A Microcalorimetry and Spectroscopy Study
on the Interaction of BSA with 2,2′-Bipyridine Octylglycinato Palladium(II)
Nitrate

MANSOORI-TORSHIZI
Hassan^1,2, ISLAMI-MOGHADDAM Mahbobe², SABOURY Ali Akbar¹*

( ¹ Institute of Biochemistry and Biophysics, University of Tehran, Tehran,
14176－14411, Iran; ² Department of Chemistry,
University of Sistan & Bluchestan, Zahedan, 98167－45345, Iran )

Abstract        The
interaction of bovine serum albumin (BSA) with a new palladium(II) complex [Pd(bpy)(Oct-Gly)]NO3
(bpy, 2,2’-bipyridine; Oct-Gly, octyl-glycine) was studied by isothermal
titration UV-visible spectrophotometry and microcalorimetry in 30 mmol/L Tris
buffer, pH 7.0. There is a set of 18 binding sites for this complex on BSA at
300 and 310 K with positive cooperativity in the binding process. The Hill
coefficients at 300 and 310 K are 2.2 and 2.4, respectively. The binding of
this palladium complex on BSA is endothermic with mean association binding
constant of 21.0 and 16.4 (mmol/L)^–1 at 300 and 310 K, respectively.
The complex can denature the protein as surfactants. The stability of BSA in
the interaction study with the complex is 84 and 58 kJ/mol at 300 and 310 K,
respectively. Also, the enthalpy of BSA denaturation due to the interaction
with the complex is 842 kJ/mol.

Key
words     serum albumin; palladium complex; isothermal titration
microcalorimetry; spectrophotometry

cis-Diamminedichloroplatinum(II)
(cisplatin), first identified as an antitumor drug in the late 1960’s[1], has been found as an
anticancer agent against testicular tumor, ovarian carcinomas, squamous
carcinomas, and a variety of sarcomas[2]. A large number of analogs of
cisplatin have also been tested. It has been reported that many active
complexes could react with DNA and inhibit its synthesis[3]. Other transition
metal complexes with favorable antitumor activity are rhodium and palladium
complexes[4,5]. The development of palladium anticancer drugs has not been
promising and their design has mainly been based on the structure-activity
relationship used for platinum anticancer drugs as well as good models for the
analogous Pt(II) complexes in solution. This is mainly because palladium
complexes are about 105 times more reactive than their Pt(II) analogs leading
to rapid hydrolysis of the leaving group/groups, thus the reactive species
formed is unable to reach their pharmacological targets[6]. This problem may be
solved by looking beyond the structure-activity relationship and identifying
novel Pd(II) compounds having chelating ligand which may not readily be
hydrolyzed and can be utilized as building blocks for palladium antitumor
drugs.

It has been
reported that the failure of cisplatin for the treatment of tumors of gastrointestinal
region is mainly due to high concentration of chloride in this region[6].
However, palladium complexes having chelating ligands are expected to be useful
for the treatment of tumors of the gastrointestinal track because they do not
interact with chloride ions. Such palladium complexes are also expected to have
low kidney toxicity than cisplatin due to inability of replacing the tightly
bound chelate ligands of Pd(II) with sulfhydryl groups of proteins kidney
tubules[7]. Among the palladium complexes including chelating ligands,
[Pd(II)(2,2’-bipyridine)(amino acid)]n+ complexes have aroused great interest
because they could be active as antiviral or antitumoral agents[8,9]. It is
thus reasonable to expect that analogs of these compounds may be found with
superior activities in animals cytotoxic studies. Also, the substitution of
chloride ligands of cisplatin with molecules such as amino acids or amino acid
derivatives that are found in biological systems, may decrease its toxic side
effects. However, less information is available about the interaction of Pd(II)
and Pt(II) anticancer complexes with proteins which may reflect the side
effects of these agents. Recently we have started this work[10,11]and found
that there is an interaction between 2,2’-bipyridineglycinatopalladium(II)
chloride and human serum albumin (HSA).

