Transcriptive complex of Newcastle disease virus

VIROLOGY

128, 105-117

(1983)

Transcriptive I. Both

Complex

L and P Proteins

MICHINARI

of Newcastle

Are Required

to Constitute

HAMAGUCHI, TESTUYA YOSHIDA, HIROSHI NARUSE, AND YOSHIYUKI Depnrtment Nagwga

of Virdogy University

Received

Disease

Virus

an Active

Complex

KAZUO NISHIKAWA, NAGAI’

and CeU Ridogy, Cancer Research Institute, School of Medicine, Nagaya 466, Japan

Janvmy

3, 1983; accepted

March

4, 1983

Virions of Newcastle disease virus (NDV) were disrupted with Triton X-100 in the presence of high salt and nucleocapsids were isolated by ultracentrifugation. The nucleocapsids had very low transcriptase activity and contained only NP as a prominent protein constituent, the bulk of L and P proteins not being retained. The L and P proteins were isolated by sequential treatment of the virions with low- and high-salt detergent followed twice by successive chromatography on phosphocellulose column and examined for their effect on RNA synthesis in a standard transcriptase system using the nucleocapsids as template. When both L and P proteins were added to the template, the RNA synthetic activity was greatly stimulated. P protein alone could not enhance but rather suppressed the activity. L protein exhibited stimulation to some extent but due to residual small amount of P protein in both L protein fraction and the template it has not been elucidated whether L protein could function as a polymerase by itself. These results indicate that both L and P proteins are required to reconstitute a fully active transcriptive complex with a functional template. Attempts have been made to isolate intracellular transcriptive complex from NDV-infected MDBK cells and to determine the protein species involved. The active complex has been recovered neither from eytoplasmic extract obtained by hypotonic disruption nor from Triton X-100 soluble fraction of the cells. However, we could isolate the complex from an extract by double detergents (Tween 40 and deoxycholate) solubiliaation. The complex contained L, P, and NP as virus specific proteins and several cellular proteins. These results support the concept that both L and P proteins are required for NDV-RNA synthesis and suggest further that the intracellular transcriptive complex may be associated with some cellular structure resistant to Triton X-100 but sensitive to the double detergents, presumably cytoskeletal frame work. INTRODUCTION

The involvement of L and P proteins in paramyxovirus RNA synthesis has not yet been established, although several mutant studies of Newcastle disease virus (NDV) have suggested this (Madansky and Bratt, 1981a, b; Peeples and Bratt, 1982; Peeples et aL, 1982). L and NS polypeptides of VSV, analogous to L and P of paramyxoviruses, respectively, have been isolated and reconstitution studies have clearly shown that both of these would be not only necI Author addressed.

to whom

requests

for

reprints

should

be

105

essary for synthesizing RNA complementary to genome RNA but also play a role in methylation, capping, and polyadenylation of the newly synthesized mRNA (Emerson and Yu, 1975; Imblum and Wagner, 1975; Naito and Ishihama, 1976). NP may serve merely as structural element, although this protein would be required for the replication of the viral genome (Blumberg et aL, 1981). Despite their importance presumed by analogy with VSV, the assignment of specific functions of paramyxovirus L and P proteins seemsto have been hampered. Due to the presence of L in very small amount, 0042-6822183 Copyrinht All rights

$3.00

0 1983 by Academic Press, fnc of reproduction in any farm reserved

106

HAMAGUCHI

this protein was not described in earlier studies of the isolated Sendai virus (SV) transcriptive complex (Stone et aL, 1972; Marx et al, 1974), or seems to be detectable so far in small amount in the complex of SV5 (Buetti and Choppin, 1978). With respect to P protein, it was detected as a prominent component in Sendai virus and SV5 transcriptive complexes (Stone et al, 1972; Buetti and Choppin, 1978), whereas it was virtually absent in those of Newcastle disease virus (NDV) (Colonno and Stone, 1976). This latter complex seems to contain L also only in trace amount, thus lacking both of the putative protein requirements for RNA synthesis. Apparently, the procedure employed for the isolation of the complex might be inadequate since, as shown in this paper, L and P are solubilized and not retained in the complex, resulting in markedly reduced recovery of biologically active complex. Under these circumstances, it appears that even the protein species involved in paramyxovirus transcriptive complex have not yet been firmly established. Therefore, an alternative approach has been undertaken which involves the analysis of the function of nucleocapsid after limited proteolysis (Chinchar and Portner, 1981a, b). These authors were able to show that RNA synthetic activity of NDV was lost concomittantly with selective cleavage of P protein, implicating this polypeptide in the viral RNA synthesis. However, no definitive direct evidence has been so far available to substantiate that both L and P would be indeed necessary for synthesis and modification of paramyxovirus RNA. The present study was carried out to isolate the NDV transcriptive complex and determine the viral protein species involved, to isolate these proteins and reconstitute an active complex with template, and to examine the location of the intracellular transcriptive complex. The results obtained indicate (1) that L and P are indeed involved in the transcriptive complex, (2) that both of these are necessary for reconstituting a functional transcriptive complex, and (3) that the intracellular transcriptive complex is recov-

