Main

Human milk contains a lipase stimulated by bile salts (BSSL, EC 3.1.1.3), which is important for fat digestion in the newborn period(1, 2). The presence of BSSL in milk appears to be predominantly restricted to carnivores, and in humans it consists of 722 amino acid residues(3–6). The structural features of this enzyme have been elucidated after cDNA cloning(5, 6), and these studies have also revealed heterogeneity of mRNA species and the existence of a pseudogene for BSSL(7, 8). BSSL has two proposed glycosylation regions, a potential N-glycosylation site at asparagine 187 and an O-glycosylation region in the 16 repeating units at the C-terminal end of the protein(9–12).

To date there has been no consensus regarding the molecular mass of BSSL. Reported molecular masses using SDS-PAGE have varied between 90 and 125 kD(13–17), and it is possible that these differences arise from variations in electrophoretic mobility when different systems are used. A higher than expected apparent molecular mass has been observed after size-exclusion chromatography(9).

More recently, two molecular forms of BSSL were found in human milk (in two of six samples analyzed) with molecular masses of 97 and 120 kD(18). The molecular mass of the single molecular species was not reported in this study.

The aim of the present study was to investigate the discrepancies in reported molecular masses and to determine the incidence of molecular forms of BSSL in preterm and term human milk. The activity of BSSL was also investigated in both types of milk to determine any variations which would be relevant for fat digestion in neonates.

METHODS

All materials were purchased from Sigma Chemical Co., St. Louis, MO, unless otherwise stated.

Subjects. Human milk was donated by 40 mothers at various times during lactation. Twenty mothers (age range 24-40 y) had preterm deliveries between 27 and 36 wk (Table 1). One of these mothers also donated milk after a subsequent term pregnancy. The remaining donors(n = 20) had term pregnancies, and the age range was 21-39 y(Table 1).

Table 1 Details of milk donors and specific activity of purified BSSL
Full size table

Purification of BSSL. Milk whey was prepared as described previously by Bläckberg and Hernell(13), and it was subjected to heparin-Sepharose chromatography using 0.02 M Tris/HCl buffer, pH 7.5, and a salt gradient up to 1 M NaCl. The esterase activity of BSSL was monitored spectrophotometrically during and after the purification procedure using p-nitrophenylacetate as a substrate(19), and the specific activity of fractions was also monitored after heparin-Sepharose chromatography. The protein content of fractions was determined using the Pierce protein assay with BSA as a standard(Pierce, Rockford, IL). The molecular mass of BSSL was determined by SDS-PAGE.

SDS-PAGE. SDS-PAGE was performed according to Laemmli(20) using 7.5% acrylamide gels. Gels were stained with Coomassie Brilliant Blue G-250(21) or periodic acid-Fuchsin (Schiff reagent)(22). Molecular mass markers included β-galactosidase (116 kD), phosphorylase b (97 kD), BSA (66 kD), and chicken egg albumin (45 kD). The molecular masses of BSSL were confirmed on at least five different gels.

Gel permeation chromatography. Gel permeation chromatography was performed using a Waters 600 Multisolvent delivery system and a Waters 990 Photodiode array detector. The 7.5 mm × 30-cm Progel-TSK G3000SW column(Supelco, Bellefonte, PA) was eluted with 0.25 M potassium phosphate buffer, pH 7.0. Molecular mass markers included β-amylase (200 kD), alcohol dehydrogenase (150 kD), BSA (66 kD), chicken egg albumin (45 kD), and lactalbumin (14 kD). An additional high molecular mass form of BSSL was frequently observed after gel permeation chromatography. As this probably arose from dimerization of BSSL under the nonreducing conditions, this molecular species was reduced with 10% SDS, 6 M guanidine hydrochloride, or DTT (0.2 or 3 M).

Approval for the study was obtained from the Research Ethics Committee of the Queen's University of Belfast and all mothers agreed to take part in the study.

Statistical analysis. Results are expressed as mean ± SD unless stated otherwise. The characteristics of the two study groups were analyzed by an unpaired t test. Blood group analyses were performed using a χ2 test.

RESULTS

The two groups of mothers who took part in the study were of similar ages and blood group types and differed only in the length of gestation(Table 1). Thirty-five percent of both the preterm and term milk samples analyzed contained two molecular forms of BSSL. The combinations of molecular forms as determined by SDS-PAGE are shown in Figure 1. The most common combination of molecular masses, 15% in each group, was 98 and 120 kD, followed by 105 and 120 kD (10%)(Table 2). BSSL, when present as a single molecular species, had a molecular mass of either 115 or 120 kD, and the incidence was similar in preterm and term milk. All the molecular forms detected were glycosylated forms of BSSL as assessed by staining gels with a carbohydrate-specific stain (Fig. 2). The difference in molecular mass between the molecular species detected using SDS-PAGE was confirmed by gel permeation chromatography, as the elution position of the smaller molecular species was always at a later retention time than was the larger species (Fig. 3). However, as in the case of size exclusion chromatography(9), we also found a higher than expected molecular mass for BSSL by gel permeation chromatography, and consequently this method was used for separation purposes only.