In this paper we
report the interaction of a new palladium(II) complex of formula
[Pd(bpy)(Oct-Gly)]NO3 (bpy, 2,2’-bipyridine; Oct-Gly, octyl-glycine) with
bovine serum albumin (BSA) by isothermal titration UV-visible spectrophotometry
and microcalorimetry. Synthesis, characterization, cytotoxic and DNA binding
studies of the above Pd(II) complex is being published else where.

1    Materials and
Methods

1.1   Materials

Bovine serum albumin (BSA) was from Sigma.
Palladium(II) complex [Pd(bpy)(Oct-Gly)]NO3(bpy, 2,2’-bipyridine; Oct-Gly,
octyl-glycine) was synthesized in this lab, its synthesis, characterization and
cytotoxic will be published else where. All other materials and reagents were
of analytical grades, and solutions were made in double-distilled water. 30
mmol/L Tris-HCl solution (pH 7.0) was used as a buffer.

1.2   Methods

1.2.1       Spectrophotometry
for ligand binding study         The
interaction of [Pd(bpy)(Oct-Gly)]NO₃ complex with BSA was followed
by difference absorption spectral technique[12]. The stock solution of BSA was
prepared by dissolving in Tris-HCl buffer by gentle stirring at room
temperature. The concentration of BSA was measured spectrophotometrically using
extinction coefficient of 6.67 for 1% protein solution by length light path of
1 cm at 279 nm[13]. The concentration of stock solution of BSA was 1.20 mg/mL.
Palladium(Pd) complex stock solution was also prepared with Tris buffer to a
concentration of 0.75 mmol/L.

Pd complex with
different concentrations (0.02－0.32 mmol/L) with BSA (1.2 mg/mL, equal to 0.0182 mmol/L) or without
it were incubated for 30 min at 300 K (or 310 K). Then the spectrophotometric
readings at 314 nm (A314) of the mixtures were taken which will keep unchanged
even for 24 h. The difference absorption of Pd complex with or without BSA
(ΔA314, ΔA at wavelength of 314 nm) was calculated.

When all the
binding sites on BSA were bound by Pd complex, ΔA was designated ΔAmax as which
can be determined by extrapolation of the plot of reciprocal of ΔA against to
the reciprocal of BSA concentration at 314 nm. 0.075 mmol/L Pd complex was
incubated without or with BSA (different fixed concentrations of 0.07－0.83 g/L), and the difference
absorption at 314 nm was recorded.

1.2.2       Spectrophotometry
for protein denaturation study        The
protein denaturing effect of Pd complex was detected using a recording
spectrophotometer (JASCO 7850 model, Japan).
The sample cell containing 1.8 mL 0.8 mg/mL BSA was incubated with Pd complex
solution from 0－0.65 mmol/L.
The reference cell was excluded of BSA. Both groups were set at a constant
temperature, 300 or 310 K. The absorption of the sample cells was recorded at
280 nm versus the reference cells.

1.2.3       Isothermal
titration microcalorimetry    Enthalpy
measurements were carried out with a four-channel commercial microcalorimetry
(Thermal activity monitor 2277, Thermometric, Sweden).
Each channel is a twin heat conduction calorimeter where the heat-flow sensor
is a semiconducting thermopile (multijunction thermocouple plates) positioned
between the vessel holders and the surrounding heat sink. The insertion vessel
was made of stainless steel. Every time, 20 μL 5.0 mmol/L ligand solution was
injected into the calorimetric stirred titration vessel which contained 1.8 mL
BSA (0.8 mg/mL) in 30 mmol/L Tris-HCl buffer (pH 7.0) by using a Hamilton
syringe. The thin (0.15 mm inner diameter) stainless steel hypodermic needle of
the syringe directly reached into the calorimeter vessel. The injection of
ligand into the perfusion vessel was repeated 20 times. In the control group,
no BSA but only 1.8 mL of Tris-HCl buffer was prepared in the calorimetric
stirred titration vessel. The rate of heat output of both groups was recorded with
an accuracy of 0.1 μW by a computerized recording system. The enthalpy change
for each injection was calculated by a “Digitam 3” computer program. The
enthalpy of dilution of the ligand solutions were measured as described above
when the BSA protein was excluded. The enthalpy of dilution for the ligand was
subtracted from the enthalpy of protein-ligand interaction. The enthalpy of
dilution of BSA is negligible. The microcalorimetry was calibrated electrically
frequently during the course of the study. The molar mass of the BSA protein
was taken as 66 kD for all calculations[14].

