Role of Metabolomics in Precision Medicine in the Context of Systemic Lupus Erythematosus and Lupus Nephritis

Abstract

Systemic lupus erythematosus (SLE) is an autoimmune, multisystemic disease, the underlying causes of which are not fully understood. Clinically, SLE is a heterogeneous disease, and its clinical spectrum overlaps with other rheumatic diseases. Even though multiple organs can be involved in the progression of the disease, the kidney is the main indicator of morbidity and mortality in SLE, a condition known as lupus nephritis (LN). The diagnosis of LN still requires an invasive procedure that becomes impractical when monitoring patients with a confirmed diagnosis, and the described biomarkers do not meet the characteristics of a good biomarker. Recently, “omics” technologies have revolutionized the field of the molecular diagnosis, including autoimmune diseases. Metabolomics has been an extremely useful tool for identifying non-invasive biomarkers in the clinical context and is proposed as a powerful and promising tool to differentiate between individuals with SLE without renal damage and patients with SLE with renal involvement. In addition, this methodology will allow the categorization of patients according to renal damage, promising personalized management of LN in the context of SLE. Here, we review the potential of untargeted metabolomic approaches using LC/GC-MS and MNR to identify potential biomarkers for renal damage in patients with SLE.

1. Introduction

Systemic lupus erythematosus (SLE) presents a high clinical complexity due to its wide variety of expression patterns. It can affect any organ and follows a relapsing-remitting course, with periods of activity and inactivity. Its multisystemic nature and the wide range of symptoms it can manifest require a careful diagnostic study and the use of differentiated therapy based on the severity of the condition [1]. The kidney is affected to varying degrees; in the majority of SLE patients, immune complex deposits are observed, only detectable at present by electron microscopy or immunofluorescence [2]. Renal involvement, or lupus nephritis (LN), can be silent or manifest with clinical patterns known as nephritic or nephrotic syndrome, high blood pressure, and progressive deterioration of kidney function, eventually leading to end-stage renal failure [3]. Approximately 75% of patients with SLE develop renal involvement at some point in the course of the disease, of which 25 to 50% develop nephropathy in the early stages of the condition [4]. The percentages for different types of LN are 25% for focal (class III according to the World Health Organization), 37% for diffuse (class IV), and 13% for membranous (class V) [5].

LN is a frequent and severe manifestation of SLE. The prevalence of LN varies depending on age, gender, ethnicity, and other factors, occurring in approximately 50–60% of SLE patients, of which up to 10% will develop end-stage renal disease [6, 7]. The incidence of SLE varies by region, with higher rates in the United States compared to Europe [8, 9], and by ethnicity, with higher rates in Black individuals (34–51%), followed by Hispanics (31–43%) and Asians (33–55%). There is also a relationship between ethnicity and the severity of symptoms, with Black individuals exhibiting more severe symptoms [10, 11].

The pathogenesis may involve a variety of mechanisms, including gene expression leading to neutrophil activation, increased production of interferon and other proinflammatory myeloid mediators, release of neutrophil extracellular traps (NETs), and complement activation. The glomerular lesion pattern observed in SLE is generally related to the binding of antibodies to multiple intrarenal autoantigens or the formation of circulating antigens, forming immune complexes containing autoantibodies deposited in various parts of the glomerulus, mainly due to antibodies against double-stranded DNA (anti-dsDNA or anti-DNA) [10, 12, 13]. Molecular phenotyping of the kidney or urine can identify the current immunological and/or injury process affecting each individual, leaving a metabolic footprint. This allows for the evaluation of diagnostic and/or monitoring biomarkers for LN and expands knowledge of pathogenesis [14, 15, 16].

Renal biopsy remains the gold standard for confirming the diagnosis of LN and determining the histological type and extent of damage, using the International Society of Nephrology/Renal Pathology Society (ISN/RPS) classification [17]. Fluid biomarkers, which are validated indicators of physiological or pathological processes or treatment response, are currently considered less invasive alternatives or complementary means of evaluating SLE-related kidney disease [18]. Traditional laboratory biomarkers include immunoserological tests such as anti-double-stranded DNA (dsDNA) levels, complement levels, and parameters associated with renal disease, such as 24-hour urinary protein or urinary protein/creatinine ratio, urine sediment, and glomerular filtration rate (GFR). They are well-established tools for clinical LN evaluation. However, they do not detect LN at early stages of the disease, presenting limited sensitivity and specificity for distinguishing between active disease and chronic lesions [19].

Recent technological advancements, including deep proteomic and metabolic analysis tests, have contributed to the evidence of new biomarkers, some of which have shown performance equivalent to traditional markers [20, 21].

The concept of metabolomics includes the systematic detection and quantification of low-molecular-weight molecules produced by cells (endogenous metabolites) that leave unique chemical traces [15, 22]. Metabonomics, on the other hand, is defined as the quantitative measurement of changes in these metabolic profiles [23]. The theoretical basis lies in the idea that disturbances in a biological system, caused by a disease, for example, will be detectable as changes in the concentrations of certain metabolites. This also provides the possibility of revealing associated underlying biochemical phenomena, thus offering insights that help in developing a better understanding of disease pathogenesis [15].

In some cases, it may be possible to identify a single robust diagnostic metabolite, as in congenital metabolic diseases. Still, in many cases, the disturbances are more subtle, involving the activation of multiple enzymatic (biochemical) pathways, making it unlikely that a single biomarker will be specific enough for diagnostic purposes [15]. However, by using multivariate statistics, it may be possible to describe changes in biomarker patterns rather than individual biomarkers [15]. Nuclear magnetic resonance spectroscopy (NMR) and liquid and gas chromatography coupled with tandem mass spectrometry (LC-MS/MS and GC-MS/MS) are the most widely used analytical techniques for metabolomic studies due to their great power to identify new metabolites in different biofluids in an unbiased manner. In this book chapter, we will review the literature related to the application of metabolomic techniques for the study of biomarkers in individuals with SLE and LN and the impact these technologies can have on personalized medicine.

