|
|
||||||||







*
Department of Engineering and Natural Sciences, University of Växjö, Växjö, Sweden;
Department of Animal Breeding and Genetics, Swedish University of Agricultural Sciences, Uppsala, Sweden; and
Department of Veterinary Microbiology, Division of Immunology, Swedish University of Agricultural Sciences, Uppsala, Sweden
| Abstract |
|---|
|
|
|---|
production in whole blood cultures, and Ab responses to two
Escherichia coli antigens) were determined in 200
F2 animals from a wild pigSwedish Yorkshire intercross.
The pedigree has been typed for 236 genetic markers covering all
autosomes, the X chromosome and the X/Y pseudoautosomal region. Through
interval mapping using a least-squares method, four QTLs with
significant effects were identified; one for total leukocyte counts,
one for mitogen-induced proliferation, one for prevaccination levels of
Abs to E. coli Ag K88, and one for Ab response to the O149
Ag. In addition, several putative QTLs were indicated. The results from
the present study conclusively show that it is possible to identify
QTLs for immune capacity traits in outbred pig populations by genome
analysis. | Introduction |
|---|
|
|
|---|
Several studies have shown that genetic polymorphism within the MHC influences variation in immune functions and/or disease resistance (6). However, non-MHC genes also regulate immune responsiveness, and a polygenic control of quantitative immune response traits has been confirmed in several experimental studies (7). In the pig, additive genetic variation has been documented for a number of immune traits, e.g., Ab response (8, 9, 10, 11, 12), proliferative and cytokine responses of mononuclear cells (12, 13, 14, 15, 16), delayed-type hypersensitivity reactions (14), and total and differential leukocyte counts (16). Medium high to high heritabilities (h2 = 0.30.8) have been estimated for several of the immune traits suggesting a large genetic impact.
Linkage mapping using dense genetic maps is a straightforward approach to locate genes that control traits and inherited diseases with a monogenic inheritance. The same approach has also successfully been used to identify quantitative trait loci (QTLs)4 that control various polygenic traits in different organisms (17, 18, 19, 20, 21, 22, 23). Regarding immune functions, several QTLs influencing Ab response have recently been identified in mice (24, 25). Detailed linkage maps of the porcine genome are now available (26, 27, 28). In Sweden, a reference pedigree, created by intercrossing the European wild pig with Swedish (Sw) Yorkshire pigs, was initiated in 1988 as a part of a Nordic project on animal gene mapping (19, 29, 30). The pedigree has up to now been typed for more than 200 genetic markers (28), and QTLs for production traits such as growth and fat deposition have been identified (19, 30).
We here report on the identification of QTLs for immune capacity in the
pig. Several measures of immune function (i.e., total and differential
leukocyte counts, neutrophil phagocytosis, mitogen-induced
proliferation and IL-2 production, virus-induced IFN-
production of
mononuclear cells, and Ab responses to Escherichia coli Ags)
were determined for the F2 animals in the pedigree. Through
interval mapping using a least-squares method, we identified four QTLs
with significant effects on some of these immune traits, all located on
different chromosomes.
| Materials and Methods |
|---|
|
|
|---|
The animals originated from a reference pedigree developed for gene mapping by crossing two European wild pig boars with eight Sw Yorkshire sows (19, 29). Four of the F1 boars were subsequently mated to 22 F1 sows generating 44 litters, from which a total of 200 F2 animals were included in the present study. The eight parental sows had earlier been included as dams in a study in which heritabilities were estimated for some of the immune parameters (16). All animals were housed under conventional conditions. The male piglets were castrated at 2 wk of age, and all pigs were weaned at 7 wk of age by removing the sow from the farrowing pen. The weaned piglets were raised litterwise until approximately 3 mo old when they were allocated to a finishing unit where they were housed in groups (n = 40) of five nonlittermates. Immediately before allocation, blood samples were collected from the cranial vena cava by using evacuated glass tubes (B-D Vacutainer, Maylan, France) without additive and with 15 mg EDTA K3 or 143 U heparin as additives. In addition, serum samples were collected 2 and 5 wk after allocation for antibody determination. Serum and EDTA blood samples for DNA analyses were stored at -70°C.
