THE ISOLATION OF TWO ENZYME-RIBONUCLEIC ACIDCOMPLEXES INVOLVED IN THE SYNTHESIS OF

FOOT-AND-MOUTH DISEASE VIRUS RIBONUCLEIC ACID

BY RALPH B. ARLINGHAUS AND JEROME POLATNICKPLUM ISLAND ANIMAL DISEASE LABORATORY, ANIMAL DISEASE AND PARASITE RESEARCH

DIVISION, AGRICULTURAL RESEARCH SERVICE, USDA, GREENPORT, NEW YORK

Communicated by Marshall Nirenberg, December 12, 1968

Abstract.-The foot-and-mouth disease virus-RNA polymerase complex wasreleased from membrane particulates present in the cytoplasm of infected babyhamster kidney cells. The soluble polymerase complex was fractionated byzonal centrifugation in sucrose gradients. Two polymerase complexes (RNAand protein complex) active in the cell-free system were isolated and had S-rateranges of 20-70S and 100-300S, respectively. The light polymerase complexcontained 20S double-stranded RNA; and the heavy polymerase complexcontained a polydisperse, partially RNase-resistant RNA. The cell-free productof these two polymerase complexes was analyzed by zonal centrifugation insucrose gradients. The light polymerase complex synthesized only 20S double-stranded RNA. The product of the heavy polymerase complex contained nodetectable 20S double-stranded RNA and only a peak of single-stranded RNAwith S-rate corresponding to 37S viral RNA. A third polymerase complexwas isolated with S-rate greater than 300S, and it contained a polydisperse,partially RNase-resistant RNA. This third polymerase complex synthesizedboth 37S viral RNA and 20S double-stranded RNA in the cell-free system,and it is probably the native polymerase complex still bound to cellularparticulates.

Introduction.-Studies by Eikhom et al.1 with coliphage Q13 RNA-dependentRNA polymerase have shown that it contains two components, both of whichare required for replication of the parent, single-stranded RNA. However,the role of these two components in the RNA replication mechanism has notbeen elucidated. Work with temperature-sensitive mutants by Lodish andZinder2 provides strong evidence that two enzymes are required to replicateF2 bacteriophage RNA. Progress in animal virus-RNA replication has beenhampered because of membrane components and excessive levels of nucleases.We have partially overcome these difficulties3'5 and have approached thestudy of the mechanism of foot-and-mouth disease virus (FMDV) RNA synthesisby analysis and purification of the active RNA-enzyme complex. Analysisof the FMDV-polymerase complex by centrifugation gave three RNA-enzyme components: a light component which synthesized 20S double-stranded RNA; a heavy component which synthesized 37S single-strandedviral RNA; and a third component, designated aggregate polymerase component,which synthesized both single-stranded and double-stranded RNA. Thelatter may represent the native FMDV-RNA replication unit.

Materials and Methods.-Nucleoside triphosphates and H3-NTP were purchased fromSchwarz BioResearch, Inc.; Tris, phosphoenolpyruvic acid (PEP), and PEP kinase from

821

Dow

nloa

ded

by g

uest

on

Sep

tem

ber

15, 2

020

822 BIOCHEMISTRY: ARLINGHAUS AND POLATNICK

Sigma Chemical Co.; membrane filters from Schleicher and Schuell; dextran-sulfate500 from Pharmacia; desoxycholate (enzyme grade) from Mann Research Laboratories;and sucrose from Fisher Scientific Corp.

Cells: Baby hamster kidney 21, clone-13 cells, obtained from the American TypeCulture Collection, were grown in 2-liter Baxter bottles rotated on a three-tiered rollermill as previously described.6 Experiments were carried out on 6- to 7-day-old cellswith each bottle containing from 6 to 8 X 108 cells.

