Tài liệu Báo cáo khoa học: Purified RPE65 shows isomerohydrolase activity after reassociation with a phospholipid membrane pdf

Purified RPE65 shows isomerohydrolase activity after
reassociation with a phospholipid membrane
Olga Nikolaeva, Yusuke Takahashi, Gennadiy Moiseyev and Jian-xing Ma
Departments of Cell Biology and Medicine Endocrinology, Harold Hamm Oklahoma Diabetes Center, University of Oklahoma Health
Sciences Center, OK, USA

Keywords
isomerohydrolase; liposome; retina; retinyl
ester; RPE65
Correspondence
G. Moiseyev, Departments of Cell Biology
and Medicine Endocrinology, Harold Hamm
Oklahoma Diabetes Center, University of
Oklahoma Health Sciences Center, 941
Stanton L. Young blvd, BSEB 302,
Oklahoma City, OK 73104, USA
Fax: +1 405 271 3973
Tel: +1 405 2718001 (ext. 48443)
E-mail:
(Received 3 February 2009, revised 23
March 2009, accepted 24 March 2009)

Generation of 11-cis-retinol from all-trans-retinyl ester in the retinal
pigment epithelium is a critical step in the visual cycle and is essential for
perception of light. Recent findings from cell culture models suggest that
protein RPE65 is the retinoid isomerohydrolase that catalyzes the reaction.
However, previous attempts to detect the enzymatic activity of purified
RPE65 were unsuccessful, and thus its enzymatic function remains controversial. Here, we developed a novel liposome-based assay for isomerohydrolase activity. The results showed that purified recombinant chicken
RPE65 had a high affinity for all-trans-retinyl palmitate-containing liposomes and demonstrated a robust isomerohydrolase activity. Furthermore,
we found that all-trans-retinyl ester must be incorporated into the phospholipid membrane to serve as a substrate for isomerohydrolase. This assay
system using purified RPE65 enabled us to measure kinetic parameters for

the enzymatic reaction catalyzed by RPE65. These results provide
conclusive evidence that RPE65 is the isomerohydrolase of the visual cycle.

doi:10.1111/j.1742-4658.2009.07021.x

In vertebrates, both rod and cone visual pigments
require 11-cis-retinal as a chromophore [1]. Upon
absorption of photon, 11-cis-retinal is photoisomerized
to all-trans-retinal, which triggers the conformational
change of opsin and subsequently activates the G-protein transducin and initiates vision [2,3]. The process
of recycling 11-cis-retinal, termed the visual cycle
(Fig. 1), is essential for the regeneration of visual
pigments [4,5]. All-trans-retinal generated by photoactivation is dissociated from opsin and converted to
all-trans-retinol by retinol dehydrogenase [6]. The
all-trans-retinol is then exported from photoreceptors
to the retinal pigment epithelium (RPE), and all-transretinol is esterified by lecithin:retinol acyltransferase
(LRAT) to all-trans-retinyl esters [7]. The key enzyme
of the visual cycle, isomerohydrolase (EC 5.2.1.7),
processes all-trans-retinyl esters into 11-cis-retinol [8].

It has been proposed that the free energy generated
from ester hydrolysis is probably used by the enzyme
to drive a thermodynamically uphill trans–cis isomerization of the retinoid double bond [9]. The chemical
nature of the isomerohydrolase has been undetermined
thus far.
RPE65 is a membrane-associated protein expressed
predominantly in the RPE [10]. The molecular mass of
bovine RPE65 as determined by MS is 61 961 Da; this
is higher than its calculated value (60 944 Da) based
on the derived amino acid sequence [11], indicating

post-translational modifications [12]. Hydropathy analysis of the RPE65 amino acid sequence revealed no
obvious hydrophobic transmembrane domains [10]. In
Rpe65) ⁄ ) mice, it has been shown that 11-cis-retinoids
are absent in the retina, and rhodopsin regeneration is
thus impaired, suggesting that RPE65 is essential for

Abbreviations
Ad-RPE65, adenovirus expressing RPE65; LRAT, lecithin:retinol acyltransferase; MOI, multiplicity of infection; PC, phosphatidylcholine; RPE,
retinal pigment epithelium.

3020

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS


O. Nikolaeva et al.

Isomerohydrolase activity of purified RPE65

Fig. 1. Scheme of retinoid visual cycle.

visual pigment regeneration in vivo [13]. Mutations in
the RPE65 gene have been linked to Leber’s congenital
amaurosis, which is an inherited disease characterized
by blindness at birth [14,15]. Recently, we and two
other groups reported that isomerohydrolase activity
was detected in cultured cells that coexpress both
RPE65 and LRAT, suggesting that RPE65 is the isomerohydrolase [16–18]. However, as isomerohydrolase
activity has never been shown using purified RPE65,
there is skepticism about whether RPE65 is indeed the

isomerohydrolase [19]. Two groups have reported independently that purified RPE65 is a retinyl ester-binding
protein [20,21]. These studies led to speculation that
RPE65 is not the isomerohydrolase itself, but rather
that it is required to present the insoluble retinyl ester
to some ubiquitous isomerohydrolase [20].
In this study, we purified recombinant chicken
RPE65 to apparent homogeneity and demonstrated its
isomerohydrolase activity, exploiting a novel enzymatic
assay system that utilizes all-trans-retinyl palmitate
incorporated into liposomes. Chicken RPE65 was
selected as the ideal homolog for this study, because
we have shown previously that chicken RPE65 has
higher expression levels than the human homolog and
higher isomerohydrolase activity than both the human
and the bovine homologs [22]. Using this system, we
have performed a kinetic analysis of the enzymatic
activity of purified RPE65.

