Accumulation of Defective Neuraminidase (NA) - Semantic Scholar

Download PDF
2 downloads 0 Views 181KB Size Report
Comment
With the use of neuraminidase (NA) inhibitors (BCX-1812, oseltamivir, or zanamivir), drug- resistant variants of influenza A viruses were generated that lacked ...
591

Accumulation of Defective Neuraminidase (NA) Genes by Influenza A Viruses in the Presence of NA Inhibitors as a Marker of Reduced Dependence on NA Marina S. Nedyalkova,1,5 Frederick G. Hayden,1,2 Robert G. Webster,3,4 and Larisa V. Gubareva1,5

1 Department of Internal Medicine, Division of Epidemiology and Virology, and 2Department of Pathology, University of Virginia School of Medicine, Charlottesville; 3Department of Virology and Molecular Biology, St. Jude Children’s Research Hospital, and 4Department of Pathology, University of Tennessee Health Science Center, Memphis; 5D. I. Ivanovsky Institute of Virology, Moscow, Russia

With the use of neuraminidase (NA) inhibitors (BCX-1812, oseltamivir, or zanamivir), drugresistant variants of influenza A viruses were generated that lacked characteristic markers of resistance, such as substitutions in the NA active center or in the hemagglutinin. Drug resistance was associated with the accumulation of defective (D) RNA segments encoding NA. This phenomenon could be explained by reduced dependence of the virus on its NA activity. Analysis of the last isolates recovered from 11 volunteers, experimentally infected with influenza virus and treated with BCX-1812, revealed that they maintained full susceptibility to the drug in the NA inhibition assay (50% inhibitory concentration, 0.35– 0.5 nM ). The presence of DRNA segments was detected in 1 of these isolates but was not found in the isolates recovered from placebo recipients (n ¼ 8). Because of a lack of cell culture– based assays for susceptibility testing of human influenza viruses, detection of DRNA segments should be considered an additional assay for monitoring of NA inhibitor resistance.

The hemagglutinin (HA) and neuraminidase (NA) proteins of influenza A and B viruses interact with cellular receptors containing terminal neuraminic (sialic) acid residues [1]. HA binds to receptors and initiates viral infection, whereas NA destroys the receptors and liberates the progeny virions. NA activity is also important for preventing the self-aggregation of progeny virions whose surface glycoproteins are sialylated by cellular enzymes [2]. The enzymatic activity of viral NA is the target

Received 7 September 2001; revised 5 November 2001; electronically published 14 February 2002.

Presented in part: “Options for the Control of Influenza III,” Hersonissos, Greece, 23– 28 September 2000 (Nedyalkova MS, Hayden FG, Webster RG, Gubareva LV. Accumulation of segment 6 sgRNAs of influenza A viruses in the presence of neuraminidase inhibitors. In: Osterhaus A, Cox N, Hampson A, eds. International congress series 1219. Amsterdam: Excerpta Medica, 2001:845–53). Written informed consent was obtained from each subject in a form approved by the University of Virginia Institutional Review Board. The guidelines for human experimentation of the US Department of Health and Human Services and the Human Subjects Committee of the University of Virginia School of Medicine were followed in the conduct of this study. Financial support: National Institutes of Health (AI-45782 to L.V.G.; AI08831 and AI-29680 to R.G.W.); R. W. Johnson Pharmaceutical Research Institute (to L.V.G.); American Lebanese Syrian Associated Charities (to R.G.W.). F.G.H. is a paid consultant for the R. W. Johnson Pharmaceutical Research Institute. Reprints or correspondence: Dr. Larisa V. Gubareva, Department of Internal Medicine, Division of Epidemiology and Virology, University of Virginia, PO Box 800473, Charlottesville, VA 22908 ([email protected]). The Journal of Infectious Diseases 2002;185:591–8 q 2002 by the Infectious Diseases Society of America. All rights reserved. 0022-1899/2002/18505-0004$02.00

of a new class of anti-influenza drugs, the NA inhibitors [3]. Two drugs of this class, zanamivir and oseltamivir, are available for use in humans, and others are under investigation [4]. The use of these novel antiviral agents prompts necessary monitoring for drug resistance in clinical settings. However, the lack of a reliable cell culture–based assay for testing the susceptibility of clinical isolates to NA inhibitors impedes the detection of the resistant virus variants [5, 6]. To elude the inhibitory effect of NA inhibitors, influenza viruses must change, facilitating the release of virions from infected cells and allowing them to spread within the respiratory tract. Substitutions at conserved residues in the active center of NA have been identified in NA inhibitor– resistant viruses selected in vitro [7– 13] or recovered from drug-treated patients [14–17]. Such variants have been detected in NA inhibition assays [18]. Another mechanism of resistance involves substitution(s) in the HA that appear to facilitate release of virus from infected cells by reducing the efficiency of the virus binding to receptors [9, 19– 22]. These virus variants have a substantially decreased requirement for NA activity, and they demonstrate cross-resistance to NA inhibitors in cell culture [18]. In addition, some of these mutant cells exhibit drug dependence that is manifested by augmentation of virus yields in the presence of low concentrations of NA inhibitor in vitro [9, 19] and in experimental animals [18]. These variants also demonstrate crossresistance to NA inhibitors in cell culture. The reduced binding efficiency of influenza viruses, resistant to NA inhibitors because of HA substitutions, decreases the dependence of these viruses on their own NA activity during the

