**Molecular Basis of Inherited Diseases in Companion Animals**

Editors

**Danika Bannasch Steven Friedenberg**

MDPI • Basel • Beijing • Wuhan • Barcelona • Belgrade • Manchester • Tokyo • Cluj • Tianjin

*Editors* Danika Bannasch University of California USA

Steven Friedenberg University of Minnesota College of Veterinary Medicine USA

*Editorial Office* MDPI St. Alban-Anlage 66 4052 Basel, Switzerland

This is a reprint of articles from the Special Issue published online in the open access journal *Genes* (ISSN 2073-4425) (available at: https://www.mdpi.com/journal/genes/special issues/Companion Animals).

For citation purposes, cite each article independently as indicated on the article page online and as indicated below:

LastName, A.A.; LastName, B.B.; LastName, C.C. Article Title. *Journal Name* **Year**, *Volume Number*, Page Range.

**ISBN 978-3-0365-0472-8 (Hbk) ISBN 978-3-0365-0473-5 (PDF)**

Cover image courtesy of Katy Robertson.

© 2021 by the authors. Articles in this book are Open Access and distributed under the Creative Commons Attribution (CC BY) license, which allows users to download, copy and build upon published articles, as long as the author and publisher are properly credited, which ensures maximum dissemination and a wider impact of our publications.

The book as a whole is distributed by MDPI under the terms and conditions of the Creative Commons license CC BY-NC-ND.

### **Contents**



**Anna Letko, Katie M. Minor, Steven G. Friedenberg, G. Diane Shelton, Jill Pesayco Salvador, Paul J. J. Mandigers, Peter A. J. Leegwater, Paige A. Winkler, Simon M. Petersen-Jones, Bryden J. Stanley, Kari J. Ekenstedt, Gary S. Johnson, Liz Hansen, Vidhya Jagannathan, James R. Mickelson and Cord Drogem ¨ uller ¨**

A *CNTNAP1* Missense Variant Is Associated with Canine Laryngeal Paralysis and Polyneuropathy Reprinted from: *Genes* **2020**, *11*, 1426, doi:10.3390/genes11121426 .................. **205**

### **About the Editors**

**Danika Bannasch** earned her DVM degree from UC Davis School of Veterinary Medicine and her PhD degree in mouse molecular genetics at Princeton University. She is currently Professor at the Department of Population Health and Reproduction in the School of Veterinary Medicine, University of California, Davis, and is the first faculty member to hold the prestigious Maxine Adler Endowed Chair in Genetics. An accomplished veterinary geneticist, Bannasch focuses on identification of the molecular causes of inherited diseases in dogs and horses in her research. Her laboratory has identified the DNA changes responsible for lethal white foal syndrome, hereditary equine regional dermal asthenia, hoof wall separation syndrome, hyperuricosuria, Alaskan Husky encephalopathy, cleft palate, cleft lip and palate, spinal dysraphism, glioma susceptibility, and chondrodystrophy. Important research findings have also led to animal models being used for similar human diseases. By studying naturally occurring diseases in animals, the Bannasch laboratory is involved in discoveries covering a triad of significant advances: the development of diagnostic tests to aid animal breeders; the identification of novel genes and pathways as candidates for human disease; and an understanding of basic molecular mechanisms of disease.

**Steven Friedenberg** is Assistant Professor of Small Animal Emergency and Critical Care Medicine and Genetics at the University of Minnesota College of Veterinary Medicine. Dr. Friedenberg received his DVM from Cornell University and his PhD from North Carolina State University. He is board-certified by the American College of Veterinary Emergency Critical Care (ACVECC) and currently educates and trains veterinary students, sees patients, and undertakes research at the University of Minnesota. Dr. Friedenberg's primary interests include autoimmune disorders such as Addison's disease and autoimmune hemolytic anemia (IMHA). His research focuses on understanding the genetic and immunologic mechanisms that cause these diseases in dogs. Additionally, he studies other complex genetic traits, in particular, various canine cardiac and neurologic disorders. Dr. Friedenberg is interested in applying "big data" methods to veterinary medicine to help improve patient care and outcomes in a wide variety of clinical disorders.

