Brown adipose tissue (BAT) possesses marked metabolic and thermogenic properties. In humans, BAT is especially abundant in newborns (1). BAT secretes regulatory proteins, termed brown adipokines or batokines, which hold autocrine, paracrine and endocrine attributes (2). The identification of batokines that could serve as circulating biomarkers of BAT activity could be useful in clinical practice. In this regard, we recently showed that BAT activity and the levels of the batokine C-X-C motif chemokine ligand 14 (CXCL14) -a purported surrogate of BAT activity- are sex-specific in the first year of life (3).

Bone morphogenetic protein-8B (BMP8B) is a member of the transforming growth factor β superfamily (TGF-β), a large group of regulatory factors involved in embryo development and adult tissue homeostasis. Experimental data in adult rodent models has revealed a thermogenic role for BMP8B, which has been demonstrated to be produced in BAT as a brown adipokine and to induce thermogenesis by acting both peripherally and centrally (4, 5). Other data in experimental models have shown that, despite BMP8B expression is minimal in healthy liver, hepatic BMP8B expression is increased under conditions of non-alcoholic steatohepatitis (6, 7). Recently, blood BMP8B levels have been found to be positively associated with body temperature and thermal stress-induced BAT activity in human neonates, thus indicating that BMP8B may be relevant for thermoregulation in human early life (8). It is known that BAT is highly present in the first years of human postnatal development (1), and considering that BAT is a major site of BMP8B release in rodent models (4), it could be hypothesized that BAT may be a major source of circulating BMP8B in human neonates. Thus, BMP8B levels might be a biomarker indicative of the extent of BAT activity in early life and, considering the association of BAT activity and a healthy metabolic status in adults (9), BMP8B could also reflect postnatal metabolic status.

Here, we determined for the first time the longitudinal outcome of circulating BMP8B levels over the first year of life of humans and analyzed their association with endocrine and metabolic parameters as well as with BAT activity.

The anthropometric data in the mothers and the longitudinal infant data split by sex are summarized in Table 1. The infants’ results confirmed the previously reported sex-specificity of lean mass, CXCL14 and BAT activity (3).

Figure 1 depicts circulating BMP8B concentrations in the infants from birth to age 12 months, as well as in the mothers of those infants in the third trimester of gestation and in healthy adults. Over the first year of life, circulating BMP8B levels decreased. When pooling BMP8B concentrations independently of sex, serum BMP8B levels were significantly higher at birth and at age 4 months as compared to age 12 months (P < 0.001) and remained higher versus those in healthy adults at all time points. In addition, serum BMP8B concentrations in pregnant women in late gestation were significantly lower than those in healthy adults or in human infants (P < 0.001) (Figure 1A). When BMP8B levels were split by sex, boys showed significantly higher concentrations than girls only at birth (P = 0.04) (Figure 1B).

Bone morphogenetic protein-8B levels decreased in a different manner from birth to age 12 months in girls and in boys (Figure 2). While in girls BMP8B levels remained elevated from birth to 4 months and dropped significantly between age 4 and 12 months (Figure 2A), boys experienced a progressive decrease throughout the first year of life (Figure 2B). Indeed, significant differences between BMP8B concentrations at birth and at 4 months were only detected in boys. Interestingly, BMP8B levels were still elevated 36 h after birth in both girls (928 ± 390 pg/ml) and boys (1,656 ± 390 pg/ml) and tended to be higher in boys.

The associations between circulating BMP8B concentrations and clinical, endocrine-metabolic, body composition and BAT activity variables are depicted in Supplementary Table 2. At 4 months, circulating BMP8B levels significantly correlated with bone mineral density (R = −0.383; P = 0.031) in the entire population, and with serum glucose levels (R = 0.467; P = 0.042) only in girls. In addition, the decrease in BMP8B levels in the entire population over the first year of life positively correlated with the increase in lean mass (R = 0.412; P = 0.036).

The expression levels of BMP8B mRNA in BAT, WAT and liver samples from human neonates and in human placenta and umbilical cord blood cells are shown in Figure 3. BMP8B expression was found to be significantly higher in BAT as compared to WAT, liver and cord blood cells.

The present study constitutes the first one reporting the levels of BMP8B–a brown adipokine involved in thermoregulation and metabolic homeostasis recently identified in experimental studies-across the first year of human development. The highest levels of BMP8B were found at birth, and declined progressively throughout the first year of life, where they were still above adult levels. This pattern is consistent with the especially high requirements of thermoregulation and high BAT activity in neonates early after birth (1). In fact, it is known that maintenance of normothermia at birth is of critical importance, as hypothermia is a challenging and not infrequent condition in newborns (15). Our findings are also consistent with a recent report indicating a positive correlation between BMP8B levels and BAT activity in neonates subjected to thermal stress stimulation in the first 2 days of life (8). Indeed, we found that BMP8B gene expression is the highest in BAT among other neonatal tissues analyzed, which provides the first direct evidence that BAT may be a potential major source of BMP8B in human neonates.

