We have identified and cloned portions of two Drosophila genes homologous to two classes of mam- malian intracellular Ca2+-release channels, the ryan- odine receptor and the inositol 1,4,5-triphosphate (IP3) receptor. The Drosophila ryanodine receptor gene (dry) encodes an approx. 15 kb mRNA. It is expressed in the mesoderm of early stage-9 embryos and subsequently in somatic muscles and their precursor cells. In adults, dry mRNA was detected in tubular muscles and at a lower level in neuronal tissues. Embryonic expression of the Drosophila IP3 receptor gene (dip) appears more dynamic and is associated with developing anterior sense organs. In adults, dip expression occurs in several tissues, and relatively high levels of dip mRNA in adult antennae suggest a role for this gene product during olfactory transduction.

The release of intracellular Ca2+ is an intermediate step in many cellular signaling processes (Berridge and Irvine, 1989; Tsein and Tsein, 1990). In vertebrates, two classes of proteins, the inositol 1,4,5-triphosphate (IP3) receptor and the ryanodine receptor, act as channels for the release of intracellular Ca2+. The IP3 receptor causes release of intracellular Ca2+ in response to IP3, which is generated during signaling mechanisms that involve activation of phospholipase C (Majerus et al., 1985). This signal trans- duction pathway is used in processes as diverse as the responses to hormones, growth factors and neurotransmit- ters (Berridge and Irvine, 1984), as well as in various sen- sory systems such as olfaction (Reed, 1992), gustation (Hwang et al., 1990) and vision (Payne et al., 1988). It must also function in the central brain, the tissue from which it was initially purified and cloned (Furuichi et al., 1989; Mignery et al., 1990). Ryanodine receptor function is best understood in vertebrate skeletal muscle, in which this pro- tein spans the junction between the T-tubules of the plasma membrane and the sarcoplasmic reticulum. It is required for the intracellular Ca2+-release that occurs prior to muscle contraction, in response to nerve impulses delivered to the muscle plasma membrane (Catterall, 1991). A second form of the ryanodine receptor has been identified in cardiac muscle and more recently in vertebrate brain (McPherson et al., 1991), where its function is poorly understood. In addition, both the IP3 and ryanodine receptors have been postulated to function during Ca2+-induced Ca2+-release in neuronal and non-neuronal tissues requiring Ca2+ oscilla- tions (Tsein and Tsein, 1990). The presence of these intra- cellular Ca2+ channels in such diverse tissues indicates that they are likely to be involved in many different cellular functions.

Drosophila offers an ideal system to examine the role of these proteins during cell signaling in development and in adult sensory and central brain processes. To this end, we have cloned and characterized two Drosophila genes that code for these two classes of intracellular Ca2+-release channels. Our results suggest that both genes are well con- served between mammals and Drosophila. We have also analyzed the expression of these genes during development and in adults. Their expression patterns suggest that they function both during embryonic development and in adult tissues. Spatial distribution of the mRNAs for the Drosophila IP3 and ryanodine receptors indicates that both channels serve neuronal functions. In addition, the ryan- odine receptor is likely to function at the neuromuscular junction.

Molecular cloning by PCR

The primers used in PCR experiments for cloning of the dry gene were designed to anneal to the cDNA sequence encoding the amino acids (T/K)CFICG and EEHNM respectively. The exact sequences were as follows: 5′ primer A(C/A)- (C/G/T/A)TG(C/T)TT(C/T)AT(C/T/A)TG-(C/T)GG; 3′ primer CAT(G/A)TT(G/A)TG(C/T)TC(C/T)TC. Total RNA from heads of Canton-S wild-type flies was extracted by standard procedures and treated with RNAase-free DNAaseI. Approximately 1 μg of this RNA was heated at 90°C for 5 minutes and added to a reverse transcription (RT) reaction containing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 0.01% (w/v) gelatin, 1 mM each dATP, dGTP, dCTP and dTTP, 40 units RNasin (Promega), 50 μM of the 3′ primer and 200 units of Super-RT reverse tran- scriptase (BRL) in a total volume of 20 μl. The RT reaction was incubated at 42°C for 60 minutes and at 95°C for 10 minutes. It was then added to 80 μl of the same reaction mix minus both RNasin and reverse transcriptase and containing 125 μM of each dNTP, 50 μM of the 5′ and 3′ primers and 2.5 units of Taq poly- merase (Boehringer-Mannheim). After an initial 2 minutes at 94°C, amplification was allowed to proceed for 5 cycles with the following regime: 94°C, 1 minute; 42°C, 2 minutes; 72°C, 3 min- utes; followed by 25 cycles of: 94°C, 1 minute; 50°C, 2 minutes; 72°C, 3 minutes. A 90 bp product from the amplification reaction was gel-purified and cloned into the SmaI site of the plasmid vector Bluescript KS from USB. The DNA sequence was deter- mined from the double stranded plasmid using a Sequenase kit (USB).

For cloning of the dip gene essentially the same procedures were employed except that poly(A)+ RNA from heads was used for the RT reaction. Primers were designed from the amino acids EHNMWHY and MTEQRKQ with the following sequences: 5′ primer - GA(G/A)CA-(C/T)AA(C/T)ATGTGGCA(C/T)TA; 3′ primer - TG(C/T)TT(C/G/T/A)C(G/T)(C/T)TG(C/T)TC- (C/G/T/A)GTCAT.

cDNA isolation and sequencing

The cloned PCR fragments for the dry and dip genes were used as probes to screen a head cDNA library in the vector lEXLX (Palazzolo et al., 1990). Larger cDNA clones obtained in this way were sequenced using sequential oligonucleotide primers from both strands. Analysis of the DNA sequence was performed with the GCG software package for VAX computers (Devereux et al., 1984).

Northern blots

The poly(A)+ RNA from embryonic developmental stages, larvae and pupae was a kind gift of S. DeSimone. Poly(A)+ RNA from adult heads and bodies was prepared by standard procedures. Approximately 2-5 μg were loaded per lane of a 0.75% formalde- hyde agarose gel and subsequently transferred to a Hybond-N nylon membrane (Amersham). Blots were hybridized to 32P- labelled cDNA fragments of the respective genes under standard conditions.

