0013-7227/03/$15.00/0
Printed in U.S.A.
The Journal of Clinical Endocrinology & Metabolism 88(1):225–229
Copyright © 2003 by The Endocrine Society
doi: 10.1210/jc.2002-020768
Contribution of Plasma Galactose and Glucose to Milk
Lactose Synthesis during Galactose Ingestion
AGNETA SUNEHAG, STELIOS TIGAS,
AND
MOREY W. HAYMOND
Department of Pediatrics, Children’s Nutrition Research Center, U.S. Department of Agriculture/Agricultural Research
Service, Baylor College of Medicine, Houston, Texas 77030
[1-13C]galactose. We observed that 69 ⴞ 6% of glucose and 54 ⴞ
4% of galactose in lactose were derived directly from plasma
glucose, whereas 7 ⴞ 2% and 12 ⴞ 2% of glucose and galactose
in lactose, respectively, were derived directly from plasma
galactose. De novo synthesis of glucose and galactose via hexoneogenesis accounted for 25 ⴞ 8% and 35 ⴞ 6%, respectively.
We conclude that during ingestion of galactose the contribution from plasma glucose to glucose and galactose in lactose
was similar to that of a short-term fasting, but part of the de
novo synthesis of glucose and galactose in the breast was replaced by direct uptake of galactose. (J Clin Endocrinol Metab
88: 225–229, 2003)
We have previously demonstrated that plasma glucose contributed 80% in the fed and 60% in the fasted state to lactose
synthesis in humans, and de novo synthesis in the breast contributing to both the glucose and galactose moieties accounted for the remaining 20% and 40%, respectively, of lactose. The present study was conducted to determine, in
lactating women, whether oral galactose is directly incorporated from plasma galactose into glucose and galactose in
milk lactose or via conversion of galactose to glucose in the
liver. Six healthy exclusively breast-feeding women (30 ⴞ 2 yr)
(mean ⴞ SE) ingested galactose at 22 ␮mol䡠kgⴚ1䡠minⴚ1 for 9 h
after an overnight fast during infusion of [6,6-2H2]glucose and
W
E HAVE RECENTLY DEMONSTRATED that plasma
glucose contributed significantly but was not the
exclusive substrate for milk lactose synthesis in humans (80%
in the fed condition and 60% during short-term fasting) (1).
In the course of these studies, we observed that the human
mammary gland is capable of de novo synthesis of both the
glucose and galactose moieties in lactose, a process we
termed hexoneogenesis (1). In those studies, plasma glycerol
contributed directly to the galactose moiety in lactose, but not
the glucose moiety, formed by this process.
Recently, Obermeier et al. (2) and Kunz et al. (3), using
[1-13C]galactose and nuclear magnetic resonance spectroscopy, reported that orally ingested 13C label was incorporated in milk, especially in lactose and neutral oligosaccharides. In some of the women, there was preferential labeling
of the galactose, compared with glucose, immediately after
galactose intake. However, the metabolic pathway(s)
through which oral galactose contributes to milk lactose production remains unknown. The present studies were carried
out to determine whether oral galactose is incorporated into
glucose and galactose in milk lactose as a result of conversion
of galactose to glucose in the liver and subsequent incorporation of this glucose into galactose and glucose in milk
lactose as we have previously demonstrated (1) or as the
result of the direct uptake of galactose into the breast and
milk glucose and galactose.
erichia coli, grade VIII, 740,000 U/g) from Sigma-Aldrich (St. Louis, MO).
BaOH, ZnSO4, d-(⫹)glucose (dextrose, corn sugar), d-(⫹)galactose, and
d-lactose (monohydrate) were purchased from Sigma-Aldrich. All organic solvents were HPLC grade. [1-13C]galactose and [6,6-2H2]glucose
were purchased from Cambridge Isotope Laboratory (Amherst, MA).
