wound healing during hibernation by black bears (ursus americanus) in the wild: elicitation of...

© 2012 ISZS, Blackwell Publishing and IOZ/CAS48

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

Integrative Zoology 2012; 7: 48–60 doi: 10.1111/j.1749-4877.2011.00280.x

ORIGINAL ARTICLE

Wound healing during hibernation by black bears (Ursus americanus) in the wild: elicitation of reduced scar formation

Paul A. IAIZZO,1 Timothy G. LASKE,1,2 Henry J. HARLOW,3 Carolyn B. McCLAY2 and David L. GARSHELIS4

1Department of Surgery, University of Minnesota, 2Medtronic, Inc., 3Department of Zoology and Physiology, University of Wyoming, Laramie, and 4Minnesota Department of Natural Resources, USA

AbstractEven mildly hypothermic body or limb temperatures can retard healing processes in mammals. Despite this, we observed that hibernating American black bears (Ursus americanus Pallas, 1780) elicit profound abilities in mounting inflammatory responses to infection and/or foreign bodies. In addition, they resolve injuries during hibernation while maintaining mildly hypothermic states (30–35 °C) and without eating, drinking, urinating or defecating. We describe experimental studies on free-ranging bears that document their abilities to completely resolve cutaneous cuts and punctures incurred during or prior to hibernation. We induced small, full-thickness cutaneous wounds (biopsies or incisions) during early denning, and re-biopsied sites 2–3 months later (near the end of denning). Routine histological methods were used to characterize these skin samples. All biopsied sites with respect to secondary intention (open circular biopsies) and primary intention (sutured sites) healed, with evidence of initial eschar (scab) formation, completeness of healed epidermis and dermal layers, dyskeratosis (inclusion cysts), and abilities to produce hair follicles. These healing abilities of hibernating black bears are a clear survival advantage to animals injured before or during denning. Bears are known to have elevated levels of hibernation induction trigger (delta-opioid receptor agonist) and ursodeoxycholic acid (major bile acid within plasma, mostly conjugated with taurine) during hibernation, which may relate to these wound-healing abilities. Further research as to the underlying mechanisms of wound healing during hibernation could have applications in human medicine. Unique approaches may be found to improve healing for malnourished, hypothermic, diabetic and elderly patients or to reduce scarring associated with burns and traumatic injuries.

Key words: black bear, denning, hair growth, healing, hibernation, histology, scarring

Correspondence: Paul A. Iaizzo, Department of Surgery, University of Minnesota, B172 Mayo, MMC 195, 420 Delaware Street SE, Minneapolis, MN 55455, USA.Email: [email protected]

INTRODUCTIONThe understanding of wound or tissue healing has pri-

marily evolved through the study of acute wounds, that is, those caused by focal trauma or induced via surgery. In general, the healing process for such injuries is well defined: (i) initial coagulation or hemostasis; (ii) inflam-

© 2012 ISZS, Blackwell Publishing and IOZ/CAS 49

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051

Wound healing in hibernating black bears

mation reaction; (iii) pronounced cell proliferation and repair of the matrix; (iv) wound epithelialization; and (v) subsequent remodeling of scar tissue. It is general-ly accepted that these steps can overlap and, under the best conditions, span numerous months (Schultz et al. 2003; Whitney & Wickline 2003; Diegelmann & Evans 2004; Arnold & Barbul 2006). If excess collagen is pro-duced during wound healing, normal tissue is replaced by distorted, non-functional, excessive accumulation of scar tissue. Chronic wounds are typically considered as a failure of the immune system to recruit the correct cells and mediate their actions for proper healing, so the damaged tissues remain in the inflammatory or cell pro-liferation stages of healing (Mangram et al. 1999; Whit-ney & Wickline 2003; Arnold & Barbul 2006). In other words, chronic non-healing wounds are characterized by defective remodeling of the associated extracellular ma-trix, failure of the epidermis to migrate to ensure closure and/or prolonged inflammation.

Humans have used ‘heat’ in various forms to promote wound healing since the beginning of recorded medical history. Hippocrates noted this: “Wounds love warmth; naturally, because they exist under shelter; and natu-rally they suffer from the opposite” (Majno 1975; Rab-kin & Hunt 1987). Experimental studies have supported the value of localized hyperthermia in treating wounds and minimizing infections; i.e., therapeutic heating has been associated with increased focal blood flow, growth factors, nutrients and oxygen tension (Rabkin & Hunt 1987; Sheffield et al. 1996; Alvarez et al. 2003). More specifically, focal heating has been reported to enhance healing by activating biochemical and enzymatic reac-tions, increasing the functional availability of immune cells and altering inhibitory factors in the local wound environment (Xia et al. 2000; Park et al. 2001; Whitney & Wickline 2003). Furthermore, focal radiant warming bandages used clinically have shown promise for treat-ing chronic ulcers (Robinson & Santilli 1998; Alvarez et al. 2003), yet note that such treatments often employ other pharmacological means to optimize outcomes.

Our observations related to wound healing abili-ties of hibernating American black bears may provide novel translational insights regarding how to promote wound healing during mild hypothermia (either whole body or limb) and/or prolonged nutritional depriva-tion. Hibernating (or overwintering) black bears may re-main in winter dens for up to 7 months, a period during which their core body temperatures are mildly hypo-thermic (30–35 °C), and they do not eat, drink, urinate or defecate (Hellgren 1998). Due to lower body temper-

ature, reduced metabolism, extreme overwinter weight loss and complete lack of protein intake (Harlow et al. 2002, 2004; Lohuis et al. 2005), it might be expected that the wounds of denning bears, such as those incurred just prior to hibernation, would not easily heal. In part, this assumption is based on decreased rates of wound healing and/or abnormal foreign body rejection respons-es in hypothermic humans (Beilin et al. 1998; Remick & Xioa 2006; Sessler 2006). Inadequate wound healing in humans leads to impaired quality of life, pain, loss of work productivity and costly treatments. Notably, inci-dences of chronic wounds are predicted to increase in developed countries due to the growing population of older individuals (Gerstein et al. 1996; Thomas 2001).

