Bioquimica

Effect of Human Serum Albumin on the Kineticsof 4-Methylumbelliferyl-b-D-N-N0-N00 TriacetylchitotriosideHydrolysis Catalyzed by Hen Egg White Lysozyme

Cristian Calderon Elsa Abuin Eduardo Lissi

Rodrigo Montecinos

Published online: 12 July 2011

Springer Science+Business Media, LLC 2011

Abstract The effect of human serum albumin (HSA)

addition on the rate of hydrolysis of the synthetic substrate

4-methylumbelliferyl-b-D-N-N0-N00 triacetylchitotrioside((NAG)3-MUF) catalyzed by hen egg white lysozyme has

been measured in aqueous solution (citrate buffer 50 mM

pH = 5.2 at 37 C). The presence of HSA leads to a decreasein the rate of the process. The reaction follows a Michaelis

Menten mechanism under all the conditions employed. The

catalytic rate constant decreases tenfold when the albu-

min concentration increases, while the Michaelis constant

remains almost constant in the albumin concentration range

employed. Ultracentrifugation experiments indicate that the

main origin of the observed variation in the kinetic behavior

is related to the existence of an HSAlysozyme interaction.

Interestingly, the dependence of the catalytic rate constant

with albumin concentration parallels the decrease of the free

enzyme concentration. We interpret these results in terms of

the presence in the system of two enzyme populations;

namely, the HSA associated enzyme which does not react

and the free enzyme reacting as in the absence of albumin.

Other factors such as association of the substrate to albumin

or macromolecular crowding effects due to the presence of

albumin are discarded. Theoretical modeling of the structure

of the HSAlysozyme complex shows that the Glu35 and

Asp52 residues located in the active site of lysozyme are

oriented toward the HSA surface. This conformation will

inactivate lysozyme molecules bound to HSA.

Keywords Enzyme kinetics Lysozyme Human serumalbumin

Abbreviations

DS Dansylsarcosine

DNSA Dansylsulfonamide

GPNA N-glutaryl-L-phenylalanine

p-nitroanilide

HSA Human serum albumin

kcat Catalytic rate constant

KM Michaelis constant

((NAG)3-MUF) 4-Methylumbelliferyl-b-D-N-N0-N00

triacetylchitotrioside

Glu Glutamic acid

Asp Aspartic acid

1 Introduction

Serum albumin is the most abundant protein in blood

plasma, (accounting for ca. 60% of the total protein; its

concentration amounts to ca. 800 lM [11, 16]. Albumins,and in particular HSA are able to bind and transport a wide

range of organic compounds, including drugs and fatty

acids [1, 2, 10, 12, 13, 17, 19, 20, 24, 27]. This capacity of

albumin to bind hydrophobic drugs in plasma modulates

their delivery to cells in vivo and reduces the free drug

concentration, a determinant parameter of its physiological

activity [14].

The presence of albumin in the reaction medium can

affect the activity of enzymes through different factors,

namely, (a) the adsorption of the substrate onto the protein,

(b) the so called crowding effect which would affect the

enzyme conformation, and, hence, its activity, and (c) an

C. Calderon (&) E. Abuin E. LissiFacultad de Qumica y Biologa, Universidad de Santiago de

Chile, Av. Bernardo OHiggins 3363, Santiago, Chile

e-mail: [email protected]

R. Montecinos

Facultad de Qumica, Pontificia Universidad Catolica de Chile,

Santiago, Chile

123

Protein J (2011) 30:367373

DOI 10.1007/s10930-011-9339-8

interaction between the enzyme and the protein which

would also alter the enzyme conformation even at low

albumin concentrations. There are several reports that take

into account factors (a) and (b), but, to the best of our

knowledge, factor (c) has not been yet considered in kinetic

studies.

