Molecular genetics of biotin metabolism: old vitamin, new scienceB
Roy A. Gravel*, Monica A. Narang
Department of Biochemistry and Molecular Biology, University of Calgary, Calgary, Alberta, Canada T2N 4N1
Received 30 March 2005; received in revised form 30 March 2005; accepted 30 March 2005
Biotin is a water-soluble vitamin that participates as a cofactor in gluconeogenesis, fatty acid synthesis and branched chain amino acid
catabolism. It functions as the carboxyl carrier for biotin-dependent carboxylases. Its covalent attachment to carboxylases is catalyzed by
holocarboxylase synthetase. Our interest in biotin has been through the genetic disease, bbiotin-responsive multiple carboxylase deficiency,Q
caused by deficient activity of holocarboxylase synthetase. As part of these studies, we made the unexpected findings that the enzyme also
targets to the nucleus and that it catalyzes the attachment of biotin to histones. We found that patients with holocarboxylase synthetase
deficiency have a much reduced level of biotinylated histones, yet the importance of this process is unknown. The dual nature of biotin, as the
carboxyl-carrier cofactor of carboxylases and as a ligand of unknown function attached to histones, is an enigma that suggests a much more
involved role for biotin than anticipated. It may change our outlook on the optimal nutritional intake of biotin and its importance in biological
processes such as development, cellular homeostasis and regulation.
D 2005 Elsevier Inc. All rights reserved.
Keywords: Biotin; Holocarboxylase synthetase; Carboxylases; Histones; Nucleus; Multiple carboxylase deficiency
1. The biotin cycle and the biotin-dependent
Biotin has been recognized as an essential nutrient since
the early part of the last century. We now appreciate that our
biotin requirement is fulfilled in part through diet, through
endogenous reutilization of biotin and perhaps through
capture of biotin generated in the intestinal flora . The
utilization of biotin for covalent attachment to carboxylases
and its reutilization through the release of carboxylase biotin
after proteolytic degradation constitutes the bbiotin cycle.Q
Biotin deficiency is associated with neurological manifes-
tations, skin rash, hair loss and metabolic disturbances that
are thought to relate to the various carboxylase deficiencies
(metabolic ketoacidosis with lactic acidosis). It has also been
suggested that biotin deficiency is associated with protein
malnutrition [2,3], and that marginal biotin deficiency in
pregnant women may be teratogenic . While these data
highlight the importance of biotin as an essential nutrient, the
role of biotin in cells remains incompletely understood.
Biotin acts as a carboxyl carrier in carboxylation
reactions . There are four biotin-dependent carboxylases
in mammals: those of propionyl-CoA (PCC), h-methylcro-
tonyl-CoA (MCC), pyruvate (PC) and acetyl-CoA carbo-
xylases (isoforms ACC-1 and ACC-2). All but ACC-2 are
mitochondrial enzymes. The biotin moiety is covalently
bound to the q amino group of a Lys residue in each of these
carboxylases in a domain 60–80 amino acids long. The
domain is structurally similar among carboxylases from
bacteria to mammals. At the center is a short peptide,
(A/V)MKM, which is the near universal biotin acceptor
sequence. HCS catalyses the transfer of biotin to all of the
Our knowledge of the reaction mechanism of HCS
comes from studies of the orthologous Escherichia coli
protein, BirA. It functions both as the biotin transfer enzyme
and as the repressor of the biotin biosynthetic operon .
Biotin is transferred, via a two step reaction involving a
biotin-5V-AMP intermediate, to the biotin carboxyl carrier
protein (BCCP) of ACC with consequent release of AMP.
0955-2863/$ – see front matter D 2005 Elsevier Inc. All rights reserved.
BThis paper was presented at the bInternational Symposium: Vitamins
as Regulators of Genetic Expression: Biotin as a ModelQ NAFTA Satellite
Meeting to the XXV National Congress of Biochemistry held December 3–
4, 2004, in Ixtapa, Zihuatanejo, Mexico. This meeting was sponsored by
Sociedad Mexicana de Bioquimica A.C.; Programa de Doctorado en
Ciencias Biomedicas, Universidad Nacional Autonoma de Mexico;
Laboratorios Roche-Syntex, Mexico; and Instituto de Investigaciones
Biomedicas, Universidad Nacional Autonoma de Mexico.
T Corresponding author. Tel.: +1 403 220 2268.
E-mail address: email@example.com (R.A. Gravel).
