11.7.8 Conclusions and comments
Studies on antifungal chemotherapy continue to lag behind those on antibacterial agents. As further information is gleaned as to the structural complexity of and biosynthetic processes in fungal cells, it is hoped that a more logical design of powerful new antifungal compounds can be achieved. Certainly, new azole derivatives can be expected to become increasingly important, but approaches such as that described in Section 11.7.7 may produce clinically useful alternatives.
The design of antiviral agents presents yet another problem. Since viruses literally 'take over' the machinery of the infected human cell, an antiviral agent must show a remarkable degree of selective toxicity to inhibit the viral cell without having concomitant action on the human cell. In contrast, the metabolism of pathogenic bacteria is sufficiently different from that of human host cells to render these microorganisms sensitive to inhibitors (e.g. penicillins) which have little or no effect on the metabolism of the host.
The genetic information for viral reproduction resides in its nucleic acid (RNA or DNA). The viral particle (virion) does not contain the enzymes required for its own reproduction and after entry into the host cell the virion either uses the enzymes already present or induces the formation of new ones. Unlike bacteria, viruses multiply by synthesis of their separate components, followed by assembly.
The amantadines exert a concentration-dependent effect: a selective strain-specific inhibition of replication of influenza A viruses at relatively low concentrations (<10 pM) and a non-specific inhibition of virus infection at concentrations >100 pM. The selective action at low concentrations does not involve a direct virucidal effect or prevent adsorption of virus to cells, but infection is inhibited at an early, pre-synthesis stage. Amantidine hydrochloride (11.142) has a very narrow spectrum and its use is usually restricted to prevention of influenza A.
Methisazone, idoxuridine and cytaribine inhibit DNA, but not RNA, viruses. Idoxuridine (IUD: 5-iodo-2'-deoxyuridine (11.143)) is a thymidine analogue which is incorporated into the viral DNA in place of the natural substrate. Cytaribine (cytosine arabinoside; Ara-C (11.144)) is active against variola; it does not prevent absorption of the virus, or its penetration or synthesis of viral DNA, and appears to inhibit synthesis of viral proteins.
Because of its toxicity, idoxuridine is unsuitable for systemic use, and it is restricted to topical treatment of herpes-infected eyes. Cytaribine is significantly more toxic than idoxuridine.
Other nucleoside analogues are at least as active as idoxuridine, e.g. adenosine arabinoside (Ara-A; vidaribine, (11.145)). Ribavirin (1-P-D-ribofuranosyl-1,2,4-triazole-3-carboxamide (11.146) is a synthetic nucleoside with a broad spectrum of activity, inhibiting both RNA and DNA viruses.
Acycloguanosine (acyclovir) is a nucleoside analogue (11.147) which becomes activated only in infected host cells. It is active against herpes viruses and less so against
varicella zoster (shingles). Initially, acyclovir was marketed as an ophthalmic product, but later an intravenous formulation was introduced for treatment in life-threatening herpes infections in immunocompromised patients.
In brief, acyclovir is active only against replicating viruses. It is activated in infected cells by a herpes-specific enzyme, thymidine kinase. This enzyme initiates conversion of acyclovir initially to a monophosphate and host kinases subsequently convert this to the antiviral triphosphate, which inhibits the action of viral (but not, to the same extent, host cell) DNA polymerase. The triphosphate is incorporated into the nascent DNA strand at its growing point causing chain termination. This is highly significant when considering the potential for toxicity. The failure of the molecule to be incorporated into mature host DNA reduces the risk of mutation in subsequent rounds of DNA replication.
Human cytomegalovirus (CMV) is only weakly inhibited by acyclovir but is more sensitive to the related acyclic nucleoside 7-(2-hydroxyl-1-(hydroxymethyl)-
ethoxy) methyl- guanine (ganciclovir (11.148)). CMV does not encode thymidine kinase but possesses a protein with an amino acid sequence similar to known protein kinases which suggests a role in protein phosphorylation for the virus gene product.
An important aspect of acyclovir action is its obligatory chain termination. This aspect has been exploited in other drugs.
