Am J Physiol Cell Physiol 2008 Scriven C1119 22

7/30/2019 Am J Physiol Cell Physiol 2008 Scriven C1119 22

1/5

doi: 10.1152/ajpcell.00133.2008294:C1119-C1122, 2008. First published 19 March 2008;Am J Physiol Cell Physiol

David R. L. Scriven, Ronald M. Lynch and Edwin D. W. MooremicroscopyImage acquisition for colocalization using optical

You might find this additional info useful...

13 articles, 2 of which you can access for free at:This article citeshttp://ajpcell.physiology.org/content/294/5/C1119.full#ref-list-1

4 other HighWire-hosted articles:This article has been cited byhttp://ajpcell.physiology.org/content/294/5/C1119#cited-by

including high resolution figures, can be found at:Updated information and serviceshttp://ajpcell.physiology.org/content/294/5/C1119.full

found at:can beAmerican Journal of Physiology - Cell PhysiologyaboutAdditional material and information

http://www.the-aps.org/publications/ajpcell

This information is current as of February 4, 2013.

0363-6143, ESSN: 1522-1563. Visit our website at http://www.the-aps.org/.

Rockville Pike, Bethesda MD 20814-3991. Copyright 2008 the American Physiological Society. ISSN:molecular physiology. It is published 12 times a year (monthly) by the American Physiological Society, 9650

is dedicated to innovative approaches to the study of cell andAmerican Journal of Physiology - Cell Physiology
http://ajpcell.physiology.org/content/294/5/C1119#cited-byhttp://ajpcell.physiology.org/content/294/5/C1119#cited-byhttp://ajpcell.physiology.org/content/294/5/C1119#cited-byhttp://ajpcell.physiology.org/content/294/5/C1119#cited-byhttp://ajpcell.physiology.org/content/294/5/C1119#cited-by


2/5

Perspectives in Cell Physiology

Image acquisition for colocalization using optical microscopy

David R. L. Scriven,1 Ronald M. Lynch,2 and Edwin D. W. Moore1

1Department of Cellular and Physiological Sciences, University of British Columbia, Life Sciences Institute, Vancouver,

British Columbia, Canada; and 2Department of Physiology, University of Arizona, Arizona Health Sciences Center,

Tucson, Arizona

Submitted 29 February 2008; accepted in final form 17 March 2008

Scriven DR, Lynch RM, Moore ED. Image acquisition forcolocalization using optical microscopy. Am J Physiol Cell Physiol294: C1119C1122, 2008. First published March 19, 2008;doi:10.1152/ajpcell.00133.2008.Colocalization, in which images oftwo or more fluorescent markers are overlaid, and coincidence be-tween the probes is measured or displayed, is a common analyticaltool in cell biology. Interpreting the images and the meaning of thisidentified coincidence is difficult in the absence of basic informationabout the acquisition parameters. In this commentary, we highlightimportant factors in the acquisition of images used to demonstratecolocalization, and we discuss the minimum information that authorsshould include in a manuscript so that a reader can interpret both the

fluorescent images and any observed colocalization.

THE ANALYSIS AND DISPLAY of colocalization in optical imagesobtained through microscopy is now a commonly used exper-imental tool. In this technique, two or more molecules arelabeled with fluorescent probes, and the images, acquired at theappropriate wavelengths, are overlaid. The extent of colocal-ization of the probes is then visualized and often quantified.This general description of a colocalization experiment coversa host of microscopic techniques, labeling methods, and imageanalysis algorithms that can be used in various combinations.To provide some uniformity in presentation, we are recom-mending several procedures for collecting colocalization data.

We also present guidelines for the information that should beincluded in the methods section of the manuscript that willallow any reader to interpret the results. We are focusing ourdiscussion on images acquired with wide-field and confocalmicroscopes since these are by far the most common ap-proaches for image acquisition.

The Nyquist Criteria

When the term colocalization is used in optical microscopy,the objects that are being compared are usually pixels or voxelsin digital images (A pixel is the smallest element in a two-dimensional digital image and a voxel is its three-dimensionalcounterpart). When we ask if two molecules are colocalized,

we are, in effect, asking Is the voxel illuminated by a fluoro-phore attached to (or part of) molecule A the same voxel as thatilluminated by another fluorophore attached to molecule B?Thus the size of the voxels is important for both planning andinterpreting the experiments. When the image is digitized(voxelized), it has the same effect as placing a grid over theimage and replacing the objects at each grid point (voxel) withthe total intensity of all of the objects contained in it. As voxel

