Neurogenesis and Cell Death in Olfactory Epithelium

Anne L. Calof, * Nobuko Hagiwara, J. David Holcomb, Jeffrey S. Mumm, and Jianyong Shou Department of Anatomy and Neurobiology and the Developmental Biology Center, University of California, Irvine, College of Medicine, Irvine, California 92717-1 275

SUMMARY

The olfactory epithelium (OE) of the mammal is ORNs. The identification of a number of factors that act uniquely suited as a model system for studying how neu- to regulate proliferation and survival of ORNs and their rogenesis and cell death interact to regulate neuron num- precursors suggests that these multiple developmental ber during development and regeneration. To identify stages may serve as control points at which cell number factors regulating neurogenesis and neuronal death in the is regulated by extrinsic factors. In vivo surgical studies, OE, and to determine the mechanisms by which these which have shown that all cell types in the neuronal lin- factors act, investigators studied OE using two major ex- eage of the OE undergo apoptotic cell death, support this perimental paradigms: tissue culture of OE; and ablation idea. These studies, and the possible coregulation of neu- of the olfactory bulb or severing the olfactory nerve in ronal birth and apoptosis in the OE, are discussed. adult animals, procedures that induce cell death and a 0 1996 John Wiley &Sons, Inc. subsequent surge of neurogenesis in the OE in viva These studies characterized the cellular stages in the ol- Keywords: neurogenesis, apoptosis, programmed cell factory receptor neuron (ORN) lineage, leading to the death, neuronal precursor cells, stem cells, transcription realization that at least three distinct stages of prolifer- factors, Mushl, Otx2, growth factors, fibroblast growth ating neuronal precursor cells are employed in generating factors, neurotrophins, transgenic mice.

INTRODUCTION for continual nerve cell renewal led a number of different groups to perform studies that provide ev- The olfactory epithelium (OE)of the mammal is idence suggesting that cell interactions can regulate uniquely suited as a model system for studying how neurogenesis and cell death in the OE in viva neurogenesis and apoptosis, or programmed cell Death of cells in the OE can be upregulated in the death, interact to regulate neuron number during adult mammal by lesioning the axons of ORNs or development and regeneration. Proliferation of ablating their synaptic target tissue, the main olfac- neuronal precursor cells, differentiation of their tory bulb of the brain ( Monti-Graziadei and Graz- progeny into olfactory receptor neurons (ORNs), iadei, 1979; Costanzo and Graziadei, 1983; Michel and death and turnover of ORNs are processes that et al., 1994; Holcomb et al., 1995). This cell death begin during embryonic development in the OE, in turn leads to increased mitotic activity in the and then continue throughout adult life (Graziadei neuronal precursor cells of the epithelium, which and Monti-Graziadei, 1978, 1979). This capacity then produce new ORNs (Camara and Harding, 1984; Schwartz-Levey et al., 1991; Gordon et al., 1995). Received December 4, 1995; accepted December 14, 1995 The ability of the OE to renew its neuronal pop- Journal of Neurobiology, Vol. 30, No. 1, pp. 67-8 1 ( 1996) ulation throughout life raises the question of what 0 1996 John Wiley &Sons, Inc. CCC 0022-3034/96/010067-15 properties of this tissue enable it, as opposed to * To whom correspondence should be addressed (e-mail: most other regions of the mammalian nervous sys- [email protected]). tem, to regenerate neurons. For example, the fact

67 68 CulQfetul that the OE of all vertebrates undergoes continuous sheets within the main body of the explanted tissue; neuron turnover and renewal throughout life postmitotic ORNs, which express the neural cell would seem to guarantee the existence of a neuro- adhesion molecule ( NCAM, a neuron-specific nal stem cell in this tissue. To try to determine if marker in this system), and migrate away from OE this is the case, and to understand the cellular and explants and extend neurites when grown on ap- molecular basis of neuron renewal in the OE, we propriate extracellular matrix substrata ( Calof and and others have established tissue culture systems Lander, 1991; Calof et al., 1994b); and keratin- to facilitate identification of the molecular factors negative, NCAM-negative cells, which like ORNs involved (Noble et al., 1984; Schubert et al., 1985; are migratory, but do not have neurites and, unlike Gonzales et al., 1985; Chuah et al., 1985, 1991; ORNs, incorporate 3H-thymidine and divide in Coon et al., 1989; Calof and Chikaraishi, 1989; culture. In serum-free, defined conditions, these Pixley and Pun, 1990; Calof and Lander, 1991; cells divide once and give rise to two daughter cells, Ronnett et al., 1991; Pixley, 1992; Mahanthappa which differentiate into ORNs and begin express- and Schwarting, 1993; Calof et al., 1994a; Hol- ing NCAM within about 12 h of the terminal S comb et al., 1995). This work led to the identifica- phase; hence we call these cells the immediate neu- tion of a number of factors that can regulate prolif- ronal prectirsors (INPs). INPs are the in vitro eration and survival of OE cells, and in addition, to equivalents of the so-called “globose” basal cells of new insights concerning the cellular stages that lead the OE, which have been shown to be the direct from undifferentiated stem cell to mature ORNs. precursors of ORNs in vivo (cf. Graziadei and In this review we describe what is known con- Monti-Graziadei, 1979; Mackay-Sim and Kittel, cerning cellular stages in the neuronal lineage of 1991; Schwartz-Leveyet al., 1991; Caggiano et al., the OE. Neuronal cell death in the OE following 1994 ). We demonstrated that INPs behave as com- synaptic target tissue ablation in vivo is discussed, mitted neuronal precursor cells, capable of un- and the identification of factors that can regulate dergoing a limited number of amplification divi- this process in vitro, and potentially in vivo, is also sions in response to appropriate exogenous factors described. We outline information gained from in (see below), and as such fit the description of neu- vitro studies on how ORNs are generated from ronal “transit amplifying cells” in the ORN lineage their precursor cells, and what is known concern- ( DeHamer et al., 1994; Hall and Watt, 1989; Pot- ing regulation of this process by extrinsic factors. ten and Loeffler, 1990). Evidence for the existence Finally, the relationship between ORN cell death of a potential neuronal stem cell, which gives rise and neurogenesis is discussed, along with possible to INPs in explant cultures, was also recently ob- bases by which these two processes may be coregu- tained, although no marker specific for this cell has lated. yet been identified (DeHamer et al., 1994; see below). More recently, we performed studies to try to CELLULAR STAGES IN THE ORN determine the true complexity of the ORN lineage. LINEAGE How many precursor cell stages lie between the self-renewing stem cells of the OE and the INPs Much of our detailed knowledge concerning cellu- that are committed to giving rise to ORNs? We lar stages in the ORN lineage comes from tissue gained insight into this question by studying the culture studies, using cell type specific markers to dynamics of expression of several different devel- identify different cells of the OE and 3H-thymidine opmentally regulated transcription factors during and/or BrDU incorporation analysis to examine OE neurogenesis in vivo and in vitro. Two of these, precursor-progeny relationships among cells. Sev- Mash],a mammalian homologue of the Drosoph- eral years ago, we established an explant culture ila achaete-mite proneural genes (Johnson et al., system using embryonic day 14.5-15.5 mouse OE, 1990), and Otx2, a murine homeogene related to in which three major cell types are distinguishable the Drosophila gene orthodenticle (Simeone et al., by antigenic markers, morphology, and differences 1992), have so far proved interesting. in their migratory behavior (Calof and Chikaraishi, Our work on Mash1 expression indicates that 1989; Calof and Lander, 1991). The three cell expression of this transcription factor demarcates types are: basal cells, which express keratin inter- a distinct stage of neuronal precursor in the ORN

mediate filaments and do not migrate in vitro, in- lineage, but that MASHl + cells are not the true stead remaining in tightly associated epithelial stem cells of the OE. Our choice of MASHl as a Birth and Death qj'Olfactory Receptor Neurons 69