ET

AL.

ered, unexpectedly, from cytoskeletal tion of infected cells. MATERIALS

AND

frac-

METHODS

virus and cells. Newcastle disease virus strain D26 was used throughout the present study. Seed stocks were grown in llday-old chick embryos and stored at -80” until use (Nagai et aL, 1980b). For purification, the virus was concentrated by centrifugation at 20,000 rpm on a Spine0 SW 27 rotor for 1 hr. Virus pellets were suspended in STE (10 mM Tris-HCl, pH 7.4, 100 mMNaCl,5 mM EDTA) and subjected to centrifugation through a linear gradient of 20 to 50% sucrose in STE in a Spinco SW 27 rotor at 24,000 rpm for 3 hr. The virus band was collected, diluted fivefold with STE, pelleted, resuspended in a small volume of storage buffer (50 mM Tris-HCl, pH 7.4, 0.1 mM dithiothreitol (DTT), 50% glycerol), and stored at -20”. The polymerase activity was stable for several months under this storage condition. MDBK cells were grown in reinforced Eagle’s medium (REM) with 10% fetal calf serum as described previously (Nagai et aZ., 1976). RNA polymerase assay. Standard reaction mixture (50 ~1) for RNA polymerase assay of whole virions is composed of 50 mM Tris-HCl (pH 7.8 at 30”), 150 mM NaCI, 0.4 mM MnC&, 2 mM DTT, 0.5 mM each of cytidine triphosphate (CTP), guanosine triphosphate (GTP), and adenosine triphosphate (ATP), 0.1 mMunlabeled uridine triphosphate (UTP), 0.77 pM [53H]UTP (Amersham Corp., 26 Ci/mmol), and 0.05% Triton X-100. In assays of isolated viral and cellular transcriptive complexes as well as reconstituted activity, Triton X-100 was omitted and salt concentration was reduced to 100 mM. For assay of intracellular transcriptive complex, actinomycin D (5 pg/ml) was added to assay mixture. In reconstitution studies, 10 mM creatine phosphate, 80 pg/ml of creatine kinase, and 0.2 mM spermidine were added to the standard mixture and ATP was used at 1 mM. This improved assay system enhanced approximately threefold the poly-

TRANSCRIPTIVE

merase activity compared with that of the standard mixture. The reaction mixtures were incubated at 30”, and reaction was terminated by addition of cold 10% trichloroacetic acid (TCA) with 10 mM sodium pyrophosphate. 3H-Labeled products were collected on Wattmann GF/C filters and counted in a toluene-based scintillant. The background counts per minute were determined for each assay by adding same amount of test sample to the reaction system at 0” followed by immediate TCA precipitation and subtracted from the value of incorporation during incubation at 30”. The background levels in the individual experiments are given in figure legends. Preparation from wri$ed

of

transcriptive

complex

Purified virions (0.5 mg/ml) were treated with 1% Triton X100 in 10 mM Tris-HCl (pH 7.4)-2 mM DTT-10% glycerol in the presence of appropriate concentrations of NaCl or KCl. After 20 min on ice, the samples were layered onto a linear gradient of sucrose (2060% (w/v)) in 30 mM NaCl-10 mM TrisHCl (pH ‘7.4)-l mMDTT-10% glycerol and centrifuged at 37,000 rpm for 1 hr at 4’ in a Spinco SW 41 rotor. The complexes were also isolated by sedimenting through 30% sucrose onto 60% sucrose cushion in the same buffer. vi&n.

Isolaticm of L and P proteins and prep aration of templates. Purified virions (0.5

mg/ml), treated first with 1% Triton X100 in 0.15 M NaCl buffer (10 mM TrisHCl, pH 7.4, 1 mM DTT, 10% glycerol), were layered onto 30% sucrose on 60% sucrose cushion in the same buffer with 30 mM NaCl and centrifuged at 37,000 rpm for 1 hr at 4” in an Spinco SW 41 rotor to remove glycoproteins. The subviral particles concentrated on 60% sucrose cushion were further treated with 1% Triton X-100 and 1 M NaCl in the same buffer. After standing on ice for 20 min, they were centrifuged as described above. L and P proteins were solubilized into the supernatant together with large amount of M protein, whereas virus nucleocapsids were concentrated on the 60% sucrose cushion. The nucleocapsids, containing NP as a major protein component, were treated again

