**4. Genetic aspects**

The genetic basis of cleft lip and palate is extremely complex due to the potential number of genes involved, their behavior (mode of inheritance, gene interaction, penetrance, expressivity, etc.), number of alleles in each gene, independent segregation (two or more genes), epistasis, and gene linkage, in addition to environmental factors that might cause phenocopies. This complexity, added to the difficulties of maintaining and handling the affected animals, has severely limited clinical and genetic studies of orofacial clefts in dogs. Consequently, few are available and these will be summarized as follows.

#### **4.1. Syndromic and nonsyndromic clefts**

Canine oral clefts may be isolated abnormalities, affecting the lip, lip and palate, or only the palate. They may also coexist with abnormalities in other areas of the body. The former are nonsyndromic clefts and the latter are syndromic clefts. The term "syndromic," as used here, is well established and corresponds to a syndrome in a general sense, i.e., a set of abnormalities that occur jointly, but does not necessarily correspond to the concept used in clinical genetics, in which a set of abnormalities can indeed be a syndrome, but also an association or sequence [39].

In dogs, there are no conclusive data on the frequency of each of these two groups. However, the clear perception of veterinary practitioners is that the nonsyndromic forms are far more common than the syndromic. In humans, approximately 70% of cleft lip and palate are isolated abnormalities, while 30% are part of multiple abnormalities due to chromosome aberrations, monogenic inheritance, teratogens, or unknown causes [56].

In veterinary clinics, the common procedure for dogs with multiple abnormalities is immediate euthanasia. This is often performed by the owners or breeders, with no records or study. Consequently, little is known about the syndromic forms of cleft lip and palate.

#### *4.1.1. Syndromic clefts*

We have seen bilateral anophthalmia and cleft lip and palate in mongrels, omphalocele, and cleft palate in Siberian huskies, and anencephaly and cleft palate in Yorkshire terriers, to name three examples. Most of the few reports available have to do with cases in which it was not possible to identify a cause. However, in four cases, a hereditary pattern was established or presumed and, in two cases, the mutation that was responsible was identified [6, 57–59].

In 2015, Wolf et al. [6] studied 13 cases of CL/P with a phenotypic spectrum ranging from bilateral cleft in the nasal wings to complete CLP in Nova Scotia duck tolling retrievers. Furthermore, 10 of the affected animals had syndactyly in the third and fourth toes, varying from incomplete in only one paw to complete in all four paws. As for the other three dogs, whether they had syndactyly was not known. These abnormalities were the result of autosomal recessive inheritance and were a syndromic form of CL/P with variable expressivity. A mutation in the *ADAMTS20* gene was associated with this phenotype. In 2014, Wolf et al. [57] had already identified another mutation in the same breed: an insertion of a LINE-1 in the *DLX6* gene, causing CP and brachygnathia with a pattern of autosomal recessive inheritance. More details on these mutations are given in the section on molecular aspects.

In 1998, Villagómez and Alonso [58] described four individuals from a litter of six Saint Bernard dogs, the offspring of normal parents. They had a cleft palate, bilateral anotia, supernumerary vertebrae and ribs, bifid tongue, and bilateral pedal preaxial polydactyly. In two of these dogs, there was also a cleft lip and one did not have polydactyly. The parents, in four previous gestations, had 28 offspring, 22 of which were normal and 6 had the same clinical phenotype as the four affected individuals. As the parents were normal and had affected male and female offspring, the authors of this report concluded that the abnormalities could be a recessive mutation of an autosomal gene, although the action of teratogens could not be discarded.

In 1985, Sponenberg and Bowling [59] studied a family of Australian shepherds in which there was a syndrome lethal only to the males. The affected animals had a cleft palate and multiple skeletal defects (scoliosis, brachygnathia, short tibia and fibula, polydactyly, syndactyly). In the females, the defects were less severe and there was no cleft palate. The authors of this report raised the hypothesis of X-linked inheritance.

