THE PENNSYLVANIA STATE UNIVERSITY SCHREYER HONORS COLLEGE

DEPARTMENT OF ANTHROPOLOGY

VARIATION IN PISIFORM MORPHOLOGY

LIA MICHELLE GAVAZZI FALL 2017

A thesis submitted in partial fulfillment of the requirements for a baccalaureate degree in Biological Anthropology with honors in Biological Anthropology

Reviewed and approved* by the following:

Timothy Ryan Associate Professor of Anthropology Thesis Supervisor and Honors Adviser

Kenneth Hirth Professor of Anthropology Faculty Reader

* Signatures are on file in the Schreyer Honors College.

ii

ABSTRACT

The pisiform is unique among carpal bones because it is the only one to form a secondary center of ossification and growth plate, a configuration that is present across nearly all mammalian lineages. The human pisiform has undergone major morphological changes including the lost of its growth plate and an ossification center. What is typically a rod-shaped bone in many mammalian species is more akin to a pea shape in our species. This drastic change in development and morphology has a number of implications for humans, however the functional consequences of pisiform growth plate loss are still not understood.

The pisiform is severely underrepresented in most skeletal literature, despite its relative importance. Unique human pisiform morphology is often correlated to locomotor behaviors or tool use, although direct associations between carpal morphology and specific behaviors have yet to be established. To understand the developmental history of the pisiform, it is imperative to look beyond the scope of mammals. Investigation of reptilian and amphibious species reveal a complicated and non-linear developmental history of the pisiform.

Evolutionary research frequently relies on a diverse number of species to act as proxy for the subject of interest. In the context of development, it is beneficial to compare species across multiple taxa with similar skeletal morphology to identify similarities or differences in development trajectories. In a remarkable case of parallel evolution, the family Xenarthra, comprised of sloths, armadillos, and anteaters contains several species who share morphological similarities with extant hominoids. This includes a reduced pisiform in each species of extant sloths. Our research indicates that the giant anteater pisiform develops along a typical mammal trajectory with the secondary center of ossification and a growth plate. The two-toed and three- toed sloth pisiform does not indicate any secondary center in ontogeny, leading us to conclude that the growth plate has additionally been lost in these species.

iii

TABLE OF CONTENTS

List of Figures ...... iv

List of Tables ...... vi

Chapter 1 A review of bone biology and growth plate formation ...... 1

Bone and growth plate development ...... 2 Wrist Anatomy and Pisiform Morphology ...... 8 Citations ...... 11

Chapter 2 Tetrapod Pisiform Ossification Patterns ...... 14

The fin-to-limb transition ...... 14 Methods and Results ...... 15 Amphibians ...... 17 Reptiles ...... 19 Mammals ...... 23 Conclusion and Discussion ...... 24 Citations: ...... 26

Chapter 3 Xenarthran Pisiform Morphology and Comparison with Hominoid Species ...... 28

Lorisine and Xenarthran Evolutionary History ...... 29 Lorisine Evolutionary History ...... 29 Two-Toed and Three-Toed Sloth Evolutionary History ...... 33 Hominoid evolution ...... 33 Methods ...... 38 Lorisoidea...... 39 Galago ...... 39 Lorises ...... 40 Xenarthan Development ...... 43 Giant Anteater ...... 43 Sloths ...... 44 Discussion ...... 46 Citations ...... 47

iv

LIST OF FIGURES

Figure 1-1. Pictoral representation of stylopod (orange), zeugopod (green), and autopod (blue). The autopod is further divided into the mesopodium (the carpals/tarsals) and the acropodium (metacarpals/metatarsals and phalanges). If this were a derived tetrapod forelimb, the orange stylopod would represent the humerus, the green zeugopod would be the radius and ulna, and the blue autopod represents the carpal bones, metacarpals, and phalanges...... 4

Figure 1-2. A µCT image of a mouse wrist (left) demonstrating an elongate pisiform and a histological image of growth plate (right). The pisiform is highlighted in blue in the µCT image. In the histological image the progression from cartilage to bone occurs from the top of this image towards the bottom...... 7

Figure 2-1 µCT image of newt forelimb generate using Avizo 8.0.1. This scan demonstrates the lack of secondary ossification centers present within the species and furthermore the lack of a pisiform ...... 18

Figure 2-2. µCT and histological imaging of the 15.25 inches long alligator. Note the lack of epiphyseal development in the µCT image. In this histological image of the alligator pisiform, Safranin O stains cartilages red and Fast Green stains bone and other tissues green. The histological image indicate cartilage with directional growth...... 20

Figure 2-3. µCT image of 48-inch alligator forelimb, indicating a epiphyseal line on the long bone and a lack thereof on the pisiform, further supporting our findings from the 15.25 inch alligator. Red arrow indicates pisiform, blue arrow indicates epiphyseal line...... 20

Figure 2-4. µCT and histological analysis of the fence lizard. Staining utilized a modified Periodic Acid Schiff protocol. Note the distinct epiphyseal lines on the µCT image of this juvenile in contrast with the distinct lack of an epiphyseal line within the pisiform. The red arrow indicates the pisiform and the blue arrow indicates an epiphyseal line, here highlighted on the ulna...... 22

Figure 2-5. µCT scan of anoles lizard and Periodic Acid Schiff staining of histological section. Like the fence lizard, there are distinct growth plate lines along the long bones of the anoles lizard autopodium and a lack of a secondary ossification center in the pisiform...... 22

Figure 2-6. Image data from mouse demonstrating an elongate, rod-like pisiform with a clear secondary center of ossification, as well as a histologically clear growth plate. Compare with images from chapter 1, also featuring a mouse autopod and pisiform growth plate, for review of cellular zones of growth plate and their correlation to this image. Blue arrow indicates the epiphyseal line on the mouse pisiform...... 24

Figure 3-1. Image data from the three galago specimens. Blue arrows indicate the pisiform in each specimen. Across several differently aged specimens we see a typical monkey-like pisiform which is elongate. Image sequence C features the

v

zeugopod/autpod boundary along with the manus. Note the clear epiphyseal lines on the radius and ulna. The pisiform is also shown in isolate...... 41

Figure 3-2. Agwantibo µCT scan. There are clear epiphyseal lines, shown by the blue arrow, and a short pisiform, indicated in red. The pisiform is nodular in this agwantibo specimen. To confirm the presence or absence of a pisiform growth plate will require more specimens...... 42

Figure 3-3. Potto µCT scan. We believe that the pisiform has been displaced during the specimen drying process and thus does not accurately represent location in-vivo. Additional specimens will be needed to confirm accurate pisiform location. Likely pisiform indicated with red arrow...... 43

Figure 3-4 Giant anteater µCT scan. Pisiform indicated with red arrow. Anteater specimen demonstrates clear secondary center of ossification as well as an epiphyseal line. This morphology is more similar to what we would expect from a quadruped and is similar to the chimpanzee pisiform...... 44

Figure 3-5. Two-toed sloth µCT scan. We believe that the pisiform is not present in this data and may have been disarticulated from the rest of the carpals during preservation or storage...... 45

Figure 3-6. Three toed sloth µCT scans. Pisiform indicated with red arrow. The pisiform is nodular. It appears that there is a subchondral surface on the pisiform, which may indicate the presence of a growth plate...... 46

vi

LIST OF TABLES

Table 3-1. List of specimens borrowed from AMNH mammology collections along with the scientific and common names for each specimen...... 38

1

Chapter 1

A review of bone biology and growth plate formation

Variation in body morphology, or body shape, is often a result of variation in gene expression during development, and it is through these subtle changes over half a billion years that “life, […] having been originally breathed into a few forms or into one, […] from so simple a beginning endless forms most beautiful and most wonderful have evolved” (Darwin 1859).

Recent advancements in genetic technologies led to the formation of the ‘new science’ of evolutionary and developmental research, henceforth referred to as evo-devo. The principle goal of evo-devo is elucidating the genetic pathways and developmental processes that inform the bauplan, or body plan, of species across diverse taxa and lineages. Evolution of anatomical diversity is a direct response of changing gene regulation rather than changes in protein sequences or coding errors (Britten and Davidson 1971). Sean Carroll stated this most succinctly, “A rapidly growing number of genetic analyses of trait divergence have demonstrated evolutionary changes at loci for which functional coding changes have been ruled out and functional cis-regulatory sequence (regions of non-coding DNA responsible for the regulation of nearby genes) changes have been implicated” (Carroll 2008). One of the systems most thoroughly studied by researchers is the skeletal system, given that evidence of the form of skeletal tissue can remain for millions of years after the death of an individual, providing an insight into morphological change that is not accessible via other body systems. This chapter provides a review of bone and growth plate development, as well as crucial background information about wrist anatomy for subsequent chapters.

2

Evo-devo studies provide us with a novel way to approach fundamental questions about evolutionary processes and their transitions throughout time and space. By combining the classic and modern morphometric methodologies of skeletal and fossorial paleontological research with the molecular, genetic, and experimental approaches of development, we are able to quantify changes in the bauplan in addition to understanding the regulatory genetic processes driving these alterations.

Bone and growth plate development

The developing embryo is comprised of three distinct layers, endoderm, ectoderm, and mesoderm, from which arise all major organ systems and tissue within a body. Embryonic mesoderm further differentiates into paraxial, intermediate, and lateral plate mesoderm. These form mesenchyme, which is comprised of ground substance with a few cells interspersed; it is able to easily migrate through the body and is formed by cells that are loosely associated with each other and lack polarity, meaning that they have yet to be assigned a developmental role (Hall and Miyake 1992). Each subdivision is responsible for the development of different systems within the body. Of key importance to this paper, a subset of mesenchymal cells from the lateral plate mesoderm define the limb field and begin proliferation to form the limb bud. In the overlapping ectoderm, fibroblast growth factors initiate further development and induce the formation of the apical ectodermal ridge (AER). Underlying the AER is a population of mitotic mesodermal cells called the progress zone, whose proliferation is responsible for the outgrowth of the limb. A second distinct population of mesenchymal cells called the Zone of Polarizing

Activity (ZPA) forms at the posterior margin of the limb bud. Sonice hedgehog (Shh) is secreted by cells within the ZPA and is vital for the development of the anterior/posterior axis in the

3 developing limbs as well as for the formation of digits on both the hands and the feet. Sonic hedgehog (Shh), is a member of a signaling pathway family with both Indian hedgehog and

Desert hedgehog. Proximo-distal limb development proceeds by a feedback loop involving the

AER, ZPA and progress zone (Tonegawa et al. 1997).

During outgrowth mesenchymal cells condense and differentiate to form the skeletal precursors that will be found in the respective proximal, intermediate and distal limb segments, the stylopod, zeugopod, and autopod. These terms refer to three major segmentations of the forelimb and hind limb shared among tetrapod lineages. The stylopod corresponds to the humerus/femur, the zeugopod is the radius/ulna or tibia/fibula, and the autopod comprises the carpals/tarsals, metacarpals/metatarsals, and phalanges (see figure 1). These terms are used in evolutionary research to discuss developmental relationships across multiple taxa and lineages where direct one-to-one comparison are not necessarily possible. This is especially prominent in the discussion of the fin-to-limb transition and assessment of basal tetrapod, where skeletal limb elements are less clearly defined relative to more derived species.

