Intravenous infusion of 5-hydroxytryptophan to mid-lactation Holstein cows transiently affects milk production and circulating amino acid concentrations

In dairy cows, the lactating mammary glands synthesize serotonin, which acts in an autocrine-paracrine manner in the glands and is secreted into the periphery. Serotonin signaling during lactation modulates nutrient metabolism in peripheral tissues such as adipose and liver. We hypothesized that the elevation of circulating serotonin during lactation would increase nutrient partitioning to the mammary glands, thereby promoting milk production. Our objective was to elevate circulating serotonin via intravenous infusion of the serotonin precursor 5-hydroxytryptophan (5-HTP) to determine its effects on mammary supply and extraction efficiency of AA, and milk components production. Twenty-two multiparous mid-lactation Holstein cows were intravenously infused with 5-HTP (1 mg/kg body weight) or saline, in a crossover design with 2 21-d periods. Treatments were infused via jugular catheters for 1 h/day, on d 1–3, 8–10, and 15–17 of each period, to maintain consistent elevation of peripheral serotonin throughout the period. Milk and blood samples were collected in the last 96 h of each period. Whole blood serotonin concentration was elevated above saline con-trol for 96 h after the last 5-HTP infusion. Dry matter intake was decreased for cows receiving 5-HTP, and on average they lost body weight over the 21-d period, in contrast to saline cows who gained body weight. Milk production and milk protein yield were lower in cows receiving 5-HTP during the 3 infusion days, but both recovered to saline yields in the days after. While milk fat yield exhibited a day-by-treatment interaction, no significant difference occurred on any given day. Milk urea nitrogen concentration was lower in 5-HTP cows on the days following the end of infusions, but not different from saline cows on infusion days. Meanwhile, plasma urea nitrogen was not affected by 5-HTP infusion. Circulating concentrations of AA were overall transiently decreased by 5-HTP, with concentrations mostly returning to baseline within 7 h after the end of 5-HTP infusion. Mammary extraction efficiency of AA was unaffected by 5-HTP infusion. Overall, both lactation performance and circulating AA were transiently reduced in cows infused with 5-HTP, despite sustained elevation of circulating serotonin concentration.


INTRODUCTION
At the onset of lactation, the mammary glands begin to synthesize large amounts of the monoamine hormone serotonin.For example, in mice, the mammary glands secrete approximately 50% of the peripheral circulating serotonin during lactation (Weaver et al., 2017).Likewise, in Holstein cows, circulating serotonin is elevated during lactation relative to the nonlactating state (Connelly et al., 2022), although an estimate of the relative contribution from the mammary glands to the circulating serotonin pool has not been reported.Mammary-derived serotonin can function in an autocrine-paracrine manner, signaling primarily through mammary serotonin receptors 5-HT 2B and 5-HT 7 to regulate cellular calcium homeostasis (Horseman, 2022).Although mammary contribution to the circulating serotonin pool has been established, its effects on peripheral metabolism remain unclear.
In cows and other species, metabolically active extramammary peripheral tissues-including the liver, pancreas, skeletal muscle, and white adipose-express a variety of serotonin receptors that are dynamic, changing serotonin's cellular and systemic effects with the physiologic stage (Kim et al., 2010, Laporta and Hernandez, 2015, Field et al., 2021).Many of these effects have been studied by using the serotonin precursor 5-hydroxytryptophan (5-HTP), which increases circulating serotonin concentration.For example, in peripartal rats,