There are also brief reports of omphalocele and bilateral cleft of primary palate in Yorkshire terriers [23], cleft lip and unilateral left-sided anophthalmia in a French bulldog [60], and bilateral cleft of the primary palate, anencephaly, and macroglossia in a dog of unspecified breed [61].

#### *4.1.2. Nonsyndromic clefts*

has severely limited clinical and genetic studies of orofacial clefts in dogs. Consequently, few

Canine oral clefts may be isolated abnormalities, affecting the lip, lip and palate, or only the palate. They may also coexist with abnormalities in other areas of the body. The former are nonsyndromic clefts and the latter are syndromic clefts. The term "syndromic," as used here, is well established and corresponds to a syndrome in a general sense, i.e., a set of abnormalities that occur jointly, but does not necessarily correspond to the concept used in clinical genetics, in which a set of abnormalities can indeed be a syndrome, but also an association or

In dogs, there are no conclusive data on the frequency of each of these two groups. However, the clear perception of veterinary practitioners is that the nonsyndromic forms are far more common than the syndromic. In humans, approximately 70% of cleft lip and palate are isolated abnormalities, while 30% are part of multiple abnormalities due to chromosome aberra-

In veterinary clinics, the common procedure for dogs with multiple abnormalities is immediate euthanasia. This is often performed by the owners or breeders, with no records or study.

We have seen bilateral anophthalmia and cleft lip and palate in mongrels, omphalocele, and cleft palate in Siberian huskies, and anencephaly and cleft palate in Yorkshire terriers, to name three examples. Most of the few reports available have to do with cases in which it was not possible to identify a cause. However, in four cases, a hereditary pattern was established or presumed and, in two cases, the mutation that was responsible was identified [6, 57–59].

In 2015, Wolf et al. [6] studied 13 cases of CL/P with a phenotypic spectrum ranging from bilateral cleft in the nasal wings to complete CLP in Nova Scotia duck tolling retrievers. Furthermore, 10 of the affected animals had syndactyly in the third and fourth toes, varying from incomplete in only one paw to complete in all four paws. As for the other three dogs, whether they had syndactyly was not known. These abnormalities were the result of autosomal recessive inheritance and were a syndromic form of CL/P with variable expressivity. A mutation in the *ADAMTS20* gene was associated with this phenotype. In 2014, Wolf et al. [57] had already identified another mutation in the same breed: an insertion of a LINE-1 in the *DLX6* gene, causing CP and brachygnathia with a pattern of autosomal recessive inheritance.

In 1998, Villagómez and Alonso [58] described four individuals from a litter of six Saint Bernard dogs, the offspring of normal parents. They had a cleft palate, bilateral anotia, supernumerary vertebrae and ribs, bifid tongue, and bilateral pedal preaxial polydactyly. In two of these dogs, there was also a cleft lip and one did not have polydactyly. The parents, in four previous gestations, had 28 offspring, 22 of which were normal and 6 had the same clinical phenotype as the

Consequently, little is known about the syndromic forms of cleft lip and palate.

More details on these mutations are given in the section on molecular aspects.

are available and these will be summarized as follows.

tions, monogenic inheritance, teratogens, or unknown causes [56].

**4.1. Syndromic and nonsyndromic clefts**

154 Designing Strategies for Cleft Lip and Palate Care

sequence [39].

*4.1.1. Syndromic clefts*

Most genetic nonsyndromic clefts occur in families in accordance with the multifactorial inheritance model. However, there are cases in which a Mendelian pattern of inheritance has been documented.

**Monogenic inheritance**. Monogenic inheritance is one that depends on a single gene and the type that has so far been confirmed in dogs is autosomal recessive. In other words, the phenotype only manifests if the individual has two copies of the mutant allele. Like all monogenic inheritance, it has a characteristic pattern as follows and is shown in **Figure 5** [62]:


This pattern of inheritance was registered in cases of nonsyndromic CL/P in dogs of the Brittany spaniel, Pyrenees shepherd, and boxer breeds.

