Using trehalose to prevent and treat metabolic function: effectiveness and mechanisms

Abstract

Purpose of Review:

Trehalose is a disaccharide with manifold industrial, commercial and biomedical uses. In the decade following its initial definition as an autophagy-inducing agent, significant advances have been realized in regard to the applicable clinical and pre-clinical contexts in which trehalose can be deployed. Moreover, the mechanisms by which trehalose exerts its metabolic effects are only beginning to gain clarity. In this review, we will highlight the most recent advances regarding the effectiveness and mechanisms of trehalose actions in metabolic disease, and discuss barriers and opportunities for this class of compounds to advance as a clinical therapeutic.

Recent Findings:

Trehalose reduced cardiometabolic disease burden in diet-induced and genetic models of atherosclerosis, dyslipidemia, hepatic steatosis and insulin- and glucose tolerance. The mechanism by which these effects occurred were pleiotropic, and involved activation of fasting-like processes, including autophagic flux and transcription factor EB. These mechanisms depend heavily on route of administration and disease-specific context. Host and microbial trehalase activity is likely to influence trehalose efficacy in a tissue-dependent manner.

Summary:

Trehalose and its analogues are promising cardiometabolic therapeutic agents with pleiotropic effects across tissue types. It is likely that we are only beginning to uncover the broad efficacy and complex mechanisms by which these compounds modulate host metabolism.

INTRODUCTION

Trehalose α,α-Trehalose, α-D-Glucopyranosyl-α-D-glucopyranoside is a naturally occurring, non-reducing disaccharide that is comprised of two glucose moieties linked via an α, α−1, 1-glucosidic bond [1]. Trehalose is ubiquitous in our natural and manufactured environments, and is found in several species of plants, algae, fungi, yeasts, bacteria, insects and other lower invertebrates. Moreover, it is also widely used in industrial and commercial applications due to its unique stabilizing, texturizing and sweetening properties [2]. Indeed, trehalose is used as a packing material, desiccant, and drug excipient, as well as a food preservative and cosmetic. In addition, significant attention over the past decade has been given to the possibility that trehalose has therapeutic effects in neurodegenerative [3] and cardiometabolic diseases. The purpose of this concise review is to discuss the most recent data specifically regarding the efficacy of trehalose in treating cardiometabolic disease in animal models and humans. To broaden the literature base from which we can draw, we also explore potential mechanisms by which trehalose acts in studies that examine trehalose action both inside and outside of its metabolic effects. We conclude by discussing limitations and opportunities to advance trehalose and its analogues as clinical treatments against cardiometabolic disease.

Mechanisms of trehalose action

Trehalose was originally identified as an agent that induces cellular macroautophagy (hereafter, “autophagy”), which is the breakdown and recycling of old or damaged macromolecules that occurs in response to cellular stresses. This mechanistic action was initially suspected becuase of trehalose’s protective effects against β-amyloid aggregation in neuronal cell lines (discussed in [3, 4]). Because autophagy regulates cellular and wholeorganism metabolic homeostasis [5–7], it is largely considered that trehalose mitigates metabolic disease by activating autophagy in specific tissue compartments [2]. The extent and mechanisms and tissue compartments in which this occurs in vivo, however, remain an area for intense interrogation [2, 8, 9]. In addition, new data indicate that other homeostatic pathways may be involved in trehalose action, which depends upon cell and tissue type. We briefly review this below as well.

Highlighting its complex biology, trehalose appears to inhibit autophagic flux in certain cell types [8, 9]. In HeLa cervical cancer cell lines, acute phase trehalose treatment mildly induced autophagic flux (e.g. at 6 and 12 h), as determined by decreased GFP:RFP ratio [10]. However, 24 h GFP:RFP ratio was increased relative to vehicle-treated HeLa cells. This indicated lower autophagic flux after chronic trehalose exposure. Similarly, in human neuroblastoma and primary rat cortical neurons, trehalose induced accumulation of LC3BII, p62 and autophagosomes with a decrease in autolysosomes. The lack of appreciable effects of the lysosomal acidification inhibitor, bafilomycin A1, on p62 and LC3BII accumulation was interpreted to mean that trehalose decreased autophagic flux in vitro. Taken together, trehalose treatment in some in vitro models may inhibit autophagic flux. However, the paucity of reports demonstrating that trehalose impairs autophagic flux in vivo highlights important differences between in vitro and in vivo modeling of autophagy regulation. This encompasses differences in effective dosage at the site of action, host and microbial trehalose metabolism, and expression and localization of trehalose target effectors in each cell type. It also underscores the importance of utilizing autophagy-deficient models when drawing conclusions on the role of autophagic flux in trehalose’s or any other agent’s mechanisms of action [11, 12].

