Quantifying cellular interaction dynamics in 3D fluorescence microscopy data - PubMed (original) (raw)

Quantifying cellular interaction dynamics in 3D fluorescence microscopy data

Frederick Klauschen et al. Nat Protoc. 2009.

Abstract

The wealth of information available from advanced fluorescence imaging techniques used to analyze biological processes with high spatial and temporal resolution calls for high-throughput image analysis methods. Here, we describe a fully automated approach to analyzing cellular interaction behavior in 3D fluorescence microscopy images. As example application, we present the analysis of drug-induced and S1P(1)-knockout-related changes in bone-osteoclast interactions. Moreover, we apply our approach to images showing the spatial association of dendritic cells with the fibroblastic reticular cell network within lymph nodes and to microscopy data regarding T-B lymphocyte synapse formation. Such analyses that yield important information about the molecular mechanisms determining cellular interaction behavior would be very difficult to perform with approaches that rely on manual/semi-automated analyses. This protocol integrates adaptive threshold segmentation, object detection, adaptive color channel merging, and neighborhood analysis and permits rapid, standardized, quantitative analysis and comparison of the relevant features in large data sets.

PubMed Disclaimer

Figures

Figure 1

Figure 1

Automated threshold segmentation. Example original (A) fluorescence two-photon image and intensity histograms showing the characteristic shape (B,C). Histograms can be divided into 1) a “peak” region (lower intensity values) and 2) a “constant” region stretching from the end of the “peak” region to higher intensity values. B) osteoclasts (green) and C) bone tissue (blue). D) Resulting segmented image after automated thresholding using maximum curvature estimation method for threshold selection illustrated in E and F. E) The curvature of the graph of the histogram fit (red) is shown in blue. Selection of the threshold at the curvature maximum is indicated by blue arrow. The approximation of the maximum curvature computation by the slope constant criterion f'(x)≈Δh/Δi=1 is depicted by the green curve and arrow. Scale bar represents 10μm. All animal procedures used in this study have been approved by the Animal Care and Use Committee, NIAID, NIH.

Figure 2

Figure 2

Adaptive channel merging. A, Original image data of T (green) and B (blue) cells. B, segmentation result, interacting T cells (red), interface (white). Magnification of blue (C) and green (D) channels with overlay (E) and segmentation result (F) from inset in A, B. Intensity profile along arrow in E is shown in G. Horizontal green and blue lines in G indicate computed threshold. Black vertical lines (1, 2, and 3) show the selection alternatives of the border between the blue and green channels under the condition that (1) either the blue or (3) the green object is given priority. (2) is the result of the border selection when comparing absolute intensities of competing channels. H, normalized intensity profile allows for balanced identification of the border between color channels (result shown in F). Scale bar represents 10 μm. All animal procedures used in this study have been approved by the Animal Care and Use Committee, NIAID, NIH.

Figure 3

Figure 3

Analysis of the impact of variations in the derivative constant on the normalized interface area (ratio interface area / total bone surface area) and on the difference between “wild-type” and “knock-out” groups. Mean values are based on 20 data sets each in each group. While changing the derivative constant by a factor of 2 in both directions relative to 1.0 results in average normalized interface area changes of ~10%, the differences change only 1.1% (0.5→1.0) and 2.2% (1.0→2.0). Scale bar represents 10 μm. See also Figure 5. (For more information on the experiments see Ishii et al., Nature, in press). All animal procedures used in this study have been approved by the Animal Care and Use Committee, NIAID, NIH.

Figure 4

Figure 4

Segmentation example results. Left column: original images; right column: segmentation results. A) Bone-osteoclast interaction (osteoclasts: green; bone: blue; red: interacting osteoclasts). B) Dendritic cells (green/red) and fiber network (blue) in lymph nodes. (white: interface area). Analysis of the whole 3-D data set reveals that 22% of the DC surface is attached to fibers. C) T-B-cell interaction in lymph nodes (T cells: green/red; B cells: blue). Scale bars represent 10 μm. All animal procedures used in this study have been approved by the Animal Care and Use Committee, NIAID, NIH.

Similar articles

Cited by

References

    1. Germain RN, et al. An extended vision for dynamic high-resolution intravital immune imaging. Seminars in immunology. 2005;17:431–441. - PMC - PubMed
    1. Stoll S, Delon J, Brotz TM, Germain RN. Dynamic imaging of T cell-dendritic cell interactions in lymph nodes. Science (New York, N.Y. 2002;296:1873–1876. - PubMed
    1. Xu X, Meier-Schellersheim M, Yan J, Jin T. Locally controlled inhibitory mechanisms are involved in eukaryotic GPCR-mediated chemosensing. The Journal of cell biology. 2007;178:141–153. - PMC - PubMed
    1. Fisher DT, et al. Hurdles to lymphocyte trafficking in the tumor microenvironment: implications for effective immunotherapy. Immunological investigations. 2006;35:251–277. - PubMed
    1. Nishida H, Okabe S. Visualization of synapse-glia dynamics. Brain and nerve = Shinkei kenkyu no shinpo. 2007;59:755–761. - PubMed

Publication types

MeSH terms

LinkOut - more resources