Awad-Elkarim et al. [4] demonstrated that p-nitrophenyl

phosphate, a substrate for Lipases, readily binds to bovine

and human serum albumin. Wang et al. [32] studied the

effect of addition of albumin on the acyl chain specificity

of a lipoprotein lipase using triacylglycerols of various

acyl-chain lengths as substrates. An unexpected finding of

this work was that the albumin ligand binding site is

accessible not only to long-chain fatty acids, but also to

short and medium chain monoacid triacylglycerol sub-

strates. It was concluded that the observed inhibitory effect

of albumin on the lipase-catalyzed hydrolysis of trihexa-

noylglycerol is probably the result of the high affinity

interaction of albumin with this substrate, but no inde-

pendent measurements of the binding affinity were per-

formed. The effect of substrate depletion due to its

adsorption onto albumin was also demonstrated in a study

of the effect of human serum albumin (HSA) on the in

vitro enzyme kinetics of the formation of hydroxytolbu-

tamide on the tolbutamide hydroxylation examined using

human liver microsomes [33]. In this system, the addition

of HSA greatly decreased the unbound concentration of

tolbutamide in the incubation medium. The value of the

Michaelis constant KM for tolbutamide, even when the

unbound concentration of the substrate is considered,

decreased from 123 lM without HSA to 73 lM in thepresence of HSA at a concentration of 5 mg/mL. In this

study it was concluded that the addition of HSA to

microsomal incubation media may yield enzyme kinetic

estimates more comparable with in vivo results than

studies carried out in its absence. A similar study on the

effect of albumin on the phenytoin hydroxylation catalyzed

by human liver microsomes has been performed by Rowland

et al. [29]. In this study, both, either bovine serum albumin

or essentially fatty acid-free HSA reduced the KM values

(based on unbound substrate concentration) with only a

minor effect on the catalytic rate constant. A discussion is

also made on the albumin effect in the reaction under study

and in vitroin vivo extrapolations.

At relatively high concentrations of albumin the space

filling or crowding effect must be also considered

[9, 26]. This effect could alter the enzyme conformation

leading to changes in its catalytic activity. Bergman and

Winzor [9] studied the effect of albumin on the reduction

of pyruvate by rabbit muscle lactate dehydrogenase. They

found that in the presence of albumin the catalysis was

enhanced as a consequence of an increase in the catalytic

rate constant with no appreciable effect on the Michaelis

constant for pyruvate or its co-factor. These authors

explained the results in terms of the crowding effect

which provokes a change in the conformation of the

enzyme. Olsen [26] studied the kinetic properties of

hexokinase in concentrated protein solutions (BSA), (up to

4 mM). All results could be accounted for by a Michaelis

Mentens approach and both, KM and kcat decreased with

increasing albumin concentration. The decrease in KM with

increasing protein concentration was ascribed to an

increase in the ratio of activity coefficients between the

native enzyme and the enzymesubstrate complex. The

decrease in kcat with increasing protein concentrations was

also explained in terms of the crowding effect, which in

this case leads to conformational changes of the enzymes

that disfavor the catalytic step. Recently, we have reported

that human serum albumin (up to 0.2 mM) do not affect the

kinetic parameters ((KM) and kcat) of the reaction of

hydrolysis of N-glutaryl-L-phenylalanine p-nitroanilide

(GPNA) catalyzed by a-chymotrypsin when they are cal-culated in terms of the unbound substrate concentration [3].

In the present work we report the results of a study on

the effect of human serum albumin on the hydrolysis of the

synthetic substrate ((NAG)3-MUF); Scheme 1) catalyzed

by egg white lysozyme. The possible occurrence of HSA-

substrate and HSAlysozyme interactions have also been

investigated, as well as the crowding effect, were taken

into consideration.

2 Experimental Section

2.1 Chemicals and Equipments

Hen egg white lysozyme (Sigma), human serum albumin

(HSA), essentially fatty acid free (Sigma), 4-methylumbel-

liferyl-b-D-N-N0-N00 triacetylchitotrioside ((NAG)3-MUF)(Sigma), 4-methylumbelliferone (Sigma) and dansyl deriv-

atives, dansylsarcosine (DS) and dansylsulfonamide (DNSA)

(Sigma) were used as received. Ultrapure water obtained from a

Modulab Type II equipment was employed to prepare all the

solutions.

Scheme 1 Structure of 4-methylumbelliferyl-b-D-N-N0-N00 triacetyl-chitotrioside ((NAG)3-MUF)

368 C. Calderon et al.

123

Absorption spectra and absorbances were recorded in a

HewlettPackard UVvisible 8453 spectrometer. Fluores-

cence measurements were recorded in an Aminco-Bowman

fluorescence spectrometer. Light scattering measurements

were done using a Zetasizer Nano S-590 (Malvern Instru-

ments). Microcentrifugations were carried out using Mil-

lipore Ultrafree-MC centrifugal filter units (molecular

weight cut off equal to 30 kDa).

