rdgBbeta encodes a soluble, cytoplasmic phosphatidylinositol transfer protein (PITP) of the Class IIB subfamily. Unlike Class I PITPs (vibrator, giotto) which bind both PI and PC, the human ortholog PITPNC1 binds and transfers phosphatidylinositol (PI) and phosphatidic acid (PA) but does not significantly bind phosphatidylcholine. The protein consists of a PITP/START-like lipid-binding domain and a short disordered C-terminal tail, lacking the transmembrane and additional domains found in other rdgB family members. rdgBbeta is broadly expressed and has been detected in adult head tissue by mass spectrometry. Genetic evidence in Drosophila links rdgBbeta to phosphoinositide metabolism in the context of neurodegeneration, as rdgBbeta alleles modify the Vap33-P58S ALS8 model phenotype.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
|
GO:0005737
cytoplasm
|
IBA
GO_REF:0000033 |
ACCEPT |
Summary: IBA annotation from phylogenetic inference. Cytoplasmic localization is well-supported across the PITP family. The human ortholog PITPNC1 was shown to be cytoplasmic by immunofluorescence (PMID:10531358), and rdgBbeta lacks transmembrane domains or secretion signals. This is consistent with a soluble lipid transfer protein.
Reason: Cytoplasmic localization is supported by the domain architecture (no transmembrane segments), the experimental localization of the human ortholog PITPNC1 (PMID:10531358), and mass spectrometry detection in Drosophila adult head (FlyBase).
Supporting Evidence:
PMID:10531358
Immunofluorescence analysis of ectopic MrdgBbeta showed cytoplasmic staining
|
|
GO:0008526
phosphatidylinositol transfer activity
|
IBA
GO_REF:0000033 |
ACCEPT |
Summary: IBA annotation from phylogenetic inference. PI transfer activity is the defining molecular function of the PITP family. The human ortholog PITPNC1 has been shown to transfer PI in vitro (PMID:10531358, PMID:22822086).
Reason: This is the core molecular function. The human ortholog was demonstrated to transfer PI in vitro with activity comparable to other PITP-like domains (PMID:10531358). The conserved PITP domain architecture (IPR001666) and START-like fold support this function. The BioReason deep research correctly identifies PI transfer as a core function (rdgBbeta-deep-research-bioreason-sft.md).
Supporting Evidence:
PMID:10531358
the ability of recombinant MrdgBbeta to transfer phosphatidylinositol in vitro was similar to other PITP-like domains
PMID:22822086
besides PI, RdgBβ binds and transfers phosphatidic acid (PA) but hardly binds phosphatidylcholine
file:DROME/rdgBbeta/rdgBbeta-deep-research-bioreason-sft.md
[BioReason correctly identifies] GO:0008526 phosphatidylinositol transfer activity
|
|
GO:0035091
phosphatidylinositol binding
|
IBA
GO_REF:0000033 |
ACCEPT |
Summary: IBA annotation from phylogenetic inference. PI binding is intrinsic to PI transfer activity and well-established for PITPNC1/rdgBbeta orthologs.
Reason: PI binding is a prerequisite for PI transfer activity and is directly demonstrated for the human ortholog (PMID:22822086). The PITP domain creates a hydrophobic cavity that accommodates PI monomers.
Supporting Evidence:
PMID:22822086
PA and PI were now incorporated into RdgBβ
|
|
GO:0005543
phospholipid binding
|
IEA
GO_REF:0000117 |
KEEP AS NON CORE |
Summary: IEA annotation from ARBA machine learning. This is a correct but very general annotation. The protein binds specific phospholipids (PI and PA), so more specific terms are more informative.
Reason: Phospholipid binding is accurate but overly broad. The protein specifically binds PI and PA (PMID:22822086). The more specific annotations for PI binding (GO:0035091) and PI transfer activity (GO:0008526) better capture the core function.
Supporting Evidence:
PMID:22822086
besides PI, RdgBβ binds and transfers phosphatidic acid (PA) but hardly binds phosphatidylcholine
|
|
GO:0005634
nucleus
|
IEA
GO_REF:0000117 |
REMOVE |
Summary: IEA annotation from ARBA machine learning. There is no experimental evidence supporting nuclear localization of rdgBbeta or its orthologs. The protein is cytoplasmic.
Reason: No published evidence supports nuclear localization for rdgBbeta or PITPNC1. The human ortholog was localized to the cytoplasm by immunofluorescence (PMID:10531358). The protein lacks nuclear localization signals. This ARBA prediction appears to be erroneous.
Supporting Evidence:
PMID:10531358
Immunofluorescence analysis of ectopic MrdgBbeta showed cytoplasmic staining
|
|
GO:0005737
cytoplasm
|
IEA
GO_REF:0000117 |
ACCEPT |
Summary: IEA annotation from ARBA machine learning for cytoplasmic localization. Redundant with the IBA annotation for the same term but independently correct.
Reason: Correct annotation consistent with experimental evidence from the human ortholog and absence of membrane-targeting signals. Redundant with the IBA annotation above.
Supporting Evidence:
PMID:10531358
Immunofluorescence analysis of ectopic MrdgBbeta showed cytoplasmic staining
|
|
GO:0015914
phospholipid transport
|
IEA
GO_REF:0000002 |
ACCEPT |
Summary: IEA annotation from InterPro domain mapping (IPR001666 PITP family). Phospholipid transport is the core biological process for this protein family.
