SEARCH

SEARCH BY CITATION

Keywords:

  • chicken embryo;
  • Gallus gallus;
  • in situ hybridization;
  • locked nucleic acid;
  • MicroRNA

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

MicroRNAs (miRNAs) are small, abundant, noncoding RNAs that modulate protein abundance by interfering with target mRNA translation or stability. miRNAs are detected in organisms from all domains and may regulate 30% of transcripts in vertebrates. Understanding miRNA function requires a detailed determination of expression, yet this has not been reported in an amniote species. High-throughput whole mount in situ hybridization was performed on chicken embryos to map expression of 135 miRNA genes including five miRNAs that had not been previously reported in chicken. Eighty-four miRNAs were detected before day 5 of embryogenesis, and 75 miRNAs showed differential expression. Whereas few miRNAs were expressed during formation of the primary germ layers, the number of miRNAs detected increased rapidly during organogenesis. Patterns highlighted cell-type, organ or structure-specific expression, localization within germ layers and their derivatives, and expression in multiple cell and tissue types and within sub-regions of structures and tissues. A novel group of miRNAs was highly expressed in most tissues but much reduced in one or a few organs, including the heart. This study presents the first comprehensive overview of miRNA expression in an amniote organism and provides an important foundation for investigations of miRNA gene regulation and function. Developmental Dynamics 235:3156–3165, 2006. © 2006 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

MicroRNAs (MiRNAs) are an abundant class of noncoding RNAs whose known functions are to regulate protein levels by binding to 3′UTRs of target mRNAs and either inhibiting translation or inducing mRNA degradation (He and Hannon,2004; Pillai,2005; Valencia-Sanchez et al.,2006). miRNAs are detected in organisms from all domains, and recent reports indicate that vertebrate genomes contain at least hundreds of miRNA genes that regulate stability or translation of 20-30% of all mRNAs (Bentwich et al.,2005; Berezikov and Plasterk,2005; Legendre et al.,2005; Xie et al.,2005).

miRNAs are transcribed as 70–100 nucleotide (nt) pri-miRNA transcripts that are processed in the nucleus to a 70-nt hairpin structure pre-miRNA. The pre-miRNA is transported to the cytoplasm and processed into a miRNA:miRNA duplex. One or occasionally both of the strands of the miRNA duplex are individually assembled into an RNA-induced silencing complex (RISC) consisting of several proteins and the mature single-stranded 21–22 nt miRNA. Although their small size has made detection of mature miRNAs somewhat problematic, Northern blot (Chen and Okayama,1987; Valoczi et al.,2004; Watanabe et al.,2005), microarray (Barad et al.,2004; Thomson et al.,2004), and most recently whole mount in situ hybridization (Aboobaker et al.,2005; Weinholds et al.,2005; Kloosterman et al.,2006) approaches have been used to monitor miRNA expression. Whereas whole mount in situ hybridization using antisense RNA probes is not sensitive enough to reliably detect short RNA sequences, Locked Nucleic Acids (LNAs), a novel type of RNA analogue, can hybridize to target RNA sequences with extremely high specificity and stability (Wahlestedt et al.,2000; Elmen et al.,2005) and have proven effective for whole mount in situ hybridization detection of miRNA expression (Weinholds et al.,2005; Kloosterman et al.,2006b). Even using LNAs, however, the ability to detect miRNAs in vivo has been variable. An in situ hybridization overview of miRNA expression in zebrafish detected the majority of miRNAs and demonstrated dynamic spatial and temporal expression patterns (Weinholds et al.,2005). By contrast, an in situ hybridization screen in mouse detected only a small number of the most highly expressed miRNAs (Kloosterman et al.,2006a). A comprehensive view of miRNA expression during embryogenesis in a higher vertebrate organism is not presently available.

The chicken embryo provides an advantageous alternative to mouse for miRNA expression analyses. As an amniote, developmental processes in chicken closely mimic those in mammalian species, including mouse and humans. Whole mount in situ hybridization protocols have also been optimized for chick to give outstanding sensitivity and low background (Nieto et al.,1996; Bell et al.,2004), and embryos are easily and inexpensively obtained. To begin to shed light on the potential functions of miRNAs during development of amniote species, we undertook a comprehensive whole mount in situ hybridization expression analysis of 111 mature miRNA sequences that are encoded by 135 distinct chicken miRNA genes (Griffith-Jones,2004; Griffiths-Jones et al.,2006). Results show that the majority of known miRNAs are expressed during the first four days of embryogenesis, in a wide variety of temporally and spatially distinct patterns that implicate miRNAs in important developmental processes.

