Daughter Dearest?

In mammals, females dissolve parts of themselves to feed their babies. Or more science-y “mammalian females pay high energetic costs for reproduction because the synthesis of milk requires mobilization of bodily reserves to nourish developing young.”

Photo by Pascal Gagneux 

BACKGROUND

In evolutionary biology, lots of research has been dedicated to understanding how mothers direct care and nourishment to their young in relation to their own physical, social, and psychological condition and where they are at in their reproductive careers (first baby, second baby, etc). Infant characteristics may also influence how mothers invest, and the characteristic of greatest interest to evolutionary biologists has been whether they are rearing a son or a daughter. Biologists have proposed numerous hypotheses “for sex-biased maternal investment,” but the most well-known and investigated remains the Trivers-Willard hypothesis.

Trivers and Willard hypothesized that a female, depending on if she is in good or poor condition, is expected to favor a son or a daughter depending on the expected return on that investment in grandbabies. In mating systems with male-male competition, they predicted that good condition females would favor sons because males are thought to benefit more from additional investment than do daughters- more competitive males get A LOT more ladies, and therefore more babies. But what does “favoring” mean? Lots of studies have investigated the birth sex-ratio (the number of sons vs. the numbers of daughters) as well as differences in maternal behavioral care. But “biased” milk synthesis has been hardly been studied at all. This is an egregious oversight to my totally unbiased mind ;-)

There is mounting evidence that the “biological recipe” for milk produced for sons differs to some extent from milk produced for daughters. This is from only a handful of studies to date, and sex-biased milk is not nec. always present. And even with the data in hand there are some serious limitations: only a couple studies have assessed both milk “quality” (amount of fat & protein, for example) and milk quantity (how much milk is produced), only a few studies have sampled over a 100 subjects and even fewer have assessed a whole lactation or multiple lactations longitudinally.

species with some evidence of "biased" milk synthesis 

animals not really to scale

And not all the studies indicate that moms are “favoring” sons or that sex bias depends on maternal condition. In rhesus monkeys I found that mothers, of all conditions- synthesized more milk on average for daughters at peak lactation (Hinde 2009). One super cool and elegant experiment in bank voles involved cross-fostering newborn pups among dams. The researchers created all-male or all-female liters that were larger (+2) and smaller (-2) than the litter that the vole mom had gestated (Koskela et al. 2009). This was a “calibrated” way to see if mothers with high energy expenditure vs. low energy expenditure were favoring sons and daughters differently. Across the board, vole moms rearing all-daughter litters synthesized more milk. (side note, rodent models are so freaking awesome). These post natal effects may have been mediated by behavioral care and nursing patterns that favored daughters. 

But what about a PRE-NATAL effect from the fetal daughter or son? This is an important question because the functional development of the mammary gland occurs during pregnancy. When dairy cows and goats are gestating twins, mammary gland development is amplified due to hormonal signals from the much larger fetal-placental unit. Carrying twins seemingly programs higher milk production to meet the needs of “double the trouble” (Nielen et al. 1989, Hayden et al. 1979). But what if fetal-placental signals aren’t just about the number of offspring, what if other features are signaled that influence milk production? Features like infant sex/gender.

THE STUDY

Just over a year ago I wrote a blog post all about “Boy Milk & Girl Milk,” explaining that it was unclear if the differences emerged due to pre-natal hormonal cues and/or post-natal nursing interactions. A few weeks later I get a twitter DM from @Anim_Nutr. It was Kansas State University professor Barry Bradford suggesting that we could look for pre-natal effects of daughters and sons using dairy cows and wanted to know if I would be interested in a collaboration. As the consummate professional I am, my tweet reply was "hell yeah!" or "shit yeah!" or some other totally inappropriate exclamation, we took the conversation over to gmail and designed our study (this study brought to you in part by online social networking; hat tip to twitter, blogspot, & gmail). Subsequently Gail Carpenter, graduate student and giant dataset programming extraordinaire, and John Clay, Director of the Dairy Records Management Systems, joined the project.

