The Interactive Fly

Zygotically transcribed genes

Odorant-binding proteins

A large family of putative odorant-binding protein (OBP) genes has been identified in the genome of Drosophila. Some of these genes are present in large clusters in the genome. Most members are expressed in various taste organs, including gustatory sensilla in the labellum, the pharyngeal labral sense organ, dorsal and ventral cibarial organs, as well as taste bristles located on the wings and tarsi. Some of the gustatory OBPs are expressed exclusively in taste organs, but most are expressed in both olfactory and gustatory sensilla. Multiple binding proteins can be coexpressed in the same gustatory sensillum. Cells in the tarsi that express OBPs are required for normal chemosensation mediated through the leg, since ablation of these cells dramatically reduces the sensitivity of the proboscis extension reflex to sucrose. OBP genes expressed in the pharyngeal taste sensilla are still expressed in the poxneuro genetic background, while OBPs expressed in the labellum are not. These findings support a broad role for members of the OBP family in gustation and olfaction and suggest that poxneuro is required for cell fate determination of labellar but not pharyngeal taste organs (Galindo, 2001).

Seven members of the Drosophila OBP family have been previously identified. With the exception of PBP2, which is secreted by epithelial cells, each is expressed in the sensillum lymph of a subset of olfactory sensilla. How large is the OBP family, and is OBP function restricted to olfactory sensilla? To address these issues, attempts were made to identify all OBP genes encoded in the Drosophila genome. 28 new candidate OBP genes were identified in the Drosophila genome project, making a total of 35 members of this gene family present in the genome of this organism. Each putative member was identified with the tBLASTn algorithm using the previously identified members of the Drosophila OBP family as probes to identify related genes in the Drosophila genome sequence. Candidate genes were screened for features characteristic of the invertebrate OBP family, including low molecular weight (13-16 kD), a predicted signal sequence, and the presence of six conserved cysteines with the invariant spacing between cysteines 2 and 3 (three residues) and cysteines 5 and 6 (eight residues) that define this family. Putative genes meeting these criteria typically are predicted to contain one or two introns with conserved splicing consensus sequences and consensus translation start sequences. Each putative OBP member is named with a number representing its chromosomal location and a letter that designates its order relative to other OBP genes at that position. It is worth noting that 35 OBP genes is a minimum number, since this analysis would miss genes with introns located between conserved cysteine motifs. It is possible, therefore, that additional OBP genes may be encoded in the Drosophila genome that were not identified here (Galindo, 2001).

The genes encoding the OBP family in Drosophila are not randomly distributed in the genome and are often found in large clusters. For example, 14 OBP genes are clustered within 825 kb located at position 56-57 on the second chromosome. A second cluster of 7 genes is located at position 83 on the third chromosome. The localization of so many members of the OBP family in a relatively discrete region of the genome suggests that these genes may have arisen by tandem duplication. Tandem duplication has been suggested to account for the two closely related genes, OBPs 83a and 83b. Several of the OBP genes are tightly linked. For example, OBP56a and OBP56b are located within 2 kb of each other, but are transcribed in opposite directions. Despite this close proximity, these genes appear to have strikingly different expression patterns. This situation is similar to that observed for clustered putative odorant receptor genes, where closely linked, chemosensory-specific genes are expressed in different olfactory organs (Galindo, 2001).

The 35 members of the Drosophila OBP family range from 9.2% to 62.6% amino acid identity, demonstrating the highly divergent nature of this protein family. Indeed this protein family is among the most diverse described. Most frequently there is a single splice immediately downstream from the DNA encoding the signal sequence. There is little or no amino acid sequence conservation among the OBP members at splice junctions. Several of the OBP genes have unique splicing patterns. Three OBP genes have no splices within the coding sequence (OBP28a, 69a, and 83d). This surprising lack of splicing conservation and low sequence similarity suggests that if gene duplication is the mechanism responsible for the large number of these genes, they evolve rapidly. All members are predicted to encode proteins with six alpha-helical domains joined by loops that vary in length. Only the six cysteines are completely conserved among all 35 members. The limited sequence similarity among members is often clustered near the cysteines. For example, all members have a hydrophobic amino acid following cysteines two, three, and six. Overall, however, there is little homology in this OBP family. This diversity is consistent with the notion that different OBPs interact with distinct sets of chemical ligands (Galindo, 2001).

