Skip to main content
NCBI home page
Journal of Histochemistry and Cytochemistry logoLink to Journal of Histochemistry and Cytochemistry
. 2011 Aug;59(8):769–779. doi: 10.1369/0022155411412185

Quantification of Ceroid and Lipofuscin in Skeletal Muscle

Hatice Tohma 1,2, Anna R Hepworth 1,2, Thea Shavlakadze 1,2, Miranda D Grounds 1,2, Peter G Arthur 1,2,✉
PMCID: PMC3261605  PMID: 21804079

Abstract

Ceroid and lipofuscin are autofluorescent granules thought to be generated as a consequence of chronic oxidative stress. Because ceroid and lipofuscin are persistent in tissue, their measurement can provide a lifetime history of exposure to chronic oxidative stress. Although ceroid and lipofuscin can be measured by quantification of autofluorescent granules, current methods rely on subjective assessment. Furthermore, there has not been any evaluation of variables affecting quantitative measurements. The article describes a simple statistical approach that can be readily applied to quantitate ceroid and lipofuscin. Furthermore, it is shown that several factors, including magnification tissue thickness and tissue level, can affect precision and sensitivity. After optimizing for these factors, the authors show that ceroid and lipofuscin can be measured reproducibly in the skeletal muscle of dystrophic mice (ceroid) and aged mice (lipofuscin).

Keywords: lipofuscin, ceroid, quantification, oxidative stress, biomarker, image analysis

Ceroid and lipofuscin are autofluorescent granules that persist in tissue (Terman and Brunk 1998) and are generated as a consequence of oxidative stress (Sohal and Brunk 1989). Ceroid accumulation is observed in oxidative stress–related pathological conditions including malnutrition, genetic defects, trauma, infections, irradiation, and hypoxia (Porta 2002). Specific conditions leading to generation of ceroid include vitamin E deficiency (Fattoretti 2002), Batten disease (Hall et al. 1991), Hermansky-Pudlak syndrome (Garay et al. 1979), and Sanfilippo’s syndrome (Dowson et al. 1989). Ceroid is composed predominantly of lipids and aggregated proteins (Elleder 1991) and can be detected by autofluorescence with a peak emission of about 539 nm when excited at 360 nm (Seehafer and Pearce 2006).

Lipofuscin is widely considered to be a marker of aging and accumulates in aging postmitotic cells such as cardiac myocytes and neurons (Samorajski et al. 1965). Lipofuscin occurs when oxidized lipids, proteins, and carbohydrates become resistant to hydrolysis by lysosomal enzymes (Brunk and Terman 2002a). Lipofuscin can be detected by autofluorescence using a range of excitation wavelengths including ultraviolet (330–380 nm), blue (450–490 nm), and green (510–560 nm) light with the barrier filters of 420, 520, and 590 nm, respectively (Brunk and Terman 2002b).

The distinction between ceroid and lipofuscin can be ambiguous in tissue sections. Ceroid typically refers to granules generated during pathological conditions, whereas lipofuscin is used to describe granules accumulating with age in postmitotic tissue (Porta 2002). Various criteria have been proposed to identify ceroid and lipofuscin; however, use of the terms can overlap (Porta 2002).

Because ceroid and lipofuscin are persistent in tissue, their quantitative measurement can be particularly useful in assessing the extent to which a tissue has been exposed to chronic oxidative stress. Ceroid and lipofuscin are observed by fluorescent microscopy as condensed granules in tissue (Nakae et al. 2008; Vila et al. 2000), and quantification commonly involves capturing images of histological sections and then using an image program to calculate the amount of ceroid/lipofuscin as the area occupied by the granules within each image. However, this method depends on qualitative judgments by the observer to define the edge of each granule. Furthermore, there has not been a detailed description of the variables that can affect the precision or sensitivity of measurement with fluorescent microscopy. These issues are addressed in the present study, where we describe a nonsubjective method of quantification and identify and resolve the measurement variables using mouse skeletal muscle. To develop this technique, ceroid was examined in dystrophic quadriceps muscles of the mdx mouse: This is a model of the human condition Duchene muscular dystrophy that is characterized by elevated oxidative stress (Dudley et al. 2006; Nakae et al. 2004). We show that this optimized technique is also applicable to the quantification of lipofuscin in skeletal muscles from aged normal mice.