Virus and infection procedure: Foot-and-mouth disease virus (type A, strain 119),passaged once in mice and 150 times in calf kidney cultures, was concentrated andpartially purified to a versene-free "aqueous phase" stage.7 Infection procedures havebeen described.8

Preparation of the membrane-free polrymerase complex: The polymerase complex(membrane-bound) was prepared from infected cells at 210 min postinfection as pre-viously described.3 All operations were done at 0-4° unless otherwise specified. Themitochondria-microsome fraction (the membrane-bound polymerase complex) from6 to 8 X 109 cells was homogenized in 12-14 ml of 0.25 M sucrose, 0.001 M MgC12.3-4 ml was layered onto each of three tubes containing 2.5 ml of 20% sucrose in 0.001 AlMgCl2, 0.01 M Tris-HCl, pH 7.5 (TM), and centrifuged 2 hr at 25,000 rpm in the SW25.1rotor at 10. The pellets were homogenized in 10-12 mil of TM and adjusted to 1 mg/mlin dextran-sulfate 500 and 0.5% in sodium desoxycholate. Each reagent was addeddropwise, and dextran sulfate was added first. Two volumes of 100% ethanol wereadded dropwise, with stirring, at -12°. The suspension was centrifuged at 10,000 X gfor 10 min at - 150, and the clear supernatant was discarded. The precipitate washomogenized in 7 ml of TM and dialyzed overnight against 1 liter of TM. The suspensionwas centrifuged at 12,000 X g for 10 min, and the supernatant was carefully removedwith a pipette. The supernatant fluid that contained the membrane-free polymerasecomplex was made 20% in glycerol and stored at -60°. The protein concentrationas determined by the method of Lowry et al.9 ranged from 4 to 8 mg/ml.

Polymerase complex assay: In a final volume of 0.7 ml, the following reagents wereadded: 10 ,smoles of Tris-HCl, pH 8.1 (230); 5 Mmoles of phospho(enol)pyruvate;20 1Ag of pyruvate kinase; 12.5 Mmoles of magnesium acetate; 25 mjAmoles each of adenosinetril)hos)hate (ATP), guanosine triphosphate (GTP), cytosine triphosphate (CTP),and uridine triphosphate (UTP); 1 Mc of H3-UTP (specific activity 1-5 curies/mmole);and 0.1-0.2 ml of the polymerase complex. The mixture was incubated at 370 for thedesired time and chilled in ice-water. One-half ml of 0.1 Al sodium pyrophosphate,300 Ag of carrier yeast-RNA, and 10 ml of 5% trichloroacetic acid (TCA) were added;and the suspension was allowed to stand for 20 min in ice-water. The precipitates werecollected on membrane filters and washed five times with 5% TCA. The membranefilter was transferred to a counting vial and prepared for radioactivity measurement.1'Samples intended for sucrose gradient analysis contained 10 Mc of H3-UTP per 0.7-ml

reaction mixture, and were not processed with the TCA wash procedure.RNA extraction: The reaction mixtures were precipitated with 2 vol of alcohol,

suspended in 2 ml of 0.1 Al NaAc, pH 5.1, containing 0.5% sodium dodecyl sulfate(SDS), and incubated for 3-5 min at 37°. The solution was layered directly on a sucrosegradient."RNase treatment of sucrose gradient fractions: 12-ml samples of each gradient fraction

were incubated with 10 ug of RNase in a final volume of 1 ml containing 0.15 M KCl,0.1 M Tris HCl, pH 7.0. The samples were incubated for 30 min at 37°, and the TCA-insoluble radioactivity was measured.

Results.-Activity of the membrane-free polymerase complex: We have recentlypublished results of experiments on the "soluble polymerase complex" fromFMDV-infected cells.4 This preparation is a mitochondria-microsome prepara-tion in which the polymerase complex and membranes were released from oneanother by detergent, although both were still present in the preparation.

PROC. N. A. S.

Dow

nloa

ded

by g

uest

on

Sep

tem

ber

15, 2

020

BIOCHEMISTRY: ARLINGHAUS AND POLATNICK

The results of that study showed that two major S-rate classes of RNA-proteincomplexes were present in the soluble polymerase complex. One was a 100-300S complex and the other a 20-70S complex. No exact determination of theS-rate limits of either complex has been made. The soluble polymerase complexwas further purified by removing membranes and the detergent (Table 1).About 50 per cent of the starting activity was lost in this purification procedurewith a two- to threefold increase in the specific activity over the crude cytoplasm.The membrane-free polymerase complex was stable when stored in 20 per centglycerol at -60° even after repeated freezing and thawing. This preparationwill be referred to hereafter as the polymerase complex. The RNA productsynthesized was similar to that made by the crude membrane-bound polymerasecomplex,3' 1 namely, 37S single-stranded viral RNA, 20S double-stranded RNA,and a heterogeneous RNA.

