Unique quadruple immunofluorescence assay demonstrates mitochondrial respiratory chain dysfunction in osteoblasts of aged and PolgA(-/-) mice

Abstract

Fragility fractures caused by osteoporosis affect millions of people worldwide every year with significant levels of associated morbidity, mortality and costs to the healthcare economy. The pathogenesis of declining bone mineral density is poorly understood but it is inherently related to increasing age. Growing evidence in recent years, especially that provided by mouse models, suggest that accumulating somatic mitochondrial DNA mutations may cause the phenotypic changes associated with the ageing process including osteoporosis. Methods to study mitochondrial abnormalities in individual osteoblasts, osteoclasts and osteocytes are limited and impair our ability to assess the changes seen with age and in animal models of ageing. To enable the assessment of mitochondrial protein levels, we have developed a quadruple immunofluorescence method to accurately quantify the presence of mitochondrial respiratory chain components within individual bone cells. We have applied this technique to a well-established mouse model of ageing and osteoporosis and show respiratory chain deficiency.

Figure 1. COX/SDH histochemistry of mouse femur.

(A) Application of sequential COX/SDH histochemistry to mouse femur at 20x magnification. The process of cutting frozen sections from bone causes tissue destruction and loss of morphology. The COX deficient cells seen within cortical bone (1) are almost certainly osteocytes based on their shape and location. However, identifying other cell types, especially those residing in trabecular bone (2) is almost impossible using this method. (B) In contrast, frozen sections of muscle retain their morphology and clear delineation of individual fibres can be seen. In this example, several fibres (1) demonstrate COX deficiency with the remaining fibres showing variable intensities of COX staining.

Figure 2. Image capture using epifluorescent light source vs confocal laser, 40x magnification.

Contrasting results of imaging can be seen when an epi-fluorescenct light source is used (top panel) compared to a laser point scanning confocal microscope (bottom panel), images taken from the same area of the same tissue section. Confocal microscopy is well known to confer several advantages over conventional microscopy. We found it to provide better clarity of target florophores with clear punctate signal emanating from target mitochondrial epitopes and lower levels of autofluorescence (as also demonstrated in Fig. 3).

Figure 3. Application of quadruple immunofluorescence assay to mouse femur, 90x magnification.

Images of bone lining osteoblasts targeted with an antibody to osteocalcin, imaged at 90x magnification using confocal laser microscopy. In comparison with aged matched control mice (upper panel), NDUFB8 (Complex I) and COX-I (Complex IV) mitochondrial respiratory chain deficiencies are seen at 11 months in PolgA^−/− mice section (lower panels) relative to mitochondrial mass (porin). The white arrows demarcate osteocytes seen within the adjacent cortical bone; the same deficiency is clear to see in these cells also.

Figure 4. Application of quadruple immunofluorescence assay to mouse osteoclasts, 90x magnification.

Application of optimised assay to mouse osteoclasts using antibody to cathepsin K, imaged at 90x magnification using confocal laser microscopy. Age matched wild type control mice demonstrate much higher signal intensities for NDUFB8 and COX-I in relation to porin levels (top panel) when compared to 11 month PolgA^−/− mice in this example (lower panel).

Figure 5. Application of quadruple immunofluorescence assay to human bone, 90x magnification.

The method also detects mitochondria in human osteoclasts with an antibody to cathepsin K, imaged at 90x magnification using confocal laser microscopy. COX-I deficiency of the osteoclasts labelled with red arrows can be seen relative to mitochondrial mass/porin signal. The osteoclast labelled with a white arrow displays normal levels of COX-I.

Figure 6. NDUFB8:Porin and COX-I:Porin ratios in mouse osteoblasts.

Following log transformation of background corrected signal intensities for porin, COX-I and NDUFB8, linear association of NDUFB8 to porin (A) and COX-I to porin (B) are evident. The strongest association exists in the youngest wild type mice at 4 months when respiratory chain deficiencies are not expected to be present at significant levels. A moderate loss of this association is seen in older wild type animals, whilst the relationship is notably weak in the PolgA^−/− osteoblasts at 11 months.

Figure 7. Representative graphs of Z-scored NDUFB8:Porin and COX-I:Porin for wild type and PolgA^−/− mice.

The mean and standard deviations of NDUFB8:porin and COX-I:porin relationships in the 4 month wild type controls are established to derive Z–scores for porin, NDUFB8 and COX-I. Z-scores for NDUFB8:porin and COX-I:porin are plotted against each other for individual mice (main graph on left of each panel). Each dot represents a single osteoblast, colour coded by the porin level (dark purple, very low; light purple, low; grey, normal; orange, high; red, very high). Typical appearances of data for 4 month old wild type mice (A), 11 month wild type mice (G), and 11 month old PolgA^−/− mice (N,T) are shown (lettering corresponds with that in Table 1). At 4 months in wild type animals, the vast majority of data points are no more than 3SD from the mean of this young control group. Reduced NDUFB8 and COX-I protein expression is seen in 11 month wild type animals and to a greater degree in 11 month PolgA^−/− mice. The corresponding percentage of osteoblasts which are positive, intermediate positive, intermediate negative or negative for each mouse are also shown on the right side of each panel, for NDUFB8 (top) and COX-I (bottom).