In situ hybridization to polytene chromosomes

Probes for in situ hybridization to polytene chromosomes were labelled by random priming with biotinylated dUTP (Enzo Diag- nostics). The hybridization was visualized with streptavidin-con- jugated horseradish peroxidase in a kit from Enzo Diagnostics.

In situ hybridization to whole-mount embryos

In situ hybridization to embryos was carried out with the Genius DNA labelling and detection kit (Boehringer Mannheim) accord- ing to the procedure of Tautz and Pfiefle (1989). Digoxigenin- labelled probes were made by random-primed labelling of Cdna fragments reduced to a size of approx. 300 bp. Random priming reactions were carried out at 15°C overnight and then at room temperature for 2 hours with an addition of 5 units of Klenow fragment.

In situ hybridization to adult frozen sections

Frozen sections of 10 μm were processed and hybridized to digoxigenin-labelled probes as described by Nighorn et al. (1991). Probes were labelled following the procedure described for embryo in situs.

Quantitative PCR (RT-QPCR)

Adult flies were snap-frozen in dry ice and shaken so as to break off their heads, antennae and legs. These tissues were separated by hand for subsequent steps. Eyes were purified by a published procedure (Matsumoto et al., 1982) and were a kind gift of L. Zwiebel. Total RNA was extracted from 50 heads, 50 antennae, 25 legs and 50 eyes using standard procedures and subjected to extensive DNAase treatment. The relative abundance of different transcripts was measured using a reverse-transcription-based com- petitive PCR assay (RT-QPCR; Gilliland et al., 1990). The assay involved the coamplification of the target RNA with a competi- tive RNA template (spike), thereby minimizing most of the vari- ability normally associated with PCR. Known amounts of spike RNA were added to the total RNA samples, and control experi- ments showed that after coamplification the resulting ratio of target DNA product to spike DNA product was an accurate measure of the amount of target transcript. Primers were designed to generate a product of 197 bp for the dry mRNA, a product of 220 bp for the dip mRNA, and a product of 360 bp for the RP49 mRNA. The spikes for dip and dry were constructed by deleting 30 and 40 bp, respectively, between the two primer sites. The spike for RP49 was generated by using the corresponding piece of genomic DNA that contains an intron of 80 bp between the two primer sites and which had been cloned in the transcription vector pSP65. The other two DNAs (dip and dry) were cloned into a transcription vector (TA from Invitrogen) using the manu- facturers instructions. All three spike DNA constructs were tran- scribed into RNA. The primers specific for each RNA were used for RT-QPCR performed with approximately equivalent amounts of RNA from each preparation. Approximately 0.5 ng of total RNA, together with an empirically determined optimal quantity of spike, was used for each reverse transcription reaction with M- MLV reverse transcriptase (BRL) and carried out according to conditions recommended by the suppliers. The 20 μl reactions contained 50 pmoles of the 3′ primer. After heat-killing, 10 μl of each reaction were added to 40 μl of a PCR mix containing the 5′ and 3′ primers at 1 μM each final concentration. PCR was car- ried out with Vent Polymerase (exo minus; New England Bio- labs) using conditions specified by the suppliers. 35S-dATP (0.25 μCi) was added to each reaction in order to visualize the PCR products with a Molecular Dynamics Phosphorimager. The PCR was performed on a Perkin-Elmer GeneAmp PCR System 9600 as follows: Each cycle consisted of 95°C, 5 seconds; 50°C, 5 sec- onds; 72°C, 45 seconds. 28 cycles were used for RP49, 31 for dry and 36 for dip. Control reactions for reverse transcription and for the PCR were performed in each case. There was no significant contamination with genomic DNA as there was no detectable signal without the addition of reverse transcriptase. Quantitation of the appropriate PCR products was performed with the phos- phorimager and the ratio of target band to spike band calculated. The amount of RP49 mRNA in each RNA preparation was measured relative to spike, and the values obtained from the dip and dry were normalized relative to RP49 values.

Isolation of the Drosophila homologs for the ryanodine receptor and the IP3 receptor gene by PCRs with degenerate oligonucleotides

The mammalian ryanodine and IP3 receptors share signif- icant homology at their carboxy-terminal (C-terminal) ends (Furuichi et al., 1989). Degenerate oligonucleotides were designed from two closely spaced regions that are well con- served between the two mammalian proteins (see Materi- als and methods and Fig. 1). These oligonucleotides were used in reverse transcription and polymerase chain reac- tions (PCR - Saiki et al., 1985) of total head RNA from Drosophila melanogaster, and a fragment of approximately the right size (90 bp) was obtained and cloned. Ten inde- pendent clones derived from two PCRs were analyzed, and in all cases the sequences between the priming oligonu- cleotides were found to be identical. Conceptual translation of this nucleotide sequence yielded protein sequence infor- mation which suggested that the PCR amplified fragment was from a Drosophila ryanodine receptor gene (dry). This was confirmed by isolating larger clones from a head cDNA library using the 90 bp fragment as a hybridization probe. These clones coded for the C-terminal end of the protein, as expected from the positions of the primers used to iso- late the original 90 bp PCR fragment. This is also evident on alignment of the Drosophila protein sequence with the mammalian protein. The rabbit cardiac muscle ryanodine receptor is composed of 4969 amino acids (Otsu et al., 1990), of which the 509 C-terminal amino acids share 61% identity and 79% similarity with the sequenced portion of the Drosophila ryanodine receptor gene, subsequently referred to as dry (Fig. 1).

Fig. 1.

Sequence alignment of the carboxy-terminal ends of the Drosophila ryanodine receptor (Dry) with the rabbit cardiac muscle ryanodine receptor (rabryr), and the Drosophila IP3 receptor (Dip) with the rat brain IP3 receptor (ratIp3). All four sequences have also been aligned. Bold letters indicate identical amino acids between two sequences of the same pair. Spaces indicated by dots have been introduced to maximize the alignments. Asterisks (*) indicate amino acids and underlines are putative membrane spanning domains that are conserved among all four proteins. Hashes (#) indicate the sequences from which primers were designed for the initial PCR.

Fig. 1.

