Abstract
Background
Hypothyroidism is clinically characterized by a decrease in levels of the circulating thyroid hormones namely thyroxine and triiodothyronine. The main treatment for hypothyroidism is thyroid hormone replacement using levothyroxine to normalize serum thyroid hormone levels.
Objectives
In this study, we explored the metabolic changes in the plasma of patients with hypothyroidism after reaching a euthyroid state with levothyroxine treatment.
Methods
Plasma samples from 18 patients diagnosed as overt hypothyroidism were collected before and after levothyroxine treatment upon reaching a euthyroid state and were analyzed by high-resolution mass spectrometry-based metabolomics. Multivariate and univariate analyses evaluated data to highlight potential metabolic biomarkers.
Results
Liquid chromatography-mass spectrometry-based metabolomics revealed a significant decrease in the levels of ceramide, phosphatidylcholine, triglycerides, acylcarnitine, and peptides after levothyroxine treatment; this could indicate a change in the fatty acid transportation system and an enhanced β-oxidation, compared with a hypothyroid state. At the same time, the decrease in the peptides suggested a shift in protein synthesis. In addition, there was a considerable rise in glycocholic acid following therapy, suggesting the involvement of thyroid hormones in stimulating bile acid production and secretion.
Conclusions
A metabolomic analysis of patients with hypothyroidism revealed significant changes in several metabolites and lipids after treatment. This study showed the value of the metabolomics technique in providing a complementary understanding of the pathophysiology of hypothyroidism and as a crucial tool for examining the molecular impact of levothyroxine treatment on hypothyroidism. It was an important tool for investigating the therapeutic effects of levothyroxine on hypothyroidism at the molecular level.
Introduction
The thyroid hormone affects almost every organ system in the body, including the cardiovascular, central nervous, skeletal, and gastrointestinal systems and metabolism. Tetraiodothyronine or thyroxine (T4) and triiodothyronine (T3) are the main hormones released by the thyroid gland (1). The hypothalamus and anterior pituitary gland produce and release several hormones that work synchronously to maintain the proper feedback mechanisms and homeostasis of T4 and T3 in the bloodstream (2).
Numerous in vitro and in vivo previously conducted studies on animals and humans have examined the effect of thyroid hormones on metabolic balance (3). The research findings prompted the development of different diagnostic tests and therapeutic agents now implemented in the clinical setting. To assess thyroid gland function, several blood tests can be used to measure the levels of thyroid hormones and other related hormones. These tests include those for thyroid-stimulating hormone (TSH), total T4 and T3, and free T4 (FT4) and T3. The other tests for thyroid antibodies and thyroglobulin identify autoimmune thyroid conditions and thyroiditis, respectively (4).
Hypothyroidism, or an underactive thyroid, is a medical condition characterized by a decrease in the thyroid hormones released into the bloodstream (5, 6). It is clinically distinguished by an indirect but broad range of signs, depending on the severity of the hormone deficiency that reflects a hypometabolic state. This metabolic condition is marked by decreased basal metabolic rate, heat production, and increased resting energy expenditure and is associated with increased cholesterol levels, decreased lipolysis, and decreased gluconeogenesis (7).
Hypothyroidism is divided into primary, secondary, and tertiary stages. In primary hypothyroidism, diminished thyroid gland production of hormones causes a compensatory increase in TSH. On the other hand, secondary hypothyroidism is characterized by decreased TSH and T3/T4 levels and is usually caused by pituitary disorders (8). Tertiary hypothyroidism is caused by hypothalamic disorders and results in decreased levels of thyrotropin-releasing hormone released from the hypothalamus, TSH, and T3/T4 (8).
Thyroid hormone replacement is currently the primary treatment method for hypothyroidism (6). According to patient feedback, levothyroxine (LT4) is a synthetic replacement for thyroid tests. The recommended dose for adults is 1.6 µg/kg, adjusted every 8 weeks, depending on the response and thyroid test results. Although patients may have partial symptom alleviation with medicine intake, they typically experience over replacement, which makes them more vulnerable to osteoporosis, liver damage, and cardiovascular diseases. Therefore, novel approaches to managing hypothyroidism are urgently needed.
Metabolomics is a well-known and widely utilized approach to understanding perturbation in an organ system under stress. This approach can provide novel insights into the end product of various biochemical processes within the body, which could reveal the physiological status of that system and identify new therapeutic targets (9, 10, 11, 12, 13, 14, 15). In hypothyroidism, metabolomics has been applied to understand the molecular mechanism in the urine of rats (16), detect metabolic changes associated with the transition from hypothyroidism to euthyroidism using NMR-based metabolomics (17), identify biomarkers to distinguish between hypothyroid and euthyroid individuals (9), and evaluate the effect of herbs with hot properties on hypothyroidism rats (18). The current study used a metabolomic technique based on liquid chromatography-mass spectrometry (LC-MS) to evaluate the metabolic changes connected with hypothyroidism therapy.
Materials and methods
Ethics approval, patient recruitment, and treatment
The study was approved by the King Khalid University Hospital (KKUH), College of Medicine, Riyadh, Saudi Arabia (registration no. E-10-172). The recruited patients, referred to the KKUH Obesity Research Center, were asked to sign a written informed consent form before enrolment. Eighteen overweight patients (10 women and 8 men) with BMI ranging from 25 to 27 kg/m2 who were diagnosed as overt hypothyroidism were included in this study. TSH values > 10 mIU/L and FT4 levels < 12 pmol/L were considered to indicate overt hypothyroidism. All patients were in good health and had no other pathological disorders, such as type 2 diabetes, hypertension, and inflammatory or autoimmune diseases.
