Effect of pH on the Protective Action of Interferon in L Cells

HALLUM, YOUNGNER, AND ARNOLD

8

7

D-J

U,

-j

6

5

4

4.5 5.0 55 6.0 6.5 7.0

pH (INITIAL)

FIG. 2. Effect ofpHrange oninterferonactioninL cells. Cultures wereincubatedfor 5 hrwith 400 units of interferon in PBSateach pH. After removalof the interferon, the cultures were washed and challenged with VSV. Virus yieldswere determinedinfluids har-vested after 6 hrof incubation of the culturesingrowth medium.

tively inhibited; atpH 6.5 apartial inhibition of the development of resistance was observed. At

pH 6.0no resistance developed. The virus yields incontrol cultures in PBS6 that lacked interferon were not significantly different from the yields from controls in PBS7 and PBS8. By contrast,

the decrease in virus yield in cultures exposed

toPBSbelowpH 6.0, withorwithoutinterferon,

resulted from the increasing toxicity for L cells of the acidified PBS solutions.

To determine whether the effects described could bereproduced withaninterferonother than that induced inL cells by NDV, similar studies

were conductedusing the inhibitor released into the circulation of mice after injection of E. coli endotoxin (endotoxin interferon). Duplicate

cul-tures were treated with 100 units of endotoxin interferon diluted in PBS6 or in PBS7 as

pre-viously described. After incubationat37Cfor 1, 3, and5hr, the cultureswerewashed, challenged with VSV, and the yield of progeny virus was determined. Control values were obtained from cultures treated with PBS at the same pH but without interferon. The results of these studies

(Table 1) showed that endotoxin interferon

be-havedsimilarlyto interferonfrom NDV-infected L cells. Resistance to challenge with VSV did not develop incultures treated with interferon in

PBS6.

TABLE 1. Comparison ofeffect of PBS6 andPBS7 on the protective

actioni

ofendotoxin-stimulated

mouse interferon in L cellsa

LogioVSV yield (PFU/ml) Diluentadded Interferon 6 hr after challenge at

toLcells dose

Ihr 3 hr 5hr

PBS-6 None 7.45 7.70 7.65

PBS-6 100 units 7.65 7.25 7.18

PBS-7 None 7.42 7.70 7.75

PBS-7 100 units 7.20 6.65 5.70

aDuplicate cultures were exposedto 100 units

of endotoxin interferon in PBS6 or in PBS7; control cultures were exposed to diluent only.

After 1,3, or5 hr at 37 C, the solutions were re-moved, the cultures were washed with growth

medium, and they were then challenged with VSV.Afterincubation for 6 hr in growthmedium at 37 C, the released virus was harvested and assayed.

The possibility that the data obtained above were the result of inactivation of interferon in PBS6 at 37 C waseliminatedbydilutingtheL-cell interferon in PBS6, PBS7, or Eagle's medium, andincubating it at 37 C or at4C for 5 hr. No

significant change

in thetiterof the inhibitor was

observed underany of these conditions.

Influence of pH

6.0 on the initial interaction

oj

interferon with L cells. A second

explanation

for the lack of

development

of resistance atpH 6.0 is,possibly, that the initial reaction between inter-feron andLcellswasblockedat this

pH.

Totest

this

possibility,

cultures

exposed

to 400 units of

interferon from NDV-infected L cells in cold PBS6, PBS7, or Eagle's medium were incubated for 15 hr at 4C. After this

time,

the interferon wasremoved and thecultures were washedthree times with cold

Eagle's

medium and transferred to 37 C. The resistance to

challenge

virus was

measured bythe

yield

inhibition method at0, 2, 4, 6, and8hrafterthetransfer of the culturesto

37 C.Theresultssummarizedin

Fig.

3show that resistance to the viruschallenge

developed

atthe

same rate

regardless

ofthe

pH

atwhich the inter-feron solution was

applied.

These data indicate that the initial interaction of interferon with L

cells takesplaceat

pH

6.0,even

though

resistance

does not

develop

under these conditions.

Maintenance of cell resistance at pH 6.0.

