- Email: [email protected]
Effect of polyethylene glycol on the rate of immune complex formation by non-precipitating antibody
Imotunoh).e.v Letter;. 4 I l q 8 2 p 4 9 E Ise~ ~et Binmed teal Pre,~*
52
EFFECT OF POLYETHYLENE GLYCOL ON THE RATE OF IMMUNE COMPLEX FORMATION BY NON-PRECIPITATING ANTIBODY Frantigek RYP.-((~EK. Dagmar SIRAKO\",-(, Panel KRATOCHV[L. Jarosla~ DROBN[K and Frantigek FRANI~K* Institute ol Macromolecular Chemistry. Czechoslorak .4cademy o f Sciences. 162 O0 Prague O. a~ltl *Institute o.tMolccular Genetics. C-eehoslo,'ak Academ.v o] Sciences. 142 20 Ptuute 4. C:echodovakia ; Received I 3 June 1981 I.Accepted 29 Jul.~ 1981p
I.
S i l i n nlar}. '
A kinetic study formation of large size complexes of non-precipitating pig-Drip anti-Drip antibody and muhl~alent di,fitrophen.~ lated serum albumin was performed using light scattering and absorption spectroscop3 of the Dnp-group. A vet 3 rapid phase of the process resulted in the formation of c o m p l e \ e s having molecular xxeight of about 2 X 10". Further increase of the COnlplex size ~as much slo~ er. Addition of PEG affected positivel3 the rate of complex growth even in concentratu.ms below 1"~-. The spectroscopic kinetic curves also showed a rapid and a slox~ phase. sensitive to the presence of PEG. The character of the kinetic data does not support the simple vlex~ that pol 5 mers enhance precipitate-formation by the steric exclusion of complexes from the pol~ met domaina. It can be assumed that the interaction of the pol 5 mer with the a n t i g e n - a n t i b o d y system consists of a subtle temporary attachment of the pol,, mer to the anubod,, molect, le resuhing in a change of the shape and/or llexibility of the antibody molecule, fa~ ouring its cros.~-hnking c a p a c i b .
2. Introduction Precipitating anti-hapten antibodies combined with muluvalent h a p t e n - p r o t e m conjugates precipi.4bbreviations: Dnp, 2,4-diniuophen.~ l: PEG. pol,~ eth.~ lene gl~ col; PBS. phosphate-buffered saline, i.e. 1511 mM NaCI. 20 mM phosphate pH 7.4. O105
2478.,'82[0f1(,i-0000,'$2 75 ,g, Else'.ier B.)med]cal Press
tate under formation of a three-dimensional lattice in which antigen molecules are bridged by bix alent il]Olecules of antibod3. The formation of the antigen antibods p~ecipitate can be potentiated b s addiuon to t i n s s.,, s t e m o f certalrl m a c r o m o l e c u l a r
substances
such as dextran [1] or PEG [2]. This effect seemed to be full.,, understood in terms of po[.,, met incompatibilit3 or a steric exchision of the complexes from the domains of the pol) mers [3]. More recently. F,an~k xxith co,workers [4.5] found that addition of h',drophilic pol.~ mer~ can also promote precipitation in the s.~stem of a h a p t e n - p r o t e i l l conjugate with noniprecipitating antibody, winch otherwise forms no insoluble product, but onl.~ small soluble complexes with the antigen. This effect occurs at a relatively low concentration of the polymer I I - o ) a n d i s s t r o n ~ 5 dependent on the molecular v, eight of the added polymer. When the polymer is ~ashed out. the precipitate remains insoluble having properties of a true a n t i g e n - a n t i b o d y lattice [5]. These findings led us to suggest that a more complicated mechanistn is imolved than simpl.~ the solvent exclusion. The aim of this v, ork has been to tollow the kmeucs of formation of complexes betu, een a h a p t e n - p r o t e i n conjugate and pig non-precipitating antibody of the lgG class in the presence of PEG. We used a low polymer concentration, in ~xhich no x isible precipitate Is formed. The size of the complexes was estimated by lid~t scattering. A complementary analysis of the gro~ th of the antigen-antibod5 lattice x~as performed b v absorption spectroscop.~ ,_ffthe hapten chromophore. 49
3. Materials and methods 3.1. Materials Non-precipitati,lg anti-Dnp antibody was isolated from sera of immunized pigs at tire decline of tile antibody response [5]. Pig serum albutnm contanling 40 Dnp-groups pet molecule {Dnpa,,-serunl albumin ~ ~as prepared according to Eisen et al. [6]. Pig se~t,m albumin containing 5.5 Dnp-groups per molecule I DnPs.5-serum albttmh~ ) ~as prepared by a modified procedure, in ~ hich 10 times less dinitrobenzene sulfonate was added than ill the standard procedure [6]. Polyeth.~ lene ~ y c o l 0000 I P E G ) ~ as purchased from Do~ (Midland. MIL Its nrolecular weight x~as tn tile range 6 0 0 0 - 7 0 0 0 . Chemicals used for buffers ~,,ere of arval3 tical purit.,, grade.
