CA1101849A - Induction of interferon production by modified nucleic acid complexes - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE

The present invention relates to the induction of interferon production in the cells of living organisms, in-cluding human beings. According to the invention, nucleic acid complexes, such as the polyriboinosinate and polycytidylate complex (rIn?rCn), are modified to yield unpaired bases (uracil or guanine) along the polycytidylate strand which render the complexes more readily hydrolyzable by nucleases present in living cells. The modified complexes retain their ability to stimulate interferon release by the cells but are rendered more vulnerable to destruction within the cells, the modified complexes being significantly less toxic than the original complexes. In addition, polyinosinate strand now has been prepared to contain 5-16% 2'-O-methyl inosinate residues, designated as (rI5-20,2'-MeI)n. The new complex (rI5-20,2'-MeI)n? rCn, exhibits 100-fold more activity than rIn?rCn as an interferon inducer in human cells.
This patent application is a divisional of Canadian patent application Serial No. 212,624 filed by Applicants on October 30, 1974 for "Induction of Interferon Production by Modified Nucleic Acid Complexes".

Description

~ACK~ROUND OF THE INVENTION
A.. Field of ~he Inven~ion The invention generally relates to therapeutic compositions of matter, methods for producing said compo-; sitions, and methods for administ:ering said composition to - living organisms, including human beings. Particularly, the several embodiments of the invent:ion allow protection of an organism against viral attack by stimulation of the cells of the organism to cause said oells to produce an antiviral protein known as interferon.
.

, .~ , .
, :
. :
,'' ~ , . .
: .

; : .
;, ~ , !
~ :: ::

' - , ~' ' . , ~ , , ' ' .
: . ' . ~ ' . : , ~' ' -', , : . ' , ' - , ,' :
'~
.
` ~; ; '.' ` ~:

B . Description of the Prior Ar t Interferon is an antiviral protein released by animal cells in response to viral infec~ion. It has long bPen known that RNA is a specific virion component which triggeri the release of lnterfexon in animal cells, both natural and synthetic double-stranded RNA's belng known to stimulate interferon production. These double-stranded RNA molecules have not found utility as chemotherapeutic agents due to the toxicities thereof, such -toxicity being related primarily to the presence of the double-helical RNA
structure. It has been recently shown that the first step in interferon induction in the cells of a liviny organism, i.e., the absorption of the nucleic aoid complex of poly-inosinic acid annealed to polycytidylic acid ~rI~ rCn), is a rapid event, thereby suggesting that an intact primary structure of the inducer complex be present only for a brief peri~d in the cell. The present invention now concludes;that, once interferon induction is triggered, the continued presence of doubl~-helical RNA is unne~essary and leads to secondary effects on the cell without increasing the magnitude of antiviral resistance~ r~he invention thus involves provision of douhle-stranded nucleic acid complexes which, while re~
taining the capability of inducing interferon production in cells, are capable of being readily hydrolyzed by nucleases in the cells. Such nucleic acid complexes are duplications of the double-stranded virus genes which promote interferon production during viral attack, but which are modified to thereby be rendered less toxic to the cells due to the ability o~ said modified complexes to be more readily destroyed within .

--2 - : ~

the cells. The compounds of this invention have been disclosed previously from the laboratories (J.Molecular Biology 70:56~[1972]).
In addition, now a new type of hypoxanthine-polynucleotides has been synthesi~ecl from a mixture of inosine 5'~diphospha~es and 2'-~me~hy}inosine 5'-dip~osphates with M. luteus polynucleotide phosphorylase~ The procedure is similar to that previously disclosed f~om the laboratories ~Biochemistry 11, 4931 [1972~). This type of hypoxanthine -polynucleo~ides contain 5-16% 2LO~methylinosine residues together with 95-84~ inosine residues correspondingly, designated as (rIs_20, 2'-MeI)n. These complexes, (rI5_20,

2'-MeI)n~rCn, are 100-fold more active than the rIn rCn as inter~eron induces to human cells. Apparently this modification of the backbone of polyinosinate strand has ;~ a pronounced enhancement effect; therefore a less amount o~ (rIs_20,2'MeI)n-rCn is needed then rIn rCn in protectlng the human cells against viral diseases.
SUMMARY OF THE INVENTION
. .
Interferon is normally produced by a living animal cell on viral attack, the double-stxanded helical virus gene triggering the production of this antiviral protein by the cells.
Th~ action of the double-stranded viral gene can be mimicked by the 1:1 complexes of polyriboinosinic and polyrihocyti~ylic acids (rIn rCn). Two generally accepted condikions for such complexes to be adequate for induction of interferon production by cells are: (1) inkactness of the double-stranded complex (with a Tm higher than the , `incubation temperature); an~, (2) adequate resistance to nucleases, i.e., to enzymatic activi-ty in the cells. It has presently been found that additional structural and con-formational requirements for the (rIn~rCn) complex exist.
Particularly, interruption of th~ rCn strand in th~ complex by unpaired bases such as uridine or guanine has less effect on the interferon induction capability of the r~sul-ting complex than does interruption of the rI~ strand. Thus, for interferon induction , the structural requirements in strand continuity and base-pairing are more stringent in the rIn strand than in the rC~ strand. Modifications to the rCn strand of ~he (rIn~rCn) complexes are thus possible with retention of interferon induction capability and sig-nificant reduction of toxicity.
The present invention provides compositions of matter wherein duplications of these double-stranded helical ~irus genes are modified to an extent whereby the ability to induce interferon production in the cells is retained but the toxicitias normally associated with such compounds are re-duced to an extent which allows their use as chemotherapeuticagents. In the first type of modification, dot~le-stranded nucleic acid complexes which normally induce cellular interferon production are modified struaturally so that the resulting molecule is readily hydrolyzed by nucleases within the cells. The modified nucleic acid complexes retain their ability to induce inter~eron production in the cells, but are rendered much less toxic than the unmodified double-stranded structures due to th~ abillty of the c~lls to hydro-lyze the modified complexes shortly after interferon production has been induced.

In t~le second type of modification, the ribosyl backbone of polyinosinic acid in the rIn rCn complex is partially replaced ~to the extent of 5-16~) by the ; 2'0-methylribosyl residues. Th~ modified complexes (rI5_20, 2'MeI)n-rCn were found to be 100 fold more active than rIn rCn as inducer for in~erferon in the human cells.
Toxicity reduction in the fixst type of modified (rIn rCn) complexes is accomplished by s~xuctural modification of said complexes, a general modification being ~he intro-10 duction ofa nucleotide into the rCn strand to prevent normalpairing between the strands, thereby to produce a "weak point"
or chemical position on the strand which is vulnerable to attack by nucleases in the cells. While the modiied complex ;~
retains s~ficient structural integrity to induce interferon production in cells, this vulnerability to enzymatic activity allows relatively rapid hydrolysis thereof withln the cells.
Thus, the modified complex structure is destroyed after its function as an interferon inducer is served, destruction of the complex preventing;harmful effects which would be ~aused thereby were said complex allowed to remain in the cell. Since the continued presence of this double-stranded, helical complex does not increase antiviral resistance in the cell once in-duction has been triggered, nothinc3 is lost by destruction of thè complex,the reduction in toxicity to the cell and to the organism allowing its use as a therapeutic agent for protection of an organism against viral attack.
In the second type of modified (rln rCn) complexes, the rI~ strand is replaced by a (rI5_20, 2'MeI)n~strand. This new complex, (rI5_20, 2'MeI)n rCn~ is 100 fold more active 30 than rI~ rCn; thereore, a much lower dosage of this modifled complex is needed for the same pxotection against viral attack ; to the human cells.

It is therefore an object o the invention to ` provide compositions of matter ancl a method for protecting against viral infection in a cell by stimulating the pro-duction of interferon by the cell.
It is another object of the invention to provide a method for structuring, on a molecular scale, double~
stranded, helical nucleic acid complexes to render said ~-; complexes capable of inducing intérferon production in an organism while being non-toxic to the organism in therapeutic ~ -10 dosages.
It is a further object of the invention to provide therapeutic agents comprised of (rIn rCn) complexes which are rapidly hydrolyable by enzymatic activity in living cells after induction o~ interferon production therein.
It is an additional object of the invention to pro-vide therapeutic agents comprised of (rIs_20, 2'Mel)n-rCn ;
which may be many-fold more active than rIn~rCn as human ~ ~
interferon inducer. ~;
Further objects and advantages of the invention will 20 become more readily apparent from the following detailed d~scription of the preferred embodiment of th~ invention.
BRIEF DESCRXPTION OF THE DR~WING5 . _ _ .. . .... .
` Fig. 1 is a graphic illustration of an elution pattern~
from a Sephade~ G50 column (2.5cm x 95 cm~ of a hydrolysate of r~C20,G~n by RNase Tl, the designations Fr. 1~6 indi-catiny "fractions 1 through 6i';
Fig. 2 is a graphic illustratisn o melting curves ; of rI~ rCn~ r(I39~U)n rCn, and r(I2l U)n r~n in minimal Eagle's salt solution, nucleotide concentration bein~ 4 x 10 5M in 30 1 + C;

