Loss-of-function mutation in carotenoid 15,15′-monooxygenase identified in a patient with hypercarotenemia and hypovitaminosis A
The enzyme carotenoid 15,15'-monooxygenase (CMO1) catalyzes the first step in the conversion of dietary provitamin A carotenoids to vitamin A in the small intestine. Plant carotenoids are an important dietary source of vitamin A (retinol) and the sole source of vitamin A for vegetarians. Vitamin A is essential for normal embryonic development as well as normal physiological functions in children and adults. Here, we describe one heterozygous T170M missense mutation in the CMO1 gene in a subject with hypercarotenemia and mild hypovitaminosis A. The replacement of a highly conserved threonine with methionine results in a 90% reduction in enzyme activity when analyzed in vitro using purified recombinant enzymes. The Michaelis-Menten constant (K(m)) for the mutated enzyme is normal. Ample amounts of carotenoids are present in plasma of persons consuming a normal Western diet, suggesting that the enzyme is saturated with substrate under normal conditions. Therefore, we propose that haploinsufficiency of the CMO1 enzyme may cause symptoms of hypercarotenemia and hypovitaminosis A in individuals consuming a carotenoid-containing and vitamin A-deficient diet.
The Journal of Nutrition
Biochemical, Molecular, and Genetic Mechanisms
Loss-of-Function Mutation in Carotenoid 15,15#-
Monooxygenase Identiﬁed in a Patient with
Hypercarotenemia and Hypovitaminosis A
Denis E. Sharvill,
and Stefan Andersson
Departments of Obstetrics-Gynecology and Biochemistry, University of Texas Southwestern Medical Center, Dallas,
Balmoral Surgery, Deal, Kent CT14 7AU, UK; and
Buckland Hospital, Dover, Kent CT17 0HD, UK
The enzyme carotenoid 15,15#-monooxygenase (CMO1) catalyzes the ﬁrst step in the conversion of dietary provitamin A
carotenoids to vitamin A in the small intestine. Plant carotenoids are an important dietary source of vitamin A (retinol) and the
sole source of vitamin A for vegetarians. Vitamin A is essential for normal embryonic development as well as normal
physiological functions in children and adults. Here, we describe one heterozygous T170M missense mutation in the CMO1
gene in a subject with hypercarotenemia and mild hypovitaminosis A. The replacement of a highly conserved threonine with
methionine results in a 90% reduction in enzyme activity when analyzed in vitro using puriﬁed recombinant enzymes. The
Michaelis-Menten constant (K
) for the mutated enzyme is normal. Ample amounts of carotenoids are present in plasma of
persons consuming a normalWestern diet, suggesting that the enzyme is saturated with substrate under normal conditions.
Therefore, we propose that haploinsufﬁciency of the CMO1 enzyme may cause symptoms of hypercarotenemia and
hypovitaminosis A in individuals consuming a carotenoid-containing and vitamin A-deﬁcient diet. J. Nutr. 137: 2346–2350, 2007.
Plant carotenoids are an important dietary source of vitamin A
(retinol and its esters) and the sole source of vitamin A for
vegetarians. The ﬁrst step in the conversion of dietary provita-
min A carotenoids to vitamin A is the cleavage of the central
carbon 15,15#-double bond in carotenoid substrates. This re-
action is catalyzed by the cytoplasmic enzyme carotenoid
(previously termed b-carotene
15,15#-monooxygenase) in the epithelial cells of the small in-
testinal mucosa (1,2). The most common carotenoid substrate
for CMO1 is b-carotene, which is cleaved by the enzyme to form
2 molecules of retinal (retinaldehyde). The retinal formed is
further converted to retinol and subsequently to retinol esters in
the epithelial cells of the intestinal mucosa and then transported
in chylomicrons to the liver, the main organ for vitamin A
storage (3). Importantly, numerous studies have shown that a
substantial amount of the absorbed dietary carotenoids are not
cleaved by the CMO1 enzyme in the human intestine, suggesting
that the CMO1 enzyme is saturated during normal dietary con-
ditions (4–8). Carotenoids that escape the CMO1 enzyme are
also incorporated into chylomicrons together with other lipids
and the majority of carotenoids circulating in the blood are
associated with low and high density lipoprotein particles and
hence are taken up in tissues via the LDL receptor (9–11).