3.2. Methods 3.2.1. Measurement of molecular weight of complexes b} licit scattering Stock solutions ofanti-Dnp antibody [0.875 mg/ml in PBSI, Dnp.,cserum aJbunlJll (0.103 nlg/nll m PBS), PEG (60 rag/nil, 40 nlg/ml, 20 nrg/ml in PBSI and PBS alone were spun for 30 mhl at 2 5 , 0 0 0 g on the ultracentrifuge MSE 65 to temo,,e dust particles. Sanlples for light scattering v, ere prepared by mLxing 4.0 ml of antibody solution with 2.0 ml of PEG solt.tton (or PBS as a control ), and finall} with 4.0 ml of Dnp.~,,-serum albunliJ1 sohttion. 1"he final concentration of mrtibod.~ ~ as 0.35 mg/ml, of Dnp4o-serum aJbumin 0.04 mg/ml, and the concentrations of PEG were 4.8 and 12 mgjml. The angular dependence of the intensity of scattered li~lt ,.,.'as measured at 25°( ' at several time intervals, starting fronl the point of mixing, with a PhotoGonio-Diffusonletre Sofica apparatus over the angular range 3 0 - 1 5 0 " ( ' . using a ,.ertically polarized primary beanl ~itll a wavelength [m ~acuo) 546 nnl. Tile experimental values interpolated for pat ticular times of react,on [7] ~ere plotted as Kc,:Reo against sin-~l®/2 h K being the optical constant, c the protein concentratmn, and R o tile Ra) le~gh ratio for the angle ofobser~ ation ®. Tile refractive index inclement used in the calculation of K was determined ~ittl a Brice-Phoenix differential refractometer, model BP-2OOO-V: its value ~as 0.189 g/nr1125°C. 546 nnl). 50
l-tie KciRe)=~,.~ w..due was obtained b} extrapolation to zero angle of obserxation [7] and its reciprocal ~as taken as the weight-a~erage molecular wei~lt. M~, of tile s)stenl. Suite M~ depends on tile concentration of the antibod~ and antigen, its ,,alue could not be obtained m tile rigorous wa.~, I.e. b) extrapolation t)f Kc,,'Rt:) to mfimte dilution. Tile error thus introduced is likel} not to be qualitative b ~ignlficant, because tile concentlation dependencies of Kc,,'R,,j m s.~stems of this type are usuall\ ~ery small. 3.2.2. Absorption spectroscop5 nleasutenlents .Mrti-Dnp antibody. PEG [or PBS ) and DnP5.5-serum albttllli~l ~ere mLxed dtrect[3 m tile test cuvette of the spectrophotometer from the stock soluttons. Tile final concenttation of tile an nbod,, ~ as 0.35 nlg/ml. that of DnP5.5-serum albt.mhl ~as 0.04 mg,,nl[ and that of PEG ~as 12 mg/nd, hnnlediatel.~ after mL\mg. tile absorphon spectra of the Dnp-moiet,, ~ere recorded in the range 3 5 0 - 3 0 0 nm, repeating the scans in 5 nlin intervals. One scan took 24 s. The wa~elength of tile absorption maximum ~as determnled from tile records graphically using tile filst deri~atixe plot. All spectroscopic measurements were done x~ith a \'arian Superscan 3 LiV-Vis specttophotometer.