The patients were treated with an appropriate dose of LT4. Samples for the hypothyroid group were taken before initiation of treatment, and the post-treatment samples were collected after the TSH levels were normalized after giving the recommended dose. The TSH levels normalized in 15 patients after 6 weeks of therapy, while 2 patients took 8 weeks and 1 patient took 10 weeks to reach euthyroid levels.
Biochemical analysis
Standard clinical tests were performed at the first visit to the hospital after diagnosis and repeated when a euthyroid state was achieved. The following parameters were evaluated using blood samples that were obtained after overnight fasting: FT4, TSH, fasting blood glucose, urea, creatinine, sodium, potassium, aspartate transaminase, alanine transaminase, alkaline phosphatase, total cholesterol, low-density lipoprotein cholesterol (LDL), triglyceride (TG), and high-density lipoprotein cholesterol. An integrated clinical chemistry autoanalyzer (Dimension® Xpand Plus; Siemens Healthcare Diagnostics) was used to determine all parameters for biochemical and hormone studies listed in Table 1.
Biochemical parameters of the hypothyroid patients at baseline and after levothyroxine therapy.
| Parameters | Hypothyroid | Euthyroid | P-value |
|---|---|---|---|
| Patient number | 18 | - | |
| Age (years) | 39 ± 11.7 | - | |
| Female/male | 10/8 | - | |
| FT4 (pmol/L) | 7.9 ± 5.8 | 17.9 ± 3.5 | <0.0001a |
| TSH (mIU/L) median (IQR) | 23.79 (36.39) | 1.20 (1.08) | <0.0001a |
| Fasting blood glucose (mmol/L) | 5.2 ± 0.5 | 5.1 ± 0.3 | 0.067 |
| Urea (mmol/L) | 4.4 ± 0.2 | 4.8 ± 0.8 | 0.318 |
| Creatinine (umol/L) | 74.5 ± 11.1 | 75.7 ± 20.4 | 0.450 |
| Sodium (mmol/L) | 140.1 ± 1.7 | 139.5 ± 0.9 | 0.79 |
| Potassium (mmol/L) | 4.6 ± 0.2 | 4.6 ± 0.1 | 1.20 |
| Aspartate transaminase (IU/L) | 35.3 ± 7.1 | 36.8 ± 7.5 | 0.450 |
| Alanine transaminase (IU/L) | 18.8 ± 6.2 | 17.9 ± 3.5 | 0.21 |
| Alkaline phosphatase (IU/L) | 95.8 ± 19.9 | 97.2 ± 21.1 | 0.86 |
| Total cholesterol (mmol/L) | 4.5 ± 0.4 | 4.9 ± 0.8 | 0.66 |
| Low-density lipoprotein cholesterol (mmol/L) | 1.4 ± 0.3 | 1.2 ± 0.2 | 0.25 |
| Triglyceride (mmol/L) | 1.5 ± 0.2 | 1.4 ± 0.4 | 0.35 |
| High-density lipoprotein cholesterol (mmol/L) | 2.3 ± 0.7 | 2.9 ± 0.5 | 0.32 |
Values are expressed as means ± s.d. median (IQR) and aindicate significant P-value.
FT4, free thyroxine; TSH, thyroid-stimulating hormone.
Sample preparation and liquid chromatography-mass spectrometry analysis
The blood samples were placed in EDTA-coated tubes (BD Vacutainer, USA) and centrifuged at 3000 g for 15 min; after that, plasma samples were kept for analysis at −80°C (19). For plasma metabolite extraction, proteins were precipitated using a mixture of 50% acetonitrile in methanol. The mixtures were processed in a thermomixer (Eppendorf, Germany) at 600 g and 4°C for 1 h, followed by centrifugation at 16,000 g for 10 min at 4°C. The supernatant was transferred into a 1.5 mL tube (Eppendorf) and then evaporated completely in a chamber (SpeedVac; Christ, Germany). The dried samples were suspended using 50% mobile phase A:50% mobile phase B.
Metabolic profiling was performed using a Waters Acquity UPLC system coupled with a Xevo G2-S QTOF mass spectrometer. Metabolites were first separated in a column (100 × 2.1 mm, 2.5 μm) (XSelect; Waters Ltd., Elstree, UK) using ACQUITY UPLC and a binary mobile phase that comprised 0.1% formic acid in dH2O as solvent A and 0.1% formic acid in 50% acetonitrile: methanol as solvent B running at a flow rate of 300 µL/min. The gradient elution was as follows: 0–16 min 95–5% mobile phase A, 16–19 min 5% mobile phase A, 19–20 min 5% to 95% mobile phase A, and 20–22 min 95% mobile phase A. The MS data for the separated metabolites were acquired using positive and negative ionization under the following conditions: source temperature, 150°C; desolvation temperature, 500°C (ESI+) or 140 (ESI−); capillary voltage, 3.20 kV (ESI+) or 3 kV (ESI−); cone voltage, 40 V; desolvation gas flow, 800.0 L/h; and cone gas flow 50 L/h. The collision energy values of the low and high functions were set at off and 10–50 V, respectively, in MSE mode. Quality control samples were prepared by pooling aliquots from all samples and were injected every 10 samples to check the reproducibility of the LC-MS system. All data were collected with Masslynx™ V4.1 workstation (Waters Inc., Milford, MA, USA) in continuum mode.