Sequence alignment of the carboxy-terminal ends of the Drosophila ryanodine receptor (Dry) with the rabbit cardiac muscle ryanodine receptor (rabryr), and the Drosophila IP3 receptor (Dip) with the rat brain IP3 receptor (ratIp3). All four sequences have also been aligned. Bold letters indicate identical amino acids between two sequences of the same pair. Spaces indicated by dots have been introduced to maximize the alignments. Asterisks (*) indicate amino acids and underlines are putative membrane spanning domains that are conserved among all four proteins. Hashes (#) indicate the sequences from which primers were designed for the initial PCR.

Since none of the clones sequenced from the 90 bp fragment were derived from a Drosophila IP3 receptor gene, oligonucleotides from other regions of the mammalian IP3 receptor were designed for the purpose of cloning this gene. This was aided by the publication of regions of identity at the C terminus of the Drosophila and mouse IP3 receptor proteins (Miyawaki et al., 1991). PCRs using oligonu- cleotides specific for the IP3 receptor were performed with poly(A)+ RNA from Drosophila heads, and a fragment of the expected 350 bp size was obtained. Cloning and sequencing of this fragment indicated that it codes for a protein similar to the rat IP3 receptor. Larger cDNA clones complementary to this fragment were isolated from a head cDNA library and sequenced.

A comparison of the IP3 receptor protein sequences from rat (ratIP3) and Drosophila (Dip) is shown in Fig. 1, and demonstrates that the two proteins share extensive homol- ogy throughout the sequenced region (73% similarity and 57% identity). This region corresponds to the C-terminal 460 amino acids of the rat IP3 receptor. The full length rat receptor protein consists of 2749 amino acids (Mignery et al., 1990).

The two Drosophila genes described here are also related to each other; like the mammalian ryanodine and IP3 recep- tors, they share significant homology near their C termini (Furuichi et al., 1989). A comparison of the 300 C-termi- nal amino acids of the cloned Drosophila ryanodine and IP3 receptor proteins shows that they are 62.6% similar and 33.6% identical (Fig. 1). The relationship between all four proteins is also shown in Fig. 1. Both ryanodine receptor genes appear equally related to the two IP3 receptor genes, indicating that an ancient duplication event probably gave rise to these two classes of intracellular Ca2+-release chan- nel genes. The regions of identity among all four proteins (depicted by asterisks) include the three putative membrane spanning domains, which have been underlined.

Chromosomal localization of dry and dip

The dry gene hybridizes to a single band on polytene chro- mosomes located near the centromeric end of chromosome 3L at band position 76C-D (data not shown). No known mutations with a neuromuscular phenotype have been found to map to this region. The dry gene also appears to be single copy by hybridization to Southern blots washed at high stringency. However, on low stringency blots we could detect in each case an additional band of hybridization, sug- gesting the presence of a related gene (data not shown). In a similar analysis with the dip gene, no differences were observed in the hybridization patterns of Southern blots hybridized and washed at either high or low stringency. The dip gene also hybridizes to a single site on polytene chro- mosomes at band position 42B1-2 on the right arm of the second chromosome (data not shown). No known mutations map to this region of the second chromosome.

Developmental expression of dry and dip by northern analysis

A characteristic feature of the ryanodine and IP3 receptor genes of mammals is the very large size of their mRNAs and encoded proteins. The rabbit cardiac ryanodine recep- tor is encoded by a 17 kb mRNA, while the mRNA for the rat IP3 receptor is 10 kb in length. From the size of the mRNAs seen for the two Drosophila genes (Fig. 2), it would appear that the Drosophila proteins are also very large. To address the question of which cellular processes are likely to involve these two Ca2+-release channel genes, we initially examined their expression by northern blot analysis (Fig. 2). With the dry probe, a band of hybridiz- ation of approximately 15 kb is first seen in 12- to 18-hour embryos. This band is more intense in 18- to 24-hour embryos. The same levels of mRNA are seen in larvae and adults, but the band is apparently absent from mid-pupae. Although the dry mRNA band is visible both in RNA from adult heads and bodies, it is clearly of greater intensity in body RNA, suggesting that as in mammals this protein may be enriched in muscle tissue.

Fig. 2.

Developmental profile of dry and dip transcripts. Poly(A)+ RNA from various stages of embryonic development (first and second instar larvae, mid-pupae, adults, and adult heads and bodies) was electrophoresed and transferred to a nylon membrane. The same membrane was sequentially hybridized to probes from the dry gene, the dip gene and a ribosomal protein gene (RP49). Position of the 9.5 kb RNA marker on the gel has been indicated.

Fig. 2.

Developmental profile of dry and dip transcripts. Poly(A)+ RNA from various stages of embryonic development (first and second instar larvae, mid-pupae, adults, and adult heads and bodies) was electrophoresed and transferred to a nylon membrane. The same membrane was sequentially hybridized to probes from the dry gene, the dip gene and a ribosomal protein gene (RP49). Position of the 9.5 kb RNA marker on the gel has been indicated.

Two bands of hybridization, differing very slightly in size, are seen with the dip probe (Fig. 2). The smaller-sized band (approx. 9.7 kb) is seen only in early embryos of 0- 6 hour. A larger band of approx. 10 kb is expressed in mid- late embryos. Unlike the ryanodine receptor mRNA levels, however, the dip mRNA levels do not increase in late embryos and a low level of expression is observed through- out development in larvae, mid-pupae and adults. Equiva- lent mRNA levels are seen in adult head and body RNAs. At present, we cannot say what accounts for the size dif- ference between the two mRNAs seen for the IP3 receptor. Although they could be derived from two different genes, it is more likely that they are alternatively spliced mRNAs transcribed from the same gene, since we were unable to detect a second gene by Southern blotting (see above). In mammals, IP3 receptor transcripts are known to be alter- natively spliced in a tissue specific manner (Danoff et al., 1991).

