Content uploaded by Jacqueline C Kent
Author content
All content in this area was uploaded by Jacqueline C Kent on Mar 20, 2014
Content may be subject to copyright.
• 5 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
• 5 •
Research
ABSTRACT
Adequate calcium intake is vital for infant health, and some cases of rickets have been
associated with a low concentration of calcium in breastmilk. The concentration of calcium
in breastmilk has been shown to vary widely both between mothers, and over the course of
lactation. To address potential concerns about the adequacy of calcium intake for infants
who are exclusively breastfed, we discuss the factors likely to be affecting the concentration
of calcium in breastmilk. We review and provide new evidence for a physicochemical model
of the interactions of calcium with other components of breastmilk, particularly phosphate,
citrate and casein. A proposed mechanism for the control of the concentration of calcium in
milk is described that highlights the influence of the concentrations of citrate and casein.
Understanding these interactions clarifies why the concentration of calcium in breastmilk is
not affected by manipulations of maternal dietary calcium and vitamin D.
Keywords: calcium, human, milk, vitamin D
Breastfeeding Review
2009; 17 (2): 5–11
Why calcium in breastmilk is independent of maternal
dietary calcium and vitamin D
Jacqueline C Kent BSc, DipEd, PhD
Peter G Arthur BSc, PhD
Leon R Mitoulas BSc, PhD
Peter E Hartmann BRurSci, PhD
INTRODUCTION
Rickets is a disease of infants and children characterised
by fragile bones and skeletal deformities. It is caused by a
deficiency of vitamin D, which is required for intestinal calcium
absorption, or a dietary deficiency of calcium or phosphorus.
The sole source of dietary calcium and phosphorus for the
exclusively breastfed infant is breastmilk. Therefore, the
concentration of calcium in breastmilk is fundamental to an
adequate calcium supply to the infant.
The median concentration of calcium in breastmilk is
252 mg/L (6.0 mM), with a very large range of 84–462 mg/
L (2.1–11.5 mM) (Dorea 1999). While the extremes are of
concern, very few samples collected between 1 and 6 months
of lactation had measured concentrations of <100 mg/L (2.5
mM) or more than 300 mg/L (7.5 mM). Compared to control
populations, teenage mothers and mothers in Egypt, the
Gambia and Zaire had low concentrations of calcium in their
milk. Despite this, the infants of these mothers showed no
adverse effects. However, cases of rickets have been reported
recently in exclusively breastfed infants <6 months old born to
vitamin D deficient mothers (Nickkho-Amiry et al 2008) and
a few breastfed Nigerian children (Thacher et al 2006). These
cases led to speculation that rickets in such infants might arise
in part because of low breastmilk concentrations of calcium
and phosphorus (Nickkho-Amiry et al 2008).
The effects of alterations of maternal dietary calcium or
vitamin D on the concentration of calcium in breastmilk
have been investigated (Basile et al 2006; Nickkho-Amiry et al
2008; Prentice 2000; Prentice et al 1995; Prentice et al 1997),
but the interventions were found to have no effect. These
studies highlight a lack of understanding of the control of the
concentration of calcium in breastmilk.
The interaction of calcium with other milk components
has been modelled and studied in milk of animals and women
(Holt, Dalgleish & Jenness 1981; Holt & Muir 1979; Kent,
Arthur & Hartmann 1998; Neville 2005; Neville et al 1994)
and a mechanism for the accumulation of calcium in milk has
been proposed (Holt 1981). In order to elucidate the control
of the concentration of calcium in breastmilk, this paper
reviews the normal changes in concentration throughout
lactation and provides new data on calcium interactions in
breastmilk. Understanding these interactions clarifies why
the concentration of calcium in breastmilk is not affected by
manipulations of maternal dietary calcium and vitamin D.
