Polish Journal of Veterinary Sciences Vol. 23, No.

2 (2020), 317–319

DOI 10.24425/pjvs.2020.133648

Short communication

Agarose gel electrophoresis pattern

of serum alkaline phosphatase isoenzymes
in Holstein cows during lactation

A. Chiba1,2, K. Hatate1,3, R. Onomi4, T. Moriyama2, A. Goto2, N. Yamagishi1,3,4

United Graduate School of Veterinary Sciences, Gifu University, Gifu 501-1193, Japan
1
2
Veterinary Medical Center, Obihiro University of Agriculture and Veterinary Medicine,
Obihiro, Hokkaido 080-8550, Japan
3
Department of Veterinary Medicine, Obihiro University of Agriculture and Veterinary Medicine,
Obihiro, Hokkaido 080-8550, Japan
4
Graduate School of Veterinary Life Sciences and Agriculture,
Obihiro University of Agriculture and Veterinary Medicine, Obihiro, Hokkaido 080-8550, Japan

Abstract
A recent study found that an agarose gel electrophoresis (AGE) method yielded two distinct
major bands corresponding to the hepatic and bone ALP isoenzymes (ALP2 and ALP3, respec-
tively) in bovine serum treated with protease and neuraminidase (PN-treatment), although there
were concerns that the intestinal ALP isoenzyme (ALP5) often overlapped with ALP3 in human
serum treated with neuraminidase. Because ALP5 was separated from ALP3 in bovine serum
treated with protease alone (P-treatment), we used a modified method employing both
P- and PN-treated bovine sera to measure the activities of the three ALP isoenzymes in 53 lacta-
ting Holstein cows: 24 primiparous and 29 multiparous. Upon electrophoresis, 51 of 53 samples
(96.2%) subjected to P-treatment yielded a distinct fraction corresponding to ALP5, as did the
control serum. All PN-treated sera yielded a definite ALP2 fraction. The ALP3 fraction was
calculated as the remainder after excluding ALP2 and ALP5. The activities of total ALP (t-ALP)
and ALP3 in primiparous cows were higher than those in multiparous cows (p < 0.001)
at early-to-peak [10–110 days in milk (DIM)] and mid (111–220 DIM) lactation. In the multi-
parous cows, the ALP3 activity at late lactation (221−477 DIM) was significantly higher than
that at early-to-peak lactation. Thus, the modified AGE method described here is able to discrimi-
nate three fractions of ALP isoenzymes in the sera of lactating cows. The AGE pattern of circu-
lating ALP isoenzymes will contribute to the understanding of the physiological bone metabolism
status in lactating cows.

Key words: agarose gel electrophoresis (AGE), alkaline phosphatase isoenzyme, lactating cow

Correspondence to: N. Yamagishi, e-mail: yamagishi@obihiro.ac.jp, tel.: +81 155 49 5541

318 A. Chiba et al.

Introduction by subtracting the percentage of the ALP5 fraction from

the percentage of the overlapping ALP3 and ALP5
A method using a commercial agarose gel electro- fraction in the PN-treated serum. The absolute activity
phoresis (AGE) kit yielded two distinct major bands (U/L) of each isoenzyme was calculated from the t-ALP
corresponding to the hepatic ALP isoenzyme (ALP2) activity measured spectrophotometrically.
and ALP3 in bovine plasma treated with protease and All statistical analyses were performed with EZR
neuraminidase (PN-treatment) (Onomi et al. 2019); software (Kanda 2013), a graphical user interface for R
however, there were concerns that the intestinal ALP (The R Foundation for Statistical Computing, Vienna,
isoenzyme (ALP5) often overlaps with ALP3 following Austria). The cows were divided into three lactation pe-
neuraminidase treatment (Van Hoof et al. 1988). riods: the early-to-peak [10−110 days in milk (DIM)],
Because ALP5 can be separated in human serum middle (111−220 DIM), and late (221−477 DIM)
treated with protease alone (P-treatment) (Ooi et al. periods. All numerical data were expressed as medians,
2007), a modified AGE method that used both P- and with the minimum and maximum values and an inter-
PN-treated sera simultaneously was explored to better quartile range (IQR). The data were analyzed using
resolve the ALP isoenzymes in bovine serum. We used the Kruskal-Wallis test with the Steel-Dwass test to
this modified AGE method to measure the serum ALP2, evaluate the differences among the lactation periods
ALP3 and ALP5 levels in lactating Holstein cows. within primiparous or multiparous cows. The Mann-
The aim of this study was to evaluate the distribution -Whitney U-test was used to compare the values
of the serum activity of each ALP isoenzyme in primi- between primiparous and multiparous cows at the same
parous and multiparous cows at different lactation lactation period. The level of significance was set at
stages. p<0.05.

