Generation of tyrosine hydroxylase-immunoreactive neurons in ventral mesencephalic tissue of Nurr1 deficient mice

Developmental Brain Research 133 (2002) 37–47 www.elsevier.com / locate / bres

Research report

Generation of tyrosine hydroxylase-immunoreactive neurons in ventral mesencephalic tissue of Nurr1 deficient mice a a, * ¨ ¨ Nina Tornqvist , Elisabet Hermanson b , Thomas Perlmann b , Ingrid Stromberg b

a Department of Neuroscience, Karolinska Institutet, S-171 77 Stockholm, Sweden Ludwig Institute for Cancer Research, Stockholm Branch, Karolinska Institutet, Stockholm, Sweden

Accepted 29 November 2001

Abstract Nurr1 is an orphan nuclear receptor belonging to the family of evolutionary conserved steroid / thyroid hormone receptors. It has been shown that Nurr1 is required for development of ventral mesencephalic dopaminergic cells in vivo and that Nurr1 regulates the tyrosine hydroxylase (TH) gene. The aim of this study was to investigate the possibility of finding ventral mesencephalic TH-positive neurons in Nurr1 deficient tissue when developed in the presence of wild type (WT) striatum. Therefore, fetal ventral mesencephalic tissue from embryonic day (E) 9.5–10.5 fetuses from Nurr1 mutant mice was co-cultured with lateral ganglionic eminence (LGE) from WT fetuses using the ‘roller-drum’ culture technique. TH-immunohistochemistry revealed similar number of positive neurons in WT, heterozygous, and Nurr1 deficient tissue, respectively. When ventral mesencephalon, dissected from E10.5 fetuses, was cultured alone without the presence of LGE, significantly more TH-immunoreactive neurons were found in WT and Nurr1 1 / 2 than that seen in Nurr1 2 / 2 cultures. In single ventral mesencephalic cultures dissected from E15.5, TH-positive neurons were found in all tissue cultures derived from knockout animals. Interestingly, the formation of TH-positive nerve fiber bundles was obvious in WT cultures while not observed in cultures of knockout tissue. When ventral mesencephalon was cultured alone in serum-free medium, almost no TH-positive neurons were found in cultures of knockout tissue. The addition of the growth factors epidermal growth factor and fibroblast growth factor-8 did not induce TH-immunoreactivity in serum-free Nurr1 2 / 2 tissue cultures. In conclusion, TH-positive neurons may be generated in ventral mesencephalic tissue of Nurr1 deficient mice, suggesting that Nurr1 is not required for TH gene expression in ventral midbrain in vitro.  2002 Elsevier Science B.V. All rights reserved. Theme: Development and degeneration Topic: Developmental genetics Keywords: Nurr1; Substantia nigra; Primary tissue culture

1. Introduction Nurr1 is an orphan nuclear receptor, which belongs to the steroid / thyroid hormone receptor superfamily [11]. It can bind to DNA target genes as a monomer or as a heterodimer with the retinoid X receptor [6,15,24]. Nurr1 mRNA is widely expressed in the brain and high levels have been found in the piriform and enthorinal cortices, hippocampus, medial habenular, paraventricular thalamic nuclei, and cerebellum [17,26,30]. Furthermore, Nurr1 is expressed in the ventral midbrain area [17,26,30]. It has been shown that Nurr1 mRNA expression is co-localized *Corresponding author. Tel.: 146-8-7287-087; fax: 146-8-7287-437. ¨ E-mail address: [email protected] (I. Stromberg).

with the vast majority of tyrosine hydroxylase (TH)-immunoreactive neurons in the retrorubral field, the substantia nigra, and the ventral tegmental area (VTA), i.e., A8, A9, and A10 neurons, respectively [1,2,4]. No colocalization of Nurr1 and TH is found in arcuate nucleus [1,2]. Nurr1 expression disappears after a lesion of the dopaminergic neurons in ventral mesencephalon, which further confirms a co-localization of Nurr1 and dopamine [30]. During development, Nurr1 mRNA appears at embryonic day 10.5 (E10.5) in the ventral aspect of the mesencephalic flexure in mice [29]. Nurr1 expression occurs 1 day earlier than the appearance of TH, which is seen at E11.5 [7,20]. Dopaminergic progenitor cells are induced by fibroblast growth factor 8 (FGF8) and sonic