8~9 Fic3. 3 is a yraphic illustration of melting curves of rIn r(C22,U)~, rIn r(C13/U)n~ rIn r~C7~U)n' rIn'r(C4 U)n, and rIn r(C20,G)n in minimal Eagle's salt solution, nucleotide concentration bei:ng 1 x 10 M in I ~ C;
Fig. 4 in a graphic illustration of melting curves of ~Ip)gI rCn (Nucleotide concentxation 1 x 10 4M)and `~ (Ip)l2I-rCn pGly-L-lysine (initial nucleotide concentration 2 x 10 5M, P/N ratio - 1) in minimal Eagle's salt solution, and (Ip)l6I rCn (nucleotide concentration 1 x 10 3MJ in 0.15m NaCl, 0.OlM MgCl~, and 0.OOlM sodium phosphate (pH7.4)j :
Fig. 5 is a yraphic illustration o~ melting curves of (1) rIn r(Cp)4~G>P, (2) rIn r(Cp)35G~P, (3) rIn X(CP)23G~P' and (4) rIn r(Cp)llG~p, in minimal Eagle's salt solution, nucleotid~ concentration being 1 x 10 M in I ~ C;
Fi~. 6 is a graphic illustration of the hydrolysis by pancrease RNase of ta) rIn rCn, (b) r(I21,U)n rCn, (c) rIn r(C4 U~n and th~ir complexes with poly-~-lysine and poly-D-lysine, the polynucleotide complex alone being re~
presented ~y the line X-X, its complex with poly-L-lysine ::
by the line ~ ~~ (P/N ratio = 2) and 0-0 (P/N ratio = 1), its complex with poly-D-lysine by the line ~ (P/N ratio = 2) and ~-0 (P/N ratio = l);
~ 'ig. 7 is a graphic illustration of the hydrolysis by pancrease RNase of rIn'rtCp)48G?p, the polynucleotide complex ~. ;
, alone being represented by the llne X~X, its complex with poly-L-lysine by the line 0-~ (P/N ration = 1), in minimal Eagle's ~alt solution, nucleotide concentration being 1 x 10-5M in I + C; and, Fig. 8 is a yraphic illustration of the hydrolysis 30 o~ r(I~n~r(Cjn, xepresented by the line ~-~ , r~I~n, repr - :
sented by the line a-o, and r(I)n'r(c20lG)n~ represented ~ -7~

8~

by the line ~-0 , with S~g/ml. of RNase Tl and pancreatic RNase A at 37C in Eagle's medium.

DE5CRIPTION OF THE PREFERRED EMBODIMæNTS
The present compositions of matt~r and methods relative to their use generally depend in the several embodiments there-of on the chemical modification of an inter~eron inducing nucleic acid complex to render said complex less toxic to a living animal cell. The chemically modified complexes dis-closed herein retain the capability of the unmodifiecl complexes ~or inducing the production of interfexon by cells while being rendered more readily hydxolyzable by nucleases in the cells relative to the unmodified complexes. The increased hydrolysis capability o the modified complexes renders said complexes less toxic to the cells due to the relatively rapid destruction of the complexes after interferon induction.
The double-stranded, helical nucleic acid complexes whlch are of concern in the present invention may be modified by mismatching of bases, thereby causing a "looping-out" from the helix at the point or points thereon where the bases are caused to be unpaired; by strand-interruption formation of oligomer-polymer complexes; and by disposition o~ hydrocarbon grvups, such as a methyl group, within the helical structure.
In the f.irst of these particular modifications, pairing of ; bases in the two strands of the complex is interrupted to~pro-duce a "weak" point or vulnerable chemical position in the complex which is~subject to attack by nucleases in the cells.
Unpaired bases such as uridine or guanine interrupt either of the strands of a nucleic acid compIex, such as the rC strand of the complexes ~f polyriboinosinic and polyrikocyticylic acids ~poly rI-rC or rIn rCn, which is polyinosin:ic acid annealed to polycytidylic acid)l without interferring with interferon induction capability (~hereby preserving antiviral function).
However, this interruption permits accelerated hydrolysis of the complex, thus reduction in toxicity thereof, because of the formation of a weakened position in the structure of the com-plex which is more subject to chemical attack and hereby hydrolysis by nucleases in -the cells. Similarly, bond break- -age in the do~le~stranded helical structure effected by strand-interruption formation of oligomer-polymer complexes or by disposition of hydrocarbon groups in the structure provides sites in the molecular structure of the complex which are more readily susceptible to hydrolyzing enzymatic attack.
The nucleic acid complex, rIn rCn, noted above proves particularly valuable as a basic structure for modification into low toxicity interferon inducers. Data to be presented hereinafter show that, for interferon induction, the struc-~tural requirements in strand continuity and base-pairing are more stringent in the rIn strand than in the rC~ strand.
2~ Thus, modified rIn rCn complexes wherein the rC~ strand is interrupted prove most useful. Since it has been found that the triggering o~ human cells for interferon production through the absorption of rIn rCn is completed within a few minutes, the intact primary structure of the inducer complex is required to be present in the cell only briefly. Once the induction has been triggered, the continued presenceof ~e double-stranded helical rIn rCn is not necessary and leads to harmful effects without increasing antiviral resistance. Accordin~ to this kinetic factor, modi~ication of the rIn rCn complex to a more `` `

readily hydrolyzable form as previously described does not reduce antiviral resistance. Two modified complexes of rIn-rCn which have been found to retain the interferon in-duction capability of the unmodiied complex but which are re~atively rapidly hydrolyzed by nucleases within the cells n ( 12-13'U)n and rIn r(~20_29~G)n where U represents uridine and G represents guanine. Other complexes so struc-tured include rIn-r(C22,U)n and rIn r(C7, )n of and characteristics of these modifi~d complexes will be described hereinafter.
Exposure of an organism to massive dosages of either the unmodified rIn rCn or the modified complexes results in toxic ~ffects due to the inability of the organism to effective-ly dispose of~said substances within a sufficiently short period of time. ~owever, those dosages normally associated with therapeutic prac~ice are rendered harmless with the modified complexes through enzymatic hydrolysis in the cells much more readily than in the unmodified rIn rCn. Modif-ication - of the backbone of the nucleic acid complexes can even produce greater induction capability as is evidenced by the (rI6_19,mI)n-Cn complex which exhibits 100-fold moxe activity than rIn rCn as an interferon inducer in human cells. Effective dosages of these modi~ied complexes vary depending on the interferon in-duction capability and hydrolysis rates thereof, a range of - from 1 to 100 ~ gm/kilogram of body weight being safely and effectively administered, for example,to mice.~ Such con-centration of the unmodified rIn rCn complex are highly toxic.
Interruption of the acid strands in these-nucleic acid complexes produce imperfect, ox modified, complexes having Tm values substantially higher than 37C~ the modified complexes being protectable from nucleases by complex formation f~

with polylysine without im~airment of the induction ability thereof. Addition of poly L-(or D-) lysine to rI~I oligocytidylate complexes, e.g., rIn~(Cp)23G~p, or to the modified complexes specificaLly pointed out hereinabove rendered said complexes resistant to enzymatic destruction without a~fecting antiviraL activity.
The followiny m thods, procedures, and materials have been used to conduct portions o the work resulting in the present invention, a discussion of which is provided for a better understanding of the practice of the invention.
Source, Composition, and Preparation of Substances Used Enæymes and nucleoside diphosphates: Polynucleotide phosphorylase ~Micrococcus luteus, lyophilized powder) was purchased from PL Biochemicals, Inc., Milwaukee, Wisconsin.
Bovine pancreases RNase, RNase Tl and Escherichia coli alkaline phosphatase were obtained from Worthington Bio-chemical Co., Freehold, New Jersey, IDP (trisodium salt)~
and UDP ~trisodium salt) were obtained from Miles Laboratories, ElXhart, Indiana. CDP (trisodium salt), CDP (trilithium salt) and GDP (disodium salt) w~re purchased from PL Biochemicals, Inc., Schwarz BioResearch, Inc., Orangeburg, New York, and from Calbiochem, Los Angeles, California, respectively.
The preparation of 2'-O-Methylinosine 5'-diphosphates or related 2'-O-alkyl nucleotides have been recently disclosed in a paper cited in Biochemistry, ~1972) 11, 4931 entitled "A Novel Procedure For the Synthesis of 2'-O-Alkyl Nucleotides", by I Tazawa et al.
Polynucleotides rIn and rCn were purchased from Miles Laboratories, Elkhart, Indiana. The maximum molar extinCtion coefficients of 10,100 (in 0 005~ sodium acetate, pH 6 0) and 6300 (in 0-01 M-sodium phosphate, pH 7~5l were used for rIn and rCn, respectively. r(I3g,U)n, r~I21lU)nt r~C22lu)n~
r~C4,U)n and r(C~0 G)n were prepared in the laboratory by enzymic polymerization of nucleoside diphosphates using polynucleotide phosphoryIase. The reaction mixture con-tained nu~leoside diphospha~es t40 mM), 0 15 M-Tris HCl~pH 8~2), 10 mM-MgCl2, 0~4 mM-EDTA and M.luteus polynucleotide phos-phorylase ~2 mg/ml. of the reaction mixture). After incubation at 37 for 5 to 7 hr, ~ l vol. of 5% sodium dodecyl sulfate and 0-1 vol~ of 10% phenol were added to ~he r~action mix- -ture, which was shaken for 5 min. Crystalline phenol (approximately 1 g/Sml. of the mixture) was added to the mixture, which was shaken viyorously. The aqueous layer was separated by contrifugation, transferred to another container, and was treated with phenol once more. The final aqueous layer was collected and dialyzed successively against 50 ~M-NaC1/5 mM-EDTA, 5 mM~NaCl/0.5 mM-EDTA and distilled water. AEter dialysis, the polymer solution was free from the nucleoslde diphosphates, as determinad by paper chromatography. The polynucleotides in aqueous solution were stored at -17C.
Base compositions o the co-polymers were determined by hydro-lysis of the polymer to its constituent nucleotides, followed by conversion to the corresponding nucleosides by E. coli alkaline phosphatase~ The nucleosides were separated by paper chromatography and quantitated by u.v. absorption. The polymers were hydrolyzed by 0 3 M-KOH except for r(C20,G)n which was hydrolyzed by a mixture of pancreatic I~ase and RNase Tl. The following solvent systems were used for paper chromatography; isopropanol/water, 7~3 for copoly(I,U)'s;

