Cell and Tissue Research

, Volume 322, Issue 3, pp 365–378

Expression patterns of nm23 genes during mouse organogenesis

Authors

  • Lilian Amrein
    • EA DRED 483, Laboratoire de Biologie de la Différenciation et du DéveloppementUniversité Victor Ségalen
  • Perrine Barraud
    • EA DRED 483, Laboratoire de Biologie de la Différenciation et du DéveloppementUniversité Victor Ségalen
  • Jean-Yves Daniel
    • EA DRED 483, Laboratoire de Biologie de la Différenciation et du DéveloppementUniversité Victor Ségalen
  • Yves Pérel
    • EA DRED 483, Laboratoire de Biologie de la Différenciation et du DéveloppementUniversité Victor Ségalen
    • INSERM E 358, Institut Francois MagendieUniversité Bordeaux 2
Regular Article

DOI: 10.1007/s00441-005-0036-9

Cite this article as:
Amrein, L., Barraud, P., Daniel, J. et al. Cell Tissue Res (2005) 322: 365. doi:10.1007/s00441-005-0036-9

Abstract

Nucleoside di-phosphate kinase enzyme (NDPK) isoforms, encoded by the nm23 family of genes, may be involved in various cellular differentiation and proliferation processes. We have therefore analyzed the expression of nm23-M1, -M2, -M3, and -M4 during embryonic mouse development. In situ hybridization data has revealed the differential expression of nm23 mRNA during organogenesis. Whereas nm23-M1 and -M3 are preferentially expressed in the nervous and sensory systems, nm23-M2 mRNA is found ubiquitously. Irrespective of the developmental state studied, nm23-M4 mRNA is only expressed at low levels in a few embryonic organs. In the cerebellum and cerebral cortex, nm23-M1, -M2, and -M3 are present in the neuronal differentiation layer, whereas nm23-M4 mRNA is distributed in the proliferating layer. Thus, nm23 mRNA is differentially expressed, and the diverse NDPK isoforms are sequentially involved in various developmental processes.

Keywords

Nucleoside diphosphate kinasenm23Embryonic developmentNervous systemDifferential expressionMouse (C57BL/6)

Introduction

The mouse nm23 (nm23-M1) gene was originally identified in murine K-135 melanoma cell lines and described as a metastatic suppressor gene (Steeg et al. 1988a). Several genes with highly homologous sequences have been characterized and shown to code for nucleoside diphospohate kinase (NDPK) in a wide variety of organisms, including the prokaryote Myxococcus xanthus (Munoz-Dorado et al. 1990), Dictyostelium discoidium (Lacombe et al. 1990), Drosophila (Biggs et al. 1990), Xenopus (Ouatas et al. 1997), and rat (Kimura et al. 1990; Shimada et al. 1993). In mouse, three closely related genes have been isolated, viz., nm23-M2 (Urano et al. 1992), nm23-M3 (Kargul et al. 2000; Massé et al. 2002), and nm23-M4 (Massé et al. 2002), whereas in human, six genes have been found, viz., nm23-H1 (Steeg et al. 1988b), nm23-H2 (Stahl et al. 1991), DR-nm23 (Venturelli et al. 1995), nm23-H4 (Milon et al. 1997), nm23-H5 (Munier et al. 1998), and nm23-H6 (Mehus et al. 1999). The nucleotide sequences of the nm23-H7 and nm23-H8 genes have been submitted to the GenBank database with accession numbers AF153191 and AF202051 but have not yet been published (Tsuiki et al. 2000).

The anti-metastatic nm23 gene family encodes NDPK proteins that catalyze the transfer of the terminal phosphate from ATP to a nucleoside diphosphate to generate a nucleoside triphosphate (Parks and Agarwal, 1973). In mouse, nm23-M1, -M2, -M3, and -M4 encode for NDPKs A, B, C, and D, respectively. Several reports suggest that, in addition to their basic enzymatic activity and probably independently of their catalytic site, NDPK isoforms are involved in other cellular functions, such as signal transduction pathways, cell growth and differentiation, embryonic development, tumor progression, metastasis, and apoptosis (for reviews, see Otero 2000; Hartsough and Steeg 2000; Kimura et al. 2000; Lacombe et al. 2000; Postel et al. 2000).

