Analysis of Lactulose Preparations by Spectrophotometric and

High Performance Liquid Chromatographic Methods

F. W. P A R R I S H , 1 K. HICKS, 2 and L. D O N E R
Eastern Regional Research Center 3
Philadelphia, PA 19118

ABSTRACT galactopyranosyl-/3-D-fructofuranose] (12, 15,

Spectrophotometric methods of analy- 21). The greater sweetness (19) and solubility
(17) of lactulose compared to lactose indicate
sis were applied to sugar mixtures pro-
that lactulose has potential as a partial replace-
duced during isomerization of lactose to
ment for sucrose in baking applications where
lactulose. A combination of four estab-
lished spectrophotometric methods was lactose has been unsatisfactory (1).
This paper describes application of established
used to quantitate galactose, tagatose,
spectrophotometric methods and presents a
lactose, and lactulose, the four major
more convenient high performance liquid
sugars present. In addition, a method
chromatographic technique for analysis of
originally developed for determining
mixtures containing lactulose and other sugars.
lactose with methylamine at 65°C and pH
12.7 was suitable for combined measure-
MATERIALS AND METHODS
ment of lactose and lactulose. After a
specific determination of lactose with
Sugar Standards
t3-galactosidase and glucose oxidase, lactu-
lose was measured by difference. A high Reagent grade anhydrous glucose, fructose,
pressure liquid chromatographic separa- tagatose, galactose, &-lactose monohydrate, and
tion of galactose, tagatose, lactose, and lactulose were dried under vacuum at 65°C for
lactulose was achieved on a commercial 16 h and then used to prepare 100 mg/ml stock
carbohydrate analysis column with a solutions.
mixture of water and acetonitrile used as Lactulose syrups were either purchased
eluent. Analysis of lactulose syrups by (Cephulac, Merrell-National Laboratories a) or
the new high pressure liquid chromato- prepared by isomerization of lactose with
graphic method yielded results that calcium hydroxide (15), triethylamine (18), or
agreed with those by the spectrophoto- sodium aluminate (11). The syrups, stored as
metric methods and required much less 70 to 80% solutions, were diluted to the
time. appropriate concentrations prior to analysis.

INTRODUCTION Spectrophotometric Analyses

As part of a program aimed at increased Total reducing sugars were measured by the
utilization of lactose and its derivatives in food 3,5-dinitrosalicylic acid method (14) with
applications, we have examined methods a-lactose monohydrate solution (0 to 1 mg/ml)
for isomerizing lactose to lactulose [4-O-fl-D- as standard; absorbance was measured at 540
nm. Ketose sugars were determined by the
thiobarbituric acid method (20) with fructose
solution (0 to 40 /~g/ml) as standard. Aldose
Received February 4, 1980. sugars were measured by a hypoiodite oxidation
1Southern Regional Research Center, Agricultural
procedure (13) with &-lactose monohydrate
Research, Science and Education Administration, US
Department of Agriculture, New Orleans, LA 70179. solution (0 to 2 mg/ml) as standard. Mono-
2Postdoctoral Fellow supported by Dairy Research saccharides were determined in the presence of
Foundation. disaccharides by the method of Tauber and
a Agricultural Research, Science and Education Kleiner (23) with fructose (0 to .4 mg/ml) as
Administration, US Department of Agriculture.
4Reference to brand or firm name does not consti- standard. Galactose (0 tO 1.25 mg/ml) was
tute endorsement by the US Department of Agri- assayed with galactose oxidase (9), and lactose
culture over others of a similar nature not mentioned. was measured, after hydrolysis with t3-galactosi-

