Peripheral Serotonin Regulates Maternal Calcium Trafficking in Mammary Epithelial Cells during Lactation in Mice

Lactation is characterized by massive transcellular flux of calcium, from the basolateral side of the mammary alveolar epithelium (blood) into the ductal lumen (milk). Regulation of calcium transport during lactation is critical for maternal and neonatal health. The monoamine serotonin (5-HT) is synthesized by the mammary gland and functions as a homeostatic regulation of lactation. Genetic ablation of tryptophan hydroxylase 1 (Tph1), which encodes the rate-limiting enzyme in non-neuronal serotonin synthesis, causes a deficiency in circulating serotonin. As a consequence maternal calcium concentrations decrease, mammary epithelial cell morphology is altered, and cell proliferation is decreased during lactation. Here we demonstrate that serotonin deficiency decreases the expression and disrupts the normal localization of calcium transporters located in the apical (PMCA2) and basolateral (CaSR, ORAI-1) membranes of the lactating mammary gland. In addition, serotonin deficiency decreases the mRNA expression of calcium transporters located in intracellular compartments (SERCA2, SPCA1 and 2). Mammary expression of serotonin receptor isoform 2b and its downstream pathways (PLCβ3, PKC and MAP-ERK1/2) are also decreased by serotonin deficiency, which might explain the numerous phenotypic alterations described above. In most cases, addition of exogenous 5-hydroxy-L-tryptophan to the Tph1 deficient mice rescued the phenotype. Our data supports the hypothesis that serotonin is necessary for proper mammary gland structure and function, to regulate blood and mammary epithelial cell transport of calcium during lactation. These findings can be applicable to the treatment of lactation-induced hypocalcemia in dairy cows and can have profound implications in humans, given the wide-spread use of selective serotonin reuptake inhibitors as antidepressants during pregnancy and lactation.


Introduction
During lactation, calcium secretion into milk by the mammary epithelial cells (MEC) increases dramatically. Regulation of maternal calcium levels during lactation, achieved through molecular and physiological adjustments in calcium homeostasis, is critical to sustain milk synthesis and to satisfy maternal calcium needs [1]. Impaired calcium homeostasis during the early periparturient period causes hypocalcemia in bovine and canine species [2][3][4]. In particular, hypocalcemia is one of the most common metabolic diseases of dairy cattle [5] with profound negative economic and welfare implications to the dairy industry [2,3,6].
The mammary gland is a highly adapted organ that consists of a complex network of cell types that can respond to different molecular and endocrine signals. Particularly during lactation, the mammary gland drives calcium homeostasis. MECs have developed a network of transporters and pumps that enables the transport of calcium from the blood into the milk [7]. The calcium sensing receptor (CaSR) and calcium release-activated calcium channel protein 1 (ORAI-1) are responsible for moving calcium from the circulation into the MEC, and the plasma membrane Ca2+ ATPase 1 and 2 (PMCA1, 2) are involved in regulation of calcium fluxes in MEC and the pumping of calcium into the milk, respectively. In intracellular compartments, the sarcoplasmic endoplasmic Ca 2+ ATPase 2 (SERCA2) stores Ca within the rough endoplasmic reticulum, and the secretory pathway Ca 2+ ATPase 1 and 2 (SPCA1 and 2) are involved in pumping calcium in and out of the Golgi apparatus. The sodium/calcium exchanger 1 (NCX1) participates in MECs trans-epithelial calcium transport, however its exact localization in the MEC is not clear [7][8][9][10][11][12].
Lactation induces the expression on 'non-classical' hormones and factors produced locally by the MECs. The monoamine serotonin (5-HT) impact milk protein gene expression, tight junction permeability, calcium and glucose homeostasis during lactation [13][14][15][16][17][18][19]. Tryptophan hydroxylase 1 (TPH1) is the ratelimiting enzyme in 5-HT synthesis and converts L-tryptophan into 5-hydroxy-L-tryptophan (5-HTP) [13], which is then converted to serotonin, by aromatic l-amino acid decarboxylase. serotonin exerts its actions by signaling through more than 15 receptor subtypes found on various tissues [20]. In lactating rat and mouse dams, serotonin induces mammary gland synthesis and secretion of parathyroid hormone related protein (PTHrP), which activates bone osteoclasts and mobilizes calcium reserved into the circulation of the dam [19,21,22]. In addition, circulating serotonin concentrations in dairy cattle on d 1 of lactation is positively correlated with circulating calcium and PTHrP concentrations, and negatively correlated with the incidence of hypocalcemia, therefore supporting serotonin involvement in calcium homeostasis [23].
Here, we tested the hypothesis that serotonin is required for the appropriate expression and localization of calcium transporters in the lactating mammary gland. We used Tph1 deficient mice to reduce peripheral 5HT synthesis. We also explore plausible downstream pathways that might be involved in the mechanism(s) by which serotonin regulates mammary gland function during lactation. Understanding how serotonin affects calcium transport within the MECs can have therapeutic implications for treatment of lactation-induced hypocalcemia in dairy cattle, and could also have implications for the treatment of depression in humans during lactation.