1 n-butanol/Eormaldehyde/wa-ter~ 77 : 10 : 13 for co-poly (C,U)'s; isopropanol/ammonia/water, 7 : 1 : 2 for r(C20,G) The molar ex-tinction coefEicien-ts of r(I3g~U) and r(I21,U)n ~ere assumed to be the same as that of rI .

Likewise, the rCn value was used for r(C22,u)n, r(C13,U) and r(C20 G) . E~tinction coe~icients of r~C7,U) and r(C4,U) have been determined to be 6700 and 6800 (in 0 01 -M-Tris HCl, pH 7 5), respectively, by phosphorus analysis.
Table 1 summarizes the preparation, base composition and sedimentation coefficlent of the copolymers prepared.

~.

' ~ .

~ ~13-, ,; . ,. :

u~ ~
:
~o ~
o~q ~
~ ~ ~ o~ co ~ ~ o ~: a) ,~ O
tn o a~

h L. r-l r-l ~1 ~ T-l r-l ~1 O ~1 ~ ) 1` 0 r-; ~ L, a) ~ C~ O~n~ ~) o ~ . ` ~ ~.
V O I~ 11 11 11 1111 11 r-l D D P ~ P V
L~ O td O ~ r-J r~ ) rl ~ (d S I
¢l r >~ : u~ u~
~ ~a) ~ , C~ ~ ô~ ~ 0 03 ~ o ~ s- F~
~ A-- ~ ~ ~ ~ F ~ ~ h ~ ~1 .,~

~ ~0) ~ ~ ô o ~ r~
r~ c,~ ~ ) o o o o ~o o ~ ~:
~ O o ~ ~
h F~3 ~ _ ~ _ o P~ ~ a a a a aa a ~d r~
u~ c) P P P P 3 P C~ ,a) ,~
.~ + + + + -~I N O O 03 co O O O
. ~ ~ O O 1` ~` O O o _~
.~ ~ ~ L~ _ _ _ , ~) ~ u~rl ~ ~a c) ca)~ a ca a ~ ~

~ rd, - :

- ~ D D P~ ^ C!~ u~ H
~1 cn r-l N ~ ` ` O t~l --a~ ~ t~ 1 ~ ~ ~ Q) ~ H H t.) C) C.) ,) _, r~ :
., O r~ r~ r~ r~ r~ ~ r~ * 0~
O O O O O O O ~H
; ~ O

': .
:

The preparation of the poly 2'-O-methylinosinic i5 al50 similar to that disclosed in Biochemistry ~1972) 11, 4931, entitled "A Novel rocedure For The Synthesis of 2'~0-alkyl nucleotides", by I Tazawa. However~ in this procedure, an attempt has been made to make a hybrid con-taining ~oth residues, i.e., the residues of inosinic as well as 2'-O-methylinosinic, both r2sidues being synthesised together in the same strand in the new type of polynucleotides.
In this case, a certain proportion of both substrates are put into the mixture, i.e., certain proportions of inosinic 5'-diphosphate and certain proportions of 2'-O-methylinosinic 5'-diphosphate were put into the enzyme reaction mixture.
Incubation followed quite closely as that reported in Biochernistry~1972) 11, 4931, entitled "A Novel Procedure For The Synthesis of 2'-C-alkyl nucliotides", by I. Tazawa, for the production of poly Z'-O-methylinosinic and as quoted in this embodiment for the synthesis of pGly 2'-0-inosinic.

Oligoinosinates Oligoinosinates were prepared by controlled alkaline hydrolysis of rIn, followed by treatment with HCl and alXaline phosphatase. They were isolated by DEAE-cellulose column chromatography and were characterized. The following extinction coefficients were used for the oligomers: 11,400 for ~Ip)2I, 10,800 for (Ip)5I, 10,400 for (Ip)gI, and 10,200 for ~Ip)l6I.

Degradation of r(C20,G)n with RNase T
About 600 optical density units (at 268nm) of r~C20,G)n were incubated with 350 units o RNase Tl at 37C

: for 3 hr in 12 5 ml. of 0 05 M-Tris EICl (pH 7-5)/lmM-EDTA.
Af-ter incuba-tion, the reaction mixture was lyophilizedr dissolved in 1 ml. of water and applied to a column of SephadexR G50 (2-5 cm x 90 cm). The column was eluted with water, the elution profile being shown in Figure 1.
Six fractions were arbitrarily selected, as shown in the Figure. Fractions 5 and 6 had similar u.v. spectra to cytidine and guanosine, respectively and were not characterized further. ~ractions 1 to 4 all showed u.v. spectra similar to that of poly C(~m~x268 nm in 0 01 M-Tris HCL, pH
7 5). The general formula of (Cp) G~p was assigned to these four fractions, considering the specificity and the amount of RNase, Tl usad. Incubation of these fractions with a mixture of pancreatic RNase, RNase Tl and E. coli alkaline phosphatase gave cytidine and guanosine exclusively. The average chain-length o~ these fractions was determinecl from the cytidine :

guanosine ratio, and is shown in Table 2. The following , . . .
extinction coefficients were arbitrarily chosen for fractions , , :
1 to 4, considering &'s of oligocytidylates and the presence -of one guanylate residue pex molecule: fraction 1, 6600;
fraction 2, 6800; fraction 3, 7000; fraction 4, 7?00, in 0- 1 M-Tris HC1, p~ 7 5, at room temperature.

: . ' ., :

; ~

t ~
' _~BLE 2 ~verage Chain Length and Base Composition o~
the various Fractions of (CpjnG~p obtained by RNase T1 hydrolysis o~ poly (C20,G)*
. . .: ,, F`rac-tion Cytidine Guanosine C:G ratio A~erage (nmoles)(nmoles) chain length _ , 1 2212 47 2 ~6-9:1 ~9 1 1753 35-6 ~9-2:1 ;
2 1405 38-~ 36 3:1 36~1 , 1559 46 5 33-5:1 1399 62-1 22-5:1 24-1 : 1312 55~7 23~6:1 ~
1001 86~5 11-6:1 :
4 979 86~6 11-3:1 12-4 __ :, *Experiments were carried out in duplicate. ~-. ~

:
,:
' ' ~:

` : --15~-! ~B;

B~9 Formation of poly-and/or oliqonucleotide complexes rIn was mixed with rCn, co-poly(C,U) 15, r(C20,G)n and (Cp)nG~p at room tempe~ature in buffer A to form the co~plexes; likewise, rCn was mixed with co~poly(I,U)'s and oligo(I)'s. Since (Ip)2I and (Ip)5I do not form complexes with rCn at room temperature, their mixtures with rCn were kept at about 4C before experiments. All mixtures were kept at least 3 hr at appropriate temperatures to ensure complete complex formation before experiments. Stoichiometry of the mixing was always 1 : 1, with respect to inosine and cytidine residues and concentrations of the complexes were . ~ expressed in terms of the I C residues.

Formation of the complexes between polylysines and the poly-andJor oligonucleotide complexes Pxeformed polynucleotide complexes in buffer A were added dropwise to ~n equal volume of the polylysine solutions in the same media, with stirring. Polynucleotide and polylysine concentrations wexe adjusted to give the desired ~ ~;
value upon mixing equal volumes of the solutions. At the concentration level of 1 to 5 x 10 5 M-residue of poly-.
nucleotide, formation of 1 : 1 complex with poly-L-lysine or poly-D-lysine at room temperature does not lead to a significant amount of precipitation. This was indicated by the ~bsence of tuxbidity ~elevation of absorbance at 320 nm) and little loss (less than 10%) of material after centrifugation at 3000 g or 15 min. However, such a solution always tuxns cloudy and begins to absorb strongly at 320nm, during the melting process, at elevated temperature, and upon cooling. It was concluded that the complex between single-30 stranded polynucleotide and polylysine is even less soluble :
~ -16-than the complex of polylysine/r (I) n- r (C) n~
For the pxeparation o.f r(Ip)l2I.rCn~poly-L-lYSir1e t~rnary complex, r(Ip~l2I rCn was made in 2 m-NaCl/O Ol ~q-NaPO4/0~ 001 M-Mgcl2 and was kept at about 5C for 1 hr.
Thsi solution was added to the cooled poly-L-lysine solution in the same solvent. This solution was c1ialyzed at 4C
successibely against 0 Ol M-NaPO4/0~ OOl M-MgCl2 containing 1 M NaCl (for 4 hr), 0 5 M-MaCl ~for 4 hr), 0 25 M NaCl (for 10 br) and i~11y 0~15 M-NaC1 (fo~ 6 hr).