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

To detect miRNAs during chicken embryogenesis, LNA modified DNA oligonucleotides (Koshkin et al.,1998; Wengel et al.,2003) antisense to 111 distinct mature chicken miRNA sequences transcribed from 135 miRNA genes (Table 1) were end labeled with digoxygenin-UTP (DIG) and used as probes for whole mount in situ hybridization analysis (Nieto et al.,1996; Bell et al.,2004) of embryos between Hamburger-Hamilton (HH; Hamburger and Hamilton,1951,1992) stages 3–26. One hundred six miRNA sequences were obtained from miRBase (Griffith-Jones,2004; Griffiths-Jones et al.,2006), and five potentially novel avian miRNAs (miR-9*, -21, -22, -144, and -363) were identified through homology searches using human miRNA gene sequences (see Supplemental Table 2, which can be viewed at www.interscience.wiley.com/jpages/1058-8388/suppmat). Eighty-four miRNAs were detected during the stages examined; nine were ubiquitously expressed and 75 showed reproducible differential expression in one or more cell layers or embryo regions, including four of the human orthologs.

Table 1. Summary of miRNA Expression Data
inline image

Early Development, Ectoderm and Endoderm

Consistent with expression studies in other organisms (Watanabe et al.,2005; Weinholds et al.,2005), few miRNAs were detected at early embryonic stages (Table 1). At successively later stages, the number of miRNAs detected increased rapidly in tissues derived from all three primary germ layers (endoderm, ectoderm, and mesoderm). This trend is consistent with studies in lower vertebrates and indicates that miRNAs play a greater role in regulating organogenesis, differentiation, and post-differentiation events than early embryonic patterning processes. At the primitive streak stage, let-7b, miR-130b, and -367 were expressed in the epiblast, and miR-130b was also detected in involuting cells of the streak (Fig. 1A–C). In zebrafish, early zygotic expression of miR-430 targets maternal RNAs for clearance from the embryo (Giraldez et al.,2005). Although miR-430 orthologues have not been identified in any amniote species, it is conceivable that let-7b, miR-130b, and/or -367b could serve a similar function.

thumbnail image

Figure 1. Whole mount in situ hybridization expression of miRNAs at gastrula stages and in the ectoderm and endoderm. Probes are listed on the panels of all figures. A–C: Dorsal views and transverse sections of stage-4 embryos showing let-7b, miR-130b, and -307 labeling in the epiblast. miR-130b also shows expression in the involuting cells of the primitive streak (B, inset). D: Trunk level transverse section at stage 15 showing miR-30e expression in the surface ectoderm and mesonephric ducts (arrow). E,F: Dorsal and lateral views of miR-205a expression in the caudal non-neural ectoderm at stage 10 (E), and at stage 20 in the lateral and ventral ectoderm, including the limb buds (arrows; F). G: Dorsal view of miR-200b label at stage 13. H: A transverse section of G shows normal localized expression in the ectoderm (arrows) and the endoderm (arrowheads). I–K: Lateral view of miR-200b at stage 21 (I) in transverse sections show localization to the lateral and ventral ectoderm (J) and to the gut endoderm (K). L: Lateral view, miR-205b labeling in the surface ectoderm with intensification in the head, pharyngeal arches, and limb buds. M: Transverse section of L. N: Enlargement of boxed area in M. O: Dorsal view showing miR-21 expression in the amnion (arrowheads, stage 13). P: Ventro-lateral view of miR-375 expression in endodermal precursors of the pancreatic insulin secreting cells (stage 15). Q: Enlargement of boxed area in P.

Download figure to PowerPoint

miR-30e transcripts were detected throughout most of the non-neural ectoderm at stage 10 (data not shown) and at later stages also were detected in the nephric ducts (Fig. 1D). miR-200a and -200b are located 2.2 kb apart in an intergenic region on chromosome 21 and show similar expression patterns in the ectoderm and endoderm at stage 12 (Fig. 1G,H) and at later stages in the ectoderm, nose, gut, and mesonephric ducts (Fig. 1I–K). miR-205a, which is located in an intergenic region, was expressed in the caudal non-neural ectoderm (Fig. 1E) and at progressively later stages in the ventro-lateral nonneural ectoderm. At limb bud stages, miR-205a became prominent in the limb ectoderm, including the apical ectodermal ridge (AER; Fig. 1F). miR-205b is located in an intron of a collagen alpha chain precursor gene, yet shows a similar expression pattern to -205a (Table 1; Fig. 1L). This is the only known case in chick of two closely related miRNA genes that are intergenic and intragenic, respectively. Apparent striping in the trunk for both probes was due to labeling of small extensions from the ectoderm that coincided with the deeper somitic segmentation pattern (Fig. 1M,N). miR-21 has not been previously reported in chicken and was expressed in the amnion (Fig. 1O; see Supplementary Table 3). miR-375, which has been implicated in insulin exocytosis regulation (Poy et al.,2004), was noted in the endodermal pancreatic rudiment (Fig. 1P,Q), and later in the pancreas. miR-375 is also expressed in the pituitary rudiment (data not shown). Ten additional miRNAs were preferentially expressed in the surface ectoderm (Table 1).

