REVIEWS

TECHNICOLOUR TRANSGENICS: IMAGING TOOLS FOR FUNCTIONAL GENOMICS IN THE MOUSE Anna-Katerina Hadjantonakis*, Mary E. Dickinson‡, Scott E. Fraser‡ and Virginia E. Papaioannou* Over the past decade, a battery of powerful tools that encompass forward and reverse genetic approaches have been developed to dissect the molecular and cellular processes that regulate development and disease. The advent of genetically-encoded fluorescent proteins that are expressed in wild type and mutant mice, together with advances in imaging technology, make it possible to study these biological processes in many dimensions. Importantly, these technologies allow direct visual access to complex events as they happen in their native environment, which provides greater insights into mammalian biology than ever before. THERMOLABILE

Unstable at moderate or increased temperatures. CHIMAERA

An animal that is generated from, and comprises, several genetically distinct populations of cells that are derived from more than one individual.

*Department of Genetics and Development, College of Physicians and Surgeons of Columbia University, New York 10032, USA. ‡ Beckman Institute, California Institute of Technology, Pasadena, California 91125, USA. Correspondence to A.-K.H. e-mail: [email protected] doi:10.1038/nrg1126

NATURE REVIEWS | GENETICS

The mouse is the premier mammalian model organism1 for studying development and disease. The mouse genome sequence is now available and the mouse germline can be manipulated in almost every way imaginable, in both a random2–7 and a directed fashion8–11, to study the phenotypic consequences of perturbations. As techniques for manipulating the genome have multiplied, so have the tools for the study of complex phenotypes. Recently, there has been a surge of interest in using fluorescence microscopy for phenotypic analysis (BOX 1). This is attributable to three converging factors. First, genetically-encoded fluorescent proteins have been developed that provide high-contrast multicolour vital markers for monitoring gene activity and protein function. Second, improvements in fluorescence-based imaging technology have allowed deeper imaging and better spectral separation in live tissues. Third, in vitro culture systems have been developed that allow the normal growth of mammalian embryos and explanted tissues, which is a requirement for vital-imaging experiments. These developments allow direct connections to be made between genetic lesions and their cellular consequences, and so pave the way for high-resolution highthroughput functional genomics. Indeed, these features are already being exploited by many laboratories in studies of dynamic biological events in several model

organisms, including Drosophila12,13 and zebrafish14–17 (see links in online links box). Having overcome some of its original challenges, such as THERMOLABILITY and quantum yield of wtGFP at 37°C, which made it unsuitable for use in mammalian systems, the mouse is fast catching up with these organisms. In this review, we outline the use of fluorescent proteins in mice and, to illustrate their power, we describe some examples of the use of transgenic mouse lines in investigating mammalian development and neurobiology. Development of GFP-based reporters

For almost two decades, lacZ, which is the bacterial gene that encodes β-galactosidase, has been the marker of choice for the generation of reporter-expressing mouse strains18,19, as it is easy to detect and has a high cellular resolution. Although advantageous for marking distinct populations of cells in CHIMAERAS or reporting the activity of specific promoters, lacZ cannot be used to mark cells in living tissues, because staining for lacZ requires tissue fixation. In 1994, green fluorescent protein (GFP), which is derived from the bioluminescent jellyfish Aequorea victoria20, was first introduced into Caenorhabditis elegans as a genetically stable fluorescent marker21 that could be detected in living tissues. The potential uses VOLUME 4 | AUGUST 2003 | 6 1 3

REVIEWS

Box 1 | Imaging requirements The advent of genetically-encoded fluorescent reporters has a Transgenic mouse model prompted researchers to find new ways to either get their Genetically-encoded fluorescent protein mice under the microscope or bring the microscope to their mice. Non-invasive time-lapse imaging of vital markers in mouse embryos, adults and organs in a physiological environment requires the availability of a transgenic mouse model that expresses one or more gene-based vital molecular markers (a), a microscope equipped with an excitation source and detection system (b), and a computer equipped with data acquisition and processing software (c). Image-processing algorithms can then be applied to extract quantitative data and generate multidimensional reconstructions. b Microscope with in vitro culture and

c Image workstation

imaging capabilities

Data acquisition and processing

Maintain specimen under physiological conditions and acquire data Heater box Gas inlet

Culture chamber

Humidifier

EMBRYONIC-STEM-CELLMEDIATED TRANSGENESIS

A method in which DNA is introduced into embryonic stem (ES) cells and integrates randomly, or through gene targeting, into the genome. Transgenic ES cells are delivered to the germline through the generation of (ES cell↔embryo) chimaeras. FATE MAP

A spatial map of the fates of different embryonic cells at a particular stage of development. CHORDONEURAL HINGE

A region at the posterior end of a vertebrate embryo that gives rise to both neural and mesodermal cells.

614

of GFP in mice were immediately obvious but issues of thermolability, low fluorescence and insolubility of fluorescent proteins hampered their initial application. These problems were overcome through mutagenesis of GFP, and the first strains of transgenic mice that expressed readily detectable GFP variants were reported approximately five years ago22,23. Fluorescent proteins were also mutated to give reporters with different spectral properties, so in vivo multicolour labelling experiments are now possible if different colour variants are expressed in the same individual. Today, several, but not all, genetically-encoded autofluorescent reporters can be used successfully in mice (BOX 2), and hundreds of different strains have been generated that carry GFP or its spectral variants. Fluorescent proteins not only make it possible to mark specific cells in living organisms, but these cells can be followed using time-lapse fluorescence-imaging techniques. So, it is now possible to describe normal events and understand aberrant cellular behaviour in mutant lines by watching the cells in question. The versatility of genetic techniques in the mouse offers several approaches for introducing fluorescent-protein reporters into the germline. Fluorescent reporters can also be incorporated in mutagenesis regimes24,25, which paves the way for higher-throughput and increased-resolution functional genomics in the mouse. Mouse strains

| AUGUST 2003 | VOLUME 4

can be engineered to show fluorescence in a constitutive26,27, conditional28,29 or lineage-restricted fashion. This affords tremendous creativity in experimental design, as illustrated in the following examples. Constitutive reporters

The first strains of transgenic mice expressed fluorescent proteins constitutively. They were generated either by traditional zygote-injection-based transgenic approaches22 or EMBRYONIC STEM (ES) CELL-MEDIATED 23 TRANSGENESIS , so the fluorescent reporters were randomly integrated into the genome. These experiments were crucial in showing the developmental neutrality of GFP-variant reporters. Mice that expressed fluorescent proteins showed widespread fluorescence and were viable, fertile and morphologically indistinguishable from their wild type (non-transgenic) littermates. Mice that constitutively express fluorescent proteins are a source of genetically-marked cells that can be explanted, recombined, transplanted and used to generate chimaeras. Cells from such mice have recently been used to refine FATE MAPS of gastrula-stage embryos30,31 and to investigate the presence and developmental potential of cells in the CHORDONEURAL-HINGE region of the developing tail bud in embryos32. Also, constitutively expressed fluorescent proteins can be used to mark cells for the isolation of stem cells,

www.nature.com/reviews/genetics

REVIEWS

TROPHOBLAST

An extraembryonic lineage that is derived from the trophectoderm of the blastocyst, which gives rise to the fetal portion of the placenta.

which is now an area of active research. Fluorescentlytagged cells can be easily identified when introduced into progeny and separated using flow cytometry33. One of the first examples of the use of fluorescent stem cells was the evaluation of putative TROPHOBLAST stem cells34. Since then, fluorescent-protein reporters have proved to be an indispensable reagent in stem-cell

research and have been used to mark several types of 35 PLURIPOTENT cell, including HAEMATOPOIETIC stem cells and neural stem cells36–38. Fluorescent mosaics. Genetic MOSAICS, such as mouse chimaeras, provide a powerful way to study development and disease processes in multicellular organisms39.

Box 2 | Genetically-encoded fluorescent proteins used in mice

NATURE REVIEWS | GENETICS

Normalized emission

Normalized excitation

Here we describe the most a b ECFP EGFP EYFP DsRed HcRed ECFP EGFP EYFP DsRed HcRed popular autofluorescent 1.0 1.0 proteins, their origin, evolution and prominent features. 0.8 0.8 Green fluorescent protein 0.6 0.6 (GFP) and other fluorescent proteins that are cloned from 0.4 0.4 COELENTERATES share a similar structure, which comprises a 0.2 0.2 β-barrel-like structure enclosing a cyclic tripeptide 0 0 375 400 425 450 475 500 525 550 575 600 450 475 500 525 550 575 600 625 650 675 700 (Ser65-Tyr66-Gly67) 109,110 Wavelength (nm) Wavelength (nm) FLUOROPHORE . Mutagenesis has been carried out to increase Colour Name Origin Features References the THERMOSTABILITY and maturation kinetics of these GFP and mutagenized variants fluorochromes and to create Green wtGFP Aequorea victoria Original (wild type) GFP 20,21 different spectral variants EGFP Phe64Leu/Ser65Thr GFP variant with increased 55 (as described in the table). thermostability and fluorescence For example, enhanced green Blue ECFP Phe64Leu/Ser65Thr/ Cyan (blue-shifted) GFP 120 fluorescent protein (EGFP), Tyr66Trp/Asn146Ile/ spectral variant which is, at present, the most Met153Thr/Val163Ala popular variant for use in Yellow EYFP Ser65Gly/Val68Leu/ Yellow/green (red-shifted) GFP 109 mice55, carries mutations that Gln69Lys/Ser72Ala/ spectral variant have increased the QUANTUM Thr203Tyr YIELD and thermostability of Yellow Venus Phe49Leu/Phe64Leu/ Reaches maturity faster and has 54 GFP. Although mutations in Ser65Gly/Val68Leu/ increased fluorescence compared Gln69Lys/Ser72Ala/ with EYFP the vicinity of the fluorophore Met153Thr/Val163Ala/ have resulted in variants with Ser175Gly/Thr203Tyr distinct spectral properties Yellow Citrine Ser65Gly/Val68Leu/ Improved photostability and sensitivity 121 (see table and figure), there is Gln69Met/Ser72Ala/ to pH changes compared with EYFP a continuing search for new Thr203Tyr sources of fluorescent proteins, DsRed1 and mutagenized variants especially those that extend Red DsRed1 Discosoma sp. Original red fluorescent protein 114 into the red part of the visible Red DsRed2 Arg2Ala/Lys5Glu/ Reaches maturity faster 115 spectrum111–113. Lys9Thr/Ala105Val/ than DsRed1 DsRed1 from the sea Ile161Thr/Ser187Ala anemone Discosoma sp. was Red DsRed-Express/ Arg2Ala/Lys5Glu/Asn6Asp/ Reaches maturity faster than 122 the first commercially DsRed-T1 Thr21Ser/His41Thr/Asn42Gln/ DsRed variant available red-fluorescentVal44Ala/Cys117Ser/ protein reporter114,115. Thr217Ala Unfortunately, it has several Red mRFP1 33 mutations introduced: Fast maturing, monomeric red 117 limitations113,116. For example, 13 internal to the β-barrel fluorescent protein it functions as an obligate and 20 external to the β-barrel tetramer and matures slowly, HcRed which limits its use for realFar red HcRed Heteractis crispa Far-red fluorescent protein 119 time gene-expression studies. ECFP, enhanced cyan fluorescent protein; EYFP, enhanced yellow fluorescent protein; mRFP1, monomeric red fluorescent protein 1. Although strains of mice that express DsRed1 in neurons have been generated71, attempts to produce lines of transgenic EMBRYONIC STEM (ES) CELLS and mice with constitutive expression of DsRed1, or its derivative DsRed2, have failed42. Several laboratories have directed their efforts towards generating an improved red fluorescent reporter116–118. There is a continuing search for new sources of spectrally-distinct fluorescent proteins111–113,118, and one such reporter that has recently become commercially available is HcRed119. Emission spectra reproduced with permission from Clontech (BD Biosciences) Catalogue (2002).

