Broad-Spectrum In Vitro Antiviral Activity of ODBG-P-RVn: An Orally-Available, Lipid-Modified Monophosphate Prodrug of Remdesivir Parent Nucleoside (GS-441524)

ABSTRACT

The necessity for intravenous administration of remdesivir confines its utility for treatment of coronavirus disease 2019 (COVID-19) to hospitalized patients. We evaluated the broad-spectrum antiviral activity of ODBG-P-RVn, an orally available, lipid-modified monophosphate prodrug of the remdesivir parent nucleoside (GS-441524), against viruses that cause diseases of human public health concern, including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). ODBG-P-RVn showed 20-fold greater antiviral activity than GS-441524 and had activity nearly equivalent to that of remdesivir in primary-like human small airway epithelial cells. Our results warrant in vivo efficacy evaluation of ODBG-P-RVn.

IMPORTANCE While remdesivir remains one of the few drugs approved by the FDA to treat coronavirus disease 2019 (COVID-19), its intravenous route of administration limits its use to hospital settings. Optimizing the stability and absorption of remdesivir may lead to a more accessible and clinically potent therapeutic. Here, we describe an orally available lipid-modified version of remdesivir with activity nearly equivalent to that of remdesivir against emerging viruses that cause significant disease, including Ebola and Nipah viruses. Our work highlights the importance of such modifications to optimize drug delivery to relevant and appropriate human tissues that are most affected by such diseases.

OBSERVATION

Remdesivir (RDV; Veklury, GS-5734) is an adenosine nucleotide analog phosphoramidate prodrug with broad-spectrum antiviral activity in vitro and in vivo (1) and is currently the only FDA-approved therapeutic for treating coronavirus 2019 disease (COVID-19) in hospitalized patients over the age of 12 (2). While RDV did not significantly reduce COVID-19 mortality, it shortened the time to recovery compared to the time for placebo controls (3). The short half-life of RDV in human and animal plasma (4–7), alongside the in vivo efficacy of the RDV parent nucleoside (RVn; GS-441524) against coronaviruses, including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (8–11), have driven proposals to utilize the RVn instead of RDV to treat COVID-19 (12). A recent comparative pharmacokinetic study in nonhuman primates, however, demonstrated higher levels of the active metabolite RVn-triphosphate (RVn-TP) in lower respiratory tract tissues of RDV-dosed animals than in RVn-dosed animals (6). A significant drawback of RDV is the requirement for intravenous administration, which limits its use to hospital contexts. To develop an orally bioavailable form of remdesivir, we recently synthesized a 1-O-octadecyl-2-O-benzyl-sn-glycerylester (ODBG) lipid-modified monophosphate prodrug of RVn (ODBG-P-RVn) (C₄₀H₆₂N₅O₉P) (Fig. 1A), which demonstrated more favorable in vitro antiviral activity than RVn and RDV against SARS-CoV-2 in Vero-E6 cells (13).

FIG 1.

Comparison of antiviral activities of RVn, RDV, and ODBG-P-RVn in African green monkey (Vero-E6), human hepatoma (Huh7), human bronchioalveolar carcinoma (NCI-H358), and primary-like human telomerase reverse transcriptase-immortalized small airway epithelial (HSAEC1-KT) cell lines using fluorescent-reporter-based, image-based, and cytopathic effect (CPE) assays. Values for representative dose-response inhibition of viral replication and induction of cellular cytotoxicity by RVn, RDV, and ODBG-P-RVn are shown. (A) The chemical structure of ODBG-P-RVn. (B) Direct measurement of reporter fluorescence intensities from recombinant Ebola virus (EBOV) expressing ZsGreen protein in Vero-E6 (left) and Huh7 (middle left) cells and recombinant Nipah virus (NiV) expressing ZsGreen protein in NCI-H358 (middle right) and HSAEC1-KT (right) cells. (C) Image-based counting of reporter fluorescence-positive cells infected with recombinant severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) expressing mNeonGreen protein (Vero-E6 and Huh7 cells) and recombinant respiratory syncytial virus (RSV) expressing enhanced green fluorescent protein (eGFP) (NCI-H358 and HSAEC1-KT cells). Values for infected cells treated with dimethyl sulfoxide (DMSO) were considered the 100% fluorescence intensity signals and 100% fluorescence-positive cell counts. (D) Compound-based inhibition of CPE induced by yellow fever virus (YFV) (Vero-E6 and Huh7 cells) and by Hendra virus (HeV) (NCI-H358 and HSAEC1-KT cells) determined by measuring cellular ATP levels (CellTiterGlo 2.0). ATP levels in uninfected cells treated with DMSO were considered 100% CPE inhibition. (E) Compound cytotoxicity/cell viability for the respective cell lines used measured by CellTiterGlo 2.0 assay. (F) Measurement of RVn-triphosphate (RVn-TP) levels in Vero E6 cells treated with RVn, RDV, or ODBG-P-RVn at various time points until 48 h posttreatment. (G) Reductions of infectious yields of EBOV-ZsG (left) and NiV-ZsG (right) by RDV and ODBG-P-RVn in HSAEC1-KT cells. y axis denotes 50% tissue culture infectious dose (TCID₅₀) expressed in logarithmic scale. Dose-response curves for antiviral assays in panels B, C, and D were fitted to the mean values of experiments performed in biological triplicate for each concentration in the 8-point, 3-fold dilution series using a 4-parameter nonlinear logistic regression curve with variable slope. Data points and error bars indicate the mean values and standard deviations from 3 biological replicates; each colored shape/line in the legend represents an independent experiment performed in biological triplicate as indicated above panel B. Infectious yield reduction assays were conducted once in biological quadruplicates. RVn and RDV used in this study were obtained from MedChemExpress (Monmouth Junction, NJ, USA).