Figure 2
figure 2

Separation of molecular forms of BSSL using SDS-PAGE and visualized using periodic acid-Fuchsin (Schiff reagent). Lane 1, 115 kD; lane 2, 120 kD; lane 3, 102 and 115 kD;lane 4, 108 and 122 kD; lane 5, 117 and 130 kD; and lane 6, 98 and 120 kD. Approximately 10 μg were added to each lane.

Full size image
Table 2 Incidence of molecular forms in preterm and term milk.
Full size table
Figure 1
figure 1

Separation of molecular forms of BSSL using SDS-PAGE and visualized using Coomassie Brilliant Blue G-250. Lane 1, 115 kD;lane 2, 120 kD; lane 3, 102 and 115 kD; lane 4, 108 and 122 kD; lane 5, 105 and 120 kD; and lane 6, 98 and 120 kD. Approximately 10 μg were added to each lane.

Full size image
Figure 3
figure 3

Separation of molecular forms of BSSL by gel permeation chromatography after heparin-Sepharose chromatography. Molecular forms of BSSL were eluted from a Progel-TSK G3000SW column with 0.25 M potassium phosphate buffer, pH 7.0 Peak 1 is a dimer of BSSL. Peak 2(upper panel) represents the 120-kD form of BSSL, and peak 3 is the 98-kD form. In the lower panel, peak 1 is a dimer and peak 2 represents the 115-kD form.

Full size image

One mother who donated milk after a preterm delivery and after a term delivery had two forms on both occasions. Multiple samples from one donor in the term group contained only one form of BSSL, suggesting no longitudinal variation in the number of forms present. The presence of either one or two molecular forms of BSSL was not related to length of lactation, maternal age, blood group, or gestation. The blood group was ascertained for 34 mothers. In the group with one form of BSSL, nine mothers (41%) were blood group O, five(23%) were A, six (27%) were B, and two (9%) were AB. In the group with two forms of BSSL, eight mothers (67%) were type O, three (25%) were A, and one(8%) was B.

The specific activities of the purified enzyme preparations were similar, whether one or two molecular forms were present (35.0 ± 19.1(n = 26); 35.8 ± 12.8 (n = 13) U/mg of protein, respectively (mean ± SEM). When two molecular forms were present, the individual forms had similar specific activities. Also, BSSL purified from preterm milk had a similar specific activity compared with that from term milk(Table 1).

After purification, the molecular forms of BSSL tended to dimerize under nonreducing conditions, e.g. in phosphate buffer after HPLC. This could occur within a few hours at 4°C and was not restricted to any particular molecular form or combination of forms. The dimerization was partially reversible by the addition of DTT at concentrations up to 0.2 M (Fig. 4). Although a greater degree of dissociation was achieved with 3 M DTT, the higher concentration appeared to interfere with absorbance measurements. Guanidine hydrochloride and SDS were less efficient as dissociating agents.

Figure 4
figure 4

Dissociation of the BSSL dimer (peak 1) with DTT to the 120-kD (peak 2) and 105-kD (peak 3) molecular forms.

Full size image

DISCUSSION

Two forms of BSSL of variable molecular mass were identified in 35% of human milk samples analyzed. The incidence is similar to that reported by Swan et al.(18), who found two molecular forms in two of six milk samples analyzed. The specific activities of the purified enzyme preparations were similar regardless of the molecular species present, so there does not appear to be any particular beneficial effect of having two molecular forms rather than one. The present study also indicates that there is no change in the number of molecular forms produced during the period of lactation, nor after different pregnancies in the same individual. We also found no correlation between maternal blood group and number of molecular forms present, even though immunochemical studies have indicated that BSSL may contain blood group-related antigenic determinants(12).

In most previous studies, the molecular mass of BSSL was determined using 10% acrylamide gels, and the corresponding molecular masses were 107 kD(10), 112 kD(14), and 120 kD(15, 16). In the present study, it has been shown that the molecular mass of this enzyme is variable, and we found that the molecular forms were not optimally separated using these electrophoretic conditions, but a more efficient separation was achieved using 7.5% gels. We also found in some instances that the bands after SDS-PAGE were diffuse, which tended to make accurate determination of the molecular mass more difficult. Another possibility is that the variation in the molecular masses reported reflected variations in the electrophoretic mobility of molecular mass markers and/or BSSL under differing electrophoretic conditions.

The variation in molecular mass and the number of forms produced could arise from differences in protein structure, differences in glycosylation, or some other phenomena. The C-terminal end of BSSL has been shown to be composed of 16 highly conserved proline-rich repeats of 11 amino acid residues that are O-glycosylated(9). The variable molecular mass could arise from differences in the number of repeating units present, as reported in other species(23) or be due to variations in glycosylation.

Results from the present investigation indicate that BSSL synthesized by the lactating mammary gland can be produced in multiple forms, and it would be of interest to determine whether this is also the case for the human pancreatic counterpart of this enzyme, carboxylic ester hydrolase. Previous studies have shown that porcine, canine and murine carboxylic ester hydrolase can exist in multiple forms, and there is also a tendency for the porcine enzyme to dimerize(14, 24). In conclusion, as BSSL can be produced in multiple forms and there are variations in molecular mass between individuals, it would be inappropriate when undertaking this type of study to pool milk samples from individual donors.