Drug Discovery of Nucleos(t)ide Antiviral Agents: Dedicated to Prof. Dr. Erik De Clercq on Occasion of His 80th Birthday

Nucleoside and nucleotide analogues are essential antivirals in the treatment of infectious diseases such as human immunodeficiency virus (HIV), hepatitis B virus (HBV), hepatitis C virus (HCV), herpes simplex virus (HSV), varicella-zoster virus (VZV), and human cytomegalovirus (HCMV). To celebrate the 80th birthday of Prof. Dr. Erik De Clercq on 28 March 2021, this review provides an overview of his contributions to eight approved nucleos(t)ide drugs: (i) three adenosine nucleotide analogues, namely tenofovir disoproxil fumarate (Viread®) and tenofovir alafenamide (Vemlidy®) against HIV and HBV infections and adefovir dipivoxil (Hepsera®) against HBV infections; (ii) two thymidine nucleoside analogues, namely brivudine (Zostex®) against HSV-1 and VZV infections and stavudine (Zerit®) against HIV infections; (iii) two guanosine analogues, namely valacyclovir (Valtrex®, Zelitrex®) against HSV and VZV and rabacfosadine (Tanovea®-CA1) for the treatment of lymphoma in dogs; and (iv) one cytidine nucleotide analogue, namely cidofovir (Vistide®) for the treatment of HCMV retinitis in AIDS patients. Although adefovir dipivoxil, stavudine, and cidofovir are virtually discontinued for clinical use, tenofovir disoproxil fumarate and tenofovir alafenamide remain the most important antivirals against HIV and HBV infections worldwide. Overall, the broad-spectrum antiviral potential of nucleos(t)ide analogues supports their development to treat or prevent current and emerging infectious diseases worldwide.


Introduction
Nucleoside and nucleotide analogues are important antiviral, antiparasitic, and anticancer therapeutics that have been widely applied in clinical practice [1][2][3][4]. As shown in Figure 1a, nucleosides are formed by a sugar moiety and a nucleobase, while nucleotides are composed of a sugar moiety, a nucleobase, and at least one phosphate or phosphate-like group [4]. Nucleoside and nucleotide analogues with modified changes can mimic the structure of nature nucleosides and nucleotides that can be recognized by cellular or viral enzymes [5,6]. Many strategies (e.g., ring-opening, N-conjugation, halogenation) have been proposed to design novel nucleoside or nucleotide analogues, such as the acyclic fleximer analogues [7], propargylated purine deoxynucleosides [8], 4 -thionucleosides [9], 3 -fluoro-5 -norcarbocyclic nucleoside phosphonates [10], 4 -modified-2 -deoxy-2 -fluoro nucleosides [11], uracil-containing heterodimers [12], l-dideoxy bicyclic pyrimidine nucleoside analogues [13], and imidazo [4,5-b]pyridine nucleoside analogues [14]. In addition to the development of N-nucleoside analogues, C-nucleoside analogues could be considered because they are stably resistant against the phosphorolytic degradation caused by phosphorylases [15]. Of note, a N-nucleoside links its sugar moiety to its nucleobase through a nitrogen, while a C-nucleoside uses a carbon atom that substitutes the nitrogen to link the sugar moiety and the nucleobase (Figure 1b). As of today, Prof. Dr. Erik De Clercq has contributed to the approval of nine smallmolecule compounds that can be applied to treat viral infections and cancers such as multiple myeloma and non-Hodgkin's lymphoma (Table 1). Based on their chemical structures, these nine approved drugs can be arbitrarily classified into adenosine nucleotide analogues (tenofovir disoproxil fumarate, tenofovir alafenamide, adefovir dipivoxil), thymidine nucleoside analogues (stavudine, brivudine), guanosine analogues (valacyclovir, rabacfosadine), one cytidine nucleotide analogue (cidofovir), and one bicyclam derivative (plerixafor). Regarding their mechanisms of antiviral action, nucleoside and nucleotide analogues mimic the structure of a natural nucleoside that can be recognized by cellular and viral enzymes such as DNA and RNA polymerases, and due to their structural modifications, they also disrupt or terminate viral replications or biological processes [3,5]. Additionally, the CXCR4 inhibitor plerixafor blocks the interaction of CXCR4 with its natural ligand CXCL12, thereby mobilizing the CD34+ stem cells from the bone marrow into the peripheral bloodstream [16,17]. Because plerixafor was previously reviewed by our team and our beloved professor [16][17][18][19][20][21], this review will focus on nucleos(t)ide antiviral agents in four categories: (i) adenosine nucleos(t)ide analogues, (ii) thymidine nucleos(t)ide analogues, (iii) guanosine nucleos(t)ide analogues, and (iv) cytidine nucleos(t)ide analogues. To celebrate the 80th anniversary of our beloved professor Erik De Clercq on 28 March 2021, this review provides an overview of nucleoside and nucleotide analogues. We searched his publications between 1967 and 2020. The reference collection of research articles and related books was extracted from databases such as PubMed, Science Direct, and Google Scholar. The drug information was extracted from the US FDA website.
Research websites (www.virusface.com, www.erikdeclercq.org) were used to update the teaching lectures and publications of our beloved professor Erik De Clercq.
Truvada was approved by the US FDA in 2012 for the prevention of HIV infection, making it the first approved regimen for HIV prevention. In the iPrEx trial that enrolled HIV-seronegative men or transgender women who had sex with men, a 44% reduction of HIV incidences was observed in the Truvada group compared with that in the placebo group (36/1251 versus 64/1248) [32]. Similar to tenofovir disoproxil fumarate plus emtricitabine (Truvada ® ), tenofovir alafenamide plus emtricitabine (Descovy ® ) are also approved for HIV-1 prophylaxis by the US FDA.

