Hypolacrimia and Alacrimia as Diagnostic Features for Genetic or Congenital Conditions

Abstract

As part of the lacrimal apparatus, the lacrimal gland participates in the maintenance of a healthy eye surface by producing the aqueous part of the tear film. Alacrimia and hypolacrimia, which are relatively rare during childhood or young adulthood, have their origin in a number of mechanisms which include agenesia, aplasia, hypoplasia, or incorrect maturation of the gland. Moreover, impaired innervation of the gland and/or the cornea and alterations of protein secretion pathways can lead to a defective tear film. In most conditions leading to alacrimia or hypolacrimia, however, the altered tear film is only one of numerous defects that arise and therefore is commonly disregarded. Here, we have systematically reviewed all of those genetic conditions or congenital disorders that have alacrimia or hypolacrimia as a feature. Where it is known, we describe the mechanism of the defect in question. It has been possible to clearly establish the physiopathology of only a minority of these conditions. As hypolacrimia and alacrimia are rare features, this review could be used as a tool in clinical genetics to perform a quick diagnosis, necessary for appropriate care and counseling.

The human eye is a complex structure, composed of an anterior and a posterior segment. The anterior eye segment is comprised of the cornea, iris, ciliary body, and lens. The anterior chamber, limited anteriorly by the cornea and posteriorly by the iris, contains the aqueous humor. The cornea is the avascular transparent tissue in contact with the external environment and covered by the tear film. Together with the lens, they form a single optical element that is instrumental in focusing light onto the retina. The posterior eye segment is composed of the vitreous humor, choroid, retina, and optic nerve.¹ 

The tear film, covering the eye surface, is composed of three layers—namely, the lipid, aqueous, and mucous layers. These three different layers are produced respectively by the meibomian glands, the lacrimal glands (LGs), and the goblet cells scattered on the conjunctival epithelium.² As the cornea is avascular, the LG supplies the necessary growth factors and cytokines for corneal maturation and homeostasis, as well as antibacterial and antiviral factors.³ Any defects in tear film quality will deteriorate corneal physiology, which ultimately can lead to opacification and white blindness.^4–7 

In humans, the LG is a tubulo-acinar gland formed of two lobes—the palpebral lobe and the orbital lobe, which are anatomically close to each other. Each lobe is divided into lobules, separated from each other by a loose connective tissue made of extracellular matrix, nerves, and blood vessels. The secreted liquid is delivered on the eye surface through six to 12 lacrimal ducts. Along with the main LG, its accessory glands, the glands of Wolfring and Krause, are localized, respectively, in the palpebral conjunctiva and in the conjunctival fornix. These glands are structurally and functionally similar to the main LGs and secrete about 10% of the tear volume.⁸ 

In humans, the LG and the lacrimal outflow pathway develop from the surface ectoderm between 5 and 7 weeks of gestation. Lacrimal gland ducts are formed at 12 weeks of gestation. Despite LGs being fully formed at birth, the mature excretory functions are only acquired postnatally, during the first 6 weeks of life.⁹ The mature LG is formed of three distinct cell types—acinar, myoepithelial and ductal compartments—and each has a role in the production, modification, and excretion of the fluids, respectively. Pyramid-shaped cells surrounding a central lumen form an acinus. The acini produce around 80% of the LG tear volume and secrete most of the proteins, electrolytes, water, and other constituents composing the aqueous part of the tear film.¹⁰ Myoepithelial cells are stellate and multiprocessed cells with contractile functions, located around the acini. They maintain acinar integrity, secrete the basal membrane components, and participate in acinar production excretion via their contraction.¹¹ LG fluid is modified and secreted by ducts. Ductal cells are water permeable and exhibit ion channels, which act upon LG fluid osmolarity. After its secretion, the lacrimal fluid is distributed over the ocular surface through eyelid movements and the presence of a tear meniscus. Overall, the lacrimal apparatus produces tears that are drained into the conjunctival sac via the puncta.¹² 

Sensory and parasympathetic innervation of the LG is provided by the lacrimal nerve.¹³ It is one of the branches of the ophthalmic nerve (V1). The latter is one of the three divisions of the trigeminal nerve. The lacrimal nerve passes through the superior part of the orbit to innervate the LG. 

Dry eye disease can be categorized into either evaporative dry eye, linked to meibomian gland deficiencies, or aqueous deficient dry eye, linked to LG deficiencies. Alacrimia, the total absence of tears, is relatively rare during childhood or young adulthood and becomes more frequent in the elderly population, being present in 15% of seniors above 65 years of age.¹⁴ The relevant processes at work in the elderly may be variable and include neurologic, immunologic, and iatrogenic processes.¹⁵ 

Hypolacrimia or alacrimia could be due to several mechanisms. First, agenesia, aplasia, or hypoplasia of the gland generates an insufficiency of tear production. Second, defective maturation of the gland, despite anatomical integrity, leads to abnormal tear production. Third, tear production relies on the integrity of the proteins involved in the secretory lacrimal system. Last, innervation or nerve degeneration affecting LG afferent or efferent nerve transmission may be implicated. 

We have attempted here to review systematically all genetic conditions and congenital disorders that include alacrimia or hypolacrimia as a feature. Where known, we have described the pathophysiological mechanism of the lacrimal defect and the consequences of the molecular defect of the gene involved at cellular, tissue, and organismal levels. We believe this will advance understanding of the physiological development of the lacrimal apparatus and provide a diagnostic tool for a better care of the patients (Table, Fig.). 

Table.

 

Summary of Different Conditions Classified by Origin of Lacrimal Defects

 

Figure.

 

Main pathophysiological processes involved in human hypolacrimia or alacrimia with corresponding anatomical defects, pathology names, and genes identified.

