Identification, purification and reconstitutions of thiamin metabolizing enzymes in human red blood cells

171

Biochimica et Biophysica Acta, 1160 (1992) 171-178

© 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4838/92/$05.00

BBAPRO 34328

Identification, purification and reconstitution of thiamin metabolizing enzymes in human red blood cells Yoshiko Egi, Shin-ya Koyama, Toshihiro Shioda, Kazuo Yamada and Takashi Kawasaki Department of Biochemistry, Hiroshima University School of Medicine, Hiroshima (Japan)

(Received 16 March 1992) (Revised manuscript received 15 May 1992)

Key words: Thiamin diphosphate; Thiamin pyrophosphokinase; Thiamin triphosphate; Cytosolic adenylate kinase; Red blood cell; (Human)

Thiamin and its mono- (TMP), di- (TDP) and triphosphate (TTP) were assayed in adult human whole blood using high-performance liquid chromatography (HPLC). TDP and TTP were detected in red blood cells (RBC), but not in plasma. After incubation with 20/~M thiamin and 5 mM glucose for 2 h, the TDP and TI'P contents of RBC increased from 111 to 222 and 0.6 to 2.2 nmol/l of packed RBC, respectively, suggesting enzymatic conversion of thiamin to TDP and then to TTP. Thiamin pyrophosphokinase (TPK, EC 2.7.6.2) had not been isolated before from human materials, nor had cytosolic adenylate kinase (AK1, EC 2.7.4.3) in human RBC been demonstrated to catalyze the phosphorylation of TDP to TTP, although AK1 from pig and chicken skeletal muscle possess Tl'P-synthesizing activity. TPK and AK1 in a human RBC lysate were therefore purified by a series of the conventional techniques. The specific activity of the purified TPK, which was obtained as a single protein, was 720 nmol TDP formed/mg protein per h at 37°C. A partially purified AK1 preparation catalyzed the formation of T/'P from TDP (specific activity, 170 nmol/mg protein per h at 37°C) in addition to its proper reaction to form ATP from ADP. After incubation of the purified TPK and AK1 with 20/zM thiamin in the presence of ATP, ADP and Mg 2+ at 37°C for 48 h, the amounts of TDP and TTP synthesized were 465 and 54.0 pmol/250 /zl reaction mixture, respectively. Neither TDP nor T/'P was formed when TPK was omitted from the reaction mixture and an omission of AK1 resulted in the formation of TDP alone.These results indicate that thiamin is converted to TDP by TPK and, subsequently, to TI'P by AK1 in human RBC.

Introduction Thiamin occurs in cells as free thiamin and as its phosphate esters, thiamin monophosphate (TMP), thiamin diphosphate (TDP) and thiamin triphosphate (T-I'P). T D P is known to act as a coenzyme for several i m p o r t a n t e n z y m e s involved in c a r b o h y d r a t e metabolism [1,2]. The existence of T I ' P in animal tissues has been reported [3-11], but its physiological role is unknown.

Correspondence to: T. Kawasaki, Department of Biochemistry, Hi-

roshima University School of Medicine, Hiroshima, Hiroshima 734, Japan. Abbreviations: TPK, thiamin pyrophosphokinase (EC 2.7.6.2); AK1, cytosolic adenylate kinase (EC 2.7.4.3); RBC, red blood ceils; TMP, thiamin monophosphate; TDP, thiamin diphosphate; qq'P, thiamin triphosphate; TTFD, thiamin tetrahydrofurfuryldisulfide; PT, pyrithiamin; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonicacid; TCA, trichloroacetic acid.

Thiamin pyrophosphokinase (TPK, EC 2.7.6.2), which catalyzes the pyrophosphorylation of thiamin to TDP in the presence of A T P and Mg 2+, is the first enzyme in the thiamin metabolic pathway. In animal tissues, partially purified TPK has been prepared from rat liver [12], pig heart [13] and it has been extracted from pig brain and purified to a single protein [14]. However, human TPK has not been purified or characterized nor has the metabolic pathway of thiamin to q-TP via T D P in human tissues been elucidated. Recently, we reported that cytosolic adenylate kinase (AK1, EC 2.7.4.3) catalyzes the phosphorylation of T D P to T-FP in muscles of the pig [8,9] and chicken [10,15,16]. Although human AK1 has been well characterized [17] and the complete nucleotide sequence of its c D N A has been reported [18], whether or not it possesses TrP-synthesizing activity has not been established. In this study, we present evidence that thiamin is metabolized in human RBC. First, thiamin was metab-

