Cystathionine γ-lyase expression during avian embryogenesis

Abstract

Cystathionine γ-lyase (CSE) is a key enzyme in the trans-sulphuration pathway for the biosynthesis of cysteine from methionine and catalyses the hydrolysis of cystathionine into cysteine. It has been reported to be expressed in mammalian liver and kidney but so far no comprehensive developmental expression analysis of CSE has been available. We cloned a 600 bp fragment of chick CSE cDNA and analysed its expression pattern during avian embryonic development until embryonic day 13. We found CSE expression in various developing organs including the notochord, eye, neural tube, limb bud mesenchyme and sclerotomal compartment of the somites. Notably, prominent expression was found in renal epithelia throughout kidney development, i.e. in the tubular structures of pronephros, mesonephros and metanephros. Our data introduce CSE as a novel marker gene to study avian kidney development.

Introduction

Cystathionine γ-lyase (CSE) (CGL, γ-cystathionase; IUBMB Enzyme Nomenclature EC 4.4.1.1) is a key enzyme of the trans-sulphuration pathway, which is essential for an adequate supply of the amino acid cysteine. CSE activity depends on pyridoxal 5′-phosphate and catalyses the final step of cysteine biosynthesis, which is the conversion of L-cystathionine into L-cysteine, α-ketobutyrate and ammonia. CSE is further involved in the synthesis of hydrogen sulphide (Yamanishi & Tuboi 1981; Stipanuk & Beck 1982), which acts as a gaseous neuromodulator (Abe & Kimura 1996) and smooth-muscle relaxant (Hosoki et al. 1997; Zhao et al. 2001; Teague et al. 2002).

The CSE gene has been characterized in many organisms including human (Lu et al. 1992; Levonen et al. 2000), amphibian (Pong et al. 2007), mouse (Ishii et al. 2004), rat (Erickson et al. 1990) and the plant Nicotiana tabacum (Clausen et al. 1999), showing high sequence identity between phylogenetically distant organisms, which indicates the evolutionary conservation of this enzyme. In humans, two isoforms of CSE are known, a longer form and a shorter form with an internal deletion of 132 bp, which is probably due to alternative splicing (Levonen et al. 2000).

Expression data from human, rat and mouse suggest a potential role of CSE during embryonic development (De Luca et al. 1974; Levonen et al. 2000; Ishii et al. 2004). CSE expression in birds has not previously been studied. In mouse and rat, CSE is expressed at high levels in the developing kidney and liver, and only weakly in other tissues like the gut. This distribution of transcripts also persists in adult rodents (Ishii et al. 2004). However, in these studies only selected organs were analysed by northern and western blot analyses. To date, a comprehensive expression study using whole-mount in-situ data is not available in any model organism.

In this study, we give the first comprehensive description of CSE mRNA expression during embryogenesis, using the chicken embryo as a model organism. We identified a CSE homologue in an avian cDNA library and examined CSE gene expression during chicken embryogenesis from Hamburger and Hamilton (HH) stage 1 up to 14 days of incubation. We found that, after an initial salt-and-pepper expression in the epiblast during gastrulation, in higher developmental stages CSE is distinctly expressed in lens, retina, notochord, neural tube, limb bud mesenchyme and sclerotome. Moreover, we observed prominent expression of CSE in the developing kidney. Our results provide a basis for further functional studies on the role of CSE during avian embryogenesis.

Material and methods

Preparation of chick embryos

Fertilized White Leghorn eggs (Gallus gallus) were incubated at 37.8 °C and 80% relative humidity. The developing chick embryos were staged according to Hamburger & Hamilton (1951).

Generation of the avian cystathionine γ-lyase probe

We isolated the CSE clone out of a subtractive avian cDNA library. The cDNA was made using the Super SMART™ cDNA synthesis kit and BD PCR select™ cDNA subtraction kit according to the protocol provided by Clontech/BD Science. The 608 bp cDNA was cloned into pDrive cloning vector (QIAGEN PCR cloning kit, cat. no. 231122). Sequence analyses were performed by the MWG Biotech Company. Comparison of the sequence was performed by BLAST search, made available by the NCBI database (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Accordingly, this fragment spanned the region between bp 911 and bp 1449 (coding sequence to bp 1292) of the GenBank sequence for chick CSE (XM_422542.2), revealing 100% sequence identity with the published sequence.

Digoxygenin-labelled sense riboprobes were made after restriction of the plasmid DNA with NotI and using Sp6 polymerase, for antisense probes after restriction with HindIII using T7 polymerase. The labelling procedure was performed according to the standard protocol.

