5.2.2. Potential Genetic Link between the Grange Syndrome and FMD

Recently, Guo and co-workers [44] conducted a whole-exome-sequencing analysis on DNA from two affected siblings of the first reported family affected by Grange syndrome. This is a unique autosomal recessive syndrome, characterized by arterial stenosis that is similar to focal FMD in angiographic appearance and distribution, congenital cardiac defects, brachydactyly, syndactyly, bone fragility and learning disabilities [45]. Using very stringent filtering criteria, and by retaining only loss-of-function mutations, whole exome sequencing led to the identification of two compound heterozygous loss-of-function mutations in the YY1 associated protein 1 gene (*YY1AP1*) in the two siblings [44]. Sanger sequencing confirmed the presence of both *YY1AP1* mutations in the third affected sibling. Interestingly, their mother, who harboured one of the variants, had a stenosis of the proximal renal artery without other syndromic manifestations. Finally, three other unrelated patients with the Grange syndrome were found to harbour distinct *YY1AP1* mutations at the homozygous state.

The *YY1AP1* gene encodes a 88 kDa protein that is part of the INO80 chromatin-remodelling complex, and may play a role in transcriptional regulation and cell cycle control. The authors showed that *YY1AP1* expression is induced during smooth muscle cell differentiation in vitro and that its knockdown prevents the expression of smooth muscle markers and the differentiation of smooth muscle cells. *YY1AP1* loss-of-function mutations may thus lead to an increased presence of proliferating smooth muscle cell precursors and reduced amounts of differentiated quiescent, contractile smooth muscle cells. In view of these elements, the authors hypothesized that heterozygous *YY1AP1* variants may be a rare predisposing allele of FMD [44]. While analysis of whole-exome-sequencing data of 282 patients with FMD and 286 matched controls failed to show an increased burden of *YYAP1* variants in FMD cases vs. controls, a single heterozygous *YY1AP1* frameshift mutation was identified in one out of the 282 cases—a female patient with involvement of both renal and carotid arteries—and none in the controls subjects [44]. Further genetic studies are required to determine the possible contribution of *YYAP1* variants in non-syndromic FMD.

#### *5.3. Non-Hypothesis Driven Genetic Association Study*

After the limited success of whole-exome-sequencing in related FMD patients and poor yield of screening for rare variants [31,44], alternative hypotheses needed to be explored. Using a genetic association study in 1154 patients with (mostly renal artery) FMD and 3895 controls of European ancestry, Kiando and co-workers reported the first consistent and replicated association. They found that allele A of a genetic variant (rs9349379) of the phosphatase and actin regulator 1 gene (*PHACTR1*), highly prevalent in the general population (~60%), was associated with a 40% increase in relative risk of FMD (OR = 1.39, *<sup>p</sup>* < 7.36 × <sup>10</sup><sup>−</sup>10) [46]. Moreover, in 2458 healthy volunteers, the authors documented an association between the *PHACTR1* at-risk allele and increased carotid intima-media thickness (IMT) (*<sup>p</sup>* < 1.65 × <sup>10</sup><sup>−</sup>4) and wall-to-lumen ratio (*<sup>p</sup>* = 0.002) [46]. Furthermore, while expression of *PHACTR1* did not differ in cultured human fibroblasts from 51 patients with FMD compared to 39 controls, stratification by rs9349379 genotype indicated increased expression in subjects with the at-risk allele, rs9349379[A] (*p* < 0.003) [46]. Finally, *zPhactr1* knockdown was associated with a mild (~8%) but significant (*p* < 0.05) vessel dilatation in zebrafish [46].

Already in 2015, the Cervical Artery Dissections and Ischemic Stroke Patients (CADISP) study, a consortium dedicated to the genetics of cervical artery dissection (CeAD), had identified *PHACTR1* as a risk locus for stroke in young adults. In particular, the rs9349379[A] allele was associated with a higher CeAD risk (OR = 1.33, *<sup>p</sup>* = 4.46 × <sup>10</sup>−10) [47]. Furthermore, in a meta-analysis of

29 genome-wide association studies (GWAS), including a total of 23285 individuals with migraine and 95,425 population-matched controls, rs9349379[A] was associated with an increased risk of migraine without aura (*<sup>p</sup>* < 2.81 × <sup>10</sup>−10) [48], clinical presentation shared by FMD and CeAD. These findings are of interest in view of the overlap between FMD, carotid artery dissection and migraine. Indeed, in a sample of 732 patients with CeAD, the prevalence of documented FMD was 5.6%, with a significant difference between patients with multiple and single CeAD (15% vs. 3.8%, *p* < 0.0001) [49]. Furthermore, headaches are a common symptom in patients with FMD (60% of patients in the US registry, with classical migraine-type headaches reported in 32%) [3]. More intriguingly, in a meta-analysis of four large genome-wide association studies including 15420 individuals with coronary artery disease (CAD) and 15,062 controls, the same allele, rs9349379[A], was associated with a decreased risk of CAD (*<sup>p</sup>* = 5.8 × <sup>10</sup>−19) [50]. The associations of the rs9349379 locus with cardio- and cerebrovascular diseases are summarized in Figure 2.

**Figure 2.** Association of rs9349379 locus with PHACTR1 and Endothelin-1 expression, cardio- and cerebrovascular diseases [46,51].