potential marker for ORN precursors was initially motivated by two findings from other labs. First, MASHl is expressed in regions of the embryonic murine nervous system where proliferating neuronal precursors are present (Lo et al., 199 1 ) . Sec- ond, and even more significantly, disruption of the Mush1 gene by homologous recombination in mice results in a profound reduction in the numbers of ORNs that are generated during development (neuronal lineages in the autonomic and en- tenc nervous systems are also affected; Guillemot et al., 1993). Usinganantibodyto MASHl, weex- amined MASH 1 expression in embryonic OE cultures, in normal adult mice, and in adult mice at various times following olfactory bulbectomy (surgical removal of the olfactory bulb, which results in the death of ORNs and a subsequent surge in neurogenesis in the OE) (Gordon et al., 1995 ). Our results indicate that MASH 1 is expressed in a population of cells morphologically similar to INPs (in vitro) or globose basal cells (in vivo), but several-fold less abundant. When neurogenesis is transiently induced in the OE in vivo following bulbectomy, a transient surge in the number of MASH 1

after bulbectomy. ( A,C,E) MASH 1 immunofluores- cence is shown; (B,D,F) corresponding phase contrast images. The photos show a marked increase in the num-

ber of MASH 1 + cells (C) on the ipsilateral side of the operated animal, compared to ( E ) the contralateral or (A) unoperated controls. (C,D) Also apparent is the de- crease in overall thickness of the OE (the result of cell loss) on the ipsilateral side of the operated animal (arrows), when compared to either of the controls. Bar- 50 pm. (G) Stained sections such as those shown in (A-

5 005 10 15 20 F) were used to determine the number of MASH 1 + cells per linear distance along the OE on both sides of animals Days Post-Bulbectomy sacrificed at various times following unilateral bulbec- Figure 1 MASHl expressing cells increase in number tomy. To normalize for interanimal variability in the

when neurogenesis is stimulated in viva Adult male mice number of MASH 1 + cells, data were converted to a ratio + were subjected to unilateral (left) bulbectomy and sacri- of MASH 1 cells/mm on the operated side to MASH 1 + ficed at various times from 2- 19 days postsurgery. Several cells/mm on the unoperated side, calculated for each an- unoperated control animals were also sacrificed. Cryostat imal. The data are plotted as a function of time following sections ( 12 pm) through the OE were taken in the hori- bulbectomy. Each data point represents the average of zontal plane (so that both left and right OE were present results obtained from at least three animals +- S.E.M. For on each section), processed for MASH 1 immunoreactiv- each animal, the data were acquired from viewing the ity as described (Gordon et al., !995), and examined by OE lying along the posterior part of the nasal septum, in fluorescence and phase contrast microscopy. (A-F) Pho- multiple sections, covering several millimeter of OE. A tos of the OE (A,B) of an unoperated animal are com- smooth, freehand curve was drawn through the data pared with photos of (C,D) the left (ipsilateral, operated points. (From Gordon et al., Mol. Cell. Neurosci. 6:363- side) and (E,F) the right (contralateral, control side) sep- 379, 0 1995 Academic Press, reprinted with permis- tal OE from a single section of an animal sacrificed 5 days sion.) 70 Culof et a/. expressing cells in the OE ensues (Fig. I). This

property of MASH 1 + cells, taken together with their high 3H-thymidinelabeling index in vitro and in vivo (e.g., in adult, unstimulated OE, -40% of

MASH 1 + cells are labeled with a 2-h pulse of 3H- thymidine administered immediately prior to sacrifice), indicate that MASH 1 ' cells behave, like INPs, as transit amplifying cells in the OE. The expansion of MASH 1 cell numbers in response to mitogenic stimulation (bulbectomy ) is characteristic 0 5 10 15 of the symmetric amplifying divisions of transit Days Post-Bulbectomy amplifying cells, rather than the asymmetric self-

renewing divisions of stem cells (Hall and Watt, Figure 2 MASH 1 + cells proliferate prior to INPs when neurogenesis is induced vivo. Adult male mice were 1989; Potten and Loeffler, 1990); and the high 3H- in subjected to unilateral (left) bulbectomy and sacrificed thymidine labeling index of MASH 1 + cells would at various times from 2-19 days postsurgery. Unoper- be highly unusual in stem cells, which typically ated control animals were also sacrificed. In all cases, an- cannot be labeled with brief pulses of S-phase imals were given two sequential injections of 3H-TdR markers due to their extremely long cell cycle times (each 20 gCi/g body weight; specific activity was 70-90 (e.g. Cotsarelis et al., 1990; Jones et al., 1995). Ci/mmol) at 2 and I h prior to sacrifice, in order to label In addition, our studies examined the numbers cells in the S phase with high sensitivity. Cryostat sec- and proliferative states of MASH 1 -expressing cells tions ( 12 gm) were taken and processed as in Figure I, under several different conditions: as neurogenesis except that sections were additionally processed for au- winds down in vitro (in normal explant cultures); toradiography as described (Gordon et al., 1995). Data when neurogenesis is sustained for long periods in were obtained from counting the number of cells/mm of explant cultures; and when neurogenesis is tran- OE that were in the basal half of the epithelium and were MASHl +, had incorporated 3H-TdR, or were both siently increased in vivo following bulbectomy. In

MASH 1 + and 3H-TdR+.For each animal, data were col- all of these situations, the results were indicative lected from several millimeters of septa1 OE in each of of a precursor-product relationship between multiple sections, and at least three animals were exam- MASHI' cells and INPs (Gordon et al., 1995). ined for each time point. The data are presented in two This is illustrated in Figure 2 showing the results of ways. The dashed line shows the number of 3H-TdR+ experiments in which unilaterally bulbectomized cells/mm in the OE on the operated (left) side, normal- mice and unoperated control animals were given ized to the number of 3H-TdRf cells/mm on the unop- pulses of 3H-thymidine 2 h prior to sacrifice at a erated (right) side. This provides a measure of overall number of different postbulbectomy time points. proliferation in the OE on the bulbectomized side. The The OE was then processed for MASH 1 immuno- solid line shows, for the bulbectomized side of each ani- reactivity and autoradiography, and the numbers mal, the percent of 3H-TdR+cells that are also MASH 1 + (calculated for each section examined and averaged over of cells in the basal half of the OE that were 3H- the total number of sections). This provides a measure

thymidine', MASH 1 + , and positive for both

of the relative contribution of MASH I + cells to overall markers were counted. The data show an early proliferation. In both cases, the data points represent

burst of proliferation in the MASH 1 + cell popula- mean * S.E.M. The data indicate that, at early times fol- + tion (solid line), which then declines as MASHl lowing bulbectomy, MASH I + cells contribute preferen- cells give rise to MASH 1 - INPs that then amplify tially to the increase in overall proliferation, whereas by

their own numbers (dashed line) prior to giving the time overall proliferation peaks, MASH 1 + cells no rise to neurons. Altogether, the data suggest that longer contribute preferentially. [Note that the value for