Intravenous infusion of 5-hydroxytryptophan to mid-lactation Holstein cows transiently affects milk production and circulating amino acid concentrations
Virginia L. Pszczolkowski,12 Meghan K. Connelly, 1 August Hoppman, 1 Amara D. Benn, 1 Jimena Laporta,12 Laura L. Hernandez,12 and Sebastian I. Arriola Apelo12* feeding the 5-HTP increases mRNA expression of hepatic gluconeogenic enzymes in concert with mammary GLUT-1 and −8 (Laporta et al., 2013).Furthermore, while intravenous infusion of 5-HTP had no effect on circulating nonesterified fatty acids (NEFA) in the first week post-calving, when cows can be in sever negative energy balance, 5-HTP did increase circulating NEFA pre-calving (Hernández-Castellano et al., 2017), and in non-pregnant, late-lactation cows, in which a decrease in circulating insulin was also observed (Laporta et al., 2015a).Similarly, we previously reported that the same mid-lactation cows used in this study, when subjected to intravenous glucose tolerance test, showed reduced insulin response and blunted NEFA decrease compared with saline-infused cows (Pszczolkowski et al., 2023).These results suggest that, at least in dairy cows, serotonin effects on energy metabolism are dependent on energy balance, with potentially more limited effects in early lactation.
In our study, 5-HTP infused cows had decreased circulating amino acid (AA) concentrations during the glucose challenge (Pszczolkowski et al., 2023).A decrease in circulating AA concentrations could be due to increased hepatic removal for gluconeogenesis, or by other peripheral tissues for anabolic or catabolic processes, including the mammary glands (Martineau et al., 2017).Milk fat synthesis may also be affected in response to serotonin signaling, with 5-HTP treatment of serotonin-deficient lactating mice resulting in transcriptomic enrichment of pathways for lipid synthesis (Laporta et al., 2015b).
Together, the literature suggests that the modulation of serotonin synthesis during lactation may contribute to improved nutrient partitioning to the mammary glands during lactation, both by increasing the availability of some nutrients and possibly improving their mammary utilization.However, previous studies in lactating cattle have been limited in duration, with only 4 d of 5-HTP infusion in Laporta et al. (2015a) and 3 d in both Connelly et al. (2021) and Pszczolkowski et al. (2023).While short-term studies are informative for dissecting out the effects of serotonin in nutrient metabolism, they may have a limited ability to detect possible changes in milk production.Indeed, the majority of studies using 5-HTP to modulate serotonin during lactation, whether during the periparturient period or during established lactation, show no effects on milk production (Laporta et al., 2015a, Weaver et al., 2016, Hernández-Castellano et al., 2017, Slater et al., 2018).Connelly et al. (2021), on the other hand, reported a decrease in milk yield on the last 2 out of 3 d of 5-HTP infusion.
Therefore, our objective with this study was to test both the metabolic and production effects of elevated circulating serotonin through intermittent jugular 5-HTP infusion for 3 weeks.We hypothesized that increasing peripheral serotonin above basal levels in mid-lactation cows would increase mammary extraction efficiency of AA and improve milk and components yields.Evaluation of the metabolic effects of serotonin on lactation performance will allow us to better understand and possibly develop strategies to manipulate nutrient partitioning and efficiency in dairy cattle.

MATERIALS AND METHODS
This experiment was performed under approval by the Institutional Animal Care and Use Committee at the University of Wisconsin-Madison (protocol #A006379).