Materials and Methods Results and Discussion

We assayed 53 frozen (-20°C) serum samples col- After the P-treatment, 51 of 53 serum samples
lected from 53 lactating Holstein cows [age 1.9-8.4 years; (96.2%) yielded a fraction corresponding to ALP5
24 primiparous and 29 multiparous (2nd - 6th parity)] on the cathode side and a poorly resolved fraction
kept at our university farm. The samples had been used of ALP2 and ALP3 on the anode side (Fig. 1B). After
previously for a metabolic profile test (MPT). All cows the PN-treatment, all samples (100%) yielded a definite
were declared healthy based on the MPT results. ALP2 fraction on the anode side and a poorly resolved
AGE was performed using an electrophoresis sys- ALP3 and ALP5 fraction on the cathode side. These
tem purchased from Helena Laboratories (Saitama, observations suggest that this AGE method can
Japan), as shown previously (Onomi et al. 2019). Con- discriminate three ALP isoenzymes using P- and
trol serum (5139; Helena Laboratories) containing PN-treated sera of cows simultaneously.
extract of bovine liver or intestinal tissue was used Fig. 1C shows the distributions of the serum activi-
as a reference for ALP2 and ALP5, respectively. Each ties of t-ALP and each ALP isoenzyme in primiparous
serum sample (60 µL) was subjected to two treatments and multiparous cows at various lactation periods.
before electrophoresis; one half (30 µL) was mixed The primiparous cows had significantly higher t-ALP
with a 300 U/mL protease cocktail (4 µL) and distilled and ALP3 activities than the multiparous cows
water (2 µL) (P-treatment), and the remainder (30 µL) (p<0.001) in early-to-peak and mid lactation. In the
was mixed with the protease cocktail (4 µL) and a sepa- multiparous cows, the ALP3 activity in late lactation
rator solution containing neuraminidase (2 µL) (PN-treat- was significantly higher than that in early-to-peak lacta-
ment). After electrophoresis (23 min at 230 V and tion. There were no statistical differences in the ALP2
15°C), the gels were stained and scanned as densito- and ALP5 activities. Primiparous cows are still grow-
metric images. The P-treated serum showed a distinct ing, and bone growth and remodeling are active
ALP5 fraction emerging on the cathode side and a frac- throughout lactation; therefore, the higher serum t-ALP
tion containing poorly separated ALP2 and ALP3 on the and ALP3 activities in the primiparous cows are indica-
anode side (Ooi et al. 2007), whereas the PN-treated tive of skeletal growth. The increase in serum ALP3
serum showed a definite ALP2 fraction on the anode activity at late lactation in the multiparous cows may
side and a poorly resolved fraction of overlapping ALP3 show the activation of bone formation due to the
and ALP5 on the cathode side (Fig. 1A). The relative decreased milk yield.
percentages of the ALP2 and ALP5 fractions were In conclusion, the modified AGE method described
determined by the optical absorbance of the bands. here was able to discriminate three ALP isoenzymes
The percentage of the ALP3 fraction was assessed in the sera of lactating cows. The AGE pattern of circu-
Agarose gel electrophoresis pattern of serum alkaline ... 319

Fig. 1. Representative densitometric images (electrophoretograms) of (A) control serum and (B) a serum sample from a 2.1-year-old
primiparous cow at 264 days of lactation and (C) box-and-whisker plots (medians, interquartile ranges, maxima, and minima) showing
the distributions of the serum activities of total ALP (t-ALP) and each ALP isoenzyme in primiparous and multiparous cows in the
early-to-peak (n = 9 and 13, respectively), middle (n=8 and 7, respectively), and late (n = 7 and 9, respectively) lactation periods.
ALP2, hepatic ALP isoenzyme; ALP3, bone ALP isoenzyme; ALP5, intestinal ALP isoenzyme; ALP2+ALP3, the overlapping ALP2
and ALP3 fraction; ALP3+ALP5, the overlapping ALP3 and ALP5 fraction; Early, early-to-peak lactation; Late, late lactation;
Mid, mid lactation; P-treatment, protease treatment; PN-treatment, protease and neuraminidase treatment. Significant differences between
the primiparous and multiparous cows at each time point are shown: ** p<0.01, *** p<0.001. The letter means a significant difference from
the values at the early to middle in multiparous cows: a p<0.05.

lating ALP isoenzymes will contribute to the under- Marrow Transplant 48: 452-458.
standing of physiological bone metabolism status Onomi R, Hatate K, Yamagishi N (2019) Determination
in lactating cows, although further studies are needed of plasma bone-specific alkaline phosphatase isoenzyme
activity in Holstein calves using a commercial agarose gel
to examine its clinical implications. electrophoresis kit. Pol J Vet Sci 22: 789-792.
Ooi K, Shiraki K, Morishita Y, Nobori T (2007) High-mole-
cular intestinal alkaline phosphatase by agarose gel
References electrophoresis. J Clin Lab Anal 21: 140-146.
Van Hoof VO, Lepoutre LG, Hoylaerts MF, Chevigné R,
Kanda Y (2013) Investigation of the freely available easy- De Broe ME (1988) Improved agarose electrophoretic
-to-use software ‘EZR’ for medical statistics. Bone method for separating alkaline phosphatase isoenzymes
in serum. Clin Chem 34: 1857-1862.