In-situ hybridization

Chick embryos from HH stage 1 to day 14 were fixed overnight at 4 °C in 4% paraformaldehyde for in-situ hybridizations on whole-mount embryos or cryosections, or in RNase-free Serra solution (Serra, 1946) for in-situ hybridizations on paraffin sections. The embryos were washed twice in PBT/PBS, dehydrated in methanol and stored at 4 °C. Whole-mount in-situ hybridization was performed as previously described by Nieto et al. (1996) and in-situ hybridizations on paraffin sections were made according to Brand-Saberi et al. (1996). Selected stained embryos at stages 4–25 and single organs were embedded in 5% agar and sectioned with a Leica vibratome at 55 µm; older stages (HH stage 28 and older) were sectioned at 100 µm. Cryosections were performed at 25 µm using a Jung and Reichert cryotome and paraffin sections were performed at 8 and 12 µm using a Jung and Reichert microtome.

Results

A clone of 608 bp in length was isolated from a chick cDNA library and subcloned into pDrive for sequence analyses and generation of probes for in-situ hybridization. Sequence analyses revealed this clone to be identical with the coding sequence (1199 bp) of chick CSE (NCBI accession no. XM_422542). Comparison of the CSE amino acid sequences among various species revealed a high degree of conservation (Fig. 1), indicating a high functional relevance of CSE throughout evolution.

Fig. 1.

Sequence comparison of cystathionine γ-lyase (CSE) protein derived from different species (chick, human, macaque, mouse, rat, zebrafish and fruit fly). The accession numbers provided by the NCBI database are given on the left. Alignment was performed using the Jotun Hein method of the MegAlign program provided by Lasergene. Identical amino acids are shaded in grey.

We observed faint CSE expression as early as HH stage 1, which corresponds to pregastrula stages shortly after laying, in the prestreak blastoderm (data not shown).

At HH stage 4, which corresponds to the definitive streak stage, CSE expression was seen in the gastrulating embryo in a salt-and-pepper pattern, staining being strongest in the epiblast (Fig. 2A,B).

Fig. 2.

Cystathionine γ-lyase (CSE) expression during chick embryogenesis revealed by in-situ hybridization. (A) Hamburger and Hamilton (HH) stage 4, dorsal view on a gastrulating embryo at definitive streak stage, cranial is to the top (arrow indicates Hensen's node). (B) HH stage 4, transverse section through primitive streak (arrowhead points at primitive groove) showing CSE expression in salt-and-pepper pattern predominantly in the epiblast (arrow). (C) HH stage 8, embryo with four pairs of somites in dorsal view, cranial is to the top (arrow points at the neural folds of the prospective forebrain). (D) HH stage 11 showing CSE expression in the neural tube and brain vesicles, cranial is to the top. (E) HH stage 14, dorsal view of the caudal portion of an embryo, cranial is to the top. CSE expression is restricted to the notochord (arrowhead), pronephric duct (asterisk) and Wolffian duct (arrow). (F) HH stage 14, transverse section showing expression in the central cells of the notochord (arrowhead) and mesonephric ducts (arrows). (G) HH stage 16, transverse section at the level of epithelial somites, illustrating that notochordal expression is strong in the central cells (arrow) but only weakly detectable in the marginal cells of the notochord. Prospective sclerotome cells in the ventral somitic wall show weak expression (arrowhead). (H) HH stage 19, expression is visible in the somites, ear placode, limb buds and mesonephros (arrow). (I) HH stage 19, transverse section. Expression in the mesonephros (arrowhead), the entire diameter of the notochord and in the sclerotomal compartment of the somites (arrow). (J) HH stage 19, section through the eye. CSE expression in the lens and outer layer of the retina (arrow). (K) HH stage 19, transverse section through the mesonephros, showing expression in the mesonephric glomerulus. (L) HH stage 23, CSE expression in the distal limb mesenchyme, mesonephros (arrow), branchial arches, eye and brain vesicles. (M) HH stage 23, longitudinal section through a wing bud. Expression in the distal limb bud mesenchyme but not in the superficial mesenchyme underlying the ectoderm, and in the ectoderm including the apical ectodermal ridge (AER) (arrowhead). (N) HH stage 25, wing bud has been removed to better reveal expression in the mesonephros (arrow). Expression in the limbs has ceased. (O) HH stage 27, transverse section showing expression in the mesonephric tubules (arrow). (P) HH stage 37, whole-mount preparation of an embryonic kidney in dorsal view. The dorsally located metanephros (arrow) is not stained, whereas the underlying mesonephros shows strong CSE expression. (Q) HH stage 37, whole-mount preparation of an embryonic kidney in ventral view. The mesonephros is intensely labelled (arrow), whereas the gonads (arrowhead) are devoid of expression. (R) Transverse section of a HH stage 37 mesonephros showing intensely labelled mesonephric tubuli.

At HH stage 8, expression was observed in the neural folds, most prominently in the prospective brain region (Fig. 2C), and persisted in the neural tube and brain vesicles at HH stage 11 (Fig. 2D).

From HH stage 12 to HH stage 16, prominent expression was observed in the central cells of the notochord but not in the outer cells in contact with the notochordal sheath (Fig. 2E–G). In the kidney anlagen, the nephric duct expressed high levels of CSE in both the pronephric and mesonephric (Wolffian duct) generation (Fig. 2E,F). Interestingly, notochordal and nephric duct expression both started in the same craniocaudal segment, at the level of the fifth somite.

