Mining the 99 Lives Cat Genome Sequencing Consortium database implicates genes and variants for the Ticked locus in domestic cats (Felis catus)

Abstract

Tabby patterns of fur coats are defining characteristics in wild and domestic felids. Historically, three autosomal alleles at one locus (Tabby): Abyssinian (Ta; a.k.a. ticked), mackerel (Tm; a.k.a. striped) and blotched (tb; a.k.a. classic, blotched) were thought to control these patterns in domestic cats and their breeds. Currently, at least three loci influence cat tabby markings, two of which are designated Tabby and Ticked. The Tabby locus is laeverin (LVRN) and affects the mackerel and blotched patterns. The unidentified gene for the Ticked locus on cat chromosome B1 was suggested to control the presence or absence of the ticked pattern (Tabby – Abyssinian (Ta; a.k.a. ticked). The cat reference genome (Cinnamon, the Abyssinian) has the ticked phenotype and the variant dataset and coat phenotypes from the 99 Lives Cat Genome Consortium (195 cats) were used to identify candidate genes and variants associated with the Ticked locus. Two strategies were used to find the Ticked allele(s), one considered Cinnamon with the reference allele or heterozygous (Strategy A) and the other considered Cinnamon as having the variant allele or heterozygous (Strategy B). For Strategy A, two variants in Dickkopf Wnt Signaling Pathway Inhibitor 4 (DKK4), a p.Cys63Tyr (B1:41621481, c.188G>A) and a less common p.Ala18Val (B1:42620835, c.53C>T) variant are suggested as two alleles influencing the Ticked phenotype. Bioinformatic and molecular modeling analysis suggests that these changes disrupt a key disulfide bond in the Dkk4 cysteine‐rich domain 1 or Dkk4 signal peptide cleavage respectively. All coding variants were excluded as Ticked alleles using Strategy B.

Full text

FULL PAPER
Mining the 99 Lives Cat Genome Sequencing Consortium database
implicates genes and variants for the Ticked locus in domestic cats
(Felis catus)
L. A. Lyons* , R. M. Buckley*, R. J. Harvey
†
, the 99 Lives Cat Genome Consortium
†
,
Danielle Aberdein
†
, Asa Ohlsson Andersson
†
, Tomas F. Bergstr€
om
†
, Adam R. Boyko
†
,
Margret L. Casal
†
, Brian W. Davis
†
, Ottmar Distl
†
, N. Matthew Ellinwood
†
, Oliver P. Forman
†
,
Edward I. Ginns
†
, Daisuke Hasegawa
†
, Vidhya Jagannathan
†
, Isabel Hernandez
†
, Maria Kaukonen
†
,
Emilie Leclerc
†
, Tosso Leeb
†
, Hannes Lohi
†
, Mark A. Magnuson
†
, Shrinivasrao P. Mane
†
,
John S. Munday
†
, Alexandra N. Myers
†
, Simon M. Peterson-Jones
†
, Clare Rusbridge
†
,
Beth Shapiro
†
, William F. Swanson
†
, Rory J. Todhunter
†
, Elizabeth A. Wilcox
†
and Yoshihiko Yu
†
*Department of Veterinary Medicine and Surgery, College of Veterinary Medicine, University of Missouri –Columbia, Columbia, MO
65211, USA.
†
School of Health and Behavioural Sciences, University of the Sunshine Coast, Sippy Downs, Qld 4558, Australia.
Summary Tabby patterns of fur coats are defining characteristics in wild and domestic felids.
Historically, three autosomal alleles at one locus (Tabby): Abyssinian (T
a
; a.k.a. ticked),
mackerel (T
m
; a.k.a. striped) and blotched (t
b
; a.k.a. classic, blotched) were thought to control
these patterns in domestic cats and their breeds. Currently, at least three loci influence cat
tabby markings, two of which are designated Tabby and Ticked. The Tabby locus is laeverin
(LVRN) and affects the mackerel and blotched patterns. The unidentified gene for the Ticked
locus on cat chromosome B1 was suggested to control the presence or absence of the ticked
pattern (Tabby –Abyssinian (T
a
; a.k.a. ticked). The cat reference genome (Cinnamon, the
Abyssinian) has the ticked phenotype and the variant dataset and coat phenotypes from the
99 Lives Cat Genome Consortium (195 cats) were used to identify candidate genes and
variants associated with the Ticked locus. Two strategies were used to find the Ticked allele
(s), one considered Cinnamon with the reference allele or heterozygous (Strategy A) and the
other considered Cinnamon as having the variant allele or heterozygous (Strategy B). For
Strategy A, two variants in Dickkopf Wnt Signaling Pathway Inhibitor 4 (DKK4), a p.Cys63Tyr
(B1:41621481, c.188G>A) and a less common p.Ala18Val (B1:42620835, c.53C>T)
variant are suggested as two alleles influencing the Ticked phenotype. Bioinformatic and
molecular modeling analysis suggests that these changes disrupt a key disulfide bond in the
Dkk4 cysteine-rich domain 1 or Dkk4 signal peptide cleavage respectively. All coding
variants were excluded as Ticked alleles using Strategy B.
Keywords Abyssinian, coat pattern, Dickkopf Wnt Signaling Pathway Inhibitor 4,DKK4,
Tabby
Introduction
The stripes, swirls and spots of cat coat patterns are well
recognized and often defining characteristics in felids. Both
wild felids and domesticated cat breeds display a wide range
of coat patterning phenotypes. Lions (Panthera leo), pumas
(Puma concolor), Abyssinians and Singapuras display basi-
cally no pattern, whereas tigers (Panthera tigris) and many
mackerel tabby cat breeds present with stripes. Leopards
(Panthera pardus), Egyptian maus and ocicats have
Address for correspondence
L. A. Lyons, Department of Veterinary Medicine and Surgery, College
of Veterinary Medicine, University of Missouri –Columbia, E109
Veterinary Medical Building, 1600 E. Rollins Street, Columbia, MO
65211, USA.
E-mail: lyonsla@missouri.edu
a
99 Lives Cat Genome Consortium authors and affiliations are provided
in Appendix 1.
Accepted for publication 02 March 2021
doi: 10.1111/age.13059
321
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.