Materials and Methods

Muscle Tissue Preparation

Dystrophic female mdx (n = 6) and normal female C57 (n = 6) mice (the C57Bl/10ScSn parental strain for mdx) aged 3 months were obtained from the Animal Resource Centre, Murdoch, Western Australia. Muscle samples were also obtained from young (3 months, n = 4) and aged (27 months, n = 4) female C57Bl/6J mice. All animal procedures were carried out in strict accordance with the guidelines of the National Health and Medical Research Council of Australia. Mice were fed a normal laboratory diet and killed by cervical dislocation under anesthesia. Quadriceps muscles were removed from each mouse, embedded in tragacanth gum (Sigma Aldrich G1128) on a small cork block, and then quenched in isopentane cooled in liquid nitrogen. Tissues were stored at –80C until used. One quadriceps muscle from each group was used for technical development, and the remaining muscles were used for the quantitative assessment of method. Frozen 10-µm sections were cut at –23C using a Leica CM 3050 S cryostat. Nine sections (10 µm thick) were taken from three different levels (three sections per level) of the quadriceps muscle of one mdx and one aged sample. Additional 5-µm and 20-µm sections were prepared from the same tissues. All sections were dried at room temperature for 1 hr prior to mounting with using DPX glue and cover slips.

Microscope and Image Capture

Sections were analyzed with a fluorescent LEICA DM RB/E microscope equipped with Q imaging micropublisher 3.3 RTV camera, a 100-W mercury lamp, a 450- to 490-nm excitation filter (blue), and a 515-nm emission barrier filter. Images of each section were captured using Imagepro (version 6.2.0424 for Windows 2000/XP). All sections were scanned using an automatic stage control setting that generated a grid structure of images covering the entire section. Magnifications of ×10 for ceroid and ×40 for lipofuscin were used with a 3-sec exposure time to capture each image unless otherwise stated.

Image Analysis

For each scanned section, images with complete tissue coverage were selected for analysis. Images with tissue edges or obvious artifacts were discarded. Quantification of ceroid and lipofuscin granules was performed with ImageJ (version 1.42q, Wayne Rasbanb, National Institutes of Health, Bethesda, MD; available at http://rsb.info.nih.gov/nih-image/). The green channel of an 8-bit RBG (red, blue, green) image was analyzed. Threshold levels were determined (see Results section for image analysis) and applied to images to discriminate ceroid and lipofuscin from background signal. The area occupied by fluorescent granules was expressed as a percentage of total image area [(area of granules/total image area] × 100).

Data Modeling

Data modeling used a bootstrapping method based on Good (2005). Bootstrapping uses the available data as the population and then takes repeated subsets from the population to determine variability. This technique was used to determine the variability in quantifying ceroid when a select number of images (image set size) were used from a larger population of images (the scanned section). Images were collected at ×10, ×20, and ×40 magnification for three sections: section 1, 24–64 useable images; section 2, 25–35 useable images; and section 3, 20–35 useable images. The bootstrapping population was the number of useable images obtained for each section, and for a combined analysis all useable images from all sections were used (69–127 images). For each bootstrapping population (section 1, section 2, section 3, and combined), image set sizes were 6, 12, 18, or 24 images (selected without replacement). For each question addressed, 1000 subsets were sampled from the relevant population. For each subset, the mean percentage area of ceroid, the median, and the standard deviation were calculated. Variation in these numbers was then compared.

Statistical Analysis

Statistical analysis was carried out in Microsoft Excel and SPSS 16.0. Comparisons of percentage area for two groups (mdx vs C57; aged vs. young) used Student’s independent samples t-test, and comparisons for three groups (day of measurement, thickness of section, and comparison of tissue levels) used one-way ANOVA. Tukey’s honestly significant difference was used as the post hoc test when significant results were seen in the ANOVA. All tests were 2-tailed. Results were considered to be significant with p < 0.05.