In Brittany spaniels, Richtsmeier et al. [63] studied dogs belonging to an intensely inbred colony. In 12 litters, 52 individuals were born, 14 of which had a cleft palate (CP). One of them also had a cleft lip (CL). In 10 of these 12 litters, the number of males and females was

**Figure 5.** Autosomal recessive inheritance. Consanguineous unions increase the probability that both individuals are heterozygotes, such as couple III-3 X III-4. The risk of recurrence in the offspring of this couple is 25%. The likelihood of having more heterozygous descendants is 2/4 (50%). However, for any one of the normal descendants (male or female) that have already been born, the likelihood is 2/3 (67%).

 registered (15 males and 29 females). Of those affected (11), there were more females than males (9 females and 2 males). In all crossings, the parents were normal.

In Pyrenees shepherd dogs, Kemp et al. [17] analyzed the records of a club for this breed over a 20-year period (1984–2004), corresponding to a population of 2104 dogs. They found 47 cases (24 males and 23 females) born in 37 litters with a total of 163 pups and normal parents. Some were only affected by a CP, while others had a cleft lip with or without a cleft palate (CL ± P).

In boxers, Moura et al. [64] found four affected dogs (two males and two females) in two litters with 11 pups born of a consanguineous union (uncle and niece) between normal individuals. All the dogs had essentially the same phenotype (bilateral CLP). Previously, Turba and Willer [15] had raised the hypothesis that in this breed, CLP had a monogenic autosomal recessive pattern of inheritance.

Bleicher et al. [65] reported a case of cleft palate in a beagle together with its pedigree, which is suggestive of autosomal recessive inheritance. There were five affected individuals of both sexes and, in all crossings, the parents were normal.

An older report on cleft palate is suggestive of autosomal recessive inheritance in bulldogs. It presents 33 pups (24 normal and nine affected) born in six litters of a supposedly heterozygous couple [66].

Regarding autosomal dominant inheritance, two reports have described possible cases in which there was nasal cleft, cleft lip, and cleft palate, occurring separately or in association in Bernese mountain dogs (Bernese sennenhund). An affected male that crossed with a normal female and then with a female German shepherd fathered 26 pups, 11 of which were affected [67, 68]. An abnormality with some similarity was also observed in a Portuguese pointer [69]. However, no further data were published to confirm the mode of inheritance in these dogs.

It should be remembered that, in principle, clefts with different patterns of inheritance could be present in the same lineage, which would hinder the interpretation of the gene segregation mechanism.

**Multifactorial inheritance**. Nonsyndromic clefts are normally distributed in families without following any monogenic pattern of inheritance, but recurrence in generations is undeniable evidence of a genetic basis. The theoretical model that explains this inheritance assumes the contribution of several genes (polygenic inheritance) with an additive effect. The presence of a determined number of liability alleles would create a critical threshold and different degrees of expression of the phenotype, which can also depend on the influence of environmental factors. For instance, if we represent four genes, segregating independently and with the liability alleles identified by the number 2, and that from five number 2 alleles the critical threshold emerges, then several genotypes would be possible (A<sup>1</sup> A2 B1 B2 C1 C2 D<sup>2</sup> D2 ; A2 A2 B1 B2 C2 C2 D1 D2 ; A1 A2 B2 B2 C<sup>2</sup> C2 D<sup>2</sup> D2 ; A<sup>1</sup> A1 B2 B2 C<sup>2</sup> C2 D1 D2 ; etc.). Thus, with any combination of five number 2 alleles, the cleft would occur, and the higher the quantity of these alleles, the more serious it would be, with environmental factors also contributing to this. **Figure 6** illustrates this example. There may also be a principal gene that would have a greater effect than the others. In real situations, the number involved is probably much higher than four genes.

registered (15 males and 29 females). Of those affected (11), there were more females than

**Figure 5.** Autosomal recessive inheritance. Consanguineous unions increase the probability that both individuals are heterozygotes, such as couple III-3 X III-4. The risk of recurrence in the offspring of this couple is 25%. The likelihood of having more heterozygous descendants is 2/4 (50%). However, for any one of the normal descendants (male or female)

