1+&778 Reprinted from Joomal of Medicinal Che9!6try' 199.3'J6' .,,

Copyright O 1993 by tbe Arirerican Chemical Society and repribted by perDission of the copyright o*'ner'

Neuraminidase-Resistant Ilenagglutination Inhibitors: Acrylemide Copolymers

Containing a C-Glycositle of N-Acetylneuraminic Acidr

Michelle A. Sparks,2 Kevin W. Williams,s and George M. Whitesides'

Department of Chemistry, Haruord llniDercity, Cambridge, Massachusetts 02138

Receiued September 14, 1992

Copolymers consisting of a polyacrylamide backbone with side chains terminated in C-glycosidicadlogs of N-acetyheluraminic acid were synthesized by free radica_l copolymerization of. a-2-C'

tB-tt2--(N-acryloyiamino)ethyllthiolpropyll-N-acetylneuraminicacid(5)withacrylamide. Unlikenatural and synihetic polyvalent materials that contain N-acetylneuraminic acid in O-glycosidicform, these i-glycosidlc copolymers resist neuraminidase-catalyzed cleavage- of the neuraminicacid residue from the copolymer backbone. Examination of these C-glycosidic copolymers in ahemagglutination inhibilion assay indicated that they are as _effective in vitro as polyvalentO-glycosidic copolymers in inhibiting agglutination of erythrocytes by influenzavirus. The minimumuul* of the inhibition constant, c;lculated on the basis of the concentration of NeubAc groupsin solution, is K;(HAI) - 10-? M for both copolymers. The inhibitory potency of the C-glycoside-based copoiymers becomes more significant at lower concentrations of NeubAc moieties in solutionthan does the inhibitory potency of the O-glycoside-based copolymer.

Polyvalency-the ability of a ligand to bind to a targetvia multiple chemical interactions-is considered animportant factor in many pathogen-host cell interactionsthat result in infection.a Influenza A, an orthomyxovirusresponsible for the most severe outbreaks of influenza,adheres to the terminal N-acetylneuraminic acid (NeuSAc)

of glycolipids and glycoproteins on the surface of mam-malian epithelial cells via the viral surface lectin, hemag-glutinin (HA).5 Monomeric a- 2-methyl-N-acetylneuramin-ic acid (1, NeubAca2Me) binds weakly to HA.6 A varietyof structurally modified, monomeric Neu5Ac derivativeshave been tested as inhibitors of hemagglutination;?-e onlyrecently have monomeric derivatives been prepared thatare significantly more effective than Neu5Aca2Me, 1.8'eThe detailed mechanism of inhibition by these compoundsremains unclear.

An alternative approach to the prevention of infectionis the development of polyualent drugs that could, inprinciple, bind to and block access to receptor sites of thepathogen.lo A number of potent, naturally-occurring,polyvalent compounds that inhibit the adhesion of in-fluenza virus to cells in vitro are known.ll In addition,we12,r3 and othersi4-16 have developed inhibitors of influ-enza virus-induced hemagglutination that are polyvalent

in NeuSAc. In vitro, these polyvalent inhibitors prevent

hemagglutination based on virus-cell adhesion at con-centrations of NeuSAc groups between 104 and 10-8 M.12-16All NeuSAc-based inhibitors, whether natural or syntheticin origin, are susceptible to glycosidic cleavage byneuraminidase (NA, EC 3.2.1.181,t2-ts a glycosidase thatis present on the surface of the influenza virus. Neuramini-dase-catalyzed cleavage of the NeubAc groups present inthe oligosaccharide moieties of glycoproteins is thoughtto allow the virus particle to escape from naturallyoccurring polyvalent inhibitorsl7-le and aid in the releaseof newly formed virion from the infected cell.20

In order to circumvent the neuraminidase-catalyzedcleavage of terminal NeuSAc residues from synthetic,polyvalent copolymers presenting Neu5Ac groups as side

chains,12,14,15 we have prepared random copolymers in-

corporating C-glycosidic NeuSAc moieties derived from

the monomeric 2- C -t3-II2- (N-acryloylamino) ethyll thiol -

propyll-N-acetylneuraminic acid21 (5, Scheme I).The copolymers of 5 and acrylamide-poly(5-co-

acrylamide)-were preparedwith a range of mole fractionsof 5 attached to an acrylamide backbone. We havemeasured the ability of these copolymers to inhibit theagglutination of chicken erythrocytes induced by influenzavirus (X-31;zz in vitro by employing a standard hemag-glutination inhibition (HAI) assay.23 These copolymershave the potential to compete with polyvalent cell surfacesfor HA (and perhaps NA) binding sites on the viral surfaceand to inhibit virus-host cell binding, without beingsusceptible to cleavage and release of the NeuSAc groups

by NA.17The synthesis of 5 began with glycosyl chlorid e2.24 This

compound was converted to an anomeric mixture ofNeu5Ac C-glycosides 3 (a-2-C-allylNeu5Ac) and 4 (0-2-

C-allylNeu5Ac) under the allylation conditions developedby Paulsen2s and Bednarski.26 Separation of 3 from 4,25'26followed by photolytic addition of 2-aminoethanethiolhydrochloride to the double bond terminus of 3,ta'zz *".a convenient method for the installation of a linkage group

with a terminal primary amine. This amine allowed for

attachment of the acrylamide moiety needed for copo-lymerization. Addition of the 2-aminoethanethiol adductto a solution of N-(acryloyloxy)succinimide in basicMeOH/HzO (pH 10) produced the desired NeuSAc-acrylamide 5 in 85 % purified yield.