Results

Method of Image Analysis

Ceroid was evident in quadriceps muscle from dystrophic (mdx) mice as autofluorescent granules (Fig. 1A), which were absent in age-matched normal C57 mice (Fig. 1B). Lipofuscin was evident as granules in tissue sections taken from aged (27-month-old) C57Bl/6J mice (Fig. 1C), which were absent in younger (3-month-old) mice (Fig. 1D). Tissue section images could also be represented in a histogram format where the number of pixels was plotted against pixel intensity (Fig. 2). For mdx and C57 images, the pixel intensity plots were broadly comparable, except for a small fraction of pixels with higher intensities in the mdx image (Fig. 2A). We interpreted this to mean that most of the pixels in mdx images were associated with a background signal comparable to C57 images (which lack ceroid), with the pixels at higher intensities representing ceroid. We examined whether it was possible to quantify ceroid in an image by counting the number of pixels associated with ceroid. This required a means of establishing a pixel intensity threshold, with the pixel intensities below the threshold representing background and the pixel intensities above the threshold representing ceroid. To identify pixels associated with the background signal, we tested the maximum pixel intensity of normal C57 images as a threshold, such that pixel intensities below this threshold in the mdx image could be classed as background. However, this method was not satisfactory, because the maximum pixel intensity varied between normal C57 images taken under the same conditions (Fig. 2B).

Figure 1.

Figure 1.

Open in a new tab

Fluorescence images of transverse frozen sections of quadriceps muscle. Example images for ceroid analysis are for 3-month-old (A) mdx and (B) C57 mice. Example images for lipofuscin analysis are for (C) 27-month-old and (D) 3-month-old C57Bl/6J mice. Fluorescent images were collected using a 450- to 490-nm excitation and 505- to 520-nm emission filter with a 3-sec exposure. Scale bar is 25 µm.

Figure 2.

Figure 2.

Open in a new tab

Representative intensity plots of images. (A) Expanded pixel intensity plots from one image collected from mdx (solid line) and one image from C57 (dashed line) muscle sections. (B) Expanded pixel intensity plots of five randomly selected images from C57 muscle sections. Inserts show the complete pixel intensity plots. Data are from images collected at ×20 magnification.

There was a clear differentiation between mdx and normal muscle when the distribution of the pixel intensity plots were transformed into sample quantile–theoretical quantile plots (q–q) (Fig. 3). The deviation from normal distribution for mdx muscle q–q plots was tested as a basis for discriminating ceroid signal from background signal. To estimate the percentage area of the image that was ceroid, a pixel intensity threshold was calculated to account for background signal. For each pixel intensity plot, the mean and standard deviation were calculated. The pixel intensity threshold was calculated as mean + (n) × standard deviation, where (n) was a threshold standard deviation multiplier.

Figure 3.

Figure 3.

Open in a new tab

Representative quantile–quantile plots. The plots contrast data from pixel intensity plots for (A) C57 and (B) mdx images with normally distributed theoretical data. The diagonal line indicates where normally distributed experimental data would fall. Data are from images collected at ×10 magnification.

The number of pixels above the pixel intensity threshold was used as an estimate of ceroid and expressed as a percentage of total pixel number. Increasing the threshold multiplier (n) resulted in an increase in the signal ascribed to background and a decrease in signal ascribed to ceroid (Fig. 4A). Choosing a threshold multiplier (e.g., n = 10) that ascribed all of the signal in the normal (control) images to background levels resulted in an underestimation of the ceroid content in the mdx images (Fig. 4A). Instead, to maximize discrimination between mdx and normal images, a threshold multiplier of 6 was chosen (Fig. 4A). A threshold multiplier of 6 captured pixels primarily associated with ceroid (Fig. 4C) with minimal detection of background pixels (Fig. 4E). This method generated data equivalent to those gathered with the usual method of manually tracing an outline of each autofluorescent granule (Fig. 5A). We also tested this approach for the analysis of lipofuscin. A threshold multiplier of 4 maximized the discrimination between aged and young images (Fig. 4B) and captured pixels primarily associated with lipofuscin (Fig. 4D) with minimal detection of background pixels (Fig. 4F). The method generated data equivalent to those gathered by manually tracing autofluorescent granules (Fig. 5B).

Figure 4.

Figure 4.

