Initially identified as oligodendrocyte specific protein (OSP), claudin-11 expression in the CNS is specific to oligodendrocytes (Bronstein et al., 1996, 1997) (Figure 2). During early development, claudin-11 is expressed in mesenchymal cells adjacent to meningeal cartilage-rich areas, while postnatally, claudin-11 expression in the CNS is restricted to early oligodendrocyte progenitors and their lineage (Bronstein et al., 2000). Topologically, claudin-11 resembles other oligodendrocyte-expressed myelin proteins such as PLP1 (Gow et al., 1997; Krause et al., 2008; Devaux et al., 2010). Unlike PLP1, however, claudin-11 is localized exclusively within the radial component between the layers of the myelin sheath (Gow et al., 1999). This specific localization is critical for regulating ion diffusion and myelin permeability, which is essential for normal electrophysiological function (Devaux and Gow, 2008; Gow and Devaux, 2008; Rasband et al., 2012). Claudin-11 expression within the CNS is restricted to the oligodendrocyte lineage, however, outside of the CNS several other cell types are found to express this unique protein.

Outside of the CNS, claudin-11 is expressed in the basal cells of the stria vascularis of the inner-ear (Gow et al., 2004; Kitajiri et al., 2004a,b) and the Sertoli cells of the testes (Gow et al., 1999; Stanton, 2016) (Figure 2). The human inner ear contains thousands of mechanoreceptors essential for hearing (Liu et al., 2017). A potassium-rich endolymph compartment within the cochlea generates the endocochlear potential (EP), an electric potential crucial for transducing acoustic stimuli into electrical signals (Gow et al., 2004; Kitajiri et al., 2004b; Liu et al., 2017; Zhang et al., 2020). By regulating ion movement across the stria vascularis, claudin-11 serves a dual role in the cochlea as both conductor and isolator, effectively functioning as a “conventional electrochemical battery” for ion transport (Liu et al., 2017). Studies in knockout mouse models have revealed the indispensable role of claudin-11 in hearing. Despite having no clear gross morphological malformations within the inner ear, claudin-11 knockout mice are deaf as confirmed through auditory brainstem analysis (Gow et al., 2004; Kitajiri et al., 2004b). Furthermore, freeze fracture electron microscopy showed the absence of TJs in the stria vascularis, highlighting the critical role of claudin-11 TJs in hearing (Gow et al., 2004; Kitajiri et al., 2004b).

In addition to the CNS and cochlea, claudin-11 is also expressed in the testes (Gow et al., 1999) (Figure 2). Several claudin family members, including claudin-1, -3, -5, -11, -12, and -13, are present in the blood-testis barrier (Stanton, 2016). Interestingly, claudin-11 is the sole essential claudin for spermatogenesis, as its absence in knockout mice leads to infertility, whereas loss of claudin-1, -3, -12, or -13 does not affect fertility (Gow et al., 1999; Stanton, 2016). The role of claudin-5 in spermatogenesis remains unresolved due to the postnatal lethality of its knockout models (Stanton, 2016).

Outside normal physiology, claudin-11 expression has been implicated in several cancer types. Interestingly, both claudin-11 silencing and overexpression have been linked to invasiveness of cancer cells. For example, in gastric cancers, hypermethylation of CLDN11 and downregulation of its expression were associated with increased cancer cell motility and invasiveness (Agarwal et al., 2009). Similar observations were reported in nasopharyngeal carcinoma cells (Li et al., 2018). Conversely, studies investigating epithelial-mesenchymal transition (EMT) mechanism implicated in cancer metastasis report increased expression of CLDN11 correlating with less favorable patient outcomes (Li et al., 2019). These findings suggest that reduced claudin-11 expression in certain cancers, such as the gastric and nasopharyngeal types, may promote cell migration through loosening of cellular junctions (Agarwal et al., 2009; Li et al., 2018). In contrast, claudin-11 upregulation in EMT-linked cancers may promote maintenance of cell clusters, thus facilitating more efficient cancer cell dissemination (Li et al., 2019).

Claudin-11 is expressed in both oligodendrocyte progenitor cells (OPCs) and oligodendrocytes (OLG) (Bronstein et al., 2000). OPCs are bipolar, migratory, and proliferative cells expressing platelet-derived growth factor receptor A (PDGFRa) and neural/glial antigen 2 (NG2), which differentiate into pre-myelinating O4- and O1-positive oligodendrocytes that ultimately differentiate into non-proliferative, myelin-producing OLGs (Pouwels et al., 2014; Kuhn et al., 2019). This myelination process entails a specialized, energy-intensive membrane expansion, increasing the membrane surface area by over six thousand-fold in order to ensheath axons (Pouwels et al., 2014).

The specialized membrane of the myelin sheath contains many key transmembrane proteins such as PLP1 and claudin-11 which are critical to its integrity and function. Claudin-11 is unique in that it is the sole claudin family member expressed in CNS myelin and therefore its loss cannot be compensated for by other claudin proteins (Gow et al., 1999; Furuse, 2009). Studies using claudin-11 null mice support this notion. Indeed, while claudin-11 null mice, which lack the radial component, do not show gross changes in ultrastructural features of myelin such as thickness or compaction, they do exhibit slower conduction velocities across myelinated axons as well as hind limb weakness (Gow et al., 1999; Devaux and Gow, 2008; Gow and Devaux, 2008). These findings demonstrate that claudin-11 is critical for the maintenance of the electrochemical gradients and electrical resistance properties within the myelin sheath (Gow et al., 1999; Devaux and Gow, 2008; Gow and Devaux, 2008).