0165-3806 / 02 / $ – see front matter  2002 Elsevier Science B.V. All rights reserved. PII: S0165-3806( 01 )00317-0

¨ et al. / Developmental Brain Research 133 (2002) 37 – 47 N. Tornqvist

38

hedgehog (Shh) which are expressed in the mid-hindbrain boundary and ventral neural tube, respectively [9,10,23,27]. Other early markers for midbrain dopamine neurons include retinaldehyde dehydrogenase 1 (RALDH1, also called aldehyde dehydrogenase 2, AHD2 [13]) and the homeodomain transcription factor Ptx3 [18,22], which are induced at E9.5 and E11.5, respectively [19,22]. In Nurr1 deficient mice TH-positive cells are absent in ventral mesencephalon [1,12,28], while Ptx3 and RALDH1 are expressed in early development [18,22]. However, RALDH1-immunoreactivity has disappeared at E15.5 in Nurr1 deficient mice, and Ptx3 expression is localized only to the medial portions of ventral mesencephalon at this stage of development [22]. TH is the rate limiting enzyme in the production of dopamine, and thus the lack of TH in the ventral mesencephalon of Nurr1 deficient brains results in absence of dopamine in the striatum [12,18,28]. In heterozygous animals the dopamine levels are significantly reduced [12,18], suggesting a deficiency of the nigrostriatal dopamine system. Furthermore, tracing studies show diverging results when using retrograde tracers to visualize a possible nigrostriatal pathway in Nurr1 mutants. Using fluorogold, there is no evidence of nigrostriatal connections while using DiI such a connection can be demonstrated [22,25]. In addition, the expression of the early marker of dopamine neurons Ptx3, shows a more medial location in Nurr1 deficient mice than that found in WT [22]. Thus, the reduced levels of dopamine in the striatum of heterozygous animals and the failure of normal distribution of the early marker Ptx3 may indicate a deficiency in migration of TH-positive neurons when lacking the Nurr1 gene. The fate of these neurons is not known at time points after birth, since the Nurr1 deficient mice die after birth. The

aim of this study was to investigate whether TH-positive neurons may be generated in Nurr1 deficient tissue in the presence or absence of normal fetal striatal environment. Furthermore, ventral mesencephalic tissue derived from Nurr1 mutants was cultured in serum-free or serum-containing media to investigate the influence of putative factors present in serum.

2. Materials and methods

2.1. Animals Fetal VM tissue was obtained from pregnant, heterozygous Nurr1 mutant mice while fetal tissue dissected for lateral ganglionic eminence (LGE) was derived from pregnant wild type animals. All animals were kept under a 12:12 h day–night cycle and provided with free access to food pellets and water.

2.2. Cell cultures Primary cell cultures were performed using the ‘rollerdrum’ method [8] or as free-floating cultures. Using the roller drum method, tissue pieces were cultured either as ventral mesencephalon / LGE co-cultures or as single ventral mesencephalon. Free-floating tissue cultures were performed as single cultures of fetal ventral mesencephalon. Fetuses from embryonic day (E) 9.5, 10.5, or 15.5 were obtained from pregnant Nurr1 mutant mice and placed in sterile petri dishes (Table 1). The ventral mesencephalic area was dissected under a dissection microscope in Dulbecco’s modified Eagle’s medium (DMEM; Gibco) under sterile conditions. Ventral

Table 1 Treatment of fetal ventral mesencephalic (VM) slice cultures dissected from Nurr1 knockout mice Tissue