size increases, resolution decreases so that separate structuresbecome lumped together in the same voxel. At some point, thevoxelized image will no longer accurately represent the origi-nal object and artifacts may appear. The voxel size at which anobject is accurately represented by an image is given by theNyquist criteria, which are a function of several variables: thenumerical aperture of the objective, the emission wavelengthof the fluorophore, the type of microscope (wide field orconfocal) and, for depth, the refractive index of the medium.Further details can be found in Castleman (2) and van derVoort and Strasters (13). The formulas and a summary of

appropriate values for a conventional microscope, using oilimmersion lenses of various numerical apertures and illumi-nating wavelengths, are given in Table 1. Confocal micro-scopes, using laser light and a typical pinhole size of250 nm,have Nyquist values that are similar to those listed (13).Ideally, imaging parameters should be optimized to reach thesevalues to obtain the best resolution for evaluating colocaliza-tion. In real systems, noise will increase the theoretical valueslisted in the table, and values within 10% of those tabulatedwill still give an acceptable result.

Voxels larger than those specified by the Nyquist criteriawill under sample the image, create artifacts, and result in falsecolocalizations. Figure 1 shows a ventricular myocyte col-lected at voxel sizes satisfying the Nyquist criteria (Fig. 1, A

and C) and the same image resampled with voxels four timeslarger in Xand Yand 1.6 times in Z(Fig. 1, B and D). It is clearthat not only does a large amount of colocalization appear inthe under-sampled image but more importantly, structuresappear that are not present in the original (compare Fig. 1, Cand D). Collecting images with voxels smaller than the Nyquistvalues requires longer exposure times to achieve a comparablesignal-to-noise ratio, increasing photobleaching and photodam-age, while producing little additional information.

For proper evaluation of a manuscript, authors should there-fore report the objectives numerical aperture, the wavelengthsthat are being used, the voxel sizes, the refractive index of themedium and the Z spacing between image planes, and show

how these values relate to the Nyquist criteria for the experiment.Diffuse Labeling

In some cases, one of the fluorophores is present in almostevery voxel. This can occur when labeling a cytosolic proteinor a widely distributed protein like actin. Under these condi-tions, colocalization by chance has almost 100% probability,making any observed or measured colocalization, not signifi-

Address for reprint requests and other correspondence: E. D. W. Moore,Dept. of Cellular and Physiological Sciences, Univ. of British Columbia, LifeSciences Institute, 2350 Health Sciences Mall, Vancouver, British ColumbiaV6T 1Z3, Canada (e-mail: [email protected]).

The costs of publication of this article were defrayed in part by the paymentof page charges. The article must therefore be hereby marked advertisementin accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Am J Physiol Cell Physiol 294: C1119C1122, 2008.First published March 19, 2008; doi:10.1152/ajpcell.00133.2008.

0363-6143/08 $8.00 Copyright 2008 the American Physiological Societyhttp://www.ajpcell.org C1119


3/5

cant. Experimental approaches other than colocalization byoptical microscopy are required to resolve this problem.

Deconvolution

Wide-field images contain much out-of-focus light, whichwill confound colocalization analysis. For this reason, it ismandatory that wide-field images be deconvolved before co-localization analysis is attempted. Authors should state whichalgorithm or software package was used. It should be notedthat the measurement of colocalization in confocal images is

also improved by deconvolution (7, 11).

Bleed Through

A fundamental assumption made when measuring colocal-ization is that the signal at each wavelength is independent of

the other. Bleed through of signal from one channel to anothercan result in an apparent colocalization when none, in fact,exists. For example, even with modern filters, there is signif-icant bleed through from FITC to rhodamine, or from cyanfluorescent protein to yellow fluorescent protein. Bleed throughshould be estimated by creating a control with labeling in onechannel and none in the other and then recording the signal

coming from the unlabeled wavelength(s). The resultant im-age should be random, having no discernable pattern, and itsintensity should be less than the chosen threshold. If, how-ever, there is significant bleed through, the control imagescan be used in concert with spectral linear unmixing (8, 14)to separate the two signals. Once corrected, the images canthen be used for analysis. Some newer instruments havespectral unmixing built into their software and correct forbleed through at the time of acquisition, although it isimportant to check how well the signals are separated beforeaccepting the results (15).