VOLUME 4 | AUGUST 2003 | 6 1 5

REVIEWS

a

d

b

e

c

f

PLURIPOTENT

Able to give rise to a wide range of, but not all, cell lineages (usually all fetal lineages and a subset of extraembryonic lineages). HAEMATOPOIETIC

Giving rise to the cellular elements of the blood, such as the white blood cells, red blood cells and platelets. MOSAIC

An organism that consists of cells of more than one genotype. The strict definition requires that the genotypically different cells all derive from a single zygote. The term mosaic is also used more broadly to describe any organism comprised of cells of different genotypes. COELENTERATE

Radially symmetrical invertebrates, which include corals, sea anemones and jellyfish.

g

FLUOROPHORE

The core portion of a molecule that is directly responsible for absorbing photons. THERMOSTABLE

Able to withstand moderate heat without the loss of characteristic properties, such as fluorescence. QUANTUM YIELD

The probability of luminescence occurring under given conditions, which is expressed as the ratio of the number of photons emitted to the number absorbed. EMBRYONIC STEM CELLS

(ES cells). Stem cells have the dual capacity to self-replicate and differentiate into several specialized derivatives. ES cells are pluripotent cells that are derived from pre-implantation stage (usually blastocyst) mammalian embryos. Mouse ES cells can be propagated and manipulated in vitro, yet still retain their pluripotency. HEMIZYGOTE

An animal with a transgene insertion on one chromosome of a homologous pair, rather than on each of the two homologous chromosomes (homozygote).

616

Figure 1 | Combinatorial fluorescent-protein-reporter detection in live chimaeras and double transgenics. Wide-field epifluorescence images illustrating dual-reporter visualization in chimeric embryos (a–c) and adult organs (d–f). Anterior view of an embryonic day (E) 7.5 double-tagged chimaera, the epiblast (lower half of embryo) and its derivatives are enhanced cyan fluorescent protein (ECFP)-positive, whereas the extra-embryonic ectoderm, the visceral endoderm and trophoblast are enhanced yellow fluorescent protein (EYFP)-positive (upper half of embryo) (a–c). Dark-field images were taken through an EYFP filter (a), an ECFP filter (b) and, for the double-exposure image, consecutive use of ECFP and EYFP filters (c). Non-invasive multiple-reporter visualization in chimeric adult organs was generated by the aggregation of reportertagged diploid embryos and embryonic stem cells (d–f). All images depict double exposures that were produced by the consecutive use of ECFP and EYFP filters under dark-field epifluorescence. Unlike in the heart (d) and pancreas (e), there seems to be greater interspersion of cells in the liver (f), which possibly reflects greater cell intermingling during the development of this organ. The laser-scanning confocal image (g) shows neurons in the brain of a Thy1-double-transgenic mouse71 that has inherited different sets of fluorescently expressing cells (ECFP versus EYFP) from each of its parents. Panels a–d and f reproduced with permission from REF. 42 © (2002) Biomed Central.

| AUGUST 2003 | VOLUME 4

Because it is not uncommon for mouse mutant phenotypes to be complex, chimaeras that consist of both wild type and mutant cells can be extremely helpful in determining the site of action of a gene product or in understanding the later consequences of a mutation that also affects early development. However, a prerequisite for all chimeric experiments is the incorporation of an independent genetic tag to distinguish mutant and wild-type cells. In the past, markers such as a constitutively expressed lacZ reporter40 or high-copy-number transgene integration41 have been used to mark one of the cellular components in a chimaera, but these could only be detected in fixed tissue. Fluorescentprotein-reporter technology now makes it possible to identify both wild type and mutant components of a chimaera in live embryos or tissues, using spectrally distinct fluorescent proteins that mark each cell type42 (FIG. 1). Fluorescent tagging of cells in chimaeras has been used in a number of studies, but none have fully exploited the potential of dynamic cell imaging in vivo that is offered by these vital reporters. Recently, a fluorescentlytagged chimeric approach has been used to investigate the action of reduced levels of vascular endothelial growth factor A (VEGFA), which has a pivotal role in the first steps of the development of the yolk sac, as well as in the establishment of the cardiovascular system. Embryos that are homozygous for a lacZ-tagged hypomorphic allele of Vegfa die around mid-gestation owing to abnormalities in yolk-sac vasculature and deficiencies in the development of the dorsal aorta. An elegant set of chimeric experiments showed that VEGFA expression in the visceral endoderm is absolutely required for the normal expansion and organization of both the endothelial and haematopoietic lineages in the early sites of vessel and blood formation43. Non-invasive sexing. In one of the first strains of transgenic mice to show constitutive expression of enhanced green fluorescent protein (EGFP), the transgene was inserted on the X chromosome. This X-linked reporter presented the first opportunity to use fluorescence to non-invasively sex embryos 44. Because only female progeny inherit and express paternally derived X-EGFP transgenes, this reporter can be used for the fluorescence-based sexing of pre-implantation-stage embryos more than ten days before an overt morphological difference between the sexes is observed44 (FIG. 2a). X-inactivation in normal, mutant and cloned mice. Females that are HEMIZYGOUS for an X-linked constitutively expressed EGFP reporter show imprinted transgene expression in extra-embryonic tissues, and the transgene is silenced as a result of X-inactivation in 50% of their somatic cells45 (FIG. 2b,c). X-linked EGFP mouse strains have been used to help investigate the actions of genes that are involved in the regulation of genomic imprinting46,47, because they allow the visualization of the status of X-linked gene expression in living embryos.

www.nature.com/reviews/genetics

REVIEWS

a

b

c

with a single null allele of Plxna3. In these experiments, only the unlabelled cells showed the mutant phenotype, whereas the EGFP-expressing neurons showed a wildtype phenotype, which indicates that mutations in Plxna3 act cell autonomously48. Over the past five years there has been an upsurge in the number of successful attempts to clone animals, and an increasing interest in the application of cloning as a method of generating genetically matched stem cells for therapeutic purposes49. One of the few biological insights into this phenomenon used X-linked EGFP mice and investigated whether cloning resets the epigenetic differences between the two X chromosomes in a female somatic nucleus46. Somatic cells were pre-selected on the basis of the X chromosome that was inactivated. As they were alive, they could be used as donors to generate cloned embryos in which X-chromosome activity was monitored. Intriguingly, epigenetic marks are removed and re-established on both X chromosomes during the cloning procedure46. Gene-expression reporters

Figure 2 | An X-linked fluorescent reporter. Wide-field epifluorescence images of a constitutively expressed X-linked enhanced green fluorescent protein (EGFP) transgene44,45. This reporter can be used for the non-invasive sexing of embryos before the appearance of overt sexual dimorphism. Also, because the transgene is subject to X inactivation, as are the endogenous genes on the X chromosome, it can be used for tracking X-inactivation. a | Live preimplantation (blastocyst stage) mouse embryos that were derived from a cross between a male carrying an EGFP transgene on his X-chromosome and a wild-type female. Only female offspring inherit the EGFP transgene and are marked by green fluorescence. b | Midbrain region from live male (left) and female (right) embryonic day (E) 10.5 hemizygous transgenic embryos, both of which carry an X-linked EGFP reporter. All cells in the male are fluorescent compared with ~50% of the cells in the female, owing to random X inactivation (it is not clear which cells express the fluorescent reporter, because of the low magnification that does not provide resolution at the cellular level). c | Newborn offspring from a cross between a female carrying one copy (hemizygous) of an X-linked EGFP transgene and a non-transgenic male. Transgenic male pups (left) have ubiquitous green fluorescence in the skin owing to the presence of a single active X chromosome. Transgenic female pups (right) have mosaic (tortoiseshell) green fluorescence in the skin owing to random inactivation of one X chromosome. Non-transgenic pups (centre) are not visible. Panel c reproduced with permission from REF. 45 © (2001) Wiley.

