Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Immunol. Author manuscript; available in PMC 2010 July 1.
Published in final edited form as:
PMCID: PMC2735236

Identification of a major susceptibility locus for lethal graft-versus-host disease in MHC-matched mice


Graft-versus-host disease (GVHD) is the major cause of morbidity and mortality after allogeneic hematopoietic cell transplantation. From a genetic perspective, GVHD is a complex phenotypic trait. While it is understood that susceptibility results from interacting polymorphisms of genes encoding histocompatibility antigens and immune regulatory molecules, a detailed and integrative understanding of the genetic background underlying GVHD remains lacking. To gain insight regarding these issues, we performed a forward genetic study. A MHC-matched mouse model was utilized in which irradiated recipient BALB.K and B10.BR mice demonstrate differential susceptibility to lethal GHVD when transplanted using AKR/J donors. Assessment of GVHD in (B10.BR x BALB.K)F1 mice revealed that susceptibility is a dominant trait and conferred by deleterious alleles from the BALB.K strain. In order to identify the alleles responsible GVHD susceptibility, a genome scanning approach was taken using (B10.BR x BALB.K)F1 × B10.BR backcross mice as recipients. A major susceptibility locus, termed the Gvh1 locus, was identified on chromosome 16 using linkage analysis (LOD = 9.1). A second locus was found on chromosome 13, named Gvh2, which had additive but protective effects. Further identification of Gvh genes by positional cloning may yield new insight into genetic control mechanisms regulating GVHD and potentially reveal novel approaches for effective GVHD therapy.

Keywords: Transplantation, Graft Versus Host Disease, Rodent


In a broadly accepted pathophysiologic model, graft-versus-host disease (GVHD) occurs as a complication of allogeneic hematopoietic cell transplantation through a three-step process (1). First, recipient conditioning leads to tissue injury and inflammation. These events trigger, in the second phase, activation of tissue-resident host antigen presenting cells which efficiently present alloantigen to donor T-cells (2, 3). Antigen-activated donor T-cells consequently undergo differentiation and expansion and, in the final phase, recruit additional effector cells to cause tissue injury via a combination of cellular and humoral inflammatory mechanisms.

This general framework of GVHD pathophysiology can be viewed, from a genetic perspective, as a complex phenotypic trait with a polygenic basis. Genetic susceptibility to GVHD is firstly conferred by deleterious alleles encoding histocompatibility antigens (HAg) that are disparate between donors and recipients. Allele effects of major HAg, encoded by classical major histocompatibility complex (MHC) genes (4), and minor HAg, arising from endogenously expressed polymorphic autosomal gene products (5, 6), are the antigenic basis for allogeneic donor T-cell activation. Whether or not mismatched HAg becomes functionally expressed as GVHD is contingent upon further interaction with polymorphisms of immune regulatory modifier genes. Examples of modifier gene effects include allelic variants of cytokines and cytokine receptors, whose protein products either dampen or augment graft-versus-host alloimmune responses (7).

While many of these genes and gene effects are well characterized, a detailed understanding of genetic control mechanisms underlying GVHD remains lacking. Further, novel molecular targets for effective GVHD therapy are still needed. To address these challenges, we initiated a forward genetic screen in a murine model of severe GVHD. In this model, irradiated BALB.K mice are susceptible whereas B10.BR mice are resistant to lethal GVHD when transplanted using hematopoietic cells from MHC-identical AKR/J donors (8). Using (B10.BR × BALB.K)F1 mice as transplant recipients, we found that lethal GVHD in these mice is a dominant gene effect with susceptibility conferred by deleterious alleles from the BALB.K strain. A backcross generated between the B10.BR parental strain and (B10.BR × BALB.K)F1 was then bred for genome scanning using linkage analysis. A major susceptibility locus on chromosome 16 was identified which we have termed the Gvh1 locus. Gvh1 appears to regulate the development of lethal as opposed to mild and clinically insignificant GVHD. Underscoring the complex regulatory mechanisms controlling GVHD, a second locus with additive but protective BALB.K allele effects was identified on chromosome 13, termed Gvh2. These results lay the groundwork for positional cloning of Gvh genes and gene discovery as a prerequisite to developing new methods for predicting, preventing or treating GVHD.


Mouse strains and crosses

Mice were bred and maintained at the Stanford University Research Animal Facility. Hematopoietic cell donor AKR/J (H2k, Thy1.1) mice were 6 – 10 weeks of age at bone marrow (BM) harvest. Recipient MHC congenic BALB.K (H2k, Thy1.2) and B10.BR (H2k, Thy1.2) mice were greater than 8 weeks of age at the time of transplantation. BALB.K male and B10.BR female mice were mated to generate a (B10.BR × BALB.K)F1 population. F1 mice were subsequently mated with females of the B10.BR parental strain to produce [B10.BR × (B10.BR × BALB.K)F1] backcross (BC) mice for linkage analysis. All work using animals was reviewed and approved by the Administrative Panel on Laboratory Animal Care at the Stanford University School of Medicine.