Expression of the dry gene in embryos

The northern analysis presented in Fig. 2 suggests that both the ryanodine and IP3 receptor genes are expressed during embryonic development. Further temporal and spatial local- ization of their RNA expression was addressed by in situ hybridization. Embryonic expression of the dry gene is first detectable in the mesodermal layer of early stage-9 embryos, at approximately 5 hours of development (Fig. 3A; m). This is earlier than predicted by the devel- opmental northern blot (Fig. 2) and could be due to a dif- ference in the sensitivity of the two methods. More enhanced expression is seen in late stage-9 embryos, when segmental indentations of the mesoderm, characteristic of this stage, are clearly visible (Fig. 3B; ind). At stage 10, when the mesoderm undergoes its third post-blastodermal mitosis, expression of the dry gene becomes restricted to the more external mesodermal cell layers, i.e. the somato- pleura, which gives rise to the somatic musculature (Fig. 3C; sp). The procephalic musculature is also seen to be expressing the dry gene at stage 10 (Fig. 3C; pm). There- after, expression is maintained in each segment, in prog- enitors of the cephalic and somatic muscles (Fig. 3D,E; pm, sm), where it is localized during the final stages of embry- onic development (stage 16; Fig. 3F; sm, cm). The resolution of the staining method prevents us from defining pre- cisely which somatic/cephalic muscles (and their progeni- tors) are positive for dry expression, although the extent of the expression suggests that dry mRNA is present in all the somatic muscles. The pattern of expression is striking in that the RNA appears in the mesoderm prior to the start of muscle differentiation and is subsequently localized to progenitors of the somatic muscles and to the muscles them- selves. There is also a notable lack of detectable expression in the visceral muscles (Fig. 3E; vm).

Fig. 3.

Expression of dry mRNA during embryonic development. Figures show the hybridization pattern of a digoxigenin-labelled dry cDNA probe to wild-type embryos at various stages of development. A-Dare lateral views with the dorsal side uppermost. E and Fare mid-dorsal and dorsal views, respectively, In all cases, anterior is to the left. (A) Early stage-9 embryn Arrow indicates faint hybridization to the mesodermal layer (m), (R) T ,ate stage-9 embryo. Arrow indicates segmental indentations of the mesoderm (ind). (C) Stage-10 embryo with hybridization to the procephalic mesoderm (pm) and the somatopleural layer of the mesoderm (sp). (D) Late stage-12 embryo with hybridization to segmentally located somatic muscle precursor cells (sm) and precursors of the cephalic muscles (pm). (E) Stage-14 embryo. Pattern of hybridization is similar to that in the em hryo Rhown in D. No hyhridi zation iR detectable among cellR of the viRceral meRoderm (vm). (F) Stage-16 embryo in which all hybridization is localized to somatic (sm) and cephalic muscles (cm). Scale bar, 40 μm. Embryo stages have been defined according to WieRchauR and NuRRlein-Volhard (1986). Further analyRiR of RtageR waR according to Campos-Ortega and Hartenstein (1985).

Fig. 3.

Expression of dry mRNA during embryonic development. Figures show the hybridization pattern of a digoxigenin-labelled dry cDNA probe to wild-type embryos at various stages of development. A-Dare lateral views with the dorsal side uppermost. E and Fare mid-dorsal and dorsal views, respectively, In all cases, anterior is to the left. (A) Early stage-9 embryn Arrow indicates faint hybridization to the mesodermal layer (m), (R) T ,ate stage-9 embryo. Arrow indicates segmental indentations of the mesoderm (ind). (C) Stage-10 embryo with hybridization to the procephalic mesoderm (pm) and the somatopleural layer of the mesoderm (sp). (D) Late stage-12 embryo with hybridization to segmentally located somatic muscle precursor cells (sm) and precursors of the cephalic muscles (pm). (E) Stage-14 embryo. Pattern of hybridization is similar to that in the em hryo Rhown in D. No hyhridi zation iR detectable among cellR of the viRceral meRoderm (vm). (F) Stage-16 embryo in which all hybridization is localized to somatic (sm) and cephalic muscles (cm). Scale bar, 40 μm. Embryo stages have been defined according to WieRchauR and NuRRlein-Volhard (1986). Further analyRiR of RtageR waR according to Campos-Ortega and Hartenstein (1985).

Embryonic expression of the dip gene

As seen by northern analysis (Fig. 2), dip transcripts are present in 0- to 6-hour early embryos. However, we were unable to detect this expression by in situ hybridization, possibly because it is low and may not be localized. The first distinct hybridization is seen at approximately 10.5 hours (in early stage-13 embryos) as a segmental pattern in cells of the lateral epidermis lying along the intersegmen- tal furrows and in posterior epidermal cells (data not shown). In late stage-13 embryos, this expression intensi- fies and extends to the head region within cells of the gnathal buds, which lie ventral to the stomodeal opening (Fig. 4A; gb). The spatiotemporal appearance of these cells in the head region suggests that they are likely to be prog enitors of anterior sense organs (Hartenstein, 1988). In stage-14 embryos, the anterior expression extends dorsally to cells in the clypeolabrum while the lateral epidermal expression decreases to barely detectable levels (Fig. 4B; cl). By stage 15, all detectable expression of the dip gene is localized to the anterior head region, presumably in pri- mordia of anterior sense organs (Fig. 4C; pl, cl, gb). As development progresses (stage 16), some of these sense organs can be identified as two pairs of dorsal structures which lie along the pharynx walls (Fig. 4D,E; do) and a more lateroventral pair, which is likely to be the labial organ (Fig. 4D,E; lo). Finally, at stage 17 only the labial organ expresses this gene (Fig. 4F; lo).

Fig. 4.

Expression of dip mRNA during embryonic development. Wild-type embryos were hybridized to a digoxigenin-labelled dip cDNA probe. A-D and Fare lateral views. E shows the anterior end of the same embryo at three different levels of focus going from ventral to dorsal and left to right. Lateral orientations and the scale are as in Fig. 3. (A) Late stage-13 embryo with hybridization to the gnathal buds (gb) in the head region. Segmental expression in lateral epidermal cells (le) has been indicated in one segment. (R) Stage-14 embryo. Anterior expression extends dorsally to cells in the clypeolabrum (cl). (C) Stage-IS embryo with expression localized to the head region in the procephalic lobe (pl), clypeolabrum (cl) and gnathal buds (gb). (D) Stage-16 embryo expressing dip in specific differentiating anterior sense organs in the dorsal region of the head (do) and more ventrally in the labial organ (lo). (E) Horizontal view of the head region of a stage-16 embryo showing, from left to right, the labial organ (ventral focus) and two dorsal organs (mid-dorsal and dorsal foci). Arrows indicate the pair of sense organs in focus. (F) Stage- 17 embryo with hybridization detectable only to the labial organ (lo).

