Skip to main content
NCBI home page
Clinical and Experimental Immunology logoLink to Clinical and Experimental Immunology
. 2004 Oct;138(1):39–46. doi: 10.1111/j.1365-2249.2004.02560.x

Multiple loci are linked with anti-red blood cell antibody production in NZB mice – comparison with other phenotypes implies complex modes of action

N J LEE *,§, R J RIGBY †,*, H GILL †, J J BOYLE ‡, L FOSSATI-JIMACK †, B J MORLEY †, T J VYSE †
PMCID: PMC1809186  PMID: 15373903

Abstract

The New Zealand Black (NZB) mouse strain is a model of autoimmune haemolytic anaemia (AHA) and systemic lupus erythematosus (SLE), characterized by the production of anti-red blood cell (RBC) antibodies and anti-nuclear antibodies (ANA), respectively. A linkage analysis was carried out in an (NZB × BALB/c) F2 cross in order to identify loci involved in the production of both anti-RBC IgM and IgG antibodies. These regions of linkage were compared with linkage data to ANA from the same cohort and other linkage analyses involving New Zealand mice. Four previously described NZB loci linked to anti-RBC antibodies were confirmed, and eight novel loci linked to this trait were also mapped: five of which were of NZB origin, and three derived from the non-autoimmune BALB/c background. A comparison between loci linked with anti-RBC antibodies and ANA demonstrated many that co-localize, suggesting the presence of genes that result in the general breaking of tolerance to self-antigen. Furthermore, the observation that some loci were associated only with the anti-RBC response suggests an antigen specific mechanism in addition to a general breaking of tolerance. A locus linked with anti-RBC antibodies and ANA on distal chromosome 7 in this cohort is orthologous to one on the q arm of human chromosome 11, a region linked to AHA and ANA in human SLE.

Keywords: antibody, autoimmune haemolytic anaemia, linkage analysis, mouse

INTRODUCTION

The New Zealand Black (NZB) strain of mouse has been studied extensively as a model of the human diseases systemic lupus erythematosus (SLE) and autoimmune haemolytic anaemia (AHA). These mice develop spontaneously an autoimmune disease with associated serological abnormalities, in particular the production of anti-nuclear antibodies (ANAs) and anti-red blood cell (RBC) antibodies [1]. The subsequent destruction of red blood cells (RBCs) mediated by these anti-RBC antibodies leads to the development of AHA [2].

The pathogenic anti-RBC antibodies produced by NZB mice react with multiple determinants exposed on the surface of intact mouse RBCs. These determinants are referred to as X antigens and have not yet been characterized fully [2–5]. Barker et al. [6] suggested that these pathogenic autoantibodies in NZB mice are specific for the erythrocyte Band 3 protein (an anion transporter). However, the fact that several NZB-derived anti-erythrocyte monoclonal antibodies seem to bind to distinct but as yet undefined RBC surface determinants indicates that there is a considerable degree of heterogeneity in the NZB humoral immune response to RBCs. The binding of these monoclonal antibodies to X antigens on RBCs has been shown to increase progressively as the RBCs age [5]. This implicates membrane alterations in aged RBCs and the subsequent increased expression of X antigens in the development of AHA.

The activity of pathogenic anti-RBC antibodies is influenced by their specificity and isotype [2,7]. The Ig heavy chain plays a critical role in AHA by influencing binding to Fcγ receptors [2,7–9]. The sequestration of agglutinated RBCs in the spleen and liver is also a major mechanism involved in the development of AHA [2]. Another factor implicated in studies of anti-RBC autoimmune responses is the binding of CD47 expressed on the surface of normal RBCs with macrophage signal regulatory protein α (SIRPα). Oldenburg et al. [10] showed that the absence of CD47 (located on mid-chromosome 16) in non-obese diabetic (NOD) mice accelerated the development and progression of AHA, provided that the mice were genetically predisposed to autoimmune disease. It has also been shown that aged RBCs express a diminished level of CD47, which may contribute to the demonstrated acceleration in antibody-dependent clearance of aged RBCs [5].

AHA occurs as a complex, multigenic phenotype in NZB mice [11]. However, there is as yet only minimal information available about the identity and location of the genes involved [12]. Previous studies have shown that the systemic autoimmunity in NZB mice is inherited as a collection of independent disorders and that at least some loci are linked differentially with separate autoantibody specificities [13,14].

Ozaki et al. [15] suggested, by using a segregation analysis, that anti-RBC antibodies are produced by a combined effect of one to three dominant predisposing NZB genes and a single dominant modifying (suppressive) gene, identified in the C57BL/6 mouse strain and designated Aem1, linked loosely to major urinary protein 1 (Mup1) on proximal chromosome 4. In a later study two more potential loci were suggested, one on chromosome 7 between 27·8 cM and 37 cM designated Aem2, and the other on chromosome 10 called Aem3 at 44 cM [12]. However, in the later study no associations were found with the Aem1 locus on chromosome 4. Knight et al. [16] described two dominant or co-dominant genes involved in the occurrence of haemolytic anaemia in NZB mice using crosses with the New Zealand Chocolate (NZC) strain. One of these, the NZB locus Aia1, was linked to the b (black/brown) coat colour locus on distal chromosome 4.