T-cell isolation

Donor T-cells were isolated from the spleen of AKR/J mice by micromagnetic bead separation using methods modified from procedures previously described (9). For pan T-cell isolation, splenocytes were labeled with phycoerythrin-conjugated anti-B220 (6B2), anti-Mac1 (M1/70), anti-DX5 and anti-Ter119 monoclonal antibodies (BD Biosciences, San Jose, CA). This was followed by second-stage labeling with anti-phycoerythrin MicroBeads for negative selection on a MidiMACS column (Miltenyi Biotech, Auburn, CA). The resulting unbound CD3+B220/Mac1/DX5/Ter119 T-cells were > 99% pure as determined by FACS analysis. For isolating T-cell subsets from spleen, phycoerythrin-conjugated anti-CD4 (GK1.5) and anti-CD8 (53-6.7) were added, respectively, to the antibody labeling cocktail for isolating CD8 (CD4CD8+) and CD4 (CD4+CD8)T-cell populations, again by negative selection.

GVHD phenotype analysis

BM transplantation and GVHD analyses were performed as previously described (8). Recipient mice were conditioned with a lethal dose of total body irradiation (TBI) delivered in 2 split fractions on day 0. A shielded room Phillips Unit Irradiator providing X-ray dose outputs of 250kV and 15mA was used for irradiation. The TBI dose was 800 cGy for BALB.K and 900 cGy for B10.BR, (B10.BR × BALB.K)F1 and BC mice. After irradiation, mice were injected with 2.5 × 106 BM cells plus defined doses of purified splenic T-cells or T-cell subsets. Survival, weight loss, and clinical signs of GVHD were monitored daily for 60 – 100 days after transplantation. T-cell chimerism in surviving mice was determined by peripheral blood FACS analysis for Thy1.1 (OX-7) and Thy1.2 (53.21) expression. Selected mice were euthanized prior to death or at the end of experiments for histopathologic analysis of liver, gut and skin where indicated.


Genomic DNA was isolated from tail tips of BC and control mice using the DNeasy Tissue Kit according to vendor instructions (Qiagen, Valencia, CA). Genotypes were performed by PCR amplification of 90 informative microsatellite markers using standard PCR conditions and cycling parameters (10). The markers spanned all 19 autosomal chromosomes and a list of microsatellite markers and genetic map position data is provided as Supplemental Table 1. The average internal spacing between markers was 13.0 cM. Proximal and distal markers for each chromosome were anchored within an average of 5.7 cM from chromosome ends. Marker position and order assignment were based on the Mouse Genome Database and accessed via the Jackson Laboratory Mouse Genome Informatics online resource ( (11). PCR amplicons were stained with ethidium bromide for size separation and allele determination on 4% agarose gels. Where necessary, markers were evaluated in duplicate until genotypes at all loci were determined for all BC mice to allow linkage analysis with no missing marker data.

Statistical and linkage analysis

A 90-marker genome scan was completed by genotyping 180 BC mice that were phenotyped for lethal GVHD. Linkage analysis was performed using R/qtl version 1.05–2, an add-on package to the R general statistical package (12). GVHD susceptibility was analyzed as a binary trait: surviving mice were scored as 0 and non-surviving mice as 1. Simple interval mapping was performed using the EM algorithm to test for maximum likelihood of single locus effects on a 1 cM grid along the genome (13). Single-locus effects were further evaluated by composite interval mapping implemented in R/qtl by performing the genome scan with the inclusion of significantly linked background loci as additive and interactive covariates. Background markers were chosen at the location of the maximum logarithm of the odds (LOD) score calculated by simple interval mapping. Hazard ratios for the GVHD phenotype according to genotype at linked loci was calculated using Cox proportional hazards regression with survival time as the dependent variable.

Genome-wide significance thresholds were determined by empirical permutation testing using 1,000 permutation replicates (14). Standard genome-wide p values to define suggestive (p < .63), significant (p < .05) and highly significant (p < .001) threshold levels of LOD scores for linkage were applied (15). For our backcross this threshold corresponded to LOD scores of 1.4, 2.7 and 4.4, respectively. Approximate confidence intervals for the locations of linked loci were obtained using the 2.0-LOD drop-off method (16).


Variable GVHD in H2k-matched mice

We previously reported a mouse model of allogeneic hematopoietic cell transplantation that uses a single donor mouse strain, AKR/J, and two MHC-congenic recipients, BALB.K and B10.BR (8). In the prior studies, GVHD was induced by co-transferring purified hematopoietic stem cells and unseparated donor splenocytes into irradiated recipients. Because mapping susceptibility to a small genomic interval requires a large number of mice, this experimental protocol was modified to permit high-throughput GVHD phenotypic evaluation for linkage analysis. Thus, hematopoietic stem cells, which require a rigorous 2-step isolation procedure for purification, was replaced with BM. In addition, splenic T-cells isolated by micromagnetic bead separation were utilized in place of whole splenocytes. As shown in Figure 1A, BALB.K mice conditioned with a lethal dose of whole body irradiation and injected with AKR/J BM along with either of two doses of T-cells developed aggressive and lethal GVHD, consistent with our prior studies. Median survival time following transplantation was 9 days. Prior to death, all BALB.K mice displayed clinical features of GVHD including bloody diarrhea, weight loss, ruffled fur and hunched posture. In contrast, similar AKR/J ➔ B10.BR transplants resulted in no mortality as survival of B10.BR mice given donor BM plus T-cells did not differ from control mice given BM alone (Figure 1B).