The copolymers were synthesized by photochemicallyinitiated copolymerization of acrylamide and 5. In allpreparations, the initial concentration of acrylamide wasfixed at 1.0 M. The initial concentration of 5 was variedfrom 4.0 mM to 4.0 M;this range corresponded to a molefraction of 5 (XNeubn.) in total acrylamide moieties of 0.005-0.80 (eq 1). We assume this mole fraction is preserved inthe final copolymer.

t5IXNeusAc =

tSl + lacrylamide](1)

These copolymers have not been fully characterized.Previous experience in the preparation of acrylamide

O 1993 American Chemical SocietY0022-2623/ 93/ 1 836-07 78$04. 00/ 0

N euraminidase - Resistant Hemagglutination Inhibitors

Scheme I. Synthesis of Copolymer 5 and Acrylamide

Journal of Medicinal Chemistry, 1993, VoL 36,1/o. 6 779

Neu5Ac

H2NCOocNH2

Neu5A

ocNH2

poly(5-co-acrylamide)

0 -Neu5Ac l r n . . r c

E F e q u i n c , r - \ 1

E F g u i n e a i a 1 : \ l

F /hovmc m - . r r

s , z h u m a n r r . \ {

ff--lyOrupt',, i,. \(. ' \- -

EFdivalent sialosrdc'

S o v r n e m u c n l r

I e tu l n ' '

n m , )n ( r vJ I e n t

| | Ncu5Ac ana logsh ' r

0 02 0.4 0.6 0.8 IX NeuSAc

Figure 1. Inhibitory potency, K,'HAL, of the C- and O-glycosidiccopolymers of NeuSAc as a function of mole fraction, XNuusA", ofNeuSAc covalently incorporated into an acrylamide copolymer.The data points presented for the C- and O-glycosidic copolymersare averages offour to six sets ofdata obtained from independentlyprepared samples of copolymer. The data for the O-glycosidiccopolymers have been previously published from this laboratory.r2K,tu'ttr is the minimum concentration of copolymer-linked Neu5Acresidues in solution that inhibits agglutination of chickenerythrocytes in vitro at 4 oC. Assay conditions are given in theExperimental Section. The horizontal line denotes the thresholdantiagglutination inhibitory potency of monomeric Neu5Aca2MeI lK,ruatt = 2.8 x 10-3 M)6 assayed under identical conditions.The error bars represent *1 well in the HAI assay for (r)C-glycosidic copolymer of Neu5Ac, and (o) O-glycosidic copol-ymers of Neu5Ac.

containing O-glycoside NeubAc groups, reported previ-

ously.12 The data are expressed in terms of K1(HAI) versusXNeubAc. For both sets of copolymers, the antihemagglu-

tination potency (measured by KrtHnt) of the C- and

O-glycosidic copolymers of Neu5Ac) reaches a maximum

value of XNeusncmax = 0.3-0.4. We have attributed the bell

shape of these curves to a combination of entropy and

optimal glycoside spacing effects.l2 The horizontal line

in Figure 1 represents K;(HAtl for Neu5Aca2Me I (2.8 x

t'[n\rt*' "'"o,lofl €

l-oa.FOAcl-OAc

,

copolymers of the parent O-glycosides of NeuSAc has,however, shown that copolymers produced by an analogousprocedure are retained by a 100 000 MW cutoff dialysismembrane.l2

Unlike their O-glycosidic counterparts, the C-glycosidecopolymers of Neu5Ac are resistant to neuraminidase.Incubation of a sample of C-glycoside copolymer (XNeusnc= 0.20) with neuraminidase prior to an HAI assay did notresult in a decrease in inhibitory potency when comparedto the HAI results of the untreated copolymer sample.Under analogous conditions, the O-glycosidic copolymerIost its inhibitory activity.le

Aliquots of the monomers 3 and 5, and of each poly(5-co-acrylamide) (XN.u5Ac = 0.005-0.8), were prepared fortesting in an HAI assay by diluting the potential inhibitorin sterile phosphate buffered saline solution (PBS) andadjusting to pH 7.2. The solutions were employed in astandard HAI assay22,23 tre evaluate the potential of thesecompounds to inhibit agglutination of chicken erythrocytesinduced by influenza virus. This assay involves addingvirus to a solution of serially diluted inhibitor in a 96-welltiter plate, incubating the mixture for 30 min, adding asuspension of erythrocytes, and determining the lowestconcentration of Neu5Ac groups in solution for whichhemagglutination is nolonger observed underthe specifiedassay conditions. The endpoint of the HAI assay, K;(HAI),is defined as this lowest concentration. This assay dependson a number of variables, not all of which are undercomplete control (especially the quality of the erythro-cytes). The assay is reasonably reproducible (+1 well)provided appropriate controls are followed. The relevanceof this in vitro assay to in vivo infectivity remains to beestablished.

Figure 1 and Table I summarize the results of the HAIassay obtained with the poly(5-co-acrylamide) copol)rmerscontaining C-glycoside Neu5Acgroups and compares thesenew results with those obtained with the copolymers

l0{

104v

l0-2

Neu5Ac,l

=Kg*'\ 5 "o'kD,*"

:'fg a--^:-"tgr

H2NCO

Neu5Ac

l'

a. hocedure of Paulsen and Bednarski, ref 25,26;b. Aminoethanethiol hydrochloride, AICV, hv;c. N-acroyloxysuccinimide, Et3N, MeOH;d. Acrylamide, AICV, H2O, hv.

C-glycoside NeuSAcO-glycoside NeuSAc

780 Journal of Medic inal Chemistry, 1993, VoI .36, No. 6

Table I. Antihemagglutination Inhibitory Potency XN"ubA"-""

and XN.usn. for the C- and O-Glycoside Acrylamide Copolymersof Neu5Ac

copolymer Kr(HAI) , pM XN.uSA.t"* o XN"ubA"o 'S b

Sparks et al.

copolymer associated with the surface of the virus as itapproached the surface of the erythrocyte. Importantly,this mechanism of stabilization does nof, in principle,require enhanced binding of Neu5Ac groups to HA bindingsites (although it does not prohibit such enhancement);itis based more on the physical characteristics of thecopolymer when adsorbed on the surface of the virus thanon the biospecificity of that adsorption. Presently avail-able, but incomplete, information suggests that bothmechanisms contribute to the inhibition of hemaggluti-nation being observed here.

