Acquisition of brains from the African elephant (Loxodonta africana): Perfusion-fixation and dissection

Page 1

AP

PDS

a

ARRA

KEBPFD

1

w(iavtn2tttpptts

0d

Journal of Neuroscience Methods 179 (2009) 16–21

Contents lists available at ScienceDirect

Journal of Neuroscience Methods

journa l homepage: www.e lsev ier .com/ locate / jneumeth

cquisition of brains from the African elephant (Loxodonta africana):erfusion-fixation and dissection

aul R. Manger ∗, Praneshri Pillay, Busisiwe C. Maseko, Adhil Bhagwandin, Nadine Gravett,on-Joon Moon, Ngalla Jillani, Jason Hemingway

chool of Anatomical Sciences, Faculty of Health Sciences, University of the Witwatersrand, 7 York Road, Parktown, 2193, Johannesburg, South Africa

r t i c l e i n f o

rticle history:eceived 8 December 2008eceived in revised form 31 December 2008ccepted 5 January 2009

eywords:volutionrainerfusionixation

a b s t r a c t

The current correspondence describes the in situ perfusion-fixation of the brain of the African elephant.Due to both the large size of proboscidean brains and the complex behaviour of these species, the acqui-sition of good quality material for comparative neuroanatomical analysis from these species is important.Three male African elephants (20–30 years) that were to be culled as part of a larger population man-agement strategy were used. The animals were humanely euthanized and the head removed from thebody. Large tubes were inserted into to the carotid arteries and the cranial vasculature flushed with arapid (20 min) rinse of 100 l of cold saline (4 ◦C). Following the rinse the head was perfusion-fixed witha slower rinse (40 min) of 100 l of cold (4 ◦C) 4% paraformaldehyde in 0.1 M phosphate buffer. This pro-cedure resulted in well-fixed neural and other tissue. After perfusion the brains were removed from the

issection skull with the aid of power tools, a procedure taking between 2 and 6 h. The brains were immediatelypost-fixed in the same solution for 72 h at 4 ◦C. The brains were subsequently placed in a sucrose solutionand finally an antifreeze solution and are stored in a −20 ◦C freezer. The acquisition of high quality neuralmaterial from African elephants that can be used for immunohistochemistry and electron microscopy is ofimportance in understanding the “hardware” underlying the behaviour of this species. This technique canbe used on a variety of large mammals to obtain high quality material for comparative neuroanatomical

studies.

. Introduction

African elephants are the largest extant terrestrial mammals,ith adult male body masses ranging between 5500 and 6000 kg

Skinner and Chimimba, 2005). Associated with this large body sizes a large brain, with reported brain masses ranging between 4000nd 6000 g (Shoshani et al., 2006). Despite this large brain mass,ery little is actually known about the structure, and thus, func-ional capacities, of the proboscidean brain. A recent review of theeuroanatomical data available for proboscidean brains (Cozzi et al.,001) reported that only 52 scientific papers have been publishedhat are specifically dedicated to structural aspects of the brain, andhat 20 of these were written in the 19th century. Moreover, 46% ofhese 52 articles were written in French, German, or Italian. Com-aratively, there is a wealth of information on the large brains of

rimates and cetaceans. It was concluded by Cozzi et al. (2001) thathe lack of interest in the proboscidean brain is: “. . .probably due tohe feeling that no ‘front line’ discovery can be derived from thesetudies. . .” and a lack of interest in support for such a study from

∗ Corresponding author. Tel.: +27 11 717 2497; fax: +27 11 717 2422.E-mail address: [email protected] (P.R. Manger).

165-0270/$ – see front matter © 2009 Elsevier B.V. All rights reserved.oi:10.1016/j.jneumeth.2009.01.001

© 2009 Elsevier B.V. All rights reserved.

funding agencies. Cozzi et al. (2001) further reason that the widegap in the amount of knowledge derived for the cetacean and pro-boscidean brain results from the military interest in the behaviourand physiology of the dolphins and whales, and the need for knowl-edge of the cetacean brain from countries that previously werewhaling countries with a specific need for commercial exploitation.

