Canine XX Disorders/Differences of Sex Development (SRY-Negative): A Potential Role for the FOXL2 Gene

Abstract

Background

Disorders/differences of sex development (DSDs) comprise a group of congenital conditions characterized by atypical gonadal and/or anatomical development of the reproductive system. Among them, XX DSD (SRY-negative) represents a rare but clinically relevant condition, defined by the presence of testicular or ovotesticular tissue in chromosomal females lacking the SRY gene. This phenotype has been described in humans and several domestic species, including dogs, cats, and goats.

Summary

Over the past decades, studies in dogs with 78,XX DSD (SRY-negative) have primarily focused on identifying causative mutations in candidate genes, particularly in SOX9. However, only a limited number of cases have been associated with SOX9 copy number variations, suggesting that additional genetic factors remain to be identified. In contrast, FOXL2, a key regulator of ovarian development and maintenance, has been relatively underexplored in canine studies. Functional evidence from mouse models highlights the importance of FOXL2 in ovarian differentiation and maintenance, as well as in craniofacial and skeletal development. 78,XX DSD has been increasingly reported in French Bulldogs, a breed predisposed to craniofacial and ocular abnormalities.

Key Messages

Current evidence indicates that FOXL2 is a biologically plausible candidate gene for 78,XX DSD in dogs, although no causative variants have yet been identified. The observed cryptic relatedness among affected French Bulldogs suggests a shared genetic background. Future studies should include comprehensive sequencing of FOXL2, particularly its GC-rich coding regions, together with detailed clinical phenotyping. Integrating molecular and clinical data may improve understanding of the genetic basis of XX DSD and support more informed breeding strategies.

Introduction

Disorders/differences of sex development (DSDs) in mammals are congenital conditions characterized by atypical development of gonadal and/or anatomical sex and are classified into sex chromosome-related DSD, XX DSD, and XY DSD [1]. In humans, sex chromosome-related DSDs include Turner syndrome (45,X) and Klinefelter syndrome (47,XXY) [2, 3]. The most common forms of XY DSDs are androgen insensitivity syndrome and defects in testosterone biosynthesis, followed by cryptorchidism, hypospadias, gonadal dysgenesis, and persistent Mullerian duct syndrome (PMDS) [4, 5]. In 46,XX DSD cases, the most common cause is a congenital adrenal hyperplasia [6]. Other forms include 46,XX (SRY-negative) testicular or ovotesticular DSD [7].

DSDs have also been documented in other mammals, including dogs, cats, and goats. These conditions represent a considerable challenge for veterinary medicine and breeding management [8]. Canine DSDs encompass a wide spectrum of congenital anomalies, including defects in gonadal development, PMDS, cryptorchidism, and hypospadias [8–11]. Certain canine breeds, such as French Bulldogs, exhibit increased incidence of XX DSD, diagnosed in dogs with a normal female karyotype (78,XX; SRY-negative) and abnormal gonadal development [12, 13]. Clinical manifestations include the presence of ovaries, testes, or ovotestes in various combinations, as well as ambiguous or male external genitalia despite the absence of the Y chromosome and the SRY gene. However, comprehensive phenotypic characterization and reliable epidemiological data remain limited, as most reports are based on case studies or small cohorts. This highlights a gap in current knowledge and the need for systematic studies.

Research on DSDs in domestic animals is important from both veterinary and breeding perspectives. This is particularly relevant in purebred dog populations, where selective breeding and closed gene pools may increase the prevalence of developmental abnormalities. For example, a large proportion of studied Miniature Schnauzers carried a mutation in the AMHR2 gene, responsible for PMDS. Approximately 50% of affected dogs are fertile and may spread the causative variant within the population [14]. PMDS is also associated with an increased risk of tumorigenesis of the reproductive system [15–17].

French Bulldogs are among the most popular canine breeds but also exhibit a high frequency of genetic abnormalities [18, 19], including brachycephalic obstructive airway syndrome, brachycephalic ocular syndrome, as well as skeletal and neurological diseases [20–22]. These findings reflect the consequences of intensive selective breeding and the accumulation of deleterious traits. Ravn-Mølby et al. [23] emphasized the need for changes in French Bulldog breeding practices in the context of brachycephalic syndrome. However, no effective short-term strategy to reduce brachycephalic syndrome prevalence and improve dog welfare has been identified. A proposed long-term strategy involves excluding severely affected dogs from breeding and gradually favoring individuals with milder phenotypes. Importantly, the accumulation of deleterious traits in this breed may also involve less visible conditions, including DSDs. Therefore, understanding the genetic basis of 78,XX DSD may contribute to more responsible breeding practices and improved canine welfare.

