Biochimica et Biophysica Acta, 1119(1992) 218-224 © 1992 Elsevier Science Publishers B.V. All rights reserved 11167-4838/92/$05.00

218

BBAPRO 34118

Subtmit IV cff human cytochrome c oxidase, polymorphism and a putative isoform Andr6 B.P. Van Kuilenburg 1, Jozef J. Van Beeumen 2, Hans Demol 2, Coby Van den Bogert ~, Ingrid Schouten 1 and Anton O. Muijsers 1 l E.C. Slater Institute for Biochemical Research, Unil'ersity of Amsterdam, Amsterdam (The Netherlands) and 2 Laboratory of Microbiology, Unit'ersity of Gent, Gent (Belgium) (Received 27 September 1991)

Key words: Cytochrome c oxidase; Enzyme i~aform; Subunit IV; Polymorphism

As part of our study of isoenzyme forms of human cytochrome c oxidase, we purified subunit IV from human heart and skeletal muscle with reversed.phase HPLC and determined the N-terminal amino acid sequences and the electrophoretic mobility. The N-terminus of human heart subunit IV proved to be ragged with 30% of the protein lacking the first three residues. Also a T y r / P h e polymorphism was observed at residue 16. No differences in N-terminal sequence and electrophoretic mobility were observed between subunit IV of cytochrome c oxidase from human heart and skeletal muscle. Therefore, our results suggest that identical subunits IV are present in cytochrome c oxidase from human heart and skeletal muscle. A putative isoform of subunit IV with a blocked N-terminus was purified from human heart cytochrome c oxidase, which proved to have a different retention time on a reversed-phase column and also a slightly higher electrophoretic mobility on an SDS-polyacrylamide gel compared to the native subunit IV. We could not demonstrate the existence of isoforms of subunit IV in human skeletal muscle.

Introduction

Cytochrome c oxidase (EC 1.9.3.1) is a multi-subunit protein that catalyses the oxidation of ferrocytochrome c by molecular oxygen and contributes to the establishment of a proton gradient across the mitochondrial inner membrane. In eukaryotes, three subunits (I-III) of cytochrome c oxidase are encoded by the mitochondrial genome [1] and constitute the functional core of the enzyme [2]. Subunits I and I1 contain the four prosthetic groups [3-5], whereas subunit III is supposed to be involved in the proton-translocating function of the enzyme [6-8]. In addition, eukaryotic cytochrome c oxidase contains several nuclearly encoded subunits, the number depending on the source of the enzyme. In Saccha-

Abbreviations: TEA. triethylamipe; TFA, trifluoroacetic acid. Correspondence: Dr. A.O. Muijsers, E.C. Slater Institute for Biochemical Research, University of Amsterdam, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands.

romyces cerevisiae, at least six dissimilar nuclearly encoded subunits have been identified [9], and 10 in mammalian cytochrome c oxidase [10]. The presence of isoforms of some of the nuclearly encoded subunits has been demonstrated for yeast [11] and mammalian cytochrome c oxidase [12,13]. For some of the nuclearly encoded subunits of mammalian cytochrome c oxidase, tissue-specific immunological and electrophoretic differences have been demonstrated in cytochrome c oxidase purified from several tissues of the same mammalian species [12,13]. The evidence for isoenzyme forms of bovine cytochrome c oxidase was confirmed by N-terminal amino acid sequence data, which revealed that different 'heart-type' and 'liver-type' isoforms of the subunits Via, Vlla and VIII were present in cytochrome c oxidase purified from bovine heart and liver tissue, respectively [13]. It has been proposed that the different isoforms of the subunits act as modulators conferring different activities to the various bovine cytochrome c oxidase isoenzymes [14]. However, attempts to demonstrate these differences in electrontransfer activity of the various bovine isoenzymes have