Central Nervous System

At least 30 miRNAs were detected in the developing nervous system (Fig. 2, Table 1), although no single miRNA was expressed in all neural tissues. miRNAs expressed in the CNS show differential expression along the dorsal-ventral, medial-lateral, and anterior-posterior axes, in localizations that suggest functions in regional patterning and within specific groups of neuronal and non-neuronal cells. The let-7 miRNA gene family codes for at least eleven miRNAs that differ by up to four nucleotides. let-7a-1, -7a-2, -7a-3, and -7j have identical mature miRNA sequences, and the let-7a/j probe detected expression in the dorsal half of the hindbrain (Fig. 2A). let-7b and -7c were expressed in the hindbrain and spinal cord, whereas let-7f and -7k transcripts were detected in the hindbrain (Fig. 2B) and let-7i and -7k were detected in the forebrain (Table 1).

thumbnail image

Figure 2. miRNA expression in the nervous system. A: Transverse section at caudal hindbrain level at stage 16, expression of let-7a in the dorsal half of the neural tube (arrow). B: Dorsal view, stage-25 embryo shows hindbrain let-7k label similar to A. C–F: Complex neural expression pattern of miR-9 in forebrain or telencephalic vesicles (C, arrow) and diencephalon (arrowhead) at stage 22. D: Dorsal view, stage 24, expression in the hindbrain varies between rhombomeres, is absent in the floor plate and stronger at the margin of the roof plate. E: A transverse section of the spinal cord of the embryo in D; expression is strongest in the dorsal margin adjacent to the roof plate (near top) and in the proliferating mantle layer (arrowheads). F: View looking down into the opened midbrain at stage 24, showing wing-like projections coming from the diencephalon (arrows), coincident with the forming optic tectum. G: Dorsal view, stage 23, miR-10b expression in the spinal cord. H: Dorsal view, stage 10, mir-17-5p expression in the head surface ectoderm and caudal neural plate. I: Dorsal view, stage 24 (same embryo as in Fig. 5D) shows expression of miR-20a in the dorsal margins of the hindbrain. J: Transverse section of I. K: Lateral view, stage 20, expression of miR-20b in all tissues, with higher levels in the wing and midbrain. L: Transverse section of embryo in K showing label in the mesencephalon. M: Dorsal view, stage 20, expression of miR-106 in the neural tube. N: Dorsal view, stage 21, shows similar expression of Mir-124a including hindbrain and midbrain (excluding floor plate, arrowhead, and roof plate, arrow. O: Dorsal view at stage 21 shows miR-124b expression in spinal cord. P: Transverse section of O, mantle zone, and exclusion from floor and roof plate. Q: Transverse section of stage-23 embryo also labeled with miR-124b showing label in the pituitary rudiment at the base of the diencephalon. R: Dorsal view, stage-25 expression of miR-135 in hind and midbrain, in the floor plate (arrowhead) and roof plate (arrow), respectively; a pattern apparently complementary to 124a (see N). S: Lateral view shows CNS expression of miR-153 at stage 21. T,U: Lateral views of miR-183 label in cranial ganglia (T, arrowheads; stage 19) and spinal ganglia (U, stage 23). V: Lateral view, mir-184 at stage 17 in the cranial ganglia (arrowheads). W: Transverse section of another embryo showing the same probe hybridizing in lens at stage 16. X: Right frontal view at stage 25; neural expression of miR-187 is limited to the dorsal closure line of the CNS. Y,Z: Lateral view, stage 20 showing miR-204 in the outer layer of the retina (neural retina) in whole mount and (Y) transverse section (arrow; Z). A': Whole mount and transverse section showing miR-200b expression in the otic placodes (arrow). B':Transverse section shows expression of miR-218 in the motor horns (mh) of the spinal cord at stage 25 (also in the roof of the hindbrain, not shown). C': Dorsal view at stage 12; D': lateral view at stage 21, expression of miR-219 in the neural tube/spinal cord. E': Transverse section of the embryo in D', expression in the roof plate and dorsal-most spinal cord, in the margin between the ventricular and mantle zone, with heightened expression at the sulcus limitans (arrowhead). F': Dorsal view in the roof of the hindbrain labeled with miR-449 at stage 20.

