'Chemical co-substrate rescue' of phytanoyl-CoA 2-hydroxylase mutants causing refsum's disease

Article abstract of DOI:10.1039/b101039p

The in vitro catalytic activity of two clinically observed mutants of phytanoyl-CoA 2-hydroxylase, an iron(II)/2-oxoglutarate-dependent oxygenase causing Refsum's Disease, was partially rescued by the use of alternatives to the natural cosubstrate, 2-oxoglutarate; this is the first demonstration of 'chemical co-substrate rescue' of mutations to an enzyme causing human disease.

Full text of DOI:10.1039/b101039p

‘Chemical co-substrate rescue’ of phytanoyl-CoA 2-hydroxylase
mutants causing Refsum’s Disease
a
a
a
a
Mridul Mukherji, Nadia J. Kershaw, Colin H. MacKinnon, Ian J. Clifton, Anthony S.
b
a
a
Wierzbicki, Christopher J. Schofield* and Matthew D. Lloyd*
a
The Oxford Centre for Molecular Sciences and The Dyson Perrins Laboratory, South Parks Road,
Oxford, UK OX1 3QY, E-mail: matthew.lloyd@chem.ox.ac.uk, christopher.schofield@chem.ox.ac.uk
Department of Chemical Pathology, King’s, Guy’s & St. Thomas’ Medical School, St. Thomas’
Hospital Campus, Lambeth Palace Road, London, UK SE1 7EH
b
Received (in Cambridge, UK) 30th January 2001, Accepted 19th April 2001
First published as an Advance Article on the web 10th May 2001
The in vitro catalytic activity of two clinically observed
mutants of phytanoyl-CoA 2-hydroxylase, an iron(II)/2-oxo-
glutarate-dependent oxygenase causing Refsum’s Disease,
was partially rescued by the use of alternatives to the natural
cosubstrate, 2-oxoglutarate; this is the first demonstration of
genesis. The desired proteins were expressed via standard
techniques in recombinant E. coli and purified as mature (i.e.
without their peroxisomal targeting sequences) enzymes to
> 95% homogeneity (by SDS-PAGE analysis). Mutations were
confirmed by DNA sequencing and ESI MS analyses. Analysis
of the secondary structure by circular dichroism suggested that
all mutants had a similar overall structure to the wild-type
enzyme.
‘
chemical co-substrate rescue’ of mutations to an enzyme
causing human disease.
Phytanic acid in the human diet is derived from the phytol
sidechain of chlorophyll, but the presence of a 3-methyl group
prevents its degradation via the fatty acid b-oxidation pathway.
Instead, a preliminary pathway effects a-oxidation of phytanic
acid, excising a methylene group to give pristanic acid.1
Within this pathway, hydroxylation of phytanoyl-CoA to
Activity of the wild-type enzyme and mutants was assayed by
1
0
conversion of phytanoyl-CoA to 2-hydroxyphytanoyl-CoA.
Assays were performed (at least in triplicate) according to a
11
modified version of a published procedure and included ATP
,2
12
to obtain maximum activity. The activity was further
enhanced ca. two-fold by the use of tris(carboxyethyl)phos-
2
-hydroxyphytanoyl-CoA is catalysed by phytanoyl-CoA 2-hy-
phine (TCEP) in place of the previously used dithiothreitol
3,4
11
droxylase (PAHX), an iron(II) and 2-oxoglutarate-dependent
oxygenase (Scheme 1). Mutations to PAHX cause ca. 45% of
(DTT). Two concentrations of 2-oxoacid (2 and 60 mM) were
5
used in order to facilitate detection of low levels of activity.
Assay mixtures for the analogues contained: 50 mM Tris-HCl,
adult Refsum’s Disease (ARD) with other cases being asso-
ciated with a second locus.6
pH 7.5, 1 mM FeSO , 50 mM synthetic (3RS,7R,11R)-
4
Two common point mutations in human PAHX, R275W and
R275Q, are present at an allele frequency of 3/40 (7.5%) and
phytanoyl-CoA, 0.44 mM b-cyclodextrin, 100 mM TCEP, 10
mM ascorbate, 4 mM ATP and ca. 20 mg enzyme (2 mM
2-oxoacid) or 10 mg enzyme (60 mM 2-oxoacid). Reactions
were quenched with 250 mM EDTA after incubation for 60 min
(2 mM 2-oxoacid) or 5 min (60 mM 2-oxoacid). Samples were
centrifuged and analysed by HPLC using a Hypersil C18 column
5
/40 (12.5%), respectively, in Dutch/Scandanavian patients
3,7
with ARD.
Sequence analyses revealed that Arg-275 is
conserved in all reported PAHX enzymes and closely related
sequences. Analysis of the sequences in the light of crystal
structures8 for two other 2-oxoglutarate oxygenases suggested
that Arg-275 binds the 5-carboxylate of the co-substrate via an
,9
11
(250 3 4.6 mm) monitoring at 254 nm.
Using the natural co-substrate, 2-oxoglutarate, the activity of
the two clinically observed mutants was < 0.5% of that of the
wild-type enzyme (Table 1). A range of 2-oxoacids was tested
in an attempt to restore the activity of the mutants. Homogen-
tisate, 4-hydroxyphenylpyruvate, indole-3-pyruvate, 2-mercap-
8
electrostatic interaction. We postulated that the PAHX R275Q
and R275W mutants were inactive due to defective 2-oxoglutar-
ate binding or utilisation, and that it may be possible to rescue
their activity using alternative co-substrates.
2
1
The two clinically observed mutants of Arg-275 in human
PAHX (R275Q and R275W)3 and a further mutant R275A,
made for comparison, were constructed by site-directed muta-
tosuccinate all gave specific activities of < 0.2 nmol min
,7
21
mg with wild-type enzyme and all mutants. The activity of
the R275Q and R275W mutants was significantly ‘rescued’
compared to the wild-type activity with 2-oxoglutarate as a co-
substrate using certain 2-oxoacids at a concentration of 60 mM
(
Table 1). Optimum rescue of mutants with hydrophobic/
aliphatic residues in place of Arg-275 is achieved using
-oxoacids with side-chains of 2–4 carbon atoms or equivalent
length side-chains. Effective examples are in the use of
2
2
5
-oxobutyrate with R275Q mutant (Table 1, entry 4) and 2-oxo-
-thiahexanoate with the R275W mutant (Table 1, entry 9). The
latter case is striking because 2-oxo-5-thiahexanoate cannot be
substituted for 2-oxoglutarate ( < 0.5%) in assays with the wild-
type enzyme. The dramatic change in co-substrate selectivity
resulting from PAHX Arg-275 mutations may be useful for the
clinical identification of these particular mutants using a
modified assay with alternative 2-oxoacids.
Even higher levels of rescue were obtained in the case of the
R275A mutant where most of the hydrophobic analogues tested
(
Table 1, entries 3–9), led to 22–28% rescue at 60 mM. An
exception was 2-oxooctanoic acid (Table 1, entry 10), which
was inactive with the wild-type enzyme, and all mutants
assayed, presumably due to its carbon chain being too large to
Scheme 1 The role of phytanoyl-CoA 2-hydroxylase in the peroxisomal
degradation of phytanic acid.
972
Chem. Commun., 2001, 972–973
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b101039p

