Samantha R. Weaver

Autocrine-paracrine regulation of the mammary gland1

The mammary gland has a remarkable capacity for regulation at a local level, particularly with respect to its main function: milk secretion. Regulation of milk synthesis has significant effects on animal and human health, at the level of both the mother and the neonate. Control by the mammary gland of its essential function, milk synthesis, is an evolutionary necessity and is therefore tightly regulated at a local level. For at least the last 60 yr, researchers have been interested in elucidating the mechanisms underpinning the mammary glands ability to self-regulate, largely without the influence from systemic hormones or signals. By the 1960s, scientists realized the importance of milk removal in the capacity of the gland to produce milk and that the dynamics of this removal, including emptying of the alveolar spaces and frequency of milking, were controlled locally as opposed to traditional systemic hormonal regulation. Using both in vitro systems and various mammalian species, including goats, marsupials, humans, and dairy cows, it has been demonstrated that the mammary gland is largely self-regulating in its capacity to support the young, which is the evolutionary basis for milk production. Local control occurs at the level of the mammary epithelial cell through pressure and stretching negative-feedback mechanisms, and also in an autocrine fashion through bioactive factors within the milk which act as inhibitors, regulating milk secretion within the alveoli themselves. It is only within the last 20 to 30 yr that potential candidates for these bioactive factors have been examined at a molecular level. Several, including parathyroid hormone-related protein, growth factors (transforming growth factor, insulin-like growth factor, epidermal growth factor), and serotonin, are synthesized within and act upon the gland and possess dynamic receptor activity resulting in diverse effects on growth, calcium homeostasis, and milk composition. This review will focus on the autocrine-paracrine regulation of the mammary gland, with an examination of both foundational work and the progress made within the last 10 to 20 yr of research.

Serotonin Regulates Calcium Homeostasis in Lactation by Epigenetic Activation of Hedgehog Signaling

Calcium homeostasis during lactation is critical for maternal and neonatal health. We previously showed that nonneuronal/peripheral serotonin [5-hydroxytryptamine (5-HT)] causes the lactating mammary gland to synthesize and secrete PTHrP in an acute fashion. Here, using a mouse model, we found that genetic inactivation of tryptophan hydroxylase 1 (Tph1), which catalyzes the rate-limiting step in peripheral 5-HT synthesis, reduced circulating and mammary PTHrP expression, osteoclast activity, and maternal circulating calcium concentrations during the transition from pregnancy to lactation. Tph1 inactivation also reduced sonic hedgehog signaling in the mammary gland during lactation. Each of these deficiencies was rescued by daily injections of 5-hydroxy-L-tryptophan (an immediate precursor of 5-HT) to Tph1-deficient dams. We used immortalized mouse embryonic fibroblasts to demonstrate that 5-HT induces PTHrP through a sonic hedgehog-dependent signal transduction mechanism. We also found that 5-HT altered DNA methylation of the Shh gene locus, leading to transcriptional initiation at an alternate start site and formation of a variant transcript in mouse embryonic fibroblasts in vitro and in mammary tissue in vivo. These results support a new paradigm of 5-HT-mediated Shh regulation involving DNA methylation remodeling and promoter switching. In addition to having immediate implications for lactation biology, identification and characterization of a novel functional regulatory relationship between nonneuronal 5-HT, hedgehog signaling, and PTHrP offers new avenues for the study of these important factors in development and disease.

Increasing serotonin concentrations alter calcium and energy metabolism in dairy cows.

A 4×4 Latin square design in which varied doses (0, 0.5, 1.0, and 1.5 mg/kg) of 5-hydroxy-l-tryptophan (5-HTP, a serotonin precursor) were intravenously infused into late-lactation, non-pregnant Holstein dairy cows was used to determine the effects of serotonin on calcium and energy metabolism. Infusion periods lasted 4 days, with a 5-day washout between periods. Cows were infused at a constant rate for 1 h each day. Blood was collected pre- and 5, 10, 30, 60, 90, and 120 min post-infusion, urine was collected pre- and post-infusion, and milk was collected daily. All of the 5-HTP doses increased systemic serotonin as compared to the 0 mg/kg dose, and the 1.0 and 1.5 mg/kg doses increased circulating glucose and non-esterified fatty acids (NEFA) and decreased beta-hydroxybutyrate (βHBA) concentrations. Treatment of cows with either 1.0 or 1.5 mg/kg 5-HTP doses decreased urine calcium elimination, and the 1.5 mg/kg dose increased milk calcium concentrations. No differences were detected in the heart rates, respiration rates, or body temperatures of the cows; however, manure scores and defecation frequency were affected. Indeed, cows that received 5-HTP defecated more, and the consistency of their manure was softer. Treatment of late-lactation dairy cows with 5-HTP improved energy metabolism, decreased loss of calcium into urine, and increased calcium secretion into milk. Further research should target the effects of increasing serotonin during the transition period to determine any benefits for post-parturient calcium and glucose metabolism.