Secretion of calcium into milk
Milk is synthesised and secreted by the lactocytes of the
mammary gland. The calcium in breastmilk is derived from
the blood, as are all components of breastmilk. The total
concentration of calcium in the blood is about 2.5 mM of
which about half is bound to proteins. Most of the remainder
is free or ionised calcium and its concentration is tightly
controlled at about 1.25 mM. The ionised calcium can be
transported from the blood plasma across the basal membrane
ReviewResearch
• 6 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
• 7 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
• 6 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
• 7 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
of the lactocytes into the cytosol where the concentration is
approximately 1 µM (Neville, Allen & Watters 1983). From
the cytosol, the ionised calcium is pumped into the Golgi
apparatus via a calcium ATPase (Watters 1984). In the Golgi
apparatus, the ionised calcium interacts with phosphate, citrate
and casein (Figure 1). The contents of vesicles derived from
the Golgi apparatus are secreted into the milk by exocytosis
(Neville, Allen & Watters 1983).
the milk will be related to the concentration of casein in the
milk. Experimental data from the milk of different species are
consistent with this theoretical relationship (Holt & Jenness
1984). The concentrations of calcium and casein in bovine
milk are 29 mM and 26 g/L, respectively (White & Davies
1958). In contrast, human milk has both the lowest calcium
and the lowest casein concentrations of all species (6 mM and
4 g/L, respectively) (Dorea 1999; Hartmann 1991).
The calcium concentration of human milk is even lower
than predicted by the general relationship between calcium
and casein. This low concentration can be attributed to
the different types of casein in human milk, which result
in human milk having only 14 moles of calcium bound per
mole of casein in the micelles (Neville et al 1994), compared
with bovine milk in which 20 moles of calcium are bound
per mole of casein (Jenness 1979). These differences in
both the concentration and type of casein in human milk
contribute to its low concentration of calcium as compared
with milk from other species. However, this normally poses
no problems for the breastfed infant. Infants take over three
months to double their birth weight and this slow growth rate,
and therefore low requirement for calcium for mineralisation
of bones, is compatible with the low concentrations of
calcium and casein in human milk. On the other hand, the
high concentrations of calcium and casein in bovine milk are
appropriate for calves that take only one month to double
their birth weight.
Calcium partitioning and modeling of interactions
The model of micelle formation explains the general
relationship between species with regards to calcium and
casein concentration (Holt & Jenness 1984). However, when
the concentration of casein is low, such as in sows’ milk
during secretory activation or in breastmilk, other components
in addition to casein make a significant contribution to the
determination of the concentration of calcium (Kent, Arthur
& Hartmann 1998). The study of the relationships between
calcium and other milk components has been facilitated by the
partitioning of milk as shown in Figure 2. Alternatively, the
diffusible fraction containing ionised calcium, free phosphate,
free citrate, calcium phosphate and calcium citrate can be
prepared by ultrafiltration of whole milk.
In the diffusible fraction of milk, the concentrations of
calcium occurring as ionised calcium, calcium phosphate and
calcium citrate are dictated by physicochemical principles. The
interactions of ionised calcium, free phosphate and free citrate
are in equilibrium and there are defined equilibrium constants.
In turn, the interactions of free phosphate and free citrate with
other components in the diffusible fraction of milk are also in
equilibrium. For example, magnesium binds both free phosphate
and free citrate, so the concentration of magnesium will affect
the concentration of calcium phosphate and calcium citrate. All
of these interactions can be quantified using physicochemical
modelling. A model has been developed and tested in milk from
cows, sows and women (Holt, Dalgleish & Jenness 1981; Kent,
Ca2+
Cit3-
CaCit-
CaHPO4HPO42-
Ca3(PO4)2
Casein
micelle
Calcium
ATP-ase
Figure 1: Schematic diagram of calcium interactions in the
Golgi apparatus
Ca2+, ionised calcium; Cit3-, free citrate; CaCit-, calcium citrate; HPO4
2-, free
phosphate; CaHPO4, soluble calcium phosphate; ●, Calcium phosphate nucleus
Ca3(PO4)2
The phosphate, citrate and casein in the Golgi apparatus are
derived from three different pathways. Phosphate is produced
in the Golgi apparatus when glucose reacts with UDP-
galactose in the presence of α-lactalbumin during the synthesis
of lactose (Neville, Allen & Watters 1983). Calcium phosphate
is sparingly soluble, and in the concentrations found in
breastmilk would precipitate without some means of keeping
it solubilised. Citrate is synthesised within the mitochondria
and then translocated to the cytosol. It is probably then
transported into the Golgi apparatus (Zulak & Keenan 1983).
In contrast to calcium phosphate, calcium citrate is highly
soluble. Caseins are milk proteins that are synthesized by the
rough endoplasmic reticulum of the lactocytes from amino
acids derived from the blood. They are then transferred to the
Golgi apparatus. Human milk contains αs-casein; β-casein and
κ-casein. In the presence of ionised calcium, the caseins will
form a micelle with colloidal calcium phosphate (Davies, Holt
& Christie 1983; Dickson & Perkins 1971). Calcium phosphate
is important for the integrity of casein micelles that, in turn,
prevent the precipitation of calcium phosphate.
