Quantitative assessments of gene expression in early embryonic organs are difficult to obtain from whole embryos. Although in situ hybridization can elucidate spatiotemperal patterns of messenger RNA (mRNA), it cannot enumerate fold differences in transcript abundance. Furthermore, bona fide quantitative techniques, such as quantitative PCR or microarray hybridization, when performed on whole embryo preparations, can ascertain transcript levels only in the aggregate. Ultimately, high-resolution measurements of gene expression in individual organs require the material to be purified away from surrounding tissues. Moreover, methods for isolating pure organs can facilitate culturing and transplantation of their constituents for in vitro studies and the in vivo evaluation of stem or progenitor cells, respectively.

Because mammalian embryos develop in utero, embryos and their internal organs are not readily accessible without laborious microsurgeries (1). The limited number of offspring also restricts tissue availability. In contrast, zebrafish embryos develop externally and are available in significantly larger quantities. Therefore, we devised a method, void of micro-dissection, for isolating abundant embryonic organs from zebrafish focusing specifically on the developing heart.

Like all zebrafish organs, the heart develops rapidly. By 2 days postfertilization (dpf), it comprises two layers and two chambers, has established flow, and has undergone looping (2). Thereafter to day 4, the ventricular wall begins to thicken via addition of myocardial cells (3,4), and cardiac valves will soon arise from developing endocardial cushions (5).

To aid in tracking hearts, we used embryos that expressed green fluorescent protein (GFP) in the myocardium under control of the cardiac myosin light chain 2 (cmlc2) promoter (6). Embryos at the desired stages were resistant to enzymatic treatment with trypsin and collagenase. Serendipitously, we discovered that the fluid forces in media repeatedly drawn into and expelled from a needle and syringe ruptured embryonic yolk membranes and extricated, among other tissues, intact hearts ((Figure 1)). Subsequent size fractionation of the disrupted embryos and manual retrieval of GFP+ hearts afforded cardiac tissue with purity that approached 100%.

Figure 1.

Embryos hemizygous for Tg(cmlc2: GFP) between 2 and 4 dpf were generated and handled as described (6). Prior to disruption, approximately 500 embryos were anesthetized, transferred to a 1.5-mL microfuge tube, washed three times with embryo disruption medium [EDM; Gibco Leibovitz's L-15 Medium (Invitrogen, Carlsbad, CA, USA) containing 10% fetal bovine serum (Hyclone, Logan, UT, USA), ice-cold at every step] and resuspended in 1.25 mL EDM. The microfuge tube, a 19-gauge needle (Precision Glide, 1 1/2 inches long, regular bevel; BD Medical, Franklin Lakes, NJ, USA) and a 6-mL syringe (Monoject syringe with a regular luer tip; Kendall Company, Mansfield, MA, USA) were secured to a ring stand ((Figure 1) A). The needle was centered in the x- and z-axes of the microfuge tube and aligned in the y-axis with the 0.1 mL volume mark.

Approximately 1 mL EDM containing embryos was drawn into the needle and syringe and immediately expelled back into the microfuge tube 30 times at a rate of 1 s per syringe motion. Fragmented embryos were applied to 105 µM nylon mesh (Small Parts, Miami Lakes, FL, USA), and the flow-through was collected in a 30-mm polystyrene Petri dish. The microfuge tube, needle, syringe, and mesh were washed with additional media that was added to the flow-through. The flow-through was subsequently applied to 40 µM nylon mesh. The Petri dish that collected the 105 µM flow-through was rinsed with additional media that was applied to the 40 µM mesh. The 40 µM mesh was inverted, and the retained material was washed off with EDM into a 30-mm Petri dish.

Intact GFP+ ventricles were identified under fluorescent light and collected selectively with a p20 micropipet under transmitted light ((Figure 1), D and E). Approximately 40 ventricles were collected each time the micropipet was filled. To minimize contamination by noncardiac tissues, the hearts were subsequently expelled into a 30-mm Petri dish containing fresh EDM and retrieved selectively a second time before being accumulated in a microfuge tube on ice. From 500 embryos, approximately 250 total hearts were retrieved in the span of 1 h. Hearts were pelleted, the media was decanted, and the preparations were frozen in a dry ice/ethanol bath prior to storage at -80°C.

At this time, visual inspection suggested that cardiac tissue purity approached homogeneity ((Figure 1), F and G). The percentage of ventricles with attached complete atria was often between 10%–30% but as high as 50% in some preparations ((Figure 1), F and G). The remainder of ventricles retained partial or no atria ((Figure 1), F and G). We were unable to identify more than a few sole atria in each preparation. The atria that were unaccounted for were either retained in the embryo's bodies or assumed to be sufficiently fragmented so as to evade identification. In general, we found that using a smaller number of syringe motions increased the percentage of ventricles with atria but decreased the percentage of hearts recovered. Nonfluorescent outflow tracts (e.g., bulbus arteriosa) were attached to the majority of ventricles ((Figure 1), F and G). Spontaneous rhythmic contractions were observed when the media was warmed to room temperature, confirming tissue viability (see Supplementary Video 1 available online at www.BioTechniques.com).