1980 J Dairy Sci 63:1809--1814 1809

1810 PARRISH ET AL.

dase (22), with a glucose oxidase assay proce- for determining total sugar (7), total reducing
dure (2). The methylamine procedure for sugars (10, 14), aldoses (13), and ketoses (20)
lactose (16) was applied also to lactulose (.5 to in sugar mixtures. In addition, enzymatic
2 mg/ml). All standard curves gave a linear methods enable specific sugars to be determined,
relationship (r~>.997) between absorbance and e.g., glucose and galactose with oxidative
sugar concentration. Absorbance was measured enzymes (9). Application of two of these
in 1-cm cuvettes with a Zeiss Model PMQII methods (9, 10) enabled quantitative determina-
spectrophotometer. Analyses of samples solu- tion of sucrose and its component monosaccha-
tions were in triplicate. rides, glucose and fructose, in mixtures of all
three sugars (6). The complexity of spectro-
High Pressure Liquid Chromatographic Analysis photometric analyses of mono and oligosaccha-
A standard mixture containing galactose, rides in mixtures, however, increases with
tagatose, lactose, and lactulose was prepared in number of sugars and number of methods that
a mixture of water and acetonitrile (1 + 1, by have to be combined to achieve the desired
volume) as solvent. All sugars were at a known result. Accuracy also is affected by differential
and equal concentration that was less than 20 responses of individual sugars to a reagent
mg/ml each. In all cases, the sugars were dis- as compared to that of the sugar used for a
solved in water before the acetonitrile was calibration standard. In this study, the lactulose
added. Sugar mixtures from isomerization of syrups contained galactose and tagatose (4, 5)
lactose were deionized with Amberlite IR-120 as well as lactose and lactulose. The presence of
(H) and Duolite A-561 (free base), and diluted these four different sugars added to the com-
so as to contain not more than 20 mg/ml of any plexity of the analysis.
component. A sample (20/al) of sugar mixture Our studies involving several isomerization
was applied by loop injection to an analytical reagents previously described (11, 15, 18) with
(3.9 mm x 30 cm) Carbohydrate Analysis varying conditions such as time, temperature,
Column (Waters, Milford, MA). Separation of and pH, have shown 3,5-dinitrosalicylic acid
all four sugars was achieved by elution with a reagent (14) is accurate for measuring total
mixture of water and acetonitrile (77/23, reducing sugar. Equal absorbance was given by
wt/wt) at 2 ml/min with a modular chromato- .875 mg galactose, .912 mg glucose, .985 mg
graphic system including an Instrumentation tagatose, .993 mg lactulose, and 1.000 mg
Specialties Co. (ISCO) metering pump model lactose. Lactose was used as the calibration
314, series 1240-003, a pressure monitor (ISCO standard because conditions for optimal yield
model 1590), and a Waters Associates model of lactulose resulted in less than 10% mono-
R401 refractive-index detector. Sugars were saccharide.
quantified by comparison of peak heights with To ascertain amounts of each of the four
those of corresponding sugars in standard sugars in lactulose syrups, the approach in
solutions. Table 1 was used. Separate analyses were
to determine aldoses (13), ketoses (20), mono-
RESULTS AND DISCUSSION saccharides (23), and lactose (2, 22). Subtraction
Spectrophotometric methods are available of monosaccharide from the sum of aldose and

TABLE 1. Spectrophotometric analysis scheme for lactulose in mixtures from isomerization of lactose.

Type of Sugars
sugar Method determined Reference

Aldoses Hypoiodite Lactose, galactose 13

Ketoses Thiobarbituric acid Lactulose, tagatose 20
Monosaccharides Acid Cu++, molybdate Galactose, tagatose 23
Lactose 3-Galactosidase + Lactose 2, 22
glucose oxidase

Journal of Dairy Science Vol. 63, No. 11, 1980

 ANALYSIS OF LACTULOSE 1811

TABLE 2. Spectrophotometric analysis of standard sugar mixtures as in the scheme of Table 1.