Ethic Statement
All experiments were performed under protocols approved by the Research Animal Care and Use Committee at the University of Wisconsin-Madison. The protocol number assigned to Dr. Laura L. Hernandez for these experiments was A1473.

Animal Handling and Experimental Design
Twenty-one pregnant female C57B6/J mice were used and maintained in a controlled environmental facility for biological research at the Animal Science Department, University of Wisconsin-Madison. Mice were maintained at a temperature of 25uC and humidity of 50%-60% controlled environment on a 12h light/dark cycle with free access to food and water. Pregnant dams were randomly assigned to individual cages from day 15 of gestation until day 10 of lactation. Mice were assigned to 3 groups: group 1 consisted of Tph1 deficient dams (Tph1 2/2 , n = 7), group 2 consisted of Tph1 2/2 dams injected i.p. with 100 mg/kg of 5-HTP (Tph1 2/2 (5-HTP), n = 7) to rescue peripheral serotonin function, and group 3 consisted of wild-type dams (WT, n = 7). WT and Tph1 2/2 dams were injected daily with saline to control for the variable of stress due to injection. Injections began on d 15 of gestation and ended on d 10 of lactation when all mice and pups were euthanized. Litter size was standardized to 6 pups per dam, regardless of their sex, to control for prolactin surge due to number of pups nursing.

Data and Sample Collection
Litter weights per dam and milk yield were recorded daily, with milk yield being estimated by the weigh-suckle-weigh method [24]. Dam weights were recorded at the beginning and end of the experiment. Serum samples were harvested from blood collected from the maxillary vein on d 1, d 5 and d 10 of lactation. On d 10 of lactation, which is the peak of lactation in female mice, all dams were euthanized and mammary gland number 4 was collected for RNA and total protein isolation. Tissue was stored at 280uC until used. The second mammary gland number 4 was fixed in 4% paraformaldehyde overnight at 4uC and then embedded in paraffin and sectioned (5 mm) for histological evaluation.

Blood serotonin and Calcium Assays
Serum serotonin concentrations were determined by ELISA (Enzo Life Sciences, #ADI-900-175), according to the manufacturer's instructions. Serum was diluted 1:50 to detect serotonin concentrations within the parameters of the assay. The intra-assay CV was 2.6%. Total serum calcium concentrations were measured with a calcium assay kit (Cayman Chemical Company, #700550) according to the manufacturer's instructions. The intra-assay CV was 3.4%.
Mammary gland RNA extraction and Quantitative Realtime RT-PCR (qPCR) Total RNA was isolated from bone and mammary gland tissue with the use of TRI-Reagent (Molecular Research) and was reverse transcribed (1 mg) to cDNA (Bio-Rad, #1708841). Quantitative RT-PCR (qPCR) was conducted with the CFX96 Touch Real-Time PCR Detection System (Bio-Rad). Reaction mixtures, cycling conditions were performed as previously described [19]. Amplification efficiencies of primers were evaluated and were within a range of 95% and 105% efficiency, and primer specificity was assessed by the presence of a single temperature dissociation peak. Primer sequences used are listed Table 1. Ribosomal protein S15 (RPS15) was used as the housekeeping gene, and analysis was conducted with the 2 2DDCt method [25].

