`
`Changes in the secondary structure of proteins labeled with 125I:
`CD spectroscopy and enzymatic activity studies
`
`Y. M. Efimova,1* B. Wierczinski,1** S. Haemers,1 A. A. van Well2
`1 Department of Radiochemistry
`2 Department of Neutron Scattering and M(cid:246)ssbauer Spectroscopy,
`Interfaculty Reactor Institute, Delft University of Technology, Mekelweg 15, 2629 JB Delft, The Netherlands
`
`(Received April 6, 2004)
`
`Bovine serum albumin (BSA) and lysozyme (LSZ) were radiolabeled with 125I. Three different methods for protein iodination with 125I were
`optimized. Parameters like incubation time and ratio of oxidizing agent and amount of protein were established. During protein iodination with
`125I, structural damages caused by the introduction of iodine into the protein may occur. These damages depend on the oxidizing agent used and
`may lead to considerable changes in the protein structure and, hence, their biological activity. Changes in secondary structure of LSZ and BSA
`were examined by circular dichroism (CD). Enzymatic activity tests were performed with lysozyme to check its biological activity. The Iodo Bead
`was found the best oxidizing agent for protein iodination.
`
`Introduction
`
`substances can be labeled by
`Many different
`radioiodination. Such labeled molecules are of major
`importance in a variety of investigations, e.g., studies of
`intermediary metabolism, quantitative measurements of
`physiologically
`active molecules
`in
`tissues
`and
`biological fluids, the adsorption kinetics and exchange
`of proteins on interfaces.
`Iodine-125 is most commonly used for iodination of
`compounds for in vitro procedures. It has a half-life of
`60 days and a (cid:1)-ray energy of 35 keV. Thus, 125I is the
`radionuclide of choice for
`radioiodination. Various
`methods have been developed for the radioidination of
`peptides and proteins.1(cid:150)5 They differ in the nature of the
`oxidizing agent
`(e.g., Chloramine-T,
`Iodogen,
`Iodo
`Beads) for converting 125I into the reactive species 125I2
`or 125I+. Those reactive species are incorporated into
`amino acid residues of the proteins. Mostly, substitution
`into tyrosine residues of the protein takes place, but
`residues, such as histidine,6
`substitution into other
`cysteine, and tryptophan may also occur.
`There are four levels of protein structure complexity.
`Primary structure is the amino-acid sequence specific for
`each protein. The shape of
`the polypeptide chain
`determines the secondary structure of the protein. There
`are four common secondary structures in proteins
`namely alpha helices, parallel or antiparallel beta sheet,
`turns, and random coil. Tertiary structure is the folding
`and twisting of secondary structures and the quaternary
`structure is defined as two or more polypeptide chains,
`which are assembled together. We assumed that changes
`in the secondary structure could be used as a probe for
`the structural stability of the protein.
`
`* E-mail: efimova@iri.tudelft.nl
`** Present address:
`Institut
`f(cid:252)r Radiochemie, Walther-Meissner-
`Strasse 3, 85748 Garching, Germany.
`
`During the process of the protein iodination by 125I
`different
`types of changes can occur in the protein
`structure.1
`(1) Damage caused by the iodination process. This is
`an important factor:
`if the labeling procedure itself
`damages the protein structure, considerable changes in
`biological activity in a variety of biological systems can
`take place. This kind of damage may depend on the
`oxidizing agent used. Therefore, we have considered
`three different oxidizing agents: Iodogen, Chloramine-T
`and Iodo Beads. It is known that Chloramine-T is a
`strong
`oxidizing
`agent
`and
`some
`proteins
`are
`the
`relatively strong oxidizing
`denaturated under
`conditions.7 Methods that use Iodogen or Iodo Beads are
`supposed to be more gentle.
`(2) Radiation damage. If very high activities of 125I
`are used for the iodination, the radiation emitted during
`the decay of 125I may cause bond breakage. Whenever
`possible, lower levels of radioactivity are used.
`(3) The introduction of an iodine atom by itself may
`have an influence on the properties of the protein.
`We have applied radiolabeling to study adsorption
`and exchange processes of proteins on solid surfaces.
