Biliary atresia (BA) is an idiopathic inflammatory cholangiopathy that occurs during early infancy. Two main types of BA exist: an embryonic subtype, associated with congenital malformations (1), and an “acquired” subtype, which is associated with no malformations and represents the vast majority of BA cases. Recently, maternal microchimerism has been demonstrated in the livers of patients with BA (2–5), suggesting a possible role of maternal cells in disease pathogenesis. Maternal-fetal cell transfer is known to be a common occurrence during uncomplicated pregnancy (6); however, the acceptance or tolerance of maternal cells into an organ of an infant would presumably require histocompatibility. An increase in human leukocyte antigen (HLA) compatibility between mother and child has been shown in other disease states, such as scleroderma (7) and systemic lupus erythematosus (8), in which maternal microchimerism has been implicated in disease pathogenesis.
Bidirectional HLA compatible relationship occurs when an offspring's inherited paternal antigen (IPA) is identical to the noninherited maternal antigen (NIMA). It is also considered “unidirectionally compatible” if the child is heterozygous at a certain HLA locus and the mother is homozygous (noninherited maternal antigen is identical to inherited maternal antigen) at that locus. In this situation the child would readily accept the maternal cells transferred. Considering a clinical report of maternal microchimerism causing hepatic graft-versus-host disease (GvHD) in children with severe combined immunodeficiency (9), maternal cells seen in patients with BA could be a cause of immunological insult seen in these patients.
Given the increase in maternal microchimerism found in BA, we hypothesized patients with BA have an increased HLA compatibility with their mothers.
A retrospective analysis of patients who had undergone liver transplantation at Kyoto University Hospital between 1995 and 2004 for end-stage liver diseases due to BA and other neonatal liver diseases were reviewed. Human leukocyte antigen typing of their mothers was performed in all cases as one of the basic investigations for liver transplantation. Basic characteristics of the patients are listed in Table 1.
The HLA data (HLA-A, B, DR, DQ) of 57 BA patient–mother pairs and 50 non-BA control pairs within Japanese population were included in this analysis. Two cases of embryonic BA associated with congenital heart disease were excluded from the analysis.
To evaluate HLA compatibility between the BA patients and their mothers, we defined 6 compatible and incompatible HLA combinations as described in Figure 1. For a statistical analysis, Fisher exact test was used to calculate P values. Odds ratios (OR) were calculated as the ratios of the proportions of the BA pairs with HLA compatibilities to the proportion of the control pairs with HLA incompatibilities.
Analysis of Bidirectional HLA Compatibility
A significantly higher matching was found between BA pairs at HLA class I (OR = 2.46, P = 0.038; Table 2). Among HLA class I loci, the prominent OR and P value were obtained from HLA-A (OR = 3.07, P = 0.017). There were 4 pairs with homozygosity for HLA-A 24 antigen in BA, whereas there was 1 pair in controls. Neither the compatibility at HLA-DR nor DQ showed statistical significance within the BA pairs (OR = 1.55; P = 0.262, OR = 0.76; P = 0.332, respectively; Table 2).
Analysis of Unidirectional HLA Compatibility
To further define unidirectional (mother-to-child or child-to-mother) compatibility and incompatibility, one-way matching was also evaluated in 4 unidirectional compatibility combinations (Fig. 1). Biliary atresia pairs were weakly associated with “child-to-mother compatiblility” at HLA class I (OR = 1.95, P = 0.067). In addition, patients with 2 or more “child-to-mother compatible” loci either at HLA class I or II were significantly higher in BA pairs. On the contrary, no compatible association (P < 0.1) was observed in the “mother-to-child” combinations. We further examined our hypothesis by analyzing incompatible relations that correspond to the “closed testing procedure.” There were significantly fewer patients who had at least 1 “child-to-mother incompatible” HLA locus in BA pairs (OR = 0.22, P = 0.043). Moreover, there were also fewer patients with more than 3 “child-to-mother incompatible” HLA loci in BA pairs (OR = 0.43, P = 0.034).
We have demonstrated an increased frequency of bidirectional HLA class I histocompatibility in BA pairs (BA patients and their mothers). In addition, there is an increased frequency of BA pairs associated with HLA class I compatibility from the perspective of the child (Table 2, 2 loci or more).
Our findings suggest that patients with BA have genetically favorable conditions to accept maternal cells. These findings are consistent with those of Berry et al (10), who reported that a high “fetus-to-mother” HLA class compatibility was associated with increased levels of maternal microchimerism. However, our results differ from their report, as well as other studies associating autoimmune diseases with HLA class II histocompatibility, in regards to the class of HLA (7,8). High compatibility found in HLA class I in our study may be due to the different ethnicity of the patients, because the HLA barrier in the Japanese population is considered to be different from that in the non-Japanese/white population (11). In the Japanese population, HLA-A and/or HLA-B matching was found to be the most important factor in preventing chronic GvHD during hematopoietic stem cell transplantation (12), whereas the importance of HLA class II matching for GvHD was reported among whites.
