An association between type A porcine endogenous retrovirus copy number and hematological parameters and gender in miniature pigs

concentration (MCHC) and platelet distribution width (PDW-CV) were significantly higher in female animals. No correlation between PERV-A env gene copy number and the coat color of animals was detected, suggesting that retroviral insertion sites and genes that determine the coat color of miniature pigs, namely KIT (chromosome 8) and MC1R (chromosome 6), are either located far apart on same chromosome or on different chromosomes. The copy number of PERV-A env gene in males was lower than in females. Presence of multiple copies of PERV-A on the X-chromosome is the most probable cause of such gender-related differences in miniature pigs. Thus, male miniature pigs of ICG SB RAS should be the source of material for xenotransplantation.

Porcine endogenous retroviruses (PERV) are RNAcontaining viruses that are capable of incorporating copies of DNA into the host cell's genome. Therefore, PERV cannot be eliminated by breeding pigs under specific pathogenfree conditions (Yudin et al., 2011). Since immunosuppressive therapy is mandatory after organ transplantation, the recipient has a high risk of developing PERV infection (Denner, 2016).
Based on the significant differences in the amino acid sequence of the receptorbinding domain of the env gene, which encodes a viral envelope protein, three main types of pig retrovirus, PERVA, PERVB and PERVC, have been recognized. PERVA and PERVB infects not only porcine cell lines, but also several human cell lines in vitro, while PERVC replicates only in porcine cells (Kimsa et al., 2014). Observations of patients treated with living porcine tissues or organs have not yet revealed the development of PERV infection in humans in vivo (Godehardt et al., 2015). It is not clear whether these data are due to real lack of virus production by pig cells or the result of effective inactivation of released virus particles by the immune system.
According to preliminary estimates, the pig genome contains 6 to 10 copies of replicable provirus, 30 to 50 fulllength PERV copies and 100 to 200 loci containing partial PERV sequence (Niebert, Tonjes, 2005). The number of retro viral integration sites, generally, positively correlates with the level of viral mRNA expression (Ka et al., 2009). The existing methods for determining the copy number of PERV in the genome based on realtime PCR require special equipment and reagents, as well as highly skilled personnel. Identification of simple phenotypic indications allowing to select for animals with low copy number of retroviruses in the genome is required for the widespread implementation of xenotransplantation.
A unique breed of Siberian miniature pig was developed at the Institute of Cytology and Genetics (ICG) SB RAS for biomedical applications by crossing pigs of Vietnamese Pot bellied breed, Large White and Landrace breeds, and wild boars of Central Asian subspecies (Tikhonov, 2010). Using material from miniature pigs, studies are carried out to test and produce bioprosthetic heart valves, blood vessels and pericardial flap for intracardiac surgery and angioplasty. A technique for obtaining chondrotransplants from newborn miniature pigs to treat cartilage tissue dystrophic and traumatic changes in idiopathic scoliosis has been developed. Yet, PERV retroviruses of all three types are present in the genomes of most miniature pigs (Aitnazarov et al., 2014).
The objective of this study was to evaluate the correlation between PERVA env gene copy number and hematological parameters, gender and coat color in miniature pigs of ICG SB RAS.

Materials and methods
1monthold miniature pigs were obtained from the Common Use Center for "Gene pools of fur and farm animals" of ICG SB RAS. Gender and coat color were visually determined. Venous blood was collected in tubes containing anticoagulant (EDTA K2). Automatic blood analyzer Hemascreen 18P (Hospitex Diagnostics, Italy) was used to determine 18 hematological parameters. The study was conducted in strict compliance with the Helsinki Declaration on Humane Treatment of Animals.
Whole blood DNA isolation was performed with protease treatment and phenol extraction (Sambrook, Russel, 2006). The copy number of PERVA env gene was determined by realtime PCR using standard samples prepared by the limiting dilution method (Aitnazarov et al., 2016) using Rotor Gene Q (Qiagen, Netherlands). Amplification was carried out using a set of realtime PCR reagents with the SYBR Green I dye (Sintol, Russia) according to the manufacturer's protocol. For each DNA sample, PCR was performed at least three times. Data was processed with RotorGene 6000 Series soft ware, version 1.8.17.5. In the calculations, the genomic DNA amount was assumed equal in pig cells and human cells (6 pg DNA per cell). Since all investigated parameters, according to the Kolmogorov -Smirnov test, were normally distributed, a oneway analysis of variance and a correlation coefficient were used to assess the impact of factors. The arithmetic mean and standard error were calculated for each experimental group. The results were processed using STATISTICA 8 software.