There are limited reported studies on the wound heal-ing abilities of hibernating mammals. Yet it has been re-ported that during hibernation, immune responses tend to be inhibited (Prendergast et al. 2002), collagen for-mation is depressed (Guth et al. 1981) and blood clot-ting times are significantly increased; the latter possi-bly occurs to prevent thrombosis that might result from a compromised (slowed) circulation (Lechler & Penick 1963). In a controlled study performed on hibernat-ing ground squirrels (Citellus lateralis), Billingham and Silvers (1960) found severe impairment of cutaneous wound healing, which was then regained upon arousal from torpor (note that these animals become deeply hy-pothermic during prolonged periods of inactivity or hi-bernation). Mori et al. (2008) reported that if the expres-sion of a protein called osteopontin in a knockout mouse model was altered, neutrophil, macrophage and mast cell expression were changed, and reduced scar forma-tion resulted during healing.

Given these findings, it was an unexpected obser-vation by Laske and colleagues that hibernating bears elicited remarkable abilities to expel recently implant-ed devices; it was also observed that these animals elic-ited rapid healing of all associated wound sites (Laske et al. 2005). These and other observations regarding the wound healing abilities of bears spurred the present bi-opsy experiments. In other words, these studies were undertaken to specifically document the progression of histological changes in monitored wound sites and to test the hypothesis that healing readily occurs during hi-bernation while these animals are hypothermic and fac-ing prolonged nitrogen deprivation. Again, it should be noted that this behavior is counter to conventional wis-dom regarding effects of nutritional status and temper-ature on wound-healing abilities in warm-blooded ani-mals (Arnold & Barbul 2006).


123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

P. A. Iaizzo et al.

MATERIALS AND METHODSWe performed multi-year investigative studies on

free-ranging black bears during annual periods of early denning (November–December, 1–2 months after they entered dens for hibernation) and late denning (March, less than 1 month before their emergence) in northern Minnesota. Animals were tracked to their dens via sig-nals from their radio-collars. They were then anesthe-tized (Telazol, Fort Dodge Animal Health, Fort Dodge, Iowa, USA; 4.4 mg/kg) and temporarily removed from their dens to assess their general health status and also to perform various studies (Table 1). These studies con-formed to the Guide for Care and Use of LaboratoryAnimals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996) and followed American Society of Mammalogist guidelines. All studies were conducted in conjunction with the Min-nesota Department of Natural Resources, and were ap-proved by the University of Minnesota’s Institutional Animal Care and Use Committee.

We induced small cutaneous wounds on 14 different bears during 4 winter seasons; 2 bears were studied for more than 1 winter (Table 1; note that data from distal

leg biopsies are not included in the table). No antibiotics were administered to any of these animals, and all were free from other injuries and/or signs of disease. The ini-tial preparation of the sites and tissue sample handling protocols were similar for all biopsy experiments. Brief-ly, 3–4 cm diameter areas of skin were exposed in either the pectoral regions of both right and left shoulders or the outer area of the right ankle by trimming and shav-ing the fur (Fig. 1). Small transcutaneous wounds were created in early winter by biopsy punch (model 33-30, Miltex, Tuttlingen, Germany) or incision. Late winter samples of the wound site were taken with the same bi-opsy punch. Biopsy samples were transferred into indi-vidually labeled cassettes and immediately immersed in a solution of 10% neutral buffered formalin.

We modified portions of the specific procedures each year, based on insights gained from the results of pre-vious years. The protocol for the first hibernation sea-son (winter 2001–2002) involved 5 animals in which we took 2 5-mm diameter full-thickness (through the epi-dermal and dermal layers down to the facial plane) bi-opsy punches from each shoulder. On each shoulder, 1 site was left untreated (open), while the second was closed with a Dacron suture. These 4 initial skin biop-

Table 1 Animal status and study information for Methods A, B and C

Animal #Age

(years)Gender

Study period (days)

Mass: early winter (kg)

Mass: late winter (kg)

Body fat: early

winter (%)

Body fat:late winter

(%)

Core temp: early winter

(°C)

Core temp: late winter

(°C)

Method AA1 923 12 Female 67 100 91 34 15 33 ndA2 2013 11 Female 76 113 98 33 31 nd 35A3 1903 2 Female 74 43 40 27 25 33 ndA4 1806 3 Female 74 57 52 33 24 30 34A5 1801 3 Female 88 53 47 30 17 33 nd

Method BB1 1903 3 Female 67 56 50 23 13 35 32B2 2007 2 Female 67 36 32 27 22 34 31B3 2064 4 Male 95 151 133 39 19 35 36B4 739 15 Female 77 124 97 32 27 35 36

Method CC1 2007 3 Female 58 56 50 33 28 30 33C2 2101 6 Male 76 133 121 33 31 31 ndC3 2102 10 Male 69 218 183 41 32 32 ndMethod A (winter 2001–2002), Method B (winter 2002–2003), Method C (winter 2003–2004). Mass was measured on a hanging scale, body fat was estimated by bioelectrical impedance (Farley & Robbins 1994), and temperature was measured in the rectum (nd = no data).


123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051


sy samples from each animal served as baseline controls for histological comparison with later healing tissue. In late winter, sites were harvested by taking 5 mm diame-ter full-thickness skin biopsies from both sutured (cen-tered over the suture) and non-sutured sites (overlapping 50% with outside tissue; Method A, Fig. 2).

The next hibernation season, we attempted to im-prove our ability to re-identify the initial wound target sites after healing. Four animals had a single 3 mm di-ameter full-thickness skin biopsy taken from both right and left shoulder areas. Then, 2 Dacron sutures were placed through the undisturbed skin at equal distanc-es superior and inferior to the initial wounding site, to serve as site markers and potential foreign bodies that could elicit an inflammatory response in otherwise unin-jured tissue. A permanent marker was also used to aid in site identification. In late winter, the previously applied full-thickness skin removal and sutured sites were sam-pled with a 5 mm biopsy tool (Method B, Fig. 2).