The first drug shown to be effective for the treatment of AIDS (caused by HIV, a retrovirus) was the nucleoside analogue 3'-azido-2',3'-dideoxythymidine (zidovudine, AZT (11.149)). The 3' hydroxyl group which provides the attachment point for the next nucleotide in the growing DNA chain during reverse transcription is replaced by an azido group. Thus this molecule is also an obligate chain terminator. AZT is incorporated in place of thymidine and is an extremely potent inhibitor of HIV replication. However, AZT is relatively toxic since it is converted to the triphosphate by cellular enzymes and therefore is also activated in uninfected cells.
Two further analogues, 2',3'-dideoxycytidine (ddC (11.150)) and 2',3'-dideoxyinosine (ddI (11.151)), which are both obligate chain terminators, have been developed for HIV. Unfortunately, these also lack selectivity and cause side-effects in man. Other potent dideoxy analogues are in development such as 2'-deoxy-3'-thiathymidine (3TC (11.152)) and 2',3'-dideoxy- 2',3'-didehydrothymidine (D4T (11.153)) and it is hoped that they may provide more selectivity. However, resistance to these drugs can build up due to mutations.
Sorivudine (E-5-(bromovinyl) arabinofuranosyluracil or BVaraU (11.154)) is the most potent inhibitor of varicella zoster virus described to date. The compound is activated by the virus thymidine kinase but the precise mechanism of inhibition of DNA synthesis is unknown. Unfortunately, a metabolite formed from its degradation, 5-bromovinyl uracil, has led to serious clinical problems when administered to patients being treated with the anti-tumour drug 5-fluorouracil (5-FU (11.155)).
Fialuridine (2'-fluoro-5-iodo-P-D-arabinofuranosyl uracil, FIAU) is active against hepatitis B, but the liver is also the major target for toxicity, and several patients enrolled in a trial of this compound died from liver failure.
Famciclovir (11.156), the pro-drug of penciclovir (7-(4-hydroxy-3-hydroxymethylbut-1-yl) guanine (11.157)) is active against varicella zoster and herpes simplex. Nucleoside analogues are often poorly bioavailable via the oral route, and with penciclovir the bioavailability was so low as to preclude its oral use. Famciclovir overcomes this problem and is metabolised in the body to produce penciclovir. The action of penciclovir is similar to that of acyclovir in that it is converted to the active triphosphate form by a herpes-specific thymidine kinase, inhibiting viral DNA synthesis. It is not, however, an obligate chain terminator and could therefore possibly be incorporated into cellular DNA. Penciclovir triphosphate has relatively weak activity against varicella zoster and herpes simplex, but has a longer half-life and achieves higher levels of the triphosphate form than acyclovir.
Valaciclovir (11.158), the valine ester pro-drug of acyclovir, has been developed recently to increase the oral bioavailability of acyclovir. Valaciclovir breaks down rapidly to produce acyclovir and valine. Plasma levels of the drug are three to five times higher using Valaciclovir than those using oral acyclovir. Valaciclovir has been shown to be more effective than acyclovir in resolving zoster-associated pain, and it is equally effective against genital herpes but requires less frequent dosing.
Foscarnet (sodium phosphonoformate) is a pyrophosphate analogue (11.159) which has potent activity against herpes simplex cold sores and is non-toxic when applied to the skin. It inhibits herpes DNA polymerase.
The phenomenon of 'viral interference' means that one virus greatly modifies the response of the host to infection with a second, immunologically distinct virus. The term
interferon is applied to a class of basic polypeptides of molecular weights ca. 2000030000 induced by viruses but the term is also applied to materials (molecular weights ca. 87000) not induced by viruses. Interferons are thus produced by the host cell in response to the virus particle, the viral nucleic acid and non-viral agents (e.g. natural and synthetic polypeptides). The interferon system involves the induction of an antiviral protein (interferon) by a well-defined inducer and its subsequent interaction with the virus, leading to the development of an antiviral state.
For many years, yields of interferons from eukaryotic cells were disappointingly low. However, the application of recombinant DNA technology and cell culture technology has allowed the production of large quantities of interferon. This work is potentially one of the most exciting applications of molecular biology to the design of potent antiviral (and anticancer) agents.
