Abstract
Aint was originally identified on the basis of its interaction in vitro with the aryl hydrocarbon nuclear receptor translocator (Arnt). Arnt is a common heterodimerization partner in the basic helix-loop–helix (bHLH)-PER-ARNT-SIM (PAS) protein family and is involved in diverse biological functions. These include xenobiotic metabolism, hypoxic response, and circadian rhythm. In addition, Arnt has a crucial role during development. Aint is a member of a growing family of transforming acidic coiled-coil (TACC) proteins and is the murine homologue of human TACC3. Here we report the spatiotemporal expression of Tacc3 mRNA and protein in embryonic, postnatally developing, and adult mouse tissues using in situ hybridization and immunocytochemistry. Tacc3 mRNA was highly expressed in proliferating cells of several organs during murine development. However, the only adult tissues expressing high levels were testis and ovary. Immunocytochemistry revealed that Tacc3 is a nuclear protein. Our results suggest that Tacc3 has an important role in murine development, spermatogenesis, and oogenesis.
Keywords
ARNT interacting protein (Aint), also now referred to as transforming acidic coiled-coil protein 3 (Tacc3), is a recently discovered protein that has been shown to interact with the aryl hydrocarbon receptor nuclear translocator protein (ARNT) in vitro (Sadek et al. 2000). ARNT is a founding member of basic helix-loop–helix/PER-ARNT-SIM (bHLH/PAS) proteins. Members of this family mediate various biological tasks, e.g., effects of dioxin-like environmental pollutants, responses to hypoxia, regulation of circadian rhythm, and also have a significant role in development (reviewed by Rowlands and Gustafsson 1997; Crews 1998). These transcription factors form heterodimeric DNA-binding protein complexes between family members. However, the number of known non-related interacting partners is increasing (Gekakis et al. 1995; Arany et al. 1996; Carver and Bradfield 1997; Ma and Whitlock 1997; Meyer et al. 1998).
Although Tacc3 was recently cloned from a day 17 mouse embryo cDNA library as a result of search for novel ARNT interacting proteins, it is not a member of the bHLH/PAS protein family (Sadek et al. 2000). Tacc3 consists of 631 amino acids and includes seven repeats of 24 amino acids and two coiled-coil regions. Glutathione S-transferase pulldown experiments showed that the C-terminus of Tacc3 interacts specifically with ARNT and its homologue, the aryl hydrocarbon receptor nuclear translocator 2 (ARNT2), and weakly with another bHLH/PAS protein, Single-minded-2 (Sim2), but not with aryl hydrocarbon receptor (AhR) or hypoxia-inducible factor (HIF-1α), indicating that Tacc3 may have a role in ARNT signaling pathways. The PAS domain of ARNT mediates the interaction with Tacc3. In addition, transient transfection experiments revealed that Tacc3 augments the transcriptional response to hypoxia in the Hif-1α signaling pathway via the hypoxia response element (HRE) (Sadek et al. 2000).
Tacc3 belongs to a new and growing family of coiled-coil proteins and is the mouse homologue of TACC3 cloned from human (Still et al. 1999b). [For simplicity within this field, we have chosen to use the term Tacc3 to refer to Aint in these and future studies. It is important to note that the sequence deposited for Aint (GenBank accession number AF156934) has differences from that currently deposited for Tacc3.] Human ERIC1 and mouse Eric1 (McKeveney et al. 2001) share high degrees of similarity with Tacc3 and are likely to be variants of the same protein. Other members of this family include TACC1 and TACC2/AZU1/ECTACC (Still et al. 1999a; Chen et al. 2000; Pu et al. 2001), Xenopus maskin (Stebbins-Boaz et al. 1999) and Drosophila TACC (D-TACC) (Gergely et al. 2000b). Northern analysis showed TACC3 expression in adult human testis and cancer cell lines (Still et al. 1999b). In embryonic mice, Tacc3 mRNA was detected at E7, E11, E15 E17 and neonatal tissues (Still et al. 1999b; Sadek et al. 2000). In addition, our preliminary results observed by Northern analysis and in situ hybridization showed a wide, dynamic expression pattern both in peripheral organs and in the central nervous system during murine embryogenesis (Sadek et al. 2000). Recently published knockout of Tacc3 showed that mice lacking Tacc3 die at mid to late gestation with abnormalities in several cell types (Piekorz et al. 2002), demonstrating its importance for normal development.