Data processing and statistical analysis
Peak picking and alignment of the detected ion (m/z, Rt) were processed using Progenesis QI v.3.0 software (Waters Technologies). Thereafter, the processed data were statistically evaluated using several approaches. After data treatment, including log-transformation, mean centering, and Pareto scaling, MetaboAnalyst v5.0 (McGill University, Montreal, QC, Canada) was used to highlight the metabolic changes associated with LT4 treatment (18). In addition, in order to maximize the covariance between the measured data (peak intensities) and the response (class assignment) within the groups, the supervised classification method orthogonal partial least squares discriminant analysis (OPLS-DA) was utilized. This test showed how two variables differed and allowed enhanced interpretation of the metabolic variations between the pre- and post-treatment groups by removing information that had no impact on the discrimination and features. A variable importance in the projection (VIP) value of >1 was considered significant. Thereafter, Mass Profiler Professional software was used for univariate analysis. A volcano plot was used to identify the metabolites that significantly differed between the pre- and post-treatment groups, based on a P-value of <0.05 and a fold change (FC) cutoff of 2. Significantly altered features were chosen based on the univariate and multivariate analyses.
Next, the metabolites were putatively annotated based on the exact mass, isotopic distribution, and fragmentation pattern in different databases, including the Human Metabolome Database (https://hmdb.ca/) and METLIN (http://metlin.scripps.edu), within a mass difference of 5 ppm. Exogenous metabolites, such as those from drugs and environmental exposures, were excluded, and the remaining identified metabolites were retained for further pathway analyses. The receiver operating characteristic (ROC) curve was used to assess the diagnostic ability of the potential biomarkers to discriminate between a hypothyroid and a euthyroid state during LT4 treatment. Sensitivity, specificity, and area under the ROC curve (AUC) were determined using the MetaboAnalyst program (https://www.metaboanalyst.ca/).
Results
Biochemical analysis
The measurements at baseline and after treatment in a euthyroid state are listed in Table 1.
Mass ion detection and significantly altered metabolites
A total of 20,406 mass ion features were detected (13,010 and 7396 in positive and negative 18 modes, respectively) (Supplementary Table 1, see section on supplementary materials given at the end of this article). After applying a frequency threshold (cutoff, 80% of all samples) and excluding 19% of missing values during the peak detection and alignment steps, 16,565 features were retained for statistical analysis. Afterward, these features were normalized by total signal median and then log-transformed and Pareto-scaled before univariate and multivariate analyses.
Based on the multivariate analysis, the OPLS-DA model (Fig. 1) clearly showed separation between the two groups; this result highlighted that the metabolic profile of patients with hypothyroidism significantly differed before and after treatment. Using an extended dataset (n = 100), a permutation test to validate the OPLS-DA model showed R2Y = 0.991 and Q2 = 0.404. The discriminant mass ions with a VIP value of >1 were responsible for the proposed separation and were subjected to univariate analysis to determine their statistical significance.
The orthogonal partial least squares discriminant analysis (OPLS-DA) model. There is evident group separation between the post- and pretreatment groups. The robustness of the created models is evaluated by the fitness of the model (R2Y = 0.991) and the predictive ability (Q2 = 0.404) values on larger data (n = 100). (green dots represent the pretreatment group, and red dots represent the post-treatment group).
Citation: European Thyroid Journal 12, 4; 10.1530/ETJ-23-0062
On univariate analysis and volcano plot, 334 features were significantly dysregulated by LT4 treatment (Fig. 2A). Of these features, 14 were upregulated and 320 were downregulated after treatment (Supplementary Table 2). Supplementary Table 3 shows a summary of 210 of the 334 compounds identified. The top 94 metabolites were identified as endogenous and retained for further pathway and biomarkers analyses (Supplementary Table 4).
(A) This is a volcano plot between the post- and pretreatment groups after applying an 80% filter on all data, (paired t-test, no correction, P ≤ 0.05, FC 2). Of 334 dysregulated metabolites, 14 (red) are upregulated and 320 (blue) are downregulated between the two groups, respectively. Orange and light blue squares refer to metabolites that failed to pass fold change cutoffs and were up- and downregulated, respectively. Gray square metabolites failed to pass both cutoffs. (B) Pathway analysis of the significantly dysregulated metabolites of patients with hypothyroidism before and after treatment. Colors (varying from yellow to red) mean the metabolites are in the data with different significance levels (P-value).
Citation: European Thyroid Journal 12, 4; 10.1530/ETJ-23-0062
The folate biosynthesis; glycerolipid, glycerophospholipid, and phosphatidylinositol signaling systems; and primary bile acid biosynthesis were the most affected pathways between the two groups (Fig. 2B). Compared with the pretreatment group, the post-treatment group had a metabolic plasma profile of significantly higher levels of glycocholic acid, lysophosphatidylinositol (LysoPI) (16:0/0:0), monoacylglyceride (MG) (0:0/24:1(15Z)/0:0), 1,2-diglyceride (DG) (8:0/13:0/0:0), and triacylglycerol (TG) (14:0/14:0/18:1(11Z)) and significantly lower levels of diacylglycerol (DG) (24:0/20:2), phosphatidylcholine (PC) (18:1(12Z)-O(9S,10R)/2:0), asparagine (Asn), aspartic acid (Asp), methionine (Met), tetracosatetraenoyl carnitine, 3-hydroxyhexanoylcarnitine, 2-octanoic, and DG (16:1n7/0:0/18:2n6). These metabolites might be considered promising biomarkers for transitioning from hypothyroid to euthyroid following LT4 treatment.
Figure 3(A) shows the specificity and sensitivity of the potential biomarkers. Three features in the ROC curve had an AUC value of 0.924. Feature ranking in OPLS-DA (Fig. 3B) revealed that LysoPI(16:0/0:0), MG(0:0/24:1(15Z)/0:0), DG(8:0/13:0/0:0), TG(14:0/14:0/18:1(11Z)), and N-acetylasparagine were the most discriminative metabolites.
Receiver operating characteristics (ROC) curve was generated by the orthogonal partial least squares discriminant analysis (OPLS-DA) model, with area under the curve (AUC) values calculated from the combination of 5, 10, 15, 25, 50, and 100 metabolites and (B) loading for significantly altered metabolites before and after hypothyroidism treatment are shown. (C) TG(14:0/14:0l18:1) (AUC, 0.901) and (D) N-acetyl asparagine (AUC, 0.951) as examples of upregulated after hypothyroidism treatment.