Sugar (mg/lO0 ml)

Lactulose Lactose Galactose Fructose a
Mixture Added Foundb Added Found Added Found Added Found

1 40 43(-+2) 140 136(-+4) 0 1(+1) 20 24(-+2)

2 80 85(-+4) 80 77(-+2) 20 22(-+1) 20 24(-+2)
3 160 171(-+7) 20 23(-+2) 20 22(-+1) 0 2(+1)

aThe more readily available ketose, fructose, was substituted for tagatose because each gave identical re-
sponses in spectrophotometric tests.
bAverage of three measurements with observed deviation.

ketose analyses gave a sum of lactose and lactulose and 1.000 mg lactose was identical.
lactulose concentrations. Lactose was measured The methylamine procedure together with the
by enzymatic hydrolysis (22) to galactose and enzymatic procedure for lactose measurement
glucose followed by estimation of glucose with permitted determination of lactulose by differ-
glucose oxidase (2). The amount of lactulose ence with more convenience and greater preci-
was obtained by subtraction of lactose from the sion than analysis involving the combination of
sum of lactose and lactulose concentrations. four assays (Table 3).
Application of these analytical procedures Because spectrophotometric analysis of
enables the concentrations of lactulose, lactose, multicomponent mixtures can be complex and
galactose, and tagatose to be calculated (Table time consuming, we examined the potential of
2). Results are in Table 3 for mixtures contain- high pressure liquid chromatography (HPLC)
ing optimal yields of lactulose by the three for this purpose. The use of HPLC in sugar
isomerization methods (11, 15, 18). analysis has been reviewed (3), and official
An alternative spectrophotometric procedure methods for the quantitation of specific sugars
for measure of lactulose (Table 3) was devel- by this chromatographic method exist (8).
oped for assays in which monosaccharide Verhaar and coworkers (24) developed a liquid
quantitation was not of interest. This was based chromatographic method for separation and
on a methylamine procedure for lactose (16). quantitation of sugars in lactulose syrups. While
Galactose produced negligible color under the their method produces baseline separation of
same conditions. The absorbance for .867 mg the various sugars, it requires a complex detec-

TABLE 3. Spectrophotometric analysis of sugars in mixtures from isomerization of lactose.

Sugars % (by wt) a

Methylamine
Isomerization Procedures of Table 1 procedure
reagent Monosaccharides Lactulose Lactose Lactulose Lactose

Calcium hydroxide 4.5(-+.5) 17(-+2) 80(-+3) 16(-+1) 76(-+3)

Triethylamine 2.5(-+.5) 38.5(-+3) 6I(-+2) 36(-+2) 62(_+2)
Sodium aiuminate 4.5(-+.5) 77(-+4) 23(+1) 73(-+3) 22(-+1)

aAverage of three measurements with observed deviation.

Journal of Dairy Science Vol. 63, No. 11, 1980

1812 PARR1SH ET AL.

100 TABLE 4. Liquid-chromatographic evaluation of

standard sugars.
0
Retention Capacity
time factor Relative
Sugar (tr, rain) (~') response

Tagatose 4.50 1.409 3.887

Galactose 6.72 2.540 1.387
80 Lactulose 13.40 6.074 1.539
Lactose 15.50 7.308 1.0130

ILl
D

Z
0
0.. tion system, accurately prepared elution
buffers, and extended times for each assay.
rr We now have developed a more rapid,
tv-
60 simpler chromatographic system based on
Lad commercially available HPLC equipment and
rv- simple binary solvent mixtures. Separation of a
O standard mixture of the four major sugars in
t.~ lactulose syrups is in Figure 1. Retention times,
iv-
b capacity factors, and relative responses to the
refractive index detector for each sugar are in
Table 4. Baseline separation is approached for
40 all sugars in less than 16 rain. Quantitation of
each sugar in an unknown mixture is accom-
plished easily by peak height comparison to the
appropriate sugar in the standard mixture. Peak
height was proportional to sugar concentration
over the range that in all cases was less than 20
mg/ml of each sugar. When the lactulose syrups
analyzed by the spectrophotometric tech-
niques in Table 3 were analyzed by the HPLC
20-- ~
I procedure (Table 5), the measures agreed with
0 5 I0 15 20 those by the former methods.
TIME, MINUTES The sugar composition of a commercial
Figure 1. High performance liquid chromatogram pharmaceutical grade lactulose syrup, Cephulac
of a standard mixture of tagatose (a), galactose (b), (Merrell-National Laboratories), was examined
lactulose (c), and lactose (d), all at 11.4 mg/ml.
Column was a Waters Associates Carbohydrate Analy- by the HPLC procedure. The syrup was deion-
sis model eluted at 2 ml/min with 77/23 acetonitrile/ ized and chromatographed as described in
H 20 (wt/wt). Materials and Methods. The chromatogram in