Numerous genes are involved in patterning the limb. As discussed above, Shh is crucial for patterning the anterior/posterior axis of the limb as well as being responsible for the differentiation of proximal to distal limb segments (Wagner et al. 2001). Hox genes also regulate initial limb patterning through mesenchymal expression (Morgan and Tabin, 1994). Hox genes are implicated in conferring embryonic bauplan segmentation by managing the genes involved in segment formation regulation (Krumlauf 1994).

4

Figure 1-1. Pictoral representation of stylopod (orange), zeugopod (green), and autopod (blue). The autopod is further divided into the mesopodium (the carpals/tarsals) and the acropodium (metacarpals/metatarsals and phalanges). If this were a derived tetrapod forelimb, the orange stylopod would represent the humerus, the green zeugopod would be the radius and ulna, and the blue autopod represents the carpal bones, metacarpals, and phalanges.

It is important to keep in mind that the skeleton is comprised of bone and cartilaginous tissue, both complex in their own right and more so in their interaction. Osteoblasts, which form bone, and chondrocytes, which form cartilage, both derive from the same mesenchyme origin

(Ducy et al. 1997). Additionally, the genes regulating skeletal formation are not the same genetic pathways that inform cellular differentiation, giving researchers the complication of studying skeletal patterning, molecular control over cellular differentiation, and the interactions between the two (Karsenty and Wagner 2002).

Cartilage is formed from the differentiation of condensed mesenchyme tissue, which are undifferentiated cells that derive from the mesoderm. Once the sites of skeletal formation are specified, the future bones initiate as mesenchymal condensations. They are often surrounded and encased in large amounts of extracellular matrix, comprised of collagen and ground substance.

Some research has indicated that Bone Morphogenetic Proteins (BMPs) are critical for the differentiation of mesoderm into mesenchymal cells and thus are essential in the induction and

5 formation of bone and cartilage, but they are also involved in the signaling of tissue differentiation (Lee et al. 2012, Winnier et al. 1995).

Bone tissue initially forms through either intramembranous or endochondral ossification.

Intramembranous ossification occurs through direct osteoblast differentiation within a mesenchymal, or undifferentiated cell, tissue (Karsenty and Wagner 2002). Bone develops within the mesenchymal cells via type I collagen matrix; this particular type of collagen is the most abundant in the human body. Mesenchymal cells first cluster together and then differentiate directly into osteoblastic cells. The developing bone forms into bony struts, called spicules.

Newly forming bone becomes vascularized and these spicules form into structures typical of compact bone. Bone spicules form to create a network called woven bone, which will become lamellar bone and directly from this structure forms a fully ossified tissue; this method of ossification is predominantly present in the skull, although endochondral ossification is present in the cranial base as well (McBratney-Owen et al. 2008).

Endochondral ossification occurs in most of the postcranial skeleton and derives from mesenchymal cells. These cells then aggregate and differentiate to form chondrocytes, which form the anlagen, or template, which, is surrounded by periosteum. The mesenchymal cells become chondrocytes and secrete type II collagen and undergo proliferation. Type II collagen is the driver behind hyaline and articular cartilage production, both of which are imperative for intra-skeletal and synovial joint lubrication and function. Periosteal cells then differentiate into osteocytes to form a bone collar where the shaft of the long bone will occur. The chondrocyte then differentiate into hypertrophic chondrocytes, which enlarge, synthesizes type X collagen, and calcify the extracellular matrix providing a matrix for invading osteocytes, the main bone maintenance cell (Kronenburg 2003). Blood vessels invade from the bone collar and provide the necessary cells for the formation of spongy bone within the shaft. This is the primary center of ossification. As ossification continues, the medullary cavity forms within the diaphysis. Around

6 this time, the secondary center of ossification begins to form in the epiphyses of the long bone following the same differentiation steps. Blood vessels invade once again and with the establishment of the epiphysis comes the development and differentiation of compact and trabecular bone. Hyaline cartilage only remains for the growth plate and the articular surface.

These two centers of ossification form an epiphyseal growth plate, which allows the bone to grow rapidly until skeletal maturity is reached.

Wnt signaling pathways are responsible for body axis patterning, cell fate, proliferation, and migration. Along with fibroblast growth factors, Wnts control embryological cell growth and development and they are activated by binding to Frizzled. Once Wnt and Frizzled are bound, the

Frizzled receptors transduce signals via the β-Catenin pathway (Tamamura et al. 2004). Frizzled serves as a receptor protein for Wnt signaling pathways and thus is another crucial component of cell proliferation. The Wnt/β-Catenin signaling pathway is essential for limb development, and it has been found that the pathway serves multiple roles in the development and proliferation of the limb skeleton, most notably in the differentiation from mesenchymal cells into osteoblastic cells

(Baron and Rawadi 2007, Glass et al. 2005). It is difficult to isolate the function of a specific Wnt in osteoblasts due to their generalized somatic effects (Nusse et al. 2008). Wnt-7a, and potentially

Wnt-3a, are expressed in the patterning of the dorsal/ventral axis in early limb development.

Growth plates are cartilaginous structures, comprised of hyaline cartilage that control nearly all post-natal longitudinal skeletal growth. A growth plate is comprised of five major zones. The zone of resting cartilage sits at the epiphyseal end of the growth plate, providing the initial cell population for the growth plate. The chondrocytes then undergo mitosis under the influence of growth hormone in the columnar zone; these cells flatten, stack on top of one another and undergo proliferation. From here, the chondrocytes cease mitotic division and begin to expand in size in the aptly named zone of hypertrophy. Next comes the zone of calcification, where these chondrocytes undergo apoptosis, also known as cell death. It is here that

7 cartilaginous matrix also begins to calcify. The final zone, the zone of ossification, is marked by the degredation of osteoclasts with the calcified cartilage. While the cartilage is broken down, osteoblasts begin to lay down bone matrix (Kronenburg 2003).

Figure 1-2. A µCT image of a mouse wrist (left) demonstrating an elongate pisiform and a histological image of growth plate (right). The pisiform is highlighted in blue in the µCT image. In the histological image the progression from cartilage to bone occurs from the top of this image towards the bottom.

Formation of growth plates during endochondral ossification is a complex event that requires the interaction of multiple genetic pathways (see Delgado and Torres 2017 for a review of major interactions in limb development). BMPs are important growth and developmental factors during ossification. These proteins are highly expressed in proliferative chondrocytes.

BMPs also mediate the expression of other proteins, such as Indian hedgehog in pre-hypertrophic chondrocytes, which increases the rate of chondrocyte proliferation and overall organismal growth. Indian hedgehog (Ihh) is a regulator of bone development, chondrocyte proliferation, and osteoblast differentiation (St. Jacques et al. 1999). Ihh upregulates parathyroid hormone-related protein (PTHrP) through a series of downstream events. PTHrP is expressed and diffused through the growth plate, which creates a critical feedback loop between Ihh and PTHrP, which controls

8 fetal growth plate development (Kronenberg 2003). Additionally, the Ihh/PTHrP feedback loop is necessary for the maintenance of growth plates throughout postnatal development (Karimian et al. 2012, Vortkamp et al. 1996).

Wnt-1, Wnt-4, and Wnt-7a antagonize Wnt-5a and -5b. Wnt-5a and Wnt-5b stimulate the differentiation of mesenchymal cells into chondrocytes while the three genes mentioned just prior act as inhibitors of cell differentiation (Tamamura et al. 2004). Activation of β-Catenin likely promotes the differentiation of mesenchymal cells into osteoblasts or chondrocytes, as inhibition of β-Catenin in turn obstructs this differentiation (Glass et al. 2005, Day et al. 2005). Tamamura et al. 2004 also found that the Wnt/β-Catenin signaling pathway may also activate catabolic reaction within the cartilage matrix

Wrist Anatomy and Pisiform Morphology

The wrist joint is formed by the short carpal bones of the hand, the radius and ulna of the forearm, and the associated soft tissue structures. These tissues create the most complex joint in the human body, capable of moving in three planes (Palmar et al. 1985). In almost all tetrapod species, the triquetral (or triquetrum) and pisiform directly articulate with the ulna to form the medial portion of the wrist joint (Westoll 1943). There are a number of species, namely extant hominoids, extant lorisine primates, and some xenarthrans, in which the pisiform has shifted distally, no longer articulating with the ulna to leave its sole attachment with the triquetral (Lewis et al. 1969, Lewis 1970, Cartmill and Milton 1977, Mendel 1979). This ulnar withdrawal and relocation of the pisiform has traditionally been associated with a suite of suspensory locomotor behaviors (Lewis 1972), although the relationship between these behavioral postures and specific morphological configuration is not satisfactorily explained.

9

The mesopodials, or carpal and tarsal bones of the wrist and ankle form via endochondral ossification and are considered ‘short bones’, but unlike long bones, they typically do not form a secondary center of ossification and growth plate. The exception in nearly all mammalian species is the pisiform, which is a large, elongate carpal that forms a growth plate (Kjosness et al. Yet, in humans the pisiform growth plate is lost and more closely resembles a pea-shaped bone. The human pisiform likely only forms as an epiphysis, a hypothesis strengthened by the research indicating that the human pisiform ossifies at approximately the same time as epiphysis in other species (Kjosness 2017). It is still an open question what the developmental underpinning of this morphological change are, but there is solid evidence to suggest that differential Hox expression may be involved (Kjosness et al. 2014, Reno et al. 2016). Hominoids, including two of our closest relatives Pan troglodytes and Australopithecus afarensis, have elongate pisiforms that are distinct from Homo sapiens (Bush et al. 1982). While the pisiform is often referred to as a sesamoid it is the only carpal with an insertion for an extrinsic flexor (flexor carpal ulnaris) and the pisiform of most species articulates with the ulna and triquetral preventing sliding mobility typical of sesamoids (Haines 1969). The pisiform also serves as an attachment site for the abductor digiti minimi muscle. Further distinguishing it from a sesamoid is the existence of the growth plate and secondary center of ossification as previously described. While a set of eight carpals is standard for our species, there are a vast number of variants noted within other tetrapod lineages. Subtle shifts in morphology can have a dramatic impact on locomotor behavior, and the association between morphology and locomotor behavior has been discussed elsewhere (Doran

1993, Thewissen et al. 1994, Orr 2017).

The large tree shrew, Tupaia tana, possesses eight carpals, including a scapholunate (a fusion between the scaphoid and lunate bones) and an os centrale (Stafford and Thorington 1998).

As a member of the order Scandentia, this species can provide valuable information for anthropological research since they are closely related to primates (Fleagle 2013). Canis

10 familiaris, the domesticated dog, has seven carpal bones including fusion between the scaphoid and lunate (this bone is called the intermedioradial in anatomical literature on Canis) (Evans and

De Lahunta 2013). This appears to be consistent across all breeds and morphological variance in dogs. The manatee displays a suite of unique morphological variations related to life as an aquatic mammal including the loss of hind limbs and a cervical vertebrae count of six instead of the nearly mammalian constant of seven (Buchholtz et al. 2007). The Floridian West Indian manatee,

Trichecus manatus latirostris, is noted to have five carpals, including a fused radiale (another name for scaphoid) and intermedium (another name for lunate) (Quiring and Harlan 1953). A pisiform is not found in manatees or the closely related dugong (Howell, 1930, Flower and

Gadow, 1885).