`For these experiments it
`is very important
`that
`the
`structure of the labeled proteins does not change during
`the labeling process. We have optimized the labeling
`procedure for two proteins: Lysozyme (LSZ) is a (cid:147)hard(cid:148)
`globular protein with a stable secondary structure and
`shape, and bovine serum albumin (BSA) which is a
`(cid:147)soft(cid:148) globular protein and may easily change its
`internal structure. Changes of the secondary structures
`after radiolabeling were determined by using circular
`dichroism (CD)
`spectroscopy
`that measures
`the
`difference in absorbance of right- and left-circularly
`
`0236(cid:150)5731/USD 20.00
`' 2005 AkadØmiai Kiad(cid:243), Budapest
`
`AkadØmiai Kiad(cid:243), Budapest
`Springer, Dordrecht
`
`Page 1 of 6
`
`YEDA EXHIBIT NO. 2051
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`Y. M. EFIMOVA et al.: CHANGES IN THE SECONDARY STRUCTURE OF PROTEINS LABELED WITH 125I
`
`polarized light. It has been shown that the CD spectra
`can be analyzed in terms of different
`secondary
`structural
`types. A number of
`review articles are
`available
`describing
`the
`technique
`and
`its
`application.8(cid:150)12 Furthermore, the enzymatic activities of
`iodinated and native LSZ were
`compared using
`Micrococcus lysodeikticus, which is a probe of the
`tertiary structure as well.
`
`Experimental
`
`Materials
`
`Proteins BSA (A4378) and LSZ (L6876) were
`purchased from Sigma and used without
`further
`purification. Tris buffer with a concentration of 50 mM
`and pH 7.5 was prepared by mixing 50 mM of Tris-
`(hydroxymethyl)-aminomethane and 50 mM of HCl in
`an appropriate ratio. The buffer solution was made using
`Super Q-Millipore water and was filtrated through a
`membrane (Rotilabo-Spritzenfilter) with a pore size of
`0.22 (cid:181)m and stored at 4 (cid:176)C until use. Oxidizing agents
`Chloramine-T (C9887) and Iodogen (T0656) were
`obtained from Sigma, Iodo Beads from PIERCE (No.
`28665). Chloroform was from J. T. Baker (No. 7386).
`The 125I in a NaOH solution (370 MBq/100 (cid:181)l) was
`from Amersham Bioscience.
`PD 10 desalting columns with Sephadex G-25 as
`medium were purchased from Amersham Pharmatia
`Biotech. Micrococcus lysodeikticus has been obtained
`from Sigma (M3770). All activities were measured
`using a Wallac 1480 (cid:147)WIZARD 3(cid:148) sodium iodide
`counter. The protein concentrations were measured on a
`Shimadzu UV-1601 spectrophotometer.
`
`Labeling procedures
`
`Optimization: For fair comparison of the different
`labeling methods the optimum combination of
`the
`incubation time and the ratio between the amount of
`oxidizing agent and protein were determined.
`Iodogen and Chloramine-T methods: Both labelings
`were performed by mixing 100 (cid:181)l of the protein solution
`([protein] = 5 mg/ml in Tris buffer), 50 (cid:181)l of Na125I
`solution, and 20 (cid:181)l of Tris buffer or
`inactive NaI
`solution in a small Eppendorf cup. In the case of
`Chloramine-T labeling, 10 (cid:181)l of Chloramine-T solution
`([Chloramine-T] = 2 mg/ml) was also added to the
`mixture
`and
`the
`total
`reaction
`volume was
`approximately 180 (cid:181)l. Exact volumes were determined
`by weighing. For the Iodogen labeling a solution of
`Iodogen in chloroform ([Iodogen] = 2 mg/ml) was
`
`evaporated in the Eppendorf cup before adding the
`reaction solutions of total volume of about 170 (cid:181)l. The
`solution is allowed to stand for 30 minutes after mixing
`by gentle shaking.
`To reduce the probability of protein denaturation the
`procedure is run on ice. Finally 300 (cid:181)l of sodium
`metabisulfite solution ([Na2O5S2] = 1 mg/ml in H2O) is
`added to stop the radioiodination reaction.
`Iodo Beads method: Put one Iodo Bead into the 125I
`buffered solution (70 (cid:181)l) and let stand for 5 minutes at
`room temperature. Add 100 (cid:181)l of the protein solution
`([protein] = 5 mg/ml in Tris buffer) and incubate for 10
`minutes. The iodination process was
`stopped by
`separation of the Iodo Bead from the reaction volume,
`which was around 170 (cid:181)l.