Glucose concentrations were determined using a biochemical analyzer model 2700 Select (Yellow Springs Instruments, Yellow Springs,
OH). Galactose concentrations were determined using a glucose and
L-lactate analyzer model 2300 Stat⫹, retrofitted to read galactose (Yellow Springs Instruments). BaOH solution (0.3N) and ZnSO4 solution
(0.3N) were purchased from Sigma-Aldrich.
Study design
Subjects. Six lactating women, between 24 and 35 yr of age (30 ⫾ 2 yr),
in good health, between 6 weeks and 3 months postpartum, took part
in these studies. Their weight was 68 ⫾ 3 kg, and body mass index was
24.7 ⫾ 1.4 kg/m2. The infants of these women were healthy and exclusively breast-fed at the time of the study. All of the volunteers underwent
a general physical examination and had normal hemoglobin and screening studies for liver and renal function and a negative pregnancy test
before they were accepted into the study.
Protocol design. Each woman and her infant were admitted to the Metabolic Research Unit (MRU) of the Children’s Nutrition Research Center
or the General Clinical Research Center on the evening before the study.
At 1800 h on the evening of admission, two iv catheters were introduced
in the mother’s antecubital fossa or forearm vein under Emla (AstraZeneca Pharmaceuticals LP, Wayne, PA) cream analgesia, one for isotope infusion and the other for blood sampling. Subjects were fed a
supper meal of 10 kcal/kg at 1800 h and were subsequently fasted,
except for water overnight. At 0600 h, baseline breast milk and blood
(5 ml) samples were obtained, and the subjects received primed
constant infusions of [6,6-2H2]glucose (20 ␮mol䡠kg⫺1䡠min⫺1, 0.33
␮mol䡠kg⫺1䡠min⫺1) to measure glucose rates of appearance and the fraction
of glucose and galactose in milk lactose derived from plasma glucose and
[1-13C]galactose (20 ␮mol䡠kg⫺1䡠min⫺1 and 0.33 ␮mol䡠kg⫺1䡠min⫺1) to
measure galactose rate of appearance and the fraction of galactose and
glucose in lactose derived from plasma galactose. In addition, unlabeled
oral galactose (22 ␮mol䡠kg⫺1䡠min⫺1) was administered every 15 min
from 0600 h until completion of the study to increase the plasma
galactose concentration; this was necessary to allow accurate measure-
Materials and Methods
Materials
Lactase (Aspergillus oryzae, 5500 ONPG, o-nitrophenyl ␤-d-galactopyranoside, units per gram) was obtained from ICN Biochemicals, Inc.
(Irvine, CA) and ␤-galactosidase (Aspergillus oryzae, 9,600 U/g and EschAbbreviations: GCMS, Gas chromatography-mass spectrometry;
PCI, positive chemical ionization; Ra, rates of appearance.
225
226
J Clin Endocrinol Metab, January 2003, 88(1):225–229
ment of galactose enrichments in plasma samples and clear separation of galactose from glucose during gas chromatography-mass
spectrometry (GCMS) analysis. The total amount of galactose ingested
over the 9-h study period would correspond to nine glasses of milk
(240 ml each).
Blood samples (6 –10 ml) were collected at 0600 h (just before start of
the tracer infusions) and then at 0900, 1200, 1245, 1300, 1315, 1330, 1345,
1400, 1415, 1430, and 1500 h. The infants were breast-fed at approximately 3-h intervals (0600, 0900, 1200, and 1500 h). Maternal blood
samples and milk samples were obtained immediately before each feeding. Indirect calorimetry was performed every 3 h. To determine the
volume of milk consumed, the difference in weight of the infant (before
and after feeding) was added to the volume of milk obtained for analyses
(generally ⬃5 cc each) (4).
