Mannan binding lectin concentration and risk of miscarriage

D.C. Kilpatrick1,3, L. Starrs2, S. Moore2, V. Souter2 and W.A. Liston2

1 Academic Unit, Department of Transfusion Medicine, 2 Forrest Road, Edinburgh, EH1 2QN, and 2 Gynaecology Clinic, The Royal Infirmary, Edinburgh, EH3 9YW, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Recurrent miscarriage is associated with low concentrations of mannan-binding lectin (MBL), but it is not known below which value relative MBL deficiency becomes a significant risk factor. The sera of 397 patients (male and female) suffering from recurrent miscarriage and 376 controls were assayed for MBL and the data analysed. It was found that the lower the cut-off value, the greater the statistical strength of the association. It was concluded that only MBL concentrations <=0.1 µg/ml were clinically significant in this context. A corollary of this conclusion is that genotyping for point mutations in the structural gene for MBL would be a much less sensitive means of identifying couples at risk of experiencing recurrent miscarriage.

Key words: mannan binding lectin/recurrent miscarriage/recurrent spontaneous abortion


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
We have previously reported an association between idiopathic recurrent miscarriage and low concentrations of mannan-binding lectin (MBL) (Kilpatrick et al., 1995Go). This relationship has since been confirmed in Danish women with recurrent miscarriage, although it was not evident in the small group of Danish male partners available for study (Christiansen et al., 1999Go). Relative MBL deficiency was not apparent in a small group of women whose recurrent miscarriages could be explained by autoimmunity, chromosomal abnormalities, etc, or in women with an unrelated gynaecological complaint, endometriosis (Kilpatrick et al., 1996Go). Additionally, another calcium-dependent opsonic lectin in serum, P35, has been found at unusually low concentrations in a small proportion of patients with recurrent miscarriage (Kilpatrick et al., 1999Go). These findings are consistent with the concept that a relative deficiency of innate immune factors predisposes to pregnancy loss, perhaps secondary to infection.

Over the last few years, much more has been learnt about the genetics of MBL and it is now clear that the assumptions we originally made about the genetic basis of MBL concentrations detectable by enzyme-linked immunosorbent assay (ELISA) were of limited accuracy. MBL concentration is determined by a combination of complex haplotypes and is influenced not just by mutations in the structural gene, but also by two dimorphic loci in the promoter region (Madsen et al., 1995Go). Nevertheless, the presence of one of several possible point mutations in the structural gene has the greatest influence on phenotype. Consequently, ~90% of individuals who are homozygous for the wild-type structural gene have MBL concentrations >0.6 µg/ml, whereas the majority of heterozygotes (one haplotype bearing one of the possible structural point mutations) have circulating MBL concentrations <0.6 µg/ml (Madsen et al., 1994Go, 1995Go).

A total of 397 patients from Edinburgh have now been tested and the data used to try and establish the most meaningful, clinically relevant cut-off value to identify couples with an increased risk of pregnancy failure.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Patients
A total of 397 patients (218 females and 179 male partners) from Edinburgh, Scotland, who had experienced recurrent miscarriage (defined as three or more consecutive miscarriages without known cause), and 376 blood donors (controls) were tested. In addition, we tested 94 patients classified as having a `bad obstetric history'; this is a heterogeneous group which does not fit the criteria for recurrent miscarriage (e.g. just two miscarriages with the same partner, three or more miscarriages but not in consecutive pregnancies, etc). Miscarriages were considered to be unexplained if both parents had a normal karyotype and the female partner was negative for anti-phospholipid antibodies; hysterosalpinography and tests of uterine cervical resistance were not routinely performed.

Enzyme-linked immunosorbent assay
MBL concentrations were measured by a sandwich ELISA as previously described (Kilpatrick et al., 1995Go). Briefly, sera were incubated in mannan-coated wells of an ELISA plate in the presence of Ca2+, and, after washing, a murine monoclonal anti-MBL antibody was added to the wells. Bound immune complexes were visualized using anti-mouse immunoglobulin conjugated to alkaline phosphatase followed by p-nitrophenyl phosphate as substrate.

