# Non-hematopoietic erythropoietin splice variant is produced in the diseased human brain and confers neuroprotection

## TL;DR

A non-hematopoietic variant of erythropoietin (hS3) is produced in the human brain and offers better neuroprotection than standard EPO in neurological diseases.

## Contribution

Discovery of a novel splice variant of EPO (hS3) with enhanced neuroprotective effects and a key functional domain in the D-helix.

## Key findings

- hS3 is upregulated in ischemic and inflammatory neurological diseases but reduced in neurodegenerative diseases.
- hS3 provides superior neuroprotection compared to EPO in an ischemia model using human stem cell-derived neurons.
- The D-helix of hS3 enhances neuroprotection when used as a small peptide (PD29).

## Abstract

Erythropoietin (EPO) is a pleiotropic cytokine with important functions in neuronal development and neuroprotection, but hematopoietic effects limit the therapeutic application of EPO in neurological diseases. We discovered human endogenous EPO splice variants that are non-hematopoietic but cytoprotective. Here, we demonstrate at the single-cell level that an alternative splice variant lacking exon 3 (hS3) is expressed in the human brain and is upregulated above EPO mRNA levels in ischemic and inflammatory neurological diseases. Conversely, hS3 mRNA expression is reduced below EPO levels in neurodegenerative disease. In an oxygen–glucose deprivation (OGD) model of ischemia, a single dose of cell-free synthesized constant glycosylated active hS3 protects neuronal cultures derived from human induced pluripotent stem cells (hiPSC) and human embryonic stem cells (hESC) more effectively than EPO. We identify the D-helix as a key functional domain of hS3 and demonstrate that the neuroprotective effect is enhanced by PD29, a novel small peptide derived from the D-helix of hS3. Long-term hS3 administration increases the neuroprotective effects in the OGD model by dose-dependent differential expression of apoptosis-related protein-coding genes and long non-coding RNAs (lncRNAs). In addition, our results suggest that hS3 induces early cell cycle inhibition without impairing differentiation of hiPSC and hESC into neuronal subtypes. In conclusion, EPO splice variant hS3 is part of the endogenous neuroprotective system in the human brain with significant therapeutic potential.