TABLE 1. The activity and preparation of the polymterase complex.Cpm/mg Total activity

Sample protein* (cpm)Lysate 3,400 7.4 X 105Nuclei 1,650 2.1 X 106Cytoplasm 3,903 7.1 X 106Mitochondria-microsome fraction 11,105 6.3 X 10578,000 X g mitochondria-microsome

supernatant -200 <6.0 X 10Membrane fraction 4.24 X 104Soluble polymerase complex in 20%

glycerol 7,654 3.4 X 105* Protein was determined by the method of Lowry et al.9

Sucrose gradient analysis of the polymerase complex: The polymerase complexwas pulse labeled for five minutes in the cell-free system with H3-UTP. Thetime course of its activity was similar to that of the membrane-bound complex.3The reaction mixture was chilled and layered directly on a 10-50 per cent linearsucrose gradient. After centrifugation at 15,000 rpm for 17 hours, the com-ponents of the polymerase complex were distributed in a 20-70S class, a 100-300S class, and a greater than 300S class (the pellet) (Fig. 1). These classeswill be referred to as the "light," "heavy," and "aggregate" polymerase com-ponents, respectively. The latter contained about 20-30 per cent ofthe input radioactivity. These results are similar to those found with thecrude soluble polymerase complex4 5 that contained desoxycholate, dextransulfate, membranes, and other constituents. Ribonuclease-resistant RNAwas found in all three fractions; and the light component was most resistantas was reported earlier with the soluble polymerase complex.5

It should be emphasized that polyribosomes are not detected in the polymerasecomplex either by optical density or by pulse-labeling determinations. Onlysingle ribosomes and ribosomal subunits are seen. Both the method of isolationand the storage of the polymerase complex in dextran-sulfate 500 preclude theexistence of polyribosomes associated with the polymerase complex.4

Analysis of the pulse-labeled RNA of the components of the polymerase complex:The polymerase complex was pulse labeled for five minutes in the cell-free system

VOL. 62, 1969 823

Dow

nloa

ded

by g

uest

on

Sep

tem

ber

15, 2

020

824 BIOCHEMISTRY: ARLINGHAUS AND POLATNICK

as shown in Figure 1. The peak tubesof each of the three polymerase com-ponents were treated separately with

140 S SDS to extract the RNA. The SDS-1500 extracted RNA was analyzed on 5-25

CPM per cent linear sucrose gradients (Fig.If 2), and a comparison of the three

£ 100gradients revealed that two classes ofRNA were present, a heterogeneous

|#/ and a homogeneous class. The aggre-S5oo ,WJ' gate component and the heavy com-

ponent (Fig. 2A and B) contained/ ,CPM +ENasRNA with heterogeneous S-rate which

was partially resistant to RNase.0 10 20 30 However, the heavy component yielded

higher S-rate RNA; and 55 per centTub. Number of the input RNA was found in the

FIG. 1.-Sucrose gradient analysis of the pellet (greater than 60S). The lightpulse-labeled polymerase complex. A 2.8- component contained mostly 20Sml reaction mixture was incubated with double-stranded RNA (Fig. 20, whichH3-UTP for 5 min in the cell-free system.The mixture was chilled and layered directly was completely resistant to RNase.on a 10-50% linear sucrose gradient con- Note that very little or no 20S double-taining 0.01 M Tris HCl, pH 7.5, and 0.004M MgCl2. The sample was centrifuged at stranded RNA was present in the heavy15,000 rpm at 10 in the SW25.1 rotor for 17 component (Fig. 2B) or the aggregatehr. The direction of sedimentation was component (Fig. 2A). Thus, therefrom right to left. The pellet contained copent (i a)uethus, ther29% of the labeled RNA. appears to be a double-stranded RNA-

protein complex and a heterogeneousRNA-protein complex in the native, unfractionated polymerase replication unit.

