Carnitine is an amino acid belonging to a quaternary ammonium cationic complex. It has two stereoisomers: bioactive L-carnitine and abiotic enantiomeric isomer D-carnitine. L-carnitine is the predominant carnitine used in biological and medical fields and is commonly referred to as “carnitine.” The human body contains about 300 mg/kg of L-carnitine, 98% of which is intracellular, with 80% present in the muscles, 5–10% in the gastrointestinal tract, and 3% in the liver (11). Although D-carnitine has no biological activity in humans, it adversely impacts biochemical processes by inhibiting the carnitine acetyltransferase (12). D-carnitine supplementation has been found to induce liver inflammation, oxidative stress, and apoptosis in animal studies (13). In addition, D-carnitine can cause secondary carnitine deficiency (SCD), so researchers often use D-carnitine supplements to feed mice to establish carnitine lack animal models (14, 15). For these reasons, D-carnitine has been rarely studied in humans.

In the average adult diet, it is estimated that about 75% of the daily carnitine requirement mainly comes from meat, fish, and dairy products (16), while the remaining 25% is derived from endogenous synthesis (1). In strict vegetarians, more than 90% of the daily carnitine requirement is obtained by endogenous synthesis (17–19), whereby humans can synthesize ~1–2 μmol carnitine/kg/day (20). The synthesized carnitine is formed when 6-N-trimethyl-lysine (TML) is released during protein degradation (Figure 1) (21). After release, TML is hydroxylated into 3-hydroxyl-6-N-trimethyl-lysine (HTML) by trimethyl dioxygenase (TMLD), which is then broken down into 4-N-trimethyl-butylaldehyde (TMABA) and glycine by HTML aldolase (HTMLA) (1). TMABA produces 4-N-trimethylaminobutyrate (γ-butyrobetaine) under the action of dehydrogenase (11). Finally, γ-butyrobetaine dioxygenase (BBD) is used to hydroxylate γ-butyrobetaine to produce endogenous carnitine (Figure 1) (22, 23). BBD is confined to the human liver, kidneys, testis and brain, thus the biosynthesis of carnitine occurs only at these locations (24). Other tissues, such as skeletal muscle, obtain carnitine from the blood (25).

The distribution and homeostasis of carnitine within the body is controlled by organic cationic transporters (OCTN) (26). OCTN act on intestinal absorption and renal reabsorption of carnitine, and plays an important role in tissue distribution by catalyzing carnitine to enter cells in vivo (27). Among these, OCTN2 is the most important physiologically transporter of carnitine due to its high affinity and wide expression (27). OCTN2 plays a crucial role in carnitine homeostasis. Notably, BB, a direct precursor of carnitine, is also a good substrate for OCTN2. The liver and kidneys have a strong ability to convert BB to carnitine. Loss or mutation of OCTN2 function results in primary systemic carnitine deficiency (PCD) with severe clinical consequences such as cardiac and skeletal myopathy, cardiac hypertrophy and NAFLD (28).

Carnitine, a natural compound closely related to the above-mentioned functions, has recently been found to alter the underlying disease pathology with fewer side effects (5, 36). It has been reported that L-carnitine as a supplement can be useful in the treatment of hypertension (37), diabetes mellitus (6, 7), NAFLD (34), heart failure (38), coronary artery disease (39), liver cirrhosis (40), muscle injury (41), dyslipidemia (42), migraine (43), Alzheimer's disease (44), and other chronic diseases.

The prevalence of NAFLD is increasing at a tremendous rate, currently affecting about 24% of the world's population. It has been pinpointed as the most common cause of liver disease globally (45). In the last decade, the clinical burden of NAFLD has been associated to liver-related morbidity and mortality and also to the extra-hepatic manifestations involving other organs and regulatory pathways. The pathogenesis of NAFLD is multifactorial and only partially understood (46). At present, the highly recognized pathogenic mechanism is a “multiple hit” hypothesis, which considers the combined effects of multiple insults on genetically susceptible subjects to induce NAFLD and provides a reasonable explanation for the development of NAFLD (47). The first hit is the accumulation of fat in the liver, followed by the development of necrotic inflammation and fibrosis. In addition, nutritional factors, intestinal microflora, and genetic and epigenetic factors exhibit significant influence (48). Inhibition of β-oxidation, mitochondrial dysfunction, and insulin resistance (IR) are the three most important links in these “hits” as well as targets for carnitine to ameliorate NAFLD. Herein, we attempt to explain the role of carnitine in the pathogenesis of NAFLD, as shown in Figure 3.