### **Preface to "Molecular Basis of Inherited Diseases in Companion Animals"**

The field of companion animal genetics has evolved rapidly since the publication of the dog genome in 2005 and the cat genome in 2007. Over the past 15 years, our community has made major advances in understanding the genetic basis of many inherited diseases in companion animals. These discoveries have helped eliminate or significantly reduce the incidence of many life-limiting conditions in our pets while also demonstrating the importance of companion animal diseases as models for similar disorders in humans. All 15 manuscripts in this Special Issue describe inherited disorders in companion animals with parallels in humans. This book celebrates the rapidly growing and evolving field of companion animal genetics by demonstrating how cutting-edge tools can be employed to help us understand the inherited basis of diseases.

> **Danika Bannasch, Steven Friedenberg** *Editors*

### *Editorial* **Special Issue "Molecular Basis of Inherited Diseases in Companion Animals"**

**Steven G. Friedenberg 1,\* and Danika L. Bannasch <sup>2</sup>**

<sup>2</sup> Department of Population Health & Reproduction, School of Veterinary Medicine, University of California Davis, Davis, CA 95616, USA; dlbannasch@ucdavis.edu

The study of inherited diseases in companion animals has exploded over the past 15 years since the publication of the first dog genome in 2005 [1] and the cat genome in 2007 [2]. Since then, countless tools and resources have been developed allowing researchers to exploit these genomes to study inherited diseases and traits in companion animals at an unprecedented pace. According to the Online Mendelian Inheritance in Animals (OMIA) database [3], as of December 2020, there are 784 single-locus diseases or traits that have been explained in dogs and 361 in cats. Identification of the genetic polymorphisms that underlie these diseases and traits has allowed us to reduce the incidence of many inherited disorders and explain much of the phenotypic diversity seen in our companion animals. Furthermore, many of these now well-characterized inherited diseases in companion animals offer potential models for similar conditions in humans.

One notable trend in companion animal genetics over the past several years has been the rapidly increasing use of whole-genome resequencing as a tool for identifying genetic variants associated with disease. Indeed, the manuscripts that comprise this Special Issue reflect this ongoing trend: Of the 15 articles in this issue, 11 employed wholegenome resequencing to identify likely causative mutations. What is even more remarkable, however, is that over half of these 11 manuscripts employed whole-genome resequencing exclusively as a means of identifying putative causative mutations without making use of traditional marker arrays.

A major driver of this trend is undoubtedly the rapidly falling costs of whole-genome resequencing, along with the increasing availability of computational resources required to process and analyze these large datasets. Perhaps an even more important driver, however, has been the development of consortium-driven resources to pool and share whole-genome resequencing data with investigators around the world. This includes resources such as the 99 Lives Cat Genome Consortium [4] and the Dog Variant Database and Biomedical Consortium [5]. These databases of known genetic variation allow researchers to quickly compare a particular genome of interest to hundreds or thousands of already sequenced animals in order to determine whether a potentially pathogenic allele is unique to an animal with a particular trait or condition. This process allows for a rapid filtering of millions of variants to hundreds or tens of variants that can then be prioritized rapidly based upon the currently understood function of a particular gene. Next-generation reference genomes built using long-range sequencing technology [6,7], along with ever-improving genome annotations, are also rapidly improving the feasibility of using whole-genome resequencing to identify variants of interest for a particular trait or condition.

A natural consequence of the increasing use of whole-genome resequencing for our companion animals is the opportunity to provide truly precision medicine for individual patients. As veterinarians and geneticists, we are often confronted with unique cases with abnormalities that may be specific to a particular animal. In some cases, sequencing the

**Citation:** Friedenberg, S.G.; Bannasch, D.L. Special Issue "Molecular Basis of Inherited Diseases in Companion Animals". *Genes* **2021**, *12*, 68. https:// doi.org/10.3390/genes12010068

Received: 21 December 2020 Accepted: 5 January 2021 Published: 7 January 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

<sup>1</sup> Department of Veterinary Clinical Sciences, College of Veterinary Medicine, University of Minnesota, Saint Paul, MN 55108, USA

**<sup>\*</sup>** Correspondence: fried255@umn.edu

animal's entire genome may provide an opportunity to arrive at a diagnosis in a way that traditional medical testing could not. This trend toward precision medicine is also reflected in this Special Issue, as four articles identified disease-causing mutations that are believed to be specific to only one animal. As whole-genome resequencing costs continue to decline and this technique is even more widely applied in veterinary hospitals around the world, we will begin to gain a better grasp on understanding the breadth of disorders we see among companion animals, along with their underlying genetic roots. Undoubtedly, this work will also lead to new spontaneous models of animal disease that can inform our understanding of similar diseases in humans and other species as well.