In Pyrenees shepherd dogs, Kemp et al. [17] analyzed the records of a club for this breed over a 20-year period (1984–2004), corresponding to a population of 2104 dogs. They found 47 cases (24 males and 23 females) born in 37 litters with a total of 163 pups and normal parents. Some were only affected by a CP, while others had a cleft lip with or without a cleft palate (CL ± P). In boxers, Moura et al. [64] found four affected dogs (two males and two females) in two litters with 11 pups born of a consanguineous union (uncle and niece) between normal individuals. All the dogs had essentially the same phenotype (bilateral CLP). Previously, Turba and Willer [15] had raised the hypothesis that in this breed, CLP had a monogenic autosomal

Bleicher et al. [65] reported a case of cleft palate in a beagle together with its pedigree, which is suggestive of autosomal recessive inheritance. There were five affected individuals of both

An older report on cleft palate is suggestive of autosomal recessive inheritance in bulldogs. It presents 33 pups (24 normal and nine affected) born in six litters of a supposedly heterozy-

Regarding autosomal dominant inheritance, two reports have described possible cases in which there was nasal cleft, cleft lip, and cleft palate, occurring separately or in association in Bernese mountain dogs (Bernese sennenhund). An affected male that crossed with a normal

males (9 females and 2 males). In all crossings, the parents were normal.

recessive pattern of inheritance.

that have already been born, the likelihood is 2/3 (67%).

156 Designing Strategies for Cleft Lip and Palate Care

gous couple [66].

sexes and, in all crossings, the parents were normal.

**Figure 6.** Polygenic inheritance. In this hypothetical pedigree, the individual who inherited at least five number 2 alleles shows the clinical phenotype.

When canine families with high degrees of consanguinity are considered, the critical threshold is more frequent than in families with less or no inbreeding (**Figure 7**). Likewise, the artificial selection process that formed certain breeds led to an increased frequency of liability alleles, making the critical threshold closer than in other breeds and, consequently, leading to a higher frequency of CL/P. As stated previously, there may be a principal gene that increases the risk, as occurs in brachycephalic breeds [70].

**Figure 7.** Distribution of genotypes in polygenic inheritance. Comparison of the threshold between the general population and consanguineous relatives or inbred lines.

#### **4.2. Molecular aspects**

Modern molecular biology techniques and the use of murine models have enabled the identification of many genes that may be associated with CL/P, and, with each new study, the number of candidate genes grows. The evidence suggests that mutations in these genes, in addition to environmental factors, can act alone or interact with several signaling pathways, negatively interfering in the development of the lip and palate [10]. These genes, and the complex signaling pathways with which they interact, are generally highly conserved in vertebrates and therefore a high degree of homology between man and dog is expected. The identification of mutations in canine genes opens up possibilities for identifying human genes and vice versa, as has happened with the discovery of mutations in mice genes [71]. **Table 3** shows several examples of candidate genes related to CL/P in humans and, potentially, in dogs.

Recently, in Nova Scotia duck tolling retrievers (NSDTR) with a cleft palate and other abnormalities, mutations have been identified in two genes: *DLX6*, located in chromosome 14 of the dog (CFA 14), and *ADAMTS20*, located in chromosome 27 (CFA 27).

In the *DLX6* gene, a LINE-1 insertion was found in the intron 2 jointly segregating with the phenotype (CP and brachygnathia) and obeying an autosomal recessive pattern of inheritance. The presence of the LINE-1 insertion disrupts the transcription of the *DLX6* gene in such a way that only 25% of the normal levels of expression occur, which is not sufficient to prevent CP and mandibular abnormalities. It is located in a noncoding region that is highly conserved,


When canine families with high degrees of consanguinity are considered, the critical threshold is more frequent than in families with less or no inbreeding (**Figure 7**). Likewise, the artificial selection process that formed certain breeds led to an increased frequency of liability alleles, making the critical threshold closer than in other breeds and, consequently, leading to a higher frequency of CL/P. As stated previously, there may be a principal gene that increases

