The Symposium on ‘Mechanisms of Ageing and Longevity’ was held at the Royal Free Campus, Royal Free and University College Medical School, London, on 17 September 1999. It was organised and sponsored by the Anatomical Society of Great Britain and Northern Ireland with generous contributions from BBSRC, AgeNet and Research into Ageing.

The Symposium was organised as a result of recent discoveries in two principal areas: the genetics of longevity regulation in the nematode worm, Caenorhabditis elegans, and the cell biology of ageing in mammalian cells. Papers were presented covering the role of insulin-like signalling, free radical damage and caloric restriction in ageing of invertebrate and mammalian systems. One of the principal aims was to examine the idea that a genetically conserved signalling pathway might regulate ageing processes at both a cellular and an organismal level.

Numerous theories of ageing have been proposed, and many have been tested experimentally, particularly using nematode models such as Caenorhabditis elegans. By combining those theories of ageing that remain plausible with recent findings from studies of C. elegans life span mutants, an integrated theory of ageing has been devised. This is formed from 3 interconnected elements: the evolutionary theory of ageing, the oxidative damage theory of ageing, and a nonadaptive programmed ageing theory. This tripartite theory of ageing gives rise to a number of predictions that may be tested experimentally.

Sublethal oxidative stresses increase the proportions of human fibroblasts positive for senescence associated β-galactosidase activity and accelerate the transition in the fibroblast morphotypes characterising fibroblast ageing. Stimulation of fibroblasts with TNF-α or IL-1α transiently increases the production of reactive oxygen species (ROS) in human fibroblasts. Here we propose that repeated stimulation of WI-38 fibroblasts with TNF-α or IL-1α can generate enough ROS to accelerate the transition in the fibroblast morphotypes and increase the proportion of cells positive for senescence associated β-galactosidase activity. The involvement of ROS is suggested by experiments where the stimulation of fibroblasts with TNF-α or IL-1α are performed in the presence of N-acetylcysteine which increases the intracellular antioxidant potential. It is proposed that the decrease in the proportions of morphotypes I and II, and the increase in the proportions of morphotypes III to VI observed after successive stimulation with TNF-α or IL1-α is attributed to an increased ROS production occurring during the stimulation.

Skeletal muscle has the unique ability to increase the rate of oxygen usage during contraction. This has led several workers to suggest that by-products of this increased oxygen consumption, oxygen-derived free radicals, may be primarily responsible for exercise-induced damage to skeletal muscle. However, because of this rapidly changing redox state, skeletal muscle has developed a number of different endogenous mechanisms which adapt rapidly following a period of exercise. These include numerous structural and biochemical changes such as increased muscle activity of antioxidant enzymes and content of stress or heat shock proteins (HSPs). This adaptation is associated with protection against the potentially damaging effects of a second period of exercise. In addition, we have recently demonstrated a significant increase in free radical production during a period of nondamaging exercise, which is rapidly followed by a significant increase in the expression of antioxidant enzymes and HSPs, suggesting that a change in redox state of the muscle may act as signal for adaptation.

Age-associated neurodegenerative diseases such as Alzheimer's disease are characterised by neuronal impairment that leads to cognitive deficits. As certain affected neurons depend on trophic factors such as neurotrophins (NTs), impairment in NT function has been suggested to be a component of neuronal damage associated with such disorders. Age-related neurodegenerative diseases are also characterised by high levels of proinflammatory cytokines such as tumour necrosis factor alpha (TNFα) in the CNS. Because TNFα receptors and certain NT receptors share a high degree of homology and are capable of activating similar signalling pathways, one possibility is that altered cytokine levels may affect NT function in the aged or diseased CNS. Here we wish briefly to review the evidence suggesting a role for cytokine and NT in the onset of age-associated neurodegenerative diseases. We propose that cytokine/NT interactions may alter neuronal homeostasis, thus possibly contributing to some of the neuronal degeneration occurring during such age-associated CNS diseases.

Approximately 30% of human and mammalian populations develop cognitive impairments with ageing. Many of these impairments have been linked to dysfunction of the hippocampus, a well studied area of the medial-temporal lobe, which is involved in episodic memory and control of the hypothalamo-pituitary-adrenal stress axis and, thus, of glucocorticoid secretion. This paper reviews the growing body of studies which explore a possible relationship between lifetime exposure to glucocorticoids and hippocampal impairment. There is now strong evidence which associates hypercortisolemia in aged men with later cognitive dysfunction and this complements a wealth of rodent and other human data. We conclude with a discussion of possible pharmacological and behavioural interventions.

