SteroidXtract: Deep Learning-Based Pattern Recognition Enables Comprehensive and Rapid Extraction of Steroid-Like Metabolic Features for Automated Biology-Driven Metabolomics.

Despite the vast amount of metabolic information that can be captured in untargeted metabolomics, many biological applications are looking for a biology-driven metabolomics platform that targets a set of metabolites that are relevant to the given biological question. Steroids are a class of important molecules that play critical roles in many physiological systems and diseases. Besides known steroids, there are a large number of unknown steroids that have not been reported in the literature. The ability to rapidly detect and quantify both known and unknown steroid molecules in a biological sample can greatly accelerate a broad range of steroid-focused life science research. This work describes the development and application of SteroidXtract, a convolutional neural network (CNN)-based bioinformatics tool that can recognize steroid molecules in mass spectrometry (MS)-based untargeted metabolomics using their unique tandem MS (MS2) spectral patterns. SteroidXtract was trained using a comprehensive set of standard MS2 spectra from MassBank of North America (MoNA) and an in-house steroid library. Data augmentation strategies, including intensity thresholding and Gaussian noise addition, were created and applied to minimize data overfitting caused by the limited number of standard steroid MS2 spectra. The CNN model embedded in SteroidXtract was further compared with random forest and XGBoost using nested cross-validations to demonstrate its performance. Finally, SteroidXtract was applied in several metabolomics studies to demonstrate its sensitivity, specificity, and robustness. Compared to conventional statistics-driven metabolomics data interpretation, our work offers a novel automated biology-driven approach to interpreting untargeted metabolomics data, prioritizing biologically important molecules with high throughput and sensitivity.

Sex-specific effects of neonatal oral sucrose treatment on growth and liver choline and glucocorticoid metabolism in adulthood.

Hospitalized preterm infants experience painful medical procedures. Oral sucrose is the nonpharmacological standard of care for minor procedural pain relief. Infants are treated with numerous doses of sucrose, raising concerns about potential long-term effects. The objective of this study was to determine the long-term effects of neonatal oral sucrose treatment on growth and liver metabolism in a mouse model. Neonatal female and male mice were randomly assigned to one of two oral treatments (n = 7-10 mice/group/sex): sterile water or sucrose. Pups were treated 10 times/day for the first 6 days of life with 0.2 mg/g body wt of respective treatments (24% solution; 1-4 µL/dose) to mimic what is given to preterm infants. Mice were weaned at age 3 wk onto a control diet and fed until age 16 wk. Sucrose-treated female and male mice gained less weight during the treatment period and were smaller at weaning than water-treated mice (P ≤ 0.05); no effect of sucrose treatment on body weight was observed at adulthood. However, adult sucrose-treated female mice had smaller tibias and lower serum insulin-like growth factor-1 than adult water-treated female mice (P ≤ 0.05); these effects were not observed in males. Lower liver S-adenosylmethionine, phosphocholine, and glycerophosphocholine were observed in adult sucrose-treated compared with water-treated female and male mice (P ≤ 0.05). Sucrose-treated female, but not male, mice had lower liver free choline and higher liver betaine compared with water-treated female mice (P < 0.01). Our findings suggest that repeated neonatal sucrose treatment has long-term sex-specific effects on growth and liver methionine and choline metabolism.

Measurement of 11-dehydrocorticosterone in mice, rats and songbirds: Effects of age, sex and stress.

Glucocorticoids (GCs) are secreted into the blood by the adrenal glands and are also locally-produced by organs such as the lymphoid organs (bone marrow, thymus, and spleen). Corticosterone is the primary circulating GC in many species, including mice, rats and birds. Within lymphoid organs, corticosterone can be locally produced from the inactive metabolite, 11-dehydrocorticosterone (DHC). However, very little is known about endogenous DHC levels, and no immunoassays are currently available to measure DHC. Here, we developed an easy-to-use and inexpensive immunoassay to measure DHC that is accurate, precise, sensitive, and specific. The DHC immunoassay was validated in multiple ways, including comparison with a mass spectrometry assay. After assay validations, we demonstrated the usefulness of this immunoassay by measuring DHC (and corticosterone) in mice, rats and song sparrows. Overall, corticosterone levels were higher than DHC levels across species. In Study 1, using mice, we measured steroids in whole blood and lymphoid organs at postnatal day (PND) 5, PND23, and PND90. Corticosterone and DHC showed distinct tissue-specific patterns across development. In Studies 2 and 3, we measured circulating corticosterone and DHC in adult rats and song sparrows, before and after restraint stress. In rats and song sparrows, restraint stress rapidly increased circulating levels of both steroids. This novel DHC immunoassay revealed major changes in DHC concentrations during development and in response to stress, which have important implications for understanding GC physiology, effects of stress on immune function, and regulation of local GC levels.