4. Results and discussion Since the enhancing effect of PEG on tile formation of insoluble arrtigen-antibodb complexes is most pronounced at low antigen to antibody ratios [5], tile molar ratio antigen to antibod). 1:4 v, as chosen fur our expemnents, file system reached the level of tile molecular weight of complexes about 2 X 10 ° practicall,,' nnmedlately. A complex of this size can be r e d resented, e.g.. b', 12 anubod.,, molecules and 3 antigen molecules. Further increase of tile complex s~ze has been found to be much slov, er (Fig. 1 ). The slow rate may indicate that rearrangement of conlplexes is necessar~ to nlake ~acant antibod,, binding sites steritally a~ailable for further interactions. Tire restriction ot further increase in size is nlost obvious in tile control system ~ithout PEG. The size of tire complexes rises so slow[ 5 that it cannot reach the molecular ~ eight necessar~ for precipitation within a realistic time. The eftect ot increased concentration of PEG
/
2,,0
i
i
8
r.7 "
I
I
E
/
C
/ w
,/¸¸¸
¢ //
,n,)
/
/ /
/ / t/'-'-
/
J "[3
1,Z,,J
a ["
~ 2'
i
. '."J
-'7") -1-i,~ IT,ICI
Fig. 1. Increase of the v, eight-a~erage m o l e c u l a r v, mght, ~1,.,,. o t c o n | p l e ' , e s formed in the s~ _,tem c o n t a i n i n g the non-precipi t a t l n g a n t i - D n p a n t i b o d 3 . D n p ~ , . - ~ e r t n a b t m i n , and PEG. I = c o n t r o l ~ i t h o u t PEG. 2 = ~L4"7 PEG; 3 = It8"" PEG: 4 = 1.2 ~ PEG.
itself in a continuous increase of the rate of for mation of b i l l molecular-~ eight complexes. At the concentration of PEG 1.2";. the weight-a~etage molecular-v, ei~at of the complexes tends to approach an infinite vMue and ~isible precipitation is formed within 3 h. The shape uf the time dependence resemmanifests
0.6
,%
I
1
,,:¢
0.4
02
300
400
500 ,nm
Fig. 2. A b s o r p t i o n spectra of free (curve 1 ) and ant,bod.~b o u n d |.curve 2~ 6 - D n p - a m i n o c a p r o i c acid. The aamples were d i s s o h e d in PBS. annbod.~ ~ a,,_ a d d e d ,n e\ce~s. X I = 364.3 nm, a = 371.8 nm.
I
I
I
4O
8O
120
160 time, m,n
Fig. 3. Increase of the ~tutt of the Dnp-group ab,orptton maxHnunl I n the s5 stem c o n t a i m n g non-precipJtatin?, a n t i - D n p antibod.~. Dnp<~-sert||n a l b u m i n and PEG. : . c o n t r o l t t i t h o u t P E G . o . 1.2"; PEG.
bles the cur~e for a cross-hnkmg reaction approaching the gel point [8]. Tile absorption spectrum of tile Dnp-gt oup is shifted to longer wavelengths ~hen the group is inserted in the combining site of the anti-Dnp antibud.v I Fte. 2 L The magnitude of the shift is proportional to the ratio of bound and flee Dnp-groups in the system. The kinetics of formation of antigen-antibod,, bonds '. ia combining sites ~ ere ff)llowed b~ measuring the shifts of maxima of the Dnp-group absorption band dt, ting the interacmon of DnP5.5-serum albumin ~ith nonprecipitating anti-Dnp antibody without and in the presence of 1.2'. PEG IFig. 31. The kinetic curves indicate that the interaction of the haptenic groups with antibod 3 combining sites ts facditated in the presence of PEG. The kinetic data. obtained b 3 light scattering on the one hand and b~ absorption spectroscop.~ on the other, are not dlrectl 3 comparable due to the different functionalit 3 of the antigens emplo.~ ed. A Io~ content of Dnp-groups v. as necessary for obtaining sufficient resohttion m absorption spectra. In spite of this difference, the data obtained a l l o ~ u s to draw certain conclusions. The h.,, pothesis based on the soh'ent exclusion effect of pu b mers can be rejected. The 51
kinetics of complex
lormation,
as s h o w n b.~ l i g h t
scatter|rig, is entirely different from the kinetics of phase separation. A gradual increase m PEG concentration causes a gradual, and not jump-like, increase in the rate of tire molecular-x~eight growth o f tire complexes, ln te~ action of the polymer ~ ith a n t i g e n antibody complex should be considered as a subtle temporar5 attachment of the pol5 mer molecule to the antibod.,, molecule. The inabilit5 of non-precipitating antibodies to precipitate multivalent antigens may be understood as a tendenc) to prefe|ential formation of c5 clic structures with limited molecular wei~]t. Due to the interaction ~ ith h 5 drophihc macromulecules, the shape and,'or llexibilitx of ant> bod3 molecules ma.~ be changed in Stlch a ~Aa} that, during formation of the lattice, the ClOss-hnkllig ~eac|ion is favoured instead of the formation of c3 ties. The lesults of absorption measurements confirm that bonds bet~ een Dnp-group~ I epitopes I and combining sites (paratopes) are im olved in the reaction, althou~l
52
the existence of additional intermolecular bonds cannot be ruled out.