Anatomic Anomaly: LG Aplasia or Hypoplasia Associated With Secretion Defects

The first reported case of a congenital lack of lacrimal secretion was described by Morton in 1884.¹⁶ A boy had hemifacial microsomia, with a right microphthalmia and a less developed right malar bone. He had never shed tears from the right eye. The LG could not be felt on either side, but the hypothesis was that the LG was congenitally absent only on the right side. In 1928, Ishikawa mentioned a 4-year-old boy with alacrimia and postulated that his condition could be explained by congenital hypoplasia of the LG. Sánchez Sevila et al.¹⁷ reported a case of documented congenital absence of the LG combined with lipoma responsible for unilateral congenital alacrimia. Isolated bilateral LG agenesis was more recently reported by Alwohaib et al.¹⁸ in two unrelated 5-year-old children, one boy and one girl, confirmed by magnetic resonance imaging (MRI). No genetic investigation was performed in these isolated cases, and there was no familial history or information about possible consanguinity in them. Although these cases had no known causes, LG anatomic anomaly has since been reported as being associated with specific syndromes. 

Bilateral dacryocele with punctal and canalicular agenesis and alacrimia form part of the lacrimoauriculardentodigital syndrome (LADD) or Levy–Hollister syndrome (MIM #149730).¹⁹ LADD is an extremely rare autosomal dominant disorder. It results from heterozygous pathogenic variations in the tyrosine kinase domains of one of the three genes encoding either for the fibroblast growth factor (FGF) receptors FGFR2 or FGFR3 or for FGF10, an FGFR2 ligand. 

Patients affected with LADD have varying degrees of hypoplasia/aplasia of the salivary glands, which can be responsible for xerostomia, dental cavities, dysphagia, dental anomalies (i.e., hypoplastic enamel, hypodontia, oligodontia, and microdontia). Furthermore, they exhibit inner and outer ear malformations, possible hearing loss, and digital anomalies including aplasia, hypoplasia, syndactyly, clinodactyly, or finger duplications.²⁰ 

Moreover, the ophthalmic involvement in this syndrome is extremely common and presents variable degrees of hypoplasia/aplasia of the lacrimal glands and ducts. It is characterized by a combination of agenesis of the lacrimal puncta, nasolacrimal duct obstruction, and deficiency of tear production. Patients with alacrimia and lacrimal puncta agenesis have no epiphora nor dacryocystitis. In the absence of LGs and normal puncta and nasolacrimal ducts, patients exhibit severe dry eye. The association of LG agenesis, dacryocystocele, and an abnormal lacrimal system is highly suggestive of LADD, even in the absence of other typical symptoms of the syndrome.¹⁹ 

Other cases with milder symptoms have been reported which could be associated with mild or rare forms of LADD. Kapoor et al.²¹ reported bilateral congenital alacrimia in a patient with giant dacryocele with punctal and canalicular agenesis. The patient did not have any other abnormality; he did not have dry mouth, ruling out a diagnosis of aplasia of major salivary glands. A report speculated that this case could represent a very mild form of LADD, with only lacrimal expression. Recently, Gupta et al.²² reported a bilateral congenital dacryocele with punctal atresia and LG agenesis in a 19-year-old woman and concluded that this represents a rare form of lacrimal outflow dysgenesis. This may consist of a mild, ocular variant of LADD, without any other system involvement. In both cases, the genetic origin of this anomaly remains speculative, as no molecular investigation was performed. 

Pathogenic variations in FGF10 can also result in aplasia of the lacrimal and salivary glands (ALSG; MIM #180920), an allelic disorder. Affected individuals may suffer from aplasia or hypoplasia of the lacrimal, parotid, submandibular, and sublingual glands and from an absence of the lacrimal puncta.²³ Such individuals are often misdiagnosed with the more prevalent disorder known as Sjögren syndrome, an autoimmune condition characterized by keratoconjunctivitis sicca and xerostomia. 

The targeted disruption of individual FGF or FGFR genes (or combinations thereof) in mice, as well as the analysis of disease-causing pathogenic variations in humans, has led to an understanding of the biological roles of individual FGFs and FGFRs. The FGFR family plays crucial roles in many developmental and physiological processes, with a variety of diseases being caused by aberrant signaling induced by FGFs or FGFRs.²⁴ FGFs have various roles in regulating cell proliferation, migration, and differentiation during development. They usually signal directionally and reciprocally across epithelial–mesenchymal boundaries. Extremely tight regulation of FGF activity and receptor specificity is required to maintain these signaling pathways. 

FGF10 is essential for postnatal life because of its critical role in development of the craniofacial complex, including salivary and lacrimal glands.²⁵ Notably, it is largely expressed first in mesenchyme and then in epithelial cells in murine salivary glands, in a strictly time-controlled fashion.²⁶ Makarenkova et al.²⁵ demonstrated that FGF10 is an inducer for LG development in mouse embryos. 

Shams et al.²⁷ studied the biological properties of FGF10 and FGFR2b mutants implicated in LADD. They showed that LADD FGF10 pathogenic variations cause inactivation of FGF10 and that the tyrosine kinase activity of FGFR2b LADD mutants expressed in cultured cells is severely compromised. Although the FGF10 pathogenic variations cause haploinsufficiency, the FGFR2b LADD mutants may exert a dominant negative interfering effect on signaling via normal FGFR2b, causing LADD, contrary to the dominant activating effect of FGFR2 pathogenic variations implicated in craniosynostosis. In a patients with LADD and his mildly affected father, Ryu et al.¹⁹ identified a heterozygous missense variation of FGFR2: c.1547C>T (p.Ala516Val); however, the functional consequences of this variation on the protein activity were neither studied nor discussed. The authors considered this variation as probably pathogenic, because the clinical phenotype of LADD was established in this family. 

Frontonasal dysplasia 2 syndrome (MIM #613451) is an autosomal recessive disorder arising from biallelic aristaless-like homeobox 4 gene (ALX4) loss-of-function pathogenic variations. Patients affected with this syndrome exhibit skull defects, sometimes in combination with encephalocele, wide nasal bridge, notched nares, depressed nasal tip, hypertelorism, and alopecia, sometimes associated with alacrimia.²⁶ 

Garg et al.²⁸ phenotypically reanalyzed one patient carrying a homozygous c.503delC pathogenic variation in exon 2 of the ALX4 gene, which resulted in the truncation of the homeobox (HD) and C-terminal OAR domain, previously reported by Kariminejad et al.²⁹ Since birth, the patient had produced no tears and had experienced multiple episodes of eye infection. An MRI revealed a bilateral absence of LG, confirming a lacrimal aplasia similar to the observed mouse-model phenotype. 