172 olized to TDP and then to TTP in fresh RBC in the presence of glucose. Then, TPK and AK1 were purified from RBC and the activity of each was characterized. Third, the thiamin metabolic system of human RBC was reconstituted with purified TPK and AK1. Materials and Methods

Chemicals. TDP, thiamin tetrahydrofurfuryldisulfide (T-I'FD), pyrithiamin (PT) and hydroxyethylthiamin were gifts from Y. Oka, Central Research Laboratory of Takeda, Osaka and TTP and chloroethylthiamin were gifts from M. Yamazaki, Central Research Laboratory of Sankyo, Tokyo. Oxythiamin and amprolium were purchased from Sigma. The chemicals used were obtained as follows: DEAE-cellulose (DE52) from Whatman; Blue Sepharose CL-6B, Sephadex G-200, PD-10 gel filtration column and CNBr-activated Sepharose 4B from Pharmacia; hydroxyapatite from Bio-Rad; LiChrosorb NH 2 from Merck and all others used were of analytical grade. Materials. Human packed RBC were supplied by K. Okada, Division of Blood Transfusion Service, Hiroshima University Hospital, Hiroshima. Fresh human blood was collected from healthy adult volunteers (age 25-50 years). TMP-agarose was prepared with CNBr-activated Sepharose 4B, as described by Wakabayashi et al. [19]. Sample preparations from human fresh blood. Fresh blood was taken using heparinized syringes from healthy adult men, who had fasted for 12 h prior to collection. The samples were centrifuged at 600 × g for 10 min, the plasma was separated from the blood cells. The buffy coat was removed from the RBC sediment by suction, which decreased contamination of white blood cells in particulate fraction to 20% of that in whole blood and TPK activity derived from white blood cells was less than 12% of the total TPK activity in the packed cells. The RBC were washed with 10 mM Hepes-buffered physiological saline (pH 7.4) and suspended in one volume of this solution, which also contained 5 mM glucose. Determination of thiamin and its phosphates in human fresh blood. Trichloroacetic acid (TCA, 50%, 1/ 4 volume) was added to each of the preparations (whole blood, plasma and RBC) described above, which were centrifuged at 10 000 x g for 10 min. The supernatant was taken off and extracted three times with diethyl ether to remove the TCA. The thiamin compounds in the aqueous layer were oxidized with BrCN and analyzed by high-performance liquid chromatography (HPLC) as described previously [4,6].

Assay of TDP and TTP synthesis in human fresh RBC. Human fresh RBC suspensions were prepared as

described above and aliquots (1.4 ml) were incubated with 20 p.M thiamin or 20/zM TTFD in the presence of 5 mM glucose at 37°C for 30 min, 2 h and 4 h. The RBC were collected by centrifugation at 700 x g for 2 min and washed twice with Hepes-buffered saline, to which was added 1/5 vol. of 50% TCA and centrifuged at 10000 × g for 10 min. The supernatant was extracted with diethyl ether and the thiamin compounds in the aqueous layer were analyzed as described above. TPK activity assay. The standard assay mixture (total volume 250 /zl) consisted of 100 mM Tris-HCl (pH 7.5), 20/zM thiamin, 5 mM ATP, 5 mM MgCI2, 1 mM dithiothreitol (DTT) and various amounts (50-200/xl) of the enzyme preparation. Each mixture, without thiamin, was preincubated at 45°C for 5 rain, then the reaction was started by adding thiamin and incubated for 60 min at 45°C. This reaction temperature was chosen to enable an accurate analysis of the initial velocity. The reaction was terminated by adding 50 p.1 50% TCA and the mixture was centrifuged at 10 000 x g for 10 min to remove the precipitated proteins. The supernatant obtained was extracted with diethyl ether and the thiamin compounds in the aqueous layer were analyzed as described above. The TPK activity was expressed as nmol TDP formed/mg protein per h. AK1 activity assay (ATP formation). The assay was performed as described previously [8,10]. The standard assay mixture (total volume of 250/xl) consisted of 100 mM Tris-HC1 (pH 7.5), 2 mM ADP, 2 mM MgClz, 1 mM DTT and various amounts of the enzyme (50 /xl enzyme preparation diluted 10-100 000-fold with buffer A, which was composed of 20 mM Tris-HC1 (pH 7.4), 1 mM EDTA and 20 mM 2-mercaptoethanol). Each mixture, without ADP, was preincubated at 25°C for 5 rain, after which the reaction was started by adding ADP and incubating for 5 rain at 25°C. The reactions were terminated by adding 50 /zl 50% TCA and the mixture was centrifuged at 10000 x g for 10 min. The supernatant was extracted three times with diethyl ether and the ATP in the aqueous layer was analyzed by HPLC as described previously [21]. The AK1 activity was expressed as mmol ATP formed/mg protein per h. Assay of TTP-synthesizing actiuity of AK1. The assay was carried out as described previously [8,10]. The standard assay mixture (total volume of 250 #1) comprised 100 mM Tris-HCl (pH 7.5), 0.1 mM TDP, 0.1 mM ADP, 0.5 mM MgCI2, 1 mM DTT and various amounts of the enzyme (100 /zl enzyme preparation, which was diluted 1-100-fold with buffer A). The mixture, without the substrate TDP, preincubated for 5 min at 37°C, the reaction was started by adding TDP and the mixture was incubated for 1 h at 37°C. The reaction was terminated by adding 50 /zl 50% TCA, which then was removed by extracting three times with diethyl ether. The thiamin compounds in the reaction