Fetal stage

Type of culture

Number of cultures

Treatment

Days in vitro

VM1LGE

VM: E9.5–10.5 LGE: E13.5–E14.5

Plated on glass

2 / 2: 5 1 / 2: 14 WT: 10

With serum

7

VM

E10.5

Plated on glass

2 / 2: 7 1 / 2: 14 WT: 12

With serum

7

VM

E15.5

Plated on glass

2 / 2: 5 1 / 2: 15 WT: 5

With serum

7

VM

E10.5

Free-floating

2 / 2: 12 1 / 2: 49 WT: 20

With serum

7

VM

E10.5

Free-floating

2 / 2: 16 1 / 2: 20 WT: 9

Serum-free

7

VM

E10.5

Free-floating

2 / 2: 10 1 / 2: 16 WT: 4

Serum-free1 EGF1FGF-8

7

¨ et al. / Developmental Brain Research 133 (2002) 37 – 47 N. Tornqvist

mesencephalon was dissected bilaterally with the mesencephalic flexure as landmark. The dissected tissue piece was plated as one piece when dissected from E9.5, cut in the midline into two pieces when dissected from E10.5, and into four pieces from E15.5 fetuses. LGE was dissected from E13.5 to 14.5 wild-type (WT) fetuses. One piece from the anterior LGE was used in each double-culture. Instruments were sterilized after each dissected fetus using 70% ethanol and a glass bead sterilizer holding 250 8C. Remaining parts of each fetus were saved and processed for DNA typing as described elsewhere [28]. After dissection the tissue pieces were kept in sterile DMEM until plated. The tissue pieces were placed on sterilized and poly-D-lysine (PDL) coated coverslips (243 12 mm) in a drop of chicken plasma, which was quickly stirred together with a drop of thrombin so that the surface of the coverslip was completely covered. For co-cultures of ventral mesencephalon and LGE, one tissue piece from each region was placed adjacent to each other. Fetal ventral mesencephalon from E15.5 was cultured as two tissue pieces from the same fetus placed on each coverslip. After drying the plasma / thrombin clot for approx. 15 min, the coverslips were placed in 15-ml Falcon tubes with 0.9 ml medium containing 0.1% antibiotics (10 000 units / ml penicillin, 10 mg / ml streptomycin, 25 mg / ml amphotericin; Gibco). The medium was prepared from DMEM, Hanks’ balanced salt solution (HBSS, Gibco), fetal bovine serum (Gibco, the final serum content of the medium used was 10%), glucose (Gibco), and HEPES (Gibco). In free-floating cultures, the tissue pieces were placed directly into the Falcon tubes containing 1 ml medium. The explants were cultured either in serum-containing medium (as described above) or serum-free medium. The serum-free medium was supplemented with 0.5 mM Lglutamine, 25 mM glutamate and 1% B27 supplement. Furthermore, free-floating cultures were treated with epidermal growth factor (EGF; Sigma) and fibroblast growth factor-8 (recombinant mouse FGF-8; R&D Systems, UK) at the concentrations of 10 and 100 ng / ml, respectively, in serum-free medium. The Falcon tubes including the cultures either plated on glass or free-floating were placed in a roller drum holding an even rotational speed of 1 turn every second minute. The roller drum was placed in an incubator keeping the temperature at 378C and the CO 2 concentration to 5%. Survival time for all cultures was 7 days, and the media were changed after 3–4 days.

39

Freez, goat anti-rabbit diluted 1:300 in 0.1 M PBS containing 1% Triton X-100). The cultures plated on coverslips were incubated in the primary antibodies in a humid chamber, and the free-floating cultures were incubated directly into the primary antibody-solution for 48 h. The cultures were then rinsed in PBS before Cy3-conjugated secondary antibodies (diluted 1:200 in 0.1 M PBS

2.3. Immunohistochemistry The cultures were fixed in 2% paraformaldehyde in 0.1 M phosphate buffer for 1 h at room temperature. They were then rinsed in phosphate-buffered saline (PBS) for 3310 min and processed for indirect immunohistochemistry using antibodies against tyrosine hydroxylase (TH; Pel

Fig. 1. Ventral mesencephalic portion of nigrostriatal co-cultures dissected from Nurr1 deficient (a), heterozygous (b), and wild-type (c) mesencephalic tissue. The stage of the fetal ventral mesencephalic tissue was E9.5 at dissection and survival time was 7 days. TH-immunohistochemistry revealed numerous neurons in all genotypes. Scale bar: a–c5100 mm.

40

¨ et al. / Developmental Brain Research 133 (2002) 37 – 47 N. Tornqvist

Fig. 2. When estimating the number of TH-positive neurons in ventral mesencephalic–lateral ganglionic eminence co-cultures from Nurr1 deficient mice there was no significant difference in number of TH-positive neurons for different genotypes (a). Significantly increased number of TH-positive neurons were found in heterozygous and wild-type tissue cultures compared to that found in 2 / 2 cultures when the ventral mesencephalic tissue was cultured as single pieces without the presence of lateral ganglionic eminence (b). When ventral mesencephalic tissue was dissected from E15.5 fetuses and cultured for 7 days, TH-positive neurons were found in all tissue derived from Nurr1 2 / 2 fetuses although significantly more TH-positive neurons were found in 1 / 2 and 1 / 1 tissue cultures. **P,0.01.

