Familial hypercholesterolemia (FH) is a monogenic disorder of the metabolism of low-density lipoprotein cholesterol (LDL-C) and is characterized by lifelong elevated levels of LDL particles and LDL-C arterial deposits (Khachadurian, 1964; Brown and Goldstein, 1974). Untreated patients carry the risk of early onset coronary artery disease (premature CAD) and the augmented risk of cardiovascular events (Nordestgaard et al., 2013; Cuchel et al., 2014b; Gidding et al., 2015; Krogh et al., 2016). The current prevalence of FH is estimated to be 1:311 individuals in the general population (Beheshti et al., 2020; Hu et al., 2020).

This monogenic disorder is inherited in two forms of autosomal dominant and autosomal recessive (De Castro-Orós et al., 2010; Cuchel et al., 2014a). The autosomal dominant form of FH is mostly due to the heterozygous and homozygous states of pathogenic variants in the low-density lipoprotein receptor (LDLR), apolipoprotein B (APOB), and proprotein convertase subtilisin/kexin type 9 (PCSK9) genes. Between 85 and 90% of the mutations harbored in LDLR that result in defective LDL receptors attenuate LDL-C clearance from the blood and, consequently, raise LDL-C plasma levels (Varret and Rabès, 2012).

Likewise, mutations in the APOB gene weaken the binding of LDL to LDL receptors, and gain-of-function mutations in the PCSK9 gene account for 5–15% of cases with FH (El Khoury et al., 2017; Sharifi et al., 2017). PCSK9, a serine protease, mediates the internalization and degradation of LDLR in the lysosome via binding to LDL receptors, leading to diminished LDLR recycling and impaired clearance of LDL-C particles from the plasma (Seidah, 2017). Recently, next-generation sequencing has demonstrated the FH phenotype due to occasional dominant mutations in apolipoprotein E (APOE) or signal-transducing adaptor family member 1 (STAP1), as well (Nordestgaard et al., 2013; Cuchel et al., 2014b; Fouchier et al., 2014; Berberich and Hegele, 2019).

The autosomal recessive form of FH is frequently in consequence of pathogenic mutations in the low-density lipoprotein receptor adaptor protein (LDLRAP) gene (Cuchel et al., 2014a). Nevertheless, whole-exome sequencing has revealed that mutations in the lysosomal acid lipase A (LIPA), ATP-binding cassette sub-family G member 5 and 8 (ABCG5/8), and cholesterol 7 alpha-hydroxylase (CYP7A1) genes phenotypically cause hypercholesterolemia similar to FH in the recessive status (Brautbar et al., 2015; Berberich and Hegele, 2019).

At present, the diagnosis of FH is frequently based on clinical phenotypes, with the most extensively utilized FH clinical criteria being the Simon Broome register (SBR) (Simon Broome Register Group., 1991) and the Dutch Lipid Clinic Network System (DLCNS) (Nordestgaard et al., 2013). In this study, we analyzed the possible mutations of the LDLR, APOB, and PCSK9 genes in Iranian patients suffering from both FH and premature CAD.

Frequent LDLR, APOB, and PCSK9 gene mutations have been described in patients suffering from hypercholesterolemia; however, genetic data on the Iranian population are still limited (Jensen et al., 1996; Fardesfahani et al., 2005; Fardesfahani and Khatami, 2010; Farrokhi et al., 2011; Fairoozy et al., 2017; Ekrami et al., 2018; Nikkhooy et al., 2018; Tajamolian et al., 2018).

In this study, we found eight different mutations among 15 patients with clinically diagnosed FH and premature CAD. Among the study population, one patient had double mutations in both LDLR and PCSK9 genes. The present study is the first investigation to report the c.1014C > G mutation. With the exception of the c.1436T > A mutation, which has been previously reported in Iran (Fairoozy et al., 2017), all the other mutations have already been reported in populations other than Iranians. Among seven mutations in the LDLR gene, only four nucleotide variants are reported in the UCL-FH database (Table 2), and none of the pathogenic mutations can be found in 1000Genome, gnomAD, and Iranome databases (Table 2).