Figure 1Figure 1
Variable graft-versus-host disease (GVHD) in MHC-matched AKR/J →BALB.K and AKR/J → B10.BR mice

Further consistent with our previous results was the observation that, while not associated with lethality, AKR/J ➔ B10.BR transplants using BM and splenic T-cells resulted in detectable mild GVHD. Clinically, this GVHD syndrome was manifested by minimal chronic weight loss in almost all recipients (data not shown). No overt skin lesions, diarrhea nor dysmotility was observed. Further evidence of mild GVHD was the finding that B10.BR mice given BM plus splenic T-cells engrafted with full donor T-cell chimerism, rather than mixed T-cell chimerism as was observed when given BM alone (Figure 1C). Lastly, histologic examination of B10.BR mice at day +60 after transplant revealed low-grade GVHD pathology restricted to the liver and not present in the skin, ileum or colon (Figure 1D). By comparison, BALB.K mice sacrificed early in the transplant course at day +5 prior to death exhibited severe GVHD pathology in both the colon and liver. No histopathologic abnormalities were seen in B10.BR mice sacrificed at the day +5 time-point.

Depending on the strain combination, GVHD mortality in MHC-identical, minor HAg-mismatched mice can be mediated by either CD4+ or CD8+ T-cells alone, by both in combination with synergistic effects, or not at all regardless of the graft cell composition (17). We characterized our GVHD model in this regards by depleting splenic T-cell subsets to produce CD4 (CD4+CD8) and CD8 (CD4CD8+) T-cell populations for transplant experiments. These studies showed that severe GVHD in AKR/J ➔ BALB.K transplants was mediated primarily by the donor CD4+ T-cell subset (Figure 2A). Lethality following co-injection of BM and CD4+CD8 T-cells was rapid and uniform. Only a weak effect was seen with CD8+ T-cells, reflected by a small proportion of mice dying after transplants of donor BM plus CD4CD8+ T-cells. Lethal GVHD could not be induced in AKR/J ➔ B10.BR transplants with either CD4+ or CD8+ T-cell subsets (Figure 2B). Only late-onset liver GVHD histopathology was detected in all B10.BR mice as well as surviving BALB.K mice given CD8+ T-cells sacrificed prior to the end of each experiment (Supplemental Figure 1). Taken together, these results show that the GVHD phenotype in AKR/J ➔ BALB.K and AKR/J ➔ B10.BR mice closely resemble other established models of murine GVHD and was thus suitable for genetic studies.

Figure 2
T-cell subsets mediating GVHD in AKR/J → BALB.K and AKR/J → B10.BR mice

Lethal GVHD susceptibility is a dominant trait

The most common strategy for performing a genome scan in mice involves breeding random genetic recombinants from parental strains that differ with regard to a trait of interest, followed by statistical analysis to identify chromosomal regions that segregate with the phenotype and are thus shared among affected individuals (18). To facilitate design of the experimental cross, we began by generating (B10.BR × BALB.K)F1 littermates to determine if GVHD is the result of dominance or additive effects, and to assess the directionality of the allele effect. As shown in Figure 3, irradiated F1 mice transplanted with AKR/J BM and donor T-cells developed rapidly aggressive GVHD with typical clinical features and uniform lethality. Only a transient latency in median survival time as compared with BALB.K parents was observed. These results show that inheritance of lethal GVHD susceptibility is governed by a dominant gene effect and susceptibility is conferred by deleterious alleles from the BALB.K strain.

Figure 3
Susceptibility to lethal GVHD is a dominant trait in AKR/J → BALB.K mice

GVHD susceptibility in a backcross

Having determined the genetic model and allele effect we next generated a [B10.BR × (B10.BR × BALB.K)F1] backcross. Detection of dominance effects is more efficient with a backcross, requiring about half the progeny size of an F2 intercross population because of lower background genetic variance (18). The backcross was made to B10.BR parents, as a segregating locus in a backcross to the BALB.K strain would not contribute to phenotypic variance. We generated 180 (F1 × B10.BR) BC littermates and the GVHD phenotype for these mice is shown in Figure 4. The mice were divided into 7 groups for lethal irradiation and injections of sex-matched AKR/J BM and T-cells, each performed as individual experiments. When survival outcome for all 180 BC mice were pooled, 28 mice (16%) were found to have a phenotype similar to the BALB.K strain and died prior to day 20 post-transplant with clinical signs characteristic of GVHD. When follow-up was extended to 100 days post-transplantation, an intermediate phenotype emerged whereby 12 additional BC mice died at time points indicated by the survival curve. The cumulative mortality prior to day 100 was thus 40 mice (22%). BC mice surviving greater than 100 days were considered to have survived the transplant without lethal GVHD. Similar to the resistant B10.BR strain, complete donor peripheral blood T-cell engraftment was seen in all surviving BC mice by FACS analysis (Supplemental Figure 2A). Surviving BC mice further resembled B10.BR parents in that variable liver abnormalities consistent with GVHD was detected in representative mice evaluated by histology (Supplemental Figure 2B).

Figure 4
Lethal GVHD susceptibility in [B10.BR × (B10.BR × BALB.K)F1] backcross (BC) mice

Interval mapping for lethal GVHD

Individual BC mice were numbered and genomic DNA was isolated from tail tip sections prior to irradiation for a 90-marker genome-wide scan involving all 180 BC mice. Genotyping for markers was performed in duplicate as necessary until 100% of mice were genotyped at all makers for a total of 16,200 genotypes. The genotype distribution was 49.4% B10.BR homozygous and 50.6% B10.BR/BALB.K heterozygous, not significantly different from the expected 50:50 distribution. We first evaluated for genotyping errors according to the method of Lincoln and Lander (19). Using the Haldane map function to convert genetic distances into recombination fraction and an assumed genotyping error rate of 0.01, we found no markers with significant error scores to suggest genotyping error. As a further test of genotype data integrity, we performed mock linkage analysis for coat color. BALB.K mice are albino and B10.BR mice have black coat color. BC mice are either black or brown. Linkage analysis for coat color in the 180 BC mice identified precise localization to the agouti locus on chromosome 2 (data not shown), which regulates coat color variation in these mice (20).