In the third denning season, we created 1 cm full-thickness skin incisions on the right and left pectoral ar-eas of 3 study animals. Cranial and caudal margins of the linear wound were marked by tattoos. Late winter, 5 mm diameter full-thickness biopsies were centered across the incision, equidistant from the tattoos (Method C, Fig. 2).

In the last winter of these studies (2004–2005), we took biopsy samples from the skin over the outer ankle of 4 bears, to test whether the colder extremity region would have poorer wound healing (Method D). We only re-examined these wounds visually in late winter, i.e. we did not assess them histologically.

Figure 1 Shoulder regions shaved for biopsy sites with images of Methods B and C inset.

Figure 2 Biopsy sites (Methods A, B and C). Method A (2001–2002): 2 punch biopsies taken from each shoulder during early win-ter, with 1 closed using Dacron suture; biopsies overlapping original site were obtained in late winter. Method B (2002–2003): 1 punch biopsy obtained from each shoulder in early denning, with sites marked by sutures and permanent marker; both original bi-opsy and suture sites were sampled. Method C (2003–2004): incision made in early winter with site marked by tattoos; biopsy was taken from incision zone in late winter.


123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

P. A. Iaizzo et al.

All excised tissue samples were processed using rou-tine histological methods, i.e., embedded in paraffin, cut into 3–5 µm thick sections, mounted on glass slides, and stained with hematoxylin, eosin and Masson’s trichrome stains (Fig. 3). Samples were then evaluated by light mi-croscopy and assessed for multiple histological parame-ters, including: (i) thickness of epidermis; (ii) compact-ness of dermal collagen layers (type I); (iii) severity of inflammatory response; (iv) distribution of inflammato-ry cells; (v) types of inflammatory cells; (vi) presence or absence of foreign materials; (vii) presence of bac-teria; and (viii) occurrence of skin appendages (Table 2). Overall severity of the inflammatory responses was semi-quantitatively scored: 0 if absent, and 0.5 through 4 if present (in increasing order as trace, minimal, mild, moderate or marked; Table 2). The study pathologist an-alyzed samples at the end of each winter season while blinded with respect to animal identity, tissue sample type and biopsy date.

We conducted additional observational studies on several denning bears that presented with cutaneous in-juries not attributable to our experiments. We photo-graphed these wounds and obtained infrared images (IRISYS Multi-Purpose Thermal Imager IRI 4010, In-fraRed Integrated Systems, Northampton, UK).

RESULTS

Gross observations on wound healing during hibernation

In our cumulative experience of observing more than 1000 radio-collared bears in winter dens over a period

of 25 years, we have identified a few animals each year with injuries resulting from: (i) gunshots or arrows from hunters; (ii) bite marks from other bears or predators; (iii) cuts of unknown origin; or (iv) lacerations from ra-dio collars (e.g. due to extreme weight gain of more than 100%). These wounds were considered to have been in-curred some time before the bears denned, and were often infected and/or inflamed when observed in ear-ly winter. Yet typically, when we revisited bears in their dens a few months later, most wounds had complete-ly resolved whether or not we debrided the wounds, su-tured the areas and/or administered antibiotics (Fig. 5).

We did not observe regrowth of hair in any body re-gion that was shaved during early winter den visits (Figs 1 and 4). This was true regardless of body loca-tion (neck, shoulder, chest, back or distal leg) or type of study performed (control, punch biopsy, full-thickness incision with or without suture; Figs 4 and 5). However, we commonly observed a fine line of hair regrowth af-ter the healing of larger wounds (i.e. those unrelated to our experiments; Fig. 5). The healing process for these wounds might have generated more localized heat to the specific skin region and, therefore, stimulated more fol-licle development than at the sites of the smaller wounds that we experimentally created (Figs 4 and 5).

Healing of cutaneous areas after biopsy

The mean length of time from the initial biopsy pe-riod to subsequent wound assessment of the 16 experi-mental animals was 75 days. Note that 2 animals were counted twice, because they were enrolled in two differ-ent study protocols: Bear 1903 in Methods A and B, and Bear 2007 in Methods B and C (Table 1). The mean du-ration between creation and examination of wounds for study protocols A, B, C and D was 76, 77, 68 and 75 days, respectively (range 58–95 days).

At the conclusion of the first year of the Method A study, complete healing of the epidermal layers was ob-served in all 5 animals; note that visual identification of the previously wounded tissue sites in late winter was very difficult. Thus, localization techniques were add-ed to more clearly demarcate target tissue sites in sub-sequent denning seasons (i.e. marking sutures and per-manent markers [Method B] and/or tattoos [Method C]). During tissue sample harvesting at the late denning time points, complete healing of the epidermis at the wound sites was observed in the 4 (Method B) and 3 (Method C) animals, respectively.

No hair regrowth was observed on any of the shaved regions subsequent to all methods of wounding: 5 mm

Figure 3 Typical examples of histological slides that were evaluated and scored in the same biopsy specimen: (a) hema-toxylin and eosin stained (HE) slide (4×) and (b) Masson’s tri-chrome slide (4×).

a b


123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051


Tabl

e 2

His

tolo

gica

l res

ults

of s

ites w

ith fu

ll-th

ickn

ess i

ncis

ions

(Met

hod

C)

Ani

mal

Site

Wou

nd/b

iops

y si

te

Com

men

tsEp

ider

mis

thic

knes

sD

erm

is

colla

gen

Infla

mm

ator

y re

spon

seFo

reig

n m

ater

ial

Seve

rity

Dis

tribu

tion

Com

posi

tion

C1

2007

Left

Inta

ctD

WC

T0

NA

NA

0H

air f

ollic

les,

adne

xaR

ight

D

isru

pted

DW

CT

with

cys

tic/

disr

upte

d in

clus

ion

2M

f, de

rmal

Mac

ro-p

hage

sLy

mph

o-cy

tes

>PM

N>>

FBG

C

Larg

e ep

ider

mal

tear

ing

artif

act;