Previous immunohistochemical studies aiming to provide a detailed analysis of NDPK expression patterns during mouse embryonic development have often been hindered by the use of antibodies unable to discriminate between the isoforms (Lakso et al. 1992). Subsequently, a few studies involving the use of specific mRNA probes have revealed the nervous expression of mRNA encoding the NDPK A isoform in Xenopus (Ouatas et al. 1998), rat (Shimada et al. 1993; Fukuchi et al. 1994), and mouse (Gervasi et al. 1998; Dabernat et al. 1999b; Massé et al. 2002). In all organisms studied, the NDPK B isoform is expressed in most tissues (Ouatas et al. 1998; Gervasi et al. 1998; Dabernat et al. 1999b; Massé et al. 2002). In mouse, nm23-M3 mRNA is preferentially expressed in the nervous system at embryonic day 15.5 (E15.5), whereas nm23-M4 mRNA is undetectable at this stage (Massé et al. 2002).

Their expression in adult and embryonic central and peripheral nervous system suggests that nm23 gene products participate in neuronal functions. Mutations in Drosophila NDPK causes abnormal development of larval neural tissue (Timmons et al. 1993), and nm23 gene products may be of importance in maintaining the appropriate differentiation and signaling pathways in neuronal cells, as suggested by the NDPK-stimulated differentiation of PC12 cells in culture (Gervasi et al. 1996; Ishijima et al. 1999).

The aim of this work has been to provide a detailed analysis of the pattern of nm23-M1, -M2, -M3, and -M4 gene expression during mouse organogenesis, especially in the developing nervous system. Here, we describe the differential expression of mRNAs encoding NDPK A, B, C and D isoforms at different stages of mouse development by in situ hybridization on whole-mount embryos and cryostat sections. Their expression varies depending upon the tissue examined and the developmental stage. Moreover, a switch in nm23 gene expression between proliferating and differentiating cells has been revealed in the cortex and cerebellum.

Materials and methods

Histochemical in situ hybridization

Tissue preparation

C57BL/6 adult and embryo mice were used for in situ hybridization. L4 and L5 dorsal root ganglia (DRGs) were rapidly dissected out from adult mouse and immediately frozen on dry ice. Mouse embryos at E13.5, E15.5, and E18.5 were obtained and also immediately frozen. Frozen tissues were cut at a thickness of 14 μm on a cryostat (Leica Instruments, Nussloch, Germany) and thaw-mounted onto Super Frost D Plus slides (CML, Nemours, France).

Probes

Oligonucleotides probes were synthesized by Eurogentec (Seraing, Belgium). Eight oligonucleotides were used for the in situ detection of nm23-M1 (n=2), nm23-M2 (n=2), nm23-M3 (n=2), and nm23-M4 (n=2) mRNAs. Sequence alignment was performed by using the CLUSTALW software from published cDNA sequences. Sequences were designed as follows:
  1. (A)
    Oligonucleotides complementary to nm23-M1 mRNA (Dabernat et al. 1999b):
    1. (i)

      5′CACTTTAATCTAACAGTGGCACGTAACACTGCACAGGGAGTCACG

       
    2. (ii)

      5′CTGTGAGAACAAGAGTAAGCAGGTAGAAAACCGGCACCGTCCTA

       
     
  2. (B)
    Oligonucleotides complementary to nm23-M2 mRNA (Dabernat et al. 1999b):
    1. (i)

      5′TAGAAAAGATGATCCATCCTGTCAGTGGGATGAAGAGCTCTGTCC

       
    2. (ii)

      5′ACCCATCAGTAGTGCTGAAAAGGATTCTGGTTTCTTCATGTCTAC

       
     
  3. (C)
    Oligonucleotides complementary to nm23-M3 mRNA (Massé et al. 2002):
    1. (i)

      5′GCTGACTGCAATAACATCTACTCATATAGCCAATGTCCGGCG

       
    2. (ii)

      5′GTCCCTGGACCGACTAGGTTGGGTTGACCTGCACTCACACAGCAC

       
     
  4. (D)
    Oligonucleotides complementary to nm23-M4 mRNA (Massé et al. 2002):
    1. (i)

      5′GGGAAGCATGGGGGCTTGACTTCTTGTTGACAGAGGTAGTAGGTC

       
    2. (ii)

      5′GGCAGGTGAAGTCACTTTGACAATGCCCCAAAGGTTGTCTGGCAC

       
     

All oligonucleotides were chosen in regions presenting few homologies with sequences of related mRNAs and were then checked against the Genbank database. The oligonucleotides were radioactively labeled as previously described (Barraud et al. 2002).

