The research report should be a detailed narrative explaining the function, biological processes, and localization of the gene product. Citations should be given for all claims.
You should prioritize authoritative reviews and primary scientific literature when conducting research. You can supplement
this with annotations you find in gene/protein databases, but these can be outdated or inaccurate.
We are specifically interested in the primary function of the gene - for enzymes, what reaction is catalyzed, and what is the substrate specificity? For transporters, what is the substrate? For structural proteins or adapters, what is the broader structural role? For signaling molecules, what is the role in the pathway.
We are interested in where in or outside the cell the gene product carries out its function.
We are also interested in the signaling or biochemical pathways in which the gene functions. We are less interested in broad pleiotropic effects, except where these elucidate the precise role.
Include evidence where possible. We are interested in both experimental evidence as well as inference from structure, evolution, or bioinformatic analysis. Precise studies should be prioritized over high-throughput, where available.
The gene symbol CG33090 is correctly associated, in the supplied UniProt record, with UniProt X2JE45 from Drosophila melanogaster, gene identifier FBgn0028916, and historical aliases BG:DS01068.2, CG15287, CG31837, and CG7469. Its supplied InterPro assignments—GH116_N, GH116_catalytic, six-hairpin glycosidase superfamily, and non-lysosomal glucosylceramidase—are mutually consistent with a glycoside hydrolase family 116 (GH116)-like protein.
However, the gene symbol “Dmel\CG33090” is literature-limited for this specific protein. Searches using CG33090, FBgn0028916, X2JE45, and its historical aliases found no paper directly demonstrating enzyme activity, physiological substrate, localization, phenotype, or pathway. The retrieved GH116 and GBA2 publications do not mention CG33090. Therefore, CG33090 must not be presented as an experimentally established glucosylceramidase. The strongest defensible annotation is:
Predicted GH116-family retaining glycoside hydrolase; potentially a non-lysosomal, GBA2-like glucosylceramidase, but its physiological substrate, cellular location, and pathway function remain unverified.
| Field | Verified interpretation |
|---|---|
| Accession | UniProt X2JE45 |
| Organism | Drosophila melanogaster, fruit fly |
| Current gene symbol | CG33090 |
| FlyBase identifier | FBgn0028916 |
| Historical aliases | BG:DS01068.2, CG15287, CG31837, CG7469 |
| Current protein description | Uncharacterized protein |
| Domain-level classification | GH116_N and GH116_catalytic regions within a six-hairpin glycosidase-like fold |
No conflicting same-symbol protein was identified in the literature search. Nevertheless, historical CG aliases can reflect changes in genome assembly or gene-model consolidation and should not be treated as independent loci without release-specific verification.
The supplied domain architecture aligns with experimentally studied GH116 proteins: the bacterial model TxGH116 has an N-terminal two-sheet β-sandwich associated with a C-terminal six-hairpin/(α/α)₆-solenoid catalytic domain. This supports the database-level classification of CG33090 as GH116-like, although it does not establish a particular substrate (pengthaisong2023reactionmechanismof pages 1-2).
GH116 enzymes are retaining glycoside hydrolases: cleavage proceeds through a covalent glycosyl-enzyme intermediate and retains the substrate’s anomeric configuration. GH116 is functionally diverse, containing characterized β-glucosidase, β-xylosidase, and β-N-acetylglucosaminidase activities. Thus, the family assignment supports glycosidase chemistry but does not, by itself, prove glucosylceramide specificity (pengthaisong2023reactionmechanismof pages 1-2).
The best current functional label for CG33090 is consequently predicted GH116 glycoside hydrolase. Calling it “non-lysosomal glucosylceramidase,” assigning EC 3.2.1.45, or asserting glucosylceramide as its natural substrate would exceed the available CG33090-specific evidence.