2.2 Lysozyme Activity Assays

The enzyme activity was measured at pH 5.2 (50 mM

citrate buffer) and 37 C, employing the synthetic substrate((NAG)3-MUF). The reaction was stopped at different

times by the addition of glycine buffer (pH 11). The rate of

the catalyzed reaction was determined by following the

formation of the fluorescent product 4-methylumbellyfer-

one (kex = 360 nm; kem = 445 nm) as a function of time.The amount of product liberated was determined using a

linear calibration curve.

2.3 Dansyl Derivatives Displacement Experiments

The possible association of (NAG)3-MUF with HSA was

evaluated by the microcentrifugation experiments (see

Sect. 2.4) an from displacement of dansyl derivatives

(dansylsulfonamide and dansylsarcosine) which are spe-

cific markers of the main binding sites present in HSA

(sites I and II, respectively). Dansyl derivatives are highly

fluorescent when they are incorporated in the hydrophobic

sites of HSA and are almost non-fluorescent in an aqueous

environment. The possible displacement of the probes was

then determined by registering the fluorescence intensity

(kex = 360 nm; kem = 475 nm) of solutions of HSAincubated in the presence of the considered dansyl deriv-

ative, before and after addition of (NAG)3-MUF.

2.4 Microcentrifugation Experiments

Microcentrifugation experiments were carried out at 37 C,pH 5.2 (50 mM citrate buffer), using Millipore Ultrafree-

MC centrifugal filter units. The molecular weight cut off of

the filter employed was 30,000 Dalton. Calibrations with

respect to filtrate volume passed through the membrane

were made as a function of the centrifugation time and

speed. Control measurements were carried out to ensure

that the molecule being monitored (the substrate or the

enzyme) was not retained by the membrane.

To examine the modes of interaction between HSA and

lysozyme the HEX v.6.3 [28] molecular docking program

was used. The structures 1E78 and 1UCO, both obtained

from Protein data Bank, were used for HSA and Lysozyme,

respectively [5, 25]. The protein surface shape and

electrostatic charges were modeled using 3D spherical

polar basis functions. This allows performing global rota-

tional and translational space scans by using fast Fourier

transformations. Structures of 500 complexes HSALyso-

zyme were ranked by energy criteria according to steric

and electrostatic surface interactions.

2.5 Molecular Dynamic Calculations

To evaluate the proteinprotein interaction as a function of

time, the five HSALysozyme complexes with the lowest

energies obtained from molecular docking were optimized

by molecular dynamics. Each HSALysozyme complex

was solvated with 22,853 water molecules and six sodium

ions to neutralize the charge, and introduced in a box

10.0 9 8.0 9 10.0 nm3 with periodic boundary conditions.

The simulations were performed using the software pack-

age GROMACS v 4.0 [8, 19, 22, 31]. For the visualization

of molecular graphics and trajectories, the program VMD

[21] was employed. HSA and lysozyme were represented

using the GROMOS96 43a2 force field [30], with SPC

water [7]. LINCS [18] was used to constrain the bond

lengths of the proteins and SETTLE [23] to restrict the

structure of the water molecules. A 1 nm cut-off was used

for the Lennard-Jones potential and the real space elec-

trostatic interactions. Long range electrostatic interactions

were calculated using PME [15]. The neighbor list was

updated every 10 time steps. All simulations were per-

formed in an isothermic-isobaric ensemble. To maintain

the temperature and pressure constant at 300 K, and 1 bar,

the weak coupling algorithm of Berendsen [6] was used,

with time constants of 0.1 and 1 ps, respectively. The time

step size in all simulations was 3 fs.

3 Results and Discussion

Figure 1a shows the results obtained for the effect of HSA

on the rate of reaction versus (NAG)3-MUF profiles. The

data shown in this figure indicate that, at all substrate

concentrations considered, the presence of HSA decreases

the rate of the lysozyme catalyzed process and that, the

decrease depends on HSA concentration. An important

decrease in activity is observed at low concentrations of

HSA (20 lM), reaching an almost complete inactivation ofthe enzyme at 100 lM HSA.