2. Role of metabolism disturbances in the context of SLE

2.1 TCA cycle

Within the many possible explanations for the pathogenesis of SLE, it has been established that abnormal T cell activation and cell apoptosis, which are highly energy-dependent processes, play a major role in the progression of the disease. SLE patients exhibit mitochondrial abnormalities that affect energy production and cell survival [24, 25]. A study conducted by Yan et al. that applied gas chromatography–mass spectrometry to analyze serum samples found that SLE patients had lower levels of fumarate and citrate (Table 1) [26], which are intermediates of the energy-producing tricarboxylic acid (TCA) cycle. This suggests that SLE disrupts cellular energy metabolism, and the extent of this dysfunction may be related to disease activity. This same study also reported that patients with SLE had lower levels of glucose in the sera compared to healthy controls, attributed to an upregulation of the pentose phosphate pathway in SLE, which requires a higher rate of glucose consumption and a subsequent disturbance of glucose utilization [26, 34].

Metabolite	Type of sample	Disease	Method applied	References
Fumarate, citrate, glucose	Serum	SLE	GC-MS	[26]
Glutathione	Serum	SLE	HPLC	[24]
Aminomalonate, threonate, alpha-tocopherol	Urine	SLE	GC-MS	[14]
L-pyroglutamic acid, adenosine	Serum	SLE	LC-MS-MS	[27]
Adenosine	Feces	SLE	UHPLC-MS-MS	[28]
Arginine	Serum	SLE	LC-MS and GC-MS	[29]
2-Hydroxyisovalerate, 2-keto-3-methylvalerate, and 2-ketoisocaproate	Serum	SLE	GC-MS	[26]
Glutamine	Brain Imaging	SLE	H single-voxel spectroscopy	[30]
Tryptophan, kynurenine	Serum	SLE	UHPLC	[31]
Kynurenine and Kynurenic acid	Peripheral Blood Lymphocytes	SLE	LC-MS-MS	[32]
2-hydroxyisobutyrate, glycerol, oleic acid, arachidonic acid, 1-monopalmitin, and linoleic acid	Serum	SLE	GC-MS	[26]
Deoxycholic acid, glycolic acid, ursodeoxycholic acid, and arachidonic acid	Feces and serum	SLE	LC-MS	[33]

Table 1.

Metabolites in SLE.

2.2 Oxidative stress

SLE is characterized not only by abnormal T cell activation but also by an increase in oxidative stress [35, 36], which has been associated to alterations in the immune system and autoantibody production, as well as to the cardiovascular complications of the disease [26, 37]. Glutathione is an important antioxidant that has been found to have significant implications in nutrient metabolism and regulation of cellular events such as gene expression, synthesis of proteins and DNA, cellular proliferation, apoptosis, signal transduction, and protein glutathionylation [38, 39]. By the subsequent actions of glutamate cysteine ligase (GCL) and GSH synthetase, this antioxidant is created from constituent amino acids including cysteine, glutamate, and pyroglutamarate [26, 38]. Glutathione is responsible for suppressing the development of Th17 cells, a subset of T cells involved in autoimmune responses, and for reducing intracellular levels of reactive oxygen species (ROS), which are known to contribute to tissue damage [24, 40]. Additionally, glutathione regulates the rise of the mitochondrial transmembrane potential, thereby influencing the activation of the mammalian target of rapamycin (mTOR) pathway in T cells of SLE patients [41, 42]. The dysregulation of the mTOR pathway is connected to the development of SLE and will be further explored later in this chapter.

A study using high-performance liquid chromatography performed by Gergely et al. [24] reported decreased levels of glutathione in the peripheral blood of SLE patients, indicating a deficiency in this important antioxidant (Table 1). In an effort to adjust glutathione levels and modulate disease severity, a couple of studies administered N-acetylcysteine (NAC), a cell-permeable precursor of cysteine and a rate-limiting component of de novo reduced glutathione synthesis, to both human patients and mouse models of SLE [42, 43]. The administration of NAC resulted in the inhibition of the mTOR pathway in T cells, resulting in a reduction in disease severity and a decrease in organ damage [38, 39, 44]. In general, SLE patients exhibit insufficient levels of both glutathione and GCL activity. In nonpathogenic states, glutathione and GCL play vital roles in protecting against oxidative stress, detoxifying xenobiotics, and maintaining cellular homeostasis. However, in SLE, these crucial components of the antioxidant defense system are compromised [45].

In addition, a study using gas chromatography/mass spectrometry on urine samples revealed that individuals with SLE exhibited reduced levels of aminomalonate, threonate, and the antioxidant alpha-tocopherol, which are crucial in regulating oxidative stress [14, 46]. Furthermore, metabolic profiling based on liquid chromatography-tandem mass spectrometry detected elevated levels of L-pyroglutamic acid in serum samples from SLE patients, indicating its potential as a disease biomarker [27]. Also known as 5-oxoproline, L-pyroglutamic acid is a derivative of glutamate via the γ-glutamyl cycle, a process in which glutathione decomposes into a γ-glutamyl amino acid and is subsequently converted to pyroglutamic acid through γ-glutamyl cyclotransferase [47]. Given the confirmed decrease in serum glutathione in SLE patients [24], it is suggested that the elevated levels of L-pyroglutamic acid in their serum result from changes in glutathione metabolism. Zhang et al. found a positive correlation between the concentrations of L-pyroglutamic acid and the erythrocyte sedimentation rate (ESR) and anti-Sm antibody in SLE patients, indicating its potential as a marker for disease progression [27]. These findings collectively underscore the imbalanced oxidant/antioxidant system in SLE patients compared to healthy controls, as evidenced by multiple studies demonstrating alterations in glutathione metabolism.