Journal of Nutritional Biochemistry 16 (2005) 428–431
Xu and Beckett  showed that binding of biotin-AMP to
BirA produces a significant conformational shift (allosteric
effect) that stabilizes the BirA-biotin-AMP complex. This
prevents unproductive synthesis and release of biotin-AMP
when BCCP is fully saturated with biotin. The BirA-biotin-
AMP complex also shifts to a dimer form which constitutes
the active repressor of the biotin operon [7–9].
The cloning of the cDNA for HCS made it possible to
compare its sequence structure with that of BirA [10,11].
While HCS and BirA share the biotin ligase function, their
structural relatedness is limited to ~130 amino acids (60%
identity) comprising the biotin transfer domain, despite
lengths of 726 amino acids for HCS and 325 for BirA. The
N-terminal half of BirA is concerned primarily with its
repressor function and contains the domain for DNA
binding. The function of the N-terminal half of HCS
remains unknown. For both proteins, the N-terminal half
can be eliminated without disrupting its biotin ligase
function [6,12]. The structure of the HLCS gene
(ENSG00000159267) has been determined. It is located
on chromosome 21q22.1 and consists of 14 exons and 13
introns in a span of 240 kb .
Among the carboxylases, genetic diseases have been
described for all four enzymes individually . In MCD, all
carboxylase activities are simultaneously deficient owing to
universally defective biotinylation. There are two genetical-
ly distinct causes: mutations in HCS or in biotinidase
(BTD). In HCS deficiency, the defect is in biotin transfer to
apocarboxylases. In BTD deficiency, the defect lies in
blocking reentry of biotin into the biotin cycle due to
defective release of biotin from biocytin (biotinyllysine), the
product of the proteolysis of carboxylases. HCS deficiency
may be life threatening during infancy. Biotinidase defi-
ciency tends to be of later onset and has milder symptoms
. Both diseases can be treated with pharmacological
supplementation with biotin, although those with BTD
deficiency retain hearing and visual deficits that can be
avoided through presymptomatic screening. Both enzymes
are found in all cells, and BTD is also a significant plasma
glycoprotein. Burri et al. [14,15] showed that most patients
with defective HCS had a decreased affinity for biotin, and,
following the cDNA cloning, we and others showed that
most patient mutations occur in the biotin binding region of
HCS, consistent with Burri’s expectations [13,16].
2. The nucleus, histones and biotin
Early studies revealed that a large proportion of
radioactive biotin injected into chicks and rats localized to
the nuclear fraction of cells . Some studies have
reported biotin in nuclei of tumor material and normal
tissues [18–20]. The main source of nuclear biotin may be
through covalent attachment to histones. Stanley et al. 
showed that all five histone classes extracted from human
lymphocytes contain biotin that was detected by Western
blot using streptavidin or anti-biotin. This experiment
derives from the earlier discovery by Hymes et al.  that
BTD will exchange biotin between biocytin and histones in
vitro. The reaction was demonstrated with serum samples
and purified BTD and was deficient when sera of patients
with BTD deficiency were used. The reverse reaction,
removal of biotin from histones, was also demonstrated
using human plasma or lymphocyte extracts and was also
deficient in patient samples . The reaction did not work
with free biotin as substrate.
Independently of these studies, we generated polyclonal
antibodies to HCS to permit subcellular localization of the
enzyme. At issue was whether biotinylation of mitochon-
drial carboxylase occurs within the mitochondria or in
transit in the cytosol. Three antibodies were produced, two
against peptides corresponding to sequences at the mature
N- and C-terminus of HCS (residues, 58–77 and 707–726,
respectively) and a third to near full-length HCS (residues,
58–726) that was expressed in E. coli. While the antibodies
were used to examine the cytoplasmic distribution of HCS,
they instead showed it primarily in the nucleus of HeLa,
Hep2 and fibroblasts with only a minor component in the
cytoplasm . This was completely unexpected, but the
same result was obtained by transfecting recombinant HCS
containing a C-terminal tag. HCS localized to the core
nuclear lamina in cells progressively extracted through
increasing salt and DNase I treatment. This parallels
immunofluorescent detection of lamin B, except that the
HCS distribution was discontinuous. DNase I did not
significantly alter its distribution, suggesting that HCS is
not tightly associated with chromatin. HCS was also
excluded from condensed chromosomes and was found to
be dispersed as particulate ring-like structures in all phases
of mitosis. Given this distribution, mitotic Hep2 and HeLa
cells were examined for possible colocalization of HCS with
lamin B. This was done in three dimensions by the analysis
of optical stacks that showed N98% colocalization of HCS
with lamin B. The in situ results were confirmed by Western
blot of fractionated HeLa and Hep2 cells. All three
antibodies recognized two protein species of 68 and 66
kDa primarily in the chromatin and nuclear matrix fractions.