On the other hand, multiple lines of evidence suggest that trehalose induces autophagic flux to exert its protective effects in several tissue types [2, 3, 13, 14], and that this is important for improvement in metabolic phenotypes. For example, trehalose blocked conjugated fatty acid-induced steatosis in wild-type cultured murine hepatocytes. However, this protective effect was reversed in hepatocytes derived from ATG16L1 hypomorphic mice [15], which have attenuated (but not absent) autophagic flux. In addition, cultured hepatocytes overexpressing kinase-dead AMPKα1 subunits were similarly not protected from fructose-induced triglyceride accumulation [15]. Similarly, high glucose incubation suppressed autophagy, and trehalose treatment reversed this autophagy suppression by high glucose incubation in HeLa and C17.2 neural stem cells, and in the neuroepithelium of high glucose-exposed murine embryos [16]. Moreover, trehalose attenuated Western diet-induced atherosclerosis in APOE-null mice was attenuated in wild-type mice treated with trehalose, but this therapeutic effect was abrogated in mice lacking macrophage ATG5-, and in mice lacking SQSTM/p62 [17, 18]. Consistent with these data, trehalose did not confer protection from LAD ligation-induced cardiomyopathy in BECN1 heterozygous mice [19].

Some evidence demonstrates, by contrast, that autophagic flux is not entirely required to convey the therapeutic effects of trehalose treatment. For example a full BECN1 complement is required for trehalose cardioprotection [19]. Conversely, ATG7 haploinsufficiency did not alter the ability of trehalose to enhance glucose tolerance and insulin sensitization in ob/ob mice [12]. Moreover, trehalose increased heat generation independent of ATG16L1, EPG5 and BECN1 [11]. Together, the data suggest that trehalose invokes autophagy-related proteins for at least some of its therapeutic effects (e.g. hepatocyte lipid accumulation, atherosclerosis). Yet, some important therapeutic outcomes, including energy expenditure [11], adipose browning [11], and glucose and insulin tolerance [20, 21] proceed independent of autophagic flux.

TFEB plays a central role in the metabolic effects of trehalose, irrespective of whether autophagy is activated [11, 22–24]. TFEB is a member of the microphthalmia family of basic helix-loop-helix- leucine-zipper transcription factors involved in lysosomal function and autophagy [17, 18]. During nutrient-rich conditions, TFEB is largely inactive. During starvation, ER or lysosomal stress, TFEB translocates to the nucleus, where it activates transcription of its target genes. Sergin and colleagues showed that trehalose activates macrophage TFEB translocation to the nucleus, which was associated with improved atherosclerotic plaque area [17, 18]. And although the direct requirement of TFEB for trehalose-mediated atherosclerotic protection remains to be determined, trehalose treatment mimicked the effects of macrophage TFEB overexpression. TFEB translocation in HeLa/TFEB cells is increased by fasting-mimetic stimuli. This includes trehalose treatment, by inhibition of Akt or PI3K, or by serine-to-alanine mutation of the Akt phosphorylation 467 residue on TFEB, (S467A) [25]. Rusmini and colleagues subsequently demonstrate PPP3-mediated TFEB activation via lysosomal permeabilization and calcium release in the presence of trehalose and other disaccharides lactulose and melibiose [22]. Hepatocyteselective TFEB knockdown by AAV8-mediated TFEB shRNA delivery blocked oral trehalose-induced thermogenesis. Further, trehalose-induced thermogenesis occurred independent of autophagy-related proteins, ATG16L1, EPG5 and BECN1 [11]. Oral trehalose administered ad libitum induced TFEB-mediated lysosomal biogenesis, CLEAR network genes, and reduced lysosomal storage burden in patient-derived JNCL fibroblasts [25]. The data together suggest that trehalose exerts its metabolic effects at least in part through TFEB-induced lysosomal regulation in several contexts. TFEB activation may be the result of lysosomal damage, fasting-like cellular responses, or a combination of the two, depending upon the level of direct trehalose exposure encountered by the target tissue [11, 17, 22, 24].

Several actions have recently been proposed regarding trehalose function. However, the relation of these actions to energy metabolism and cardiometabolic disease remain to be directly explored. This includes inhibition of cell stress pathways, such as matrix metalloproteinases, the p38 MAPK pathway, NRF2-KEAP1, and unfolded protein responses [26, 27]. The contributions of these pathways to the metabolic efficacy of trehalose therapy remains to be further explored.

