Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, March 2006, p. 984-993, Vol. 50, No. 3
0066-4804/06/$08.00+0 doi:10.1128/AAC.50.3.984-993.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Kelly L. Warfield,2,
Dana L. Swenson,2
David A. Stein,3
Jeffery L. Smith,2
C. Scott Gamble,2
Andrew D. Kroeker,3
Patrick L. Iversen,3
Sina Bavari,2* and
Elke Mühlberger1*
Department of Virology, Philipps-University Marburg, Hans-Meerwein-Str. 3, 35043 Marburg, Germany,1 U.S. Army Medical Research Institute of Infectious Diseases, Frederick, Maryland 21702-5011,2 AVI BioPharma Inc., Corvallis, Oregon3
Received 29 June 2005/ Returned for modification 5 August 2005/ Accepted 3 January 2006
|
|
|---|
|
|
|---|
Most MARV and EBOV species cause a severe hemorrhagic fever associated with fatality rates up to 90% in humans and nonhuman primates (8, 32). Due to the high mortality rates and the lack of any approved effective human therapeutic or vaccine (5), filoviruses have been classified as biological safety level 4 (BSL-4) agents. Many approaches have been employed in attempts to develop effective therapies for EBOV, including the administration of nucleoside analogues (i.e., ribavirin), immune globulins, type I interferons and other cytokines, anticoagulants, and therapeutic vaccines (10, 13, 19, 20, 24, 37, 39, 40). To date, the highest survival rate (33%) observed in EBOV-infected nonhuman primates was obtained by the administration of an anticoagulant molecule, recombinant inhibitor of factor VIIa/tissue factor (11). Collectively, previous work suggests that treatment of disease symptoms or processes may increase the likelihood of survival. However, an approach that directly inhibits EBOV replication as a way to more effectively treat filovirus infections seems highly desirable.
Phosphorodiamidate morpholino oligomers (PMO) are a class of single-stranded antisense agents that have the same purine and pyrimidine bases as those of DNA but possess a backbone modified to contain a phosphorodiamidate linkage and morpholine ring (36). The molecular character of the PMO backbone results in near-complete resistance to nucleases (18) yet allows maintenance of aqueous solubility. PMO function by Watson-Crick base pairing and are most frequently designed to target mRNAs by complementary base pairing with the translation start site region, thus forming a steric blockade (reviewed in reference 15). To enhance uptake into cells, PMO have been conjugated to cell-penetrating arginine-rich peptides (6, 26). Both nonconjugated PMO and peptide-PMO (P-PMO) have been shown to successfully inhibit replication of several viruses in cell culture, including Vesivirus (35), mouse hepatitis virus (31), severe acute respiratory syndrome coronavirus (30), and several flaviviruses (6, 21).
For this study, PMO and P-PMO were designed to target six sequences in the genomic or antigenomic RNA of the EBOV species Zaire. We sought to select the most efficacious of the oligomers, based on their in vitro activities, for further evaluation in a mouse model of EBOV infection. Compared to other sequences, a VP35-specific P-PMO was able to inhibit EBOV replication in cell culture and greatly increase the survival of EBOV-infected mice when administered either prophylactically or therapeutically. A corresponding nonconjugated PMO was also effective as a prophylactic in mice. These data suggest that PMO compounds can specifically interfere with EBOV replication in vivo and thus hold interest for further therapeutic development.
|
|
|---|
1,000 PFU of a mouse-adapted strain of EBOV (4). Design and synthesis of PMO and P-PMO. PMO compounds were designed as sequences complementary to Zaire EBOV (GenBank accession number AF086833) and named for their target nucleotides within the EBOV genome or antigenome (Table 1). All PMO were synthesized and purified by AVI BioPharma Inc. (Corvallis, OR), as previously described (36). A set of PMO were covalently conjugated at the 5' end to an arginine-rich peptide, NH2-RRRRRRRRRFFC-CONH2, designated as R9F2. The conjugation, purification, and analysis of R9F2-PMO compounds were performed as previously described (26). In this report, peptide-conjugated PMO are referred to as P-PMO and were used in all of the cell culture experiments. Both P-PMO and nonconjugated PMO were used in the in vitro translation and animal experiments and are duly indicated. A "nonsense" sequence (named "scramble") was prepared as a PMO and as a P-PMO to serve as a control for off-target effects of the respective chemistries.