These results suggest that particulate structures containing
HCS derive from the disassembly of the nuclear envelope
and indicate that their macromolecular nature is retained,
rather than disaggregated, during mitosis.
Assay of biotin transfer to p67, a peptide comprised of the
C-terminal 67 amino acids of the PCC a subunit, showed
that HCS from the nuclear matrix and insoluble fractions
retain biotinylating activity. Given the report of biotinylated
histones , we assayed purified recombinant enzyme and
showed that it could transfer14C-biotin to all histone classes
(H1, H2A, H2B, H3 and H4) in the presence of ATP .
Significantly, the substrate was free biotin. This result led us
to examine the state of histone biotinylation in MCD cells
deficient in HCS activity. Histones were extracted from
control and patient fibroblasts and evaluated by avidin and
anti-biotin Western blot. In control cells, all five histone
R.A. Gravel, M.A. Narang / Journal of Nutritional Biochemistry 16 (2005) 428–431
classes were detected by Coomassie stain and all contained
biotin. In contrast, mutant cells showed similar histone levels
but were profoundly diminished in biotinylation across all
five classes. These results provide convincing evidence that
HCS is responsible for histone biotinylation. We suggest that
BTD, which requires biocytin as substrate, does not attach
biotin to histones in vivo. Rather, it may well be respon-
sible for removal of biotin from histones as it does for
Unexpectedly, we also observed that nuclear HCS
contains attached biotin—unexpected because HCS
does not contain a consensus biotin attachment site (i.e.,
(A/V)KMK). Nevertheless, anti-biotin capture of proteins
from HeLa cell nuclei followed by Western blot with
streptavidin revealed a series of biotin-containing proteins,
including bands with the pattern of histones . One of
the bands, at 68 kDa, was also detected with anti-HCS.
The reciprocal experiment, captured with anti-HCS and
Western blot with antibiotin, also revealed a 68-kDa band.
Blotting with anti-HCS confirmed the presence of both the
68- and 66-kDa HCS species. We also showed that the
66-kDa species from permeabilized interphase or mitotic
cells could be solubilized with the nonionic detergent
NP-40 while the 68-kDa species remained in the pellet.
This reinforces the view that the 68 kDa species is
associated with other proteins or cell structures.
These results suggest additional roles for biotin and
HCS beyond attachment of the biotin cofactor to
carboxylases. There have been long standing suggestions
of a role for biotin in genetic regulation. Biotin has been
reported to stimulate the synthesis of hepatic glucokinase
[25,26] and to repress phosphoenolpyruvate carboxyki-
nase activity  in rat liver. It was reported to stimulate
an increase in the mRNA for 6-phosphofructokinase
following biotin administration to biotin-deficient rats
[28,29]. It was also shown to increase the level of the
asialoglycoprotein receptor in hepatoma cells using a
mechanism that appears to function through cGMP and
guanylate cyclase [30–32]. More recently, Stanley et al.
and Crisp et al. [21,33] suggested that cell proliferation is
linked to biotin content.
Significantly, biotin appears to regulate the expression of
HCS and some carboxylases. Solorzano-Vargas et al. 
reported that in biotin-starved human hepatoma cells, the
level of the mRNAs for HCS, ACC-1 and the a subunit of
PCC were all reduced and were restored to initial levels on
resupplementation with biotin. They showed that the effect
was RNA synthesis dependent and that cells from a patient
with MCD required 100? the biotin level used for control
cells to restore the mRNAs to starting levels. They showed
that cGMP can bypass the biotin requirement, in control or
MCD cells, and that inhibitors of soluble guanylate cyclase
or of cGMP-dependent protein kinase inhibited the cGMP
effect. Given the involvement of HCS in the sequence of
events, they suggested that biotin-AMP, the intermediate
product of the HCS reaction, is a component of the
stimulatory process. Independent studies on biotin-deficient
rats demonstrated a similar effect on HCS mRNA and
protein and on PCC and PC protein levels . Subsequent
studies in a rat model revealed that this mechanism of
regulation was restricted to tissues such as liver and kidney,
while in the brain, the biotin cycle remained constitutively
expressed. Leon Del Rio proposed that this pattern of
regulation is aimed at restricting biotin utilization in
peripheral tissues while sparing the brain during periods
of biotin deprivation. These studies suggest a role for biotin
that goes beyond maintenance of functional carboxylase
activities and suggest further that biotin-AMP may be an
intermediate in this process.