''`

.:~

'~ ' ' ' :~

:~ .
:

~ :~ .
* ~
~:~i QI I I I I I I I I I I I 1 ~ 1 ~ I I V I I I ,~ .
~ 1 ~n u~
~1 O H ~ I O O I O O I I I I I I I 03 0 O ~ I ~Q O

3 ~ ~ ~
a) . .
~ - -~ :~
h . .~
5~ _ . ~.
,~ ~ ~ ,.
O ~ X ~ ~ -~: I I I I i I I ~ ,1 o C~ X ~ ~9 ~d ~ . u~
o ~ a) (U lc O O O O O O O O O O c) ' '~ ' H~l ~ ~m x x ~c x x I I I I x x x I I I I I x I I x s~u~ ~ ~ ~ ~ I o o r~ o ~ ~1 ~
O ~ ~ X ... ., .,, . , ~ ..
r~ ~ ~ S~ ~1 ~) ~

~ ~ h ~ ~ X X I ~'Xl I I X X X
1~1 .~
.
" ~ ~ ~ ~ .
Q,ol o`P r~I` 00 0 1` ~ O I 1~ 0 ~ ~1 1` Ll-) I I I I I ~1 1 1 - Ql ~ r o~) ~) ~1 ~`I co ' C3~1 ~ X ~1 : a) rol ~ ~ ' .
Q ~ * . C~) ~¢ O O O ~ 1 ~ o o o o o o o o o co oo o r-1 ~) .H h X ,1 ~ o ~1 ~) ~ ,1 ~ I I ul I o o o o o o ~ ~n o r-l ~ ~ .
.~ O U~ . .~
O ~ u~ ~ ~Ln ~r~ r~ n~ ~ u~ m~r ~u~ ~r~ ~ ~ o H--' -- I I I I I I I I I I I I I I I I I I I I I
O ~ o o o o o o o o o o o o o o o o o o o o o h U~ XXXXXXXXXXXXXXXXXXXXX
O ~ ~ 1 .' ~ . , ~
~> U U O C~ ^
~C ~ h 5~ P P ~ P P Q
Q) 1~ ) ~ t'l E3 ~ p ~ ~ ~ p ~.) U O o ~ ) U ---U O U w O ~ ~ ~ ~ ~ O~
C) ~ ~ ~ ~ r~ a H H H H H H ; ~ ~ ~ HS ~H~ H H H H H H H H H O *
h SI ~ h :.

u~
Q) o ~1 ~H
o a rlJ5~
rl)a).c V
d a) H1: --I
O ~
~H :~
~H ~1 ~) U7 ~rl ~d 3 u~ ~
a) ~
$
td a) S~
s~
o d ~ 5~ X
3 ~H O a) O
5~ rd ~
u~ ~ O
tO h ,.1 C) 5 ~
d :~ rd ~:
O ~
~1 ~ 0 ~ X t~
n s~
a) o o ~H S I rl ~1 o~o ,~: 3 ~rl ~
d ,~
~-r1 r~r~ ~
r~ r~ t~
LI "
$ ~
~; V ~ ~ td -O
--- ~d ~ ~ ,~
~ ~ O
r~ ~
c~ -rl r~ O
o a ~ td a) x ~: 0 td W~r~ 3~r ..

. . .

8~91 1 ~Iydrolysis of Nucleic ~cid Complexes and the Ternary Complexes with Polylysines by Pancreas RNase rIn r~nr r(I21~U)n~rCn, rIn r(C4,U)n, rIn (CP)48 and their complexas wi-th I- and D-polylysine were prepared as mentioned above. The nucleotide concen-tration used was lx10-5~, and the ratio of phospha-te (nucleotide) to nitrogen ~lysine)was 1 : 1. The low nucleotide concentration was employed to avoid precipitation of the complex by polylysine. Bovine pancreas RNase was dissolved in distilled water to a concen~ration of 1 mg/ml.; 60 ~
of the enzyme solution was then added to 36 ml. of the complex solution and mixed well. The mixture was kept at room tem-perature; its absorption spec-tra was recorded at appropriate time intervals using a cell with a 10-cm path-length.
Experiments have been done to ensure that the lack of increase of absorbance of the polynucleotide complexes ; with polylysine in the presence of the nuclease is a reliable indicator that the polynucleotides remain intact. This was done by addition of saturated (NH4)2SO4 solution to a final concentration of 20% saturation. Under this condition, the nuclease activity is prohibited and the polylysine-poly-nucleotide complex dissociates. No increase in absorbance of the polylysine-po~nucleotide complex (1:1) after incubation with the enZyme due to the addition of (NH4)2SO4 was found (after adjustment for dilution). Addition of (NH4)2SO4 ;~ did not change the increase of absorbance caused by poly-nucleotide degradation after incubation with enzyme in the absence of polylysine.
~ 30 ~.

Physico-Chemical Properties ~ bsorp-tion spectra were measured in a well-known fashion.
The Tm was defined as the temperature at which half of the total op-tical change occurred. Sedimenta-tion coefficients were determined on a Spinco model E analytical ultracentrifuge e~uipped with a photoelec-tric scanner.
The concentration of these samples was adjusted to give 0-5 to 0 8 optical density unit a-t 265 nm.

Biological Studies Solutions Buf~er A is 0 15 M NaCl, 0 01 M-sodium phosphate (pH 7 2), 0 001 M-MgC12. This buffer was used for both physical and biological studies so that the results could be compared directly. Minimal Eagle medlum was prepared to contain either 12% or 6~ ~etal cal~ serum as specified, and glutamine (2mM), penicillin G (200 units/ml.? and streptomycin (200 ~g/ml/) wera aaded just before use. Saline -D is 0 14 M-NaCl, 0.0017 M.Na2~PO4, 0-0054 M RCl, 0-0023 M KH2PO~ and 0-6mg of phenol red per ml. Trypsin ~0 2%) was prepared in saline D to contain 0 005 M.ethylenediamine-tetra acetic acid. National Cancer Institute medium 2X
was used as the plaque-overlay medium.
; Cells ~luman neonatal fibroblasts, grown as monolayers, were maintained in 75 cm2 plastic flasks; for interferon studies, they were transferred to plastic panels (35 mm x 10 mm).
These cells, which produce relatively small amounts of extracellular interferon in vitro were harvested and passed in the standard manner.

~ .

., . ~ .

8f~

1 Preparation of virus stocks Bovine vesicular stomat.itis virus, New Jersey serot~pe, was harvested Erom in:Eected mouse embryo and mouse amnion cells to yield a titer of 1 -to lOxlO~ plaque-formin~ units/ ~-.
ml.; the virus stock, ususally diluted 100-Eold, was stored at -70C, '~

, ' -'~ ' ' .' ' .
.

1 Interferon Induc-tion a~cl ~ssa~
To ob-ta:in a sensitive measure oE interference, several measurements of the antiviral state were quanti-tated simul-taneously; ~a) interferon assays and two in-dices oE intracellular interference; (b) colorimetric assay of viral cy-topathicity and (c) reduction in viral yield. The latter method enables us to compare the an-ti-viral protection in the regions near 0% and 100% of cell survival, where the former method fails. Usually the cells `
were used between passages 7 and 25, because the sensitivity of these cells to rIn rCn changes from time to time, each set of experiments contains an internal reference, usually rIn-rCn studied at 10-3 to 10-5M.
Cells were e~posed to polynucleotide complexes in minimal Eagle medium at concentrations specified for 1 hr and then washed three times before reincubation in fresh medium at 37C. (a) Interferon was harvested from the e~tracellular fluids 18 hr later and measured colorimetri-cally using bovine vesicular stomatitis virus as ~he ~
challenge virus. Assays were carried out in duplicate or txiplicate and reference interferons (supplied by the Biologic Resources Branch, National Institutes of Health) were also processed from time to time to confirm the sen-sitivity of the assay system. Assays of intracellular interference: (b) Intracellular resistance as determined colorimetrically. Generally a multiplicity of infection of approximately 1 plaque-forming unit/cell was usea.
Resistance (~)was defined as the ratio of number ~f viable c~lls (after virus infection), to the number of living cells~
(only mock infected). The colorimetric titration (in duplicate) was usually made 72 hr after infection.

,:~

'.~
.~ . .

1 (c) Reduction in virus yield as a measurement o-f intracellular interference. Cells were treated with bovine vesicular sto~atitis virus (107 plaque-forming units per dish, multipli-ci-ty of in~ection approxima~ely 20) for 60 min and the unabsorbed virus then removed; the cells were washecl twice and the cultures incubated for 20 hr in fresh medium. Virus titer was then determined (in duplicate~ in mouse cells by plaque titrations.
Criteria for Idenki~ication o~ Human Interferon ~he interferons were shown initially to fulfill these ~ `
criteria: they were trypsin-sensitive, non-dialysable,~and did not sediment at 105,000g for 2 hr. They demonstrated species speci- ~`
ficity, as shown by the absence o~ activi-ty in mouse or chick cells. Representative interEerons obtained in ~hese studies were chromatographed on SephadexR G150 and G200, yielding a single peak of activity corresponding to a molecular weight of approxi-mately 96,000 daltons.
Interpretation of Results The General Approach It has been generally recognized that in order for the ribosyl polynucleotides to be e~fective interferon inducers, they must possess the secondary structure of a double-stranded helix. In addition, polynucleotides need to be resistant to ; nucleases in order to remain as macromolecules for a ' ~ .

