Current trends in small molecule discovery targeting key cellular signaling events towards the combined management of diabetes and obesity

Background

Deregulation of biological pathways leads to a host of diseases inhuman. Thus, identification of molecules as drugs that modulatesthe impaired molecular targets is of importance despiteconsiderable progress over decades. This is even morechallenging in combating multi-genic diseases such as cancer, diabetes and inflammatory disorders. In such cases, modulatingthe activity of multiple targets is essential to achieve optimaltherapeutic benefit [1]. Hence, a single drug or a combination ofdrugs that simultaneously impact multiple targets are preferredfor control of complex disease systems. [2].

T2D is a multi-factorial disorder resulting in abnormalities ofcarbohydrate and lipid metabolism [3]. The incidence of T2D israpidly becoming a global pandemic and has been projected toafflict more than 300 million individuals worldwide by the year2025 [4]. The epidemic of diabetes with its associatedcomplications has a huge impact on healthcare as well as humanmorbidity and mortality. Although the primary cause of thisdisease is unknown, it is clear that insulin resistance plays anearly role in its pathogenesis. Insulin resistance occurs viamultiple mechanisms and mediators. Among them, elevated freefatty acid (FFA) levels have been reported as a key cause ofinsulin resistance along with an altered glucose output anduptake [5]. Increased levels of FFA decrease insulin sensitivity viaimpairing the anti-oxidant defense by generation of reactiveoxygen species, and an over expression of inflammatory markers[6]. Elevated FFAs lead to a reduction in intracellular glutathionelevels [7] and along with inflammation and oxidative stress arethe factors that implicate adipo-genesis and insulin resistance.

NON-INSULIN DEPENDANT DIABETES MELLITUS(NIDDM) management

Our current understanding of NIDDM has helped identifyseveral potential and specific targets for the management ofNIDDM and its related secondary disorders. There are ranges ofmolecular drug targets identified on the basis of modulation ofone or more key aspects of the pathogenesis of diabetes [8]. Theseinclude (i) reduction in the excessive hepatic glucose production(Glucagon, Hexokinase, G6P, F16BP, GSK3β); (ii) increasedglucose-stimulated insulin secretion (DPP-IV inhibitors, GLP-1and GLP-1 receptor agonists); (iii) inhibition of specific moleculartargets involved in insulin signaling pathway (resistin, PTP1Binhibitors and SHIP2 inhibitors); (iv) control of obesity andaltered lipid metabolism (Adiponectin, AMPK and PTP1Binhibitors).

It should be noted that targets that control carbohydrate as wellas lipid metabolism are gaining increased attention in recentyears. PTP1B is the known key enzyme that attenuates insulintransduction cascade by desensitizing the insulin receptor [9].Since diabetes is a multi-genic disorder, drugs designed to act against individual molecular targets cannot be effective. Hence,successful strategies for control of the metabolic disorder wouldrequire a better mechanistic understanding followed bydevelopment of drugs (either single or in combination) thatimpacts multiple targets simultaneously. The combination drugscurrently employed are primarily of rational design, but theincreased efficacy they provide justifies in vitro discovery effortsfor identifying novel multi-target mechanisms. This reviewpresents a systematic approach towards identifying interactionsbetween molecular pathways that could be leveraged for themanagement of diabetes and obesity using several phytoactiveprinciples; thereby offering enhanced therapeutic potential withbroader specificity on multiple targets.

Drug templates for diabetes and obesity from plants

Natural products and their derivatives traditionally serve astherapeutic agents due to their structural diversity, biochemicalspecificity and maximum therapeutic efficacy with minimumside effects [10]. Most of these active principles originate fromedible plants, so their inclusion in diet would undoubtedly be ofvalue due to their hypoglycemic potential. Several phytomoleculesincluding flavonoids, alkaloids, glycosides, saponins,glycolipids, dietary fibres, polysaccharides, peptidoglycans,carbohydrates, amino acids, triterpeniods, steroids, xanthone,coumarins, iridoids, alkyl disulphides, inorganic ions andguanidines obtained from various plant sources have beenreported to exhibit anti-diabetic activity [11].