Previous research on the Iranian population (Jensen et al., 1996; Fardesfahani et al., 2005; Fardesfahani and Khatami, 2010; Khan et al., 2011; Fairoozy et al., 2017; Ekrami et al., 2018; Nikkhooy et al., 2018; Tajamolian et al., 2018) has identified 15 different mutations harbored in exons 3, 4, 9, 10, 11, 12, 14, and 17 of the LDLR gene (Supplementary Table 4). In 2017, an investigation drawing upon target next-generation sequencing identified seven different mutations (C.389C > G, c.1436T > A, c.1599G > A, c.1729T > C, c.2001_2002delinsGT, and c.2146dupG) that were not previously reported in the Iranian population (Fairoozy et al., 2017). Two different mutations (c.285C > G and c.415G > C) were found in another study conducted on a sample of the Iranian population (Ekrami et al., 2018). The c.445G > T, c. 1246C > T, c.1478-1479delCT, c.660-661InsCC, c.105A > T, and c.1474G > A mutations have also been reported in different studies (Fardesfahani et al., 2005; Fardesfahani and Khatami, 2010; Khan et al., 2011; Nikkhooy et al., 2018; Tajamolian et al., 2018).

The detected mutations in the LDLR gene in the present study were scattered over exons 4, 6, 7, 10, and 12, and they encoded the functional domain, the ligand-binding domain, and the epidermal growth factor (EGF)–like of the mature LDLR protein (Figure 1). The prevalence of the identified mutations was uppermost in exon 4 (3/7 = 43%) of the LDLR gene (Heath et al., 2001; Villéger et al., 2002). Exon 4, along with exons 5 to 7, of the LDLR gene encodes the ligand-binding domain of the LDLR protein; nevertheless, exon 4 encodes the critical ligand-binding section of the protein, and any changes in the sequence may affect the folding of the protein and change the functionality of the mature protein. The consequences of missense mutations in this exon may lessen the receptor affinity of LDLR toward both APOE and APOB (Esser et al., 1988). Our data revealed c.514C > G, c.443G > A, and c.647G > A mutations in exon 4; none of these mutations has been previously reported among Iranians. Based on the previously detected mutations in the Iranian population, 5 of the 15 (33.3%) identified mutations occurred in exon 4 of LDLR (Supplementary Table 4). Accumulating evidence has demonstrated that individuals carrying mutations in exon 4 have a more severe FH phenotype and congenital heart disease risks; mutations in this region are therefore predominantly deleterious to receptor function (Gudnason et al., 1994; Graham et al., 1999; Humphries et al., 2006). The c.443G > A and c.647G > A mutations have both been described in Germany, the United Kingdom, and Spain (Nauck et al., 2001; Dedoussis et al., 2004; Humphries et al., 2006; Meriño-Ibarra et al., 2007).

The genetic analysis on an 11-year-old girl from the southern Iranian city of Kerman demonstrated a homozygous loss-of-function mutation in the PCSK9 gene (c.158C > T [p.Ala53Val]) and a pathogenic homozygous mutation in the LDLR gene (c.514C > G [p.Asp172His]). The c.158C > T variant in PCSK9 has also been previously reported in South Africa (Thiart et al., 2000). Several studies have revealed that loss-of-function mutations in PCSK9 culminate in low LDL-C levels and lead to a diminished risk of CAD (Benjannet et al., 2012; Seidah, 2017; Bayona et al., 2020). Apropos the effect of loss-of-function PCSK9 variants, some clinical trials have also confirmed that PCSK9 inhibitors lower LDL-C concentrations in the serum (Cohen et al., 2006; Stein et al., 2014; Vuorio et al., 2016; Naeli et al., 2017). Although the index patient carried the homozygous loss-of-function mutation in PCSK9, she presented severe phenotypes such as an uncontrollable cholesterol level; several elbows, knees, and hip xanthomas; and aortic valve disease, which necessitated valve replacement surgery. The reason for the patient’s presentations may have been completely defective LDLR activity owing to homozygous mutations. In addition, she was the offspring of a consanguineous marriage between a couple who presented with moderate hypercholesterolemia but had no cardiovascular diseases. The parents carried both mutations in LDLR and PCSK9 in heterozygous forms. Thus, it seems that the loss-of-function mutation in PCSK9 has a protective effect against the development of the severe forms of phenotypes (Seidah, 2019) in such individuals, notwithstanding the heterozygous mutation in LDLR. The proband had two brothers, who also carried the heterozygous mutation in the LDLR gene. In regard to the PCSK9 gene, the elder brother, similar to the parents, featured a heterozygous mutation, whereas the younger one was a wild type.