The model of micelle formation explains how large
amounts of calcium and phosphate are transported in the
milk in a stable form to the infant (Holt 1994). It also predicts
that the concentration of calcium in the colloidal fraction of
• 6 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
• 7 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
• 6 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
• 7 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
Whole milk
Centrifuge 4 000 - 6 000g 30 min
Fat Skim milk
Centrifuge 4°C
100 000g 2h
or 150 000g 1h
Casein Supernatant
(whey proteins and
water soluble components)
Ultrafilter
10 000 to 30 000 Da cutoff
Whey proteins Diffusible fraction
(ionized calcium
free phosphate
free citrate
calcium phosphate
calcium citrate)
Figure 2: Fractionation of milk samples
Centrifugation for 30 min at 4 000 to 6 000g brings to the surface the fat layer
over the skim milk fraction. Centrifugation of the skim milk at 150 000g for 1 h
(Fransson & Lönnerdal 1982, 1983) or 100 000g for 2 h at 4°C (Neville et al 1994)
has been used to precipitate the casein fraction (Kunz & Lönnerdal 1992) while the
aqueous supernatant contains the whey proteins and water soluble components.
Ultrafiltration of the aqueous supernatant through a membrane with a 10 000 to
30 000 Da cutoff separates the whey proteins from the fraction containing low
molecular weight components, and the latter is termed the diffusible fraction.
Arthur & Hartmann 1998; Neville et al 1994). Using this model,
the effect of changing the concentration of any component of
milk on the concentrations of the different forms of calcium in
milk can be calculated.
We will describe the changes in concentration of calcium in
breastmilk throughout lactation and consider the interactions
that are associated with these changes.
Initiation of lactation — first five days after birth
During the initiation of lactation, termed secretory activation
(Pang & Hartmann 2007), in women the total concentration
of calcium in breastmilk increases rapidly from less than 6
mM on the first day after birth to more than 8 mM on day
five (Kent et al 1992; Neville et al 1991). In order to study
the causes of the increase in calcium we have measured
pH, total calcium, diffusible calcium, ionised calcium and
diffusible citrate during secretory activation in daily samples
collected from 12 term mothers (Kent et al 1992). The data
for days one and five are presented in Figure 3. The increase
in the total concentration of calcium was not explained by an
increase in ionised calcium because ionised calcium remains
relatively constant between 2.3 and 2.9 mM (Kent et al 1992;
Neville et al 1991).
In the diffusible fraction after partitioning milk, the
concentration of inorganic phosphate on days 1 and 5 was 0.3
mM and 1.5 mM, respectively (Kent et al 1992). It was thus
postulated that the formation of calcium phosphate contributes
to the increase in total calcium. However, using physicochemical
modeling, we calculated that during the first five days of lactation
in women, the concentration of calcium phosphate remains less
than 0.2 mM (Kent, Arthur & Hartmann 1998). This finding
confirms that phosphate is not a significant contributor to the
total concentration of calcium (Neville et al 1994). Indeed, an
increase in the concentration of calcium phosphate would be
undesirable because of the low solubility of calcium phosphate
and the need to prevent precipitation of calcium phosphate in the
mammary gland. There was a relationship between calcium and
citrate as predicted by the physicochemical model (Holt, Dalgleish
& Jenness 1981), consistent with the increase in citrate being at
least partly responsible for the increase in calcium. Calculations
using this model suggest that calcium citrate is responsible for
more than 50% of the increase in total calcium.
During this neonatal period the concentration of casein
increases from being undetectable immediately after birth to
reaching a peak of 5.9 g/L at 6–10 days after birth (Kunz &
Lönnerdal 1992). Neville and co-workers (1994) studied 13 term
mothers and showed an increase in the proportion of calcium
bound to protein during this period. Moreover, our data show an
increasing differential between the concentrations of diffusible
and total calcium (Kent et al 1992). This finding is consistent with
the measured increase in the concentration of casein (Kunz &
Lönnerdal 1992) and increased micelle formation. Therefore, it
can be concluded that an increase in casein was responsible for
the remaining (almost 50%) increase in calcium.