Adefovir Dipivoxil
In 2002, adefovir dipivoxil 10 mg (Hepsera ® ) once daily orally with or without food was approved for HBV infections. Adefovir dipivoxil was initially designed to inhibit HIV, but its HIV application was abandoned because of its nephrotoxicity and its inferiority to tenofovir disoproxil fumarate [38]. Adefovir dipivoxil was subsequently pursued for HBV treatment, and its dosage of 10 mg/day was effective to treat HBeAg-positive and HBeAg-negative patients [39,40]. In randomized clinical trials, 48 weeks of adefovir dipivoxil 10 mg/day offered significant virologic, histologic, and biochemical improvement, while adefovir-associated resistance mutations were not identified in the HBV DNA polymerase [39,40]. However, a daily dose of adefovir dipivoxil 10 mg was inferior to a daily dose of tenofovir disoproxil fumarate 300 mg through 48 weeks [41]. Five-year treatment of tenofovir disoproxil fumarate also offered better safety and efficacy in HBV-infected patients [42]. According to the AASLD and EASL guidelines of HBV management, adefovir dipivoxil has been virtually replaced by tenofovir disoproxil fumarate and tenofovir alafenamide due to their better clinical efficacy and safety profiles [43,44].

Thymidine Nucleos(t)ide Analogues
As of today, there are at least seven thymidine nucleos(t)ide analogues approved for clinical use, including (i) brivudine (Zostex ® ) for HSV-1 and VZV treatment, (ii) stavudine (Zerit ® ) for HIV treatment, (iii) idoxuridine (Dendrid ® ) for HSV-1 treatment, (iv) trifluridine (Viroptic ® ) for HSV treatment, (v) zidovudine (Retrovir ® ) for HIV treatment, (vi) telbivudine (Tyzeka ® ) for HBV treatment, and (vii) sofosbuvir (Sovaldi ® ) for HCV treatment (Figures 2 and 3). As described by previous reviews [48,49], the career of Prof. Dr. Erik De Clercq began with interferon inducers and then shifted to nucleos(t)ide analogues such as stavudine and brivudine ( Figure 3). Here, we focus on brivudine and stavudine, while other compounds have been reviewed in our previous study [3].  and stavudine (c). A case of acute herpes zoster was treated by brivudine (d). In the first 3 days after symptom onset, a corticosteroid topical cream (mometasone furoate) offered no improvement. Based on the advice of our beloved professor Erik De Clercq, three once-daily tablets of brivudine (Zostex ® ) were subsequently used to cure the herpes zoster, and all crusts fell off 3 weeks later. Front and back views were taken before the brivudine treatment and 2 months after the treatment.