If CG33090 is functionally equivalent to metazoan GBA2, its candidate reaction would be:
β-D-glucosylceramide + H₂O → ceramide + D-glucose
A GBA2-like enzyme might also hydrolyze glucosylsphingosine. Human GBA2 is a GH116 non-lysosomal enzyme reported to hydrolyze both glucosylceramide and glucosylsphingosine (pengthaisong2023reactionmechanismof pages 1-2). Human GBA2 additionally acts on bile-acid glucosides and can transfer sugar to cholesterol using glucosylceramide or galactosylceramide as donors, demonstrating that family members can support reactions beyond simple glucosylceramide hydrolysis (jatooratthawichot2020effectofexpression pages 1-3).
These observations make glucosylceramide a biologically plausible candidate substrate for CG33090, not an established one. GH116’s broader substrate diversity leaves open the possibility that the fly enzyme acts on a soluble glycoside or another glycoconjugate.
The most important recent mechanistic development is the 2023 ACS Catalysis study by Pengthaisong and colleagues, published in April 2023 (DOI: 10.1021/acscatal.3c00620). In TxGH116, Glu441 is the catalytic nucleophile and Asp593 is the general acid/base. These residues are approximately 8 Å apart, compared with roughly 4.5–6.5 Å in conventional retaining glycosidases that protonate laterally (pengthaisong2023reactionmechanismof pages 1-2).
Structures of a D593N mutant bound to cellobiose and laminaribiose, together with QM/MM calculations, support an unusual perpendicular protonation trajectory. The −1-subsite glucose binds in a relaxed ⁴C₁ chair, progresses through a ⁴H₃ half-chair transition state, and is protonated by the acid from above the pyranose-ring plane; a constrained gauche/trans orientation of glucose C6OH helps organize this geometry (pengthaisong2023reactionmechanismof pages 1-2).
For CG33090, retaining chemistry and analogous catalytic carboxylates are reasonable predictions. Exact catalytic-residue numbers cannot be transferred from TxGH116 without a sequence alignment, and catalytic equivalence should ultimately be tested by site-directed mutagenesis.
Substrate specificity is the central unresolved issue.
Human GBA2 provides the strongest metazoan analogy. A COS-7 expression study tested nine human GBA2 isoforms; only isoform 1 showed clearly elevated activity, whereas the others remained near background. The artificial-substrate assay used 3.5 mM 4-methylumbelliferyl-β-D-glucoside, pH 5.8, 37 °C, for 30 minutes, with or without 10 μM conduritol B epoxide (jatooratthawichot2020effectofexpression pages 1-3, jatooratthawichot2020effectofexpression pages 14-16). Human GBA2 has also been reported as a 927-residue, approximately 104-kDa enzyme with an activity optimum around pH 5.5–6.0; unlike lysosomal GBA1, it is CBE-insensitive and does not require saposin C (kim2021studiestowardsfinding pages 13-17).
Lipidomics after active GBA2 isoform-1 expression showed decreases in Hex1Cer species 34:1, 40:1, 42:1, and 42:2 and increases in corresponding ceramides. Cer/Hex1Cer ratios increased for all four matched species, supporting activity across multiple ceramide lipid moieties. Sphingomyelin/Hex1Cer ratios rose for 40:1, 40:2, and 42:2, consistent with downstream conversion of some released ceramide to sphingomyelin, although sphingomyelin abundance itself did not change significantly (jatooratthawichot2020effectofexpression pages 6-9).
The assay could not analytically distinguish glucosylceramide from galactosylceramide, although negligible GalCer in the tested COS-7 system made the HexCer signal likely to represent GlcCer. Effects were strongest at 48 hours and were not significant at 72 hours, emphasizing the context dependence of overexpression lipidomics (jatooratthawichot2020effectofexpression pages 6-9, jatooratthawichot2020effectofexpression pages 14-16).
None of these quantitative findings concerns CG33090 directly. They define what should be tested, rather than what can already be asserted about the fly protein.
No CG33090-specific microscopy, fractionation, topology, or endogenous-tagging study was found. Its cellular location therefore remains unknown.