Several growth and carcass traits were recorded during the rearing period and at slaughter of the pigs (19, 30). Intestinal specimens were collected after slaughter, and all animals were typed for the presence of the intestinal receptor for E. coli K88ab, K88ac (31).
Genetic markers
The pedigree has been typed for 236 genetic markers, including blood group loci, protein markers, RFLPs, and microsatellites (28). The established linkage map, which covers nearly all parts of the pig genome, has a total sex-average map-length of 2300 centiMorgans (cM). Linkage groups have been assigned to all 18 autosomes, the X chromosome and the X/Y pseudoautosomal region.
Total and differential white blood cell counts
The total number of white blood cells (WBC) was determined in a celloscope, and differential WBC counts were carried out by using blood smears stained with Giemsas and May Grünewalds solutions. The analyses were performed at the Department of Clinical Chemistry, Swedish University of Agricultural Sciences. From these data the total numbers of polymorphonuclear leukocytes (PMNL), band neutrophils, segmented neutrophils, eosinophils, and lymphocytes per liter blood were calculated.
Phagocytic capacity of blood PMNL
The phagocytic capacity of blood PMNL was determined in whole blood by a luminol enhanced chemiluminiscence (CL) assay as earlier described (16, 32). The phagocytosis was induced with opsonized zymosan particles, and the peak CL value was chosen as a measure of the phagocytic capacity.
Mitogen-induced proliferation and IL-2 production
The mitogen-induced proliferation was determined in whole blood cultures incubated as previously described (33). Cultures were stimulated with Con A (5 µg/ml; Pharmacia, Uppsala, Sweden), PHA (1 µg/ml; Wellcome, Dartford, U.K.), and PWM (10 µg/ml; Boehringer Mannheim, Mannheim, Germany), respectively, or grown in plain growth medium for 72 h, the last 24 h in the presence of 0.5 µCi [3H]thymidine (specific activity, 5 Ci per mmol; Amersham International, Amersham, U.K.). The proliferation was expressed as mean cpm values of triplicate wells.
The levels of IL-2 activity were determined in supernatants collected from the cultures above after 48 h of incubation. A bioassay measuring the proliferation of an IL-2-dependent cytotoxic T lymphocyte line was used (15), and the IL-2 activity was expressed as percentage of a laboratory standard included at every test occasion.
Aujeszkys disease virus (ADV)-induced IFN-
production
The IFN-
production of PBMC was induced with ADV
(34) and adapted to cultures of whole blood
(33). In brief, 200 µl of heparinized blood was diluted
1:10 in growth medium and incubated on monolayers of fixed, ADV
(Phylaxia strain)-infected porcine kidney cells (PK-15; Flow
Laboratories, Irvine, Scotland, U.K.) in flat-bottomed microtiter
plates. After 18 h at 37°C, the culture supernatants were
collected and the IFN-
content was measured by a sensitive
immunoassay for porcine IFN-
(35).
Immunization procedure and detection of serum antibodies to E. coli Ags
The F2 animals were immunized intramuscularly with an E. coli vaccine (Porcovac, Hoechst, Germany), containing the K88ab, K88ac, and O149 Ags (31). One animal per pen (n = 40) was used as a nonvaccinated control. Blood samples were collected immediately before immunization and 3 wk later. Serum IgG Abs to the K88ab, K88ac, and O149 Ags, respectively, were determined by ELISA (EC Diagnostics, Uppsala, Sweden), and the antibody levels are expressed as optical density values at a wave length of 492 nm, OD492 values.