¨ et al. / Developmental Brain Research 133 (2002) 37 – 47 N. Tornqvist

containing 1% Triton X-100; Jackson ImmunoResearch Labs.) were applied. Incubation was performed for 1 h at room temperature. After additional rinsing the cultures were mounted in 90% glycerin diluted in PBS. In addition,

41

some cultures were processed for immunohistochemistry using anti-mouse TH antibodies (Boehringer Mannheim) and Cy3- (goat anti-mouse, Jackson ImmunoResearch Labs.) conjugated antibodies, and further processed for

Fig. 3. TH-immunohistochemistry from ventral mesencephalic cultures derived from wildtype (a), heterozygous (b), and Nurr1 deficient (c–d) tissue. A distinct formation of a densely packed TH-positive nerve fiber bundle was found in wild-type cultures. In heterozygous tissue cultures TH-positive nerve fibers were frequently found, but these had formed a nerve fiber network rather than a dense fiber bundle (b). In Nurr1 knockout tissue cultures TH-positive nerve fibers never formed a nerve fiber bundle (c), and at higher magnification a nerve fiber network was obvious (d). Scale bar: a–b5100 mm, c5200 mm, d580 mm.

42

¨ et al. / Developmental Brain Research 133 (2002) 37 – 47 N. Tornqvist

RALDH1 (raised in rabbit [14], and Cy2- (anti-rabbit) conjugated secondary antibodies (Jackson ImmunoResearch Labs.).

2.4. Statistics Cell counts of TH-positive neurons were performed. The thickness of the tissue cultures did not permit calculation of the actual cell number, since the cell density was too high and it became difficult to distinguish individual neurons. Therefore, TH-positive cell numbers were estimated in approximate numbers. Below 50 TH-positive neurons was rated as 1, between 50 and 150 cells counted were given 2, 150–250 represent 3, 4 represents 250–350 cells, and 5 represents more than 350 TH-positive neurons. Comparisons were made using one-way analysis of variance (ANOVA) followed by Fisher’s PLSD post hoc test and expressed as means6S.E.M.

3. Results

3.1. Ventral mesencephalic ( E9.5 – 10.5) –LGE ( E13.5 – 14.5) co-cultures TH-positive neurons were found in nigrostriatal cocultures from knockout, heterozygous, and WT animals (Fig. 1). There were slightly more TH-positive neurons found in WT than that found in Nurr1 1 / 2 and in 2 / 2 (Fig. 2a). Nerve fibers were found in all ventral mesencephalic portions of the co-cultures, but the neurites never reached into the striatal portion of the co-explants. In cultures from WT tissue the nerve fibers had formed a bundle (seen in eight out of 10 cultures). This bundle was found in two out of 14 cultures from Nurr1 1 / 2 and in none (n55) of the cultures of Nurr1 2 / 2 tissue (Fig. 3). There was no obvious difference in the number of THpositive neurons when comparing cultures from E9.5 and E10.5, and therefore cell counts for TH-positive neurons were pooled.

3.2. Ventral mesencephalic single cultures from E10.5 Since TH-positive neurons were found in all ventral mesencephalic tissue when co-cultured with LGE, the presence of TH-positive neurons was studied in single cultures of ventral mesencephalon. In single ventral mesencephalic cultures from E10.5 fetuses the results revealed that TH-positive neurons were present in all but one culture from Nurr1 2 / 2 tissue (Fig. 4). However, significantly more neurons were found in WT and Nurr1 1 / 2 tissue cultures than that found in Nurr1 2 / 2 cultures (P,0.01; Fig. 2b). TH-positive neurons were located in a cluster within the tissue piece, which was similar to that found in co-cultures with LGE. The formation of a nerve fiber bundle was not as frequently

Fig. 4. TH-immunohistochemistry revealed that numerous neurons were found in single free-floating tissue cultures from Nurr1 deficient (a), heterozygous (b), and wild-type (c) tissue cultured for 7 days in vitro in serum-containing medium. Scale bar: a–c5200 mm.

found in single cultures as that seen in co-cultures, although it was sometimes found in WT tissue.