Cows and experimental design
Twenty-two multiparous mid-lactation Holstein dairy cows (187 ± 34 SD days in milk; 46 ± 5 kg/day milk yield; average parity 3, range 2-4) were housed in tie-stalls at the University of Wisconsin-Madison Dairy Cattle Center, with constant access to fresh water.Cows were milked 2x/day (0430h and 1630h) and fed ad libitum with feed delivered 1x/day after morning milking.The diet (Table 1) was formulated with the Cornell Net Carbohydrate and Protein System (Van Amburgh et al., 2015).Cows were acclimated to the facility no less than 7 d and were adapted to the trial diet from the standard herd diet over a 7-d period, with 30% trial diet fed on the first 2 d, 60% fed on the next 3 d, and 100% fed thereafter, with experimental infusions beginning on the 8th day after diet adaptation began.
Cows were randomly assigned to one of 2 treatment sequences in a 2-period crossover design.Treatments consisted of jugular infusions of either 5-HTP (1 mg/ kg body weight, catalog no.H9772, MilliporeSigma, Burlington, MA) dissolved in 1 L of sterile saline and filtered through a 0.22 µm membrane filter or the same volume of sterile saline.Each period consisted of 21 d, with a 7-d washout between periods.Jugular catheters (catalog no.B1220, Mila International Inc., Florence, KY) were placed one day before the beginning of each period, as previously described (Laporta et al., 2015a).Briefly, cows were sedated with xylazine (intravenous, 10 mg).A shaven area over the jugular vein was scrubbed with alternating 2% chlorhexidine scrub and 70% ethanol, and blocked with subcutaneous lidocaine.An incision was made through the skin over the jugular vein.A 14g, 1.5in needle was used to pierce the jugular vein, and a wire-guide was inserted through the needle.The needle was then removed, and the catheter placed over the wire-guide.After patency was confirmed, the wire was removed and the catheter sutured to the skin at anchor points.Catheters were carefully maintained for the 21-d period and replaced as needed.Within each 21-d period, treatments were infused into the jugular vein only on d 1-3, 8-10, and 15-17.Days 4-7, 11-14, and 18-21 were considered rest days, with no infusions performed, and cows were only monitored and/or sampled.During infusion days, treatment infusions were administered over one hour beginning at 0700h, following morning milking with a veterinary intravenous infusion pump (catalog no.6203H, Heska Corporation, Loveland, CO), with catheters flushed with heparinized saline before and after each infusion.A cyclical infusion design was chosen for 2 reasons.First, jugular infusion of 5-HTP for several consecutive days has been shown to maintain the elevation of circulating serotonin for at least 96 h following termination of infusions in periparturient cows (Hernández-Castellano et al., 2017).Because we designed our experiment under the hypothesis that circulating peripheral serotonin is a major driver of serotonergic metabolism, we expected that the sustained elevation of circulating serotonin provided by intermittent 5-HTP infusions would be sufficient to induce sustained secondary effects.Second, we sought to avoid causing serotonin syndrome in response to chronic infusion of 5-HTP (Turner et al., 2006), for which a maximum safe dose has not been established in cattle.Therefore, to maintain an expected steady elevation of circulating serotonin while attempting to minimize the risk of morbidities with the 3-week duration of the experiment, a pattern of 3 infusion-days and 4 d of rest was chosen.The 21-d period length was chosen to capture potential changes in milk production.

Production and intake measurements
Body weight was recorded at the beginning and end of each period using a walk-through cattle scale.Feed intake was recorded on d 15-21 of each period, and milk production was recorded daily.Total mixed ration feed samples were collected weekly and analyzed by near infrared spectrometry and wet chemistry at Rock River Laboratory (Watertown, WI, USA), and recorded as averages across the experiment.Milk samples were collected each milking on d 15-20 of each period and analyzed by mid-infrared spectroscopy at AgSource Laboratories (Marshfield, WI, USA) for components, with each milking day consisting of the afternoon milking and the following morning milking.Feed refusal samples were collected on d 17-19, and averaged refusals dry matter was used to estimate dry matter intake (DMI) for d 15-21.

Blood sampling
Starting on d 17 (period 1), blood samples were collected at 0, 3, 7, 24, 48, and 96 h following the cessation of saline or 5-HTP infusion.These samples were collected from a coccygeal vessel, to represent arterial blood (Emery et al., 1965, Zhang et al., 2016), and a subcutaneous abdominal vein (referred to mammary vein hereafter), to represent mammary venous blood (Thivierge et al., 2000).Na Heparin and K2 EDTA were used as anticoagulants.Coccygeal samples collected into K2 EDTA-treated samples were mixed with 10 mg/mL ascorbic acid (Connelly et al., 2020), and stored at −20°C for whole blood serotonin analysis.Na Heparin-treated samples were refrigerated at 4°C, centrifuged at 2500 × g for 15 min, with separated plasma stored at −80°C until analyzed.
Data analysis was performed in RStudio using packages nlme and emmeans with ANOVA.Pre-analysis data checks were performed using R packages lmerTest and MASS.In cases of non-normality, the response variable (Y) was transformed either as log e (Y) or 1/ (Y) following the Box-Cox transformation test, and back-transformed for results presentation.In cases of repeated measures, models also included the random effect of cow, the effect of time (or day), and time (or day) × treatment interaction, with correlation structure corAR1.When there was a significant interaction between treatment and time (or day), post-hoc contrasts between treatments at each time point were performed, with Bonferroni p-value adjustment.Based on our power analysis, a second period was required for production variables, for which we used a crossover design.In those cases, models also included fixed effect of period and random effect of cow within period.Significance was declared at P ≤ 0.05, and tendency at 0.05 < P ≤ 0.10.Results are presented as estimated marginal means ± standard error.