Human GBA2 is reported to be a tightly membrane-associated but nonintegral protein located on the cytoplasmic faces of the endoplasmic reticulum and Golgi; family-level discussion also includes endosomes. This localization gives the catalytic domain access to cytosol-facing glycosphingolipid pools outside lysosomes (pengthaisong2023reactionmechanismof pages 1-2, kim2021studiestowardsfinding pages 13-17).
Accordingly, cytosolic association with ER, Golgi, or endosomal membranes is a reasonable working hypothesis for CG33090 only if it is genuinely GBA2-like. The “non-lysosomal glucosylceramidase” InterPro label does not by itself demonstrate membrane association or exclude another compartment.
Under the GBA2-like hypothesis, CG33090 would participate in non-lysosomal glycosphingolipid catabolism, controlling the balance between glucosylceramide and ceramide. Because ceramide can feed sphingomyelin and other sphingolipid branches, altered activity could affect a wider lipid network. Human-cell lipidomics supports this pathway direction: active GBA2 reduced multiple Hex1Cer species, increased matched ceramides, and altered selected sphingomyelin/Hex1Cer ratios without globally disrupting lipid-class distribution (jatooratthawichot2020effectofexpression pages 6-9).
Nevertheless, no CG33090 knockout, rescue, tissue-specific perturbation, metabolomics, or flux experiment was identified. It is therefore premature to assign CG33090 definitively to the fly sphingolipid pathway or to infer neuronal, developmental, fertility, or behavioral functions from mammalian GBA2 phenotypes.
The key distinction between established observations and inference is summarized below.
| Proposed statement | Evidence level | Supporting observations | Annotation implication |
|---|---|---|---|
| X2JE45 corresponds to Drosophila melanogaster CG33090 (FBgn0028916), with historical aliases BG:DS01068.2, CG15287, CG31837, and CG7469. | Direct database identity | The supplied UniProt record links the accession, organism, gene identifier, and aliases; exact-identifier searches found no conflicting protein assignment. | Identity is sufficiently resolved for annotation, but historical aliases should not be treated as separate genes without checking genome-release history. |
| CG33090 contains GH116-related regions, including GH116_N and GH116_catalytic, within a six-hairpin glycosidase-like fold. | Direct database identity / GH116 family evidence | The supplied InterPro assignments are IPR024462, IPR006775, IPR008928, IPR012341, and IPR052566. Experimentally studied GH116 proteins have an N-terminal β-sandwich and C-terminal six-hairpin or (α/α)₆ catalytic domain (pengthaisong2023reactionmechanismof pages 1-2). | The strongest defensible molecular annotation is predicted GH116-family glycoside hydrolase; domain membership alone does not identify the physiological substrate. |
| The protein probably uses retaining glycosidase chemistry. | GH116 family evidence | GH116 enzymes are retaining glycoside hydrolases. In TxGH116, Glu441 is the nucleophile and Asp593 is the acid/base; their approximately 8 Å separation supports unusual perpendicular protonation, compared with approximately 4.5–6.5 Å in conventional lateral protonators (pengthaisong2023reactionmechanismof pages 1-2). | Retention of anomeric configuration and analogous catalytic carboxylates are credible predictions, but the corresponding CG33090 residues must be confirmed by sequence alignment and mutagenesis. |
| CG33090 may hydrolyze glucosylceramide to ceramide and glucose and possibly glucosylsphingosine. | Human GBA2 analogy | Human GBA2 is a GH116 non-lysosomal enzyme that hydrolyzes glucosylceramide and glucosylsphingosine (pengthaisong2023reactionmechanismof pages 1-2). However, GH116 also includes β-glucosidase, β-xylosidase, β-N-acetylglucosaminidase, and other substrate specificities (pengthaisong2023reactionmechanismof pages 1-2). | Plausible but unproven. CG33090 should not be named glucosylceramidase or assigned EC 3.2.1.45 until activity against defined fly glucosylceramides is demonstrated. |
