Graduate School of Pharmaceutical Sciences, Kyushu University

九州大学学術情報リポジトリ

Kyushu University Institutional Repository

リゾチームの微細な構造変化の解析 : モノクローナ ル抗体と高分解能NMRを用いて

阿部, 義人

Graduate School of Pharmaceutical Sciences, Kyushu University

https://doi.org/10.11501/3110887

出版情報：Kyushu University, 1995, 博士（薬学）, 課程博士 バージョン：

権利関係：

CHAPTER IV

Reduction of Disulfide Bonds in Proteins by 2-Am i noth iophe no I under Weakly Acidic Conditions

ABSTRACT

1\. method for reducing disulfide bonds in proteins under weakly acidic

conditions by use of 2-aminothiophcnol was developed. The disulfide bonds in I ·b 1 1\. _{and soybean trypsin inhibitor were} hen egg-white ysozyme, ⁿ onuc ease '

quantitatively re d uced by 2-aminothiophenol In phosphate buffer, pH 6, I ^{· · o} 8 _{M Gd11-HCI} I mM EDTA and 20% ethanol, for 60 min at 40 oc. _On

con aimnb ,

analysis of the recerse-phas c HPLC patterns of tryptic peptides, which were derived from reduced and S _-alkylated lysozyme and ribonuclease A at pH 6, it was confirmed that no side reaction occurred. Moreover, the reduction und_er weakly acidic conditions was demonstrated to be applicable for the location of such a labile residue as 0 -acetylated tyrosine.

INTRODUCTION

In this CHAPTER, applied 2-aminothiophenol as a reducing reagent under the acidic pH region. Methods for reducing disulfide bonds in proteins

I _I (c _{f. ld t} I 1963.) _or dithiothreitol

by use of 2-mercaptoet 1nno rest 1e ^e a.,

(Konigsberg 1972) have a rca y I d b een es a t bl. IS h ed . Bc)th these reductions of disulfide bonds in proteins and subsequent alkylation of the liberated thiols have been carried out under weakly alkaline conditions. However, some posttranslationally modified amino acid residues, for example, E- m e t hY 1 arginine (Kim & _Paik, _{1970) and} y-carboxymethyl ester of glutamic acid (Tkahashi et a/., 1967), _{are easily} hydrolyzed to arginine and glutamic acid at pH 9-9.5, re. pectively. And then. the deamidation of Gln and Asn was occurred easier in alkaline condition (see CHAPTER Y). Since these chemical bonds are

more stable in a low pH region than a high pH one, it would he convcnienr to develop a method for reducing disulfide bonds in n l^owe r pH region in order to elucidate the locati^on of the modified site^.

In _CHAPTER IV, a method for reducing disulfide bonds in proteins under weakly acidic conditions by use of 2-aminothiophenol was described.

MATERIALS AND METHODS M at eri als.

Bovine pancreatic ribonuclease A (RNase A) and soybean trypsin inhibitor (STI) were purchased from Sigma. Bio-Gel P-4 was obtained from Bio-Rad Laboratories. 2-Aminothiophenol and maleimide were purchased from Nakalai Tesque Chemical Ltd. (Kyoto). After 2-aminothiophenol had been unsealed, _it was resealed in several

tubes after being degassed by aspiration. Trifluoroethanethiol was purchased from Aldrich. L-Cysteine monohydrochloride monohydr ate was purchased from Wako Pure Chemical Industry (Osaka). All other chemicals were of analytical grade for biochemical use.

Analytical Method.

Amino acid analysis was performed with a Hitachi L-8500 after rapid- vapor-phase acid hydrolysis of a peptide sample in 6 N _{HCl containing} 0.1%

phenol at I 50 °C for h. Peptide mapping was performed according to the method described in CHAPTER 11.

Determination of the Amou11t of 2-Aminothiophello l During In c ubat ion.

2-Aminothiophenol (80 �tl) was dissolved in 2 ml of 0.1 M phosphate buffer (pH 6.0) containing 8 M Gdn-HCI and 20% EtOH, and then incubated at 40

°C under nitrogen. Aliquots were taken at appropriate time intervals and analyzed by reverse-phase HPLC. The elution conditions were the same as the

d. · Th perce11t"< ge o f the amount of 2- peptide mapping con 1t1on. e ^ll

· · b t. t. ·1st the initial level aminothiophenol after an appropnate 1ncu a 10n 1me aga11.

was determined on the basis of the peak height of 2-aminothiophenol.

Prepara tion of S-Succinimido Cysteine and S-( 1 ,2-Dicarboxyethyl) Cysteine.

s -Succinimido cysteine was prepared as follows. L-CYs t e i n e monohydrochloride monohydrate ( 165 mg, mmol) was dissolved in 0.5 ml of 0.1 M phosphate buffer (pH 6.0) containing 8 M Gdn-HCI, I mM EDT/\ and 20%

EtOH, :1nd then incubated with maleimide (97 mg, I mmol) at room temperature for 30 min. s _-( 1 ,2-Dicarboxyethyl) cysteine was prepared by acid hydrolysis of s -succinimido cysteine. These samples were purified by reverse-phase HPLC, as described above, and I yophi I ized.

f I _C'olor Factors of .S-SzLccz'nimido Cysteine and S- Determina tion o t 1e

( 1 ,2-Dicarboxyethyl) Cysteine.

The color factors of S-succinimido cysteine and S- ( I ,2 -dicarboxyethyl) cysteine were determined. with a Hitachi L-8500 amino acid analyzer, to be 92%

nnd 67% relative to that of alanine, respectively, by comparison with the peak area of the weighted sample.

Prep a r a t i _{o n of} 0-_{A c e t y} I _{a ted Ty r o s} i _{11 e.}

0 -Acetylation of N -acetyl tyrosine ethylester with N-acetylimidazole was

carried out according to the method of Riordan & Vallee (I 965). Briefly, 20 mg of N -acetyl tyrosine ethylester was dissolved in 5 ml of 0.0 l M Tris-HCI buffer (pH 7.5). The solution was mixed with a 100-fold molar excess of solid N­

acctylimidazole and then vigorously stirred at 25 oc for 60 min. The resulting 0 -acetylated tyrosine was purified by reverse-phase HPLC, as described above, and lyophilized.

RESULTS AND DISCUSSION

Reduc tion of Lysozyme with 2-Aminothiophenol.

Three milligrams of lysozyme (0.21 !111101) was reduced with 2- aminothiophenol and then S-alkylated with maleimide. After acid hydrolysis of reduced and S-succinimidate<.l lysozyme, the number of S-succinimido cysteines was determined. Under the acid hydrolysis conditions used S- s u c c i n i m i d o cysteine was converted t o S-( I ,2-dicarb oxyethyl) c ysteine. S-(1,2 - Dicarboxyethyl) cysteine could be detected between S-carboxymethyl cysteine and aspartic acid, with a Hitachi L-8500 amino acid analyzer, under the routine conditions (Fig. lV-1). The number of S-(1,2-dicarboxyethyl) cysteines increased with incubation time up to 60 min and then gradually decreased after 60 min (Fig. IV-2) (the reason will be discussed later). The reduction of disulfide bonds in lysozyme was found to be completed in 60 min, at which time the number of S-( I ,2-dicarboxyethyl) cysteines almost reached the maximum (theoretically 8.0). In Table IV -I, the amino acid compositions of reduced nnd S -succinimidated lysozyme are shown. Then, the method was applied to reduce the disulfide bonds of ribonuclease A (RNase A) and soybean trypsin inhibitor (STI). The amino acid compositions of these reduced proteins are also shown in Table IV-1. The results clearly indicated that the disulfide bonds 111 these proteins were also quantitatively reduced.

