LC–MS/MS-based quantification of tryptophan, kynurenine, and kynurenic acid in human placental, fetal membranes, and umbilical cord samples

Tryptophan breakdown metabolites formed along the kynurenine pathway play a significant role in pregnancy and fetal development. To understand their involvement, it is crucial to quantify the levels of tryptophan (TRP), kynurenine (KYN), and kynurenic acid (KYNA) in relevant biological samples such as the placenta, fetal membranes, and umbilical cord. This study used liquid chromatography-tandem mass spectrometry (LC–MS/MS) to determine TRP, KYN, and KYNA levels. The LC–MS/MS method was optimized for high sensitivity and specificity, demonstrating good reproducibility with a precision of < 10% CV and an accuracy of 85–115%. The lower limit of quantification for both TRP and KYN was 0.5 µg/ml, while for KYNA, it was 0.5 ng/mL. The method exhibited linearity within the examined range of concentrations in the homogenate, ranging from 0.5 to 30 µg/ml for TRP and KYN and from 0.5 to 25 ng/ml for KYNA. Using this method, we found significant differences in the concentrations of these substances in investigated maternal–fetal compartments. Placenta samples exhibited higher KYN and lower KYNA concentrations than the umbilical cord and fetal membrane, indicating a potentially important role for kynurenines in late pregnancy. Collectively, this finding may facilitate further research and provide inside into the involvement of the kynurenine pathway of TRP metabolism in fetal development.


Results
In this study, placentae from 9 healthy pregnant women aged 22 to 37 (median: 34, IQR: 4) were included, and the gestational weeks ranged from 37 weeks and 0 days to 40 weeks and 6 days (median: 38 weeks and one day; IQR gestational weeks = 1 and IQR gestational days = 4). No medical complications or infections were reported during or after birth.
LC-MS/MS chromatography was carried out, as detailed in our methods, to measure the concentration of TRP, KYN, and KYNA in the maternal placenta, umbilical cord, and fetal membrane tissue samples. The retention times for TRP, KYN, and KYNA in the three samples were less than 4 min. The chromatograms for the three tissue samples are depicted in Figs. 1 and 2. The analytes exhibited excellent linearity, with regression coefficients (R 2 ) greater than 0.99, a precision of less than 10% CV, and an accuracy of 85-115% (Fig. S1). The lowest limit of quantification (LLOQ) was 0.5 µg/mL for TRP and KYN and 0.5 ng/mL for KYNA. The concentrations in the homogenates used for calibration were 0.5-30 µg/mL for both TRP and KYN and 0.5-25 ng/mL for KYNA. Whenever the signal exceeded the range, we diluted the samples before processing or adjusted the injection volume to remain within the range of method linearity. TRP, KYN, and KYNA concentrations were quantified using LC-MS/MS (Figs. 3, S2, S3 and S4). First, we determined the amount of TRP in placental, umbilical cord, and fetal membrane tissues. The concentrations varied between samples, with the highest concentration found in fetal membrane tissues (3.63 ± 1.34 µg/mL), followed by umbilical cord tissue (2.80 ± 0.20 µg/mL) and maternal placenta (1.35 ± 0.35 µg/mL). The difference appeared to be statistically significant between the placenta and umbilical cord. However, no significant differences were found between the fetal membrane, the maternal placenta, and the umbilical cord, probably due to the high variability of intra-group measures (Fig. 3A) www.nature.com/scientificreports/ KYN concentrations varied among the samples. The highest concentration was found in maternal placenta tissue samples (62.89 ± 6.35 µg/mL), followed by fetal membrane samples (13.59 ± 5.30 µg/mL) and umbilical cord samples (2.20 ± 0.54 µg/mL). In both umbilical cord and fetal membrane tissues, KYN levels were significantly lower than in the maternal placenta (Fig. 3B). The KYNA concentrations were higher in the fetal membrane (26.30 ± 3.80 ng/mL), followed by umbilical cord tissues (18.08 ± 2.30 ng/mL) and maternal placenta samples (2.58 ± 0.32 ng/mL). In the placenta, KYNA levels were significantly lower than in both umbilical cord and fetal membrane tissue (Fig. 3C). Noteworthy is the considerable value of gradient between the concentrations of the investigated substances in the maternal-fetal compartments studied, which is especially evident for KYN and KYNA (Fig. 3D). Based on these measurements, KYN/TRP and KYNA/KYN ratios were calculated, which reflect the activity of IDO and KAT enzymes, respectively. The KYNA/TRP ratio was also calculated. In the human placenta, we determined a significantly higher KYN/TRP ratio (7755.2 ± 1838.4) in comparison to the umbilical cord (77.5 ± 16.47) and fetal membrane (1777.0 ± 1035.2) (Figs. 4A and S5). KYNA/KYN ratio was significantly higher in the umbilical cord (0.0120 ± 0.0031) than in the placenta (0.000046 ± 0.000009) and fetal membrane (0.0044 ± 0.0011) (Figs. 4B and S6). The KYNA/TRP ratio was higher in the fetal membrane (0.0251 ± 0.0116) than in the maternal placenta (0.0037 ± 0.0011) and umbilical cord (0.0067 ± 0.0009) (Figs. 4C and S7).