2.3 Purine metabolism

Alterations in the metabolism of purines have also been correlated to the pathogenesis of SLE. A study performed by Zhang et al. that utilized liquid chromatography-tandem mass spectrometry revealed that serum adenosine levels were lower in SLE patients compared to healthy individuals [27]. This outcome aligns with a metabolomic investigation involving feces samples from SLE patients [28]. Adenosine, an endogenous purine nucleoside, holds great significance as Treg cells produce it as a potent immunosuppressive agent, thereby reducing immune responses to oneself, regulating tolerance to tissue grafts, and providing protection against autoimmune conditions [48, 49]. Gao et al.’s study [50] also revealed a notable rise in the blood adenosine deaminase activity, an enzyme involved in adenosine hydrolysis, among individuals with SLE, which may account for the reduced blood adenosine levels observed in these patients. The pathophysiology of SLE is frequently characterized by an imbalance between Th17 and Treg cells. This imbalance involves a reduction in Treg cells and an elevation in Th17 cells, which commonly exhibit functional abnormalities [51]. Consequently, the reduction in adenosine could in addition be linked to the imbalance of Treg cells and could be used to predict the population of Treg cells in individuals with SLE [27].

Finally, in relation to purine metabolism, a study investigating the intestinal dysbiosis in individuals with SLE [52] identified a clear association between purine metabolism and the prevalence of the genus Streptococcus, which was significantly higher in SLE patients. The known role of adenosine in facilitating intestinal epithelial repair and anti-inflammation has been established [53, 54], suggesting that fecal purine metabolism serves as an informative indicator of the gut microbiome in SLE patients. These results imply that adenosine or purine metabolism, which affects gut flora, inflammation, and systemic immunity, has a role in the pathogenesis of SLE [27, 28].

2.4 Amino acid metabolism

In the context of SLE, amino acids are key nutrients for proliferating T cells as they serve both as a source of energy and as precursors for protein and nucleic acid biosynthesis. Glutaminolysis is essential for the production of Th1 and Th17, pro-inflammatory effectors T cells. The enzymes involved in this pathway are of great interest and thus have been examined in the context of pathogenic states [38]. Several investigations have also suggested that amino acid transporters are essential components for T cell clonal proliferation and differentiation in response to antigen presentation. Particularly, it has been demonstrated that the loss of the Large Neutral Amino Acid Transporter 1 (LAT-1), also known as the L-system Slc7a5 transporter, which is a transporter devoted to the transport of essential amino acids, inhibits CD4+ and CD8+ T cell differentiation and proliferation but does not impair CD4+ T cell differentiation into regulatory T cells. T cells without Slc7a5 trancarrier do not increase glutamine and glucose uptake and do not change their metabolism to aerobic glycolysis after stimulation of the T cell receptor. This lack of availability of amino acids leads to inadequate mTORC1 activation, which is required for CD4+ cell differentiation into the T-helper (Th1) and Th17 subsets [55, 56].

Branched-chain amino acids (BCAAs) include valine, leucine, and isoleucine. More than half of the body’s overall metabolism is accounted for by the metabolism of amino acids in muscle tissue, with skeletal muscle being the primary site of BCAA catabolism [26, 57]. Branched amino acids are known to function as signaling molecules that control the synthesis of proteins, lipids, and glucose [38]. Through unique signaling networks, particularly the phosphoinositide 3-kinase signaling – protein kinase B – mTOR pathway (PI3K/AKT/mTOR), they can also contribute to gut health and immunity [58].

The activity of the target of rapamycin (mTOR) is regulated by several elements such as the availability of amino acids, the energy levels, and growth factors. In mammalian cells, mTOR forms two distinct complexes: the mTORC 1 complex (mTORC1) and the mTORC 2 complex (mTORC2). The mTORC1 complex detects several signs of stress, including the accumulation of amino acids such as leucine, isoleucine, and glutamine. According to studies, Th17 cells and T cells that produce IL-4 have increased mTORC activity, which contributes to the proinflammatory profile seen in SLE patients. Additionally, mTOR is necessary for Th 17 cell development because it promotes the synthesis of hypoxia inducible factor 1α (HIF1α), which enhances inflammatory cells’ glycolysis in the pseudo-hypoxia state that is characteristic of subjects with SLE [59].

A study performed on murine models showed that the leucine antagonist, N-acetyl-leucine amide (NALA), can inhibit mTORC1 activity and T cell function by compromising the production of IL-2 and IFγ in Th1 cells [60]. It has been reported that leucine is essential for Treg cell function since it promotes mTORC1 activity in Treg cells through the small G RagA/B and Rheb1/2 proteins, inducing its suppressive activity by boosting the expression of the inducible T cell co-stimulator (ICOS) and CTLA4. According to the study, mice with Rheb1-Rheb2- or RagA-RagB-deficient Treg cells experienced a diminished effector activity of Treg cells and developed an autoimmune illness resembling scurvy [58].

The first steps in the breakdown of BCAAs involve decarboxylation by the branched-chain alpha-keto acid dehydrogenase (BCKDH) complex or transamination by aminotransferases (BCATs). Following these processes, metabolites of BCAAs are changed into succinyl- and acetyl-CoA, which are then involved in the tricarboxylic acid cycle (TCA). BCAT has a negative regulatory effect on glycolysis and mTOR in CD4+ T cells. Activated T cells from mice with cytosolic branched chain aminotransferase (BCATc) deficiency showed an increase in mTORC1 activation compared to control healthy mouse T cells [61]. Furthermore, a different study found that oral delivery of the leucine analog ERG240 specifically suppressed BCAT1 activity, lessening the severity of proliferative glomerulonephritis and collagen-induced arthritis in mice [62].

A GC-MS metabolomics study additionally identified a decrease in the concentrations of BBCAs and their metabolites such as 2-hydroxyisovalerate, 2-keto-3-methylvalerate, and 2-ketoisocaproate, indicating SLE may be involved in increased energy production, gluconeogenesis, and protein synthesis or decreased protein degradation or both [26].