`
`Separation from free iodide
`
`radiolabeled proteins were
`labeling the
`After
`separated from free iodine by use of a PD 10 desalting
`column. Before use the column is equilibrated with
`25 ml Tris buffer. The solution containing the free
`iodide and the labeled protein is applied onto the column
`and the mixture is eluted with Tris buffer. Fractions of
`about 0.5 ml are collected. Figure 1 shows a typical
`elution profile for the separation of free iodide and
`radiolabeled protein.
`
`Enzymatic activity test of LSZ
`
`The rate of lysis of Micrococcus lysodeikticus using
`turbidity measurements was determined.13 Native and
`radiolabeled LSZ were dissolved at a concentration of
`150(cid:150)500 units/ml with reagent grade water. Pipette
`2.9 ml of Micrococcus lysodeikticus cell suspension
`([Mic lys.] = 0.3 mg/ml in Tris buffer pH 7.5) into a
`cuvette and after adding 0.1 ml of dissolved LSZ record
`the changes in A450 (absorbance at wavelength 450 nm)
`per minute from the initial linear part of the curve. One
`unit of LSZ activity is equal to a decrease in turbidity of
`0.001 per minute at 450 nm, pH 7.5 and 25 (cid:176)C and the
`specific enzymatic activity of LSZ was calculated by:
`
`Units
`mg
`
`=
`
`1000
`(enzyme
`
`(cid:2)
`PA
`/min
`450
`mixture)
`reaction
`in
`
`mg
`
`(1)
`
`Circular dichroism (CD) spectroscopy
`
`The CD spectra were recorded on a JASCO (J-715)
`spectropolarimeter. The temperature was kept constant
`at 20–0.1 (cid:176)C using a JASCO PTC-348WI thermostat.
`The spectra were scanned and then analyzed by
`Dicroprot 2000 software.14
`
`92
`
`Page 2 of 6
`
`YEDA EXHIBIT NO. 2051
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`Y. M. EFIMOVA et al.: CHANGES IN THE SECONDARY STRUCTURE OF PROTEINS LABELED WITH 125I
`
`Fig. 1. Separation of free iodine and radioiodinated protein with a PD-10 column. Fractions of 0.5 ml were collected
`and activities of each fraction were measured
`
`Results and discussion
`
`Labeling procedure
`
`NI/NI* is the ratio of inactive and radioactive iodine. Na
`[mol(cid:150)1] is the Avogadro(cid:146)s constant. The protein mass in
`the reaction mixture is mpr [g] and the Mpr [Da] is its
`molecular weight. (MLSZ = 14307 Da, MBSA = 66430 Da).
`The (cid:147)master(cid:148) curves for BSA and LSZ using three
`different labeling methods are presented in Fig. 2. The
`curves were fit with a linear
`function with an
`approximate slope 1 which justified our assumption that
`the labeling is a first order reaction. Table 1 gives the
`exact fitting results for the curves and the average of the
`fits for the three different procedures are presented as
`solid lines in Fig. 2 for both proteins. This means that
`Iodogen, Chloramine-T and Iodo Beads methods have a
`similar
`linear dependence of an increasing labeling
`efficiency with increasing amount of iodine atoms in the
`reaction mixture.
`
`Table 1. Slopes from fitting results of master curves for each method
`and their average for LSZ and BSA (Fig. 2, solid lines)
`
`Labeling method/ protein
`Chloramine-T
`Iodo Beads
`Iodogen
`Average:
`
`LSZ
`1.02 – 0.01
`1.04 – 0.03
`1.03 – 0.01
`1.03 – 0.01
`
`BSA
`1.02 – 0.02
`1.03 – 0.01
`1.06 – 0.02
`1.06 – 0.02
`
`93
`
`the process of
`For a better understanding of
`incorporation of iodine into the protein molecules we
`have plotted (cid:147)master(cid:148) curves for each method and each
`type of protein. We define the (cid:147)master(cid:148) curve as the
`relative number of labeled protein molecules Npr*/Npr
`(labeling efficiency) versus
`the
`total number of
`radioactive and inactive iodine atoms (NI*/NI) in the
`reaction mixture. From a chemical point of view
`radioactive 125I and inactive iodine are identical.
