Increasing serotonin bioavailability alters gene expression in peripheral leukocytes and lymphoid tissues of dairy calves

Dairy calves are born with a naïve immune system, making the pre-weaning phase a critical window for immune development. In the U.S., 40–60% of dairy farms feed milk replacer to pre-weaned calves, which are devoid of bioactive factors with immunological roles. Serotonin is a bioactive factor with immunoregulatory properties naturally produced by the calf and present in milk. Human and rodent immune cells express the serotonin machinery, but little is known about the role of serotonin in the bovine immune system. Supplementing milk replacer with 5-hydroxytryptophan (serotonin precursor) or fluoxetine (reuptake inhibitor) increases serotonin bioavailability. We hypothesized that increased serotonin bioavailability promotes serotonergic signaling and modulates the expression of immune related genes in peripheral leukocytes and immune-related tissues of dairy calves. The present experiment targeted candidate genes involved in serotonin production, metabolism, transport, signaling and immune regulation. We established that bovine peripheral leukocytes express all known serotonin receptors, and can synthesize, uptake and degrade serotonin due to the expression of serotonin metabolism-related genes. Indeed, we showed that increasing serotonin bioavailability alters gene expression of serotonin receptors and immune-related genes. Further research will determine whether manipulation of the serotonin pathway could be a feasible approach to bolster dairy calves’ immune system.

Dairy calves are born with a naïve immune system. Feeding newborn calves high quality colostrum is a practice readily implemented on most U.S. dairy farms 1,2 . While colostrum is important for calf immune protection and survival, the remaining pre-weaning phase consists of a liquid diet of either whole milk or milk replacer. In the U.S., 40-60% of dairy farms feed milk replacers to pre-weaned dairy calves 3 . Dairy calves' adaptive immune system develops gradually, and the pre-weaning phase has been shown to be critical for immune system development and maturation 4 . Emerging data demonstrate that milk not only delivers nutrients, but also primes the newborn's growth and development through delivery of bioactive factors 5,6 . Although milk replacer formulation has improved over the years, it still lacks bioactive components naturally present in milk that could aid in the development of the dairy calf immune system. Therefore, there is a need to explore novel bioactive factors that when added to milk replacers can enhance dairy calf immune development.
Serotonin is a bioactive factor with immunoregulatory properties 7-12 that is present in cow milk and is also endogenously synthesized by the calf [13][14][15] . However, little is known about the immunologic role of serotonin in milk or in cattle. Serotonin is derived from the conversion of L-tryptophan to 5-hydroxytryptophan (5-HTP) by the rate limiting enzyme tryptophan hydroxylase (TPH1, in peripheral tissues, and TPH2, in the brain), which is subsequently converted to serotonin by the aromatic amino acid decarboxylase enzyme (AADC/DDC) 16 . There are 7 serotonin receptor families with more than 10 G-protein coupled receptor (GPCR) subtypes and 3 ion-gated channel receptor subtypes 17 . Depending on which serotonin receptor subtype (G s , q/11 or i/o ) is activated, signaling cascades including adenylyl cyclase (AC), protein kinase C (PKC), inositol trisphosphate (IP3) and mitogen and extracellular signal regulated kinase (MERK) are activated 18 to modulate the activity of proteins or to regulate gene transcription. Serotonin action is terminated by the serotonin transporter (SERT), which removes circulating serotonin from the extracellular space to be recycled or degraded by monoamine oxidase (MAO).

Effects of 5-HTP and FLX on Proliferation, Apoptosis, and Cell Metabolism Genes. Peripheral
leukocytes. The expression of genes related to cell proliferation, apoptosis, cell metabolism and cell cycle in peripheral leukocytes after 10-d of 5-hydroxytryptophan or fluoxetine supplementation was evaluated. Supplementation of 5-HTP upregulated PTEN (P = 0.005), whereas FLX upregulated CCND1 (P = 0.02) and tended to upregulate LAMP2, CASP8 and PCNA (P < 0.10; Supplementary Table S1).