.

, 8q~

1 sufEicient lencJth of -time; polynucleo-tides are usually much less sensitive to att~ch when -they are in a helical complex. Therefore, -these two basic requirements have to be recognized for an~ modification of the rIn-rCn complex.
For instance, the modified rIn-rCn complex (in Eagle's balanced salt solution, buffer A) should have .~-a Tm substantially above the incuba~ion t:emperature (37C) of the cells. The sensi~ivities of these modified complexes to nucleases have been assayed; when these compounds have been found to be more susceptible than the parent rIn-rC
complex, measures were taken to increase their resistance.
The remedial measure adopted for such a purpose is the introduction of polylysine (both the D-form and the L-form~
for the formation of polycation-~Olynuclqotide complexes, which have been shown to be very resis-tant to nucleases~
Thi~ leads to an investiga-tion of the biological activity of the polylysine-polynucleotide compléx. It is important to show that the addition of the polylysine to rIn-rCn does not significantly alter the interferon-inducing capability ~ :~
of the original polynucleotide complex. With the above -strategy in mind, we have exc~mined the modified rIn-rCn . complex with respect to its thermal melting properties, its biological activity as a~ interferoll inducer in human cells, its sensitivities to nucleases and the effect~oE
. -compléx formation with polylysine on all these aspects. ~:

- rI rCn Complexes containing unpaired n Bases ~- Uridylate residues were introduced into the rIn strand and the rCn str~d as the unpaired bases andin one special case guanylate was introduced to the xCn strand ~or the same :

; 30 purpose. The composition of these pres~mably random co-,~
~ -21-~ .

.
.

.

8~3 1 polymers and their sec1imentati.on coe~icien-ts are shown in Table 1. These polynucleo-tides have a molecular weight ranging from 30,000 to 100,000 as judged from their sedimentation coefficient values~ The melting profiles and the Tm values of seven imperfect I~C complexes in bu:Efer A
(Eagle' 5 salt solution), along with the parent rIn~rCn are shown in Figures ~ and 3. All of the complexes gave highly co-operative profiles with the Tm ~50 to 60C) sli~htly lower than that of rIn rCn(64 8C)/ bu-t substantially above 37C, the incubation tempera-ture. The small reduction in Tm and slight broadening of the helical-coil transition profile are to be expected from -these imperfect complexes.

~; : ',' ' -;'' ," '' : ' , _ ;

.

' 1 When these imperfect complexes were evaluated for antiviral func~ions (Table 3~, it was immediately apparent that the introduction of U into rIrl strand caused a much greater reduction in ac-tivity of the resultant complex than when introduced into the rCn strand. The biological acti-vities of rIn-r(C13,U)n and rI~ r(C7,U)n are significantly higher than those of r(I3g,U)n rCn and r(I21,U)n rCn in two separa-te experiments (Table 3), even though all of these complexes are helical at 37C. The Tm of r~I39,U)n rCn was only 2 deg.C lower than that of the rIn~rC~ (Fig.2), but the biological activity is reduced by about 100-fold.
On the other hand, while the rIn~r(C13,u)n ha~ a Im of about 5 deg~C lower than that of rIn-rCn, the biological ac-~ivity of this imperfect complex is only slightly xeduced. Even the rIn r(C7,U)n is definitely active, although the activity o~ rIn-r(C~,U)n is just marginal and is not resE~onsive to an incrëasè in concentration. Previously, lt was noted that formation of an r(C,U)n co-pol~mer with a 1:1 ratio results in a biologically inert rInor(C,U)n complex.
Within experimental error, rIn r(C~2,U) is jUSt as active ;~ as rIn r(C)n Oligo I-rCn and rIn-oligo C complexes, the E~fect of S-trand interruption It has been shown that the biological activities o-f poly (C)-hexainosinate complex were only negligib:Le without the enhancing effect of DEAE-dextran.
In the present studies, we examined this system in more detail and tried to establish the requirement of chain~
lenyth of eithex oligo(I~ or oligo (C), w~ich, when complexed with the complementary polymer, would xetain the interferon--inducing ability of the poly (I,C) complex. The thermodynamic ~nd optical properties of oligo ~I) poly~C) comp:Lexes have ~i . .
' 1 been described in detail.
The Tm o:E the oligomer-polymer complex is dependent on -the o].igomer chain-length and on the concentration oE the complexes and ls less than that of the polymer-polymex complex. The melting profiles of r(Ip)9I rCn (assayed by absorbance) and of r(Ip)l6I rCn ~assayed by optical rotation) are shown in Flgure 4, and some of the Tm values of the complexes are listed in Table 4. The complexes of r(I)n~o rCn have Tm values lower -than 37~C; thereforer the lack of biological activity of these complexes (Table 4) is no-t surprising. In the case of rCn r(Ip)l6I, the incubation temperatuxe during the exposure ~1 hr) was reduced to 30C instead of 37C
~a practice which has been found in our laboratory to have no significant effect so that the oligo I rCn complex (Tm ~ 47C) could remain helical during the exposure. Under this condition, the biological activity o-f this oligo I.rCn complex was found to be only marginal, even at a concent-ration of 10 3M. For ther(Ip)l8I~rCn complex (Tm~y50C), definite indication of biological activity was observed :~
(Table 3), even though it is at least 100-fold less than -~
rIn-rCn.

. .

h ~ o o o I I I I ~ r~ I o H~-- O
* ~ :
~ I~ I' m O o X I I I I I X X I I X X ~
I L~ N Ll') N ~ rl O
~1 a) t:n t!OCt) ~ ~I ~I ~ ~
rl rl O O O O O O O O U~ ~ .
a) ~ C I ~ C X X X X X 1~
h X o o o o II O o N o Z
~ F'l ~ 1 a~

. h ,IJ
~C ~
,1 a~ o\o co ~D ~ ~ R :~

O ~ h X 1~ h h~ ~l ~r l ~i h ~ 1~ ~r~
H ~) O co ~ a~ ,-1 o o o o o o u7 $
¦ ~) H ~/ X r-l r-l ~ ~I r-l I O O 0 ~1 r O

: ~ ¦ ~ rl O
E~ ~ ~1 ~ ~
~;: Il) O g ~1 O h h ~, ~ ~ . d~ ~ ~ tY~ Lr) r~) t~ ~r ~;11 Lr~ 4-1 ~rl h--I ~ I I I I I ~1 1 I I I
0 0 0 o ~I O Q o O O C_) Ul S I r~ ~¦ r-~ 1 0 ~ ¦ 1=l ~I r-l r-l O O
~rl O h X X X X X ~ X X X X X ~ ~9 ,~ ~; E~ ~--1 ~I r~l ~1 N ~1--I ~1 ~ --I 1~ 0 (d _ ~I C) ~ ~ a) O
r O
. o (~r ~_ O u~ O 0 0 .~) O O 1~0 r~ ~ ~O
. ~ E - ~ V
~D ~U $
: ~ (1~ a) I'~s S~
H H Hc~ 1 r-l 3 3 O

h '~ Qi Q Q~ * ~ X
.) H H H H H 1~1 ~ I ~ ~I r-l . ~ :1 ~ ~ ~ V V ~ ) S L~ rJ
rc) h S-l S-l 5-! h h 5-1 S l h a) h S ~ u~3 ~ ~)~H H H O C.) ~ h ~ S-l S-l S J h h h Z 1~
.

~ '',~ .

4~

1 The interruption of the rCn strand was carxied out by co-polymerization of G~ into rCn, followed by the RNase Tl en2yme degradation. The G residues serve both as vulnerable positions for specific hydrolysis and as identification of chain-length, since the end residues in all khe oligomers are the G residue. Therefore, the rIn~r(Cp)I~p complexes not only have a strand interruption but also a looping-out of the G residue at every interruption. The melting profiles of all the rIn~r(Cp)IG>p complexes are shown in Figure 5. The rIn~oligo C complexes do have ~!

