THE JOURNAL OF COMPARATIVE NEUROLOGY 395:273–280 (1998) Two Olfactory Marker Proteins in Xenopus laevis PATRICIA RÖSSLER, MARIO MEZLER, AND HEINZ BREER* University Stuttgart-Hohenheim, Institute of Physiology, D-70593 Stuttgart, Germany ABSTRACT Mature olfactory receptor neurons of mammals are characterized by the expression of the highly conserved olfactory marker protein (OMP) encoded by single copy genes. In Xenopus laevis, two homologous genes encoding olfactory marker proteins have been identified that share a sequence identity with mammalian OMPs of about 50%. Sequence comparison revealed significant variability in the N-terminus and C-terminus regions; in contrast, two internal domains were highly conserved between amphibian and mammalian OMPs, suggesting some functional relevance. The two OMP subtypes were regionally expressed in the olfactory nasal epithelium of Xenopus. XOMP1 transcripts were more abundant in the lateral diverticulum and XOMP2 in the medial diverticulum. The lateral location of XOMP1 and medial location of XOMP2 correspond to the suggested locations of olfactory receptor neurons responsive to water-borne and air-borne odorants, respectively. J. Comp. Neurol. 395:273– 280, 1998. r 1998 Wiley-Liss, Inc. Indexing terms: mature olfactory receptor neurons; amphibia; subtypes; expression pattern The olfactory marker protein (OMP) is an abundant 19-kDa cytoplasmic protein that is almost exclusively expressed in vertebrate mature olfactory sensory neurons (Margolis, 1985, 1988). Although the tissue-specific expression and some recent observations on OMP-deficient mice (Buiakova et al., 1996) imply an involvement in chemosensory signal transduction, the biological function of OMP remains enigmatic. Deciphering the amino acid sequence of rat OMP revealed a unique structure; there is no evidence for any related protein (Margolis, 1985). A comparison of the sequences of OMP from rat, mouse, and human revealed high conservation among the mammalian OMPs (Buiakova et al., 1994) but provided no hints to some conserved and potentially important functional domains. Determining the primary structure of OMP in nonmammalian vertebrate species may be more promising for discovering selective conservation of distinct domains during vertebrate evolution. Because some previous studies have demonstrated that antibodies against rat OMP showed weak but significant immunoreactivity in the nasal epithelium of nonmammalian vertebrates (Keller and Margolis, 1975; Margolis, 1980; Rama Krishna et al., 1992; Riddle and Oakley, 1992), we have set out to identify the primary structure of OMP in Xenopus laevis. MATERIALS AND METHODS The experiments comply with the Principles of Animal Care, Publication 85–23 (revised 1985) of the National r 1998 WILEY-LISS, INC. Institute of Health and also with the current laws of Germany. Screening of an olfactory cDNA library of Xenopus laevis Olfactory tissue was isolated from nasal cavities of 50 adult Xenopus laevis (Kähler, Hamburg). The tissue comprised sensory epithelium of the medial diverticulum (MD), the lateral diverticulum (LD), and the vomeronasal organ (VNO). The tissue was used for isolation of poly(A)1 RNA and subsequent construction of a directional oligo (dT)-primed Xenopus olfactory library in phage vector l Zap Express (Stratagene, La Jolla, CA). DNA of 3.6 3 105 independent recombinant phage clones was transferred to nylon membranes (Pall, East Hills, NY) and were subsequently prehybridized at 37°C for 3 hours in a solution containing 30% formamide, 53 standard saline citrate (SSC; 13 SSC 5 150 mM sodium chloride, 15 mM sodium citrate, pH 7.5), 0.02% sodium dodecylsulfate (SDS), 2% blocking reagent (Boehringer, Mannheim, Germany), 0.1% laurylsarcosin, and 100 µg/ml herring sperm DNA. Hy- Grant sponsor: Deutsche Forschungsgemeinschaft; Grant number: Br 712/16–1; Grant