Until recently, the developmental mechanisms for the loss of the pisiform growth plate in humans were not well understood and the functional consequences of this change are still debated. Skeletal morphological traits are frequently hypothesized to correlate with functional adaptations and thus observations of hand and wrist bone shape are frequently correlated to locomotor and postural behaviors. Excluding obligate bipeds like humans, most tetrapod species utilize the hand and wrist to navigate environmental substrates in a mechanically effective manner. The ancient gait patterns of walking and running are present in nearly all tetrapod clades, likely correlating to the first terrestrial quadrupeds in the Permian (299 – 251 MYA) (Reilly et al.

2006). The genetic mechanisms that control bone growth are the primary regulators of skeletal morphology, though locomotor and postural behaviors do modify bone structure to some degree

(Pearson et al. 2004,). The exact relationship between internal developmental factors and external environmental factors on bone growth and morphology is not clearly understood and thus represents a critical area in need of research (Young 2017).

11 Citations

Baron R and Rawadi G. Targeting the Wnt/β-Catenin Pathway to Regulate Bone Formation in the Adult Skeleton. Endocrinology 148(6):2635–2643. 2007. Britten, R. J., & Davidson, E. H. (1971). Repetitive and non-repetitive DNA sequences and a speculation on the origins of evolutionary novelty. The Quarterly review of biology, 46(2), 111-138. Buchholtz, E. A., Booth, A. C., & Webbink, K. E. (2007). Vertebral anatomy in the Florida manatee, Trichechus manatus latirostris: a developmental and evolutionary analysis. The Anatomical Record, 290(6), 624-637 Bush, M. E., Lovejoy, C. O., Johanson, D. C., & Coppens, Y. (1982). Hominid carpal, metacarpal, and phalangeal bones recovered from the Hadar Formation: 1974–1977 collections. American Journal of Physical Anthropology, 57(4), 651-677. Carroll, S. B. (2008). Evo-devo and an expanding evolutionary synthesis: a genetic theory of morphological evolution. Cell, 134(1), 25-36. Cartmill, M., & Milton, K. (1977). The lorisiform wrist joint and the evolution of “brachiating” adaptations in the Hominoidea. American Journal of Physical Anthropology, 47(2), 249- 272. Day, T. F., Guo, X., Garrett-Beal, L., & Yang, Y. (2005). Wnt/β-catenin signaling in mesenchymal progenitors controls osteoblast and chondrocyte differentiation during vertebrate skeletogenesis. Developmental cell, 8(5), 739-750. Darwin, C. (1859). On the Origin of Species by natural selection. Delgado, I., & Torres, M. (2017). Coordination of limb development by crosstalk among axial patterning pathways. Developmental Biology. Doran, D. M. (1993). Comparative locomotor behavior of chimpanzees and bonobos: the influence of morphology on locomotion. American Journal of Physical Anthropology, 91(1), 83-98. Ducy, P., Zhang, R., Geoffroy, V., Ridall, A. L., & Karsenty, G. (1997). Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. cell, 89(5), 747-754. Evans, H. E., & De Lahunta, A. (2013). Miller's Anatomy of the Dog-E-Book. Elsevier Health Sciences. Glass, D. A., Bialek, P., Ahn, J. D., Starbuck, M., Patel, M. S., Clevers, H., ... & Karsenty, G. (2005). Canonical Wnt signaling in differentiated osteoblasts controls osteoclast differentiation. Developmental cell, 8(5), 751-764. Fleagle, J. G. (2013). Primate adaptation and evolution. Academic Press. Flower, W. H. & Gadow, H. 1885. An Introduction to the Osteology of the Mammalia. Macmillian and Co. Haines, R. W. (1942). The evolution of epiphyses and of endochondral bone. Biological Reviews, 17(4), 267-292. Haines, R. W. (1969). Epiphyses and sesamoids. Biology of the Reptilia, 81-115. Hall, B. K., & Miyake, T. (1992). The membranous skeleton: the role of cell condensations in vertebrate skeletogenesis. Anatomy and embryology, 186(2), 107-124

12 Howell, A. B. 1930. Aquatic mammals: their adaptations to life in the water, Baltimore Md;Springfield, Ill;, C. C. Thomas. Karsenty, G., & Wagner, E. F. (2002). Reaching a genetic and molecular understanding of skeletal development. Developmental cell, 2(4), 389-406. Kivell, T. L., & Begun, D. R. (2009). New primate carpal bones from Rudabánya (late Miocene, Hungary): taxonomic and functional implications. Journal of human evolution, 57(6), 697-709. Kjosness, K. M. 2017. Evolutionary and Developmental Mechanisms of Pisiform Growth Plate Loss. Doctor of Philosophy, The Pennsylvania State University. Kjosness, K. M., Hines, J. E., Lovejoy, C. O., & Reno, P. L. (2014). The pisiform growth plate is lost in humans and supports a role for Hox in growth plate formation. Journal of anatomy, 225(5), 527-538. Kronenberg, H. M. (2003) Developmental regulation of the growth plate. Nature 423, 332–336. Krumlauf, R. (1994). Hox genes in vertebrate development. Cell, 78(2), 191-201. Lewis, O. J. (1969). The hominoid wrist joint. American Journal of Physical Anthropology, 30(2), 251-267. Lewis, O. J., Hamshere, R. J., & Bucknill, T. M. (1970). The anatomy of the wrist joint. Journal of anatomy, 106, 539. Lewis, O. J. 1972. Osteological features characterizing the wrists of monkeys and apes, with a reconsideration of this region in Dryopithecus (Proconsul) africanus. Am J Phys Anthropol, 36, 45-58. McBratney-Owen, B., Iseki, S., Bamforth, S. D., Olsen, B. R., & Morriss-Kay, G. M. (2008). Development and tissue origins of the mammalian cranial base. Developmental biology, 322(1), 121-132. Mendel, F. C. (1979). The wrist joint of two‐ toed sloths and its relevance to brachiating adaptations in the Hominoidea. Journal of Morphology, 162(3), 413-424. Morgan, B. A., & Tabin, C. (1994). Hox genes and growth: early and late roles in limb bud morphogenesis. Development, 1994(Supplement), 181-186. Nusse, R., Fuerer, C., Ching, W., Harnish, K., Logan, C., Zeng, A., D. Ten Berge & Kalani, Y. (2008, January). Wnt signaling and stem cell control. In Cold Spring Harbor symposia on quantitative biology(Vol. 73, pp. 59-66). Cold Spring Harbor Laboratory Press. Orr, C. M. (2017). Locomotor hand postures, carpal kinematics during wrist extension, and associated morphology in anthropoid primates. The Anatomical Record. Palmer, A. K., Werner, F. W., Murphy, D., & Glisson, R. (1985). Functional wrist motion: a biomechanical study. The Journal of hand surgery, 10(1), 39-46. Quiring, D. P., & Harlan, C. F. (1953). On the anatomy of the manatee. Journal of Mammalogy, 34(2), 192-203. Reilly, S. M., McElroy, E. J., Odum, R. A., & Hornyak, V. A. (2006). Tuataras and salamanders show that walking and running mechanics are ancient features of tetrapod locomotion. Proceedings of the Royal Society of London B: Biological Sciences, 273(1593), 1563-1568.

13 Reno, P. L., Kjosness, K. M., & Hines, J. E. (2016). The role of Hox in pisiform and calcaneus growth plate formation and the nature of the zeugopod/autopod boundary. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution, 326(5), 303-321 St. Jacques B, Hammerschmidt M, McMahon AP "Indian hedgehog signaling regulates proliferation and differentiation of chondrocytes and is essential for bone formation". Genes Dev. 13 (16): 2072–86. (August 1999). Stafford, B. J., & Thorington, R. W. (1998). Carpal development and morphology in archontan mammals. Journal of Morphology, 235(2), 135-155. Tamamura, Y., T. Otani, N. Kanatani, E. Koyama, J. Kitagaki, T. Komori, Y. Yamada, F. Costantini, S. Wakisaka, M. Pacifici, M. Iwamoto, and M. Enomoto-Iwamoto. "Developmental Regulation of Wnt/ -Catenin Signals Is Required for Growth Plate Assembly, Cartilage Integrity, and Endochondral Ossification." Journal of Biological Chemistry 280.19 (2005): 19185-9195. Thewissen, J. G., Hussain, S. T., & Arif, M. (1994). Fossil evidence for the origin of aquatic locomotion in archaeocete whales. Science-AAAS-Weekly Paper Edition-including Guide to Scientific Information, 263(5144), 210-211 Tonegawa A, Funayama N, Ueno N, Takahashi Y. "Mesodermal subdivision along the mediolateral axis in chicken controlled by different concentrations of BMP-4". Development 124(10): 1975–84. (1997) Vortkamp A, Lee K, Lanske B, Segre GV, Kronenberg HM, Tabin CJ (August 1996). "Regulation of rate of cartilage differentiation by Indian hedgehog and PTH-related protein". Science 273 (5275): 613–22 Wagner, G. P., & Chiu, C. H. (2001). The tetrapod limb: a hypothesis on its origin. Journal of Experimental Zoology Part A: Ecological Genetics and Physiology, 291(3), 226-240. Westoll, T. S. (1943). The origin of the primitive tetrapod limb. Proceedings of the Royal Society of London B: Biological Sciences, 131(865), 373-393. Young, N. M. (2017). Integrating “Evo” and “Devo”: The Limb as Model Structure. Integrative and Comparative Biology.

14

Chapter 2

Tetrapod Pisiform Ossification Patterns

The rise of the tetrapod limb, along with the rise of asymmetry and body segmentation, marks one of the most significant developments in evolution of the vertebrate bauplan during the past 500 million years (Hallgrímsson et al. 2012). While the importance of limb formation has been emphasized for years the only methodologies previously available emphasized general shape change in skeletal or fossil remains. Recent advancements in genetic technology have provided a plethora of new methodologies for analysis of major evolutionary trends (Watson

1917, Carroll 2008, Young 2017). Developmental methodologies are providing a novel perspective on broad evolutionary questions that were previously only informed by paleontological research. Major shifts in vertebrate morphology are correlated with changes in the developmental genetic regulatory systems, and all tetrapod appendages are organized by a similar gene suite (Young 2017). All tetrapod species maintain a generalized, standard pattern of gene expression and it is through differential expression of this highly regulated design that morphological diversity can arise. It is worth noting that variation in skeletal morphology tend to involve changes to distal skeletal elements, like the carpals and phalanges, more frequently than proximal elements (Shubin et al. 1997).

The fin-to-limb transition

The forelimb and hind limb provide excellent models for understanding the relationship between evolutionary and developmental processes. The deep homology of limb organization provides a basis of comparison for the remarkable locomotor diversity that is observed across

15 tetrapod lineages. Hox genes are likely a critical factor in the regulation and maintenance of serially homologous appendages (Shubin 1995, Shubin et al. 1997). The development of the tetrapod limb is linked to a temporal and spatial shift in Hox gene expression, and this likely occurred during the Devonian. It is in this period that the fossil record demonstrates a clear fin to limb transition in many prototetrapod and sarcopterygian species (Shubin et al. 1997). There is strong evidence to suggest that the closest relative to all tetrapods are the lungfish which together with Tetrapoda form the Rhipidistia clade (Takezaki and Nishihara 2017, Amemiya et al. 2013).