|
View this table: [in a new window] |
TABLE 1. PMO compounds and target locations in EBOV RNA
|
TCID50 assay. Vero cells were grown in 96-well plates to 30 to 40% confluence. The cells were then inoculated with 10-fold serial dilutions of supernatant from P-PMO-treated and infected cells. At 10 to 12 days postinfection (dpi), when the cytopathic effect (CPE) had stabilized to a constant rate, the dilutions showing CPE were evaluated by light microscopy. The TCID50/ml was calculated using the Spearman-Kärber method (17).
MTT cytotoxicity assay. Vero cells were grown in 96-well plates to 60 to 70% confluence in normal growth medium supplemented with 10% fetal bovine serum. The cells were then washed with phosphate-buffered saline (PBS) and incubated with serum-free medium containing water or PMO compounds at the indicated concentrations. After 24 h, the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] reagent was added, and the cells were assayed as previously described (21).
RT-PCR. Vero cells were seeded into 24-well culture plates at a concentration of 104 cells per well and allowed to adhere overnight. At various time points, as stated in the figure legends and Results, cells were treated with the indicated amounts of P-PMO in 2 ml serum-free DMEM. Cells were infected with EBOV at an MOI of 5 TCID50 per cell. After 1 hour, the virus inoculum was removed and replaced with DMEM containing 2% FCS and P-PMO, where applicable. At 24 or 48 h postinfection (hpi), the cell supernatant was removed, and total RNA was isolated using the RNeasy mini kit (QIAGEN, Valencia, CA) following the manufacturer's instructions. A tenth of the eluted RNA was used in a OneStep reverse transcription-PCR (RT-PCR) (QIAGEN) according to the manufacturer's instructions, with EBOV-specific primers to amplify nucleotides 3660 to 4778 of genomic EBOV RNA (VP35/VP40) (GenBank accession number AF086833). MARV viral RNA (GenBank accession number Z29337) was detected by amplifying nucleotides 5890 to 6521 (GP gene). The PCR was performed for 27 cycles and included amplification of GAPDH (glyceraldehyde-3-phosphate dehydrogenase) mRNA as a control. Ten percent of the RT-PCR products was resolved on a 2% agarose gel and stained with ethidium bromide.
Cell-free translation assay. The nucleotide sequence corresponding to a segment of EBOV VP35 mRNA (98 to +39 relative to the translation initiator AUG codon; genomic bases 3020 to 3157), immediately followed by the protein coding sequence for firefly luciferase (without its initiator AUG codon), was subcloned into the polylinker of plasmid pCiNeo (Promega, Madison, WI). In vitro-transcribed RNA was generated via the plasmid T7 RNA polymerase promoter using the T7 MEGAscript kit (Ambion, Austin, TX). In vitro translations using a rabbit reticulocyte system (Promega) were carried out by mixing different concentrations of PMO with in vitro-transcribed RNA present at a final concentration of 1 nM. Luciferase light emission was read on an FLx800 microplate luminometer (Bio-Tek Instruments, Inc., Winooski, VT). The relative light units produced by each reaction (n = 3 per data point) were normalized to the mean of the water control reactions and expressed as percent inhibition of luciferase translation. Fifty percent effective concentration (EC50) values were determined with GraphPad Prism (San Diego, CA) graphing software.
Mice.
