KEY CONCEPTS
- •
Keratoconus is a progressive corneal disease with genetic susceptibility.
- •
Genome-wide linkage studies have identified genes responsible for familial keratoconus.
- •
A number of keratoconus genes are also involved in other multisystem genetic disorders and ocular syndromes.
- •
Genome-wide association studies of keratoconus cases and population-wide studies of variation in central corneal thickness have identified new susceptibility genes and biological pathways.
- •
Transcriptomic and expression studies identify tissue-specific effects of keratoconus genes and noncoding RNAs.
- •
Genomic insights can be implemented into keratoconus clinical practice.
Evidence for Genetic Predisposition and Familial Inheritance of KC
Keratoconus (KC) is characterized by the progressive thinning and protrusion of the cornea, which assumes a conical shape in the most advanced cases ( Box 4.1 ). It is most commonly detected at puberty and is progressive until the third to fourth decades of life, when it usually arrests. Population prevalence of KC in different parts of the world ranges from 1:375 to 1:2000. Some countries report higher prevalence owing to specific genetic and possibly environmental conditions.
- ■
Complex/multifactorial
- ■
Genetically heterogeneous (familial transmission and “sporadic” presentation)
- ■
Gene-environmental interactions (i.e., eye rubbing)
- ■
Progressive
- ■
Population frequency between 1/375 and 1/2000 in different countries and ethnic groups
- ■
Clinically heterogeneous with variable age of onset and degree of visual acuity loss
- ■
Irreversible without treatment
- ■
If undiagnosed leads to post-LASIK ectasia
KC is a genetically heterogeneous disorder with family history being a major risk factor. Having a first-degree relative with KC remains the most significant risk factor for developing KC. , The estimated KC prevalence in first-degree relatives was found to be 15 to 67 times higher than that in the general population. Genetic contribution is further supported by higher concordance of KC severity in monozygotic twins versus dizygotic twins, and consanguineous marriage (first-cousin marriage) is a risk factor for KC in the offspring. Familial KC most commonly presents with an autosomal-dominant inheritance pattern and occasionally with a recessive pattern, especially in families with consanguineous marriages. Multiple segregation analyses in families have conclusively proved genetic determination and suggested complex genetic inheritance. ,
Clinically, even in the absence of fully developed KC, abnormal corneal topography (KC suspect) is frequently diagnosed in family members of KC patients , ( Fig. 4.1 ). In a longitudinal study, unaffected relatives with certain imaging parameters showed significantly greater risk of progression to KC.
GENOME-WIDE LINKAGE STUDIES
Multiple genome-wide linkage studies (GWLS) in individual families and family sets have been undertaken. GWLS involve genotyping families affected by a certain disease using a collection of genetic markers across the genome and examining how those genetic markers segregate with the disease across multiple families. GWLS, also called linkage studies ( Box 4.2 ), have been applied successfully to identify genetic variants that contribute to rare disorders such as familial breast cancer, Huntington disease, cystic fibrosis, and others (reviewed in Altshuler et al.). For decades, these studies were generally conducted using 300 to 400 microsatellite markers spaced at 10 to 20 centimorgans (cM) apart. These multiallelic markers were robust and highly informative; however, their genotyping was a time-consuming process. Shortly after single nucleotide polymorphisms (SNPs) were discovered to be abundant polymorphic markers uniformly distributed throughout the human genome, dense SNP arrays quickly became the genotyping platform of choice, owing to highly unparalleled interrogation and accurate scoring. Testing of genotyping data also evolved from being model based (recessive, dominant, etc.) to the use of robust nonparametric alternatives.
A genetic linkage study is a study aimed at identification of genetic markers inherited together with a locus for a specific trait or a disease, because of their proximity to one another on the same chromosome.
A genetic association study is a study aimed at testing whether genetic markers (i.e., single nucleotide polymorphisms [SNPs]) differ between two groups of individuals (cases vs. controls).
A genomic sequencing study is aimed at allowing researchers to identify complete nucleotide sequence of genes and intergenic regions of the genome.
A genetic expression study is aimed at detecting and quantifying messenger RNA (mRNA) levels of a specific gene.
A genetic transcriptomic study is aimed at detecting and quantifying the complete set of RNA transcripts that are produced by the genome.
A genetic functional study is aimed at studying the biochemical, cellular, and physiological properties of gene products, sometimes including intergenic regions of the genome.