Elevation of circulating serotonin improves calcium dynamics in the peripartum dairy cow.

Hypocalcemia is a metabolic disorder that affects dairy cows during the transition from pregnancy to lactation. Twelve multiparous Holstein cows and twelve multiparous Jersey cows were intravenously infused daily for approximately 7 days prepartum with either saline or 1.0mg/kg bodyweight of the immediate precursor to serotonin synthesis, 5hydroxy-l-tryptophan (5-HTP). On infusion days, blood was collected before, after, and at 2, 4, and 8h postinfusion. Blood and urine were collected daily before the infusion period, for 14 days postpartum and on day 30 postpartum. Milk was collected daily during the postpartum period. Feed intake and milk yield were unaffected by 5-HTP infusion postpartum. Cows infused with 5-HTP had elevated circulating serotonin concentrations prepartum. Infusion with 5-HTP induced a transient hypocalcemia in Jersey cows prepartum, but not in any other treatment. Holstein cows infused with saline had the highest milk calcium on the day of and day after parturition. Postpartum, circulating total calcium tended to be elevated, and urine deoxypyridinoline (DPD) concentrations were elevated in Holstein cows infused with 5-HTP. Overall, Jerseys had higher urine DPD concentrations postpartum when compared with Holsteins. Taken together, these data warrant further investigation of the potential therapeutic benefit of 5-HTP administration prepartum for prevention of hypocalcemia. Further research should focus on delineation of mechanisms associated with 5-HTP infusion that control calcium homeostasis during the peripartum period in Holstein and Jersey cows.

Increased serum serotonin improves parturient calcium homeostasis in dairy cows

Hypocalcemia in dairy cows is caused by the sudden increase in calcium demand by the mammary gland for milk production at the onset of lactation. Serotonin (5-HT) is a key factor for calcium homeostasis, modulating calcium concentration in blood. Therefore, it is hypothesized that administration of 5-hydroxy-l-tryptophan (5-HTP), a 5-HT precursor, can increase 5-HT concentrations in blood and, in turn, induce an increase in blood calcium concentration. In this study, 20 Holstein dairy cows were randomly assigned to 2 experimental groups. Both groups received a daily i.v. infusion of 1 L of either 0.9% NaCl (C group; n = 10) or 0.9% NaCl containing 1 mg of 5-HTP/kg of BW (5-HTP group, n = 10). Infusions started d 10 before the estimated parturition and ceased the day of parturition, resulting in at least 4 d of infusion (8.37 ± 0.74 d of infusion). Until parturition, blood samples were collected every morning before the infusions, after parturition samples were taken daily until d 7, and a final sample was collected on d 30. Milk yield was recorded during this period. No differences between groups were observed for blood glucose, magnesium, and β-hydroxybutyrate. Cows receiving the 5-HTP infusion showed an increase in fatty acid concentrations from d -3 to -1 before parturition. Serum 5-HT concentrations were increased at d -4 related to parturition until d 5 postpartum in the 5-HTP group compared with the C group. In addition, cows from the 5-HTP group had increased 5-HT concentrations in colostrum, but not in mature milk, on d 7 postpartum. Serum calcium concentrations decreased in both groups around parturition; however, calcium remained higher in the 5-HTP group than in controls, with a significant difference between groups on d 1 (1.62 ± 0.08 vs. 1.93 ± 0.09 mmol/L in control and 5-HTP groups, respectively) and d 2 (1.83 ± 0.06 vs. 2.07 ± 0.07 mmol/L in control and 5-HTP groups, respectively). Additionally, colostrum yield (first milking) was lower in the 5-HTP group compared with the C group, but without consequences on colostrum IgG concentrations. Milk yield did not differ between groups during the rest of the experiment. The study data were consistent with the concept that infusion of 5-HTP to dairy cows increases blood 5-HT concentrations, which in turn is a significant regulatory component in the chain of effectors that affect calcium status around parturition, hence the occurrence of clinical or subclinical hypocalcemia.