| If CG33090 is GBA2-like, it could act on the cytosolic faces of ER, Golgi, or endosomal membranes rather than in lysosomes. | Human GBA2 analogy | Human GBA2 is described as a nonintegral, membrane-associated enzyme on cytoplasmic ER/Golgi surfaces; family literature also reports association with endosomes (pengthaisong2023reactionmechanismof pages 1-2, kim2021studiestowardsfinding pages 13-17). | This is a testable localization hypothesis, not a CG33090 annotation. Cytosolic membrane association must not be inferred solely from the “non-lysosomal glucosylceramidase” InterPro label. |
| A GBA2-like CG33090 would participate in glycosphingolipid catabolism and regulate the glucosylceramide–ceramide balance. | Human GBA2 analogy | Expression of active human GBA2 isoform 1 increased ceramide and decreased Hex1Cer species 34:1, 40:1, 42:1, and 42:2; matched Cer/Hex1Cer ratios rose, supporting pathway flux from glucosylceramide toward ceramide (jatooratthawichot2020effectofexpression pages 6-9). | Comparative pathway placement is biologically coherent, but no CG33090-specific lipidomic, genetic, or flux evidence currently establishes this role in flies. |
| Substrate breadth cannot be predicted reliably from GH116 membership alone. | GH116 family evidence / unknown | Human GBA2 shows broad specificity for glucosylceramides with different lipid components and can act on bile-acid glucosides; GH116 enzymes as a whole accept chemically diverse glycosides (jatooratthawichot2020effectofexpression pages 1-3, pengthaisong2023reactionmechanismof pages 1-2). | Annotation should retain “predicted glycosidase” and list glucosylceramide as a candidate substrate rather than an established substrate. |
| No CG33090-specific biochemical or localization publication was found. | Unknown | Searches using CG33090, FBgn0028916, X2JE45, and the historical aliases found no publication reporting purified-enzyme activity, substrate specificity, localization, expression-linked mechanism, phenotype, or pathway assignment. The retrieved GH116 and GBA2 papers do not mention CG33090 (pengthaisong2023reactionmechanismof pages 1-2, jatooratthawichot2020effectofexpression pages 1-3). | Functional and localization claims remain computational or comparative; absence of a retrieved publication is not proof that no unpublished or newly deposited evidence exists. |
| Major experiments needed are recombinant-enzyme assays, catalytic-residue validation, lipidomics, and localization analysis. | Unknown | Priority tests include hydrolysis of structurally defined fly glucosylceramides and glucosylsphingosine; screening against soluble β-glucosides, xylosides, and β-N-acetylglucosaminides; product and stereochemistry analysis; catalytic carboxylate mutagenesis; endogenous tagging and membrane-topology studies; and CG33090 knockout/rescue lipidomics. Human GBA2 assays used 3.5 mM 4MUG at pH 5.8 and 37 °C for 30 minutes, but such artificial-substrate activity alone would not establish physiological specificity (jatooratthawichot2020effectofexpression pages 14-16). | These experiments are required to distinguish a GBA2-like glucosylceramidase from another GH116 glycosidase and to determine its actual cellular compartment and pathway role. |
Table: This table separates verified CG33090 identity and domain information from GH116-family inference, human GBA2 analogy, and unresolved claims. It highlights why glucosylceramidase activity, membrane localization, and pathway placement remain hypotheses requiring direct experiments.
The major 2023 advance was mechanistic rather than gene-specific: structural complexes and QM/MM simulations established GH116’s unusual perpendicular protonation geometry. This finding has implications for designing inhibitors that distinguish perpendicular-protonating GH116 enzymes such as GBA2 from conventional lateral protonators such as GBA1 (pengthaisong2023reactionmechanismof pages 1-2).