CMO1 has been shown to be highly expressed in the epithelial
cells of a variety of extraintestinal tissues, which suggests that
the enzyme may constitute a back-up pathway for vitamin A
synthesis during times of insufﬁcient dietary intake of vitamin A
(12). An important feature of the enzyme is that only 1 unsub-
stituted b-ionone ring half-site is required for efﬁcient 15,15#-
double bond cleavage; hence, in addition to b-carotene, there are
;50 additional known provitamin A carotenoids found in na-
ture that can each form 1 molecule of retinal (13).
Retinoids are vital for normal embryonic and fetal develop-
ment as well as for normal physiological functions in children
and adults. Vitamin A deﬁciency in humans leads to impaired
dark adaptation, xerosis of the conjunctiva and cornea, kerato-
malacia, and impaired function of the immune system. In some
developing countries, dietary or primary deﬁciency of vitamin A
remains the major cause of blindness in young children and
carries with it a high mortality rate (3,14).
Whether vitamin A deﬁciency in humans can be caused by
primary failure of enzymatic cleavage of carotenoids to retinal is
unknown. So far, only a few cases have been reported in which
individuals were thought to be vitamin A deﬁcient due to an
impairment in the conversion of carotenoids to retinal in the
intestine (5,15–20). These subjects had orange-yellowish skin
and variable degrees of mild vitamin A deﬁciency and in all
cases, the biochemical hallmark was high levels of circulating
This work was supported by grant DK62192 from the NIH.
Author disclosures: A. Lindqvist, J. Sharvill, and S. Andersson, no conﬂicts of
Supplemental Table 1 and Supplemental Figure 1 are available with the online
posting of this paper at jn.nutrition.org.
Deceased November 15, 2003.
Abbreviations used: ACO, apocarotenoid 15,15#-oxygenase; CMO1, carote-
noid 15,15#-monooxygenase; OTG, octylthioglucoside; SNP, single nucleotide
* To whom correspondence should be addressed. E-mail: stefan.andersson@
2346 0022-3166/07 $8.00 ª 2007 American Society for Nutrition.
Manuscript received 8 August 2007. Initial review completed 29 August 2007. Revision accepted 6 September 2007.
by guest on June 5, 2013jn.nutrition.orgDownloaded from
Supplemental Material can be found at:
carotenoids. The fact that the patients had slightly lowered
serum vitamin A levels and reportedly consumed diets normal to
low-normal in carotene intake excludes dietary carotenemia.
Importantly, all patients responded well to treatment with
vitamin A for the mild deﬁciency. The accumulation of carot-
enoids in the skin of the patients was slowly alleviated by low-
carotenoid diets. These reports all relate to nonvegetarian
individuals and the effects of reduced vitamin A production
due to a dysfunctional CMO1 enzyme would become more ap-
parent in people who adhere to a strict vegetarian diet lacking
preformed vitamin A, albeit rich in provitamin A carotenoids.
The incidence of individuals with impaired conversion of carot-
enoids to vitamin A is unknown, because in most cases, Western
diets contain sufﬁcient amounts of preformed vitamin A and
hence compensate for the reduced production of retinoids from
In 1970, one of the present authors (D.E.S.) (17) published
a case in which the subject had increasing yellowness of the
skin but otherwise appeared healthy. The patient had very high
serum b-carotene concentrations; the mean of 10 tests was
14.8 mmol/L, whereas published normal values range from
0.9–3.7 mmol/L (17). The patient’s serum vitamin A level was
low to low normal. Excessive intake of b-carotene–rich food
was excluded as the cause of the hypercarotenemia and it was
suggested that the symptoms were caused by an abnormality
of the conversion of b-carotene into vitamin A. In this study,
we identiﬁed a novel T170M mutation in the BCMO1 gene in
this patient. The replacement of the highly conserved threo-
nine with methionine results in a ;90% reduction in activity
when analyzed in vitro using puriﬁed recombinant enzymes.
We also describe 1 novel single nucleotide polymorphism (SNP) in
apparently normal individuals.