sponsor: Human Science Frontier Program; Grant sponsor: Fond der Chemischen Industrie. *Correspondence to: Prof. Dr. Heinz Breer, University StuttgartHohenheim, Institute of Physiology, Garbenstrasse 30, D-70593 Stuttgart, Germany. E-mail: email@example.com Received 6 October 1997; Revised 7 January 1998; Accepted 23 January 1998 274 P. RÖSSLER ET AL. Fig. 1. A: Deduced amino acid sequences of Xenopus olfactory marker protein (OMP) genes XOMP1 and XOMP2. Differences in the amino acid sequences are shaded. Most of the 14 amino acid substitutions are conservative exchanges and are scattered over the sequence. B: Alignment of Xenopus OMPs XOMP1 and XOMP2 with OMP proteins from human, mouse, and rat. Amino acid residues conserved in all sequences are indicated by an asterisk. Amino acids sharing .50% identity in all sequences are shaded. Dashes represent gaps in the amino acid sequence. bridization was performed overnight at 37°C in the same buffer with 2.5 ng/ml of digoxigenin (Dig)-labeled DNA encoding the complete coding region of the rat OMP (Rogers et al., 1987). The DNA was labeled by using the method of Feinberg and Vogelstein (1983) with the random priming Dig-DNA labeling kit (Boehringer). After hybridization, filters were rinsed for 2 3 2 minutes in 53 SSC, 0.1% SDS at room temperature, followed by a wash in 23 SSC, 0.1% SDS at 37°C for 4 3 20 minutes. Chemiluminescent detection was performed by following the Dig system user’s guide for filter hybridization of Boehringer. Filters were exposed to Fuji RX films for 16 hours. corresponded to Xenopus XOMP1. Four clones were identical and corresponded to Xenopus XOMP2. Sequencing primers were T3 primer—58 ATT AAC CCT CAC TAA AGG GA 38, T7 primer—58 TAA TAC GAC TCA CTA TAG GG 38, primer position 551—58 AAT (G/A)TT CCA TTG GCT TCA 38, primer position 843 for XOMP1, 815 for XOMP2—58 GT(A/G) TGT TAC (C/T)AA GTG TAG 38, primer position 1358 specific for XOMP1—58 GAC TAC TGG TGG GTT CAG 38, and primer position 1339 specific for XOMP2—58 GGT ACC ATC TCC CAC TG 38. Analysis of sequence data was performed by using the HUSAR 3.0 software package based on the sequence analysis software package 7.2 from the Genetic Group (Madison, WI). Isolation of full-length cDNA clones for DNA sequencing and analysis of sequence data Eleven positive phage plaques were isolated; of these, five l clones showed positive hybridization signals in the rescreen. These were recovered as phagemid clones by using the ExAssist helper phage, and DNA was prepared by alkaline lysis by using the Qiagen Midi kit (Chatsworth, CA). Both strands of each recombinant phagemid were sequenced by using the RR Dye Deoxy Terminator cycle sequencing kit (Perkin Elmer, Weiterstadt, Germany). Automatic sequencing was performed on an ABI 310 genetic analyzer. One of the five analyzed clones Southern blot analysis High-molecular-weight DNA was prepared from muscle tissue of an individual adult Xenopus according to the method of Ausubel et al. (1987). Ten micrograms of genomic DNA were digested with restriction endonucleases EcoRI and KpnI, size fractionated on a 1% agarose gel, and transferred to Hybond N1 nylon membrane (Amersham, Braunschweig, Germany) by using standard protocols (Sambrook et al., 1989). A 382-bp fragment of XOMP1 cDNA encoding amino acids 19–146 was labeled by using XENOPUS OLFACTORY MARKER PROTEINS 275 the Dig DNA labeling kit (Boehringer). The blot was hybridized under high stringency conditions to the Dig-labeled DNA probe, as described by Engler-Blum et al. (1993), after the optimized hybridization and detection protocol (blocking reagent and anti-Dig AP-Fab fragments were obtained from Boehringer). Subsequently, the membrane was exposed to Fuji RX film for 12 hours. Northern blot analysis For the analysis of tissue-specific expression of XOMP1 and XOMP2, total RNA from Xenopus olfactory epithelium (OE), olfactory bulb (OB), brain (B), and liver (L) was prepared by tissue homogenization in TRIzol reagent (Gibco BRL, Eggenstein, Germany) according to the manufacturer’s instructions. Five micrograms of total RNA were separated on a 1.2% agarose gel containing formaldehyde and were blotted (Sambrook et al., 1989). Prehybridization and hybridization was performed at 68°C by using Dig easy hyb buffer (Boehringer). The hybridization buffer contained 100 ng/ml Dig-labeled RNA of either XOMP1 or XOMP2 full-length clones. The antisense RNA probes were generated by using the T3/T7 RNA transcription system according to the manufacturer’s specifications (Boehringer), with recombinant pBK-CMV phagemid vectors as a template. In brief, 2 micrograms of linearized vector were transcribed in the presence of 70 nmol digoxigenin-11uridine-58-trisphosphate, followed by partial alkaline hydrolysis of the RNA according to Angerer and Angerer (1992), thereby producing fragments of about 200 bases in length. After hybridization, the membrane was washed in 0.13 SSC, 0.1% SDS at 68°C. After chemiluminescent detection, films were exposed to Fuji RX film for 11 hours. Reverse transcriptase–polymerase chain reaction (RT-PCR) To analyze the expression of both XOMP subtypes in the different subcompartments of the nasal cavity, total RNA was extracted from the olfactory bulb and sensory epithelia of the VNO, LD, and MD of single animals, as described above. Total RNA was dissolved in 20 µl of RNase-free water followed by a DNase digestion (DNase I, Gibco BRL). After the spectrometric determination of the amount of total RNA, equal amounts of mRNA were isolated (Dynabeads, Dynal, Oslo, Norway). Messenger RNA redissolved in 10 µl RNase-free water was reverse transcribed by using a first-strand cDNA synthesis kit according to the manufacturer’s instructions (Pharmacia, Freiburg, Germany). To compare the amount of template used in each reaction, empirical tests were performed basically as described previously (Rappolee et al., 1988; Deutsch Murphy et al., 1990; Wilson and Melton, 1994). To make sure that equal amounts of cDNA were used as a template throughout all tissue samples in subsequent PCR reactions, additional PCR experiments were performed by using the housekeeping gene L8 as endogenous control. To amplify either XOMP1 or XOMP2, specific oligonucleotide antisense primers corresponding to the 38 untranslated region of each subtype were combined with a general XOMP sense primer. The primers were sense for XOMP1 and XOMP2 (position 469)—58 CTT TCT TAG ATG GCG CTG ACC 38, antisense for XOMP1 (position 872)—58 GTG GTT ATT TCT CTA CAC TTG G 38, antisense for XOMP2 (1017)—58 ACA CAC TTT TTT GTC TTG GG 38, primers for L8 sense—58 TTG CAT TCC GTG ATC CTT ACA GG 38, Fig. 2. Southern blot analysis of Xenopus XOMP genes. Xenopus genomic DNA isolated from muscle tissue was digested with restriction endonuclease EcoRI (lane 1) and KpnI (lane 2), electrophoresed on a 1% agarose gel, and blotted on a nylon filter. The blot was hydridized with a Dig-labeled probe that corresponded to a 382-bp fragment from XOMP1 cDNA encoding amino acids 19–146, which shows no internal recogntion sight for EcoRI or KpnI. Hybridizing fragments share a sequence homology of .90%. The position of HindIII and HindIII/EcoRI digested lambda-DNA is shown on the left in kilobase pairs. antisense—58 ATC TCT CCT GAT GGT TGA GGG 38. To have a linear correlation between template concentration and PCR products, the optimal PCR cycle number was determined. Amplification was carried out in 50 µl of 10 mM Tris-HCl (pH 8.8), 50 mM KCl, 1.5 mM MgCl2, with 200 µM of each dNTP, 100 pmol of each primer, and 2 U of Taq DNA polymerase (Gibco BRL). PCR was performed by using the following conditions: 94°C for 2 minutes, 57– 60°C for 2 minutes (depending on the primers), and 72°C for 4 minutes (1 cycle); 94°C for 30 seconds, 65°C for 1 minutes, and 72°C for 50 seconds (20–35 cycles); and 72°C for 8 minutes. The optimal number of cycles were 20 cycles for L8 at an annealing temperature of 60°C and 35 cycles for both XOMP subtypes at 57°C. To determine the specificity of the PCR primers and for quantification purposes, the PCR products were separated on 1% agarose gels and blotted, as described above. Dig-labeled probes corresponding to the 38 untranslated region of either XOMP1 or 276 P. RÖSSLER ET AL. Fig. 3. A: Northern blot analysis of XOMP1 and XOMP2 expression in the nasal organ, the olfactory bulb, and nonolfactory tissues of Xenopus. Hybridization was performed under high stringency conditions with 5 µg of total RNA from Xenopus nasal cavity, olfactory bulb, brain, and liver by using antisense riboprobes of both full-length clones. The blots were exposed for 11 hours. The position of the 18-S and 28-S ribosomal RNA bands are indicated. XOMP mRNA of both subtypes is expressed in the olfactory epithelium (OE), whereas no expression of either XOMP1 or XOMP2 is detectable in the olfactory bulb (OB), brain (B), and liver (L). In the left blot, the native 1.8-kb XOMP1 mRNA is visible; in the right blot, expression of XOMP2 mRNA of a slightly smaller size (1.6 kb) is detectable. Note the higher level of XOMP2 mRNA. B: Compartment-specific expression of XOMP genes revealed by semiquantitative reverse transcriptase–polymerase chain reaction (RT-PCR). RNA from all three compartments including vomeronasal organ (VNO), lateral diverticulum (LD), medial diverticulum (MD), and olfactory bulb (OB) of one Xenopus nasal cavity was subjected to RT-PCR analysis by using primer pairs specific for the 38-untranslated region of either XOMP1 or XOMP2. To compare equal amounts of cDNA, additional PCR experiments were performed by using the housekeeping gene L8 as an endogenous control. The PCR samples of L8 were separated by electophoresis to compare band intensities (lower panel). The linearly amplified XOMP PCR products were analyzed on Southern blots by using labeled XOMP-subtype specific probes corresponding to the 38-untranslated regions. The amount of probe was similar for both XOMP subtype probes. In the LD, a preferential expression of XOMP1 was recognized. In the MD, XOMP2 was preferentially expressed. In the VNO, small amounts of XOMP1 and XOMP2 transcripts were observed. In the OB, no XOMP expression was detected. XOMP2 were hydridized as described by Engler-Blum et al. (1993). The PCR products of the housekeeping gene L8 were measured with ethidium bromide staining. and rapidly frozen in an N2 cooled isopentane bath. Coronal sections 10 µm thick were cut on a Reichert & Jung cryostat (model 2800 E) at 230°C, adhered to Superfrost plus (Fisher, Orangeburg, NY) microslides, and air dried for 2 hours. For in situ hybridization, tissue sections were covered with 10 µl of hybridization solution containing 50% deionized formamide and 3–5 ng Diglabeled antisense RNA, and coverslipped. Hybridization was carried out at 55°C for 16 hours in closed humid boxes. After incubation, sections were washed twice for 30 min- In situ hybridization Adult Xenopus laevis were decapitated, and the upper jaws were fixed in 4% paraformaldehyde, 1% picric acid in phosphate buffered saline, pH 7.3, for 2 hours. After cryoprotection in 25% sucrose buffer overnight at 4°C, the jaws were embedded in Tissue Tek (Miles, Elkhart, IN) XENOPUS OLFACTORY MARKER PROTEINS 277 utes in 0.13 SSC at 60°C. Hybridization was visualized by using an anti-Dig AP antibody (1:750, Boehringer) for 30 minutes at 37°C, followed by two washes in Tris-buffered saline (100 mM Tris-HCl, 150 mM NaCl, pH 7.0) for 15 minutes. Bound antibodies were visualized by using nitroblue tetrazolium and bromo-chloro-indolyl phosphate (Biomol, Hamburg, Germany) as