In situ hybridization procedure for sections

For the detection of nm23-M1, nm23-M2, nm23-M3, or nm23-M4 mRNA, the sections were incubated as described earlier (Dagerlind et al. 1992), without any pretreatment, with 0.5 ng per slide of one or both [35S]-labeled specific oligonucleotides in a hybridization solution containing 50% formamide (Sigma, St Louis, Mo., USA), 4× SSC (1×SSC=150 mM NaCl, 15 mM sodium citrate, pH 7.0), 1× Denhardt's solution, 1% sarcosyl (N-lauryl sarcosine, Sigma), 0.02 M phosphate buffer (pH 7.0), 10% dextran sulfate (Sigma), 250 μg/ml yeast tRNA (Sigma), and 500 mg/ml sheared heat-denatured salmon sperm DNA (Sigma). After hybridization, the slides were rinsed in 1× SSC (4×15 min) at 55°C followed by 30 min at room temperature.

Sections were then air-dried and dipped into Ilford K5 nuclear emulsion (Ilford, Mobberly, Cheshire, UK) diluted 1:1 with distilled water, exposed for 7–30 days, developed in Kodak D19 (Kodak, Rochester, N.Y., USA) for 3 min, and fixed in Kodak 3000 for 8 min. After being mounted in glycerol, the sections were analyzed on a Zeiss Axiophot 2 microscope (Zeiss, Jena, Germany) equipped with a dark-field condenser. Some sections were counterstained with cresyl violet or bisbenzimide to assess the labeled areas.

Whole-mount hybridization

Tissue preparation

Cesarean section was performed after euthanasia of C57BL/6 females on gestational day 10.5. Embryos were rapidly dissected out, immersed in fixative solution (4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4) for 30 min, rinsed in phosphate-buffered saline (PBS) with 0.1% Tween 20, dehydrated in increasing concentrations (25%–100%) of methanol, and frozen in absolute methanol at −20°C.

Probes

A 163-bp nm23-M1 specific probe containing in the 3′ untranslated region of the nm23-M1 gene was generated by polymerase chain reaction and gel purification and subcloned into pBluescript II SK-vector (Stratagene, La Jolla, Calif., USA). The nm23-M3 riboprobe was complementary to the 3′ untranslated region of the nm23-M3 gene.

Antisense and sense riboprobes were generated by in vitro transcription with T7 and T3 RNA polymerases, respectively (Promega, Madison, Wis., USA). Probes were labeled by incorporation of digoxygenin-11-UTP (Roche Biochemicals, Mannheim, Germany).

In situ hybridization procedure for whole-mounts

After rehydration in decreasing concentrations (75%–25%) of methanol, embryos were treated with 2% H2O2 and incubated in 10 μg/ml proteinase K (Gibco BRL Life Technologies, Rockville, Md., USA) and then in a glycine solution (2 mg/ml). They were subsequently fixed by immersion in 0.2% glutaraldehyde. Next, embryos were prehybrided with hybridization solution consisting of 50% formamide (Sigma), 5× SSC, 50 μg/ml yeast tRNA (Sigma), 1% SDS, 200 mM dithiothreitole at 70°C for 1 h and incubated overnight at 65°C with 10 μg digoxygenin-11-UTP-labeled specific probes in 500 μl hybridization solution. Embryos were then washed twice for 30 min each at 70°C in solution I (50% formamide, 5× SSC, 1% SDS), followed by 10 min at 70°C in a mix (1/1) of solution I/solution II (0.5 M NaCl, 10 mM TRIS HCl pH 7.5, 0.1% Tween 20), and three times at room temperature in solution II. Embryos were finally treated with 100 μg/ml RNase in solution II at 37°C for 30 min, rinsed in solution II for 5 min at room temperature, 5 min in solution III (50 % formamide, 2× SSC), followed by two washes for 30 min at 65°C in solution III. They were finally incubated three times in 2 mM Levamisol in TRIS-buffered saline with 0.1% Tween 20 at room temperature.

Probes were detected by using alkaline-phosphatase-conjugated anti-digoxigenin Fab fragments (Roche). Enzyme activity was visualized with nitro blue tetrazolium/5-bromo-4-fluoro-3-indolyl phosphate as described by the manufacturer (Gibco BRL Life Technologies).

Double labeling combining in situ hybridization and BrdU detection

Cesarean section was performed after euthanasia of C57BL/6 females on gestational day 18.5. At 1 h before collection of E18.5 embryos, each female received an intra-peritoneal injection of 250 mg/kg 5′-bromodeoxyuridine (BrdU) in PBS. E18.5 embryonic brains were rapidly dissected out, immersed in the fixative solution overnight at 4°C, and rinsed for at least 24 h in 0.1 M phosphate buffer (pH 7.4) containing 12% sucrose and 0.01% sodium azide (Sigma). They were then frozen in isopentane, cooled to −70°C on dry ice. Frozen tissues were cut at a thickness of 10 μm on a cryostat and thaw-mounted onto Super Frost D Plus slides.