Recent studies suggest that the TACC proteins are a conserved family of centrosome- and microtubule-interacting proteins (Gergely et al. 2000a,b; Cullen and Ohkura 2001; Lee et al. 2001; Theurkauf 2001). Human TACC3 has been shown to be strongly concentrated at centrosomes during mitosis (Gergely et al. 2000a). When the TACC3 protein is overexpressed in human cells in culture, large polymeric structures interacting with microtubules and tubulin are formed and the number of microtubules is increased (Gergely et al. 2000a). Therefore, TACC3 may have a role in regulating interaction between centrosomes and microtubules in a similar manner to D-TACC protein, which is believed to be required for stabilizing microtubules in Drosophila (Gergely et al. 2000b; Cullen and Ohkura 2001; Lee et al. 2001).
Tacc3 interacts in vitro with ARNT, ARNT2 and, to a lesser degree, with Sim2, and these proteins have developmental importance (Sadek et al. 2000). Therefore, we wanted to study in detail the expression of Tacc3 mRNA during murine embryogenesis, during postnatal development, and in adult tissues. Furthermore, we have examined the localization of Tacc3 protein in embryonic and early postnatal mouse to obtain more information regarding its role during development and in adult tissues.
Materials and Methods
In Situ Hybridization
In this study we used embryonic (E9-E17), 36-hr, 3-, 8-, and 14-day-old and adult NMRI mice. Embryos and tissues were frozen on a block of dry ice and sectioned with a Microm HM-500 cryostat to serial 14 μm thick sections, thawed onto polysine (Menzel-Gläser; Hamburg, Germany) glasses and stored at −20C until used.
In situ hybridization was carried out as previously described by Dagerlind et al. (1992) and by Kononen and Pelto-Huikko (1997). Two oligonucleotide probes directed against the mouse Tacc3 mRNA (nucleotides 162–191 and 1660–1689) were used. The probes used also identify the recently reported shorter form, Tacc3b, which is expressed in ovary (Hao et al. 2002). The probe sequences exhibited less than 70% similarity to all known sequences in the GenBank database. Both probes gave similar expression patterns when used separately and were usually combined to intensify the hybridization signal. Several probes for non-related mRNAs with known expression patterns and with similar length and GC content were used as controls. Addition of a 100-fold excess of non-labeled probes extinguished all signal. The probes were labeled with α-[33P]-dATP (DuPont-NEN; Boston, MA) to a specific activity of 109 cpm/μg using terminal deoxynucleotidyltransferase (Amersham; Poole, UK). The sections were briefly air-dried at room temperature before hybridization. Hybridization was performed in a humidified chamber at 42C for 18 hr with 5 ng/ml of the probe in a mixture containing 4 X SSC (1 X SSC = 0.15 M NaCl, 0.015 M sodium citrate), 50% formamide, 1 X Denhardt's solution (0.02% polyvinyl-pyrroline, 0.02% bovine serum albumin, and 0.02% Ficoll), 1% sarcosyl, 0.02 M phosphate buffer (pH 7.0), and 10% dextran sulfate. After hybridization the sections were washed four times at 55C in 1 X SSC for 15 min each and then left to cool for 1 hr at room temperature. The sections were rinsed in distilled water, dehydrated with 60 and 90% ethanol, and air-dried. Thereafter the sections were covered with Kodak MR5 autoradiography film (Kodak; Rochester, NY) and exposed at −20C for 30–60 days. The autoradiography films were developed using LX24 developer and AL4 fixative (Kodak). Alternatively, the sections were dipped in NTB2 emulsion (Kodak) diluted 1:1 with distilled water and exposed at 4C. After 60–90 days of exposure, the sections were developed with D19 developer (Kodak), fixed with G333 fixative (Agfa Gevaert; Leverkusen, Germany). The sections were counterstained with hematoxylin-eosin and examined with a Nikon FXA microscope equipped with epipolarization and bright-field optics and a PCO Sensicam digital camera (PCO; Kelheim, Germany). Autoradiographs were digitized with a Nikon Coolscan II filmscanner. Images were processed using Corel Draw software and printed with a Hewlett Packard 995c printer.
