Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

In-depth proteomic analysis and structural modelling of protein-protein interactions identify known and novel molecular binding partners for EMP3 in erythroid cells

Alattar, A G LU orcid ; Ghosh, S LU orcid ; Chouhan, B LU ; Pimkova, K LU ; Atkinson, G LU ; Flygare, J LU ; Hansson, J LU orcid ; Storry, J LU and Olsson, Martin L LU orcid (2024) In Vox Sanguinis 119(S1).
Abstract
Background: The absence of Epithelial Membrane Protein 3 (EMP3) on red blood cells (RBCs) underlies the MAM-negative (MAM−) phenotype. EMP3 is a ubiquitous protein found at low levels in most tissues but with higher expression in hematopoietic cells. Low levels of CD44 are a signature of MAM− RBCs due its proposed physical interaction with EMP3, indicating that EMP3 stabilises CD44 in the membrane (Thornton, Nature Communications, 2020). Erythroid culture of CD34+ cells from peripheral blood from MAM− individuals showed an unexpected expansion of erythroid cells, suggesting that EMP3 acts as a brake on the erythropoietic pathway. However, the function of EMP3 in erythroid cells is yet to be elucidated.

Aims: To enhance our... (More)
Background: The absence of Epithelial Membrane Protein 3 (EMP3) on red blood cells (RBCs) underlies the MAM-negative (MAM−) phenotype. EMP3 is a ubiquitous protein found at low levels in most tissues but with higher expression in hematopoietic cells. Low levels of CD44 are a signature of MAM− RBCs due its proposed physical interaction with EMP3, indicating that EMP3 stabilises CD44 in the membrane (Thornton, Nature Communications, 2020). Erythroid culture of CD34+ cells from peripheral blood from MAM− individuals showed an unexpected expansion of erythroid cells, suggesting that EMP3 acts as a brake on the erythropoietic pathway. However, the function of EMP3 in erythroid cells is yet to be elucidated.

Aims: To enhance our understanding of the function of EMP3 by investigating its molecular partners in erythroid cells, including but not limited to CD44.

Methods: Global proteomics were performed on three examples of MAM− RBCs from our in-house collection of rare reagent RBCs and compared to three anonymized MAM-positive (MAM+) samples. Flow cytometric analyses were performed on RBCs and cultured cells. In silico analysis of the expression differences of blood group proteins incl. EMP3 was performed in a dataset from CD44 overexpression in HEK293 cells. 3xFLAG-tagged EMP3 (fused to either the N- or C-terminus) was overexpressed in erythroid cells cultured in vitro from CD34+ hematopoietic stem and progenitor cells and used in co-immunoprecipitation (Co-IP) experiments with anti-FLAG-conjugated beads. A library of all detected proteins during erythroid maturation, reticulocytes and RBCs was compiled. This library was used for in silico co-folding analysis of EMP3 with the entire spectrum of erythroid proteins utilizing AlphaFold protein co-folding prediction technology.

Results: Using comprehensive proteomics analysis of MAM− versus MAM+ RBC samples, we confirmed that CD44 is one of the significantly down-regulated proteins in the RBC proteome. In addition, we identified numerous other proteins that are significantly up- or downregulated in MAM− RBCs. When analysing RBCs of MAM+ individuals using flow cytometry, we found that CD44 is highly correlated to MAM antigen expression levels on RBCs. EMP3-mRNA was significantly upregulated in HEK293 cells when CD44 is overexpressed. The co-IP experiments identified multiple potential EMP3 binding partners, including CD44, LGALS3, LAPTM5 and TFRC. Certain proteins were preferentially precipitated with N-terminally FLAG-tagged EMP3, for example, CD36, indicating interactions at the EMP3 C-terminus. Utilizing AlphaFold2 we found a high co-folding score between EMP3 and CD44 mediated by salt bridges between EMP3 (residues Glu35; Asp42) and CD44 (Arg643). Furthermore, despite stringent conditions, EMP3 was predicted to interact with several other erythroid proteins. However, only two candidate proteins overlapped between the two principal experimental approaches (co-IP and two AlphaFold2 prediction scores), CD44 and ENO1, both expressed early in erythropoiesis.