References [ I ] Hellsing, K. and Laurem, T. C. C19641 Acta Chem. Stand. 1~, 1 3 1 l j - ] 3114. [2] H,Jrrlll~t,an, J. C., Fenb)n, J. ~... II and Pert, J. H. ~1971 ~ lnm]unoc hemi~!.r.~ 8, 4 1 3 - 4 2 ] . [3] Hell~in~,K. I I 9 6 9 ) Biochen1. J. I 1 2 , 4 7 5 48]. [41 Fran.~k. F. i [ 9 7 3 } ill: Antibod.~ Structure and Molecular Immunolog.~. FE BS Symposia, Vol. 36 (Gerge[.~, ,I. and Medg~esi, G. ~ . e d , . j p p . 63 "5, Akad&nial Kl3.dt:~, Budape>t. [31 Fr,m~'k, F., O1;'o* ~k,i, Z. and Simek, L. 11979) Eur. J. Immunol. 9 , 6 9 6 - 7 0 1 . 161 Eisen. H. N., Belman, S. and Caz_,ten, M. E. i 1953'I J. Am. Chem. Soc. ~5. 4 3 8 3 - 4 5 8 3 . [ ~'1 Kratoch~ il. P., Munk. P. and Sedl:K~ek, B. i 196] I Coll. Czech. Chem. Commun. 20, 1274 - 1282. [81 Flor~., P. J. I19531 Principle,, of Pol) nler Chembtr.~, p. 383. Cornell Universu.', Pre~, Ithaca, Ne~ York.
52
EFFECT OF POLYETHYLENE GLYCOL ON THE RATE OF IMMUNE COMPLEX FORMATION BY NON-PRECIPITATING ANTIBODY Frantigek RYP.-((~EK. Dagmar SIRAKO\",-(, Panel KRATOCHV[L. Jarosla~ DROBN[K and Frantigek FRANI~K* Institute ol Macromolecular Chemistry. Czechoslorak .4cademy o f Sciences. 162 O0 Prague O. a~ltl *Institute o.tMolccular Genetics. C-eehoslo,'ak Academ.v o] Sciences. 142 20 Ptuute 4. C:echodovakia ; Received I 3 June 1981 I.Accepted 29 Jul.~ 1981p
I.
S i l i n nlar}. '
A kinetic study formation of large size complexes of non-precipitating pig-Drip anti-Drip antibody and muhl~alent di,fitrophen.~ lated serum albumin was performed using light scattering and absorption spectroscop3 of the Dnp-group. A vet 3 rapid phase of the process resulted in the formation of c o m p l e \ e s having molecular xxeight of about 2 X 10". Further increase of the COnlplex size ~as much slo~ er. Addition of PEG affected positivel3 the rate of complex growth even in concentratu.ms below 1"~-. The spectroscopic kinetic curves also showed a rapid and a slox~ phase. sensitive to the presence of PEG. The character of the kinetic data does not support the simple vlex~ that pol 5 mers enhance precipitate-formation by the steric exclusion of complexes from the pol~ met domaina. It can be assumed that the interaction of the pol 5 mer with the a n t i g e n - a n t i b o d y system consists of a subtle temporary attachment of the pol,, mer to the anubod,, molect, le resuhing in a change of the shape and/or llexibility of the antibody molecule, fa~ ouring its cros.~-hnking c a p a c i b .