Discussion
The role of serotonin as an immunoregulatory molecule has been widely demonstrated in human and rodents [7][8][9][10][11]25,27 , however, evidence supporting its role in livestock species is lacking. We previously reported that increasing serotonin bioavailability in dairy calves is possible through the supplementation of 5-hydroxytryptophan or fluoxetine 35 . Herein, we report the effects of increased serotonin bioavailability on circulating WBC count and the gene expression of peripheral leukocytes and secondary lymphoid organs of dairy calves undergoing immune system maturation. To our knowledge, this is the first experiment to characterize how the serotonin axis regulates the bovine immune system.
In this experiment, WBC and WBC subfractions including neutrophil, monocyte, lymphocyte, eosinophil, and basophil counts were within the normal physiological ranges for growing dairy calves and similar among treatment groups before and after 10 days of treatment supplementation. This indicates that increasing serotonin bioavailability for 10 days did not significantly alter immune cell populations. Even though we did not see an increase in neutrophil counts, it is possible that serotonin is improving neutrophil function. For instance, human neutrophils cultured in vitro with high concentrations of serotonin have higher motility than neutrophils grown in low serotonin conditioned media 36 . Platelet expression of FcγRIIA, a receptor that recognizes immune complexes, plays a role in inflammation by activating neutrophils and enhancing endothelial vasodilatation 37 . Furthermore, neutrophils from wild type mice have been shown to have improved tissue infiltration compared to TPH1 knockout mice 25 . Thus, further in vitro experiments evaluating serotonin's role in neutrophil motility and function (i.e. oxidative burst and phagocytic capacity) are needed in bovine.
For over 20 years, researchers have investigated the significance of serotonin receptors in human and murine immune cells and their possible implication in autoimmune diseases. In our experiment, the entire serotonergic machinery, including genes involved in serotonin synthesis, mechanism of action, and metabolism were expressed in the circulating leukocytes of all calves. This indicates that peripheral leukocytes of dairy calves can synthesize, metabolize, uptake and degrade serotonin. Supplementation of either 5-HTP or FLX increased serotonin bioavailability 35 and upregulated several genes involved in serotonin machinery in peripheral leukocytes. The ubiquitous aromatic decarboxylase enzyme, DDC, that converts 5-HTP to serotonin, and monoamine oxidase enzyme, MAOA, that metabolizes serotonin, were significantly upregulated in the peripheral blood leukocytes of both 5-HTP and FLX fed calves. This suggests an overall increase of serotonin metabolism in immune cells of calves under supplementation.
Notably, supplementing 5-HTP upregulated the gene expression of 9 out of the 13 serotonin receptor subtypes evaluated in peripheral leukocytes compared to controls. Interestingly, all serotonin receptors from family 1 subtypes (5-HT1), including -1A, -1B, -1D, and -1F were significantly upregulated, suggesting a positive feedback loop to increase ligand binding. Serotonin receptor family 1 proteins couple mainly through G i/o proteins to inhibit adenylyl cyclase in various cells and tissues and have high affinity towards serotonin 18 . Serotonin receptor 5-HT1 subtypes are expressed in various human and mouse immune cells including monocytes/macrophages, dendritic cells, neutrophils, mast cells, eosinophils, B cells and T cells 7 . Data in human and/or rodents shows that 5-HT1A receptor subtype modulates adhesion and chemotaxis of mast cells 30 and enhances phagocytosis by murine macrophages 38 . Upregulation of serotonin receptors subtypes from family 1 in peripheral leukocytes after 5-HTP supplementation could support calves' immune function by promoting adhesion and chemotaxis of mast cells, and/or enhancing phagocytosis. However, future functional studies targeting this specific receptor should be conducted in the bovine.