Mammary Gland Protein Isolation and Protein Assays
Protein was isolated from the mammary gland tissue using radioimmunoprecipitation assay buffer (RIPA) plus 10 mL/mL of Halt Protease and Phosphatase Inhibitors Cocktail (Thermo scientific #78441). Protein concentrations were determined with the bicinchoninic acid assay (Pierce Chemicals #23227).

Statistical Analyses
Statistical analysis was conducted with Prism version 6.0 b (GraphPad Software). Gene and protein expression, alveoli size and number, dam body weight, blood serotonin and calcium, were analyzed using one-way ANOVA followed by Tukey's post hoc test to test differences between the groups. If sample populations did not have equal variances, Welch's correction was applied. Normality and outlier tests were performed using Shapiro-Wilk test and Grubbs test (ESD method), respectively. Litter weights and milk yield data were analyzed by two-way ANOVA with group, time, and the interaction between group and time serving as main effects. Multiple comparisons were tested with the Holm-Sidak method to detect differences between treatment groups. Differences between means were considered significant at P,0.05. All values are reported as means 6 SEM.

Results
Tph1 gene ablation does not affect dam and litter growth or dam milk yield serotonin at high concentrations can cause mammary gland involution [26] potentially affecting milk yield and pup growth. Therefore, we first evaluated if Tph1 gene ablation affected dam and litter weights, and milk yield. Dam body weight was similar between all group comparisons, both at the beginning of the experiment and on d10 of lactation (31.263.5 and 26.461.5 g average of all groups, respectively; P.0.24). Pup growth and dam milk yield/bout of nursing were affected by time (days of lactation, P,0.001), reflecting offspring's normal growth pattern (7.760.24, 10.560.7, 1661.5 g, mean 6 SEM of three groups for d1, 5, and 10, respectively) and dam milk production (0.03560.01, 0.2560.03 and 0.4260.9, mean 6 SEM of milk yield/bout of nursing for all three groups) but there were no significant differences between groups (P.0.05) at any time point comparison.

Tph1 gene Ablation alters Mammary Epithelial Cell Morphology and Proliferation during Lactation
We then evaluated whetherTph1 ablation affected normal mammary gland histology and cell proliferation. Alveolar size (diameter) was reduced by approximately 50% in Tph1 2/2 mice, regardless of exogenous 5-HTP administration, when compared to WT alveoli size (P,0.001; Figure 1a, b). Alveolar number was not different in Tph1 2/2 mice compared to WT, but upon 5-HTP administration to Tph1 2/2 mice, an increase in alveolar number was observed compared to WT (P,0.001; Figure 1a, c). The number of proliferating cells decreased about 6-fold in Tph1 2/2 mice compared to WT (P = 0.001), and the number was restored through the administration of exogenous 5-HTP to Tph1 2/2 mice (Figure 1d, e).

Tph1 K.O Mice had Decreased Circulating serotonin and Calcium concentrations during Lactation
As expected, Tph1 2/2 dams were deficient in circulating serotonin compared to WT animals on both d1 and d10 of lactation (P,0.038). Administration of exogenous 5-HTP to the Tph1 2/2 mice restored the circulating serotonin deficiency (P, 0.012, Figure 2a). Herein, we assessed serotonin concentrations in the mammary gland. Tph1 2/2 dams had decreased mammary serotonin synthesis (P = 0.018) and this deficiency was restored to the WT levels by administering exogenous 5-HTP to the Tph1 2/2 dams (Figure 2b). In line with previous findings [19] we demonstrated that peripheral 5HT is also necessary to maintain calcium homeostasis during lactation in this experimental model. Tph1 2/2 mice had decreased serum total calcium concentrations compared to WT mice on d 1 and d 10 of lactation, and this decrease was attenuated by giving Tph1 2/2 mice exogenous 5-HTP on both days (P,0.05; Figure 2c).