Download figure to PowerPoint

thumbnail image

Figure 5. miRNA expression in the craniofacial region. A: Diagram of lateral view of head and pharyngeal arches at day 3–4. Named labeled arches are (I) maxillary/mandibular, and (II) hyoid. B: Expression of miR-10b in the mesenchyme of all pharyngeal arches at stage 23. C: Expression of miR-19a at stage 19 in face and maxillary/mandibular arch. D: Expression of mir-20a in surface ectoderm of the face and arch I (from below), stage 24. E: Same embryo, lateral view. F: Expression of miR-106 in all arches, stage 22. G: Transverse section from F showing surface ectoderm and subcutaneous mesenchyme (arrowhead), and a labeled mandibular adductor muscle (arrow). Intermandibular muscles also expressed miR-106 (not shown). Several miRNAs were expressed in the caudal half of the hyoid arch (II), including (I) miR-128a/b, stage 19; (H,J,L,N,O) stage 21. K: Frontal view, expression in several areas of the face and neck, stage 24. M: Expression of miR-205a in surface ectoderm of the arches (and elsewhere), stage 22.

Download figure to PowerPoint

The miR-9 probe, recognizing identical miRNAs transcribed from two miR-9 genes, revealed the strongest and most diverse expression pattern of the neurally expressed miRNAs, labeling subsets of cells in all brain vesicles as well as in the spinal cord by stage 19 (Fig. 2C–F). Although miR-9* has been reported in some species (Weinholds et al.,2005), we did not detect expression of a chicken miR-9* miRNA at the embryonic stages examined (data not shown). Several miRNAs showed restricted expression in both neural and non-neural tissues. The current draft of the chicken genome assembly places miR-10b on the sense strand in intron 2 of the HoxD10 gene. However, the genome assembly has some ambiguity in the Hox cluster. Transcripts of miR-10b were detected along the full length of the spinal cord (Fig. 2G) and forelimb mesenchyme (see Fig. 6B,C,K). miR-17-5p was detected in the head region and posterior neural folds at stage 10 (Fig. 2H), and at later stages in many tissues but reduced in the heart (see Figs. 3J, 4, Table 1). miR-20a and -20b were strongly expressed in the midbrain vesicle, dorsal hindbrain, and spinal cord (Fig. 2I–L), and in facial structures (see Fig. 5D,E), but were also expressed at lower levels throughout the embryo with lowest expression in the heart (see Fig. 3K,4). miR-106 was expressed in the spinal cord (Fig. 2M), hindbrain, and pituitary rudiment, and in several mesodermal structures (see Fig. 5F,G). miR-106 is polycistronic with -18b and -20b, and possibly also -363, and their superficial limb, CNS, and pharyngeal expression appears similar.

thumbnail image

Figure 6. miRNA expression in the limbs. A: let-7a expression at stage 16, more intense in wing than leg bud (arrowheads). B,C: Higher miR-10b expression at stages 16 and 24, the wing versus leg buds (see also K, L). D: miR-18b expression at stage 17 in wing versus leg buds (arrowheads). E: Stage 21, mesodermal expression of mir-18b in the margin of both wing and leg buds. F: miR-363 expression in both wing and leg buds at stage 18. G–I: Limb bud mesoderm expression in wing (shown) and leg buds. A similar pattern was obtained with miR-128a (wing), miR-130b (wing and leg), and miR-217 (wing). J: miR-17-5p expression is nearly ubiquitous but excludes the heart (see Fig. 3J) and AER on the limbs (stage 23). K,L: Higher magnifications of wing and leg buds from C. M: Strong expression of miR-199a in the leg bud, and also other tissues (see Fig. 4). N: miR-200b expression is restricted to surface ectoderm in the limb including the AER.

Download figure to PowerPoint

thumbnail image

Figure 3. miRNA expression in mesodermal derivatives. A: Ventral view, miR-1 expression in the forming heart tube (arrowhead, stage 9), stage 13 (B), and stage 16 (C) in both the looped heart and somitic myotome. D: At stage 22, prominent labeling is observed in the heart and somites (arrowheads). E: Transverse section through D; expression is limited to the myotome of somites (arrowheads) and myocardium (asterisks). F: Expression of miR-1b in the heart and somites at stage 22 is similar to miR-1. G: Dorsolateral view at stage 15 of miR-133a expression in the myocardium and myotome. H: miR-34a expression in the atria (arrow) at stage 23. I: miR-206 expression at stage 20 in the somite (myotome) but not the heart. J,K: Examples of miRNAs that are expressed generally throughout the embryo except in the heart (stage 17 and 22). Other mesodermal expression included miR-126 in vasculature and blood islands (arrowheads) at stage 13 (L,M) and at stage 22 in individual capillaries (arrows) of the pharyngeal arches (N; arch numbers shown). O: Dorsal view, miR-144 expression limited to the blood islands at stage 8. P,Q: miR-7b and -140 at stages 14–15 in notochord, (arrowheads). R: Transverse section, miR-30a expression in nephric ducts (asterisks), stage 22.