Analyses
Plasma samples. Plasma glucose was measured using an enzyme-specific
method (glucose analyzer, YSI, Inc.). Plasma insulin and C-peptide were
measured using commercially available RIA kits (Linco Research, Inc.,
St. Charles, MO). Plasma free fatty acids and ␤-hydroxybutyrate were
determined by microfluorometric enzyme analyses as previously described using a Cobas Fara II analyzer (Roche Diagnostic Systems, Inc.,
Montclair, NY). The pentaacetate derivatives of glucose and galactose
were prepared as described previously (5). The isotopic enrichments of
[6,6-2H2]glucose, [1-13C]galactose, and 13C-glucose, derived from
[1-13C]galactose, were measured by GCMS using a quadrupole instrument (HP 6890/HP5973, Hewlett-Packard Co., Palo Alto, CA) and an
SPB-17 column (30 m ⫻ 0.25 mm internal diameter, 0.25 ␮m phase
thickness, Supelco, Inc., Bellefonte, PA). Positive chemical ionization
(PCI) mode was used with methane as the reagent gas and selected ion
monitoring of m/z 331–333 for both glucose and galactose. The conditions for the GCMS glucose and galactose analyses were: the temperature program used was an initial temperature, 70 C for 1 min, a temperature ramp at 30 C/min up to 280 C, followed by a 4-min hold. The
13
C isotopic enrichments in breath CO2 were measured by isotope ratio
mass spectrometry (MS instrument, Tracermass Robeprep G, Europa
Scientific, Crewe, UK). All measurements were made in the Stable Isotope Core Laboratory of the CNRC. To verify that the [1-13C]enrichment
from galactose remained in position 1 when galactose was converted to
glucose in the liver, GCMS was also employed in the electron impact
mode monitoring the m/z 242–244 fragments, in which carbon 1 is
cleaved off (6).
Milk samples. Milk samples were defatted, deproteinized, and hydrolyzed with lactase, using the method recently described (1). The hydrolyzed products were then derivatized (1) and analyzed by GCMS using
a mass selective detector (GC model 6890 series, MS model 5973,
Hewlett-Packard Co.) with an SPB-17 column (30 m ⫻ 0.25 mm internal
diameter, 0.25 ␮m phase thickness, Supelco). The conditions for the gas
chromatography were the same as for the analysis of the plasma samples. The PCI ions monitored were 331–333 for glucose and galactose.
Sunehag et al. • Lactose Synthesis from Plasma Galactose
共E关13C兴glucose/E关13C兴galactose兲 ⫻ glucose Ra 共␮mol 䡠 kg⫺1䡠min⫺1)
Potential sources for milk glucose and galactose in lactose are plasma
glucose, plasma galactose, plasma glucose derived from the hepatic
conversion of galactose to glucose, and finally the lactose could be
produced within the breast, a process we have denoted hexoneogenesis
(1). The M⫹2 glucose and M⫹2 galactose in lactose (derived from the
[6,6-2H2]glucose tracer) represent the direct contribution from plasma
glucose. Thus, the proportion of glucose (M⫹2 Glucmilk) and galactose
(M⫹2 Galmilk) in lactose derived from plasma glucose can be calculated
using the following equation:
Glucmilk4plasma M⫹glucose ⫽ M⫹2 Glucmilk/M⫹2 Glucplasma
where M⫹2 Glucplasma is the M⫹2 in plasma glucose derived from
[6,6-2H2]glucose and M⫹2 Glucmilk is the M⫹2 in glucose in lactose
derived from [6,6-2H2]glucose.
Galmilk4plasma M⫹2glucose ⫽ M⫹2 Galmilk/M⫹2 Glucplasma
where M⫹2 Gal milk is the M⫹2 in galactose in lactose derived from
[6,6-2H2]glucose.
The M⫹1 of plasma glucose (M⫹1 Gluplasma) and M⫹1 of glucose
(M⫹1 Glumilk) and M⫹1 of galactose in lactose (M⫹1 Galmilk) are derived from the [1-13C]galactose tracer. Thus, the M⫹1 of plasma glucose
represents glucose derived from galactose, i.e. galactose converted to
glucose in the liver. The M⫹1 of glucose and galactose in lactose could
be derived directly from plasma galactose or plasma glucose generated
by hepatic conversion of galactose to glucose. The proportion of the
glucose and galactose in lactose derived from plasma glucose should be
equal whether calculated from the [6,6-2H2]glucose or the [13C]glucose
data, i.e.
