Kirchgessner, 1991). Thus, it appears that
Fe is unique among the essential trace
elements in that endogenous excretion by
either faeces or urine does not play a
significant role in homeostasis.
Selenium
The metabolism of Se varies with its
chemical form (Sunde, 1997). Se in animal
tissue is predominantly in the form of
protein-bound selenocysteine, whereas
plant Se is predominantly seleno-
methionine. Selenium in inorganic supple-
ments is normally selenite or selenate.
Circulating selenite is taken up by erythro-
cytes and rapidly reduced to selenide,
which is a common intermediate in the
metabolism of both inorganic and amino
acid forms of Se. Selenocysteine and
selenomethionine may be incorporated
directly into proteins or also converted to
selenide prior to incorporation of Se as
selenocysteine into Se-specific seleno-
proteins such as glutathione peroxidase.
Since animals are able to synthesize
selenocysteine but not selenomethionine
and since dietary selenomethionine may be
incorporated into protein (mainly muscle
protein), there are two kinetically distinct
components of Se metabolism in animals:
selenite-exchangeable Se and seleno-
methionine Se (Sunde, 1997). Since the
selenomethionine pool cannot be labelled
with selenite, the kinetics of labelled
selenomethionine differ from those of
labelled selenite. A selenite-exchangeable
metabolic pool has been found to be highly
correlated with whole-body Se in rats and
to respond to dietary Se restriction by
decreasing in size in humans. Most whole-
body Se is found in muscle. Analysis of 12
major tissues of lactating dairy cows
showed that 61% of Se in tissues analysed
was in skeletal muscle, 7% in the hide, 6%
in the liver, 5% in the udder, 4% in bone,
4% in smooth muscle of the reticulorumen
and 3% or less in each remaining tissue.
Stable isotope tracer studies with
selenite and selenomethionine in humans
have helped to elucidate certain aspects of
Se metabolism (Patterson et al., 1989;
Swanson et al., 1991). According to kinetic
models developed from the tracer experi-
ments, 84% of orally administered selenite
Se was absorbed and 90% of absorbed
selenite Se was taken up by the liver. More
than half, 58%, of the Se taken up by the
liver was secreted into the gastrointestinal
tract via the bile, from where most of it was
reabsorbed. Selenium tracer released from
the liver into plasma was taken up by a
peripheral tissue compartment which
eventually released the tracer for excretion
into the urine without further recycling to
liver or other tissues. Within 12 days, 18%
of ingested dose was recovered in the faeces,
including unabsorbed tracer, and 17% was
recovered in the urine. Metabolism of
selenomethionine Se differed from selenite
Se in the kinetic model. Ninety-eight per
cent of orally administered seleno-
methionine was absorbed and it was
removed more rapidly from the blood by
the liver than selenite Se. Slightly less
tracer was returned to the gastrointestinal
tract via the bile. Forty-three per cent of the
selenomethionine Se released from the
liver to plasma was taken up by a slowly
turning over peripheral tissue compart-
ment not accessed by the selenite tracer.
Virtually all of the selenomethionine Se
released from tissue compartments was
recycled back to the liver. Over 12 days,
4% of ingested selenomethionine tracer
was recovered in faeces and 11% was
recovered in urine. In both the selenite and
selenomethionine models, endogenous Se
was excreted predominantly in the urine,
which agrees with other studies. The
combined effect of the differences in
metabolism of the two Se forms was that
the whole-body half-life of seleno-
methionine Se was 252 days while that of
selenite selenium was 102 days. The
percentage absorption and differences in
endogenous faecal excretion of selenite and
selenomethionine in rats are similar to the
results obtained with humans.
Absorption of Se can be high, but it
varies considerably among species. Se
absorption by rats is >90% (Kirchgessner et
al., 1997), whereas measurements of178 W.T. Buckley