In this study, we extended our in vitro comparisons to include 14 viruses from 7 virus families responsible for causing diseases of significant human public health concern. These were Ebola virus (EBOV) and Marburg virus (MARV) from the family Filoviridae, Nipah virus (NiV), Hendra virus (HeV), human parainfluenza virus 3 (hPIV3), measles virus (MV), mumps virus (MuV), and Sosuga virus (SoSuV) from the family Paramyxoviridae, respiratory syncytial virus (RSV) from the family Pneumoviridae, yellow fever virus (YFV) from the family Flaviviridae, Lassa virus (LASV) from the family Arenaviridae, Crimean-Congo hemorrhagic fever virus (CCHFV) from the family Nairoviridae, and SARS-CoV-2 from the family Coronaviridae (14–18). We utilized the following three previously described assays to compare the antiviral activities of RVn, RDV, and ODBG-P-RVn against this panel of viruses (15, 17): (i) directly measuring the fluorescence of a reporter protein expressed by recombinant viruses (REP) (Fig. 1B), (ii) quantitating focus-forming units (FFU) via fluorescent-reporter imaging (Fig. 1C), and (iii) indirectly measuring cytopathic effect (CPE) based on cellular ATP levels (CellTiterGlo 2.0; Promega) (Fig. 1D), which was also used to evaluate compound cytotoxicity (Fig. 1E). Assay conditions varied based on virus replication kinetics and on the specific assay used; the multiplicities of infection (MOIs) ranged from 0.01 to 0.25, and endpoint measurements were conducted between 72 and 144 h postinfection (hpi) (see Methods in the supplemental material). We conducted dose-response experiments using 8-point, 3-fold serial dilutions of RVn, RDV, and ODBG-P-RVn against our panel of viruses in Vero-E6 cells and showed that ODBG-P-RVn consistently had greater antiviral activity than RVn and RDV against all viruses susceptible to RVn/RDV inhibition, with 50% effective concentration (EC₅₀) values ranging from 0.026 to 1.13 µM (Fig. 1B to D, left; Table 1; Fig. S1 in the supplemental material). RVn and ODBG-P-RVn induced partial cytotoxicity but only at the highest concentration tested (100 µM) and without reaching 50% cytotoxic concentration (CC₅₀). To understand the comparatively greater potency of ODBG-P-RVn in Vero E6 cells, we measured the levels of RVn-TP in cells treated with RVn, RDV, or ODBG-P-RVn. We observed that RVn-TP levels indeed correlated with antiviral activity, with ODBG-P-RVn consistently accumulating to higher levels than both RVn and RDV across 3 time points (Fig. 1F). We then compared these antivirals in human hepatoma (Huh7) and bronchioalveolar carcinoma (NCI-H358) cell lines, which represent more-relevant cell types that are targeted by subsets of viruses used in our study. In both human cell lines, although ODBG-P-RVn showed EC₅₀ values comparable to those observed in Vero-E6 cells and was 3- to 5-fold more active than RVn, it consistently showed 6- to 20-fold less activity than RDV (Fig. 1B to D, middle left and middle right; Table 1; Fig. S2 and S3). Whereas the CC₅₀ values for RDV in Huh7 and NCI-H358 cells were 54.2 and 77.2 µM, respectively, ODBG-P-RVn was less cytotoxic in Huh7 cells (CC₅₀ = 93.4 µM) and did not show measurable cytotoxicity in NCI-H358 cells even at the highest concentration tested (100 µM) (Fig. 1E, middle right; Table 1).