To the best of our knowledge, this is the first report on the coincidental inheritance of a pathologic mutation in LDLR (p.Asp172His) and a loss-of-function mutation (p.Ala53Val) in PCSK9 in an Iranian family. The loss-of-function mutation (c.137G > T [p.Arg46Leu]) in PCSK9 and a pathologic mutation in LDLR (c.902A > G [p.Asp301Gly]), both in the heterozygous form, have been previously reported in a Caucasian family (Bayona et al., 2020).

The c.906C > G FH-causing variant (p.Cys302Trp), which was found in this study in exon 6 of LDLR, has also been previously identified in an Iraqi–Turkish patient (Webb et al., 1996). The 40-year-old woman carrying this mutation originated from the north of Iran. Given the close geographical and cultural links between Iraq, Turkey, and Iran, it is probable that the two patients shared some common founder mutations.

The novel mutation found in this study (c.1014C > G), which is located in exon 7 of the LDLR gene, changes cysteine to tryptophan. According to the LOVD database, approximately 3% of the total 3731 variants of LDLR happen on exon 7 (Fokkema et al., 2005). This variant has not been previously identified; still, the same location features another change that has been previously reported (c.1014C > A). This variation, which is a stop-gained mutation, alters cysteine in position 338 to the termination code and creates a truncated protein (Kotzer and Baudhuin, 2009). Exon 7, along with exons 8–12, of the LDLR gene participates in the production of the EGF precursor homology domain of LDLR. This domain of LDLR shares such conserved sequences as cysteine-rich sequences, YWTD repeats, and EGF repeats to EGF (Davis et al., 1987; Springer, 1998). Furthermore, the EGF precursor-like domain comprised two EGF homology domains (EGF-A and EGF-B), dispersed from the third EGF-like domain by a B-propeller domain. Exon 7 of LDLR encodes the EGF-A part of the EGF-like domain of LDLR, which encompasses 40-amino-acid residues and six cysteine amino acids, forming three disulfide bonds (Springer, 1998; Jeon et al., 2001). In the molecular mechanisms of LDLR recycling in hepatocytes, PCSK9 binds to the EGF-A domain of LDLR in a PH- and calcium-dependent manner and through lysosomal degradation decreases the total LDLR level in the liver (Zhang et al., 2007; Bottomley et al., 2009). A deficiency in this region interferes with the release of the binding ligand of the internalized LDLR and inhibits its recycling to the cell surface (Davis et al., 1987). On account of the fact that this missense mutation exists in a cysteine residue, it may result in misfolding and defective protein receptors. In addition, all in silico predictive programs describe this variant as deleterious, probably damaging, and disease causing. The segregation analysis of this variant in the index patient’s sister, who had hypercholesterolemia, confirmed the pathogenicity of this variant as well.