3.3. Ventral mesencephalic single cultures from E15.5 The presence of TH-positive neurons in single cultures derived from E10.5 raised the question of whether the early stage at dissection had any influence on the TH expression. Therefore later stages of fetuses were used for

¨ et al. / Developmental Brain Research 133 (2002) 37 – 47 N. Tornqvist

tissue culture. In ventral mesencephalic single cultures dissected from E15.5, TH-positive neurons were found in all cultures from Nurr1 2 / 2 tissue. The number of THpositive cells in Nurr1 2 / 2 tissue was significantly lower than that seen in Nurr1 heterozygous and WT tissue cultures (Fig. 2c). Moreover, co-existence of RALDH1 and TH was found in Nurr1 2 / 2 (Fig. 5) as well as in WT cultures.

3.4. Ventral mesencephalic free-floating single cultures in serum-free vs. serum-containing medium Cultures were performed free-floating to avoid interference with the plasma / thrombin substrate used for plating. The generation of TH-positive neurons in Nurr1 2 / 2 tissue (E10.5) cultured in serum-free medium was poor (Fig. 6a). In most serum-free cultures of knockout tissue no TH-positivity was found although the tissue was viable. In some cultures (n52 out of 16) from 2 / 2 tissue, a few (up to 30) TH-positive neurons were found (Fig. 7). TH-positive neurons were found in cultures from WT and heterozygous fetuses. In contrast, in ventral mesencephalic tissue cultured free-floating with serum-containing medium

43

the number of TH-positive neurons was not significantly different in Nurr1 knockout, WT, and 1 / 2 cultures (Fig. 6b). Thus, TH-positive neurons were found also in Nurr1 knockout tissue cultures.

3.5. Ventral mesencephalic free-floating cultures treated with FGF-8 and EGF The addition of the trophic factors FGF-8 and EGF to serum-free medium did not increase the number of THpositive neurons in ventral mesencephalic cultures. Thus, in one out of 10 ventral mesencephalon cultures derived from knockout fetuses, a few TH-positive neurons were found and the rest of the knock-out cultures were THnegative. However, in heterozygous tissue cultures the number of TH-positive neurons was comparable with that seen in WT cultures.

4. Discussion The results from the present study show that TH-positive neurons may be generated in Nurr1 2 / 2 tissue

Fig. 5. Tissue cultures from Nurr1 deficient fetuses dissected and plated at E15.5. At 7 days in vitro TH-positive neurons appeared well developed and some TH-positive neurons co-expressed RALDH1 (b, c). Scale bar: a5100 mm, b–c560 mm.

44

¨ et al. / Developmental Brain Research 133 (2002) 37 – 47 N. Tornqvist

Fig. 6. Estimated cell counts of TH-positive neurons in ventral mesencephalic tissue cultured free-floating in serum free (a) or serum-containing medium (b). In cultures from Nurr1 deficient mice almost no TH-positive neurons were found (a), while there was no difference in the number of TH-positive neurons found in tissue cultures from the different genotypes when tissue was cultured in serum-containing medium (b). ***P,0.001.

cultures with or without the presence of LGE. In VM single cultures, the presence of serum in the medium was needed to induce TH-positivity in knockout tissue cultures. Furthermore, the formation of TH-positive axon bundles was absent in Nurr1 knockout tissue but frequently found in tissue cultures derived from WT. The expression of Nurr1 gene in A9 and A10 dopamine neurons in normal adult mice is restricted to a degree higher than 95% [2]. Within the dissected piece of tissue used for the cultures also A8 was included. In this nucleus the co-existence of TH and Nurr1 is 91% in normal mice [2]. Although there are no TH-positive neurons found in Nurr1 2 / 2 mice in situ [1,12,28], there is a small subpopulation of dopamine neurons in the ventral mesencephalon that do not normally

express the Nurr1 gene, and may then constitute the TH-positive neurons that were found in these Nurr1 knockout cultures. However, the presence of RALDH1 in TH-positive neurons from cultures derived from Nurr1 2 / 2 tissue, strengthen the possibility that dopamine neurons normally expressing Nurr1 were present in Nurr1 null mutant tissue cultures, since RALDH1 is co-expressed in Nurr1 and TH-positive neurons located in A9 and A10 nuclei [22]. It has been demonstrated that the Nurr1 gene is required to generate TH-positive neurons in ventral mesencephalon [12,18,28]. Although neuronal development, as visualized with neuronal markers such as TUJ1- and NeuN-immunoreactivity, appears normal during early stages, prior to the