Open in a new tab

Use of standard deviation multipliers to estimate percentage area of ceroid. (A) Percentage area of signals from C57 and mdx images (left axis) and the difference between mdx and C57 percentage area (right axis) according to threshold standard deviation multipliers of 5 to 10 (n = 5). (B) Percentage area of signals from young and aged images (left axis) and the difference between aged and young percentage area (right axis) according to threshold standard deviation multipliers of 3 to 10 (n = 5). For A and B, a solid line shows the measured percentage area and a dashed line shows the difference of the area. (C) Quantile–quantile plots for 3 mdx images (1, 2, 3) with the pixel intensity threshold shown as a horizontal line (threshold multiplier = 6). (D) Quantile–quantile plots for three images from aged mouse tissue (1, 2, 3) with the pixel intensity threshold shown as a horizontal line (threshold multiplier = 4). (E) Quantile–quantile plots for three C57 images (1, 2, 3) with the pixel intensity threshold shown as a horizontal line (threshold multiplier = 6). (F) Quantile–quantile plots for three images from young mouse tissue (1, 2, 3) with the pixel intensity threshold shown as a horizontal line (threshold multiplier = 4).

Figure 5.

Figure 5.

Open in a new tab

Comparison of methods to measure autofluorescent granule content. (A) Ceroid content was estimated in five mdx images using a threshold multiplier of 6 and by manually tracing all visible granules. (B) Ceroid content was estimated in 5 aged images using a threshold multiplier of 4 and by manually tracing all visible granules.

Effect of the number of images on measurement precision

The distribution of ceroid in the sections was heterogeneous (Fig. 6A). Although it would be possible to estimate the total ceroid content of a tissue section by quantifying the ceroid in each individual image, this would be time consuming. Instead, we estimated the number of images required to generate a reasonable estimate of ceroid content in the section.

Figure 6.

Figure 6.

Open in a new tab

Precision of the estimated ceroid value in mdx tissue sections. (A) Variability in estimated ceroid content from 30 images captured from a single mdx quadriceps section using a threshold standard deviation multiplier of 6. (B) Variation in quantifying ceroid when using different numbers of images. Boxes show first and third quantiles. Vertical lines show minimum and maximum values. Horizontal bars in each box are median values. Dashed line represents the value of ceroid content estimated by the analysis of all usable images from three sections. (C) Reproducibility of ceroid measurement performed on three different days (10-µm thickness, n = 24). Data are from images collected at ×10 magnification. Bar indicates p < 0.05 relative to Day 21.

To determine the variability of the images taken from the mdx and normal C57 muscle sections, comparisons were made within and between three sections of one muscle at ×10, ×20, and ×40 magnification. These comparisons were performed using a bootstrap method as described in Materials and Methods. Comparable results were obtained for the different magnifications, with results for ×10 magnification shown in Fig. 6B. As the number of images per section increased, there was reduced measurement variation (Fig. 6B). Selection of images from multiple sections gave a better approximation of the mean ceroid value of the tissue than was gained by taking the same number of images from a single section (Fig. 6B). Given these observations, the following procedure was adopted: Images for three sections were collected, and eight images from each section were randomly selected for analysis (24 in total) from the useable images.

Precision of repeat measurements

To establish whether repeated ceroid measurement over time gave consistent results, ceroid analysis was performed three times at 10-day intervals by the same operator. At each of these times, 24 images from the same three sections of mdx muscle were analyzed for ceroid content at ×10 magnification. For the first 10-day period, the results were consistent (Fig. 6C), indicating that the fluorescent properties of ceroid were stable for at least for 10 days. However, in the subsequent measurement, there was a significant reduction in the percentage area ascribed to ceroid, suggesting that long-term storage of slides may result in underestimation of ceroid (Fig. 6C).

Variables Affecting Estimated Autofluorescent Granule Content

Having established criteria for the measurement of autofluorescent granule content, we next identified variables that affected the sensitivity and the precision of estimating granule content for ceroid and lipofuscin.