The increase in histocompatability between patients and mothers could also help explain the ethnicity-dependent influence on BA incidence. Intriguingly, the incidence of BA in the Japanese population living in Japan (1.0–1.2/10,000 live births) is close to that seen in Japanese living in Hawaii (0.8 ± 0.3), but not to whites (0.6 ± 0.2), Filipino (2.0 ± 0.9), or Chinese (2.0 ± 0.9) living in Hawaii (13,14). This ethnic population dependent difference may be explained by the different degree of increased mother–patient HLA compatibility of different ethnic groups. The higher incidence of BA among these ethnic populations could be due to a less genetic admixture occurring in these populations. An Egyptian article, which reports that half of BA patients undergoing Kasai were from consanguineous marriage, also supports this explanation (15). Although familial cases have been reported (16), familial occurrence is not generally seen in BA. This would not contradict our hypothesis, as the development BA may be a complex immunologic process requiring a “second hit.” This “second hit” may be immune or infectious including viral or bacterial triggers.
Although the presence of increased maternal microchimerism has been confirmed by multiple investigators, the exact role of these cells in BA is unclear. Recently we identified the existence of maternally derived CD8+ T cells (5). Such cells could attack fetal tissues including developing bile ducts during early gestation, which would lead to BA. However, how these maternal cells are taking part in the pathogenesis of BA needs further investigation.
There are several limitations in our study. First, our control group did not include healthy individuals. Therefore, it is difficult to know the true implications of the patient–mother histocompatibility. Second, BA patients in this cohort were only from those requiring liver transplant. We may be looking at only poor prognostic subgroup of BA. Materno-fetal HLA compatibility may not be expanded generally to all of the patients with BA and the mothers. Third, the number of cases evaluated was limited. A larger population-based investigation is needed to draw any conclusion. Fourth, HLA analysis was not done by high resolution. We do not think this will change our conclusion; however, high-resolution–based analyses will allow us to perform more precise compatibility estimations. Despite these limitations, this is the first report regarding an association of HLA class I compatibility in Japanese patients with BA and their mothers. Our preliminary results warrant further investigation regarding maternal HLA compatibility with maternal microchimerism in BA.
The authors thank Dr David Suskind, Seattle Children's Hospital and Dr Saji, HLA Research Laboratory, Kyoto for their valuable discussion.
1. Davenport M, Tizzard SA, Underhill J, et al
. The biliary atresia
splenic malformation: a 28-year single-center retrospective study. J Pediatr 2006; 149:393–400.
2. Suskind DL, Rosenthal P, Heyman MB, et al
. Maternal microchimerism
in the livers of patients with biliary atresia
. BMC Gastroenterol 2004; 4:14.
3. Kobayashi H, Tamatani T, Tamura T, et al
. Maternal microchimerism
in biliary atresia
. J Pediatr Surg 2007; 42:987–991.
4. Hayashida M, Nishimoto Y, Matsuura T, et al
. The evidence of maternal microchimerism
in biliary atresia
using in situ hybridisation. J Pediatr Surg 2007; 42:2097–2101.
5. Muraji T, Hosaka N, Irie N, et al
. Maternal microchmerism in underlying pathogenesis of biliary atresia
: quantification and phenotypes of maternal cells in the liver. Pediatrics 2008; 121:517–521.
6. Lo YM, Lo ES, Watson N, et al
. Two-way cell traffic between mother and fetus: biologic and clinical implications. Blood 1996; 88:4390–4395.
7. Nelson JL, Furst DE, Maloney S, et al
. Microchimerism and HLA-compatible relationships of pregnancy in scleroderma. Lancet 1998; 351:559–562.
8. Stevens AM, Tsao BP, Hahn BH, et al
. Maternal HLA class II compatibility in men with systemic lupus erythematosus. Arthritis Rheum 2005; 52:2768–2773.
9. Susanna M, Ege M, Pottharst A, et al
. Transplacentally acquired maternal T lymphocytes in severe combined immunodeficiency: a study of 121 patients. Blood 2001; 98:1847–1851.
10. Berry SM, Hassan HS, Russell E, et al
. Association of maternal histocompatibility at class II HLA loci with maternal microchimerism
in the fetus. Pediatr Res 2004; 56:73–78.
11. HLA-A2 allele compatibility and its effect to clinical outcome in hematopoietic cell transplantation (HCT) from unrelated donors. In: Analysis of HCT Component in the 13th International Histocompatibility Workshop HLA2005
12. Morishima Y, Sasazuki T, Inoko H, et al
. The clinical significance of human leukocyte antigen
(HLA) allele compatibility in patients receiving a marrow transplant from serologically HLA-A, HLA-B, and HLA-DR matched unrelated donors. Blood 2002; 99:4200–4206.
13. Shim WK, Kasai M, Spence MA, et al
. Racial influence on the incidence of biliary atresia
. Prog Pediatr Surg 1974; 6:53–62.
14. Japanese Biliary Atresia
Society. The data of nation-wide survey of biliary atresia
in 2004. J Jap Soc Pediatr Surg
15. Hassan HA-K, El-Ayyouti M, Hawas S, et al
. HLA in Egyptian children with biliary atresia
. J Pediatr 2002; 141:432–433.
16. Smith BM, Laberge JM, Schreiber R, et al
. Familial biliary atresia
in three siblings including twins. J Pediatr Surg 1991; 26:1331–1333.