Influx and Efflux of Calcium in the Lactating Mammary Gland is Impaired by serotonin Deficiency
Lactation is characterized by an extensive trans-cellular flux of calcium, from the basolateral side of mammary alveolar epithelium (blood calcium) into lumen apical side of the cell (milk calcium). We investigated serotonin's role in the regulation and localization of calcium channels and pumps during lactation. The CaSR has been described as master regulator of systemic calcium metabolism, and is largely responsible for calcium influx into the mammary gland [10]. Mammary gland CaSR mRNA expression was decreased inTph1 2/2 mice compared to WT (P = 0.007) and was markedly increased after exogenously administration of 5-HTP compared to Tph1 2/2 and WT (P,0.050; Figure 3 a1). Similarly, mammary CaSR protein level was decreased in Tph1 2/ 2 dams compared to WT (P = 0.048), and recovered after administering 5-HTP (Figure 3 a2). The store-operated channel ORAI-1, central to calcium signaling in mammalian cells, is restricted to basolateral domains of the mammary plasma membrane and is involved in entrance of calcium to the cell [8]. Herein, ORAI-1 mRNA and protein expression were decreased in  the Tph1 2/2 mammary glands compared to the WT (P,0.011), and were restored to WT levels after administration of 5-HTP in both cases (Figure 3 b1, b2). PMCA2 is responsible for pumping about 60% of calcium directly across the mammary secretory cells apical membrane [11]. In this study, PMCA2 mRNA and protein expression were decreased in Tph1 2/2 mammary glands compared to the WT (P,0.051). After admiration of 5-HTP mRNA and protein levels in the mammary tissue were increased compared to WT and Tph1 2/2 (P,0.048; Figure 3 c1, c2).

Calcium Sequestration and Compartmentalization is altered by serotonin Deficiency
Calcium is an essential and ubiquitous second messenger. Changes in cytosolic calcium trigger downstream signaling cascades which regulate a wide range of cellular functions. Additionally, intracellular calcium levels must be tightly controlled to avoid cytoplasmic calcium toxicity [27]. Intracellular calcium stores are regulated by the actions of active calcium transporters including SPCA1, SPCA2, SERCA2 and PMCA1 [7]. We observed that serotonin deficiency decreases mammary gland mRNA expression of several calcium transporters (P,0.006; Figure 3d). Additionally, SERCA2 and SPCA1 mRNA expression increased 2-fold after exogenous 5-HTP was administered to Tph1 2/2 mice (P,0.019; Figure 3d).

