HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Layalle, S. Articles by Dambly-Chaudière, C. Search for Related Content PubMed PubMed Citation Articles by Layalle, S. Articles by Dambly-Chaudière, C.

Laboratoire de Neurogénétique, INSERM E343, Cc 103, Université Montpellier II, Place E.Bataillon, 34095 Montpellier, France

	Abstract

Top Abstract Introduction Results Discussion Experimental procedures References

 
We identified the gene smooth (sm) in a screen for genes that are specifically expressed within the lineage that forms the adult chemosensory bristles. sm is expressed in most or all differentiating neurones during embryogenesis, but is specifically expressed in the neurones of the adult chemosensory organs on the wings and legs during metamorphosis. The inactivation of sm results in axonal defects in the chemosensory neurones, in the inability of mutant flies to feed and in their precocious death. As sm belongs to a family of heterogeneous nuclear ribonucleoprotein (hnRNP), we propose that the control of axonal navigation and connectivity is partly achieved at the level of mRNA splicing or exporting.

	Introduction

Top Abstract Introduction Results Discussion Experimental procedures References

 
Two types of sensory bristles are found on the wings and legs of Drosophila, the mechanosensory bristles, which mediate touch, and the chemosensory bristles, which mediate taste. Mechanosensory bristles are innervated by a single neurone, the dendrite of which contacts the base of the shaft and responds to shaft deflection (tactile stimulus). Chemosensory organs are innervated by five neurones, four of which are chemosensory neurones and extend their dendrites inside the shaft up to its pored tip where they are exposed to various stimuli like sugars, water and amino acids (Morita & Yamashita 1959; Pollack & Balakrishnan 1997). The fifth neurone connects to the base of the shaft and is supposed to be mechanosensory (Nayak & Singh 1983; Nottebohm et al. 1994).

Both mechano- and chemosensory neurones of a given leg extend their axons along the leg nerve and into the central nervous system where they establish characteristic axonal projections in the corresponding leg neuromere of the thoracico-abdominal ganglion (Murphey et al. 1989a). Mechanosensory axons extend in the periphery of the neuromere; the terminal arbor is not branched and stays in the same plane as the nerve root (Murphey et al. 1989b). Chemosensory axons follow a more medial path along the ventral side of the neuromere and penetrate deeply in the neuromere where they establish a highly branched terminal arbor (Nottebohm et al. 1992).

The differences between mechano- and chemo-sensory organs depend on the gene pox-neuro (poxn), which encodes a paired-box transcriptional factor (Bopp et al. 1989; Dambly-Chaudière et al. 1992). poxn is expressed in the chemosensory precursor cells and in their progeny and acts as a selector gene for the chemosensory cell. Studies of gain- and loss-of-function combinations have shown that poxn is responsible for all differences between the mechano- and chemosensory organs that have been studied so far (external morphology, lineage, innervation, central projection) (Nottebohm et al. 1994; Awasaki & Kimura 1997).

In order to better understand how poxn controls the various aspects of chemosensory organ development, we performed a differential screening to recover effector genes that presumably act downstream of poxn to implement the chemosensory ‘programme.’ The screening was carried out on a subtractive library built with RNA from a poxn gain-of-function line (hsp-poxn) from which we have subtracted RNA from the poxn loss-of-function mutant line (poxn⁷⁰) (Layalle et al. 2004). The recovered genes are likely to be expressed in a poxn dependant way, and are therefore considered as putative effector genes.

One of the candidates that we recovered is the gene smooth (sm). sm has been isolated by Mackay (Mackay 1985), in a genetic screen for quantitative trait loci (QTL) affecting bristle number. Adult flies homozygous for the original sm allele have an average of two abdominal bristles per abdominal sternite, when wild-type flies have between 17 and 20 sternital bristles. A molecular analysis revealed that the sm gene encodes a protein with a high level of similarity with RNA-binding proteins (zur Lage et al. 1997). The closest human relative to sm is the gene coding for the heterogeneous nuclear ribonucleoprotein L (hnRNP L).

Here we focus on the role of sm in the development of chemosensory organs. We show that sm is specifically expressed in chemosensory neurones in the embryo and pupa, and that its loss of function impairs feeding. The analysis of chemosensory neurones in sm mutants revealed no defects in the dendrites, but showed that the axons establish an inappropriate central projection.