Materials and Methods
Mutation detection. Genomic DNA was extracted from white blood
cells 5 d after blood was collected from the patient using the Puregene
DNA Isolation kit (Gentra Systems) and subsequently analyzed by PCR
ampliﬁcation and DNA sequencing. Exon-speciﬁc primers were designed
to anneal to sequences in the ﬂanking introns; hence, this method of
screening allows us to identify mutations that affect classical intron-exon
splice junctions as well as those within exons. The description of the
primers used for PCR ampliﬁcation of BCMO1 exons is presented in
Supplemental Table 1. For exons 5, 7, 8, 9, and 10 of the BCMO1 gene,
the initial denaturation cycle lasted 5 min at 94C, followed by 35 cycles
of 30 s at 94C, 30 s at 55C, and 45 s at 72C, and then a ﬁnal extension
for 5 min at 72C. The conditions were the same for exons 2, 3, 4, 6, and
11 except the annealing temperature was 60C. For exon 1, the anneal-
ing temperature was 57C and the extension time 60 s. The primers used
to amplify part of the BCMO1 promoter were P218: 5#-AAAGAGC-
TCGGCAACATAACAGCAGGCACG-3# and P3pr: 5#-TTTCTCGA-
GTTTTGGCTGGCT-CTCACTTGTCC-3#, and the conditions were as
for the exons but with an annealing temperature of 58C and an ex-
tension time of 30 s. The study was approved by the Institutional Review
Board of the University of Texas Southwestern Medical Center, and
consent was obtained from the subject.
Plasmid constructions. The pCMV-hCMO1 T170 His vector has been
described previously (21) and the vector was used here as template when
threonine 170 was mutated to methionine using a QuikChange Site-
Directed Mutagenesis kit (Stratagene). Primers used for the mutation
were sense 5#-CGGTAAA TCTGGCAA
TGTCACA TCCC-3# and anti-sense
ATTGCCAGATTTACCG-3#. The pCMV-hCMO1
T170 c-myc and pCMV-hCMO1 M170 c-myc plasmids were constructed
by PCR amplifying the 3# part of hCMO1 with primers that introduce 10
amino acids (EQKLISEEDL) for the c-myc tag before the stop codon,
sense 5#-AAGTCTACTGCCAGCCGGAA T TTCTTTATGAAGGC-3# and
TTCTGCTCGGTCAGAGGAGCCCCGTGGC-3#. The c-myc tagged 3#
part was then introduced into the respective pCMV-hCMO1 His-
containing vectors using AccI/SalI restriction enzyme sites, resulting in
an exchange of the His tag for a myc tag. The pFastBac1-hCMO1 T170
c-myc, pFastBac1-hCMO1 M170-His, and pFastBac1-hCMO1 M170
c-myc donor plasmids, for subsequent production of recombinant bacu-
lovirus, were constructed by transfer of the pCMV vector inserts into
pFastBac1 (Life Technologies) using EcoRI and SalI.
Insect cell culture, expression, and puriﬁcation of recombinant
human CMO1 protein. These techniques were performed as previously
described (21), except that 16,000 g supernatant was prepared by adding
2 volumes of homogenization buffer (300 mmol/L NaCl and 50 mmol/L
, pH 7.0) to the insect cell pellet and the cell membranes were
disrupted by 20 strokes using a Kontes SZ 23 tissue grind pestle (Kontes)
followed by centrifugation at 16,000 3 g; 30 min at 4C.
Gel ﬁltration chromatography. The puriﬁed enzymes (5 mg) were
applied on a Sephacryl S-300 (Pharmacia) column (7 3 280 mm) equil-
ibrated with 50 mmol/L Tricine-KOH buffer (pH 8.0), containing 100
mmol/L NaCl, 1 mmol/L dithiothreitol, and in some experiments, 1%
(w:v) octylthioglucoside (OTG) (Pierce) was included in the buffer. The
column was run with a ﬂow rate of 21 mL/h and 0.25-mL fractions were
collected. Then 7.5 mL of each fraction was loaded on 10% SDS-PAGE
and analyzed by immunoblotting for presence of CMO1 protein (21).