Behavioural studies of African elephants have demonstratedsome exceptional capacities, including ultra-low frequency soundcommunication (Garstang, 2004), exceptional long-term memory(Markowitz et al., 1975), very complex social structures (Payne,1998; McComb et al., 2000, 2001), and even basic tool construc-tion and use (Anderson, 2002). For the most part, these behaviouralstudies do not refer to the structure (and inferred functional capac-ities) of the brain, as the information required to make this sort ofinterpolation is just not available. Detailed neuroanatomical stud-ies of the African elephant brain will begin to unlock the neuralarchitecture subserving many of the behaviours recorded, and mayin fact provide clues pointing behavioural studies in new directions,

leading to a deeper insight and understanding of the African ele-phant. While many scientists study the behaviour of elephants, bothin the wild and in captivity, there is at present no concerted effortdirected towards understanding the structure of the brain. Tworecent studies have improved our knowledge of elephant brains,

Page 2

uroscie

ds(tfpah

cTtffWrtf

2

mAttwsN

F(it

of animals that were to be culled as part of a larger populationmanagement program to be undertaken by the Malilangwe Trust.The bodies of the three animals were subsequently butchered toprovide 100 000 meals to the people living in the region. Thus, the

P.R. Manger et al. / Journal of Ne

escribing in detail the gross anatomy (Shoshani et al., 2006) andtructural anatomy detectable with magnetic resonance imagingHakeem et al., 2005). While certain clues relating structure to func-ion have been determined, these papers have difficulty makingurther inferences due to the conditions of fixation of the tissue,ost-mortem immersion fixation between 12 and 24 h after death,nd the subsequent inability to apply techniques such as immuno-istochemistry or electron microscopy to this tissue.

We set out to obtain perfusion-fixed African elephant brains thatan be used for a variety of modern neuroanatomical techniques.he current paper describes the method we used to perfusion fixhe brains in situ and the treatment of the tissue following per-usion. We successfully obtained three well perfusion-fixed brainsrom three male African elephants in the age range of 20–30 years.

hile the method described here may have been used by otheresearchers, in other animals, previously, we are unaware of thisechnique being applied to the largest terrestrial mammal, and thuselt our experience is worth relating.

. Permits and ethical issues

Prior to undertaking this study, permission to sacrifice the ani-als was granted by the Zimbabwe Parks and Wildlife Management

uthority. Ethical permission was obtained from the University ofhe Witwatersrand Animal Ethics Committee and the animals were

reated and used according to the guidelines of this committee,hich parallel those set down by the NIH for use of animals in

cientific experiments. Permission was granted by the Malilangweature Conservation Trust to obtain African elephants resident on

ig. 1. (A) Insertion of large tubes into the carotid artery of the African elephantindicated by arrow). Note the size of the carotid artery, being approximately 15 mmn diameter. (B) Position of the perfusion tubes in relation to the larynx and body ofhe second cervical vertebra (C2).

nce Methods 179 (2009) 16–21 17

their property near Chiredzi in south-eastern Zimbabwe. Lastly,permission was obtained from the Department of Agriculture,South Africa, to allow the importation of formalin fixed tissuefrom Zimbabwe into South Africa. All work with the elephants wasperformed under the direction and supervision of an extensivelyexperienced wildlife veterinarian employed by the MalilangweTrust.

The elephants selected for use were all solitary males that wereold enough to be independent of the maternal herd, but not oldenough to be dominant breeders in the region. By selecting animalsof this type the potential effect of removing three animals from thepopulation was minimized. In addition, these animals were a subset

Fig. 2. Overall depiction of the perfusion set up. The tripod to the right enabled thelifting of the large tank containing the saline rinse or the fixative with a block andtackle. The length of each leg of the tripod was 6 m. A length of flexible plastic tubingwas taken from the bottom of the tank to the head of the elephants where this tubewas split with each split going to one carotid artery. The height of the tank in thisimage was that used for the perfusion of the fixative and allowed a slow flow offixative through the cranial vasculature.

Fig. 3. Image of the cut aspect of the neck showing both the flow of fixative throughthe cranial vasculature and the artery clamps used to seal off the bilaterally pairedvertebral arteries.