mul

tiple

foci

of

fore

ign

body

resp

onse

in d

erm

is; h

air f

ollic

les,

adne

xa

C2

2101

Left

Inta

ctD

WC

T0

NA

NA

0H

air f

ollic

les,

adne

xa, a

dipo

seR

ight

Fu

ll-th

ickn

ess f

ocal

cy

stic

spac

e in

to

derm

is

DW

CT

with

cy

stic

spac

e co

ntig

uous

w

ith e

pide

rmis

0N

AN

A0

Sect

ion

tear

ing

artif

act;

hair

folli

cles

, adn

exa;

ke

ratin

cys

t

C3

2102

Left

Non

e pr

esen

tD

WC

T0

NA

NA

0H

air f

ollic

les,

adne

xaR

ight

In

tact

with

foca

l te

arin

g ar

tifac

t, la

rge

intra

epid

erm

al

incl

usio

n-se

ptic

m

icro

absc

ess,

foca

l hy

perk

erat

osis

&

dysk

erat

osis

DW

CT

2f,

intra

-ep

ider

mal

PMN

Irre

gula

r bl

ack

or

wax

y go

lden

am

orph

ous

form

ing

sphe

rule

s or

elon

gate

Sept

ic m

icro

absc

ess w

ith c

olon

ies r

esem

blin

g St

aphy

loco

ccus

and

fore

ign

mat

eria

l inc

lusi

on;

hair

folli

cles

, adn

exa

Adn

exa,

spe

cial

ized

stru

ctur

es a

risin

g fr

om th

e ep

ider

mis

; DW

CT,

den

se w

hite

con

nect

ive

tissu

e; d

yske

rato

sis,

pre

mat

ure

kera

tiniz

atio

n of

indi

vidu

al e

pide

rmal

cel

ls;

hype

rker

atos

is, i

ncre

ased

thic

knes

s of s

tratu

m c

orne

um (a

nucl

ear l

ayer

of e

pide

rmis

); N

A, n

ot a

pplic

able

; sep

tic, p

atho

geni

c or

gani

sms i

n tis

sue;

f, fo

cal;

mf,

mul

ti-fo

cal;

FBG

C, f

orei

gn b

ody

gian

t cel

ls, P

MN

, pol

ymor

pho-

nucl

eocy

te.


123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

P. A. Iaizzo et al.

sites (Table 3). Statistically significant differences in in-flammation existed between each characterized group (P < 0.05): control samples less than no suture sites; control less than suture sites; and no suture less than su-ture sites (as classified by the study pathologist).

Non-haired skin appeared intact, smooth and without discoloration or scar formation at all wound and suture sites. Of the 34 wound samples evaluated from the 10 animals in Methods A–C, there was histologic evidence of reepithelialization in 29 biopsy sites. Slides from 3 samples could not be fully evaluated owing to epithelial tearing as a sectioning artifact during histologic process-ing. The epithelium was missing on 2 slides of wound sites such that sampling deficiency, processing artifact or ineffective healing could not be differentiated. His-tologic evidence of focal epidermal hyperplasia was ob-served in low numbers of control and post-wounding biop-sies, including: 3 of 30 pre-wounding control sites; 4 of 24 post-wounding sites without suture closure; 2 of 10 post-wounding sites closed with sutures; and 4 of 8 suture only sites without wounding. In addition, in 1 case (Bear 1801), post-wounding sites were re-biopsied in summer (July), more than 200 days after initial biopsy. In all 4 sam-ples, intact epidermis showed mild hyperplasia, whereas this finding was observed in only 1 of these sites previ-ously. Incidence of epidermal dysplasia was found mi-croscopically in 4 samples each from control sites and post-wounding without suture biopsies and in 2 samples

Figure 4 Bilateral cutaneous areas shaved and punch biopsied using Method A. Photographs were taken in late winter just prior to biopsying for histological assessment. White arrows denote sutured sites, and black arrows are non-sutured sites. Note the lack of hair growth. Bear 1806 exhibited the greatest inflammation at a non-sutured site, of all sites investigated (black arrow on right side; scored as ‘4’ on Table 3).

punch biopsies with or without sutures, 3 mm punch bi-opsies without sutures, or 1 cm full-thickness excisions without sutures. Yet, despite the lack of grossly visi-ble hair regrowth at 2–3 months post-wounding, there was histologic evidence of hair follicles present in 20 of 42 biopsy samples. Lack of hair follicles found in biop-sy samples might not necessarily be attributed to incom-plete healing; rather, other factors in biopsy sampling (size and depth of tissue) and histologic sectioning pro-cess might limit observance of follicles.

Shaved hair eventually grew back after denning. For Bears 1903 and 2007, which were studied for 2 con-secutive years, we re-shaved the shoulder areas to cre-ate the wound in the second year. Several bears were trapped during May and June, at which time we ob-served that all shaved hair had regrown (i.e. we could not detect any prior areas of hair removal).

Injury sites progressed completely through the in-flammation and proliferation stages during hibernation with: (i) minimal signs of granulation or infection; (ii) detectable conversion of type III to type I collagen; (iii) remodeling of the dermal layers, including the for-mation of new hair follicles; and (iv) contraction of the wound sites, but with minimal expression of scar-ring (Tables 2 and 3, Figs 4 and 5). It should be noted that the biopsy sites that were closed with Dacron sutures elicited more inflammatory and errythemic responses (i.e. foreign body responses; 89% of sites) than unsutured


123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051


Figure 5 (a) Injury incurred before denning (Bear 2081), first observed in December (left) and after healing 12 weeks lat-er (right). This animal’s core body temperature was 35.4 °C in December and 32.8 °C in March. (b) Paired infrared imag-es from those shown in (a), with the temperature scale of 25 °C (black) to 35 °C (white). Subsequent to initial December pho-tos (a, left), a local subcutaneous injection of lidocaine was ad-ministered along the injury border, the injury was debrided, and single sutures were used to close the skin edges (approx-imately 1 cm spacing). (c) Close-up view before (left) and af-ter (center) suturing, as well as after a healing period (right). In the late winter photos (right), a small line of new hair can be detected (1–3 mm wide), just along the line where the skin was sutured.

from suture only sites. Epidermal inclusion cysts were observed microscopically in 1 case from each of the 3 study protocols post-treatment; samples from 2 animals (Bear 2013 and 2064) were harvested at suture sites, while the sample from Bear 2101 was harvested at a site without sutures. There was no inflammation associated with the inclusion cysts and 2 samples contained kera-tin.

