High resolution mapping of chromosomal regions controlling resistance to gastrointestinal nematode infections in an advanced intercross line of mice


Fine mapping of quantitative trait loci (QTL) associated with resistance to the gastrointestinal parasite Heligmosomoides polygyrus was achieved on F6/F7 offspring (1076 mice) from resistant (SWR) and susceptible (CBA) mouse strains by selective genotyping (top and bottom 20% selected on total worm count in week 6). Fecal egg counts were recorded at weeks 2, 4, and 6, and the average was also analyzed. Blood packed cell volume in weeks 3 and 6 and five immunological traits (mucosal mast cell protease 1, granuloma score, IgG1 against adult worm, IgG1, and IgE to L4 antigen) were also recorded. On Chromosome 1 single-trait analyses identified a QTL with effects on eight traits located at about 24 cM on the F2 mouse genome database (MGD) linkage map, with a 95% confidence interval (CI) of 20-32 cM established from a multitrait analysis. On Chromosome 17 a QTL with effects on nine traits was located at about 18 cM on the MGD map (CI 17.9-18.4 cM). Strong candidate genes for the QTL position on Chromosome 1 include genes known to be involved in regulating immune responses and on Chromosome 17 genes within the MHC, notably the Class II molecules and tumor necrosis factor.


Genetic resistance to gastrointestinal nematode parasites has been clearly demonstrated in both livestock (Gasbarre and Miller 2000; Gray et al. 1995) and mice (Wakelin 2000). Exploitation of this variation in breeding programs may be assisted by the mapping of the quantitative trait loci (QTL) that control resistance to the parasites in question. Recently, we have exploited the contrasting features of resistance to gastrointestinal (GI) nematode infection in SWR and CBA mouse strains, which are respectively resistant and susceptible to infection with Heligmosomoides polygyrus, a strongyloid nematode parasite of mice (Behnke et al. 2000). A genome-wide scan of an F2 cross between SWR and CBA detected 14 QTL on 12 different chromosomes affecting parasitological traits (Iraqi et al. 2003) and the immunological traits (Menge et al. 2003) associated with H. polygyrus infection. The QTL were mapped mostly within a relatively large genomic interval of 20-30 cM. Among the different strategies that have been proposed for high-resolution mapping of QTL (Darvasi 1998; McPeek 2000), we have used the advanced intercross line (AIL) approach described by Darvasi and Soller (1995). This approach was successful in the high-resolution mapping of QTL associated with susceptibility to trypanosomosis (Iraqi et al. 2000), pulmonary adenoma (Wang et al. 2003a), HDL cholesterol (Wang et al. 2003b), and malaria (Hernandez-Valladares et al. 2004). This article reports the fine mapping of QTL on Chromosomes 1 and 17 for both parasitological and immunological traits measured in F6 and F7 cross mice. The QTL with the largest effects were identified on these two chromosomes in the F2 mice (Iraqi et al. 2003; Menge et al. 2003).

Materials and methods

Generation of the advanced intercross line

SWR and CBA mice were purchased from Harlan UK Ltd. Reciprocal crossing of SWR and CBA produced F1 mice. Thirty pairs of F1 mice produced 50 breeding pairs of F2 mice. In subsequent generations of inter se mating, a minimum of 50 randomized nonrelated pairs produced the next generation. Five hundred F6 mice were produced from 100 mating pairs of F5 mice. Similarly, 600 F7 mice were produced from 100 mating pairs of F6 mice.

Assay of resistance

The strain of H. polygyrus used for this work corresponds to H. p. bakeri as described by Behnke et al. (1991). The parasite was maintained originally in Nottingham in CFLP mice (Jenkins and Behnke 1977) for more than 20 years and was imported to ILRI in 2000, where it was maintained subsequently in CBA mice by passage about every 2-3 months.

The trickle infection protocol used for infecting experimental mice is described fully and evaluated in Behnke et al. (2003). The experimental design comprised ten batches of 50 F6 mice and 12 batches of 50 F7 mice that were infected orally with 125 infective larvae of H. polygyrus suspended in 0.2 ml of water once every 7 days for 6 weeks (i.e., seven infections), starting at ten weeks of age. Ten SWR and ten CBA mice (five of each sex) were included in each batch as controls and were treated the same as the F6 and F7 mice.

Fecal egg counts (FEC) from individual mice were carried out on freshly collected feces from mice that were temporarily isolated on the morning of fecal collection and were recorded as eggs per gram of feces at 2, 4, and 6 weeks (FEC2, FEC4, and FEC6) after first infection. Approximately 60 μl of blood were collected from the tail vein of each mouse using a 75-μl heparinized capillary tube 3 and 6 weeks after first infection, and packed cell volume (PCV3 and PCV6) was recorded using microhematocrit. Blood samples were also collected on Day 0 of infection from the last four batches of F7 mice. Mice were euthanized 44 days after first infection and total worm count (TWC) recorded as the sum of adult and larval worms. Worms were extracted using the Baerman procedure in which the entire small intestine was removed, split longitudinally and incubated, and suspended in a gauze in a 50-ml beaker containing Hanks saline at 37°C for a minimum of 4 h. The worms migrate through the gauze and accumulate at the bottom of the beaker, from where they can be transferred to a Petri dish for counting under a dissecting microscope. The order of infecting and of recording of mice was randomized. Two additional CBA mice were infected in each batch; these two mice were euthanized and total worm count recorded 14 days after first infection to confirm the viability of the infective larvae. Tails were collected at necropsy and preserved at −20°C for subsequent DNA extraction. The three FEC measurements and TWC were transformed by taking logarithms to achieve approximately normal distributions of residuals (LFEC2, LFEC4, LFEC6, LTWC).