Supplementation with carnitine may have a positive effect on the β-oxidation of fatty acids in mitochondria, thereby promoting lipid metabolism by boosting the uptake of fatty acids and eventually reducing fat accumulation in hepatocytes. It is known that NAFLD is associated with an imbalance in a variety of metabolic networks, among which abnormal lipid metabolism is the core pathological metabolic process (49). The immediate cause of NAFLD is high levels of triglycerides (TG) and serum free fatty acids (FFAs). Excess TG comes from hepatic fat production and dietary fat supply, while FFAs accumulate due to lipolysis of visceral adipose tissues (50). The most important strategy to reduce NAFLD is to decrease dietary fat consumption and promote the catabolism of FFAs. β-oxidation is the core process of FFA decomposition, and carnitine acts as an FFA transporter. Carnitine plays an extremely important role in β-oxidation through the “carnitine shuttle,” as described above (51). The carnitine-mediated entry process is a rate-limiting step for mitochondrial LCFA oxidation and, thus, a major regulating point. Increased levels of carnitine may improve the originally inhibited β-oxidation, which is a great way to reduce fat accumulation in the liver (34). Recent studies have emphasized the key role of mitochondrial dysfunction in the occurrence and development of NAFLD (52). The effect of carnitine on mitochondrial dysfunction has been confirmed by various experiments (5). Increased FFAs can lead to hepatic lipotoxicity, generation of reactive oxygen, and damage to the mitochondrial membrane (52). Carnitine treatment could increase the mRNA expression of carnitine palmitoyltransferase 1A and peroxisome proliferator-activated receptor-γ (PPAR-γ), while preventing lipid membrane peroxidation and ROS. This further leads to the reversal of mitochondrial dysfunction, thereby increasing mitochondrial β-oxidation and reducing intracellular oxidative stress to prevent hepatic lipotoxicity (53, 54).

An important mechanism by which carnitine improves IR is enhancing the oxidation of mitochondrial long-chain acyl-CoAs (Figure 2) (55). Accumulation of long-chain acyl-CoAs and other FA metabolites impairs insulin signaling and leads to the development of IR, a crucial pathophysiological factor for carbohydrate metabolism disorders consistent with the development of NAFLD (49). Since abnormal mitochondrial function is associated with IR (56), carnitine therapy can also reduce IR by improving mitochondrial function (56).

Other possible mechanisms to improve insulin sensitivity in NAFLD reported in the literature include the regulation of the acetyl-CoA/CoA ratio in mitochondria, modulation of the pyruvate dehydrogenase complex (PDHC) activity, altering the expression of glycolytic and gluconeogenic enzymes as well as the expression of genes associated with the insulin signaling cascade, and stimulation of the insulin-like growth factor-1 (IGF-1) axis and IGF-1 signaling cascade (34, 55). As mentioned above, carnitine improves the glucose metabolism by reducing acetyl-CoA/CoA and reduces IR. Both insulin and IGF-1 act through their homologous receptor tyrosine kinases to coordinate metabolism and cellular responses to nutrient supply (57). When IGF-1 axis is activated, multiple signaling pathways corresponding to its kinase domain are also activated, including PI3K/Akt pathway and Raf/MEK/ERK level linkage pathway, which can ultimately reduce IR and prevent cell apoptosis (58, 59), thereby improving NAFLD.

