A composite bar graph showing age-diet effect as means with ±1 SD for all of the variables studied from 6 groups of rats. AL indicates ad libitum–fed; CR, calorie restriction; GAG, glycosaminoglycan. Young adult rats were 4 months old; adult rats, 12 months old; and old rats, 24 months or older.
Histological sections of the abdominal skin from adult rats in the calorie restriction (A) and ad libitum–fed (B) groups, stained with the sequence of hematoxylin-eosin and phloxine. Diminution in epidermal width is evident in the calorie-restricted animal (A). D indicates dermis; E, epidermis; P, pilosebaceous unit (original magnification ×200).
Section from an adult rat in the calorie-restriction group stained for elastic fibers (arrow) and collagen bundles (C) by the Verhoeff–van Gieson method (original magnification ×400).
Illustration of colloidal iron staining (blue-green) showing the presence of glycosaminoglycans (asterisk) interspersed between collagen bundles (C) (original magnification ×400).
Bhattacharyya TK, Merz M, Thomas JR. Modulation of Cutaneous Aging With Calorie Restriction in Fischer 344 RatsA Histological Study. Arch Facial Plast Surg. 2005;7(1):12-16. doi:10.1001/archfaci.7.1.12
Author Affiliations: Department of Otolaryngology–Head and Neck Surgery, University of Illinois at Chicago.
Correspondence: Tapan K. Bhattacharyya, PhD, Department of Otolaryngology–@Head and Neck Surgery, University of Illinois at Chicago, 1855 W Taylor St, Chicago, IL 60612 (firstname.lastname@example.org).
Objective To examine whether histological changes in skin owing to intrinsic aging in a laboratory rodent model are modulated by caloric restriction (CR).
Methods The abdominal skin from colony-raised ad libitum–fed Fischer 344 rats and age-matched rats subjected to CR was studied in the light microscope using histological morphometric methods. Animals 4, 12, and 24 months or older were used in this study. We studied the skin to obtain (1) quantitative data on the depth of the epidermis, dermis, and fat layer, the epidermal cellular density, the percentage fraction of dermal collagen, elastic fibers, pilosebaceous units, and capillaries, and the fibroblast density; and (2) qualitative assessment of histological staining for dermal glycosaminoglycans. We analyzed data by means of general linear model 2-way analysis of variance to obtain significance for the effects of age, diet, and age-diet interaction.
Results The ad libitum–fed rats showed age-related increase in the depth of the epidermis, dermis, and fat layer. Calorie restriction prevented these changes, but epidermal nuclear density appeared to be stimulated. A trend toward increased values for collagen and elastic fibers, fibroblasts, and capillaries in skin samples from CR rats was observed. Pilosebaceous units were not modified. Moderately reduced staining for the dermal glycosaminoglycans in the skin of CR rats was noticed.
Conclusions Histomorphological changes resulting from intrinsic aging affected some of the studied variables in the rat skin, and these changes were delayed or prevented by CR. Some stimulatory effects, such as increased densities of fibroblasts and capillary profiles and higher values of connective tissue fibers resulting from CR, were also observed. Cutaneous morphological changes due to natural aging in this rat model seem to be modified by physiological or metabolic alterations imposed by CR.
During the past few decades, calorie restriction (CR) has been used as an experimental paradigm to increase the life span of many laboratory animals and to minimize the incidence of tumors and other pathologic conditions. At the same time, numerous biological variables or physiological characteristics associated with aging have been modified by the imposition of CR in rodents.1-5 It has been suggested that an age-related increase in oxidative damage during the normal aging process is attenuated by CR in rodents.6 These kinds of changes were also observed in nonhuman primates, and several biomarkers of aging were modified in response to CR.7
The aging of human skin consists of the phenomenon of intrinsic aging superimposed with environmental insults, and it is often difficult to separate the 2 mechanisms.8 This has prompted biological studies of aging mammalian skin from colony-raised animals, mainly to separate environmental influences from the mechanism of chronological aging.9-12 However, it is not known whether dietary restriction imposed in such age-graded colony-raised animals can modify the normal aging process of the skin, especially in terms of its microanatomic response. The present report describes the histological changes of the aging skin in Fischer 344 rats undergoing ad libitum feeding (AL group) and parallel age-matched cohorts raised with CR.
We purchased Fischer 344 male rats from the National Institute on Aging, Bethesda, Md, and the following 3 age groups from the AL and CR colonies were used in this study: 4 months (young adult), 12 months (adult), and 24 months or older (old). We used a total of 36 animals, with 6 animals in each age group. We followed institutional guidelines regarding humane use and handling for animal experiments, and animals were housed in the animal care facilities of the Biological Resources Laboratory of the University of Illinois at Chicago for 1 week.