substrates. Subsequently, sections were mounted in Euparal (Roth, Karlsruhe, Germany) and examined under a Zeiss Axiphot microscope by using Nomarski phase-contrast optics. In control experiments, sense RNAs were transcribed and hybridized to tissue sections, as described for antisense probes. No signals were observed in any of these controls. RESULTS An olfactory cDNA library from Xenopus laevis was screened by means of a rat OMP DNA probe. Two different full-length cDNA clones, XOMP1 and XOMP2, were isolated under low stringency conditions. The complete nucleotide sequences of these cDNA clones are deposited in the database. The open reading frames of both clones comprise 474 bp and are flanked by 50 bp of untranslated region on the 58 side. The 38 flanking regions differ in length; whereas XOMP1 comprises a 38 tail of 1,275 nucleotides, XOMP2 contains a 38 untranslated region of 1,048 bp (including the stop codon). The open reading frames of XOMP1 and XOMP2 encode polypeptide chains of 158 amino acids with calculated molecular masses of 18,724 Da (XOMP1) and 18,445 Da (XOMP2). A comparison of the amino acid sequences of XOMP1 and XOMP2 (Fig. 1A) demonstrated an identity of 91%, which is similar to the sequence identity among OMPs from mammalian species. However, the sequence identities between Xenopus and mammalian OMPs ranged only between 51% and 55%. Aligning the amino acid sequences of Xenopus OMPs and OMPs from mammals (rat, mouse, human) demonstrated a striking conservation of two domains (Fig. 1B). These regions, ranging from positions 75 to 100 and from 118 to 141, were highly conserved and demonstrated a significantly higher sequence identity of 92% and 83%; in contrast, both the N-terminus and the C-terminus differed significantly between OMPs from mammals and amphibia. Furthermore, both Xenopus OMPs exhibited three deletions in the N-terminus and also missed the final leucine at the C-terminus. In rat, OMP is encoded by a single copy gene (Danciger et al., 1989). To investigate how many OMP genes may exist in Xenopus, Southern blot analyses were performed by probing genomic DNA with a 382-bp fragment from XOMP1 cDNA encoding amino acids 19–146, which does not contain sites for EcoRI or KpnI. Thus, because the OMP genes are not supposed to contain introns within the coding regions, the number of hybridizing bands should indicate the number of OMP genes. Under high stringency conditions in each restriction endonuclease lane, four bands of different intensity were visualized, suggesting the existence of at least four different OMP genes in the Xenopus genome (Fig. 2). To examine, whether both XOMP genes are actually transcribed and whether the two XOMP subtypes display any tissue- or region-specific expression, riboprobes corresponding to the full-length clones of both XOMP1 and XOMP2 were hybridized to Northern blots from Xenopus Fig. 4. Cellular localization of XOMP gene expression in adult Xenopus olfactory epithelium by in situ hybridization. A: Section through the medial diverticulum (MD) incubated with a digoxigeninlabeled antisense riboprobe of XOMP2. Labeled cells are confined to the sensory part of the mucosa; no signals were detected in the nonsensory part. Arrowheads indicate the transition between the olfactory epithelium (OE) and adjacent respiratory epithelium (RE). B: A section through the lateral diverticulum (LD) probed with XOMP1. Within the epithelium, specific hybridization is restricted to olfactory receptor neurons (ORN) but not to sustentacular cells (SC) or basal cells (BC). Signals are absent from the lamina propria (LP). Scale bars 5 100 µm in A, 40 µm in B. olfactory epithelium, olfactory bulb, brain, and liver under high stringency conditions. The results (Fig. 3A) indicate that both probes only hybridized to RNA from the olfactory epithelium. A strong hybridization signal was obtained