have been identified, including cation/proton exchange, cation uniport, and both sodium‐dependent and sodium‐independent zwitterion transport. A proposed nucleotide binding site is present on the intracellular side of OCTNs, which may stimulate transport of some substrates (i.e., TEA). OCTN1 substrates include the prototypical cation TEA, zwitterions (e.g., ergothioneine, acetylcholine, carnitine), and a number of drugs, including quinidine, verapamil, ipratropium, amisulpride, sulpiride, gabapentin, cytarabine, and ethambutol (Table 2.2). OCTN1 has the highest affinity for ergothioneine with a K m of 21 μM and transports the compound in a sodium‐dependent fashion [2]. In vitro, OCTN1 is inhibited by over 30 compounds, both endogenous and exogenous. Some of the most potent inhibitors appear to be lidocaine (K i = 0.83 μM), verapamil (K i = 8.38 μM), ergothioneine (K i = 9 μM), cobicistat (IC50 = 2.5 μM), carnitine (K i = 24 μM), and imatinib (IC50 = 31 μM). In humans, no clinical drug–drug interactions have been documented for OCTN1 to date. However, it has been suggested that OCTN1‐mediated secretion of some cationic and zwitterionic substrates may contribute to interindividual variation in drug response [81].
Genetic variants in OCTN1 can alter affinity for substrates and inhibitors, as illustrated by the common polymorphism, p.L503F, which has a higher affinity for xenobiotics but a decreased affinity for carnitine compared to the reference transporter [2]. The p.L503F variant also appears to influence drug excretion in vivo, as evidenced by altered renal clearance of gabapentin, detailed below.
2.8.2 Regulation
OCTN1 expression in rat liver is responsive to the PPARα agonist clofibrate, although to a lesser degree than OCTN2 [82]. OCTN1 contains an intronic binding sequence for the transcription factor RUNX1, disruptions of which have implications in rheumatoid arthritis (RA) [4]. OCTN1 is regulated by PDZK1 [83].
2.8.3 Animal Models
An Octn1 −/− knockout mouse was developed with a complete deletion of exon 1 [84]. Octn1 −/− mice are fertile, viable, and have no gross physiological abnormalities. These mice exhibit systemic depletion of ergothioneine compared to wild‐type, which has high tissue accumulation of ergothioneine in the small intestine and kidney. Octn1 −/− mice subjected to intestinal oxidative stress have significantly lower tolerance for the oxidative stress and worse survival outcomes compared with wild‐type mice, suggesting that ergothioneine’s antioxidant properties contribute to protection or recovery from oxidative stress. Additionally, knockout of Octn1 results in increased kidney damage measured by oxidative stress and interstitial fibrosis in a mouse model of chronic kidney disease (CKD) [85].
2.8.4 Human Genetic Studies
In genome‐wide association studies (GWAS), genetic variants in OCTN1 have been associated with blood cell counts (monocytes, neutrophils, white blood cells, and granulocytes) [86], blood metabolite measurements, particularly acylcarnitines [87], and inflammatory bowel disease [31], among others. However, as with many GWAS, these associations are primarily drawn from samples of European ancestry. Thus, further studies in diverse populations are needed to continue to clarify the role of OCTN1 in health and disease.
OCTN1 is associated with a number of pathologies relating to inflammation and autoimmunity. OCTN1 exists within the IBD5 locus, a roughly 250 kb segment of chromosome 5.31q that has been associated with inflammatory bowel disease (IBD), particularly Crohn’s disease (CD) and ulcerative colitis [2]. While these are complex polygenic diseases, a haplotype containing the p.L503F variant of OCTN1 and a promoter region variant (−207G>C) of OCTN2 is associated with susceptibility to CD [2].
In addition to IBD, an intronic variant of SLC22A4 (OCTN1) has been associated with RA in a Japanese population [2]. This variant disrupts a consensus sequence for the hematopoietic transcription factor RUNX1, which is also associated with RA. As a result, transcription of SLC22A4 is suppressed when this intronic variant is present, increasing susceptibility to RA, although full understanding of these associations and mechanisms remains to be elucidated.