Statistical analysis
An analytical method based on least-squares for the identification of QTLs segregating in crosses between outbred lines (36, 37) was used. In this analysis, the probabilities for every individual being each of the three genotypes (homozygous for alleles from the wild pig or from the Sw Yorkshire, or inheriting one allele from each breed) were estimated by using marker information for a putative QTL in a given position in the linkage group. The phenotypic score (i.e., an immune capacity trait) of an individual was regressed onto the additive and dominance coefficients of the QTL calculated from these genotype probabilities. The additive effects were estimated as half of the deviation of animals homozygous for the wild pig alleles from those homozygous for the Sw Yorkshire alleles, and the dominance effects as the deviation of heterozygous animals from the mean of the two homozygotes. The F-ratio values obtained by repeating the procedure at 1-cM intervals were plotted against the chromosomal position, with the highest point of the curve representing the most likely position of a QTL. The significance levels used are the significance thresholds obtained by permutation tests (38) for the immune traits. Covariates and fixed effects (i.e., family, sex, live weight, age) were analyzed using the GLM procedure of SAS (39) and included in the QTL analysis when appropriate.
The difference between the parental populations was also tested by calculating the average proportion of wild pig genome (i.e., the average additive and dominance coefficients from the QTL analysis) of each F2 animal. This parameter was then used as covariate in an analysis of covariance, and the difference between wild pig and Sw Yorkshire was estimated as twice the additive effect.
| Results |
|---|
|
|
|---|
6.0 are included. The F-values
corresponding to suggestive significance thresholds, i.e., one false
significant result per genome analysis (p
0.05 for each of the 18 autosomes), were between 4.3 and 5.9 for the
individual chromosomes. It is likely that several of the suggested QTLs
represent type I errors, but they are included to facilitate
comparisons with results of future studies and to reduce the risk that
true QTLs which did not reach the genome-wide 5% significance
threshold in the present study are ignored.
|
The total number of blood leukocytes (23.0 ± 6.4 x
109/L, mean value ± SD, n = 199) in
the F2 wild pig x Sw Yorkshire intercrosses was
slightly higher than previously reported for purebred Sw Yorkshire pigs
(20.6 ± 4.4 x 109/L blood, n =
124 (16)). The analysis of this parameter revealed one
significant QTL located 78 cM from the proximal end of chromosome 1
(Table I
; Figs. 1
and 2
). Only the
additive effect was significant, and the wild pig allele at this QTL
was associated with a higher number of leukocytes (Table I
), the
difference between the two homozygotes being 6.3 x
109 leukocytes per liter blood.
|
|
Differential leukocyte counts revealed that the circulating
leukocytes were equally distributed into PMNL (46.1 ± 12.2%) and
lymphocytes (47.0 ± 12.1%). The total number of PMNL (10.8
± 5.0 x 109/L blood, n = 199) was
somewhat higher for the wild pig intercrosses than for purebred Sw
Yorkshire pigs (8.7 ± 2.9 x 109/L blood,
n = 124 (16)). For both PMNL and
lymphocytes, QTLs were indicated in the same region of chromosome 1 as
for total numbers of WBC (F = 5.7 and 4.8,
respectively, p > 0.2). The PMNL were further
subdivided into segmented neutrophils (10.5 ± 4.8 x
109/L blood), band-formed neutrophils (0.3 ± 0.7
x 109/L blood), eosinophils (0.8 ± 0.8 x
109/L blood), and basophils (0.09 ± 0.17 x
109/L blood), respectively. No significant QTL was found
for differential cell counts, but several putative (0.05 <
p
0.20) QTLs were indicated (Table I
and Fig. 1
).
Proliferation and IL-2 production
A large inter-individual variation was found for the spontaneous and mitogen-induced proliferation and IL-2 production. As regards proliferation (spontaneous, 3,010 ± 3,450 cpm; PHA induced, 22,200 ± 13,900 cpm; PWM induced, 28,600 ± 10,800 cpm; Con A induced, 52,500 ± 22,300 cpm; n = 199), the coefficient of variation was similar to that previously recorded for purebred Sw Yorkshire pigs. In contrast, the coefficient of variation for the IL-2 production in these cultures (spontaneous, 16.2 ± 17%; PHA induced, 27.9 ± 30.7%; PWM induced, 186.8 ± 110.3%; Con A induced, 54.8 ± 57.9%; n = 199) was higher than that for purebred pigs (16). Because the previous study was carried out with purified PBMC and the present analyses were performed using whole blood cultures, a comparison of the absolute values is not valid.