Calcium Sensing and Trafficking is Controlled by Serotonin
The lack of peripheral serotonin in lactating mammary tissue from Tph1 2/2 mice markedly decreased the abundance of CaSR (Figure 4 a3) compared to the lactating mammary tissue from WT and the Tph1 2/2 to which 5-HTP was exogenously administered (Figure 4 a1, a5). The WT CasR phenotype is rescued by administering 5-HTP to the Tph1 2/2 mice (Figure 4 a5-6).
Tph1 2/2 lactating mammary tissue exhibited less ORAI-1 (Figure 4 b3) compared to WT and Tph1 2/2 (5-HTP) lactating mammary tissue (Figure 4 b1, b5). PMCA2 is normally present in the apical membrane of the MEC. However, in Tph1 2/2 deficient dams, we observed a shift of this calcium pump to the intracellular compartments, as seen by the unexpected slight co-localization with the b-Cat basolateral marker (Figure 4 c4). PMCA2 did not colocalize with b-Cat in WT and Tph1 2/2 (5-HTP) mammary glands (Figure 4 c2, c6). We next evaluated the expression of Htr2b since the regulation of PTHrP and calcium by serotonin is driven by this receptor subtype [21]. Mammary gland Htr2b mRNA expression decreased in Tph1 2/2 dams compared to WT (P = 0.019), and was rescued by administering 5-HTP to the Tph1 2/2 mice (P = 0.005, Figure 5a). Binding of serotonin to its Gq/11-coupled receptor 2b, activates phospholipase C (PLC), initiating a rapid release of inositol triphosphate resulting in increased intracellular Ca levels, as well as protein kinase C (PKC) and Ras-Raf-Mek-ERK signaling pathways [28,29]. To determine whether serotonin was acting through Htr2b to elicit downstream signals, we tested and determined that mammary gland PLCb3 protein levels were decreased in Tph1 2/2 dams compared to WT (P = 0.036) and increased in the 5-HTP dams (P = 0.045) reaching similar protein levels to the WT mice (Figure 5b). Mammary gland PKC protein levels did not change with 5-HT deficiency but tended to be increased in the 5-HTP dams (P = 0.09; Figure 5b). The mitogenactivated protein (MAP) kinases p44ERK 1 and p42ERK 2 are crucial components of the regulatory machinery underlying normal and malignant cell proliferation [30]. Peripheral serotonin deficiency in the Tph1 2/2 mice decreased ERK 1 (42 KDa) and ERK 2 (44 KDa) expression in their mammary glands (P,0.048, Figure 5c). Administration of exogenous serotonin was able to partially restore ERK 1 but not ERK 2 protein levels in the mammary glands of Tph1 2/2 mice (Figure5c).

