The molecular basis of spinal muscular atrophy (SMA), an autosomal recessive neuromuscular disorder, is the homozygous loss of the survival motor neuron gene 1 (SMN1). A nearly identical copy of the SMN1 gene, called SMN2, modulates the disease severity. The functional difference between both genes is a translationally silent mutation that, however, disrupts an exonic splicing enhancer causing exon 7 skipping in most SMN2 transcripts. Only 10% of SMN2 transcripts encode functional full-length protein identical to SMN1. Transcriptional activation, facilitation of correct SMN2 splicing, or stabilization of the protein are considered as strategies for SMA therapy. Among various drugs, histone deacetylase inhibitors such as valproic acid (VPA) or 4-phenylbutyrate (PBA) have been shown to increase SMN2-derived RNA and protein levels. Recently, in vivo activation of the SMN gene was shown in VPA-treated SMA patients and carriers. Clinical trials are underway to investigate the effect of VPA and PBA on motor function in SMA patients.