Despite the advances afforded by whole-genome resequencing, one area of genetics where this technology alone is unlikely to provide a complete understanding is complex inherited traits. These traits, which are polygenic by nature, are likely to continue to require a combination of marker arrays, whole-genome resequencing, and other approaches such as selection mapping in order to fully understand the contribution of genetic variation to the incidence of disease. Two of the articles in this Special Issue, one on diabetes [8] and the other on obesity [9], reflect the challenges associated with understanding the genetic basis of complex traits in companion animals. Because many common diseases we see in companion animal medicine are likely complex traits (e.g., autoimmune disorders, breed-associated cancers), these areas remain ripe for ongoing research as we continue to improve the tools and resources we have at our disposal as geneticists.

One last point regarding the articles in this Special Issue which we would be remiss to overlook is the degree of overlap between companion animal and human disorders. In fact, all 15 manuscripts in this Special Issue describe inherited disorders in companion animals with parallels in humans. This observation underscores the value of ongoing "One Health" approaches to medicine, which are meant to recognize the connections between the health of people, animals, and the environment. Notably, in this Special Issue, these connections were made by collaborations between veterinarian scientists and, in most cases, Ph.D. scientists. That all of these manuscripts were co-authored by veterinarians highlights the many advantages of dogs and cats over more traditional model organisms: a shared living environment, the breadth and depth of quality medical diagnostics and treatments, and the dedication of animal owners around the world that continues to drive this field forward. The disease parallels between animals and humans highlight the importance of companion animals in providing sources of spontaneous disease models for similar conditions in humans that would be difficult to re-create in a laboratory setting.

In summary, this issue celebrates the rapidly growing and evolving field of companion animal genetics by demonstrating how cutting-edge tools can be employed to help us understand the inherited basis of diseases. As new reference genomes and improved sequencing technologies continue to emerge and enhance our ability to understand inherited disorders, we are confident that many more exciting discoveries in the world of companion animal genetics are certain to emerge.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### **A Missense Variant A**ff**ecting the C-Terminal Tail of UNC93B1 in Dogs with Exfoliative Cutaneous Lupus Erythematosus (ECLE)**

**Tosso Leeb 1,2,\*, Fabienne Leuthard 1,2, Vidhya Jagannathan 1,2, Sarah Kiener 1,2, Anna Letko 1,2, Petra Roosje 2,3, Monika M. Welle 2,4, Katherine L. Gailbreath 5, Andrea Cannon 6, Monika Linek 7, Frane Banovic 8, Thierry Olivry 9, Stephen D. White 10, Kevin Batcher 11, Danika Bannasch 11, Katie M. Minor 12, James R. Mickelson 12, Marjo K. Hytönen 13,14,15, Hannes Lohi 13,14,15, Elizabeth A. Mauldin <sup>16</sup> and Margret L. Casal <sup>16</sup>**


Received: 20 December 2019; Accepted: 31 January 2020; Published: 3 February 2020

**Abstract:** Cutaneous lupus erythematosus (CLE) in humans encompasses multiple subtypes that exhibit a wide array of skin lesions and, in some cases, are associated with the development of systemic lupus erythematosus (SLE). We investigated dogs with exfoliative cutaneous lupus erythematosus (ECLE), a dog-specific form of chronic CLE that is inherited as a monogenic autosomal recessive trait. A genome-wide association study (GWAS) with 14 cases and 29 controls confirmed a previously published result that the causative variant maps to chromosome 18. Autozygosity mapping refined the ECLE locus to a 493 kb critical interval. Filtering of whole genome sequence data from two cases against 654 controls revealed a single private protein-changing variant in this critical interval, *UNC93B1*:c.1438C>A or p.Pro480Thr. The homozygous mutant genotype was exclusively observed in 23 ECLE affected German Shorthaired Pointers and an ECLE affected Vizsla, but absent from 845 controls. UNC93B1 is a transmembrane protein located in the endoplasmic reticulum

and endolysosomes, which is required for correct trafficking of several Toll-like receptors (TLRs). The p.Pro480Thr variant is predicted to affect the C-terminal tail of the UNC93B1 that has recently been shown to restrict TLR7 mediated autoimmunity via an interaction with syndecan binding protein (SDCBP). The functional knowledge on UNC93B1 strongly suggests that p.Pro480Thr is causing ECLE in dogs. These dogs therefore represent an interesting spontaneous model for human lupus erythematosus. Our results warrant further investigations of whether genetic variants affecting the C-terminus of UNC93B1 might be involved in specific subsets of CLE or SLE cases in humans and other species.