To determine the spatial and temporal expression pattern of each putative OBP gene, several kilobases of upstream regulatory sequence for each OBP gene were fused to a reporter gene encoding a nuclear-localized ß-galactosidase. Transgenic flies were generated that are expected to express the reporter gene in the same cells that normally express that particular OBP. The advantages of this approach over in situ hybridization analysis are that it can detect expression in tissues not amenable to in situ hybridization (like the wing and legs) and it has little background and excellent sensitivity. Similar fusions have been shown to precisely reproduce wild-type gene expression (Galindo, 2001).

Transgenic flies carrying reporter constructs fused to each OBP promoter were generated and the flies were stained for ß-galactosidase activity. Surprisingly, in addition to the expected olfactory expression, a wide variety of gustatory organs were labeled in many lines. On the basis of reporter gene expression, the members of the OBP family can be classified in one of five classes. Class 1 is composed of putative OBP genes (nine members) expressed exclusively in subsets of chemosensory sensilla. Class 2 genes (four members) are expressed exclusively in gustatory sensilla. Class 3 genes (nine members) are expressed in subsets of olfactory and gustatory sensilla. Class 4 (five members) OBP promoters drive LacZ expression in broad areas that include regions that do not contain chemosensory organs. These genes may be functionally related to PB-PRP2, a gene expressed by epithelial cells that may function as a scavenger protein. Class 5 (seven members) includes OBP genes in which no LacZ expression was detectable. These genes may be pseudogenes, or the promoter fragments used may lack essential regulatory elements required for expression. One putative OBP gene, OBP83e, was not analyzed because no appropriate initiation methionine could be identified. This gene may have a large 5' intron or may be a pseudogene (Galindo, 2001).

Surprisingly, only three new class 1 OBP genes were identified -- OBP19a, 57a, and 99a. OBP19a is expressed exclusively in a subset of chemosensory sensilla on the third antennal segment. For OBP19a, a bilaterally symmetric pattern of LacZ staining is visible in the third antennal segment, but not the maxillary palps or larval chemosensory organs. This expression pattern is consistent with previously reported members of the OBP family that are also expressed in subsets of olfactory sensilla. OBP57a and OBP99b are expressed in subsets of sensilla in both olfactory organs, the maxillary palps, and third antennal segments (Galindo, 2001).

Class 3 genes are expressed in both olfactory and gustatory organs. Nine members of this class were identified. For members of this class, LacZ expression is primarily restricted to the olfactory organs, but the genes are also expressed in at least one gustatory organ. Transgenic flies expressing LacZ under control of OBP56d and OBP57c are expressed in all olfactory sensilla, including all sensilla on the antenna and maxillary palp. The expression of these OBPs in the antenna is unique because all other previously reported members are expressed in subsets of sensilla. These lines are not identical, however, since OBP56d is also expressed in the wing and tarsal gustatory sensilla and dorsal organ, and OBP56c is expressed in the wing and the larval olfactory organ, the dorsal organ (Galindo, 2001).

A gustatory-specific, class 2 OBP, OBP19c, is expressed exclusively in six cells, including two cells in the labral sense organ (LSO). The LSO consists of nine sensilla that sample the lumen of the pharynx just distal to the oral opening. Three of the LSO sensilla contain chemosensory neurons; the other six are purely mechanosensory. OBP19c is expressed strongly in a single pair of bilaterally symmetric cells in the LSO and weakly in a single cell in each ventral cibarial sense organ (VCSO) and dorsal cibarial sense organ (DCSO) in the adult. The larval dorsal organ also stains for LacZ in these transgenic flies (Galindo, 2001).

OBP56b is also expressed exclusively in the pharyngeal gustatory organs, including two cells in the LSO and two cells in the DCSO. This expression pattern is very similar to OBP19c in the LSO. These expression patterns are consistent with expression of these OBPs in support cells of gustatory sensilla, although these cells cannot be identified precisely. In Drosophila larvae, volatile odorants are detected by neurons that reside in the dorsal organ, while gustatory responses appear to be mediated by neurons in the terminal organ, in chemosensory neurons located in the ventral pits present on each thoracic hemisegment, and in some of the neurons in the dorsal organ. Both the terminal organ and dorsal organs express LacZ in OBP56b-nlsLacZ transgenic flies (Galindo, 2001).