for the XOMP2 probe, whereas the XOMP1 probe only produced a weak response. Because both probes produced equal signal intensities on standardized templates, this result suggests that XOMP2 is more strongly expressed in the olfactory tissue. No hybridization was observed with RNA from the olfactory bulb, brain, and liver. The size of XOMP2 mRNA is slightly smaller than that of XOMP1 mRNA. The length of the hybridizing RNA is about 1.8 kb for XOMP1 and 1.6 kb for XOMP2; this is in good agreement with the size of the cloned cDNA, thus supporting the notion that XOMP1 and XOMP2 represent full-length cDNAs. 278 Fig. 5. In situ hybridization of adult Xenopus coronal sections through the sensory epithelia of all subcompartments probed with either XOMP1-specific or XOMP2-specific antisense RNA. A,B: Adjacent sections through the lateral diverticulum (LD) hybridized with XOMP1 and XOMP2. XOMP1 is strongly expressed in the sensory cells of the LD (A). The expression of XOMP2 is significantly weaker (B). C,D: Consecutive sections through the medial diverticulum (MD) probed with XOMP1 and XOMP2. C: XOMP1 strongly hybridized to only single neurons randomly distributed within the MD. The major- P. RÖSSLER ET AL. ity of neurons are weakly stained. D: Hybridization with the XOMP2specific probe intensively labeled many cells in the olfactory neuron layer. E: Coronal section of the vomeronasal organ (VNO) annealed with XOMP1. The probe labeled single, randomly distributed chemosensory cells in the VNO. These XOMP1-expressing neurons are surrounded by cells with a diffuse and very weak staining. Arrowhead indicates a single mature VNO neuron expressing XOMP1. Scale bars 5 50 µm. XENOPUS OLFACTORY MARKER PROTEINS The olfactory system of Xenopus consists of three separate and functionally different compartments: the main olfactory cavity, which is subdivided into the lateral diverticulum (LD) and the medial diverticulum (MD), and the vomeronasal organ (VNO; Föske, 1934; Altner, 1962). To approach the question if the two OMP subtypes were expressed differently in the nasal subcompartments, the appropriate olfactory tissue from an individual animal was dissected. Because of the very small quanitites of tissue, samples were analyzed by following a semiquantitative RT-PCR protocol (Zamorano et al., 1996) with specific primers corresponding to the 38 untranslated region of XOMP1 and XOMP2. The results shown in Figure 3B indicate that the endogenous internal standard, the housekeeping gene L8 (Shi and Liang, 1994), was amplified at constant levels throughout all tissue samples, demonstrating the suitability of the procedure. XOMP1 was expressed in all three cavities, but the highest level of expression was found in the lateral diverticulum; XOMP2 transcripts were also detected in the three cavities, but XOMP2 was preferentially expressed in the medial diverticulum. In contrast to mammals (Wensley et al., 1995), in Xenopus there was no indication of XOMP mRNA in the olfactory bulb (Fig. 3B). To analyze the topographic localization of XOMPexpressing cells in the nasal epithelium of adult Xenopus, in situ hybridization experiments were performed by using digoxigenin-labeled antisense probes from the full-length clones under high stringency conditions. Labeled cells were only found in the chemosensory epithelium of the nasal cavity. No hybridization was detected in the adjacent respiratory epithelium (Fig. 4A). Within the olfactory epithelium, reactive cells were situated in the layer of the neuroepithelium, which consists of mature receptor neurons (Fig. 4B). The stretch of mucosa above the basal membrane, where immature neurons are located, was devoid of labeled cells. Sustentacular cells in the apical layer of the neuroepithelium showed no staining. Hybridization signals were also absent from the lamina propria beneath the epithelium (Fig. 