Two observations from the present work are relevant tothe mechanism of action of these copolymers. First, theconcentration of NeuSAc groups present in the C-glycosidecopolymer (at low values of XNeusAc) required to achievea given extent of inhibition of hemagglutination is lowerthan for the corresponding O-glycosidic copolymer; second,the maximum inhibition capacity of the C- and O-glycosidiccopolymers is very similar. We suggest as a hypothesis tobe tested in future work, that this difference reflects theability of the C-glycoside copolymer to interact with theneuraminidase binding site, without hydrolysis, in additionto interacting with the HA binding site. This ability wouldprovide an additional set of interactions that would attachthe C-glycosidic copolymer to the virus at low concen-trations of NeubAc groups. The observation that thisenhanced binding at low values of XN"usn. is not reflectedin an increase in the maximum value observed for K;(HAIIis qualitatively comparable with a mechanism for inhi-bition of hemagglutination based on steric stabilization;understanding its relevance to entropically enhancedbinding is more complicated.

Conclusion

Carbohydrates present on the surface of mammaliancells are ligands for surface lectins for a number ofbacteria2e and viruses.sO In most cases that have beenexamined, the monosaccharides themselves do not bindtightly and do not significantly inhibit pathogen-celladhesion. This general observation supports the hypoth-esis that it is the polyvalent character of the pathogen-host interaction that results in tight binding and thatpromotes infection.3l A number of studies have examinedthe effects of polyvalencys2'er and have reported thesyntheses of compounds that are polyvalent in knownbacterial and/or virus carbohydrate-based ligands.12-16 l1is clear that, in order to be effective in vivo, thesecarbohydrate-based compounds must also be resistant toglycosidic cleavage by pathogen and host-cell hydrolases.The C-glycosidic copolymers of NeuSAc described hereare as effective as the analogous O-glycosidic copolymersof Neu5Acl2 in inhibiting influenza virus-induced agglu-tination of erythrocytes in vitro and have the addedadvantage of being resistant to cleavage by viral neuramini-dase. At present we do not know whether these compoundsand copolymers are biologically active in vivo, and theyare currently being evaluated in infectivity assays.

Experimental Section

Materials and Methods. All solvents and reagents were fromAldrich and used without further purification unless otherwisenoted. Reaction mixtures were stirred magnetically and mon-itored by thin-layer chromatography on silica gel precoated glassplates (E. Merck, Darmstadt). Flash column chromatographywas performed on silica gel60rzsa (230-400 mesh, E. Merck) usingthe solvents indicated. Size-exclusion chromatography was

0.010.09

o XN",5A.-"* is the mole fraction of copolymer-linked NeuSAc thatyields the maximum antiagglutination potency. b XN.'5A"05 is theminimum mole fraction of copolymer-linked NeuSAc required toexert half of the maximum inhibitory potency above the thresholdinhibitory level given by monomeric Neu5Aca2Me in solution (K,tHAIt= 2.8 x 10-3 M).

i0-3 M)o and is shown as a reference for the KrtHAtt o1monomeric analogs of Neu5Ac.

Figure 1 permits two important comparisons betweenthe C- and O-glycosidic copolymers. First, both copoly-mers have similar maximum inhibitory potency in vitro.Four identical, independently prepared poly(5-co-acryl-amide) polymers having XNeubAc = 0.4 showed inhibitionof hemagglutination with &(HAr)

- | x 10-7 to 6 x 10-7 Min Neu5Ac groups. A parallel set of experiments conductedwith copolymers containing O-glycosidic Neu5Ac gaveKr(HAlt of 2 x 10-7 to 3 x 10-7 M in NeubAc groups.l2 Theslightly weaker inhibition of hemagglutination for theC-glycosidic copolymer (at XNeubAc > 0.1) relative to theO-glycosidic copolymer suggested by Figure 1 is withinthe margin of experimental error in the hemagglutinationassay (+1 well). Second, the minimum mole fraction thatis required toexerthalf of the maximum inhibitory potencyabove the threshold inhibitory level (i.e. l(rtHel) - 2.5 x10-3 M for NeubAca2Me, 1),6 XN",rsn.0'5, is shifted from0.09 for the O-glycosidic copolymer to 0.01 for theC-glycosidic copolymer. Together with XN.ubA.-*, the molefraction of copolymer-linked Neu5Ac that yields themaximum antiagglutination potency, the XNeuse.0'5 indexprovides a convenient means to compare the effectivenessof different types of copolymers containing neuraminicacid (Table I).

The monomeric c-C-glycosides 3 and 5 were also testedin an HAI assay. For these monomers,Kr(HAtr fell between2.5 and 5 mM; these values are similar to the values ofKr{HAtr obtained for most monomeric NeuSAc analogs.6-8The B-analogs 4, and the analogous NeubAc-acrylamide5 that was prepared from 4, showed no inhibition atconcentrations (20 mM, the highest concentration em-ployed in the HAI assays. Copolymers prepared from theB-analog of 5 (0.05-0.2 mole fraction of 0-NeuSAc 5 inacrylamide copolymer) also showed no inhibition by HAI.