Reason: Phospholipid transport is the direct biological process consequence of PI transfer activity. The PITP domain mediates monomeric lipid transfer between membranes. This is well-supported by the domain architecture and ortholog biochemistry.
Supporting Evidence:
PMID:22822086
RdgBβ, when containing PA, regulates an effector protein or can facilitate lipid transfer between membrane compartments
|
|
GO:0005737
cytoplasm
|
ISS
GO_REF:0000024 |
ACCEPT |
Summary: ISS annotation by manual curator judgment based on sequence similarity to human PITPNC1 (Q9UKF7). Cytoplasmic localization is correct and well-supported.
Reason: Based on sequence similarity to human PITPNC1, which was experimentally shown to localize to the cytoplasm (PMID:10531358). This is the strongest non-experimental annotation for localization.
Supporting Evidence:
PMID:10531358
Immunofluorescence analysis of ectopic MrdgBbeta showed cytoplasmic staining
|
|
GO:0007165
signal transduction
|
ISS
GO_REF:0000024 |
KEEP AS NON CORE |
Summary: ISS annotation based on sequence similarity to human PITPNC1 (Q9UKF7). Signal transduction is a plausible but very broad annotation. PITPs can influence signaling by supplying PI for phosphoinositide synthesis, and the human ortholog interacts with signaling components (ATRAP, 14-3-3).
Reason: Signal transduction is plausible but overly broad. PITPNC1 participates in phospholipase D signaling by binding PA (PMID:22822086) and interacts with ATRAP in the angiotensin signaling pathway (PMID:21728994). However, the specific signaling pathways for Drosophila rdgBbeta are unknown. The annotation is not wrong but is too general to be informative about core function.
Supporting Evidence:
PMID:22822086
RdgBβ, when containing PA, regulates an effector protein or can facilitate lipid transfer between membrane compartments
PMID:21728994
the PITP domain of RdgBβ interacts with the integral membrane protein ATRAP
|
|
GO:0008526
phosphatidylinositol transfer activity
|
ISS
GO_REF:0000024 |
ACCEPT |
Summary: ISS annotation based on sequence similarity to human PITPNC1 (Q9UKF7). PI transfer activity is the core molecular function, well-established for the ortholog.
Reason: Redundant with the IBA annotation for the same term. PI transfer activity is the defining function of this protein family and is directly demonstrated for the human ortholog (PMID:10531358, PMID:22822086).
Supporting Evidence:
PMID:10531358
the ability of recombinant MrdgBbeta to transfer phosphatidylinositol in vitro was similar to other PITP-like domains
|
Q: What is the in vivo function of rdgBbeta in Drosophila? No mutant phenotype has been characterized beyond genetic interaction with Vap33-P58S. Does rdgBbeta loss-of-function produce any developmental or physiological phenotypes?
Q: Does Drosophila rdgBbeta also bind and transfer phosphatidic acid, as demonstrated for the human ortholog PITPNC1? What are the specific lipid species it handles in vivo?
Q: How does rdgBbeta modify the Vap33-P58S ALS phenotype? Does it act by modulating phosphoinositide levels, or through an independent mechanism?
Experiment: Generate rdgBbeta null mutants using CRISPR and characterize phenotypes in development, neuronal function, and lipid metabolism. Assess phosphoinositide and PA levels in mutant tissue.
Hypothesis: rdgBbeta loss-of-function may cause subtle defects in lipid homeostasis that become apparent under stress conditions or in sensitized genetic backgrounds.
Experiment: Perform in vitro lipid binding and transfer assays with purified Drosophila rdgBbeta to determine whether it shares the PI/PA dual specificity of human PITPNC1.
Hypothesis: Drosophila rdgBbeta will show PI and PA binding/transfer activity comparable to human PITPNC1, but minimal PC binding, consistent with Class IIB PITP specificity.
Experiment: Characterize the genetic interaction between rdgBbeta and Vap33-P58S in detail, measuring phosphoinositide levels and neurodegeneration markers.
Hypothesis: rdgBbeta alleles rescue Vap33-P58S by reducing phosphoinositide delivery to membranes, partially compensating for the loss of Sac1-mediated PI dephosphorylation.
---
provider: bioreason-pro
model: SFT
source: huggingface.co/datasets/wanglab/protein_catalogue
date: '2026-04-04'
uniprot_id: Q9U9P7
gene: rdgBbeta
organism: Drosophila melanogaster (Fruit fly)
---
## Thinking Trace
The architecture begins with IPR023393 (START-like domain superfamily, residues 2–263), a fold that creates a deep hydrophobic cavity for lipid sequestration and exchange. Nested within this scaffold is IPR055261 (Phosphatidylinositol transfer protein, N-terminal domain, residues 2–244), which defines the specific PITP topology that opens and closes a lipid-binding pocket. Multiple family-level signatures of the phosphatidylinositol transfer protein lineage are embedded across the sequence—IPR001666 (PITP family) at residues 3–254, 17–36, 84–104, 110–125, 193–208, and 213–232—marking conserved motifs that line the cavity and the gating elements that control lipid capture and release. The ordered nesting of a PITP-specific N-terminal domain inside a START-like superfamily fold causes high-affinity, monomeric lipid transfer between membranes without vesicle budding or fusion.