As for alkylation of liberated thiol groups, N -ethy I male imide, a typic a!

blocking reagent for cysteine residues under weakly acidic conditions (Feeney

& Whitaker, 1987), was not suitable for blocking cysteine residue under the

reduction conditions employed here due to the considerable precipitation. It was impossible to separate the S-alkylated protein quantitatively from the crystalline precipitate of the reagent. On the other hand, monoiodeacetamide or monoiodoacetic acid was less reactive with cysteine residues than makimide under the conditions employed here. Therefore, we chose maleimide for

· res1'dues under the conditions, while S-succinimidation of blocking cysteine

cysteine residues produced diastereomers (Fig .. IV -3).

t .

tl ^{tl c}

�i

^{'t: �}

� ;..,

^·�

�� ·�

�!

^f3

i �

^'I;!

!J

^�

-

J; ^·�

_CJ

]

=

"'Q)

I

�

�

�

^Q

�

�

.s

a

e

^{0.. �}

^� !\ ^A

v\...

_I

0 �

"0 .9

:§ �

.8 ^;...

!

^CJ

0

30

Retention time (min)

Fig. IV_ 1. Chromatography of amino acids with a Hitachi

L-8500

amino acid analyzer under the routine conditions.

�

10

� 0 .J:J

J,..; • ..

...

�

� I

N ^"

5

� '-' � _I =

� ...

� � 0 .... _trJ J,..; �

� u .J:J .-.,

=�

_:=..c:

₀

^{I I}

z� 0 60 120

Incubation time (min)

Fig. IV -2. Time course of the reduction of disulfide bonds of lysozyme (0.21 �mol) with 2-aminothiophenol

(0.36

M) at

40

°C in

0.1

M phosphate buffer

(pH 6.0)

containing I mM EDT A, 8 M Gdn^-

H

CI and 20% ethanol.

0

-NHCHC-II

CH2 1

SH I

0 0

-

II II

-NHCHC- -NHCHC-

0 N _{I I}

H CH2 CH2

---1·� ^{I I}

pH 6.0

rl

^{or ,S}

�

^NH

� f:io

H

S -succinimindo cysteine

0

-NHCHC-II

acid hydrolysis ..

CH2 1 I

0

;:-\

^{OH OH 0}

S -(1 ,2-dicarboxyethyl) cysteine

Fig. IV -3. Reaction of cysteine with maleimide and the acid hydrolyzed

product.

A

��I\

0

Table IV-1. Amino acid comEosition of S-malcimidated Eroteins. n

Lysozyme RNase A STI

� c- Mf

_{T c}

MI

_T c MI Asp 21 20.2 20.0 26 26.2 26.9 15 15.0 15.4 Thr 7 6.5 6.5 7 6.1 6.8 10 9.2 9.9 Ser 10 8.8 9.2 11 10.1 10.8 15 13.0 13.9 Glu 5 5.3 5.7 18 20.6 21.3 12 12.8 13.4 Gly 12 12.0 12.5 16 15.8 16.1 3 3.5 3.5 Ala _{12 12.0 12.0 8 8.0 8.0 12 12.0 12.0} Va_{l 6 5.3 5.3 14 10.4 10.4 9 8.3 8.6} Met 2 2.0 2.0 2 2.7 2.2 4 3.9 3.8 Ilc 6 5.1 5.4 15 12.6 12.8 3 2.2 2.3 Leu 8 _{7.9 8.0} 14 14.5 14.5 2 2.7 2.7 Tyr 3 _{3.1 3.2 4 4.5 4.2 6 5.8 5.8} Phe 3 2.8 2.9 9 8.4 9.0 3 3.0 2.8 Lys 6 5.7 5.8 2 2.5 2.0 10 10.3 9.9 His 1 _{1.0 1.1} 9 _{8.4 9.0 4 3.7 3.4} Arg _{10 10.1 10.7 10 10.3 10.8 4 4.1 4.1} MI

C

_{8 8.0 4 3.7 8 8.3}

n All values arc expressed as molar rations normalized to a value of _alanine. Reduction of proteins in the presence of b excess of 2-aminothiophenol for 60 min at pH 6.0 and 40 °C.

c Theory.

0 Control.

c S -Maleimidated _proteins.

The number of S-maleimidatcd cystcines.

0 0 _c

� ^"

..c 0. 0 ..c 0. 0

:.c :s

0 _c ⁰ c:

E _..

-�

,\ �\

]

..,

'>l 0

c

^... "

... "

i E·�·

>.. t: "

�-

� -�

t ^��

� .. _v

i. •

'-- \..--- � -

60 120

0�---:::60:---__; 120 0 Retention time (min)

60

Fig.

IV-�.

_!Ypical

rcv�rsc-phase HPL�

patterns for (A) oxidation of 2-aminothiophenol; (B) purification ofS-s�cctn11111do cystemc and S-( I ,2-dlcarboxyethyl) cysteine; and (C) hydrolysis of O,N -diacetyl tyrosmc cthylcstcr. The column (Wakopak 5C 18, 4 x 250 mm) was eluted with a gradient formed from

50 ml of I% acetonitrile and 50 ml of 50% acetonitrile, both containing 0.1% HCI, at the flow rate of 0.8 ml/min.

!21

In order to determine why the number of S-( I ,2- d i c a rbo x y e t h y l ) cysteines decreased after 60 min in Pig. IV-2, the amount or 2-aminothiophenol in the reduction solution wns determined by reverse-phase HPLC. A typical reverse-phase HPLC pattern of 2-aminothiophenol after proper incubation in the reduction solution is shown in Pig. IV-4A. /\s shown in Fig. lV-5, the amount of 2-aminothiophenol gradually decreased after 60 min. This suggested that the decrease in the number of S -(I ,2-dicarboxyethyl) cysteines after 60 min in Fig.

IV-I mny be due to nir oxidation of 2-aminothiophcnol itself by unremoved oxygen. This was consistent with the previous report that an aromatic thiol group was much more easily oxidized than an aliphatic thiol group ( Kharasch

et al., 1951). ^{However, the} concentration of 2-aminothiophenol in the reduction solution could not be increased due to its low solubility. Therefore, 60 min-reduction time was suitable for reduction of the disulfide bonds in proteins by use of 2-aminothiophenol under the present conditions.

100

....--, ^•

� 0 Q Q)

.c:: �

50

:a

_... 0

.5

0

<

s

_I

N

0

0 60 120 180

Incubation time (min)

Fig.