Summary / Conclusions: We confirm that EMP3 influences CD44 expression in erythroid cells and predict the precise site of protein-protein interaction. Utilizing different in vitro and in silico experimental approaches, we also found that EMP3 interacts with proteins expressed during early stages of development, suggesting potentially important roles in erythroid biology. (Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
Vox Sanguinis
volume
119
issue
S1
article number
PA26-L06
publisher
Wiley-Blackwell
external identifiers
  • pmid:38922723
  • scopus:85197191009
ISSN
1423-0410
DOI
10.1111/vox.13652
language
English
LU publication?
yes
id
4f21434f-ab46-4ab5-88f6-365bece2d6e8
date added to LUP
2024-08-23 13:22:10
date last changed
2024-08-24 04:03:12
@misc{4f21434f-ab46-4ab5-88f6-365bece2d6e8,
  abstract     = {{Background: The absence of Epithelial Membrane Protein 3 (EMP3) on red blood cells (RBCs) underlies the MAM-negative (MAM−) phenotype. EMP3 is a ubiquitous protein found at low levels in most tissues but with higher expression in hematopoietic cells. Low levels of CD44 are a signature of MAM− RBCs due its proposed physical interaction with EMP3, indicating that EMP3 stabilises CD44 in the membrane (Thornton, Nature Communications, 2020). Erythroid culture of CD34+ cells from peripheral blood from MAM− individuals showed an unexpected expansion of erythroid cells, suggesting that EMP3 acts as a brake on the erythropoietic pathway. However, the function of EMP3 in erythroid cells is yet to be elucidated.<br/><br/>Aims: To enhance our understanding of the function of EMP3 by investigating its molecular partners in erythroid cells, including but not limited to CD44.<br/><br/>Methods: Global proteomics were performed on three examples of MAM− RBCs from our in-house collection of rare reagent RBCs and compared to three anonymized MAM-positive (MAM+) samples. Flow cytometric analyses were performed on RBCs and cultured cells. In silico analysis of the expression differences of blood group proteins incl. EMP3 was performed in a dataset from CD44 overexpression in HEK293 cells. 3xFLAG-tagged EMP3 (fused to either the N- or C-terminus) was overexpressed in erythroid cells cultured in vitro from CD34+ hematopoietic stem and progenitor cells and used in co-immunoprecipitation (Co-IP) experiments with anti-FLAG-conjugated beads. A library of all detected proteins during erythroid maturation, reticulocytes and RBCs was compiled. This library was used for in silico co-folding analysis of EMP3 with the entire spectrum of erythroid proteins utilizing AlphaFold protein co-folding prediction technology.<br/><br/>Results: Using comprehensive proteomics analysis of MAM− versus MAM+ RBC samples, we confirmed that CD44 is one of the significantly down-regulated proteins in the RBC proteome. In addition, we identified numerous other proteins that are significantly up- or downregulated in MAM− RBCs. When analysing RBCs of MAM+ individuals using flow cytometry, we found that CD44 is highly correlated to MAM antigen expression levels on RBCs. EMP3-mRNA was significantly upregulated in HEK293 cells when CD44 is overexpressed. The co-IP experiments identified multiple potential EMP3 binding partners, including CD44, LGALS3, LAPTM5 and TFRC. Certain proteins were preferentially precipitated with N-terminally FLAG-tagged EMP3, for example, CD36, indicating interactions at the EMP3 C-terminus. Utilizing AlphaFold2 we found a high co-folding score between EMP3 and CD44 mediated by salt bridges between EMP3 (residues Glu35; Asp42) and CD44 (Arg643). Furthermore, despite stringent conditions, EMP3 was predicted to interact with several other erythroid proteins. However, only two candidate proteins overlapped between the two principal experimental approaches (co-IP and two AlphaFold2 prediction scores), CD44 and ENO1, both expressed early in erythropoiesis.<br/><br/>Summary / Conclusions: We confirm that EMP3 influences CD44 expression in erythroid cells and predict the precise site of protein-protein interaction. Utilizing different in vitro and in silico experimental approaches, we also found that EMP3 interacts with proteins expressed during early stages of development, suggesting potentially important roles in erythroid biology.}},
  author       = {{Alattar, A G and Ghosh, S and Chouhan, B and Pimkova, K and Atkinson, G and Flygare, J and Hansson, J and Storry, J and Olsson, Martin L}},
  issn         = {{1423-0410}},
  language     = {{eng}},
  month        = {{06}},
  note         = {{Conference Abstract}},
  number       = {{S1}},
  publisher    = {{Wiley-Blackwell}},
  series       = {{Vox Sanguinis}},
  title        = {{In-depth proteomic analysis and structural modelling of protein-protein interactions identify known and novel molecular binding partners for EMP3 in erythroid cells}},
  url          = {{http://dx.doi.org/10.1111/vox.13652}},
  doi          = {{10.1111/vox.13652}},
  volume       = {{119}},
  year         = {{2024}},
}