At autopsy, ventral skin samples were excised and immersed in Bouin-Hollande fixative for 48 hours before further processing. Samples were paraffin embedded, and 5-μm sections were stained with a sequence of hematoxylin-eosin and phloxine, and the Verhoeff–van Gieson staining technique was used to distinguish collagen and elastic fibers. Glycosaminoglycans (GAGs) underwent the Mowry colloidal iron staining method.
The methods of morphometric measurement were essentially similar to those used in earlier studies of skin histology.12-14 All observations were made manually in a light microscope with ×4, ×10, and ×45 objective lenses and a ×100 oil immersion lens in a totally blinded manner. The following variables were used for assessment of histological changes of the skin: (1) depth of the epidermis, dermis including fat layer, and fat layer (in micrometers); (2) epidermal nuclear density; (3) area fraction of collagen bundles, elastic fibers, capillaries, and pilosebaceous units; (4) fibroblast density; and (5) qualitative staining intensity for GAG reaction. Thickness of the epidermis (excluding stratum corneum), dermis, and adipose layer was measured using a calibrated ocular micrometer scale. We used a point-counting morphometric method to assess change in the percentage of dermal structures in the skin samples. A 10 × 10-square grid (105 × 105 μm) with 121 intersecting points was placed over the section at a magnification of ×1000. The number of intersection points falling on a given dermal element (collagen, elastic fibers, blood vessels, and pilosebaceous units) was recorded. The number of points overlying the point of interest divided by the total number of points yielded the relative percentage or area fraction of that particular dermal element. For estimating epidermal nuclear density, the same grid was projected over the basement membrane of the epidermis by aligning the base of the square parallel to the membrane. Nuclear profiles of the epidermis falling within the grid were recorded. Dermal fibroblasts within the square grid were counted, and the data were expressed as the number of cells per square millimeter of dermis. For estimation of GAG reaction, arbitrary values were assigned to estimate the staining intensity in histological sections. For all variables, we compared the group mean values of the age groups from AL and CR rats using general linear model 2-way analysis of variance to detect significant effect of age, diet, and age-diet interaction. Differences were considered to be significant if P<.05, and all data were analyzed with Excel data Analysis Tool Pac (Microsoft Corp, Redmond, Wash) and SPSS version 12.0 software (SPSS Inc, Chicago, Ill).
The descriptive statistics for the measurement and counts of several skin characteristics with group mean ± SDs of all measures for the 6 groups of rats are given in Table 1. Table 2 details statistical significance with observed power for age, diet, and age-diet interaction. Figure 1 is a composite panel of bar charts for all quantitative characteristics used in this investigation.
Age-induced increase in epidermal thickness was noted in the 3 groups of AL rats, but this trend was not obvious in CR animals (age, F = 4.96 [P<.01]; Figure 2). On the contrary, epidermal cell density showed higher values in CR animals compared with AL animals (age, F = 10.34 [P<.001]; diet, F = 12.81 [P<.001]). The thicknesses of the dermis and fat layer showed pronounced diet effects. Increasing thickness in both of these criteria related to aging was inhibited by CR (dermis, F = 9.26 [P = .005]; fat, F = 5.89 [P = .02]).
Collagen percentage showed higher increasing values with age (F = 7.72 [P = .002]) and a significant age-diet interaction effect (F = 4.49 [P = .02]). Elastic fiber fraction area (Figure 3) showed only an age effect (F = 3.3 [P = .05]). The response of both forms of fibrous connective tissue was triphasic, but the diet effect resulted in higher values, especially in young and adult animals. There was no perceptible quantitative difference in percentage of pilosebaceous units in any experimental group. The capillary profiles showed a trend of decreasing values with aging, although this trend was statistically insignificant. Calorie restriction led to higher values in all 3 age groups (age-diet interaction, F = 11.64 [P = .002]). A pronounced effect of diet and age-diet interaction was observed in the dermal fibroblast population (diet, F = 6.73 [P = .01]; age-diet interaction, F = 6.96 [P = .003]). Higher scores were observed in the fibroblast population in CR rats. The GAG reaction (Figure 4) was estimated only by subjective qualitative assessment, and somewhat lower staining intensity was noticeable in CR animals. Significant effects of age, diet, and age-diet interaction were observed in this reaction (age, F = 4.52 [P = .02]; diet, F = 11.66 [P = .002]; age-diet interaction, F = 7.38 [P = .002]).