ALX4 encodes a homeodomain transcription factor important for many developmental processes.³⁰ Homeobox proteins are implicated in early embryonic development. ALX4 protein is required for development of skull, head, and face. Biallelic pathogenic variations of ALX4 are thus responsible for defects in skull and face formation. 

Garg et al.²⁸ showed that ALX4 is required for LG development in mice. In their experiments, inactivation of Alx4 disrupted FGF10 expression and downstream FGF signaling, causing failure of LG development. They identified ALX4 as the key effector of Shp2 signaling to control the expression of FGF10 in the periocular mesenchyme. They analyzed Alx4^lst-J mice carrying a frameshift pathogenic variation that removed both the homeodomain and the downstream C-terminal orthopedia, aristaless, and rax (OAR) domain, reproducing the human ALX4-related pathology. In homozygous Alx4^lst-J animals, at E14.5, there was a drastic reduction of FGF10 adjacent to the LG bud, accompanied by a downregulation of FGF target genes ETS variant transcription factor 4 (Etv4) and Etv5 in the LG epithelium. By P1, no LG was detectable. They demonstrated that ALX4 binds a terrestrially conserved FGF10 genomic element to regulate its expression in the LG mesenchyme. 

PCWH syndrome (peripheral demyelinating neuropathy, central dysmyelination, Waardenburg syndrome, and Hirschsprung disease; MIM #609136), is an autosomal dominant condition that arises from heterozygous SRY-box transcription factor (SOX)10 pathogenic variations, resulting in a multisystemic disorder.³¹ 

There is a huge phenotypic heterogeneity in SOX10-related disorders, including notably three types of Waardenburg syndrome (WS) including PCWH syndrome, as well as Hirschsprung disease, Kallman syndrome, and non-syndromic deafness.³⁰ WS is characterized by the association of deafness and pigmentation anomalies of the skin and eyes, with variable penetrance and expressivity.³² In type 2, there is no other associated feature, whereas type 4 includes Hirschsprung disease. Patients with PCWH have Hirschsprung disease, deafness, iris heterochromia, hypomelanic skin patches, and neurologic involvement with possible nystagmus, myoclonus, cerebellar ataxia, spasticity, hyporeflexia, arthrogryposis, distal muscle wasting, and severe intellectual disability. Touraine et al.³³ reported alacrimia as a feature in the first reported patients with PCWH. Alacrimia, as well as asialia, was first related to dysautonomia. Elmaleh-Berges et al.³⁴ retrospectively reviewed imaging studies from 14 WS probands whose subjects all had different pathogenic variations in the SOX10 gene, including two patients with WS2, six with WS4, and six with PCWH. Hypoplastic or absent LGs were incidentally identified in 65% and 14% of these patients, respectively. This abnormality was associated with aplasia of the parotid glands in some patients and was present in patients with all types of SOX10-related Waardenburg, except WS2.³⁵ However, clinically relevant alacrimia was reported only in patients with PCWH and not in patients with other types of WS. 

SOX10 belongs to the SOX family of transcription factors, which are involved in multiple developmental processes, such as neurogenesis and neural crest development, where they control stemness, cell fate, and differentiation.^36,37 The SOX10 transcription factor is indeed a characteristic marker for migratory multipotent neural crest progenitors, as well as for various neural crest derivatives. SOX10 also plays a role during differentiation of myelinating Schwann cells and oligodendrocyte, by inducing stage-restricted transcriptional regulators.^38,39 

There is a large proportion of truncating pathogenic variations in PCWH that are located in the C-terminal part, responsible for an escape from non–sense-mediated RNA decay. A few non-stop pathogenic variations have been described, notably one that generates a specific inframe new C-terminus with the loss of normal termination. Functional studies performed in mouse models have implicated a gain-of-function deleterious effect.⁴⁰ 

Blepharophimosis, ptosis, and epicanthus inversus syndrome (BPES; MIM #110100) is an autosomal dominant condition linked to loss of function of the FOXL2 gene. Patients with BPES exhibit the association of ptosis, epicanthus inversus, and telecanthus. Women sometimes exhibit premature ovarian failure. Hypolacrimia and alacrimia are often incorrectly evaluated due to eyelid malformation but have been reported in association with hypoplasia or aplasia of the LG. In a series of 21 patients reported by Duarte et al.,⁴¹ MRI evaluation showed a lack of LG in 52.3%; of these cases, 80% had a bilateral absence of LG. Of the cases exhibiting LGs, 33% showed a reduced LG volume. There was a clear association between these radiologic findings and tear hyposecretion, diagnosed by Schirmer's test, which might be complicated with keratopathy. 

Human forkhead box L2 (FOXL2) codes 376 amino acids, including a 110-amino acid DNA-binding forkhead domain (FHD) and a polyalanine tract of 14 residues (poly-Ala).⁴² It belongs to the winged helix/forkhead transcription factor family. FOXL2 is a nuclear protein specifically expressed in the mesenchyme of developing eyelids and in fetal and adult ovarian follicular cells.⁴³ Pathogenic variants in FOXL2 generate a change in DNA binding and transactivation capacity, causing an abnormal localization of the protein. Some missense variants lead to the mislocation and intranuclear aggregation of the FOXL2 protein, whereas non-sense variants are usually associated with protein retention in the cytoplasm. Expansions of the polyalanine tract also cause mislocation from the nucleus to the cytoplasm and aggregation of the protein.⁴⁴ Such changes affect the interaction of FOXL2 with target promoters, thus leading to decreased expression of several genes involved in stress response, including apoptosis, transcriptional regulation, mediation of inflammation, cholesterol metabolism, and reactive oxygen species detoxification. 