C57Bl/6 mice were obtained from the National Cancer Institute, Frederick Cancer Research and Development Center (Frederick, MD). The mice used in these studies were females aged 8 to 10 weeks at the start of these experiments. Mice were housed in a BSL-4 laboratory at USAMRIID in microisolator cages and provided water and chow ad libitum. Mice were treated with a 500-µg dose of the indicated PMO compound at both 24 and 4 h before or 24 h after intraperitoneal injection with
1,000 PFU, or
30,000 99% lethal dose, of mouse-adapted EBOV (4) diluted in PBS. After challenge, animals were observed for illness at least twice daily for at least 28 days. Some mice were weighed once daily as a measure of illness (4). Blood samples (n = 6 mice per group) were obtained on 3 and 6 dpi under anesthesia by cardiac puncture. Viral titers in the serum were determined by traditional plaque assay (25), and IFN-
levels were determined using a commercially available enzyme-linked immunosorbent assay per the manufacturer's directions under BSL-4 conditions (PBL Laboratories, Piscataway, NJ).
Research was conducted in compliance with the Animal Welfare Act and other federal statutes and regulations relating to animals and experiments involving animals and adhered to principles stated in the Guide for the Care and Use of Laboratory Animals (28a). The facility where this research was conducted (USAMRIID) is fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International.
Statistical analysis.
A paired Student t test was used to directly compare treated and mock-treated samples (viral titers and cytokine levels). The proportions of VP35-specific P-PMO-treated and scramble P-PMO- or PBS-treated mice were compared by one-tailed Fisher exact tests within experiments. A P value of
0.05 was considered significant.
|
|
|---|
![]() View larger version (20K): [in a new window] |
FIG. 1. Inhibition of viral replication by EBOV-specific P-PMO. (A) Schematic drawing of the EBOV Zaire genome and PMO target sequences. Genes that encode proteins directly involved in viral RNA synthesis are shaded in light gray. PMO compounds were designed to target either genomic () RNA ( ) or antigenomic (+) RNA and mRNA ( ). Compounds binding to plus-stranded RNA were directed against the translation start site regions of either the VP35 gene (3136) or the L gene (11588) and the trailer region (18959). The 3' end of the genome is magnified to show the target locations of the three negative-strand sequences in detail: leader terminus (leader), transcription start signal of the NP gene (tss-NP), and a region promoting efficient replication (rep pro, 108). Nucleotide numbers refer to GenBank accession number AF086833. (B) Vero cells were pretreated with 5.0 µM of the indicated compound for 3 h and subsequently infected with EBOV at an MOI of 5 TCID50 per cell. After the inoculum was removed, DMEM with 2% FCS containing 5 µM P-PMO was added to the cells. Oligomer sequences and their target sites are listed in Table 1. At 48 hpi, supernatant was used to infect Vero cells in 96-well plates for a TCID50 assay. On day 12 postinfection, the CPE was evaluated and the TCID50/ml calculated. The experiment was performed twice with similar outcomes.
|
![]() View larger version (49K): [in a new window] |
FIG. 2. Evaluation of cytotoxicity and efficacy of EBOV-specific P-PMO compounds. (A) Vero cells were incubated with 0 to 20 µM of the various P-PMO. After 24 h, the cells were examined for viability using an MTT assay. The data are plotted as percent changes in viability in comparison to untreated control Vero cells. Standard deviations of the means of triplicate samples are indicated by bars. (B) Light microscopy was employed to observe the CPE caused by either EBOV infection or cytotoxicity of the oligomers. Vero cells were preincubated with 5.0 µM of 3136 or scramble P-PMO for 3 h. Cells were then infected with EBOV at an MOI of 5 TCID50 per cell. The inoculum was removed, fresh DMEM was supplemented with 2% FCS, and the appropriate P-PMO at 5.0 µM was added. Cells were checked daily for CPE and photographed under light microscopy. Pictures shown here were taken at 6 dpi.
|
95%, compared to infected cells without P-PMO (Fig. 3A, right panel). Together, these results suggest that RT-PCR of viral RNA could be used as an alternative method to TCID50 analysis to determine the effect of P-PMO on viral replication. However, the lowest detectable titers using RT-PCR were
104 TCID50/ml. Since infection with an MOI of 0.05 yielded viral titers of less than 104 TCID50/ml (Fig. 3A), viral RNA could not be detected by RT-PCR in these samples. We suspect that the low sensitivity of the RT-PCR assay could have been caused by the length of the amplicon (1,119 nucleotides).