In summary, these experiments reveal a dual role for
HCS: its traditional role in the biotinylation of carboxylases
and a novel role in the attachment of biotin to histones. Our
studies suggest a complex pathway of nuclear localization,
retention and organization and deficient histone biotinyla-
tion in MCD cells. The current studies of the role of biotin
in cells may change our outlook on its optimal nutritional
intake and the meaning behind the symptoms of individuals
with biotin deficiency, genetic or acquired. A benefit of
these studies will be to increase our understanding of the
disease in MCD and the long-term impact of treatment with
pharmacological doses of biotin.
Grant support was from the Canadian Institute of Health
Research. M.N. received a fellowship from the Alberta
Heritage Medical Research Foundation.
 Wolf B. Disorders of biotin metabolism. In: Scriver CR, Beaudet AL,
Sly WS, Valle D, editors. The metabolic and molecular bases of
inherited disease. New York7 McGraw-Hill; 2001. p. 3935–62.
 Velazquez A. Biotin deficiency in protein-energy malnutrition:
implications for nutritional homeostasis and individuality. Nutrition
 Velazquez A, Martin-del-Campo C, Baez A, Zamudio S, Quiterio M,
Aguilar JL, et al. Biotin deficiency in protein-energy malnutrition.
Eur J Clin Nutr 1989;43:169–73.
 Zempleni J, Mock DM. Marginal biotin deficiency is teratogenic.
Proc Soc Exp Biol Med 2000;223:14–21.
 Cronan Jr JE. The E. coli bio operon: transcriptional repression by an
essential protein modification enzyme. Cell 1989;58:427–9.
 Xu Y, Beckett D. Evidence for interdomain interaction in the
Escherichia coli repressor of biotin biosynthesis from studies of an
N-terminal domain deletion mutant. Biochemistry 1996;35:1783–92.
 Kwon K, Streaker ED, Ruparelia S, Beckett D. Multiple disordered
loops function in corepressor-induced dimerization of the biotin
repressor. J Mol Biol 2000;304:821–33.
 Streaker ED, Gupta A, Beckett D. The biotin repressor: thermody-
namic coupling of corepressor binding, protein assembly, and
sequence-specific DNA binding. Biochemistry 2002;41:14263–71.
 Streaker ED, Beckett D. Coupling of protein assembly and DNA
binding: biotin repressor dimerization precedes biotin operator
binding. J Mol Biol 2003;325:937–48.
R.A. Gravel, M.A. Narang / Journal of Nutritional Biochemistry 16 (2005) 428–431
 Suzuki Y, Aoki Y, Ishida Y, Chiba Y, Iwamatsu A, Kishino T, et al.
Isolation and characterization of mutations in the human holocarbox-
ylase synthetase cDNA. Nat Genet 1994;8:122–8.
 Leon-Del-Rio A, Leclerc D, Akerman B, Wakamatsu N, Gravel RA.
Isolation of a cDNA encoding human holocarboxylase synthetase by
functional complementation of a biotin auxotroph of Escherichia coli.
Proc Natl Acad Sci U S A 1995;92:4626–30.
 Campeau E, Gravel RA. Expression in Escherichia coli of N- and
C-terminally deleted human holocarboxylase synthetase. Influence of
the N-terminus on biotinylation and identification of a minimum
functional protein. J Biol Chem 2001;276:12310–6.
 Yang X, Aoki Y, Li X, Sakamoto O, Hiratsuka M, Kure S, et al.
Structure of human holocarboxylase synthetase gene and mutation
spectrum of holocarboxylase synthetase deficiency. Hum Genet 2001;
 Burri BJ, Sweetman L, Nyhan WL. Mutant holocarboxylase synthe-
tase:evidence for the enzymedefect in early infantile biotin-responsive
multiple carboxylase deficiency. J Clin Invest 1981;68:1491–5.
 Burri BJ, Sweetman L, Nyhan WL. Heterogeneity of holocarboxylase
synthetase in patients with biotin-responsive multiple carboxylase
deficiency. Am J Hum Genet 1985;37:326–37.