Discussion
To our knowledge, this is the first study to compare TRP, KYN, and KYNA concentrations in pregnant women's maternal placental, fetal membrane, and umbilical cord tissue samples using LC-MS/MS. Although KYN and its metabolites are known to play a role in fetal development, there is a lack of studies investigating TRP and KP metabolites in these tissues. Most researchers measure the level of substances in the blood, whereas it is crucial to know the content of substances in the tissue, where their potential molecular targets are located.
This report revealed significantly lower KYNA concentrations but higher KYN concentrations in the maternal placenta compared to the other tissues tested. TRP was lower in the placenta than in the umbilical cord. Moreover, the maternal placenta exhibited a significantly higher KYN/TRP ratio than the umbilical cord. On the other hand, the KYNA/KYN ratio was significantly higher in the umbilical cord than in the maternal placenta or fetal membrane. KYNA/TRP ratio was lower in the placenta than in the fetal membrane.  www.nature.com/scientificreports/ Regarding the technical aspects of the analytical method used to assess TRP, KYN, and KYNA, we achieved an accuracy of 85-115% and a precision of < 10% CV, with an acceptable LLOQ (0.5 µg/mL for TRP and KYN and 0.5 ng/mL for KYNA) and linearity over the studied concentration range. The study of this method meets the criteria for bioanalytical methods, and our results were similar to those of previous studies on the analysis of kynurenines in serum (Hu et al. 29 : < 10% CV; LLOQ: 1 µg/mL for TRP, 100 ng/mL for KYN, and one ng/mL for KYNA; van Zundert et al. 30 : < 13% CV; LLOQ: 0.01 µmol/L for KYN and 0.04 µmol/L for TRP); however there are no studies which assess kynurenines in placental tissue by using LC/MS.