TABLE 1.

Mean antiviral activities of RVn, RDV, and ODBG-P-RVn in Vero E6, Huh7, and NCI-H358 cell lines

Virus family	Virusa	Species/variantb	Assayc	Mean value (µM) ± SD for indicated antiviral in indicated cellsd
				Vero E6									Huh7/NCI-H358e
				RVn (GS-441524)			RDV (GS-5734)			ODBG-P-RVn			RVn (GS-441524)			RDV (GS-5734)			ODBG-P-RVn
				EC50	EC90	SI (CC50 > 100 µM)	EC50	EC90	SI (CC50 > 100 µM)	EC50	EC90	SI (CC50 > 100 µM)	EC50	EC90	SI (CC50 > 100 µM/>100)	EC50	EC90	SI (CC50: 54.2 ± 6.0/77.2 ± 5.3)	EC50	EC90	SI (CC50: 93.4 ± 3.0/>100)
Filoviridae	EBOV	Rec. Makona-ZsG	REP	2.03 ± 0.50	7.54 ± 1.09	49	5.15 ± 1.09	17.31 ± 0.89	>19	0.39 ± 0.10	1.71 ± 0.25	>258	1.84 ± 0.31	6.91 ± 1.79	>54	0.020 ± 0.003	0.16 ± 0.02	2,710	0.37 ± 0.06	2.13 ± 0.37	251
	MARV	Rec. Bat371-ZsG	REP	0.96 ± 0.09	4.05 ± 1.42	104	2.16 ± 0.27	10.22 ± 2.02	>46	0.19 ± 0.04	0.81 ± 0.12	>521	1.92 ± 0.06	4.47 ± 0.48	>52	0.025 ± 0.002	0.075 ± 0.003	2,128	0.33 ± 0.02	0.99 ± 0.09	285
Paramyxoviridae	NiV-M	Rec. Malaysia-ZsG	REP	1.10 ± 0.40	2.20 ± 1.05	73	5.87 ± 0.19	9.82 ± 0.43	>16	0.31 ± 0.04	0.78 ± 0.28	>196	2.43 ± 0.31	5.95 ± 1.10	>41	0.075 ± 0.001	0.31 ± 0.04	1,026	0.50 ± 0.06	2.83 ± 1.39	>198
			CPE	0.48 ± 0.06	0.78 ± 0.19	207	3.34 ± 0.34	5.39 ± 0.29	>30	0.19 ± 0.01	0.30 ± 0.04	>522	ND	ND	NA	ND	ND	NA	ND	ND	NA
	NiV-B	Bangladesh	CPE	0.52 ± 0.02	1.14 ± 0.02	192	2.84 ± 0.10	5.81 ± 0.44	>35	0.17 ± 0.01	0.38 ± 0.04	>599	3.42 ± 0.005	5.41 ± 0.29	>29	0.12 ± 0.0004	0.19 ± 0.01	661	0.82 ± 0.053	1.38 ± 0.05	>122
	HeV	1996	CPE	1.43 ± 0.17	12.06 ± 3.14	70	4.56 ± 0.20	17.58 ± 3.91	>22	0.37 ± 0.04	3.93 ± 1.98	>270	3.68 ± 0.08	6.33 ± 0.18	>27	0.16 ± 0.02	0.25 ± 0.03	491	0.95 ± 0.12	1.42 ± 0.03	>105
	MV	Rec. rMVEZGFP(3)	REP	0.58 ± 0.20	1.71 ± 0.07	172	4.97 ± 0.25	6.12 ± 0.3	>20	0.16 ± 0.03	0.21 ± 0.01	>609	0.88 ± 0.16	6.99 ± 1.90	>113	0.025 ± 0.007	0.13 ± 0.09	3,074	0.12 ± 0.003	0.86 ± 0.22	>803
	hPIV3	Rec. JS-GFP	FFU	0.14 ± 0.01	0.28 ± 0.02	70	0.43 ± 0.09	0.90 ± 0.03	>232	0.026 ± 0.002	0.050 ± 0.002	>3,896	1.43 ± 0.16	1.98 ± 0.05	>70	0.031 ± 0.002	0.052 ± 0.01	2,458	0.22 ± 0.01	0.43 ± 0.02	>457
	MuV	Rec. IA2006-eGFP	FFU	5.11 ± 0.20	7.80 ± 0.64	18	16.81 ± 1.23	25.1 ± 1.97	>4.9	1.13 ± 0.04	2.53 ± 0.25	>56	9.3 ± 0.30	13.71 ± 0.24	>11	0.20 ± 0.003	0.24 ± 0.003	266	1.85 ± 0.11	2.24 ± 0.23	50
	SoSuV	Rec. 2012-ZsG	REP	1.00 ± 0.10	2.72 ± 0.62	100	5.31 ± 1.8	19.10 ± 9.31	>19	0.31 ± 0.089	0.80 ± 0.06	>325	2.06 ± 0.09	7.76 ± 1.11	>48	0.052 ± 0.01	0.13 ± 0.02	1,042	0.52 ± 0.10	1.08 ± 0.15	180
Pneumoviridae	RSV	Rec. rgRSV0224 (A2)	FFU	0.49 ± 0.05	0.62 ± 0.01	206	1.80 ± 0.08	2.40 ± 0.27	>55	0.10 ± 0.02	0.22 ± 0.03	>997	1.93 ± 0.02	2.36 ± 0.08	>51	0.078 ± 0.004	0.17 ± 0.02	991	0.55 ± 0.057	1.41 ± 0.09	>180
Coronaviridae	SARS-CoV-2	Rec. icSARS-CoV-2 mNG (WA1)	FFU	0.42 ± 0.09	0.60 ± 0.06	236	1.77 ± 0.13	2.81 ± 0.78	>56	0.10 ± 0.005	0.16 ± 0.01	>997	0.69 ± 0.01	1.50 ± 0.20	>144	0.011 ± 0.001	0.035 ± 0.002	5,073	0.12 ± 0.02	0.69 ± 0.07	778
Flaviviridae	YFV	17D	CPE	3.52 ± 0.24	30.25 ± 10.08	28	19.86 ± 1.73	>50	>5	0.87 ± 0.043	7.37 ± 1.59	>114	36.83 ± 2.85	>50	>2.7	0.88 ± 0.057	3.09 ± 1.47	62	14.11 ± 0.90	>50	6.6
Arenaviridae	LASV	Rec. Josiah-ZsG	REP	NI	NI	NA	NI	NI	NA	31.14 ± 7.79	>50	>3	NI	NI	NA	2.87 ± 0.61	5.17 ± 0.33	19	NI	NI	NA
Nairoviridae	CCHFV	Rec. IbAr10200-ZsG	REP	NI	NI	NA	NI	NI	NA	NI	NI	NA	NI	NI	NA	NI	NI	NA	NI	NI	NA