Cellular responses

Levels of mucosal mast cell protease 1 (mMCP1), reflecting the activation of mucosal mast cells, were measured in plasma samples collected from tail tips by heparinized capillary tubes at week 3, by antigen capture ELISA with reagents supplied by R&D Systems and Moredun Scientific Limited, respectively (Behnke et al. 2003). In responder strains of mice the levels of mMCP1 peak at this time and show maximum divergence from the considerably lower response of poor responder strains (Behnke et al. 2003). Data were logarithm transformed before analysis to normalize the variance of this trait (LmMCP1). Granuloma formation on the intestinal mucosa was assessed as a granuloma score (GS) on a scale of 0-4 (low to high) at post-mortem.

Antibody responses

Sera were collected at the end of week 6 of infection and were stored at −40°C until used. Levels of immunoglobin G1 (IgG1) against adult worms and those of both IgG1 and immunoglobin E (IgE) to L4 antigen were measured by ELISA. Mouse immunoglobulin isotype-specific, alkaline phosphatase-conjugated antisera were purchased from Serotec and the methods used for ELISA and for preparation of antigen have been described previously (Wahid and Behnke 1993; Wahid et al. 1994). Titers in all cases are expressed as the relative response index (RRI) calculated for each individual sample in relation to values obtained on each ELISA plate from sera from naïve and hyperimmune mice. The same pools of reference sera were used throughout this work.

Phenotypic analyses and genotyping strategy

Initially each trait was analyzed by least-squares fitting a statistical model that included fixed effects for line (CBA, SWR, and F6/F7), batches (22), sex (male and female), and statistically significant (p < 0.05) first-order interactions.

The F6/F7 data that were used for QTL analyses were analyzed separately, fitting the effects of batch, sex, and the interaction between batch and sex. Each trait was then corrected for the significant fixed effects and interactions estimated from the least-squares analysis, and then standardized by dividing by the residual standard deviation (values used in Table 1) so that all traits were compared on the same scale. Average fecal egg count (AVFEC), the average of the three standardized LFEC at weeks 2, 4, and 6, was also analyzed. Residual correlations among all the traits were also estimated.

Table 1 Numbers of mice and least squares means (standard error) for the CBA and SWR inbred lines (parental controls) and the F6/F7 crossbred mice

A selective genotyping strategy was used (Darvasi and Soller 1992; Ronin et al. 1998). The top and bottom 20% of the F6 and F7 mice were selected on standardized LTWC adjusted for the fixed effects (Fig. 1). Selection of mice for genetic analysis was carried out when phenotyping of the F6 mice was complete (batches 1-10) and then when the F7 mice had been phenotyped (batches 11-22). Thus, the final selection was a close approximation to the overall top and bottom 20% of progeny for adjusted LTWC. Selection of the resistant mice with the lowest standardized residuals was much easier to perform in this population where 19% of the F6/F7 mice had zero egg counts compared with the F2 population in which 82% of the mice had shown zero egg counts (Iraqi et al. 2003).

Fig. 1

Distribution of adjusted logarithm transformed worm counts presented as residual effects in standard deviation units.

DNA was obtained from mouse tail samples using a phenol chloroform extraction (Sambrook et al. 1989) and assayed for microsatellite genotypes using standard procedures on an ABI 377 slab gel automatic sequencer or on an ABI 3730 capillary sequencer (Applied Biosystems). Markers were initially chosen at 5-10-cM intervals, spanning the confidence intervals of the QTL regions on Chromosomes 1 and 17 detected by Iraqi et al. (2003) and Menge et al. (2003). Additional markers were added as required to close gaps where QTL were confirmed in the F6/F7 population.

Genotype verification, linkage, and mapping analyses

Linkage maps were constructed from marker data using MapMaker Express software (Lander 1987; Lincoln et al. 1994), using a Haldane map function. The resulting maps were compared with the mouse genome assembly, build 33 (http://www.ensembl.org/Mus_musculus/) and to the Mouse Genome Database (MGD) F2 map (http://www.informaticsjax.org/). The mouse genome assembly was assumed to provide the correct marker order. After allowing for the approximately 3-3.5-fold map expansion expected in a mixture of F6 and F7 animals compared with the F2 map, two markers that caused an unexpectedly large map expansion were eliminated from the data set because such map expansion generally indicates a substantial frequency of genotyping errors. The remaining markers were then included in QTL mapping analyses. This article reports the QTL mapping results for Chromosomes 1 and 17 which were the two chromosomes in the F2 mice where QTL with the largest effect were detected.

Single-trait maximum likelihood (ML) QTL mapping was performed using MapMaker Express/QTL. Single-trait least-squares (LS) interval mapping was performed using QTL Express (Haley and Knott 1992). Map distances were based on the linkage map distances resulting from the F6 and F7 data, which for ease of interpretation and comparison with the F2 results are shown interpolated back to the murine F2 MGD linkage map positions and distances. Two-QTL models were also fitted to each chromosome separately in the LS analyses. When the two-QTL models provided a significantly better fit than a one-QTL model, the largest QTL was fixed as a cofactor and the two-QTL model was run again to test significance of a third QTL on each chromosome separately. Chromosome-wide permutation tests were performed in the one-QTL model to determine significance thresholds, and these significance thresholds were applied to the two- and three-QTL significance tests. Bootstrapping was applied in the one-QTL model to obtain 95% confidence intervals (CI) for QTL location. The MapMaker QTL analyses were used for model and QTL validation, and the ML additive and dominance effects were tabulated and compared to the LS estimates from QTL Express. The selective genotyping strategy is expected to create a substantial upward bias in LS estimates of QTL effects compared with the ML estimates in which all the mice that were phenotyped are included in the analysis (Ronin et al. 1998).