Whole blood peripheral serotonin concentration is increased for 96 h following 5-HTP treatment
Whole-blood serotonin concentrations remained increased up to 96 h in cows receiving 5-HTP after the final infusion on d 17, averaging 1426 ng/mL higher at 0 h and 1211 ng/mL higher at 96 h compared with saline cows (treatment P < 0.001, Figure 1).While the elevation in whole-blood serotonin concentration for 5-HTP cows does numerically drop after the 24 h mark, there is not a significant effect of time (P = 0.53).There is, intriguingly, a numerical elevation of serotonin in saline cows at the 0 h mark.While not significant, this minor elevation in serotonin in the saline group at the 0 h mark may suggest stress-induced serotonin release during the jugular infusion protocol, as has been shown in rats in response to a one-hour long stress test (Malyszko et al., 2009).
Regardless, the persistent relative elevation is consistent with what has previously been demonstrated in periparturient dairy cattle, where increased serum serotonin concentrations were maintained for 4 d following the end of 5-HTP infusions (Hernández-Castellano et al., 2017) and demonstrates the overall efficacy of the intermittent approach with respect to circulating serotonin.Despite circulating serotonin being elevated in response to 5-HTP for at least 96 h, most other responses were far more acute, with effects lasting less than 24 h after the cessation of infusions.
Serotonin synthesis is derived from a 2-step process originating from the essential AA tryptophan first being hydroxylated to 5-HTP.This hydroxylation is catalyzed by one of 2 isoforms of tryptophan hydroxylase (TPH) and is considered the rate-limiting reaction of serotonin synthesis.Following hydroxylation, decarboxylation by aromatic AA decarboxylase (AADC) to serotonin occurs (Kuhn and Hasegawa, 2020).When serotonin is released into the circulation by serotonin-synthesizing cells, it is efficiently scavenged by circulating platelets through the serotonin transporter SERT, where it is either stored as dense granules or oxidatively degraded (Pletscher, 1987, Mercado andKilic, 2010).In human platelets, serotonin half-life is estimated at about 5-6 d (Heyssel, 1961), such that it is possible that much of the 96 h of elevation of serotonin in our cows was due wholly to serotonin stored in platelets in the hours following 5-HTP infusion.Tissues that actively secrete serotonin also store it: although not yet identified in the mammary glands, tissues that synthesize serotonin via TPH express vesicular monoamine transporters to facilitate the storage and secretion of serotonin (Erickson et al., 1996).Field et al. (2023) provided evidence of prolonged storage of serotonin in the mammary gland, with increased mammary epithelial cell serotonin content in the days and weeks following intramammary 5-HTP infusion in cows at the time of dry-off.In this way, it is possible that metered release of serotonin by the mammary glands and other tissues may also be partly responsible for the continued elevation of cir-