Another

possible

explanation

for the failure of cell resistance to

develop

at

pH

6.0 is that the metabolic products

required

toexpress the

anti-viral action of interferon

(13)

were unstable at

thispH. Ifinterferon

protection

were

dependent

upon the so-called "second

protein"

and if this material were unstable in cells

exposed

to

pH 6.0,

--- CONTROLS-NO INTERFERON _

- a INTERFERON TREATED

_ m _

F

I

774 J. VIROL.

on November 11, 2019 by guest

http://jvi.asm.org/

pH AND INTERFERON ACTION

E ~

N ---PBS-6

N -0-- PBS- 7

LL

0. ---A---GROWTHMEDIUM

z_ \

0

z

O \

-0~ ~ ~ 0

2

2 1

0 2 4 6 8

Time(HourAInGrowth MediumFollowing 4°Incubation

FIG.3.Lackof effectofpH6.0oninitialinteraction ofinterferonand L cells. L cellswereincubatedfor15 hr at 4 C withinterferon in PBS6, PBS7, or inEagle's

medium. After this period, the cultures were washed three timeswith coldmedium,3mlofEagle'smedium wereadded,and thecultureswere transferredto37 C. At the timesshown, thecultures were challengedwith VSVbytheyieldinhibition method. Virusyieldin con-trol cultures withoutinterferonwas2.1 X 107 PFU/ml.

cells in which interferon

protection

had been established would become more

susceptible

to

infectionwhenheldatthis

pH.

Totestthis

possi-bility,

L-cell cultures were inoculated with 270

unitsof L-cell NDV interferon in

Eagle's

medium and incubated for 15 hr at 37 C. The cultures were then washed with

PBS6, PBS7,

or

Eagle's

medium, covered with3 mlof thesame

solution,

andincubated

again

at37C.After 0, 4, 8, and24 hr of additional incubation at 37

C,

thecultures

were

challenged

with VSV and the

yield

of

prog-enyviruswasdeterminedasdescribed

previously.

Theresults of this

experiment, plotted

in

Fig.

4, show thatinterferon

protection,

once

established,

is

equally

stable in cells

exposed

toPBS6 orPBS7.

The results from cultures

exposed

to

Eagle's

me-diumwere

identicalto

thedata

obtained

with PBS6 andPBS7.

Influence of PBS6 onmacromolecularsynthesis in L cells. Another

possible

explanation

for the failure of cells treated with interferon at pH 6.0 to develop resistance to virus is that macro-molecularsynthesis might be inhibited at this pH, thus preventing the formation of significant amounts of "second protein." This possibility was testedby measuringthedifferencein the rate

of

incorporation

of

'4C-leucine

and 3H-uridine

into proteinand RNA,respectively, as a function of the pH of the PBS to which the cells were exposed. Replicatecultures of L cells were treated

E

D

a-0

-i

w

0

0 -J

8

7

0 4 8 t 24

Time(hours)After Tronsferto PBS-6 orPBS-7 FIG. 4. Maintenance ofcell resistance atpH 6.0. Cultures treatedfor 15 hr at 37 C with 270 units ofinter-feron in Eagle'smedium were washed andtransferred

toPBS6, PBS7, or tofreshEagle'smedium,and incu-bation was continued.Aftervarioustimes inthesemedia,

thecultures werechallengedwith VSVaccording to the

yield inhibitionmethod. TheresultswithEagle's medium

weresimilar to those in the otherdiluentsand are not shown.

0 Io

80

gr 60

<oz

,_r-ZI 40

20

az

co

OUL

0 1 2 3 4

TIME (HOURS)IN PBS-6

FIG. 5. Influence ofexposureofLcellstoPBS6on proteinand RNA synthesis. Culturesweretreated with

PBS6 or PBS7 for the intervals shown, then pulsed with 14C-leucine or 3H-uridine. At each time interval,

the trichloroacetic acid-precipitable countsin the cells exposedtoPBS6werecomparedtothecountsincells in PBS7. (Average ofcountsincorporated into PBS7 controlcultures: "4C-leucine, 1,717countsperminper culture; 3H-uridine, 524countsperminperculture.)

with eitherPBS6orPBS7 and after intervalsof 10

mi to 5 hr they were pulsed with radioactive

precursorsasdescribed above.Theresultsof these experiments, summarized in Fig.5,clearly demon-stratedthattherates of both 14C-leucineand