Table 1 Specific activities of (3RS,7R,11R)-phytanoyl-CoA hydroxylation as catalysed by mature recombinant wild-type and mutant PAHX enzymes in the
presence of various 2-oxoacids (nmol min2 mg protein). Assays were carried out with 2 mM or 60 mM 2-oxoacid
1
21
PAHX mutant
R275 (wild-type)
2
R275A
2
R275Q
2
R275W
2
2-oxoacid/mM
60
60
60
60
1
. 2-oxoglutarate
2
R = (CH
) CO
2 2 2
265
129
885
235
2.63
3.14
8.7
128
0.59
< 0.2
< 0.2
54.0
106
< 0.2
< 0.2
< 0.2
11.0
55.2
63.9
17.0
66.2
56.2
98.2
< 0.2
2
. 2-oxoadipate
2
R = (CH
2 3
) CO
2
57.0
< 0.2
< 0.2
0.6
3
. pyruvate
R = CH
. 2-oxobutyrate
R = CH CH
. 2-oxovalerate
R = (CH CH
. 2-oxoisovalerate
R = CH(CH
. 2-oxocaproate
R = (CH CH
. 2-oxoisocaproate
R = CH CH(CH
. 2-oxo-5-thiahexanoate
R = (CH SCH
0. 2-oxooctanoic acid
R = (CH CH
3
6.8
78.6
201
250
210
192
235
217
226
< 0.2
< 0.2
2.7
4
2
3
2.1
1.6
69.4
40.1
45.3
20.1
17.8
< 0.2
< 0.2
5.7
0.9
5
)
2 2
3
4.8
3.0
71.6
72.2
42.4
45.4
52.8
< 0.2
1.93
0.49
2.4
6
)
3 2
0.65
0.7
4.30
5.3
1.52
1.33
1.9
7
)
2 3
3
8
2
3
)
2
0.8
4.8
2.2
9
)
2 2
3
< 0.2
N/D
4.8
1.6
2.6
1
)
2 5
3
N/D
N/D
N/D
be accommodated in the proper orientation in the active site.
The R275A mutant was also less selective than the wild-type or
other tested mutants, presumably due to a relaxation of steric
and electrostatic constraints reflecting the presence of a small,
hydrophobic and neutral side-chain.
kinases.15 The in vitro work that is reported here is the first
demonstration of the ‘chemical co-substrate rescue’ of muta-
tions in an enzyme (Scheme 2) implicated in a human disease.
In vivo studies directed towards demonstrating the technique in
cell lines are in progress.
2-Oxoacids are metabolically related to proteinogenic amino
acids via transamination reactions. Several of the 2-oxoacids
which rescue the activity of the clinically observed R275W and
R275Q mutants are thus accessible via in vivo amino acids, e.g.
Notes and references
2-oxovalerate from valine and 2-oxo-5-thiahexanoate from
1
D. Steinberg, in ‘Refsum Disease’, The metabolic and molecular basis
of inherited metabolic disease, ed. C. R. Scriver, A. L. Beaudet, W. S.
Sly and D. Valle, New York, 1995.
methionine. Thus, certain forms of ARD might be treated via
dietary supplements containing the appropriate amino acids.
Maple syrup urine disease is caused by a deleterious accumula-
tion of excess 2-oxacids, and is treated, in the reverse of this
2 N. M. Verhoeven, R. J. A. Wanders, B. T. Poll-The, J. M. Saudbubray
and C. Jakobs, J. Inherit. Metab. Dis., 1998, 21, 697.
3
4
S. J. Mihalik, J. C. Morrell, D. Kim, K. A. Stackster, P. A. Watkins and
S. J. Gould, Nat. Genet., 1997, 17, 185.