Abnormal Maturation of the LG

Abnormal maturation is defined by anatomical presence and normal volume of the LG but abnormal differentiation resulting in defective tear production. 

Ectodermal dysplasia (ED) represents a diverse group of inherited disorders characterized by a congenital defect in two or more ectodermal structures, which derive from embryonic ectoderm, such as hair, teeth, nails, sweat glands, mammary glands, external ear, melanocytes, cornea, conjunctiva, and lacrimal apparatus. All of these can be variably involved.⁴⁵ 

There is within ED a huge variability in genetic causes and clinical phenotypes. Wright et al.⁴⁵ classified the different conditions according to the molecular pathways involved: ectodysplasin-A (EDA)/nuclear factor-kappa B (NF-κB), tumor protein (TP) 63, and structure group. Abnormalities of tear production were described in hypohidrotic ectodermal dysplasia linked to the EDA gene (EDA/NF-κB pathway group) and in p63-related disorders. 

X-linked hypohidrotic ectodermal dysplasia (XLHED; OMIM #305100) is an X-linked recessive condition linked to EDA. Dry eye diseases observed in patients with XLHED are linked to LG defects.⁴⁶ A loss-of-function variant of the EDA gene, which belongs to the tumor necrosis factor (TNF) superfamily, was the first genetic alteration identified as being causative for ED.⁴⁷ EDA signaling acts principally in a paracrine manner and activates NF-κB during skin appendage formation by triggering the formation of an NF-κB–associated switching defective/sucrose nonfermenting (SWI/SNF, or BAF) complex leading to subsequent gene regulation, initiating a signaling to facilitate transcription during organogenesis.⁴⁸ In Eda^−/− mutant mice, also called tabby, atrophied LGs were detected during embryogenesis, and a dry eye phenotype was detected in postnatal stages.⁴⁹ However, Grüneberg⁴⁹ attributed the related dry eye phenotype observed in patients to an atrophy of the meibomian glands and subsequent accelerated tear film evaporation, without evaluating the effect on LG formation and function of Eda pathogenic variation. Kuony et al.⁵⁰ characterized the physiological and molecular defects of LG in mice resulting from an EDA loss-of-function pathogenic variation. They demonstrated that, despite normal morphology, Eda^−/− LGs exhibited altered terminal differentiation in comparison with control. Their results suggest that EDA signaling has little or no impact on LG embryonic morphogenesis but results in impaired LG maturation and defective LG secretory function. 

In a dog model of XLHED, caused by a point pathogenic variation in the splice-acceptor site of intron 8 that results in a truncated non-functional protein, Casal et al.⁵¹ found that XLHED dogs produce ∼25% less lacrimal fluid than wild-type dogs, which may contribute to frequent neonatal eye infections and keratoconjunctivitis sicca in older affected dogs. They corrected these anomalies by administrating recombinant EDA. More recently, Schneider et al.⁵² reported normal development of human fetuses with EDA mutations after injecting in utero a recombinant protein including the receptor-binding domain of EDA. They injected this protein intra-amniotically into two affected human twins at 26 and 31 weeks of gestation and into a single affected human fetus at 26 weeks of gestation. They observed a partial correction of teeth and sweating defects and a normalization of the meibomian gland number. 

The second group of ED syndromes includes all TP63 pathway syndromes, including five clinically defined different syndromes. These are autosomal dominant disorders arising from TP63 heterozygous pathogenic variations, which can be associated with ectodermal defects, cleft lip/palate, and ectrodactyly.⁵³ 

The ocular phenotype of patients suffering from ectrodactyly ectodermal dysplasia cleft lip/palate syndrome (EEC) linked to TP63 (MIM #604292) was studied by Di Iorio et al.,⁵⁴ who aimed to determine the pathogenic basis of visual phenotype. Twenty-three patients underwent a full ocular assessment, and all had ocular defects, the most common being an anomaly of the meibomian glands, which was present in all 23 cases (100%). These meibomian gland defects resulted in evaporative dry eye, which was evidenced in all assessed patients by unstable tear film, as measured by the tear film breakup time. The second most common ocular anomalies consisted of lacrimal drainage system defects (absence, occlusion, or stenosis), which were present in 21 of 23 cases (91.3%). Anomalies of tear meniscus height and tear function index revealed the presence of aqueous tear deficiency in 56.5% of the patients (13/23). 

TP63 (also known as p63) is a transcription factor of the p53 family.⁵⁵ Multiple isoforms of p63 have been discovered and have diverse functions, including lineage specification, proliferative potential, differentiation, cell death and survival, DNA damage response, and metabolism. It is a key regulator in epithelial and ectodermal structure commitment and development. TP63 is a convergence of numerous signalization pathways and is directly regulated by the wingless-type MMTV integration site family (Wnt)/β-catenin pathway.⁵⁶ It is also an important downstream target of Hedgehog signaling.⁵⁷ In addition to the role of NF-κB in regulating p63 in response to DNA damage, NF-κB–mediated repression of TP63 also plays a role in epithelial cell differentiation.⁵⁸ The tyrosine kinase receptor epidermal growth factor receptor can also induce TP63 expression.⁵⁹ TP63 is additionally regulated in a cell-type–specific manner by Notch, which is a key regulator of maintenance and cell fate.⁶⁰ 

Abnormal Nerve Function

Congenital Defects of Innervation

Sjögren and Eriksen¹⁶ were the first to provide a historical review of cases of alacrimia congenita and reported cases associated with cranial nerves aplasia. In 1900, Heubner described a boy, 11 years old with healthy parents, who suffered from double-sided paralysis of the abducent, facial, and hypoglossal nerves, the left side being more affected than the right, with a complete lack of tears. Postmortem examination revealed that the cause was a congenital aplasia of the brain where the nuclei of these nerves lie. Thus, the left hypoglossus, the left facialis, and both abducens nuclei were lacking. In the right hypoglossus and facialis nuclei, the cells were fewer than usual. In 1921, Kayser reported a case of congenital trigeminal paralysis with total lack of tears. In 1931, Lutz suggested that the cause of alacrimia might be anaplasia of the petrosal bone with co-existing aplasia of the nervus petrosus superficialis major. 