Fig. 4.

Expression of dip mRNA during embryonic development. Wild-type embryos were hybridized to a digoxigenin-labelled dip cDNA probe. A-D and Fare lateral views. E shows the anterior end of the same embryo at three different levels of focus going from ventral to dorsal and left to right. Lateral orientations and the scale are as in Fig. 3. (A) Late stage-13 embryo with hybridization to the gnathal buds (gb) in the head region. Segmental expression in lateral epidermal cells (le) has been indicated in one segment. (R) Stage-14 embryo. Anterior expression extends dorsally to cells in the clypeolabrum (cl). (C) Stage-IS embryo with expression localized to the head region in the procephalic lobe (pl), clypeolabrum (cl) and gnathal buds (gb). (D) Stage-16 embryo expressing dip in specific differentiating anterior sense organs in the dorsal region of the head (do) and more ventrally in the labial organ (lo). (E) Horizontal view of the head region of a stage-16 embryo showing, from left to right, the labial organ (ventral focus) and two dorsal organs (mid-dorsal and dorsal foci). Arrows indicate the pair of sense organs in focus. (F) Stage- 17 embryo with hybridization detectable only to the labial organ (lo).

Expression of dry and dip in adults

Expression of the dry and dip genes in structures of the adult head and thorax was investigated by two methods. In situ hybridization to frozen cryostat sections revealed that the dry gene is expressed maximally in tubular muscles of the thorax, namely, the ‘jump’ muscle or the ter- gotrochanteral muscle and the leg muscles (Fig. 5A-C; ttm, lm). In addition, expression is also seen in the cortex of the ventral ganglion and the brain (Fig. 5B,D; vg, bc). Some expression was also observed in muscles of the head and proboscis (data not shown). However, there is no discernible hybridization to any of the fibrillar or indirect flight muscles. These observations are significant, since the tubu- lar and fibrillar thoracic muscles are known to differ in their ultrastructure and biochemistry of contraction (Crossley, 1978).

Fig. 5.

Adult expression of dry in sections of the thorax, legs and head. Sections were hybridized to digoxigenin- labelled probes. (A) Sagittal section at a mid-lateral position through the thorax. Anterior is to the top and the dorsal side is towards the left. Hybridization is to the tergotrochanteral muscle (ttm) and some of the leg muscles (lm). (B) Horizontal section through the thorax including the ventral ganglion. Hybridization to the cortical regions of the ventral ganglion (vg) has been indicated. Anterior is towards the top. (C) Sagittal section through a part of a leg showing hybridization to leg muscles (lm). (D) Sagittal section through a lateral part of the adult head with hybridization to the cortical region of the central brain (bc). The orientation is the same as in Fig. 5A. Scale bars, 80 μm.

Fig. 5.

Adult expression of dry in sections of the thorax, legs and head. Sections were hybridized to digoxigenin- labelled probes. (A) Sagittal section at a mid-lateral position through the thorax. Anterior is to the top and the dorsal side is towards the left. Hybridization is to the tergotrochanteral muscle (ttm) and some of the leg muscles (lm). (B) Horizontal section through the thorax including the ventral ganglion. Hybridization to the cortical regions of the ventral ganglion (vg) has been indicated. Anterior is towards the top. (C) Sagittal section through a part of a leg showing hybridization to leg muscles (lm). (D) Sagittal section through a lateral part of the adult head with hybridization to the cortical region of the central brain (bc). The orientation is the same as in Fig. 5A. Scale bars, 80 μm.

Expression of the dip gene in adults occurs in the cortex of the central brain and the antennae (Fig. 6A; bc, ant). In sections from the same animal, no hybridization was detectable to any region of the eye (Fig. 6B). This was sur- prising since there is genetic evidence linking IP3 with the visual transduction pathway in Drosophila (Bloomquist et al., 1988). We therefore used a second method, quantita- tive PCR (RT-QPCR), to compare the abundance of dry and dip, as well as the control gene RP49 (encoding a Drosophila ribosomal protein), in a few relevant adult struc- tures (heads, antennae, legs and eyes).

Fig. 6.

Expression of the dip gene in sections of the adult head. Hybridization to digoxigenin-labelled probes was as described in the legend to Fig. 4. Orientation of the sections is as stated for Fig. 5A. (A) Sagittal section with hybridization to the third segment of the antenna (ant) and to the brain cortex (bc). (B) Sagittal section more lateral to that in A. The compound eye has been indicated with an arrow. Hybridization to the brain cortex (bc) is detectable. Scale bars, 60 μm.

Fig. 6.

Expression of the dip gene in sections of the adult head. Hybridization to digoxigenin-labelled probes was as described in the legend to Fig. 4. Orientation of the sections is as stated for Fig. 5A. (A) Sagittal section with hybridization to the third segment of the antenna (ant) and to the brain cortex (bc). (B) Sagittal section more lateral to that in A. The compound eye has been indicated with an arrow. Hybridization to the brain cortex (bc) is detectable. Scale bars, 60 μm.

The RT-QPCR involves the addition to the PCR reaction of a known amount of synthetic RNA (the spike) that can be amplified by reverse transcriptase and PCR in the same reaction and with the same primers as the transcript of inter- est. The sequence between the primers is also identical to the tissue-derived transcript except for a small deletion or insertion that allows resolution of the two amplified DNA fragments by gel electrophoresis. The ratio of the two bands, when between approximately 0.2 and 5.0, estimates the amount of transcript in the RNA preparation. More importantly, the ratio of the two bands, compared between RNA derived from two tissues (i.e., the ratio of the ratios), measures the relative abundance of the transcript of inter- est between the two tissues (see Materials and methods). For example, the dip transcript is twice as abundant in RNA from heads as in RNA from legs, as the ratio of the dip RNA band to its spike band in RNA from heads is twice that of RNA from legs (Fig. 7A).

Fig. 7.