592

Nedyalkova et al.

release of progeny virions from infected cells. In this respect, they are comparable to the propagation of influenza A viruses in the presence of exogenous (i.e., bacterial) NA, another situation that diminishes the need for the viral enzyme. Work elsewhere showed that virus variants, selected in the presence of exogenous NA, accumulated defective RNA segments encoding the viral NA [23–25]. They were derived from the progenitor RNA segment in which a massive internal region was deleted, whereas the 50 and 30 gene termini were retained. Therefore, these defective segments lacked the region encoding the NA active center. Although defective viral RNAs seem to arise because of errors made by the viral RNA polymerase [26], their accumulation appears to result from a diminished need for viral NA enzymatic activity. Because of these observations, we wanted to investigate whether the reduced binding efficiency of NA inhibitor– resistant viruses would lead to accumulation of defective (D) RNA segments encoding NA in vitro and in infected humans. Consequently, we examined the genomes of the resistant viruses adapted to grow in the presence of the NA inhibitor and assayed samples from experimentally infected humans given the neuraminidase inhibitor BCX-1812.

Materials and Methods Compounds. The NA inhibitors oseltamivir carboxylate (GS4071) and BCX-1812 (RWJ-270201) were provided by BioCryst. They were resuspended in distilled water and stored at 220 C before they were further diluted for use in the enzyme assay. Viruses and cells. Clinical isolates of influenza A (H1N1) virus were recovered from patients at the University of Virginia Health Sciences Center during the 1994– 1995 influenza season. Clinical isolates were propagated twice in Madin-Darby canine kidney (MDCK) cells by using a standard procedure involving Eagle MEM containing 10% fetal bovine serum [27, 28]. A/turkey/Minnesota/833/80 (H4N2) and its zanamivir-resistant variant (resistant mutant [RM]– 1) [9] were obtained from the repository at St. Jude Children’s Research Hospital (Memphis). Virus isolates were recovered from volunteers experimentally infected with A/Texas/36/91 (H1N1) and treated with either BCX1812 or a placebo [29]. The treatment was started 24 h after virus inoculation and continued for 5 days. The last isolates recovered from the nasal washes of the BCX-1812 recipients (n ¼ 11) and the placebo recipients (n ¼ 8) who shed the virus for >5 days were propagated in MDCK cells in the absence of drug and stored at 270 C until used for the analysis. Influenza virus propagation in MDCK cells in the presence of NA inhibitor. To generate influenza virus resistant to the NA inhibitor BCX-1812 in vitro (the 31-RM variant), the clinical isolate A/Charlottesville/31/95 (H1N1) was passaged 18 times in MDCK cells in the presence of increasing concentrations of BCX-1812 (0.3– 300 mM ) or in the absence of the drug (as a control) by a procedure described elsewhere [9]. Viral infection of monolayers of MDCK cells was done at a low MOI (0.001 TCID50/cell). The NA inhibitor was added to media after virus adsorption for 1 h at room temperature. Virus yields harvested after 72 h of infection

JID 2002;185 (1 March)