COMPLEX

OF

NDV

107

with Triton-1 M NaCl and pelleted. The nucleocapsids were dialized against 10 mM Tris-HCl (pH 7.4)-0.1 M NaCl-50% glycerol and used as the template. The supernatant was dialized against the 0.5 M NaCl buffer (10 mM Tris-HCl, pH 7.4, 1 mMDTT, 10% glycerol) and applied onto a phosphocellulose column (PC) (Whatmann Pll) equilibrated with the same buffer. M protein was bound to the column and isolated by elution with 1 M NaCl. The L and P proteins flowed through the column and this fraction was dialyzed against 0.1 MNaCl buffer and applied onto PC column equilibrated with the same buffer. L and P proteins were finally separated from each other by elution with a linear gradient of NaCl (0.1-0.5 M) as described in the text. Isolation of intracellular transcriptive complex. Cells were infected at an input

multiplicity of 10 PFU/cell and harvested after 14 hr of incubation at 37” in REM. After swelling in RSB (10 mM Tris-HCl, pH 7.4, 10 mM NaCl, 1.5 mM MgClz) for 20 min on ice, the cells were disrupted by homogenization and centrifuged for 10 min at 800 g. The supernatant was used as cytoplasmic extract. The pellet was washed once with 1% Triton X-100 in RSB and then resuspended in RSB containing 1% Tween 40 and 0.5% sodium deoxycholate (DOC), incubated for 5 min on ice, and centrifuged for 10 min at 800 g. The supernatant was used as a double detergent extract. Both extracts were layered onto 8 ml of sucrose gradients (20 to 60% (w/w)) in 10 mM Tris-HCl (pH 7.4)-30 mM NaCl1 mM DTT and centrifuged at 37,000 rpm for 2 hr at 4’ in a Spinco SW 41 rotor. Fractions were collected from the bottom of the tube, and aliquots were immediately determined for RNA polymerase activity. In experiments where virus specific polypeptides were examined, the infected cells were incubated in Eagle’s minimum essential medium (MEM) containing l[35S]methionine (New England Nuclear Corp., 0.5 mCi/mmol) instead of methionine from 6 to 7 hr p.i.. The cells were harvested and fractionated as described above.

108

HAMAGUCHI

ET

Pol?dacryamide gel electrophoresis. Ten percent polyacrylamide slab (160 X 1 mm) gels were used in Tris-glycine buffer with sodium dodecyl sulfate (Laemmli, 1970).

AL.

say system as described under Materials and Methods. As shown in Fig. iA, peak of the recovered RNA polymerase activity was associated with several fractions near the bottom of the tube. In parallel, the same amount of virions were disrupted with the detergent in the presence of reduced concentration of salt (0.15 M), which is optimum for in vitro polymerase assay, and fractionated similarly. As shown in Fig. lB, although the peak fractions shifted to little lighter density, significantly higher polymerase activity was recovered than in the case of high-salt disruption. Both peak fractions obtained after highand low-salt extraction were collected and their protein compositions were compared with each other by polyacrylamide gel electrophoresis. In agreement with the previous results (Scheid and Choppin, 1973), low-salt detergent disruption resulted in solubilization of only the glycoproteins HN and possibly Fl (not resolved

RESULTS

Inqflectiveness of High-Salt Detergent Method in Isolation of Active Transcriptive Complex of NDV To isolate transcriptive complex of NDV, we employed the previously reported procedure, disruption of the virions with Triton X-100 in the presence of high salt followed by isopycnic centrifugation (Colonno and Stone, 1976). Purified virions grown in chick embryo were treated with 1% Triton X-100 in the presence of 0.75 M KC1 and centrifuged through a linear gradient of sucrose (20-60s). Portions of each fraction were determined for [3H]UTP incorporation in an in vitro polymerase as-

abc

c) ‘0 Z

6

% 04

5

10 FRACTION

15

20

NUMBER

FIG. 1. Isolation of subviral particles from NDV grown in chick embryo. Virions disrupted with 1% Triton X-106 in the presence of 0.75 (A) or 0.15 (B) M KC1 were centrifuged through ZO-60% (w/v) sucrose gradients, and aliquots (10 pl) of each fraction (0.5 ml) were assayed for [aH]UTP incorporation during 3 hr incubation at 30” as described under Materials and Methods. The subtracted background level ranged between 106 and 206 cpm. Polyaerylamide gel electrophoresis of proteins, (a) and (b), were from peak fractions (m) of A and B, respectively. (c) Proteins of intact virions. The gels were stained by Coomassie blue.

TRANSCRIPTIVE

because this comigrates with P), leaving the other viral proteins in the isolated subviral particles. In contrast, high-salt disruption solubilized not only M and glycoproteins but also L and P, leaving only NP and its cleavage product NP’ in the core. This polypeptide pattern resembles closely that of the transcriptive complex previously isolated (Colonno and Stone, 1976). If one assumes that L and P may be necessary for RNA synthesis of NDV, the markedly reduced recovery of polymerase activity under high-salt conditions may be due to liberation of these two polypeptides from the nucleoeapsid. We examined then the effect of salt concentration on the recovery of active transcriptive complex. In this experiment we used MDBK-grown virus to facilitate identification of P protein (M, 56,000) on standard polyacrylamide gels; the glycoprotein F. (iI& 68,000) of the strain used is not proteolytically processed in these cells, whereas it is cleaved to Fl (M, 56,000) and F2 (Mr 12,000) in chick embryo, the former comigrating with P (Nagai et al, 1980a). In preliminary studies of mapping by partial proteolytic digestion, it was shown that the digestion patterns of P had no correspondence to any of the other virus specific proteins or actin incorporated into the virion, confirming that P is an unique abundant virus specific protein, analogous to NAP or P previously identified (Chambers and Samson, 1980; Smith and Hightower, 1981). The purified virions grown in MDBK cells were treated with Triton X-100 (1%) in the presence of 0.15,0.4, and 0.75 MKCl, and the virus cores were sedimented through 30% sucrose onto 60% sucrose cushion. Their protein composition and RNA-synthetic activity were determined. As shown in Fig. 2, the bulk of glycoproteins and a portion of M were solubilized by the detergent in the presence of 0.15 M KCl. By increasing the salt concentration, M was efficiently liberated, whereas L and P were gradually lost from the core. NP and its cleavage product NP’, as well as actin, were highly resistant and retained in the core even at the highest salt con-