Effect of magnification

To evaluate the effect of magnification on ceroid detection, sections were assessed at three magnifications. The most distinct high pixel intensity values were obtained at the highest (×40) magnification, whereas there was minimal signal discrimination at the lowest (×10) magnification (Fig. 7). The relationship between ceroid pixel intensity and magnification was also reflected in images for a heat map of pixel intensities (Fig. 7). These data show that increasing the magnification improved the sensitivity of ceroid detection.

Figure 7.

Figure 7.

Open in a new tab

Ceroid signal at different magnifications. Pixel intensity plots of C57 (n = 3) and mdx (n = 3) images at (A) ×10, (B) ×20, and (C) ×40 magnification. Heat maps of ceroid spots in images at (D) ×10, (E) ×20, and (F) ×40 magnification. Images collected for tissue of 10-µm thickness, with a 3-sec exposure. Scale bar 25 µm.

To assess the effect of magnification on quantification of ceroid, selected images (n = 30 images) from the same sections of mdx muscle were analyzed at three different magnifications (×10, ×20, and ×40). There were no significant changes in the percentage of area ascribed to ceroid in mdx images for all three magnifications (Fig. 8A).

Figure 8.

Figure 8.

Open in a new tab

Variability in estimated granule content. (A) Percentage area of ceroid in mdx tissue sections at three magnifications (n = 30). (B) Ceroid content in sections at different thickness taken from the same tissue level (×10 magnification, n = 24). (C) Calculated lipofuscin content at different tissue thickness (n = 24). Bar indicates p < 0.05.

Lipofuscin was not detectable at ×10 magnification (data not shown), so the effect of magnification on lipofuscin quantification was not evaluated. All subsequent analyses for lipofuscin were performed at ×40 magnification.

Effect of tissue thickness on granule quantification

To determine whether the thickness of a section influenced the estimation of ceroid, 24 images (8 images per section) were analyzed at thicknesses of 5 µm, 10 µm, and 20 µm (three sections per magnification). As the thickness of the section increased, there was an increase in ceroid signal, which allowed for better discrimination of the mdx images from the normal C57 images (Fig. 8B).

The quadriceps of young and aged muscle were also analyzed to determine the effect of thickness (5 µm, 10 µm, and 20 µm) on lipofuscin quantification. In contrast to the ceroid data, increased tissue thickness gave variable results for lipofuscin measurement in aged tissue (Fig. 8C). Additionally, as tissue thickness increased, the apparent signal ascribed to lipofuscin in young tissue increased, which decreased discrimination between young and aged tissues (Fig. 8C). Consequently, maximum discrimination between young and aged tissues was achieved with the thinnest section (5 µm). Taken together, these data indicate that the different thicknesses of tissue should be tested for maximal discrimination.

Photobleaching by exposure to blue light

Ceroid and lipofuscin were tested for resistance to photobleaching by exposing tissue sections to blue light (e.g., 450–490 nm) at ×10 magnification for ceroid and at ×40 magnification for lipofuscin. Following exposure to blue light, there was evidence of photobleaching (Fig. 9A,B) and there was an overall decrease in the fluorescence in both mdx and aged mouse images (Fig. 9C). Nevertheless, the calculated percentage area of ceroid remained relatively consistent (Fig. 9D). For lipofuscin, photobleaching caused a substantial decrease in estimated percentage area (Fig. 9E). Possibly this was a consequence of the higher magnification leading to a more intense exposure of the tissue to light. In summary, photobleaching has the potential to reduced estimated granule content, which could be minimized by using short (up to 15 min) scanning times.

Figure 9.

Figure 9.

Open in a new tab

The effect of photobleaching on quantification of ceroid and lipofuscin. An image taken from mdx mouse tissue (A) before and (B) after exposure to blue light. Arrows indicate the same patch of ceroid. (C) Reduction in the integrated optical density of the images taken from mdx and aged tissue following continuing exposure to blue light. Calculated percentage area of (D) ceroid and (E) lipofuscin in images taken from mdx and aged tissue sections, respectively, with a standard deviation multiplier of 6. Scale bar is 25 µm.