In summary, the increase in the concentration of total calcium
during the first five days after birth in breastmilk is caused by
increases in the concentrations of both casein and citrate.
Throughout lactation — 1 to 15 months after birth
During the period of established lactation, the changes in the
concentration of calcium are less dramatic than during the
Figure 3: Concentrations of calcium in fractions of milk,
citrate measured in ultrafiltrate, and pH measured in fresh
whole milk from 12 mothers during secretory activation
(day 1
**
** **
!
*
and day 5
**
** **
!
*
) and from another 12 mothers in late
lactation (>15 months
**
** **
!
*
)
**
** **
!
*
One-way analysis of variance: * different from >15 months, P < 0.001; ** different
from day 1 and >15 months, P < 0.001; † different from day 1, P < 0.05.
• 8 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
• 9 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
initiation of lactation (Figure 4). Neville and co-workers (1986,
1991) continued to collect milk samples from the 13 mothers
studied during the neonatal period described above, and found
that from 6 to 15 months of lactation the total concentration
of calcium declined. Using a physicochemical model of the
interactions of the ions in the diffusible fraction of milk, it was
calculated that during this time the proportion of calcium bound
to citrate decreased as the concentration of citrate decreased
(Neville et al 1994). Neville and co-workers (1994) also calculated
that in whole milk the protein-bound fraction of calcium
decreased. The latter finding is intriguing, because as lactation
proceeds beyond six months casein increases as a proportion of
total protein, reaching 50 % by 12 months (Kunz & Lönnerdal
1992). Since casein accounts for almost all of the protein-bound
calcium it is difficult to explain a decrease in protein-bound
calcium in late lactation.
± 0.1 mM and 2.43 ± 0.04 mM, respectively), and then declined
at 15 months to 5.7 ± 0.1 mM and 0.87 ± 0.08 mM, respectively
(Kent 2000, Figure 4), similar to previous results (Neville et al
1986, 1991). There was a significant relationship between the
concentrations of total calcium and total citrate for all samples
collected (r 2 = 0.61, P = 0.0001) independent of time after birth.
This data confirms that the decline in the concentration of citrate
can explain at least part of the decline in the total concentration
of calcium.
The contribution of casein to the total concentration
of calcium will be dependent to a large degree on the total
concentration of protein in the milk. We measured a decrease in
the concentration of protein in the milk between 4 and 6 months
of lactation from 9.9 ± 0.4 g/L to 8.0 ± 0.4 g/L (P = 0.001), with
no significant change at 12 months (Mitoulas et al 2002), after
which we measured an increase (P = 0.006) to 10.0 ± 0.7 g/L at
15 months (Kent 2000). This increase is consistent with previous
results (Neville et al 1991) and should have resulted in an increase
in the proportion of calcium bound to protein at this stage of
lactation. This contradicts the findings of Neville et al (1994) that
the protein-bound fraction of calcium decreased at this time.
The discrepancy suggests that the interactions of calcium in
late lactation required further study. We collected, partitioned
and analysed one sample of milk from each of 12 mothers
who had been lactating for at least 15 months (Kent 2000).
The pH, and measured concentrations of calcium and citrate
in samples collected in late lactation are shown in Figures 3
and 4. Physicochemical modeling was used to calculate the
concentration of calcium citrate in the diffusible fraction,
and the contributions of ionised calcium, calcium citrate and
casein-bound (colloidal) calcium to the total concentration are
shown in Figure 5.
The concentration of calcium citrate was calculated to be
0.91 ± 0.11 mM, similar to that calculated previously (Neville
et al 1994). By subtraction of diffusible calcium from total
calcium, the protein-bound calcium was calculated to be 1.54
± 0.15 mM. This amount is significantly higher than that
previously found for samples collected at 15 months (Neville
et al 1994), confirming a higher contribution of casein to the
total concentration of calcium. Despite this higher proportion
of casein-bound calcium, the decrease in the concentration of
citrate has a predominant effect and is the cause of the decline
in the total concentration of calcium.
The decrease in the concentration of calcium in breastmilk
after six months, combined with the decrease in the volume
of breastmilk consumed by infants as complementary foods
are introduced, results in a decreased intake of calcium from
breastmilk and highlights the importance of considering the
calcium content of complementary foods.