2    Results and Discussion

The results of
the binding of Pd complex to BSA at different temperature of 300 and 310 K are
shown in Fig.1(A). ΔA was the difference absorption at 314 nm and was positive
related to the binding of ligand to BSA. In Fig.1(B), the reciprocal
relationship between ΔA at 314 nm and the concentration of BSA at a fixed Pd
complex concentration (0.075 mmol/L) was shown, and two linear equations can be
deduced from the results as following.

(ΔA)^–1=0.96[BSA]^–1+11.00  at 300 K (R²=0.99)

(ΔA)^–1=1.75[BSA]^–1+6.97    at 310 K (R²=0.99)

R², the linear correlation coefficient

When [BSA]^－1=0, ΔAmax equals to 0.091 at 300 K and
0.143 at 310 K, The concentration of bound ligand, [L]b in Fig.1(A) can be
obtained as multiplying ΔA314〖〗ΔAmax by
[L]t (0.075 mmol/L) in Fig.1(B)[12]. The binding isotherm was constructed by
treatment of the data using standard computer program of linear least-square
fitting according to the equation (1).

v=g(K[L]f)ⁿ/1+(K[L]f)ⁿ                            
(1)

Where ν is the average number of bound ligands to one macromolecule of BSA
and [L]f is the concentration of free ligand.

Fig.1       The
difference absorption of ligand binding reaction with fixed BSA or Pd complex
at 314 nm

(A) The difference absorption for two analogous sets of ligand
solutions with and without BSA at 314 nm, ΔA314, given by
mili-absorbance (mAbs). The concentration of BSA is fixed in 1.2 g/L (equal to
0.0182 mmol/L). The total concentration of ligand, [L]t, is in the range of 20－380 μmol/L. (B) The
linear plot of the reciprocal of ΔA314 versus the reciprocal of
[BSA] for difference absorption of two analogous sets of ligand solutions
(fixed concentration of 0.075 mmol/L) with and without BSA (different
concentrations of 0.07－0.83 g/L).

This binding
isotherm could be plotted as v versus lg[L]f as shown in Fig.2(A), In
this case, the Scatchard plots[15] are downward curves at 300 and 310 K, as
shown in Fig.2(B). Therefore, the binding of Pd complexes to BSA is
cooperative[16,17] at both temperatures. The number of binding sites (g), the
apparent equilibrium constant (K) and the Hill coefficient (n),
as a criterion of cooperativity in the binding process) can be obtained by
fitting of experimental data to the Hill equation [equation (1)][18,19]. The
binding data shown in Fig.2(A) for the binding of ligand to BSA have been
fitted to Hill equation using a computer program for nonlinear least-square
fitting[20]. The results are tabulated in Table 1. The maximal error on
experimental values of ν, according to those values obtained by using these results (Table
1) in equation (1), is 0.5. There is a set of 18 binding sites for Pd complex
on BSA at 300 and 310 K. The Hill coefficients at 300 and 310 K are 2.2 and
2.4, respectively. So the cooperativity of binding is positive.

Fig.2       Binding
isotherm in ν versus lg[L]f plot , and Scatchard
plot

(A) Binding isotherms in ν versus lg[L]f plot for
ligand of 2,2’-bipyridine octylglycinato palladium(II) nitrate on the
interaction with BSA (pH 7.0). (B) Scatchard plot. ν, the average
number of bound ligand to one macromolecule of BSA; [L]f, the free
concentration of ligand in “mmol/L”.

Table 1   Values
of binding parameters in the Hill equation for interaction between Pd(II)
complex and BSA

The parameters were in 30 mmol/L
Tris-HCl buffer, pH 7.0. g, the number of binding sites; K, the apparent equilibrium
constant; n, the Hill coefficient (as a criterion of cooperativity).