Our study revealed 7-and tenfold lower KYNA concentrations in maternal placenta samples than in the other two analyzed tissue samples, namely the umbilical cord and fetal membrane. The significantly lower concentration of KYNA found in maternal placental samples may have physiological relevance. It has been shown that KYNA does not cross the placental barrier between the mother and the fetus, as evidenced in pregnant mice 31 . However, the concentration of KYNA as high as 1.13 μM in last-trimester human amniotic fluid was reported 32 . This phenomenon may indicate that both compartments are isolated and KYNA production in the fetus occurs independently of the maternal plasma concentration. Indeed, a study conducted on mice tissue slices revealed that the fetal brain might produce 3-4 times more KYNA than the maternal brain and placenta, though less than the maternal and fetal liver 33 . However, it has also been suggested that elevated KYNA levels in the placenta are linked to maternal pathological conditions, such as preeclampsia, and abnormal fetal development, including restricted fetal growth and impaired brain development 26 . Future studies should consider evaluating KYNA www.nature.com/scientificreports/ concentrations under medical or psychological conditions, such as depression or perinatal psychological stress since KYNA could also play a role in brain development 26 and its increase is related to stress 26 . Herein, we show a 5-and 25-fold increase in the concentration of KYN in the maternal placenta compared to the fetal membrane and umbilical cord tissue, respectively. Such marked compartmentalization of KYN points to its local high physiological importance. Mechanistically, this is somehow an unexpected phenomenon because KYN is easily transported across biological barriers. Thus, the underlying cause is presumably local KYN production. Indeed, previous studies have demonstrated that increased IDO levels resulted in an enzymatic activity that leads to an increase in KYN concentration in placental tissues 34 . It has also been shown that KYN production starts in the first trimester, as demonstrated in an ex vivo study with placental cell culture 35 . By the end of pregnancy, the development of the endothelial system in the placenta is responsible for an increase in KYN concentration due to higher IDO enzyme activity 34 . Our KYN/TRP ratio calculation, a substitute of IDO activity, indicated increased activity of this enzyme in the maternal placenta samples, consistent with previous studies 1, 34 . Moreover, this enzyme is known to be localized in the syncytiotrophoblast 34,36 and increases from the 14th week of gestation in the human placenta 1 , reaching higher concentrations by the end of pregnancy 34 . Karahoda and colleagues also reported significant IDO enzyme expression and activity in term placentas 34,37 . Our study is consistent with previous research as we found an increased placental KYN/TRP ratio compared to the umbilical cord samples. In addition, our results are consistent with those of other studies, which highlight the critical role of TDO/IDO in maintaining a pregnancy 22,23 . Our study presented the IDO estimate, calculated using the formula [KYN]/[TRP]. This ratio has been used to express changes in IDO activity in isolated tissues or culture media [38][39][40] . It is important to note that this indicator does not strictly evaluate enzyme activity in vivo, as relevant physiological conditions are not considered. Instead, the KYN/TRP ratio used in our study is an easyto-calculate diagnostic index representing the actual content of both substances, which is the sum of the supply, metabolism, and excretion of TRP and KYN.
Other ratios, such as KYNA/KYN ratio and KYNA/TRP ratio, demonstrate dynamic changes in the proportion of KP metabolites in different tissues. Our study found an increased KYNA/KYN ratio, a substitute of Data are presented as a mean ± SEM, number of subjects = 9 per group (A-C). The intercompartmental gradient was calculated as the ratio of the investigated substance's umbilical cord or fetal membrane content against its placental content (adopted as 1). *p < 0.05, ***p < 0.001 versus indicated groups. A nonparametric Kruskal-Wallis test indicated a statistical difference in TRP concentration between at least two tissues (p = 0.031). Dunn's multiple comparisons showed lower TRP content in MP versus UC (p < 0.05). A nonparametric Kruskal-Wallis test showed a statistical difference in KYN concentration between at least two tissues (p < 0.001). Dunn's multiple comparisons test evidenced lower KYN content in both UC (p < 0.001) and FM (p < 0.05) in comparison to MP. One-way ANOVA analysis revealed a statistical difference in KYNA concentration between at least two tissues (p < 0.001). Tukey's multiple comparisons tests indicated that the mean KYNA concentration was significantly higher in UC (p < 0.001) and FM (p < 0.001) in comparison to MP. www.nature.com/scientificreports/ kynurenine aminotransferase (KAT) activity, in umbilical cord tissue but not in the maternal placenta or fetal membrane samples. One possible explanation for this finding is the increased amino acid metabolism of the human umbilical cord mesenchymal stem cells, which is essential for its biological function and proliferation 41 .