Probe labeling and hybridization procedure were performed as described previously (Barraud et al. 2002; Landry and Calas, 2002).

After posthybridization washes, immunohistochemistry was performed by using a Vector M.O.M kit (Vector, Burlingame, Calif., USA), with the following modifications. After the secondary biotinylated anti-mouse IgG, slides were rinsed in PBS and incubated for 2 h with horseradish-peroxidase-conjugated streptavidin (Vector) diluted 1:200 in 2% normal goat serum in PBS. 3-3′-Diaminobenzidine was used as the horseradish-peroxidase substrate. Radioactive probes were then visualized by autoradiography as described above.

Controls

For in situ hybridization control experiments, excess of non-labeled probes (100-fold) was added to the mixture. For whole-mount hybridization, sense riboprobes were used as controls.

Results

The four nm23 genes whose expression was investigated in this study belong to the same gene family and share high sequence homology (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig1a_HTML.gifhttps://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig1b_HTML.gif
Fig. 1

Comparison of nm23-M1, -M2, -M3, and -M4 mRNA primary sequences. The first methionine codon and the stop codon are show in bold and are boxed (asterisks conserved nucleotides found at all sites, dashes gaps). Oligonucleotide probes used for in situ hybridization are complementary to the yellow [oligonucleotides (i)] and red [oligonucleotides (ii)] highlighted sequences

General distribution of the four nm23 mRNAs

To analyze nm23-M1 and -M3 in the earliest stages of mouse organogenesis, nm23-M1 and -M3 mRNAs were hybridized on E10.5 embryos by using digoxigenin labeled riboprobes (Fig. 2). mRNA encoded by nm23-M1 (Fig. 2b–d) and -M3 (Fig. 2d) gene were highly expressed in diencephalon (Fig. 2b–d). nm23-M1 mRNA was also detected in the neural tube and in developing DRGs (Fig. 2c,e). Heart was labeled at this developmental stage (Fig. 2c).
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig2_HTML.jpg
Fig. 2

Detection of nm23-M1 mRNA (b, c, e) and nm23-M3 mRNA (d) expression by whole-mount in situ hybridization on E10.5 mice, with digoxygenin-11-UTP labeled antisense riboprobes. Control experiments were performed with specific nm23-M3 sense riboprobes (a). mRNAs were seen in developing dorsal root ganglia (double arrowhead in c), heart (arrow in c), and brain (arrowheads in b, c, d). A dorsal view showed positive signal in the neural tube (arrows in e). Bars 0.5 mm

At later stages of mouse organogenesis, expression of the four nm23 genes was investigated by using oligoprobes on E13.5 (Fig. 3), E15.5, and E18.5 embryo sections. Results are summarized on Table 1. In all late embryonic stages, nm23-M1 mRNA labeling was strong in the nervous system and sensory organs but weak in other tissues. In contrast, nm23-M2 mRNA was highly expressed in most tissues. The nm23-M3 messenger was widely expressed in the nervous system. Its expression was faint in the sensory system and almost undetectable in other organs. nm23-M4 mRNA was only detected in some brain regions and heart or intestine at E18.5.
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig3_HTML.jpg
Fig. 3

Photomicrographs showing the distribution of nm23-M1 (a), -M2 (b), -M3 (c), and -M4 (d) mRNA in parasagittal sections of mouse embryo at E13.5; in situ hybridization with 35S-labeled oligoprobes. nm23-M1 (a) and nm23-M3 (c) mRNA expression is prominent in the nervous system whereas nm23-M2 mRNA (b) displays a wide distribution throughout the embryo. All messengers are strongly expressed in the dorsal root ganglion (DRG; ac). nm23-M4 mRNA (d) is hardly detectable at this stage. Head-to-rump length 6 mm

Table 1

Summary and comparison of nm23-M1, -M2, -M3, and -M4 mRNA expression in embryonic tissues

 Site

nm23 mRNA expression during mouse organogenesis

nm23-M1

nm23-M2

nm23-M3

nm23-M4

E13.5

E15.5

E18.5

E13.5

E15.5

E18.5

E13.5

E15.5

E18.5

E13.5

E15.5

E18.5

Olfactory bulbs

-Granular cell layer

nd

+++

+++

nd

+++

+++

nd

+++

++

nd

nd

0

-Subependymal zone

nd

+

+

nd

++

++

nd

+++

++

nd

nd

0

Cerebellum

+++

+++

 

+++

+++

 

+

++

 

+

++

 

External granular layer

  

+

  

++

  

+

  

+++

Internal granular layer

  