Acquired Defects of Innervation

Autoimmune Process

Autoimmune polyendocrine syndrome type 1 (APS1), or autoimmune polyendocrinopathy–candidiasis–ectodermal dystrophy (APECED; MIM #240300), is a rare polyendocrinopathy with autosomal recessive inheritance. It results from defects in the human autoimmune regulatory (AIRE) gene. 

The three major manifestations of APS1 are mucocutaneous candidiasis, hypoparathyroidism, and Addison's disease. Ophthalmic manifestations include keratoconjunctivitis, dry eye, iridocyclitis, and cataract.⁶¹ 

APS1-associated keratopathy appears to be an early manifestation, with the age of onset ranging from 2 to 9 years. Most cases have dry eye with a variable degree of pain and photophobia.⁶² 

The AIRE gene encodes a protein, the function of which has been discovered through the generation and study of Aire knockout (KO) mice.⁶³ The broad spectrum of self-antigens expressed by medullary thymic epithelial cells and presented to the developing thymocytes is linked to AIRE expression.⁶⁴ Chen et al.⁶⁴ used a spontaneous mouse model deficient in the Aire gene. The mice developed spontaneous, CD4⁺ T-cell–mediated exocrinopathy and aqueous-deficient dry eye, which were associated with a loss of nerves causing innervation of the cornea and LG. Changes in innervation and tear secretion were accompanied by increased proliferation of corneal basal epithelial cells, limbal expansion of keratin 19 (KRT19)⁺ progenitor cells, increased vascularization of the peripheral cornea, and reduced nerve function in the LG. In addition, they found extensive loss of muscle, intestine, and stomach expression 1 (MIST1)⁺ secretory acinar cells in the Aire^−/− LGs, suggesting that acinar cells are also among the primary targets of the disease. The topical application of ophthalmic steroid enabled the effective restoration of corneal innervation in Aire^−/− mice, demonstrating the link between nerve loss and local inflammation in the dry eye. An Aire KO rat model also recapitulated features observed in patients with APECED, including visual features, organ lymphocytic infiltrations, and the production of autoantibodies.⁶⁵ 

Nerve Degeneration Process

Triple A syndrome (MIM #231550) was previously known as 4A syndrome. This syndrome is an autosomal recessive condition arising from biallelic pathogenic variations in the achalasia–addisonianism–alacrima syndrome (AAAS) gene, whose predicted protein is ALADIN (alacrimia–achalasia–adrenal insufficiency neurologic disorder).⁶⁶ Alacrimia is the first and most consistent feature of this syndrome, resulting from a deterioration of the autonomic innervation of the LG rather than an impaired corneal innervation. Other features may appear within a variable time frame. These include achalasia, features of autonomic dysfunction, neurodegeneration, and insensitivity to adrenocorticotropic hormone, which can result in adrenal insufficiency and isolated glucocorticoid deficiency. In addition to alacrimia, the ophthalmic manifestations include keratoconjunctivitis sicca, corneal melts, LG atrophy, pupillary abnormalities including sluggish pupils, tonic pupils with hypersensitivity to dilute miotics, accommodative dysregulation, amblyopia, and optic atrophy. Lacrimation (both reflex and basal), pupillary miosis, and the process of accommodation are under parasympathetic control.⁶⁷ Mullaney et al.⁶⁸ studied LG in three patients with triple A syndrome, and, by means of orbital imaging, they observed a reduction in LG size; on LG biopsy, this was associated with a reduced number of serous secreting cells. 

Aragona et al.⁶⁹ reported the complete ophthalmologic investigation results of an 18-year-old male patient. Using the Schirmer's test of 0 mm/5 minutes in both eyes, they documented his dry eye condition, which was accompanied by tear hyperosmolarity, mild meibomian gland dysfunction, reduced break-up time, mucus filaments in the tear film, and conjunctival epithelium metaplastic changes. Activated keratocytes and a normal nerve pattern were observed. They concluded that the dry eye appeared to arise from a tear aqueous deficiency. 

By contrast, Botella García et al.⁷⁰ successfully used topical cyclosporine A 0.05% in a patient with hypolacrimia arising from Allgrove syndrome. By inhibiting IL-2 activation of lymphocytes, cyclosporine A produces immunomodulatory and antiinflammatory effects. The authors hypothesized that the patient may suffer from the anatomical underdevelopment of both the main LG and the functional accessory LG of Krause and Wolfring. Cyclosporine A treatment was effective. It decreased the overall inflammation of the ocular surface and recovered goblet cells. 

The precise role of Aladin is not understood. It normally localizes to the cytoplasmic face of the nuclear membrane, as part of the nuclear pore complex (NPC), which is involved in nucleocytoplasmic transport. Bitetto et al.⁷¹ explored the consequences of the homozygous AAAS pathogenic variation c.464G>A (p.R155H) in the central nervous system tissues and fibroblasts of a patient presenting motor neuron disease, cerebellar ataxia, and autonomic dysfunction. By neuropathological analyses, a significant reduction in the perinuclear expression of Aladin was demonstrated, associated with a severe loss of motor neurons and Purkinje cells. 

Alacrimia, achalasia, and mental retardation syndrome (AAMR; MIM #615510) is an autosomal recessive condition caused by biallelic pathogenic variations in the guanosine diphosphate (GDP)-mannose pyrophosphorylase A (GMPPA) gene. Alacrimia, achalasia, and developmental delay occur at birth or in early infancy. Other features can include hypotonia, gait disturbance, anisocoria, and hearing and visual deficits. Although the disorder resembles the triple A syndrome, it never includes adrenal insufficiency. The clinical features of the patients with pathogenic variations indicate that GMPPA plays a significant role in neurons, in autonomic nerve fibers and in the innervation of the distal esophageal sphincter and the LG.⁷² 

GDP-mannose pyrophosphorylase B (GMPPB) catalyzes the conversion of mannose-1-phosphate and guanosine-5′-triphosphate (GTP) to GDP-mannose, which is required for glycosylation. Hypoglycosylation of α-dystroglycan (α-DG) linked to GMPPB defects causes muscle disease.⁷³ Alpha-DG belongs to a protein complex that stabilizes myofibers by linking the extracellular matrix with the cytoskeleton.⁷⁴ Franzka et al.⁷⁵ showed that Gmppa KO mice recapitulate cognitive and motor deficits observed in patients with AAMR. GMPPA is an allosteric feedback inhibitor of GMPPB. GMPPA defects generate an enhancement of mannose incorporation into glycoproteins, including α-DG in mice and men. Thus, α-DG turnover is increased and α-DG abundance is decreased. 

