X-linked spinobulbar muscular atrophy (SBMA, Kennedy disease, OMIM no. 313200) is an adult-onset motor neuronopathy that typically causes slowly progressive, symmetric wasting and weakness, initially of the proximal muscles of the hip and shoulder. Muscle cramps, hand tremors, and fasciculations are often associated. Eventually, motor nuclei of the brain stem become involved, leading to speech and swallowing difficulties. Male hypogonadism, usually represented by gynecomastia and testicular atrophy and attributable to mild androgen insensitivity, is not infrequent. Variable expressivity is prominent.
SBMA is the archetype and prototype of the class of Mendelian, adult-onset neuronodegenerative diseases caused by proteins with expanded polyglutamine (polyGln) tracts. It is archetypical for two reasons: (a) it was the first member of the class to be identified when it was discovered that exon 1 of the androgen receptor gene ( AR ) in SBMA patients had an expanded (CAG) n tract, and (b) it was the first member of the class in which the normal functions of the culpable protein were well known. Since subjects with complete androgen insensitivity, including those with complete AR deletions, do not develop SBMA, this knowledge mandated the logic that the polyCAG-expanded AR or the polyGln-expanded AR protein is somehow motor neuronotoxic by a gain, not a loss, of function.
SBMA is prototypical of the class for several reasons: (a) it is now clear that the polyGln-expanded AR protein, and probably the expanded polyGln tracts themselves, with certain flanking segments of the various parental proteins, are the essential pathogenetic agents in this class of neuronodegenerative disease, (b) the preferential death of certain motor neurons in SBMA exemplifies the fact that selective death of certain neuronal populations is responsible for the clinical diversity among the known diseases in the class, (c) there is extremely variable clinical expressivity (age of onset, rate and pattern of progression) of SBMA, and compared with other members of the class, relatively little of this variability is attributable to different degrees of polyGln expansion generated by meiotic or mitotic instability of (CAG) n -repeat size, and (d) SBMA illustrates that expansion to beyond approximately 38 repeats endows a polyGln tract with a threshold property (or more than one) that is selectively lethal to certain neurons. The biochemical, histopathologic, and neurophysiologic features of SBMA are, unremarkably, those secondary to motor denervation. Endocrine laboratory evidence of mild androgen insensitivity (normal or elevated serum testosterone and luteinizing hormone levels) may or may not be found.
The mild androgen insensitivity component of SBMA is clearly due to a loss of function by the polyGln-expanded AR protein. The loss may be attributable to one or more of the following: decreased steady-state levels of AR, decreased androgen-binding affinity of the mutant AR, or decreased transcriptional regulatory competence of the mutant AR. These factors may contribute to different extents in various androgen target cells. Since androgens are both motor neuronotropic and motor neuronotrophic, it is possible that the polyGln-expanded AR protein loses a function that is necessary, but not sufficient, for the motor neuronopathy of SBMA generally or for death of certain motor neurons specifically.
As of 2002, data bearing on the pathogenesis of SBMA, and of the polyGln-expanded neuronopathies as a group, are accumulating rapidly. These data are being generated by research with the following underlying themes: (a) polyGln expansion alters the folding of a parental protein and therefore its sensitivity to proteases, proteosomes, or chaperones, and, consequently, the nature or extent of the polyGln-expanded fragments derived from it, (b) these polyGln-expanded fragments oligomerize, directly or through intermediates, either noncovalently (by hydrogen-bonded polar zippering) or covalently (by transglutaminase-catalyzed isodipeptide formation), to yield inclusions (aggregates) that accumulate in or around the nucleus of certain neurons, (c) sequestration of critical factors alters transcriptional regulation, and (d) the entire process is slow, so it takes years to see the clinical consequences. Why neurons are selectively vulnerable to the toxicity of polyGln-expanded proteins when these proteins (e.g., the AR) are widely distributed in many nonneuronal cells and why, as in SBMA, only certain motor neurons are affected remain entirely speculative.
It is reasonable to postulate that cells that are spared are “protected” in one or more ways. Such protection may be due to the paucity of a particular protease (perhaps one of the caspases that become active during apoptosis), to the abundance of certain chaperones (either proteins or small molecules) that ensure the proper proteolysis of certain polyGln-expanded fragments and thus avert their pathogenicity, or to other defense mechanisms that flow logically from the aforementioned current research themes.
Such pathogenetic concepts favor the development of certain approaches to the prophylaxis of SBMA. These include inhibitors of transglutaminases, caspases, or histone deacetylases and potentiators of proteosomes or chaperones. Neither androgen nor antiandrogen treatment is indicated; the therapeutic potential of androgen withdrawal must be thoroughly tested in animal models.