IV

_-5. Air oxidation of 2-aminothiophenol. The percentage of the amount of 2-arninothiophenol against the initial level (0.36

M)

was determined after incubation for appropriate times at 40 °C in a 0.1

M

phosphate buffer solution (pH 6.0) containing

8 M

Gdn-HCl and 20% ethanol.

Tryptic Hydrolysis and Peptide Mapping of Reduced and S-

Succinimido Lysozyme.

The elution pattern of tryptic peptides of 2-aminothiophenol reduced and S-succinimidated lysozyme at pH 6.0 on reverse-phase HPLC is shown in Fig. IV- 6B. For companson, that of 2-mercaptoethanol reduced and S - c a r b o x y- methylated lysozyme at pH 8.6 is also shown in Fig. IY-6A. These tryptic pcptidcs were assigned by amino acid analyses and the results are added to Fig.

IV-6A. The peptides including cysteine residues (for example, T'J7+ 18, T'3, T'9+ 1 Q. T'g, T'JS+ 16. T'6 and T'J 1) in Fig. lV-68 were split into two or more peak fractions since diastereomcrs were produced on S-maleimidation of cysteine residues (f-'ig. IV-3), while these peptides were recovered quantitatively on amino acid analysis (data not shown). Moreover, except for these peptides, there was no significant difference in the pattern between Fig. 1V-6A and B, indicating that no side reaction occurred. This was supported by the results of amino acid analysis of S -succinimido lysozyme (see Table IV -1 ). Although the amount of tryptophan could not be determined by amino acid analysis under the acid hydrolysis conditions, it is clear that tryptophan residues did not react with 2-aminothiophcnol under the present conditions on comparison of the reverse-phase HPLC patterns 1n Fig. IV-6A and B. Therefore, it was confirmed that 2-aminothiophenol was a suitable reagent for reducing disulfide bonds in lysozyme under weakly acidic conditions. fn order to test the wide applicability of this method, similar experiments were carried out on RNase A. The elution patterns on reverse-phase HPLC of tryptic peptides derived from 2- mercaptoethanol reduced and S -carboxymethylated RNase A at pH 8.6, and 2- aminothiophenol reduced and S -succinimidated RNase A at pH 6.0 are shown in

Fig. IV-6C and ^D, respectively. These tryptic peptides were also assigned by amino acid analysis and are shown in Fig. IV-6C an^d D. In the case of RNase A, there was no significant difference in the pattern between Fig. IY-6C and D, except for the peptides including cysteine residues (T' 1 Q, T' 1 2+ 1 3, T' 13, T'4+5, T'4,

A

1-o t--

1-o ., ... .!

1-o

c

..₊ .

� ₀

..:

1-o ..

r-< ..,

1-J\,) ^�

l

^1'-

J

0

!-< ...

B

!-< ...

....

�

^�·

.. ....

1-o ,.,

� ... �

60 0

Retention time (min)

::l

�

D

f-t r-< ..

"' 1-o �

�

r-< ::l

).

l w�U�

60 0

)..

Retention time (min)

60

-::l

!-<

�

-...

� -

r-< -

... �

-0 r-<

� -

r-< ::l

1 J..

�

60

Fig. IV _-6. reverse-phase HPLC of tryptic pep tides derived from (A) 2-mercaptoethanol r^edu^ce^d and S -

�

arbox!methylated lysozyme at pH 8.6 (T1+2, Lys 1-Arg 5; T3, S -carboxymethylated Cys 6-Lys 13: Ts, H

�

^{s 15-A}rg 21; T6, Gly 22-Lys 33; T11+7, Gly 22-Arg 45; T7, Phe 34-Arg 45; T8, Asn 46-Arg 61, Ty, T1p 62-Arg 68; T9+IO• Trp 62-Arg 73; T11, Asn 74-Lys 96· T11 12 Asn 74-Lys 97· T

L 97 ys -A· ^. 1g 112· , T 13• II e 98-Arg I _{12; T15+111,}

�

-car

?

oxymethylated Cys ' + 11' 5-A^rg 125; T111, Gly 117-' 12+13•

Arg 125, TI7+1R• Gly 126-Leu 129),

(B)

2-ammothwphenol reduced and S-succinimidated lysozyme at pH 6.0, (C) 2-mercaptoethanol reduced and S-carboxymethylated RNase A at pH 8 6 (T Lys 1- Lys 7;

!3·

Phe 8-Arg 10; _T4, Gin 11-Lys 31 ;T4+5, Gin 11-Arg 33; T6, Asn 34-Lys 37.; T10

:

+

J

'sn 62- Lys

�

6, Tl.l• Asn 67-Arg 85; T12+!3• Glu 86-Ly s98; T13, Thr 99-Lys 104), and

(D)

2-amtnotht?phenol r�duced andS -succinimidated RNase A at pH 6.0. Since T 11 in lysozyme contains three cystemes, T 11' ¹¹¹ B became broad due to S-succinimidation of cysteines.

and T' 1 1 ). Therefore, this rengcnt nlso successfully reduced the disulfide bonds in RNasc A.

De terminatio11 of the Color Fa ctor of S-( 1,2-Dicarboxyetlzyl) Cysteine.

Rapid-vapor-phase acid hydrolysis for determining the amounts of S­

succinimido cysteine in lysozyme. RNase A and STI was used here, because the

I · · 1 _t Tl _e ,·,llJ.de bonds in S-succinimido cysteine were

nnn ys1s t1me 1s s1or . 1

completely hydrolyzed under the conditions used (Fig. IY-3). Then, the color factors of S-succinimido cysteine and S-( I ,2-dicarboxyethyl) cysteine, and the recovered yield of S-(1,2-dicarboxyethyl) cysteine derived from S-succinimido cysteine on acid hydrolysis were tried to determine. S-Succinimido cysteine and S-(1,2-dicnrboxyethyl) cysteine were synthesized (see MATERIALS AND METHODS), and purified by reverse-phase H PLC (Fig. I Y -4 B). These color fnctors, relntive to that of alanine. were determined to be 92% and 67%, respectively. On cnlculation with use of these values, the recovered yield of S­

( 1 ,2-dicarboxyethyl) cysteine was determined to be 82% under the conditions used.

For the wide application, we determined the recovered yield of S-( 1, 2- dicarboxyethyl) cysteine derived from S-succinimido cysteine under the conventional conditions of acid hydrolysis (110 °C, 20 h). Under these conditions, S-succinimido cysteine was completely hydrolyzed, nnd the recovered yield of S -(I ,2-dicarboxyethyl) cysteine was determined to be 57%

(the rest of the product could not be detected with a Hitachi L-8500 amino acid nnnlyzer under routine conditions).

Comparison of 2-Aminothiopheno/ with Other Reductive Reagents.

The pKa of the sulfhydryl group of 2-aminothiophenol was about 2.5 (pH meter reading), as judged on pH titration in 8 M Gdn-HCl and 20% EtOH, while

those of 2-mercaptoethanol and thiophenol were 6.7 and 5.9, respectively.