+++

  

+++

  

+++

  

++

Cortex

+++

  

+++

  

+++

  

0

  

Cortical plate

 

+++

+++

 

+++

+++

 

+++

+++

 

++

+

Ventricular zone

 

+++

+

 

+++

+

 

++

+

 

+++

+++

Mesencephalon

+++

+++

+++

+++

+++

+++

+

++

+++

0

+

+

Spinal cord

+++

+++

++

+++

+++

+++

++

++

++

0

0

0

Trigeminal ganglion

+++

+++

+++

+++

+++

+++

+++

+++

+++

+

0

0

DRG

+++

+++

+++

+++

+++

+++

+++

+++

+++

0

0

0

Tongue

++

++

++

+++

+++

+++

++

+

++

nd

0

+

Eye

nd

+++

+++

nd

++

++

nd

+

nd

nd

+

nd

Hair follicles

nd

++

++

nd

+++

+++

nd

++

++

nd

+

+

Heart

++

+++

+

+++

++

+++

+

++

++

0

+

+++

Thymus

++

+

+++

nd

+

+++

++

+

++

0

0

+

Lung

+

+

+

+++

+++

++

+

+

+

0

+

0

Intestine

+

+

+

+++

+++

+++

+

+

+

+

++

++

nd not done, 0 no expression, + low, ++ moderate, +++ abundant

Expression in the developing nervous system

Central nervous system

From E13.5 to E15.5, the nm23-M1, -M2, and -M3 genes were highly expressed in the nervous system. In all the developmental stages, the amount of nm23-M1 mRNA was variable in the different brain areas. For instance, in the olfactory bulbs, nm23-M1 mRNA was highly expressed in differentiating cells of the granule cell layer, but labeling was weak in the subependymal zone (Fig. 4a,b). nm23-M2 and -M3 mRNAs were homogeneously expressed in various areas, such as the spinal cord (Fig. 4c) and mesencephalon (Fig. 4d), respectively. In contrast, nm23-M4 mRNA was only detected in discrete regions.
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig4_HTML.jpg
Fig. 4

Detection of nm23-M1, -M2, and -M3 mRNAs expression by in situ hybridization with 35S-labeled oligoprobes in the central nervous system. At E18.5, nm23-M1 mRNAs were detected in the granule cell layer of the olfactory bulbs (arrows in a). Cresyl-violet counterstaining shows the cellular distribution of nm23-M1 in the olfactory bulb (arrows in b). Strong expression of nm23-M2 and -M3 mRNAs was detected in the mesencephalon at E15.5 (c) and spinal cord at E13.5 (d). Bar 50 μm

The developing cortex at E18.5 exhibited differential distribution of nm23-M4 versus nm23-M1, -M2, and -M3 mRNA (Fig. 5). mRNAs for isoforms A, B, and C were preferentially expressed in the cortical plate, the location of differentiating cells (Fig. 5a–c). Isoform D mRNA was detected in the ventricular zone, a region mainly containing proliferating cells (Fig. 5d). Double labeling combining in situ hybridization and immunohistochemistry (Fig. 6) confirmed that BrdU-positive and nm23-M1 mRNA-expressing cell populations were segregated (Fig. 6a–c). In contrast, nm23-M4 mRNA was found in BrdU-positive proliferating cells (Fig. 6d–f). Accordingly, nm23-M1 and -M4 mRNA were weakly expressed in the proliferating area and cortical plate, respectively.
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig5_HTML.jpg
Fig. 5

Differential expression patterns of nm23-M1, -M2, -M3, and -M4 mRNA detected by in situ hybridization in E18.5 cortex (v ventricle). nm23-M1 (a), -M2 (b), and -M3 (c) mRNA were preferentially expressed in the cortical plate (cp), whereas nm23-M4 mRNA was detected in the ventricular zone (vz). Bar 50 μm

https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig6_HTML.jpg
Fig. 6

Autoradiograms of E18.5 brain cortex sections after double labeling combining in situ hybridization with 35S-labeled oligoprobes complementary to nm23-M1 and -M4 mRNAs and immunoperoxydase staining for BrdU. nm23-M4 mRNA was preferentially expressed in BrdU-positive cells of the ventricular zone (arrows in df), whereas nm23-M1 mRNA (arrowheads in a, c) was detected in non-proliferating cells of the cortical plate (arrows in a, b). Bar 12 μm

The same segregation between nm23-M4 versus nm23-M1, -M2, and -M3 mRNA was observed in the cerebellum. Isoforms A, B, and C mRNA were preferentially expressed in the internal granule cell layer (Fig. 7a–c,e), whereas isoform D messenger was preferentially located in the external germinal layer (Fig. 7d,f).
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig7_HTML.jpg
Fig. 7