Citation: European Thyroid Journal 12, 4; 10.1530/ETJ-23-0062
Discussion
Thyroid hormones affect nearly every cell in the body when they cross the plasma membrane and bind to receptors on the mitochondria that control metabolism, including energy production and glucose oxidation (20, 21). In 1930, hypothyroidism was first reported to have a huge impact on blood lipid components (22). Since then, several studies that examined the association and impact of thyroid hormones on metabolism have shed light on the pathophysiology of hypothyroidism and the mechanism of action of LT4.
In this study, we demonstrated distinct sample clustering and group separation between the hypothyroid and euthyroid groups as a result of LT4 treatment, as shown in the OPLS-DA model, which had acceptable goodness of fit and prediction (23). Moreover, LysoPI(16:0/0:0), MG(0:0/24:1(15Z)/0:0), DG(8:0/13:0/0:0), and TG(14:0/14:0/18:1(11Z)) were the highest ranking metabolites that could discriminate between hypothyroidism and euthyroidism before and after treatment, respectively. Therefore, these metabolites may be further evaluated to monitor treatment with LT4. The significant increase in the T4 level, which is expected to significantly decrease the TSH level, in the euthyroid state from the T4 level in the hypothyroid state reflected a positive therapeutic effect of LT4 on hypothyroidism.
Inadequate levels of circulating thyroid hormones secondary to hypothyroidism decrease hepatic cholesterol synthesis, probably by inhibition of hydroxy methylglutaryl coenzyme (HMG-CoA) reductase. Furthermore, thyroid hormones are known to stimulate transcription of the LDL receptor and HMG-CoA reductase genes. HMG-CoA reductase is expressed in the liver and stimulates the production of cholesterol. However, a decrease in hepatic cholesterol synthesis is outweighed by the action of thyroid hormones on LDL cholesterol receptor expression, resulting in net accumulation of serum LDL cholesterol. Therefore, compared with healthy subjects, patients with hypothyroidism usually present with increased TC and LDL concentrations. Moreover, they manifest elevated TG levels secondary to a decrease in the activity of lipoprotein lipase, which is an enzyme that degrades circulating TG in the bloodstream (24, 25, 26, 27). LT4 can be extremely effective for the treatment of hyperlipidemia associated with hypothyroidism (26). However, our results showed no notable changes in the serum concentrations of TC, LDL, and TG after LT4 treatment. This implied that a longer time may be needed to observe for changes in the lipid profile, even if a euthyroid state has been achieved. Another reason for the absence of significant changes in the lipid profile after LT4 treatment may be the fact that the recruited patients had lipid profiles within the accepted range before treatment.
In addition to dyslipidemia, which is consistently seen in hypothyroidism, considerably greater plasma levels of glycerophospholipids in individuals with clinical and subclinical hypothyroidism than in euthyroid individuals were previously reported (9). Moreover, several animal studies reported that thyroid hormones could greatly affect sphingolipid metabolism, de novo ceramide synthesis, glycerophospholipid metabolism, and fatty acid β-oxidation in the liver (28, 29). In other studies, the level of ceramides increased upon LT4 administration to hypothyroid tissue and was even higher in hyperthyroid hepatocytes, liver tissue, and heart muscle (29, 30). LT4 is believed to increase the activity of ceramide phosphocholine transferase, which is an enzyme that transfers the phosphocholine group from phosphatidyl cholines (PC) to ceramide to generate sphingomyelin (SM) and diglycerides (DG) (31). In a recent plasma metabolomics profiling study by Feifei Shao et al., the level of ceramide was higher in individuals with clinical and subclinical hypothyroidism than in the control group (9). Consistent with previous reports, this present study found low plasma levels of ceramides, trihexosyl ceramide (d18:1/22:0), and trihexosyl ceramide (d18:1/24:0) upon LT4 treatment. In addition, the downregulation of ten metabolites identified as PC following LT4 treatment could reflect that more PCs were redirected to the liver and consumed during SM synthesis, while the liver lipoprotein-associated PC efflux decreased. This provided evidence of the key role of thyroid hormones in lipid synthesis and metabolism, particularly sphingolipid metabolism in the liver. Moreover, monitoring plasma and/or liver ceramides and probably SM could serve as gold standard regulators of hypothyroid treatment and may be used to monitor LT4 overdose.
An elevated level of DG is a common finding in hypothyroidism. DG accumulation had been attributed to multiple routes, including lipid efflux, ceramide, polyphosphoinositide, and PC hydrolysis, in addition to reduced hepatic lipogenesis, all of which resulted in reduced levels of PC and TG (32). Sustained elevation of DG level is believed to be associated with decreased protein kinase C (PKC) activity (33). PKC comprises a family of closely related enzymes that are activated by the second messenger calcium (Ca2+) and DG, depending on the duration and magnitude of these signals. Inappropriate DG buildup promotes different cellular transformations, including hepatocarcinogenesis, hyperglycemia, hypertriglyceridemia, other cardiovascular complications of diabetes, and hypothyroidism (34, 35, 36, 37, 38). In addition, the work of Bansode R et al. on mouse liver and skeletal muscle showed reductions in adipose tissue depots and TG content upon deletion of the PKCβ isoform (37, 38). Therefore, the level of intracellular DG must be closely regulated. Upon LT4 administration, PKC is reactivated, the amount of DG synthesized de novo will be decreased, and the level of TG is expected to increase (33, 39, 40). In the current study, the levels of three of four identified DGs were downregulated and the level of one of two identified TGs was upregulated after treatment. Although this result coincides with that of previous studies, not all DGs and TGs follow the same trend, thereby, suggesting that a complete metabolic change was not achieved after reaching a euthyroid state with treatment, because T3 and T4 levels do not reflect all the changes reversed in hypothyroidism. Consistent with our findings, the results of C. Piras et al. showed that upon LT4 administration to 18 patients with overt primary hypothyroidism, the metabolic changes persisted despite the normalization of serum TSH and thyroid hormone concentrations (17). Therefore, the metabolomics technique may help integrate traditional hormone assays and more effectively assess the accomplishment of a euthyroid state.