During amino acid catabolism, the carbon skeleton and amino groups are processed in separate but interconnected pathways: the TCA and the urea cycles, respectively. Amino acids entering the TCA cycle contribute to energy generation, but in humans, the oxidative energy derived from amino acid catabolism comprises only a small fraction. Thus far, peripheral blood samples from SLE patients have shown a reduction in most of the amino acids examined, including gluconeogenic and ketogenic amino acids. The amino group’s catabolic result is ammonia, which is then transformed into urea via the urea cycle and eliminated in the urine. A metabolomic study carried out the measurement of the metabolites related to the urea cycle, finding that arginine, the immediate precursor of urea, and urea were increased in subjects with SLE, suggesting increased urea cycle activity in SLE, compared to healthy individuals [29, 63]. On top of this, a GC-MS metabolomics study identified a decrease in the concentrations of BBCAs and their metabolites such as 2-hydroxyisovalerate, 2-keto-3-methylvalerate, and 2-ketoisocaproate. This suggests that SLE may be associated with increased energy production, gluconeogenesis, protein synthesis, decreased protein degradation, or both [26].

Glutamate appears to be a particularly critical amino acid for T cell activation because, upon their activation, T cells consume glutamine at rates that are comparable to or higher than those of glucose [64]. Through the amino acid transporter 2 (ASCT2), CD4+ T cells are able to bind glutamine, which in turn affects the in vivo and in vitro production of proinflammatory Th1 and Th17 cells. It has been noted, meanwhile, that neither Th2 nor regulatory T cell immunological responses are impacted by the genetic ablation of ASCT2. ASCT2−/− T cells exhibit a reduction in oxygen consumption, lactate generation, and glucose uptake, indicating that glutamine is crucial for the way T cells respond to abrupt changes in their metabolic needs [37, 65].

Glutamine is also a fundamental component of protein synthesis and has a role in the production of fatty acids, nucleotides, and redox regulation, among other critical activities that promote T cell proliferation. Citrate, which is produced from glycolytic pyruvate, is exported from the mitochondria and employed in the synthesis of lipids in activated lymphocytes. Through the cycle of tricarboxylic acids, α-ketoglutarate, which is produced from glutamine, contributes to the formation of citrate. After that, citrate is used to produce acetyl groups for fatty acid synthesis. By means of this process, α-ketoglutarate supplies the building blocks needed to synthesize polyamines, which are essential for the synthesis of nucleotides. Ultimately, glutamate—the initial byproduct of glutamine oxidation—acts as a metabolic bridge for the production of glutathione and has a significant impact on lymphocytes’ state of oxidative stress [37, 56].

Glutaminase, the enzyme that is responsible for converting glutamine into glutamate, promotes the proliferation and activation of Th17 cells through various mechanisms. The cAMP responsive element modulator, which is overexpressed in T cells of SLE patients and MRL/lpr mice prone to SLE, regulates the production of glutaminase [66]. Multiple investigations have been carried out to elucidate the impact of this enzyme on the pathophysiology of SLE. Bis-2-(5-phenylacetamide-1,3,4-thiadiazole-2-yl sulfide) ethyl (BPTES), a glutaminase 1 hybridizer, has been shown to reduce Th1 cell development and disease activity in mice with experimental autoimmune encephalomyelitis [67]. This inhibitor, BPTES, also improves disease activity SLE in MRL/lpr mice [55]. Lastly, another enzyme, glutamate oxaloacetate transaminase 1 (GOT1), which converts glutamate to α-ketoglutarate, helps to potentiate Th17 differentiation through epigenetic processes. Selective inhibition of GOT1 by small hairpin (shRNA) RNA silencing or aminooxyacetic acid (AOA) therapy decreased notably the Th17 differentiation of murine T cells [55].

Furthermore, a study based on H single-voxel spectroscopy specifically identified glutamine as a potential biomarker for cerebral disease activity in SLE patients since the levels of glutamine were found to be altered in the brain of SLE patients even before neurologic and imaging manifestations emerged [26, 30].

Another metabolite that has been associated with SLE pathogenesis is tryptophan. This is an essential amino acid whose bioavailability is strictly based on dietary supplement and the subsequent degradation by the intestinal microbiota [43]. Tryptophan is the precursor of several metabolites produced by microbial and endogenous enzymes. The tryptophanase enzymes (Tnase) in the gut microbiota convert dietary tryptophan into bioactive catabolites, including indole and other indole derivatives via several pathways. Moreover, tryptamine, which is produced by some bacteria via the enzyme tryptophan decarboxylase, controls intestinal transit by binding to serotonin receptors [68].

Multiple mechanisms have been proposed to explain how tryptophan contributes to the development and progression of lupus. Low tryptophan concentrations but high kynurenine levels in the serum of lupus-prone mice were indicative of a skewed tryptophan metabolism through the kynurenine pathway, as reported by Brown et al. [31]. In addition, Pearl et al. [32] showed, thanks to a quantitative metabolome analysis based on peripheral blood lymphocytes, that N-acetylcysteine (NAC) therapy may be able to reverse the accumulation of kynurenine and kynurenic acid in patients with SLE. Exogenous kynurenine has been shown in some studies to enhance Th1 polarization of CD4+ T cells and decrease Treg cell polarization of cytotoxic T cells, indicating that kynurenine supports pro-inflammatory T cell phenotypes. According to the same study [32], kynurenine causes the activation of the target of rapamycin (mTOR) in human T cells, which contributes to the high level of mTOR activation that characterizes CD4+ T cells in patients with SLE. Proinflammatory cytokines, such as interferon gamma (IFNγ) and interleukin 17 (IL-17), which are essential to the pathophysiology of SLE, have also been observed to be over-expressed and over-activated in CD4+ T cells with active hypomethylation of genes in the mTOR pathway [69].

Lastly, many tryptophan-derived metabolites, including indole-3-aldehyde, indole-3-acetic acid, 3-methylindole, and tryptamine, are aryl hydrocarbon receptor (AhR) ligands [31, 70]. Through the regulation of P53, FasR, Bcl-2, and cell cycle kinases, AhR activation controls numerous essential biological functions, including cell cycle progression, apoptosis, and cell proliferation. AhR stimulation increases the regulation of genes encoding cytokines such as IL-10, which regulate immune tolerance in SLE [31].