The liver is the main organ responsible for detoxification and metabolism of various compounds that produce reactive oxygen species (ROS). As such, liver disease may lead to increased ROS production and, subsequently, increased peroxidation of lipid membranes and inflammatory factors production, leading to hepatocyte injury and cell death (60–63). Injury or death of hepatocytes results in increased release of liver enzymes (64). Patients with chronic liver disease, especially liver cirrhosis, usually suffer from secondary carnitine deficiency (65, 66). Because the liver is one of the main sites for carnitine synthesis, liver disease can impair carnitine synthesis, which further aggravates the above-mentioned pathological processes and promotes disease progression. A recent meta-analysis showed that L-carnitine supplementation improves liver function by reducing histological steatosis and NAS scores in patients with NASH (67). In addition, Okabayashi et al. demonstrated that L-carnitine could improve postoperative liver function in hepatectomized patients (68). L-carnitine reduces inflammatory response via the transfer of β-oxidized long-chain fatty acids in mitochondria and the excretion of toxic substances in fatty acid metabolism (69). It also plays the role of free radical scavenger by reducing the production of ROS (70). Hence, L-carnitine may contribute by reducing oxidative stress, decreasing inflammatory response, and activating enzymes involved in the defense against oxidative damage, thereby reducing elevated serum liver enzymes in patients with liver disease.

Primary carnitine deficiency (PCD) is an autosomal recessive FA oxidation disease (34). Both serum and intracellular carnitine levels are very low in PCD patients. Therefore, FAs cannot be accumulated as an energy source (71), leading to increased ROS production (72). Different from PCD, secondary carnitine deficiency (SCD) is mostly caused by drugs (such as valproic acid, anticancer drugs, omeprazole, amphoteric drugs, etc.) or diseases (such as fatty acid oxidation disorder, organic acidemia, etc.) (1). These drugs or diseases increase the excretion of carnitine in the form of acylcarnitine in urine, reduce renal tubule reabsorption and endogenous biosynthesis, and inhibit carnitine transporters, resulting in increased carnitine consumption (73, 74). Carnitine deficiency impairs the mitochondrial β-oxidation of FAs, leading to acute metabolic decompensation, elevated aminotransferase, and hepatic encephalopathy similar to NAFLD (75). It has been reported that long-term oral carnitine treatment for 6 months in a patient with carnitine deficiency resulted in increased muscle strength, significant reduction in heart size, relief from cardiomyopathy, and partial repletion of carnitine levels in plasma and muscle with complete repletion of liver functions (71). L-carnitine has been approved by the US Food and Drug Administration for the treatment of PCD and SCD (76).

Alcoholic fatty liver disease is another major cause of fatty liver disease, for which carnitine supplementation may also be beneficial for improvement (77). The endogenous biosynthesis of FAs is the main cause of hepatic steatosis in chronic alcoholism (78). Hepatic steatosis is further aggravated by impaired fatty acid oxidation in an ethanol-impaired liver (79). Sachan et al. evaluated the lipid-lowering effects of carnitine and its precursors (lysine and methionine) in rats with chronic alcoholism and concluded that carnitine effectively prevented alcohol-induced hyperlipidemia and liver fat accumulation. Carnitine was also found to be more effective than its precursors (77).

The efficacy of L-carnitine supplements in the treatment of chronic hepatitis B (CHB) remains controversial. A Korean study compared the combination of entecavir and carnitine with entecavir alone to determine their therapeutic effects in CHB. Results indicated that the ALT normalization rate in the combination group was higher than that in the entecavir group, but the HBV-DNA normalization rate and change in serum HBV-DNA levels were similar (80). However, a recent trial showed that besifovir dipivoxil maleate combined with L-carnitine did not lead to any improvement in hepatic steatosis in CHB patients (81). On the other hand, carnitine has a higher recognition for the therapeutic effect in chronic hepatitis C (CHC). A recent meta-analysis showed that long-term use of low-dose (≤2,000 mg/day) L-carnitine could reduce TC and TG in overweight patients with liver disease, especially those with chronic hepatitis C (82). Another study also suggested that L-carnitine may be an effective adjuvant for anti-HCV therapy, because it reduces the oxidative stress induced by JFH-1 infection, inhibits HCV assembly, and exhibits anti-HCV activity through its anti-adipogenic activity in HCV-infected cells (83).