distinctive spots, and clouded leopards (Neofelis nebulosa),
marbled Bengals and American shorthairs (blotched, classic
tabbies) can present with intriguing blotches and swirls
(Fig. 1). In the wild, undoubtedly, these patterns confer a
camouflage tailored to the felid’s ecological niche and have
been a focus of artificial selection in the domesticated cat
breeds. Considering species taxonomy, the blotched tabby
cat was considered the ‘voucher’ specimen for the domestic
cat (Abdel-Malek et al. 2001), hence, this pattern is also
known as a ‘classic’ tabby (Fig. 1c; Linneaus 1758).
For over a century of domestic cat breeding, the Tabby
locus, which was originally thought to control all of the
domestic felid pattern and markings, was defined by three
autosomal alleles: Abyssinian (T
a
; a.k.a. ticked), mackerel (T
m
;
a.k.a. striped) and blotched (t
b
; a.k.a. classic,blotched; Lomax &
Robinson 1988). Traditional patterns were represented by
three phenotypes governed by the following genotypes: T
a
,
which produces few markings and possible stripes on the
head, legs and tail but not on the torso; T
m
T
m
or T
m
t
b
, which
produces stripes on the head, legs, tail and torso; and t
b
t
b
,
which produces stripes on the head, legs and tail but circular
patterns on the torso. Cats homozygous for the Abyssinian
allele (T
a
T
a
) may have less barring on the legs and can often
be distinguished from heterozygotes. Hence, the T
a
allele can
be considered co-dominant to T
m
and t
b
when considering
the leg barring, but this phenotype has variable expression
and is difficult to clearly distinguish. The allelic series
T
a
>T
m
>t
b
was suggested; however, this series came under
dispute because the distinctive spotted patterns of Egyptian
maus, ocicats and other cats could not be explained by this
single, locus multi-allelic model (Cat Fanciers’ Association
2004; Kaelin & Barsh 2010; Eizirik et al. 2010).
To localize genes associated with Tabby, a genome scan
was performed on a large pedigree of cats segregating for
tabby coat markings, specifically for the Abyssinian (T
a
) and
the blotched (t
b
t
b
) phenotypes (Lyons et al. 2006). A genetic
linkage between the phenotypes and eight short tandem
repeat (STR) markers on cat chromosome B1 was indicated,
the most significant linkage was between cat short tandem
repeat marker FCA700 and ‘Tabby’ phenotype (Z=7.56,
h=0.03). The linked markers covered a 17 cM region and
flanked an evolutionary breakpoint, suggesting that the
Tabby gene has a homolog on either human chromosome 4
or 8, which share synteny with cat chromosome B1.
A second linkage analysis confirmed the B1 location of a
patterning locus in cats and identified a second locus
controlling patterning on cat chromosome A1 (Eizirik et al.
2010; Kaelin et al. 2012). Using cross-species analyses, five
SNPs showed significant association (Prange =9910
4
to
3.2 910
9
) within a 180 kb interval on cat chromosome
A1. The associated gene and variants were refined and the
candidate gene was identified as laeverin (LVRN;a.k.a.
transmembrane aminopeptidase Q,Taqpep), which encodes a
membrane-bound metalloprotease and plays a regulatory
role in extravillous trophoblast migration (Maruyama et al.
2007; Horie et al. 2012). Three variants, p.S59X
(c.176C>A), p.D228N (c.682G>A) and p.W841X
(c.2522A>G), were suggested to cause the blotched tabby
patterns across breeds (Kaelin et al. 2012). An additional
variant was suggested to be the cause of the rare king
cheetah phenotype, in which spots coalesce into blotches
and stripes (Buckley et al. 2020b).
The tabby markings of cats are also influenced by the
pigment switching signaling pathway including agouti-
signaling protein (ASIP) and melanocortin 1 receptor (MC1R;
Barsh et al. 2000; Abdel-Malek et al. 2001; Baxter, Loftus &
Pavan 2009; Baxter & Pavan 2013). Cats with tabby
patterning have solid-colored hairs as part of the fur
composing the stripes, swirls and spots, whereas the fur in
between the pattern is ‘ticked’ with bands of pheomelanin
and eumelanin (i.e. agouti –named after the South
American rodent which also has bands of both pigment
types), giving an illusion of brown coloration. Ticked
tabbies, such as an Abyssinian, do not have any pattern,
but have ticked, agouti hairs throughout the coat. These
hairs can have limited ticking, with one band of eumelanin
at the tip and a base band of pheomelanin, or the fur can
have several alternating bands of eumelanin and pheome-
lanin. The number and extent of alternating bands of
pigment is postulated as a separate locus suggested known
as Wide-band (Robinson 1991).
After the publication of the variants in LVRN on cat
chromosome A1 for the now defined Tabby locus (OMIA
001429-9685), the associated locus identified on cat
chromosome B1 was termed the ‘Ticked’ locus (Ti) (Eizirik
et al. 2010; Kaelin et al. 2012). The Ticked locus is more akin
to an absence of pattern common to the Abyssinian and
Singapura cat breeds, and probably interacts with the ASIP-
MC1R-LVRN pathway in a manner influencing the presence
or absence of coat pattern, regardless of the type of pattern
dictated by the Tabby locus. Thus, the Ticked locus is epistatic
to the Tabby locus. The Ticked allele (Ti/), which has no
pattern, is considered autosomal dominant; however, as
patterning is highly prevalent in domestic cats and present
in the progenitor wildcat species, the wt allele for the Ticked
locus is the presence of pattern (i.e. non-ticked (ti
+
ti
+
)).
Analyses of the single nucleotide variant and phenotypic
data of the 99 Lives Cat Genome Sequencing dataset was used
to refine the associated region on cat chromosome B1 and
identify putative functional variants that cause the lack of
patterning in cats and probably the gene for the Ticked locus.
Materials and methods
The STRs from the published linkage analyses (Lyons et al.
2006) were remapped to the Felis catus V 9.0 genome
assembly (Buckley et al. 2020a) using the basic local
alignment search tool (BLAST; Altschul et al. 1990). An
extended area around the 17 cM linked region on cat
chromosome B1 was analyzed for candidate variants.
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Lyons et al.322

Variant filtering
The details of the 99 Lives dataset, including sequence data
processing and the variant calling workflow have been
previously published (Buckley et al. 2020a; Buckley et al.
2020b; Yu et al. 2020). The 99 Lives VCF was filtered for
candidate variants using VARSEQ (Golden Helix). The VCF
was annotated using both Ensembl 101 release (20 August
2020) and NCBI Felis catus annotation release 104 (10
December 2019). The ticked (Ti) allele is considered
dominant (absence of pattern); therefore, a ticked cat can
(a)
(d)
(g) (h)
(e) (f)
(b) (c)
Figure 1 Tabby patterns of cats and the Ticked cats in the 99 Lives Cat Genome database. (a) The wt domestic cat is a brown mackerel (striped)
tabby with relevant color genotypes A,B,C,D,E,gg,ii,ss,Ta
M
,ti
+
ti
+
and X
o
X
o
. (b) The Somali (longhaired Abyssinian) is the hallmark
breed with a Ticked phenotype (Ti
A
). (c) An American shorthair with the blotched (a.k.a. classic) tabby (ti
+
ti
+
,ta
b
ta
b
), which represents the voucher
specimen for a domestic cat although not the common wt specimen, which is a mackerel tabby. (d) The Egyptian mau breed has been selected for
spots. The ticked cats in the data analyses include (e) Art Deco –red ticked Oriental shorthair (image courtesy of Winter Trussell, DVM); (f) Nina –
blue ticked random bred; (g) Cali –obligate heterozygote bred from a Somali; and (h) Monkey –obligate heterozygote bred from a Somali. Images of
Somali, the American shorthair and the Egyptian mau courtesy of Chanan Photography, Richard Katris.
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Ticked cats with DKK4 variants 323