Glucmilk4plasma M⫹1 glucose
⫽ 共M⫹2 Glucmilk/M⫹2 Glucplasma兲*M⫹1 Glucplasma
Galmilk4plasma M⫹1 glucose ⫽ 共M⫹2 Galmilk/M⫹2 Glucplasma兲*M⫹1 Glucplasma
Glucose and galactose in lactose derived directly from plasma galactose were calculated using the following equation:
Glucmilk4plasma galactose ⫽ M⫹1 Glucmilk ⫺ Glucmilk4plasma M⫹1 glucose
where Glucmilk4plasma galactose is the glucose in lactose derived directly
from plasma galactose.
Galmilk4plasma galactose ⫽ M⫹1 Galmilk ⫺ M⫹1 Galmilk4plasma M⫹1glucose
where Galmilk4plasma galactose is the galactose in lactose derived directly
from plasma galactose.
The fraction of glucose and galactose in lactose derived from the
direct uptake of plasma galactose was then estimated on the basis of
standard product/precursor relationship using the following equations:
Fraction of glucose in lactose directly from plasma galactose
Calculations
⫽ Glucmilk4plasma galactose/M⫹1 Galplasma
The rates of appearance (Ra) of glucose and galactose into the systemic circulation were calculated under near steady-state condition using standard equations:
⫺1
Ra ⫽ 关共EI/Ep兲 ⫺ 1兴 䡠 I 共␮mol 䡠 kg
⫺1
䡠 min 兲
where Ra is the rate of appearance, Ei and Ep are the enrichments of the
isotope in the infusate and the plasma, respectively, and I is the rate of
infusion of the isotope of interest.
Absolute splanchnic extraction of galactose was calculated by subtracting galactose Ra from ingested galactose, and the fractional splanchnic extraction was calculated by dividing absolute splanchnic extraction
by ingested galactose.
The fraction of glucose Ra derived from galactose was calculated as
E[13C]glucose/E[13C]galactose, where E[13C]glucose is the plasma enrichment of
13
C glucose derived from [1-13C]galactose, and E[13C]galactose is the plasma
enrichment of [1-13C]galactose. The flux of galactose converted to glucose was calculated as follows:
Fraction of galactose in lactose directly from plasma galactose
⫽ Galmilk4plasma galactose/M⫹1 Galplasma
Galoxidation ⫽ 共13CO2 䡠 VCO2兲/共关1-13C兴galplasma 䡠 0.8兲
where Galoxidation is the rate of galactose oxidation, 13CO2 is the
13
C enrichment inexpired CO2, VCO2 is the rate of CO2 expired
(␮mol䡠kg⫺1䡠min⫺1), [1-13C]galplasma is the enrichment of [1-13C]galactose
in the systemic circulation, and 0.8 corrects for recovery of labeled CO2.
Statistical analysis
All data are expressed as mean ⫾ se. Substrate and hormone data
from the overnight 12-h fast (baseline, 0600 h) were compared with the
values measured from 1200 to 1500 h, i.e. after 6 –9 h of ingesting galactose using paired t test. The glucose and galactose kinetic data (glucose and galactose Ra, fraction of glucose Ra derived from galactose,
Sunehag et al. • Lactose Synthesis from Plasma Galactose
J Clin Endocrinol Metab, January 2003, 88(1):225–229 227
splanchnic extraction of dietary galactose, and oxidation rate of
[1-13C]galactose) represent the average of the values measured between
1200 and 1500 h, i.e. after 6 –9 h of galactose feeding. Isotope enrichments
in milk glucose and galactose used for calculation of product-precursor
relationships represent the mean of the values from the milk samples
obtained at 6 and 9 h into the study.
Results
Plasma substrate and hormone concentrations
Following 6 –9 h of continuous galactose ingestion, plasma
galactose concentration increased from 0.07 ⫾ 0.01 to 0.55 ⫾
0.15 mm, (P ⬍ 0.05). Plasma glucose concentrations did not
change [5.74 ⫾ 0.32 (baseline) vs. 5.96 ⫾ 0.27 mm, P ⫽ NS].