Discussion

The mechanism by which these copolymers influencehemagglutination is stiilbeingdefined. Twobroad classesof mechanisms are now being evaluated. In the one, thepresence of multiple Neu5Ac moieties on the copolymerwould result in entropic enhancement of binding and ina higher occupancy of HA binding sites on the surface ofthe virus than in the presence of the same concentrationof NeuSAc groups present as monomers. In the other, thepresence of a hydrophilic copolymer that binds to viruswould result in the formation of a layer of copolymer atthe viral surface and would inhibit binding of virus toerythrocytes by a combination of interactions lumpedtogether under the term *steric stabilization'.28 Stericstabilization includes both enthalpic and entropic changesthat would accompany compression of the water-swollen

0.40.2

0.3-o.40.3-o.4

N euram,inidase - Resistant Hemagglutination Inhibitors

performed on Biogel P2 resin or Sephadex G10 using distilledwater as the eluant. Ion-exchange chromatography was per-formed on Dowex 50W-X8 (H+ form) cation exchange resin.Vortexing of copolymer samples was done with a Fischer vortexGenie II.

All melting points were obtained using a Mel-temp apparatusand are uncorrected. Proton and carbon NMR spectra weremeasured on a Bruker AM-400 MHz NMR spectrophotometer.Chemical shifts are reported in ppm relative to the solvent: CHCI3in CDCIg at7.24 ppm, HOD in DzO at 4.80 ppm, and CDH2OHin CD3OD at 3.30 ppm for the proton spectra. For carbon spectrathe references are 77.0 ppm for CDCI3,49.9 ppm for CD:OD, and1.30 ppm for CHgCN in DzO.

N-Acetylneuaminic acidwas obtained from extraction of edibleChinese swiftlet's nest.3a Erythrocytes from 2 week old chickenswere from Spafas Inc. Influenza virus (X-31) was obtained fromthe laboratory of Professor John Skehel. Phosphate-bufferedsaline used in the HAI assays was prepared from 80 g of NaCl,2 g of KCl, 11 g of NazHPO4, and 2 g of KHzPOT in 1 L of distilledHrO. This stock solution was diluted 1 part in 10 in distilledH:O and then adjusted to pH 7.2 with 1 N NaOH.

Hemagglutination Inhibition Assay.22'23 The HA titer ofa PBS stock solution of X-31 influenza virus was determined byserial dilution of 50 s.L of the virus solution through 12 microtiterplate wells (of a 96-well microtiter plate) each containing 50 pLof PBS. A suspension of chicken erythrocytes in PBS (100 rrl,)were added to each well, mixed, and incubated at 4 oC for t h.The HA endpoint is defined as the last dilution well beforeerythrocyte pellets begin to form and is expressed as a reciprocalof the endpoint dilution.

The monomeric inhibitors 3 and 5 were diluted to 80 mM inpBS (pH 7.2). The copolymers were also diluted in PBS withconcentrations of C-glycosidic Neu5Ac 5 ranging from 5 x 10-4to 4 x 10-2 M. The stock solution of the inhibitor (50 rrl,) wasserially diluted through 12 microtiter plate wells containing 50pL of PBS. To each well was added 50 rr.L of a suspension ofX-31 virus that had been diluted to the concentration of theendpoint in the titer, determined as described above. After a30-min incubationperiod (4 oC), 100pLof a suspension of chickenerythrocytes was added to each well, and the plate was gentlyagitated and then incubated at 4 oC for an additional 2 h. TheHAI endpoint is the last well where a red pellet is observed. Thisendpoint [KrtunItl is defined as the lowest concentration of sialicacid residues in solution that gave a 50% inhibition of hemag-glutination (incomplete pellet formation) by influenza virus (X-

31).35 The inhibition of hemagglutination by monomeric andcopolymeric sialosides was measured relative to Neu5Aca2Me I

[Krtuett = 2.8 mM].65-Acetamido-2,6-anhydro-3,5-dideoxy-2- G(2-propenyl )-D-

erythrer-mann(>non-2-ulosonic acid (3) was prepared asdescribed previously'25'26 mp 104 oC dec; tH NMR (400 MHz,CD3OD) 6 5 .91 (m, 1 H) ,4 .99 (m,2 H) ,3 .85-3.78 (m,2 H) ,3 .73-3.66 (m,2 H),3.61-3.45 (m, 3H),2.64 (dd, 1 H, J = 4.8,12.6H2),2.39 (m, 2H),1.92 (s, 3 H), 1.40 (dd, 1 H, J = 12.4,12.6H2);13CNMR (100 MHz, CD3OD) 6 179.4, L75.5,135.1, IL7.4 ,82.5,75.3,72.9,70.4,69.9, 64.5, 54.6, 46.0, 42.3, 22.6; high-resolution massspectrum (FAB) mlz 334.t502 t(M + H)*, calcd for Cr+HzrOaN,334.15021.

5-Acetamido-2,6-anhydro-3,5-d ideoxy -2- C(2- propenyl )-D-erythror-glueonon-2-ulosonic acid (4) was prepared asdescribed previously'25,26 mp 110-114 oC dec; lH NMR (400 MHz,DzO) 6 5.72 (m,1 H), 5.14 (m, 2 H), 4.04 (m, 1 H), 3.89-3.78 (m,4 H) , 3 .61 (dd, 1 H, J = 6 .3 , 12.6 Hz) , 3 .51 (d , 1 H, J = 8 .8 Hz) ,2.92 (dd,1 H, J = 6.0, 15.0 Hz), 2.53 (dd, 1 H, J = 8.0, 15.0 Hz),2 .32 (dd , 1H , J = 4 .70 ,13 .0 Hz ) ,2 .01 ( s ,3 H ) , 1 .74 (dd , 1H , J= 12.2,13.0 Hz); t3C NMR (100 MHz, DzO with CH3CN as aninternal standard, 1.3 ppm) 6 176.7,175.3, 132.0, 119.6, 80.1, 71.0,?0.5, 68.9, 67.2, 63.6, 53.0, 39.3, 35.9,22.6; high-resolution massspectrum (FAB) mlz 334.1512 t(M + H)*, calcd for CuHzaOaN,334.15021.

