Journal of Medical Virology 36:217-221 (1992)

Infection Enhancement of Influenza A NWS Virus in Primary Murine Macrophages by Anti-Hemagglutinin Monoclonal Antibody Hiroshi Ochiai, Masahiko Kurokawa, Shoko Matsui, Tatsuru Yamamoto, Yoshito Kuroki, Chiharu Kishimoto, and Kimiyasu Shiraki Department of Virology (H.O., M.K., S.M., T.Y., Y.K., K.S.) and 2nd Department of Internal Medicine (C.K.), Toyama Medical and Pharmaceutical Uniuersity, Toyama, Japan

and continuous M@-likecell lines in the presence of subneutralizing antibody has been demonstrated for several viruses [Takeda et al., 1988; Traavik et al., 1988; Gimenez et al., 1989; for reviews, see Halstead, 1982; and Porterfield, 19861. These non-beneficial and paradoxical infectious pathways might contribute to the pathogenesis of viruses [Halstead, 19821. We have shown that antibody-enhanced influenza virus infection occurs in a murine M@-likecell line P388D1 via Fc receptor-mediated entry in the presence of subneutralizing concentrations of rabbit antiserum or anti-hemagglutinin (HA) monoclonal antibody (mAB) [Ochiai et al., 1988, 19901. However, this phenomenon has not yet been investigated on primary M@, since influenza virus replication has been reported to be abortive in the primary M@in which viral antigens are produced but no infectious virus is released [Nugent and Pesanti, 1979; Rodgers and Mims, 1981,19821. On the other hand, it is evident that proteolytic cleavage of the precursor HA of influenza viruses into HA1 and HA2 fragments is essential for infectivity [Klenk et al., 1975; Lazarowitz and Choppin, 1975; Rott et al., 19801. Hence, we consider that one of the possible reasons for the abortive infection in primary M@might be due to the lack of proteolytic cleavage of HA and, if it is true, the antibody-enhanced infection might occur in the presence of appropriate protease for cleavage. To test KEY WORDS: influenza virus, resident and these possibilities, the effect of trypsin was investithioglycolate-elicited peritoneal gated on the infectivity of influenza A NWS virus macrophages, acquisition of ingrown in primary murine M@ and antibody-mediated fectivity by trypsin, antibodygrowth in these [email protected] we show that multiple-cycle enhanced infection, Fc receptor growth of NWS virus occurs by the acquisition of infectivity in the presence of trypsin, and the infection is enhanced in the presence of appropriate concentrations INTRODUCTION of anti-HA mAB in addition to the presence of trypsin Macrophages (M@)are able to internalize effectively in murine primary M@. antibody-coated particles through Fc receptors and digest them [Mantovani et al., 1972; Unkeless and Eisen, 19751. Moreover, it was demonstrated that the enhanced internalization of antibody-coated viruses via Accepted for publication September 18,1991. Fc receptors results in enhanced infections under sevAddress reprint requests to Hiroshi Ochiai, Department of eral conditions [Halstead, 1982; Porterfield, 19861. Re- Virology, Toyama Medical and Pharmaceutical University, cently, enhanced infection of cultured peripheral M@ 2630 Sugitani, Toyama 930-01, Japan. Antibody-dependent enhancement (ADE) of influenza A NWS virus infection was investigated in primary murine macrophages (Mp,)using antihemagglutinin(HA1 monoclonal antibody (mAB). Contrary to previous reports of abortive influenza virus infection in primary Ma, this study demonstrated that the NWS virus replicated productively in both resident peritoneal Mp, and thioglycolate-elicited peritoneal Mp, providing cleavage of the HA was achieved by trypsin; 5 pg/ml of trypsin was the optimum concentration for the induction of infectivity. Under multiplecycle growth conditions in the presence of mAB at various concentrations in trypsin-containing media, ADE was demonstrated in both Mp, in the presence of subneutralizing concentrations of mAB. Flow cytometric analysis showed that the mechanism of virus entry into Mp, could be through HA to specific virus receptors, or HA plus antibody to Fc receptors. These results indicate that ADE of the NWS virus infection actually occurs on Fc receptor-bearing primary murine Mp, depending on the concentration of antibody in the presence of the appropriate protease for cleavage of viral HA.