2.5 Gut-microbiome-derived metabolism

Lupus patients often experience gastrointestinal symptoms that are atypical and nonspecific. Until 2014, rheumatologists and researchers dismissed the idea that gut dysbiosis could be associated with SLE due to the lack of clear evidence. However, a breakthrough came in 2014 when Hevia et al. [71] demonstrated, for the first time in a group of 20 SLE patients, a distinct dysbiosis characterized by a higher abundance of Bacteroidetes and a significant decrease in the Firmicutes/Bacteroidetes ratio compared to healthy individuals. Since then, subsequent studies have revealed similarities in SLE-related gut microbiota across different regions, such as Spain and China. In these populations, SLE patients exhibited an enrichment of Prevotellaceae compared to their healthy counterparts [70, 72]. Moreover, He et al. identified an increased prevalence of specific genera, including Rhodococcus, Eggerthella, Klebsiella, Prevotella, Eubacterium, Flavonifractor, and Incertae sedis, along with a decrease in Dialister and Pseudobutyrivibrio [72]. A study by Azzouz et al. [73] found that SLE patients had reduced species richness diversity, particularly pronounced in those with a high SLE Disease Activity Index (SLEDAI). This study highlighted a greater representation of Ruminococcus gnavus (RG) of the Lachnospiraceae family in SLE patients, which is known to contribute to gut barrier impairment. The combination of increased intestinal barrier permeability and gut dysbiosis in SLE patients has been linked to the translocation of pathogenic bacteria, bacterial endotoxins, toxic metabolites, and an elevation in circulating microbial components. Although these changes may not manifest clinically as infections, they contribute to pathological alterations and immune system dysregulation [74].

The gut microbiota constitutes an intricate assembly of bacteria crucial for digestion, playing a key role in facilitating the breakdown of nutrients through enzymes absent in the human genome [75]. Consequently, bacterial metabolites significantly influence the metabolic processes related to nutrient absorption [70]. Numerous studies have substantiated this concept by delineating the abnormal metabolism of amino acids, lipids, and carbohydrates observed in individuals with SLE [76, 77, 78].

A metabolomic investigation using GC-MS revealed a notable elevation of 2-hydroxyisobutyrate in SLE patients experiencing active disease compared to both healthy individuals and SLE patients with inactive disease. This finding suggests a potential link between altered gut microbial metabolism and SLE morbidity, particularly in cases of active SLE, as 2-hydroxyisobutyrate is primarily derived from the degradation of valine by gut microbes [26]. In an ongoing study by our group, we also found a group of bacterial metabolites associated to SLE by urinary metabolomic profiling using LC/GC-MS-MS (unpublished data).

2.6 Lipid metabolism

Examinations of lipid metabolism, conducted in both the serum and feces of SLE patients, indicate a heightened susceptibility to developing lipid profile disorders compared to individuals without the disease [33, 79]. The proposition that dyslipidemia plays a role in the pathogenesis of SLE is substantiated by the elevated occurrence of ischemic heart disease among SLE patients, ranging from 3.8–16%, a figure nearly 10 times greater than that observed in the general population and 50 times higher than the prevalence seen in young women of reproductive age. Furthermore, it has been established that dyslipidemia is prevalent in 68–100% of adult individuals with SLE [80]. Studies exploring lipid metabolism and its association with SLE revealed that more than 60% of altered metabolites in the serum of SLE patients are lipids [33].

Metabolomic and lipidomic investigations have revealed elevated levels of medium-chain fatty acids (MCFA) and free fatty acids (FFA), along with a reduction in long-chain fatty acids (LCFA) in the profiles of individuals with SLE. Short-chain fatty acids (SCFA) can efficiently enter the mitochondrial matrix for beta oxidation directly through the inner mitochondrial membrane, whereas LCFA require assistance from carnitine transport, a process found to be diminished in SLE patients compared to healthy controls [63, 80, 81]. Regarding lipid membranes, a decrease in most phospholipids was observed in individuals with SLE, possibly indicating an increase in cellular turnover. Lipid metabolism in SLE has also been linked to heightened oxidative stress, evidenced by elevated levels of lipid peroxidation products such as malonaldehyde acid (MDA), 9-hydroxyctadecadienoic acid (9-HODE), and 13-HODE, coupled with a decrease in antioxidants such as α-tocopherol, glutathione, and their precursors. Lipid peroxidation, associated with increased free radical activity and damage to cellular membranes, organelles, and/or DNA, has been implicated in cardiovascular and renal complications in individuals with SLE [82, 83, 84].

In an investigation conducted by Yan et al. it was observed that individuals with SLE had notably lower levels of glycerol, oleic acid, and arachidonic acid, while 1-monopalmitin levels were significantly higher compared to control subjects. Glycerol is an essential constituent of triglycerides and phospholipids, whereas 1-monopalmitin falls under the category of monoacylglycerols. In addition, this study revealed that patients with active SLE exhibited lower levels of linoleic acid in comparison to the control group [26].

The impact of lipid metabolism on the functions of immune cells and the pro-inflammatory processes in SLE has been extensively documented. Cells of the innate immune system, like macrophages, engage with oxidized Low-Density Lipoprotein (LDL) particles through scavenger receptors in atherosclerotic plaques. This interaction results in lipid saturation, the production of pro-inflammatory cytokines, and the recruitment of immune cells. In SLE, this process is exacerbated due to elevated circulating levels of LDL and alterations in macrophage function. This dysfunction is induced by the direct activation of hepatic X receptors (LXRs) by lipids, which regulate cellular cholesterol levels and immune functions, influencing factors such as the production of IL-23 and IL-17 and phagocytic pathways [85, 86].

The hyperlipidic environment in individuals with SLE has also been reported to impact T cells. Elevated levels of oxidized LDL have been shown to indirectly enhance T cell activation through monocyte uptake [85, 87]. Furthermore, SLE patients exhibit increased cholesterol and glycosphingolipids in the membranes of T cells, leading to alterations in the composition of signaling platforms known as lipid rafts. These rafts are crucial for T cell receptors to provide stimulatory signals that regulate cell function and inflammation. This alteration is partly attributed to the expression of genes responsible for lipid metabolism in individuals with SLE, but it could also result from changes in the cellular uptake of LDL/VLDL cholesterol and modifications in cholesterol efflux to High-Density Lipoprotein (HDL) [85, 88].