Recently, X-linked EGFP mice were used to investigate the defects in plexin-A3 mutant mice48. Plexin-A3 is a receptor for semaphorin ligands. Mutants show defective pruning of specific axonal branches in the developing brain. The natural mosaicism that is created by the process of X-inactivation can be used for mutant analysis. In such an experiment, X-linked EGFP mice were crossed with mice that carried a targeted mutation in the X-linked Plxna3 gene, so that one X chromosome contained the EGFP marker and a wild-type allele of Plxna3, whereas the other was null for Plxna3. Because X-inactivation creates a genetic mosaic, EGFP-expressing cells were wild type and unlabelled cells were mutant

NATURE REVIEWS | GENETICS

Mouse lines can be generated with fluorescent-protein reporters that reproduce the expression of specific genes. To do this, a reporter construct that contains the cis-regulatory elements from both the gene of interest and the fluorescent protein is randomly inserted into the genome, or, alternatively, the fluorescent protein is targeted to a specific locus. A reporter construct can be randomly integrated into the mouse genome by the pronuclear injection of DNA into zygotes, or the introduction of DNA into ES cells, and these methods often result in high-levels of reporter-gene expression. However, this approach requires knowledge of the cis-acting regulatory elements of the gene. If minimal information is available about the cis-acting regulatory elements that drive spatiotemporal gene expression, a bacterial artificial chromosome (BAC) transgenic approach can be adopted. In this strategy, a fluorescent-protein reporter is introduced by recombination into the region of interest of a BAC vector. This large construct, which probably contains all necessary regulatory sequences, is then randomly inserted into the mouse genome50–52. A BAC-modification strategy using clones that contain the regulatory elements of the Nkx2.5 gene53 was recently used to drive expression of a fluorescent reporter in the embryonic heart. The resulting spatiotemporal pattern of green fluorescence was similar to that obtained by RNA in situ hybridization. Alternatively, a gene-targeting approach that uses homologous recombination in ES cells can recapitulate the expression of the gene of interest. However, this will usually result in the incorporation of only a single copy of the fluorescent reporter, which might not provide adequate fluorescence intensity, particularly at loci that are not strongly expressed. Gene-targeting strategies can be designed to insert a fluorescent-protein reporter into a locus so as to disrupt the function of the gene product but leave its regulation intact. This knock-in strategy provides a fluorescent marker for mutant cells with disrupted gene function, and allows the identification of

VOLUME 4 | AUGUST 2003 | 6 1 7

REVIEWS

Box 3 | Static-culture systems for time-lapse imaging Similar to most mammals, mouse embryos develop in the relatively inaccessible environment of the maternal reproductive system. Vital-imaging experiments require that protocols be modified to allow normal development to take place in static cultures on the microscope stage. Existing protocols for the culture of whole embryos and individual organs all require a robust temperature-controlled oxygenated environment for the duration of the experimental period. Whole pre-implantation-stage embryos (from fertilization to embryonic day (E) 3.5) can be easily grown in static culture and allow the observation of the intricacies of cleavage, compaction and cavitation to form the blastocyst at E3.5 (see figure)123–125. For examples of whole embryos that have been imaged whilst being cultured in vitro see supplementary movies 1 and 2 online. Post-implantation

Pre-implantation Ovary Oviduct

Uterus

E10.5–E12.5

Implantation Cleavage Ovulation

E2.5

E3.5

E2 E1.5

E4.5

Compaction and cavitation Fertilization

E8.5

Gastrulation and neurulation

Hatching

Organogenesis

Culturing post-implantation mouse embryos is more difficult, and protocols for the static culture of these stages are based on serum-rich media preparations. Methods for culturing early post-implantation embryos between E5.5–E8.5 (REFS 107,108) and E8–E10.5 (REF. 67) in serum-rich media are in use in several laboratories. Later-stage embryos (E11.5–E12.5) can be grown in serum-free conditions126. Also, protocols for the explantation and in vitro culture of parts of embryos, including germ layers from early gastrulating embryos127, are common and can be adapted to work on a microscope stage. At later stages, whole embryos cannot be maintained in culture; however, their increasing size makes them more opaque, so the culture of organ primordia is more effective for imaging experiments. Protocols for the in vitro culture of several organ systems, such as ureteric buds61, lung buds128, limb buds129 and the neural tube130,131, have been developed. Figure reproduced with permission from REF. 141.

cells in which the promoter of interest is active. However, low levels of expression can be a complicating factor for some loci. For example, it has been difficult to recapitulate reporter expression under standard epifluorescent conditions when lacZ reporters are replaced with fluorescent proteins (A.-K.H. and V.E.P., unpublished observations). This is usually because fluorescence is a direct readout of gene activity, and is not detected using a chromogenic signal-amplification reaction (as with lacZ), so fluorescent reporters can be less sensitive. The development of new and brighter fluorescentprotein reporters, for example the use of ‘venus’ 54

in contrast to enhanced yellow fluorescent protein (EYFP)55,56 (BOX 2), should address this issue. Another way to increase the sensitivity is by concentrating the total available fluorescent protein using subcellularly localized reporters (as discussed below). Another limitation of many gene-based reporters is PERDURANCE of the reporter protein. If the protein has a long half-life it might not faithfully report the temporal domain of the regulatory elements that drive its expression. A solution to this problem has been achieved by fusing the fluorescent protein to the degradation domains of other proteins to accelerate degradation57.

PERDURANCE

The ongoing stability and activity of a protein in the cellular environment. EPIBLAST

An embryonic lineage that is derived from the inner-cell mass of the blastocyst, which gives rise to the body of the fetus.

618

0 hours

24 hours

60 hours

84 hours

Figure 3 | Green fluorescence as a marker of branching morphogenesis in the developing kidney. Wide-field epifluorescence images of an in vitro cultured embryonic kidney taken at various time-points during the same experiment. Fluorescence is confined to the ureteric-bud derivatives of the kidney. The mouse embryo carried a HoxB7-EGFP transgene61. The kidney primordium was dissected out of an embryonic day (E) 11.5 embryo and maintained in culture for 84 hours. Reproduced with permission from REF. 61 © (1999) Wiley.

| AUGUST 2003 | VOLUME 4

www.nature.com/reviews/genetics

REVIEWS

Box 4 | Imaging fluorescent proteins using laser-scanning microscopy In many of the examples described a b Single-photon excitation Multi-photon excitation in this review, multiphoton 132 microscopy has been used for vital and deep-tissue imaging. Introduced over a decade ago, this form of laserscanning microscopy (LSM) uses Laser illumination Laser illumination near-infrared light to excite everywhere everywhere fluorochromes and results in excitation only at the plane of focus Excitation only Excitation (region of highest photon flux), at highest photon everywhere flux which eliminates noise that is created by out-of-focus emission photons. This is in contrast to conventional wide-field fluorescence microscopy and confocal LSM (CLSM)133, which use visible light to excite dyes. These techniques cause excitation of fluorescent dyes above and below the plane of focus, which results in thick samples appearing blurred. Confocal microscopes reduce this problem and increase the resolution of the image by using a pinhole in front of the detector to eliminate out of focus photons. In multiphoton microscopy134,135, out of focus photons are never generated, which provides optical sectioning without a pinhole. This is because fluorescence is only produced at the plane of focus. In standard fluorescence microscopy, a single high-energy photon is absorbed by a fluorochrome and a photon of slightly lower energy is emitted when it returns to ground state. For example, for a fluorescent protein such as GFP (panel a), a blue photon (blue arrows) is absorbed and a green photon (green arrows) is emitted. In multiphoton excitation, the single high-energy photon is replaced by the quasi-simultaneous absorption of two (two-photon excitation) or three (three-photon excitation) lower-energy near-infrared photons. So, for GFP, rather than a single blue excitation photon, two near-infrared photons (red arrows) can be used to excite the fluorochrome, which leads to the emission of a green photon (panel b). Fluorescence is generated only at the point of focus, at the highest photon flux, because multiple photons must be absorbed simultaneously. Note that the emission of photons (green arrows) is actually in all directions, although only those that can be collected by the objective lens are shown here. Multiphoton excitation has the advantage of less overall PHOTOBLEACHING and less damage to the tissue by repeated illumination because, when creating a three-dimensional stack of images along the z-axis, each focal plane is excited only once. By contrast, in confocal microscopy all focal planes are exposed to excitation light each time an optical plane is collected. Confocal scanning can lead to extensive photobleaching and a greater chance of toxicity if multiple z-stacks are taken repeatedly over time. The use of near-infrared photons for excitation has a number of benefits. Near-infrared light penetrates deeper into the sample than visible light, which allows the imaging of deeper structures than in confocal microscopy. Near-infrared light is also less harmful to tissue in general, because few endogenous molecules absorb at near-infrared wavelengths, which greatly limits the potential for tissue damage. This has been shown for pre-implantation hamster embryos136. Hamster embryos that were imaged using confocal microscopy did not undergo normal cleavage, whereas embryos imaged using multiphoton microscopy not only developed to the blastocyst stage, but could be implanted into a pseudo-pregnant host for further development and, at least in one case, the imaged embryo gave rise to a healthy adult hamster.

PHOTOBLEACHING

The irreversible destruction of a fluorophore that is under illumination.

NATURE REVIEWS | GENETICS

Such reporters might provide a better temporal representation of promoter activity; however, there could be a reduction in sensitivity if insufficient protein accumulates to ensure robust fluorescence. At present, few reports describe the use of destabilized GFP-variant reporters in mice. One example is the successful use of regions upstream of the Axin2 gene to direct the tissue-specific expression of d2EGFP (an EGFP fusion with a two hour half-life) in transgenic mouse embryos58. Both targeted and randomly integrated (injectionand ES-cell-based) transgenic reporters have been used to generate strains of mice that express fluorescent proteins in a lineage-specific fashion. Next, we describe some of the most recent studies that have used transgenic mice for multidimensional analyses of embryonic and neuronal development.

Imaging primordial-germ-cell migration in embryos. One of the first reports that exploited a fluorescentprotein reporter to follow the movement of cells in a living mouse embryo (see BOX 3 for details of culturing embryos) used a transgenic mouse line with EGFP under the regulation of the Oct4 promoter (Oct4-EGFP). This promoter is first active in the cells of the inner-cell mass of the blastocyst and the EPIBLAST of the pregastrulation embryo, and is subsequently restricted to the primordial germ cells (PGCs) of the post-implantation embryo59,60. In this study, fluorescently labelled PGCs, which are the founder cells of the gametes, were followed using time-lapse microscopy of living postimplantation-stage embryos. By imaging living embryos, as well as investigating the shapes and positions of PGCs in fixed tissues, the site of origin of PGCs was determined and these cells were shown to

VOLUME 4 | AUGUST 2003 | 6 1 9

REVIEWS

a

loop between the ureteric-bud epithelium and the stromal mesenchyme, and showed that vitamin A controls epithelial–mesenchymal interactions through the receptor tyrosine kinase Ret expression63. Time-lapse imaging and three-dimensional (3D) reconstructions of HoxB7-EGFP transgene expression were used to show that ureter maturation depends on formation of the ‘trigonal wedge’ — an epithelial outgrowth from the base of the Wolffian duct — and that the urogenital malformations observed in Rara–/–Rarb–/– mice probably result from the failure of this process64.

b

H2B

EGFP

c

EGFP

d

myr

EGFP

tau

EGFP

Figure 4 | Fluorescent fusion proteins as markers of subcellular compartments. Three-dimensional (3D) reconstructions of confocal laser-scanning-microscopy serial sections of transgenic mouse embryonic stem (ES) cells that express various subcellularly-localized fluorescent proteins. a | Native enhanced green fluorescent protein (EGFP), which labels the entire cell. b | A fusion of EGFP with human histone H2B, which labels chromatin. c | A fusion of EGFP with an N-terminally engineered myristoylation sequence (myr), which labels cell membranes. d | A fusion of EGFP with the bovine microtubule-binding protein tau, which labels microtubules (supplementary movies 3–6 online show 3D reconstructions of these images).