0 1992 WILEY-LISS, INC.

218

MATERIALS AND METHODS Virus Influenza A NWS virus (HlN1) strain was propagated in the allantoic cavity of 10-day-old embryonated hen eggs for 48 to 72 hr a t 35°C and used for the infection. The allantoic fluids were stored in small aliquots a t -80°C after clarification at 1,OOOg for 20 min. The virus titer of the allantoic fluids was 2.0 x 10' plaqueforming units (PFUYml. Cells Resident peritoneal M# (RPM)and thioglycolate-elicited peritoneal M# (TPM) were obtained from an outbred specific-pathogen-free strain of ICR female mice aged 5 to 6 weeks (Shizuoka Laboratory Animal Center, Japan) by the method described previously [Rodgers and Mims, 19811. These peritoneal cells were suspended a t the indicated density in Dulbecco modified Eagle minimal essential medium (MEM)supplemented with 0.2% bovine serum albumin (BSA) and 0.1% glucose (serum-free medium) and incubated a t 37°C in the presence of 5% CO'. After 2 to 6 hr, adherent monolayers were washed twice with phosphate-buffered saline (PBS) and infected. In both M# preparations, more than 96% of the adherent cells were positive for nonspecific esterase, a marker enzyme for M# [Koski et al., 19761, stained specifically with esterase assay kit (Sigma). MDCK cells were grown in MEM supplemented with 10%fetal bovine serum. Anti-HA mAB, Serological Test, and Flow Cytometry Anti-HA mAB B-4 [immunoglobulin G (IgG) subclass, IgGl] was used throughout the experiments. The origin and character of this mAB were described elsewhere [Ochiai et al., 1988,19901.Both ascitic fluid containing mAB and normal mouse serum were previously treated with receptor-destroying enzyme (Denka Institute of Biological Science, Tokyo, Japan) and then diluted with PBS containing 1%BSA. Hemagglutination inhibition (HI)test was carried out by standard methods [Casals, 19671. For flow cytometric analysis to examine viral and Fc receptor-mediated attachment, fluorescein isothiocyanate (FITCblabeled virus (HA titer, 1,024) was mixed with an equal volume of mAB (HI titer, 400) or normal mouse serum (dilution, 1:40) and then incubated for 30 min at 37°C. This virus was designated AS-V or NS-V, respectively. To confirm Fc receptor-mediated attachment, FITC-labeled virus was exposed to a Fab fragment of anti-HA IgG (HI titer, 400) as above and designated Fab-V. Methods for the preparations of FITC-labeled NWS virus and Fab fragment were described previously [Ochiai et al., 19881. Adherent monolayers (3 x lo6 cells in 3.5 cm dish) were washed twice with cold PBS and infected with these viruses for 1hr on melting ice. After the inoculum was removed, the cells were collected with cell scraper, washed with cold PBS, and then applied to a flow cytometer (FACScan, CAI. Total amounts of attached vi-

Ochiai et al. ruses per lo4 cells were calculated by multiplying mean fluorescence intensity per cell by the peak area in the flow cytometer data as described previously [Ochiai et al., 19881.