be heterozygous or homozygous for the ticked allele and
hence the variants controlling the allele. In addition, as the
Ticked locus has not been examined genetically in detail;
phenocopies may be present or more than one allele may
produce the same phenotype. For example, four variants are
known to cause long hair in cats (Dr€
ogem€
uller et al. 2007;
Kehler et al. 2007). A independent WGS of the reference cat,
Cinnamon, is available in the 99 Lives dataset. Because the
cat reference genome represents an Abyssinian cat (Lin-
neaus 1758; Lomax & Robinson 1988; Lyons et al. 2006;
Kaelin & Barsh 2010), Cinnamon, who has a ticked
phenotype, Cinnamon can be either homozygous or
heterozygous for the variant controlling the ticked allele
(Buckley et al. 2020a; Li et al. 2016; Montague et al. 2014;
Pontius et al. 2007). If Cinnamon is homozygous, then the
ticked allele would be the reference allele, which is not the
wt allele in cats as ticked is rare. However, if the reference
cat (Cinnamon) is heterozygous, either the ticked or non-
ticked allele could have been randomly selected as the
reference allele. Therefore, two filtering strategies (A and B)
were considered to identify candidate variants: (B) Abyssini-
ans (including Cinnamon’s WGS data) are heterozygous or
homozygous for the variant allele; and (B) Abyssinians
(including Cinnamon’s WGS data) are heterozygous or
homozygous for the reference allele. The cats used for the
filtering steps within the two strategies are outlined in
Table 1.
Only cats in the 99 Lives dataset with known phenotypes
(ticked or non-ticked), documented by the investigators by
visual inspection and or images, were used to filter the
variants for candidates (Table 1), but all cats were consid-
ered in the final variant considerations and discussions.
Breeds defined by their patterning, such as spotted cats
(Egyptian mau, ocicat), striped cats (toyger) or other tabby
markings (Bengal and savannah) were considered non-
ticked. Except for one solid Bengal, cats of solid coloration
(File S1) determined by the 2 bp deletion in the Agouti locus
(ASIP A3:25086566; XM_019826162.2 c.264_265delCA,
p.Met89Glufs*59) causing melanism or listed as white were
not included in the variant filtering as the ticked phenotype
would be masked (Eizirik et al. 2003).
The variant filtering steps used the same cats for both
strategies A and B. The two strategies differed by
changing the expected variant zygosity of the cats for
either the reference or the variant allele. Filtering steps
included: (i) eliminate intergenic, intronic and synony-
mous variants; (ii) consider the variant zygosity for the
WGS entry for Cinnamon; (iii) consider the variant
zygosity for the tabby breeds (n=9); (iv) consider the
variant zygosity for the known tabby cats (n=40); (v)
consider the variant zygosity for the known Abyssinians,
which are ticked (n=3); (vi) consider the variant zygosity
for the two known heterozygous ticked cats; and (vii)
consider the variant zygosity for the two additional cats
with the ticked phenotype (not listed as a specific breed).
Two phenotypically ticked cats were known to be
heterozygous (obligate heterozygous) for the ticked allele
as they were produced by a breeding of a tabby cat
(Bengal) mated with a pedigreed ticked Somali (Fig. 1g,h;
Cogne et al. 2020). The cats used in each filtering step are
presented in Table 1.
A previous analysis of structural variants was conducted
on the 99 Lives WGS dataset containing 54 cats, which
included two of the Abyssinians for this study (Buckley et al.
2020b https://doi.org/10.1371/journal.pgen.1008926.
s025). This dataset was examined for consistent structural
variants found in the two Abyssinians but absent in the
other non-ticked cats, which are present in the current
dataset.
Bioinformatic analysis of the impact of Dkk4 variants
Signal peptide cleavage was predicted using the SIGNALP 5.0
server (Almagro Armenteros et al. 2019) for both wt Dkk4
protein and the p.Ala18Val variant. Disulfide bonds in the
Dkk4 CRD1 domain were visualized using the structure of
residues 19–224 of human Dkk4 (Patel et al. 2018) in the
ChimeraX molecular graphics system (Pettersen et al.
2021). The p.Cys63Tyr variant was modeled with the
swapaa command using the Dunbrack backbone-dependent
rotamer library and taking into account the lowest clash
score, highest number of H-bonds and highest rotamer
probability.
Table 1 Coding variant analysis of the Ticked linked region at B1: 35–
65 Mb.
Number of
variants
Strategy A
Filter 1 –without intergenic, intronic, synonymous 10 675
Filter 2 –Cinnamon as heterozygous or homozygous
variant
372
Filter 3 –tabby breeds (n=9) (Ocicat, E. mau, Toyger,
Bengal, Savannah) –as homozygous reference
38
Filter 4 –known tabbies (40) –as homozygous reference 8
Filter 5 –Abyssinian as heterozygous or homozygous
variant (PennyLane, Dot, CVB15215)
1
Filter 6 –Cali and Monkey set as heterozygous 1
Filter 7 –ticked cats (Nina, Art Deco) as heterozygous or
homozygous variant
0
Strategy B
Filter 1 –without intergenic, intronic, synonymous 10 675
Filter 2 –Cinnamon as heterozygous or homozygous
reference
10 590
Filter 3 –tabby breeds (n=9) (Ocicat, E. mau, Toyger,
Bengal, Savannah) –as homozygous variant
44
Filter 4 –known tabbies (40) –as homozygous variant 3
Filter 5 –Abyssinian as heterozygous or homozygous
reference (PennyLane, Dot, CVB25215)
0
Filter 6 –Cali and Monkey as heterozygous 0
Filter 7 –ticked cats (Nina, Art Deco) –as heterozygous
or homozygous reference
0
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Lyons et al.324

Results
The 99 Lives dataset includes 195 domestic cats. Cat
signalment, short read archive (SRA) submission identifiers
and investigator contacts are presented in File S1. The 99
Lives dataset included an independent ILLUMINA-based
sequence for the reference cat, listed as Cinnamon, thus,
both alleles for the reference cat can be determined in the
99 Lives data. Considering the ASIP 2 bp deletion (Eizirik
et al. 2003), 79 cats were identified as solid and one as
white, thus, their ticked –tabby phenotype could not be
determined (File S1). Three Egyptian maus, one ocicat, one
savannah, one toyger and three Bengals represented breeds
known for their tabby markings and 40 additional cats were
known tabbies. Six additional cats were hairless or pointed;
therefore the ticked phenotype could not be accurately
determined. Fifty-three cats did not have a definitive
phenotype as no images were available.
The dataset included two obligate heterozygous ticked
cats (Cali and Monkey; Fig. 1g, h) identical by descent for
their ticked allele as they are parent–offspring bred from a
ticked Somali. Three known ticked Abyssinians (Cinnamon,
Dot and Penny Lane) and three cats with the ticked
phenotype (CVB15215, Nina (Fig. 1f) and Art Deco
(Fig. 1e)) were also in the dataset; therefore, at least eight
cats were known to have the ticked phenotype.
Eight STRs from the published linkage analysis were
remapped to cat chromosome B1q within the region of
positions 49–58 Mb (Table S1; Lyons et al. 2006). The
linkage region is near the centromere and has known
recombinants, thus the region was extended to include
genes on the centromeric portion of B1p, including single
variant analyses for all sequences (genic and intergenic)
from 35 to 65 Mb on B1. This region contained 897 372
variants, with approximately 53% as intergenic variants,
45% as intronic variants and 0.5% as synonymous variants
(Table 2, File S2a). After applying filter 1, the intergenic,
intronic and synonymous variants were excluded and
10 675 variants remained to be filtered (File S2b); however,
all variants were considered in the analyses. The reductions
of the non-coding and coding variants at each step of the
analyses are summarized in Table 2.
Five additional filtering steps were used to eliminate
variants not associated with the ticked phenotype (Table 1,
Files S2b and S3a,b). Two strategies were considered to
identify candidate variants: (A) Cinnamon and other
Abyssinians (the reference genome cat) as heterozygous or
homozygous for the variant allele; and (B) Cinnamon and
other Abyssinians as heterozygous or homozygous for the
reference allele. Therefore, the zygosities for the remaining
cats with known phenotypes were altered accordingly,
depending on which of the two strategies was being
considered. For strategy A (Table 1, File S2b), Cinnamon
must be heterozygous or homozygous for the variant allele
(Filter 2), then only 372 variants remained as candidates.
After filter 3 (Tabby breeds (n=9) must be homozygous
reference), only 38 variants remained. Filter 4 set the
zygosity for the 40 known tabby cats to homozygous
reference and reduced the possible variants to 8. After
applying filters 5 and 6, which considered the three known
Abyssinians and the two obligate heterozygous cats (Cali
and Monkey), only one variant remained, namely a variant
in DKK4. This variant was at position B1:41621481 and
was a guanine to alanine change (XM_023252567.1;
ENSFCAT00000034752: c.188G>A), causing a
p.Cys63Tyr amino acid change. The allele count for this
variant was consistent with the ticked allele being rare in
the domestic cat population and identified in seven cats, two
as a homozygote and five as a heterozygote. The only two
homozygous cats were the two pedigree Abyssinians, and
five heterozygous cats were identified, comprising the two
known heterozygotes (Cali and Monkey; Fig. 1g,h), Nina
(Fig. 1f), CVB15215 and Cinnamon.
After applying filter 7, which meant two ticked cats had
to be heterozygous or homozygous for the variants, one
ticked Oriental cat (Art Deco; Fig. 1e) did not have the
p.Cys63Tyr variant, suggesting a misidentification of this
cat or possible heterogeneity. However, an obtained pho-
tograph confirmed the phenotype. As identified in filter step
4, this cat had a second, slightly more common missense
variant in DKK4, a p.Ala18Val that was also identified in
Cinnamon and consistent with Cinnamon being a com-
pound heterozygote for two DKK4 variants (see the discus-
sion below). This variant was identified in 15 cats, 11 as a
homozygote and four as a heterozygote.
Considering the second strategy B and filter 2 (File S3a,b),
when Cinnamon must be heterozygous or homozygous for
the reference allele, 10 590 variants remained for consid-
eration. Filter 3 (tabby breeds (n=9) as homozygous
variant) reduced the candidates to 44 variants. Filter 4 set
the zygosity for the 40 known tabby cats to a homozygous
variant, which reduced the possible number variants to five.
Considering the three Abyssinian cats as heterozygous or
homozygous for the reference allele (filter 5), no variants
remained, except for intergenic and intronic. No structural
variants were identified that were specific to the two
Abyssinians but absent from the other subset of 52 non-
ticked cats (File S4).
DKK4 had 51 variants within the 99 Lives dataset (File
S5). However, 30 variants had an allele count of three or
less and had poor genotyping qualities, and thus, were not
considered valid variants. Four DKK4 variants were syn-
onymous changes and were not further considered. Five
additional missense variants were present (p.Ile86Met,
p.Arg71Lys, p.Lys132Arg, p.Ille201Thr and p.Ser208Gly),
which were excluded as candidates along with the addi-
tional variants as they were present in several tabby cats.
The p.Cys63Tyr variant was concordant in all ticked cats,
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Ticked cats with DKK4 variants 325