Mash1 expression demarcates a neuronal transit the percent of 'H-TdR+ cells that are also MASH 1 + at 5 amplifying cell, which lies immediately upstream days is significantly different from the values at either 0 of the INP, but downstream of the stem cell, in the days (p= 0.006) or 6 days (p= 0.001 ), whereas the values at 0 and 6 days are not significantly different from ORN lineage. Furthermore, our results suggest that each other]. Thus, in response to bulbectomy, increased the crucial role ofMashl in olfactory neurogenesis

proliferation of MASH 1 + cells precedes increased prolif- (Guillemot et al., 1993) is due to expression of eration of cells in the basal half of the OE as a whole (80- MASH 1 by cells within the ORN lineage itself. 90% of which are MASH 1 -). (From Gordon et al., Mol. Mush1 has a very restricted pattern of expres- Cell. Neurosa. 6: 363-379, 0 1995 Academic Press, re- sion, both in the OE and elsewhere in the develop- printed with permission.) Birth and Death ofO(&ory Receptor Neurons 71

ing nervous system. In contrast, expression of Otx2 appears to be much more widespread among neural precursor cells within the anterior regions of the developing nervous system, including the OE, during periods when neurons are actively generated (e.g., E12.5; Simeone et al., 1993). To determine if Otx2 might be a useful marker for ORN precursors, we first performed an analysis of levels of expression of Mashl, Otx2, and Otxl, another mu- nne orthodentick homologue (Simeone et al., 1992 ), using RNA made from nasal turbinates isolated from E14.5-15.5 mouse embryos, the same age and tissue that serves as a source of cells for our OE explant cultures. This analysis, illustrated in Figure 3, shows that Otx2 and Mashl messages are both relatively abundant, while Otxl appears to be the least abundant message of the three tested. In Figure 3 RT-PCR analysis of hlushl, Otxl. 01x2, and situ hybridization for Otx2 mRNA, shown in Fig- actin mRNA levels in E14.5-15.5 OE. Total RNA was ure 4, indicates that Otx2 is expressed by a signifi- isolated from E 14.5- 15.5 nasal turbinates using Ultras- cant proportion of migratory cells (47.1 2.1 % pec RNA isolation mixture (Biotex). First strand cDNA + was synthesized from 10 pg of total RNA in 20 fiL of S.E.M.) in 12-h OE explant cultures. Our previous reaction mixture containing 1X RT buffer, 1 mM studies showed that the migratory ceIls in these cul- dNTPs. 100 pmol random hexamer, and 10 U of AMV tures consist entirely of ORNs and ORN precur- RT (Promega) for I h at 37°C. Desired amounts of sors (Calof and Chikaraishi, 1989; Calof and cDNA mix were added to a PCR mix containing 1 X PCR Lander, 199 1 ). However, essentially no Otx2 ex- buffer, 1 mMMgCI2. 250 pMdNTPs, 0.1 mg/mL BSA, pressing cells have neurites ( 1 out of 327 migratory 0.5 pM each of forward and reverse primers, and 2.5 U cells counted in 10 random fields), suggesting that of Taq DNA polymerase (Gibco-BRL) to a final volume Otx2 is not expressed by differentiated ORNs. Al- of 20 pL. The cycling parameters were denaturation at though confirmation of Otx2 as a precursor-spe- 94°C for I min, annealing at 60°C for 1 min, and elonga- cific marker awaits 3H-thymidine incorporation tion at 72°C for 1 min. Cycle numbers used were 40 cy- analysis and double-labeling experiments using cles for Mushl, 01x1, Otx2; 30 cycles for actin. After reverse transcription with random primers, separate other OE cell type specific markers (e.g., anti- PCR reactions were performed for each set of specific MASH 1 ), both the number and morphology of primers. Specific primers were: Otx2-expressing cells suggest that Otx2 may be a global marker for ORN progenitors, a possibility Mashl: 5'-CTC TTA GCC CAG AGG AAC-3' that we are currently investigating. (forward) The theme that is emerging from these studies is 5'-GGT GAA GGA CAC TTG CAC-3' the complexity of this neuronal lineage: there ap- (reverse) pear to be at least two distinct stages of proliferating Otxl: 5'-CGT ATC TAG CTC TGC TTC-3' neuronal precursor cells interposed between the (forward) 5'-CCT GGC CAT AGG ACA TAG-3' (reverse) Otx2: 5'-CTC TAG TAC CTC AGT CCC A-3' (forward) lanes are, from left, a DNA size marker X 174), PCR 5'-GTC CAG GAA GCT GGT GAT (Cp products from 4, 2, I, 0.5, and 0.25 KLof cDNA input. G-3' (reverse) For actin, lanes are, from left, a DNA size marker, PCR Bactin: 5'-TCA TGA AGT GTG ACG TTG products from 2, 1, 0.5, 0.25, 0.125, and 0.0625 pL ACA TCC-3' (forward) cDNA input. Expected sizes of amplified fragments are: 5'-GTA AAA CGC AGC TCA GTA Otxl, 290 bp; Otx2, 242 bp; Mushl, 454 bp; actin, 290 ACA GTC-3' (reverse) bp. No-RT controls gave no amplification products in A series of different cDNA amounts was used for PCR, the subsequent PCR reactions (not shown). The sizes of and amplified products were separated on 2% agarose the DNA marker fragments (4X 174)are from top: 1353 gels. In the photographs of Otxl, Otx2, and Mush1 gels, bp, 1078 bp, 872 bp, 603 bp, 310 bp, 281 bp, and271 bp. 72 Culolfet ul.

