9-(trans-2′,trans-3′-dihydroxycyclopent-4′-enyl)-3-deazaadenine (DHCDA) is a neplanocin A analog and functions as an inhibitor of S-adenosylhomocysteine (AdoHcy) hydrolase, with broad-spectrum antiviral potential against vesicular stomatitis virus, vaccinia virus, parainfluenza virus, reovirus, and rotavirus [25]. Inspired by the observations that both _D-like and _L-like neplanocin derivatives possess potent antiviral activities, Chen lab has pioneered to synthesize a series of DHCDA analogues, all of which contain _L-type configuration and an adenine nucleobase but lack the 5′-hydroxyl group for nucleoside kinases to recognize [26].

The synthesized _L-neplanocin derivatives contained a cyclopentenol pseudo-sugar and an adenine nucleobase analogue, which were coupled via a Mitsunobu coupling reaction to provide the target compounds (Figure 1). The syntheses of halogenated derivatives, including bromonation at its nucleobase and 5′-fluorination/bromonation at sugar, were also accomplished. These _L-like analogues of natural carbocyclic nucleoside neplanocin A were evaluated as potential inhibitors against norovirus and Ebola viruses. Compounds 1a and 2b showed potent antiviral activity against norovirus (EC₅₀ = 4.2 µM and EC₅₀ = 3.0 µM, respectively). Compound 1b was found to possess effective antiviral activity against the Ebola virus (EC₅₀ = 8.3 µg/mL), while the analogues 2a and 2b were completely inactive. With the additional exocyclic derivatizations at the 5′-position, compound 5 displayed the potent activity against Pichinde (EC₅₀ = 0.9 µg/mL) and Tacaribe (EC₅₀ = 1.3 µg/mL). Both viruses are negative single-stranded RNA viruses belonging to arenaviridae, the same family as the Lassa fever virus. Compound 4a and compound 4b were also active against the Ebola virus. Although 4b showed no activity against those two arenaviruses, it is two-fold more active against the Ebola virus than compound 5.

It is noteworthy that the conformational behavior of “sugar” puckering (north/south) and nucleobase orientation (syn/anti) may contribute to the antiviral activity differences. The crystallographic studies revealed that the sugar in compound 1b adopted a 3′-exo conformation, while the more conformational, rigid bicyclic sugar in compound 2b would lock it into 2′-exo conformation. In addition, because of the steric hindrance, the 3-bromo substitution played an important role in anti-Ebola activity by forcing the nucleobase to adopt the less congested anti-conformation over the syn-conformation. No activity was found for single-stranded positive viruses, which suggests that _L-like neplanocin analogues are more effective on single-stranded negative-sense RNA viruses [27].

Immucillin-H (ImmH, also known as Forodesine) is a transition-state analog inhibitor of purine nucleoside phosphorylase (PNPases). It has been extensively studied for the treatment of patients with T-cell acute lymphoblastic leukemia (T-ALL), and its C-nucleoside hydrochloride form is in phase II clinical trials as an anti-T-cell leukemia agent [28]. Various Immucillin analogs modified at the 2′-, 3′-, or 5′-positions of the aza sugar moiety or at the 6-, 7-, or 8-positions of the deazapurine, have been synthesized and tested for their inhibition of human PNPases [29]. Inspired by the nucleoside-like structures of Immucillin analogues and their binding modes with PNPases by crystal structures [30,31], their _L-enantiomers have been investigated as novel pharmaceuticals against T-cell mediated disorders [32]. The synthetically achieved (1R)-1-(9-Deazahypoxanthin-9-yl)-1,4-dideoxy-1,4-imino-_L-ribitol (Figure 2, 6a) was an _L-enantiomer of natural _D-ImmH, and its hydrochloride complex was revealed to be a slow-onset tight-binding inhibitor of PNPases of human, bovine, and Plasmodium falciparum. Although this compound showed less activity than _D-ImmH when inhibiting the selected enzymes, it still demonstrated more excellent binding potency compared to 3′- and 5′-modified _D-ImmH. Moreover, the _L-enantiomer of second-generation Immucillin analogue, 4′-deaza-1′-aza-2′-deoxy-1′-(9-methylene)-Immucillin-H (DADMe-ImmH) [33], was also synthesized. The _L-DADMe-ImmH (Figure 2, 6b) also displayed lower activities as an inhibitor, when binding to the three enzymes. However, it was interesting to observe the sub-nanomolar binding capacities of these two _L-formed Immucillin analogues, plus they had the potential to be applied in different circumstances.