move by active locomotion from the primitive streak (a region in the posterior epiblast) into the adjacent endoderm, from where they migrate to the gonad. Imaging branching morphogenesis. Urogenital abnormalities are common in human infants, and are often caused by malformations of the kidneys or irregular connections between the ureter and the bladder. Mice present an excellent model for urogenital development, and kidney primordia can be cultured in vitro (BOX 3) where they recapitulate branching morphogenesis. Fluorescent labelling of various cell types in the developing kidney allows dynamic imaging of these events61,62 (FIG. 3). A HoxB7-EGFP transgene61 that fluorescently labels the developing ureteric bud was used in a series of experiments to investigate the phenotype of Rara –/–Rarb –/– mice, which, as a result of impaired retinoid (vitamin A) signalling, develop syndromic urogenital malformations that are similar to those seen in humans. This led to the identification of a signalling

620

| AUGUST 2003 | VOLUME 4

Quantifying blood-flow dynamics in live embryos. Traditional protocols for the in vitro development of post-implantation mouse embryos involve rotating cultures65,66. These techniques have been recently adapted to enable fluorescence time-lapse imaging of embryos up to mid-gestation, which are maintained in static culture on the microscope stage67 (BOX 3). In mouse embryos, blood cells, along with endothelial cells, first form at embryonic day (E)7.5 in the yolk sac, which is an extra-embryonic structure that surrounds the embryo. This site is the sole source for bloodcell production until definitive blood cells are produced in the embryo itself several days later. The heart begins beating at E8.5 and a notable remodelling of the vasculature, especially in the yolk sac, takes place over the next 24 hours. Using a transgenic mouse line that expresses GFP under the control of the embryonic globin (ε-globin) promoter68, the formation and flow of newly formed blood cells in live intact mouse embryos have been studied by confocal67 (single-photon) and multiphoton microscopy (BOX 4). Because the flow of blood is dynamic, such studies would have been impossible in fixed samples. Brightly labelled spherical blood cells first appear at E7.5 (REF. 68) and the expansion of these cells, which are initially located in the blood islands, has been followed through time-lapse microscopy67. A large number of mutant mice have various defects in the development of the cardiovascular system. Through genetic manipulation, many genes have been found to have a role in early cardiovascular development. As blood and endothelial-cell development occur in concert with blood flow, there is need for a model system to study the interplay between physical and genetic forces in the early cardiovascular system in mice. Using laser-scanning microscopy and fluorescent reporters, accurate measurements of blood-cell velocities can be determined in early embryos as soon as circulation begins (E. A. Jones, M. H. Baron, S.E.F and M.E.D., unpublished observations). By measuring velocity, and other important parameters that have been made accessible using GFP-marked blood cells, the forces between blood cells and surrounding cells can be determined in both wild type and mutant backgrounds. This methodology represents a significant advance in studying quantitative blood flow in mammalian embryos. Other techniques that are used to study blood flow in mouse embryos, such as Doppler ultrasound69 and Doppler optical-coherence tomography70, do not offer enough spatial or temporal resolution to study microcirculation

www.nature.com/reviews/genetics

REVIEWS

Box 5 | Simultaneous imaging of several fluorescent proteins

Relative intensity

When several genetically-encoded fluorescent Acquire a λ-stack proteins are combined in the same imaging λ experiment, more information can be x obtained from the same image session and the relationship between different cells or different processes can be discovered. y However, many fluorescent-protein variants have similar excitation and emission Requires special hardware, such as the Zeiss LSM 510 META properties, so it can be difficult to unambiguously separate the signals from Select reference spectra from the spectral database each reporter. In fluorescence microscopy, emission signals are usually separated using 1.0 glass filters, but in situations of significant 0.8 spectral overlap, glass filters are ineffective. In 0.6 much the same way, excitation curves are also 0.4 highly overlapping, which makes it difficult to choose excitation wavelengths that would 0.2 excite only single variants. So, spectral 0.0 cross-talk between most genetically 460 480 500 520 540 560 580 600 620 encoded fluorescent proteins is inevitable. Emission wavelength (nm) However, there are mathematical Apply linear unmixing algorithms to create cross-talk-free multichannel images approaches that can be used to separate spatially and spectrally overlapping signals, the most effective of which is linear unmixing137–139.A series of images are collected from the sample at different wavelengths using a specialized detector to generate a λ-stack. Linear unmixing makes use of the fact that variations in the intensity of fluorescence spectra are specific to each fluorochrome. Therefore, reference spectra can be selected from a special database. This information is used to convert the data from the λ-stack into a multichannel image, with channels that contain the signal from each fluorescent component. This approach has been used with great success to separate spectrally and spatially overlapping fluorescent-protein signals and can be used by either sampling emission spectra or excitation spectra.

at the onset of development, and can only provide information about large vessels or blood flow through the heart. Methods that are based on laser-scanning microscopy (BOX 4) are especially attractive as they make it possible to combine quantitative blood-flow analysis with the study of other cell types simultaneously, by using other fluorescent-protein variants or vital dyes Vital imaging of neurons in mice. A growing number of studies have used transgenic mice that express different fluorescent proteins42 (BOX 2), in particular those under the control of the Thy1 promoter, which is active in neurons71. Interestingly, specific lines express fluorescent proteins in different subsets of neurons (expression mosaics in contrast to genetic mosaics), and ‘designer’ mice that have different combinations of neurons labelled with spectral variants can be generated by crossing specific strains (FIG. 1g). These mice are frequently used because many populations of cells can be studied using different spectral combinations. Mice that express fluorescent proteins that are driven by the Thy1 promoter have been used to explore the formation and plasticity of synapses. For this analysis, the same neuromuscular junction was repeatedly imaged every 24 hours using multiphoton microscopy

NATURE REVIEWS | GENETICS

to visualize dynamic changes in innervation and synapse elimination. These studies take full advantage of mice that express genetically-encoded fluorescent proteins, and have given insights into the guidance of neurons during injury and the mechanisms of neuronal branching and synapse maturation. For example, axons that are damaged by crushing will make new synapses on the same cells that they originally innervated, but if the nerve is cut, this specificity is lost72. Time-lapse multiphoton imaging has also provided new information about axon-branch trimming at the synapse. Although single muscle cells are originally innervated by several motor axons, these connections are refined as the synapse matures and many axons are retracted, leaving, in most cases, a single axon. By watching the same synapses over time, it is clear that there is no spatial or temporal bias to branch elimination, which indicates that local competitive interactions are involved in the elimination of one branch and the stabilization of the other73. Using neurons that are labelled with different fluorescent-protein colour variants, it is clear that the retraction of one axon is accompanied by the expansion of another, which explains the increase in synaptic strength that is shown by the ‘winning’ axon74. Recently, Thy-1 fluorescent reporter constructs in mice have been used to examine synaptic plasticity in the cortex75,76. In these studies, the brain of a living animal is directly imaged using multiphoton microscopy (BOX 4). The structures of dendritic arbors were imaged over the course of days and even months, which shows that there is considerable long-term stability of synaptic connections in adult mice, although there were dynamic changes in the synapses of younger mice. These results indicate a mechanism for long-term memory and experience. Such approaches have not only been used in brain imaging77,78, but also for the in vivo imaging of tumour formation79. Fusion proteins allow subcellular resolution. A popular application of fluorescent-protein reporters in cell biology is their incorporation into fusion proteins, which can be used as vital labels for subcellular compartments80 or for tracking protein dynamics81. Also, to provide subcellular resolution, fluorescent-protein fusions can help to concentrate the reporter, which increases its sensitivity and allows visualization of lower levels of reporter protein. However, at present, the use of subcellularlylocalized fluorescent reporters is limited to a handful of published reports that describe mice with constitutive and lineage-restricted expression of a bovine tau–EGFP fusion82–84, constitutive expression of a GPI-anchored EGFP fusion85 and expression of EGFP with nuclearlocalization sequences86 (FIG. 4). Subcellularly localized fluorescent-protein reporters provide superlative resolution compared with other genetically-encoded reporters and vital stains. Moreover, fusion of a fluorescent protein to a nucleosomal component, such as a histone, offers a higher level of resolution compared with a standard nuclear-localization sequence. When samples from transgenic embryos that express a nuclear-localized fluorescent reporter are imaged using confocal or multiphoton microscopy (BOX 4), individual

VOLUME 4 | AUGUST 2003 | 6 2 1

REVIEWS cells can be unequivocally identified in a group of cells, and cells in mitosis can easily be discerned from cells in interphase, as can those undergoing cell death. Nuclearlocalized reporters also allow nuclear morphology to be visualized, and as this is specific to different cell types it can be used to identify cell lineages. This raises the possibility of confocal or multiphoton ‘histology’ of live specimens and paves the way for the development of high-resolution multidimensional histological atlases. Furthermore, subcellularly localized fluorescent proteins are more amenable to fixation and preservation of the fluorescent signal (A.-K.H. and V. E. P., unpublished observations). Aa

Ab

Ac

Subcellularly localized reporters allow the visualization of dynamic subcellular events in an organismal context. For example, cytoskeletally localized fluorescent proteins (such as actin, tubulin and tau fusions) allow the dynamic visualization of changes in cell shape. By contrast, nuclear-localized fluorescent proteins (such as those with a nuclear-localization sequence or fusions to nuclear proteins such as histones), allow the unequivocal identification and ‘tracking’ of individual cells in a field/population of cells that all express the same fluorescent-protein reporter. Nuclear-localized fluorescent proteins also allow the vital imaging of active chromatin, such that interphase nuclei can be distinguished from nuclei in mitosis and from cellular debris (FIG. 4b). Histone–GFP fusion proteins were first reported in cultured cells, in which their use is now widespread87. Histone fusions have also been used for high-resolution imaging in whole-organism systems, including C. elegans 88,89, Drosophila90,91 and zebrafish92,93. Looking to the future

Ad

Ae

B

Af

0 µm

C

20 µm 40 µm 60 µm 80 µm 100 µm

Figure 5 | Towards a high-resolution multidimensional atlas of a living mouse. Multiphoton microscopy images of a live embryonic day (E) 6.5 mouse embryo constitutively expressing a histone–enhanced green fluorescent protein (EGFP) fusion protein and maintained in an in vitro culture. This reporter labels active chromatin and facilitates the high-resolution visualization of cell nuclei including mitotic cells and fragmenting nuclei of dying cells. Every tenth section is depicted from a sequential optical z-stack comprising 60 optical sections taken at 2 µm intervals of the distal end of a living mouse embryo, starting at the surface (Aa) and penetrating into the specimen (Ab–f). The embryo is cup shaped and comprises two cell layers: an outer layer of visceral endoderm cells surrounding an inner layer of epiblast cells. Deep to superficial (B) and superficial to deep (C) colour-coded three-dimensional (3D) reconstructions generated from the z-stack encompassing the entire image series. The colours of the scale bar denote fluorescence at different depths within the embryo (red being outermost, and blue being deepest). This is one way to present 3D data in two-dimensions (supplementary movies 7–9 online show z-stacks and 3D images from live embryos). Panel c reproduced with permission from REF. 140.