For the mitogen-induced proliferation, the highest F value
(8.6) was found for a QTL on chromosome 4 with effect on the
PWM-induced proliferation (Table I
). Animals carrying wild pig alleles
at this QTL showed a lower PWM-induced proliferation, with an additive
gene effect corresponding to a difference between the homozygotes of
8890 cpm.
As regards mitogen-induced IL-2 production, the wild pig alleles were
on average associated with a higher PHA-induced IL-2 activity when this
trait was regressed on the average proportion of wild pig genome
(p < 0.01). However, no individual QTL
affecting IL-2 production reached the genome-wide significance
threshold (Table I
).
Phagocytic capacity and ADV-induced IFN-
production
The phagocytic capacity in whole blood, expressed as peak CL per
PMNL, was 3.3 ± 1.8 mV per 106 cells (mean value
± SD, n = 167). The coefficient of variation was
similar to that previously recorded for purebred Sw Yorkshire (6.3
± 3.4 mV per 106 cells, n = 72
(16)). The IFN-
production showed a pronounced
phenotypic variation (26.8 ± 41.8 U IFN-
per ml,
n = 179). Similarly, a large inter-individual variation
was previously found for this trait using purified PBMC of purebred Sw
Yorkshire pigs (122.8 ± 212.9 U IFN-
per ml, n
= 124 (16)). No QTL influencing the phagocytic capacity
was identified. Neither was any QTL found for the ADV-induced IFN-
production in whole blood cultures, although a higher production was
indicated (p < 0.05) for wild pigs when the
effect of the average proportion of wild pig genome was tested.
Specific IgG levels to E. coli Ags
The E. coli immunization provoked a significant
(p
0.001) increase in Ab levels to the
E. coli Ags tested, from an OD492 value of
0.33 ± 0.01 to 0.64 ± 0.02 for K88 and from 0.37 ±
0.01 to 0.61 ± 0.01 for O149, respectively (least-square
means ± SE). The nonvaccinated control animals (n
= 40) remained at prevaccination levels (OD492 value,
0.36 ± 0.04 and 0.38 ± 0.04 for IgG levels to K88 and O149,
respectively).
A QTL with significant effect on the specific Ab response to O149 was
found approximately 68 cM from the proximal end of chromosome 6 (Table II
and Fig. 1
). The wild pig alleles were associated with higher IgG
response to the O149 Ag, the additive gene effect leading to a
difference in Ab response (day 21-0) between the homozygotes of 0.22
(OD492 value). Also, prevaccination IgG levels differed
between pigs carrying wild or Yorkshire alleles. A significant QTL for
prevaccination levels of IgG to the Ag K88 was found on chromosome 5
with the Sw Yorkshire alleles associated with higher titers. Additional
putative QTLs influencing IgG levels to the E. coli Ags
before and after immunization were also identified (Table II
and
Fig. 1
).
|
| Discussion |
|---|
|
|
|---|
A QTL with a large effect on the total number of WBC was identified on chromosome 1, and a number of other QTLs influencing specific leukocyte populations were indicated. Earlier studies have shown that the levels of circulating neutrophilic granulocytes are under genetic control in humans (40, 41, 42), cattle (43), rats (44), and mice (45). For purebred Sw Yorkshire pigs a high heritability (h2 = 0.87 ± 0.41) has been suggested for the total number of circulating neutrophilic granulocytes along with a medium high heritability estimate for the total number of WBC (16). In concordance with these heritability estimates from purebred Yorkshire pigs, the total WBC numbers were found to vary between animals carrying different Yorkshire alleles at the QTL on chromosome 1. Moreover, the present results support earlier findings suggesting that several loci are involved in the regulation of blood leukocyte levels (44, 45). The present material did not allow identification of the actual gene(s) involved in the regulation of leukocyte numbers, but those regulating leukopoiesis and/or the release of leukocytes into the circulation will be likely candidates.