Namely, the sui fhydryl group of 2-am i nothiophenol had an about 4 _{pH units} lower pKa than that of 2-mercaptoethanol, which is one of the conventional reduction reagents. On account of this lower pKa, the disulfide bonds in proteins could be reduced under weakly acidic conditions (even in 0.1 M AcOH­

NaOH buffer, pH 4, containing 8 M Gdn-HCI, 20% EtOH and I mM EDTA, data not sho wn).

The sulfhydryl group of trifluoroethanethiol was expected to have a lower pKa due to the electronegative trifluoromethyl residue. Actually, its pKa was estimated to be about 3 under the above conditions. This reagent may be useful for reducing the disulfide bonds under weakly ncidic conditions, similar to 2-aminothiophenol. Moreover, the reagent has the advantage of being miscible in the Gdn-HCl solution without ethanol. However, it has a low boiling point (34-35 °C) and is volatile, and therefore it would be hnrd to use under routine conditions.

Ad vantage of Amino thiop hen ol

the under

Reduction of Weakly Acidic

Disulfide Bonds by 2- Conditions.

The 0 -acetyl bond in 0 -acetylated tyrosine is comparntively unstable and is easily hydrolyzed at pH 9.0 (Riordan & Vallee, 1965; Nakne & Hamnguchi, 1972). When the site of 0 -acetylation In proteins with disulfide bonds is determined, this labile bond may be cleaved with conventional reduction methods. fn order to examine the utility of the reduction conditions employed here, the hydrolysis of 0 -acetylated tyrosine was examined in the presence or absence of reduction reagents at pH 6.0 or 8.6.

First, N, 0 -diacetyl tyrosine ethyl ester was prepared according to the literature by the reaction of N -acetyl tyrosine ethylester with acetylimidazole (see MATERIALS AND METHODS). Lyophilized N,O -diacetyl tyrosine ethylester (0.3 mg) was dissolved in 0.1 ml of the reduction buffer under weakly alkaline

conditions [0.584 M Tris-HCl buffer, pH 8.6, containing 8.125 M urea and 5.7>7 mM EDTA (routine reduction conditions) or 0.1 M Tris-HCl buffer, pH 8.6, containing 8 M Gdn-HCl I mM EDTA and 20% EtOH] with or without 0.18 M 2- mercaptoethanol. On the other hand, the same amount of N.O -diacetyl tyrosine ethylester was dissolved in 0.1 ml of the reduction buffer under weakly acidic conditions (0.1 M phosphate buffer, pH 6.0, containing 8 M Gdn-HCI, I mM EDTA and 20% ethanol) with or without 0.7>6 M 2-aminothiophcnol. After incubation of the reaction mixture at 40 °C _for 60 min, the extent of hydrolysis of 0 _-acetyl bonds ¹¹¹ N. 0 -diacetyl tyrosine ethyl ester under each set of conditions was analyzed by reverse-phase HPLC (Table IV-2). A typical reverse-phase HPLC pattern is also shown in Fig. IV-4C. Since the ester bond in N _{-acetyl tyrosine} ethylcstcr was stable under the conditions employed here (data not shown), the decrease in the peak height of the 0 -acetyl bond of N.O _{-diacetyl tyrosine} ethylester was attributable to hydrolysis of the 0 -acetyl bond.

Table ^IV -2. Hydrolysis of 0- acetyl bond in O ,N-diacetylated tyrosine ethylester under various conditions at ^40°C in ⁶⁰ m in.

pH

8.63 8.63 8.6b 8.6b 6.0c 6.0c

reducing reagent

None

2-Mercaptoethanol None

2-Mcrcaptoethanol None

2-Am inothiophenol

Hydrolysis (%)

67 97 67 97 5 17

a0.584 M Tris-HCI buffer, pH ^8.6, containing ^8.125 M urea and 5.37 mM EDT ^A.

bO.l l\1 Tris-HCI buffer, pH ^8.6, containing ⁸ M Gdn HCI, ¹ mM EDTA and 20% EtOH.

co.I M phosphate buffer, pH ^6.0, containing ⁸ M Gdn-HCI, ¹ mM EDT A and 20% EtOH.

As shown in Table IV-2, in each reduction solution at 8.6, the hydrolysis of the 0 -acetyl bond was identical. This indicated that the hydrolysis of the 0 -acetyl bond was independent of the kind of denaturant, the presence or absence of EtOH, and the concentration of EDT/\, and accelerated 111 the presence of a reduction reagent. The hydro! ysis at pH 6 was slower than that at pH 8.6, and less accelerated in the presence of a reduction reagent. Therefore, the reduction of disulfide bonds in proteins by use of 2-nminothiophenol at pH 6 was found to be useful for determining the position or a labile residue like 0 -acetylated tyrosine in proteins with disulfide bonds.

Recommended Procedure for the Reduction of Disulfide bonds in Proteins by Use of 2-Aminothiophenol.

Proteins (3-5 mg) with disulfide bonds were dissolved in 2 rnl of 0.1 M phosphate buffer (pH 6.0) containing 8 M Gdn-HCI, I mM EDTA and 20% ethanol.

The solution was degassed by aspiration. To the solution, 80 111 or 2- ami nothi ophenol, which had been preserved in a sealed tube after being degassed by aspiration, was <1dded, and then the solution was incubated for I h at 40 °C under nitrogen. Then, 200 mg of maleimide was added in the reduction solution, followed by incubation for 'JO min at room temperature. The reaction mixture was then passed through a 1.5 x 15 em column of Bio-gel P-4 equilibrated with 0.1 M phosphate buffer containing 6 M Gdn-HCI, I mM EDTA and 20% ethanol. The eluted protein was dialyzed against I 0% acetic acid and lyophilized.

As demonstrated above, due to expansion of the reduction of the disulfide bonds in proteins from a weak alkaline to a weak acidic pH region, the reduction conditions could be selected according to the nature of the material to be reduced. [t is believed that the method developed in this study will be helpful for obtaining intact information on the primary structures of proteins.

REFERENCES

Crestfield, A. _M., Moore, S., & _{Stein, W. H. ( 1963).} J. _Bioi. Clrem., 238, 622-627.

Feeney, R. E. & Whitaker, J. R. ( 1987). Protein Tailoring for Food and Medical Uses by the American Chemical Society.

Kharasch, M. S., Nudenberg, W., & Mantell, G. J. (I 95 I). J. _{Org. Chem.,} 16, 524- 532.

Kim, S. & _{Paik. W. K.} (I _970). J. _Bioi. Cllem., 245, I 806-1813.

Konigsberg, W. (1972). Method in Enz.ynw!or;y, 25, 185-188. Academic Press.

Nell' York,

Nakac, Y. & Hamaguchi, K. ( _1972). J. Biochem. (Tokro), 72. I I 55-I I 62.

Riordan, .1. r-. & Vallee, B. L. (1965). Bioc:hemistrr, 9, 1758-1765.

Takahashi, K .. Stein, W. H., & Moore, S. ( 1967 ). J. _Bioi. Clzem., 242, 4682-4690.