The column was calibrated with proteins of known molecular weight,
bovine serum albumin (67 kDa), catalase (232 kDa), and Blue Dextrane
(;2 3 10
CMO1 enzyme assay and HPLC analysis of reaction products. The
reactions were performed as described (21), except that in the in vitro
kinetic analyses, the b-carotene concentration in the reactions varied
between 1 and 32 mmol/L and the puriﬁed CMO1 enzyme amount
varied between 0.25 and 1.6 mg per reaction. Equal amounts of wild-
type and M170 CMO1 enzymes were used in the in vitro wild type/
M170 mixing experiments. In experiments with 16,000 g supernatants
expressing CMO1 enzymes, 16 mL supernatant (460 mg total protein)
and 8 mmol/L b-carotene were used in a 60-min reaction. The reactions
were terminated and analyzed by HPLC as described (21). We calculated
kinetic constants and measured protein concentrations as described (21).
Protein levels of CMO1 protein in 16,000-g supernatant fractions were
quantiﬁed after immunoblotting and exposure to HyBlot CL autoradi-
ography ﬁlm (Denville Scientiﬁc). The ﬁlms were scanned using a Hewlett
Packard ScanJet 5100C and blot images were then quantiﬁed using Multi
Gauge Version 3.0 software (Fujiﬁlm).
Protein electrophoresis and immunoblotting blotting. Protein
electrophoresis and immunoblotting with undiluted hybridoma culture
medium containing mouse anti-CMO1 antibody MAb 1–11 or mouse
anti-c-myc antibody raised against a EQKLISEEDL peptide (gift from Dr.
Mike Roth, UT Southwestern Medical Center at Dallas), or puriﬁed
polyclonal rabbit anti-6xHis antibody (Research Diagnostics) diluted
1:10,000 in Tris buffered saline containing 0.2% Tween 20 was per-
formed as previously described (21).
Amino acid sequence analysis. Amino acid sequences for CMO1 from
8 different species plus the amino acid sequence for apocarotenoid 15,15#-
oxygenase (ACO) from cyanobacterium were included in a CLUSTALW
multiple alignment (22).GenBank accession numbers forCMO1sequences
used: human, AF294900_1; chicken, CAB90825.1; mouse, AF271298_1;
rat, BAB60807.1; dog, XP_546815.1; cow, AAY25023.1; pufferﬁsh,
CAF95764.1; chimpanzee, XP_523435.2; and cyanobacterium Synecho-
cystis sp. 6803 ACO, P74334.
Identiﬁcation of a heterozygous mutation in the gene
encoding CMO1 in a patient with hypercarotenemia and
hypovitaminosis A. DNA sequence analysis of ampliﬁed exons
Mutation in human BCMO1 causes hypercarotenemia 2347
by guest on June 5, 2013jn.nutrition.orgDownloaded from
showed the individual to be heterozygous for an ACG / ATG
allele in exon 5 of the gene, resulting in a T170M substitution in
CMO1 (Supplemental Fig. 1), thus replacing a highly conserved
threonine with a methionine (Fig. 1). To investigate whether the
T170M substitution represents a SNP, we analyzed exon 5 in the
BCMO1 gene in 30 unrelated, apparently normal individuals.
The results showed that all 30 subjects were homozygous for the
C allele that encodes the threonine residue.
Biochemical analysis of recombinant enzymes. Enzyme
assays revealed that the T170M mutation severely compromised
the CMO1 enzyme activity, because the speciﬁc activity of M170
was 0.75 nmol retinalmg protein
compared with 7.02
nmol retinalmg protein
for the wild-type enzyme, cor-
responding to a turnover number (k
) of 0.05 min
, respectively (Table 1; Fig. 2). The Michaelis-Menten
constant ( K
) was 6 mmol/L for both enzymes. When M170 and
the wild-type enzymes were mixed in equal amounts, the k
was 0.22 min
, i.e. ;50% of wild-type activity; hence, the
M170 enzyme did not exert a dominant negative effect on the
In a previous publication, we determined that the puriﬁed
wild-type enzyme migrated on a Sephadex S-300 gel ﬁltration
column as a ;230-kDa oligomer using buffer conditions iden-
tical to those used for enzyme assays, i.e. including the detergent
OTG (21). To determine whether the impaired enzyme activity
of the M170 enzyme was due to a change in the protein’s olig-
omeric state, we subjected the puriﬁed wild-type and M170 en-
zymes to gel ﬁltration chromatography in the enzyme assay
buffer containing the detergent OTG. Both wild-type CMO1
(Fig. 3A) and CMO1-M170 (Fig. 3B) migrated as ;230-kDa
oligomers. To ascertain whether the CMO1 proteins are mono-
mers or oligomers in a detergent-free solution, gel ﬁltration
chromatography experiments were also performed in enzyme
assay buffer without OTG. Both wild type (Fig. 3C) and M170
(Fig. 3D) migrate as ;65-kDa monomeric proteins under
detergent-free conditions, suggesting that the OTG detergent
creates detergent-protein micelles containing oligomeric CMO1
complexes. Furthermore, when assayed for enzyme activity, the
wild-type and M170 enzyme fractions that were eluted from the
column exhibited similar activities regardless of whether the en-
zymes had been chromatographed in the presence or absence of
OTG detergent (data not shown).