The anti-diabetic effect of six different plants Cichorium intybus,Cinnamomum cassia, Costus pictus, Mangifera indica, Adathodavasica, Cassia fistula and Syzygium cumuni, the bioactive moleculesisolated through bioactivity-guided fractionation is reviewed.Chlorogenic acid (CGA), a major hydrolysable tannin wasisolated from methanolic extract of Cichorium intybus, cinnamicacid was isolated from the hydro-alcoholic extract ofCinnamomum cassia, methyl tetracosanoate (MT), a saturated fattyacid was isolated from the methanol extract of Costus pictus, 3β-taraxerol a pentacyclic triterpenoid (14-taraxeren-3-ol) waspurified from ethyl acetate extract of Mangifera indica, (3β)-stigmast-5-en-3-ol is a triterpenoid isolated from the ethyl acetateextract of Adathoda vasica, Aloe emodin-8-O-glycoside (AEG), ananthroquinone from the methanol extract of Cassia fistula andVitalboside A (VBA) or Oleanolic acid-3-O-glucoside, apentacyclic triterpene glycoside was isolated from the methanolicextract of Syzygium cumuni (Table 1).

Insulin sensitisers

Insulin sensitivity is determined by the ability of insulin topromote glucose uptake and utilization. Insulin activates a seriesof signal transduction events that results in specific biologicalaction on insulin sensitive tissues. Insulin binds to the insulinreceptor leading to auto-phosphorylation of the receptor throughits tyrosine kinase activity and, subsequent phosphorylation ofinsulin receptor substrates (IRS). Phosphorylation of IRS leads toactivation of the phosphatidyl-inositol 3-kinase (PI3K) pathway,which plays an important role in glucose transport and disposal.Insulin resistance is an impairment of the signaling cascade thatresults in inappropriate glucose and lipid metabolism. The mostconclusive evidence for defective insulin sensitivity in type 2diabetes comes from euglycemic and hyperinsulinemic clampstudies, in which the total body glucose clearance is shown to bereduced in type 2 diabetic subjects [12]. Furthermore, the primarysite of reduced insulin-mediated glucose uptake is located in theperipheral muscle tissue [13, 14]. Thus, insulin insensitivitydescribed as insulin resistance with its associated abnormalitieshas become a major challenge and has prompted pharmaceuticalcompanies to search for new insulin sensitizing agents.

3β-taraxerol, a pentacyclic triterpenoid (14-taraxeren-3-ol)purified from Mangifera indica has been observed to show insulin sensitizingaction in an in vitro adipocyte model. The sensitizingeffect of the natural compound was revealed through itssignificant effect on insulin stimulated glucose uptake, in anIRTK/PI3K dependant activation that facilitated an effectiveGLUT4 translocation and activation [15]. Methyl tetracosanoate(MT) exhibited its promising anti-diabetic activity through itsinsulin sensitizing action by activation of key targets of insulinsignaling cascade in a similar manner [16]. Vitalboside A, a smallmolecule derived from Syzygium cumini showed activity as aninsulin sensitizer by increasing glucose uptake in bothdifferentiated L6 myocytes and 3T3-L1 adipocytes throughPI3K/AKT signaling pathway [17]. TZDs are anti-diabetic drugsthat enhance insulin sensitivity and so the insulin sensitizingeffects of compounds discussed in this review were validatedwith a derivative of TZD - Rosiglitazone. Since thesebiomolecules are known to modulate the insulin signaltransduction pathway their activity could be therapeuticallyrelevant.

We have also postulated that cinnamic acid activates glucosetransport in L6 myotubes independent of PI3K activation [18].Adisakwattana et al. [19] have reported the insulin secretagogueeffect of cinnamic acid and its derivatives. This dual activity ofcinnamic acid on both insulin signaling and secretion reveals it asan ideal prototypic compound for structural and functionalstudies.