173 mixture were analyzed as described above and the enzyme activity was expressed as nmol TTP formed/mg protein per h. The reaction proceeded over 3 h and the rate was proportional to enzyme amount under the standard assay condition. Purification of TPK and AK1 from RBC. The entire procedure was carried out below 4°C. 2 1 of human packed RBC were lysed for 30 min with 9 volumes of 10 mM Tris-HCl (pH 7.4), 1 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride and 20 mM 2-mercaptoethanol. The hemolysate was filtered through 2 sheets of Advantec No.2 paper filter, activated DE52 (400 g wet weight) was added to the filtrate, the mixture was stirred for 30 min and filtered through a glass filter. The TPK was adsorbed by the DE52 and the AK1 was dissolved in the final filtrate. The TPK-containing DE52 was packed into a column (7 × 35 cm) and washed first with 1.5 I buffer A and then with 1.5 1 buffer A containing 0.1 M NaCI, after which the TPK was eluted with a linear gradient of 0.1 to 0.2 M NaC1 in buffer A. The DEAE-cellulose fraction which corresponded to the TPK activity peak was collected, the TPK was precipitated by adding ammonium sulfate (55-70% cut), collected by centrifugation at 10000 × g for 20 min, the precipitate was dissolved in buffer A and this fraction was desalted with PD-10 (ammonium sulfate fraction). This fraction was then applied to a Sephadex G-200 column (2.5 x 100 cm), which was equilibrated with buffer A and eluted with the same buffer. The peak fractions of TPK activity obtained were pooled (Sephadex G-200 fraction) and dialyzed overnight against 20 mM acetate buffer (pH 4.5), centrifuged to remove the precipitate and then dialyzed against 20 mM phosphate buffer (pH 6.8) (acid-treated fraction). This fraction was applied to a hydroxylapatite column (1 x 20 cm), which was equilibrated to 20 mM phosphate buffer (pH 6.8) and eluted with the same buffer. The unadsorbed fraction, which contained the TPK activity (hydroxylapatite fraction), was dialyzed against buffer A and applied to a TMP-agarose affinity column (5 ml) [14,19], which was equilibrated with buffer A. The column was washed first with buffer A and then with buffer A containing 2 mM ATP and the TPK was eluted with a linear gradient of 2 to 20 mM ATP in buffer A. The fractions were analyzed by SDS-PAGE. Fractions containing single TPK protein were collected (lanes 5-8 in Fig. 3) and stored at 4°C until required for use. The AK1 was purified as described by Nealon [22]. The filtered fraction, which was not adsorbed by the DEAE-cellulose, was applied directly to a Blue Sepharose CL-6B column (1.5 × 15 cm) equilibrated with buffer A. The column was washed first with 200 ml buffer A, followed by 200 ml buffer A containing 2 mM NADH and finally with 100 ml buffer A contain-

ing 0.25 M KCI. The AK1 was eluted with a linear gradient of 0.25 to 1.0 M KCI in buffer A. The AK1 activity peak were pooled and passed through a PD-10 column to desalt and exchange buffer A for buffer B (10 mM Hepes-NaOH (pH 7.8), 1 mM MgCI 2 and 0.1 mM DTT) (the first Blue Sepharose fraction). The resultant enzyme preparation was applied to a Blue Sepharose CL-6B column, which was equilibrated with buffer B, then washed with 100 ml buffer B and the AK1 was eluted with buffer A containing 0.4 M KCI. The second Blue Sepharose fraction was collected and used as partially purified AK1.