Download figure to PowerPoint

thumbnail image

Figure 4. Northern analysis of miRNA expression in embryonic tissues. Lane 1: Twenty nanograms of DNA oligonucleotide corresponding to the indicated miRNA sequence. Lane 2: Negative control, 20 ng of DNA 22mer from alpha actin 3′UTR. Lanes 4–7: 40 μg each of mRNA isolated from stage 24 heart, limb, head, and remaining body (embryos minus hearts, limbs, and heads). miR-17-5p signal in the heart mRNA lane is 12–15-fold lower than the signal in the body mRNA lane. miR-20b with comparable fivefold reduction in heart. miR-1 lane demonstrates localized expression in the heart lane for miR-1 (positive control for heart). RNA lane shows loading control.

Download figure to PowerPoint

The miR-124a and -124b probes produced identical labeling patterns in the hindbrain and in the midbrain, in the lateral regions of the spinal cord, and in the pituitary rudiment (Fig. 2N–Q). miR-124a lies in an intergenic region on chromosome 2, while two miR-124b genes are located 2.3 kb apart on chromosome 23. The miR-124a and -124b probe sequences differ by a single internal nucleotide, and because a single nucleotide mismatch did not limit hybridization for miR-124 in zebrafish (Kloosterman et al.,2006), each probe may recognize both transcripts. Additional data on miRNAs expressed in the central and peripheral nervous systems can be found in Figure 2R-F', and in Table 1.

Derivatives of the Mesoderm

Several miRNAs were detected specifically in cardiac muscle and myotomal skeletal muscle cells (Fig. 3). miR-1 (transcribed from two genes, miR-1-1 and miR-1-2) differs from miR-1b by just one internal nucleotide, and so each probe may recognize both transcripts (Kloosterman et al.,2006). The miR-1 probe detected transcripts in the forming heart tube beginning at the onset of cardiac myocyte differentiation at stages 9–10, and expression persisted in the heart at least through stage 22 (Fig. 3A–E). The miR-1 probe also detected transcripts in the somitic myotome beginning at stage 14, coincident with the onset of skeletal muscle cell differentiation. A similar though somewhat weaker labeling pattern was observed with the miR-1b probe (Fig. 3F). miR-1 is one of the most highly studied miRNAs (Zhao et al.,2005; Chen et al.,2006), and shows interesting variations in heart expression in different organisms. miR-1 is detected in the dorsal vessel in Drosophila, is not detected in hearts of fish or amphibians, but is highly expressed in hearts of chickens, mice, and humans (Zhao et al.,2005; Kloosterman et al.,2006). A comprehensive evolutionary comparison can be found in Ason et al. (2006).

miR-1-1, -1-2, -1b, and -206 are located immediately 5′ to miR-133a-2, -133a-1, -133a-3, and 133b, respectively. Since each pair is apparently expressed as a polycistron (Chen et al.,2006), it is not surprising that the miR-133a probe (which differs from miR-133b by a single terminal nucleotide) localizes to both the heart and myotomal skeletal muscles (Fig. 3G). Several additional miRNAs were expressed above background in the atria (Table 1), including miR-34a (Fig. 3H). although trapping and background staining complicate interpretation of true signal in the atria.

Although miR-206 and miR-1 are evolutionarily related, miR-206 is expressed in myotomal skeletal muscle cells but not in the heart (Fig. 3I). Zhao et al. (2005) identified a Mef2 cis element in the 5′ promoter region of the mouse miR-1-1 gene that is required for cardiac specific expression. An identical Mef2 element (core sequence 5′-CTAAATATGG-3′) is present in the 5′ flanking region of the chicken miR-1-1 gene (Zhao et al.,2005). Analysis of a 5-kb 5′ flanking region of chicken miR-206 identified a Mef2-like element containing a single nucleotide substitution (5′-CcAATATGG-3′) that has been shown to abolish Mef2 binding (Andres et al.,1995).

miRNAs expressed in other mesodermal derivatives include miR-126 in blood islands and the vascular endothelium (Fig. 3L–N). miR-144 also has not been previously reported in chicken and was detected transiently in the blood islands (Fig. 3O). let-7b and miR-140 (Fig. 3P,Q), and several other miRNAs (Table 1) were detected in the notochord. miR-30a and -30e were localized to the nephric ducts (Fig. 3R, and see Fig. 1D) at levels above background. However, trapping and background staining complicated interpretation of label in the notochord and nephric ducts.