Additional ML multiple-trait interval mapping analyses were performed using MultiQTL software (Korol et al. 1995, 2001). Multiple-trait analyses were performed for all 11 traits recorded (as shown in Table 1, but excluding AVFEC) and then for different subsets of the traits based on possible groupings identified in the single-trait analyses. Significance thresholds were obtained for LOD score estimates using chromosome-wide permutation testing involving 1000 random permutations of the phenotype data, and CIs were obtained by bootstrapping (Korol et al. 2001).


Phenotypic analyses

Table 1 shows that relative to the susceptible CBA mice the resistant SWR mice had lower FEC at all measurement times, lower TWC and IgG1-Ad, but higher PCV, GS, mMCP1, Ig-G1-L4, and IgE-L4, with all these differences being highly significant (p < 0.001). For many traits (i.e., FEC2, FEC4, FEC6, TWC, PCV6, GS, LmMCP1) the F6/F7 mice had mean values significantly different from the midparent average in the direction of the resistant SWR mice, indicating heterosis for resistance. For PCV3, IgG1-L4, and IgE-l4 the F6/F7 mean values were significantly (p < 0.001) higher than the mean for the SWR mice, indicating particularly high degrees of heterosis in the direction of the resistant parent. In the case of IgG1-Ad, the F6/F7 mean value was not significantly different (p > 0.05) from the susceptible CBA mice, indicating negative heterosis for susceptibility. PCV was also recorded at Day 0 in the last four batches of mice and there was no significant difference (p > 0.05) among the three strains (i.e., CBA, SWR, and F6/F7).

Within the F6 and F7 mice there were positive correlations among the LFEC measurements and with LTWC ranging from 0.24 to 0.74 (Table 2). PCV6 was negatively correlated with the FEC measurements and LTWC (ranging from −0.09 to −0.33), while the comparable correlations for PCV3 were slightly negative but not significantly different (p > 0.05) from zero (Table 2). The correlation of LTWC with the immunological traits was negative for GS, LmMCP1, and IgE-L4 but positive for IgG1-Ad and IgG1-L4 (Table 3). The three immunoglobin assays were positively correlated, while GS was positively correlated with IgG1-L4, negatively correlated with IgG1-Ad, and uncorrelated with LmMCP1 and IgE-L4 (Table 3). The correlation of LmMCP1 with the three immunoglobin assays was close to zero and ranged from 0.05 to −0.06 (Table 3).

Table 2 Residual correlationsa among logarithm transformed fecal egg counts (LFEC), logarithm transformed total worm count (LTWC), and blood packed cell volume (PCV) for the F6 and F7 mice
Table 3 Residual correlationsa among logarithm transformed total worm count (LTWC) and the immunological traitsb for the F6 and F7 mice

Quantitative trait loci

The map expansion of our F6/F7 map compared with the MGD F2 map was 3.8 for Chromosome 1 and 3.9 for Chromosome 17. This compares with an approximate 3-3.5-fold expansion expected in an F6/F7 population.

Summaries of test statistics and map locations for all significant QTL detected on Chromosomes 1 and 17 using QTL Express are presented in Table 4. Where a two-QTL model gave a significantly better fit to the data, the locations of the two QTL are also shown in Table 4. The estimate of the 95% CI of the QTL in Table 4 apply to the one-QTL model.

Table 4 Location,a confidence interval, and LOD score of significant quantitative trait loci detected on Chrs 1 and 17 using QTL Express

On Chromosome 1 the single-trait QTL analyses identified a QTL with substantial effects on FEC4, FEC6, AVFEC, and TWC, and QTL with smaller but still significant effects on FEC2, PCV6, IgG1-L4, and IgG1-Ad (Fig. 2). The very similar location of the QTL for most of the traits and the results of the multiple-trait analyses in Table 5 suggest that this is a single QTL located at about 24 cM on the F2 MGD map and with a 95% CI ranging from about 20 to 32 cM (Table 5). The smallest CI was 9.5 cM for TWC and ranged from 26.3 to 35.8 cM for the other traits.

Fig. 2

F ratios of association of microsatellite markers with parasitological (A) and immunological (B) traits on Chromosome 1 (MMU1). All markers are simple sequence repeats that have been ordered into a high-resolution linkage map. The F ratios were derived from maximum likelihood analyses of single traits in QTL Express. Abbreviations for traits correspond to those used in the text. We also include a physical map of the region below the horizontal axis based on the Ensembl database, with distance marked in units of 106 nucleotide base pairs from the proximal end of the chromosome, and provide locations of some of the candidate genes.

Table 5 Location, confidence interval (CI), and LOD scores of quantitative trait loci on Chrs 1 and 17 from multiple trait interval mapping analyses performed using MultiQTL software

The single-trait QTL analyses for Chromosome 17 identified a QTL with significant and substantial effects on FEC4, FEC6, AVFEC, TWC, GS, IgG1-Ad, and IgE-L4 and a much smaller effect on mMCP1 and IgG1-L4 (Fig. 3). The similarity of the locations of the QTL for the different traits suggests the presence of a single QTL with pleiotropic effects. Based on the multiple-trait analysis of all traits (Table 5), the estimated location of this QTL is at about 18 cM with a CI of 0.5 cM (17.9-18.4 cM). The CIs were larger for the QTL identified for the individual traits and were 0.7 cM for IgE-L4, 4.5 cM for GS, and 6.7 cM for TWC (Table 4). A two-QTL model gave a significantly better fit to the data than a one-QTL model for FEC4 (p < 0.05). The first QTL mapped to approximately the same location as in the one-QTL model, while the second QTL mapped distal to the first QTL at about 35 cM.