	Results

Top Abstract Introduction Results Discussion Experimental procedures References

 
Expression pattern of sm during development

We analysed the pattern of expression of sm by in situ hybridization throughout development. sm is not expressed until late stages of embryogenesis, when transcripts are observed in most or all neurones of the CNS (Fig. 1A,B) and in a subset of chordotonal neurones (Fig. 1B) as well as, at a lower level, in other sensory neurones. No other tissue or structure is labelled during embryogenesis.

View larger version (97K): [in this window] [in a new window]   Figure 1  Pattern of expression of sm as revealed by in situ hybridization. (A–B) During late embryogenesis, (C–E) Early during metamorphosis. (A) At about 16 h after egg laying, sm is expressed in most or all CNS neurones. (B) At the same stage, sm is also expressed in the PNS, most notably in the chordotonal neurones (arrows) and at a lower level in other sensory neurones (arrowhead). (C–E) At 20–24H after puparium formation (APF), sm is expressed in clusters of cells in the anterior wing margin and tibia. (C,D) In the wing, sm is expressed in regularly spaced groups of cells (arrowheads) in a pattern that corresponds to that of adult chemosensory organs. (E) In the leg, sm is also expressed in regularly spaced clusters of cells (arrowheads). (F) The sensory neurones of 20–24H APF sm pupae immunolabelled with the 22C10 antibody. The morphology and the number of chemosensory neurones (arrowheads) and of mechanosensory neurones (arrows) are as in the wild-type. The clusters of chemosensory neurones labelled by 22C10 are similar in position and number to the clusters of sm expression. The mechanosensory neurones do not appear to express sm.

In summary, the position and number of the labelled cells along the anterior wing margin, tibia and tarsus suggests that the gene sm is expressed in the differentiating adult chemosensory neurones but not in the mechanosensory neurones.

sm mutant flies have morphologically normal sense organs

The original sm allele (Bridges & Brehme 1944) has become fully lethal, presumably owing to the acquisition of one or more secondary mutations. A P-induced allele, sm³, was recovered as a quantitative trait locus affecting bristle number on the abdomen and allowed the cloning of the gene (zur Lage et al. 1997). We have used a second P-element insertion (Karpen & Spradling 1992), sm⁴, where a PlacZ transposon has inserted in exon 1 of the gene (zur Lage et al. 1997). sm⁴ shows the same loss of abdominal bristles as the other alleles, to the exception of the weaker allele sm².

Since the only known phenotype of this mutation is an absence of sternital bristles, we examined the pattern of mechano- and chemosensory bristles on wing and leg of mutant flies.Both chemo- and mechanosensory bristles are present in normal numbers and patterns and present a normal morphology. This was not unexpected since sm is apparently not expressed in the sensory precursor cells, nor in the support cells that will form the external structures of the organs. Given that sm is expressed during the differentiation of the chemosensory neurones, we investigated the neuronal component of the adult sensory organs by immunolabelling with the neuronal antigen, 22C10. We observed that at 20–24H APF, the chemosensory neurones are present in normal numbers and shapes, and the mechanosensory neurones are also normal (Fig. 1F).

Behavioural analysis of sm mutant flies

We observed that homozygous sm adults do not survive more than 2–4 days, a very reduced life expectancy given the average of 1–2 months for normal adult flies. This observation led us to probe for behavioural defects in the mutants, that might impair feeding and therefore survival.

We submitted the sm mutants to a simple nutrition test by incubating wild-type and mutant flies on food media containing methylene blue. Flies normally accumulate food in the crop, a gut diverticulum where food is stored and progressively returned to the gut for digestion. After one night (16 h) on colored food, wild-type flies show a bright blue crop whereas the sm mutants remain uncolored (Fig. 2). We conclude that sm flies do not feed properly.

View larger version (114K): [in this window] [in a new window]   Figure 2  Nutrition test. Wild-type and sm mutant flies were maintained overnight on food stained with methylene blue. (A) Wild-type flies show a blue coloration in their crop, indicating a normal feeding behaviour whereas (B) sm mutant remain uncolored, suggesting that they do not feed.

The observation that sm mutants do not eat led us to examine more closely the chemosensory neurones in the mutant. We first used retrograde DiI labelling to examine the dendrites (see Experimental procedures). We observed that several neurones underlie each chemosensory bristle, as already indicated by the 22C10 immunolabelling, and that they extend their dendrites properly into the shaft (Fig. 3).