In this experiment, the 5-HT2B receptor subtype was upregulated following 5-HTP supplementation. Activation signals among 5-HT2 family receptor subtypes are different, but mainly couple through G q/11 proteins. For instance, the 5-HT2A receptor subtype signals through the activation of PLC-β in tissues and cells, whereas the 45% homologous 5-HT2B receptor subtype signal through various phospholipases (i.e. PLC-β, PLA) 18 . This 5-HT2B receptor subtype has been widely studied in human dendritic cells (DCs) where it is reported to promote anti-inflammatory functions. Human monocytes cultured in the presence of serotonin, as well as IL-4 and granulocyte-macrophage colony-stimulating factor, differentiate into DCs with reduced expression of co-stimulatory molecules (i.e. CD86) which are needed for antigen-presenting cells (APC; i.e. DCs) and T cell cognate interactions 39 . Similarly, 5-HT2B receptor activation was found to downregulate monocyte derived DC expression of co-stimulatory molecules that activate naïve T cells, and possibly preventing inflammation by regulating both innate and adaptive immune systems 40 . Apart from 5-HT2B, 5-HT2A expression is upregulated on activated CD4+ and CD8+ T cells 41 . Furthermore, 5-HT2A antagonist treatment inhibits T cell activation and diminishes IL-2 and interferon gamma (IFN-γ) production in a dose dependent manner 41 . Thus, it appears that 5-HT2A acts as a proinflammatory serotonin receptor whereas 5-HT2B acts as an immunosuppressive serotonin receptor.
Herein, both IL2 and IFNG gene expression were upregulated even though 5-HT2A was not differentially expressed. Interleukin-2 is an immunoregulatory cytokine, mainly produced by CD4+ T cells, which enhances T cell proliferation, regulates T helper cell differentiation 42 , and limits immune responses by enhancing T-regulatory cells function 43 . Moreover, IL-2 induces the transcription of IFN-γ in T cells 44 . Interferon gamma is an important inflammatory cytokine that induces maturation and licensing of APC that in turn recruit and prime T cells, and increases the expression of major histocompatibility complex 45 . Translation of IFN-γ by T cells skews B cells to enhance antibody production and induces isotype switching from IgM to IgG2a 46 . Upregulation of IL2 and IFNG by 5-HTP supplementation, if translated to protein, could act to support adaptive immune responses. Further studies are warranted to confirm and determine the implications of these findings. (2020) 10:9712 | https://doi.org/10.1038/s41598-020-66326-w www.nature.com/scientificreports www.nature.com/scientificreports/ The serotonin receptors -3B and -3C, and -4 were upregulated following 5-HTP supplementation. The 5-HT3B receptor subtype is a ligand-gated ion channel 47 and is 41% homologous to the 5-HT3A receptor 48 . This receptor has been linked to nausea in patients undergoing chemotherapy 49 . To our knowledge the function of 5-HT3B receptor subtype has not been linked to immunity, probably because 5-HT3B can be expressed as a heteromeric receptor, 5-HT3A/B, with properties differing from those of 5-HT3A receptor subtype 50 .
Gene expression of key signaling molecules downstream of serotonin receptors, including ADCY1, PLCB2, MAPK3 and AKT, was upregulated by 5-HTP supplementation. Adenyl cyclase (ADCY1) is a major downstream signaling gene for 5-HT4, -6 and -7 following G s coupling activation 18 . Yet, in some cell types, activation of 5-HT1A receptor inhibits ADCY1 51 . Since both 5-HT1 and -4 receptor families were upregulated in our experiment we cannot determine which specific serotonin receptor subtype might be upregulating the expression of ADCY1 intracellularly. Furthermore, both PLCB2 and 5-HT2B were differentially expressed and PLCB2 is known to be the major signaling downstream molecule for 5-HT2B 52 . Yet, further investigation is needed to characterize the effects of specific serotonin receptors and downstream pathways on specific circulating immune cells in the bovine.