Tetrapod limbs first appear in the fossil record in species such as Ichthyostega during the

Devonian periods approximately 370 million years ago (Ahlberg and Milner 1994).

Paleontological evidence supports that sarcopterygian fishes and their paired fins are precursors to the tetrapod limb. Additionally, there is strong evidence to suggest that the development of the mesopodium occurred only once and belies a singular evolutionary event (Laurin 1998).

However, the) The transition from fin-to-limb occurred piecemeal over millions of years with initial autopodial development, the formation of the polydactyl limb, and then transformation of the archaic tetrapod autopodium to the pentadactyl autopod seen in most extant species today

(Coates 1996).

The pentadactyl tetrapod limb appeared approximately 350 million years ago, nearly 25 million years after the initial fin-to-limb transition (Shubin et al. 1997). The five-digit limb that is now canalized in all tetrapod populations arose out of a shift in Hox expression, including

Hoxa13 and Hoxd13 (Khergjemil et al. 2017). Digits appear in the fossil record before carpals, indicating that the acropodium evolved before the formation of the mesopodium. This hypothesis is supported by developmental and molecular research examining the expression of Hox genes in the autopod (Johanson et al. 2007). Other research (Reno et al. 2016) has suggested instead that carpal homologues may precede metacarpals and phalanges based on the expression of Hox at the zeugopod/autopod boundary. Hoxa13 and Hoxd13 are more highly expressed in the distal portion

16 of the limb and at the zeugopod/autopod boundary while Hoxa11 and Hoxd11 modify pisiform formation (Kjosness et al. 2014, Reno et al. 2016). Hox genes are expressed in a collinear pattern in the stylopod and zeugopod meaning that as the limb forms proximo-distally, the Hox gene number increases.The reversal of this pattern only occurs in the autopod – an event typically known as reverse collinearity (Reno et al. 2016, Tschopp and Duboule 2011).

Mouse modeling indicates that dosage reduction of Hox11 and Hox13 or knock-out events increase the number of digits present in an individual, thus highlighting the importance of

Hox gene regulation in the fin-to-limb transition (Sheth et al. 2012, Wagner and Chiu 2001).

Overexpression of the Hoxd13 gene in zebrafish resulted in a fin with excess cartilaginous tissue and a reduced fin fold, two morphological variants hypothesized to initiate the fin-to-limb transition (Paço and Freitas 2017).

Both gene expression and knock-out research of Hoxa11 and Hoxd11 demonstrate the importance of these genetic regions at the zeugopod/autopod boundary and their role in growth plate formation (Reno et al. 2016 Pineault et al. 2015). Reno et al. (2016) shows that Hoxa11 and

Hoxd11 genes, which are typically considered markers of the zeugopod/autopod boundary, are expressed around the mouse pisiform giving this carpal bone more developmental similarities to the zeugopod than the autopod. This type of research demonstrates the importance of a developmental approach in conjunction with a morphological analysis to best understand the evolutionary processes that have influenced species diversity.

It is the goal of this study to elucidate the ancestral tetrapod state of the pisiform and identify the presence or absence of a growth plate in this carpal utilizing a variety of reptilian and amphibian model species such as the alligator and the red eft newt. It is imperative to determine whether or not a growth plate and long-bone like ossification occur within the non-mammalian pisiform because we currently have no known baseline for comparison of pisiform morphology

17 and development. The evolutionary significance of pisiform growth plate formation or loss within certain taxa is not clearly understood, and thus it is necessary to elucidate the ancestral condition before further detailed research can be conducted. This is preliminary research to identify general morphology, histology, and evolutionary history of the tetrapod pisiform.

Methods and Results

Five species across three major linages were used in this analysis; Eastern newt

(Notophthalmus viridescens), American alligator (Alligator mississippiensis), Eastern fence lizard

(Sceloporus undulatus), Carolina anoles lizard (Anolis carolinensis), and mouse (Mus musculus).

The fence lizard, anoles lizard, and newt specimens were generously donated by Dr. Tracy

Langkilde, Professor of Biology at The Pennsylvania State University, University Park, PA.

Alligator specimens provided by Dr. Philip Reno. An FVB line mouse from the Reno mouse colony, previously analyzed by Dr. Kelsey Kjosness, was used for this analysis. The pisiform ossification patterns of the alligators, lizards, newts, and mice were compared using micto-CT data in addition to histological analysis. Specimens were scanned on a GE v|tome|x L 300 high- resolution microfocus CT system voxel size 0.015. All micro-CT data was processed and analyzed with Avizo 8.0.1. Pisiform ossification patterns were visualized in histological sections using Saffranin O/Fast Green and a modified Periodic Acid Schiff/Light Green protocol to distinguish cartilaginous components from bone tissue (Serrat 2007).

Amphibians

The amphibian sample was limited to one species, Notophthalmus viridescens, a common

Eastern newt in the family Salamandridae, which falls under the order Urodela. This; these

18 species functioned as an outgroup for this project. µCT data of the newt sample revealed a lack of secondary ossification centers in any long bones or carpals of the forelimb. Additionally, after histologically clearing and staining, we were unable to definitively locate the pisiform or a pisiform-like structure; neither Safranin O nor a modified PAS protocol revealed any distinct pisiform morphology. This lack of ossification has been noted in other aquatic species such as some species of whales (Cooper et al. 2007). Previous analyses have concluded that the pisiform never develops in amphibious specimens (Fabrezi and Alberch 1996).

Early tetrapods appear to maintain cartilaginous epiphyses (Haines 1942). We can tentatively support this hypothesis with the amphibious µCT data – no bony epiphyses are present on the carpals, phalanges, or radius and ulna. If the epiphyses are cartilaginous, they would not be visible in the µCT scan. Haines hypothesized that the lack of epiphyses allowed for a larger amount of cartilage within the limb structure, functioning as a shock absorption system while navigating a rocky substrate (Haines 1942). Further analysis using a more diverse range of species is required to identify the potential homologies with the pisiform in amphibians.

Figure 2-1 µCT image of newt forelimb generate using Avizo 8.0.1. This scan demonstrates the lack of secondary ossification centers present within the species and furthermore the lack of a pisiform

19 Reptiles

Three reptilian species were used for analysis of carpal developmental trajectories: the

American alligator, the Eastern fence lizard, and Carolina anole. In the alligator specimens, the

µCT data indicated a lack of secondary centers of ossification within the pisiform and the long bones. The bone surfaces of the primary centers were porous in an organized manner, perhaps indicating that ossification and growth were still occurring at the time of death. Of the four specimen sizes, 14.75 inch alligator, 15.25 inches, 48 inches, and 82 inches, we chose to focus our efforts on the 15.25 inches long alligator as it was the most intermediate in age and thus most likely to provide growth plate data. Histological analysis utilized a Safranin O staining protocol.

The stained slides indicate an organized ossification pattern within the alligator pisiform. This conclusion corroborates with data provided in Reno et al. (2007), which shows that both mice and alligators have highly organized chondrocyte zones in the growth plate, but that alligators are distinct in having an irregular ossification zone (Reno et al. 2007). Comparison across several differently sized alligator specimens indicates irregular ossification of the alligator forelimb. We observe directional ossification in the alligator pisiform and cellular organization similar to a growth plate, but the alligator is lacking in a secondary ossification center. Figure 2-1 demonstrates a curvilinear, almost elliptical shape of the cartilage to bone transition, demarcated by red for cartilage and green for bone. Within the cartilaginous portion there is a zone of hypertrophy, as well as a perceived columnar zone, where several flattened cells are stacked upon each other. It is especially interesting to call attention to the unique ossification patterns in the

µCT image of the alligator specimen below, which is notably porous towards the ends of the long bones.

20

Figure 2-2. µCT and histological imaging of the 15.25 inches long alligator. Note the lack of epiphyseal development in the µCT image. In this histological image of the alligator pisiform, Safranin O stains cartilages red and Fast Green stains bone and other tissues green. The histological image indicate cartilage with directional growth. The pisiform is indicated by the red arrow.

Figure 2-3. µCT image of 48-inch alligator forelimb, indicating a epiphyseal line on the long bone (blue arrow) and a lack thereof on the pisiform, further supporting our findings from the 15.25 inch alligator. Red arrow indicates pisiform, blue arrow indicates epiphyseal line.

To round out the reptilian component of this study, we used two different species of lizard – a fence lizard species and anoles lizard. Phylogenetically, these species are more closely

21 related to each other than either are to alligators. This close proximity is clearly demonstrated in the developmental trajectories of these species, as both µCT and histological sections of the anoles and fence lizard closely resemble each other (figures 2-3 and 2-4). Whereas the histological images from the alligator specimen showed a distinct and directional ossification pattern, neither lizard species appears to establish any directional ossification in the pisiform.

Additionally, the µCT data for the lizards demonstrates clearly recognizable epiphyseal lines in the long bones, leading us to conclude that our observations do not indicate growth plate development in the pisiform. Given these results, it is likely that the pisiform does not develop a secondary center of ossification nor a growth plate in the suborder Iguania, to which both lizard species belong.

This may be a morphological and developmental feature exclusive to this lineage or potentially indicate a more generalized reptilian condition. The ability of certain species, like the green anolis lizard, to adapt to multiple niches seems to correlate with plasticity in body plan (Di-

Poï et al. 2009). A pisiform-like element has been noted in species like Seymouria sanjuanensis, a reptile-like tetrapod dating to the early Permian, approximately 280 – 270 million years ago

(Berman et al. 2000). Both of the specimens listed in the Berman study were identified as adults; epiphyseal fusion lines are unlikely to be present and it is impossible to determine the presence or absence of a pisiform growth plate.

22

Figure 2-4. µCT and histological analysis of the fence lizard (left). Staining utilized a modified Periodic Acid Schiff protocol (right). Note the distinct epiphyseal lines on the µCT image of this juvenile (blue arrow) in contrast with the distinct lack of an epiphyseal line within the pisiform (red arrow). The red arrow indicates the pisiform and the blue arrow indicates an epiphyseal line, here highlighted on the ulna.

Figure 2-5. µCT scan of anoles lizard (left) and Periodic Acid Schiff staining of histological section (right). Like the fence lizard, there are distinct growth plate lines along the long bones of the anoles lizard autopodium (blue arrow) and a lack of a secondary ossification center in the pisiform (red arrow).

23 We observe epiphyseal lines on the radius and ulna of the lizard specimens, however, the pisiform does not appear to develop a secondary center of ossification or growth plate. In the alligator, there appears to be a small growth plate surface which is different from the periosteal surface. We conclude that alligator pisiform ossification can be more complex than a typical carpal and thus developmentally resembles pisiform growth patterns in mammals. The location of the pisiform within the lizard manus is worth examination. The typical mammalian pisiform is rod-shaped and articulates with both the ulna and the triquetral (see figure 2-6). In extant hominoid species, the pisiform has shifted distally and articulates solely with the triquetral, thus losing contact with the ulna. Contrary to both of these known wrist joint configurations, the anolis and fence lizard pisiform appears to be located proximally, with no contact with the triquetral and instead resting on the medial portion of the ulna. This may reflect the ancestral condition of the pisiform, which is thought to have originated as a bud from the ulna by some researchers

(Stafford and Thorington 1998).