View larger version (125K): [in this window] [in a new window]   Figure 3  Sensory dendrites in sm mutants. Sensory neurones on the tibia of wild-type flies (A) and of adult sm4 mutants (B) were labelled with DiI. The tip of the mechanosensory dendrites (arrowheads) terminate at the base of the shaft and never enter it, while the dendrites of the chemosensory neurones extend into the thin, recurved shaft of the gustatory bristles (arrows). The figures were mounted from Z-series (step = 3 microns) taken with a Nipkow disk confocal attachment; this explains the interrupted aspect of the chemosensory dendrites in (B), where the interruptions correspond to the gap between consecutive confocal images. The chemosensory bristle seen in (A) is in a more lateral position, hence the presence of the nerve in the field, whereas the bristle in (B) is closer to the upper surface of the tibia, hence the slightly different aspect of both chemosensory and mechanosensory dendrites.

View larger version (53K): [in this window] [in a new window]   Figure 4  Central projections of mechano- vs. chemo-sensory bristles. (A) Typical projections of a chemosensory bristle (c) and of a posterior mechanosensory bristle (m) within the thoracico-abdominal ganglion (dotted outline), as seen from the ventral side. The outlines of the prothoracic, mesothoracic and metathoracic leg neuromeres are represented by the dashed lines. Also shown are the wing and abdominal neuromeres. In order to provide a realistic drawing, the outline of the ganglion and neuromeres was obtained from a dissected adult ganglion labelled with the neuropil marker 1E8 (Teugels & Ghysen 1983). The two projections are camera lucida drawings based on two peroxidase-filled neurones. (B) Transformed chemosensory projection in a sm mutant (also shown at a higher resolution in Figure 5H). Scale is approximately the same as for the drawing in panel A. The white dots outline the prothoracic and mesothoracic leg neuromeres. Arrows: mesothoracic leg nerve.

View larger version (64K): [in this window] [in a new window]   Figure 5  Central projections of adult sensory neurones. (A,B) Two examples of projections from leg chemosensory neurones. In (A) two neurones are labelled as evidenced by the presence of two axons. The axons establish an extended terminal arbor deep inside the neuromere. (C,D) Two examples of projections from mechanosensory neurones. The axons establish a relatively simple, flat terminal arbor. (E–H) Projections of chemosensory neurones in sm mutant flies. The axons runs in an intermediate or lateral position and establish a simple terminal arbor. (E,F) Two examples of a reduced arborization. (G,H) Two examples of a mechanosensory-like arborization. In (F), * indicates the position where the projection was broken due to minor damage during the dissection. In all panels the stereo views will provide a view as seen from the ventral side if seen with uncrossed eyes, or from the dorsal side if seen with crossed eyes.

	Discussion

Top Abstract Introduction Results Discussion Experimental procedures References

 
sm is expressed in differentiating chemosensory neurones

During embryogenesis, sm is much more strongly expressed in differentiating neurones than in any other tissue. This is similar to the murine homolog of sm, which is widely expressed in mouse tissues but shows a higher expression in the brain (Kamma et al. 1995). Once the differentiation of larval neurones is achieved, sm is not expressed anymore either during larval life or in the first stage of metamorphosis, during the fixed divisions that will generate the adult sensory organs. Expression resumes in chemosensory neurones at the time when they have all been generated and begin to differentiate. In view of this pattern of expression, the overt phenotype of the mutant, a decrease in the number of abdominal sensory bristles, is surprising. As the abdominal segments are derived from inconspicuous masses of cells rather than from well-defined imaginal discs, we have not been able to document the pattern of expression of sm in this tissue, nor to define what step in the development of these abdominal organs is affected.

sm is an effector of poxn that is involved in defining the axonal behaviour

We recovered sm in a screen for genes that are preferentially expressed in leg and wing tissues that over-express the transcriptional regulator poxn. We considered therefore sm as a putative effector of poxn. Our observation that sm is specifically expressed in the chemosensory neurones on the wings and legs is entirely consistent with this role. Of course we do not know whether the control of poxn on sm is direct or indirect, yet the pattern of expression of sm suggests a role in the identity or differentiation of adult chemosensory neurones.