2. Introduction Precipitating anti-hapten antibodies combined with muluvalent h a p t e n - p r o t e m conjugates precipi.4bbreviations: Dnp, 2,4-diniuophen.~ l: PEG. pol,~ eth.~ lene gl~ col; PBS. phosphate-buffered saline, i.e. 1511 mM NaCI. 20 mM phosphate pH 7.4. O105
2478.,'82[0f1(,i-0000,'$2 75 ,g, Else'.ier B.)med]cal Press
tate under formation of a three-dimensional lattice in which antigen molecules are bridged by bix alent il]Olecules of antibod3. The formation of the antigen antibods p~ecipitate can be potentiated b s addiuon to t i n s s.,, s t e m o f certalrl m a c r o m o l e c u l a r
substances
such as dextran [1] or PEG [2]. This effect seemed to be full.,, understood in terms of po[.,, met incompatibilit3 or a steric exchision of the complexes from the domains of the pol) mers [3]. More recently. F,an~k xxith co,workers [4.5] found that addition of h',drophilic pol.~ mer~ can also promote precipitation in the s.~stem of a h a p t e n - p r o t e i l l conjugate with noniprecipitating antibody, winch otherwise forms no insoluble product, but onl.~ small soluble complexes with the antigen. This effect occurs at a relatively low concentration of the polymer I I - o ) a n d i s s t r o n ~ 5 dependent on the molecular v, eight of the added polymer. When the polymer is ~ashed out. the precipitate remains insoluble having properties of a true a n t i g e n - a n t i b o d y lattice [5]. These findings led us to suggest that a more complicated mechanistn is imolved than simpl.~ the solvent exclusion. The aim of this v, ork has been to tollow the kmeucs of formation of complexes betu, een a h a p t e n - p r o t e i n conjugate and pig non-precipitating antibody of the lgG class in the presence of PEG. We used a low polymer concentration, in ~xhich no x isible precipitate Is formed. The size of the complexes was estimated by lid~t scattering. A complementary analysis of the gro~ th of the antigen-antibod5 lattice x~as performed b v absorption spectroscop.~ ,_ffthe hapten chromophore. 49
3. Materials and methods 3.1. Materials Non-precipitati,lg anti-Dnp antibody was isolated from sera of immunized pigs at tire decline of tile antibody response [5]. Pig serum albutnm contanling 40 Dnp-groups pet molecule {Dnpa,,-serunl albumin ~ ~as prepared according to Eisen et al. [6]. Pig se~t,m albumin containing 5.5 Dnp-groups per molecule I DnPs.5-serum albttmh~ ) ~as prepared by a modified procedure, in ~ hich 10 times less dinitrobenzene sulfonate was added than ill the standard procedure [6]. Polyeth.~ lene ~ y c o l 0000 I P E G ) ~ as purchased from Do~ (Midland. MIL Its nrolecular weight x~as tn tile range 6 0 0 0 - 7 0 0 0 . Chemicals used for buffers ~,,ere of arval3 tical purit.,, grade.
3.2. Methods 3.2.1. Measurement of molecular weight of complexes b} licit scattering Stock solutions ofanti-Dnp antibody [0.875 mg/ml in PBSI, Dnp.,cserum aJbunlJll (0.103 nlg/nll m PBS), PEG (60 rag/nil, 40 nlg/ml, 20 nrg/ml in PBSI and PBS alone were spun for 30 mhl at 2 5 , 0 0 0 g on the ultracentrifuge MSE 65 to temo,,e dust particles. Sanlples for light scattering v, ere prepared by mLxing 4.0 ml of antibody solution with 2.0 ml of PEG solt.tton (or PBS as a control ), and finall} with 4.0 ml of Dnp.~,,-serum albunliJ1 sohttion. 1"he final concentration of mrtibod.~ ~ as 0.35 mg/ml, of Dnp4o-serum aJbumin 0.04 mg/ml, and the concentrations of PEG were 4.8 and 12 mgjml. The angular dependence of the intensity of scattered li~lt ,.,.'as measured at 25°( ' at several time intervals, starting fronl the point of mixing, with a PhotoGonio-Diffusonletre Sofica apparatus over the angular range 3 0 - 1 5 0 " ( ' . using a ,.ertically polarized primary beanl ~itll a wavelength [m ~acuo) 546 nnl. Tile experimental values interpolated for pat ticular times of react,on [7] ~ere plotted as Kc,:Reo against sin-~l®/2 h K being the optical constant, c the protein concentratmn, and R o tile Ra) le~gh ratio for the angle ofobser~ ation ®. Tile refractive index inclement used in the calculation of K was determined ~ittl a Brice-Phoenix differential refractometer, model BP-2OOO-V: its value ~as 0.189 g/nr1125°C. 546 nnl). 50