GWLS analyses have identified a number of genomic loci linked with KC located on multiple chromosomes: 1p36.23-36.21, 2p24, 2q13, 3p14-q13, 5q14.3-q21.1, 5q21.2, 5q32-q33, 8q13.1-q21.11, 9q34, 13q32, 14q11.2, 14q24.3, 15q15.1, 15q22.33–24.2, 16q22.3-q23.1, and 20p13-p12.2, 20q12. However, for the vast majority, identification of genomic positions of the KC genes did not lead to conclusive evidence for specific genes. Genes and variants started to be identified by implementing a tool testing for genetic association. Recent advances in next-generation sequencing (NGS) methods (allowing simultaneous interrogation of billions of nucleotides) and multiplex genotyping (allowing simultaneous interrogation of millions of nucleotides at the specific positions in the genome) led to the identification of mutations and variants ( Table 4.1 ). Some of the identified genes and mutations were found to be located in the previously identified linkage regions.
| Gene Name | Position | Variant(s) | Gene Function | Method | References | ||
|---|---|---|---|---|---|---|---|
| CAST | 5q15 | rs4434401 | Calpain/calpastatin proteolytic degradation | GWLS, FM | |||
| COL4A3 | 2q36.3 | Multiple | Collagen type IV, alpha-3 chain | S | |||
| COL4A4 | 2q36.3 | Multiple | Collagen type IV, alpha-4 chain | S MA | , , | ||
| COL5A1 | 9q34.2-34.3 | Multiple | Collagen type V, alpha-1 chain | GWLS, FM, GWAS, MA, S | , , , , , | ||
| DOCK9 | 13q32.3 | c.2262A>C p.Gln754His | Guanine nucleotide exchange factor | GWLS, GWAS, S | |||
| FNDC3B | 3q26.31 | rs4894535 | Fibronectin extracellular matrix protein | GWAS, MA | , , , | ||
| FOXO1 | 13q14.1 | rs2721051 | Transcription factor | GWAS, MA, TG | , , , , , | ||
| HGF | 7q21.1 | rs3735520 | Involved in corneal wound healing | GWAS, TG | , , | ||
| IL1A | 2q13 | rs2071376 | Interleukin 1 alpha, cytokine | TG, S | |||
| IL1B | 2q13 | Multiple | Interleukin 1 beta, cytokine | TG, S | |||
| IMMP2L | 7q31.1 | rs757219 | Inner mitochondrial membrane peptidase subunit 2 | GWAS, MA | , | ||
| LOX | 5q23.2 | Multiple | Lysyl oxidase, participates in collagen cross-linking | GWLS, LD, FM, S, MA, TG | , , , , | ||
| MAML2 | 11q21 | rs10831500 | Transcription factor | GWAS, TG | , | ||
| MIR184 | 15q25.1 | c.57 C>T | MicroRNA | GWLS, TG | , | ||
| MPDZ/NF1B | 9p23 | rs1324183 | Intergenic region | GWAS, TG | , , , , , | ||
| PNPLA2 | 11p15.5 | rs61876744 | Participates in triglyceride hydrolysis | GWAS, TG | , | ||
| PPIP5K2 | 5q21.2 | Multiple | Kinase/phosphatase | GWLS, LD, FM, S | |||
| RAB3GAP1 | 2q21.3 | rs4954218 | Regulates exocytosis | GWAS, TG | , , | ||
| SOD1 | 21q22.1 | Multiple | Superoxide dismutase 1, cytoplasmic antioxidant enzyme | S | , | ||
| TGFBI | 5q31.1 | Transforming growth factor beta induced | S | , | |||
| TIMP3 | 22q12.3 | c.476 C>T p.Pro458Ser | Tissue inhibitor of metalloproteinases | S | |||
| TSC1 | 9q34 | Multiple | Hamartin, regulates cell growth | S | |||
| VSX1 | 20p11.2 | Multiple | Visual system homeobox 1, transcription factor | S | |||
| WNT10A | 2q35 | rs12190810 | WNT signaling | GWAS | |||
| ZEB1 | 10p11.2 | c.1920G >T p.Gln640His | Zinc finger transcription factor | S | , | ||
| ZNF469 | 16q24.2 | Multiple | Transcription factor | GWAS, TG, S, TA | , , , , |
NGS enables interrogation of genomic sequence in the absence of linkage. Whole exome sequencing (WES) is used to identify variants in the protein-coding regions (about 1% of the human genome), whereas whole genome sequencing (WGS) explores variants in the whole genome, as well as structural changes such as copy number variants.
