Intravital imaging

Abstract

Until recently, the idea of observing life deep within the tissues of a living mouse, at a resolution sufficient to pick out cellular behaviors and molecular signals underlying them, remained a much-coveted dream. Now, a new era of intravital fluorescence microscopy has dawned. In this Primer, we review the technologies that made this revolution possible and demonstrate how intravital imaging is beginning to provide quantitative and dynamic insights into cell biology, immunology, tumor biology, and neurobiology.

Figure 1. IVM to assess fundamental cell biological events in vivo

Cell division Top: Time-lapse IVM of HeLa H2b–mCherry/tubulin–EGFP cells reveals stages of mitosis in vivo. Bottom left: A higher magnification image reveals kinetochore fibers in a normal mitotic cell and microtubule rosettes surrounded by chromosomes in a taxol-arrested mitotic cell. Bottom right: Time-lapse IVM can be used to derive quantitative information, such as mitotic indexes in response to anti-cancer drugs. Images reproduced from (Orth et al., 2011). Cell death. Top: A variety of reporters reflecting cell status are available. From left to right: H2b-XFP assesses nuclear morphology (Orth et al., 2011); fluorescent intermembrane space proteins (IMS-XFP) track mitochondrial outer membrane permeabilization (MOMP); Bcl2-XFP follows programmed cell death; and LC3-XFP identifies the formation of autophagosomes. Bottom left: Time-lapse IVM allows the status (dead or alive) of a surrogate cancer cell to be tracked after attack by a cytotoxic T lymphocyte (green). The target cell had been previously labeled with a red cytoplasmic (CMTMR) and a blue nucleic (Hoechst33342) dye. A lethal hit leads to motility arrest of the target cell followed by a decreased red/blue ratio; this is caused by the loss of cell membrane integrity leading to the release of the cytoplasmic dye into the extracellular milieu (visible at the 33 minute 30 second time point). Images reproduced from (Mempel et al., 2006). Bottom right: Cytotoxic T lymphocyte-mediated tumor cell apoptosis is detected using a Forster resonance energy-transfer-based (FRET-based) reporter of caspase 3 activity. Tumor cells express the CFP and YFP molecules linked by a peptide containing the caspase 3 cleavage motif DEVD. Apoptosis-induced caspase 3 activation cleaves the DEVD motif and induces FRET loss. Live tumor cell: orange; apoptotic tumor cell: green; cytotoxic T lymphocyte: red. Images reproduced from (Breart et al., 2008). Cell migration. Left: Repeated imaging at defined intervals enables single cell tracks to be reconstructed (top) and quantitative information derived (bottom). The data demonstrate that reservoir splenic monocytes increase their motility in response to a distant injury. Right: IVM can also be used to study intravasation or extravasation processes. This example shows extravasation of splenic monocytes for redistribution to distant tissues. Images reproduced from (Swirski et al., 2009). Cell communication. Top: Images of rare long lasting physical interactions between two immune cell types (images reproduced from (Mempel et al., 2006)). Bottom: Simultaneous visualization of multiple cell types reveals complex tissue interactions. This image shows a lymph node dendritic cell interacting with five different lymphocytes. Time-lapse imaging can be used to derive quantitative information, such as the interaction times between different cell types.

Figure 2. Mapping of long-term cell fate by IVM

(a) Green to red photoswitching of Dendra2-expressing tumor cells in avascular (left) and vascular (right) regions in vivo. White dotted lines delineate the position of the vessel. Photoswitched cells can be tracked for up to 5 days and show distinct behaviors depending on their original position. Images reproduced from (Kedrin et al., 2008). (b) Orange to far-red photoswitching of PSmOrange in HeLa cells. Micrographs show PSmOrange-tubulin co-expressed with nuclear localization signal (NLS)-mCherry. Images reproduced from (Subach et al., 2011). (c) The Brainbow transgenic technique for combinatorial expression of fluorescent proteins. The image, reproduced from (Livet et al., 2007), is of individual neurons in the hippocampus, where neurons express either CFP, YFP or RFP transgenes. Co-expression of multiple copies of fluorescent reporters can generate up to 90 colors. (d) RGB marking with LeGO vectors encoding mCherry (red), Venus (yellow-green) and Cerulean (blue) fluorescent proteins represents another recombination strategy that can be used for mapping cell fate. Simultaneous expression of these fluorescent proteins creates a spectrum of colors that can be then used to identify single clones. Images reproduced from (Weber et al., 2011).

Figure 3. IVM for assessing drug distribution in vivoat the single cell level

(a) Bioorthogonal chemistry allows a fluorochrome to be irreversibly “clicked” onto drugs. Here, the poly-ADP-ribose-polymerase (PARP)-1 inhibitor AZD2281 was modified to incorporate the permeable fluorochrome BODIPY. (b) In vivo distribution of the fluorescent PARP-1 inhibitor (red) 90 minutes after its intravenous injection into fibrosarcoma-bearing mice; tumor cells are expressing fluorescent H2b (green). IVM could identify drug distribution in most tumor cells (white arrows), where the drug is most likely accumulating within the nucleoli. Within the same microenvironment, the drug also accumulates in high concentrations in non-tumor cells (red arrows).