Fig. 3

F ratios of association of microsatellite markers with parasitological (A) and immunological (B) traits on Chromosome 17 (MMU17). See legend to Fig. 2 for further details.

Both the LS and ML additive and dominance effects for the QTL for the individual traits are shown in Table 6. The additive effect is estimated for the allele from the resistant SWR line and a negative value indicates that the SWR allele lowers the mean value of the trait while a positive value indicates that the SWR allele increases the trait. All the additive effects for the individual FEC measurements, AVFEC, TWC, and IgG1-Ad, on both Chromosomes 1 and 17 were negative, indicating that the SWR allelle lowered mean FEC, TWC, and IgG1-Ad (Figs. 4 and 5). The additive effects were positive for PCV6 on Chromosome 1 and positive for GS, mMCP1, and IgG1-L4 on Chromosome 17. All these additive effects are consistent with the phenotypic differences between the resistant SWR and susceptible CBA parent strains (Table 1). The negative additive effects for IgG1-L4 on Chromosome 1 and for IgE-L4 on Chromosome 17 are in opposite directions to the differences in parent strain phenotype. In the case of IgG1-L4 and IgE-L4 on Chromosome 17, the additive effects are small or moderate and the QTL exhibits positive overdominance (i.e., the dominance effect is larger than the positive additive effect), consistent with the high degree of heterosis observed for the parental strains. Most traits, except FEC2 and FEC4, exhibit substantial heterosis, but the dominance estimates for the QTL are not highly consistent with the heterosis exhibited by the F6/F7 crossbred mice.

Table 6 Least squaresa (LS) and maximum likelihoodb (ML) estimates of the additive and dominance QTL effects
Fig. 4

The contribution to the phenotype of the alleles of the two QTL for four parasitological traits [(A) fecal egg counts in week 2; (B) fecal egg counts in week 4; (C) fecal egg counts in week 6; and (D) worm counts in week 6]. Additive values are shown for each trait for mice homozygous with CBA alleles (CC), homozygous for SWR alleles (BB), and heterozygous (BC) for two markers nearest to the center of each QTL, D1mit478 (MMU1) and D17mit176 (MMU17).

Fig. 5

The contribution to the phenotype of the alleles of the two QTL for five immunological traits [(A) granuloma score in week 6; (B) plasma mucosal mast cell protease in week 3; (C) specific anti-L4 larval antigen IgG1 serum antibody concentration in week 6; (D) specific anti-adult worm antigen IgG1 serum antibody concentration in week 6; (E) specific anti-L4 larval antigen IgE serum antibody concentration in week 6]. For further details see legend to Fig. 4.

The effect of the selective genotyping that causes an upward bias in QTL effects estimated by LS is most marked for TWC, which was the trait on which mice were selected for genotyping. The additive effect for TWC on Chromosome 1 was biased upward by 85% and that for TWC on Chromosome 17 by 92%. This was also true for traits closely correlated with TWC such as FEC6 and FEC4 (Table 2), but the upward bias was not as large and ranged from 11% to 42%. However, for the immunological traits that were not closely correlated with TWC (Table 4), most of the LS and ML QTL effects were quite similar, consistent with an effectively random sample of mice having been genotyped for these traits.


The significant difference between the SWR and CBA parental lines for all the parasitological and immunological traits measured confirm previous studies (Behnke et al. 2003; Iraqi et al. 2003; Menge et al. 2003). The heterosis for resistance manifested by the F6/F7 mice also confirms the earlier F2 results (Iraqi et al. 2003; Menge et al. 2003). Protocols for measurement of FEC were modified following the earlier F2 experiments (i.e., fresh fecal samples were processed within one or two days of collection rather than being stored in formalin for measurement at a later date) to improve the accuracy of the resistance assay. The correlations among the three FEC measurements and with TWC achieved here (Table 2) were two to five times higher than those obtained in the earlier F2 QTL mapping experiment (Iraqi et al. 2003), indicating that the new protocols resulted in much higher accuracy of assay.

In general, the QTL detected on Chromosomes 1 and 17 in the F6/F7 mice confirm the QTL detected in the original F2 QTL mapping population (Iraqi et al. 2003; Menge et al. 2003). In the F2 population significant QTL were detected on Chromosome 1 for FEC4, AVFEC, TWC, and mMCP1 located at 18, 21, 21, and 85 cM, respectively. In the F6/F7 population a single QTL with pleiotropic effects on eight traits (Table 4) was detected and located at about 24 cM on the F2 MGD map, which falls within the CI of 27 cM (range = 18-45 cM) for TWC in the F2 mice. The CI for TWC in the F6/F7 mice was reduced by about one third to 9.5 cM (range = 26.3-35.8 cM). However, the QTL for mMCP1 located at 85 cM (range = 73-106 cM) in the F2 mice was not confirmed in the F6/F7 mice.

In the F2 mice there appeared to be two QTL on Chromosome 17, one toward the proximal end affecting TWC and four immunological traits (GS, mMCP1, IgG1-Ad, and IgE-L4) and the other toward the distal end affecting FEC2 and AVFEC. The QTL toward the distal end of chromosome 17 was not confirmed in the F6/F7 mice but that towards the proximal end was confirmed with the most likely interpretation being a single QTL located at about 18 cM with pleiotropic effects on nine traits which include FEC and TWC (Tables 4 and 5). The CI from the multiple-trait analysis of the five immunologic traits was reduced to 0.4 cM (range = 17.9-18.3 cM) with a LOD score of 62.7, which is very strong evidence for a QTL in this region of the chromosome. However, there was a larger CI of 6.7 cM (range = 16.4-23.1 cM) for TWC (Table 4). The largest reduction in the CI for the traits with significant QTL on Chromosome 17 was for IgE-L4, which was 0.7 cM in the F6/F7 mice (Table 4) compared with 30 cM in the F2 mice (Menge et al. 2003).