Figure 4 In sitzi hybridization for 01x2 mRNA in OE explant cultures. E14.5 OE explants were cultured in serum-free medium on polylysine- and merosin-coated coverslips as described (DeHamer et al., 1994)for 12 h and in sitzi hybridization with dioxygenin (DIG)-labeled Otx2 (A)antisense or (B) sense probes was performed. DIG-labeled Otx2 antisense and sense probes (-250 bp) were synthesized from pMOTX2 (gift of Dr. A. Simeone through Dr. S. McConnell). Hybridization was performed at 60°C in 1OX Denhardt’s, 50% formamide, 4X SSC, and 250 pg/mL yeast tRNA containing 300 ng/mL ofantisense or sense probe. Unbound probe was removed by wash at 55°C in 50% formamide and 5X SSC, RNase treatment (20 pg/ mL) at 37T, again washed in 50% formamide and 2X SSC, finally at room temperature in 2X SSC. The remaining probe was reacted with anti-DIG antibody and color was developed in NBT/ BClP by following the manufacturer’s recommendation (BMB, DIG nucleic acid detec- tion kit) with modification described by Schaeren-Wiemers and Gerfin-Moser ( 1993). Note that OtxZ positive cells are nonprocess bearing cells. Bar = 50 pm. postmitotic ORN and the stem cell that is ulti- This appears to result from newly generated ORNs mately responsible for the ability of the OE to con- being able to survive for only a short time when the tinually generate neurons. The role of these olfactory bulb is absent (Schwob et al., 1992; Carr multiple developmental stages is not known. How- and Farbman, 1992, 1993). In addition, it was re- ever, studies discussed below suggest that each cel- ported that the rate of generation of new ORNs is lular stage in the ORN lineage serves as a point at permanently elevated following bulbectomy, sug- which extrinsic factors can act to control cell num- gesting that this ongoing ORN death might some- ber and, perhaps, cell fate. how play a role in regulating proliferation of neuronal precursors in the OE (Carr and Farbman, 1992). APOPTOSIS IN THE NEURONAL Until recently, it was not known whether the LINEAGE OF THE OE death that cells undergo in the OE following bulbectomy is apoptotic cell death, or even what cell Previous work from a number of different labora- types are induced to die. Although it had been re- tories demonstrated that when one olfactory bulb ported that DNA fragmentation occurs in the OE is removed from an adult animal (unilateral when the olfactory bulb is removed, this work was bulbectomy), nearly all ORNs in the ipsilateral OE done using agarose gel electrophoresis to evaluate die (e.g., Costanzo and Graziadei, 1983). The OE fragmentation, and therefore could not provide in- then decreases in thickness as cells degenerate. De- formation about the numbers or types of cells in- spite the fact that cells in the basal compartment of duced to die (Michel et al., 1994). To approach the OE then proliferate and many new ORNs are these questions, we performed an analysis of cell generated ( Schwartz-Levey et al., 199 1; Gordon et death in the OE under three conditions: in normal al., 1995), in the absence of its synaptic target tis- adult mice; in adult mice subjected to unilateral 01- sue the OE never reaches its original thickness. factory bulbectomy; and in primary cell cultures Birth and Death of Olfactory Receptor Neurons 73 derived from embryonic mouse OE. To quantify sis to identify the types of cells undergoing apop- apoptotic cell death, we used the TUNEL tech- totic death, we found that cells at all stages in the nique ( DNA end-labeling with deoxynucleotide OE neuronal lineage (proliferating neuronal pre- terminal transferase and dUTP-biotin) to test for cursor cells, immature ORNs, and mature ORNs) DNA fragmentation in cells in vivo and in vitro undergo apoptosis in the acutely bulbectomized (Arends and Wyllie, 1991; Gavrieli et al., 1992; animal (24 h following bulbectomy ). Bulbectomy Deckwerth and Johnson, 1993). We combined this does not induce apoptosis of two other cell types technique with the use of cell-type specific anti- whose role, if any, in the ORN lineage is uncertain; body markers and 'H-thymidine incorporation to keratin-expressing horizontal basal cells, and sup- determine which cells in the ORN lineage die, and porting or sustentacular cells (Holcomb et al., with what time course, when they are deprived of 1995). In the chronically bulbectomized OE, the synaptic contact with the olfactory bulb. increase in apoptosis that is observed is accounted In Figure 5, an example of an in vivo experiment for entirely by an increase in the death of mature is shown. In this experiment, TUNEL labeling was ORNs, suggesting that the factors that mediate the combined with anti-NCAM (neural cell adhesion survival of mature ORNs may differ from those molecule) immunohistochemistry (Calof and Chi- mediating the survival of immature ORNs and karaishi, 1989) so that the number of NCAM+ neuronal precursors (Holcomb et al., 1995). Thus, ORNs induced to die 24 h following bulbectomy apoptosis appears to regulate neuronal number in could be quantified. Figure 5 ( A-C) shows septal the OE at multiple stages in the neuronal lineage, OE on the side ipsilateral to the surgery; (D,E) an observation that has been made for other neu- show the septal OE immediately opposite, on the ronal lineages as well (e.g., Birren and Anderson, contralateral, unoperated side. In (A), the TUNEL 1993; DiCicco-Bloom et al., 1993; Verdi and An- staining (white dots; e.g., arrow) shows the nuclei derson, 1994). of cells with fragmented DNA; many cells are Our studies in vitro confirmed that embryonic clearly undergoing DNA fragmentation, and there- ORNs and their precursors also undergo cell death fore apoptotic cell death, by 24 h postsurgery. The when explanted into culture. Like the cell death double labeling with anti-NCAM in Figure 5 (B) that occurs following ablation of the olfactory bulb shows that the great majority of apoptotic cells are in vivo, pharmacological experiments and TUNEL ORNs. As shown in Figure 5 (D),there was virtu- staining demonstrate that olfactory neuronal cell ally no TUNEL staining in the contralateral, unop- death in vitro has the characteristics of apoptosis. erated OE. It is also apparent from these pho- Interestingly, the onset of apoptosis shows a similar tographs that a high level of TUNEL staining is time course in vivo, following bulbectomy, and in evident before overt signs of morphological vitro, following explanation of dissociated olfac- degeneration are observed in the OE [compare Fig. tory neuronal cells into culture (Fig. 7). 5 (C)with (F)], indicating that the onset of exten- To begin to identify factors that might mediate sive cell death in the OE occurs several days earlier olfactory neuronal cell survival in vivo, we used in than estimates previously indicated, based on mea- vitro assays to test agents that prevent apoptosis in surements of OE thickness alone. other cells, including aurintricarboxylic acid We assessed the time course and extent of ( ATA), a membrane-permeant ana!og of cyclic apoptosis in the OE following bulbectomy in ani- AMP (CPT-CAMP),and members of the neuro- mals sacrificed from I2 h to 84 days after surgery. trophin family of polypeptide growth factors Data from these experiments are summarized in (Holcomb et al., 1995). [Neurotrophins are Figure 6. In OE on the bulbectomized side, the known to be expressed in the olfactory bulb, mak- number of TUNEL' cells increases sharply by 12 ing them good candidates for potential target-de- h postsurgery and is maximal at 2 days. The num- rived trophic factors in this system (e.g., Large et ber of TUNEL' cells then declines rapidly to near- al., 1986; Maisonpierre et al., 1990; Guthrie and normal levels but still remains elevated over that Gall, 199 1 ) .] ATA and CPT-CAMP are each able seen in the contralateral OE at all time points to promote survival of a fraction ofcultured ORNs, tested. The mean thickness of OE on the operated as are three neurotrophins-brain derived neuro- side is also shown (cf. Costanzo and Graziadei, trophic factor, neurotrophin-3, and neurotrophin- 1983; Schwartz-Levey et al., 1991). When we 5, but not nerve growth factor (Holcomb et al., combined TUNEL staining with immunohisto- 1995 ). We used immunohistochemical methods to chemistry and 'H-thymidine incorporation analy- determine if the neurotrophin tyrosine kinase re- 74 Calof et al.

Figure 5 DNA fragmentation in the olfactory epithelium following unilateral olfactory bulbectomy. Adult male mice were anesthetized and a small suction tube was used to selectively remove the left olfactory bulb without causing injury to the contralateral olfactory bulb or to the brain. At time points ranging from 12 h to 84 days following surgery, animals were sacrificed and the region ofthe nose containing the OE was dissected and fixed by freeze substitution as described (Holcomb et al., 1995). After decalcification for 7 days in -390 mMEDTA, pH 7.1, the tissue was sectioned in the horizontal plane in 12-pm sections with a cryostat. Sections of OE were stained for DNA fragmentation using deoxynucleotide terminal transferase end labeling of DNA fragments with biotinylated dUTP and a fluorescent avidin, a modification of the TUNEL technique of Gavrieli et al. ( 1992). (A-C) Photos show the bulbectomized OE and (D-F) contralateral OE immediately opposite from an animal sacrificed at 24 h postbulbectomy. (A,D) Fluorescein optics showing TUNEL staining; arrow in (A) indicates TUNEL+ cell. (B,E) Rhodamine optics showing NCAM immunoreactivity in the same sections. (C,F) Nomarski optics. Bar = 50 pm. (From Calof et al., Growth Factors as Drugs for Nezirological and Sensory Disorders, 0 1995 The Ciba Foundation, reprinted with permission.) Birth and Death ofO1factor.v Receptor Neurons 75

\ complete ORN survival (see also Mahanthappa * Peak of apoptosis and Schwarting, 1993). In light of our finding that OE neuronal precursors undergo apoptosis, we recently began to analyze cell death in the OE of Mashl-I- embryos. We found that few ORNs ever appear in Mushl OE, indicating that the lack of ORNs at birth reflects a defect in neuronal production (as opposed to neuronal survival). Interestingly, we observe an elevated level of cell death in the NCAMP cells in the E14-15