**Keywords:** *Canis familiaris*; dermatology; immunology; animal model; skin; TLR7; toll-like receptor; syndecan binding protein; syntenin-1; systemic lupus erythematosus; SLE; CLE

#### **1. Introduction**

In humans, cutaneous lupus erythematosus (CLE) represents a group of lupus erythematosus (LE)-associated autoimmune skin diseases exhibiting a cell-rich interface dermatitis leading to erosions and ulcerations with subsequent scarring, disfiguration and decreased quality of life [1–4]. CLE can affect only the skin or be present as part of a diverse range of potentially life-threatening and debilitating symptoms in patients with systemic lupus erythematosus (SLE) [1–4].

The incidence of CLE has been reported at ~4 cases per 100,000 persons per year [5–8]; 10% to 30% of human patients with CLE exhibit a transition from cutaneous into SLE forms, suggesting shared pathways and genetic background relevant to both cutaneous and systemic manifestations [5,6,9].

It has been proposed that some CLE forms, similarly to SLE, have an underlying genetic predisposition that combines with environmental factors to elicit an abnormal immune response with a continuous activation of the innate immune system. Several genetic associations have been identified in human CLE, with the majority of them involving type I interferon pathways, cell death and clearance of cell debris, antigen presentation and immune cell regulation [10,11]. To date, a single monogenic form of CLE caused by heterozygous variants in the *TREX1* gene encoding the three prime repair exonuclease has been identified in human patients with familial chilblain lupus erythematosus [12]. The pathogenic *TREX1* variants lead to chronic hyperactivation of the type I interferon system via cytosolic DNA recognition pathways [11,13]. A rare monogenic form of SLE in humans is caused by variants in the *DNASE1* gene encoding deoxyribonuclease 1 [14]. Mice deficient for Dnase I also develop an SLE-like autoimmune disease [15].

Dogs may also suffer from various forms of CLE, some of which resemble or are identical to their human homologs [4]. The so-called exfoliative cutaneous lupus erythematosus (ECLE) is a dog-specific variant of chronic CLE that has a very strong hereditary component and appears to be inherited as a monogenic autosomal trait [16–18]. Despite its current designation, signs of ECLE are not restricted to the skin. In most patients, ECLE starts with characteristic skin lesions in juvenile or young adult dogs (Figure 1). In later stages, ECLE often additionally affects the joints with severe pain, but a progression to classic antinuclear antibody-positive SLE is usually not seen [4,16–18]. The treatment of ECLE-affected dogs with immunomodulatory drugs often is insufficient to achieve long-lasting control of the disease, leading to a guarded prognosis [18,19]. Dogs affected with ECLE often are euthanized due to the severity of their disease. ECLE has been observed in several closely related hunting dog breeds, German Shorthaired Pointers, Braques du Bourbonnais, and Vizslas.

**Figure 1.** Exfoliative Cutaneous Lupus Erythematosus (ECLE) phenotype. (**A**) Scarring alopecia, generalized hair loss and adherent crusts on the face of a 2-year-old male dog. (**B**) Erythematous lesions on the back of a 1.5-year old male dog. (**C**) Close up of patchy lesions on the abdomen. (**D**) Haired skin from an ECLE affected dog with typical histological changes that include a cell-rich interface inflammation with frequent basal keratinocyte apoptosis (arrows). Hematoxylin and eosin stain.

A previously reported genome-wide association study (GWAS) mapped the causative genetic defect for ECLE to chromosome 18, but the causative variant has not yet been identified [20]. The best-associated marker was located at position 53,913,829 (CanFam 2) [20], which corresponds to 50,888,317 in the current CanFam 3.1 assembly.

In the present study, we performed a new GWAS followed by a whole genome sequencing approach with the goal to identify the causative genetic variant for ECLE in dogs.