The last decade has witnessed a significant turn in our understanding of the mechanisms responsible for the decline of cognitive functions in aged brain. As has been demonstrated by detailed morphological reassessments, the senescence-related changes in cognition cannot be attributed to a simple decrease in the number of neurons. It is becoming clearer that a major cause of age-induced deterioration of brain capability involves much subtler changes at the level of synapses. These changes are either morphological, i.e. reduction in the number of effective synapses and/or functional alterations, i.e. changes in the efficacy of remaining synapses. Important questions are now raised regarding the mechanisms which mediate these synaptic changes. Clearly, an important candidate is calcium, the cytotoxic role of which is already firmly established. The wealth of evidence collected so far regarding the changes of Ca2+ homeostasis in aged neurons shows that the overall duration of cytoplasmic Ca2+ signals becomes longer. This is the most consistent result, demonstrated on different preparations and using different techniques. What is not yet clear is the underlying mechanism, as this result could be explained either through an increased Ca2+ influx or because of a deficit in the Ca2+ buffering/clearance systems. It is conceivable that these prolonged Ca2+ signals may exert a local excitotoxic effect, removing preferentially the most active synapses. Uncovering of the role of Ca2+ in the synaptic function of the aged brain presents an exciting challenge for all those involved in the neurobiology of the senescent CNS.

Following the discovery that the proto-oncogene protein kinase B (PKB) functions as a downstream element in signalling from phosphoinositide 3′-kinase (PI3-kinase) (Burgering & Coffer, 1995), PKB has been shown to mediate a diverse array of PI3-kinase dependent cellular responses. Most recently PKB-dependent phosphorylation of 3 members of the family of Forkhead transcription factors has been demonstrated to play a role in PI3-kinase dependent effects on transcription. This review focuses on this newly discovered function of PKB in conveying the diversity of PI3-kinase dependent cellular responses.

Research studies clearly indicate that age-related changes in cellular and tissue function are linked to decreases in the anabolic hormones, growth hormone and insulin-like growth factor (IGF)-1. Although there has been extensive research on the effects of these hormones on bone and muscle mass, their effect on cerebrovascular and brain ageing has received little attention. We have also observed that in response to moderate calorie restriction (a treatment that increases mean and maximal lifespan by 30–40%), age-related decreases in growth hormone secretion are ameliorated (despite a decline in plasma levels of IGF-1) suggesting that some of the effects of calorie restriction are mediated by modifying the regulation of the growth hormone/IGF-1 axis. Recently, we have observed that microvascular density on the surface of the brain decreases with age and that these vascular changes are ameliorated by moderate calorie restriction. Analysis of cerebral blood flow paralleled the changes in vasculature in both groups. Administration of growth hormone for 28 d was also found to increase microvascular density in aged animals and further analysis indicated that the cerebral vasculature is an important paracrine source of IGF-1 for the brain. In subsequent studies, administration of GHRH (to increase endogenous release of growth hormone) or direct administration of IGF-1 was shown to reverse the age-related decline in spatial working and reference memory. Similarly, antagonism of IGF-1 action in the brains of young animals impaired both learning and reference memory. Investigation of the mechanisms of action of IGF-1 suggested that this hormone regulates age-related alterations in NMDA receptor subtypes (e.g. NMDAR2A and R2B). The beneficial role of growth hormone and IGF-1 in ameliorating vascular and brain ageing are counterbalanced by their well-recognised roles in age-related pathogenesis. Although research in this area is still evolving, our results suggest that decreases in growth hormone and IGF-1 with age have both beneficial and deleterious effects. Furthermore, part of the actions of moderate calorie restriction on tissue function and lifespan may be mediated through alterations in the growth hormone/IGF-1 axis.

The disposable soma theory suggests that longevity is determined through the setting of longevity assurance mechanisms so as to provide an optimal compromise between investments in somatic maintenance (including stress resistance) and in reproduction. A corollary is that species with low extrinsic mortality are predicted to invest relatively more effort in maintenance, resulting in slower intrinsic ageing, than species with high extrinsic mortality. We tested this prediction in a comparative study of stress resistance in primary skin fibroblasts and confirmed that cells from long-lived species are indeed more resistant to a variant of stressors. A widely studied example of within-species variation in lifespan is the rodent calorie restriction model. Food-restricted animals show elevations in a range of stress response mechanisms, and it has been suggested that this is an outcome of natural selection for life history plasticity. We have developed a theoretical model for dynamic optimisation of the allocation of effort to maintenance and reproduction in response to fluctuations in food availability. The model supports the suggestion that the response to calorie restriction may be an evolutionary adaptation, raising interesting questions about the hierarchy of genetic control of multiple stress response systems. The model identifies ecological factors likely to support such an adaptation that may be relevant in considering the likely relevance of a similar response to calorie restriction in other species. Comparative and theoretical studies support the role of somatic maintenance and stress response systems in controlling the rate of ageing.