Mammals

The mammalian lineage is represented by the mouse in this study. The presence of a secondary ossification center and growth plate has been confirmed in the mouse pisiform previously, which is what we would expect from a mammalian species given previous research both on the mouse and on other mammalian species across diverse lineages (Kjosness et al. 2014,

Kjosness 2017). The µCT and histological data provide a clear image of a developing growth plate morphologically and at the cellular level, with all of the defining zones of growth visible in the histological section (see figure 1-2 for review). A clear secondary center of ossification is visible on the elongate pisiform in the µCT data, with the epiphyseal line demarcating the space in which the growth plate is maintained. Previous work observing the mouse metacarpals and

24 pisiform was confirmed as we identified the presence of an active growth plate within this carpal

(Reno et al. 2007, Kjosness et al. 2014, Kjosness 2017).

Figure 2-6. Image data from mouse demonstrating an elongate, rod-like pisiform (left) with a clear secondary center of ossification (blue arrow), as well as a histologically clear growth plate (right). Compare with images from chapter 1, also featuring a mouse autopod and pisiform growth plate, for review of cellular zones of growth plate and their correlation to this image. Blue arrow indicates the epiphyseal line on the mouse pisiform.

Conclusion and Discussion

Given the differences between the mammalian and reptilian specimens, the developmental and evolutionary history of the pisiform is more complicated than originally considered. We observe no pisiform formation as well as a lack of secondary ossification centers in the forelimb in the amphibian, directional ossification in the alligator, a lack of formal cartilage organization in the lizards, followed by the formation of a growth plate and secondary ossification center in the mouse and then subsequent loss of the growth plate in humans. These results indicate a complex history of pisiform ossification.

25 Ophiacodon, a basal synapsid from the Late Carboniferous to the Early Permian

(approximately 306 – 280 MYA), demonstrates a pisiform that ‘stands out like a shelf posteriorly’ (Eaton 1962). The pisiform may be a remnant of a similar skeletal element found in

Rhipidistian fish (Eaton 1962), but further work is required to establish homology. A subadult crocodile specimen from Las Hoyas, an early Cretaceous cite in Spain, was noted to have an elongate, oval-shaped pisiform located along the medial side of the ulnare (Buscalioni 1997), a configuration similar to what we have observed in the lizard specimens. If these species do share common morphological traits, functional analysis may elucidate the relationship between certain behavioral patterns and pisiform shape and location. The extant alligator pisiform is short and unlike that of the Cretaceous specimen. However, it does appear to undergo directional endochondral ossification, which is the hallmark of growth plate activity.

Moving forward, it would be beneficial to explore other reptilian and amphibious species, to gain a more complete understanding of pisiform formation and evolution across these diverse groups, as well as subsequent analysis of the reptilian species utilized here for a complete developmental timeline. Additionally, research involving Hox gene expression within these species could illuminate some of the morphological variants noted within these reptilian and amphibious specimens and better contextualize the molecular pathways that underpin gross morphological variation.

26 Citations:

Ahlberg, P. E., & Milner, A. R. (1994). The origin and early diversification of tetrapods. Nature, 368(6471), 507-514. Amemiya, C. T., Alföldi, J., Lee, A. P., Fan, S., Philippe, H., MacCallum, I., ... & Organ, C. (2013). Analysis of the African coelacanth genome sheds light on tetrapod evolution. Nature, 496(7445), 311. Berman, D. S., Henrici, A. C., Sumida, S. S., & Martens, T. (2000). Redescription of Seymouria sanjuanensis (Seymouriamorpha) from the Lower Permian of Germany based on complete, mature specimens with a discussion of paleoecology of the Bromacker locality assemblage. Journal of Vertebrate Paleontology, 20(2), 253-268. Cooper, L. N., Berta, A., Dawson, S. D., & Reidenberg, J. S. (2007). Evolution of hyperphalangy and digit reduction in the cetacean manus. The Anatomical Record, 290(6), 654-672. Di-Poï, N., Montoya-Burgos, J. I., & Duboule, D. (2009). Atypical relaxation of structural constraints in Hox gene clusters of the green anole lizard. Genome research, 19(4), 602- 610. Eaton, T. H. (1962). Adaptive features of the fore limb in primitive tetrapods and mammals. American Zoologist, 157-160. Fabrezi, M., & Alberch, P. (1996). The carpal elements of anurans. Herpetologica, 188-204. Haines, R. W. (1942). The evolution of epiphyses and of endochondral bone. Biological Reviews, 17(4), 267-292. Hallgrímsson, B., Jamniczky, H. A., Young, N. M., Rolian, C., Schmidt‐ ott, U., & Marcucio, R. S. (2012). The generation of variation and the developmental basis for evolutionary novelty. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution, 318(6), 501-517. Kherdjemil, Y., & Kmita, M. (2017). Insights on the role of hox genes in the emergence of the pentadactyl ground state. genesis. Kjosness, K. M., Hines, J. E., Lovejoy, C. O., & Reno, P. L. (2014). The pisiform growth plate is lost in humans and supports a role for Hox in growth plate formation. Journal of anatomy, 225(5), 527-538. Kjosness, K. M. 2017. Evolutionary and Developmental Mechanisms of Pisiform Growth Plate Loss. Doctor of Philosophy, The Pennsylvania State University. Laurin, M. (1998). A reevaluation of the origin of pentadactyly. Evolution, 52(5), 1476-1482. Paço, A., & Freitas, R. (2017). Hox D genes and the fin‐to‐limb transition: Insights from fish studies. Genesis. Pineault, K.M., Swinehart, I.T., Garthus, K.N., Ho, E., Yao, Q., Schipani, E., Kozloff, K.M. and Wellik, D.M. (2015). Hox11 genes regulate postnatal longitudinal bone growth and growth plate proliferation. Biology open, bio-012500. Reno, P. L., Horton, W. E., Elsey, R. M., & Lovejoy, C. O. (2007). Growth plate formation and development in alligator and mouse metapodials: evolutionary and functional implications. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution, 308(3), 283-296.

27 Reno, P. L., Kjosness, K. M., & Hines, J. E. (2016). The role of Hox in pisiform and calcaneus growth plate formation and the nature of the zeugopod/autopod boundary. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution, 326(5), 303-321. Serrat, M. A. 2007. Environmentally-Determined Tissue Temperature Modulates Extremity Growth in Mammals: A Potential Comprehensive Explanation of Allen's Rule. Doctor of Philosophy, Kent State University. Sheth, R., Marcon, L., Bastida, M. F., Junco, M., Quintana, L., Dahn, R., ... & Ros, M. A. (2012). Hox genes regulate digit patterning by controlling the wavelength of a Turing-type mechanism. Science, 338(6113), 1476-1480. Shubin, N. (1995). The evolution of paired fins and the origin of tetrapod limbs. Evolutionary Biology, 28, 39-86. Shubin, N., Tabin, C., & Carroll, S. (1997). Fossils, genes and the evolution of animal limbs. Nature, 388(6643), 639-648. Stafford, B. J., & Thorington, R. W. (1998). Carpal development and morphology in archontan mammals. Journal of Morphology, 235(2), 135-155. Takezaki, N., & Nishihara, H. (2017). Support for Lungfish as the Closest Relative of Tetrapods by Using Slowly Evolving Ray-Finned Fish as the Outgroup. Genome biology and evolution, 9(1), 93-101. Wagner, G. P., & Chiu, C. H. (2001). The tetrapod limb: a hypothesis on its origin. Journal of Experimental Zoology Part A: Ecological Genetics and Physiology, 291(3), 226-240.

28 Chapter 3

Xenarthran Pisiform Morphology and Comparison with Hominoid Species

The extant hominoid wrist joint is primarily defined by a reduction of the distal ulna, also called ulnar withdrawal, and a relocation of the pisiform. This modification of the wrist joint eliminates contact of the pisiform and triquetral with the ulna, leaving a single articulation between the pisiform and triquetral. This suite of features is typically associated with suspensory locomotor or postural behaviors, distinct from the classic primate palmigrade or digitigrade quadrupedal morphology, where the ulna directly articulates with the triquetral and pisiform

(Mivart 1867, Lewis 1969). It has been proposed that there was a single morphological shift towards ulnar withdrawal at the beginning of the hominid lineage and the configuration has been maintained throughout the individual evolution and development of each species (Lewis 1969,

Lewis 1970, Lewis et al. 1970).

Additionally, the pisiform is the only carpal to form a growth plate and develop like a long bone (Kjosness et al. 2014). This is a feature we see in many mammals, though it is conspicuously lost in human development. The human pisiform only develops from one center of ossification, much like the other carpal bones, and resembles a nodular pea-shape versus the elongate pisiform observed in other mammals, including Pan troglodytes and Australopithecus afarensis (Kjosness et al. 2014).

Ulnar withdrawal within the hominoid lineage has typically been associated with a history of brachiation and below-branch suspension, however the exact origins of this morphological configuration and the related function is contested (Lewis 1969). Ulnar withdrawal is present in all extant hominoids, but it is not a feature unique to the lineage as previously thought (Lewis 1969, 1970). Ulnar withdrawal has also been identified within the lorisine

29 primates, two-toed sloths, and three-toed sloths, presenting a remarkable case of convergent or parallel evolution (Cartmill and Milton 1977, Mendel 1979).

It has been hypothesized that ulnar withdrawal and the superposition of the pisiform on the triquetral functions to increase wrist flexibility in relation to brachiation within hominoids

(Lewis 1969). The discovery of ulnar withdrawal in both lorisoids and sloths confounds this explanation, as neither family has a known brachiating ancestor (Cartmill and Milton 1977,

Mendel 1979). I hypothesize that this similarity is due to the reduction of compressive force and stress on the joints, facilitating all forms of below-branch locomotion. Over time, the wrist configuration of these species has reorganized to better suit this below-branch lifestyle and the developmental processes leading to limb formation may reflect this ecological shift as it has driven evolutionary change. The goal of this study is to analyze the efficacy of Xenarthra and lorisine primates as models in an evolutionary and developmental context to explore the relationship between primate pisiform form and function. Numerous morphological similarities have been identified between these distinct lineages, but it is not well understood if the shortened pisiforms reflect the similar changes in skeletal development and ossification.

Lorisine and Xenarthran Evolutionary History

The family Xenarthra is comprised of sloths, anteaters, and armadillos. Sloths are one of the few non-primate examples of obligate suspensory postural and locomotor behavior, making their forelimb morphology a candidate for research in parallel evolution. Additionally,

Myrmecophaga tridactyla, the giant anteater, is the only known non-hominoid species to engage in knuckle-walking. Several functional morphology studies have demonstrated the remarkable similarities in xenarthran wrist morphology and hominoid morphology, as both two-toed and

30 three-toed sloths demonstrate ulnar withdrawal and the anteater shares many knuckle-walking specific traits with chimpanzees, such as robust medial phalanges (Cartmill and Milton 1977,

Mendel 1979, Sarmiento 1988, Orr 2005). The aforementioned structural skeletal changes in brachiating hominoids likely serve to allow a greater range of flexibility within the wrist joint than the standard palmigrade morphology, which is critical in these suspensory behaviors

(Mendel 1979, Schmitt et al. 2016). Additionally, the ability to rotate the wrist greater than 180˚, in conjunction with a high degree of flexion and extension, allows branch substrate at varying positions in a complex three-dimensional environment to be navigated with ease. It is commonly thought that the dramatic wrist change in hominoids has to do with adaptation for other forms of locomotion to access novel niches such as leaping, knuckle-walking, climbing, as well as a myriad of postural behaviors (Lewis 1972). There are other species, such as the spider monkey or the howler monkey, that also engage in brachiation. However, ulnar withdrawal and relocation of the pisiform is not universal brachiating platyrrhines.