at 34 C were used for the next round of propagation. Similarly, the virus isolate A/Charlottesville/28/95 (H1N1) was propagated 5 times in the presence of increasing concentrations (0.1– 2000 mM ) of oseltamivir carboxylate (GS4071) to obtain the resistant variant 28-RM. Plaque assay. Viral resistance to NA inhibitors was evaluated on monolayers of MDCK cells by a plaque reduction assay described elsewhere [9]. Confluent monolayers were inoculated with virus diluted in Eagle MEM to give 50 plaques per well. After the virus adsorption to cells was completed and the residual virus was removed, 10-fold dilutions (0.1– 100 mg/mL) of the NA inhibitor were added in the agar overlays. The size and number of plaques formed in the absence of the NA inhibitor were used as controls. NA inhibition assay. Viruses (grown in the absence of an NA inhibitor) from clarified cell culture supernatants were used as a source of NA activity in the modified fluorometric assay of Potier et al. [30] for measuring influenza NA activity and its inhibition by NA inhibitor. The assay measures 4-methylumbelliferone released from the fluorogenic substrate 20 -(4-methylumbelliferyl)-aD -N-acetylneuraminic acid (Sigma) by the influenza virus enzyme. The reaction was done in 0.33 mM 2-(N-morpholino)ethanesulfonic acid containing 4 mM CaCl2 at the final substrate concentration of 100 mM, as described elsewhere [31]. Reverse-transcriptase (RT) polymerase chain reaction (PCR) amplification of the NA and HA genes. Viral RNAs were extracted from the virus-containing cell supernatants with Trizol (Life Technologies) and used with a universal primer (50 -AGCAAAAGCAGG-30 ) for the synthesis of cDNA according to procedures described elsewhere [15]. Two overlapping fragments of the full-length RNA segment 6 encoding NA were amplified by use of the primer pair N1.2 (50 -CAAAAGCAGGAGTTTAAAATG-30 ) and N1.702R (50 -ACACAGACACATTCAGACTC-30 ) and a second primer pair, N1.600 (50 -TGGCTAACAATCGGAATTTC-30 ) and N1.1439R (50 -AACGGACTACTTGTCAATGG-30 ). The PCR products were purified with the use of a QIAquick PCR purification kit (Qiagen) and subjected to sequence analysis. To identify DRNA segments in the 31-RM variant and clinical isolates, we used PCR amplification with primers complementary to the 30 and 50 termini of the NA gene (N1.2 and N1.1439R). The products of the reactions were analyzed in a 1% agarose gel stained with ethidium bromide according to standard procedures. PCR products that were smaller than the full-length RNA segment (the DRNA) were extracted from the gel according to the manufacturer’s protocol (QIAquick gel extraction kit; Qiagen) and subjected to sequence analysis. The HA gene was analyzed by a procedure similar to that for the NA gene, the only difference being the use of primers H1.11 (50 -GGGGAAAATAAAAACAACC-30 ), H1.1101R (50 -TCCATCTATCATTCCAGTCC-30 ), H1.913 (50 -GCTATAAACAGTAGTATTCC-30 ), and H1.1725R (50 -GAAATTCTGGTCTCAGATGC-30 ). Sequence analysis of the HA and NA genes. Corresponding PCR products were sequenced by use of Taq Dye Terminator chemistry and then analyzed on an ABI 373 DNA sequencer (Applied Biosystems), according to the manufacturer’s instructions, at the Center of Biotechnology at the University of Virginia. Sequencher 4.0 software (Gene Codes) was used for the analysis and translation of nucleotide data. The HA and NA sequences of A/Charlottesville/ 31/95 (H1N1) and A/Charlottesville/28/95 (H1N1) were submitted

JID 2002;185 (1 March)

Marker of Reduced Dependence on NA Activity

to the GenBank database (accession nos. AF398868, AF398869, AF398870, AF398872, AF398873, AF398874, AF398875, and AF398878).

Results Generation of a BCX-1812–resistant mutant and analysis of its HA and NA sequences. The variant of A/Charlottesville/ 31/95 (H1N1), selected after 18 passages in the presence of BCX1812 (designated 31-RM), was analyzed for drug resistance by a plaque reduction assay, and these results were compared with those for the wild-type and MDCK-passaged counterparts. The 31-RM variant was highly drug resistant because no reduction in plaque size was observed when BCX-1812 was present in the agar overlay (table 1). In contrast, the plaque size of the 2 counterpart viruses was reduced by 50% at .1000-fold lower concentration of the inhibitor (table 1). Moreover, the 31-RM variant was cross-resistant to the NA inhibitors zanamivir and oseltamivir carboxylate in cell culture (IC50 [drug concentration that inhibits 50% of enzyme activity or reduces plaque size], .100 mg/mL). However, when the NA of the 31-RM variant was examined in the NA inhibition assay, no increase in the IC50 values was detected (table 1). Therefore, its enzyme remained fully susceptible to the drug. Substitutions in either HA or NA or in both surface glycoproteins are the recognized molecular markers of resistance of influenza viruses to NA inhibitors. Therefore, we compared the sequences of the HA and NA genes of the 31-RM variant with those of its MDCK-passaged counterpart by amplifying and sequencing 2 overlapping fragments of the cDNA encoding each

Table 1.

593

gene. The results of this analysis predicted no amino acid substitutions in HA, whereas the NA of the 31-RM variant contained 2 substitutions, Asn58!Asp and Ile211!Thr (table 1). Neither of these substitutions involved the conserved residues that form the active center of NA [2]. The absence of substitutions in the NA active center of the 31-RM variant is consistent with the retention of NA sensitivity to BCX-1812, as assessed in the NA inhibition assay. Detection of defective NA genes in the virus passaged in the presence of NA inhibitors. To test our hypothesis—namely, that reduced virus susceptibility to NA inhibitors and, therefore, reduced virus dependence on NA activity—would lead to accumulation of DRNA segments, we examined the genomes of the drug-resistant viruses. To distinguish between a full-length RNA segment 6, which encodes NA, and DRNA segments that have an internal deletion, we amplified cDNA by PCR with primers complementary to the 30 and 50 termini of the RNA segment. The PCR products obtained from the cDNAs of the wild-type and MDCK-passaged viruses were the same size as that predicted for the full-length segment (1400 bp; figure 1A, lanes 1 and 2 ). In contrast, when the cDNA of the 31-RM variant was amplified, the band corresponding to the full-length segment was very faint or not detected on the gels (figure 1A, lane 3 ). Instead, a prominent, smaller band (600 bp) was present. As we hypothesized, sequence analysis of this smaller PCR product confirmed that it arose from an NA-encoding RNA segment that had undergone a massive internal deletion. We then tested whether defective copies of the RNA segment encoding HA could be detected in the 31-RM variant. Defective HA genes were not detected with primers complementary to the

Characterization of the influenza virus variants selected in vitro in the presence of neuraminidase (NA) inhibitors.