Trehalose Efficacy in Animal Models of Metabolic Disease

Because trehalose induces autophagy, and because activating autophagy exerts well-described therapeutic metabolic effects, trehalose was tested as a potential metabolic therapeutic agent (Figure 1). First, several animal models demonstrated that trehalose mediates protection from atherosclerosis and dyslipidemia. Sergin and colleagues treated APOE-null mice fed a Western diet (16wk) with combined oral and intraperitoneal trehalose (8wk). This reduced aortic root plaque size in trehalose-treated mice without reducing plasma cholesterol, when compared with untreated mice [17, 18]. Stachowicz and colleagues similarly showed that orally dosed trehalose (16wk) reduced aortic atherosclerotic lesion area in chow-fed APOE-null mice [28]. Interestingly, the anti-atherosclerotic effects of oral trehalose were diminished in subsequent mouse APOE-null cohorts that were also fed a Western diet [17] or a high-fat diet (HFD) [28]. The anti-atherosclerotic effects of trehalose were recapitulated in rabbits fed a high-cholesterol diet (HC, 1% cholesterol, 8wk). Rabbits fed HC and treated with intravenous trehalose (350mg/kg 3x/week) exhibited reduced atherosclerotic plaque size without changes in body weight or plasma lipids when compared with untreated animals fed the same diet [29]. In the same set of studies at 12 wk dietary intervention, trehalose also reduced atherosclerotic plaque development and intimamedia thickness ratio [29]. In contrast, mice fed a high-fructose diet (60% fructose, 10d) exhibited reduced plasma triglycerides and cholesterol in a prevention model in which trehalose therapy was initiated prior to starting high-fructose dietary intervention [15]. In HFD-fed APOE-null mice, oral trehalose reduced plasma triglycerides were reduced after 16wk without altering plasma cholesterol [28]. Taken together, trehalose attenuates atherosclerosis in several distinct models, whereas the effects on circulating lipids may be specific to substrate, route and/or timing of trehalose administration.

Figure 1.

Metabolic effects of trehalose in multiple tissue types. Trehalose inhibits atherosclerotic plaque formation and secondary cardiomyopathies. Moreover, trehalose activates fasting signaling, reduces hepatic steatosis and activates adipose tissue browning. Gut microbial and brush border trehalases are likely to limit trehalose maximal trehalose efficacy as a metabolic therapeutic agent, which creates an opportunity for trehalose analogues that resist – or even inhibit – trehalase activity.

In a murine model of cardiac remodeling after acute ischemic insult, mice were treated with or without combined oral and IP- trehalose (4wk) following left anterior descending arterial ligation [19]. Trehalose-treated mice exhibited reduced LV dilation and increased LV systolic function after LAD ligation when compared with untreated mice. Trehalose treatment after LAD ligation was also associated with decreased pulmonary congestion, reduced cardiomyocyte apoptosis, and cardiac fibrosis. In contrast, sucrose treatment in LAD-operated mice did not exert the same cardioprotective effects. Together, trehalose improves cardiac function in acute ischemic insult and chronic cardiomyopathy models of left ventricular remodeling.

In addition to its cardiovascular effects, trehalose improved energy homeostasis and glucose homeostasis in distinct obese models. In ATG7-deficient ob/ob mice, intraperitoneal trehalose (3 doses/wk, 2g/kg) reduced fasting glucose, and glucose and insulin intolerance [12]. Moreover, trehalose significantly enhanced insulin-stimulated Akt phosphorylation in liver and skeletal muscle. In the liver, 8 wk trehalose treatment also reduced hepatic triglyceride content, and serum AST and ALT as markers of hepatocyte injury in ob/ob Atg7-heterozygous mice. Similarly, in a high fructose diet-fed model, oral trehalose (3% trehalose in water fed ad libitum) blocked fructose-induced hepatic triglyceride accumulation [15]. These therapeutic effects of trehalose on hepatic steatosis were subsequently demonstrated in a distinct HFD-fed (16wk) APOE-null mouse model. Oral trehalose (16 wk, 2.5g/kg in chow) reduced serum ALT levels, histological macrosteatosis, and quantitative triglyceride accumulation when compared with untreated, HFD-fed APOE-null controls [28].