TABLE 5. High pressure liquid chromatographic analysis of sugars in mixtures from isomerization of lactose.

Isomerization Sugars % (by wt)

reagent Tagatose Galactose Lactulose Lactose

Calcium hydroxide 2.3 2.5 15.1 80.1

Triethylamine 3.7 2.2 32.8 61.3
Sodium aluminate 5.4 4.3 72.7 17.6

Journal of Dairy Science VoL 63, No. 11, 1980

 ANALYSIS OF LACTULOSE 1813

TABLE 6. Analysis of major sugars in Cephulac. a

Trial Statistic
Sugar 1 2 3 4 5 x SD c v (%)

Tagatose 1.12 .98 1.08 1.10 .94 1.04 .O8 7.60

Galactose 12.58 12.32 12.35 12.63 12.31 12.44 .15 1.24
Lactulose 79.45 80.03 79.95 79.10 80.03 79.71 .42 .52
Lactose 6.86 6.66 6.61 7.17 6.72 6.80 .22 3.31

aln relative weight percent.

Figure 2 indicates several minor unknown graphic techniques for quantitative determina-
compounds. The identified sugars, tagatose, tion of sugars in lactulose preparations.
galactose, lactulose, and lactose, were quanti-
REFERENCES
tated and are reported in relative weight percent
in Table 6. The precision of the analysis also is 1 Ash, D. J. 1976. Research on lactose indicates uses,
limitations as a substitute for sucrose in bakery
given for standard deviations and coefficients of
goods. Food. Prod. Dev. 10(6):85.
variation for five successive analyses. 2 Bergmeyer, H. U., and E. Bernt. 1963. D-Glucose.
The HPLC method described here is a rapid, Determination with glucose oxidase and peroxidase.
convenient, precise, and specific alternative to Page 123 in M e t h o d s of e n z y m a t i c analysis. H. U.
existing spectrophotometric and chromato- Bergmeyer, ed. Academic Press, New York, NY.
3 Conrad, E. C., and J. K. Palmer. 1976. Rapid
analysis of carbohydrates by high-pressure liquid
chromatography. F o o d Technol. 30:84.
4 Corbett, W. M., and J. Kenner. 1953. The degrada-
60- c tion o f carbohydrates by alkali. Part 2. Lactose. J.
Chem. Soc. 2245.
5 Corbett, W. M., and J. Kenner. 1954. The degrada-
tion of carbohydrates by alkali. Part 5. Lactulose,
maltose, and maltulose. J. Chem. Soc. 1789.
6 DellaMonica, E. S., M. J. Calhoun, and P. E.
50- McDowell. 1974. T h e quantitative determination
LIA
03 of glucose, fructose, and sucrose in fruits and
z potatoes. J. F o o d Sci. 39:1062.
o
7 Dubois, M., K. A. Gilles, J. K. Hamilton, P. A.
03
Rebers, and F. Smith. 1956. Colorimetric m e t h o d
r~ for determination of sugars and related substances.
ow 4 0 - Anal. Chem. 28:350.
LIA 8 Engle, C. E., and P. M.Olinger. 1979. High pressure
e-x
liquid chromatographic determination of saccha-
O rides in corn sirups: Collaborative study, J. Assoc.
Off. Anal. C h e m . 62:527.
9 Finch, P. R., R. Yuen, H. Schachter, and M. A.
~" 3 0 - Moscarello. 1969. Enzymic m e t h o d s for the micro
assay of D-mannose, D-glucose, D-galactose, and
L-fucose from acid hydrolyzates o f glycoproteins.
Anal. Biochem. 31 : 296.
10 F u r u h o l m e n , A. M., J. D. Winefordner, F. W.
20-...j Knapp, and R. A. Dennison. 1964. The quantitative
analysis o f glucose and fructose in potatoes. J.
Agric. Food Chem. 12:109.
I I I 11 Guth, J. H., and L. T u m e r m a n . 1970. Method of
making lactulose. US Pat. 3,546,206.
12 Isbell, H. S., and W. W. Pigman. 1938. Pyranose-
TIME, MINUTES furanose interconversions with reference to the
Figure 2. High performance liquid c h r o m a t o g r a m m u t a r o t a t i o n s of galactose, levulose, lactulose, and
o f Cephulac: tagatose (a), galactose (b), lactulose (c), turanose. J. Res. Nat. Bur. Standards 20:773.
lactose (d). 13 Miller, G. L., and A. L. Burton. 1959. Spectro-