Reconstitution of the thiamin metabolic system in vitro. Each complete reaction mixture (250 /~1) comprised 100 mM Tris-HC1 (pH 7.5), 20 IzM thiamin, 2 mM ATP, 0.2 mM ADP, 2 mM MgCI 2, 1 mM DTT, 46 ng purified TPK and 1.7/zg purified AK1 were incubated at 37°C for various time periods (0-48 h). The reactions were terminated by adding 50/xl 50% TCA, the mixtures were centrifuged, the supernatants were extracted with diethyl ether to remove TCA, and TDP and TTP were analyzed as described above. The proteins were determined by the method of "Lowry et al. [23]. SDS-PAGE. The purity of the enzyme preparation was assessed by SDS-PAGE by the method of Laemmli [24]. The sample protein used was 1 to 2/~g each and a slab gel was composed of 12.5% acrylamide. After electrophoresis, proteins were stained with the 2D silver-stain kit from Daiichi (Tokyo, Japan). Results

Content of thiamin and its phosphates in human blood Table I shows the distribution of thiamin and its phosphates in human blood. Human adult whole blood contained thiamin, TMP, TDP and T-I'P (3.7, 5.3, 62.5 and 2.3 nmol/1, respectively). Both TDP and TTP existed exclusively in the RBC (123.0 and 3.5 nmol/1 packed cell volume, respectively); TDP and TTP were not detected in plasma. TABLE I

Content of thiamin and its phosphate esters in human blood H u m a n blood was freshly collected. Extractions and determinations of thiamin and its phosphate esters from whole blood, plasma and R B C are described in Materials and Methods. The contents are shown as average values_+ S.D. for 8 volunteers. Whole blood (nmol/l)

Red blood cell ( n m o l / l )

Plasma (nmol/l)

TTP TDP TMP Thiamin

2.34__.1.24 62.5 _+8.61 5.30 _+0.78 3.70_+0.99

3.53_+ 1.15 123.02_+ 16.55 0.42 -+ 0.42 1.25_+ 1.29

0 0 9.05 + 2.47 4.96+2.94

Total

73.83 _+8.72

128.22 +_ 16.98

14.01 _+3.16

174

,-,

[ i

A.

16

|

i

i

P

,,,va''': i

0q

0

|

/ 100,

!

i

300

I--

.,e,

•

B.

2

4

?

•

4~

i

~0 ~

4

Ttn~ (h) Fig. 1. Time-dependence of T D P and T T P syntheses by R B C suspension. The R B C suspension was incubated in the presence of 2 0 / ~ M free thiamin (closed symbols) or 20 /~M T T F D (open symbols) for 1-4 h at 37°C as described in Materials and Methods. The contents of thiamin (11, D) (A), T D P (e, o ) and T T P ( A , A) (B) were determined as described in Materials and Methods. A typical result with duplicated m e a s u r e m e n t is shown and the results obtained in three separate experiments were found to be reproducible.

TDP was the main form of thiamin present in whole blood and RBC (85 and 96% of the total thiamin, respectively) and TTP in whole blood and RBC accounted for 3.2 and 2.8% of the total thiamin, respectively. These results support the assumption that human RBC contain enzymes involved in synthesis of TDP and TI'P.

Time-course of thiamin metabolism in human RBC Freshly separated RBC were incubated with 20/zM thiamin and 5 mM glucose in Hepes-buffered physiological saline at 37°C. Thiamin entered the RBC in a time-dependent manner (Fig. 1A) and the amounts of TDP and TTP in the RBC increased after a 4-h incubation period from 112 to 264 and from 0 to 4.1 nmol/l, respectively (Fig. 1B). This result confirms that thiamin was converted to its phosphorylated forms and accumulated in RBC.

When the RBC were incubated with TTFD, a lipidsoluble thiamin derivative that permeates the RBC membrane more rapidly than thiamin, thiamin was accumulated very rapidly in the RBC, reached a peak within 1 h and then declined gradually (Fig. 1A). The intracellular levels of thiamin in RBC incubated with TTFD were several times higher than in those incubated with thiamin. Nevertheless, the RBC contents of TDP and TTP were the same at all assay points after incubation with both compounds (Fig. 1B). These results indicate that the synthetic rates of TDP and T]?P were not limited by the concentration of free thiamin in RBC and, therefore, the rate-limiting step may be the enzyme(s) activities involved in the metabolic pathway of thiamin.