The profiles of denaturation of BSA by
2,2’-bipyridine octylglycinato palladium(II) nitrate (Pd complex) are shown in
Fig.3. The concentration of ligand in the midpoint of transition, [L]1/2, is
decreased by improving temperature, from 0.55 mmol/L at 300 K to 0.40 mmol/L at
310 K. So, the increasing of temperature also lower the stability of the
protein against denaturation caused by this ligand.

Fig.3       The
change of absorbance of BSA (0.8 mg/mL) at λmax=280 nm
due to increasing the total concentration of ligand [L]t

The standard Gibbs
free energy of protein unfolding (ΔG°, the work
needed for protein denaturation) was calculated as a function of ligand
concentration by assuming two-state mechanism and using the equations
(2,3)[21,22]:

F_D=[A_N－A_obs]/[A_N－A_D]                        
                     (2)

ΔG°=[－RTlnF_D]/[1－F_D]                                          
(3)

Where Aobs
is the observed absorbance used to follow unfolding in the transition region,
AN and AD are the values of absorbance to the native and denatured conformations
of the protein, respectively. FD represents the fraction of the protein present
in the denatured state. Fig.4 shows the free energy of unfolding, which is
calculated from equation (3) based on the data in the Fig.3, which varies
linearly with ligand concentration in the limited region.

Fig.4       The
standard Gibbs free energy of unfolding ΔG° vs. [L]t

ΔG°, calculated from equation (3) by assuming a two-state mechanism; [L]_t,
total concentration of ligand at two temperatures in the base of data shown in
Fig.3. In fact, ΔG° is the work needed for protein denaturation.

The simplest
method of estimating the conformat-ional stability in the absence of denaturant
ligand, ΔG° (H₂O), in two
temperatures, is to assume that linear dependence continues to zero
concentration and to use a least-square analysis to fit the data to the
equation (4)[23]:

ΔG°=ΔG°(H₂O)－m[L]                                     
     (4)

Where m is a
measure of the dependence of ΔG° on ligand concentration. The values of
ΔG°(H2O) are 84.0 and (58.0±0.2) kJ/mol at two temperatures of 300 and 310 K,
respectively. Therefore, BSA is more stable at 300 K and the existence of
ligand led to less stability of the protein. Moreover, values for m are 1.51×105 and 1.45×105 (kJ/mol)・(mol/L)－1 at two temperatures of 300 and
310 K, respectively. The m value is a measure of ligand strength for protein
denaturation. These m values for this ligand are very similar to surfactants[24－26].

The standard enthalpy of denaturation by
the Pd complex (ΔH°) can be calculated using
Gibbs-Helmholtz equation (5)[27]:

ΔH°={[ΔG°(T₁)/T₁]－[ΔG°(T₂)]T₂}/[1/T₁－1/T₂]                     
(5)

Where ΔG°(T₁)
and ΔG°(T2) are the Gibbs free energies of protein denaturation at two
temperatures of T1 and T2, respectively, which were shown in Fig.4. In fact,
ΔH° is the heat needed for protein denaturation. The enthalpy curve of protein
denaturation in different concentration of the complex is shown in Fig.5. The
simplest method of estimating the enthalpy of protein denaturation in the
absence of denaturant ligand, ΔH°(H2O) is to assume that linear dependence
continues to zero concentration of the ligand. Hence, the enthalpy of BSA
denaturation by the complex is 842 kJ/mol.

Fig.5       ΔH°
curve of BSA denaturation due to the interaction with Pd(II) complex

The standard enthalpy (ΔH°) curve of BSA denaturation 2,2’-bipyridine octylglycinato
palladium(II) nitrate at a temperature region 300－310 K. [L]t, the total
concentration of ligand. In fact, ΔH° is the heat
needed for protein denaturation.

In addition, the
entropy (ΔS°) of protein denaturation by complex can be calculated as follows.

ΔS°(H₂O)=[ΔH°(H₂O)－ΔG°(H₂O)]/T=[(842－84) kJ/mol]/(300 K)=2.53 (kJ/mol)K^–1

The positive
value of entropy change related to the more disorder of denatured protein respect
to the native protein.