Dry matter intake is reduced with concomitant loss of body weight under 5-HTP treatment
Cows receiving 5-HTP treatment consumed on average 2.3 kg/day less dry matter than did their saline control counterparts (P = 0.005, Figure 2A).Because the day × treatment interaction was not significant (P = 0.58; Figure 2A), we did not test by-day differences.Serotonin signaling in the brain is involved in regulation of feeding behavior in mammals, producing anorectic effects through various serotonin receptors across multiple regions of the brain (Bacqué-Cazenave et al., 2020).While serotonin itself does not generally cross the blood-brain barrier, 5-HTP does so readily (Hardebo and Owman, 1980), and peripheral intra-venous infusion of 5-HTP has been shown to increase brain serotonin content in rats (Nakatani et al., 2008).In addition to the anorectic effects of serotonin signaling from brain-derived serotonin, cow feed intake in the present study may also have been decreased by neuronal serotonin signaling in the gut: stimulus of serotonin receptor 5-HT 3 on vagal sensory neurons in the gut signal satiety and reduce food intake (Savastano and Covasa, 2007).In other studies where lactating cows were treated with 5-HTP, DMI was not reported (Laporta et al., 2015a, Connelly et al., 2021), and so we cannot make comparisons in that respect.However, in prepartum animals, both Holsteins and Jerseys displayed a drop in feed intake relative to control animals as calving approached, but only after several days of 5-HTP infusion (Weaver et al., 2016).Meanwhile, in nonlactating cows, ruminal 5-HTP infusion did not alter intake (Connelly et al., 2020), but high intravenous infusion rates of 2.5 and 5 mg/kg decreased intake in steers (Valente et al., 2021).
It is not clear whether the drop in DMI for our 5-HTP-treated cows is due to continued serotonergic activity in the brain or gut.As there is a direct connection between the gut and brain via enteric signaling, particularly with respect to serotonin (Margolis et al., 2021), it may well be that both systems were involved, but we were unable to test this either way given our experimental approach.
Cows receiving 5-HTP had a negative average daily gain (ADG) of −0.9 ± 0.2 kg over the full 21-d period, while cows receiving saline instead had an ADG of 0.4 ± 0.2 kg (P < 0.001, Figure 2B).The decline in DMI is at least partly responsible for the observed differences in ADG across the experimental period in 5-HTPtreated cows, especially on the non-infusion days when milk and ECM yield was similar between treatments (Figure 3A,B), thus placing 5-HTP-infused cows in negative energy balance.Unfortunately, DMI was only recorded during d 15-21 of the 21-d period, such that we cannot confirm if and how DMI was affected by 5-HTP before d 15 of each period.
Although not recorded for statistical analysis in this experiment, it should be noted that all cows receiving 5-HTP had an increased incidence of diarrhea-like loose manure during infusion days, starting on d 1 of infusions, and recovering on rest days.This is consistent with results observed by Laporta et al. (2015a) in late-lactation dairy cows and by Valente et al. (2021) in steers.Signaling from enterochromaffin cell derived serotonin stimulates gut motility (Mawe and Hoffman, 2013), and so such a response is not unexpected when driving rapid serotonin synthesis with a bolus-type 5-HTP treatment.Excessively high gut motility results in reduced digestion, reduced nutrient absorption, and increased water loss, which all may have contributed to the lower ADG over the period and milk yield on infusion days for 5-HTP-infused cows in our study (Golher et al., 2021).
The loss in body weight in the 5-HTP-infused cows also suggests mobilization of body fat reserves, which is corroborated by the elevated NEFA observed in the cows during the glucose tolerance test experiment (Pszczolkowski et al., 2023).Although any mobilization of body fat reserves was not captured by a corresponding increase in milk fat production, similar fat yield between treatments suggests again that elevated serotonin may have been acting to promote energy partitioning toward the mammary glands for milk fat synthesis.