Results
Expression and solubilization of recombinant
RPE65 with isomerohydrolase activity
To optimize the expression of chicken RPE65, 293ALRAT cells were infected with different titers of adenovirus expressing RPE65 (Ad-RPE65) [multiplicity of
infection (MOI) 5–500] and harvested 24 h after infection. The cells were disrupted by sonication, and
RPE65 was solubilized using Chaps. RPE65 expression
levels were evaluated by western blot analysis, using the
same amount (20 lg) of either total cellular protein
(Fig. 2A) or the Chaps-soluble fraction (Fig. 2B). As
shown by western blot analysis, RPE65 expression
levels increased with MOI, and reached a plateau at
MOI 150–500 both in total cell homogenates and in the

Chaps-soluble fractions. The cells expressing RPE65
were treated with different concentrations of Chaps to
determine the optimal amount for solubilizing RPE65.
As shown by western blot analysis, Chaps at concentrations of 0.1–0.5% solubilized significant amounts of
recombinant RPE65 in the cells, whereas lower concentrations of the detergent did not adequately solubilize
RPE65 from the membrane (Fig. 2C).
We also determined the effect of increasing Chaps
concentrations on the enzymatic activity of RPE65.
For these measurements, a novel enzymatic activity

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS

3021


Isomerohydrolase activity of purified RPE65

O. Nikolaeva et al.

A

D

80
60

RPE65

50

β-actin
50
10
0
15
0
20
0
30
0
50
0
BM
F

5

25

0

40

A320 (1 × 10–2 AU)

Total cell lysate

kDa

3.0

0% CHAPS
1

2.0
3

1.0
2
0
0

5

10

MOI

15

20

25

Time (min)

E
CHAPS soluble fraction

80

60

RPE65

50
β-actin
0
15
0
20
0
30
0
50
0

3.0

1

0.5% CHAPS

2.0
1.0
2

3

0

10

50

25

5

0

40

A320 (1 × 10–2 AU)

B
kDa

0

5

MOI

10

15

20

25

Time (min)

F

80
RPE65

60

50

β-actin
5
0.

3
0.

1
0.

01
0.

0.

00

1

0

40

11-cis retinol (pmol)

C
kDa

600
500
400
300
200
100
0
0

CHAPS (%)

0.1

0.2

0.3

0.4

0.5

0.6

CHAPS (%)

Fig. 2. Optimization of expression and solubilization of recombinant RPE65. (A, B) The 293A-LRAT cells were infected with Ad-RPE65 with
increasing MOI, and harvested at 24 h after infection. Equal amounts (20 lg) of proteins from total cell lysates (A) and the Chaps (0.1%)-solubilized supernatant after ultracentrifugation (B) were analyzed by western blot analysis using an antibody specific for RPE65, and normalized
by b-actin levels. Proteins of the bovine RPE microsomal fraction (1 lg) were included as a control. (C) To determine the effects of Chaps
concentration on RPE65 solubility, the cells were infected with Ad-RPE65 at MOI 100, and harvested 24 h after infection. The cell lysates
were incubated with increasing concentrations of Chaps for 1 h at 4 °C, and then centrifuged at 200 000 g for 30 min. Equal amounts (2 lg)
of total proteins from the supernatant fractions were blotted with antibody against RPE65. (D, E) The effect of Chaps concentration on the
isomerohydrolase activity of RPE65 was evaluated using in vitro isomerohydrolase assays. Liposomes preloaded with the all-trans-retinyl
ester (50 lM lipids, 0.66 lM all-trans-retinyl palmitate) were incubated with 500 lg of total proteins from the cells expressing RPE65 in the
presence of 0% (D) or 0.5% (E) Chaps for 2 h. The generated retinoids were analyzed by HPLC, and peaks were identified as follows: 1, retinyl esters; 2, all-trans-retinal; and 3, 11-cis-retinol. (F) Dependence of the isomerohydrolase activity on Chaps concentration was measured
for total cell lysates (4) and Chaps-soluble fractions ( ) and plotted. The activity was calculated from the peak areas of the generated
11-cis-retinol in HPLC profiles (mean ± standard deviation, n = 3).

assay with liposomes containing all-trans-retinyl ester
was developed (see Experimental procedures). In the
absence of Chaps, incubation of the total cell homogenate expressing RPE65 with all-trans-retinyl palmitate
incorporated into liposomes generated a significant
amount of 11-cis-retinol (Fig. 2D). The addition of
0.5% Chaps to the assay system almost completely
abolished the 11-cis-retinol formation (Fig. 2E).
To define the Chaps concentration that sufficiently
solubilizes RPE65 while preserving its enzymatic activity, we measured the dependence of isomerohydrolase
activity on Chaps concentration, both for total 293A-

3022

LRAT cell homogenates expressing RPE65 and for
Chaps-solubilized fractions (Fig. 2F). For total cell
lysates, the production of 11-cis-retinol gradually
decreased with increasing Chaps concentrations. When
the Chaps-soluble fractions were used for the isomerohydrolase assay, an initial plateau of enzymatic activity
was observed up to 0.1% of Chaps. In line with previous data [20], 11-cis-retinol generation was drastically
decreased by 0.3% Chaps (Fig. 2F). Taken together,
these results suggest that 0.1% Chaps is optimal for
solubilizing RPE65 while preserving its enzymatic
activity, and this concentration was therefore employed

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS


O. Nikolaeva et al.

for the following RPE65 purification and enzymatic
assays.

Isomerohydrolase activity of purified RPE65

A kDa

Purification of recombinant RPE65
To facilitate the purification of recombinant chicken
RPE65, a histidine-hexamer tag (6 · His) was fused to
the N-terminus of RPE65 and expressed using
Ad-RPE65 at MOI 500. The recombinant RPE65 was
solubilized with 0.1% Chaps and purified through an
Ni2+–nitrilotriacetic acid column. The purified RPE65

appeared to be homogeneous, as shown by
SDS ⁄ PAGE followed by Coomassie Brilliant Blue
staining (Fig. 3A). The identity of the purified RPE65
was confirmed by western blot analysis, using antibodies against RPE65 (Fig. 3B) and the His-tag (Fig. 3C).

