Developmental and Comparative Immunology 40 (2013) 180–184
Contents lists available at SciVerse ScienceDirect
Developmental and Comparative Immunology journal homepage: www.elsevier.com/locate/dci
IgA deficiency in wolves Marcel Frankowiack a, Lars Hellman b, Yaofeng Zhao c, Jon M. Arnemo d,e, Miaoli Lin f, Katarina Tengvall g, Torsten Møller h, Kerstin Lindblad-Toh g, Lennart Hammarström a,⇑ a
Department of Laboratory Medicine, Division of Clinical Immunology, Karolinska Institute at Karolinska University Hospital Huddinge, SE-14186 Stockholm, Sweden Department of Cell and Molecular Biology, Uppsala University, SE-75124 Uppsala, Sweden c State Key Laboratory of Agrobiotechnology, College of Biological Sciences, China Agricultural University, Beijing 100193, PR China d Department of Forestry and Wildlife Management, Hedmark University College, NO-2418 Elverum, Norway e Department of Wildlife, Fish and Environmental Studies, Swedish University of Agricultural Sciences, SE-901 83 Umeå, Sweden f State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yet-Sen University, CN-510060 Guangzhou, PR China g Department of Medical Biochemistry and Microbiology, Uppsala University, SE-75124 Uppsala, Sweden h Kolmården Animal Park, SE-61892 Kolmården, Sweden b
a r t i c l e
i n f o
Article history: Received 10 December 2012 Revised 7 January 2013 Accepted 7 January 2013 Available online 22 January 2013 Keywords: Serum IgA concentrations IgA deficiency Dog Wolf
a b s t r a c t Low mean concentrations of serum immunoglobulin A (IgA) and an increased frequency of overt IgA deficiency (IgAD) in certain dog breeds raises the question whether it is a breeding-enriched phenomenon or a legacy from the dog’s ancestor, the gray wolf (Canis lupus). The IgA concentration in 99 serum samples from 58 free-ranging and 13 captive Scandinavian wolves, was therefore measured by capture ELISA. The concentrations were markedly lower in the wolf serum samples than in the dog controls. Potential differences in the IgA molecule between dogs and wolves were addressed by sequencing the wolf IgA heavy chain constant region encoding gene (IGHA). Complete amino acid sequence homology was found. Detection of wolf and dog IgA was ascertained by showing identity using double immunodiffusion. We suggest that the vast majority of wolves, the ancestor of the dog, are IgA deficient. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Selective IgA deficiency (IgAD) is the most common primary immunodeficiency in humans with a prevalence of 1 in 600 in Caucasians (Hammarström and Smith, 2007). The disorder is defined as serum immunoglobulin A (IgA) concentrations equal to or less than 0.07 g/l with normal levels of serum IgG and IgM in children 4 years of age or older (Al-Herz et al., 2011). Even though most of the patients with IgAD are asymptomatic, 30% suffer from recurrent infections at mucosal sites (Al-Herz et al., 2011). The genetic background of IgAD is complex, but there is a strong association between IgAD and genes within the MHC region (Ferreira et al., 2012) and non-MHC genes, the latter including interferon induced with helicase C domain 1 (IFIH1), also known as MDA5 (melanoma differentiation-associated protein 5) C-type lectin domain family 16, member A (CLEC16A), and, albeit to a lesser extent, several other autoimmunity risk alleles (Ferreira et al., 2010). Domestication of the dog is associated with two narrow genetic bottlenecks around 15,000 years, and, more recently, 200 years ago (Lindblad-Toh et al., 2005; Karlsson and Lindblad-Toh, 2008). At ⇑ Corresponding author. Fax: +46 852483588. E-mail address:
[email protected] (L. Hammarström). 0145-305X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dci.2013.01.005