':

' ~

~; 30 ~ ~

,, . . , ~ 23A--'` ~ ' ~, ' ': " . .

lo~er Tm values -th~n the rCn oligo 1 with oligomer oE the sa~e chaitl length. There~ore, in terms of thermal stability, rIII r(Cp~3,G~p is approximately equivalen-t to r(Ip)~
rCn under similar conditions. The biological activities of these rIn-r(Cp)IG~p complexes are lower than the original rIn r(C20,G)n complex, which is almost the same as the rIn rCn. The complex with the largest o:ligomer, rIn r(Cp)48 G ~p, has a very high T (59 6C), but is tenfold less active than rIn rtc2o~G)~ rIn r(~ )23G~p is about tenfold more active than r(Ip)l8I-rCn. r~n r(Cp)llG~ p showed only a small amount of activity, as may be antlcipated, since the Tm of this complex is below the 37~C incubation temperature.
Two conclusions can be drawn from the above results.
First, interruption o one of the strands in the helical complex definitely reduced the interferon-inducing activity, as exemplified by the comparison between rIn-r(C20,G)n versus rIn r(Cp)4~G~p (Table 5). In this case, the rIn r(Cp)4g G> p has almost the same Tm (59-6C) as that of rIn-r(C20,G)n (60 9C) and the rI~ r(Cp)48G> p contains even less frequency of unpaired G; yet the biological activity of rIn r(Cp)~8 ;~
G~p is -tenfold less than that of the uninterrupted rIn r(C20,G)n. Second, the rIn oligo C complex is more biologically active than the oligo I-rCn complex, even when they are similar in thermal stability and not greatly dif~erent in oligomer chain length. Thls is illustratèd in the comparison between rIn r(Cp)23G~p versus r(Ip)l8~I rCn.
in which the former conplex is tenfold more active than the latter. This observation reinforces the conclusion that modification on the rIn strand has a larger biological e~fect that that on the xCn strand.

)q :

.

~ 8~3i o _ ~n h ~ ~_ r-~ r~ O O
~1 ~ ~ ¦ I O I I O r-J

H . :
IH
O
_ "_ _~_ co ~ O O X
r~ r r ¦ ~¦
X X ~JI ~
J r-l ~ ~ C~l aJ r-l r~
r Ir) ~ O~ r~ r l u~ o o oo o o o ~o) o o o s~ n (I) tO ~ r-l r~l r-l r~¦ ~I r-l r-l r~ l Id _ X ~ X X ~ ~ I I X X X X ~C X X I I I I ~C I . ~.~
~1 d~ ~ It) IS') O ~) ~ ~ V
r~l rl ~ a) ~H
~ t~ 1~ r-¦ w c~) r-l 1~ r~ r-l 8 o ~ ~
~ . a) ~

r l O r~
V ~1 .IJ: co ~1 . OQ~ .
~ a) ~u , _, _ o s~
~ri v -1-1 o~ O O 000 0 ~lO 0 0 Ln t~ ~ r~ O . r(:~rl Ln~ I id O--' .o o c;~o o r-iO O ~i I C17 1 ~ ~ U~
O 5~ J r-i r-i r-i r-i r-l r-i r-l Fl HS ~ P~ . ~ r'`l I?
~¢ h . Ul ~ :
~1 :' X ~ ~ .
11) : r-l ~rl 0 : ~ ~1 3 S~ . ~ ~ O ~
,~ . ~r~ ~ C) ~
U~ U~ ~rl (I) ~ ~ ~ o a ~
r~l q_l ~d f;l r~ ~rl ~
_ r-l O a) l ~ Ln ~ ei~ Ln ~) ~ Ln Ln ~ ~* Ln Ln t~) ~ Ln L~ ~j ri a) l l l l l l l l l l l l l l l l l l q~
., t.) o o o o o o o o o o o o o o o o o o a) ~ri :~ ~ r-i ~i r-i r-l ~i r~ i r-i r-i r-i ~i r-i r-l r-i r-i r~ I ~ ~i ' ` O 3C X X X X X X X X X ~C ~C X X X X ~ X ~ O
~1 r-i r-t r-i r-i r l r-i ~i r-i r-i r-i r-i r-i r-i r-i r-i r-l ~I r-l O ,~

-_>~, ;:
r-i ~ri ~ )~ V t~ ~1 ~ri Si _ _ ~ 0~ ) LnLn In ~) r~) ~) ~) r-i r~i r~ Ul. ~ C~) V ~ ~ ~r ~) ~) ~) ~~ i ~ ~i r-i r~ S~
O O ~ P~ ~ ~ Q~ Ql Q~ O ~ U~ h ~Ll ~ S~ ~ ~~' ~ ~ ~ ~ `~ ~ ~~ ~~ ~ _~ _ _. .
h $ ~ h ~I S-i h h S i 5-1 hi h S-i h h h h h r~
~d ~ h l-i ~ H H t-l H H 1-l H H H H 1~1 1--I H l~i H H 1--1 rl `~
1:1 h h h ) I h S-l h h h h h h h h h h 5~ i ,!~.,,~ .~'.:d ; . . ' .
.~ , ' ' ' . ~' ~, . . .

:
1 The above modifica-tions introduced to -the rIn rCn complex (mismatching and interruption) invaria~ly led to a loss of biological activity in varying degrees. Before this observation can be interpreted properly, we should be certain that this loss of biological activity is not simply a reflection oE the increase in sensitivity of these ~odified complex~s to attack of nucleases. As indicated in Figures 6 and 7, the biologically inactive r(I21,u) rcnand rIn r~C4,U)n, as well as the less active rIn~r(cp)48G ~p, are considerably more sensitive to attac~ of pancreatic RNase A than the highly .
active rIn~ rCn .

:

~ ' ~

~ ' ~ .
:' , , ~ -25A-.. l.. ~ ' ., .

One ~nown procedure to protect RNA from nucleases is by complex formation with a polycation. Interaction o~
poly-L-lysine with rIn rCn was first investigated by Tsuboi, Matsuo & Tslo (1966); they reported a stoichiometry of I
lysine: 2 nucleotides in a solution of dilute sal-t (0 05~q-NaCl, 0 001 M-sodium citrate/ p~f 7 0). ~his stoichiometry was reflec-ted by the proportionality of a two-step melting profile, lower transition (57C) for the uncomplexed rIn rCn and a higher transition (^~90C) for the ternary poly-L-lysine + rIn rCn complex.
It has been shown that the binding of poly-L-lysine to RNA(lmg/ml.) on an equivalent basis resulted in the -forma-tion of insoluble complexes at low sal~ concen-tration.
Soluble complexes are formed, howevex, at a low poly~L-lysine/RNA ratio. Digestion of the soluble complexes by .
pancreatic ribonuclease, ribonuclease T1, non-specific ribonuclease from Bacillus cereus, and Micrococcal nuclease - yielded a precipitate wi-th a lysine/nucleotide ratio of `
l:l Together with other supporting expeximents, it was concluded that the complex formation between RNA and poly- ~;
L-lysine protected the RN~ from attac~ by these nucleases.
Our recent reinvestigation of this interaction showed that, in Eagle's salt solution (0-15M-NaCl, 0-00l M-MgC12 and 0-01 ~-PO4,pH7.2), the stoichiometry of the ternary complex of poly-L-lysine to rIn rCn is l lysine: l nucleotide, instead of 1 : 2 observed earlier in dilute salt.
A two-step transition profile was observed in ~i solu-tions with poly-L-lysine present in less than 1 : l stoichiometry. The Tm of the ternary complex is a~out 30 83 ~ 1 deg.C. At a low concentration of rIn-rCn(less than , 5x10~5M), by adding an equal volume of nucleic acid solution ,' ' ' ., i . ., : ' ' 34~ ~

l slowly with mixing to a poly-L-lysine solution at room temperature, precipi-ta-tion can be kept to a minimum ~less than 10%). EIowever, such a solution invariably became cloudy at elevated temperature, especially near the Tm~
This phenomenon, the precipitation of the melted complex in the presence of poly-1-lysine, was veri-fied by the observation that the single-stranded rIn and rCn are much ; less soluble than the double-stranded complex (rIn rCn) in the presence of poly-L-lysine. We also have investi-gated the formation of rIn rCn with poly-D-lysine. Many ; properties of this ~ernary complex are similar to those ~:
of poly-L-lysine, except that the Tm f this ternary com- ~; ;
plex is only about 68C.
As shown in Figrues 6 and 7, in the ternary complex ~ with either poly-L-lysine or poly-D-lysine in a l:l ratio (N/P), the modified or unmodified rIn.rCn complex was mostly protected from the pancreatic RNase. Interestingly, in the complex with a l lysine : 2 nucleotide ratio, merely ;. half of the polynucleotide complexes were protected. This ~; 20 observation confirmed -the l:l stoichiometry of these ternary ~ .. ;
~ complexes.
: Table 6 shows that~oly-L-lysine alone does not prevent the multiplication of bovine vesicular stomatitis virus.
Ternary complex of rIn~rC~ with poly-L-lysine iA

: ~ ' ' :' `: ' ~
' ' ' ' ;` ~ 30 ; .

-26A- ~

~-,, ' ~' ' o o o o o o Q .~ .~ ~ r~l ~ X X .~ X X

~H
~ ~ .. O O O O r~ ~ ri O
~ .~ r-l r-l r-l r~
;~ ~t , ~
'.' Pl ;, .
. ~ .` ` ' ' .- . O u~
U) ~-1~ ~ l~t . X~ P' 'I ~i r~ I I I I I . ' ~;7 r~
O . ; . ' '~
~ . ~.
~rl . ~ ~ ' O O O O O O
t~ ~t X X X X X X
O ~_ U~ O L~ U~ O U) ~:
~St t~ tt~ t~l C~t . ',~

h . .
Q ,~
Q) F, rl rl I ,1 ~ h~ O ~ .
,0 Ul U U~ Ut Ut ~ ~
-rl ~1 ~ r l r~ Or~ r~ O Q) ~ " .