a-2 - C-13 -l[ 2 - ( N- Ac ryloy I am i no ) et hy U t hio ] p ropy U - N-acetylneuraminic Acid (5). The intermediate a-[3-[(2-ami-noethyl)thiolpropyll-C-glycoside was prepared via the methodof.Leez7 and Roy:ra C-glycoside 3 (85.9 mg,0.26 mmol) was addedto a 5-mL round-bottom Pyrex flask and dissolved in distilledHzO (1 mL) at room temperature. Aminoethanethiol hydro-

Journal of Medicinal Chemistr l ' , 1993, Vol. 36,,4/o. 6 781

chloride (2.0 equiv, 0'52 mmol, 58.3 mB) and 1y','1y'' 'azobis(iso-cyanovaleric acid) (AICV, 1 mg) were added to the reactionmixture, and the solution was purged with argon for 10 min. Themixture was irradiated with a medium-pressure Hg arc lamp for8 h at which time TLC and lH NMR examination of the crudereaction mixture indicated that the starting material had beenconsumed completely. Purification of the reaction mixture ona Biogel P2 size-exclusion column and lyophilization of the pooledfractions afforded the desired primary amine (90 mg, 0.22 mmol,85% yield) as a white foam: mp 240 oC dec; tH NMR (400 MHz,DzO) 6 3 .85-3.52 (m, 9 H) , 3 .18 ( t , 2H, J = 6 .3 Hz) , 2 .81 ( t , 2 H,J = 6 .3 Hz ' t , 2 .59 (m ,2 H ) ,2 .03 ( s ,3 H ) , 1 .83 -1 .43 (m ,4 H1 ; t r 6

NMR (100 MHz, CD30D) 6 179.8, 175.9, 82.4, i4 .3 ,72,9, 69.6,69. 1, 63.5, 53.2, 41.9,39.4, 39. 3, 3 1.5, 28. 7, 24.3, 22.8;high- resolut ionmass spectrum (FAB) mlz 433.1621 [(M + Na), calcd forCroHsoNzOeSNa, 433. 16041.

The a-Z-C -[ 3- [ (2-aminoethyl)thio] propyll -N-acetylneuramin-

ic acid (183 mg,0.43 mmol) was dissolved in MeOH (1.5 mL) andHzO (4.5 mL) at room temperature. Triethylamine (300 pL) andN-(acroyloxy)succinimide (prepared from acroyl chloride andN-hydroxysuccinimide, 1.3 equiv,0.56 mmol, 94 mg) were added,and the reaction was stirred for 4 h at room temperature. Thereaction mixture was maintained at pH - 10 with EtrN. Puri-fication

782 Journal of Medicinal Chemistry, 1993, Vol. 36, No. 6

solution purged with argon for 20 min. The test tube containingthe reaction solution was moved to a photoreaction chamberwhere the solution was irradiated with a UV lamp (365 nm) for6 h; as the copolymerization proceeded the solution becameviscous. The solution of copolymer was transferred by pipetinto an Eppendorf tube and diluted to 0.b mL with pBS (pH7.0). The copolymer solutions prepared in this fashion wereusually acidic (pH 3.0-a.5) and had to be titrated to pH 7.0 with1.0 N NaOH. Each solution containing dissolved copolymer wasvortexed briefly 110 s) prior to use in an HAI assay.

Ref.erences

(1) This work is part of a collaboration with Professors J. J. Skehel,J. R. Knowles, M. Karplus, and D. C. Wiley. Support was providedby NIFI Grant GM 39589.

(2) NIH postdoctoral fellow 1991-1993 (CA 0912?-01).(3) NSF postdoctoral fellow 1990-1992 (CHE-9002653).(4) Matrosovich, M. N. Towards the Development of Antimicrobial

Drugs Acting by Inhibition of Pathogen Attachment to Host Cells:A Need for Polyvalency. FEBS Lett. 1989,252,I-4.

(5) Weis, W.; Brown, J. H.; Cusack, S.; Paulson, J. C.; Skehel, J. J.;Wiley, D. C. Structure of the Influenza Virus HemagglutininComplexed with Its Receptor Sialic Acid. Na"ture 1988, JJJ, 426-431. Wiley, D. C.; Skehel, J. J. The Structure and Function of theHemagglutinin Membrane Glycoprotein of Influenza Virus. Annu.Reu. Biochem. L987,56, 365-394. Ruigrok, R. W. H.; Andre, P. J.;Hoof Van Huysduynen, R. A. M.; Mellema, J. E. Characterizationof Three Highly Purified Influenza Virus Strains by ElectronMicroscopy. J. Gen. Virol. 1985, 65,799-802.

(6) Sauter, N. K.; Bednarski, M. D.; Wurzburg, B. A.; Hanson, J. E.;Whitesides, G.M.;Skehel, J. J.;Wiley, D. C. Hemagglutinins fromTwo Influenza Variants Bind to Sialic Acid Derivatives withMillimolar Dissociation Constants: A 500-MHz Proton NuclearMagnetic Resonance Study. Biochemistry lg8g, 28, 8988-8896.Pritchett, T. J.; Brossmer, R.;Rose, U.; Paulson, J. C. Recognitionof IV[onovalent Sialosides by Influenza Virus H3 Hemagglutinin.Virology 1987, 160, 502-506.

(7) DeNinno, M. P. The Synthesis and Glycosidation of N-Acetyl-neuraminic Acid. Synf hesls 1991, 483-493. Appropriately placedaromatic substituents have been reported to enhance monomericbinding: Toogood, P. L. ;Gal l iker , P. K. ;Gl ick, G. D. ;Knowles, J.R. Monovalent Sialosides That Bind Tightly to Influenza A Virus.J. Med. Chem. 1991, 34, 3140-3143. See also ref b and Paulson,J. C. Interactions of Animal Viruses with Cell Surface Receptors.In The Receptors; Conn, P. M., Ed.; Academic: Orlando, 1985;Vol . 2, pp 131-219.

