Spectral Changes in a Cat?onic Dye Due to Interaction with

In the course of examining the use of carbocyanine dyes as agents for the detection of trace amounts of protein and other macromolecules, the spectral...
0 downloads 0 Views 1MB Size
1896

R. E. KAY,E. R. WALWICK, AND C. K. GIFFORD

Spectral Changes in a Cat?onic Dye Due to Interaction with Macromolecules. I.

Behavior of Dye Alone in Solution

and the Effect of Added Macromolecules’

by Robert E. Kay, E. Richard Walwick, and Cheryl K. Gifford Philco Research Laboratories, -Yewport Beach, California

(Receined February 10, 196Q)

I n the course of examining the use of carbocyanine dyes as agents for the detection of trace amounts of protein and other macromolecules, the spectral changes resulting from the interaction of various iiiacromolecules with the dye 4,5,4’,5’-dibenzo-3,3’-diethyl-Y-methylthiacarbocyanine bromide were observed, and the effects of environmental factors on the absorption spectrum of the free dye were determined. The aqueous dye solution was stable over the pH range 3.8-9.6 and unaffected by storage at temperatures below 60”) but it was unstable when exposed to light. The effects of pH, solvent, dye concentration, temperature, and inorganic ions on the wave length of the dye absorption maximum were ascertained. The pH had no effect on the position of the absorption maximum, but other variables such as the composition of the solvent system or changes in the dye conceiitration produced changes in the wave length of the maximum. AIaxinia were observed a t 575, 555, 535, 510, 150, or 650 nip (J-band) and these maxima are believed to represent increasing degrees of aggregation of the dye in the order: 575. 535, 510, 450, and 650 nip. The %555nip band appears to be associated with the J-band maximum and probably does not represent the first increase in aggregation from the monomer. The interactions of the dye with inorganic salts, polypeptides, simple proteins, conjugated proteins, synthetic polypeptides, nucleic acids, carbohydrates, amino acids, pyrimidine and purine bases. nucleosides, and nucleotides were all investigated. I n trace amounts (less than 0.002y0) only proteins. synthetic polypeptides, nucleic acids, and substituted polysaccharides caused changes in the absorption spectrum of the dye. Xono-, di-, and trisaccharides, purine and pyrimidine bases, amino acids, and nucleosides had no effect. Polypeptides and nucleotides were usually effective only a t higher concentrations, and the action of the inorganic salts was found to depend upon the nature of the anion. Divalent anions were very effective, and small amounts induced the formation of J-bands. On the other hand, monovalent anions u ere niuch less effective, and relatively large amounts were required to induce the formation of a J-band.

aggregation of the dye molecules when adsorbed on the

Introduction

recent review by visibility in tissue preparations. Schubert and Hai~lerl1~a~~2 has extensively covered t,he chelnical chromasia.

and histochemical use Of metaI n general, these color changes are due to

T h e Journal of Physical Chemistry

(1) This research was supported by the National Aeronautics and Space &4drninistration,under Contract No. NASr-84. (2) M. Schubert and D. Hamerman, J . Histochem. Cytochem., 4 , 159 (1956).

BEHAVIOR OF CATIOXIC DYEIN SOLUTION

spectrum of the simple dye molccules. Likewise, thle interaction of acridine dyes with polyanions in solution produces spectral changes which have been explained as a function of the aggregation of the dye bound to the p~lyanion.~-’I n this study we have been especially concerned with the interaction of cyanine dyes with biological material for the purpose of using these dyes to detect trace amounts of protein and other macromolecules. The cyanine dyes are important sensitizing dyes which have found wide use in the photographic process18and consequently their behavior in solution and as films has been studied extensively. When these dyes are prepared in aqueous solutions, increasing their concentration produces an aggregation of the dye molecules, and most cyanine dyes exhibit new bands a t shorter wave lengths itban the monomolecular spectrum. However, some of these dyes which have a planar structure, on further increase in concentration of the aqueous solution, develop an intense, sharp absorption maximum a t longer wavc length than the niolecular absorption, This effect has been observed by a number of investi-. gatorsq-16 and the new intense absorption band is often referred to as a J-band. The J-bands appear to result from interaction of large numbers of niolecules in an orderly array and Jelley16concludes that the dye molecules are arranged parallel to each other. Sheppard1’ suggests that water molecules linking the polar groups of the dyes a t the edges of pairs of molecules assist in formation of the aggregate state. Of especitll significance to the present study are the observations by Sheppardg and IYatanson” that traces of protein favor aggregation of certain carbocyanine dyes to give a J-band. The principal objectives of this study were to: (a) determine the effects of some environmental factors on the absorption spectrum of the dye 4,5,4’,5’-dibenzo3,3’-diethyl-S-niethylthiacarbocyanine bromide and (b) observe the spectral changes resulting from interaction of various macromolecules with the dye and relate these changes to those which occur with the dye alone in solution.

Methods All test materials were obtained conimercially. Two samples of the dye 4,5,4‘,5 ’-dibenzo-3,3‘-diethyl-% methylthiacarbocyanine bromide were gifts of F. W. Mueller, Ansco, and J. A. Leermakers, Eastman Kodak Co. Absorption spectra were determined from 400 to 700 mp with the use of either a Cary Model 14 recording spectrophotometer or a Beckman DK-2A recording spectrophotometer. Absorption cells having path lengths of 0.1-10 cm. were employed and the reference solution used is indicated in each experiment. Dye and

1897

test materials were prepared from weighed amounts of the materials, except as otherwise indicated.