The behavioural and neuronal phenotypes of sm mutants show that this gene is indeed required for the proper differentiation of chemosensory neurones. Interestingly, however, the mutant shows no morphological defect at the level of the dendrites. The difference between chemosensory and mechanosensory dendrites is striking: the former do extend up to the tip of the shaft, while the latter do not enter the shaft. Because we did not observe any indication of dendritic defect neither after immunolabelling with 22C10 nor after labelling the neurones with DiI, we believe that sm is specifically required for the proper development of the chemosensory axons.

sm may act through RNA splicing or export control

Neurone-specific splice variants are known for a large number of genes. Since Sm is a protein of the heterogeneous nuclear ribonucleoprotein (hnRNP) family, it might act at the level of splicing specificity. The closest human homolog to the Drosophila smooth gene is the hnRNP L (zur Lage et al. 1997). hnRNP L contains two RNA recognition motifs (RRMs) and is a component of a large family of proteins found associated with premRNA (Pinol-Roma et al. 1989; Matunis et al. 1992; Dreyfuss et al. 1993). hnRNP proteins are thought to be implicated in a large number of RNA packaging and processing operations like splicing control, RNA stability and even nucleo-cytoplasmic transport (Pinol-Roma & Dreyfuss 1991, 1992). hnRNP L can bind to a polypyrimidine-rich region of the mRNA of the human vascular endothelial growth factor (VEGF), leading to mRNA stabilization (Shih et al. 1999). Another recent study shows that hnRNP L recognizes CA repeats and specifically activates splicing of the human endothelial nitric oxide synthase (eNos) premRNA (Hui et al. 2003).

It may be, therefore, that sm is responsible for editing a subset of premRNAs in a manner that makes them functional to define specific aspects of axonal navigation and/or connectivity. Since the expression of the gene poxn is known to convert mechano- into chemosensory organs, it seems likely that the expression of sm likewise converts the properties of axons from the default state found in mechanosensory neurones to the alternative state proper to chemosensory neurones. Our observations on DiI-labelled neurones support this idea, as the mutant projections are clearly transformed, albeit not always completely, into mechanosensory projections.

Alternatively, we cannot exclude that the effect of sm is not on the splicing but on the addressing of mRNAs. sm would then be specifically in charge of mRNA exporting towards the axonal compartment and its absence would alter the delivery of chemosensory-specific mRNAs to the mutant axons. Whatever the detailed mechanism of sm action, our results suggest that the control of neuronal development may be segmented and that the control of dendritic and of axonal properties may be partly separate.

	Experimental procedures

Top Abstract Introduction Results Discussion Experimental procedures References

 
Drosophila strains

Flies were raised on standard medium. The wild-type is Oregon R strain and is maintained at 25 °C. The smooth mutant line (sm⁴) was obtained from the Bloomington Drosophila Stock Center (Bloomington, IN, USA) and is raised at 18 °C.

Immunostaining

Pupal tissues were isolated and prepared as described in Layalle et al. (2004). 22C10 antibody (DBSH, 1 : 500) and anti-mouse antibody conjugated to peroxidase (Sigma, 1 : 800) were used. Peroxidase activity was revealed by incubation in DAB and urea hydrogen peroxide (according to Sigma advice), and tissues were mounted in glycerol 80% for microscopic observation.

In situ hybridization

20–24H APF pupae were fixed overnight at 4 °C in 4% paraformaldehyde/0.1% Tween 20/PBS. Pupal legs and wings were dissected and washed in PBS/0.1% Tween 20 (PBTw). Pupal tissue was incubated in ammonium acetate (300 mM)/ethanol (1 : 1), washed in ethanol for 10 min, incubated 10 min in a solution of xylene/ethanol (1 : 1), washed in ethanol three times and finally shifted in methanol. Tissues were fixed in methanol/4% paraformaldehyde in PBTw (1 : 1) for 15 min and washed in PBTw extensively. Pupal tissues were blocked in prehybridization buffer (50% formamide, 5x SSC, tRNA 100 µg/mL, heparin 50 µg/mL, 0.1% Tween 20 at pH 4.5) for at least 1 h at 55 °C. Hybridization was performed at 55 °C overnight. Digoxigenin-labelled probes for RNA in situ hybridization were prepared using the cDNA clone GH05823 (from the Berkeley Drosophila Genome Project, BDGP) corresponding to the gene smooth (CG9218). Signal was detected using an anti-DIG antibody conjugated to the alkaline phosphatase (1 : 1000) revealed with NBT/BCIP (Roche). Pupal tissues were mounted in 80% glycerol and analysed under conventional microscope (Nikon Microphot).