Nearly Ubiquitous Expression

A novel class of miRNA expression patterns was identified that showed moderate to high levels generally throughout the embryo but at much reduced levels in one or a few tissues, including the heart. These include miR-17-3p and -17-5p, -18a, -19a -19b, -20a, -20b, -92, -106, -107, and -363 (Fig. 3J,K; and data not shown). miR-17, -18a, -19a, -20a -19b, and -92 are part of the miR-17-92 polycistron, and miR-106, -20b, -18b, and -363 are located within a paralagous cluster on chromosome 4 (Hayashita et al.,2005; He et al.,2005). Northern analyses confirmed that miR-17-5p and -20b are expressed broadly in the embryo but at reduced levels in the heart (Fig. 4). At later stages, expression of the miR-17-92 polycistron miRNAs was also reduced in portions of the ectoderm, especially the AER of the limb (Fig. 6J). Elevated expression of the miR-17-92 cluster miRNAs is associated with proliferation and tumor progression (Hayashita et al.,2005). However, cardiac myocytes replicate throughout fetal life (Soonpaa and Field,1997), so proliferative capacity of myocytes does not strictly correlate with miR-17-92 expression.

Pharyngeal Arches and Facial Structures

Thirteen miRNAs were detected in the craniofacial region (Fig. 5A). miR-10b, -19, -20a, -106, -125b, -128a/b, -130b, -135b, -184, -205a, -217, -218, and -222b showed expression in one or more pharyngeal arches (Fig. 5B–J, L–O; and data not shown). Whereas most of these miRNAs were detected in the mesenchyme of multiple arches, miR-125b, -128a/b, -130b, -135b, -217, 218, and 222b also showed localized expression in the caudal portion of arch II, the hyoid arch. miR-106 was detected in all four arches in the mesenchyme immediately adjacent to the ectoderm and in the developing arch muscles (Fig. 5F,G), whereas miR-140 was expressed in the mesenchyme of the frontal prominence, the medial and lateral nasal prominences, and in the maxillary prominence (Fig. 5K). miR-130b was detected in both ectoderm and mesoderm of the pharyngeal arches, but only in the mesenchyme of the pharyngeal clefts. miR-184 was expressed in the pharyngeal arch ectoderm. A similar number of miRNAs are detected in zebrafish pharyngeal arches (Weinholds et al.,2005). However, of the combined 21 miRNA genes that are expressed in pharyngeal arches of either species, only miR-30c, -140, and -205a are commonly expressed. Eight of the remaining genes are found only in one species and the remaining 11 miRNAs are expressed in different embryonic locations. These results highlight a general concept that miRNA expression patterns, and by implication the mRNAs that they regulate, can show considerable divergence across evolution. It should be mentioned that the zebrafish screen (Weinholds et al.,2005) extended to a later relative developmental time than this study in chick, and so it is possible that additional miRNAs are commonly expressed in the pharyngeal arches at later stages (Ason et al.,2006).

Limbs

Twenty-one miRNAs show differential expression in the developing limbs in patterns that suggest functions in limb growth and patterning. let-7a, miR-10b, -18b, and -363 were detected in the limb primordia beginning around stage 14 (Fig. 6A,B,D,F). let-7a, miR-18b, and -363 show expression first in the wing bud and slightly later in both the wing and leg buds (Fig. 6A,D–F). miR-10b showed significantly higher expression levels in the wing versus leg bud at stage 18 that persisted at least through stage 24 (Fig. 6B,C,K,L). In contrast, miR-199a is located in an intron of the Dynamin-1 gene and was expressed strongly in both the wing and leg buds (leg shown, Fig. 6M) and in other mesodermal locations (data not shown). Additional miRNAs detected in limb bud mesenchyme include miR-15a, -18b, -125b (Fig. 6G–J), -128a -130b, and -217 (data not shown). Several of these genes showed localized or enhanced expression in a region of the posterior limb bud approximating the zone of polarizing activity (ZPA). miR-17-5p was detected throughout large regions of the embryo but was excluded from some regions of the ectoderm, and in particular from the AER (Fig. 6J). Conversely, miR-200a and -200b were expressed throughout the ectoderm including the limb bud and AER (see Figs. 1G–J, 6N). Lancman et al. (2005) recently reported that chicken lin-41 orthologue (clin-41) is expressed in the limb buds. In C. elegans, lin-41 is apparently regulated by let-7 and lin-4, and their avian homologues let-7a and miR-125b are co-expressed in chicken limb buds with clin-41 (Lancman et al.,2005). miRNA regulation of lin-41 may be highly conserved from C. elegans to chicken. Sixteen additional miRNAs were detected in the developing limbs (Table 1). These findings represent the first comprehensive screen of miRNA expression in developing limbs.