Growth Assay To assay single-cycle growth, adherent monolayers (1x lo6 cells per well) in a 24-well plate were infected a t a multiplicity of infection (MOI)of 5 PFU/cell for 1hr at 37°C. The cells were washed four times with PBS and cultured in 1ml of serum-free medium in the presence or absence of trypsin (type 3, Sigma) at 37°C. At 18 hr postinfection, the culture supernatant was collected by centrifugation a t 600g for 10 min. Finally, supernatant infectivity was calculated by subtracting the unadsorbed virus titers at 0 hr postinfection from released virus titers determined by plaque assay on MDCK cells [Tobita et al., 19751. To assay multiple-cycle growth, the adherent monolayers in a 25 cm2 flask (1x lo7 cells per flask) were infected a t an MOI of 0.01 PFU/cell and incubated in 8 ml of serum-free medium in the presence or absence of trypsin. At the indicated times postinfection, 0.3 ml of culture supernatant were collected to assay the infectivity titers. After collection of the culture supernatants, fresh media were not added. To examine the antibody-mediated growth, the virus, which was previously exposed t o mAB showing 50% plaque reduction titer (dilution, 1:106) a t 37°C for 30 min, was used for the infection instead of the virus alone in the assay system for multiple-cycle growth as described above. Alternatively, the adherent monolayers (1 x lo6 cells in each well of a 24-well plate) were infected with the virus alone a t an MOI of 0.01 PFU/cell and cultured in 1ml of serum-free medium in the presence of both trypsin (5 p,g/ml)and mAB a t various concentrations. At 65 hr postinfection, the culture supernatants were collected to assay the infectivity titers. RESULTS Infectivity Titers of NWS Virus Grown in Primary Mg in the Presence or Absence of Trypsin Initially, we examined the effect of trypsin on the infectivity in the cultures of TPM under single-cycle growth conditions. At 18 hr postinfection, the supernatant of trypsin-free culture showed no detectable infectivity titer ( < 5 x 10' PFU/ml), whereas the supernatants of trypsin-containing cultures (0.3 to 20 pg/ml) showed detectable titers. The highest titer (3.2 x lo4 PFU/ml) was obtained a t 5 pg/ml. Consequently, 5 pg/ml of trypsin was used for the experiments described below. Once infectivity of the progeny viruses was acquired by trypsin, a multiple-cycle growth might occur in the presence of trypsin. Thus, time-related infectivity titers were compared between trypsin-free and cultures with trypsin under multiple-cycle growth conditions. As shown in Figure 1, no infectivity was detected in each trypsin-free culture supernatant at any time postinfec-

Infection Enhancement of Influenza Virus

219 cells per well). The data are summarized in Table I. In the absence of mAB, supernatant infectivity titers from TPM and RPM were 1.4 x lo5 and 1.8 x lo4 PFU/ml, respectively. In the presence of mAB, infectivity titers increased depending on the concentrations of mAB; no infectivities were detected a t the highest concentration of mAB (dilution, 1:lo6), whereas considerably higher titers were detected in a lower concentration range of mAB. The highest enhancement ratios (infectivity titer with/without mAB) for TPM (6.6-fold) and RPM (23.9fold) were obtained at the dilution of 1:109 and 1:10lo, respectively. Because mAB showed 50%plaque reduction titer a t a dilution of 1:106against the virus titer of 1 x lo4 PFU, these antibody dilution ranges might correspond to the subneutralizing concentrations for the virus titers released from M# under multiple-cycle growth conditions. Thus, in the presence of subneutralizing concentrations of mAB, the degree of ADE increased in both M# compared with the above experimental system. In this experimental system, ADE was also more readily demonstrable in RPM than TPM as shown by higher enhancement ratios in a wider dilution range of mAB.

107,

Mechanisms of Virus Entry Into TPM and RPM The mechanism of virus entry into Mp)was examined Fig. 1. Multiple-cycle growth in the presence of trypsin. TPM (cir- by flow cytometric analysis. As shown in Figure 2B, cle) and RPM (triangle)were infected a t an MOI of 0.01 PFU/cell and NS-V, with an HA titer of 512, could attach to both Mp) cultured in the presence (closed symbol) or absence (open symbol) of trypsin ( 5 pg/ml). At the indicated times postinfection (abscissa), 0.3 (fluorescein intensities were 18.8 and 11.2 for TPM and ml of culture supernatants were collected to assay the infectivity titers RPM, respectively), indicating that sialic-acid-containon MDCK cells. ing viral receptors are distributed on the surfaces of both [email protected] could also attach t o both Mp) (fluorescein intensities were 12.6 and 9.9 for TPM and RPM, respection. However, in the presence of trypsin, infectivity tively) in spite of the lack of HA activity (HA titer, titers in the supernatants from TPM increased loga-