The potential effect of the M170 enzyme on the wild-type
enzyme was further examined by coexpressing c-myc-tagged wild-
type CMO1 and His-tagged M170 enzymes, as well as express-
ing the 2 variants individually. The relative amounts of the wild
type and the mutant enzyme in the 16,000-g supernatants were
determined by immunoblotting, utilizing antibodies against the
CMO1 enzyme, c-myc, and 6xHis, respectively (Fig. 4B). The ac-
tivity of the wild-type CMO1 16,000-g supernatant was 2.6 pmol
retinal formedmg protein
, which represents 100% ac-
tivity (Fig. 4A, lane 1). The results demonstrate that the activity
of CMO1 M170 was ;8% of wild-type CMO1 (Fig. 4A, lane
3), whereas the fraction that contained 85% of M170 and 15%
of wild-type CMO1 enzyme had an activity corresponding to
19% of the wild-type protein (Fig. 4A, lane 2). Furthermore,
when His-tagged wild-type CMO1 was coexpressed in Sf-9 cells
with c-myc-tagged CMO1 M170 and subsequently Co
afﬁnity puriﬁed, CMO1 M170 was not copuriﬁed with the wild-
type enzyme (data not shown).
Promoter sequence analysis. Because the subject was found
to be heterozygous for the T170M mutation, we wanted to
explore the possibility of mutations in the BCMO1 promoter by
sequence analysis of a 218-bp region (291 to 127; 1 1 being the
transcriptional start site) of the most proximal part of the
promoter, which includes a potential TATA-box (23) as well as a
potential response element for the nuclear PPAR. These sites in
the proximal mouse BCMO1 promoter have been shown to be
sufﬁcient to drive both basal and speciﬁc promoter activity (24).
The analysis revealed that both alleles in the patient were iden-
tical with the corresponding BCMO1 promoter region in the 2
human genomic sequences (NCBI accession nos. NT010498 and
NW926528), which include the BCMO1 gene on chromosome
16q21–23 (data not shown).
Novel BCMO1 SNP. When we analyzed the BCMO1 gene in
our patient DNA and the sample collection from 30 normal
individuals, we discovered a novel R228C SNP in exon 6 (Fig. 5).
The polymorphic Tallele at position 228 (
CGC / TGC) encodes
TABLE 1 Kinetic constants of CMO1 variants
CMO1 variant K
Wild type 6.0 6 1.9 0.45 6 0.05
Wild type/M170 5.0 6 1.5 0.22 6 0.02
M170 6.0 6 1.4 0.05 6 0.003
value in the table represents the mean 6 SD of 2–6 independent
experiments performed in duplicate.
FIGURE 1 Partial amino acid sequence alignment of 8 different
CMO1 isozymes and 1 carotenoid cleaving enzyme from a cyanobac-
terium. The highly conserved threonine residue is indicated with an
arrowhead and the His involved in binding of the catalytic iron in the
active site is highlighted with black. The amino acid residues are
numbered to the right of the sequences.
FIGURE 2 In vitro kinetic analysis of wild-type and M170 CMO1
2348 Lindqvist et al.
by guest on June 5, 2013jn.nutrition.orgDownloaded from
a protein with cysteine rather than arginine and was only found
in heterozygous form. The frequency of the C allele (Arg) was 0.98
and that of the T allele (Cys) was 0.02. We also found the 2 pre-
viously characterized SNP, R267S, and A379V (NCBI’s dbSNP;
refSNP ID rs12934922 and rs7501331, respectively). The indi-
vidual with hypercarotenemia described in this article was ho-
mozygous for R228, S267, and A379.
