Animal Viruses Are Able to Fuse with Prokaryotic Cells FUSION BETWEEN SENDAI OR INFLUENZA VIRIONS AND MYCOPLASMA*

Sendai and influenza virions are able to fuse with mycoplasmata. Virus-Mycoplasma fusion was demonstrated by the use of fluorescently labeled intact virions and fluorescence dequenching, as well as by electron microscopy. A high degree of fusion was observed upon incubation of both virions with Mycoplasma gallisep- ticum or Mycoplasma capricolum. Significantly less virus-cell fusion was observed with Acholeplasma laidlawii, whose membrane contains relatively low amounts of cholesterol. The requirement of cholesterol for allowing virus-Mycoplasma fusion was also dem- onstrated by showing that a low degree of fusion was obtained with M. capricolum, whose cholesterol con- tent was decreased by modifying its growth medium. Fluorescence dequenching was not observed by incu- bating unfusogenic virions with mycoplasmata. Sendai virions were rendered nonfusogenic by treatment with trypsin, phenylmethylsulfonyl fluoride, or dithiothre- itol, whereas influenza virions were made nonfuso- genic by treatment with glutaraldehyde,

Sendai and influenza virions are able to fuse with mycoplasmata. Virus-Mycoplasma fusion was demonstrated by the use of fluorescently labeled intact virions and fluorescence dequenching, as well as by electron microscopy. A high degree of fusion was observed upon incubation of both virions with Mycoplasma gallisepticum or Mycoplasma capricolum. Significantly less virus-cell fusion was observed with Acholeplasma laidlawii, whose membrane contains relatively low amounts of cholesterol. The requirement of cholesterol for allowing virus-Mycoplasma fusion was also demonstrated by showing that a low degree of fusion was obtained with M. capricolum, whose cholesterol content was decreased by modifying its growth medium. Fluorescence dequenching was not observed by incubating unfusogenic virions with mycoplasmata. Sendai virions were rendered nonfusogenic by treatment with trypsin, phenylmethylsulfonyl fluoride, or dithiothreitol, whereas influenza virions were made nonfusogenic by treatment with glutaraldehyde, ammonium hydroxide, high temperatures, or incubation at low pH. Practically no fusion was observed using influenza virions bearing uncleaved hemagglutinin. Trypsinization of influenza virions bearing uncleaved hemagglutinin greatly stimulated their ability to fuse with Mycoplasma cells. Similarly to intact virus particles, also reconstituted virus envelopes, bearing the two viral glycoproteins, fused with M. capricolum. However, membrane vesicles, bearing only the viral binding (HN) or fusion (F) glycoproteins, failed to fuse with mycoplasmata. Fusion between animal enveloped virions and prokaryotic cells was thus demonstrated.
A long-standing and unresolved problem in cell biology is the initial stages of virus-cell interaction (Rott and Klenk, 1977;Choppin and Scheid, 1980;White et al., 1983). In spite of extensive and continuous research, the detailed mechanism by which animal enveloped virions fuse with membranes of eukaryotic cells is still obscure (Rott and Klenk, 1977;White et al., 1983).
Enveloped viruses belonging to the Paramyxovirus group, such as Sendai virus, fuse with the cell plasma membranes at * This work was supported by a grant from the National Council for Research and Development, Jerusalem, Israel, and a grant from the Gesellschaft fiir Strahlenforschung, Munich, West Germany. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. neutral or basic pH values (Rott and Klenk, 1977;Choppin and Scheid, 1980;White et al., 1983). Influenza virions, on the other hand, are also able to fuse with the cell plasma membrane, but this occurs only when the pH of the incubation medium is lowered to between 5.0 and 5.2 (White et al., 1983).