Thyroid hormones are known to increase the transcription of fibroblast growth factor 21 (FGF21), which is a coded protein that serves as an endocrine factor and a key metabolic regulator of glucose and lipids in adipose tissue (41). A reduced level of FGF21 was reported in hypothyroidism, whereas an elevated FGF21 level was observed in hyperthyroidism (42, 43). FGF21 inhibits the synthesis of fatty acids, cholesterol, and TG and stimulates fatty acid uptake and β-oxidation, which enhance thermogenesis and increase energy expenditure (44, 45). β-oxidation is a multistep process that takes place in the mitochondria, where fatty acids are broken down by different tissues to produce acetyl-CoA units that can be employed in the tricarboxylic acid (TCA) cycle to produce ATP. Therefore, accumulation of plasma long-chain acylcarnitine in patients with hypothyroidism supports the decrease in the activity of fatty acid oxidation, which may indicate dysfunction in the long-chain fatty acid transport protein (FATP) 1 (28) or incomplete β-oxidation relative to the TCA cycle flux (46). In this study, a low level of long-chain acylcarnitine was detected upon treatment; this reflected enhanced β-oxidation in euthyroid patients than in hypothyroid patients. Thyroid hormones activate Ca2+/calmodulin-dependent protein kinase-β, which, in turn, mediates phosphorylation of AMP-activated protein kinase (AMPK) and increases fatty acid transport from the periphery into the cells by translocation of CD36 through activation of Ras-related protein Rab-8A (Rab8a) or FATP on the plasma membrane. In addition, AMPK activates acetyl-CoA carboxylase (ACC), which is an enzyme that catalyzes the conversion of acetyl-CoA to malonyl-CoA, which promotes fatty acid synthesis (47). However, when activated by AMPK, ACC activity is inhibited, thereby decreasing the amount of malonyl-CoA released and activating carnitine palmitoyl transferase 1, which is an outer mitochondrial membrane protein that converts acyl-CoA to acyl carnitine, accompanied by an increase in fatty acid trafficking into mitochondria (47). In addition, thyroid hormones increase the production of several mitochondrial enzymes required for β-oxidation, including mitochondrial uncoupling protein (48), pyruvate dehydrogenase kinase (49), and medium-chain acyl-CoA dehydrogenase (50).
Furthermore, glycocholic acid was increased upon restoration of a euthyroid state. The thyroid hormone facilitates the conversion of cholesterol to bile acid by stimulating the rate-limiting enzyme cholesterol 7α-hydroxylase through hepatic TH β receptors and stimulates bile acid secretion in the liver and intestine (51, 52, 53). Therefore, an increase in the glycocholic acid concentration was expected in the post-treatment samples.
Compared with a hypothyroid state, a euthyroid state led to the downregulation of small peptides upon achieving normal plasma FT4; this suggested a change in protein metabolism. Since 1956, whole-body protein breakdown and decreased synthesis had been documented upon disruption in thyroid hormone levels (either hypothyroidism or hyperthyroidism). However, the reversal of these changes following treatment indicated the importance of sustaining normal thyroid hormone levels for maintaining anabolic effects (54, 55).
The findings from the study have to be taken into consideration keeping in mind the small sample size and lack of gender base analysis and the time duration to reach euthyroid levels. Another aspect that requires consideration is the observation that our study demonstrates notable enhancements in the metabolomic profile following thyroxine treatment. However, it remains uncertain at which levels of FT4 there is a full normalization of the metabolomic profile, representing genuine cellular euthyroidism, rather than simply falling within the laboratory's normal FT4 range. Additional research is imperative to validate the findings obtained in this study.
Conclusions
A metabolomic analysis of patients with hypothyroidism revealed significant changes in the level of several metabolites and lipids after treatment. This study showed the value of the metabolomics technique in providing a complementary understanding of the pathophysiology of hypothyroidism and as a crucial tool for examining the molecular impact of LT4 treatment on hypothyroidism. However, it is important to note that metabolic changes might persist upon reaching a euthyroid state after treatment.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/ETJ-23-0062.
Declaration of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be considered as potential conflict of interest.
Funding
This project was funded by the National Plan for Science, Technology and Innovation (MAARIFAH), King Abdulaziz City for Science and Technology, Saudi Arabia (Project No. 08-MED513-02).
Data availability statement
Metabolomics data were deposited to the EMBL-EBI MetaboLights database with the identifier MTBLS7600. The complete dataset can be accessed at https://www.ebi.ac.uk/metabolights/MTBLS7600 (30 March 2023).
Statement of ethics and informed consent statement
The study was conducted in accordance with the Declaration of Helsinki, and all procedures and protocols, including the clinical samples, were reviewed and approved by the ethics committee of the College of Medicine, King Saud University, Riyadh, Saudi Arabia (registration no. E-10-172). Informed consent was obtained from all participants involved in the study.