This PITP fold enforces a molecular function centered on selective lipid exchange. The conserved cavity and gating residues accommodate phosphatidylinositol and phosphatidylcholine, enabling GO:0008526 phosphatidylinositol transfer activity and GO:0008525 phosphatidylcholine transporter activity. Because the cavity is optimized for neutral and anionic glycerophospholipids, the mechanism is a cycle of membrane docking, lipid extraction into the hydrophobic pocket, transit through the cytosol, and deposition at a target membrane.
Lipid transfer of PI and PC has direct consequences for membrane composition and signaling during cell division. By delivering PI to specific membrane subdomains, the protein sustains phosphoinositide synthesis and curvature-sensitive recruitment of cytokinetic machinery; by shuttling PC, it maintains bilayer integrity and supports membrane expansion at the ingressing furrow. These activities drive the processes formalized as GO:0007110 meiosis I cytokinesis and GO:0007111 meiosis II cytokinesis, and extend to GO:0007112 male meiosis cytokinesis and GO:0048137 spermatocyte division, where precise membrane remodeling is essential. The same lipid supply and signaling flux coordinate with the actomyosin apparatus, promoting GO:0000916 actomyosin contractile ring contraction and GO:0036090 cleavage furrow ingression by ensuring a competent, lipid-rich cortex that can withstand contractile forces and by positioning PI pools that recruit small GTPases and effectors. Proper PI availability also influences microtubule–cortex communication and membrane–microtubule interfaces, contributing to GO:0000212 meiotic spindle organization.
The domain architecture lacks transmembrane segments and secretion signals, indicating a soluble protein that cycles on and off membranes. Its action requires proximity to sites of intense membrane turnover and signaling. Thus, it operates in the cytoplasm and transiently associates with the GO:0005794 Golgi apparatus to acquire and dispatch lipids, with the GO:0032154 cleavage furrow where cytokinesis occurs, and with the GO:0031965 nuclear membrane during meiotic stages when nuclear envelope dynamics and membrane continuity are critical. This distribution is consistent with a soluble lipid-transfer factor that concentrates at division sites and organelles engaged in lipid trafficking.
Mechanistically, the protein likely docks to donor membranes via basic surface patches and amphipathic elements, captures PI or PC into its START-like cavity, and releases the lipid at target membranes to sustain phosphoinositide synthesis and bilayer homeostasis. At the cleavage furrow and nuclear envelope, it plausibly collaborates with small GTPases (e.g., ARF and RAB family members) and phosphoinositide kinases to position PI pools that recruit actomyosin regulators and membrane-sculpting factors. It may also engage cytokinetic scaffolds at the furrow to synchronize lipid supply with contractile ring constriction and with spindle–cortex communication during meiosis.
## Functional Summary
A soluble lipid-transfer factor in fruit fly that uses a deep hydrophobic cavity to extract and shuttle phosphatidylinositol and phosphatidylcholine between membranes. By supplying these lipids to division sites and organelles, it sustains phosphoinositide signaling and bilayer integrity required for spindle organization, contractile ring function, and membrane expansion during meiotic and spermatocyte cytokinesis. It operates in the cytoplasm and transiently associates with the Golgi, the cleavage furrow, and the nuclear envelope to coordinate lipid flux with cell division mechanics.
## UniProt Summary
Catalyzes the transfer of phosphatidylinositol and phosphatidylcholine between membranes.
## InterPro Domains
- IPR023393: START-like domain superfamily (homologous_superfamily) [2-263]
- IPR055261: Phosphatidylinositol transfer protein, N-terminal (domain) [2-244]
- IPR001666: Phosphatidylinositol transfer protein (family) [3-254]
- IPR001666: Phosphatidylinositol transfer protein (family) [17-36]
- IPR001666: Phosphatidylinositol transfer protein (family) [84-104]
- IPR001666: Phosphatidylinositol transfer protein (family) [110-125]
- IPR001666: Phosphatidylinositol transfer protein (family) [193-208]
- IPR001666: Phosphatidylinositol transfer protein (family) [213-232]
## GO Term Predictions
### Molecular Function
### Biological Process
### Cellular Component
rdgBbeta belongs to the PtdIns transfer protein family, PI transfer class IIB subfamily.
It contains:
- IPR023393: START-like domain superfamily (residues 2-263)
- IPR055261: Phosphatidylinositol transfer protein, N-terminal domain (residues 2-244)
- IPR001666: Phosphatidylinositol transfer protein (family) (residues 3-254 and multiple sub-regions)
Key structural feature: rdgBbeta contains ONLY the PITP domain plus a short disordered C-terminal tail.
It lacks transmembrane domains and the conserved C-terminal domain found in other rdgB-family proteins
(rdgB/RdgBalpha, Nir2, Nir3). This makes it a soluble, cytoplasmic lipid transfer protein.
The human ortholog PITPNC1 has been characterized biochemically PMID:22822086. Critical finding:
- PITPNC1/rdgBbeta binds and transfers phosphatidylinositol (PI) and phosphatidic acid (PA)
- It does NOT significantly bind phosphatidylcholine (PC)
- This distinguishes it from Class I PITPs (PITPalpha, PITPbeta/vibrator/giotto) which bind both PI and PC
PITPNC1/RdgBbeta is the first lipid-binding protein identified that can bind and transfer PA.