l-tie KciRe)=~,.~ w..due was obtained b} extrapolation to zero angle of obserxation [7] and its reciprocal ~as taken as the weight-a~erage molecular wei~lt. M~, of tile s)stenl. Suite M~ depends on tile concentration of the antibod~ and antigen, its ,,alue could not be obtained m tile rigorous wa.~, I.e. b) extrapolation t)f Kc,,'Rt:) to mfimte dilution. Tile error thus introduced is likel} not to be qualitative b ~ignlficant, because tile concentlation dependencies of Kc,,'R,,j m s.~stems of this type are usuall\ ~ery small. 3.2.2. Absorption spectroscop5 nleasutenlents .Mrti-Dnp antibody. PEG [or PBS ) and DnP5.5-serum albttllli~l ~ere mLxed dtrect[3 m tile test cuvette of the spectrophotometer from the stock soluttons. Tile final concenttation of tile an nbod,, ~ as 0.35 nlg/ml. that of DnP5.5-serum albt.mhl ~as 0.04 mg,,nl[ and that of PEG ~as 12 mg/nd, hnnlediatel.~ after mL\mg. tile absorphon spectra of the Dnp-moiet,, ~ere recorded in the range 3 5 0 - 3 0 0 nm, repeating the scans in 5 nlin intervals. One scan took 24 s. The wa~elength of tile absorption maximum ~as determnled from tile records graphically using tile filst deri~atixe plot. All spectroscopic measurements were done x~ith a \'arian Superscan 3 LiV-Vis specttophotometer.
4. Results and discussion Since the enhancing effect of PEG on tile formation of insoluble arrtigen-antibodb complexes is most pronounced at low antigen to antibody ratios [5], tile molar ratio antigen to antibod). 1:4 v, as chosen fur our expemnents, file system reached the level of tile molecular weight of complexes about 2 X 10 ° practicall,,' nnmedlately. A complex of this size can be r e d resented, e.g.. b', 12 anubod.,, molecules and 3 antigen molecules. Further increase of tile complex s~ze has been found to be much slov, er (Fig. 1 ). The slow rate may indicate that rearrangement of conlplexes is necessar~ to nlake ~acant antibod,, binding sites steritally a~ailable for further interactions. Tire restriction ot further increase in size is nlost obvious in tile control system ~ithout PEG. The size of tire complexes rises so slow[ 5 that it cannot reach the molecular ~ eight necessar~ for precipitation within a realistic time. The eftect ot increased concentration of PEG
/
2,,0
i
i
8
r.7 "
I
I
E
/
C
/ w
,/¸¸¸
¢ //
,n,)
/
/ /
/ / t/'-'-
/
J "[3
1,Z,,J
a ["
~ 2'
i
. '."J
-'7") -1-i,~ IT,ICI
Fig. 1. Increase of the v, eight-a~erage m o l e c u l a r v, mght, ~1,.,,. o t c o n | p l e ' , e s formed in the s~ _,tem c o n t a i n i n g the non-precipi t a t l n g a n t i - D n p a n t i b o d 3 . D n p ~ , . - ~ e r t n a b t m i n , and PEG. I = c o n t r o l ~ i t h o u t PEG. 2 = ~L4"7 PEG; 3 = It8"" PEG: 4 = 1.2 ~ PEG.
itself in a continuous increase of the rate of for mation of b i l l molecular-~ eight complexes. At the concentration of PEG 1.2";. the weight-a~etage molecular-v, ei~at of the complexes tends to approach an infinite vMue and ~isible precipitation is formed within 3 h. The shape uf the time dependence resemmanifests
0.6
,%
I
1
,,:¢
0.4
02
300
400
500 ,nm
Fig. 2. A b s o r p t i o n spectra of free (curve 1 ) and ant,bod.~b o u n d |.curve 2~ 6 - D n p - a m i n o c a p r o i c acid. The aamples were d i s s o h e d in PBS. annbod.~ ~ a,,_ a d d e d ,n e\ce~s. X I = 364.3 nm, a = 371.8 nm.