Insulin mimetics

Another class of compounds that improves insulin action is theinsulin mimickers. They act similar to insulin and exhibit theireffects on insulin sensitive tissues independent of insulinstimulation. An uptake of glucose into the insulin targeted tissuesis one of the mechanisms implicated in the blood glucoselowering effect of these agents, which seems to result frombeneficial extra-pancreatic actions on glucose clearance rather than from enhanced insulin release [20]. Sujatha et al. havereported (3β)-stigmast-5- en-3-ol, a plant phytosterol to possessinsulin like effects [21]. This insulin mimetic effect of (3β)-stigmast-5-en-3-ol was initially established based on its efficacy instimulating basal glucose uptake in differentiated L6 myotubesand, was later confirmed through molecular mechanistic studieson the targets involved in the insulin signaling cascade. Asequential activation of IR-β, IRS-1, PI3K, AKT/PKB, PKCα andGLUT4 were observed in rat skeletal muscle cells. Since therestoration of glucose uptake activity by (3β)-stigmast-5-en-3-olwas evidenced without stimulation of insulin, it led us to suggestthe involvement of insulin-like property to be the mechanismunderlying the in vitro anti-diabetic activity [21]. Wang et al. [22]have demonstrated the cholesterol-lowering efficacy of (3β)-stigmast-5-en-3-ol. Based on these evidences, we hypothesize that(3β)-stigmast-5-en-3-ol can be of interest as an anti-diabetictherapeutic, since it improves insulin resistance through itsinsulin mimetic property coupled with cholesterol loweringpotential. Similarly, AEG exhibits insulin mimetic activity by upregulatingglucose uptake irrespective of insulin stimulationthrough the activation of IR-β, IRS1, p85 subunit of PI3K, PKBand finally GLUT4 translocation [23].

Activators of glucose storage

In T2D, insufficient action of insulin in liver and elevatedcirculating levels of glucagon could account for the high rate ofhepatic glucose production, contributing to increased bloodglucose levels [24]. Modulation of the production of hepaticglucose could be an important strategy towards inducing glucosestorage through glycogen synthesis. Glycogen synthesis occursby activation of the insulin-signaling pathway in which PI3Kphosphorylates protein kinase B (Akt) that in turn inhibits GSK3βactivity. These sequential events result in the activation ofglycogen synthase, subsequently converting glucose to glycogen[25]. T2DM is associated with increased levels of GSK3β, andinhibitors of its activity can reproduce the effects of insulin.Therefore, inhibition of GSK3β might represent a potential targetto enhance glycogen synthesis. Although, there are very fewagents that have an effect on glycogen stores (Metformin), theystill possess the risk of developing side effects like hypoglycemia.3β-taraxerol and AEG two structurally dissimilar compounds ofdifferent origin have been reported to induce glycogen synthesisindependent of insulin stimulation. Both the compounds showedsignificant inhibitory effects on GSK3β activity through activationof PKB [15, 23]. Likewise, MT a saturated fatty acid also exhibitedsignificant inhibition of GSK3β activity [16].

Inhibitors of adipogenesis

The progression of diabetes and its associated complications hasbeen observed to increase with obesity. However, currentlyavailable anti-diabetic drugs like Thiazolidinediones andSulphonylureas promote obesity while reducing blood glucoselevels. Hence, drugs that regulate carbohydrate metabolismwithout disturbing lipid metabolism are the current need of thehour [26]. As herbal preparations are anticipated to offer strongtherapeutic potential, we have attempted to address this issuewith Cichorium intybus, a salad crop. Since the methanol extract ofthe plant (CME) promoted glucose transport in a PI3K dependantmanner combined with inhibition of adipocyte differentiation in 3T3L1 adipocytes, we attempted to isolate the bioactive principleresponsible for the dual activity. As tannins formed the majorcomposition of the extract the role of tannins present in themethanol extract was evaluated after detannification (CME/DT).But surprisingly the activity exhibited by the detannified extractwas exactly vice versa to the methanol extract [27]. Thisinteresting observation coincides with early reports on inhibitoryrole of tannins in adipocyte differentiation [28].