The QTL located on Chromosome 17 is at almost the identical position as that detected in a similar experiment that mapped QTL for resistance to trypanosome infection, where the CI of the largest QTL was reduced to 0.9 cM (range = 17.4-18.3 cM) in an F6 AIL (Iraqi et al. 2000) compared with approximately 10 cM for the same QTL in the F2 population (Kemp et al. 1997). This very small CI in part was due to the fact that both the C57BL6/J*A/J and the C57BL6/J*BALB/c F6 populations exhibited extremely large map expansion involving a large number of markers in the region that contained the QTL. This provided an unexpected substantial increase in mapping accuracy in exactly the region where the major QTL lay.

The estimates of QTL effect in this article are not subject to ascertainment bias because the QTL were originally detected in a previous experiment and the current experiment was designed to refine QTL location and obtain unbiased estimates of their effect. The ML estimation procedures should also have corrected for possible bias in estimates of QTL effects introduced by the selective genotyping procedures. The combined additive effects of the QTL on Chromosomes 1 and 17 (Table 6, Fig. 4) on FEC, TWC, and PCV explain between 21% (for PCV6) and 86% (for FEC4) of the observed parental strain difference in phenotype (Table 1). For the immunological traits, the range was from −41% for IgE-L4 to 163% for IgG1-Ad. Apart from errors of estimation, it is possible to explain more than 100% of the strain difference because the parental strains may well carry QTL with effects that are opposite to the strain difference, which are not assayed in this experiment. The estimates of QTL effect can be used to obtain estimates of their contribution to genetic variance (V g) in the AIL, assuming that the allele frequencies of the CBA and SWR alleles are each 0.5, as

$$ Vg = 0.5(a_{1}^{2}\;+\;a_{17}^{2})\;+\;0.25(d_{1}^{2}\;+\;d_{17}^{2}) $$

where a 1 and a 17 and d 1 and d 17 are the estimated additive and dominance effects of Chromosomes 1 and 17. This estimate of variance is biased upward because the estimates of a and d include the errors of estimation, but it remains instructive for comparison to the estimate of contribution of QTL to the strain differences. The estimates of V g range from 3% (for PCV6) to 31% (for FEC4) of the AIL residual variance given in Table 1. For the two key disease traits, FEC6 and TWC, the additive effects explained 56% and 54% of the difference in phenotype of the parental strains and the estimates of V g were 16% and 15% of the F6/F7 residual variance. The heritability of these traits in an AIL is unknown, but in sheep the heritabilities for FEC have been estimated to range from 0.23 to 0.41 (Dominik 2005). The results thus suggest that the QTL on Chromosomes 1 and 17 leave unexplained a substantial proportion of the genetic differences between CBA and SWA strains. This is likely true of most other traits, with the possible exception of IgG1-Ad, where the additive effects explained 160% of the strain difference and 20% of the F6/F7 variance. The heritability of IgG1-Ad is unknown, so the possibility remains that the QTL on Chromosomes 1 and 17 explain most of the genetic difference in IgG1-Ad response between the two parental strains.

The single QTL with pleiotropic effects on Chromosome 1, mapping to approximately 24 cM, lies in a relatively gene-poor region of this chromosome and is close to the insulin-dependent diabetes 5.1 (Idd5.1) locus for which candidate genes include Ctla4 (cytotoxic T-lymphocyte-associated protein 4 at 30.1 cM), Icos (inducible T-cell costimulator at 32.0 cM), Als2cr19 [amyotrophic lateral sclerosis 2 (juvenile) chromosome region, candidate 19 (human)], and Nrp2 (neuropilin-2) (Wicker et al 2004). Idd5.1 and Idd5.2 are believed to be the loci responsible for the non-obese diabetic (NOD) mouse mutation that suffers from autoimmune destruction of the islets of Langerhans (Lamhamedi-Cherradi et al. 2001) and have been implicated in experimentally induced autoimmune encephalomyelitis in mice (Greve et al. 2004). Ctla4, Icos, and CD28 (30.1 cM), the latter mapping just proximal to these loci, are all members of the immunoglobulin superfamily, expressed on T cells, and involved in interactions with B7 ligands on antigen-presenting cells (Loke et al. 2005), but previous experiments using knockouts and/or in combination with blockade of receptors/ligands by specific antibodies have indicated that gene products are not essential for resistance to GI nematodes in mice (Harris et al. 1999; Kopf et al. 2000; Scales et al. 2004), including H. polygyrus (Ekkens et al. 2002). Nevertheless, the possibility that SWR mice have alleles of some of these genes that enhance the response cannot be excluded on the basis of these studies. Cflar (Casp8 and FADD-like apoptosis regulator (FLIP) at 30.1 cM), which has been shown to have a role in Th2 polarization of immune responses (Tseveleki et al. 2004), also maps to this region. Proximally, on the periphery of the 95% confidence limits for TWC, FEC6, and FEC4 are genes involved in signal transduction and gene activation associated with Th1 lymphocytes, including activation of the IFN-γ pathway (STAT1 and STAT4; Kaplan 2005) at 25.9 cM (genetic distances from Mouse Genome Informatics Database; http://www.informatics.jax.org) and several QTL associated with susceptibility to infection (Salmonella typhimurium) and autoimmune and inflammatory diseases overlap with our QTL.