Results and discussion
Data on PERVA env gene copy number in blood cells from different groups of miniature pigs of ICG SB RAS are given in Table 1. The mean copy number for all studied animals (26.1, n = 40) significantly exceeds the value, which we previously obtained for pigs of the same breed (4.5, n = 10) (Aitnazarov et al., 2016). Influence of such factors as "boar genotype" and "sow genotype" on the PERVA env gene copy number was insignificant (F 6.33 = 0.65, p = 0.69 and F 14.25 = 1.08, p = 0.42, respectively). The minimum /maximum values were 12.3/38.1 in the present study and 0.6/58.9 in the previous study, which confirms data from other authors stating significant differences between animals in this parameter (Liu et al., 2011). The ob served differences might be related to a relatively small number of animals in our first study.
Determination of retroviral genome copy number in vivo or in silico remains a difficult task. The alignment of primers used in our study to the Duroc pig genome, version Sscrofa10.2/ susScr3 in the UCSC genomic browser, using Blat procedure allowed us to detect a total of seven copies: on chromosome 1 (two copies) and chromosomes 7, 8, 12, 13 and X (one copy each). However, this result may change when using another procedure and/or parameters for alignment. In addition to the individual differences mentioned above, it has been shown that the copy number of PERV retrovirus per cell may depend on the breed (Yu et al., 2007;Ma et al., 2010;Lee et al., 2011) and the examined organ Mazurek et al., 2013).
Earlier, based on the study of the prevalence of retrovirus in animals of different gender, we suggested that a copy(ies) of PERVA is localized on the Xchromosome in pigs of the Large White breed of the Achinsk type (Aitnazarov et al., 2006). In our study, the copy number of PERVA env gene in males was 16 % lower than in females ( p < 0.02), which is about 4 copies per cell. The most likely cause of the observed differences is the localization of several PERVA copies on the Xchromosome of miniature pigs. According to cytoge netic studies, PERVA is localized on the Xchromosome in Australian Westran pigs, but is absent on the same chromosome in Large White and native Korean pig breeds (Jung et al., 2010).
A similar gender effect has been observed in other species of mammals. The copy number of env gene of endogenous human retrovirus, which has been associated with the development of multiple sclerosis, in peripheral blood mononuclear  Ассоциация числа копий эндогенных ретровирусов PERV A с гематологическими показателями cells is significantly higher in women as compared to men (GarciaMontojo et al., 2013). According to the authors, this may be due to localization of at least two copies of retrovirus on the human Xchromosome. In cats, on the contrary, the copy number of env gene of endogenous retrovirus is higher in males than in females (Tandon et al., 2007). This is probably due to the localization of 3-5 provirus copies on the cat Ychromosome (Roca et al., 2005). We did not identify reliable correlation between the PERVA copy number per cell and the coat color of animals. In mi niature pigs of ICG SB RAS and in pigs of other breeds, the white coat color is controlled by the KIT gene (genotypes I/I and I/i) located on chromosome 8. MC1R gene located on chromosome 6 controls blackpied (genotype E p /E p ) and black (genotypes E d /E d , E d /E and E d /E p ) coat colors (Nikitin et al., 2016). It seems that retroviral insertion sites and genes that determine the coat color of miniature pigs are either located far apart on same chromosome or on different chromo somes.
A dilute viral (dv) coatcolor mutation is known in inbred mice, caused by the insertion of murine leukemia virus Emv 3 into the intron of the Myo5a gene (Seperack et al., 1995). Insertion of retrovirus leads to alternative splicing and formation of a shorter form of myosin heavychain protein, which is encoded by Myo5a gene. It is believed that this protein is necessary for maintaining normal structure of dendrites and organelle transport in melanocytes. Feline leukemia virus is more common in animals with a solid, rather than spotted fur color (McMichael et al., 1986). Insertion of retrovirus into the 5′ region of the aromatase gene in the Sebright chicken breed leads to the development of plumage color of roosters similar to the plumage color of hens (McPhaul et al., 1991).
Reference values for 18 haematological parameters of mi niature pigs of ICG SB RAS have been determined (Table 2). All blood parameters were in agreement with the accepted physiological norms (Kondrakhin, 2004) and data obtained from other breeds of pigs (Rispat et al., 1993;Ekser et al., 2012;Kawaguchi et al., 2012). The white blood cell count (WBC), monocyte, basophil and eosinophil count (MID#) and gra nu locyte count (GRA#), as well as the red blood cell distribu tion width (RDWCV) were significantly higher in males than in females. Females had higher mean corpuscular he mo globin concentration (MCHC) and platelet distribution width (PDW CV) as compared to males.
A study of 23 blood parameters in Japanese micro mini pigs has shown that basophil and lymphocyte counts were higher in females, but neutrophil count was higher in male animals (Kawaguchi et al., 2012). In Yucatan miniature pigs, the only detected difference between males and females was the higher platelet count in males (Rispat et al., 1993). The observed gender differences in hematological parameters between pigs of different breeds might be due to the influence of both genetic and environmental factors (feed, external conditions, etc.), as well as due to the different ages of studied animals.
No correlation between hematological parameters and the copy number of PERVA env gene in blood cells of male miniature pigs has been detected. In females, a significant negative correlation was found between the copy number and the mean corpuscular hemoglobin concentration (MCHC) (Figure). Possibly, a copy(ies) of an active form of PERVA retrovirus causes disruption of genes that regulate hemoglobin level in females to a greater extent than in males. It is known that hemoglobin level is on average 12 % lower in women than in men and this ratio is also observed in many species of mammals, birds and reptiles (Murphy, 2014). Since both genders have similar erythropoietin levels, it was hypothesised that oestrogens dilate and androgens constrict the renal microvasculature. Dilatation and vasoconstriction respectively increase and decrease the hematocrit in blood, providing a mechanisms for varying the red cell mass without changes in erythropoiesis.
Thus, material for xenotransplantation should be taken from male miniature pigs of ICG SB RAS, since they have a much lower copy number of PERVA env gene as compared to females. It seems also promising to use some hematological parameters for this purpose, but this issue requires further investigation. Correlation between PERV-A env gene copy number in blood cells and the mean corpuscular hemoglobin concentration (MCHC, g/L) in female miniature pigs.