The interaction and fusion of Sendai and influenza virions with animal cells require the presence of sialic acid-containing receptors such as sialoglycolipids and sialoglycoproteins (Gottschalk, 1957;Markwell and Paulson, 1980). Recently, however, we have shown that Sendai virions are able to fuse with erythrocyte membranes lacking virus receptors provided that the membrane phospholipids are exposed to the viral glycoproteins by osmotic stress  or with lipid vesicles composed only of phosphatidylcholine and cholesterol . Similar observations have been made following the interaction between influenza virions and phospholipid vesicles (Maeda et al., 1981;Doms et al., 1985). Based on these observations, it is tempting to speculate that viral envelopes are able to fuse with any biological membrane whose lipid molecules are exposed to the viral glycoproteins, without the need of any specific virus receptors. A most attractive system to study this hypothesis is the interaction between animal enveloped virions and prokaryotic cells, especially mycoplasmata. Mycoplasmata are deprived of a rigid cell wall and are known to lack any sialic acid-containing components (Razin, 1975). In addition, the composition of their membrane phospholipids can be manipulated, thus reflecting, to a certain degree, the lipid composition of the growth medium (Razin and Tully, 1970;Rottem, 1980). The recent use of fluorescently labeled virions (Hoekstra et al., 1984; and energy transfer or fluorescence dequenching methods to follow virus-membrane fusion processes allows the study of fusion between animal virions and membranes of cells other than eukaryotic cells. In this work, we have demonstrated, for the first time, fusion between Sendai or influenza virions and Mycoplasma cells. A high degree of fusion was observed with the cholesterol-requiring Mycoplasma gallisepticum and Mycoplasma capricolum cells, whereas a low degree of fusion was observed with Acholeplasma laldlawii cells, whose membranes contain low amounts of cholesterol. from Sigma. Octadecyl rhodamine B chloride (R18) was purchased from Molecular Probes.
Viruses-Sendai and influenza (A/PR/8/34) strain, H,Nl, PR8) viruses were isolated from the allantoic fluid of fertilized chicken eggs (Peretz et al., 1974;Klenk et al., 1975). Influenza A,possessing H A (virus N(H10N7)), was grown in Madin-Darby canine kidney cells and isolated as previously described for chick embryo cells (Klenk et al., 1975). The viral hemagglutinating units and hemolytic activities were determined as described previously (Peretz et al., 1974;Huang et al., 1981). If not otherwise stated, egg grown viruses possessing cleaved hemagglutinin were used.
Cells-M. capricolum, M. gallisepticum, and A. luidluwii were grown in a modified Edward's medium (Razin and Rottem, 1976). The cells were harvested at the midexponential phase of growth (Awo = 0.20-0.25), washed (12,000 X g for 10 min at 4 "C) in Solution Na (250 m M NaCl, 10 mM Tris-HC1, pH 7.4), and resuspended in the same buffer. For fusion experiments, the cells were used immediately after harvesting.
Preparation of Fluorescent Virus Particles-Sendai and influenza virions were labeled with Rls as described elsewhere . Briefly, 5-7 pl of a 6 mg/ml ethanolic solution of Rls were rapidly injected into 700 pl of Solution Na containing 1.7 mg of either Sendai or influenza virions. After 15 min of incubation at room temperature in the dark, the virions were washed in 20 volumes of Solution Na (100,000 X g for 30 min at 4 "C) and resuspended in the same buffer to give a protein concentration of 0.2-0.3 mg/ml. Under such conditions, Rls was inserted into the viral membranes at selfquenching surface density (about 3 mol % of total viral phospholipids), and its decrease was shown to be proportional to the fluorescence dequenching (Hoekstra et aL, 1984;. Incubation of Sendai and Influenza Virions with Mycoplasmata: Fluorescence Measurements-Fluorescent Sendai or influenza virions (2.0 pg of protein of Sendai virions and 6 pg of protein of influenza virions) were incubated with Mycoplasma cells in a final volume of 400 and 200 pl, respectively, under the conditions described for each experiment. In a weight ratio of 1 pg of viral protein/100 pg of Mycoplasma proteins, the virus:cell ratio is about 1:2. After incubation, 1 ml of Solution Na was added to the reaction mixture, and the degree of fluorescence (excitation at 560 nm, emission at 590 nm) of each sample was estimated before and after solubilization with 0.1% Triton X-100. The fluorescence degree obtained in the presence of the detergent was considered to represent 100% dequenching, i.e. infinite dilution of the probe (Hoekstra et al., 1984;. All fluorescence measurements were carried out with a Perkin-Elmer MPF-4 spectrofluorometer with a 520-nm high pass filter and narrow excitation slits to reduce light scattering. Protein Determination-Protein was determined by the method of Lowry et al. (1951) with bovine serum albumin as a standard.  (Razin and Tully, 1970). The cholesterol is incorporated unchanged into the cell membrane of the former two species, reaching levels of up to 50 mol % (Rottem, 1980). The view that the fluorescence dequenching observed indeed reflects fusion between the viral envelope and the Mycoplasma membrane was further supported by the results presented in Table I. As can be seen in Table I, fluorescence dequenching was observed only when hemolytic Sendai virions were incubated with either one of the two cholesterolcontaining Mycoplasma species. It has been well established that virus-induced hemolysis reflects a process of virus-membrane fusion (Maeda et al., 197713). On the other hand, incubation of nonhemolytic (ie. DTT-, PMSF-, or trypsin- capricolum cells: effect of pH. The pH of the incubation mixture was varied between 4.0 and 12.0 using 250 mM NaCl buffered with 10 mM glycine HCl, pH 4.0, sodium acetate, pH 5.0-6.0, Tris-HCI, pH 7.5, and glycine/NaOH, pH 9.0-12.0. All other experimental conditions were as described for A , except that 200 pg of cell protein were used. C, fusion of Sendai virus with M. capricolum cells: effect of temperature. The fluorescence dequenching measurements were performed at pH 7.5 as described for B, except that incubation temperatures were varied between 5 and 50 "C. treated) Sendai virions with Mycoplasma resulted in very little fluorescence dequenching (Table I). It is well established that the above treatments cause inactivation of the viral fusogenic activity (Ozawa et al., 1979;Israel et al., 1983).

Fusion of
Similarly, very little fluorescence dequenching was observed when active fusogenic Sendai virions were incubated with mycoplasmata that were pretreated with 0.1% glutaraldehyde (Table I). Thus, glutaraldehyde-fixed cells should not be susceptible to fusion with Sendai virions but should allow lipid-lipid exchange between the viral envelope and the biological membranes (Maeda et al., 1977a).
Additional direct support for the claim that the fluorescence dequenching observed is indeed due to a fusion process be-

TABLE I Requirement for hemolytic Sendai virions to allow fusion with
Mycoplasma cells Sendai virions were labeled with Rla as described under "Materials and Methods." Fluorescently labeled Sendai virions were treated with DTT (3 mM), PMSF (7 mM), or trypsin (60 pg/mg of viral proteins) as described before (Israel et al., 1983). Hemolysis was determined by incubating fluorescently labeled Sendai virions (5 pg, treated and untreated) with 2.5% (v/v) human erythrocytes for 15 min at 37 "C in a final volume of 250 p1 of 150 mM NaCI, 10 mM Tris-HC1, pH 7.4, as described before (Peretz et al., 1974). In Experiment 11, cells (1 mg of proteins) were incubated with glutaraldehyde (0.1% final concentration in Solution Na) for 30 min at 37 'C in a final volume of 1 ml. At the end of the incubation, the cells were washed three times (10,000 X g for 10 min at 5 "C) with 5 volumes of Solution Na, and the final pellet was suspended in 250 pl of Solution Na (to give about 4 mg of protein/ml). For fluorescence measurements, RIBlabeled Sendai virions (2 pg of viral protein) were incubated with 200 pg of Mycoplasma cells in a final volume of 400 pl of Solution Na. All subsequent steps and experimental conditions were as described under "Materials and Methods" and for Fig. 1

TABLE I1
Requirement for both Sendai virus envelope glycoproteins for fusion with Mycoplasma cells RSVE or vesicles bearing the viral F (F-vesicles), NH (NH-vesicles), or both F and NH (l:l, w/w; F-NH-vesicles) glycoproteins were prepared as described by Chejanovsky et al. (1986). The viral preparations (2 pg) were labeled with Rla and incubated with M. capricolurn cells (300 pg), and the extent of fluorescence dequenching (RIB dequenching) was determined as described for Fig. l for Sendai virions. PMSF-treated RSVE (RSVEPMSF) or F-NH-vesicles (F-NH-vesiclespMsp) were obtained as described for Table I  tween the viral envelope and the Mycoplasma membranes was obtained from the experiments described in Table 11. As can be seen, incubation of fluorescently labeled For HN-vesicles with M. capricolum resulted in a low degree of dequenching. I t is noteworthy that due to the presence of the viral binding protein, the HN-vesicles are able to attach to cell surfaces in the same manner as intact virions (Fukami et al., 1980). Fluorescence dequenching was observed only when fluores-?s with Mycoplasma Cells 463 cent vesicles possessing the two viral envelope glycoproteins, i.e. the F-HN-vesicles, were incubated with the Mycoplasma cells (Table 11). The results in Table I1 show that the degree of dequenching obtained following incubation with the coreconstituted F-HN-vesicles was very close to that obtained with intact virions or with reconstituted Sendai virus envelopes (RSVE). As expected, only a low degree of dequenching was obtained following incubation with nonfusogenic RSVE or F-HN-vesicles, namely with reconstituted viral vesicles treated with PMSF (Table 11).