In general, there was minimal gross evidence of scab formations or superficial infections in any of the 10 ani-mals in Methods A–C (Fig. 4). More specifically, histo-logic evidence of eschar (scab) formation with acute in-flammation was present in only 2 cases: Bear 2007 at

2 sites, 1 with suture only and 1 post-wounding with-out suture closure; and Bear 1903 at 1 site post-wound-ing without suture. Histologic examination of tissue sections confirmed only bacterial colonies, morpholog-ically resembling Staphylococcus spp., primarily in the epidermal layers of 10 wound samples from 7 animals. Of the 5 animals in Method A, 4 developed detectable superficial infections at post-wounding sites closed with suture, and 1 animal also developed an infection at a post-wounding site without suture closure (note that such results were observable histologically, but not visually) . Of the 4 animals in Method B, 3 had evidence of infections in 4 samples, including 2 post-wounding sites not closed with sutures and 2 suture only sites without wound-ing. Of note is an additional sample from a suture only site revealing microabscess formation, but lacking de-tectable bacteria. One of 3 animals in Method C had a scored infected post-wounding site not closed with su-ture. None of the 34 control samples appeared infected. Visible evidence of eschars was present in 3 of 4 wound sites in the Method B animals (Fig. 6).

The presence of chronic or mixed chronic active in-flammation was variably present in both control and treatment samples. Of the 24 control sites in Method A, 5 showed very minimal to mild inflammation, primari-ly within the dermis. For the treatment sites in Method A, 2 of 10 post-wounding samples without suture clo-sure had very minimal inflammation, while 7 of 10 post-wounding samples with suture closure elicited mini-mal to marked chronic or chronic active inflammation. For the treatment sites in Method B, 6 of 8 post-wound-ing with suture closure and 7 of 8 suture only samples showed minimal to moderate chronic or chronic active inflammation. For the treatment sites in Method C, 1 of 6 post-wounding samples without suture closure showed moderate chronic inflammation.

In Methods A, B and C, all biopsies were taken from the pectoral regions of the animals where skin temper-atures tended to be relatively warm due to blood circu-lation and the curled position of the denning bear, how-ever all were still considered to be mildly hypothermic (Table 1). In the last series of studies on wounds within the ‘colder extremities’ (Method D), we did not perform a second series of biopsies of the wound sites on the an-kles, but visually these appeared to heal as well as the shoulder sites. Core temperatures of these animals were 34.1 + 1.2 °C at the December den visits and 34.5 + 1.9 °C during March visits; from infrared imaging, we found that these biopsy sites were 4–6 °C lower than the animals’ core temperatures (Fig. 6).

a

b

c


123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

P. A. Iaizzo et al.

DISCUSSIONOur observations indicate that American black bears,

hibernating in the wild with mildly hypothermic core temperatures, are routinely able to fully heal wounds and elicit dramatic foreign body responses. These findings are consistent with our previous reports of wound heal-ing in hibernating bears after they expelled subcutaneous-ly-implanted devices through the skin (Laske et al. 2005, 2010, 2011). Here, we observed that biopsied sites, with respect to both secondary intention (non-sutured) and primary intention (sutured), healed with: 1) minimal evi-dence of either eschar formation or scarring ; 2) minimal

dyskeratosis (inclusion cysts); 3) replaced epidermis and dermal layers (subsequent remodeling); and 4) abilities to produce hair follicles. This continued healing ability of wild black bears during hibernation (i.e. in the face of mild hypothermia and nitrogen restriction) provides added insights regarding the adaptation of these ani-mals and might spur further investigations to discover the unique biological mechanism(s) for this profound wound healing.

Others investigators (Echols et al. 2004; Jessup & Koch 1984) had previously observed expulsion of sub-cutaneously-implanted devices (transmitters) in bears during their active season (spring–autumn). Important-ly, such responses are in stark contrast to those in hu-mans and previously studied research animals (e.g. dogs, sheep, swine, calves and primates, all non-hiberna-tors) that have been implanted with pacemakers and/or de-fibrillators. In other words, in these other species, such devices stay within a formed fibrotic tissue pocket for decades, with relatively few reported complications. Typically, these implanted devices are contained within hermatically-sealed alloy canisters specifically designed to cause minimal foreign body responses and conse-quent rejections (Byrd 2000; Byrd & Wilkoff 2000).

It is generally accepted that for full-thickness cuta-neous wounds to heal optimally, the patient’s core tem-perature and/or the temperature surrounding the wound area should be 37 °C or above. Thus, in many major surgeries, medical devices are used that increase a pa-tient’s body temperature, to reduce risks of post-surgi-cal site infection and to promote healing (Mangram et al. 1999). Furthermore, although therapy for cutaneous burns might be initially aided by focal cooling, this pro-cedure has been reported to inhibit early wound repair (Esclamado et al. 1990; Martineau & Shek 2005). Be-yond focal temperature responses, if a patient becomes hypothermic unintentionally, such as during surgery (a common complication of anesthesia), numerous ad-verse effects on several systems might occur, including increased wound healing times and infection rates (Bei-lin et al. 1998; Remick & Xioa 2006; Sessler 2006). Ac-cordingly, individuals who are slightly hypothermic due to hypothyroidism commonly elicit decreased wound healing abilities (Ladenson et al. 1984). Again, this con-trasts with our observations of healing in hibernating bears whose core temperatures were typically 3–4 °C below their summer active temperatures.