We then proceeded with genome-wide linkage analysis for lethal GVHD by interval mapping. We considered but ultimately rejected evaluating for linkage using survival days as a quantitative trait because nearly 80% of BC mice survived longer than 100 days, a clear departure from the standard assumption of normal distribution for interval mapping. As shown in Supplemental Figure 3, a log10 transformation of the survival time failed to resolve the skewed phenotype distribution. Therefore lethal GVHD was analyzed as a binary trait: surviving mice were scored as 0 and dying mice as 1. Separate analyses were performed for mice with survival < 20 days (n = 28 affected) or survival < 100 days (n = 40 affected, cumulatively). Linkage analysis was performed by interval mapping using the EM algorithm to test for maximum likelihood of single locus effects on a 1 cM grid along the genome. This procedure allows localization of genomic regions linked to the phenotype by analyzing coinheritance of genetic markers and phenotype. As shown in Figure 5, a single highly significant (p < .001) lethal GVHD susceptibility locus on chromosome 16 was found for both survival < 20 days and survival < 100 days. A minor peak associated with markers on chromosome 5 met suggestive but not significant thresholds for survival < 20 days. Another minor peak was found at a locus on chromosome 13 that met suggestive, but again not significant, thresholds for survival < 100 days. No suggestive linkage was identified elsewhere throughout the genome. To obtain additional evidence for linkage, we applied a second mapping analysis using a different statistical test. We performed individual marker regression implemented by the Map Manager QTX package (21) and found an identical result, which was a single highly significantly linked locus on chromosome 16 (Supplemental Table 2). The additive statistic for this locus was a positive value (0.37), confirming that susceptibility was conferred by BALB.K alleles.

Figure 5
Genome-wide scan for lethal GVHD reveals a highly significant susceptibility locus on chromosome 16

All BC mice were genotyped for 10 additional markers along chromosome 16 and interval mapping was repeated. Results are shown in Figure 6. For survival < 100 days, a broad and complex linkage pattern was observed with an LOD peak of 9.1 at map position 29 cM, flanked by markers D16Mit4 and D16Mit138. This linkage was followed by a secondary peak with a LOD score of 8.1 at position 52.2 cM, proximal to D16Mit189. For survival < 20 days, the LOD peak was marginally offset in the centromeric direction at position 28.0 cM and the peak LOD score was slightly higher at 9.4. The 2 LOD confidence interval for survival < 100 days, extending from 17.0 cM to 57.6 cM, was used as boundaries for a new GVHD susceptibility locus which we term the Gvh1 locus. Numerous mouse histocompatibility H loci have been mapped to autosomal chromosomes, as shown in Figure 8A. These gene regions include the H2 locus on chromosome 17and the well characterized H60 locus on chromosome 10 (22). The Gvh1 locus is unique among these for being located on chromosome 16. The 2 LOD confidence interval for the Gvh1 locus extends from 23.6 – 86.1 Mb on the physical map, as shown in Figure 8B.

Figure 6
Interval mapping for lethal GVHD on chromosome 16
Figure 8
Physical location of the Gvh1 locus

Composite interval mapping for lethal GVHD

One explanation for the wide confidence interval for the Gvh1 locus as that more than one adjacent susceptibility loci was located on the same chromosome. To test for this possibility and to identify additional loci, a form of composite interval mapping was implemented by including D16Mit4 marker genotype as an additive cofactor for linkage analysis. This model assumes that the effect of a putative trait locus is not dependent on D16Mit4 marker genotype and allows detection of background loci with weak main effects. D16Mit4 at map position 27.3 cM on chromosome 16 was the individual marker with highest LOD score. As shown in Figure 7, composite interval mapping controlling for additive D16Mit4 effects revealed that distal markers on chromosome 16 had a peak LOD score of 1.9 at position 55.2 cM. This finding indicates suggestive linkage for a second GVHD susceptibility locus on chromosome 16, but the significance threshold was not reached. Thus, whether or not more than one susceptibility loci is present on chromosome 16 could not be resolved with the current backcross. Interestingly, the chromosome 13 locus became significantly linked when all non-surviving mice were scored as affected, including those with death between days 20 – 100, suggesting that this locus may have contributed to the intermediate phenotype. The peak LOD score was 3.4 and, in contrast to the Gvh1 locus, protective rather than deleterious effects were conferred by BALB.K alleles at the chromosome 13 locus. That is, BC mice homozygous for B10.BR alleles at this locus had higher susceptibility than mice that were heterozygous with both B10.BR and BALB.K alleles. This locus is designated Gvh2. No evidence for epistatic interactions between this chromosome 13 locus and Gvh1 was observed when composite interval mapping included D16Mit4 marker genotype as an interactive covariate, or when BC mice were partitioned into two groups according to D16Mit4 genotype for simple interval mapping (Supplemental Figure 4).