OCTN1 contains a number of polymorphisms in various populations. The most common are p.I306T, with a global allele frequency (GAF) of 0.583 according to the GnomAD database [88], and p.L503F, which exists at a GAF of 0.291 with varying frequencies in all ancestral populations [88] (Table 2.3). In vitro, neither polymorphism results in significant alteration in the transport of OCTN1 substrates, betaine or TEA [81]. Conversely, p.L503F has been shown to reduce OCTN1 transport of the anticonvulsant drug gabapentin in vitro and influence the clearance of the gabapentin in vivo. Individuals with the 503L/503L genotype exhibit net secretion of gabapentin by OCTN1 at the apical membrane of the kidney, facilitating elimination of the drug, while individuals homozygous for 503F exhibit no net secretion of gabapentin, with renal clearance roughly equivalent to the rate of glomerular filtration [89].
2.9 OCTN2
2.9.1 Substrate and Inhibitor Selectivity
In vitro, hOCTN2 is multi‐specific and has been shown to transport a number of endogenous compounds and xenobiotics [2]. Primarily, OCTN2 is a sodium‐dependent high‐affinity L‐carnitine transporter with a K m of 4 μM. To a lesser extent, OCTN2 transports some short‐chain acylcarnitines, including acetyl‐L‐carnitine and the drug metabolites pivaloylcarnitine and valproylcarnitine. OCTN2 transports the prototypical cation, TEA, in a sodium‐independent manner. Other substrates of OCTN2 include drugs ipratropium, mildronate, amisulpride, sulpiride, etoposide, ethambutol, cephaloridine, quinidine, and verapamil (Table 2.2). In vitro, many approved drugs act as inhibitors of OCTN2. Transport of L‐carnitine is inhibited by β‐lactam antibiotics including cefepime, cefoselis, cephaloridine, cefuroxime, cephalexin, and cefazolin with varying IC50 values, likely attributed to the presence of a quaternary amine functional group similar to carnitine. Other strong inhibitors span many drug classes, including cardiac drugs verapamil, quinidine, and amiodarone, proton‐pump inhibitors such as omeprazole, and anticancer agents including tamoxifen, gefitinib, and cedirinib, among others.
In vivo, OCTN2 has not been reported as a target for drug–drug interactions to date. However, multiple drugs have been observed to cause carnitine deficiency through various mechanisms. Administration of the anticonvulsant valproic acid and the antibiotic pivalic acid causes reduced plasma carnitine levels and, in some cases, clinically relevant carnitine deficiency. Multiple mechanisms have been proposed. One possibility is that valproate and pivalate directly inhibit the binding pocket for carnitine in OCTN2. Other studies have suggested that rather than inhibit OCTN2 directly, these drugs form carnitine conjugates and are likely effluxed out of the kidney with poor reabsorption, resulting in carnitine wasting and depletion. Alternatively, the valproyl‐ and pivaloyl‐carnitine esters could block reabsorption of free carnitine at OCTN2. Regardless of mechanism, these cases of drug‐induced carnitine deficiency have been fatal in patients with carnitine transporter deficiency who already have reduced systemic carnitine levels.
In recent years, OCTN2 has become a target of drug delivery optimization strategies. Multiple properties make it an attractive drug target. First, it is theorized to increase oral bioavailability of targeted drugs due to high expression in the small intestine. Second, it has the potential to increase blood–brain barrier permeability of substrates due to expression at the BBB. Third, it allows for the targeting of drugs to the kidney. And fourth, it has been hypothesized to increase delivery of asthma therapeutics to the lung [90]. Multiple carnitine‐conjugated prodrugs have been developed, including butyrate used in treatment for gut inflammation, nepotic acid used to treat seizures, and the chemotherapeutic drug, gemcitabine. Carnitine‐conjugated