Peroxisome proliferator activator receptor (PPAR),CCAAT/enhancer binding protein (C/EBP), and sterolregulatory element binding protein (SREBP) families are the welldocumentedprimary adipogenic transcription factors involved inadipocyte differentiation. Among them PPARγ is mostextensively studied for its therapeutic utility in the treatment ofNIDDM [29]. The methanol extract of Cichorium intybus (CME)controls the expression of all the three adipogenic markersandthe molecular mechanisms of the anti obesity and anti diabeticeffects of flavonoids have been well studied [30]. The detannifiedextract (CME/DT), however did not exhibit an inhibitory activity[27]. Based on this Chlorogenic acid (CGA) was isolated fromCME [27]. Despite isolating the pure molecule (CGA) afterenrichment from the crude extract (CME) there was no significantenrichment in adipogenesis inhibition [27]. Wagner et al hasreported that the therapeutic superiority of crude extract over anisolated single molecule would be caused due to additive effectsof the other molecules present in the extract [31]. The plant C.intybus has been reported for the presence of cichoric acid (CHA),traces of caffeic acid (CFA) and cinnamic acid [32, 33]v. Hence, wehypothesize that association of CFA and CHA would be addingto the net biological activity of CME. This observationsubstantiates the earlier hypothesis on higher efficacy of tanninspresent in CME than the non-tannins present in CME/DT (withPPARγ agonism).

Similarly, the methanol extract of Costus pictus has been reportedto possess dual activities like anti-diabetic and anti-adipogenicactivities. But the identified bioactive principle methyltetracosanoate (MT) did not exhibit adipogenic inhibitory effect[16]. The prospect of identifying the anti-adipogenic moleculefrom CPME favoring treatment options for obesity-induceddiabetes exists. We have also reported Aloe emodin-8-Oglycoside(AEG), a plant derived anthroquinone to possess apromising effect on adipogenesis inhibition thus highlighting theimportance of it as a glucose-lowering agent without inducingadipogenesis [23]. The methanolic extract of Syzygium cumini(SCME) and its dual bioactive compound, Vitalboside A (VBA)were found to exhibit both anti-diabetic and anti-adipogeniceffect. Interestingly, VBA inhibited partial PPARγ2transcriptional activity in transiently transfected HEK-293 cellsresulting in partial inhibition of lipid accumulation in adipocyteswithout affecting synthesis and secretion of adiponectin, [17].Further, investigation of the in vivo efficacy of SCME on insulinresistance in high fat diet (HFD) induced obese C57BL/6J miceshowed significant reduction in insulin resistance, dyslipidemiaand body weight gain with concomitant increase in glucosetolerance. Elucidation of bifunctionality of SCME was found toexhibit inhibition of PTP1B and partial activation of PPARγ activity in skeletal muscle and visceral adipose tissue of HFD fedobese mice.

PTP1B inhibitors

Targeting the common form of insulin resistance in type 2diabetes is difficult since it involves multiple non-specific insulinsignaling defects [3]. Usually, most intracellular signaling takesplace via cascades of phosphorylating enzymes (kinases) and thede-phosphorylation reactions are catalyzed by proteinphosphatases. Protein tyrosine phosphatases (PTPs) belong to thefamily of classical and dual specificity phosphatases and, haveemerged as new and promising class of signaling targets since thediscovery of PTP1B [34]. Protein tyrosine phosphatases(PTPases), act as physiological negative regulator of insulinsignaling by dephosphorylating the activated insulin receptor(IR), thereby limiting the insulin-signaling pathway. Anincreased expression and activation of PTPases has beenobserved in muscle and fat from rodents, obese and diabetichumans and are believed to be involved in the pathogenesis ofinsulin resistance [3]. PTP deleted mice have been reported to behealthy, lean and obesity resistant with reduced levels of insulinand glucose. This phenotype has been ascribed to restore insulinsensitivity through insulin receptor (IR) and insulin receptorsubstrate (IRS1) in the absence of PTP1B, thus projecting theimportance of PTP1B as drug target for diabetes and obesity [35][36]. PTPs have control over both leptin and insulin signaling [37],therefore inhibition of PTPase activity seems to be an appealingstrategy for obesity and diabetes.

