Primary bone microanatomy records developmental aspects of life history in catarrhine primates

Abstract

A central challenge in human origins research is to understand how evolution has shaped modern human life history. As fossilized remains of our ancestors provide the only direct evidence for life history evolution, efforts to reconstruct life history in paleontological contexts have focused on hard tissues, particularly on dental development. However, among investigators of other vertebrate groups, there is a long tradition of examining primary bone microstructure to decipher growth rates and maturational timing, based on an empirical relationship between the microanatomy of primary bone and the rate at which it is deposited. We examined ontogenetic variation in primary bone microstructure at the midshaft femur of Chlorocebus aethiops, Hylobates lar, and Pan troglodytes to test whether tissue type proportions vary in accordance with predictions based on body mass growth patterns described previously. In all taxa, younger age classes were characterized by significantly higher percent areas of fibro-lamellar and/or parallel-fibered tissues, while older age classes showed significantly higher proportions of lamellar bone. In prior experimental studies, fibro-lamellar and parallel-fibered tissue types have been associated with faster depositional rates than lamellar bone. Principal components analysis revealed differences among taxa in the timing of this transition, and in the particular tissue types observed among individuals of similar dental emergence status. Among M1 and M2 age classes, higher proportions of parallel-fibered and fibro-lamellar tissues were observed in those taxa characterized by reportedly faster body mass growth rates. Further, persistence of fibro-lamellar tissue throughout DECID, M1 and M2 age classes in chimpanzees contrasts with the pattern reported previously for modern humans. Despite the necessary limitations of our cross-sectional study design and the secondary remodeling of bone in primates, large areas of primary bone remain intact and represent a valuable and independent source of information about the evolution of growth and development in the fossil record.

Introduction

The study of life history is concerned with those events and developmental processes that occur during the life cycle, and which determine the manner in which organisms allocate energy towards growth, reproduction and maintenance to optimize reproductive effort over their life span (Smith, 1992, Leigh and Blomquist, 2007). Ontogeny has figured prominently in the study of primate life histories, as offspring growth trajectories are governed by energetic trade-offs that determine changes in allocation towards the maturation of different body systems throughout life. Prior work has shown that different anatomical organs/systems (e.g., brain, body, dentition, and reproductive) may be developmentally dissociated from one another, and their growth rates and timing subject to independent selection (Watts, 1990, Leigh and Park, 1998, Bolter and Zihlman, 2003, Pereira and Leigh, 2003, Godfrey et al., 2004, Leigh and Bernstein, 2006, Dirks and Bowman, 2007). These diverse ‘modes’ of development underlie important differences in primate life history strategies (Pereira and Leigh, 2003). Shifts in the rate and timing of different body systems, namely dental versus skeletal growth, also characterize fossil hominins, suggesting that modularity may underlie life history evolution in the human lineage (Smith, 1993, Pereira and Leigh, 2003, Dean and Smith, 2009).

A major focus of our efforts to understand how evolution has shaped primate life history strategies has been on hard tissue features observable in fossils. Among investigators of other vertebrate groups, including mammals, there is a long tradition of examining the microscopic organization of primary bone tissues (i.e., those deposited during growth, as opposed to secondary remodeling) in both modern and fossil samples as a means of revealing life history information, including growth rates, maturational timing, and longevity (e.g., Enlow, 1966, Castanet et al., 1993, Ricqlès, 1993, Klevezal, 1996, Padian et al., 2001, Chinsamy-Turan, 2005, Erickson, 2005, Castanet, 2006, Köhler and Moyà-Solà, 2009, Padian, 2013, Kolb et al., 2015). The microscopic organization of bone records a history of its growth and development, and thus represents a potentially rich source of information about the various ontogenetic, environmental, local and phylogenetic factors that influence its formation and maintenance during life (Enlow, 1963, Enlow, 1966, Ricqlès, 1993, Martin et al., 1998, Currey, 2002, Legendre et al., 2013). Yet, its significance for revealing developmental aspects of life history in primates has been largely unexplored, with a few notable exceptions. The current study begins to address this shortcoming. Identifying shifts in dental development has been critical to our understanding of human life history evolution from evidence available in the fossil record (e.g., Bromage and Dean, 1985, Smith, 1992, Dean, 2010, Schwartz, 2012). However, a foundation based on the incorporation of studies of tooth and bone microanatomy in the same individuals would allow us to more fully address questions concerning developmental modularity and the evolution of other unique human life history traits, such as our slow somatic growth during childhood (Bogin and Smith, 1997, Leigh, 2001, Dean and Smith, 2009, Bromage et al., 2009b).