Plasma lactate increased from 0.81 ⫾ 0.12 (baseline) to 1.65 ⫾
0.15 mm, (P ⬍ 0.05). Plasma glucose, ␤-hydroxybutyrate, free
fatty acids, insulin, C-peptide, and glucagon remained unchanged throughout the study (Table 1).
Plasma glucose and galactose kinetics
The results on glucose and galactose kinetics are based on
the data provided in Table 2 using the calculations described
in Calculations. The concentration of plasma galactose was
too low to assess accurately the plasma galactose enrichment
in two subjects (low abundance). As a consequence, the analyses are based on an n ⫽ 6 for the maternal glucose kinetic
data and an n ⫽ 4 for plasma galactose kinetics and the milk
results. Plasma glucose and galactose Ra were 13.35 ⫾ 1.05
(n ⫽ 6) and 9.15 ⫾ 1.11 (n ⫽ 4) ␮mol䡠kg⫺1䡠min⫺1, respectively.
The fraction of plasma glucose Ra derived from galactose
was 41% ⫾ 3%, and the rate of galactose converted to glucose
was 5.51 ⫾ 0.45 ␮mol䡠kg⫺1䡠min⫺1 (n ⫽ 6). The nongalactosederived glucose production (i.e. after deducting the contribution of galactose to glucose Ra from the total glucose Ra)
was 8.02 ⫾ 1.28 ␮mol䡠kg⫺1䡠min⫺1. The rate of splanchnic
TABLE 1. Plasma substrate and hormone concentrations
following a 12-h overnight fast (baseline) and 6 –9 h of continuous
feedings with galactose
12-h fast (baseline)
Glucose (mM)
Galactose (mM)
Lactate (mM)
␤-OH-B (mM)
FFA (mM)
Insulin (␮U/ml)
C-peptide (ng/ml)
Glucagon (pg/ml)
5.74 ⫾ 0.32
0.07 ⫾ 0.01
0.81 ⫾ 0.13
0.21 ⫾ 0.09
0.47 ⫾ 0.12
5.08 ⫾ 1.74
1.46 ⫾ 0.40
63 ⫾ 4
Galactose feeding
5.96 ⫾ 0.27
0.55 ⫾ 0.16a
1.65 ⫾ 0.16a
0.05 ⫾ 0.01
0.45 ⫾ 0.06
4.40 ⫾ 0.79
1.07 ⫾ 0.10
57 ⫾ 3
extraction of galactose was 12.85 ⫾ 1.11 ␮mol䡠kg⫺1䡠min⫺1,
and the fractional extraction was 58% ⫾ 5%.
Milk glucose and galactose enrichments
Table 2 provides the enrichments of M⫹1 and M⫹2 in
plasma glucose, galactose, and glucose in lactose between 6
and 9 h of study (6 –9 h of isotope infusion). The M⫹2 enrichment of glucose in lactose was 69 ⫾ 6% of that of plasma
[6,6-2H2]glucose enrichment (Fig. 1), and the M⫹2 enrichment of galactose in lactose was 54 ⫾ 4% of that of the plasma
[6,6-2H2]glucose (i.e. 69% of the glucose and 54 of the galactose in lactose were derived directly from plasma glucose).
The portion of glucose and galactose in milk lactose derived directly from plasma galactose was 7 ⫾ 2% and 12 ⫾
2%, respectively (Fig. 1). Consequently, during ingestion of
galactose at 22 ␮mol䡠kg⫺1䡠min⫺1, “hexoneogenesis” contributed 25 ⫾ 8% of the glucose in lactose and 35 ⫾ 6% of the
galactose in lactose (Fig. 1). Thus, plasma glucose contributes
61 ⫾ 5%, plasma galactose 9 ⫾ 2% and “hexoneogenesis” 30
⫾ 7% of the substrate for lactose synthesis during ingestion
of galactose at 22 ␮mol䡠kg⫺1䡠min⫺1. Collectively, plasma glucose and galactose contributed 70% of the carbon for lactose
formation, a value in between the values found during feeding and fasting in our previous study (feeding 80% and
fasting 60%) (1).