A combined study utilizing 16S sequencing and LC-MS metabolomics characterized alterations in the gut microbiome, as well as fecal and serum metabolomes in SLE patients. The findings indicated a significant correlation between specific bile acids, including deoxycholic acid, glycolic acid, ursodeoxycholic acid, and arachidonic acid, and the SLE Disease Activity Index (SLEDAI) score in patients. Moreover, these bile acids exhibited a strong predictive ability for disease activity [33]. Besides their well-known role in lipid metabolism, bile acids are also signaling molecules that exercise their functions through the activation of the bile acid receptors. A study reported a decrease in FXR receptors in subjects with SLE and in MRL/lpr models of lupus with liver dysfunction [89]. Notably, the administration of chenodeoxycholic acid, an FXR receptor agonist, in mice demonstrated suppression in the expression of inflammatory cytokines such as TNF-α, IFN-γ, and IL-6.

In addition to their impact on the liver, bile acids were identified to modulate intestinal immunity in a separate study. This research revealed that certain metabolites derived from lithocholic acid (LCA), such as 3-oxoLCA and isoalloLCA, could inhibit the differentiation of Th17 cells by directly binding to a crucial transcription factor, retinoid-related orphan receptor-γt (RORγt). These metabolites were found to enhance FOXP3 gene expression by generating mitochondrial reactive oxygen species (mitoROS), leading to the expansion of regulatory T cells [70, 90].

3. Metabolomics in urine and its alteration in the context of LN

One of the studies in LN metabolomics was conducted by Kalantari et al., who used 1H nuclear magnetic resonance (NMR-1H) to compare the urinary metabolic profile of LN patients, patients with SLE without renal damage, and healthy controls (HC). In the aforementioned study, biomarkers were identified to assess their diagnostic effectiveness in search of operative characteristics of the receptor (ROC). Three metabolites were used including 3,4-dihydroxyphenylacetdehyde, beta-alanine, and 2,2-dimethylsuccinic acid of which the combination of 2,2-dimethylsuccinic and DOPAL was associated with greater diagnostic sensitivity when determining acute renal injury, suggested as a diagnostic panel for LN with an AUC of 0.89, sensitivity of 81%, and specificity of 100% [91].

Among the pathways impaired in LN pathogenesis, twenty-four pathways were significant. Nicotinate and nicotinamide metabolism, AXIN degradation, and DNA nucleotide excision repair pathways were the top three impaired pathways in LN patients [91]. Souliotis and colleagues also reported DNA repair issues in active LN patients, resulting in higher levels of DNA damage than those with inactive disease [92].

In another study based on gas chromatography/mass spectrometry in patients with SLE and LN, 81 metabolites were detected, of which mainly highlights amino acids, nucleotides, organic acids, amines, fatty acids, glyceric acid, creatinine, p-cresol, myoinositol, treonate, hypurate, and urate [22]. However, this study did not differentiate between patients with SLE and LN.

Β-alanine is one of the two amino acids (along with histidine) that produce carnosine. The synthesis and metabolism of carnosine develops mainly at the renal level. Its metabolism favors eliminating reactive oxygen species. A significant decrease in urinary excretion of beta-alanine has been observed in LN patients, which could lead to a decrease in the carnosine content of the kidneys in LN patients, making their kidneys susceptible to oxidative stress insults [91].

2,2-Dimethylsuccinate (2,2-DMS) performs a plasma and urinary function by selectively inhibiting sodium-dependent dicarboxylate transporters (NaDC-3 and NaDC-1), which are located at the level of the proximal renal tubule in the luminal and basolateral plasma membranes. Transporters import intermediates from the Krebs cycle for the cells of the proximal tubule to use as energy substrates, among these compounds is succinate. The import of 2,2-DMS is done with higher speed compared to the suction given at greater affinity with the conveyor. Therefore, a decrease in 2,2-DMS levels in the urine of patients with LN may be secondary to increased activity of these transporters in patients with LN [91].

Class IV of the LN favors the proliferation of endocapillary and extracapillary cells, which would explain the energy requirement. There is also evidence that increased levels of nicotinate and nicotinamide (NMN), which participate in the metabolism of vitamin B3, is directly related to energy metabolism. When increasing dicarboxylate transporters, there is a decrease in levels of 2,2-DMS [91].

3.1 Lipids

Within the sphingolipid family are glycosphingolipids that favor cell proliferation and inflammation. At the renal level, they are related to a variety of diseases including glomerulonephritis, renal neoplasia, and renal polycystosis [93]. Tamara and colleagues aimed to identify the role of glycosphingolipids in LN. They utilized the LMR/MpJ-FasLpr/J (LMR/lpr) mouse model of LN, as well as human renal biopsies and urine samples [93]. Elevations of renal and urinary glycosphingolipids occurred in mice prone to lupus; a relationship was determined with the positive transcriptional regulation of neuraminidase that initiates its function in early stages of the disease and progressively increases its activity favoring the elevation of glycosphingolipids. Urinary levels of neuraminidase and lactosilceramide are higher in patients with lupus and LN. Complex glycosphingolipids are formed with the participation of lactosilceramide, example of which is the formation of gangliosides with the addition of sialic acid residues in conjunction with lactosilceramide. Glycosphingolipids are transported within the cell and are mainly located in the plasma membrane. Neuraminidases are enzymes that eliminate sialic acids and proteins and are located in various cell structures at the level of the lysosome, cytosol, plasma membrane, and mitochondria (NEU 1, 2, 3, 4, respectively), of which NEU1 has the ability to secrete itself extracellularly to act on the plasma membrane. Given the above, there is a regeneration of lactosilceramide through NEU enzymes. These data strongly suggest that glycosphingolipids may serve as early markers of LN [93].

With the evolution of the disease, the histological characteristics of the LN are changing; in the follow-up, good or poor response to treatment is evidenced. However, at the time, it is not possible to determine changes in the LN dynamically and in real time, and it may affect the start of therapy and follow-up and thus the prognosis [23].