These in vivo effects were also recapitulated using in vitro models of hepatic triglyceride accumulation, which indicates cell-autonomous mechanisms of trehalose action on hepatocytes. Trehalose treatment (100mM) reduced fat accumulation in fructose- and BSA-conjugated fat-treated primary murine hepatocyte cultures [15]. In a distinct genetic model of hepatocyte fat accumulation, trehalose treatment reduced triglyceride accumulation in cultured hepatocytes from MTTP-deficient mice, which accumulate hepatocyte lipids due to an impaired lipid efflux mechanisms [15]. More recently, genetic overexpression of effectors downstream of trehalose stimulation in hepatocytes (e.g. the lipoxygenase ALOXE3 and the arginine-metabolizing enzyme, Arginase 2) mimicked fasting-like effects of trehalose itself on hepatic and systemic energy metabolism [20, 21]. Together, recent data examining trehalose effects in murine metabolic disease demonstrate broad effects of oral and parenteral trehalose in atherosclerotic disease, cardiac remodeling, dyslipidemia and hepatic steatosis.

Trehalose Utility in Human Metabolic Disease

Because trehalose carries a “generally regarded as safe” (GRAS) designation by the United States Food & Drug Administration, recent successes in treating metabolic disease in animals have prompted the use of trehalose in humans. The range of cardiometabolic disease against which trehalose is proposed to have therapeutic efficacy is broad. Several of these clinical contexts are already examined in small human trials, and span both acute and intermediate effects on glycemia and vascular biology. In healthy Japanese subjects, acute trehalose gavage increased blood glucose to a lesser extent than glucose gavage [30]. Consistent with this, trehalose gavage acutely invoked a significantly lower insulin and plasma active gastric inhibitory peptide GIP response when compared with oral glucose gavage [30]. In context of intermediate term exposure, in a double-blind parallel treatment group comparison study, 34 healthy subjects (BMI > 23) were fed 10 g/d trehalose over a 12 wk time period had a significantly lower peak serum glucose concentration during oral glucose tolerance testing, when compared with baseline peak OGTT [31]. Stratified analysis of patients within this cohort who had increased truncal fat exhibited significant reductions in truncal fat, body weight, waist circumference and systolic blood pressure after trehalose treatment when compared with sucrose-treated controls. Taken together, acute trehalose exposure in humans reduced glycemic and insulinotropic responses, when compared with sucrose. Intermediate-term exposure modestly improves cardiometabolic risk factors and vascular function in select populations.

Trehalase

Despite promising data that demonstrate trehalose utility in modulating human metabolism, an important factor that theoretically limits trehalose’s clinical utility is its degradation by host and microbial trehalases. Trehalase is an evolutionarily conserved enzyme that hydrolyzes trehalose into two glucose molecules [1]. In vertebrates, trehalase is required for the degradation of ingested trehalose, and is predominantly expressed in the intestinal brush border and kidney. Additional reports suggest that trehalase is expressed to a lesser extent in brain and liver [17, 32]. The enzymatic function of trehalase to generate glucose from trehalose is adaptive in contexts in which glucose is scarce. However, trehalase activity becomes a liability in clinical settings where high trehalose bioavailability is desired. In mice, oral trehalose is ~99% enterohepatically cleared [11, 17]. That suggests that only ~1% of an oral bolus trehalose load is absorbed and is measurable as intact trehalose by mass spectrometry [11, 17]. Detailed biodistribution of oral trehalose has not yet been measured directly. However, the distribution of oral trehalose is the net sum of fates that includes cleavage by gut and microbial trehalase activity [18] (Figure 1), hepatocyte transport [2, 22], direct excretion [33], and absorption and action by renal and serum trehalases.

Low peripheral oral bioavailability notwithstanding, however, oral trehalose attenuates NAFLD and insulin-resistance, and secondary cardiomyopathies [15, 19]. This suggests that high peripheral bioavailability is not categorically required to exert clinically important metabolic effects. Further, oral trehalose actions on thermogenesis, white adipose tissue browning and UCP1 expression invoke hepatocyte TFEB and FGF21, in addition to inducing novel hepatocyte genes, ALOXE3 and Arginase 2, which augment the hepatocyte fasting-like response to improve insulin sensitivity and reduce hepatocyte steatosis [20, 21]. In contrast, oral trehalose alone was not protective against diet-induced atherosclerosis [17]. Other treatment models utilized routes that encompass combined enteral and parenteral, or parenteral-only administration, which precludes us from drawing conclusions on the role of enterohepatic trehalases and their role in trehalose action. One working hypothesis is that at least part of trehalose action on hepatic and extrahepatic energy metabolism requires hepatocyte-specific fasting-like responses (e.g. TFEB activation, AMPK activation, and autophagic flux), although this remains to be fully examined [11]. Nevertheless, depending on the specific metabolic effect that is quantified, enterohepatic trehalase clearance seems to attenuate part, but not all trehalose effects.

