial culture
A. Preparation Of Bacterial (N. meningitidis) Suspension
B. Cell Culture Infection
Staining of intracellular bacteria and α-actinin can be performed sequentially or simultaneously by the use of appropriate primary and secondary antibodies as follows; all procedures can be carried out in 12-well plates.
2. Confocal Laser Scanning Microscopy (CLSM)
To observe and capture images of intracellular bacteria and cytoskeletal elements, we used immunolabelled samples and captured images using a Leica SP5-AOBS confocal laser scanning microscope attached to a Leica DM I6000 inverted epifluorescence microscope. All images were collected using a 63x NA 1.4 oil immersion objective and process with Leica software.
3. Quantification of Colocalisation
Statistical analyses of the confocal scanning microscope images are performed with Volocity software (Improvision, PerkinElmer). This software provides a tool designed specifically for colocalisation analysis as described by Manders et al.(1993)5. Colocalisation in the context of digital fluorescence imaging can be described as the detection of a signal at the same voxel (pixel volume) location in each channel. The two channels are made up of images of two different fluorochromes taken from the same sample area (Volocity user guide). Statistical analyses are performed with Volocity software (Improvision, PerkinElmer) using Quantitative Colocalisation Analysis described below.
Quantitative Colocalisation Analysis
4. Representative Results
Intracellular localisation of Opc-expressing Neisseria meningitidis and α-actinin
Confocal imaging of human brain microvascular endothelial cells infected with Nm for 3 and 8 hours as described above indicated colocalisation of α-actinin and Nm which appeared to be less frequent in 3 h infection experiments (not shown) compared with cultures infected for 8 h (Figure 1 A-F). A demonstrable colocalisation of α-actinin with Opc-expressing meningococci was observed each time in >5 replicate experiments. Statistical analysis of colocalisation using several confocal images was carried out as described above. Overall, in HBMEC infected with Opc-expressing meningococci, >25% overlap of the green (α-actinin) and red (Nm) pixels was obtained (Figure 2B, Overlap coefficient R). In contrast to α-actinin, experiments in which labeling of internalised bacteria and either actin or vimentin was performed, occasional colocalisation was observed with actin but that with vimentin was rare (Figure. 2B).
The data were also analyzed using the coefficient My, which takes into account the relative abundance of each moiety. My is a measure of the frequency of occurrence of the more abundant signal (in this case green, α-actinin) each time the less abundant signal (in this case red, bacteria) occurs. This measure shows a striking level of occurrence of α-actinin in the vicinity of internalised meningococci (Figure 2A and C).
Figure 1. Confocal laser scanning microscopy to assess intracellular interactions of N. meningitidis with α-actinin. A-H. Confluent endothelial monolayers grown on coverslips were infected with the Opc-expressing (A-F) N. meningitidis. After 8 h, non-adherent bacteria were washed off, cells fixed with paraformaldehyde and permeabilised with 0.1% Triton X-100. Subsequently, bacteria and α-actinin were stained as described above (α-actinin, green; bacteria, red).
A-C. One field showing x-y images of the location of Nm (A) or α-actinin (B). The overlay image in (C) indicates several regions in which yellow-orange color appears suggesting colocalisation. Arrows in (A) and (B) show regions where a high degree of α-actinin accumulation appears to have occurred around bacteria.
D. Optical dissection of an infected HBMEC monolayer indicating colocalisation around intracellular bacteria located at the base of a cell.
Again, this colocalisation is not due to accidental proximity of α-actinin, as the general stain of α-actinin in this region is low.
E and F. Three-dimensional images of infected HBMEC monolayers processed as above. An oblique view of the apical surface (E) shows adherent bacteria stained red (red arrow) whereas several bacteria located towards the basal surfaces of endothelial cells (yellow arrow) are distinctly orange/yellow in color. Basal location can be more clearly seen in (F) which is an end-on X-Z cross-section.
G. A negatively stained electron microscope image of N. meningitidis showing its predominant diplococcal from. Each coccus is approximately 0.5 μm in diameter.
Figure 2. Localisation and distribution of α-actinin, actin and vimentin in HBMEC cells.
A. Infected monolayers of HBMEC were treated as described in the legend above but in addition to α-actinin, some coverslips were used for the detection of actin or vimentin by procedure similar to that for α-actinin. As above, α-actinin concentrated around several internalised bacteria (white arrows). Vimentin and actin did not colocalise with bacteria to appreciable levels. Bar represents 20 μm.
B. & C. The values for the coefficients R and My were obtained from more than three experiments using Volocity software as described above.
The possibility of binding of internalised Opc-expressing Neisseria meningitidis to α-actinin was explored using HBMEC by the examination of the colocalisation of bacteria and the cytoskeletal protein in infected cells after 3 and 8 h incubation period. By confocal microscopy, colocalisation of Neisseria meningitidis with α-actinin could be demonstrated. Notably, although bacteria were internalised at 3 h, there was little colocalisation with α-actinin at this time point. Bacterial association with the cytoskeletal protein appeared to require a longer period of intracellular residence as after 8 h infection period, significant numbers of bacteria had α-actinin apparently in close association. Alpha-actinin is a multifunctional protein, and bacterial interactions with the cytoskeletal element could have significant influence on the target cell function which is a subject of current studies.
Quantification of colocalisation as described above requires meticulous specimen preparation. Particular attention should be given to specimen fixation, blocking period and antibody dilutions. For the best signal to noise ratio, each primary and secondary antibody should be titrated in preliminary experiments to determine the optimum concentrations. In our experience, the mounting medium Mowiol produced better images.
No conflicts of interest declared.
The studies were funded by the Wellcome Trust and Meningitis UK. HBMEC cell line was provided by Dr K.S. Kim. Confocal imaging and electron microscopy were performed in the Wolfson Bioimaging Facility, University of Bristol. We would also like to thank Mr Alan Leard, Dr. Mark Jepson (University of Bristol), and Mr. Alan Tilley (PerkinElmer) for their help and advice.
|12 well plate||Corning||BC311|
|Glass Coverslips||VWR international||631-0152|
|BHI AGAR||LAB M||LAB048|
|RPMI 1640||Lonza Inc.||BE12-167E|
|MEM Non Essential Aminoacids||Sigma-Aldrich||M7145|
|HANK’S balanced salt solution||Sigma-Aldrich||H9269|
|MEM Vitamin solution||Sigma-Aldrich||M6895|
|Anti€‘Nm antiserum||Laboratory raised||Rabbit serum|
|TRITC anti-rabbit Ig||Sigma-Aldrich||T6778|
|Anti Î±€‘actinin mAb [7H6]||Abcam||Ab32816||IgG|
|Anti actin mAb||Sigma-Aldrich||A4700||IgG2a|
|Anti vimentin mAb||Sigma-Aldrich||V6630||IgG1|
|ALEXA FLUOR 488 anti€‘mouse IgG||Invitrogen||A11001|
|FITC anti-mouse IgG||Sigma-Aldrich||F-2012|
1. Confocal Laser Scanning Microscopy (CLSM):
Leica SP5 confocal imaging system: This system, by using a combination of AOTF (acousto-optical tuneable filter) and an AOBS (acousto-optical beam splitter), simplifies excitation with specific wave lengths.
Leica confocal software LCS, Leica Microsystems, Germany.