`Therefore, to study the high iodine concentrations only a
`small part of radioactive 125I was used and the rest
`consisted of inactive iodine. The amount of labeled
`protein molecules after iodination (Npr*) and the initial
`amount of proteins (Npr) can be calculated by:
`A
`pr
`(cid:3)
`mN
`a
`M
`
`NN
`
`N
`
`pr
`
`*
`
`=
`
`(cid:1)
`
`N =
`pr
`
`pr
`
`pr
`
`I
`
`I
`
`*
`
`(2)
`
`(3)
`
`where Apr [Bq] is the activity of the proteins after the
`iodination process, (cid:3) [s(cid:150)1] is the decay constant of 125I,
`
`Page 3 of 6
`
`YEDA EXHIBIT NO. 2051
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`Y. M. EFIMOVA et al.: CHANGES IN THE SECONDARY STRUCTURE OF PROTEINS LABELED WITH 125I
`
`Fig. 2. Ratio of labeled proteins and the initial amount of protein molecules as a function of the total number of iodine atoms (inactive iodine plus
`active) in the reaction mixture (volume of reaction mixtures is given in (cid:147)Labeling procedures(cid:148)) for LSZ (a) and BSA (b); S Iodogen method, T
`Chloramine-T method, U Iodo Beads method. The relative errors of the measurements were a few percent and are not visible in these graphs. The
`solid lines are the averages of the fits for the three different procedures. The dashed lines indicate the numbers of tyrosine (the most sensitive for
`iodine) residues in the proteins
`
`94
`
`Page 4 of 6
`
`YEDA EXHIBIT NO. 2051
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`Y. M. EFIMOVA et al.: CHANGES IN THE SECONDARY STRUCTURE OF PROTEINS LABELED WITH 125I
`
`Fig. 3. CD spectra of the native proteins (S) LSZ (a) and BSA (b) compared to the corresponding spectra
`after using different labeling methods: V Chloramine-T, W Iodogen, X Iodo Beads
`
`Table 2. Secondary structures of BSA and LSZ in native state, labeled with Iodogen, Iodo Beads andChloramine-T
`
`Structure
`
`Native
`
`LSZ
`Iodogen
`
`(cid:1)-Helix, %
`(cid:2)-Sheet, %
`(cid:2)-Turns, %
`Unordered, %
`
`28*
`20
`18
`34
`
`26
`22
`18
`34
`
`Iodo
`Beads
`24
`23
`19
`34
`
`Chlora-
`mine-T
`22
`25
`19
`34
`
`Native
`
`BSA
`Iodogen
`
`67
`7
`12
`14
`
`58
`8
`13
`21
`
`Iodo
`Beads
`59
`8
`13
`20
`
`Chlora-
`mine-T
`55
`9
`13
`23
`
`* In our experience, the relative error of the experiments within the same batch is a few percent.
`
`95
`
`Page 5 of 6
`
`YEDA EXHIBIT NO. 2051
`MYLAN PHARM. v YEDA
`IPR2015-00644
`
`
`
`Y. M. EFIMOVA et al.: CHANGES IN THE SECONDARY STRUCTURE OF PROTEINS LABELED WITH 125I
`
`Table 3. Specific enzymatic activites with their standard deviations (determined from three measurements)
`of native LSZ and LSZ labeled by different methods
`
`Specific enzymatic activity,
`(cid:215)1000 units/mg
`
`Native LSZ
`15 – 1
`
`Chloramine-T
`12 – 1
`
`Iodogen
`11 – 1
`
`Iodo Beads
`15 – 1
`
`LSZ contains 3 tyrosine residues, BSA contains 19.
`The dashed lines in Fig. 2 indicate these numbers.
`Figure 2 shows that the experimentally found maxima of
`the substitution degrees of iodine atoms per protein
`molecules are scattered around these lines. It has to be
`mentioned that not only monoiodotyrosines can be
`formed but also diiodotyrosines. The relative ratios of
`mono- and di-iodotyrosines in different proteins have
`been a subject of controversy for some time.15,16 For
`example, human serum albumin (HSA)16 has 12
`tyrosine residues available for iodination. If the apparent
`substitution level is 100% up to 20–10% of the HSA
`tyrosines will be diiodinated. Assuming that BSA
`resembles HSA we have included a dotted line into Fig.
`2 which shows the correct maximum of iodine atoms per
`BSA molecule. The substitution degree of iodine atoms
`per LSZ molecule is even smaller, except
`for
`the
`Iodogen method, than the number of tyrosine groups in
`LSZ. We
`assumed that
`the probability to find
`diiodotyrosines in LSZ is quite low.
`
`Structural changes
`
`the
`that
`important
`is
`it
`As mentioned above,
`secondary structure of the labeled proteins is kept intact.