Author contribution statement
Hicham Benabdelkamel, Anas M. Abdel Rahman, and Assim A. Alfadda conceived the idea and designed the study. Assim A. Alfadda and Afshan Masood were involved in patient recruitment. Hicham Benabdelkamel, Anas M. Abdel Rahman, Reem H. AlMalki, and Mohthash Musambil performed the metabolomics laboratory work. Anas M. Abdel Rahman, Reem H. AlMalki, and Lina A. Dahabiyeh carried out data analysis. Hicham Benabdelkamel, Malak A. Jaber, Lina A. Dahabiyeh, Afshan Masood, Reem H. AlMalki, Anas M. Abdel Rahman, and Assim A. Alfadda wrote the manuscript. All authors have read and approved the final manuscript.
Acknowledgements
We wish to acknowledge Mr Majed Kenneth and Ms Amina Fallata for their assistance in the laboratory work.
References
- 1↑
Carvalho DP, & Dupuy C. Thyroid hormone biosynthesis and release. Molecular and Cellular Endocrinology 2017 458 6–15. (https://doi.org/10.1016/j.mce.2017.01.038)
- 2↑
Chiamolera MI, & Wondisford FE. Minireview: thyrotropin-releasing hormone and the thyroid hormone feedback mechanism. Endocrinology 2009 150 1091–1096. (https://doi.org/10.1210/en.2008-1795)
- 3↑
Duntas LH. Thyroid disease and lipids. Thyroid 2002 12 287–293. (https://doi.org/10.1089/10507250252949405)
- 4↑
Feldt-Rasmussen U, & Klose M. Clinical strategies in the testing of thyroid function. In Feingold KR, Anawalt B, & Boyce A, Eds. Endotext. South Dartmouth , MA, USA: MDText.com, Inc, 2020. (available at: https://www.ncbi.nlm.nih.gov/books/NBK285558/)
- 5↑
Patil N, Rehman A, & Jialal I. Hypothyroidism. In StatPearls : Treasure Island , FL, USA: StatPearls Publishing, 2022. (available at: https://www.ncbi.nlm.nih.gov/books/NBK519536/)
- 6↑
Jonklaas J, Bianco AC, Bauer AJ, Burman KD, Cappola AR, Celi FS, Cooper DS, Kim BW, Peeters RP, Rosenthal MS, et al.Guidelines for the treatment of hypothyroidism: prepared by the American Thyroid Association task force on thyroid hormone replacement. Thyroid 2014 24 1670–1751. (https://doi.org/10.1089/thy.2014.0028)
- 7↑
Mullur R, Liu YY, & Brent GA. Thyroid hormone regulation of metabolism. Physiological Reviews 2014 94 355–382 (https://doi.org/10.1152/physrev.00030.2013)
- 8↑
Garber JR, Cobin RH, Gharib H, Hennessey JV, Klein I, Mechanick JI, Pessah-Pollack R, Singer PA, Woeber KA & American Association of Clinical Endocrinologists and American Thyroid Association Taskforce on Hypothyroidism in Adults. Clinical practice guidelines for hypothyroidism in adults: cosponsored by the American Association of Clinical Endocrinologists and the American Thyroid Association. Endocrine Practice 2012 18 988–1028. (https://doi.org/10.4158/EP12280.GL)
- 9↑
Shao F, Li R, Guo Q, Qin R, Su W, Yin H, & Tian L. Plasma metabolomics reveals systemic metabolic alterations of subclinical and clinical hypothyroidism. Journal of Clinical Endocrinology and Metabolism 2022 108 13–25 (https://doi.org/10.1210/clinem/dgac555)
- 10↑
Zhang A, Sun H, Yan G, Wang P, & Wang X. Metabolomics for biomarker discovery: moving to the clinic. BioMed Research International 2015 2015 354671 (https://doi.org/10.1155/2015/354671)
- 11↑
Saigusa D, Matsukawa N, Hishinuma E, & Koshiba S. Identification of biomarkers to diagnose diseases and find adverse drug reactions by metabolomics. Drug Metabolism and Pharmacokinetics 2021 37 100373 (https://doi.org/10.1016/j.dmpk.2020.11.008)
- 12↑
Ashrafian H, Sounderajah V, Glen R, Ebbels T, Blaise BJ, Kalra D, Kultima K, Spjuth O, Tenori L, Salek RM, et al.Metabolomics: the stethoscope for the twenty-first century. Medical Principles and Practice 2021 30 301–310. (https://doi.org/10.1159/000513545)
- 13↑
Dahabiyeh LA, Malkawi AK, Wang X, Colak D, Mujamammi AH, Sabi EM, Li L, Dasouki M, & Abdel Rahman AM. Dexamethasone-Induced Perturbations in Tissue metabolomics Revealed by Chemical Isotope Labeling LC-MS analysis. Metabolites 2020 10. (https://doi.org/10.3390/metabo10020042)
- 14↑
Aleidi SM, Al-Ansari MM, Alnehmi EA, Malkawi AK, Alodaib A, Alshaker M, Benabdelkamel H, & Rahman AM. Lipidomics profiling of patients with low bone mineral density (LBMD). International Journal of Molecular Sciences 2022 23. (https://doi.org/10.3390/ijms231912017)
- 15↑
LA MM, AlMalki RH, Arafat T, Benabdelkamel H, Alfadda AA, & Rahman A. Lipids Alterations Associated with Metformin in Healthy Subjects: An Investigation Using Mass Spectrometry Shotgun Approach. International Journal of Molecular Sciences 2022 23 11478. (https://doi.org/10.3390/ijms231911478)
- 16↑
Wu S, Tan G, Dong X, Zhu Z, Li W, Lou Z, & Chai Y. Metabolic profiling provides a system understanding of hypothyroidism in rats and its application. PLoS One 2013 8 e55599 (https://doi.org/10.1371/journal.pone.0055599)