Uy, R. & _Wold, F. (1977). Science. 198. 890-896.

CHAPTER V

Preparation of Lysozyme with 13 C-Enriched Methionine Residues and Analysis of Its Folded State in the Presence of Substrate Analog

ABSTRACT

Jones et al. have reported that half of the £-carbons of methionine residues

111 myoglobin are enriched with stable isotopes in two-step reactions, methylation using ^I 3 C H 3 I and demethylation using dithiothreitol lJones, W .C., Rothgeb, T.M ., &

Gurd, F. _R.N. (I 976) J. _{Bioi. Chem.} 251, 7452-7460]. Based on their method, we were not very successful in preparing a lysozyme where the £-carbons of methionine residues (methionine 12 and methionine I 05) in lysozyme are enriched with stable isotopes because of the deamidations of Asn or Gin during the long demethylation under alkaline conditions (37 °C, 18 h and pH l 0.5). Therefore, we used 2-aminothiophenol, which is a reducing reagent reactive In weakly acidic conditions (see CHAPTER l _{V), during the} de me thy Ia t ion reaction because deamidation of Asn or Gin was depressed under weakly acidic conditions. From the analyses of acid urea PAGE, amino acid analyses, peptide mappings under non­

reducing conditions on reverse-phase HPLC and two-dimensional NMR spectra, the structure of the I 3 C-enr iched methionine lysozyme obtained by using 2- aminothiophenol was identical to that of the intact lysozyme. The total yield was

30%. Using this lysozyme with enrichment of the £-carbon of the methionines, I 3c-NMR analyses were carried out to examine local structure and mobility around the methionines in the presence or absence of a trimer of N-acetyl glucosamine, (NAG) 3. From analysis of the relaxation process of the 1 � _{C H 3 residue in} methionines, the flexibility around Met 105 of the lysozyme was found to be depressed in the presence of (NAG)].

INTRODUCTION

As was described in CHAPTER II and III, NMR spectroscopy had an advantage to detect the subtle structural changes of protein in solution. In this CHAPTER, I describes NMR spectroscopy was also utilized for the detection of the mobility of amino ncid residue in proteins. The amino acid residues in proteins are mobile on various time scales. Especially, rapid mobility was suggested to be required for protein functions such as nctivity (Tsou, 1993). Due to the development of the NMR, this sort of mobility in protein (pico second order-micro second order mobility) be detected by means of measuring relaxation times using NMR and has been can

vigorously discussed. 13c-NMR is convenient for detecting the mobility of protein because the relaxation mechanism of the I 3c nuclei is simpler than that of the 1 H nuclei (Void et a/., 1968). However, the sensitivity in 13c-NMR is low as the nucleus has a low natural abundance and hns a small gyro magnetic ratio. Therefore, l3c- enrichment using stable isotopes is required to effectively obtain the information of protein mobility (Jones et a/., 1976; Richarz et a!., 1980; Nicholson et a/., 1992).

Recently. the enrichment of amino acid residues in protein with stable isotope is easy to carry nut by means of genetic engineering. However, for this purpose, it is also necessary to prepare the particular expression system of the desired protein (appropriate minimum cultivation medium, etc). Two decades ago, Jones et al.

( 1976) demonstrated that £-C H 3 of the two methionines in myoglobin could be enriched with I 3 C H 3 I using chemical modification (scheme I).

tzcH3

I

s

I

(CHz)z -CH-

I

13CH�2CH3 's/

I

•

(CHz)z -CH-

I

Scheme I. Methylation and dcmcthylation

R-SH

This method IS convenient for the preparation or 13c-enriched methion ine residues in proteins because the paticular expression system or proteins is not required. However, 111 the process of dcmcthylation of methylated methionine obtained by the reaction of myoglobin with I) C H )l, they used dithiothreitol under alkaline conditions. Under these conditions (pH 10.5, .�7 °C, 18 h), we wondered if various chemical reactions may occur, for example, deamidation <rrom Asn and Gln to Asp and Glu) (Tyrer-Closs & Schirch, 1991 ). Actually, their method has not been popular for preparing proteins with the I J C-enriched methionine residue while it has an advantage. Therefore, at present, it should be worth while to improve their method, especially the demethy I at ion process.

In CHAPTER V, a successful method for the preparation of a lysozyme with l 3c_ enriched methionine residues using 2-aminothiophenol, which is a reductant reactive under weakly acidic conditions (see CHAPTER IV), was described using the process of demethylation of the methylated methionine residue 111 lysozyme.

Moreover, it was suggested the subtle changes of structure and mobility around the methionine residues in the presence of a substrnte analog, (NAG)3, using two enriched methionines as probes.

MATERIALS AND METHODS

Materials.

Dithiothreitol was purchnscd from Nakalai Tasque (Kyoto). I lc H 31 was the product of euriso-top. A trimer of N-acetyl glucosamine was purchased from Seikagaku Kogyo Co., Ltd. All other chemicals were of analytical grade for biochemical use.

Methylation of Methionine Residue in Lysozyme.

The methylation of methionine residue in lysozyme was performed according to the literature (Link & Stark, 1968). Namely, 20 mg or lysozyme was

dissolved in 0.1 N KNOJ (pH 3.3) containing 6 M Gdn-HCI. 13c-labeled methyl iodide (0.1 M) w^;1s added to the solut ion. After the soluti on was vigorously stirred at room temperature for I 0 h i n the dark, it was dialyzed against l 0% AcOH and lyophilized.

A11alytica/ Method.

fab mass spectrum of n pept ide sample was measured using a JMS DX-300 (JEOL) mass sp^ectromete^r. Acid urea P/\GE was performed according to the method of Hollcckcr and Creighton (Hollecker & Creighton, 1980). The condition was the same to thei rs expect for pH (pi-1=3.6). Amino acid <Jnalysis and peptide mapping were carri ed out accord ing to the methods described in CHAPTER ¹¹ and ^111.

Demethylation of Methyl Methionine Residue.

A:by Use of Ditltiothreitol.

Demethylat i on us i ng dithiothreitol was performed according to the literature (Jones et a/., 1976).

B:by Use of 2-Aminothioplrenof.

Lysozyme w i th methylated methionin^e residue ( 10 mg) was dissolved in 1 ml of 0.1 M MOPSO buffer (pH 6.0) containing 8 M Gdn-HCI, l mM EDTA and 20% EtOH.

After 2-am i nothi ophenol (2 ml) was added to the solution, and the solution degassed us i ng a vacuum pump and then sealed. The reaction mixture

was was incubated at 50 °C for an app^roximate time. To remove 2-arninothiophenol, the reacti on soluti on was appli ed to a 8io-Gel P-4 column ( 1.5 X 20 em) equ i l ibrated with 20% ethanol solution containing 6 M Gdn-HCI. The eluted protein was dialyzed against I 0% /\cOH at 4 °C ^and I yo phi lizcd.

R e 11 at u ratio 11 of Red 11 c e d Lysozyme.

Renaturation of lysozyme was performed acco^rding to the method described i n CHAPTER lll.

Purification of the Renaturated Lysozyme.