except for the ticked Oriental shorthair cat (Art Deco;
Fig. 1e) and not identified in any other cats.
The p.Ala18Val variant had a low allele count in the data
and was identified as heterozygous in four cats and
homozygous in 11 cats. The two obligate heterozygous
cats (Cali and Monkey; Fig. 1g,h) both excluded this variant
as these cats were homozygous for the reference allele.
However, if a second allele accounts for ticking, Art Deco
(Fig. 1e) was heterozygous for the p.Ala18Val variant, as
was one Burmese from Australia and Gizmo, a Maine coon
cat from Cornell, which was listed with a coloration of
black, smoke and white but was not a solid cat as
determined by the ASIP data. This coloration was reported
in the patient information and no image was avail-
able. Eleven cats were homozygous for the p.Ala18Val
allele (B1:42620835; ENSFCAT00000034752:c.53C>T;
XM_023252567.1:c53C>T), all Burmese from Australia.
Further bioinformatic analysis of the DKK4 variants
suggest these missense changes are likely to influence the
function of Dkk4. Dkk4 is secreted Wnt antagonist, and the
precursor protein consists of an N-terminal signal peptide
(amino acids 1–19) that consists of two independent folded
cysteine-rich domains (CRD1 and CRD2) joined by a highly
flexible, non-structured linker (Patel et al. 2018; Fig. 2a).
Both Ala18 and Cys63 are highly conserved residues in
mammalian Dkk4 sequences (Fig. 2a,b). Using the SIGNALP
5.0 server (Almagro Armenteros et al. 2019), the p.Ala18-
Val variant is suggested to influence predicted signal peptide
cleavage (Fig. 2c). The wt Dkk4 is normally cleaved
between amino acids 18 and 19: LSA-LV with a probability
of 0.6007. However, for Dkk4 p.Ala18Val, cleavage was
predicted between amino acids 20 and 21, VLV-LD, with a
much lower probability of 0.3275. Using the structure of
the Dkk4 N-terminal cysteine-rich domain (CRD1;
Table 2 Variant summary of the Ticked linked region (B1: 35–65 Mb) for 195 cats from the 99 Lives dataset.
Strategy
None
Number of variants
AB
Variant type 1
1
2 3 4 567
2
2 3 4567
2
30UTR 4919 4919 190 21 4 0 0 0 4880 24 3 0 0 0
50UTR start gain 55 55 1 0 0 0 0 0 55 0 0 0 0 0
50UTR 1650 1650 79 6 1 0 0 0 1641 6 0 0 0 0
Disruptive inframe deletion 2 2 0 0 0 0 0 0 2 0 0 0 0 0
Disruptive inframe insertion 6 6 0 0 0 0 0 0 6 0 0 0 0 0
Downstream gene 13 13 4 0 0 0 0 0 13 0 0 0 0 0
Frameshift 152 152 9 2 0 0 0 0 139 0 0 0 0 0
Inframe deletion 20 20 0 0 0 0 0 0 20 0 0 0 0 0
Inframe insertion 32 32 3 0 0 0 0 0 32 0 0 0 0 0
Initiator codon 4 4 0 0 0 0 0 0 4 0 0 0 0 0
Intergenic 477 950 –26 001 375 0 9 9 9 467 917 3226 581 71 71 48
Intron 404 352 –19 314 468 0 32 31 31 396 052 2013 218 38 31 20
Missense 1813 1813 18 2 3 1 1 1 1810 5 1 0 0 0
Non-coding exon 1308 1308 30 0 0 0 0 0 1303 5 1 0 0 0
Splice acceptor 24 24 1 0 0 0 0 0 23 0 0 0 0 0
Splice donor 22 22 1 0 0 0 0 0 21 0 0 0 0 0
Splice region 618 618 35 0 0 0 0 0 605 4 0 0 0 0
Stop gained 31 31 1 0 0 0 0 0 30 0 0 0 0 0
Stop lost 4 4 0 0 0 0 0 0 4 0 0 0 0 0
Stop retained 2 2 0 0 0 0 0 0 2 0 0 0 0 0
Synonymous 4395 –42 0 0 0 0 0 4395 7 0 0 0 0
Total B1: 35–65 Mb 897 372 10 675 45 729 4138 750 41 40 40.0 878 954 5290 804 109 102 68.0
1
Filter 1 eliminated intergenic, intronic and synonymous variants, thus the values were the same for strategies A and B. However, for this table only,
filters 2–7 include all variants.
2
Additional of the ticked cat Art Deco eliminates all variants in strategies A and B for filter 7.
Figure 2 Bioinformatic analysis of DKK4 variants. (a) Alignment of cat, mouse and human Dkk4 proteins showing conservation of the N-terminal
signal peptide (blue shading) and cleavage site and cysteine-rich domains CRD1 and CRD2. Key disulfide bonds in CRD1 and CRD2 are indicated. (b)
Ala18 and Cys63 are highly conserved residues in key mammalian Dkk4 sequences. (c) Signal peptide cleavage analyzed using the SIGNALP 5.0 server
(Lomax & Robinson, 1988). The wt Dkk4 is typically cleaved between amino acids 18 and 19 with a probability of 0.6007. For the Dkk4 variant
p.Ala18Val, signal peptide cleavage was predicted between amino acids 20 and 21 with a lower probability of 0.3275. (d) Structural analysis reveals
that the Dkk4 p.Cys63Tyr variant is predicted to disrupt a key disulfide bond between amino acids Cys47 and Cys63, one of five disulfide bonds in
CRD1 that play a major role in stabilizing the Dkk4 tertiary structure. (e) Substitution of p.Cys63Tyr also results in clashes with residues Cys41, Asp46
and Cys47. [Correction added 7 April 2021, after first online publication: The figure 2 has been updated in this version.]
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Lyons et al.326