Procedure and Results Selection of the Experimental Dye. The carbocyanine dyes listed in Table I were prepared as aqueous solutions and mixed with equal volumes of water, 0.02% deoxyribonucleic acid (DIYA), or 0.00270 gelatine solutions. The absorption spectra of the DIYA and gelatine solutions were determined and compared with that of the aqueous dye solution. The absorption spectra of dyes 3, 4, 6, 7, and 8 were altered by the presence of DKA. On the other hand, only dyes 6, 7, and 8, all of which are substituted a t the 5,s’ and/or the 4,4‘ position, were affected by the addition of gelatine, and all of these dyes formed J-bands a t about 650 mp. Dye number 8, 4,5,4’,5’-dibenzo-3,3’-diethyl-9-methylthiacarbocyanine bromide, formed the niost intense J-band with gelatine; and it also exhibited the greatest changes in the presence of DSA. Since this dye appeared to interact most favorably with both DKA and gelatine, it was selected as the test dye for use in subsequent experiments. Its structural formula is shown in Fig. 1.

Figure 1. Structural formula of 4,5,4’,5’-dibenzo3,3‘-diethy1-9-methylthiacarbocyaninebromide.

(3) L. Michaelis. J . Phys. Colloid Chem., 54, 1 (1950). (4) D. F. Bradley and G. Felsenfeld, Xature, 184, 1920 (1959). (5) D. F. Bradley and M. K. Wolf, Proc. Natl. Acad. Sci. U . S., 45, 944 ( I 959). (6) R. E. Boyie, S.S. Kelson, F. R. Dollish. and & J. I. Olsen, Arch. Biochem. Riophys., 96, 47 (1962) (7) V. W. Appel and G. Scheibe. 2. .~aturjorsch., 13, 359 (1958). (8) C. E. K. Mees, “ T h e Theory of the Photographic Process,’’ T h e Maemillan Co., New York, S. Y . , 1959, (9) S.E. Sheppard, Rev. Mod. Phys., 14, 303 (1942). (10) E. E. Jelley, .Vatwe, 138, 1009 (1939). (11) S.Nntanson, Acta Physicochim. U.R.S.S.. 21, 451 (1946). (12) H. 0 . Dickinson, Trans. Faraday Soc., 43, 486 (1947). (13) G. Scheibe, Kolloid-2.. 82, 1 (1938). (14) S.M.Solov’ev, Z h . F C . Khim. 19, 459 (1946). (15) E. Sheppard and A. L. Geddes, J . Am. Chem. Soc., 6 6 , 2003

s.

(1944).

(16) E. E. Jelley, Suture, 139, 631 (1937). (17) S. E. Sheppard and €1. 11. Brigham, J . Am,. Chem. Soc., 6 6 , 380 (1944).

Volume 68,.Vumber 7

Julv, 1964

1898

R. E. KAY,E. R. WALWICK, AND C. K. GIFFORD

Table I : Carbocyanine Dyes Tested for Their Interaction with Macromolecules" Absorption maxima of aqueoua dye solution plus------. Water, 0.01% 0.001% m DNA, m gelatin, m

w--

No.

Concentration of dye, M

Dye

1

3,3',9-Triethylselenacarbocyanine

2

3,3',9-TriethyIthiacarbocyanine

x 10-4 4 x 10-6

3

3,3',9-Triethyloxacarbocyanine

1

x

4

3,3'-Diethyl-9-methylthiacarbocyanineiodide

4

x

5

3,3'-Diethyl-9-methyloxacarbocyanine iodide

6

5,5'-Dichloro-S-ethy1-3,3'-dimethylthiacarbocyanine

1

555

555

545 505

545 505

545 505

10-4

475

468 490

475

10-6

503 540

510 545

503 540

x 10-4 4 x 10-6

467

467

467

510 550

555

510 550 650b

10-5

506

575

510 656

4 x 10-6

510

57.Y 535d

510 650

1

7 8

555

4 4,5,4',5'-Dibenzo-3,3 '-diethyl-9-methylthiacarbocyanine

x

' Dilute solutions of each dye were mixed with equal volumes of 0.020% DNA, 0.002% gelatin, or water and the absorption spectra of the solutions determined in a 1-cm. cell against a water blank. * Concentration of gelatin must exceed 0.001%. Native DNA. Denatured DKA.

The Efects of Environment on Lhe Absorption Spectrum of the Dye. In order to assess properly the influence of added materials on the absorption spectruni of the dye, it is necessary to know how the environment affects the dye in solution. There are two principal parameters by which the effects of environmental changes on the dye solution can be evaluated : absorbance, and the wave length at which maximum absorption takes place. We have used changes in absorbance to estimate the chemical stability of the dye and have ascertained the effects of light radiation, temperature, and changes in pH on this parameter. On the other hand, changes in the position of the absorption maximum have been used as criteria for alterations in the degree of aggregation of the dye. Effect of Light, p H , and Temperature on the Absorbance o f the Dyt! Solution. A 2 X 10-s M solution of the dye when exposed to a high light intensity, delivered from a 375-w. incandescent lamp after filtering through 11 cni. of water, quickly bleached from a deep fuchsia color to a light yellow color and the absorbance at 510 nip decreased from 1.060 to 0.060 in 75 sec. Exposure to normal room light also caused the dye to bleach, but this was noticeable only after exposures of about 30 min. The effect of temperature on the aqueous dye solution was ascertained by keeping solutions of the The Journal of Physical Chemistry

dye for 48 hr. a t temperatures of 4-100". A decrease in absorbance during the first 24 hr. was observed at all temperatures tested. However, the changes were not large unless the temperature was greater than 40°, and it appears that the initial decreases at lower temperatures may be due to adsorption of the dye on the containers since small amounts of adsorbed dye were found on the glass at the end of the experiment and storage a t 22' for an additional 37 days did not cause a further decline in absorbance. The absorbance of the dye solution was not appreciably altered by changes in pH over the range 3.8-9.6, but at a pH of less than 3.8 or more than 9.6, absorbance decreased rapidly (Fig. 2 ) . Because of the unfavorable influence of low or high pH and light on the free dye in solution, subsequent experiments were carried out, where possible, in pH 6.9-7.0, 0.017 M cacodylic acid buffer, or pH 8.8, 0.001 M tris buffer, and in containers wrapped in aluminum foil. E$ect of p H , Solvent, Dye Concentration, Temperature, and Inorganic Ions on the Wave Length of the Dye Absorption Maximum. Changes in pH over the range 2.3-12 had no effect on the wave length of the absorption maximum of an aqueous 2 x 10-5 Ad dye solution. The absorption spectra of the dye dissolved in water and ethanol are shown in Fig. 3. The two spectra are

1899

BEHAVIOR OF CATIONIC DYEIN SOLUTION

2.0-

rrrml,

A

I.9-

,

B-

IF-

16-

1.5-

,.'I-

1.2-

PH.

w

Figure 2. Effect of pH 011 the absorbance of the aqueous dye solution; two aliquots of a 2 X 10-5 A4 dye solution were used. The pH of one sample was decreased by the addition of 1 AT HC1 and the pH of the second aliquot was increased by the addition of 1 Ai NaOH. Changes in volume due to the additions of HC1 or NaOH were negligible. At each p H indicated in the figure, the absorption spectrum of the dye solution was determined and the absorbance at 510 mp mcertained. Measurements were made in a 1.0-cm. cell against a water blank.

quite different. The absorption maximum is at 575 mp in ethanol and 510 mp in water, and the intensity of the absorption peak is 570/, greater in the ethanol solu-. tion than in the water solution. When the dye is; dissolved in a series of solutions in which the ethanol content is continually diminished, the absorption spectrum does not undergo a change until the ethanol content is 50% or less (Table 11). However, as the ethanol content is brought below this level the wave length of the ma,ximum shifts toward the blue end of the spectrum and the absorbance at, the maximum diminishes. At an ethanol content of 30%, two distinct maxima appear, one a t 572 mp and the other a t 533 mp (Fig. 3). As the ethanol content is described further, a maximum also appears at about 510-515 mp. Thus, there appear in succession, one replacing the next, a series of three maxima: one a t about 570 mp, one a t 525-533 mp, and one a t 510-515 my. In aqueous solution, the position of the dye absorption peak varied with the concentration of the dye. The wave length of the absorption maximum continually increased as the concentration of the dye was decreased over the range 1.2 X to 5 X lo-' M and in very dilute solutions two absorption maxima were evident; one a t about 530 mp and the other at 570 mp (Table ITI). With decreasing dye concentration there was initially a very gradual Elhift in the posi-

1.1

-

2

..j

1.0

-

09-

08-

0.1

-

0.6

-

0.3

-

Figure 3. Effect of solvent composition on the absorption spectrum of the dye; in each case, the dye concentration was 2 X 10-6 M and the measurements were made using a 1.0-em. cell and a water blank.

tion of the absorption maximum from 500 to 510 mp, but a t a dye concentration of 5 X M a more abrupt shift to the 530 mp region of the spectrum occurred, and M a second peak was a t a concentration of 1 X evident a t 570 mp. Alterations in temperature likewise produced changes in the position of the absorption maximum. With an increase in temperature there was a gradual shift in the position of the absorption maximum from 510 m p at 11' to 515 mp a t 43'. At 51' two peaks appeared, one a t 520 mp and the other at 530 mp. With an increase in termperature to 6 9 O , the 520 mp peak disappeared, the 530 mp maximum Volume 68, Number 7 July, 1964

1900

R. E. KAY,E. R. WALWICK, A N D C. K. GIFFORD

Table 11: Effect of Solvent Composition on the Absorption Maximum of the Dye in Ethanol-Water Solution" % ethanol

Wave length of absorption maxima,

100 90 80 70 60 50 40 30 20 10 0

575 575 575 575 575 573 572 572,533 570,530 525,515 510

mp

Table IV: Effect of Teniperature on the Position of the Absorption Maximum in an Aqueous Dye Solution"

Absorbance a t maximum

Temp.. ' C .

Absorption maxima, mp

Absorbance a t absorbance maximum

1.930 1.970 1.970 1.970 1.970 1,906 1.760 1 . 2 7 0 , l .340 0 . 7 2 0 , l .490 1.210,1.215 1.230

11 19 23 29 34 43 51 69 79 96

50 1 505 506 510 513 515 520,530 575,530 575,533 575, 533

1,224 1,178 1 117 0.980 0.930 0,900 0,859,O.849 0.432,0,900 0.530,O.850 0.672,0.701

2 X 10-5 M dye solutions having the indicated ethanol content were prepared by diluting a concentrated ethanol solution of the dye with water. The absorption spectrum of each solution was determined using 1.0-cm. cells and water as a blank. The positions and intensities of the absorption maxima were obtained by inspection of the absorption spectra.

became more intense, and a new peak was found a t 575 mp. Increasing the temperature still higher caused an increase in absorbance a t 579 mp and a decrease in the intensity of the 533 mp maximum (Table IV).

The absorption spectra of 2 X 10-6 M solutions of the dye in pH 7.0, 0.017 M cacodylic acid buffer a-ere determined at the indicated temperatures and the wave lengths of the absorption maxima were obtained by inspection. Measurements were made against water using cells with a path length of 1.0 cm.

'"I

1.1

~~

Table 111: Effect of Dye Concentration on the Position of the Absorption Maximum in an Aqueous Dye Solution" Dye concentration, iM

1.2 x 8 X 6 X 4 x 2 x 1x 5 x 1x 5 x

10-4 10-6 10-5 10-6 10-6 10-8 10-5 10-7

Absorption maxima, mp

500 502 505 506 509 510 527 528 and 570 530 and 570

a hqueous solutions having the indicated dye concentrations M solution of the dye. were prepared by diluting a 1.2 X The absorption spectra of these solutions were determined a t 25" and the wave length of the absorption maximum obtained by inspection of the absorption curves. Measurements were made against water in cells having path lengths of 0.1-10 em. All solutions were buffered a t pH 7.0 with 0.017 M cacodylic acid buffer.

It is well known that the addition of electrolytes to cyanine dye solutions will cause the dye molecules to aggregate.8~9~12~16 This phenomenon is illustrated in Fig. 4 which shows the absorption spectrum of a dye solution to which hIgClz has been added. Compared to The Journal of Physical Chemistry

O ' -

,J,

I

I

I

I

I

the aqueous dye solution, the absorption spectrum has been greatly altered. A very intense, narrow absorption maximum (J-band) appears at 650 mp, a second

1901

BEHAVIOR OF CATIONIC DYEIN SOLUTION

maximum occurs a t 555 mp, and prominent shoulders are seen at 512 and 450 mp. Decreases in dye concentration, incresses in temperature, and increases in the ethanol content of the solvent systems are all factors which tend to increase the dispersion of dye molecules, whereas the addition of electrolyte usually promotes aggregation. In each case, when the factor investigated was altered so as to increase dispersion, the absorption maximum was shifted to longer wave lengths. As the dispersive effect was strengthened, definite absorption maxima appeared a t the following wave lengths and in the sequence indicated: 510, 533, and 575 mp. On the other hand, the effect of adding electrolyte was quite different. Sew absorption peaks were found at both longer and shorter wave lengths than that exhibited by the aqueous dye solubion. These new peaks are a t 650, 555, and 450 mp. From this information, it appears that at least six distinct dye species, representing different states of dye aggregation, can be formed. The aggregate represented by the 650 mp maximum is evidently relatively large since it is sedimented by centrifugation, whereas the species represented by the other absorption maxima are not. This information was obtained by centrifuging different dye solutions, some of which showed spectrophotometric evidence for the presence of the species represented by the 650 mp maximum and others of which did not. The dye was sedimented only when the species represented by the 650 mp maximum u as spectrophotometrically evident. Interaction of the Dye with Inorganic Salts. In aqueous solutions a t pH 7.0, the dye is cationic in nature and is present in a somewhat aggregated state as evidenced by the information obtained in the previous experiments. Therefore, it is reasonable to expect that the formation of higher dye aggregates will be greatly affected by the concentration of negative ions. The presence of negative ions should favor aggregation; and thus, a change in the position of the absorption maximum of the dye to 650 and/or 450 mp regions of the spectrum. The amounts of electrolyte necessary to induce aggregation mere determined by adding solutions containing various quantities of electrolyte to a buffered solution of the d;ye. The absorption spectrum of the solution was then deteimined against a water blank, and when changes in the spectrum indicating aggregation were observed, the electrolyte concentration was noted. Under these circumstances, increases in aggregation were always indicated by the formation of SL J-band in the 650 mp region of the spectrum when an appropriate amount of electrolyte was present. There

was also often an additional absorption maximum in the region 555-585 mp. Figure 4 shows an absorption spectrum obtained in the presence of sufficient electrolyte to cause aggregation. Table V gives the approximate minimum electrolyte concentration which was effective in causing the dye to form a J-band. It will be seen that small amounts of the electrolytes are

Table V : Minimum Electrolyte Concentration Effective in Causing Aggregation of the Dye” Electrolyte

Conrentration, m M

Concentration, mM

0.06 0.08 0.17 0.17 0.17 0.17 0.17 0.17 0.17 0.40 0.40 0.40 1.70

1.70 1.70 4.0 4.0 4.0 >8.0 >17.0 >17,0 >17.0 >17,0 >17.0

a Aggregation was detected by the formation of a J-band. The dye concentration was 4 X 10-6 M and the solutions were buffered at pH 7.0 in 0.017 A4 cacodylic acid buffer.

required and, further, that their potency is dependent upon the nature of the negative ion. In fact the effects of the electrolytes seem to be quite similar to their precipitating action on colloids.’* It was also found that in the presence of 10% ethanol, the salt concentration required to induce aggregation, as evidenced by J-band formation, was increased about tenfold. This may be due to the greater solubility of the dye in alcohol which gives rise to increased dispersion and decreased interaction of the dye molecules. Interaction of Dye with Proteins and Protein Derivatives. It has already been shown that the addition of some proteins to the dye solution gives rise to the formation of new maxima in the absorption spectrum of the dye. The manner in which a number of different proteins interact with the dye is therefore of interest. This, in turn, makes the reaction of the dye with the constituents of proteins, e.g., peptides and amino acids, important for understanding some of the factors which influence the reaction of the dye with proteins. I n order to obtain data which could be (18) F. H. Getman and F. Daniels, “Outlines of Theoretical Chemlstry,” John W5ley and Sons, Inc., Kew York, N. Y . , 1937, p. 216.

Volume 68,Number 7

J u l y , 1064

1902

useful in analyzing the reaction of the dye with proteins, the influence of a large number of proteins, peptides, and amino acids on the absorption spectrum of the dye was ascertained. Thirty-two proteins were investigated. The influence of the proteins on the absorption spectrum of the dye is summarized in Table VI. The results were quite dependent upon the particular protein used, and only in the case of carboxpeptidase and native ribonuclease did the addition of protein fail to cause the formation of new absorption maxima. The changes occurred with trace amounts of protein and the new maxima were always found at about 480, 570, 600, or 650 mp. It appears that for a great many of the protein-dye complexes the exact wave lengths of the induced maxima may be characteristic of the protein. The effects of 37 different peptides on the absorption spectrum of the dye were determined. Only glycylDL-methionine produced changes in the dye spectrum a t a'concentration of less than 0.02%. However, a t higher concentrations (0.1%) most of the peptides caused alterations in the absorption spectrum. The wave lengths of new maxima which vere observed with a peptide concentration of 0.1% are recorded in Table VII. The alterations, as in the case of the proteins, involved combinations of increases in absorbance at wave lengths of about 480, 570, 600, and 650 mp. However, in the vast majority of the cases, the absorbances of the induced maxima were not very great and the main effect was the formation of a new maximum a t about 560 to 570 mp, which indicates that the effect was one of deaggregation rather than aggregation of the dye. The occurrence of hydrophobic or hydrophilic side chains and the ionizability of the hydrophilic side chains in the peptides did not appear to have any consistent influence upon the nature of the changes in the absorption spectrum. However, the sequence of the amino acids in the peptide was of considerable importance; e.g., glycylleucine caused the formation of an intense absorption peak a t 605 mp, whereas leucylglycine mas without effect. Likewise, glycylalanine induced a new absorption peak at 540 mp, but alanylglycine did not influence the absorption spectrum. The effect of 0.1% solutions of 40 different amino acids or amino acid derivatives on the absorption spectrum of the dye was investigated in unbuffered solutions, 0.017 M cacodylic acid buffered solutions, pH 7.0, and 0.001 M tris buffered solutions, pH 8.8. Xew maxima (645 and 540 mp) mere observed only in the case of L-mesolanthionine. Interaction of the D y e with Synthetic Polypeptides. The study of the reaction of synthetic polypeptides The Journal of Physical Chemistry

R. E. KAY,E. R. WALWICK, AND C. K. GIFFORD

Table VI : Iiew Absorption Maxima Formed by Dye in the Presence of Trace Amounts of Protein"

Protein

Commercial sources*

Vave lengths of new maxima,

w Ribonuclease, oxidized, bovine pancreas Oxytocin, pituitary (heterogeneous sources) Urease, jack bean meal a-Globulin, bovine fra.ction IV 0-Globulin, bovine fraction I11 Acetylcholine esterase, bovine erythrocytes Carbonic anhydrase Casein Plasminogen, bovine fraction 111 Gelatin Pepsin, swine gastric mucosa Pepsinogen, swine gastric mucosa a-Lipoprotein, *bovine 0-Lipoprotein, bovine fraction 111-0 0-Lactoglobulin, bovine Myoglobin, horse heart Glycoprotein, bovine fraction VI Deoxyribonuclease, bovine pancreas Cellulase, fungal Pectinase, fungal Follicle stimulating hormone, porcine 0-Glucuronidase, bovine liver Glutenin Insulin, bovine pancreas Trypsin, bovine pancreas Albumin, bovine Albumin, human fraction V a-Chymotrypsin, bovine pancreas Hemoglobin, human Somatotrophic hormone, porcine Ribonuclease, native, bovine pancreas Carboxpeptidase, pancreas

hlann

650

Calbio Calbio NBCo. KBCo.

620 526, 570 650,570 650,570

hlann Calbio Calbio XBCo. Knox Worth Worth Calbio Calbio NBCo. NBCo. NBCo. Calbio Calbio Calbio

637,570 642, 580 646, 560 642,570 650,580 615,559 620,558,480 637, 570,480 630, 560, 480 646,560,480 640,600,570 650,600,570 650, 600, 570,450 600, 570 600, 570

Calbio Mann SBCo. Calbio Worth SBCo. Calbio Worth NBCo. Calbio

600, 570 600,556 600,570 600,555,480 570 480 480 480 480 480

Mann Calbio

None Kone

a The absorption spectrum of a 2 X 10-6 A4 solution of the dye in the presence of 0.00270 protein was determined using a 2 X 10-5 d l solution of the dye as a blank. All solutions were buffered in 0.001 M tris buffer, pH 8.8, and measurements were made in l.O-cm. cells. Abbreviations used in the table are as follows: NBCo., Sutritional Biochemicals Corp.; Calbio, California Corp. for Biochemical Research; Mann, Mann Research Laboratories, Inc. ; Worth, Worthington Biochemical Corp.

with the dye presents an opportunity to examine, in a more exact manner, the influence of various types of repeating side chains on the changes in the absorption spectrum of the dye. By choosing appropriate polypeptides where all the side chains in any one polypeptide are the same, it is possible to ascertain the

1903

BEHAVIOR OF CATIONIC DYEIX SOLTJTIOR

~

Table VI1 : New Absorption Maxima Formed by Dye in the Presence of Peptides=

Peptide

o-Leucy lgly cylgly cine DL- Alanylglyc ylgly cine' DL-Alanyl-DL-phenylalanine" DL-Alanyl-D 1,-a1anin.e DL-Alanyl-DL-asparagined Glycyl-L-asparagine' Gly cyh-asparagine' Glycylglycylglycine Glycyl-DL-methionine' Glycyl-L-leucine Glycylglycine DL-Alanyl-DL-leUcine DL- Alanyl-DL-norleucine DL-Leucyl-DL-phenylalanine Histidylhistidine Glycyl-DL-leucine Glycyl-DL-pl.ienylala.nine Glycyl-DL-norleucine Glycyl-DL-valine Glycylglycylglycylglycine Glycyl- tyrosine DL-Leucylgl y cylgly cine DL-Leucyl-or.-norvaline DL- Alany 1-DL-valine Glycyl-L-tryptophane Glycyl-DL-norvaline Glycylglycine.HC1 D-Leucylglycine DL-Alanyl-DL-norvaline Glycyl-DL-ala,nined Glycyl-DL-serine DL-Alanylglycine DL-Leucylglycine o-Leucyl-L-t yrosine L-Leucyl-L-tyrosine DL-Benzoylahine Benzoylgly cine

Wave lengths of new maxima, mp

650,600,565 650,562 648,585 650, ,560, 482 650,537 650,560 650,554 620,560 620,565,445 612,570 610, 565 610, 560 610,560 610,560 608,565 605,560 605,565 605,565 605,568 605,570 605,560 605,560 605,560 605,560 604,560 602,565 600,560 600,565 600,565 570,540 LTone Xone None None None Yone Xone

a The absorption spectrum of a 2 X 10-6 or 4 X 10-5 M solution of the dye in the presence of 0.1% of each peptide was determined in a 1.0-em. (-ell using the aqueous dye solution or distilled water as a blank. Measurements were made in unbuffered solutions and in solutions buffered with 0.001 M , pH 8.8 tris buffer, or 0.017 M , pH 7.0 cacodylic acid buffer. All of the peptides were obtained from Xutritional Biochemical Corp. The indicated response was obtained only in the unbuffered and cacodylic arid buffered solution. The peptide solution was saturated and probably contained considerably less than 0.1 % peptide. The indicated response was obtained only in the cacodylic arid buffered solution.

relationship of a particular repeating functional group to the changes in the absorption spectrum of the dye. The following long-chain polypeptides dissolved in 0.017 M cacodylic acid buffer, pH 7.0, were examined:

WAVELENGTH

(>!iLLlNIcRoir$

Figure 5. Effect of poly-L-aspartic acid on the absorption spectrum of the dye; the dye was 2 X M and the polyL-aspartic acid 0.00027,. The mixture was buffered a t pH 7.0 in 0.017 iM cacodylic acid buffer, and measurements were made in a 1.0-cm. cell against a water blank. The aqueous dye spectrum is shown for comparison.

(a) poly-L-lysine hydrobromide (mol. wt. 100,000200,000), having ionizable cationic H2S (CH2)d- side groups, (b) poly-L-aspartic acid (mol. nit. 500010,000) with ionizable anionic HOCOCH2- side groups, (e) poly-L-hydroxyproline (mol. wt. 50,000100,000) with hydrophilic but iioiiioiiizable HOCH(CHJ- side chains, and (d) poly-L-proline (mol. wt. 25,000-50,000) having hydrophobic H2C(CH2)%-side groups. The presence of poly-L-hydroxyproliiie, poly-L-proline, or poly-~-lysiiiehydrobromide a t a coiiceiitration of 0.02% did not alter the absorption spectrum of the aqueous dye solution, whereas a 0.000270 solution of poly- aspartic acid produced a very intense absorption peak a t about 530 mp (Fig. 5 ) . Interaction of the D y e with Deoxyribonucleic Acid ( D N A ),Ribonucleic A c i d ( R N A ) ,and Their Derivatives. Since the regularly spaced anionic sidc chains of polyL-aspartic acid are extremely effective in altering the aggregation of the dye, other substaiices having reguVolume 68, Number 7

J u l y , 1964

1904

R. E. KAY,E. R.WALWICK, AXD C. K. GIFFORD

1.70

1.60 1.50 1.40 1.30

1.20 1.10 1 .oo

w

3

0.90

0

g

0.80

4

0.70

0.60

0.50 0 .LO 0.30

WAVELENGTH (MILLIMICRONS)

Figure 6. Absorption spectra of DIiA- and RXB-dye complexes; solut,ions of DNA and RNA were prepared in 0.017 ilf cacodylic acid buffer, pH 7.0, containing 0.5 mil4 MgC12. The dye is 4 X 10-6 ‘1.I and the D N 4 and RNA concentrations are about 0.007yc. Measurements were made against water in a 1.0-cm. cell.

RKA was dissolved in buffer in the same manner as the DXA, but the preparation was not centrifuged. Each solution of DNA and RYA was mixed with an equal amount of 8 X 10-5 ‘14 aqueous dye solution and the absorption spectra were determined using water as a reference. Both D S A samples induced new intense absorption peaks at 575 mp (Fig. 6). The calf thymus DSA-dye mixture formed a suspension of blue fiberlike particles which precipitated after standing for several hours. On the other hand, the salmon sperm DSA-dye mixture formed a clear blue solution and no precipitate was evident. In contrast to the 575 mp peaks induced by DSA solutions, RSA caused the formation of a new intense absorption peak at 535 mp (Fig. 6). The exact concentrations of the D S X and R S A solutions were not known, since the degree of hydration and the losses due to centrifugation were n6t ascertained. However, based on the weight of material used, a 0.0002% solution of DKX or R X h produced new absorption peaks at 575 or 535 mp. Because of the marked interaction of DSA and R S A with the dye, it was of interest to investigate the interaction of the dye with their constituent parts, the pyrimidine and purine bases, nucleosides, and nucleotides. Each substance was tested initially at a concentration of 0.02% in 0.017 111, pH 7.0 cacodylic acid buffer. Only the nucleotides altered the absorption spectrum of the dye. In every case an intense absorption peak was formed a t about 650 mp, and additional peaks were present in the 570 or 540 mp regions of the spectrum. Table VI11 lists the nucleotides tested, and indicates the wave lengths of new

Table VI11 : New Absorption Maxima Formed by Dye in the Presence of Nucleotides”

larly spaced ionizable anionic groups may also be potent agents. The nucleic acids possess these attributes and they are also key building blocks in biological systems. Therefore, the interactions of the dye with salmon sperm DYA (California Corp. for Biochemical Research), calf thymus DXA (Worthington Biochemical Corp.), and yeast RXA (California Corp. for Biochemical Research) were investigated. DNA samples were prepared by dissolving, with gentle agitation at 4”, weighed amounts of D S A in 0.05 M cacodylic acid buffer, pH 7.0, containing 1.0 m X 31gClz. The DNA solution was then centrjfuged at 19,000 times gravity and 4’ for 60 min. and the supernatant solution decanted from the sediment. dliquots of the supernate were diluted with the buffer solution to give a series of solutions having different D S A concentrations. A weighed amount of yeast The Journal of Phusicul Chemistry

Nucleotide

Wave lengths of new absorption maxima

Minimum effective nucleotide concentration, molesil.

3 ’( 2 ’)-Adenylic acid 3’(2‘)-Guanylic acid 3’(2’)-Cytidylic acid 3 ’(2’)-Cridylic acid 5’-Deoxyadenylic acid 5‘-Deoxpguanylic acid 5’-Deoxycytidylic acid 5’-Thymidylic acid Cytidine 0.5H2S04

660,585,535 648,540 655,530 655,600,530 655,535 660,575,530 655] 580,530 650,580,530 652,575

1 . 3 x 10-5 1 z x 10-5 6 X 6 X 5 6 X 10-5 6 X 1 . 5 x 10-4 1 2 x 10-4 3 4 x 10-4

a The absorpt,ion spect,rum of a 4 X 10-8 .If solution of the dye in the presence of various amounts of nucleotides was determined using water as a blank. All solutions were buffered a t pH 7.0 in 0.017 M cacodylic acid buffer and measurement’s were made in 1.0-em. cells.

1905

BEHAVIOR OF CATIOKIC DYEIN SOLETIOX

maxima and the minimum nucleotide concentrations which caused alterations in the absorption spectrum of the dye. The minimum effective concentration for the nucleotides varied considerably. However, in each case, the 3’(2’)-ribose compounds were considerably more effective than their corresponding 5’-deoxyribose nucleotides, and within groups having the phosphoric acid a t the same position, the purine containing nucleotides were more effective than the pyrimidine containing nucleotides. Interaction of the Dye with Carbohydrates. The results of the studies on proteins and nucleic acids strongly indicate that anionic sites and a highly organized structure are important factors which influence the ability of a substance to cause s, change in the aggregation of the dfe. The carbohydrates include a large number of substances which vary greatly in regard to complexity and the occurrence of anionic sites. Appel and Scheibe7 have observed marked changes in the absorption spectrum of N,N’-djethylpseudocyanine iodide in the presence of small amounts of heparin, chondroitin sulfate, and hyaluronic acid. All of these substances are large complex polysaccharides which ]possess numerous anionic sites. In view of this demonstrated effect of polysaccharides on the absorption spectrum of a cyanine dye, it was of interest to examine the interaction of a number of carbohydrates with the dye used in the present experiments. Forty-two carbohydrates were tested in unbuffered solutions and in solutions buffered a t pH 7.0 with 0.017 M cacodylic acid. Kone of the mono-, di-, or trisaccharides altered the absorption spectrum of the dye in any manner, but some of the polysaccharides caused marked changes. Long-chain polysaccharides, such as inulin and dextrin, mhich do not possess anionic sites had no effect. On the other hand, heparin, polygalacturonic acid, agar, chondroitin sulfate, alginic acid, hyaluronic acid, and carboxymethyl cellulose ether, all1 of which have large numbers of anionic groups, caused the formation of new absorption maxima. I n most cases, these new maxima appeared in the presence of 0.0002-0.00006% of the substance tested and each macromolecule affected the dye spectrum in a somewhat different way. As an example of the changes observed, Fig. 7 shows the effect of a trace amount of alginic acid on the absorption spectrum of the dye.

Discussion It has been shown that changes of the temperature and solvent composition will cause the cationic dye 4,5,4’,5’-dibenzo-3,3’-diet hyl-9-met hylt hiacarbocyanine bromide to form a number of aggregated species

I 00.50

-

-

8

2

P 4

0.30-

n.ac-

0.10

-

0-

400

425

450

475

500

WAVELENGTH

550

600

700

(M L 1 i 1” ic rons)

Figure 7. Absorption spectrum of alginic acid-dye complex; the solution was buffered in 0.017 A4 cacodylic acid buffer, pH 7.0, the dye concentration was 1 X 10-5 M , and the alginic acid concentration 0 . 0 0 0 2 ~ . Measurements were made against water in a 1.0-cm. cell. The absorption spectrum of a 1M dye solution in buffer is shown for comparison. 1X

in solution. This phenomenon is not unique, since many other dyes are known to evhibit this property. However, the fact that a t least six different maxima can be demonstrated is noteworthy. Because of the response of this dye to factors which are known to influence aggregation, these maxima may be arranged in a manner analogous to that found for other cationic dyes,g in order of increasing degree of dye aggregation as follows: 570, 535, 510, 450, and 650 mp. In addition, a maximum, which appears under the same conditions as the J-band, is often found a t 555 mp. Since the 570 mp peak is found in organic solvents, and a t low dye concentrations and high temperatures in aqueous solutions, it apparently represents the lowest state of aggregation (monomer), whereas the aggregate represented by the 650 mp maximum is probably large since it can be sedimented by centrifugation. The band at 535 mp is displaced by about 1300 cni.-‘ from the monomeric band a t 575 mp. This is about the same value of displacement of the dimeric band from the monomer which is seen in a number of thjacarbocyanine dyes.7-9 The 535 mp band, therefore, seems to be the dimeric band. Thus, the maximum a t 555 mp probably does not represent the first increase in aggregation from the monomer, but is rather in some way associated with the J-band. On the other hand, the maxima a t 510 and 450 mp presumably are characVolume 68, .Vumber 7

J u l y , 1964

1906

teristic of species which have aggregation states between the dimer and the J-band aggregate. Over the concentration range which was used in this study, the dye exists in aqueous solution as the species having an absorption maximum at 510 mp (i.e., in a state of intermediate aggregation). Thus, it is possible for the dye normally present in an aqueous solution to either increase or decrease its state of aggregation, and to do so in several discrete steps. Therefore, reactions of the dye with macromolecules, which possess a large number of anionic sites, in aqueous solution can result in the formation of new absorption maxima due to: (a) further polymerization of the dye, (b) depolymerization of the dye, or (c) a combination of (a) and (b). However, the mere occurrence of a large number of anionic groups does not, in itself, determine the nature of the response which is obtained when the dye comes in contact with an anionic substance. l l a n y factors appear to be involved and these can best be considered in terms of the effects of various substances on the absorption spectrum of the dye. Both peptides and amino acids contain ionizable amino and carboxylic groups; however, at pH 8.8, peptides will probably carry a negative charge since the pK’s of their amino groups are near 8, whereas amino acids will be neutral since their pK’s are near 10. With the exception of glycyl-DL-serine, the peptides which failed to cause changes in the absorption spectrum of the dye show one of two characteristics. They either contain a large aromatic structure or have glycine as the donor of the carboxylic end group. It is possible that the aromatic group hinders the close approach of the dye molecule to the peptide; and thus, the anionic site on the peptide is unable to influence the aggregation state of the dye. In regard to the apparently unique effect of the glycine entity in dipeptides, it is noteworthy that there is no aliphatic side chain on the a-carbon of the peptide only when glycine acts as the donor of the carboxylic end group. Perhaps an aliphatic side chain is necessary to orient the dye molecule properly with regard to the peptide cationic site. The proteins are far more effective, on a weight basis, than the peptides as agents which influence the aggregation state of the dye. Compared with peptides, the terminal carboxylic groups of proteins are separated farther from the end amino groups. However, since there are so few terminal groups per mass of protein, it seems unlikely that this increased separation can be responsible for the greatly increased effectiveness of proteins to provide focal points for dye aggregation. Rather, it is generally recognized2 that the effective sites are the anionic side chains which are available along the backbone of the protein. The great sensiThe Journal of Physical Chemistrg

R. E. KAY,E. R. WALWICK, ASD C. I