We used 2.39 million lactation records from 1.49 million Holstein dairy cows from 1995 to 1999 to investigate whether the sex of the fetus influences the capacity of the mammary gland to synthesize milk during lactation. Standardized husbandry practices, systematic milking procedures, detailed record-keeping, and large sample sizes make the dairy cow a powerful model for the exploration of milk synthesis from both mechanistic and evolutionary perspectives. Notably, calves are removed from the dam shortly after birth, allowing us to specifically investigate pre-natal mechanisms of sex-biased milk synthesis independent of post-natal maternal care and suckling behavior. Importantly, bulls are much bigger than heifers, and aurochs, the ancestor of cows, were characterized by male-male competition for access to mates so cows make a good candidate for investigating potential sex-biased maternal investment.

Holstein dairy cows. Photo by Scott Bauer/USDA

Holsteins biased milk production in favor of daughters, not sons. Across a standard 305-day lactation, total milk production was significantly higher after gestating a daughter, ranging between ~100-150 kgs more milk across lactation (220-310 lbs). This effect was strongest in cows on their first lactation, likely because substantial changes occur in the mammary gland the first time it “gears up” for milk synthesis. The percent fat and protein in milk did not differ between cows that had gestated a son or daughter, so the “quality” of milk was the same. But because the “quantity” was higher after gestating a daughter, the total kilograms of milk fat and protein after gestating a daughter were higher than after gestating a son.

But fetal sons and daughters don’t just affect the current lactation. Dairy cows are often concurrently pregnant and lactating, typically 200+ days of the 305-day lactation. We restricted a new analysis to a smaller conservative, longitudinal dataset of the first and second lactations for individual cows (N=113,750 cows) with no cases of dystocia (calving score 1 or 2), and no reported administration of hormones to increase milk production. Fetal sex interacted dynamically across lactations, in part because the second pregnancy overlapped with the first lactation. Cows that gestated a son and then another son synthesized significantly less milk on the first lactation. Cows that first gestated a son and then a daughter on the second pregnancy, could partially “rescue” milk production, but remained substantially lower than for cows that had a daughter on their first pregnancy. Additionally cows that had a son on their first pregnancy were handicapped in their milk production even on their second lactation, especially if they also had a son on their second pregnancy.  Specifically, cows with two daughters back-to-back produced ~445 kgs (~980 lbs) more milk across the first two lactations than did cows with back-to-back sons.

Yeah it’s super complicated so we made a “conceptual model”

WHAT DOES THIS MEAN?

For the General Public: Our results, for the first time, identify adramatic and sustained programming of mammary function by offspring sex in utero. Maternal nutrition and hormones are known to have pronounced and long-term effects on offspring, but the ways in which offspring has sustained physiological effects on mom are less known. How does the fetus mechanistically influence milk synthesis? It’s likely that hormones from the fetus and placenta may differ between fetal sons and daughters, subsequently enter the maternal bloodstream and affect the milk producing cells in the mammary gland.

zOMG Baby

Importantly, no study has yet measured differences in milk volume for sons and daughters, but in the last couple years there have been a handful of studies among humans reporting differences in milk composition between sons and daughters (Powe et al. 2010; Fujita et al. 2011; Thakkar et al. 2013; bit see Quinn 2013 for no differences in milk for sons and daughters). Humans have a very invasive placenta that would allow fetal hormones to pass into maternal circulation and possibly influence mammary gland development. But it hasn’t yet been systematically studied. One can only speculate about the "mixed" signals from fraternal multiples of different genders- is it a wash? or double daughter twinsies, is it extra amplified? Who knows! 

This research in cows demonstrates that the fetus can influence the milk the mother produces during lactation and limited evidence suggests that similar processes may be operating in humans. We need to systematically understand if breast milk reflects the differing developmental priorities of sons and daughters. Such work has potential implications for nutrition management of babies in neonatal intensive care units and selection of donor milks. And such research can improve infant formulas tailored more specifically to the physiological needs of sons and daughters. To put it into perspective, we make deodorants specifically for men and women, why aren’t we talking about sex-specific early life nutrition?