We studied an (NZB × BALB/c) F2 intercross to determine loci influencing the production of both serum anti-RBC IgG and anti-RBC IgM antibodies. The effects two of these loci have in isolation were also investigated in BALB/c.NZB congenic mice. The data show that some AHA-linked loci are also linked to other autoimmune traits in New Zealand and BALB/c mice, suggesting that some mechanisms of autoimmunity are acting in a non-antigen-specific manner. However, other loci seem to be associated specifically with the anti-RBC response. Therefore, AHA in this cohort of mice seems to be the result of multiple genes and several autoimmune mechanisms.

MATERIALS AND METHODS

Mice

NZB/BINJ (NZB) and BALB/cByJ (BALB/c) mice were purchased from Harlan Olac Ltd (Bicester, Oxfordshire, UK) and maintained in the Biological Services Unit of Imperial College Faculty of Medicine (London, UK). These mice were crossed, and the resulting F1 progeny intercrossed to produce an (NZB × BALB/c) F2 cohort (n = 222 female mice).

Two BALB/c mouse lines, congenic for different regions of NZB chromosome 4, were bred to backcross six using the speed congenic method [17] and the interval fixed by intercrossing heterozygous carriers of the congenic interval. The BALB/c.NZB.C4a (C4a) congenic line contains an NZB region from the centromere to 30 cM of chromosome 4 on a BALB/c background, and the BALB/c.NZB.C4b (C4b) congenic line an NZB region from 34 cM to 66 cM of chromosome 4 on a BALB/c background. As in the F2 cohort, only female mice were studied.

The (NZB × BALB/c) F2 cohort were bled from the tail every 2 months from 6 months of age until 14 months old, and the congenic strains every 2 months from 3 months of age until 15 months old, or until a one-off proteinuria level of ≥5 g/l (3 +) on Combur3 urine dipstick (Roche Diagnostics, Lewes, UK) or a proteinuria level of 1 g/l (2) for 2 consecutive months resulted in the sacrifice of the individual. Blood was incubated at 37°C for 30 min, centrifuged for 10 min at 13 500 rpm at room temperature and the serum fraction removed. Sera were stored at −70°C until required.

DNA extracted from tail biopsies was amplified in a standard 35-cycle polymerase chain reaction (PCR) reaction with oligonucleotides flanking microsatellite repeat regions polymorphic between NZB and BALB/c. The resulting PCR product was electrophoresed on polyacrylamide gels (Mini-Protean II electrophoresis system, Bio-Rad, Hemel Hempstead, UK) at 2·25 V/mm for 90–120 min, stained with ethidium bromide solution, viewed under UV light and photographed digitally.

Anti-RBC antibody assay

The levels of both RBC-binding IgM and IgG antibodies in the serum of the mice were examined using flow cytometry and are referred to in this paper as anti-RBC IgM or IgG antibodies. The flow cytometric assay used was similar to that previously described by Fossati-Jimack et al. [8]. Briefly, mouse serum was diluted in PBS−1% bovine serum albumin (BSA) to 1 : 100 for the IgG assay and to 1 : 200 for the IgM assay. One hundred µl of the relevant diluted serum was then incubated in microtitre plates with a 1·25% RBC suspension in PBS−1% BSA for 1 h on ice. The RBCs were freshly obtained from young (<3 months), healthy BALB/c mice and washed three times in PBS prior to suspension. After washing three times with PBS−1% BSA, bound anti-RBC antibodies were detected by incubating for 1 h on ice with 30 µl of either antimouse IgG FITC (Serotec, Oxford, UK) at a concentration of 1 : 50 or biotinylated antimouse IgM (Sigma, Poole, UK) at 1 : 1000 in conjunction with 10 µl PE-conjugated streptavidin (Sigma, Poole, UK) at 1 : 10. Fluorescence was determined using flow cytometry (FacsCalibur, Becton Dickinson, San Jose, CA, USA) and analysed using WinMDI 2·8 analysis software (http://facs.scripps.edu/software.html). Antibody concentrations were determined by assigning nominal values (100–0·8 U) to a standard curve of pooled NZB and then calculating sample values from this standard curve after subtracting the background. Any samples that were over 100 U were diluted to bring them onto the standard curve, rerun and the result multiplied by the dilution factor.

Histology

Liver tissue was processed overnight and embedded in paraffin wax. Two µm sections were cut, rehydrated and stained with Perl's iron stain [2% hydrochloric acid and 2% potassium hexacyanoferrate(II) trihydrate (Sigma)] for 20 min. Following washing in dH20, sections were counterstained with 1% neutral red (Sigma) in 1% glacial acetic acid and mounted. The extent of iron deposits was determined blind by J. J. B. using the following criteria: 0 = absent, 1 = 1–2 haemosiderin-laden Kupffer cells per ×10 field, 2 = >2 haemosiderin-laden Kupffer cells per ×10 field, 3 = parenchymal cell and Kupffer cell haemosiderin deposits.

Statistical analysis

Linkage was determined by the program MapManager QT [18]. Thresholds for suggestive, significant and highly significant linkage were calculated by using the permutation test in MapManager QT with 1000 permutations and are specific for each data set. The thresholds ranged from χ2-values of 8·6–9·5 for suggestive linkage, 14·1–16·6 for significant linkage, and 18·2–26·3 for highly significant linkage.