Effect of PT on thiamin uptake and thiamin metabolism in RBC In order to establish whether or not TPK is involved in the synthetic pathway of TDP from thiamin in RBC, PT was added to the RBC suspension, which contained 2 IzM thiamin and 5 mM glucose. The synthesis of TDP was inhibited almost completely by PT (Fig. 2A), which was accompanied by complete inhibition of TTP synthesis (Fig. 2B). The decrease in the total thiamin accumulation in RBC observed in the presence of PT was accounted for quantitatively by the decrease in the TDP content (Fig. 2A). Purification of TPK and AK1 The enzymes TPK and AK1 were purified from a RBC lysate using conventional methods (the steps involved are summarized in Tables II and III). TPK and AK1 were separated effectively at an early step of the purification (DEAE-cellulose chromatography). The result of the final TMP-agarose chromatography is shown in Fig. 3. Fractions containing a single TPK protein (lanes 5-8) were collected for use of kinetic analysis. The specific activity of TPK increased by 3 • 106-fold from the lysate (2.5/xmol TDP formed/mg

T A B L E II

Purification of TPK from human RBC lysate T P K was purified from 2 I of packed h u m a n RBC. Purification and assay of T P K activity were done according to the experimental procedures described in Materials and Methods.

Lysate DEAE-cellulose A m m o n i u m sulfate Sephadex G-200 Acid-treated Hydroxylapatite TMP-agarose

Total protein

T D P formed

(mg)

Total activity (nmol/h)

Specific activity ( n m o l / m g of protein per h)

(%)

Yield

1093 000 3290 661 46.9 41.8 4.3 0.094

951 869 764 423 433 346 235

0.00087 0.264 1.16 9.03 10.4 80.5 2500

100 91.4 80.3 44.5 45.6 36.4 24.7

175 A

800

A.

a

8

B.

l

'

a

2

3

4

5

6

7

8

MW (kDa

600

~6

~: 400

A,*

97.4 66.2

4

45.0 | 200

2

0

•

C 0

2

4

0 Ttl

31.0

4

?

(h)

Fig. 2. Effect of PT on thiamin transmembrane transport and metabolism by RBC. The RBC suspension was incubated with 2 p,M free thiamin in the presence (closed symbols) or absence (open symbols) of 20/zM PT for 1-4 h at 37°C as described in Materials and Methods and the contents of total thiamin (11, U), TDP (o, ©) (A) and TTP ( A, zx) (B) were determined as described in Materials and Methods. A typical result with duplicated measurement is shown and the results obtained in three separate experiments were found to be reproducible.

21.5 14.4 Fig. 3. SDS-PAGE of TPK preparations obtained from TMP-agarose chromatography. TMP-agarose fractions eluted with a linear gradient of 2 to 20 mM ATP in buffer A (lanes 1-9), were separated on 12.5% gel under conditions described in Materials and Methods. The left outer lane shows molecular mass markers of 14.4, 21.5, 31.0, 45.0, 66.2 and 97.4 kDa.

Properties of purified TPK protein per h at 45°C) and the yield was 24.7% (Table II). The activity of AK1 was recovered from the DE52unbound fraction. Subsequent purification by repeated Blue Sepharose chromatography resulted in a 1.3 • 103fold increase in its specific activity for ATP synthesis (84.8 mmol ATP formed/mg protein per h at 25°C) (Table III). The AK1 preparation showed one main band with two minor bands on SDS-PAGE. The qTPsynthesizing activity was co-purified with AK1, which yielded a specific activity of 170 nmol TI'P formed/mg protein per h at 37°C (Table III). These activities of the purified RBC AK1 were comparable to those obtained with chicken AK1 purified to a single protein (90 mmol ATP formed/mg protein per h at 25°C and 125 nmol TTP formed/mg protein per h at 37°C, respectively) [15].