The point mutations (p.leu479Gln and p.G593Afs^∗72) were respectively harbored in exons 10 and 12, which encode different parts of the LDLR EGF precursor-like domain. Exon 10 encodes part of a 5-repeat region of 40–60 amino acids that has a conserved motif (YWTD). This part is also known to play a role in the intracellular trafficking and recycling of LDLR, which explains why mutations in this section are acknowledged to be the cause of transport-defective or recycling-defective proteins (Brown and Goldstein, 1974; Davis et al., 1987). The p.leu479Gln mutation has been previously described in Iran (Fairoozy et al., 2017) and the United Kingdom (Heath et al., 2001). Our segregation analysis of this mutation in the first degree-relatives of the proband (P8) confirmed the pathogenicity of the mutation too. The translation product of the allele carrying the frameshift mutation (p.G593Afs^∗72) leads to the truncation of the proteins of 663 amino acids and is deemed a receptor-negative mutation (Bertolini et al., 2017). With respect to the pathogenicity of this mutation, confirmed via the molecular diagnosis of the variation in the family of the proband (P9), the result of the screening of the family demonstrated that the father carried this mutation in a heterozygous state. It is also probable that the proband’s mother was also heterozygous for this variation. This mutation was originally found in Italian and Northern Ireland populations (Bertolini et al., 2017).

Furthermore, we succeeded in detecting 31 different nucleotide variants in the PCSK9 gene. While most of them have been previously reported, there are some unreported variants scattered in exons, introns, and 3′-UTRs (Supplementary Table 3). The UTR parts (5′ and 3′) include regulatory elements that interact with such different regulatory factors as transcription factors and microRNAs. Altering the nucleotides in the UTRs of genes may interfere with their interactions and causes the misexpression of their targeted gene (Chatterjee and Pal, 2009; Azad et al., 2020). The existing evidence notwithstanding, this hypothesis still requires more functional analyses to be confirmed.

In the present study, we detected none of the common APOB mutations, suggesting that the frequency of APOB mutations in Iran may differ from that in other countries. Further research on this point is warranted.

Among our study population, which comprised 15 subjects with clinically diagnosed FH and premature CAD, we detected eight patients who were negative for FH mutations. Accumulating evidence has shown that, in addition to mutations in the LDLR and PCSK9 genes, mutations in other single or multiple genes can contribute to the creation of the FH phenotype (Smilde et al., 2001; Nordestgaard et al., 2013; Cuchel et al., 2014b; Fouchier et al., 2014; Brautbar et al., 2015; Berberich and Hegele, 2019). The accumulation of common small-effect LDL-C–raising alleles can elevate LDL-C levels and, consequently, cause FH (Talmud et al., 2013). Accordingly, the absence of mutations in the FH-mutation–negative probands in our study may be attributable to nucleotide alterations in other genes, nucleotide variations in other parts of the APOB gene (not addressed in this study), or a combination of LDL-C-raising alleles. Be that as it may, previous studies on other diseases have demonstrated that the presence and combinations of some polymorphisms can render individuals more susceptible to disease (Khajali and Khajali, 2014; Mesic et al., 2019; Pinheiro et al., 2019).

We succeeded in finding several nucleotide variations/polymorphisms in both LDLR and PCSK9 scattered in exons, introns, and 5′–3′ UTRs. These nucleotide variations/polymorphisms could be the combination of different polymorphisms, predisposing FH-mutation–negative individuals to hypercholesterolemia.

The existing literature contains a paucity of information regarding genetic testing on nucleotide alterations in the Iranian population. The finding of seven different mutations in this study, along with other studies on samples of the Iranian population, demonstrates the high degree of genetic heterogeneity and the wide spectrum of mutations in Iranians. This domain undoubtedly requires further in-depth research. In daily practice, patients presenting with premature CAD are rarely screened for the genetic analysis of FH; this is a missed opportunity for preventive therapy. Given that the first-degree relatives of patients with FH have a 50% chance of affliction, a targeted molecular genetic screening of individuals with premature CAD and FH is an effective strategy both to prevent cardiovascular diseases in their nascent stages and to confer more desirable management plans.

The accession number of the novel and reported variants are as follows:

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

The studies involving human participants were reviewed and approved by Rajaie Cardiovascular, Medical and Research Center Ethics Committees. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

AM: experiment design, lab work, data production, data interpretation, and MS edit. MMale, ZG, ZK, and FN: clinical evaluation of patients, data interpretation, and MS edit. MMo: lab work, data production, and MS edit. SK: data interpretation and MS edit. SM and NS: experiment design, data interpretation, and MS edit. MMala: experiment design, data interpretation, writing the first draft of MS, and manuscript final edit. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.