3H-CONTROLS: -0-PBS-6

---

PBS-7-6

-- INTERFERON TREATED:

--_-

PBS-6-5 -

PBS-7-

-->K_/

-/

-4I

-e *--C1LEUCINE INCORPORATION -0- H3URIDINE INCORPORATION

O . -0 --. -O

- *

5

775

VOL. 2, 1968

on November 11, 2019 by guest

http://jvi.asm.org/

[image:2.490.267.412.80.280.2]

HALLUM, YOUNGNER, AND ARNOLD

(I 100 0--0

990 I

I

z 60_

t-50

o

R 80 5 .

cr~ ~ ~~p

0 0) 3

50C4 EUIN

FI 40 3

at37-CUttepAl o rIDINen usdwt

O0 II

0

0 50 6 U65 70

p H

FIG. 6. Influence ofPBS atdifferent pHonl macro-molecularsynthesisin L cells. Cultureswere incubated

at 37 C at the pH shiownfior 5 hr, then pulsed with

either '4C-leucine or 3H-uridine. The trichloroacetic acid-precipitable counts werecomparedtocultures in-cubatedfbr 5 hr in PBS7. (Counts incorporatedinto PBS7 control cultures: 14C-leucine, 1,247 coun2ts per

minperculture; 3H-uridine, 1,801 counitsper miii per culture.)

uridine incorporation were decreased by about 70%0 within 20min ofexposureto PBS6.

Additional incorporation studies were carried

out to determine the effect on macromolecular

synthesis of exposure to a range ofpH. These

experiments at pH values from 4.5 to 7.0 were conducted as described for PBS6, except that a single 5-hrincubation periodat37 Cwasapplied

ateach pH. The results of these studies (Fig. 6)

show that the incorporation of both "4C-leucine and 3H-uridine was progressively reduced at

decreasing pH. In PBS5 and PBS5.5, no

signifi-cant synthesis of either RNA or protein was

observed.

DISCUSSION

The data presented show that exposure of L

cells to PBS at pH 6.0 results in a rapid and reversible inhibition of both RNA and protein synthesis. Control of macromolecular synthesis byexposuretoPBSatpH6.0 offers certain advan-tages over the useofthe usual chemical or anti-biotic inhibitors. In addition to the speed and

reversibility of the inhibition of synthesis, the treatment doesnot involve entry into the cell of

molecules orions thatare notordinarily present. The mechanism by which exposure of L cells to PBS6 depresses RNA and protein synthesis is under investigation; preliminary studies indicate

that exposure of L cells for 20 min to PBS6

causes a disaggregation of polysomes. Parallel

studies of chick embryo cell cultures showedthat

these cells are more resistant to the effects of

lowered pH than are L cells. The basis for the species and cell differences is being explored.

The results confirmed that interferon can interact fully with Lcells even though resistance

does not develop (3, 9). We used a lowered range ofpH to demonstrate a model system for separat-ing the initial interferon-cell interaction fromthe

synthetic processes essential for the development of resistance.

Previous studies of the effect of change of pH ontheaction of interferon have been reported by Gifford(6) and by De Maeyer and De Somer (1). The pH range covered in these reports was from 6.8 to 7.6. Gifford (6) concluded that in chick embryo cells the plating efficiency of the vaccinia virus challenge was altered by changes in pH, but that the action of interferon was not signi-ficantly affected. De Maeyer and De Somer (1), using a continuous line of rat tumor cells, re-ported an increase in interferon titer at pH 6.8 compared to pH 7.2. However, since both papers covered only a narrow physiological range of pH, theinhibition of interferon activity at the low pH usedinthis study was not observed.

ACKNOWLEDGMENT

This investigation wassupportedby Public Health Service research grant AI-06264 from the National Institute of Allergy and Infectious Diseases.

LITERATURE CITED

1. DeMaeyer, E., andP. DeSomer. 1962.Influence of pH on interferon production and activity.

Nature194:1252-1253.

