The effects of supravalvular aortic stenosis elastin gene mutation on voice production.(Clinical report)

Supravalvular aortic stenosis (SVAS) is an autosomal dominant disease that can occur as a part of Williams syndrome or as an isolated condition. It is caused by loss of function mutations within the elastin (ELN) gene leading to a quantitative reduction of elastic fibers due to abnormal elastogenesis. Elastic fibers are known to be present in normal vocal folds. To test the hypothesis that elastic fibers are necessary for normal physiology of the human vocal fold, we have investigated quantitative characteristics of voice production in six individuals (five adults and one child) with positive elastin mutation status compared to age and gender matched normal controls. The voice in all adult SVAS participants was characterized by a low speaking fundamental frequency and greater spectral tilt. The male SVAS participants, who manifested the greatest degree of SVAS phenotypic expression, also produced a limited physiological frequency range. We conclude that elastin haploinsufficiency uniquely influences vocal production characteristics, highlighting the role of elastin in regulating the pitch and range of human voice by contributing to specific vibrational properties of the vocal fold.

**********

Supravalvular aortic stenosis (SVAS) is a genetic disorder that may occur as an isolated condition (nonsyndromic SVAS) (Eisenberg, Young, Jacobsen, & Boito, 1964) or as part of a related disorder, Williams syndrome (WS; Beuren, Apitz, & Harmjanz, 1962; Williams, Barratt-Boyes, & Lowe, 1961). Isolated SVAS is caused by point mutations (Curren et al., 1993; Li et al., 1997; Tassabehji et al., 1997) and deletions (Olson et al., 1995) in the elastin gene (ELN), while WS is caused by a genomic deletion of 1.5 Mb in chromosome band 7q11.23 that encompasses 24 genes (DeSilva et al., 2002; Tassabehji, 2003), including ELN. Mutations in ELN can result in differing degrees of genetic expression, so that some individuals with the mutation may develop SVAS while some do not. The primary pathophysiological consequence of SVAS is impaired elastogenesis and increased proliferation of smooth-muscle cells (Urban et al., 2002). The result is variable obstructive vascular disease that can include pulmonary, carotid, cerebral, renal, and coronary arteries.

Although there have been no specific clinical reports documenting dysphonia in nonsyndromic SVAS, dysphonia has been consistently reported as a general clinical characteristic of WS (Axelsson, 2005; Cherniske et al., 2004; Duba et al., 2002; Marler, Elfenbein, Ryals, Urban, & Netzloff, 2005; Morris & Mervis, 2000; Stewart, Dalzell, McReid, & Cinnamond, 1993; Vaux, Wojtezak, Benirschke, & Jones, 2003). Research has yet to elucidate the specific structural, acoustic, and auditory-perceptual characteristics of dysphonia in this population, and it is unclear if phonatory irregularities in WS are caused by elastin haploinsufficiency (the hemizygous ELN mutation clinically affects the individual because a single copy of the normal gene is unable to stimulate adequate protein production) or by other mechanisms (e.g., muscular hypotonicity). The principal research focus on arterial morphology and pathology in nonsyndromic SVAS has shown conclusively that obstructive vascular disease in both syndromic and nonsyndromic forms is caused by elastin deficiency (Urban et al., 2002). Due to elastin's role in vocal fold physiology, we suspect that in individuals with SVAS, the amount and structure of vocal fold elastin would be different from unaffected individuals and as a result affect voice quality and vocal abilities. The first part of this supposition (difference in amount and structure of elastin) is supported by recent histological evidence which has demonstrated that vocal fold elastin was both decreased and disorganized in an individual with ELN deletion (Vaux et al., 2003).

The macrostructure of the human vocal fold is comprised of a layered organization, covered with stratified squamous epithelium followed by three layers of lamina propria and, at the deepest level, the vocalis muscle. The concentration of elastin and collagen differs throughout the layers of the lamina propria, and their concentration levels have been used as the basis for identification of the layered vocal fold divisions (Gray, 2000; Volic, Klapan, Seiwerth, & Ibrahimpasic, 2004). Elastin and collagen are the two predominant fibrous proteins found in the vocal folds. In the superficial layer of the lamina propria (SLLP), there are a small number of immature elastic fibers (oxytalan and elaunin) present, along with ground substance consisting of proteoglycans (Hahn, Kobler, Starcher, Zeitels, & Langer, 2006; Hammond, Gray, Butler, Zhou, & Hammond, 1998; Hammond, Zhou, Hammond, Pawlak, & Gray, 1997). The intermediate layer of the lamina propria (ILLP) is characterized by a larger number of elastin fibers, while the deep layer of the lamina propria (DLLP) consists of a greater ratio of collagen to elastin (Gray, 2000; Hammond et al., 1997; Madruga de Melo et al., 2003). The ILLP and DLLP together form the vocal ligament, which anchors the outer epithelium and SLLP to the vocalis muscle.

The cover-body theory of vocal fold vibration has proposed that the layers of the vocal folds act functionally as two layers, a cover (epithelium and SLLP) and a body (vocalis muscle), with the loose cover oscillating over the stiff body during phonation (Hirano & Kakita, 1985). The cover is anchored to the body via the vocal ligament, which serves as a transition zone characterized by a gradient in stiffness: as one moves deeper into the lamina propria, tissue properties become more rigid (Hirano & Kakita, 1985; Madruga de Melo et al., 2003). Contemporary research in voice science has been built upon this well-supported theory (Berke & Gerratt, 1993; Jiang, Lin, & Hanson, 2000; Titze, 1994; Titze & Story, 2002). Histological and physiological investigations of human vocal folds have supported our understanding that the composition of the vocal fold lamina propria, most especially the extracellular matrix, influences vocal fold biomechanics, voice quality, and singing range (Grey, 2000; Hammond et al., 1997; Hartnick, Rehbar, & Prasad, 2005; Sato, Hirano, & Nakashima, 2002).

Acoustic analyses of voice production are often utilized to elucidate characteristics of phonatory function. Numerous publications have reported altered acoustic properties of voice with both organic and functional vocal fold abnormalities (Awan & Roy, 2006; Hartelius, Buder, & Strand, 1997; Linville & Rens, 2001; Yiu, Worrall, Longland, & Mitchell, 2000). Typically, time-based measures of vocal fundamental frequency ([F.sub.o]), variability in [F.sub.o] (referred to as frequency perturbation or jitter), variability in vocal amplitude (referred to ...