Sequence analysis of genes involved in carotenoid me-
tabolism and vitamin A biosynthesis. Because the subject
was heterozygous for the T170M version, we analyzed other genes
known to be involved in carotenoid metabolism and vitamin A
synthesis. By using our standard protocol for primer design and
PCR ampliﬁcation, we analyzed exons and ﬂanking regions in the
genes encoding carotenoid 9#,10#-monooxygenase, cellular ret-
inol binding protein I and II, and retinol dehydrogenase types 11,
12, and 14. All these genes were found to be normal in the subject
(data not shown).
In this study, we reported the identiﬁcation of a mutation in the
BCMO1 gene in a subject with hypercarotenemia and mild hy-
povitaminosis A. The T170M mutation results in a ;90% loss
edge of a loss-of-function mutation in the human BCMO1 gene.
Threonine at position 170 is one of the more conserved res-
idues among the CMO1 isozymes of different species. Although
the function of this residue is not clear, substitution of this small
hydrophilic threonine with a bulky hydrophobic methionine re-
sults in a severe attenuation of enzymatic activity. T170 is posi-
tioned in the proximity of H172, which is most likely 1 of the
4 His that coordinate the iron atom in the catalytic center of the
enzyme (25). It should be mentioned that the structure of ACO
from the cyanobacterium Synechocystis sp. 6803 was recently
characterized (26). ACO belongs to the same protein family as
CMO1 and several other iron-containing carotenoid cleaving
enzymes and the catalytic iron (II) ion that deﬁnes the active
center in ACO is buried in a tunnel-like cavity that holds the
substrate. One of the iron-coordinating His residues in ACO
corresponds to H172 in CMO1 and this His residue forms a
stabilizing hydrogen bond with the serine residue in ACO that
corresponds to T170 in human CMO1 (Fig. 1). Hence, the me-
thionine at position 170 of the enzyme may destabilize binding
or change the position of the iron (II), thereby compromising the
enzyme’s catalytic capacity. The observation that the M170 en-
zyme’s afﬁnity for b-carotene is identical to the wild-type T170
enzyme suggests that this residue is not crucial for substrate
The T170M mutation was only found in a single allele of the
subject and no other mutations were identiﬁed. With our method
of screening, we would have found any mutations that affected
the exons or the classical exon-intron splice junctions. However,
it is conceivable that the other allele contains a mutation that
creates a splice donor or acceptor site distant from the exon-
intron junctions. This would, in turn, cause the formation of an
aberrantly spliced CMO1 messenger RNA, a hypothesis that
could have been tested by RT-PCR, but CMO1 messenger RNA-
containing tissue from the patient was not available. Another
explanation would be that the wild-type allele could contain a
mutation in the promoter region of the CMO1 gene, a hypoth-
esis we tested by sequencing 218 bp of its proximal 5#-ﬂanking
region; however, this region was wild type in both alleles. This
observation does not, however, exclude the possibility of the
existence of a more distant promoter mutation.
In a previous study, we described that the puriﬁed recombi-
nant CMO1 enzyme exists as a 230-kDa tetramer in solution as
revealed by gel ﬁltration chromatography (21). Because our aim
then was to ensure that the protein fractions detected by immu-
noblotting were in a catalytically active form, we performed the
FIGURE 5 Mutation and SNP in the human BCMO1 gene. Exons
are indicated by blocks and Roman numerals. Introns are not drawn to
scale with the exception of intron 6. The star denotes the TGA stop
FIGURE 4 Enzyme activity assays of wild-type and M170 CMO1
enzymes in 16,000-g supernatants of infected Sf-9 cells (A) and
relative amounts of respective proteins visualized by immunoblotting
(B). Each value represents the mean 6 SD of 1 experiment performed
FIGURE 3 Gel filtration chromatography of wild-type and M170
CMO1 enzymes visualized by immunoblotting analysis. Arrowheads
mark where the standards, bovine serum albumin (67 kDa), and cata-
lase (232 kDa) eluted.