Purified RPE65 showed isomerohydrolase
activity that was dependent upon association
with liposomes
Although all-trans-retinyl ester has been established as
the substrate of the isomerohydrolase [24], the poor
solubility of hydrophobic all-trans-retinyl ester has historically hindered its use as a substrate for assays of
isomerohydrolase activity. In this study, a novel isomerohydrolase activity assay was developed in which alltrans-retinyl ester was incorporated into liposomes,
and all-trans-retinyl ester-containing liposomes were
then used as the substrate for measuring the isomero-

2

3

4

5

RPE65

50
37
25
20

B kDa

1

2

3

4

5

1

2

3

4

5

220
120
100
80
60
50

Reassociation of RPE65 with a phospholipid

membrane
To investigate the interaction of RPE65 with the lipid
membrane, we performed a liposome flotation assay.
Using this technique, others have shown that centrifugal force causes liposomes to float to the top of the
sucrose gradient, owing to inherent buoyancy, separating liposomes from unbound protein, which remains in
the bottom fractions [23]. All-trans-retinyl palmitate
was incorporated into 1,2-dioleoyl-sn-glycero-3-phosphocholine ⁄ 1,2-dilauroyl-sn-glycero-3-phosphocholine
liposomes at a lipid ⁄ retinyl palmitate ratio of 75 : 1.
After centrifugation, liposomes were predominantly
present in the fractions from the top of the gradient,
regardless of the presence of RPE65 protein (Fig. 4A).
In the absence of liposomes, RPE65 was located only
in the bottom fractions of the gradient (Fig. 4D,E).
However, in the presence of liposomes, significant
amounts of RPE65 floated to the top of the gradient
(Fig. 4B,C), demonstrating that RPE65 efficiently binds
to liposomes containing all-trans-retinyl palmitate.

1

250
150
100
75

40
30

C kDa
220

120
100
80
60
50
40
30
20

Fig. 3. Purification of recombinant RPE65. The 293A-LRAT cells
were infected with Ad-RPE65 expressing chicken RPE65 at an MOI
of 500. RPE65 was solubilized by 0.1% Chaps and purified by
Ni2+–nitrilotriacetic acid affinity chromatography. (A) SDS ⁄ PAGE
with Coomassie Brilliant Blue staining. (B) Western blot analysis
with antibody specific for RPE65. (C) Western blot analysis with
antibody specific for the His-tag. Lane 1: total cell lysate. Lane 2:
Chaps-solubilized supernatant after centrifugation at 200 000 g for
1 h. Lane 3: flow-through fraction not bound to the Ni2+–nitrilotriacetic acid column. Lane 4: purified recombinant chicken RPE65.
Lane 5: bovine RPE microsomal proteins. The amounts of protein
used for SDS ⁄ PAGE were 20 lg for lanes 1, 2, 3, and 5, and 5 lg
for lane 4. For western blot analysis, the amount of protein was
500 ng for each lane.

hydrolase activity of purified RPE65. As shown in
Fig. 5A, incubation of purified RPE65 with the liposomes containing all-trans-retinyl palmitate generated a

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS

3023

Isomerohydrolase activity of purified RPE65

O. Nikolaeva et al.

A

D

B
kDa

1

2

3

4

5

6

P

kDa

80

2

3

4

5

6

P

80

60

1

60

E

C

D
2.5

A320 (1 × 10–2 AU)

A320 (1 × 10–2 AU)

A
1

2.0
1.5
1.0

3
5

0.5

2

4

0
0

5

10

15

20

1.2
5

1.0
0.8
0.6
0.4
0.2
0
0

25

5

10

Time (min)
A320 (1 × 10–2 AU)

E

Absorbance
(1 × 10–3 AU)

2.0
1.5
1.0
0.5
0
280

300

320

340

360

1.2

25

1

0.8
0.6
0.4
0.2

2

5

0
0

380

5

10

15

20

25

20

25

Time (min)

F

1.2

A320 (1 × 10–2 AU)

A320 (1 × 10–2 AU)

20

1.0

Wavelength (nm)

1.0
0.8

0.6
0.4
0.2
0
0

5

10

15

Time (min)

3024

15

Time (min)

B

C

Fig. 4. Interaction of purified RPE65 with
liposomes. Purified RPE65 protein (25 lg)
was incubated with the 14C-labeled liposomes (100 lM lipids, 1.3 lM all-trans-retinyl
palmitate) for 2 h at 37 °C. The mixture was
placed at the bottom of a sucrose gradient
and centrifuged. Six 500 lL fractions were

collected from the top of the gradient. (A)
The lipid amount in each flotation fraction
was quantified by scintillation counting of
[14C]PC and expressed as a percentage of
the total amount of [14C]PC in the gradient
(means ± standard deviation, n = 3). (B–E)
Purified RPE65 was incubated with (B, C)
and without (D, E) 14C-labeled liposomes,
and centrifuged in the gradient as described
above. The same volumes from each fraction (30 lL) and pellets (6 lL) were examined by western blot analysis using antibody
against RPE65. RPE65 levels in each of the
flotation fractions with liposomes (C) and
without liposomes (E) were quantified by
densitometry and averaged from three independent experiments (mean ± standard
deviation, n = 3).

20

25

12
10
1

8
6
4
2

2

0
0

5

10

15

Time (min)

Fig. 5. Isomerohydrolase activity of purified
RPE65 reconstituted in liposomes. Purified
RPE65 (25 lg) was incubated with the
following substrates in the presence of
25 lM cellular retinaldehyde-binding protein
and 0.5% BSA for 2 h at 37 °C. The generated retinoids were analyzed by HPLC. (A)
Ten microliters of liposomes (250 lM lipids,
3.3 lM all-trans-retinyl palmitate); (B) UV
absorbance spectrum recorded for the
indicated 11-cis-retinol peak from (A); (C) no
substrate was added; (D) 3.3 lM all-transretinol; (E) 3.3 lM all-trans-retinyl palmitate
added in 2 lL of N,N-dimethylformamide
without liposomes; (F) 10 lL of liposomes
(250 lM lipids, 3.3 lM all-trans-retinyl palmitate) in the absence of RPE65 protein.
Peaks were identified as follows: 1, retinyl
esters; 2, all-trans-retinal; 3, 11-cis-retinol; 4,
13-cis-retinol; 5, all-trans-retinol.