The data
obtained from isothermal titration micro-calorimetry of BSA interaction with
ligand is shown in Fig.6. Fig.6(A) shows the heat of each injection (q) and
Fig.6(B) shows the heat related to each total concentration of ligand. “q” is
the heat of interaction, which related to two processes simultaneously: (1)
ligand binding to the protein; (2) the protein denaturation. Fig.6(B) shows a
distinct maxima represents an endothermic unfolding process within an
exothermic binding process. Concentrations of ligand related to the region of
minima and maxima are also in good coincidence with profile of protein
denaturation obtained by spectrophotometric method (Fig.3).

On the basis of
these results it can be concluded that the Pd(II) complex can denture the
protein by exothermically binding process including endothermically
denaturation process, and there is a set of 18 binding sites for ligand on the
protein with positive cooperativity in the binding process. The strength of
this complex for protein denaturation is similar to the surfactant. There are
two sets of binding sites for surfactant interaction with protein. The
electrostatic interaction, which is accompanied by a preliminary hydrophobic
interaction, occurs initially and is followed by a more extensive pure
hydrophobic interaction[19,26]. The predominant unfolding of a protein is
related to the first interaction, in which neutralization of charges at the
surface of the protein perturb the balance of forces in the protein structure[19,28].
However, in protein denaturation by Pd(II) complex, there is only one set of
binding sites with positive cooperativity in the binding process. Hence, the
binding process is different for Pd(II) complex in comparison with surfactant.
We have also observed the difference in the binding process for another Pd
complex[10,11]. Also, it is expected that one of the side effects of antitumor
palladium group metal complexes might be due to their interaction with
proteins.

Fig.6       Heat change related with the
ligand injection to BSA

(A) The heat of ligand interaction (q)
on BSA for 20 automatic cumulative injections, each of 20 μL, of ligand solution
5 mmol/L, into the sample cell containing 1.8 mL BSA solution at a
concentration of 0.8 mg/mL at pH=7.0 and 300 K. (B) The cumulative heat related
to each total concentration of ligand [L]. In fact, q is related to two
processes simultaneously: (1) ligand binding to the protein; (2) the protein
denaturation.

References

1     Rosenberg
B, van Camp L. The successful regression of large solid sarcoma 180 tumors by
platinum compounds. Cancer Res, 1970, 30(6), 1799－1802

2     Hollis
LS, Amundsen AR,
Stern EW. Chemical and biological properties of cis-diammineplatinium(II)
antitumor agents containing three nitrogen donors: cis-[Pt(NH3)(N-donor)Cl]+. J
Med Chem, 1989, 32(1): 128－136

3     Barton
JK, Lippard SJ. Heavy metal interactions with nucleic acids. In: Spiro TG ed.
Metal Ions in Biology. Vol. 1, New York, USA:
Wiley-InterScience, 1980, 31－114

4     Barnard
CFJ, Cleare MJ, Hydes PC. Second generation anticancer platinum compounds. Chem
Brit, 1986, 22(11): 1001－1004

5     K�epf-Meier P, K�epf H. Non-platinum group metal antitumor
agents. History, current status, and perspectives. Chem Rev, 1987, 87(5): 1137－1152

6     Hacker
MP, Douple EB, Krakoff IH. Platinum Coordination Complexes in Cancer
Chemotherapy. Boston, MA: Nijhoff Publisher, 1984, 267－278

7     Zhao
G, Sun H, Lin H, Zhu S, Su X, Chen Y. Palladium(II) complexes with
N,N’-dialkyl-1,10-phenanthroline-2,9-dimathan-amine: Synthesis,
characterization and cytotoxic activity. J Inorg Biochem, 1998, 72(3-4): 173－177

8     Farrel
N. Transition Metal Complexes as Drugs and Chemotherapeutic Agents, Dordrecht,
Netherlands: Kluwer
Academic Publishers, 1989

9     Noji
M, Okamoto K, Kidani Y, Tashiro T. Relation of conformation to antitumor
activity of platinum(II) complexes of 1,2-cyclohexanediamine and
2-(aminomethyl) cyclohexylamine isomers against leukemia P388. J Med Chem,
1981, 24(5): 508－515

