Vitamin A and cancer

Vitamin A and cancer
7.10.1 Retinoids and apoptosis
Apoptosis is a tightly regulated natural process of cell
death, which involves changes in the expression of
distinct genes. The physiological role of apoptosis in
cancer prevention is to eliminate DNA-damaged cells
that would otherwise replicate and lead to mutations
and possibly cancer. Retinoic acid induces apoptosis
in many tumour cell types and therefore retinoids
have potential use for cancer chemotherapy and prevention.
According to Warrell et al. (1993), all-trans
retinoic acid appears to be the most effective single
agent for the treatment of any type of acute leukaemia.
The relationship between terminal differentiation
and apoptosis may conform to one of the following:
(1) retinoids fi rst induce differentiation, and then the
differentiated cells undergo apoptosis; (2) retinoids
induce differentiation and apoptosis concurrently;
(3) retinoids induce apoptosis in a process that is independent
of differentiation (Lotan, 1995).
7.10.2 Anti-cancer effects
A majority of primary human cancers arise in epithelial
tissues that depend upon retinoids for normal
cellular differentiation. The inhibition of carcinogenesis
by retinoids in various epithelial tissues is well
documented. Retinoids suppress malignant transformation
of cells in culture irrespective of whether the
transformation is induced by ionizing radiation or by
chemical carcinogens. Moreover, retinoids are potent
inhibitors of phorbol ester-induced tumour promotion.
Other studies have demonstrated a relationship
between retinoid defi ciency and cancer. Unfortunately,
chronic pharmacological administration of
retinoids is limited by their potential toxicity.
In humans, a signifi cant body of epidemiological
evidence correlates the intake of carotenoid-rich
fruits and vegetables with protection from some
forms of cancer. The potential use of carotenoids for
the chemoprevention of cancer in humans would
appear to be of great signifi cance as β-carotene is
essentially non-toxic. However, data obtained from
experimental studies using rats or mice (the welldefi
ned cancer models) cannot be extrapolated to
humans on account of the rodent’s limited ability
to absorb carotenoids as intact molecules and by the
rapid metabolism of any absorbed carotenoid.
Studies of carotenoids in cell culture have been hindered
in the past by diffi culties in supplying the highly
lipophilic carotenoids in a bioavailable form. Bertram
et al. (1991) overcame this problem by the use of
tetrahydrofuran as solvent, allowing the delivery of
diverse carotenoids to cultured cells at high concentration
in a bioavailable, micelle-like form. Using this
Vitamin A: retinoids and carotenoids 177
delivery system and the 10T½ line of transformable
mouse fi broblasts, Bertram’s group demonstrated
that many dietary carotenoids can inhibit neoplastic
transformation in the post-initiation phase of carcinogenesis.
7.10.3 Carotenoids as biological
antioxidants
The anticarcinogenic effect of carotenoids may be
due, at least partly, to their antioxidant activity. This
is attributable to two main properties of carotenoids:
(1) the ability to trap peroxyl free radicals implicated
in lipid peroxidation (Burton & Ingold, 1984) and
(2) the ability to deactivate singlet oxygen by physical
quenching (Foote & Denny, 1968). The extensive
conjugated double bond system of carotenoids is important
for both of these properties, but provitamin A
activity is irrelevant.
Trapping of peroxyl radicals
β-Carotene acts as a chain-breaking antioxidant in
lipid peroxidation by trapping the chain-propagating
lipid peroxyl radical, LOO•, in an addition reaction
(reaction 7.1). The chain-breaking activity of β-carotene
is weak compared with that of α-tocopherol, the
major lipid-soluble antioxidant.
LOO• + β-carotene → LOO–β-carotene• (7.1)
The carbon-centred β-carotenyl radical thus formed
is resonance-stabilized because of the delocalization
of the unpaired electron in the conjugated polyene
system. The β-carotenyl radical can react with another
peroxyl radical, leading to a termination reaction
(reaction 7.2).
LOO–β-carotene• + LOO• → LOO–β-carotene–OOR
(non-radical) (7.2)
Alternatively, the β-carotenyl radical can react with
molecular oxygen to form a β-carotene–peroxyl radical
(reaction 7.3).