Extensive clinical trials have shown interferon to have limited usefulness to date. Although shown to be antiviral in animal models, results in humans have been disappointing. It is currently licensed in the USA for treating hairy cell leukaemia, Kaposi sarcoma and refractory condyloma acuminata. A number of studies are currently in progress, many to determine the effectiveness of interferons in combination with other antiviral agents.
Until relatively recently, few antiviral drugs were available and little was known about their mode of action. Now, not only are more drugs in clinical use but also information about their mechanism of antiviral activity has improved substantially, allied to which is the increasing depth of knowledge about resistance mechanisms.
Many of the features of bacterial resistance, viz. plasmid-encoded insusceptibility, reduced drug uptake, drug efflux or drug inactivation, do not apply to viruses. Drug resistance of viruses usually results from point mutations which lead to alterations in proteins (enzymes) that normally either activate drugs or are inhibited by them.
Activation of Ara-A to the triphosphate, Ara-ATP, occurs in both herpes-infected and uninfected cells, although the herpes DNA polymerase is a target for Ara-A, and this contributes to its selectivity. Resistant mutants specify DNA polymerases which are less susceptible to Ara-ATP than the enzyme specified by the wild-type virus. Acyclovir triphosphate is a more potent inhibitor of herpes simplex DNA polymerase than of the cellular enzyme. Acyclovir resistance arises as a result of mutations in the thymidine kinase or DNA polymerase genes. Two thymidine kinase mutations have been recognised for herpes simplex. One produces strains deficient in thymidine kinase, resulting in reduced or no phosphorylation of acyclovir. However, thymidine kinase-deficient mutants tend to be less pathogenic, though immunocompromised patients may still be at risk. The second mutation results in altered binding properties for acyclovir. DNA polymerase mutations may confer only marginal shifts in drug susceptibility, but the mutant virus can retain its virulence.
Ganciclovir is up to 100 times more active than acyclovir against CMV. Resistance to ganciclovir has been associated with point mutations in the catalytic domain of the phosphotransferase gene, which results in no phosphorylation of ganciclovir.
Foscarnet-resistant mutants of CMV and herpes simplex have been found clinically. Resistance is due to mutation of the viral DNA polymerase gene. Interestingly, foscarnet-resistant herpes simplex tends to be susceptible to acyclovir, suggesting that the drugs may have different binding sites on the viral DNA polymerase.
Amantadine prevents uncoating of influenza A, blocking release of viral RNA into the cytoplasm. Resistance is due to a mutation in the M2 (matrix) protein.
Prolonged treatment of HIV with AZT results in resistance to the drug. Various mutations in the viral reverse transcriptase gene are responsible for resistance. The number and combinations of the various mutations relate directly to the level of resistance achieved, and there tends to be no cross-resistance to other dideoxynucleotide analogues. Drug combinations are now in use to attempt to overcome this problem, and results from ongoing trials are awaited with interest.
In many instances (e.g. P-lactams), the development of chemotherapeutic agents for use in the treatment of bacterial, fungal or viral infections has generally followed an enlightened empirical pattern, based upon experimental testing and observations of series of compounds, rather than upon theoretical design.
With antibacterial drugs, for example, screening of moulds or Streptomyces for the production of antibiotics or enzyme inhibitors has often been followed by the isolation, by further chemical means, of material that is suitable for the design of additional drugs. With increasing knowledge about their mechanism of action, and of bacterial resistance to them, at the molecular level, more sophisticated drugs should be designed with, for example, improved intracellular penetration, enhanced enzyme resistance or increased binding at the sensitive site in the cell.
Some areas are worthy of future consideration. The first utilizes permease exploitation and in principle has the concept of fooling a transport system into accepting an antibiotic as if it were the natural substrate (see Section 11.4.4). The antibiotic may act as a carrier for a smaller molecular weight 'warhead' function which is selectively released only at its intracellular target site. A second relies on suicide inhibitors (see chapter 8) initially being recognised and processed by the target enzyme as if they were a natural substrate. This is followed by molecular rearrangement of the inhibitor to produce a reactive 'warhead' species which irreversibly inhibits the enzyme. A third area, which is particularly appropriate in hospital settings where the possibility of bacterial resistance is greater, involves the construction of hybrid molecules which are metabolized to produce two different types of antibacterial agents. Hybrid cephalosporin-quinolone molecules have been synthesized which produce the active cephalosporin and quinolone moieties when administered. It is too early to say whether such constructs will be useful in a clinical setting.