Marom et al.⁷⁶ reported on a consanguineous Israeli Arab family with five males in two interrelated families with intellectual disabilities, alacrimia, achalasia, and mild autonomic dysfunction, naming this the X-linked mental retardation 17 syndrome (MRX17) condition. The phenotype was similar to those described in AAMR. The pedigree was compatible with either X-linked or autosomal recessive inheritance. Linkage to chromosome X was suspected, given that genotyping of affected family members identified a 16.4-Mb continuous segment of identical alleles shared by the patients between markers rs2748314 and rs5906782 on Xp11.23-p21. However, no molecular anomaly has been identified since then, nor has any other family compatible with the X-linked phenotype been described. It is thus probable that MRX17 is not a real clinical entity. 

Hereditary sensory and autonomic neuropathy type III (HSAN3; MIM #223900) is an autosomal recessive neurodegenerative disorder linked to the elongator complex protein 1 (ELP1) gene, with onset soon after birth and affecting mostly the Ashkenazi Jewish population. Progressive depletion of sensory proprioceptive and autonomic neurons leads to progressive symptoms, including gastrointestinal dysfunction, gastroesophageal reflux, vomiting crises, recurrent pneumonia, seizures, gait abnormalities leading to loss of ambulation, kyphoscoliosis, orthostatic hypotension, hypertension crises, absence of fungiform tongue papillae, decreased deep tendon reflexes, defective lacrimation, insensitivity to pain and temperature, and finally death. Only 50% of patients reach 40 years of age.⁷⁷ 

This neuropathy is caused by a “leaky” mRNA splicing defect that results in reduced levels of the inhibitor of kappa B kinase complex-associated protein (IKBKAP, also referred to as IKAP and ELP1).⁷⁸ Homozygous mutant cells express both wild-type and mutant inhibitor of IKBKAP mRNA and produce small amounts of full-length functional IKAP, as the pathogenic variation weakens but does not completely abolish the 5′ splice site of exon 20.⁷⁹ There is a variability of the relative expression of wild-type and mutant IKBKAP transcripts between tissues. The lowest levels of wild-type IKBKAP mRNA and IKAP are observed in the central and peripheral nervous systems. In vivo studies have highlighted that the loss of IKAP (ELP1) leads to neuronal cell death because of failed tissue innervation rather than abnormal neuronal migration.⁸⁰ 

HSAN6 (MIM #614653) is an autosomal recessive disease linked to biallelic dystonin (DST) pathogenic variations. It was first identified in infants with lethal autonomic sensory neuropathy presenting clinical features similar to familial dysautonomia associated with contractures.⁸¹ Symptoms can begin at birth, during childhood, or during adulthood in milder forms. Patients who present the disease at birth have been reported as having hypotonia, respiratory distress, absent deep tendon reflexes, and autonomic instability including hyperthermia and hypothermia; blood pressure and heart and respiratory rate lability; and alacrimia with corneal injury. In milder forms, clinical features consist predominantly of dysautonomia, insensitivity to pain, and peripheral sensory neuropathy, complicated with ulceration of the skin of the soles and palms.^82–85 

The human DST gene has multiple promoters that lead to different isoforms, with a specific tissue and subcellular distribution that has yet to be fully characterized. Dystonin is a large cytoskeleton linker protein.⁸⁶ HSAN6 appears to be caused by the disruption of neuronal isoform dystonin-a2. It is speculated that the truncation of pathogenic variations that result in the loss of expression of all neuronal isoforms produces a severe disorder with congenital defects and early lethality, whereas pathogenic variation combinations that partially maintain dystonin protein expression and function will have less severe clinical expression and will generate neuropathies without major reduction of lifespan.^87,88 In vitro studies have shown that dystonin is significantly more abundant in cells of familial dysautonomia patients bearing IKBKAP pathogenic variations in comparison with control fibroblasts.⁸¹ This suggests that upregulation of dystonin could be an adaptive cellular response in HSAN3 patients. 

Recessive Dst pathogenic variations in mice are responsible for dystonia musculorum (Dst^dt), a sensory neuropathy.⁸⁹ Affected mice exhibit ataxia, autonomic disturbances, and ultimately death linked to massive degeneration of the sensory neurons in the dorsal root ganglion. The phenotype is partially rescued by restoring the dystonin-a2 expression in neuronal tissues. Recent investigation of Dst^dt sensory neurons revealed an accumulation of autophagosomes, suggesting an autophagy defect.⁹⁰ Defects in cell adhesion and migration consequent to Dystonin pathogenic variations may interfere with neurite outgrowth and guidance and could contribute to neurodegeneration by cytoskeletal impairment and consequent axonal transport defects. 

Hereditary sensory and autonomic neuropathy type 1B (HSAN1B), with cough and gastroesophageal reflux (MIM #608088), was described in 2002 by Spring et al.⁹¹ Adult-onset patients suffered from unexplained chronic coughing that could lead to syncopes and gastroesophageal reflux, associated with sensory loss starting with the lower limbs and a loss of tendon reflexes due to axonal neuropathy concerning unmyelinated and myelinated axons. Some patients also suffered from sensorineural hearing loss. In 2005, Spring et al.⁹² precisely described the phenotype in 17 affected individuals from two families. Alacrimia, as diagnosed by Schirmer's test, was identified in two individuals of the same family. 