Figure 7
Composite interval mapping for lethal GVHD

Regression analysis of Gvh1 and Gvh2

Cox regression was used to calculate hazard ratios for lethal GVHD in BC mice according to their genotypes at D16Mit4 and D13Mit248. These markers were, respectively, the individual markers with peak LOD value for Gvh1 and Gvh2. As shown in Table 1, we considered a proportional hazard model where baseline risk was conferred by the presence of homozygous B10.BR alleles at both loci. Compared with this reference group, BC mice with a deleterious BALB.K allele at Gvh1 had a 20-fold risk of death prior to day +100 post-transplantation (HR 20.3, 95% CI: 4.8 – 85.3, P < .001) if Gvh2 remained B10.BR homozygous. The risk of lethal GVHD was reduced to 6-fold, however, if a BALB.K allele at Gvh1 was also accompanied by heterozygosity at Gvh2 (HR 6.3, 95% CI: 1.4 – 29.3, P = .02), where the presence of a BALB.K allele was protective.

Table 1
Regression analysis of Gvh genotype and lethal GVHD


The overall picture of susceptibility that emerged from our genome scan is consistent with the idea that severe GVHD is a complex trait, since no one-to-one gene to phenotype relationship was identified in our mice. That said, major susceptibility was conferred by deleterious alleles from the BALB.K background with strong effect at a single locus on chromosome 16, the Gvh1 locus. Genetic control of GVHD was influenced by a second locus on chromosome 13, the Gvh2 locus, which acted independently of Gvh1 and contributed an additive but opposite allele effect. The Gvh2 locus was not significant when mice dying between days 20 – 100 were excluded from analysis, suggesting a contribution to the intermediate phenotype of delayed GVHD lethality. Lastly, the fully susceptible BALB.K parental strain by itself expresses both deleterious and protective alleles from Gvh1 and Gvh2, respectively. This is likely indicative of additional background gene effects not detected with the current genome scan.

In this study, we performed a classical linkage analysis for lethal GVHD involving an experimental backcross in a mouse model. Data from linkage analyses and recent advances in mouse genome informatics, congenic and in silico haplotype mapping techniques, and high-throughput methodologies, such as gene expression profiling, have allowed accelerated positional cloning of many genes underlying susceptibility loci that are responsible for important physiologic traits in mice. These include the cloning of latexin as a regulator of hematopoietic stem cell size, histamine receptor H1 as a mediator of T-cell responses in autoimmune disease, complement factor 5 as a modifier of liver fibrosis, and the serine-threonine kinase ROP18 as a key host effector mechanism controlling Toxoplasma gondii virulence (2326). It is our anticipation that similar strategies may allow identification of Gvh genes that will be clinically useful in the management of GVHD complications.

Forward genetic approaches for finding GVHD susceptibility genes in mouse models have otherwise been sparingly used. As of this writing, the only previously published series of genetic linkage analyses investigated the trait variance in GVHD that results from alternatively using C57BL/6 versus DBA/2 parents as donors for transplants into (C57BL/6 × DBA/2)F1 recipients (2730). In this model, transfer of C57BL/6 lymphoid cells result in development of acute GVHD, whereas transfer of DBA/2 lymphoid cells result in chronic GVHD with features of weight loss, auto-antibody production, and nephritis. Several quantitative trait loci linked to chronic, rather than acute, GVHD were identified that mapped to chromosomes 1, 2, 4, and 17. The current report adds to these efforts by identifying novel loci on chromosomes 13 and 16 with major effects in conferring susceptibility to severe acute GVHD. Further, the mice used in our studies are MHC-matched and minor HAg-mismatched, a genetic combination more similar to clinical allogeneic hematopoietic cell transplantation.

This major susceptibility locus, Gvh1, was associated with a complex linkage pattern and a wide confidence interval that spanned a large segment of chromosome 16. While it is possible that closely linked allelic genes on the same chromosome account for this broad linkage pattern, we found that Gvh1 could not be fractionated into more than one detectable sublocus with the current mouse cross. In this regards, prior efforts describing congenic mapping of quantitative trait loci are insightful for the diverse genetic control mechanisms that may be uncovered. A single trait locus can in fact map to a single chromosomal segment and provide significant refinement of a confidence interval defined by standard linkage analysis (31). More often, however, two or even three separately linked subloci are found to underlie the single linkage-derived trait locus (3234). Alternatively, a trait locus may be revealed to encompass multiple discrete gene effects with additive as well as epistatic elements, or two separate loci with additive but negating opposite allelic effects (35, 36). These and other possible outcomes can be envisioned when the Gvh1 locus is further interrogated by fine mapping.

Based on current understanding of GVHD pathophysiology, we hypothesize that strain-specific polymorphisms of Gvh genes encoding immnodominant minor HAg and/or immune regulatory molecules are responsible for causing lethal GVHD in our model. We may eventually find a minor HAg-mediated basis underlying Gvh gene effects through positional cloning of susceptibility loci linked to lethal GVHD in our model. Because MHC-restriction for murine minor HAg have no direct human MHC peptide binding equivalent, more clinically relevant may be discovering Gvh genes encoding novel immune regulatory molecules with a human homologue. The significance of important modifier genes that interact with mismatched major or minor HAg to influence GVHD is best illustrated by studies of cytokine polymorphisms. In humans, numerous studies have now implicated significant associations between GVHD and polymorphisms of genes for the IL-1, IL-2, IL-4, IL-6, IL-10, IL-18, IFN-γ, TGF-β, and TNF-α cytokines and, in some instances, for their respective receptors as well (7). Tissue injury resulting from recipient conditioning, particularly in the gastrointestinal tract, is a major source of inflammatory cytokines in the immediate post-transplant period (37). Mouse strains differ in their sensitivity to irradiation (38), thus it is possible that variable response to TBI conditioning may contribute to GVHD susceptibility as well.