Statistical analyses
These were carried out using Prism for Windows software from Graph Pad (San Diego, CA, USA).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Any cut-off concentration is to some extent arbitrary, but we have analysed our data in relation to three thresholds, each of which can be logically justified: (i) <=0.01 µg/ml is the sensitivity limit of the assay and represents the virtually undetectable; (ii) <=0.05 µg/ml appeared to distinguish a discrete population of MBL deficient subjects evident in scattergrams of both Danish and Scottish blood donors (Christiansen et al., 1998); (iii) <=0.6 µg/ml provides a substitute for genotyping, since the two-thirds of the population above that value are mostly homozygotes for the wild-type structural gene while the other third of the population consists mainly of heterozygotes.

The lower the cut-off value, the stronger was the association of MBL deficiency with recurrent miscarriage, although the most significant difference statistically was at <=0.05 µg/ml (Table IGo). Examination of the odds ratio (OR) as a function of different thresholds revealed a biphasic relationship, with a steep curve down to a concentration of ~0.2 µg/ml, after which it levelled off completely (data not shown). OR above unity below the median value were entirely due to the contribution of values <=0.1 µg/ml.


View this table:
[in this window]
[in a new window]
 
Table I. Recurrent miscarriage (RM) risk in relation to mannan-binding protein (MBL) concentrations
 
The strengths of association were similar for male and female partners of recurrent miscarriage couples: for female patients the proportions with MBL <=0.01, <=0.05 and <=0.6 µg/ml were 9.2, 12.8 and 33.5% respectively; the corresponding figures for their male partners were 8.4, 12.8 and 35.8%. The heterogeneous group of couples with a bad obstetric history had almost as strong an association with MBL deficiency; at the same three cut-off values the percentages were 7.4% (OR = 1.81), 11.7% (OR = 1.71) and 34% (OR =1.30), although statistical significance was not obtained with the smaller numbers.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
These results confirm and extend our original observation that low concentrations of MBL are over-represented in both male and female partners of couples experiencing recurrent miscarriage. It is now apparent that only the lowest concentrations of MBL are associated with recurrent miscarriage. The reason for this association is not known, but it is reasonable to assume that a MBL-deficient fetus may be more susceptible to sub-clinical infectious and/or inflammatory events in utero.

In conclusion, we suggest that only MBL values <=0.1 µg/ml should be considered as a clinically-significant risk factor for spontaneous abortion. It follows that investigations of miscarriage patients only by genotyping for the presence of a mutant structural gene are unlikely to provide statistically significant results unless very large groups of patients and controls are analysed. However, for individual couples, genotyping (in addition to measuring the protein concentrations) could provide useful information by establishing zygosity. Only by genotyping both potential parents can the risk to the fetus of inheriting a genotype encoding a low MBL phenotype be accurately assessed, and that may be the crucial factor.


    Notes
 
3 To whom correspondence should be addressed Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Christiansen, O.B., Kilpatrick, D.C., Souter, V. et al. (1999) Mannan-binding lectin deficiency is associated with unexplained recurrent miscarriage. Scand. J. Immunol., 49, 193–196.[ISI][Medline]

Kilpatrick, D.C., Bevan, B.H. and Liston, W.A. (1995) Association between mannan binding protein deficiency and recurrent miscarriage. Mol. Hum. Reprod., 1, see Hum. Reprod., 10, 2501–2505.

Kilpatrick, D.C., Bevan, B.H., Liston, W.A. et al. (1996) Lectins in the human feto–placental unit: a review. In Van Driessche, E. et al. (eds), Lectins–Biology, Biochemistry, Clinical Chemistry. Vol. 11. Textop, Hellerup, pp. 161–167.

Kilpatrick, D.C., Fujita, T. and Matsushita, M. (1999) P35, an opsonic lectin of the ficolin family, in human blood from neonates, normal adults and recurrent miscarriage patients. Immunol. Lett., 67, 109–112.[ISI][Medline]

Madsen, H.O. Garred, P., Kurtzhals, J.A.L. et al. (1994) A new frequent allele is the missing link in the structural polymorphism of the human mannan-binding protein. Immunogenetics, 40, 37–44.[ISI][Medline]

Madsen, H.O., Garred, P., Thiel, S. et al (1995) Interplay between promoter and structural gene variants control basal serum level of mannan-binding protein. J. Immunol., 155, 3013–3021.[Abstract]

Submitted on January 1, 1999; accepted on April 15, 1999.