There have been extensive studies in search of chemotherapeutics to effectively cure pathogenic Human Immunodeficiency Virus (HIV) [34]. The dominant experimental and clinical attempts are focused on the development of modified nucleoside analogues, which bind to HIV reverse transcriptase and interfere with the synthesis of DNA copying of the viral genome [35]. Successful examples include small molecule drugs of AZT [36], ddI [37], ddC [38], and d4T [39] that have been clinically used to treat AIDS patients. The lack of 2′- and 3′-hydroxyl groups in nucleoside sugar can cause the termination of HIV reverse transcription and inhibit the viral life cycle. In particular, d4T has the double bond in its pseudosugar ring to rigidify the ring to planar conformation. In order to engender more potent inhibitor against HIV with great activity and less cytotoxicity, a number of _L-nucleoside have been reported, most of which bear the chemically derivatized pyrimidine and dideoxy _L-ribose.

Some β-_L-2′,3′-didehydro-2′,3′-dideoxythymidine (β-_L-d4T) analogues (Figure 3, 7a) have been synthesized, all bearing a tether on the C-5 position of the uracil ring, and they were evaluated in vitro for anti-HIV-1 activity [40]. The results revealed that the _L-d4T derivative, containing 12 methylene units at the 5-position, displayed some activity in the CEM-SS cells (IC₅₀ 2.3 µM), probably due to the more lipophilic nature of the nucleoside. Meanwhile, another innovative _L-d4T derivative, _L-MCd4T, containing the conformationally rigid methanocarba (MC) nucleoside, was synthesized, and its bicyclo[3.1.0]hexane moiety was restrained to North conformation and the 2′,3′-double bond further reduced the structural flexibility (Figure 3, 7b) [41]. _L-MCd4T was found to be a potent anti-HIV-1 inhibitor (EC₅₀ 6.76 µg/mL) without significant cytotoxicity, which is comparable to clinical drug ddI [42].

Additionally, some _L-enantiomers of 2′,3′-dideoxycytidine analogues were reported to selectively inhibit HIV in various cell cultures. For example, _L-2′,3′-dideoxycytidine, _L-2′,3′-dideoxy-5-fluorocytidine (_L-ddC and _L-FddC, Figure 3, 7c), _L-2′,3′-didehydro-2′,3′-dideoxycytidine, and _L-2′,3′-didehydro-2′,3′-dideoxy-5-fluorocytidine (_L-d4C and _L-Fd4C, Figure 3, 7d) were found to have impressive inhibitory activity but significantly less toxicity when treated in different cells, including rat glioma, lung carcinoma, lymphoblastoid, and skin melanoma cells [43,44,45]. The _L-enantiomers of 2′,3′-dideoxy-3′-thiacytidine and its 5-fluoro-derivative (_L-3TC and _L-FTC, Figure 3, 7e), when having a sulfur atom in place of the 3′-carbon, were discovered to be a potent inhibitor against HIV-1 in peripheral blood mononuclear cells and were also effective in thymidine kinase-deficient CEM cells. Meanwhile, nontoxicity was observed in human lymphocytes and other cell lines at up to 100 µM [46,47,48].

Various azanucleoside analogues have been pioneered, in which the natural carbone atoms in nucleobases are substituted with bioisosteric nitrogens, and these innovative compounds exhibited promising antitumore activity [49,50]. Inspired by this observation, Sartorelli et al. synthesized various dioxolane azanucleosides with _L-configuration (Figure 4, 8a–d) and bioevaluated them against HBV (Hepatitis B virus) [51]. The synthetic strategies involved the condensation of dioxolane derivative with silyl-protected azacytosine and azathymine. Interestingly, the in vitro HBV activity assay revealed that only (-)-(2S,4S)-1-[2-(hydroxy-methyl)-1,3-dioxolan-4-yl]-5-azacytosine (8a) possessed superior activity against HBV (EC₅₀ = 0.6 µM), whereas its _D-analogue was found inactive. No significant antiviral activity was observed in _L-azathymidine analogues.