The observation that hibernating bears rejected en-cased foreign objects and that the wounds fully healed seems remarkable. However, it is clearly adaptable for

Table 3 Severity scores of inflammation assessed in late winter for injuries imposed in early winter

Inflammation severity (0 – 4)

SampleDuration

(days)Control

No suture

Suture

Method AA1 923 67 0, 0, 0, 0 0, 0 1, 1A2 2013 76 0, 0, 0.5, 0.5 0, 0 0.5, 4A3 1903 74 0, 0, 0, 0 0, 0 1, 4A4 1806 74 0, 0, 0, 0 0, 4 2, 4A5 1801 88 0, 0, 0, 0 0, 0 0, 3

Method BB1 1903 67 0, 0 1, 3 3, 4B2 2007 67 0, 0 0, 4 1, 4B3 2064 95 0, 0 1, 1 0, 1B4 739 77 0, 0 0, 2 1, 2

Method CC1 2007 58 — 0, 2 —C2 2101 76 — 0, 0 —C3 2102 69 — 0, 2 —Average 74 0.04 0.83 * 2.03 **,***

Control samples were obtained from initial tissue biopsies (none for method C, Fig. 2) and undisturbed tissue biopsied in late winter. Late winter biopsies included skin that was previously cut or biopsied and skin that was sutured (Fig. 2). Duration, time between imposed injury in early winter and re-examination. Scoring for severity of inflammation: 0 = none; 0.5 or 1 = minimal; 2 = mild; 3 = moderate; 4 = marked. Statis-tically significant differences existed between each character-ized group (P < 0.05): *control versus no suture sites; **con-trol versus suture sites; ***no suture versus suture sites.


123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051


Figure 6 Digital images showing a representative example of an ankle biopsy site (Method D) at the den visit during early (De-cember, a) and late (March, b) hibernation. Shown in (c) are in-frared images from those shown in (a); these are infrared images of this ankle site in March. Temperature scale was 20 °C (black) to 30 °C (white).

an animal that spends half the year in a state of hiberna-tion to be able to continue to heal wounds incurred just prior to hibernation. It should be noted that we have ob-served very few chronic wounds in wild bears, even in

those animals that have been observed for over a de-cade. Bears are also known to produce elevated levels of hibernation induction trigger (delta-opioid receptor agonist) and ursodeoxycholic acid (major bile acid with-in plasma, mostly conjugated with taurine) during hiber-nation (Sola et al. 2008). It is considered that increased circulating concentrations of such hormones/molecules might play a role in the healing abilities of bears during hibernation. More specifically, it has been hypothesized that seasonal variations in plasma bile acid composition might explain the possible role that ursodeoxycholic acid has in cellular protection in these hibernating mam-mals (Sola et al. 2006).

The typical response to an injury by a hibernating small mammal is to fill the site with granulation tissue, after which the wound is reepithelialized (Billingham & Silvers 1960). Notably, during hibernation, these small mammals typically arouse every 10–20 days for sev-eral hours, warm up, and then go back into deep hypo-thermia (Carey et al. 2003). Non-mammalian poikilo-thermic vertebrates, such as snakes and fish, assume the cool temperature of their environment with concomi-tantly low metabolism where injuries do resolve, but at much reduced rates (Anderson & Roberts 1975; Smith et al. 1988). It has been reported that the rates of the wound healing process vary linearly with temperature (e.g. between 10 and 30 °C) (Anderson & Roberts 1975; Smith et al. 1988). Accordingly, in homeotherms, heal-ing rates of wounds in an extremity (e.g. horse limb) ex-posed to the cold are delayed compared to limbs at high-er ambient temperatures (Smith 1973). In contrast, here we describe that wounds healed with reepithelialization and minimal scar formation over short time periods in hibernating bears. This was true even in their extremi-ties, which were 4–5 °C colder than their already mildly hypothermic core temperatures.

The state of hibernation entails many molecular, cel-lular and physiological adaptations to protect the body, i.e., during this extended period without nourishment (Boyer & Barnes 1999; Carey et al. 2003). Additional-ly, numerous adaptations of bone and muscle physiolo-gy of hibernating bears have been described previous-ly (Milbury et al. 1998; Harlow et al. 2001; Donahue et al. 2003, 2006a; Lohuis et al. 2005; Lohuis et al. 2007). However, special protective adaptations of the skin in bears or any other mammal are not well understood (Ruben 1982). Unlike previous reports of other hiberna-tors (Ruben 1982), we did not observe the regrowth of shaved hair in these hibernating black bears. Rather, re-growth of hair occurred some time shortly after hiber-

a

b

c


123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

P. A. Iaizzo et al.

nation. It appears that hibernating bears focus their lim-ited energy resources on regenerating the tissue types most important for their survival. Our observations that hibernating bears are able to maintain wound healing abilities are consistent with numerous reports indicat-ing that hibernating bears also maintain bone tissue and muscle masses. For example, recent studies by Fedorov and co-workers (2009, 2011) suggest that the differen-tial modulations of gene expression are associated with these abilities. In other words, reduced body tempera-tures associated with hypothermia in humans (e.g. in the surgical or diabetic patient) will have very different physiological consequences compared to the controlled reduction in body temperature that is associated with black bear hibernation.

There is a clear survival advantage to wound healing during hibernation, as bears unable to do so would like-ly suffer from: (i) critical loss of body fluids; (ii) great-ly increased metabolic demands and additional weight loss; and/or (iii) toxicity from resultant infections dur-ing these periods of depressed renal function. We do not know whether these impressive healing abilities are any better than those expressed by bears during their ac-tive seasons, as we have no comparative data; howev-er, we believe that this does not diminish the uniqueness of the finding with respect to hibernation. We specu-late that several compounding factors facilitate this ad-aptation in bears, including: (i) greater mass and ther-mal inertia compared to other hibernators; (ii) higher hibernating body temperature; and (iii) unique adapta-tion to winter birthing (which requires healing proper-ties). Perhaps, most importantly, hibernating bears have also been shown to elicit daily episodes of elevated skin temperature, which might be associated with increases in peripheral blood flow, thus carrying nutrients and ox-ygen to the potential wound sites (Harlow et al. 2004). We have measured periodic bouts of muscle electro-myographic EMG activity with power spectrums indic-ative of isometric, isotonic and shivering activity (data not published). These muscle contractions could pro-duce elevated skin temperature, but not dramatical-ly increase core body temperature (Harlow et al. 2004). Muscle contractions might not only facilitate the reten-tion of skeletal muscle tone during hibernation (Harlow et al. 2001), but also enhance peripheral vasodilation and a vigorous supply of blood to the skin, which could be an aid to wound healing. Other mechanisms for these unique healing abilities might include elevated levels of circulating hormones or proteins and/or increased lev-els of plasma bile acids (e.g. ursodeoxycholic acid) that

have been observed in denning bears (Ruit et al. 1987; Bolling et al. 1997; Sola et al. (2006). These compounds have shown promise in benefiting other human medical applications (Chien et al. 1989; Rodrigues et al. 1999). In this regard, we believe that further investigation into wound healing processes in bears might provide unique insights relative to novel techniques or biological mate-rials that could be employed for treating wounds in hu-mans (perhaps even with reduced scarring); this would be especially important for malnourished, hypothermic, diabetic and/or elderly patients.