Sample anti-RBC antibody levels were calculated from the standard curve using a third order polynomial equation, and the significance of differences in antibody levels between the strains studied was determined by the Mann–Whitney U-test. The significance of difference in hepatic iron deposits in the F2 cohort was determined using Fisher's exact test. These were performed using GraphPad Prism version 3·03 for Windows (GraphPad Software, San Diego, CA USA).

RESULTS

NZB mice develop AHA associated with high serum levels of anti-RBC IgM and IgG antibodies [1], whereas BALB/c mice have low levels of both antibodies. In order to study the genetic control of anti-RBC antibody production in NZB mice, an (NZB × BALB/c) F2 (B-F2) cohort was bred, mapped with microsatellite markers and linkage analysis carried out using the computer program MapManager QT. Two congenic lines with NZB regions on a BALB/c background were then investigated to confirm the F2 linkage results.

Serum anti-RBC antibody levels in parental and F2 mice

In a cohort of 9-month-old female NZB mice (n = 15) the median serum anti-RBC IgM level was 27·8 U and the median anti-RBC IgG level was 19·1 U. In comparison, a cohort of 8–9-month-old BALB/c mice (n = 18) had a significantly lower median serum anti-RBC IgM level (2·8 U; P < 1·0 × 10−4). The median serum anti-RBC IgG level was also significantly lower (2·7 U; P < 1·0 × 10−4) than that of NZB in a cohort of 8–9-month-old BALB/c mice (n = 15).

Serum anti-RBC IgM levels were measured at three time-points (6, 8 and 10 months old) and anti-RBC IgG levels at two time-points (8 and 10 months old) in (NZB × BALB/c) F2 mice (n = 222) (Fig. 1a,b). At 8 and 10 months old the median serum anti-RBC IgM levels were 5·0 U and 7·0 U, respectively, and thus significantly lower (P < 1·0 × 10−4) than that at 6 months old (13·1 U). In comparison, the median serum anti-RBC IgG level of 4·7 U at 10 months old is significantly higher (P < 1·0 × 10−4) than the median level at 8 months old (2·2 U). These results show a definite trend for the serum anti-RBC IgM levels to decrease with age while the serum anti-RBC IgG levels increase with age.

Fig. 1.

Fig. 1

Open in a new tab

(a) Distribution of anti-RBC IgM antibodies in NZB, BALB/c and (NZB × BALB/c) F2 mice at 6, 8 and 10 months old. Horizontal bars represent median point of data set. (b) Distribution of anti-RBC IgG antibodies in NZB, BALB/c and (NZB × BALB/c) F2 mice at 8 and 10 months old. Horizontal bars represent median point of data set.

Linkage of serum anti-RBC antibody levels in (NZB × BALB/c) F2 mice

The (NZB × BALB/c) F2 cohort was genotyped with 96 microsatellite markers across the autosomes at an average intermarker distance of approximately 17 cM and linkage analysis carried out over the multiple time-points for both serum anti-RBC IgM and IgG levels.

Linkage to serum anti-RBC IgM antibodies was observed at a number of loci, detailed in Table 1. Linkage was observed at a locus on proximal chromosome 12 at both 6 and 8 months of age (P = 4·5 × 10−3 and P = 7·8 × 10−4, respectively). This is in a position described previously in this strain combination to be associated with both high titres of ANAs and the development of lupus nephritis [19]. Linkage was also observed at a locus on proximal chromosome 7, again at both 6 and 8 months of age (P = 1·1 × 10−3 and P = 5·0 × 10−4, respectively). Proximal chromosome 7 is a region that has been linked consistently with both AHA and SLE traits in both NZB mice and other lupus-prone strains [12,20–24]. Additionally, linkage to anti-RBC IgM antibodies at 10 months of age was observed at loci on proximal chromosome 1 (P = 5·2 × 10−3, linkage to BALB/c allele), mid-chromosome 13 (P = 7·8 × 10−3) and distal chromosome 7 (P = 3·0 × 10−3). The locus on distal chromosome 7 has been associated previously with anti-double-stranded (ds) and anti-single-stranded (ss) DNA IgG antibodies in this strain combination [19].

Table 1.

Linkage of anti-RBC IgM antibodies in (NZB × BALB/c) F2 mice

Ch Age (months) Peak linkage P at peak linkage Region of > suggestive linkage (cM) Allele Known NZ and BALB AHA loci in region (position, traits, reference) Known NZ and BALB SLE loci in region (position, traits, reference)
1 10 19·5 5·2 × 10−3 19·5–24·5 BALB None None
7 6 4·1–5·4 1·1 × 10−3 3·4–14·3 NZB Aem2 (27·8–37 cm, AHA [12]) Sle3 (19 cm, ANA/GN [20,21]); Lbw5 (23 cM, Mo [22]); Nba3 (31 cm, GN [23]); Yaail (9–16 cM, ANA, GN [24])
7 8 6·8–8·2 5·0 × 10−4 3·4–14·3 NZB Aem2 (27·8–37 cm, AHA [12]) Sle3 (19 cm, ANA/GN [20,21]); Lbw5 (23 cM, Mo [22]); Nba3 (31 cm, GN [23]); Yaail (9–16 cm, ANA, GN [24])
7 10 58·3–59·7 3·0 × 10−3 52·7–65·3 NZB Aem2 27·8–37 cm, AHA [12] Nba5 (63 cm, ANA [23])
12 6 1 4·5 × 10−3 1 NZB None Nbwa1 (6 cm, ANA/GN/IgG [19])
12 8 1 7·8 × 10−4 1–2·3 NZB None Nbwa1 (6 cm, ANA/GN/IgG [19])
13 10 43·4–46 7·8 × 10−3 40–49·5 NZB None Yaa1 (41 cm, ANA [24])
Open in a new tab