At HH stage 19, CSEwas strongly expressed in the entire diameter of the notochord, without a difference between inner and outer notochordal cells (Fig. 2I). In the mesonephros, robust expression was detected in the Wolffian duct and forming mesonephric tubules (Fig. 2H,I,K). Moreover, at lower level, expression was detected in the sclerotome of the somites, mesenchyme of the limb buds, neural tube, lens epithelium, aortic wall and outer cell layers of the retina (Fig. 2H–J).

Notochordal expression was downregulated starting at around HH stage 20 (not shown).

At HH stage 23 CSE expression was found in the brain vesicles, branchial arches and limb buds (Fig. 2L). Expression in the limbs was detectable in the central limb mesenchyme and in the ectoderm including the apical ectodermal ridge (AER), whereas the peripheral limb mesenchyme underlying the ectoderm was devoid of CSE transcripts (Fig. 2M). In the developing kidney, the mesonephros was strongly labelled (Fig. 2L). Moreover, slight expression was detectable in the neural tube and dorsal root ganglia (not shown).

At HH stage 25, strong expression persisted in the mesonephros as well as faint expression in the sclerotome and dorsal root ganglia (not shown), whereas expression in the limb buds was beyond detection (Fig. 2N).

At HH stage 27, there were high levels of transcripts in the mesonephric tubules (Fig. 2O), together with further domains in the neural tube, dorsal root ganglia, intestine and liver.

In subsequent stages, expression persisted at low levels in all of the regions listed above. In contrast, strong and distinct expression was present in the embryonic kidney until the last stages examined, corresponding to HH stage 39. In the kidney of HH stage 37 embryos, at 11.5 days of development, the mesonephros is covered dorsally by the developing metanephric tissue. At this stage, the mesonephros showed high levels of CSE expression in the tubular epithelia, whereas the metanephros showed only faint expression in the newly formed metanephric tubuli. The ventrally abutting gonads were devoid of expression (Fig. 2P–R). At HH stage 39, with the increasing amount of metanephric tubuli, the metanephros showed similarly intense expression as the mesonephric kidney (not shown).

In summary, we present a comprehensive analysis of CSE expression during embryonic development. In the avian embryo, the most conspicuous expression domains are the expression in the notochord and kidney. In the notochord, expression is highly specific regarding both the narrow time window and the distribution in cortical and medullary cells. In the kidney, we found prominent expression in the pronephric, mesonephric and metanephric kidney generations, which is limited to the tubular components of the developing excretory system.

Discussion

In this study, we present the first comprehensive analysis of CSE expression during embryonic development and the first data on CSE expression in the chick embryo. We found CSE expression in a number of diverse embryonic organs, most of which had not previously been known to express CSE, like notochord, sclerotome, limb bud mesenchyme and neural tissue.

The striking expression pattern of CSE in the avian kidney as reported here is congruent with data from mammals. Ishii et al. (2004) detected a low level of CSE activity in mouse liver and kidney at embryonic day 12.5. In adult murine kidneys, they observed stronger CSE expression in the renal cortex than in the medulla, especially in the renal tubules of the inner cortex. In the rat, House et al. (1997) described CSE enrichment in the inner cortex and outer medulla, with strong evidence of an enrichment in cells of the proximal straight tubule. Moreover, CSE levels seem to vary according to the physiological situation, as Akahoshi et al. (2006) have shown variable CSE expression in the kidneys of mouse dams during gestation and lactation. These data are difficult to correlate with our findings in the chick because the avian kidney is organized differently from the mammalian situation, as for instance the medulla and cortex are not strictly centripetally separated but rather intermingled (Hamilton 1952).

The retinal expression observed here is in line with immunohistochemical data in salamander detecting CSE protein in retinal Müller cells, whereas, in contrast, mouse retina did not show CSE protein (Pong et al. 2007). Interestingly, in rat lenses the absence of CSE due to oxidative stress has been correlated with cataract formation (Sastre et al. 2005).

The function of CSE during development is not known. Unfortunately there are no CSE knockout mice available to investigate possible renal defects in the absence of CSE. In-vitro studies using HEK-293 cells or human aortic smooth muscle cells to overexpress CSE resulted in ERK/MAPK activation, downregulation of cyclin D1 expression and increased hydrogen sulphide production rates. This resulted in inhibition of cell proliferation and DNA synthesis, and increased apoptosis (Yang et al. 2004, 2006).

Our expression data introduce CSE as a marker for epithelial tubules of all kidney generations, including pronephros, mesonephros and metanephros. The function of CSE in kidney development remains elusive. Intriguingly, in other contexts like the sclerotome or limb mesenchyme, CSE is also expressed in mesenchymal cells, thus excluding a potential general role in epithelia formation. Functional studies in vivo, including misexpression experiments in the chick, will be needed to elucidate the role of CSE during embryonic development.