ACKNOWLEDGMENTSWe thank PL Coy, KV Noyce, JR DeJong, and BJ

Dirks for their technical support, MA Mahre for assis-tance with manuscript preparation, and the Camp Ripley National Guard Training Site.

REFERENCES Alvarez OM, Rogers RS, Booker JG, Patel M (2003).

Effect of noncontact normothermic wound therapy on the healing of neuropathic (diabetic) foot ulcers: an interim analysis of 20 patients. Journal of Foot and Ankle Surgery 42, 30–35.

Anderson CD, Roberts RJ (1975). A comparison of the effect of temperature on wound healing in a tropical and temperate telost. Journal of Fish Biology 7, 173–82.

Arnold M, Barbul A (2006). Nutrition and wound heal-ing. Plastic and Reconstructive Surgery 117, 42S–58S.

Beilin B, Shavit Y, Razumovsky J, Wolloch Y, Zeidel Z, Bessler H (1998). Effects of mild perioperative hypo-thermia on cellular immune responses. Anesthesiolo-gy 89, 1133–40.

Billingham RE, Silvers WK (1960). A note on the fate of skin autografts and homografts and on the healing of cutaneous wounds in hibernating squirrels. Annals of Surgery 152, 975–86.

Bolling SF, Tramontini NL, Kilgore KS, Su TP, Oeltgen PR, Harlow HJ (1997). Use of ‘natural’ hibernation induction triggers for myocardial protection. Annals of Thoracic Surgery 64, 623–7.

Boyer BB, Barnes BM (1999). Molecular and metabol-ic aspects of mammalian hibernation. BioScience 49, 713–24.

Byrd CL (2000). Management of implant complications. In: Ellenbogen KA, Kay GN, Wilkoff BL, eds. Clin-


123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051


ical Cardiac Pacing and Defibrillation. W.B. Saun-ders Co., Philadelphia, PA, pp. 669–94.

Byrd CL, Wilkoff BL (2000). Techniques and devic-es for extraction of pacemaker and implantable car-dioverter defibrillator leads. In: Ellenbogen KA, Kay GN, Wilkoff BL, eds. Clinical Cardiac Pacing and Defibrillation. W.B. Saunders Co., Philadelphia, PA, pp. 695–705.

Carey HV, Andrews MT, Martin SL (2003). Mammalian hibernation: cellular and molecular responses to de-pressed metabolism and low temperature. Physiolog-ical Reviews 83, 1153–81.

Chien S, Diana JN, Oeltgen PR, Todd EP, O’Connor WN, Chitwood Jr WR (1989). Eighteen to 37 hours’ preservation of major organs using a new autoperfu-sion multiorgan preparation. Annals of Thoracic Sur-gery 47, 860–67.

Diegelmann RF, Evans MC (2004). Wound healing: an overview of acute, fibrotic and delayed healing. Frontiers in Bioscience 9, 283–9.

Donahue SW, Vaughan MR, Demers LM, Donahue HJ (2003). Bone formation is not impaired by hiberna-tion (disuse) in black bears Ursus americanus. Jour-nal of Experimental Biology 206, 4233–9.

Donahue SW, Galley SA, Vaughan MR et al. (2006a). Parathyroid hormone may maintain bone formation in hibernating black bears (Ursus americanus) to pre-vent disuse osteoporosis. Journal of Experimental Bi-ology 209, 1630–8.

Donahue SW, McGee ME, Harvey KB, Vaughan MR, Robbins CT (2006b). Hibernating bears as a model for preventing disuse osteoporous. Journal of Biome-chanics 39, 1480–8.

Echols KN, Vaughan MR, Moll D (2004). Evaluation of subcutaneous implants for monitoring American black bear cub survival. Ursus 15, 172–80.

Esclamado RM, Damiano GA, Cummings CW (1990). Effects of local hypothermia on early wound repair. Head and Neck Surgery 116, 803–8.

Farley SD, Robbins CT (1994). Development of two methods to estimate body composition of bears. Ca-nadian Journal of Zoology 72, 220–6.

Fedorov VB, Goropashnaya AV, Tøien O et al. (2009). Elevated expression of protein biosynthesis genes in liver and muscle of hibernating black bears (Ursus americanus). Physiological Genomics. 37, 108–118.

Fedorov VB, Goropashnaya AV, Tøien O et al. (2011). Modulation of gene expression in heart and liver of

hibernating black bears (Ursus americanus). BioMed Central Genomics. 12, 171.

Gerstein AD, Phillips TJ, Rogers GS, Gilchrest BA (1996). Wound healing and aging. Dermatologic Clinics 11, 749–57.

Guth L, Barrett CP, Donati EJ, Deshpande SS, Albu-querque EX (1981). Histopathological reactions and axonal regeneration in the transacted spinal cord of hibernating squirrels. Journal of Comparative Neu-rology 203, 297–308.

Harlow HJ, Lohuis T, Beck TD, Iaizzo PA (2001). Mus-cle strength in overwintering bears. Nature 409, 997.

Harlow HJ, Lohuis T, Grogam RG, Beck TDI (2002). Body mass and lipid changes by hibernating repro-ductive and nonreproductive black bears (Ursus americanus). Journal of Mammalogy 83, 1020–5.

Harlow HJ, Lohuis T, Anderson-Sprecher C, Beck TDI (2004). Body surface temperature of hibernating black bears may be related to periodic muscle activi-ty. Journal of Mammalogy 85, 414–9.