The requirements of the TPK reaction for nucleoside triphosphates and divalent cations are shown in Table IV. ATP and GTP were equally effective as pyrophosphate donors, whereas UTP and ITP were 70% as effective as ATP. Of the divalent cations, Co 2÷ was slightly more effective than Mg 2+ and Mn 2÷ was 50% as effective as Mg 2÷. Although TPK demonstrated fairly broad specificity for nucleoside triphosphates and divalent cations, ATP and Mg 2÷ are assumed to be required for the TPK reaction under physiological conditions in cells. The TPK activity demonstrated a broad plateau at the pH range of 6-9 and the optimal temperature for its activity was 60°C (data not shown). When TPK activity was assayed under the standard assay conditions using varying concentrations of either thiamin or ATP-Mg 2÷, the reaction demonstrated sat-

TABLE III

Purification of AK1 from RBC AK1 was purified from 100 ml of packed human RBC. Purification of AK1 and assay of AK1 activity were done according to the experimental procedures described in Materials and Methods. Total protein (mg)

Lysate DEAE-cellulose Blue Sepharose first fraction second fraction

57000 56 000 1.30 0.0846

Total activity

Specific activity

ATP (mmol/h)

TTP (nmol/h)

382 306

428 366

30.3 7.18

58.0 14.4

ATP (mmol/mg per h) 0.0067 0.0055 23.3 84.8

Yield (%) TTP (nmol/mg per h) 0.00752 0.0066 44.6 170

ATP

TTP

100 80.1

100 85.6

7.90 1.88

13.5 3.37

176 TABLE IV Requirement of nucleoside triphosphates (NTPs) and divalent cations for RBC TPK activity TPK activity was assayed under standard conditions as described in Materials and Methods, except for metals and NTPs. The % activity is given !n parentheses. TDP formed (izmol/mg protein per h)

TDP formed (/~mol/mg protein per h)

Metals

Activity

NTPs

Activity

Mg2+ Co2÷ Mn2+ Zn2÷ Cu2+ Ca2+

1.49 (100) 1.59 (107) 0.75 (50) 0.21 (14) 0 (0) 0 (0)

ATP GTP CTP UTP ITP

1.43 (100) 1.38 (97) 0.34 (24) 1.01 (71) 1.02 (71)

uration kinetics. The K m values for thiamin and A T P were calculated to be 100 nM (standard deviation was 12.5 nM) and 2.5 mM, respectively. This K m value for thiamin was smaller than but essentially in the same range of that of pig brain TPK (430 nM) [14]. The Vmax value of TPK was calculated to be 2.5/zmol T D P / m g protein per h at 45°C (data not shown). The effects of adding a 10-fold molar excess of thiamin analogs to the standard reaction mixture were investigated. PT inhibited the activity of TPK completely and hydroxyethylthiamin inhibited it by 15%. Other thiamin analogs (chloroethylthiamin, oxythiamin and amprolium) exerted no inhibitory effects, whereas TMP inhibited TPK activity completely (data not shown). Sulfhydryl reagents ( p - c h l o r o m e r c u r i b e n z o a t e (0.01-0.5 mM) and N-ethylmaleimide (1-10 mM)) inhibited TPK activity in a concentration-dependent manner. The molecular mass of the purified TPK was calculated to be 56 kDa by gel filtration and a single band with a molecular mass of 28 kDa was detected by S D S - P A G E (Fig. 3), which suggests that this enzyme has a dimeric structure. Properties o f purified AK1 The requirements for the TTP-synthesizing reaction catalyzed by the AK1 preparation were TDP, A D P and

MgCI 2. The optimal temperature for TTP synthesis was 37°C and the specific activity at pH 10 was 2.5-fold higher than that at pH 7.5 (data not shown), which was comparable to the pH response characteristics observed with pig [8] and chicken [15] purified AK1. The TTP-synthesizing activity of the purified AK1 was determined using a standard assay. With varying concentrations of TDP, the K m value for TDP and Vmax value of AK1 were calculated to be 2.1 mM and 402 nmol T T P / m g protein per h at 37°C, respectively. This Vmax value was relatively low compared to the specific activity shown in Table III, which probably ascribed to inactivation of the enzyme. The molecular mass of the purified AK1 was calculated to be 22 kDa by gel filtration and SDS-PAGE (data not shown). Reconstitution of the thiamin metabolic system Purified TPK and AK1 were incubated in the complete reaction mixture, which contained 2 mM ATP, 0.2 mM A D P and 2 mM MgC12, as described in Materials and Methods. The concentrations of TDP and TTP in the reaction mixture increased in a timedependent manner as the incubation period increased (Table V). When TPK was omitted from the reaction mixture, neither T D P nor TTP was detected, whereas when AK1 was omitted T D P was formed in a time-dependent manner, but no TTP was synthesized from TDP. In the absence of the phosphate donors A T P and ADP, neither T D P nor TTP was synthesized. After incubation of the complete reaction mixture for 48 h, 465 p m o l / 2 5 0 / z l (1.68 /xM) T D P and 54.0 p m o l / 2 5 0 /~l (0.216 ~ M ) TTP were formed. These results indicate that thiamin is metabolized to T D P and then to TTP by coupling of TPK and AK1 in vitro and that the AK1 reaction is the rate-limiting step of the overall reaction from thiamin to TTP, although TDP is a biochemically active coenzyme.