Brivudine
While brivudine ((E)-5-(2-bromovinyl)-2 -deoxyuridine) was discovered in 1976 at the University of Birmingham, its antiviral activities in cell cultures and clinical reports were first reported in 1979 by Erik De Clercq et al. [50]. After its phosphorylation by the viral thymidine kinase and nucleoside-diphosphate kinase, the brivudine 5 -triphosphate blocks the incorporation of viral DNA and inhibits the activity of viral DNA polymerases [3]. In both cell cultures and animal models, brivudine exerted a remarkable inhibitory effect on the replication of HSV-1 and VZV [50,51]. Moreover, the anti-VZV activity of brivudine was much more potent than that of (val)acyclovir, ganciclovir, and penciclovir [49,52].
After its rocky journey of clinical development [48], brivudine was approved in many countries (e.g., Germany, Belgium, Czech Republic, and Greece) to treat herpes zoster caused by VZV and HSV-1 (note that HSV-2 often causes genital herpes). In clinical practice, the once-daily tablets of brivudine should be administered as early as possible, preferably within 72 h from the first cutaneous manifestations such as red blistery patches. According to the label instructions, the safety and efficacy of brivudine in a 7-day course of therapy have been validated in adults, but its clinical effectiveness in the pediatric population (age 0 to 18 years) remains unclear. Figure 3d shows a case of herpes zoster treated by brivudine-as experienced firsthand by our team member who received the antiviral drugs contributed by Prof. Erik De Clercq.

Stavudine
Shortly after the anti-HIV discovery of zidovudine (AZT) in October 1985, stavudine (d4T) was discovered simultaneously at three locations: the Rega Institute, Yale University, and Yamamoto's Laboratory in Tokyo [49]. However, the anti-HIV activity of stavudine (2 ,3 -didehydro-2 ,3 -dideoxythymidine) was first described in 1987 by Masanori Baba et al. [53], who was, at that time, a favorite student of Prof. Erik De Clercq. Stavudine lacks the 3 -hydroxyl group (Figure 3c) which is indispensable for chain elongation. The incorporation of stavudine into nascent viral DNA causes the termination of the HIV transcription [54]. In 1994, stavudine was approved by the US FDA. However, a high level of drug resistance can be induced through the increased phosphorolytic excision of the incorporated monophosphate of stavudine [54]. Due to its off-target toxicity levels and drug resistance, stavudine was discontinued and removed from the market in 2020.

Valacyclovir
Valacyclovir (or valaciclovir) is the derivative of acyclovir with the valine ester (Figure 4a,b), which was co-discovered in 1983 by Leon Colla, Erik De Clercq, Roger Busson, and Hubert Vanderhaeghe at the Rega Institute [55]. This strategy was later applied to design valganciclovir by adding the valine ester to ganciclovir (Figure 4d,e). Before the advent of valacyclovir, acyclovir was the gold standard in the 1980s and was widely applied in the treatment of herpesvirus infections [56]. At the very beginning, Prof. Erik De Clercq thought the modifications of an existing compound to its prodrug were not very innovative [56], but valacyclovir with the aminoacyl ester surprisingly showed better oral absorption than acyclovir because of its increased aqueous solubility (51% to 54%) compared to the parent compound [55,57]. This also supports the use of oral applications of valacyclovir to treat cold sores (2 g every 12 h for 1 day) over acyclovir (5 times per day for 5 days).
In 1995, valacyclovir was approved by the US FDA for the treatment of cold sores, herpes zoster, and chickenpox. Compared with the oral acyclovir, valacyclovir enhances the bioavailability of acyclovir (10% to 20% according to dose), while acyclovir and valacyclovir share comparable safety profiles in the treatment of HSV infections [58]. Moreover, valacyclovir may play a promising role in the antenatal treatment of congenital cytomegalovirus [59] as well as the antiviral prophylaxis to prevent the late onset of HCMV infections in kidney-pancreas EBV-seronegative kidney recipients [60].