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS


O. Nikolaeva et al.

Kinetics of the isomerohydrolase activity
of purified RPE65
To determine the steady-state kinetics of RPE65 activity, the assay conditions were optimized to ensure that
measurements were taken within the linear range. First,
we plotted the time course of 11-cis-retinol generation
after incubation of the liposomes containing all-transretinyl palmitate with 25 lg of purified RPE65 for
various time intervals. The time course of 11-cis-retinol
production was linear in its initial period (Fig. 6A), and
all of the further experiments in this study were therefore conducted within this range. Second, to establish
the dependence of 11-cis-retinol production on the concentration of purified RPE65, the liposomes containing
all-trans-retinyl palmitate were incubated with increasing amounts of purified RPE65. The production of
11-cis-retinol was found to be a linear function of
RPE65 concentration within a range of 20–250 lgỈmL)1
RPE65 (Fig. 6B). Finally, to analyze the substrate
dependence of the RPE65 isomerohydrolase, we measured the initial reaction velocity using different concentrations of retinyl ester incorporated into the liposomes.
Lineweaver–Burk analysis of these data yielded the

11-cis retinol (pmol)

A 600
500
400
300
200
100

0
0

20

40

60

80

100

120

140

Time (min)
11-cis retinol production
rate (pmol·h–1)

B 300
250
200
150
100
50
0
0

50

100

150

200

250

300

RPE65 (µg·mL–1)

C 0.35
1/V (pmol–1·min·mg)

significant amount of 11-cis-retinol (Fig. 5A). The
identity of the 11-cis-retinol peak was validated by
recording the UV spectrum during chromatography
(kmax = 319 nm) (Fig. 5B) and also confirmed by coelution with the 11-cis-retinol standard (data not
shown). As a control, no 11-cis-retinol was generated
when the purified RPE65 was incubated alone in the
absence of the added liposomes (Fig. 5C), suggesting
that the purified recombinant protein did not contain
endogenous all-trans-retinyl ester. To exclude the possibility that trace amounts of LRAT were copurified
with RPE65, all-trans-retinol was examined as a
substrate. Neither retinyl ester nor 11-cis-retinol was
produced after incubation of all-trans-retinol with the
purified RPE65 (Fig. 5D), confirming that LRAT

activity was absent from the system. This result also
provides further evidence confirming that all-trans-retinol is not an intrinsic substrate for RPE65.
When the liposomes containing all-trans-retinyl
palmitate were incubated in the absence of RPE65, no
11-cis-retinol was generated (Fig. 5E), verifying that
nonspecific thermal isomerization did not occur. Interestingly, in the absence of liposomes, RPE65 did not
generate 11-cis-retinol from nonincorporated all-transretinyl palmitate (Fig. 5F). These results indicate that
association of RPE65 with liposomes containing
the retinyl ester substrate is essential for the efficient
isomerohydrolase activity of RPE65.

Isomerohydrolase activity of purified RPE65

0.30
0.25
0.20
0.15
0.10
0.05
0
0

2

4

6

1/S (µM–1)
Fig. 6. Kinetic analysis of isomerohydrolase activity of purified

RPE65. (A) Time course of 11-cis-retinol generation. Liposomes
containing all-trans-retinyl palmitate (250 lM lipids, 3.3 lM all-transretinyl ester) were incubated with purified RPE65 (25 lg) for the
indicated time intervals, and the generated 11-cis-retinol was quantified by HPLC. (B) Dependence of isomerohydrolase activity on
RPE65 protein concentration. Various amounts of purified RPE65,
as indicated, were incubated with liposomes (250 lM lipids, 3.3 lM
all-trans-retinyl palmitate) for 2 h. The 11-cis-retinol generated from
the reaction was calculated from the area of the 11-cis-retinol peak
(mean ± standard deviation, n = 3). (C) Lineweaver–Burk plot of
11-cis-retinol generation by RPE65. Liposomes with increasing
concentrations (S) of all-trans-retinyl palmitate were incubated with
equal amounts of purified RPE65 (25 lg). Initial rates (V) of
11-cis-retinol generation were calculated according to 11-cis-retinol
production recorded by HPLC.

kinetic parameters kcat and Km for this reaction: the
Michaelis constant (Km) was 3.7 lm and the turnover
number (kcat) was 1.45 · 10)4 s)1 for purified RPE65
(Fig. 6C).

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS

3025


Isomerohydrolase activity of purified RPE65

O. Nikolaeva et al.

Discussion
A key step of the retinoid visual cycle is the conversion

of all-trans-retinyl ester to 11-cis-retinol, which is catalyzed by isomerohydrolase. Although the isomerohydrolase activity was first reported over 20 years ago
[25], the enzyme has eluded definite identification until
now. Recently, we and others [16–18] have shown that
cell lysates coexpressing RPE65 and LRAT can generate 11-cis-retinol from all-trans-retinol, assuming that
the long-sought isomerohydrolase in the visual cycle is
RPE65. As the isomerohydrolase activity has never
been demonstrated in purified RPE65, some researchers in this field are still not convinced that RPE65 is
the isomerohydrolase [20]. The present study established a novel in vitro isomerohydrolase assay that utilizes all-trans-retinyl ester incorporated into liposomes
as substrate for the isomerohydrolase. Using this
assay, we demonstrated that purified RPE65, when
reassociated with lipid membranes, directly converts
all-trans-retinyl ester to 11-cis-retinol, leading to
the conclusion that RPE65 is the isomerohydrolase.
Furthermore, this enzymatic activity assay allowed us
to measure the kinetic parameters of purified RPE65.
A major reason why previous studies failed to demonstrate isomerohydrolase activity using purified RPE65 is
that the isomerohydrolase activity of the protein is
highly sensitive to all detergents that were previously
used for solubilization of RPE65 [26]. In addition, early
attempts to detect the isomerohydrolase activity of
RPE65 were complicated because both all-trans-retinol
and retinyl ester were proposed as possible substrates
for the isomerase [8,27]. Although we later established
that all-trans-retinyl ester is the substrate for isomerohydrolase [24], its insolubility in hydrophilic milieu
limits its application in isomerohydrolase assays. Consequently, experiments employing ectopic coexpression of
LRAT and RPE65 in mammalian cells previously provided only indirect evidence of the isomerohydrolase
activity of RPE65 [16–18]. We have also shown that
colocalization of LRAT and RPE65 in the same membrane is essential for isomerohydrolase activity [17]. This
represents another challenge to reconstituting the isomerohydrolase activity of RPE65 in vitro, as LRAT has
not been purified as a full-length protein. To overcome