Modern molecular biology techniques and the use of murine models have enabled the identification of many genes that may be associated with CL/P, and, with each new study, the number of candidate genes grows. The evidence suggests that mutations in these genes, in addition to environmental factors, can act alone or interact with several signaling pathways, negatively interfering in the development of the lip and palate [10]. These genes, and the complex signaling pathways with which they interact, are generally highly conserved in vertebrates and therefore a high degree of homology between man and dog is expected. The identification of mutations in canine genes opens up possibilities for identifying human genes and vice versa, as has happened with the discovery of mutations in mice genes [71]. **Table 3** shows several

**Figure 7.** Distribution of genotypes in polygenic inheritance. Comparison of the threshold between the general

Recently, in Nova Scotia duck tolling retrievers (NSDTR) with a cleft palate and other abnormalities, mutations have been identified in two genes: *DLX6*, located in chromosome 14 of the

In the *DLX6* gene, a LINE-1 insertion was found in the intron 2 jointly segregating with the phenotype (CP and brachygnathia) and obeying an autosomal recessive pattern of inheritance. The presence of the LINE-1 insertion disrupts the transcription of the *DLX6* gene in such a way that only 25% of the normal levels of expression occur, which is not sufficient to prevent CP and mandibular abnormalities. It is located in a noncoding region that is highly conserved,

examples of candidate genes related to CL/P in humans and, potentially, in dogs.

dog (CFA 14), and *ADAMTS20*, located in chromosome 27 (CFA 27).

the risk, as occurs in brachycephalic breeds [70].

158 Designing Strategies for Cleft Lip and Palate Care

population and consanguineous relatives or inbred lines.

**4.2. Molecular aspects**

**Table 3.** Examples of genes (human and dog orthologs) that have been associated with CL/P in humans.

disturbing a binding domain for SUZ12, a molecule that plays a significant regulatory role in the development of the embryo [57]. Dlx genes form an important family for the development of the first branchial arch, regulating genetic programs that direct the formation of the pattern of the maxilla and mandible [72]. The inactivation of *Dlx5* and *Dlx6* in mice causes serious defects in the craniofacial, axial, and appendicular skeleton, leading to perinatal death [73].

In the *ADAMTS20* gene, a deletion of two nucleotides (AA) was found, segregating together with the phenotype (CL/P and syndactyly) and adhering to an autosomal recessive pattern of inheritance. This deletion represents a frameshift mutation in the metalloprotease domain and should cause the truncation of 1461 amino acids of a protein of 1916 amino acids [6]. The *ADAMTS20* gene is a member of a gene family that encode zinc-dependent proteases. In mouse embryos, its expression is detected in the first branchial arch and between the medial nasal processes [74]. In the palatal mesenchyme, it directs the formation and extension of the palatal shelves [5].

In parallel with the study on NSDTR dogs, Wolf et al. [6] conducted a family-based genomewide association analysis in a population of native Guatemalans. They identified a significant association between cases of CL/P and the *ADAMTS20* gene, lengthening the list of candidate genes for the etiology of oral clefts in humans.

#### **4.3. Genetic counseling**

Like any genetic abnormality, the main recommendation in cases of CLP in dogs is that affected individuals should not be crossed, nor should normal couples with affected descendants ever be crossed again. As the majority of oral clefts in dogs appear to be multifactorial or recessive, it should be noted that owners of normal dogs who have had affected offspring are not always willing to follow this recommendation, especially when the dogs have characteristics of their breed that are highly valued. Therefore, if the owners/breeders decide to cross them again, and are sure that the cleft lip or palate is genetic in nature, the risk of recurrence should be seriously taken into consideration [35].

To avoid autosomal recessive clefts, an important strategy is never to cross individuals that are known to be heterozygotes one with another, such as those that have already had affected offspring. When there is a family history of recessive cleft and the zygosity of an individual is not known, consanguineous unions should be avoided. For X-linked recessive phenotypes, normal female offspring of affected father are all carriers, i.e., heterozygotes, and should not be crossed even when the males are normal. For multifactorial clefts, the main strategy is to avoid crossing dogs that have any relationship. This will reduce the probability of reaching the critical threshold [35].