4B). In detail analysis of adjacent sections through the lateral diverticulum demonstrated that XOMP1 is strongly expressed in many cells of the LD epithelium (Fig. 5A), whereas a significantly weaker staining was obtained with the XOMP2 probe (Fig. 5B). Employing a XOMP1-specific probe to coronal sections through the medial diverticulum demonstrated only few intensely hybridizing neurons randomly distributed within the epithelium; the majority of cells produced very weak signals (Fig. 5C). Probing adjacent sections through the epithelium of the medial diverticulum with a XOMP2specific riboprobe resulted in hybridization signals of high intensity in the layer where mature neurons are situated (Fig. 5D). These observations are in line with the RT-PCR results. In the vomeronasal organ, expression of both XOMP subtypes appeared to be restricted to very few cells on consecutive sections. The punctate hybridization pattern of XOMP1 in the VNO is shown in Figure 5E. The specific probe for XOMP2 produced similar results (data not shown). The significant differences concerning the number and intensity of hybridization signals in all three subcompartments are consistent with the different amounts of transcripts determined by semiquantitative RT-PCR experiments. 279 DISCUSSION In the present study, two subtypes of olfactory marker protein were identified in Xenopus laevis. Sequence comparison demonstrated a moderate overall homology with the highly conserved OMPs from three mammalian species; however, two domains emerged that are identical in all OMP subtypes. Conservation of the primary structure during vertebrate evolution suggests that these domains may be of considerable relevance for structure and/or function of the OMPs. Furthermore, elucidation of these highly conserved domains may open the way for novel experimental approaches, such as the yeast two-hybrid system for identification of target proteins, and ultimately disclose the functional role of these proteins in olfactory neurons. Whereas rat OMP seems to be encoded by a single copy gene (Danciger et al., 1989), Southern blot analysis of Xenopus genomic DNA demonstrated four bands. This result suggests that four different OMP genes exist in Xenopus, possibly as a consequence of genome duplication as a result of tetraploidization. Because intensive screening of the Xenopus olfactory cDNA library led to the isolation of only two different OMP cDNA clones and both XOMP subtypes generated a single band in Northern blots, it seems likely that only two XOMP genes are expressed within the nose of the frog. However, the observation that in the VNO only very few cells express XOMP1 or XOMP2 may indicate that additional XOMP subtypes are expressed. The preferential expression of XOMP1 in mature olfactory neurons of the lateral diverticulum and XOMP2 in the sensory cells of the medial diverticulum may contribute to the functional compartmentalization of the olfactory system. In Xenopus, the lateral diverticulum is supposed to be specialized for watersoluble odors (Altner, 1962) and selectively express ‘‘fishlike’’ receptors, whereas the medial diverticulum seems to be responsible for the reception of volatile odorants and is equipped with ‘‘mammalian like’’ receptors (Freitag et al., 1995). Therefore, each XOMP subtype may be expressed in a distinct subset of functionally specialized olfactory neurons. Although the functional role of OMP in mature olfactory neurons still awaits elucidation, the identification of OMP from nonmammalian vertebrates may shed some new light on the evolution of OMP genes. LITERATURE CITED Altner, H. (1962) Untersuchungen über die Leistungen und Bau der Nase des südafrikanischen Krallenfrosches Xenopus laevis (Daudin, 1803). Z. Vergl. Physiol. 