Cytidine Nucleos(t)ide Analogues
As of February 2021, at least four cytidine nucleos(t)ide analogues have been approved for clinical use, including (i) cidofovir (Vistide ® ) for treating HCMV retinitis in AIDS patients, (ii) zalcitabine (Hivid ® , which is now discontinued) for HIV treatment, (iii) emtricitabine (Emtriva ® ) for HIV treatment, and (iv) lamivudine (Epivir ® ) for HIV and HBV treatment ( Figure 5). This section focuses on cidofovir, a contribution of our beloved professor, while other drugs have been reviewed in our previous article [3]. Cidofovir is an acyclic nucleoside phosphonate (Figure 5b) licensed as an anti-DNA viral agent [66]. In 1987, cidofovir was first reported as an acyclic nucleoside derivative called (S)-HPMPC that effectively inhibited HCMV in human embryonic lung cells (minimum antiviral concentration: 0.08 µg/mL, selectivity index: 625) [61]. In 1996, the intravenous infusion of cidofovir was approved in the treatment of HCMV retinitis in AIDS patients-a severe complication that almost no longer exists thanks to the successful application of HAART treatments [67]. Its off-label use is primarily considered for treating DNA viruses such as Epstein-Barr virus (EBV), poxvirus infections (e.g., molluscum contagiosum), adenovirus infection, and human polyoma infections [67,68]. For instance, two patients with locally recurrent EBV-associated nasopharyngeal carcinoma were successfully treated by cidofovir [69].

Conclusions
Among the list of eight nucleos(t)ide analogues (Table 1) co-contributed by our beloved professor, adefovir dipivoxil, stavudine, and cidofovir are virtually discontinued for clinical use, whereas tenofovir disoproxil fumarate and tenofovir alafenamide remain the most important antivirals in the treatment of HIV and HBV infections as well as the prophylaxis of HIV infections. By saving millions of HIV-infected or HBV-infected patients worldwide, the popularity of tenofovir disoproxil fumarate and tenofovir alafenamide in clinical practice has proved the potential of nucleotide and nucleoside analogues.
Nucleoside and nucleotide antiviral agents are an important drug class with broadspectrum antiviral activities to treat current and emerging infectious diseases worldwide, thereby encouraging future development of novel nucleos(t)ide analogues. As a popular example of broad-spectrum antiviral activities, remdesivir inhibits many RNA viruses such as ebolavirus and respiratory pathogens including Middle East respiratory syndrome coronavirus, severe acute respiratory syndrome coronavirus (SARS-CoV), and SARS-CoV-2 [70]. In regards to the worldwide spread of COVID-19 [71][72][73], the US FDA approved remde-sivir (Veklury ® )-a monophosphoramidate prodrug of an adenosine analogue (Figure 1e) that acts as a direct-acting antiviral to treat COVID-19 [74]. Remdesivir, unlike other Nnucleoside analogues from Figures 1-5, is a C-nucleoside analogue that is stably resistant against phosphorolytic degradation caused by phosphorylases [15].
In addition to the discovery of broad-spectrum antivirals with better potency, it is important to develop highly selective nucleos(t)ide analogues with minimal toxicity because antiviral nucleos(t)ides may exert toxicity through the disruption of natural nucleoside triphosphate pools and interfere with activities of host polymerases such as human mitochondrial DNA/RNA polymerases (e.g., Pol γ, POLRMT) [75]. Future development of nucleos(t)ide analogues should focus on the reduced toxicity, the increased aqueous solubility, and the reduced risk of drug resistance.

Conflicts of Interest:
The authors declare no conflict of interest.