 Dupuis L, Leon-Del-Rio A, Leclerc D, Campeau E, Sweetman L,
Saudubray JM, et al. Clustering of mutations in the biotin-binding
region of holocarboxylase synthetase in biotin-responsive multiple
carboxylase deficiency. Hum Mol Genet 1996;5:1011–6.
 Dakshinamurti K, Mistry SP. Tissue and intracellular distribution of
biotin-C14OOH in rats and chicks. J Biol Chem 1963;238:294–334.
 Nakatani Y, Kitamura H, Inayama Y, Ogawa N. Pulmonary
endodermal tumor resembling fetal lung. The optically clear nucleus
is rich in biotin. Am J Surg Pathol 1994;18:637–42.
 Yokoyama S, Kashima K, Inoue S, Daa T, Nakayama I, Moriuchi A.
Biotin-containing intranuclear inclusions in endometrial glands during
gestation and puerperium. Am J Clin Pathol 1993;99:13–7.
 Hasegawa Y, Ishida Y, Kato K, Ijiri R, Miyake T, Nishimata S, et al.
Pancreatoblastoma. A case report with special emphasis on squamoid
corpuscles with optically clear nuclei rich in biotin. Acta Cytol
 Stanley JS, Griffin JB, Zempleni J. Biotinylation of histones in human
cells. Effects of cell proliferation. Eur J Biochem 2001;268:5424–9.
 Hymes J, Fleischhauer K, Wolf B. Biotinylation of histones by human
serum biotinidase: assessment of biotinyl-transferase activity in sera
from normal individuals and children with biotinidase deficiency.
Biochem Mol Med 1995;56:76–83.
 Ballard TD, Wolff J, Griffin JB, Stanley JS, van Calcar S, Zempleni J.
Biotinidase catalyzes debiotinylation of histones. Eur J Nutr
 Narang MA, Dumas R, Ayer LM, Gravel RA. Reduced histone
biotinylation in multiple carboxylase deficiency patients: a nuclear
role for holocarboxylase synthetase. Hum Mol Genet 2004;13:15–23.
 Dakshinamurti K, Cheah-Tan C. Biotin-mediated synthesis of hepatic
glucokinase in the rat. Arch Biochem Biophys 1968;127:17–21.
 Chauhan J, Dakshinamurti K. Transcriptional regulation of the
glucokinase gene by biotin in starved rats. J Biol Chem 1991;266:
 Dakshinamurti K, Li W. Transcriptional regulation of liver phospho-
enolpyruvate carboxykinase by biotin in diabetic rats. Mol Cell
 Deodhar AD, Mistry SP. Regulation of glycolysis in biotin-deficient
rat liver. Life Sci II 1970;9:581–8.
 Dakshinamurti K, Tarrago-Litvak L, Hong HC. Biotin and glucose
metabolism. Can J Biochem 1970;48:493–500.
 Collins JC, Paietta E, Green R, Morell AG, Stockert RJ. Biotin-
dependent expression of the asialoglycoprotein receptor in HepG2.
J Biol Chem 1988;263:11280–3.
 Stockert RJ, Paietta E, Racevskis J, Morell AG. Posttranscriptional
regulation of the asialoglycoprotein receptor by cGMP. J Biol Chem
 Stockert RJ, Morell AG. Second messenger modulation of the
asialoglycoprotein receptor. J Biol Chem 1990;265:1841–6.
 Crisp SE, Griffin JB, White BR, Toombs CF, Camporeale G, Said
HM, et al. Biotin supply affects rates of cell proliferation, biotinylation
of carboxylases and histones, and expression of the gene encoding the
sodium-dependent multivitamin transporter in JAr choriocarcinoma
cells. Eur J Nutr 2004;43:23–31.
 Solorzano-Vargas RS, Pacheco-Alvarez D, Leon-Del-Rio A. Holo-
carboxylase synthetase is an obligate participant in biotin-mediated
regulation of its own expression and of biotin-dependent carboxylases
mRNA levels in human cells. Proc Natl Acad Sci U S A 2002;99:
 Rodriguez-Melendez R, Cano S, Mendez ST, Velazquez A. Biotin
regulates the genetic expression of holocarboxylase synthetase and
mitochondrial carboxylases in rats. J Nutr 2001;131:1909–13.
R.A. Gravel, M.A. Narang / Journal of Nutritional Biochemistry 16 (2005) 428–431