Virus strain, variant

NA inhibitor

Plaque size reduction, IC50 (mM )

A/Charlottesville/31/95 (H1N1) Wild type MDCK passaged 31-RM

None None BCX-1812

0.3 0.3 .300

A/Charlottesville/28/95 (H1N1) Wild type MDCK passaged 28-RM A/turkey/Minnesota/833/80 (H4N2)c Wild type MDCK passaged RM-1

None None Oseltamivir carboxylate None None Zanamivir

ND ND ND

0.001 ,0.01 10

NA inhibition, IC50 (nM )

Substitution(s) in HAa

Substitution(s) in NAb

Detection of DRNA

0.3 0.2 0.1

None None None

None None Asn58!Asp, Ile211!Thr

No No Yes

0.6 0.6 0.5

None HA1; Ser140!Pro HA1; Ser140!Pro

None None None

No No Yes

8.0 7.0 6.0

None HA2; Gly114!Lys HA2; Gly75!Glu

None None Arg249!Lys

No No Yes

NOTE. Plaque reduction and NA inhibition assays were done in the presence of the NA inhibitor used for selection of resistant variant in Madin-Darby canine kidney (MDCK) cells. IC50 values (drug concentration that inhibits 50% of enzyme activity or reduces plaque size) are averages from at least 3 experiments. HA, hemagglutinin; ND, not determined; RM, resistant mutant; DRNA, defective RNA segments encoding NA. a H3, HA subtype numbering. b N2, NA subtype numbering. c Data for A/turkey/Minnesota/833/80 and its RM-1 variant have been reported elsewhere [9, 10, 22].

594

Nedyalkova et al.

JID 2002;185 (1 March)

Figure 1. Detection of defective genes in influenza A virus genome. Products of reverse-transcriptase polymerase chain reaction amplification with primers complementary to 30 and 50 termini were analyzed by electrophoresis through 1% agarose gel and were stained with ethidium bromide. Molecular markers (Mw): HaeIII fragments of fX174 DNA and 1-kb DNA ladder. A and B, Amplification of neuraminidase (NA) and hemagglutinin (HA) genes, respectively. Lane 1, Wild-type virus A/Charlottesville/31/95 (H1N1); lane 2, wild-type virus after passage in MadinDarby canine kidney (MDCK) cells; lane 3, 31-RM variant obtained by passaging wild-type virus in presence of NA inhibitor BCX-1812. C, Amplification of NA gene. Lane 1, 31-RM variant obtained after 18 passages in MDCK cells in presence of NA inhibitor BCX-1812; lane 2, 31-RM variant propagated for 5 additional passages in absence of BCX-1812. D, Amplification of NA gene. A/Charlottesville/28/95 (H1N1) was passaged in MDCK cells in presence of NA inhibitor oseltamivir carboxylate. Lanes 1–5, Virus after culture in presence of NA inhibitor for 1– 5 passages, respectively; lane 6, virus passaged in MDCK cells in absence of NA inhibitor.

30 and 50 termini of the HA gene (figure 1B), whereas full-size HA products were detected readily. Therefore, accumulation of DRNA segments encoding NA—but not HA—was observed in the drug-resistant 31-RM variant. Next we examined how withdrawal of the NA inhibitor affected the detection of the RNA segments in the viral genome. The 31-RM variant was propagated for 5 additional passages in the absence of the NA inhibitor and then reexamined. The PCR product corresponding to the full-length RNA segment was readily visible on the gel (figure 1C). Therefore, propagation of 31-RM in the absence of the NA inhibitor increased the proportion of full-length RNA segments encoding NA. To determine whether accumulation of the DRNA segments was limited to the particular virus or was the result of the particular NA inhibitor that we used, we passaged the clinical isolate A/ Charlottesville/28/95 (H1N1) in a manner similar to that of the 31-RM variant in the presence of the NA inhibitor oseltamivir carboxylate. Viruses were harvested after each of the 5 subsequent passages and analyzed. No DRNA segments encoding mutant NA were detected in the virus, before and after 5 passages in the absence of the NA inhibitor (figure 1D, lane 6 ). The emer-

gence of DRNA segments was detected after the first passage of the virus in the presence of the NA inhibitor (figure 1D, lane 1 ). The DRNA segments were 250– 700 nt. The accumulation of a single dominant DRNA segment became prominent after 4 passages, when the PCR product for the full-length RNA segment was no longer visible on the gel (figure 1D, lanes 5 and 6 ). The virus harvested after 5 passages in the presence of the NA inhibitor was designated as 28-RM, and the HA and NA sequences of this virus were compared with those of the MDCKpassaged counterpart. No changes in these 2 sequences of the viruses were detected, although both of the viruses contained an HA substitution that was absent in the wild-type HA (table 1). In addition, we examined the virus variant of A/turkey/Minnesota/833/80 (H4N2), which we previously selected in MDCK cells in the presence of zanamivir and the properties of which have been described elsewhere [9, 10, 22] (table 1). In light of the results of the plaque reduction assay, the virus variant selected after 8 passages (designated RM-1) had reduced susceptibility to zanamivir, but there were no substitutions in the active center of its NA. RT-PCR analysis of the NA gene revealed the presence of 2 DRNA segments (390 bp and 380 bp) (table 1).

