Minor Groove Binders for DNA as Antitrypanosomal Agents: the Veterinary Context

Animal African trypanosomiasis (or nagana) is a wasting livestock disease found in sub-Saharan Africa and caused by protozoan parasites Trypanosoma congolense, T. vivax and T. brucei. Chemotherapy (diamidine diminazene aceturate) and chemoprophylaxis (phenanthridine isometamidium) are essential for disease control. Current treatments have reduced efficacy due to increased drug resistance, the need for new veterinary trypanocides becomes a high-priority. All three of the above African species are susceptible to minor groove binding drugs from Strathclyde (S-MGBs) which is important because compounds useful in the field must have activity across the range of infectious species so that characterisation of the infection is not required at diagnosis. Over 100 S-MGBs have been evaluated in Glasgow and Switzerland, and compounds with development potential have been identified. As an example, S-MGB 234 has been shown to be curative in in vivo models of trypanosome infection in mice. Importantly S-MGB 234 does not show cross resistance with other antitrypanosomal drugs such as diminazine, isometamidium, or ethidium bromide, which is consistent with a different route into the parasite’s cell. S-MGBs that contain alkene links, such as in S-MGB 234, are the most active sub-class of S-MGB and point the way towards structural optimisation.


Introduction
From the early stages of our MGB programme we have had evidence for activity against kinetoplastid parasites, especially Trypanosoma brucei.Over the last five years the study has been developed in collaboration with the University of Glasgow and the Swiss Tropical Health Institute, Basel, to reach the stage at which advanced leads and potential development candidates have emerged.In reaching this point, some generalities concerning structure-activity relationships have been observed.This paper summarises the salient features by selecting specific small groups of closely related compounds.
An effective anti-trypanosomal compound for use against African Animal Trypanosomiasis (AAT) must be active against all of the species causing disease so that identification of a species is not a necessary part of diagnosis.Moreover such a compound must also be effective against strains of trypanosomes resistant to currently used drugs such as isometamidium and diminazine.Evidence is presented to show that S-MGBs do not show cross resistance.
Lastly, a development candidate must be active in an in vivo model of disease.Data showing that S-MGBs are curative in a mouse model of trypanosomiasis are presented.
Part of the S-MGB library (about 100 compounds) was screened in vitro against T. congolense (IL3000 BF) and ex-vivo against T. vivax (STIB 719/ILRAD 560 BF) for trypanocidal activity.To evaluate selectivity against mammalian cells, the toxicity of each compound to the L6 murine cell line was determined.
The S-MGB library consists of compounds of general structure illustrated by S-MGB 234, shown on the next slide.Three different N-terminal (head groups) are included, amide, amidine, and alkene, of which the alkene in general shows the best anti-infective activity.Examples of differences cause by these structural changes are shown in the following slides.
Five S-MGBs had IC 50 less than 100 nM against T. congolense and 7 had IC 50 less than 100 nM against T. vivax.There was a broad but not exact correlation between the two activities (see following slides).On the basis of their determined activities and selectivity indices, 7 S-MGBs were chosen for further evaluation of which 3, S-MGBs 234, 235, and 248 were progressed to the in vivo disease model stage.
The data are summarised on the following chart, which shows the activity against T. congolense in comparison with that against T. vivax.The bubble size represents the effect on L6 cells; the larger the bubble, the lower the toxicity.Perhaps unsurprisingly some of the most active compounds (lower left quadrant) are also the most toxic.However, as the data presented in the following slides show, several S-MGBs have sufficiently large selectivity indices to warrant further investigation.

Structural features of S-MGBs
Strathclyde MGBs (S-MGBs) of all three structural types have been investigated.Tertiary amine and amidine tail groups have been included.The following chart shows the activity for the S-MGBs against Trypanosoma congolense and T. vivax; the bubbles represent the toxicity against the L6 cell line, the larger the bubble, the lower the toxicity.Significant compounds in this study have been numbered.Whilst some activities in the pairs above are very similar, the selectivity indices (SIs) differ greatly.Two factors may contribute strongly to this. 1.The morpholino tail groups are weakly basic because of self association of the MGB and largely unprotonated at physiological pH leading to highly lipophilic compounds.2. In contrast, the amidine tail groups are essentially permanently protonated giving more soluble compounds that could be substrates for specific carrier proteins.