Differential expression pattern of nm23-M1, -M2, -M3 and -M4 mRNA detected by in situ hybridization in E18.5 cerebellum. nm23-M1 (a), -M2 (b), and -M3 (c) mRNAs were preferentially expressed in the internal granule cell layer (arrowheads in ac), whereas nm23-M4 (d) mRNA was abundant in the external germinal layer (arrow in d). Cresyl-violet counterstaining shows the cellular distribution of nm23-M2 (arrowhead in e) and -M4 (arrow in f) in the cerebellum. Bar 50 μm

In peripheral nervous system

Our study of sensory systems focused on trigeminal ganglia and DRGs. Little expression of nm23-M4 mRNA was seen in trigeminal ganglia at E13.5, with none at E15.5 and E18.5. Isoform A, B and C mRNAs were strongly expressed in sensory ganglia at all stages studied (Table 1). In DRG, nm23-M1, -M2, and -M3 mRNAs were detected as soon as E10.5 (data not shown). Their expression increased until E13.5 and E18.5, before decreasing in adult ganglia (Fig. 8). Quantification by cellular grain-counting methods confirmed this result and showed a maximal amplitude of variation for nm23-M2 mRNA (Fig. 9). The amount of nm23-M1 and -M3 mRNAs was similarly regulated (Fig. 9). All differentiating cells expressed a given isoform mRNA. Thus, the regulation of expression affected cellular mRNA levels rather than the number of positive cells.
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig8_HTML.jpg
Fig. 8

Autoradiograms of mouse DRG sections during ontogenesis at E13.5 (a, d, g), E15.5 (b, e, h), and adult (c, f, i) after in situ hybridization with 35S-labeled oligoprobes complementary to nm23-M1 (ac), -M2 (dh), and -M3 (gi) mRNA. All differentiating neurons were labeled. Bar 50 μm

https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig9_HTML.gif
Fig. 9

Quantification of in situ hybridization signal for nm23-M1, -M2, or -M3 mRNAs in embryonic and adult DRGs during ontogenesis. nm23 mRNA expression was highest between E15.5 and E18.5, before decreasing in adult ganglia. *P<0.05, ***P<0.01

Expression during sensory system development

All sensory organs investigated in this work expressed nm23 genes. mRNA expression levels were different from one isoform to other one but remained constant during organogenesis for a given isoform. In hair follicles, NDPK B messenger was strongly expressed (Table 1), whereas the expression levels of mRNA encoding isoforms A and C were moderate. Similar expression was shown in the tongue (Table 1). In the retina, nm23-M3 and -M4 mRNAs were expressed at low levels and only at E15.5 but remained undetectable at E13.5 and E18.5. In contrast, nm23-M1 messenger was highly expressed in the retina of late embryos.

Expression during the development of other organs

Nervous and sensory systems were not the only tissues expressing mRNAs for NDPK isoforms. Despite its prominent expression in the nervous system, nm23-M1 mRNA was also detected in the heart. Indeed, nm23-M1 mRNA expression was maximal at E15.5 and then decreased at E18.5 (Fig. 10a,b). nm23-M2 and -M3 expression level remained constant in heart. However, in situ hybridization exhibited strong signal for nm23-M2 mRNA, whereas nm23-M3 mRNA was found only at low levels. The heart was one of the few organs in which nm23-M4 mRNA was detected by our method. Its expression increased until a maximum at E18.5 (Fig. 10c,d).

The immune system, like the nervous and circulatory system, expressed nm23 mRNA in a regulated manner. Between E13.5 and E15.5, nm23-M1 mRNA expression increased in the thymus. In this organ, nm23-M1 and -M2 expression was highest at E18.5 (Fig. 11). Despite its weak expression, we also showed regulation of nm23-M3 mRNA in the thymus. Under our experimental conditions, nm23-M4 messenger remained undetectable.

The lung exhibited high levels of nm23-M2 mRNA. Only small amounts of nm23-M1 and -M3 mRNAs were detected in this organ (Table 1). As in most of the embryonic organs, nm23-M4 remained undetectable.