COMPLEX

OF

NDV

109

HN,jHNFoNPPNP’AM-

FIG. 2. Polyacrylamide gel electrophoresis of the proteins in subviral particles isolated from [*S]methionine-labeled virions grown in MDBK cells by treatment with 1% Triton X-100 in the presence of 0.15 (Z), 0.4 (3). and 0.75 (4) M KCl. Lane 1, pattern of the same amount of intact virions.

centration. Above response of the individual proteins to salt was quantitated and summarized in Fig. 3A. RNA-polymerase activity in the isolated core was found to decrease gradually depending on the salt concentration (Fig. 3B). A comparison of the data shown in Fig. 3A and B suggests that the decrease in the activity is paralleled most likely with the release of L and P proteins from the core. From the data taken together, it was concluded that high salt-Triton X-100 treatment would be ineffective in isolation of fully active transcriptive complex of NDV. Similar results

110

HAMAGUCHI

ET

AL.

(Fig. 4B), and thus M could not be separated from the other proteins. We attempted, therefore, an alternative way of separation on a phosphocellulose (PC) column. The supernatant was dialized against 0.1 MNaCl buffer and applied onto the PC column. The flow-through material contained little virus-specific protein and L,

SAMPLE

VCDE

NO.

FIG. 3. Polypeptides and RNA polymerase activity associated with the subviral particles obtained at various salt concentrations. The amounts of L (0), P (O), NP (A), M (A), HN, (U), and Fo (m) retained in the individual subviral particles shown in Fig. 2 were quantitated by scanning the film by a densitometer (Densitron, Model PAN, Jookoo Sangyo Co. Ltd.) and their relative amounts to the intact virion were expressed by percentage (A). No. of each sample corresponds to the gel No. shown in Fig. 2. RNA polymerase activity of the subviral particles obtained under the same conditions from approximately fivefold-concentrated nonlabeled virions was shown by percentage of activity directed by control virions which showed 5174 cpm at 30” for 5 hr (B). The subtracted background levels ranged between 224 and 304 cpm.

-L-

l

-P-/ ‘NP-

P

were obtained with several other NDV strains (Miyadera, Bl and Ulster). However, Baudette C strain used by Colonno and Stone (1976) was not examined because it has been unavailable. Isolation of L and P Proteins Since P and L proteins are solubilized together with M into supernatant after high-salt detergent disruption of NDV virions, we attempted to isolate the individual proteins from the supernatant. The supernatant containing P, L, and M proteins were prepared from purified virions by two successive treatments with Triton X-100 in the presence of low and high salt as described under Materials and Methods (Fig. 4A). The supernatant was dialyzed against 10 mM phosphate buffer (pH 7.2) and centrifuged at 10,000 g for 20 min to precipitate M protein (Scheid and Choppin, 1973). However, together with M, the bulk of L and P proteins were precipitated

FIG. 4. Polyacrylamide gel electrophoresis of the isolated L and P fractions. Chick embryo-grown virions disrupted with 0.15 M NaCl-1% Triton X-100 were centrifuged to remove soluble materials. The pellet was further treated with 1 M NaCl and the detergent and supernatant (A) was obtained by centrifugation. The supernatant was then dialyzed against 10 mM phosphate buffer (pH 7.2) and the precipitate (B) was collected by centrifugation. The initial supernatant was dialyzed against 0.5 M NaCl buffer and loaded onto a phosphocellulose (PC) column. The M protein selectively bound to the column was eluted with 1 MNaCl (C). The flow-through fraction containing L and P was dialyzed against 0.1 M NaCl buffer and loaded onto a PC column again. The P and L proteins were eluted around 0.2 M (D) and 0.4 M(E) of NaCl, respectively. V, intact virions for control.