2. Friedman,R. M.1967.Interferonbinding:thefirst step in establishment of anti-viral activity. Science 156:1760-1761.

3. Friedman, R. M., and J. A. Sonnabend. 1964. Inhibition of interferon action by p-fluoro-phenylalanine. Nature203:366-367.

4. Friedman, R. M., and J. A. Sonnabend. 1965. Inhibition of interferon by puromycin. J. Immunol. 95:696-703.

5. Fujioka, M., M. Koga, and 1. Lieberman. 1963. Metabolism of ribonucleic acid after partial

hepatectomy.J.Biol. Chem. 238:3401-3406. 6. Gifford, G. E. 1963. Effect of environmental

changes upon antiviral action of interferon. Proc. Soc. Exptl. Biol. Med. 114:644-649. 7. Hallum,J.V., andJ.S.Youngner. 1966.

Quantita-tiveaspectsofinhibition of virusreplication by

interferon in chick embryo cell cultures. J. Bacteriol. 92:1047-1050.

8. Levine, S. 1964. Effect of actinomycin D and

puromycin dihydrochloride on action of inter-feron.Virology24:586-588.

776 _{J. VIROL.}

on November 11, 2019 by guest

http://jvi.asm.org/

pH AND INTERFERON ACIION

9. Levine,S. 1966. Persistenceofactive interferon in cells washed after treatment with interferon. Proc. Soc. Exptl. Biol. Med. 121:1041-1045. 10. Lockart,R. Z., Jr.1964.The necessityforcellular

RNA and protein synthesis forviral inhibition resulting from interferon. Biochem. Biophys. Res.Commun. 15:513-518.

11. Merchant,D.J.,R.H.Kahn, andW. H.Murphy. 1960. Handbook of cell and organ cultures.

BurgessPublishing Co., Minneapolis, Minn. 12. Stinebring, W. R., and J. S. Youngner. 1964.

Patterns of interferon appearance in mice

in-jected with bacteria or bacterial endotoxin.

Nature 204:712.

13. Taylor, J. 1964. Inhibition of action ofinterferon by actinomycin. Biochem. Biophys. Res.

Com-mun.14:447-451.

14. Youngner, J. S., A. W. Scott, J. V. Hallum, and W. R. Stinebring. 1966. Interferon production by inactivated Newcastle disease virus in cell cultures and in mice. J. Bacteriol. 92:862-868.

VOL. 2, 1968

777

on November 11, 2019 by guest

http://jvi.asm.org/

JOURNALOFVIROLOGY, Aug.1968,p.778-786

Copyright @ 1968 American Society for Microbiology _{Printed in U.S.A.}Vol. 2, No. 8

Protective

Effects

of Specific Immunity

to

Viral

Neuraminidase

on

Influenza

Virus

Infection

of

Mice

JEROME L. SCHIJLMAN, MANIJEH KHAKPOUR, AND EDWIN D. KILBOURNE Division of Virus Research, Department of Public Health, Cornell University Medical College, New York,

New York 10021

Received forpublication29April 1968

Antibody specific for viral neuraminidasecanbe demonstratedin micefollowing

(i) pulmonary infection with influenza virus, (ii) immunization with

ultraviolet-in-activated influenza virus, (iii) immunization with isolated neuraminidase of

in-fluenzaA2 virus, and (iv) passive immunization withseraofrabbitsimmunized with

isolated A2 neuraminidase. Neuraminidase antibody produced by any of these

methodsexertsaprofound inhibiting effectonvirus replication in the lungs of mice

challenged with strains of virus having homologousneuraminidase protein,evenin

the absence of hemagglutinatinginhibiting antibody tothechallenge virus, and

re-sults in markedly decreased pulmonary virus titers and diminished lung lesions.

These observationssuggestthat antineuraminidase immunitymayplayasignificant

role inthe protection against influenza virus challenge observed inmice after infec-tionor artificial immunization.

It is now well established that

hemagglutinin

and neuraminidase are

antigenically

distinct proteins

of

the

envelope

of influenza virus, and

that

by

genetic

recombination

hybrid

(recom-binant) virusescanbe producedin which hemag-glutinin is derived from one parental virus and neuraminidase from the other

(7, 8,

11).