#### **2. Materials and Methods**

#### *2.1. Ethics Statement*

All the dogs in this study were privately owned and samples were collected with the consent of their owners. The collection of blood samples was approved by the "Cantonal Committee for Animal Experiments" (Canton of Bern; permit 75/16).

#### *2.2. Animal Selection*

This study included 877 dogs. They consisted of 552 German Shorthaired Pointers (26 ECLE cases/526 controls), 52 unaffected German Longhaired Pointers, 210 unaffected German Wirehaired Pointers, 7 unaffected Braques du Bourbonnais, and 56 Vizslas (1 ECLE case/55 controls). The 27 ECLE cases were diagnosed by licensed veterinarians. The 850 dogs classified as unaffected represented population controls without reports of severe immunological or skin-related health issues. Peripheral blood samples were collected in EDTA vacutainers and stored at −20◦C. Additional details on samples are given in Table S1.

#### *2.3. DNA Extraction and SNV Genotyping*

Genomic DNA was either available from a previous study [20], isolated from EDTA blood with the Maxwell RSC Whole Blood Kit using a Maxwell RSC instrument (Promega, Dübendorf, Switzerland), or from formalin-fixed paraffin-embedded (FFPE) tissue samples using the Maxwell RSC DNA FFPE kit according to the manufacturer's instructions. DNA from 14 ECLE cases and 29 controls was genotyped on illumina\_HD canine BeadChips containing 220,853 markers (Neogen, Lincoln, NE, USA). The raw SNV genotypes are available in File S1. We did not have complete pedigree information on all 43 dogs that were genotyped on the SNV arrays. Some of the dogs were closely related, including, for example, 5 cases that were full siblings. Table S2 lists the pairwise IBD between all dogs and gives an objective measure of the relatedness between the genotyped dogs. A multiple dimension scaling (MDS) plot is shown in Figure S1.

The previously published GWAS [20] had been done with Affymetrix v2 127 k SNV genotyping arrays. A total of 6 cases and 2 controls were shared between the two analyses. The other 35 samples herein were from dogs different from those of the previous study.

For some dogs from the previous study [20] only very little DNA was left. The remaining DNA of 8 German Shorthaired Pointers was used up for SNV genotyping on the illumina\_HD canine BeadChips. In these dogs, no specific targeted genotyping could be performed (see Section 2.8 below).

#### *2.4. GWAS and Autozygosity Mapping*

We used PLINK v.1.9 for basic file manipulation of the SNV genotypes [21]. We removed markers and individuals with less than 90% call rates. We further removed markers with minor allele frequency of less than 10% and markers deviating from the Hardy–Weinberg equilibrium in controls with a p-value of less than 10<sup>−</sup>5. An allelic GWAS was then performed with the GEMMA 0.98 software using a linear mixed model including an estimated kinship matrix as covariable to correct for the genomic inflation [22]. Manhattan and QQ plots of the corrected p-values were generated in R [23].

For autozygosity mapping, the genotype data of 14 ECLE cases were used. A tped-file containing the markers on chromosome 18 was visually inspected in an Excel spreadsheet to find a homozygous shared haplotype in the cases (Table S3).

#### *2.5. Whole Genome Sequencing of Two A*ff*ected German Shorthaired Pointers*

Illumina TruSeq PCR-free DNA libraries with ~450 bp insert size of two affected German Shorthaired Pointers without known relationships were prepared. We collected 277 and 160 million 2 × 150 bp paired-end reads on a NovaSeq 6000 instrument corresponding to 29.3× and 17.9× coverage, respectively. Mapping and alignment were performed as described previously [24]. The sequence data were deposited under study accession PRJEB16012 and sample accessions SAMEA5657398 and SAMEA6249504 at the European Nucleotide Archive.

#### *2.6. Variant Calling*

Variant calling was performed using GATK HaplotypeCaller [25] in gVCF mode as described [24]. To predict the functional effects of the called variants, SnpEff [26] software together with NCBI annotation release 105 for the CanFam 3.1 genome reference assembly was used. For variant filtering we used 654 control genomes, which were either publicly available [27,28] or produced during other projects of our group [24] (Table S4).

