2 Mammary odor sources
The human nipple/areolar region is supplied with all known types of skin gland. The nipple abounds in apocrine and sebaceous glands with ducts opening on its tip and giving off secretions during lactation (Montagna and MacPherson, 1974; Perkins and Miller, 1926). Eccrine sweat glands and large sebaceous glands can also be found on the areolae (Montagna and MacPherson, 1974), the surface of which is additionally dotted with small prominences (Morgagni's corpuscles) that host underlying MG (Montgomery, 1837), composed of sebaceous glands coalesced with miniature mammary acini (Montagna and Yun, 1972; Smith et al., 1982; Vorherr, 1974). A quantitative assessment of MG prevalence, distribution, and patent activity in 121 postparturient women engaging breastfeeding (Doucet et al., 2010b; Schaal et al., 2006) indicated that 97% of (Caucasian) women bore more than 1 MG per areola, and 83%, from 1 to 20 units per areola. These MG can give off a latescent fluid that is visible in about 1 out of 5 women. This polymorphism in MG number/activity may be balanced in part by their nonrandom distribution on the areolae: secretory MG appear indeed most localized over the upper and lateral areolar quadrants, the zones toward which infants' nose are typically directed during nursing (Doucet et al., 2010b; Schaal et al., 2006).
Colostrum and milk released from main lactiferous ducts add their intrinsic olfactory qualities to the areolae. The quality/intensity of lacteal secretions are in part influenced by odorous compounds transferred from maternal diet. Empirical evidence that food aromas or experimental odorants easily pass into milk has been obtained by regularly sampling milk after mothers ingested them (Hausner et al., 2008; Mennella and Beauchamp, 1991a,b, 1996; Schaal, 2005); subsequent odor changes of milk are detectable to infants who modify their sucking pattern and consumption accordingly. In the same way, maternally inhaled odorants (e.g., combusting tobacco) can also be transferred into milk (Mennella and Beauchamp, 1998). Other potential contributions to the mammary area are imported from extramammary sources, conveyed by the newborn and the mother. Regarding the newborn's contribution, such extraneous sources labeling the areolae may be vaginal secretions and blood, amniotic fluid and vernix caseosa covering the skin of the newly born infant; tears, mucus, saliva, and saliva–milk coagulate spread at first sucking episodes. Finally, mothers may add some artificial scent as they often smear their areolae with locally prescribed emollient creams (e.g., Delaunay-El Allam et al., 2006, 2010).
Taken together, these varied sources of mammary substrates create a multifaceted and dynamic areolar odor blend. The lipid fraction from keratinizing epidermis, sebum originating from free sebaceous glands and from MG, as well as fatty acids from milk, may act as odor fixatives that improve the chemical and temporal stability of the olfactory complex formed on the areolae. The intricate arrangement of sebaceous and lacteal sources within the MG may favor the mingling of sebum and areolar milk secretion during sucking episodes. In addition, local biochemical and thermal processes may act on the selective release of given compounds or categories of compounds from the mixture. Namely, salivary enzymes deposited by the suckling infant may speed up the release of odor-active compounds (Büttner, 2002), or areolar skin temperature fluctuations may differentiate volatiles having contrasted vapor pressures.
The complexity of the human areolar–nipple area in terms of multiplicity/intricacy of sources of biological substrates renders the isolation of active odor cues/signals uneasy. A recent study (Doucet et al., 2007) attempted to evaluate whether morphologically differentiable areas of the breast could be related to distinct behavioral effects in newborns. These areas were “fractionated” in applying an odor-free plastic film directly onto the breast, various openings in it allowing to subtract the areolar odor contribution from the whole-breast odor and to separate the areolar odor from that of the nipple or of oozing colostrum/milk. The behavioral impact of the isolated areola was tested in positioning waking newborns next to such a selectively masked breast in an unconstrained nursing posture before a feed. Natural odor cues from the lactating breast modulated arousal states and promoted appetitive oral activity in 3 day olds in terms of mouth opening, tongue protrusion, rooting (see also Koepke and Bigelow, 1997, for similar results in a similar situation). The infants responded similarly to whole breast versus areola versus nipple versus milk odors, suggesting equivalent attractive potencies, all equaling that of human milk. Such behavioral uniformity of breast stimuli may be caused either by overlapping compounds due to shared exocrine sources or cross-contamination, or by perceptually distinct compounds having acquired similar attractiveness due to similar conditions of reward (Delaunay-El Allam et al., 2006). In a finer grained study (Doucet et al., 2009), 3-day-old newborns were directly exposed to native MG secretion and to its components, milk and sebum (from forehead; both stimuli not from their own mother). Neonatal reactivity to these areolar odor cues was also tested against several control stimuli (e.g., solvent, vanilla, fresh cow's milk, cow's milk-based formula). It came out that pure MG secretion equaled fresh human milk in reactogenic potency regarding the respiratory variable, but it had higher activity than human milk, sebum, and all other reference stimuli, in the elicitation of oral–facial responses.