these occasions, only a limited number of wild ancestors contributed to generate a restricted gene pool for the domestic dog, leading to long haplotypes and long disequilibrium linkages, making the canine genome a suitable model for tracing the genetic background of complex human diseases (Lindblad-Toh et al., 2005; Karlsson and Lindblad-Toh, 2008). Multiple studies on serum IgA concentrations in dogs have found ‘‘normal’’ IgA concentrations in most dog breeds tested (Felsburg et al., 1985; German et al., 1998; Griot-Wenk et al., 1999; Clemente et al., 2010). However, low concentrations, or even overt deficiency, have been found in selected breeds including the German shepherd dog (Whitbread et al., 1984; Griot-Wenk et al., 1999; Littler et al., 2006), shar pei dogs (Moroff et al., 1986; Rivas et al., 1995) and beagle dogs from selected kennels (Felsburg et al., 1985; Glickman et al., 1988). Whereas low serum concentrations of IgA are not associated with clinical symptoms in a majority of animals (Griot-Wenk et al., 1999), infections, especially of the skin and the upper respiratory tract as well as intestinal bacterial overgrowth, are known to be associated with IgAD in dogs (Felsburg et al., 1985; Moroff et al., 1986; Batt et al., 1991). With the exception of serum IgE concentrations, being approximately twice the concentration of those in dogs (Ledin et al., 2008), little is known about Ig concentrations in wolves. This is
M. Frankowiack et al. / Developmental and Comparative Immunology 40 (2013) 180–184
of particular interest, since the gray wolf is considered to be the ancestor of the dog (Vonholdt et al., 2010). We therefore set out to determine whether the low concentrations of serum IgA in selected dog breeds was passed on from their wolf ancestors or whether it is a consequence of selective breeding during domestication.
181
products were cloned into a pMD19-T vector and sequenced. For analyzing the sequence homology between dog and wolf, the obtained nucleotide sequence was aligned using MegAlign (DNASTAR). The used primer pair enables amplification of all four variants. 2.4. Immunodiffusion
2. Material and methods 2.1. Samples Ninety-nine serum samples originating from 71 wolves, obtained between 1998 and 2004, were used. Fifty-eight animals were free-ranging and 13 individuals were captive. The former represent approximately 30% of the wolf population in Scandinavia (Ledin et al., 2008). Between one and four samples were taken per wolf at different occasions. For 53 of the animals (including the captive ones) only one sample was available. The time between repetitive sampling ranged from around one year to more than 4 years. All samples were kept at 18 °C until used. 2.2. Canine IgA ELISA Antibodies previously successfully applied for measuring serum IgA concentrations in dogs (polyclonal goat anti-dog IgA antibodies (AbD Serotec), monoclonal mouse anti-dog IgA antibodies (AbD Serotec) and polyclonal, AP-conjugated goat anti-mouse antibodies (Jackson Immunoresearch)) were used as antibody set A. A second set of antibodies, antibody set B (polyclonal goat anti-dog IgA antibodies (AbD Serotec) and alkaline phosphatase (AP)-conjugated goat anti-dog IgA antibodies (Bethyl Laboratories)), was also used. IgA levels in all serum samples were measured by capture ELISA. The protocols were previously successfully used to screen dog serum samples for IgA deficiency in our laboratory (Tengvall et al., 2013). In short, 100 ll of primary antibody, diluted 1:2,000 in carbonate/bicarbonate buffer, was coated onto polystyrene plates (Corning) and incubated at 4 °C overnight. The plates were then washed with phosphate-buffered saline with 0.05% Tween (PBST) and serum samples diluted 1:5,000, 1:10,000 and, 1:20,000 in PBST were added and incubated overnight at 4 °C. The plates were subsequently washed and the secondary antibody, diluted 1:2000 in Tris-buffered saline with 0.05% Tween (TBST) was added, followed by incubation for 2 h. The plates were then washed again and the substrate para-Nitrophenylphosphate (PNPP) was added (antibody set B). For antibody set A, the tertiary antibody was added in a 1:2000 dilution and incubated for 2 h before the plates were washed and substrate (PNPP) was added. Serum samples from dogs kindly provided by Professor M.J. Day, Bristol University (England) with previously determined IgA concentrations were used as controls for each individual measurement. All samples were measured twice using antibody set B and at least once using antibody set A.