The results in Fig. 1B show that maximum fluorescence dequenching (45%) was observed between pH 6.0 and 9.0. At pH values below 6 or above 9, a relatively low degree of fluorescence dequenching was observed (Fig. 1B). Other workers have demonstrated that the Sendai virus fusion factor is maximally activated at pH 6.0-9.0 (White et al., 1983;Chejanovsky and Loyter, 1985).
Fusion of Sendai virions with mycoplasmata is extremely dependent on the incubation temperature, as demonstrated in Fig. 1C for M. capricolum. Maximum fusion (fluorescence dequenching) was observed a t 37 "C, whereas very little fusion occurred a t 17 "C and below (Fig. 1C). In addition, a decrease in the extent of fusion was observed at temperatures above 42 "C. This is probably due to thermal inactivation of the viral fusion protein Chejanovsky and Loyter, 1985).
Electron Microscopic Observatwns of Fusion between Sendai Virions and M. capricolum-Fusion between intact Sendai virions and mycoplasmata was also demonstrated by electron microscopy using thin-sectioned (Fig. 2) preparations. The micrographs depicted in Fig. 2

show that Sendai virions ( V )
as well as Mycoplasma cells (MP) can clearly be identified in thin sections. The viral ribonucleoproteins can be distinguished in these preparations as long threads within the viral vesicle. No such structures are seen in the Mycoplasma cells whose intracellular space is filled with the darkly stained floccular material, as was reported before (Razin, 1975). Fu-  Fig. 1. After incubation, thin sections were prepared and stained with uranyl acetate as described elsewhere (Marti and Webster, 1986). a, cells incubated with PMSF-treated Sendai virions (prepared as described for Fig. 1)  influenza virus A, as described before (Klenk et al., 1975). Trypsinization of the H& virions was performed by incubating 200 pg of virus with 3 pg of trypsin (bovine pancreas, Type 111, Sigma) in a final volume of 200 pl of PBS (10 mM sodium phosphate and 150 mM NaCl), pH 7.4, for 20 min at 37 "C essentially as described before (Klenk et al., 1975). Influenza virions (6 pg of PR8, N-HA,,, and trypsinized (Tryp) N-HA,,) were incubated with 300 pg of M. capricolum at pH 5.2 (open bars) or 7.4 (closed bars) as described under "Materials and Methods." At the end of the incubation period, the degree of fluorescence dequenching was estimated as described under "Materials and Methods" and for A above. sion between the virus envelopes and the Mycoplasma cells was not observed in preparations containing nonfusogenic (PMSF-treated) Sendai virions (Fig. 2a). The same picture was obtained from preparations made of fusogenic viruses which were incubated with the Mycoplasma cells in the cold (not shown). A preliminary step in the fusion process between Sendai virions and M. capricolum appears in the thin section shown in Fig. 2b. A virus particle and a Mycoplasma cell are seen in tight contact. At the point of attachment (arrow), no clear viral or Mycoplasma membranes can be identified. Similar observations have also been found when the fusion areas between viral envelopes and plasma membranes of eukaryotic cells were examined (Toister and Loyter, 1973). Viral particles whose envelopes had completely fused with the Mycoplasma membranes are seen in Fig. 2 ( c and d ) . This is particularly clear in the electron micrograph presented in Fig. 2d. The viral envelope and the Mycoplasma membrane appear to be connected (Fig. 2d, arrow), forming one continuous bilayer.