Such

antigenically

hybrid

viruses in which

hemag-glutinin and neuraminidase

proteins

from differ-ent subtypes have been

segregated

have

proved

useful in the

isolation

of neuraminidase free of demonstrable

hemagglutinin protein (11)

and in

the

production

of

specific antibody

to viral

neuraminidase

(7).

In

aneuploid

cell

culture,

in ovo, and inchick

shell-membranesystems, neuraminidaseantibody isnon-neutralizingexcept in

high concentrations,

but it

partially

inhibits virus

replication by

its effect on therelease and

yield

of influenza virus from cells(7, 14,16;R. G.Webster,W. G.Laver, and E. D. Kilbourne, submittedfor publication). In earlier studies ofmice immunized

by

infec-tion with

antigenically

hybrid

influenzaA

viruses,

equivalent

protection

was found after

challenge

with viruses which contained either the same

hemagglutinin or the same "minor"

(neuramini-dase) antigen

possessed by

the

immunizing

virus (9).

We now report a further

comparison

of the

relativeeffectiveness of specific immunitytoviral neuraminidase and

immunity

to viral hemag-glutinininprotecting mice against challengewith influenza virus infection. In these experiments, antibodytoneuraminidase wasproduced in mice (i) by infectionwithrecombinantviruses in which hemagglutinin and neuraminidase were derived from parents ofdifferent subtypes, (ii) by injec-tion of ultraviolet-inactivated preparations of the

same

viruses, (iii) by injection

of

purified

A2

neuraminidase,

or (iv) by passive immunization

with rabbit

antibody

tothe enzyme.

MATERIALS AND METHODS

Cellsandplaqueassay. Clone1-5C4,derivedfrom theWong-Kilbournevariant ofthehumananeuploid

Chang conjunctival cell line, was used for plaque-reductiontests(15).Themethods for assay ofplaques

and for plaque inhibition with antisera have been

publishedin detail(5, 15).

Viruses. Mostof the viruses employed have been described in earlier reports (9). These include Aol

NWS andA2/Jap. 305(mouseadapted) and the

fol-lowingrecombinants: unadapted andmouse-adapted

strains ofX-7, X-3,

X-lL,

and X-9 (17 passages in

mice),X-7(F1),andX-15 (hybridofA/equine1

con-taining A2neuraminidase) (6).Theantigenic

designa-tionof theseviruses(Table 1)wasmade in accordance withasystemdescribedpreviously (7, 8).

Ultraviolet inactivation ofvirus. Insomeinstances,

allantoic fluidseedviruses wereinactivatedby ultra-778

on November 11, 2019 by guest

http://jvi.asm.org/

INFLUENZA VIRUS INFECTION OF MICE

violet(UV) irradiationpriortoparenteral inoculation into mice. Fluids werecentrifuged at 8,000rev/min, dialyzed atpH 7.2overnight, and then subjectedto UVlight from a 7.5-w GElampat adistanceof 17.8 cm for 2 min. Afterthisprocedure,residualinfectivity

for eggs could be demonstrated only with undiluted

fluidsin the presence ofcortisone.Thefluidsthen were

adjustedto identical hemagglutinationtiters priorto

inoculationof mice.

Mice. Specific pathogen-free male Manor Farm mice (MF-1) 8 to 16 weeks of age wereused.

Eggs. White Leghorn chick embryos 10 to 12 days

of age were used to assess the infectivity titer of all

tissues.

Hemagglutination-inhibition tests.Hemagglutination

tests with mouse andrabbitsera were performed by

methods describedpreviously (17).

Enzyme-inhibition tests. Enzyme-inhibition tests

with mouse and rabbit sera were carried outwith a fetuin substrateaccording tomethods described pre-viously (7, 11). Measurements of inhibition of A2

neuraminidase (E) were made with X-7(FI) virus adjustedtogive optical density readings of300 to 800 at 549 nm in a Bausch and Lomb Spectronic-20 colorimeter or Beckman DU-2 spectophotometer.

Measurements ofinhibition ofAoneuraminidase (e) were made with the same substrate using an 18-hr

incubation period of X-9 (A2e) virus and substrate. Titers of antisera are expressed as the dilution at

which 50% inhibition of neuraminidase activity

occurred.