Indirect calorimetry data
The rate of galactose oxidation (regardless of the pathway of metabolism) was estimated to be 3.69 ⫾ 0.64
␮mol䡠kg⫺1䡠min⫺1, or 40 ⫾ 3% of the galactose Ra and 17 ⫾
1% of the ingested amount of galactose.
Milk volume and lactose production
During the 9 h of galactose ingestion, milk production was
374 ⫾ 58 ml. Assuming a lactose concentration of 7% or 220 mm
in milk and that 61% [(69 ⫹ 54)/2] of lactose is derived from the
plasma glucose in the fed state, the rate of glucose converted to
lactose was approximately 1.40 ␮mol䡠kg⫺1䡠min⫺1 and the rate
of galactose conversion to lactose 0.21 ␮mol䡠kg⫺1䡠min⫺1. This
represents approximately 11% and 2% of the systemic rates
of appearance of glucose and galactose, respectively. The rate
of plasma clearance of glucose and galactose into lactose
synthesis can be estimated from the rates of conversion of
plasma glucose and galactose into lactose and the plasma
concentrations. The clearance of plasma glucose and galactose into lactose synthesis was 0.24 and 0.38 ml/kg⫺1䡠min⫺1,
TABLE 2. Percentage enrichments M⫹1 and M⫹2 in plasma glucose, milk galactose, and milk glucose between 6 and 9 h of galactose
feeding (n ⫽ 4)
Plasma
Glucose
Galactose
Milk
Glucose
Galactose
Milk/plasma ratios
Gal/Glu
Glu/Glu
M⫹1
M⫹2
Fasting (previous)
[U-13C]glucose (n ⫽ 6)
Fed (previous)
[U-13C]glucose (n ⫽ 6)
1.58 ⫾ 0.12
3.55 ⫾ 0.36
2.49 ⫾ 0.21
1.33 ⫾ 0.09
0.57 ⫾ 0.05
1.31 ⫾ 0.16
1.28 ⫾ 0.18
M ⫹ 1/M ⫹ 1
0.80 ⫾ 0.06
0.82 ⫾ 0.04
1.75 ⫾ 0.30
1.36 ⫾ 0.20
M ⫹ 2/M ⫹ 2
0.54 ⫾ 0.04
0.69 ⫾ 0.06
0.98 ⫾ 0.10
0.68 ⫾ 0.07
M ⫹ 6/M ⫹ 6
0.51 ⫾ 0.03
0.72 ⫾ 0.04
0.51 ⫾ 0.04
0.36 ⫾ 0.04
M ⫹ 6/M ⫹ 6
0.68 ⫾ 0.07
0.98 ⫾ 0.03
228
J Clin Endocrinol Metab, January 2003, 88(1):225–229
Sunehag et al. • Lactose Synthesis from Plasma Galactose
FIG. 1. Fraction of glucose and galactose derived from
hydrolyzed human lactose derived from plasma glucose
(white bar), galactose (gray bar), and de novo synthesis
(hatched bar) during ingestion of galactose at 22
␮mol䡠kg⫺1䡠min⫺1. SE bars reflect the variance for each
individual part of the contribution to lactose synthesis.
respectively (i.e. the clearances are similar for glucose and
galactose despite a 10-fold difference in their plasma
concentrations).
Discussion
Using an iv infusion of [1-13C]galactose and oral administration of unlabeled galactose, the present study demonstrated that ingestion of galactose results in direct uptake of
galactose by lactating mammary tissue with conversion to
both glucose and galactose in lactose and uptake of plasma
glucose (as a result of presumed hepatic conversion of galactose to glucose) and incorporation into glucose and galactose in milk lactose. In principle, these data are in agreement with those recently reported by Obermeier et al. (2) and
Kunz et al. (3) in which they observed that following ingestion of [1-13C]galactose, the label was recovered in glucose
and galactose in lactose. However, their study design precluded them from determining whether the labeled galactose
was converted to labeled glucose in the liver and subsequently the labeled plasma glucose converted to labeled glucose and galactose in milk lactose or the direct uptake of
labeled galactose by the breast.