The QTL on Chromosome 17, spanning the murine MHC and associated loci, was expected not only on the basis of our earlier study (Iraqi et al. 2003) but also because experiments with congenic and recombinant mice over a decade earlier had established that both primary infections (Behnke and Wahid 1991) and secondary infections (Behnke and Robinson 1985; Enriquez et al. 1988) are regulated by genes within the H-2 region. SWR mice carry the H-2q haplotype, which is typically characteristic of other resistant strains (e.g., NIH) to H. polygyrus and also to other nematodes such as Trichinella spiralis (e.g., BUB/Bn and B10.G; Wakelin and Donachie 1983; Wassom et al. 1983). In contrast, mouse strains such as CBA and C3H, carrying the H-2k haplotype, and those with the H-2b haplotype (e.g., C57BL10) show poor resistance and sustain long chronic infection with H. polygyrus (Wahid and Behnke 1993), while expelling acute infections such as T. spiralis more slowly than resistant strains (Wakelin and Donachie 1983). Enriquez et al. (1988) concluded that at least two genes were involved in the case of H. polygyrus, one associated with the Class II, I-Eα locus (18.66–18.7 cM) and the other mapping toward the distal H-2D end of the MHC, close to the TNF locus (19.06 cM), and the positions of our QTL are in close concordance with these loci. However, it should be stressed that the QTL (Tir1) for resistance to Trypanosoma congolense in mice (Iraqi et al 2000), which was originally believed to lie within the MHC, was eventually shown to map just proximal to the MHC at 18 cM. Because the present multiple-trait analysis gave a position of 18.1 cM and the confidence limits for the positions of the QTL for most of the individual traits overlap with this region, we cannot exclude the possibility that genes proximal or distal to the MHC are also involved.

Taken together, the regions of Chromosomes 1 and 17 within which our QTL are located contain several candidate genes that are believed to play a regulatory role in immune responses. Which of these or other genes (the confidence limits for some traits encompass other possible candidates) plays the decisive role in determining susceptibility/resistance to H. polygyrus will be clarified only when the regions are dissected through the production of recombinant congenic strains by introgression of the responder alleles from SWR onto the CBA background. However, from our current (F6/F7) and earlier studies (F2), the work being conducted to map QTL for resistance to GI nematodes in sheep (Beh et al. 2002; Diez-Tascon et al. 2005) and that on the human GI nematode Ascaris lumbricoides (Quinnell 2003; Williams-Blangero et al. 2002), a picture is emerging that clearly indicates multigenic control of resistance to GI nematodes in mammals. The challenge for the future will be to identify the crucial combinations of genes whose alleles make the greatest difference to resistance, without compromising resistance to other pathogens and important production traits and which endow protection to a wide range of GI nematodes and other helminths. Clearly, MHC genes do not fall into this category, because diversity in MHC genes is beneficial in the face of natural challenge by a vast range of potential pathogens (Cooke and Hill 2000; Hedrick 1994; Penn et al. 2002; Wegner et al. 2003). However, it is conceivable that among the non-MHC genes there are combinations that are responsible for a significant proportion of the remaining variation in resistance phenotypes, with potential for exploitation in improving breeds of livestock for the future. Identification of these genes, their combinations, and alleles are priorities.


  1. Beh KJ, Hulme DJ, Callaghan MJ, Leish Z, Lenane I, et al. (2002) A genome scan for quantitative trait loci affecting resistance to Trichostrongylus colubriformis in sheep. Anim Genet 33:97–106

    PubMed  CAS  Article  Google Scholar 

  2. Behnke JM, Robinson M (1985) Genetic control of immunity to Nematospiroides dubius: a 9-day anthelmintic abbreviated immunizing regime which separates weak and strong responder strains of mice. Parasite Immunol 7:235–253

    PubMed  CAS  Google Scholar 

  3. Behnke JM, Wahid FN (1991) Immunological relationships during primary infection with Heligmosomoides polygyrus (Nematospiroides dubius): H-2 genes determine worm survival. Parasitology 103:157–164

    PubMed  Article  Google Scholar 

  4. Behnke JM, Keymer AE, Lewis JW (1991) Heligmosomoides polygyrus or Nematospiroides dubius? Parasitol Today 7:177–179

    PubMed  CAS  Article  Google Scholar 

  5. Behnke JM, Lowe A, Menge D, Iraqi F, Wakelin D (2000) Mapping genes for resistance to gastrointestinal nematodes. Acta Parasitol 45:1–13

    CAS  Article  Google Scholar 

  6. Behnke JM, Lowe A, Clifford S, Wakelin D (2003) Cellular and serological responses in resistant and susceptible mice exposed to repeated infection with Heligmosomoides polygyrys bakeri. Parasite Immunol 25:333–340

    PubMed  CAS  Article  Google Scholar 

  7. Cooke GS, Hill AVS (2000) Genetics of susceptibility to human infectious disease. Nat Rev Genet 2:967–977

    Article  CAS  Google Scholar 

  8. Darvasi A (1998) Experimental strategies for the genetic dissection of complex traits in animal models. Nat Genet 18:19–24

    PubMed  CAS  Article  Google Scholar 

  9. Darvasi A, Soller M (1992) Selective genotyping for determination of linkage between a molecular marker and a quantitative trait locus. Theor Appl Genet 85:353–359

    Article  Google Scholar 

  10. Darvasi A, Soller M (1995) Advanced intercross lines, an experimental population for fine genetic mapping. Genetics 141:1199–1207

    PubMed  CAS  Google Scholar 

  11. Diez-Tascon C, Keanne OM, Wilson T, Zadissa A, Hyndman DL, et al. (2005) Microarray analysis of selection lines from outbred populations to identify genes involved with nematode parasites resistance in sheep. Physiol Genomics 21:59–60

    PubMed  CAS  Article  Google Scholar 

  12. Dominik S (2005) Quantitative trait loci for internal nematode resistance in sheep: a review. Genet Sel Evol, 37:s65–s82