Mutation in human BCMO1 causes hypercarotenemia 2349
by guest on June 5, 2013jn.nutrition.orgDownloaded from
gel ﬁltration experiments under buffer conditions identical to
those used in the enzyme assays with 1% OTG detergent in the
elution buffer; hence, experimentally created detergent-protein
micelles could be the cause of the migration as a tetrameric
complex. In an effort to revisit this question and to explain the
observation that a heterozygous mutation causes a disease phe-
notype, we tested whether the M170 enzyme could have a dom-
inant negative effect on the wild-type T170 subunit as has been
demonstrated with several tetrameric enzymes, e.g. mitochon-
drial acetaldehyde dehydrogenase (27,28). If it is assumed that
both T170 and M170 are present in equal numbers and that the
4-subunit assembly is random, the mix of different combinations
would result in a total enzyme activity much less than 50% of
wild-type activity if the M170 subunit has a dominant negative
effect on the complex. On the other hand, if there is no
interaction between the wild-type and mutant enzymes, the
heterozygote will have a total activity of ;50% of the wild-type
activity. Reconstitution experiments with puriﬁed T170 and
M170 enzymes and coexpression of the 2 enzymes in cultured
cells revealed an additive, not dominant negative, effect of the
M170 enzyme on the wild-type T170 enzyme, suggesting that
the polypeptide chains catalyze chemistry independent of each
other. Furthermore, gel ﬁltration chromatography demonstrated
that CMO1 is a monomeric enzyme in the absence of detergent.
Taken together, these experiments strongly suggest that the
CMO1 molecules operate as monomers independently of each
other, which leads us to the question whether CMO1 deﬁciency
is another example where a heterozygous loss-of-function allele
causes a disease phenotype, i.e. haploinsufﬁciency.
Haploinsufﬁciency can be exempliﬁed by the various forms
of porphyria, which are inherited and acquired disorders in which
the activities of the enzymes of the heme biosynthetic pathway
are partially deﬁcient (29). There are 8 enzymes involved in the
synthesis of heme and an enzymatic defect in any of the 8 en-
zymes in the pathway, with the exception of the ﬁrst enzyme,
results in a 50% reduction in activity and causes a porphyria
phenotype. Because ample amounts of b-carotene and other ca-
rotenoids are present in plasma of numerous species, including
humans, our observations supports the notion that the CMO1
enzyme is normally saturated with substrate. Thus, even though
we have found a heterozygous mutation in only 1 subject, our
data suggest that CMO1 deﬁciency manifests itself in an autoso-
mal dominant fashion.
In conclusion, haploinsufﬁciency of the CMO1 enzyme may
cause symptoms of hypercarotenemia and hypovitaminosis A in
individuals consuming a carotenoid-containing and retinol-
inadequate diet. Especially susceptible would be individuals on
a pure vegetarian diet lacking preformed vitamin A, a situation
common in developing countries where the major source of vi-
tamin A is dietary provitamin A carotenoids. Further studies of
individuals with hypercarotenemia and hypovitaminosis A will
be necessary to assess the correlation of mutations in CMO1
with these phenotypes.
1. Goodman D, Huang H. Biosynthesis of vitamin A with rat intestinal
enzymes. Science. 1965;149:879–80.
2. Olson JA, Hayaishi O. The enzymatic cleavage of b-carotene into vita-
min A by soluble enzymes of rat liver and intestine. Proc Natl Acad Sci
3. Blaner WS, Olson JA. Retinol and retinoic acid metabolism. In: Sporn
MB, Roberts AB, Goodman DS, editors. The retinoids. New York: Raven
Press; 1994. p. 232–40.
4. Goodman DS, Blomstrand R, Werner B, Huang HS, Shiratori T. The
intestinal absorption and metabolism of vitamin A and b-carotene in
man. J Clin Invest. 1966;45:1615–23.
5. Blomstrand R, Werner B. Studies on the intestinal absorption of
radioactive b-carotene and vitamin A in man. Conversion of b-carotene
into vitamin A. Scand J Clin Lab Invest. 1967;19:339–45.
6. Dimitrov NV, Meyer C, Ullrey DE, Chenoweth W, Michelakis A,
Malone W, Boone C, Fink G. Bioavailability of b-carotene in humans.
Am J Clin Nutr. 1988;48:298–304.
7. van Vliet T. Absorption of b-carotene and other carotenoids in humans
and animal models. Eur J Clin Nutr. 1996;50 Suppl 3:S32–7.
8. Castenmiller JJ, West CE. Bioavailability and bioconversion of carot-
enoids. Annu Rev Nutr. 1998;18:19–38.