#### *2.7. Gene Analysis*

We used the CanFam 3.1 dog reference genome assembly and NCBI annotation release 105. Numbering within the canine *UNC93B1* gene corresponds to the NCBI RefSeq accession numbers XM\_540813.6 (mRNA) and XP\_540813.3 (protein).

#### *2.8. Sanger Sequencing*

The *UNC93B1*:c.1438C>A variant was genotyped by direct Sanger sequencing of PCR amplicons. On high-quality genomic DNA samples, a 399 bp PCR product was amplified from genomic DNA using AmpliTaqGold360Mastermix (Thermo Fisher Scientific, Waltham, MA, USA) together with primers 5'-ATC CGT GTC TGT GCC CTC A-3' (Primer F) and 5'-CGA CCT GAG ACG CGG TAA A-3' (Primer R). For FFPE-derived DNA samples, a smaller amplicon of 124 bp was amplified with the primers 5'-CCT CGT ACC TGT GGA TGG AG-3' (Primer F2) and 5'-CTC TCG TCG GAG TTG TCC TC-3' (Primer R2). After treatment with exonuclease I and alkaline phosphatase, amplicons were sequenced on an ABI 3730 DNA Analyzer (Thermo Fisher Scientific). Sanger sequences were analyzed using the Sequencher 5.1 software (GeneCodes, Ann Arbor, MI, USA).

#### **3. Results**

#### *3.1. Mapping of the ECLE Locus*

We performed a GWAS with genotypes from 43 German Shorthaired Pointers. After quality control, the pruned dataset consisted of 14 ECLE cases, 29 controls and 116,891 markers. We obtained a single association signal with 35 markers exceeding a suggestive significance threshold of *<sup>p</sup>* <sup>=</sup> <sup>5</sup> <sup>×</sup> <sup>10</sup>−<sup>5</sup> after adjustment for genomic inflation. All associated markers were located on chromosome 18 within an interval spanning from 49.0 Mb–53.9 Mb. The top-associated marker at Chr18:49,835,345 had a *<sup>p</sup>*-value of 1.5 <sup>×</sup> <sup>10</sup>−<sup>6</sup> (Figure 2).

**Figure 2.** Mapping of the ECLE locus by genome-wide association. (**A**) Manhattan plot illustrating a single signal on chromosome 18. The dashed red line indicates the threshold for suggestive significance at *p* = 5 <sup>×</sup> 10−<sup>5</sup> according to [29]. The best associated marker did not reach the stringent Bonferroni significance threshold (*p*Bonf. = 4.3 <sup>×</sup> 10<sup>−</sup>7) due to several close relationships and extreme genomic inflation in the dataset. The genomic inflation factor was 1.90 before and 0.99 after the correction. (**B**) The quantile–quantile (QQ) plot shows the observed versus expected –log(*p*) values. The straight red line in the QQ plot indicates the distribution of p-values under the null hypothesis. The deviation of *p*-values at the right side indicates that these markers are stronger associated with the trait than would be expected by chance. This supports the biological significance of the association.

To narrow down the identified region, we visually inspected the genotypes of the cases and performed an autozygosity mapping. We searched for homozygous regions with allele sharing and found one region of ~493 kb which was shared between all 14 cases. The critical interval for the causative ECLE variant corresponded to the interval between the first flanking heterozygous markers on either side of the homozygous segment or Chr18:49,545,431-50,038,225 (CanFam 3.1 assembly).

#### *3.2. Identification of a Candidate Causative Variant*

We sequenced the genome of two affected dogs at 29.3 × and 17.9 × coverage and called SNVs and small indel variants with respect to the reference genome. We then compared these variants to whole genome sequence data of 8 wolves and 646 control dogs from genetically diverse breeds. This analysis identified two private homozygous variants in the critical interval in the affected dogs (Table 1). A visual inspection of the short read alignments ruled out any additional structural variants affecting protein-coding sequences in the critical interval in the two sequenced cases.


**Table 1.** Variants detected by whole genome re-sequencing of two ECLE-affected dogs.

One of the two private variants in the critical interval, Chr18:49,733,311C>T, was located in an intron of the *CHKA* gene and thus not investigated further. The other variant, Chr18:49,834,825C>A was a missense variant in the last exon of the *UNC93B1* gene. The formal designation of this variant is XM\_540813.6:c.1438C>A or XP\_540813.3:p.(Pro480Thr). It is predicted to change a highly conserved amino acid in the C-terminal tail of the UNC93B1 protein. We confirmed the variant by Sanger sequencing (Figure 3).