Separation of the polymerase complexes: The above experiments with the five-minute pulse-labeled polymerase complex suggest that two polymerase complexesmake up the FMDV-RNA replication unit. These two polymerase complexeswere demonstrated by fractionating the active polymerase complex on a sucrosegradient and assaying for polymerase activity across the gradient (Fig. 3).The results show that polymerase complex activity was present in the two majorS-rate classes in the gradient (namely, the heavy and light components) andin the pellet at the bottom of the tube (aggregate component). The lattercontained about 22 per cent of the activity, and the total recovery of the poly-merase activity was 93 per cent.RNA products of the polymerase components: The sucrose gradient profiles

of Figure 2 identify the nascent chain RNA (or precursor RNA) of the threecomponents of the polymerase complex. The final product of the three poly-merase components should give further information about their physiologicalfunction. Aliquots of the pellet (the aggregate component), fraction number 8(the heavy component), and fraction number 28 (the light component) of thegradient shown in Figure 3 were incubated with H3-JTP in the cell-free systemfor 60 minutes at 37°. The reaction mixtures were concentrated by alcohol

PROC. N. A. S.

Dow

nloa

ded

by g

uest

on

Sep

tem

ber

15, 2

020

BIOCHEMISTRY: ARLINGHAUS AND POLATNICK

240 240

200 37 S / 200

160 20 S 37S 20S 30 0 l 160u CPM f+

120 I 120o~~~~~~~~~M0a

IG. CPMlaeR

40 Vol-~;~> /CM+Ra.40plus RMasS plus R~ar

0 0

0 10 20 30 0 10 20 30 0 10 20 30

Tube Number

FIG. 2.-Sucrose gradient analysis of the template RNA of the polymerase com-ponents. 0.8 ml of polymerase complex was pulse-labeled for 5 min in the cell-freesystem (2.8-ml reaction mixture), layered directly on a 10-50% linear sucrose gradient,and processed as described in Fig. 1. RNA was extracted with SDS from the peaktubes of the "aggregate component" (>300S), the "heavy component" (100-300S),and the "light component" (20-70S) and RNA profiles are shown in (A), (B), and (C),respectively. The gradients were 5-25% linear sucrose gradient in 0.01 M NaAc,pH 5.1. The gradients were centrifuged for 17 hr at 24,000 rpm in the SW25.1 rotorat 10. The percentages of labeled RNA in the pellet of (A), (B), and (C) were 21, 55,and 2%, respectively. The radioactivity at the top of the tube in (C) is residualTCA-soluble substrate.

precipitation (2 vol), and the RNA was extracted with SDS at pH 5.1. Thesucrose gradient analysis of the RNA product of each polymerase componentis shown in Figure 4. In all three profiles, radioactivity was found at the topof the tube. This material is residual substrate radioactivity which is not seenif the fraction is dialyzed versus 0.2 M NaAc, pH 5.1, prior to TCA precipitationor if the original reaction mixture is concentrated by ammonium sulfate precipita-tion. The kinetics of polymerization of RNA by all three polymerase compo-nents were similar during the 60-minute period of observation with two-thirdsto three-fourths of the total polymerization occurring in the first 30 minutes.The results are striking and show that the heavy component synthesized 37Ssingle-stranded RNA (Fig. 4B). Essentially no 20S double-stranded RNAwas seen. It should be emphasized that the heterogeneous RNA3 10, 11 mayalso be present, since it is polydisperse in sucrose gradients and overlaps 37Sviral RNA.5 The light component synthesized mainly 20S double-strandedRNA with little or no 37S single-stranded RNA (Fig. 4C). Note that the20S double-stranded RNA is completely RNase-resistant. The aggregatecomponent synthesized both 37S single-stranded RNA and 20S double-strandedRNA (Fig. 4A), and it may represent the native physiologically active RNAreplication unit. This fraction sediments to the bottom of the tube whencentrifuged on a 10-30 per cent linear sucrose gradient for one hour at 24,000

VOL. 62, 1969 825

Dow

nloa

ded

by g

uest

on

Sep

tem

ber

15, 2

020

826 BIOCHEMISTRY: ARLINGHAUS AND POLATNICK

rpm. The S-rate is thusestimated to be more than

2.0 500S. However, it may bethat this component maystill be membrane-bound or

140 s still attached to some cellu-1.S. t | lar material. In addition,

2800 when the membrane-freeE polymerase complex is notaO0 stored in 20 per cent glyc-_ .0 f \ 2 A erol prior to fractionation,

U 1.0 1 S \ 2000 °V 90 per cent of the poly-

5 f iE hrme 3 merase activity is found inthe greater than 300S frac-O / t \ 1200 > tion (the pellet of Fig. 3).