Comments on SAR
The trends observed in the previous 3 slides are representative of observations from the full collection of S-MGBs studied.Different structural features turn out to contribute to the best profiles discovered so far for several different applications of S-MGBs (antibacterial, antifungal and antiparasitic).The currently available generalisations are described in the accompanying paper 'Selectivity in anti-infective minor groove binders'.
With respect to this representative set of compounds and the antitrypanosomal activity, there is no evident correlation between the activity and typical medicinal chemical parameters such as clogP and tPSA.Indeed S-MGBs are characterised by large polar surface areas (110 -150 for the most active alkene-containing compounds) and S-MGBs with clogP values in the range 2.5 to 5.0 have all been found to be active.An added complication is the known tendency of S-MGBs to aggregate in solution which makes an SAR basis in physicochemical properties difficult to define (see Parkinson et al., Med. Chem. Commun., 2013, 4, 1105-1108).
The range of activities observed might be understood through different DNA-binding properties and by different uptake / efflux into and from target cells.Because of the multiple local targets expected on trypanosomal DNA there are no data with which to discriminate between compounds for DNA binding.The following biological evaluation, however, implicates uptake as important in the activity of S-MGBs.In vitro trypanocidal activity of selected S-MGBs against two diminazene-resistant T. congolense lines (DimR and EMS MUT DimR) as compared to wild type (WT) (mean ± SEM, n≥3) were determined at the University of Glasgow.Ratios (RF) close to unity indicate that the two strains compared are of essentially equal sensitivity to the S-MGB.The data suggest that the S-MGBs do not act by the same mechanism as diminazene; both differences in uptake and differences in intracellular activity may contribute.
(Note: The EC 50 data here differ from those reported in the SAR comparisons because the latter were obtained at the Swiss Tropical Health Institute under slightly different conditions.) Evidence for lack of cross resistance to diminazene S-MGB 234 and 235 have significant in vivo activity against T. congolense STIB 736/IL1180 mouse model of infection.The positive curative effects and the significant length of survival (at least to end of experiment) are notable.In view of this and the lack of cross resistance, understanding the mechanism of action of S-MGBs becomes important.Some relevant information is presented in the following slides.

In vivo activity against T. congolense
Conventional diamidine minor groove binding trypanocidal drugs cause destruction of the kinetoplast, which is an essential component of their mechanism of action.S-MGBs might be expected to show similar effects.This has been tested using ISMR1, which is a T. b. brucei in vitrogenerated isometamidium resistant line with reduced mitochondrial membrane potential and loss of kinetoplast.
Evidence for significance of kinetoplast in S-MGB activity S-MGB 234 is as active against akinetoplastic T. b. brucei , ISMR1, as it is against wild type.This strain is resistant to diminazene and to the phenanthridines, isometamidium and ethidium bromide.
The presence of the kinetoplast therefore is not necessary for the trypanocidal activity of the S-MGBs.This is a distinctive feature of their mechanism of action.

Antitrypanosomal activity compared with antibacterial
Evidence has been obtained from metabolomics studies which show that many metabolite levels are altered on treatment of T. b. brucei with S-MGB 234 (5 × EC 50 , 8 h).The major changes in identified metabolites were in the nucleoside / nucleotide pool together with some derived cofactors (pyridoxine and pyridoxamine).2-Oxoglutarate was also found to have a significant change.These changes relate to two of those observed in the bacterial experiments using RNAseq methodology.
The metabolomics observations can be compared with the results from RNA-seq experiments using another S-MGB (3) attacking Staphylococcus aureus (see paper 'Why antibacterial minor groove binders are a good thing.).In the latter case, challenge with S-MGB-3 caused hundreds of changes in gene expression, up and down regulation, that could be mapped on to several areas of metabolism.One of the most prominent of these was nucleotide metabolism.Energy production and lipid metabolism were also affected.There are thus some similarities between the effects of the S-MGBs on the two very different species.
It can be suggested, therefore, that S-MGB 234 by binding to trypanosomal DNA causes multiple biological effects in a manner related to that by which S-MGB 3 affects S. aureus.Since the selectivity of S-MGB 234 appears to be due to an uptake mechanism different from that of current drugs, a change in transporter availability or effectiveness may lead to resistance.

S-MGBs as a class have significant antitrypanosomal activity that shows clear structure-activity patterns.
• Amidine tail groups provide the best balance between activity and in vitro toxicity.
• Weakly basic tail groups are associated with compounds with higher toxicity.
• The most active compounds have arylalkenyl head groups and amidine tail groups.S-MGBs are promising leads for the treatment of animal African trypanosomiasis.
• Good in vitro activity combines with proved in vivo efficacy against relevant Trypanosoma species; • They are active in vitro against parasites resistant to currently used veterinary trypanocides.
Despite being both minor groove binders, S-MGBs and diamidines are unlikely to be crossresistant.
• S-MGBs do not share the same transporters involved in diamidine uptake and resistance in T. b.
brucei.• The kinetoplast, which is disintegrated by diamidine exposure, is not the main target of the S-MGBs.

S-MGBs are cytocidal and affect cell division.
• Treated parasites do not proliferate and are unable to undergo correct cytokinesis.
• Some nucleotides accumulate in treated parasites, possibly due to decrease in DNA synthesis or to DNA disintegration.
MGB 234 does not cause destruction of the kinetoplast it has major effects on the DNA-containing organelles.Treatment of T. b. brucei with S-MGB 234 causes aberrations in the DNA-containing organelles, their segregation and their morphology as well as an apparent block of cytokinesis.An accumulation of cells with multiple kinetoplasts and nuclei (MKMN) is observed over time in the treated parasite population.Exposure to S-MGB 234 (5 × EC 50 ) for 10 h has irreversible effects on T. congolense survival showing that this S-MGB, and presumably others, is cytocidal.do not act fast.Exposure to 5×EC 50 S-MGB 234 kills most T. b. brucei within 48 h and sterilises T. congolense cultures within 72 h.Abnormal, sick cells are visible before death occurs.Treated cells do not proliferate, which is further evidence for cytocidal activity.