JID 2002;185 (1 March)

Marker of Reduced Dependence on NA Activity

Therefore, passage of 3 strains of influenza A viruses in the presence of an NA inhibitor (BCX-1812, oseltamivir carboxylate, or zanamivir) resulted in the accumulation of DRNA segments. Examination of influenza virus isolates recovered from volunteers treated with an NA inhibitor. The activity of NA is essential for efficient viral propagation in vivo [23, 25]. Therefore, we wanted to examine whether accumulation of viral DRNA segments could be detected in humans treated with an NA inhibitor. For this purpose, we analyzed virus isolates recovered from volunteers experimentally infected with influenza A (H1N1) virus and treated with BCX-1812 [29]. When tested in the NA inhibition assays, drug susceptibility was similar among the viruses recovered from both groups of subjects and among the challenge virus (IC50, 0.35– 0.5 nM ). Full-length RNA segments were detected in all isolates (figure 2A and 2B). However, the isolate recovered from 1 subject who received BCX-1812 contained an additional band of a smaller size, indicating the presence of a DRNA segment (figure 2B, lane 7 ). The sequence analysis of this DRNA segment revealed that, as in the viruses generated in vitro, it contained the se-

595

quences of the 50 and 30 ends (the sequence from nt 1–127 connected to the sequence from nt 1027–1461) but lost the internal part of the segment encoding NA (900 nt shorter than the wildtype RNA segment). To confirm the presence of the DRNA segment in the virus before its propagation in cell culture, we performed RT-PCR amplification of viral RNA extracted directly from the clinical sample. The size of the DRNA segment detected in the virus of the original clinical sample was similar to that of the virus grown in MDCK cells (550 bp). When the individual virus clones recovered from the nasal wash were subjected to RT-PCR, only 1 of 8 clones contained the DRNA segment (not shown). The virus titers in this subject’s nasal washes recovered on days 1–6 of shedding were 104, 104.8, 103.2, 103, 100.8, and 103 TCID50/mL, respectively. No virus was detected on day 7 after infection. Therefore, the virus containing the DRNA segments was cleared. Next, we examined how early during the treatment course the DRNA segments could be detected in the clinical samples recovered from this NA inhibitor recipient. The DRNA segments were detected in the viruses recovered on days 2, 5, and 6 of

Figure 2. Amplification of the neuraminidase (NA) gene with primers complementary to the 30 and 50 gene termini. Products of reversetranscriptase polymerase chain reaction amplification were analyzed by electrophoresis through 1% agarose gel and were stained with ethidium bromide. Molecular markers (Mw): HaeIII fragments of fX174 DNA. A and B, Virus isolates were recovered on last day of shedding from volunteers experimentally infected with A/Texas/36/91 (H1N1) and treated with either placebo or BCX-1812, respectively; C, isolates recovered on days 1– 6 of shedding from volunteer experimentally infected with A/Texas/36/91 (H1N1) and treated with BCX-1812.

596

Nedyalkova et al.

shedding but were not detected on days 3 and 4 (figure 2C). The DRNA segments found in the isolates recovered on days 2, 5, and 6 differed in size (250, 500, and 550 bp, respectively). The size of the DRNA segment of the viruses recovered on days 5 and 6 is more consistent with the length of the DRNA segment detected in in vitro studies. The absence of the DRNA segments in the virus isolates recovered on days 3 and 4 could be due to the low titer of virus variants carrying DRNA segments in the nasal washes. Discussion In the present study, we describe the rapid accumulation of defective RNA segments encoding NA as a result of the passage of influenza virus in the presence of several NA inhibitors. The DRNA segments were detected in influenza A viruses carrying the NA of N1 and N2 subtypes. In addition, we identified DRNA segments in clinical samples recovered from a subject experimentally infected with influenza A virus and treated with an NA inhibitor. Detection of DRNA segments in clinical samples demonstrates the need for additional studies to evaluate the incidence of their emergence in patients who receive NA inhibitors for natural influenza A virus infections. If the majority of influenza A viruses resistant to NA inhibitor acquired reduced dependence of their NA enzymatic activity and all of them accumulated DRNA segments encoding NA, detection of such defective genes directly in clinical samples could be a useful test for emerged resistance. Although the presented results indicate existence of the discrete mechanism of influenza virus resistance to NA inhibitors that does not involve substitutions in the NA active site or in HA, they do not reveal the molecular basis of this mechanism. The genome of influenza A virus consists of 8 RNA segments of negative polarity, of which RNA segment 6 encodes a single protein, NA [1]. The mechanism leading to the accumulation of defective RNA segments may be associated with the formation of virus aggregates in the presence of the NA inhibitors [32]. Therefore, coinfection by virions carrying a complete genome and virions carrying any defective RNA segment could allow assembly of the defective RNA segments into the genomes of the progeny virions. If a defective segment were small, it would have a replicative advantage; thus, the number of the progeny virions carrying the defective segment would increase with passage. Accumulation of defective RNAs in the genome of influenza viruses also may be a result of virus propagation at a high MOI (von Magnus effect) [26, 33]. In the present study, we did not control the MOI, although we used high dilutions of the virus yields for subsequent passage of the virus. Either of the described feasible mechanisms of accumulation of defective genes is based on coinfection of the same cell, with virions containing a complete set of segments and virions carrying the defective ones. Of importance, these mechanisms do not account for the