When purified from E. coli, it was preloaded with PA and phosphatidylglycerol.
PMID:21728994 The C-terminus of RdgBbeta binds 14-3-3 proteins via two tandem phosphorylated sites
(Ser274 and Ser299 in human). The PITP domain interacts with ATRAP (angiotensin II type I
receptor-associated protein), an integral membrane protein that recruits RdgBbeta to membranes.
14-3-3 shields PEST sequences; loss of 14-3-3 binding accelerates proteasomal degradation.
[PMID:10531358 Fullwood et al. 1999] Cloned human rdgBbeta by EST database searching for rdgB homologs.
Key findings:
- 333 amino acids (human), N-terminal PITP-like domain + short C-terminal domain
- No transmembrane domains (unlike other rdgB family members)
- Cytoplasmic localization by immunofluorescence
- Ubiquitous expression (highest in heart, muscle, kidney, liver, leukocytes)
- In vitro PI transfer activity comparable to other PITP-like domains
PMID:23492670 Drosophila ALS8 model: Vap33-P58S expression causes neurodegeneration through
increased phosphoinositide levels. Sac1 phosphatase is sequestered into aggregates.
While this paper does not mention rdgBbeta directly, the FlyBase genetic interaction data
suggests rdgBbeta alleles modify this phenotype, consistent with its role in PI metabolism.
Important: rdgBbeta should NOT be confused with:
- rdgB/RdgBalpha (FBgn0003218): membrane-associated PITP with additional domains (FFAT, DDHD),
essential for photoreceptor maintenance, required for phototransduction
- vibrator (vib): Class I PITP, required for cytokinesis (cleavage furrow ingression) PMID:16684816
- giotto (gio): Class I PITP, required for both mitotic and meiotic cytokinesis PMID:16431372
The cytokinesis, meiosis, and spermatocyte division phenotypes in Drosophila are associated with
vibrator and giotto mutants, NOT rdgBbeta.
PMID:26977884 PITPNC1 is amplified in breast cancer and overexpressed in metastatic breast,
melanoma, and colon cancers. It promotes malignant secretion by binding Golgi-resident PI4P and
localizing RAB1B to the Golgi, driving secretion of pro-invasive and pro-angiogenic factors
(HTRA1, MMP1, FAM3C, PDGFA, ADAM10).
The BioReason deep-research file makes several claims that need verification:
GO:0008525 phosphatidylcholine transporter activity -- INCORRECT. Literature shows rdgBbeta/PITPNC1
"hardly binds phosphatidylcholine" PMID:22822086. This is a Class I PITP property, not Class IIB.
Cytokinesis functions (GO:0007110, GO:0007111, GO:0007112, GO:0000916, GO:0036090, GO:0048137) --
NOT SUPPORTED for rdgBbeta. These functions are associated with vibrator and giotto (Class I PITPs),
not rdgBbeta. No published evidence connects rdgBbeta to cytokinesis or meiotic division.
GO:0000212 meiotic spindle organization -- NOT SUPPORTED. No evidence for rdgBbeta.
GO:0005794 Golgi apparatus -- Partially supported for human PITPNC1 in cancer context PMID:26977884,
but not demonstrated for Drosophila rdgBbeta.
GO:0032154 cleavage furrow -- NOT SUPPORTED. Cleavage furrow association is for vibrator, not rdgBbeta.
GO:0031965 nuclear membrane -- NOT SUPPORTED. No evidence for rdgBbeta.
PI transfer activity (GO:0008526) -- CORRECT. Well-supported by ortholog data.
Cytoplasmic localization (GO:0005737) -- CORRECT. Supported by multiple lines of evidence.
The BioReason model appears to have conflated the functions of different PITP family members,
attributing Class I PITP functions (cytokinesis, PC binding) to this Class IIB PITP.
Source: rdgBbeta-deep-research-bioreason-sft.md
The BioReason functional summary describes rdgBbeta as:
A soluble lipid-transfer factor in fruit fly that uses a deep hydrophobic cavity to extract and shuttle phosphatidylinositol and phosphatidylcholine between membranes. By supplying these lipids to division sites and organelles, it sustains phosphoinositide signaling and bilayer integrity required for spindle organization, contractile ring function, and membrane expansion during meiotic and spermatocyte cytokinesis. It operates in the cytoplasm and transiently associates with the Golgi, the cleavage furrow, and the nuclear envelope to coordinate lipid flux with cell division mechanics.
This summary contains multiple factual errors that arise from conflating rdgBbeta (a Class IIB PITP) with Class I PITPs (vibrator/giotto) and from fabricating biological processes unsupported by any evidence.
Correctness issues:
Phosphatidylcholine binding is incorrect. The summary states rdgBbeta shuttles "phosphatidylinositol and phosphatidylcholine." Experimental evidence for the human ortholog PITPNC1 (PMID:22822086) demonstrates that RdgBbeta "hardly binds phosphatidylcholine," which is what distinguishes Class IIB PITPs from Class I PITPs. Instead, rdgBbeta binds and transfers phosphatidic acid (PA) alongside PI. This is a fundamental biochemical error.