To objectively assess the content and purity of heart preparations, we measured the relative amounts of seven tissue-specific markers in three tissue fractions with quantitative PCR (see (Table 1) and References 7–14). Total RNA was isolated with TRIZOL® reagent (Invitrogen), with each heart yielding between 1.5 and 3 ng. Total RNA was random primed and used as templates in first-strand synthesis (SuperScript™ First Strand Synthesis System for RT-PCR; Invitrogen). Quantitative PCR was performed with SYBR® green DNA dye and an ABI PRISM® 7000 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's instructions. All signals were normalized to the ubiquitous β-actin mRNA (5′-GCTGTTTTCCCCTCCATTGTT-3′, 5′-TCCCATGCCAACCATCACT-3′).

Three tissue fractions were prepared from Tg(cmlc2:GFP) embryos on day 3 postfertilization as described and analyzed by quantitative PCR. Signals in all samples were normalized to the ubiquitous messenger RNA (mRNA), -actin, and reported as fold-differences relative to whole embryos. Bolded numbers identify the fraction(s) enriched with each transcript. six3b mRNAs were used as a control for tissues not predicted to detach from the embryos. R forward primer; R, reverse primer.

^aFold differences in the heart sample represent the averages of two experiments.

^bcmlc2 is expressed in both atrial and ventricular myocardium.

Heart preparations harbored 340-fold more cmlc2 mRNA than whole embryos, indicating a substantial enrichment for myocardium ((Table 1)). They also contained 79-fold more atrial myosin heavy chain mRNA, confirming the presence of atria. Lastly, purified hearts retained endocardium, as evidenced by a 23-fold greater quantity of endothelial-tie2 transcripts in the heart fraction. Negligible amounts of other tissue-specific mRNAs were present in the heart fraction ((Table 1)), thereby confirming tissue purity suggested by visual inspection ((Figure 1), F and G). Enrichment of specific tissues in other size fractions suggests that manual or automated retrieval of these organs with a micropipet or fluorescence-activated cell sorting (FACS), respectively, could yield pure tissue preparations for other organs ((Table 1)).

We present a simple method for isolating intact embryonic hearts from zebrafish. This technique will permit quantitative comparisons of gene expression in wild-type and mutant hearts. Single transcripts can be assessed using quantitative PCR. Alternatively, expression profiling of myriad genes can be accomplished with microarrays or serial analysis of gene expression (SAGE). This technique will be particularly useful for identifying downstream effectors of proteins that, when mutated, cause defects in cardiomyocyte proliferation or endocardial cushion maturation.

2.) Stainier Y. D. Fouquet B. Chen N. J. Warren S. K. Weinstein M. B. Meiler E. S. Mohideen A. M. Neuhauss, L C. S., Mutations affecting the formation and function of the cardiovascular system in the ze-brafish embryo, Development, P285 - P292

3.) Antkiewicz S. D. Burns G. C. Carney A. S. Peterson E. R. Heideman W., Heart malformation is an early response to TCDD in embryonic zebrafish, Toxicol. Sci., P368 - P377

4.) Mably D. J. Mohideen A. M. Burns G. C. Chen N. J. Fishman C. M., heart of glass regulates the concentric growth of the heart in zebrafish, Curr. Biol., P2138 - P2147

5.) Beis D. Bartman T. Jin W. S. Scott C. I. D'Amico A. L. Ober A. E. Verkade H. Frantsve J., Genetic and cellular analyses of zebrafish atrioventricular cushion and valve development, Development, P4193 - P4204

6.) Burns G. C. Milan J. D. Grand J. E. MacRae A. C. Fishman C. M., High-throughput assay for small molecules that modulate zebrafish embryonic heart rate, Nat. Chem. Biol., P263 - P264

7.) Yelon D. Horne A. S. Stainier Y. D., Restricted expression of cardiac myosin genes reveals regulated aspects of heart tube assembly in zebrafish, Dev. Biol., P23 - P37

8.) Berdougo E. Coleman H. Lee H. D. Stainier Y. D. Yelon D., Mutation of weak atrium/atrial myosin heavy chain disrupts atrial function and influences ventricular morphogenesis in zebrafish, Development, P6121 - P6129

9.) Motoike T. Loughna S. Perens E. Roman L. B. Liao W. Chau C. T. Richardson D. C. Kawate T., Universal GFP reporter for the study of vascular development, Genesis, P75 - P81

10.) Brownlie A. Hersey C. Oates C. A. Paw H. B. Falick M. A. Witkowska E. H. Flint J. Higgs D., Characterization of embryonic globin genes of the zebrafish, Dev.Biol., P48 - P61

11.) Denovan-Wright M. E. Pierce M. Sharma K. M. Wright M. J., cDNA sequence and tissue-specific expression of a basic liver-type fatty acid binding protein in adult zebrafish (Danio rerio), Biochim. Biophys. Acta, P227 - P232

12.) Argenton F. Zecchin E. Bortolussi M., Early appearance of pancreatic hormone-expressing cells in the zebrafish embryo, Mech. Dev., P217 - P221

13.) Andre M. Ando S. Ballagny C. Durliat M. Poupard G. Briancon C. Babin J. P., Intestinal fatty acid binding protein gene expression reveals the cephalo-caudal patterning during zebrafish gut morphogenesis, Int. J. Dev. Biol., P249 - P252

14.) Seo C. H. Drivenes S. Ellingsen S. Fjose A., Expression of two zebrafish homologues of the murine Six3 gene demarcates the initial eye primordia, Mech. Dev., P45 - P57