For the Dairy Farmers: Dairy herd management decisions can be informed by the effects we report here Artificial insemination is standard practice in dairying and sex-selected semen is widely used. The use of sexed semen with a virgin heifer has additional economic value in terms of long-term milk production. In cows whose first pregnancy yielded a bull calf, the use of sexed semen creates additional economic benefit in terms of milk production. In cows whose first pregnancy yielded a heifer calf, the use of sexed semen does not seem to create additional economic benefit in terms of milk production. Rough “back of the napkin” calculations, taking into account the current wholesale value of milk, the number of 2-year old heifers added to US dairy herds annually, the production advantage across the first two lactations of conceiving a daughter on the first pregnancy, and the increased probability of conceiving a daughter from sex-selected semen, suggests a gross value in the neighborhood of $200 million in milk production across the first two lactations in the United States alone. Organic dairy farmers, because they do not use synthetic hormones to increase milk production, may be especially keen to consider techniques to influence the conception of daughters.

zOMG Cow

Conventional wisdom in dairy management has really emphasized how dystocia and energy balance influence milk production, which we considered in our manuscript. Although gestation of a son increases risk of dystocia, in our conservative sample (N=113,750) we restricted the sample to cows with the easiest labors. We retained only lactations following a calving score of 1 (no problem) or 2 (slight assistance), extreme difficulty is a calving score of 5, but just to be safe, since milk produce decreases with increasing difficulty, we didn’t include the 3s or 4s either  (Dematawena and Berger 1997). The higher milk production after gestating a daughter remained.

Perhaps the costs of gestating a son just take more energy, leaving less for milk production. This doesn’t stick either. In our study we showed that reduced milk synthesis from gestating a son on the first pregnancy is partially recovered on lactation 1 if the cow is gestating a daughter on the second pregnancy. Now if having a son was so costly in terms of energy and calving difficulty, the signature of that should persist but it doesn’t, not on lactation 1 if the cow is simultaneously gestating a fetal daughter. And not on lactation 2, months or years later. The signature of a fetal son or daughter on the first pregnancy continues to leave a mark on lactation 2, with first-born daughters enhancing and first-born sons decreasing milk production even after a second offspring is conceived & delivered and 305 days of milking on lactation 2 has ensued. Importantly, this “sex-biased” milk did not occur when farmers used bST (recombinant bovine somatropin) on multiparous cows, indicating that bST exerts a stronger effect on the mammary gland of a cow that gestated a son. But bST use for cows on the first lactation did not “overcome” the effect of fetal daughters. Taken together these results suggest that cows on their first pregnancy are particularly sensitive to fetal sex and that these effects are mediated through hormonal pathways to directly affect the mammary gland.

Yes, dystocia matters... a lot. Yes, energy balance matters… a lot.

But those can not, on their own, explain the phenomena that we are describing here. Something more is going on, and we think it is how fetal hormones affect the mammary gland architecture during functional development of the mammary gland.

For the Evolutionary Biologists

Why favor daughters? One evolutionary explanation is that mammalian mothers may have physiological adaptations to respond to fetal daughters. More milk for daughters may allow them to mature faster and initiate reproduction at relatively younger ages than do sons. A daughter’s lifetime reproductive output, a target of natural selection, is limited by time because pregnancy and lactation are of a largely fixed duration. In species in which males compete for access to females, males generally have a longer period of development before reproducing, so they have more time to compensate for deficits in the milk they received from their mother. As such mothers may be able “under-invest” in a son with relatively less consequence for the number of offspring he will go on to sire.

A bull can slow play entering the reproductive fray, 

a heifer is sunk if she waits for the river.

These results suggest that uniformly biased milk production favors daughters across maternal conditions- using parity as a proxy for condition- does not support the Trivers-Willard hypothesis. This opens the door for more discussion of the many other hypotheses about sex-biased investment: local resource competition, “advantaged daughters”, local resource enhancement, the “safe bet”/reproductive value, and sex-differentiated sources of mortality (see further reading list below). TWH needs to stop headlining the tour and get sent to the state fair circuit where it belongs!

zOMG Darwin

Just kidding. But seriously, we need to stop letting TWH dominate the interpretation of data on differentiated care/effort and offspring outcomes. Not only because TWH can be a competitive inhibitor for considering other hypotheses, but also because these investigations prioritize the role of the mother and sex-biased allocation of maternal resources, underestimating offspring as agents in their own development. Our results may NOT be adaptive maternal investment.  Mechanisms within offspring may be sexually-differentiated. In this way, daughter-biased milk synthesis may not necessarily reflect adaptations in the mother, but instead may be adaptations in the fetus. Daughters may be better able to send hormonal signals to affect the mammary gland, whereas sons are constrained in their production of such hormones. Or maybe there are adaptations operating in both a la Parent-Offspring Conflict (Trivers 1974). And these results MAY NOT EVEN BE ADAPTIVE! The effect of fetal daughters on milk production may just be a constraint of placental physiology and mammary gland development. 