RT-QPCR detection of the dry and dip transcripts in RNA from heads, eyes, antennae, and legs. (A) Total RNA from heads and legs was assayed for the dry transcript after the addition of the dry spike RNA. RNA was prepared and the reactions carried out as described in Materials and methods. Lane 1, no RNA; lane 2, leg RNA plus 100 fg (femtograms) of spike; lane 3, leg RNA plus 33 fg of spike; lane 4, head RNA plus 100 fg of spike; lane 5, head RNA plus 33 fg of spike. The dry-derived band and the spike-derived band are indicated. (B) Normalized levels of dip expression in the four tissues. (C) Normalized levels of dry expression in the four tissues.

Fig. 7.

RT-QPCR detection of the dry and dip transcripts in RNA from heads, eyes, antennae, and legs. (A) Total RNA from heads and legs was assayed for the dry transcript after the addition of the dry spike RNA. RNA was prepared and the reactions carried out as described in Materials and methods. Lane 1, no RNA; lane 2, leg RNA plus 100 fg (femtograms) of spike; lane 3, leg RNA plus 33 fg of spike; lane 4, head RNA plus 100 fg of spike; lane 5, head RNA plus 33 fg of spike. The dry-derived band and the spike-derived band are indicated. (B) Normalized levels of dip expression in the four tissues. (C) Normalized levels of dry expression in the four tissues.

A summary of experiments of this nature shows that the dip transcript is detectable in all 4 tissues examined (Fig. 7B), suggesting that the dip gene is expressed in eyes, as well as in heads, legs, and antennae. The most striking quantitative feature is the four-fold difference in abundance between RNA from antennae and RNA from heads or eyes as well as an even greater eight-fold difference in abun- dance between antennae and legs. The data are consistent with the in situ hybridization experiments that detected intense signals in antennae. The same approach shows that the dry transcript is also present in eyes, heads, legs, and antennae (Fig. 7C). In this case there is an approximately four-fold difference in abundance between RNA from legs and RNA from heads, eyes, or antennae (which are all approximately the same), also consistent with the in situ hybridization experiments shown above. Antennae and legs were purified on two separate occasions and RNA from both preparations gave essentially the same results, i.e., maximal abundance for dry RNA in legs and for dip RNA in antennae.

Sequence analysis presented here for a portion of two Drosophila Ca2+-release channel proteins indicates a high level of conservation with the corresponding mammalian proteins. For both mammalian proteins, the C-terminal end forms the membrane spanning domain required for tetramerization and assembly into a functional Ca2+ chan- nel (Mignery and Sudhof, 1990; Catterall, 1991); it seems likely that the same structural organization occurs in Drosophila. Based on mRNA size, the remaining non- sequenced portions of these Drosophila genes appear com- parable to the corresponding regions of the mammalian genes, suggesting that they are likely to encode very large cytoplasmic domains, like those shown to regulate the open- ing and closing of the mammalian channels (Mignery and Sudhof, 1990; Caterall, 1991). Although further sequence analysis is required to determine the extent to which the cytoplasmic domains are conserved, partial information for this region of the dip gene has been published and is con- sistent with a high level of conservation throughout most of the cytoplasmic domain (Miyawaki et al., 1991). The characterization of complete cDNA clones will be required to address the possibility of multiple dry or dip receptor proteins, generated perhaps by differential splicing or from multiple genes (see below).

The pattern of expression seen for the dry gene during embryonic development raises some interesting possibili- ties regarding the function of this gene in muscle development. Its expression in the mesoderm starts well before the expression of other muscle-specific genes such as the myosin heavy chain (Michelson et al., 1990). In fact, the appearance of dry mRNA appears to be synchronous with that of a muscle determining gene nau, which is the myoD homolog in Drosophila (Michelson et al., 1990). One reason for this early expression may be that the Dry pro- tein is required for mesodermal development, perhaps in cell-signaling processes that precede commitment to the somatic muscle lineage. This idea is supported by the finding that expression of the ryanodine receptor also occurs in very early differentiating muscle cells in a vertebrate tissue culture model (Airey et al., 1991). It is also of interest that dry expression is detectable only in somatic muscles of late- stage embryos and in tubular muscles of the adult. Both of these muscle types are rich in sarcoplasmic reticulum (SR), while the larval visceral muscles and the adult fibrillar muscles have a very scanty SR (Crossley, 1978). There is thus a correlation between muscles that detectably express dry and those with an extensive SR, suggesting that the Dry protein is located in Drosophila muscle SR and therefore functions as the excitation-contraction coupling protein of the neuromuscular junction, as in the case of vertebrate muscles (Catterall, 1991).

The correlation between the amount of SR and the level of dry RNA raises the possibility that dry is being expressed in other muscles at levels commensurate with their low levels of SR. Alternatively, as in mammals, there may be other ryanodine receptor genes in Drosophila, and it is pos- sible that these are expressed differentially in physiologi- cally different muscle types (Otsu et al., 1990). This pos- sibility is supported by the presence of additional bands of hybridization on a low-stringency Southern blot with the dry probe (data not shown).

Expression of dry in adult neuronal tissues, observed by both in situ hybridization and the PCR method, suggests that, as in vertebrates, this channel also serves as yet poorly understood neuronal functions (McPherson et al., 1991). It should be possible to elucidate these functions in Drosophila by obtaining and analyzing mutants in this gene, as well as by localizing the protein within the brain.

Expression of the Drosophila IP3 receptor gene suggests that it functions during sensory transduction in both larvae and adults. Initial embryonic expression of dip in lateral and posterior epidermal cells occurs at about the same time that epidermal cells undergo mitosis and move inward to form sensory precursors. However, expression appears more epidermal than subepidermal and is also fairly wide- spread, indicating that it may not be restricted to or even occurring within sensory precursor cells. Since this expression is also transient, it is difficult to speculate on its role. Expression of dip in the embryonic head region can be followed more readily within some of the presumptive sensory precursor cells and the anterior sense organs. The functions of the three pairs of sense organs, within which it is detected at stage 16, are not well understood, though one pair of dorsal structures, seen in Fig. 4E, may be the equivalent of the ‘stemmata’ described in other fly species. The stemmata are light-sensing larval organs that have also been described for Drosophila (Steller et al., 1987). This expression pattern does not indicate whether the IP3 receptor is required in these cells for their development, as suggested by the transient expression of its mRNA, or for their sensory functions, or both. We have begun to address this question by looking for the presence of the Dip protein in late embryos and larvae using anti-Dip antibodies. The results to date have been inconclusive (data not shown).