The curves shown in Fig. 1a correspond to the fitting of

the data to the MichaleisMenten equation, from which the

values of the Michaelis constant (KM) and the catalytic rate

constant (kcat) were obtained. The results obtained show that

KM remains almost unmodified in the presence of albumin

(1 9 10-5 2.5 9 10-6 M-1 in the absence of albumin

and 8.0 9 10-6 1.2 9 10-6 at [HSA] = 100 lM) while

Effect of Human Serum Albumin on the Kinetics 369

123

(kcat) decreases with albumin concentration as shown in

Fig. 1b (by a factor ca. 10).

The possible interaction of the substrate with albumin

was assessed by microcentrifugation experiments. Results

obtained are shown in Fig. 2a. As can be seen in this figure,

the concentration of (NAG)3-MUF in the filtrand and the

filtrate remains almost the same, irrespective of HSA

concentration, indicating a lack of adsorption of the sub-

strate onto albumin.

In line with this result, dansyl derivatives displacement

experiments (Fig. 2b) shows no incorporation of (NAG)3-

MUF to the main binding sites present on HSA. It clearly

turns out the absence of substrate binding to albumin.

These results are quite satisfactory with the insensitivity of

KM upon the concentration of albumin and clearly indicate

that the effect of albumin on the activity of lysozyme is not

related to a deprivation of the substrate to the enzyme in

the presence of HSA.

The possible interaction between HSA and lysozyme is

an interesting alternative worth to be explored in order to

interpret the origin of the lysozyme inactivation by HSA.

Microcentrifugation experiments (Fig. 3) show a decrease

in the concentration of free lysozyme with the increase of

albumin concentration (being the enzyme almost totally

associated at 100 lM). The formation of a HSAlysozymecomplex could induce conformational changes leading to

distortion of the active site of the enzyme leading a less

active conformation.

Fig. 1 a Effect of HSA on the reaction rate for the hydrolysis of(NAG)3-MUF catalyzed by lysozyme (2 lM). Kinetic data fittedaccording to the MichaelisMenten equation. No HSA (filled circle);20 lM (filled square); 50 lM (filled triangle); 80 lM (filled invertedtriangle); 100 lM (filled diamond). b Dependence of the catalytic rateconstant with the concentration of HSA for the hydrolysis of (NAG)3-

MUF catalyzed by hen egg white lysozyme

Fig. 2 a Concentration of ((NAG)3-MUF) in the filtrand (black bars)and the filtrate (white bars), measured in microcentrifugationexperiments. b Effect of the addition of ((NAG)3-MUF) on thefluorescence intensity of dansyl derivatives. ([Dansyl deriva-

tive] = 20 lM, [HSA] = 2 lM); DS as probe (open square); DNSAas probe (filled circle)


123

The more relevant result of the present study is that the

decrease of free lysozyme and the catalytic rate constant

show the same dependence with albumin concentration as

shown in Fig. 3. We interpret this finding as indicative of

the presence in the system of two lysozyme populations:

the free enzyme catalyzing the reaction as in the absence of

albumin and the HSA-associated lysozyme which would be

completely inactive.

3.1 Theoretical Modeling

In order to obtain further insight on the molecular basis of the

lysozyme inactivation by HSA, theoretical modeling tools

were used. First, molecular docking calculations were per-

formed to model the formation of the complex between

lysozyme and the synthetic substrate (NAG)3-MUF (Fig. 4).

The modeling shows the incorporation of the substrate to the

enzyme across a cleft that runs through the enzyme molecule,

cleft in which the catalytic active site is located.

The HSAlysozyme complexes, obtained by molecular

dynamics simulations, do not alter significantly their struc-

tures after 50 ns. MD energy profile indicates that the

interaction proteinprotein is mainly electrostatic. Figure 5

shows the HSAlysozyme complex with the lowest potential

energy after 49.5 ns of simulation. The structure of this

complex shows that the Glu35 and Asp52 residues located in

the active site of lysozyme are oriented toward the HSA

surface. This conformation would restrict the binding of the

substrate to the active site of lysozyme, leading to an inactive

enzyme population, in line with the experimental results.

Molecular docking calculations were performed

between (NAG)3-MUF and the solution for the HSA/

Lysozime complex obtained by molecular dynamics sim-

ulations. The results obtained (Fig. 6) demonstrate that the

incorporation of the substrate to the active site of the

enzyme is hindered by formation of the HSA/Lysozyme

complex, mainly due to an occlusion of the cleft in which

the active site is located.