9. Matthews-Roth MM, Gulbrandsen CL. Transport of b-carotene in
serum of individuals with carotenemia. Clin Chem. 1974;20:1578–9.
10. Bjornson LK, Kayden HJ, Miller E, Moshell AN. The transport of
a-tocopherol and b-carotene in human blood. J Lipid Res. 1976;17:
11. Forman MR, Johnson EJ, Lanza E, Graubard BI, Beecher GR, Muesing
R. Effect of menstrual cycle phase on the concentration of individual
carotenoids in lipoproteins of premenopausal women: a controlled die-
tary study. Am J Clin Nutr. 1998;67:81–7.
12. Lindqvist A, Andersson S. Cell type-speciﬁc expression of b-carotene
15,15#-mono-oxygenase in human tissues. J Histochem Cytochem. 2004;
13. Bendich A, Olson JA. Biological actions of carotenoids. FASEB J. 1989;
14. Sommer A. Clinical research and the human condition: moving from
observation to practice. Nat Med. 1997;3:1061–3.
15. Cohen H. Observations on carotenemia. Ann Intern Med. 1958;48:
16. Frenk E. Etat keratodermique avec taux serique abaisse de la vitamine A
et hypercarotinemie. Dermatologica. 1965;132:96–8.
17. Sharvill DE. Familial hypercarotinaemia and hypovitaminosis A. Proc R
Soc Med. 1970;63:605–6.
18. McLaren DS, Zekian B. Failure of enzymic cleavage of b-carotene. The
cause of vitamin A deﬁciency in a child. Am J Dis Child. 1971;121:
19. Monk BE. Metabolic carotenaemia. Br J Dermatol. 1982;106:485–7.
20. Svensson A, Vahlquist A. Metabolic carotenemia and carotenoderma in
a child. Acta Derm Venereol. 1995;75:70–1.
21. Lindqvist A, Andersson S. Biochemical properties of puriﬁed recombinant
human b-carotene 15,15#-monooxygenase. J Biol Chem. 2002;277:
22. Higgins DG, Thompson JD, Gibson TJ. Using CLUSTAL for multiple
sequence alignments. Methods Enzymol. 1996;266:383–402.
23. Yan W, Jang G-F, Haeseleer F, Esumi N, Chang J, Kerrigan M,
Campochiaro M, Campo chiaro P, Palczewski K, et al. Cloning and
characterization of a human b,b -carotene-15,15#-dioxygenase that is
highly expressed in retinal pigment epithelium. Genomics. 2001;72:
24. Boulanger A, McLemore P, Copeland NG, Gilbert DJ, Jenkins NA, Yu
SS, Gentleman S, Redmond TM. Identiﬁcation of b-carotene 15,15#-
monooxygenase as a peroxisome proliferator-activated receptor target
gene. FASEB J. 2003;17:1304–6.
25. Poliakov E, Gentleman S, Cunningham FX Jr, Miller-Ihli NJ, Redmond
TM. Key role of conserved histidines in recombinant mouse b-carotene
xygenase-1 activity. J Biol Chem. 2005;280:29217–23.
26. Kloer DP, Ruch S, Al-Babili S, Beyer P, Schulz GE. The structure
of a retinal-forming carotenoid oxygenase. Science. 2005;308:267–9.
27. Weiner H, Wei B, Zhou J. Subunit communication in tetrameric class 2
human liver aldehyde dehydrogenase as the basis for half-of-the-site
reactivity and the dominance of the oriental subunit in a heterotetramer.
Chem Biol Interact. 2001;130–132:47–56.
28. Crabb DW, Edenberg HJ, Bosron WF, Li TK. Genotypes for aldehyde
dehydrogenase deﬁciency and alcohol sensitivity. The inactive ALDH2(2)
allele is dominant. J Clin Invest. 1989;83:314–6.
29. Anderson KE, Sassa S, Bishop DF, Desnick RJ. Disorders of heme
biosynthesis: X-linked sideroblastic anemia and porphyrias. In: Scriver
CR, Beaudet AL, Sly WS, Valle D, editors. The metabolic and molecular
bases of inherited disease. New York: McGraw-Hill; 2001. p. 2991–
2350 Lindqvist et al.
by guest on June 5, 2013jn.nutrition.orgDownloaded from