219 led to contradictory results so far [15]. The question of the function of the l0 smaller, nuclearly encoded subunits of the mammalian oxidase remains as yet unsettled. Studies on the subunit composition and isoform pattern of human cytochrome c oxidase revealed that the pattern of expression of the isoforms found in bovine cytochrome c oxidase was not conserved between mammalian species [16-19]. Determination of the complete primary structure of subunit VIII of human heart cytochrome c oxidase [16] and gel electrophoresis of immunopurified cytochrome c oxidase preparations from various human tissues [18] suggested that no isoforms of subunit VIII exist. Although these gel electrophoretic studies of immunopurified cytochrome c oxidase suggested that isoforms of subunit Via existed [18], firm evidence for the existence of isoforms has only been provided for subunit Vlla [17]. The identification at the polypeptide level of two different isoforms of subunit VIIa in a preparation of cytochrome c oxidase from human skeletal muscle showed that the human isoenzymes are not strictly tissue-specific [17]. So far, no sound evidence has been presented for the occurrence of isoforms of subunit IV in mammalian cytochrome c oxidase. Immunological differences between subunit IV of cytochrome c oxidase from fetal and adult rats suggested the occurrolce of fetal and adult isoforms of subunit IV [12]. In cows, subunit IV from heart and liver tissue appeared to be identical [13], whereas recent immunohistochemical data suggest that fibre-type specific isoforms of subunit IV exist in human skeletal muscle [20]. Subunit V is the only subunit of the yeast enzyme of which isoforms exist. Yeast subunit V shows sequence homology to subunit IV of mammalian cytochrome c oxidase [11]. Recently, it has been shown that these isoforms o~ yeast subunit V, designated as Va and Vb, modulate the catalytic properties of yeast cytochrome c oxidase in vivo by altering the rate of one or more intramolecular electron transfer reactions [21]. In order to investigate whether or not isoforms of subunit IV exist in human cytochrome c oxidase, we determined the N-terminal amino acid sequences of subunit IV of cytochrome c oxidase from human heart and skeletal muscle. Our study revealed that an identical subunit IV protein was present in human heart and skeletal muscle tissue. A putative isoform of subunit IV was found in cytochrome c oxidase from human heart. Materials and Methods

pooled skeletal muscle, pooled hearts or from a single heart, each of which had become available within 24 h post mortem. Human cytochrome c oxidase was purified as described before [16], essentially by differential extraction of submitochondrial particles with deoxycholate. Bovine heart cytochrome c oxidase was purified according to the method of Fowler [22], as modified in our laboratory [23]. The nuclearly encoded subunits of human cytochrome c oxidase were separated from those encoded mitochondrially using the method of Power et al. [24] with the following modifications: after preparing supernatant S1, the final pellet (P1) was resuspended in 1.25% triethylamine (TEA) and 1.25% trifluoroacetic acid (TFA). 1 vol. of the mixed solvent acetonitrile/1-propanol (1 : 1, 0.05% TEA, 0.05% TFA) was added to the suspension and the mixture was stirred on ice for 3 h. After centrifugation, the supernatant was retained and the pellet extracted twice more. The final extraction was carried out overnight. All supernatants were pooled and designated supernatant $2. Supernatant $2 was lyophilized and saved for further analysis.

Purification of subunit IV with recersed-phase HPLC Subunit IV was purified from supernatant $2. An aliquot of supernatant $2 was dissolved in 8 M guanidine HCI/0.05% TFA and incubated for 1 h at room temperature. Dithiothreitol (10 mM) was added 15 min prior to injection in the HPLC system. HPLC was performed on a Brownlee (Santa Clara, CA) column (type RP-300, Aquapore octyl, 300 ,~ size, 7 ~m spherical, 220 × 4.6 nm) and a guard column (type RP-8 Aquapore ODS, 7 ~m spherical). The HPLC system consisted of two LKB 2150 pumps, a 2152 LC controller, a Rheodyne 7125 injector and an LKB Uvicord 2138 S detector operating at 280 nm. The solvents used for chromatography consisted of 0.05% TFA/0.05% T E A / 5 % acetonitrile/water (solvent A) and 0.05% TFA/0.05% T E A / 5 0 % acetonitrile/50% 1-propanol (solvent B). Subunit IV of human cytochrome c oxidase purified from a single heart could not be purified from supernatant $2 in a single chromatographic run. The subunit IV preparation obtained from supernatant $2 was lyophilized and dissolved in 8 M guanidine HCl/0.05% TFA. Dithiothreitoi (10 mM) was added 15 min prior to injection in the HPLC system. The solvents used for chromatography consisted of 0.05% TFA/0.05% T E A / 5 % acetonitrile/water (solvent A) and 0.05% TFA/0.05% TEA in acetonitrile ~solvent B).

Purification of cytochrome c oxidase and fractionation of the small subunits

Polyacrylamide gel electrophoresis

Human hearts and skeletal muscle (quadriceps of mixed fibre-types) were obtained at obduction. Batches of human cytochrome c oxidase were purified from

Tricine-SDS gel electrophoresis was performed according to Schiigger and Von Jagow [25], gels (30 × 18 ×0.15 cm) were prepared with glycerol instead of

220 urea. The samples were incubated at room temperature for 20 h in sample buffer containing 4% SDS, 20% glycerol in 10 mM Tris-HCl (pH 6.8) and 1%/3-mercaptoethanol. We added up to 2% fl-mercaptoethanol 30 rain prior to the start of the gel-electrophoresis run. Gels were either stained overnight with Coomassie brilliant blue or directly during gel electrophoresis [26], as indicated in the figure legends. Silver staining was performed according to Wray et al. [27].