- 17↑
Piras P, Pibiri M, Leoni VP, Balsamo A, Tronci L, Arisci N, Mariotti S, & Atzori L. Analysis of metabolomics profile in hypothyroid patients before and after thyroid hormone replacement. Journal of Endocrinological Investigation 2021 44 1309–1 310. (https://doi.org/10.1007/s40618-020-01434-y)
- 18↑
Wang YY, Sun YP, Yang BY, Wang QH, & Kuang HX. Application of metabolomics and network analysis to reveal the ameliorating effect of four typical "hot" property herbs on hypothyroidism rats. Frontiers in Pharmacology 2022 13 955905 (https://doi.org/10.3389/fphar.2022.955905)
- 19↑
Jaber MA, Benabdelkamel H, Dahabiyeh LA, Masood A, AlMalki RH, Musambil M, Alfadda AA, & Abdel Rahman AM. The metabolomics approach revealed a distinctive metabolomics pattern associated with hyperthyroidism treatment. Frontiers in Endocrinology (Lausanne) 2022 13 1050201. (https://doi.org/10.3389/fendo.2022.1050201)
- 20↑
Alfadda AA, Benabdelkamel H, Masood A, Jammah AA, & Ekhzaimy AA. Differences in the plasma proteome of patients with hypothyroidism before and after thyroid hormone replacement: A proteomic analysis. International Journal of Molecular Sciences 2018 19. (https://doi.org/10.3390/ijms19010088)
- 21↑
Masood A, Benabdelkamel H, Jammah AA, Ekhzaimy AA, & Alfadda AA. Identification of protein changes in the urine of hypothyroid patients treated with thyroxine using proteomics approach. ACS Omega 2021 6 2367–2378. (https://doi.org/10.1021/acsomega.0c05686)
- 22↑
Rizos CV, Elisaf MS, & Liberopoulos EN. Effects of thyroid dysfunction on lipid profile. Open Cardiovascular Medicine Journal 2011 5 76–84. (https://doi.org/10.2174/1874192401105010076)
- 23↑
Westerhuis JA, Hoefsloot HCJ, Smit S, Vis DJ, Smilde AK, van Velzen EJJ, van Duijnhoven JPM, & van Dorsten FA. Assessment of PLSDA cross validation. Metabolomics 2008 4 81–89 (https://doi.org/10.1007/s11306-007-0099-6)
- 24↑
Choi JW, & Choi HS. The regulatory effects of thyroid hormone on the activity of 3-hydroxy-3-methylglutaryl coenzyme A reductase. Endocrine Research 2000 26 1–21. (https://doi.org/10.1080/07435800009040142)
- 25↑
Jia L, Betters JL, & Yu L. Niemann-pick C1-like 1 (NPC1L1) protein in intestinal and hepatic cholesterol transport. Annual Review of Physiology 2011 73 239–259. (https://doi.org/10.1146/annurev-physiol-012110-142233)
- 26↑
Pearce EN. Hypothyroidism and dyslipidemia: modern concepts and approaches. Current Cardiology Reports 2004 6 451–456. (https://doi.org/10.1007/s11886-004-0054-3)
- 27↑
Liu H, & Peng D. Update on dyslipidemia in hypothyroidism: the mechanism of dyslipidemia in hypothyroidism. Endocrine Connections 2022 11. (https://doi.org/10.1530/EC-21-0002)
- 28↑
Wu S, Chen S, Dong X, Tan G, Li W, Lou Z, Zhu Z, & Chai Y. Lipidomic profiling reveals significant alterations in lipid biochemistry in hypothyroid rat cerebellum and the therapeutic effects of sini decoction. Journal of Ethnopharmacology 2015 159 262–273. (https://doi.org/10.1016/j.jep.2014.11.033)
- 29↑
Babenko NA. Long- and short-term effects of thyroxine on sphingolipid metabolism in rat liver. Medical Science Monitor 2005 11 BR131–BR138.
- 30↑
Mikłosz A, Łukaszuk B, Chabowski A, Rogowski F, Kurek K, & Żendzian-Piotrowska M. Hyperthyroidism evokes myocardial ceramide accumulation. Cellular Physiology and Biochemistry 2015 35 755–766. (https://doi.org/10.1159/000369735)
- 31↑
Vater Y, Levy A, Martay K, Hunter C, & Weinbroum AA. Adjuvant drugs for end-stage liver failure and transplantation. Medical Science Monitor 2004 10 RA77–RA88.
- 32↑
Nishizuka Y. Intracellular signaling by hydrolysis of phospholipids and activation of protein kinase C. Science (New York, NY) 1992 258 607–614. (https://doi.org/10.1126/science.1411571)
- 33↑
Krasilnikova OA, Kavok NS, & Babenko NA. Drug-induced and postnatal hypothyroidism impairs the accumulation of diacylglycerol in liver and liver cell plasma membranes. BMC Physiology 2002 2 12. (https://doi.org/10.1186/1472-6793-2-12)
- 34↑
Gu X, Al Dubayee M, Alshahrani A, Masood A, Benabdelkamel H, Zahra M, Li L, Abdel Rahman AM, & Aljada A. Distinctive metabolomics patterns associated with insulin resistance and type 2 diabetes mellitus. Frontiers in Molecular Biosciences 2020 7 609806. (https://doi.org/10.3389/fmolb.2020.609806)
- 35↑
Qu X, Seale JP, & Donnelly R. R. Tissue and isoform-selective activation of protein kinase C in insulin-resistant obese Zucker rats - effects of feeding. Journal of Endocrinology 1999 162 207–214. (https://doi.org/10.1677/joe.0.1620207)
- 36↑
Okumura K, Matsui H, Kikuchi M, Mukawa H, Toki Y, & Ito T. Triiodothyronine decreases the accumulation of 1,2-diacylglycerol in rat hearts. Canadian Journal of Cardiology 1995 11 565–572.