Cat Dkk4 MAVVVLLGLSWFCAPLSALVLDFNNIKSSADVHGARKGSQCLSDKDCSSR 50
Human Dkk4 MVAAVLLGLSWLCSPLGALVLDFNNIRSSADLHGARKGSQCLSDTDCNTR 50
Mouse Dkk4 MVLVTLLGLSWFCSPLAALVLDFNNIKSSADVQGAGKGSLCASDRDCSEG 50
Cat Dkk4 KFCLKPQDERPFCATCRGLRRRCQRNAMCCPGTLCINDVCTTMEDATPIL 100
Human Dkk4 KFCLQPRDEKPFCATCRGLRRRCQRDAMCCPGTLCVNDVCTTMEDATPIL 100
Mouse Dkk4 KFCLAFHDERSFCATCRRVRRRCQRSAVCCPGTVCVNDVCTAVEDTRPVM 100
Cat Dkk4 ERQMDDQDDIETKGTTEHPIQENKPKRKPNIKKPQGGKGQEGERCLRTLD 150
Human Dkk4 ERQLDEQDGTHAEGTTGHPVQENQPKRKPSIKKSQGRKGQEGESCLRTFD 150
Mouse Dkk4 DRNTDGQDGAYAEGTTKWPAEENRPQGKPSTKKSQSSKGQEGESCLRTSD 150
Cat Dkk4 CGAGLCCARHFWTKICKPVLLEGQVCSRRGHKDTAQAPEIFQRCDCGPGL 200
Human Dkk4 CGPGLCCARHFWTKICKPVLLEGQVCSRRGHKDTAQAPEIFQRCDCGPGL 200
Mouse Dkk4 CGPGLCCARHFWTKICKPVLREGQVCSRRGHKDTAQAPEIFQRCDCGPGL 200
Cat Dkk4 ICRNQVTSNQQHTRLRVCQKI* 221
Human Dkk4 LCRSQLTSNRQHARLRVCQKIEKL* 224
Mouse Dkk4 TCRSQVTSNRQHSRLRVCQRI* 221
C63Y
A18V
(c)
(d)
Cat DKK4 wild-type MAVVVLLGLSWFCAPLSALVLDFNNIKSSADVHGARKG
Cat DKK4 Ala18Val MVAAVLLGLSWLCSPLGVLVLDFNNIRSSADLHGARKG
Predicted cleavage sites
Signal peptide
CRD1
CRD2
C63
C47
Y63
C47
D46
C41
C53
(e)
(a)
Cat Dkk4 MAVVVLLGLSWFCAPLSALVLDF....KFCLKPQDERPFCATCRGLRRRC
Cow Dkk4 MVVVVLLGLGWLCAPLGALVLDS....KFCLQRYDEKPFCATCRGPRRRC
Dog Dkk4 MVVVVLLGLSWFCAPLGALVLDF....KFCLKPQDEKPFCATCRGLQRRC
Horse Dkk4 MEVVVLLGLSWFCAPLGALVLDF....KFCLKPRHEKAFCATCRRLRRRC
Pig Dkk4 MAVVVLLGLGWLCTPLGALVLDF....KFCLKPQDEKPFCATCRGLRRRC
C63
A18
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Ticked cats with DKK4 variants 327

Montague et al. 2014), the p.Cys63Tyr variant was
predicted to disrupt a key disulfide bond between amino
acids Cys47 and Cys63 (Fig. 2a,d), one of five disulfide
bonds in CRD1 that play a major role in stabilizing the Dkk4
tertiary structure (Patel et al. 2018). The introduction of a
tyrosine at position 63 also results in predicted clashes with
residues Cys41, Cys47 and Asp46 (Fig. 2e).
Discussion
At least two different loci have been associated with coat
patterning in domestic cats, Tabby and Ticked (Eizirik et al.
2010; Kaelin et al. 2012; Lyons et al. 2006). Only the gene
and variants for the Tabby locus on cat chromosome A1
have been identified and the Tabby gene, LVRN, is suggested
to have interactions with other genes, such as endothelin 3
(EDN3), for influencing pattern (Kaelin et al. 2012).
However, Ticked (previously known as Tabby), was the first
locus to be regionally positioned to the centromeric region
of cat chromosome B1q (Lyons et al. 2006). The ticked
allele, formerly known as tabby –Aby (T
a
), presents as a cat
devoid of patterning, implying no stripes, blotches, swirls or
spots –no tabby markings within the cat coat coloration
(Fig. 1b). This allele is dominant, suppressing patterning,
and hence epistatic to the LVRN Tabby locus, which controls
the type of pattern, including mackerel (stripes) or classic
(blotched) pattern (Fig. 1a,c). The control of spotted pat-
terns common to the Egyptian mau (Fig. 1d) and ocicat
breeds is probably influenced by yet unidentified loci (Eizirik
et al. 2010; Kaelin et al. 2012). Ticked is a rare phenotype
and is infrequent in random-bred cats and other breeds;
however, the phenotype is fixed in the Abyssinian and
Singapura cat breeds. However, breeds that have temper-
ature-sensitive variants, such as the tyrosinase (TYR)
variants associated with Siamese, Burmese and related
breeds, could be ticked tabbies (Lyons et al. 2005). ‘Ghost’ or
bleed-through tabby patterns can appear in the coats of cats
with the TYR variants as they age, which is an undesirable
presentation for these breeds with temperature-sensitive
coat colorations.
The development of the 99 Lives cat genome sequencing
database and improved assembly and annotation of the cat
have led to vastly improved variation identification for cats
and the ability to define candidate genes and variants based
on cat phenotypes (Buckley et al. 2020a; Buckley et al.
2020b; Cogne et al. 2020; Jaffey et al. 2019; Mauler et al.
2017; Pettersen et al. 2021; Yu et al. 2020). As the cat used
for the reference genome has the ticked phenotype of
interest, and because this variant is autosomal dominant
and a ticked cat, such as the reference cat, can be
heterozygous, the wt allele (non-ticked) may or may not
be the reference allele in the cat genome assembly.
Therefore, two different strategies were necessary to identify
candidate variants in which strategy A was successful. For
strategy A, Cinnamon was considered heterozygous or
homozygous for the variant allele; all variants in the region,
including intergenic and intronic, were eliminated as
candidates except for a p.Cys63Tyr variant in DKK4. This
variant was at position B1:41621481, near the centromere
and at an extreme end of the linked region in the linkage
analyses. The allele count suggested that no other cats in
the 99 Lives dataset, including 127 solid-coloration cats
and 79 cats with unknown phenotypes, have a ticked
phenotype. Forty intergenic and intronic variants were not
excluded. Considering variant discovery strategy B, Cinna-
mon was considered as heterozygous or homozygous for the
reference allele. All variants in the region, except approx-
imately 68 intergenic and intronic, were eliminated as
candidates. In addition, a previous study examined the
structural variants of a 54-cat subset of the current WGS
dataset and included two Abyssinian cats. No structural
variants were identified that were unique to the two
Abyssinian cats and absent from the 52 non-ticked cats;
therefore, further structural variant analyses were not
performed.
One cat, a ticked Oriental shorthair (Art Deco; Fig. 1e),
excluded the p.Cys63Tyr as causal for all ticked phenotypes.
This cat, as well as the reference cat, Cinnamon, was
heterozygous for a different allele, the p.Ala18Val variant,
which also had a low allele count and was identified as
heterozygous in four other cats and homozygous in 11 cats.
However, the two obligate heterozygous cats both excluded
this variant as explaining all cases of ticked coats as these
cats were homozygous for the reference allele at the same
genomic position. If a second allele accounts for ticking in
cats, Cinnamon was a compound heterozygous for the
p.Cys63Tyr/p.Ala18Val variants. A previous study on
hairlessness in the lykoi cat breed described a similar
experience with new phenotypes presenting as compound
heterozygotes (Buckley et al. 2020b). One Burmese from
Australia and a cat from Cornell (Gizmo), which is listed
with a coloration of black, smoke and white, were also
heterozygous for the p.Ala18Val allele and thus should be
ticked. A black smoke cat suggests a solid-colored cat with
at least one copy of the Inhibitor allele (I). The ASIP data
suggested that this cat was not solid-colored, and perhaps
the amount of white on it obscured the tabby markings.
Eleven cats were homozygous for the p.Ala18Val allele,
which were all Burmese from Australia, but as these cats
were solid, the ticked phenotype could not be determined.
The breeding colony that produced the reference cat,
Cinnamon, was established in Sweden (~1983) and several
cats were moved to the University of Missouri –Columbia in
2001 to re-establish the colony (Narfstrom 1983; Menotti-
Raymond et al. 2010). Outcrossed, natural matings with
European short-haired cats were performed to increase the
heterozygosity in the colony, and 55 backcrossed cats were
generated as part of the project to map the Ticked and Tabby
loci. However, overall, Cinnamon was selected as the
reference cat from over 1000 genotyped cats because she
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Lyons et al.328