Cytokinesis functions are entirely fabricated. The BioReason prediction attributes meiosis I cytokinesis (GO:0007110), meiosis II cytokinesis (GO:0007111), male meiosis cytokinesis (GO:0007112), actomyosin contractile ring contraction (GO:0000916), cleavage furrow ingression (GO:0036090), spermatocyte division (GO:0048137), and meiotic spindle organization (GO:0000212) to rdgBbeta. There is NO published evidence connecting rdgBbeta to any of these processes. These are known phenotypes of the Drosophila Class I PITPs vibrator (PMID:16684816) and giotto (PMID:16431372), which are completely different genes. The BioReason model appears to have conflated different PITP family members.
Subcellular localizations are fabricated. The claims of association with the Golgi apparatus (GO:0005794), cleavage furrow (GO:0032154), and nuclear membrane (GO:0031965) have no experimental support for Drosophila rdgBbeta. Cleavage furrow association is documented for vibrator, not rdgBbeta. The Golgi connection comes from human PITPNC1 cancer biology (PMID:26977884) but has not been shown for the Drosophila protein.
GO:0008525 phosphatidylcholine transporter activity is incorrect. The thinking trace explicitly predicts this term, which directly contradicts the biochemical evidence showing rdgBbeta does not bind PC (PMID:22822086).
What BioReason got right:
Completeness issues:
No mention of phosphatidic acid binding and transfer. This is the most distinctive biochemical feature of PITPNC1/rdgBbeta (PMID:22822086), making it unique among lipid transfer proteins. The BioReason model completely missed this.
No mention of 14-3-3 binding and ATRAP interaction. The C-terminal regulatory mechanism involving 14-3-3 proteins and ATRAP-mediated membrane recruitment (PMID:21728994) is absent.
No mention of the ALS/neurodegeneration connection. The most interesting Drosophila-specific finding is that rdgBbeta alleles modify the Vap33-P58S ALS8 model phenotype (FlyBase genetic interaction data), connecting it to phosphoinositide homeostasis in neurodegeneration.
No mention of the cancer biology of PITPNC1. Human PITPNC1 is amplified in breast cancer and drives metastatic secretion via RAB1B/Golgi network (PMID:26977884). While this is human-specific, it provides functional insight into PITP-Golgi interactions.
No mention of proteasomal regulation. RdgBbeta is degraded by the proteasome with a 4-hour half-life, regulated by 14-3-3 shielding of PEST sequences (PMID:21728994).
The InterPro2GO annotation is:
- GO:0015914 phospholipid transport (from IPR001666)
This single annotation is accurate and conservative. The BioReason narrative adds some valid mechanistic context (PITP domain architecture, lipid cavity) but introduces far more errors (PC binding, cytokinesis, fabricated localizations) than it adds value. The InterPro2GO annotation is more reliable than the BioReason prediction for this gene.
The thinking trace reveals systematic problems in the BioReason reasoning:
Family-level conflation. The trace correctly identifies the PITP domain architecture but then attributes functions of other PITP family members (vibrator, giotto) to rdgBbeta. This is analogous to attributing hemoglobin functions to myoglobin because they share the globin fold.
Hallucinated biological processes. The cytokinesis, meiosis, spermatocyte division, and spindle organization functions are not from any annotation database or publication for rdgBbeta. They appear to be fabricated from the general PITP literature, where these functions belong to different genes.
PC binding error contradicts available biochemistry. The thinking trace states the cavity "accommodates phosphatidylinositol and phosphatidylcholine," which is precisely wrong for this Class IIB PITP. The key biochemical finding (PMID:22822086) is that Class IIB PITPs bind PA instead of PC. This suggests the model lacks access to or failed to incorporate the PITPNC1-specific biochemistry literature.
Overconfident localization claims. The trace confidently assigns Golgi, cleavage furrow, and nuclear membrane localizations based purely on functional reasoning ("it must go where lipids are needed"), without any experimental evidence. This is speculative reasoning presented as factual.
Expression data contradicts claims. FlyBase shows rdgBbeta has a weak negative testis specificity index (-0.19), meaning it is actually underrepresented in testis. This directly contradicts the model's emphasis on spermatocyte and meiotic functions.
The BioReason prediction for rdgBbeta demonstrates the failure mode of domain-architecture-only reasoning when a protein's specific biology diverges from the general family narrative. Class IIB PITPs have distinct lipid specificity and biological roles compared to Class I PITPs, and these distinctions cannot be captured by domain-level analysis alone.
id: Q9U9P7
gene_symbol: rdgBbeta
product_type: PROTEIN
status: DRAFT
taxon:
id: NCBITaxon:7227
label: Drosophila melanogaster
description: >-
rdgBbeta encodes a soluble, cytoplasmic phosphatidylinositol transfer protein (PITP) of the
Class IIB subfamily. Unlike Class I PITPs (vibrator, giotto) which bind both PI and PC, the
human ortholog PITPNC1 binds and transfers phosphatidylinositol (PI) and phosphatidic acid (PA)
but does not significantly bind phosphatidylcholine. The protein consists of a PITP/START-like
lipid-binding domain and a short disordered C-terminal tail, lacking the transmembrane and
additional domains found in other rdgB family members. rdgBbeta is broadly expressed and has
been detected in adult head tissue by mass spectrometry. Genetic evidence in Drosophila links
rdgBbeta to phosphoinositide metabolism in the context of neurodegeneration, as rdgBbeta alleles
modify the Vap33-P58S ALS8 model phenotype.