this difficulty, the present study established a novel
assay with which to characterize the enzymatic activity
of purified RPE65 by embedding the highly hydrophobic substrate – all-trans-retinyl palmitate – into liposomes that serve as a carrier of the substrate to the
enzyme. Our results showed that utilization of liposomes
dramatically enhanced the magnitude of RPE65 isomerohydrolase activity.
3026

Solubilization of membrane-associated proteins is
the critical first step in their purification. Although the
amounts of solubilized RPE65 increase with increasing
concentrations of Chaps, higher concentrations of
Chaps also abolished the enzymatic activity of RPE65.
By careful titration, we found that Chaps at a concentration of 0.1% was optimal for solubilizing RPE65
while preserving its catalytic activity. Interestingly,
several previous studies reported that RPE65 efficiently
binds retinyl ester substrate even at 1% Chaps [20,21].
It is likely that high concentrations of Chaps (e.g.
0.5%) may partially disturb the RPE65 conformation,
abolishing its catalytic activity, while leaving its substrate-binding ability intact. In this case, the all-transretinyl ester is probably bound nonproductively and
cannot be converted to 11-cis-retinol. Nonproductive
binding has been previously reported in other enzymes
[28]. Retinyl ester is a hydrophobic substance and does
not freely exchange between membranes [29]. In RPE
cells, retinyl esters are confined either to microsomal
membranes or lipid droplets [30]. It is unlikely that the
hydrophobic substrate diffuses from the membrane to
the aqueous phase to interact with the protein. Therefore, it is necessary for RPE65 to interact with the
lipid membrane to extract the hydrophobic substrate.
Indeed, it has been previously reported that RPE65
demonstrates high affinity for phospholipid vesicles [31].

RPE65 may bind to phospholipids through an attached
palmitoyl group [32] or through a hydrophobic patch
on the protein surface [33]. In the current work, we confirmed, using the liposome flotation assay, that RPE65
efficiently binds to liposomes containing retinyl ester. It
is possible that RPE65 regains its catalytically active
conformation upon binding to the liposomes. Other
examples of this phenomenon exist, such as the recent
finding that binding to lipid membranes induces a conformational change in Bax protein [23]. This could
explain why RPE65 cannot catalyze the conversion of
all-trans-retinyl ester substrate alone but displays robust
activity when it is incorporated into liposomes.
The current study shows that no 11-cis-retinol was
generated by purified RPE65 when N,N-dimethylformamide-solubilized all-trans-retinyl palmitate was
added in the absence of liposomes. This may seem to
contradict previous results published by us and others,
showing that a small amount of 11-cis-retinol was
generated when N,N-dimethylformamide-solubilized
all-trans-retinyl palmitate was added to an isomerohydrolase assay system using RPE65 in bovine [24] or
mouse [34] RPE microsomes. However, the low level
of isomerohydrolase activity observed in those assays
could be explained by the presence of lipid-containing
microsomes, which not only served to contain RPE65

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS


O. Nikolaeva et al.

in its membrane-bound, active conformation, but
could also allow a small proportion of retinyl ester to

be incorporated into the lipids of the microsomal
membrane to serve as a substrate for RPE65. In contrast, the present study was performed using purified
RPE65 in a membrane-free environment. Therefore,
this disparity can be ascribed to the lack of microsomal membranes in the present study.
Interestingly, previous attempts to reconstitute
RPE65 in proteoliposomes have been unsuccessful;
that is, isomerohydrolase activity has not been restored
[20]. It is possible that retinyl palmitate incorporated
into liposomes can promote the formation of the
catalytically active conformation of RPE65 upon its
reassociation with liposomes.
The data presented in this article suggest that interaction of RPE65 with lipid membrane is essential for its
isomerohydrolase activity. Previously, it has been proposed that light can regulate RPE65 function, switching
it between inactive soluble and active membrane-associated forms using a palmitoylation mechanism [32]. At
that time, the authors interpreted RPE65 as a retinyl
ester-binding protein that presents its substrate to an
unknown isomerohydrolase [32]. We assume that the
membrane association is probably essential for extracting highly hydrophobic retinyl ester substrate from the
membrane. Although a crystal structure of RPE65 is
not yet available, a computer model using a carotenoid
oxygenase as a template suggests that retinyl ester is
bound inside a hydrophobic tunnel [35]. It is likely that
RPE65 binds to the retinyl ester-containing membrane
in such a manner that the entrance of the tunnel would
be located close to the membrane surface. Such an
interaction would allow for substrate to transfer from
the hydrophobic milieu of the membrane to the hydrophobic tunnel of the RPE65 active site. This transfer
would be energetically favorable, as it would allow the
hydrophobic substrate to avoid unfavorable interactions with water.
The exact mechanism for the interaction of RPE65

with the membrane is currently unknown. It has been
suggested that palmitoylation of the three Cys residues
may be responsible for the membrane association [32].
However, it was later shown that these Cys residues
are not palmitoylated [34]. Recently, a new palmitoylation site (Cys112) was found to be essential for membrane association of RPE65 [12]. It has also been
shown that a fragment of RPE65 containing residues
126–250 interacts with the lipid monolayer substantially more strongly than other fragments [33], suggesting that the sequence of RPE65 located between
residues 126 and 250 residues might be very important
for binding to the membrane.