Immunohistochemistry
For immunocytochemistry, whole embryos (E9, E13, and E16) and tissues of 2-day-old (P2) mice were fixed by immersion in the 4% paraformaldehyde in PBS for 3–6 hr. The tissue samples were cryoprotected with 15% sucrose in PBS, frozen with carbon dioxide, and 10-μm sections were cut. The embryos were embedded in paraffin, cut into 5-μm sections, and deparaffinized. Endogenous peroxidase activity was inhibited by immersing the sections in 0.5% hydrogen peroxide in PBS for 20 min. A rabbit polyclonal antibody was raised against a peptide, (VTFQTPLRDPQTHRILSP, hTACC3 amino acids 55–72; accession number NM006342 and AF156934 for mouse Tacc3) designed to detect both hTACC3 and mTacc3 and purified by passage over a CNBr-Sepharose 4B column (Sadek et al. in press). Sections were first incubated with the anti-TACC3 antibody at 4C overnight in a buffer containing 1% BSA and 0.3% Triton X-100. After several washes the sections were incubated with biotinylated goat anti-rabbit IgG and ABC complex (Vectastain Elite Kit; Vector Laboratories, Burlingame, CA) for 30 min each. Nickel-intensified diaminobenzidine was used as a chromogen to visualize the sites expressing Tacc3 immunoreactivity. The sections were dehydrated and embedded in Entellan. Controls included omission of the primary and secondary antibodies and preabsorption of the primary antibodies with respective proteins. Only the reactions that disappeared after preadsorption of the antibody with the antigen are described. In some organs the antibody gave clear cytoplasmic staining that could not be abolished by presaturation and is therefore regarded as nonspecific.
Results
Tacc3 mRNA Expression
In situ hybridization analysis of Tacc3 mRNA during embryonic and postnatal development and in adult mouse revealed a tightly regulated expression pattern in different organ systems. Tacc3 mRNA was detected at all time points studied in distinct organs, but the expression was most widespread throughout various tissues and at the highest level during embryonic days E11, E13, and E15. In some organs, the expression continued postnatally. However, in adult mouse high amounts of Tacc3 mRNA were present only in testis and ovary and lower levels were detected in thymus and spleen. (Tables 1 and 2)
Expression of Tacc3 mRNA in developing murine central nervous system a
+++, strong signal; ++, clear signal; +, low but positive signal; −, no signal; no sign, not studied.
Expression of Tacc3 mRNA in peripheral organs of mouse
Nervous System
At the initial stages of prenatal brain development, at E9 and E10, strong expression of Tacc3 mRNA was seen in the rapidly growing primary proliferative matrix, the neuroepithelium. The neuroepithelium is almost the only component of the developing nervous system at this time and gives rise to all neural elements of the brain (Altman and Bayer 1995). At E11, 13, 15, and 17, when the neuroepithelium is still prominent and active, it continued to express Tacc3 mRNA conspicuously. At this time also the proliferative secondary germinal matrices surrounding the ventricular lumen (also called subventricular zones) and from E13 onwards the external germinal layer of cerebellum that contains the precursors of the majority of the neurons in cerebellar cortex, the granular cells, showed a signal for Tacc3 mRNA. The progressively enlarging brain parenchyma, which consists of differentiating neurons, did not express Tacc3 mRNA. The expression in primary germinal matrices became more and more limited towards the perinatal period, whereas in some subventricular zones (neocortical, striatal, and olfactory bulb subventricular zones) a weak signal was still seen at postnatal day 3. In the external germinal layer of cerebellum, the signal was detectable until 14 days after birth. In summary, Tacc3 mRNA expression was seen in brain areas containing dividing neuroblasts, and the expression ceased when the young neurons left these proliferation sites to migrate to their final destination and lost the potential to divide. This was elegantly seen in the external germinal layer of developing cerebellum, where the expression of Tacc3 mRNA emerged at the same time (E13) as the number of cells increased (Miale and Sidman 1961). This layer, from which the pregranular neurons migrate into the developing cerebellar white matter to produce granular cells, does not disappear until 20–25 days after birth, in agreement with the late expression of Tacc3 in this area. For the detailed distribution of Tacc3 mRNA, see Table 1 and Figures 1 and 2a–2i. In adult brain, no Tacc3 mRNA could be detected.