Spatial expression of the dip gene in adult Drosophila has been analyzed by the usual procedure of in situ hybrid- ization as well as by PCR with RNA isolated from dis- sected tissues. The latter approach was taken in part to verify that little or no expression is detectable in eyes. Con- trary to expectation, the RT-QPCR assay gave rise to sim- ilar dip RNA levels in eyes and in heads. We cannot com- pletely rule out the possibility that the eyes were contaminated with a small amount of brain tissue or that the PCR product was derived from genomic DNA (but see Materials and methods). It is also possible that another IP3 receptor gene exists in Drosophila and that eye mRNA derived from this gene gave rise to the PCR signal; as noted above, attempts to identify such a gene by low stringency Southern blots have been unsuccessful (data not shown). More likely, in our view, is the possibility that the in situ hybridization to eye tissue is relatively insensitive, perhaps because of some peculiarity of this mRNA’s localization. The presence of dip gene expression in eyes is expected from the important role of phospholipase C in visual pho- totransduction in Drosophila (e.g., Yoshioka et al., 1985) and is also supported by our recent finding that antibodies to the Dip protein stain regions of the photoreceptor cells as well as cortical regions of the brain and antennae (data not shown).

Expression of dip in the adult antennae suggests that the IP3 receptor could be required during olfaction in Drosophila. A role for IP3 as the second messenger during olfactory sensory transduction has been suggested by vari- ous experiments in both vertebrates and invertebrates (Breer et al., 1990; Reed, 1992). Interestingly, the IP3 receptor may be involved primarily in the development and/or func- tion of sensory structures in larvae, since its RNA is not detected in either the brain or ventral nerve cord, even at very late stages of embryonic development (Fig. 4E,F). In contrast, adults express dip RNA in the central brain and ventral ganglion (data not shown), as well as in sensory organs like antennae, legs and eyes. This difference in expression may reflect the greater functional complexity of the adult brain as compared to the larval brain. The use of antibodies and mutants should help resolve these issues, especially those related to olfaction.

We are grateful to Kalpana White, Veronica Rodrigues and K. Vijay Raghavan for help in interpreting the in situ data. A. Mehta and I. Edery were instrumental in setting up the RT-QPCR assay, and A. Mehta performed the experiments shown in Fig. 7. The work was supported by a grant from the National Institutes of Health to M.R. (GM33205).