Steroid profiling in human primary teeth via liquid chromatography-tandem mass spectrometry for long-term retrospective steroid measurement.

Steroid hormones are important modulators of many physiological processes, and measurements of steroids in blood, saliva, and urine matrices are widely used to assess endocrine pathologies and stress. However, these matrices cannot be used to retrospectively assess early-life stress and developmental endocrine pathologies, because they do not integrate steroid levels over the long term. A novel biological matrix in which to measure steroids is primary teeth (or "baby teeth"). Primary teeth develop early in life and accumulate various endogenous molecules during their gradual formation. Here, we developed and validated the first assay to measure steroids in human primary teeth using liquid chromatography-tandem spectrometry (LC-MS/MS). Our assay is highly sensitive, specific, accurate, and precise. It allows for the simultaneous quantification of 17 steroids in primary teeth (16 of which have not been examined previously in primary teeth). Overall, steroid levels in primary teeth were relatively low, and 8 steroids were quantifiable. Levels of dehydroepiandrosterone, cortisol, and progesterone were the highest of the 17 steroids examined. Next, we used this assay to perform steroid profiling in primary teeth from males and females. The same 8 steroids were quantifiable, and no sex differences were found. Levels of androgens (androstenedione and testosterone) were positively correlated, and levels of glucocorticoids (cortisol, cortisone, corticosterone, 11-dehydrocorticosterone) were also positively correlated. These data demonstrate that multiple steroids can be quantified by LC-MS/MS in human primary teeth, and this method potentially provides a powerful new way to retrospectively assess early-life stress and developmental endocrine pathologies.

Isoflurane stress induces region-specific glucocorticoid levels in neonatal mouse brain.

The profound programming effects of early life stress (ELS) on brain and behavior are thought to be primarily mediated by adrenal glucocorticoids (GCs). However, in mice, stressors are often administered between postnatal days 2 and 12 (PND2-12), during the stress hyporesponsive period (SHRP), when adrenal GC production is greatly reduced at baseline and in response to stressors. During the SHRP, specific brain regions produce GCs at baseline, but it is unknown if brain GC production increases in response to stressors. We treated mice at PND1 (pre-SHRP), PND5 (SHRP), PND9 (SHRP), and PND13 (post-SHRP) with an acute stressor (isoflurane anesthesia), vehicle control (oxygen), or neither (baseline). We measured a panel of progesterone and six GCs in the blood, hippocampus, cerebral cortex, and hypothalamus via liquid chromatography tandem mass spectrometry. At PND1, baseline corticosterone levels were high and did not increase in response to stress. At PND5, baseline corticosterone levels were very low, increases in brain corticosterone levels were greater than the increase in blood corticosterone levels, and stress had region-specific effects. At PND9, baseline corticosterone levels were low and increased similarly and moderately in response to stress. At PND13, blood corticosterone levels were higher than those at PND9, and corticosterone levels were higher in blood than in brain regions. These data illustrate the rapid and profound changes in stress physiology during neonatal development and suggest that neurosteroid production is a possible mechanism by which ELS has enduring effects on brain and behavior.

Glucocorticoid Production in Lymphoid Organs: Acute Effects of Lipopolysaccharide in Neonatal and Adult Mice.

Glucocorticoids (GCs) are critical modulators of the immune system. The hypothalamic-pituitary-adrenal (HPA) axis regulates circulating GC levels and is stimulated by endotoxins. Lymphoid organs also produce GCs; however, it is not known how lymphoid GC levels are regulated in response to endotoxins. We assessed whether an acute challenge of lipopolysaccharide (LPS) increases lymphoid levels of progesterone and GCs, and expression of steroidogenic enzymes and key HPA axis components (eg, corticotropin-releasing hormone [CRH], adrenocorticotropic hormone [ACTH]). We administered LPS (50 µg/kg intraperitoneally) or vehicle control to male and female C57BL/6J neonatal (postnatal day [PND] 5) and adult (PND90) mice and collected blood, bone marrow, thymus, and spleen 4 hours later. We measured progesterone, 11-deoxycorticosterone, corticosterone, and 11-dehydrocorticosterone via liquid chromatography-tandem mass spectrometry. We measured gene expression of key steroidogenic enzymes (Cyp11b1, Hsd11b1, and Hsd11b2) and HPA axis components (Crh, Crhr1, Pomc, and Mc2r) via quantitative polymerase chain reaction. At PND5, LPS induced greater increases in steroid levels in lymphoid organs than in blood. In contrast, at PND90, LPS induced greater increases in steroid levels in blood than in lymphoid organs. Steroidogenic enzyme transcripts were present in all lymphoid organs, and LPS altered steroidogenic enzyme expression predominantly in the spleen. Lastly, we detected transcripts of key HPA axis components in all lymphoid organs, and there was an effect of LPS in the spleen. Taken together, these data suggest that LPS regulates GC production by lymphoid organs, similar to its effects on the adrenal glands, and the effects of LPS might be mediated by local expression of CRH and ACTH.