Purification of the renaturated lysozyme us i ng gel and reverse-phase chromatography was performed according to the method described in CHAPTER III. For further pur i f i cation, the protein fraction was applied to the ion exchange HPLC column of CM Toyopearl 650M (5 X 400 mm). The column was eluted with a gradient formed with 100 ml of 0.05 M phosphate buffer (pH 7.0) and 100 ml of 0.05 M phosphate buffer (pH 7.0) conta ining 0.5 M NaCl at a flow rate of 1.5 ml/min. To remove the 10 1-� lysozyme, a lysozyme deri vative with a �-aspartyl sequence at Asp 1 0 1-G ly I 02 (Yam ada ^et cd., l 985), the eluted protei n fraction a ftc r dialysis again^st water, was then applied on the HPLC column of CM Toyopearl 650M (5 X ⁴00 mm) at pH 5.0 according to the literature (Tomizawa et a/., 1994). The column was eluted with a gradient formed with I 00 ml of 0.05 M acetate buffer (pH 5.0) and I 00 ml of 0.05 M acetate buffer (pH 5.0) containing 0.5 M N^aCI at a flow rate of 1.5 ml/min. The eluted protein fraction was dialyzed against water and lyophil ized.

NMR Measurement.

IH_I3c HSQC experiment (Boudenhauzen & Ruben, 1980) and IJc-edited NOESY (Bax & Weiss, I 987) experiments at pH 3.8 were carr i ed out using the standard procedure. ln the I H-I 3c HSQC experiment, 16 transients were typically

recorded for each of 256 increments. A digital resolution of 2.4 Hz per point in the I H dimension and 1 1.7 Hz per point in the I 3 C dimension was used. In the I 3 C­

edited NOESY, 16 transients were typ ically recorded for each of the 512 increments. A digital resolution of 2.4 Hz per po int in both dimensions was used.

Mixing time was 150 msec ( 13C-edited NOESY). Spin-lattice relaxation time (T 1) was obtained by the inversion recovery method (Gross, 1967), and the sp1n-sp1n relaxation time (T2) was obtained by the CPMG method (Meiboom & Gill, 1958). The other NMR experiments were carried out according to the methods described in CHAPTER II and III.

RESULTS AND DISCUSSION

Methylation of Methionine Residues in Lysozyme.

Hen egg-white lysozyme bas two methionines (lmoto ^et al., 1972). One is methionine 12, which is in an a-helix, and the other is methionine 105 which is in the center of the "hydrophobic box" (Fig. V-1 ). l t was postulated that €-C H] of

these methionines were difficult to react using chemic<11 reagents in the native state because these residues are present in the interior of the lysozyme molecule.

Therefore, lysozyme was reacted with I] _{C H} 3 _{I at pH} 3.3 in the presence of 6 _{M Gdn-} HCl (see Mi\ TERIALS AND _METHODS).

Met 105

Fig. V -1. Crystal structure of hen egg-white lysozyme [set 16RS, Diarnmond, 1974:

coordinates taken from Protein Data Bank (2LZT)]. Side chain of methionines are indicated.

S i nee methy I methionine could be detected between am mania and histidine with a Hitachi L-8500 amino acid analyzer under the routine conditions, we determined the number of methyl methionines by this method. Twenty-two percent of the methyl methionine was hydrolyzed to methionine during acid hydrolysis under the conditions employed. The number of methyl n1ethionines was determined after the correction using this value. The number of methyl methionincs increased with reaction time (Fig . V-2A). The modification of methionine in lysozyme was found to be completed in l 0 h. Since its ami no acid composition except for cystine or tryptophan, which could not be estimated by means of acid hydrolysis, was identical to that of the native lysozyme, no side reaction occurred during the reaction.

100

r-.... 100 r-....

B

'F;f. 'F;f.

--- ---

= =

0 0

·-_� ·.c

� 50 � 50

- -

� �

..c ..c

� �

� �

� 8

_�

Q

0

5 1 0 0 1 0 20

Reaction time (hour) Reaction time (hour)

Fig. V -2.

A)

Time course of the methylation of methionine in 0.1 M KN03 _(pH 3.3) containing 6 M Gdn-HCI and 0.1 M 13CH31 at room temperature.

B)

Time course of the demethylation of methylated lysozyme with 2-aminothiophenol (6.24 M) in 0.1 M MOPSO buffer (pH 6.0) containing 1 mM EDT A, 8 M Gdn-HCI and 20% ethanol at 50 _°C.

In acid-urea PAGE, proteins are separated by the difference in the net charge in the protein (Hollecker & Creighton, 1980). Therefore, it may be suitable to analyze whether methylation occurred during the reaction time. Before the samples were applied to acid-urea PAGE, they were reduced and S- carboxymethylated to decrease their net charge for better separation. Both the modified and native Iysozymes mainly showed one band (Fig. Y-3). In comparison of the mobility of the modified lysozyme with that of the native lysozyme, the modified lysozyme had a greater positive net charge than the intact lysozyme because the mobility of the modified lysozyme was larger than that of native lysozyme. This result indicated that only a slight side reaction occurred under the above conditions.

A B c D

'•

Fig. V -3. Acid-urea PAGE of reduced and S-carboxymethylated lysozymes.

A) from native lysozyme, B) from demethylated lysozyme using 2-aminothiophenol in weakly acidic condition (details in MATERIAL and METHOD), C) from

demethylated lysozyme by Jones's method, D) from methylated lysozyme.

In order to confirm that two methionine residues 111 the lysozyme were methylated with methyl iodine, the primary structure of the modified lysozyme was analyzed. The elution pattern of the tryptic peptides of the reduced ami S- carboxymethylated modified lysozymes on reverse-phase H PLC 1s shown in J-ig. Y-

4B. For comparison, that of the native lysozyme is shown in Fig. Y-4A. The peaks,

T 3, T 1 2 + 1 3, T 1 3, which contain the methionine residue disappeared, but new peaks,

T a, Tb, T c, appeared. From the amino acid analyses or these peptides, Ta, Tb, and T c had the same amino acid composition as that of T] +4, T 1 2 + l 3, and T 1] except for methionine, respectively. These results suggested that new peptides with the methionine residue that had a positive charge due to the modification of methyl iodine eluted faster than the original peptides. Similar behavior on reverse-phase HPLC was observed in a previous paper in a peptide whose methionine residue has a positive charge due to the modification of methyl iodine (Sasagawa et n!., 198::1).

On the other hand, since the peptide pattern derived from native and modified lysozyme except for these three peptides were identical to each other, it was confirmed that there was no reaction on amino acid residues except for the methionine residue in the modified lysozyme. From these results, It was concluded that two methionine residues in the lysozyme completely reacted with methyl iodine in I 0 h. The yield of the methylated lysozyme was 95%.

Demethylati on of Methyl a ted Met hi on ine Residues in Lysozyme.