TRANSCRIPTIVE

P, and M were all bound to the column. With a linear gradient of NaCl (0.1-l M), P was eluted first and the bulk of M by 1 M NaCl, both in relatively pure form (not shown). However, L was eluted broadly across the gradient always associated with significant amount of M protein (not shown), and thus L could not be isolated. The separation of L from M has been considerably difficult and unsuccessful by other ion-exchange (CM-cellulose and DEAE-cellulose) and molecular sieve chromatography. During these attempts, it appeared that the affinity of L to M protein might depend on the salt concentration. These proteins formed complex at concentrations less than 0.2 M but were separated from each other at 0.5 M. Therefore, the supernatant was dialized, first of all, against 0.5 M NaCl buffer. When applied onto PC column equilibrated with the same buffer, the bulk of M was bound to the column and isolated by elution with 1 M NaCl (Fig. 4C), whereas L and P proteins flowed through. The flow-through fraction was then dialyzed against 0.1 M NaCl buffer and applied onto the column equilibrated with the same buffer. When elution was performed with a linear gradient of NaCl (0.1-0.5 M), P was eluted first at salt concentration around 0.2 M, whereas L eluted around 0.4 M. When the polypeptide patterns of L and P fractions (Figs. 4D, E) are compared with that of whole virion (Fig. 4V), it is clear that both fractions were greatly enriched with the respective proteins. However, they seem to be still contaminated with several proteins in trace amount. Particularly, fraction L has been always contaminated with a small amount of P on repeated experiments but attempts for further purification by ionexchange and molecular sieve chromatography as well as sucrose gradient centrifugation have been so far unsuccessful. Reconstitution NDV

of RNA Synthesis System of

The L and P fractions obtained were dialyzed against 0.1 M NaCl buffer and examined for their effect on RNA-synthetic

COMPLEX

OF

111

NDV

activity in an improved assay system as described under Materials and Methods. As template, the nucleocapsids were used which contained NP as prominant component. In order to remove as much L and P proteins as possible, the high-salt detergent extraction was repeated as described under Materials and Methods. The isolated nucleocapsids still contained small amount of L and P and consequently exhibited 10 to 15% RNA polymerase activity of the control virions disrupted by low salt detergent. Neither L nor P fraction, nor the mixture of these two fractions exhibited in vitro RNA-synthetic activity in the absence of the template (not shown). The results of a reconstitution study were summarized in Fig. 5. When P fraction was added to the template, the RNA polymerase activity remained at a low level or rather was reduced to about 60 to 70% of the background activity exhibited by the template alone. On the other hand, L fraction stim-

I

I

10

15

I 5

added

amount

I 20

25

(pl)

FIG. 5. Reconstitution of RNA polymerase activity by addition of L and P proteins to the functional template. The nucleocapsid (3 mg/ml) containing only NP as a prominent protein component was prepared as described under Materials and Methods and used as the template. To 5 ~1 of the template, the indicated volume of the isolated L (55 pg/ml, 0), P (173 *g/ml, A), M (655 pg/ml, A), or a mixture of equal volumes of L and P (0) were added and the final volume of assay mixture was adjusted to 56 ~1. RNA polymerase activity of these mixture was determined as described under Materials and Methods. The subtracted background levels were 135 cpm for the template alone and ranged between 144 and 155 cpm when the protein samples were added.

112

HAMAGUCHI

ulated the activity to some extent. However, whether or not L protein alone can stimulate RNA synthesis has not been elucidated because of contamination with P in the L fraction as well as in the templates. When both L and P fractions were added together, much more efficient and distinct stimulation was observed. The activation took place almost linearly and then reached a plateau, suggesting that there might be optimum molar ratio of L, P, and the template for maximum stimulation. At the highest activation shown in Fig. 5 (arrow), L, P, and the structural element (NP) of the template were in an approximate ratio 1:40:100. Such a ratio is approximately twofold higher with respect to L and P than that in the intact virion which gives 0.5:20:100. To determine the specificity of the RNA synthesized in our reconstitution system, the labeled products were annealed to purified NDV virion RNA. When the whole reaction mixture was extracted with SDSphenol-chloroform and then incubated under annealing conditions (Kingsbury, 1966) without addition of virion RNA, 61% of the total radioactivity (10,550 cpm) was rendered resistant to ribonuclease, presumably because of the presence of template RNA in the system. When annealed with added excess virion RNA, 92% became resistant to the enzyme. The background enzyme resistance without annealing was about 18%. It was therefore suggested that the RNA synthesized in reconstitution system might be virus specific. These results indicate that both L and P proteins would be required to reconstitute an active transcriptive complex of NDV with a functional template. Isolation of Transcriptive Complex of NDV from Infected MDBK Cells To complement the above results that both L and P may be involved in NDVRNA synthesis, we examined whether these proteins are contained in intracellular transcriptive complex. NDV-infected (m.o.i. 10 PFU/cell) and mock-infected cells were harvested at 14 hr p.i. Cytoplasmic

ET

AL.