The cells, vascular canals, and collagen fiber matrix of bone combine to produce stereotypical patterns of microanatomical organization, often referred to as bone ‘tissue types’. While bone microanatomy varies along a continuum, investigators have historically used typological classifications to characterize diversity and facilitate comparisons among groups (e.g., Quekett, 1855, Foote, 1916, Amprino and Godina, 1947, Enlow and Brown, 1956, Smith, 1961, Enlow, 1963, Pritchard, 1972, Ricqlès, 1975, Francillon-Vieillot et al., 1990). Figure 1 depicts different tissue types that have been described previously, or are observed here, to occur in primate bone. These tissue types are recognized on the basis of multiple criteria, such as developmental origin, collagen fiber organization, and vascularization (Enlow, 1963, Ricqlès, 1969, Ricqlès, 1975, Ricqlès, 1977, Francillon-Vieillot et al., 1990, Ricqlès et al., 1991, Enlow and Hans, 1996). (See Supplementary Online Material [SOM] for a more comprehensive description of tissue type diversity as it pertains to the current study.) Bone tissue type distributions vary considerably among vertebrate taxa, individuals of different ages, and within single skeletons (Quekett, 1855, Foote, 1916, Demeter and Mátyás, 1928, Amprino and Godina, 1947, Enlow and Brown, 1956, Enlow and Brown, 1957, Enlow and Brown, 1958, Singh et al., 1974). This diversity can be largely understood as a reflection of ontogenetic, local (e.g., biomechanical), environmental (e.g., nutritional) and phylogenetic factors (Enlow, 1963, Ricqlès, 1993, Martin et al., 1998, Castanet et al., 2001, Currey, 2002, Pearson and Lieberman, 2004, Ricqlès, 2007, Gosman, 2012, Maggiano, 2012).

As bones increase in linear and cross-sectional size during growth through endochondral and intramembranous ossification, their characteristic adult morphologies are achieved through a sequence of modeling changes at the microscopic level (termed 'growth remodeling' by Enlow, 1963, Frost, 1973). Modeling involves spatially and temporally coordinated patterns of resorption and formation on different endosteal, cancellous and periosteal bone surfaces; it is distinct from the turnover process of bone remodeling, which entails the coupled activity of osteoblasts and osteoblasts at single bone sites (Enlow, 1963, Frost, 1973, Parfitt, 1983, Jee et al., 2007). Local dynamics of bone modeling during growth are subject to the interacting influence of genetic, hormonal and environmental (e.g., nutrition) factors, and requirements to produce changes in regional and whole bone structural properties that are mechanically appropriate (Enlow, 1963, Enlow, 1966, Enlow, 1976, Parfitt, 1983, Frost, 1990a, Frost, 1990b, Martin et al., 1998, Currey, 2002, Pearson and Lieberman, 2004, Jee et al., 2007, Gosman, 2012, Maggiano, 2012).

Modeling results in the sequential relocation of bone tissue areas in space as growth proceeds, serving to maintain or to achieve local adjustments in bone shape with increases in overall bone size. Cortical drift is one such modeling process involved, for example, in the development of long bone diaphyseal curvature; through coordinated patterns of bone deposition and resorption on complementary endosteal and periosteal surfaces, it results in the directional movement or ‘drift’ of the bone cortex in morphological space. Modeling may entail differential rates of osteogenesis across bone regions, and local destruction of tissues deposited during earlier growth stages; together with secondary remodeling, it determines the regional distribution of tissue types within cross-sections. As a result, bone has a highly stratified and heterogeneous appearance in cross-section, where each stratum characterizes a stage of deposition that may vary in its microanatomical organization or growth direction from adjacent strata (Enlow, 1963). Histological ‘signatures’ of modeling (e.g., cement lines indicating reversals in growth direction) were clarified by Enlow (Enlow, 1962, Enlow, 1963, Enlow, 1976, Enlow, 1982), and their consistency with features associated with formative and resorptive activity on bone surfaces observed in scanning electron microscopy have been demonstrated (Saunders, 1985), thus making it possible to reconstruct the sequential changes that occurred throughout the growth history of a given cross-section of bone for the purposes of revealing information about developmental aspects of life history.

The heterogeneous microscopic appearance of bone in cross-section preserves a record of growth rate variability during its development, which, when interpreted within the context of local bone modeling and remodeling circumstances, represents a unique source of information about individual developmental history that cannot be gleaned from macroscopic studies of the skeleton alone. Amprino (1947) was the first to propose a relationship between the microscopic organization of primary bone tissues, particularly tissues of periosteal origin, and the rates at which they are deposited. In a series of comparative studies, including examinations of human bone, he observed that primary bone tissue types show considerable variation during ontogeny within species; deposition of loosely organized woven bone characterizes periods of rapid growth, while a transition to more highly ordered lamellar bone characterizes periods of slow growth (Amprino and Godina, 1947). Further, he observed that animals characterized by different rates and patterns of somatic growth also differ substantially in the timing and succession of primary bone tissue types deposited in their limb skeletons throughout ontogeny.