622

| AUGUST 2003 | VOLUME 4

The continued evolution of genetically-encoded fluorescent-protein reporters, coupled with advances in opticalimaging modalities, will make reporter-expressing mice an essential tool for the multidimensional analysis and understanding of biological processes in wild type, mutant and pathological contexts. Technological advances. In the field of fluorescentprotein-reporter development, new spectral variants that mature faster with increased fluorescence, and improved excitation and emission spectra, will continue to become available as will methodologies that allow the separation of proteins with similar spectral profiles (BOX 5). These improvements will facilitate the increased use of technicolour mice that express multiple fluorescent reporters. So far, few published reports illustrate combinatorial imaging with dual fluorescentprotein-reporter detection in individual embryos42 or adult mice42,71, but there are already several lines of mice that have been specifically designed for anticipated dual-tagging experiments29. Cross-talk between fluorescent proteins has limited multicolour experiments to specific combinations (for example, enhanced cyan fluorescent protein (ECFP) and EYFP). However, recent advances in imaging technology that use both excitation and emission fingerprinting to separate closely overlapping markers, such as linear unmixing (BOX 5), allow signal from different fluorescent reporters to be resolved. This opens the door to increased multiple fluorochrome experiments in the future. As imaging technology improves, there is a general trend towards making imaging hardware more portable, increasing the types of cell that can be imaged in situ. The progress in fibre-optic confocal imaging (FOCI) and two-photon fibre-optic imaging, together with recent advances in implantable devices78, such as a miniature head-mounted multiphoton microscope94, raises the possibility of monitoring body functions

www.nature.com/reviews/genetics

REVIEWS

MUSHROOM BODIES

The region of the Drosophila brain that is required for olfactory learning and memory.

through internalized miniature surveillance cameras. These technologies combine the benefits of highconstrast high-resolution light microscopy with the intravital advantages that are offered by other less invasive imaging technologies, such as magnetic-resonance imaging (MRI ) (see links in online links box). Also, to make progress in imaging techniques, the development of reporter proteins with special features, including photoactivatable proteins and biochemical sensors, is anticipated to expand the use of fluorescent mice. Recently, fluorescent reporters have emerged that can be modulated photochemically. When irradiated with light of the appropriate wavelength, these proteins fluoresce at a specific wavelength, so irradiated cells and their progeny can be visualized. These fluorescent highlighters include ‘PA-EGFP’, which is a GFP variant95, and ‘kaede’96, which was cloned from Trachyphyllia geffroyi. There are no reports, at present, of the use of fluorescent highlighters in mice; however, this class of reporter is particularly attractive because it could be used in fatemapping experiments in which photo-activated reporters could label cells non-invasively and be followed in situ. One ‘final frontier’ in real-time imaging of live specimens is a reporter acting as a direct read-out of a physiological process. Several functional fluorescentprotein-based biochemical-sensor reporters have been developed97–100, which include sensors for heterotrimeric G-protein activity101, phosphoinositide102 and calcium103,104 signalling. Calcium sensors (often called camaroos) have recently been used in Drosophila to measure signalling in the MUSHROOM BODIES of the brain105 and to decipher the logic of odour perception through the use of two-photon microscopy (BOX 4) for the analysis of odour-evoked patterns of neural activity106. It is only a matter of time until biological-sensor reporters are used in mice. Applications. Developments in fluorescent-reporter technologies should lead to the widespread incorporation of technicolour transgenics into functional genomic approaches in the mouse, including phenotype-driven screens (such as N-ethyl-N-nitrosourea (ENU) mutagenesis2,3,5,6) or expression-driven screens (such as genetrap mutagenesis4). This should help focus screens by making it easier to identify mouse mutants that disrupt

1.

2.

3.

4.

5.

6.

Anderson, K. V. & Ingham, P. W. The transformation of the model organism: a decade of developmental genetics. Nature Genet. 33 (Suppl.), 285–293 (2003). Justice, M. J., Noveroske, J. K., Weber, J. S., Zheng, B. & Bradley, A. Mouse ENU mutagenesis. Hum. Mol. Genet. 8, 1955–1963 (1999). Anderson, K. V. Finding the genes that direct mammalian development: ENU mutagenesis in the mouse. Trends Genet. 16, 99–102 (2000). Stanford, W. L., Cohn, J. B. & Cordes, S. P. Gene-trap mutagenesis: past, present and beyond. Nature Rev. Genet. 2, 756–768 (2001). Nolan, P. M. et al. A systematic, genome-wide, phenotypedriven mutagenesis programme for gene function studies in the mouse. Nature Genet. 25, 440–443 (2000). Brown, S. D. & Balling, R. Systematic approaches to mouse mutagenesis. Curr. Opin Genet. Dev. 11, 268–273 (2001).

NATURE REVIEWS | GENETICS

7.

8.

9. 10.

11. 12.

13.

specific biological processes, as well as increasing the rate of mutant characterization. Fluorescent-protein reporters will also allow mouse development to be studied in more detail. This approach is already helping to investigate the genetic control that underlies the patterning and morphogenetic movements of early post-implantation development. This stage is characterized by an increase in cell proliferation and during gastrulation, cell movements occur that are integral to the establishment of the mammalian body plan. A wealth of available mutations perturb early development, and most lead to the untimely death of the embryo. However, little is known about the dynamics of the onset and propagation of many early mutant phenotypes, as much of the information comes from ‘snapshots’ of embryos. Early post-implantation embryos are small, and can easily be cultured (BOX 3)107,108, so they represent attractive subjects for vital imaging. These features, coupled with the availability of high-resolution vital markers, facilitates the multidimensional acquisition of data, in which cell position, movement, proliferation and death can be catalogued both spatially through 3D reconstructions of z-stacks (FIG. 5) and temporally by acquiring a time-lapse series of z-stacks. It is possible to imagine a future in which high-resolution four-dimensional atlases become commonplace and are used for high-throughput evaluations of mouse-mutant phenotypes. Conclusion

The introduction of genetically-encoded vital fluorescent reporters in mice, together with the development of robust imaging protocols, now provides the essential bridge between molecular genetic analysis and highresolution multidimensional phenotypic analysis. With so many possibilities, this is an exciting time in fluorescence-based vital imaging of healthy and diseased mice. Advances in optical-imaging techniques have provided enhanced resolution for the study of cellular and even subcellular events, improved depth penetration for imaging events deep within biological specimens, and better spectral flexibility and separation. The upsurge in the availability and use of transgenic mice that carry a technicolour assortment of jellyfish genes means that the future of mammalian biology as modelled in mice is not only bright, it is extremely colourful.

Hrabe de Angelis, M. H. et al. Genome-wide, large-scale production of mutant mice by ENU mutagenesis. Nature Genet. 25, 444–447 (2000). Nagy, A., Perrimon, N., Sandmeyer, S. & Plasterk, R. Tailoring the genome: the power of genetic approaches. Nature Genet. 33 (Suppl), 276–284 (2003). Le, Y. & Sauer, B. Conditional gene knockout using Cre recombinase. Mol. Biotechnol. 17, 269–275 (2001). Yu, Y. & Bradley, A. Engineering chromosomal rearrangements in mice. Nature Rev. Genet. 2, 780–790 (2001). Lewandoski, M. Conditional control of gene expression in the mouse. Nature Rev. Genet. 2, 743–755 (2001). Jacinto, A. et al. Dynamic analysis of actin cable function during Drosophila dorsal closure. Curr. Biol. 12, 1245–1250 (2002). Sepp, K. J. & Auld, V. J. RhoA and Rac1 GTPases mediate the dynamic rearrangement of actin in peripheral glia. Development 130, 1825–1835 (2003).

14. Koster, R. W. & Fraser, S. E. Direct imaging of in vivo neuronal migration in the developing cerebellum. Curr. Biol. 11, 1858–1863 (2001). 15. Weinstein, B. Vascular cell biology in vivo: a new piscine paradigm? Trends Cell Biol. 12, 439–445 (2002). 16. Ritter, D. A., Bhatt, D. H. & Fetcho, J. R. In vivo imaging of zebrafish reveals differences in the spinal networks for escape and swimming movements. J. Neurosci. 21, 8956–8965 (2001). 17. O’Malley, D. M., Zhou, Q. & Gahtan, E. Probing neural circuits in the zebrafish: a suite of optical techniques. Methods 30, 49–63 (2003). 18. Goring, D. R., Rossant, J., Clapoff, S., Breitman, M. L. & Tsui, L. C. In situ detection of β-galactosidase in lenses of transgenic mice with a γ-crystallin/lacZ gene. Science 235, 456–458 (1987). 19. Cui, C., Wani, M. A., Wight, D., Kopchick, J. & Stambrook, P. J. Reporter genes in transgenic mice. Transgenic Res. 3, 182–194 (1994).