¨ et al. / Developmental Brain Research 133 (2002) 37 – 47 N. Tornqvist

45

Fig. 7. Tissue slices cultured free-floating in serum-free medium for 7 days. Tissue cultures dissected from Nurr1 knockout mice (a, b) were mostly TH-negative (a), but in a few cultures some small TH-positive neurons were found (b). In cultures dissected from heterozygous (c) and wild-type (d) mice numerous TH-positive neurons were found. Scale bar: a–d5100 mm.

expected expression of TH, as well as in newborn mice of Nurr1 null mutants, the expression of TH fails to be demonstrated in Nurr1 deficient mice [22,25]. RALDH1, which is another early marker of dopaminergic neurons is present in Nurr1 knockout fetuses as early as E9.5, but has disappeared at E15.5 [22]. It is not clear from previous studies whether the RALDH1-positive neurons have died or lost their RALDH1 expression. Notably, neuronal death occurs in knockout ventral midbrain although it is evident only at later (E18.5) stages, when RALDH1 expression has already disappeared [22]. When dissected tissue from E10.5 fetuses has been cultured for 1 week, the uppermost age of the cultured tissue is E17.5. Thus, neurons in the early stage cultures correspond to an age where apoptosis is found and they may still undergo cell death. The late dissected (E15.5) cultured tissue was assumed to correspond to the stage of newborn when analyzed. Still, THpositive neurons were found in Nurr1 2 / 2 cultures. At this time point the apoptotic profiles have declined in situ [25]. In addition, recent data show TH-positive neurons in

cultures from newborn Nurr1 knockout tissue [5]. Thus, despite evidence for early cell death of developing dopamine neurons in ventral mesencephalon found in earlier in situ studies, the present result suggest that these neurons may be generated, express TH, and survive at least over the stage of newborn. Interestingly, there was a clear difference in the distribution of nerve fibers; in WT cultures the formation of thick, well-defined nerve fiber bundles were found. In 1 / 2 cultures this phenomenon was not as pronounced, while in knockout cultures only a nerve fiber network had been developed. In earlier studies trying to prove the presence of the nigrostriatal pathway in Nurr1 knockout animals using retrograde tracers, the results have been divergent and shown both presence and absence of nigrostriatal connections [22,25]. The reason for these discrepancies is unknown and could result from, e.g., different genetic background of the Nurr1 knockout mouse strains or differences in how the gene targeting of Nurr1 was achieved. The present data support the conclusion that

46

¨ et al. / Developmental Brain Research 133 (2002) 37 – 47 N. Tornqvist

dopamine axon pathfinding may require Nurr1 since no pathway was found in knockout tissue. These findings were indeed achieved from the same line of Nurr1 knockout animals previously shown to lack nigrostriatal connections [22]. Thus, formation of the nigrostriatal fiber bundle may exert important cues for long-term survival of dopamine neurons. The mechanisms that induce the TH-expression has been suggested to be dependent on the presence of Shh and FGF-8 [3,9,10,27]. However, for dopamine neurons located in the ventral mesencephalon also Nurr1 gene expression has been suggested to be needed [16]. Furthermore, EGF creates an environment permissive for proliferating progenitor cells [21]. The induction of TH in Nurr1 2 / 2 tissue cultures was not dependent on the presence of their normal striatal target, since TH-positive neurons were found in nigrostriatal co-cultures as well as in single ventral mesencephalic cultures. Knockout tissue responded to serum-treatment by an increase in the number of THpositive neurons, while WT tissue surprisingly did not. Although it cannot be determined from this study which serum-derived compound that influences TH-expression in knockouts, it is probably not FGF8 or EGF since neither of these two factors could induce TH-expression in knockout tissue when added to serum-free cultures. Taken together, TH may be induced in ventral mesencephalic tissue cultures dissected from Nurr1 deficient mice before and after the normal time period during which midbrain dopamine neurons normally are born, since THpositive neurons were found in cultures from both E10.5 (i.e., prior to development) and E15.5 fetuses (postmitotic). Thus, Nurr1 is not absolutely required for generation of TH-positive ventral mesencephalic neurons in vitro. The failure to develop nerve fiber bundles in Nurr1 deficient cultures suggests the necessity of the Nurr1 gene in guiding the formation of the nigrostriatal pathway.