Page 3

18 P.R. Manger et al. / Journal of Neurosci

Ftl

te

3

(idtltattcstolirat

Fm

ig. 4. Cleaned posterior aspect of the skull of one of the African elephants used inhis study. This photograph is taken from behind showing the extent of the nuchaligament attachment and attachment area of the nuchal musculature.

hree elephants that were sacrificed for this study were used in theirntirety.

. Sacrifice and perfusion

The targeted animals were darted with an overdose of Scolinesuxamethonium chloride, an anaesthesia non-toxic for humanngestion) and when immobile were shot through the heart. Uponeath the head was cut free from the remainder of the body athe anterior level of the upper limb, immediately posterior to thearynx. Using large, sharp knives, one side of the neck was cut andhen, with the aid of a crane truck, the animal was turned overnd the remainder of the neck cut. A chainsaw was used to sectionhrough the second cervical vertebrae. Once the head was free fromhe body (approximately 45 min), the head was hoisted onto therane truck and driven to the area set up for perfusion. At the perfu-ion site the head was unloaded and positioned such that the top ofhe head was in contact with the ground, and the sectioned regionf the neck faced backwards and upwards. The carotid arteries were

ocated in an anatomical position superior and lateral to the larynx,nferior and lateral to the sectioned cervical vertebrae. During theemoval of the head from the remainder of the body, the carotidrteries would retract deep within the rostrally located tissue, andhus it would require a small amount of dissection to reveal these

ig. 5. (A) Posterior view of the partially dissected skull of one of the African elephants usantle of the skull that surrounds the brain. (B) Lateral view of the same.

ence Methods 179 (2009) 16–21

large arteries. Large tubes (15 mm outer diameter, 12 mm innerdiameter) were inserted into both left and right carotid arteries andsecured in place with cable ties (Fig. 1). These tubes were connectedto each other and then to a large stainless steel tank that contained100 l of cold 0.9% saline. A tap at the base of the tank served to con-trol the flow of fluid. A large cold room at the perfusion site allowedus to maintain the saline and fixative at 4 ◦C prior to use. The tankof saline was raised 5 m above the ground using a block and tackleattached to a large custom-built tripod. At a distance of approxi-mately 3 m above the ground, the saline was found to flow throughthe cranial vascular system (Fig. 2). At this stage we clamped thebilaterally paired vertebral arteries to prevent the rapid escapeof saline via these arteries and force the saline to flow out of thecranial vasculature via the jugular vein (Fig. 3). Once the vertebralarteries were clamped the saline was raised to the 5 m level. It tookapproximately 20 min for 100 l of saline to flow through the cranialvasculature and clear the head of blood (it should be noted herethat much of the blood within the head was drained during theremoval of the head from the body). Following this, the tank waslowered and 100 l of cold 4% paraformaldehyde in 0.1 M phosphatebuffer was poured into the tank, and the tank raised approximately3 m above the ground. This height was sufficient to provide a steadyflow of fixative through the cranial vasculature and provided fora longer perfusion of the head with the fixative. At this height the100 l of fixative took approximately 40 min to flow through thecranial vasculature. This procedure provided good clearance ofblood from the head and adequate fixation of the tissue.

4. Extraction of the brain

After perfusion the tissue surrounding the skull was removed.In the current instance we were collecting tissue in addition to thebrain for related studies, thus we initially removed the larynx fol-lowing which we removed the eyes. After this tissue was removedwe cut both ears and the trunk away from the head. At this stagewe were able, with some effort, to roll the head over so that it wassitting on the jaw and anterior inferior surfaces of the tusks, pro-viding a stable base. Once this was achieved, the tissue surroundingthe skull was stripped clean using large, sharp knives and scalpels.