A correlated examination of graphically represented data and statistical analysis signifies certain trends in cutaneous morphological response from chronological aging and the effects of CR on such alterations in age-matched animals. A predominant trend of increasing width was noted in the epidermis and the dermis across the age groups of AL rats, and this growth pattern is reminiscent of 2 previous studies of aging rat skin. In male Fischer 344 rats, epidermal thickness remained nearly constant from 3 to 22 months of age after an initial increase. The dermal thickness in these animals likewise remained constant from 2 to 24 months of age.10 A similar trend was also seen in Wistar rats.11 The hallmark of normal skin aging in the rat seems to be age-dependent increase in epidermal width, which is presumably maintained until old age. The aging rat epidermis thus presents a different profile than human skin, which shows many striking changes, such as flattening of the epidermal-dermal interface and loss of structural support that could also affect skin absorption.10 However, recent quantitative work involving noninvasive methods has questioned many earlier human data on normal skin aging.12
A noteworthy effect of CR was the prevention of age-graded expansion of the epidermis. This epidermal effect of CR on age-induced expansion may be related to be numerous metabolic, physiological, or behavioral changes that have been described in Fischer 344 rats.15 Curiously, despite the atrophic effect of CR on the epidermal width, epidermal nuclear density in these animals showed a rather paradoxical stimulating effect of dieting. In the rodent literature, CR has shown inhibitory as well as proliferative effects on epithelial nuclei. In female mouse tissue, CR inhibited cell proliferation.16 In Fischer 344 rats, CR reduced proliferative response of liver cells derived from young hosts, but long-term CR caused enhanced proliferation of hepatocytes in aged cells.17 Simultaneous inhibition and stimulation of cell proliferation by dietary restriction in these rats was also reported by Lu et al.18
The inhibitory effect of CR on the increase of total dermal width in AL aging rats was presumably caused by shrinkage of the underlying adipose layer. The shrinkage of the adipose layer itself in these CR rats is in keeping with reports that CR is known to reduce lipid levels in adipose tissue, liver, and whole-animal bodies.19 A fall in serum levels of different lipid components in Fischer rats after dietary restriction has been described.20 Thus, a diminution of the subdermal fat layer seems to be part of a generalized effect of CR of fat loss in many organs.
The area fraction of the dermal collagen and elastic fibers showed progressively increased values in some of the normal-aging AL rats. In the aging Wistar rat, the pattern of collagen bundle surface density was estimated by means of image analysis and showed a comparable effect of normal aging.11 In the present experiment, CR somewhat accentuated this effect on the collagen percentage area in the 2 younger groups of rats. This trend was also noted in percentage values of dermal elastic fiber, and raises the question whether CR in these rats might have a modulating effect in decelerating breakdown of fibrous connective tissue in aging animals. A normal aging effect is a decline in extractable collagen from most connective tissues due to formation of cross-links between fibrils, and low food intake retards this aging effect in many laboratory animals.19 Glycation of body proteins has been implicated in the aging process, and CR was found to reduce age-related accumulation of glycoxidation products of skin collagen in rats.21-22 An associated or a probably interrelated finding was a higher fibroblast density in CR animals. Fibroblasts are responsible for elastogenesis, and CR may be beneficial in preserving a higher activity of the cellular population as a means to preserve the integrity of dermal connective tissue. In the aging human skin, fibroblasts become quiescent with accompanying loss of collagen and interstitial matrix.23 Qualitative grading of histochemical reaction of GAGs also showed lower staining intensity in CR rats and possibly indicated a greater surface area of collagen bundles and a relative sparseness of these matrix molecules. A study of GAGs in sun-protected and sun-exposed human skin showed loss of collagen in exposed skin with copious amounts of GAGs as shown histochemically.24 The normal rat skin with a relatively thin dermis does not have massive amounts of demonstrable GAGs, and these observations need to be confirmed with immunohistochemistry.
It appears that rodent skin morphology is an effective age-sensitive variable that is influenced by CR, and these changes may reflect a plethora of physiological modifications resulting from CR imposition. The observations made in this study, however, need to be interpreted with caution, because this is only a cross-sectional study with a few representative stages of the life cycle of these animals. Because of a high cost factor, a rather limited number of these “precious” animals were used, and a high degree of interanimal variability was evident. Nevertheless, the quantitative assessment of skin structures in 2 diet groups from representative aging animals shows some interesting patterns that might have resulted from altered metabolic pathways elicited by CR. Despite retardation of some effects of chronological aging in this species (ie, shrinkage of skin layers), some beneficial effects of CR in the skin of aging animals were observed with respect to increased vascularity, increased surface density of fibrous connective tissue elements, and greater activity of the fibroblast population. These findings are worthy of verification in other strains of CR colony-raised rodents.
Correspondence: Tapan K. Bhattacharyya, PhD, Department of Otolaryngology–Head and Neck Surgery, University of Illinois at Chicago, 1855 W Taylor St, Chicago, IL 60612 (email@example.com).
Accepted for Publication: September 22, 2004.
Acknowledgment: We thank Minu Patel, PhD, of the Biostatistics Department, University of Illinois–Chicago, for his advice and suggestions during the course of the experiment.