Tissue distribution of granules

Estimates of ceroid and lipofuscin could be affected by granule heterogeneity throughout the tissue. To evaluate the impact of granule heterogeneity, sections were taken from three portions of tissue 500 µm apart for mdx and 100 µm for aged tissue. These data show that the distribution throughout the skeletal muscle tissue was relatively consistent for ceroid (Fig. 10A) but not for lipofuscin (Fig. 10B). This observation indicates that tissue sections for lipofuscin should be consistently sampled from similar parts of the muscle to minimize this type of variability.

Figure 10.

Figure 10.

Open in a new tab

Quantification of ceroid and lipofuscin in dystrophic and aged muscles. Measurement of ceroid and lipofuscin content throughout quadriceps muscle from (A) mdx and (B) aged mouse (n = 24 images for each level). Ceroid content (C) was estimated from images taken at ×10 magnification for 3-month-old mdx mice (n = 6 mice). To minimize measurement variability, image collection for each mdx mouse was matched with image collection from a 3-month-old normal (C57) mouse to estimate background signal. Lipofuscin content (D) was estimated from images taken at ×40 magnification in 27-month-old C57Bl/6J mice (n = 4 mice). Image collection for each aged mouse was matched with image collection from a young (3 month old) mouse to estimate background signal. Bar indicates p< 0.05.

Application of the Method

Having identified and optimized the variables affecting the measurement of granules, we applied the method to estimate ceroid and lipofuscin content in muscle tissue. For dystrophic mdx muscle, ceroid content was about 4-fold higher than the background signal in normal muscle (Fig. 10C). The criteria used for optimizing ceroid quantification were also applied to lipofuscin quantification, and comparative analysis showed a significant (about 3-fold) increase in lipofuscin content in aged (27-month-old) compared with young (3-month-old) skeletal muscles (Fig. 10D). Together, these data show that ceroid and lipofuscin can be reproducibly estimated in mouse skeletal muscle using the quantitative technique that we have developed.

Discussion

Autofluorescent granules in muscle have been quantified previously by image analysis of histological sections (Nakae et al. 2004). We have refined this approach by developing an operator independent method using tissue sections of diseased and aged skeletal muscle. Furthermore, we have identified and resolved variables affecting the sensitivity and precision of measuring autofluorescent granules.

The shortcomings of previous methods to measure autofluorescent aggregates have been documented. As has been described in earlier publications, one approach is to manually trace an outline of each autofluorescent aggregate (Bluhm et al. 2002; Kodama et al. 2005). However, this requires the operator to select the edge of the aggregate, and thus operator bias can influence the measurement. To eliminate operator bias, double analysis of images has been performed, although this is clearly time consuming (Peixoto et al. 2002). Background autofluorescence can also affect the estimated granule content in tissue (Sundelin and Nilsson 2001). To avoid operator bias and to address the issue of background autofluorescence, we developed a simple statistical method to define threshold intensity, above which the pixels were taken to be derived from autofluorescent granules. Other algorithms presumably could be applied to identify autofluorescent granules, but these will need to be experimentally tested (Sezgin and Sankur 2004).

The heterogeneity of distribution of autofluorescent granules in tissue sections poses a challenging issue for accurate estimates in muscle. Previous attempts to address heterogeneity included analyzing multiple images, with examples of up to 6 images for heart (Nakano 1992) and 14 for brain (Fonseca et al. 2005). Multiple sections have also been analyzed, with use of up to 10 sections per animal in a study of shrimps (Kodama et al. 2005). These studies did not test whether multiple images or sections improved measurement precision, whereas our data demonstrate that selecting multiple images can substantially improve measurement precision. However, increasing the number of images to be measured also increased workload. We found that analysis of 24 images for ceroid in muscle tissue of mdx mice and for lipofuscin in muscle tissue of aging mice provided a balance between workload and acceptable measurement precision.

Our interest in measuring ceroid and lipofuscin stems from evidence indicating that ceroid and lipofuscin are generated as a consequence of oxidative stress (de Gritz et al. 1994). Because ceroid and lipofuscin are persistent in tissue, their measurement may provide a lifetime history of cumulative exposure to oxidative stress. This type of measure would have several benefits. First, it would be easier to track the oxidative stress where exposure to oxidative stress is transient and unpredictable. For example, dystrophy in mdx mice is characterized by regular bouts of necrosis during growth that may be linked to transient changes in oxidative stress (Disatnik et al. 1998). Our data support previous observations that ceroid is elevated in mdx muscle (Nakae et al. 2004), and with our method we found that ceroid could be measured reproducibly between adult mdx mice. Second, accumulating ceroid or lipofuscin may provide the additional sensitivity required to track low levels of chronic oxidative stress that have the potential to cause tissue dysfunction over the longer term. In this context, we found that our method could measure lipofuscin, long considered a marker of aging, with good reproducibility.