Discussion
Despite the numerous studies focusing exclusively on the role of neuronal serotonin on mood and behavior, increasing evidence supports the fact that the monoamine serotonin regulates important non-neuronal functions in various organs throughout the body [31,32]. serotonin is produced by MEC, acting via autocrine-paracrine mechanisms to regulate mammary gland development, milk protein gene expression, milk secretion, and the regulation of tight junctions [13,16,17,26]. Indeed, serotonin regulates maternal calcium levels during lactation through the mammary induction of PTHrP and directly acts on the bone to induce remodeling and support calcium homeostasis [19,21,33].
Herein, we demonstrate that non-neuronal serotonin actively participates in MEC calcium transport and in needed for the expression of key calcium transporters and pumps, maintaining alveolar structure and supporting MEC proliferation during lactation. How calcium is transported from the maternal circulation into the MEC and into the lumen of the alveoli during lactation has been extensively investigated [34,35], and to date the key channels and pumps involved in calcium flux across MECs are known to increase during lactation [7,36,37]. However, the mechanisms underlying and potentially regulating these events are only partially understood. Of particular interest was how serotonin affects calcium transport within the mammary gland and the associated molecular mechanisms. We demonstrate that peripheral serotonin deficiency during lactation markedly decreases the expression of key calcium pumps and channels (CaSR, ORAI-1 and PMCA2) both at the protein and transcript level. It is possible that this observation might be an adaptation of the mammary epithelium to the observed decrease in circulating calcium concentrations seen in TPH1 deficient dams.
The lactating mammary gland has the capacity to sense the extracellular calcium concentrations and adjust its secretion into the milk and the secretion of PTHrP into the circulation via the CaSR [36,38] and therefore the CaSR has been described as a master regulator of systemic calcium metabolism. CaSR-null mice secrete 50% less calcium into the milk [39]. In our study, we demonstrate that lack of serotonin results in decreased mammary CaSR as well as PTHrP (data not shown), but also decreased mRNA expression of other calcium transporters and pumps, including PMCA2 which is responsible for more than 60% of apical transference of Ca to the milk [9]. Even though we did not evaluate calcium content in the milk, we assume a decrease in key calcium transporters will decrease milk calcium. We not only detected decreased PMCA2 (both at the mRNA and protein levels) but also we detected a partial shift in its cellular localization from the basolateral to the intracellular compartment of the mammary epithelium. The anatomic, cellular and subcellular distribution, abundance and localization of a calcium transporter will determine its final function. Potentially, this disruption in the cellular localization of PMCA2 cellular location could be explained by the effects serotonin has on alveolar structure and might be reflected in a decreased efflux of calcium to the milk.
ORAI-1 was recently found to be markedly up-regulated during lactation acting as an important contributor to calcium influx from maternal circulation into the MEC [8,37]. Our studies indicate that given the absence of peripheral serotonin, ORAI-1 mRNA expression is markedly decreased, in a similar fashion to CaSR, which is also located in the basolateral membrane of MEC. This indicates that the influx of calcium to the mammary gland is compromised due to the deficiency in serotonin levels during lactation. Interestingly, we detected an increased mRNA expression of CasR, ORAI-1, NCX1, SERCA2, PMCA2 and SPCA1 after administration of exogenous 5-HTP to the Tph1 2/2 dams, supporting the hypothesis that serotonin may positively affect calcium transport both directly, by increasing its transport and indirectly through CaSR and PTHrP interaction [36,38]. The results reported in this study support the concept that increasing 5-HT during the periparturient period could be beneficial for improving calcium transport across the MEC in addition to signaling the calcium release from bone during lactation by PTHrP [19,21]. Abnormal cellular calcium homeostasis, including dysregulation of mammary gland calcium transporters, can cause alterations in normal cell functions such as uncontrolled cellular proliferation and/or cell death which can lead to diseases such as cancer [37,40].
The major gap in understanding non-neuronal 5-HT's actions in the mammary gland are in terms of which receptor subtype and downstream molecular pathway is modulating specific physiological responses. The mammary glands of different species express unique serotonin receptor patterns with the Htr7 and 2b receptor subtypes being expressed and conserved within the mammary glands of mice, humans, and cattle [21,41]. In particular, the receptor subtype Htr2b has been shown to be responsible for serotonin induction of PTHrP in the bovine mammary gland [21]. Herein, the decrease in this receptor subtype in the mammary glands of Tph1 2/2 coincides with lower calcium concentrations, therefore confirming its participation in calcium regulation in the murine mammary gland. We also show that serotonin deficiency during lactation provokes a decrease in mammary gland expression of PLC-b and PKC signaling pathways, which are downstream of Htr2b receptor. Peripheral serotonin deficiency decreases mammary expression of ERK 1/2 as well as its phosphorylation capacity within the MEC. This result is reflected in decreased cell proliferation observed in the Tph1 K.O mammary glands. The suppression of these signaling cascades might also explain the decreased alveolar number and diameter observed in the Tph1 2/2 dams. However, because exogenous administration of 5-HTP to these dams only partially rescued the WT phenotype, we cannot rule out the participation of other extracellular stimuli (i.e. growth factors, cytokines, mitogens, hormones) in these observed outcomes or 5-HT actions through a different receptor subtype. Abnormal regulation of the MAPK pathways has been reported for a wide range of diseases including many cancers, obesity and diabetes [42], therefore these outcomes can have profound implications not only for lactation.
In summary, the results described here demonstrate the necessity of non-neuronal serotonin in achieving or maintaining adequate MEC calcium influx and efflux as well as normal mammary gland morphology and MEC proliferation, all of which are essential for a successful lactation. Our better understanding of the molecular mechanism(s) and pathways involved in the regulation of mammary gland function by serotonin during lactation may open new avenues of research for the potential pharmacological manipulation of this pathway to improve calcium status and therefore reduce the incidence of lactation induced hypocalcemia as well as other disorders that had been positively correlated with serotonin status at the onset of lactation [23]. Additionally, these findings may be relevant for women taking selective serotonin reuptake inhibitors during lactation.