Journal of Dairy Science Vol. 63, No. 11, 1980

1814 PARRISH ET AL.

p h o t o m e t r i c determination of aldoses b y an relative to sucrose. J. Food Sci. 44:813.

iodometric procedure. Anal. Chem. 31 : 1790. 20 Percheron, F. 1962. Colorimetric determination of
14 Miller, G. L., R. Slater, R. Birzgalis, and R. Blum. fructose and fructofuranosides b y thiobarbituric
1961. Application of different colorimetric tests to reaction. Compt. Rend. 255:2521.
cellodextrins. Anal. Biochem. 2: 521. 21 Perlin, A. S., P. Herve du Penhoat, and H. S. Isbell.
15 Montgomery, E. M., and C. S. Hudson. 1930. 1973. Carbon-13 and hydroxyl proton NMR
Relations between rotating power and structure in spectra of ketoses. Page 39 in Carbohydrates in
the sugar group. Part 27. Synthesis of a new solution. Adv. Chem. Series No. 117. A m . Chem.
disaccharide ketose (lactulose) from lactose. J. A m . Soc., Washington, DC.
C h e m . Soc. 52:2101. 22 Reithel, F. J. 1963. Lactose. Page 103 in M e t h o d s
16 Nickerson, T. A., I. F. Vujicic, and A. Y. Lin. o f enzymatic analysis. H. U. Bergmeyer, ed.
1976. Colorimetric estimation of lactose and its Academic Press, New York, NY.
hydrolytic products. J. Dairy Sci. 59: 386. 23 Tauber, H., and I. S. Kleiner. 1932. A m e t h o d for
17 Oosten, B. J. 1967. Solubility diagram of lactose the determination of monosaccharides in the
and lactulose in water. Rec. Tray. Chim. 86:675. presence of disaccharides. J. Biol. Chem. 99:249.
18 Parrish, F. W. 1970. Isomerization of glucose, 24 Verhaar, L.A.Th., M.J.M. Van Der Aaist, J.A.W.M.
maltose, and lactose with amino compounds. US Beenackers, and B.F.M. Kuster. 1979. Ion-exchange
Pat. 3,514,327. c h r o m a t o g r a p h y o f lactose-lactulose isomerization
19 Parrish, F. W., F. B. Talley, K. D. Ross, J. Clark, m i x t u r e s using a boric acid-borate eluent. J.
and J. G. Phillips. 1979. Sweetness of lactulose Chromatogr. 170: 363.

Journal of Dairy Science Vol. 63, No. 11, 1980