This initially involved the removal of the 2.5 cm thick skin from thelevel of the anterior aspect of the skull down to the level of theexternal acoustic meatus. Nuchal musculature and the enormousnuchal ligament were cut away to expose the near vertical poste-rior aspect of the skull (Fig. 4). The large temporalis muscles were

ed in this study with the outer mantle and sinus bone removed to expose the inner

Page 4

uroscience Methods 179 (2009) 16–21 19

sOswcasas

aurwbbDtc

FL

Fv

P.R. Manger et al. / Journal of Ne

tripped from the skull to below the level of the zygomatic arch.nce this extra-cranial tissue was removed we could use a chain-

aw to work through the large sinus bone overlying the brain. Weere able to judge the location of the brain by following the spinal

ord through the foramen magnum. The most inferior point of thettachment of the nuchal ligament to the posterior surface of thekull was observed to mark the upper most level of the brain (in thenatomical position), and served as a valuable landmark to preventawing into the cranial cavity.

The use of a chainsaw was invaluable in the removal of the brain,s the large size of the skull makes the use of manual instrumentsnfeasible. We would saw through the skull in a piecemeal fashion,emoving approximately 20 cm of tissue at a time. By doing this weere able to locate the inner mantle of the skull that covered the

rain. The chainsaw was used to “chip-off” the remaining sinus tra-eculae from the braincase, leaving the braincase exposed (Fig. 5).uring removal of the sinus trabeculae we would on occasion cut

hrough the braincase exposing the brain. Fortunately, in theseases, the dura mater was thick enough (between 5 and 15 mm)

Fig. 6. Removal of the anterior portions of the inner cranial mantle surrounding thebrain using large rongeurs. Latex gloves were used to protect the spinal cord fromdamage and desiccation during the removal of the bone surrounding the brain.

ig. 7. (A) Posterior view of the brain of one of the African elephants used in this study with meninges in place following removal of the inner mantle of the cranial bone. (B)ateral view of the same.

ig. 8. (A) Posterior view of the brain of one of the African elephants used in this study following removal of the thick dura mater, but with arachnoid still in place. (B) Lateraliew of the same.

Page 5

2 urosci

tittpcautavot

F(cct

0 P.R. Manger et al. / Journal of Ne

o protect the brain from any damage. Once a small hole was maden the braincase, we could use large rongeurs to dissect the bone ofhe braincase overlying the frontal and temporal lobes away fromhe dura mater (Fig. 6). The braincase overlying the cerebellum andosterior aspects of the cerebral cortex was up to 10 mm thick andould not be removed using the rongeurs. To expose the posteriorspect of the brain we used the chainsaw as one would typicallyse a dental drill during a craniotomy, to make small cuts through

he bone which we could then dissect free of the underlying durand lift from the surface of the brain. This method proved to beery effective in rapidly exposing the brain, and given the thicknessf the dura mater did not lead to damage of the underlying neuralissue (Fig. 7).

ig. 9. Extraction of the brain of one of the African elephants used in this study.A) Sectioning of upper cervical spinal nerves and lower cranial nerves to free theerebellum from the cranial base. (B) Cutting of the large trigeminal and vestibulo-ochlear nerves. (C) Sectioning of the pituitary gland and olfactory bulbs to allowhe lifting of the remainder of the brain out of the cranial base.

ence Methods 179 (2009) 16–21

Once the bone of the braincase was clear from the brain we couldremove the very thick dura mater (Fig. 8). This tissue was readily cutthrough with a scalpel, but we found as we were removing the duramater that there was a lot of vascular connectivity to the underlyingarachnoid. These vascular connections were up to 2 mm diameterand needed to be cut with a scalpel to prevent damage to the neuraltissue. These vascular connections were not only found between thedura mater and arachnoid on the superior aspect of the brain (suchas the arachnoid granulations in humans), but were found at alllocations where the arachnoid and dura mater were in contact. Thedura mater forming the tentorium cerebelli was extremely deepand careful cuts through this dural fold with a sharp scalpel wasrequired to free the tentorium from the surrounding neural tissue.

