linc-mipep and linc-wrb encode micropeptides that regulate chromatin accessibility in vertebrate-specific neural cells

Thousands of long intergenic non-coding RNAs (lincRNAs) are transcribed throughout the vertebrate genome. A subset of lincRNAs enriched in developing brains have recently been found to contain cryptic open reading frames and are speculated to encode micropeptides. However, systematic identification and functional assessment of these transcripts have been hindered by technical challenges caused by their small size. Here we show that two putative lincRNAs (linc-mipep, also called lnc-rps25, and linc-wrb) encode micropeptides with homology to the vertebrate-specific chromatin architectural protein, Hmgn1, and demonstrate that they are required for development of vertebrate-specific brain cell types. Specifically, we show that NMDA receptor-mediated pathways are dysregulated in zebrafish lacking these micropeptides and that their loss preferentially alters the gene regulatory networks that establish cerebellar cells and oligodendrocytes - evolutionarily newer cell types that develop postnatally in humans. These findings reveal a key missing link in the evolution of vertebrate brain cell development and illustrate a genetic basis for how some neural cell types are more susceptible to chromatin disruptions, with implications for neurodevelopmental disorders and disease. Flash-frozen pooled brains were prepared based on Protocol CG000366 – Rev D (Protocol 2) from 10x Genomics (available at https://www.10xgenomics.com/support/single-cell-multiome-atac-plus-gene-expression/documentation/steps/sample-prep/nuclei-isolation-from-embryonic-mouse-brain-tissue-for-single-cell-multiome-atac-plus-gene-expression-sequencing). It is critical to keep samples cold and/or on ice for all steps. Briefly, all samples were processed identically and simultaneously, to minimize batch effects. Chilled 0.1X Lysis Buffer (500μl) was immediately added to frozen samples, and samples were homogenized using a glass dounce tissue grinder with glass pestle. Samples were incubated on ice for 5 minutes, gently pipetted 10x, then incubated again for 5 minutes. Chilled Wash Buffer (500μl) was gently added to samples. After pipetting the mix 5x, the samples were passed through 70μm-porosity Flowmi tips into new ice-cold low-bind 1.5ml tubes. Each suspension was subsequently passed through a 40μm-porosity Flowmi tip into a new ice-cold low-bind 1.5ml tube. Samples were centrifuged at 500 rcf 5 minutes at 4°C. The supernatant was gently removed, without disturbing the nuclei pellet. Chilled Wash Buffer (1ml) was added, and the nuclei were gently resuspended 5x. This wash and resuspension step was repeated one more time. On the final step, nuclei were resuspended in Diluted Nuclei Buffer. Quality and number of nuclei (as assessed by >90% Trypan Blue staining) for each sample was assessed using a hemocytometer, and were immediately used for tagmentation step using the 10x Genomics platform.