(8) Hydrophobic N-acyl analogs of neuraminic acid have been preparedand examined by HAI: Sparks, M.A.; Will iams, K. W.; Lucaks,C.; Schrell, A.; Priebe, G.; Spaltenstein, A.; Whitesides, G. M.Synthesis of Potential Inhibitors of Hemagglutination by InfluenzaVirus: Chemoenzymic Preparation of N-5 Analogs of N-Acetyl-neuraminic Acid. Tetrahedron 1993,49, 1-12. Schmid, W.;Avila,L. Z.; Will iams, K. W.; Whitesides, G. M. Synthesis of Methyla-Sialosides N-Substituted with Large Alkanoyl Groups, andInvestigation of Their Inhibition of Agglutination of Erythrocytesby Influenza A Virus, BioMed. Chem./,eft., submitted.

(9) A new Neu5Ac derivative, 6'-(((napthylmethyl)amino)carbonyl)-hexyl 5-acelamido-3,5-dideoxy-4-O- ( ( (5- (dimethylamino)napth- t-yl) sulfonyl ) glycyl ) - o-g/yc e r o - a-D- galacf o-nonulopyranosidonic acid(i), has been prepared that inhibits hemagglutination at 3.? x 10 6M: Knowles, J. R.;Weinhold, E. Design and Evaluation of a TightlyBinding Fluorescent Ligand for Influenza A Hemagglutinin. ./.Am. Chem. Soc. 1992. 114.927V9275.

Sparks et al.

gosaccharide-Acrylamide Copolymers. Glycoconjugate J. lg8g, 8,597-611. Roy, R.; Tropper, F. D. Synthesis of Copolymer AntigensContaining 2-Acetamido-2-deoxy-a- or B-n-glucopyranosides . G ly -coconjug,ate J. 1988, 5,203-2A6. Roy, R.;Tropper, F. D. Synthesisof Antigenic Carbohydrate Polymers Recognized by Lectins andAntibodies. J. Chem. Soc., Che.m. Commun. 1988, 10b8-1060.Horeksi, V.; Smolek, P.; Kocourek, J. Water-soluble O-glycosylPolyacrylamide Derivatives for Specific Precipitation of Lectins.Biochim. Biophys. Acta 1978,538, 293-298.

( 1 1) Equine d2-macroglobulin (equine azM) is the most potent inhibitorof influenza viral adhesion known, see: Pritchett, T. J.; Paulson,J. C. Basis for the Potent Inhibition of Influenza Virus Infectionby Equine and Guinea Pig a2-Macroglobulin. J. Biol. Chem.lg9g,264, 9850-9858. Other naturally occurring polyvalent inhibitorsinclude orosomucoid (Morawiecki, A.; Lisowska, E. PolymerizedOrsomucoid-An Inhibitor of Influenza Virus Hemagglutination.Biochem. Biophys. Res. Commun. 1965, 18,606-€10), fetuin (Spiro,R. G. Studies on Fetuin, a Glycoprotein of Fetal Serum. J. Biol.Chem. 1960, 235, 2860-2869), and mucin (Gottschalk, A.; I\{cKenzie,H. A. Studies on Mucoproteins VIII. On the Molecular Size andShape of Ovine Submaxillary Gland Mucoprotein. Biochim.Biophys. Acto 1961, 54,226-235). See also ref 15.

(12) Spaltenstein, A.; Whitesides, G. M. Polyacrylamides BearingPendant a-Sialoside Groups Strongly Inhibit Agglutination ofErythrocytes by Influenza A Virus. J. Am. Chem. Soc. lggl, i/J,686-68?.

(13) Liposomes that incorporate a sialyl ganglioside inhibit hemagglu-tination as well as or better than the best-known natural inhibitors(see ref 11): Kingery-Wood, J. E.; Will iams, K. W.; Sigal, G. B.;Whitesides, G. M. The Agglutination of Erythrocytes by InfluenzaVirus is Strongly Inhibited by Liposomes Incorporating an Analogof Sialyl Gangliosides. J. Am. Chem. Soc. 1992, 114,7908-7905.

(14) Roy, R.; Laferriere, C. A. Synthesis of Antigenic Copolymers ofN-Acetylneuraminic Acid Binding to Wheat Germ Agglutinin andAntibodies. Carbohydr. Res. 1988, 177,C1-C4.

(15) Matrosovich, M.N.;Mochalova, L. V.; Marinina, V. P.;Byramova,N. E.; Bovin, N. V. Synthetic Polymeric Sialoside Inhibitors ofInfluenza Virus Receptor-Binding Activity. FEBS Lett. lgg0, 272,209-212. Roy, R.; Laferriere, C. A. Synthesis of Protein Conjugatesand Analogues of N-Acetylneuraminic Acid. Can. J. Chem. lgg0,68,2045-2054. Roy, R.;Laferriere, C. A. Michael Addition as theKey Step in the Synthesis of Sialyloligosaccharide-Protein Con-jugates from N-Acroylated Glycopyranosylamines. J . Chem. Soc.,Chem. Commun.1990, 1709-1711. Gamian, A. ; Jennings, H. J. ;Laferriere, C. A.; Roy, R. Inhibition of Influenza A Virus Hemag-glutinin and Induction of Interferon by Synthetic SialylatedGlycoconjugates. J. Carbohydr. Chem.1987, 6, 161-165.