Once the dura mater was removed we could lift the brain freefrom the underlying inner cranial base. We began by sectioning thespinal nerves from the upper cervical spinal cord and lifting this freefrom the underlying dura mater. We could slide our fingers under-neath the brainstem and cerebellum and gently lift the posteriorhalf of the brain (Fig. 9). As one person was doing this, another wasusing a long pair of surgical scissors to cut the cranial nerves and anyvascular connections between the dura mater and arachnoid. Thelarge trigeminal nerve (up to 10 mm diameter at the exit from thepons) proved to be the most difficult cranial nerve to section. As thecranial nerves were cut progressively more anterior aspects of thebrain could be lifted free of the inner cranial base. The large pinealstalk and optic chiasm were readily sectioned; however, the largeolfactory bulbs proved very difficult to remove together with thebrain. We found it easier to section the olfactory tract and removethe olfactory bulbs as separate entities following complete removalof the brain from the skull. The pineal gland was also removed as aseparate entity.

5. Post-fixation treatment of the tissue

Immediately upon removal from the skull (a process that tookbetween 6 h [first animal] and 2 h [third animal]) the brain wasweighed and it and the other neural tissue (olfactory bulbs andpituitary gland) were placed in 20 l of cold fixative. The three brainsobtained weighed 5145 (specimen number LA1, Loxodonta africanaone), 5250 (LA2), and 4835 g (LA3). The brains were stored in post-fix solution for 72 h at 4 ◦C, by which time a solid fixation of theentire tissue was observed (as judged by qualitative touch). Thebrains were then transferred to a cold (4 ◦C) solution of 7% sucrosein 0.1 M phosphate buffer (30 l/brain) for 5 days, by which time theyhad equilibrated in this solution. The brains were then transferredto an antifreeze solution (30 l/brain) comprised of 30% glycerol, 30%ethylene glycol, 30% distilled water and 10% 0.244 M phosphatebuffer. The brains remained in this solution at 4 ◦C for a period ofapproximately 3 weeks before they equilibrated, and at this stagethey were placed in a −20 ◦C freezer in this antifreeze solution forstorage. This process of tissue handling allows the maintenanceof antigenicity necessary for immunohistochemical staining proce-dures. When tissue is required for sectioning, the parts of interestwill be dissected free from the remainder of the brain and placed ina 30% sucrose in 0.1 M phosphate buffer solution until equilibrated,then frozen in dry ice and sectioned using a freezing microtome.

6. Discussion

The current paper describes a technique for the perfusion-

fixation in situ of the brain of large mammals. While this techniquehas been used previously (various personal communications) weare unaware of this technique being used for the perfusion of theelephant brain, and we are not aware of any formal reporting of thistechnique. The described technique has the advantage of enabling

Page 6

P.R. Manger et al. / Journal of Neuroscie

FfSS

tffpTasspbpsttwdfimtb2taait

ltab

ig. 10. (A) Low power photomicrograph of a 50 �m section of the cerebral cortexrom one of the elephants used in the current study (LA3) stained with cresyl violet.cale bar = 500 �m. (B) High power photomicrograph of nissl stained layer 3 neurons.cale bar = 100 �m.

he fixation of tissue in a manner that allows this tissue to be usedor modern neuroanatomical techniques. In this sense, specificallyor proboscideans, this is a great advantage over the tissue obtainedreviously for study (e.g. Hakeem et al., 2005; Shoshani et al., 2006).ranscardial whole body perfusion could be used for perfusion fix-tion even in the largest mammals; however, the amount of fixativeolution required to undertake transcardial perfusion and the pres-ure required to successfully perfuse a large mammal would berohibitive and impractical. That we can perfuse and post-fix therain of an animal as large as an African elephant with 120 l of 4%araformaldehyde makes the current technique very practical andhould allow for the collection of a great deal more usable tissuehat to date has not been readily accessible. One improvement onhe current technique would have been to perfuse the elephantsith twice the volume of fixative; however, given the practicalifficulties of mixing and transporting this amount of fluid to theeld, and the reasonably rapid time to immersion post-fixation, itay be more pragmatic to use a smaller amount. We have used

he currently described technique successfully to perfusion-fix therains of a black rhinoceros (Diceros bicornis, 20 l rinse and fixative,.5 m height), 3 giraffe (Giraffa camelopardalis, 10 l rinse and fixa-ive, 2 m height), 2 Scimitar-horned oryx (Oryx dammah, 10 l rinsend fixative, 2 m height), and 4 blesbok (Damaliscus dorcas, 5 l rinsend fixative, 2 m height). In these latter cases the size of the tubesnserted into the carotid arteries were smaller than those used forhe African elephants.