Identity between wolf and dog IgA in serum samples was investigated by double immunodiffusion as previously described (Ouchterlony et al., 1950). Briefly, polyclonal goat anti-dog IgA antibodies (AbD Serotec) and monoclonal mouse anti-dog IgA antibodies (AbD serotec) were used to analyze serum samples from wolves and dogs. Antisera and serum samples were added to holes with a defined distance, punched in an 0.9% agarose matrix. The samples were allowed to diffuse in the matrix for 72 h at 4 °C. 3. Results 3.1. Immunoglobulin concentrations IgA levels in all serum samples were measured by capture ELISA. A high correlation between the IgA concentrations obtained for 99 serum samples taken from the 71 wolves using two different sets of antibodies was observed (r2 = 0.91%, Fig. 1). The sample with the highest IgA concentration was found have 0.55 g/l (antibody set A) and 0.61 g/l (antibody set B) respectively. This sample may be considered an outlier, since its IgA concentration was three times higher than the second highest sample. Complete absence of IgA was observed for two of the samples using both antibody sets. Furthermore, five samples showed an IgA concentration between 0.23 and 0.11 g/l. All the remaining samples (n = 93) were found to have an IgA concentration of 0.1 g/l or less. The IgA concentrations in the wolves were compared with those of 50 samples from various dog breeds (10 standard poodles, 10 miniature poodles, 10 Tibetan terriers, 10 Rottweilers, and 10 Boxers) and 154 serum samples from German shepherd dogs, a breed know to have low serum IgA concentrations (Whitbread et al.,
2.3. Sequencing of the IGHA gene Genomic DNA was prepared from blood samples. The IgA heavy chain constant region gene (IGHA gene), encoding the heavy chain constant region (CH1 to CH3) was amplified in two wolves using primers targeting the IGHA gene in dog (GenBank: L36871): 50 CATGAAGACCTGTGCATTTCTCA 30 , 50 AGGGACTCAATTGTGAGGAGGAA 30 . The PCR conditions were: 94 °C for 5 min for 1 cycle; followed by 30 cycles of 94 °C for 30 s, 60 °C for 30 s, and 72 °C for 120 s and a final extension at 72 °C for 7 min. The resulting 1.7 kb PCR
Fig. 1. IgA concentrations obtained for 99 serum samples from wolves using two different sets of antibodies. Samples with similar concentrations are displayed by one marker only.
182
M. Frankowiack et al. / Developmental and Comparative Immunology 40 (2013) 180–184
IGHA gene in wolf and one (variant B) of the four published allelic variants of the IGHA gene in dogs (Peters et al., 2004) revealed a 100% identity (Fig. 3). A comparison to the whole length of this dog IGHA variant (B) is not possible, due to limited length of the published sequence. However, there are only two nucleotide substitutions for the dog IGHA GenBank sequence (variant A, accession number: L36871) (Peters et al., 2004) outside the hinge and the CH2 region at positions 179 and 581 in the cDNA alignment (Fig. 4), and there are no differences in the predicted amino acid sequence for the sequenced part of the IgA molecule (Fig. 5). The sequences obtained for the two different DNA samples from wolf were identical. The IgA concentrations of these samples were 0.08 and 0.12 g/l, respectively. 3.3. Immunodiffusion Fig. 2. Mean IgA concentrations obtained for serum samples from wolf (n = 99), German shepherd dogs (GSH) (n = 154) and randomly selected control samples from other dog breeds (n = 50). N Miniature poodles, Tibetan terriers, j standard boxers, rottweilers. poodles,
1984; Batt et al., 1991; Littler et al., 2006) (Fig. 2). The mean concentrations (±2 SD) were found to be 0.05 g/l (±0.12 g/l) for the wolf serum samples, 0.25 g/l (±0.33 g/l) for the German shepherd dogs and 0.30 g/l (±0.41 g/l) for the serum samples from dogs of varies other breeds. No sex- or age-dependency could be detected for serum IgA concentrations (data not shown). Furthermore, the IgA concentration did not differ between animals kept in captivity (n = 13) or living in the wild (n = 58) (data not shown).