^aEBOV, Ebola virus; MARV, Marburg virus; NiV-M, Nipah virus Malaysia strain; NiV-B, Nipah virus Bangladesh strain; HeV, Hendra virus; MV, measles virus; hPIV3, human parainfluenza virus 3; MuV, mumps virus; SoSuV, Sosuga virus; RSV, respiratory syncytial virus; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; YFV, yellow fever virus; LASV, Lassa virus; CCHFV, Crimean-Congo hemorrhagic fever virus.

^bRec., recombinant; ZsG, ZsGreen fluorescent protein; GFP, green fluorescent protein; eGFP, enhanced GFP; mNG, mNeonGreen.

^cREP, CPE, and FFU assays were conducted between 72 and 144 hpi. REP, fluorescent reporter; CPE, cytopathic effect; FFU, focus-forming units.

^dValues were derived from 3 independent experiments performed in biological triplicates, except for assays of NiV-B (NCI-H358 cells), HeV (NCI-H358 cells), and YFV (Vero E6 cells), which were performed twice in biological triplicates. EC₅₀, EC₉₀, and CC₅₀ values were calculated using GraphPad Prism 9 software. EC₅₀ and EC₉₀, 50% and 90% effective concentrations; CC₅₀, 50% cytotoxic concentration; SI, selective index (EC₅₀/CC₅₀); ND, not determined; NI, no inhibition; NA, not applicable. The CC₅₀ values for each compound in the respective cell lines are indicated in parentheses above the column indicated for SI values.

^eData in boldface were derived from Huh7 cells, and underlined data were derived from NCI-H358 cells.

