Article Text

Download PDFPDF

Protection against rheumatoid arthritis by HLA: nature and nurture
  1. A L Feitsma1,2,
  2. A H M van der Helm-van Mil1,
  3. T W J Huizinga1,
  4. R R P de Vries2,
  5. R E M Toes1
  1. 1
    Department of Rheumatology, LUMC, Leiden, The Netherlands
  2. 2
    Department of Immuno- haematology and Blood Transfusion, LUMC, Leiden, The Netherlands
  1. Miss A L Feitsma, Department of Rheumatology, C1-R, Albinusdreef 2, 2333 ZA Leiden, The Netherlands; A.L.Feitsma{at}


Rheumatoid arthritis (RA) is a complex genetic disorder in which the HLA region contributes most to the genetic risk. HLA-DRB1 molecules containing the amino acid sequence QKRAA/QRRAA/RRRAA (ie, HLA-DRB1*0101, *0102, *0401, *0404, *0405, *0408, *0410, *1001 and *1402) at position 70–74 in the third hypervariable region of the DRB1 chain are associated with susceptibility to RA. HLA-DRB1 molecules containing the amino acids “DERAA” (ie, HLA-DRB1*0103, *0402, *1102, *1103, *1301, *1302 and *1304) at the same position are associated with protection from RA. Interestingly, not only inherited but also non-inherited HLA-antigens from the mother can influence RA susceptibility. A protective effect of “DERAA”-containing HLA-DRB1 alleles as non-inherited maternal antigen (NIMA) has recently been described. The underlying mechanism of this protective effect is currently unknown, although a possible explanation is covered below. In this review, an overview of the current knowledge on protection against RA is given and the inherited and NIMA effect of “DERAA”-containing HLA-DRB1 alleles are compared.

Statistics from


Rheumatoid arthritis (RA) is a complex genetic disorder in which the HLA region contributes most to the genetic risk. Especially HLA-DRB1 molecules sharing a common epitope, R(Q)K(R)RAA (ie, the amino acids arginine, (glutamine), lysine, (arginine), arginine, alanine, alanine) at position 70–74 in the third hypervariable region of the DRB1 chain, the so-called shared epitope (SE), are associated with both susceptibility to, and severity of, RA.14 The shared epitope is present in the HLA-DRB1*0101, *0102, *0401, *0404, *0405, *0408, *0410, *1001 and *1402 molecules.

At the same position as the SE, the amino acids “DERAA” (ie, the amino acids aspartic acid, glutamic acid, arginine, alanine, alanine) can be present in other HLA-DRB1 molecules (ie, HLA-DRB1*0103, *0402, *1102, *1103, *1301, *1302 and *1304). People carrying HLA-DRB1 alleles that express this “DERAA” sequence display a lower susceptibility to develop RA and have less severe disease than people with “neutral” (SE- and “DERAA”-negative) HLA-DRB1 alleles. The odds ratio of people carrying HLA-DRB1 alleles that express the “DERAA” sequence compared with that of people with “neutral” (SE- and “DERAA”-negative) HLA-DRB1 alleles to develop RA is 0.5–0.7, indicating that “DERAA”-positive subjects have a lower susceptibility to develop RA.58 The protective effect associated with “DERAA” is also found after stratification for the presence or absence of HLA-SE alleles. This indicates that the protective effect associated with “DERAA” expression cannot be explained by an over-representation of SE alleles in patients, resulting automatically in a lower frequency of other HLA alleles in patients with RA. Thus, the “DERAA”-containing HLA-DRB1 alleles are independently associated with a reduced risk of developing RA.5

It is unclear whether the entire “DERAA” motif is essential for the protection or whether only certain amino acids of this motif confer the same effect. In contrast to several reports showing the protective effects by “DERAA”-containing HLA-DRB1 alleles to the development and severity of RA,5 9 10 other reports hypothesise that the amino acids “RAA” at position 72–74 in the third hypervariable region influence the susceptibility to RA development, whereas the amino acids at positions 70 and 71 modulate this effect.11 12 In these articles it is indicated that HLA alleles expressing the 70ERAA74 sequence or the aspartic acid (D) at position 70 both have a lower frequency in patients with RA than in healthy controls. Further, it has also been described that protection is mainly associated with the aspartic acid (D) at position 70.8 13

Thus, despite these differences in nomenclature and stratification, it is becoming increasingly clear that some HLA alleles confer susceptibility, whereas others are associated with protection.