We could only find one prior paper describing the fixation ofarge brains in large mammals. In this paper (Knudsen et al., 2002),he authors describe the fixation of the brains of minke whales usingtechnique that fixes the brain in place but does not perfuse therain either with a rinse of saline or fixative. The disadvantage of

nce Methods 179 (2009) 16–21 21

this technique is that by not using the perfusion method, a great dealof blood will remain in the tissue. This blood will cause interferencewith many staining methods, and the length of the fixation period(up to 72 h) will lead to significant degradation of proteins withinthe brain that may be of specific interest for study.

The technique described in the current paper allows for fixa-tion of the brain in a manner that renders the specimens usefulfor a range of techniques, from the basic neuroanatomical stainsto immunohistochemistry and electron microscopy (with post-fixation in glutaraldehyde). The flushing of the blood from thebrain by the saline rinse will also augment the quality of the tissue(Fig. 10). The acquisition of high quality brain tissue from large ani-mals for comparative neuroanatomical studies is of importance forthe understanding of the evolution of large brain size and potentialassociated cognitive abilities, and will allow the field to progresswith less speculation regarding the brains of a variety of mam-mals.

Acknowledgements

The authors would like to extend their deep gratitude to thefollowing people for the assistance they rendered during the prepa-ration and undertaking of this project. In the School of AnatomicalSciences, University of the Witwatersrand, we would like to thankMr. Jacob Mekwa for his construction of the hardware requiredfor this project, and Mrs. Glynis Veale and Mrs. Alison Mortimerfor their assistance with the many administrative matters relatedto acquiring the necessary consumables. We would also like tothank Dr. Hilary Madzikanda of the Zimbabwe Parks and WildlifeManagement Authority for his untiring assistance in acquiring thenecessary permits to undertake this work in Zimbabwe. And lastlyan enormous debt of gratitude is owed to Dr. Bruce Fivaz andthe team at the Malilangwe Trust for the enthusiastic and nearunending assistance provided to us before, during and after ourstay. Without all these people and their help the current projectwould not have been the success it turned out to be. This studywas supported by a grant from the South African National ResearchFoundation to PRM.

References

Anderson JR. Gone fishing: tool use in animals. Biologist (Lond) 2002;49:15–8.Cozzi B, Spagnoli S, Bruno L. An overview of the central nervous system of the ele-

phant through a critical appraisal of the literature published in the XIX and XXcenturies. Brain Res Bull 2001;54:219–27.

Garstang M. Long-distance, low-frequency elephant communication. J Comp PhysiolA 2004;190:791–805.

Hakeem AY, Hof PR, Sherwood CC, Switzer RC, Rasmussen LEL, Allman JM. Brain of theAfrican elephant (Loxodonta africana): neuroanatomy from magnetic resonanceimages. Anat Rec 2005;287A:1117–27.

Knudsen SK, Mørk S, Øen EO. A novel method for in situ fixation of whale brains. JNeurosci Methods 2002;120:35–44.

Markowitz H, Schmidt M, Nadal L, Squier L. Do elephants ever forget? J Appl BehavAnal 1975;8:333–5.

McComb K, Moss C, Sayialel S, Baker L. Unusually extensive networks of vocal recog-nition in African elephants. Anim Behav 2000;59:103–9.

McComb K, Moss C, Durant SM, Baker L, Sayialel S. Matriarchs as repositories of socialknowledge in African elephants. Science 2001;292:491–4.

Payne K. Silent thunder: in the presence of elephants. New York: Simon and Schuster;1998. p. 228.

Shoshani J, Kupsky WJ, Marchant GH. Elephant brain. Part I. Gross morphology,functions, comparative anatomy, and evolution. Brain Res Bull 2006;70:124–57.

Skinner JD, Chimimba CT. The mammals of the Southern African subregion. 3rdedition Cambridge, UK: Cambridge University Press; 2005. p. 814.