0.5 ( We conclude from all theOD i results presented here that

N two polymerase complexes400 - are present in the FMDV-

0 5 RNA replicating unit. One0 S 10 15 20 225 30 synthesizes double-stranded

RNA and the other syn-Tub. Number thesizes 37S single-stranded

FIG. 3.-Polymerase-complex activity profile in a su- RNA. These two poly-crose gradient. The polymerase complex from about 1 X merase complexes may work109 cells (1.5 ml containing 4.9 mg/ml protein) was ad- in tandem in a high mole-justed to 0.004 M MgCI92 and 10% glycerol and layereddirectly on a 10-50% sucrose gradient in TM. The cular weight multistrandedsample was centrifuged for 17 br at 15,000 rpm at 10 and RNA complex, possibly sim-every other tube was assayed for polymerase activity in ilar to that described bythe cell-free system for 60 min at 37°. Bishop et al.'2

Discussion.-Spiegelman and co-workers' have isolated two protein fractionsfrom QB coliphage-infected bacteria. Neither of these two protein fractionshas been shown to have a clear-cut physiological role in Q$l RNA replication.However, both are absolutely required to synthesize QfB RNA. One componentwill synthesize poly G from a poly C primer and the other was found in uninfectedcells.The mechanism of FMDV-RNA replication was investigated by isolating

the functioning polymerase complex found in the infected cell, purifying it tosome extent, and making use of template differences to fractionate the poly-merases. Protein-RNA complexes were fractionated but not proteins them-selves. The results showed that separation of the two active polymerasecomponents was easily achieved (Figs. 3 and 4). One polymerase componentwhich synthesized 37S single-stranded RNA contained a heterogeneous, partiallyRNase-resistant RNA (Fig. 2B) which may be the replicative intermediateor Franklin structure.'3 The polymerase component which synthesized the20S double-stranded RNA contained a 20S RNA completely resistant to RNase

PRoc. N. A. S.