AHA = autoimmune haemolytic anaemia, ANA = anti-nuclear antibodies, GN = glomerulonephritis, IgG = total serum IgG, IgM = total serum IgM, Mo = mortality.

As above, linkage to serum anti-RBC IgG antibodies was identified at a number of loci (Table 2). Linkage was observed at a locus on proximal chromosome 7 at 8 months of age (P = 4·7 × 10−3). This represents the only common point of linkage between anti-RBC IgG and IgM antibodies in this cohort. A complex pattern of linkage was observed on chromosome 4. At 8 months of age a region of suggestive linkage on proximal chromosome 4 (P = 6·7 × 10−3 at 4·2 cM) was observed, with a bimodal pattern of significant linkage on mid to distal chromosome 4 (P = 2·7 × 10−4 at 47·8 cM, P = 1·8 × 10−4 at 59·9 cM). However, at 10 months of age linkage was only observed at the distal chromosome 4 locus (P = 3·9 × 10−3), although there was a trend for linkage at the proximal locus (P = 3·0 × 10−2). The complexity of the linkage on chromosome 4 may reflect the fact that numerous autoimmunity loci, including two AHA loci, have been identified at various points across the chromosome.

Table 2.

Linkage of anti-RBC IgG antibodies in (NZB × BALB/c) F2 mice

Ch Age (months) Peak linkage P at peak linkage Region of > suggestive linkage (cM) Allele Known NZ and BALB AHA loci in region (position, traits, reference) Known NZ and BALB SLE loci in region (position, traits, reference)
3 8 16·5–17·8 7·8 × 10−3 16·5–22·5 NZB None None
4 8 6·1–7·2 7·8 × 10−3 3·5–8·9 NZB Aem1? (proximal of Mup1 at 27·8 cM, AHA [15]) Nbwa2 (0–20 cM, ANA/GN/IgG [19])
4 8 47·8 2·4 × 10−4 11·8–79* NZB None Sle2 (44·5 cM, GN/ANA/IgM [20,26])
4 8 59·9 1·8 × 10−4 11·8–79* NZB Aia1 (75 cM, AHA [16]) Lbw2 (55·6 cM, GN/ANA/Mo [22]); Nba1 (59–69 cM, GN [14,29])
4 10 64·6 2·3 × 10−3 58·7–75·4 NZB Aia1 (75 cM, AHA [16]) Lbw2 (55·6 cM, GN/ANA/Mo [22]; Nba1 (59–69 cM, GN [14,29])
6 8 21·3–22 3·9 × 10−3 13·6–24·8 NZB None Lxw2 (26 cM, GN [36])
7 8 13–13·6 4·7 × 10−3 9·5–14·3 NZB Aem2 (27·8–37 cM, AHA [12]) Sle3 (19 cM, ANA/GN [20,21]), 3 cM, Mo [22]); Nba3 (31 cM, GN [23]); Yaail (9–16 cM, ANA, GN [24])
9 10 31·2–31·7 1·3 × 10−2 30·6–32·3 BALB None Baa1 (28 cM, ANA [25])
10 10 58·7 1·7 × 10−3 54–66·8 NZB Aem3 (44 cM, AHA [12]) Sle12 (69 cM, GN [21])
16 8 54 5·0 × 10−3 51·7–56·8 BALB None None
Open in a new tab

AHA = autoimmune haemolytic anaemia, ANA = anti-nuclear antibodies, GN = glomerulonephritis, IgG = total serum IgG, IgM = total serum IgM, Mo = mortality.

*

Bimodal linkage pattern.

Additional regions of linkage to anti-RBC IgG antibodies at 8 months of age were observed on proximal chromosome 3 (P = 7·8 × 10−3), mid-chromosome 6 (P = 3·9 × 10−3) and distal chromosome 16 (P = 5·0 × 10−3, linkage to BALB/c alleles), and to anti-RBC IgG antibodies at 10 months of age on distal chromosome 10 (P = 1·7 × 10−3), close to the AHA locus Aem3, and to a BALB/c derived locus on mid-chromosome 9 (P = 1·3 × 10−2). This locus co-localizes with the BALB/c-derived SLE associated locus Baa1 [25].