A common thread between lorises and sloths is their unique mode of cautious climbing, also called slow arborealism. Both groups are famously known for their slow locomotion through the trees; they seemingly move at a pace incongruous with the fast-paced lifestyles of nearly every other species around them. Both two-toed and three-toed sloths locomote using suspensory, inverted quadrupedalism and are the only extant species to do so. Lorises facultatively alternate between above-branch slow quadrupedalism and inverted quadrupedalism, while sloths are obligated to locomote as below-branch quadrupeds (Cartmill and Milton 1977, Ashton and

Oxnard 1964). Hominoids, lorises, and sloths are all separated by millions of years of independent evolution, yet share this common thread of suspensory postures in a case of convergent evolution. To date, there are only three published articles that examine the similarities in locomotor behavior and skeletal morphology between xenarthrans, hominoids, and lorisines.

Cartmill and Milton (1977) first noted the morphological similarities between the Lorisidae (slow

31 loris, slender loris, angwantibo, and potto) and the extant hominoids. The morphological similarities between these species are suggested to stem from their cautious arboreal behaviors.

All hominoids and lorises maintain arboreal substrate contact with three limbs while the fourth limb searches for purchase; once contact is made, the other limbs will be moved lockstep as a way to ensure tree limb security during locomotion. Mendel (1979) took this analysis a step further and included the two-toed and three-toed sloths for analysis. Sloths also exhibit ulnar withdrawal, arboreal lifestyles, and cautious locomotion. Orr (2005) provided a functional morphological analysis of the giant anteater forelimb as it relates to the chimpanzee forelimb as they are both terrestrial knuckle-walkers that share skeletal and muscular morphological similarities in the wrist and hand. While these studies have provided critical information regarding the functional anatomy and gross morphology of these species, no work to date has analyzed the similarities between xenarthrans, lorisines, and hominoids from an ontogenetic perspective.

Lorisine Evolutionary History

There is a plethora of misinformation regarding lorisine primates, often regarded as the lesser of the strepsirrhine family and relegated to live in the shadow of the lemur. The lorises are a diverse and vastly underrepresented group within anthropological and biological literature, which leaves them often excluded from studies of primate evolution as they are considered the most primitive group within the family. However, given their remarkable convergent evolution with the hominoids, often touted as the most derived family, the lorises can provide a fascinating look into the evolution of primates, hominoids, and even humans. Extant lorisine primates currently exist in African and Asian tropical forests. The family Lorisidae is comprised of four species overall – the potto and angwantibo of Eastern Africa as well as the slender loris and slow

32 loris of the Asian tropical forests. The Galagidae are also a member of the superfamily Lorisoidea that can be used to understand the evolutionary history of Lorisidae. Galagos maintain a small degree of ulnar withdrawal like the lorises; however, galagos will leap between tree gaps while lorises employ bridging postures to cross these spaces.

There is some debate regarding whether the lorisine fossil ancestors are the adapoids or omomyids (Szalay and Katz 1973, Kunimatsu et al. 2017). Naturally, there is always the possibility that the most ancestral loris has been lost through time or has yet to be discovered.

One of the many issues that arise when defining fossil lorisines stems from the use of soft tissues such as a wet nose to identify the Strepsirrhini. Strepsirrhini fossil primates are difficult to identify, given that the suborder has been traditionally used as a catch-all for primate species that cannot be classified as Haplorhini. Little material from fossil lorisines has been recovered from the Eocene and Oligocene. Current lorisines all exhibit slow climbing quadrupedalism, despite the lorises being Asian primates and the pottos being located in Africa. There is considerable debate within the literature regarding the alignment of certain Miocene fossils such as Karasinia with the lorisines or the galagines (Seiffert 2007, Walker 1969, Walker 1974, Gebo 1986).

Although less frequently cited, there is a possibility that the lorisine/galagine split occurred after the Miocene, which would explain the confusion surrounding the fossil remains (Rasmussen and

Nekaris 1998). It is also worth mentioning that the galagos represent a recent adaptive radiation while the lorises appear to split much earlier with only a few modern species remaining

(Rasmussen and Nekaris 1998).

The lorisine locomotor pattern requires the wrist to be highly flexible and the shoulder joint to orient itself in a brachiator-like manner, which would make sense if ulnar withdrawal equates to increased joint flexibility. There is some indication that the lorisines are facultative inverted quadrupeds, much like the sloth. (Walker 1969, Yalden 1972, Ashton and Oxnard 1964).

33 Once the ancestors and behaviors of the loris are more thoroughly understood, we can use that information to understand the development of ulnar withdrawal in hominoids more thoroughly.

Two-Toed and Three-Toed Sloth Evolutionary History

Although it is commonly thought that the two-toed and three-toed sloths are closely related, there is actually an extended period of independent evolutionary development in both species (Nyakatura 2012). It was originally thought that the tree dwelling two-toed sloth species,

Tardigradus, were distinct from the extinct ground sloths, the Gravigrada (Owen 1842). The sloth fossil record is complicated but, as opposed to Owen (1842), there is strong evidence that two-toed and three-toed sloths are the descendants of two separate and distinct species of extinct ground sloths, with two-toed sloths descending from Megalonychidae and three-toed sloths sharing ancestry with Megatheriidae (Delsuc et al. 2004).

Fossil sloths exhibit an extremely wide range of behaviors. Though there are only six species living today, roughly 50 species have been uncovered from the fossil record (Raj Pant et al. 2014). It is estimated that nearly 90% of sloth species went extinct following The Great

Megafaunal Extinction Event during the Pleistocene (Steadman et al. 2005). There is much contention surrounding the prime mover behind this event, with human arrival being cited by some while others claim climate change proved too great for American megafauna (Barnosky et al. 2004, Barnosky and Lindsey 2010). Prior to this event, there is growing evidence that sloths were exploiting a highly diverse range of locomotor behaviors and morphologically adapted to these behaviors. There is evidence of terrestrial, semi-arboreal, aquatic, and even facultatively bipedal sloths, though all extant sloths are exclusively arboreal (Melchor et al. 2015, Casinos

1996). There is no known evidence of suspensory behaviors in fossil sloths (Nyakatura 2012). In

34 addition to diversity in their movements, there is a suggestion that they exploited a variety of dietary adaptations as well (Bargo et al. 2006).

None of the extinct fossil sloths exhibit suspensory behavioral adaptations, and it is hypothesized that the last common ancestor between the two modern sloth groups was probably terrestrial and likely non-suspensory (White 1993). The two-toed sloth and three-toed sloth are morphologically similar, yet separated by approximately 30 million years of independent evolution. This, combined with the shared obligatory suspensory posture and locomotion in modern sloths creates "one of the most striking examples of convergent evolution known among mammals" (Gaudin 2004, pg 255).

Like lorisines, sloths did not have a history of brachiation but utilized below branch quadrupedalism. Napier and Napier (1967) call it slow-climbing quadrupedal locomotion. Three- toed sloths have also been observed engaging in vertical climbing behaviors (Mendel 1985).

Neither species has a history of brachiation, but this cautious arborealism requires them to have extremely flexible limbs as they search for a substrate to make purchase on. Despite their size, roughly three to five kilograms, both groups of sloths show a marked preference for moving under small supports such as tertiary branches or on vine-laden trees (Mendel 1985, Mendel

1981, Montgomery and Sunquist 1978). Both two-toed sloths and three-toed sloths have incredibly rigid phalanges, represented as hooks, and incredibly lax articulations between the carpal bones as well as between the proximal row of carpals and the radius/ulna (Mendel 1979,

1981). Mendel noted loose articulations between the humerus and the scapula (Mendel 1979).

This combination of flexible joints in both the shoulder and the wrist ostensibly gives sloths the ability to move their arms in all directions unhindered, which is a critical development necessary for their below-branch suspensory quadrupedalism. Both groups locomote below-branch (sloths are obligate, lorises are facultative). Two-toed sloths utilize the smaller secondary and tertiary branches of tropical forests, which bend and intersect in complex ways. To navigate this ecology,

35 they would need a highly flexible joint. Their forelimb and scapular anatomy look similar to hominoids in other ways.

In another case of parallel evolution, the giant anteater, Myrmecophaga tridactyla engages in knuckle-walking locomotion, a behavior typically associated solely with chimpanzees, bonobos, and gorillas. Morphologically, the giant anteater shares a number of forelimb skeletal similarities with chimpanzees and provides a model of convergence for these knuckle-walking behaviors (Orr 2005). The features include, but are not limited to, a robust third metatarsal for stabilization and primary substrate contact, a distally projecting dorsal ridge on the scaphoid, a fused os centrale, and, most importantly for this study, an elogante, palmarly oriented pisiform

(Orr 2005). Additionally, like the chimpanzee and unlike the sloths, the anteater pisiform articulates with both the ulna and the triquetral, thus resembling a more common mammalian configuration over the highly derived morphology of ulnar withdrawal and relocation of the pisiform.

Hominoid evolution

Ulnar withdrawal is a critical adaptation in hominoid evolution and it is thought that this morphological adaptation is an adaptation to suspensory behaviors (Lewis 1969, Lewis 1970,

Carmill and Milton 1977). Brachiating adaptations in the musculoskeletal system of all extant hominoids have been known for a long time and are often considered to be a strong piece of evidence for shared familial status (Gregory 1934). Naturally, there are a handful of paleoanthropologists who stress the reconfiguration of the human skeleton for bipedality and cite the loss of several brachiator-specific skeletal features like a more dorsally oriented scapula

(Fleagle 1983, Selby and Lovejoy 2017). It is considered great luck when any limb bones can be

36 recovered from a fossil species, making our knowledge of Miocene ape locomotor behaviors sparse. It is uncommon to find phalanges or carpals within the fossil record, given their small size and thus make a direct comparison between the extinct and extant apes difficult. Using knowledge gained from extant hominoids in combination with lorises and sloths can provide a deeper understanding of the development of these features.

The likely stem catarrhine, Aegyptopithecus zeuxis, recovered from the Fayum and dating to roughly 35 million years ago, shows a mixture of features that do not clearly align with either extant cercopithecines or with the suspensory behaviors of apes (Ciochon and Corruccini, pp 301-

324). The distal end of the ulna has yet to be recovered from the fossil record, yet the features on the proximal end of the bone show indication that the ulna of A. zeuxis is "clearly that of a slowly moving, relatively short-limbed arboreal quadruped" (Bown et al. 1982). Propliopithecus chirobates, also sometimes referred to as Aeolopithecus in some literature, is an Oligocene primate for which limb bone material has been recovered (Gebo and Simons 1987). The limb skeleton of this species appears relatively similar to that of A. zeuxis, as many of the features of these recovered long bones are indicative of an arboreal quadrupedal lifestyle. It has been suggested that P. chirobates was capable of hindlimb suspension by some and that their limb skeleton is more primitive than seen in A. zeuxis (Fleagle 1983). The postcranial remains of A. zeuxis were compared to Alouatta, the howler monkey, and P. chirobates was likened to

Notharctus, an adapiform primate from the Eocene often considered a strepsirrhine (Bown et al.