Mammalian sex determination is regulated by a complex gene regulatory network involving hundreds of genes and multiple signaling pathways (see review by Eggers and Sinclair [24]). Initially, XX and XY embryos are sexually indistinguishable and develop bipotential gonadal ridges. In XY embryos, the male sex development pathway is initiated by SRY, which upregulates SOX9 expression. The SOX9 gene encodes a transcription factor that drives testis differentiation. In the absence of SRY, SOX9 is not upregulated and remains at basal levels insufficient to activate the testis differentiation pathway. Consequently, the ovarian developmental pathway is established through the expression of ovarian-promoting genes, including RSPO1, FOXL2, and WNT4. SOX9 promotes testicular development and inhibits ovarian pathways, while FOXL2 represses testis-specific genes, including SOX9, ensuring proper ovarian differentiation (Fig. 1a). Thus, sex determination relies on the antagonistic interplay between key regulatory genes [25]. Disruption of this balance can result in DSDs, such as XX testicular or ovotesticular DSD. This review provides an overview of 78,XX (SRY-negative) testicular or ovotesticular DSD, with an emphasis on the potential role of the FOXL2 gene, which remains one of the least investigated genes in canine XX DSD.

Fig. 1.

Schematic representation of selected genes involved in sex determination and gonadal development. a Key genes in male and female pathways leading to testis and ovary development, respectively. b The balance between SOX9/DMRT1 and FOXL2 is essential for proper gonadal development. *These genes also contribute to the maintenance of gonadal identity in adult gonads.

XX DSDs in Dogs: A Literature Overview

Research on 78,XX DSDs began in the 1980s, when the first clinical cases were described [26–28]. Early reports focused on anatomical and cytogenetic characterization of affected dogs. The first molecular investigations of 78,XX DSD were subsequently performed in an American Cocker Spaniel and a German Shorthaired Pointer, and confirmed the absence of the SRY gene [29, 30]. These findings supported the hypothesis that canine XX DSD is caused by mutation(s) in autosomal genes involved in sex determination and differentiation pathways.

Early studies focused on known selected candidate genes and applied targeted Sanger sequencing in limited breed-specific cohorts (particularly American Cocker Spaniels extensively studied by Meyers-Wallen’s laboratory). These analyses did not identify potential causative variants in genes involved in sex determination and gonadal development, including GATA4, FOG2, LHX1, SF1, SOX9, PISRT1, WT1, DMRT1, and LHX9 [31–35]. Additional analyses in individual cases also failed to identify causative variants in RSPO1, SOX9, MAMLD1, or SRD5A2 [36–38].

On the other hand, a single-nucleotide variant (G insertion) upstream of SOX9 has been reported in dogs with XX DSD. In the same study, expression analyses of selected genes, including RSPO1, were performed, and altered RSPO1 expression was reported in affected individuals. However, the proposed regulatory link between the G insertion and RSPO1 expression has not been functionally validated, and a subsequent study did not confirm an association between this variant and the XX DSD phenotype across multiple breeds [39, 40].

In parallel, genome-wide linkage analysis in a pedigree founded by carrier American Cocker Spaniel and Beagle females suggested that a locus for “XX sex reversal” (SRY-negative) may be located on chromosome 29 (CFA29) [41]. As canine chromosome numbering is conserved across breeds, this locus on CFA29 may represent an additional region associated with XX DSD.

Cytogenetic analyses identified structural chromosomal abnormalities in some XX DSD cases, including a Robertsonian translocation involving chromosomes 5 and 23 in a Bernese Mountain Dog with a male phenotype [42]. Based on this finding, the authors suggested that a potential causative genomic region for this XX DSD case may reside on chromosome 23, where the genes FOXL2, PISRT1, and CTNNB1 are located. Additional rare cases, including French Bulldogs with XX DSD and leukocyte chimerism or a Robertsonian fusion involving chromosome 23, further highlight the potential contribution of chromosomal and structural variation to the DSD phenotype [13, 43]. Although such chromosomal aberrations are not considered a common cause of XX DSD, they emphasize the importance of cytogenetic analyses in affected dogs.

Cytogenetic studies suggested that Robertsonian translocations in XX DSD dogs may involve genes located on CFA23, including CLASP2, UBP1, FBXL2, CTNNB1, and FOXL2. However, targeted DNA sequencing of these genes in 15 XX DSD dogs from various breeds did not identify DNA variants associated with the DSD phenotype [44, 45]. These results highlighted the limitations of early candidate-gene approaches, which were largely restricted to coding regions and small cohorts.