Adherence to surfaces, mediated by bacterial surface appendages, is essential for bacterial colonization. Adherent bacteria are less sensitive to antibiotics and to natural host defence systems, although several existing antibiotics will affect adhesion. Competitive inhibition of bacterial adhesion can be achieved in vitro, but problems remain in vivo. Selective interference with adherence could be a useful concept in drug design, although it has yet to be demonstrated that such an approach would be clinically useful.
Antifungal compounds have often been developed following chance observations, the screening of antibacterial agents or of substrates that have been discarded because of a lack of effect on bacteria. Modifications of chemical structure and an empirically based testing procedure have usually followed. Information on the mechanism of action of antifungal compounds has increased considerably, as has that on the underlying reasons for tissue toxicity. Taken with an improved knowledge of the fungal structure and molecular biology, these facts lead to the hope that more effective, less toxic antifungal agents can be designed.
Antiviral drugs have often proved to be disappointing, the major problem being the achievement of a selectively toxic effect on the infecting virus without a concomitant harmful effect on the host. However, the introduction of acyclovir and newer nucleoside analogues as antiviral agents augurs for a brighter future, though the dramatic changes in biological properties induced by apparently simple changes in these molecules means that, for the moment, each compound must be judged on its own merits.
Bowden, K., Harris, N.V. and Watson, C.A. (1993) Structure-activity relationships of dihydrofolate reductase inhibitors. Journal of Chemotherapy 5, 377-388.
Bugg, C.E., Carson, W.M. and Montgomery, J.A. (1993) Drugs by design. Scientific American 269, 92-98.
Dreyfuss, M.M. and Chapela, I.H. (1994). Potential of fungi in the discovery of novel, low-molecular weight pharmaceuticals. Biotechnology 26, 49-80.
Greer, J., Erickson, J.W., Baldwin, J.J. and Varney, M.D. (1994) Application of the three-dimensional structures of protein target molecules in structure-based drug design. Journal of Medicinal Chemistry 37, 1035-1054.
Hamilton-Miller, J.M. (1994) Dual-action antibiotic hybrids. Journal of Antimicrobial Chemotherapy 33, 197-200.
Hooper, D.C. and Wolfson, J.S. (1993) Quinolone Antimicrobial Agents, 2nd edn. Washington DC. American Society for Microbiology.
Horan, A.C. (1994) Aerobic actinomycetes: a continuing source of novel natural products. Biotechnology 26, 3-30.
Hunter, P.A., Darby, G.K. and Russell, N.J. (1995) Fifty years of antimicrobials: past perspectives and future trends. Society for General Microbiology, Symposium 53. Cambridge. Cambridge University Press.
Hunter-Cevera, J.C. and Belt, A. (1994) Bacteria as a source of novel therapeutics. Biotechnology 26, 31-47.
Kuntz, I.D. (1992) Structure-based strategies for drug design and discovery. Science 257, 1078-1082.
Rossmann, M.G. and McKinlay, M.A. (1992) Application of crystallography to the design of antiviral agents. Infectious Agents and Disease 1, 3-10.
Russell, A.D. and Chopra, I. (1996) Understanding antibacterial action and resistance, 2nd edn. London: Chapman & Hall.
Vaara, M. (1992) Agents that increase the permeability of the outer membrane. Microbiological Reviews 56, 395-411.
Whittle, P.J. and Blundell, T.L. (1994) Protein structure-based drug design. Annual Review of Biophysics & Biomolecular Structure 23, 349-375.
Yarborough, G.G., Taylor, D.P., Rowlands, R.T., Crawford, M.S. and Lasure, L.L. (1993) Screening microbial metabolites for new drugs—theoretical and practical issues. Journal of Antibiotics 46, 535-544.
Was this article helpful?