Fusion of Influenza Virions and Mycoplasmata-Incubation of fluorescently labeled influenza virions, similar to incubation of Sendai virions, with M. gallisepticum and M. capricolum resulted in a relatively high degree (40-55%) of fluorescence dequenching, whereas incubation with A. laidlawii resulted in a much lower degree of fluorescence dequenching (10-15%) (Fig. 3A). Surprisingly, almost the same degree of fluorescence dequenching was observed following incubation at either low (pH 5.2) or neutral pH values (Fig. 3A). A low pH is required for activation of the influenza virus' infectivity as well as its ability to fuse with biological membranes (Huang et al., 1981;White et al., 1983).
The view that the fluorescence dequenching observed indeed reflects a process of virus-membrane fusion is strongly supported by the results in Fig. 3B, showing that a relatively low degree of fluorescence dequenching was obtained by incubating influenza virions, bearing HA,, with M. capricolum. Activation of the nonfusogenic HA-containing influenza virions can be achieved by trypsin digestion which specifically cleaves the HA,, glycoprotein (Klenk et al., 1975). The results in Fig. 3B show a marked increase in the degree of fluorescence dequenching following trypsinization of the HA,-containing influenza virions. As can be seen (Fig. 3B), the extent of fluorescence dequenching observed upon incubation of trypsinized HA, influenza virions with M. capricolum was very close to that obtained with the egg-grown influenza virions.
The assumption that the fluorescence dequenching observed at both pH 5.2 and 7.4 is a quantitative reflection of a virus-membrane fusion process is further strengthened by the results summarized in Table 111. Incubation of active, fusogenic influenza virions with human erythrocyte ghosts at pH 5.2, but not at pH 7.4, resulted in fluorescence dequenching; whereas with M. capricolum, an increase in fluorescence was observed at both pH values. On the other hand, a low degree  Fig.  3. Human erythrocyte ghosts were prepared as previously described (Peretz et al., 1974) and were incubated (per 250 pg of protein) with fluorescently labeled influenza virions (6 pg) for 10 min at 4 'C at pH 7.4. All subsequent steps, including adjustment of the pH of the medium, were as described for incubation of influenza virions with mycoplasmata (Fig. 3). For inactivation of the viral fusogenic activity, 400 pg of viral protein in 200 pl were treated as follows. For heat and glutaraldehyde inactivation, a virus suspension in PBS, pH 7.4, was incubated at 85 "C for 30 min and in 0.1% glutaraldehyde for 30 min at 37 "C, respectively. Inactivation by low pH was performed essentially as described before (Sato et al., 1983) by incubating a virus suspension in 0.5 M sodium acetate, pH 5.0, for 30 min at 37 "C. For treatment with NH,OH, a virus suspension in 1 M NHzOH, pH 6.5, was incubated for 30 min at 37 "C as previously described (Schmidt and Lambrecht, 1985). At the end of the incubation, the virus in the various systems was washed twice with 10 volumes of PBS, pH 7.4, resuspended in 200 pl of PBS, and labeled with Rls as described above. It should be noted that at the concentrations used, CCCP, by itself, slightly reduced the degree of fluorescence. Therefore, CCCP was also added to control systems incubated in the absence of cells to estimate its effect on the degree of fluorescence. The use of higher concentrations of CCCP was avoided in order to minimize its effect on the fluorescence measurements.
of fluorescence dequenching was observed upon incubation of inactivated influenza virions with either human erythrocyte ghosts or M. capricolum (Table 111). Influenza virions were rendered nonfusogenic by treatment with hydroxylamine (Schmidt and Lambrecht, 1985) or with glutaraldehyde or by preincubation at low pH (pH 5.2) or at high temperatures (Sat0 et al., 1983).
The results in Table I11 clearly indicate that: (a) almost the same decrease in the degree of fluorescence dequenching (percent of inhibition) was observed by incubating treated virions with either human erythrocyte ghosts or M. capricolum, and (b) the same degree of inhibition was observed following incubation of treated influenza virions with mycoplasmata either at pH 5.2 or 7.4.