Localization of Tacc3 mRNA in embryonic and newborn mouse shown by in situ hybridization. At E10 (
In situ hybridization analysis of Tacc3 mRNA expression in developing cerebellum at E17 (
Digestive System
In the alimentary tract, Tacc3 mRNA was expressed strongly in most of the hepatic parenchymal cells from E11 to E15, whereas later no signal could be detected (Figures 1b–1f, 3c, and 3d). In all parts of the intestines, signal was seen both in mucosa, including epithelial cells, and in the muscular layer (Figures 1c, 1g, 3a, 3e, and 3f). The intestinal expression was high at E11 and E13, weak from E15 to postnatal day 1.5, and undetectable at P8, P14, and adult. In esophagus, Tacc3 mRNA was seen from E13 to P1.5 in the mucosa and muscular layer (Figure 3h). At E13 and E15 there was a signal in the tongue epithelium and muscle, while at E17 and P1.5 the expression was restricted to the epithelium (Figures 1c, 1e, and 1g). In the oral cavity, Tacc3 mRNA was seen from E13 to E17. In tooth, the expression could be observed from E13 to P1.5. The signal was seen in odontoblasts, inner and outer enamel epithelium, and in stratum intermedium (Figure 4c). Salivary gland showed low to moderate levels of Tacc3 mRNA from E15 to P1.5 (Figure 3b).
In small intestine of a newborn mouse, expression of Tacc3 mRNA is seen in all layers of intestinal wall; the dotted line separates the muscular layer (mus) and mucosa (mu) (
Respiratory System
Tacc3 mRNA was expressed at a moderate level in nasal process at E11 (Figure 1b). In lung, the signal could be seen from E11 to E17 (Figures 1c, 1d, and 3g). At E13 the expression was strong, but subsequently it diminished and no signal could be detected postnatally. Tacc3 mRNA was seen both in lung mesenchymal and bronchial epithelial cells and in all layers of the pharyngeal, laryngeal, and tracheal (from E13 to E17) walls (Figure 3h).
Genitourinary System
Tacc3 mRNA was abundant in the primitive cortical region of kidney at E13 and E15, and weak expression was also seen in the medullary region (Figures 1e and 4f). From E17 to P3 the signal was moderate, at P8 weak, and by P14 the expression had ceased. In urinary bladder, weak to moderate signal could be seen from E13 to P1.5 (Figure 1g). The expression was strongest in the muscular layer and the transitional epithelial cells were also positive, whereas the lamina propria was negative. The genital tubercle was strongly positive at E13 (Figure 1c) and the urethra was positive at E15. In testis, a weak signal for Tacc3 mRNA could be detected from E17 to P3. Thereafter, the expression gradually increased and reached a very high level in adult testis. Most of the seminiferous tubules exhibited a strong signal in adult. The strongly positive cells were primary and secondary spermatocytes and a lower signal was seen in early round spermatids (Figures 4j, 4k, and 4l). The germ cells at earlier stages of development, Sertoli cells, and interstitial Leydig cells were negative. In ovary, moderate expression was seen in the interstitial cells and germinal epithelium at the age of 1 and 2 weeks and subsequently the expression diminished to nondetectable levels. Strong expression was observed in oocytes of primordial, primary, and secondary follicle stages at the age of 1 and 2 weeks and in adult (Figures 4n–4p). A weak expression was seen also in granulosa cells of follicles at the same stages. Ovary was not studied during embryonic development.
Strong expression of Tacc3 mRNA is seen in limb bud (lb) at E11 (
Cardiovascular System
From E11 to P1.5, Tacc3 mRNA was expressed weakly in heart muscle cells and from E13 to P1.5 in the walls of large vessels such as the aorta (Figure 1c).