Milk production is transiently decreased by 5-HTP treatment
While whole blood serotonin remained increased for at least 96 h, the secondary effects of 5-HTP treatment were not as long-lasting.As such, the concentration of serotonin in whole blood cannot explain the transient nature of the other responses to treatment.Likely, the fact that the 5-HTP treatment was delivered over a brief period of time, essentially as a bolus, explains much of the variation.The rapid influx of 5-HTP to serotonin-sensitive tissues via intravenous circulation during 5-HTP infusion provides an immediate supply of serotonin through AADC activity, whether or not those tissues possess a TPH isoform.If increased serotonin synthesis is only present for the duration that there is 5-HTP substrate, autocrine-paracrine action of serotonin in these tissues may only occur until the newly synthesized serotonin is sequestered in the tissue or released into the circulation, where it will be readily scavenged by platelets (Pletscher, 1987).For instance, in the mammary glands themselves, the high rate of serotonin synthesis occurring at the time of 5-HTP infusion may be sufficient to trigger the early stages of involution (Horseman and Collier, 2014), and subsequently be reversed once free serotonin is no longer available.This is demonstrated by change in milk yield during and after 5-HTP infusions.Except for d 2, all days of infusion resulted in milk yield being depressed by a mean of 5.2 kg/day (P < 0.05) for cows receiving 5-HTP, with rest-day milk yields returning to saline levels (Figure 3A  nously infused with 1.5 mg/kg body weight 5-HTP for 3 consecutive days.In designing the present study, we had chosen 1 mg/kg as opposed to 1.5 mg/kg bodyweight specifically with this in mind, with the intent to avoid the possible toxic effects evident at the 50% higher dose.Using the same dose as our present study, revious work by Laporta et al. (2015a) reported markers of improved nutrient partitioning without impact on milk yield over 4 d of infusion, which we expected to translate into greater milk yield over our longer infusion timeline.That negative effects on milk production were instead evident at the lower dose is notable, and demands further research into both the mechanisms of serotonin and the dosing of its precursor.
Despite lower dry matter intake (Figure 2A), energycorrected milk (ECM) yield was less affected by 5-HTP than the uncorrected milk yield, with only d 15 (the first day of the final 3 d of infusions) being significantly decreased, at 4.9 kg (P = 0.003, Figure 3B).This reduced effect on ECM can be attributed to the absence of significant change in milk fat yield (Figure 3C).The fact that milk fat production was less depressed suggests that elevated serotonin may have been acting to promote nutrient partitioning toward the mammary glands for milk fat synthesis, despite negative effects on other aspects of milk synthesis (Figure 3).During the glucose tolerance test we performed on a subset of these cows, we observed increased fasting NEFA, as well as a reduced response of NEFA to insulin (Pszczolkowski et al., 2023), allowing for greater mammary NEFA availability.Additionally, in lactating mice with a whole-body knockout of tryptophan hydroxylase 1, administration of 5-HTP induced increased mammary gene expression of lipid metabolism pathways that favor milk fat synthesis (Laporta et al., 2015b), indicating that milk fat synthesis may be improved by serotonin from the standpoints of both nutrient supply as well as cellular metabolism.
Milk protein yield exhibited a pattern similar to uncorrected milk yield, with an average reduction of 102 g/day (P ≤ 0.02, Figure 3D) on the days of infusion before subsequent return to saline levels on rest days.Gene expression of both caseins and whey proteins in the mammary epithelium are reduced by local serotonin signaling (Hernandez et al., 2011, Field et al., 2023), as part of serotonin-mediated involution.Expression and activation of different serotonin receptors result in markedly different outcomes (Kim et al., 2010), and so it is not necessarily a contradiction that reduced protein production was observed in concert with an indifferent response in milk fat production.As well, although we suspect that serotonin signaling was likely only enhanced around the time of 5-HTP infusion, transcriptional changes (i.e., casein gene expres-sion) and their subsequent physiological responses (i.e., lower milk protein production) lag behind the initial stimulus, such that a brief increase in serotonin signaling may have been sufficient to drive detectable drop in daily milk protein yield.
The response in milk urea nitrogen (MUN) concentration is intriguing, as it is the only response to 5-HTP treatment that began after the cessation of treatment infusions (Figure 3E).Cows receiving 5-HTP did not differ in MUN concentration on infusion d 15, 16, or 17, but exhibited a reduced MUN concentration on d 18 and 19, averaging 1.3 mg/dL lower on those days (P ≤ 0.05).Concentrations of MUN are predictive of urinary nitrogen excretion (Toledo et al., 2021), suggesting that the 5-HTP-treated cows were likely losing less nitrogen through urine on days post termination of infusions.Potentially, this is the result of the combined effect of milk protein synthesis returning to normal (Figure 3D) while dry matter intake (DMI), discussed in greater detail below, remained low (Figure 2A).As milk protein synthesis returned to the level of saline control cows, this would have pulled more AA from the circulation for sequestration as milk protein, thereby preventing recycling for hepatic catabolism and potentially increasing N efficiency (Pszczolkowski and Arriola Apelo, 2020).