VOLUME 4 | AUGUST 2003 | 6 2 3

REVIEWS 20. Prasher, D. C., Eckenrode, V. K., Ward, W. W., Prendergast, F. G. & Cormier, M. J. Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111, 229–233 (1992). This paper describes the initial cloning and sequencing of wtGFP. 21. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W. & Prasher, D. C. Green fluorescent protein as a marker for gene expression. Science 263, 802–805 (1994). The first report to detail the use of a genetically encoded autofluorescent protein (GFP) in a heterologous system (Caenorhabditis elegans). 22. Okabe, M., Ikawa, M., Kominami, K., Nakanishi, T. & Nishimune, Y. ‘Green mice’ as a source of ubiquitous green cells. FEBS Lett. 407, 313–319 (1997). The first report of the successful generation of injection-based transgenic mice engineered to express readily detectable EGFP. 23. Hadjantonakis, A. K., Gertsenstein, M., Ikawa, M., Okabe, M. & Nagy, A. Generating green fluorescent mice by germline transmission of green fluorescent ES cells. Mech. Dev. 76, 79–90 (1998). The first report of the successful generation of ES-cell-mediated transgenic mice engineered to express readily detectable EGFP. 24. Gagneten, S., Le, Y., Miller, J. & Sauer, B. Brief expression of a GFP cre fusion gene in embryonic stem cells allows rapid retrieval of site-specific genomic deletions. Nucleic Acids Res. 25, 3326–3331 (1997). 25. Horie, K. et al. Efficient chromosomal transposition of a Tc1/mariner-like transposon Sleeping Beauty in mice. Proc. Natl Acad. Sci. USA 98, 9191–9196 (2001). 26. Niwa, H., Yamamura, K. & Miyazaki, J. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108, 193–199 (1991). 27. Zambrowicz, B. P. et al. Disruption of overlapping transcripts in the ROSA β-geo 26 gene trap strain leads to widespread expression of β-galactosidase in mouse embryos and hematopoietic cells. Proc. Natl Acad. Sci. USA 94, 3789–3794 (1997). 28. Novak, A., Guo, C., Yang, W., Nagy, A. & Lobe, C. G. Z/EG, a double reporter mouse line that expresses enhanced green fluorescent protein upon Cre-mediated excision. Genesis 28, 147–155 (2000). 29. Srinivas, S. et al. Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev. Biol. 1, 4 (2001). 30. Kinder, S. J. et al. The orderly allocation of mesodermal cells to the extraembryonic structures. and the anteroposterior axis during gastrulation of the mouse embryo. Development 126, 4691–4701 (1999). 31. Kinder, S. J. et al. The organizer of the mouse gastrula is composed of a dynamic population of progenitor cells for the axial mesoderm. Development 128, 3623–3634 (2001). 32. Cambray, N. & Wilson, V. Axial progenitors with extensive potency are localised to the mouse chordoneural hinge. Development 129, 4855–4866 (2002). 33. Hadjantonakis, A. K. & Nagy, A. FACS for the isolation of individual cells from transgenic mice harboring a fluorescent protein reporter. Genesis 27, 95–98 (2000). 34. Tanaka, S., Kunath, T., Hadjantonakis, A. K., Nagy, A. & Rossant, J. Promotion of trophoblast stem cell proliferation by FGF4. Science 282, 2072–2075 (1998). 35. Faust, N., Varas, F., Kelly, L. M., Heck, S. & Graf, T. Insertion of enhanced green fluorescent protein into the lysozyme gene creates mice with green fluorescent granulocytes and macrophages. Blood 96, 719–726 (2000). 36. Wang, S. et al. Isolation of neuronal precursors by sorting embryonic forebrain transfected with GFP regulated by the T-α1 tubulin promoter. Nature Biotechnol. 16, 196–201 (1998). 37. Tropepe, V. et al. Direct neural fate specification from embryonic stem cells: a primitive mammalian neural stem cell stage acquired through a default mechanism. Neuron 30, 65–78 (2001). 38. Roy, N. S. et al. Promoter-targeted selection and isolation of neural progenitor cells from the adult human ventricular zone. J. Neurosci. Res. 59, 321–331 (2000). 39. Rossant, J. & Spence, A. Chimeras and mosaics in mouse mutant analysis. Trends Genet. 14, 358–363 (1998). 40. Friedrich, G. & Soriano, P. Promoter traps in embryonic stem cells: a genetic screen to identify and mutate developmental genes in mice. Genes Dev. 5, 1513–1523 (1991). 41. Lo, C. W., Coulling, M. & Kirby, C. Tracking of mouse cell lineage using microinjected DNA sequences: analyses using genomic Southern blotting and tissue-section in situ hybridizations. Differentiation 35, 37–44 (1987). 42. Hadjantonakis, A. K., Macmaster, S. & Nagy, A. Embryonic stem cells and mice expressing different GFP variants for multiple non-invasive reporter usage within a single animal. BMC Biotechnol. 2, 11 (2002). This paper describes the generation of transgenic ES cells and mice that express readily detectable ECFP and EYFP, and shows combinatorial reporter visualization in ES cells, embryos and adult organs.

624

| AUGUST 2003 | VOLUME 4

43. Damert, A., Miquerol, L., Gertsenstein, M., Risau, W. & Nagy, A. Insufficient VEGFA activity in yolk sac endoderm compromises haematopoietic and endothelial differentiation. Development 129, 1881–1892 (2002). 44. Hadjantonakis, A. K., Gertsenstein, M., Ikawa, M., Okabe, M. & Nagy, A. Non-invasive sexing of preimplantation stage mammalian embryos. Nature Genet. 19, 220–222 (1998). 45. Hadjantonakis, A. K., Cox, L. L., Tam, P. P. & Nagy, A. An X-linked GFP transgene reveals unexpected paternal X-chromosome activity in trophoblastic giant cells of the mouse placenta. Genesis 29, 133–140 (2001). 46. Eggan, K. et al. X-chromosome inactivation in cloned mouse embryos. Science 290, 1578–1581 (2000). 47. Wang, J. et al. Imprinted X inactivation maintained by a mouse Polycomb group gene. Nature Genet. 28, 371–375 (2001). 48. Bagri, A., Cheng, H. J., Yaron, A., Pleasure, S. J. & Tessier-Lavigne, M. Stereotyped pruning of long hippocampal axon branches triggered by retraction inducers of the semaphorin family. Cell 113, 285–299 (2003). 49. Hadjantonakis, A. K. & Papaioannou, V. E. Can mammalian cloning combined with embryonic stem cell technologies be used to treat human diseases? Genome Biol. 3, 1023 (2002). 50. Yang, X. W., Wynder, C., Doughty, M. L. & Heintz, N. BAC-mediated gene-dosage analysis reveals a role for Zipro1 (Ru49/Zfp38) in progenitor cell proliferation in cerebellum and skin. Nature Genet. 22, 327–335 (1999). 51. Heintz, N. BAC to the future: the use of bac transgenic mice for neuroscience research. Nature Rev. Neurosci. 2, 861–870 (2001). 52. Gong, S., Yang, X. W., Li, C. & Heintz, N. Highly efficient modification of bacterial artificial chromosomes (BACs) using novel shuttle vectors containing the R6Kγ origin of replication. Genome Res. 12, 1992–1998 (2002). 53. Hidaka, K. et al. Chamber-specific differentiation of Nkx2.5-positive cardiac precursor cells from murine embryonic stem cells. FASEB J. 17, 740–742 (2003). 54. Nagai, T. et al. A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nature Biotechnol. 20, 87–90 (2002). 55. Cormack, B. P., Valdivia, R. H. & Falkow, S. FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173, 33–38 (1996). 56. Sawano, A. & Miyawaki, A. Directed evolution of green fluorescent protein by a new versatile PCR strategy for site-directed and semi-random mutagenesis. Nucleic Acids Res. 28, 78 (2000). 57. Li, X. et al. Generation of destabilized green fluorescent protein as a transcription reporter. J. Biol. Chem. 273, 34970–34975 (1998). 58. Jho, E. H. et al. Wnt/β-catenin/Tcf signaling induces the transcription of Axin2, a negative regulator of the signaling pathway. Mol. Cell Biol. 22, 1172–1183 (2002). 59. Anderson, R. et al. Mouse primordial germ cells lacking β1 integrins enter the germline but fail to migrate normally to the gonads. Development 126, 1655–1664 (1999). 60. Anderson, R., Copeland, T. K., Scholer, H., Heasman, J. & Wylie, C. The onset of germ cell migration in the mouse embryo. Mech. Dev. 91, 61–68 (2000). This is one of the first reports to use time-lapse imaging of EGFP in a live mouse embryo. 61. Srinivas, S. et al. Expression of green fluorescent protein in the ureteric bud of transgenic mice: a new tool for the analysis of ureteric bud morphogenesis. Dev. Genet. 24, 241–251 (1999). This is one of the first reports to use time-lapse imaging of EGFP in a live organ explant. 62. Imai, E. et al. Glowing podocytes in living mouse: transgenic mouse carrying a podocyte-specific promoter. Exp. Nephrol. 7, 63–66 (1999). 63. Batourina, E. et al. Vitamin A controls epithelial/mesenchymal interactions through Ret expression. Nature Genet. 27, 74–78 (2001). 64. Batourina, E. et al. Distal ureter morphogenesis depends on epithelial cell remodeling mediated by vitamin A and Ret. Nature Genet. 32, 109–115 (2002). 65. Sadler, T. W. & New, D. A. Culture of mouse embryos during neurulation. J. Embryol. Exp. Morphol. 66, 109–116 (1981). 66. Sturm, K. A. & Tam, P. P. L. in Methods in Enzymology Vol. 225 164–190 (Academic Press, San Diego, 1993). 67. Jones, E. A. et al. Dynamic in vivo imaging of postimplantation mammalian embryos using whole embryo culture. Genesis 34, 228–235 (2002). This paper reports the development of a static-culture system for post-implantation mouse embryos that is specifically designed for use on a microscope stage. 68. Dyer, M. A., Farrington, S. M., Mohn, D., Munday, J. R. & Baron, M. H. Indian hedgehog activates hematopoiesis and vasculogenesis and can respecify prospective neurectodermal cell fate in the mouse embryo. Development 128, 1717–1730 (2001).