Lymphatic System
In thymus, very strong expression was seen from E13 to P1.5 (Figures 1g, 1h, 4g, and 4h). During embryonic development the signal appeared in both medullary and cortical thymocytes. Postnatally, Tacc3 mRNA was located abundantly in the thymocytes of the outer cortex and a weak signal was seen also in the medulla. After P1.5, the expression declined. However, low levels were still observed in adult. In the spleen, Tacc3 was expressed both in the white and red pulp from E17 to P3 and low signal was present in the red pulp in adult (not shown).
Integumentary System
In skin, we found Tacc3 mRNA as early as E13, and the expression continued throughout embryonic development and could still be seen in the newborn mouse but not afterwards. The signal was particularly strong in the epidermis and in the hair follicles and vibrissae in the dermis (Figure 4i).
Musculoskeletal System
In somites, clear expression was observed at E10 (not shown). In limb bud, moderate expression could be seen at E10 and E11 (Figures 1b, 4a, and 4b). The branchial arches, which later form the chin, upper jaw, part of the auditory ossicles and hyoid bone, were positive at E11 (Figure 1b). Tacc3 mRNA was found in the mesenchyme adjacent to the primordial bones (in limbs shown in Figures 1d and 1f), in the developing ligaments, and in intervertebral discs (Figures 4d and 4e) from E13 to E17. In the developing muscles there was a clear expression from E13 to E17 (not shown).
Organs of Special Sense
In developing eye, Tacc3 mRNA expression was observed from E13 onwards and the expression was still detectable at P14, although absent in the adult (Figures 1d, 1f, and 2m–2r). The mRNA was located in the inner nuclear layer of retina (Figure 2l). In the developing inner ear, the expression was seen at E13 and E15 in mesenchymal cells (Figures 1d, 1f, and 2j), and by E17 the expression had ceased. The expression was detected in olfactory epithelial cells from E13 to P1.5 (Figures 1c and 2k).
Endocrine System
In thyroid gland, Tacc3 mRNA expression was seen at E17 (not shown). Weak expression was detected in adrenal gland at E15 and postnatally (P1.5) (not shown). Low expression was seen in pituitary from E13 to P1.5 (not shown).
Brown Fat
Brown fat showed low expression of Tacc3 mRNA at E17 and later was negative (Figure 1g).
Placenta and Fetal Membranes
We studied placenta at E9 and E10. There was a clear signal in all layers of placenta: in the chorionic plate, the labyrinth region, in spongiotrophoblasts, and more weakly in decidual cells on the maternal side (not shown). The giant trophoblasts were weakly positive. In fetal membranes, moderate expression was seen in yolk sac (not shown).
Tacc3 Immunoreactivity in Mouse Embryos and Postnatally
We studied Tacc3 immunoreactivity (IR) during murine development at embryonic days 9 (E9), 13 (E13), and 16 (E16) and 2 days after birth (P2). Tacc3 IR was nuclear and often localized in a distinct punctate pattern. The antibody gave cytoplasmic staining in some cells. However, this staining could not be abolished by presaturation with the peptide used for immunization, thus indicating nonspecific signal.
On the embryonic side of the placenta, protein expression was detected in the cells of the chorionic plate, labyrinthine part, and in spongiotrophoblasts of the basal layer (Figure 5a). In trophoblast giant cells, Tacc3 was localized to distinct points in the nucleus (Figure 5b). The staining was seen in nuclei of endothelial cells of the placental vessels. Tacc3 was expressed in yolk sac and amnion (Figure 5c).
Immunocytochemical staining for Tacc3 in E9, E13, and E16 mouse embryo shows that the protein is expressed in several tissues and is located in the nucleus. Nuclear staining is seen on the embryonic side of placenta in the cells of chorionic plate (cp) and labyrinthine part (la) as well as in the spongiotrophoblasts (stb) of basal layer. Arrowheads point to the labeled nuclei of trophoblast giant cells (
Tacc3 IR could be detected widely in mesenchymal cells at E9. In the neuroepithelium of the neural tube, most of the neuroblasts expressed Tacc3 (Figure 5d). The epithelium of developing intestinal canal and the surrounding mesenchymal cells were also labeled at E9.