In summary, our results show that a number of factors such as magnification, tissue section thickness, photobleaching, and tissue levels can affect the precision and sensitivity of ceroid and lipofuscin measurement. Based on these data, we developed a protocol that maximized measurement precision, sensitivity, and efficiency for assessing ceroid and lipofuscin content. Reproducible measurements for ceroid and lipofuscin will enable investigators to test whether ceroid and lipofuscin can be used to sensitively track the long-term development of oxidative stress, with potential applications to track the efficacy of interventions that reduce oxidative stress.

Acknowledgments

We thank Jessica Terrill for assistance with animal sampling and Guy Ben Array for assistance with microscopy.

Footnotes

The author(s) declared no potential conflicts of interest with respect to the authorship and publication of this article.

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was made possible by funding from the National Health and Medical Research Council of Australia (MG, PA and TS) and a scholarship from the Education Ministry of Turkey to Hatice Tohma. Anna Hepworth is funded from an unrestricted grant from Medela Ag (Switzerland) to the University of Western Australia.

References

  1. Bluhm BA, Beyer K, Niehoff B. 2002. Brain structure and histological features of lipofuscin in two Antarctic caridea (decapoda). Crustaceana. 75:61–76 [Google Scholar]
  2. Brunk UT, Terman A. 2002a. The mitochondrial-lysosomal axis theory of aging: accumulation of damaged mitochondria as a result of imperfect autophagocytosis. Eur J Biochem. 269:1996–2002 [DOI] [PubMed] [Google Scholar]
  3. Brunk UT, Terman A. 2002b. Lipofuscin: Mechanisms of age-related accumulation and influence on cell function. Free Radic Biol Med. 33:611–619 [DOI] [PubMed] [Google Scholar]
  4. de Gritz BG, Rahko T, Korpela H. 1994. Diet-induced lipofuscin and ceroid formation in growing pigs. J Comp Pathol. 110:11–24 [DOI] [PubMed] [Google Scholar]
  5. Disatnik MH, Dhawan J, Yu Y, Beal MF, Whirl MM, Franco AA, Rando TA. 1998. Evidence of oxidative stress in mdx mouse muscle: studies of the pre-necrotic state. J Neurol Sci. 161:77–84 [DOI] [PubMed] [Google Scholar]
  6. Dowson JH, Wilton-Cox H, Oldfors A, Sourander P. 1989. Autofluorescence emission spectra of neuronal lipopigment in mucopolysaccharidosis (Sanfilippo’s syndrome). Acta Neuropathol. 77:426–429 [DOI] [PubMed] [Google Scholar]
  7. Dudley RW, Khairallah M, Mohammed S, Lands L, Des Rosiers C, Petrof BJ. 2006. Dynamic responses of the glutathione system to acute oxidative stress in dystrophic mouse (mdx) muscles. Am J Physiol Regul Integr Comp Physiol. 291:R704–710 [DOI] [PubMed] [Google Scholar]
  8. Elleder M. 1991. Primary extracellular ceroid type lipopigment: a histochemical and ultrastructural study. Histochem J 23:247–258 [DOI] [PubMed] [Google Scholar]
  9. Fattoretti P. 2002. Morphometry of age pigment (lipofuscin) and of ceroid pigment deposits associated with vitamin E deficiency. Arch Gerontol Geriatr. 34:263. [DOI] [PubMed] [Google Scholar]
  10. Fonseca DB, Sheehy MR, Blackman N, Shelton PM, Prior AE. 2005. Reversal of a hallmark of brain ageing: lipofuscin accumulation. Neurobiol Aging. 26:69–76 [DOI] [PubMed] [Google Scholar]