The results in Fig. 4A show that almost the same high degree of fluorescence dequenching (40-45%) was obtained following incubation of influenza virions and M. capricolum between pH 6.0 and 10.0. The degree of fluorescence (fluorescence dequenching) observed at pH 5.2 was always higher than that at pH 7.4 (Fig. 4, A and B). The addition of CCCP (5 p~) markedly decreased the extent of fluorescence observed at pH values between 6.0 and 10.0, but had very little effect on that observed at pH 5.2 (Fig. 4, A and B). No effect of CCCP could be detected on the fluorescence dequenching at pH 7.4 following incubation of Sendai virions and M. capricolum (Fig. 4B).
Susceptibility of M. capricolum to the Fusogenic Activity of Sendai and Influenza Virions: Effect of Cholesterol-The results in Fig. 5 confirm previous observations (Rottem, 1980) showing that the cholesterol content of M. capricolum is highly dependent on its amount in the culture medium. As can be seen (Fig. 5), cells that were grown in a medium containing a low percentage of serum (0.5%) contained relatively low amounts of cholesterol (7.0 pg/mg of cells), whereas  (Razin and Rottem, 1976). The amount of cholesterol in the cell membranes was determined as previously described (Rude1 and Morris, 1973). A, fusion with Sendai virus. R18-labeled, intact (closed bars) or trypsinized (open bars) virions (2 pg) were incubated with 300 pg of cell protein as described for Fig. 1. E, fusion with influenza virus. RIBlabeled, intact virions (6 pg) were incubated with 300 pg of cell protein either at pH 7.4 (closed bars) or at pH 5.2 (hatched bars) as described for Table I. Also, hydroxylamine-treated virions (open bars) were prepared (see Table 11) and incubated at pH 5.2 with Mycoplasma cells as described above for intact virions.

DISCUSSION
Fusion of enveloped viruses with recipient animal cells was studied and demonstrated mainly by following the ability of such viruses either to cause infection or, alternatively, to promote cell-cell fusion and induce lysis of recipient cells. The development of energy transfer and fluorescence dequenching methods allowed the investigation, on a quantitative basis, of fusion of animal viruses either with lipid vesicles or with biological membranes lacking virus receptors (Hoekstra et al., 1984;. Utilizing these methods, the results of this work demonstrate that fusogenic animal enveloped viruses are able to fuse with the plasma membranes of prokaryotic cells. Fusion was inferred from experiments showing a relatively high degree of fluorescence dequenching following the incubation of fluorescently labeled Sendai and influenza virions with two species of Mycoplasma. Much less fluorescence dequenching was obtained with virions whose fusogenic activity had been inactivated. Sendai virions were rendered nonfusogenic by treatment with trypsin, DTT, or PMSF. It is noteworthy that PMSF-treated virions, similarly to viruses bearing the uncleaved fusion factor, are able to attach to but not to fuse with animal cells (Israel et al., 1983).
The present results also showed that fusion of Sendai virus with mycoplasmata (similarly to its fusion with animal cultured cells (Fukami et al., 1980)) required the presence of the two viral envelope glycoproteins. Membrane vesicles bearing only the Sendai virus-binding (NH) or the fusion (F) glycoprotein failed to fuse with M. capricolum. This was inferred from the experiments showing that only a low degree of dequenching was observed following incubation of Mycoplasma cells with HN-or F-vesicles, as opposed to a high degree of dequenching obtained with co-reconstituted F-HNvesicles. These results further support the view that the increase in the extent of fluorescence resulted from fusion between the virus envelopes and the Mycoplasma cells. Using different experimental systems, it has been demonstrated (Miura et al., 1982;Heath et al., 1983;Gitman et al., 1985) that the HN glycoprotein, beside being the viral binding protein, also plays an active role in the fusion process itself. Its presence, together with the F glycoprotein, was required even for fusion of Sendai virus envelopes with membrane vesicles lacking virus receptors. Had the fluorescence dequenching observed been caused by lipid transfer or by lipidlipid exchange processes, it should have also occurred following incubation of the fluorescent HN-or F-vesicles with Mycoplasma cells.
The degree of fluorescence dequenching is a quantitative measure of the extent of virus-membrane fusion Chejanovsky and Loyter, 1985); therefore, it should be inferred that about 50-60% of the virus particles in the population fused with the mycoplasmata. This percentage is very close to that observed for fusion of Sendai virions with cultured cells and was also confirmed by electron microscopy. Essentially, the same micrographs were also obtained using influenza virions (data not shown).