Whilst there is an increased understanding of the cell biology of nerve regeneration, it remains unclear whether there is a direct interrelationship between vascularisation and efficacy of nerve regeneration within a nerve conduit. To establish this is important as in clinical surgery peripheral nerve conduit grafting has been widely investigated as a possible alternative to the use of nerve autografts. The aim of this study was to assess whether vascular endothelial growth factor (VEGF), a highly specific endothelial cell mitogen, can enhance vascularisation and, indirectly, axonal regeneration within a silicone nerve regeneration chamber. Chambers containing VEGF (500–700 ng/ml) in a laminin-based gel (Matrigel) were inserted into 1 cm rat sciatic nerve defects and nerve regeneration examined in relation to angiogenesis between 5 and 180 d. Longitudinal sections were stained with antibodies against endothelial cells (RECA-1), axons (neurofilament) and Schwann cells (S-100) to follow the progression of vascular and neural elements. Computerised image analysis demonstrated that the addition of VEGF significantly increased blood vessel penetration within the chamber from d 5, and by d 10 this correlated with an increase of axonal regeneration and Schwann cell migration. The pattern of increased nerve regeneration due to VEGF administration was maintained up to 180 d, when myelinated axon counts were increased by 78% compared with plain Matrigel control. Furthermore the dose-response of blood vessel regeneration to VEGF was clearly reflected in the increase of axonal regrowth and Schwann cell proliferation, indicating the close relationship between regenerating nerves and blood vessels within the chamber. Target organ reinnervation was enhanced by VEGF at 180 d as measured through the recovery of gastrocnemius muscle weights and footpad axonal terminal density, the latter showing a significant increase over controls (P < 0.05). The results demonstrate an overall relationship between increased vascularisation and enhanced nerve regeneration within an acellular conduit, and highlight the interdependence of the 2 processes.

To clarify the nature of M cells, the detailed ultrastructural characteristics and lectin-binding properties of M cells were investigated in follicle-associated epithelium (FAE) of chicken caecal tonsils. M cells presented various outlines from columnar to dome shaped. Their polymorphism was dependent on the number of harboured intraepithelial migrating cells. The lighter and larger nuclei of M cells were situated at more apical levels in the epithelial lining compared with those of neighbouring microvillous epithelial cells. The microvilli, which were significantly shorter and thicker than those of adjacent microvillous epithelial cells, were sparsely distributed or completely absent on the apical surfaces of M cells. In general, the apical cytoplasm of M cells without microvilli protruded slightly into the intestinal lumen. Numerous small vesicles were often contained in the apical cytoplasm. The numerous small invaginations of the apical and lateral cell surfaces suggested active transportation of luminal substances. No canaliculi existed in the apical cytoplasm of M cells whereas they were often detected in the neighbouring microvillous epithelial cells. A noteworthy finding was the frequent detection of multivesicular bodies in the apical cytoplasm of M cells. These multivesicular bodies suggest some degradation of ingested luminal substances during transcytoplasmic transportation. WGA and 4 other lectins strongly reacted with all epithelial cells except for M cells, this negativity suggesting a means of detecting M cells in chicken caecal tonsils. Three lectins, DSL, ConA and Jacalin, reacted weakly with the glycocalyx on M cells. The positive reactivity might allow chicken M cells to be utilised for specific antigen delivery into the mucosal immune system in some parenteral vaccinations.

The structure and vascular pattern of the human tibialis anterior tendon was investigated using injection techniques, light and transmission electron microscopy and immunohistochemistry. From the well vascularised peritenon, blood vessels penetrate the tendon tissue and anastomose with a longitudinally oriented intratendinous network. The distribution of blood vessels within the tibialis anterior tendon was not homogenous. The posterior part of the tendon had a complete vascular network that extends from the musculotendinous junction to the insertion at the first metatarsal and medial cuneiform bones. In the anterior half, the tissue was avascular in a zone with a length of 45–67 mm. This zone was covered by a single layer (∼30 μm) of oval shaped cells. Transmission electron microscopy showed that these cells have the characteristics of chondroid cells. This region was stained by Alcian blue at pH 1 which indicates a high concentration of acid glycosaminoglycans and immunohistochemical staining for chondroitin-4-sulphate, chondroitin-6-sulphate and aggrecan was positive. However, immunostaining for the typical cartilage specific type II collagen within this zone was negative. The location of the avascular zone corresponds to the region where the tibialis anterior tendon wraps around the superior and inferior retinacula which serve as fibrous pulleys. This is the region where most spontaneous ruptures of the tibialis anterior tendon occur. The presence of fibrocartilage within gliding tendons is a functional adaptation to compressive and shearing forces. In contrast to reports from the literature about the structure of gliding tendons wrapping around a bony pulley, the gliding zone of the tibialis anterior tendon has only a narrow layer of chondroid cells and proof of type II collagen is lacking.