The study of Wang and colleagues reported a high expression of the KAT II isoenzyme in the human umbilical cord mesenchymal stem cells, which catalyzes the conversion of KYN to KYNA 42 . Therefore, the higher KYNA/ KYN ratio observed in the umbilical cord tissue sample in our study may be attributed to the mesenchymal stem cells. Further studies are needed to confirm these results and compare KAT activity and its expression between normal-term pregnancies and pathological conditions or dysfunctions such as preeclampsia or perinatal maternal psychological stress. Finally, we find a significant difference in the KYNA/TRP ratio, a marker for the elimination/excretion of kynurenine metabolites indicative of kidney disease 43 , between the placenta and fetal membrane but not umbilical cord samples. The explanation for this phenomenon needs to be substantiated by further research. We speculate that under physiological circumstances, there may be an equilibrium in the placental system regarding the excretion of kynurenine metabolites. Therefore, it would be interesting to compare the excretion of kynurenine metabolites between healthy pregnant participants and those with medical conditions such as preeclampsia or perinatal depression. Although our results confirm the feasibility of using LC-MS/MS for quantifying TRP and other tryptophan metabolites in the maternal placenta, fetal membrane, and umbilical cord samples, several limitations should be considered. These include the small sample size, the lack of comparison with tissues obtained from patients with medical conditions such as preeclampsia, perinatal depression, or perinatal psychological stress, and the lack of umbilical blood or maternal blood samples for further comparisons. Due to the ethics committee's consent conditions, no person-related data were collected. Therefore, our analysis does not include numerous records, including anesthesia and the use of other drugs during elective cesarean delivery. Technical issues, such as the limit of quantification and analysis throughput, should also be considered limitations. Increasing the assay's sensitivity will be necessary if the concentrations of the metabolites assayed are too low. Automated sample preparation may be considered to increase analysis throughput. However, for small numbers of samples, this throughput is satisfactory. Another limitation of this study was the lack of a comparison with genetic material to analyze the production of enzymes related to TRP breakdown or an immunohistological assessment of receptors targeted by KYNA and other TRP metabolites in our study.
In conclusion, we demonstrated that TRP and its metabolites are measurable reliably in fetoplacental tissues. We have developed a highly sensitive and specific LC-MS/MS method for quantifying TRP, KYN, and KYNA in human placental and umbilical cord samples. The method demonstrated good reproducibility, with < 10% CV precision and 85-115% accuracy. This study underscores the significance of precise TRP, KYN, and KYNA quantification in biological samples and the potential application of this method in future studies investigating the involvement of the KP in fetal development. Additionally, we found a high KYNA/KYN ratio in the umbilical cord, high KYN concentrations, and KYN/TRP ratio but low KYNA concentrations in the maternal placenta using this method in material collected during cesarean sections performed on healthy women.  www.nature.com/scientificreports/ Future studies could investigate the relationship between the content of kynurenines and various placental and fetal characteristics, such as gestational age, birth weight, and maternal health. Considering the higher content of KYN in the maternal placenta and higher content of KYNA in the fetal compartment and the multiple biological effects exerted by KYN and KYNA, further investigation of their role in the placenta, including synthesis, metabolism, and regulation, would help explain their importance to fetal development. Our method provides a valuable tool for future studies to understand the role of TRP and kynurenines in placental and fetal health and medical or psychological conditions (i.e., perinatal depression or perinatal psychological stress) during pregnancy.

Materials and methods
Study design. The following is a methods-based research study with a small sample size of human placenta samples collected anonymously from the Department of Obstetrics at the University Hospital of Mannheim, Germany. This research aimed to analyze the concentrations of TRP, KYN, and KYNA. Additionally, we examined the KYN/TRP ratio (reflecting IDO activity), KYNA/KYN ratio (reflecting KAT activity), and the KYNA/ TRP ratio of the human placenta, fetal membranes, and umbilical cord samples. Since the samples were collected anonymously, there was no contact with the participants, and the midwives provided information regarding age and gestation time. The study physician had no contact with the participants. Finally, this study was conducted following the Helsinki Declaration and approved by the Ethics Committee of the Medical Faculty of Mannheim, University of Heidelberg (2022-574).

Study participants and sample collection.
Term placentas (n = 9) were collected after elective cesarean delivery between the 38th and 40th weeks of gestation. All participants who underwent cesarean delivery were medically stable, without any medical conditions, scheduled for the surgical procedure, had previous cesarean delivery experience, and had no medical or emergency indications. Samples of maternal placentae, fetal membranes, and umbilical cords were obtained immediately after birth and dissected. Maternal placentae of patients who underwent emergency procedures and were critically ill or had any medical or obstetric diseases (including fetal death) were excluded from the study.