The corresponding _L-4′-thionucleosides (Figure 5, 9a–d) were synthesized by starting from 1,2,3,5-tetra-O-acetyl-4-thio-β-L-ribofuranose [52]. All tested compounds showed a moderate growth inhibitory activity against HTB14 human glioma cells. Notably, compounds 9b and 9c exhibited a significant growth stimulatory activity towards NB4 and T47D cells at concentrations 0.78–1.56 µM.

It has been reported that a type of cytidine analogue, Cytarabine or Ara-C, can be used as an effective chemotherapy against acute myeloid leukemia, acute lymphocytic leukemia, chronic myelogenous leukemia, and non-Hodgkin’s lymphoma [53]. Cytarabine has the arabinoside sugar to mimic the native deoxycytidine, and it is rapidly phosphorylated into cytosine arabinoside triphosphate (Ara-CTP) to interfere with the DNA synthesis in the S phase of the cancer cell cycle [54]. However, the drug resistance of Ara-C is quite common, because the effective Ara-CTP can be easily deaminated to an inactive uridine metabolite by cytidine deaminase (CDA) [55]. To address this issue, an _L-nucleoside analogue of Ara-C, 5-fluorotroxacitabine (5FTRX) has been developed and activity tested in Acute Myeloid Leukemia (AML) cell lines [56].

5FTRX has an _L-configuration, containing a fluorinated cytosine nucleobase and dioxolane ring as sugars (Figure 6, 10). Experimental results suggested that 5-FTRX could also be phosphorylated to its 5′-triphosphate nucleotide, leading to significant DNA chain termination during DNA replication and cancer cell death. In addition, compared to Ara-C, 5FTRX was observed to overcome the drug resistance induced by CDA overexpression to some extent. Meanwhile, no signs of significant toxicity were displayed in a mouse experiment.

The LdT drug, also known as Telbivudine and invented by Novartis Inc., is an FDA-approved (in 2006) anti-viral drug used in the treatment of hepatitis B infection (Figure 7, 11) [57]. Telbivudine has the _L-converted structure of native thymidine, and it impairs the HBV DNA replication by leading to chain termination. Clinical trials have fully demonstrated its effect of viral suppression in patients and less viral resistance.

Lamivudine and Emtricitabine (commonly called 3TC and FTC, Figure 8, 12a,b) are _L-type cytidine analogues both containing oxathiolane rings as sugars. Emtricitabine has an additional Fluoro-modification at the 5-position of its cytosine nucleobase. Both compounds have been FDA approved for the treatment of human HIV and HBV infections, which can be administered individually or in combination with other inhibitors [58]. As the cytidine analogues, 3TC and FTC share the similar mechanism of action, by inhibiting the HIV reverse transcriptase and hepatitis B virus polymerase functions. The lack of 3′-OH groups prevents viral DNA elongation and terminates viral DNA growth. As the non-natural _L-nucleoside, both drugs were identified as less toxic agents in mitochondrial DNA [59].

Inspired by the success of Lamivudine and Emtricitabine and the functional AZT-containing 3′-azido group, the Schinazi lab has prepared various _L-3′-azido-2′,3′-dideoxypurine nucleosides (Figure 9, 13a,b), and evaluated their activity against HIV and HBV [60]. Eleven different _L-nucleosides were obtained through microwave-assisted optimized transglycosylation reactions. These _L-nucleoside analogues could be metabolized to corresponding nucleoside 5′-triphosphate compounds in primary human lymphocytes. Weak antiviral activities against HIV-1 and HBV were exhibited, even though no significant toxicity was observed.