10    Saboury
AA, Shamsaei AA, Moosavi-Movahedi AA, Mansuri-Torshizi H. Thermodynamics of
binding 2,2’-bipyridineglycinato palladium(II) chloride on human serum albumin.
J Chin Chem Soc, 1999, 46(6): 917－922

11    Shamsaei
AA, Saboury AA, Moosavi-Movahedi AA, Mansori-Torshizi H. Human serum albumin
binding studies of 2,2’-bipyridineglycinato palladium(II) chloride. Biophys J,
1999, 76(1): A101

12    King
AMQ, Nicholson BH. The interaction of aflatoxin B1 with polynucleotides and its
effect on ribonucleic acid polymerase. Biochem J, 1969, 114(5): 679－687

13    Harmsen
BJM, de Bruin SH, Janssen LHM, Rodrigues de Miranda JF, van Os AJ. pK changes
of imidazol groups in bovine serum albumin due to conformational change at
neutral pH. Biochemistry, 1971, 10(17): 3217－3221

14    Privalov
PL. Stability of proteins: small globular proteins. Adv Prot Chem, 1979, 33
(1): 167－234.

15    Scatchard
G. The attractions of proteins for small molecules and ions. Ann NY Acad Sci,
1949, 51(4): 660－672

16    Saboury
AA, Moosavi-Movahedi AA. Evaluation of Hill coefficient from Scatchard and
Klotz plots. Biochem Educ, 1994, 22(1): 48－49

17    Bordbar
AK, Saboury AA, Moosavi-Movahedi AA. The shapes of Scatchard plots for systems
with two sets of binding sites. Biochem Educ, 1996, 24(3): 172－175

18   Hill
AV. The possible effects of the aggregation of the molecules of haemoglobin on
its oxygen dissociation curve. J Physiol (Lond), 1910, 40(1): 4－7

19    Saboury
AA, Bordbar AK,
Moosavi-Movahedi AA. Resolution of two sets of binding sites for cationic
surfactant-urease interaction. Bull Chem Soc Jp, 1996, 69(11): 3031－3035

20    James
ML, Smith GM, Wolford JC. Applied Numerical Methods for Digital Computer, 3rd
ed, New York, USA:
Harper and Row Publisher, 1985

21    Pace
CN. Measuring and increasing protein stability. TIBTECH, 1990, 8(1): 93－98

22    Saboury
AA, Moosavi-Movahedi AA. Derivation of the thermodynamic parameters involved in
the elucidation of protein thermal profiles. Biochem Educ, 1995, 23(3): 164－167

23    Greene
RF, Pace CN. Urea and guanidine hydrochloride denaturation of ribonuclease,
lysozyme, α-chymotrypsin, and β-lactoglobulin. J Biol Chem, 1974, 249(17): 5388－5393

24    Moosavi-Movahedi
AA, Nazari K, Saboury AA. Thermodynamic denaturation of horseradish peroxidase
with sodium n-dodecyl sulphate and n-dodecyl trimethylammonium bromide.
Colloids and surf B: Biointerfaces, 1997, 9(1): 123－130

25    Saboury
AA, Karbassi F. Thermodynamic studies on the interaction of calcium with
α-amylase. Thermochim Acta, 2000, 362(1): 121－129

26    Moosavi-Movahedi
AA, Saboury AA. Elucidation of binding sites for protein denaturation by
surfactant. Jour Chem Soc Pak, 1999, 21(3): 248－259

27    Atkins
PW. Physical Chemistry, 6th ed, Oxford, UK:
Oxford University Press, 1998, Chapter 5

28    Saboury
AA, Bordbar AK,
Moosavi-Movahedi AA. The enthalpy of unfolding for jack bean urease with
interaction of n-alkyl trimethylammonium bromides. J Chem Thermodyn, 1996,
28(10): 1077－1082

_________________________________________________

Received: March 13, 2003Accepted: July 9,
2003

This work was financially supported by the
Research Council of the University of Tehran and University of Sistan
& Bluchestan

*Corresponding author: Tel, +98-21-6956984;
Fax, +98-21-6404680; e-mail, [email protected]

Updated at: 2003-10-05