Analysis of serum anti-RBC antibody levels in congenic lines

In order to investigate the roles of two of the loci identified in the linkage analysis in isolation, serum anti-RBC antibody levels were assayed in BALB/c mice congenic for NZB proximal chromosome 4 (C4a) and NZB distal chromosome 4 (C4b). No linkage to anti-RBC IgM antibodies was observed on proximal chromosome 4. This was corroborated in the C4a congenic line, which did not exhibit significantly raised anti-RBC IgM antibodies compared to BALB/c controls (P > 0·05, Fig. 2a). Conversely, the C4b congenic line had significantly raised anti-RBC IgM antibodies compared to BALB/c controls (P = 0·0003, Fig. 2a). The genomic region encompassed by the C4b congenic demonstrated a trend for linkage to anti-RBC IgM antibodies in the B-F2 cohort (P = 0·015 at 6 month, P = 0·027 at 8 months), and has been shown previously to be linked very strongly to total serum IgM in NZB mice [19,26]. Neither of the congenic lines had significantly raised serum anti-RBC IgG antibody levels compared to BALB/c mice (Fig. 2b).

Fig. 2.

Fig. 2

Open in a new tab

(a) Distribution of anti-RBC IgM antibodies in NZB, BALB/c, C4a and C4b congenic mice. n.s. = not significant, *P = 0·0003, **P < 0·0001. Horizontal bars represent median point of data set. (b) Distribution of anti-RBC IgM antibodies in NZB, BALB/c, C4a and C4b congenic mice. n.s. = not significant, **P < 0·0001. Horizontal bars represent median point of data set.

Histology

In order to determine if raised anti-RBC IgG antibodies have a haemolytic effect, individuals in the B-F2 cohort were ranked according to the sum of serum anti-RBC IgG antibodies at both 8 months and 10 months. Liver sections from the highest 14 (median value 23·64 U) and lowest 14 (median value 1·97 U) were stained for iron deposits with Perl's stain and counterstained with neutral red. By grouping the liver sections as unaffected (score of 0–1) and affected (2–3), the cohort with the highest anti-RBC IgG levels had a significantly higher (P = 0·0019) degree of iron deposition than the cohort with the lowest anti-RBC IgG levels, as determined by Fisher's exact test.

DISCUSSION

This study provides insight into the genetic control of AHA in NZB mice by identifying loci involved in the production of anti-RBC antibodies using linkage analysis in an (NZB × BALB/c) F2 cohort.

We have confirmed the linkage of a number of loci that were previously associated with AHA in NZB mice. Two loci on chromosome 4, Aem1 (proximal of Mup1 at 27·8 cM) and Aia1 (at 75 cM) have been linked previously to AHA in NZB mice [15,16]. In this cross we have observed linkage to anti-RBC IgG antibodies in the region of both these loci. Additionally, we have observed a trend for linkage to anti-RBC IgM antibodies on distal chromosome 4. Ochiai et al. identified two loci linked to AHA in NZB mice − Aem2 on chromosome 7 and Aem3 on chromosome 10 [12]. In this study we have observed linkage of anti-RBC antibodies to these regions, thus confirming the findings of the previous study.

In this investigation we have also identified novel loci linked to anti-RBC antibody production in NZB and BALB/c mice. Loci derived from the NZB background linked to anti-RBC antibodies were identified on proximal chromosome 3, proximal chromosome 6, distal chromosome 7, proximal chromosome 12 and mid-chromosome 13. Interestingly, three loci linked to anti-RBC antibodies were derived from the ‘non-autoimmune’ BALB/c background. These were situated on proximal chromosome 1, proximal chromosome 9 and distal chromosome 16. All the novel anti-RBC loci were linked with a similar probability (P < threshold for suggestive linkage) to the loci that were found to corroborate the Aem and Aia loci mapped by other investigators. All these loci are summarized in Fig. 3.

Fig. 3.

Fig. 3

Open in a new tab

Linkage of SLE and AHA traits in NZB and BALB/c mice. Loci in italics represent previously described loci (AHA loci in bold italics). Solid blocks represent NZB linkage regions, and open blocks BALB/c linkage regions in the (NZB × BALB/c) F2 cohort. Linked traits are shown adjacent to linkage regions. Chromosomes demonstrating no linkage in this cohort are omitted.

Of all the loci identified in this work linked to anti-RBC antibodies, the majority overlap with loci linked previously with anti-nuclear antibodies (ANAs), that is anti-ssDNA, anti-dsDNA and antichromatin antibodies, or a combination thereof. Loci on proximal, and to a greater extent distal chromosome 4 have been linked to ANAs, and other lupus traits, in this cohort [19] and a number of other investigations in New Zealand mice [14,20,22,26–30]. Similarly, chromosome 7, which was linked to both anti-RBC IgG and IgM antibodies in this cohort, was also linked to ANAs and other lupus traits both in this cohort [19] and a number of other studies [20–22,24]. This pattern of linkage, i.e. anti-RBC loci co-localizing with ANA/lupus loci, is also seen on proximal chromosome 12, at a locus that is derived from NZB but is dependent on the presence of the BALB/c genetic background, on distal chromosome 10 [19] and on mid-chromosome 13 [24,31]. Of the loci derived from the BALB/c genome, only that on chromosome 9 was in a region linked previously to lupus traits, named Baa1 by Vyse et al. [25].