69. Turnbull, D. H. Ultrasound backscatter microscopy of mouse embryos. Methods Mol. Biol. 135, 235–243 (2000). 70. Schenk, J. O. & Brezinski, M. E. Ultrasound induced improvement in optical coherence tomography (OCT) resolution. Proc. Natl Acad. Sci. USA 99, 9761–9764 (2002). 71. Feng, G. et al. Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron 28, 41–51 (2000). This paper describes the generation of a series of transgenic mice that express readily detectable ECFP, EGFP, EYFP or DsRed1 under the control of the Thy1 promoter, and also shows combinatorial reporter visualization in the postnatal brain. 72. Nguyen, Q. T., Sanes, J. R. & Lichtman, J. W. Pre-existing pathways promote precise projection patterns. Nature Neurosci. 5, 861–867 (2002). 73. Keller-Peck, C. R. et al. Asynchronous synapse elimination in neonatal motor units: studies using GFP transgenic mice. Neuron 31, 381–394 (2001). 74. Walsh, M. K. & Lichtman, J. W. In vivo time-lapse imaging of synaptic takeover associated with naturally occurring synapse elimination. Neuron 37, 67–73 (2003). 75. Grutzendler, J., Tsai, J. & Gan, W. B. Rapid labeling of neuronal populations by ballistic delivery of fluorescent dyes. Methods 30, 79–85 (2003). 76. Trachtenberg, J. T. et al. Long-term in vivo imaging of experience-dependent synaptic plasticity in adult cortex. Nature 420, 788–794 (2002). 77. Helmchen, F. & Denk, W. New developments in multiphoton microscopy. Curr. Opin Neurobiol. 12, 593–601 (2002). 78. Helmchen, F. Miniaturization of fluorescence microscopes using fibre optics. Exp. Physiol. 87, 737–745 (2002). This review shows the potential of miniature fluorescence microscopes. 79. Jain, R. K., Munn, L. L. & Fukumura, D. Dissecting tumour pathophysiology using intravital microscopy. Nature Rev. Cancer 2, 266–276 (2002). 80. Tsien, R. Y. & Miyawaki, A. Seeing the machinery of live cells. Science 280, 1954–1955 (1998). 81. Lippincott-Schwartz, J., Snapp, E. & Kenworthy, A. Studying protein dynamics in living cells. Nature Rev. Mol. Cell Biol. 2, 444–456 (2001). 82. Rodriguez, I., Feinstein, P. & Mombaerts, P. Variable patterns of axonal projections of sensory neurons in the mouse vomeronasal system. Cell 97, 199–208 (1999). This is one of the first reports to use gene targeting to deliver an autofluorescent protein reporter to a locus of interest. It is also one of the first studies to use an autofluorescent protein fusion (microtubule binding protein tau–EGFP) in a mouse. 83. Pratt, T., Sharp, L., Nichols, J., Price, D. J. & Mason, J. O. Embryonic stem cells and transgenic mice ubiquitously expressing a tau-tagged green fluorescent protein. Dev. Biol. 228, 19–28 (2000). 84. Potter, S. M. et al. Structure and emergence of specific olfactory glomeruli in the mouse. J. Neurosci. 21, 9713–9723 (2001). 85. Kondoh, G. et al. Tissue-inherent fate of GPI revealed by GPI-anchored GFP transgenesis. FEBS Lett. 458, 299–303 (1999). 86. Zhou, Q. & Anderson, D. J. The bHLH transcription factors OLIG2 and OLIG1 couple neuronal and glial subtype specification. Cell 109, 61–73 (2002). 87. Kanda, T., Sullivan, K. F. & Wahl, G. M. Histone–GFP fusion protein enables sensitive analysis of chromosome dynamics in living mammalian cells. Curr. Biol. 8, 377–385 (1998). 88. Praitis, V., Casey, E., Collar, D. & Austin, J. Creation of low-copy integrated transgenic lines in Caenorhabditis elegans. Genetics 157, 1217–1226 (2001). 89. Rogers, E., Bishop, J. D., Waddle, J. A., Schumacher, J. M. & Lin, R. The aurora kinase AIR-2 functions in the release of chromosome cohesion in Caenorhabditis elegans meiosis. J. Cell Biol. 157, 219–229 (2002). 90. Clarkson, M. & Saint, R. A His2AvDGFP fusion gene complements a lethal His2AvD mutant allele and provides an in vivo marker for Drosophila chromosome behavior. DNA Cell Biol. 18, 457–462 (1999). 91. Savoian, M. S. & Rieder, C. L. Mitosis in primary cultures of Drosophila melanogaster larval neuroblasts. J. Cell Sci. 115, 3061–3072 (2002). 92. Koster, R. W. & Fraser, S. E. Tracing transgene expression in living zebrafish embryos. Dev. Biol. 233, 329–346 (2001). 93. Pauls, S., Geldmacher-Voss, B. & Campos-Ortega, J. A. A zebrafish histone variant H2A.F/Z and a transgenic H2A.F/Z:GFP fusion protein for in vivo studies of embryonic development. Dev. Genes Evol. 211, 603–610 (2001). 94. Helmchen, F., Fee, M. S., Tank, D. W. & Denk, W. A miniature head-mounted two-photon microscope. High-resolution brain imaging in freely moving animals. Neuron 31, 903–912 (2001). 95. Patterson, G. H. & Lippincott-Schwartz, J. A photoactivatable GFP for selective photolabeling of proteins and cells. Science 297, 1873–1877 (2002).

www.nature.com/reviews/genetics

REVIEWS 96. Ando, R., Hama, H., Yamamoto-Hino, M., Mizuno, H. & Miyawaki, A. An optical marker based on the UV-induced green-to-red photoconversion of a fluorescent protein. Proc. Natl Acad. Sci. USA 99, 12651–12656 (2002). 97. Llopis, J., McCaffery, J. M., Miyawaki, A., Farquhar, M. G. & Tsien, R. Y. Measurement of cytosolic, mitochondrial, and Golgi pH in single living cells with green fluorescent proteins. Proc. Natl Acad. Sci. USA 95, 6803–6808 (1998). 98. Nakanishi, T., Ikawa, M., Yamada, S., Toshimori, K. & Okabe, M. Alkalinization of acrosome measured by GFP as a pH indicator and its relation to sperm capacitation. Dev. Biol. 237, 222–231 (2001). 99. Ting, A. Y., Kain, K. H., Klemke, R. L. & Tsien, R. Y. Genetically encoded fluorescent reporters of protein tyrosine kinase activities in living cells. Proc. Natl Acad. Sci. USA 98, 15003–15008 (2001). 100. Zhang, J., Ma, Y., Taylor, S. S. & Tsien, R. Y. Genetically encoded reporters of protein kinase A activity reveal impact of substrate tethering. Proc. Natl Acad. Sci. USA 98, 14997–5002 (2001). 101. Janetopoulos, C., Jin, T. & Devreotes, P. Receptor-mediated activation of heterotrimeric G-proteins in living cells. Science 291, 2408–2411 (2001). 102. Brock, C. et al. Roles of Gβγ in membrane recruitment and activation of p110-γ/p101 phosphoinositide 3-kinase-γ. J. Cell Biol. 160, 89–99 (2003). 103. Miyawaki, A. et al. Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature 388, 882–887 (1997). 104. Nakai, J., Ohkura, M. & Imoto, K. A high signal-to-noise Ca(2+) probe composed of a single green fluorescent protein. Nature Biotechnol. 19, 137–141 (2001). 105. Yu, D., Baird, G. S., Tsien, R. Y. & Davis, R. L. Detection of calcium transients in Drosophila mushroom body neurons with camgaroo reporters. J. Neurosci. 23, 64–72 (2003). 106. Wang, J. W., Wong, A. M., Flores, J., Vosshall, L. B. & Axel, R. Two-photon calcium imaging reveals an odor-evoked map of activity in the fly brain. Cell 112, 271–282 (2003). 107. Beddington, S. P. An autoradiographic analysis of the potency of embryonic ectoderm in the 8th day postimplantation mouse embryo. J. Embryol. Exp. Morphol. 64, 87–104 (1981). 108. Thomas, P. & Beddington, R. Anterior primitive endoderm may be responsible for patterning the anterior neural plate in the mouse embryo. Curr. Biol. 6, 1487–1496 (1996). 109. Ormo, M. et al. Crystal structure of the Aequorea victoria green fluorescent protein. Science 273, 1392–1395 (1996). 110. Tsien, R. Y. The green fluorescent protein. Annu Rev. Biochem. 67, 509–544 (1998). 111. Yu, Y. A. & Szalay, A. A. A Renilla luciferase-Aequorea GFP (ruc-gfp) fusion gene construct permits real-time detection of promoter activation by exogenously administered mifepristone in vivo. Mol. Genet. Genomics 268, 169–178 (2002). 112. Lippincott-Schwartz, J. & Patterson, G. H. Development and use of fluorescent protein markers in living cells. Science 300, 87–91 (2003). 113. Zhang, J., Campbell, R. E., Ting, A. Y. & Tsien, R. Y. Creating new fluorescent probes for cell biology. Nature Rev. Mol. Cell Biol. 3, 906–918 (2002). 114. Matz, M. V. et al. Fluorescent proteins from nonbioluminescent Anthozoa species. Nature Biotechnol. 17, 969–973 (1999). This important reference describes the isolation of coelenterate-derived red fluorescent proteins. 115. Terskikh, A. V., Fradkov, A. F., Zaraisky, A. G., Kajava, A. V. & Angres, B. Analysis of DsRed mutants: space around the fluorophore accelerates fluorescence development. J. Biol. Chem. 277, 7633–7636 (2002). 116. Yarbrough, D., Wachter, R. M., Kallio, K., Matz, M. V. & Remington, S. J. Refined crystal structure of DsRed, a red fluorescent protein from coral, at 2.0-A resolution. Proc. Natl Acad. Sci. USA 98, 462–467 (2001). 117. Campbell, R. E. et al. A monomeric red fluorescent protein. Proc. Natl Acad. Sci. USA 99, 7877–7882 (2002).

NATURE REVIEWS | GENETICS

118.

119. 120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

131.

132.

133. 134.

135.

136.

137.

This report describes the mutagenesis of DsRed into a monomeric form that is expected to be more amenable to use in mice. Yanushevich, Y. G. et al. A strategy for the generation of non-aggregating mutants of Anthozoa fluorescent proteins. FEBS Lett. 511, 11–14 (2002). Gurskaya, N. G. et al. GFP-like chromoproteins as a source of far-red fluorescent proteins. FEBS Lett. 507, 16–20 (2001). Heim, R. & Tsien, R. Y. Engineering green fluorescent protein for improved brightness, longer wavelengths and fluorescence resonance energy transfer. Curr. Biol. 6, 178–182 (1996). Griesbeck, O., Baird, G. S., Campbell, R. E., Zacharias, D. A. & Tsien, R. Y. Reducing the environmental sensitivity of yellow fluorescent protein. Mechanism and applications. J. Biol. Chem. 276, 29188–29194 (2001). Bevis, B. J. & Glick, B. S. Rapidly maturing variants of the Discosoma red fluorescent protein (DsRed). Nature Biotechnol. 20, 83–87 (2002). Biggers, J. D., Moore, B. D. & Whittingham, D. G. Development of mouse embryos in vivo after cultivation from two-cell ova to blastocysts in vitro. Nature 206, 734–735 (1965). Biggers, J. D., McGinnis, L. K. & Raffin, M. Amino acids and preimplantation development of the mouse in protein-free potassium simplex optimized medium. Biol. Reprod. 63, 281–293 (2000). Erbach, G. T., Lawitts, J. A., Papaioannou, V. E. & Biggers, J. D. Differential growth of the mouse preimplantation embryo in chemically defined media. Biol. Reprod. 50, 1027–1033 (1994). Moore-Scott, B. A., Gordon, J., Blackburn, C. C., Condie, B. G. & Manley, N. R. New serum-free in vitro culture technique for midgestation mouse embryos. Genesis 35, 164–168 (2003). Ang, S. L. & Rossant, J. Anterior mesendoderm induces mouse Engrailed genes in explant cultures. Development 118, 139–149 (1993). Weaver, M., Dunn, N. R. & Hogan, B. L. Bmp4 and Fgf10 play opposing roles during lung bud morphogenesis. Development 127, 2695–2704 (2000). Naiche, L. A. & Papaioannou, V. E. Loss of Tbx4 blocks hindlimb development and affects vascularization and fusion of the allantois. Development 130, 2681–2693 (2003). Dickinson, M. E., Selleck, M. A., McMahon, A. P. & Bronner-Fraser, M. Dorsalization of the neural tube by the non-neural ectoderm. Development 121, 2099–2106 (1995). Echevarria, D., Vieira, C. & Martinez, S. Mammalian neural tube grafting experiments: an in vitro system for mouse experimental embryology. Int J. Dev. Biol. 45, 895–902 (2001). Denk, W., Strickler, J. H. & Webb, W. W. Two-photon laser scanning fluorescence microscopy. Science 248, 73–76 (1990). One of the first papers to show that multiphoton excitation can be used for fluorescence microscopy. Sheppard, C. J. & Wilson, T. The theory of the direct-view confocal microscope. J. Microsc. 124, 107–117 (1981). Denk, W. & Svoboda, K. Photon upmanship: why multiphoton imaging is more than a gimmick. Neuron 18, 351–357 (1997). Piston, D. W. Imaging living cells and tissues by two-photon excitation microscopy. Trends Cell Biol. 9, 66–69 (1999). Squirrell, J. M., Wokosin, D. L., White, J. G. & Bavister, B. D. Long-term two-photon fluorescence imaging of mammalian embryos without compromising viability. Nature Biotechnol. 17, 763–767 (1999). Lansford, R., Bearman, G. & Fraser, S. E. Resolution of multiple green fluorescent protein color variants and dyes using two-photon microscopy and imaging spectroscopy. J. Biomed. Opt. 6, 311–318 (2001). One of the first reports to describe the principle and application of emission fingerprinting of spectrally overlapping fluorochromes using laser scanning microscopy.