![]() View larger version (37K): [in a new window] |
FIG. 3. Inhibition of EBOV replication by the VP35-specific P-PMO is dose, time, and sequence dependent. (A) Vero cells were pretreated with 5 µM of the 3136 or scramble P-PMO for 3 h prior to infection with EBOV (MOI of 5 and 0.05 TCID50 per cell, respectively). After 1 h, the inoculum was removed and replaced with DMEM containing 2% FCS and P-PMO at 5 µM. At 48 hpi, supernatants from the P-PMO-treated and EBOV-infected cells were used to inoculate Vero cells in 96-well plates for a TCID50 assay. The CPE remained constant after 11 dpi, and the TCID50/ml was calculated (left panel). The cells were lysed, and total RNA was isolated and used as a template for RT-PCR, with primers designed to amplify 1,119 nucleotides of EBOV genomic RNA comprising sequence from the 5' region of VP35 gene and 3' region of VP40 gene, or GAPDH mRNA as a control. Products were separated by agarose gel electrophoresis and quantified after staining. The amount of EBOV-specific RNA was normalized with the GAPDH control, and reduction was calculated by setting EBOV-specific signals from infected cells without P-PMO treatment at 100% (right panel). Error bars represent the standard errors of the means of three replicates. (B) Compounds were added to Vero cells seeded in 24-well plates at the indicated time points: 3 h prior to infection, immediately after removal of inoculum (0 hpi), or at 4, 8, or 24 h after infection with EBOV (MOI = 5 TCID50 per cell). Cells were harvested at 48 hpi, total RNA was isolated, RT-PCR was performed, and the products were resolved on a 2% agarose gel stained with ethidium bromide. (C) Vero cells were preincubated with 5 or 10 µM of 3136 or scramble P-PMO. After 3 h, cells were infected with MARV at an MOI of 5 TCID50 per cell. The inoculum was removed at 1 hpi and replaced with fresh DMEM containing 2% FCS and the respective P-PMO. At 24 hpi, cells were harvested and total RNA was isolated. RT-PCRs (of a segment of GP) from genomic viral RNA template and cellular GAPDH mRNA, as a control, were performed. The experiments shown in this figure are representative of three experiments of similar design and outcome.
|
To directly examine the effect of PMO and P-PMO on translation of EBOV sequence RNA, a cell-free in vitro translation system was used. In the interest of developing an optimal antiviral compound, the efficacy of the 22-mer VP35-specific 3136 P-PMO was compared to that of 3136 PMO varying in length from 16 to 22 base residues. A reporter plasmid was constructed with a sequence corresponding to the 5' portion of the VP35 mRNA (bases 98 to +39; in relation to the AUG start site) fused to the coding sequence of firefly luciferase. Various concentrations of the PMO compounds were added to reaction mixtures containing EBOV-luciferase RNA and rabbit reticulocyte lysate, and the light emissions were measured. High levels of light emission were observed in untreated in vitro translation reactions or when a scramble PMO compound was included in the reaction (Fig. 4). In contrast, the 22-residue VP35-specific 3136 PMO and P-PMO strongly inhibited light emission (Fig. 4). The EC50 for 22-mer 3136 PMO was approximately 1 µM, and the corresponding 3136 P-PMO was approximately 0.07 µM. The shorter 16- and 19-residue 3136 PMO generated only minor inhibition of the reporter signal compared to untreated control samples, even at the highest concentration tested (30 µM). VP35-specific 3136 P-PMO of various lengths were not available for potency testing; however, Nelson et al. reported observing that increasing P-PMO length resulted in increasing potency using in vitro translation assays similar to those described above (29).