existing_annotations:
- term:
id: GO:0005737
label: cytoplasm
evidence_type: IBA
original_reference_id: GO_REF:0000033
review:
summary: >-
IBA annotation from phylogenetic inference. Cytoplasmic localization is well-supported
across the PITP family. The human ortholog PITPNC1 was shown to be cytoplasmic by
immunofluorescence (PMID:10531358), and rdgBbeta lacks transmembrane domains or
secretion signals. This is consistent with a soluble lipid transfer protein.
action: ACCEPT
reason: >-
Cytoplasmic localization is supported by the domain architecture (no transmembrane
segments), the experimental localization of the human ortholog PITPNC1 (PMID:10531358),
and mass spectrometry detection in Drosophila adult head (FlyBase).
supported_by:
- reference_id: PMID:10531358
supporting_text: "Immunofluorescence analysis of ectopic MrdgBbeta showed cytoplasmic staining"
- term:
id: GO:0008526
label: phosphatidylinositol transfer activity
evidence_type: IBA
original_reference_id: GO_REF:0000033
review:
summary: >-
IBA annotation from phylogenetic inference. PI transfer activity is the defining
molecular function of the PITP family. The human ortholog PITPNC1 has been shown
to transfer PI in vitro (PMID:10531358, PMID:22822086).
action: ACCEPT
reason: >-
This is the core molecular function. The human ortholog was demonstrated to transfer
PI in vitro with activity comparable to other PITP-like domains (PMID:10531358). The
conserved PITP domain architecture (IPR001666) and START-like fold support this function.
The BioReason deep research correctly identifies PI transfer as a core function
(rdgBbeta-deep-research-bioreason-sft.md).
supported_by:
- reference_id: PMID:10531358
supporting_text: "the ability of recombinant MrdgBbeta to transfer phosphatidylinositol in vitro was similar to other PITP-like domains"
- reference_id: PMID:22822086
supporting_text: "besides PI, RdgBβ binds and transfers phosphatidic acid (PA) but hardly binds phosphatidylcholine"
- reference_id: file:DROME/rdgBbeta/rdgBbeta-deep-research-bioreason-sft.md
supporting_text: "[BioReason correctly identifies] GO:0008526 phosphatidylinositol transfer activity"
- term:
id: GO:0035091
label: phosphatidylinositol binding
evidence_type: IBA
original_reference_id: GO_REF:0000033
review:
summary: >-
IBA annotation from phylogenetic inference. PI binding is intrinsic to PI transfer
activity and well-established for PITPNC1/rdgBbeta orthologs.
action: ACCEPT
reason: >-
PI binding is a prerequisite for PI transfer activity and is directly demonstrated
for the human ortholog (PMID:22822086). The PITP domain creates a hydrophobic
cavity that accommodates PI monomers.
supported_by:
- reference_id: PMID:22822086
supporting_text: "PA and PI were now incorporated into RdgBβ"
- term:
id: GO:0005543
label: phospholipid binding
evidence_type: IEA
original_reference_id: GO_REF:0000117
review:
summary: >-
IEA annotation from ARBA machine learning. This is a correct but very general annotation.
The protein binds specific phospholipids (PI and PA), so more specific terms are more
informative.
action: KEEP_AS_NON_CORE
reason: >-
Phospholipid binding is accurate but overly broad. The protein specifically binds PI
and PA (PMID:22822086). The more specific annotations for PI binding (GO:0035091) and
PI transfer activity (GO:0008526) better capture the core function.
supported_by:
- reference_id: PMID:22822086
supporting_text: "besides PI, RdgBβ binds and transfers phosphatidic acid (PA) but hardly binds phosphatidylcholine"
- term:
id: GO:0005634
label: nucleus
evidence_type: IEA
original_reference_id: GO_REF:0000117
review:
summary: >-
IEA annotation from ARBA machine learning. There is no experimental evidence
supporting nuclear localization of rdgBbeta or its orthologs. The protein is
cytoplasmic.
action: REMOVE
reason: >-
No published evidence supports nuclear localization for rdgBbeta or PITPNC1. The
human ortholog was localized to the cytoplasm by immunofluorescence (PMID:10531358).
The protein lacks nuclear localization signals. This ARBA prediction appears to be
erroneous.
supported_by:
- reference_id: PMID:10531358
supporting_text: "Immunofluorescence analysis of ectopic MrdgBbeta showed cytoplasmic staining"
- term:
id: GO:0005737
label: cytoplasm
evidence_type: IEA
original_reference_id: GO_REF:0000117
review:
summary: >-
IEA annotation from ARBA machine learning for cytoplasmic localization. Redundant
with the IBA annotation for the same term but independently correct.
action: ACCEPT
reason: >-
Correct annotation consistent with experimental evidence from the human ortholog
and absence of membrane-targeting signals. Redundant with the IBA annotation above.
supported_by:
- reference_id: PMID:10531358
supporting_text: "Immunofluorescence analysis of ectopic MrdgBbeta showed cytoplasmic staining"
- term:
id: GO:0015914
label: phospholipid transport
evidence_type: IEA
original_reference_id: GO_REF:0000002
review:
summary: >-
IEA annotation from InterPro domain mapping (IPR001666 PITP family). Phospholipid
transport is the core biological process for this protein family.
action: ACCEPT
reason: >-
Phospholipid transport is the direct biological process consequence of PI transfer
activity. The PITP domain mediates monomeric lipid transfer between membranes.