was highly inbred, leaning towards having minimal allelic
variation (Pontius et al. 2007). Alternatively, Abyssinians
and other cats may have more than one Ticked allele; thus,
even if they are highly inbred, they could be segregating for
different alleles that produce the same phenotype. In
addition, many cat breeds with temperature-sensitive, as
well as solid but dilute, colorations, such as Siamese,
Burmese and blue cats (korats, chartreux and Russian
blue), and cats with Orange coloration, may also be Ticked
because ‘ghost patterning’ can bleed through in these coat
colorations when the cat is young or as it ages, which is not
ideal for many cat breed competitions. Thus, breeders may
select for the Ticked allele in these breeds, even though the
cats are solid in coloration and patterning is not of interest.
The allele count suggested that several Burmese cats in the
99 Lives dataset do not have the p.Cys63Tyr allele but
instead the p.Ala18Val Ticked allelic variant.
DKK4 is a member of the dickkopf (Dkk) family of
cysteine-rich secretory proteins that are antagonists of Wnt
signaling pathways, involved in antero-posterior axial
patterning, limb development, somitogenesis and eye
formation (Niehrs 2006). Four Dkk proteins exist in
humans (Dkk1-4) and all contain two CRDs, each of
which contains five disulfide bonds. Dkk1, Dkk2, and Dkk4
inhibit Wnt signaling by binding to the Wnt co-receptors
LRP5/6 via the CRD2 domain. In humans, DKK4 is
localized to 8p11.21, which is near the centromere but on
the short arm of the chromosome, suggesting a re-
positioning of the cat centromere. The Dkk4 p.Cys63Tyr
and p.Ala18Val alleles are biologically plausible candidates
for influencing the Ticked phenotype in domestic cats. The
p.Ala18Val is predicted to interfere with correct signal
peptide cleavage, which in turn could affect the efficient
secretion of Dkk4. In contrast, p.Cys63Tyr disrupts a key
disulfide bond responsible for stabilizing the tertiary struc-
ture of Dkk4. Interestingly, Dkk4 expression has also been
previously implicated in hair follicle spacing and density
(Sick et al. 2006), acting as an inhibitor of primary hair
placode formation (Cui et al. 2010). Taken together, this
provides a plausible biological link between DKK4 dys-
function and the Ticked phenotype.
Supportive data has also been suggested in a detailed
recent study by Kaelin et al. (2020). This study combines
variant analyses from the published 99 Lives sequence data
with an elegant investigation using single-cell gene expres-
sion analysis in fetal skin of domestic cats to define the
location and timing of pattern development in felids.
Association analysis of additional DNA samples across
(n=115) and within breeds (n=238) provides further
support that variation in Dkk4 causes Ticked.
Phenotypic and genotypic data from the 99 Lives cat
genome sequencing project has supported the identification
of candidate variants for the Ticked phenotype in domestic
cat breeds. Additional genotyping of the proposed variants,
in a large cohort of phenotyped cats, as well as supportive
functional data, would clarify the role of these variants in
cat coat pattern development. The identified variants do not
clarify the pathways leading to the production of the spotted
coat phenotype in cats, suggesting that additional genes
influence other tabby patterns in domestic cats. The allelic
series for the Ticked locus is suggested as Ti
A
=Ti
CK
>Ti
+
,
where the Ti
A
allele represents the p.Cys63Tyr variant and
the Ti
CK
allele represents the p.Ala18Val variant.
Acknowledgements
Funding for this project has been provided in part the
University of Missouri College of Veterinary Medicine
Gilbreath–McLorn endowment (L.A.L.) and Winn Feline
Foundation (W16-030, MT18-009, MTW18-009 and
MT19-001) in support of the 99 Lives Cat Genome Project
(https://www.winnfelinefoundation.org/). We appreciate
the provision of cat images by cat breeders and owners to
document the tabby phenotypes in domestic cats. We
appreciate technical assistance and support with the
manuscript from Thomas R. Juba and assistance with
figures from Karen Clifford. We thank Dr. S-J Luo for
providing the phenotypic data of cats in the SRA. In the 99
Lives Consortium (2020 cat analysis –99Lives195/284)
the organizer was Leslie A. Lyons and the data analyst was
Reuben M. Buckley. Each member of the 99 Lives Consor-
tium (2020 cat analysis –99Lives195/284) provided at
least one greater than 159coverage genome of the
domestic cat or a wild felid to support the analyses of the
dataset.
Conflict of interest
The authors disclose there are no conflicts of interest. The
funders had no role in study design, data collection, data
analysis, interpretation of results or decision to publish.
Author contributions
Conceptualization, L.A.L.; methodology, R.J.H., L.A.L.; soft-
ware, R.M.B.; validation, R.M.B., L.A.L.; formal analysis,
R.J.H., L.A.L.; resources, L.A.L.; data curation, R.M.B.,
R.J.H., L.A.L.; writing –original draft preparation, L.A.L.;
writing –review and editing; R.M.B., R.J.H., L.A.L.; 99 Lives
Consortium; supervision, L.A.L.; project administration,
L.A.L.; funding acquisition, L.A.L. All authors have read
and agreed to publish this manuscript.
Data availability statement
Genome sequence and signalment for the 195 cats are
available in the NCBI SRA (www.ncbi.nlm.nih.gov) and
accession nos are provided in File S1. All supplementary
files have been uploaded to a figshare data sharing site:
https://doi.org/10.6084/m9.figshare.13611188.
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Ticked cats with DKK4 variants 329