The genetic variation earlier documented for mitogen-induced proliferation and IL-2 activity in the pig (12, 13, 14, 15, 16) was confirmed as several putative QTLs were found for these traits; for mitogen-induced proliferation on chromosomes 4, 6, and 7, and for IL-2 production on chromosomes 7 and 12. In addition, a QTL for spontaneous proliferation was indicated on chromosome 13. A chromosomal region harboring genes that control the maintenance of IL-12 responsiveness and thus induce Th1 cell development was recently identified on murine chromosome 11 (46). Several genes with potential influence on T cell differentiation are located within this region, including IL-4, IL-5, IL-3, and IFN regulatory factor. A region syntenic with this part of murine chromosome 11 has been localized to human chromosome 5q31:1, which in turn corresponds to pig chromosomes 2q21-qter or 16 (47). No effect of these chromosomal regions was however detected in the present study.
Up to a 100-fold variation in IFN-
production between individuals
has been reported in both mice and humans (48). In mice,
this variation is controlled by a number of nonstructural If
genes that influence the amount of IFN yield per cell. The murine
If-1 locus, located on chromosome 3, affects the Newcastle
disease virus-induced expression of IFN-
, TNF-
, and IL-6 genes
(49). Also in pigs, a large phenotypic variation in levels
of virus-induced IFN-
has been found (16) in addition
to breed differences in IFN-
yield per cell (50). In
spite of that, a very low heritability was found for the ADV-induced
production of IFN-
among purebred Sw Yorkshire pigs, and no QTL was
found for this trait in the present study. However, the large variation
in IFN-
production found among the Sw Yorkshire population implies
that alleles with effects on IFN-
production might segregate in the
founder population. In such case, the possibility to detect putative
QTLs among the intercrosses is limited with the current experimental
design.
Ab response was one of the first immune capacity trait to be examined by genetic analysis. The establishment of selection lines of mice with high and low Ab production to certain Ags (7) was followed by similar studies in domestic animals (12, 51, 52, 53, 54, 55). In most of these studies, medium high heritabilities were estimated. For the IgG response to the E. coli Ags O149 and K88, heritabilities of 0.29 and 0.45, respectively, were estimated in a population of crossbred Sw Yorkshire x Sw Landrace pigs (11). In addition, single genes/gene complexes influencing the Ab response have been identified. The effect of MHC loci on Ab response has been thoroughly studied in mice (6) and also in the pig (9, 56, 57, 58). A higher IgG response to the K88 Ag has been found in pigs expressing a dominantly inherited intestinal receptor for this Ag encoded by the K88acR locus on chromosome 13 (31). In the present study, two QTLs with strong effect on IgG levels to the E. coli Ags were identified in addition to several putative QTLs. One of these QTLs influenced prevaccination levels of IgG to K88 with pigs carrying domestic alleles at this locus having higher IgG levels, whereas the other QTL influenced the production of Abs to O149 after vaccination. However, none of these QTLs were located on chromosomes 7 (MHC) or 13 (K88acR), suggesting that, for these loci, there were no fixed differences in allele frequencies between the populations or that the effect was too small to be detected in this limited sample. Several studies in mice show that more than 10 loci, including the MHC and Igh loci, influence the Ab response in mice to various antigens (7, 24, 25). The role of MHC molecules in recognition of processed Ags and thus its influence on the elicited Ab response is well documented, but in concordance with the studies in mice, our results show that additional non-MHC loci are important for the Ab response. In that context it is notable that the locus that affected the production of Abs to K88 after natural, oral infections with E. coli did not influence the Ab production to K88 after intramuscularly immunization. Due to differences in route of entry, adjuvant activity, dose, and nature of the Ag, the two types of Ab responses studied most likely involve distinct events in Ag handling, which are under different genetic influence.
The pedigree used in the current study was designed to facilitate the
mapping of genes controlling phenotypic differences between the
European wild pig and domestic Sw Yorkshire pigs. Domestic and wild
pigs show marked differences for a number of traits such as growth and
fat deposition, and QTLs for those traits have also successfully been
mapped (19, 30). Although differences in immune capacity
between wild and domestic pigs have so far not been studied, it is
likely that such differences exist due to the markedly different
environmental conditions to which these populations are exposed. In the
present study it would thus have been beneficial to measure the
phenotypic traits also for the parental and F1 populations.