In the context of these potential Gvh gene characteristics, we note that the 2 LOD confidence interval for the Gvh1 locus extends from 23.6 – 86.1 Mb on the physical map, as shown in Figure 8B. A discussion of genes of interest is premature but the T-cell co-stimulatory molecules CD80 and CD86 merit highlighting. Polymorphisms of the CD86 gene in humans have been characterized and emerging data suggests a functional effect by these variants on transplantation tolerance and allergic and autoimmune disease (3941). The αvβ5 integrin is expressed in human dendritic cells and is important for antigen cross presentation (42). Stefin A1, A2 and A3 (stfa1, stfa2, and stfa3) inhibit cysteine endo- and exopeptidases important in antigen processing, such as cathepsin L and S, and may influence mHAg processing (43). Specific cathepsins have been shown to be important for initiating autoimmune disease, such as diabetes in the NOD model (44), and pharmacologic inhibitors may be therapeutically promising in those settings (45). The CD200 surface antigen and receptor system is active in regulating immune responses related to allergic and autoimmunity conditions (46). The GA binding protein (GABP) is essential for the regulation of IL-7 receptor expression in T cells (47). Previously mapped within this region are quantitative trait loci for autoimmune ovarian dysgenesis (Aod1), collagen-induced arthritis (Lp1), Leishmania susceptibility (Lmr12), IL-4 and IL-10 production (Cypr1), and experimental allergic encephalomyelitis (Eae11) (4852).

We present here a unique mouse model of GVHD where the genetic basis may be dissected to the level of gene identification by positional cloning of linked loci found in this study. Successful identification of a Gvh minor HAg would yield new insight into the biology of GVHD and unveil a snapshot of the genetic architecture underlying this complex trait. Alternatively, identification of a novel Gvh immune regulatory molecule with a human homolog could directly lead to testing of clinical hypotheses and possibly new therapies.

Supplementary Material



We gratefully acknowledge Lucino Hidalgo for excellent care of our mouse colony and Dr. Guidot Tricot for critique of the manuscript.

SUPPORT: Supported in part by grants PO1CA049605 and R01HL087240 (J.A.S.), and K08HL067847 (T.M.C.) from the National Institutes of Health.