Glucocorticoid production in the thymus and brain: Immunosteroids and neurosteroids.

Glucocorticoids (GCs) regulate a myriad of physiological systems, such as the immune and nervous systems. Systemic GC levels in blood are often measured as an indicator of local GC levels in target organs. However, several extra-adrenal organs can produce and metabolize GCs locally. More sensitive and specific methods for GC analysis (i.e., mass spectrometry) allow measurement of local GC levels in small tissue samples with low GC concentrations. Consequently, is it now apparent that systemic GC levels often do not reflect local GC levels. Here, we review the use of systemic GC measurements in clinical and research settings, discuss instances where systemic GC levels do not reflect local GC levels, and present evidence that local GC levels provide useful insights, with a focus on local GC production in the thymus (immunosteroids) and brain (neurosteroids). Lastly, we suggest key areas for further research, such as the roles of immunosteroids and neurosteroids in neonatal programming and the potential clinical relevance of local GC modulators.

Isoflurane stress induces glucocorticoid production in mouse lymphoid organs.

Glucocorticoids (GCs) are secreted by the adrenal glands and locally produced by lymphoid organs. Adrenal GC secretion at baseline and in response to stressors is greatly reduced during the stress hyporesponsive period (SHRP) in neonatal mice (postnatal day (PND) 2-12). It is unknown whether lymphoid GC production increases in response to stressors during the SHRP. Here, we administered an acute stressor (isoflurane anesthesia) to mice before, during, and after the SHRP and measured systemic and local GCs via mass spectrometry. We administered isoflurane, vehicle control (oxygen), or neither (baseline) at PND 1, 5, 9, or 13 and measured progesterone and six GCs in blood, bone marrow, thymus, and spleen. At PND1, blood and lymphoid GC levels were high and did not respond to stress. At PND5, blood GC levels were very low and increased slightly after stress, while lymphoid GC levels were also low but increased greatly after stress. At PND9, blood and lymphoid GC levels were similar at baseline and increased similarly after stress. At PND13, blood GC levels were higher than lymphoid GC levels at baseline, and blood GC levels showed a greater response to stress. These data demonstrate the remarkable plasticity of GC physiology during the postnatal period, show that local steroid levels do not reflect systemic steroid levels, provide insight into the SHRP, and identify a potential mechanism by which early-life stressors can alter immunity in adulthood.

Steroid profiling of glucocorticoids in microdissected mouse brain across development.

Corticosterone is produced by the adrenal glands and also produced locally by other organs, such as the brain. Local levels of corticosterone in specific brain regions during development are not known. Here, we microdissected brain tissue and developed a novel liquid chromatography tandem mass spectrometry method (LC-MS/MS) to measure a panel of seven steroids (including 11-deoxycorticosterone (DOC), corticosterone, and 11-dehydrocorticosterone (DHC) in the blood, hippocampus (HPC), cerebral cortex (CC), and hypothalamus (HYP) of mice at postnatal day (PND) 5, 21, and 90. In a second cohort of mice, we measured the expression of three genes that code for steroidogenic enzymes that regulate corticosterone levels (Cyp11b1, Hsd11b1, and Hsd11b2) in the HPC, CC, and HYP. There were region-specific patterns of steroid levels across development, including higher corticosterone levels in the HPC and HYP than in the blood at PND5. In contrast, corticosterone levels were higher in the blood than in all brain regions at PND21 and PND90. Brain corticosterone levels were not positively correlated with blood corticosterone levels, and correlations across brain regions increased with age. Local corticosterone levels were best predicted by local DOC levels at PND5, but by local DHC levels at PND21 and PND90. Transcripts for the three enzymes were detectable in all samples (with highest expression of Hsd11b1) and showed region-specific changes with age. These data demonstrate that individual brain regions fine-tune local levels of corticosterone during early development and that coupling of glucocorticoid levels across regions increases with age.