Unfortunately, it was not possible to maintain reasonably large
populations of wild pig and F1 animals due to practical and
economical reasons. We therefore used a genome-wide test to look for
differences between the two populations. The F2 animals
were on average 50% wild pig, but the exact proportion of wild pig
alleles varies between individuals due to Mendelian segregation. Using
all marker data, the proportion of wild pig genome for each
F2 animal was estimated. A significant effect of this
parameter on the traits studied is expected if there is a net
difference, summed over many loci, between alleles from the two
populations. However, only PHA-induced IL-2 production and ADV-induced
IFN-
production showed a significant difference between the two
parental populations, with the wild pigs having higher levels. The
power of the experiment to detect QTLs for immune capacity is thus
lower than for traits like growth and fat deposition.
Although the pigs were clinically healthy, one cannot exclude that the observed genetic effects are exerted on responses to infectious agents or reflects genetic influence on susceptibility to infections because the study was performed on conventionally reared pigs. Differences in immune capacity due to divergent responses to potential "stressors" within the management (e.g. weaning (59, 60), transport (61, 62), and mixing and allocation (34, 63)), may also influence the results. Whether the observed QTL effects reflect differences in immune capacity between domestic and wild pigs, in susceptibility to subclinical infections and/or in the ability to cope with potential stressors, cannot be disentangled in the present study. Nevertheless, chromosomal regions harboring loci with influence on the examined traits were identified and warrants further studies to identify the actual genes involved.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Inger Edfors-Lilja, Department of Engineering and Natural Sciences, University of Växjö, S-351 95 Växjö, Sweden. ![]()
3 Current address: Animal Health Trust, Center for Preventive Medicine, Lanwades Park, Kentford, CB8 7UU United Kingdom. ![]()
4 Abbreviations used in this paper: QTL, quantitative trait locus; Sw, Swedish; WBC, white blood cells; PMNL, polymorphonuclear leukocytes; CL, chemiluminiscence; ADV, Aujeszkys disease virus. ![]()
Received for publication June 2, 1997. Accepted for publication March 23, 1998.
| References |
|---|
|
|
|---|
in serum and signs of reduced immune function in pigs after transport and installation in a fattening farm. Vet. Immunol. Immunopathol. 23:345.[Medline]
production and tissue localization of interferon-
/ß producing cells after intradermal administration of Aujeszkys disease virus-infected cells in pigs. Scand. J. Immunol. 41:121.[Medline]
producing (NIP) cell frequency and of interferon yield per cell. Vet. Immunol. Immunopathol. 37:113.[Medline]
This article has been cited by other articles:
![]() |
S. Timmusk, E. Merlot, T. Lovgren, L. Jarvekulg, M. Berg, and C. Fossum Regulator of G protein signalling 16 is a target for a porcine circovirus type 2 protein J. Gen. Virol., October 1, 2009; 90(10): 2425 - 2436. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. H. Wikstrom, B. M. Meehan, M. Berg, S. Timmusk, J. Elving, L. Fuxler, M. Magnusson, G. M. Allan, F. McNeilly, and C. Fossum Structure-Dependent Modulation of Alpha Interferon Production by Porcine Circovirus 2 Oligodeoxyribonucleotide and CpG DNAs in Porcine Peripheral Blood Mononuclear Cells J. Virol., May 15, 2007; 81(10): 4919 - 4927. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. L. Vincent, B. J. Thacker, P. G. Halbur, M. F. Rothschild, and E. L. Thacker An investigation of susceptibility to porcine reproductive and respiratory syndrome virus between two genetically diverse commercial lines of pigs J Anim Sci, January 1, 2006; 84(1): 49 - 57. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Marklund, J. Kijas, H. Rodriguez-Martinez, L. Rönnstrand, K. Funa, M. Moller, D. Lange, I. Edfors-Lilja, and L. Andersson Molecular Basis for the Dominant White Phenotype in the Domestic Pig Genome Res., August 1, 1998; 8(8): 826 - 833. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |