Interactions between PAMAM dendrimers and bovine serum albumin

Biochimica et Biophysica Acta 1648 (2003) 115 – 126 www.bba-direct.com

Interactions between PAMAM dendrimers and bovine serum albumin Barbara Klajnert a, Lidia StanislCawska a, Maria Bryszewska a,*, BartlComiej PalCecz b a

Department of General Biophysics, University of C Lo´dz´, ul. Banacha 12/16, C Lo´dz´ 90-237, Poland Department of Physical Chemistry, University of C Lo´dz´, ul. Pomorska 165, C Lo´dz´ 90-236, Poland

b

Received 30 July 2002; received in revised form 31 January 2003; accepted 20 February 2003

Abstract Dendrimers are a new class of polymeric materials. They are globular, highly branched, monodisperse macromolecules. Due to their structure, dendrimers promise to be new, effective biomedical materials as oligonucleotide transfection agents and drug carriers. More information about biological properties of dendrimers is crucial for further investigation of dendrimers in therapeutic applications. In this study the mechanism of interactions between polyamidoamine (PAMAM) dendrimers and bovine serum albumin (BSA) was examined. PAMAM dendrimers are based on an ethylenediamine core and branched units are constructed from both methyl acrylate and ethylenediamine. We used three types of PAMAM dendrimers with different surface groups ( – COOH, – NH2, – OH). As BSA contains two tryptophan residues we were able to evaluate dendrimers influence on protein molecular conformation by measuring the changes in the fluorescence of BSA in the presence of dendrimers. Additionally experiments with a fluorescent probe 1-anilinonaphthalene-8-sulfonic acid (ANS) were carried out. The differential scanning calorimetry (DSC) was chosen to investigate impact on protein thermal stability upon the dendrimers. Our experiments showed that the extent of the interactions between BSA and dendrimers strongly depends on their surface groups and is the biggest for amino-terminated dendrimers. D 2003 Elsevier Science B.V. All rights reserved. Keywords: PAMAM dendrimer; Serum albumin; Intrinsic fluorescence; Red edge excitation shift; Fluorescence quenching; Tryptophan fluorescence; ANS; Energy transfer; Differential scanning calorimetry

1. Introduction Dendrimers are a specific class of polymers. Unlike linear polymers they have well-defined structure. Dendrimers consist of a central core and branched monomers. The cyclic manner in which they are built results in a globular shape and a large number of end groups known as surface or terminal groups. The more layers of branched units are added the higher generation of dendrimer is obtained. Many families of dendrimers have been synthesised with various core molecules and building monomers. Polyamidoamine (PAMAM) dendrimers are based on an ethylenediamine core, and branched units are constructed from both methyl acrylate and ethylenediamine [1]. It causes that each layer is built in two steps. First, amino

* Corresponding author. Tel.: +48-42-635-44-74; fax: +48-42-635-4474. E-mail address: [email protected] (M. Bryszewska).

groups react with methyl acrylate monomers and then ethylenediamine is added: NH2 CH2 CH2 NH2 þ 4CH2 CHCOOCH3 ! NCH2 CH2 NðCH2 CH2 COOCH3 Þ4

ð1Þ

NCH2 CH2 NðCH2 CH2 COOCH3 Þ4 þ 4NH2 CH2 CH2 NH2 ! NCH2 CH2 NðCH2 CH2 CONHCH2 CH2 NH2 Þ4 þ 4CH3 OH:

ð2Þ

The half-generations of PAMAM dendrimers possess surfaces of carboxylate groups and full-generations—surfaces of amino groups. In the current studies we used polyamidoamine dendrimers both third and a half- (PAMAM G3.5) and fourth-generation (PAMAM G4) and also a fourth-generation hydroxy-terminated polyamidoamine dendrimers (PAMAM-OH G4) (Fig. 1). All these dendrimers

1570-9639/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S1570-9639(03)00117-1

116

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126

Fig. 1. The PAMAM dendrimer structure: a—PAMAM G4; b—PAMAM G3.5; c—PAMAM-OH G4.

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126

117

Fig. 1 (continued ).

have 64 end groups on their surface and molecular weight for PAMAM G3.5, PAMAM G4 and PAMAM-OH G4 equals 12 419, 14 215 and 14 279 Da, respectively. Their ˚. diameters are similar and equal to approximately 40 A The specific structure makes dendrimers suitable for a variety of biomedical applications. Among them the use of dendrimers as a drug delivery system has been of great interest. There are two interesting properties of dendrimers on the molecular level. First, it has been shown that drug molecules can be attached to dendrimers through covalent bonds. Due to a large number of terminal groups to which drug molecules can be conjugated, one molecule of dendrimer is capable to carry drugs at a high density [2– 5]. Such a use of dendrimers is based on the multiplication of certain functionalities to attain higher activity. Additionally, dendrimers offer optimal properties as linker molecules to radiolabel antibodies. Modifying antibodies with radioisotopes or cytotoxic drugs is very promising in targeted therapy. Unfortunately, modifications of an antibody molecule often eliminate its biological activity. An attractive approach is to use dendrimers as linker molecules. Dendrimer surface can bear many isotopes or drugs and then the dendrimer can be attached to the antibody. Thus, only a single site on the surface of the antibody is changed [4]. Another strategy to design dendrimers for drug delivery is using them as containers which encapsulate drug molecules [6– 9]. It is possible

because there are large, empty cavities inside dendrimers (Fig. 2). It has been shown that dendrimers with a hydrophobic interior and a hydrophilic surface are able to solubilize hydrophobic drugs in aqueous solutions [6], so they act analogously to micelles. However, dendrimers have an important advantage: they are static systems formed by covalent bonds. That is why they have been called unimolecular micelles. Drugs entrapped inside the dendrimer can be released slower. The slow release is important for antitumour drugs because it reduces their toxicity [7]. Considering the use of dendrimers for drug delivery, it is necessary to know their biological properties such as toxicity and biocompatibility. Although applications of dendrimers

Fig. 2. Two strategies for drug delivery. Black dots represent drug molecules.

118

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126

have been studied in a very broad range, their influence on biological systems has remained largely unexplored. In the present work, we investigated whether polyamidoamine dendrimers might interact with bovine serum albumin (BSA) and change its conformation. Serum albumins are the most abundant proteins in plasma. As the major soluble protein constituents of circulatory system, they have many physiological functions and play a key role in the transport of many endogenous and exogenous ligands. For many drugs binding to serum albumin is a critical determinant of their distribution and pharmacokinetics [10]. BSA possesses a high helical content (about 67%) and molecular weight of 66.4 kDa (calculated from the amino acid composition). BSA consists of 583 amino acids in a single polypeptide chain. It is built from three structurally homologous domains (I, II and III). Each domain is the product of two subdomains (IA, IB, etc.) [11]. Serum albumin is postulated to have a heart-shaped structure with dimensions ˚ [12]. of 80  80  30 A

2. Materials and methods 2.1. Materials Essentially-fatty-acid-free (fraction V) BSA and 1-anilinonaphthalene-8-sulfonic acid (ANS) were purchased from Sigma (USA). BSA was used without further purification. PAMAM dendrimers (generation 3.5 and 4), PAMAM-OH dendrimer (generation 4), acrylamide, caesium chloride, potassium iodide and L-tryptophan were obtained from Aldrich (UK). All other chemicals were of analytical grade. Water used to prepare solutions was double-distilled. 2.2. Fluorescence emission spectra of tryptophan residues in BSA BSA possesses two tryptophan residues. One of them is located on the bottom of hydrophobic pocket in subdomain IIA (Trp213), whereas the other is on the surface of the molecule in subdomain IB (Trp134) [13]. BSA was dissolved in phosphate-buffered saline (PBS: 150 mmol/l NaCl, 1.9 mmol/l NaH2PO4, 8.1 mmol/l Na2HPO4, pH 7.4) at a concentration of 5 Amol/l. Fluorescence spectra were taken with a Perkin-Elmer LS-50B spectrofluorometer. Samples were thermostatted at 25 jC. Excitation wavelength of 295 nm was used to avoid the contribution from tyrosine residues. The emission spectra were recorded from 300 to 440 nm. The excitation and emission slit widths were set to 10 and 3.4 nm, respectively. Samples were contained in 1-cm path length quartz cuvettes and were continuously stirred. Next, increasing concentrations of dendrimers, ranging from 2.5 to 85 Amol/ l, were added to BSA from a stock solution in PBS (0.9 mmol/l) and fluorescence of tryptophan residues was measured.

Before examining the fluorescent properties of the protein, it was checked that dendrimers were not excited by 295-nm wavelength and did not emit fluorescence. 2.3. pH-effect on BSA fluorescence in the presence of PAMAM G4 dendrimers BSA at a concentration of 5 Amol/l was dissolved in phosphate buffers (pH 5.7, 6.3, 6.9, 7.4 and 8.0) with 150 mmol/l NaCl (the final salt concentration equalled to 160 mmol/l). Then the experiment was carried out as above. 2.4. Fluorescence quenching measurements Fluorescence quenching studies were carried out with a neutral quencher acrylamide, and two ionic quenchers: potassium iodide and caesium chloride. Increasing aliquots of the quencher were added from a stock solution in water to 5 Amol/l sample of BSA. The stock solutions for acrylamide, KI and CsCl were 1, 5 and 10 mol/l, respectively. A stock solution of KI contained 0.1 mmol/l Na2S2O3 to prevent oxidation of I to I3. The intensity of fluorescence at the emission maximum (348 nm) was measured after excitation at 295 nm. The emission slit width was kept at 10 nm, whereas the excitation slit width was 3.4 nm. Quenching data were collected for native BSA dissolved in PBS and for BSA after adding two different concentrations of dendrimers (0.03 and 0.06 mmol/l). 2.5. Red edge excitation shift (REES) in fluorescence of BSA REES is a useful method to monitor motions around chromophores. In protein studies, it is a powerful tool to gain information about the environment and organization of tryptophans. REES is a shift in the emission maximum toward a higher wavelength caused by a shift in the excitation wavelength toward the red edge of the absorption band [14]. The excitation of tryptophan residues results in a redistribution of electronic charge inducing a significant change in both direction and strength of their dipole moment. The red edge excitation effect is a consequence of distribution of electronic transition energies of tryptophan. Excitation at the red-edge of the absorption spectrum results in a photoselection of those tryptophan residues which have the lowest electronic transition energies [15]. The REES is due to the electronic coupling between the tryptophan indole ring and neighbouring dipoles and occurs when there are slow relaxations of solvent media. If the environment of tryptophan residues is fluid the relaxation comes before tryptophan emission, and emission maximum from the relaxed state does not change with the excitation wavelength. However, if tryptophan is presented in a motionally restricted environment there is a slow rate of solvent relaxations around the excited state of tryptophan indole, and emission maximum from a nonrelaxed state depends on excitation wavelength.

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126

The fluorescence emission spectra of 5 Amol/l BSA dissolved in PBS were scanned at various excitation wavelengths ranging from 280 to 305 nm with 5-nm increase. The excitation and emission slits were set to 2.5 nm. The experiments were also performed after adding dendrimers at two different concentrations: 0.09 and 0.18 mmol/l. 2.6. Dendrimers’ impact on free L-tryptophan fluorescence L-Tryptophan was dissolved in PBS at a concentration of 50 Amol/l. The excitation wavelength of 280 nm was used and the emission spectra were recorded from 300 to 440 nm. The excitation and emission slit width were 10 and 2.5 nm, respectively. Next, increasing concentrations of dendrimers were added and fluorescence spectra were recorded.

2.7. ANS fluorescence measurements ANS was dissolved in PBS at a concentration of 100 Amol/l. The excitation wavelength of 360 nm was used and the emission spectra were recorded from 400 to 700 nm. The excitation and emission slit width were 10 and 3 nm, respectively. Then increasing aliquots of BSA were added from a stock solution in PBS until there were no changes in spectra properties. It corresponded to the state of binding of all ANS molecules. Next, increasing concentrations of dendrimers were added and changes in the fluorescence intensity for emission maximum (466 nm) were registered. The analogous experiment was carried out for the excitation wavelength of 295 nm (ANS molecules were excited via energy transfer from tryptophan residues). 2.8. BSA thermal stability The differential scanning calorimetry (DSC) was chosen to investigate changes in protein thermal stability upon the addition of dendrimers. All scans were taken with a MicroDSC III Setaram calorimeter. BSA solution (0.15 mmol/l) in phosphate buffered saline with different concentrations of dendrimers (ranging from 0.14 to 0.95 mmol/l) was loaded into a 600-Al steel cell. The reference cell was filled with phosphate buffered saline. A scan rate of 0.5 K/min was employed in a temperature range of 20– 80 jC. The maximum of the transition endotherm (Tm) was registered.

119

BSA contains two tryptophan residues, one can evaluate the interactions between protein and dendrimer molecules by studying changes of fluorescence spectra. In our studies the wavelength at the maximum of emission (kmax), the width at a half-height of maximum (D1/2) and the fluorescence intensity in the maximum were taken for the evaluation. The decrease in the fluorescence intensity was the most marked change in the fluorescence spectrum observed upon addition of dendrimers (Fig. 3). For all types of dendrimers, their increasing concentrations caused a linear reduction in the fluorescence of tryptophan residues. The effect was the strongest for PAMAM dendrimers of generation 4, less pronounced for PAMAM dendrimers of generation 3.5, and the weakest for PAMAM-OH dendrimers (Fig. 4). However, all dendrimers decreased the fluorescence intensity so significantly that their behaviour could be compared to the quenchers and the data analysed by Stern –Volmer equation: F0 =F ¼ 1 þ KSV ½Q

ð3Þ

where F0 and F are, respectively, fluorescence intensities in the absence and presence of quencher, KSV is the Stern –Vol mer dynamic quenching constant and [Q] is the concentration of the quencher. The equation assumes a linear plot of F0/F versus [Q] and the slope equals to KSV. The Stern – Volmer constants express chromophore accessibility to the quencher [16]. The Stern – Volmer constants for the quenching of tryptophan fluorescence by different dendrimers are presented in Table 1. KSV were calculated from plots shown in Fig. 4. The wavelength of the fluorescence maximum for BSA was at 348 nm, which indicated that the tryptophan residue (Trp134) is in contact with bound water molecules [17]. Commonly used method to study the environment of tryptophans is measuring the shift in the wavelength of emission maximum. The shift in the position of emission maximum corresponds to the changes of the polarity around chromo-

3. Results 3.1. Fluorescence emission spectra of tryptophan residues in BSA The basic information contained in fluorescence measurements relates to the molecular environment of the chromophore. Fluorescence of tryptophan residues is very sensitive to the changes in their vicinity, thus it is widely used to study variations of the molecular conformations of proteins. As

Fig. 3. PAMAM G4 dendrimers’ influence on BSA spectrum, kexc = 295 nm. 1—BSA solution (5 Amol/l); 2—after addition of PAMAM G4 at a concentration 22 Amol/l; and 3—after addition of PAMAM G4 at a concentration 84 Amol/l.

120

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126

Fig. 5. The scheme of double-wavelength method to evaluate the position of emission maximum. Fig. 4. Stern-Volmer plots for tryptophan fluorescence quenching by dendrimers; kexc = 295 nm, n = 6.

phore molecule. The red shift indicates that tryptophans are, on average, more exposed to the solvent, whereas the blue shift is a consequence of transferring tryptophan residues into a more hydrophobic environment [18]. After adding dendrimers, the position of emission maximum changed only very slightly (variations did not exceed 1 nm). Thus we decided to register it as a ratio of fluorescence intensities at two wavelengths (328 and 368 nm): on the left ( FL) and on the right ( FR) slopes of the spectrum [19] (Fig. 5). Fig. 6 shows the effect of dendrimers on the position of emission maximum. A slight blue-shift upon addition of PAMAM dendrimers of generation 4 and 3.5 was observed. Dendrimers of generation 4 alter the position of emission maximum stronger than dendrimers of generation 3.5 over the same concentration range. This shift indicates that the bigger contribution to the fluorescence of BSA has this tryptophan residue which is placed in a less polar environment [16]. No changes were observed for PAMAM-OH dendrimer. The last studied parameter to characterise the spectrum was the width at a half-height of maximum. According to Burstein’s [20] classification, there are five states of tryptophan residues in proteins. Each class corresponds to different location of tryptophan, from buried in extremely rigid and hydrophobic environment to totally exposed to bound or free water molecules. Spectral classes are characterised by the position of emission maximum and by its half-width. Generally, the wider the spectrum the bigger the red shift. However, in our studies a slight blue shift was accompanied by the widening of the spectrum. The effect was the biggest for PAMAM dendrimers of generation 4 and was not observed for hydroxy-terminated dendrimers (Fig. 7).

3.2. pH effect on BSA fluorescence in the presence of PAMAM G4 dendrimers The quenching of BSA intrinsic fluorescence caused by PAMAM G4 was investigated for different pH values. For all cases data were analysed by Stern –Volmer Eq. (3) and the Stern –Volmer dynamic quenching constant KSV was

Table 1 Stern – Volmer constants of the quenching of tryptophan fluorescence by dendrimers Stern – Volmer constant, KSV [mM 1] PAMAM G4 PAMAM G3.5 PAMAM-OH G4

8.38 F 0.84 3.83 F 0.56 2.87 F 0.29

Results are expressed as means F S.D. of six experiments.

Fig. 6. The effect of dendrimers on the position of emission maximum registered as a ratio of fluorescence intensities at two wavelengths: on the left ( FL) and on the right ( FR) slopes of the spectrum; n = 6.

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126

121

Fig. 8. The effect of pH on the Stern-Volmer dynamic quenching constant by PAMAM G4 dendrimers; n = 6.

directly dependent on the extent to which they can approach tryptophan residues in the protein. Quenching data for all the quenchers used in this study were analysed by Stern –Volmer equation (Eq. (1)). The Stern –Volmer plots were linear over the studied concentration range. The calculated Stern – Volmer constants are given in Table 2. The strongest quenching of BSA fluorescence was observed for acrylamide. Addition of the charged quenchers KI and CsCl led to a smaller degree of quenching. Upon addition of PAMAM dendrimers the observed quenching was considerably smaller. The most dramatic decrease was observed for quenching by caesium chloride. The least marked effect was for acrylamide quenching. The quenching by acrylamide obtained for different dendrimers was similar. For all quenchers the Stern –Volmer constant decrease was dependent on dendrimer concentrations and values of KSV were smaller for higher dendrimer concentrations. Fig. 7. The effect of dendrimers on the spectrum width at a half-height of maximum; n = 6.

evaluated. The primary amino groups on the surface of PAMAM dendrimers have a pK of 9.5 [21]. The extent of ionisation of amino groups of PAMAM dendrimers was estimated using Henderson –Hasselbalch equation: pH ¼ pK þ log

½NH2 : ½NHþ 3

ð4Þ

At studied pH range (from 5.7 to 8.0) almost all (97 – 100%) amino groups were positively charged. The sharp increase of the KSV value was observed for the quenching of fluorescence of BSA dissolved in basic buffer (pH 8.0) (Fig. 8). 3.3. Fluorescence quenching measurements Fluorescence quenching studies on BSA, both in the absence and in the presence of dendrimers, have been carried out using three different quenchers: acrylamide, caesium chloride and potassium iodide. Fluorescence quenching by these molecules proceeds by physical contact of the quencher with tryptophan indole ring and hence is

3.4. REES in fluorescence of BSA In our studies BSA showed a 2.5-nm REES, indicating that, on average, tryptophan residues were not in a motionally restricted environment (Fig. 9). Addition of PAMAMOH dendrimers of generation 4 did not significantly change the situation and for the lower concentration (0.09 mM) no effect was observed. On the contrary, PAMAM dendrimers of generation 4 and 3.5 (respectively: 4.5- and 4-nm REES for higher concentration—0.18 mM) had greater impact on Table 2 Stern – Volmer constants of the quenching of tryptophan fluorescence before and after addition of dendrimers

BSA PAMAM G4 0.03 mM PAMAM G4 0.06 mM PAMAM G3.5 0.03 mM PAMAM G3.5 0.06 mM PAMAM-OH G4 0.03 mM PAMAM-OH G4 0.06 mM

Acrylamide

KI

CsCl

14.20 F 0.25 12.88 F 0.41* 12.43 F 0.24** 13.23 F 0.60* 12.67 F 0.26** 12.88 F 0.53*

3.82 F 0.05 3.54 F 0.02** 3.17 F 0.02** 3.34 F 0.02** 3.20 F 0.03** 3.44 F 0.01**

0.22 F 0.01 0.17 F 0.01** 0.17 F 0.01** 0.15 F 0.01** 0.13 F 0.01** 0.15 F 0.01**

12.49 F 0.05** 3.43 F 0.03** 0.14 F 0.01**

Results are expressed as means F S.D. of six experiments. Statistical significance was assessed using Student – Fischer test, *P < 0.05, **P < 0.001.

122

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126

Fig. 11. The effect of dendrimers on the fluorescence intensity of ANS registered as a ratio F0/F, where F0 and F are fluorescence intensities measured at the maximum of the emission band, in the absence and presence of dendrimers.

Fig. 9. Red edge excitation curves for BSA before and after addition of dendrimers; n = 5.

3.5. Dendrimers’ impact on free L-tryptophan fluorescence

the mobility of tryptophan microenvironment and tryptophan residues had to face motional restrictions from their surrounding.

The control experiments with fluorescence of free tryptophan were carried out. The changes in the fluorescence intensity upon addition of dendrimers were registered. We observed the increase of the fluorescence intensity for PAMAM dendrimers G4 and G3.5 (Fig. 10). The intensity

Fig. 10. The effect of dendrimers on the fluorescence intensity of tryptophan; kexc = 280 nm, n = 4.

Fig. 12. Dendrimers impact on ANS fluorescence after deducting quenching effect, kexc = 295 nm.

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126

123

Fig. 13. DSC heating scans of BSA solution (0.15 mM) in the presence of dendrimers at concentration of 0.95 mM.

increase was accompanied by a red shift (results not shown). The changes in the fluorescence spectra were not observed for PAMAM-OH dendrimers G4. 3.6. ANS fluorescence measurements Changes in the fluorescence intensity of ANS binding to BSA upon dendrimers were evaluated as a ratio F0/F, where F0 and F are fluorescence intensities measured at the maximum of the emission band, in the absence and presence of dendrimers, respectively. When the samples were excited by 360-nm wavelength, we observed the intensity enhancement for PAMAM G4 dendrimers; the effect was less pronounced for PAMAM G 3.5 and no changes were observed for PAMAM-OH G4 dendrimers (Fig. 11). Using excitation wavelength of 295 nm caused reverse tendency. ANS molecules were excited by energy transfer from tryptophan residues. The intensity of this energy transfer depends on the fluorescence properties of donor molecules. Upon addition of dendrimers, tryptophan fluorescence was

quenched. The quenching effect was evaluated using the formula: 

F0 F



 

ANS

F0 F



 þ1 ¼

Trp

F0 F

 KSV  c;

ð5Þ

ANS

where KSV—Stern– Volmer constant for quenching of tryptophan residues by a dendrimer; c—dendrimer concentration. Results are presented in Fig. 12. 3.7. BSA thermal stability Heat denatures most of soluble proteins. It means that they lose their native secondary and tertiary structures and form random aggregates. This process can be considered as a thermotropic phase transition and the temperature of the maximum of the transition endotherm corresponds to the denaturation temperature (Fig. 13). Our experiments showed that increasing concentrations of dendrimers cause a loss of protein stability because denaturation occurs in lower temperatures. The extent of the interactions between BSA and dendrimers strongly depends on their surface groups and is the biggest for amino-terminated dendrimers (Fig. 14).

4. Discussion

Fig. 14. Changes in the denaturation temperature upon addition of dendrimers; n = 3.

It has been demonstrated that PAMAM dendrimers with amino surface were cleared rapidly from the circulation when administered intravenously [22,23]. However, it was not checked if they can bind to plasma proteins. The main purpose of this work was to study the effect of polyamidoamine dendrimers possessing different terminal groups on BSA conformation. There were three reasons for choosing this protein: (a) it is abundant in plasma and plays an important physiological role; (b) it is a model protein with

124

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126

well-known structure; (c) it possesses two tryptophan residues in different positions (on the surface and in the hydrophobic pocket), which is important for fluorescent measurements. The structure of dendrimers depends on their generation: lower generations of dendrimers possess an asymmetric shape and open structure, while the higher generations possess a nearly globular shape and densely packed end groups [24]. In our studies we chose to investigate the fourth generations of dendrimers. They are spherical but have enough flexibility to form on the flat surface a structure like a drop of water on a glass [25]. Fluorescence spectroscopy is one of the most popular methods in the studies of protein structure. The basic information contained in fluorescence measurements relates to the molecular environment of the chromophore. As BSA contains two tryptophan residues, we were able to study dendrimers’ impact on protein conformation by assaying the fluorescence spectra of BSA. The changes of the emission spectrum shape and the position are very sensitive indicators of tryptophan microenvironment. The observed widening of the emission spectrum indicated that tryptophans were more exposed to the solvent. However, such exposure should also be accompanied by a red shift in fluorescence spectrum. In our studies we used precise, double-wavelength method to observe spectral shift which gave a slight blue shift upon addition of PAMAM dendrimers of generation 4. In our studies, rather the changes in the spectral shape than the spectral shift were measured by the double-wavelength method. It is well known that the emission spectrum is a result of fluorescence of two tryptophan residues (Trp134 and Trp213) and the quantum yield of each fluorophore is dependent on its microenvironment. Hence, each tryptophan residue may contribute differently to the total fluorescence of the protein in changing conditions [26]. The biggest changes in the spectral shape and its position were observed for PAMAM dendrimers of generation 4. No effect was found for hydroxy-terminated dendrimers. These results are in agreement with quenching experiments. BSA fluorescence was strongly quenched by amino-terminated dendrimers, but was only poorly quenched by PAMAM-OH dendrimers and PAMAM dendrimers of generation 3.5, as shown by KSV values. In all cases the data were well fitted by a straight line, typical of a simple collision quenching mechanism. These observations support the hypothesis that surface groups and their charges are an important determinant of the dendrimer – protein interactions. It is believed that both carboxylate and amino end groups are ionised at pH 7.4. Only hydroxy-terminated dendrimers bear no charge on their surface. We carried out the control experiments with the free tryptophan solution to eliminate the hypothesis that the observed effect was only a consequence of simple diffusion processes of the dendrimers in a solution. If an accidental collision process had quenched the emission of tryptophan

residues, then the same quenching effect should have been observed for the free tryptophan. On the contrary, we obtained the increase of the fluorescence upon addition of PAMAM G4 and G3.5 dendrimers and no effect for hydroxy-terminated dendrimers. It is due to electrostatic interactions between the dendrimer surface and tryptophan molecules which in neutral pH have both amino and carboxylate groups ionised [27]. For PAMAM G4 dendrimers the pH dependence was also investigated. The KSV value sharply increased for pH 8.0. It can be a result of two facts. First, BSA is an acidic protein. The isoionic point is about 5.2. At this pH, all basic groups are protonated and all carboxyl groups are negatively charged [28]. At pH 8.0, the net charge is strongly negative, thus the electrostatic interactions between dendrimers’ amino groups and the protein molecule are enhanced. Second, BSA undergoes expansion above pH 8.0 and at pH 9.0 it changes conformation to the basic form [29]. It is a result of breaking salt bridges between domain I and domain III. It causes stronger ligand binding to subdomain IIB [30]. Therefore, better availability of Trp213 residue to the dendrimer may be involved in enhancing of quenching effect. The connection of the type of the dendrimer’s end groups with the impact on BSA structure was further confirmed by REES experiments. The extent of REES allowed for useful conclusions about the degree of BSA –dendrimer interactions. The strongest REES effect was observed for aminoterminated dendrimers. It suggests that tryptophan residues (also hydrophilic Trp134) were placed in a motionally restricted environment. It can be due to interactions between charged residues present on the surface of BSA with dendrimer charged groups because the least changes in the REES effect were for PAMAM-OH dendrimers. The use of quenchers like acrylamide, potassium iodide and caesium chloride helped reveal the link between the change in the surrounding of tryptophan residues and their accessibility for quencher molecules. Acrylamide is a polar but uncharged quencher. It is assumed that acrylamide can penetrate into the interior of proteins through diffusive processes enabled by small fluctuations in the polypeptide conformation [31]. As a consequence of this property, acrylamide has accessibility not only to Trp134 but also to Trp213 which is buried in hydrophobic pocket. Quenching of tryptophan fluorescence by ionic quenchers occurs due to heavy ion effect and requires direct collision of the ions (I or Cs+) with the excited indole ring [32]. The ionic quenchers are not expected to penetrate into the protein matrix, so they can only quench tryptophan residues which are located on the protein surface. The results show that after addition of dendrimers, BSA tryptophans become less accessible for the quencher and KSV values are considerably smaller. It is known that quenching requires molecular contact between the quencher and the fluorophore. Our data indicate that the presence of dendrimers protected BSA tryptophans from the contact with quenchers. The protective effect was the least for

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126

acrylamide quenching because acrylamide is able to penetrate the interior of the protein molecule and can also quench tryptophan residues buried in the protein. The intrinsic fluorescence studies were supplemented with experiments with the fluorescent probe ANS. ANS is one of the most universally used compounds in protein research [33]. The fluorescence enhancement after addition of BSA to ANS solution is due to the binding of ANS to protein molecules. ANS is believed to locate not too deep in the protein interior [34]. In our previous studies we showed that ANS fluorescence was increased on adding dendrimers (with an exception for PAMAM G3.5) to ANS solution [27]. This enhancement was caused by a movement of ANS molecules into dendrimer interior. Studies on protein conformational changes upon dendrimers were performed when all ANS molecules were bound to BSA. Adding dendrimers (especially PAMAM G4 and in lesser extent PAMAM G3.5) caused the increase of ANS fluorescence intensity. It indicates that the probe was placed in a more hydrophobic environment. To summarise, spectrofluorometric experiments (based on intrinsic BSA fluorescence and with ANS) show that the strength of interactions between protein and dendrimers depends on the type of dendrimer surface and can be ordered: PAMAM G4 > PAMAM G3:5 > PAMAM-OH G4: ð6Þ Some obtained results (lesser quenching of BSA fluorescence by ionic quenchers in the presence of dendrimers, increase in REES effect, enhancing ANS fluorescence) lead to assumption that a kind of layer exists on protein surface, which acts like a screen. The interactions between dendrimers and BSA are probably of electrostatic nature. That is why they are the weakest for neutral hydroxy-terminated dendrimers. At pH 7.4 (the pH of blood), BSA has the negative net charge (  17); therefore, amino-terminated cationic dendrimers have the biggest impact on the protein. However, serum albumin molecule is not uniformly charged within the structure and in physiological pH domain III is weakly positively ionised ( + 1), giving the place for possible contact between BSA surface and anionic carboxy-terminated dendrimers (Fig. 15). In the case of PAMAM G4 dendrimers, the interactions lead to exposure of hydrophobic protein region toward the surface. The hypothesis is based on strong quenching of intrinsic fluorescence by these dendrimers. Probably not only Trp134, located on the surface, but also hydrophobic Trp213 have a contact with dendrimer molecules. The conformational changes were monitored by transfer of excitation energy from tryptophan residues (donor) to ANS (acceptor). The intensity of this process depends on donor and acceptor mutual orientations. For PAMAM G3.5 and PAMAM-OH G4 results, obtained by excitation with 360 and 295 nm were similar (Fig. 12). However, after addition of PAMAM G4 dendrimers when samples were excited with

125

Fig. 15. The models of interactions between serum albumin and (A) PAMAM G4 dendrimers, (B) PAMAM G3.5 dendrimers.

295 nm, the increase in ANS fluorescence was bigger (comparing with excitation with 360 nm). This indicates changes in the distance between binding ANS molecules and tryptophan residues due to protein conformational changes. Further evidence for protein conformational disturbances were brought by DSC experiments. When the concentration of dendrimers increased, a fall in denaturation temperature (Tm) was observed. It indicates a significant destabilisation of the protein molecule. Moreover, the extent of changes followed the order (Eq. (6)) obtained from spectrofluorometric analysis. In conclusion, our experiments showed that there are interactions between PAMAM dendrimers and BSA, and the extent of the impact depends strongly on their surface groups. It is very likely that the interactions are of electrostatic nature and cause BSA conformational changes, but more detailed analysis is needed to solve this problem.

References [1] D.A. Tomalia, H. Baker, J.R. Dewald, M. Hall, G. Kallos, S. Martin, J. Roeck, J. Ryder, P. Smith, A new class of polymers: starburstdendritic macromolecules, Polym. J. 17 (1985) 117 – 132. [2] D. Zanini, R. Roy, Practical synthesis of Starburst PAMAM a-thiosialodendrimers for probing multivalent carbohydrate – lectin binding properties, J. Org. Chem. 63 (1998) 3486 – 3491. [3] R.X. Zhuo, B. Du, Z.R. Lu, In vitro release of 5-fluorouracil with cyclic core dendritic polymer, J. Control. Release 57 (1999) 249 – 257. [4] J.C. Roberts, Y.E. Adams, D. Tomalia, J.A. Mercer-Smith, D.K. Lavallee, Using starburst dendrimers as linker molecules to radiolabel antibodies, Bioconjug. Chem. 2 (1990) 305 – 308. [5] R.F. Barth, D.M. Adams, A.H. Soloway, F. Alam, M.V. Darby, Boronated starburst dendrimer-monoclonal antibody immunoconjugates: evaluation as a potential delivery system for neutron capture therapy, Bioconjug. Chem. 5 (1994) 58 – 66. [6] L.J. Twyman, A.E. Beezer, R. Esfand, M.J. Hardy, J.C. Mitchell, The synthesis of water soluble dendrimers, and their application as possible drug delivery systems, Tetrahedron Lett. 40 (1999) 1743 – 1746. [7] C. Kojima, K. Kono, K. Maruyama, T. Takagishi, Synthesis of polyamidoamine dendrimers having poly(ethylene glycol) grafts and their ability to encapsulate anticancer drugs, Bioconjug. Chem. 11 (2000) 910 – 917. [8] M. Liu, K. Kono, J.M.J. Fre´chet, Water-soluble dendritic unimolecu-

126

[9]

[10]

[11] [12]

[13] [14]

[15]

[16]

[17]

[18]

[19]

[20] [21]

[22]

B. Klajnert et al. / Biochimica et Biophysica Acta 1648 (2003) 115–126 lar micelles: their potential as drug delivery agents, J. Control. Release 65 (2000) 121 – 131. M. Liu, J.M.J. Fre´chet, Preparation of water-soluble dendritic unimolecular micelles as potential drug delivery agents, Polym. Mater. Sci. Eng. 80 (1999) 167 – 168. A.A. Bhattacharya, S. Curry, N.P. Franks, Binding of the general anesthetics propofol and halothane to human serum albumin, J. Biol. Chem. 275 (2000) 38731 – 38738. T.J. Peters, All About Albumin, Biochemistry, Genetics and Medical Applications, Academic Press, San Diego, 1996. O.J. Bos, J.F. Labro, M.J.E. Fischer, J. Witling, L.H.M. Janssen, The molecular mechanism of the neutral-to-base transition of human serum albumin, J. Biol. Chem. 264 (1989) 953 – 959. D.C. Carter, X.J. Ho, Structure of serum albumins, Adv. Protein Chem. 45 (1994) 153 – 203. A.P. Demchenko, Site-selective excitation: a new dimension in protein and membrane spectroscopy, Trends Biochem. Sci. 13 (1988) 374 – 377. A.H.C. de Oliveira, J.R. Giglio, S.H. Andriao-Escarso, R.J. Ward, The effect of resonance energy homotransfer on the intrinsic tryptophan fluorescence emission of the bothropstoxin-I dimer, Biochem. Biophys. Res. Commun. 284 (2001) 1011 – 1015. M.R. Eftink, Fluorescence quenching reactions: probing biological macromolecular structures, in: T.G. Dewey (Ed.), Biophysical and Biochemical Aspects of Fluorescence Spectroscopy, Plenum, New York, 1991, pp. 1 – 41. A. Filenko, M. Demchenko, Z. Mustafaeva, Y. Osada, M. Mustafaev, Fluorescence study of Cu2 +-induced interaction between albumin and anionic polyelectrolytes, Biomacromolecules 2 (2001) 270 – 277. P.R. Callis, 1La and 1Lb transitions of tryptophan: applications of theory and experimental observation to fluorescence of proteins, Methods Enzymol. 237 (1997) 113 – 150. E.A. Lapshina, I.B. Zavodnik, Microcalorimetric and fluorescent studies of pH-induced transitions in erythrocyte membranes, Biol. Membr. 7 (1994) 155 – 163. E.A. Burstein, Equilibrium dynamics of native protein structure (in Russian), USSR Acad. Sci. Pushchino (1977) 60 – 72. O.M. Milhem, C. Myles, N.B. McKeown, D. Attwood, A. D’Emanuele, Polyamidoamine Starburst dendrimers as solubility enhancers, Int. J. Pharm. 197 (2000) 239 – 241. J.C. Roberts, M.K. Bhalgat, R.T. Zera, Preliminary biological evaluation of polyaminoamine (PAMAM) Starburstk dendrimers, J. Biomed. Mater. Res. 30 (1996) 53 – 65.

[23] N. Malik, R. Wiwattanapatapee, R. Klopsch, K. Lorenz, H. Frey, J.-W. Weener, E.W. Meijer, W. Paulus, R. Duncan, Dendrimers: relationship between structure and biocompatibility in vitro, and preliminary studies on the biodistribution of 125I-labelled polyamidoamine dendrimers in vivo, J. Control. Release 65 (2000) 133 – 148. [24] G. Caminati, N.J. Turro, D.A. Tomalia, Photophysical investigation of starburst dendrimers and their interactions with anionic and cationic surfactants, J. Am. Chem. Soc. 112 (1990) 8515 – 8522. [25] M. Wells, R.M. Crooks, Interactions between organized, surface-cofined monolayers and vapour-phase probe molecules. Preparation and properties of chemically sensitive dendrimer surfaces, J. Am. Chem. Soc. 118 (1996) 3988 – 3989. [26] S.S. Komath, M.J. Swamy, Fluorescence quenching, time-resolved fluorescence and chemical modification studies on the tryptophan residues of snake gourd (Trichosanthes anguina) seed lectin, J. Photochem. Photobiol., B Biol. 50 (1999) 108 – 118. [27] B. Klajnert, M. Bryszewska, Interactions of tryptophan and ANS with PAMAM dendrimers, Cell. Mol. Biol. Lett. (in press). [28] F.W. Putman (Ed.), The Plasma Proteins, Academic Press, New York, 1960. [29] S. Era, K.B. Itoh, M. Sogami, K. Kuwata, T. Iwama, H. Yamada, H. Watari, Structural transition of bovine plasma albumin in the alkaline region—the N – B transition, Int. J. Pept. Protein Res. 35 (1990) 1 – 11. [30] O.J. Bos, J.F. Labro, M.J. Fischer, J. Wilting, L.H. Janssen, The molecular mechanism of the neutral-to-base transition of human serum albumin. Acid/base titration and proton nuclear magnetic resonance studies on a large peptic and a large tryptic fragment of albumin, J. Biol. Chem. 264 (1989) 953 – 959. [31] R.M. France, S.H. Grossman, Acrylamide quenching of apo- and holo-a-lactalbumin in guanidine hydrochloride, Biochem. Biophys. Res. Commun. 269 (2000) 709 – 712. [32] S. Ercelen, D. Kazan, A. Erarslan, A.P. Demchenko, On the excitedstate energy transfer between tryptophan residues in proteins: the case of penicillin acylase, Biophys. Chemist. 90 (2001) 203 – 217. [33] J. Slavik, Anilinonaphthalene sulfonate as a probe of membrane composition and function, Biochim. Biophys. Acta 694 (1982) 1 – 25. [34] E. Scho¨nbrunn, S. Eschenburg, K. Luger, W. Kabsch, N. Amrhein, Structural basis for the interaction of the fluorescence probe 8-anilino1-naphthalene sulfonate (ANS) with the antibiotic target MurA, Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 6345 – 6349.