(16) Several compounds that are divalent in Neu5Ac have been preparedand shown to inhibit hemagglutination approximately 102 betterthan NeuSAca2Me (1) : Gl ick, G. D. ;Toogood, P. L. ;Wi ley, D. C. ;Skehel, J. J.; Knowles, J. R. Ligand Recognition by InfluenzaVirus: The Binding of Bivalent Sialosides. J. Biol. Chem. 199L,266,2366A-23669. Glick, G. D.;Knowles, J. R. Molecular Recog-nition of Bivalent Sialosides by Influence Virus. J. Am. Chem.Soc. 1991, 113, 470L-4703. For the synthesis and HAI dataconcerning cluster inhibitors (divalent in Neu5Ac) see: Sabesan,S.; Duus, J. O.;Domai l le, P. ; Kelm, S. ;Paulson, J. C. Synthesis ofCluster Sialoside Inhibitors for Influenza Virus. J. Am. Chem.Soc. 1991. /13. 5865-5866.

(17) Burmeister, W. P.; Ruigrok, R. W.; Cusack, S. The 2.2 A ResolutionCrystal Structure of Influenza B Neuraminidase and Its Complexwith Sialic Acid. EMBO J. 1992, 11,49-56. Air, G. M.; Laver, W.G. The Neuraminidase of Influenza Virus. Proteins lg8g, 6, 941-356.

(18) NA activity presumably prevents these molecules from acting asinhibitors in vivo, as several NeuSAc-containing glycoproteins havebeen shown to prevent binding of virus to cell in vitro: see ref l1and Hanaoka, K.; Pritchett, T. J.; Takasaki, S.; Kochibe, N.;Sabeson, S.; Paulson, J. C.; Kobata, A. 4-O-Acetyl-N-acetyl-neuraminic Acid in the N-Linked Carbohydrate Structures ofEquine and Guinea Pig a2-Macroglobulins, Potent Inhibitors ofInfluenza Virus Infection. J. Biol. Chem. 1989,264.9842-9849.

( 19) Polymers containing side chains terminating in O-glycosidic NeuSAcgroups are rendered ineffective in inhibiting viral hemagglutinationof chicken red blood cells by pretreatment with NA: Spaltenstein,A.; Whitesides, G. M. Unpublished results.

(20) Palese, P.;Tobita, K.;Ueda, M.;Compans, R. W. Characterizationof Temperature Sensitive Influenza Virus Mutants Defective inNetrraminidase. Virology I974, 61,397-410.

(21) The complete IUPAC name for this compound is 5-acetamido-2,6-anhydro-3,5-dideoxy-2- C- I3- t t2-(N-acry loylamino)ethyl l -thiol propyll -o- e ry thro-r- rnanno-nonulosonic acid.

(22) Ki lbourne, E. D. Bul l . W.H.O.1969, 41, 643.(23) Rogers, G. N.; Pritchett, T. J.; Laue, J. L.; Paulson, J. C. Differential

Sensitivity of Human, Avian, and Equine Influenza A Viruses toa Glycoprotein Inhibitor of Infection: Selection of Receptor SpecificVariants. Virology 1983, 131, 394-408. W.H.O. Tech. Rep. Ser.1953.64. r-32.

NM%

t(10) Polymers of other carbohydrates have been prepared in attempts

to mimic ligands for a variety of lectins: Roy, R.; Tropper, F. D.;Romanowska, A. New Strategy in Glycopolymer Synthesis. Prep-aration of Antigenic Water-Soluble Poly(acrylamide-co-p-acryl-amidophenyl d-lactoside). Bioconj. Chem. 1992, 3, 256-2G1.Chernyak, A. Y.; Sharma, G. V. M.; Kononov, L. O.; Krishna, P.R.; Rao, A. V. R.; Kochetkov, N. K. Synthesis of Glycuronamidesof Amino Acids, Constituents of Microbial Polysaccharides andTheir Conversion into Neoglycoconjugates of Copolymer Type.Glycoconjugate J. 1991, 8, 82-89. Kosma, P.; Waldstatten. P.:Daoud, L.;Schulz, G.;Unger, F. M. Synthesis of Poly(acrylamide)-copolymers Containing 3,5-Dideoxy-n-arabino-2-octulopyranosy-Ionic (5-deoxy-KDO) Residues. Carbohydr. Res. 1989, 194,145-154. Kallin, E.;Lonn, H.;Norberg, T.; Elofsson, M. DerivatizationProcedures for Reducing Oligosaccharides, Part 3: Preparation ofOligosaccharide Glycosylamines, and Their Conversion to Oli-

N euraminidase- Resistant Hemagglutination Inhibitors

(24) Ogura, H.; Furuhata, K.; Itoh, M.; Shitori, Y. Synthesis of 2-O-Glycosyl Derivatives of N-Acetyl-o-Neuraminic Acid. Carbohydr.Res. 1986, 1 58, 37-bI.

(25) Paulsen, H.; Matschulat, P. Synthesis of a C-glycoside of N-Acetyl-neuraminic Acid and Derivatives. Liebigs Ann. Chem.l991, 487-495.

(26) Nagy, J. O.; Bednarski, M. D. The Chemical-Enzymatic Synthesisof a Carbon Glycoside of N-Acetylneuraminic Acid. TetrahedronLett. 1991, 32, 3953-3956.

(27) Lee, R. T.; Lee, Y. C. Synthesis of 3-(2-Aminoethylthio)propylGlycosides. Carbohydr. Res. L97 4, 37, L93-201.

(28) Everett, D. H. Basic Principles of Colloid Science; Royal Societyof London: London, 1988; pp 45-51.