Finally, to take into account the possibility of the

operation of a crowding effect, we calculated the volume

0 20 40 60 80 1000,0

0,5

1,0

1,5

2,0

[HSA] (M)

[Lys

ozym

e] (

M)

0,0

3,0x10-4

6,0x10-4

9,0x10-4

1,2x10-3

1,5x10-3

1,8x10-3

kC

AT ( s

-1)

Fig. 3 Comparison between free lysozyme concentration (filledcircle) and the decrease on kcat (filled square) as a function of HSAconcentration

Fig. 4 Complex formation between lysozyme and the syntheticsubstrate (NAG)3-MUF, obtained by molecular docking calculations.

Front view (left panel); side view (right panel). The aminoacidicresidues Glu35 and Asp52 present in the active site of lysozyme are

shown as space filling models

Fig. 5 HSA (green)/Lysozyme(orange) complex obtained bymolecular dynamic simulations.

Glu35 (purple) and Asp52(blue) located at the active siteof lysozyme are represented

indicating their van der Waals

radii. For simplicity water

molecules and ions were

removed from the visualization.

Side view (left panel); upperview (right panel) (Color figureonline)


123

occupied by albumin molecules. To do this calculation of

the hydrodynamic radius of HSA (6 nm approx.) was

determined employing dynamic light scattering measure-

ments (Fig. 7). The results obtained shows that even at the

highest albumin concentration used in the lysozyme

activity studies (100 lM HSA), the volume occupied bythe albumin molecules slightly exceeds 1% of the total

solution volume. This is by far lower than those presented

in several works on the subject were macromolecular

crowding effect is operative. Furthermore, the distances

between albumin molecules are 34 times larger than the

lysozyme molecules diameter (approx. 3 nm), ruling out a

physical effect on the enzyme structure due to an excluded

volume effect. According with these results, the effect of

macromolecular crowding can be neglected under the

conditions employed in the precedent work.

4 Conclusions

The presence of HSA (up to 100 lM) leads to a decrease inthe rate of hydrolysis of 4-methylumbelliferyl-b-D-N-N0-N00

triacetylchitotrioside catalyzed by hen egg white lysozyme.

The reaction follows a MichaelisMenten mechanism under

all the conditions employed. The decrease in the rate of

hydrolysis is related to a decrease in the catalytic rate

constant (kcat) with albumin concentration rather than a

modification of the Michaelis constant (KM). The decrease

of kcat arises from the occurrence of HSAlysozyme inter-

actions leading to an inactivation of the enzyme, so

decreasing the concentration of active enzyme available for

catalysis. These results are relevant, since as far as we

know, the effect on the kinetic parameters of complex for-

mation between albumin and enzymes has not been

explicitly demonstrated. No depletion of the substrate by

albumin or macromolecular crowding effects is observed.

Acknowledgments Thanks are given to Dicyt (USACH) andFondecyt (Grant No. 1095036) for financial support. C. Calderon

acknowledges CONICYT for a doctoral fellowship.

References

1. Abuin E, Aspee A, Lissi E, Leon L (2007) J Chil Chem Soc

52:11961197

2. Abuin E, Calderon C, Lissi E (2008) J Photochem Photobiol A

Chem 195:295300

3. Abuin E, Lissi E, Ahumada M, Calderon C (2011) Protein J

30:143147

4. AwadElkarim A, Means GE (1988) Comp Biochem Physiol

91B:262272

5. Bhattacharya AA, Curry S, Franks NP (2000) J Biol Chem

275:3873138738

6. Berendsen HJC, Postma JPM, van Gunsteren WF, Dinola A,

Haak JR (1984) J. Chem Phys 81:36843690

7. Berendsen HJC, Postma JPM, van Gunsteren WE, Hermans J

(1981) Interaction models for water in relation to protein

hydration. In: Pullman B (ed) Intermolecular Forces. Reidel,

Dordrecht, pp 331342

Fig. 6 Interaction of thesynthetic substrate (NAG)3-

MUF with the HSA/Lysozyme

complex obtained by molecular

docking calculations. Global

view of the complex (leftpanel); Zoomed view of thecomplex, showing the

surroundings of the occluded

active site of the enzyme (rightpanel)

