Functional regulation of xanthine oxidoreductase expression and localization in the mouse mammary gland: evidence of a role in lipid secretion

Abstract

Xanthine oxidoreductase (XOR), a key enzyme of purine metabolism, has been implicated in the secretion of the milk fat droplet in lactating mammary epithelial cells, possibly through structural interactions with other milk fat globule proteins including butyrophilin (Btn) and adipophilin (ADPH). To help determine the mechanism by which XOR is regulated, we examined the expression and localization of XOR in the non-secretory states of late pregnancy and induced involution compared with the state of active secretion. XOR mRNA levels started to increase at mid-pregnancy, turned sharply upwards at the onset of lactation and decreased rapidly with forced involution, indicating transcriptional control of the enzyme level by differentiation and secretory function. During pregnancy and involution the enzyme was diffusely distributed in the cytoplasm, but moved rapidly to the apical membrane of the cells when secretion was activated, where it colocalized with both Btn and ADPH, similar to the situation in the milk fat globule itself. Size-exclusion chromatography of solubilized milk fat globule membrane proteins showed that XOR formed a sulphydryl-bond-dependent complex with Btn and ADPH in the milk fat globule membrane. XOR returned to a diffuse cytoplasmic localization shortly after induced involution, while Btn remained localized to the apical membrane, suggesting that localization of XOR is not dependent on the presence of Btn in the apical membrane. Our findings indicate that the expression and membrane association of XOR in the mammary gland are tightly regulated by secretory activity, and suggest that the apical membrane association of XOR regulates the coupling of lipid droplets to the apical plasma membrane during milk lipid secretion.

Figure 1. Steady-state levels of xanthine oxidoreductase (XOR) and milk protein mRNA in the mammary gland during pregnancy and lactation

The average steady-state levels of XOR mRNA (▪), β-casein (♦), whey acidic protein (WAP; •) and α-lactalbumin (alpha-lac, ▴) in mammary tissue are shown for the indicated days post coitus. RNA levels were determined by RT-PCR using the primer pairs described in Methods and normalized to those of β-actin. The points show the mean ± s.e.m. for 3-4 animals. The error bars for some points lie within the symbols. The inset shows the change in the average steady-state levels of XOR in the mammary gland (closed bars) and liver (open bars) during pregnancy, lactation and involution, as determined by S1 nuclease protection. RNA values are averages of duplicates from different animals expressed as the amount of radioactivity (cpm) associated with protected bands after normalization to 25 μg of total RNA. Individual values were within 10 % of the averages. Parturition in our colony occurs between days 19 and 20 post coitus.

Figure 2. Lobuloalveolar morphology and XOR localization in alveolar epithelial cells from pregnant and lactating mice

Haematoxylin/eosin staining (a-c; scale bar + 25 μm) of mammary gland sections from mice at day 19 of pregnancy (a), day 1 of lactation (b) and day 10 of lactation (c). The presence of large intracellular lipid droplets (arrow, a), the small lumen (Lu) size and the lack of luminal material indicate the absence of significant secretory activity at this period. The significant increase in luminal size and the loss of large intracellular lipid droplets in glands at day 1 of lactation indicate active secretion (b). At day 10 of lactation, lipid droplets in the process of being secreted can be seen at the apical membrane (arrows, c). Immunolocalization of XOR (d-f; low power images, scale bar + 50 μm) in perfusion-fixed mammary glands from mice at day 19 of pregnancy (d), day 1 of lactation (e) and day 10 of lactation (f). Insets in d-f show higher power images of XOR localization (scale bar + 1 μm). Arrowheads indicate apical membrane association of XOR, arrows show XOR labelling of secreted lipid droplets. The arrow in the inset in panel f indicates XOR on the surface of a lipid droplet in the process of being secreted.

Figure 3. Effect of ovariectomy on XOR localization

Immunofluorescence staining of XOR in alveolar epithelial cells at 4 h (a), 12 h (b) and 20 h (c) after ovariectomy of mice at day 17 of pregnancy. Panels d-e show the corresponding Hoffman contrast images of a-c. Arrowheads indicate the apical membrane. Scale bar + 5 μm.