Subsequent research increasingly focused on SOX9, a gene crucial for testicular development [46, 47]. In humans, several genetic alterations involving the SOX9 have been reported in 46,XX DSD patients [48, 49], primarily including gene duplications or disruption of long-range upstream regulatory regions that result in ectopic SOX9 expression. These findings suggested that the XX DSD (SRY-negative) phenotype in dogs may also result from SOX9 duplication or alterations affecting regulatory regions.

In two 78,XX DSD dogs, a heterozygous duplication of a chromosome 9 fragment containing the SOX9 gene was identified using microarray-based comparative genomic hybridization [50]. These findings shifted attention toward copy number and structural variation analyses using multiple molecular approaches, including multiplex ligation-dependent probe amplification, fluorescence in situ hybridization, droplet digital PCR, and microarray-based comparative genomic hybridization. Subsequently, three copies of SOX9 were detected in an American Bully and a French Bulldog with 78,XX DSD [12, 40]. Several CNVs upstream of SOX9 have also been reported [13, 40, 51–54], although their functional relevance remains unconfirmed. Copy number variation analyses of NR5A1, DAX1, RSPO1, and SOX3 in 55 XX DSD dogs, including 17 French Bulldogs, did not identify structural variants associated with the phenotype [55]. Studies on SOX9 in canine XX DSD are summarized in Table 1.

Table 1.

Summary of SOX9 and WGS studies in canine 78,XX DSD (2014–2025)

Reference	Karyotype	Studied gene(s)	Methods	Most important results	Case number	French Bulldog number	DSD phenotype
SOX9 gene
[50]	78,XX	SOX9	Array-CGH, qPCR	2 cases with heterozygous duplication of CFA9 fragment with SOX9 coding sequence	7	0	Testicular tissues in XX dogs
[52]	78,XX	SOX9	FISH	No SOX9 duplication detected. Increased CNVR1 copy number (upstream of SOX9) observed in DSD dogs, but not specific to cases	16	1	8 dogs with testis or ovotestis; 8 dogs with unknown gonadal histology
			MLPA
[54]	78,XX	SOX9	FISH	Size variation in the CNVR upstream of SOX9	1	1	Abdominal testes, structures similar to uterus horns and epididymis, male external genitalia
[53]	78,XX	SOX9	GWAS targeted NGS	DNA variants were not associated with DSD. CNV upstream of SOX9 with sequence similarity to MAGI2 5′UTR, potentially containing regulatory element	17 (GWAS)	1 (GWAS)	8 dogs (NGS) with testis or ovotestis; French Bulldog used in GWAS with testes
			Sanger seq		8 (NGS)	0 (NGS)
[13]	78,XX	SOX9	ddPCR	No association with DSD for exon 3, 5′UTR, and CNVR1; one case diagnosed with XX/XY leukocyte chimerism	6	6	5 dogs with 78,XX DSD: 4 cases with testicular tissue; 1 case with ovotesticular tissue
							1 case with leukocyte chimerism (XX/XY)
[40]	78,XX	SOX9	ddPCR	Increased SOX9 upstream copy number in cases, but not across all breeds; one American Bully with 3 copies of SOX9. Previously reported G insertion not associated with DSD in studied breeds	45	12	Cases representing 23 breeds. 11 dogs with ovotestis; 16 with testis; 3 with testis + ovary or ovotestis; 15 with unknown gonadal histology
			FISH
			Sanger seq
[51]	78,XX	SOX9	Sanger seq, qPCR, array-CGH	No causative RSPO1 variant identified; 11 CNVs detected, including a duplication in the SOX9 promoter region in three cases	4	2	A female phenotype, abnormal ano-genital distance, an enlarged clitoris with penis bone. French Bulldogs had uterine horn-like structures
		RSPO1
[12]	78,XX	SOX9	Sanger seq, ddPCR	A single French Bulldog with 3 copies of SOX9	27	27	19 cases with abdominal testes; 4 with testis and ovotestis; 1 with testis and ovary; 1 with ovotestes; 2 with unknown gonadal histology. Dog with 3 copies of SOX9 exhibited skeletal abnormalities
WGS
[56]	78,XX	Whole genome + selected genes	WGS, ddPCR Sanger seq	Increased PADI6 allele frequency in 78,XX DSD dogs, but not specific to DSD	17	11	8 French Bulldogs with testes; 1 with ovotestes; 2 with testis + ovary. 3 American Staffordshire Terriers with testes; 1 with ovotestes; 2 with unknown gonadal histology
[57]	78,XX	Whole genome	WGS	Cryptic relatedness identified in French Bulldogs; 19 chromosomal regions detected with candidate genes (AKAP2, PIWIL2, POLR3A, SH2D4B) potentially involved in DSD	17	11	As above

Array-CGH, microarray-based comparative genomic hybridization; MLPA, multiplex ligation-dependent probe amplification; FISH, fluorescence in situ hybridization; ddPCR, droplet digital PCR; CNVR, CNV region.