![]() View larger version (30K): [in a new window] |
FIG. 4. Sequence-specific inhibition with VP35 PMO compounds in cell-free translation assay. RNA representing the 5' 137 nucleotides of VP35 mRNA sequence, which includes the 3136 target site followed by the coding sequence for firefly luciferase, was in vitro transcribed from a reporter plasmid. Various concentrations of oligomers were added to rabbit reticulocyte lysate in vitro translation reactions in the presence of 1 nM RNA (see figure inset for oligomers tested). The graphed line represents the mean light units of three treatment wells per data point. Data are represented as the mean percent inhibitions of reporter signals of triplicate treatment wells in comparison to the means of water-treated control reactions.
|
![]() View larger version (16K): [in a new window] |
FIG. 5. Prophylactic treatment with antisense PMO compounds targeting VP35 increases survival of mice infected with EBOV. (A) Kaplan-Meier survival curve showing mice pretreated via intraperitoneal injection at 24 and 4 h before EBOV infection with 500-µg doses of 3136 22-mer ( ), leader ( ), or scramble (x) PMO. The mice were challenged with 1,000 PFU of mouse-adapted EBOV, and the results are plotted as percent survival for each group (n = 20 to 30 per group). (B) Survival of mice treated intraperitoneally with 500-µg doses of 3136 22-mer ( ), 19-mer ( ), 16-mer ( ), or scramble (x) PMO at 24 and 4 h before infection with 1,000 PFU of mouse-adapted EBOV. The data are plotted as percent survival for each group (n = 10 to 20 per group). (C) Survival curves of mice treated 24 and 4 h via intraperitoneal injection before EBOV infection with 500-µg doses of 3136 22-mer PMO ( ), 3136 22-mer P-PMO ( ), scramble PMO (x), or scramble P-PMO ( ). The data are presented as Kaplan-Meier survival curves (n = 10 to 20 per group). All mice were observed for at least 28 dpi, and no changes in survival were noted from 14 to 28 dpi.
|
levels between the scramble and VP35-specific P-PMO-treated mice (Fig. 6D). Together, our in vitro and in vivo studies suggest that the VP35-specific PMO and P-PMO generate protection by reducing virus amplification.
![]() View larger version (17K): [in a new window] |
FIG. 6. Prevention of lethal outcome in EBOV-infected mice treated therapeutically with VP35-specific P-PMO. (A) Kaplan-Meier survival curve showing mice treated intraperitoneally with a single 500-µg dose of VP35-specific 3136 22-mer P-PMO ( ) or the scramble P-PMO (x) at 24 h after receiving 1,000 PFU of mouse-adapted EBOV infection. The results are plotted as percent survival for each group (n = 12 per group). All mice were observed for at least 28 dpi, and no changes in survival were noted from 21 to 28 dpi. (B) The 3136 ( ) or scramble (x) P-PMO mice were weighed daily following lethal EBOV infection. Data are representative of the mean weight of the surviving mice in each group, which were weighed daily en masse (n = 12 per group). (C and D) Blood samples from the 3136 (white bars) or scramble (black bars) P-PMO mice were obtained on day 3 and day 6 under anesthesia by cardiac puncture. Viral titers in the serum were determined by plaque assay (C), and IFN- levels were determined using a commercially available enzyme-linked immunosorbent assay (D). Data are shown as means (n = 6 mice per group) and standard deviations (viral titers) or standard errors of the means (IFN). Significant differences (P < 0.05) between the VP35-specific 3136 and scramble P-PMO-treated group are indicated by an asterisk.