In addition to concerns regarding bioavailability, trehalase activity raises at least two additional opportunities for further exploration as trehalose advances as a metabolic therapy. These are categorized broadly as (i) serum trehalase activity and its correlation with diabetes, and (ii) microbial trehalose utilization. Correlative data in humans thus far are difficult to reconcile. Recent metabolomics data obtained as part of the ARIC cohort indicated that increased serum trehalose predicted incident diabetes [34], even after correcting for sociodemographic factors, study design features, established diabetes risk factors and fasting glucose. Moreover, polymorphisms in the trehalase locus in the African American subset of the ARIC cohort were associated with incident diabetes and serum trehalose levels. And yet, in a cohort of Pima Indians in whom trehalase polymorphisms were linked to incident T2DM [35], no correlation was observed between plasma trehalase activity and incident T2DM in this population. Based upon these data, the authors suggested that the trehalase variants might be tagging a nearby bona fide T2DM locus [35]. Finally, the data are discordant with earlier studies demonstrating elevated plasma trehalase activity in diabetic mouse models and in diabetic patients [36]. The pathophysiological mechanisms by which serum trehalases might act is simply by directly increasing serum glucose, or by scavenging circulating bioavailable trehalose away from its target tissues. Overall, the role of basal trehalose levels, trehalase genetic polymorphisms, and trehalase enzymatic activity in metabolic risk remains inconclusive. Subsequent mechanistic studies, perhaps in models with lesions in microbial and mammalian trehalases, are required to better define this complex and clinically important interaction.

Lastly, recent data suggest that microbial treA, a trehalose-6-phosphate phosphotrehalase, enhances the ability of specific Clostridioides difficile ribotypes (027 and 078) to metabolize trehalose in the absence of glucose [33, 37–39]. By contrast under similar conditions, wild-type isolates would not grow if only trehalose was given as a carbon source in vitro. Generally, glucose is abundant in the colon and stool of most mammalian hosts [40], such that the low glucose media in which ribotypes 027 and 078 were examined may not fully apply to typical clinical settings [41]. Nevertheless, the concern that these ribotypes use trehalose as a carbon source raised the consideration for trehalase-resistant trehalose analogues as next-generation metabolic therapies that (i) retain or extend trehalose’s metabolic effects while (ii) reducing concerns regarding microbial drug metabolism [1]. One example is lactotrehalose (α-D-galactopyranosyl-(1,1)-α-D-glucopyranoside), an analogue that resists porcine trehalase [1, 42], and conveys enhanced hepatocyte fasting-mimetic responses. These include activating PGC1a, FGF21, ALOXE3, and Arg2 to a greater extent than native trehalose [20, 21]. Thus, trehalose analogues are promising metabolic therapies because they ameliorate concerns over potential interactions between trehalase, microbial carbon metabolism and oral drug bioavailability.

CONCLUSION

Trehalose and its analogues are promising neurodegenerative and metabolic therapies. Primary barriers exist regarding the actions of host and microbial metabolism, which may be addressed by trehalase-resistant analogues that retain the metabolic effects of native trehalose. Activation of autophagy, hepatocyte fasting signals, and hepatocyte and peripheral TFEB are central to the bulk of its actions. However, the full cadre of mechanisms by which trehalose ameliorates metabolic disease are likely to be multi-faceted and specific to the outcome measure. Continued efforts to define these mechanisms not only informs the utility of this compound in pre-clinical and clinical contexts, it opens the possibility to elucidate new pathways to leverage against cardiometabolic diseases.

SUMMARY

• Trehalose has pleiotropic, autophagy-dependent and autophagy-independent therapeutic effects on host metabolism.

• A unifying mechanism across these effects is activation of fasting-like or stress-activated processes. This includes AMPK and TFEB, and FGF21 activation, as well as lysosomal biogenesis.

• Trehalose degradation by host and microbial trehalases represent a potential limitation to bioavailability, safety and potency of trehalose as a metabolic therapeutic, which can be addressed by trehalase-resistant analogues or trehalase inhibition.

• Significant work remains to be completed in order to elucidate the full biological effects and mechanisms of action of trehalose class compounds.

• Understanding the trehalose mechanisms is likely to lend new insights into fundamental glucose and fasting, metabolism, cell stress responses and lysosomal biogenesis regulation.