Whole-Genome Sequencing Studies in French Bulldogs with XX DSD

Initial studies examined only a limited number of affected dogs with testicular or ovotesticular 78,XX DSD. In recent years, this condition has been diagnosed more frequently in French Bulldogs and American Staffordshire Terriers, particularly in Poland, as also highlighted in a previous review [13]. Whole-genome sequencing (WGS) studies have expanded previous research focused on structural variation and copy number changes in canine XX DSDs. WGS-based studies, performed mainly in French Bulldogs, are summarized in Table 1.

A reanalysis of WGS data revealed cryptic relatedness among French Bulldogs diagnosed in Poland [57]. This finding suggests that the condition in these dogs may be driven by a mutation in a shared genomic region or gene. WGS analyses identified an allele of the PADI6 gene with higher frequency in affected cases, as well as the AKAP2, PIWIL2, POLR3A, and SH2D4B genes, selected based on identity-by-descent analysis [56, 57]. These results indicate a shared genetic background within the studied cohort; however, the identified candidate genes are not directly linked to the canonical FOXL2-SOX9 regulatory pathway, and their contribution to the DSD phenotype remains to be determined.

Although WGS has been applied in these closely related French Bulldogs, a causative DNA variant for 78,XX (SRY-negative) DSD may remain undetected. Standard WGS protocols, particularly those relying on PCR-based library preparation, may provide insufficient coverage of GC-rich regions [58], including FOXL2 (see online suppl. Fig. 1; for all online suppl. material, see https://doi.org/10.1159/000553153), as applied in the study by Nowacka-Woszuk et al. [56].

Alternative approaches, such as PCR-free Illumina library preparation or long-read sequencing technologies (PacBio HiFi or Oxford Nanopore), may improve coverage of GC-rich regions. Krzeminska [59] highlighted that several canine genes involved in mammalian sex development, including AMH, INSL3, NR5A1, FOXL2, and SOX9, contain GC-rich sequences and tandem repeats, which may limit their representation in standard WGS datasets.

Accordingly, read coverage of these genes should be carefully evaluated in WGS analyses. In cases of insufficient coverage, targeted validation approaches – such as modified PCR conditions (additives and “slowdown” cycling) or oligonucleotide-based “disruptor” reagents [60] – may be considered to enable subsequent Sanger sequencing.

FOXL2 as a Key Factor in Ovarian Development: The Least Studied Gene in 78,XX DSD Dogs

As noted by Salamon et al. [44], FOXL2 encodes a transcription factor involved in multiple developmental processes in vertebrates, including ovarian differentiation [61]. It is one of the earliest markers associated with ovarian development and is essential for promoting and maintaining female gonadal identity [62].

Experimental studies based on FOXL2 induction in human somatic cell models have shown that increased FOXL2 activity induces broad transcriptional changes, including downregulation of testis-specific genes such as SOX9, NR0B1, and DHH, as well as reduced expression of early gonadal ridge-associated transcription factors (LHX9 and GATA4). Changes were also observed in ovarian pathway-associated genes, including RSPO1 and WNT4 [63]. The results suggest that FOXL2 contributes to ovarian somatic cell identity by repressing male pathway genes and reprogramming early gonadal somatic cells toward a transcriptional profile characteristic to supporting cells. These results are consistent with dynamic regulation of LHX9 expression during lineage commitment and stage-dependent regulation of RSPO1 and WNT4 during granulosa cell development [64].

FOXL2 also functions postnatally as a “guardian” of ovarian identity by maintaining granulosa cell fate and repressing testicular differentiation pathway [65]. A similar antagonistic role is observed for DMRT1 in testes, where its loss leads to reprogramming of adult Sertoli cells into granulosa-like cells in mouse model [66].

Functional studies in animal models further support the importance of FOXL2. In mice, Foxl2 ablation results in reprogramming of granulosa cells into Sertoli-like cells and formation of seminiferous tubule-like structures in adult ovaries [65]. Foxl2-deficient mice also exhibit craniofacial, skeletal, and eyelid abnormalities, including defects in bone and cartilage formation during embryogenesis and altered postnatal growth [67, 68]. In goats, deletions affecting FOXL2 and PISRT1 (a noncoding RNA associated with the polled intersex syndrome locus) expression result in polled intersex syndrome, characterized by “XX sex reversal” and hornlessness [69]. Biallelic loss of FOXL2 is sufficient to induce XX DSD phenotype, with species-specific differences in the timing of FOXL2 function: in goats, it is essential during fetal development, whereas in mice it is required primarily for postnatal maintenance of ovarian identity [65, 70]. These findings support a conserved role of FOXL2 in ovarian development across mammals; however, they also indicate species-specific differences in its functional timing and associated phenotypes.