The mechanism of protection is unknown, but it has been proposed that it is mediated by the recognition by T cells of peptides containing the “DERAA” sequence presented by HLA-DQ molecules.14 Whether these T cells have a regulatory phenotype or are deleted in the thymus by negative selection is still a subject of research.


In 1954 Owen et al reported that Rhesus D (RhD) negative children were tolerant to the RhD antigen when they had a RhD-positive mother, probably owing to exposure to the RhD antigens during pregnancy.15 This was the first time that a biological effect of non-inherited maternal antigen (NIMA) was described. Figure 1 illustrates this terminology. Confrontation of the fetal/newborn immune system with the NIMA may have a lifelong influence on the immune response of the child. This phenomenon has considerable implications for transplantation, and most studies on NIMA are coming from the transplantation field. Claas et al reported that renal transplant patients often do not generate antibodies against the mismatched HLA antigens of their mother in comparison with those of their father and are therefore tolerant for this HLA mismatch when they are transplanted. This is associated with a longer transplant survival.1618 This is exemplified best in a study by Burlingham showing that haploidentical NIMA-mismatched sibling transplants have a graft survival similar to that of HLA-identical siblings, whereas non-inherited paternal antigen (NIPA)-mismatched sibling transplants did as poorly as the recipients of maternal and paternal grafts.19

Figure 1 Terminology of non-inherited maternal antigen (NIMA) and non-inherited paternal antigen (NIPA). The terminology is orientated from the point of view of the child. ◊ gender can be male or female.

We have recently shown that HLA-DRB1 molecules that contain the amino acid sequence “DERAA” when presented as NIMA also have a protective effect on the development of RA.20 We expected that if non-inherited “DERAA”-containing HLA-DRB1 alleles of the mother protect the child from RA development, then the prevalence of mothers of patients with RA bearing a “DERAA”-containing HLA-DRB1 allele would be lower than in the general population. Indeed, using a cohort of Dutch patients with RA together with their parents, we showed that the mothers of patients with RA had a significantly lower frequency (16.1%) of “DERAA”-containing HLA-DRB1 alleles than the Dutch control population (29.3%; p = 0.02). In contrast, the frequencies of “DERAA”-containing HLA-DRB1 alleles in fathers of the patients with RA (26.2%) and members of the healthy control group were comparable. These findings were replicated in the English multicase families from Manchester.

To further ascertain that the observed difference in frequency of “DERAA”-containing HLA-DRB1 alleles between mothers and fathers of patients with RA could indeed be attributed to an effect of non-inherited HLA-antigens, the “DERAA”-positive families with a “DERAA”-negative child (the patient with RA) were selected for further analysis. For this analysis, the patients from the UK and the Netherlands were pooled. The odds ratio (OR) for “DERAA”-negative patients with RA of having a “DERAA”-positive mother compared with a “DERAA”-positive father was 0.25 (95% CI 0.09 to 0.65; p = 0.003). These results together show that there is a protective effect of “DERAA”-containing HLA-DRB1 alleles as NIMA on development of RA in the child.