Enveloped virions belonging to the Orthomyxovirus group require a low pH environment to activate their fusogenic glycoprotein (Choppin and Scheid, 1980;White et al., 1983). Based on fluorescence dequenching measurements, the results of this work show that influenza virions fuse with erythrocyte membranes only at pH 5.2, but not at pH 7.4. This is expected from previous observations (Huang et al., 1981) showing that influenza virions induced hemolysis at low pH values only. On the other hand, fusion of influenza virions with mycoplasmata was observed also at neutral and high pH values. This may indicate the existence of a local, low pH environment at the outer surface of the Mycoplasma membrane, which may be attributed to a potent proton pump known to be present in these cells (Linker and Wilson, 1985). This view is supported by the experiments showing that the [H'lionophore CCCP which causes the collapse of a pH potential across biological membranes (Harold, 1982) strongly inhibits influenza-Mycoplasma fusion occurring at high pH values, although practically without any effect on the fusion observed at pH 5.2. Furthermore, under the conditions used, fusion of Sendai virions (whose fusogenic protein is also active at pH 7.4) with mycoplasmata was not affected by the addition of CCCP. Thus, it appears that a local, low pH environment may activate the fusogenic activity of those influenza virions which are in close proximity to the Mycoplasma membranes.
However, alternative explanations cannot as yet be excluded.
Evidently, the fusion observed in this work cannot be mediated by membrane-associated virus receptors because the prokaryotic cells used lack any sialic acid residue-containing components (Razin, 1975). In addition, it should be mentioned that the same degree of fluorescence dequenching was obtained following incubation of Sendai or influenza virions with untreated or neuraminidase-treated Mycoplasma cells (not shown). Hence, the process observed should be attributed to a direct association and interaction between the viral glycoproteins and the Mycoplasma membrane, especially its phospholipid bilayer. In this respect, virus-Mycoplasma fusion resembles the previously reported fusion between Sendai or influenza virions and phospholipid vesicles lacking any specific virus receptors . It is noteworthy that fusion with liposomes composed of only phosphatidylcholine and cholesterol and lacking virus receptor required the presence of two Sendai viral glycoproteins, namely the HN and F polypeptides . As opposed to fusion with cultured cells bearing virus receptors, fusion of Sendai virions with liposomes composed of only phosphatidylcholine and cholesterol was found to be a nonleaky process . The same results were observed using influenza virions (Schmidt and Lambrecht, 1985). Indeed, preliminary results in our laboratory have also indicated that fusion of Sendai and influenza virions with Mycoplasm cells does not result in the release of the Mycoplasma content as was monitored by following leakage of intracellular enzymes or [3H]thymidine (not shown). These results may indicate that under the conditions used (about one virus particle/one to two cells of Mycoplasma), the Mycoplasma cells remained intact. Recent results in our laboratory also showed that under the conditions used, fusion with Sendai virions caused inhibition in the growth rate of mycoplasmata. Experiments are under way to study whether this is due to virus-induced cell death or to a decrease in the growth rate of cells fused with virus envelopes.
The failure of Sendai as well as influenza virions to fuse effectively with cells of A. luidluwii can be attributed to the relatively low amount of cholesterol present in the membranes of these cells (Razin and Tully, 1970). The presence of a certain amount of cholesterol has been found to be obligatory to allow fusion of Sendai virions as well as other enveloped viruses with phospholipid vesicles (White et al., 1983;. The ability to change the lipid composition of the Mycoplasma cells by modifying the lipid precursors present in the growth medium (Rottem, 1980) makes these cells an exceptionally excellent tool to study the relationship between the lipid composition of biological membranes and their ability to interact functionally with fusogenic animal virions. In this work, we have altered the cholesterol content of M. capricolum by culturing these cells in a medium containing increasing amounts of serum. Our results indeed show that almost a linear correlation exists between the cholesterol content of M. capricolum and the susceptibility of these cells to fusion with Sendai and influenza virions. Our results demonstrate, for the first time, that cholesterol is required to allow fusion between enveloped viruses and biological membranes.