In 2003, in a study involving in two families, Kok et al.⁹³ used genome-wide screening to identify a 3.42-cM interval containing the locus for HSAN1B on chromosome 3p22-24. The locus contained 28 mapped genes. Two candidate genes were then discussed: topoisomerase II β (TOP2B) and solute-carrier family 4 member 7 (SLC4A7). However, these were not taken into consideration in HSAN1B in later studies. To our knowledge, the molecular basis of HSAN1B remains unknown, with no additional individuals with HSAN1B and alacrimia having been reported subsequently. 

Fabry disease (FD; MIM #301500) is a treatable multisystem X-linked disease caused by a defect in the alpha-galactosidase A (GLA) gene and the consequent accumulation of toxic metabolites such as globotriaosylceramide (Gb3) and globotriaosylsphingosine (lyso-Gb3).⁹⁴ 

The classic disease manifestations of FD in males include the characteristic angiokeratoma, corneal opacity (cornea verticillata), neuropathic pain (acroparasthesias) due to small-fiber neuropathy, intolerance to heat, inability to sweat, micro-albuminuria, and increased intima media thickness. Later in life, these patients develop progressive kidney disease with progressive renal failure, cardiac symptoms that may include progressive hypertrophic cardiomyopathy, conduction defects and arrhythmia, atrial fibrillation, valvular disease and coronary artery stenosis, and cerebrovascular disease (stroke). In heterozygous FD females, clinical manifestations, which are highly variable, are usually less severe than in heterozygous FD males, and can be restricted to one specific organ.⁹⁵ 

Ocular signs of FD, including corneal verticillata and corneal dysesthesia resulting from small-fiber neuropathy, are typically among the first symptoms to be identified. Sivley et al.⁹⁶ performed a longitudinal study of ocular manifestations in a cohort of 13 patients with FD. Twelve of the 13 patients had evidence of dry eye, nine of whom were symptomatic. They had inflammatory eyelid disease, including blepharitis, as well as meibomian gland dysfunction, superficial punctate keratitis, and excessive watering of the eyes, or some combination of these. Nine of the 12 with dry eye reported symptoms of foreign body sensation, chronic irritation, and/or photophobia. Among them, 89% also had conjunctival lymphangiectasia. 

Putative Neurodegenerative Process

Schinzel–Giedion syndrome (SGS; MIM #269150) is a rare autosomal dominant condition arising from SET binding protein 1 (SETBP1) heterozygous de novo pathogenic variations. This is a multiple congenital malformation syndrome defined by an evocative midfacial retraction, kidney and urinary malformations, and multiple skeletal abnormalities associated with a neurodegenerative process.⁹⁷ Minn et al.⁹⁸ reported two infant patients with SGS with alacrimia, confirmed by Schirmer's test, and corneal hypoesthesia associated with severe visual impairment. The exact origin of the alacrimia and corneal hypoesthesia has not yet been identified but could be related to neurodegenerative processes at the level of the atrophic brainstem, in particular the pons. 

SETBP1 pathogenic variations responsible for SGS occur in a four-amino-acid mutational hotspot.⁹⁷ These variations result in an increased half-life of the mutated SETBP1 protein, causing its accumulation and the subsequent inhibition of the protein phosphatase 2A (PP2A) oncosuppressor through the SETBP1–SET–PP2A axis. SETBP1 is a DNA-binding protein. SETBP1 interacts with genomic DNA through its AT-hook domains, forming a multiprotein complex with host cell factor C1 (HCF1), lysine-specific N-methyltransferase 2A (KMT2A), plant homeodomain-like finger protein 8 (PHF8), and PHF6. SETBP1 gene ontology analysis of deregulated SETBP1 target genes indicates that they are key controllers of visceral organ development and brain morphogenesis. In mouse models, in utero brain electroporation of mutated SETBP1 impairs neurogenesis and is responsible for a profound delay in neuronal migration.⁹⁹ 

Possible Lacrimal Secretory Function Defect

N-glycanase 1 (NGLY1) deficiency (MIM #615273), an autosomal recessive congenital disorder of deglycosylation, is caused by pathogenic variations in the NGLY1 gene. NGLY1 deficiency is characterized by developmental delay, hypolacrimia or alacrimia, seizure, intellectual disability, movement disorders, and other neurological phenotypes.¹⁰⁰ 

The NGLY1 protein has a pivotal role in endoplasmic reticulum–associated degradation processes, cleaving N-glycans from misfolded glycoproteins in the cytosol before they can be degraded by the proteasome. Loss of NGLY1 leads to accumulation of cytoplasmic ubiquitinated proteins, a marker of misfolded proteins in the neurons of the central nervous system.¹⁰¹ 

Tambe et al.¹⁰² identified a resistance to hypotonic lysis in Ngly1-null mouse embryonic fibroblasts, NGLY1 KO human cells, and patient fibroblasts. Ngly1-deficient mouse embryonic fibroblasts swell more slowly and have reduced aquaporin 1 mRNA and protein expression. In NGLY1-deficient mouse embryonic fibroblasts, the activating transcription factor 1 (ATF1)/cAMP-response element binding protein (CREB) signaling pathway was disrupted, leading to reduced aquaporin 1 expression. The aquaporin 1 mRNA level was reduced in patient fibroblasts and NGLY1 KO cells, thus suggesting that NGLY1 may regulate the expression of multiple aquaporins.¹⁰² These results led to the identification of a non-enzymatic, regulatory function of NGLY1 in aquaporin transcription, possibly related to alacrimia and neurological symptoms. Aquaporins are a family of 13-membrane proteins expressed in most exocrine and endocrine secretory glands.¹⁰³ Their role is essential in the transport of fluids across the cell plasma membrane. Aquaporin 5^−/− mice have dry eye symptoms and a modified structure of LG epithelial cells.¹⁰⁴ 

However, the data currently available are not sufficient to exclude the implication of other mechanisms, such as neural degeneration, in the occurrence of alacrimia in patients with an NGLY1 deficiency. 