1. Ferrara JL. Pathogenesis of acute graft-versus-host disease: cytokines and cellular effectors. Journal of hematotherapy & stem cell research. 2000;9:299–306. [PubMed]
2. Korngold R, Sprent J. Lethal GVHD across minor histocompatibility barriers: nature of the effector cells and role of the H-2 complex. Immunological reviews. 1983;71:5–29. [PubMed]
3. Shlomchik WD, Couzens MS, Tang CB, McNiff J, Robert ME, Liu J, Shlomchik MJ, Emerson SG. Prevention of graft versus host disease by inactivation of host antigen-presenting cells. Science. 1999;285:412–415. [PubMed]
4. Klein J. Natural History of the Histocompatibility Complex. New York, NY: John Wiley & Sons; 1986.
5. Wallny HJ, Rammensee HG. Identification of classical minor histocompatibility antigen as cell-derived peptide. Nature. 1990;343:275–278. [PubMed]
6. Falk K, Rotzschke O, Rammensee HG. Cellular peptide composition governed by major histocompatibility complex class I molecules. Nature. 1990;348:248–251. [PubMed]
7. Mullally A, Ritz J. Beyond HLA: the significance of genomic variation for allogeneic hematopoietic stem cell transplantation. Blood. 2007;109:1355–1362. [PubMed]
8. Cao TM, Lo B, Ranheim EA, Grumet FC, Shizuru JA. Variable hematopoietic graft rejection and graft-versus-host disease in MHC-matched strains of mice. Proceedings of the National Academy of Sciences of the United States of America. 2003;100:11571–11576. [PubMed]
9. Yu P, Lee Y, Liu W, Chin RK, Wang J, Wang Y, Schietinger A, Philip M, Schreiber H, Fu YX. Priming of naive T cells inside tumors leads to eradication of established tumors. Nature immunology. 2004;5:141–149. [PubMed]
10. Dietrich W, Katz H, Lincoln SE, Shin HS, Friedman J, Dracopoli NC, Lander ES. A genetic map of the mouse suitable for typing intraspecific crosses. Genetics. 1992;131:423–447. [PubMed]
11. Eppig JT, Bult CJ, Kadin JA, Richardson JE, Blake JA, Anagnostopoulos A, Baldarelli RM, Baya M, Beal JS, Bello SM, Boddy WJ, Bradt DW, Burkart DL, Butler NE, Campbell J, Cassell MA, Corbani LE, Cousins SL, Dahmen DJ, Dene H, Diehl AD, Drabkin HJ, Frazer KS, Frost P, Glass LH, Goldsmith CW, Grant PL, Lennon-Pierce M, Lewis J, Lu I, Maltais LJ, McAndrews-Hill M, McClellan L, Miers DB, Miller LA, Ni L, Ormsby JE, Qi D, Reddy TB, Reed DJ, Richards-Smith B, Shaw DR, Sinclair R, Smith CL, Szauter P, Walker MB, Walton DO, Washburn LL, Witham IT, Zhu Y. The Mouse Genome Database (MGD): from genes to mice--a community resource for mouse biology. Nucleic acids research. 2005;33:D471–D475. [PMC free article] [PubMed]
12. Broman KW, Wu H, Sen S, Churchill GA. R/qtl: QTL mapping in experimental crosses. Bioinformatics. 2003;19:889–890. [PubMed]
13. Lander ES, Botstein D. Mapping mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics. 1989;121:185–199. [PubMed]
14. Churchill GA, Doerge RW. Empirical threshold values for quantitative trait mapping. Genetics. 1994;138:963–971. [PubMed]
15. Lander E, Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nature genetics. 1995;11:241–247. [PubMed]
16. Van Ooigen JW. Accuracy of mapping quantitative trait loci in autogamous species. Theor Appl Genet. 1992;84:803–811. [PubMed]
17. Korngold R, Sprent J. T cell subsets in graft-versus-host disease. In: Burakoff SJ, Deeg HJ, Ferrara JL, editors. Graft-vs.-Host Disease: Immunology, Pathophysiology, and Treatment. New York, NY: Dekker; 1990. pp. 31–50.
18. Darvasi A. Experimental strategies for the genetic dissection of complex traits in animal models. Nature genetics. 1998;18:19–24. [PubMed]
19. Lincoln SE, Lander ES. Systematic detection of errors in genetic linkage data. Genomics. 1992;14:604–610. [PubMed]
20. Silver LM. Mouse Genetics: Concepts and Applications. New York, NY: Oxford Unviersity Press; 1995.
21. Manly KF, Cudmore RH, Jr., Meer JM. Map Manager QTX, cross-platform software for genetic mapping. Mamm Genome. 2001;12:930–932. [PubMed]
22. Malarkannan S, Shih PP, Eden PA, Horng T, Zuberi AR, Christianson G, Roopenian D, Shastri N. The molecular and functional characterization of a dominant minor H antigen, H60. J Immunol. 1998;161:3501–3509. [PubMed]
23. Liang Y, Jansen M, Aronow B, Geiger H, Van Zant G. The quantitative trait gene latexin influences the size of the hematopoietic stem cell population in mice. Nature genetics. 2007 [PubMed]
24. Ma RZJ, Gao N, Meeker D, Fillmore PD, Tung KS, Watanabe T, Zachary JF, Offner H, Blankenhorn EP, Teuscher C. Identification of Bphs, an autoimmune disease locus, as histamine receptor H1. Science. 2002;297:620–623. [PubMed]
25. Hillebrandt S, Wasmuth HE, Weiskirchen R, Hellerbrand C, Keppeler H, Werth A, Schirin-Sokhan R, Wilkens G, Geier A, Lorenzen J, Kohl J, Gressner AM, Matern S, Lammert F. Complement factor 5 is a quantitative trait gene that modifies liver fibrogenesis in mice and humans. Nature genetics. 2005;37:835–843. [PubMed]
26. Taylor S, Barragan A, Su C, Fux B, Fentress SJ, Tang K, Beatty WL, Hajj HE, Jerome M, Behnke MS, White M, Wootton JC, Sibley LD. A secreted serine-threonine kinase determines virulence in the eukaryotic pathogen Toxoplasma gondii. Science. 2006;314:1776–1780. [PubMed]
27. Fast LD. Identification of a single non-H-2 gene regulating graft-versus-host disease response. J Immunol. 1990;144:4177–4182. [PubMed]
28. Allen RD, Dobkins JA, Harper JM, Slayback DL. Genetics of graft-versus-host disease, I. A locus on chromosome 1 influences development of acute graft-versus-host disease in a major histocompatibility complex mismatched murine model. Immunology. 1999;96:254–261. [PubMed]
29. Harper JM, Slayback DL, Dobkins JA, Allen RD. A locus on chromosome 2 influences the development of acute graft-versus-host disease in a murine model. Bone marrow transplantation. 1999;23:1183–1190. [PubMed]