We have also isolated bioactive molecules from several plantswith potential PTP1B inhibitory effect. Chlorogenic acid (CGA)was isolated from C. intybus through bioactivity-guidedpurification and reported by our group [27]. It is an ester formedbetween caffeic acid and (L)-quinic acid and is a majorhydrolysable tannin compound present in coffee [38]. CGA hasearlier been reported for glucose uptake [39], which was later,confirmed by Muthusamy et al [27]. Rodinone et al [40] observeda decrease in adipogenesis with a down regulation of alladipogenic genes in animals ablated with PTP1B, and postulatedthe pivotal role of PTP1B in the development of obesity.Similarly, we have reported that the PTP1B inhibitory potential ofCGA would be the linking factor that is responsible for theobserved increase in glucose uptake and decreased adipogenesis[27]. Analysis of CGA and other caffeoyl derivatives of themethanolic extract of cichory salad leaves were found to exhibitbetter insulin sensitivity in both in vitro and in vivo experimentsthrough inhibition of PTP1B [41]. Our study further confirmedthe non-competitive binding nature of CGA and CHA, whichwas established by co-incubating it with sodium orthovanadate(SOV) a protein tyrosine phosphatase inhibitor [27]. Vanadatebased inhibitors inhibit PTPs non-selectively and have been usedwith some success in clinical trials with only moderate toxicity.This suggests that PTP1B inhibiting drug need not have absoluteselectivity [34]. Computational approaches to study the bindingand interaction of these two compounds at the Allosteric site ofPTP1B located ~20 A away from the catalytic site was performedusing molecular docking and MD simulations. Results showedsignification allosteric interactions and modulation at theallosteric alpha 7 helix restraining the activation of the protein, in comparison to Compound-2, a standard allosteric inhibitor [42].However, recent experimental evidence using NMR- chemicalshift analysis accompanied with docking and MD simulationsshow a non-competitive binding of CGA and a competitiveinhibition by CHA [43]. Although, this could be due to the factthat these compounds were not subjected to co-incubation assaywith a strong catalytic inhibitor such as SOV, it requires furtherinvestigation.

We have also reported the PTP1B inhibitory potential of 3β-taraxerol, a triterpenoid from Mangifera indica and Methyltetracosanoate (MT), a saturated fatty acid from Costus pictus incomparison to SOV [15, 16, 44]. The Aloe emodin glycoside(AEG) isolated from Cassia fistula was observed to exhibit amoderate effect on PTP1B inhibition eliminating the possibility ofAEG as a potent PTP1B inhibitor [23]. The methanolic extract ofSyzygium cumini (SCME) and the dual active moleculeVitalboside A (VBA) it showed a dose-dependent inhibition ofPTP1B. Molecular docking and co-incubation studies show themto selectively inhibit PTP1B at its allosteric site [17]. Thus, it isevident that compounds with different structures isolated fromplant species can act as active moieties and serve as templates forthe development of PTP1B inhibitors.

Conclusion

Our continued interest to identify therapeutics for diabesity hasresulted in the isolation of several active principles from differentplant sources. These principles are structurally dissimilar withdifferent characteristic features and yet they exhibit significantbiological activity. It is evident that each of the active moietiesisolated so far have shown potential in leveraging multipletargets involved in diabesity with their unique mode of actiontowards commercial exploitation by pharmaceutical industry.Hence, it is prudent that further evaluation of the properties andapplications of these compounds will greatly enhance theirperspectives in drug discovery.