[3] [4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

Acknowledgements [17]

This study was supported by the Swedish Medical Research Council grants 09917, 13233, Magnus Ber¨ gvall’s, Karolinska Institutet’s, and Goran Gustafsson’s Foundations, and the US Army Medical Research grant. E.H. was supported by a fellowship from the Swedish ´ National Network in Neuroscience. We thank Dr. Wallen for valuable comments on the manuscript.

References [1] J.S. Baffi, M. Palkovits, S.O. Castillo, E. Mezey, V.M. Nikodem, Differential expression of tyrosine hydroxylase in catecholaminergic neurons of neonatal wild-type and Nurr1-deficient mice, Neuroscience 93 (1999) 631–642. ¨ ´ B.J. Hoffer, M. Morales, A [2] C. Backman, T. Perlmann, A. Wallen,

[18]

[19]

[20]

[21]

selective group of dopaminergic neurons express Nurr1 in the adult mouse brain, Brain Res. 851 (1999) 125–132. P.H. Crossley, S. Martinez, G.R. Martin, Midbrain development induced by FGF8 in the chick embryo, Nature 380 (1996) 66–68. ¨ K. Fuxe, Evidence for the existence of monoamineA. Dahlstrom, containing neurons in the central nervous system. I. Demonstration of monoamines in the cell bodies of the brain stem neurons, Acta Physiol. Scand. Suppl. 232 (62) (1964) 1–55. J.B. Eells, J.E. Rives, S.K. Yeung, V.M. Nikodem, In vitro regulated expression of tyrosine hydroxylase in ventral midbrain neurons from Nurr1-null mouse pups, J. Neurosci. Res. 64 (2001) 322–330. B.M. Forman, K. Umesono, J. Chen, R.M. Evans, Unique response pathways are established by allosteric interactions among nuclear hormone receptors, Cell 81 (1995) 541–550. ¨ G. Foster, M. Schultzberg, T. Hokfelt, M. Goldstein, H.C. Hemmings, C.C. Ouimet, S.I. Walaas, P. Greengard, Ontogeny of the dopamine and cyclic adenosine-39,59-monophosphate-regulated phosphoprotein (DARPP-32) in the pre- and postnatal mouse central nervous system, Int. J. Dev. Neurosci. 6 (1988) 367–386. ¨ B.H. Gahwiler, M. Capogna, D. Debanne, R.A. McKinney, S.M. Thompson, Ordanotypic slice cultures: a technique has come of age, Trends Neurosci. 20 (1997) 471–477. M. Hynes, J.A. Porter, C. Chiang, D. Chang, M. Tessier-Lavigne, P.A. Beachy, A. Rosenthal, Induction of midbrain dopaminergic neurons by sonic hedgehog, Neuron 15 (1995) 35–44. M. Hynes, K. Poulsen, M. Tessier-Lavigne, A. Rosenthal, Control of neuronal diversity by floor plate: contact-mediated induction of midbrain dopaminergic neurons, Cell 80 (1995) 95–101. S.W. Law, O.M. Conneely, F.J. DeMayo, B.W. O’Malley, Identification of a new brain specific transcription factor, Nurr1, Mol. Endocrinol. 6 (1992) 2129–2135. W.-D. Le, O.M. Conneely, L. Zou, Y. He, O. Saucedo-Cardenas, J. Jankvic, D.R. Mosier, S.H. Appel, Selective agenesis of mesencepahlic dopaminergic neurons in Nurr1-deficient mice, Exp. Neurol. 159 (1999). ¨ H. Li, E. Wagner, P. McCaffery, D. Smith, A. Andreadis, U. Drager, A retinoic acid synthesizing enzyme in ventral retina and telencephalon of the embryonic mouse, Mech. Dev. 95 (2000) 283–289. P. McCaffery, U.C. Drager, High levels of a retinoic acid-generating dehydrogenase in the meso-telencephalic dopamine system, Proc. Natl. Acad. Sci. USA 91 (1994) 7772–7776. T. Perlmann, L. Jansson, A novel pathway for vitamin A signaling mediated by RXR heterodimerization with NGFI-B and Nurr1, Genes Dev. 9 (1995) 769–782. K. Sakurada, M. Oshima-Sakurada, T.D. Palmer, F.H. Gage, Nurr1, an orphan nuclear receptor, is a transcriptional activator of endogenous tyrosine hydroxylase in neural progenitor cells derived from the adult brain, Development 126 (1999) 4017–4026. O. Saucedo-Cardenas, O.M. Conneely, Comparative distribution of NURR1 and NUR77 nuclear receptors in the mouse central nervous system, J. Mol. Neurosci. 7 (1996) 51–63. O. Saucedo-Cardenas, J.D. Quinta-Hau, W.-D. Le, M.P. Smidt, J.J. Cox, F. De Mayo, J.P.H. Burbach, O.M. Conneely, Nurr1 is essential for the induction of the dopaminergic phenotype and the survival of ventral mesencephalic late dopminergic precursor neurons, Proc. Natl. Acad. Sci. USA 95 (1998) 4013–4018. ˆ J.J. Tremblay, J.J. Cox, M.P. Smidt, H.S.A. van Schaick, C. Lanctot, A.A.M. van der Kleij, G. Wolternik, J. Drouin, J.P.H. Burbach, A homeodomain gene Ptx3 has highly restricted brain expression in mesencephalic dopaminergic neurons, Proc. Natl. Acad. Sci. USA 94 (1997) 13305–13310. Y. Solberg, W. Silverman, Y. Pollack, Prenatal ontogeny of tyrosine hydroxylase gene expression in the rat ventral mesencephalon, Dev. Brain Res. 73 (1993) 91–97. A.L. Vescovi, B.A. Reynolds, D.D. Fraser, S. Weiss, bFGF regulates the proliferative fate of unipotent (neuronal) and bipotent (neuronal / atroglial) EGF-generated progenitor cells, Neuron 11 (1993) 951– 966.