OBP56g is expressed exclusively in the labellum in the adult. The two outer rows of gustatory bristles are LacZ positive. No other chemosensory organs are labeled in these transgenic flies. The promoter for OBP56h expresses LacZ in approximately five sensilla on each third antennal segment, in the pharyngeal organs and in the dorsal organ, the terminal organ, and the ventral pits of the third instar larvae. This OBP, therefore, may function in both olfactory and gustatory systems (Galindo, 2001).

With the exception of the vaginal plate chemosensory sensilla, members of the OBP family in all chemosensory organs have been identified. Is only one OBP expressed in each gustatory sensillum or can multiple OBPs be coexpressed? Several transgenic lines driving LacZ with different OBP promoters appear to have overlapping expression patterns (Galindo, 2001).

The promoters for OBP57d and 57e drive LacZ expression exclusively in four cells of the tarsi associated with curved chemosensory bristles on each of the six legs in flies expressing nuclear LacZ by the promoter of OBP57d and 57e, respectively. The expression patterns of OBP57d and 57e are also consistent with expression in support cells associated with gustatory sensilla. Interestingly, these genes are located <1 kb apart from each other at the end of the 56-57 gene cluster, have significant amino acid homology to each other in the C-terminal half, and may represent a relatively recent gene duplication event. Data indicate that OBP57d and 57e are coexpressed in the same cells in the tarsi (Galindo, 2001).

The promoters for three OBP genes, OBP56b, OBP56h, and OBP19c, drive LacZ expression in single pairs of cells in the pharyngeal LSO. Are any of the OBPs expressed in the LSO coexpressed in the same cells? OBP56h or 19c are coexpressed in the same cells and OBP56h is expressed in neighboring cells that do not express OBP56b (Galindo, 2001).

Transgenic flies expressing the pro-apoptotic protein Grim exclusively in the OBP57e-positive cells in the tarsi were used to evaluate the biological importance of these cells in gustation. Wild-type and pOBP57eGal4-VP16;UAS Grim-expressing flies were tested for their ability to detect sucrose applied to the tarsi using the proboscis extension reflex (PER). Normally when sucrose is applied to the tarsi, the fly extends the proboscis to feed. Expressing Grim in the cells expressing OBP57d and 57e results in a dramatic loss of sensitivity to sucrose as determined by this assay. Wild-type control flies have strong PERs to sucrose concentrations as low as 10^-6 M sucrose. However, flies lacking the cells that express OBP57d and 57e have dramatically reduced PERs to sucrose concentrations from 10^-6 to 10^-2 M. Interestingly, at concentrations of 10^-1 M, there is no difference in the probability of eliciting the PER between the two groups. This demonstrates that the cells that normally express OBP57d and 57e are important for normal sensitivity of the tarsi to sucrose, but at high sucrose concentrations the loss of these cells has no effect. Therefore the PER reflex is intact in the Grim-expressing animals, but the sensitivity to sucrose is blunted. These results confirm that the cells expressing OBP57d and 57e are required for normal gustatory sensitivity of the tarsi to sucrose (Galindo, 2001).

poxneuro is a paired domain transcriptional regulator. Mutants defective for poxneuro have an abnormal number of leg segments and conversion of labellar gustatory sensilla to mechanosensory bristle phenotype. RT-PCR experiments reveal that poxneuro mutants fail to express most, but not all, putative gustatory receptors. To determine if poxneuro transforms the support cells that make OBPs, it was determined whether reporter genes regulated by the promoter for OBP56g are still expressed in the poxneuro mutant background (Galindo, 2001).

LacZ expression was examined in labellar sensilla in wild-type flies expressing LacZ under control of the OBP56g promoter. When the P element carrying this construct is crossed into the poxneuro genetic background, LacZ expression is completely absent in the labellum. These results indicate that the cells in labellar sensilla do not express OBP56g in the poxneuro genetic background. This supports the notion that poxneuro acts early in the development of chemosensory sensilla to delegate chemosensory identity on all cells in the sensillum, including the cells that synthesize and secrete OBP56g (Galindo, 2001).