Peak of 'H-incorporation Mushl -'-OE (compared to wild-type and heterozy- II I by precursors I gote controls) as revealed by TUNEL staining (Calof et al., 1995; J.S. and A.L.C., data not shown). Be- 0 0246 56 84 cause the majority of the NCAM- cells in the epithelium at this age (E14- 15 ) are neuronal precursors Days after removal of olfactory bulb Figure 6 Time course of DNA fragmentation following unilateral bulbectomy. Cryostat sections of OE from un- A. In Vitro operated (control) and bulbectomized mice (sacrificed at postoperative time points indicated) were processed for TUNEL as described (Holcomb et al., 1995). TUNEL' cells were counted in sections of septa1 OE on the bulbectomized side (open circles), contralateral side (open triangles), and from unoperated animals (time = 0). Mean numbers of TUNEL+ cells/mm OE (f S.E.M.) are plotted, together with changes in the average thickness of the bulbectomized OE (solid circles), over time following bulbectomy. Where error bars are not seen, the error was small enough to be obscured by the symbol representing the data point. Differences be- Hours in culture tween bulbectomized and contralateral OE were statisti- cally significant for all times I 2 and 2 6 days (pI 0.02 B. In Vivo except at 56 days, where p = 0.055; Student's t test: Glantz, 1992). (From Holcomb et al., Dev. Biol. 172: g 200 307-323, 0 1995 Academic Press, reprinted with per- E mission.) E 150 -.v) +$ 100 -I ceptors trkB and trkC are expressed in OE, and, W 5 50 consistent with the effects we see on ORN survival I- we found that these two receptors are ex- in vitro, 0 pressed by fractions of ORNs scattered throughout neonatal OE (Holcomb et al., 1995; cf. Deckner et Hours post-surgery al., 1993). However, we do not think it likely that Figure 7 DNA fragmentation in target-deprived olfac- neurotrophins are the only growth factors that me- tory neuronal cells occurs over similar time courses in diate survival of ORNs: the survival-promoting vitro and in viva (A) Dissociated olfactory neuronal cells effects of individual neurotrophins in vitro are from E16.5-17.5 CD-I embryos were plated at a density never as large as those seen with ATA or CPT- of -3 X lo3 cells/well in 96-well tissue culture plates CAMP; and when we have tested combinations of as described (Holcomb et al., 1995). Cells were fixed at indicated times and stained for TUNEL. For the t = 0 different neurotrophins, we did not observe an in- time point, cells were incubated for 30 min at 37°C prior crease in cell survival (Holcomb et al., 1995; to fixation. The percentage of total cells per well that are J.D.H. and A.L.C., unpub. observ.). The most TUNEL' is plotted for each time point. Data show mean plausible explanation for this is that other factors, f S.E.M. of triplicate wells. (B) Data regraphed from in addition to neurotrophins, are necessary for Figure 6. 76 Cuhf et ul

(Calof and Chikaraishi, 1989), we hypothesize that, that FGF supports the proliferation and/ or sur- in Mushl animals, neuronal precursor cells are vival of this cell (DeHamer et al., 1994 ). Our esti- produced, but most undergo apoptosis without gen- mates of the abundance of this cell put it at a fre- erating ORNs. A role for Mushl in regulating expres- quency of - 1 /2500 INPs, suggesting that these cells sion of genes that mediate survival of neuronal pre- may lie far “upstream” of INPs in the ORN lineage cursor cells in the ORN lineage is a possibility that we (J.S.M., J.S., and A.L.C., unpub. observ.). Our cur- are currently investigating. rent hypothesis is that this cell may be the neuronal stem cell that is ultimately responsible for the ability of the OE to continually generate neurons. Stimulation of ORN production in OE cultures REGULATION OF NEUROGENESIS was reported by two other groups. Coculture of IN THE OE neonatal rat OE with astrocytes was shown to pro- long the generation of ORNs (Pixley, 1992), an Neurogenesis ceases after about 24 h in OE ex- effect that may be attributable to production of plants cultured in serum-free, defined medium in FGFs by these cells (Woodward et al., 1992; Baird, the absence of exogenous growth factors (Calof and 1994). Mahanthappa and Schwarting ( 1993) re- Chikaraishi, 1989). In contrast. neurogenesis in ported that TGF-P2 stimulates neurogenesis in vivo occurs continually throughout the lifetime of postnatal rat OE cultures, an effect that we were not the organism. This abrupt termination of neuro- able to reproduce in our assay system (DeHamer genesis in vitro was used by our laboratory as the et al., 1994). However, the study by Mahanthappa basis for a screen to identify polypeptide growth and Schwarting ( 1993) left open the possibility factors capable of promoting prolonged neurogen- that the role of TGF-P2 is to serve as a survival fac- esis in OE explant cultures. Our findings indicate tor for newly generated ORNs, rather than as a pro- that members of the fibroblast growth factor (FGF) liferation factor for neuronal precursors. In addi- family are able to promote prolonged neurogen- tion, differences in the age of tissue assayed, and/ esis, and further that FGFs act in two ways or complications arising from endogenous produc- (DeHamer et al., 1994).First, FGFs act to increase tion of TGF-P2 by OE cells differentially present in the likelihood that INPs divide twice, rather than the two culture systems (cf. Millan et al., 199 1 ), once, before generating ORNs. This action re- may have contributed to the different results ob- quires exposure of INPs to FGFs by early G, of tained in the two studies. their cell cycle, the phase at which their commit- Recently, we developed an approach to obtain ment to terminal differentiation would be expected more direct evidence for the existence of early pro- to occur (Soprano and Cosenza, 1992). This is genitors (potential stem cells) in the OE and to similar to the way in which FGFs are believed to learn more about growth conditions for these cells. act in myogenesis, where FGFs act to promote my- To isolate early progenitors, a neuronal cell frac- oblast proliferation by repressing terminal differ- tion containing purified INPs and ORNs is sepa- entiation in G,, thereby allowing cells to progress rated from the basal cells in E 14- l 5 OE suspension through additional cell cycles (Clegg et al., 1987). cultures on the basis of differential cell adhesion Further, observation of the action of FGFs in pro- (Calof and Lander, 1991; Calof et al., 1996); then moting divisions of INPs while leaving their neuro- dishes coated with antibodies to NCAM (a specific nal fate unaffected (INPs quantitatively give rise to marker for differentiated ORNs) are used to re- ORNs even when cultured in FGFs; neuronal move ORNs (Hagiwara et al., 1995; J. S. Mumm, differentiation is simply delayed by the additional J. Shou, and A. L. Calof, submitted). When the re- cell cycles) is responsible for our notion of INPs as sulting purified INP fraction is cultured, most cells neuronal transit amplifying cells ( DeHamer et al., quickly give rise to neurons. However, if the cells 1994). are plated on top of stromal fibroblasts from the The second action of FGFs is to cause a distinct embryonic OE, small numbers of colonies of un- subpopulation of OE explants to continually gen- differentiated cells are visible after 7 days in culture erate large numbers of neurons for at least several [Fig. 8 (A)]. [Typically, the INP fraction is isolated days. Only 5-870 of explants continue to generate from OE purified from the Rosa 26 transgenic large numbers of ORNs at late times in culture, mouse strain (Friedrich and Soriano, 199 1 ), which suggesting that an early, rare progenitor, possibly a expresses lucZ in all cells, while the stromal cells neuronal stem cell, is present in these explants, and are isolated from wild-type mice; colonies are then Birth and Death ofO(factory Receptor Neurons 77

visualized using P-galactosidase histochemistry ( Xgal staining) .] When these colonies are dissociated from their stromal feeder layers, and replated under conditions that promote neuronal differentiation, cells that are morphologically identical to ORNs are formed [Fig. 8 (B)]. Staining with antibodies to NCAM demonstrates that these P-galactosidase-expressingcells are in fact ORNs, and 3H- thymidine incorporation analysis indicates that the precursors of these ORNs are still dividing after 7 days in coculture with stromal fibroblasts (J. S. Mumm, J. Shou, and A. L. Calof, submitted). These results suggest very early progenitors in the ORN lineage can be isolated and that factors produced by stromal cells (possibly including FGFs; cf. Mason et al., 1994) can maintain these cells and their capacity to generate neurons for at least 7 days in culture.

ARE NEURONAL BIRTH AND DEATH COREGULATED IN THE OE?