In the evaluation of the patient, the differentiation of proliferative LN (Class III/IV) and pure membranous LN (Class V) is complex, since they present signs and symptoms in common; both are associated with proteinuria, alterations in blood pressure and kidney function. Pronounced proteinuria is the hallmark of focal segmental glomerulosclerosis (FSGS). One histological characteristic of FSGS is podocyte injury, leading to varying degrees of proteinuria and potentially hypoalbuminemia. These clinical and histological characteristics can also occur with active LN [23].

Romick et al. conducted a study with the aim of identifying urinary metabolites that could discriminate between proliferative LN (Class III/IV), pure membranous LN (Class V), and primary FSGS, using metabolomics based on NMR spectroscopy [23]. It was determined that the urinary metabolites citrate and taurine allow differentiating between proliferative LN and pure membranous LN (Table 2). In Class V, citrate levels are up to eight times lower and taurine levels normal, compared to Class III/IV where citrate levels may be normal and taurine up to 10 times lower. Additionally, the levels of urinary hypurate were evaluated, which allowed us to distinguish with precision between Class V that presents normal levels of taurine, compared with FSGS, with the absence of hypurate [23]. In a study conducted by Ganguly et al., it was demonstrated that LN patients exhibit significantly low levels of urinary creatinine/citrate compared to clinically healthy individuals, levels that increased significantly after 6 months of cyclophosphamide treatment. In the same study, it was shown that LN patients had significantly high levels of the acetate/creatinine ratio in urine, levels that did not change after immunosuppressive therapy [95].

Metabolite	Type of sample	Disease	Method applied	References
3,4-Dihydroxyphenylacetdehyde, beta-alanine; 2,2-dimethylsuccinic acid	Urine	LN	NMR-1H	[91]
Neuraminidase, glucosylceramide (GlcCer) and lactosylceramide (LacCer), Neu1	Urine human renal biopsies; LMR/MpJ-FasLpr/J (LMR/lpr) mouse model of LN	LN	Matrix-assisted laser desorption/ionization imaging MS (MALDI-IMS)	[93]
Citrate and taurine hypurate	Urine	LN	NMR spectroscopy	[23]
LDL/VLDL lipoproteins, sorbitol, glycolic acid acetate, cortisol, creatinine, L-aspartyl-L-phenylalanine	Serum	LN	UPLC-HRMS	[94]

Table 2.

Metabolites in SLE and LN.

The relationship of the metabolites citrate and taurine with kidney function has been determined. Normally, the kidney filters and reabsorbs metabolites, which allows a metabolic balance, so in case of functional deterioration, an alteration in the metabolic profile will be observed [96].

3.2 Serum metabolomics

In LN, a wide range of alterations in metabolic pathways are observed, including glycolysis, amino acid metabolism, and lipid metabolism, resulting in a consequential elevation or reduction that generates a metabolic signature allowing the distinction of patients with SLE. However, variations are observed depending on the region, population, and disease state. In general terms, patients with LN exhibit elevated serum levels of lipid metabolites (including LDL/VLDL lipoproteins), sorbitol, and glycolic acid metabolites and reduced levels of acetate, cortisol, creatinine, and L-aspartyl-L-phenylalanine [94]. Compared to SLE, metabolic variations are observed, such as an increase in LDL/VLDL lipoproteins (triglycerides and fatty acids) and reduced serum levels of acetate [97].

In lupus nephritis, reduced levels of several glucogenic amino acids (such as glycine, alanine, valine, glutamate, proline, and histidine) and ketogenic amino acids (such as leucine) have been determined. This could lead to abnormal catabolism of amino acids and biosynthesis of important proteins in various biological processes, such as cell cycle advancement, genetic transcription, inflammatory responses, and autoimmunity. It has been described that lower serum histidine levels could be closely related to protein waste, energy, inflammation, and oxidative stress. In general, elevated serum glucose levels and reduced levels of most amino acids guide changes in energy production, such as reduced aerobic glycolysis and the use of metabolites as amino acids and ketone bodies as a source of energy [98, 99].

Li J, Xie X et al. conducted a metabolomic study on human serum that revealed a wide range of differential metabolic signatures in patients with LN and idiopathic nephrotic syndrome (INS). Significant metabolic alterations were found in five metabolites, including cortisol, creatinine, sorbitol, L-aspartyl-L-phenylalanine, and glycolic acid [94]. Furthermore, combined forms of biomarkers proved to be more effective in the diagnosis of LN than a single one.

The AUC (area under the curve – ROC) was 0.85, the highest obtained, when combining theophylline, oxidized glutathione, and capric acid, which indicates a very good diagnostic accuracy, with a sensitivity and specificity of 87.50 and 67.86%, respectively; according to the study data in general, the positive predictive value (PPV) and the predictive value negative (NPV) were 75.68 and 82.61%, respectively [94].

3.3 Cortisol

Low serum levels of cortisol play a proinflammatory role, being negatively correlated with the degree of systemic inflammation. Zietz et al. reported that serum cortisol is negatively correlated with the degree of systemic inflammation in SLE. The study by Li J, Xie X et al. indicated that a significant decrease in the amount of cortisol is associated with a high risk of LN and perhaps an impact on the patient’s prognosis [94].

Straub et al., in their study, related that a low level of serum cortisol may be due to a reduction in adrenal steroidogenesis secondary to the inflammatory process and adrenal insufficiency [97].

On the other hand, in the work carried out by Judd et al., it was concluded that proinflammatory cytokines (such as tumor necrosis factor (TNF)) repress relevant enzymatic steps in steroidogenesis that occurs in adrenalocortical cells. The decrease in cortisol levels in patients with LN may be due to inflammation and adrenal insufficiency in patients with LN [94, 97].