Behavioural nutrition test

Flies were incubated overnight on instant fly food medium supplemented with a 0.03% methylene blue solution (pH 5.6 in 0.3 M acetate). Oregon and sm mutants were observed under a dissection microscope and images were taken with a Nikon Coolpix4500 camera.

Labelling of sensory neurones

Chemosensory neurones of the tibia were stained with DiI C12, a lipophilic fluorescent dye (Molecular probes, Inc.). Thorax were fixed overnight at 4 °C in 3.7% formaldehyde solution in 0.1 M sodium carbonate buffer, pH 9.5. For labelling the cell bodies and dendrites, legs were cut at the level of the coxa, the dye (dissolved in 95% ethanol) was applied to the cut and the legs were returned to the fixative for 2 days. For labelling axonal projection, chemosensory bristles were cut and the dye was applied to the stump with a micropipette using a micromanipulator and a fixed stage microscope. This procedure usually results in the labelling of 1–2 neurones, presumably because the cut does not result in an equal exposition of the four chemosensory dendrites. Mechanosensory projections were labelled by dislodging the bristles from their socket and applying dye to the socket as for the chemosensory bristles. Staining was allowed to proceed for 2 days at 25 °C in sodium carbonate buffer. Central nervous systems were then dissected and mounted in glycerol 80%. Observation of the dendrites was done on a Zeiss microscope equipped with a CARV Nipkow disk confocal attachment. Images were taken with a Coolsnap HQ camera. Observation of the axonal projections was performed using a rhodamine filter set on a Nikon microscope, and images were taken with a Princeton Pentamax camera. In both cases Z-series were taken using an Advanced Scientific Instrument Z-drive and a Vincent Associates shutter, both running under IPLab 3.6 software. Image sequences were deconvolved using IPLab 3.9 rapid deconvolution procedure. In the case of dendrites the different planes of focus were asembled using Adobe Photoshop 7.0; in the case of axonal projections brightest-pixel projections were computed with a tilt to the left and to the right to provide stereo pictures.

	Acknowledgements

 
We are grateful to G. Ragone and A. Giangrande for help in the construction of the subtractive library, to N. Cubedo for stock maintenance, and to J. van Helden and R. Vidaud for help with the imaging set up. This work was supported by a grant to AG from the Association pour la recherche sur le Cancer (ARC, France).

	Footnotes

 
Communicated by: Lily Yeh Jan

^* Correspondence: E-mail: alain.ghysen{at}univ-montp2.fr

	References

Top Abstract Introduction Results Discussion Experimental procedures References

 
Awasaki, T. & Kimura, K. (1997) pox-neuro is required for development of chemosensory bristles in Drosophila. J. Neurobiol. 32, 707–721.[CrossRef][Medline]

Bopp, D., Jamet, E., Baumgartner, S., Burri, M. & Noll, M. (1989) Isolation of two tissue-specific Drosophila paired box genes, Pox meso and Pox neuro. EMBO J. 8, 3447–3457.[Medline]

Bridges, C.B. & Brehme, K.S. (1944) The mutants of Drsosophila. Washington DC: Carnegie Institute of Washington, Monograph, 522, 257 p.

Dambly-Chaudière, C., Jamet, E., Burri, M., et al. (1992) The paired box gene pox neuro: a determinant of poly-innervated sense organs in Drosophila. Cell 69, 159–172.[CrossRef][Medline]

Dreyfuss, G., Matunis, M.J., Pinol-Roma, S. & Burd, C.G. (1993) hnRNP proteins and the biogenesis of mRNA. Annu. Rev. Biochem. 62, 289–321.[CrossRef][Medline]

Hui, J., Stangl, K., Lane, W.S. & Bindereif, A. (2003) HnRNP L stimulates splicing of the eNOS gene by binding to variable-length CA repeats. Nature Struct. Biol. 10, 33–37.[CrossRef][Medline]

Kamma, H., Portman, D.S. & Dreyfuss, G. (1995) Cell type-specific expression of hnRNP proteins. Exp. Cell Res. 221, 187–196.[CrossRef][Medline]