Figure 4. Effects of milk stasis on XOR expression and localization

A, relative steady-state mRNA levels of XOR (▪) and β-casein (□) in mammary glands from paired unablated (control) and nipple-ablated (ablated) animals at 18 and 42 h after nipple ablation on day 15 of lactation. RNA levels determined by RT-PCR are shown as the percentage of the respective control values at each time point. B, immunolocalization of XOR in alveolar epithelial cells (a and b, scale bar + 2 μm) from unablated mammary glands (a) and nipple-ablated mammary glands 18 h after ablation on day 15 of lactation (b). The large arrowhead in a indicates XOR at the apical membrane, and the arrow in a shows XOR staining around a lipid droplet in the process of being secreted into the lumen. Note the absence of XOR at the apical membrane in ablated tissue. C, histological and terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labelling (Tunel) analysis of mammary tissue after nipple ablation. Panels a-c show haematoxylin/eosin staining of mammary glands from unablated animals (a), animals nipple ablated for 24 h (b) and 48 h (c). Scale bar + 20 μm. Panels d-e show Tunel staining in similar sections from unablated glands (d) and glands 24 h (e) and 48 h (f) after nipple ablation. The arrowheads indicate Tunel-positive nuclei. Scale bar + 40 μm.

Figure 5. Colocalization of XOR, butyrophilin (Btn) and adipophilin (ADPH) in lactating mouse mammary tissue

XOR was separately immunolocalized with Btn or ADPH in paraffin-embedded mammary tissue from mice at day 10 of lactation. XOR immunostaining was detected with cy3-labelled secondary antibodies (red), and Btn and ADPH immunostaining were detected with fluorescein isothiocyanate (FITC; green)-labelled secondary antibodies, as described in Methods. a, merged image of XOR and Btn staining; c, XOR only staining; e, Btn only staining; b, merged image of XOR and ADPH staining; d, XOR only staining; f, ADPH only staining. Nuclei (N) and luminal regions (Lu) are indicated. Scale bar + 10 μm.

Figure 6. Colocalization of XOR, Btn and ADPH on milk lipid globule membranes

A, XOR was separately immunolocalized with Btn or ADPH in paraformaldehyde-fixed mouse milk. XOR was detected with FITC-labelled antibodies (green) and Btn and ADPH were detected with cy3-labelled antibodies (red) as described. a, Btn only staining; b, XOR only staining; c, merged image of XOR and Btn staining; d, ADPH only staining; e, XOR only staining; f, merged image of XOR and ADPH staining. Note that XOR is largely localized internal to Btn and that all ADPH staining appears to overlap with XOR, as shown by the yellow areas. B, Coomassie-blue-stained gel of mouse milk lipid droplet membrane proteins separated by SDS-PAGE. XOR, Btn and ADPH bands are indicated; their identities were verified by N-terminal microsequence analysis. C, size-exclusion chromatography (SEC) analysis of interactions between XOR, Btn and ADPH isolated from milk lipid globule membranes. Panel a shows the elution properties of XOR enzymatic activity (dashed line) and 220 nm absorbing material (continuous line) following SEC of Triton-X-100-solubilized milk lipid globule membrane proteins. The insets show silver-stained gels of proteins in peaks I and II separated by SDS-PAGE. The identities of XOR, Btn and ADPH bands were confirmed by N-terminal microsequence analysis and are indicated. XOR bands in peaks I and II are indicated by the arrow at the right of the peak II lane. Panel b shows the elution properties of XOR, Btn and ADPH after incubating peak I with 5 mm DTT. XOR activity and A220 nm absorbance are indicated by the dashed and continuous lines, respectively. Silver-stained gels of the proteins in peaks I and II are shown in the insets and the positions of XOR, Btn and ADPH are indicated as above.

Figure 7. Effect of milk stasis on Btn localization

Immunolocalization of Btn in alveolar epithelial cells (× 60) from unablated mammary glands (a) and nipple-ablated mammary glands (c) 24 h after ablation on day 15 of lactation. Panels b and d are corresponding Hoffman contrast images of a and c, respectively. Arrowheads indicate the apical surfaces, the arrow in c shows Btn staining around milk fat globules in the lumen (Lu).