In the intestine, one of the few organs expressing nm23-M4 mRNA, we detected a moderate but constant expression of this transcript during embryonic development (Table 1). Comparable with other embryonic organs, few nm23-M1 and -M3 mRNAs were detected in the intestine, whereas nm23-M2 was strongly expressed.
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig10_HTML.jpg
Fig. 10

Detection of nm23-M1 and -M4 mRNA expression by in situ hybridization with 35S-labeled oligoprobes in the heart. nm23-M1 mRNA expression reached a maximum at E15.5 (a) and decreased at E18.5 (b). At E15.5, only a low level of expression of nm23-M4 mRNA was detected in this organ (c), whereas a maximum was seen at E18.5 (d). Bar 50 μm

https://static-content.springer.com/image/art%3A10.1007%2Fs00441-005-0036-9/MediaObjects/441_2005_36_Fig11_HTML.jpg
Fig. 11

Detection of nm23-M1 and -M2 mRNA expression by in situ hybridization with 35S-labeled oligoprobes in thymus. Expression of nm23-M1 (a, b) and -M2 (c, d) increases between E15.5 (a, c) and E18.5 (b, d). Bar 50 μm

Discussion

Regulation of nm23 gene expression during embryogenesis

We have performed a quantitative analysis of the expression of nm23 mRNA only for the DRG, and a qualitative analysis for the other organs. The lack of in situ hybridization signal does not necessarily imply a lack of messenger but rather a low transcription activity and/or a high degradation rate. However, since our results corroborate previous immunohistochemistry data, we consider that the measured mRNA levels reflect NDPK protein content, at least partially.

nm23 gene expression patterns during embryogenesis are well documented in invertebrate organisms, such as Drosophila melanogaster (Dearolf et al. 1988; Biggs et al. 1990), and in vertebrates, such as Xenopus laevis (Ouatas et al. 1998). However, few studies have been carried out on mouse organogenesis. Moreover, the techniques used have not been able to discriminate between the different nm23 isoforms. Nevertheless, some data have previously shown NDPK protein expression in the developing nervous system and heart, the first embryonic tissues to differentiate, on embryonic day E10.5 (Lakso et al. 1992). Cloning and sequencing of various mouse nm23 cDNAs have allowed specific probes to be used to study the expression of nm23-M1, -M2, -M3, and -M4 mRNAs (Dabernat et al. 1999a,b; Massé et al. 2002). However, no data are available for a comparison of nm23 gene expression patterns during mouse embryogenic development.

Our work confirms previous studies: (1) nm23 genes are actively transcribed in the nervous system and heart on embryonic day E10.5 (Lakso et al. 1992), (2) nm23-M2 mRNA is widely expressed in all tissues investigated (Dabernat et al. 1999a,b; Massé et al. 2002) (3) nm23-M1 and -M3 are preferentially expressed in the nervous system (Dabernat et al. 1999a,b; Massé et al. 2002).

At all developing stages, nm23-M2 gene expression is ubiquitous in most of the tissues studied, as previously described in adult mouse (Dabernat et al. 1999a). This ubiquitous distribution is consistent with the putative function of nm23-M2 as a transcription regulator (Postel et al. 1993; Arcinas and Boxer 1994; Berberich and Postel 1995). This hypothesis is supported by morphological data showing a nuclear localization for NDPK B in vitro (Kraeft et al. 1996; Pinon et al. 1999) and in vivo (Barraud et al. 2002). Other isoforms exhibit tissue-specific expression. In some organs, such as hair follicles, intestine, or lungs, their expression remains constant during organogenesis. These isoforms are thus unlikely to play key roles in the formation and/or function of these organs.

On the other hand, nm23 mRNA expression is regulated during differentiation in other tissues. In heart and thymus, expression changes according to the developing stage examined, with a minimum and maximum that differ from one isoform to the other. Lakso et al. (1992) have studied NDPK expression patterns during mouse development by using antibodies unable to distinguish between the different NDPK isoforms. They have demonstrated the high expression of these proteins in the heart as soon as E10.5. NDPK expression subsequently remains unchanged until adulthood (Lakso et al. 1992). The strong signal for nm23-M1 and -M2 found in heart suggests that Lakso et al. (1992) had mainly detected isoforms A and B, rather than isoforms C and D. Furthermore, our study provides new insights regarding the possible differential regulation of the various isoforms at the mRNA level. Thus, the expression patterns of nm23-M1 and -M4 fluctuate in an opposing manner in late embryos, with the highest expression for nm23-M1 and -M4 being at E15.5 and E18.5, respectively. These opposing expression patterns suggest that NDPK A and D proteins are involved in different processes during heart development.

Taken together, these data show that, during mouse organogenesis, nm23-M1, -M2, -M3, and -M4 gene expression depends on the organ considered and on the isoform itself. Thus, each isoform or isoform combination may be involved in different cellular processes and exhibit specific properties possibly related to nucleotide channeling, transcription, and transduction (Mesnildrey et al. 1998; Barraud et al. 2002). Such specificities have been suggested for other gene families with expression patterns similar to the nm23 genes. This includes the various activin and inhibin subunits, which are members of the transforming growth factor-β gene family and their receptors (Feijen et al. 1994) or of the fibroblast growth factor gene family (Colvin et al. 1999).