After the dissection of the placental structure, the biological samples were stored immediately in dry ice for transportation to storage at a research facility at the Central Institute of Mental Health in Mannheim (Germany). The samples were sent in dry ice to the research facility at the Department of Bioanalytics, University of Lublin, Poland, for further analysis. For sample preparation, placentas, fetal membranes, and umbilical cords were collected from study subjects after cesarean section and stored at − 80 °C. 0.2 g of each sample was weighed into a 2 mL plastic centrifuge tube and homogenized in ultrapure water (1:3 w/v) using a laboratory ball homogenizer (Bead Mill MAX, VWR, part of Avantor, Radnor PA, United States). Disintegration was performed using 2.8 mm diameter ceramic bead balls at 6 m/s during one cycle of 30 s. After centrifugation at 20,598 × g for 10 min at 4 °C, 25 µL of the working solution of isotopically labeled internal standards: TRP [ 13 C 11 , 15 Table 1 and Fig. 1 Statistical analysis. All numeric variables were presented as mean ± standard error of the mean -SEM-).
Data that followed a Gaussian distribution were assessed using one-way ANOVA to evaluate the significance of differences between groups. Tukey's test for multiple comparisons was used as a post hoc analysis if a significant difference was found. Data with a non-Gaussian distribution were evaluated using the Kruskal-Wallis (nonparametric ANOVA) followed by post hoc Dunn's multiple comparisons test if a significant difference was found.
Adjusted p-values were calculated for these differences. This study defined statistical significance whenever the two-tail-p-value was less than or equal to 0.05. For the statistical analyses and the descriptive data, we used the R-based software jamovi 2.0.0 44 . Graphs were generated using Prism 8 GraphPad (GraphPad Software Inc., California, United States of America).
Ethics approval, informed consent and consent to participate. Since the samples were collected anonymously, there was no contact with the participants, and the midwives provided information regarding age and gestation times. The study physician had no contact with the participants. As we exclusively utilized anonymized patient data, individual patient consent was not necessary or possible. The study protocol and all study procedures were reviewed and approved by the ethics committee of the Medical Faculty of Mannheim. Additionally, this feasibility study was conducted in accordance with the principles outlined in the Helsinki Declaration for non-identifiable samples. Furthermore, it adhered to the definitions for non-identifiable biological materials as set forth by the Committee of Ministers to member states on research on biological materials of human origin of the Council of Europe (958th meeting of the Ministers' Deputies).
Regarding informed consent, we would like to clarify that since the samples were collected anonymously and, consequently, non-identifiable, there was no contact with the participants . The study physician obtained fresh placentas immediately after birth from the procedures and laboratory room of the Department of Obstetrics at the University Hospital of Mannheim. The midwives provided age and gestation time information without names or other identifiable data. The study physician had no contact with the participants. As we only utilized anonymized patient data and did not perform genetic analysis on the placenta samples, individual patient consent was not necessary or possible. This statement and the study protocol received approval from the Ethics Committee of the Medical Faculty of Mannheim, which waived the need for informed consent. The Ethics Committee II of the Ruprecht-Karls-University of Heidelberg (Medical Faculty Mannheim) reviewed the research project (574-22) from an ethical and legal standpoint and did not raise any concerns regarding the implementation. As biomaterial, obtained through residual material, is considered pseudonymous, the Ethics Committee granted the study without informed consent because genetic analyses are not conducted, and personally identifiable data are not utilized. Consequently, contacting the organ donor or accessing medical records for medication or patientrelated information was prohibited. Furthermore, the Ethics Committee restricted contact between participants and study team members. Lastly, the material was collected directly from the laboratory of the Department of Gynecology and Obstetrics at the University Hospital of Mannheim and was delivered by the hospital's midwives. The committee provided the following instructions, which were strictly adhered to: (1) the project is a feasibility study/methods study using residual material for the establishment of a laboratory method, and (2) since no genetic studies were planned, the biomaterial could be considered pseudonymous at most. Upon completion, the biomaterial was destroyed.

Data availability
The data sets generated and analyzed during the study are not publicly accessible due to the applicable data protection law of the State of Baden-Württemberg, but they are available from the corresponding author on justified request.