The gene expression of clusters of differentiation CD4, CD8 and CD14 in peripheral leukocytes were not affected by 5-HTP supplementation, suggesting that increased serotonin bioavailability had no effect on immune cell concentrations. However, serotonin has been linked to the migration of specific immune cell types. For instance, Müller et al. 53 reported that 5-HT1B receptor induces the migration of human immature DCs. Additionally, human DCs have been shown to secret IL-1 β after activation of receptors 5-HT3, -4 and -7 54 . Interestingly, in our experiment, peripheral blood leukocytes had greater expression of serotonin receptor subtypes 5-HT3B, -3C and -4, although upregulation of IL1B was not observed. Hernandez-Castellano et al. 55 supplemented 5-HTP to newborn dairy calves for 15 days and reported an upregulation of both IL-1 β and nuclear factor kappa beta (NF-κβ) genes in blood but no differences in IgG production were observed. In our experiment, REL (NF-κβ subunit) was not affected by 5-HTP supplementation, however discrepancies between Hernandez-Castellano's and our results could be attributed primarily to calf age linked to different facets of immune system development.
To explore serotonin's role on immune system activation, we also evaluated the gene expression of surface molecules and cytokines. Supplementation of 5-HTP for 10 days tended to upregulate CD80 gene expression and upregulated CTLA4, its preferential binding partner 56 . The cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) is a protein expressed by regulatory T cells and activated T cells that exerts negative feedback to diminish T cell responses, whereas CD80 is a costimulatory molecule expressed on APC 57 . As previously mentioned, 5-HTP supplementation upregulated IL2 gene expression, which plays an important role in T-helper cell proliferation and survival, and T-regulatory cell activation 42 . Our findings suggest that greater serotonin bioavailability might activate serotonin receptors on immune cells, promoting IL-2 production, T cell activation, and eventually CTLA4 expression to prevent excessive T cell activation 42,43 . Thus, we propose that the serotonin axis may play a role in balancing the immune system by promoting protective immune responses and preventing potentially dangerous inflammation.
Fluoxetine oral supplementation increases serotonin bioavailability by blocking SERT 58 . Fluoxetine supplementation downregulated 5-HT2A and -3A receptor subtypes by more than 20-fold in peripheral leukocytes. Contradictory findings have been reported linking the use of SSRIs to 5-HT2A downregulation in the rodent frontal cortex 59 . Nevertheless, the effects and implications of the downregulation of 5-HT2A receptor by SSRIs in the immune system remain unknown [59][60][61] . Monocytes and T cells express the 5-HT3A receptor subtype 62 , as well as naïve and activated B cells, predominantly by differentiating B cells at the germinal centers of lymph nodes 63 . Interestingly, the use of the 5-HT3A antagonist, tropisetron, inhibits T cell activation and production of IL-2 64 . In 1994, Fan 65 reported that 5-HT3 receptor is a target of fluoxetine, which in turn decreases serotonin influx into the cell. Moreover, fluoxetine blocks 5-HT3 receptors by interactions during both open and closed channel states, although the clinical relevance of this effect is still unknown 66 . Thus, further research is needed to understand 5-HT3B downregulation by fluoxetine supplementation.
The thymus is a central lymphoid organ where T cells develop, while the spleen and lymph nodes are important secondary lymphoid organs where immune responses are generated 67 . Therefore, we sought to explore the effects of increased serotonin bioavailability in these tissues. Similar to peripheral leukocytes, all tissues, independent of treatment supplementation, expressed the serotonergic machinery indicating that serotonin could be playing a role in the development and deployment of adaptive immune responses. We demonstrated that increased serotonin bioavailability exerts a less pronounced effect in these tissues compared to peripheral leukocytes, at least at the mRNA level. However, it is important to mention that tissue data was collected from a subset of animals (n = 4/treatment) euthanized after 10 days of treatment supplementation and that the statistical power to detect significant differences in tissues was 66%. Nevertheless, it is notable that we observed a significant downregulation of CD14 in the popliteal lymph node after 5-HTP supplementation and FLX tended to downregulate CD8B in spleen tissue, while both treatments upregulated CTLA4 in the thymus and spleen tissues. CD8 downregulation has been shown to occur when CD8 effector T cells are switching functions 68 . There is limited data exploring the effect of serotonin on thymus, lymph node, and/or spleen gene expression, thus, these results are novel and warrant deeper investigation.