Isomerohydrolase activity of purified RPE65

The isomerohydrolase activity of purified RPE65
obeyed classic Michaelis–Menten kinetics for a singlesubstrate enzyme-catalyzed reaction. Thus, kcat and Km
values for the purified RPE65 were determined and
compared with those of the other enzymes that process
retinoids and carotenoids enzymes. The kcat value was
calculated to be 1.45 · 10–4 s)1. Although this value
seems low, it is still higher than the kcat for the purified
truncated form of LRAT (4.8 · 10–5 s)1) [36]. The kcat
for full-length LRAT has not been determined, as it
has never been purified. The kcat for human b-carotene
oxygenase was reported to be 0.011 s)1, which is
75-fold higher than that of RPE65 [37]. However, it
should be taken into account that the purified RPE65
in the assay was not completely bound to liposomes
and, furthermore, liposome-bound RPE65 may be
incompletely refolded into its active conformation. The
Km value for purified recombinant chicken RPE65
was approximately 10-fold higher than that for

LRAT [36] and two-fold lower than the Km measured
for unpurified human RPE65 [16]. Previously, it has
been estimated that RPE65 has a specific activity at
least 25 000-fold lower than that of LRAT [16], suggesting that the high abundance of RPE65 in the RPE
may be necessary to compensate for its low catalytic
capacity.
It is likely that the kcat value for RPE65 measured
in this work is a lower estimate of RPE65 isomerohydrolase activity in the RPE, which can be higher for
several reasons. First, a change in conformation of
purified RPE65 upon reassociation with liposomes
may limit the reaction rate. Second, retinyl ester might
adopt various physicochemical forms in the complex
mixtures in the RPE (i.e. emulsions, membrane
vesicles, mixed micelles). This may also affect the
enzymatic activity of RPE65 in RPE cells.
In summary, the present study demonstrates that
purified RPE65 possesses intrinsic isomerohydrolase
activity, and provides conclusive biochemical evidence
that RPE65 is the isomerohydrolase of the visual cycle.
It also reveals that retinyl ester must be incorporated
into the phospholipid membrane to serve as a substrate for RPE65 isomerohydrolase. This finding opens
new opportunities to study the specificity of RPE65
for modified retinyl esters and to elucidate the chemical mechanism of the isomerohydrolase reaction.

Experimental procedures
Construction of Ad-RPE65 with a His-tag
The chicken RPE65 cDNA was cloned as described previously [22]. A DNA sequence encoding a histidine-hexamer

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS

3027


Isomerohydrolase activity of purified RPE65

O. Nikolaeva et al.

(6 · His) was inserted at the N-terminus of the chicken
RPE65 cDNA by PCR, using the following primers: forward primer, 5¢-GCGGCCGCCACCATGCATCATCACCA
TCACCATTACAGCCAGGTGGAGC-3¢ containing a NotI
site (underlined) and the Kozak sequence (bold); and
reverse primer, 5¢-AAGCTTCATGCTCTTTTGAAGAGTC
CATGG-3¢, containing a HindIII site (underlined). Preparation, amplification and titration of the recombinant adenovirus (Ad-RPE65) were performed as described previously
[17].

Evaluation of the effect of Chaps concentration
on the efficiency of RPE65 solubilization
Recombinant RPE65 was expressed as described previously
[22]. The 293A-LRAT cells [35] expressing RPE65 were
harvested, resuspended in Buffer R (10 mm BTP, pH 8.0,
100 mm NaCl), homogenized by sonication, and aliquoted.
Each portion was supplemented with various Chaps concentrations (0%, 0.001%, 0.01%, 0.1%, 0.3%, and 0.5%).
After incubation for 1 h, each homogenate was centrifuged
at 200 000 g for 30 min at 4 °C to obtain the solubilized
fractions. Two micrograms of protein from each fraction
was used for western blot analysis with the antibody
against RPE65 [11] to quantify RPE65.

Purification of recombinant RPE65
The cells expressing chicken RPE65 were resuspended in

Buffer A (50 mm sodium phosphate, pH 8.0), lysed by three
freeze–thaw cycles, and centrifuged at 100 000 g for 30 min
at 4 °C. The pellet was resuspended in 35 mL of Buffer B
(50 mm sodium phosphate, pH 8.0, 150 mm NaCl, 10%
glycerol, 0.1% Chaps), sonicated, incubated for 1 h at
4 °C, and centrifuged at 125 000 g for 1 h at 4 °C. The
supernatant was loaded onto an Ni2+-nitrilotriacetic acid
agarose (Qiagen Inc., Valencia, CA, USA) column. The column was washed with Buffer C (50 mm sodium phosphate,
pH 8.0, 300 mm NaCl, 10% glycerol, 0.1% Chaps) containing 10 mm imidazole. Protein was eluted with Buffer C containing 250 mm imidazole. The RPE65 elution pattern and
the purity of RPE65 were examined by Coomassie Brilliant
Blue staining and western blot analysis. The RPE65
protein-enriched fractions were pooled, concentrated, and
dialyzed at 4 °C against Buffer D (50 mm sodium phosphate, pH 8.0, 100 mm NaCl, 10% glycerol, 0.1% Chaps).
The concentration of the purified RPE65 was determined
by Bradford assay [38].

Western blot analysis
The same amount of total protein (20 lg) was blotted with
antibody against RPE65 (1 : 1000 dilution) or antibody
against His-tag (Sigma-Aldrich, St Louis, MO, USA) as

3028

previously described [24]. The membrane was briefly
washed with the stripping buffer (Pierce, Rockford, IL,
USA) and reblotted with a monoclonal antibody for b-actin
(Abcam, Cambridge, MA, USA) where it was specified
(1 : 2500 dilution). Western blot images were captured with
the imager Chemi-Genius2 (Syngene, Frederick, MD,
USA).