Colocalization in X-Y versus Z

In all optical microscopes, except the 4Pi, the resolvingpower in Z is only one-third to one-quarter of that in X and Y(9). The consequence is that colocalization is most powerfulwhen performed on structures that run along the XY plane,parallel to the coverslip, whereas it is least accurate for objectsthat are running parallel to the Z-axis of the microscope. Theconsequences of this are best seen in a structure that runs bothparallel and perpendicular to the XYplane. This is diagrammedin Fig. 2A, which shows a cross-section through a cell in whichfluorophores of the same color are separated by 200 nm, andthe two different fluorophores are interspersed at 100-nmintervals. Figure 2B shows the result of superimposing theseplanesthe pixels that lie in the transverse plane are notcolocalized, but those at the edges, lying along the Z-axisappear to be. This effect can be observed in real cells. Figure2C shows a rat ventricular myocyte labeled with antibodiesspecific for vinculin (green) and caveolin-3 (red). A 1-mthick section from the cells center is displayed in Fig. 2Ci anddemonstrates almost complete colocalization at the cell surface

Fig. 1. The effects of pixel size: An image,2 m thick, of a rat ventricular myocytelabeled with antibodies for vinculin (green)and caveolin-3 (red), with colocalized voxelswhite. The scale bar is 5 m and the voxeldimensions are: (A) 100 100 250 nm,

which satisfies the Nyquist criteria and (B)400 400 400 nm, which does not. Thecolocalization of vinculin with caveolin-3(calculated as the percentage of voxels la-beled for vinculin that also contained caveo-lin-3) was 19% in A and 61% in B and thecolocalization of caveolin-3 with vinculinwas 6% in A and 32% in B. (C, D) a blowupof the inserts (white square) in A and Brespectively. Note the structural artifacts inthe under-sampled image D that do notappear in the original (A, C).

Table 1. Lateral and axial sampling distances determinedby the Nyquist criteria

Fluor ophore Emiss ion W avel engt h, nm

NA

1.2 1.3 1.4

DAPI/Alexa 350 450 97 89 83394 316 248

Cy2 505 105 97 90426 341 269

FITC/Alexa 488 520 108 100 93441 353 278

Cy3/Alexa 555 565 118 109 101479 383 302

TRITC/Alexa 546 575 120 111 103487 390 307

Texas red/Alexa 594 615 128 118 110521 417 329

Cy5/Alexa 647 670 140 129 120568 455 360

The x and y sampling distances are calculated (in nm) using the formula/(4NA) (4) where is the emission wavelength (in nm) and NA is thenumerical aperture of lens. The axial (z) sampling distances (in italics) are foran oil (n 1.515) immersion objective and are calculated with the formula,

z /[2n (1 cos)], where sin1 (NA/n) and n is the refractive indexof the medium (13). DAPI, 6-diamidino-2-phenylindole; TRITC, tetramethyl-rhodamine isothiocyanate, Cy2, Cy3, Cy5, different fluorescent cyanine dyes.


C1120 IMAGE ACQUISITION FOR COLOCALIZATION

AJP-Cell Physiol VOL 294 MAY 2008 www.ajpcell.org


4/5

running parallel to the Z-axis (arrowheads). Figure 2Cii is animage of the top surface of the same cell running parallel to theXY axis, which shows 5% colocalization (boxed regions). Astereo pair of the complete data stacks (Fig. 2Ciii) shows thatcolocalization is confined to those regions of the sarcolemmathat are oriented largely parallel to the microscopes Z-axis, butthese molecules are, in fact, not colocalized.

Sometimes the objects to be colocalized have little or no X-Ycomponent (e.g., synapses or gap junctions), and in this casethe measured colocalization has less predictive power than itwould for objects oriented in the X-Y plane, since the colocal-

ized objects could be separated by 500 nm or more (i.e., 2 theNyquist limit in Z, see Table 1).

Sampling the Cell

The great advantage of using optical microscopy to assesscolocalization, as opposed to other techniques including elec-

tron microscopy, is that images of entire cells or their substruc-tures can be readily collected and analyzed, and regionalvariations can be visualized. When focusing an analysis onspecific regions of a cell, care needs to be taken when deciding

Fig. 2. Edge Effects causing Colocalization.(A and B): A diagram showing how the effectoccurs. (A) A cross-section in which fluoro-phores of the same color are separated by200 nm, and the two different fluorophoresare interspersed at 100 nm intervals. The gridrepresents 4 image planes, 250 nm apart in Zand 100 100 nm in X and Y. (B) Projec-tion of the image planes onto a line of vox-

els; those in which colocalization occurs arecolored yellow. (C) A rat ventricular myo-cyte labeled for vinculin (green) and caveo-lin-3 (red), colocalized voxels are white. 2 mthick sections from the center of the cell (i)and the cell surface (ii). The arrowheadsindicate sarcolemma that is parallel to theoptical axis, and the arrows indicate sarco-lemma that is perpendicular to the opticalaxis. The boxed regions, in which only 4% ofthe vinculin and 5% of the caveolin-3 werecoincident, were used to numerically assesscolocalization. (iii) Stereo-pair, 6 rotation,of the entire data set, 15 m deep.