Mechanism of control of calcium in milk throughout
lactation
The measurements we have made on milk samples collected
throughout lactation confirm a sharp increase in the concentration
of calcium during the first few days of lactation (Kent et al 1992)
Figure 4: The concentrations of total calcium (■), citrate
(●) and ionised calcium (X) in breastmilk samples collected
throughout lactation
0
2
4
6
8
10
12
0 3 6 9 12 15 >15
Time after birth (m)
Concentration (mM)
a
b
b
c
d
a,e a,e
cc
d
a,e
a
b
c
eff
cc
d
g
a
h
h
Symbols with the same letter are not significantly different. Different letters
denote difference from previous time point (one-way analysis of variance): b, c, d,
e P < 0.001, f, g P < 0.01, h P < 0.05. Data from 12 mothers day 1 to 5 (Kent et al
1992), 5 mothers 1–15 months and 12 mothers >15 months (Kent 2000).
In order to clarify this situation, we have previously studied
five mothers who exclusively breastfed their infants for at least
six months (Kent et al 1999). All the infants were introduced
to solid foods between 6 and 12 months of age, and partial
breastfeeding continued for at least 15 months. The mothers
measured their 24-hour milk production and collected milk
samples at 1, 2, 4, and 6 months of lactation, and then at 3-
monthly intervals until 15 months.
The mean 24-hour milk production during the first six
months was 425 ± 21 g per breast, similar to previous results
(Neville et al 1986, 1991), and then declined to 160 ± 52 g per
breast at 15 months.
We analysed four samples of whole milk from each mother
at each time point. The concentrations of calcium and citrate
that we measured were constant between 1 and 4 months (7.5
• 8 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
• 9 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
• 9 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
followed by a gradual decline for as long as lactation continues
(Kent 2000) (Figure 4). These changes in calcium can be explained
by using the physicochemical model (Kent 2000; Kent, Arthur &
Hartmann 1998) to understand the relationship between calcium
and other milk components.
The concentration of ionised calcium remains relatively
constant throughout lactation (Figure 3 and Neville 2005).
Therefore, changes in the concentration of calcium in the milk
cannot be attributed to changes in ionised calcium.
In women, immediately after birth the concentration of
casein in milk is very low and contributes to the very low
concentration of total calcium. The concentration of casein
increases to reach a peak at 6–10 days after birth (Kunz &
Lönnerdal 1992) consistent with the increased amount of casein
binding an increased amount of calcium (Neville et al 1994).
Between 6 days and 6 months of lactation, the concentration of
casein in human milk decreases from 0.26 to 0.15 mM (Kunz
& Lönnerdal 1992). This result is reflected in the decrease in
the concentration of colloidal calcium from 3.2 to 1.2 mM
(Neville et al 1994) which in turn contributes to the decrease
in total calcium from 10 to 6.5 mM (Figure 3). At 15 months
the concentration of casein increases to 0.20 mM (Kunz &
Lönnerdal 1992) consistent with the increase in colloidal calcium
to 1.5 mM (Figure 5).
However, these changes in the concentration of casein are
not sufficient to account for all the changes in the total calcium
concentration and the contribution of the concentration
of citrate must be considered. The theoretical relationship
between calcium and citrate in the diffusible fraction of milk
Figure 5: Measured concentration of ionised calcium (Ca2+),
calculated concentration of calcium citrate (CaCit-), and
colloidal calcium calculated as the difference between the
total and diffusible calcium
0
1
2
3
Ca2+ CaCit-Cacoll
Concentration (mM)
The milk was collected from 12 mothers who had been lactating for at least 15
months. The samples were collected with minimal exposure to air and fractionated
by ultracentrifugation through a membrane with a MW cut off of 30 000 Da
(Kent et al 1992).
has been confirmed when the concentrations are changing
rapidly during initiation of lactation in sow milk as well as in
human milk (Kent, Arthur & Hartmann 1988; Kent et al 1992;
Neville et al 1994). The increase in the concentration of citrate
between day 1 and day 5 (Kent et al 1992) (Figures 3 and 4)
results in an increase in the concentration of calcium citrate and
contributes to the increase in the total calcium concentration.
The decrease in the total concentration of citrate in human
milk between 5 days and 6 months contributes to the decline
in the total concentration of calcium. The concentrations of
both calcium and citrate decreased between 6 months and
12 months while the concentration of protein was stable,
indicating that the decreased concentration of citrate was
causing the decrease in the concentration of calcium. After
12 months, the concentration of citrate continued to decrease
while the concentration of casein increased, thus resulting in a
stable concentration of total calcium.