The mechanics of mammalian intervertebral joints are complicated by the viscoelastic nature of the connective tissues joining vertebrae, and by multiple vertebral articulations and complex morphologies. Further, interspecific variation in these structures can greatly compound their functional variation between species, making comparative mechanical analyses even more difficult. Despite these sources of variation however, mammalian intervertebral joints universally exhibit a creep relaxation behaviour based on the viscoelastic nature of the soft tissue joint. We have evaluated, in 6 degrees of freedom, the mechanical signature of a novel mammalian lumbar intervertebral joint found in the Scutisorex spine, and compared it with a more typical mammalian joint in the Rattus (rat) lumbar spine. Scutisorex, the hero shrew, is an East African species of shrew with what is likely the most highly modified vertebral morphology in the entire history of mammals. Thus we decided to evaluate the mechanical behaviour of the intervertebral joint of this species, comparing it with a more representative mammal species in Rattus. We built a custom, 6 degrees of freedom, intervertebral joint transducer and a combined axial moment and load application system in order to quantify and compare the complex mechanical behaviour of these joints. Our results suggest that the Scutisorex joint is 5 times more resilient to simple axial torsion per body mass unit than Rattus, and that the complex load (combined axial compression and torsion) mechanical signature of Scutisorex is probably novel among all mammalian intervertebral joints. Under significant but physiological axial compression the Scutisorex intervertebral joint demonstrates no creep relaxation behaviour, simulating the mechanical behaviour of a rigid construct rather than a viscoelastic joint. The purpose of this rigid intervertebral joint in the ecology of Scutisorex remains unknown.

Development of desmin-positive hepatic stellate cells was studied in mice using double immunofluorescent techniques and in vitro cultures with special attention given to their cell lineages. Several studies recently reported on the presence of cells that are immunologically reactive with both antidesmin and anticytokeratin antibodies in young fetal rat livers, and suggested the possibility that these cells give rise to hepatocytes and hepatic stellate cells. At early stages of mouse liver development, stellate cells with desmin-positive filaments were scattered in the liver parenchyma. However, the stellate cells definitely differed from hepatoblasts and hepatocytes in terms of their morphology and expression of desmin and hepatoblast and hepatocyte-specific E-cadherin in the liver. Fetal hepatoblasts and hepatocytes did not react with antidesmin antibodies, nor did desmin-positive stellate cells express E-cadherin in vivo and in vitro. Thus it is likely that desmin-positive stellate cells and hepatoblasts belong to different cell lineages. In the fetal liver, the desmin-positive stellate cells surrounded blood vessels, and extended their processes to haematopoietic cells and megakaryocytes. Many, but not all, hepatoblasts and hepatocytes were observed to be associated with the stellate cells. At fetal stages, cellular processes positive for desmin in the stellate cells were also thick compared with those in the adult liver, in which desmin-positive stellate cells lay in Disse's space and were closely associated with all hepatocytes. These developmental changes in the geography of desmin-positive cells in the liver parenchyma and their morphology may be associated with their maturation and interactions with other cell types.

There are limited data detailing the pattern of age and gender-related changes to the thoracic vertebral bodies and intervertebral discs. A retrospective MR investigation, involving T1-weighted midsagittal images from 169 cases, was undertaken to examine age influences on the anterior wedge (anteroposterior height ratio or Ha/Hp), biconcavity (midposterior height ratio or Hm/Hp), and compression indices (posterior height/anteroposterior diameter or Hp/D) of the thoracic vertebral bodies. Disc degenerative changes in the annulus, nucleus, end-plate and disc margin were noted on T2-weighted sagittal images for the 169 cases, based on a 3-level grading system. A linear age-related decline in the Ha/Hp and Hm/Hp indices was noted. The Hp/D index increased during the first few decades of life, then decreased gradually thereafter. The prevalence of abnormal findings in the annuli, nuclei and disc margins increased with increasing age, particularly in the mid and lower thoracic discs. Greater disc degenerative changes were observed in males. These findings provide further insight into the nature of thoracic vertebral shape changes across the lifespan, and the typical patterns of degeneration of the thoracic intervertebral discs.