Elevated serotonin transiently reduces concentrations of circulating AA without affecting mammary extraction efficiency
Plasma AA are the substrate mammary glands use to synthesize milk protein, as well as signaling molecules that stimulate that process (Pszczolkowski and Arriola Apelo, 2020).Therefore, we wanted to assess how serotonin would affect circulating concentrations and mammary extraction efficiency of AA over time.We observed that circulating concentrations of the majority of AA, both individually and grouped, decreased for at least a short duration following 5-HTP infusion on d 17, with specific AA affected for a longer duration.The concentration of grouped total AA was reduced by 745 µmol/L (P = 0.007, Figure 4A) in response to 5-HTP treatment immediately after infusion (0 h), but returned to saline control concentration by 7 h after infusion (P = 1.00).Grouped essential AA (EAA) and non-essential AA (NEAA) followed similar patterns, but the average for EAA was still lower 3 h hours after the end of infusion (Figure 4B,C).For grouped branched-chain AA (BCAA), the effect was more subtle but longer-lasting, with a treatment effect (P = 0.02) and no time × treatment interaction (P = 0.51, Figure 4D).During the glucose challenge immediately after d 3 infusions detailed in our previous manuscript Pszczolkowski et al.: 5-hydroxytryptophan and milk production (Pszczolkowski et al., 2023), we also observed lower circulating AA concentrations, with effects lasting for the 120 min after the glucose bolus, when blood parameters were measured.
Reduced intestinal absorption following 5-HTP infusion may be involved in the observed decreased plasma AA concentrations.Salvador and collaborators (1997) identified that serotonin signaling reduces Na +dependent Leu transport across the jejunum in rabbits, presumably via the Na + -dependent AA transporter SLC6A19 (Böhmer et al., 2005).This transporter is also responsible for intestinal absorption of the other large neutral AA, namely His, Ile, Met, Phe, Thr, Trp, Tyr, and Val.All of these AA, except Trp and Val, were significantly reduced in plasma in response to 5-HTP infusions, but whether this reduction was time-dependent varied across AA (Table 2).Reduced intestinal absorption of AA may also have been caused by excessive gut motility induced by 5-HTP treatment, as discussed above with respect to ADG and milk production (Figure 2B).
Increased uptake by peripheral tissues is another possible explanation for the transiently decreased AA.Although we hypothesized that hepatic catabolism of AA for gluconeogenesis would increase in response to increased serotonin (Laporta et al., 2013, Pszczolkowski et al., 2023), that does not appear to be the case in the present study.Hepatic BCAA extraction is nearly absent in cows (Hanigan et al., 2004), while we observed significant or marginal decreases in all 3 BCAA in response to 5-HTP (treatment P ≤ 0.07, Figure 4D and Table 2).Likewise, some AA that are more heavily extracted by the liver, namely Ala, Asp, and Glu (Hanigan et al., 2004), all glucogenic AA, were relatively unaffected by 5-HTP treatment in our study (treatment P ≥ 0.11, Table 2), suggesting that hepatic AA extraction would not be the main mechanism explaining the decrease in circulating AA concentration in 5-HTP-infused cows.
Muscle utilization of AA in response to 5-HTP is another possibility, although this is a largely understudied area of research in dairy cows.Expression at the mRNA level of various serotonin receptors has been reported in the skeletal muscle of dairy calves, and calves fed 5-HTP show skeletal muscle transcriptomic changes in line with cellular growth (Field et al., 2021).In vitro treatment of rat skeletal muscle with serotonin causes reduced release of Ala and Gln, but without any associated increase in protein synthesis (Garber, 1977), suggesting that skeletal muscle protein accretion may not be responsible for the drop in circulating AA observed in the present study.
Because serotonin function and receptor expression can differ dramatically in lactation compared with the nonlactating state (Kim et al., 2010, Laporta andHernandez, 2015), it is possible that the mechanisms explaining the observed drop in circulating AA have not yet been described in the literature.Mammary extraction efficiency was not different between treatments for any AA (Table 3), although this does not mean that mammary utilization did not change.Moreover, serotonin is a regulator of blood flow (Busk et al., 1999) and therefore blood flow was likely unstable during and after treatment infusions, such that we were unable to make use of the Fick principle (Cant et al., 1993) to assess mammary uptake of AA.Increased mammary AA uptake does not align with the observed drop in milk protein yield, but it does so with the decrease in MUN observed on the days following the last 5-HTP infusion.Increased flux of AA toward the mammary glands with same fractional removal would mean reduced recycling of AA to the liver and urea production (Arriola Apelo et al., 2014).