1982).

In Proconsul, it has been noted that these basal apes likely utilized generalized or slow- moving arboreal quadrupedal behavior, evidenced by their lightly constrained joint and ligament morphology recovered from the fossil record (Napier and Davis 1959, Aiello 1993, Ciochon and

Corruccini pp 325-351). Analysis of the semicircular canal reaffirms the paleontological material

(Ryan et al. 2012). In their 1959 paper, Napier and Davis explain the great importance of the

37 Proconsul wrist and its impact on our understanding of hominoid adaptations, "It's clear that these bones belong to one of the most significant periods of primate evolution: a period when the generalized Catarrhine stock was emerging from a prolonged phase of arboreal quadrupedalism

[…] and was entering upon a phase that would provide a diversity of environmental opportunity leading ultimately to the emergence of four distinct patterns of locomotion among the

Anthropoidea" (Napier and Davis 1959, pg 1). There is little debate that Proconsul was a quadruped, though some researchers also suggest that the beginnings of brachiator adaptations can be noted in their morphology (Lewis 1971, Lewis 1972, Napier and Davis 1959). It almost goes without saying that the locomotor patters of Proconsul are hotly debated by many scholars.

Sivapithecus, a Miocene ape, often touted as the progenitor species for modern orangutans, shows few suspensory adaptations (Morgan et al. 2015). This could indicate that the skeletal modifications associated with brachiation occurred in parallel between the African and

Asian apes (Ward 2015). If parallel evolution is indeed what transpired, then reduction of the ulna and the relocation of the pisiform would have also occurred in parallel. Both Proconsul and

Sivapithecus are classified and dated as Miocene primates, yet they are separated by approximately 10 million years.

Forelimb remains from Pierolapithecus are relatively complete with all carpals recovered except for the pisiform. Ulnar withdrawal has been confirmed in this species along with a suite of derived skeletal morphologies, such as a shortened lumbar column, similar to extant apes.

However, other features of the hand skeleton indicate Pierolapithecus engages in little, if any, suspensory behavior (Almecijia et al. 2009). This suggests that some iconic hominoid traits, including ulnar withdrawal, may not have evolved within the context of below branch suspension.

Remains from this species are dated to approximately 12 million years ago, placing this species temporally near the proposed last common ancestor of the great apes (Moy à-Solà et al. 2004).

38 All known evidence points to Aegyptopithecus zeuxis as a slow, climbing, arboreal primate and likewise the remain of Propliopithecus show similar adaptations. The origins of the hominoid lineage may be murky, but the beginnings of brachiating or slow-climbing behaviors are already being noted. The lack of limb remains make it difficult to concretely identify ulnar withdrawal or relocation of the pisiform. Understanding the morphological history of the wrist joint can inform our perception of the developmental trajectories in each species. In turn, a strong understanding of developmental processes in extant species provides information that allows us to research hominoid evolution through a novel lens.

Methods

All Xenarthran and lorisine specimens were scanned on a GE v|tome|x L 300 high- resolution microfocus CT system at the Center for Quantitative Imaging, The Pennsylvania State

University. The µCT data was then analyzed using Avizo 8.0.1. All but one specimen were generously loaned by the Department of Mammalogy at the American Museum of Natural

History (New York, NY). One galago specimen was used from the Department of Anthropology collection at The Pennsylvania State University (University Park, PA). Complete list of specimens is featured in table 3-1. Voxel size is approximately 0.015 with one specimen scanned at a resolution of voxel size 0.02. All specimens selected are considered juvenile based on the presence of epiphyseal lines and dental eruption patterns.

Table 3-1. List of specimens borrowed from AMNH mammology collections along with the scientific and common names for each specimen.

Species name (common name) AMNH Loan Number

39

Bradypus tridactylus (three-toed sloth) 77894

Bradypus tridactylus (three-toed sloth) 77895

Myrmecophaga tridactyla (giant anteater) 100136

Myrmecophaga tridactyla (giant anteater) 130242

Myrmecophaga tridactyla (giant anteater)

Arctocebus calabarensis (agwantibo) 212954

Perodicticus potto (potto) 269907

Galago senegalensis (galago) Ryan Lab Specimen

Galago senegalensis (galago) 212956

Galago senegalensis (galago) 212957

Choloepus hoffmanni (two-toed sloth) 209941

Lorisoidea

Galago

Three galago species were analyzed as a complement to the lorisine primates, given their phylogenetic proximity. As mentioned in the introductory material, galagoes also demonstrate some degree of ulnar withdrawal like the lorises, however, they are not cautious arborealists, instead preferring to leap from tree limb to limb.

Our results show an elongate pisiform in all three galago specimens with an articulation between the pisiform and ulna (see figures 3-1 to 3-4). The pisiform from specimen 212956 is additionally isolated, further exemplifying the rod-like morphology that is typical of a generalized quadrupedal primate pisiform (Kjosness 2017).

40

41 Figure 3-1. Image data from the three galago specimens. Blue arrows indicate the pisiform in each specimen. Across several differently aged specimens we see a typical monkey-like pisiform which is elongate. Image sequence C features the zeugopod/autpod boundary along with the manus. Note the clear epiphyseal lines on the radius and ulna. The pisiform is also shown in isolate.

Lorises

Of the four lorisine subgroups, we only have data for Arctocebus (agwantibo) and the potto. Both the potto and the agwantibo are African lorisines, with Nycticebus (slow loris) and

Loris (slender loris) residing in southeastern Asia and Polynesia. What distinguishes the lorisine primates from the galagoes is the absence of leaping or jumping behaviors in their locomotor repertoire and instead a focus on bridging or cantilever postures (Nekaris and Bearder 2007). All lorisines rely on their flexible, elongated limbs to maneuver through forested regions (Nekaris

2001). There is no evidence that the Asian slow loris and slender loris have demonstrably distinct morphological features from the African potto and agwantibo. Despite the varying home ranges, diet preferences, and niche occupation between these species, there is no indication that their locomotor behaviors are drastically different. Thus, while all literature cited up to this point has focused on the Asian lorises, we feel confident that conclusions made based on African loris morphology is likely applicable to the slow and slender loris. We look forward to testing this in the near future.

The Arctocebus calabarensis specimen (AMNH specimen number 212954) demonstrates ulnar withdrawal just as we would expect given the data published in Cartmill and Milton (1977).

The agwantibo pisiform is distally relocated, and does not articulate with the ulna. Figure 3-5 clearly shows a reduced, shortened pisiform. The epiphyseal fusion lines are clear on the radius, though none are noticeably present on either the pisiform or the metacarpals. Given this absence, it is likely that we have missed the ontogenetic timing of pisiform ossification and inadvertently chosen a specimen that is too old for the scope of this project. An addition of other agwantibo

42 specimens at diverse juvenile ages should provide a better temporal resolution to our questions. It is interesting to note that the agwantibo pisiform is shortened, although this does not immediately indicate the loss of the growth plate – orangutans have shortened pisiforms that still form a

secondary center of ossification and growth plate

(Kjosness

2017).

Figure 3-2. Agwantibo µCT scan. There are clear epiphyseal lines, shown by the blue arrow, and a short pisiform, indicated in red. The pisiform is nodular in this agwantibo specimen. To confirm the presence or absence of a pisiform growth plate will require more specimens.

43

Figure 3-3. Potto µCT scan. We believe that the pisiform has been displaced during the specimen drying process and thus does not accurately represent location in-vivo. Additional specimens will be needed to confirm accurate pisiform location. Likely pisiform indicated with red arrow.

Xenarthan Development

Giant Anteater

The giant anteater µCT data provides strong evidence for a secondary center of ossification. It appears that the pisiform was slightly displaced during the drying process in all three specimens, although this did not affect the interpretation of our data. Figure 3-7 provides an overview of the giant anteater autopod. Figure 3-8 shows a number of epiphyses on the metacarpals and we can clearly see that secondary center of ossification on the pisiform. In addition to the adult morphological and functional similarities noted by Orr (2005), the ontogenetic processes of carpal development in the giant anteater autopod align closely with the processes of chimpanzee skeletal development, further emphasizing the homoplasy and parallel evolution between these distinct species (Kjosness 2017, Kivell and Schmitt 2009, Kivell et al.

2013).

44

Figure 3-4 Giant anteater µCT scan. Pisiform indicated with red arrow. Anteater specimen demonstrates clear secondary center of ossification as well as an epiphyseal line. This morphology is more similar to what we would expect from a quadruped and is similar to the chimpanzee pisiform.

Sloths

The two-toed sloth has a highly unusual skeleton, with many studies focusing on the vertebral column (see Hautier et al. 2017 for a review of xenarthran developmental research on this subject). The two-toed sloth, Choloepus hoffmani (AMNH specimen number 209941) used for this study provided an equally odd carpal morphology. Figure 3-9 shows a number of fused carpals. It is likely that the proximal row of carpals is missing from this μCT data although this will need to be confirmed in future work with additional specimens. Notice that the most distinct epiphyseal lines are present on metacarpals that are highly reduced in the adult two-toed sloth – this may be an example of heterochrony within the autopod and warrants further investigation with a larger sample size, given the plethora of morphological oddities noted in this species.

45 Figure 3-5. Two-toed sloth µCT scan. We believe that the pisiform is not present in this data and may have been disarticulated from the rest of the carpals during preservation or storage.

The three-toed sloth specimens provide an equally interesting picture of pisiform development. Figures generated from µCT data show that the pisiform is small, nodular, and distally relocated away from the ulna. It appears that there is subchondral bone present on the pisiform, which could potentially indicate the presence of a growth plate although as mentioned previously there is no secondary center of ossification present in either of these scans.

46 Figure 3-6. Three toed sloth µCT scans. Pisiform indicated with red arrow. The pisiform is nodular. It appears that there is a subchondral surface on the pisiform, which may indicate the presence of a growth plate.

Discussion

Since the number of individuals for which we have μCT data is limited, it is difficult to determine the ontogenetic processes within the three-toed sloth pisiform. The subchondral surface implies the loss of the pisiform epiphysis; however, it is also possible that the secondary center of ossification occurs later than what we might initially expect. More data from differently aged juvenile specimens are needed to elucidate the processes at work.

There are a number of other morphological and developmental interests here, especially in the lorisine specimens, that warrant additional research involving a more complete ontogenetic timeframe as well as a diversity of lorisine species. Low species counts, while initially informative and relatively inexpensive to analyze, may indicate anomalies that are not actually indicative of the broader developmental trajectories.

Throughout this analysis, I have been focusing most heavily on the structural similarities and variations between the lorises, sloths, and humans. Currently, this is where most research regarding these species has focused. At the cellular level, no comparisons have been made. This is a project I hope to take on in the future, analyzing growth plate structure and development using tissue samples. However, given the numerous challenges that come with acquiring bone tissue from these endangered animals while they are still juveniles and sub-adults, this presents a nearly insurmountable setback in histological comparison.