To assess the role of poxneuro in the differentiation of the pharngeal chemosensory organs, flies carrying the OBP56b promoter driving LacZ expression in specific pharyngeal organs were crossed into the poxneuro genetic background. poxneuro does not disrupt expression of LacZ regulated by the pharyngeal OBP promoter OBP56b. Together, these data indicate that poxneuro is required for expression of OBP56g in the labellar gustatory sensilla, but not for OBP56b expressed in pharyngeal gustatory organs. This suggests that different developmental mechanisms are required for the proper specification of pharyngeal and labellar gustatory sensilla (Galindo, 2001).

Two features of this gene family are extraordinary: the low degree of sequence similarity among the family members and the location of so many members in large gene clusters. The diversity of the family based on amino acid homology is striking. Only the six conserved cysteines are conserved in all members, probably reflecting a requirement for proper disulfide bonding for functional tertiary structure. Most of the genes have a single splice junction located after the signal sequence. This may reflect a common ancestor for many of these genes. However, many genes have two splices, and often these occur in novel positions. Other than immunoglobulin gene families that underwent an explosive increase in number and diversity in early jawed vertebrates, the other large gene family that has undergone rapid diversification is another chemosensory-specific gene family, the odorant receptors. Seven transmembrane receptor families mediate chemical detection in vertebrates, Caenorhabditis elegans, and Drosophila. These mechanisms could have arisen independently in the three animal groups or, more likely, evolved from a common ancestor early in the animal lineage and diverged. While there is little or no sequence similarity between the receptor families in these three groups of animals, there are similarities between receptor genes within an organism, indicating that diversity within an organism may arise by gene duplication. The OBP family described here has little similarity among the Drosophila members. The presence of most of these genes in clusters suggests these genes arose by tandem duplication. Almost half of the OBP genes are located within 825 kb of genomic DNA located at chromosomal position 56-57 on the right arm of the second chromosome. This clustering is consistent with a gene duplication mechanism for generating the large size of the gene family. OBP83a and OBP83b are closely related and juxtaposed in the Drosophila genome, but there is only a single related gene present in other Drosophila species. This strongly supports a gene duplication mechanism accounting for the size of this family. However, the low sequence identity among the members suggests that there is rapid evolution of these sequences following duplication. Perhaps the OBP genes are evolving more rapidly than the receptor family. If true, this could reflect constraints on the evolution rate of the receptor family, which must interact with other signaling molecules, or there may be unappreciated positive selective pressure to have a diverse OBP population (Galindo, 2001).

The pheromone-binding protein from the moth B. mori undergoes pH-dependent conformational changes thought to reflect a mechanism for loading and unloading the pheromone. In particular it has been suggested that two adjacent histidines present in several pheromone-binding proteins just before alpha-helix 4 may be important for these conformational changes and, thus, for the function of these proteins (references in Galindo, 2001). If this is true, this feature should be conserved among other members of the family. When the Drosophila proteins are aligned with the B. mori pheromone-binding protein, it is found that these histidines are not conserved in the Drosophila sequences. In fact none of the 35 predicted Drosophila proteins contain this motif, although some putative proteins have histidines in this general vicinity. Either the moth pheromone proteins function differently from the other OBP family members or the conformational changes suggested to be important for loading and unloading ligands do not specifically require these residues (Galindo, 2001).

Recent studies analyzing the expression of members of the odorant receptor and gustatory receptor (GR) families suggest that the difference between smell and taste in Drosophila is not based on the receptor family expressed, but on the location and connections of the neurons that express the receptor. Indeed members of the GR family probably mediate olfactory responses. In olfactory and gustatory sensilla, ligands must enter the sensillum lymph to interact with receptor molecules located on the neuronal dendrites. Therefore, in sensilla that mediate both taste and smell the ligand is in solution. Tastants are generally soluble molecules, but many odorants are hydrophobic with low water:oil solubility ratios. Therefore, the role of the OBPs has previously been thought to be odor specific and possibly associated with the special problem of getting odorants into solution. The presence of OBPs in gustatory sensilla forces a rethinking of this issue. Indeed, lush mutants have defective behavioral responses to ethanol, a molecule that should have little difficulty entering aqueous or hydrophobic environments. Perhaps OBPs function to protect chemicals ligands from enzymatic modification in the sensillum lymph, since a variety of enzymes are expressed in the sensillum lymph, including cytochrome P450 enzymes. Whatever the biochemical role of the olfactory OBPs, it seems likely that this function is important for both olfactory and gustatory physiology (Galindo, 2001).