An overview of miRNA expression patterns highlights several themes. Expression of some miRNAs is restricted to or excluded from specific germ layers and their derivatives. For example, miR-200a and -200b are expressed in the ectoderm and the endoderm but largely excluded from the mesoderm, while miR-30e and 205a are broadly expressed only in the ectoderm. miR-199a appears to be expressed in almost all mesodermal derivatives, though not at the earliest stages following appearance of mesoderm, whereas other miRNAs show localized expression in specific mesoderm derivatives, including the notochord, cardiac and skeletal muscles, vascular endothelial cells, mesonephric ducts, and mesenchyme of the limbs and craniofacial region. In general, many miRNAs are expressed within forming embryonic structures such as the limb buds, pharyngeal arches, and central nervous system, whereas only one to several miRNAs are detected in individual differentiated cell types such as the myocardium (miR-1, -1b, 134a, -367), vascular endothelial cells (-126), or specific subsets of neurons (miR-218, -219).

Four potentially novel chicken miRNAs were identified by homology comparisons with human miRNA gene sequences (see Supplementary Table 2). miR-21, -22, -144, and 363 have not been previously reported in avians, and each was detected by whole mount in situ hybridization. miR-21 is detected in the amnion (see Fig. 1O), miR-22 is ubiquitously expressed, miR-144 is expressed in blood islands (see Fig. 3O), and miR-363 is expressed in multiple tissues including the ectoderm, pharyngeal arches, notochord and limb bud mesenchyme (see Fig. 6F). Expression patterns of miR-21 and miR-144 are conserved between zebrafish and chicken, while miR-21 is expressed in different patterns between the two species. With the rising estimates of the number of miRNAs (Bentwich et al.,2005; Berezikov and Plasterk,2005; Legendre et al.,2005; Xie et al.,2005), it is likely that many novel avian miRNAs remain to be identified. Additional information concerning probe sequences, homology, and similarity can be found in the Supplemental Material.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

Embryo Collection and Preparation

Fertile chicken eggs (HyLine, Iowa; not a commercially available source) were incubated in a forced-draft, humidified incubator at 37.5°C for 0.5 to 5 days, depending on the stages desired. Embryos were collected into chilled chick saline (123 mM NaCl in nanopure water), removed from the vitelline membrane, and cleaned of yolk. Extra-embryonic membranes and large body cavities (brain vesicles, atria, allantois, eye) were opened to minimize trapping of the in situ reagents. Embryos were fixed in fresh, cold 4% paraformaldehyde in PBS over night at 4°C. Embryos were maintained at or near 4°C during collection, because significant loss of labeling was correlated with increased time at room temperature during collection, especially for miRNAs.

Embryos were rinsed in PBS, then in PBS plus 1% Tween-20 (PBT), and dehydrated by steps (25, 50, 75, 100, 100%) into methanol before being cooled to −20°C overnight (or up to 10 days). Rehydration reversed this series. Embryos were rinsed 2× in PBS and older embryos were treated with proteinase K: stages 8–13 and 14–18 at 10 μg/ml of proteinase K for 10 and 20 min, respectively; stages 19 and older at 20 μg/mL of proteinase K for 20 min. Embryos were rinsed repeatedly in PBT to stop the digestion, and were moved into prehybridization (see below). Embryos were stored until use either at the methanol step or in prehybridization at −20°C for fewer than 10 days. Embryos stored for more than 10 days showed a considerable decrease in hybridization signal and increase in background, especially with miRNA.

Probe Preparation

Locked Nucleic Acid modified DNA oligonucleotides (LNAs) complementary to the mature miRNAs were supplied by Exiqon A/S. Digoxigenin-labeled UTP was added to the 3′ end of the LNAs using a DIG Oligonucleotide 3′ end labeling kit (Roche).

In Situ Hybridization

Prepared embryos were transferred to a standard prehybridization solution (50% formamide, 5× SSC, 2% blocking powder, 0.1% Tween-20, 0.1% CHAPS, 50 μg/ml yeast RNA, 5 mM EDTA, 50 μg/ml heparin, DEPC water). Pre-hybridizations were for 2 hr in 24-well plates (1 ml /well) in a shaking hybridization oven at a temperature between 21°C and 23°C below the reported melting temperature of the LNAs. Recent work indicates that up to a 5°C spread in annealing temperature (20–25°C below the melting temperature) is consistent with hybridization of LNA to miRNA (Kloosterman et al.,2006). Probe was added to 1 ml fresh prehyb buffer and hybridization occurred overnight at the prehyb temperature. Embryos were transferred after hybridization to 6- or 12-well plates containing 15- or 24mm Netwell Inserts, respectively, with attached 74-μM polyester mesh bottoms (Corning, Inc., Cat. No. 3477, 3479) in 2× SSC, 0.1% Chaps prewarmed to the hybridization temperature. Inserts helped maximize wash volume and minimize embryo handling and damage for high-throughput screening.