JID 2002;185 (1 March)

absence of defective RNA segments encoding HA in the genome of the virus variants selected in the presence of NA inhibitor in our study. Of note, the number of DRNA segments was not necessarily greater than the number of full-length RNA segments in the genomes of the 31-RM (figure 1A, lane 3 ) or 28-RM (figure 1D, lane 6 ) variants, as well as in the clinical isolate (figure 2B, lane 7 ). The virions carrying DRNA segments could possibly exist as a small proportion of the total virus load, and the overwhelming abundance of the products synthesized from the DRNA segments is likely to be a result of the more efficient amplification of a short (rather than long) template by RT and Taq polymerase. We believe that the emergence of the DRNA segments encoding NA was caused by the virus’s reduced dependence on its own enzyme. This property was acquired by the virus variants during propagation in the presence of an NA inhibitor and was manifested by resistance to the drug in cell culture. Sequence analysis of the NA of the 31-RM variant revealed 2 substitutions: one (Asn58!Asp) resulted in abolishment of a glycosylation site in the stalk region, whereas the other (Ile211!Thr) led to the creation of a potential glycosylation site in the head region of the molecule. Because of the location with respect to the enzyme active site, either of the substitutions is unlikely to be involved in emergence of viral resistance to NA inhibitors. The fact that the second NA inhibitor–resistant variant, 28-RM, did not contain any substitutions in its NA supports such a conclusion. An NA-independent mechanism of emerged resistance was supported by the results of the NA inhibition assays, which demonstrated unaltered susceptibility to BCX-1812. The observed cross-resistance of the BCX-1812– selected virus to 2 other NA inhibitors in cell culture also supports its NA-independent mechanism of drug resistance [18]. It seems unlikely that the emergence of resistant viruses with reduced efficiency of binding would have a consequence for the clinical use of BCX-1812. In our experiments in a ferret model, we demonstrated that the virus carrying DRNA segments encoding NA has reduced infectivity and virulence (unpublished data). Results of several studies support the possible role of the HA substitutions in the development of resistance to zanamivir and oseltamivir carboxylate [13–15, 19]. It appears that HA substitutions, by reducing dependence of the virus on the enzymatic activity of NA, could serve as prerequisites for the changes in the NA enzyme itself [18]. Confirmation of the role of HA substitutions for emerged resistance to NA inhibitors is difficult [18]. Of importance, propagation of egg-adapted human or avian influenza viruses in MDCK cells [34–37] could be responsible for at least some of the changes identified in the HA of the NA inhibitor–resistant variants in previous studies [7, 12, 19, 38], including ours [9, 10, 22]. We also reported selection of HA variants in humans experimentally infected with egg-adapted influenza A virus [15]. In contrast to egg-grown influenza viruses, human influenza viruses passaged in MDCK cells typically show few HA

JID 2002;185 (1 March)

Marker of Reduced Dependence on NA Activity

sequence variations [39, 40]. For this reason, in the present study, we used clinical isolates A/Charlottesville/28/95 and A/ Charlottesville/31/95, which were isolated and propagated exclusively in MDCK cells. Compared with the HA sequences of viruses passaged in parallel in the absence of NA inhibitor, the HA sequences of viruses passaged in the presence of NA inhibitor contained no substitutions. Thus, the molecular basis of the viral resistance lies outside the HA sequence. Overall, we did not detect any of the characteristic changes described previously for NA inhibitor–resistant viruses, such as amino acid substitutions in either or both of the surface glycoproteins, HA and NA. In the reports by Tai et al. [12] and Staschke et al. [7], the NA inhibitor– resistant virus variants had substitutions in NA but not in HA. The substitutions in NA that confer drug resistance to these virus variants at the same time compromise the function of the enzyme [7, 10, 41, 42]. This situation leaves unanswered the question of compensatory changes for the reduced NA activity of these resistant variants. The virus variant could acquire the ability to replicate in the presence of the NA inhibitor because of the reduced efficiency of binding. We speculate that changes in genes other than the HA gene could be responsible for this feature. For example, it was shown that virus interactions with cellular receptors could be influenced by the M1 protein [43]. Our findings, together with those of others, necessitate further investigation of the mechanism(s) by which influenza virus compensates for the reduced NA activity in the presence of NA inhibitors.