Two NZB-derived and two BALB/c-derived anti-RBC antibody-associated loci were linked to regions of the genome unlinked previously with ANAs. These NZB derived loci, on proximal chromosome 3 and proximal chromosome 6, and BALB/c-derived loci on proximal chromosome 1 and distal chromosome 16, may represent loci specifically associated with the anti-RBC response, and not a general autoimmune response. Loci linked to an antigen-specific autoimmune response may be the result of a polymorphic antigenic determinant at that loci. Therefore, we examined the regions linked only to the anti-RBC response for possible candidate genes that may be influencing this phenotype. A possible candidate gene, Kel, maps to proximal chromosome 6 at 20·5 cM [32], 0·8 cM away from the peak of linkage in this region. The gene encodes the blood group antigen Kell, and a polymorphic structural element of this erythrocyte membrane glycoprotein would be a viable candidate for a region linked solely to an anti-RBC antibody response [32]. The fact that this protein is highly polymorphic in humans lends weight to this theory, and we are currently investigating whether NZB mice generate antibodies to the Kell protein.

Conversely, loci associated only with ANAs in this cohort [19], namely distal chromosome 1 at Sle1/Nba2, proximal chromosome 5 around Sle6 and distal chromosome 17 at Agnz1, may be associated specifically with an anti-nuclear response. It is of note that Sle1 has been shown previously to be restricted to a specific antigen, H2A/H2B/DNA complexes [33]. In conclusion, these data suggest two general mechanisms of autoantibody production in NZB mice − a general loss of tolerance that is not antigen-specific, and an antigen-specific response.

A number of loci are linked to anti-RBC IgM antibodies. Of these loci (on chromosomes 1, 7, 12 and 13) only proximal chromosome 7 is also linked to total serum IgM; however, distal chromosome 4 exhibited a trend for linkage to anti-RBC IgM antibodies and is linked very strongly to total IgM production [19,29]. Therefore, the anti-RBC response associated with these regions may be part of a general unregulated B-cell hyperactivity, rather than be directed at a specific antigen or set of antigens.

The analysis of anti-RBC antibody levels in BALB/c mice congenic for NZB proximal chromosome 4 (C4a) and distal chromosome 4 (C4b) confirmed some of the linkage data. Anti-RBC IgM antibodies were linked to distal chromosome 4; the congenic line carrying this interval had significantly raised anti-RBC IgM levels at 9 months (P = 0·0003, Fig. 2a). The C4a congenic line did not have significantly raised anti-RBC IgM levels, perhaps unsurprising considering the lack of linkage in this region. Despite the linkage of anti-RBC IgG at loci on both proximal and distal chromosome 4, the congenic lines carrying these regions did not exhibit significantly raised anti-RBC IgG antibodies (Fig. 2b). Therefore, in isolation, the susceptibility genes present within these loci do not seem to disrupt the mechanism of tolerance sufficiently to orchestrate fully an anti-RBC IgG response.

A number of linkage analysis experiments have been carried out in human SLE, linking a numerous disease loci to the development of SLE or disease phenotypes. A locus on 11q14 was associated with both AHA and ANA in an African-American cohort [34–35]. This locus is orthologous to distal mouse chromosome 7, which was linked to both ANA and AHA in this study, and to ANA in NZB mice in a previous study [23]. Additionally, Tsao et al. have shown that thrombocytopenia and AHA occur together in families with SLE, with thrombocytopenia in SLE often being the result of antiplatelet antibodies [37].

In conclusion, loci that are linked with autoimmunity in NZB mice seem to do so in both an antigen-specific and antigen non-specific manner. Antigen-specific loci may harbour genes encoding molecules that directly influence antigen availability, for example; we propose the Kel gene on chromosome 6 as a strong candidate in AHA. Conversely, loci linked to non-antigen specific autoimmunity, such as on proximal chromosome 7, may harbour genes that function in the general activation and regulation of the immune system. The exclusive linkage of a specific class of immunoglobulin to an autoimmune-associated locus may also provide valuable information into the role of the underlying gene or genes − for example, the linkage of both total serum IgM and anti-RBC IgM antibodies to the mid-region of chromosome 4 suggests gene/s involved in B-cell activation, or B1 cell function.

Acknowledgments

The authors would like to thank Mrs Margarita Lewis for technical assistance. A Project Grant from the Arthritis Research Campaign awarded to B. J. M and T. J. V. and a Welcome Trust Senior Fellowship Grant, awarded to T. J. V. supported this work.