GENES FOR FAMILIAL KC
LOX
One of the first discovered and most promising KC genes is the one coding for the LOX (lysyl oxidase) gene. It is located in the 5q32-q33 genomic region identified by a two-stage GWLS. Multiple SNPs rs10519694 and rs2956540 located in the intron (noncoding portion) of the LOX gene and SNPs rs1800449 and rs2288393 located in the exons (coding portion) of the gene were found to be genetically associated in KC families. They were also found to be associated in case-control panels of KC patients without known family history. LOX initiates the cross-linking of collagens and elastin by catalyzing oxidative deamination of the epsilon-amino group in certain lysine and hydroxylysine residues. LOX defects can potentially lead to the reduction of cross-linking of collagen fibers of the corneal stroma, thus leading to biomechanical weakening of the cornea. Multiple samples of independently collected KC patients around the world confirmed the LOX association (see Table 4.1 ). LOX involvement is also supported by functional data that showed its attenuation in the corneal epithelium of KC patients at levels corresponding to disease severity and revealed changes in LOX distribution and its decreased activity in KC corneas.
DOCK9
Single mutation rs191047852 (c.2262A > C, p.Gln754His) in the DOCK9 (dedicator of cytokinesis 9) gene was identified by targeted Sanger sequencing of the 13q32 chromosomal region linked to KC in several Ecuadorian families. Dozens of affected individuals were found to carry this mutation whereas almost no controls were carriers. Functional investigation in vitro found aberrant splicing of the DOCK9 gene that leads to exon skipping, resulting in the introduction of a premature stop codon, disrupting the functional domains of DOCK9 protein, which may alter its biological function as an important regulator of corneal wound repair.
PPIP5K2
Recently, two familial variants of a novel KC gene PPIP5K2 (diphosphoinositol pentakisphosphate kinase 2) were identified in a four-generation family identified in 1992, linked by GWLS to 5q14.3-q21.1 (Tang et al.), and by genetic association to the 95 to 100 Mb region on chromosome 5 and in another unrelated family ( Fig. 4.2A and B ). These two variants, rs35671301 (c.1255 T > G, p.Ser419Ala, S419A) and rs781831998 (c.2528 A > G, p.Asn843Ser, N843S), are located in the exonic portion of the PPIP5K2 gene. Biochemical effects of these variants were investigated in vitro that identified significant reduction in the phosphatase activity and elevated levels of the kinase activity. Further investigation in the mouse model carrying truncated PPIP5K2 gene (in vivo) found irregularities on the anterior corneal surface, reduced anterior chamber depth, and corneal opacity in mice heterozygous and homozygous for the affected allele ( see Fig. 4.2C ).
MIR184
MicroRNAs (miRNAs) are small noncoding RNAs that suppress posttranscriptional gene expression by pairing with their target messenger RNAs (mRNAs), inducing either translational repression or mRNA degradation of their targets. The biologically active part of these molecules (seed) is only 18 to 25 nucleotides long. Mutation c.57 C > T in the seed region of miR-184 has been identified in familial hereditary ocular diseases, including familial KC with lens abnormalities and stromal thinning , and EDICT (endothelial dystrophy, iris hypoplasia, congenital cataract, and stromal thinning) syndrome. This mutation represents a single known germline inherited miRNA defect associated with a variety of ocular abnormalities. Analysis of the biological effects of this mutation in human lens epithelial cells (in vitro) identified significantly altered expression of genes mostly coding for proteins of the cell membrane and those involved in calcium ion transfer.
GENES INVOLVED IN KERATOCONUS AND OTHER SYNDROMES
In about 97% of cases, KC is seen in its non-syndromic form, that is, without involvement of other tissues; however, in about 3% of cases, it has been identified in patients with multisystem diseases such as Down syndrome, Ehlers-Danlos syndrome (EDS), Bardet-Biedl syndrome, nail-patella syndrome, and others ( Box 4.3 ). A study has found that as many as 85% of the known KC-related genes may also be involved in connective tissue disorders or other ocular diseases ( Fig. 4.3 ).