Determination of the N-terminal amino acid sequences The amino acid sequence analysis was carried out on a 477A pulsed-liquid sequencer (Applied Biosystems, U.S.A.) with on-line analysis of the phenylthiohydantoin amino acids on a 120A PTH analyzer (Applied Biosystems).

Tg"

'~ 0192[

~

N ] '~ 0128~co

I

"

/

,

~bO •

z 0016

20.

~ 0008

~c 006t.

01""/, 0-

,

,

gO 6O

-- 002&

It.i . . . . .

!

B

/0

oi

20 26 32 RETENTION TIME(minl

*

~0

3'z

36

RETENTION TIHE(min]

Results

Fig. 1. Purification of subunit IV of cytochrome c oxidase from human heart and human skeletal muscle.The 280 nm absorbance of parts of the elution profiles of supernatant S2 obtained from cytochrome c oxidase from human heart (A) and human skeletal muscle (B) is given by a solid line. The dashed line represents the percentage of solvent B in the gradient. The solvents used for chromatography consisted of 0.05% TFA/0.05% TEA/5% acetonitrile/water (solvent A) and 0.05% TFA/0.05% TEA/50% acetonitrile/50% l-propanol (solventB).

In order to purify subunit IV of cytochrome c oxidase from human heart and skeletal muscle with reversed-phase HPLC, we separated the nuclearly encoded subunits from those encoded mitochondrially by first incubating the holoenzyme with acetonitrile at neutral pH followed by extraction with an acetonitrile/ 1-propanol mixture at acidic pH (supernatant $2). Subunit IV was further purified from supernatant $2. Part of the elution profiles of supernatant $2 are shown in Fig. 1. Both subunit IV of cytochrome c oxidase from

human heart and skeletal muscle eluted as a large peak at approx. 38% solvent B. The purity of the polypeptides was tested by SDS-gel electrophoresis using cytochrome c oxidase from bovine heart, human heart and human skeletal muscle as references (Fig. 2). The bands were numbered according to the system of Kadenbach et al. [28]. The left-hand panel of Fig. 2 (Coomassie staining) shows the distinct electrophoretic mobility of the subunits Vlla, Vllb and

T--~

. . . . .~IIT-. ~

i

--~]][°'~' . . . . . .

--I

lg--~

--lg

~[~i

---Ya,b

YlTT-~-~

~_V.Ea, ---VIII" b,¢

Fig. 2. Gel electrophoresis of the purified subunit IV of cytochrome c oxidase from human heart and skeletal muscle. Cytochrome c oxidase from human heart, human skeletal muscle and bovine heart were used as references. The subunit nomenclature of Kadenbach was used [28]. SD$-glycerol gel electrophoresis was performed according to Seh~gger and Von Jagow [25]. Left-hand part: the gel was directly stained with Coomassie brilliant blue during the gel electrophoresis as described in Ref. 26. Additional Coomassie staining of the gel was performed overnight. Right-hand part: silver staining according to Wray et al. [27].

221

VIlc of human and bovine cytochrome c oxidase, indicating the high resolution that is obtained using this gel system• Both Coomassie staining (left-hand part) and the highly sensitive silver staining (right-hand part) of the gels showed that the purified subunit IV of cytochrome c oxidase from human skeletal muscle was not contaminated with other polypeptides. The analysis by gel electrophore~sis of the human heart subunit IV revealed an additional minor band with a slightly higher electrophoretic mobility• Under our conditions, tl e additional polypeptide present in the human heart subunit IV preparation always co-eluted with the native subunit IV during the reversed-phase column chromatography. No differer~cc in electrophoretic mobility was observed between subunit IV of cytochrome c oxidase from human hear~ and that from human skeletal muscle. In Fig. 3 the N-terminal amino acid sequences of the subunits IV from human heart and skeletal muscle

Human Human Human Human Human Bovine

heart heart heart(I) muscle l i v e r (cDHA) heart

Human

heart

Human h e a r t Human h e a r t (I) Human muscle H u m a n l i v e r (cDNA) Bovine heart Human Human Human Human Human

heart heart h e a r t (I) muscle l i v e r (cDNA) Bovine h e a r t

are presented together with the corresponding part of the sequence of subunit IV from human liver [29] and bovine heart [30]. The sequence given for human liver subunit IV was derived from the c D N A sequence. Amino acid sequence analysis allowed unambigiuous identification of 50 residues of the N-terminal amino acid sequence of subunit IV of cytochrome c oxidase from h u m a n muscle. The sequence proved to be completely identical to the amino acid sequence derived from a c D N A specifying a presumed human liver subunit IV [29]. The amino acid sequence analysis of human heart subunit IV was followed up to His-29. The exact nature of the residues 26 and 30 was not clear. The N-terminal amino acid sequence of h u m a n heart subunit IV was identical to that determined for the h u m a n muscle protein, except at cycle 16 where both Tyr and Phe were identified. Approx. 2 / 3 of the protein contained a Tyr, whereas the remaining 1 / 3 of the protein contained a Phe at position 16. Further-