Jones et al. ( 1976) demethylated the methylated methionine residues in myoglobin using dithiothreitol. First, the methylated methionine residues in the lysozymes was tried to demethylate according to their method. Since the amino acid composition of the methylated lysozyme after demethylation using dithiothreitol was identical to that of the native lysozyme, the side reaction could not be detected during the reaction except for cystine or tryptophan which could not be estimated by means of acid hydrolysis. However, in acid-urea PAGE, the

A

I

l

I

B

l

�

l

c

l

1

l

0

�ff")

,... ,....(

�

^M _,...

+'�

� �

l

_l

₁ 1 ^{l �LL�}

<:.1

�

� ,CJ

� �

A

1\.

)l

'---'

�

�

"'0

�

l _t 1 l 1

_�

1.1

.60

Retention time (min)

120

Fig. V -4. Reverse-phase HPLC of tryptic peptide from reduced and S-carboxymcthylated A) native lysozyme, B) methylated lysozyme and C) demethylated lysozyme on W akosil SC 18 ( 4.6 x 250 rnm).

T3 wasS-carboxymethylated Cys 6-Lys 13; T12+13 was Lys 97-Arg 112; T13 was Ile 98-Arg 112. T,l' T1v and T<.: were methylated peptides from T3, T12+13, and T13, respectively. T0 and Tc were

demethylated peptides from Ta and Tt:, respectively. The column was eluted with the gradient of 40 ml of 40% acetonitrile containing 0.1% HCl at a flow rate of 0.6 ml/min.

demethylated lysozyme showed a broader band than the intact lysoLyme (Fig. V- 3C). This indicated that the demethylated lysozyme consisted of several species.

Moreover, the lysozyme demethylated by using dithiothreitol was found to have little activity because the demethylated lysozyme could not refold. The alkaline condition (pH !0.5) and long incubation (18 h) may cause deamidation of asparagine or glutamine as has been previously suggested (Tyrer-Closs & Schircll, 1991 ). Recently, Maeda ^et a/. ( !994) demonstrated that the renaturation yield of the reduced lysozyme decreased with the decrease in positive net charges of the reduced lysozyme. Therefore, the reason why the lysozyme demethylated using dithiothreitol could not refold may depend on the decrease in the positive net charge due to deamidation. From these results, it was suggested that demethylation of the methylated methionine residues in the lysozyme surely took place as was shown in the literature (Jones ^et al., 1976) but unfavorable side reactions also occurred during the demethylation. Therefore, the yield of the I 3 C-e n r i c h e d methionine lysozyme using dithiothreitol was 0% of the methylated lysozyme.

Since deamidation of Gin or Asn were depressed in weakly acidic conditions, demethylation of the methylated lysozyme with l3c H 3! using 2-aminothiophcnol.

which is a reduction reagent in weakly acidic conditions (see CHAPTER IV), was carried out (see MATERIALS AND METHODS). After the amino acid analysis of the reacted lysozyme, the number of methionines was evaluated. The number of methionines increased with the reaction time and reached a plateau in 24 h (Fig.

V-28). Therefore, the demethylation of methyl methionine in lysozyme was found to be completed in 24 h at 50 oc. Since the amino acid composition of the lysozyme demethylated using 2-aminothiophenol was identical to that of the native lysozyme, no side reaction also occurred during the demethylation using 2- amino thiophenol.

In acid-urea PAGE, the band of the demethylated lysozyme was predominantly single and had the same mobility to that of the intact lysozyme. The demethylated lysozyme was suggested to be a single species with the same net

charge :1s the intact lysozyme (Fig. V-38). This result indicated that during the demethylation under weakly acidic conditions, deamidations were considerably depressed. Moreover, in order to confirm that two methionine residues in the lysozyme were demethylated using 2-aminothiophenol, the primary structure of the lysozyme dcmethylated using 2-aminothiophenol was analyzed.

The sample was reduced and S -carboxymethylated, and the elution pattern of the tryptic peptides on reversed-phase HPLC is shown in f-ig. Y-4C. There was no difference in the patterns between Fig. V-4A and 4C. Therefore, it was confirmed that two methionine residues were demethylated and there was no side reaction on the amino acid residues under the conditions employed. Peptides Te and Td s h o w e d the same amino acid composition of T3 and Tjj, respectively. The molecular weight of the pcptidcs was cxnn1incd using FAB-Mass spectroscopy. The FAB-mass spectrum of peptide Td indicated the two peaks at 894 and 895 daltons, while the intact peptide Tr3 showed the 894 dalton. These results indicated that £-C H 3 of the

methionine residues in the lysozyme was confirmed to be enriched with 1 3 C nuclei. The yield of the unfolded lysozyme enriched with I 3c thus obtained was 85% of the methylated lysozyme.

Conformation of the Renatu rated Lysozyme That Contai ns I 3 c _

Enri ched Methioni11e Res idues.

The rennturation of the lysozyme with 13c-enriched methionine residues (I 3 C-enrichcd methionine lysozyme) was carried out according to the method described in CHAPTER I I I . The 13c-enriched methionine lysozyme was purified

· 1 1 t h d cve1·se pl1ase HPLC :1ccord.ng t

to the method us1ng ge c noma ograp y an r· - ^,

described in CHAPTER Ill, and further purified using ion exchange chromatography. The elution position of the renaturated sample on reverse-phase HPLC nnd ron exchange chromatography was similar to that of the intact lysozyme. Since its activity was also identical to that of the native lysozyme, the

structure of the renaturared lysozyme was t.:lucidated to he similar to that of the native lysozyme.

Furthermore, in order to examine the tertiary structure, I H-I II phase sensitive DQF-COSY spectrum of the renaturated lysozyrne was measured since the NMR is sensitive enough to detect the difference in the tertiary structure of the proteins. Comparing the spectrum of the renaturated lysozyme with that of the intact lysozyme, there was no difference between these patterns. The result was identical to the result in CHAPTER lii. The total yield of lysozym�.; with I] C - enriched methionine residues based on the native lysozyme was 30%. The decrease in yield depended on the failure of the ref olding. Even tn weakly acidic conditions, an unfavorable chemical reaction such as the formation of the 13- aspartylglycyl sequence at the Asp-Giy sequence 111 lysozyme (Tomizawa et a/., 1994) is known to occur. Anyway, I could obtain enough of the lysozyme with I 3c - enriched methionines to measure the I 3 C-NMR spe ctrum using 2- aminothiophenol 1n the demethylation process of the methylated methionine residue.

Assignment of 13 C-Enriched Methionine Lysozyme in I 3 C -N MR.

The 13c-NMR spectrn of the I 3c-enriched methionine lysozyme at pD 3.8 and 35 °C are shown in Fig. V -SA. There were two signals. The resonances of the E-C H 3

of methionine 12 and methionine I 05 were assigned by means of I H-I] C HSQC (Fig.

V- 6) because the £- C H 3 proton resonances of both methionines have already been assigned (Redfield & Dobson, 1988). From the I H-13c coupling resonance, it was clear that the 13c resonances originated from the E-CH3 in the two methionines.

The chemical shift of £-CH 3 in methionine 12 was 16.90 ppm and that in methionine I 05 was 14.06 ppm at 35 °C and pD 3.8.