extracts were prepared by hypotonic disruption of the cells and sedimented through gradients of sucrose as described under Materials and Methods. Assay for RNA polymerase activity in the presence of actinomycin D showed no significant peaks of virus-specific activity exceeding the level of mock-infected cells (Fig. 6A). Therefore, it appeared that the procedure might be inefficient to recover the transcriptive complex from infected cells, or that the specific activity may remain in the cells at too low a level to be detected. To investigate the former possibility, we searched the polymerase activity in the nuclear pellets which had been routinely discarded following the preparation of cytoplasmic extracts. The pellets were first washed with Triton X-100 (1%) in RSB but no detectable activity was recovered into the supernatants. Then the pellets obtained after Triton X-100 washing were further treated with double detergent solution (1% Tween 40 and 0.5% DOC in RSB), a procedure originally developed for extraction of cytoskeleton (Lenk et al, 1977), and fractionated into supernatants and pellets. The pellets contained little virus specific proteins and thus were discarded. The supernatants were sedimented through sucrose gradients and RNA-synthetic activity across the gradient was examined in the presence of actinomycin D. As shown in Fig. 6B, a significant peak of activity was formed near the bottom of the tube, which was detectable only in the infected cells. This activity was dependent on the presence of all four nucleotide triphosphates in the assay mixture. Further, the bulk of RNA synthesized by the peak fractions has been hybrydized to excess virion RNA, 79% of 18,330 cpm becoming RNase resistant. In this experiment the background enzyme resistance without annealing and the value of self-annealing were 3.5 and 28%, respectively. Near the top of the gradients, another peak of radioactivity was detected both in infected and mock-infected cells. This activity was not dependent on the four nucleotide triphosphates (Fig. 6B), and might -be similar to poly(U) polymerase

TRANSCRIPTIVE

COMPLEX

OF

113

NDV

FRACTIOHS FIG. 6. Intracellular transcriptive complex MDBK cells were fractionated into cytoplasmic described under Materials and Methods. After activity across the gradient was determined background levels ranged between 100 and 600 UTP as the only triphosphate was determined (A) cells.

of NDV. NDV-infected (0) and mock-infected (0) extract (A) and double detergent extract (B) as sucrose gradient centrifugation, RNA polymerase in the presence of Actinomycin D. The subtracted cpm. The incorporation in an assay system containing with some fractions of infected (A) and mock-infected

(Villarreal and Holland, 1974). It has to be noted that double detergents causes some 50% reduction in the virion-associated RNA polymerase activity. Thus, the intracellular transcriptive complex might have theoretically higher activity than detected here. Similar density gradient centrifugation was carried out with hypotonic cytoplasmic and double detergents extracts from infected cells labeled with [35S]methionine, and the proteins in each fraction were analyzed by polyacrylamide gel electrophoresis. Most of the glycoproteins and M protein and portion of NP were extracted into the cytoplasmic fraction distributing quite broadly across the gradient, whereas little or only trace amounts of L and P proteins were present throughout the gradient (the arrow indicate not L but a cellular protein) (Fig. 7A). As shown in Fig. 7B, exactly corresponding to the peak fractions of RNA polymerase activity (Fig. 6B), the double detergent extract formed a single sharp peak which was composed of NP, NP’, L, and P

proteins. In addition, a number of presumably cellular proteins might be concentrated to various extent in these fractions. Figure ‘7C shows a protein pattern of such a peak (fractions 7, 8, and 9 of infected cells of Fig. 6B) visualized by direct staining with Coomassie blue. The data indicate again that L, P, and NP are involved in the complex and that several cellular proteins are associated with it. NP seems to be more abundant in pulse-labeled complex (cf. Fig. 7B). These results confirm that both L and P would be involved in the transcriptive complex of NDV. In addition, it was suggested that the intracellular transcriptive complex might be associated with cytoskeletal structure. DISCUSSION

Polypeptides L and P have long been presumed to be involved in RNA synthesis of paramyxoviruses, but no direct evidence for this has been available. Using NDV, we have investigated here (1) whether these proteins are indeed contained in iso-

114

HAMAGUCHI

ET

AL.

,“No - HN

-P

>NP --A 9-M

t b

6

t t

P,L

NP Nd

J r

P

‘NP

t b FIG. 7. Polyacrylamide gel electrophoresis of cytoplasmic (A) and double detergent (B) extracts from NDV-infected MDBK cells after centrifugation through 20-60s sucrose gradients. The cells were labeled with [asS]methionine at 6-7 hr p-i. b, bottom; t, top, of the gradients. (C) Proteins of transcriptive complex isolated from fractions 7,8, and 9 in Fig. 6B and visualized by staining with Coomassie blue. V, control MDBK-grown virion.

lated complex which retains the RNA-synthesis activity and (2) whether these components can reconstitute transcriptase activity when added to a functional template.

Isolation of NDV nucIeocapsids retaining transcriptase activity was first carried out by high-salt detergent solubilization of the virions grown in chick embryo fol-