45:272–306. Angerer, L.M. and R.C. Angerer (1992) In situ hybridization to cellular RNA with radiolabelled RNA probes. In D.G. Wilkinson (ed): In Situ Hybridization. Oxford, England: IRL, pp. 15–32. Ausubel, F.M., R. Brent, R.F. Kingston, D.D. Moore, J.G. Seidmann, J.A. Smith, and K. Struhl (1987) Current Protocols in Molecular Biology. New York: Greene and Wiley. Buiakova, O.I., N.S. Rama Krishna, T.V. Getchell, and F.L. Margolis (1994) Human and rodent OMP genes: Conservation of structural and regulatory motifs and cellular localization. Genomics 20:452–462. Buiakova, O.I., H. Baker, J.W. Scott, A. Farbman, R. Kream, M. Grillo, L. Franzen, M. Richman, L.M. Davis, S. Abbondanzo, C.L. Stewart, and F.L. Margolis (1996) Olfactory marker protein (OMP) gene deletion causes altered physiological activity of olfactory sensory neurons. Proc. Natl. Acad. Sci. U.S.A. 93:9858–9863. Danciger, E., C. Mettling, M. Vidal, R. Morris, and F.L. Margolis (1989) Olfactory marker protein gene: Its structure and olfactory neuronspecific expression in transgenic mice. Proc. Natl. Acad. Sci. U.S.A. 86:8565–8569. 280 Deutsch Murphy, L., C.E. Herzog, J.B. Rudick, A.T. Foje, and S.E. Bates (1990) Use of the polymerase chain reaction in the quantification of mdr-1 gene expression. Biochemistry 29:10351–10356. Engler-Blum, G., M. Meier, J. Frank, and G.A. Müller (1993) Reduction of background problems in nonradioaktive Northern and Southern blot analyses enables higher sensitivity than 32P-based hybridizations. Anal. Biochem. 210:235–244. Feinberg, A.P. and B. Vogelstein (1983) A technique for radiolabelling DNA restriction fragments to high specific activity. Anal. Biochem. 132:6–13. Föske, H. (1934) Das Geruchsorgan von Xenopus laevis. Z. Anat. Entweil. 103:519–550. Freitag, J., J. Krieger, J. Strotmann, and H. Breer (1995) Two classes of olfactory receptors in Xenopus laevis. Neuron 15:1383–1392. Keller, A. and F.L. Margolis (1975) Immunological studies of the rat olfactory marker protein. J. Neurochem. 24:1101–1106. Margolis, F.L. (1980) A marker protein for the olfactory chemoreceptor neuron. In R.A. Bradshaw and D.M. Schneider (eds): Proteins of the Nervous System, 2nd ed. New York: Raven Press, pp. 59–84. Margolis, F.L. (1985) Olfactory marker protein: From PAGE band to cDNA clone. Trends Neurosci. 8:542–546. Margolis, F.L. (1988) Molecular cloning of olfactory specific gene products. In F.L. Margolis and T.V. Getchell (eds): Molecular Neurobiology of the Olfactory System. New York: Plenum, pp. 237–265. Rama Krishna, N.S., T.V. Getchell, F.L. Margolis, and M.L. Getchell (1992) P. RÖSSLER ET AL. Amphibian olfactory receptor neurons express olfactory marker protein. Brain Res. 593:295–398. Rappolee, D.A., C.A. Brenner, R. Schultz, D. Mark, and Z. Werb (1988) Develompental expression of PDGF, TGF-a, and TGF-b genes in preimplantation mouse embryos. Science 241:1823–1825. Riddle, D.R. and B. Oakley (1992) Immunocytochemical identification of primary olfactory afferents in rainbow trout. J. Comp. Neurol. 324:575– 585. Rogers, K.E., P. Dasgupta, U. Gubler, M. Grillo, Y.S. Khew-Goodall, and F.L. Margolis (1987) Molecular cloning and sequencing of a cDNA for olfactory marker protein. Proc. Natl. Acad. Sci. U.S.A. 82:5218–5222. Sambrook, J., E.F. Fritsch, and T. Maniatis (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Shi, Y.-B. and V.C.-T. Liang (1994) Cloning and characterization of the ribosomal protein L8 gene from Xenopus laevis. Biochim. Biophys. Acta 1217:227–228. Wensley, C.H., D.M. Stone, H. Baker, J.S. Kauer, F.L. Margolis, and D.M. Chikaraishi (1995) Olfactory marker protein mRNA is found in axons of the olfactory receptor neurons. J. Neurosci. 15:4827–4837. Wilson, P.A. and D.A. Melton (1994) Mesodermal patterning by an inducer gradient depends on secondary cell–cell communication. Curr. Biol. 4:676–686. Zamorano, P.L., V.B Mahesh, and D.W. Brann (1996) Quantitative RT-PCR for neuroendocrine studies. Neuroendocrinology 63:397–407.