3.4 Glycolic acid

Glycolic acid, or glycocholic acid, is a crystalline bile acid that plays a role in the emulsification of fats. It is an acyl glycine produced during the enzymatic metabolism of bile acids in the colonic environment. This compound acts as a fat solubilizer, facilitating its absorption. Additionally, it contributes to the homeostasis of bile acids, which act as signaling molecules with endocrine functions, thereby promoting the homeostasis of triglycerides, cholesterol, and glucose. In a study, the level of glycolic acid was reduced in patients with LN. The decrease in the level of glycolic acid in plasma could be due to the reduced concentration of glycolic acid in the plasma of LN patients [94].

4. Conclusion

The pathogenesis of SLE and LN is of complex etiology. There are no sufficiently specific biomarkers that allow for the noninvasive or minimally invasive prediction of renal involvement in the context of SLE. With the advent of “omics” approaches, a series of metabolites, amino acids, lipids, and proteins, among others have been discovered that could potentially serve as biomarkers for SLE and LN in the future. Metabolomics has unique characteristics that make it a methodology with enormous potential to discover metabolites associated with imbalances in crucial metabolic pathways for the organism. Large-scale validation of these metabolites discovered by these non-targeted approaches in patient cohorts from different regions will be necessary to uncover their role and impact in personalized medicine.

Acknowledgments

We are truly grateful to the rheumatology department and the renal unit of the Clinica de la Costa in Barranquilla City (Colombia) for the recruitment and sample collection of the patients enrolled in the study that was support for this chapter book.

Conflict of interests

The authors declare no conflict of interests.

Funding

This chapter book was supported by the Ministerio de Ciencia, Tecnología e Innovación (MINCIENCIAS) (SIGP code: 125384467468, Contract: RC No. 852 of 2019).

Abbreviations

2,2-DMS	2,2-dimethylsuccinate
9-HODE	9-hydroxyoctadecadienoic acid
13-HODE	13-hydroxyoctadecadienoic acid
AhR	aryl hydrocarbon receptor
Anti-DNA	antibodies against DNA
Anti-dsDNA	antibodies against double-stranded DNA
Anti-Sm Antibody	anti-smith antibody
AOA	aminooxyacetic
ASCT2	alanine serine cysteine transporter 2
AUC	area under the curve - ROC
BCAAs	branched-chain amino acids
BCAT1	branched-chain amino acid transaminase 1
BCATs	branched-chain amino acid aminotransferases
BCKDH	branched-chain alpha-keto acid dehydrogenase
Bcl-2	B-cell lymphoma 2
BPTES	bis-2-(5-phenylacetamido-1,3,4-thiadiazol-2-yl)ethyl sulfide
cAMP	cyclic adenosine monophosphate
CTLA4	cytotoxic T-lymphocyte-associated antigen 4
ERG240	leucine analogue
ESR	erythrocyte sedimentation rate
FasR	Fas receptor
FFA	free fatty acids
FOXP3	transcription factor forkhead box protein 3
FSGS	focal segmental glomerulosclerosis
FXR receptors	farsenoid X receptors
GC-MS-MS	gas chromatography with tandem mass spectrometry
GC-MS	gas chromatography - mass spectrometry
GCL	glutamate cysteine ligase
GFR	glomerular filtration rate
GOT1	glutamate oxaloacetate transaminase 1
GSH	glutathione
HC	healthy controls
HDL	high-density lipoprotein
HIF1α	hypoxia inducible factor 1 subunit alpha
HPLC	high-performance liquid chromatography
ICOS	inducible t cell co-stimulator
IFγ	interferon gamma
IL-2	interleukin 2
IL-4	interleukin 4
IL-6	interleukin 6
IL-10	interleukin 10
IL-17	interleukin 17
IL-23	interleukin 23
INS	idiopathic nephrotic syndrome
ISN/RPS	International Society of Nephrology/Renal Pathology Society
LacCer	lactosilceramide
LAT-1	large neutral amino acid transporter 1
LC-MS-MS	liquid chromatography with tandem mass spectrometry
LCA	lithocholic acid
LCFA	long-chain fatty acids
LDL	low-density lipoprotein
LMR/lpr	LMR/MpJ-FasLpr/J
LN	lupus nephritis
LXRs	liver X receptors
MCFA	medium-chain fatty acids
MDA	malonaldehyde acid
mitoROS	mitochondrial reactive oxygen species
MRL/lpr mice	mice are homozygous for the lymphoproliferation spontaneous mutation
mTOR	mammalian target of rapamycin
mTORC 1	mammalian target of rapamycin complex 1
mTORC 2	mammalian target of rapamycin complex 2
NAC	N-acetylcysteine
NaDC	sodium-dependent dicarboxylate transporters
NALA	N-acetyl-leucine Amide
NETs	neutrophil extracellular traps
NEU1	neuraminidase 1 lysosome
NEU2	neuraminidase 2 cytosol
NEU3	neuraminidase 3 plasma membrane
NEU4	neuraminidase 4 mitochondria
NMN	nicotinamide
NMR-1H	1H nuclear magnetic resonance
NMR	nuclear magnetic resonance spectroscopy
NPV	negative predictive value
p53	tumor protein p53
PI3K/AKT/mTOR pathway	phosphoinositide 3-kinase signaling - protein kinase B - mTOR pathway
PPV	positive predictive value
Rag A/B	Ras-related GTP-binding protein A/B
RG	ruminococcus gnavus
Rheb1/2	Ras homolog enriched in brain protein 1/2
ROC	operative characteristics of the receptor
RORγt	retineic-acid-receptor-related orphan nuclear receptor gamma
ROS	reactive oxygen species
SCFA	short-chain fatty acids
shRNA	short hairpin RNA
Slc7a5 transporter	large neutral amino acid transporter 1 (LAT-1)
SLE	systemic lupus erythematous
SLEDAI	systemic lupus erythematous disease activity index
TCA	tricarboxylic acid
Th1 cells	type 1 T helper cells
Th2 cells	type 2 T helper cells
Th17 cells	type 17 T helper cells
Tnase	tryptophanase enzymes
TNF-α	tumor necrosis factor alpha
TNF	tumor necrosis factor
Treg cells	regulatory T cells
UHPLC-MS/MS	ultra-high performance liquid chromatography tandem mass spectrometry
VLDL	very low-density lipoprotein