Acknowledgments

We thank Amy L. B. Frazier and Julia Cay Jones (St. Jude Children’s Research Hospital, Memphis) for editorial work and Douglas Schallon (University of Virginia, Charlottesville) for excellent technical assistance.

References 1. Lamb RA, Krug RM. Orthomyxoviridae: the viruses and their replication. In: Knipe DM, Howley PM, eds. Fields’ virology. 4th ed. Philadelphia: Lippincott Williams &Wilkins, 2001:1487–532. 2. Colman PM. Neuraminidase: enzyme and antigen. In: Krug RM, ed. The influenza viruses. New York: Plenum Press, 1989:175–210. 3. Gubareva LV, Kaiser L, Hayden FG. Influenza virus neuraminidase inhibitors. Lancet 2000; 355:827–35. 4. Babu YS, Chand P, Bantia S, et al. BCX-1812 (RWJ-270201): discovery of a novel, highly potent, orally active, and selective influenza neuraminidase inhibitor through structure-based drug design. J Med Chem 2000; 43:3482–6. 5. Tisdale M. Monitoring of viral susceptibility: new challenges with the development of influenza NA inhibitors. Rev Med Virol 2000; 10:45–55. 6. Zambon M, Hayden FG, Global Neuraminidase Inhibitor Susceptibility Network. Position statement: Global Neuraminidase Inhibitor Susceptibility Network. Antiviral Res 2001; 49:147–56. 7. Staschke KA, Colacino JM, Baxter AJ, et al. Molecular basis for the resistance of influenza viruses to 4-guanidino-Neu5Ac2en. Virology 1995; 214:642–6.

597

8. Blick TJ, Tiong T, Sahasrabudhe A, et al. Generation and characterization of an influenza virus neuraminidase variant with decreased sensitivity to the neuraminidase-specific inhibitor 4-guanidino-Neu5Ac2en. Virology 1995; 214:475– 84. 9. Gubareva LV, Bethell R, Hart GJ, Murti KG, Penn CR, Webster RG. Characterization of mutants of influenza A virus selected with the neuraminidase inhibitor 4-guanidino-Neu5Ac2en. J Virol 1996; 70:1818–27. 10. Gubareva LV, Robinson MJ, Bethell RC, Webster RG. Catalytic and framework mutations in the neuraminidase active site of influenza viruses that are resistant to 4-guanidino-Neu5Ac2en. J Virol 1997; 71: 3385– 90. 11. McKimm-Breschkin JL, Sahasrabudhe A, Blick TJ, et al. Mutations in a conserved residue in the influenza virus neuraminidase active site decreases sensitivity to Neu5Ac2en-derived inhibitors. J Virol 1998; 72: 2456– 62. 12. Tai CY, Escarpe PA, Sidwell RW, et al. Characterization of human influenza virus variants selected in vitro in the presence of the neuraminidase inhibitor GS4071. Antimicrob Agents Chemother 1998; 42:3234– 41. 13. Barnett JM, Cadman A, Burrell FM, et al. In vitro selection and characterisation of influenza B/Beijing/1/87 isolates with altered susceptibility to zanamivir. Virology 1999; 265:286– 95. 14. Gubareva LV, Matrosovich MN, Brenner MK, Bethell RC, Webster RG. Evidence for zanamivir resistance in an immunocompromised child infected with influenza B virus. J Infect Dis 1998; 178:1257–62. 15. Gubareva LV, Kaiser L, Matrosovich MN, Soo-Hoo Y, Hayden FG. Selection of influenza virus mutants in experimentally infected volunteers treated with oseltamivir. J Infect Dis 2001; 183:523–31. 16. Treanor JJ, Hayden FG, Vrooman PS, et al. Efficacy and safety of the oral neuraminidase inhibitor oseltamivir in treating acute influenza: a randomized controlled trial. US Oral Neuraminidase Study Group. JAMA 2000; 283:1016– 24. 17. Whitley RJ, Hayden FG, Reisinger KS, et al. Oral oseltamivir treatment of influenza in children. Pediatr Infect Dis J 2001; 20:127– 33. 18. McKimm-Breschkin JL. Resistance of influenza viruses to neuraminidase inhibitors. Antiviral Res 2000; 47:1–17. 19. McKimm-Breschkin JL, Blick TJ, Sahasrabudhe A, et al. Generation and characterization of variants of NWS/G70c influenza virus after in vitro passage in 4-amino-Neu5Ac2en and 4-guanidino-Neu5Ac2en. Antimicrob Agents Chemother 1996; 40:40– 6. 20. Penn CR, Barnett JM, Bethell RC, et al. Selection of influenza virus with reduced sensitivity in vitro to the neuraminidase inhibitor GG167 (4guanidino-Neu5Ac2en): changes in the haemagglutinin may compensate for loss of neuraminidase activity. In: Brown LE, Hampson AW, Webster RG, eds. Options for the control of influenza III. Amsterdam: Elsevier Science, 1996:735– 40. 21. Sahasrabudhe A, Black TJ, McKimm-Breschkin JL. Influenza virus variants resistant to GG167 with mutations in the haemagglutinin. In: Brown LE, Hampson AW, Webster RG, eds. Options for the control of influenza III. Amsterdam: Elsevier Science, 1996:748– 52. 22. Gubareva LV, Bethell RC, Penn CR, Webster RG. In vitro characterization of 4-guanidino-Neu5Ac2en– resistant mutants of influenza A virus. In: Brown LE, Hampson AW, Webster RG, eds. Options for the control of influenza III. Amsterdam: Elsevier Science, 1996:753–60. 23. Liu C, Eichelberger MC, Compans RW, Air GM. Influenza type A virus neuraminidase does not play a role in viral entry, replication, assembly, or budding. J Virol 1995; 69:1099–106. 24. Liu C, Air GM. Selection and characterization of a neuraminidase-minus mutant of influenza virus and its rescue by cloned neuraminidase genes. Virology 1993; 194:403–7. 25. Hughes MT, Matrosovich MN, Rodgers ME, McGregor M, Kawaoka Y. Influenza A viruses lacking sialidase activity can undergo multiple cycles of replication in cell culture, eggs, or mice. J Virol 2000; 74: 5206– 12.