At E13, the most intense immunoreactivity in the CNS was seen in the proliferating neocortical neuroepithelium (Figure 5e). Tacc3 IR was also detected in the nuclei of the neuroblasts of differentiating field. A weaker reaction was also observed in striatial, rhinencephalic, hippocampal, pallidal, tegmental, and tectal neuroepithelium and in differentiating field of thalamus, hypothalamus, and spinal cord. At this time, the cerebellar neuroepithelium appeared negative. No labeling was detectable in the dorsal root ganglia.
The majority of the hepatocytes exhibited Tacc3 IR at E13. The staining was moderate and located in the nucleus. At the same time, the intestinal epithelia did not express the protein, whereas weak staining was seen in the muscle cells of the intestinal wall. Similarly, the nuclei of pancreatic acinar cells were negative and the mesenchymal cells were weakly positive. Bronchial epithelium did not show any Tacc3 IR, whereas weak staining was observed in the mesenchymal cells of lung. In the genital tubercle, the epithelial cells covering the tubercle exhibited strong Tacc3 IR. In heart, weak staining was observed in the nuclei of heart muscle cells. The nuclei of endothelial cells of the vessels were stained. In thymus, a large number of cells were labeled. In the segmental interzone of backbone, future intervertebral disc, the nuclei exhibited weak dotted staining. In olfactory epithelium, weak nuclear staining was seen.
At E16, intense immunoreactivity for Tacc3 was detected in neuroepithelium and especially in the subventricular zone surrounding the lateral ventricles (Figure 5j). The nuclei of neuroblasts in differentiating fields of cerebral hemispheres exhibited weak labeling. In cerebellar hemispheres, moderate nuclear staining was observed in differentiating field (Figure 5i). At this stage, stained nuclei could be seen in both the cerebellar neuroepithelium and the external germinal layer. The ependyma of the choroid plexus was also positive. In spinal cord the neuroblasts displayed Tacc3 IR (Figure 5o).
Expression in the digestive system changed between E13 and E16. The intestinal epithelium expressed Tacc3 protein at E16 (Figure 5l), as did the pancreatic acinar cells (Figure 5m). The nuclei of muscle cells of the intestinal wall were also stained. Most of the cells in the liver expressed Tacc3 protein (Figure 5i). In salivary gland, both epithelial and mesenchymal cells were stained. The bronchial epithelial cells and mesenchymal cells of lung exhibited moderate Tacc3 IR (Figure 5g). The nuclei of endothelial cells of vessels (Figure 5f) and muscle cells of heart continued to express Tacc3 protein at E16. Moderate labeling was seen in thymocytes. The nuclei of striated muscle cells showed strong immunoreactivity (Figure 5h). In eye, moderately stained nuclei were observed in the inner nuclear layer of retina (Figure 5n). Nuclear staining was also seen in the single cells of cochlear epithelium of the inner ear and in brown fat and thyroid gland.
Postnatally, at P2 (2 days after birth), Tacc3 IR was observed widely in cerebral and cerebellar sections. In cerebellum, strong staining was observed in the nuclei of cells of the external germinal layer, in agreement with the in situ hybridization results. Separate stained cells were also seen in differentiating field (Figure 6d). The cerebral areas with strong Tacc3 IR included thalamic nuclei, hippocampus, meninges (arachnoid and pia mater), and ependyma. In eye, the most intensively stained cells were still the neuroblasts of the inner retinal layer. However, positive cells were also seen among the neuroblasts of the outer nuclear layer. Immunoreactivity was observed in corpus ciliare and epithelial cells of Hardenian glands. In peripheral nerves, the nuclei of Schwann cells were positive.