The relatively constant concentration of ionised calcium
in milk of women from initiation until late lactation suggests
that there is a homeostasis of ionised calcium in milk, as there
is in blood. The minimum concentration of ionised calcium
must be 1 mM for casein micelle stability (Holt, Davies & Law
1986) and 2–4 mM for lactose synthesis by lactose synthase
(Powell & Brew 1976). The upper limit is determined by
the free phosphate concentration, since precipitation of
calcium phosphate in the mammary gland must be avoided.
The ionised calcium concentration in the Golgi apparatus is
maintained within these limits by calcium-ATPase against a
concentration gradient from the cytosol. The regulation of the
activity of the calcium-ATPase is thus an important factor in
understanding the control of the secretion of calcium in milk
that warrants further exploration. While vitamin D promotes
intestinal calcium absorption it has no known effect on calcium
transport across the membrane of the Golgi apparatus.
Calcium in human milk is regulated indirectly by
regulating the concentration of citrate and casein in
the milk (Neville et al 1994). We propose the following
mechanism, illustrated in Figure 6. During the initiation of
lactation, as increasing amounts of casein are synthesised
and accumulate within the Golgi apparatus, ionised calcium
is bound to casein and phosphate in the formation of
casein micelles. Simultaneously, as increasing amounts
of citrate are transported across the Golgi membrane,
it also binds ionised calcium to form calcium citrate. As
the ionised calcium is bound, calcium-ATPase transports
ionised calcium from the cytosol into the Golgi apparatus
and maintains the concentration of ionised calcium at 2.6
mM. During extended lactation in women, as less citrate is
transported into the Golgi apparatus, less calcium citrate is
formed and the transport of ionised calcium decreases.
Holt (1981) suggested that citrate accumulated in the Golgi
apparatus as a consequence of the active transport of calcium
ions. Since the transport of citrate into the Golgi apparatus was
unaffected by the addition of 5 mM calcium to the medium
(Zulak & Keenan 1983), we suggest the alternative hypothesis,
• 10 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
• 11 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
• 10 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
• 11 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
namely that it is the transport of citrate into the Golgi apparatus
that determines the concentration of diffusible calcium in
milk. Thus, the regulation of the transport of citrate into the
Golgi apparatus is important in the secretion of calcium but is
not yet understood.
CONCLUSION
It is important to understand the determinants of the
concentration of calcium in breastmilk if there is a possibility
that a low concentration of calcium is a contributing factor in
the development of rickets. To this end, we have reviewed the
normal increase in the concentration of calcium in breastmilk
during the first week of lactation and the subsequent gradual
decline until late lactation. The application of a physicochemical
model has demonstrated that the total concentration of calcium
is dependent on the concentrations of both casein and citrate,
and change in the concentration of citrate is the predominant
factor in breastmilk. We have provided evidence for the
hypothesis that the transport of citrate into the Golgi apparatus
determines the concentration of calcium citrate in milk, which
comprises a significant proportion of the total concentration of
calcium in breastmilk.
Future research may discover a means of controlling the rate
of transport of citrate into the Golgi apparatus. However, even
if there were known mechanisms that would make this possible,
then excessive increases in concentrations of citrate or casein
may not be desirable.
Alterations in maternal dietary calcium or vitamin D will have
no effect on the physicochemical interactions between calcium
and casein, citrate and phosphate, and therefore cannot influence
the total concentration of calcium in breastmilk.
ACKNOWLEDGMENTS
Ethics approval for the studies performed by the authors and
reported in this review was granted by the Human Research
Ethics Committee at The University of Western Australia. The
work was funded by NH&MRC and Medela AG.
REFERENCES
Basile LA, Taylor SN, Wagner CL, Horst RL, Hollis BW 2006,
The effect of high-dose vitamin D supplementation on serum
vitamin D levels and milk calcium concentration in lactating
women and their infants. Breastfeed Med 1(1): 27–35.
Davies DT, Holt C, Christie WW 1983, The composition of
milk. In Mepham TB (ed) Biochemistry of Lactation. Elsevier,
Amsterdam.
Dickson IR, Perkins DJ 1971, Studies on the interactions between
purified bovine caseins and alkaline-earth-metal ions. Biochem
J 124: 235–240.