    Article  CAS  Google Scholar 

  13. Ekkens MJ, Liu Z, Liu Q, Foster A, Whitmire J, et al. (2002) Memory Th2 effector cells can develop in the absence of B7-1/B7-2, CD28 interactions, and effector Th cells after priming with an intestinal nematode parasite. J Immunol 168:6344–6351

    PubMed  CAS  Google Scholar 

  14. Enriquez FJ, Brooks BO, Cypess RH, David CS, Wassom DL (1988) Nematospiroides dubius: Two H-2 -linked genes influence levels of resistance to infection in mice. Exp Parasitol 67:221–226

    PubMed  CAS  Article  Google Scholar 

  15. Gasbarre LC, Miller JE (2000) Genetics of helminth resistance. In: Breeding for Disease Resistance in Farm Animals, 2nd ed., Axford REF, Bishop SC, Nicholas FW, Owen JB (eds.) (Wallingford, Oxfordshire, UK: CAB International) pp 129–152

    Google Scholar 

  16. Gray GD, Woolaston RR, Eaton BT (1995) Breeding for resistance to infectious diseases of small ruminants [Canberra, Australia: Australian Centre for International Agricultural Research (ACIAR)), Monogr No. 34

  17. Greve B, Vijayakrishnan L, Kubal A, Sobel RA, Peterson LB, et al.(2004) The diabetes susceptibility locus Idd5.1 on mouse chromosome 1 regulates ICOS expression and modulates murine experimental autoimmune encephalomyelitis. J Immunol 173:157–163

    PubMed  CAS  Google Scholar 

  18. Haley CS, Knott SA (1992) A simple regression method for mapping quantitative trait loci in line crosses using flanking markers. Heredity 69:315–324

    PubMed  CAS  Google Scholar 

  19. Harris NL, Peach RJ, Ronchese F (1999) CTLA4-Ig inhibits optimal T helper 2 cell development but not protective immunity or memory response to Nippostrongylus brasiliensis. Eur J Immunol 29:311–316

    PubMed  CAS  Article  Google Scholar 

  20. Hedrick PW (1994) Evolutionary genetics of the major histocompatibility complex. Am Naturalist 143:945–964

    Article  Google Scholar 

  21. Hernandez-Valladares M, Naessens J, Gibson JP, Musoke AJ, Rihet P, et al.(2004) Confirmation and dissection of QTL controlling resistance to malaria in mice. Mamm Genome 15:390–398

    PubMed  Article  Google Scholar 

  22. Iraqi F, Clapcot S, Kuman P, Haley C, Kemp S, Teale A (2000) Fine mapping of trypanosomiasis resistance QTLs in mice using advanced intercross lines. Mamm Genome 11:645–648

    PubMed  CAS  Article  Google Scholar 

  23. Iraqi F, Behnke J, Menge D, Lowe A, Teale A, et al. (2003) Chromosomal regions controlling resistance to gastro-intestinal nematode infections in mice. Mamm Genome 14:184–191

    PubMed  CAS  Article  Google Scholar 

  24. Jenkins SN, Behnke JM (1977) Impairment of primary expulsion of Trichuris muris in mice concurrently infected with Nematospiroides dubius. Parasitology 75:71–78

    PubMed  CAS  Google Scholar 

  25. Kemp SJ, Iraqi F, Darvasi A, Soller M, Teale AJ (1997) Localization of genes controlling resistance to trypanosomiasis in mice. Nat Genet 16:194–196

    PubMed  CAS  Article  Google Scholar 

  26. Kopf M, Coyle AJ, Schmitz N, Barner M, Oxenius A, et al. (2000) Inducible costimulator protein (ICOS) controls T helper cell subset polarization after virus and parasite infection. J Exp Med 192:53–61

    PubMed  CAS  Article  Google Scholar 

  27. Korol AB, Ronin YI, Kirzhner VM (1995) Interval mapping of quantitative trait loci employing correlated trait complexes. Genetics 140:1137–1147

    PubMed  CAS  Google Scholar 

  28. Korol AB, Ronin YI, Itskovich AM, Peng J, Nevo E (2001) Enhanced efficiency of quantitative trait loci mapping analysis based on multivariate complexes of quantitative traits. Genetics 157:1789–1803

    PubMed  CAS  Google Scholar 

  29. Lamhamedi-Cherradi SE, Boulard O, Gonzalez C, Kassis N, Damotte D, et al. (2001) Further mapping of the Idd5.1 locus for autoimmune diabetes in NOD mice. Diabetes 50:2874–2878

    PubMed  CAS  Google Scholar 

  30. Lander E, Green P, Abrahamson J, Barlow A, Daley M, et al. (1987) MAPMAKER: An interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1:174–181

    PubMed  CAS  Article  Google Scholar 

  31. Lincoln S, Daly M, Lander ES (1994) MAPMAKER/QTL 1.1 Whitehead Institute Technical Report Vol. 2 (Cambridge, MA: Whitehead Institute)

  32. Loke P, Zang X, Hsuan L, Waitz R, Locksley RM, et al. (2005) Inducible costimulator is required for type 2 antibody isotype switching but not T helper cell type 2 responses in chronic nematode infection. Proc Nat Acad Sci U S A 102:9872–9877

    PubMed  CAS  Article  Google Scholar 

  33. McPeek MS (2000) From mouse to human: fine mapping of quantitative trait loci in model organism. Proc Natl Acad Sci U S A 97:12389–12390

    PubMed  CAS  Article  Google Scholar 

  34. Menge DM, Behnke JM, Lowe A, Gibson JP, Iraqi F, et al. (2003) Quantitative trait loci for resistance to gastro-intestinal nematode infections in mice. Parasite Immunol 25:341–349