  11. Garay SM, Gardella JE, Fazzini EP, Goldring RM. 1979. Hermansky-Pudlak syndrome: pulmonary manifestations of a ceroid storage disorder. Am J Med. 66:737–747 [DOI] [PubMed] [Google Scholar]
  12. Good PI. 2005. Introduction to statistics through resampling methods and R/S-PLUS Hoboken, NJ: Wiley-Interscience [Google Scholar]
  13. Hall NA, Lake BD, Dewji NN, Patrick AD. 1991. Lysosomal storage of subunit c of mitochondrial ATP synthase in Batten’s disease (ceroid-lipofuscinosis). Biochem J 275(Pt 1):269–272 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kodama K, Yamakawa T, Shimizu T, Aoki I. 2005. Age estimation of the wild population of Japanese mantis shrimp Oratosquilla oratoria (Crustacea: Stomatopoda) in Tokyo Bay, Japan, using lipofuscin as an age marker. Fish Sci. 71:141–150 [Google Scholar]
  15. Nakae Y, Hirasaka K, Goto J, Nikawa T, Shono M, Yoshida M, Stoward P. 2008. Subcutaneous injection, from birth, of epigallocatechin-3-gallate, a component of green tea, limits the onset of muscular dystrophy in mdx mice: a quantitative histological, immunohistochemical and electrophysiological study. Histochem Cell Biol. 129:489–501 [DOI] [PubMed] [Google Scholar]
  16. Nakae Y, Stoward PJ, Kashiyama T, Shono M, Akagi A, Matsuzaki T, Nonaka I. 2004. Early onset of lipofuscin accumulation in dystrophin-deficient skeletal muscles of DMD patients and mdx mice. J Mol Histol. 35:489–499 [DOI] [PubMed] [Google Scholar]
  17. Nakano M. 1992. Accumulation of cardiac lipofuscin depends on metabolic rate of mammals. J Gerontol. 47:B126. [DOI] [PubMed] [Google Scholar]
  18. Peixoto S, Aguado N, D’Incao F, Wasielesky W, Cousin JC. 2002. Preliminary identification and quantification of the age-pigment lipofuscin in the brain of Farfantepenaeus paulensis (Crustacea: Decapoda). Braz J Biol. 62:871–876 [DOI] [PubMed] [Google Scholar]
  19. Porta EA. 2002. Pigments in aging: an overview. Ann N Y Acad Sci. 959:57–65 [DOI] [PubMed] [Google Scholar]
  20. Samorajski T, Ordy JM, Keefe JR. 1965. The fine structure of lipofuscin age pigment in the nervous system of aged mice. J Cell Biol. 26:779–795 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Seehafer SS, Pearce DA. 2006. You say lipofuscin, we say ceroid: defining autofluorescent storage material. Neurobiol Aging. 27:576–588 [DOI] [PubMed] [Google Scholar]
  22. Sezgin M, Sankur B. 2004. Survey over image thresholding techniques and quantitative performance evaluation. Journal of Electronic Imaging. 13:146–168 [Google Scholar]
  23. Sohal RS, Brunk UT. 1989. Lipofuscin as an indicator of oxidative stress and aging. Adv Exp Med Biol. 266:17–26; discussion 27–19 [DOI] [PubMed] [Google Scholar]
  24. Sundelin SP, Nilsson SE. 2001. Lipofuscin-formation in retinal pigment epithelial cells is reduced by antioxidants. Free Radic Biol Med. 31:217–225 [DOI] [PubMed] [Google Scholar]
  25. Terman A, Brunk UT. 1998. On the degradability and exocytosis of ceroid/lipofuscin in cultured rat cardiac myocytes. Mech Ageing Dev 100:145–156 [DOI] [PubMed] [Google Scholar]
  26. Vila Y, Medina A, Megina C, Ramos F, Sobrino I. 2000. Quantification of the age-pigment lipofuscin in brains of known-age, pond-reared prawns Penaeus japonicus (Crustacea, decapoda). J Exp Zool. 286:120–130 [DOI] [PubMed] [Google Scholar]

Articles from Journal of Histochemistry and Cytochemistry are provided here courtesy of The Histochemical Society

RESOURCES