¨ et al. / Developmental Brain Research 133 (2002) 37 – 47 N. Tornqvist ´ R.H. Zetterstrom, ¨ [22] A. Wallen, L. Solomin, M. Arvidsson, L. Olson, T. Perlmann, Fate of mesencephalic AHD2-expressing dopamine progenitor cells in Nurr1 mutant mice, Exp. Cell Res. 253 (1999) 737–746. [23] M.Z. Wang, P. Jin, D.A. Bumcrot, V. Margio, A.P. McMahon, E.A. Wang, T. Woolf, K. Pang, Induction of dopaminergic neuron phenotype in the midbrain by sonic hedgehog protein, Nat. Med. 1 (1995) 1184–1188. [24] T.E. Wilson, T.J. Fahrner, J. Milbrandt, The orphan receptors NGFIB and steroidogenic factor 1 establish monomer binding as a third paradigm of nuclear receptor–DNA interaction, Mol. Cell Biol. 13 (1993) 5794–5804. [25] J. Witta, J.S. Baffi, M. Palkovits, E. Mezey, S.O. Castillo, V.M. Nikodem, Nigrostriatal innervation is preserved in Nurr1-null mice, although dopaminergic nerons precursors are arrested from terminal differentiation, Mol. Brain Res. 84 (2000) 67–78. [26] Q. Xiao, S.O. Castillo, V.M. Nikodem, Distribution of messenger RNAs for the orphan nuclear receptors Nurr1 and Nur77 (NGFI-B)

[27]

[28]

[29]

[30]

47

in adult rat brain using in situ hybridization, Neuroscience 75 (1996) 221–230. W. Ye, K. Shimamura, J.L.R. Rubenstein, M.H. Hynes, A. Rosenthal, FGF and Shh signals control dopaminergic and serotonergic cell fate in the anterior neural plate, Cell 93 (1998) 755–766. ¨ R. Zetterstrom, L. Solomin, L. Jansson, B. Hoffer, L. Olson, T. Perlmann, Dopamine neuron agenesis in Nurr1-deficient mice, Science 276 (1997) 248–250. ¨ L. Solomin, T. Mitsiadis, L. Olson, T. Perlmann, R.H. Zetterstrom, Retinoid X receptor heterodimerization and developmental expression distinguish the orphan nuclear resptors NGFI-B, Nurr1, and Nor1, Mol. Endocrinol. 10 (1996) 1656–1666. ¨ R.H. Zetterstrom, R. Williams, T. Perlmann, L. Olson, Cellular expression of the immediate early transcription factors Nurr1 and NFGI-B suggests a gene regulator role in several brain regions including the nigrostriatal dopamine system, Mol. Brain Res. 41 (1996) 111–120.