Isolation ofA2neuraminidase. Thepurificationand

disruption of X-7(F1) virus and the isolation and

elution of the neuraminidase protein from cellulose

acetatestrips were carried out bymethods described previously (11).

Virus titrations. Lungswere removed atdesignated periods afterinfection andgroundinglass tubes

ac-cording to techniques described previously (12).

Serial 10-fold dilutions of ground lung suspensions

wereinoculated into eggs, andafter40 hrof

incuba-tion allantoic fluids were harvested and were tested

for hemagglutinationwith human "O"redcells at a 1:4 dilution. Titrationendpoints were calculated in

terms ofthedilution oflung tissue infecting 50% of

thechickembryos.

Scoringofpulmonarylesions.Amodification ofthe

maximal scoremethod wasused, inwhichthe extent ofpulmonary lesions was expressed as a percentage

ofthetotallung surface (4).

Aerosolinfection ofmice. The apparatus and pro-cedures usedto generateaerosols ofinfectious virus

have been described previously (12). During the-periods in which they were inside the aerosol chamber, mice wereexposed to an estimated 10 to 100 mouse infective doses of thechallenge viruses employed.

Intranasal inoculation. In someexperiments,rabbit antiserum was delivered intranasally to mice. Mice werelightly anesthetized with ether, and three drops (0.05 ml) of the appropriate serum was delivered into the nostrilsthrough a 26-gauge needle.

Bronchial washings. Mice were killed by cervical

fracture, and the trachea was dissected free of the esophagus and surrounding connective tissue. A

TABLE 1. Antigenicdesignation ofvirusesemployed

in the presentstudy

Virus Hemagglutinin Neuraminidase Antigenic

subtype subtype designation

X-7... Ao A2

AoE-X-7(F1)... Ao A2 AoE

X-3 ... Ao Ao Aoeb

X-1L... A2 A2 A2E

X-9 ... A2

Ao

A2e

X-15... A/equine A2 Eq E

1/56

NWS... A

Ao

Aoe

Jap.305... A2 A2 A2E

a E = A2 neuraminidase.

b e = Ao neuraminidase.

no. 20adapter wasinsertedintothetracheathrougha

small holeandtied inplace bythread. Anamountof

1 mlofsterile saline (0.1% gelatin) wasinjectedinto the trachea andlungsand then wasaspiratedbackinto asterile syringe (0.4to0.8 ml wasrecovered).These

fluids were treated as undiluted bronchial washings,

and all subsequent dilutions were expressed accord-ingly.

RESULTS

Demonstration of neuraminidase antibody in mice after influenza virus infection. Mice were

infected with 100 mouse infective doses of

Ao/NWS

or A2/Jap. 305 virus or were exposed

to an aerosol of saline. Serum

specimens

and bronchialwashings from 10 mice in each group wereseparately pooled4weeks later for

antibody

determination. Hemagglutinating-inhibiting

(HI)

activity was measured against the two

infecting

virusesandagainst recombinantX-15 (equine E) virus [a virus inhibitable in

hemagglutination-inhibition

tests with

antibody

to

A2

neuramini-dase

(E)

(6)].

Enzyme-inhibiting

antibody

was measuredagainst X-7 (Fl)

(AoE)

andX-9 (A2e) recombinant

viruses,

and

plaque

size-reducing activity was titrated in human conjunctival cells infected with X-9 (A2e) or X-7

(AoE)

viruses under

antibody

in agar

overlays.

The results (Table 2) demonstrate that, following influenza virus infection

of

mice,

hemagglutinating-inhibit-ingantibody appeared in bronchial secretions, as well as in the serum, and that antibody to the neuraminidase component ofthe infecting virus waspresentintheserum.Thisneuraminidase an-tibodywasdemonstrated by inhibition of enzy-matic activity of intact X-9 (A2e) or X-7

(Fl)

(AoE)

virusandby plaque size reductionof viruses

containing

neuraminidasehomologousto the

neur-aminidase antibody. The HI activity against X-1 5 virus in sera of mice previously infected with A2/Jap. 305 virus is additional evidence of A2

VOL.

2,

1968 779

on November 11, 2019 by guest

http://jvi.asm.org/