`Figure 3 shows the changes in CD spectra of LSZ and
`BSA after the use of different oxidizing agents. The
`simplest method of extracting secondary structure
`content from CD data is to assume that a spectrum is a
`linear combination of CD spectra of each contributing
`secondary structure type (e.g., (cid:147)pure(cid:148) alpha helix, (cid:147)pure(cid:148)
`beta, (cid:147)pure(cid:148) turn, etc.) weighted by its abundance in the
`polypeptide conformation. From CD spectra analysis it
`was concluded that the use of Chloramine-T results in
`some loss of the alpha-helix content in both LSZ and
`BSA. This was expected from the strong oxidizing
`agent. At the same time the beta-sheet contents do not
`change while the unordered portion slightly increases.
`Table 2 shows that Iodogen and Iodo Beads do not have
`a significant influence on the secondary structure of LSZ
`and BSA and can be considered as gentle oxidizing
`agents for protein iodination.
`
`Enzymatic activity test
`
`The effect of present of oxidizing agents on LSZ
`biological activity can be investigated by comparing the
`enzymatic activity before and after iodination. This test
`was performed for lysozyme which causes lysis of
`Micrococcus lysodeikticus. The decrease in turbidity of
`
`96
`
`after
`suspensions
`cell
`lysodeikticus
`Micrococcus
`addition of LSZ solution was measured in time. The
`specific enzymatic activities of native and labeled LSZ
`were calculated by Eq. (1). Results are presented in
`Table 3.
`A comparison of the specific enzymatic activities of
`labeled LSZ with LSZ in native state shows that labeling
`of LSZ with Iodo Bead method does not change the
`enzymatic specific activity. However,
`Iodogen and
`Chloramine-T labeling methods decrease the enzymatic
`activity of LSZ significantly.
`
`Conclusions
`
`The use of Chloramine-T, Iodogen and Iodo Beads
`as oxidizing agents in the radiolabeling of BSA and LSZ
`experiments shows the same efficiency of the methods
`and allows the production of iodinated proteins with
`different substitution levels of iodine atoms per protein
`molecule. From CD experiments it was concluded that
`Chloramine-T has an influence on the secondary
`structure of both proteins. The enzymatic activity test of
`LSZ has also revealed some changes in enzymatic
`activity of LSZ labeled with the use of Chloramine-T.
`Iodogen as oxidizing agent does not
`influence the
`secondary structure for both BSA and LSZ but it causes
`a decrease of the biological activity of LSZ as an
`enzyme. From any point of view the Iodo Bead seems to
`be the best oxidizing agent for protein iodination.
`
`References
`
`1. A. E. BOLTON, Radioiodination Techniques, 2nd ed., Amersham,
`England, 1985, p. 7.
`2. E. REGOECZI, Iodine-Labeled Plasma Proteins, Vol. I, CRC Press,
`1984, p. 35.
`3. P. R. SALACINSKI, C. MCLEAN, J. E. C. SYKES, Anal. Biochem.,
`117 (1981) 136.
`4. G. S. DAVID, R. A. REISFELD, Biochemistry, 13 (1974) 1014.
`5. J. M. WALKER, Methods in Molecular Biology, Vol. 32,
`J.M. Walker, Human Press, Tonowa, NJ, 1994, p. 441.
`6. C. H. LI, J. Am. Chem. Soc., (1944) 66.
`7. W. M. HUNTER, F. C. GREENWOOD, Nature, 194 (1962) 495.
`8. N. SREERAMA, R. W. WOODY, J. Molec. Biol., 242 (1994) 497.
`9. P. MANALAVAN, W. C. JOHNSON, Anal. Biochem., 167 (1987) 76.
`10. R. W. WOODY, Eur. Biophys. J., 23 (1994) 253.
`11. W. C. JOHNSON, Proteins, 7 (1990) 205.
`12. S. W. PROVENCHER, J. GLOCKNER, Biochemistry, 20 (1981) 33.
`13. D. SHUGAR, Biochim. Biophys. Acta, 8 (1952) 302.
`14. G. DELEAGE, program Dicroprot 2000 http://dicroprot-pbil.ibcp.fr
`15. B. K. SEON, O. A. ROHOLT, Biochim. Biophys. Acta, 221 (1970) 114.
`16. R. F. PENNISI, Biochim. Biophys. Acta, 213 (1967) 486.
`
`Page 6 of 6
`
`YEDA EXHIBIT NO. 2051
`MYLAN PHARM. v YEDA
`IPR2015-00644