F ~-1 r-l r J ~--1 r l r l r-l Q P~ O O O O O O 3 H O
Q) :
~ ~ ~ :
_ l l l l I¢
~ r-l r-l r-l r~
,' 1~1U')10U') ' ~ ~','.
.'` . '. :~
. ~ ~ 1:~ ~ F ~:
O ~ ) U ~ S-l h h . ~ ~ ~ O Q) r~:S ~ F.
. H H H H O O O O

' ~

, ~ .

1 1:0-5 (P/N) ra-tio, 1:1 ratio, and 1:5 ratio, all have about the same biologica:L activi-ty as the uncomplexed rIn rCn.

Complex formation wi.th poly L lysln~, however, does not significantly enhance the ac~ivity of rIn.rCn either, when rIn.rCn was supplied in suboptimal concentra-tion (lxlO 6M~ Table 7).
Results given in Tables 4, 5 and 7 show that in the ternary complex formation with either poly I, lysine or poly D lysine, in 1:1 ratio or 2:1 ratio (N/P), the ~o illterferon-inducing activity of the modi~ied rIn-rCn complexes has not been increased. For example, though the susceptibility to nucleases of r(I21,U)~.rCn and rIn.r(C4,U)n are greatly reduced by complex formation with poly L lysine (Fig. 5`, the biological activities of these inactiVe, modified complexes were not enhanced (Table 7). Similarly, ' ~:

~ ; -27A-LB~9 I ~
o o o o o o o oo o u~ ~a m ,~ r-l ~I r-l ~r-i X I I X X ~ X X~C I I X X I I I I I X
~ ~ ~ cn w fr) o o t~
-rl rl PJ .
~> ~X r~ Or-ll~ ) ' ~, ' :` ' .

~ .:
o ~ 1~
I I I r~l ~i r-l ~1 I I ~ ~ I I I I I r-l ~1 X~ I : '' (~ 1 ,~ ~ .
O
~l ~\ :: i tn o v v a) '~ ~ O
~, ~ ~ ~ ~ O O ~ Ci~ O 1~ ~r lD ~ ~ i H . I I I I I r-l ~ I ! ~
Ql O X .
v ~1 ~ .~
~: ~rl . . ` . l C) ..

H r-l . .
O r-l r-l ~I r-7 ~I r-l r-l r-l r-l r-l r~ l .; :
O ,S ~ ~J ~ ~ ~I r-l r-l ~ l ~ a"~ . . . - : ~: .
." I X ~ S-l ' ~ ~ ~

O ~1 :
.

O S~ L' L' ,. ~
(D ~1 ~ rl ~rl ~rl rl rl ~rl rl rl rl : ~:
1~ V r-l r-l r-l r-l r I r~ r-lr J H rt rt rt ~ ~1 ~1 ~
a, a ~ a ~ ~ . : :
a~ ~ ~ :

~: I . .
h O ~: :
,H: ~ ~

H ~ I I I I I I I I I I I I I I I I I I .
V~ O O O O O O O O O O O O O O O O O O I :::
: ~ Li ~ r~l ,-1 rl r-l r~l r-l r~l r-l r-l r-l r-l rt r-l r-l rt rt r-l rt . . : '~, .~ _ . ,:
.
. ~ .
r~
L'i O
~ C.) O C ) C,) ~ 1 :' ~ 1 5~\ O p p p ~) ~IJ
~1 ` ~ r o .~ O ~ L~ L' P t) P P P ~) U C ) ~) O U ~ ) U V
~ 1-1 5-1 h 5-J h ~ ~ ~ c~ ~ cs, ~ h ~ i~ r-l .; r~ ~ ~ r~ r-l ~ r~ (d .: ~ L' ~ L'i ~ ~ H H H H H H L' S ~
H ~ H H H H ~ ~ ~ ~ H H H H H H H ~,1 h S-l S l h h h I J ~J h ~I h 1~ 1 h 1-l h ~-1 h ~
. i .
.~

. ., ,i .--' . . . ; ~ -: ~ , :: . ;.: . -1 the activities of the moderately active rIn rtcp)35G~p and rIn-r(Cp)23G>p were no-t promoted by fcrmation of a ~:1 : complex with poly^L lysine (Table S)~ even though this tern-ary complex should have reduced the susceptibility oE
the polynucleotides to nucleases, as extrapolated from the results on rIn r(Cp)48G (~ig. 7). In addition, the Tm o the rCn-r(Ip)l2I has been increased to 45C~
significantly above the i.ncuba-tion temperature, by complex formation with poly L lysine, and this rCn oligo I complex remained inactive (Table 4). Therefore, the reduction of susceptibility to nucleases and the enhancement in Tm brought about by the ternary complex with polylysine (~ or D) did no-t transEorm the inactive, modified rIn-rCn comp:exes into an active state.
- , ',~
; , ~
~ , ''~ ", ~

, . . .

,~ . ' :, '~

.,, ~: 30 :-~ 28A- ~

1 Rela-tive Rates o Enzymic Hydrolys_s of Several C plexes With Similar Inter~eron Induc ng Ac-tivities :~
As described in Tables 3 and 5, xIn r(C13,U) and ~In.r(C~O~G) are essen-tially as active in interferon induction as unmodified rI rC ~ It is of both theoretical and practical interest to compare the susceptibility of these complexes toward nucleases. The Tl ribonuclease, an endonuclease for the rI strand and G
10 residue, and the pancreatic RNase A, an endonuclease for the .
rCn strand, were both used .in this experiment. . .~.
The data in Figure 8 clearly show that both :.
.
. modified complexes were hydrolyzed at rates o~ 5~to 8-fold ~ .
. . , ~
; more rapidly than rIn rCn; indeed, approximately 50- of their total hydrolysis occurred within the first 12 minutes ~ .
of incubation. Previous work has suggested that high - ~.. ~.`.-antiviral activity of synthetic polynucleotide inducers . :~ :
is integrally associated with their prolonged stability .
in biological fluids; and, as greater antiviral activities 20 are achieved further toxicities simultaneously acarue. .
It has been ~ound that it is possible to prepare :~ a polynucleotide complex which is highly active as an - ::
interferon inducer but also highly susceptible to nucleases.
. .
Two types of structural modifications were made ~ :~
~ : :
to the rIn rCn ccmplex in this investigaticn~
Mis~!atching of kases, which cause a looping out from the helix; and (2) strand interruption. The interferon-inducing ac-tivity of rIn rCn is lowered by these modifications in varying degrees; the dec.re~se is much moxe significant when the perturbation is imposed upon the rIn strand than upon the . ~

: ' ' ; ~ ~29- .
:~ :

1 rC s-trand. This lowering of the biological activi-ty of rIn.rCn cannot be explained by the two sim~le requirements for thermal stability, and susceptibility to nucleases.
The Tm values of the modi-Fied rIn rCn complexes are only slightly less than that o~ rIn rCn and are well above the incubation temperature o~ the cells~ Ternary complex formation with poly L lysine significantly increases the Tm oE these polynucleotide helices~ but has little effec-t on their biological activities. These modified complexes are more susceptible to nucleases: however, modified rIn rCn complexes can be prepared which have biological activities comparable to those of the original rIn-rCn, but are much more susceptible to nucleases. Xn addition, ternary complex formation with polylysine (both L and D ~. -forms) virtually protected all these polynucleotide helices from nuclease, yet their biological activities were not changed significantly. These observations and reasonings indicate that the structura2. modifications introduced here :~
may be directly related to the structural requirement of receptors in the cells responsible for the triggering .
mechanism of .interferon production.

' .`

, , ' ' .
,~
, ~ .

. ~ -29A-..

1 The ~pparently stringent req-lirement for the polynucleotide complex in order to be an effective interEeron inducer is well known. However, it has been reported that ternary complex formation with DEAE-dextran significantly enhanced -the biological activity of rIn-rCn. Most of the effec-t was attribu-ted by the investigators to an increase of uptake and protection ~rom endonucleases. It has been ;~
proposed that the change of mass to charge ratio may acilitate the uptake process, which is paxt of the consideration related to the question of specificity.
; However, the fate of such a complex inside the cell has not been discussed. It is not known in what manner the DE~E-dextran can be dissociated from the rIn rCn complex or can be degraded. Similarly, in this investigation, ternary complex formation with polylysine ~both ~ and D), did not ~; chanye the biological activity of rIn rCn as interferon inducer. A relevant study on the physico-chemical properties ~, . .
of DNA, RNA, rAn rUn and rIn rCn upon complex formation has been reported. The circular dichroism spectrum Oe rIn rCn has been drastically changed upon complex formation with poly L lysine and the ternary complex has al50 been investigated by high-angle X-ray diffraction, low-angle X-ray dif~raction and by electron microscopy. While definite information about the helical structure cannot yet be obtained, it is most likely that the ternary complex is a multiple~stranded fiber. It has also been reported that the inEectious RNA of two equine encephalitis viruses lost their infectivity upon compIex formation with poly L lysine, though the activity of such a complex can be recovered after . .
pronase treatment or dissociation by strong salt. This polylysine-RNA complex was resistant to inactivation of the nuclease.