Expression patterns of mRNA nm23 in the developing nervous system

Central nervous system

In the central nervous system, some tissues show a heterogeneous distribution of nm23 transcripts during embryogenesis, particularly in the olfactory bulb, brain cortex, and cerebellum. In the olfactory bulb, nm23-M1 mRNA has been mainly detected in the granule cell layer, whereas nm23-M3 mRNA is present in both the granule cell layer and the subependymal zone. The subependymal zone of the olfactory bulb contains dividing cells (Altman 1969; Smith and Luskin, 1998) exhibiting immunohistochemical (Menezes et al. 1995), morphological (Luskin, 1993; Jankovski and Sotelo, 1996), and functional (Jankovski and Sotelo, 1996; Stewart and Rodaros, 1999) neuronal features. They become postmitotic and attain their neuronal differentiation in the granule cell layer (Luskin 1993; Lois and Alvarez-Buylla 1994; Coskun and Luskin 2002; Stewart et al. 2002). NDPK C isoform in the olfactory bulb might be involved in both in cell proliferation and differentiation, whereas the NDPK A isoform might play a role only in neuronal differentiation.

In the cerebellum and cortex, the four nm23 genes show a complementary distribution. nm23-M1, -M2, and -M3 mRNAs are located in the outer part of the cortex, whereas nm23-M4 mRNA is detected in the inner area.

Morphological modifications occurring during cell division imply an involvement with cytoskeleton proteins; indeed, interactions of NDPKs with some cytoskeleton proteins have been previously described (Biggs et al. 1990; Lombardi et al. 1995; Otero 1997; Pinon et al. 1999). In humans, isoform C expression appears to be associated with neuroblastoma cells and can induce cellular adhesion by the regulation integrin expression (Amendola et al. 1997). NDPK C might be involved in remodeling cell adhesion during cortical neurogenesis and/or migration.

Cortical neurogenesis is achieved through asymmetric divisions, in which one daughter cell of a cortical progenitor becomes a postmitotic neuron, while the other remains as a precursor cell (Price and Thurlow 1988; Rakic 1988; Walsh and Cepko 1988; Reid et al. 1995; Caviness and Takahashi 1995; Chenn and McConnell 1995; Luskin 1998). mRNAs for nm23-M1, -M2, and -M3 are expressed in differentiating neurons in the cortical plate, whereas nm23-M4 mRNA is restricted to BrdU-positive dividing progenitors in the ventricular zone. A similar situation is observed in the cerebellum: the internal granule cell layer expressing isoforms A, B, and C corresponds to differentiating cells, whereas the external germinal layer expressing isoform D corresponds to proliferating cells (Kuhar et al. 1993).

In human cells, nm23-H1 has been shown to co-immunoprecipitate and interact with PRUNE (Reymond et al. 1999). Moreover, the increase in cellular motility is correlated to the interaction between PRUNE and the human isoform A (D'Angelo and Zollo 2004). These data support the role of nm23 genes in developmental processes and suggest that nm23-M1 might be involved in motility processes underlying cell migration.

Peripheral nervous system

In mouse, at E10.5, DRG precursors migrate from the neural crest and aggregate in ganglia before they differentiate. nm23 mRNAs are detected immediately at the end of cell migration and before the start of cell differentiation. Quantification of nm23 mRNAs labeling intensity has been performed from E13.5, corresponding to neuronal precursor division in the DRG (Lawson and Biscoe 1979), to the adult. nm23 mRNA levels increase during cell differentiation until E15.5-E18.5. At this stage, sensory neurons establish contact with and innervated their targets (Murphy et al. 1993), their fate being dependent on nerve growth factor (NGF).

NGF may cause an increase in nm23 mRNA levels. This hypothesis is supported by results obtained in PC12 cell culture experiments, which have shown that nm23 overexpression induces NGF-dependent differentiation (Gervasi et al. 1996).

Concluding remarks

This study presents, for the first time, a simultaneous analysis of nm23-M1, -M2, -M3, and -M4 gene expression patterns during mouse development. We report either overlapping or complementary nm23 expression, notably in the developing nervous system. In particular, from its mRNA distribution, nm23-M4 might have distinct functions from nm23-M1, -M2, and -M3 and could be mainly involved in cell proliferation during ontogenesis.

Copyright information

© Springer-Verlag 2005