Immunohistochemical staining for Tacc3 in tissues of 2-day-old mouse. In bile duct, the columnar epithelial cells are intensively stained (arrowheads) (
In small intestine, Tacc3 was located in the nuclei of epithelial cells, cells of lamina propria, and muscle cells. In liver, a minority of hepatocytes were weakly positive, and moderately stained granulocytes and lymphocytes were also detected. In bile duct and gallbladder, the columnar epithelial cells showed strong Tacc3 IR (Figure 6a). The antibody continued to display strong staining in the pancreatic acinar cells. In kidney, labeled nuclei were seen in the epithelium of collecting tubules and loops of Henle, in both cortex and medulla. In glomeruli, the nuclei of podocytes were stained. Transitional epithelial cells of the renal pelvis expressed Tacc3 intensely (Figure 6b).
Tacc3 was seen postnatally in thymus and spleen. Tacc3 IR was strong in cortical thymocytes of the outermost layers of cortex (Figure 6c). Epithelial-reticular cells and thymocytes in medulla were also weakly stained. In spleen, some lymphocytes were moderately stained. In skin, the basal layer of epidermis, hair follicles, and the striated muscle cells expressed Tacc3. The nuclei of adipocytes were positive (Figure 6e).
At P2, labeling was seen in the epithelial cells of endometrium and the smooth muscle cells of myometrium of the uterus (Figure 6f). Both ovary and testis expressed Tacc3 IR at this point (Figures 6g and 6h). In ovary, strong labeling was evident in oocytes, granulosa cells of primordial follicles, interstitial cells, and cells of the germinal epithelium. In testis, the strongly stained cells included Sertoli cells, primitive germ cells, interstitial cells, peritubular cells, and fibroblasts of the tunica albuginea.
Discussion
The present study describes the expression of Tacc3 mRNA during embryonic and postnatal development and in adult mice to shed light on the possible functions of Tacc3 during development and in adult tissues. Immunocytochemistry was performed at certain developmental time points to show Tacc3 protein expression. Previous studies have shown the importance of Arnt and Arnt2 in development (reviewed by Rowlands and Gustafsson 1997; Crews and Fan 1999; Gu et al. 2000). Tacc3 has also been implicated in AhR and Hif-1α signaling pathways, which are known to be crucial for normal development (Fernandez-Salguero et al. 1995; Schmidt et al. 1996; Iyer et al. 1998; Sadek et al. 2000). On the other hand, Tacc3 is a member of a novel and growing family of transforming acidic coiled-coil-containing proteins. Members of the TACC family have been shown to have cell-transforming capacity and are candidates for cell growth regulators (Still et al. 1999a,b; Chen et al. 2000). The coiled-coil C-terminus of Tacc3 has a high degree of homology with TACC1 and TACC2. TACC1 was identified from a breast cancer amplicon, is highly expressed during early embryogenesis, and aberrant expression causes alterations in cell morphology and cell growth (Still et al. 1999a). Other reported members of this family appear to be involved in development (Stebbins-Boaz et al. 1999; Chen et al. 2000; Gergely et al. 2000a,b). The role for Tacc3 during development may in part be related to AhR/ARNT/ARNT2/HIF signaling pathways and may in part be associated with other functions.
Expression of Tacc3 was detected in tissues of endodermal, ectodermal, and mesodermal origin. The expression in the CNS and in peripheral organs of mouse is highest and most widespread during fetal development from E13 onwards and continuing to 1.5 days after birth. This period of mouse development is the time of most active proliferation and differentiation in many organs. Because the expression of Tacc3 is associated with regions of rapid proliferation, it can be suggested that Tacc3 has a role in cell proliferation during mouse development. This is supported by studies suggesting a role for Tacc3 in interacting with microtubules during mitosis (Gergely et al. 2000a). In addition, a high rate of p53-mediated apoptosis in hematopoietic cells is evident in Tacc3 knockout mice (Piekorz et al. 2002).