Liposome preparation
All phospholipids used in this study were purchased from
Avanti Polar Lipids (Alabaster, AL, USA). Chloroform
stocks of 1,2-dioleoyl-sn-glycero-3-phosphocholine and
1,2-dilauroyl-sn-glycero-3-phosphocholine were mixed at
85 : 15 (mol ⁄ mol) and supplemented with all-trans-retinyl
palmitate to produce a 75 : 1 lipid ⁄ retinyl ester ratio. l-a-1Palmitoyl-2-arachidonyl-phophatidylcholine ([14C]PC) was
used to label liposomes in proportions of 0.1 lCiỈmL)1.
The organic solvent was removed by argon flow under dim
red light, and the dried lipids ⁄ all-trans-retinyl palmitate film
was dispersed in Buffer R by vortexing. This mixture was
exposed to five freeze–thaw cycles and passed through a
polycarbonate membrane (0.1 lm) with a Mini-Extruder
(Avanti Polar Lipids). The total lipid concentration of the
resulting liposome suspension was 5 mm.

Liposome flotation assay to detect membrane
binding of purified RPE65
The purified recombinant RPE65 (25 lg) was incubated
with 20 lL of liposomes (100 lm lipid, 1.3 lm all-trans-retinyl palmitate) in Buffer R for 2 h at 37 °C in the dark. The
mixture (50 lL) was adjusted to a final sucrose concentration of 1.8 m (final volume 450 lL), placed at the bottom
of a 3.5 mL ultracentrifuge tube, and overlaid consecutively
with 850 lL portions of 1.35, 0.8 and 0.25 m sucrose in the
same buffer. The gradient was centrifuged at 250 000 g for
3 h at 10 °C, and 500 lL fractions were then drawn from
the top. The pellets were resuspended in 100 lL of Laemmli
sample buffer to detect aggregated and sedimented protein.
Aliquots of each fraction (30 lL) and pellets (6 lL) were
analyzed by immunoblotting with the antibody against

RPE65. The RPE65 content in each fraction was analyzed
by densitometry. The lipid distribution was determined by
liquid scintillation counting of [14C]PC radioactivity.

In vitro isomerohydrolase activity assay
The 293A-LRAT cells expressing RPE65 were lysed in
Buffer R. For each reaction, the liposomes (250 lm lipids,
3.3 lm all-trans-retinyl palmitate) and either 500 lg of total
proteins of cell lysates, 250 lg of Chaps-solubilized supernatant proteins or 25 lg of the purified RPE65 was added
to 200 lL of Buffer R containing 0.5% BSA and 25 lm

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS


O. Nikolaeva et al.

cellular retinaldehyde-binding protein. After 2 h of incubation at 37 °C in the dark, the generated retinoids were
extracted with 300 lL of methanol and 300 lL of hexane
and analyzed by normal-phase HPLC as described previously [24].

Acknowledgements
This study was supported by NIH grants EY012231
and ET015650, grant P20RR024215 from the National
Center for Research Resources, a research award from
JDRF, a grant from ADA, and a research grant from
OCAST HR07-067.

Isomerohydrolase activity of purified RPE65

12

13

14

15

References
1 Baylor D (1996) How photons start vision. Proc Natl
Acad Sci USA 93, 560–565.
2 McBee JK, Palczewski K, Baehr W & Pepperberg DR
(2001) Confronting complexity: the interlink of phototransduction and retinoid metabolism in the vertebrate
retina. Prog Retin Eye Res 20, 469–529.
3 Lamb TD & Pugh EN Jr (2004) Dark adaptation and
the retinoid cycle of vision. Prog Retin Eye Res 23,
307–380.
4 Saari JC (2000) Biochemistry of visual pigment regeneration: the Friedenwald lecture. Invest Ophthalmol Vis
Sci 41, 337–348.
5 Rando RR (2001) The biochemistry of the visual cycle.
Chem Rev 101, 1881–1896.
6 Rattner A, Smallwood PM & Nathans J (2000) Identification and characterization of all-trans-retinol dehydrogenase from photoreceptor outer segments, the visual
cycle enzyme that reduces all-trans-retinal to all-transretinol. J Biol Chem 275, 11034–11043.
7 Saari JC & Bredberg DL (1989) Lecithin:retinol
acyltransferase in retinal pigment epithelial microsomes.
J Biol Chem 264, 8636–8640.
8 Rando RR (1991) Membrane phospholipids as an
energy source in the operation of the visual cycle. Biochemistry 30, 595–602.
9 Deigner PS, Law WC, Canada FJ & Rando RR (1989)
Membranes as the energy source in the endergonic
transformation of vitamin A to 11-cis-retinol. Science

244, 968–971.
10 Hamel CP, Tsilou E, Pfeffer BA, Hooks JJ, Detrick
B & Redmond TM (1993) Molecular cloning and
expression of RPE65, a novel retinal pigment epithelium-specific microsomal protein that is post-transcriptionally regulated in vitro. J Biol Chem 268,
15751–15757.
11 Ma J, Zhang J, Othersen KL, Moiseyev G, Ablonczy
Z, Redmond TM, Chen Y & Crouch RK (2001)