No association between FOXL2 variants and ocular or skeletal abnormalities has been reported in dogs, cattle, cats, or pigs, indicating a potential gap in comparative studies. FOXL2 and SOX9/DMRT1 act as antagonistic regulators of sex development (Fig. 1a), and their balanced expression is essential for proper development (Fig. 1b) [25, 66]. Beyond gonadal function, FOXL2 and SOX9 are also implicated in broader developmental processes, including skeletal development [67, 71, 72]. Notably, SOX9 copy number variation has been identified in a dog with 78,XX DSD accompanied by skeletal abnormalities [12].

In humans, FOXL2 mutations are associated with blepharophimosis, ptosis, and epicanthus inversus syndrome (BPES), confirming its role in eyelid development. In some cases, BPES is associated with premature ovarian insufficiency, highlighting FOXL2's dual rolein craniofacial and gonadal development [71, 72].

Given its involvement in eyelid and craniofacial development in humans and mice, FOXL2 may represent a potential candidate gene for investigation in brachycephalic dog breeds, which are predisposed to craniofacial and ocular abnormalities. Brachycephalic dog breeds, such as French Bulldogs, are among the most popular companion animals [19, 73]. Their characteristic craniofacial morphology results from the accumulation of the numerous genetic variants fixed during breeding. In addition to skeletal abnormalities and brachycephalic obstructive airway syndrome, these breeds are predisposed to ocular conditions, including corneal ulcers, corneal pigmentation, corneal fibrosis, and entropion [74]. Entropion, characterized by the inward rolling of the eyelid, is frequently associated with abnormalities in eyelid length or shape. Several breeds, including Basset Hound, Bloodhound, Chow Chow, Great Dane, Golden Retriever, Labrador Retrievers, and Bulldog, show increased susceptibility to entropion and ectropion [75]. These phenotypes involve tissues in which developmental pathways related to FOXL2 activity may be relevant. However, further studies are required to clarify this association in dogs.

Despite its biological relevance, FOXL2 has been poorly characterized in canine genetic studies. To date, the FOXL2 gene has been analyzed in a single molecular study of 78,XX DSD dogs, including one French Bulldog [44]. The authors reported that only part of the gene could be analyzed due to technical limitations. According to current genomic databases and gene annotations, primers used in that study targeted a region downstream of the coding sequence (online suppl. Fig. 2). In addition, canine FOXL2 coding sequence is more GC-rich than the human ortholog (see online suppl. Fig. 3), which may complicate sequencing and read mapping. Therefore, potential FOXL2 variants may have been overlooked in previous WGS studies, particularly when standard protocols were applied without evaluating read coverage for the FOXL2 locus [56, 57]. This highlights a methodological gap in prior analyses and supports the need for targeted sequencing of the entire FOXL2 coding region in further studies.

Conclusion

Studies indicate that mutations in SOX9, the most extensively investigated candidate gene, explain only a small fraction of testicular or ovotesticular 78,XX (SRY-negative) DSD cases in dogs. Although multiple candidate genes have been studied, the genetic basis of this condition remains largely unknown. In contrast, FOXL2, a key regulator of ovarian development, has received limited attention in canine studies.

The observed relatedness among affected French Bulldogs in Poland suggests a shared genetic variant may contribute to this condition. Considering its roles in ovarian, craniofacial, and skeletal development, FOXL2 represents a biologically plausible candidate gene for further investigation.

Future studies should include comprehensive sequencing of the entire FOXL2 gene using methods optimized for GC-rich regions, together with expression and epigenetic analyses. Integration of detailed clinical phenotyping would improve genotype-phenotype correlations and enhance diagnostic accuracy for 78,XX DSD. Overall, combining molecular and clinical approaches may provide a clearer understanding of the genetic background of this condition. Given the antagonistic roles of FOXL2 and SOX9 in gonadal development, further investigation of both genes is warranted.

Acknowledgments

The author declares that ChatGPT (OpenAI) was used solely for language editing and stylistic improvement. The author takes full responsibility for the content of the manuscript.

Conflict of Interest Statement

The author has no conflicts of interest to declare.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

P.K. – writing and editing the manuscript.

Funding Statement

This study was not supported by any sponsor or funder.