**Figure 3.** Details of the *UNC93B1* missense variant. (**A**) Representative Sanger electropherograms from dogs with the three different genotypes at c.1438C>A are shown. The amino acid translation is indicated. (**B**) Evolutionary conservation of the SDCBP binding domain [30]. The proline at position 480 of the canine UNC93B1 protein is strictly conserved across all vertebrates. The sequences were derived from the following database accessions: dog XP\_540813.3, human NP\_112192.2, cow XP\_540813.3, horse XP\_023510352.1, mouse NP\_062322.2, rat NP\_001101983.1, chicken XP\_004941322.1, frog NP\_001093723.1, zebrafish XP\_0026660582.1. (**C**) Scanning-alanine mutagenesis in mouse macrophages identified four mutants that disrupt SDCBP binding and lead to upregulated TLR7 signaling [30]. The altered residues in these mutants are underlined.

#### *3.3. Genotype Phenotype Association of the UNC93B1:p.Pro480Thr Variant*

We genotyped 544 German Shorthaired Pointers for the p.Pro480Thr variant and found a near perfect association of the genotypes at this variant with ECLE (*pFisher* = 1.2 <sup>×</sup> 10−39). None of the 520 genotyped controls were homozygous for the mutant A/A genotype. However, one of the 24 genotyped cases was not homozygous A/A. We speculate that this single discordant dog is likely due to a phenotyping error as it had an atypically late age of onset and was not clinically confirmed as having ECLE by a board certified veterinary dermatologist (Table S1). The analysis of additional animals from related hunting dog breeds revealed the presence of the mutant allele in German Longhaired Pointers and Vizslas. A single ECLE-affected Vizsla also had the homozygous mutant A/A genotype (Table 2).


**Table 2.** Association of the genotypes at *UNC93B1*:c.1438C>A with ECLE.

#### **4. Discussion**

In this study, we identified UNC93B1:pPro480Thr as a candidate causative variant for ECLE in dogs. We performed a new GWAS and obtained the strongest association signal on the same chromosome, but approximately 1 Mb more proximal than the location in the previously reported GWAS [20]. Given that linkage disequilibrium within breeds can span several Mb, we consider our new result a confirmation and refinement of the previously reported association. Compared to the previous study [20], we detected a different ~500 kb homozygous haplotype block harboring the 7 top markers of our GWAS that was shared among all 14 investigated ECLE cases.

Whole genome sequencing data of two cases and 654 controls revealed a single private protein changing variant in the critical interval, UNC93B1:pPro480Thr. All but one of the designated ECLE cases were homozygous for the mutant allele with the single discordant dog believed to represent a phenotype mismatch. Conversely, the mutant allele was not found in a homozygous state in more than 1000 control dogs.

The mutant allele was also detected in heterozygous status in controls of two related breeds, German Longhaired Pointers and Vizslas. These breeds share a common ancestry with German Shorthaired Pointers. This provides indirect support for the previous observation that ECLE also can affect dogs from breeds related to the German Shorthaired Pointer. The hypothesis of a common genetic defect in these breeds was confirmed by our finding of an ECLE affected Vizsla that was also homozygous mutant at the *UNC93B1* variant.

The *UNC93B1* gene encodes a protein named "unc-93 homolog B1, TLR signaling regulator". The human UNC93B1 consists of 597 amino acids and is a 12 transmembrane domain containing protein located in endosomal membranes [31]. It acts as a trafficking chaperone of the intracellular nucleic acid-sensing Toll-like receptors (TLRs) 3, 7 and 9 [32–35]. These TLRs are essential components of the innate immune system and activated when pathogen derived nucleic acids appear in endolysosomes. *UNC93B1* mediates the correct trafficking and localization of these TLRs to endolysosomes [32]. Complete loss-of-function of UNC93B1 results in a severe immune deficiency in human patients [36] and the *3d* mouse mutant [37].