Thus, together these data indicate that both “DERAA”-containing HLA-DRB1 alleles inherited from one of the parents and the presence of “DERAA”-containing HLA-DRB1 alleles as a NIMA protect against the development of RA. The question that arises from these observations is how the strengths of both effects compare with each other. To answer this question, both effects were compared in the same set of patient and control families. Only Dutch families20 were included in this analysis for a proper comparison with the control families. The data depicted in table 1 indicate that, indeed, the effect of “DERAA”-containing HLA-DRB1 alleles as NIMA is as strong as the effect observed when the “DERAA” alleles are inherited directly from one of the parents. Although not significant (over 7000 families would be required to determine whether the inherited and non-inherited protection differ significantly or not), these data indicate that both effects are of the same magnitude. The comparable effect size described here is similar to the observations made in the transplantation setting.19

Table 1 Comparison of the inherited and non-inherited maternal antigen (NIMA) effect of “DERAA” (the Dutch families were used for this analysis)20

This result is in line with the assumption that only very few cells can exert the protective effect. One of the few cell populations that can give rise to many different cell types and has a lifelong existence is the stem cell population. During pregnancy the immune systems of mother and child are in close contact and trafficking of cells, antibodies or antigens, or both, can occur. Therefore the most plausible explanation for the observed NIMA effect of “DERAA”-containing HLA-DRB1 molecules is maternal microchimerism.

Moreover, because the NIMA effect is not taken into account in most studies analysing the contribution of the HLA system to RA susceptibility, these data also indicate that the association would be even more prominent if the effect of “DERAA” as NIMA was considered.


During pregnancy there is a bidirectional maternal–fetal lymphocytic transfer.21 Occurrence of these cells starts after about 3 months of gestation and persists until delivery.22 It is shown that the levels of fetal DNA in the circulation of the mother increase during these 6 months and disappear for the most part after delivery.23 During pregnancy, also, cells of the mother migrate to the fetus and may induce lifelong microchimerism in the child.21 24 25 Maternal microchimerism has been shown in mice to induce neonatal B-cell,26 and probably also T-cell,27 tolerance and is therefore one of the possible mechanisms for NIMA effects.28 Although speculative, we postulate, therefore, that the protective effect of the DERAA-containing HLA-DRB1 alleles as NIMA on the development of RA is most probably mediated by maternal cells entering the bloodstream and tissues of the child which exert their effect through a change in the immune repertoire and most likely the T-cell repertoire of the child. These maternal cells might influence thymic selection or act in the peripheral lymphoid organs, for example, as a consequence of the sustained presence of cells from the mother in the child. It has been shown that maternal microchimeric cells can be present in many different cell subsets29 in both healthy and diseased subjects30 31 in which they may exert different effects.32 33 Likewise, immune regulatory mechanisms might be directly induced in the fetus as it has recently been shown that the fetus can already develop cytotoxic T cells directed at a maternal minor H antigen in utero34 or becomes sensitised against foreign antigens to which the mother is exposed during pregnancy.35 Although the presence of maternal microchimerism is not rare, there are several reports that the amount of microchimerism influences the sensitivity of a person to certain diseases.30 31 36 37

The observation that the inherited and NIMA effect of “DERAA” have an approximately similar effect size strengthens the idea that the NIMA effect is caused by lifelong circulating microchimeric cells that play a role in the thymic selection and therefore influence the T-cell repertoire. Only when the inherited and the NIMA-acquired “DERAA” have the same mechanism of induction of protection, can the similar strength be explained.

Overall, we conclude from the data presented in this review that the presence of “DERAA”-containing HLA-DRB1 molecules can protect a person against the development of RA. The “DERAA”-containing HLA-DRB1 molecules can either be present because an individual has inherited them directly or because an individual has a “DERAA”-positive mother and acquired some of the “DERAA”-containing HLA-DRB1 molecules during fetal or neonatal life, or both. The protective effect that is acquired in either way is of similar strength, which suggests that already a small number of cells can initiate this protective effect. Further research is required to elucidate the mechanism of protection of both the inherited and the NIMA effect of the “DERAA”-containing HLA-DRB1 molecules. Such research might be rewarding as it may guide the way to the development of new treatments initiating protection in a manner similar to that provided by “DERAA”-positive mothers to their “DERAA”-negative children.



  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.