The present experiment targeted candidate genes involved in serotonin production, metabolism, transport, signaling and immune regulation. We established that bovine peripheral blood leukocytes and immune tissues express components of the serotonin signaling pathway, including TPH1, SERT, DDC, MAO and serotonin receptors. This indicates that these cells and tissues have the potential to synthesize, transport, respond to and/or degrade serotonin. We demonstrated that at the mRNA level, increased serotonin bioavailability exerts a pronounced immunomodulatory response, particularly in peripheral leukocytes and spleen tissue. Indeed, specific Scientific RepoRtS | (2020) 10:9712 | https://doi.org/10.1038/s41598-020-66326-w www.nature.com/scientificreports www.nature.com/scientificreports/ serotonin receptors and cytokines were differentially expressed upon 5-HTP or FLX supplementation, which could potentially influence the developmental trajectory and maturation of immune cells in dairy calves at a young age. Differences in the modulatory effects of 5-HTP and FLX in the peripheral immune system could be attributed to intrinsic differences in their molecular mechanism of action. Indeed, we previously reported that 5-HTP calves had greater circulating serotonin concentrations when compared to fluoxetine calves 35 , thus we hypothesize that greater serotonin bioavailability exerts different effects. While promising correlations between serotonin and the immune system exist in other species, and now in the bovine at the transcriptional level, it is currently unknown whether promoting serotonin translates into both innate and adaptive immune system orchestration. Ultimately, the fact that 5-HTP is a biogenic modified amino acid whereas fluoxetine is a synthetic drug should be taken into consideration when developing strategies to enhance livestock health and development.

Methods
Animals and experimental design. All methods and procedures performed in this study were carried out in accordance with relevant guidelines and regulations approved by the Institutional Animal Care and Use Committee at the University of Florida (protocol # 201709851). Experimental design and treatments are described in detail by Marrero et al. 35 . Briefly, Holstein bull calves (n = 24, 18 ± 2 d of age, 47 ± 3.2 kg) were assigned to one of three treatments in a complete randomized block design (8 pens, 2.3 m x 2.5 m; n = 3 per pen, one of each treatment). Calves received 4 L of milk replacer (Southeast Milk Inc, Okeechobee, FL) at 0700 h and 1700 h. Treatments were administered once daily (0700 h feeding) by supplementing milk replacer with 5-hydroxytryptophan (5-HTP, 90 mg/d, n = 8, Sigma, St. Louis, MO, USA; #H9772), fluoxetine (FLX, 40 mg/d, n = 8, Spectrum Chemical, Gardena, CA, USA; #F1200) or saline (CON, n = 8) for 10 consecutive days. Treatment was applied individually to each calf. Supplementing 5-HTP and FLX increases serotonin bioavailability by different mechanisms: exogenous 5-HTP bypasses the rate liming enzyme TPH1, allowing its conversion to serotonin by AADC 69 , whereas FLX binds to SERT inhibiting endogenous serotonin reuptake within the cell 58,70 . Hematology analysis. Whole blood samples were collected from the jugular vein before (d0) and on the 10 th day of treatment supplementation 4 h after 0700 h feeding in tubes containing K2 EDTA (BD, Franklin Lakes, NJ, USA, #368047). Within 2 h of collection, blood samples were analyzed for hematology parameters including white blood cells (WBC), neutrophil, lymphocyte, monocyte, eosinophil and basophil count/μL using the Idexx ProCyte Dx analyzer (IDEEX Laboratories Inc., Westbrook, ME). peripheral blood mononuclear cell isolation. Blood