LOO–β-carotene• + O2 LOO–β-carotene–OO•
(7.3)
Reaction 7.3 is reversible and dependent on the partial
pressure of oxygen (pO2) in the system. If pO2 is suffi
ciently low, the equilibrium shifts to the left, reducing
the amount of β-carotene–peroxyl radical. On the
other hand, if pO2 is high, the equilibrium shifts to
the right, increasing the amount of β-carotene–peroxyl
radical, which is capable of propagating the
lipid peroxidation chain reaction. The physiological
pO2 in most tissues is below 20 torr, but in lung it is
much higher – around 160 torr. Kennedy & Liebler
(1992) investigated the effect of pO2 on the ability
of β-carotene to inhibit peroxidation in phospholipid
liposomes (a model membrane bilayer). They
found β-carotene to be an effective antioxidant up
to 160 torr oxygen, with no difference in effectiveness
between 160 and 15 torr; at 760 torr inhibition of peroxidation
markedly declined. These data imply that
β-carotene could be as effective an antioxidant in lung
as in other tissues. Canthaxanthin and astaxanthin,
which possess oxo groups at the 4 and 4´-positions in
the β-ionone ring, are purported to be more effective
than β-carotene in trapping peroxyl radicals (Terao,
1989). It may be no coincidence that the length of the
β-carotene molecule approximates to the width of a
typical cell membrane. With an orientation perpendicular
to the plane of the membrane, the conjugated
double bond system of the carotene molecule could
trap radicals at any depth in the membrane.
Quenching of singlet oxygen
Carotenoids deactivate singlet oxygen (1O2) by
physical quenching, which involves the transfer of
excitation energy from 1O2 to the carotenoid, forming
ground state triplet oxygen (3O2) and triplet excited
carotenoid (reaction 7.4).
1O2 + carotenoid → 3O2 + 3carotenoid (7.4)
The energy is harmlessly dissipated through rotational
and vibrational interactions between the triplet
carotenoid and the surrounding medium to regenerate
ground state carotenoid (reaction 7.5).
3carotenoid → carotenoid + thermal energy (heat)
(7.5)
The rate of quenching is a function of the length of the
conjugated polyene chain and parallels the protective
action of carotenoids (Foote et al., 1970). Lycopene,
an open chain carotenoid, is approximately twice as
effective as β-carotene in quenching 1O2 (Di Mascio
et al., 1989).
Prooxidant actions of carotenoids
The antioxidant activity of carotenoids may shift into
pro-oxidant activity, depending on the redox poten-
178 Vitamins: their role in the human body
tial of the carotenoid molecules as well as on the biological
environment in which they act. If an inappropriate
prooxidant activity were to develop in normal
cells, the reactive oxygen metabolites generated would
lead to damage of cellular lipids, proteins and DNA,
and possibly induce neoplastic transformation. In
contrast, if carotenoids were to act as prooxidants in
already transformed cells, the production of cytotoxic
products would be benefi cial (Palozza, 1998).
7.10.4 Induction of gap junctional
communication
There is increasing evidence that one biochemical
mechanism underlying the chemopreventive action of
retinoids is enhanced gap junctional communication of
growth controlling signals. Hossain et al. (1989) showed
good correlation between retinoid inhibition of neoplastic
transformation and enhanced gap junctional communication
in carcinogen-initiated 10T½ cells. Retinoid-
enhanced junctional communication is achieved
in mouse 10T½ fi broblasts (Rogers et al., 1990) and in
human skin (Guo et al., 1992) through the synthesis of
the gap junction protein, connexin43, and its mRNA.
Zhang et al. (1991) demonstrated excellent correlation
between the inhibition of neoplastic transformation
of 10T½ cells by carotenoids and their
ability to enhance gap junctional communication.
The magnitude of induced junctional communication
was similar to that induced by retinoids, but
required higher concentrations of carotenoids (up
to 1000-fold) and longer treatment times (3–4 days
vs. 1 day for retinoids). The order of potency was β-
carotene/canthaxanthin (approximately equipotent),
lutein, lycopene and α-carotene. The antioxidant
activity of the carotenoids was found not to correlate
with their ability to inhibit neoplastic transformation
or their ability to enhance junctional communication.
Furthermore, the antioxidant α-tocopherol exhibited
only limited activity in the transformation assay and
methyl-bixin was inactive (Table 7.6). These data
suggest that the antioxidant properties of carotenoids
are not signifi cant in their action as inhibitors
of neoplastic transformation of 10T½ cells. Three
of the carotenoids tested (canthaxanthin, lutein and
lycopene) are reported to be without provitamin A
activity in mammals, thus their activity in the 10T½
cells is presumably not attributable to conversion to
retinoids. The carotenoids enhanced gap junctional
communication by up-regulating connexin43 gene
expression (Zhang et al., 1992). This effect appeared
unrelated to their provitamin A or antioxidant properties.
Whereas a synthetic retinoid up-regulated
expression of the RARβ gene at the message level (as
expected), canthaxanthin did not produce this effect.
These results imply that carotenoids and retinoids
function through separate but overlapping pathways.