Discussion Human whole blood contains thiamin and its phosphate esters and RBC contain T D P and TTP, which were not detected in plasma (Table I).

TABLE V Reconstitution of the thiamin metabolicpathway in vitro The purified TPK and AK1 were incubated at 37°C under the reaction conditions described in Materials and Methods. All values are expressed as pmol formed/250/.d. Time

Complete

(h) 3 24 48

TDP 80.6 257 465

- AK TTP 0 14.3 54.0

TDP 83.8 310 568

- TPK TTP 0 0 0

TDP 0 0 0

- ATP, - ADP TTP 0 0 0

TDP 0 0 0

TTP 0 0 0

177 Kimura and Itokawa [25] assayed thiamin and its phosphate esters in human whole blood using a reversed-phase HPLC method with a post-column derivatization procedure and showed that the total thiamin concentration was 137 +_ 6.7 nmol/l, 70% of which was distributed in TDP with less (30%) in TTP and TMP and thiamin were undetectable. Conflicting results were obtained by Brunnekreeft et al. [26], who reported that the total thiamin mean concentration in human whole blood was 132 nmol/1 and that of TTP was less than 4.0 nmol/1. The results shown in Table I indicate that the total thiamin concentration we obtained (73.8 + 8.7 nmol/1) is somewhat lower than those cited above, which probably is due to the 12-h period of starvation before blood sampling used in our study. The TI'P content relative to that of total thiamin was calculated to be 3.2%, which is consistent with that reported by Brunnekreeft et al. (3.0%) [26]. Figs. 1 and 2 show clearly that thiamin was metabolized first to TDP and then to TTP in RBC, which suggests the involvement of at least one known enzyme, i.e., TPK. This is supported by the observation that PT, an antimetabolite of thiamin, suppressed the conversion of thiamin to TDP in RBC completely (Fig. 2A). The results obtained with PT also indicate that newly synthesized TDP from thiamin was phosphorylated to TTP in RBC, because PT also suppressed the increase in the level of TTP (Fig. 2B). In mammals, TPK is the only enzyme known to synthesize TDP from thiamin. TPK activity has been detected in human leukocytes, but not in a RBC lysate by a radiometric assay [27]. The activity of TPK in RBC has been shown to be lower in patients with thiaminresponsive megaloblastic anemia than in normal controls [28]. In order to elucidate the pathophysiological role of TPK in such metabolic disorders, the isolation, purification and characterization of human TPK is essential. In this paper, we report the isolation and purification of TPK from a human RBC lysate. The low specificity of the purified TPK for nucleoside triphosphates and divalent metal ions we observed (Table IV) is similar to that of the enzyme from pig heart [13], pig brain [14] and yeast [29]. Based on the normal cellular concentrations of nucleoside triphosphates and metal ions, TPK in human RBC would appear to utilize ATP and Mg 2+ physiologically. The high affinity of the purified TPK for thiamin is suitable to enable thiamin to be accumulated efficiently as TDP in RBC under conditions of a relatively low thiamin intake. The K m value for ATP determined fell within its physiological concentration range. Recently, we demonstrated that AK1 from pig [8] and chicken [10] skeletal muscles and recombinant chicken AK1 [15,16] catalyzed the synthesis of TFP

from TDP and ADP in the presence of Mg 2+. Human RBC AK1 has been well characterized [17] and its complete amino-acid sequence has been determined [18], although the TTP-synthesizing activity of RBC AK1 had not been demonstrated. The partially purified RBC AK1 preparation (Table Ill) clearly demonstrated a TTP-synthesizing activity with similar properties to the enzyme obtained from pig [8] and chicken [15,16] skeletal muscle. In order to establish thiamin metabolic pathway in RBC, the TTP-synthesizing system was reconstituted using the purified enzymes, TPK and AK1 (Table V). Both TDP and TTP were synthesized in the complete reaction mixture in a time-dependent manner. The conversion ratio of thiamin to TDP was 8.3% and that of TDP to TTP 11.8% after incubation for 48 h, which indicates that the overall conversion ratio of thiamin to TTP was of the order of 1%. These results prove that the TPK and AK1 enzymes are involved in thiamin metabolism in human RBC. Under physiological conditions, the main product of this thiamin metabolic pathway in RBC is TDP, which is an essential coenzyme for carbohydrate metabolism. The physiological role of TYP, the final product of the pathway, should be investigated further.