Karpen, G.H. & Spradling, A.C. (1992) Analysis of subtelomeric heterochromatin in the Drosophila minichromosome Dp1187 by single P element insertional mutagenesis. Genetics 132, 737–753.[Abstract]

zur Lage, P., Shrimpton, A.D., Flavell, A.J., Mackay, T.F. & Brown, A.J. (1997) Genetic and molecular analysis of smooth, a quantitative trait locus affecting bristle number in Drosophila melanogaster. Genetics 146, 607–618.[Abstract]

Layalle, S., Ragone, G., Giangrande, A., Ghysen, A. & Dambly-Chaudière, C. (2004) Control of bract formation in Drosophila: poxn, kek1, and the EGF-R pathway. Genesis 39, 244–256.

Mackay, T.F. (1985) Transposable element-induced response to artificial selection in Drosophila melanogaster. Genetics 111, 351–374.[Abstract/Free Full Text]

Matunis, E.L., Matunis, M.J. & Dreyfuss, G. (1992) Characterization of the major hnRNP proteins from Drosophila melanogaster. J. Cell Biol. 116, 257–269.[Abstract/Free Full Text]

Morita, H. & Yamashita, S. (1959) Generator potential of insect chemoreceptor. Science 130, 922.[Abstract/Free Full Text]

Murphey, R.K., Possidente, D., Pollack, G. & Merritt, D.J. (1989a) Modality-specific axonal projections in the CNS of the flies Phormia and Drosophila. J. Comp. Neurol. 290, 185–200.[CrossRef][Medline]

Murphey, R.K., Possidente, D.R., Vandervorst, P. & Ghysen, A. (1989b) Compartments and the topography of leg afferent projections in Drosophila. J. Neurosci. 9, 3209–3217.[Abstract]

Nayak, S.V. & Singh, S.N. (1983) Sensilla on the tarsal segments and mouthparts of adult Drosophila melanogaster Meigen (diptera: Drosophilidae). Int. J. Insect Morph. Embryol. 12, 273–291.[CrossRef]

Nottebohm, E., Dambly-Chaudière, C. & Ghysen, A. (1992) Connectivity of chemosensory neurons is controlled by the gene. Poxn. Drosophila. Nature 359, 829–832.

Nottebohm, E., Usui, A., Therianos, S., Kimura, K., Dambly-Chaudière, C. & Ghysen, A. (1994) The gene poxn controls different steps of the formation of chemosensory organs in Drosophila. Neuron 12, 25–34.[CrossRef][Medline]

Pinol-Roma, S., Swanson, M.S., Gall, J.G. & Dreyfuss, G. (1989) A novel heterogeneous nuclear RNP protein with a unique distribution on nascent transcripts. J. Cell Biol. 109, 2575–2587.[Abstract/Free Full Text]

Pinol-Roma, S. & Dreyfuss, G. (1991) Transcription-dependent and transcription-independent nuclear transport of hnRNP proteins. Science 253, 312–314.[Abstract/Free Full Text]

Pinol-Roma, S. & Dreyfuss, G. (1992) Shuttling of pre-mRNA binding proteins between nucleus and cytoplasm. Nature 355, 730–732.[CrossRef][Medline]

Pollack, G.S. & Balakrishnan, R. (1997) Taste sensilla of flies: function, central neuronal projections, and development. Microsc. Res. Techn. 39, 532–546.[CrossRef][Medline]

Possidente, D. & Murphey, R.K. (1989) Genetic control of sexually dimorphic axon morphology in Drosophila sensory neurons. Dev. Biol. 132, 448–457.[CrossRef][Medline]

Shih, S.C., Chien, C.L., Tseng, K.W., Lin, S.C. & Kao, C.R. (1999) Immunohistochemical studies of transforming growth factor-beta and its receptors in the gastric mucosa of patients with refractory gastric ulcer. J. Formos. Med. Assoc. 98, 613–620.[Medline]

Teugels, E. & Ghysen, A. (1983) Independence of the numbers of legs and leg ganglia in Drosophila bithorax mutants. Nature 304, 440–442.[CrossRef]

Received: 5 August 2004
Accepted: 8 November 2004

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Layalle, S. Articles by Dambly-Chaudière, C. Search for Related Content PubMed PubMed Citation Articles by Layalle, S. Articles by Dambly-Chaudière, C.