Amprino's hypothesis has since been confirmed in examinations of a wide range of vertebrates (e.g., Enlow and Brown, 1956, Enlow and Brown, 1957, Enlow and Brown, 1958, Enlow, 1962, Ricqlès et al., 1991). Vital fluorochrome labeling studies have shown that non-vascular lamellar bone is characterized by the lowest rates of periosteal deposition, typically less than 0.3 micron/day in most vertebrates studied, including many mammals (see Ricqlès et al., 1991 and references therein). Mammalian lamellar bone growth rates vary at this slow end of the continuum, depending upon surface topography, osteoblast recruitment and life history (Bromage et al., 2009b, Bromage et al., 2011). While there is some overlap, non-vascular parallel-fibered bone may be deposited at slightly faster rates, reportedly 0.10–0.5 microns/day, while depositional rates of woven bone are most commonly between 0.3 and 5.0 microns/day (Ricqlès et al., 1991, Castanet et al., 2004). Bone tissues of endosteal and secondary origin appear to be deposited at faster rates than periosteal tissues having the same collagen fiber matrix and vascular organization (Yen and Shaw, 1977, Yen et al., 1978, Buffrénil and Pascal, 1984, Ricqlès et al., 1991, Balena et al., 1992, Margerie et al., 2002, Parfitt, 2002). Further, the blood supply of primary bone has also been shown to vary in accordance with depositional rate, as more highly vascularized tissues are associated with faster deposition than less vascularized tissues having a similar collagen fiber organization (but see Starck and Chinsamy, 2002). In a series of systematic studies of extant avians by Castanet and colleagues (Castanet et al., 1996, Castanet et al., 2000, Margerie et al., 2002, Margerie et al., 2004), vascular fibro-lamellar tissue types were the most rapidly deposited primary bone tissues, characterized by rates of osteogenesis that were typically greater than 5–10 μm/day. These findings are in agreement with studies of mammalian bone, including primates (Newell-Morris and Sirianni, 1982, Buffrénil and Pascal, 1984, Ricqlès et al., 1991, Castanet et al., 2004).

Primates have been included in a number of broadly comparative studies of vertebrate bone histology to date (Quekett, 1855, Foote, 1916, Demeter and Mátyás, 1928, Amprino and Godina, 1947, Enlow and Brown, 1958, Jowsey, 1966, Jowsey, 1968, Singh et al., 1974, Brits et al., 2014), which demonstrated considerable ontogenetic and taxonomic variability in both the microscopic organization of primary bone tissues and the extent and distribution of secondary remodeling. While these studies have led to an appreciation of general trends and the range of variation observed among species, they have tended to utilize bone samples collected opportunistically from a variety of anatomical locations, animals frequently of unknown sex and/or age, and taxa represented by only a handful of specimens (see Enlow, 1966, Ricqlès, 1993). More recent systematic examinations of long bone microanatomy in nonhuman primates have tended to focus on secondary remodeling (e.g., Schaffler and Burr, 1984, Burr, 1992, Paine and Godfrey, 1997, Mulhern and Ubelaker, 2003, Havill, 2004, Skedros et al., 2011, Streeter, 2012) and the utility of bone histology for distinguishing human from nonhuman bone or in archaeological contexts (Hillier and Bell, 2007, Maggiano et al., 2011, Mulhern and Ubelaker, 2012). The value of primary bone microstructure for addressing questions of life history significance has been rarely explored in primates (Enlow, 1966), with a few notable exceptions.

Two investigations have employed fluorescent vital labeling techniques to investigate bone growth in nonhuman primates, and demonstrated significant changes in the microanatomical structure of periosteal bone deposits coincident with changes in somatic and bone growth rates during early ontogeny. Newell-Morris and Sirianni (1982) observed variation in the collagen fiber matrix and a decline in primary osteon number at the midshaft humerus of pig-tailed macaques (Macaca nemestrina), which was associated with a decline in periosteal bone depositional rates during the late fetal and early neonatal period. They reported mean periosteal depositional rates for vascularized woven bone (equivalent to fibrolamellar complex here) at 9.8 microns/day during the fetal period, vascular parallel-fibered bone at approximately 4 microns/day in 11-week neonates, and vascular lamellar bone at approximately 1.5 microns/day in 60-week animals. In a study of captive gray mouse lemurs (Microcebus murinus), Castanet and colleagues (2004) reported a dramatic decrease in bone depositional rates at the midshaft tibia as animals approached the end of their first growing season in other body measurements; this slow-down in osteogenesis was associated histologically with an abrupt transition from deposition of parallel fibered bone containing numerous vascular canals (at rates of 3–8 microns/day) to avascular parallel-fibered bone (rates as low as 0.1 micron/day). Other studies of existing skeletal samples (not subject to vital labeling) have also demonstrated correspondence between primary bone microanatomy and somatic growth patterns in primates, including humans (Pfeiffer, 2006, Warshaw, 2008, Goldman et al., 2009, Cambra-Moo et al., 2014).

These studies have demonstrated the potential of primary bone microanatomy for revealing important information about developmental aspects of life history in primates, including variability in growth rates over the course of ontogeny. However, we still lack basic information concerning quantitative variability in primary bone microanatomy across many nonhuman primate species, throughout postnatal development and adulthood, and across different skeletal sites. The current study begins to address these shortcomings. If the utility of primary bone microanatomy for interpreting variation in somatic growth patterns can be demonstrated in a sample of Old World anthropoid primates, including apes, this can contribute towards an important comparative framework and independent line of inquiry into the evolution of human life history from the fossil record.