138. Dickinson, M. E., Bearman, G., Tilie, S., Lansford, R. & Fraser, S. E. Multi-spectral imaging and linear unmixing add a whole new dimension to laser scanning fluorescence microscopy. Biotechniques 31, 1272, 1274–1276, 1278 (2001). This paper describes the design and use of a specialized detector for the rapid imaging of λ-stacks for emission fingerprinting. 139. Dickinson, M. E., Simbuerger, E., Zimmerman, B., Waters, C. W. & Fraser, S. E. Multiphoton excitation spectra in biological samples. J. Biomed. Optics 8, 329–338 (2003). The first report to describe the principle and application of multiphoton-excitation fingerprinting of spectrally overlapping fluorochromes. 140 Papiannou, V. E. & Hadjantonakis, A.-K. in Cold Spring Harbor 2002 Mouse Molecular Genetics Meeting Abstract Book (Cold Spring Harbor Laboratory, New York, 2002). 141 Papiannou, V. E. & Johnson, R. in Gene Targeting: A Practical Approach (ed. Joyner, A.) 107–146 (IRL at Oxford Univ. Press, New York, 1993).

Acknowledgements We thank F. Costantini, L. Jones and J. Lichtman for generously providing figures. We apologize to the authors of many important findings and strains of mice that are not discussed here owing to space limitations. Our work is supported by grants from the National Institutes of Health, National Science Foundation and Muscular Dystrophy Association. A.-K.H. is a fellow of the American Heart Association.

Online links DATABASES The following terms in this article are linked online to: LocusLink: http://www.ncbi.nlm.nih.gov/LocusLink Axin2 | Nkx2.5 | Plxna3 | Rara | Rarb | Thy1 | Vegfa FURTHER INFORMATION BD Biosciences Inc.: http://www.clontech.com/gfp/index.shtml Bio-Rad Microscience: http://microscopy.bio-rad.com/ products/live_cell.htm C. elegans movies: http://www.bio.unc.edu/faculty/goldstein/lab/movies.html Carl Zeiss Advanced Imaging Microscopy: http://www.zeiss.de/us/micro/home.nsf Coherent Inc.: http://www.cohr.com Drosophila atlases: http://pbio07.uni-muenster.de Drosophila brain atlas: http://flybrain.unifreiburg.de/Flybrain/html FishScope: http://depts.washington.edu/fishscop Leica Microsystems: http://www.leicamicrosystems.com/website/lms.nsf Mouse MRI Atlas at the Biological Imaging Center: http://quad.bic.caltech.edu/mouseatlas NCBI mouse genome resources: http://www.ncbi.nlm.nih.gov/genome/guide/mouse Scott Fraser’s Laboratory: http://bioimaging.caltech.edu Spectra-Physics: http://www.spectraphysics.com The Edinburgh Mouse Atlas Project (EMAP): http://genex.hgu.mrc.ac.uk The Jackson Laboratory Induced Mutant Resource: http://www.jax.org/imr/index.html The Jackson Laboratory Mouse Genome Database (MGD): http://www.informatics.jax.org Virginia Papaioannou’s laboratory: http://cpmcnet.columbia.edu/dept/genetics/papaioannoulab.html Zebrafish vascular anatomy atlas: http://mgchd1.nichd.nih.gov:8000/zfatlas/Intro%20Page/ intro1.html Access to this interactive links box is free online.

VOLUME 4 | AUGUST 2003 | 6 2 5

ONLINE Anna-Katerina Hadjantonakis is interested in understanding how the mammalian body plan is established during embryogenesis, and in the development and use of tools for genome alterations and biological imaging in mice. Her graduate work at Imperial College, London, focused on the identification of the Celsr gene family, which are the vertebrate homologues of the Drosophila planar-cell-polarity gene flamingo. As a postdoctoral researcher at the Samuel Lunenfeld Research Institute, Toronto, she generated some of the first strains of fluorescent-protein-expressing mice and showed combinatorial reporter detection in living embryos. She is now working with Virginia Papaioannou at Columbia University, New York, using vital imaging as a tool for unravelling the genetic basis of mouse embryonic development. She will be joining the Program in Developmental Biology at the Sloan-Kettering Institute, New York, at the start of 2004. Mary E. Dickinson is developing new optical-imaging methodologies and applying imaging techniques to the study of mouse and chicken embryology. She trained as a mouse embryologist, first as an undergraduate at Vanderbilt University with Brigid Hogan, then as a graduate student at the Roche Institute, New Jersey, and Harvard University with Andy McMahon, where she worked on Wnt signalling. She came to the California Institute of Technology to work with Scott Fraser to combine advanced imaging technology with vertebrate development. She is now a Member of the Professional Staff in the Biological Imaging Center, at the California Institute of Technology, where she has been developing new imaging modalities for studying vertebrate development and using vitalimaging studies to characterize dynamic aspects of early mouse embryogenesis. Scott E. Fraser is the Anna L. Rosen Professor of Biology at the California Institute of Technology. He has a long-standing interest in following important lineage and migration events in embryos as they develop. For the past decade he has directed the Biological Imaging Center at the California Institute of Technology, which is dedicated to the application of advanced optical and magnetic resonance microscopies to intravital imaging in systems that range from sea urchin and fish embryos to mouse models of disease. Together with an active set of collaborators, he has applied his training in physics and biophysics to challenges such as imaging the patterns of cell migration in the developing cerebellum and testing the role of blood flow in the shaping of the developing heart. He has been active in science education, previously serving as the Co-director of the embryology course at the Marine Biological Laboratory in Woods Hole, Massachusetts.

• Optical-imaging technologies provide a way to perform live-cell analyses in an organismal context. Continued improvements in fluorescence-based imaging technologies allow deeper imaging and better spectral separation of fluorescent-protein reporters in living specimens. • Genetically-encoded fluorescent proteins that are expressed in normal and mutant mice represent high-resolution high-contrast multicolour vital markers for monitoring gene activity and protein function. They pave the way for the multidimensional multispectral imaging of living specimens and the generation of multidimensional atlases of model organisms. • An increasing number of fluorescent-protein reporters are available for use in mice. These fall into four main categories: spectral-variant reporters, subcellularly localized reporters, photo-activatable reporters and reporters that act as biochemical sensors. • The development of static in vitro culture systems for mammalian embryos and explanted tissues promotes normal ex vivo growth, which is necessary for vital-imaging experiments. • Researchers now possess a powerful toolbox that offers the potential to visualize and quantify biological processes at the cellular and subcellular level, both in vitro and in intact living organisms.

LocusLink LocusLink http://www.ncbi.nlm.nih.gov/LocusLink Axin2 http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=12006 Nkx2.5 http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=18091 Plxna3 http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=18846 Rarb http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=218772 Thy1 http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=21838 Vegfa http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=22339 Further Information

Virginia E. Papaioannou is a Professor in the Department of Genetics and Development at Columbia University. She has been interested in the genetic control of early mammalian development and the nature of embryonic stem cells since her graduate and postdoctoral studies at the Universities of Cambridge and Oxford in the UK, where she began using micromanipulation to study cell fate and cell lineage. As a faculty member of Tufts University, Boston, she applied the technique of gene targeting to mutate genes that are expressed during early embryonic development. She has been involved in the study of the T-box family of transcription factor genes since their discovery, and has generated and studied mutations in many of these genes, which have profound effects on development. • The mouse is the premier mammalian model organism and provides an unparalleled platform for modelling mammalian development and disease. Comprehensive functional annotation of the mouse genome relies on an integrative approach, using an array of tools for unravelling the molecular and cellular basis of normal and mutant phenotypes.

BD Biosciences Inc. http://www.clontech.com/gfp/index.shtml Bio-Rad Microscience http://microscopy.bio-rad.com/products/live_cell.htm C. elegans movies http://www.bio.unc.edu/faculty/goldstein/lab/movies.html Carl Zeiss Advanced Imaging Microscopy http://www.zeiss.de/us/micro/home.nsf Drosophila atlases http://pbio07.uni-muenster.de Drosophila brain atlas

ONLINE http://flybrain.uni-freiburg.de/Flybrain/html

FishScope http://depts.washington.edu/fishscop Leica Microsystems http://www.leica-microsystems.com/website/lms.nsf Coherent Inc. http://www.cohr.com Mouse MRI Atlas at the Biological Imaging Center http://quad.bic.caltech.edu/mouseatlas NCBI mouse genome resources http://www.ncbi.nlm.nih.gov/genome/guide/mouse Scott Fraser’s Laboratory http://bioimaging.caltech.edu Spectra-Physics http://www.spectraphysics.com The Edinburgh Mouse Atlas Project (EMAP) http://genex.hgu.mrc.ac.uk The Jackson Laboratory Induced Mutant Resource http://www.jax.org/imr/index.html The Jackson Laboratory Mouse Genome Database (MGD) http://www.informatics.jax.org Virginia Papaioannou’s laboratory http://cpmcnet.columbia.edu/dept/genetics/papaioannou-lab.html