Dorea JG 1999, Calcium and phosphorus in human milk. Nutr
Res 19(5): 709–739.
Fransson G-B, Lönnerdal B 1982, Zinc, copper, calcium, and
magnesium in human milk. J Pediatr 101: 504–508.
Fransson G-B, Lönnerdal B 1983, Distribution of trace elements
and minerals in human and cow’s milk. Pediatr Res 17(11):
912–915.
Hartmann PE 1991, The breast and breast feeding. In Philipp E,
Setchell M, Ginsburg JG (eds) Scientific Foundations of Obstetrics
and Gynaecology. 4th edn. Butterworth-Heinemann, London.
Holt C 1981, Some principles determining salt composition and
partitioning of ions in milk. J Dairy Sci 64: 1958–1964.
Holt C 1994, The Biological Function of Casein? Yearbook Hannah
Research Institute, Ayr.
Holt C, Dalgleish DG, Jenness R 1981, Calculation of the ion
equilibria in milk diffusate and comparison with experiment.
Anal Biochem 113: 154–163.
Holt C, Davies DT, Law AJR 1986, The effects of colloidal
calcium phosphate content and milk serum free calcium ion
concentration on the dissociation of bovine casein micelles. J
Dairy Res 53: 557–572.
Holt C, Jenness R 1984, Interrelationships of constituents and
partition of salts in milk samples from eight species. Comp
Biochem Physiol 77A: 275–282.
Holt C, Muir DD 1979, Inorganic constituents of milk I
Correlation of soluble calcium with citrate in bovine milk. J
Dairy Sci 46: 433–439.
Jenness R 1979, Comparative aspects of milk proteins. J Dairy Res
46: 197–210.
Kent JC 2000, The roles of citrate and casein in determining the
calcium concentration of the milk of women and sows. PhD
thesis, The University of Western Australia, Perth.
Kent JC, Arthur PG, Hartmann PE 1998, Citrate, calcium,
phosphate and magnesium in sows’ milk at initiation of
lactation. J Dairy Res 65: 55–68.
Figure 6: Proposed schematic representation of the
determination of calcium concentration of breastmilk by
casein and citrate
Casein micelle
Ca2+
(2.6 mM)
Ca2+
(1 !M)
CitrateCitrate
CaseinCasein
Golgi
Calcium
ATPase
Cytosol
Milk
Ca2+
(2.6 mM)
Calcium citrate
Lactocyte
Calcium
citrate
Casein
micelle
The concentration of ionised calcium (Ca2+) in the cytosol of the lactocyte is 1
µM. As casein is synthesized and accumulates within the Golgi apparatus, Ca2+ is
bound to casein (and phosphate) in the formation of casein micelles. As citrate is
transported across the Golgi membrane, it also binds Ca2+ to form calcium citrate. As
the Ca2+ is bound, calcium-ATPase transports Ca2+ from the cytosol and maintains
the concentration of Ca2+ at 2.6 mM. The concentrations of casein micelles, calcium
citrate and Ca2+ in the milk are the same as in the Golgi apparatus.
• 10 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
• 11 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
• 10 •
BREASTFEEDING REVIEW VOL 17 NO 2 2009
• 11 •
VOL 17 NO 2 2009 BREASTFEEDING REVIEW
Kent JC, Arthur PG, Retallack RW, Hartmann PE 1992, Calcium,
phosphate and citrate in human milk at initiation of lactation.
J Dairy Res 59: 161–167.
Kent JC, Mitoulas LR, Cox DB, Owens RA, Hartmann PE 1999,
Breast volume and milk production during extended lactation
in women. Exp Physiol 84: 435–447.
Kunz C, Lönnerdal B 1992, Re-evaluation of the whey protein/
casein ratio of human milk. Acta Paediatr 81: 107–112.
Mitoulas L, Kent JC, Cox DB, Owens RA, Sherriff JL, Hartmann
PE 2002, Variation in fat, lactose and protein in human milk
over 24h and throughout the first year of lactation. Br J Nutr
88: 29–37.
Neville MC 2005, Calcium secretion into milk. J Mammary Gland
Biol Neoplasia 10(2): 119–128.
Neville MC, Allen JC, Archer PC, Casey CE, Seacat J, Keller
RP, Lutes V, Rasbach J, Neifert M 1991, Studies in human
lactation: milk volume and nutrient composition during
weaning and lactogenesis. Am J Clin Nutr 54: 81–92.