0,1 1 10 1000

5

10

15

20

25

30

35

Inte

nsi

ty (

%)

Diameter (nm)

Fig. 7 Size distribution by intensity obtained by dynamic lightscattering measurements. HSA (black bars); lysozyme (white bars).The measurements of each protein were preformed independently, but

for simplicity the distributions are shown in the same graph


123

8. Berendsen HJC, van der Spoel D, van Drunen R (1995) Comp

Phys Comm 91:4356

9. Bergman DA, Winzor DJ (1989) Eur J Biochem 185:9197

10. Bordbar A-K, Taheri-Kafrani A (2007) Colloid Surf B Bioin-

terfaces 55:8489

11. Carter D, Ho JX (1994) Adv Protein Chem 45:153203

12. de Wolf FA, Brett GM (2000) Pharm Review 52:207236

13. Daz X, Abuin E, Lissi E (2003) J Photochem Photobiol A Chem

155:157162

14. Epps DE, Raub TJ, Caiolfa V, Chiari A, Zamai M (1999) J Pharm

Pharmacol 51:4148

15. Essman U, Perela L, Berkowitz ML, Darden T, Lee H, Pedersen

LG (1995) J Chem Phys 103:85778592

16. Gelamo EL, Silva CHTP, Imasato H, Tabak M (2002) Biochim

Biophys Acta 1594:8499

17. Hajduk PJ, Mendoza R, Petros AM, Huth JR, Bures M, Fesik SW,

Martin YC (2003) J Comput Aided Mol Des 17:93102 (and

references cited therein)

18. Hess B, Bekker H, Berendsen HJC, Fraaije JGEM (1997)

J Comput Chem 18:14631472

19. Hess B, Kutzner C, van der Spoel D, Lindahl E (2008) J Chem

Theory Comput 4:435447

20. Hongwei Z, Min G, Zhaoxia Z, Wenfeng W, Guozhong W (2006)

Spectrochim Acta Part A 65:811817

21. Humphrey W, Dalke A, Schulten K (1996) J. Mol Graph 14:3338

22. Lindahl E, Hess B, Van der Spoel E (2001) J Mol Model

7:306317

23. Miyamoto S, Kollman PA (1992) J Comput Chem 13:952962

24. Moosavi-Movahedi Z, Safarian S, Zahedi M, Sadeghi M, Saboury

AA, Chamani J, Bahrami H, Ashraf-Modarres A, Moosavi-

Movahedi AA (2006) Protein J. 25:193201

25. Nagendra HG, Sudarsanakumar C, Vijayan M (1996) Acta

Crystall Sect D52:10671074

26. Olsen SN (2006) Thermochim Acta 448:1218

27. Peters TJ (1995) All about albumin: biochemistry, genetics and

medical applications. Academic Press, New York

28. Ritchie DW, Venkatraman V (2010) Bioinformatics 26:

23982405

29. Rowland A, Elliot DJ, Knights KM, Mackenzie PI, Miners JO

(2008) Drug Metab Dispos 36:870877

30. Scott WRP, Hunenberger PH, Tironi IG, Mark AE, Billeter SR,

Fennen J, Torda JAE, Huber T, Kruger P, van Gunsteren WF

(1999) J Phys Chem A 103:35963607

31. Van der Spoel D, Lindahl E, Hess B, Groenhof G, Mark AE,

Berendsen H (2005) J Comput Chem 26:17011718

32. Wang C-S, Bass H, Whitmer R, McConathy WJ (1993) J Lipid

Res 34:20912095

33. Wang J-S, Wen X, Backman JT, Neuvonen PJ (2002) J Phar-

macol Exp Ther 302:4349


123

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
c.10930_2011_Article_9339.pdfEffect of Human Serum Albumin on the Kinetics of 4-Methylumbelliferyl- beta -D-N-Nvprime-NPrime Triacetylchitotrioside Hydrolysis Catalyzed by Hen Egg White LysozymeAbstractIntroductionExperimental SectionChemicals and EquipmentsLysozyme Activity AssaysDansyl Derivatives Displacement ExperimentsMicrocentrifugation ExperimentsMolecular Dynamic CalculationsResults and DiscussionTheoretical ModelingConclusionsAcknowledgmentsReferences

Bioquimica

Documents

Transcript of Bioquimica