Dow

nloa

ded

by g

uest

on

Sep

tem

ber

15, 2

020

BIOCHEMISTRY: ARLINGHAUS AND POLATNICK

700 A AB37 S a C 700

600 600500 l-CPM

Soo ~ 05205 500osa. a.-U + ~~~~~~~37S +400 ~~~~~~~~~~~~~~~~~~40.0

~~~.400~~~~~205 I

300 103002a

a

200 CM200

CPM100 CPMR~s. 100

-B A /oe0 CPM + RNase 11 CPM + RtNa.00 10 20 30 0 10 20 30 0 10 20 30

Tube Number

FIG. 4.-The RNA product synthesized by the polymerase complex components in the cell-free system. The polymerase complex was fractionated on a sucrose gradient as shown inFig. 3 and a 1-ml aliquot of the 5-ml pellet fraction (A). Tube number 8 (B) and tube number28 (C) that correspond to the "aggregate," the "heavy," and the "light" components, respec-tively, were incubated in a 1.8-ml cell-free reaction mixture containing 125 ug/ml dextransulfate 500 for 60 min at 37°. The reaction mixtures were concentrated by alcohol precipita-tion, and the RNA was extracted with SDS. The SDS extract was layered directly on thegradient as shown in Fig. 2. The percentages of labeled RNA in the pellet of gradients A, B,and C were 15, 15, and 3%, respectively. The labeled material at the top of each gradientprofile is residual acid-soluble material not seen if the fraction is dialyzed in high salt priorto TCA precipitation, or if the reaction mixture is concentrated by ammonium sulfate precipi-tation prior to RNA extraction.

which must be incomplete double-stranded RNA (Fig. 2C). Further workis required to determine if any slight S-rate differences exist between the double-stranded RNA template and product of the light component (cf. Figs. 2Cand 4C).The evidence presented here indicates that single-stranded 37S viral RNA

synthesis can occurindependently of 20S double-stranded RNA synthesis. There-fore, it follows that two enzyme-RNA complexes are present in the FMDV-RNAreplication complex. If it is assumed that there are two unique enzymes, thenative polymerase complex may be a large multistranded double-strandedRNA complex in which both polymerases function. One polymerase may beworking on templates which are still being completed by the other. In otherwords, the incomplete nascent viral-RNA chain product of the one polymerasebecomes the template for the other polymerase. This model has recently beenput forward by the Spiegelman group.'2 Therefore, it is tentatively concludedthat the aggregate component contains RNA which is multistranded (Franklinstructure) and contains double strands forming on nascent viral-RNA chains.It may well be that the aggregate component is the RNA replicating unit andthat the purification scheme has dissociated this unit into two different polymerasecomponents, the light and heavy.

82".Q)7VOL. 62, 1969

Dow

nloa

ded

by g

uest

on

Sep

tem

ber

15, 2

020

828 BIOCHEMISTRY: ARLINGHAUS AND POLATNICK

It has been proposed by Plagemann and Swim14 that double-stranded RNAis the product of some side reaction that is not actively involved in the synthesisof viral RNA. This concept suggests that the heavy component (100-300SRNA-enzyme complex) is the true viral-RNA replication unit since double-stranded RNA is not present in the product of the heavy component (Fig. 4B)or in the heavy component itself (Fig. 2B). However, the presence of double-stranded RNA in both whole cell and cell-free synthesized virus-specific RNArequires some detailed explanations.

Dextran-sulfate 500 has been shown to be an efficient deproteinizing agentof ribosomes yielding intact ribosomal RNA.4 5. 15, 16 Since the polymerasecomplex is stored in high levels of dextran-sulfate 500, attachment of ribosomesto the polymerase complex is not likely; furthermore, no polyribosomes weredetected in the preparation. However, in vivo, the polymerase complex mayfunction together with ribosomes. Work is in progress to isolate and studya polyribosome-polymerase complex preparation.Summary.-Two polymerase-RNA complexes have been isolated from

FMDV-infected cells. One, a 100-300S component, synthesizes 37S single-stranded viral RNA. The other, a 20-70S component, synthesizes 20S double-stranded RNA. These two polymerase components are found in a greaterthan 300S polymerase-RNA complex that may represent the physiologicallyactive viral-RNA replicating unit.

We are deeply indebted to Dr. Richard Ascione and Dr. Howard Bachrach for manyhelpful discussions, to Mrs. B. Montgomery for excellent technical assistance, and toMrs. B. Miska for assistance in manuscript preparation.

Abbreviations used: FMDV, foot-and-mouth disease virus; PEP, phosphoenolpyruvic acid;TM, 0.001 M MgC12, 0.01 M Tris-HCI, pH 7.5; SDS, sodium dodecyl sulfate.

',Eikhom, T. S., D. J. Stockley, and S. Spiegelman, these PROCEEDINGS, 59, 506 (1968).2L'odish, H., and N. Zinder, Science, 152, 372 (1966).'Polatidck, J., and R. B. Arlinghaus, Virology, 31, 601 (1967).4Arlinghaus, R. B., and J. Polatnick, Science, 158, 1320 (1967).6 Arlinghaus, R. B., and J. Polatnick, Virology, 37, 252 (1969).Polatnick, J., and H. L. Bachrach, Appl. 3Microbiol., 12, 368 (1964).

7Bachrach, H. L., R. Trautman, and S. S. Breese, Jr., Am. J. Vet. Res., 25, 333 (1964).8 Polatnick, J., G. F. Vande Woude, and R. B. Arlinghaus, Arch. Gem. Virusforech., 23,

218 (1968).9 Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall, J. Biol. Chem., 193, 265

(1951).10 Arlinghaus, R. B., J. Polatnick, and G. F. Vande Woude, Virology, 30, 541 (1966).11 Arlinghaus, R. B., H. L. Bachrach, and J. Polatnick, Biochim. Biophys. Acta, 161, 170

(1968).12 Bishop, D. H. L., N. R. Pace, and S. Spiegelnan, these PROCEEDINGS, 58, 1790 (1967).18 Franklin, R. M., these PROCEEDINGS, 55, 1504 (1966).14Plagemann, P. G. W., and H. E. Swim, Bacteriol. Rev., 30, 288 (1966)."T Miyazawa, F., 0. R. Olijnyk, C. J. Tilley, and T. Tamaoki, Biochim. Biophys. Acta, 145,

96 (1967).I' Petrovic, S., J. Petrovic, and M. E. Bayer, Biochim. Biophys. Acta, 145, 193 (1967).

PROC. N. A. S.

Dow

nloa

ded

by g

uest

on

Sep

tem

ber

15, 2

020