The unique 5′-cap structure in mRNA is essential for effective binding to ribosome [61]. Interference with the formation of cap structure could inhibit the replication process and provide the potential strategy for viral treatment. The 5′-cap formation relies on the catalysis by methyltransferases, which has become a popular target for anti-viral drug design [62]. It has been found that many adenosine analogues displayed the interesting antiviral activity by inhibiting S-Adenosyl-_L-homocysteine (SAH) hydrolase, because SAH hydrolase is a key regulator of many S-adenosyl-_L-methionine (SAM) dependent biological methylation processes [63]. Various 5′-ethylenic and acetylenic substituted _L-adenosine derivatives were synthesized (Figure 10, 14a–e), and some of them showed modest inhibition of SAH hydrolase at 100 μM, when tested in the growth of HeLa cells or Bel-7420 cells [64].

Following the similar principle of _L-d4T and _L-ddC to treat HIV by inhibiting the viral reverse transcription with reduced toxicity, the Chu lab has developed a series of _L-nucleoside analogues containing a cyano group at the 3′-position (Figure 11, a–l). Some of the compounds also contained 2′,3′-unsaturated ribose [65]. The synthesized nucleosides were tested for anti-HIV activity in human PBM cells in vitro, and five of them (derivatives a, b, g, h and j) showed modest antiviral activity.

Ribavirin is a nucleoside analogue used to treat respiratory syncytia viral infection [66] and HCV infection [67]. It is reported to have the effects of inducing type 1 cytokine bias and enhancing the T cell-demiated immunity in vivo [68]. Structurally, Ribavirin is similar to nucleoside, which has the nucleobase replaced by 1,2,4-triazole-3-carboxamide. Ramasamy et al. have synthesized a series of _L-nucleoside analogues of Ribavirin and evaluated their activity of stimulating type 1 cytokine and enhancing T cell-demiated immunity [69]. One of the compounds prepared, which had 1,2,4-triazole-3-carboxamide (Figure 12, 17a) as its nucleobase, was found to be the most uniformly potent compound with interesting immunomodulatry potential.

Zika virus is a mosquito-born flavivirus that can cause the symptoms of fever, rash, joint pain, and red eyes [70]. Zika virus shares the same replication cycle as other flaviviruses, suggesting the potential of using nucleoside analogues as antiviral agents to terminate Zika virus DNA elongation. The Brancale lab has screened a targeted small molecule pool against Zika virus in vitro [71]. Several modified adenosine compounds have been identified to significantly inhibit the virus-induced cytopathic effect. Interestingly, one additional prodrug, 18a, which exhibited a _L-nucleoside conformation, was also screened out (Figure 13). This analogue contained _L-dideoxyribose sugar and a bicyclic pyrimidine as its nucleobase, and it had a potent synergistic effect of inhibiting the vaccinia and measles viruses when applied together with other adenosine phosphoramidate compounds.

Above, researchers listed some of the _L-nucleoside analogues with a solid demonstration of their therapeutic activities. Besides, there are many other nucleoside-like small molecules designed and evaluated as antiviral therapeutics, which had an _L-configuration similar to its modified nucleoside, including cyclobutene _L-nucleoside analogues [72], _L-erythro-hexopyranosyl nucleosides [73,74], _L-4′-C-ethynyl-2′-deoxypurine nucleosides [75], _L-ribo-configured Locked Nucleic Acid [76,77,78,79], pyrrolo, pyrazolo, or imidazo-modified _L-nucleoside [80], et al. There have also been many methods developed to efficiently synthesize carbocyclic _L-nucleoside analogues [52,81,82,83], and they have been summarized elsewhere [84,85].

In the future development of _L-nucleoside as therapeutic molecules, one direction is to design a broad-spectrum antiviral drug essential for rapid and efficient disease treatment. The recent viral outbreaks in the past decade have urged this need. Generally, the development of _L-nucleosides with broad-spectrum antiviral activities is more challenging because of the different behaviors among viruses, especially after their infections to the host. To design a broad-spectrum antiviral nucleoside, a comprehensive investigation is needed to discover the biological features of multiple viruses and design chemically modified _L-nucleosides to target these features. In addition, the combinatory treatment using different nucleoside drugs, or a nucleoside with another biological agent, will be necessary to decrease the drug resistance.