In the CNS, the development of a properly functioning neural network requires coordinated birth, migration, and death of neurons. The expression of Tacc3 mRNA in the proliferating neuroepithelium of neural tube and later in primary and secondary germinal matrices surrounding ventricles and in external granular layer of cerebellum, at the time when proliferation of neuronal cells occurs, suggests an important role for Tacc3 in birth of neural cells. Tacc3 IR was also seen in differentiating neurons, although at lower levels. When migration and further differentiation of neurons occur after the cells have left the germinal matrices, there is a downregulation of Tacc3, also supporting the role for Tacc3 at a temporally limited period of proliferation and differentiation. Current studies (Sadek et al. unpublished) have shown an upregulation of Tacc3 during early differentiation of PC12 cells into neurons and NIH 3T3-LI into adipocytes, and the levels diminish as the cells become terminally differentiated. Similarly, Tacc3 has been shown to be highly expressed during proliferation and at low level during differentiation (Raouf and Seth 2002). We have also shown that Tacc3 IR is present in adult human neurons, suggesting that Tacc3 may have additional functions in nonproliferating cells (Sadek et al. unpublished).
It is interesting that Tacc3 is expressed highly and simultaneously with Hif-1α in the neuroepithelium of neural tube and that Hif-1a-/- embryos manifested failure of neural tube closure with cystic degeneration and prolapse of the neural folds at this time (Iyer et al. 1998; Jain et al. 1998). Tacc3 expression was also temporally and spatially parallel with the expression of its interaction partners Arnt and Arnt2 and with AhR in developing neuroepithelium (Abbott et al. 1995; Abbott and Probst 1995; Hirose et al. 1996; Jain et al. 1998). The simultaneous expression of Tacc3 with these genes enables us to speculate that Tacc3 is needed for neural development, perhaps through participation in AhR/Arnt/Arnt2/Hif signaling pathways. However, Arnt and Tacc3 are not always expressed in parallel in the nervous system. For example, in the dorsal root ganglia, Arnt mRNA expression is very strong, whereas no Tacc3 mRNA is observed (Sadek et al. 2000).
Tacc3 was expressed in several peripheral organs during embryogenesis. For example, in endoderm-derived hepatocytes, expression is high at the time of active proliferation. At E11, when expression starts, the size of the liver rapidly enlarges several-fold as the hematopoietic activity is transferred to liver from the yolk sac and as it matures functionally in several ways (reviewed by Zaret 1996). Interestingly, AhR and Arnt are expressed contemporarily with Tacc3 in embryonic mouse liver (Abbott et al. 1995; Abbott and Probst 1995; Jain et al. 1998). We have previously shown that at E13 and E15 most mouse hepatocytes express both Tacc3 and Arnt mRNA, suggesting that both transcripts are present in the same cells (Sadek et al. 2000).
In contrast to the other organs studied, Tacc3 mRNA levels in testis were low during prenatal and early postnatal development. However, when the mice matured and became fertile, Tacc3 mRNA levels increased remarkably. Tacc3 mRNA was highly expressed in primary spermatocytes and secondary spermatocytes, which are going through first and second meiotic divisions. High expression of TACC3 has also been detected in human testis with Northern analysis and immunocytochemistry (Still et al. 1999a,b; Sadek et al. in press). These findings suggest that Tacc3 has an important role in mammalian spermatogenesis. In ovary, expression of Tacc3 was prominent postnatally and in adult. High expression of two forms of Tacc3 in ovary was recently reported (Hao et al. 2002). Our results are somewhat divergent from those published, as we saw a wider distribution of Tacc3 mRNA and protein in developing and adult ovary than was reported by Hao et al. The high expression of Tacc3 in oocytes suggests that it may have a role in oogenesis. The essential role for Tacc3 in spermatogenesis and oogenesis is supported by the finding that both sexes of combined Tacc3/p53 knockout mice that survive to adulthood are infertile (Piekorz et al. 2002).
In conclusion, the expression of Tacc3 is associated with regions of rapid proliferation, which suggests that it has a role in cell proliferation during normal mouse development. In addition, the high levels of Tacc3 during spermatogenesis and oogenesis provide evidence for a role in these pathways in the adult. Further studies will clarify the exact role of Tacc3 in different tissues.
Footnotes
Acknowledgements
Supported by the Medical Research Fund of Tampere University Hospital, the Research and Science Foundation of Farmos, the Finnish Medical Foundation, and the Swedish Cancer Fund.
We thank Ulla-Margit Jukarainen and Riika Salmela for competent technical assistance. We thank Dr M. Tujague for his help with antibody preparation and testing.