Proliferation of neuronal precursors in the OE (detected as 3H-TdR incorporation by cells in the basal compartment of the epithelium) increases Figure 8 Purified ORN progenitors form colonies and following bulbectomy, reaching a peak 5-6 days continue to generate neurons for 7 days in culture. (A) postsurgery in the mouse (Fig. 6; cf Schwartz-Levey NCAM progenitor cells were purified from the dissoci- et al., 199 1 ) . Cell loss in the OE (measured as epithe- ated neuronal cell fraction of E14.5-15.5 Rosa26 lial thickness) follows a similar time course: epithelial transgenic mice, as described in the text. After coculture thickness reaches its minimum at about 5 days post- over a monolayer of mitomycin-treated ( 10 Fg/rnL for 2 bulbectomy, then increases again to approximately h) stromal fibroblasts for 7 days, cultures were fixed ( 15 70% of its original value (Fig. 6; Costanzo and Graz- min in 0.5% glutaraldehyde, 2 mM MgCI2, 5% sucrose iadei, 1983; Schwartz-Levey et al., 199 1 ). This tem- in PBS, pH 7.5), permeablized with Triton, 0.01% 0.1% poral correlation between neuronal cell loss in the OE deoxycholic acid, and 2 mM MgCI2 in PBS (pH 7.5) for 1 h, and then placed in X-Gal staining solution [ 0.1 % and 3H-TdR incorporation by ORN precursors sug- Triton, 0.01% deoxycholic acid, 2 mM MgCI2, 5 mM gests a possible causal relationship between the two K,Fe(CN),, and 5 mM &Fe(CN), in PBS (pH 7.5), events: differentiated ORNs might somehow provide plus 320 mg/mL X-Gal(5-bromo-4-chloro-3-indolyl-~- a signal that feeds back to inhibit proliferation oftheir D-galactopyranoside] for 24 h at 37°C. Colonies of cells own precursors. Such a regulatory mechanism was arise at the rate ofapproximately 1 /800 panned, purified suggested for larval frog retina as well (Reh and progenitor cells initially plated. The majority of these Tully, 1986). According to this view, maximum pro- colonies contain cells with morphological characteristics liferation would be associated with maximum cell of undifferentiated progenitor cells: they are small, loss (at 5-6 days postbulbectomy ), and the elevated round, and do not bear neurites. (B) Progenitor cell col- level of proliferation that is maintained in the chron- onies were first cuitured for 7 days on stromal fibroblasts, then dissociated from feeder cell layers and replated over E14.5 CD-I derived OE explants that had been precul- tured for 24 h. Cells with morphological characteristics of ORNs can be seen by X-Gal staining. The black arrow shows a small clump of 3 lucZ-positive cells, indicating from the CD- 1 explants, for morphological comparison. that these cells were derived from replated Rosa26 pro- Bar = 50 pm. Micrograph in (B)is approximately 4X the genitors. White arrows indicate the lacZ-negative ORNs magnification of micrograph shown in (A). 78 Calof et al.

Stromal cells,

Feedback Inhibrrion' FGFs , other factors? (survival and/or proliferation) MASH 1+ 1 Precursorsa (OTX2+?) u Amplifrca tron 7

INPs (OTX2+? ) Feedback Inhibition7 Amplification FGFs (proliferation)

//I I

Neuro trophins, other factors?

Ne uro trophins, other factors?

Figure 9 Cell interactions regulating neurogenesis and cell death in the olfactory epithelium Evidence in support of this model is discussed in detail in the text.

ically bulbectomized OE would be associated with proliferation at 5-6 days postbulbectomy that is ob- the reduced number of ORNs that are maintained in served could simply reflect expansion of transit am- the OE in this condition. plifying cells (MASH1' cells and/or INPs) in re- The results of our studies on apoptosis in the OE sponse to an early mitogenic stimulus. suggest the possibility of an additional relationship This in turn suggests the interesting possibility that between neuronal death and precursor proliferation actively dying ORNs may provide a positive signal in the OE, however. The peak of apoptosis of OE regulating bulbectomy-induced neurogenesis in the neuronal cells occurs 2 days postbulbectomy (Fig. OE. In some cells, including neurons, an early step in 6), preceding both maximum cell loss and the peak the apoptotic pathway is the expression of cysteine of induced 3H-TdR incorporation by 3-4 days. We proteases similar to interleukin- 1P-converting en- know that the genesis of ORNs proceeds through at zyme (ICE), an enzyme which cleaves the inactive least three precursor cell stages (the stem cell, the IL-10 precursor protein to generate the active cyto-

MASH 1 + cell, and the INP; DeHamer et al., 1994; kine (Yuan et al., 1993; Miura et al., 1993; Kumar et Gordon et al., 1995). If the signal for induced neuro- al., 1994; Gagliardini et al., 1994). If ORNs express genesis following bulbectomy acts on an early pro- ICE-like enzymes while undergoing apoptosis, such genitor cell, perhaps the stem cell, then the peak in enzymes could potentially activate factors that might Birth and Death of OlJactory Receptor Neurons 79 then act as positive regulators of proliferation by early BAIRD,A. ( 1994). Fibroblast growth factors: activities progenitors in the ORN lineage. and significance of non-neurotrophin neurotrophic growth factors. Curr. Opin.Neurobiol. 4:78-86. CONCLUSIONS BIRREN,S. J., LO, L.-C., and ANDERSON,D. J. (1993). Sympathetic neuroblasts undergo a developmental A summary of some of our current ideas concern- switch in trophic dependence. Development 119:597- ing the regulation of neurogenesis and cell death in 610. the OE is provided in Figure 9. This model pro- CAGGIANO,M., KAUER,J. S., and HUNTER,D. D. ( 1994). poses three stages of proliferating neuronal precur- Globose basal cells are neuronal progenitors in the olfactory epithelium: a lineage analysis using a replication- sors in the ORN lineage: a self-renewing stem cell incompetent retrovirus. Neuron 13:339-352. (for which no molecular marker yet exists); and CALOF,A. L., ADUSUMALLI,M. D., DEHAMER,M. K., two stages of neuronal transit amplifying cells, GUEVARA,J. G., MUMM,J. M., WHITEHEAD,S. J., MASH 1-expressing cells, and their progeny, the and LANDER,A. D. ( 1994a). Generation, differenti- INPs (Gordon et al., I995 ). Both types of amplify- ation, and maturation of olfactory receptor neurons ing precursors may express Otx2 INPs giving rise to in vitro. In: Olfaction and Taste XI K. Kurihara, N. immature ORNs, which are postmitotic cells that Suzuki, and H. Ogawa, Eds. Springer-Verlag, Tokyo, express NCAM and GAP43 (Calof and Chikara- pp. 36-40. ishi, 1989; Verhaagen et al., 1990). Immature CALOF,A. L., CAMPANERO,M. R., OREAR,J. J., YUR- ORNs undergo further differentiation to become CHENCO,P. D., and LANDER,A. D. ( 1994b). Do- mature ORNs, which express both NCAM and ol- main-specific activation of neuronal migration and factory marker protein (OMP; Margolis, 1980) but neurite outgrowth-promoting activities of laminin. Neuron 13:117-130. not GAP43 (Verhaagen et al., 1990); both imma- CALOF,A. L. and CHIKARAISHI,D. M. ( 1989).Analysis ture and mature ORNs appear to be at least par- of neurogenesis in a mammalian neuroepithelium: tially dependent on neurotrophins for their sur- proliferation and differentiation of an olfactory neu- vival (Holcomb et al., 1995). Our initial culture ron precursor in vitro. Neuron 3:115-127. studies on isolated OE progenitor cells suggest that CALOF,A. L., GUEVARA,J. G., and GORDON,M. K. the neuronal stem cell of the OE may be dependent ( 1996). Mouse olfactory epithelium. In: Primary Neu- upon stroma-derived factors, which may include ronal Cell Culture:A Practical Approach. J. Cohen and FGFs (DeHamer et al., 1994; J. S. Mumm, J. Shou, G. Wilken, Eds. Oxford, IRL Press. and A. L. Calof, submitted). At a later stage in the CALOF,A. L., HOLCOMB,J. D., MUMM,J. S., HAGIWARA, lineage, FGFs appear to regulate the amplification N., TRAN,P., SMITH,K. M., and SHELTON,D. L. divisions of INPs (DeHamer et al., 1994). Addi- (1995). Factors affecting neuronal birth and death in tional factors acting on the different stages of this mammalian olfactory epithelium. In: Growth Factors as lineage have yet to be identified, although feedback Drugs fir Neurological and Sensov)i Disorders. Wiley, Chichester (Ciba Foundation Symposium 196). inhibition by ORNs is one of the proposed mecha- CALOF,A. L. and LANDER,A. D. ( 199 I ). Relationship nisms for regulating the division of precursors. between neuronal migration and cell-substratum ad- The authors thank Susan McConnell for Otxl and hesion: laminin and merosin promote olfactory neu- Otx2 probes, and Francois Guillemot for Mash l+’- ronal migration but are anti-adhesive. J. Cell Biol. transgenic mice. We are grateful to Mary Burkman, Me- 115:779-794. linda Gordon, Kirk Smith, and Peter Rim for help with CAMARA,c. G. andHARDING,J. W. ( 1984). Thymidine various experiments, and to Arthur Lander for many incorporation in the olfactory epithelium of mice: helpful suggestions. This work was supported by grants early exponential response induced by olfactory neu- to A. L. Calof from the Institute on Deafness and Other rectomy. Brain Rex 308:63-68. Communication Disorders of the NIH (DC 02 180), the CARR,V. M. and FARBMAN,A. I. ( 1992). Ablation of National Institute of Neurological Disorders and Stroke the olfactory bulb up-regulates the rate of neurogenesis of the NIH (NS32174), the Council for Tobacco Re- and induces precocious cell death in olfactory epithe- search (3663), and a March of Dimes/Basil O’Connor lium. Exp. Neurol. 11555-59. Scholar Award. CARR,V. M. and FARBMAN,A. I. ( 1993). The dynamics of cell death in the olfactory epithelium. Exp. Neurol. REFERENCES 124~308-314. CHUAH,M. I., DAVID,S., and BLASCHUK,0. (1991 ). ARENDS,M. J. and WYLLIE,A. H. (1991). Apoptosis: Differentiation and survival of rat olfactory epithelial mechanisms and roles in pathology. Int. Rev. Exp. Pa- neurons in dissociated cell culture. Brain Rex 60: 123- thol. 136:593-608. 132. 80 Cahfet al.