    PubMed  CAS  Article  Google Scholar 

  35. Penn DJ, Damjanovich K, Potts WK (2002) MHC heterozygosity confers a selective advantage against multiple-strain infections. Proc Nat Acad Sci U S A 99:11260–11264

    PubMed  CAS  Article  Google Scholar 

  36. Quinnell RJ (2003) Genetics of susceptibility to human helminth infection. Int J Parasitol 33:1219–1231

    PubMed  CAS  Article  Google Scholar 

  37. Ronin YI, Korol AB, Weller JI (1998) Selective genotyping to detect quantitative trait loci affecting multiple traits: interval mapping analysis. Theor Appl Genet 97:1169–1178

    Article  Google Scholar 

  38. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratory manual, 2nd ed. (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press)

    Google Scholar 

  39. Scales HE, Lerna MX, Gutierrez-Ramos JC, Coyle AJ, Garside P, et al. (2004) Effect of inducible costimulator blockade on the pathological and protective immune responses induced by the gastrointestinal helminth Trichinella spiralis. Eur J Immunol 34:2854–2862

    PubMed  CAS  Article  Google Scholar 

  40. Tseveleki V, Bauer J, Taoufik E, Ruan C, Leondiadis L, et al. (2004) Cellular FLIP (long isoform) overexpression in T cells drives Th2 effector responses and promotes immunoregulation in experimental autoimmune encephalomyelitis. J Immunol 173:6619–6626

    PubMed  CAS  Google Scholar 

  41. Wahid FN, Behnke JM (1993) Immunological relationships during primary infection with Heligmosomoides polygyrus (Nematospiroides dubius): parasite specific IgG1 antibody responses and primary response phenotype. Parasite Immunol 15:401–413

    PubMed  CAS  Google Scholar 

  42. Wahid FN, Behnke JM, Grencis RK, Else KJ, Ben-Smith AW (1994) Immunological relationships during primary infection with Heligmosomoides polygyrus: Th2 cytokines and primary response phenotype. Parasitology 108:461–471

    PubMed  CAS  Google Scholar 

  43. Wakelin D. (2000) Rodent models of genetic resistance to parasitic infections In: Breeding for Disease Resistance in Farm Animals, 2nd ed., Axford REF, Bishop SC, Nicholas FW, Owen JB (eds.) (Wallingford, Oxfordshire, UK: CAB International) pp 107–126

    Google Scholar 

  44. Wakelin D, Donachie AM (1983) Genetic control of immunity to Trichinella spiralis: influence of H-2-linked genes on immunity to the intestinal phase of infection. Immunology 48:343–350

    PubMed  CAS  Google Scholar 

  45. Wang M, Lemon W J, Liu G, Wang Y, Iraqi F, et al. (2003a) Fine mapping and identification of candidate pulmonary adenoma susceptibility 1 genes using advanced intercross lines. Cancer Res 63:3317–3324

    CAS  Google Scholar 

  46. Wang X, Le Roy I, Nicodeme E, Li R, Wagner R, et al. (2003b) Using advanced intercross lines for high-resolution mapping of HDL cholesterol quantitative trait loci. Genome Res 13:654–664

    Article  CAS  Google Scholar 

  47. Wassom DL, Brooks BO, Cypess RH, David CS (1983) A survey of susceptibility to infection with Trichinella spiralis of inbred mouse strains sharing common H-2 alleles but different genetic backgrounds. J Parasitol 69:1033–1037

    PubMed  CAS  Article  Google Scholar 

  48. Wegner KM, Reusch TBH, Kalbe M (2003) Multiple parasites are driving major histocompatibility complex polymorphism in the wild. J Evol Biol 16:224–232

    PubMed  CAS  Article  Google Scholar 

  49. Wicker LS, Chamberlain G, Hunter K, Rainbow D, Howlett S, et al. (2004) Fine mapping, gene content, comparative sequencing, and expression analyses support Ctla4 and Nramp1 as candidates for Idd5.1 and Idd5.2 in the nonobese diabetic mouse. J Immunol 173:164–173

    PubMed  CAS  Google Scholar 

  50. Williams-Blangero S, VandeBerg JL, Subedi J, Aivaliotis MJ, Rai DR, et al. (2002) Genes on chromosomes 1 and 13 have significant effects on Ascaris infection. Proc Natl Acad Sci U S A 99:5533–5538

    PubMed  CAS  Article  Google Scholar 

Download references


The authors acknowledge the technical inputs in the helminthology laboratory of Sam Njomo, Fredrick Moseti, and Sarah Kanyingi; mouse husbandry by Jane Ikanyi, John Kiragu, and Pauline Mbuthia under the supervision of Bob King; and John Wambugu, Moses Ogugo, Daniel Mwangi, and Nemuel Nyamweya for the genotyping in the molecular genetics laboratory. The authors are particularly grateful to Clare Kemp for preparing the color versions of Figs. 2 and 3. This research was funded by a grant from the Wellcome Trust (063810) and by member donors of the CGIAR, and by program-restricted grants to ILRI from the EU and Department for International Development (DfID), UK. Helpful advice was generously provided by Prof. A.B. Korol and Prof. Jan Bradley.

Author information



Corresponding author

Correspondence to Jerzy M. Behnke.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Behnke, J.M., Iraqi, F.A., Mugambi, J.M. et al. High resolution mapping of chromosomal regions controlling resistance to gastrointestinal nematode infections in an advanced intercross line of mice. Mamm Genome 17, 584–597 (2006). https://doi.org/10.1007/s00335-005-0174-0

Download citation


  • Quantitative Trait Locus
  • Quantitative Trait Locus Effect
  • Significant Quantitative Trait Locus
  • Quantitative Trait Locus Location
  • Single Quantitative Trait Locus