Temporarily decreasing progesterone after timed artificial insemination decreased expression of interferon-tau stimulated gene 15 (ISG15) in blood leukocytes, serum pregnancy-specific protein B concentrations, and embryo size in lactating Holstein cows

Our objective was to evaluate the effects of temporarily decreasing progesterone (P4) after timed artificial insemination (TAI) on embryonic growth in dairy cows. Lactating Holstein cows (n = 80) were submitted to a Double-Ovsynch protocol for first TAI and were assigned randomly to receive 12.5 mg of PGF2α 5 d after the last GnRH treatment (LowP4) or remain untreated (control). Blood samples were collected thrice weekly from 5 to 29 d after TAI for all cows and from 32 to 67 d for pregnant cows, and were analyzed for P4 and pregnancy-specific protein B concentrations. Expression of interferon-tau stimulated gene 15 (ISG15) was assessed in blood leukocyte mRNA 18 and 20 d after TAI. Pregnancy diagnosis was performed weekly using ultrasound from 32 to 67 d after TAI, and embryonic crown-rump length was measured 32, 39, and 46 d after TAI. Data were analyzed by ANOVA and logistic regression using the MIXED and GLIMMIX procedures of SAS. The LowP4 cows had less P4 than control cows from 6 to 11 d after TAI; however, pregnancy outcomes 32 d after TAI and pregnancy loss from 32 to 67 d after TAI did not differ between treatments. Control cows diagnosed pregnant 32 d after TAI had greater expression of ISG15 20 d after TAI than LowP4 cows diagnosed pregnant 32 d after TAI, and pregnant control cows had greater pregnancy-specific protein B concentrations from 25 to 67 d after TAI than pregnant LowP4 cows. Embryo size did not differ between treatments 32 and 39 d after TAI, but control cows had larger embryos 46 d after TAI. In conclusion, temporarily decreasing P4 after TAI decreased embryonic growth during early pregnancy in lactating Holstein cows but did not affect pregnancies per artificial insemination or pregnancy loss.

Serotonin Deficiency Rescues Lactation on Day 1 in Mice Fed a High Fat Diet.

Obesity is an inflammatory state associated with delayed lactogenesis stage II and altered mammary gland morphology. Serotonin mediates inflammation and mammary gland involution. The objective of this study was to determine if a genetic deficiency of tryptophan hydroxylase 1, the rate-limiting enzyme in peripheral serotonin synthesis, would result in an improved ability to lactate in dams fed a high fat diet. Twenty-six female mice were fed a high (HFD) or low fat (LFD) diet throughout pregnancy and lactation. Fourteen mice were genetically deficient for Tph1 (Tph1-/-), and twelve were wild type. Milk yield, pup mortality, and dam weights were recorded and milk samples were collected. On day 10 of lactation, dams were sacrificed and mammary glands were harvested for RT-PCR and histological evaluation. HFD dams weighed more than LFD dams at the onset of lactation. WT HFD dams were unable to lactate on day 1 of lactation and exhibited increased pup mortality relative to all other treatments, including Tph1-/- HFD dams. mRNA expression of immune markers C-X-C motif chemokine 5 and tumor necrosis factor alpha were elevated in WT HFD mammary glands. Mammary gland histology showed a reduced number of alveoli in WT compared to Tph1-/- dams, regardless of diet, and the alveoli of HFD dams were smaller than those of LFD dams. Finally, fatty acid profile in milk was dynamic in both early and peak lactation, with reduced de novo synthesis of fatty acids on day 10 of lactation in the HFD groups. Administration of a HFD to C57BL/6 dams produced an obese phenotype in the mammary gland, which was alleviated by a genetic deficiency of Tph1. Serotonin may modulate the effects of obesity on the mammary gland, potentially contributing to the delayed onset of lactogenesis seen in obese women.

Characterization of mammary-specific disruptions for Tph1 and Lrp5 during murine lactation

Serotonin is a homeostatic regulator of the mammary gland during lactation. The contribution of mammary-derived serotonin to circulating serum serotonin concentrations was previously unknown. We have developed mice with mammary-specific disruptions of tryptophan hydroxylase 1 (Tph1) or low-density lipoprotein receptor-related protein 5 (Lrp5) that are induced during late pregnancy and lactation via use of the whey acidic protein (WAP)-Cre cre-lox system. Our objective was to characterize dams with a lactation- and mammary-specific disruption of Lrp5 (WAP-Cre × Lrp5FL/FL) or Tph1 (WAP-Cre × Tph1FL/FL). Milk yield and pup weights were recorded throughout lactation. Dams were euthanized on d10 postpartum and mammary glands and duodenal tissue were harvested. WAP-Cre × Lrp5FL/FL dams had elevated serotonin concentrations in both the mammary gland and circulation compared to controls. In contrast, WAP-Cre × Tph1FL/FL dams had decreased mammary gland and serum serotonin concentrations compared to controls. Alveolar morphology, milk yield, and pup weights were similar. Mammary-derived serotonin makes a significant contribution to circulating serotonin concentrations during lactation, with no effect on milk yield or alveolar morphology. These transgenic models can and should be confidently used in future lactation studies to further elucidate the contribution of serotonin to the maintenance of lactation.

Elevating serotonin pre-partum alters the Holstein dairy cow hepatic adaptation to lactation.

Serotonin is known to regulate energy and calcium homeostasis in several mammalian species. The objective of this study was to determine if pre-partum infusions of 5-hydroxytryptophan (5-HTP), the immediate precursor to serotonin synthesis, could modulate energy homeostasis at the level of the hepatocyte in post-partum Holstein and Jersey dairy cows. Twelve multiparous Holstein cows and twelve multiparous Jersey cows were intravenously infused daily for approximately 7 d pre-partum with either saline or 1 mg/kg bodyweight of 5-HTP. Blood was collected for 14 d post-partum and on d30 post-partum. Liver biopsies were taken on d1 and d7 post-partum. There were no changes in the circulating concentrations of glucose, insulin, glucagon, non-esterified fatty acids, or urea nitrogen in response to treatment, although there were decreased beta-hydroxybutyrate concentrations with 5-HTP treatment around d6 to d10 post-partum, particularly in Jersey cows. Cows infused with 5-HTP had increased hepatic serotonin content and increased mRNA expression of the serotonin 2B receptor on d1 and d7 post-partum. Minimal changes were seen in the hepatic mRNA expression of various gluconeogenic enzymes. There were no changes in the mRNA expression profile of cell-cycle progression marker cyclin-dependent kinase 4 or apoptotic marker caspase 3, although proliferating cell nuclear antigen expression tended to be increased in Holstein cows infused with 5-HTP on d1 post-partum. Immunofluorescence assays showed an increased number of CASP3- and Ki67-positive cells in Holstein cows infused with 5-HTP on d1 post-partum. Given the elevated hepatic serotonin content and increased mRNA abundance of 5HTR2B, 5-HTP infusions may be stimulating an autocrine-paracrine adaptation to lactation in the Holstein cow liver.

Use of the RatLoft decreases pup mortality in lactating mice

Lactating mice in laboratory conditions are thought to be under considerable stress. Dams may manifest this stress through a decrease in milk yield and/or increase in infanticide. The purpose of this study was to examine the effect of access to an environmental enrichment device called the RatLoft on milk yield, circulating serotonin, and pup mortality using both wild-type mice and mice genetically deficient in tryptophan hydroxylase 1 (TPH1−/−; the rate-limiting enzyme in the non-neuronal serotonin synthesis pathway). Presence or absence of the RatLoft did not affect milk yield or circulating serotonin concentrations overall, but serotonin concentrations decreased throughout the 21-day lactation period. Serotonin concentrations were increased in TPH1-deficient mice with access to the RatLoft compared with those without the RatLoft on day 21. Pup mortality tended to decrease for dams with access to the RatLoft as compared to no RatLoft. Within the TPH1−/− groups, dams with access to the RatLoft tended to kill less pups per dam than dams without the RatLoft. These results demonstrate that access to the RatLoft during lactation decreases pup infanticide by dams. This environmental enrichment may be particularly beneficial to TPH1−/− dams that, in addition to decreased infanticide, had increased serotonin concentrations, possibly indicating enhanced well-being. Use of the RatLoft could prove beneficial to researchers working with lactating mouse models to increase the number of pups weaned and positively impact the welfare of the dam.