REFERENCES

  • 1.Theofilopoulos AN, Dixon FJ. Murine models of systemic lupus erythematosus. Adv Immunol. 1985;37:269–390. doi: 10.1016/s0065-2776(08)60342-9. [DOI] [PubMed] [Google Scholar]
  • 2.Izui S. Autoimmune hemolytic anemia. Curr Opin Immunol. 1994;6:926–30. doi: 10.1016/0952-7915(94)90015-9. [DOI] [PubMed] [Google Scholar]
  • 3.DeHeer DH, Linder EJ, Edgington TS. Delineation of spontaneous erythrocyte autoantibody responses of NZB and other strains of mice. J Immunol. 1978;120:825–30. [PubMed] [Google Scholar]
  • 4.Diiulio NA, Fairchild RL, Caulfield MJ. The anti-erythrocyte autoimmune response of NZB mice. Identification of two distinct autoantigens. Immunology. 1997;91:246–51. doi: 10.1046/j.1365-2567.1997.00248.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Fossati-Jimack L, Azeredo DS, Moll T, et al. Selective increase of autoimmune epitope expression on aged erythrocytes in mice: implications in anti-erythrocyte autoimmune responses. J Autoimmun. 2002;18:17–25. doi: 10.1006/jaut.2001.0563. [DOI] [PubMed] [Google Scholar]
  • 6.Barker RN, Sa Oliveira GG, Elson CJ, Lydyard PM. Pathogenic autoantibodies in the NZB mouse are specific for erythrocyte band 3 protein. Eur J Immunol. 1993;23:1723–6. doi: 10.1002/eji.1830230750. [DOI] [PubMed] [Google Scholar]
  • 7.Fossati-Jimack L, Ioan-Facsinay A, Reininger L, et al. Markedly different pathogenicity of four immunoglobulin G isotype-switch variants of an antierythrocyte autoantibody is based on their capacity to interact in vivo with the low-affinity Fcgamma receptor III. J Exp Med. 2000;191:1293–302. doi: 10.1084/jem.191.8.1293. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Fossati-Jimack L, Reininger L, Chicheportiche Y, et al. High pathogenic potential of low-affinity autoantibodies in experimental autoimmune hemolytic anemia. J Exp Med. 1999;190:1689–96. doi: 10.1084/jem.190.11.1689. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Azeredo dS, Kikuchi S, Fossati-Jimack L, et al. Complement activation selectively potentiates the pathogenicity of the IgG2b and IgG3 isotypes of a high affinity anti-erythrocyte autoantibody. J Exp Med. 2002;195:665–72. doi: 10.1084/jem.20012024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Oldenborg PA, Gresham HD, Chen Y, Izui S, Lindberg FP. Lethal autoimmune hemolytic anemia in CD47-deficient nonobese diabetic (NOD) mice. Blood. 2002;99:3500–4. doi: 10.1182/blood.v99.10.3500. [DOI] [PubMed] [Google Scholar]
  • 11.Howie JB, Simpson LO. Autoimmune haemolytic disease in NZB mice. Series Haematol. 1974;7:386–426. [PubMed] [Google Scholar]
  • 12.Ochiai K, Ozaki S, Tanino A, et al. Genetic regulation of anti-erythrocyte autoantibodies and splenomegaly in autoimmune hemolytic anemia-prone new zealand black mice. Int Immunol. 2000;12:1–8. doi: 10.1093/intimm/12.1.1. [DOI] [PubMed] [Google Scholar]
  • 13.Bocchieri MH, Cooke A, Smith JB, Weigert M, Riblet RJ. Independent segregation of NZB immune abnormalities in NZB × C58 recombinant inbred mice. Eur J Immunol. 1982;12:349–54. doi: 10.1002/eji.1830120417. [DOI] [PubMed] [Google Scholar]
  • 14.Vyse TJ, Drake CG, Rozzo SJ, Roper E, Izui S, Kotzin BL. Genetic linkage of IgG autoantibody production in relation to lupus nephritis in New Zealand hybrid mice. J Clin Invest. 1996;98:1762–72. doi: 10.1172/JCI118975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Ozaki S, Honda H, Maruyama N, et al. Genetic regulation of erythrocyte autoantibody production in New Zealand black mice. Immunogenetics. 1983;18:241–54. doi: 10.1007/BF00952963. [DOI] [PubMed] [Google Scholar]
  • 16.Knight JG, Adams DD. Genes determining autoimmune disease in New Zealand mice. J Clin Lab Immunol. 1981;5:165–70. [PubMed] [Google Scholar]
  • 17.Wakeland EK, Morel L, Achey K, Yui M, Longmate J. Speed congenics: a classic technique in the fast lane (relatively speaking) Immunol Today. 1997;18:472–7. doi: 10.1016/s0167-5699(97)01126-2. [DOI] [PubMed] [Google Scholar]
  • 18.Manly KF, Olson JM. Overview of QTL mapping software and introduction to map manager QT. Mamm Genome. 1999;10:327–34. doi: 10.1007/s003359900997. [DOI] [PubMed] [Google Scholar]
  • 19.Rigby RJ, Rozzo SJ, Boyle JJ, Lewis M, Kotzin BL, Vyse TJ. New Loci from new zealand black and new zealand white on chromosomes 4 and 12 contribute to lupus-like disease in the context of BALB/c. J Immunol. 2004;172:4609–17. doi: 10.4049/jimmunol.172.7.4609. [DOI] [PubMed] [Google Scholar]