References
Abdel-Malek Z.A., Scott M.C., Furumura M., Lamoreux M.L.,
Ollmann M., Barsh G.S. & Hearing V.J. (2001) The me-
lanocortin 1 receptor is the principal mediator of the effects of
agouti signaling protein on mammalian melanocytes. Journal of
Cellular Science 114, 1019–24.
Almagro Armenteros J.J., Tsirigos K.D., Sonderby C.K., Petersen
T.N., Winther O., Brunak S., von Heijne G. & Nielsen H. (2019)
SignalP 5.0 improves signal peptide predictions using deep neural
networks. Nature Biotechnology 37, 420–3.
Altschul S.F., Gish W., Miller W., Myers E.W. & Lipman D.J. (1990)
Basic local alignment search tool. Journal of Molecular Biology
215, 403–10.
Barsh G., Gunn T., He L., Schlossman S. & Duke-Cohan J. (2000)
Biochemical and genetic studies of pigment-type switching.
Pigment Cell Research 13,48–53.
Baxter L.L., Loftus S.K. & Pavan W.J. (2009) Networks and
pathways in pigmentation, health, and disease. Wiley Interdisci-
plinary Reviews: Systems Biology and Medicine 1, 359–71.
Baxter L.L. & Pavan W.J. (2013) The etiology and molecular
genetics of human pigmentation disorders. Wiley Interdisciplinary
Reviews: Developmental Biology 2, 379–92.
Buckley R.M., Davis B.W., Brashear W.A. et al. (2020a) A new
domestic cat genome assembly based on long sequence reads
empowers feline genomic medicine and identifies a novel gene for
dwarfism. Plos Genetics 16, e1008926.
Buckley R.M., Gandolfi B., Creighton E.K. et al. (2020b) Werewolf,
there wolf: variants in Hairless associated with hypotrichia and
roaning in the lykoi cat breed. Genes (Basel) 11, 682.
Cai X., Yao Z., Li L. & Huang J. (2018) Role of DKK4 in
tumorigenesis and tumor progression. International Journal of
Biological Sciences 14, 616–21.
Cat Fanciers’ Association (2004) The Cat Fanciers’ Association
Complete Cat Book. Harper Collins Publishers, New York, NY.
Cogne B., Latypova X., Senaratne L.D.S. et al. (2020)
Mutations in the Kinesin-2 Motor KIF3B cause an autoso-
mal-dominant ciliopathy. American Journal of Human Genetics
106, 893–904.
Cui C.Y., Kunisada M., Piao Y., Childress V., Ko M.S. & Schlessinger
D. (2010) Dkk4 and Eda regulate distinctive developmental
mechanisms for subtypes of mouse hair. PLoS One 5, e10009.
doi: https://doi.org/10.1371/journal.pone.0010009
Dr€
ogem€
uller C., R.€
ufenacht S., Wichert B., Leeb T. (2007) Muta-
tions within the FGF5 gene are associated with hair length in
cats. Anim Genet. 38(3), 218–21. https://doi.org/10.1111/j.
1365-2052.2007.01590.x
Eizirik E., David V.A., Buckley-Beason V., Roelke M.E., Schaffer
A.A., Hannah S.S., Narfstrom K., O’Brien S.J. & Menotti-
Raymond M. (2010) Defining and mapping mammalian coat
pattern genes: multiple genomic regions implicated in domestic
cat stripes and spots. Genetics 184, 267–75.
Eizirik E., Yuhki N., Johnson W.E., Menotti-Raymond M., Hannah
S.S. & O’Brien S.J. (2003) Molecular genetics and evolution of
melanism in the cat family. Current Biology 13, 448–53.
Horie A., Fujiwara H., Sato Y., Suginami K., Matsumoto H.,
Maruyama M., Konishi I. & Hattori A. (2012) Laeverin/aminopep-
tidase Q induces trophoblast invasion during human early
placentation. Human Reproduction 27, 1267–76.
Jaffey J.A., Reading N.S., Giger U., Abdulmalik O., Buckley R.M.,
Johnstone S., Lyons L.A. & Lives Cat Genome C (2019) Clinical,
metabolic, and genetic characterization of hereditary methe-
moglobinemia caused by cytochrome b5 reductase deficiency in
cats. Journal of Veterinary Internal Medicine 33, 2725–31.
Kaelin C. & Barsh G. (2010) Tabby pattern genetics –a whole new
breed of cat. Pigment Cell and Melanoma Research 23, 514–6.
Kaelin C.B., McGowan K.A. & Barsh G.S. (2020) Developmental
geneticsof color pattern establishment incats. https://www.biorxiv.
org/content/ https://doi.org/10.1101/2020.11.16.385609v1.
Kaelin C.B., Xu X., Hong L.Z. et al. (2012) Specifying and sustaining
pigmentation patterns in domestic and wild cats. Science 337,
1536–41.
Kehler J.S., David V.A., Schaffer A.A., Bajema K., Eizirik E., Ryugo
D.K., Hannah S.S., O’Brien S.J. & Menotti-Raymond M. (2007)
Four independent mutations in the feline fibroblast growth factor 5
gene determine the long-haired phenotype in domestic cats.
Journal of Heredity 98, 555–66.
Li G., Hillier L.W., Grahn R.A. et al. (2016) A high-resolution SNP
array-based linkage map anchors a new domestic cat draft
genome assembly and Provides detailed patterns of recombina-
tion. G3 (Bethesda) 6, 1607–16.
Linneaus C. (1758) Systema Naturae per Regna Tria Naturae,
Secundum Classes, Ordines, Genera, Species, Cum Characteribus,
Differentiis, Synonymis, Locis. Laurentii Salvii, Holmiae.
Lomax T. & Robinson R. (1988) Tabby pattern alleles of the
domestic cat. Journal of Heredity 79,21
–3.
Lyons L.A., Bailey S.J., Baysac K.C. et al. (2006) The Tabby cat locus
maps to feline chromosome B1. Animal Genetics 37, 383–6.
Lyons L.A., Imes D.L., Rah H.C. & Grahn R.A. (2005) Tyrosinase
mutations associated with Siamese and Burmese patterns in the
domestic cat (Felis catus). Animal Genetics 36, 119–26.
Maruyama M., Hattori A., Goto Y., Ueda M., Maeda M., Fujiwara H.
& Tsujimoto M. (2007) Laeverin/aminopeptidase Q, a novel
bestatin-sensitive leucine aminopeptidase belonging to the M1
family of aminopeptidases. Journal of Biological Chemistry 282,
20088–96.
Mauler D.A., Gandolfi B., Reinero C.R., O’Brien D.P., Spooner J.L.,
Lyons L.A. & Consortium L (2017) Precision medicine in cats:
Novel Niemann-Pick type C1 diagnosed by whole-genome
sequencing. Journal of Veterinary Internal Medicine 31, 539–44.
Menotti-Raymond M., Deckman K.H., David V., Myrkalo J., O’Brien
S.J., Narfstr€
om K. (2010) Mutation discovered in a feline model of
human congenital retinal blinding disease. Invest Ophthalmol Vis
Sci 51(6), 2852–9. https://doi.org/10.1167/iovs.09-4261
Montague M.J., Li G., Gandolfi B. et al. (2014) Comparative analysis
of the domestic cat genome reveals genetic signatures underlying
feline biology and domestication. Proceedings of the National
Academy of Sciences 111, 17230–5.
Narfstrom K. (1983) Hereditary progressive retinal atrophy in the
Abyssinian cat. Journal of Heredity 74, 273–6.
Niehrs C. (2006) Function and biological roles of the Dickkopf
family of Wnt modulators. Oncogene 25, 7469–81.
Patel S., Barkell A.M., Gupta D. et al. (2018) Structural and
functional analysis of Dickkopf 4 (Dkk4): new insights into Dkk
evolution and regulation of Wnt signaling by Dkk and Kremen
proteins. Journal of Biological Chemistry 293, 12149–66.
Pettersen E.F., Goddard T.D., Huang C.C., Meng E.C., Couch G.S.,
Croll T.I., Morris J.H. & Ferrin T.E. (2021) UCSF ChimeraX:
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Lyons et al.330