Interestingly, BMP8B levels at birth were higher in boys than in girls, and this trend persisted for at least the next 36 h. This may be in part related to reports of a lower body temperature in boys at birth (16) which may result in enhanced thermal stress, high BAT activation and subsequent BMP8B release. Most reports point to higher BAT activity in adult women than in adult men (17) and there are experimental data indicating that estrogens activate BAT thermogenesis (18) and BMP8B gene expression (19). Along these lines, we disclosed the presence of higher posterior-cervical BAT activity in 1 year-old girls relative to boys (3). In contrast, studies on sex differences in BAT activity in pre-school children have provided heterogeneous conclusions (20–22). Further research is warranted as there are currently no data on sex-related differences in BAT activity in human neonates in the first months of postnatal life.

We did not find significant correlations between BAT activity and BMP8B levels at age 1 year. This may indicate that the reported association between BMP8B levels and inducible BAT activity in 1–2-day-old neonates (8) is restricted to this very early development period characterized by intense thermoregulatory needs and does not happen later in development. It is also possible that the association between BMP8B and BAT activity is evidenced only when BAT activity is determined under acute induction of thermal stress, as in Urisarri et al., (8), and not in basal conditions. Moreover, we did not find significant associations between BMP8B levels and the main auxological, metabolic and endocrine parameters across the first year of life. Indeed, only marginal associations were found between BMP8B levels and each bone mineral density and gains in lean mass in the whole population; these associations could be reminiscent of the role of BMP8B, as member of the bone morphogenetic protein sub-family, in the ossification process (23) and also reflect the effects of gonadal steroids on both lean mass and circulating BMP8B levels (19). Most likely, though, the lack of strong associations could be due to the narrow range of variation in metabolic and endocrine parameters as well as in BAT activity among individuals in the first year of life under healthy conditions. In any case, and according to our findings, variations in BMP8B levels do not appear to be particularly informative of the individual metabolic status and BAT activity in apparently healthy human infants in the first year of life.

An additional noteworthy result in our study is the intense downregulation of BMP8B levels observed during late pregnancy relative to BMP8B levels in non-pregnant women. It has long been known that the thermogenic activity of BAT is decreased at term gestation in rodent models (24, 25). Although no such data on BAT activity are currently available in human pregnancy, low BMP8B levels in pregnant women may reflect reduced BAT thermogenic activity and subsequently reduced BMP8B secretion. On the other hand, the very low BMP8B levels in pregnant women suggest that the substantial expression of BMP8B mRNA in placenta does not result in relevant release of BMP8B to maternal circulation.

The current study has several limitations mostly related to obvious ethical restrictions for the invasiveness of some analytical procedures in healthy neonates. For example, BAT activity data couldn’t be obtained in unrestrained neonates immediately after birth and tissue BAT sampling had to be limited to necropsies from severely ill neonates and was obviously unfeasible in healthy 1-year old infants. However, the longitudinal nature of the study, with extensive information about endocrine-metabolic and body composition parameters to be related to BMP8B levels is a strength of this study, which is to our knowledge the first reporting the outcome of circulating BMP8B levels in humans in the first year of life.

In conclusion, changes in BMP8B levels appear to reflect overall changes in BAT activity in development. However, circulating BMP8B concentrations do not seem to be particularly useful as biomarker of individual variations in the metabolic-endocrine status and BAT activity of healthy infants in the first year of life.

CG-B and LI are clinical investigators of CIBERDEM (Centro de Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas Asociadas, Instituto de Salud Carlos III, Madrid, Spain). AL-B is a clinical investigator of the I3 Fund for Scientific Research (Ministry of Science and Innovation, Spain). FV is an ICREA Academia researcher.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by the Institutional Review Board of University of Barcelona, Sant Joan de Déu University Hospital. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

CG-B contributed to literature research, design of figures and tables, data collection, analysis, and interpretation, and wrote the manuscript. JV, CP, AG-N, and RC contributed to data collection, analysis, and interpretation. PC contributed to data analysis and interpretation. FZ and AL-B contributed to data interpretation, and reviewed and edited the manuscript. LI and FV contributed to study design, data interpretation, wrote the manuscript, and reviewed and edited the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer ML, declared past co-authorship with the authors AG-N, RC, and FV, and the absence of any ongoing collaboration with any of the authors to the handling editor.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.