I
I
I
4O
8O
120
160 time, m,n
Fig. 3. Increase of the ~tutt of the Dnp-group ab,orptton maxHnunl I n the s5 stem c o n t a i m n g non-precipJtatin?, a n t i - D n p antibod.~. Dnp<~-sert||n a l b u m i n and PEG. : . c o n t r o l t t i t h o u t P E G . o . 1.2"; PEG.
bles the cur~e for a cross-hnkmg reaction approaching the gel point [8]. Tile absorption spectrum of tile Dnp-gt oup is shifted to longer wavelengths ~hen the group is inserted in the combining site of the anti-Dnp antibud.v I Fte. 2 L The magnitude of the shift is proportional to the ratio of bound and flee Dnp-groups in the system. The kinetics of formation of antigen-antibod,, bonds '. ia combining sites ~ ere ff)llowed b~ measuring the shifts of maxima of the Dnp-group absorption band dt, ting the interacmon of DnP5.5-serum albumin ~ith nonprecipitating anti-Dnp antibody without and in the presence of 1.2'. PEG IFig. 31. The kinetic curves indicate that the interaction of the haptenic groups with antibod 3 combining sites ts facditated in the presence of PEG. The kinetic data. obtained b 3 light scattering on the one hand and b~ absorption spectroscop.~ on the other, are not dlrectl 3 comparable due to the different functionalit 3 of the antigens emplo.~ ed. A Io~ content of Dnp-groups v. as necessary for obtaining sufficient resohttion m absorption spectra. In spite of this difference, the data obtained a l l o ~ u s to draw certain conclusions. The h.,, pothesis based on the soh'ent exclusion effect of pu b mers can be rejected. The 51
kinetics of complex
lormation,
as s h o w n b.~ l i g h t
scatter|rig, is entirely different from the kinetics of phase separation. A gradual increase m PEG concentration causes a gradual, and not jump-like, increase in the rate of tire molecular-x~eight growth o f tire complexes, ln te~ action of the polymer ~ ith a n t i g e n antibody complex should be considered as a subtle temporar5 attachment of the pol5 mer molecule to the antibod.,, molecule. The inabilit5 of non-precipitating antibodies to precipitate multivalent antigens may be understood as a tendenc) to prefe|ential formation of c5 clic structures with limited molecular wei~]t. Due to the interaction ~ ith h 5 drophihc macromulecules, the shape and,'or llexibilitx of ant> bod3 molecules ma.~ be changed in Stlch a ~Aa} that, during formation of the lattice, the ClOss-hnkllig ~eac|ion is favoured instead of the formation of c3 ties. The lesults of absorption measurements confirm that bonds bet~ een Dnp-group~ I epitopes I and combining sites (paratopes) are im olved in the reaction, althou~l
52
the existence of additional intermolecular bonds cannot be ruled out.
References [ I ] Hellsing, K. and Laurem, T. C. C19641 Acta Chem. Stand. 1~, 1 3 1 l j - ] 3114. [2] H,Jrrlll~t,an, J. C., Fenb)n, J. ~... II and Pert, J. H. ~1971 ~ lnm]unoc hemi~!.r.~ 8, 4 1 3 - 4 2 ] . [3] Hell~in~,K. I I 9 6 9 ) Biochen1. J. I 1 2 , 4 7 5 48]. [41 Fran.~k. F. i [ 9 7 3 } ill: Antibod.~ Structure and Molecular Immunolog.~. FE BS Symposia, Vol. 36 (Gerge[.~, ,I. and Medg~esi, G. ~ . e d , . j p p . 63 "5, Akad&nial Kl3.dt:~, Budape>t. [31 Fr,m~'k, F., O1;'o* ~k,i, Z. and Simek, L. 11979) Eur. J. Immunol. 9 , 6 9 6 - 7 0 1 . 161 Eisen. H. N., Belman, S. and Caz_,ten, M. E. i 1953'I J. Am. Chem. Soc. ~5. 4 3 8 3 - 4 5 8 3 . [ ~'1 Kratoch~ il. P., Munk. P. and Sedl:K~ek, B. i 196] I Coll. Czech. Chem. Commun. 20, 1274 - 1282. [81 Flor~., P. J. I19531 Principle,, of Pol) nler Chembtr.~, p. 383. Cornell Universu.', Pre~, Ithaca, Ne~ York.