References 1 Gubler, C.J. (1984) in Handbook of Vitamins (Machlin, L.J., ed.), pp. 245-279, Marcel Dekker, New York. 2 Bisswanger, H. and Ullrich, J. (eds.) (1991) Biochemistry and Physiology of Thiamin Diphosphate Enzymes, pp. 1-453, VCH, Weinheim. 3 Rindi, G. and De Guiseppe, L. (1961) Biochem. J. 78, 602-606. 4 Ishii, K., Sarai, K., Sanemori, H. and Kawasaki, T. (1979) J. Nutr. Sci. Vitaminol. 25, 517-523. 5 Egi, Y., Koyama, S., Shikata, H. Yamada, K. and Kawasaki, T. (1986) Biochem. Int. 12, 385-390. 6 Kawasaki, T. (1986) Methods Enzymol. 122, 15-20. 7 Matsuda, T., Tonomura, H., Baba, A. and Iwata, H., (1989) Comp. Biochem. Physiol. 94B, 405-409. 8 Shikata, H., Egi, Y., Koyama, S., Yamada, K. and Kawasaki, T. (1989) Biochem. Int. 18, 943-949. 9 Shikata, H., Egi, Y., Koyama, S., Yamada, K. and Kawasaki, T. (1989) Biochem. Int. 18, 933-942. 10 Miyoshi, K., Egi, Y., Shioda, T. and Kawasaki, T., (1990) J. Biochem. 108, 267-270. 11 Kawasaki T. (1992) in Modern Chromatographic Analysis of the Vitamins (De Leenheer, A., Lambert, W. and Nelis, H., eds.), 2nd Edn., pp. 319-354, Marcel Dekker, New York. 12 Mano, Y. (1960) J. Biochem. 47, 283-289. 13 Hamada, M. (1969) Seikagaku 41,837-849. 14 Wakabayashi, Y. (1978) Vitamins 52, 223-236. 15 Shioda, T., Egi, Y., Yamada, K. and Kawasaki, T. (1991) Bioehim. Biophys. Acta 1115, 30-35. 16 Shioda, T., Egi, Y., Yamada, K., Yamada, M., Nakazawa, A. and Kawasaki, T. (1991) Biochim. Biophys. Acta 1115, 36-41. 17 Hamada, M., Sumida, M., Kurokawa, Y., Sunayashiki-Kusuzaki, K., Okuda, H., Watanabe, T. and Kuby, S.A. (1985) J. Biol. Chem. 260, 11595-11602.

178 18 Matsuura, S., Igarashi, M., Tanizawa, Y., Yamada, M., Kishi, F., Kajii, T., Fujii, H., Miwa, S., Sakurai, M. and Nakazawa, A. (1989) J. Biol. Chem. 264, 10148-10155. 19 Wakabayashi, Y., Iwashima, A. and Nose, Y. (1979) Methods Enzymol. 62, 105-107. 20 Koyama, S., Egi, Y., Shikata, H., Yamada, K. and Kawasaki, T. (1985) Biochem. Int. 11,371-380. 21 Sugino, K., Yamada, K. and Kawasaki, T. (1986) J. Chromatogr. 361,427-431. 22 Nealon, D.A. (1986) Arch. Biochem. Biophys. 250, 19-22. 23 Lowry, O.H., Rosebrough, N.J., Farr. A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275.

24 Laemmli, U.K. (1970) Nature 227, 680-685. 25 Kimura, M. and Itokawa, Y. (1985) J. Chromatogr. 332, 181-188. 26 Brunnekreeft, J.W.I., Eidhof, H. and Gerrits, J. (1989) J. Chromatogr. 491, 89-96. 27 lwashima, A., Kinugasa, A. and Nose, Y. (1983) Acta Vitaminol. Enzymol. 5, 41-45. 28 Poggi, V., Rindi, G., Patrini, C., De Vizia, B., Longo, G. and Andria, G. (1989) Eur. J. Pediatr. 148, 307-311. 29 Kaziro, Y. (1959) J. Biochem. 46, 1587-1596.