Maternally derived anti-helminth antibodies predict offspring survival in a wild mammal

The transfer of antibodies from mother to offspring provides crucial protection against infection to offspring during early life in humans and domestic and laboratory animals. However, few studies have tested the consequences of variation in maternal antibody transfer for offspring fitness in the wild. Further, separating the immunoprotective effects of antibodies from their association with nutritional resources provided by mothers is difficult. Here, we measured plasma levels of total and parasite-specific antibodies in neonatal (less than 10 days old) wild Soay sheep over 25 years to quantify variation in maternal antibody transfer and test its association with offspring survival. Maternal antibody transfer was predicted by maternal age and previous antibody responses, and was consistent within mothers across years. Neonatal total IgG antibody levels were positively related to early growth, suggesting they reflected nutritional transfer. Neonatal parasite-specific IgG levels positively predicted first-year survival, independent of lamb weight, total IgG levels and subsequent lamb parasite-specific antibody levels. This relationship was partly mediated via an indirect negative association with parasite burden. We show that among-female variation in maternal antibody transfer can have long-term effects on offspring growth, parasite burden and fitness in the wild, and is likely to impact naturally occurring host–parasite dynamics.


= (sample OD − blank OD) (positive control OD − blank OD)
Where the numerator was set to zero if the blank OD was greater than the sample OD in order to avoid negative values. The number of samples that failed quality control per assay was 16 for anti-T.
Our neonatal antibody dataset initially comprised of 3,491 lambs which had blood samples taken in the lambing seasons between 1990 and 2015. No April samples or data were collected in 2001 due to foot-and-mouth disease precautions. Five individuals had two samples, and these duplicates were removed, taking the first measurement if samples were collected on different days, or picking one of the duplicates at random if not. One lamb born late in the season (July) was removed from the dataset.
We further restricted analyses to neonates caught within 10 days of birth for three reasons: first, to accurately estimate birth weight from capture weight since the relationship between age and capture weight changes after this; second, due to a sharp decline in antibody levels in older lambs ( Figure S1); and third, to account for low sample sizes in older capture age groups (n = 85). Our final dataset comprised of neonatal antibody measures for 3,379 lambs from 845 mothers, 52% of the lambs were female and 24% were twins.
Mammals have five classes (isotypes) of antibodies: IgA, IgE, IgD, IgG and IgM. IgG is the most abundant isotype in serum, and is effective at neutralising toxins, activating complement and the agglutination and opsonisation of pathogens. IgM is the second-commonest isotype in serum, and is the major isotype produced following first exposure to an antigen. IgM forms pentamers making them effective activators of the complement system but prevents them readily leaving the bloodstream. The third most common isotype in serum is IgA, which is the most common isotype in secretions including mucosal surfaces and it acts as a strong neutralising antibody and can agglutinate pathogens.
IgD can be found on B cells and this isotype tends not to be secreted. Finally, IgE is present at very low concentrations in serum, has the lowest half-life of all antibodies, but is key in mediating acute inflammatory responses. Colostrum is rich in immunoglobulins, predominantly IgG in ewes, followed by IgM and IgA. In milk, the concentration of immunoglobulins drops and IgG continues to predominate but IgA becomes the second most dominant isotype [3]. In peri-parturient ewes in this population we found uniformly positive correlations between anti-Tc IgG, IgA, IgM and total IgG, IgA and IgM, suggesting that maternally-transferred antibody levels are likely to be positively correlated in neonates [4].      (Table S3). Boxes show the median and the interquartile range (IQR) with whiskers extending from the hinges to values no further than 1.5*IQR from the hinge (outliers not shown). Figure S6. Associations between birth weight (corrected for capture age, see Methods) and neonatal anti-T. circumcincta IgA, IgE and IgG and total IgG levels. Plots show raw data with LMM predictions and standard errors estimated for female singleton lambs with average values for all continuous fixed effects in the minimal model (Table S3).  (Table S3).     Table S4. AIC comparisons of models investigating different capture age relationships with antibody levels in neonatal Soay sheep. Different capture age relationships include linear and quadratic terms, plus threshold age models with a single threshold at day 1 to 9 ("t=") for anti-T. circumcincta IgA, IgE, IgE and total IgG levels. Models included birth weight, maternal age (quadratic), sex and twin status of the lamb as fixed effects in addition to maternal identity, year and plate as random effects. The best fitting models with the lowest AIC for each isotype are highlighted in bold.