To further evaluate cell type-specific effects on the antiviral activities of RVn, RDV, and ODBG-P-RVn, we tested them against a smaller subset of reporter viruses in primary-like human telomerase reverse transcriptase (hTERT)-immortalized human microvascular endothelial (TIME) cells (19, 20). In TIME cells, we observed a trend in antiviral activity similar to those in Huh7 and NCI-H358 cells, with ODBG-P-RVn showing 15- to 22-fold greater activity than RVn but 5- to 8-fold less activity than RDV in reporter-based assays (Table 2; Fig. S4A). We further compared the activities of RDV and ODBG-P-RVn by infectious yield assay and observed that both compounds equivalently reduced the infectious yield of EBOV expressing ZsGreen protein (EBOV-ZsG), by up to 4 log₁₀, and that of of NiV-ZsG, by approximately 2 log₁₀, in a dose-dependent manner, with EC₅₀ values closely mirroring the values determined in reporter assays (Table 2; Fig. S4B). However, RDV was more cytotoxic (CC₅₀ = 17.2 µM) than ODBG-P-RVn (CC₅₀ > 50 µM), which is reflected in its biphasic inhibition of NiV-ZsG, with cytotoxic inhibition by RDV shown at 16.6 µM (Fig. S4C).

TABLE 2.

Mean antiviral activities of RVn, RDV, and ODBG-P-RVn in TIME and HSAEC1-KT cell lines

Virus family	Virusa	Species/variantb	Assayc	Mean value (µM) ± SD for indicated antiviral in indicated cellsd
				HSAEC1-KT									TIME
				RVn (GS-441524)			RDV (GS-5734)			ODBG-P-RVn			RVn (GS-441524)			RDV (GS-5734)			ODBG-P-RVn
				EC50	EC90	SI (CC50 > 100 µM)	EC50	EC90	SI (CC50 > 100 µM)	EC50	EC90	SI (CC50 = 20.5 ± 0.29 µM)	EC50	EC90	SI (CC50 > 100 µM)	EC50	EC90	SI (CC50 = 17.2 ± 0.42 µM)	EC50	EC90	SI (CC50 > 50 µM)
Filoviridae	EBOV	Rec. Makona-ZsG	REP	10.7 ± 2.62	21.79 ± 3.16	>9.3	0.17 ± 0.02	0.41 ± 0.14	>587	0.21 ± 0.02	1.06 ± 0.18	98	14.88 ± 0.28	17.24 ± 0.16	>3.36	0.13 ± 0.04	0.2 ± 0.01	132	0.99 ± 0.063	1.96 ± 0.043	>50
			VTR	ND	ND	NA	0.11	0.82	>909	0.21	0.95	98	ND	ND	NA	0.032	0.064	530	0.15	0.39	>324
	MARV	Rec. Bat371-ZsG	REP/FFU	35.53 ± 7.07	71.35 ± 1.28	>2.8	0.75 ± 0.19	2.92 ± 0.14	>133	0.71 ± 0.11	3.67 ± 0.49	29	5.2 ± 0.26	6.89 ± 0.86	>9.61	0.04 ± 0.003	0.086 ± 0.004	430	0.23 ± 0.036	0.66 ± 0.032	>213
Paramyxoviridae	NiV-M	Rec. Malaysia-ZsG	REP	16.46 ± 0.04	19.12 ± 0.05	>6.1	0.23 ± 0.01	0.31 ± 0.06	>440	0.57 ± 0.013	0.97 ± 0.21	36	13.53 ± 2.44	17.52 ± 0.77	>3.70	0.10 ± 0.01	0.20 ± 0.01	172	0.75 ± 0.05	2.01 ± 0.30	>66
			CPE	16.12 ± 4.21	78.1 ± 35.08	>6.2	0.31 ± 0.04	0.075 ± 0.004	>318	0.90 ± 0.07	10.22 ± 4.99	23	ND	ND	NA	ND	ND	NA	ND	ND	NA
			VTR	ND	ND	NA	0.26	0.36	>379	0.47	0.77	44	ND	ND	NA	0.054	0.07	319	0.26	0.77	>195
	NiV-B	Bangladesh	CPE	11.23 ± 0.63	33.6 ± 1.58	>8.9	0.21 ± 0.063	0.62 ± 0.20	>379	0.41 ± 0.039	1.71 ± 0.66	50
	HeV	1994	CPE	11.52 ± 1.49	26.11 ± 4.44	>8.7	0.22 ± 0.04	0.65 ± 0.11	>463	0.42 ± 0.023	1.19 ± 0.061	49
	MV	Rec. rMVEZGFP(3)	REP	4.98 ± 0.37	12.02 ± 2.7	>20	0.063 ± 0.02	0.128 ± 0.016	>1,587	0.082 ± 0.026	0.29 ± 0.043	251
	hPIV3	Rec. JS-GFP	FFU	4.96 ± 0.05	5.77 ± 0.06	>20	0.063 ± 0.001	0.074 ± 0.002	>1,582	0.091 ± 0.009	0.20 ± 0.008	226
Pneumoviridae	RSV	Rec. rgRSV0224 (A2)	FFU	4.92 ± 0.47	8.09 ± 0.68	>20	0.088 ± 0.026	0.21 ± 0.033	>,1134	0.12 ± 0.008	0.34 ± 0.047	176