Currently Unknown Mechanisms

Bosch–Boonstra–Schaaf optic atrophy syndrome (BBSOAS; MIM #615722) is a rare congenital syndrome arising from pathogenic de novo variants in the nuclear receptor subfamily 2 group F member 1 (NR2F1) gene.¹⁰⁵ 

The main features of this syndrome are optic atrophy and intellectual disability. The neurodevelopmental issues are usually severe, with moderate to severe intellectual disability, autistic features, and in certain cases epilepsy. Other features include facial dysmorphism and deafness (20%–40%). Absent or decreased reflex tears are reported in 78% of patients.¹⁰⁶ As with the associated deafness, the exact mechanism of alacrimia in this condition has not yet been investigated, neither in patients nor in mouse models. Cerebral MRIs performed on patients have not focused on LGs. Having noticed that the molecular and cellular mechanisms contributing to visual impairment were still poorly characterized, Jurkute et al.¹⁰⁷ gave a precise report of the ocular phenotype in patients with BBSOAS and in mouse models, but they did not study LG or tear secretions. Regardless, it should be stressed that alacrimia is a clue symptom, and as such can help to diagnose BBOAS in patients with intellectual disability. 

NR2F1 belongs to the nuclear hormone receptor family of steroid hormone receptors. Paired box protein 6 (PAX6) is a key regulator in LG development in mouse and NR2F1 loss of function affects PAX6 expression, suggesting a possible defect of LG development in patients with BBSOAS.²⁵ Conversely, mitochondrial involvement with secondary complex IV deficiency of the mitochondrial respiratory chain has also been documented in patients with BBOAS.¹⁰⁸ Moreover, NR2F1 strongly affects neurogenesis. Thus, a neurogenic mechanism could also be involved in the physiopathology of alacrimia in patients with BBOAS. 

Pathogenic variations in trafficking protein particle complex subunit 11 (TRAPPC11) have been described as causing limb-girdle muscular dystrophy type 2S (MIM #615356).¹⁰⁹ Four patients from two unrelated consanguineous Turkish families described by Koehler et al.¹¹⁰ exhibited achalasia from childhood, congenital alacrimia, no adrenal insufficiency, hyperkeratosis, psychomotor retardation with intellectual disability, epilepsy, muscular weakness and dystrophy, gait abnormalities, short stature and ponderal hypotrophy, and scoliosis, thus expanding the phenotype of limb-girdle muscular dystrophy type 2S. The authors underlined the similarities with triple A syndrome, consisting of a differential diagnosis, as well as AAMR. Although the mechanism of alacrimia remains unexplained, it should be noted that there was no sign of autonomic dysfunction at the ocular level. 

Koehler et al.¹¹⁰ identified a TRAPPC11 homozygous splice pathogenic variation (c.1893+3A>G), resulting in a splicing defect in patients. In doing so, they were the first to describe alacrimia as a feature of this defect. The transport protein particle (TRAPP) is a multiprotein complex with several related but compositionally distinct forms, implicated in multiple processes including endoplasmic reticulum to Golgi transport, intra-Golgi and endosome to Golgi transport, and autophagy. TRAPPC11 belongs to the mammalian TRAPP III complex. Quantification of mRNA and protein suggested that the dysfunction of the protein was due to a very low amount of intact TRAPPC11 transcript in combination with an aberrant splicing product. TRAPPC11 protein levels have been associated with hypoglycosylation of lysosomal-associated membrane protein 1 (LAMP1), which participates in the maintenance of lysosomal integrity and function. Analysis of fibroblasts from patients revealed delayed exit through the Golgi apparatus of a marker protein, indicating a defect in secretory trafficking. However, the exact physiopathology and molecular consequences of TRAPPC11 defect have not yet been fully characterized. 

Both inherited isolated autosomal recessive alacrimia (MIM #601549) and autosomal dominant alacrimia (MIM #103420) have been reported. Hegab and al-Mutawa¹¹¹ described three consanguineous cousins with alacrimia, suggesting an autosomal recessive inheritance of a dysfunction of the LG. 

Hegab et al.¹¹² also reported a Kuwaiti Arab family in which the father and one of his two sons had severe hypolacrimation with blotchy staining of the cornea and punctate staining of the interpalpebral bulbar conjunctiva by fluorescein and rose bengal staining, suggesting an autosomal dominant isolated hypolacrimia. The only other recorded pedigree with autosomal dominant transmission is an Irish family, with five affected members from four generations. These differed from the Kuwaiti family in that they also had atopic dermatitis.¹¹³ 

Conclusions

Hypolacrimia and alacrimia are rare features in which different mechanisms can be involved. For a minority of conditions in which they figure, the pathophysiology is clearly established. However, because LGs have been inadequately studied in human development and childhood physiology, as well as in pathologic situations, most existing explanations remain putative or insufficiently proved. In clinical genetics, hypolacrimia and alacrimia are rarely encountered. Their presence may therefore be useful in supporting a quick diagnosis, leading to the provision of appropriate care and counseling. Their presence may be especially relevant in a number of conditions, including Allgrove syndrome, where hypolacrimia or alacrimia precedes the occurrence of adrenal insufficiency; in N-glycanase deficiency, where other symptoms can mimic mitochondriopathy and lead to invasive tests; and in BBOAS, where the association with optic atrophy in the context of intellectual disability appears to be very specific. 

This review highlights the need to widen the study of hypolacrimia or alacrimia in patients suffering from these conditions, including the clinical evaluation of lacrimation and, where possible, the study of the pathophysiological mechanisms involved. Tests involved could include the measurement of LG volume by MRI and the measurement of tear secretion by Schirmer's tests. Moreover, animal models exist for some of these pathologies. Lacrimal defect should be studied in these models to clarify the complex pathophysiology involved and to increase awareness of the hitherto disregarded ocular conditions of the above-mentioned pathologies. 

Acknowledgments

Disclosure: M. Willems, None; C.F. Wells, None; C. Coubes, None; M. Pequignot, None; A. Kuony, None; F. Michon, None