Outside its role in the radial component of myelin, claudin-11 has been linked to OLG proliferation and migration. Using a yeast two-hybrid system, claudin-11 was found to interact and form a complex with OSP/claudin-11-associated protein (OAP-1) and beta-1 integrin (Tiwari-Woodruff et al., 2001). Overexpression of claudin-11 and OAP-1 increased proliferation in a conditionally immortalized mouse oligodendrocyte (CIMO) cell line (Bronstein et al., 1998; Tiwari-Woodruff et al., 2001). Antibodies targeting claudin-11, OAP-1, and beta-1 integrin led to a marked reduction in CIMO cell migration (Tiwari-Woodruff et al., 2001). These findings suggest that the claudin-11/OAP-1/beta-1 integrin complex may be involved in the regulation of migratory and proliferative behavior of OLGs (Tiwari-Woodruff et al., 2001). Despite these in vitro observations, overt impairment or abnormal proliferation and migration of OLGs have not been reported in claudin-11 null mice (Devaux et al., 2010).

White matter diseases are classified into subcategories based on the predominant nature of the myelin pathology observed and include demyelinating, myelinolytic, and dysmyelinating conditions (Naidu, 1999; Knaap and Bugiani, 2017). Myelinolytic diseases, exemplified by Canavan disease, are characterized by unique vacuolar myelinopathy (Naidu, 1999). Demyelinating conditions, such as multiple sclerosis, involve destruction of biochemically normal myelin (Naidu, 1999). Dysmyelinating diseases, notably leukodystrophies, often have genetic causes affecting myelin formation or function (Naidu, 1999). The term “leukodystrophy,” derived from “leuko” meaning white and “dystrophy” meaning wasting, is a fitting descriptor highlighting the degeneration and atrophy of CNS white matter observed in these conditions (Vanderver et al., 2015). The first documented leukodystrophy was described in 1885 and 1910 by Pelizeaus and Merzbacher, identifying a familial progressive white matter disorder known as PMD (Koeppen and Robitaille, 2002). HLDs, such as PMD and HLD22, are progressive disorders (Naidu, 1999; Knaap and Bugiani, 2017). Clinically, they manifest as developmental delay, ataxia, spasticity, hypotonia, and varying degrees of intellectual disability (Pouwels et al., 2014; Knaap and Bugiani, 2017). A defining feature of HLDs is the abnormal development or deposition of myelin (Pouwels et al., 2014). This is typically accompanied by variety of pathological alterations, such as neuronal loss, dysmyelination, demyelination, and axonal damage, that manifest to varying degrees depending on the underlying genetic defect (Sima et al., 2009; Gruenenfelder et al., 2020). HLDs are typically diagnosed through magnetic resonance imaging (MRI) manifesting as lower T2 hypointensities without much reduction in T1 hyperintensities (the latter is typically seen in other non-HLD leukodystrophies) (Pouwels et al., 2014).

CLDN11 was identified as a novel cause of HLD in 2020 through trio exome sequencing in three unrelated pediatric patients (2 male, 1 female) (Riedhammer et al., 2020). These patients carried de novo, heterozygous stop-loss variants: two had a c.622T>C, p.(*208Glnext*39) variant, while the third patient had the c.622T>G, p.(*208Gluext*39) variant (Riedhammer et al., 2020). Although additional disease-associated variants are listed in databases such as ClinVar and gnomAD, with associated phenotypes like marfanoid habitus and intellectual disability (National Center for Biotechnology Information, 2022), only the stop-loss variants are categorized as pathogenic to date.

The de novo stop-loss variants are predicted to result in a 39-amino acid extension at the cytoplasmic C-terminal end of claudin-11 (Riedhammer et al., 2020). This extension is predicted to form an alpha helix that does not integrate into the cytoplasmic membrane (Riedhammer et al., 2020). Imaging findings of the patients showed that T1-weighted images had signal intensity near normal, whereas T2-weighted images showed reduced myelination of white matter structures (Riedhammer et al., 2020). All three individuals presented with delayed developmental milestones, contractures, nystagmus, drooling, hypermetropia, severe disarticulation, and require mobility support when walking and climbing stairs, features that are shared with PMD and other HLDs (Riedhammer et al., 2020). Interestingly, hypermetropia is not a common clinical feature of HLD, suggesting its potential as a diagnostic marker for HLD22 (Riedhammer et al., 2020). HLD22 affects paediatric patients and is both early onset and progressive (Riedhammer et al., 2020). The underlying mechanisms for this disease are currently not well understood, and effective treatments are currently lacking.

Since the description of the first three patients with CLDN11-related HLD22, at least three more individuals have been identified with a clinical phenotype compatible with HLD22 and with CLDN11 variants leading to an extended protein (unpublished). Currently, it is not clear whether CLDN11 variants causing other types of changes in the protein are associated with a similar phenotype in humans.