|
|
|
|---|
With the use of minigenome and infectious clone systems, a critical role for VP35 in the EBOV and MARV replication and transcription complexes has previously been demonstrated (38). Recently, an alternate antiviral approach employing transfection of small interfering RNAs against NP, VP30, and VP35 before MARV infection resulted in decreased production of viral proteins (9). The most plausible mechanism of action of the VP35-directed PMO compounds is the formation of a duplex with VP35 mRNA and subsequent interference with the assembly of the translation initiation complex at the initiator AUG region, thereby reducing or preventing translation of VP35 protein (12). As the VP35 oligomers target the sequence found in the full-length antigenomic replicative intermediate, as well as in the VP35 mRNA, we cannot rule out interference in viral genomic RNA synthesis as possibly contributing to the observed reduction in EBOV amplification. It is currently unclear whether interference with the interferon antagonist function of VP35 (1) contributed to in vivo efficacy of the 3136 oligomers; however, our current data do not support this notion as there were no significant differences in IFN-
levels between EBOV-infected mice treated with the 3136 or scramble P-PMO (Fig. 6D). To date, in vitro demonstrations of the interferon antagonistic properties of VP35 have not been repeated using in vivo animal models. Of note, EBOV infection of nonhuman primates and mice results in high levels of serum IFN-
(Fig. 6D) (16, 23). Future studies will include the utilization of PMO compounds as tools to investigate the functional role of VP35 and other viral proteins and RNA elements in the filovirus life cycle.
All of the viral sequences targeted by the P-PMO in these studies are located in domains thought to be critical for efficient viral amplification, yet only the VP35 (3136) and trailer-targeted (18959) oligomers showed efficacy. Although 18959 showed high efficacy at 5 µM, the inexplicable cytotoxicity exhibited at 20 µM lessened our enthusiasm for further testing of this compound. Why the four other antisense P-PMO were ineffective against EBOV in this cell culture model is unclear at this time. Such negative results leave the question of whether an oligomer was unable to base pair effectively to its intended target or whether base pairing did occur but did not result in significant interference with any critical viral molecular events. We suspect that the three compounds that were designed to target EBOV genomic RNA may have been hindered or unable to duplex with their target sequences due to the physical inaccessibility of negative-strand viral RNA. The tight encapsidation of filovirus genomic RNA has been demonstrated by stability of the nucleocapsid in high salt concentrations (3) and its resistance to micrococcus nuclease activity after treatment with nonionic detergents (27, 28).
In the current studies, we have used both peptide-conjugated and nonconjugated PMO. The R9F2 peptide used here enhances uptake of PMO by cells in culture (26, 31). Furthermore, P-PMO have been shown to generate a higher degree of antisense efficacy than corresponding nonconjugated PMO of the same base sequence, independent of delivery considerations (29), and this phenomenon was evident here as well in the results of the cell-free translation comparison between the 3136 PMO and 3136 P-PMO 22-mers. However, at the dose employed in this study (0.5 mg), the peptide conjugate does not appear to be critical in achieving a high degree of in vivo protection against lethal EBOV infection in this mouse model, and there was not a large difference in survival between mice treated with the conjugated or nonconjugated 22-mer 3136 compounds (Fig. 5C). Not surprisingly, an incremental increase in length (from 16 to 22 base residues) of VP35-specific PMO generated corresponding incremental increases in cell-free and in vivo efficacy.
This report is one of the first demonstrations of in vivo protection against a member of the Mononegavirales by using antisense-mediated viral gene expression knockdown. Previously, Leaman et al. showed that a modified phosphorthioate antisense oligomer designed to target repetitive gene start sequences within the respiratory syncytial virus genome could reduce viral titers in respiratory syncytial virus-infected monkeys up to 10,000-fold (22). Our current studies suggest that the VP35-directed PMO and/or P-PMO hold potential as therapeutics for EBOV infection and merit consideration for efficacy testing in a nonhuman primate model.
K.L.W and D.L.S are both recipients of a National Research Council Associateship award. A portion of the research described herein was sponsored by the Defense Threat Reduction Agency JSTO-CB (F_X012_04_RD_B) and the Medical Research and Material Command research plan no. 03-4-7J-012. This work was partially supported by the FAZIT-Stiftung and the Fonds der Chemischen Industrie (to S. Enterlein) and by the Deutsche Forschungsgemeinschaft (SFB 535). S. Enterlein performed this work in partial fulfillment of the requirements for a Ph.D. degree from the Philipps-University Marburg, Marburg, Germany.
Both authors contributed equally to this work. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»