To assess the contribution of apoptosis to the mechanism of synovial joint cavitation, and to clarify morphological cellular changes during cavitation, we investigated the development of the rat knee joint by light and electron microscopy, TUNEL methods, and electrophoresis of DNA fragments. Although cavitation occurred within the interzone, which consists of 2 outer and a middle layer termed the intermediate zone, no morphological or biochemical signs of cell death, in particular apoptosis, were seen in the interzone at any embryonic stage. Microscopic and ultrastructural alterations affecting cell differentiation were clearly observed in the interzone, i.e. mesenchymal cells gradually showed elongation, cytoplasmic vacuolation and pyknosis in the intermediate zone where the elongated cells were arranged in parallel in some strata. Some of these cells were further flattened into spindle cells and the number of strata decreased to 2. The rest of the cells were incorporated secondarily into the outer layers, becoming chondroblasts. Collagen fibrils were arranged in a network structure in the outer layers, which obviously differed from the directional pattern parallel to the long axis of elongated cells in the intermediate zone. In addition, the density of collagen fibrils was higher in the outer layers than in the intermediate zone. During cavitation, the initial separation was detected between the elongated cells in the intermediate zone in paraffin sections at E16.5 and the spindle cells in epoxy sections at E18.5. The spindle cells lining the cavity, namely, the surfaces of the epiphysis and meniscus, finally became chondrocytes. The diminution of proteoglycans and collagen fibrils and the synthesis of hyaluronan in the extracellular matrix are now generally believed to be parts of the mechanism for cavitation based on the concept of ‘loss of cohesion’. The microscopic and ultrastructural alterations in the interzone seemed to reflect differences in the arrangement and density of collagen fibrils and the developmental condition of the extracellular matrix between layers. Also it did not seem likely that these alterations inhibit the synthesis of hyaluronan at the presumptive joint line because this synthesis takes place at the plasma membrane. Separation between spindle cells should therefore represent the mechanism for developmentally programmed cavitation. Reorganization of the extracellular matrix is probably necessary for the cellular metamorphoses in the interzone involved in the process of cavitation.

In the infant the region lateral to the sella turcica which is traditionally termed the ‘cavernous sinus’ is composed of 3 individual compartments. The aim of this study was to demonstrate a similar compartmentation of the adult cavernous sinus and to identify and quantify the adipose bodies which are located within the compartments. The region of the cavernous sinus in 136 adults was analysed by microdissection and histology. We demonstrate that in 66% of the cavernous sinus in adults is composed of at least 2 compartments and in 22% it is made up of 3. Assimilating the renaming in infants we termed the compartments ‘pterygopalatine’, ‘orbital’ and ‘lateral sellar’. The pterygopalatine and orbital compartments are connected with extracranial tissue spaces via the superior orbital fissure and contain characteristic adipose bodies. Exact topographic descriptions and measurements of the compartments and their adipose bodies are provided. Our study clearly defines compartmentation of the adult parasellar space in most individuals and thus changes the anatomical view of this space. The direct connection of 2 of these compartments with extracranial tissue spaces and the measurements of their adipose bodies are of interest for surgeons and neuroradiologists.

Articular hyaline cartilage takes on the contours of the subchondral bone on which it lies, but its thickness varies between joints, within a single joint and within a single articular surface. Previous studies have correlated articular cartilage thickness distribution with the degree of stress and weight bearing on joint surfaces, but few studies have considered the thickness of the calcified cartilage in relation to these parameters. Here we report a correlation between the cartilage thickness distribution and weight bearing distribution on the head of the 1st metatarsal bone, a component of one of the major weight bearing joints in the lower extremity during the gait cycle. The greatest total and uncalcified articular cartilage thickness was found on the central and lateral distal aspects of the metatarsal head, a region that receives maximal ground reactive force during the propulsive phase of the normal gait cycle. Although the thickness of the calcified cartilage was correlated with the thickness of the uncalcified cartilage, it varied to a lesser extent across the articular surface than did that of the uncalcified cartilage.

Anatomical Society of Great Britain and Ireland

Future meetings

3–5 January 2001. Royal Holloway College. Winter meeting including a symposium on ‘The evolution of developmental mechanisms’ (organised by Anthony Graham and Gillian Morriss-Kay). Speakers include J. Brockes, A. Gould, A. Graham, B. Hill, L. Holland, P. Holland, C. Kimmel, N. Le Douarin, G. Morriss-Kay, O. Pourquie, P. Sharpe, A. Simeone, K. Smith, C. Stern, S. Wilson.

11–13 July 2001. University College, Dublin. Summer meeting, including a symposium on ‘The respiratory system’.