CHUAH,M. I., FARBMAN,A. I., and MENCO,B. P. ( 1985). Cell and explant culture of olfactory chemo- ( 1985). Influence of olfactory bulb on dendntic knob receptor cells. J. Neurosci. Methods 14:77-90. density of rat olfactory receptor neurons in vitro. GORDON,M. K., MUMM,J. S., DAVIS,R. A., HOLCOMB, Brain Res. 338:259-266. J. D., and CALOF,A. L. ( 1995). Dynamics of MASH- CLEGG,C. H., LINKHART,T. A., OLWIN,B. B., and 1 expression in vitro and in vivo suggest a non-stem HAUSCHKA,S. D. ( 1987). Growth factor control of cell site of MASH-I action in the olfactory receptor skeletal muscle differentiation: commitment to termi- neuron lineage. Mol. Cell. Neurosci. 6:363-379. nal differentiation occurs in GIphase and is repressed GRAZIADEI,P. P. C. and MONTI-GRAZIADEI,G. A. by fibroblast growth factor. J. Cell Biol. 105:949-956. ( 1978). Continuous nerve cell renewal in the olfactory COON, H. G., CURCIO,F., SAKAGUCHI,K., BRANDI, system. In: Handbook of Sensory Physiology, Volume M. L., and SWERDLOW,R. D. (1989). Cell cultures /X: Developmmt of Sensory Systems. M. Jacobson, of neuroblasts from rat olfactory epithelium that show Ed., Springer-Verlag, New York, pp. 55-83. odrant responses. Proc. Natl. Acad. Sci. USA 86: 1703- GRAZIADEI,P. P. C. and MONTI-GRAZIADEI,G. A. 1707. ( 1979). Neurogenesis and neuron regeneration in the COSTANZO,R. M. and GRAZIADEI,P. P. C. (1983). A olfactory system of mammals. I. Morphological as- quantitative analysis of changes in the olfactory epi- pects of differentiation and structural organization of thelium following bulbectomy in hamster. J. Comp. the olfactory sensory neurons. J. Neurocytol. 8: 1- 18. Neurol 215:370-38 1. GUILLEMOT,F., Lo, C.-C., JOHNSON,J. E., AUERBACH, COTSARELIS,G., SUN, T.-T., and LAVKER,R. M. A., ANDERSON,D. J., and JOYNER,A. L. ( 1993). ( 1990). Label-retaining cells reside in the bulge are0 Mammalian achaete-scute homolog 1 is required for of pilosebaceous unit: implications for follicular stem the early development of olfactory and autonomic cells, hair cycle, and skin carcinogenesis. Cell 61: neurons. Cell 15:463-476. 1329-1337. GUTHRIE,K. M. and GALL,C. M. ( 199 I ). Differential DECKNER,M.-J., FRISEN,J., VERGE,V. M. K., HOK- expression of mRNAs for the NGF family of neuro- FELT, T., and RISLING,M. (1993). Localization of trophic factors in the adult rat central olfactory system. neurotrophin receptors in olfactory epithelium and J. Comp. Neurol. 313:95-102. bulb. NrriioReport 5:30 1-304. HAGIWARA,N., MUMM,J. S., and CALOF,A. L. (1995). Characterization and immortalization of early olfactory DECKWERTH,T. J. and JOHNSON,E. M., Jr. (1993). receptor neuron progenitor cells isolated from mouse ol- Temporal analysis of events associated with pro- factory epithelium. Soc. Neurosci. Abstr. 21:287. grammed cell death (apoptosis) of sympathetic neu- HALL,P. A. and WATT, F. M. (1989). Stem cells: the rons deprived of nerve growth factor. J. Cell Biol. 123: generation and maintenance of cellular diversity. De- 1207-1222. velopment 106:6 19-633. DEHAMER,M. K., GUEVARA,J. L., HANNON,K., OL- HOLCOMB,J. D., MUMM,J. S., and CALOF, A. L. WIN, B. B., and CALOF,A. L. ( 1994). Genesis of olfac- ( 1995). Apoptosis in the neuronal lineage of the tory receptor neurons in vitro: regulation of progenitor mouse olfactory epithelium: regulation in vivo and in cell divisions by fibroblast growth factors. Neuron 13: vitro. Dev. Biol. 112:307-323. 1083- 1097. JOHNSON,J. E., BIRREN,S. J., and ANDERSON,D. J. DICICCO-BLOOM,E., FRIEDMAN,W. J., and BLACK, ( 1990). Two rat homologues of Drosophila achaete- I. B. ( 1993). NT-3 stimulates sympathetic neuroblast scute specifically expressed in neuronal precursors. proliferation by promoting precursor survival. Neuron Nature 346:858-86 1. 11:1101-1111. JONES,P. H., HARPER.S., and WATT, F. M. (1995). FRIEDRICH,G. and SORIANO,P. ( 199 1 ). Promoter traps Stem cell patterning and fate in human epidermis. Cell in embryonic stem cells: a genetic screen to identify 80~83-93. and mutate developmental genes in mice. Genes Dev. KUMAR,S., KINOSHITA,M., NODA, M.. COPELAND, 51513-1 523. N. G., and JENKINS,N. A. (1994). Induction of GAGLIARDINI,V., FERNANDEZ,P.-A., LEE, R. K. K., apoptosis by the mouse Nedd2 gene, which encodes a DREXLER,H. C. A., ROTELLO,R. J., FISHMAN,M. C., protein similar to the product of the Cnrnorhabditis and YAN,J. ( 1994). Prevention of vertebrate neuro- elegans cell death gene ced-3 and the mammalian IL- nal death by the crmA gene. Science 263:826-828. I -b-converting enzyme. Genes Dev. 8: 16 13- 1626. GAVRIELI,Y., SHERMAN,Y., and BEN-SASSON,S. A. LARGE,T. H., BODARY,S. C., CLEGG,D. O., WESKAMP, ( 1992). Identification of programmed cell death in G., OTTEN,U., and REICHARDT,L. F. (1986). Nerve situ via specific labeling of nuclear DNA fragmenta- growth factor gene expression in the developing rat tion. J. Cell Biol. 119:493-501. brain. Science 234: 35 2-3 5 5. GLANTZ,S. A. ( 1992). Primer ofBiostatistics, 3rd ed. Lo, L.-C., JOHNSON,J. E., WUENSCHELL,C. W., SAITO, McGraw-Hill, New York. T., and ANDERSON,D. J. ( 1991 ). Mammalian GONZALES,F., FARBMAN,A. I., and GESTELAND,R. C. achaete-scute homolog 1 is transiently expressed by Birth und Deulh of Olfactory Receptor Nc.ur0n.c 81

