New work from Gideon Dreyfuss and colleagues from the University of Pennsylvania in Philadelphia gives a clue. In today’s Cell, Dreyfuss and coworkers report evidence that the absence of SMN leads to widespread defects in messenger RNA (mRNA) splicing. The splicing mistakes occur in all tissues in SMN-deficient mice, but the exact genes affected are different in each tissue. Cell type-specific changes in gene expression, as a result of defective mRNA processing, could possibly explain the distinctive loss of motor neurons in SMA.

The SMN protein is known to factor in messenger RNA splicing. It forms part of a large protein complex that assembles small nuclear RNA-ribonucleoproteins (snRNPs), which in turn participate in pre-mRNA splicing. First authors Zhenxi Zhang and Francesco Lotti led the way to look at snRNPs and their constituent snRNAs in HeLa cells engineered with an inducible SMN RNAi knockdown. They found that SMN deficiency led to reductions in snRNPs, but surprisingly found little overall change in total levels of their constituent snRNAs. When the researchers looked more closely at eight different snRNAs, they saw that some were modestly decreased, but others were unaffected. Interestingly, they found a different profile of snRNA changes after loss of SMN in motor neurons versus HeLa cells. When they looked in an SMA mouse model featuring a moderate decrease in SMN protein, they found tissue-specific alterations in snRNA repertoires as well.

To look at the effects of these changes on mRNA splicing, the investigators analyzed mRNA from normal and SMA mouse tissues on exon arrays. In spinal cord mRNA, they found 259 genes that showed differential exon expression. From brain, they picked out 73 genes and 633 from kidney that showed possible splicing pattern changes because of loss of SMN. A large-scale RT-PCR analysis of 22 of the genes revealed widespread splicing variations beyond those indicated by the exon array data. Some of the changes looked like alternative splicing, but many others appeared to stem from aberrant reactions.

The splicing mistakes were tissue-specific, and covered many types of genes. For the 30 top genes from each tissue studied, only two genes were found in common. However, the top genes from all tissues included many for transporter proteins, extracellular matrix proteins, and other genes with large messages and many exons, hence many introns to splice.

“These findings reveal a key role for the SMN complex in RNA metabolism and in splicing regulation and indicate that SMA is a general splicing disease that is not restricted to motor neurons,” the authors conclude.

How, then, does the work explain the selective toxicity of SMN deficiency to motor neurons? The answer may lie in the complex regulation of tissue-specific mRNA splicing. One explanation the authors propose is that, rather than the SMN protein itself having a cell-specific function, other cell-specific factors come into play to influence the relative levels of different snRNPs when SMN becomes limiting. That leads to the observed cell-specific splicing defects, which could result in a critical shortage of a protein or proteins required by motor neurons. “The attrition of motor neurons may be caused by one or more aberrant transcripts or by the cumulative effect of many splicing defects,” the authors write. Even changes in neighboring cells could cause or contribute to motor neuron death, they speculate.—Pat McCaffrey.

Reference:
Zhang Z, Lotti F, Dittmar K, Younis I, Wan L, Kasim M, Dreyfuss G. SMN Deficiency Causes Tissue-Specific Perturbations in the Repertoire of snRNAs and Widespread Defects in Splicing. Cell. 2008 May 16;133:585-600. Abstract