samples were collected from the jugular vein on the 10 th day of treatment supplementation 4 h after the 0700 h feeding using heparin blood collection tubes (BD, Franklin Lakes, NJ, USA, # 366430) and kept on ice until laboratory arrival. Blood was centrifuged at 1,200 g for 18 min at 20 °C, plasma layer was discarded, and peripheral leukocytes (i.e., buffy coat) was transferred to a 15 mL conical tube. To lyse residual red blood cells, a hypotonic solution (lyse buffer:1.5 g Na2HPO4 (Fisher; #BP332-1) and 0.3 g NaH2PO4 (Fisher; #BP329-1) at pH 7.2) was used. To restore cells, restore buffer was used (27 g NaCl (Fisher cat. #BP358-1) added to the lyse buffer solution and adjusted to pH 7.2). To each sample, 8 mL of lyse buffer and 4 mL of restore buffer were added. Tubes were centrifuged at 650 g for 5 min at 4 °C to form a pellet. Pellets were resuspended in 200 µL of RNAlater (Invitrogen, Carlsbad, CA, USA; #AM7021) and stored at -80 °C until RNA extraction. euthanasia and tissue collection. After the 10 th d of treatment supplementation, 4 calves per treatment (n = 4 pens) were euthanized at the University of Florida abattoir. Calves were sedated by intravenous administration of 0.2 mg/kg xylazine and euthanized using a captive bolt pistol followed by jugular exsanguination. Spleen, popliteal lymph node and thymus tissues (approximately 1 g each) were harvested, rinsed in sterile PBS, transferred to a cryotube containing RNAlater and stored at −80 °C until RNA extraction.
peripheral Leukocyte and tissue RnA extraction. Peripheral leukocytes resuspended in RNAlater were centrifuged at 650 g for 10 min at 4 °C to reform the pellet and RNAlater was removed. For tissues, RNA was extracted from 60 mg each of spleen, popliteal lymph node and thymus. Each sample was placed in 1 mL of QIAzol Lysis reagent (Qiagen, cat. #79306) and homogenized using a tissue homogenizer (Tissue Master 125, Omni International, GA, USA). A commercial RNA extraction kit (RNeasy Plus Universal Mini Kit, Qiagen, cat. #73404) was used according to the manufacturer's instructions. RNA concentration and quality were determined using a NanoDrop (NanoDrop Spectrophotometer, Thermo Scientific, USA; #ND-2000). RNA samples were stored at −80 °C until gene expression analysis. primer design, validation and, gene expression analysis. We quantified the expression of genes related to serotonin machinery and signaling (i.e., synthesis and metabolism, serotonin receptors, and downstream pathways), immune-related genes (i.e., cytokines), and genes involved in metabolic and cellular processes (i.e., apoptosis, cell cycle, among others) in peripheral leukocytes, spleen, thymus and popliteal lymph node tissues. For this, high-throughput Multiplex RT-qPCR BioMark Dynamic Array Integrated Fluidic Circuits (IFCs) was used (Fluidigm Corporation, South San Francisco, CA). Briefly, 96 primers targeting 91 genes of interest, 4 reference genes (ACTB, GADPH, RSP9 and HPRT1), and one reference structural gene were assayed (see Supplementary  Table S1). An initial quantification run was performed for primer validation using an 8-point, two-fold dilution series (in triplicate) using RNA pools per tissue of interest. The linearity between RNA quantity and cycle threshold (Ct) was tested and efficiency of amplification was calculated. Primers were considered validated if they passed 5 points with an efficiency of 0.8-1.3 and an R 2 ≥ 0.92. Specificity of amplification for each primer pair was evaluated by plotting the dissociation-characteristics of double-stranded DNA. A single peak following melt curve analysis indicated a pure, single amplicon. For gene expression of immune-related tissues and peripheral leukocytes, RNA