30. Slayback DL, Dobkins JA, Harper JM, Allen RD. Genetic factors influencing the development of chronic graft-versus-host disease in a murine model. Bone marrow transplantation. 2000;26:931–938. [PubMed]
31. Nathan BM, Hodges CA, Supelak PJ, Burrage LC, Nadeau JH, Palmert MR. A quantitative trait locus on chromosome 6 regulates the onset of puberty in mice. Endocrinology. 2006;147:5132–5138. [PubMed]
32. McAllister RD, Singh Y, du Bois WD, Potter M, Boehm T, Meeker ND, Fillmore PD, Anderson LM, Poynter ME, Teuscher C. Susceptibility to anthrax lethal toxin is controlled by three linked quantitative trait loci. The American journal of pathology. 2003;163:1735–1741. [PubMed]
33. Baguet A, Epler J, Wen KW, Bix M. A Leishmania major response locus identified by interval-specific congenic mapping of a T helper type 2 cell bias-controlling quantitative trait locus. The Journal of experimental medicine. 2004;200:1605–1612. [PMC free article] [PubMed]
34. Lin E, Pappenfuss T, Tan RB, Senyschyn D, Bahlo M, Speed TP, Foote SJ. Mapping of the Plasmodium chabaudi resistance locus char2. Infection and immunity. 2006;74:5814–5819. [PMC free article] [PubMed]
35. Youngren KK, Nadeau JH, Matin A. Testicular cancer susceptibility in the 129.MOLF-Chr19 mouse strain: additive effects, gene interactions and epigenetic modifications. Human molecular genetics. 2003;12:389–398. [PubMed]
36. Roper RJ, McAllister RD, Biggins JE, Michael SD, Min SH, Tung KS, Call SB, Gao J, Teuscher C. Aod1 controlling day 3 thymectomy-induced autoimmune ovarian dysgenesis in mice encompasses two linked quantitative trait loci with opposing allelic effects on disease susceptibility. J Immunol. 2003;170:5886–5891. [PubMed]
37. Hill GR, Ferrara JL. The primacy of the gastrointestinal tract as a target organ of acute graft-versus-host disease: rationale for the use of cytokine shields in allogeneic bone marrow transplantation. Blood. 2000;95:2754–2759. [PubMed]
38. Roderick TH. The Response of Twenty-Seven Inbred Strains of Mice to Daily Doses of Whole-Body X-Irradiation. Radiat Res. 1963;20:631–639. [PubMed]
39. Marin LA, Moya-Quiles MR, Miras M, Muro M, Minguela A, Bermejo J, Ramirez P, Garcia-Alonso AM, Parrilla P, Alvarez-Lopez MR. Evaluation of CD86 gene polymorphism at +1057 position in liver transplant recipients. Transpl Immunol. 2005;15:69–74. [PubMed]
40. Abdallah AM, Renzoni EA, Anevlavis S, Lagan AL, Munkonge FM, Fonseca C, Black CM, Briggs D, Wells AU, Marshall SE, McHugh N, du Bois RM, Welsh KI. A polymorphism in the promoter region of the CD86 (B7.2) gene is associated with systemic sclerosis. Int J Immunogenet. 2006;33:155–161. [PubMed]
41. Corydon TJ, Haagerup A, Jensen TG, Binderup HG, Petersen MS, Kaltoft K, Vestbo J, Kruse TA, Borglum AD. A functional CD86 polymorphism associated with asthma and related allergic disorders. J Med Genet. 2007;44:509–515. [PMC free article] [PubMed]
42. Albert ML, Pearce SF, Francisco LM, Sauter B, Roy P, Silverstein RL, Bhardwaj N. Immature dendritic cells phagocytose apoptotic cells via alphavbeta5 and CD36, and cross-present antigens to cytotoxic T lymphocytes. The Journal of experimental medicine. 1998;188:1359–1368. [PMC free article] [PubMed]
43. Bode W, Huber R. Structural basis of the endoproteinase-protein inhibitor interaction. Biochimica et biophysica acta. 2000;1477:241–252. [PubMed]
44. Maehr R, Mintern JD, Herman AE, Lennon-Dumenil AM, Mathis D, Benoist C, Ploegh HL. Cathepsin L is essential for onset of autoimmune diabetes in NOD mice. The Journal of clinical investigation. 2005;115:2934–2943. [PMC free article] [PubMed]
45. Gupta S, Singh RK, Dastidar S, Ray A. Cysteine cathepsin S as an immunomodulatory target: present and future trends. Expert Opin Ther Targets. 2008;12:291–299. [PubMed]
46. Gorczynski RM. CD200 and its receptors as targets for immunoregulation. Curr Opin Investig Drugs. 2005;6:483–488. [PubMed]
47. Xue HH, Bollenbacher J, Rovella V, Tripuraneni R, Du YB, Liu CY, Williams A, McCoy JP, Leonard WJ. GA binding protein regulates interleukin 7 receptor alpha-chain gene expression in T cells. Nature immunology. 2004;5:1036–1044. [PubMed]
48. Wardell BB, Michael SD, Tung KS, Todd JA, Blankenhorn EP, McEntee K, Sudweeks JD, Hansen WK, Meeker ND, Griffith JS, et al. Aod1, the immunoregulatory locus controlling abrogation of tolerance in neonatal thymectomy-induced autoimmune ovarian dysgenesis, maps to mouse chromosome 16. Proceedings of the National Academy of Sciences of the United States of America. 1995;92:4758–4762. [PubMed]
49. Bauer K, Yu X, Wernhoff P, Koczan D, Thiesen HJ, Ibrahim SM. Identification of new quantitative trait loci in mice with collagen-induced arthritis. Arthritis and rheumatism. 2004;50:3721–3728. [PubMed]
50. Badalova J, Svobodova M, Havelkova H, Vladimirov V, Vojtiskova J, Engova J, Pilcik T, Volf P, Demant P, Lipoldova M. Separation and mapping of multiple genes that control IgE level in Leishmania major infected mice. Genes and immunity. 2002;3:187–195. [PubMed]
51. Kosarova M, Havelkova H, Krulova M, Demant P, Lipoldova M. The production of two Th2 cytokines, interleukin-4 and interleukin-10, is controlled independently by locus Cypr1 and by loci Cypr2 and Cypr3, respectively. Immunogenetics. 1999;49:134–141. [PubMed]
52. Butterfield RJ, Blankenhorn EP, Roper RJ, Zachary JF, Doerge RW, Sudweeks J, Rose J, Teuscher C. Genetic analysis of disease subtypes and sexual dimorphisms in mouse experimental allergic encephalomyelitis (EAE): relapsing/remitting and monophasic remitting/nonrelapsing EAE are immunogenetically distinct. J Immunol. 1999;162:3096–3102. [PubMed]