1 5 10 Ala-Bis-Glu-Ser-Val-Val-Lys-Ser-Glu-Asp-I - ........... Ser-Va1-Val-Lys-Ser-Glu-Asp-J Ala-Bis-Glu-Ser-Val-Val-Lys-Ser-Glu-Asp-~ Ala-His-Glu-Ser-Val-Val-Lys-Ser-Glu-Asp-~ &la-His-Glu-Ser-Val-Val-Lys-Ser-Glu-Asp-J Ala-Bis~Ser-Val-Va1-Lys-Ser-Glu-Asp-J Ii 15 20 Phe-Ser-Leu-Pro-Ala-Tyr-Met-Asp-Arg-Arg• Phe Phe-Ser-Leu-Pro-Ala-Tyr-Met-Asp-Arg-ArgPhe Phe-Ser-Leu-Pro-Ala-Ty~-Met-Asp-Arg-ArgPhe-Ser-Leu-Pro-Ala-TFr-Met-Asp-Arg-ArgPhe-Ser-Leu-Pro-Ala-Tyr-~6t~Asp-Arg-ArgTFr-Ala~Leu-Pro~Ser~Tyr~ValSAsp-&rg-Arg21 " 25 30 Asp-His-Pro-Leu-Pro-(Glu)-Val-Ala-His-(Val)Asp-Bis-Pro-Leu-Pro-(Glu)-Val-Ala-His-(Val)Asp-His-Pro-Leu-. ........................... Asp-Bis-Pro-Leu-ProGlu- Val-Ala-His- Val Asp-His-Pro-Leu-ProGlu- Val-Ala-His- Val Asp-Tyr~Pro-Leu-Prof--~ V a l - A l a - B i s - Val -

I

31

35

40

Ht~an heart Human h e a r t Human h e a r t ( I )

I.......................................

Human

Lys-His-Leu-Ser-Ala-Ser-Gln-Lys-&la-LeulLys-BisoLeu-Ser-Ala-Ser-Gln-Lys-Ala-LeuL[.L]~-Asn~Leu-Ser-Ala-Ser-Gln-Lys-Ala-Leu-

muscle l i v e r (cDNA) Bovine h e a r t Human

41 Human Human Human Human Human Bovine

heart heart hearttI) muscle l i v e r (cDNA) heart

45

50

I

LFs-Glu-Lys-Glu-Lys-&la-Ser-Trp-Ser-Ser-

Fig. 3. The N-terminal amino acid sequence of subunit IV of cytochrome c oxidase purified from human heart and human skeletal muscle. The N-terminal amino acid sequence of subunit IV from human heart and skeletal muscle oxidase as determined by us, was compared with that from bovine heart [30], and human liver [29]. Cytochrome c oxidase purified from a single human heart was designated as H.heart(1). The partial sequence given for human liver subunit IV was derived from the eDNA sequence. Identical residues in the sequence of subunit IV of cytochrome c oxidase from human heart, human skeletal muscle, human liver and bovine heart have been boxed. Parts of the. sequences that have not yet been determined are indicated by dotted lines. The ragged N-terminus found for the human heart subunit IV is represented by a dashed line. The exact nature of the residues 26 and 30 was not clear and they were therefore placed in parentheses.

222 more, the N-terminus of the human heart protein proved to be ragged: approx. 1/3 of this protein lacked the first three residues at the N-terminus and therefore the sequence began with Ser-Val-Val-. Surprisingly, there was also evidence for small amounts of subunit IV with N-terminal sequences starting at Ser-8 and Ser-12. To investigate whether or not the sequence heterogeneity found at residue 16 represents a naturally occurring polymorphism, we purified subunit IV from cytochrome c oxidase isolated from a single human heart. Part of the elution profile of supernatant $2 and the analysis by gel electrophoresis of two peak fractions obtained from this supernatant $2 are shown in Fig. 4. Cytochrome c oxidase purified from pooled human hearts and from a single human heart were used as references. Subunit IV, which again eluted around 38% solvent B (peak 1), could not be purified in a single chromatographic run on a reversed-phase column. Analysis by gel electrophoresis showed that it was still contaminated with other proteins. From supernatant $2 we could also purify another protein (peak 2), which had a different retention time on a reversed-phase column as well as a slightly higher electrophoretic mobility on SDS-polyacrylamide gels than subunit IV. The N-terminal amino acid sequence of this polypeptide, which might be an isoform of subunit IV,

i

I

I

I

1

006

60

t

E

--r

i

1 2 ~

O

~OOt~ M.I

40

z