^aEBOV, Ebola virus; MARV, Marburg virus; NiV-M, Nipah virus Malaysia strain; NiV-B, Nipah virus Bangladesh strain; HeV, Hendra virus; MV, measles virus; hPIV3, human parainfluenza virus 3; RSV, respiratory syncytial virus.

^bRec., recombinant; ZsG, ZsGreen fluorescent protein; GFP, green fluorescent protein.

^cREP, FFU, CPE, and VTR assays were conducted at 72 hpi. REP, fluorescent reporter; VTR, virus titer reduction; CPE, cytopathic effect; FFU, focus-forming units.

^dValues were derived from 3 independent experiments performed in biological triplicates. TIME, primary-like human telomerase reverse-transcriptase (hTERT)-immortalized human microvascular endothelial cell line; HSAEC1-KT, hTERT-immortalized small airway epithelial cell line; EC₅₀, EC₉₀, and CC₅₀ values were calculated using GraphPad Prism 9 software. EC₅₀ and EC₉₀, 50% and 90% effective concentrations; CC₅₀, 50% cytotoxic concentration; SI, selective index (EC₅₀/CC₅₀); ND, not determined; NI, no inhibition; NA, not applicable. The CC₅₀ values for each compound in the respective cell lines are indicated in parentheses above the column indicated for SI values.

Since ODBG lipid nucleoside modification enhances in vivo lung tissue distribution via the chylomicron pathway (21, 22), we compared the activities of the three compounds against filoviruses, paramyxoviruses, and RSV in another primary-like, hTERT-immortalized small airway epithelial cell line (HSAEC1-KT) (23). Notably, the dose-response curves of RDV and ODBG-P-RVn were strikingly similar, with EC₅₀ values in the submicromolar range within a 3-fold range of each other; the EC₅₀ values for some viruses were almost identical (Fig. 1B to D, right; Table 2; Fig. S5). Furthermore, RDV and ODBG-P-RVn reduced the infectious yields of EBOV-ZsG and NiV-ZsG in HSAEC1-KT cells equivalently, by 5 log₁₀ and 3 log₁₀, respectively, and their EC₅₀ values reflected the limited differential in antiviral activities between them (Fig. 1G; Table 2). Although ODBG-P-RVn was more cytotoxic (CC₅₀ = 20.5) than RDV (CC₅₀ > 100) in HSAEC1-KT cells (Fig. 1D, right; Table 2), it also effectively reduced the virus yields at noncytotoxic concentrations. We also evaluated the antiviral activity of the ODBG lipid alone and observed no detectable antiviral activity against any of the viruses tested in HSAEC1-KT cells (data not shown).

Our results demonstrate that ODBG-P-RVn has greater antiviral activity than RVn and has cell type-dependent activity levels that range from moderately lower than to nearly equal to those of RDV. In vivo, RDV is converted rapidly to RVn (4–7), which has 0.5 to 2 log₁₀ less activity than RDV against most of the viruses tested. In contrast, ODBG-P-RVn was stable in plasma for >24 h and reached therapeutic plasma levels (above the EC₉₀ for SARS-CoV-2) after oral administration of 16.9 mg/kg of body weight to Syrian hamsters; it also did not produce virologically significant levels of RVn (13). Thus, one would predict sustained in vivo antiviral activity with ODBG-P-RVn, without substantial generation of RVn, the less active metabolite, in plasma. Taken together, our results support further optimization of ODBG-P-RVn and future in vivo evaluation of such monophosphate lipid-modified analogs of RVn for their efficacy against viruses significant to human health.