Hellgren EC (1998). Physiology of hibernation in bears. Ursus 10, 467–77.

Jessup DA, Koch DB (1984). Surgical implantation of a radiotelemetry device in wild black bears, Ursus americanus. California Fish and Game 70, 163–6.

Ladenson PW, Levin AA, Ridgway EC, Daniels GH (1984). Complications of surgery in hypothyroid pa-tients. American Journal of Medicine 77, 261–6.

Laske TG, Harlow HJ, Werder JC, Marshall MT, Iaizzo PA (2005). High capacity implantable data recorders: system design and experience in canines and denning black bears. Journal of Biomechanical Engineering 127, 964–71.

Laske TG, Harlow HJ, Garshelis DL, Iaizzo PA (2010). Extreme respiratory sinus arrhythmia enables over-wintering black bear survival – physiological insights and applications to human medicine. Journal of Car-diovascular Translational Research 5, 559–69.

Laske TG, Garshelis DL, Iaizzo PA (2011). Monitoring the wild black bear’s reactions to human and environ-mental stressors. BioMed Central Physiology 11, 13.

Lechler E, Penick GD (1963). Blood clotting defect in hibernating ground squirrels (Citellus tridecemlinea-tus). American Journal of Physiology 205, 985–8.

Lohuis TD, Beck TDI, Harlow HJ (2005). Hibernating black bears have blood chemistry and plasma amino acid profiles that are indicative of long-term adaptive fasting. Canadian Journal of Zoology 83, 1257–63.


123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051

P. A. Iaizzo et al.

Lohuis TD, Harlow HJ, Beck TDI, Iaizzo PA (2007). Hibernating bears conserve muscle strength and maintain fatigue resistance. Physiological and Bio-chemical Zoology 80, 257–69.

Majno G (1975). The Healing Hand: Man and Wound in the Ancient World. Harvard University Press, Cam-bridge, MA.

Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR (1999). Guideline for prevention of surgical site infection, 1999. Infection Control and Hospital Epi-demiology 20, 247–78.

Martineau L, Shek PN (2005). Evaluation of a bi-layer wound dressing for burn care I. Cooling and wound healing properties. Burns 32, 70–76.

Milbury PE, Vaughan MR, Farley S, Matula Jr GJ, Con-vertino VA, Matson WR (1998). A comparative bear model for mobility-induced osteopenia. Ursus 10, 507–20.

Mori T, Shaw TH, Martin P (2008). Molecular mech-anisms linking wound inflammation and fibrosis: knockdown of osteopontin leads to rapid repair and reduced scarring. Journal of Experimental Medicine 205, 43–51.

Park HY, Phillips T, Kroon C, Murali J, Seah CC (2001). Noncontact thermal wound therapy counteracts the effects of chronic wound fluid on cell cycle-regulato-ry proteins. Wounds 13, 216–22.

Prendergast BJ, Freeman DA, Zucker I, Nelson RJ (2002). Periodic arousal from hibernation is neces-sary for initiation of immune responses in ground squirrels. American Journal of Physiology – Regula-tory, Integrative and Comparative Physiology 282, R1054–62.

Rabkin JM, Hunt TK (1987). Local heat increases blood flow and oxygen tension in wounds. Archives of Sur-gery 122, 221–5.

Remick DG, Xioa H (2006). Hypothermia and sepsis. Frontiers in Bioscience 11, 1006–13.

Robinson C, Santilli SM (1998). Warm-up active wound therapy: a novel approach to the management of chronic venous stasis ulcers. Journal of Vascular Nursing 16, 38–42.

Rodrigues CM, Ma X, Linehan-Stieers C, Fan G, Kren BT, Steer CJ (1999). Ursodeoxycholic acid prevents

cytochrome c release in apoptosis by inhibiting mito-chondrial membrane depolarization and channel for-mation. Cell Death and Differentiation 6, 842–54.

Ruben RL (1982). Early skin responses of hibernating and nonhibernating ground squirrels to topical appli-cations of DMBA. Experientia 38, 612–4.

Ruit KA, Bruce DS, Chiang PP, Oeltgen PR, Wel-born JR, Nilekani SP (1987). Summer hibernation in ground squirrels (Citellus tridecemlineatus) induced by injection of whole or fractured plasma from hiber-nating black bears (Ursus americanus). Journal of Thermal Biology 12, 135–8.

Schultz GS, Sibbald RG, Falanga V et al. (2003). Wound bed preparation: a systematic approach to wound management. Wound Repair and Regenera-tion 11, S1–28.

Sessler DI (2006). Non-pharmacologic prevention of surgical wound infection. Anesthesiology Clinics 24, 279–97.

Sheffield CW, Sessler DI, Hopt HW et al. (1996). Cen-trally and locally mediated thermoregulatory respons-es alter subcutaneous oxygen tension. Wound Repair and Regeneration 4, 339–45.

Smith DA, Barker IK, Allen OB (1988). The effect of ambient temperature and type of wound on healing of cutaneous wounds in the common garter snake (Thamnophis sirtalis). Canadian Journal of Veteri-nary Research 52, 120–28.

Smith IA (1973). (I) Some factors affecting wound heal-ing. Equine Veterinary Journal 5, 47–51.

Sola S, Garshelis DL, Amaral JD et al. (2006). Plas-ma levels of ursodeoxycholic acid in black bears, Ursus americanus: seasonal changes. Comparative Biochemistry and Physiology, Part C Toxicology & Pharmacology 143, 204–8.

Thomas DR (2001). Age-related changes in wound heal-ing. Drugs & Aging 18, 607–20.

Whitney JD, Wickline MM (2003). Treating chronic and acute wounds with warming. Journal of Wound Osto-my & Continence Nursing 30, 199–209.

Xia A, Sato A, Hughes M, Cherry G (2000). Stimulation of fibroblast growth in vitro by intermittent radiant warming. Wound Repair and Regeneration 8, 138–44.

wound healing during hibernation by black bears (ursus americanus) in the wild: elicitation of...

Documents