We have previously demonstrated (1) that plasma glucose
contributed approximately 90% of the glucose and approximately 60% of the galactose in lactose in the fed state, regardless of whether the feeding is glucose alone or a mixed
liquid caloric meal. During fasting these values decreased to
approximately 60% and approximately 50%, respectively (1).
In the present study, we observed values of 69 ⫾ 6% and 54
⫾ 4% for the contribution of plasma glucose to milk glucose
and galactose, respectively. These values are slightly higher
than our previous fasted values but lower than the fed data,
which is not surprising because the women in the present
study were really not fasting but receiving oral galactose at
22 ␮mol䡠kg⫺1䡠min⫺1. Despite the galactose ingested, it is
interesting to note that the plasma glucose rates of appearance in this and our previous fasting studies (7) were essentially the same (i.e. the oral galactose did not increase maternal plasma glucose appearance rates).
Using the single- and double-labeled plasma glucose en-
richments, we were able to partition the contributions of
plasma glucose to milk glucose and galactose. In addition, we
were able to determine the fractional contribution of plasma
galactose to both carbohydrate moieties in milk lactose.
Plasma galactose made a greater contribution to milk galactose than milk glucose, suggesting a ready pathway of transport and phosphorylation of the galactose in the mammary
tissue. In our previous study (1), plasma glucose contributed
approximately 60% and hexoneogenesis approximately 40%
of milk lactose in the fasted state. During ingestion of galactose, the contribution from plasma glucose to lactose was
virtually identical to that of the fasting state but the contribution from hexoneogenesis was approximately 10% lower
(⬃30%). This difference is accounted for by the direct conversion of plasma galactose to lactose.
In studies by Kunz et al. (3), an oral bolus dose of labeled
galactose was administered (a total of 27 g). Assuming that
the women weighed approximately 70 kg, this would correspond to 386 mg/kg or 2140 ␮mol/kg of which 7% was
13
C-labeled galactose. We have previously demonstrated (8)
that the uptake of galactose by the liver is saturable at about
approximately 15 ␮mol䡠kg⫺1䡠min⫺1. Thus, it is impossible to
determine how much of this bolus was extracted by the liver.
In the present study, we administered the oral galactose over
a period of 9 h. As a result, direct comparison of the results
from these studies is of no value. The authors’ observation
of the 13C appearing in the C-1 position in galactose and
glucose in lactose (on the basis of their nuclear magnetic
resonance data) does not prove that plasma [1-13C]galactose
was the primary or only precursor for lactose synthesis.
Using GCMS employing both electron impact (which cleaves
C-1) (6) and PCI (which preserves the entire carbon skeleton)
(6), we demonstrated that the plasma glucose (derived from
galactose) was labeled exclusively in the C-1 position. This is
consistent with direct hepatic conversion of [1-13C]galactose
to [1-13C]glucose. Because Obermeier et al. (3) most likely
achieved higher plasma galactose concentrations following
their bolus ingestion than we did with our 9-h constant
ingestion, they may have had somewhat higher direct conversion of plasma galactose to milk lactose than we observed.
Sunehag et al. • Lactose Synthesis from Plasma Galactose
However, we demonstrated that under our study conditions,
approximately 40% of the plasma glucose appearance rate
was derived from galactose. With higher administration
rates of galactose in the Kunz study (3), the vast majority of
glucose would be derived from galactose (8).