A 2023 review of glycoside hydrolases and the 2023 mechanistic work also reinforce that GH116 comprises subfamilies with distinct substrate preferences. This makes domain-only annotation inadequate for assigning the physiological substrate of CG33090. No 2023–2024 primary study directly characterizing CG33090 was retrieved.
The appropriate expert interpretation is therefore conservative: the domain evidence is strong for catalytic class, moderate for retaining GH116 chemistry, but weak for a GBA2-specific substrate and weak for any cellular localization or physiological pathway. An InterPro hit to “non-lysosomal glucosylceramidase” should be treated as a homology-derived hypothesis, not equivalent to biochemical validation.
There is no known CG33090-specific clinical, industrial, or biotechnology implementation. Its current value is principally as:
Broader GH116 research has practical relevance to selective inhibitor design, glycosidase chemical biology, glycosynthase engineering, and lipid-metabolism studies, but these applications cannot yet be attributed specifically to CG33090 (pengthaisong2023reactionmechanismof pages 1-2, jatooratthawichot2020effectofexpression pages 1-3).
The highest-priority experiments are:
Recommended name: “Predicted GH116-family glycoside hydrolase.”
Permissible note: “Contains GH116_N and GH116 catalytic domains; potentially related to non-lysosomal glucosylceramidases.”
Do not yet assign as established: non-lysosomal glucosylceramidase/GBA2, EC 3.2.1.45, glucosylceramide as physiological substrate, ER/Golgi/endosomal localization, or a defined glycosphingolipid pathway role.
Overall, CG33090 is a well-supported GH116-domain protein but remains an experimentally uncharacterized Drosophila enzyme. Its proposed glucosylceramidase activity and membrane-associated sphingolipid role are useful, testable hypotheses—not established functional annotations.
References
(pengthaisong2023reactionmechanismof pages 1-2): Salila Pengthaisong, Beatriz Piniello, Gideon J. Davies, Carme Rovira, and James R. Ketudat Cairns. Reaction mechanism of glycoside hydrolase family 116 utilizes perpendicular protonation. ACS Catalysis, 13:5850-5863, Apr 2023. URL: https://doi.org/10.1021/acscatal.3c00620, doi:10.1021/acscatal.3c00620. This article has 29 citations and is from a highest quality peer-reviewed journal.
(jatooratthawichot2020effectofexpression pages 1-3): Peeranat Jatooratthawichot, Chutima Talabnin, Lukana Ngiwsara, Yepy Hardi Rustam, Jisnuson Svasti, Gavin E. Reid, and James R. Ketudat Cairns. Effect of expression of human glucosylceramidase 2 isoforms on lipid profiles in cos-7 cells. Nov 2020. URL: https://doi.org/10.3390/metabo10120488, doi:10.3390/metabo10120488. This article has 10 citations.
(jatooratthawichot2020effectofexpression pages 14-16): Peeranat Jatooratthawichot, Chutima Talabnin, Lukana Ngiwsara, Yepy Hardi Rustam, Jisnuson Svasti, Gavin E. Reid, and James R. Ketudat Cairns. Effect of expression of human glucosylceramidase 2 isoforms on lipid profiles in cos-7 cells. Nov 2020. URL: https://doi.org/10.3390/metabo10120488, doi:10.3390/metabo10120488. This article has 10 citations.
(kim2021studiestowardsfinding pages 13-17): J Kim. Studies towards finding selective substrates for human β-glucosidases. Unknown journal, 2021.
(jatooratthawichot2020effectofexpression pages 6-9): Peeranat Jatooratthawichot, Chutima Talabnin, Lukana Ngiwsara, Yepy Hardi Rustam, Jisnuson Svasti, Gavin E. Reid, and James R. Ketudat Cairns. Effect of expression of human glucosylceramidase 2 isoforms on lipid profiles in cos-7 cells. Nov 2020. URL: https://doi.org/10.3390/metabo10120488, doi:10.3390/metabo10120488. This article has 10 citations.