' ~ -30-1 At presen-t, we have no knowledge about the fate of the polylysine-rIn rC complex outside the cell duri~g the incubation period of one hour, or of the complex remaining ~ithin -the cell at the end of exposure aft.er washing.
It does not appear to be a simple, easy process for the removal o~ polylysine from the ternary complex, either by dissociation or by degradatlon (especially the D-analog).
It remains a possibility that the "receptors" in the cell can be triggered by the entire ternary complex containing both polylysine and rIn-rCn. However, as described above, the conformation of rIn-rCn has been greatly changed by the association with polylysine.

~''''-:1 ': . ' .~ , - .
' '~- ' :
:`; :;:

''; . ' ' ' ~' ;~
~ :

. ~ .
' ' ':

~ 30A- ;~

1 L~1849 An analysis o~ thc composition revealed that the actual synthesis produced polymers which have about one half as mucl as the 2'-Q-MethylinOsinic originally put into the reaction mixture. For example, in the composition given in Tables VIII
and IX, these compositions were given, as indicated in the footnotes of those tables, as an input concentration into the enzyme mixture during the preparation procedure but the analysis of the result indicates the amount of mI is only half of that put in. Therefore, the correct chemical composition, in the cases determined, always contains half of the mI as put into the original en~yme mixture. For example, rI10,mI which is now cited in Table VIII represents the inpu-t ratio. The chemical ratio of the product actually is rI20,mI, which is footnoted in Table VIII.
A biolo~-cal activity of poly 2'-0-Methylinosinic as well as the hybrid strands containing both iOsinic residue and 2'0-Methlinosinic residue were tested on human cells. These -~ compounds first will form complex with rC polyribocytidylic in a manner descri~ed in the experiments. The results indicate clearly that the poly 2'-0-Me~ylinosinic polyrib~cytidylic 1 to 1 complex has little or no activities as human cells interferon inducer as compared to the original rI rC. However, in the case of the hybrid molecules, the resultsare totally different. As recorded in Tables VIXI and IX, the human cells and cultures respond to the complex rC as well as the hybrid molecules of polyinosine and 2'-0-Me inosine produce significantly much better results in tests against viral attacks as compared to the original rI-rC.
It i5 to be noted from the footnotes of Tables VIII and IX
that the composition of the products are certainly not identical to the input of the oxiginal substrates in the enzyme preparation.
; It can be summarized fr3m Tables VIII and IX that 1 to 1 complex of rC and polyinosinic containing 5 to 16% 2'-0-Me inosine can be 100 fold more as effective as rI^rC in inducing .. ,.. . .... . ~ . , ~

rd O ~ ~ ~ P ~ ~ ~r ~ ~ r-- ~D
t~ ~ aJ c~ o o o o o o o o o c~ o Q) ~
S O r~ r--J r-¦ r ¦ r lr-l r-~ r-l r-l r-l r-l r-~ ~1 ~ 1 0 r~ U U~ XX X X X X XX X X #~ X o r-l 0 ~3 ~ ~ ~1 r-l r-~U~ ~1 cOr-l ~) U ~ r-l O w 11~ rl w ~ ~ ~ ~ ~ ~ ~ ~ ~o ~ ~ ~ ~ r~ 1 ~ O
U~ w~-r1 S~ r-l ;~ ~ r-lr-t ~ ~ ~ a~
O P~ ~ 5J H
~) a ~ h r-l ~ w O _l ~
~ . P~
w ~r-l O
w ,a ~: rt r~l w r1 ¦-1 r~ rl n 111 ~ O rC ~ rl ~ IJ 1 ~ r ~ IJ~ ~ Ot~ ~ ~ n h O ~ C)~ O ~ O r~ O ~ O ~D ~D n~ I S ~ ~
U U nw r~ r-l r-l r~l r-~ X ~ ~ ~
~ O ~ ~ U
U~ ~1 0 P~
H ,~ n- 5 C~ ~: U~ nw ,S~
: O ~,1 :~

O ~

5 1 n~ H ~ ~ :
~ ~ u ) r O l l l ~ _ r~
O O O r~ t~) nw r~ rJ
~ r-~ r-l r-~ HH H U~ U ~ m O U H H H H H H O O O r/ ~w ~J
r ( ~ X X X ~I H H H H H H H H Q~ h r-l :
~ `; h O ~ r-l 111 8 ~ ~ Wi ., _ . U ~ X
,~ i-t H H ,~ N O
H H ~ tu ~ r~
~i H O w rd ~rl O
n ~ : O O

rl C.) ~) C.) ~ r~ ! r~ rl t~ . . ~-1 ~ h h ~ n 1~ ~: ~ ~ ? Ul ,_ ~ ~ s ~
H H H E~ ~ril ,~: ~1 ~ ~) n ~ ¦: El ~3 13 1~ 0 ,1: ~w ~t u ~ ~ ~ # a) u~
h 1-1 O O O ,S O O ~ r~
~) ~ ~ ~J --1 ~ ~ r~ E;
1:: H H H ~1, ~ "w h H 1-l h 1-l ~ rl (~ Ul .C Q~

_ _ _ _ ' ~ :~

:
' ~ ~ ' ,: , &~ ~ ~ ~ a~ ~ ~ ~ ~ ~ co ~D~ , ~ c.
~. ~ I ~,~ oo I v u~ I ~J ~ ~o i ~1 l ~ I ~ ~ ~1 0 ~
~ ~ ~ 8 ~, a~
O ~ ~ ~ U H
~:: qJ ~ _~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~I o ~ a~ o 0~ O O O O O O O o O O C~ O C~
O rl ~ r~l ~ r-1 1~1 ~ --~ r~ r-l ~ 1 ~1 ~1 r~ 1 ~ p~ O ~ H
t X X I X X I X X I X ~ I X X I X ~ r~ 0 0`--

6~ ~ u~ r~ ~ ~r ~ rl ~ ~D ~ ~1 Q~ . . . . . . . . . . . . ~-~'I N -1 ~ ~ ~ l ~ q~ ~ ~ a~m~o~

~U . 3,~,00 U~ ~ ~ O ~ h a ...... : ~N ~0 I ~ Cl~
'~P co ~ ~ a~ o o o c~ ~ ~D ~O O
$dP ~ ~ a~ oo o ~ ~ a~ o I o u~ I ~ ~ o F ~ ~
s~
~
o ~
~ ~ :~ X ~ . o Q, u~ ~ D.~
~:S`l : O C) O ~) ~ ~ `
,~ ~ r-l ~1 ~1 H H H H H H O ~
,4 t.) ~ X ~ H HH HH b~lH H 1-1 H H H Q. ~ ~ 1~ Q
~ O ~I ~-1 H H H H H H H 1-1 H H H H H H H H H H E~
lU O l~ l Uln a) UO~O~O
:~ 1~ GJ ~ ~J ~
3 ~ ~
... : U~ U' ~,~' ~,~ ~ U~ o~ o ::.

lU . h . C.l . O U ' . ~ L: 1:~ ~ ~:: . _ a~ ~1 0 0 .: ~ .. _ _~ .. _ ~_ ~ H ~ ~
~: H t-l H _ H _~ ~ ~ ~t ~ D~ Dt a H
~-I ~ ~ E~ . E~ --I E3 ~ H ~ h ~ :
u~ ~
o o O _ 0_, ~ . ~ E ~ o . ~ s~l ~1 ~ U~ t~l 1: 1-1 H H H H H ~ ~ N ~ H
H 1 1 ~ ~ J.~ ~ ~
.~ h _ _ _ _ _ ~_ .rt ~ o V~`' :~ :

8~ .
~nterferons from h~unan cells. This is a very important finding, ~perhaps allowing the administration of -the drug a hundred fold le~s in a dose level producing the therapeutic value.

. :

:. :
.: .

. ~ :

: ~ ~
; -34- `

~ , . . .. .. .. . . ....... .. . . . . . . ..

Claims (14)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A process for forming a polynucleotide sub-stance comprising adding the appropriate nucleoside diphosphates in an incubation mixture containing an enzyme polynucleotide phosphorylase to form the polynucleotide substance and then purifying said polynucleotide substance, said polynucleotide substance being a copolynucleotide selected from the group consisting of poly(Cn,U) and poly (Cn,G) wherein n is an integer having a value of from 4 to 29.

2. The process of claim 1 wherein said copoly-nucleotide is poly(C4,U).

3. The process of claim 1 wherein said copoly-nucleotide is poly(C7,U).

4. The process of claim 1 wherein said copoly-nucleotide is poly(C13,U).

5. The process of claim 1 wherein said copoly-nucleotide is poly(C22,U).

6. The process of claim 1 wherein said copoly-nucleotide is poly(C20,G).

7. The process of claim 1 wherein said copoly-nucleotide is poly(C29,G).

8. A polynucleotide substance when prepared by the process of claim 1.

9. A polynucleotide substance wherein the copoly-nucleotide is poly(C4,U) when prepared by the process of claim 2.

10. A polynucleotide substance wherein the copolynucleotide is poly(C7,U) when prepared by the process of claim 3.

11. A polynucleotide substance wherein the copolynucleotide is poly(C13,U) when prepared by the process of claim 4.

12. A polynucleotide substance wherein the copolynucleotide is poly(C22,U) when prepared by the process of claim 5.

13. A polynucleotide substance wherein the copolynucleotide is poly(C20,G) when preapred by the process of claim 6.

14. A polynucleotide substance wherein the copolynucleotide is poly(C29,G) when prepared by the process of claim 7.