There is a remarkable diversity of locomotor behaviors between various armadillo species – plantigrade, unguligrade, and potentially digitigrade postures are all represented within this group. Moving forward, the authors are interested in elucidating the developmental

47 mechanisms behind each species and contextualizing these differences within a reconstructed phylogeny.

Citations

Aiello, L. C. (1993). The fossil evidence for modern human origins in Africa: a revised view. American Anthropologist, 95(1), 73-96. Almecija, S., Alba, D. M., & Moya-Sola, S. (2009). Pierolapithecus and the functional morphology of Miocene ape hand phalanges: paleobiological and evolutionary implications. Journal of Human Evolution, 57(3), 284-297. Ashton, E. H., & Oxnard, C. E. (1964, January). Locomotor patterns in primates. In Proceedings of the Zoological Society of London (Vol. 142, No. 1, pp. 1-28). Blackwell Publishing Ltd. Bargo, M. S., Toledo, N., & Vizcaíno, S. F. (2006). Muzzle of South American Pleistocene ground sloths (Xenarthra, Tardigrada). Journal of Morphology, 267(2). 248-263. Barnosky, A. D., Koch, P. L., Feranec, R. S., Wing, S. L., & Shabel, A. B. (2004). Assessing the causes of Late Pleistocene extinctions on the continents. science, 306(5693), 70-75. Barnosky, A. D., & Lindsey, E. L. (2010). Timing of Quaternary megafaunal extinction in South America in relation to human arrival and climate change. Quaternary International, 217(1), 10-29. Bown, T. M., Kraus, M. J., Wing, S. L., Fleagle, J. G., Tiffney, B. H., Simons, E. L., & Vondra, C. F. (1982). The Fayum primate forest revisited. Journal of Human Evolution, 11(7), 603-632. Cartmill, M., & Milton, K. (1977). The lorisiform wrist joint and the evolution of “brachiating” adaptations in the Hominoidea. American Journal of Physical Anthropology, 47(2), 249-272. Casinos, A. (1996). Bipedalism and quadrupedalism in Megatheriurn: an attempt at biomechanical reconstruion. Lethaia, 29(1), 87-96. Ciochon, R. L., & Corruccini, R. S. (Eds.). (1983). New interpretations of ape and human ancestry. New York: Plenum Press. Delsuc, F., Vizcaíno, S. F., & Douzery, E. J. (2004). Influence of Tertiary paleoenvironmental changes on the diversification of South American mammals: a relaxed molecular clock study within xenarthrans. BMC Evolutionary Biology, 4(1), 1. Fleagle, J. G. (1983). Locomotor adaptations of Oligocene and Miocene hominoids and their phyletic implications. In New interpretations of ape and human ancestry (pp. 301-324). Springer US.

48 Gaudin, T. J. (2004). Phylogenetic relationships among sloths (Mammalia, Xenarthra, Tardigrada): the craniodental evidence. Zoological Journal of the Linnean Society, 140(2), 255- 305. Gebo, D. L. (1986). Miocene lorisids–the foot evidence. Folia Primatologica, 47(4), 217- 225. Gebo, D. L., & Simons, E. L. (1987). Morphology and locomotor adaptations of the foot in early Oligocene anthropoids. American Journal of Physical Anthropology, 74(1), 83-101. Godfrey, L. R., & Jungers, W. L. (2003). The extinct sloth lemurs of Madagascar. Evolutionary Anthropology: Issues, News, and Reviews, 12(6), 252-263. Gregory, W. K. (1934). Polyisomerism and anisomerism in cranial and dental evolution among vertebrates. Proceedings of the National Academy of Sciences, 20(1), 1-9. Kivell, T. L., & Schmitt, D. (2009). Independent evolution of knuckle-walking in African apes shows that humans did not evolve from a knuckle-walking ancestor. Proceedings of the National Academy of Sciences, 106(34), 14241-14246. Kivell, T. L., Barros, A. P., & Smaers, J. B. (2013). Different evolutionary pathways underlie the morphology of wrist bones in hominoids. BMC evolutionary biology, 13(1), 229. Kjosness, K. M., Hines, J. E., Lovejoy, C. O., & Reno, P. L. (2014). The pisiform growth plate is lost in humans and supports a role for Hox in growth plate formation. Journal of anatomy, 225(5), 527-538. Kjosness, K. M., & Reno, P. L. (2017). 9 Using Comparisons between Species and Anatomical Locations to Discover Mechanisms of Growth Plate Patterning and Differential Growth. Building Bones: Bone Formation and Development in Anthropology, 77, 205. Kunimatsu, Y., Tsujikawa, H., Nakatsukasa, M., Shimizu, D., Ogihara, N., Kikuchi, Y., ... & Ishida, H. (2017). A new species of Mioeuoticus (Lorisiformes, Primates) from the early Middle Miocene of Kenya. Anthropological Science, 125(2), 59-65 Lewis, O. J. (1969). The hominoid wrist joint. American Journal of Physical Anthropology, 30(2), 251-267. Lewis, O. J. (1970). The development of the human wrist joint during the fetal period. The Anatomical Record, 166(3), 499-515. Lewis, O. J., Hamshere, R. J., & Bucknill, T. M. (1970). The anatomy of the wrist joint. Journal of anatomy, 106, 539. Lewis, O. J. (1972). Osteological features characterizing the wrists of monkeys and apes, with a reconsideration of this region in Dryopithecus (Proconsul) africanus. American journal of physical anthropology, 36(1), 45-58. Melchor, R. N., Perez, M., Cardonatto, M. C., & Umazano, A. M. (2015). Late Miocene ground sloth footprints and their paleoenvironment: Megatherichnum oportoi revisited. Palaeogeography, Palaeoclimatology, Palaeoecology, 439, 126-143. Mendel, F. C. (1979). The wrist joint of two‐toed sloths and its relevance to brachiating adaptations in the Hominoidea. Journal of Morphology, 162(3), 413-424. Mendel, F. C. (1981). The hand of two‐toed sloths (Choloepus): its anatomy and potential uses relative to size of support. Journal of Morphology, 169(1), 1-19 Mivart, G. (1867). On the appendicular skeleton of the primates. Philosophical Transactions of the Royal Society of London, 157, 299-429.

49 Montgomery, G. G., & Sunquist, M. E. (1978). Habitat selection and use by two-toed and three-toed sloths. The ecology of arboreal folivores, 329-359. Morgan, M. E., Lewton, K. L., Kelley, J., Otárola-Castillo, E., Barry, J. C., Flynn, L. J., & Pilbeam, D. (2015). A partial hominoid innominate from the Miocene of Pakistan: Description and preliminary analyses. Proceedings of the National Academy of Sciences, 112(1), 82-87 Moyà-Solà, S., Köhler, M., Alba, D. M., Casanovas-Vilar, I., & Galindo, J. (2004). Pierolapithecus catalaunicus, a new Middle Miocene great ape from Spain. Science, 306(5700), 1339-1344 Napier, J. R., & Davis, P. R. (1959). The fore-limb skeleton and associated remains of Proconsul africanus. British Museum. Nekaris, A., & Bearder, S. K. (2007). The Lorisiform primates of Asia and mainland Africa. Primates in Perspective. Oxford University Press, New York, 24-45. Nyakatura, J. A. (2012). The convergent evolution of suspensory posture and locomotion in tree sloths. Journal of Mammalian Evolution, 19(3), 225-234. Orr, C. M. (2005). Knuckle‐walking anteater: A convergence test of adaptation for purported knuckle‐walking features of african Hominidae. American journal of physical anthropology, 128(3), 639-658. Owen, R. (1842). Description of the Skeleton of an Extinct Gigantic Sloth: Mylodon Robustus, Owen, with Observations on the Osteology, Natural Affinities, and Probable Habits of the Megatherioid Quadrupeds in General. R. and JE Taylor, sold by J. van Voorst. Pant, S. R., Goswami, A., & Finarelli, J. A. (2014). Complex body size trends in the evolution of sloths (Xenarthra: Pilosa). BMC Evolutionary Biology, 14(1), 184. Ryan, T. M., Silcox, M. T., Walker, A., Mao, X., Begun, D. R., Benefit, B. R., ... & Moyà-Solà, S. (2012). Evolution of locomotion in Anthropoidea: the semicircular canal evidence. Proceedings of the Royal Society of London B: Biological Sciences, 279(1742), 3467- 3475. Sarmiento, E. E. (1988). Anatomy of the hominoid wrist joint: its evolutionary and functional implications. International Journal of Primatology, 9(4), 281-345. Schmitt, D., Zeininger, A., & Granatosky, M. C. (2016). Patterns, Variability, and Flexibility of Hand Posture During Locomotion in Primates. In The Evolution of the Primate Hand(pp. 345-369). Springer New York. Seiffert, E. R. (2007). Early evolution and biogeography of lorisiform strepsirrhines. American Journal of Primatology, 69(1), 27-35. Selby, M. S., & Lovejoy, C. O. (2017). Evolution of the hominoid scapula and its implications for earliest hominid locomotion. American Journal of Physical Anthropology, 162(4), 682-700. Szalay, F. S., & Katz, C. C. (1973). Phylogeny of lemurs, galagos and lorises. Folia primatologica, 19(2-3), 88-103. Walker, A. (1969). The locomotion of the lorises, with special reference to the potto. African Journal of Ecology, 7(1), 1-5. Walker, A. C. (1974). Locomotor adaptations in past and present prosimian primates. Primate locomotion, 390, 349-382.

50 Ward, C. V. (2015). Postcranial and locomotor adaptations of hominoids. In W. Henke & I. Tattersall (Eds.), Handbook of Paleoanthropology (2 ed., Vol. 2, pp. 1363-1386). Berlin: Springer-Verlag. White, J. L. (1993). Indicators of locomotor habits in xenarthrans: evidence for locomotor heterogeneity among fossil sloths. Journal of Vertebrate Paleontology, 13(2), 230-242. Yalden, D. W. (1972). The form and function of the carpal bones in some arboreally adapted mammals. Cells Tissues Organs, 82(3), 383-406. Young, R. W. (2003). Evolution of the human hand: the role of throwing and clubbing. Journal of Anatomy, 202(1), 165-174.

51

Academic Vita of Lia Gavazzi [email protected]

Education: Major(s) and Minor(s): B.S. Biological Anthropology, B.A. Italian Language and Literature, M.A. Anthropology Honors: B.S. Biological Anthropology

Thesis Title: Variations in Pisiform Morphology Thesis Supervisor: Timothy Ryan and Philip Reno

Grants Received: Liberal Arts Enrichment Fund (Fall 2016, Spring 2017)

Professional Memberships: American Association of Physical Anthropologists (AAPA)

Presentations: Variations in Tetrapod Pisiform Ossification – Penn State Undergraduate Exhibition Variations in Tetrapod Pisiform Ossification – AAPA Undergraduate Research Symposium

International Education: Study Abroad Experience Spring 2015 semester at L’Università per Stranieri di Perugia in Perugia, Italy INTAG 499 – Experience in The Republic of Trinidad and Tobago – Summer 2014

Language Proficiency: Conversational in Italian (Fluency at C1 level)

Experiences: Penn State Blue Band – Member 2013-2016 The Penn State Women’s Varsity Fencing Team – Roster member 2014-2016 Member of the Evo-devo anthropology (Reno) research lab 2015-present