Airey
,
J. A.
,
Baring
,
M. D.
and
Sutko
,
J. L.
(
1991
).
Ryanodine receptor protein is expressed during differentiation in the muscle cell lines BC3H1 and C2C12
.
Dev. Biol
.
148
,
365
374
.
Berridge
,
M. J.
and
Irvine
,
R. F.
(
1984
).
Inositol triphosphate, a novel second messenger in cellular signal transduction
.
Nature
312
,
315
320
.
Berridge
,
M. J.
and
Irvine
,
R. F.
(
1989
).
Inositol phosphates and cell signalling
.
Nature
341
,
197
205
.
Bloomquist
,
B. T.
,
Shortridge
,
R. D.
,
Schneuwly
,
S.
,
Perdew
,
M.
,
Montell
,
C.
,
Steller
,
H.
,
Rubin
,
G. M.
and
Pak
,
W. L.
(
1988
).
Isolation of a putative phospholipase C gene of Drosophila, norpA, and its role in phototransduction
.
Cell
54
,
723
733
.
Breer
,
H.
,
Boekhoff
,
I.
and
Tareilus
,
E.
(
1990
).
Rapid kinetics of second messenger formation in olfactory transduction
.
Nature
345
,
65
68
.
Campos-Ortega
,
J. A.
and
Hartenstein
,
V.
(
1985
).
The Embryonic Development of Drosophila melanogaster
.
Springer-Verlag
,
New York
.
Catterall
,
C.
(
1991
).
Excitation-contraction coupling in vertebrate skeletal muscle. A tale of two Ca++ channels
.
Cell
64
,
871
874
.
Crossley
,
C.
(
1978
).
The morphology and development of the Drosophila muscular system
.
In The Genetics and Biology of Drosophila
2b
(eds.
M.
Ashburner
and
T.R.F.
Wright
), pp.
499
560
.
New York
:
Academic Press
.
Danoff
,
S. K.
,
Ferris
,
C. D.
,
Donath
,
C.
,
Fischer
,
G. A.
,
Munemitsu
,
S.
,
Ullrich
,
A.
,
Snyder
,
S. H.
and
Ross
,
C. A.
(
1991
).
Inositol 1,4,5-triphosphate receptors: Distinct neuronal and non-neuronal forms derived by alternative splicing differ in phosphorylation
.
Proc. Natl. Acad. Sci. USA
88
,
2951
2955
.
Devereux
,
J.
,
Haeberli
,
P.
and
Smithies
,
O.
(
1984
).
A comprehensive set of sequence analysis programs for the VAX
.
Nucl. Acids Res
.
12
,
387
395
.
Furuichi
,
T.
,
Yoshikawa
,
S.
,
Miyawaki
,
A.
,
Wada
,
K.
,
Maeda
,
N.
and
Mikoshiba
,
K.
(
1989
).
Primary structure and functional expression of the inositol 1,4,5-triphosphate-binding protein P400
.
Nature
342
,
32
38
.
Gilliland
,
G.
,
Perrin
,
S.
,
Blanchard
,
K.
and
Bunn
,
H. F.
(
1990
).
Analysis of cytokine mRNA and DNA: Detection and quantitation by competitive polymerase chain reaction
.
Proc. Natl. Acad. Sci. USA
87
,
2725
2729
.
Hartenstein
,
V.
(
1988
).
Development of Drosophila larval sensory organs: spatiotemporal pattern of sensory neurons, peripheral axonal pathways and sensilla differentiation
.
Development
102
,
869
886
.
Hwang
,
P. M.
,
Verma
,
A.
,
Bredt
,
D. S.
and
Snyder
,
S. H.
(
1990
).
Localization of phosphatidylinositol signaling components in rat taste cells: Role in bitter taste transduction
.
Proc. Natl. Acad. Sci. USA
87
,
7395
7399
.
Majerus
,
P. W.
,
Wilson
,
D. B.
,
Connolly
,
T. M.
,
Bross
,
T. E.
and
Neufeld
,
E. J.
(
1985
).
Phosphoinositide turnover provides a link in stimulus-response coupling
.
Trends Biochem. Sci
.
10
,
168
171
.
Matsumoto
,
H.
,
O’Tousa
,
J. E.
and
Pak
,
W. L.
(
1982
).
Light-induced modification of Drosophila retinal polypeptides in vivo
.
Science
217
,
839
841
.
McPherson
,
P. S.
,
Kim
,
Y-K.
,
Vladivia
,
H.
,
Knudson
,
C. M.
,
Takekura
,
H.
,
Franzini-Armstrong
,
C.
,
Coronado
,
R.
and
Campbell
,
K. P.
(
1991
).
The brain ryanodine receptor: A caffeine-sensitive calcium release channel
.
Neuron
7
,
17
25
.
Michelson
,
A. M.
,
Abmayr
,
S. M.
,
Bate
,
C. M.
,
Arias
,
A. M.
and
Maniatis
,
T.
(
1990
).
Expression of a MyoD family member prefigures muscle pattern in Drosophila embryos
.
Genes Dev
.
4
,
2086
2097
.
Mignery
,
G. A.
,
Newton
,
C. I.
,
Archer
III,
B. T.
and
Sudhof
,
T. C.
(
1990
).
Structure and expression of the rat inositol 1,4,5-triphosphate receptor
.
J. Biol. Chem
.
265
,
12679
12685
.
Mignery
,
G. A.
and
Sudhof
,
T. C.
(
1990
).
The ligand binding site and transduction mechanism in the inositol-1,4,5-triphosphate receptor
.
EMBO J
.
9
,
3893
3898
.
Miyawaki
,
A.
,
Furuichi
,
T.
,
Ryou
,
Y.
,
Yoshikawa
,
S.
,
Nakagawa
,
T.
,
Saitoh
,
T.
and
Mikoshiba
,
K.
(
1991
).
Structure-function relationships of the mouse inositol 1,4,5-triphosphate receptor
.
Proc. Natl. Acad. Sci. USA
88
,
4911
4915
.
Nighorn
,
A.
,
Healy
,
M. J.
and
Davis
,
R. L.
(
1991
).
The cyclic AMP phosphodiesterase encoded by the Drosophila dunce gene is concentrated in the mushroom body neuropil
.
Neuron
6
,
455
467
.
Otsu
,
K.
,
Huntington
,
F. W.
,
Khanna
,
V. K.
,
Zorzato
,
F.
,
Green
,
N. M.
and
MacLennan
,
D. H.
(
1990
).
Molecular cloning of cDNA encoding the Ca 2+ release channel (ryanodine receptor) of rabbit cardiac muscle sarcoplasmic reticulum
.
J. Biol. Chem
.
265
,
13472
13483
.
Palazzolo
,
M. J.
,
Hamilton
,
B. A.
,
Ding
,
D.
,
Martin
,
C. H.
,
Mead
,
D. A.
,
Mierenderf
,
R. C.
,
Raghavan
,
K. V.
,
Meyerowitz
,
E. M.
and
Lipshitz
,
H. D.
(
1990
).
Phage lambda cDNA cloning vectors for subtractive hybridization, fusion-protein synthesis and Cre-loxP automatic subcloning
.
Gene
88
,
25
36
.
Payne
,
R.
,
Walz
,
B.
,
Levy
,
S.
and
Fein
,
A.
(
1988
).
The localization of calcium release by inositol triphosphate in Limulus photoreceptors and its control by negative feedback
.
Phil. Trans. R. Soc. Lond. B
320
,
359
379
.
Reed
,
R. R.
(
1992
).
Signaling pathways in odorant detection
.
Neuron
8
,
205
209
.
Saiki
,
R. K.
,
Scharf
,
S.
,
Faloona
,
F.
,
Mullis
,
K. B.
,
Horn
,
G. T.
,
Ehrlich
,
H. A.
and
Arnheim
,
N.
(
1985
).
Enzymatic amplification of β-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anaemia
.
Science
230
,
1350
1354
.
Steller
,
H.
,
Fischbach
,
K. F.
and
Rubin
,
G. M.
(
1987
).
Disconnected: A locus required for neuronal pathway formation in the visual system of Drosophila
.
Cell
50
,
1139
1153
.
Tautz
,
D.
and
Pfeifle
,
C.
(
1989
).
A non-radioactive in situ hybridization method for the localization of specific RNAs in Drosophila embryos reveals translational control of the segmental gene hunchback
.
Chromosoma
98
,
81
85
.
Tsein
,
R. W.
and
Tsein
,
R. Y.
(
1990
).
Calcium channels, stores and oscillations
.
In Annual Review of Cell Biology
, vol.
6
. (ed.
G.E.
Palade
), pp.
715
760
.
California
:
Annual Reviews, Inc
.
Yoshioka
,
T.
,
Inoue
,
H.
and
Hotta
,
Y.
(
1985
).
Absence of phosphatidylinositol phosphodiesterase in the head of a Drosophila visual mutant norpA (no receptor potential A)
.
J. Biochem. (Tokyo)
97
,
1251
1254
.
Wieschaus
,
E.
and
Nusslein-Volhard
,
C.
(
1986
).
Looking at embryos
.
In Drosophila: A Practical approach
(ed.
D.B.
Roberts
), pp.
199
227
.
Oxford
:
IRL Press
.

A recent publication (Journal of Biological Chemistry, 267, 16613-16619; August 15, 1992) reports the molecular cloning and characterization of the Drosophila IP3 recep- tor gene. The two sets of results are mostly in agreement, with the exception of the localization of the gene by in situ hybridization to polytene chromosomes. As a consequence, we repeated this experiment and found that our original assignment (42B1-B2) was incorrect. The correct position (83A5-A9) is as reported in the J. Biol. Chem. publication.