In summary, during ingestion of galactose at 22
␮mol䡠kg⫺1䡠min⫺1, 9.2 ␮mol䡠kg⫺1䡠min⫺1 appeared in the systemic circulation, 5.5 ␮mol䡠kg⫺1䡠min⫺1 appeared as plasma
glucose, and 4 ␮mol䡠kg⫺1䡠min⫺1 was oxidized (i.e. we accounted for 85% of the oral galactose). Because plasma lactate
concentrations doubled during ingestion of galactose, a portion of oral galactose was most likely converted to plasma
lactate. Plasma galactose directly contributed approximately
10% of milk lactose with a larger proportion directed toward
galactose in lactose, compared with glucose in lactose (12%
vs. 7%). The direct contribution from plasma glucose to lactose was similar to that observed in the fasted state in our
previous study (1). Thus, the carbon provided by plasma
galactose replaced part of the carbons provided by hexoneogenesis during fasting. This study demonstrates that although plasma glucose is the primary source of milk lactose,
plasma galactose is a potential substrate for milk lactose
during ingestion of galactose. Milk is the primary dietary
source of galactose. However, in milk the galactose is contained in the lactose molecule (i.e. galactose ⫹ glucose). We
have previously demonstrated (8) that coingestion of galactose with glucose results in increased splanchnic uptake of
galactose and thus very low plasma concentrations of
galactose. Therefore, to achieve plasma concentrations of
galactose that would enable us to determine whether galactose could contribute to milk lactose, in the present study,
galactose was ingested alone at a rate that would correspond
to nine glasses of milk (240 ml/each) over the 9-h study
period.
J Clin Endocrinol Metab, January 2003, 88(1):225–229 229
Acknowledgments
We acknowledge and thank the technicians in our laboratory (Patricia
Breitenfelder, Shaji Chacko, Cindy Clarke, Dan Donaldson, and Kathryn
Louie), our nurse coordinator (Andrea Jones, R.N.), and the staff in the
Metabolic Research Unit and kitchen and the General Clinical Research
Center who greatly facilitated the execution of these studies.
Received May 17, 2002. Accepted September 12, 2002.
Address all correspondence and requests for reprints to: M. W. Haymond, M.D., Children’s Nutrition Research Center, 1100 Bates Street,
Houston, Texas 77030-2600. E-mail: [email protected].
This work was supported by grants from U.S. Department of Agriculture Cooperative Agreement #58-6250-6-001, NIH RO1DK 55478, and
NIH HD37857. This work is a publication of the USDA/ARS Children’s
Nutrition Research Center, Department of Pediatrics, Baylor College of
Medicine (Houston, TX). The contents of this publication do not necessarily reflect the views of policies of the U.S. Department of Agriculture, nor does mention of trade names, commercial products, or organizations imply endorsement from the U.S. Government.
References
1. Sunehag A, Louie K, Tigas S, Bier J, Haymond MW 2002 Hexoneogenesis in
the human breast during lactation. J Clin Endocrinol Metab 87:297–301
2. Obermeier S, Rudloff S, Pohlentz G, Lentze MJ, Kunz C 1999 Secretion of
13C-labelled oligosaccharides into human milk and infant’s urine after an oral
[13C]galactose load. Isotopes Environ Health Stud 35:119 –125
3. Kunz C, Rudloff S, Baier W, Klein N, Strobel S 2000 Oligosaccharides in
human milk: structural, functional, and metabolic aspects. Annu Rev Nutr
20:699 –722
4. Motil KJ, Kertz B, Thotathuchery M 1997 Lactational performance of adolescent mothers shows preliminary differences from that of adult women. J
Adolesc Health 20:442– 449
5. Argoud GM, Schade DS, Eaton RP 1987 Underestimation of hepatic glucose
production by radioactive and stable tracers. Am J Physiol 252:E606 –E615
6. Biemann K, DeJongh DC, Schnoes HK 1963 Application of mass spectrometry to structure problems. XIII. Acetates of pentoses and hexoses. J Am Chem
Soc 85:1763–1771
7. Tigas S, Sunehag A, Haymond MW 2002 Metabolic adaptation to feeding and
fasting during lactation in humans. J Clin Endocrinol Metab 87:302–307
8. Sunehag A, Haymond MW 2002 Splanchnic galactose extraction is regulated
by co-ingestion of glucose in humans. Metabolism 51:827– 832