16

17

18

19

20

21

22

23

24

Expression, purification, and MALDI analysis of
RPE65. Invest Ophthalmol Vis Sci 42, 1429–1435.
Takahashi Y, Moiseyev G, Ablonczy Z, Chen Y,
Crouch RK & Ma JX (2009) Identification of a novel

palmitylation site essential for membrane association
and isomerohydrolase activity of RPE65. J Biol Chem
284, 3211–3218.
Redmond TM, Yu S, Lee E, Bok D, Hamasaki D, Chen
N, Goletz P, Ma JX, Crouch RK & Pfeifer K (1998)
Rpe65 is necessary for production of 11-cis-vitamin A in
the retinal visual cycle. Nat Genet 20, 344–351.
Thompson DA & Gal A (2003) Genetic defects in vitamin A metabolism of the retinal pigment epithelium.
Dev Ophthalmol 37, 141–154.
Thompson DA, Gyurus P, Fleischer LL, Bingham EL,
McHenry CL, Apfelstedt-Sylla E, Zrenner E, Lorenz B,
Richards JE, Jacobson SG et al. (2000) Genetics and
phenotypes of RPE65 mutations in inherited retinal
degeneration. Invest Ophthalmol Vis Sci 41, 4293–4299.
Jin M, Li S, Moghrabi WN, Sun H & Travis GH
(2005) Rpe65 is the retinoid isomerase in bovine retinal
pigment epithelium. Cell 122, 449–459.
Moiseyev G, Chen Y, Takahashi Y, Wu BX & Ma JX
(2005) RPE65 is the isomerohydrolase in the retinoid
visual cycle. Proc Natl Acad Sci USA 102, 12413–
12418.
Redmond TM, Poliakov E, Yu S, Tsai JY, Lu Z & Gentleman S (2005) Mutation of key residues of RPE65 abolishes its enzymatic role as isomerohydrolase in the visual
cycle. Proc Natl Acad Sci USA 102, 13658–13663.
Xue L, Jahng WJ, Gollapalli D & Rando RR (2006)
Palmitoyl transferase activity of lecithin retinol acyl
transferase. Biochemistry 45, 10710–10718.
Mata NL, Moghrabi WN, Lee JS, Bui TV, Radu RA,
Horwitz J & Travis GH (2004) Rpe65 is a retinyl ester
binding protein that presents insoluble substrate to the
isomerase in retinal pigment epithelial cells. J Biol Chem

279, 635–643.
Gollapalli DR, Maiti P & Rando RR (2003) RPE65
operates in the vertebrate visual cycle by stereospecifically binding all-trans-retinyl esters. Biochemistry 42,
11824–11830.
Moiseyev G, Takahashi Y, Chen Y, Kim S & Ma JX
(2008) RPE65 from cone-dominant chicken is a more
efficient isomerohydrolase compared with that from
rod-dominant species. J Biol Chem 283, 8110–8117.
Yethon JA, Epand RF, Leber B, Epand RM &
Andrews DW (2003) Interaction with a membrane
surface triggers a reversible conformational change in
Bax normally associated with induction of apoptosis.
J Biol Chem 278, 48935–48941.
Moiseyev G, Crouch RK, Goletz P, Oatis J Jr, Redmond
TM & Ma JX (2003) Retinyl esters are the substrate for
isomerohydrolase. Biochemistry 42, 2229–2238.

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS

3029


Isomerohydrolase activity of purified RPE65

O. Nikolaeva et al.

25 Bernstein PS, Law WC & Rando RR (1987) Isomerization of all-trans-retinoids to 11-cis-retinoids in vitro
[see comments]. Proc Natl Acad Sci USA 84, 1849–
1853.
26 Barry RJ, Canada FJ & Rando RR (1989) Solubilization and partial purification of retinyl ester synthetase

and retinoid isomerase from bovine ocular pigment
epithelium. J Biol Chem 264, 9231–9238.
27 McBee JK, Kuksa V, Alvarez R, de Lera AR, Prezhdo
O, Haeseleer F, Sokal I & Palczewski K (2000) Isomerization of all-trans-retinol to cis-retinols in bovine retinal
pigment epithelial cells: dependence on the specificity of
retinoid-binding proteins. Biochemistry 39, 11370–
11380.
28 Huang WC, Westlake AC, Marechal JD, Joyce MG,
Moody PC & Roberts GC (2007) Filling a hole in cytochrome P450 BM3 improves substrate binding and catalytic efficiency. J Mol Biol 373, 633–651.
29 Ho MT, Pownall HJ & Hollyfield JG (1989) Spontaneous transfer of retinoic acid, retinyl acetate, and retinyl
palmitate between single unilamellar vesicles. J Biol
Chem 264, 17759–17763.
30 Imanishi Y, Batten ML, Piston DW, Baehr W &
Palczewski K (2004) Noninvasive two-photon imaging
reveals retinyl ester storage structures in the eye. J Cell
Biol 164, 373–383.
31 Tsilou E, Hamel CP, Yu S & Redmond TM (1997)
RPE65, the major retinal pigment epithelium micro-

3030

32

33

34

35

36

37

38

somal membrane protein, associates with phospholipid
liposomes. Arch Biochem Biophys 346, 21–27.
Xue L, Gollapalli DR, Maiti P, Jahng WJ & Rando
RR (2004) A palmitoylation switch mechanism in the
regulation of the visual cycle. Cell 117, 761–771.
Trudel E, Beaufils S, Renault A, Breton R & Salesse C
(2006) Binding of RPE65 fragments to lipid monolayers
and identification of its partners by glutathione S-transferase pull-down assays. Biochemistry 45, 3337–3347.
Jin M, Yuan Q, Li S & Travis GH (2007) Role of
LRAT on the retinoid isomerase activity and membrane
association of Rpe65. J Biol Chem 282, 20915–20924.
Takahashi Y, Moiseyev G, Chen Y & Ma JX (2005)
Identification of conserved histidines and glutamic acid
as key residues for isomerohydrolase activity of RPE65,
an enzyme of the visual cycle in the retinal pigment
epithelium. FEBS Lett 579, 5414–5418.
Bok D, Ruiz A, Yaron O, Jahng WJ, Ray A, Xue L &
Rando RR (2003) Purification and characterization of a
transmembrane domain-deleted form of lecithin retinol
acyltransferase. Biochemistry 42, 6090–6098.
Lindqvist A & Andersson S (2002) Biochemical properties of purified recombinant human beta-carotene
15,15¢-monooxygenase. J Biol Chem 277, 23942–23948.
Bradford MM (1976) A rapid and sensitive method for
the quantitation of microgram quantities of protein
utilizing the principle of protein-dye binding. Anal

Biochem 72, 248–254.

FEBS Journal 276 (2009) 3020–3030 ª 2009 The Authors Journal compilation ª 2009 FEBS