598

Nedyalkova et al.

26. Jennings PA, Finch JT, Winter G, Robertson JS. Does the higher order structure of the influenza virus nucleoprotein guide sequence rearrangements in influenza viral RNA? Cell 1983; 34:619–27. 27. Hayden FG, Jennings L, Robson R, et al. Oral oseltamivir in human experimental influenza B infection. Antivir Ther 2000; 5:205–13. 28. Rollins BS, Elkhatieb AH, Hayden FG. Comparative anti-influenza virus activity of 20 -deoxy-20 -fluororibosides in vitro. Antiviral Res 1993; 21: 357–68. 29. Hayden FG, Treanor JJ, Qu R, Fowler C. Safety and efficacy of an oral neuraminidase inhibitor RWJ-270201 in treating experimental influenza A and B in healthy adult volunteers [abstract 1156]. In: Program and abstracts of the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy (Toronto). Washington, DC: American Society for Microbiology, 2000. 30. Potier M, Mameli L, Belisle M, Dallaire L, Melancon SB. Fluorometric assay of neuraminidase with a sodium (4-methylumbelliferyl-a-D -Nacetylneuraminate) substrate. Anal Biochem 1979; 94:287–96. 31. Hayden FG, Gubareva LV, Monto AS, et al. Inhaled zanamivir for the prevention of influenza in families. N Engl J Med 2000; 343:1282–9. 32. Palese P, Compans RW. Inhibition of influenza virus replication in tissue culture by 2-deoxy-2,3-dehydro-N-trifluoro-acetyl-neuraminic acid (FANA): mechanism of action. J Gen Virol 1976; 33:159–63. 33. Nayak DP, Chambers TM, Akkina RK. Defective-interfering (DI) RNAs of influenza viruses: origin, structure, expression, and interference. Curr Top Microbiol Immunol 1985; 114:103–51. 34. Rott R, Orlich M, Klenk HD, Wang ML, Skehel JJ, Wiley DC. Studies on the adaptation of influenza viruses to MDCK cells. EMBO J 1984; 3: 3329–32.

JID 2002;185 (1 March)

35. Orlich M, Khatchikian D, Teigler A, Rott R. Structural variation occurring in the hemagglutinin of influenza virus A/turkey/Oregon/71 during adaptation to different cell types. Virology 1990; 176:531–8. 36. Gubareva LV, Wood JM, Meyer WJ, et al. Codominant mixtures of viruses in reference strains of influenza virus due to host cell variation. Virology 1994; 199:89–97. 37. Lin YP, Wharton SA, Martin J, Skehel JJ, Wiley DC, Steinhauer DA. Adaptation of egg-grown and transfectant influenza viruses for growth in mammalian cells: selection of hemagglutinin mutants with elevated pH of membrane fusion. Virology 1997; 233:402– 10. 38. Bantia S, Ghate AA, Ananth SL, et al. Generation and characterization of a mutant of influenza A virus selected with the neuraminidase inhibitor BCX-140. Antimicrob Agents Chemother 1998; 42:801–7. 39. Katz JM, Naeve CW, Webster RG. Host cell– mediated variation in H3N2 influenza viruses. Virology 1987; 156:386–95. 40. Robertson JS. An overview of host cell selection. Dev Biol Stand 1999; 98:7– 11. 41. Colacino JM, Chirgadze NY, Garman E, et al. A single sequence change destabilizes the influenza virus neuraminidase tetramer. Virology 1997; 236:66–75. 42. McKimm-Breschkin JL, Mcdonald M, Blick TJ, Colman PM. Mutation in the influenza virus neuraminidase gene resulting in decreased sensitivity to the neuraminidase inhibitor 4-guanidino-Neu5Ac2en leads to instability of the enzyme. Virology 1996; 225:240–2. 43. Tong N, Nobusawa E, Morishita M, Nakajima S, Nakajima K. M protein correlates with the receptor-binding specificity of haemagglutinin protein of reassortant influenza A (H1N1) virus. J Gen Virol 1998; 79: 2425–34.