spatially restricted subsets of early neuroepithelial and sine hydroxylase-containing amacrine cell number in neural crest cells. Genes Dev. 5: 1524- 1537. larval frog retina. Dev. Biol. 114:463-469. MACKAY-SIM, A. and KITTEL,P. ( 199 1 ). Cell dynamics RONNETT,G. V., HESTER,L. D., and SNYDER,S. H. in the adult mouse olfactory epithelium: a quantitative ( 199 1 ). Primary culture of neonatal rat olfactory neu- autoradiographic study. J. Neurosci. 11:979-984. rons. J. Neurosci. 11:1243-1255. MAHANTHAPPA,N. K. and SCHWARTING,G. A. ( 1993). SCHAEREN-WIEMERS,N. and GERFIN-MOSER,A. Peptide growth factor control of olfactory neurogen- ( 1993). A single protocol to detect transcripts of van- esis and neuron survival in vitro: roles of EGF and ous types and expression levels in neural tissue and TGF-0s. Neuron 10:293-305. cultured cells: in situ hybridization using dioxigenin- MAISONPIERRE,P. C., BELLUSCIO,L., FRIEDMAN,B., labelled cRNA probes. Histochemistry 100:43 1-440. ALDERSON,R. F., WIEGAND,S. J., FURTH,M. E., SCHUBERT,D., STALLCUP,W., LACORBIERE,M., KIDO- LINDSAY,R. M., and YANCOPOULOS,G. D. ( 1990). KORO, Y., and ORGEL, L. ( 1985). Ontogeny of elec- NT-3, BDNF, and NGF in the developing rat nervous trically excitable cells in cultured olfactory epithelium. system: parallel as well as reciprocal patterns of expres- Proc. Nutl. Acad. Sci. USA 82:7782-7786. sion. Neuron 5:50 1-509. SCHWARTZ-LEVEY,M., CHIKARAISHI,D. M., and MARGOLIS,F. L. ( 1980). A marker protein for the olfac- KAUER,J. S. ( 1991). Characterization of potential tory chemoreceptor neuron. In: Proteins of the Ner- precursor populations using immunocytochemistry vous System. R. A. Bradshaw and D. Schneider, Eds. and autoradiography. J. Neurosci. 11:3556-3564. Raven, New York, pp. 59-84. SCHWOB,J. E., MIELESZKOSZUMOWSKI, K. E., and MASON,I. J., FULLER-PACE,F., SMITH, R., and DICK- STASKY,A. A. ( 1992). Olfactory sensory neurons are SON, C. ( 1994). FGF-7 (keratinocyte growth factor) trophically dependent on the olfactory bulb for their expression during mouse development suggests roles prolonged survival, J. Neurosci. 12:3896-39 19. in myogenesis, forebrain regionalisation and epithe- SIMEONE,A., ACAMPORA,D., GULISANO,M., STORNAI- lial-mesenchymal interactions. Mech. Dev. 45: 15-30. UOLO,A., and BONCINELLI,E. ( 1992). Nested expres- MICHEL, D., MOYSE,E., BRUN,G., and JOURDAN,F. sion domains of four homeobox genes in developing ( 1994). Induction of apoptosis in the rat olfactory rostra1 brain. Nature 358:687-690. neuroepithelium by synaptic target ablation. Neuro- SIMEONE, A., ACAMPORA,D., MALLAMACI, A,, STOR- Report 5: 1329-1 332. NAIUOLO, A,, D’APICE,M. R., NIGRO,V., and BONCI- MILLAN,F. A., DENHEZ,F., KONDAIAH,P., and AK- NELLI, E. ( 1993). A vertebrate gene related to ortho- HuRST, R. ( 199 1 ). Embryonic gene expression pat- denticle contains a homeodomain of the bicoid class terns of TCFP 1, 02 and 03 suggest different develop- and demarcates anterior neuroectoderm in the gastru- mental functions in vivo. Development 111: I3 1- 144. lating mouse embryo. EMBO J.12:2735-2747. MIURA,M., ZHU,H., ROTELLO,R., HARTWEIG,E. A., SOPRANO,K. J. and COSENZA,S. C. ( 1992). Cell cycle and YUAN,J. (1993). Induction of apoptosis in fi- and cell-growth control. In: Molecular und Cellular broblasts by IL- 1b-converting enzyme, a mammalian Approaches to the Control of Proliferation and Differ- homolog of the C. elegans cell death gene ced-3. Cell entiation. G. s. Stein and J. B. Lian, Eds. Academic 75~653-660. Press, Inc., San Diego, pp. 73-1 14. MONTI-GRAZIADEI, G. A. and GRAZIADEI, P. P. C. VERDI,J. M. and ANDERSON,D. J. ( 1994).Neurotroph- ( 1979). Neurogenesis and neuron regeneration in the ins regulate sequential changes in neurotrophin recep- olfactory system of mammals. 11. Degeneration and re- tor expression by sympathetic neuroblasts. Neuron 13: constitution of the olfactory sensory neurons after ax- 1359-1372. otomy. J. Neurocytol. 8:197-2 13. VERHAAGEN,J., OESTREICHER, A. B., GRILLO, M., NOBLE,M., MALLABURN,P. S., and KLEIN,N. (1984). KHEW-GOODALL, Y.-S., GISPEN,W. H., and MAR- The growth of olfactory neurons in short-term cultures GOLIS,F. L. ( 1990). Neuroplasticity in the olfactory of rat olfactory epithelium. Neurosci. Lett. 45193-1 98. system: differential effects of central and peripheral le- PIXLEY,S. K. ( 1992). CNS glial cells support in vitro sions of the primary olfactory pathway on the expres- survival, division, and differentiation of dissociated 01- sion of B-501GAP43 and the olfactory marker pro- factory neuronal progenitor cells. Neuron 8: I I9 1- tein. J. Neurosci. Res. 26:3 1-44. 1204. WOODWARD,W. R., NISHI,R., MESHUL,C. K., WIL- PIXLEY,S. K. and PUN,R. Y. (1990). Cultured rat ol- LIAMS, T. E., COULOMBE, M., and ECKENSTEIN, F. P. factory neurons are excitable and respond to odors. ( 1992). Nuclear and cytoplasmic localization of basic Dev. Brain Res. 53: 125- 130. fibroblast growth factor in astrocytes and CA2 hippo- POTTEN,C. S. and LOEFFLER,M. ( 1990). Stem cells: campal neurons. J. Neurosci. 12: 142-1 52. attributes, cycles, spirals, pitfalls and uncertainties. YUAN,J., SHAHAM,S., LEDOUX,S., ELLIS,H. M., and Lessons for and from the crypt. Development 110: HORVITZ,H. R. (1993). The C. elegans cell death 1001-1 020. gene ced-3 encodes a protein similar to mammalian REH,T. A. and TULLY,T. ( 1986). Regulation of tyro- interleukin- 1P-converting enzyme. Cell 75:64 1-652.