Prewarming the wash solutions to the hybridization temperature before washing was crucial for maximum signal-to-background ratio and is not available using some robots. Embryos in the Netwell inserts could be moved quickly into plates filled with pre-warmed wash buffer, minimizing cooling for high throughput processing. Embryos were washed 3× 20 min in the high salt wash, then 3× 20 min in 0.2× SSC, 0.1% Chaps. Embryos were rinsed twice in KTBT (50 mM Tris, pH 7.5, 150 mM NaCl, 10 mM KCl, 1% Tween-20) and transferred back into clean 24-well plates to minimize volume for the antibody step. Embryos were pretreated in 20% sheep serum in KTBT at 4°C for 2–3 hr or longer. Anti-DIG antibody binding (1:2,000–1:4,000) was carried out in 24-well plates at 4°C on a nutator. Final washes were in KTBT in large Netwell inserts at room temperature for a minimum of 5 changes over 5 hr, but often including overnight at 4°C. Embryos were shifted back to 24-well plates into fresh NTMT (two solutions changes × 10 min; 100 mM NaCl, 100 mM Tris pH 9.5, 50 mM MgCl2, 0.1% Tween-20). Color reactions (NBT/BCIP in NTMT) were for 1–6 hr at room temperature on a nutator until signal or background became visible, followed by overnight washing in KTBT. A second or third round of color reaction followed until each probe had yielded a strong signal, or until the negative control began to show background label. Reactions were stopped with KTBT and embryos were then washed in PBS, dehydrated by a Methanol series to remove background and enhance signal, then rehydrated and stored in PBS plus 0.1% sodium azide.

Imaging and Histology

Embryos were photographed on a Leica PlanApo stereomicroscope using a digital acquisition system and transmitted, lateral and/or direct illumination. Some embryos were dehydrated into methanol (25, 50, 75, 100, 100% in PBS), transferred to xylene (2 changes for 10 min), and embedded in paraplast (Kendall). Sections were cut at 14 μm, mounted using Cytoseal XYL (Richard-Allan Scientific), and photographed using DIC optics on a Leica DMRXE microscope. Images were modified using Adobe Photoshop only to correct brightness, contrast, color balance, and to remove particulates.

Northern Analysis

Heart, limb, head, and trunk tissues were dissected from stage-24 embryos, suspended in Trizol (Invitrogen Corp), and total RNA was isolated by following the manufacturer's protocol. Forty micrograms of total RNA and 5.8 pmol of 22mer DNA control oligos were fractionated by 15% PAGE and transferred to Nytran N+ membrane. Nucleic acids were UV cross-linked to membranes and baked at 80° for 30 min. Blots were pehybridized in 5× SSC, 20 mM Na2HPO4, pH 7.2, 7% SDS, 40 μg/ml yeast tRNA, and 2× Denhardt's solution at 50°C for 2 hr, followed by hybridization overnight in 25 ml of fresh hybridization solution containing 25 pmol of DIG-labeled LNA probe at 50°C. Blots were rinsed 2× in 45 ml of buffer (3× SSC, 25 nM NaH2PO4, pH 7.5, 5% SDS, and 10× Denhardt's solution) at 25°C, followed by 30 min at 50°C, 30 min at 50°C in 1× SSC and 1% SDS, and 2× for 30 min in 0.1× SSC and 0.1% SDS at 65°C. To visualize bound probe, blots were washed briefly with 150 ml of KTBT at room temperature, then in 200 ml blocking solution (5% Skim Milk Powder, 20% sheep serum in KTBT) overnight at 4°C, then incubated in blocking solution containing a 1:2,500 dilution of anti-Digoxigenin-AP Fab Fragment (Roche) for 2 hr at room temperature. Blots were washed 3× at room temperature on a nutator in KTBT, followed by two washes in NTMT. Blots were stained with BOLD APB Chemiluminescent substrate (Molecular Probes) according to the manufacturer's protocol. Chemiluminescent signal was detected by exposing on BioMax Light X-ray film 15 min to 3 hr.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

We thank Brandon Ason, Cynthia Lance-Jones, Drew Noden, and Jennifer LaVail for comments on the data or manuscript, and Sean Davey for database support. We also thank Hy-Line International for supplying fertile chicken eggs. Additional images showing these and other chicken miRNA expression patterns can be viewed at GEISHA, the online repository for chick gene expression data (www.geisha.arizona.edu). This work was supported by NIH R01HD044767 to P.B.A.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS AND DISCUSSION
  5. EXPERIMENTAL PROCEDURES
  6. Acknowledgements
  7. REFERENCES
  8. Supporting Information

The Supplementary Material referred to in this article can be viewed at www.interscience.wiley.com/jpages/1058-8388/suppmat .

FilenameFormatSizeDescription
dvdy20956-DarnellSupportingMaterial.doc240KSupporting Information file dvdy20956-DarnellSupportingMaterial.doc

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.