  • 20.Morel L, Rudofsky UH, Longmate JA, Schiffenbauer J, Wakeland EK. Polygenic control of susceptibility to murine systemic lupus erythematosus. Immunity. 1994;1:219–29. doi: 10.1016/1074-7613(94)90100-7. [DOI] [PubMed] [Google Scholar]
  • 21.Morel L, Mohan C, Yu Y, et al. Multiplex inheritance of component phenotypes in a murine model of lupus. Mamm Genome. 1999;10:176–81. doi: 10.1007/s003359900964. [DOI] [PubMed] [Google Scholar]
  • 22.Kono DH, Burlingame RW, Owens DG, et al. Lupus susceptibility loci in New Zealand mice. Proc Natl Acad Sci USA. 1994;91:10168–72. doi: 10.1073/pnas.91.21.10168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Xie S, Chang SH, Sedrak P, Kaliyaperumal A, Datta SK, Mohan C. Dominant NZB contributions to lupus in the (SWR × NZB) F1 model. Genes Immun. 2002;3(Suppl. 1):S13–20. doi: 10.1038/sj.gene.6363886. [DOI] [PubMed] [Google Scholar]
  • 24.Santiago ML, Mary C, Parzy D, et al. Linkage of a major quantitative trait locus to Yaa gene-induced lupus-like nephritis in (NZW × C57BL/6), F1 mice. Eur J Immunol. 1998;28:4257–67. doi: 10.1002/(SICI)1521-4141(199812)28:12<4257::AID-IMMU4257>3.0.CO;2-H. [DOI] [PubMed] [Google Scholar]
  • 25.Vyse TJ, Morel L, Tanner FJ, Wakeland EK, Kotzin BL. Backcross analysis of genes linked to autoantibody production in New Zealand White mice. J Immunol. 1996;157:2719–27. [PubMed] [Google Scholar]
  • 26.Mohan C, Morel L, Yang P, Wakeland EK. Genetic dissection of systemic lupus erythematosus pathogenesis: Sle2 on murine chromosome 4 leads to B cell hyperactivity. J Immunol. 1997;159:454–65. [PubMed] [Google Scholar]
  • 27.Waters ST, Fu SM, Gaskin F, et al. NZM2328: a new mouse model of systemic lupus erythematosus with unique genetic susceptibility loci. Clin Immunol. 2001;100:372–83. doi: 10.1006/clim.2001.5079. [DOI] [PubMed] [Google Scholar]
  • 28.Drake CG, Babcock SK, Palmer E, Kotzin BL. Genetic analysis of the NZB contribution to lupus-like autoimmune disease in (NZB × NZW) F1 mice. Proc Natl Acad Sci USA. 1994;91:4062–6. doi: 10.1073/pnas.91.9.4062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Hirose S, Tsurui H, Nishimura H, Jiang Y, Shirai T. Mapping of a gene for hypergammaglobulinemia to the distal region on chromosome 4 in NZB mice and its contribution to systemic lupus erythematosus in (NZB × NZW) F1 mice. Int Immunol. 1994;6:1857–64. doi: 10.1093/intimm/6.12.1857. [DOI] [PubMed] [Google Scholar]
  • 30.Morel L, Tian XH, Croker BP, Wakeland EK. Epistatic modifiers of autoimmunity in a murine model of lupus nephritis. Immunity. 1999;11:131–9. doi: 10.1016/s1074-7613(00)80088-6. [DOI] [PubMed] [Google Scholar]
  • 31.Rozzo SJ, Vyse TJ, Menze K, Izui S, Kotzin BL. Enhanced susceptibility to lupus contributed from the nonautoimmune C57BL/10, but not C57BL/6, genome. J Immunol. 2000;164:5515–21. doi: 10.4049/jimmunol.164.10.5515. [DOI] [PubMed] [Google Scholar]
  • 32.Lee S, Russo DC, Pu J, Ho M, Redman CM. The mouse Kell blood group gene (Kel): cDNA sequence, genomic organization, expression, and enzymatic function. Immunogenetics. 2000;52:53–62. doi: 10.1007/s002510000251. [DOI] [PubMed] [Google Scholar]
  • 33.Mohan C, Alas E, Morel L, Yang P, Wakeland EK. Genetic dissection of SLE pathogenesis. Sle1 on murine chromosome 1 leads to a selective loss of tolerance to H2A/H2B/DNA subnucleosomes. J Clin Invest. 1998;101:1362–72. doi: 10.1172/JCI728. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Kelly JA, Thompson K, Kilpatrick J, et al. Evidence for a susceptibility gene (SLEH1) on chromosome 11q14 for systemic lupus erythematosus (SLE) families with hemolytic anemia. Proc Natl Acad Sci USA. 2002;99:11766–71. doi: 10.1073/pnas.182162399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Sawalha AH, Namjou B, Nath SK, et al. Genetic linkage of systemic lupus erythematosus with chromosome 11q14 (SLEH1) in African-American families stratified by a nucleolar anti-nuclear antibody pattern. Genes Immun. 2002;3(Suppl. 1):S31–S34. doi: 10.1038/sj.gene.6363904. [DOI] [PubMed] [Google Scholar]
  • 36.Kono DH, Park MS, Theofilopoulos AN. Genetic complementation in female (BXSB × NZW) F2 mice. J Immunol. 2003;171:6442–7. doi: 10.4049/jimmunol.171.12.6442. [DOI] [PubMed] [Google Scholar]
  • 37.Tsao BP, Grossman JM, Riemekasten G, et al. Familiality and co-occurrence of clinical features of systemic lupus erythematosus. Arthritis Rheum. 2002;46:2678–85. doi: 10.1002/art.10519. [DOI] [PubMed] [Google Scholar]

Articles from Clinical and Experimental Immunology are provided here courtesy of British Society for Immunology

RESOURCES