CONCLUSIONS
Overall, we found that increasing circulating serotonin via intermittent, jugular 5-HTP bolus-like infusions caused pronounced but transient reductions on milk production and circulating AA concentrations without altering overall mammary extraction efficiency of AA.Meanwhile, MUN decreased on days post-infusion, suggesting an increase in N efficiency for 5-HTP infused cows after the acute effects of bolus 5-HTP infusion are overcome.Direct effects of elevated serotonin signaling on AA metabolism in mammary tissue in concert with 5-HTP infusion is possible, but changes in metabolism in the gut and other peripheral tissues also likely contributed to the observed responses.Furthermore, ubiquitous expression of AADC limits our ability to determine if those likely effects of serotonin on peripheral tissues are autocrine/paracrine or endocrine and from gut-or mammary-derived serotonin.Importantly, lactation research incorporating the use of 5-HTP as a substrate for serotonin synthesis must consider possible 'off-target' effects of 5-HTP infusions, such as excessive gut motility.These changes can confound responses at sites of interest, particularly regarding nutrient absorption in the intestine, water loss, and possible microbiome depopulation.
Pszczolkowski et al.: 5-hydroxytryptophan and milk production culating serotonin in our study, but more research is needed.

Figure 1 .
Figure 1.Whole blood serotonin remained elevated for at least 96 h after d 17 (final) treatment infusion in mid-lactation Holstein cows treated with 5-HTP (1 mg/kg body weight/day, red triangles) compared with saline (light blue circles).Results presented as estimated marginal means ± standard error, n = 8/treatment (period 1 only).

Figure 2 .
Figure 2.Both DMI and ADG were lower in mid-lactation Holstein cows treated with 5-HTP (1 mg/kg body weight/day, red triangles or bar) compared with saline (light blue circles or bar)."Infusion" (light gray background) refers to the days when cows received jugular infusions of treatments, and "Rest" (white background) refers to the days when cows did not receive infusions.Results presented as estimated marginal means ± standard error, n = 21/treatment (periods 1 and 2).A) Dry matter intake (DMI).B) Average daily gain (ADG).

Figure 3 .
Figure 3. Milk yield was decreased in mid-lactation Holstein cows intravenously infused with 5-HTP (1 mg/kg body weight/day, red triangles) compared with saline (light blue circles)."Infusion" (light gray background) refers to the days when cows received jugular infusions of treatments, and "Rest" (white background) refers to the days when cows did not receive infusions.Results presented as estimated marginal means ± standard error, n = 21/treatment (periods 1 and 2).Asterisks (*) indicate significant difference (P ≤ 0.05) and pound signs (#) indicate a tendency for difference (0.05 < P ≤ 0.10) at each indicated time point, with Bonferroni adjustment.A) Milk yield.B) Energy corrected milk yield (ECM).C) Milk fat yield.D) Milk protein yield.E) Milk urea nitrogen (MUN) concentration.

Table 1 .
Pszczolkowski et al.: 5-hydroxytryptophan and milk production Diet components and analysis

Table 3 .
Mammary extraction efficiency of amino acids (AA) in mid-lactation cows receiving intravenous 5-HTP or saline