Mainstream lacteal secretions, colostrum and transitional milk, were also tested separately as sources of active odor cues for infants ranging in age from minutes to weeks after birth. The odors of colostrum (from postpartum days 1–2) and milk (from postpartum days 3–4) were shown to elicit reliable positive head-turning responses (Marlier and Schaal, 2005; Marlier et al., 1998). They were also reported to be strong activators of facial and oral responses (mouthing, protruding tongue, rooting) indicative of their attractive value to infants born at gestational term (Russell, 1976; Soussignan et al., 1997) or preterm (Bingham et al., 2003a, 2007). The sucking pattern of infants aged 10–14 days, fed formula since birth, was quantified while being exposed to the odors of their mother's fresh milk, their familiar formula, and a neutral control (water) during a regular formula feed. It resulted that, during the first minute of exposure, the infants exposed to their mother's milk odor evinced higher sucking frequency and positive pressure, leading to a higher sucking efficiency index (i.e., milliliters milk transferred per minute; Mizuno and Ueda, 2004). Further, preliminary data indicate that the odor of colostrum or milk elicits cortical activation, but its specificity remains to be established in comparison to reference odorants (as assessed by EEG or near-infrared spectroscopy; Bartocci et al., 2000; Yasumatsu et al., 1994). Finally, the odor of human milk has been shown in some studies to dampen pain responses more than control stimuli (water or formula; although sucrose appears more efficient; e.g., Mellier et al., 1997; Nishitani et al., 2009; Örs et al., 1999; Rattaz et al., 2005; Upadhyay et al., 2004). The gist of the above results is that odors from human lacteal secretions are clearly detectable to infants aged from about 2 months before term to more than 1 month postbirth, and that they have a particular behavioral effects in terms of arousal, general attraction, appetitive actions, and self-regulatory responses.
Nevertheless, available studies are heterogeneous in many ways (infant age, sampling-conservation methods, design and reliability of testing). For example, it may be noted that, as in the case of rabbit milk described above, the time elapsing between milk sampling and odor testing has a clear influence on the attractive potency of milk odor to newborns: while milk odor remains attractive after standing 30 min at ambient temperature, its attractiveness is significantly reduced after 3 h of ambient storage (Couegnas, 2003). Thus, future efforts toward behavioral assaying and chemoanalysis of lacteal odors will have to take care of the conditions of milk sampling, conditioning, and conservation (cf. Spitzer and Büttner, 2009). So far, only a handful of attempts have been run to chemically characterize the volatile compounds of human milk (Bingham et al., 2003b; Büttner, 2007; Pellizari et al., 1982; Shimoda et al., 2000; Stafford et al., 1976), colostrum, or areolar secretion of MG (Schaal et al., 2008b). Although these analyses found a variety of common odorous compounds (n = 109: Shimoda et al., 2000; n = 5: Bingham et al., 2003b; n = 28: Büttner, 2007), underlying methods are so diverse (pooled vs. individual milk samples; fresh vs. frozen; solvent extracted vs. headspace sampled with stir bar sorptive extraction; compared to formula milk or UHT treated cow's milk) that no clear view can be drawn. More standardized studies are thus needed to built a representation of the volatile compounds (and their precursors) of milks from diverse ethnic-cultural backgrounds, of their chemical overlap, and of their specificity regarding the milk of other selected mammals. Such a comparative approach should also investigate compounds of lacteal secretions that might overlap (or not) with extramammary substrates of semiochemical interest in the nursing situation (such as amniotic fluid and infant saliva).