Recently, the molecular mechanisms of the interaction of UNC93B1 with TLRs were studied in great detail. A 33 amino acid sequence motif in the cytoplasmic C-terminal domain of UNC93B1 binds to syndecan binding protein (SDCBP), also called syntenin-1. SDCBP interacts with both UNC93B1 and TLR7 [30]. This interaction dampens TLR7 signaling and prevents autoimmune activation of TLR7 by endogenous nucleic acids [30,35]. Gene-edited mice expressing a mutant Unc93B1 in which three critical amino acids of this C-terminal domain were altered (530-PKP/AAA-532) developed hallmarks of systemic inflammation and autoimmunity [30], similar to what has been observed in Tlr7 overexpressing mice [38–40]. In summary, the available literature suggests that complete loss of function of UNC93B1 leads to an immune deficiency, while UNC93B1 variants that only affect the C-terminal tail containing the SDCBP binding domain lead to upregulation of TLR7 signaling with subsequent development of systemic autoimmune disease.

The detailed functional knowledge on the role of the C-terminal tail of UNC93B1 for the regulation of TLR7 signaling strongly suggests that ECLE in dogs is due to dysregulated TLR7 signaling caused by the canine UNC93B1:p.Pro480Thr variant.

To the best of our knowledge, ECLE affected dogs represent the first spontaneous *UNC93B1* mutant that develops an autoimmune disease of the lupus group. Therefore, these dogs represent an interesting model for human CLE and/or SLE. As already suggested by [30], it seems possible that lupus erythematosus or other related autoimmune diseases in some human patients might be due to comparable genetic variants in *UNC93B1*.

#### **5. Conclusions**

We identified the spontaneously arisen UNC93B1:p.Pro480Thr variant as likely causative for ECLE in dogs. Knowledge of this variant will facilitate genetic testing of dogs to prevent the non-intentional breeding of ECLE-affected dogs. This unique canine form of CLE in dogs represents an interesting model for lupus erythematosus and potentially other autoimmune diseases in humans.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4425/11/2/159/s1, Figure S1: Multiple dimension scaling plot (MDS plot). File S1: SNV genotypes of 43 German Shorthaired Pointers. Table S1, details on 632 dogs in the study. Table S2: Pairwise IBD among the 43 dogs that were genotyped on SNV arrays. Table S3, results of autozygosity mapping. Table S4, accession numbers of 648 dog and 8 wolf genome sequences.

**Author Contributions:** Conceptualization, T.L.; Data curation, V.J.; Investigation, T.L., F.L., S.K., A.L., K.B., K.M.M., M.K.H.; M.L.C.; Methodology, V.J., A.L.; Resources, K.L.G., A.C., M.L., F.B., T.O., S.D.W., D.B., J.R.M., H.L., E.A.M., M.L.C.; Supervision, T.L.; Visualization, T.L.; Writing—original draft, T.L.; Writing—review and editing, T.L. F.L., V.J., S.K., A.L., P.R., M.M.W., K.L.G., A.C., M.L., F.B., T.O., S.D.W., K.B., D.B., K.M.M., J.R.M., M.K.H., H.L., E.A.M., and M.L.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by a grant from the Swiss National Science Foundation. H.L. and M.K.H. are supported by the Academy of Finland.

**Acknowledgments:** The authors are grateful to all dog owners who donated samples and shared health and pedigree data of their dogs. We thank Eva Andrist, Nathalie Besuchet Schmutz, Sini Karjalainen, Sabrina Schenk, and Daniela Steiner for expert technical assistance, the Next Generation Sequencing Platform of the University of Bern for performing the high-throughput sequencing experiments, and the Interfaculty Bioinformatics Unit of the University of Bern for providing high performance computing infrastructure. We thank the Dog Biomedical Variant Database Consortium (Gus Aguirre, Catherine André, Danika Bannasch, Doreen Becker, Brian Davis, Cord Drögemüller, Kari Ekenstedt, Kiterie Faller, Oliver Forman, Steve Friedenberg, Eva Furrow, Urs Giger, Christophe Hitte, Marjo Hytönen, Vidhya Jagannathan, Tosso Leeb, Hannes Lohi, Cathryn Mellersh, Jim Mickelson, Leonardo Murgiano, Anita Oberbauer, Sheila Schmutz, Jeffrey Schoenebeck, Kim Summers, Frank van Steenbeek, Claire Wade) for sharing whole genome sequencing data from control dogs. We also acknowledge all researchers who deposited dog or wolf whole genome sequencing data into public databases.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