structure visualization for researchers, educators, and developers.
Protein Science 30,70–82.
Pontius J.U., Mullikin J.C., Smith D.R. et al. (2007) Initial sequence
and comparative analysis of the cat genome. Genome Research 17,
1675–89.
Robinson R. (1991) Genetics for Cat Breeders. Pergamon Press,
Oxford.
Sick S., Reinker S., Timmer J. & Schlake T. (2006) WNT and DKK
determine hair follicle spacing through a reaction-diffusion
mechanism. Science 314, 1447–50.
Yu Y., Creighton E.K., Buckley R.M., Lyons L.A. & Consortium L
(2020) A deletion in GDF7 is associated with a heritable forebrain
commissural malformation concurrent with ventriculomegaly and
interhemispheric cysts in cats. Genes (Basel) 11, 672.
1 Appendix
99 Lives Consortium (2020 cat analysis –
99Lives195/284)
Marie Abitbol
1
, Danielle Aberdein
2
, Paulo C. Alves
3,4
, Asa
Ohlsson Andersson
5
, Rebecca R. Bellone
6
, Tomas F.
Bergstr€
om
7
, Nuket Bilgen
8
, Adam R. Boyko
9
, Jeffrey A.
Brockman
10
, Margret L. Casal
11
, Marta G. Castelhano
12
,
Brian W. Davis
13
, Lucy Davison
14
, Ottmar Distl
15
, Nicholas
H. Dodman
16
, N. Matthew Ellinwood
17
, Jonathan E.
Fogle
18
, Oliver P. Forman
19
, Dorian J. Garrick
2
, Edward I.
Ginns
20
, Jens H€
aggstr€
om
5
, Daisuke Hasegawa
21
, Bianca
Haase
22
, Vidhya Jagannathan
23
, Philippa Lait
24
, Isabel
Hernandez
25
, Marjo K. Hyt€
onen
26
, Maria Kaukonen
26
,
Tomoki Kosho
27
, Emilie Leclerc
28
, Teri L. Lear
29
, Tosso
Leeb
23
, Ronald H.L. Li
30
, Hannes Lohi
26
, Maria Longeri
31
,
Mark A. Magnuson
32
, Richard Malik
33
, Shrinivasrao P.
Mane
34
, Rondo Middleton
35
, John S. Munday
2
, William J.
Murphy
13
, Alexandra N. Myers
13
, Niels C. Pedersen
36
,
Simon M. Peterson-Jones
37
, Max F. Rothschild
38
, Clare
Rusbridge
39
, Jeffrey J. Schoenebeck
40
, Beth Shapiro
41
,
Joshua A. Stern
36
, William F. Swanson
42
, Karen A. Terio
43
,
Rory J. Todhunter
12
, Wesley C. Warren
44
, Elizabeth A.
Wilcox
12
, Julia H. Wildschutte
45
, Yoshihiko Yu
21
1
Universit
e de Lyon, VetAgro Sup, 69280, Marcy-l’Etoile,
France; Universit
e de Lyon, CNRS UMR5310, INSERM
U1217, Universit
e Claude Bernard Lyon I, Institut Neuro-
MyoG
ene, 69008 Lyon, France
2
School of Veterinary Science, Massey University,
Palmerston North 4442, New Zealand
3
CIBIO/InBIO, Centro de Investigac
ß
~
ao em Biodiversidade
e Recursos Gen
eticos/InBIO Associate Lab and Faculdade de
Ci^
encias, Universidade do Porto, Campus e Vair~
ao, 4485-
661 Vila do Conde, Portugal
4
Wildlife Biology Program, University of Montana, Mis-
soula, MT 59812, USA
5
Department of Clinical Sciences, Faculty of Veterinary
Medicine and Animal Science, Swedish University of Agri-
cultural Sciences, Uppsala SE-750 07, Sweden
6
Department of Population Health and Reproduction,
Veterinary Genetics Laboratory, University of California –
Davis, Davis, CA 95616, USA
7
Department of Animal Breeding and Genetics, Swedish
University of Agricultural Sciences, 750 07 Uppsala, Sweden
8
Department of Genetics, Veterinary Faculty, Ankara
University, 06110 Ankara, Turkey
9
Department of Biomedical Sciences, College of
Veterinary Medicine, Cornell University, Ithaca, NY
14853, USA
10
Hill’s Pet Nutrition Inc., Topeka, KS 66601, USA
11
Reproduction, and Pediatrics, School of Veterinary
Medicine, University of Pennsylvania, Philadelphia, PA
19104, USA
12
Department of Clinical Sciences, College of Veterinary
Medicine, Cornell University, Ithaca, NY 14853, USA
13
Department of Veterinary Integrative Biosciences, Col-
lege of Veterinary Medicine, Texas A&M University, College
Station, TX 77845, USA
14
Clinical Sciences and Services, Royal Veterinary College,
Hawkshead Lane, North Mymms, Hertfordshire AL9 7TA,
UK
15
Institute for Animal Breeding and Genetics, University
of Veterinary Medicine Hannover, 30559 Hannover, Ger-
many
16
Department of Clinical Sciences, Cummings School of
Veterinary Medicine, Tufts University, Grafton, MA 01536,
USA
17
National MPS Society, PO Box 14696, Durham, NC
27709, USA
18
Department of Population Health and Pathobiology,
College of Veterinary Medicine, North Carolina State
University, Raleigh, NC 27607, USA
19
WALTHAM Centre for Pet Nutrition, Freeby Lane,
Waltham on the Wolds, Leicestershire, LE14 4RT, UK
20
Department of Psychiatry, University of Massachusetts
Medical School, Worcester, MA 01655, USA
21
Laboratory of Veterinary Radiology, Nippon Veterinary
and Life Science University, Musashino, Tokyo 180-8602,
Japan
22
Sydney School of Veterinary Science, Faculty of Science,
University of Sydney, Sydney, NSW 2006, Australia
23
Vetsuisse Faculty, Institute of Genetics, University of
Bern, 3001 Bern, Switzerland
24
Langford Vets, University of Bristol, Langford, Bristol
BS40 5DU, UK
25
Pediatrics and Medical Genetics Service, College of Veteri-
nary Medicine, Cornell University, Ithaca, NY 14853, USA
26
Department of Veterinary Biosciences; Department of
Medical and Clinical Genetics, University of Helsinki and
Folkh€
alsan Research Center, Helsinki 00014, Finland
27
Department of Medical Genetics, Shinshu University
School of Medicine, Matsumoto, Nagano 390-8621, Japan
28
SPF - Diana Pet food –Symrise Group –56250 Elven,
France
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Ticked cats with DKK4 variants 331

29
Department of Veterinary Science, University of Ken-
tucky –Lexington, Lexington, KY 40506, USA (in memoriam)
30
Department of Surgical and Radiological Sciences,
School of Veterinary Medicine, University of California
Davis, One Shields Ave, Davis, CA 95616, USA
31
Dipartimento di Medicina Veterinaria, University of
Milan, 20122 Milan, Italy
32
Departments of Molecular Physiology and Biophysics,
Cell and Developmental Biology, and Medicine, Vanderbilt
University, School of Medicine, Nashville, TN 37232, USA
33
Centre for Veterinary Education, University of Sydney,
Sydney, NSW 2006, Australia
34
Elanco Animal Health, Greenfield, IN 46140, USA
35
Nestl
e Purina Research, Saint Louis, MO 63164, USA
36
Department of Medicine and Epidemiology, School of
Veterinary Medicine, University of California at Davis,
Davis, CA 95616, USA
37
Small Animal Clinical Sciences, College of Veterinary
Medicine, Michigan State University, East Lansing, MI
48824, USA
38
Department of Animal Sciences, College of Agricultural
and Life Sciences, Iowa State University, Ames, IA 20011,
USA
39
School of Veterinary Medicine, Faculty of Health and
Medical Sciences, University of Surrey, Guildford, Surrey
GU2 7AL, UK
40
Royal (Dick) School of Veterinary Studies and Roslin
Institute, University of Edinburgh, Easter Bush, Midlothian
EH25 9RG, UK
41
Department of Ecology and Evolutionary Biology,
University of California, Santa Cruz, Santa Cruz, CA
95064, USA
42
Center for Conservation and Research of Endangered
Wildlife (CREW), Cincinnati Zoo & Botanical Garden,
Cincinnati, OH 45220, USA
43
Zoological Pathology Program, University of Illinois,
Brookfield, IL 60513, USA
44
Division of Animal Sciences, College of Agriculture,
Food and Natural Resources, School of Medicine, University
of Missouri, Columbia, MO 65211, USA
45
Department of Biological Sciences, Bowling Green State
University, Bowling Green, OH, 43403, USA
Supporting information
Additional supporting information may be found online in
the Supporting Information section at the end of the article.
File S1 99 Lives Cat Genome Sequencing dataset signal-
ment, accessions and contacts.
File S2 (a) All variants in B1: 35–65 Mb for association
with Ticked. (b) Filtering of variants in B1: 35–65 Mb for
association with Ticked –strategy A.
File S3 (a) Filtering of variants in B1: 35–65 Mb for
association with Ticked –strategy B. (b) All variants in B1:
35–65 Mb for association with Ticked –strategy B (filter 2
only).
File S4 Structural variants in B1: 35–65 Mb for association
with Ticked.
File S5 DKK4 variants in the 195 cat 99 Lives dataset.
Table S1 Cat assembly V9.0 locations of the previously
linked STRs associated with the Ticked locus
1
.
©2021 The Authors. Animal Genetics published by
John Wiley & Sons Ltd on behalf of Stichting International Foundation for Animal Genetics, 52, 321–332
Lyons et al.332