(29) Lindahl, M.;Brossmer, R.;Wadstrom, T. Carbohydrate ReceptorSpecificity of K99 Fimbriae of Enterotoxigenic Escherichia coli.Glycoconjugate J. 1987, 4,51-58. Ishikawa, H.; Isayama, Y.Evidence for Sialyl Glycoconjugates as Receptors for Bordetellabronchiseptica on Swine Nasal Mucosa. Inf ect. Immun.1987, 55,1607-1609. Lindberg,A. A.; Brown, J. E.; Stromberg, N.; Westling-Ryd, M.; Schultz, J. E.; Karlsson, K.-A. Identification of theCarbohydrate Receptor for Shiga Toxin Produced by Shigelladysenteriae Type 1. J. Biol. Chem.lg87 ,262,1779-1785. Murray,P. A.; Levine, M. J.; Reddy, M. S.; Tabak, L. A.; Bergey, E. J.Preparation of a Sialic Acid Binding Protein from Strepf ococcuEmif is KS32AR. Infect . Immun.1986, 53, 359-365. Bose, S. K. ;Smith, G. B.; Paul, R. G. Influence of Lectins, Hexoses andNeuraminidase on the Association of Purified Elementary Bodiesof Chlamydio Trachomotis UW-31 with HeLa Cells. Infect.Immun.1983, 40, 1060-106?. Levy, N. J. Wheat Germ AgglutininBlockage of Chlamydial Attachment Site: Antagonism by N-Acetyl-o-Glucosamine. Infect. Immun. 1979, 25, 946-953. Ofek, I.;Mirelman, D.;Sharon, N. Adherenceof Escherichia coli to HumanMucosalCellsMediated byMannose Receptors. Nature 1977,265,623-625. Kuo, C.-C.; Wang, S.-P.; Grayston, J. T. Effect ofPolycations, Polyanions and Neuraminidase on the Infectivity ofTrachoma-Inclusion Conjunctivitis and Lymphogranuloma Ve-nereum Organisms in HeLa Cells: Sialic Acid Residues as PossibleReceptors for Trachoma Inclusion Conjunctivitis. Infect. Immun'1973, 8, 74-79. Carbohydrates in bacterial adhesion, reviews:Isberg, R. R. Discrimination Between Intracellular Uptake andSurface Adhesion of Bacterial Pathogens. Science 1991,252,934-938 and references cited therein. Sharon, N. Bacterial Lectins,Cell-Cell Recognition and Infectious Disease. FEBS Lett.1987,217,145-15?. Reid, G.; Sobel, J. D. Bacterial Adherence in thePathogenesis of Urinary Tract Infection: A Review. Reu. Inf ect.

Journal of Medic inal Chemistr t . 199,3, \ ' r , i . ;36. . \ 'o 6 783

Dis.1987,9,47V487. Beachey, E. H. Bacter ia l Adherence Adhesrn-Receptor Interactions Mediating the Attachment oi Bacieria toMucosal Surfaces. J. Infect . Dis. 1981. 1.13.3:5- .115.

(30) Tavakkol , A. ; Burness, A. T. H. Evidence for a Direct Role iLrrSialic Acid in the Attachment of Encephaloml'ocarditis Virus trrHuman Erythrocyte s. Biochemistry I 990, 29, 1 0684- I 069(l Eatnn.B. T.; Crameri, G. S. The Site of Bluetongue Virus Attachment t.,Glycophorins from a Number of Animal Erythrocltes. J. Gen.Vlrol. 1989, 70,3347-3353. Paul, R.W.;Lee, P. W. K. Gly'cophorrnis the Reovirus Receptor on Human Erythrocytes. Virologl 1987.159,94-lC]r. Allaway, G. P.;Burness, A. T. H. Site of Attachmentof Encephalomyocarditis Virus on Human Erythrocytes. J. Virol.1986,59, 768-770. Fried, H.;Cahan, L. D.;Paulson, J. C. PolyomaVirus Recognizes Specific Sialyloligosaccharide Receptors on HostCells. Virology 1981, 109, 188-192.

( 3 1 ) See ref 15, Matrosovich et al. Rao, B. N. N.; Moreland, M.; Brandley,B. K. The Potential of Carbohydrates as Cell Adhesion Inhibitors.Med. Chem. ftes. 1991, .1, 1-8.

(32) Ponpipom, M. M.; Bugianesi , R. L. ;Robbins,J. C. ;Doebber,T.W.;Shen, T. Y. Cell-Specific Ligands for Selective Drug Delivery toTissues and Organs. J. Med. Chem.1981,24,1388-1395. Lee, Y.C. Synthesis of Some Cluster Glycosides Suitable for Attachmentto Proteins and Solid Matrices. Carbohydr. Bes. 1978, 67,509-514. Kawasaki, T.; Etoh, R.; Yamashina, I. Isolation and Char-acterization of Mannan-Binding Proteins from Rat Liver. Bioc hem.Bio phy s. R e s. C ommun. 1978, 81, 1018-1024' Stahl, P. D.; Rodman,J. S.; Miller, M. J.; Schlesinger, P. H. Evidence for Receptor-Mediated Binding of Glycoproteins, Glycoconjugates and Lysos-omal Glycosidases by Alveolar Macrophages. Proc. Natl. Acad.Sci. U.S.A. 1978. 75, 1399-1403. Kornfield, R.; Kornfield, S.Comparative Aspects of Glycoprotein Structure. Annu. Reu.Biochem. 197 6, 45, 217 -237 .

(33) Singer, S. J. Intercellular Communication and Cell-Cell Adhesion.Science 1992,255, 167l-1677. For discussions concerning poly-valency in cellular communication of the immune system seeSpringer, T. A. Adhesion Receptors of the Immune System. N ature1990.346. 425-434.

(34) Czarniecki, M. F.; Thornton, E. R. Carbon-13 Nuclear MagneticResonance Spin-Lattice Relaxation in the N-Acylneuraminic Acids.Probes for Internal Dynamics and Conformational Analysis. J.Am. Chem. Soc. 1977. 99. 8273-8279.

(35) Since the inhibition values (K,tuntt; are calculated based on theconcentration of NeuSAc functional groups in solution they can becompared directly for monomeric and polymeric derivatives ofNeu5Ac.

371(1).PDF371(2).PDF371(3).PDF371(4).PDF371(5).PDF371(6).PDF