This is well-supported by the domain architecture and ortholog biochemistry.
supported_by:
- reference_id: PMID:22822086
supporting_text: "RdgBβ, when containing PA, regulates an effector protein or can facilitate lipid transfer between membrane compartments"
- term:
id: GO:0005737
label: cytoplasm
evidence_type: ISS
original_reference_id: GO_REF:0000024
review:
summary: >-
ISS annotation by manual curator judgment based on sequence similarity to human
PITPNC1 (Q9UKF7). Cytoplasmic localization is correct and well-supported.
action: ACCEPT
reason: >-
Based on sequence similarity to human PITPNC1, which was experimentally shown to
localize to the cytoplasm (PMID:10531358). This is the strongest non-experimental
annotation for localization.
supported_by:
- reference_id: PMID:10531358
supporting_text: "Immunofluorescence analysis of ectopic MrdgBbeta showed cytoplasmic staining"
- term:
id: GO:0007165
label: signal transduction
evidence_type: ISS
original_reference_id: GO_REF:0000024
review:
summary: >-
ISS annotation based on sequence similarity to human PITPNC1 (Q9UKF7). Signal
transduction is a plausible but very broad annotation. PITPs can influence
signaling by supplying PI for phosphoinositide synthesis, and the human ortholog
interacts with signaling components (ATRAP, 14-3-3).
action: KEEP_AS_NON_CORE
reason: >-
Signal transduction is plausible but overly broad. PITPNC1 participates in
phospholipase D signaling by binding PA (PMID:22822086) and interacts with ATRAP
in the angiotensin signaling pathway (PMID:21728994). However, the specific
signaling pathways for Drosophila rdgBbeta are unknown. The annotation is not
wrong but is too general to be informative about core function.
supported_by:
- reference_id: PMID:22822086
supporting_text: "RdgBβ, when containing PA, regulates an effector protein or can facilitate lipid transfer between membrane compartments"
- reference_id: PMID:21728994
supporting_text: "the PITP domain of RdgBβ interacts with the integral membrane protein ATRAP"
- term:
id: GO:0008526
label: phosphatidylinositol transfer activity
evidence_type: ISS
original_reference_id: GO_REF:0000024
review:
summary: >-
ISS annotation based on sequence similarity to human PITPNC1 (Q9UKF7). PI transfer
activity is the core molecular function, well-established for the ortholog.
action: ACCEPT
reason: >-
Redundant with the IBA annotation for the same term. PI transfer activity is the
defining function of this protein family and is directly demonstrated for the human
ortholog (PMID:10531358, PMID:22822086).
supported_by:
- reference_id: PMID:10531358
supporting_text: "the ability of recombinant MrdgBbeta to transfer phosphatidylinositol in vitro was similar to other PITP-like domains"
references:
- id: GO_REF:0000002
title: Gene Ontology annotation through association of InterPro records with GO terms
findings:
- statement: PITP family domain (IPR001666) associated with phospholipid transport
- id: GO_REF:0000024
title: Manual transfer of experimentally-verified manual GO annotation data to orthologs by curator judgment of sequence similarity
findings:
- statement: Annotations transferred from human PITPNC1 (Q9UKF7)
- id: GO_REF:0000033
title: Annotation inferences using phylogenetic trees
findings:
- statement: Phylogenetic inference of PITP family conserved functions
- id: GO_REF:0000117
title: Electronic Gene Ontology annotations created by ARBA machine learning models
findings:
- statement: Machine learning predictions for phospholipid binding, cytoplasm, and nucleus localization
- id: PMID:10531358
title: "Cloning and characterization of a novel human phosphatidylinositol transfer protein, rdgBbeta."
findings:
- statement: Human rdgBbeta (PITPNC1) cloned and characterized as a novel PITP lacking transmembrane domains
supporting_text: "In contrast to other rdgB-like proteins, MrdgBbeta contains no transmembrane motifs or the conserved carboxyl-terminal domain"
- statement: Cytoplasmic localization demonstrated by immunofluorescence
supporting_text: "Immunofluorescence analysis of ectopic MrdgBbeta showed cytoplasmic staining"
- statement: PI transfer activity demonstrated in vitro
supporting_text: "the ability of recombinant MrdgBbeta to transfer phosphatidylinositol in vitro was similar to other PITP-like domains"
- statement: Ubiquitous expression with highest levels in heart, muscle, kidney, liver, and leukocytes
supporting_text: "MrdgBbeta mRNA is ubiquitously expressed"
- id: PMID:22822086
title: "Phosphatidylinositol transfer protein, cytoplasmic 1 (PITPNC1) binds and transfers phosphatidic acid."