When comparing the shapes of these two resonances, there was a minor resonance around the main resonance at methionine I 05 (Fig. V-5A). The m1nor resonance may have originated from methionine I 05 because the signal area of

A

I I

19

I 1 1 I I I

18 17

I I I , 1 l I I I I I

16 15 14

, 1 r r ' , 1

13 12

I I

11 ppm

Fig. V -5. Low field ISO

MHz 13C-NMR

spectra of 1�C-enriched lysozyme in the absence

(A)

and in the presence

(B)

^{of 1.61 m}

M (NAG)3

^{at pD 3.8 and 35}

°C.

1."2 (ppm) -0.2

0.0 0.2 0.4 0. 6 - 0.8 1.0 1.2 1.4 1.6-- 1. 8- 2.0 2.2

19 18

Met105

<>

Met 12

17 16 15 14 13 12 11 10

Fl (ppm)

Fig. V-6. 1H-13C HSQC spectrum of 13C-enriched methionine lysozyme at pH 3.8 _and 35 _°C.

The 1 H-13C _cross peaks are labeled according to the residue number of methionine.

the methionine 12 resonance was equal to the total area (major ⁺ minor) of methionine I 05 . After the measurement for 24 h at pD j .8 and ]5 °C, the sample was applied to the ion exchange HPLC column of CM Toyopearl 650M (5 X 400 _mm) at pH 7. The column was eluted with a gradient formed with 100 ml of 0.05 _M phosphate buffer (pH 7.0) and 100 ml of 0.05 M phosphate buffer (pH 7.0) containing 0.5 M NaC I at a flow rate of 1 .5 ml/min. The minor peak that eluted later than the major peak of the native lysozyme on the ion exchange chromatogram was identified to be I 0 1-succinimide lysozyme where the aspartylg1ycyl sequence at Asp 101-Giy 102 formed a cyclic imide. The isomerization from the native lysozyme to the I 0 1-succinimede lysozyme was suggested to occur non- enzymatically under acidic conditions at 40 °C (Tomizawa et a/., 1994). In our previous report, the I H chemical shift of £-C _{H 3 of Met I} 05 of lysozyme in ^{I H-}N M ^R was shown to be altered by 0.04 ppm by isomerization from the ^a-asp arty I g l y c y I sequence at Asp10 1-Giyl02 to the 0-one (Endo et a/., _{1987) .} Therefore, the local conformational change around Asp I 0 I may affect the chemical shift at Met I 05.

From these results, the minor signal in the 13c-NMR of I 3c-enriched methionine lysozyme may result from the formation of the I 0 1-succinimide lysozyme during the accumulation.

Binding of Substrate Analog to 13 C-Enriched Methiofline Lysozyme.

Lysozyme is one of the 0-1 ,4-N-acetylmuramidases. A trimer of N- acetylglucosamine, (NAG):,, is a constituent of the substrate and nonproductively binds to the upper part of the active site cleft in the lysozyme (lmoto et al., _{1972) .} Using a I 3c-enriched methionine lysozyme, it is possible to detect a change in the local structure and the dynamic behavior around the methionine residues when a substrate analog binds to the lysozyme. Met 12 is in an a-helix and Met 105 is in a hydrophobic region (lrnoto et a/., _1972). One-dimensional I 3 C-NM R _and I 3 C-edi ted NOESY spectra of the 13c-enriched methionine lysozyme at pH 3.8 and 35 co are shown in Fig. Y-5A and Fig. _V-7A, respectively. These spectra in which most of

F2 - (ppm) -0.2

0.0 0.2 ..

0.4

0.6 o.8- 1. 0.

1. 4•

1.6 1.8

F:d -:

(ppm)_

-0.2 o. o·

0. 2 . 0. 4:

0.6- 0. 8.:

1.0 1.2 1. 4- 1. 6- 1.8

WIIIC:211 W28C411

t t

0 ^{• 0} W28CG

..

^II WI08C4ll

..

B

7 6

Wl!IC1ll WZRC-Ill

t. t

0 • '

.. ..

W!OOC·III \Vi08C511

A

7

I·

5

6 5

W28uCII N27f\CIII N27flCIII

t

.

�

t t

MI0.5ctCII

..

W28uCI I ^W2Bf\CI 11

t t

f) II � 10 V29nCII

..

--, .. �-·,--,·-r-r-1__.,�,-,-

4 3

P1 (ppm)

W28aCll

t

.. ��

�1105uC!!

W28aCI I W28flCII1

4

t t

o o � a V29aCll

..

3 F1 (ppm)

0 •

2

2

. I

.,/\ 1 1 1 I -1 -" ,.

1 -0 -1

1 -0 ·1

Fig. V -7. 13C-edited NOESY spectrum of 1 �C-enriched methionine lysozyme in the absence (A) and in the presence

(B)

of 1.61 mM (NAG)} at pH 3.8 and 35

°C

with a l 50 msec mixing time. The 1 H-1 H

(£-CH3

proton of methionine residues) NOEs are labeled according to residue number.

Table V -1. 13C nuclei relaxation time of methionine EC in 13C enriched methionine lysozyme on the 150 MHz spectrometer at pD 3.8 and 35

° C

in the presence or absence of 1.61 mM

(NAG)

Absence of

(NAG)j

Presence of

(NAGh

Met 12 Met 105

1.87 1.49

0.01 0.02

aLongitudinal relaxation time constant.

bTranverse relaxation time constant.

1.83 1.03

().01

<0.01

the lysozyme forms the (NAG)]-Iysozyme complex arc also shown in Fig. V-58 and Fig. V-78. In the presence of (NAG)], although the chemical shifts of the £-CH] in the two methionine residues did not change, the resonance or Met I 05 became broader than that in the absence of (NAG)]. The line shape of the resonance is strongly related to the relaxation time of the nuclei. Also the latt�r dir�ctly reflects the mobility of the nuclei (Doddrell et a/., 1972). Therefore, the relaxation times (T 1, T2) of the resonances in £- C H 3 of these two methionine residues in the absence or presence of (N AG)j were measured (Table Y - I). Both T 1 and T2 at £- C H 3 of Met 105 in the absence of (NAG)3 are smaller than those in the presence of (NAG)]. These results indicated that the mobility around Met 105 was altered when a substrate analog is bound to the active site cleft while Met I 05 is apart from the active site cleft (Cheetham et ol., 1992). Jones et a!. ( 1976) have discussed the mobility of aquoferrimyoglobin and carbonmonoxymyoglobin by analyzing both T 1 and T2 of the resonance £- C H 3 enriched with methionine residues. In their analyses, the model suggests that the methionine £-carbon nuclei undergoes rotational reorientation characterized by an internal correlation time, TG, within a rigid protein matrix that is reorienting with an overall correlation time of 'tR. The relationship between TG and T1 or T2 strongly depends on TR (Doddorell et

a!., 1972). Considering the value of 'tR to be 5.7 ns according to the literature (Buck et a/., 1995), a value for TG increases with the decrease in T 1 or T2 (Doddorell et al., 1972; Richerz eta/., 1980; Buck et a!., 1995). Therefore, the mobility of the