TRANSCRIPTIVE

lowed by isopycnic centrifugation (Colonno and Stone, 1976). In the present study, however, we have shown that this technique allows only markedly reduced recovery of the enzymatically active complex (Fig. 1). Although the presence of high salt is a precondition to solubilize M protein (Scheid and Choppin, 19’73), this has been accompanied with such reduced recovery of RNA polymerase activity possibly due to release of L and P proteins from the isolated core. Correlation between preservation of L and P in the viral cores and their RNA synthetic activity has been more clearly shown by using MDBKgrown virions which allow separation of P protein from fusion glycoprotein by conventional polyacrylamide gel electrophoresis and facilitate quantitation of all the six polypeptides (Figs. 2,3). In agreement with these data, an earlier report showed that the subviral NDV particles obtained by disruption with Triton NlOl alone retained RNA polymerase activity, whereas purified nucleocapsids did not (Meager and Burke, 1973). Therefore, we must conclude that high-salt detergent procedure would not be suitable for the isolation of fully active RNA polymerase complex from NDV. Although L, P, and M were solubilized by high-salt detergent treatment of the virions, the isolation of the former two proteins has been considerably difficult. The difficulty seemed to be caused not only by the presence of L protein in very low amounts on one hand and on the other the great abundance of M protein in the sample, but also by relatively strong interaction of L protein with the other two proteins, particularly with M protein. Thus, we employed a multistep process of dialysis against buffer with intermediate (0.5 M) and low (0.1 44) concentrations of salt combined with ion-exchange column chromatography, paying particular attention to remove, first of all, M protein from the solubilized sample. The selective binding of M protein to PC column in the presence of 0.5 M salt may be due to its basic isoelectric point (Chambers and Samson, 1980) and facilitated further step of isolation. Once M was removed, L and P could

COMPLEX

OF

NDV

115

be separated from each other by the second step of column chromatography following dialysis against 0.1 M salt. Both L and P proteins were required to reconstitute the maximum transcription activity with a functional template containing NP as prominent protein constituent, suggesting that both proteins may be required for NDV-RNA synthesis (Fig. 5). However, it has been yet unknown why the L fraction alone exhibited stimulatory effect. L protein may function as a polymerase by itself or alternatively, P protein contaminating the template as well as the L fraction may also result in the stimulation by L fraction. It has to be also elucidated why P protein did not stimulate but rather inhibited the activity. To answer these questions, it seems necessary to prepare completely enzyme-free and transcriptionally inactive template on one hand and on the other to obtain L protein in much purer form. In spite of the above problems to be resolved, the present results have shown that L, P, and NP existed in the reconstituted system of the highest activity attained with a molar ratio which is only twofold higher with respect to L and P than and thus not very different from that in the intact virion. In addition, the stimulated activity under such conditions has reached nearly 50% of the whole virion, when calculated on the basis of the relative template copies estimated by the amount of NP, and the products have been complementary to virion RNA. Thus the reconstitution system may reflect fairly faithfully a natural process of NDV transcription and be helpful to elucidate the role of L and P proteins. Both could interact as subunits of a heteropolymeric enzyme or each could act as independent proteins, functioning not only in the RNA synthesis complementary to the viral genome but also in its modification such as capping, methylation, and polyadenylation (Colonno and Stone, 1976). Using the above reconstitution system, we are currently investigating the effect of monoclonal antibodies directed to these proteins on such multifunctions. The involvement of L and P in NDVRNA synthesis has been also supported by

116

HAMAGUCHI

the analysis of an active complex isolated from infected cells (Figs. 6,7). In addition, we obtained results suggesting that the intracellular complex has been extracted not into cytoplasmic fraction obtained by hypotonic disruption but after solubilization of the nuclear fraction by combination of two detergents (Tween 40 and DOC). As the double detergent methods has been developed for extraction of cytoskeletal proteins (Lenk et al., 19’77), the above results suggest that the intracellular transcriptive complex may be associated with the cytoskeletal structure. It is of interest to examine whether the cellular proteins recovered with the viral complex (Figs. 7B and C) would be the constituents of cytoskeleton or not. Evidence available indicates that poliovirus replication complex with its double-stranded RNA as well as VSV mRNAs are extracted together with cytoskeleton from infected cells (Lenk and Penmann, 1979; Cervera et al, 1981). However, it seems also possible with NDV that an active transcriptive complex forms so large structure that it may be simply trapped in the cage of the cytoskeleton. In addition, the intracellular transcriptive complexes of the other paramyxoviruses could be obtained simply by hypotonic disruption (Mahy et aL, 1970; Buetti and Choppin, 1978). Thus, to elucidate possible relevance of cytoskeleton in NDV-RNA synthesis, further studies will be required. ACKNOWLEDGMENTS We are grateful to K. Kojima for his continuous encouragement and support. For valuable suggestions and continuous interest, we thank A. Ishihama, the Institute for Virus Research, Kyoto University, Japan. We also thank Y. Kuno and K. Nakamura for their excellent technical assistance. This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, and by Kudo Foundation and Gashu Foundation. REFERENCES BLUMBERG, B. M., LEPPERT, M., and KOLAKOFSKY, D. (1981). Interaction of VSV leader RNA and nucleocapsid protein may control VSV genome replication. CeU 23. 837-845. BUETTI, E., and CHOPPIN,P. W. (1977). The transcrip-