findings:
- statement: PITPNC1/RdgBbeta binds and transfers both PI and phosphatidic acid (PA)
supporting_text: "besides PI, RdgBβ binds and transfers phosphatidic acid (PA) but hardly binds phosphatidylcholine"
- statement: Does not significantly bind phosphatidylcholine, unlike Class I PITPs
supporting_text: "the lipid binding properties of this protein are distinct to Class I PITPs because, besides PI, RdgBβ binds and transfers phosphatidic acid (PA) but hardly binds phosphatidylcholine"
- statement: Purified RdgBbeta preloaded with PA and phosphatidylglycerol from E. coli
supporting_text: "RdgBβ when purified from Escherichia coli is preloaded with PA and phosphatidylglycerol"
- statement: PA binding increases at the expense of PI binding when phospholipase D is activated
supporting_text: "After an increase in PA levels following activation of endogenous phospholipase D...binding of PA to RdgBβ was greater at the expense of PI binding"
- id: PMID:21728994
title: "The phosphatidylinositol transfer protein RdgBβ binds 14-3-3 via its unstructured C-terminus, whereas its lipid-binding domain interacts with the integral membrane protein ATRAP (angiotensin II type I receptor-associated protein)."
findings:
- statement: RdgBbeta C-terminus binds 14-3-3 at two phosphorylated sites
supporting_text: "the C-terminus contains two tandem phosphorylated binding sites (Ser(274) and Ser(299)) for 14-3-3"
- statement: 14-3-3 binding shields PEST sequences and stabilizes the protein
supporting_text: "The C-terminus also contains PEST sequences which are shielded by 14-3-3 binding"
- statement: PITP domain interacts with ATRAP to recruit RdgBbeta to membranes
supporting_text: "the PITP domain of RdgBβ interacts with the integral membrane protein ATRAP"
- statement: RdgBbeta is degraded by proteasome with half-life of 4h; faster without 14-3-3 binding
supporting_text: "RdgBβ is degraded with a half-life of 4 h following ubiquitination via the proteasome"
- id: PMID:26977884
title: "PITPNC1 Recruits RAB1B to the Golgi Network to Drive Malignant Secretion."
findings:
- statement: PITPNC1 amplified in breast cancer and overexpressed in metastatic cancers
supporting_text: "PITPNC1 as a gene amplified in a large fraction of human breast cancer and overexpressed in metastatic breast, melanoma, and colon cancers"
- statement: PITPNC1 binds Golgi-resident PI4P and recruits RAB1B to the Golgi
supporting_text: "PITPNC1 promotes malignant secretion by binding Golgi-resident PI4P and localizing RAB1B to the Golgi"
- statement: Drives secretion of pro-invasive and pro-angiogenic factors
supporting_text: "PITPNC1-mediated vesicular release drives metastasis by increasing the secretion of pro-invasive and pro-angiogenic mediators HTRA1, MMP1, FAM3C, PDGFA, and ADAM10"
- id: file:DROME/rdgBbeta/rdgBbeta-notes.md
title: Research notes on rdgBbeta gene review
findings:
- statement: rdgBbeta alleles (EP2360, G8057) ameliorate ALS-like phenotype in Vap33-P58S model
- statement: BioReason prediction incorrectly attributes Class I PITP functions (cytokinesis, PC binding) to this Class IIB PITP
core_functions:
- description: >-
rdgBbeta is a soluble cytoplasmic lipid transfer protein that shuttles phosphatidylinositol
and phosphatidic acid between membranes. It uses its PITP/START-like hydrophobic cavity to
extract lipid monomers from donor membranes and deliver them to acceptor membranes. Unlike
Class I PITPs, it does not significantly bind phosphatidylcholine. Through PI and PA transfer,
it likely contributes to phosphoinositide homeostasis and lipid signaling.
molecular_function:
id: GO:0008526
label: phosphatidylinositol transfer activity
directly_involved_in:
- id: GO:0015914
label: phospholipid transport
locations:
- id: GO:0005737
label: cytoplasm
proposed_new_terms: []
suggested_questions:
- question: >-
What is the in vivo function of rdgBbeta in Drosophila? No mutant phenotype has been
characterized beyond genetic interaction with Vap33-P58S. Does rdgBbeta loss-of-function
produce any developmental or physiological phenotypes?
- question: >-
Does Drosophila rdgBbeta also bind and transfer phosphatidic acid, as demonstrated for
the human ortholog PITPNC1? What are the specific lipid species it handles in vivo?
- question: >-
How does rdgBbeta modify the Vap33-P58S ALS phenotype? Does it act by modulating
phosphoinositide levels, or through an independent mechanism?
suggested_experiments:
- description: >-
Generate rdgBbeta null mutants using CRISPR and characterize phenotypes in development,
neuronal function, and lipid metabolism. Assess phosphoinositide and PA levels in mutant tissue.
hypothesis: >-
rdgBbeta loss-of-function may cause subtle defects in lipid homeostasis that become
apparent under stress conditions or in sensitized genetic backgrounds.
- description: >-
Perform in vitro lipid binding and transfer assays with purified Drosophila rdgBbeta
to determine whether it shares the PI/PA dual specificity of human PITPNC1.
hypothesis: >-
Drosophila rdgBbeta will show PI and PA binding/transfer activity comparable to
human PITPNC1, but minimal PC binding, consistent with Class IIB PITP specificity.
- description: >-
Characterize the genetic interaction between rdgBbeta and Vap33-P58S in detail,
measuring phosphoinositide levels and neurodegeneration markers.
hypothesis: >-
rdgBbeta alleles rescue Vap33-P58S by reducing phosphoinositide delivery to
membranes, partially compensating for the loss of Sac1-mediated PI dephosphorylation.