3.2. Sequence analyses Very low IgA concentrations (60.1 g/l) were obtained in the vast majority of wolf serum samples. Hence, the question arose whether there is a problem in the detection of wolf IgA. Even though the antibodies used are polyclonal, major differences in the amino acid sequence could result in a reduced detectability of wolf IgA. To investigate whether sequence differences between wolf and dog exist, the IGHA gene (genomic sequence encoding the CH1 to CH3) was amplified and sequenced in two wolves. An alignment of the hinge region and the CH2 encoding exon of the
As shown in Fig. 6, a line of precipitate between goat anti-dog IgA antibody (primary antibody of the antibody sets A and B) and serum samples from dog and wolf can be observed. The connecting lines of precipitation imply that the antibodies bound to the same epitopes on IgA from dog and wolf. No precipitation was observed using mouse anti-dog IgA antibodies, possibly due to assay dependent properties. 4. Discussion The aim of this study was to investigate whether the low IgA concentrations in some dog breeds are a breeding-enriched phenomenon or whether similarly low IgA concentrations can be detected in wolves. Since there is no standard for wolf IgA, the obtained low IgA concentrations raised concerns whether polyclonal antibodies targeting dog IgA are suited to detect wolf IgA under the conditions used. However, molecular identity shown by double immunodiffusion as well as a high sequence similarity for the IGHA gene and the predicted identity in amino acid sequence in dogs and wolves prove the detectability of wolf IgA using anti-dog IgA capture ELISA. Although the dog and wolf a chain encoding sequences are similar/identical, we cannot exclude that the low concentration of IgA in the wolf samples is due to as yet uncharacterized variants in regulatory elements affecting the expression of IgA.
Fig. 3. Nucleotide and amino acid sequence comparison of 4 IGHA allelic variants in dog with the sequence in wolf and GenBank sequence L36871. Figure adapted from Peters et al. (2004). The IGHA nucleotide and amino acid sequence in wolf is identical to the B variant of the 4 allelic variants of the IGHA gene.
M. Frankowiack et al. / Developmental and Comparative Immunology 40 (2013) 180–184
183
Fig. 4. cDNA alignment of the IGHA gene in wolf and dog. Blue line: exon 1, orange line: exon 2, green line: exon 3. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 5. Amino acid alignment of the IGHA in wolf and dog.
An association between a locus in the bone morphogenetic protein receptor type-1B (BMPR1B) on dog chromosome 32 and low IgA (60.1 g/l) concentrations has recently been observed (Tengvall et al., 2013). BMPR1B encodes a BMP receptor also involved in T-cell differentiation and IgA switching and thus, is a good candidate for affecting the concentrations of IgA.
Synteny between the dog chromosome 32 and the human chromosome 4 has previously been observed (Ostrander, 2007). However, association with IgAD in humans have only been found for the MHC region (Ferreira et al., 2012) as well as IFIH1 and CLEC16A on chromosome 2 and 16, respectively on chromosome 6 (Ferreira et al., 2010). Moreover, unpublished data from our group show a
184
M. Frankowiack et al. / Developmental and Comparative Immunology 40 (2013) 180–184
Fig. 6. Double immunodiffusion showing binding of goat anti-dog IgA antibody (ab 1) to IgA in serum samples from both dog and wolf. Binding could only be detected for ab 1 and IgA from dog and wolf, visualized as white precipitation lines. The fused precipitation lines imply binding to the same epitope(s) in sera from dog and wolf.
genetic association with factors on chromosome 9 and low, albeit not overt IgAD, IgA concentrations (0.08–0.7 g/l). In humans, IgAD is characterized by serum IgA concentrations 60.07 g/l (Al-Herz et al., 2011) with normal levels of IgG and IgM in children 4 years of age or older. There is currently no generally accepted cut-off for IgAD for dogs or wolves. Previous findings have suggested two cut-offs, at