C1121IMAGE ACQUISITION FOR COLOCALIZATION



5/5

which part of the cell is sampled, because this can dramaticallyaffect the results. Specifically, two recent papers (3, 10) haveshown that regional variations in colocalization can occur inthe absence of any obvious structural differences.

Guidelines for Publication

To enable readers to interpret images and to evaluate themeaning of colocalization from a given experiment, manu-scripts should include the following information:

The pixel sizes, the Z spacing between image planes, therefractive index of the medium, and how this relates to theNyquist criteria for the experiment (Table 1).

The objectives magnification and numerical aperture. For images obtained with a confocal microscope, either

the pinhole size or the axial resolution should be reported. The emission wavelengths of the fluorophores and the

filter sets used. Scalebars should be included in each image group.

In addition, the following conditions should be satisfied. Wide-field images must be deconvolved, and the method-

ology reported. Images in which the colocalization is primarily along the

Z-axis should, where possible, be examined for edge artifacts. Bleed through should be evaluated and corrected for, if

present. If corrections are made, details of the methodologyused should be provided.

Conclusion

Colocalization can be a powerful method for determiningwhether molecules have the potential to associate or interact,but it cannot confirm that they do. The inherent limitations ofthe technique mean that confirmation may require electronmicroscopy (1, 5) or other microscopic techniques such as

fluorescence resonance energy transfer (6, 12).

GRANTS

This work was funded by the American Heart Association GIA 93014700and National Heart, Lung, and Blood Institute Grant HL-66044 (to R. M.

Lynch), a Grant-in-Aid from the Heart and Stroke Foundation of BritishColumbia and the Yukon, and by grants from the National Sciences andEngineering Research Council and the Canadian Institutes of Health Research(to E. D. W. Moore).

REFERENCES

1. Anderson JB, Carol AA, Brown VK, Anderson LE. A quantitativemethod for assessing co-localization in immunolabeled thin section elec-

tron micrographs. J Struct Biol 143: 95106, 2003.2. Castleman KR. Resolution and sampling requirements for digital image

processing, analysis, and display. In: Electronic Light Microscopy, editedby Shotton D. New York: Wiley-Liss, 1993, p. 7193.

3. Dan P, Lin E, Huang J, Biln P, Tibbits GF. Three-dimensional distri-bution of cardiac Na-Ca2 exchanger and ryanodine receptor duringdevelopment. Biophys J 93: 25042518, 2007.

4. Gonzalez RC, Woods RE. Digital Image Processing. New York: Addi-son-Wesley, 1992.

5. Griffiths G, Parton RG, Lucocq J, van Deurs B, Brown D, Slot JW,Geuze HJ. The immunofluorescent era of membrane traffic. Trends Cell

Biol 3: 214219, 1993.6. Jares-Erijman EA, Jovin TM. FRET imaging. Nat Biotechnol 21:

13871395, 2003.7. Landmann L. Deconvolution improves colocalization analysis of multi-

ple fluorochromes in 3D confocal data sets more than filtering techniques.J Microsc 208: 134147, 2002.

8. Nadrigny F, Rivals I, Hirrlinger PG, Koulakoff A, Personnaz L,Vernet M, Allioux M, Chaumeil M, Ropert N, Giaume C, Kirchhoff F,Oheim M. Detecting fluorescent protein expression and co-localisation onsingle secretory vesicles with linear spectral unmixing. Eur Biophys J35:533547, 2006.

9. Schrader M, Bahlmann K, Giese G, Hell SW. 4Pi-confocal imaging infixed biological specimens. Biophys J 75: 16591668, 1998.

10. Scriven DR, Klimek A, Asghari P, Bellve K, Moore ED. Caveolin-3 isadjacent to a group of extradyadic ryanodine receptors. Biophys J 89:18931901, 2005.

11. Sedarat F, Lin E, Moore ED, Tibbits GF. Deconvolution of ConfocalImages of Dihydropyridine and Ryanodine Receptors in DevelopingCardiomyocytes. J Appl Physiol 97: 10981103, 2004.

12. Sekar RB, Periasamy A. Fluorescence resonance energy transfer (FRET)microscopy imaging of live cell protein localizations. J Cell Biol 160:629633, 2003.

13. van der Voort HTM, Strasters KC. Restoration of confocal images forquantitative image analysis. J Microsc 178: 165181, 1995.14. Zimmermann T, Rietdorf J, Pepperkok R. Spectral imaging and its

applications in live cell microscopy. FEBS Lett 546: 8792, 2003.15. Zucker RM, Lerner JM. Wavelength and alignment tests for confocal

spectral imaging systems. Microsc Res Tech 68: 307319, 2005.


C1122 IMAGE ACQUISITION FOR COLOCALIZATION


Documents

Am J Physiol Cell Physiol 2008 Scriven C1119 22