ET AL. tase complex of the paramyxovirus SV5. virdogy 82,493-508. CERVERA, M., DREYFUSS, G., and PENMAN, S. (1981). Messenger RNA is translated when associated with the cytoskeletal framework in normal and VSVinfected HeLa cells. CeU 23, 113-120. CHAMBERS, P., and SAMSON, A. C. R. (1980). A new structural protein for Newcastle disease virus. J. Ge7h Viral 50, 155-166. CHINCHAR,V. G., and PORTNER, A. (1981a). Functions of Sendai virus nucleocapsid polypeptides: Enzymatic activities in nucleocapsids following cleavage of polypeptide P by Staphylococcus aureus protease V8. Virology 109, 59-71. CHINCHAR,V. G., and PORTNER,A. (1981b). Inhibition of RNA synthesis following proteolytic cleavage of Newcastle disease virus P protein. Virology 115, 192-202. COLONNO,R. J., and STONE, H. 0. (1976). Isolation of a transcriptive complex from Newcastle disease virions. J. viral. 19, 1080-1089. EMERSON, S. U., and Yu, Y.-H. (1975). Both NS and L proteins are required for in vitro RNA synthesis by vesicular stomatitis virus. J. Fir01 15,X348-1356. IYBLUM, R. L., and WAGNER, R. R. (1975). Inhibition of viral transcriptase by immunoglobulin directed against the nucleocapsid NS protein of vesicular stomatitis virus. J. Viral 15, 1357-1366. KINGSBURY, D. W. (1966). Newcastle disease virus RNA, II. Preferential synthesis of RNA complementary to parental viral RNA by chick embryo cells. J. Mol. Biol 18, 204-214. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227, 680485. LENK, R., RANSOM, L., KAUFMANN, Y., and PENMAN, S. (1977). A cytoskeletal structure with associated polyribosomes obtained from HeLa cells. CeU 10, 67-78. LENK, R., and PENMAN, S. (1979). The cytoskeletal framework and poliovirus metabolism. Cell l&289301. MADANSKY, C. H., and BRATT, M. A. (1981a). Noncytopathic mutants of Newcastle disease virus are defective in virus-specific RNA synthesis. J. Vid 37,317-327. MADANSKY, C. H., and BRA’IT, M. A. (1981b). Relationship among virus spread, cytopathogenicity, and virulence as revealed by the noncytopathic mutants of Newcastle disease virus. J. viral. 40, 691702. MAHY, B. W. J., HUTCHINSON,J. E., and BARRY, R. D. (1970). Ribonucleic acid polymerase induced in cells infected with Sendai virus. J. viral 5. 663-671. MARX, P. A., PORTNER, A., and KINGSBURY, D. W. (1974). Sendai virion transcriptase complex: Polypeptide composition and inhibition by virion envelope proteins. J. fid 13, 107-112.

TRANSCRIPTIVE MEAGER, A., and BURKE, D. C. (1973). Studies on the structural basis of the RNA polymerase activity of Newcastle disease virus particles. J. Gen Viral. 18, 305-317. NAGAI, Y., KLENK, H.-D., and ROTT, R. (1976). Proteolytic cleavage of the viral glycoproteins and its significance for the virulence of Newcastle disease virus. Virology 72, 494-508. NAGAI, Y., HAMAGUCHI, M., MAENO, K., IINUMA, M., and MATSIJMOTO, T. (1980a). Proteins of Newcastle disease virus. A comparison by partial protease digestion among the strains of different pathogenicity. Virology 102, 463-467. NAGAI, Y., YOSHIDA, T., HAMAGUCHI, M., NARUSE, H., IINUMA, M., MAENO, K., and MATSUMOTO, T. (1980b). The pathogenicity of Newcastle disease virus isolated from migrating and domestic ducks and the susceptibility of the viral glycoproteins to proteolytic cleavage. Microbid Immurwl 24, 173-177. NAITO, S., and ISHIHAMA, A. (1976). Function and structure of RNA polymerase from vesicular stomatitis virus. J. Bid Chem, 251, 4307-4314. PEEPLES, M. E., and BRATS, M. A. (1982a). UV irradiation analysis of complementation between, and

COMPLEX

OF

117

NDV

replication of, RNA-negative tive mutants of Newcastle 41.965-973.

temperature-sensidisease virus. J. Vid

PEEPLES, M. E., RASENAS, L. L., and BRATT, M. A. (1982b). RNA synthesis by Newcastle disease virus temperature-sensitive mutants in two RNA-negative complementation groups. J. G-o/. 42, 9961006. SCHEID, A., and CHOPPIN, P. W. (1973). purification of the envelope proteins disease virus. J. Viral. 11, 263-271.

Isolation and of Newcastle

G. W., and HIGHTOWER, L. E. (1981). Identification of the P proteins and other disulfide-linked and phosphorylated proteins of Newcastle disease virus. J. Viral. 37, 256-267.

SMITH,

STONE, H. O., KINGSBURY, D. W., and DARLIGTON, R. W. (1972). Sendai virus-induced transcriptase from infected cells: polypeptides in the transcriptive complex. J. Viral 10, 1037-1043. VILLARREAL, L. P., and HOLLAND, J. J. (1974). Transcribing complexes in cells infected by vesicular stomatitis virus and rabies virus. J. Viral. 14,441450.