G. A. Jansen, R. Ofman, S. Ferdinandusse, L. IIjlst, A. O. Muisers, O. H.
Skjeldal, O. Stokke, C. Jakobs, G. T. N. Besley, J. E. Wraith and R. J. A.
Wanders, Nature Genet., 1997, 17, 190.
proposed therapy, by a diet low in branched-chain amino
acids.13 Restoration of complete wild-type activity may not be
required, as 5% of wild-type activity is apparently sufficient to
effectively correct inherited homocystinuria14 with vitamin
B6.
To our knowledge the only other example of the rescue of
enzyme activity with a modified co-substrate has involved the
elegant use of ATP analogues to study in vivo activity of
5
6
A. G. Prescott and M. D. Lloyd, Nat. Prod. Rep., 2000, 17, 367.
A. S. Wierzbicki, J. Mitchell, M. Lambert-Hammill, M. Hancock, J.
Greenwood, M. C. Sidey, J. de Belleroche and F. B. Gibberd, Eur. J.
Hum. Genet., 2000, 8, 649.
7
8
9
G. A. Jansen, E. M. Hogenhout, S. Ferdinandusse, H. R. Waterham, R.
Ofman, C. Jakobs, O. H. Skjeldal and R. J. A. Wanders, Hum. Mol.
Gen., 2000, 9, 1195.
K. Valegård, A. C. Terwisscha van Scheltinga, M. D. Lloyd, T. Hara, S.
Ramaswamy, A. Perrakis, A. Thompson, H.-J. Lee, J. E. Baldwin, C. J.
Schofield, J. Hajdu and I. Andersson, Nature, 1998, 394, 805.
Z.-H. Zhang, J. Ren, J. K. Stammers, J. E. Baldwin, K. Harlos and C. J.
Schofield, Nat. Struct. Biol., 2000, 7, 127.
1
1
1
1
0 M. Mukherji, N. J. Kershaw, I. J. Clifton, C. J. Schofield, A. S.
Wierzbicki and M. D. Lloyd, manuscript in preparation.
1 S. J. Mihalik, A. M. Rainville and P. A. Watkins, Eur. J. Biochem.,
1
995, 232, 545.
2 K. Croes, V. Foulon, M. Casteels, P. P. Van Veldhoven and G. P.
Mannaerts, J. Lipid Res., 2000, 41, 629.
3 S. E. Snyderman, P. M. Norton, E. Roitman and L. Holt, Pediatrics,
1964, 34, 454.
Scheme 2 ‘Chemical co-substrate rescue’ of a PAHX mutant as exemplified
for R275W. R = hydrophobic/aliphatic group. Wild-type enzyme showing
interaction of guanidino group of Arg-275 and 5-carboxylate of 2-oxo-
glutarate (above). Unfavourable interaction between aromatic sidechain of
Trp-275 and 5-carboxylate of 2-oxoglutarate; rescue of activity via
hydrophobic interactions in 2-oxoacid binding site (below). The relative
arrangement of the iron ligands is that of deacetoxycephalosporin C
14 J. P. Kraus, M. Janosik, V. Kozich, R. Mandall, V. Shih, M. P.
Sperandeo, G. Sebastio, R. de Franchis, G. Andria, L. A. Kluijtmans, H.
Blom, G. H. Boers, R. B. Gordon, P. Kamoun, M. Y. Tsai, W. D.
Kruger, H. G. Koch, T. Ohura and M. Gaustadnes, Hum. Mut., 1999, 13,
362.
15 Y. Liu, K. Shah, F. Yang, L. Witucki and K. M. Shokat, Chem. Biol.,
1998, 5, 91.
8
synthase (DAOCS).
Chem. Commun., 2001, 972–973
973