As per usual, more research is needed. Luckily as scientists we are never in need of questions to study, just grant funding with which to do it. #amiright #oramiright 

Further Reading:

Hinde K, Carpenter AJ, Clay JS, Bradford BJ (2014) Holsteins Favor Heifers, NotBulls: Biased Milk Production Programmed during Pregnancy as a Function ofFetal Sex. PLoS ONE 9(2): e86169. doi:10.1371/journal.pone.0086169

Clark AB (1978) Sex ratio and local resource competition in a prosimian primate. Science 201(4351): 163-165.

Cockburn, A., Legge, S., & Double, M. C. (2002). Sex ratios in birds and mammals: can the hypotheses be disentangled. Sex ratios: concepts and research methods, 266-286.

Dematawena, C. M. B., & Berger, P. J. (1997). Effect of dystocia on yield, fertility, and cow losses and an economic evaluation of dystocia scores for Holsteins. Journal of dairy science, 80(4), 754-761.

Emlen ST, Emlen JM, Levin SA (1986) Sex-ratio selection in species with helpers-at-the-nest. Am Nat 127(1): 1-8.

Fujita, M., Roth, E., Lo, Y. J., Hurst, C., Vollner, J., & Kendell, A. (2012). In poor families, mothers' milk is richer for daughters than sons: A test of Trivers–Willard hypothesis in agropastoral settlements in Northern Kenya. American journal of physical anthropology, 149(1), 52-59.

Hayden, T. J., Thomas, C. R., & Forsyth, I. A. (1979). Effect of number of young born (litter size) on milk yield of goats: role for placental lactogen. Journal of dairy science, 62(1), 53-57.

Hinde, K., Carpenter, A., Clay, J., Bradford B. 2013. Holsteins Favor Heifers, Not Bulls: Biased Milk Production Programmed during Pregnancy as a Function of Fetal Sex. PLOS One.

Koskela, E., Mappes, T., Niskanen, T., & Rutkowska, J. (2009). Maternal investment in relation to sex ratio and offspring number in a small mammal–a case for Trivers and Willard theory?. Journal of Animal Ecology, 78(5), 1007-1014.
Leimar O (1996) Life-history analysis of the Trivers and Willard sex-ratio problem. Behav Ecol 7(3): 316-325

Nielen, M., Schukken, Y. H., Scholl, D. T., Wilbrink, H. J., & Brand, A. (1989). Twinning in dairy cattle: a study of risk factors and effects. Theriogenology,32(5), 845-862.

Powe, C. E., Knott, C. D., & Conklin‐Brittain, N. (2010). Infant sex predicts breast milk energy content. American Journal of Human Biology, 22(1), 50-54.

Quinn, E. A. (2013). No evidence for sex biases in milk macronutrients, energy, or breastfeeding frequency in a sample of filipino mothers. American journal of physical anthropology, 152(2), 209-216.

Shibata F, Kawamichi T (2009) Female-biased sex allocation of offspring by an Apodemus mouse in an unstable environment. Behav Ecol Sociobiol 63(9): 1307-1317.

Silk JB (1983) Local resource competition and facultative adjustment of sex ratios in relation to competitive abilities. Am Nat 121(1): 56-66.

Silk JB, Brown GR (2008) Local resource competition and local resource enhancement shape primate birth sex ratios. Proc Biol Sci 275(1644): 1761-1765.

Simpson MJ, Simpson AE (1982) Birth sex ratios and social rank in rhesus monkey mothers. Nature 300(5891): 440-441.

Smith JM (1980) A new theory of sexual investment. Behav Ecol Sociobiol 7(3): 247-251.

Thakkar, S. K., Giuffrida, F., Cristina, C. H., Castro, C. A., Mukherjee, R., Tran, L. A., ... & Destaillats, F. (2013). Dynamics of human milk nutrient composition of women from singapore with a special focus on lipids. American Journal of Human Biology, 25(6), 770-779.

Trivers, R. L. (1974). Parent-offspring conflict. American zoologist, 14(1), 249-264.

Trivers RL, Willard DE (1973) Natural selection of parental ability to vary the sex ratio of offspring. Science 179(4068): 90-92.