Neville MC, Allen JC, Watters C 1983, The mechanisms of
milk secretion. In Neville MC, Neifert MR (eds) Lactation:
Physiology, Nutrition, and Breast-feeding Plenum Press, New York
and London.
Neville MC, Casey CE, Keller RP, Archer P 1986, Changes in
milk composition after six months of lactation: the effects
of duration of lactation and gradual weaning. In Hamosh
M, Goldman AS (eds) Human Lactation 2: Maternal and
Environmental Factors. Plenum Press, New York.
Neville MC, Keller RP, Casey C, Allen JC 1994, Calcium
partitioning in human and bovine milk. J Dairy Sci 77(7):
1964–1975.
Nickkho-Amiry M, Prentice A, Ledi F, Laskey MA, Das G, Berry
JL, Mughal MZ 2008, Maternal vitamin D status and breast
milk concentrations of calcium and phosphorus. Arch Dis
Child 93(2): 179.
Pang WW, Hartmann PE 2007, Initiation of human lactation:
secretory differentiation and secretory activation. J Mammary
Gland Biol Neoplasia 12(4): 211–221.
Powell JT, Brew K 1976, Metal ion activation of
galactosyltransferase. J Biol Chem 251: 3645–3652.
Prentice A 2000, Calcium in pregnancy and lactation. Ann Rev
Nutr 20: 249–272.
Prentice A, Jarjou LM, Cole TJ, Stirling DM, Dibba B, Fairweather-
Tait S 1995, Calcium requirements of lactating Gambian
mothers: effects of a calcium supplement on breast-milk
calcium concentration, maternal bone mineral content, and
urinary calcium excretion. Am J Clin Nutr 62(1): 58–67.
Prentice A, Yan L, Jarjou LM, Dibba B, Laskey MA, Stirling DM,
Fairweather-Tait S 1997, Vitamin D status does not influence
the breast-milk calcium concentration of lactating mothers
accustomed to a low calcium intake. Acta Paediatr 86(9): 1006–
1008.
Thacher TD, Pettifor JM, Fischer PR, Okolo SN, Prentice A
2006, Case-control study of breast milk calcium in mothers
of children with and without nutritional rickets. Acta Paediatr
95(7): 826–832.
Watters CD 1984, A Ca2+-stimulated adenosine triphosphatase
in Golgi-enriched membranes of lactating murine mammary
tissue. Biochem J 224: 39–45.
White JCD, Davies DT 1958, The relation between the chemical
composition of milk and the stability of the caseinate
complex. J Dairy Res 25: 236–255.
Zulak IM, Keenan TW 1983, Citrate accumulation by a Golgi
apparatus-rich fraction from lactating bovine mammary gland.
Int J Biochem 15(3): 747–750.
ABOUT THE AUTHORS:
This work was conducted in the School of Biomedical,
Biomolecular and Chemical Sciences at The University of
Western Australia.
Jacqueline Kent has been undertaking research on the
biochemistry and physiology of lactation and breastfeeding in
babies for over 20 years under the supervision of Peter Hartmann.
The aim is to provide scientific evidence that can be used by
health professionals to support breastfeeding mothers.
Peter Arthur completed his PhD on novel methods for
investigating milk synthesis in lactating mothers under the
supervision of Peter Hartmann. He has significant expertise in
physicochemical modelling, and his current interests include the
effect of reactive oxygen species on cell function.
Leon Mitoulas completed his PhD at The University of
Western Australia on short- and long-term variation in the
production, content and composition of human milk fat under
the supervision of Peter Hartmann. He is now the Head of
Breastfeeding Research at Medela, Switzerland.
Peter Hartmann has been at the forefront of breastfeeding
research for over 30 years. His awards include the Elizabeth Mills
Award from the Nursing Mothers’ Association of Australia,
Macy-György Award from the International Society for Research
in Human Milk and Lactation, the La Leche League International
Award for Excellence and he was recently made a Fellow of the
Australian Nutrition Society (Inc.).
Correspondence to
Jacqueline Kent
School of Biomedical, Biomolecular and Chemical Sciences
Faculty of Life and Physical Sciences
The University of Western Australia
M310, 35 Stirling Highway
Crawley WA 6009
Australia
© Australian Breastfeeding Association 2009