- 37↑
Bansode RR, Huang W, Roy SK, Mehta M, & Mehta KD. Protein kinase C deficiency increases fatty acid oxidation and reduces fat storage. Journal of Biological Chemistry 2008 283 231–236. (https://doi.org/10.1074/jbc.M707268200)
- 38↑
Schmitz-Peiffer C. The tail wagging the dog--regulation of lipid metabolism by protein kinase C. FEBS Journal 2013 280 5371–5383. (https://doi.org/10.1111/febs.12285)
- 39↑
Lawrence WD, Schoenl M, & Davis PJ. Stimulation in vitro of rabbit erythrocyte cytosol phospholipid-dependent protein kinase activity. A novel action of thyroid hormone. Journal of Biological Chemistry 1989 264 4766–4768. (https://doi.org/10.1016/S0021-9258(1883655-9)
- 40↑
Kavok NS, Krasilnikova OA, & Babenko NA. Thyroxine signal transduction in liver cells involves phospholipase C and phospholipase D activation. Genomic independent action of thyroid hormone. BMC Cell Biology 2001 2 5 (https://doi.org/10.1186/1471-2121-2-5)
- 41↑
Szczepanska E, & Gietka-Czernel M. FGF21: A novel regulator of glucose and lipid metabolism and whole-body energy balance. Hormone and Metabolic Research 2022 54 203–211 (https://doi.org/10.1055/a-1778-4159)
- 42↑
Xiao F, Zeng J, Huang P, Yan B, Zeng X, Liu C, Shi X, Wang L, Song H, Lin M, et al.Independent association of serum fibroblast growth factor 21 levels with impaired liver enzymes in hyperthyroid patients. Frontiers in Endocrinology (Lausanne) 2018 9 800. (https://doi.org/10.3389/fendo.2018.00800)
- 43↑
Wang G, Liu J, Yang N, Hu Y, Zhang H, Miao L, Yao Z, & Xu Y. Levothyroxine treatment restored the decreased circulating fibroblast growth factor 21 levels in patients with hypothyroidism. European Journal of Internal Medicine 2016 31 94–98. (https://doi.org/10.1016/j.ejim.2016.03.028)
- 44↑
Chau MD, Gao J, Yang Q, Wu Z, & Gromada J. Fibroblast growth factor 21 regulates energy metabolism by activating the AMPK-SIRT1-PGC-1alpha pathway. Proceedings of the National Academy of Sciences of the United States of America 2010 107 12553–12558. (https://doi.org/10.1073/pnas.1006962107)
- 45↑
Jeon SM. Regulation and function of AMPK in physiology and diseases. Experimental and Molecular Medicine 2016 48 e245. (https://doi.org/10.1038/emm.2016.81)
- 46↑
Ribel-Madsen A, Ribel-Madsen R, Brøns C, Newgard CB, Vaag AA, & Hellgren LI. Plasma acylcarnitine profiling indicates increased fatty acid oxidation relative to tricarboxylic acid cycle capacity in young, healthy low birth weight men. Physiological Reports 2016 4. (https://doi.org/10.14814/phy2.12977)
- 47↑
Sayre NL, & Lechleiter JD. Fatty acid metabolism and thyroid hormones. Current Trends in Endocrinology 2012 6 65–76.
- 48↑
Jekabsons MB, Gregoire FM, Schonfeld-Warden NA, Warden CH, & Horwitz BA. T(3) stimulates resting metabolism and UCP-2 and UCP-3 mRNA but not nonphosphorylating mitochondrial respiration in mice. American Journal of Physiology 1999 277 E380–E389. (https://doi.org/10.1152/ajpendo.1999.277.2.E380)
- 49↑
Holness MJ, Bulmer K, Smith ND, & Sugden MC. Investigation of potential mechanisms regulating protein expression of hepatic pyruvate dehydrogenase kinase isoforms 2 and 4 by fatty acids and thyroid hormone. Biochemical Journal 2003 369 687–695. (https://doi.org/10.1042/BJ20021509)
- 50↑
Djouadi F, Riveau B, Merlet-Benichou C, & Bastin J. Tissue-specific regulation of medium-chain acyl-CoA dehydrogenase gene by thyroid hormones in the developing rat. Biochemical Journal 1997 324(Pt 1) 289–294. (https://doi.org/10.1042/bj3240289)
- 51↑
Layden TJ, & Boyer JL. The effect of thyroid hormone on bile salt-independent bile flow and Na+, K+ -ATPase activity in liver plasma membranes enriched in bile canaliculi. Journal of Clinical Investigation 1976 57 1009–1018. (https://doi.org/10.1172/JCI108342)
- 52↑
Bonde Y, Plösch T, Kuipers F, Angelin B, & Rudling M. Stimulation of murine biliary cholesterol secretion by thyroid hormone is dependent on a functional ABCG5/G8 complex. Hepatology 2012 56 1828–1837. (https://doi.org/10.1002/hep.25861)
- 53↑
Bonde Y, Breuer O, Lütjohann D, Sjöberg S, Angelin B, & Rudling M. Thyroid hormone reduces PCSK9 and stimulates bile acid synthesis in humans. Journal of Lipid Research 2014 55 2408–2415. (https://doi.org/10.1194/jlr.M051664)
- 54↑
Crispell KR, Parson W, & Hollifield G. A study of the rate of protein synthesis before and during the administration of L-triiodothyronine to patients with myxedema and healthy volunteers using N-15 glycine. Journal of Clinical Investigation 1956 35 164–169. (https://doi.org/10.1172/JCI103261)
- 55↑
Morrison WL, Gibson JN, Jung RT, & Rennie MJ. Skeletal muscle and whole body protein turnover in thyroid disease. European Journal of Clinical Investigation 1988 18 62–68. (https://doi.org/10.1111/j.1365-2362.1988.tb01167.x)
